This is the manual of UP↑SCALE, the scale app in the browser and the firmware of the UP↑SCALE devices: every mode, option and setting explained, with examples. Here you find some info about the different options and possibilities of your smart scale. Check it out!
To start the scale, just power it and take a device with Bluetooth (and a web-browser)! Check if there are new firmware-updates first.
Power ON
For 420/710 Scales
... if you are using our hardware scale models:
Our scales do not have any battery, display or buttons at all! All you will find is a USB-C plug. Simply connect the scale by USB-C-Cable directly with your mobile phone to power it - or use any standard USB-C charger/wall-plug, your computer's USB-port or even a small mobile power-bank! If you plug your scale, it will switch on automatically! There is no automatic power-off, the scale will stay online as long as it is plugged in! By the way: The cable is only used for the energy supply, but not to transfer any data! Just plug the scale to a power-bank or wall-plug and you can move around freely with your mobile phone / display.
For 232 Upscalers
... if you are using our scale upgrade box:
Just connect your 232UPSCALER with a nullmodem-cable to the RS232-Port of your scale and the USB-C port of your 232UPSCALER to an energy source like a wallplug, powerbank or your mobile - or to your computer/printer directly.
For the UP↑SCALE DIY on the ESP32-C3 RS232 Adapter
Plug the adapter into the RS-232 port of your scale and power it with USB-C (charger or power bank) - see UP↑SCALE DIY.
For the UP↑SCALE DIY (your own ESP32-S3/C3 + RS-232 board)
Power the microcontroller board with USB-C (charger or power bank); it gives 3.3 V to the RS-232 board - see UP↑SCALE DIY.
Range or battery - you choose: in the settings you can raise or lower the power of the wireless connection (Bluetooth Power). Use the highest setting in a bigger kitchen, a shop or a workshop, where the scale is on a wall plug anyway. Use a lower setting when you are close to the scale and maybe power it from your phone. And use a very low setting when the scale records something all alone on a battery - out in the woods, for example ;-)
Device & Progressive Web App (PWA)
The UP↑SCALE PWA - which is a simple HTML-page - will run on all mobile phones (Android, iPhone, Windows etc.) as well as on notebooks or desktop computers. There is no App-download and installation, no registration and not even an internet-connection needed to use the scale. (Well, on Apple devices you need an extra app to use our PWA: on iPhone & iPad - on the iPhone the free beacio extension for Safari, on the iPad the Bluefy browser (or WebBLE)! How to start it everywhere and what to do if it does not connect: app setup & help; what it does, with screenshots: the UP↑SCALE app.) All data will stay local and private! Neither the scale nor the app will send any data to us and nothing will get tracked or stored from us at all! Not even cookies! There is no cloud, there are no subscriptions and nothing from us needed to run the scale! Like that it will still work in 50 years, even without us! Just go sure you've downloaded your PWA once ;-)
ONLINE: You can use our scale app directly online. Just use the "Install-To-Homescreen" option in your browser to get a button to the live UP↑SCALE PWA on your screen (how to install it). After you have opened it once and it has loaded completely, the app is stored in your browser and also starts without an internet connection - installing is not required, but an installed app is kept more reliably (a private window keeps nothing, clearing the website data of the browser removes it, and Safari removes it after 7 days without use). The virtual scale (test weights, scales of other makers) is stored the first time you use it online. If there is a new app version, it gets downloaded in the background and a line "APP UPDATE AVAILABLE" shows at the bottom of the screen; its INSTALL UPDATE opens a window with the details and the "INSTALL UPDATE" button (✕ on the line or "LATER" in the window: then it waits in the menu ☰ under ✰ SCALE INFO as ⟳ INSTALL APP UPDATE). Messages at the bottom: every note and warning of the app is a line in one stack at the bottom of the screen, on every page - red warnings at the top (no ✕: they go when the cause is gone), then yellow reminders, then plain notes; ✕ hides a reminder or note until the app is opened again. It also tells you which scale firmware the new app version needs, and warns you if your connected scale has an older one. Nothing changes without that click, so you can also keep your current app version (e.g. together with an older scale firmware). Version numbers: the app has the version of the firmware it belongs to plus a build number for its fixes, e.g. 01.01312.3 (the third build for firmware 01.01312); the update window shows your version and the new one, the head of the ☰ menu and ✰ SCALE INFO show the version you run. Reports and exports carry the line App … · FW … · LEGAL …. iPhone with beacio: don't install it on the Home Screen - from a Home Screen icon beacio gives the app no Bluetooth (online or offline). Save a bookmark in Safari instead: opened from the bookmark in a normal Safari tab, the app also starts without internet, with Bluetooth.
OFFLINE: Additionally you can download the UP↑SCALE PWA as a single file offline version directly to your device and open it from there later! Like this you also can use the scale totally offline! It is recommended to download the latest file once, but you also can find the links later on our update-page!
Screen stays on: while a program runs - recipe, coffee, mold or container filling, check-weighing, counting, sorting or tracking - the app keeps the screen of your phone, tablet or laptop on (screen wake lock), so it doesn't go dark in the middle of a recipe. In the normal weighing mode and on the setup pages the screen goes off as usual. Browsers without this function (or with the battery saver on) just behave like before.
Firmware Updates
This scale will get regular firmware updates to enhance its functions and to add extra modes and options! You can check for updates within the scale app by self - and if you are using the (always updated) online UP↑SCALE PWA, then you also will get a notification at the bottom of the app if there is a new firmware available! If the app needs a newer firmware, it asks you to update the scale. To update the scale, just connect it by USB-C to a computer or notebook, open our update-website and click a button! It could not be more easy!
⚖ Free and PRO version (licence)
Every scale and adapter runs the free version of our firmware - and the PRO version or the ENTERPRISE & COMPLIANCE edition with a licence. The licence is made for one device (its Device ID) and unlocks it there - once and forever (lifetime) or for a month or a year (see month, year or once). No account, no internet needed. Every phone, tablet and computer that connects gets the unlocked version. Firmware and app updates stay free. Without UP↑SCALE hardware, the virtual scale (the app with your own scale) has its own licence: PRO VIRTUAL.
The four versions side by side (all prices incl. VAT):
green better with UP↑SCALE hardware · red a limit of the virtual scale
FREE free for ever
PRO VIRTUAL APP ONLY
PRO with UP↑SCALE hardware
ENTERPRISE & COMPLIANCE includes PRO
What you need
an UP↑SCALE device - or only the app with your own scale (the virtual scale)
no UP↑SCALE hardware: your own scale of another maker (the supported Bluetooth, USB and serial scales) and a browser
24.90 EUR a year or 89.90 EUR once - one licence for 2 browsers (e.g. phone + laptop), each its own virtual scale
from 4.43 EUR a month, 46.90 EUR a year or 119.90 EUR once - per UP↑SCALE device, with as many phones, tablets and computers as you like
from 13.00 EUR a month, 144.00 EUR a year or 599.00 EUR once - per UP↑SCALE device, with as many phones, tablets and computers as you like
Try it first
-
the full PRO for the first 3 days, by itself
14 days of PRO and ENTERPRISE on your device (start it in the app: scale info → LICENCE)
the same 14 days
Weighing
all units (also volumes), tare, prices and nutrition values, counting, simple check-weighing, recipes, coffee timer, labels as an image, the calculators, 1 extra scale; tracking 15 minutes in a row
everything of FREE plus the PRO functions: extended check-weighing, sorting, filling, density, statistics, the sales, CCLUB and ROASTERY modes, stock and lots, printing labels and the EU nutrition table, more extra scales, tracking without pauses, CSV exports, food by barcode and quick add in the diary, gold prices and infusions saved as THINGS
the same PRO functions, plus the connections of the device: the USB serial output (ASCII as HEX or plain text), the Bluetooth broadcast of the weight, ESP-NOW devices and machine control, the Decent Scale protocol for coffee machines
everything of PRO
THINGS and recipes saved
3 THINGS and 1 recipe on a device - 1 THING and 1 recipe on the virtual scale
up to 1000 THINGS and 250 recipes
up to 1000 THINGS and 250 recipes
up to 1000 THINGS and 250 recipes
Records for inspectors and auditors
-
- (a browser can not keep records tamper-evident)
-
yes: users with roles and login, activity log in a hash chain, the UP↑SCALE SEAL on reports and exports, equipment register, signatures, validation, compounding assistant, pipette calibration
Your data
on the device - on the virtual scale in this browser
in this browser: another browser or cleared site data gives a new, empty virtual scale - back up your THINGS and recipes
on the device, for every phone and computer
on the device, the activity log also archived by the app
Several phones and computers
on a device: yes, all at once
no: every browser is its own virtual scale (no sync); it runs only while the app is open
yes, as many as you like, all on the same device at once
yes, as many as you like, each user with their own login
Not included
the PRO and ENTERPRISE functions (gray in the app)
ENTERPRISE; the USB serial output and Bluetooth broadcast, ESP-NOW and machine control, the Decent Scale protocol; calibration in the app (your scale calibrates itself); not verified for trade
the ENTERPRISE functions
-
FREEfree for ever 0 EUR, no time limit
What you need
an UP↑SCALE device - or only the app with your own scale (the virtual scale)
Try it first
-
Weighing
all units (also volumes), tare, prices and nutrition values, counting, simple check-weighing, recipes, coffee timer, labels as an image, the calculators, 1 extra scale; tracking 15 minutes in a row
THINGS and recipes saved
3 THINGS and 1 recipe on a device - 1 THING and 1 recipe on the virtual scale
Records for inspectors and auditors
-
Your data
on the device - on the virtual scale in this browser
Several phones and computers
on a device: yes, all at once
Not included
the PRO and ENTERPRISE functions (gray in the app)
PRO VIRTUALAPP ONLY 24.90 EUR a year or 89.90 EUR once - one licence for 2 browsers (e.g. phone + laptop), each its own virtual scale
What you need
no UP↑SCALE hardware: your own scale of another maker (the supported Bluetooth, USB and serial scales) and a browser
Try it first
the full PRO for the first 3 days, by itself
Weighing
everything of FREE plus the PRO functions: extended check-weighing, sorting, filling, density, statistics, the sales, CCLUB and ROASTERY modes, stock and lots, printing labels and the EU nutrition table, more extra scales, tracking without pauses, CSV exports, food by barcode and quick add in the diary, gold prices and infusions saved as THINGS
THINGS and recipes saved
up to 1000 THINGS and 250 recipes
Records for inspectors and auditors
- (a browser can not keep records tamper-evident)
Your data
in this browser: another browser or cleared site data gives a new, empty virtual scale - back up your THINGS and recipes
Several phones and computers
no: every browser is its own virtual scale (no sync); it runs only while the app is open
Not included
ENTERPRISE; the USB serial output and Bluetooth broadcast, ESP-NOW and machine control, the Decent Scale protocol; calibration in the app (your scale calibrates itself); not verified for trade
PROwith UP↑SCALE hardware from 4.43 EUR a month, 46.90 EUR a year or 119.90 EUR once - per UP↑SCALE device, with as many phones, tablets and computers as you like
14 days of PRO and ENTERPRISE on your device (start it in the app: scale info → LICENCE)
Weighing
the same PRO functions, plus the connections of the device: the USB serial output (ASCII as HEX or plain text), the Bluetooth broadcast of the weight, ESP-NOW devices and machine control, the Decent Scale protocol for coffee machines
THINGS and recipes saved
up to 1000 THINGS and 250 recipes
Records for inspectors and auditors
-
Your data
on the device, for every phone and computer
Several phones and computers
yes, as many as you like, all on the same device at once
Not included
the ENTERPRISE functions
ENTERPRISE & COMPLIANCEincludes PRO from 13.00 EUR a month, 144.00 EUR a year or 599.00 EUR once - per UP↑SCALE device, with as many phones, tablets and computers as you like
What you need
an UP↑SCALE device, like PRO
Try it first
the same 14 days
Weighing
everything of PRO
THINGS and recipes saved
up to 1000 THINGS and 250 recipes
Records for inspectors and auditors
yes: users with roles and login, activity log in a hash chain, the UP↑SCALE SEAL on reports and exports, equipment register, signatures, validation, compounding assistant, pipette calibration
Your data
on the device, the activity log also archived by the app
Several phones and computers
yes, as many as you like, each user with their own login
Not included
-
FREE and PRO on an UP↑SCALE device in detail:
FREE version
PRO version
weighing in all units, tare, HOLD, prices, volumes, counting, simple check-weighing (zero check), the coffee timer and pour-over, recipes, calibration and all settings
everything of the free version
the connections: several phones and computers, RS-232 setup, serial output, backup - and 1 extra scale
the extended scale tests and a plain print of a single test; the ℮ prepackage control; the reports: print, HTML and plain CSV (sales log, ledger, statistics, stock, tracking, club - not sealed, not checkable), and the statistics series sent by the scale
labels saved as an image (PNG), with a small line "UP↑SCALE FREE VERSION" below
printing labels (print dialog of the browser and label printers directly), labels without that line, the EU nutrition table on labels
coffee timer and pour-over with own pours and steps, the ready-made recipes, sharing recipes
the early stop for the target yield (with learned lag), the timer start by espresso apps and machines, your own coffee recipe library on the scale
up to 3 THINGS and 1 recipe on the scale (more can be loaded, overwritten or deleted)
no limits (only the memory of the scale)
Try PRO first: the PRO functions are gray in the app - a click on one tells you what it is and how to get it. To test them with your own scale and your real work, start the free ENTERPRISE trial: everything of PRO and ENTERPRISE & COMPLIANCE for 14 days, once per scale. PRO pays for the development of UP↑SCALE: no ads, no accounts, no data collection, no subscription needed (every licence is also sold once, for ever). The free version stays free forever and has no time limit.
Tracking is the only mode with a free limit: 15 minutes of recording in a row, then tracking pauses for 10 minutes. The pause only counts while the scale is on, a restart does not skip it; the scale info (LICENCE) shows the minutes left. An ESP-NOW setting saved on the scale stays saved - the scale starts ESP-NOW again as soon as it has PRO. Tools of the app that need no scale (nutrition diary, food search, baker's, gold and infusion calculators, cocktails, conversions) are free. The PRO limits of the app (reports, extra scales, the diary synchronization) apply while a free scale is connected; printing labels, the EU nutrition table on labels, the ℮ prepackage control, the barcode lookup, ⚡ QUICK ADD of the diary and saving a gold price or an infusion as a THING also without a scale, unless the last scale the app saw was a PRO scale.
UPCENT↑420 (Half Decent Scale): all coffee functions of the scale - the coffee timer, pour-over, the early stop for the target yield, the coffee recipe library, the timer start by Decent Scale clients and ESP-NOW (for the pump relay) - are free on this firmware, so its users are not limited in their coffee work. PRO unlocks the other modes there.
See and activate your licence
✰ SCALE INFO is one page of cards: ⚖ ABOUT THIS SCALE (Software ID, Hardware ID, the website), 🔑 LICENCE, 🛡 AUDIT TOOLS (ENTERPRISE: the device qualification form and the user access register), 🛒 BUY (what this scale can get here, each with ORDER, and 🎁 TRY ENTERPRISE), ⟳ UPDATES (the versions of the app and the firmware, on RS-232 adapters also the legal part of LEGAL MODE, and CLICK TO CHECK ONLINE) and 📖 MANUAL & FAQ; the key at the bottom is 💎 LICENCES & PRICES. The card 🔑 LICENCE shows FREE, PRO or ENTERPRISE & COMPLIANCE, until when (a licence for a month or a year), the exact time row and the Device ID of your scale (a tap copies it); the row 🔑 Licence key has 📷 SCAN and ⋯ (PASTE KEY, 📂 KEY FILE, 💾 SAVE KEYS AS FILE, 🔎 FIND MY LICENCE ONLINE / GET MY NEWEST KEY, the lost key link). A warning about the clock of this device or a licence the app can not confirm also stands as a red line at the bottom of the screen while the scale is connected. A free scale also shows ⭐ FREE VERSION - GET PRO (or the offer of the free trial) at the bottom of the screen (a tap or 💎 opens 💎 licences & prices, ✕ hides it until the app is opened again) - switch it off in the settings (LICENCE: PRO / ENTERPRISE notifications, saved in this browser); 7 days before a subscription ends, the bar reminds you to renew (also with the notifications switched off; see month, year or once).
💎 Licences & prices: tap the licence in the head of the ☰ menu, ☛ UPGRADE in the menu, the ⭐ line at the bottom of the screen, 💎 LICENCES & PRICES in a hint ("this is part of PRO") or in the LICENCE box: one page shows what your scale can get, every package with its price (incl. VAT) and an ORDER button that opens the order form with your Device ID and this package filled in. A FREE scale gets PRO and ENTERPRISE & COMPLIANCE (3, 6 or 9 months, 1 or 2 years, for ever); PRO with an end: renew it, PRO for ever or ENTERPRISE; PRO for ever: the upgrade to ENTERPRISE for ever for the difference (or ENTERPRISE for months / years); ENTERPRISE with an end: renew it or for ever; ENTERPRISE for ever: nothing more to buy, just a thank you; the virtual scale: PRO VIRTUAL and what ENTERPRISE gives on our adapters and scales. The order form shows the final price (an offer of the day, the volume discount from 5 boards).
Get PRO: the PRO licence costs from 4.43 EUR a month (3, 6 or 9 months: 19.90 / 33.90 / 39.90 EUR), 46.90 EUR a year or 119.90 EUR once (incl. VAT) for the ESP32-C3 RS232 Adapter and the DIY boards (one firmware, UP↑SCALE DIY). Enter the Device ID when you order - nothing is shipped. The virtual scale (the app with your own scale, no UP↑SCALE hardware) gets PRO VIRTUAL: 24.90 EUR a year or 89.90 EUR once, one licence for 2 browsers.
Find the Device ID: connect the device in the app, tap the licence in the head of the ☰ menu (the 💎 licences & prices page) or open ✰ SCALE INFO: both show the Device ID (12 characters, e.g. 24EC4A1DD634). Its GET PRO button opens the order form with the Device ID already filled in. Without the app (developers): esptool chip-id shows the same 12 characters as MAC.
Activate: you get a link and a QR code by email. Open the link on the device with the app (the app keeps the licence and activates it when the scale connects), or tap 📷 SCAN in the scale info (card LICENCE, row 🔑 Licence key) and scan the QR code, or PASTE KEY (behind ⋯ in the same row). The scale checks the licence by itself and saves it - also after a restart or an update it stays PRO. If an install has erased the scale, it asks for a new calibration first; the app brings the licence back right after the calibration (the scale may restart once for it).
Offline copy of your keys: in the scale info (card LICENCE, ⋯ of the row 🔑 Licence key) tap 💾 SAVE KEYS AS FILE - a small text file with the keys of this scale. Keep it. To install them again without the internet (e.g. after an install with erase): connect the scale and choose the file with 📂 KEY FILE. The keys work only on the device they were made for.
Installed the firmware again with "erase"? Just activate the licence again (the app sends a licence it already knows by itself).
Several scales: only the scale that runs the program needs PRO - free scales can join as additional scales. A free main scale works with 1 extra scale.
Lost the key (e.g. after a new install with "erase")? The app remembers it and sends it again by itself. An app that does not know the key yet (e.g. a new phone) asks you when it saw the board as PRO before and it is free now (erased) whether to look it up on up-scales.com - or tap 🔎 FIND MY LICENCE ONLINE (scale info, card LICENCE, ⋯ of the row 🔑 Licence key). Only with your OK, and only the Device ID is sent; the app never does this by itself. Or the lost licence page (also linked behind the same ⋯) shows your licence again as QR code and link - enter the Device ID. A licence with an end date: 🔎 GET MY NEWEST KEY behind the same ⋯ looks up the key of your last payment the same way.
The app checks the key too: a scale that reports PRO or ENTERPRISE without a valid licence key on it (unofficial firmware) counts as the free version in the app, and the app says so.
Month, year or once
All prices incl. VAT. PRO and ENTERPRISE: one licence per UP↑SCALE device (board), with as many phones, tablets and computers as you like - PRO VIRTUAL: one licence for 2 browsers (the saving is against 3 months):
PRO VIRTUAL APP ONLY, your own scale - 2 browsers per licence
PRO per UP↑SCALE device - unlimited phones & computers
ENTERPRISE & COMPLIANCE includes PRO, per UP↑SCALE device
3 months
-
19.90 EUR 6.63 EUR a month
57.00 EUR 19.00 EUR a month
6 months
-
33.90 EUR 5.65 EUR a month, save 15 %
90.00 EUR 15.00 EUR a month, save 21 %
9 months
-
39.90 EUR 4.43 EUR a month, save 33 %
117.00 EUR 13.00 EUR a month, save 32 %
1 year
24.90 EUR 2.08 EUR a month, for 2 browsers
46.90 EUR 3.91 EUR a month, save 41 %
144.00 EUR 12.00 EUR a month, save 37 %
2 years
-
79.90 EUR 3.33 EUR a month, save 50 %
264.00 EUR 11.00 EUR a month, save 42 %, + PRO for ever
lifetime
89.90 EUR for 2 browsers
119.90 EUR
599.00 EUR upgrade from PRO lifetime: 479.10 EUR
Already have PRO lifetime? The upgrade to ENTERPRISE & COMPLIANCE lifetime costs only the difference: 479.10 EUR (row in the card 🛒 BUY of the scale info and on the licence page; only for a board with a PRO lifetime licence - also a gift or the PRO included with ENTERPRISE for a year).
Several boards: enter how many boards you want and the Device ID of each one (one per line) - one order, one payment, one email with all keys and QR codes. Every licence costs less from 5 boards in one order: 5 % off from 5, 10 % off from 10, 15 % off from 15 boards (months, years and lifetime). Months and years bought together (the same email, ordered within 2 days) end on the same day, so you renew them together. Business customers tick "I order as a business" and enter the company, address and VAT ID; with an EU VAT ID of your country you pay no VAT (reverse charge). Payment:BUY NOW (it shows the total) takes you straight to the secure card payment (also Apple Pay / Google Pay) - the key comes at once, on the page and by email. Left the payment page? The confirmation email has the link to pay later. With an EU VAT ID the shop checks it at once in VIES (the EU VAT service): only a valid one gets the reverse charge.
All prices incl. VAT. In the order form you can buy 3, 6 or 9 months (the longer, the less a month) or 1 or 2 years at once (one key for the whole time). If the scale already has a paid period of the same level (PRO or ENTERPRISE) that has not ended, the new time starts at its end, no day is lost. A month or a year has no automatic charge and we send no reminder emails: 7 days before the end the app shows a yellow line at the bottom - "Your PRO / ENTERPRISE subscription runs out in X days!" with RENEW (opens the order form for the next period) and ✕ (hides it until the app is opened again); after each payment you get a new key (link and QR code) - the next period starts at the end of the paid one, no day is lost. The key works 7 days longer than the paid time, so a late payment does not stop your work. ENTERPRISE for 2 years (paid at once) includes a PRO licence for ever for the same board (a second key in the same email, only if the board has no lifetime licence yet): when ENTERPRISE ends, the scale stays PRO. The ORDER keys of the month and year rows in the card 🛒 BUY of the scale info open the order form for the renewal.
The free ENTERPRISE & COMPLIANCE trial
Every FREE or PRO scale can try ENTERPRISE & COMPLIANCE for 14 days, with no key and no order: 🎁 TRY ENTERPRISE 14 DAYS FREE in the card 🛒 BUY of the scale info (or the main key of the licence page; it asks first, then a line with a turning ⚖ says Starting the ENTERPRISE trial on the scale … until the scale answers, which takes a few seconds; a message tells you that the trial started, or that the scale did not answer). The scale counts the 14 days by itself from the start (also while it is switched off), once per scale; a scale that had ENTERPRISE before gets no trial. During the trial everything of ENTERPRISE and PRO is unlocked. The seal certificate of the scale (seal) comes with a paid licence: seals made in the trial can still be checked with the key of the scale, they only show "no UP↑SCALE certificate".
How the scale knows the date
The scale has no clock. For a licence with an end date and for the trial it takes the time of the phone, tablet or computer that connects (every app sends it at the connect and every hour) and keeps its own one-way clock: it only moves forward, also counting the running time of the scale without any app. It works completely offline.
A device whose clock is behind the scale is ignored - the scale keeps its own time and the card LICENCE of the scale info (and a red line at the bottom of the screen) shows ⚠ The clock of this device is … behind the scale. Check the date and time settings of that device.
🕒 EXACT TIME FROM UP-SCALES.COM (card LICENCE on the scale info page, row 🕒 Exact time, key FROM UP-SCALES.COM; under such a warning with a hint above it): one tap, and after your OK (GET THE TIME) the app asks up-scales.com once for the exact time (from several official time servers), signed for this scale. The scale then sets its clock to it - also back, if it was ahead. Only the Device ID and a one-time random number are sent; it needs an internet connection only for that moment. A line at the bottom of the screen shows while it asks. Tick automatically at every connect in the row below (it asks once first) and the app does it by itself at every connect while this device is online (a setting of this browser); offline the scale simply takes the time of the device as always.
After the scale was switched off, the first time of an app may move the clock forward by the days it was off. While the scale runs, a time far ahead is refused (and written to the activity log on ENTERPRISE scales).
Would that first time move the clock more than 2 days ahead and end your time licence, gift or trial? Then the scale keeps its own time and the app asks first: Is the date of this device right? Only after a yes does the scale take the new date (written to the activity log). Until then the licence is paused: ENTERPRISE works as in records mode (the activity log, logins and the archive go on, the other tools wait) and a PRO subscription as FREE. Set the date of the phone right and connect again, or press 📅 THE DATE IS RIGHT in the LICENCE box. So a wrong date on a phone can't end a licence by mistake.
With users on the scale only a logged-in device sets this clock.
Was the clock of the scale still set too far ahead (a wrong date confirmed by mistake), so that your time licence or gift ended too early? Write to us: you get a clock repair key that sets the clock right once and changes nothing else on your licence.
Limits, honestly: someone who erases the flash with developer tools can get around the date (an erase also deletes all data of the scale). The rules are made against mistakes and casual tricks, not against firmware hackers.
When a licence ends
PRO (a month or a year, not renewed): the scale goes back to the free version. Your THINGS, recipes and all data stay on it; the limits of the free version apply again until you renew.
ENTERPRISE & COMPLIANCE (a month, a year or the trial): the scale goes back to PRO (with a PRO licence - also the one included with the yearly ENTERPRISE licence) or to the free version and keeps all records - records mode: the activity log goes on, users can still log in, the archive of the log and the record files can still be saved and checked. New reports, seals, signatures, reviews, the user and device administration and new scale profiles above 3 need ENTERPRISE again; the gates can only be switched off.
PART 2 Weighing & Modes
⚖ Weighing Modes & Functions
Overview
After you are connected with the scale, you will land directly in the normal, simple weighing mode! Here you already have a lots of additional options - but you also can choose between different scale modes in the menu! Each of them is specially made for different use-cases and includes a variety of different sub-modes and options!
The start page
Before a scale is connected the app shows its start page. At the top the empty weight display (– – – g, NO SCALE CONNECTED) and right below it, highlighted, the key CLICK TO PAIR WITH A SCALE!: the browser lists the scales around you (UP↑SCALE scales and adapters and the scales of other makers) - choose yours. While it connects, the key shows the steps. Where this browser can not pair, the key is gray with a line below it - tap it anyway to see why and what to do (another browser, iPhone / iPad, a switch of the browser). Below it OTHER WAYS TO CONNECT (the i explains them): 🔌 USB cable, ⌨ Keyboard scale, 📡 Broadcast scale and 🧰 NO SCALE? TRY IT WITH TEST WEIGHTS, each with its key - a way your browser can not use is gray, says why, and its i tells more. Then the line about the scales of other makers (does mine work?), the field WHAT DO YOU WEIGH?, a list with the 📒 nutrition diary (only when this browser already has diary entries), privacy, offline use (with 📲 INSTALL THE APP where the browser offers it) and help - a tap unfolds each -, the links to the website (about the app, FREE & PRO, ⭐ UPGRADE, firmware, Instagram), a short introduction and the FAQ. The language is the select in the head; ⌂ HOMEPAGE and ▤ MANUAL are the keys at the bottom. A link with a recipe of the 🔖 bookmarklet shows a line at the bottom until you connect.
The head of the page
The dark bar at the top is the same on every page. ☰ at its left opens the menu. Its two lines tell you where you are: the small line is the mode, in short (for example ⚖ WEIGHING or 🧑🍳 RECIPES), the large line is what is open - the title of the page, the loaded recipe, or else the THING or name you typed, or the scale itself. A text that is too long ends with … (point at it to read all of it). After a lost connection the head shows the same as before. Short answers to a tap - saved, copied, done, loading - appear as a small note below the head, green when it worked, and leave by themselves; warnings and offers (for example the PRO trial or a missing clock) stay as lines at the bottom of the screen, above the keys. On the right: the link dot (green: the weight comes in; red: no connection or no data - not on the pages with their own ✕, e.g. the settings: there the menu tells the link, and its scale name turns red; tap the dot for a small connection menu right there: ⚖ DISCONNECT or ⚖ RECONNECT, ⚖ SWITCH SCALE and ⚖ ADD MORE SCALES), the battery of scales that have one (red when it is almost empty - tap it for the voltage) and the person who is logged in (role sign and initials). The QR key shows while the large line is a recipe saved on the scale (a tap opens its label), and the round red ✕ at the right end leaves the page you are on.
The weight at the top and the keys at the bottom
On every page with a weight, the dark display below the head stays in place while the page scrolls. Its top row: ↰ TAKE-AWAY at the left (takeaway mode), 卄 HOLD at the right (hold) - both marked while they are on, gray while there is no connection - and between them the marks of the display: NET (a tare is active), STABLE (the weight is stable), →0← (the scale is exactly at zero, see center of zero) and the key ⛶: the weight big on the whole screen (SPACE tares the scale there, ✕ or Esc closes it). A held weight shows in the 卄 HOLD key, which is lit then. The number never changes its colour: while the weight moves it is dimmed and STABLE is off. More connected scales stand in a row above the number (⚖2 weight, ⚞T⚟ tares only this scale, ✕ disconnects it). Right below the number: the name of the THING or the name you typed in the THING / NAME field (not while a recipe is loaded: its name is in the head). Below that: the lines that belong to the weight (the NET line with its i, the stability light, the container that is taken off with 👻 and ✕, a virtual weight in red with ✕, what is left in the takeaway mode, the hold, the take-from with 🍽 and ✕, the extra line of an RS-232 scale, the true mass). Without data the number shows – – –. A tap on the number opens the window About the weight: what the marks mean, the scale behind the weight (with its Max and d), the virtual weight, ⚙ SETTINGS and 📖 MANUAL. Below the display the unit cards show your units live (two in a row on a phone, the numbers at the right; a price highlighted, in the sales modes the whole width) - a tap on a card opens its goal; ✎, 👻 and ✕ of a unit are at its chip. Below the cards: the sums of the log (Weighed 2x · Total and ∅ - a tap opens the log). The keys are in one row at the bottom of the screen: ⚞ TARE ⚟ at the left end, the action of the page at the right end (∑ LOG IT on the weighing and counting page, the next step of a running program - e.g. OVERTAKE in the check-weighing and the factory -, END THE PROGRAM in the sorting run; elsewhere a program ends with the round ✕ of the head); SPACE = TARE, ENTER = ∑ LOG IT. On a narrow screen (a phone) the keys between the two ends show only their sign (↶, 📊, 🔔 …) - the name is in their tooltip. When the next step needs a key at the bottom, that key is green and pulses slightly (e.g. ▶ START after you have filled in the fields at the top, OVERTAKE when the weight is right, SAVE after a change of the settings); a key that ends what is running is red (e.g. STOP of the coffee timer, END THE PROGRAM). The messages of the app stand right above the keys.
☰ The menu
Everything that is not a step of the page you are on is in one menu - on every page. On a phone or tablet, tap ☰ at the left of the head: the menu slides in (it closes with ✕, Esc, a tap beside it or when you choose an entry). On a computer (a wide window) it stays open as a sidebar on the left: « makes it a narrow row of icons, » or a click on an icon makes it wide again (your browser remembers it). The head of the menu shows the connected scale; below it the view, the licence (FREE, PRO or ENTERPRISE - tap it for 💎 licences & prices), the battery, who is logged in, the link (connected with a green dot or not connected with a red one, right-aligned above the key) and the key DISCONNECT over the whole width, which disconnects the scale at once. The scale name at the top left of the menu turns red while there is no link or no data.
The menu is sorted into categories. A tap on a category opens it and folds the one that was open, so the names of all categories always stay in view; a second tap folds it again. Your browser remembers the open category. The place you are in is marked in the colour of your view.
⚖ MODES: ★ your last used modes on top, ALL MODES (folded, a tap opens it) below in groups (basic, kitchen, production, lab, trade) - the i beside the name of a group explains what its modes do. At the end of ALL MODES: ☑ TESTING MODE, the page of all tests of the scale. The mode you are in is marked; PRO and ENTERPRISE modes are gray with a ⭐ in the free version. You change the mode on the weighing or counting page, and straight from the pages where nothing runs: ✰ SCALE INFO, the settings pages (with changes that are not saved, the app asks first), the page of all tests and the result of a test. While a program runs (factory, tracking, a recipe, a test …) the modes are gray: finish or cancel it first.
🗂 RECORDS: one entry for each part of the records page - 🛡 ACTIVITY, ☑ TESTS, ✓ ROUTINE, 🖥 DEVICES, 👥 ACCESS, 🧾 LOTS, ⚕ BATCHES, 💶 SALES and, when your view has them, 🥦 CLUB, ☕ ROASTERY, 🏪 DISPENSARY and 💧 PIPETTES: a tap opens the records right there.
🛠︎ SCALE SETTINGS:⚙ ALL SETTINGS opens the settings page of the scale; 🔌 RS-232 SETUP (on the RS-232 adapter) opens its own page. The other entries go straight to one part of the settings, sorted into five groups that each fold open and shut on their own (a tap on the title; the app remembers which ones you left open): SCALE (scale data, basic settings, USB / Bluetooth / ESP-NOW, 🏷 LABEL PRINTER - its own window), WEIGHING (calibration, speed & stability, creep compensation), ENVIRONMENT (sensors, environment test, air buoyancy), AUDIT & DATA (seal, register & compliance, LIMS / ELN) and BUSINESS (inventory, cannabis club). Entries that go to the settings page are gray where they would leave a running program.
✰ SCALE INFO: ⟳ INSTALL APP UPDATE (only while a new app version waits), ✰ SCALE INFO (gray where it would leave a running program), ▤ SCALE MANUAL (this manual), ☛ UPGRADE TO PRO (a FREE scale, also during the free trial) or ☛ UPGRADE TO ENTERPRISE (a PRO scale) - the 💎 licences & prices page - and ⌂ HOMEPAGE (up-scales.com in a new tab).
🤝 CONNECTION: ⚖ DISCONNECT / ⚖ RECONNECT, ⚖ SWITCH SCALE, ⚖ ADD MORE SCALES and 📡 SENSORS.
🌐 VIEW: the app view (kitchen, diet, coffee, cannabis, shop, packing, laboratory, audit & metrology or the full app), the colour scheme (light or dark), the weight unit and the language.
💬 SPEECH OUTPUT: the app speaks the weight. Opening the group loads the voices of your browser (a short loading window; if your browser has none, it tells you there) - then choose a voice, or STOP.
📒 NUTRITION DIARY (apart, always at the very bottom of the menu - it opens upwards): 📖 OPEN THE DIARY and, in the weighing mode, ➕ ADD THE WEIGHT and 🍽 TAKE-FROM (before & after).
Leaving a page: every page of a mode or a program that you leave again (the start page of the recipe or coffee mode, a setup, the solution or tracking page …) has a round ✕ at the right end of the head - it goes back (to the page before or to the weighing). Before you leave a running pipette run or a pipette result that is not saved, and before END THE PROGRAM, the app asks in its own window. The title of such a page is the large line of the head; its parts are boxes, and the longer explanations are behind the ℹ️ of each box.
A red dot on ☰ means: the scale is not connected or sends no data, its battery is almost empty or an app update is ready. When the link is lost, a red line NO CONNECTION TO THE SCALE with the key RECONNECT stays at the bottom of the screen on every page (and ⚖ RECONNECT is in the menu).
If the connection drops (out of range, scale restarted), the app opens it again by itself: first after a moment, then every few seconds for ten minutes, without any message. While it tries, the ⚖ in front of the red line turns. You can press RECONNECT at any time to try at once. DISCONNECT stops it, the app only connects again when you press RECONNECT. When the connection still stands but no data come (the red line NO DATA FROM THE SCALE), its RECONNECT ends that connection and opens it again. A scale with a password asks for it again after every new connection.
The app can do a lot - and most people only need a part of it. An app view shows only the modes and tools of one group, as tiles under YOUR VIEW at the end of the weighing page (a mode of a higher edition gray with its mark), its own colour and a badge in the head of the menu. The layout stays the same, and nothing is changed on the scale: the view is kept in your browser, for each scale on its own.
Choose it: on the start page, the field WHAT DO YOU WEIGH? below the ways to connect: 🍳 🥗 ☕ 🥦 🛒 📦 🧪 🔍 or ⚖ for the full app (its i explains it). The line below it then tells what the view holds, in the colour of the view. It counts for the pair key and every other way to connect alike. If you have never chosen one, the app asks once after the first connect. A link can choose it too: up-scales.com/app/?view=coffee (kitchen, diet, coffee, cannabis, shop, packing, lab, audit or full).
Change it: ☰ → 🌐 VIEW → App view, or tap the badge in the head of the menu. ➕ ALL MODES (FULL APP) at the end of ⚖ MODES opens everything again.
The views: 🍳 KITCHEN (recipes, cocktails, tracking for dough and fermentation, the 📒 diary) · 🥗 DIET & HEALTH (the 📒 diary, recipes) · ☕ COFFEE (☕ ESPRESSO MACHINE and ☕ FILTER CONE as buttons of their own, the roastery, 🕘 brew history) · 🥦 CANNABIS (CCLUB, the DISPENSARY MODE for Metrc, sales, recipes for edibles, drying with the tracking mode, counting, the 🌿 infusion calculator, stock, labels) · 🛒 SHOP & SALES (sales, counting, density for gold, stock, labels, records) · 📦 PACKING & FACTORY (check-weighing, factory, sorting, tracking, 📊 statistics & ℮, labels, stock) · 🧪 LABORATORY (check-weighing, density, solutions, compounding, pipettes, ☑ tests, statistics, records, 📡 sensors) · 🔍 AUDIT & METROLOGY for auditors, QA, inspectors and service technicians (☑ the tests of the scale and their history, the routine checks, the device qualification, the validation pack, the activity log, the user access, 🔍 CHECK A REPORT FILE for a sealed PDF or a saved HTML report (its seal, the attached files, the signatures - no scale and no licence needed), scale data, calibration, the environment test, the register, 📡 sensors, 📊 statistics & ℮, the pipettes with their register and - on an RS-232 adapter - 🔏 LEGAL MODE; in the menu it leaves out the stock, the label printer, the cannabis club, the dispensary and the sales records) · ⚖ FULL APP (everything, as without a view).
What stays: the weighing mode, the settings, the logs and the connection are in every view. The licence works as always: PRO and ENTERPRISE modes stay gray with a ⭐ in the free version, the pipette modes need a lab balance on the RS-232 adapter (or the virtual scale) - on other scales the LABORATORY view shows them as gray buttons that tell you so. If the scale is in a mode outside your view (for example another phone switched it), the menu shows that mode with the note not part of this view.
🌗 Light or dark
☰ → 🌐 VIEW → Colour scheme: 🌗 SYSTEM (the app follows your phone or computer: light by day, dark when your device is set to dark), 🌕 LIGHT or 🌚 DARK. Your choice is kept in this browser and holds from the first moment the app opens. The weighing colours (orange, yellow, green, red) and the charts stay readable in both. Printed reports, PDFs, labels and the label preview stay black on white, as on paper.
Weighing ⚖ & Counting 🔢 Mode
The Weighing ⚖ Mode
This is the standard scale mode and your starting point if you open the scale App. If you want to weigh things - here you go! Additionally you have the following options:
💶 Sales Log - daily sales per THING (pieces, weight, revenue, profit), 366 days saved on the scale, every record numbered and chained (checkable), print and CSV: with the inventory and the labels your scale is also a sales computer
If you have many small things of the same weight, like screws, bags or whatever, and if you want to know how many without counting every single one manually - here you go. In the counting mode you are able to:
💶 Sales Log - daily sales per THING (pieces, weight, revenue, profit), 366 days saved on the scale, every record numbered and chained (checkable), print and CSV: with the inventory and the labels your scale is also a sales computer
Well, this is a scale - so you can weigh things! Use the tare-button to - well - tare the scale, as example with a container on it. A tare waits (up to 3 seconds) until the weight has stopped moving, so a hand or a swinging load does not end up in the zero - the same goes for the zero when the scale starts.
→0← ZERO and ⚞ TARE ⚟ - the difference:ZERO means "the empty scale weighs 0": crumbs, a wet plate or a little drift are gone, and no tare is left (no NET, the gross weight is 0 again). TARE means "take this container away": the weight is shown as NET, and the scale still knows the gross weight (net + tare) - that is what a label with gross / tare / net needs. That is why ZERO only works close to the zero the scale had at its start (up to 4 % of Max, OIML R 76-1): with a container on the scale it is refused - use TARE for that. After a tare, take the container off and press ZERO: the tare is gone and the empty scale shows 0. Like a tare, ZERO waits for a stable weight (tare guard). Where: in the window of the weight (a click on the weight) on every page where TARE tares; next to TARE in the key row in the sales modes, the tracking, with Zero Memory on and in the strict audit mode. Also USB Z and the WATCH LINK (z). RS-232 adapters: ZERO sends the zero command of the connected scale (Mettler Toledo, Kern, Ohaus, Radwag, Sartorius, CAS in its command mode, Dini Argeo, Rice Lake, SMA, NCI, Toledo, Toledo POS, Shinko / Vibra) - the scale checks its own zero range; when it shows 0, the tare of the adapter goes too. Not with listen only, and not for scales with one combined ZERO/TARE key (A&D, Shimadzu, Adam): use TARE there.
🔢 Counting
In the counting-mode you can count things. At first you need to tell the scale how much 1 of the things is weighing:
By direct input
If you know the weight, just input it.
Or let the scale check...
If you don't know the weight, just check-weigh a couple of your things. Take a known amount, for example 10 of the screws you are going to count, place them on the scale and there you go!
After that the scale will show you the amount of things on the scale, as like as the total- and the average weight of them! Also you can log and count everything in pieces now.
The steps: the title of every step is in the head (its ✕ goes back), below it one card says what to do now in bold, with an ℹ️ for the details. The first page shows the two ways as rows (⌨ BY DIRECT INSERT · ⚖ OR WEIGH THINGS). The field of a step has its unit beside it (pcs or your weight unit); the key that goes on is at the right end of the key row at the bottom (🆗 OK ▸, 🆗 OK, START COUNTING, 🆗 OK) - Enter in the field does the same. A wrong value gets a red line right under the field: the weight of one piece has to be bigger than 0, the amount of pieces a whole number of 1 or more. The step "Let's weigh 10 things" names the amount you typed (a second device that joined later just says "Let's weigh the things"). While counting, the pieces are the big number in the display; the lines below it show the average and total weight, where the piece weight comes from and up to how many pieces the count is exact - the warnings (very light pieces, more pieces than the exact range, 🎯 add more pieces) come as yellow lines above the keys. ⚞ TARE ⚟ on the counting page tares all connected scales, as in the weighing mode. ∑ LOG IT checks the weight against the tolerance of the scale in grams, whatever weight unit you use.
🔗 Piece weight from an other scale (two-scale reference counting)
Small, light pieces in a big amount? Count the sample on a fine scale and the bulk on a big one - like the counting systems in industry. Connect the fine scale as additional scale, start counting on the big (main) scale, and on its start page put the counted pieces on the fine scale (tared, without a container). Enter how many they are and press USE ⚖2: the app takes the stable weight of the fine scale, divides it by the pieces and sends this piece weight to the main scale - like a typed in weight per piece ("Piece weight from ⚖2 ... (50 pcs)"). A finer scale gives a more exact piece weight - so the count of the bulk is more exact too. The fine scale then stays a device that counts its pieces with the same piece weight - and they are added to the big number (see below).
⚖+⚖ Count on many scales at once: one sum, live
Connect more scales to the app (1, 2, many scales) and start counting on the main scale: now all scales count together. Every scale shows its own weight and pieces small on top ("⚖1: 409.2 g = 100 pcs", "⚖2: 40.9 g = 10 pcs"), the big number in the middle is the sum of all of them - and so is everything else: the average and total weight below it (of all pieces on all scales), your units per piece (prices, mg, kcal...), the 🏷 label and the ∑ log. Put pieces on any scale, take them from one to the other: the sum follows live, whichever scale changes. Every scale gets rounded to whole pieces on its own, so the pieces on top always add up exactly to the big number. All scales count with the piece weight of the main scale - typed in, sampled on it, or taken from a finer scale with USE ⚖2 (above). Each scale has its own ⚞T⚟ tare button. The takeaway mode works with the sum too: it remembers what is on all scales and counts what you take from any of them.
⚖+⚖ WHAT NO NORMAL COUNTING SCALE CAN DO
Counting systems with a reference scale and a bulk scale are known from industry - with cables, a fixed setup and a price to match. Here it is wireless and as big as you need it: count the sample on the fine 710 (0.01 g), the bulk on the 420 - or on your old floor scale with the 232 UP↑SCALER - and add more scales whenever the amount does not fit on one. One phone shows the pieces of every scale and the exact total, with prices, label and log.
Example: 5,000 small screws for a delivery, in three bins on three scales. Count 50 of them on the 710 and press USE ⚖2 - fill the bins on all scales at once and watch the big number grow up to 5,000. Print the label with the exact count and the price of all of them, log it - done, without counting one bin after the other.
🎯 More exact with every piece: piece weight optimisation
Like professional counting scales, the scale makes the piece weight more exact by itself while you are sampling: after the sample (e.g. 10 pieces), simply add more pieces - up to 3 times the sample (here 30 pieces) on the scale. As soon as the count is stable for a second and surely exact, the scale takes it as the new sample and calculates the piece weight again ("🎯 Piece weight optimised with 25 pcs!"). Add more pieces again and it goes on - the bigger the sample, the more exact the count. The optimisation ends as soon as you take pieces away, tare or log (∑) - so a container, the takeaway mode or virtual weights can never change the piece weight. A piece weight that you typed in is never changed.
Below the totals you see where the piece weight comes from ("Piece weight from 25 pcs") and up to how many pieces the count is exact ("exact up to ≈ 870 pcs", calculated with the precision of your scale). Two warnings help you: if the pieces are very light for the scale (less than 10 digits each, e.g. under 1 g on the 420), use a bigger sample - or a finer scale like the 710. And if there are more pieces than the exact range, the count can be off by a piece: count a bigger sample (or let it get optimised) for amounts like this.
⬆️ Takeaway Mode
If you switch on the Takeaway Mode (with the key ↰ TAKE-AWAY at the top left of the weight, in the weighing and counting mode - marked while it is on), then the scale will remember how much is on the scale and weigh how much you are taking from it instead of what you are adding. Place a full container/bowl on the scale, hit the button and just count how much you get out from it! (The scale will show you also how much is left on it in that case.) In the 🛍️ sales mode a booking takes what you took away (with its price), not what is left on the scale; the scale can not check this weight against its own reading (it is the difference of two weighings), so the record is not marked as verified by the scale.
卄OLD: freeze the weight
For animals, wobbly or big things - or when you can't see the display while weighing: the key 卄 HOLD at the top right of the weight (in the weighing and counting mode) holds the weight. 1 click: the next weight is held (the key shows HOLD 1×), 2 clicks: every weight is held (HOLD ∞), 3 clicks: off; the key is marked while it is on, and the mark HOLD of the display is lit while a weight is held. As soon as the weight is stable, the app averages it for 2 seconds - small movements (a pet on the scale) stay in the average, up to 5 % - and then freezes it: the display, your units, the label and the ∑ log use the held value (in the counting mode the held pieces). A line below the weight shows what happens (with an ℹ️ that explains the hold in short; "waiting for a stable weight…", "averaging… 1.4 s" with a countdown, "250.3 g held.") - the weight is held after the 2 seconds, even if nothing moves on the scale. Take it off: with "next weight" the hold ends, with "every weight" the app waits for the next one. Adding more while a weight is held does not change it - take it off first. The hold is done in the app, so it works with every firmware.
🍽 TAKE-FROM Mode (leftovers, jars, boxes)
An extended takeaway mode: you weigh before and after, and in between the scale can be empty - you can take your plate to the table, or your jar back to the shelf. Put the full plate, jar or box on the scale (with its units on the page, e.g. a loaded THING with nutrition values or THC, or the processed food from batch processing) and press 🍽 TAKE-FROM (before & after) in the menu (☰ → 📒 NUTRITION DIARY) (weighing mode only). The app remembers the weight and the values, the entry turns green, and the full weight is written into the log right away ("🍽 510.4 g on the scale - waiting for the 2nd 🍽") - so you can still see it later. This line counts nowhere yet: not in the totals, the diary or the servings calculator. After eating (or taking something out), put it back on the scale and press the 🍽 in the line below the weight (or 🍽 TAKE-FROM in the menu again): the same line becomes the difference, with all units calculated for it ("🍽 from 510.4 g", what is left is shown as "Rest"). No name needed: it gets the name from the name field, if there is one. The log opens with the new line - and if it has nutrition values, the 📒 diary dialog opens with just this line. Without units only the weight gets logged. The first weight is kept in your browser, so the app can be closed in between (the log is not: after a restart the difference gets a new line). ✕ next to the info cancels it (asked first) and removes the waiting line from the log.
Add units / conversions to the screen
You can add an unlimited amount of units to the screen. This units will be updated live together with the weight and will be also included in the log-files. You can choose between many preset units - but also add your own, customized ones easily. Here an overview about the different units (every unit with short examples: THINGS & UNITS):
In the THING / NAME card (below the unit cards; on a wide screen the right column) is the field UNITS: choose a kind in its list + add a unit (PRICE PER WEIGHT 💰, INGREDIENTS (METRIC) 🧪, CANNABINOIDS 🌿, INFUSION CALCULATOR 🧈 (opens the calculator), FOOD NUTRITION VALUES 🍉, WEIGHT <--> VOLUME 🥛, HYDRATION (DOUGH) 💧, PERCENT WEIGHING 💯, CUSTOM UNITS 🔣, LOT NUMBER 🧾, CONTAINER/JAR WEIGHT 🫙) and its window opens (the list shows its first line again). The ℹ️ right behind its title explains it, ✕ (or Esc, or a click beside it) closes it; the fields are labelled, the key to add is at the bottom right and stays gray until the input is complete - a wrong input is said in the line above the keys. After a unit is added, the window closes by itself and the new unit shows up below the weight. The windows of the camera, the lot list and the baker's tools open on top of it. The ℹ️ behind UNITS gives a short overview of all kinds. Every unit on the page is a small chip below the list:✎ opens its editor in place of the chips to change its value (OK or Enter saves it, ✕ or Esc closes it, a wrong number is said below the field; the chip of Carbohydrate also chooses Carbohydrate / Bread Units / Carbohydrate Units there, the chip of Sodium Salt / Sodium - the name of its card), 👻 hides its card below the weight (the chip then shows the name struck through and 👻 in an underlined box - a tap on 👻 shows it again; the hidden units stay in the THING and the logs), ✕ removes it. A percent unit has no ✎. With more than one unit, the last chip ALL hides them all (👻), removes only the hidden ones (🧹) or removes them all (✕). Removing a unit that is in the current log (✕ or ALL ✕) asks first: hide it with 👻 instead if you want to keep its sum in the log. A price card shows the computed price of the weight and, below it, the price per weight unit of the app (e.g. 9.50 €/g; marked 🧾 when it is the price of the chosen lot) - in every mode with live cards, also the sales and CCLUB mode. While a THING is loaded the card also shows a small box with its mini QR code and 🏷 LABEL: a click opens the label window of the THING.
Conversion Units can be added in counting- and weighing mode and can be simply a price per weigh-unit, or the calories of a food product or the amount in milliliters/cups if you are weighing something liquid. Once defined (for example as 4.20 EUR per Gram) this units will be shown on the scale and updated with the weight on it (at 100g == 420.00 EUR, to stay with the example.) You also can setup goals for fill-weighing based on that amounts now! Like this you can automatically calculate, show and track different units additional to the weight live on the scale, show the volume of something or subtract a container weight!
You can add an unlimited amount of units to the scale at the same time! Additionally you are able to save this once defined units together with a name as a THING on the scale in order to reload them with a click later! Like that you can add your custom products, foods or containers easily!
💰 Prices / Costs
To add a price and to show it live with the weight, simply select one of the currencies from the selection and add the price per gram (or per piece in the counting modes). After a click on the button, the price will be shown on top below the weight and update automatically with every weight change. You also can choose just "🪙" or "💰" instead of $ or or EUR. And for sure you can add multiple currencies at the same time. Like that you can show your price in Dollars, Euros and maybe in Bitcoin at the same time - just as you prefer! Buying or selling gold? The 🪙 GOLD PRICE CALCULATOR key in the price window calculates the price per gram from the karat, the spot price and your payout.
🪙 Gold price calculator
Open 🪙 GOLD PRICE CALCULATOR in the window of the 💰 price units (left of its ADD key). It works like the 🌿 infusion calculator, in three steps:
1. The gold
Choose the karat: 24 K (999 ‰), 22 K (916), 21 K (875), 20 K (833), 18 K (750), 14 K (585), 10 K (417), 9 K (375), 8 K (333) - or type an own fineness in ‰ (e.g. 730).
2. The price
Type in the spot price of fine gold per troy ounce (31.1 g), gram or kilogram, in one of the currencies of the price units - the app has no price feed and never goes online, so you type in today's price from your bank or a price website. And your payout in % of the gold value (e.g. 90 %; 100 % is the pure gold value). The calculator keeps karat, price, currency and payout in this browser.
3. The piece (optional)
Weigh the piece (⤓⚖ WEIGHT FROM SCALE, like in the infusion calculator): the result shows its fine gold, its value and the payout.
💾 Save as THING
Saves a THING named like "18 Karat Gold" (or your own name) with two price units per gram: Spot-Price (the gold value of the alloy: spot per gram × fineness) and Payout (the value × your payout %). Load it in the weighing mode and every piece on the scale shows both. New spot price tomorrow? Open the calculator, type it in and save the THING again (overwrite). ⚖ TO WEIGHING shows the prices with the weight without saving.
⭐ PRO:💾 Save as THING and using the gold price for weighing (the calculation itself stays free) - gray in the free version; try it with the free trial (free and PRO).
🧪 Fixed metric units
Sometimes you want to show or keep track the amount of something. For example Magnesium in milligrams - even if you run the scale with ounces. With this option you can add the metric amount and the name of the ingredient to show its absolute amount on top below the weight. This is specially interesting for producers of infused edibles who like to mention/track/indicate the exact amount of THC or CBD in milligrams contained in the weighed product on top of the scale. Don't know the mg per g of your butter or oil yet? The 🌿 INFUSION CALCULATOR key in the window of the metric units (left of its ADD key) calculates it (with the cannabis club settings on, it is a key of the 🌿 CANNABINOIDS window instead).
Each box of a metric unit also shows the share of the goods beside its name, exact to two decimals, e.g. THC 17.25 % (in a box on the left before the name: 17.25 % THC). Not in the counting mode: there the amount is per piece.
🥛 Weight <----> Volume
If you want to measure 100ml of Milk, Honey, Alcohol or Coconut-Oil - or if you would need the weight of your flour, nuts or sugar measured in Cups, Table- or Teaspoons... Then you normally first would need to know the density of the particular liquid and then calculate the weight in grams or ounces manually for every weighing. Well, with google and your calculator - but still... what a mess...
With an UP↑SCALE you simply select a liquid or food-product from the integrated database with the most common kitchen- and laboratory products,
or from the additional database with 8000+ food products. Or add an own value here, if your products density is not listed until now, no problem at all. (in g/ml or ml/g)
Save everything as a THING on the scale then, and you can re-use it later with one click!
Either way: One click, and the equivalent amount in milliliters, spoons or cups will be shown live with the weight! That's it - no guessing or messing around with measuring cups - you can just weight and log directly in volume now!
You can convert from all weighing units on the scale (gram, ounces, 1/8s etc.) directly to:
You want to weigh 2.5 US Cups of corn flour? Simply select it from the database, choose US Cups, and hit the button. The flours specific weight - in this case 0.62 g/ml - will be overtaken from the database automatically and you will see the amount of flour in US Cups live at the scale. Additionally you could now add a weighing goal for the US Cups, in our case 2.5 - and the scale will not only tell you how much you need to weigh (372g) - but also will show you a percent-weighing-bar! This indicate the amount you have on the scale in a percentage of your goal - easy to fill 97... 98... 99... 100 percent then!
💧 Hydration (doughs and breads)
With HYDRATION (DOUGH) 💧 in the UNITS list (THING / NAME card) you can add the liquid share of a product. Add it to the products of your doughs and save them as THINGS - then your logs and recipes show the total flour, the total liquids and the true hydration of the dough (water + milk etc. ÷ flour), and the baker's percentages refer to the total flour. There is one hydration per THING, and it is always shown in percent (not with the weight):
💧 Liquid
The share of liquid in the weight: water 100 %, milk about 87 %, eggs about 75 %, butter about 16 %.
🫙 Preferment
Levain (sourdough starter), poolish, biga, sponge... with its own hydration, like bakers say it: water ÷ flour, so 100 % = the same weight of water and flour, a stiff levain 50-60 %. The water in it counts to the liquids, the rest to the total flour.
🍞 Flour
Marks a product as flour (0 % liquid, it counts to the total flour). Flours without it are found by their names ("flour", "Mehl", "harina", "rye", "Dinkel"...). If no flour is found at all, the biggest ingredient without hydration counts as flour.
Products without this unit count with 0 % liquid (salt, yeast, sugar...). Recipes find the hydration by the same name as the THING (like the nutrition values). The hydration unit is saved on the scale like all units. The 🌡 DDT calculator and 🫙 starter feeding can be opened right below its input, too.
💯 Percent Weighing / Baker's Percentage
The percent-weighing-option enables you to define whats 100% - and for the next weightings you will see the percentage of that below your current weight.
This is practical for many uses and specially for bakers a great option! The so-called "baker's percentage" is a dough formula wherein the total weight of the flour represents 100%, and the weight of all other ingredients is calculated as a percentage of that total.
With our scale you not need to do any calculations if you want to follow recipes in bakers percentage - you just add your amount of flour as 100% reference weight - and all the rest works automatically! Just weigh your other ingredients step by step without any extra calculations! No place for mistakes - and more time for cake!
Also this option is very useful for fill-weighing because the percentage will be shown in a live-bar, with an extra scale for the last 10% to allow exact filling even at bigger amounts! And last but not least: You also can do percent-weighing based on a goal of an additional unit instead of the weight! For that check out the Fill-Weighing-Goals option!
Create your own recipes!
If you log/sum-up minimum 2 different things, then our scale will show the amounts of your mix also as a Bakers Percentage Recipe automatically! It was never so easy to create and use this type of recipes! How amazing is that?
🍉 Food Nutrition Values
Maybe you are on a diet - or you just want to know how many calories, carbohydrates, proteins, fat, saturated fat, cholesterol, dietary fiber, sugar and/or salt (sodium) is in the stuff you are eating every day - or in the recipe you are just mixing? Here you go! By the way: The nutrition values of your different weightings of different foods can be simply logged and summed together with the summing- and logging-function then - and our Traffic Light Rating system will show you how healthy your recipe is in that case! Additionally you have a couple of options how to select, insert, save or output this values, check this out:
Simply use the search-option to select one of more then 8000 food products with their most important nutrition values from our inbuilt database! With one click you can overtake all nutrition values and the name of the product to the scale - it could not be more simple!
The search finds all words in any order and with or without plural endings ("raw blackberry" also finds "Blackberries, raw"). The best matches are shown first and in bold.
📷 ++ Scan the barcode of a product
⭐ PRO: the barcode lookup, here and in the nutrition diary (the food database and your own values stay free) - gray in the free version; try it with the free trial (free and PRO).
Below the database search you can scan the barcode (EAN-13, EAN-8, UPC-A or UPC-E) of a packed food product with the camera of your mobile or computer (📷 button) - or just type the number below the barcode and click 🔎. The nutrition values per 100 g are taken from Open Food Facts, a free and open food database (data under the Open Database License, ODbL), and filled into the form for own values. Please check them with the package - then click the button to add the product to the scale (and save it as a THING if you need it more often).
Good to know:
This is the only function of the app that needs internet. Only the barcode number is sent to Open Food Facts - nothing else.
If a product is not in their database (yet), simply enter the values from the package. You can also add the product at Open Food Facts yourself - that helps everybody.
In the counting mode the app asks for the weight of one piece, because there the values are needed per piece.
The barcode is read directly by your browser if it can do that (most Android devices) - otherwise by our own barcode reader in the app. Hold the barcode straight, about 10-20 cm in front of the camera, with good light.
Add own values or foods
Its very easy and fast to add own food products with custom values to the scale! You simply can add one (or more) of the nutrition values directly live to the scale - but you also can add all the info for your most used food products easily, add a custom name, and save them as a THING on the scale. Next time you can load it directly from there with a click!
Salt or sodium? Choose what is printed on your package. The scale calculates with sodium, and like the EU food labelling rules we use: salt = sodium × 2.5.
🍞 ++ Show Bread- & Carbohydrate Units
Carbohydrates can be counted also as Bread Units (1 BU = 12g carbohydrate) or as Carbohydrate Units (1 CU = 10g carbohydrate). People with diabetes often use CU or BU instead of gram to calculate their insulin doses. If you have Carbohydrates on the display of this scale, you will be able to show them directly in CU or BU, and also the logging/summing will include the amount of that selected unit later! Like this you see live how many Carbohydrate-Units you have on the scale at the moment - and altogether in the pot! No calculations needed any more - even if you weigh in Ounces!
🚥 ++ Traffic light rating
If you log your weightings with nutrition value units - like fat, saturated fat, sugar and salt (sodium) - then you will find an indication for the "healthiness" of your composition. We are using the common food traffic light system to show you if the values are 🔴 high, 🟠 medium or 🟢 low and how much of the ingredients you got per 100 g - or per 100 ml for drinks. The limits are the UK front-of-pack criteria (Department of Health, 2016): food per 100 g - also when it is sold by volume, like a soup - fat 3 / 17.5 g, saturates 1.5 / 5 g, sugars 5 / 22.5 g, salt 0.3 / 1.5 g (🟢 up to the first value, 🔴 above the second); drinks per 100 ml fat 1.5 / 8.75 g, saturates 0.75 / 2.5 g, sugars 2.5 / 11.25 g, salt 0.3 / 0.75 g.
Food or drink: a log of a drink recipe (a cocktail, or a recipe of the kind "drinks") is rated as a drink, everything else as food. The switch 🍽 FOOD | 🥤 DRINK on the card of the log page changes it for this log (a milkshake weighed on the weighing page, for example). The ml of a drink come from the volume unit of a line, else from the density of the ingredient (its THING, the density list or the food database) - with an estimated density the card says so. Red per portion: a finished recipe with a portion over 100 g (food) or 150 ml (drinks) is 🔴 for a nutrient when one portion has more than 21 g fat, 6 g saturates, 27 g sugars or 1.8 g salt (drinks: 10.5 / 3 / 13.5 / 0.9 g) - whatever its value per 100. The rating of the diary is per 100 g of everything you ate that day.
🔣 Custom Units
You want to add a price in a strange currency which we do not have in our list? Or maybe you like to show the percentage of love, included in every ounce of your product? Or the amount of single Raisins in a piece of cake? No problem at all! You can add a own unit, a name for it and the amount - and for sure - you can add as many units you want! Btw. The amount can be added:
In proportion to the weight.
The amount can be added in proportion to the weight on the scale - for example as 12.0g per 1g. Or as 12.0g per 10g, or per 0.0023 gram - just as you like. You can preselect some common units (like 1 or 10) - and you can change you amount-per-gram value later at any time!
As a percentage.
The amount in percent is preselected and not need much explanation. Add the amount of your things in percent of the weight - and there you go.
As absolute weight.
If you select "as absolute weight", then you can insert the weight of one of your things - like in our example - the average weight of a raisin. Then the amount of raisin in your weighed goods will be shown directly under the weight! The same works for the points of a diet program (the grams of one point), counted in your nutrition diary.
🧾 Lot Number of a THING
With LOT NUMBER 🧾 in the UNITS list (THING / NAME card) you can give the goods on the scale a lot number (batch number): type it in or 📷 scan it from the package (a GS1 code with (10) works too). Up to 20 letters, digits and - . / _ + : ( ) - the size of a GS1 lot number, so it fits into every barcode. Save it with the THING (💾), and every time you load the THING its lot number is back on the screen.
🔍 IDENTITY (ENTERPRISE), below the lot number in the same window: the approved codes of this THING (GTIN / EAN / article number - type, 📷 scan or a hand scanner; of a GS1 code only its GTIN is kept; max. 5), the approved label names (what the label of the container says, max. 3) and the grade (e.g. USP / Ph. Eur.). Save them with the THING (💾). They are the same for every lot of the THING. The ⚕ COMPOUNDING ASSISTANT checks every container with them - they never go into the weighing log, the sums or the labels.
Weighing log: every line of the THING shows its lot (🧾).
Labels: the lot is a line of the price labels (on the label from the start, you can switch it off) and the placeholder [lot] ([charge], [lote]) in your own texts; the QR label of the THING has it too.
Recipes: if an ingredient of a recipe is a THING with a lot number, its lot field starts with it. Change it there for this run only - the THING keeps its lot number.
No value: the lot number is never added up and never used for prices, nutrition values or recipe values. The 📦 inventory stays per THING.
A THING has max. 20 units on the scale - the lot number counts as one of them.
Lot code on the label: the option 🧾 LOT CODE prints a small QR code of the lot in a frame with the word LOT and its number beside it - so it is never mixed up with the EAN-13 price barcode or the QR code of the THING. The code is a web link (up-scales.com/lot?nr=...): a normal QR scanner (any phone camera) opens a page with the lot number, the product, the date and your business, readable for your customers - the UP↑SCALE app reads the lot number from it and goes on in the app. The page only shows what is in the code: we keep no data of your lots.
Your business: set it once on the settings page: the first line of the company data in QUALITY RECORDS & USERS → Reports (clubs: the name of the club in the CCLUB 🥦 MODE → ⚙ CLUB SETTINGS). It is saved on the scale for every phone and computer. While there is none, the 🧾 box shows a warning with a button to the settings. It goes onto the page of the lot code and is a line of the price labels ([business], not on the label until you check it).
🧾 Lot store of the scale (PRO)
With PRO the scale keeps up to 420 lots: THING, lot number, who made it (with users), when, and an optional 🔒 internal note of up to 420 characters (origin, supplier, best before ...) - only for your own records: it is never printed on labels or shown on the page of the lot code. All phones and apps of the scale see the same lots.
Generator: a pattern like {T3}{YY}{MM}{DD}-{N3} - {T3} the first 3 letters of the THING, {YYYY} {YY} {MM} {DD} the date and one counter: {N3} per day or {C3} per THING (3 digits, more if needed). The scale counts, so two phones never get the same number. Or type in a completely own number. The same lot number twice for one THING is refused.
🔍 LOTS: a window with all lots of the THING (a chip with ✕ shows all of them) - a row each: 🟢 / ⚪ / ⛔, the lot number, the THING, the date and the use-by date, newest first, and ⤓ at the end. Typing in the search field filters the list at once, 🔍 (or Enter) asks the scale for exactly that number (not there: a red line under the field), or 📷 scan it. Tap a lot for its card (state, date, who, number, its QR code, use by, price, lab values and the internal note) with ← back to the list; ⤓ in the list or ➜ THING (the main key of the card) loads it with this lot. 🟢 ON SALE is a key of the card, the rest (🏷 LABEL first, the reports, the package label, ⛔ LOCK, ✕ DELETE) is behind ⋯. Loaded this way, the lot is the one on the page (log, labels) - it is not saved with the THING, so you can sell the rest of an older lot.
THINGS scanner: scan a lot number (or the lot code of a label) instead of the THING code - the scale finds the lot and the app loads its THING (if it still exists) with this lot.
Straight into the right mode: a lot (or batch) of the ROASTERY ☕ MODE, the CCLUB 🥦 MODE or the DISPENSARY 🏪 MODE scanned with any of the app's scanners (also the lot code of a label, or the Metrc tag of a package) starts that mode and loads its THING with this lot - ready to book. The app tells them apart by the lines of the lot (roast, blend and green coffee lines: roastery; the harvest line: club; the Metrc lines: dispensary); if a lot has none of them and only one of these modes is available, it is that one. A scanned text of 24 letters and digits is tried as a Metrc tag first; in the DISPENSARY MODE, the lot list also has 📷 METRC TAG. Modes that are not available are never started; other lots work as before.
📥 IMPORT a lot list (PRO, the button of the lot list): a CSV, TSV or JSON file with your lots - the batch list of a supplier or of your own software, or the package list of Metrc. The app finds the columns by their names (lot, THING, quantity, unit, use-by date, THC, CBD, price, text) - check them or pick the right column; the choice is remembered for files with the same header. The THING column names the THING of each lot (a Metrc package list links each item to a THING first - remembered). The preview shows the first 40 rows (a row with an error in red; a lot number longer than 20 characters is shown as an error) and how many places the lot store has left. The lots are made one by one with their use-by date, lab values, price and text - with a progress line, and at the end what was created, updated, skipped or failed.
Several lots on sale at the same time: a THING can have many lots in the store - every lot that is 🟢 ON SALE (a new lot is) is offered when you load the THING: with more than one, the 🧾 line on the screen becomes a select (★ = the lot saved with the THING; without one the newest lot is chosen). The weighing log and the labels take the lot you chose. Set a sold out lot to ⚪ NOT ON SALE on its card. In the recipes, the lot fields suggest all lots of an ingredient that are on sale.
⛔ LOCK a lot (blocked or recalled): it stays in the store and shows as locked everywhere. ✕ DELETE follows the deleting protection of the ADMIN.
💰 Price of a lot: every lot can have its own price (on its card, per weight unit of the app, in the currency of the sales log) - saved on the scale, so every phone knows it. While the lot is chosen, its price replaces the price of the THING on the screen, on the labels and in the sales modes - 🛍️ sales, CCLUB 🥦 and ROASTERY ☕ (marked 🧾); the THING keeps its own price. Empty: the price of the THING is used. A price saved in another currency is not used.
⌛ Use by: every lot can have a use-by date (on its card) - saved on the scale, so every phone knows it. The lot list marks it ⌛ within 30 days and ⛔ when it is past; the lot box of the weighing page, the lot fields of the recipes and the labels ([expiry], and the lot label) show it, and the page of the lot code too. A ∑ or a booking with an expired lot asks first; a sale of an expired lot (🛍️ sales mode, the dispensing of the CCLUB 🥦 mode, the roastery and the dispensary) is refused: BOOK NOW stays gray with the reason when the app already sees the date is past, and the UP↑SCALE refuses it too (by its own clock; the use-by day itself still sells) - destroying or writing off an expired lot still works; the ⚕ compounding assistant does not start with one. A lot typed in without the lot store keeps its date in this browser.
🧪 Lab values (cannabis): a lot can have its own THC / CBD in % from its lab report (on its card: 🧪 + LAB VALUES - open at once in the CCLUB 🥦 mode, for a lot that has values and while its THING with a THC / CBD unit is on the page). While the lot is chosen, they replace the THC / CBD of the THING (more in the club mode). Lots of a roastery, a bakery or a shop never show them: the lists, the CSV and the PDF have THC / CBD columns only in the club mode or when a lot has a lab value.
Lot numbers from the generator never collide: if a number is taken already for the THING (typed in by hand or brought from another scale), the scale takes the next free one.
All lots at a glance: the 🗂 RECORDS page, tab 🧾 LOTS, lists every lot of the store in small text (number, THING, date, status - and THC / CBD where cannabis is weighed) - click one for its card. ⤓ EXPORT ALL LOTS saves them as a plain CSV file (with ENTERPRISE & COMPLIANCE sealed, as CSV or JSON in your own format), and with ENTERPRISE & COMPLIANCE 📄 PDF OF ALL LOTS makes a sealed PDF with your company header, the report number and the integrity code - both with the user and the internal note.
To another scale:back up or ⇄ sync your THINGS - all their units and lot numbers go with them, and you are asked once whether the lots of the store should go too.
Full? When all 420 places are used, new lots are refused - nothing is ever overwritten, because traceability comes first. From 378 lots the app warns you.
With the ENTERPRISE & COMPLIANCE edition every new, changed, locked or deleted lot goes into the activity log.
🫙 Container / Jars
Not with an RS-232 scale in EXACT WEIGHT: there the app never changes the weight of the verified scale - tare the container on the scale.
Here you can subtract a know weight. For example: If your container, bowl, bag or jar is already filled with your weighing goods and you don't like to refill everything to tare the scale with the empty container. Here you simply can subtract the weight of your container from the total weight on the scale.
Tip: Best to add your most used kitchen containers/bowls once with this option and save them as a THING to the scale. Then you are able to reload them later with only one click and you don't have to remember their weights any more!
Extra-Tip: You can add your storage tupperware/containers for your different foods/ingredients as a THING including all additional units like prices, nutrition values etc. Like that you simply can load a product, together with your units and its container-weight with just one click. Just take the tupper from the kitchen cabinet, place it on the scale and see how much is left in there! For a whole shelf or bar, use the 📦 inventory.
If you use the Takeaway Mode now, you simply can weigh, count and log what you are taking from the container - without the need to remember or calculate anything!
🥅 Fill-Weighing Goals for Units
If you add a unit to the screen, for example 3.57$ per gram, then you maybe want to know how much would be the weight for 420.00$ now - for example to fill a bottle exactly with that amount.
Tap the card of the unit (the faint 🥅 in its corner says so): a small window opens right below it - 🥅 Weigh up to [420] $ · SET. While you type, a line shows what the goal is as a weight ("= 117.65 g of Oats (at 3.5700 $/g)"). SET (or Enter) sets the goal: the card shows the percent of the goal ("16 % of 420 $") and a bar, and the 🥅 in its corner is full. Up to 90 % the bar fills in the ink colour; from 90 % on, the last 10 % stretch over the whole bar (to fill the last bit exactly) in the colours of the weighing: orange below 98 %, yellow up to the goal, green on the goal (within the PERFECT range of the check weighing you set up in this app, at least half a digit), red over it. One goal at a time: a new goal replaces the old one. Tap the card again: REMOVE GOAL removes it, ✕ (or Esc) closes the window. Percent and hydration cards have no goal. This is also very useful for manual fill-weighing by the way...
➕ Virtual Weights
Add a "virtual weight" to the scale - or remove it! Where: tap the weight at the top of the weighing or counting page - the field VIRTUAL WEIGHT with ➕ is at the end of the window that opens (Enter = ➕); the virtual weight then shows in red below the weight, ✕ removes it. This can be helpful for many things (not with an RS-232 scale in EXACT WEIGHT and not in the strict audit mode: there it is switched off - the window says so, its i tells why -, a virtual weight is no weighing result):
Check some values
You want to know how many calories would be in 3.5 oz of some food? Or how many THC is in some amount of dough? Easy! Just select a product from our food-database (or load your pre-saved THING) or insert the nutrition values from the package - and then add your amount as virtual weight to the scale!
Create your recipes
Now you also can use the (∑ LOG IT) Summing-Up / Logging functions! Like this you can create your recipes and check how healthy they are - without actualling weighing something! Just add the names and the "virtual weights" for each product to log the amounts together with all units and conversions! Now you also can use the 🧮 Servings Calculator, to calculate all portions to the details and to create your labels!
Or just use it
to remove some weight, because you forgot to tare... again... ;-)
Recipe 🧑🍳 Mode
The Recipe 🧑🍳 Mode
Save your recipes on the scale, weigh all ingredients step by step and up- or downscale your recipes on the fly without any calculations needed! The scale will do just everything for you - just enter your desired weight or amount of portions - and the Recipe Mode will guide you through all ingredients one after next! And if you add too much of something - a click on the button can re-calculate all amounts on the fly! This amazing mode will not only save you a lots of work, it also will make your products more consitant and safe!
The recipe mode enables you to create recipes consisting of 2 or more ingredients including their amounts in proportion to the total weight. After selecting one of the recipes and inserting the desired weight to produce, the recipe mode will guide through the weighing process for all ingredients.
The start page of the mode (RECIPES) has one card 🧑🍳 YOUR RECIPES (ℹ️ right behind its title: what the mode can do) with one row each - 📋 ALL RECIPES (load, edit, delete), 📷 SCAN A RECIPE LABEL, 🌐 IMPORT FROM A WEBSITE and 📖 FROM A COOKBOOK. Below it three cards that fold (a click on the title opens or closes them, your browser remembers it; ℹ️ explains each): 🥗 NUTRITION VALUES (how many THINGS with values the app knows, 📥 LOAD VALUES OF ALL THINGS), 🍞 BAKER'S TOOLS, 🌿 INFUSION CALCULATOR and 🍦 ICE CREAM & SWEETS (with 🧪 BRIX & SYRUP). The key at the bottom right is + NEW RECIPE. Every recipe page has its keys in the row at the bottom of the screen: the main action at the right end, more behind ⋯; the ✕ in the head goes back.
Creating Recipes
The page of a new (or edited) recipe has its title in the head of the page and four cards: 📝 BASIC DATA (food or cocktail, the product for the 🍦 / 🍬 values, name, weight of one portion), 📄 short instructions (optional - folded while empty), 🥣 INGREDIENTS (the values as baker's percent, total percent or weight, one card per ingredient, + ADD AN INGREDIENT, one tolerance for all) and SAVE (save it on the scale - or not). The main key at the bottom right follows your choice: START, START + SAVE - or for an edited recipe SAVE CHANGES, START + SAVE or START (NOT SAVED). The ℹ️ right behind the title of a card explains it. When something is not right (fewer than 2 ingredients, a sum that is not 100 %, a row with an amount but no name, a row above 100 % in total percent, a missing density), a window names all problems and the rows in question get a red edge. Leaving the form with ✕ after a change asks first - what you typed is not kept.
To create a new recipe, open the add-a-new-recipe page and simply add:
Recipes name and description
...to find it later again. Something like "My Great Infused 100MG cookies", for example. Also add a description for the recipe if you like (max. 2048 characters)
Portion size
Optionally you can add the size of a portion/serving in your selected weight unit (like Grams). If you do so, you later can directly insert the amounts of portions you want to make and not need to calculate the full weight of everything any more!
And then, for each ingredient:
A name
...like Butter, Sugar, Cannabis Oil etc...
...or 🔗 a THING of the scale: the selection below the name lists your THINGS, and choosing one gives the ingredient the name of that THING. An ingredient with the same name as a THING is connected with it (its name field turns green): the 🥗 nutrition values of the recipe come from that THING. Choose "no THING" to type an own name. If a name is only similar to a THING (like "Butter, salted" and "Butter"), a 🔗 Butter? button suggests it - one click connects it. The app knows the list of THINGS of each scale only while it runs (nothing is saved in your browser), from every THINGS list read from it (the THINGS menu, a backup, the recipe menu). If it was not read from this scale yet, the form reads it by itself when it opens (once per app start) - the form stays as it is. Without a connected scale the selections show the THINGS with values loaded so far.
...or 🔎 the food database: an ingredient without a THING gets a 🔎 food database button beside its THING selection. Search and click a food to link the ingredient with it (an ingredient without a name gets the name of the food - so you can also start a new recipe right from the database): its 🥗 nutrition values come from there. The link is remembered in your browser for all recipes (by the name of the ingredient), ✕ removes it again.
The amount
...you can enter in 3 different ways!
1.) Bakers Percentage: Here the main ingredient (like the flour) will be added with 100%, and all other ingredients in relation to that. An ingredient can also have more than 100%, like the water in a bread dough with 110% hydration.
2.) Total Percentage: Just add the percentage of the ingredients as total percentange of the recipe. (All ingredients together have to be 100% then!)
3.) Absolute Weight: Just add the total amounts in a weight unit like Gram. All ingredients have to be added in the same weight unit in that case and our program will translate that to percentages later.
🥄 Cups and spoons: with the absolute weight, every ingredient gets a unit selection below its amount: the weight unit - or a measure of a cookbook: US cup (240 ml), metric cup (250 ml: Australia, Canada, New Zealand...), UK imperial cup (284 ml: old British books), Japanese cup (200 ml), tablespoon (15 ml), dessertspoon (10 ml), teaspoon (5 ml), US / UK fluid ounces or ml. Type the amount as in the book - fractions work: 1 1/2, 1-1/2, 1½, ¾, 1,5. A volume needs the density of the ingredient (g per ml): it gets filled in below the amount - from the THING with the same name (its volume unit), else from the density list of the volume units (the list you also choose from for a THING; common ingredients in German, Spanish, Japanese, Chinese and Korean find it too: Mehl, harina, 小麦粉...), else from the food database (the food the ingredient is 🔎 linked with, or a clear hit of a search). The longest name counts ("brown sugar" before "sugar"); another food or a word like "condensed" next to it (rice milk, almond flour, peanut butter, condensed milk) is not taken as that food - then only the food database helps, or you type it. ≈ marks a value that is not sure - hover over it to see where it comes from, and put in the right one if you know it. Without a density the recipe can't be saved. Next to the density you see the weight it makes ("= 127 g"). The scale saves the recipe in baker's percentages as always; the units and densities stay in this browser (saved with the recipe name) and come back when you load or edit the recipe here. The line ("1 ½ cups flour") goes into the short info of an ingredient that has none, so another phone or browser shows it too (there the recipe is weighed in your weight unit).
A tolerance
If you want to add some tolerances for your weighing, here you go. Just add the tolerance in percent or keep it at 0! This tolerances have influence in when the weighing bar becomes "green" and the ingredient will be marked as "ready". For example, if you need to weigh 100.0g of Butter, and if you have added a tolerance of 2%, then the butter will be marked as "ok" between 98 and 102 grams!
A short info
...add some short instructions for every ingredient! (max. 125 chars)
Additionally
...you can add some info if the ingredient have to be added in the same- or in another bowl and about the action byself, like Add, Pour, Fill-Up, Garnish With etc...
The order
Every ingredient is shown as a card: its action (short, e.g. STIR), name and amount on top, below them the 🔗 THING selection with the 🔗 suggestion and the 🔎 food database beside it. ⚙ more opens the rest in the same card - the action moves there (in full, e.g. STIR IN) beside the amount type and tolerance, the bowl, the short instruction - and ▲ ▼ to move it one place up or down and 🗑 to remove it.
...and if the amount is the given amount (ready within its tolerance), a minimum (ready from the amount minus its tolerance on, more is ok too) or a maximum (ready as soon as something is added, but not more than the amount plus its tolerance).
Import Recipes
Import recipes from calculators or 3rd-party websites with 1 click!! For example, you can use the free 420BASE.net Edibles Recipe Calc-O-Lator to create your recipes, and then just export them with one click to your scale from there! Try it out, it could not be more easy!
🌐 IMPORT FROM A WEBSITE: almost all recipe websites (BBC Good Food, Cookpad, Chefkoch, food blogs with a recipe plugin...) have their recipe as data in the page for search engines (schema.org/Recipe, as JSON-LD or microdata). Open 🌐 in the recipe menu and paste the 🔗 link of the recipe: the app loads the page directly and reads that data. This works only with websites that allow other apps to load their pages (e.g. BBC Good Food). If a website blocks it, the dialog shows the other ways as tabs (also with "Other ways" below the link field; the third one, 📖 COOKBOOK, is below):
🔖 BOOKMARK (works on every website, 1 click): drag the button 🔖 UP↑SCALE RECIPE to the bookmarks bar of your browser once (on a phone: copy its code and paste it as the address of a new bookmark). Then open a recipe on any website and click that bookmark: it reads the recipe data on the page and opens the app in a new tab with the recipe. Connect your scale (and unlock it, if it has a password) and the recipe form opens. No copying, and nothing goes through a server: the recipe travels in the # part of the link, which browsers never send to a server. (The offline app can't be opened by a website, so the bookmark made there opens the online app.)
📋 PAGE SOURCE: put view-source: in front of the address of the recipe (or right-click > View page source), select all, copy, and paste it in the field - the recipe data alone (JSON-LD) works too.
Nothing goes through our server.
📖 FROM A COOKBOOK: a recipe from a printed, old or foreign cookbook - in cups and spoons. Click 📖 FROM A COOKBOOK in the recipe menu (or the tab 📖 COOKBOOK in the 🌐 window) and type or paste the ingredient list as it is in the book, one ingredient per line: "2 cups flour", "1 ½ tsp salt", "1 EL Zucker", "250 ml de leche", "大さじ2 砂糖", "3 eggs". Lines in English, German, Spanish, Japanese, Chinese and Korean work, with fractions and ranges; bullets, empty lines and headings like "For the dough:" are left out. Give it a name and the servings if you like, and choose the cups in this book: as written (cup 240 ml, Tasse / taza 250 ml, カップ / 컵 200 ml), or one size for every cup - US cup 240 ml, metric cup 250 ml (Australia, Canada, New Zealand...), imperial cup 284 ml (old British books: fl oz and pints become imperial too) or Japanese cup 200 ml. Then TO THE RECIPE FORM: the list goes the same way as a recipe of a website (below) - check it in the form, add the steps to the description, start or save it. Only the ingredients go into the field, not the steps.
Name, servings, ingredients and the steps (into the description, max. 2048 characters with the source address) are taken over, max. 50 ingredients. Recipes of the category cocktail/drink become drinks.
All amounts get a weight: weights directly (also "1 cup (120 g)" and "2 x 400 g"), cups, spoons, ml, oz, lb, pinches... with the density of the ingredient - the same as in the recipe form: from your THING, the density list of the volume units or the food database (unknown ones like water, marked ≈) - eggs, onions, garlic cloves, lemons and some more by their usual weight, a stick of butter as 113 g. Fractions (½, 1 1/2) and ranges (2-3: the middle) work, in English, German and Spanish (and Japanese, Chinese and Korean lines like "砂糖 大さじ2"). Old US abbreviations too: c. or C = cup, T. / Tbl = tablespoon, t. = teaspoon (capital T and small t count), and words like heaped, level, scant, gestrichen, gehäuft, colmada or rasa (the amount stays as written). Two parts like "1 cup + 2 tbsp" or "1 lb 2 oz" are added up; a second part that can't be added ("1 cup + 20 g") gives ⚠.
Cups and spoons stay cups and spoons: an ingredient in cups (US, metric, imperial, Japanese), tablespoons, dessertspoons (10 ml), teaspoons, fl oz or ml keeps that unit and its amount in the form ("1 cup + 2 tbsp": 1 ⅛ cups), with the density used next to it (≈ = a guess: like water when the ingredient is unknown - put in the right one) and the weight from it. "1 cup (120 g) flour" keeps the cup with the density of that line. dl, cl, l and an imperial gill (142 ml) stay as ml; pints, quarts, dashes and pinches become a weight only.
Lines without a weight ("salt to taste") get ⚠ and a minimum of 0.01 - set them in the form. The form opens with the other amounts in your weight unit ("Input as": g, oz...), saving makes the baker's percentages. Every ingredient keeps its original line as instruction. The portion weight is the total weight divided by the servings.
Check everything in the form before you save it, like with every import.
The imported recipe opens in the form, so you can check and change it before you start (and save) it. Before that, the app reads the list of THINGS from the scale and connects the ingredients with your THINGS: an ingredient with the same name (upper/lower case, accents and spaces don't matter) gets connected and takes the exact name of the THING. Ingredients with a similar name get a 🔗 …? suggestion to connect them with one click. Above the form you see how many ingredients are connected. The same happens for cocktails imported from the cocktail database.
Edit Recipes ✏️
Change a recipe that is saved on the scale: click ALL RECIPES in the recipe menu (or ALL DRINKS in the cocktail mode) and ✏️ beside the recipe - or load it as usual and click ✏️ EDIT in the key row of its start page. The recipe opens in the form with everything it has: name, type, portion size, description and all ingredients with their amounts (as baker's percentages), tolerances, actions, bowls, minimum/maximum and instructions. Ingredients connected with a THING show it in their 🔗 selection - choose an other THING to connect the ingredient with it (it takes the name of that THING). Move or remove ingredients with ⚙ more ▲ ▼ 🗑, add new ones, change the input unit for the amounts if you like. A recipe with 🥄 cups and spoons saved in this browser opens with the weights and its cups and spoons as you typed them.
Overwrite
...saves the changes in the same place on the scale (the name may stay the same). Paused runs of this recipe get deleted, because their weights belong to the old ingredients.
Save as a new recipe
...keeps the old recipe and saves the changed one next to it (with a number like "Cookies #2" if the name is the same).
Don't save
...starts the changed recipe once ("START (NOT SAVED)"): the changes are not saved on the scale and get lost afterwards.
Weighing a Recipe
If you select one of your recipes (or start it directly from the input form) you get its start page - the name of the recipe in the head. The card AMOUNT comes first: with a portion size you see one portion and type the portions you want to make (and an extra %), the total amount follows; without one, type the total amount. The card INGREDIENTS shows the needed amount of every ingredient at the right (the % of the recipe until a total is typed) - check if you have everything - and START WEIGHING at the bottom right starts the program (no total: a red line under the field says so). ✏️ EDIT next to it opens the form, ⋯ holds the 🏷 QR CODE LABEL. The cards in order: AMOUNT, 📦 I ONLY HAVE … (folds), INGREDIENTS, 🍞 DOUGH (doughs only), 🍦 ICE CREAM / 🍬 SUGAR VALUES (recipes with a product), 🥗 nutrition values, ⎙ PORTIONS & LABELS and 💰 COSTS & MARGIN. A new or changed recipe that was saved tells it in a line at the bottom of the screen.
📦 I only have ...
Not enough of an ingredient at home? Click 📦 I ONLY HAVE ... below the total amount, insert how much you have of an ingredient (+ ONE MORE INGREDIENT for more of them) - and the app shows the biggest amount you can make with it, and which ingredient runs out first (for a recipe with a portion size: as many whole portions). ✓ USE IT takes it as the total amount (or the number of portions).
Step-By-Step
The program will start with the first ingredient: below the weight you see its action (ADD, POUR …) small and its name big, a new bowl or glass, its instruction and the chips ⏱ (and 🌡 WATER TEMPERATURE for water), then a bar with how much of the goal is on the scale. The result box below it shows what is still missing: yellow (ADD - or ADD MINIMAL / MAXIMAL) while there is too little, green READY! within the tolerance (with a short beep if 🔊 is on - the key in the corner of the box switches it; a minimum amount stays READY! also above it), red REMOVE when there is too much. The keys at the bottom: TARE (left), ◀ LAST, NEXT ▶ and LOCK & TARE (right, with the new total below it): it takes the weight on the scale for this ingredient, tares and asks whether to go on to the next one. ⋯ holds ⏸ PAUSE & REPEAT LATER (saved recipes), 🧮 RE-CALCULATE, 📄 INSTRUCTIONS (the short instructions of the recipe, shown at the start) and the sound. The ✕ in the head ends the recipe - after something was weighed it asks first, then the weighed ingredients become a log.
🥄 Cups and spoons
An ingredient typed in cups or spoons shows its goal in that unit next to the weight ("goal 1 ½ cups · 180 g") and how much is on the scale so far in that unit ("so far 1 ¼ cups"); what is still missing also shows in the unit ("≈ 2 tbsp"), and the list of all ingredients has the goal in the unit below the weight. When you make more or less (portions, total amount, 📦 I ONLY HAVE), the start page and the weighing show the scaled amounts - in kitchen fractions (eighths and thirds: ⅛ ¼ ⅓ ⅜ ½ ⅝ ⅔ ¾ ⅞; ≈ in front when the fraction is more than 5 % off; less than ¼ cup in tablespoons or teaspoons). Once an ingredient is ready (within its tolerance), "so far" shows its goal. What is on the scale is rounded down until the goal is reached, and the cups line of the short info is left out while the goal is scaled (the scaled amount shows above it). The amounts come from the weight and the density: you still weigh, no measuring cup needed. When every ingredient has a density, the total amount also shows the volume of the whole recipe.
Within the running program you also have a couple of options:
Progress
The card PROGRESS shows all the ingredients of your recipe at any time (↯ marks the one of now), with how much you have already weighed of the goal and how far away you are from it - the edge of a row is orange while there is too little, green when it is ready and red when there is too much. The last row is the total with ✅ / ❌. A tap on a row jumps to that ingredient.
Weighing Parts
You have to weigh 5 kg flour but your scale only have 3kg capacity? Or your pot is already nearly full and the next ingredient would exeed the maximum weight of your scale? No problem at all. Simply press LOCK & TARE with what you have on the scale already and answer STAY when it asks whether to go on to the next ingredient, get a new, empty bowl on the scale, press TARE and keep on weighing! Things will get summed up automatically!
Jump
You can jump between the ingredients freely, no problem at all! Weigh a part of the sugar, then add the flour, go back to add some more sugar - all easy!
Recalculation On-The-Fly
Do you have added too much oil accidentally?? And you can't get it out from the pot any more? No worries! Simply re-calculate the recipe: ⋯ > 🧮 RE-CALCULATE, choose the ingredient it is based on (only the ones with too much or too little can be chosen) and confirm. The needed extra amounts for all other ingredients will be calculated automatically based on the new amount of oil! Now just add whats needed and produce a bit more - but everything in the right proportions!
⏱ Timers
Times in the short instruction of an ingredient (like "stir for 3 minutes", "rest 1-2 h", "30 min ruhen lassen") become buttons below it: one click starts the timer (for a range like 20-25 min the shorter time). ⏱ TIMER starts an own one (in minutes). Running timers are shown as chips in a line at the bottom of the screen - also while you weigh on or switch to another mode - each with its ✕ to cancel it. At the end the timer turns red, beeps for a minute (always, also with the sounds switched off 🔇), vibrates the phone and shows a notification (if you allowed it). The timers run in the app: closing the app ends them.
🧾 Lot numbers
Below PROGRESS: the card 🧾 LOT NUMBERS (it folds; the number of lots typed in behind its title) has a field for every ingredient - type in the lot (batch) number of the package you use, or scan it with 📷 (QR code or barcode); the use-by date of a known lot shows below it. They are kept until the recipe is done (also while it is paused), go into the weighing log and into the batch report - and the last 5 lot numbers of an ingredient are suggested next time. An ingredient that is a THING with a lot number starts with that lot (change it for this run only, the THING keeps its own).
🧾 Batch report
When a recipe is done, the weighing log gets the button 🧾 BATCH REPORT: a report of this batch with the start and end time, the scale and - for every ingredient - its lot number, the target amount (and tolerance), the weighed amount, the deviation in % and ✓ if it is within its tolerance (judged by the UP↑SCALE), the totals, the portions and the nutrition values. Add a batch number (date and time are preset), who made it, notes and your company (the header of all reports, the same as in the lot report), then ⎙ PRINT it (or save it as PDF in the print dialog) or ⇩ SAVE it as an HTML file - the window shows the report as a paper page while you type (✕ or Esc asks first when something is typed). Good for small producers who have to show what went into a batch.
🍞 Baker's tools
For bread, pizza and all doughs: the recipe menu has the box 🍞 BAKER'S TOOLS with two calculators and a journal - and the recipe pages link to them where they help. The three open in one window with the tabs 🌡 DDT | 🫙 STARTER | 📒 JOURNAL under its title (ℹ️ behind the title explains the tool, ✕ or Esc closes it): °C | °F at the top, the cards below, the keys at the bottom - ⚖ TARE at the left, the main key of the tool at the right. A weight field has the key ⤓ joined to it: it shows the live weight on the scale (pale with ⏳ while it is not stable) and takes it with one tap.
🍞 Dough & hydration
The start page of a dough recipe (with flour or a preferment in it) shows 🍞 DOUGH & HYDRATION below the ingredients: every ingredient as baker's % of the total flour (also the flour in preferments), the total flour and liquids (in g, once you set the total amount) and the true hydration - and how much of the flour is prefermented. It uses the 💧 hydration of the THINGS with the same names as the ingredients; without it, the box tells you how to add it. The finished recipe log shows the same hydration of the dough.
🌡 DDT calculator (desired dough temperature)
The water temperature decides how warm the dough is after mixing - and how fast it ferments. Add your desired dough temperature, the flour and room temperature (and the preferment temperature, if you use one) and how you mix: the app shows the right water temperature with the standard formula of professional bakers: water = DDT × 3 (× 4 with a preferment) − flour − room − (preferment) − friction. The friction factor has presets for mixing by hand, a stand mixer and a spiral mixer (in °F like in most baking books: 6, 26, 22) - or measure your own: mix a dough, insert the water temperature you used and the dough temperature right after mixing, and CALCULATE AND SAVE. Temperatures in °C (standard) or °F. Your last inputs are kept in this browser. While running a recipe, the step of the water gets the button 🌡 WATER TEMPERATURE (DDT).
🫙 Starter feeding
Weigh the rest of your sourdough starter (⤓ 12.3 g at its field shows the live weight and takes it - in another mode the app asks to switch the scale to the weighing mode first) or type it, choose the feed ratio starter : flour : water (presets 1:1:1, 1:2:2, 1:3:3, 1:5:5, 1:10:10, 1:5:2.5 or your own) - and get the flour and water to add, the fresh starter in total and its hydration. Need a certain amount? Add the fresh starter you need: only the needed part gets fed, the rest is shown as discard. ⏱ START THE TIMER starts a timer to the peak of the starter - preset with a rough estimate from the ratio and the room temperature (5 h for 1:1:1 at 24 °C), change it to what you know from your starter. It runs like the ⏱ recipe timers in the line at the bottom of the screen.
📒 Starter journal
Next to 🫙 STARTER FEEDING: write down every feed of your sourdough starter - the ratio (1 : flour : water, preset with the one of the feeding), how much starter you kept (⤓ works here, too), the hours it took to the peak (optional), the room temperature and a note - and 💾 SAVE THE ENTRY. The card FEEDS shows all feeds, newest first (up to 365; ✕ at the end of a row deletes it after a question), and the average peak time of your last feeds. Closing the window while a new entry is typed asks first. Next feed in (hours) sets the ⏰ NEXT FEED (✕ at the end of its countdown clears it, after a question): the window and the baker's tools box in the recipe menu count down to it, and with 48 hours or less it also runs as a ⏱ timer in the line at the bottom of the screen (type e.g. 168 for a starter in the fridge). The app can only ring while it is open - if you come back later it shows the feed as overdue. The journal is saved in this browser only, not on the scale.
Most edible calculators stop at a number on a screen - and you still guess with spoons and old notes. Here your scale weighs every step of the batch, and the result does not stay a number: it becomes a THING on your scale. From then on you weigh your infusion directly in mg THC (and every other cannabinoid), your recipes show the mg per portion, the servings calculator splits a batch into portions of exactly 5 or 10 mg, your labels print the exact cannabinoid content and the ingredients, and the nutrition diary tracks your daily intake. One tool from the flower to the labeled edible - for home cooks who want reliable doses, and for small edible makers who have to document them.
Make infused butter, oil or tinctures with a known strength - and use them in all your recipes. Open 🌿 INFUSION CALCULATOR in the recipe menu or below the input of the 🧪 metric units (with the cannabis club settings of the CCLUB 🥦 MODE on: at the end of the 🌿 CANNABINOIDS box). It is one window with a card per step (ℹ️ behind a title or a label explains it, ✕ or Esc closes it). Every weight field has the key ⤓ 12.3 g joined to it - the live weight on the scale (pale with ⏳ while it is not stable), one tap takes it into the field - and ⚖ TARE is at the left of the keys of the window, so you see and control what you weigh (in another mode the app asks to switch the scale to the weighing mode first). Every app start begins a new batch: no cannabis chosen, no lab values (only the solvent settings are kept). The card 💾 SAVE AS THING is shown from the start, the keys ⚖ TO WEIGHING and 💾 SAVE AS THING (the main key at the right) get usable as soon as the steps 1 to 4 are filled in (or only the 🔬 lab values of your infusion).
1. The cannabis
Choose what you infuse (nothing is chosen at the start): cannabis flower, trim, kief, hashish, aged hashish, rosin, distillate, extract - or other with an own name (e.g. "Bubble hash"). Choosing it prefills the THC row with rough average values (e.g. flower: THCA 18 %, THC 1 %) - always use the values of your lab label if you have them (values you typed into the THC row yourself are only replaced when you say so). Weigh it (or type its weight).
2. The lab label
Add the cannabinoids of the lab label, one row each: THC, CBD, CBG, CBN, CBC, THCV, CBDV or an own one (e.g. THCP) - with the % of its acid form (THCA, CBDA ...) and of the neutral form. + ONE MORE CANNABINOID adds a row, ✕ at its end removes it (asked first when it holds values). Total = neutral + acid × 0.877: decarbing turns THCA into THC, which is 12.3 % lighter (the same for CBDA → CBD ...). Lower the decarb factor if your decarb was not complete. Already decarboxylated cannabis (distillate, aged hashish ...): add its amounts directly as THC, CBD ... (the neutral form) - and both, if it is partly decarboxylated.
3. The solvent
Butter, ghee, coconut oil, MCT oil, olive oil, cocoa butter, glycerin, alcohol (tincture) or other (with an own name, e.g. "Hemp seed oil" - for the ingredients and the name of the THING: "Canna-Hemp seed oil, 4.11 mg/g") - each with a preset extraction efficiency (how much of the cannabinoids get into the solvent: rough values, 70-85 % for fats and alcohol, 50 % for glycerin). Change it if you know better, e.g. from a lab test of your butter. Weigh the solvent used too (before infusing): it is needed for the ingredients of your THING. For alcohol add its strength in % vol: the solvent gets it in its name ("Alcohol 96% vol"), the THING gets the unit Ethanol (the pure alcohol in it, on labels as % of the weight) and the spoons use the density of the mixture. Every solvent has its density (g/ml) preset from the food database of the app (butter 0.911, coconut oil 0.93, glycerin 1.20, alcohol by its % vol ...) - change it if you know better (↺ at the field takes the preset again). Amounts in: type or weigh the solvent and the infusion in your weight unit or in ml (the scale weight gets converted with the density). The density is used for the ml, the spoons and the 🥛 milliliter unit of the THING.
4. The strained infusion
Weigh the butter, oil or tincture after straining (the plant material keeps some of it): the 📊 RESULT shows the mg of every cannabinoid (and with more than one: the Total Cannabinoids) in the cannabis, in the infusion, per g, per teaspoon and per tablespoon - and the ingredients of the product (e.g. "Butter 91.7 %, Rosin 8.33 %", from the fat used and the cannabis).
🔬 Lab-tested infusion (optional)
Got your butter, oil or tincture tested by a lab? Add its value of every tested cannabinoid in mg/g or % (1 % = 10 mg/g; empty: the estimate). A lab value replaces the estimate of that cannabinoid: the RESULT marks it with 🔬, the THING, the name ("Cannabutter, 9.5 mg/g"), the dosing and the labels use it. With the cannabis and its label filled in too, the RESULT compares both and shows the real extraction efficiency of this batch (e.g. "THC: lab 9 mg/g, estimate 12.98 mg/g - real extraction efficiency of this batch: 49 %") - a good value for the efficiency of your next batches. With lab values of all cannabinoids, the steps 1 to 3 are optional: e.g. for a bought, tested oil, only the cannabinoids, their lab values and the weight are needed (the ingredients of the THING need the cannabis and the solvent used).
💾 Save as THING
Saves a THING on the scale with these units: every cannabinoid in mg per g (🧪 metric units), the Total Cannabinoids (with more than one) and the ingredients as weight units (the solvent and the cannabis in g per g - on its label as % of the weight and mg/g, e.g. "91.67 % (917 mg/g)") and a 🥛 Milliliter unit (weight <----> volume with the density: weigh it and see its ml, cups or spoons) - in this order: solvent, cannabis, (ethanol), milliliter, the cannabinoids, Total Cannabinoids. Then you weigh with it in the normal weighing mode directly in mg of every cannabinoid, and print labels with the exact cannabinoid content and the ingredients. It gets the name in the field above it: proposed like "Cannabutter, 12.3 mg/g" (it follows your values) - or type your own (↺ at the field brings the proposal back; without a name the key stays gray). An existing name asks whether to overwrite it or save it as a new THING. Use this name for the ingredient of your recipes: the recipe values, portions & labels, the logs and the diary then show the mg of THC per portion. ⚖ TO WEIGHING shows the values with the weight without saving. Both need a connected scale.
5. How much infusion for my recipe? (optional)
Plan your edible (this card is folded until you open it): add the THC you want per portion (without THC: the first cannabinoid - the label of the field names it), the portions and the fat (butter, oil ...) your recipe needs in total. The app tells you how much of your finished infusion (this batch) goes into the recipe - and how much plain, not infused fat fills up the rest of its fat (e.g. "14.6 g of your finished infusion go into the recipe + 85.4 g plain butter"). Without a wanted THC: the THC per portion if all the fat of the recipe is your infusion.
⚠ Only an estimate: the real strength depends on the flower, the decarb and the infusion! For a known strength, get your infusion tested by a lab and add its values. ⚠
⭐ PRO:💾 Save as THING and using the infusion for weighing (the calculation, the lab values and the dosing of step 5 stay free) - marked 🔒 PRO in the free version (a tap tells why); try it with the free trial (free and PRO).
🍦 Ice cream & 🍬 sweets: the values of your recipes
For ice cream and gelato makers, jam, syrup and candy makers: give a recipe its product, and its start page shows the values that decide whether the batch works - with a light 🟢 🟠 🔴 for the usual range of the product. The finished batch keeps them, calculated from what you really weighed.
The product is a field of 📝 BASIC DATA in the recipe form: ice cream, premium ice cream, gelato, sorbet, plant-based frozen dessert, jam & jelly, syrup, caramel & fudge, hard candy, ganache & filling, gummies & jelly sweets - or none. It is saved with the recipe on the scale, so every phone or computer that loads the recipe shows the same values. The list of the recipes shows its sign after the name.
🍦 ICE CREAM VALUES
For ice cream, premium ice cream, gelato, sorbet and plant-based desserts (below the ingredients): fat, milk solids non-fat (MSNF: protein, milk sugar and minerals of the milk), sugars (without lactose), sugar alcohols and alcohol (if there are any) and the total solids in % of the mix; the sweetness (POD) and the anti-freezing power (PAC) of all sugars as sucrose equivalents per 100 g of mix (sucrose = 100: dextrose is 70 sweet but 190 anti-freezing, so it makes ice cream softer without making it sweeter); the freezing point of the mix, the serving temperature (where about two thirds of the water is ice) and the water frozen at −18 °C, from the freezing curve of the mix (the method of H. D. Goff, University of Guelph: the sugars as sucrose equivalents per 100 g of water, plus the milk salts); and the lactose per water (much of it can make ice cream sandy).
🍬 SUGAR VALUES
For sorbet, jam, syrup, caramel, hard candy, ganache and gummies: the °Brix of the mix (dissolved solids per 100 g of the liquid part), total solids and water, the water activity (an estimate of how well it keeps: lower is longer), the glucose : sucrose ratio (glucose syrup, glucose and fructose per sucrose: more keeps sugar from crystallizing) and for gummies the gelling agent and acid in %. With a target °Brix (preset for the product: jam 65, syrup 80, caramel 87, hard candy 99 - type your own) you get the end weight of the batch and the water to boil off - boiling down only takes water - the sugar stage of the target with its temperature and the water activity at the end. Gummies and hard candy: type the number of pieces (moulds) for the weight of one piece. 🧪 BOIL DOWN opens the Brix tool with this batch, for gummies also 🌿 INFUSION (the dose per piece is there).
Where the values come from
Every ingredient is listed with its source: ✓ THING - a THING with the same name that has composition units (weight units per g named Water, Sucrose, Glucose, Fructose, Lactose, Maltose, Milk solids non-fat, Glucose syrup solids DE 42, Erythritol, Gelatin, Citric acid ... in any language of the app; Water is needed, the fat comes from its nutrition value) -, 📘 the built-in technical ingredients by their name (dextrose, glucose syrup - with "DE 36", "DE 42" or "DE 62" in the name the nearest one -, dried glucose syrup, invert sugar syrup 70 %, maltodextrin, lactose, inulin, trehalose, erythritol, allulose, sorbitol, maltitol, isomalt, xylitol, glycerol, citric acid, gelatin, pectin, agar, salt, water), 🔎 the food database (the food you linked in the recipe form, or common names like milk, cream, sugar, honey, milk powder, egg yolk, butter, cocoa, strawberries in the languages of the app) - or ✕ no composition: those ingredients are named with their share, the values then count only the others. A food whose sugars are not known one by one (e.g. skim milk powder) is named as estimated.
The lights
🟢 in the published range of the product, 🟠 close to it, 🔴 outside; no light where no published range was found (e.g. the composition of gelato or plant-based desserts). The ranges: standard ice cream 10 - 12 % fat and 36 - 38 % total solids, premium 12 - 15 % and 38 - 40 %, MSNF 9 - 12 %, sweetness 14 - 15, sugars 8 - 20 %, PAC 20 - 25 (scoops well at −18 °C); gelato served at −12 to −11 °C; sorbet 30 - 35 % total solids, sweetness 27 - 28.5; jam at least 60 °Brix (EU directive 2001/113/EC: 60 % soluble dry matter; US 21 CFR 150.160: 65 %) and a water activity of 0.80 - 0.85; caramel and fudge 0.45 - 0.60, hard candy 0.25 - 0.40, gummies 0.50 - 0.75 with 5 - 10 % gelatin and 0.5 - 1.25 % acid. Sources: H. D. Goff (University of Guelph), J. Tobias (University of Illinois), W. M. A. Mullan (dairyscience.info), Tharp & Young, Tetra Pak Dairy Processing Handbook, L. Hobbs (Starch: Chemistry and Technology), the Exploratorium (sugar stages), Ergun, Lietha & Hartel 2010 (water activity of confections), FAO (Norrish equation), EU Regulation 231/2012 and 21 CFR 168 (technical ingredients), makers' data sheets.
The finished batch
The log of a finished recipe with a product shows the same values in a card 🍦 ICE CREAM VALUES or 🍬 SUGAR VALUES - from the grams you weighed, so a batch that went over or under shows its real values. They are in the text of the log too.
💾 THE BATCH AS A THING
Saves a new THING with the values per g of the recipe: its nutrition values and its composition (water, sugars, milk solids ...). Use it as an ingredient of other recipes - a base mix, a variegate, a sauce - and its values are known there.
🧪 BRIX & SYRUP
A window with four tabs (in the recipe menu, on the start page of the kitchen view and from the 🍬 card): BOIL DOWN - the batch at the start (g), its °Brix and the target give the end weight (and with the pot); weigh the empty pot once and then the pot with the batch while it boils: the °Brix now, the sugar stage and how much is still to boil off (or how far it went over). DILUTE - the water for a lower °Brix. ADD SUGAR - the sugar for a fruit purée or juice measured with a refractometer (e.g. a sorbet base). BLEND - how much of two syrups give the total at the target. Every weight field has ⤓ for the weight on the scale, ⚖ TARE is at the left of the keys. The fields are kept in this browser.
⚠ All values are estimates from the composition of the ingredients: °Brix is exact for sucrose solutions (other sugars, acids and fruit read a little different on a refractometer), the water activity is a calculation, not a measurement. Check a new recipe with a test batch. ⚠
⭐ PRO:💾 SAVE AS THING (the values, the lights and the Brix tool stay free) - marked 🔒 PRO in the free version; try it with the free trial.
🥗 Nutrition values of your recipes
When you start a recipe, the page with the total amount also shows 🥗 NUTRITION VALUES per 100 g, per portion (if the recipe has a portion size) and for the whole amount you are going to make - calculated from the shares of the ingredients. The values come from your THINGS: an ingredient gets the values of the THING with the same name (upper/lower case, accents and spaces don't matter: "Flour type 405" = "flour type 405"). The scale can save THINGS whose names differ only in upper/lower case (like "TestThing" and "Testthing") - then an ingredient takes the one with exactly its spelling.
Where the app knows your THINGS from
Straight from the scale - and only while the app runs: nothing is saved in your browser, closing the app forgets them. When you open the recipe menu, the app reads the values of all THINGS it doesn't know yet by itself - in the background, the scale stays in the recipe menu - the first time all of them, later only new or changed ones. Every THING you load on the page (weighing or counting mode) is added too, and so are all THINGS of a THINGS backup or a restore. Every new THINGS list read from the scale refreshes them: THINGS that are gone are dropped, and a THING you saved again gets read again. The recipe start page does the same for its own ingredients (e.g. a recipe loaded from another device or after the app was closed): it reads just the THINGS of these ingredients (in the background as well). 📥 LOAD VALUES OF ALL THINGS in the recipe menu reads all of them again.
No THING with this name?
Every ingredient is listed below the values with where its values come from (✓ THING, 🔎 food database or ✕ no values). This page only shows them: to link an ingredient with the built-in food database, click ✏️ EDIT THIS RECIPE (see Edit Recipes) and use 🔎 food database beside the THING selection of the ingredient (see the recipe form) - the link is remembered for all recipes. Ingredients without values are named in red together with their share of the recipe: the values are too low then.
Raw or processed?
These are the values of the raw ingredients. Cooking or baking changes the weight (water evaporates, rice and pasta take up water) - for the values of the processed food use 🍲 batch processing - or insert the loss in ⎙ PORTIONS & LABELS.
Save it as a THING
💾 SAVE AS THING in the block "SAVE THIS RECIPE AS THING" below the list of ingredients saves the values per g of the recipe (like in the table, raw) as a new THING with the name of the recipe - e.g. to weigh a portion of the dough later, or to use the recipe as an ingredient of another recipe. It asks for the name first (the name of the recipe is preset), then the scale goes to the weighing mode and saves it.
The recipe in the log
When a recipe is done (or ended), everything you have weighed becomes a weighing log - one line per ingredient with its values. Below the totals you see the number of portions and what one portion contains. From there you can add it to the 📒 diary, use the 🧮 servings calculator or weigh the pot after cooking or baking (🍲 PROCESSED). A paused recipe is logged when you finish it.
⎙ Portions & labels of your recipes
Below the 🥗 nutrition values of the recipe page, the block ⎙ PORTIONS & LABELS works like the 🧮 servings calculator of the log - but with the total amount you insert above, before you weigh anything. The result is shown right away with every change:
Amounts for / servings of
Insert the number of servings - or choose your weight unit to insert the size of one serving (e.g. 250 g), or a unit of the recipe (e.g. 500 kcal per serving). For a recipe with a portion size, the number of portions to make is preset.
Loss while processing
Cooking and baking make food lose water. Check the box and insert the loss in % of the raw weight (ℹ️ explains it): this amount of water is taken out of the recipe before the servings, the ingredients, the calories and all other values are calculated (e.g. 1000 g dough, 15 % loss = 850 g bread). Don't know it? Weigh the pot before and after the next time, or use 🍲 batch processing.
The result
Like in the log: the total weight (and after processing), the number of servings, the weight of one serving, what one serving is equal to (all units of the recipe, like kcal, sugar or the price) and what one serving contains (every ingredient with its exact amount, biggest first).
Label for one portion
🛠 PREP, ⇩ SAVE and ⎙ PRINT make a label for one calculated portion - with its weight as net weight, all units of the recipe and the EU nutrition table (checked from the start). It is its own label with its own setup and size (saved in your browser, used for all recipes) and works like the 🏷 LABELS with live values: picture, own texts with placeholders, order, highlighting, size. The first time it takes the size of the weighing label (e.g. for your label printer). Additionally there is the line Ingredients: all ingredients of the recipe, biggest amount first, broken into lines between the words. The ⚠️ allergens in it are written in capitals. Change the text in 🛠 PREP if you like - it is kept until you load another recipe. With a label printer set up, ⎙ PRINT prints it right away.
💰 Costs & margin of your recipes
Below the portions, the recipe page shows 💰 COSTS & MARGIN. Like the 🥗 nutrition values, every ingredient takes the 💰 price unit (price per g) of its THING with the same name, times its weight in the batch: "✓ 1.20 €/kg · 600 g = 0.72 €". An ingredient without a price shows ✕ no price and is named in red with its share of the recipe (the costs are too low then) - give its THING a price unit, e.g. 0.0012 € per g. If the THINGS have prices in more than one currency, you choose the currency (each ingredient uses its price unit with that symbol).
Typed in per recipe
Ingredients total (instead of the sum - e.g. from your invoice; empty: the sum), packaging per batch (jars, bags, labels), labour per batch, the portions (empty: the total amount / portion size of the recipe) and your selling price per portion. The app keeps them per recipe in this browser.
The result
The costs of the batch (ingredients + packaging + labour), per portion and per kg, and with a selling price the profit per portion and per batch and the margin (profit in % of the selling price). A selling price below the costs is shown as a loss. Without a total amount, the card shows the price per kg of the mix.
🏷 QR Code Labels for Recipes
Like the labels of THINGS: as soon as a recipe saved on the scale is loaded, the key QR is shown next to its name in the head of the page (on its start page and while you weigh it) - and the start page has 🏷 QR CODE LABEL behind ⋯ in its key row. Click one of them to make a label for your recipe binder, the box of a pre-mix or the shelf: the QR code, and optionally the name, the portion size, the ingredients (biggest amount first, ⚠️ allergens in capitals), the EU nutrition table per 100 g (if all ingredients have their values) and the description. Size, ⇩ download and ⎙ print work like for THINGS (also with a label printer).
Open a recipe with the camera:📷 SCAN A RECIPE LABEL in the recipe menu - or any other of our scanners (see every label works in every scanner). The app loads the list of recipes from the scale and opens the start page of the scanned recipe. The QR code contains only the name (e.g. "RECIPE:Basic Cookies"), so the recipe has to be saved on the connected scale. The same spelling is taken first, otherwise the same name with other upper/lower case. If the scale shows another page (e.g. the start page of an other recipe), it goes to the weighing mode for a moment first.
⚠️ Allergens
The 14 allergens of the EU food information regulation (1169/2011) - gluten (cereals), crustaceans, egg, fish, peanuts, soy, milk, nuts, celery, mustard, sesame, sulphites, lupin and molluscs - have to stand out in the list of ingredients. The app finds them by the names of your ingredients (in English, German and Spanish) and writes them in capitals: the word that names the allergen itself in capitals ("WHEAT flour", "WEIZENmehl", "HAZELNUTS"), and for other foods the allergen in brackets behind them ("Butter (MILK)", "Parmesan (MILK)", "Flour (GLUTEN)"). All texts of a label are bold, so capitals are what makes them stand out. Things like "coconut milk", "peanut butter" or "gluten-free flour" are not marked as milk or gluten.
The ingredients of a recipe on a label (⎙ PORTIONS & LABELS and 📥 FROM A RECIPE OF THE SCALE in the 🏷 LABELS) are marked right away.
⚠️ MARK ALLERGENS below the ingredients text of a label marks a text you typed in, and shows all allergens found in it.
The list of ingredients below the 🥗 nutrition values shows the allergens of every ingredient (⚠️ MILK).
Please check it yourself: the app can only guess by the names - it doesn't know what is inside a product (e.g. "Chocolate" with milk, or traces of nuts). You are responsible for the labels you print.
Pause ⏸️ Recipes
You can pause a weighing process at any time in order to repeat it later. For example if you have parts which should react over some hours before adding the next components to the mix. Simply click ⋯ > ⏸ PAUSE & REPEAT LATER, optionally add a name, and thats it. Now you can select the paused recipe from the list of all your recipes.
You can pause
up to 50 weighing-processes at the same time! This can be also different preparations from the same recipe (bowl 1, bowl 2, etc...) Add a name for every paused recipe to stay organized.
Repeat later
You can repeat the weighing process at any time, the scale will remember whats already in your pot! Just select the paused recipe from the list.
Pause again
In the moment you repeat a paused recipe, the paused version will get deleted from the scale automatically. Simply press the pause-button again if you want to store the updated version!
All 50 used?
Then the recipe will not be paused (and keeps on running) - load and finish one of your paused recipes first.
Save up to 250 recipes!
You are able to save up to 250 recipes including max. 50 ingredients each - directly on the scale! Like that you can use your recipes from each connected device without the need of any online-accounts or logins somewhere! Just plug your scale, connect with a mobile or notebook, load your recipe list from the scale and start working!
The recipe list (ALL RECIPES in the recipe menu, ALL DRINKS in the cocktail mode) shows one recipe per row with its kind (🍽️ food, 🍸 drink, 🧪 solution, ⚕ formula), its paused versions (⏸️ chips) below the name and ⤓ (load), ✏️ (edit) and ✕ (delete, asks first in the app's window) at the right - a click on the name does what the list was opened for, the row waits with a turning ⚖ until the scale answers. On top: the kinds as chips (tap one to show or hide it), a filter by name and + NEW. No recipe on the scale yet: a card with ⬆ RESTORE and the key + NEW RECIPE. The lines appear while the scale is still sending them: with many recipes you don't have to wait - click one at any time, the rest of the list stops loading (↻ LOAD ALL loads it again).
Names are unique on the scale: if you save a second recipe with a name that is already used, it gets a number behind its name, like "Cookies #2", "Cookies #3" and so on. A name can have up to 60 Latin letters - letters of other alphabets take more room: about 30 Cyrillic, Greek or Arabic letters, about 20 Chinese, Japanese or Korean characters. So a long name gets shortened for its number, like "Grandma's Sunday Sourdough With Rye, Spelt And Toasted... #2". After saving, the start page shows the name the recipe was saved with. If all 250 slots are used, the recipe is not saved (you can still run it once) - delete a recipe first if you want to save it.
Several devices on one scale: if more than one device is connected to the same scale, only the device where you opened the list (or pressed SEND for a new recipe) talks to the scale while the recipes are transferred - the others only watch. If a recipe is running on the scale that is not loaded in the app (e.g. after an app restart, or started on an other device), the app shows a ⟳ LOAD THE RUNNING RECIPE button instead of loading it automatically - so the other device is not interrupted.
💾 Backup: behind ⋯ in the key row of the recipe list (ALL RECIPES) you find ⬇ BACKUP and ⬆ RESTORE. BACKUP reads all recipes (with ingredients, instructions and description) from the scale and downloads them as one file. RESTORE saves the recipes of such a file on this scale - or on an other one. You choose which recipes, and what happens if a recipe with the same name is already there: skip it, save it again (it gets a number like "Cookies #2") or replace it. Paused recipes are not part of the backup. With a second scale connected you also find ⇄ SYNC there: see sync between scales.
⇄ Sync between scales: copy THINGS, RECIPES or MOLDS directly from one scale to an other one - without a file. Connect your scale, add the other one with ⚖ ADD MORE SCALES, and open the THINGS list, the recipe list (ALL RECIPES: behind ⋯ of its key row) or the mold list: next to BACKUP and RESTORE you find ⇄ SYNC (it comes and goes by itself when the second scale connects or disconnects). Every step is a window of the app with ✕ at the top right: first the other scale, the direction and the options (the ℹ️ behind the title explains them), then READ FROM THE SCALES (a loading window with a bar - its ✕ cancels), then one card per direction with a checkbox per item; the copy key stays gray while nothing is chosen and asks once more before it writes. The result comes in the message window of the app, one line per direction.
Other scale
...if more than one additional scale is connected, choose the one to sync with.
Direction
Copy from this scale to the other one, from the other one to this one - or both ways: then only the items missing on the other scale get copied (in both directions), nothing gets replaced. Afterwards both scales have everything.
Same name already there
(one direction only) skip it, save it again (with a number like "Cookies #2") or replace it - like a restore. Molds are compared by their values: skip or save again.
Lots
Also the lots of the lot store (checked in the THINGS sync): the 🧾 lots of the THINGS you choose go with them - like a backup.
THINGS and RECIPES together
The ingredients of recipes are connected with THINGS by their names. Check sync also the RECIPES (in the THINGS sync) or sync also the THINGS (in the recipe sync) to copy both: the THINGS first.
How it works
⇄ READ FROM THE SCALES reads the items of both scales (one after the other), then you choose which ones to copy. While the app talks to the other scale, its weight is not shown and it shows its transfer pages - afterwards it goes back to helping your main scale. Keep the device close to both scales; CANCEL stops it at any time.
Factory 🏭 Mode
⭐ PRO: the factory mode - gray in the free version; try it with the free trial (free and PRO).
The Factory 🏭 Mode
This mode allows you to fill molds step by step, just like on a conveyor belt. Perfect if you need to fill up many cavities/bags with always the same amount of something.
The factory-mode allows you to fill molds, one after the other, always with the same amount of filling, while counting, taring and logging everything automatically for you!
Fill molds step by step
First, define your molds with the amount of cavities, the amount of filling for each cavity as like as a range for over- and under-filling. The start page of the mode: SELECT A CONTAINER (the main key at the bottom right), DEFINE A NEW CONTAINER beside it, and behind ⋯DELETE A CONTAINER - the same list, with a 🗑 at the end of every row (asked first; the row goes at once). In the list, ⋯ also holds ⬇ BACKUP, ⬆ RESTORE and ⇄ SYNC of all containers. On the page of a new container the three optional parts (active ingredient, OK & PERFECT range, auto-overtake time) are rows with ▸ that unfold in place; tick "Save the container on the scale" above the key row, then START at the bottom right.
Now the scale will ask you to add the first container (or mold) and fill to the first cavity. For single bags/jars/containers (1 cavity) the program talks about "containers", for molds with multiple cavities about "molds". The scale will overtake the weight and get tared automatically if the filling is in the perfect range (and has been stable for at least 0.6 seconds, also with a shorter auto overtake time: so the tare for the next cavity is exact) - or just click the button manually (it counts at once).
The scale detects automatically when you put a container on it or take a full one away (also if you swap the full container directly with an empty one). Containers need to weigh a bit more than 5% of the filling (min. 0.5 g on the 420, 0.05 g on the 710, max. 2.5 g) to be detected - for lighter ones just press the main key ✔ YES, I DID.
Colors on the display and (optionally) sounds will indicate if you are close or perfect to your goal and the percentage-indication-bar will make filling easy for you. The colors are the same in the factory mode, the check-weighing and the fill weighing: orange too less, yellow in the OK range (a bit too less or too much), green PERFECT, red too much. Between the fillings (add the first mold, remove the full mold ...) the box is light gray - it keeps its size, so the page does not jump. The key row at the bottom: TARE · 📊 STATISTICS & ℮ · ⋯ (🔊 / 🔇 the sound - it is the sound setting of the whole app) · the main key of the moment: OVERTAKE while you fill, ✔ YES, I DID between the fillings. The program ends with the ✕ in the head (asked first: CANCEL keeps it running). The auto overtake time has its own card below the box.
The totals box below tells you how many molds and cavities you have filled altogether, the total and the average per mold and per cavity - and the over- or under-filling: the last unit (its weight and how far it was from the target, in g and %) and the average of all units against the target (per unit and in total: what you gave away or what is missing). Over is red, under orange (the weighing colours). 📊 STATISTICS & ℮ in the key row opens the statistics of all fillings without ending the program.
🛑 Predictive cut-off for filling machines
Fill with a machine - a valve, a pump or an auger driven by an ESP32? The scale can stop it via ESP-NOW. A machine never stops at once: the valve needs a moment to close and material is still falling when the stop comes. So the scale sends the stop early: as soon as weight + flow rate × lag reaches the filling of the cavity/container (the industrial "in-flight" or "dribble" compensation).
Where it aims - never under: the scale learns how much your machine scatters and aims 2 × that scatter above the target - or at the middle of the PERFECT range, if that is higher (with PERFECT -0 % / +3 % it fills +1.5 % above the target, the usual setting for prepackages). Never above the PERFECT range. Together with the learning below, about 1 filling in 100 is a little under the target (and then only by a hair) - in the worst case it gives a little more. The more exact your machine (e.g. with the slow feed), the less it gives away.
Tick 🛑 With a filling machine (cut-off over ESP-NOW) on the overview of the mold (its i explains it), above 🆗 START FILLING, the main key that starts the filling (this browser keeps it). If ESP-NOW is off, a yellow line at the bottom of the page says so while the program runs. Only then the running program shows the cut-off below the auto overtake time and switches it on - without the tick it stays hidden and off, so filling by hand keeps its page short. ESP-NOW has to be on in the scale settings.
Lag: the time from the stop signal until the last material is on the scale (start with 0.3 s). The flow rate is measured by the scale all the time (the slope of its last 16 weights, about 0.4 seconds, from the weight before the rounding - the same flow engine as the coffee modes). While the cut-off is on, the scale keeps its filters fast (a new weight every 0.125 seconds), also if slower filters are set in the settings.
Learn the lag (on by default): after every stop the scale waits until the weight is stable. Too much? It stops earlier next time - too less? Later. It learns never to underfill: a filling under the aim corrects the lag in full, one over it only by a quarter - so it settles a little above the aim (about as much as your machine scatters) and in the worst case gives a little more, almost never less. Single outliers don't throw it off. A learned lag is saved at the end of the program, the settings at once.
Stopped too early (below the PERFECT range)? The machine gets the signal to fill again (top up), until the cut-off comes again.
Slow down: x % before the target (coarse / fine dosing, optional, 0 = off): many filling machines have two speeds - a big flow for most of the filling and a small one for the end. Set how much before the target the slow feed should start, in % of the filling (e.g. 10 %: at 450 g of a 500 g bag). The scale sends fill_slow at that point (also early, by flow rate × lag), the machine goes on slowly, and the cut-off stops the small, steady flow - so less material is still falling at the stop and the fillings are closer to the target. Right after the slow-down the scale waits the lag and half a second more before it stops by the flow rate (the measured flow is still the fast one until the machine got slower) - meanwhile it only stops at the target itself. Choose the slow part so it runs for at least about one second. Machines with only one speed simply ignore fill_slow.
Pause: x s before the next filling (0 = none, up to 120 s): after every count (and its tare) the machine waits this long before it gets fill_run again - time to move the mold to the next cavity by hand, or for your own mold transport (it gets fill_state 5 meanwhile). After the last cavity of a mold there is no pause: the machine gets "mold full".
While the scale tares (after a count, after a new mold), the cut-off waits: the full weight that is still shown in that moment does not stop or slow down the next filling.
Your machine gets fill_run (fill now), fill_slow (slow down now), fill_stop (stop now - both sent at once in an extra packet), a stop number, the flow rate and the lag - and fill_state, the step of the program: 0 no filling program (ended), 1 START (waiting for a mold or container: the first one, and the next one after a full one), 2 filling, 3 slow feed, 4 stopped (wait for the count), 5 pause, 6 mold full (take it off). Every change of the state is sent at once. The MasterBlaster example sketch switches a valve or relay with them (set FILL_VALVE_PIN, and FILL_FAST_PIN for a coarse valve or the speed input of a two-speed auger or pump) - and closes them if no data comes from the scale for 0.5 seconds. Auto overtake and tare work as before: after the count the next cavity/container starts and fill_run gets true again (moving to the next cavity is the job of your machine).
Setup your Molds
Press "DEFINE A NEW CONTAINER" and setup your containers with the
Container type
Choose "Single Bags / Jars / Containers" (default) if you want to fill single bags, jars or cups. This is useful if you want to fill bags/cups with different weights with always the same amount! After every correct filling the scale will ask you to remove the cup/bag and will tare the next one automatically.
Choose "Containers/Molds with multiple cavities" if each container has more than one cavity, for example a chocolate mold.
Amount of cavities per mold
Only for containers/molds with multiple cavities: here you define the amount of "cavities" per "mold". For example for the 36 single bonbons in a chocolate mold.
Filling per bag/jar/container or per cavity
How much have to be filled in each of the bags/jars/cups or cavities? Add the amount here...
MG per cavity (optional)
This is an optional setting for edible makers. Just add the amount of your active ingredients in milligram-per-weightunit. The scale now will show you the averages per mold and per cavity - also the total amounts will get tracked, too. Like that you see at once if your portions are even (weight). The scale can only weigh - mix your batch well, so that even portions also mean even doses.
PERFECT Range
The "Perfect range" defines your weighing window. Weighings in this range will be marked green at the screen and a (optional) sound will indicate that you are perfect now. The scale will tare automatically now to get ready for the next filling (the time is adjustable within the running program later) - or press the button to start filling the next cavity.
If you are looking for an exact filling, just add 0 (zero) and 0 (zero) to the minimum- and maximum settings. In that case only a perfect filling will be marked as perfect and trigger the tare.
If you have some leeway with your fillings, add a little bigger range here. For example, add 0% for the under-filling and 3% for the overfilling to ensure that every bag have at least the minimum- but a little higher maximum amount.
OK Range
The "OK range" will work similar to the perfect one - but the scale NOT will not get tared automatically! Instead it only will indicate that you are close to your perfect weight, optionally also with a sound.
Save / Delete molds
You can save up to 42 different containers/molds directly on the scale! Simply check the box while defining a new one. Now you just can select them from your list to start filling them again. If all 42 places are used, a new one will not be saved (the overview will tell you) - delete an old one first.
💾 Backup: behind ⋯ in the key row of the mold list you find ⬇ BACKUP and ⬆ RESTORE. BACKUP downloads all saved molds as one file, RESTORE saves the molds of such a file on this scale - or on an other one. Molds that are already there (same values) are skipped, unless you choose to save them again. The auto overtake time is part of the backup. Copy molds directly between two connected scales with ⇄ SYNC: see sync between scales.
Counting & Logging
Within the program the scale will give you info about the amount of already filled molds and cavities as like as the average amount of filling for each cavity and mold. Like that you get a nice overview of how exact you was working at the end, as like as a sum of all used product, milligrams etc...
When you end the program, the totals (filled containers/molds and cavities, total weight, averages and milligrams) are written to a new logfile. Everything you log afterwards goes to a new logfile again.
Coffee ☕ Modes
The Coffee ☕ Modes
Espresso or filter coffee: both modes keep your recipe on the UP↑SCALE, show the weight, the time, the flow rate and the ratio live, draw your brew against the curve of your recipe and log every cup. The scale does the maths - the flow rate too: it measures it from 40 weights a second, more exact and with less delay than an app can from the Bluetooth packets.
The page, top to bottom: the head says ☕ COFFEE and the page (COFFEE TIMER, POUR-OVER, WEIGH YOUR COFFEE), its ✕ leaves the page. Under the weight the tabs LIVE VIEW (its ℹ️ right behind it: what is measured and what is an illustration) and 📈 LIVE CHART, then the view - on top of it a strip with the instruction of now (green only for READY, red only for STOP NOW and STOP POURING NOW, orange when there is more coffee than the dose) - the values, and the step card: what to do now in words. The key row at the bottom: TARE on the left, the action of the step on the right (START / STOP / WEIGH THE FULL CUP for a shot, JUG ✓ … START … FINISH for a pour-over), with WEIGH CUP or 📈 CHART between them. After a brew comes HOW DOES IT TASTE?, then the folded cards ☕ RECIPE and 🕘 BREWS & REPORTS. An alarm is a red line above the keys: its 🔇 (or a tap on it) stops it; what the recipe library did (loaded, saved, renamed) is a short note there.
☕ Recipe on the scale: dose, ratio and times are saved on the UP↑SCALE - every phone and tablet sees the same recipe. It is the folded card ☕ RECIPE below the step card, with your recipe in short on its second line (e.g. 15 g · 1:16 · 240 g · 3:30 min) - tap it to open its groups: READY-MADE, the recipe, the pours and steps (pour-over), MORE (the water temperature, the grind and the beans of the recipe, the spoken steps, automatic tare, start and stop, the alarm, vibrate, the bloom water), 🎯 EARLY STOP and 📡 SENSORS (espresso) and 📚 RECIPES (your own library, share, scan). A recipe loaded by name shows its name first on the line. While a pour-over brews, the card stays but its fields are locked. A function of PRO shows a 🔒 PRO chip on the FREE version - a tap says what it needs.
Values: the coffee in the cup (or the water poured), the time, the flow rate and the ratio - each with its target from the recipe below it; the last brew stays there until the next one starts.
Alarm options beep at the shot time, at a weight, at the yield, when the coffee is ready (espresso) - or after a time or at a ratio (pour-over).
Automatic log-file: every brew goes to the weighing log with its values - and its chart (📈 BREW CHARTS in the card WITH THIS LOG of the 📝 WEIGHING LOG).
How does it taste? After every brew a card asks: sour, bitter, sour + bitter, weak, strong or good - one tap gives an advice for the next one (finer, coarser, a longer or shorter ratio, more or less coffee, a more even puck or pour). For a shot the UP↑SCALE adds its own hint: the shot was on time, too fast or too slow for your recipe (see the dial-in below), and the advice takes it into account. With the brew history (PRO) the taste is kept with the brew, and the card gets two keys: 📝 NOTES (this brew in the history) and 🕘 HISTORY.
🔬 TDS - extraction yield (PRO; free on the UPCENT↑420): measure the TDS of the brew with a refractometer and type it in the card HOW DOES IT TASTE? after the brew - the UP↑SCALE works out the extraction yield = coffee in the cup × TDS / dose and shows where the brew is on the brewing control chart: the green box is the classic target (filter: EY 18 - 22 % at TDS 1.15 - 1.35 %; espresso: EY 18 - 22 % at a common TDS of 8 - 12 %), the dashed line is your brew ratio (a finer or coarser grind moves the brew along it). Under-, over-extracted, weak or strong: it says so with an advice. A pour-over without a lift-off at the end uses the estimated coffee in the jug (marked). With the brew history the TDS and the EY are kept with the brew.
🫘 Beans (PRO, with the inventory on the scale): under MORE pick the bag of beans from your 📦 inventory and the day it was roasted. Every brew of the recipe books its dose out of the bag - the UP↑SCALE does it at the end of the brew (espresso: the dose of the recipe; pour-over: the coffee you weighed in). MORE shows the days since the roast and about how much is left in the bag, the log line of the brew too. A 📦 check of the bag (weigh it) sets the stock right again.
📡 Sensors (espresso timer, PRO; free on the UPCENT↑420): in the card ☕ RECIPE of the espresso timer, the group SENSORS connects a pressure sensor or a thermometer of another maker next to the UP↑SCALE - over Bluetooth from the same browser, the UP↑SCALE stays connected. Its value runs as a line of its own in the brew chart (pressure 0 - 12 bar or more, the temperature with its own range, the value of now at its end) and is kept with the brew in the log and the history; the highest pressure and the mean temperature are values of the shot (history, CSV, weighing log). Pressure: Bookoo Espresso Monitor, Pressensor PRS, ESPROFILE, Popsicle (DIY), Transducer Direct; temperature: ETI BlueTherm (Thermapen Blue, ThermaQ Blue), MEATER, the probe of the ESPROFILE and every Bluetooth Health Thermometer. All sensors and their roles: 📡 sensors. ZERO sets the pressure of now as 0 bar - before the shot, with nothing in the group. Tell us how it works with yours.
📳 Vibrate at every step (MORE, both modes, on by default, saved in this browser): the phone vibrates with every beep of a brew - pour now, all the water in, a step to do, stop now. Phones and tablets with Android; an iPhone or a computer can not vibrate from a web app, there the switch is not shown.
🔊 Speak the steps (MORE, both modes, saved in this browser): the app says every new instruction aloud while you brew - pour now, wait, stir, lift off - so your eyes can stay on the kettle. It uses the language of the app and the speech output of the app (☰ → 💬 SPEECH OUTPUT, where you choose the voice; the ℹ️ of the switch takes you there): the switch starts as the speech output is set, and switching it on starts the speech output too. While the steps are spoken, the weight itself is not spoken on the coffee pages. STOP POURING IN 3 … is said once, then only 2, 1, now.
Espresso ☕ Timer
Monitor your espresso machine with the Espresso Timer, even if your scale does not fit under it! Optional automatic options detect the cup, the first drops of coffee and the end of the shot - place your cup on the scale, start the machine and the rest works by itself. A cup put on or swapped while the scale waits for the first drops is tared again (it starts no shot), and a cup taken off makes it wait for a cup again.
☕ The recipe
Dose (the ground coffee; ⥆⚖ weighs it into the portafilter or out of the bag), ratio (drink : coffee, e.g. 1:2; the chips 1:1 · 1:1.5 · 1:2 · 1:2.5 · 1:3 below set it with one tap) - the UP↑SCALE works out the yield (18 g × 2 = 36 g), the target of the chart, the alarm and the early stop - shot time (the whole shot) and pre-infusion: the seconds your machine wets the coffee before the first drops. The scale cannot see them - when the first drops start the timer (Auto Start), they are added, so the timer shows the same time as your machine. A start with the button or by the machine adds nothing. When the shot time becomes as short as the pre-infusion or shorter (e.g. a short recipe loaded), the pre-infusion goes to 0. Ready-made (at the top of the recipe line): five published espresso recipes with their own numbers - Espresso · Five Senses (22 g, 44 g in 28 s, 94 °C), Normale · La Marzocco (18 g, 36 g in 30 s, 93 °C), Flair 58 · Flair (the starting recipe of the Flair 58: 18 g, 40 g in 30 s), Turbo · Cameron et al. (the study "Systematically Improving Espresso", 2020: 15 g, 40 g, ground much coarser, 14 s; the study brewed at 6 bar) and Allongé · S. Rao (Scott Rao: 15 g, 73 g in 25 s). None of them gives a pre-infusion: loading one keeps the pre-infusion of your machine. See Recipes.
⥆⚖ Weigh the dose: your portafilters and dosing cup
⥆⚖ opens the coffee weighing: INTO THE PORTAFILTER or OUT OF THE COFFEE BAG (a card each; the weight shows – until you choose); USE THE DOSE ✓ at the right end of the key row takes the weight as the dose of the recipe (the ✕ of the head goes back without it). While you weigh, the ratio chips are on this page too and the line below them shows what the dose on the scale makes, e.g. → 36.0 g in the cup at 1:2 - USE THE DOSE takes the dose and the ratio chosen here, and the UP↑SCALE sets the yield. Into the portafilter, the card PORTAFILTERS & CUPS remembers up to 4 empty holders on the UP↑SCALE (for every phone and tablet): put the empty portafilter or dosing cup on the scale, press + REMEMBER when the weight is still and give it a short name (up to 16 bytes - ä and emojis take more than one). Later choose it and put it on the scale filled: the UP↑SCALE shows the dose in it at once - the weight above the empty scale minus the empty holder, no tare needed (a tare done before does not matter; with an RS-232 scale that tares itself, start with the empty scale at zero). The holder you chose last is chosen again next time. If the weight is less than the empty holder, the hint asks whether another holder is on the scale; above 40 g in it, whether something else is on the scale too (a yellow line above the keys) - USE THE DOSE stays gray until there is a dose. ✎ RENAME, ⚖ WEIGH AGAIN (a new basket, a cleaned portafilter) and 🗑 DELETE work on the holder chosen; ✋ TARE BY HAND works as before (the empty holder on the scale, TARE, fill it).
Place the scale
If your scale fits, place it directly below the portafilter to see the live flow of your coffee.
External checkmode
If your scale does not fit, the external checkmode times the shot anyway: weigh the empty cup, then the link at the end of the step card asks (in the app's window) to take the cup off - SCALE IS EMPTY starts it. Take the time with the keys, then weigh the full cup. While the cup is under the machine the weight shows – and TARE is gray.
Tare your cup
A click on the button tares the cup. With Auto ☕ Tare: just place your cup on the scale.
Coffee detection
Press the button when the extraction starts - or let the scale detect the first drops with Auto Start (1.5 seconds of weight).
First drops and peak flow
When the shot was started before the coffee comes - with START, by your machine (Decent Scale protocol) or with the timer key of your scale - the UP↑SCALE measures the time to the first drops: from the start until the coffee in the cup reaches 5 d and stays there for 1.5 seconds (a cup bumped and let go does not count; a cup put on after the start, or coffee already in the cup at the start, gives no time). It shows in the time tile (drops at 6.2 s, in place of the pre-infusion of the recipe), in the log and in the brew history - compare it with the pre-infusion of your recipe. With Auto Start there is no such time: the first drops start the timer there, and the pre-infusion is the number of your recipe, not a measurement. The peak flow is the highest flow rate of the shot - from 1 second after the first drops (the drops landing do not count) until the stop; after STOP NOW of the early stop it only drips, and above 20 g/s it is no espresso (a hand, a cup put on). It shows next to the average flow of the result, in the log, the brew history and its CSV.
Flow rate
The flow rate is measured by the UP↑SCALE itself: the slope of the weight over the last second, from 40 weights a second before the rounding, with single drops hitting the cup left out. It shows the flow of now - the tile shows the average flow of the shot when it is ready. In the chart the flow is drawn half a second earlier (half its measuring time), so it lines up with the weight.
☕ LIVE VIEW
The tab ☕ LIVE VIEW shows your machine with the portafilter and the glass on the scale. Without a cup the glass waits above its place; with the cup on, the group head lights up while the UP↑SCALE waits for the first drops. During the shot a stream runs from the spouts - its width and speed follow the flow the scale measures (single drops at a low flow, nothing without a measured flow) - and the glass fills with the coffee the scale weighs, against the yield of your recipe (🎯 on the glass and in the gauge on the left; a green ✓ when it is reached, orange when it is well over it). The display of the scale shows the weight and the shot time. STOP NOW of the early stop flashes up in red. At the end the result flashes up (the coffee in the cup and the shot time), and the cup stays with its crema and a little steam until you take it off. With a pressure sensor or a thermometer connected (sensors), a dial and a small display on the machine show their values live; without one there is none. In the manual check mode the glass waits above the scale during the shot (it is under the machine) and lands when you put the full cup on the scale. A shot started by a machine or app (Decent Scale protocol) looks the same, with a small note. The machine, the glass and the stream are an illustration (the ℹ️ under the gauge says it too) - measured are the coffee in the cup, the flow and the time. 📈 LIVE CHART shows the brew chart; switch at any time. ☕ LIVE VIEW is the standard view: a new cup on the scale shows it again, a chart you picked stays for that shot, and the live view stays after the shot.
The end
Press the button when the coffee is ready - or let Auto Stop detect it (4 seconds of a still weight). The full cup of the shot never counts as a new cup: the next cup is taken (tared) only after the scale was empty once - until then it says READY.
Dial-in hint
At the end of a shot the UP↑SCALE compares it with your recipe: the time it would have taken for the yield of the recipe (a shot stopped earlier or later is scaled to the yield; the pre-infusion is not scaled) against the shot time of the recipe. Within 10 %: ⏱ on time; faster: ⏩ grind a little finer; slower: ⏪ grind a little coarser. Below 1 g in the cup there is no hint. It shows in the result next to the taste row.
Machine timer
With the Decent Scale protocol on, a machine or app (e.g. GaggiMate, the DE1) starts and stops the timer: the scale switches to the Espresso Timer by itself, and the shot is logged.
🎯 Target yield (early stop)
Switch on Early stop (under MORE): the target is the yield of your recipe. The scale gives STOP NOW a little before the target: when the weight plus flow rate × lag reaches it - the same maths as the predictive cut-off of the factory mode. So the coffee that still comes after the stop (pump, dripping, the delay of the scale) ends up right at your target.
STOP NOW goes to the app (beep), to a Half Decent Scale with UPCENT↑420 (beep) and via ESP-NOW to a pump relay (fill_stop, see the MasterBlaster example sketch): the relay lets the pump run with the cup on the scale and during the shot, and stops it at STOP NOW.
Learn the lag: after every shot the scale compares the final weight with the target and corrects the lag by half of the difference. You can also set the lag by hand (0 - 4 seconds, start: 1 second).
The shot ends by itself when everything is in the cup (stable weight, max. 8 seconds after the stop); its time ends with the stop.
Machines with the Decent Scale protocol (DE1, GaggiMate...): the protocol has no stop command - set the target weight in the machine, it stops by itself. A shot started by such a machine is not learned.
Not in the external checkmode (no live weight there).
Pour-over ☕ Mode
For the filter cone (V60, Kalita, Chemex...): the UP↑SCALE guides you step by step and goes on by itself - OK always works too.
☕ The recipe
Dose, ratio - all the water : the coffee, the bloom water included, as in the usual recipes (15 g at 1:16 = 240 g water; change the water and the dose follows) - bloom (the time from the first water until you pour on) and total time. Under MORE: the bloom water = the dose × a factor (3 by default, 0 = no bloom). That is the classic recipe: the bloom water, then all the water in one pour. For more pours see below. ⚖️ Water follows the weighed coffee (under MORE, on by default, saved on the UP↑SCALE): when you weigh in a little more or less coffee than the dose, all the water, the bloom water and every pour follow it - the ratio of the recipe stays, so the coffee tastes the same (21.2 g instead of 20 g at 1:16.7: 354 g of water instead of 334 g). Switched off, the water of the recipe stays as it is (e.g. for the size of your carafe) and the real ratio shows in the result.
Pours and steps
In the recipe line, POURS AND STEPS: ➕ POUR turns the classic recipe into your own pours (the bloom becomes the first one) - each pour has its start (from the first water, m:ss) and the water poured by its end; the last pour is always all the water of the recipe. ➕ STEP adds a step at a time: 🥄 stir, 🌀 swirl, ⬇ press or ☕ release. Up to 12 steps - at 12 the two keys turn gray (e.g. the 4:6 method: five pours of 60 g every 45 s). While you brew, the bar on top says what to do - POUR 2/5 - UP TO 120 G, WAIT 12 S - THEN POUR TO 180 G, STIR NOW - it beeps at every new pour and step, and the scale goes on by itself: at the start of a step, or a little earlier when you start the next pour after more than half of the wait. A pause in a pour - the filter is full, you let it sink - is fine: the picture holds the kettle up without a stream, and the pour goes on when you pour again; only after 15 seconds without pouring does a pour that is short of its water end (no timer of the recipe stops for it). A pour that was too short is made up by the next one (every pour counts all the water). The water of a pour is kept as a part of all the water: change the dose or the ratio and every pour follows. ↺ CLASSIC goes back to the bloom + one pour (asked first: your pours and steps go). 🗑 at the end of a row removes that pour or step (asked first). Method above the rows (its ℹ️ right behind it). A time per pour (optional, the small field in of a pour, e.g. 0:10; empty = a free pour): then the UP↑SCALE compares the water with a steady pour over that time and says POUR FASTER or POUR SLOWER (orange, also spoken) when you are more than 5 g or 10 % of the pour behind or ahead - not in the first second, not in the last 2 seconds, not while STOP POURING IN counts down and not once the pour is in. The first pour runs from the first water; a later pour from the moment you really start it, at the rate of the recipe (its water in its time). 🎯 Pour trainer (free): after every timed pour the UP↑SCALE scores how close your water stayed to that line - 0 - 100: 100 right on the line, 40 points less for every 5 g or 10 % of the pour of mean distance (within the band of POUR FASTER / SLOWER on average: 60 or more; an overshoot or a short pour counts until you stop). The bar shows the score of each pour for a few seconds (🎯 POUR 2: 87 / 100), and after the brew the row below the result shows the score of the brew (the mean of its timed pours), every pour with its mean distance and a tip; the brew history keeps it (PRO). Below it, with two pours or more, a small table YOUR POURS: every pour with its start, the water it brought (and its target), the time you were really pouring (a pause is not counted), its mean flow and its trainer score. Practise with the ready-made V60 · J. Hoffmann: both its main pours have a time (30 s each). STOP POURING IN 3 … 2 … 1: near the water of a pour the bar counts down the seconds to its target at your pace of now (every pour, timed or not, with the water still in the air counted; not with the dripper in a stand beside the scale), then STOP NOW. The ready-made recipes bring their published times; the chart draws its curve with them.
Method
Dripper on a jug (V60, Kalita, Chemex …): jug, dripper, coffee, water - as described here. Dripper in a stand on the scale: the stand with the cup stands on the scale, the dripper sits in the stand - it works like a dripper on a jug (lift the dripper out of the stand at the end to weigh the coffee). Dripper in a stand, only the cup on the scale (the stand beside the scale, a holder on the wall): first put the dripper with its filter on the empty scale and weigh the coffee into it; then put it into the stand - the empty scale is zeroed - and put the cup on the scale. The UP↑SCALE weighs the coffee in the cup all the time, exactly, no lift-off needed - but it cannot see the water in the dripper above it: press START with the first water (the first drops in the cup start the timer too, a bit late), the pours of the recipe go by the time, and the brew ends by itself when the drops stop. The water poured is not weighed in this setup, so nothing is learned there. With a second scale under the stand (PRO; free on the UPCENT↑420): connect it to the app as another scale and choose it under MORE - Second scale under the stand. Its weight goes to the UP↑SCALE four times a second, so the water poured is known again (the water in the dripper + the coffee in the cup): the first water into the dripper starts the brew, every pour goes by its water, STOP POURING IN 3 … 2 … 1 and the pour trainer work, and the tile shows the water (💧) below the coffee in the cup. When its weight stops coming, the pours go by the time again. AeroPress on the cup: the same as the dripper on a jug with the AeroPress on your cup (its weight is taken in its own step); lift it off after pressing to weigh the coffee in the cup. At its press step the bar says TAKE THE CUP WITH THE AEROPRESS OFF THE SCALE: press beside the scale, not on it - the force of a press can harm its load cell. While everything is off, the bar says PRESS NOW - NEXT TO THE SCALE; nothing is weighed and nothing ends. Then put the cup back alone: it is weighed at once and the brew is done 3 seconds later - or put it back with the AeroPress on it to let it drip; lifting the AeroPress off later weighs the cup as usual. Immersion (French press, Clever, AeroPress upside down): the brewer alone on the scale - there is no dripper step. When all the water is in, the bar counts down STEEP to the total time; at the end take the brewer off the scale (pour it out, put the Clever on your cup, flip the AeroPress) and the brew ends there - the coffee itself is not weighed (or press FINISH). Pressing a plunger, stirring or swirling pushes on the scale - that is not counted as water: water counts only while the scale rises steadily like a pour (and a still reading just after a pour, up to 3 g). If you are still pouring when a stir or swirl step begins, that pour still counts; during a press nothing is added. A press step holds the water until the end, a stir or swirl step for 8 s.
1. The jug
Start with an empty scale and put the jug (or a cup) on it - the scale takes it as soon as it stands still (a jug left on the scale from the last brew starts nothing: take it off once). Rinse the paper filter before, if you do.
2. Dripper and filter
Put the dripper with the filter on the jug. (Jug and dripper put on together: the coffee step comes next.)
3. The coffee
Fill in your dose - spoon by spoon is fine: it goes on when the dose is in (it says how much is missing). More than the dose: press COFFEE ✓ to take it. The weighed coffee counts for this brew, the recipe keeps its dose.
4. The first water
Start pouring the bloom water: the timer starts with the first water (a real flow - pressing on the coffee bed starts nothing, and a false start is taken back).
5. Bloom - pour now - done
The scale follows the phases: bloom (pour the bloom water), wait (until the bloom time; it beeps at POUR NOW), pour up to all the water (it beeps when the target is reached). The scale does not tare after the bloom: the water poured is counted in one piece, bloom included. When all the water is in, the bar on top counts down LET IT DRIP to the total time of your recipe - a smaller line under it reminds you that lifting the dripper weighs the real coffee in the jug - and then says LIFT THE DRIPPER TO WEIGH.
Lift-off: the coffee in the jug
Lift the dripper off for a moment: the scale then weighs the coffee in the jug - measured (✓ in the picture and a mark in the chart) as soon as your hand is still for half a second; while it still drips, the number goes up live. Put it back and go on, nothing is lost. At the end lift it off and press FINISH, or just keep it lifted: after the total time of the recipe the brew ends about 3 seconds after the coffee in the jug is measured; before that (all the water in, the time not over yet) it waits 10 seconds - the bar counts down: the result shows the water poured, the coffee in the jug, the water the coffee kept and both ratios. Finished without a lift-off, the coffee in the jug is not measured. Then the picture shows your coffee ("YOUR COFFEE IS READY!") until you take the jug off the scale - after that the next brew can start. 📈 CHART OF THIS BREW in the key row shows its chart at once.
🫖 LIVE VIEW
The tab 🫖 LIVE VIEW shows the scale with the jug, the dripper and a kettle: the next item waits above its place, the coffee bed grows with your dose, the kettle comes up from the counter and pours while you pour and goes back when you stop, a gauge shows the water with the bloom and the target marks. A spoon or a hand on the scale does not make it pour - only water the scale counts. With the AeroPress only a few drops seep through until you press; pressed next to the scale, the picture shows the cup with the AeroPress in front of the scale. When the dose, the bloom water and all the water are right, a short ✓ OK flashes up. The gauge on the left shows the water poured (blue) and in it the coffee in the jug (brown) - with the number when it is measured (✓, the dripper lifted). In the gauge: 🎯 the goal of all the water and, dashed, ≈ ☕ the coffee to expect in the jug - an estimate, not a goal to reach; the same dashed mark (≈) is on the jug. When the coffee is measured at the end, only its weight with ✓ stays. Both follow the coffee you really weighed in, not only the dose of the recipe. The coffee to expect is all the water minus the water the coffee keeps: 2 g per g of coffee at first, then learned by the UP↑SCALE from every brew that ends with a lift-off (half of the way to what it measured; a value below 0.5 or above 4 g per g - lifted off too early - is not taken), separately for the dripper and the AeroPress. The levels in the dripper and the jug are an illustration (the scale weighs the sum; the ℹ️ under the gauge says it too) - measured are the ground coffee, the water poured, the flow and the coffee in the jug while the dripper is lifted. 📈 LIVE CHART shows the brew chart while you brew; switch at any time. 🫖 LIVE VIEW is the standard view: every brew starts with it, and it stays at the end (the chart of the brew is one tap away and in the log).
📈 The brew chart
Both modes draw the brew live: the weight (espresso: the coffee in the cup, pour-over: all the water poured) in blue, the flow rate of the scale in orange (right axis), the curve of your recipe dashed in green with a band for ± 10 % of its time, the target as a line and the target flow dashed - and your last brew in grey behind it, to compare (or the brew you made your reference). Marks show the pre-infusion, the phases of the pour-over, the early stop and every lift-off with the measured coffee in the jug. Every brew keeps its chart in the weighing log: 📈 BREW CHARTS in the card WITH THIS LOG of the 📝 WEIGHING LOG shows them again.
Zoom: the buttons below the chart show the whole brew (ALL) or the last 60 / 30 / 15 seconds - a zoomed chart follows the brew live. Drag it to move back in time, zoom with two fingers (or Ctrl + mouse wheel). ⤢ (at the end of the buttons) opens the chart over the whole screen - turn the phone for a big chart; it keeps running live; its ✕ or Esc closes it.
📚 Recipes: ready-made, your library, share
The ready-made recipes are the first choice in the ☕ RECIPE line, your own recipes are the part 📚 RECIPES at its end (both modes). The brew history and the cafe report are in the card 🕘 BREWS & REPORTS at the bottom of the page (it stays open or closed as you left it).
Ready-made (at the top of the recipe line, both modes): 53 published recipes with their own numbers, in one block per brewer - Espresso (5), Hario V60 with the Switch and the Mugen (12), Kalita Wave (8), Chemex (2), Origami (6), Clever Dripper (1), French press (8) and AeroPress (11). The name says the brewer and whose recipe it is (a roaster, a maker, a barista, a World AeroPress Championship place: WAC 2022 #3 = third place 2022); the window before LOAD shows its numbers. Each brings what its source gives: the dose, the water or the yield, the pours with their times, stir, swirl, flip (release) and press, the total time, and the water temperature and the grind where the source states them (a source that gives the temperature only in °F has none here: set it under MORE). A step without its own time in the source is left out. Pour-over: V60 · J. Hoffmann (James Hoffmann's "Ultimate V60 Technique": 30 g coffee, 500 g water, medium fine; bloom with 60 g for up to 45 s, to 300 g in the next 30 s, to 500 g in the next 30 s, stir at 1:45, drawdown by 3:30) and 4:6 · T. Kasuya (Tetsu Kasuya's 4:6 method: 20 g coffee, 300 g water at 92 °C, coarse; 50, 70, 60, 60 and 60 g at 0:00, 0:45, 1:30, 2:10 and 2:40, the dripper off at 3:30). Also: V60 1 cup · J. Hoffmann (15 g, 250 g, medium-fine: 50 g, a swirl at 0:10, then to 100, 150, 200 and 250 g at 0:45, 1:10, 1:30 and 1:50, a swirl at 2:00, about 3:00), AeroPress · J. Hoffmann (11 g, 200 g just boiled, AeroPress on the cup: all the water, a swirl at 2:00, press at 2:30), V60 Switch · Hario (15 g, 250 g at 92 °C, medium: 50 g bloom with the valve shut, to 250 g at 0:30, open the valve at 2:00) and WAC 2021 · T. Merikanto (the World AeroPress Champion 2021, inverted: 18 g, 200 g at 80 °C, 50 g, a stir at 0:10, to 200 g, a stir at 0:50, flip at 1:40 and press). Added: V60 from the guides of La Cabra (15 g, 250 g at 96 °C), Equator (24 g, 360 g), Verve (25 g, 375 g), Four Barrel (22 g, 330 g) and Hario's ambassador I. Al Mallouhi (20 g, 300 g), Tetsu Kasuya's one-pour recipe for the Mugen (25 g, 300 g, 1:30); Kalita from Equator (23 g, 390 g), Verve (20 g, 320 g) and Stumptown (21 g, 345 g); Chemex from Equator (45 g, 720 g) and Stumptown (42 g, 700 g); Origami from Stumptown (21 g, 340 g); Clever from Equator (25 g, 375 g, on the cup at 2:00); French press from Stumptown (56 g, 850 g, press at 4:00), Stone Creek (48 g, 720 g at 96 °C), Sightglass (38 g, 600 g) and 3FE (15 g, 250 g at 93 °C, the press at 9:00); AeroPress from Verve (14 g, 210 g, upright) and the World AeroPress Championship: WAC 2022 #3 · Ho (inverted, 18 g, 200 g at 84 °C), WAC 2018 #3 · Pinchukov (upright, 15 g, 230 g at 80 °C) and WAC 2010 #2 · Hasager (inverted, 18 g, 230 g at 75 °C). Also from roasters and makers: Onyx Coffee Lab - their recipes for Monarch, Sources, Power Nap and Southern Weather (Kalita), Geometry and Tropical Weather (Origami) and Eclipse (French press), with the grind in µm; ORIGAMI (a recipe of Daichi Hagiwara, 18 g, 300 g at 92 °C, six pours) and The Barn (Origami with its own cone paper and with Kalita paper, 94 °C); Ineffable Coffee (Kalita 18 g / 300 g at 94 °C, AeroPress inverted 16 g / 240 g at 96 °C, French press 25 g / 415 g at 99 °C); Climpson & Sons (V60 13 g / 200 g, cafetiere 40 g / 600 g, AeroPress inverted 14 g / 200 g, all at 95 °C); Square Mile (V60 15 g / 250 g in five pours, AeroPress 17.5 g / 250 g, French press 70 g / 1 l); AeroPress from Partners (17 g / 300 g) and Coffee Collective (16 g / 230 g at 93 °C). The French press, the Clever and an inverted AeroPress stand alone on the scale (method immersion): taking them off ends the brew. Espresso: Espresso · Five Senses (the standard recipe of Five Senses Coffee: 22 g, 44 g in 28 s, 94 °C), Normale · La Marzocco (La Marzocco's "using espresso brew ratios": 18 g coffee, 36 g espresso in 30 s, 93 °C - the temperature its machines are set to), Flair 58 · Flair (Flair Espresso's starting recipe for the Flair 58: 18 g, 40 g in 30 s), Turbo · Cameron et al. (from the study "Systematically Improving Espresso", Matter 2020: 15 g instead of 20 g, 40 g out, ground much coarser, shots of 14 s in the cafe that used it; the study brewed at 6 bar) and Allongé · S. Rao (Scott Rao's allongé: 15 g, 73 g in 25 s). They bring the dose, the ratio, the shot time and the temperature or the grind where the source gives them; your pre-infusion stays as it is. Only recipes whose numbers are published by their authors or makers are offered. Pick one, look at it and press LOAD: the UP↑SCALE makes it your recipe of now (dose, ratio, times, pours and steps, method and its name). Then change what you like.
My recipes (at the end of the recipe line; PRO, free on the UPCENT↑420): 💾 SAVE AS … keeps the recipe of now under a name on the UP↑SCALE - up to 30 recipes, espresso and pour-over together, seen by every phone and tablet. The name is typed in the app's own window (Enter saves); up to 24 characters - letters like ä or é and emojis take more room, the window shows what is left. The same name again replaces it. LOAD makes one the recipe of now, ✎ gives it a new name (the recipe of now with the old name takes the new one too), 🗑 deletes it (asked first; deleting works without PRO too).
🔗 SHARE: the recipe of now as a QR code and a link (SEND on a phone, COPY LINK on a computer). Right after a pour-over (its result still in the tiles) it shares the brew as you made it: with the coffee you weighed in as the dose - the ratio, the pours and the times of the recipe stay, so the water is as in your brew (the window says so). The name: type another one in the window - the code and the link follow at once (as long as the name on the UP↑SCALE: 24 bytes). A loaded recipe that you changed (✎ behind its name) or a brew shared as you made it gets (edited) behind its name - its numbers are not the ones of its author any more; when the name is too long for that, the author is shortened to the last name or left out ("V60 · Hoffmann (edited)", "V60 1 cup (edited)"). Whoever opens the link or scans the code with the phone camera gets the app with the recipe offered - it is loaded into the coffee mode it belongs to, after a tap on LOAD. Not connected yet: the app says so and waits - right after the UP↑SCALE (or the virtual scale) is connected, that coffee mode opens by itself and offers the recipe. 📷 SCAN reads the code of a recipe with the camera inside the app.
Not while brewing: a recipe is loaded before or after a brew.
🕘 Brew history
The part 🕘 BREW HISTORY in 🕘 BREWS & REPORTS at the bottom of the page (both modes; PRO, free on the UPCENT↑420) keeps every finished brew in this browser: its values, its chart and the recipe it was made with. After a brew, 📝 NOTES in the card HOW DOES IT TASTE? opens it at once:
Your notes: a rating (1 - 5 ★, the same star again takes it back), the bean, the grind, the taste (also set by the taste row after the brew), the water temperature and free notes - kept as you type. The bean is filled in for you: the bean of the recipe (an item of your inventory, with the days since the roast), else the bean you typed for the newest brew of that mode.
🫘 Per bean: above the list a chip for every bean of the brews - one tap shows only the brews of that bean, each with its grind and taste, and a summary line: how many brews, the last grind, the grind of the best-rated brew and the days since the roast. The same bean in other capitals counts as one.
🕘 OPEN lists the brews in a window of its own, the newest first (tabs All · Espresso · Pour-over; ✕ or Esc close it). A tap on a brew opens it, ← goes back to the list. Tick two brews of the same coffee mode and press ⚖ COMPARE: one chart (the older brew as the grey line) and their values side by side.
⭐ USE AS REFERENCE: the chart then shows this brew as the grey line while you brew, instead of the last brew - your best brew to aim at. One reference per coffee mode; tap again to go back to the last brew.
☕ LOAD ITS RECIPE makes the recipe of that brew your recipe of now again (not while brewing), ⋯ holds ⬇ VISUALIZER and, as its last entry, 🗑 DELETE (asked first).
📊 DAY shows the brews of one day (today first; pick another date): per mode the number of brews, the time and the coffee in the cup (mean ± spread), the ratio, the shots on time for the recipe (with the fast and slow ones, from the dial-in hint of the UP↑SCALE) and the mean rating - then every brew with its time of day, recipe, dose, coffee in the cup, time, hint and rating. 🖨 PRINT / PDF prints it (with your logo). Who brewed them is in the cafe report of ENTERPRISE.
⬇ CSV saves all brews as a table (Excel in the language of the app; with ENTERPRISE in your export format). Fixed column keys: date, mode, recipe, dose_g, yield_g (espresso: in the cup, pour-over: all the water), ratio, time_s, avg_flow_gs, peak_flow_gs, first_drop_s (espresso, only when measured), beverage_g, retained_g, bloom_g, rating, grind, temp_c, bean, notes, scale, pressure_max_bar, measured_temp_c (📡 sensors), taste (sour, bitter, both, weak, strong, good), roast_days.
⬇ VISUALIZER saves brews as shot files for visualizer.coffee, a website for shot charts and shot comparisons: in one brew this brew, below the list the ticked brews (none ticked: all brews of the list, at most 30, the newest first) - one file per brew (more than one: the app asks first, and your browser may ask once to allow several downloads). Upload them by hand on visualizer.coffee (your account there, Upload, choose the files); the app itself sends nothing. A file holds the chart (weight, the flow of the scale, the pressure and temperature of a sensor when one was used, the marks as stage lines), the start time, the recipe name as the title, dose, drink (espresso: in the cup, pour-over: in the jug, else the water), grind, bean, rating (the stars as 20 - 100), TDS / EY and your notes with a line of the other values (recipe, water, ratio, temperature, peak flow, taste, scale). Only brews with a chart: visualizer.coffee takes no brew without one. Pour-overs go up the same way; visualizer.coffee shows them with its espresso labels (weight, weight flow).
The history stays in this browser (not on the UP↑SCALE): another phone keeps its own, and clearing the data of the site deletes it. The FREE version keeps no history - the chart shows your last brew as before.
🏢 Cafe report and recipe rights (ENTERPRISE)
With the ENTERPRISE & COMPLIANCE edition the UP↑SCALE writes every finished brew into its activity log: the mode, the coffee in the cup (pour-over: in the jug, or the water poured when it was not measured), the time, the dose, the dial-in hint of a shot (on time, fast, slow) and the user who was logged in. The part 🏢 CAFE REPORT in 🕘 BREWS & REPORTS at the bottom of the page (both modes) shows a day:
Per barista and mode: the number of brews, the time (mean ± spread), the coffee in the cup (mean ± spread), the ratio and how many shots were on time for the recipe (with the fast and slow ones).
Every brew below: the time of day, who, the mode, the dose, the coffee in the cup, the time and the hint.
📊 SHOW puts it in the part, 🖨 PRINT / PDF makes a report with its number, integrity code and the seal of the UP↑SCALE, like the other reports of the edition.
The brews come from the archive of the activity log in this app and the newest entries of the UP↑SCALE - archive the log now and then so older days stay complete.
Recipe rights: with users on the UP↑SCALE an operator brews with the recipe but cannot change it - the dose, the ratio, the times, the pours, a recipe loaded or saved, the beans: only a QM or an ADMIN can (the app says so when the UP↑SCALE refuses). The rules of your change control apply as for every other change.
Keys and timer of your coffee scale
The UP↑SCALE with buttons (UPCENT↑420 on a Half Decent Scale): the round key tares, the square one starts the coffee timer on the weighing page, starts / stops the shot and goes to the next step of the pour-over. On a page of the app without its own use of the key it presses ∑ LOG IT.
A coffee scale of another maker on the virtual scale: its keys work the same way (tare, the timer key), and where its protocol has timer commands, the timer on its display runs along with the coffee modes (started with the shot, stopped at the end, back to 0 for the next one) - e.g. BOOKOO, Acaia, Felicita, Timemore, Varia, Skale, Eureka, the Decent Scale.
Cocktail 🍹 Mode
The Cocktail 🍹 Mode
If you like to drink a cocktail from time to time but you can't remember the ingredients and amounts, or if you are a bartender but your stuff can't remember anything - the cocktail-mode gets you sorted!
The cocktail mode is not only great if you just want to make a perfect cocktail from time to time - but also if you run a bar! Just add your own recipes directly to the scale and your staff is able to replicate them without any special schooling!
Create or Import Cocktails
Like at the Recipe Mode, you are able to create your own recipes. Or search and import them directly from cocktail databases! For more detailed info about how to create a recipe, check out the Recipe Mode, its the same, just for food instead of drinks!
Create Own Recipes
Add up to 250 own cocktail recipes, including description and maximum 50 ingredients for each recipe! For each ingredient you also can add a short description, as like as info about the amounts to add... Simply add the total amounts here, add a serving-weight - and in future you simply can upscale your recipe without any calculations needed!
The start page
Two cards (ℹ️ behind each title): 🍸 YOUR DRINKS with the row 🍸 ALL DRINKS › (the list of the drinks on the scale), and SEARCH AND IMPORT A COCKTAIL with the field for the URL or ID of a drink of TheCocktailDB and its key ⬇️ IMPORT, and the search field with its key 🔎 SEARCH (Enter works in both). The results are rows: a picture, the name (a tap imports it), 🌍 (the drink on TheCocktailDB in a new tab) and ⬇️ IMPORT. The keys at the bottom: 🍸 ALL DRINKS and + NEW DRINK (an empty recipe form for a drink). Search and import need the internet: while the app asks TheCocktailDB, a line at the bottom says so - without an answer (offline) a note tells you.
Search & Import Recipes
Search and browse public cocktail-databases and overtake the recipe with a click! You are able to edit all amounts and descriptions as you like - and to save the recipe after directly on your scale. Or just run it once! The ingredients get connected with your THINGS by their names.
Edit Saved Drinks
Change your saved drinks any time with ✏️ beside the drink in 🍸 ALL DRINKS - see Edit Recipes.
Cocktail Database Included
There are already 600+ cocktails and drinks from all around the world select-able!!
Run the Cocktail Mode
The Cocktail Mode will guide you through all the cocktail making process! Step by step it will ask you to add the ingredients, and its check-weighing options will help you to fill the exact amounts!
Upscale easily
Just add the amount of cocktail you want to make - or the amount of servings/glasses - and there you go! No calculations needed!
Live corrections
You have poured in much of something accidentally? Just hit the button and all other ingredients will be re-calculated. You will have more cocktail altogether then - but it will be perfect!
Exact proportions
Even if you pour the first ingredient roughly by eye - just hit the re-calculation button - and add the rest of the ingredients in the right proportions!
Check-Weigh & Check-Count ✅ Modes
The Check-Weighing ✅ Modes
Use one of the checkweighing modes if you want to control if the weight of things you are adding (or removing!) from the scale is within a given range. There are 5 different checkweigh- and checkcount modes included!
Here you can check if things have the right weight or amount - or you can use this modes to weigh what you get off from the scale instead of what you put on! (for example to fill bags without the need to weigh each single one every single time.) The checkweighing-programs will indicate if the amount is perfect (or in an optional OK-range) with colors and - if you like - with a sound. A percent-bar also will show how close you are - very useful if you are using this option for a manual fill-weighing! Also the scale will tare, count and log for you automatically in the checkweighing modes!
Simple (Zero) Check Weighing
Simple checkweighing. Add things to the scale, and it will show you if its (and how much) too much or too less. Then you can remove the things again manually.
Positive- or Negative Checkweighing
Compare the weighing item with specified limit values. (the target weight and optionally a range for OK and PERFECT). You can set the PERFECT range either in percent or directly as weight ("PERFECT from X up to Y g" below the setpoint) - the other values will be calculated for you. If this limits are exceeded or undershot, it will be highlighted and shown on the display (in the same colors as the factory mode and the fill weighing: orange too less, yellow OK, green PERFECT, red too much). Positive checkweighing determines whether the weight of the added material is within the target range. Similarly, negative checkweighing checks the weight of the removed material.
Positive- or Negative Checkcounting
In check counting, it is not the weight but the number of pieces that is compared with limit values. You can add the amount of pieces as like as the weight for one of it - and the scale will do the rest! The PERFECT range can be set in percent or as amount of pieces here too ("PERFECT from X up to Y x" below the setpoint). For sure we will show you not only the difference in percent, but also in pieces and as weight!
Options for the extended modes:
Auto-Overtake: Perfect weights can be overtaken/counted automatically after a given amount of time or with a click on the button. The automatic count waits until the weight has been stable for at least 0.6 seconds (also with a shorter time): a heavy item swings a moment after you put it on, and the tare after the count must not take that swing - otherwise the next zero would be off. The button counts at once.
After counting: choose how you work (positive check-weighing):
Tare - stack the next pack on top (default): the scale tares after every count, the next pack goes on top of the counted ones.
Single packs - zero when taken off: put every pack on alone. After the count the app shows "Counted - take it off!", and the scale zeroes itself as soon as it is empty and stable again. Until then the same pack can not be counted twice (no auto overtake, no button). A manual TARE ends the waiting.
Counting / Averaging: If counted per button or automatic option, a total weight as like as the average amounts of everything will be shown automatically!
The setup page: the card PLEASE SELECT A MODE (the three modes as chips - the extended ones with 🔒 PRO on a free scale - its ℹ️ explains the chosen mode), the setpoint with its unit and the ℮ key; the card of the extended options (weight per piece in check counting, positive / negative, auto-overtake with its seconds, after counting - only for positive check-weighing); the card OK & PERFECT RANGE. ✅ START CHECKWEIGHING is the main key at the right end of the key row (Enter in a field does the same). A wrong value gets a red line right under its field - also when the OK range is narrower than the PERFECT range (set both OK values at least as big as the PERFECT ones, or leave both empty). Your values stay while the app is open - also when you switch the language or the weight unit (the weights are converted). The next page names the program ("POSITIVE CHECK-WEIGHING", "NEGATIVE CHECK-COUNTING" ...) and says what to do; ▶ START ... is its main key.
While it runs: the weight (or the pieces) on top, below it the result box in the weighing colours: the word (TOO LESS, OK · TOO LESS, PERFECT!, OK · TOO MUCH, TOO MUCH), the percent of the setpoint and a bar - from 90 % on the last 10 % fill the whole bar (a mark at its left says so). 🔊 / 🔇 in the corner of the box switches the sounds, in every check mode. Below it what to do ("Please add 12 g"), in the extended modes the card THIS RUN (counted, total weight, averages - right-aligned) and 📊 STATISTICS. In the extended modes 🆗 OVERTAKE NOW (or IN n with the auto-overtake) is the main key - gray while a counted single pack is still on the scale; Enter counts too. The ✕ of the head ends the program (it asks first). Additional scales show their state in the display as stations (TOO LESS in orange, OK yellow, PERFECT green, TOO MUCH red).
℮ button: next to the setpoint - enter the nominal quantity of your ℮ prepackages and the setpoint and ranges are set for you (see ℮ prepackage control). Every counted pack goes into the 📊 statistics.
Fill 💯 Weighing
There is no extra fillweighing-mode, but we have different fillweighing options included in the other modes. All this options will show you a percentage-bar as indication for the right weight - with an extra scale for the last 10% - to allow exact filling even with bigger amounts!
You can use the simple percent-weighing options in the weighing- and counting modes. Here you can add your goal in units, in a weighing unit - or even if every other custom unit like dollars or calories with the help of the Fill-Weighing-Goals option!
Additionally you can use one of the Check-Weighing modes to fillweigh, here you can setup also ranges and get indications and automatic counting/logging! Also the Factory-Mode is a good option for fillweighing!
📊 Statistics & ℮ Prepackage Control
How exact do you fill? The statistics show it for every series of values - with one page for all modes. Open it with 📊 STATISTICS & ℮ in the menu (☰ → 📝 LOG & STATISTICS) or its tile in the packing, lab and audit views (it opens the series of the program that runs, else the current log), with the 📊 of the weighing log card and the log page, with 📊 STATISTICS & ℮ in the key row of the factory mode and of the running sorting program, or in the card THIS RUN of the check-weighing. The statistics are a page of their own over the page you are on: its title is in the head, ✕ in the head (or Esc) closes it and brings you back. Below the head the tabs: OVERVIEW (the numbers and charts), ℮ CHECK (the prepackage control - called HB 133 with the US rules) and ⚖ COMBINATION (only for a weighing log with lines named like A+B). Under the tabs you choose the series - it counts for every tab; a small green dot beside it means the running program of the scale sends its values right now and the page follows them. Opened from a log that has no statistics (coffee, factory, tracking log), the page says which series it shows instead. The series:
Check-weighing: every pack that gets counted (by button or auto overtake) - its weight, in check counting its pieces. With the PERFECT and OK range of your setup.
Factory mode: every filled cavity or container, with the ranges of the mold.
Sorting: every counted item, with the limits of the categories (undo and reset take the values back).
Weighing- and counting logs: every logged line (∑) - in the counting log the pieces of every line.
You see the number of values (n), the mean, the standard deviation (and in % of the mean), min/max and the range, the deviation of the mean from the target and how many values were in PERFECT, in OK and outside. With ranges you also get Cp and Cpk: how well your spread fits into the range (Cp) and how well it is centered too (Cpk) - from 1.33 green, from 1.0 orange, below red. The run chart shows every value in its order, in the weighing colours (PERFECT green, OK yellow, below T1 or the MAV line orange, outside the ranges or below T2 red) - so you see a trend, e.g. when a filling machine drifts. The histogram shows how the values spread. With ℮ switched on the numbers are net: a net chip in the head of the NUMBERS card says that the packaging weight (with the US tare sample: the average tare) is subtracted. A check-weighing started on another device brings its target but not its ranges - then there are no PERFECT / OK rows and no Cp, and the card says so. The last card, ⤓ EXPORT, saves all values of the series as one file with ⬇ SAVE (⭐ PRO; the format select and ⚙ need ENTERPRISE): number, time, name (logs with names), the value in gram (pieces in the counting series), with ℮ on the net value, the scale line (alibi) and the user when there are any. A locked key or select carries a 🔒 and tells you why when you tap it.
Check-weighing, factory and sorting values are collected by the scale: it keeps the last 2000 values of the running program until the next program or a restart, and sends every new value to all connected apps at once. An app that reconnects - or joins a running program later, e.g. a second phone - gets all of them from the scale. The weighing- and counting logs are kept in the app. Nothing is saved for later - save the values with ⤓ EXPORT if you need them.
Worked out by the UP↑SCALE: the numbers of a series, the limits of its ranges, the ℮ and FPackV check, the lot check, the HB 133 check (with its tare sample) and the masses of a combination weighing are computed by the connected UP↑SCALE - also by the virtual scale. While the answer is on its way the cards say The UP↑SCALE works out the numbers …; without a connected UP↑SCALE they ask you to connect it (the charts still show the values). A long series goes to the scale in parts, so the answer takes a moment longer.
⚖ Combination weighing (weighing design)
You need the masses of several small objects as exactly as your scale allows? Weigh them alone and together and let the statistics combine all weighings: every weighing then helps every object. Type what is on the scale into the THING / NAME field before every ∑ LOG IT: A, B, A+B, A+C, B+C … (+ joins the objects; spaces and capital letters don't matter; A+A counts A twice). As soon as one line of a weighing log has a + in its name, the tab ⚖ COMBINATION shows up in 📊 STATISTICS for that log (the ℹ️ beside its title explains the naming):
the mass of every object, fitted to all named lines at once (least squares), with its standard uncertainty u (k = 1),
u / 1 weighing: how much smaller the u of the object is than the u of one single weighing - e.g. 0.816 (−18 %) for A, B and A+B once each, 0.632 (−37 %) for 3 objects alone and in all 3 pairs (6 weighings),
Alone: how many lines had only this object on the scale,
the standard deviation of one weighing from the residuals (this needs more weighings than objects; with few degrees of freedom it is only a rough estimate) - or type in the one of your repeatability test,
folded: every weighing with its reading, the fitted value and the residual - a big residual shows a weighing that went wrong.
Good to know: if you only need one object, weighing it alone several times is better (3 times: 0.577). A combination design pays off when you need the mass of every object. If the weighings can't tell two objects apart (e.g. A and B only ever weighed together), there is no result: weigh one of them alone or in another combination. The combination only lowers the random spread: the errors of the calibration, linearity and eccentricity and the air buoyancy stay - check them with the scale self-check and use the air buoyancy correction for true masses. Tare before every weighing, place the objects in the middle and wait for a stable weight (or the stability light). The CSV of a log with names has a name column. The UP↑SCALE works out the masses - at most 8 objects from at most 16 named lines (more: the card says so). The maths is part of the validation cases (WD-).
℮ Prepackage control (EU 76/211/EEC, OIML R 87)
⭐ PRO: the ℮ prepackage control - gray in the free version; try it with the free trial (free and PRO).
℮ only with a verified scale: packs you sell with the ℮ mark have to be filled and checked on a verified (legal for trade) scale. The 420 and 710 are not verified - use them to plan and to practise. For your ℮ packs: your verified scale + our 232 UP↑SCALER = ℮ control, labels and statistics.
Packs with the ℮ mark (and most prepackages in the EU, sold with a fixed nominal quantity like "500 g") have to follow 3 rules - the packer is responsible for them:
The mean of the net contents of a batch must be at least the nominal quantity (Qn).
Only a few packs - max. 2.5 % - may contain less than T1 = Qn - TNE.
No pack may contain less than T2 = Qn - 2 × TNE.
The tolerable negative error (TNE) comes from the table of the directive: 9 % for 5 - 50 g, 4.5 g for 50 - 100 g, 4.5 % for 100 - 200 g, 9 g for 200 - 300 g, 3 % for 300 - 500 g, 15 g for 500 g - 1 kg, 1.5 % for 1 - 10 kg (the table ends at 10 kg; the % values rounded up to 0.1 g). Example: 500 g honey - TNE 15 g, T1 485 g, T2 470 g.
On the tab ℮ CHECK switch on "Check the series as ℮ prepackages" (with a free scale it is gray with a 🔒 and stays off), enter the nominal quantity and - if you did not tare the packaging - the packaging weight (it is subtracted from every value). The card RESULT shows the TNE, T1 and T2 (also as lines in the charts: packs below T1 orange, below T2 red) and the 3 rules with a green ✓ or a red ✗, then the verdict. The settings are saved in your browser and used for all weight series.
Fill with the check-weighing: in the check-weighing setup, the ℮ button next to the setpoint asks for the nominal quantity (in the app's own window; the TNE comes from the connected UP↑SCALE - a quantity outside 5 g - 10 kg is named in a message) and sets up a positive check-weighing for you: PERFECT from Qn up to Qn + TNE, OK from T1 up to Qn + 2 × TNE - so every pack below T1 is red. It also sets After counting to Single packs - zero when taken off (you can change it in the setup). Every pack you count then goes into the statistics with the ℮ rules. Only PERFECT packs are counted automatically: count the others (too light or too heavy) with 🆗 OVERTAKE NOW, or the statistics miss exactly the packs that matter.
Please note: this helps you to check your own packs, it is not a legal certificate. Inspectors check samples of a batch (e.g. 30 or 50 packs) and also take the spread into account - so fill with a safe distance to Qn. For selling prepackages your scale has to be legal for trade (a 232 UP↑SCALER on your verified scale works with all of this). Check the rules of your country.
📋 Lot check with sample plan & lot report
Check a whole lot like the inspectors do. On the ℮ CHECK tab, choose the rules (ℹ️ beside the label explains them): EU ℮ packs (76/211/EEC), Germany: FPackV for all packs with a nominal quantity, also without ℮ (max. 2 % of the packs below T1), International: OIML R 87:2016 (the same 3 rules with its own table of the tolerable deficiency T: also above 10 kg, the % values above 1 000 g rounded up to the whole g; the box title says PREPACKAGE CONTROL (OIML R 87:2016)) - or USA: NIST Handbook 133 (below). In the card LOT CHECK enter the lot size (all packs of one filling: max. one hour of a filling line, otherwise max. 10 000 packs; OIML R 87: max. 100 000) and if you open the packs (destructive test; not shown with OIML R 87, which has one plan for every test).
The app shows the sample plan. EU and FPackV: the reference method of 76/211/EEC annex II - a double plan from 100 packs: weigh the 1st sample; up to the acceptance number below T1 the lot passes this rule, from the rejection number it is rejected, in between a 2nd sample of the same size is weighed and the packs below T1 of both samples count together.
Packs in the lot
1st sample
accepted up to / rejected from
2nd sample (in all)
accepted up to / rejected from
mean criterion
100 - 500
30
1 / 3
30 (60)
4 / 5
30 packs, k = 0.503
501 - 3 200
50
2 / 5
50 (100)
6 / 7
50 packs, k = 0.379
3 201 and over
80
3 / 7
80 (160)
8 / 9
the first 50 packs, k = 0.379
Lots under 100 packs are checked pack by pack. Opened packs (destructive test): 20 packs, accepted up to 1, rejected from 2, k = 0.640 - only for lots of 100 packs or more; for a smaller lot the card says to check it without opening the packs. OIML R 87:2016: one sample from its Table 2 / Annex I for every lot size (e.g. 49 packs of 100 with 2 allowed below T1, 81 of 500 with 4, 98 from 600 packs with 5) and the sample correction factor SCF instead of k; 20 packs or fewer: every pack, none below T1.
Take the sample at random from the lot and weigh the packs in the order you drew them - the first values of the series are the sample (the 1st sample, then the 2nd one if the plan asks for it; the mean criterion uses the first packs of the sample). The verdict checks the 3 rules: mean + k × s ≥ Qn (the mean criterion: k or SCF comes from the plan, s is the standard deviation of the sample - so the sample mean may be a little below Qn, by k × s, the uncertainty of a sample), the packs below T1 and none below T2. You see ✓ LOT ACCEPTED, ✗ LOT REJECTED (also before the sample is complete, as soon as it can not pass anymore), weigh the 2nd sample or how many packs are still missing. The sample plan and the verdict of the EU, FPackV and OIML rules are worked out by the UP↑SCALE (also the virtual scale) - connect it to see them.
Drained weight of handmade packs: with the German rules you can check "Drained weight of a handmade pack or one with naturally grown food" - § 10 (3) FPackV: for packs labelled with a drained weight that are mostly made by hand or contain naturally grown food, no pack may be below Qn - 3 × TNE (instead of 2 × TNE). It only covers the drained weight of these packs - all other rules stay the same.
Results in colour: in the lot report - and in every other report, its print and its PDF - a result shows at first sight: ✓, OK, PASS, met, ACCEPTED in dark green, ✕, ERROR, FAIL, NOT met, REJECTED in dark red, ⚠ warnings in orange, always bold. The verdict of the lot (✓ LOT ACCEPTED / ✗ LOT REJECTED) is big, in a box of its colour. The colours are dark enough that a print in black and white stays readable.
🇺🇸 USA: package check along NIST Handbook 133
Packing for the USA? On the same tab choose the rules USA: NIST Handbook 133, then Labeled by (ℹ️): weight, count or volume. It follows the procedure of the US weights and measures inspectors, NIST Handbook 133-2026, Category A (all places except USDA-inspected meat and poultry plants) - the tab is then called HB 133 and its first card PACKAGE CHECK (USA, NIST HB 133). Every option has a small ℹ️ that explains it. The tables, the tare sample and the verdicts are worked out by the connected UP↑SCALE (also the virtual scale) from the gross weights (or the pieces) and the tares.
MAV (maximum allowable variation) instead of T1 / T2: one limit per pack from Table 2-5 of the handbook - e.g. 12.7 g for 227 g, 11/16 oz for 16 oz, 10 % below 36 g, 2 % above 24.67 kg. A pack more than the MAV below the nominal is an unreasonable minus error. The charts show the line MAV.
Unit on the label (g / oz / lb, ℹ️): the table has its own column for each unit and the values differ a little - choose the unit of your label (with both units on it: the one with the smaller MAV, to be safe). At first the weight unit of the app.
Moisture allowance (ℹ️): products that lose water after packing get an allowance from Table 2-3 - cannabis (plant material only) 3 %, flour 3 %, pasta products 3 %, dry pet food 3 %, fresh poultry 3 %, franks 2.5 %, bacon / fresh sausage / luncheon meats 0 % (the last three with wet tare: the free liquid and the wet packaging count as packaging weight) - or an own % with the guidance of your jurisdiction. The allowance lowers the nominal the packs are compared with (28 g cannabis with 3 %: 27.16 g, line Qn - MA in the charts); the MAV stays the one of the labeled quantity (HB 133 2.3.8.1).
Without a lot the card RESULT shows: the mean must not be below the (adjusted) nominal, and no pack below the MAV line. And a warning if the division of your scale is larger than MAV / 6 - the handbook asks for a finer scale then.
LOT CHECK (lot size with ℹ️): the sample plan of Table 2-1 - up to 11 packs all of them, 12 to 250 packs: 12, 251 to 3 200: 24, more: 48. Packs allowed over the MAV: 0 (1 in a sample of 48). The average error (mean - nominal) passes when it is plus; a minus only within the sample error limit SEL = s × the sample correction factor (0.635 for 12 packs, 0.422 for 24, 0.290 for 48). The lot report names the rules, the unit, the MAV, the allowance and all rows; packs below the MAV line are marked "> MAV".
Labels and fill weighing: with the US rules the label carries the nominal quantity without the ℮ (℮ is EU only) - switch on US + metric for the US units. 🛍 Fill weighing has the tolerance USA NIST HB 133 (MAV) too: green from the nominal, the MAV line instead of T1 / T2.
Labeled by count (50 gummies, 100 capsules): open a series of the counting mode and enter the labeled count (ℹ️). 50 items or fewer (Table 2-11): no average requirement - the sample may have 1 pack with fewer items (12 packs; 24: 2; 48: 3); a pack short by more than the MAV of Table 2-7 (17 or fewer: 0, 18 to 50: 1) is shown as defective even if the lot passes - repack or relabel it. More than 50: like weight, in pieces - the MAV of Table 2-7 (3 for 100, 7 for 250 ...) and the average within the sample error limit. The handbook asks for a scale that can tell one item (ℹ️) - else count by hand.
Labeled by volume (mL or US fl oz, ℹ️): the gravimetric procedure of HB 133 3.2 - choose the reference temperature (ℹ️, Table 3-1: 20 °C / 68 °F for most liquids, 4 °C for refrigerated food and beer, 15.56 °C for spirits), fill a volumetric flask with the liquid of 2 packs at that temperature and enter the flask capacity and both liquid weights (ℹ️ with the steps). The app takes the average density, turns every net weight into a volume (net ÷ density) and checks it with the MAV of Table 2-6 (14.7 mL for 500 mL, 29 mL for 1 L, 0.5 fl oz for 16 fl oz). If the 2 packs differ by more than one division of the scale, it warns: the density varies too much, the handbook asks for the volumetric procedure (not in the app).
⚖ TARE SAMPLE (weight and volume, ℹ️): for packaging that is not always the same (glass, cans). Weigh the closed packs (gross, do not tare the packaging), enter the lot size, open the first 2 packs of the sample (glass or aerosol with 24 or 48 packs: 3) and enter their packaging weight in the card TARE SAMPLE - typed in or ⚖ TAKE (the weight on the scale, shown in the head of the card; ⚖ TAKE is gray while nothing is on the scale). The app compares the range of their package errors (Rc) with the range of the tares (Rt) and tells you from Table 2-3 how many packs to open in all (from the same 2 up to the whole sample); the average tare of the opened packs is then subtracted from every pack, also in the charts and the report. ✕ CLEAR TARES starts again (after a question). Until the average tare is known, the rules, the verdict and the report keys wait for it. Off: one packaging weight for all packs.
Not covered: Category B (USDA-inspected meat and poultry plants), the volumetric procedures for liquids whose density varies, packs labeled by length or area, borax. Your own check before the inspector comes - not an official inspection; check the rules of your state.
Lot report: in the card LOT REPORT add your company (several lines: name, address... - one header for all reports: lot report, batch report and tracking report, also kept on the scale for every app), the product and the lot number. The keys at the bottom of the page: ⇩ SAVE (HTML), 📄 PDF (ENTERPRISE: the sealed PDF with the HTML inside) and ⎙ PRINT LOT REPORT (or save it as PDF in the print dialog); saving and printing are ⭐ PRO - with a free scale all three carry a 🔒. It holds the nominal quantity, the rules, TNE, T1 and T2, the lot size, the sample plan, the scale with its firmware and calibration date, the results with ✓/✕, the verdict, every value of the sample (packs below T1 or T2 marked) and a line for the signature. Weighed with a 420 or 710, the report and the app say that it is not a verified scale. Weighed with a 232, the report states the mode in a boxed line - "Weighed on a verified scale, in listen-only plus exact-weight mode" when both are on (otherwise which of them were off). The app learns the two settings when you open the RS232 scale settings once after connecting; until then it asks you to do that and the report leaves the line out. The line shows the settings at the time of the report. The settings are saved in your browser. This is your own check - not an official certificate.
Sorting 📦 Mode
⭐ PRO: the sorting mode - gray in the free version; try it with the free trial (free and PRO).
The Sorting 📦 Mode
Sort things by weight into your own categories - for example eggs into S, M, L and XL! Put one thing after the next on the scale: the scale sorts each of them into its category, counts it and sums up the weights, and gets ready for the next one as soon as you take it away.
The sorting mode sorts single things by their weight into categories, counts them and sums them up. Select "SORTING MODE" in the menu (☰ → ⚖ MODES): if no sorting program is running, you will land on the setup page, otherwise directly in the running program.
Create a sorting program
Program name
Give your program a name (max. 50 characters), for example "Eggs from the farm".
Categories
Add 2 to 42 categories. You will see the inputs for the first 2 categories - add more with "➕ ADD A CATEGORY" beside the title of the categories, or remove one with the "➖" at the end of its row (asked first when the row has a label or a limit). Each category gets a label (max. 15 characters, like S, M, L, XL) and a maximum weight.
Limits
Each category goes up to its maximum weight (the maximum itself still belongs to the category), the next category starts above it. The first category can have an optional minimum weight, the maximum of the last category is optional (leave it empty to keep the last category open).
UNDER / OVER
Things lighter than the minimum of the first category are counted as "UNDER", things heavier than the maximum of the last category as "OVER".
Weighing units
Insert and edit all weights in the weighing unit selected in the app. The programs are always saved in gram, so you can switch the unit at any time.
Active Ingredient (optional)
Tap the row "🧪 TRACK AN ACTIVE INGREDIENT ▸": it unfolds the card of the ingredient (its 🗑 removes it again, asked first). Add a name for an ingredient (for example THC), its amount per weighing unit and its unit (ng, µg, mg or g). Then you can choose to insert the category limits as weight or as ingredient amount: the matching weight (or ingredient amount) is calculated on the fly and shown next to each input. The counted ingredient amounts (average and total per category, and the total of everything) are shown in the running program and written to the log.
Presets, save & delete
Your sorting programs are saved in your browser, not on the scale. Tick "Save this program in this browser" above the key row of the setup page (the program is saved when you start it with "▶ START SORTING", the main key at the bottom right), or use "💾 SAVE/OVERWRITE THIS PROGRAM IN THIS BROWSER" behind ⋯ in the running program.
Select a saved program or one of the presets (egg sizes for Europe, USA, Japan and Taiwan, avocados, mangoes, baking potatoes (USDA), table potatoes (EU) and fresh truffles (UNECE)) in the card 📋 PROGRAM at the top of the setup page: the form will be filled with its categories and weights. Change what you like and start it or save it again.
If you save a program you have loaded with its unchanged name, you can choose to overwrite it or save it as a new program. If the name is already used (also by a preset), the new program will be saved as "Name #2", "Name #3" and so on.
Delete one of your programs by selecting it and tapping the 🗑 at the end of the program row (asked first). Presets can not be deleted. Overwrite or save as new? The app asks in its own window (CANCEL · SAVE AS NEW · OVERWRITE).
Programs saved in one browser are not available in other browsers or on other devices.
Run the Sorting Mode
Click "START SORTING" and put the first thing on the scale:
A bar shows all categories in different colors (with their name, or their number if the name does not fit), from zero up to a bit more than the last limit. An arrow on top of it points to the current weight, so you can see if a thing is at the lower end of a category, in the middle or close to the next one. The current category gets highlighted.
The current weight is shown at the top like in all other modes (grey as long as it is not stable). Below it you see the program name and the status: "Weighing..." with the category of the thing on the scale, then "Counted!" and "Remove it now!". As soon as the weight is stable, the thing is counted in its category automatically. Take it away and the scale tares quickly and is ready for the next one. Use TARE at the left end of the key row at any time if needed: it does not change any counts.
The card COUNTED shows the total of everything in its head and for every category how often it was counted, the average weight and the total weight. The key row: TARE · ↶ UNDO LAST (takes the last count back) · 📊 STATISTICS & ℮ · ⋯ · END THE PROGRAM. Behind ⋯: "RESET COUNTERS" sets all counts to zero and starts a new logfile (the old logfile keeps its numbers - asked first), "NEW PROGRAM" and "SAVE".
A logfile is created automatically with an overview of the run (total weighed, and count, average and total weight per category). It gets refreshed with every count.
The running program stays on the scale while it is switched on: if the app gets disconnected or reloaded, it will continue right where you left it. "END THE PROGRAM" (the main key, asked first) or "NEW PROGRAM" (behind ⋯, asked first) will end it.
Multiple devices
If multiple devices are allowed, other devices connecting to the scale will switch to the running sorting program automatically and get all counts that are already done. Counts, undo and reset will be shown on all devices at the same time.
Control other machines
Via ESP-NOW the scale sends the weight, the status (weighing, counted, taring), the category of every item and all counts to your own devices - and an extra ping in the moment an item gets counted, with its category. So a microcomputer can sort the items automatically by triggering a gate, a conveyor or any other machine. TARE, OK and RESET can be sent back to the scale.
Tracking Mode 📈 (Interval Recording)
The Tracking 📈 Mode (Interval Recording)
Record the weight at a fixed interval - for minutes, days or weeks: drying and curing, plants (when do they need water?), dough and fermentation, evaporation tests... The scale records by itself, also with the app closed, and the app shows live charts, statistics, daily summaries and a prediction when your target will be reached.
Select TRACKING MODE in the menu (☰ → ⚖ MODES). The scale records one point per interval: the mean of the shown weight over that interval (not just one reading), rounded to the scale division - with your speed, filter, locking and display settings. So a short bump or a breeze does not count, and a weight that stays stable on the display is a flat line in the charts. The scale records by itself: you can close the app or take your phone with you - when you connect again, the app loads all points from the scale. If a run is on the scale, the mode opens it directly.
Start a run
Put the empty container on the scale and tare it (tare button), then put your item on it.
Name (optional, saved in your browser) and the interval: one point every 0.25 seconds up to every 24 hours. The page shows how long the full detail lasts (8000 points, e.g. 33 minutes at 0.25 s or 5.5 days at 1 minute). Below 1 second the finest points stay only on the scale: when its memory is full they become 0.5 s, then 1 s points; from 1 second on, the app also keeps the full detail in your browser.
Start delay (optional): the recording begins this many seconds after the start - placing the item, closing a lid and the first movements do not count. The run page counts down; the zero and the first point are taken at the end of the delay.
Optional target and alarms, then START TRACKING. The first point is the weight at the start (or at the end of the start delay).
Changed your mind? The ✕ at the top right of the page leaves the setup without a run.
Live view, charts & statistics
The run page is live while the scale records, and stays as the result after you stop it. TARE at the left end of the key row tares the scale during the run (the space key does not tare here); the curve does not jump, the tare shows up as a marker. Below the weight: the name of the run with its badge (RECORDING, STARTS IN … or STOPPED) and ⏱ Sampling time, the current interval - it doubles each time the memory of the scale is full, and while recording the page shows when that happens next. Below that four tabs: 📈 RUN (alarms, tiles and charts), ⛰ EVENTS, 📅 DAYS and Σ STATISTICS. While the app loads the points from the scale, a line at the bottom of the page shows how far it is:
Tiles: the weight now, the change since the start (in your unit and in %), the rate of the last hour (per hour and per day - in the first hour of a run the rate so far, per minute and per hour), the time until the target with the expected date and time (from the current trend) - once it is reached, a ✓ with the time (several times: how often, the first and the last time), the duration and the points with the time until the next one. A run started with the drying preset also shows the estimated moisture now.
Weight chart: all points over the time with a time axis (all, 14 days, 1 day, 6 h, 1 h, 30, 15, 10, 5, 2.5 or 1 minute, 30 or 10 seconds), with the highest and lowest point, the target line, the average (Ø) and its band, the peaks (purple ▲ / green ▼), the biggest rises and drops (circles), tares (a dashed line with a round T - the curve does not jump with a tare), restarts after a power cut (⚡) and the current weight (dark dot). Click or tap one of these markers (also the dots in the zero point chart): a small box shows what it is, its value and the exact time - and the alarm, if the scale raised one at that point. While a run records, the charts stand still while the mouse is on a marker or an info box is open (a thin dashed outline shows it), so the markers do not move away under the pointer. The ℹ️ behind the title of the weight chart explains how to move and zoom.
Gross and net - what is the difference? When a run starts, the scale remembers its zero at that moment: the start zero. Then there are two ways to look at the weight:
GROSS = the weight against the start zero. A TARE during the run does not change it: the item on the scale still weighs the same, so the gross curve stays where it is. It always shows the real change of your item - no matter who pressed TARE when (you, another device, a machine).
NET = the weight like the display shows it. After a TARE the display shows 0 - and the net curve goes to 0 too.
As long as nobody tares during the run, both are exactly the same and you see only one line. An example - you dry 200 g of herbs:
display
gross
net
start of the run
200.0 g
200.0 g
200.0 g
you press TARE
0.0 g
200.0 g
0.0 g
later: 5 g of water are gone
-5.0 g
195.0 g
-5.0 g
The gross curve tells you "your herbs weigh 195 g now", the net curve tells you "5 g less than at the tare". Both are right - they only count from a different zero.
What you see: the chips in the head of the weight chart. BOTH (standard): the net curve (like the display) as the solid blue line, and the gross curve as a thin dashed line behind it - it only shows up after a tare, when the two are different. NET (display): only the net curve. GROSS (start zero): only the gross curve. In BOTH and NET, the change, rate, statistics and markers use the net weights; in GROSS the gross weights. After a tare, the "Now" tile also shows the other weight, and a note below the chart says what you see. The CSV then has both columns. Your choice is saved in this browser.
Why does the gross curve not go back to exactly 0 when the scale is empty? Every scale moves its zero a little over time (warming up, temperature, a restart after a power cut). The display hides that after a TARE or with the automatic zero tracking - the gross curve shows it honestly, e.g. -0.2 g with an empty plate. On a very fine scale (0.01 g) this is easy to see.
Only real tares count: the small automatic zero corrections (autozero, creep) are not a tare. They are only shown in the zero point chart, not as a T and not in the events.
Alarms use (in the alarm settings, also changeable during the run) chooses which weight the alarms of the scale (target, constant weight, jump, peak, band) use:
gross (standard, the safe choice): a tare during the run changes nothing for the alarms - no false alarm, and the run is never stopped by mistake.
net (like the display): the alarms follow the display. A tare itself is not counted as a jump - but careful: if someone tares while the item is on the scale, the net weight drops to 0 at once. That can reach a ≤ target right away and, with auto stop, end the run.
The target line and the band in the chart are always drawn where the scale really raises the alarm - in the gross and in the net view.
Zoom and move: the chips above the weight chart (a row that scrolls sideways) set how much time you see (counted back from now). Drag a chart left or right to move the window to any time of the run, Shift + mouse wheel (or two fingers on a phone) zooms in and out around the pointer, a double click zooms in (Shift + double click out), and − / + zoom around the middle. The thin strip below the weight chart shows the whole run with the shown window: drag the window, drag its edges to make it longer or shorter, or click beside it to jump there. The rate and zero point charts below always show the same window, and can be dragged and zoomed the same way. ▶| Live jumps back to the newest points and follows the run again. The zero point chart is always shown - a flat line means no tare and no drift of the zero in the window.
Rate chart: how fast the weight changed over the run - bars above the zero line when it goes up (darker), below it when it goes down (lighter). Great to see when drying slows down, or the day and night rhythm of a watered plant.
Zero point chart: the zero (tare point) of the scale over the time - when it was tared (●) or the autozero corrected it (·). It is always shown; a flat line means the zero did not move. The weight curve does not move with it: the scale measures the whole run against its own zero. The autozero corrections are only shown here (not as a tare, not in the events). During a run the autozero only works while the scale is empty for the run too: with an item tared away (display 0), it stays off, so slow changes of the item (drying, creep) are not zeroed away.
Peaks and events (tab ⛰ EVENTS): highest and lowest weight, the biggest rises and drops between two points (at least 1 d), the peaks, the tares and zero corrections and the restarts - each with date and time, the newest first. Zero corrections (drift) right after each other are one line: the newest, with how many (e.g. "· Zero drift (12×)"). The page shows the latest 10; ▼ SHOW ALL opens the whole list.
Daily summary (tab 📅 DAYS, from 6 hours of recording on): a bar chart with the change per day and a table with the first, last, lowest and highest weight of every day.
Statistics (tab Σ STATISTICS): start weight, last point, total change (in % if the start weight is at least 10 d), mean, standard deviation, range, mean rate, jumps, peaks and how often the weight left the average band.
⇩ CSV: all points with date, time and Unix time for your spreadsheet (after a tare during the run: the gross weight and the display weight). The key row: TARE · 🔔 ALARMS · ⋯ (⎙ PRINT, ⇩ REPORT, SAVE AS PDF, ⇩ CSV, the export format, ↺ RESET, ✚ NEW RUN) · the main key ■ STOP while the run records or waits for its start - after the stop ✚ NEW RUN. Every one of STOP, RESET and NEW RUN asks first. ■ STOP ends the recording (the points stay on the scale). ↺ RESET (while recording) clears the points and statistics and starts new at once with the same interval, target and alarms (and the same name): the weight now is the new first point, without a new setup or start delay. ✚ NEW RUN deletes the run from the scale. The ✕ at the top right of the page goes to the weighing mode. While a run records (or waits for its start), the tracking page can't be left: press ■ STOP first. This also applies to the modes, ✰ SCALE INFO and 🛠︎ SCALE SETTINGS in the menu. Closing the app or disconnecting is fine: the scale goes on recording, and when you reconnect (with any device) the app opens this page again with all the data.
Target, constant weight & jump alarms
Target: a weight - or a percentage of the start weight (e.g. 75 % = 25 % weight loss, perfect for drying). Choose the direction: ≤ raises the alarm when a recorded point drops to or below the target (drying, evaporation), ≥ when it rises to or above it (filling up, growing). For a percentage the target is calculated from the first point of the run, so the start weight never triggers it.
🌿 Drying / curing preset: enter the dry matter of your material (your own estimate: fresh cannabis flowers and herbs have about 20–25 %, i.e. 75–80 % water; partly dried material more) and the final moisture you want (10–12 % is usual for dried flowers and herbs). The app calculates the target: start weight × dry matter ÷ (100 % − final moisture), e.g. 22 % dry matter and 12 % moisture = 25 % of the wet weight (75 % loss). With the item on the scale it shows the target weight right away, and the weights for 12 … 10 % moisture. USE AS TARGET sets the target (≤, % of the start weight) and a point every 10 minutes (if the interval was shorter than 1 minute). The run page then shows the estimated moisture as a tile, and the alarm comes when the target is reached. Tare the empty tray, rack or jar before you put the material on. The moisture is only as exact as your dry matter estimate: to learn your value, dry a small sample completely once (constant weight) and divide its dry weight by its wet weight.
Constant weight: like a drying balance - "done" when the weight changes less than your value within your time (e.g. less than 0.2 g within 120 minutes).
Jump alarm: when the weight changes more than your value from one point to the next (watering, something was taken away...).
Peak alarm: at every high or low of at least your value - a peak counts when the weight went back at least that much from it, so small wiggles are ignored (e.g. the daily high and low of a plant pot).
Average band: when the weight goes more than your percentage above and/or below the average of the run (at least 1 d) - again every time after it was back inside the band.
Auto stop: for the target and the constant weight you can tick "stop the recording then" - the run ends by itself when it is done (e.g. drying finished), the points stay on the scale.
Fixed limits (silo, tank, reorder level): a LOW and a HIGH limit in your weight unit, each with a hysteresis. The alarm comes once when the weight falls to the LOW limit (or rises to the HIGH limit), and again only after the weight went back past the limit by more than the hysteresis - e.g. LOW 500 kg with a hysteresis of 50 kg: one alarm at 500 kg, the next one only after the silo was filled above 550 kg and is empty again. So a weight that hovers at the limit gives no new alarms. Both limits are lines in the weight chart (when the chart shows the weight the alarms use) and rows in ⛰ EVENTS. The scale checks the fixed limits itself, also with the app closed, and keeps them over a power cut like the other alarms: the alarm comes in the red card with the date and time of its point, as a system notification and through ESP-NOW and the watch link. For a silo or tank that always stays on the scale, switch on Zero Memory: the scale keeps its zero over a restart, so the weight of the full silo is right after a power cut.
System notification: when an alarm comes while the page is hidden (another tab, the screen off, the app in the background), the app shows a notification of your system. The browser asks once (at START or SAVE ALARMS); the tick System notification when an alarm comes and the page is hidden in the alarms switches it off. A browser that blocks the notifications of the page says so below the tick.
The scale checks the alarms itself, also with the app closed. The app shows them with the date and time of the point where they happened (with a sound and vibration) as soon as it is connected, in a red card on the tab 📈 RUN, until you press OK, SEEN. Change the alarms during a run with 🔔 ALARMS in the key row: a window with the same fields and SAVE ALARMS (its ✕ asks when you changed something).
The result: log, report & print
When a run is finished (STOP or auto stop), the app writes its result into the log - start and end, duration, start and end weight, the change (also in %), lowest and highest weight and the events. If the app was closed when the run ended, it does that the next time it opens the run. A finished run shows a card on top of the tab 📈 RUN with ⎙ PRINT, ⇩ REPORT, SAVE AS PDF (ENTERPRISE) and ⇩ CSV (PRO); the same are behind ⋯ of the key row, also while the run records. With ⎙ PRINT you print a report (with your company header, if you added one - see the lot report) with all charts over the whole run, the statistics, the daily summary and the events - ⇩ REPORT saves the same report as a file (to keep it or send it). Both also work while the run is still going (as an interim result), and ⇩ CSV gives you all points.
Memory, power cuts & time
The scale holds up to 8000 points in full detail - at one point per 0.25 seconds that is 33 minutes, per second 2 h 13 min, per 10 seconds 22 hours, per minute 5.5 days, per 10 minutes 55 days, per hour 11 months (also while ESP-NOW is switched on). When the memory is full, the run goes on: two neighbouring points become one and the interval doubles - so the whole run always fits, with less detail for the older part. A device that opens a running run loads all points from the scale first - a line at the bottom of the page shows the percentage and a bar, and the charts fill up while it loads. The app keeps the full detail: every point it got while it was connected stays in this browser (the last 3 runs), and the charts, statistics and the CSV use the finest data there is for every time. The points are saved in the scale about every 240 points, at most every minute and at least every 15 minutes (e.g. every minute at 0.25 s, every 4 minutes at 1 s, every 15 minutes from 4 s on), and at start and stop - few enough writes that the flash memory does not wear out. After a power cut the scale continues the run by itself (without a tare) and marks the gap. It waits 5 seconds before it records again, so the first point after the restart is not a still settling weight (no false jump, peak or band alarms); at most the points since the last save are missing (about 240 points, max. 15 minutes). The scale has no clock (solved by the device qualification): the app sends the time at the start and every time it opens the run. If the scale restarted while no app was connected, the times after the restart are estimated until an app connects again (marked in the CSV).
Alarms to other devices (ESP-NOW)
Via ESP-NOW the scale sends the state of the run in every packet (recording or stopped, points, last point, start weight, target, interval) and the alarms - with one extra packet in the moment a new alarm comes. So your own microcomputer can switch a buzzer, a lamp or a relay, or send a message. The MasterBlaster example sketch has an ALARM_PIN for that.
Density Mode 💎 (Archimedes)
⭐ PRO: the density mode - gray in the free version; try it with the free trial (free and PRO).
The Density 💎 Mode (Archimedes)
Find the density of gold, silver, coins, gemstones, minerals, meteorites, wood or 3D prints by weighing in air and in water - guided step by step, with built-in reference tables. Made for gold stackers, coin collectors, scrap metal buyers, jewelers and hobbyists.
Select DENSITY MODE in the menu (☰ → ⚖ MODES). The scale guides you through the measurement and takes every value by itself - all connected apps show the same state, so you can start on the scale's phone and watch (or tare, or measure again) on a tablet. It needs the firmware that has the density mode. The density is the weight of a solid divided by its volume. The volume comes from the water it pushes aside: the weight of the displaced water is the same number in grams as its volume in cm³ (the UP↑SCALE corrects it for the temperature of the water).
The three steps
1 · The dry item in air: the scale tares itself when the mode starts. Put the dry item on the plate: its weight in air is taken by itself when it is stable (1.5 seconds). The scale always waits until it has seen the empty plate with zero first, so a wrong tare can not spoil the result.
2 · The container with water: take the item off, put a container with water on the plate and press TARE (the TARE key at the bottom of the app, or on the scale). Use enough water to cover the item completely.
3 · The item in the water: hold the item on a thin string or needle completely under the water, not touching the bottom or the walls. The scale now shows the weight of the displaced water, which is taken by itself when it is stable (2 seconds). Items that float (wood, plastics) are held under with a thin needle in a stand.
The keys at the bottom: TARE (left; marked when it is the thing to do) and ↶ START OVER (new dry item) while you measure; when it is measured ↺ MEASURE AGAIN (same item, e.g. after bubbles - lift it out, wait for zero, submerge again), 💾 SAVE, ⋯ (🏷 SAVE AS THING) and ➕ NEXT ITEM (right). The ✕ top right closes the density mode on every connected device. The settings (liquid, temperature, material) are a card that folds; the saved measurements are the card SAVED (⋯: 📋 COPY TABLE, 🗑 at the end of a row asks first). With a liquid other than water (isopropanol, ethanol, salt water or your own, with its density) you can measure items that dissolve or react with water. For water, type in its temperature in whole degrees (default 20 °C): the UP↑SCALE takes the density of the water at that temperature from its table. A thermometer in the water fills it with 📡 (the scale takes whole degrees; the history keeps the exact value and the sensor).
The result
The UP↑SCALE works out the result itself - the density, the volume and the uncertainty - and every connected app shows the same numbers. The app shows the weight in air, the displaced liquid, the weight in the liquid, the volume and the density in g/cm³ (and kg/m³). The uncertainty is calculated from the resolution of your scale (0.1 g on the 420, 0.01 g on the 710) - it does not include handling errors, so small items are measured less exact: if the result is only good to about 3 % or worse, the app warns you. It also warns if both weights are almost the same (the item touched the bottom or floated).
Closest materials: the app lists the 4 materials whose typical density is closest (✓ = your value is inside the range of the material, uncertainty included).
Compare with: select a material (for example "Gold 18 K", "Quartz" or "PLA") and you get a verdict: inside the range, possible within the uncertainty, or no match - and the difference in %.
Fineness of gold and silver: for gold and silver the app estimates the fineness (‰ and karat) from the density, with a range - assuming a silver/copper alloy. If your scale is too coarse for this item, it says so instead.
Porosity: for plastics (3D prints), metals, gold and silver, a density below the solid material shows the share of air inside (infill, pores, casting holes).
Wood grade: very light to very heavy, for air-dry wood.
🏷 SAVE AS THING makes a THING from the measured density: its volume unit (ml per gram = 1 / density) shows the volume of any piece of this material you weigh later. Give it a name (like every THING name: up to 60 Latin letters; the closest material is proposed) and, if you like, a price per weight unit (g, kg, oz t, oz or lb, in one of the currencies of the price units): the weighing mode then shows the price with the weight, like a price-calculating scale. SAVE ON THE SCALE uploads it (an existing name asks to overwrite), ONLY TO THE WEIGHING MODE just uses it. Either way the density mode ends.
💾 SAVE stores the measurement with an optional name in this browser (the last 30), 📋 COPY TABLE copies them tab-separated for a spreadsheet.
Reference tables
The tables are built into the app (typical density and range in g/cm³): gold 24, 22, 21, 20, 18, 14, 10, 9 and 8 K (yellow gold with silver and copper), silver 999, 925, 900, 835 and 800, 20 metals (platinum, palladium, tungsten, lead, copper, brass, bronze, nickel, steel, aluminium, titanium ...), 23 gemstones (diamond, moissanite, cubic zirconia, ruby/sapphire, emerald, topaz, quartz, jade, amber, glass ...), minerals and rocks, meteorites (stony, stony-iron, iron), 20 woods (air-dry) and 17 plastics (PLA, PETG, ABS, TPU, nylon, resin ...). Gold alloys are calculated from the pure metals - white, rose and green gold are a few % off.
Tips & limits
No air bubbles on the item: a drop of dish soap in the water helps. Use a thin string and always the same depth, so its volume does not matter.
Do not touch the container or the string with your fingers while the weight is taken - and let waves settle.
Small items (under about 5 g on a 0.1 g scale) are not exact: use the 710 or a bigger item. The bigger the displaced water, the better.
Density alone does not prove gold: tungsten has almost the same density as gold (19.25 against 19.32). Gold plating on tungsten, hollow parts, stones and solder fool this test - combine it with other tests (drilling, acid or electronic test, XRF). The same is true for diamond and topaz (both about 3.52).
Wood, porous stones and open 3D prints take up water: measure fast or seal them (wax, varnish). Soft stones (opal, turquoise, pearl, coral) should not stay in the liquid, and never in alcohol.
Solution Mode 🧪 (standard and stock solutions)
⭐ PRO: the solution mode - gray in the free version. ENTERPRISE & COMPLIANCE adds its report (print, HTML, sealed PDF with electronic signatures).
The Solution 🧪 Mode
Weigh a few ingredients in fixed proportions - a standard or stock solution in the lab, resin and hardener, paint and thinner, plant food. Goals as weight, % or parts; weigh one ingredient a bit too much and the app calculates all other goals again, so the proportion stays right. The scale holds the solution: every connected phone or computer shows the same.
For a solution of a few ingredients in fixed proportions - e.g. 10 mg of a powder in 50 ml of a solvent (0.2 mg/ml). If you weigh 10.4 mg instead of 10.0 mg, the app does not ask you to remove powder: 🧮 calculates all other goals again in the same proportion - add 52 ml (52 g of water) and the concentration is the same. Select 🧪 SOLUTION MODE in the menu (☰ → ⚖ MODES). The scale holds the solution: every connected phone or computer shows and uses the same, like a running recipe. It needs the firmware that has the solution mode.
What else is it good for? Everything that is mixed in a fixed proportion, not only laboratory solutions: two-component resins and glue (epoxy, polyurethane, silicone - e.g. resin and hardener as 100 : 45 parts), paints and inks with a thinner, cosmetics, cleaners, plant food, aquarium mixes or a sourdough and dough batch. Choose parts if the data sheet says "100 : 45", % for a recipe in percent, or g for fixed amounts. Poured too much by accident? Do not scoop it out: take the value and choose ⋯ → 🧮 RECALCULATE on that ingredient. All other ingredients are calculated again in the same proportion - the confirmation lists their new amounts and the new full weight - so the mix stays right and nothing is wasted.
The rows: 2 to 10 ingredients, each with a name and a goal. Type the name or pick a THING with the 📦 button left of the name (or from the list while typing). Goal as a weight (in the selected weight unit), in % (they must add up to 100 %) or in parts (e.g. 4 and 1 for 4:1) - with % and parts also the total. A THING with a volume unit (its density) gets a small select: the goal and the weight in ml (or its volume unit) instead of the weight.
On the goal means within one digit (d) of the scale - or within the tolerance in ± … % of each goal (the field Tolerance ± in the card GOAL, empty = 1 d), when that is wider. Every row shows the band the UP↑SCALE accepts in g next to its goal (e.g. Goal 50.00 g ± 0.25 g). A tolerance narrower than d (e.g. ± 0.2 % of 10 g = ± 0.02 g on a scale with d = 0.1 g) cannot be shown by any reading: the UP↑SCALE flags that row, accepts it within ± 1 d, and the page shows an orange box ⚠ This scale cannot resolve the tolerance of these rows with the tolerance and the band it accepts. Take a finer balance, a larger batch or a wider tolerance - or weigh anyway: the report names the row and the band that was accepted.
⚖ TARE the empty vessel, then ⚖ WEIGH NOW on the row of the first ingredient: the button shows the live weight - green on the goal, red under or over, grey while it still moves - and the row shows a bar with the percent of its goal. A 2nd click on the button takes the weight (stable, or confirmed); WEIGH NOW on another row takes it too. Who took it and when is kept with the value. When a value is taken, the scale tares by itself: the next ingredient starts at 0 in the same vessel.
⋯ → 🧮 RECALCULATE on a row that is over or under its goal: all other goals are calculated again in its proportion (the confirmation shows the new total, and warns if an ingredient that is already weighed would then be too much). Gray when the row is on its goal.
TOTALS: below the ingredients a table shows what is weighed now, its share in % and in parts next to the goal (in brackets), and how much each row and the total is too little or too much. The big weight and the stability light stay on top. When all rows are green the ✔ FINISH key pulses until you click it; after FINISH it reads ✔ FINISHED.
✔ FINISH works when every row is weighed and green: a summary goes to the weighing log (one line per ingredient) and the THINGS go out of the inventory with it - once. Then the values are locked; ↺ AGAIN starts the same solution with empty values.
The keys at the bottom: TARE (left), 📂 LOAD, 💾 SAVE, ⋯ (🔊 / 🔇 sound on / off, ↺ AGAIN, ⎙ PRINT, ⇩ HTML, 📄 PDF) and ✔ FINISH (right, it pulses when every row is green). 🧮 RECALCULATE is in the ⋯ of a row. The ✕ of the head goes back to weighing.
ENTERPRISE & COMPLIANCE: after FINISH, ⎙ PRINT, ⇩ HTML and 📄 PDF (in ⋯) make the solution report: scale, calibration date, start and end, every ingredient with goal, weight, deviation, who and when, the tolerance, the SOP and the line to sign. The PDF is numbered and sealed by the scale and can be signed electronically by a logged-in user.
ENTERPRISE & COMPLIANCE: with the protection "only approved recipes can run" (see users) only approved, saved solutions are weighed; typed rows can still be saved for the approval.
ENTERPRISE & COMPLIANCE: with the air buoyancy correction on, the goals of THINGS with a density are true masses: the scale judges every row with the correction (🌬 on the row), the report shows the true mass with the reading and the factor.
💾 SAVE keeps the solution on the scale with a name and a short SOP / description (shown above the rows), as a recipe of the kind "solution" - in the recipe list it has a 🧪 and opens in this mode. 📂 LOAD opens a saved one. Saving starts the solution again (empty values).
ENTERPRISE & COMPLIANCE: the special mode for laboratories and pharmacies - gray in the FREE and PRO versions (14 days free to try).
The Compounding ⚕ Assistant (Rezeptur- & Abfüllassistent)
From the master formula to the signed batch record in one place: the assistant leads every batch step by step - raw materials with lot and use-by date, weighing that only goes on when it is right, checklists, in-process controls, filling with the uniformity of mass, the label - and writes everything into a batch record that the scale seals and two people sign.
Master formulas (% shares, tolerance per ingredient, SOP, shelf life, process parameters, approval)
Choose ⚕ COMPOUNDING ASSISTANT in the menu (☰ → ⚖ MODES). It works on the same rows as the 🧪 solution mode: the scale holds the batch, every connected phone or computer shows the same rows, goals and values, and the stability light, the 🌬 air buoyancy correction and the minimum weight work as there. The UP↑SCALE runs the batch itself: it checks the start (lots, use-by dates, blocked lots, checklist, approval, the identity data of the THINGS), lets only the ingredient of the current step be weighed, unlocks a container only after it checked its code or the label was confirmed, judges every in-process value and filling, and finishes only when every step is done. So every connected device shows the same batch and can carry on - a reload, another phone or a power loss loses nothing (open the formula again: the batch goes on where it was). One batch at a time per UP↑SCALE: while one runs, another formula shows it with OPEN ITS FORMULA or ✕ ABORT (with a reason). It is made for the records pharmacy and GMP work asks for - such as the manufacturing record of a pharmacy preparation - but whether it fits your SOP is decided by your QA: the batch record is a record that people sign, never a certificate.
⚕ Master formulas
➕ NEW FORMULA (or ⋯ → ✏️ EDIT FORMULA of an open one): name, formula ID, standard batch size, dosage form, shelf life in days, storage, the SOP (the description), process parameters (e.g. stirring time 10 min, 40 °C) and the ingredients - each with its share in % (they add up to 100 %), its own tolerance in % (empty: one digit d of the scale; a tolerance narrower than d at the batch size is flagged - see below) and an article number (taken from the THING when it has a unit named like Art.-Nr., PZN or SKU). Up to 16 ingredients.
The steps in their order: every ingredient has its ⚖ weighing step; add 📋 instructions ("warm the base to 40 °C"), ☑ checks (yes / no), 🔬 in-process controls (name, target ± tolerance, unit - e.g. pH 5.5 ± 0.3) and 🫙 filling, and put them in order with ▲ ▼. A preparation checklist ("workplace cleaned?") is ticked before the start.
🔍 Identity check of the containers:off, ✋ confirm the label (or scan) or 📷 scan to unlock. The approved codes and label names are not in the formula but in each THING (the same for every lot of it) - so every ingredient must be a THING of the scale that has them (scan: codes; label: names or codes). The editor shows ✓ / ⚠ per ingredient; 📥 READ THE THINGS FROM THE SCALE reads the ones this app does not know yet. The formula keeps a fingerprint of these data: if a THING gets other codes or names later, the batch does not start until the formula is saved (and approved) again.
💾 SAVE keeps the formula on the scale as a recipe of the kind "compounding" - with its whole process plan. In the recipe list it has a ⚕ and opens in this mode.
Approval (four eyes): with users, a saved formula is not approved until a QM approves this version (📂 OPEN A FORMULA lists the formulas in a window, a row each with ✔ approved / ⏳ waiting - tap a row to open it; a QM approves one with 🔏 there or in the recipe list). The approval covers the rows, the SOP and the whole process plan: change anything and it needs a new approval. With the protection "only approved recipes can run" the scale opens no other formula.
1. Preparation
Batch size: type the amount you want (e.g. 500 g) - the scale calculates every goal from the % at once.
Batch number of the product: type it, or 🧾 GENERATE the next number of the lot generator of the scale (the formula is the THING of the lot - the batch then has its own lot card, and its use-by date is set when you finish).
Raw materials: the lot of every ingredient (📷 scan it or type it - the lots of the THING and of the lot store are suggested) and its use-by date (a lot of the lot store brings it). An expired lot blocks the start, and so does a lot that is locked (blocked or recalled) in the lot store.
Tick the checklist, then ▶ START BATCH (bottom right; gray until everything is there - a red line above the keys says what is still missing). The UP↑SCALE checks everything again (also the use-by dates against its own clock and the lots of the lot store) and says what blocks the start. An ingredient whose tolerance is narrower than d at this batch size shows the orange ⚠ box above and asks once more at the start (it does not block - the batch record names the row and the band that was accepted).
2. The guided steps
On top while the steps run: the name of the formula with ℹ️, below it the batch number and the card of the step you are at (Step 2 of 5). The big weight of the page is hidden then - the weight is in the check box of the weighing step. The keys at the bottom: TARE (left) and the key of the step (right, e.g. ⚖ OVERTAKE); ⋯ holds ✕ ABORT BATCH, ✔ FINISH and the report (gray until every step is done), 🗂 BATCHES and the sound 🔊 / 🔇.
🔍 Container check first (when the formula has the identity check): the step is locked - no check box, no OVERTAKE - until the container is checked. Scan its code with 📷 or a hand scanner (it types into the field, Enter), or (with "confirm the label") ✋ CHECK THE LABEL: a window lists the approved label names, the grade and the lot of the batch, tick "I checked the label" and THAT'S IT!. The UP↑SCALE compares the code with the approved codes of the THING, the lot of the code with the lot of the batch and its use-by date with today - a code that is not approved, a GS1 code with another lot than the one of the batch, or with a use-by date in the past gives ⛔ wrong container - the step stays locked. Each check (how: scanned code or label confirmed) and each wrong scan goes into the batch record. Every time a weigh step comes up again, it is locked again.
⚖ Weighing: the scale tares by itself at the start of the step. A check box like in the check-weighing modes: the live weight, its % of the goal and the bar, colored yellow = add more (it says how much), green = on the goal (within the tolerance of the ingredient), red = too much; the weight is pale while it still moves. Below it the goal (with Already weighed: … (Δ %) when a part of this ingredient was taken before - it counts on live with what you add). (No AUTO TARE here: all ingredients go into one vessel - use ⚖ TARE when you need it.) The key ⚖ OVERTAKE (bottom right) shows what is on the scale now (a part weighed before is not in it, it is in the bar, the light and the table) and works when the weight is stable: green - the scale takes the value with who and when, and the next step comes. Yellow - the amount stays in the container as a part of this ingredient and counts, and the next open ingredient comes (the rest later). Red - a message what you can do (see "Too much?" below).
↪ Any ingredient, any time: you are not tied to the order. Tap an ingredient in the card INGREDIENTS below the step to weigh it now - what is already in the container stays and counts (the container stays on the scale, you only TARE again). An ingredient that was already weighed can get more the same way. The scale adds the part to the reading, the traffic light and the bar judge the total, and the batch record lists every part and every addition (who, when). Go from ingredient to ingredient until everything fits - the ones that are not done yet stay open and come up again.
Too much? Take some out - or press 🔏 QM RELEASE: a user with the role QM - never the user logged in on this device - chooses the name, types the password and a reason on this device. The scale checks the password (wrong tries count and lock like a login) and writes the release into its activity log; the batch record shows who released which value and why. Or 🧮 calculates all other goals again in the proportion of this one (noted in the batch record).
📋 Instructions are confirmed with ✔ DONE, ☑ checks with ✔ YES - a ✕ NO asks why, goes into the batch record as a deviation and the step stays open.
🔬 In-process controls: type the measured value (pH, temperature, ...) and press ✔ SAVE VALUE. The UP↑SCALE judges it: outside target ± tolerance it asks for the reason (a deviation in the batch record).
⋯ → ✕ ABORT BATCH asks for the reason and closes the batch record as aborted.
🫙 Filling and the uniformity of mass
The filling step weighs every container: tare the empty one, fill it, ✔ TAKE 1 / 20 (stable; ↶ UNDO takes the last one back) - the UP↑SCALE keeps every unit, works out the uniformity of mass and tares by itself with AUTO TARE - for the number of containers of the step (e.g. 20 capsules or 12 jars). The same colored check box shows each fill against the target. With AUTO TARE (the switch below the box, on by default, ℹ️ explains it) you only swap containers: take the filled one off, put the next empty one on - when it stands still for 1.5 s it is tared by itself (the bar pulses and says "hold still …" meanwhile); ⚖ TARE is optional. Below it the statistics like the 📊 statistics of the modes: the verdict, number, mean, standard deviation, min / max, units within the limit / outside it / outside twice the limit, and a run chart with the target and the limit lines (± limit green, ± twice the limit orange; a histogram from 10 units on). Every unit opens the list of each value with its deviation.
Uniformity of mass along Ph. Eur. 2.9.5 for capsules, granules and powders (single-dose), tablets, suppositories and pessaries, powders for parenteral use and for eye drops: 20 units, not more than 2 may deviate from the average by more than the limit of the table (e.g. capsules under 300 mg: 10 %), none by more than twice the limit. Or an own limit in ± % around the target fill (e.g. jars of a cream): every unit must be inside. The maths is checked by the validation cases UM-01 to UM-07, computed by the UP↑SCALE firmware (check page).
Not met? The app asks why the batch may go on (a deviation in the batch record).
3. Finish, batch record, label
✔ FINISH ends the batch on the scale (activity log), books every ingredient into the weighing log with its own lot and the person who weighed it and takes the raw materials out of the 📦 stock.
The batch record: formula, ID and approval (fingerprint), batch number, start and end, operator, batch size, use-by date of the product, the scale and its calibration, every ingredient with article number, lot and use-by date, goal, weighed amount, deviation in %, tolerance, who and when (and 🔏 who released it and why; ⚠ a tolerance narrower than d with the band the scale accepted), the checklist with times and names, every step with its result (in-process values, checks), the single fills with the uniformity of mass, all deviations and recalculations, the SOP and two signature lines. With a room sensor connected the record also keeps the room (air temperature, humidity, air pressure, with the sensor) at the start and the end of the batch; an IPC step in °C, %RH or hPa can take the value of a sensor with 📡.
🔖 The alibi number of a verified RS-232 balance stands below every ingredient it was printed for (EXTRA LINE: press the print key of the balance, then ⚖ OVERTAKE - or let the UP↑SCALE press it: ALIBI ON TAKE), next to every part of an ingredient and next to every fill of a filling step - the UP↑SCALE keeps them with the batch, so every device has them. Every value of the record can be traced to the alibi memory of the balance. In Germany this (or certified software) is what the verification offices ask for when software documents the weighings of a verified balance - see what stays legal.
⎙ PRINT, 📄 PDF (SEALED) and (in ⋯) ⇩ HTML with your company header, the report number and the integrity code; the PDF is sealed by the scale and carries the record as a JSON file. Right after the PDF the operator can ✍ SIGN it (⋯) (performed); later a QM signs it as reviewed / approved - the scale refuses a QM who made the report (four eyes). Check every file on the check page.
The finished batch shows the ingredients table with the weighed amounts and lots and, for every filling step, the units, their average, the limit, the units outside it and the verdict, with the single fills.
The record comes from the UP↑SCALE: the device that pressed FINISH (or aborted the batch) keeps it; any device can open it while the batch is still on the scale. ↺ NEW BATCH keeps the record in this browser and removes the batch from the scale.
🗂 RECORDS → BATCHES keeps every batch record in this browser: print it again, make the sealed PDF, sign it (the next signature goes on top of the newest signed file), the label, and a CSV of all batches. Keep the PDF files on your own read-only storage - the browser is no archive.
⋯ → 🏷 LABEL: product name, batch number with the lot code (QR), dosage form, quantity, made on, use by (end + shelf life), the ingredients with a click - their share in % and / or the weighed amount -, storage and a free text (patient, prescriber, use) - printed on your label printer.
To your LIMS / ELN by itself: with the forwarding of "finished programs" every batch record goes to your server at the end - as JSON, and if you want also as the file sealed by the scale.
↺ NEW BATCH starts the same formula again with empty values (a reason, a scanned code or a QM release reason can have up to 72 characters).
Pipette Check 💧 and Pipette Calibration ⚗ (labs)PRO / ENTERPRISE
For laboratories with a lab balance: the pipettes of your lab, of every maker, checked and calibrated on your balance through an UP↑SCALE RS-232 adapter (DIY / 232). 💧 PIPETTE CHECK is PRO (also with the virtual scale), ⚗ PIPETTE CALIBRATION is ENTERPRISE & COMPLIANCE (14 days free to try). The modes are only in the list when an RS-232 adapter or the virtual scale is connected: a kitchen or parcel scale is far too coarse for pipettes.
Pipettes: the quick check and the calibration
The UP↑SCALE firmware guides every step, takes every value from the balance by itself and does the maths; the app keeps the register, the records and the reports. Every option of the setup, every button of the run and every field of the register and the lab settings has an ℹ️ with a short explanation.
💧 Pipette check: an intermediate check between calibrations - one volume, a few deliveries, PASS / FAIL
⚗ Pipette calibration: gravimetric along ISO 8655-6:2022, as found / as left, every channel
🗂 Pipette register: codes, labels, history, due dates - shared with your other devices
⇩ Import of the pipette list of your old system, with an import record
Dispensers and piston burettes: the same check and calibration for bottle-top and repeating dispensers (ISO 8655-5) and piston burettes (ISO 8655-3)
Volumetric glassware: flasks, measuring cylinders, one-mark and graduated glass pipettes and glass burettes, built along ISO 4787:2021
Not only pipettes - dispensers and piston burettes: the register knows the instrument of every entry: a pipette (ISO 8655-2), a bottle-top dispenser or a repeating dispenser (ISO 8655-5) or a piston burette (ISO 8655-3, manual or motor-driven, e.g. the dosing unit of a titrator). Check and calibration are the same gravimetric procedure of ISO 8655-6:2022 with the same modes, records, CMC table and labels; the instrument decides the rest:
Limits: motor-driven burettes from ISO 8655-3:2022 table 1 (the official sample of the standard); manual burettes and bottle-top dispensers from a maker's procedure (secondary sources - check them against the official text); repeating dispensers have no default - enter the maker's or your own limits. The µL of the nominal volume hold at every test volume, at most 25 % of the test volume (ISO 8655-3:2022 formula 1). The report names the source.
The steps: no tips, but prime the instrument and discard the first stroke (or step) into waste; the checklist and the instructions follow the instrument. The vessel may leave the balance for a delivery (many maker procedures dispense into a vessel held under the outlet): the UP↑SCALE waits until it is back and takes the difference to the reading before - the app shows vessel off the balance meanwhile. Only EMPTY / OK takes a new vessel reading.
Single channel, test volumes in mL from 1 mL up; the balance must fit the volume (ISO 8655-6 table 1: d ≤ 0.1 mg up to 10 mL, d ≤ 1 mg above 10 mL - a precision balance on the adapter is enough above 10 mL).
Wording: every title and button names the instrument (Calibration of the piston burette as found, START THE CHECK OF THE DISPENSER) - it is never the calibration of the balance or of the UP↑SCALE.
Uncertainty: the handling share of EURAMET cg-19 (at least 0.1 % of the volume) is meant for piston pipettes; dispensers and burettes have their own lab setting (0 by default: 0.1 % would be as big as their limits).
Volumetric glassware: the same two modes check volumetric flasks, measuring cylinders, one-mark pipettes, graduated pipettes and glass burettes, built along ISO 4787:2021. A check is a receipt or intermediate check (PRO); a calibration gives the volume with its uncertainty (ENTERPRISE). Volumes are shown in mL.
In the register (✎ EDIT, instrument = the item): the nominal capacity in mL, the accuracy class A, AS or B (or no class: own limits), the tolerance table ISO or ASTM / USP (lab setting, or per item), to contain (In) or to deliver (Ex) (flasks and cylinders contain, pipettes and burettes deliver - change it for a TD cylinder), the material (borosilicate glass 3.3, soda-lime glass, or another material with its own γ), the inner neck Ø at the ring mark (one-mark items) or the reading resolution (graduated items: the smallest step you read, e.g. 0.02 mL on a burette with 0.1 mL divisions), the marked waiting time and delivery time, and the number of the maker's certificate.
Limits: the class tolerance of the table, the same at every point (class B = 2 × class A; AS = A), no random limit. The tables come from public secondary sources (maker catalogues, USP <31>) - check them against your standard. A size that is not in the table needs own limits. The setup, the result and the report name the standard and the class.
To contain (flask, cylinder): put the empty, dry item on the balance → OK; the UP↑SCALE weighs it empty (it waits until an item is on the pan). Take it off, fill it a few mm above the mark, set the meniscus down to the mark and put it back - the scale takes the value. Then it asks to empty and dry the item and put it back empty → OK → it is weighed again before the next fill. Up to 2 kg of water: a 1000 mL flask on a precision balance works.
To deliver (pipette, burette): the weighing vessel on the balance, the meniscus at the mark (or at zero) → OK. Take the vessel off, hold it at about 30° under the tip, deliver, wait the waiting time (⏱ counts it down) and touch the tip off; put the vessel back - the scale takes the value. A burette delivers from zero down to each test point.
No early values: the scale takes a glass value only after the item or the vessel was lifted from the balance - or after ✓ MENISCUS SET (also SPACE / ENTER or a foot pedal) when you set the meniscus with the item on the balance. So a meniscus being set or the waiting time of a burette never gives a value too early. TAKE NOW still works (marked by hand).
Test points and fills: one-mark items the nominal capacity; burettes 100 / 80 / 60 / 40 / 20 %, other graduated items 100 / 50 / 20 % (up to 5 points, change them in the setup). A check takes 3 fills (a receipt check may take 1), a calibration 10 (one-mark) or 5 per point (graduated) - ⚙ LAB SETTINGS → VOLUMETRIC GLASSWARE.
Uncertainty: the model of EURAMET cg-19: the expansion γ of the material at the water temperature, referred to 20 °C (or 27 °C, lab setting), and the setting of the meniscus from the neck Ø (u_p about half the line width, with a magnifier as low as 0.05 mm - lab setting, default 0.05 mm) or from the reading resolution (read once, or twice for burettes and graduated pipettes). Without neck Ø / reading resolution the meniscus is not in U, and the report lists it.
The rules of ISO 4787 are checked on a calibration instead of those of ISO 8655-6: humidity 35 - 80 %, at most 1 °C change of the room, the air thermometer U ≤ 0.2 °C, the water within 0.5 °C of the air, and the balance of ISO 4787 table 1 (d 0.1 mg up to 10 mL, 1 mg up to 1000 mL, 10 mg above). Deviations are listed, and the report then says "based on ISO 4787:2021 with deviations".
Wording: the record says within the class A tolerance of ISO 1042 (with the source of the value) - never "certified"; a maker's batch certificate is the maker's statement.
The balance
Connect your lab balance to the RS-232 adapter (RS-232 settings) and set its Max and d in ⚙ SETTINGS → SCALE DATA. The adapter reads every line of the balance exactly, digit by digit (also 0.01 mg and 0.001 mg balances, and balances with two ranges) and uses the balance's own stable mark.
ISO 8655-6 asks for a balance that is fine enough for the volume: d 0.001 mg below 20 µL, 0.01 mg from 20 µL, 0.1 mg from 200 µL to 10 mL. The app shows for every test volume whether your balance is fine enough. The coarser of the two counts: the last digit of the lines or the d of the scale data. A calibration does not start with a balance that is too coarse; a check warns.
Switch the balance's own auto-zero / zero tracking off: it can swallow a small delivery. A balance that only sends on its PRINT key works too: press PRINT after each delivery.
💧 Pipette check (PRO)
What it is: an intermediate check between calibrations (ISO/IEC 17025, 6.4.10) to keep confidence in a pipette between two calibrations. It is no calibration and gives no uncertainty; it moves only the date of the next check.
Choose the pipette (▾ CHOOSE from the register, 📷 SCAN its label - QR, Code 128, Code 39 -, a USB hand scanner, or + NEW), the test volume (default: the nominal volume) and the number of deliveries (default 4). The limits come from the pipette: the ISO 8655-2 table, the maker specification or your own limits.
A multichannel pipette: check channel 1, every channel, or a rotating group - 1 to 4 channels per check, and the next check takes the next ones, so every channel comes round (e.g. 3 of 12: channels 1-3, then 4-6, 7-9, 10-12). The register keeps where the next group starts; the record and the report name the channels that were checked.
Enter the water temperature (and the room values if you have them) - or take them from a sensor with 📡 (an empty field fills itself, see conditions from sensors) -, confirm the checklist, then ▶ START THE CHECK.
The steps on the scale: set the volume, fit a new tip and pre-wet it into the reservoir → OK (the weight is ignored meanwhile, so pre-wetting into the vessel does no harm). The scale takes the vessel reading when the balance is stable; then aspirate and dispense against the wall of the vessel: the scale takes every delivery by itself when the balance is stable - your hands stay on the pipette. SPACE / ENTER or a USB foot pedal is the OK. A delivery far from the expected mass asks KEEP or REDO (with a reason; a redone value stays in the record).
The result: mean volume, systematic error (µL and %), random error (s and CV %), judged against the limits → PASS / FAIL. The app that started the check saves it at once in the register (💾 SAVE NOW while it is not saved); 🖨 REPORT prints the record, ⋯ → 🏷 STICKER prints a sticker (date, next due, PASS); a failed pipette can go into quarantine with a status tag (⋯ → 🏷 STATUS TAG). + NEW RUN (or ▶ NEXT INSTRUMENT of a worklist) is the main key at the right.
⚗ Pipette calibration (ENTERPRISE & COMPLIANCE)
What it is: a calibration along the gravimetric reference procedure of ISO 8655-6:2022 - the result with its uncertainty. Whether it is traceable depends on your calibrated instruments (balance, thermometer, barometer, hygrometer - entered in the lab settings and printed on the report); whether it is accredited depends on your ISO/IEC 17025 accreditation and its scope (the CMC table). UP↑SCALE calls nothing "accredited" or "certified".
Test volumes: nominal, 50 % and the lower limit (10 % of the nominal or the lowest settable volume), 10 deliveries each, every channel of a multichannel pipette (aspirate with all channels, dispense one channel into the vessel). After the last delivery of a channel the next channel starts by itself - the tips stay on and are pre-wetted already, and the reading of the last value is the vessel reading; with ⚙ LAB SETTINGS → "Ask for an OK before every next channel" the UP↑SCALE waits for an OK instead. A 12-channel pipette is still 12 × 10 deliveries per volume: with one balance ISO 8655-6 counts every channel as a pipette of its own. As found = before any adjustment; after the maintenance (cleaning, seals, adjustment - recorded) ⋯ → 🔧 MAINTENANCE → AS LEFT of the result runs it again.
Evaporation: before the first volume the scale measures the loss of the vessel (10 s to 15 min, lab setting) and corrects every delivery by its own cycle time. A new tip at least once per volume (option: before every delivery).
Environment: water and air temperature, air pressure and humidity at the start and again at the end; the mean of both goes into the Z factor (water density after Tanaka 2001 - for air-saturated water as it stands in an open reservoir, the lab setting "Water is air-saturated", on by default; off = air-free water, about 2.5 ppm less volume -, air density after OIML R 111).
⋯ → ■ STOP AS FOUND ends an as-found run early (e.g. it failed already). Each volume shows its result right after its last delivery.
The calibration report: your own title (e.g. "Calibration certificate") and accreditation text, the pipette, method, instruments, conditions, every result with the limits, U (k = 2) and U in use, the decision rule, all single values and the uncertainty budget; sealed PDF with the report number of the scale, signatures (performed / reviewed - four eyes), the JSON record inside the PDF. Every finished calibration is written into the activity log of the scale.
Deviations from ISO 8655-6:2022: the standard says that a calibration may claim it only when every requirement is followed. The app checks what it can see: the d of the balance, your instruments (lab settings: at most 0.2 °C for the water thermometer, 0.3 °C for the air, 5 %RH, 10 hPa as U with k = 2; an instrument whose calibration was due, see the register), the room at the start and the end (20 ± 3 °C, or 27 ± 3 °C by lab setting; 45 - 80 %RH; at most 0.5 °C change), the water within 0.5 °C of the air, the test volumes and 10 deliveries per volume, a run stopped early (the UP↑SCALE checks them; the app adds the instruments whose calibration was due). Before START the setup lists what is missing; you can still calibrate. Then the result and the report list every deviation, and the report says "based on ISO 8655-6:2022 with deviations" and claims no conformity to the standard. Equilibration, water quality and handling stay on your checklist.
📡 Conditions from sensors: with a room sensor or a thermometer connected, START takes the values of now (a typed value stays), the run page shows the room and the water live and warns when the air moved more than 0.5 °C (glass 1 °C) since the start or the water is more than 0.5 °C away from the air, and the end form shows the values of now. When a sensor of the start sends no more (out of range, switched off, disconnected), the run page says so (no values since 14:32) and its field in the end form stays empty - type the value of now; if you can't, ✓ asks whether the values of the start count for the end, and the record and the report then say as at the start. The record and the report name the source of every value - a sensor (with its register entry, certificate and correction) or typed (with the instrument of the register you picked next to the field). A thermometer, barometer or hygrometer of the register brings its U into the budget when it is bigger than the lab's value.
Uncertainty and decision rule
The scale computes the uncertainty of every mean volume along ISO 8655-6:2022 / ISO/TR 20461:2023: repeatability, balance (calibration, linearity, resolution), evaporation, water and air temperature, air pressure, humidity, density of the weights, the water and air density formulas, the expansion coefficient and the handling of the instrument (EURAMET cg-19: at least 0.1 % of the volume for piston pipettes - the default, its own line in the report; dispensers and burettes: their own lab setting, 0 by default) - each with its contribution in the report. U = k × uc with k = 2. With an expansion coefficient γ the pipette counts at the water temperature, and the volume is referred to the reference temperature of the lab settings (20 °C, or 27 °C where adopted).
U in use is the uncertainty of a single delivery: it adds the scatter of one delivery and the operator, laboratory and process effects from your own studies (lab settings).
The standard uncertainties of your instruments and these effects are yours: enter them in ⚙ LAB SETTINGS from your certificates. The defaults of the thermometers, the barometer and the hygrometer are the largest values ISO 8655-6:2022 allows (as k = 1); the others are only examples.
Decision rule (lab setting): simple acceptance (|e| ≤ limit) or a guard band with U (ILAC G8): PASS, FAIL or INCONCLUSIVE. The report names the rule.
The balance and instrument requirements and the room conditions of ISO 8655-6:2022 are taken from the standard itself. The limits of ISO 8655-2:2022 come from public secondary sources - check them against the official text: single-channel from two sources, the multichannel table from one (above 2000 µL: 2 × single). At every test volume the µL of the nominal volume apply, at most 25 % of the test volume.
Positive displacement pipettes: D1 or D2. ISO 8655-2:2022 has two types, and the register asks for it (✎ → Positive displacement type): D1 - re-usable plunger and capillary (e.g. a glass capillary with a steel piston) - has the limits of air displacement pipettes; D2 - disposable plunger and capillary (capillary pistons used once) - has its own table: 5 µL 2.5 / 1.5 %, up to 10 µL 2.0 / 1.0 %, up to 20 µL 2.0 / 0.8 %, up to 100 µL 1.4 / 0.6 %, up to 1000 µL 1.2 / 0.4 % (systematic / random, from two secondary sources; multichannel: 2 × single). There is no default: a positive displacement pipette without D1 / D2 can not be saved, and an older entry without it gets no limits of the app until you choose (your own limits table holds for every pipette). The card, the report and the register report show D1 / D2.
Your own limits table (⚙ LAB SETTINGS → limits): one line per nominal volume, nominal;systematic;random in µL, optionally followed by the two multichannel limits (100;0.8;0.3;1.6;0.6), and lines for the limits at one test volume, nominal@volume;systematic;random (100@10;0.8;0.3). With your copy of ISO 8655-2:2022 you type its multichannel table and the limits at your test volumes once; a line with an error is shown and nothing is saved. The setup, the result and the report name the source of the limits: the table of the UP↑SCALE (secondary sources), your lab table, the maker specification (with the maker's document) or own limits of the pipette. The tables of the standards (the limits, the balance a volume needs, the test volumes) are in the UP↑SCALE firmware: the setup shows the limits and the needed d when the UP↑SCALE has answered (– until then), and with the table of a standard the UP↑SCALE takes the limits itself at START. An empty limits table in the lab settings starts with the table of the UP↑SCALE (BACK TO THE DEFAULT TABLE loads it again; the UP↑SCALE must be connected).
CMC table
⇩ UPLOAD A CMC TABLE (CSV) in the register (⋯ → CMC): the volume ranges of your scope with the CMC in µL, in % or as a + b × V, optionally per type (air / positive displacement) and channels. The app checks overlaps and gaps.
Every upload is a new version (file, SHA-256, who, when); the old ones stay, and each report names its version. With an ENTERPRISE UP↑SCALE connected, the upload is written into the activity log.
The reported U is never smaller than the CMC of the volume (ILAC P14: raised and marked). A volume without a CMC row is marked outside the scope, and the accreditation text is left off that line. The UP↑SCALE looks up the CMC and raises U when the result is there (the run is saved with it); the scale calibration does the same with the rows of type balance when you COMPUTE the result - after a new CMC table, COMPUTE again.
🗂 The pipette register
Every pipette (or dispenser / burette - see instruments): asset no., the instrument, maker, model, serial no., type, channels, volume range, tip, location, owner, status (in use, quarantine, out of service, retired), check and calibration interval, limits (ISO, maker specification, own), notes - and the next check / calibration due. Filters: due, overdue, quarantine, without a label code, and the customer.
Codes: keep your own labels, use ours, or both. Every code finds the pipette: your asset labels (Code 128 / 39 / QR), a maker's code (Data Matrix / GS1 with GTIN + serial - the GTIN fills the model from your model library), our QR sticker. One code belongs to one pipette only.
🏷 Labels with the label creator and your label printer: asset label (QR or your code as Code 128), the sticker of the last check / calibration, the OUT OF SERVICE / QUARANTINE tag; for many pipettes at once (tick them in the list).
Models and the makers' specifications (⋯ → MODELS): your own models with their defaults, the maker specification and product codes (GTIN), and a list of the makers' specifications in the app - the limits of common pipettes of Eppendorf, Gilson (PIPETMAN, PIPETMAN M), Rainin (Pipet-Lite XLS+, E4 XLS+, Classic), Thermo Scientific (Finnpipette), Sartorius, BRAND and Socorex as the makers print them in their operating manuals and specification sheets, each with the document (title, revision). Type or choose a model for a new pipette: it takes the volume, the channels, the tip, the maker's limits and the document, and a calibration takes the maker's test volumes. The app does not change these values: check them against the papers of your pipette, as a maker may change them. COPY puts a model of the list into your models, where you can change it. The report names the maker's document next to "maker specification".
👥 Customers (ENTERPRISE, ⋯ → CUSTOMERS) - for a calibration service: customer no., name, address and contact. A pipette can belong to a customer (empty: your own lab). The list of the register shows all pipettes, those of your lab or those of one customer; the register report, the CSV, the labels, the calendar file and the register file then follow the choice (a customer's register file holds only its pipettes, their runs and the customer - no lab settings). Each run keeps the customer as it was: the report shows its name and address, and the file name starts with the customer no. A deleted customer stays in the history of its pipettes.
⏰ Reminders: what is overdue or due soon - the days ahead are a lab setting (14 by default) - is shown once a day when an RS-232 adapter connects, with a button to the list. 📅 DUE DATES AS CALENDAR (.ics) saves every check and calibration due date as an all-day event with an alarm the same days ahead: import it into Outlook, Google Calendar or Apple Calendar, and the reminders come without the app.
History: every change of a pipette, of the lab settings, of a model and of a customer with who, when, old → new and - with strict audit - the reason. Each entry carries a SHA-256 check value of the entry before it and of the record after the change, so the entries form a chain: the line on top of the history checks it and shows a changed or removed entry, or a record changed outside the app, as broken. The register report has a column with this check for every pipette. With an ENTERPRISE UP↑SCALE connected, each change also goes into its activity log, together with the first bytes of the check value of the entry - the history in the browser is tied to the hash-chained log of the device. This makes changes visible (tamper evidence); it does not stop someone with access to the browser from deleting the register as a whole - keep register files as a backup.
Changes only with the UP↑SCALE connected: when an UP↑SCALE in the strict audit mode is connected once, the register keeps the rule (⚙ LAB SETTINGS → Audit): from then on the pipettes, lab settings, models and customers change only while an ENTERPRISE UP↑SCALE is connected and a user is logged in, so every change reaches its activity log. Without it the register is read only (search, reports, labels and files still work). Only a change made with the UP↑SCALE connected switches the rule off, and that change is in the history. Only an ADMIN or a QM may change the register when the scale has users.
Where it is kept: in this browser. It is shared live with your other devices that joined (⇄ JOIN), while they are connected to the same UP↑SCALE at the same time - nothing of it is kept on the UP↑SCALE. A pipette changed on two devices is shown with both versions and you choose. ⇧ SAVE REGISTER FILE / ⇩ LOAD REGISTER FILE for devices that never meet, for a backup and for a new computer.
🗂 RECORDS → 💧 PIPETTES: what is due (within the reminder days) and every saved check and calibration.
⇩ Import of a pipette list
Export the list of your old system as CSV (comma, semicolon or Tab; UTF-8 or Windows text) or JSON - from Excel: save as CSV.
The app guesses the columns from the headers (English, German, Spanish) and remembers your mapping; dates, decimal commas and volume texts such as "10–100 µl" are read. A preview shows what happens to every row: new, update, unchanged or refused (no volume, an asset no. twice, a code that is taken). Nothing is deleted.
The old last calibration and due date come along, and columns that are not mapped are kept as extra fields. A customer column (customer no. or name) puts the pipette with that customer of the register; an unknown customer is shown in the preview and the pipette stays with your own lab.
The import record (file, SHA-256, rows, mapping, every row with its result) is the evidence of the data transfer - print it, with ENTERPRISE as a sealed PDF. Afterwards: 🏷 labels for the pipettes without a code.
PART 3 THINGS, Logs & Labels
⚖ THINGS, Logging, Diary & Labels
Save what you weigh as THINGS, log it, keep a nutrition diary, a stock and a sales log - and print labels with the live values. These functions work in the weighing and counting modes (and partly in the recipe mode).
∑ Summing-Up / Logging
With a keyboard:SPACE tares the scale (whenever ⚞ TARE ⚟ is on the page) and ENTER presses ∑ LOG IT (in the roastery: its main key, in a blend run 🔀 BOOK THE BLEND; on the coffee pages SPACE is their TARE and ENTER their main key - START / STOP, the OK of a pour-over step, READY of the coffee weighing) - not while you type into a field, not while a window of the app is open (the ⛶ big view of the weight is the exception: SPACE tares there too), and not when a button has the keyboard focus (then the key presses that button, as usual). If you hit ∑ LOG IT (at the right end of the key row at the bottom, ⚞ TARE ⚟ at its left end) in weighing- or counting mode, the current weight will be added to the log-file! A small window in the middle says OK! GOT THE WEIGHT! (or COUNT) - EMPTY THE SCALE. After you have removed the logged goods, the scale will tare automatically and is ready for the next weighing. The window has two buttons: LOG & TARE tares the scale right away (it does not get stable by itself, or you want to go on with the goods on it), JUST LOG goes back to the weighing without a tare. (In the Takeaway Mode the scale will jump to 0 automatically and also will show you how much is still left on the scale or in your container!)
📝 Weighing log, 📊 statistics and 🗂 records: in the menu
The groups 📝 LOG & STATISTICS and 🗂 RECORDS of the menu (☰) hold all logs, on every page: 📝 WEIGHING LOG opens the log (greyed out while it is empty, highlighted as soon as it has entries; after every new entry a short NEW LOG ENTRY! shows at the bottom of the screen - tap it to open the log). As soon as the log has entries, it is also right on the weighing and counting page as the card 📝 WEIGHING LOG below the THING / NAME card (above the tiles of your app view, which are always the last thing of the page): its last lines (the name left, the weight - in the counting mode the count and the weight - right), then Sum (with the number of lines) and ⌀ (the average); ℹ️ behind the title explains the card in short, ➕ starts a new, empty log right there (the old one stays on the log page; the sums start again), 📊 its statistics and OPEN the whole log (copy, the other logs, diary, portions), 📊 STATISTICS & ℮ opens the statistics, and each entry of 🗂 RECORDS opens the page of the records that are kept for later on its tab: 🛡 the activity log of the scale with its archive, ☑ the test history of the scale, ✓ the routine checks of this browser, 🖥 the devices (the register of the scale and the form of this device), 👥 the user access register, 🧾 the lots of the lot store, ⚕ the batch records of the compounding assistant, 💶 the sales log and - when their mode is on - 🥦 the club reports, ☕ the roastery reports, 🏪 the reports of the dispensary and 💧 the pipette register. The weighing log only lives in the app until you close it; the records are kept on the scale or in your browser (all logs at a glance). Without the ENTERPRISE & COMPLIANCE edition the page shows what your scale keeps, without the sealed exports - a PRO scale saves its lots and its sales log as plain CSV files (the activity log and the test history are kept by ENTERPRISE scales only).
The records page lies over the page you are on: the head shows 🗂 RECORDS small and the open tab large, the ℹ️ behind it says what the tab holds, ✕ (or the Esc key) closes it. The tabs are right under the head (swipe them sideways when they do not fit). Every tab is made of cards and lists - one row per entry, the numbers on the right; a row with ▸ opens something (a lot card, a report, a pipette). The exports and prints are keys at the bottom: the main action of the tab at the right end, at most two more keys beside it, the rest behind ⋯. A key your licence does not include carries 🔒 PRO or 🔒 ENTERPRISE and tells you why when you tap it - it is never the main key. The page keeps its tab and its scroll position while new data comes in. The tabs: 🛡 ACTIVITY (the user filter and REFRESH above the log; with ENTERPRISE the card Archive of the activity log with ARCHIVE NOW, the export format and Check exported archive files, the card Audit trail review, the keys ⬇ EXPORT ARCHIVE and 🗄 SAVE FOR THE RECORDS), ☑ TESTS (REFRESH, the list, the trend), ✓ ROUTINE (the list; EXPORT LOG and PDF REPORT), 🖥 DEVICES (this device, the device register; 🖥 DEVICE QUALIFICATION FORM, 📄 REGISTER AS SEALED PDF, ⋯ 🖨 BLANK FORM (PAPER) and ⬇ TEMPLATES FOR YOUR IT), 👥 ACCESS (the users and the last access review; 👥 OPEN THE REGISTER, 📄 SAVE SEALED PDF, ⋯ blank form), 🧾 LOTS (every lot; ⤓ EXPORT ALL LOTS, 📄 PDF OF ALL LOTS), ⚕ BATCHES (the batch records of the compounding assistant; ⇩ CSV OF ALL), 💶 SALES (💶 OPEN THE SALES LOG), 🥦 CLUB / ☕ ROASTERY / 🏪 DISPENSARY (their reports as rows in cards; the dispensary with 📤 TO METRC as its main key) and 💧 PIPETTES (🗂 OPEN THE REGISTER, 📄 REGISTER REPORT). On a narrow screen (a phone) the keys beside the main key show only their sign (🔒 behind it when locked) - the name is in their tooltip.
📝 The log page
📝 WEIGHING LOG (in the menu, the tile of your view, OPEN on the log card of the weighing page, a tap on the sums below the weight or on NEW LOG ENTRY!) opens the log as a page of its own: its title and ✕ are in the head of the page (Esc closes it too), the ℹ️ behind the title explains it in short. Without any entry the app says so at the bottom of the screen instead. From the top:
The log
With more than one log a list at the top: LOG 2 · 14:05 · weighing - the number, the time it started and its kind (weighing, counting, coffee, factory, sorting, tracking). Pick one to see it (the sums below the weight follow). Beside it ➕ starts a new log and 📊 opens its statistics (when 📊 is not the big key at the bottom).
The lines
One row per logged line: the time, the number and the name at the left, the weight (counting: the count, the piece weight and the weight below it) at the right. Below a line in small: its units (price, nutrition values, other units), the rest on the scale (takeaway), 🍽 of the TAKE-FROM mode (the first weight waits in gray for the 2nd 🍽), 🧾 the lot, 🔖 the alibi number of the scale, 🌡 the room, 🌬 the true mass, 👤 who weighed it (with several users) and 📒 when it is in the diary. A line taken off the scale counts as minus. Then the ∑ row: how many lines, the total and the average (⌀).
The cards
TOTALS (total weight, average, the user, the volume and the sums of all other units such as costs), NUTRITION VALUES (total and per 100 g, with the 🚥 traffic light - for drinks per 100 ml, the switch 🍽 FOOD | 🥤 DRINK changes it), BAKER'S PERCENTAGES (2 or more different names; ℹ️ in its head explains the conversion factor), 🥗 RECIPE and 🧾 LOT NUMBERS after a recipe run, one card per run for coffee, factory, sorting and tracking logs, PORTIONS (the 🧮 servings calculator) and WITH THIS LOG: 📒 NUTRITION DIARY, 🍲 PROCESSED and 💾 SAVE AS THING (a log with nutrition values), 🧾 BATCH REPORT (after a recipe run), 📈 BREW CHARTS (a coffee log with charts). Each card folds with a tap on its head.
The keys at the bottom
📋 copies the log shown as text (the same text as always, to paste it anywhere) - with the result of the servings calculator if there is one. The big key: 📒 ADD TO DIARY for a log with nutrition values, else 📊 STATISTICS.
Every key works on the log that is shown. The page shows the texts in the language of the app (also after you switch it).
⚠ When the data stops: no weight, nothing recorded
The app never records a weight it doesn't know for sure. The scale sends its weight several times per second on the weighing and counting page and in every program with a live weight (factory filling, check-weighing, recipes and cocktails, the coffee timer, coffee weighing and pourover, density, solution and the ⚕ compounding assistant), also when nothing changes (at least every 0.25 seconds). If that stops for about 1.25 seconds - the scale is switched off, too far away, the battery is empty or the connection broke - the app shows – – – instead of the last weight (in the programs over their whole panel, whose buttons still work), a red line NO DATA FROM THE SCALE appears at the bottom of the screen (a tap explains it; its RECONNECT opens the connection again) and the link dot in the head turns red, and ∑, sales and labels with the live values are refused until fresh data come in. After a lost connection the same happens at once. Right after connecting, the scale is busy for a moment with the new app (settings and lists): for about 3 seconds after the first data a slowly turning ⚖ shows instead of the weight, then the current weight. Nothing is ever recorded with the "last known weight"; sales that were finished before are kept and sent when the scale is back. This is built to the loss-of-connection rule for weights sent to a computer (as described for WELMEC and NTEP). The UPCENT↑420 locks the weight only when the connection is lost. The heartbeat lock is part of ENTERPRISE & COMPLIANCE (also in the 14-day trial): on FREE and PRO scales the scale sends no heartbeat and the apps lock only when the connection is lost (the switch is greyed). Heartbeat Lock in the settings (tab GENERAL, card BASIC, below Reboot Scale) switches it: it is on by default and saved on the scale for all apps. Switched off, the scale sends its weight only when it changes and the apps lock only when the connection is lost - fine for a kitchen or a hobby, but keep it on for trade, sales and every kind of record. It is always on (the switch is greyed) with EXACT WEIGHT. While it is off (with strict audit), a red warning line at the bottom of the screen shows in the 🛍️ SALES MODE, the CCLUB 🥦 MODE and in the factory, check-weighing, sorting and density programs. Legal for trade is always the whole system with its type approval - the app alone doesn't make a scale legal for trade.
STABLE only says that the shown weight stands still. In a lab a weight is often saved a moment too early, while draft or vibration still move it inside one digit. Switch on Stability light in the basic settings (saved in this browser, off by default) and a line below the weight shows MOVING (red: not stable), SETTLING (yellow, pulsing: stable, but the scale still sees the weight move - the number is the spread of the last ~0.8 seconds in d) and SETTLED (green: the spread stayed at 0.5 d or less for 1 second). The scale decides this itself, so every connected app shows the same. The signal of a load cell is never perfectly still: SETTLED means within this limit, not zero. While the light is on, labels, the inventory check, the nutrition diary and taking a weight into a field wait for SETTLED: a window shows the light and saves by itself as soon as it turns green - or press SAVE ANYWAY (the activity log of the scale notes it with the weight and the user), or CANCEL. With the sound on, a short beep tells you the weight has settled. Needs the newest firmware; with an older one the app asks as before when the weight is not stable.
Add a name
Optionally you can add a name for your weighed product before you sum up, in that case it will be overtaken to the log-file to make it better readable later.
New logfile
Click ➕ beside the log at the top of the log page to start a new, empty logfile (the page says so until the next ∑ fills it). Your next weighings will be logged there, and the sums below the weight start from zero again. Another kind of weighing (e.g. counting after weighing) starts a new log by itself. The old logfiles are not deleted: select them in the list at the top at any time to see them again or to add more weighings to them.
Log Conversions & Units
All units/conversions will be logged and summed, too! Like that you will get not only an overview about what you have weighed, you also will see how much was the costs, ingredients or nutrition values for each of the weightings - and altogether!
Healthiness Info
If you count sugar, salt, fat or saturated fat, we will add a 🚥 Traffic Light Rating, to show you, how healthy your mix got (and how many percent of everything are inside)!
📒 Add it to the Nutrition Diary
If your log contains nutrition values (or metric/custom units like "THC"), the big key "📒 ADD TO DIARY" is at the right end of the key row of the log page. Only what you add with this key lands in your Nutrition Diary - your weighings for baking or meal prep do not end up there by accident. In the card WITH THIS LOG, 💾 SAVE AS THING saves the values per g of the whole log (all weighed lines together, e.g. your own muesli mix) as a new THING - a window asks for the name first (preset: its first line or the recipe; 💾 SAVE AS THING is gray while the name is empty).
Amounts as [B/M]AKERS PERCENTAGE
If you have weighed and logged 2 or more different products/things, then we will show you also the amounts of them as a Bakers Percentage Recipe! This is not only useful for baked-, but actually for all recipes. This way of recipe creation make it very simple to re-use your recipes later-on, for any batch sizes, and without any extra calculations for all the ingredients! Give it a try, you will love it!
Easy to use with our scale!
We also calculate the exact conversion factor for your recipe directly! With this you are able to calculate the amount of your initial 100%! (You want to make 1200g dough? Just divide 1200 by this factor!) This amount you now can add with the 💯 Percent Weighing Option to the scale, and then you can weight everything step by step with the given percentages! Without any further calculations for every single ingredient!
Hydration of the dough
If at least one weighed product has a 💧 hydration (from its THING, or the THING with the same name), 100 % is the total flour - also the flour in your levain or poolish - and the log shows HYDRATION OF THE DOUGH: the total flour, the total liquids (each with the part in preferments) and the true hydration in %. Without any hydration data, the biggest ingredient is 100 %. Every log line also shows the hydration of its product in %.
++ Portions 🧮 Servings Calculator
It is the card PORTIONS of the log page (weighing and counting logs with units or with 2 or more lines): type a number, choose what it means (servings, the weight of one serving, ml of one serving with volumes, or an amount of a logged unit - the label above says AMOUNTS FOR or SERVINGS OF) and press 🧮 (or Enter). The result shows in the card; 📋 copies it with the log. You can use the servings-calculator to calculate your portion amounts and sizes based on a weight, a number, on the amount of an ingredient - or by a price, for example! It also will show you the amount of all logged conversions/units for your calculated portion-size - ideal if you want to create labels for your products!
Portion Sizes / Servings can be calculated for:
An amount of servings
In this case, the total weight will be devided with your requested amount of servings, and the weight for each servings as like as the amount of all ingredients for each serving will be shown.
For a given weight
In this case we will show you how many servings you will get in total, as like as the amount of ingredients in one of them.
For an amount of an ingredient
Lets say you want to make 250-MG-Cookies. You have weighted 2325.43g dough with the scale, and you know, there are 13275.6 MG THC inside. Now, don't calculate everything by hand - just add 250 mg THC to the calculator! The scale will show you not only the exact amount of dough needed for each cookie, but also how many cookies you can make altogether - and even all other values (costs, nutrition values etc...) for each of the cookies! And ready is not only your calculation - but also your label! By the way... You can use every conversion unit you've added to your logging as a base for your calculation - including Milliliter, spoons, prices, calories, etc. pp!!!
The card PORTIONS shows up on the log page as soon as you have logged at least 2 weights in the weighing mode (also without any conversion units: portions by servings or by weight then) - or something with conversion units. Insert your desired portion size or amount, press 🧮 and your calculation shows in the same card.
The same calculation for a recipe before weighing it (from the amount you are going to make, optionally minus the water lost while cooking or baking - and with labels for one portion): ⎙ Portions & labels of your recipes.
📒 Nutrition Diary
What did you eat today, and how close are you to your goals? The nutrition diary keeps everything you ate - day by day, for each person of your family, with own daily goals. You open it with 📖 OPEN THE DIARY in the menu (☰ → 📒 NUTRITION DIARY, at the very bottom), in the weighing log - and, as soon as there is a diary in your browser, also directly on the start page (the row 📒 NUTRITION DIARY, without a connected scale).
The diary is a page of its own over the page you are on: its title is in the head, the small line above it shows the person you see. ✕ in the head (or the Esc key) closes it and brings you back. Below the head are the tabs 📒 DAY | 📅 HISTORY | 👪 PERSONS and a row with one chip per person - a tap on a chip shows the diary of that person. At the bottom are the keys of the diary: "+ ADD FOOD" at the right (in the day view) and "🔄 SYNC" (while a scale that passes on data is connected). The last card of every tab, "⤓ EXPORT", holds CSV, BACKUP and IMPORT.
Add food to the diary
Weigh your food with nutrition values on the screen and log it with (∑). Then click "📒 ADD TO DIARY" in the weighing log and choose:
What
All or only some of the logged weighings - as single entries, or summed up as one meal with its own name (e.g. "Lasagne"). Weighings that are already in the diary are marked with 📒 in the log, and the app asks before adding them twice.
For whom
For you, for another person - or shared between persons: give each person a number of servings (e.g. 2 and 1: two thirds and one third of the pot), and everybody gets the right share of all values.
When
Breakfast, lunch, dinner or snacks (chosen by the time automatically) - and the date and time, so you can also add things later.
Only one thing, without logging? In the weighing mode, "➕ ADD THE WEIGHT" in the menu (☰ → 📒 NUTRITION DIARY) adds what is on the scale right now - with the nutrition values of the page (shown units) - directly to the diary, with the same choices as above (for whom, meal, date and time). The button is greyed out as long as the scale is empty or the page has no nutrition values for it (load a THING with nutrition values or add them as units) - click it then for a short info.
Only what you really ate? With the 🍽 TAKE-FROM mode you weigh your plate before and after eating: only the difference gets logged, and the diary dialog opens with just this line.
You ate something without weighing it? Use "+ ADD FOOD" at the bottom right of the day: a window opens where you add the name and the amount in gram, then ADD. The values per 100 g can come from:
the food database (search with min. 3 letters), or type them from the package
📷 at the end of the search field: one camera for both kinds of codes. The barcode of a product gets its values from Open Food Facts (you can also type the barcode number into the search field) - and the QR code on the label of a THING loads that THING from your connected scale.
"⚖️ THING FROM THE SCALE" (only while a scale is connected): the window shows the list "Which THING?" (← goes back to the form) - choose one of your saved THINGs and its nutrition values are taken over. Its other units (like THC) are added with the amount, too. This only works while the scale is in the weighing mode; the scale switches back to it afterwards, and the THING on your weighing page stays as it was.
Close the window with ✕ - after you typed something the app asks first. Tick "★ Save to my foods" before you add it, and the food goes into your own food list (with its values and this amount): next time it is one tap, and the search of this form finds it first (★). ⭐ PRO, see quick add.
⭐ PRO: adding with one tap, new favourites (also "★ Save to my foods") and new meal templates; your saved foods stay in the search of the form and can be changed or deleted - gray in the free version; try it with the free trial (free and PRO).
Below the date of the day view, "⚡ QUICK ADD" logs food with one tap - for the person and the day you see:
★ My foods (favourites): your own food list in this browser, next to the THINGS of the scale. Add a food with the tick "★ My foods" in the ✏️ window of an entry (tap the entry; untick it to remove it again), or with "★ Save to my foods" when you add food without weighing. A tap adds it with its saved amount.
🍽 Meal templates ("My breakfast"): click 💾 next to a meal of the day and give it a name - all its foods are saved together. A tap adds all of them again, to the same meal.
🕘 Recent: the newest foods of this person (from the diary itself), with the amount of last time.
After a tap a line shows what was added: EDIT opens it to change the amount (all values change with it), UNDO takes it out again. ⚙ lets you change the amount of a favourite and delete favourites and templates (the diary itself stays as it is). Your foods and templates are shared by all persons and are part of the JSON backup of the diary.
Day view & daily goals
For every day you see the totals of calories, protein, fat, saturated fat, carbohydrate (also as BU, CU or carb choices, if set for the person), sugar, dietary fiber, salt (or sodium) and cholesterol, the meals with all entries, and a 🚥 traffic light rating of everything you ate. Each entry is a row: time, name and its values, the calories at the right and 🗑 at the end (it asks before it deletes). The day changes with ◀ ▶ around the date (tap the date for a calendar, TODAY goes back to today).
A tap on an entry opens the window "✏️ Edit entry": change the name, time, amount (and pieces), the meal, the person (SAVE moves the entry to that person), "★ My foods" and all values of an entry, including the other units: rename them, change their amount, remove them (empty amount) or add a new one. You ate only a part? Just change the amount - all values change with it (you can switch that off).
At the top you switch between the persons, and between the day, the history and the persons. In "👪 PERSONS" you add persons (with name and icon) and their daily goals - calories, protein, carbohydrate, fat and fiber - and daily limits for sugar, saturated fat and salt. Bars show you how far you are, the line on a bar is the goal: the bar is dark while you are below it, yellow near the goal (from 80 %, a limit from 90 %), green on it (90 - 110 %) and red above it (above a limit: red).
🍞 Carbohydrate units for diabetics
If you count carbohydrates for your insulin, choose your unit for each person in "👪 PERSONS": Bread Units (BU, 12 g), Carbohydrate Units (CU, 10 g) or Carb Choices (CC, 15 g, common in the USA). The grams per unit can be changed (e.g. 10 g per BU in Switzerland) - use what your diabetes team told you. The diary then shows the units of every entry, of every meal and of the whole day. If you follow a fixed meal plan, add your units per meal (e.g. 4 BU for breakfast) and you see "3.5 / 4 BU" at every meal.
Net carbs: check "Net carbs" to count the carbohydrates minus the dietary fiber. Use it only if the values of your foods count the fiber as carbohydrate - like the built-in food database (USDA) and US package labels. EU package labels and most values from the barcode lookup already show the carbohydrates without fiber: there the fiber would be subtracted twice. Fat-protein units (FPU): if you use an insulin pump with an extended bolus, check "Show fat-protein units": 100 kcal from fat and protein are 1 FPU (fat 9 kcal per g, protein 4 kcal per g). The diary shows them next to the carbohydrate units of every entry, meal and day, and in the CSV.
The diary only adds up - it does not calculate insulin doses. Please note: the carbohydrates of the built-in food database (USDA) include the dietary fiber, while EU package labels (and most products from the barcode lookup) do not. For foods with a lot of fiber (like wholemeal bread), the database shows more carbohydrates than the package.
Other units (like THC) & their daily limits
Besides the nutrition values, the diary also keeps your metric units (like "THC" or "caffeine" in mg - always saved in mg) and your custom units (by name and unit). They are shown at every entry and as "Other units" in the day totals and the history. In "👪 PERSONS" you can give each of them a daily limit (e.g. 30 mg THC) - a red bar and a ⚠️ in the history show when it was too much. Prices, volumes and percentages are not saved in the diary.
🎯 Points of your diet program
Does your diet program give your foods points and you a daily budget? Track them with a custom unit: load or create the THING of a food, add CUSTOM UNITS 🔣 with a name (e.g. "Diet points") and a short tag (e.g. "pt"), and choose as absolute weight: the grams of one point (a serving of 140 g with 5 points = 28 g per point). Save it with the THING (💾). Use the same name and tag for all your foods - the diary adds them up by name and tag. Then weigh the food and add it with 📒 ADD TO DIARY in the log or ➕ ADD THE WEIGHT (☰ → 📒 NUTRITION DIARY), and give "Diet points (pt)" your daily budget as its daily limit in "👪 PERSONS". The day then shows your points against the budget with a bar (red above it), the history a ⚠️ at the days above it.
How many points a food is worth comes from you - the app has no points formula of any diet program. It only takes your value per weight, works out the points of the weight on the scale and adds them up in the diary. Take the values from your program: programs change them from time to time and round them in their own way (the diary shows decimals, e.g. 4.6 pt where your program may say 5). Foods that count 0 points need no unit.
History
"📅 HISTORY" shows all days of a person with their totals, and the averages of the last 7 and 30 days. Click a day to open it. The list shows 90 days; "SHOW OLDER DAYS" shows 90 more each time.
📊 Week chart: at the top of the history, one bar chart each for calories, protein, carbohydrate and fat - a bar for every day from Monday to Sunday, the daily goal of the person as a dashed line, and the average of the days with entries next to the goal. The bars have the colors of the goal bars of a day: dark below the goal, yellow near it (from 80 %), green on it (90 - 110 %), red above it, grey without a goal. Hover a bar for the exact value, click it to open the day, and switch the week with ◀ ▶.
Your data stays with you
The diary is saved only in the browser on your device - no account, no cloud, nothing is sent anywhere. Because of that:
"⬇ BACKUP" downloads the whole diary (all persons) as a file. With "⬆ IMPORT" you can restore it - or move it to another device or browser. Entries that are already there are not added twice. A person of the file goes to the person of the same name here; a person whose name is not here becomes a new person.
"⬇ CSV" downloads all entries as a spreadsheet file (for Excel, LibreOffice, your doctor or your nutrition app) - with the carbohydrate units and one column for each other unit.
Deleting the browser data (or using a private window) deletes the diary too - so make a backup from time to time!
🔄 Synchronize between your devices
⭐ PRO: the synchronization - the licence of the scale the devices are connected to counts - gray in the free version; try it with the free trial (free and PRO).
Your family uses the scale with several phones or tablets? "🔄 SYNC" brings the diaries together - through the scale, without internet, account or cloud. Connect all devices to the same scale (they can be connected at the same time). Then click "🔄 SYNC" in the diary on one device. The app finds the other devices, and you choose the persons and what should happen:
Combine all devices
The entries of these persons from all devices get combined, then every device gets the same diary. For double entries (the same entry, or the same food with the same time and weight) the entry of the device that started the synchronization wins.
Overwrite the other devices
The diary of these persons on the other devices gets deleted and replaced with the one of your device.
Nothing happens without a confirmation: every other device shows what will happen, and a check code - confirm only if it is the same code as on the device that asks. There you also choose which of your persons gets which data (e.g. "Anna" of the other device is "Anna" here, "Klaas" is a new person - or skip a person). Persons that are not chosen stay as they are. After the first synchronization the persons are known on all devices and get found by themselves.
Only devices connected to your scale get the data (and if your scale has users, only logged-in ones can send). The scale only passes the data on and does not save it.
🍲 Batch processing: weigh the pot
You cook a big pot of chili for the whole week, bake a bread or mix a dough? Weigh and log the raw ingredients with their nutrition values (or run a recipe), cook, bake or mix - and then click 🍲 PROCESSED in the card WITH THIS LOG of the log page. Cooking and baking change the weight (water evaporates, rice and pasta take it up), so 100 g of the finished food have other values than 100 g of the raw mix - this window calculates them (ℹ️ behind its title explains it in short):
1. Raw ingredients
All lines of the log are checked - uncheck what is not in the pot (e.g. the water: it has no values, but it counts for the raw weight).
2. The pot
The pot, bowl or baking tin the food is in. Choose one of your saved pots, insert its weight - or put the empty pot on the scale and click ⚖ WEIGH EMPTY POT (with a name): its empty weight is saved in your browser for the next time. If you tared the scale with the pot, choose "No pot". 🗑 at the end of the row of a saved pot deletes it (the app asks first).
3. Processed
Put the pot with the processed food on the scale and click ⚖ TAKE THE WEIGHT (the pot is subtracted) - or insert the weight of the processed food.
4. Result
The yield factor (processed / raw, e.g. 0.82 = 18 % lost, mostly water), all values per 100 g processed with the 🚥 traffic light, and the split into containers: how much goes into each of them and what each one contains. Prices and other units (like THC) are spread over the processed food too.
The two keys at the bottom of the window are gray until card 4 has a result. ✕ or Esc closes the window - after a change the app asks first. When the weight is not stable yet, ⚖ asks whether to take it anyway (with the stability light it waits for SETTLED). ⚖ USE FOR WEIGHING switches the scale to the weighing mode (if it is not there yet) and puts the values per g of the processed food on the weighing page - with the pot as container, so the scale shows the food in the pot without the pot. Now weigh the portions into your containers, log them and add them to the diary, print a label with the EU nutrition table for each box - or use the 🍽 TAKE-FROM mode for your meal. 💾 SAVE AS THING does the same and saves it as a new THING on the scale (e.g. "Chili, processed" - change the name as you like) with all its units and the pot.
You know the loss already (e.g. 15 % for your bread)? Then you can calculate the portions of a recipe with it before weighing: ⎙ Portions & labels of your recipes.
➕ 1000 THINGS
You can save THINGS on the scale, to re-load them from there later. (What a THING can carry, with examples: THINGS & UNITS.)
With THINGS you can save your containers/bowls to remember their weight for later. Or add just a name and a price. But you can also save your storage tuppers/jars of food or weed, including their nutrition values and ingredients to check later how much you have without taring/unfilling/refilling everything first. (And to weigh directly out of that containers with the Takeaway Mode then). Or just save your common kitchen foods including their nutrition values like calories, proteins etc., to reload them later fastly! The options are endless!
A THING need a (unique) name, which you can enter on the page - and also at least one conversion/unit, like a price, nutrition value or a container weight. Then you just hit 💾 next to the name field of the THING / NAME card to save it permanently on the scale (without a name the field says so in red). Now you are able to load this THING at any time to any mobile/device, just click ▾ next to the name field and select the THING from the list (📷 next to it scans the QR code of a THING at once). The list (THINGS of the scale, ℹ️ behind its title) shows one THING per line, sorted by name, with LOAD and 🗑 (delete - asked in a window first) at the right; a click on the name loads it, ⏳ shows while a THING loads or is deleted. 📷 next to the filter loads a THING by its QR code. The filter on top shows only the THINGS whose name contains what you type. The lines appear while the scale is still sending them (with the turning scale and how many of them came): with many THINGS you don't have to wait - click one at any time. The list opens as its own page on top of the app; ✕ top right closes it. Fast and in the background: the scale stays where you are (weighing or counting) while the list and the THINGS come. While the app is open, it keeps the list and every THING you loaded (only in the memory of the app - nothing is saved in your browser): the next time it only asks the scale whether something changed - if not, nothing has to be sent again, and a THING you loaded before is there at once (also after a reconnect). After the app was closed, the list is read once again. The app asks only when you need the list (▾, 🗑, a THING select, a QR code ...), never just because you connected. Every THING you save or delete - on this or any other connected device - shows up on all connected devices at once (lists, selects).
You can save up to 1000 THINGS on the scale, each THING with a maximum of 20 Conversions/Units! A THING name can have up to 60 Latin letters - letters of other alphabets take more room: about 30 Cyrillic, Greek or Arabic letters, about 20 Chinese, Japanese or Korean characters (the name field stops there). If you save a THING with a name that is already used, it gets a number behind its name, like "Flour #2" (a long name gets shortened for it). If all 1000 slots are used, a new THING is not saved - delete one first. Edit a THING: load it, change, add or hide its units (keep its name) and press 💾 again - then a window asks: OVERWRITE the saved THING (it keeps its name and its place in the list, but gets only the units shown now), SAVE AS NEW THING or CANCEL. A note at the bottom of the screen says when it is saved, overwritten or deleted.
Several devices on one scale: if more than one device (e.g. your phone and your computer) is connected to the same scale, every device loads its own list and THINGS without disturbing the others. While a THING is saved (💾) or deleted (🗑), only that device talks to the scale; the other devices show a loading line ("Another device is saving or deleting THINGS on this scale …") until it is done - and get the change right after.
💾 Backup: at the bottom of the THINGS list (▾) you find ⬇ BACKUP and ⬆ RESTORE. BACKUP reads all THINGS with all their units from the scale and downloads them as one file. RESTORE saves the THINGS of such a file on this scale - or on an other one. You choose which THINGS, and what happens if a THING with the same name is already there: skip it, save it again (it gets a number like "Flour #2") or replace it. Keep the app open and the device close to the scale while it runs - it can take a while for many THINGS. With a second scale connected, ⇄ SYNC copies THINGS (and recipes) directly to or from it: see sync between scales.
Prices, cannabinoids and lots go with the THINGS (they are units of the THING). If the scale has a 🧾 lot store, you are asked once in the app's window (WITH THE LOTS / WITHOUT) whether its lots of these THINGS should go too - with number, date, status (on sale / locked), internal note and lab values. They are saved on the other scale with the same number and date (as an own number, made by the user of this app there); a lot that is there already for its THING is skipped, and a THING saved as "Haze #2" takes its lots with it. The other scale needs a lot store with PRO.
📦 Inventory of THINGS (bottles, jars, bags)
⭐ PRO: the inventory on the scale - in the free version 📦 STOCK is gray with 🔒 PRO and a tap tells you why; try it with the free trial (free and PRO).
Like the inventory scales of bars: weigh your open bottles, jars and bags and the app tells you what is left in them. Click 📦 STOCK in the THING / NAME card (or its tile in the views 🌿 cannabis, 🛒 shop and 📦 packing): the window 📦 INVENTORY opens - the list of your THINGS in stock; a THING, ✎ STOCK, ⚖ CHECK and 🚚 GOODS IN open as pages of the window with ← back (a page that weighs shows the weight at its top and has TARE at the left end of its keys). Leaving a form with something typed in asks first; ✕ (or Esc) closes the window. The stock is saved on the scale (up to 250 THINGS, next to the THINGS themselves, which stay as they are): every phone and every app connected to it sees the same stock, and a change in one app shows up in the others right away. To keep the connection quick, the app loads the whole stock from the scale only when it needs it: when you open 📦, when it books a THING it does not know yet, or right after connecting when LIMS forwarding of the stock is on. What the apps used since the last check and the last 20 checks (with the expected content and the loss) are saved on the scale too - so the loss is right even if several people pour from the same bottle. Without a connected scale you see the last state this app got from it, but you can't check or change it.
🍸 FOR BARS: the inventory scale of your shelf
Save your Gin as a THING with a 💰 price and the empty bottle as container unit, then add it in 📦: empty bottle 520 g (the container button fills it in), full bottle 1,190 g, 700 ml, 6 sealed ones in stock. Pour your drinks with the recipe and cocktail mode or weigh them into the log - every pour gets subtracted.
At the end of the shift press ▶ CHECK ALL: bottle on the scale, saved, bottle off, the next one - the whole shelf in a few minutes. You see the ml and % left of every open bottle, the total stock and its value - and the loss: what left the bottle without being poured with the app (over-pouring, free drinks, spills). Every phone behind the bar sees the same stock, because it is saved on the scale. Send the CSV to your office.
🌿 FOR CANNABIS SHOPS AND CLUBS: never lose your overview
Save your Haze as a THING with its 💰 price and the weight of the empty jar as container unit, print its automatically made 🏷 QR code label for the jar and add it to the 📦 inventory.
Now just put the jar on the scale, scan the code (📷) and switch on the takeaway mode: the scale shows you directly how much you take out - and its exact price. Print a 🏷 label for the bag right away, with the exact weight, price and all other values, and log the sale (∑).
When you check your inventory after a lot of sales, you directly see what has been sold and how much should be left - and how much has "evaporated" in the meantime (drying, crumbs, mistakes). All counters of your shop see the same numbers.
The list
One row per THING: ☑, the name with the date of its last check below it, the stock now at the right (the last check minus what was used since, plus the full ones) - red below 0 (more used than the last check had: time for a check), – when it was never weighed - and 🗑 at the end (it asks first). A tap on a row opens the THING: its stock (big), the last check (when, by whom, weighed or typed in), the full ones, the total, the value, the loss and what was used since - with the keys ✎ STOCK, ✎ CHANGE, ⋯ (🗑 DELETE) and ⚖ CHECK. The last checked come first (or sort by name), 20 per page with the page numbers below the list, and the field on top filters by name; + ADD A THING is next to it. Below the list it says how many are shown (e.g. 10 of 45 THINGS or Page 2 | THINGS 21 - 40 (of 324)). The keys of the window: 🚚 GOODS IN, ⋯ (⤓ EXPORT) and ▶ CHECK ALL.
+ ADD A THING
The name (best the same as the THING on the scale - the list suggests them; if a THING is loaded on the page, the form opens with its name), the empty weight (the bottle or jar with or without lid or pour spout, the way you weigh it; 0 for a bag without a container) - type it in, take the weight on the scale with ⤓⚖ at the right of the field, or take the container unit of the THING (the chip under the field shows whether it is with or without lid); the full weight has its own ⤓⚖. A missing or wrong value is marked red under its field - and optionally the full weight of a new one, its volume when full (e.g. 700 ml) and how many full ones (sealed) you have in stock.
What is left
Content = weight on the scale − empty weight. With a full weight you also get the % left, with a volume (or a volume unit of the THING) the ml. The full ones count with their full content: total = open one + full ones, and with a 💰 price unit of the THING you see its value.
⚖ CHECK / ▶ CHECK SELECTED / ▶ CHECK ALL
Check one THING (⚖ CHECK on its card), the ones you ticked in the list (☑ in front of the names, the box in the head ticks all that match the filter - the main key becomes ▶ CHECK SELECTED (n) and checks them in the order of the list), or all of them one by one: put it on the scale - it saves by itself once the weight is stable and close to the expected content (within 2 %, min. 2 g), or click ✓ SAVE NOW - take it off, the scale gets tared when it is empty and the next THING comes. While it waits you see the last check (when, by whom, weighed or typed in) and what was used in the app since then. Further off - less or more? It does not save by itself: it shows the weight, how much less or more than expected (g and %), what the expected content comes from, and asks - ✓ SAVE THIS WEIGHT or ⟲ WEIGH AGAIN. Tick "Always ask before a check is saved" in ⚙ INVENTORY SETTINGS below the list to confirm every check (saved in this browser). You can also correct the number of full ones while checking. The scale has to be on the weighing page. The page of the check shows the weight on the scale at its top (pale while it moves) and the result below it - less than expected orange, more red, as expected green; SKIP goes to the next THING, ■ STOP (behind ⋯, or ←) ends the check - the saved ones stay.
✎ STOCK
For a THING you can not weigh: type in the content left - in the weight unit, in % of a full one (needs the full weight) or in its volume - and the full ones. The app shows what it means (%, ml, total, the loss against the expected content) before you save. It is saved on the scale like a check, marked ✎ by hand: the expected content and the loss are calculated from it, and the used counter starts again. Every check keeps who did it (the logged-in user); with ENTERPRISE the activity log lists a stock typed in by hand with its amount.
Counting by weight (pieces)
For goods you count rather than weigh - small parts, candy, tablets, pre-rolls: with a piece weight a check also shows the pieces (the content / the piece weight, rounded) and the expected pieces, the question before saving looks at the pieces (other pieces than expected, or a weight between two counts, asks first), and the CSV has the columns Pieces and Piece weight (g). The piece weight comes from the lot (saved with it when a lot of the store is loaded), from the bag size of a THING of the fill weighing, or you set it in the check: type it, or put a number of pieces on the scale, type how many they are and take their weight with ⤓ TAKE in the 🔢 COUNT BY WEIGHT box - while the number is typed in, and until the pieces are off the scale, nothing is saved by itself. The app warns when the pieces are too light for the scale to count them surely - count by hand then. The stock itself stays in grams.
🚚 GOODS IN
Weigh a delivery against its delivery note - for shops, kitchens, roasteries (green coffee), labs, and for the transfer check of the DISPENSARY 🏪 MODE: the button 🚚 GOODS IN next to ▶ CHECK ALL. Type the rows of the note - the THING (the list suggests yours), the lot, the expected amount, the number of packages and the tare per package (at first the container of the THING) - with + ROW, or load them with 📥 FROM CSV (a file of your supplier; the app finds the columns by name). Then put the rows on the scale one by one, as in ▶ CHECK ALL: once the weight is stable the app shows the net (the weight minus packages × tare) and the difference to the expected amount in g and % (and in pieces with a piece weight), and saves it when it is close (within 2 %, min. 2 g) - further off it asks first. Every row adds its net weight to the 📦 stock and marks its lot as received (a new lot goes into the lot store, PRO). A row in several parts: + ANOTHER PART; too big for the scale: ⋯ → ✎ TYPE IN THE WEIGHT; ■ STOP (⋯ or ←) ends the delivery with the rows weighed so far - none weighed yet: back to your delivery note, nothing typed is lost. At the end a list of all rows - expected, weighed, difference, pieces, ✓ OK or ⚠ OFF - to print or to save as CSV (PRO, 🔒 in the free version); this browser keeps the last 20 deliveries (Earlier goods-in checks below the delivery note).
Loss
"Subtract from the inventory in" chooses in which modes a weight logged with the name of an inventory THING counts as used - tick any of them: 🛍️ sales modes (the bookings of the 🛍️ sales mode the dispensings of the CCLUB 🥦 mode and the weigh-outs of the DISPENSARY 🏪 mode - the only one on by default, so a weighing just to look does not count), ⚖ weighing modes (every ∑ of the weighing and counting mode - a kitchen, where everything weighed is taken out) and 📜 recipe modes (every recipe run, also cocktails, the 🧪 solutions and the compounding batches). The setting is saved on the scale (the same for every app) and is also in the block 📦 INVENTORY of the settings page, together with "Always ask before a check is saved". At the next weighing you see the expected content and the loss: what went missing without the app (over-pouring, spills, evaporation...).
⤓ EXPORT
⋯ → ⤓ EXPORT (PRO): choose the file format (ENTERPRISE: the presets, JSON or your own settings with ⚙), then ⤓ SAVE. The inventory as a table for your spreadsheet: last weighing, content, %, volume, full ones, total, value, expected and loss of every THING, who checked it and whether it was weighed or typed in by hand.
ℹ️
The options of the inventory and the sales log (what is subtracted, always ask, full ones, the time and the grouping of the sales) each have a small ℹ️ that explains exactly what they do.
Several apps and users
Every app adds what it logs to the same used counter on the scale, and a check in any app starts it new for all. The scale calculates the expected content and the loss of every check, so every phone shows the same numbers - also for the checks the others did while this app wasn't connected (each app keeps up to 100 checks per THING, the scale the last 20).
💶 Sales Log: every sale and the daily report per THING
Click 📦 STOCK: the window "Inventory & Sales Log" has the tab 💶 SALES LOG under its title (it only shows up if your scale has the sales report). There you see what you sold per THING, day or user - or every single sale with its time - today, the last 7 days, this month, this year or your own time range. Every booking with BOOK NOW in the 🛍️ sales mode (and the sales of the CCLUB and ROASTERY modes) is added up on the scale (not in your browser) - under the name of the THING: all phones, apps and users of this scale see the same report, even after clearing the browser. A ∑ in the weighing or counting mode is no sale (it stays in the weighing log and the inventory), so no accidental weighing gets into the sales log; recipes and the other modes are not counted either. The 📦 inventory still counts only THINGS.
🛒 YOUR SALES COMPUTER: shop, market stall, bar or club
With the THINGS (name, price, container weight), the QR code labels, the takeaway mode, the price labels, the 📦 inventory and this 💶 sales log your scale also does everything a small shop needs (on top of all its other modes) - without an extra cash register, without a PC, without a monthly fee:
Weigh: scan the QR code of the jar (📷), put it on the scale, the scale shows the weight and the exact price (and the costs, if you saved them as "Costs" unit).
Sell: in the 🛍️ sales mode press BOOK NOW and confirm: the sale is logged - and saved on the scale with its time, THING, lot, weight, price and costs.
Label: print the label for the bag with weight, price and barcode.
Report: 📦 → 💶 SALES LOG: pieces, weight, revenue, costs and profit with margin - per THING, per day, per user or per member ID (as soon as there are sales with one); today, this month, this year or any time range - or the list of every single sale with its time. Print it or save it as CSV for your accountant.
Chained and checkable: the scale gives every record its own number (never used twice) and saves it with the SHA-256 fingerprint of the record before it, and it checks the weight against its own reading before it saves it. On an RS-232 dongle a checked weighing also gets the alibi number of the verified scale (EXTRA LINE) - the chain covers it too. A changed or inserted record no longer fits the chain, a deleted or lost one leaves a gap in the numbers - the 🔏 sealed report (ENTERPRISE & COMPLIANCE) shows both, and its sealed file can be checked by anybody on up-scales.com/check (in the browser, nothing is uploaded).
Stock: in the same dialog the 📦 inventory shows what is left of every jar and bottle, what has been sold and what got lost.
All of it is saved on the scale (every single sale, up to 366 days): every phone, tablet and counter sees the same numbers, the history survives clearing a browser or changing the phone - and the scale warns you when its memory gets full.
Please note: the sales log is your own record and overview. It is not a certified cash register or fiscal device (even if it does the same! ;-)), and the app itself is not verified - for receipts, price labels and your bookkeeping check the rules of your country (see What stays legal).
What is counted
Pieces (a weighing counts as 1, in the counting mode the counted pieces), weight and revenue from the price unit of the THING (🪙 or a currency). If the THING also has a "Costs" unit (💰), the report shows the costs, the profit and the margin too - the app takes both as the same currency, and calculates the profit only for sales that had costs.
Mode
Several sales modes on one scale (e.g. the shop counter and the roastery, or a club that also sells to guests): every mode books its own kind of sale, so the sales log can show All sales or only the ones of one mode (🛍️ Shop, 🥦 Club, ☕ Roastery, 🏪 Dispensary) - the select appears as soon as more than the shop is used. The CSV and the printout follow the choice (the file name says which mode). The 💶 SALES button of a mode opens the log with its own sales.
Single sales
Choose Group by: Single sales to see every sale with its time, THING, weight, price, costs and profit (the latest 300 on the screen - print and CSV have all of them). Every sale has its own number, so a sale is never saved twice, even if the wireless connection drops.
🖨 PRINT / ⤓ CSV
The keys at the end of the tab: ↻ (load again), ⋯ (⤓ CSV, LEDGER CSV, 🔏 SEALED REPORT, the export format, the delete rows) and 🖨 PRINT. The report as a printout (with your company header, if you have one) or a table for your spreadsheet: one line per day and THING - or, with Single sales, one line per sale with date, time and sale number.
Storage
The scale keeps the last 366 days, up to 300 different THINGs. It saves a new sale about 5 seconds after the last one, so don't pull the plug right after a sale. The scale does not have a clock: the day is the date of your phone. The name of a THING is freed again as soon as no saved sale uses it any more. A day can hold up to 2000 sales. ✕ DELETE SHOWN (⋯ of the tab, red, asked first) deletes the days of the shown time range on the scale, ✕ DELETE ALL clears the whole report.
⚠ Memory getting full
THINGS, recipes, the inventory and the sales share the memory of the scale. If less than 20 % of it is free, the app warns you when you save a THING or a recipe and in the sales log (⚠ on the tab): download all sales with ⤓ CSV (EVERYTHING), then delete the old days (Own range, ✕ DELETE SHOWN) and the THINGS or recipes you don't need.
🔒 Protection against deleting
So that nobody (an employee, a helper) can delete your sales or your stock, the ENTERPRISE & COMPLIANCE edition has users: the ADMIN switches deleting only by an ADMIN on (it needs an ADMIN with a password), and every sale is saved with the name of the user who was logged in. The former separate delete password can't be set any more; one that already exists still works for deleting and can be removed here (REMOVE PASSWORD). If a password is lost, nobody can read or recover it - for a forgotten ADMIN password another ADMIN sets a new one; if the only ADMIN forgets it, only flashing the firmware completely (update page) helps, which erases ALL data of the scale.
Several apps
If the wireless connection drops, the app keeps the sale and sends it again after reconnecting - the scale counts every sale only once.
SALES 🛍️ MODE: the sales display (shop, market stall, bar)
⭐ PRO: the sales mode and the sales log - gray in the free version; try it with the free trial (free and PRO).
The Sales 🛍️ Mode
Your scale as a sales display: big weight, the price highlighted, BOOK NOW with a confirmation - every sale saved on the scale with time, THING, lot, weight and price. The live units of the THING and the labels stay as in the weighing mode. Cannabis social clubs use the CCLUB 🥦 MODE, coffee roasters the ROASTERY ☕ MODE, US dispensaries the DISPENSARY 🏪 MODE (Metrc) - the same screen with the bookings of their trade. All of them work next to each other on one scale.
Choose SALES 🛍️ MODE in the menu (☰ → ⚖ MODES) (it stays marked there as long as the sales mode is on). Several devices: when another phone or computer connects to the scale while a device is in the sales mode, it joins it by itself and lands on the same screen with everything loaded: the THING with its units and lot, the customer ID and the settings of the sales mode; the currency too, if none is chosen on the new device yet. Nothing is sent to the scale for it, so the device that is selling is not interrupted. The scale weighs as in the weighing mode, but the screen becomes a sales display: big weight, the price highlighted, everything else that you don't need for selling (the THING / NAME field, ∑ LOG IT, the UNITS bar, 📦 STOCK) is hidden.
The page: below the weight and its units one card - THING, LOT and (with customer IDs) MEMBER - and the tiles RECENT at the end. The key row at the bottom: TARE at the left, BOOK NOW at the right with the live weight and price in its second line; ⚙ in the head opens the settings, ✕ beside it leaves the mode and goes back to the weighing mode (while the confirmation or a sheet of the mode is open it asks first: CANCEL | LEAVE - nothing is booked). The same ✕ is in the head of the CCLUB, ROASTERY and DISPENSARY modes. While something is missing, BOOK NOW is gray and a line above the keys says why (a tap on it says it too); the warnings of the sales log (days dropped soon, the memory of the scale) are lines there too, with ⤓ CSV.
THING: choose it from the list (alphabetical) or 📷 scan its QR code or a lot code. Your last 12 THINGS are tiles at the end of the page (name and price) - one click loads them again. With several lots on sale, choose the lot below the weight.
BOOK NOW opens the confirmation (also with nothing on the scale). BOOK NOW in that window (CANCEL beside it) takes the stable, positive weight of the scale - or a ✍ manual weight you type in there, e.g. for a pre-packed bag: it replaces the weight of the scale for this booking (log, inventory, price, sales log), is marked as manual in the record, the activity log and the sealed report, and prints no automatic label. Not with an RS-232 scale in EXACT WEIGHT. A weight of zero or below zero (empty plate, tared wrong, container too heavy) is never booked, also not with pieces: the scale itself refuses a zero or minus weight, a minus price and - when its own weight is booked - less than one digit on the plate. First comes a confirmation with THING, lot, weight and price - you can change the price per weight unit or the total for this sale (also 0, e.g. for a club). The UP↑SCALE works out the amount of the sale from the price and the weight it books, rounded to the cent. Auto print prints the label right after booking.
Every booking goes into the weighing log, the inventory and the sales log on the scale - with time, THING, lot, weight (shown with the d of your scale: g, mg or µg, kg), price, the logged in user and the member ID.
The scale checks the weight: when a booking arrives, the scale compares it with its own stable reading (minus the container of the app) and saves its own value - the app only proposes it. The record also gets a number of the scale (1, 2, 3 ... never used twice), the tare (T weighed on the scale, PT the container of the app), the unit price, the PLU (a unit named PLU in the THING, e.g. 4011) and the calibration and configuration counters. See the sealed report.
Currency: choose the currency of the sales log once: ⚙ in the head of the sales mode → SETTINGS OF THE SALES MODE, or the row Currency in the basic settings of the scale (sales are blocked until then). It is saved on the scale, so every phone and computer of this scale books in the same currency, and the scale refuses a sale in another one. A THING can have prices in several currencies: only its price in this currency is booked (the "Costs" unit counts in it too), and every record keeps its currency (ISO code in the sealed report). Change it only when you start a new sales log: the sums don't convert between currencies.
THING required (on) and Member ID required (off) in the settings (card REQUIRED) stop bookings without them.
🧾 Price of a lot: a lot of the lot store can have its own price (on its card). While this lot is chosen, its price replaces the price of the THING on the screen, on the labels and in the confirmation (marked 🧾); the tiles keep the price of the THING.
The label line stays at the bottom (live label generator, save, print; the label printers are set up in the generator).
The sales display warns you before old sales get deleted (the scale keeps 366 days with sales) and when the memory of the scale gets full - with the button to download everything.
The sales log of the shop: 📦 → 💶 SALES LOG.
⚙ Settings of the sales mode (⚙ in the head: a window - the currency, customer IDs, REQUIRED, the member cards; ℹ️ the privacy note):
Customer IDs (off by default - e.g. for regular customers or customer accounts): a field and a scanner for the customer ID. Only IDs are saved, never names. As soon as an ID is there, its sales of the last 3 months are shown (folded): every sale with lot and weight, the sums of today, of this month and in total.
🪪 Create member cards: one ID, a list (one per line) or a range like M-0001 - M-0050 - each card has a QR code of the ID, the ID in big letters, your business and an optional line (e.g. "valid until"). Print one as a label, or all of them on A4 sheets with credit card sized cards (85.6 × 54 mm) to cut out and laminate, or have them printed on PVC cards. Your own member cards work too: any QR code or barcode whose content is the ID (Code 128/39 where the browser can read it).
Next to the other sales modes: the SALES MODE stays in the mode list also when the club, the roastery or the dispensary mode is available. Every mode books its own kind of sale, and the scale checks each booking with the rules of its own mode: a sale of the shop needs no member ID or THC / CBD, a dispensing of the club still does. The reports of the club and the roastery take only their own bookings, the 💶 sales log can show all sales or the ones of one mode, and a customer ID of the shop is never counted as a member of the club. The 📦 inventory is one per THING - use different THING names when two modes sell different goods.
Please note: the sales mode and the sales log are a documentation aid - please check your record keeping against the rules of your country. The customer IDs are kept in the memory of the scale without encryption: export the sales log regularly (with ENTERPRISE & COMPLIANCE as sealed files, see tamper-evident records) and keep the list of who has which ID apart.
🔏 Sealed Report: every record for an inspection (ENTERPRISE & COMPLIANCE)
The button 🔏 SEALED REPORT in the sales tab (for the period chosen there) and in the CCLUB 🥦 MODE and 🗂 RECORDS → 🥦 CLUB (this year) collects every record of the period - the sales and the other bookings (harvest, dried, destroyed) - in the order of the numbers the scale gave them. The records of other scales loaded in this browser follow, each scale with its own checks. As a sealed PDF (the JSON of all records inside) or as CSV / JSON with the UP↑SCALE SEAL.
The report of one mode: started from the CCLUB, the ROASTERY or the DISPENSARY mode (or their tab in 🗂 RECORDS), the PDF lists only the bookings of that mode and says how many other bookings of the scale are in the period. The checks on top (chain, missing numbers, counters) are still made over all records - a club report can never hide a gap by leaving out the shop sales - and the data file inside the PDF and the CSV / JSON keep all records with their mode (column trade), so up-scales.com/check can check the whole chain again.
Per record: number of the scale, record ID, date and time (also as UTC, from the time zone of the app clock) with the time check (column time_check: scale = the scale compared it with its own clock, differs = more than 5 minutes off, marked ⚠ in the report - then set the clock, best with the signed exact time, see clock guard), Device ID, kind, THING, PLU, lot, member ID or note, user, gross, tare with its type (T / PT), net in the unit of the sale and its d, pieces, unit price, total, whether the scale checked the weight, the calibration and configuration counter at the booking, THC / CBD (mg), the chain and the alibi number of the verified RS-232 scale (column alibi, also in the PDF next to the THING).
The checks on top: the chain of the records (each record holds the SHA-256 fingerprint of the record before it, together with its own alibi number: a changed, inserted or removed record - or a changed or removed alibi number - breaks it), missing numbers (deleted or lost records - the activity log shows the deletion), records the scale could not check (sent again later), and every change of a counter inside the period (between which records and on which day).
The instrument: firmware, the counters now and the certificate data you typed in once. They are also printed at the bottom of every other report (self-check, sales, lots, tracking ...).
Built along the record points of NIST Handbook 44 (audit trail with event counters for calibration and configuration), OIML R 76 (gross / tare / net, tare marked T or PT) and WELMEC Guide 7.2 (event logger; the legally relevant weight comes from the scale and the record is protected). Built to them - not certified, and our own scales have no type approval.
Not a fiscal export: no DSFinV-K (Germany) or SAF-T: they belong to a certified cash register with its own security module. No payment method and no tax - your cash register stays separate and can take the CSV / JSON.
Saved on the scale, for every device: the currency, the settings of the sales mode (customer IDs, "a THING is needed", "a customer ID is needed"), the cannabis club settings of the CCLUB 🥦 MODE (on / off, the limits per member - not the ones for members under 21, they stay in each browser), the roastery settings of the ROASTERY ☕ MODE (on / off, batch number and green lot needed, the coffee tax with its rate and currency), the dispensary settings of the DISPENSARY 🏪 MODE (on / off, a customer ID needed, harvest by batch, the state, the tag prefix of the facility and the scale variance), what the inventory takes as used, the pattern of the lot generator and the company header of the reports (its first line is the name of your business or club) are saved on the scale and sent to every phone and computer that connects. A change on one device is at once the setting of all of them. The scale also checks every booking against them and refuses one that misses something (currency, member ID, lot, THC / CBD, the reason of a destruction or waste, the name of the club, the batch number of a roast, the plant tag of a harvest): another browser or an old app can't go around them. With users, changing them follows the rules of the settings (ADMIN gate, not by an operator, a reason in the strict audit mode) and is noted in the activity log. Only the automatic printing of the confirmation stays a setting of each device (every device has its own printer).
CCLUB 🥦 MODE: dispensing, harvest, drying and destruction for cannabis social clubs (PRO)
⭐ PRO: the CCLUB mode - gray in the free version; try it with the free trial (free and PRO).
The CCLUB 🥦 Mode
Made for cannabis social clubs: the app knows the record-keeping duties of the KCanG (§ 26) and helps you keep the documentation it asks for. The dispensing to the members on one tab, harvest, drying and destruction on the other - all linked by the lot number, with THC / CBD, saved on the scale. It works like the 🛍️ sales mode (live price and units, labels, manual weight, several devices), and all club settings are in the mode.
Choose CCLUB 🥦 MODE in the menu (☰ → ⚖ MODES). There is no switch: the mode is in the menu when your licence is PRO and your view has it (the cannabis view or the full view). The first time, open ⚙ CLUB SETTINGS (⚙ in the head) and enter the name of the club (it is also the first line of the report header and of the lot codes) - until then BOOK NOW says that it is missing. The window holds all club settings: the name, the currency, REQUIRED, THC / CBD (the way to the 🌿 CANNABINOIDS window), the member cards and the limits; its ℹ️ the legal and privacy note. They are saved on the scale for every phone and computer; the scale checks every booking of the club for the club records (the SALES MODE and the ROASTERY MODE stay available: their bookings have their own rules). Without PRO, or in a view without the mode, the page shows only the card 🥦 CCLUB MODE. The box 🥦 CANNABIS CLUB on the settings page only opens the mode; with a free licence, or a view without the mode, it shows the PRO hint instead. When another device connects while one is in the CCLUB mode, it joins it on the same tab.
Two tabs:🌿 DISPENSING books the dispensing to a member (member ID, its dispensings of the last 3 months, the limits, the price). 🌱 HARVEST & STOCK has the kinds 🌱 FRESH (harvest), 🍂 DRIED, 🗑 DESTROYED FRESH and 🗑 DESTROYED DRIED as a row of chips (with what the chosen kind books below them), then the card of the kind with THING, lot and the NOTE / REASON field - no member (the stock itself is counted in the inventory). The tabs stand right under the weight; the card says which kind you book. The key row: TARE · 📋 the club reports · ⋯ (🗂 CLUB RECORDS, 🛍 FILL WEIGHING, ⇄ OTHER SCALES) · BOOK NOW. Every booking is weighed on the scale with a stable weight, confirmed and saved on the scale - with time, user, THING, lot number, weight (by d) and a note. A harvest (fresh) has no price and no THC / CBD. Dried takes THC / CBD and a price if you have them (both optional); once it is saved the app asks whether to add the weight to the stock of the inventory (a THING that is not there yet is created). A destruction needs a reason (note field, e.g. mold). Book fresh material that is destroyed before the drying (mold, pests, male plants) as destroyed fresh: it has no THC / CBD, it does not come off the dried stock or the rest, and the drying loss is counted without it. Dried goods that are destroyed are destroyed dried: they come off the rest and are taken out of the inventory like a dispensing. The activity log names each kind (e.g. "Harvest weighed (fresh)").
+ NEW STRAIN and + NEW BATCH (tab 🌱 HARVEST & STOCK, beside the THING and beside the lot) - before the first harvest, without going to the THING editor: + NEW STRAIN opens a short form with the name, the dispensing price per g (the price unit in the currency of the sales log), the costs per g (the "Costs" unit in the same currency, optional), THC / CBD in % (optional - the dispensing needs a value; the lab values of a batch replace it later), the first batch (lot number, proposed by the lot generator from the name) and the 📦 stock of dried goods that are there already (optional, g). SAVE ON THE SCALE saves the strain like + ADD, puts the batch into the lot store and loads the strain on the page. + NEW BATCH makes a further batch of the strain on the page: number from the lot generator (or your own), optional lab values THC / CBD and a note (e.g. plants, tent) - and chooses it at once.
The club reports are behind 📋 of the key row (in groups: logs, statistics, sales & proof - the last one used is marked), and in 🗂 RECORDS, tab 🥦 CLUB (only with the club settings on; 🗂 CLUB RECORDS in ⋯ leads there). Every report is a window: the lines at the top, the rows with the weight at the right (a tap shows the other columns; the window shows the newest 200 bookings and says so - print, PDF and CSV have all), the keys CANCEL · ⋯ (PDF, CSV) · 🖨 PRINT. 🌱 Harvest log: only the dispensing is a sale - the sales log and its exports show the sales only. Harvest, dried and destroyed are in the HARVEST LOG, grouped by THING and lot. A harvest is usually weighed in several parts: tick the weighings of a lot (all are ticked; ONLY FRESH / ONLY DRIED / ONLY DESTROYED (fresh and dried) / ALL with one click) and make one report of them - the fresh harvest, later the dried buds of the same lot, or both together with the drying loss in %. The drying loss is (fresh - destroyed fresh - dried) / (fresh - destroyed fresh). The 🗑 DESTRUCTION LOG lists every destruction of a year over all THINGS and lots, fresh or dried, with the reason and the sums (fresh, dried, all).
⇄ Other scales: harvest and drying on one scale, dispensing on another (or green coffee on the floor scale and the bags on the table scale of a roastery)? Every scale keeps its own bookings - they are never copied into the log of another scale. On the harvest scale, the button ⇄ OTHER SCALES (in the club or roastery reports) → 📤 SAVE TRANSFER FILE saves all its bookings as one file (on an ENTERPRISE scale sealed by the scale). In the app of the dispensary scale, 📥 LOAD TRANSFER FILE checks it (seal, chain and gaps of the numbers, lot and note texts against the records) and keeps it in this browser: from then on the reports of the club (lot report, harvest log, destruction log, sums of a year) and of the roastery (roast log, green ledger, batch trace, overview, statistics, coffee tax) show the bookings of both scales together - one file holds all bookings of a scale, each report takes the ones of its mode, each row of the other scale with its name. Load the newest file any time - bookings already there are joined, nothing is counted twice. Every other phone or computer has to load the file once too; ✕ REMOVE takes it out of the browser only. The 🔏 SEALED REPORT of the dispensary scale lists them too: its own records first, then each other scale with its own checks (chain and gaps of its numbers, counter changes, the seal of its transfer file) and its Device ID in every row; the transfer files are inside the PDF, so their own seals can be checked on their own. The seal of the report is made by the connected scale over the whole report - the records of the other scales stay records weighed and numbered by those scales. Lots and THINGS go to the other scale with the backup / sync of the THINGS; the same THING name and lot number connect the bookings.
🛍 FILL WEIGHING (⋯ of the key row) - pre-pack portions into bags or jars out of a strain, with the box, AUTO TARE and the live statistics, own − / + tolerances (e.g. −0 / +0.1 g; the ℮ rules start at 5 g). FINISH moves the filled weight in the 📦 stock from the strain to the portion THING with the lot of the batch - no dispensing is booked. All details: 🛍 fill weighing.
The lot links everything, also weeks apart: create the lot at the harvest (lot store), print its label, and scan it again when you weigh the dried flowers or dispense - the lot is chosen again and all bookings belong together. Several containers of one lot are several bookings that add up. Weigh the net weight (tare the container first).
THC and CBD: in the units of a THING the window 🌿 CANNABINOIDS (+ add a unit) takes the content in percent from your lab report (the THING keeps it as milligram per gram, 1 % = 10 mg/g) - THC and CBD on top, below CBG, CBN, THCV and more. Only THC and CBD go into the records (one of the two missing counts as 0); the other substances are normal units for the live display and labels, and you can add any other substance in the box INGREDIENTS (METRIC) above. A lot can have its own lab value: on its card in the lot store enter THC / CBD (in %); it wins over the value of the THING, is shown on the page of the lot code and goes into the records as it was when you booked. While that lot is chosen (also when you switch between several lots on sale below the weight), its values replace the THC / CBD of the THING in the boxes below the weight, the % beside them and on the labels, marked 🧾 - so the screen, the label and the booking always show the same value. The THING keeps its own values (saving it with 💾 saves them, not the lot's); a lot without a lab value shows the THING's own values.
The check before every booking: the club name and THING are needed, and the lot number; for a dispensing also the price, the member ID and THC / CBD. If something is missing the button stays blocked and says what.
Required: ☑ Lot number | ☑ Member ID (⚙ CLUB SETTINGS, card REQUIRED): both are on at first. A club that does not need the full documentation switches them off - then a booking without a lot, or a dispensing without a member ID, is possible. The switches are saved on the scale (the same for every phone and computer), and the scale checks them at every booking. The reports of a lot and the limits per member need them, of course.
History of a lot: on the lot card the button LOT REPORT shows the harvest, drying (with the loss in %), destruction, dispensing and the rest (dried - dispensed - destroyed dried) and every booking. 📅 SUMS OF A YEAR adds up one calendar year per THING: harvest, destroyed fresh, dried, dispensed with its THC and CBD, destroyed dried and the rest - as a help to fill in the report you hand in once a year. The reports give THC and CBD as the amount in the weighed goods (mg, for large sums g or kg - one unit for the whole report), not in %.
Limits per member (⚙ CLUB SETTINGS, off by default): limits per day and month in g. Above a limit the confirmation of a dispensing shows red lines above its keys - only a warning, booking stays possible; the limits are yours, please check them against the law.
Members under 21 (with the limits on): the German KCanG (§ 19 (3)) allows members from 18 to under 21 less per calendar month and only cannabis with at most 10 % THC. Tick ☐ Under 21 below the member ID of a dispensing: for this member the confirmation then warns with the limit per month, members under 21 (default 30 g; the day limit stays the same) and when the THC of the lot or of the THING is above THC at most, members under 21 (default 10 %) - the same red lines, again only a warning. Only the member ID is kept (no birth date): untick it when the member turns 21. The marks and these two limits are kept in this browser only, not on the scale - mark the member on every device you dispense with.
🪪 Member cards (⚙ CLUB SETTINGS): the same cards as in the sales mode - a QR code of the ID, never a name.
Print and export: print and a plain CSV are PRO; JSON, your own format, the sealed CSV and the sealed PDF are ENTERPRISE like the other exports. Their header names the club, the scale and the logged-in user; see tamper-evident records for what the seal does and does not mean.
Please note: this is a documentation aid, not legal advice. Please check with the law (KCanG § 26) and your authority what you have to record and report. The law asks for the name, first name and year of birth of a member: the scale saves the member ID only, so keep your member list with the names yourself. The records can be checked for changes - keep the exported files safe.
ROASTERY ☕ MODE: green coffee, roast batches, blends and sales for coffee roasters (PRO)
⭐ PRO: the roastery mode - gray in the free version; try it with the free trial (free and PRO).
The Roastery ☕ Mode
Made for small coffee roasters: green coffee in, the green charge into the roaster, the roasted coffee out with the roast loss of every batch, sorted out, samples, waste, blends and the sale of the roasted coffee - all linked by the green lot and the batch number, saved on the scale for every phone of the team. It works like the 🛍️ sales mode (live price and units, labels, manual weight, several devices) and runs next to the other sales modes on the same scale.
Choose ROASTERY ☕ MODE in the menu (☰ → ⚖ MODES). The first time, the page shows only one card below the weight: switch on the roastery for this scale and enter the name of the roastery (the first line of the report header and of the lot codes; ℹ️ behind the title says what the mode does). Both are saved on the scale for every phone and computer, and the scale checks every booking of the roastery with its own rules. The page: the head shows ☕ ROASTERY and the THING on the page, ⚙ in the head opens ROASTERY SETTINGS (the switch, what is required, the currency, the roasting costs, customer IDs and their cards, the coffee tax - every change is saved at once; ℹ️ behind its title: the legal and the privacy note). The key row: TARE | 📋 the reports | ⋯ 🗂 RECORDS, 🛍 FILL WEIGHING, ⇄ OTHER SCALES, ℮ PACK CHECK | BOOK NOW with the live weight and price below it (gray while something is missing - the line above the keys says what, a tap too). The reports are also in 🗂 RECORDS, tab ☕ ROASTERY. The THINGS you used last are tiles at the end of the page. Save your coffees as two THINGS: the green coffee (e.g. "Guji green") and the roasted coffee with its selling price (e.g. "Guji").
Two tabs below the weight:☕ SALES books the sale of roasted coffee (one card: THING with 📷, its batch as the lot, an optional customer ID with 📷 and ✕ and its sales of the last 3 months, the price) - exactly like the sales mode. 🔥 ROASTING has the kinds as a row of chips (it scrolls sideways) with the hint of the chosen one below: 🌱 GREEN IN (green intake: the sacks of a green lot, optional moisture), 🔥 ROAST IN (the green charge before the roast), ☕ ROAST OUT (the roasted coffee after cooling), 🪨 SORTED GREEN (defects of a green lot), 🫘 SORTED ROASTED (quakers, stones of the destoner), 🥄 SAMPLE / OWN USE (cupping, samples, the coffee of your own café), 🗑 WASTE (a reason is needed), 🔀 BLEND (the blend view below) and 📤 GREEN OUT (green coffee returned to the importer or sold green, price optional). Every booking is weighed with a stable weight (or a ✍ manual weight), confirmed in a window (CANCEL | BOOK NOW) and saved on the scale with time, user, THING, lot and note. The form of the chosen chip is a card on the page (the weight stays on top, TARE is the key at the left): THING, lot, the fields of the kind and the note (its label says what it is for: supplier, reason ...). + NEW beside the THING makes a new green or roasted coffee in a window.
Units: the roastery mode follows the weight unit of the app. With grams (or kg) the amounts are shown in g below 1 kg and in kg from 1 kg, prices, costs, the roasting costs and the coffee tax rate per kg, small amounts per 100 g; with ounces or pounds in oz below 1 lb and in lb from 1 lb, everything per lb, small amounts per 1 oz. The totals you type (blend, stock, planner) are in kg or lb. The CSV exports have their columns in kg or lb (named _kg / _lb). The scale keeps the coffee tax rate and the roasting costs per kg - switching the unit never changes them.
Green and roasted coffee are THINGS: one THING per green coffee (e.g. Ethiopia Guji green; each delivery is a lot of it) and one per roasted coffee or blend (e.g. My Super Coffee (Blend) #3; each roast is a batch = a lot of it). Use different names for the green and the roasted coffee. You can make them in the usual THING editor, or right in the mode: + NEW GREEN COFFEE (on GREEN IN), + NEW ROASTED COFFEE (on ROAST OUT) and + NEW BLEND (on BLEND) open a short form - name, selling price per kg (the price unit in the currency of the sales log: what you sell it for), costs per kg (the "Costs" unit in the same currency: what it costs you; for green coffee what you pay incl. freight), the lot (green: the lot of the delivery, with its origin / EUDR data folded below; roasted: the batch of the roast out; blend: a new blend number) and the 📦 stock there now (optional, kg - only what is there already, the weighings of the mode ask by themselves). SAVE ON THE SCALE saves the THING on the scale like 💾 and loads it on the page. ROAST OUT warns when the green coffee of the roast in is still loaded and offers the roasted coffee of the last roast out of that green coffee (⤓ name) - or + NEW ROASTED COFFEE.
Costs of a batch: the costs of the green coffee are saved with each booking (the "Costs" of the green THING × the weight; a roast in without costs takes the mean costs of the green intakes of its lot). The costs per kg roasted = costs of the green in (raised by the share of its green lot that was sorted out) / roasted kg that can be sold (without the roasted sorted out) + the roasting costs per kg (energy, work, bags - optional, ⚙ ROASTERY SETTINGS, saved on the scale). ROAST OUT shows the expected costs of the batch (green in without costs in its bookings: with the "Costs" of the green THING), + NEW ROASTED COFFEE proposes them as the costs of the THING, the roast log and the overview show them per batch and per coffee, and the batch trace shows the margin of the sold coffee - and a hint when the "Costs" of the roasted THING are more than 5 % away from the real costs of the batch.
Green lot and batch number: GREEN IN, ROAST IN, SORTED GREEN and GREEN OUT book against the lot of the green THING (choose it below the weight or create it in the lot store). ROAST IN proposes the batch number of the new roast (R + date + counter, e.g. R261103-01; ↻ makes a new one, you can also type your own): several sacks into one charge are booked one by one with the same batch number. A pre-roast blend (green coffees mixed before the roast): book one ROAST IN per green coffee with the same batch number - the batch keeps all of them with their shares (e.g. Brazil green 60 % + Ethiopia green 40 %), the roast loss and the costs count for the whole charge, the label shows all origins and the 📈 planner splits the green kg by these shares. A post-roast blend (roasted coffees mixed): see 🔀 Blends below. ROAST OUT offers the open batches (roasted in, not roasted out yet, last 14 days): choose it, put the roasted coffee on the scale and the confirmation shows the roast loss = (green in - roasted out) / green in. The loss and the optional roast colour (e.g. Agtron) are saved with the booking, and the batch becomes a lot of the roasted THING in the lot store (text: green: <lot>) - so the label of every bag and its lot code carry the batch. Sorted roasted, samples, waste and sales book against that batch. Required: ☑ Green lot | ☑ Batch number in ⚙ ROASTERY SETTINGS (on at first, saved and checked on the scale).
🌱 GREEN LOT DATA: beside the lot of GREEN IN (and on every lot card) - origin, farm / co-op, process, crop year, importer, the EUDR reference (the reference number of the due diligence statement of your importer), moisture and density. They are saved as lines in the text of the lot on the scale and read by the green ledger, the batch trace and the exports.
The reports (📋 of the key row: one list in groups - logs, statistics, sales & proof - each row with what it shows, the last one used marked; and 🗂 RECORDS → ☕ ROASTERY). A report with a period first asks for it: THIS MONTH, LAST MONTH, THIS YEAR, LAST YEAR, or your own From / To with MAKE (gray while the end is before the start). A report opens in a window: the notes behind ℹ️, one row per batch, lot or THING with its main number at the right - tap a row for its other columns -, the total in bold at the end; keys CANCEL | ⋯ (sealed PDF - ENTERPRISE, CSV) | 🖨 PRINT (print PRO, CSV / sealed PDF like all reports):
🔥 ROAST LOG: every batch of a month or year - date, batch, green coffee and lot, green in, roasted out, roast loss, colour, sorted out, samples, waste, user, costs per kg roasted. The CSV names its columns like the roast properties of roast software (batch number, weight in, weight out, weight loss), so you can line the weights of the scale up with your roast curves (Artisan, Cropster ...) by batch number and date.
🌱 GREEN LEDGER: per green lot intake - roast in - sorted out - green out - waste = rest, with origin, process, importer and EUDR reference. Its CSV has one row per green lot (also with the price per kg of the lot) - for your own spreadsheet or the import of green lots into other roastery software.
🔎 BATCH TRACE: a batch → its green lots with their data and EUDR reference → every booking of it (sales, samples, waste) and the blends that used it, with the costs of the batch and the margin of its sales. Also from the lot card of the batch, and from 🔎 at the end of a row of the roast log (then ← in its title goes back to the log). Its window groups the lines: the batch, its green lots (or the parts of a blend), the costs, the sums - then one row per booking.
📅 OVERVIEW: a month or year per THING and in total: green in, roast in, roasted out, mean roast loss (weighted by the green weight, with the lowest and highest batch), sorted out, samples, waste, blends, sold and revenue, costs per kg roasted, costs of the sales and profit - a year also per month.
📊 STATISTICS: the roast loss of every batch over time as a chart (one line per coffee), mean and spread per coffee, the yield from green to sold, the batches per week - and the batches far from the mean of their coffee (only an info: worth a cupping).
💶 COFFEE TAX (switch it on in ⚙ ROASTERY SETTINGS; rate and currency are saved on the scale, 2.19 per kg at first as in Germany): for a month the roasted coffee produced, the taxable removals (sales + samples / own use) and kg × rate, waste listed apart. Green coffee is not counted. It is the basis for your own declaration, not a declaration - please check the rate and the rules with your customs office (in Germany: Hauptzollamt). 📈 PLANNER: type the roasted kg you need and get the green kg = roasted / (1 - mean roast loss of the last 10 batches of the coffee), with the rest of its green lots.
🔀 BLEND (tab 🔥 ROASTING) - a post-roast blend of roasted coffees. Setup in the mode: the name of the blend (a THING, or a new name), at least two parts (roasted coffee + its batch from the bookings of the last days, or typed + share in %; ⚖ = 100 % evens the shares out), the total, the tolerance of every part (1 % at first) and the blend batch number (proposed: B + date + counter). Every part shows its rest, price and costs per kg, the blend its price and costs per 100 g and per kg with the margin (from the price unit and the "Costs" of the THINGS of the parts - 📥 READ PRICES AND COSTS for the ones the app does not know yet; a part without "Costs": the costs of its batch). Below it the selling price and the costs of the blend per kg: the computed values are filled in - change them as you like (an empty field = the computed value); + NEW BLEND takes them. The setup opens as the window 🔀 THE BLEND by itself when you choose the chip (one card per part with ✕ at its end - it asks when the part has a coffee; + NEW beside the blend opens + New blend as a page of it with ← back; ✕ closes it, nothing is lost); the page keeps a card of the blend (its parts with their shares, the total) with ✎ to open it again, and START WEIGHING is the main key (gray while something is missing - the line above the keys says what). START WEIGHING opens the blend on the solution page of the scale: every connected phone sees the same parts. ⚖ WEIGH NOW a part in any order, take it (the scale tares by itself), a taken part can get more (its value stays, the rest is added), 🧮 recalculates the other goals from a part that is off; while nothing is taken the shares and the total can still be changed. Price and costs of every part and of the blend follow the weights live. Nothing is booked while weighing. 🔀 BOOK THE BLEND (every part taken, a name given - pulsing when every part is on its goal within the tolerance): the scale books it itself - one record into a blend per part (its roasted coffee, its batch, the blend batch as the note, its costs) and the blend (the sum of the parts, the blend batch, the costs of the parts), all with the values the scale took itself (verified, marked RUN in the sealed report). On the run page the head says ☕ ROASTERY / Blend · batch …, every part is a card (the one on the scale framed, with its meter), WEIGH NOW shows the taken value in the weighing colours, the key row is TARE | 🔀 BOOK THE BLEND. Shares off? A yellow line above the keys shows how much of which part is missing for the right proportions (e.g. add 185.4 g of Brazil) - put it in, or book the blend anyway: it asks once (BOOK ANYWAY) and books it with the shares as they are (the lot text keeps them). If one of the records would be refused by the rules of the scale, nothing is booked and the app says why. Then the app puts the blend batch into the lot store (blend / from / parts / blended - the label shows the blend date and the origins of all parts, the batch trace goes blend → parts → roast batches → green lots), takes the parts out of the 📦 stock of their coffees and puts the blend in, and goes back to the ROASTERY MODE. A new name: + NEW BLEND opens with the price and the costs of the weighed parts proposed - change them before you save. ✕ in the head ends the blend without booking anything (with values taken it asks first: END THE BLEND). The parts booked into a blend are no sales and no removal for the coffee tax: the sale of the blend is. A pre-roast blend (green coffees roasted together) is booked with ROAST IN (see above). ℮ PACK CHECK: switches the ℮ prepackage control on with the batch on the page as the lot number of its report and opens the check mode - set the nominal quantity (250 g, 500 g, 1 kg) and weigh the bags.
🛍 FILL WEIGHING with live statistics (PRO; ⋯ of the key row of the ROASTERY ☕ MODE and of the CCLUB 🥦 MODE, ⋯ of the key row of the DISPENSARY 🏪 MODE) - fill your bags one by one out of a bulk THING. Setup: the bulk THING you fill out of (e.g. the roasted Guji) and its batch (the loaded one is proposed), the bag size (one click on 250 g / 500 g / 1 kg, with oz / lb 8 oz ... 5 lb; in the club 1 ... 25 g, or typed in), the bag THING you sell later (proposed: Guji 250 g), the number of bags (optional) - the app shows the stock of the bulk THING and how many bags it gives - and the tolerance: the ℮ rules (EU or German FPackV: T1, T2, mean ≥ nominal, green from the nominal up to one TNE above it or your own limit), OIML R 87:2016 (the same with its own table of T, the label without ℮), the USA NIST HB 133 rules (MAV line instead of T1 / T2, the label without ℮) or your own − / + (one click on −0 / +1 %, ±2 %, −0 / +0.1 g ...). The UP↑SCALE runs the filling: it works out the limits of the rule itself (the ℮, FPackV and OIML R 87 tables, the HB 133 MAV table, the moisture allowances), judges every bag, keeps up to 500 bags per filling (also after a restart or a power loss) - on an RS-232 dongle each bag with the alibi number the verified scale printed for it (EXTRA LINE: press its print key before ✔ TAKE, or ALIBI ON TAKE does it; in the report, the CSV and on the label of that bag) -, tares by itself and gives the verdict - every connected device shows the same filling and can carry on. Filling: the box turns orange (fill on), yellow (℮: between T1 and the nominal), green (OK) or red (over) - the same colors as in the factory mode and the check-weighing, with the % and the bar; AUTO TARE tares every empty bag by itself (⚖ TARE still works); ✔ TAKE BAG (the main key at the right of the key row) takes the stable weight (a bag outside the limits asks first: TAKE ANYWAY), ↶ takes back the last one, ⋯ has + CREATE PREPACKED PRODUCT AS THING and ✕ DISCARD (asked first when a bag is taken: nothing is moved in the stock); the head shows 🛍 FILL WEIGHING and the bag · batch, its ✕ goes back to the weighing page (the filling stays open on the scale). After ▶ START FILLING a line says that the filling starts until the scale sends it. Below it the statistics grow with every bag: mean, standard deviation, min / max, bags below T1 / T2 / the nominal, over the green range, the total and the overfill you give away (g and %), run chart and histogram. 🏷 Labels: while you fill, the bag THING is on the page (with the batch as its lot), so the label line of the bag at the bottom (🛠 PREP, ⇩ SAVE, ⎙ PRINT) gives the same label as the 🛍️ sales mode for that THING - your label setup, size and printer, its price, EAN, units, roast date and origin lines - but not with the live weight: with the ℮ rules the label carries the nominal with ℮ (e.g. 250 g ℮) and so the price of a bag, with your own tolerance the weight taken for that bag (and its price). ☑ Print labels automatically (in the card with AUTO TARE) prints the label of every bag at its TAKE. A bag THING that does not exist yet gives a label with name, weight and lot only - + CREATE PREPACKED PRODUCT AS THING (in the setup, below the result box and in ⋯ while you fill, and after FINISH) makes it from the bulk THING without the THING editor: a form with the name, the selling price and the costs per bag (proposed from the ones per g of the bulk THING - add the bag, the label and the work), the batch as its lot and all units of the bulk THING (nutrition values, THC / CBD, ...; untick what does not belong to the bag) - change what you like, it is saved on the scale and comes onto the page, so the next label already has its price and EAN. Selling the bags: a THING made this way carries its bag size and its bulk THING (a unit with the sign 🛍, named like the bulk THING, 1 🛍 per bag - it also counts the bags on the scale and is never a line of the label). One bulk THING can have as many bag THINGS as you like (5 g, 10 g, 250 g, 1 kg ...). Choose it in the sales mode (☕ SALE in the roastery, 🌿 DISPENSING in the club, or the 🛍️ SALES MODE), press BOOK and type the number of bags into 🛍 Bags of 250 g (nothing has to be on the scale): the confirmation shows the 🛍 price per bag (from the THING - change it if you like; price per kg and total follow, or type the total), and the sale is booked with the weight of the bags (bags × nominal, marked ✍ like a manual weight), that price and the number of bags as its pieces. The stock comes out of the bag THING - the filling took it out of the bulk THING already, so nothing is counted twice. In the 📦 stock every bag THING shows its bags ("10 bags of 250 g in stock · filled out of Guji"), and its bulk THING the bagged amount with the bags of each size and the total with the bags (open coffee + bags) - also as columns of the stock CSV. A reload, another device or a restart of the UP↑SCALE goes on with the open filling. ✔ FINISH (in the card THIS RUN, asked first) takes the filled weight (with the overfill) out of the bulk THING in the 📦 stock of the scale and puts the bags into the bag THING at their nominal (e.g. 12 × 250 g = 3 kg) - no sale is booked, the coffee in total only goes down with the sale (the sale of a bag later takes it from the bag THING, so nothing is counted twice) - gives the bag THING the lot of the batch (label, lot code), shows the report (print, PDF, CSV with every bag) and offers + CREATE PREPACKED PRODUCT AS THING for a bag THING that does not exist yet. The last 30 fillings of this browser stay in the setup for their report. Fill weighing is not the ℮ control: it watches the filling while the bags are still open - an open bag can still lose something (it falls over, coffee stays in the funnel) before it is closed. The ℮ prepackage control is made afterwards with the closed bags: ℮ PACK CHECK, its lot report is the record for the authority. The filling step of the ⚕ compounding assistant uses the same box, AUTO TARE and statistics with the limits of the dosage form (Ph. Eur. 2.9.5).
📦 Stock: after GREEN IN and ROAST OUT the app asks in its window whether to add the weight to the stock of the inventory (ADD; a note above the keys confirms it); a booked blend moves the stock by itself (the parts out, the blend in); roast in, sorted out, samples, waste and sales take it out (when the inventory takes the sales modes as used).
Several scales: sacks on a floor scale, bags on a table scale, the shop on a third one? ⇄ OTHER SCALES joins their bookings in the reports of this scale; a batch roasted in on one scale and roasted out on another is paired by its batch number.
Live forwarding: with the LIMS / ELN forwarding every roastery booking goes as JSON to your own server (with its kind, trade and values such as the roast loss).
Please note: this is a documentation aid, not legal or tax advice. Please check with your customs office (Kaffeesteuer, tax warehouse) and your food authority what you have to record and report. The records can be checked for changes (hash chain, seal with ENTERPRISE) - keep the exported files safe.
METRC (DISPENSARY) 🏪 MODE: weigh-out, harvest, waste and stock check with the CSV files for Metrc - for US dispensaries (PRO)
⭐ PRO: the DISPENSARY MODE, its CSV files for the Metrc upload, the plain reports (print and CSV) and the package labels - gray in the free version; try it with the free trial (free and PRO). ENTERPRISE & COMPLIANCE adds the sealed PDF of the reports and the 🔏 sealed report, the ✍ witness signature of the waste log, the seal file beside every CSV file and the LIMS forwarding of the bookings.
The Dispensary 🏪 Mode (Metrc)
Made for US dispensaries and cultivators that report to Metrc, the track-and-trace system of their state. Every weight is booked on the UP↑SCALE while you weigh: the deli-style weigh-out by the Metrc tag of the bulk package, the wet weight of every plant, the packages of a harvest, the waste with the reasons of your state and a stock check of every package against Metrc. 📤 TO METRC then shows every row before it goes out and writes the CSV files for the Upload CSV window of Metrc. The mode works like the 🛍️ sales mode (live price and units, labels, manual weight, several devices) and runs next to the other sales modes on the same scale. The weights for Metrc come from your NTEP scale - with the 232 UP↑SCALER plugged into it (see legal use).
Choose DISPENSARY 🏪 MODE in the menu (☰ → ⚖ MODES), group trade. There is no switch: the mode is in the menu when your licence is PRO and your view has it (the cannabis view or the full view). The page of the mode: the weight on top, the three tabs right under it, the card of the tab, and the key row at the bottom - ⚖ TARE at the left, 📋 (the reports), ⋯ (the tools, with the number of the rows not sent to Metrc yet as a badge) and BOOK NOW at the right end with the live weight and price below it. While BOOK NOW waits for something (a tag, a reason ...), it is gray and a line above the keys says what is missing. On a FREE licence the mode shows only one card (🔒 PRO): your state, the tag prefix of your facility and the scale variance. All of it is saved on the scale for every phone and computer, and the scale checks every booking of the mode with its own rules (the SALES, CCLUB and ROASTERY MODES stay available: their bookings have their own rules). When another device connects while one is in the DISPENSARY MODE, it joins it on the same tab. ⚙ in the head of the page opens the window ⚙ DISPENSARY SETTINGS (its ℹ️: the legal and the privacy note), each part in its own card with an ℹ️ - every change is saved at once:
🏪 THE MODE: the name of the dispensary (the first line of the report header and of the labels) and the currency of the sales log.
🧾 BOOKINGS:customer / patient IDs (a field and a scanner on the weigh-out - only the ID is saved, never a name; with the 🪪 member cards), a customer / patient ID needed for every weigh-out and harvest by batch (a harvest without plant tags, if your state allows it). The two checks are saved on the scale, and the scale refuses a booking that misses them.
🏷 METRC (saved on the scale): the state of your licence - after a download the app opens the address of your state's Metrc site (change it in 📤 METRC FILES if yours differs) and the reasons start with the preset of your state; the symbol on a label is chosen in the label dialog (symbol slot). The list makes no claim about which states use Metrc. The tag prefix of your facility: the first characters of your Metrc tags, the same on all tags of your facility (4 to 16 capital letters and digits). The scale variance in g per lb (0.9 at first: ±0.9 g on 1 lb, about ±0.06 g on 1 oz) - how far a package may differ from its Metrc quantity before the stock check asks for a reason. Some states name about 0.9 g per lb: check the rule of your state.
🔖 ALIBI ON TAKE (on a 232 UP↑SCALER): the switch of ALIBI ON TAKE right in the mode - the ticket or alibi number of your NTEP scale then goes with every booking (see legal use).
📤 METRC FILES (kept in this browser - EXPORT saves them as a file, IMPORT loads them on the other phones): the Metrc license number (in the file names), the address of the Metrc site, the drying location of the harvest weights and the location of new packages (as named in Metrc), the days of bookings on the TO METRC screen (30 at first), the employee numbers (per user of the UP↑SCALE and a default - some states want the employee license or permit number with every package adjustment), the adjustment reasons and harvest waste types and the CSV columns per file type.
Three tabs:
⚖ WEIGH-OUT (deli-style): scan the Metrc tag of the bulk package (📷 or a hand scanner) - its THING and its lot come by themselves -, weigh the amount for the customer and BOOK NOW: a sale with the exact weight, the price (optional), THC / CBD of the THING or the lot as in the CCLUB 🥦 MODE and the customer / patient ID (when switched on). The label: as in the sales mode (the label bar, print automatically in the confirmation). Below the tag the app shows the balance of the tag: the quantity of the package list minus the weigh-outs and the package waste, plus the differences of the stock checks since then (this scale and the ⇄ other scales loaded in this browser) - compare it with Metrc and your POS. Your POS reports the sale to Metrc: ring it up there as usual. The DISPENSARY MODE writes no sales file, so no sale goes to Metrc twice. + NEW THING (+ at the end of the THING row, and in ⋯ of the key row) makes the THING of a product right in the mode, without the THING editor: the name, the price and the costs per g (optional), THC / CBD in % (optional - the lab values of a lot replace them later), the Metrc tag of its package or a lot number of your own (optional) and the 📦 stock that is there already (optional, g). SAVE ON THE SCALE saves it like 💾 and loads it on the page; a typed or scanned tag becomes the lot of the package (lot number = the last 20 characters, the full tag in the lot text - see Metrc tags).
🌱 HARVEST & WASTE with a row of three chips under the tabs; the chosen one shows its card right on the page, below the weight (THING, lot, then its own fields):
🌱 HARVEST (wet weight): the harvest name (proposed from the THING and the day, e.g. OGKUSH-20261112; ↻ makes a new one, or type your own - the harvest name of Metrc links the plants, the packages and the waste of a harvest, also weeks apart), the drying location (the first one comes from 📤 METRC FILES) and per plant its plant tag (scan it) and its weight. The UP↑SCALE keeps the last 20 characters of the plant tag; the Harvest Plants file puts the full tag together with the tag prefix of the settings - the app warns at once when the prefix does not fit the scanned tag. With harvest by batch on, the plant tag is optional.
📦 PACKAGE (package from a harvest): the harvest it comes from, the tag of the new package (scan it - a tag that is in the lot store already is not taken), the Metrc item (optional, e.g. OG Kush - Flower; empty: the name of the THING) and the weight of the package; THC / CBD come from the THING or the lot. After the booking the package becomes a lot of the THING with its Metrc tag, its harvest and its item, and the app asks whether to add the weight to the 📦 stock.
🗑 WASTE: of a 📦 PACKAGE (scan its tag; it is taken out of the stock and goes into the Package Adjustment file) or of a 🌱 HARVEST (no Metrc CSV: it goes on the waste list to type into Metrc) - weigh the waste and choose the reason (a package) or the waste type (a harvest). The scale refuses a waste without a reason.
📋 STOCK CHECK: every package of the lot store with a Metrc tag (from the package list, or scanned), with ✓ for the ones checked today in this browser. Scan its tag or pick it (↻ in the head of the card reads the lots of the scale again; ✕ of the card of a package goes back to the list), check the Metrc quantity now (the app proposes the package list minus the bookings since then - type what Metrc shows if it differs) and put the package on the scale: the app shows the counted quantity in the unit of the package (Grams, Ounces, Each ...) and the difference = counted minus the Metrc quantity, with its sign, as Metrc wants it. Within the scale variance the reason is the scale variance reason of your state; over it you choose the reason. The difference shows in the colours of a weighing result: green no difference, orange within the scale variance, red over it. Then BOOK NOW at the right end of the key row. Every difference is a row of the Package Adjustment file of 📤 TO METRC; no difference, no row. Packages in Each are counted by weight. An empty or missing package can not be weighed: adjust it in Metrc directly.
The tabs of the mode are teal, so you always see where you are. Every booking is weighed with a stable weight (or a ✍ manual weight, which is marked and warned), confirmed and saved on the scale with time, user, THING, lot (the Metrc tag or the harvest name), weight and note. The check before every booking: a lot (tag or harvest name) for every booking, the currency of the sales log for a weigh-out (and the customer / patient ID when it is needed), the plant tag of a harvest (unless by batch), the new tag of a package, the reason of a waste and the Metrc quantity of a stock check. If something is missing, the button stays blocked and says what - and the scale checks the lot, the currency, the customer / patient ID, the plant tag and the waste reason once more (settings saved on the scale).
Metrc tags: a Metrc tag has 24 letters and digits and carries a Code 128 barcode (and an RFID chip, which phones can not read). 📷 beside the tag field reads the barcode with the camera of your phone or tablet; a USB or Bluetooth hand scanner types it into the tag field (Enter). Anything that is not a 24-character tag is not taken, and the camera keeps scanning. In the lot store of the UP↑SCALE the lot number of a package is the last 20 characters of its tag (a lot number has up to 20 characters) and the full tag is a line of the lot text - so the lot card shows the Metrc tag, and the 🧾 lot list finds a package by its tag. Two tags that end with the same 20 characters can not both be lots: the app refuses the second one and says so. A tag that is not in the lot store yet: the app offers to make it a lot of the THING on the page - or import the package list first. Scanned with the THINGS scanner or in the tag field of the mode, a tag finds its lot and starts the DISPENSARY MODE with its THING - straight into the right mode; 📷 METRC TAG in the lot list opens the card of its lot.
📥 PACKAGE LIST (⋯ of the key row, and below the list of the stock check): in Metrc, save the list of your active packages as a CSV file (a spreadsheet file such as .xlsx: save it as CSV first) and choose it here. The app finds the columns by their names (tag, item, quantity, unit, packaged date ...) - check them, or pick the right column; the choice is remembered for files with the same header. Then each Metrc item gets its THING (e.g. OG Kush - Flower → your THING OG Kush; remembered for the next import). The preview shows the first 40 rows (a row without a valid tag in red) and how many places the lot store has left. The packages become lots one by one, each with its Metrc tag, its item, its quantity of today (the start of the balance and of the stock check) and its packaged date (printed on the 🏷 PACKAGE LABEL); a package that is a lot already gets the new quantity. At the end: created, updated, skipped, failed. The same import works for any lot list: 📥 IMPORT in the lot list.
📤 TO METRC (the first row of ⋯ of the key row, with the number of the open rows - the same number as a badge on ⋯; also in 🗂 RECORDS → 🏪 DISPENSARY): a window with every booking of this UP↑SCALE that is not in a file for Metrc yet (↻ in its title reads them again) - one card per file type, row by row, before anything goes out:
File (Metrc File Upload Type)
What goes in
Harvest weights (Harvest Plants)
The wet weight of every plant: plant tag, weight, drying location, harvest name, date.
Packages from a harvest (Packages from Harvest)
Every new package: its tag, item, weight, harvest name, location, date. Made only after its columns were pasted or confirmed once.
Package adjustments (Package Adjustment)
The stock checks with a difference (counted minus the Metrc quantity, in the unit of the package, with its sign) and the waste of a package (minus its weight) - each with its reason, a note, the date and the employee number.
Harvest waste (list)
Metrc has no CSV upload for it: a list (CSV and 🖨 PRINT THE LIST) to type into Metrc as the waste of each harvest.
Upload them in this order (harvest weights first, then the packages, then the adjustments). The weigh-outs go to Metrc through your POS, the received packages are accepted in Metrc by hand - neither has a file here.
Every row shows the tag (or plant tag), the item, the quantity and its unit, the reason, the date, who booked it and the ticket number 🔖. Red = an error - Metrc would reject the whole file: no 24-character tag, a lot without a Metrc tag, a quantity of 0, an unknown unit, a reason that is not in your list, no employee number while the columns have an employee column, no harvest name or drying location. Yellow = a warning about where the weight comes from: typed in by hand, not confirmed by the scale, relayed by another app, weighed on an UP↑SCALE with its own load cell (no NTEP), no NTEP CC number in SCALE DATA, no ticket (alibi) number while ALIBI ON TAKE is on. Warnings never stop a file: they are listed once more before the download and stay in the SOURCE column of the reports. The CSV file holds only the columns of Metrc.
Untick what should not go (ALL, NONE, UNTICK ERRORS); the totals follow. Every row of a file with a note column has a note for Metrc (at first the kind and the number of the booking, with the ticket number) - change it right there (kept in this browser). Some states want a note with every adjustment.
⤓ DOWNLOAD + OPEN METRC: the app fetches the rows from the scale once more (a file marked on another phone counts), checks them (a ticked row with an error stops it, with the list), shows the warnings, downloads the file - e.g. metrc-package-adjustment-<license number>-2026-11-12.csv, at most 500 rows per file, more go into numbered files - and opens the Metrc site of your state in a new tab. In Metrc: Data Import → Upload CSV → File Upload Type → pick the file. With ENTERPRISE & COMPLIANCE the scale seals the file: a small .seal.json comes beside it (seal) - keep it, upload only the CSV.
Back in the app:Did Metrc take the file?YES books the sent marker on the UP↑SCALE, NO marks nothing - fix what Metrc rejected and download it again.
The screen shows the bookings of the last 30 days (📤 METRC FILES); SHOW EARLIER adds 90 days. If no file of a type was marked in the days shown, the screen says so and asks from which day the open rows start.
🧾 COLUMNS - the header row of your state: the column names of the Metrc CSV upload differ per state and change from time to time, and Metrc shows them only inside Metrc. In Metrc: Data Import → Upload CSV → choose the File Upload Type - the CSV column headers are shown there. Copy that header row and paste it into CSV columns per file type (⚙ DISPENSARY SETTINGS → 📤 METRC FILES, or 🧾 COLUMNS on the TO METRC screen) and press TAKE THIS HEADER (this confirms it): the app finds the value of each column by its name; a column it does not know stays empty in every row (Metrc wants such columns present and blank) - check every column. CONFIRM keeps the default columns of a file type. 🧾 COLUMNS shows on a box of the TO METRC screen that has open rows. The default columns come from public sources and are not confirmed by Metrc - paste the header row of your state before the first upload. The packages from a harvest are made only after their header row was pasted (or the default confirmed) once.
Adjustment reasons and waste types: the reasons of a package adjustment, one per line (📤 METRC FILES). The list starts with the preset of your state - California, Oregon, Nevada and Ohio have their own, every other state starts with a generic list - and the lists of the states change: copy the exact names from your Metrc account, Metrc rejects a file with an unknown reason. The app proposes the reason itself where it can: the scale variance reason for a stock check within the scale variance (e.g. Scale Variance), the waste reason for the waste of a package (e.g. Waste (Unusable Product) in California). A booking keeps the first 20 bytes of its reason, so two reasons that start the same are refused by the editor. PRESET takes the list of your state again. The harvest waste types (at first Plant Material, Fibrous, Root Ball - not confirmed by Metrc) are set by your state: type them as named in your Metrc account.
The sent marker:YES books a marker 📤 SENT TO METRC on the UP↑SCALE - no weight, only the file type, the number of rows and the number of the last booking in the file, with the time and the user like every record (also in LEGAL MODE: there is no weight to store). From then on every phone and computer of this scale offers only the rows after it, so no row goes to Metrc twice. The reports, the sums and the stock leave the marker out; the sealed report lists it like every record.
Unticked rows inside a file count as handled once the file is marked: the question says how many, and so does the note of the marker. To send one later: EARLIER FILES → tick only that row → DOWNLOAD AGAIN → upload it in Metrc.
Metrc took only some rows? Answer NO, untick the rows Metrc took, fix the others and download them again - then answer YES.
EARLIER FILES lists the rows of every marked file, newest first. A file can be downloaded again after a warning: uploading it again books it twice in Metrc.
The reports (📋 of the key row: one list in groups - logs, statistics, sales & proof -, the last one used marked; ⋯ has 🗂 DISPENSARY RECORDS - and 🗂 RECORDS → 🏪 DISPENSARY, only while the mode is on). A report opens in a window: its rows show the tag and the number that matters at the right, a tap unfolds the other columns; CANCEL | ⋯ (PDF, CSV) | 🖨 PRINT:
🌱 HARVEST LOG: the harvest (wet), the packages and the harvest waste per THING and harvest name. A harvest is weighed in several parts: tick the weighings (ONLY WET / ONLY PACKAGES / ONLY WASTE / ALL) and make one report of them - the wet weight, later the packages of the same harvest, or both with the drying loss in %, as in the club. 🗑 HARVEST WASTE LOG: every harvest waste of a year with its waste type.
🗑 WASTE LOG: every waste of the period, of packages and of harvests, with its reason. With ENTERPRISE & COMPLIANCE: save the sealed PDF, then ✍ WITNESS SIGNATURE - a second person signs it as witness with their password on the UP↑SCALE (electronic signatures).
⚖ WEIGH-OUTS PER TAG: per Metrc tag the count, the sum and the balance - to compare with your POS.
📋 STOCK CHECKS: per package the weight, the counted quantity, the Metrc quantity before the check, the difference and its reason.
🚚 RECEIVED: the packages of incoming transfers - shipped quantity and weighed (or counted) here.
📅 SUMS OF A YEAR: per THING the harvest, the packages, the received packages, the weigh-outs with their THC and CBD, and the waste.
The waste log, the stock checks, the weigh-outs per tag and the received report have your company header, the scale and the NTEP CC number and serial number of your scale from SCALE DATA on top, the ticket number 🔖 and ✍ for a manual weight beside a weight, and a SOURCE column (where each weight comes from). Print and a plain CSV are PRO, the sealed PDF is ENTERPRISE. With ⇄ OTHER SCALES the bookings of a second scale (e.g. the harvest scale in the back) are in the reports of this one.
🏷 Package and jar labels: on the card of a package in the lot store (while the DISPENSARY MODE is on), 🏷 PACKAGE LABEL makes a label of its own - tick what goes on it: the Metrc tag as a QR code or a Code 128 barcode with the tag below, the item, THC / CBD in %, the net weight (of the newest package, stock check or received booking of the lot - only while ⚖ LIVE WEIGHT AND PRICE ON THE LABEL is on), the harvest name, packed and use-by dates, the price per g (jar labels for the weigh-out), the first two lines of your company header (the licensee) and the symbol slot (symbol slot): choose your state - every US state with legal cannabis sales is in the list. A state with its own symbol prints that symbol with its warning statements; for a state without a public symbol you choose the symbol yourself (any state symbol, our 3-symbol warning or none) and the statement of any state. On a colour printer the symbols print in their official colours. The label of every weigh-out comes from the label bar as in the sales mode - with the 🏷 Metrc tag of the package and the same symbol slot. A Code 128 of 24 characters needs a wide label: the label dialog warns when its bars get too thin to read - choose the QR code or a wider label then. Our label is not a Metrc tag or a Retail ID, and it does not replace the label your state asks for: check its rules.
🚚 TRANSFER CHECK (⋯ of the key row) - weigh an incoming transfer before you accept it:
In Metrc, open the incoming transfer and save its packages as a CSV file.
Choose the file here and check the columns (tag, item, shipped quantity and unit, manifest).
Weigh every package in the 🚚 goods-in check: the app shows the difference to the shipped quantity, checked against the scale variance, and books it as RECEIVED with its Metrc tag and the manifest number - the package becomes a lot with its tag, and its weight goes into the 📦 stock.
Accept the transfer in Metrc with the received quantities of the list (🚚 RECEIVED report): receiving is typed into Metrc, there is no CSV for it.
Count check of Each packages (pre-rolls, packed eighths): a package whose Metrc unit is Each is counted by weight in the stock check and the transfer check. The weight of one piece comes from the lot, from the bag size of the THING (fill weighing), or you type it (in the transfer check also with ⤓ TAKE in the 🔢 COUNT BY WEIGHT box: put some pieces on the scale and type how many they are). The difference is in pieces. When the pieces are light for the scale, the app tells you to count them by hand to be sure. The same count by weight works in every stock check of the 📦 inventory.
More in ⋯ of the key row:+ NEW THING (also the + at the end of the THING row of the weigh-out); 🧰 DAILY SCALE CHECK - the routine check with your test weights before use; 🛍 FILL WEIGHING - pre-pack eighths, grams or jars out of a bulk package with the box, AUTO TARE and the live statistics, with the USA NIST HB 133 rules or your own tolerance (fill weighing); ℮ PACK CHECK (HB 133) - the USA package check of closed packages with the lot on the page and the moisture allowance of cannabis; ⇄ OTHER SCALES. With several scales on one app (e.g. a floor scale for the harvest and a fine scale for the weigh-out), see multiple scales.
Live forwarding: with the LIMS / ELN forwarding (ENTERPRISE) every booking of the mode goes as JSON to your own server - with its kind and the full Metrc tag, so a partner system can take the weights.
⚖ LEGAL USE: your NTEP scale stays the legal reading
The weights for Metrc and for the sale come from a legal-for-trade scale: an NTEP scale (Certificate of Conformance), registered and inspected as your state and county require - dispensaries mostly need accuracy class II. Our 420 and 710 have no NTEP. Plug the 232 UP↑SCALER (or the UP↑SCALE DIY on the ESP32-C3 RS232 Adapter) into your NTEP scale and switch on EXACT WEIGHT (set it up in 5 steps).
The display or the ticket of the NTEP scale is the legal reading. The app and the adapter have no NTEP and are no customer display. The adapter sends the NTEP scale only key commands (weight request, tare, print) over its protective interface - never an adjustment or a setting (what stays legal).
The ticket number with every record: with ALIBI ON TAKE the UP↑SCALE presses the print key of your NTEP scale at every booking: the scale stores the weight in its alibi memory or prints its ticket, and its number goes with the record - into the reports and the note of the Metrc rows. A scale without an alibi memory or a printer: leave it off.
Your instrument on every report: type the NTEP CC number and the serial number of your scale once into 📜 certificates & counters.
Weights typed in by hand or weighed on a scale without NTEP are warned before the upload and marked in the reports (SOURCE) - it is your decision what goes to Metrc.
Selling at the counter: the customer sees the display of the NTEP scale, and your POS keeps the sales and reports them to Metrc. Switch off ⚖ LIVE WEIGHT AND PRICE ON THE LABEL for labels at the counter (your choice).
Keep the records as your state requires; the CSV columns follow the Upload CSV window of your state. No legal advice: check the rules of your state.
A live link to Metrc - the 232 UP↑SCALER sends the weights to Metrc itself, with your own Metrc user key - is planned for the 232 UP↑SCALER. Until then, the DISPENSARY MODE writes the CSV files for the Metrc upload. The 232 UP↑SCALER →
Metrc is a trademark of Metrc LLC. UP↑SCALE is not affiliated with or endorsed by Metrc LLC.
Please note: this is a documentation aid, not legal advice. Please check with the rules of your state and your weights and measures office what you have to weigh, record, label and report. The CSV files and their columns follow what Metrc shows in the Upload CSV window of your state - check every file in Metrc before you rely on it. The records on the scale can be checked for changes (hash chain, seal with ENTERPRISE) - keep the exported files safe.
🏷 QR Code Labels for THINGS
As soon as a THING is loaded from the scale (or saved on it), its name stands in the head - and, in the THING / NAME card of the weighing page, a small box shows the mini QR code of the THING and the link 🏷 LABEL (next to the 📦 STOCK key, both the same height and together the full width of the card; it goes away when the name is cleared or changed). Click on the QR code or on 🏷 LABEL to create a black and white label for your jar, bag or container:
The label window (the same for every label of the app - THINGS, recipes, lots, the LABELS with live values, compounding, pipettes, member cards): the title with ✕ at the top right (Esc or a click beside the window close it too - everything is saved at once, nothing gets lost), the label itself on a white card, below it its size ("480 × 320 dots = 60 × 40 mm at 203 dpi.") and what is wrong with it (⚠ - in red when the label breaks a rule, e.g. a symbol smaller than the state allows), then the card 📋 LINES ON THE LABEL (one row per line: tick it, values on the right, its settings below it), the cards of the label (e.g. 🖼 PICTURE, TEXTS & MORE) and at the end the folded card 📐 SIZE & PRINTER. The keys at the bottom: ⇩ SAVE (the label as a PNG image) and ⎙ PRINT - in the free version ⎙ PRINT is gray with 🔒 PRO and ⇩ SAVE is the main key; every saved label of the free version has the line "UP↑SCALE FREE VERSION" below it.
The label always has the QR code (its row in the list has no checkbox). Optionally add the name of the THING below it (scaled to the width of the label) and one or more of its units in smaller text: nutrition values per 100g like on food packages (like "Protein: 12 g/100g"), your own weight units per 1 of your selected weight unit (like "THC: 0.22 g/g" or "6.237 g/oz"), the container weight in grams (as "CONTAINER (+ lid)" or "CONTAINER (no lid)"), percent weighing as "100 % = 200 g", milliliters as density in both directions (like "0.533 g/ml = 1.876 ml/g"), prices per 1 of your selected weight unit (like "0.125 €/g" or "3.54 €/oz"), and all others per 100g, like "kcal: 350 kcal/100g" (in counting mode: per piece).
Set the width and optionally the height in pixels - or in millimeters together with the DPI of your printer - or in printer dots: then the label is made in exactly the dots of your label printer (the DPI comes from the printer setup). As soon as you set up a label printer for direct printing, all label dialogs start with its print width and resolution - change it as you like, it stays until you change the printer setup again. The width starts with 600 px (max. 3000 px), the height is filled in automatically with the size of the content (no maximum). Empty the height field to get back to the automatic height.
If you set a height that is bigger than the content, you can choose to spread the free space between the QR code and the text or to keep it white at the bottom. If it is smaller, the label gets cut at the bottom (the dialog tells you).
The QR code always fills the whole width of the label. Around everything there is only a thin white border - just as much as a camera needs to read the QR code reliably (2 QR dots). A label with only the QR code is always square, its height is set to its width.
Click ⇩ SAVE to save the label as an image, or ⎙ PRINT to send it to your printer (one page in the exact size of the label). With a label printer set up for direct printing, ⎙ PRINT prints right away - without the print dialog (while it prints, ⎙ PRINT waits).
Load a THING with the camera: click 📷 ("SCAN THING's QR") next to the THING / NAME field - the camera starts right away. (Or click ▾ in the field and then 📷 beside the search field at the top of the list.) Hold the label in front of the camera: if the THING is saved on the connected scale, it gets loaded with all its units right away. If not, you get a message and can scan the next one.
Good to know: a THING is identified only by its name - the QR code contains nothing else (e.g. "THING:Flour"). So the THING has to be saved on the connected scale to load it with a label. Spelling counts: "Flour" and "flour" are different THINGS for the scanner.
Every label works in every scanner: scan the label of a recipe with the THINGS scanner (or in the 📒 diary) and the recipe opens - scan a THING label with the recipe scanner and the THING gets loaded. Only the barcode scanner of the nutrition values (food products) stays with product barcodes, and the scanner of the lot numbers takes every other code as a lot number. While a recipe is running, THING and recipe labels can't be loaded (finish or pause it first).
Several scales: you can save the same THING on more than one scale by yourself - just load it, connect the other scale (its units stay on the page) and press 💾 there. Make sure it gets exactly the same name on every scale: if that scale already has a THING with this name, it saves the new one as "Flour #2" - then delete the old one there first (▾, then 🗑 in its line). Then one label works with all your scales, and you can even use different units or values on each of them.
⭐ PRO: printing labels (the free version saves them as an image) - gray in the free version; try it with the free trial (free and PRO).
Set up your own label once - and then print it for every weighing with the current values. The LIVE LABEL line sits at the very bottom of the screen above the keys (in the weighing and in the counting mode) - always visible, always one line: 🏷 LIVE LABEL ℹ️ and the keys 🛠 PREP, ⇩ SAVE and ⎙ PRINT (gray with 🔒 in the free version); where the words do not fit, the keys show only their signs. The lines of messages and the 🧰 test weights that also show at the bottom are stacked below it, so nothing covers it. Weigh your products piece by piece and print an exact label for each of them - including the price, the mg of THC, the nutrition values and everything else you have added as a unit.
🛠 PREP opens the setup with a live preview (calculated with the weight on the scale right now). Check the lines you want on the label and change their order with the ▲/▼ buttons:
Picture (optional): click 🖼 ADD PICTURE (card 🖼 PICTURE, TEXTS & MORE) to upload your own logo or photo (✕ REMOVE PICTURE asks first). It gets converted to grey and is always as wide as the label. Choose photo (grey becomes fine black dots) or logo (sharp black and white) - label and thermal printers can only print black.
Name: the name in the name field at the top (the same as for THINGS and logs) - as big as the width of the label allows. Load another THING and the label gets its name automatically.
NET WEIGHT in your selected weight unit - and in the counting mode also the NET COUNT (pieces). Both start as "big with frame". You can select them separately, but with the same highlighting they always share the same box.
Units: every unit on the page - also the hidden ones (👻) - with its value for the current weight, like "THC 10208.00 mg" or "Price 5.80 €". Nutrition values also get their share of the net weight in %, like "Sugar 23.20 g (50 %)" - and so do THC and CBD, like "THC 1052.50 mg (21.05 %)" (below 1 % with two decimals). An inverted box (white on black) right above or below a framed box, the nutrition table or the lot code is joined to it, without a white gap. New units are checked if they are shown on the page.
NUTRI-TABLE (EU nutrition table): if all 7 values of the EU nutrition declaration (food information regulation 1169/2011, "big 7") are on the page - energy, fat, saturated fat, carbohydrate, sugar, protein and salt (sodium) - you can check this line to print them as a table instead of the single lines: energy in kJ and kcal, "of which" lines indented, rounded like the EU guidance says (e.g. "<0.5 g"). Choose per 100 g or per 100 g + net weight (counting mode: + per piece). Its 7 single lines are left out then and shown gray in the list (uncheck the NUTRI-TABLE to get them back), all other units (like the price) stay. It is in the language of the app. Please note: the values are only as good as the values on your page - you are responsible for your labels. ⭐ PRO: the NUTRI-TABLE is gray in the free version - its 7 values are printed as single lines then (also on the label of a recipe).
Own texts: click + TEXT (card 🖼 PICTURE, TEXTS & MORE) to add as many as you like (✕ at the end of its row removes one - the app asks first). Use placeholders in square brackets: "With [sugar] sugar!" becomes "With 23.24 g (50 %) sugar!". While you type, all placeholders that work right now are shown below the text field together with their current values - click one to insert it (or click "more info"). Always possible are [weight], [name] and in the counting mode [count], [scaleline] (extra line of an RS-232 scale, e.g. its alibi number); upper/lower case and accents don't matter, nutrition values work in all languages ([sugar] = [zucker] = [azúcar]). Dates: [date] (today), [date+7] (in 7 days - e.g. for a best before date, any number of days works, also [date-1]) and [time], written like your browser writes dates ([datum], [fecha], [zeit], [hora] work too). A placeholder that matches no unit stays as you typed it, so you see a typo in the preview.
🧾 Lot: the lot number of the THING on the page (on the label from the start). Also as placeholder: [lot]. 🧾 LOT CODE (not on the label until you check it): its QR code, small and marked with LOT and the number, a link to the readable lot page. Your business (not on the label until you check it): the first line of your company header, also as placeholder [business].
Unit price (not on the label until you check it): for every price unit on the page there is a line "Price per kg" (per lb if you weigh in oz or lbs, per piece in the counting mode), like "12.90 €/kg" - in the EU, goods sold by weight need the price per kg next to the price. Also as placeholder: [price per kg]. The "Costs" unit (what the goods cost you) is its own line "Costs per kg" and is never on the label (nor in the EAN-13 barcode) until you check it.
Roast date, origin, harvest date, Metrc tag: the lines that the ROASTERY ☕ MODE, the CCLUB 🥦 MODE and the DISPENSARY 🏪 MODE keep in the text of the lot on the scale - Roast date (the day of the roast out of the batch), Blend date, Harvest date (the day of the first fresh harvest of a club batch, or of a Metrc harvest), the Metrc tag of a package (as text), and of a coffee batch from its green lot: Origin, Farm / co-op, Process, Crop year and the Green lot (a blend: of all its batches). Each one is its own line: check it, move it, make it small, big, framed or inverted like any other line; the dates and the origin are on the label from the start, the others when you check them. Also as placeholders in your own texts: [roast date], [blend date], [harvest date], [origin], [farm / co-op], [process], [crop year], [green lot], [metrc tag]. The app reads the lot texts from the scale before it shows or prints a label.
🏷 METRC TAG (only offered when the lot on the page is a package with a Metrc tag - DISPENSARY 🏪 MODE; not on the label until you check it): the full 24-character tag as a QR code or a Code 128 barcode (choose it beside the arrows). A Code 128 of 24 characters is long: the info line of the label dialog warns when its bars get thinner than 2 printer dots, which a scanner may not read - choose the QR code or a wider label then. This label is not a Metrc tag or a Retail ID; the weight on it follows ⚖ LIVE WEIGHT AND PRICE ON THE LABEL.
Name of the price line (under the price line of the setup): the currency (e.g. Euro), Price or Equivalent - the "Costs" never get printed as a price.
⚖ LIVE WEIGHT AND PRICE ON THE LABEL (the first card of the label window, on at first): off, the label leaves out every measured value - the net weight, the price, weight and volume units, an EAN-13 with price or weight, and the placeholders of these ([weight], the price) print as "–". Name, lot, dates, the unit price per kg, THC / CBD and the ingredients stay; ℮ prepacks keep their nominal quantity. It is your choice - the ℹ️ beside it explains why it can matter: when you sell by weight with the customer in front of you, in the EU every printer for the customer must be verified together with the scale (in the USA: NTEP), and the app is not verified. Off is the safe setting for sales at the counter: the weight and the price for the customer then come from your verified scale or cash register (see What stays legal).
US + metric (☑ in the NET WEIGHT line, off at first): the net weight also in the other system, like 227 g (8.00 oz) or 8.00 oz (226 g). Grams on the scale get the ounces (from 1 lb: 24.00 oz / 1 lb 8.00 oz, from 4 lb the pounds), a scale in oz, lbs or ⅛ gets the grams (from 1 kg: kg). The second value is rounded down, so the label never says more than the scale weighed. Also in the placeholder [weight]. In the USA consumer packages need the net quantity in both systems (FPLA, NIST Handbook 130) - some products follow only the rules of your state (cannabis, for example).
℮ mark (☑ in the NET WEIGHT line, on at first): a label of a ℮ prepack carries the nominal quantity with the ℮ mark beside it (e.g. 500 g ℮) - the price and an EAN-13 with price or weight follow the nominal. A ℮ prepack: while the ℮ prepackage control is on (nominal set), and the bags of 🛍 fill weighing with the ℮ rules (with the US rules and OIML R 87: the nominal quantity without ℮). Off: the weighed weight, no ℮.
EAN-13 BARCODE (not on the label until you check it): a barcode with the price, the weight (g) or the pieces in it, for the barcode scanner of your checkout. Shops use the numbers 20-29 at the start of an EAN-13 for their own goods: choose the prefix (20-29), your 5 digit item number and what it contains - the app adds the value (price in cents, max. 999.99; weight in g, max. 99999) and the check digit, e.g. 21 00001 00323 4 for 3.23 €. With more than one currency on the page, choose the one for the barcode. The bars are made in whole printer dots, so they stay sharp. Some checkout systems use other formats (e.g. 4 digits price with an own price check digit) - please check yours before you print many labels.
CANNABIS WARNING / SYMBOL (not on the label until you check it; not offered in the ROASTERY ☕ MODE): a row with a warning symbol, pure black and white for label printers; with the small select beside its arrows you choose ◻ black on white or ◼ inverted (white on black). Its settings (shown only while the line is checked):
State: every US state (and DC) with legal cannabis sales - adult use or medical. A state brings its rules (the size and colours of its symbol) and its warning statements, word for word from its official rules (the select Warning chooses the statement for the kind of goods - flower, edibles, concentrates, topicals, medical... - and, where a state asks for them, the rotating warnings). States with several official symbols are listed once per symbol (New York: square, vertical, horizontal; Utah: outline, triangle, diamond; New Jersey: package symbol and the imprint for each serving; Minnesota: adult use and medical). A state that asks for two symbols gets both side by side (Connecticut, Massachusetts, Minnesota, Vermont).
Symbol: the symbol of the state - drawn from its official file and only resized, never changed or cropped -, our 3-symbol warning or none (only the text, over the full width). Our 3-symbol warning: left a no driving sign (a crossed out car), in the middle a warning triangle with "! THC" and right an 18+ sign - our own symbols, not official ones of any state or country, made for home makers, private clubs and "not for children" labels. You choose freely: any state symbol, ours or none, with the statement of any state as the text. A state without a public symbol (e.g. Illinois, Pennsylvania, Hawaii, or Ohio and Delaware, whose symbols are only given to licensees) starts with none and its statements. Only in the DISPENSARY 🏪 MODE a state with its own symbol always prints that symbol.
Colours: most states ask for their symbol in colour. On a colour printer it prints in its official colours, on a black-and-red printer in black and red, on every other printer in black (dark colours black, light colours white). The line below the settings says when the rule of the state asks for colour.
Size:100 % (the symbol as wide as the label), 50 %, 33 % or 25 % of the width. The symbol is always on the left, and a text flows around it: beside the symbol as long as it is high, then below it over the full width (normal size - or as big as fits beside the symbol). Instead of a text, a small picture (photo or logo - your logo, a pictogram ...) can stand in a framed box beside it (not at 100 %). This is for every label with this line, also without a state. When you pick a state, the symbol goes to 50 % with the state's warning flowing around it (the long warning texts stay readable) - change the width if you like.
The warning of the state: choosing a state fills the text of the box with its warning statement, word for word from its rules (e.g. California: the GOVERNMENT WARNING for cannabis or the one for cannabis products) - when the box is empty or still holds the warning of another state. You can change the text, or pick the statement of any other state in the select.
The size on paper: the info line of the label dialog shows how high the symbol prints (mm and in) and warns when it is smaller than the minimum of the state - make the label bigger or give the symbol a bigger part.
The ℹ️ names the source, the colours, the size rule and the date the state was checked. A symbol and a warning alone do not make a label legal: please check which symbols, warnings, sizes and colours the law of your state or country requires. For other official symbols: simply add them as a picture (the picture line, the box beside the symbol or the 🔣 SYMBOL WITH A BOX line). The package labels of the DISPENSARY 🏪 MODE have a simpler slot: the symbol, its width and the warning text.
🔣 SYMBOL WITH A BOX (not on the label until you check it; on every label, also in the ROASTERY ☕ MODE): your own picture as a symbol - an organic, allergen or hazard pictogram, a logo - 100 %, 50 %, 33 % or 25 % as wide as the label (always on the left), only resized and printed in black and white (logo: sharp edges, photo: grey as dots). A text flows around it (beside it, then below it; normal size or as big as fits beside the symbol; placeholders like [weight] or [date] work) - or a second small picture stands in a framed box beside it (not at 100 %). With the small select beside its arrows: ◻ black on white or ◼ inverted (white on black, in one black band). The info line of the label dialog shows how high the symbol prints. Official symbols often have a minimum size and fixed colours: check the rules that apply to you. The pictures are saved in your browser with the label setup.
Ingredients (not on the label until you check it): type them in - or click 📥 FROM A RECIPE OF THE SCALE and choose one of your recipes: the text gets filled with "Ingredients: " and all its ingredients, biggest amount first (you can still change it). Allergens are written in capitals right away (see ⚠️ Allergens), and ⚠️ MARK ALLERGENS below the text does the same for a text you typed. On the label it is broken into lines between the words. Reading the recipes needs the weighing mode (the scale shows the recipe pages for a moment and comes back); in the counting mode you can type the text. The text is saved with your label.
Every dot is used: these labels have no border - the picture and the texts go from edge to edge, and there is no space above the first or below the last line. If your printer or labels need some space at the edges, enter a white border (in px/dots) in the setup: then the content keeps exactly this distance on all four sides.
Highlight what you want: every line (except the picture) has a small select left beside its arrows: b small (normal text size, not stretched to the width), B normal, ▢ big with frame (black frame around it) or ◼ big inverted (white text on black). Highlight the name, a unit like the price or the THC, an own text - or several of them. Highlighted units (and net weight/count) one after the other with the same highlighting share one box, the name and own texts always get their own box.
Every line is printed as big as possible without breaking it: all unit lines in the same size (the longest one fills the width). In your own texts every line gets its own size and goes exactly from the left to the right edge - short lines get bigger, long lines smaller (they only break where you put a line break). Highlighted lines are always the biggest texts, the normal lines stay smaller than them.
Size: like for the QR code labels - width and height in pixels, in millimeters with the DPI of your printer, or in printer dots. For label rolls set a fixed height: it is remembered for this label (empty field = automatic height), and you can spread the free space between the lines or keep it white at the bottom. Content that is higher than a fixed height is made smaller to fit. With a label printer set up, the size starts with its print width and resolution (tape printers: the label lies on the tape).
✂ Cut label after printing (in 📐 SIZE & PRINTER): only shown if your label printer has a cutter - the same switch as "Cut after printing" in the printer window.
⇩ SAVE downloads the label as PNG and ⎙ PRINT sends it to your printer (one page in the exact size of the label) - both with the weight on the scale at the moment you click and all values calculated with it. If the weight is not stable yet, the app asks first. With a label printer set up for direct printing, ⎙ PRINT prints right away - without the print dialog.
The setup (including the picture) is saved in your browser, so it is there again next time - also with another scale.
Print with one click: connect your label or thermal printer directly (wireless, USB or serial port). It has its own window 🖨 LABEL PRINTER (🖨 PRINTER ▸ in 📐 SIZE & PRINTER of the label window, ☰ → 🛠︎ SCALE SETTINGS → 🏷 LABEL PRINTER, or the SCALE SETTINGS page) - more than 15 printer languages from receipt printers to Brother, Dymo, Zebra, Niimbot or Phomemo, and colour printing (black and red, or all colours) with a colour for every line. Then ⎙ PRINT in the LIVE LABEL bar prints right away, without the print dialog of the browser. All about it: Label Printers.
Labels for a recipe: on the recipe page, ⎙ PORTIONS & LABELS makes a label for one calculated portion (before you weigh anything) - with its own setup and the ingredients of the recipe.
🖨 Label Printers
⭐ PRO: label printers - gray in the free version; try it with the free trial (free and PRO).
All labels of the app - the LABELS with live values of the weighing- and counting mode and the QR code labels of THINGS and recipes - can be printed with the print dialog of your browser. That works with every printer that is installed on your device (also network printers), but you have to confirm the dialog every time. Connect your label- or thermal printer directly instead, and every label is printed with one click.
Open the window 🖨 LABEL PRINTER with 🖨 PRINTER ▸ - in the card 📐 SIZE & PRINTER of the label window (next to the state of the printer: print dialog, connected, connects at the next print, printing...) -, with ☰ → 🛠︎ SCALE SETTINGS → 🏷 LABEL PRINTER, or on the 🛠 SCALE SETTINGS page (tab GENERAL, row 🏷 LABEL PRINTER). The ℹ️ behind its title explains the connections, printer languages and colours. Every change is saved at once (there is no SAVE key); ADVANCED is folded at the end, the state of the printer stands above the keys. The settings are saved in your browser, not on the scale - so they stay the same when you connect another scale. Direct printing works in Chrome and Edge on PC (Windows, macOS, Linux, ChromeOS) and in Chrome or Samsung Internet on Android mobiles and tablets (wireless, and USB via an OTG adapter). On iPhone and iPad only wireless printers work, in the Bluefy browser. The Dymo LabelManager (HID) works on PC only.
Connections
Print dialog of the browser - all printers.
Wireless - most mobile label and thermal printers.
USB - printers with a direct USB connection (computers). If your system has its own driver for the printer, it can block the access.
USB (HID) - for the Dymo LabelManager.
Serial port / wireless serial - USB printers that work as a serial port, and wireless printers with a serial connection: pair them in the Bluetooth settings of your computer first. Niimbot printers with USB cable work as serial port too.
Printer Languages
Select the language of your printer - the list only shows the languages that work with the selected connection:
Language
Printers
Connection
ESC/POS
receipt and thermal printers (58/80 mm), most cheap wireless printers
BLE, USB, serial
TSPL
label printers of TSC, Xprinter, Munbyn, HPRT, iDPRT, Polono, Jadens and many more
BLE, USB, serial
ZPL
Zebra and compatible label printers (also many Honeywell, Bixolon, Godex...)
BLE, USB, serial
CPCL *
mobile printers (Zebra mobile, HPRT, Honeywell...)
BLE, USB, serial
EPL2 *
older Zebra/Eltron and many cheap label printers
USB, serial
Star *
receipt printers of Star Micronics (TSP100/TSP143, mC-Print, SM-L...) in graphic mode
BLE, USB, serial
Brother QL *
Brother QL label printers (raster mode)
USB, serial
Brother P-touch *
Brother PT tape printers (tape printer)
USB, serial
Dymo LabelWriter *
Dymo LabelWriter
USB
Dymo LabelManager *
PnP, 280, 420P (tape printer)
USB (HID)
Dymo LetraTag *
LetraTag 200B (tape printer)
BLE
Phomemo *
M110, M120, M220...
BLE, serial
Phomemo D30 *
D30, D35, Q30 mini label makers (tape printer)
BLE, serial
Niimbot *
D11, D110, B21, B3S... - and "Niimbot new" for B1, B18, B21 Pro...
*: experimental - made after the documentation, not tested with a real printer yet. Please tell us if yours works!
Colour printing
Colours in the printer setup chooses what your printer can print: black (every printer), black and red (two-colour printers) or all colours. With black and red or all colours, every line in the label dialogs gets its own colour: a small select ⬛ black / 🟥 red on a two-colour printer, a colour picker on a colour printer (pictures have no colour choice - they keep their own colours on a colour printer). An inverted line uses its colour as background with white text. The colours are saved per label in your browser and also apply to ⎙ PRINT in the LIVE LABEL bar. The state symbols of the cannabis line print in their official colours on a colour printer and in black and red on a two-colour printer.
Colours
Printers
How
Black and red *
Brother QL-800, QL-810W, QL-820NWB with the two-colour tape DK-22251
Brother QL, USB or serial (the QL-820NWB also over Bluetooth as serial port)
Black and red *
receipt printers with two-colour paper: Epson TM-T88IV, TM-T90, TM-L90, Bixolon SRP-350plus/352plus, Citizen CT-S601II/801II and other ESC/POS printers with two-colour paper - switch the printer to two-colour paper with its own utility first
every colour printer with its own driver: Epson ColorWorks and TM-C3500, Primera LX series, Afinia L301/L502, Brother VC-500W, Brother QL two-colour, inkjet and laser printers on label sheets
print dialog of the browser
*: experimental - made after the official command references, not tested with a real printer yet. Printers that only print black on coloured labels (most small Niimbot, Phomemo and similar printers) stay at black.
Settings
Colours - black, black and red or all colours (see colour printing); only offered for the printer languages that can do it and for the print dialog.
Resolution (dpi) of your printer - mostly 203, Brother QL and Dymo 300, Brother P-touch and Dymo LabelManager 180, Epson ColorWorks 600 (check the value of your model).
Print width (dots) - the width of the printhead (58 mm printers: 384, 80 mm: 576, Niimbot D11: 96, Brother 62 mm: 696, Dymo LabelWriter: 672). Wider labels are made smaller, receipt printers get the label in the middle. For Brother P-touch and Dymo LabelManager you select the tape width instead.
Darkness, labels with gap (die-cut labels) and the gap in mm, feed, copies and turn the label by 90° - depending on the printer language.
✂ Cut after printing - for printers with a cutter (ESC/POS, TSPL, ZPL, Star, Brother QL and P-touch, Epson ColorWorks). The same option is in the card 📐 SIZE & PRINTER of the label window - and only shown if your printer can cut.
Advanced: packet size and pause for wireless printers (try smaller packets or longer pauses if a printer stops in the middle), baud rate for serial ports, margins and line length for Brother QL (90 bytes: QL-500 to QL-820, 162 bytes: QL-1050/1060N/1100).
Size of the labels
As soon as you set up a printer, all label dialogs start with its print width and resolution, and the size is shown in printer dots: then the label is made in exactly the dots of your printer - nothing gets scaled when printing. Change the size as you like, it stays until you change the printer setup again. The dialog warns if a label is wider than the printer.
Tape printers (Brother P-touch, Dymo LabelManager and LetraTag, Phomemo D30/D35/Q30): the printhead goes across the tape, so the label lies on the tape - its height is the width of the tape, its width the length of the label (e.g. LetraTag: 150 × 30 dots, P-touch 12 mm: 350 × 70 dots). The label is turned automatically when printing, and content that is higher than the tape is made smaller to fit.
Connect & Print
🔗 CONNECT (the main key at the right) opens the list of your browser to choose the printer - connected, the key becomes ✕ DISCONNECT; 🧪 TEST PRINT prints a small test label. Changing the connection forgets the printer.
PRINT in all label dialogs and in the LIVE LABEL bar prints right away on your printer - without the print dialog of the browser (the LIVE LABEL bar with the weight on the scale at the moment you click). With "Print dialog of the browser" selected as connection, PRINT opens the print dialog.
After a restart of the app the printer connects again with the next print (Chrome may ask you to choose it again).
Not printing? Switch the printer on, make sure it has labels and is not connected to another app or device. Try the next printer language, a smaller wireless packet size or a longer pause. And the print dialog of the browser still works with every installed printer.
Printer languages after: Epson ESC/POS (two-colour graphics), Epson ColorWorks ESC/Label, Brother QL-800 series raster command reference (two-colour printing), TSC TSPL/TSPL2, Zebra ZPL II / CPCL / EPL2 programming guides, Brother QL and P-touch raster command references, Dymo LabelWriter technical reference and the open source projects brother_ql, phomemo-tools, niimprint, niimbluelib and Cat-Printer, python-StarTSPImage, phomemo_d30, dymoprint, dymo-bluetooth and peripage-python (no code was taken from them).
PART 4 Settings & Checks
⚖ Main Settings
This scale is operate-able with plenty of weighing units and in different languages! Calibration is fast and simple! Also it will allow you to operate it with multiple displays and devices at the same time, it even can be connected with your home automation systems or other devices!
Within the app you are always able to switch between weight units or to select a different language - but you are also able to save your preferred settings directly on the scale in order to load them automatically - independtless of the device you are using to connect with the scale! Go to the 🛠 SCALE SETTINGS page to change your settings and save them on the scale directly! The head shows the name of your scale and the page; the ℹ️ behind the title explains the page, the ✕ at the top right goes back to the scale. The tabs under the head sort the settings into cards:
GENERAL: BASIC (weight unit, language, currency, net / gross, indicator, stability light, restart, heartbeat lock, zero memory, new scale number), the PRO / ENTERPRISE notes of the free version and 🏷 LABEL PRINTER (the row opens the window of the label printer).
SCALE: 🔌 RS-232 setup ▸ (RS-232 adapter), SCALE DATA (with its folds: the data of the attached scale, 📜 certificates & counters, minimum weight), UP↑SCALE SEAL, ⚖ CALIBRATION (calibrate now, the date of the last calibration and its reminder) and 🔍 Scale self-check ▸ (the test page).
LINK: USB, BLUETOOTH & ESP-NOW (multiple masters, wireless for all, WATCH LINK, Bluetooth power, encrypted link with the registered apps, the USB cable, 📡 ESP-NOW devices ▸, USB serial output), SELECT YOUR OUTPUT (only with EXTENDED++) and BOARD HARDWARE (UPCENT↑420).
SPEED: SPEED & STABILITY (the slider, autozero, locking, display), the folded ⚙ expert filter settings, creep compensation, tare guard, environment filter and environment test.
QUALITY: air buoyancy, 📡 sensors, quality records & users, LIMS / ELN forwarding, 📦 inventory and the cannabis club.
Every row has its ℹ️ right behind its label. A card with its own keys (for example the tare guard, the seal or the register) saves by itself; everything else goes to the scale with 💾 SAVE at the right end of the key row at the bottom. While a field differs from the scale, a line above the keys says Not saved yet: n changes (n = the rows you changed; the six parts of a MAC address are one row). Leaving the page then asks first: STAY, DISCARD or 💾 SAVE (✕ and Esc stay). A change that restarts the scale (a new scale number, the board hardware, the Decent mode, WATCH LINK) asks first. On a phone a row that is too long puts its control below its label.
Straight to one part of the settings: ☰ → 🛠︎ SCALE SETTINGS lists every part of the settings page (and the RS-232 setup and the label printer); an entry opens the tab of that part and unfolds it. The entries work on the weighing and counting pages, on ✰ SCALE INFO, on the page of all tests and the result of a test, and on the settings pages themselves. If you leave the settings page - or the RS-232 setup or the ESP-NOW page - with changes that are not saved yet, the app asks first: STAY, DISCARD or 💾 SAVE (it goes on only once the scale took the settings; a restart you cancel keeps you on the page). A new settings packet of the scale (for example from another app) draws the page again, but keeps what you typed and did not save.
Weighing Units
This scale is able to weigh in Gram as like as in many other units. Ounces or Pounds? No problem! We even can weigh in Eights! (No joke!) But also you can find all the other more or less common or useful units, like Carats, Troy Ounces or - for our American friends - Giraffes per Footballfield.
Here a list of all weighing units:
g Grams
oz Ounces
⅛ Eights
lbs Pounds
oz t Troy Ounces
ct Carats
GN Grains
tla Tolas
dwt Pennyweights
gpf Giraffes per football field
kg Kilograms (also for big RS-232 scales that send in kg)
mg Milligrams
µg Micrograms (for micro and ultra-micro balances on an RS-232 adapter)
t Tonnes (for big platform scales and weighbridges on an RS-232 adapter)
The app shows every unit with as many decimals as the scale can show: for g, kg, mg, µg and t straight from the smallest step d of the scale (a weighbridge with d = 20 kg shows 60.02 t, not 60.020000 t). mg, µg and t also work as the unit an RS-232 scale sends, and saved as your unit on the scale.
And by the way, its also possible to weigh directly in volume, such as Milliliter or Cups, regardless of the scales weight unit!
Languages
The app is available in English, German, Spanish, French, Italian, Chinese (Simplified), Russian and Arabic (Arabic texts run from right to left). You can change the language on the fly, or save your preferred language on the scale at the settings page. On the start page the language is the select in the head. The app remembers the last language on your device, so the start page already comes in it; when a scale connects, the language saved on the scale counts - not on the virtual scale (🧰 test weights, ⌨, 📡, scales of other makers): it keeps the language of your browser. Numbers show a decimal point in every language (1.50 $/g) - on the page, on labels and in reports; you can type a comma or a point. Only the voice and the spreadsheet export with the decimal sign of the language use the comma where the language has one.
Speech Output: You can switch on a speech-output! The scale now will read out loud the current weight each time it has changed! Depending to your selected language, device and browser you maybe even have different voices and dialects to choose from! If no voices are installed for your language, the app shows a message instead. On Linux, Chrome only finds the system voices when it is started with the option --enable-speech-dispatcher.
Calibration
Its generally not a bad idea to re-calibrate the scale from time to time, specially after not using it for a long time or if you travel with it... Our inbuilt calibration program will make things easy for you.
Single-Point: You just need to have known weight ready to hand - then click the button at the settings page and follow the instructions! For a single-point-calibration, ideally use a weight which is min. a third of the range of the scale!
The calibration page: the head shows ⚖ CALIBRATION small and SCALE CALIBRATION large; the cards: LAST CALIBRATION (the date and the reminder), MODE (1, 2 or 3 calibration points, ℹ️ explains the weights) and one card per point with its range, the ideal weight and the field Your weight. ▶ START CALIBRATION at the right end of the key row checks the fields (a wrong one gets a red frame) and asks if the scale is empty - then every step has its own card: "Please wait" with the turning ⚖ while the scale tares or checks, PUT WEIGHT n ON THE SCALE with ⚖ WEIGHT n: CALIBRATE NOW, EMPTY THE SCALE with NEXT ▸; "Step n of m" above the card. The ✕ in the middle of the steps asks first - stopped there, the scale keeps its old calibration (it saves the new one only at the end). When it is saved, a note "⚖ Calibrated" shows on the weighing page. On an RS-232 adapter the reminder line has SET DATE ▸ instead of CALIBRATE NOW: it opens the ⚖ CALIBRATION card of the settings, where you save the date of the calibration you did with the menu of the scale.
Which weight? The calibration is only as exact as the weight value you type in - the scale takes it as the truth. Best are test weights (OIML R 111, e.g. class M1 or F2). No test weights at hand? Weigh any thing on a different, calibrated scale and use that value. Not a known weight: the weight printed on a package, or the reading of this scale itself (taking the shown value as the true one changes nothing - the scale would just keep its error). For a scale used for trade, the standard OIML R 76-1 (3.7.1) asks for test weights of OIML R 111 with an error of at most 1/3 of the permissible error of the scale, and the verification is done by the authorities - our calibration and self-check are no verification.
Multi-Interval / Multi-Range: Optionally its possible to calibrate the scale with multiple weights! This option will make the scale even more exact over its full range! You can use a 2-point-calibration or even a 3-point calibration which makes your scale a real multi-interval-scale (Well, we're always showing the smallest, best possible interval - but with a higher precision then, based on the current gross weight). In this case you will need a separate known weight for each calibration-step, but no worries, our calibration wizard will guide you through that!
The scale also remembers the weights of a multi-point calibration and goes in a straight line from one calibration point to the next one - so there are no steps in the weight when the load passes from one range to the next.
Exact zero and weights: for the zero and for each calibration weight, the scale takes the mean of 2 seconds - and only when the weight has been still for these 2 seconds: no hand on the plate, and a heavy weight has stopped swinging (else the error would stay in the calibration until the next one). So after you put the weight on and press the button, it can take a few seconds more; if the scale does not get still at all, it takes the value after 10 seconds anyway. Then check the result with the calibration check.
PS: Your scale will come pre-calibrated with 3 different calibration points from the factory!
📅 Calibration date & reminder: with every calibration the app sends the date to the scale, and the scale saves it (the scale has no clock of its own; for audits see the device qualification). So every phone or computer that connects sees the same "Last calibration: 15.07.2026 (94 days ago)", at the top of the calibration page and in the ⚖ CALIBRATION block of the scale settings. Scales calibrated before this firmware show "unknown" until their next calibration. Below the date you choose after how many days you want to be reminded (30, 60, 90 - the default -, 180 or 365 days, or never; this is a setting of the browser). When the calibration is older than that, a yellow line at the bottom of the screen tells you so when you connect, with CALIBRATE NOW, REMIND ME IN 30 DAYS and ✕ (hidden until the app is opened again). The date (with ⚠ when a calibration is due) is also printed in the header of the reports, e.g. the batch report and the tracking report. A 232 UP↑SCALER can't calibrate the connected scale itself: calibrate it with the scale's own menu, then save that date in the scale settings ("Calibrated with the menu of the scale itself on").
Center of zero: on the weighing page the mark →0← of the display lights up when the scale is exactly at zero - within a quarter of the verification interval e (OIML R 76-1 4.5.5; e = d unless you set a larger e in SCALE DATA), not only rounded to 0 (like on lab and legal-for-trade scales). With more scales connected, the sum is shown, so the mark stays off then. The scale decides this itself from its unrounded weight, not the app. Not with an RS-232 scale in EXACT WEIGHT: its weight comes already rounded to d, so "0" only means within half a digit - the real center of zero is known by that scale alone (look at its own display), and the self-check skips the zero indicator there. A click on the weight (on every page with the weight) opens a window: what the marks mean (STABLE off and the number dimmed = still moving, STABLE = stable, →0← = exactly at zero; the current one is marked), and what is behind the weight - the max. load and d of the scale, the speed preset with its averaging time and when the weight is stable, autozero, locking and creep compensation on or off, the last calibration (⚠ when the reminder is due) and an overload. From there, →0← ZERO sets the zero of the empty scale (ZERO and TARE), ⚙ SETTINGS opens the scale settings and 📖 MANUAL this page.
⚠ Overload: if the load on the scale is more than its max. load (+ 9 digits), a red line "⚠ OVERLOAD" with the load and the max. load shows at the bottom of the screen (on every page - a tap on it shows the whole warning). The weight is still shown, but above the max. load it is no longer exact, and the load cell can be damaged: take weight off. The load counts against the empty scale, so a tared container counts too. Also when the sensor reaches the end of its range (then the shown weight is wrong). Scales with a load cell; an RS-232 scale shows its own overload.
Tare: the new zero is the mean weight of the half second before you pressed TARE - so the press itself does not get into it, and the display shows 0 at once. If the weight was still moving (you pressed while putting a bowl down or pouring), the scale waits until it is still (max. 3 seconds) and then tares: the weight stays visible until then, and the zero is right. The same at the start: the scale zeros itself as soon as the plate is still (max. 3 seconds after switching on), so a hand on the plate does not get into the zero. This is for your own tares (the app, the button of the scale, "T" via USB); tares of programs and machines (coffee, recipes, factory, check-weighing, sorting, espresso machines, ESP-NOW) are done at once. After a tare the scale shows 0 for a moment before it is stable again: 1 second after a button on the scale (the plate still shakes from the press), and only 0.1 seconds after all other tares (the app, USB, programs and machines) - so the first drops of an espresso are shown right away.
Automatic zero tracking (autozero) (on by default): while the scale shows zero, it keeps it exactly at zero - small drifts (warming up, temperature) are corrected by themselves, in tiny steps (max. half a digit per second, only within less than one digit around zero). This also works after you tared a bowl: as long as nothing is added, the display stays at 0. So small amounts you add slowly are not zeroed away, and the display does not jump. During a tracking run it stays off while an item is tared away. With the creep compensation on, it works more strictly (like legal-for-trade scales, OIML R76): only at the real zero of the empty scale, after taring something light (up to 4 % of Max), or when a tared container was taken off again - and all its corrections together stay within 4 % of Max.
Scale data and scale profiles
The block SCALE DATA at the top of the settings holds everything about the scale itself. On our own scales max. load, d and the Device ID as serial number are fixed and only shown. The verification interval e for the tests can be set there (empty = d): every test of the test suite judges against e - the error limits of the class, the zero-setting, the creep, the zero return and the stable equilibrium. Larger than d it must be a power of ten of the gram (0.01 g, 0.1 g, 1 g ...) and at most 10 d (OIML R 76-1 3.4.2); the app refuses other values. A larger e allows larger errors. Below it the accuracy class for the tests: detect (the default) takes the loosest class that fits e and the number of steps n and shows it in brackets. You can also test against any class yourself: the list marks the classes that fit e and n with ✓ and the others with "does not fit e and n" - they can be chosen all the same, the tests then judge against their limits. Unclassified (maker's tolerance) is for scales without an accuracy class, like most kitchen and coffee scales: type in the tolerance from the data sheet, ± in g and, if the data sheet gives one, + % of the load (e.g. ±(0.05 % + 0.1 g)). The errors are then judged against it, without the extra limits of instruments in use; creep, zero and the other limits that only the standard defines are shown, not judged. Without a tolerance the tests show the errors and their uncertainty with no pass or fail, like a calibration. The loads, Min and the number of weighings of an unclassified scale follow the loosest class that fits. The same applies to the e and the class of a scale on an RS-232 adapter. On an RS-232 adapter (232 / DIY) you set the max. load and d of the scale it is attached to here (from the data plate or the manual; this is needed to weigh), and the connection itself (baud rate, data format, scale type, tare signal) on the RS-232 page. Maker / model, serial number, e and class are optional: you do not need them to weigh - they only go into the test records and reports of the test suite and the quality records.
Scale profiles (RS-232 adapters, PRO: 3 profiles, ENTERPRISE & COMPLIANCE: 50): do you move one adapter between several scales? Everything is in one place: on top of the SCALE DATA block is the select of the scale in use, below it the data of the attached scale. Select a scale and the adapter applies its RS-232 settings, max. load and d (with its ranges) at once (it restarts if the baud rate or data format changes); its data is filled in and folded away, so it takes no room - open it to change something and press SAVE CHANGES. With - custom - the fields are open: enter or change the data and press SAVE, or SAVE AS NEW PROFILE (it takes the present RS-232 settings, max. load and d, plus the name and the optional data). The profiles are saved on the adapter, so every phone and computer connected to it sees the same list and the same scale in use. The records follow the scale: the register entry, the test history and the routine check are kept per profile on the adapter, and the test data in your browser too, so a scale never inherits the test dates of another one. "Custom" has records of its own; when you save the attached scale as a new profile while "custom" is selected, the records made so far go with it. Deleting a profile deletes its records. A change on the RS-232 page while a profile is in use updates that profile. With users, saving and deleting follow the rules of the settings (ADMIN gate, reason in the strict audit mode, not by an operator); selecting a scale is allowed for every logged-in user. The activity log notes every selection, so it shows which scale was attached from when on. In the free version the select is shown greyed out (saving and switching profiles are PRO); max. load and d can always be set.
Back up and move profiles:⬇ EXPORT PROFILES (below the select) saves a file with the data of every profile: name, maker / model, serial number, baud rate and data format, scale type, tare signal, unit, decimals, Max and d table, e and class, the certificate data, the asset name and test interval of the register, and the permanent number. It never contains the logs, the test history or the sales. ⬆ IMPORT PROFILES reads such a file on the same or another adapter. An import never deletes a profile and never changes the meaning of an old log entry: a profile of the file with the same name and serial number as one on the adapter is written into it (it keeps its slot and its permanent number), every other one is added as a new profile with a new slot and a new number. The adapter asks first and shows which profiles are updated (=) and which are added (+); the licence limit (PRO 3, ENTERPRISE 50) applies, and with users and in the strict audit mode the usual rules for saving a profile apply (ADMIN gate, a reason). Every written profile is noted in the activity log. Keep the file safe: it holds the data of your scales.
📜 Certificates & counters (folded, at the end of the SCALE DATA block): type in once what is on the data plate or certificate of the weighing instrument - the CE / metrology marking with the notified body, the EU type examination or OIML certificate, the NTEP Certificate of Conformance, the serial number of the instrument and (RS-232 adapters) the firmware of the connected scale. It is saved on the scale (adapters: per scale profile) and printed on every report and the sealed report, so an inspector can compare it with the plate. The line on top shows the two event counters of the scale: the calibration counter (+1 for every calibration saved) and the configuration counter (+1 for every change of the settings, the creep / tare guard / environment filter, the density settings or the scale profile). They count in every edition and can not be reset. Fill in only what your instrument really has - our own scales have no type approval. The reports of the DISPENSARY 🏪 MODE carry the NTEP CC number and the serial number in their header, and 📤 TO METRC warns while the NTEP CC number is missing.
⚖ Minimum weight (USP <41>) (ENTERPRISE & COMPLIANCE, on by default; folded, at the end of the SCALE DATA block): below the minimum weight of a balance the relative error of a weighing is too big for analytical work - e.g. 2 mg on a balance whose minimum weight is 10 mg. Type in the minimum weight of your balance from its calibration certificate or its qualification (USP <41> / Ph. Eur. 2.1.7) and press SUBMIT. From then on, the UP↑SCALE compares every weight above 0 with it, and below it the app asks first before it is recorded: ∑ (also a booking of the SALES MODE) and a label say "BELOW THE MINIMUM WEIGHT" with both values. Recorded anyway, the UP↑SCALE notes it in the activity log with its own weight (a manual weight of a booking: that weight). Without a value nothing is checked - the block then reminds you to type it in, so a shop or a club that never weighs at the limit of its scale is not bothered. The value is saved on the scale for every phone and computer (an RS-232 adapter keeps one per scale profile). With users only an ADMIN changes it (with a reason in the strict audit mode); every change and switching it off is an entry of the activity log. The routine check with the USP or Ph. Eur. profile shows the minimum weight it measures (2000 x s) as a help - the value you type in is yours. FREE and PRO show the block greyed out.
One scale or everything of the adapter: records of an adapter with several scales
An RS-232 adapter that is moved between scales keeps one activity log, one sales log and one archive for all of them. So that a report or an export never mixes up the scales, every record is assigned to the scale it was made on, and you choose the scope wherever it matters:
Per scale on the adapter (the profile of the scale, scale profiles): the register entry, the test history, the routine check, the certificate data and the two event counters (calibrations, configuration changes). Selecting another scale is no configuration change.
Assigned by the app: the entries of the activity log (the log notes which scale is attached: at every start, every selection, every saved or deleted profile) and the sales (every sales record carries the scale it was booked on).
The same for all scales (marked adapter): users and their roles, logins, the device register, the clock, the settings of the adapter, the seal and the licence.
Scope: a selection Scope appears on the 🗂 RECORDS page (tab 🛡 ACTIVITY) and in the sales report: This scale (the default: only the records of the scale that is attached now plus the entries about the adapter) or Everything of this adapter (all scales, a column tells which scale each entry belongs to). It applies to the lists, the CSV exports, the printouts, the sums by day / THING / user and the sealed report. For one scale the sums are made of its single sales, because the UP↑SCALE adds up the sums of the whole adapter. The download everything button of the memory warning always saves all of it.
Records for the files:SAVE FOR THE RECORDS always saves the complete log of the adapter - the chain can only be checked over all entries, so these files are the evidence. For an auditor or a customer of one scale there is EXTRACT OF THIS SCALE: a sealed file with only the entries of that scale and the ones about the adapter. Every entry carries the fingerprint before it, so Check exported archive files (and up-scales.com/check) can check each entry on its own, and against the complete files if you choose them as well. The extract also has its own chain: every row is chained to the row before it in the extract, and every row says how many entries of other scales lie before it. So the extract is a gapless trail of that scale on its own: a changed, removed or swapped row, or a hole that is not declared (also a cut-off start or end of the log: the file names the first and last entry of the whole log), is found by the check. What the other scales did is not shown; that the declared holes really belong to other scales can only be proven with the complete files (keep them, see the record files above). The file is named UPSCALE-Audit-Extract_<adapter>_<scale>_<date>_No-<from>-<to>.json, so in a folder the extracts sort by adapter, then by scale, then by date, and the extracts of different scales never mix. Choose the extract together with the complete record files (Check exported archive files, or the check page) and the check goes further: the complete files must hold the whole log from the first to the last entry named in the extract with an unbroken chain, every row of the extract must be there unchanged, nothing that belongs to this scale or the adapter may be left out, and the skipped entries must belong to other scales. Only then is the extract proven complete; if the complete files cover only a part of the log, the check says so and does not call the extract complete. An extract is no replacement for the complete files.
The sealed report of sales checks the numbers and the chain over all records of the adapter first, then shows the scope; for one scale it says how many of the records of the adapter in the period it shows. The counters inside it are compared with the records of the same scale.
Permanent numbers: every scale profile gets a number that the adapter never gives twice (P000001, ...), every device in the device register a device number (S000001, ...), like the user numbers. They stay unique when a profile is deleted and its place is used again later (new profiles take the places in a circle) or a device is removed from the list; old entries show the right scale and device.
Limits: entries of the log that are older than the first note of the attached scale cannot be assigned and are not shown for a single scale (the number is told). The sums of the sales per scale need the sales records of the newer firmware.
RS232 Scale Settings
If you are running a 232 UP↑SCALER Scale Adapter, then you can/should setup your scale once in the RS232 Scale Settings! Here you are able to configure the serial connection to your scale as like as the precision, the capacity and how you are going to tare the scale.
Open it with 🔌 RS-232 setup ▸ in the tab SCALE of the settings (or ☰ → 🛠︎ SCALE SETTINGS → RS-232). It is a page of the settings: ← under the head goes back to the settings, ✕ to the scale. On top the card 🔏 LEGAL MODE (RS-232 adapters, ENTERPRISE & COMPLIANCE), then the cards THE SCALE (weight source, scale type, hardware tare, listen only, exact weight, extra line, alibi on take, 🔍 AUTO-DETECT) and DATA FORMAT (unit, precision, capacity, baud rate and data config - a change of the two marked with * restarts the adapter), and the row 📌 Board pins ▸ (DIY only: its own page with ⇄ SWAP TX/RX and 💾 SAVE PINS). The key row: ↺ RESET reads the settings of the scale again, 💾 SAVE saves them. A field that is not valid turns red and the app says why. Leaving the page with changes that are not saved yet asks first (STAY, DISCARD, 💾 SAVE). While LEGAL MODE is on, every field and key of the page is gray (the scale refuses the changes anyway); only the LEGAL MODE card stays.
Scale Type (Presets)
Select the maker of your scale: METTLER TOLEDO (MT-SICS), KERN (KCP), OHAUS, A&D, ADAM, SARTORIUS (SBI), RADWAG, SHIMADZU, CAS, DINI ARGEO, RICE LAKE (RLWS), SMA (many US indicators: Fairbanks, Cardinal ...), NCI (Avery Weigh-Tronix, Brecknell), TOLEDO CONTINUOUS, YAOHUA (XK3190), METTLER PM / SYSTEC, ESC p (G&G and many other balances), TOLEDO POS (8213/8217/PS60) (US POS scales: Mettler Toledo Viva / Ariva, Brecknell and other scales that answer W with a weight frame), SHINKO / VIBRA (CSP) (ALE balances: switch their RS-232 output on first, setting 411 ACTIVATE = ON; 1200 baud 8N2 at the factory) - or one of the price-computing scales: CHECKOUT DIALOG 02, CAS AP-1, TVP EASY WEIGH, NCI PRICE (X), AVAWEIGH. The preset fills baud rate, data config and tare signal with the usual factory settings (e.g. A&D 2400 baud 7E1, SARTORIUS 9600 baud 7O1, SHIMADZU 1200 baud, NCI and TOLEDO 9600 baud 7E1, the others 9600 baud 8N1), and the dongle speaks the maker's protocol:
METTLER TOLEDO: the dongle starts the weight stream with SIR (and again after 2 seconds without data) and only takes the weight answers ("S S" stable, "S D" dynamic): no error or tare answers as weight.
RADWAG, SHIMADZU: the dongle starts the weight stream (C1, D03) and reads the stable mark of the scale (RADWAG "?", SHIMADZU "S"/"U"). Verified SHIMADZU balances have no stream: use their print or auto print output with LISTEN ONLY.
KERN (KCP), A&D, SARTORIUS, DINI ARGEO, SMA, NCI, TOLEDO POS, METTLER PM / SYSTEC, ESC p: the dongle asks for the weight all the time (SI, Q, ESC P, READ, W, ESC p), so the scale doesn't have to send by itself. KERN: set Mode ASK and protocol KCP in the scale's menu. Overload, error and status lines are skipped. NCI scales send a weight only while it is stable (in motion they send only their status).
OHAUS, ADAM, CAS, RICE LAKE, TOLEDO CONTINUOUS, YAOHUA: set the scale's output to continuous (OHAUS: Auto Print - Continuous; CAS: F31 = 1). The lines of a print that are no weight (OHAUS tare lines, ADAM No./Total/Date lines) are skipped. CAS: give the scale a device ID other than 10, 13 and 44 (CAS sends it as a raw byte, and these three bytes look like a line end or a separator).
RS232 WEIGHING SCALE: every other scale that sends its weight continuously.
⚖ PRICES FROM YOUR SCALE (price-computing scales): a price-computing scale works out the price itself and shows it. With these presets the UP↑SCALE takes the weight and that price (the unit price and the total) from your scale: the price unit of the THING (in the currency of your sales log) then shows the price of the scale, the SALES MODE books it (the price fields can't be changed then) and labels print it. Give the THING a price unit in your currency - its value then comes from the scale.
CHECKOUT DIALOG 02 (checkout scales of Bizerba, Mettler Toledo, Kern, Dibal and others; 9600 baud 7O1): the UP↑SCALE works like a cash register - it sends the unit price of the THING (per kg, per lb on a scale in lb) to your scale, your scale computes the price, shows it and sends it back. Set the scale to Checkout Dialog 02 (not 06). After each weighing the scale wants the item taken off before it gives the next price.
CAS AP-1, TVP EASY WEIGH, NCI PRICE (X) (NCI scales with the price command X): you type the unit price on the scale; the UP↑SCALE asks for weight, unit price and total. AVAWEIGH (334PCS): its output setting HOST, the price comes with each press of its PRINT key.
Tare on the scale: the scale prices its own net weight, so the UP↑SCALE never tares itself with these presets. CHECKOUT DIALOG: ⚞ TARE ⚟ sends the weight now as the tare to the scale; NCI PRICE: with a tare signal TARE goes to the scale; the others: TARE says "tare on your scale".
In LEGAL MODE the price of the scale is stored with the line of the scale and its number, and a label prints exactly that price.
Not part of FIND ALL - choose the preset yourself. These protocols are taken from the makers' documents; tell us how it works with your scale.
The models of a maker differ: the preset is a start, your scale's menu must have the same settings (check its manual). For every type:
a line ends with CR, LF or both;
the weight is the number of the line with a decimal point (not every digit of the line: device numbers, range digits, check characters and dates are left out), and a decimal comma works too (123,45);
lines without any digit (like "S +" for overload) and lines in pieces, percent or density (pcs, %, g/cc) are never taken as weight;
the stable mark of the scale is read where its protocol has one (for EXACT WEIGHT), e.g. "ST"/"US", "WT" (A&D), the unit that SARTORIUS and A&D send only when stable, and the motion bit of TOLEDO, SMA and RICE LAKE.
Serial Connection
If the scale is not working out of the box, check your manual for the exact serial connection settings (baud rate (often 9600), start bits, stop bits and parity) and overtake them to the app.
🔍 Auto-Detect
Don't know the settings? Press 🔍 AUTO-DETECT. With the scale type RS232 WEIGHING SCALE (you don't know the protocol) the dongle finds everything in up to 40 seconds: it listens first (does the scale send by itself?), then it asks the scale in the ways of all presets (MT-SICS, KERN, A&D, SARTORIUS, RADWAG, SHIMADZU, DINI ARGEO, SMA/NCI, ESC p) at the usual baud rates. It shows the type of the scale, the baud rate and the data config (8N1, 7E1, 7O1, 8E1, 8O1), a line of the received data, and the unit and the d (the last decimal of the weight lines) - all filled in, with the preset's tare signal. A scale with steps of 2 or 5 (d = 0.02, 0.5 ...): set d by hand. With another type selected, the dongle only finds baud rate and data config for that type (about 10 seconds). Press SAVE to use the values (a new baud rate or data config restarts the dongle). With LISTEN ONLY on, nothing is sent: only scales that send by themselves are found. Nothing found? The app tells you why: no signal at all - the scale doesn't send (switch on its output, e.g. continuous output or the print key) or the wires are crossed the wrong way (try a null-modem adapter, or a straight cable instead of one); data, but unreadable - choose the preset of your scale or set data bits and parity by hand. On the virtual scale with a scale on the USB port of the computer the search starts by itself (LOOKING FOR YOUR SCALE): press USE IT and the scale is ready.
Hardware/Software Tare
While many RS-232 scales are able to receive a TARE signal, some cheaper models are not able to do so! You are able to select between the most common hardware tare signals (like T, t, @, ESC T or ESC U for SARTORIUS, T + CR for SHIMADZU and SMA, TARE for DINI ARGEO, KTARE for RICE LAKE, !KT for ADAM Luna/Nimbus, ESC t for ESC p balances, D00KT for CAS in command mode) within the settings, best try out if one of them is working, this would be the best option! If not, no worries, you always can use the software tare option!
Board Pins (UP↑SCALE DIY only)
Only shown on a DIY adapter (also on the ESP32-C3 RS232 Adapter, which needs no change here): the GPIOs of your board that go to the RS-232 board. Choose your board (ESP32-C3 RS232 Adapter, Seeed XIAO ESP32S3/C3, Adafruit QT Py ESP32-S3/C3) - its pins are filled in - or CUSTOM and type the GPIO numbers (not the D/A numbers printed on the board) of TX (to TXD, data to the scale), RX (to RXD, data from the scale) and optionally RTS (leave it empty for none; the firmware holds it on, for scales with a hardware handshake). ⇄ SWAP TX/RX exchanges the two for wires the wrong way round. SAVE PINS saves them on the adapter, which restarts. Pins that would stop the board (flash, PSRAM and USB pins) are refused. A new adapter starts with the pins of the XIAO ESP32S3 (TX GPIO 43, RX GPIO 44, RTS GPIO 2) on an S3, or of the ESP32-C3 RS232 Adapter (TX GPIO 10, RX GPIO 4, RTS GPIO 0) on a C3 - on a XIAO ESP32C3 choose it here once.
Weight Source: RS-232 Pins or USB Port (ESP32-S3 adapters)
On an adapter with an ESP32-S3 (232 UP↑SCALER on the S3, DIY on an S3 board) the row Weight source on top of the RS-232 settings lets you choose where the weight comes from: RS-232 PINS (the RS-232 board or plug) or USB PORT: a USB-serial cable (with a CH340, CP210x or FTDI chip - the usual cheap USB-to-RS-232 cables) or a scale with its own USB port (USB serial, USB HID scales, balances that type their weight like a keyboard), plugged into the USB port of the board. Then no RS-232 board is needed. All settings below work the same way: presets, auto-detect, listen only, EXACT WEIGHT, tare. A balance in keyboard mode: set its keyboard layout to US/English, if it has that setting. Power: the board gives the device no power - use a USB OTG Y-cable with a power input (one end to the board, one to the scale or cable, one to a USB power supply or power bank). The adapter restarts when you change the source. Below the choice the app shows what is on the USB port (e.g. "CH340 cable", "USB scale (HID)", "Nothing on the USB port"). While USB is chosen, the USB port belongs to the scale: the app connects by Bluetooth (not over the USB cable), and there is no serial console. Updates still work: plug the board into your computer - when a computer is on the USB port at the start, the board stays a normal USB device for that start (the app then says "Off for this start"); or hold the BOOT button while plugging it in. The ESP32-C3 has no USB host: there the row only shows the RS-232 cable.
Listen Only (for scales without a protective interface)
Switch on LISTEN ONLY and the dongle never sends anything to your scale: no weight requests of the presets, no hardware tare (the app uses the software tare, the hardware tare select is locked), also not while auto-detecting. It only reads what the scale sends. Your scale must then send its weight continuously by itself: set its output to continuous in the scale's menu (with the presets that ask the scale for its weight, like METTLER TOLEDO, KERN, A&D or SARTORIUS, the app shows a reminder). You only need it for a scale without a protective interface. Almost every verified (legal for trade) scale has one (German: rückwirkungsfreie Schnittstelle): over its data port it takes only what its keys do too - weight request, tare, print - and nothing sent there can change its weighing result or its legal settings (see the scale's manual). Without LISTEN ONLY the dongle sends only such commands, never an adjustment, a setting, a unit or a preset tare - so the hardware tare, ALIBI ON TAKE and LEGAL MODE with the tare of the balance work (LEGAL MODE also works with LISTEN ONLY: the software tare is stored with the line). The display of the verified scale stays the legal reading: weights, prices and labels from the app are not verified by themselves. For weight or price labels used in trade, ask your verification office (in Germany the Eichamt) which setup is allowed (e.g. a scale with an alibi memory). For your own checks (like the ℮ control), statistics, counting or recipes it works as usual. The setting is saved on the dongle.
No Weight From the Scale (– – –)
When the scale on the cable has sent its weight all the time (a stream or the requests of a preset, also the weight an app relays) and then stops - switched off, its output off, the cable out - the adapter waits 2 seconds (or 4 times the usual gap between its lines), then the weight counts as unknown: the app shows – – – instead of the last weight and the red line "NO WEIGHT FROM THE CONNECTED SCALE" at the bottom of the screen, and nothing is recorded (∑, sales, labels), like a lost connection. The weight comes back with the next line of the scale. A scale that only sends when its print key is pressed keeps its last printed weight.
Exact Weight (without our filters)
Switch on EXACT WEIGHT and the app shows exactly the weight your scale sends: no averaging, no stable lock, no display filter and no automatic zero tracking - the speed and stability settings don't change the weight then. Stable is what your scale says, if its protocol sends a stable mark: METTLER TOLEDO and KERN ("S S" stable, "S D" moving), A&D, ADAM and many other scales ("ST"/"US"), OHAUS and RADWAG ("?" while moving), SHIMADZU ("S" / "U"), METTLER PM / SYSTEC ("S" / "SD"), SARTORIUS, ESC p and the A&D KF format (the unit only when stable), the motion bit of TOLEDO CONTINUOUS, SMA, RICE LAKE and DINI ARGEO, NCI (a weight only when stable). Without a mark, the app checks the stability of the exact values itself. The software tare takes the last exact value of the scale (no average). Best set your scale to grams and the precision to the scale's d: other units are converted, which can change the last digit. Together with listen only, the app shows the scale's own weight and stable mark, minus your software tare. Nothing of the app changes that weight: a virtual weight and the container of a THING are switched off in EXACT WEIGHT (they would change the result outside of the verified scale) - tare the container on the scale itself. The app also shows no center of zero here. Saved on the dongle.
Indicator view (EXACT WEIGHT)
The weight big on a second screen, e.g. for your staff or a packing table. It is no verified display: when you sell to customers, they must see the display of the verified scale itself (see What stays legal). Switch on Indicator in the basic settings (off by default, saved in this browser; switching EXACT WEIGHT on switches it on too); then in EXACT WEIGHT mode the weighing page adds below the weight the marks GROSS (no tare) and the tare (T) to the marks NET, →0← (exactly at zero) and STABLE of the display, updated with every value of the scale. ⛶ in the last line of the display (with every scale) shows the weight big on the whole screen with all marks - for a second screen or tablet (SPACE tares the scale, ✕ or Esc closes it). When the data stop, it shows – – – (see when the data stops).
Extra Line (e.g. alibi number)
Some scales send an extra line when you press their print key, e.g. the alibi number of a verified scale ("Alibi 000123"). Type into Extra line of the scale how this line starts (e.g. Alibi, up to 12 characters, upper and lower case are the same). The dongle then does not read the line as a weight, but keeps the last one (up to 32 characters) and sends it to the app. A line that starts with the word Alibi is never read as a weight, also when the field is empty.
One number, one weighing: an alibi number belongs to the weight that was on the scale when it was printed. So press the print key of your scale while the load is on it, and then record the weighing in the app - or let the UP↑SCALE press it at every take (ALIBI ON TAKE). The number shows below the weight (weighing and counting, "🔖 Alibi 000123"; ✓ once it is used) and in the weighing step of the compounding assistant - as long as this weight stays on the scale.
Where it goes: the next record of this weight gets it, each number once per kind of record:
kept by the UP↑SCALE (every app sees them, also after a reload): every sale and booking weighed now and checked by the scale (SALES, CCLUB and ROASTERY MODE - in the journal, the CSV, the sealed sales report and the transfer files of the club), every bag of fill weighing (sheet, report, CSV, its label) and every ingredient - and every part of one - of a compounding batch (batch record, also the fills of its filling step);
kept by the app: the entry you log (🍽/OK, "🔖 Alibi 000123" in the log), a value of the statistics (lot report and CSV export, an extra column only if there are numbers), an ingredient of a solution (report).
On your labels it goes with the placeholder [scaleline]: the label of a bag carries the number of that bag, a label of the weighing page the number of the weight on the scale now (a label can be printed again).
Never the wrong number: the UP↑SCALE and the app follow the same rule - once the weight changed (load off, more added, tared), the number is not used any more, also not if the same weight comes back. A record then has no alibi number rather than the one of another weighing. The last line the scale sends again when an app connects is only shown in the settings.
Tamper-evident: the number of a sale is part of the hash chain of the sales log - a changed or removed number breaks the chain like any other change (see sealed records).
This only reads, so it works with LISTEN ONLY too. The app copies the text of the scale: it does not create or check an alibi number, the alibi memory of the verified scale stays the legal record.
Alibi on take & balance ID
With ALIBI ON TAKE the UP↑SCALE presses the print key of your balance itself, at every take: a sale or booking (also every booking of the DISPENSARY 🏪 MODE - its ⚙ settings have the switch too, and the ticket number goes into the reports and the note of the Metrc rows), a fill bag, an ingredient of a compounding batch, a ∑ LOG IT line or a label. A balance with an alibi memory stores the value there and sends its alibi number; the record gets it (🔖), just as if you had pressed the key - only for the value that is still on the balance. It needs the EXTRA LINE (how the alibi line starts) and does not work with listen only.
The print command of the scale: the preset knows it for OHAUS, A&D, ADAM, SARTORIUS, RADWAG, SHIMADZU, CAS, DINI ARGEO and RICE LAKE. Other scales: choose own command and type the one from the manual of your balance (\r = CR, \n = LF, \e = ESC, \x1B = a byte in hex; up to 12 bytes).
Continuous output: if the usual request of the preset prints too (SARTORIUS ESC P), the dongle stops asking for the weight - every request would be an entry of the alibi memory. Set the balance to continuous output (auto print) in its menu; the app says so.
TEST PRINT sends the command once and shows what came back, and whether an alibi line was among it.
No alibi number came (within a second): the record is kept without one and the app says so - check the alibi memory of the balance, the cable and the command.
READ BALANCE ID asks the balance for its serial number and model (METTLER TOLEDO, KERN, OHAUS, A&D, SARTORIUS, RADWAG) and offers to save the serial number in the scale data - it then goes into the reports and the QA documents.
The commands only do what a key of the balance does: never an adjustment, a unit, a setting or a preset tare. In LEGAL MODE the setting is frozen like the other RS-232 settings.
🔏 LEGAL MODE (ENTERPRISE & COMPLIANCE)
For a verified balance in a pharmacy, a lab or a shop: LEGAL MODE makes the UP↑SCALE a data storage for the values of the balance - also for a balance without an alibi memory of its own. Open 🔏 LEGAL MODE at the top of the RS-232 settings. What it does:
Every value that is used is stored first: a sale or booking, a fill bag, a row or a part of a solution or compounding batch, a check-weighing / factory / sorting value, a pipette or density reading, a ∑ LOG IT line and a weight label. The UP↑SCALE keeps the line of the balance exactly as it came (with its stable mark), the time, the user, your software tare if there is one, and gives it its own running number (🔖 UP-000123). Only then does the value go into the record, and the record carries that number - so every value can be traced to the store. There is no button to skip it.
Only values of the balance: the value must be the value of a line the balance just sent (within half a d), and not marked as moving. A typed (✍ manual) weight, a value an app relays, a weight that is not on the balance any more or a derived amount (take-away rest, a part of a recipe) is refused - and not recorded; the app says why.
Prices of the balance: with a price-computing scale the unit price and the total your scale computed are stored with its line (a price entry right after it), and a label or a sale gets exactly that price. Other scales: the app computes the price as before, and you decide whether it goes on the label.
Tare: with a hardware tare set, the balance tares itself on the app's TARE (its net value is stored). If no stable zero comes from the balance within 3 seconds, the app warns that the balance did not tare. Without one (or with LISTEN ONLY) the software tare of the UP↑SCALE stays, and the entry keeps both: the line of the balance and the tare.
Frozen settings: EXACT WEIGHT goes on, the RS-232 settings (also Max and d from the SCALE DATA block), the scale profiles and the auto-detect are refused, and values relayed by an app (another maker's Bluetooth scale) are refused - only the cable counts.
Only an ADMIN with a reason switches it on or off (users with an ADMIN are needed); both are in the activity log and in the store itself (a switch-off leaves a visible gap, the stored values stay). A new firmware leaves a mark in the store too.
Keeping: at least 92 days (3 months; more if you set it). The size follows the free memory of the UP↑SCALE and is shown with the fill level (a warning at 80 %, with the days left at the current rate). A value is only overwritten when it is older than the keep time or archived - when the store is full of newer values that are not archived, new values are refused, nothing is lost. The UP↑SCALE must know the time (connect an app once after a power loss). While LEGAL MODE is on, a big jump of the time of an app is not taken, and old values are only overwritten after the UP↑SCALE got the signed time of up-scales.com since its last start (licence page: 🕒 EXACT TIME FROM UP-SCALES.COM, or switch on its automatic use) - so a wrong clock can't make young values look old.
Its own legal part: LEGAL MODE runs in a separate legal part of the firmware with its own version and software ID (shown in the window and the RS-232 settings as legal part v1.0 · ID …, and in the validation pack). Updates that don't touch it keep its ID; a new legal part leaves a mark in the store. Before an update you can see whether it changes: the update page shows the legal part of every firmware version (unchanged, or in red CHANGED with the version before), the first line of its changelog says it too, and the app's message about a newer firmware says whether the legal part stays or changes. App updates never change it. Designed along WELMEC 7.2.
Checking: the window shows the newest values and the software ID of the UP↑SCALE, and CHECK THE CHAIN reads every entry and checks its SHA-256 chain (a changed or removed entry breaks it). EXPORT ALL (CSV + CHECK FILE) first lets the UP↑SCALE seal the head of its chain itself, then saves every entry: a CSV to read and a JSON file with the saved bytes of each entry. Anybody can check the JSON file on up-scales.com/check: the numbers must follow each other, every readable field must fit its bytes, and the chain over all entries must end in the head the UP↑SCALE sealed.
Copies and the archive device: every qualified device that connects keeps a copy of the store in its app (at least the keep time), quietly in the background - not while a factory, check-weighing, sorting or fill program runs. One of them is the archive device (shown in the window; the first qualified device that archives, another one only by MAKE THIS THE ARCHIVE DEVICE - an ADMIN with a reason): it keeps everything and saves a sealed record file once a day by itself (e.g. UPSCALE-Legal_DIY-76439_UP-000001-UP-000250_2026-11-12.json, checkable on up-scales.com/check). Only after that file may the UP↑SCALE overwrite those entries before their keep time when it needs the space - so the store never runs full, and one device always has the whole chain, not parts on several devices. SAVE RECORD FILE NOW / ARCHIVE + FREE SPACE do the same at once. Devices that are not qualified keep no copy.
Backup: the record files are the proof for the time the UP↑SCALE no longer keeps. Let them land on backed-up, write-protected storage: the browser policy of the device qualification puts the downloads of the archive device into your records share by itself (see record storage). The copy inside the app (the browser) is no backup - a cleared browser loses it. Whether an archive outside the device is enough for your retention rules is decided by your verification office.
Records say it: a solution report or a batch record with 🔖 numbers ends with a note that its weights can be traced to the store (UP- numbers) or to the alibi memory of the balance (other numbers).
LEGAL MODE is ready today for every record you want to trace back to the balance (QA, the lab, production, your own documentation) - in GMP, GLP or ISO/IEC 17025 labs your QA validates and approves it like any lab software - the validation pack helps (validation pack). A certificate is only needed where the law demands an approved alibi memory (e.g. the documentation of a pharmacy, see what stays legal) and the balance has none of its own - with one, the EXTRA LINE puts its alibi number into every record, and that works today. For the balances without one it is planned for our 232 UP↑SCALER - until then, your verification office decides. The display of the verified balance stays the legal reading.
Scale Settings: max. load, d and the weight unit
Add the maximum capacity (in g) and the d of your scale (the smallest step it shows, on its label or in its manual) and the unit your scale sends (grams, kilograms, milligrams, ounces...). d is always 1, 2 or 5 × 10ⁿ - from 0.1 µg (ultra-micro balances) to 500 kg (weighbridges), e.g. 0.2 g, 5 g or 20 kg. The list only shows the d that fit the max. load (100 to 100 million steps, grouped in µg / mg / g / kg / t); ▾ show all d shows the rest. d is in the unit the scale sends.
Several ranges (multi-interval or multiple range scales, e.g. many shop scales: up to 6 kg d = 2 g, up to 15 kg d = 5 g): choose 2 RANGES or 3 RANGES, the d of each range and where each range ends (the last one ends at the max. load). Multi-interval: d follows the load up and down. Multiple range: the bigger d stays until the scale is back at zero. The adapter and the app round every weight to the d of its range, both in exactly the same way. The test suite judges a multi-interval scale with the e of the range of each load (and Table 4 of OIML R 76-1 for the classes); a multiple range scale with the range the scale was really in at each reading - each range is a scale of its own (3.2): on the way down the weighing test still uses the bigger d, e and error limits until the zero.
High resolution: a scale with more than 4 million steps in one range (e.g. a microbalance with 6.1 g and d = 1 µg, or an ultra-micro balance with 2.1 g and d = 0.1 µg) has more digits than the normal weight value can carry (about 7). It works with EXACT WEIGHT only (the app switches it on): the adapter then also reads the weight as an exact count of d from the digits of the scale, so the weighing page, the logs, labels and the tests get every digit. Limits: the tracking, statistics, check-weighing, factory, sorting and density modes keep working with about 7 digits (the app says so once when you open one of them); several ranges can not be combined with high resolution; the max. load of the old setting (for older apps) is a whole number of grams.
The table is saved on the adapter and goes with its scale profiles.
Connectivity
You have the option to allow (or disallow) the connection to the scale for multiple devices! This can be very useful if several persons are using the same scale at the same time - but could course trouble in public environments, so you can decide when you need it. Also there is a Weight-For-Everybody function. If this option is on, then the current weight or count on the scale will be continuously send via the public manufacturer data of the scale and can be read from every device nearby - without any prior connection to the scale! (Same like the find-me or iBeacon devices). Beside of that you will find multiple options to transmit data and receive commands via USB (Serial Port) and/or ESP-NOW! (To connect with scale software, printers or own devices!) Your labels you can print directly on label printers via wireless, USB or serial port.
Expert Configuration
You are able to control all the sensor- and averaging settings which are controlling how fast and exact the weight will be shown. You can keep all settings at automatic, but sometimes you maybe want to change something, for example to stabilize the weight better if you are in a windy environment. Click here to read more about this expert-scale-settings.
Speed presets: at the top of "Speed & Stability" in the settings, one slider sets all of them at once - from A (AUTO, the factory setting: the scale adapts the filters by itself) via 1 (super fast, nearly unfiltered) and 3 (fast, for filling and the factory mode: quick while you fill, calm when stable) to ★ 4 STANDARD, and on to 8 and 9 for animals (the mean weight over about 4 or 8 seconds, for calm or restless animals) and 10 (the calmest value, averaged over about 18 seconds - for wind outdoors or long measurements). Below the slider you see what the selected preset does, how long it averages and when the weight is stable. The single sensor and filter settings are below it under "⚙ Expert filter settings", followed by Autozero, Locking and Display (always visible) - change one of them and the slider shows "own settings". Save the settings as usual.
Zero Memory (silo, tank): for a load that stays on the scale - a silo, a tank or a container that is always on it. Normally the scale sets its zero at every start with what is on it, so the plate must be empty then. With Zero Memory on (settings, top box, below Heartbeat Lock; off by default) the scale saves its zero and a tare and starts with them instead: you can switch it off and on again with the load on it, and it shows the same weight as before. The zero is saved about 2 seconds after a tare, a slow change of the autozero at most every 10 minutes. If the load at the start is clearly below the saved zero (more than 4 % of Max), the scale does not trust the saved zero and sets a new one as usual; the settings show a red note then. A tracking run that goes on after a restart keeps its own zero. Load cell scales only (not the RS-232 adapters: the connected scale sets its own zero). Saved on the scale for all apps.
✅ Routine check before use: the first block of the ☑ TEST page (see Scale Self-Check). The short check you do before you work with the scale - every day, every shift or at the interval of your own procedure: put a known weight on a few times, the app takes every value by itself as soon as it is stable (or with ✓ TAKE NOW), each value counts minus the empty plate before it, and you get PASS or FAIL. There is no one rule for every kind of work, so you choose a profile:
Quick check (kitchen, food, HACCP): one reference weight 3 times in the middle of the plate; PASS when the mean is within your tolerance (default 1 % of the weight, at least 2 d - set what your HACCP plan asks for). There is no standard for kitchen scales: this is the common practice.
Trade (OIML R 76, in use): one weight (and optionally a second one) 3 times, 6 times for the classes I/II; every single value must be within the limit for scales in use, 2 x the maximum permissible error of the class of your scale. What your country asks for is set by the national rules.
Industry QA (two weights): a small weight 10 times for the repeatability and a big weight (at least 5 % of Max, best near Max) once for the sensitivity, with limits from your process tolerance (default 0.10 %).
Lab: USP <41> and Lab: Ph. Eur. 2.1.7 (balances for analytical work): one weight 10 times; the repeatability is met when 2 x the standard deviation divided by the smallest net weight you weigh is at most 0.10 % (a standard deviation below 0.41 d counts as 0.41 d). Ph. Eur. 2.1.7 takes a repeatability test weight of at most 5 % of Max and at least 100 mg: with this profile the app proposes one (the biggest weight of the 1-2-5 series in that range, e.g. 10 g on a 220 g balance) and the scale refuses a weight 1 outside it. Accuracy (USP, 0.10 %) or sensitivity (Ph. Eur., 0.05 %): one weighing of a weight of at least 5 % of Max - the second weight, else the first weighing of the series (with Ph. Eur. 2.1.7 type in the second weight: the first one is too small for it). The app shows the minimum weight of your scale (2000 x s): below it, the pharmacopoeias say the scale is not good enough for your weighings. Neither pharmacopoeia asks for this test every day - how often is set by your own risk assessment.
Custom (own SOP): the number of weighings (3 to 30), a limit for the mean error and, if you want, a repeatability limit.
Every check with 10 or more weighings shows the mean, the standard deviation and the minimum weight. If your test weights may be off by more than a third of a limit (the weight tolerance of the self-check data), the result says so: then the check partly tests the weight. A failed check puts a red NOT RELEASED FOR USE bar on the weighing page and the test page until a check passes: find the cause (dirt, level, place, test weight, calibration - CALIBRATE NOW is offered) and press ↻ RE-TEST. The bar only warns, the scale keeps weighing. REMOVE WARNING takes it away with a reason that is logged - on a scale with users only an ADMIN can do that, without users (or without an ADMIN) everybody. Minimum interval (off, every 4 / 8 / 12 hours, every day, every 7 / 30 days) is kept by the scale, together with the time of the last passed check and an open failure, so every phone and computer sees the same: when the interval has passed since the last passed check, a yellow line with a START button shows at the bottom of the weighing page of every phone and computer until the check is done (✕ hides it until the app is opened again). A failed check shows the red line ⛔ NOT RELEASED FOR USE (RE-TEST, REMOVE WARNING) on every device. The lines stay until a check passes or an ADMIN removes the warning or switches the interval off. On a scale with users only an ADMIN can set the interval (the scale refuses it for the others, in the strict audit mode with a reason); without users everybody. Changes of the interval, results and removed warnings are entries of the activity log on ENTERPRISE scales. The last checks of this browser are listed with the result, s, the minimum weight and the user (the browser keeps the last 1000; all of them are on the 🗂 RECORDS page, tab ✓ ROUTINE). With ENTERPRISE & COMPLIANCE, EXPORT LOG saves them as CSV / JSON, and PDF REPORT makes a sealed PDF of all of them: the period, how many checks passed and failed, whether every failure was followed by a passed re-test (or a removed warning with its reason), how often the minimum interval was missed, and every single check with the user. It documents that the scale was checked before use (with the report number, the integrity code and the UP↑SCALE SEAL like the other reports); right after a check, 📄 PDF OF THIS CHECK saves that one check (scale, time, user, every value, the limits and PASS / FAIL) the same way; on ENTERPRISE scales every check and every removed warning also goes into the test history of the scale (a removed warning shows ⚑) and into the report of the test suite. The check runs with your settings, like in daily use - it is your own check, no verification and no calibration; the names USP and Ph. Eur. only say which rules the profile follows.
🕒 Clock of the balance (RS-232 adapters): a balance with its own clock puts its time on its printouts and in its alibi memory. Switch Clock of the balance to ASK AT EVERY CHECK in the routine check block, and every check first asks whether that clock shows the right date and time (look at its display or a printout). ✓ YES, RIGHT or NO / NOT SURE is kept with the check - in the list of checks, the reports, the CSV and the LIMS / ELN forwarding; the check runs either way. A wrong clock is set in the menu of the balance. The UP↑SCALE keeps its own time for its records, so this is only about the balance. Saved in this browser.
⚖ Calibration check: on the ☑ TEST page (see Scale Self-Check). Is the calibration still right? Type in your test weight (best a real test weight, e.g. 1 kg, or the weight you calibrated with), start the check with an empty plate (the scale goes to the weighing mode and is tared once), and put the weight on the middle of the plate 3 or 6 times - each value counts minus the empty plate before it, and everything is taken by itself as soon as it is stable (or with ✓ TAKE NOW). You get the mean difference to the test weight in the unit, in d and in %, and the spread of the values. How many loadings: like the verification in OIML R 76-1 (A.4.10): 3 for scales of the classes III and IIII, 6 for the classes I and II - the app takes the class the self-check finds for your scale (the 420 and the 710 fit class II: 6). The repeatability test of the self-check uses 10 loadings, like the type approval. Up to 0.05 % of the weight (min. 2 d) the calibration is right, up to 0.1 % (min. 5 d) acceptable - a test weight has its own tolerance (e.g. 1 kg of class M1: ±50 mg), so a stricter limit would test the weight, not the scale. Where these limits come from: no standard sets one verdict for a calibration check. The 0.1 % is the accuracy limit of USP <41> for balances (there: one weighing of a weight between 5 % and 100 % of the max. load); OK at half of it and the minimums of 2 d and 5 d are our own rule. The judgement by OIML R 76-1 is the class result: in the report, the mean error is also judged against the limits of the classes, like every other test: the uncertainty is half a step for rounded readings (a tenth of a step for readings before rounding) plus the tolerance of your test weight. Taking the mean does not make rounded readings more exact (6 loadings that show the same value all carry the same rounding error), so the mean of 6 loadings counts as uncertain as one reading. If the error is more than acceptable, CALIBRATE NOW starts the calibration. The reminder to calibrate again (every 30 days to 1 year) is in the settings too.
⌖ Eccentricity test (corner load): part of the Scale Self-Check on the test page (tab 🔍 SELF-CHECK). It checks if a weight shows the same in the middle and on the 4 corners of the plate. Take one weight of about 1/3 of the max. load, start the test with an empty plate (the scale goes to the weighing mode and is tared once) and put the weight in the middle, then front left, front right, back right and back left - each time in the middle of that quarter of the plate, not on the edge. The test goes 3 times around, and everything is taken by itself: first the empty plate (the zero), then the weight as soon as it is stable for a second. Each value counts minus the empty reading before it (like the usual test procedures: zero before each loading), so there is no tare in between. Then lift the weight off and put it on the next position. A status line shows what the test waits for, the current weight and when the next value is taken. If a value is not taken by itself (e.g. the weight does not get stable long enough), press ✓ TAKE NOW: it takes the current weight for the step that is running. The overview shows the 3 exact values of each position, their mean, their spread and the difference to the middle in d (1 d = the smallest step of the scale), and at the end the repeatability (standard deviation) and the zero drift. If Autozero is on, the test shows a warning: Autozero moves the zero by itself while the plate is empty, so switch it off for the most exact test. The result: the biggest difference of a corner mean to the middle mean (the method of EURAMET cg-18, it needs no calibrated weight) is compared with the maximum permissible error of OIML R 76 for the test load, in the class of the scale (the class you entered in the self-check, else the loosest class that fits e and n; e.g. class III: 0.5 e up to a load of 500 e, 1 e up to 2000 e, 1.5 e above). The uncertainty of the readings is taken into account: rounded readings make the difference of two means uncertain by up to 1 d (the mean of 3 rounded readings is not more exact than one), readings before rounding by about 0.2 d - always the d of the range the test weight is in. ✓ = surely within the limit, ≈ = cannot be decided with these readings, ~ = above the limit but surely within twice the limit (the limit for instruments in use), ✗ = above that. With the true mass of the weight (optional, the field above the start button: the conventional mass from its calibration certificate, not the nominal value) every single corner loading (each of the 3 rounds, not their mean: OIML R 76-1 judges the error of each weighing by itself) is also judged against that mass like OIML R 76 A.4.7 - only the corners, the middle is not part of it; the uncertainty adds the rounding of the readings and the tolerance of the weight that you entered in the self-check data. Then that result counts for the report and the classes, and the difference to the middle is shown for information. Without the mass, the comparison with the middle is the result. This is a self-check - not a verification of the scale. More than that: check that the plate sits freely and the scale stands level and firm. The result also depends on the speed and stability settings, so test again after you change them.
⚖ Creep compensation (optional, off by default, in the "Speed & Stability" block of the settings - the scales with a load cell, not the RS-232 adapters):
🌍 Environment test (at the end of the "Speed & Stability" block in the settings; load-cell scales and the RS-232 adapters): how steady is the place where your scale stands? Put the scale on its table (empty or with a fixed load), press START ENVIRONMENT TEST and do not touch the table for 3 minutes. The scale measures its own unrest (how much the readings swing within 10 s for averaging times from 25 ms to 12.8 s, plus the slow drift of the weight - weighing standards have no such test) and the app shows the noise for every averaging time, a graph of the weight during the test with its trend, what kind of unrest it is (noise, vibration or a soft table, drift, or too restless) and the fastest speed preset that keeps the display steady (noise up to 0.18 d: the display then swings less than about one step) next to the one you use now. USE THE RECOMMENDED SETTING takes it over and saves the settings - nothing changes without your click. It works in both directions: a place that is steadier than you thought gets a faster preset. With drift or a too restless place no preset is offered (no filter fixes that). If something is put on or taken off the plate during the test, it starts again; it stops when the scale leaves the weighing/settings page. A new scale is still warming up: in the first 30 minutes the drift can be warm-up, the app tells you - repeat the test later. At an RS-232 adapter the test measures the values your scale sends (the adapter filters them again in the app); in EXACT WEIGHT mode the adapter filters are off, so only the steadiness is shown and no preset is offered. The result is a recommendation for this place and these 3 minutes, not a measurement result.
🌬 Air buoyancy correction (ENTERPRISE & COMPLIANCE, off by default; own block after the "Speed & Stability" block in the settings): a balance is adjusted with steel weights (8.0 g/cm³), so it shows the conventional mass (OIML D 28). A sample of a lower density pushes away more air and shows a bit less than its true mass: water about +0.105 %, ethanol about +0.137 %, a powder of 1.5 g/cm³ about +0.065 %, steel 0. It is a relative error, so it grows with the weight: on a scale with d = 0.01 g, 10 g of water are already one digit; on an analytical or micro balance at an RS-232 adapter it is many digits. Switched on, the UP↑SCALE calculates true mass = reading x (1 - ρ air / 8.0) / (1 - ρ air / ρ sample) for every THING with a density (the app sends the density, the scale answers with the factor of its air density) - its volume unit, typed in as g/ml or saved by the density mode. Without a density nothing is corrected (never a guess). The reading itself is never changed: the true mass is shown and recorded next to it. Air density: the preset is 0.00120 g/cm³; type in your own value, or choose "from temperature, pressure and humidity" and the UP↑SCALE calculates it with the formula of OIML R 111 (the app shows it before you save) (use the pressure at the place of the balance, not the sea level value of a weather report; the air changes with the weather - set it again regularly). 📡 FROM THE SENSORS fills temperature, pressure and humidity from a room sensor; 📡 Keep up to date from the sensors (this app, off until switched on) then sends new values by itself when the air density changed by 0.2 % or more - at most every 15 minutes, never while a recipe or a solution is open; with the strict audit trail it asks first (APPLY, then the reason). The weighing log and the LIMS forwarding then name the sensors of the air. When the sensors send no more, the scale keeps the last air density it got and the chip of the modes says so (sensors gone - the air of 14:32 stays). SUBMIT saves it on the scale for every phone and computer (one value per scale, also with several scale profiles: it is the air of the room); the block shows when and by whom it was set last. With users only an ADMIN changes it (with a reason in the strict audit mode); every change is an entry of the activity log with the old and the new air density, and it counts as a configuration change. While a recipe or a solution is open on the scale the values stay fixed. Not used where the conventional mass is the rule: calibration, the tests and the routine check, the minimum weight, check-weighing, sorting, counting, sales, the club and the stock. Where it is used:
Weighing: with a THING loaded, a line under the big weight shows "🌬 true mass" with the density and the factor (only when the correction is at least half a digit). Its volume units (ml, cups ...) are calculated from the true mass.
Solution mode and recipes: the goals are true masses. The app sends the density of every ingredient to the scale when the solution or recipe opens, the scale works out the correction and judges green / red with it - so a row is green when its true mass is right, and 🧮 recalculates in true masses. The rows show the true mass (🌬), the report lists the true mass with the reading, the density and the factor below it and the air density in its head. The app needs the values of the THINGS: when a solution opens (after starting the mode and typing the rows, 📂 LOAD or 💾 SAVE) with THINGS whose values this app has not read yet, and nothing is weighed yet, it reads them by itself in the background (the solution stays open) - a THING added after the first weighing is corrected from the next LOAD or SAVE on. Recipes read them on their start page.
Density mode: the result includes the air: density = weight in air / displaced liquid x (ρ liquid - ρ air) + ρ air - so a THING saved from it has the true density. Saved results that were corrected have a 🌬 in the list (the air density is in its tooltip and in the 📋 copy); results without it use the plain formula.
The weighing log (∑) adds the true mass to a line of a THING with a density, and the LIMS / ELN forwarding sends it as air_buoyancy (true_mass_g, factor, density_g_cm3, air_density_g_cm3) next to the reading grams - for weighings and the live weight. The inventory is always booked with the reading.
🌍 Environment filter (AUTO / OFF, off by default, below the environment test; load cells and RS-232 adapters): with AUTO the UP↑SCALE does the environment test by itself from time to time, in the background and only while it stands still, and adapts the averaging: slower where the table is restless, faster where it is very steady. It starts 30 minutes after the scale was switched on (warm-up), changes something only when two measurements in a row (10 minutes apart) ask for the same preset - and only if the present setting is clearly too noisy (above 0.20 d) or clearly over-filtered (under 0.09 d) - and never when the weight drifts or the place is too restless. Only the sensor speed, the floating average and the smoothing change; stability, lock and display settings stay yours, and your saved settings are never changed (a restart goes back to them). The app shows a note as soon as the scale changed something, with an UNDO button (the automatism then waits 60 minutes). If you change the averaging in the settings yourself, it steps back. Not in EXACT WEIGHT mode of an RS-232 adapter.
🛑 Tare guard (next to the environment filter, on by default): a scale must not tare or set zero while the weight is still moving (OIML R 76-1 4.4.2, 4.5.4, 4.6.8). With the guard on, every tare waits up to 3 seconds for a weight that is stable for at least a quarter of a second; if the table or the load keeps moving, the tare is refused and the app shows a short note. That is a tare from the app, the key of the scale, USB or a WATCH LINK - and also the tares of programs and machines (coffee, recipes, check-weighing, factory, sorting, counting, the TEST MODE, ESP-NOW, the Decent Scale protocol of espresso machines). A short disturbance is no problem: the scale waits and then tares. A program step that needs its tare (the empty container of the counting, the start of the check-weighing or the sorting) goes on only when the tare is done - refused, the step stays and you press its key again. A tare inside a step (the next ingredient of a recipe, the cup of a coffee, a mold of the factory mode) is done by the scale itself as soon as the same load is calm again; if another load is on by then, only TARE does it. A machine that tares while it still vibrates (e.g. an espresso machine at the start of a shot) can be refused then: switch the guard off for it - the one switch is for all tares. The zero at power-on waits for a stable weight too, but is never refused. In EXACT WEIGHT mode of the RS-232 adapter the guard is off by itself.
What is creep? Put a weight on your scale and leave it there: over the next minutes the reading slowly changes a little - every load cell does this, cheap ones more. Take the weight off, and the zero slowly comes back. On a fine scale (0.01 g) you can see it easily: a few hundredths of a gram over half an hour.
Autozero or creep compensation? The autozero only corrects at zero (empty scale, or a tared bowl with nothing added). The creep compensation corrects the creep at every weight, from 0 to full load - also while you weigh, and also the slow return after taking a load off.
How it knows your scale: every load cell creeps differently, so it is measured once: ⚖ MEASURE CREEP opens the tracking mode with the creep test preset (a point every second, no alarms). Follow the steps on the run page: empty plate for 1 minute, put a weight of half to full load on the scale and do not touch it, leave it 30 minutes, take it off, wait 15 minutes, then CALCULATE CREEP. Do not tare during the test. The app shows how much the scale crept, what is left with the compensation, and checks the return after taking the weight off. SAVE AND SWITCH ON saves the values on the scale. Repeat the test after a new calibration or a repair.
While it is on, the autozero works more strictly (see above): the creep under a tared bowl is removed by the compensation, and slow real changes of an item (drying, evaporation) are never zeroed away. The tracking mode is PRO (the free version: 15 minutes in a row, then a 10-minute pause - enough for a short test, a full creep test needs PRO).
Scale Number
If you connect with your scale, it will identify with a number (like #45678). This number was randomly generated at the first start of the scale and is only stored there, we do not know (or store) it at all. If you don't like your number, or if you want to change it for any reason, select the checkbox at the settings page and a new number will be generated for you!
⚖ Scale Self-Check: test your scale and get a report
Where: all tests, the checklist and the reports are on their own page: ☑ TESTING MODE in the menu (☰ → ⚖ MODES → ALL MODES). The head shows ☑ TESTS small and SCALE TEST large with its ℹ️, the ✕ goes back to weighing. Below the head are five tabs (the app remembers the last one): ✅ ROUTINE (the routine check before use, the short check of every day, with the last checks of this browser and their export), 🔍 SELF-CHECK (the checklist of the tests, the strict procedure, a card per test with its loads and ▶ START, the scale data, the inspector and the report), ⚖ CALIBRATION (the date of the last calibration and the reminder, the calibration check, and the tests that belong to settings: environment test and creep - a row opens the settings at that place), 📑 VALIDATION (the quality records, the validation of the calculations, the validation pack and the VSR) and 🖥 DEVICES (the device qualification). Each tab has its own key row at the bottom, the main action at the right end (✅ ROUTINE: ▶ START ROUTINE CHECK, or ↻ RE-TEST after a failed check; 🔍 SELF-CHECK: 📄 FINAL REPORT with ENTERPRISE & COMPLIANCE, open report and HTML behind ⋯; ⚖ CALIBRATION: CALIBRATE NOW | ▶ START CHECK; 🖥 DEVICES: 🖨 BLANK FORM | 📋 DEVICE REGISTER | ⋯ (templates) | 🖥 DEVICE QUALIFICATION FORM). The ▶ / ↻ at the end of a row of the checklist goes to the card of that test (also on another tab). A field the scale refuses (a load, the test weight) gets a red frame.
While a test runs its card is right below the weight: the head shows ☑ TEST small and the test large, the card the thing to do now in large letters, the weight now below it, and "Value n of m" with a thin bar (the eccentricity test also shows the plate with the place of the weight). The key row: ABORT at the left end (asked first: the values so far are not kept) and ✓ TAKE NOW at the right end - no TARE, the scale tares by itself. The result stays on the page with the verdict in the colours of the weighing results (✓ green, ≈ yellow, ~ orange, ✗ red), the values, the classes and the help folded at the end; the keys: ↺ RESTART and ⎙ OPEN REPORT (PRO: a plain print of this test; ENTERPRISE & COMPLIANCE: numbered and sealed) - the ✕ of the head goes back to the test page, at the tab of the test.
On the test page, the tab 🔍 SELF-CHECK lets you test a scale with the steps of the official test procedures (OIML R 76 / EN 45501 and the EURAMET guide cg-18) and gives you a report you can print or save. What each edition gets is in the editions of the tests. It works with all UP↑SCALE variants and - with the 232 / 323 adapter - with the scale you have connected to it, e.g. a verified (legal for trade) scale.
The UP↑SCALE runs the tests: it guides you step by step, takes every value by itself when the weight is stable, judges it and saves the settings of the tests and the last result of every test. Every test starts with a tare of the empty plate: if something is still on it (e.g. the weight of the test before, also when it is tared and the display shows 0), the test waits until you take it off (if the plate is empty but shows a little, press ✓ TAKE NOW). Every phone, tablet or computer connected to it shows the same test and can take part (TAKE NOW, ABORT, the changeover steps ΔL), and any of them can make the report later. A running test goes on when you close the app (the creep test takes up to 4 hours): open the app again and it shows where the test is. It ends when it is complete, with ABORT, when you choose another page on the UP↑SCALE, or when the UP↑SCALE restarts. With the adapter, the settings and results belong to the scale profile in use.
It is a self-check, not a verification and not a calibration certificate. Only an authorised body can verify a scale. But the self-check can show you early what such a body would find:
Before you call the verification office: a failed appointment costs the call-out and the re-verification, and the time your scale is out of service. If the self-check shows a clearly bad corner, a drifting zero or an error that is too big, you can first fix the cause (level, a loose plate, dirt, calibration) - and you only book the appointment when the scale passes.
Before you buy a used scale: a second-hand lab or legal-for-trade scale costs a lot of money. Bring your test weights and your phone and check it on the spot - repeatability, all four corners, up and down - before you pay. A scale that fails the self-check may cost you more in repair or re-verification than the price difference to a good one.
Regularly: run the same tests every few months with the same weights and keep the reports - then you see a change early, not when a customer complains.
Scale data and classes (all optional)
The limits come from the maximum permissible errors (mpe) of the accuracy classes: 0.5 e up to a first load step, 1 e up to the second and 1.5 e above (for class III: up to 500 e, up to 2000 e, above; class II: 5000 e and 20000 e; class IIII: 50 e and 200 e; class I: 50000 e and 200000 e), where e is the verification scale interval on the data plate. The app takes d (the smallest step) from the scale and uses e = d unless you set a larger e - on an RS-232 adapter (232 / DIY) the e of the data plate of its scale, on our own scales an e for the tests (a power of ten of the gram, at most 10 d). The e, the class and the serial number are set in the SCALE DATA block of the settings (per scale profile on an adapter); the test page shows them with a link there. From e and the number of steps n (max. load divided by e) the app works out which classes are possible for this scale and tests against the loosest one - or the class you pick in SCALE DATA (from the data plate, or any class you want to test against), or the maker's tolerance of an unclassified scale. On an unclassified scale every check whose limit is the mpe of a class (also creep and warm-up) is judged against the maker's tolerance; a check with a fixed limit of its own in the standard is shown as not judged. In the report you see for every class that fits if the scale meets its limits, only meets the limits of instruments in use (twice the mpe), cannot be decided or fails. So you do not have to know the class before you start. The serial number (for our own scales the Device ID, which is the MAC address, is filled in for you), remarks and the tolerance of your test weights are optional too (the tolerance of the weights is added to the uncertainty of every result).
The tests
Every test is guided: the status line tells what to do (leave the plate empty - put this weight on - take it off) and each value is taken by itself as soon as it is stable for a second. Each value counts minus the empty reading before it, like in the usual procedures. The scale goes to the weighing mode and is tared once at the start. Use ✓ TAKE NOW if a value is not taken by itself. The newest value is always on top of the table (the report lists them from the first to the last).
Test settings: while a test runs, the scale switches the filters that would change a reading by itself - only for the test and only in its memory, your saved settings are not touched: the stable lock and the display filter are off and the stability tolerance is the default (1 d). Autozero is on for the repeatability test and the calibration check (OIML R 76-1 A.4.10 asks for it: the scale is tested as it is used), off for the eccentricity test (the zero must not move between the corners), and as you set it for the weighing and tare tests. The panel shows 🧪 while the test settings are on. They end when the test ends or is aborted, when you leave the test, or when the scale restarts. All apps connected at the same time see the test and the unfiltered weight. The routine check runs with your own settings: it checks the scale as you use it. The creep compensation stays as you set it: it is part of how the scale measures.
Where the filters are used and where not: the stable equilibrium test always runs with your own settings (stable lock, display filter and tolerance are exactly what it tests, and the shown values are what counts). The weighing, tare and warm-up tests use the weight before rounding with the filters off - or, with the strict procedure and the legal display, your filters and the shown weight with the changeover step ΔL. The zero-setting, creep & zero return and eccentricity tests run with the autozero off (the zero must not move by itself). The discrimination test reads the display, because it asks what the display shows.
🎯 Strict procedure (switch in the tab 🔍 SELF-CHECK of the test page, off by default): makes the tests follow OIML R 76-1 (Annex A.4) as closely as a self-check can, and tells in the report which rules were met. Loads: filled in as the standard asks (Min, the loads where the mpe changes, Max; at least 5 for a verification, 10 for a type examination, A.4.4.1; the tare test with a tare of half of Max, A.4.6.1). Preload to Max once before each weighing test (A.4.1.10). Legal display: the shown weight is used, and the error before rounding comes from the changeover step +ΔL at every load (A.4.4.3: add small weights of about 1/10 e until the display jumps up by one step, and count them with the + / − buttons of the row: one click is 1/10 e; the zero readings get a row of their own with the same buttons, because the standard corrects every error by the error at zero E0 found the same way - without it the rounding of the zero stays in the uncertainty) - or, if you prefer, the weight before rounding as the reading of a device with a smaller interval (A.4.1.6); the filters of the scale stay as you use it. Repeatability by A.4.10: one series with about 0.8 Max, 3 weighings (classes III and IIII) or 6 (classes I and II) for a verification; two series with about 50 % and 100 % of Max, 10 weighings each, for a type examination - a known test weight is needed. Warm-up: the scale should be on for 30 minutes (our rule: R 76-1 asks that the scale is switched on throughout the tests, A.4.1.3, and its warm-up test A.5.2 ends after 30 minutes) - the app knows the time since the scale started and asks before it starts earlier. Test weights: OIML R 111 with an error of at most 1/3 of the mpe (3.7.1): enter their tolerance, the checklist checks it. Room temperature at the start and the end (optional; with a room sensor the app takes them at the start and the end of the test when the fields are empty - a typed value stays): steady means a change of at most 1/5 of the temperature range of the class, not more than 5 °C (A.4.1.2). The checklist in the report lists every rule with ✓ met, ✗ not met or ? not known. Not part of the self-check: temperature, tilt and voltage tests (A.5), the range of the zero-setting (A.4.2.1) and the checks of the manufacturer's documents. It stays a self-check: no verification, no legal value.
Repeatability: the same load 10 times, with two loads (about half of the max. load and about the max. load, or only one if that is all you have) - like the type approval in OIML R 76-1 A.4.10 (a verification uses only one series of 3 weighings for classes III/IIII, 6 for classes I/II). No error may be bigger than the mpe, and the difference between the biggest and the smallest error must stay within the mpe. No weight with a known value? Leave the load field empty: the test asks you to put any weight on (best about half of the max. load up to the max. load), takes the first loading as the reference (rounded to d, it is not counted as a test), and all 10 loadings after it are compared with it. Then only the spread is judged - the repeatability itself (R 76-1 3.6.1 asks only that the results of the same load agree) - not the error against a known weight. Normal conditions or raw: the checkbox above the load field (only on scales with filter settings; PRO licensed scales always test under normal conditions, RS-232 adapters always use the unfiltered value) lets the test run with your filters on, like in daily use, instead of the raw readings; the report tells which one was used. Results in plain words: after a test the app shows the biggest deviation in percent with an example at a third of the max. load, explains what "undecided" and "in use limit" mean (a scale in that range is still accurate enough for everyday, non-legal use) and, if the tested class is not met, the same checks against the next looser class that is met. The classes are explained in a short list there and in the report.
Accuracy & linearity test (weighing test up & down): the loads rise step by step and fall again, then the empty plate (zero return). The error at every load must stay within the mpe, going up and going down. The difference between up and down at the same load (hysteresis) is shown for information only: OIML R 76-1 sets no limit of its own for it. The result also shows the linearity: the largest deviation of the values from the best-fit straight line, as digits and in % of the biggest load, and the gain error (slope). This is the weighing test of OIML R 76-1 A.4.4 - the standard has no separate linearity test; the error curve against the mpe is the proof. The standard uses at least 5 loads (10 for the first test of a type), including Min (only if Min is at least 100 mg), the loads where the mpe changes and Max.
Tare test: the same with a tare weight that is put on and tared first: does the tare function keep the accuracy?
Eccentricity test (corner load) and the calibration check: see above - their results go into the report too.
Discrimination test (type examination, d of at least 5 mg; OIML R 76-1 A.4.8.2 and 3.8.2.2): at three loads (Min, 1/2 Max, Max) you put the load and ten small weights of 1/10 d on the scale, take the small weights off one after the other until the display drops by one step, put one back and gently add a weight of 1.4 d - the display must then show one step more than at the start. The app follows the display and tells you each step; it needs small weights of 1/10 d and 1.4 d (10 mg and 140 mg on a 0.1 g scale). On a scale with several ranges a load is never right at the end of a range (Max1, Max2): the small weights would carry the display into the next range, where it steps by a larger d, and the test would fail although the scale is right. So the suggested loads stay at least 2 d below the end of a range (e.g. 149 g instead of 150 g), and the scale does not start the test with such a load of your own. R 76-1 has no threshold test for digital scales (the sensitivity test A.4.9 is only for non-self-indicating instruments).
Creep & zero return (classes II, III, IIII; A.4.11): half an hour with a load close to Max, the app reads the indication every minute. Creep: at most 0.5 e change within the 30 minutes and at most 0.2 e between minute 15 and 30. If they are not met, tick go on to 4 hours before the start: the standard then allows the change within four hours to stay within the maximum permissible error at the load (3.9.4.1) - the app goes on by itself after the 30 minutes (a reading every 5 minutes), the 30 minute rows are then shown for information and the 4 hour result decides. Keep the page open and the room temperature within 2 °C for the four hours. Zero return: after taking the load off, the zero may differ by at most 0.5 e from before. On a multiple range scale (3.9.4.2) it is 0.5 e of the range of the load, and the test adds 5 minutes with the empty plate: the zero then may not move by more than e of the first range. The autozero is off during this test. On our own scales the readings are taken before the rounding (a reading device finer than 1/5 e, A.4.1.6) - with readings rounded to d these small limits can not be decided. Both are evaluated from the same half hour. Keep the table still.
Accuracy of the zero-setting (A.4.2.3.1, 4.5.2): with a load on the plate the app presses the zero key (a tare within 4 % of Max is a zero-setting, the range of the zero-setting device, 4.5.1); the indication at zero must then be within ±0.25 e. Up to five different loads, exchanged each time. With the weight before rounding the zero is read directly; with the rounded display the changeover step ΔL gives it. Also checked: the zero indicator (4.5.5) - the dark green zero must be on exactly when the unrounded zero is within ±0.25 e (only with the weight before rounding). The autozero is off during this test.
Zero tracking (autozero, A.4.2.3.2): no test of its own. The standard measures the error at about 10 e, outside the range where the tracking works; the weighing test does that at every load (the first is Min: 10 to 100 d, depending on the class) with the autozero as you set it, so a tracking that eats into the load shows up as an error. The report lists it in the checklist as "not tested separately".
Warm-up time test (A.5.2): for a scale that was off for at least 8 hours. Switch it on, connect the app and start the test at once: zero and a load close to Max right after switching on, then after 5, 15 and 30 minutes (the app asks you shortly before each time to take the load off and put it back; no preload). Each result is corrected for the zero error of that moment. The time comes from the scale (its minutes since the start); the report shows the real minute of every measurement. The app asks whether the scale was off for 8 hours - it cannot know that itself.
Stable equilibrium test (A.4.12, 4.4.2): about 50 % of Max on the plate; you disturb the equilibrium once (tap the table or push the plate), five times in a row. The moment the scale says "stable" again counts as the print command: the values shown in the next 5 seconds may be no more than two adjacent ones (a span of at most 1 e). The zero and tare keys during a disturbance are not tested here: the tare guard refuses them.
Rounded readings and the changeover step
Weight before rounding: while a test runs, the scale also sends its weight before rounding (and before the stable lock and the display filter) - the tests take that value. (Outside the tests it sends only the shown weight: fewer packets while the weight stands still.) That is what the standard allows for a scale with a display of a smaller step (OIML R 76-1 A.4.4.3, at most 1/5 e): no changeover step is needed, and the report says the weight before rounding. The weight on the display stays rounded. On the RS-232 adapter the value is the one of your scale: switch on EXACT WEIGHT in the adapter setup, then each value is taken exactly as your scale sends it - if your scale rounds it, the app sees that all values sit on the steps (below); if it sends finer values (a service mode), the app sees that too. Without EXACT WEIGHT the adapter averages the values, and an average of rounded values only looks finer: the tests then use the shown weight. Everything below is for scales that send only their rounded display.
A scale shows its weight rounded to the step d, so one reading is uncertain by half a step. The app checks this from your values: if every reading sits exactly on the steps of the scale, the readings are rounded and the half-step uncertainty is used; if some readings fall between two steps (e.g. a verified scale with a service mode that shows ten times the resolution), the scale gives finer values and only a tenth of a step is used. A result is therefore marked ok only if it is within the limit even in the worst case, undecided if it is that close to the limit that the rounding could decide, and too big if it is clearly over. For an undecided row the app shows an input +ΔL: use the changeover method of the standard - add small weights of 0.1 d one after the other until the display jumps to the next step, and type in the total of the added weights. The app then calculates the exact error (E = I + d/2 - ΔL - L) and judges again. You only need the fine weights for the rows that are undecided. Example: a 420 shows 1000.0 g with a 1 kg test weight; you add 0.01 g pieces and the display jumps to 1000.1 g after 0.03 g: the value before rounding is 1000.0 + 0.05 − 0.03 = 1000.02 g, so the error is +0.02 g = +0.2 d. In practice few people have weights of 1/10 d - the reason why the scale sends the weight before rounding.
The scale does the following by itself while a test runs (see test settings); to set it by hand: switch the stable lock and the display filter off and set the stability tolerance to the default in the scale settings (the test warns you if the lock or autozero is on) - a big tolerance calls the weight stable while it is still settling, the display filter holds the shown value until it differs by 1 d for a second, the lock holds it until it moves by 0.75 d. On an RS-232 scale use LISTEN ONLY and EXACT WEIGHT in the adapter setup, so the adapter passes on the weight exactly as the scale shows it. Use test weights of class M1 or better, put the scale on a firm and level ground, and let it warm up first.
Checklist, inspector data and final report
At the top of the tab 🔍 SELF-CHECK the checklist lists every test with its status (not done / done with the result ✓ ≈ ~ ✗ and the time), a START / AGAIN button that jumps to the test, and for each test whether it is asked for a verification / conformity assessment (OIML R 76-1 Annex A: needed, or type examination = only asked when a type is examined), an inspection in use (typical: a usual set - what is asked depends on your country, please check your national rules) and a calibration (EURAMET cg-18: repeatability, eccentricity, errors of indication). The columns are short so the table fits a phone: VER (verification / conformity), USE (inspection in use), CAL (calibration) with ● asked, ○ typical, △ only for a type examination, – not asked; the first column shows the status (⬜ not done, ✓ ≈ ~ ✗ the result), explained in a legend below the table. A filter shows only the tests of one procedure, and a line per procedure counts how many asked tests are done. The same table opens every report.
Inspector data (optional, below the tests): name, organisation, inspector ID, place, report number, remarks and a signature that can be drawn with finger or mouse. It is printed on every report and saved only in this browser, never sent anywhere. The FINAL REPORT button builds one record from the checklist and all results: header with the scale and its settings, summary per procedure (which asked tests are done, whether the done ones stay within the limits, what is missing - no verdict), all test tables, and at the end fields for date, signature, stamp and remarks to complete by pen, with the line "values checked against the display of the scale". With an inspector name or signature the report says that the signing inspector is responsible for its content (and names the EXACT WEIGHT mode if it is on); without, it keeps the note that it is a self-check without legal value. Can an inspector use it? The inspector reads the legal display of the scale himself while testing; the app (with the RS-232 adapter in EXACT WEIGHT mode: the value of the scale unchanged) only records the readings. Whether such a record is accepted is the decision of the authority or the accredited body - UP↑SCALE is not affiliated with any of them and the report never claims to be an official certificate.
Why test your own scale? Good for everybody, not only for officials
You do not need to be an inspector. The tests are the same ones an official would do, so you can find out for yourself - and save money and trouble:
Before you call a verificator or a service technician: a visit costs the call-out, and a scale that fails costs a repair or adjustment and a second appointment. If you test it first, you know whether it will pass, you can fix the obvious (levelling, a dirty or loaded plate, a drifting zero, the calibration) beforehand and call only when it is ready - or when the test shows a real defect.
Before you buy a used scale (or accept a new one): the test shows in half an hour whether it is accurate, repeatable and the same in every corner, and which accuracy class it really reaches - on the spot, before the money is gone, and as proof if you want to return it or claim under warranty.
Instead of guessing: is the scale still right after a move, a repair, a heavy overload or a year of use? Does it drift when it is cold? Is the table the problem? The results and the checklist answer that, and a report from last year shows whether it is getting worse.
For your own quality system or customers: shops, labs, workshops and packers can show with a dated record that their scale was checked - and decide on facts whether a recalibration or a new scale is really needed.
It stays a self-check: for legal use (trade, packs, medicine) only the display of a verified scale counts, and only the responsible body can verify or calibrate. But a scale that passes your own test will very likely pass theirs.
The third licence level (from 13.00 EUR a month, 144.00 EUR a year or 599.00 EUR once per UP↑SCALE device, incl. VAT, includes everything of PRO; 14 days free to try, see month, year or once) is made for businesses, service companies, laboratories and inspectors who have to prove that their scales are tested, unchanged and documented. The measuring itself is never held back: weighing, taring, calibration and all tests work in every version. You find the tools in the settings of the scale, tab QUALITY, in the card QUALITY RECORDS & USERS (straight there: ☰ → 🛠︎ SCALE SETTINGS → 🛡 REGISTER & COMPLIANCE; the tab 📑 VALIDATION of the test page shows the date of the last test and the next test due, with a row that opens the card). The test history, the activity log and its archive are on the 🗂 RECORDS page (in the menu ☰ → 🗂 RECORDS, one entry per tab; the card has a link to it). Without the edition the buttons stay visible but greyed out; a click tells you how to get it.
Equipment register: give the scale an asset name and a test interval (30 days to 2 years). The scale saves them with the date of its last test and calculates the next test due (✅ valid, ⚠ due within 30 days, ⛔ overdue). The register lists all scales this browser has seen, with their due dates; export it as CSV or JSON for your quality system (ISO 9001 measuring equipment list).
⚖ Register card and QR sticker: every scale of the register has a card: name and asset, maker, model and serial number (of the scale profile on an RS-232 adapter), the last test with its result, the next due date, the status (in use or ⛔ OUT OF ORDER), the problems and releases, and your 🔗 links to SOPs and manuals (+ ADD LINK). Open it with ⚖ REGISTER CARD · STICKER in the card QUALITY RECORDS & USERS, or tap a scale in the list of all scales. 🏷 STICKER (behind ⋯) prints the name or asset with a QR code of the scale (its Device ID, and the profile on an RS-232 adapter): put it on the scale, and the 📷 scanner of the app opens the card. The card opens without the scale too - with the data this browser saved the last time; a yellow line then says from when the data is. The links and the problems are kept in this browser and go into its 📦 data file and the CSV / JSON export of the register.
⚠ Report a problem and OUT OF ORDER:REPORT A PROBLEM (on the register card and in the card QUALITY RECORDS & USERS) takes the category (wrong values, damage, does not start, other), a short description and who reports it. With the scale connected the report is written into its activity log at once (with the user who is logged in); without the scale it waits in this browser and goes into the log at the next connect. The log keeps the category and a fingerprint of the description; the description itself stays in the app, goes to the other apps connected to the same scale and into the data file. Tick Mark the scale OUT OF ORDER and the scale itself keeps this state (for each scale profile of an RS-232 adapter): every phone and computer that connects shows a red line on every page, and until an ADMIN or a QM releases the scale it starts no program (recipes, factory, check-weighing, sorting, density, solution, tracking, coffee, fill weighing, pipettes) and books no sale (the booking stays in the app and goes to the scale after the release); the app does not log ∑ LOG IT values of the weighing and counting pages either. ✅ RELEASE FOR USE (⋯ of the register card, with the scale connected) asks for the reason and writes the release into the activity log (without users, anybody can release). The tests of the TEST MODE stay open, so you can check the scale after a repair. ⛔ OUT OF ORDER TAG prints a "do not use" label. For a scale that is not connected, the card shows the state the scale sent the last time. ✉ SEND TO SERVICE opens your e-mail program with the data of the scale and the newest problem, addressed to your 📇 service contact (name, e-mail, phone; optional, in the card QUALITY RECORDS & USERS and in the lab settings of the pipette register) - the app sends nothing itself.
Test history: every finished test run of the scale test suite adds a short record to the scale itself (date, test, result ✓ ≈ ~ ✗, worst error in steps d, initials of the inspector): the last 40 runs. It is shared by all phones and computers that connect, and a small chart shows the trend per test - so you see if a scale gets worse from test to test. The full measurements stay in the reports.
Users: optional, up to 50. The first user is the ADMIN; every user has a unique name (case sensitive), a title or role (boss, employee, bar staff, ...) and an optional password (the ADMIN too). With users, every device logs in: after connecting, every phone or computer asks who is working and logs in by itself - several users can be logged in at the same time on different devices, all sharing the one screen of the scale, and each action is recorded under the user of the device that did it. A user without a password is chosen by name only (the name just goes into the logs); a user with a password needs it on every device. There are no guests: a device that has not logged in gets nothing from a scale that has users (its screen shows the login), and the logged-in user is shown with a sign and the initials in the head of the app and by name in the ☰ menu (☻ user, ♛ ADMIN, ☑ QM, ⚒ operator). The ADMIN switches the protections (all off at first): deleting sales and stock only by an ADMIN, settings, calibration, licence key and register only by an ADMIN, and strict audit (every change needs a reason, see below). A protection can only be switched on if an ADMIN has a password (the fourth, only approved recipes, also needs a QM with a password), otherwise anybody could just pick the ADMIN; a user without a password can be chosen by anybody, so the login only keeps people out if the users have passwords. The ADMIN adds, renames, deletes users, sets other users' passwords (a forgotten ADMIN password can be reset by another ADMIN; the last ADMIN can't be deleted; if the only ADMIN forgets the password, only a complete reset of the scale helps) and reads the activity log per user. The user name appears in the weighing log (on every line only if several users weighed, else once at the total), in the sales views and CSV, in the statistics values and their CSV (the scale keeps the user of every value), and in the header of printed reports. USB serial and ESP-NOW have no login (no password is sent over the air): they act with the rights of the user who logged in last on a connected app, else as GUEST.
The login: after connecting, the app opens the window Who is weighing?: choose your name (a 🔑 behind a name: this user has a password), type the password if you have one and press LOG IN (or Enter). The window has no ✕ - there are no guests. A wrong password, a locked user or a missing password under strict audit is told in a message, and the window stays for the next try. If the scale is closed for devices without a login, the app shows the page 🔒 SCALE LOCK instead: the same login in a card in the middle and LOG IN at the bottom right (no TARE, no weight on this page). If your password was set by an ADMIN or is older than the password rules allow, a window PASSWORD: your name asks for a new password twice; SAVE AND LOG IN changes it and logs you in with it (this change can't be skipped). LOG OUT is in the ☰ menu next to your name and in the users card of the settings (there also PASSWORD for your own password). There is no automatic logout after a time: a login lasts until LOG OUT or until the connection ends.
Passwords never go over the air in plain text: every scale has its own key (the one of its seal). The app encrypts each password for the login, a new password and a signature with a one-time key made for this scale (ECDH + AES), and the scale takes it only once, within 30 seconds - a recording of the radio is no use to anybody. The app remembers the key of each scale; if it ever changes (a scale with an erased memory, or another device that poses as your scale), the app asks before it sends a password.
Roles: every user has one role. User: weighs, runs programs and tests, saves recipes and THINGS (the normal case). Operator: weighs, runs programs and tests and can sign a report as performed, but changes nothing: the scale itself refuses settings, calibration, saving or deleting THINGS and recipes, deleting and the other protected actions (the refusal is written to the activity log; the app greys out the settings button and asks for no reason). A recipe an operator runs is not saved. QM / approver: a user who additionally approves recipes and signs reports as reviewed or approved. ADMIN (also possible together with QM): manages users and protections. The role is chosen when a user is created or edited. With strict audit every user needs a password to log in (the user list marks users without one with ⚠). Roles are technical controls of the scale: it does not know who a person really is, so the passwords and the organisation around them matter.
Approval of recipes (four eyes): with users, the list of recipes on the scale shows for each recipe whether it is approved (by whom, when) or not approved (saved by whom, when). A QM presses APPROVE, confirms the statement and signs with the password; the scale checks the password again, keeps who approved which version (a fingerprint of the recipe) and writes the signature into its activity log. Every later save of the recipe makes it "not approved" again, and nobody can approve a version that he saved himself. WITHDRAW takes an approval back. The fourth protection only approved recipes can run (needs a QM with a password) lets the scale refuse to start any recipe that is not approved; the refusal is logged. It covers the solutions too (they are saved as recipes): a saved solution that is not approved does not open, and a solution typed into the app is not weighed - 💾 save it, a QM approves it in the recipe list, then 📂 load it. Recipes that were saved before approvals existed show as not approved until a QM approves them.
⚗ Pipette calibration: the gravimetric calibration of pipettes along ISO 8655-6:2022 on a lab balance (RS-232 adapter): as found / as left, every channel, evaporation, the uncertainty of every volume along ISO/TR 20461:2023, the CMC table of your scope, the sealed calibration report with signatures; the pipette register with import of your old list (the 💧 pipette check is PRO).
⚕ Compounding assistant: the special mode for labs and pharmacies - master formulas with process plan and approval, raw-material lots with use-by dates, guided weighing with the 🔏 QM release of a value over its goal (a second person with password and reason, logged by the scale), in-process controls, filling with the uniformity of mass, the sealed batch record with the signatures of the operator and a QM, and the product label.
Electronic signature of a saved report: in the card QUALITY RECORDS & USERS (settings), the row Sign a saved report file (at the end of the card, under REPORTS) takes a saved HTML report of this edition and adds a signature with the meaning performed (every logged-in user with a password), reviewed or approved (a QM who neither took the report number nor performed the report). The user enters the password, the scale checks it and writes the signature (name, date and time, meaning, report number and the integrity code of the report) into its activity log, so it is part of the SHA-256 chain and of the archive and record files. The signed file is saved again (..._signed-1.html); the integrity code does not include the signatures, so a report can collect several. Signed files are checked on the public page up-scales.com/check. Printed reports are signed by hand as before.
Activity log: the scale records the last 10,000 events with time and user: every sale saved with the sum button, changes of the stock (saved, deleted), start and end of the programs (recipe, factory, check-weighing, sorting, density, tracking, coffee), finished tests with their result, deleting sales and stock, calibration (with the old and the new calibration factor), saved settings, licence key, creep / tare guard / filter changes, recipes and THINGS saved or deleted, changes of the density settings (old and new liquid density), printed reports (with their number), changes of the users, log-ins, log-outs and wrong passwords, changes the scale refused (with the reason why), clock jumps and the devices (qualified, retired, a device without qualification connected). Not every single weighing. Every entry has a running number, and the entries are linked by a SHA-256 chain: the app checks with it that no entry was changed or taken out (see Archive below). The ADMIN sees the log per user on the 🗂 RECORDS page (tab 🛡 ACTIVITY). No command can delete the log. Power cuts: every entry is written in steps that a power cut cannot mix up; if the power goes while an entry is being written, the scale checks the chain at the next start and finishes the entry or leaves it out, so the chain stays whole - and it records that it happened ("Log write interrupted (power loss)"). The entries already saved, the users and the settings are not affected, and a lost Bluetooth connection changes nothing: the scale writes its log by itself.
Reason for a change and strict audit: the ADMIN can switch on a third protection, strict audit (needs an ADMIN with a password). Then every change of settings, calibration, licence key, register, creep / tare guard / filter and every deleting of sales or stock needs a reason: the app asks for it once (routine, correction of an error, new method or product, service or repair, other) with an optional note, and the scale itself refuses the change if no reason came with it, if it does not know the time yet, or if the log would overwrite entries that no app has archived. In that last case it also refuses new programs and new sales (the app keeps the sales and sends them again after archiving) - weighings and programs that are running always go on. Recipes, THINGS and density settings are asked for too and logged with the reason. The virtual weight is switched off (it is no weighing result). Without strict audit, equal setting changes within a minute are one entry; with it, every change is its own entry.
Archive and record files of the activity log: the scale keeps the last 10,000 entries. An ENTERPRISE app archives them by itself whenever this is needed - in the background, nothing is interrupted. It checks every minute (one short question to the scale) and archives right away when strict audit or LIMS forwarding of the activity log is on, or when more than half of the log has not been archived by any app yet. Otherwise it archives as soon as you open the 🗂 RECORDS page, save a record file or press ARCHIVE NOW, and then for the rest of the connection. A round loads only the entries this browser does not have yet, checks the chain (no entry changed or taken out, no number missing), keeps every entry in the archive of its browser with the note of its reason, and tells the scale how far it is archived. The 🛡 ACTIVITY tab of the 🗂 RECORDS page shows the number of entries, the chain result (✓ unchanged, or ✗ broken between two entry numbers), the entries on the scale that are not archived yet, alarms (a missing range, a broken chain, an archived entry that is different on the scale now), entries the scale had to overwrite before any app archived them, whether the browser keeps its data for good, and the date of the last saved record. A line at the bottom of every page warns when 80 % (yellow, ✕ hides it until the app is opened again) / 95 % (red) of the log on the scale are not archived yet. With strict audit the log is no rolling buffer (FIFO): before an entry that no app has archived would be overwritten, the scale stops new sales, new programs and changes (work that is running goes on) until an app has archived the log - connect the app and it goes on, no entry is lost. Without strict audit the oldest unarchived entries are overwritten, but never silently: the scale records the lost numbers and the app shows them. A round only loads the entries that are new since the last one (about 65 entries per second): a long round (from 200 new entries on, e.g. the first one or after many days without an app) shows a progress bar with the percentage and the time left - keep the app connected until it is done. SAVE FOR THE RECORDS (after archiving the newest entries) saves a record file with the entries since the last record of this scale, named so that the files of a scale stand in order in a folder: UPSCALE-Audit_420-42149_No-0000040-0000041_2026-10-23.json (scale, entry numbers, date). Every record file names the previous record file and its SHA-256 and carries the chain value at the end of it, so the files form one chain: a missing, changed or exchanged file shows up. When nothing is new since the last record file, the app first asks whether to SAVE COMPLETE - the whole archive again as a separate complete record. Right after saving, a line at the bottom names the file, and the app offers the printout of the same entries in its own window: with ENTERPRISE 📄 SEALED PDF (the record file attached to it) or 🖨 PRINT (the print of the browser) - ✕ for neither; in records mode 🖨 PRINT or CANCEL. The printout holds: scale, range, chain result, missing numbers, alarms, the previous file with its SHA-256, the name and the SHA-256 of this record file, all entries with time, user, event, reason and note, and lines for signatures - so the paper and the file belong together provably. Keep the record files unchanged in your folder (the record is the file, not a copy opened and saved by another program). Check exported archive files (several at once) shows the SHA-256 of every file, checks the chain of the entries inside them, the continuity between the files (no number missing) and the link of every file to its previous file. EXPORT ARCHIVE (CSV / JSON in the chosen format) is for reading in a spreadsheet only - a spreadsheet can change values when it saves, so it is not the checkable record. The chain is public - it shows changes and gaps, it is no secret seal: someone with the tools to rewrite the flash memory of the scale could write a new consistent log, but not the record files that are already saved.
One archive device for the log: once a qualified device has archived the log, it is the archive device of this scale (🗂 RECORDS shows it): only its archive frees entries on the scale, so one device always has the whole log - other apps keep their copies, but a copy on a phone frees nothing. It archives in every round and saves a sealed record file once a day by itself into the records share of the device (no dialog, no printout; SAVE FOR THE RECORDS still works at any time). A new PC, or the old one is broken: MAKE THIS THE ARCHIVE DEVICE on another qualified device (an ADMIN with a reason, in the activity log). Without any qualified device every app archives. The regular rounds in the background wait while a factory, check-weighing, sorting or fill program runs. Back up the records share: it holds the complete record.
Report extras for all reports: printed or saved reports (test suite, final report, sales report, ℮ lot report, tracking report) get a header with your company data and logo, a running report number counted by the scale (the same in every app, never repeated), an integrity code in the footer and the UP↑SCALE SEAL. The footer also names the firmware, the calibration and configuration counters and the certificate data and the address of the public check: up-scales.com/check - anybody can check the file there in the browser, nothing is uploaded. If a saved file is changed afterwards, the check there shows it. (With PRO, your logo already goes on top of printed and saved reports - without number, integrity code or seal, so these files cannot be checked.)
Exports in your format: all table exports (test suite, sales, statistics, stock list, tracking, equipment register) belong to this edition. Next to every export button you choose the format: UP↑SCALE standard (comma, dot, ISO dates, fixed English column names, no BOM), Excel in the language of the app (semicolon, decimal sign of the language, BOM), Excel English, JSON (numbers and dates exactly as they are, for programs and ERP), or own settings (⚙: separator, decimal sign, date format dd.mm.yyyy / dd/mm/yyyy / mm/dd/yyyy / ISO, BOM, column names fixed or in the app language). The fixed columns: sales date, time, sale_id, thing, user, pieces, weight_g, revenue, costs, profit (per day: without time and sale_id); stock list name, date, empty_g, full_g, weighed_g, content_g, left_pct, volume, volume_unit, full_units, total_g, value, currency, expected_g, loss_g, used_since_g; statistics no, time, value_g (or value_pcs), net_g, alibi; tracking no, time, unix_time, weight, display_net, interval_s, note; register device, scale, asset, last_test, interval_days, due, state, last_seen, edition. The LEDGER CSV of the sales (date, description, amount; one line per day and THING; always comma, dot and dd/mm/yyyy) has the three columns that accounting programs document for importing transactions (QuickBooks Online, Xero via column mapping). We do not promise ready-made templates of third-party programs (Square, Shopify, DATEV, ...): their templates are strict and change - use JSON or the standard CSV with the column mapping of your program.
🔗 LIMS / ELN forwarding: in addition to the live weight and tracking of PRO, the app sends the activity log, the sales log, the stock, the sealed reports, the status of the scale for use and the weighing log as JSON to a server of your lab, while it is open and connected (LIMS / ELN forwarding).
Validation of the calculations (IQ/OQ): reference cases worked out by hand from OIML R 76-1, computed by the code that judges your tests (the UP↑SCALE firmware and the app), with a printable record and code fingerprints for your validation file - see Validation of the calculations.
Protection of deleting and settings: instead of separate passwords, the ADMIN switches two protections for all users (see Users): deleting sales and stock only by an ADMIN, and settings, calibration, licence key and register only by an ADMIN. The old passwords for deleting the sales log and for locking the settings are gone and can't be set any more (an existing delete password still works and can be removed in the sales tab).
Good to know: the date and time come from the app (the scale has no clock - the device qualification closes this for audits): every app sends its time when it connects and every hour while it is connected, and the scale logs every jump of more than 2 minutes (clock guard). With one click (the row 🕒 Exact time, key FROM UP-SCALES.COM, in the card 🔑 LICENCE of ✰ SCALE INFO) the scale takes the signed exact time of public time servers instead and checks the devices against it. Users, protections and the activity log are kept on the scale and are the same for every app and every browser. Every phone, tablet and PC that runs the scale is qualified once with the 🖥 DEVICE QUALIFICATION form and kept in the device register.
👥 Who may use the scale: users, roles and the user access register (ENTERPRISE & COMPLIANCE)
Auditors ask for a controlled list of the people who may use a system. They want to know the user ID of each person, their role, who authorised them, since when they have access, and when the list was last reviewed. 21 CFR Part 11, EU GMP Annex 11 (§12), ISO/IEC 17025 and GLP all ask for this kind of access control. The scale and the app keep it for you:
One user per person, never reused: every user gets a permanent user number from the scale (shown as U000001, U000002, …), and the scale never gives the same number twice. This also holds after years and after any number of users came and went. Use one user per person and no shared logins.
The history is kept by the scale: creating a user, every change of the role, a new password and removing a user are written to the hash-chained activity log, with the user number, the old and the new role and the ADMIN who did it. The app archives the log, so the access history cannot be rewritten without it showing. Old log entries keep the person who was the user at that time.
The people behind the users: the 👥 USER ACCESS REGISTER lists every current user of the scale with number, user name, title, role and whether a password is set. Your QA adds the real name, the function, who authorised the access, the training date and remarks. These entries are kept on the scale, the same for every app. Only an ADMIN or a QM can change them, and every change is written to the activity log. They stay after a user is removed, so the former users keep their real names. Users who were removed stay in the list as former users.
Access review: a QM signs it with the password (✍ SIGN THE ACCESS REVIEW), and the scale checks the password every time. The scale keeps the review with the date, the reviewer and the next due date (3 to 24 months), and writes it to the activity log.
Output for QA and inspectors:
SAVE SEALED PDF makes a numbered PDF, sealed by the scale; the reviewer then signs it electronically (✍ SIGN). It contains the current users, the former users, the access history from the archived log, the access review, signature lines for the reviewer and the system owner, and a column for the signature specimen of every user. The register is attached as JSON.
PRINT prints the same record, and CSV / JSON exports it for your QA system.
BLANK FORM (free) is the paper version.
The register is a window of the app (✕ at the top right, ℹ️ behind its title): a card per part (current users with their five fields, former users, access history, access review, export format). 📄 SAVE SEALED PDF is the main key at the right; after it ✍ SIGN REPORT n comes beside it; PRINT stays beside it, CSV / JSON and BLANK FORM are behind ⋯. Without ENTERPRISE the locked keys carry 🔒 and wait behind ⋯, and BLANK FORM is the main key.
Where to find it: the settings, card QUALITY RECORDS & USERS (the row 👥 USER ACCESS REGISTER below the user list); the 🗂 RECORDS page, tab 👥 ACCESS; and the licence of an ENTERPRISE scale.
Your part: check who a person is before you give them a user and an electronic signature. Keep the signed register with your SOP. If your authority asks for it, send the statement on electronic signatures yourself; for the FDA that is the letter under 21 CFR 11.100(c).
🔑 Password rules: lockout, maximum age, a password set by an ADMIN (ENTERPRISE & COMPLIANCE)
The scale checks every password itself: at the login, when you change your own password, with every electronic signature and with every review.
Lockout: after 5 wrong passwords in a row the user is locked for 15 minutes. A restart of the scale does not end the lock. Every wrong try and the lock are written to the activity log. An ADMIN can unlock a user at once (🔓 UNLOCK in the user list). The user list shows wrong tries and locked users.
Maximum age: the ADMIN can set how long a password may be used: never, 90, 180 or 365 days (Password rules below the user list). An older password has to be changed at the next login, before the user can work. The user list shows how old each password is.
A password set by an ADMIN for somebody else is changed by that person at the first login, when the password rules or strict audit are on. Only that person knows their password.
A new password must be a different one from the current one.
Changes of the rules are logged with the old and the new value. Minimum length, complexity and the training of your users are rules of your own SOP.
🧾 Review of the audit trail (ENTERPRISE & COMPLIANCE)
Audit trails must be reviewed regularly (EU GMP Annex 11 §9). On the 🗂 RECORDS page, tab 🛡 ACTIVITY, the box Audit trail review shows the last review, by whom, up to which entry, and how many entries have come since then.
A QM reads the entries since the last review (changes with old and new values, refusals, logins, wrong passwords, clock jumps, device events) and presses ✍ SIGN THE AUDIT TRAIL REVIEW.
The QM signs with the password, and chooses when the next review is due (1, 3, 6 or 12 months).
The scale checks the password, keeps the review, and writes it into the hash-chained activity log with the number of the last reviewed entry. So the reviews themselves are part of the record files.
A line at the bottom confirms the signed review. An overdue review is shown in red.
🔗 LIMS / ELN forwarding: your weights and records to your own server (PRO / ENTERPRISE & COMPLIANCE)
Which edition: with PRO the app forwards the live weight and tracking (start, end, alarms and points). ENTERPRISE & COMPLIANCE adds the finished programs and batches (with their sealed report files if you want), the activity log, the sales log, the stock, the sealed reports, the status of the scale for use, the validation runs (IQ/OQ) and the weighing log (marked 🏢 on a PRO scale).
The app can send what happens at the scale straight to a server of your lab - your LIMS, your ELN, a small archive server in your network - as JSON, one event per POST. Nothing goes to us or to any other cloud: you give the address, and only your server gets the data. Settings > 🔗 LIMS / ELN FORWARDING.
Important: the app must be open and connected to the scale on that device. The scale itself sends nothing to the network - it is the app that forwards. If the phone or computer is off, the app is closed or a phone has put it to sleep in the background, nothing is sent at that time. What is lost and what comes later:
Activity log: nothing is lost. The scale keeps its entries, and the next time the app is connected it fetches the new lines (it does this every minute anyway) and forwards each line once.
Sales: the scale hands every new sale to all connected apps, so the forwarding app also sends the sales of the other phones and tablets at the scale - while it is connected itself. Sales of a time without it are not sent (each one is still in the activity log, which catches up).
Weighings, live weight and tracking events: only what happens while the forwarding app is open, and the weighings only of that app.
Only the app where you switch it on sends. The settings (address, token, events) are saved in that browser, not on the scale. Switch it on in one app only - for example on the lab computer next to the scale. You choose with a checkbox for each kind of event what is forwarded:
📜 Activity log of the scale: every line with its sequence number, time, user, text and the chain hash - your server can check the chain by itself (hash chain).
💶 Sales log of the scale: every sale the scale saves (sales mode), from any app, with its record number, the chain value of the sealed record, the THING, weight, price, user, member, lot and the raw bytes. A booking of the DISPENSARY 🏪 MODE also carries the full Metrc tag of its lot (metrc_tag, when the app knows the lot), the plant or package tag of a harvest or package (metrc_plant_or_package), the Metrc quantity and unit of a stock check or a received package (metrc_qty, metrc_unit), whether a waste was of a package or a harvest (waste_of) and, for a sent marker, the file type, the rows and the last booking it covers (metrc_file_type, metrc_rows, metrc_last_seq) - so a partner system can take the weights with their tags.
📦 Stock on the scale (inventory): every change the scale hands to the apps - grams used (∑, sales, recipes, solutions) or added (a dried harvest), the number of full ones, every stock check with the expected content and the loss, a new or deleted item - from every phone at the scale, also what was booked while the forwarding app was closed (it compares with the state it knew).
🔏 Sealed reports and files: every seal the scale makes (UP↑SCALE SEAL) - reports, PDFs, exports and record files, and every electronic signature - with what was sealed (title, file name, report number, for a signature its meaning and signer), the SHA-256 of the file, the time, the user, the number of its entry in the activity log, the seal itself, the certificate and the public key of the scale. Your server can check every seal on its own, the same way as the public check page. Without more, the file itself is not sent: your server gets its fingerprint, and the file that is saved or printed can be matched against it at any time. Tick with the file itself below it and the finished file comes too, as its own event sealed_file right after its seal: the same log_seq as its sealed event, the file name, the type (mime), the size, the SHA-256 of the finished file (sha256_file - for a PDF or report it differs from the sealed SHA-256, because the seal is written into the file) and the file as base64 (content_base64). Until it is sent, the file waits in the storage of this browser (IndexedDB), and only a reference waits in the queue; if the browser has lost it, the event comes with content_missing: true. A file can be large: allow POST bodies of a few MB on your server (PHP: post_max_size).
🚦 Status of the scale for use (event status, field kind): every finished routine check (routine_check: pass or fail, profile, re-test, n, mean, s, minimum weight and every single check with its error and limit in d), the release state whenever it changes and once when the app connects (release: released, due - the minimum interval has passed - or not_released after a failed check, with the state before and the last passed check), a removed warning with its reason (warning_removed), and every weight recorded below the minimum weight after the warning (below_min_weight). So your server can mark every weighing made while the scale was not released.
📝 Weighing log of this app: every ∑ and the lines of the programs (recipes, solutions, factory mode …). A line below the minimum weight has below_min_weight: true and min_weight_g.
⚕ Finished programs (event programs, field kind): every batch of the compounding assistant (batch: the whole batch record as JSON UPSCALE-BATCH-1 - lots, use-by dates, weights, releases, checks, in-process values, filling, deviations), every finished solution (solution), every finished recipe run with its lots (recipe) and every saved pipette check / calibration (pipette: results, setup, verdict) of this app - by itself, at the end. With "with the report as a file" the app also makes the report itself (batch record, solution report, recipe batch report) - numbered, with the integrity code and sealed by the scale, without a download - and sends it as base64 in its own event (…:file, with its SHA-256 and report number; checkable on the check page). The batch record made this way is also kept in 🗂 RECORDS → BATCHES, ready to sign. ENTERPRISE.
📡 Room log of the sensors (event room, field kind): one mean value a minute (minute: air_c, humidity_pct, air_hpa and the sensors) every crossing of a room limit (limit: code and text) and every room sensor that stopped sending or sends again (sensor: code gone / back and text) - see room. With a sensor connected every event also carries env.
☑ Validation of the calculations (event validation): every run of the validation of the calculations (IQ/OQ) in this app - date, app version, code fingerprint, browser, scale, firmware, Device ID, edition, the number of cases passed and failed, and every case (results: id, clause, pass, expected, calculated, error). Ready for your QMS without reading a PDF. ENTERPRISE.
⚖ Live weight of every mode - what the big weight shows (weight, pieces in the counting mode - then also with the weight in g and the piece weight - ml in the coffee mode), with the mode it was in. All values are plain JSON numbers; when it settles (once per weight or count; with the stability light on, when it is SETTLED) or every 1, 5 or 60 seconds. With the weight in g and in your unit, stable, the stability state and spread in d, d, the THING - and the line of the serial data format (the same as in the manufacturer data and on USB) as plain ASCII text ("serial": ST,+0000019.60 g + CR LF, counting SQ,+0000000012 PC), so a program that already reads the scale over USB or Bluetooth can use the same parser. With butler scales connected, every scale comes on its own ("scales": number, name, g, stable) together with the total of all of them ("total_g"; the serial line then shows the total). With a minimum weight set (ENTERPRISE & COMPLIANCE), a weight below it has below_min_weight: true and min_weight_g (with butler scales: the weight of this scale).
📈 Tracking: start, end and every alarm of a tracking run (target reached, constant weight, jump, restart, peak, above / below the band) - and, as an extra choice, every new point.
Every event looks like this (with a sensor connected it also has "env": the room air, humidity, air pressure and the liquid of now, each with its source):
The id is fixed for each event (the activity log: device + sequence number). If an event comes twice - after a lost answer, or when an app connects again during a tracking run - your server simply drops the second one.
The app never waits for your server. A POST that gets no answer is given up after 8 seconds (a sealed file: 60 seconds); a failed one is sent again later (after 5 s, then longer, up to every 5 minutes), in the right order. Up to 1000 events wait in the browser, also over a reload. A yellow line at the bottom of the screen (a tap explains it, ✕ hides it until the app is opened again) warns when 800 of them are used (soon the oldest are dropped - they stay on the scale and in the archive files) or when your server was not reachable for an hour; it goes away by itself once the queue empties. The live weight never waits: an old weight is worth nothing, so it is only sent when nothing else is on its way. 📡 TEST sends a test event ("ping") and shows the answer; the status line shows how many were sent, how many wait and the last error.
What your server needs (browsers are strict here):
https. The app runs on https; browsers block a POST from it to a plain http:// address. A certificate of your own lab (internal CA) is fine.
CORS headers that allow the app: Access-Control-Allow-Origin (https://up-scales.com; the offline app sends the origin null), Access-Control-Allow-Headers: Content-Type, Authorization, and for an address in your local network also Access-Control-Allow-Private-Network: true. Some browsers ask once whether the app may reach devices in the local network - allow it.
An answer with a 2xx status. Anything else counts as not delivered and is sent again.
A receiver without CORS (some hook services, chat webhooks): switch on Receiver without CORS headers (simple webhook). The app then sends the JSON as text/plain and puts the token into the address (?token=...). The browser shows no answer in this mode, so every event counts as delivered and nothing is sent again - for records that must arrive, use your own server with the CORS headers.
A complete receiver in PHP that writes every event as one line into a file:
<?php
// upscale-receiver.php: every event as one line in events.jsonl
header('Access-Control-Allow-Origin: https://up-scales.com');
header('Access-Control-Allow-Methods: POST, OPTIONS');
header('Access-Control-Allow-Headers: Content-Type, Authorization');
header('Access-Control-Allow-Private-Network: true');
if($_SERVER['REQUEST_METHOD'] == 'OPTIONS') {exit;}
if(!hash_equals('Bearer YOUR-TOKEN', (string)($_SERVER['HTTP_AUTHORIZATION'] ?? ''))) {http_response_code(401); exit;}
$event = json_decode(file_get_contents('php://input'), true);
if(!is_array($event) || empty($event['id'])) {http_response_code(400); exit;}
file_put_contents(__DIR__ . '/events.jsonl', json_encode($event) . "\n", FILE_APPEND | LOCK_EX);
echo 'OK';
The CORS headers on other servers (the browser first asks with an OPTIONS request - your server answers it with these headers and no body; then the POST follows):
Apache (.htaccess in the folder of the receiver, with mod_headers and mod_rewrite):
Header always set Access-Control-Allow-Origin "https://up-scales.com"
Header always set Access-Control-Allow-Methods "POST, OPTIONS"
Header always set Access-Control-Allow-Headers "Content-Type, Authorization"
Header always set Access-Control-Allow-Private-Network "true"
RewriteEngine On
RewriteCond %{REQUEST_METHOD} OPTIONS
RewriteRule ^ - [R=204,L]
n8n, Node-RED and similar tools: in the webhook node set the allowed origin to https://up-scales.com (n8n: Allowed Origins (CORS)), check the token in the first step, answer with 200. The offline app (the single file) sends the origin null: allow it only if you use that file.
Check it: 📡 TEST in the settings of the forwarding. "Not sent" with a working address almost always means a missing CORS header - the browser console names the one it missed.
A second copy outside the scale and the browser: forward the activity log to a storage of yours where entries can only be added (for example an append-only file, or a bucket with a retention lock). Together with the chain hashes this keeps an independent copy of every line, and a later change on the scale side or in a browser archive shows. How long and how strictly that copy is kept is set up in your storage.
The UP↑SCALE SEAL (ENTERPRISE & COMPLIANCE)
Every scale makes its own seal key (ECDSA P-256) a few seconds after its first start with this firmware. The private part never leaves the scale. When an ENTERPRISE app prints or saves a report, a record file of the activity log, an export or a PDF, it sends the SHA-256 of the file to the scale, and the scale seals it: it signs the SHA-256 together with its Device ID, the time, the report number, the logged-in user and the number of the entry that it writes into its activity log for this seal ("Sealed by the scale"). Changing one character of the file breaks the seal, and no app or computer can make a new seal without the key of the scale - also not with a changed app or a pretended scale.
Where the seal is: a report (HTML) shows it in its footer ("Sealed by the scale ..., log entry #..."); a JSON record file or export has it as its last field seal; a CSV export gets a second small file <name>.seal.json - keep both; a PDF has it in its document data and shows the report number and the integrity code on every page.
Signatures: a performed / reviewed / approved signature on a saved report file is sealed too, with the name of the user and the meaning of the signature: the meaning or the name can not be changed later.
PDF reports:SAVE AS PDF (test suite, validation record, lot report, tracking report, and the printout of a record file) makes an A4 PDF with the report as text and tables (charts are left out); the HTML report itself - with its integrity code, seal and signatures - and, for the activity log, the JSON record file are attached to the PDF (most PDF viewers show them in their attachments panel). The PDF is sealed as a whole.
Seal certificate: UP↑SCALE confirms with its licence key that the seal key belongs to the scale with this Device ID and the ENTERPRISE edition. It comes with the link of the key email of an ENTERPRISE order (the app sends it to the scale), or the app asks for it once when the scale has none (right after the ENTERPRISE key is installed, and again when an erase made a new seal key) - only after you confirm: with the link of your order it comes at once; without it, a link goes by email to the address of the order: open it on the device with the connected scale and press its button - only then the certificate is made, and the app opens with it. The settings page has the block UP↑SCALE SEAL (below SCALE DATA): a short description, the state and the public seal key, a red warning while the scale has no certificate, and GET SEAL CERTIFICATE to ask for it. A scale that has a certificate shows CHECK SEAL CERTIFICATE instead: the app asks up-scales.com whether this certificate is in our list; if it is not found, you can ask for a new one (same way as above), and it replaces the old one on the scale. A new key (after the flash memory of the scale was erased) needs a new certificate; old certificates stay valid for the old files. Without a certificate a seal shows only "the same key as before" - the check says "key not certified".
Checking: sealed files are checked on the public page up-scales.com/check, by anybody (an authority, a customer, an auditor): PDF, HTML reports, JSON record files, a CSV with its .seal.json. It shows whether a file is unchanged, which scale sealed it, when and who was logged in, the certificate and the signatures. It runs in the browser, nothing is uploaded. In the app, the audit view has the button 🔍 CHECK A REPORT FILE for a PDF or an HTML report (its seal, the attached files and the signatures); JSON record files are checked in 🗂 RECORDS → 🛡 ACTIVITY.
Sealed, not signed: why the scale's mark is a seal, how it is made and how an auditor checks it: sealed or signed?
What a seal is not: no qualified electronic signature (eIDAS), no official certificate, and no proof that the measurement was right or that the firmware of the scale is unchanged. Seals made outside the scale have no fitting entry in the archived activity log, so keep the archive and save the record files regularly.
🔏 Sealed or signed?
Two words that are often mixed up. In UP↑SCALE they mean two different things, and a record can carry both:
Sealed = by the scale, by itself, for every report, PDF, record file and export: "this file comes unchanged from the scale with this Device ID, made at this time, with this user logged in, logged as entry #...". In EU law (eIDAS) this idea is an electronic seal: the mark of an organisation or a system that shows where a document comes from and that it is unchanged.
Signed = by a person, on purpose, with their own password: "I, name, performed / reviewed / approved this report" (see signatures). That is an electronic signature in the sense of 21 CFR Part 11 §11.50 / §11.70: the printed name, the date and the meaning, linked to the record. The scale checks the password, writes the signature into its activity log and then seals it, so the name and the meaning can't be changed later.
Why it is called "sealed": a signature needs a person who means something by it. The scale means nothing - it only vouches that the file is the one it saw. If the scale's mark were called a signature, an auditor would read that a person approved the report when nobody did. So the app calls a file sealed as soon as the scale has sealed it (SAVE SEALED PDF), and signed only after a person has signed it (✍ SIGN REPORT …). For the same reason nobody - not we and not you - may call it a qualified electronic seal: that needs a certificate from a qualified trust service on the EU trusted list and a certified signing device. Our seal uses the same idea and the same cryptography, but it isn't one.
Why nobody else can make one: the private key is made by the scale and never leaves it; our licence key confirms which scale the public key belongs to (the seal certificate). Without the key of the scale no app, computer or person can make a seal that fits a changed file - we can't either, we never have that key.
Industry standard, open format: SHA-256 (NIST FIPS 180-4) and ECDSA on the curve P-256 (NIST FIPS 186-5), the algorithms that also protect HTTPS, PDF signatures, software updates and electronic passports. Only the place of the seal in the file is ours, and it is open:
a seal = 80 bytes of data + 64 bytes of ECDSA signature (r | s) over the SHA-256 of these 80 bytes. The data: version, kind (1 report, 2 file, 3 signature), Device ID (6), the number of the log entry of the seal, the time (epoch, UTC), the report number, the user and its name, the meaning of a signature, and the SHA-256 of the document.
the seal certificate = 80 bytes (version, edition, Device ID, certificate number, date, the public key of the scale) + 64 bytes of signature of our licence key; our public key is built into the check page.
the document hash: an HTML report without its code, seal and signature fields; a JSON file without its last field seal; a CSV file as it is (the seal is in the .seal.json next to it); a PDF with its seal field /UPSCALESeal filled with zeros.
The script of the check page is plain, readable JavaScript, so an auditor's IT can re-check a seal with any standard crypto library, without us.
How an auditor checks a file in under a minute: open up-scales.com/check on any computer or phone, choose the files (several at once), and read for each one: unchanged or not, which scale, when, which user, which log entry, whether the key is certified, and the signatures of the people. Nothing is uploaded, no account or app is needed. For the complete check the auditor asks for the record files of the activity log of the same period: the log entry named in the seal must be in them.
What it brings: for you, every report is sealed by itself, without extra signing software, bought certificates or a subscription, and it shows that the record comes from the scale and not from a typed or edited file. Auditors check your records themselves, at once and independently of you and of us, and ask for the right log entries instead of searching.
Adobe Reader and other PDF viewers do not list our seal in their signature panel: that panel shows only PAdES signatures with the certificate of a trust service. If you need one, sign a copy of the PDF with your own signature tool and keep the sealed original: any change to a PDF (an added signature too) makes the seal of the whole PDF no longer fit, but the attached report still checks.
For regulated laboratories (ISO/IEC 17025, GLP/GMP, 21 CFR Part 11)
Laboratories that work under a quality system have to show that their balances are tested, that their records are complete and unchanged, and who did what. No software is "compliant" by itself: compliance comes from the validated system together with the procedures (SOPs) of the laboratory. UP↑SCALE gives you technical controls and records for that; this list says exactly what they do today, so that you can judge them in your own validation.
Records on the scale, not in one browser: users, protections, the test history and the activity log are kept in the scale itself and are the same for every connected phone or computer (see ENTERPRISE & COMPLIANCE).
Who may use the scale: every user has a permanent user number that is never given twice, and every change of the users and roles is in the chained activity log. The user access register adds the real names, authorisations, training and the access review, as a sealed PDF for your QA and the inspectors.
Passwords and reviews: lockout after 5 wrong passwords, an optional maximum age, a password set by an ADMIN is changed at the first login (password rules), and QM-signed reviews of the user access and of the audit trail (audit trail review).
Who, when, what and why: with users, every connection has to log in, and each logged action carries the user, the time and - with strict audit - the reason; calibration, density and every setting that changes the weight or the records keep the old and the new value. Every entry has a running number and is linked by a SHA-256 chain, and the app keeps an archive of the log in its browser (see ENTERPRISE & COMPLIANCE).
Report integrity and the UP↑SCALE SEAL: every printed or saved report carries a running number counted by the scale and an integrity code (SHA-256), and the scale seals it: it signs the SHA-256 of the report with its own key (ECDSA P-256) and writes the seal into its activity log. Record files, exports and PDF reports are sealed the same way. Anybody can check a sealed file on the public page up-scales.com/check (in the browser, nothing is uploaded) - see the UP↑SCALE SEAL.
Pipettes:intermediate checks and calibrations of your pipettes on your lab balance - every value taken by the UP↑SCALE, the uncertainty along ISO/TR 20461:2023, the reported U never below your CMC, the pipette register with its history and the import record of your old list.
Test procedures: the test suite follows OIML R 76-1 Annex A; the limits are calculated by the app and every rule is shown as met, not met or unknown.
Validation of the calculations: a built-in set of reference cases worked out by hand from OIML R 76-1, run against the code of the app, with a printable IQ/OQ record and a code fingerprint (see Validation of the calculations).
Complete on the product side. These are the technical controls. Everything an audit can expect from a balance and its software is built in:
an audit trail in the scale with who, when, what, why and the old and the new value, hash-chained, for every connected device (Part 11 §11.10(e), Annex 11 §9)
access control: permanent user numbers and roles, lockout, password age, and QM-signed reviews of the access and of the audit trail (§11.10(d), (g), §11.300, Annex 11 §12)
electronic signatures with the name, the date and the meaning, checked by the scale and linked to the record (§11.50, §11.70)
sealed records that anybody can check (§11.10(b), (c); sealed or signed?)
the test suite, the routine check and the validation of the calculations
What stays with you is what no product can do for you: your validation, your SOPs, training, checking who your users are, and the storage and time servers of your IT. The app brings the forms and templates for these too. How it saves time:
For your QA: the binder documents come out of the app, filled in by the app, sealed and with signature lines: device records, the user access register, the device register, test records, routine check logs, review records, the IQ/OQ record and the validation pack with plan, references, requirements, risk assessment, test scripts and the Part 11 / Annex 11 checklist. You execute and sign the tests - no consultant for the paperwork and no subscription.
For the auditor: checking a file takes a minute on the public check page. The reviews, the signatures and the log entry of every seal are in the records, so the auditor asks for certain entries instead of leafing through logbooks.
All of this is built to the expectations of 21 CFR Part 11, EU GMP Annex 11, ISO/IEC 17025, GLP and ALCOA+. It is not a certification: compliance comes from your validated system together with your procedures.
Limits today (please take them into account in your validation): the scale keeps the last 10,000 log entries and the app archives them while it is connected (without strict audit or LIMS forwarding only when you open 🗂 RECORDS, or when half of the log is not archived yet) - save the record files regularly; with strict audit the log is no rolling buffer: the scale refuses changes, new programs and new sales before it would overwrite entries that no app has archived (without strict audit it overwrites them, but records which numbers were lost). The log records which setting was changed, by whom, when and why, with the old and the new value for calibration, density, the weight unit, the filters, zero tracking, the stable lock, d and the ranges, the RS-232 setup, the tare guard, the creep compensation, the environment filter, the heartbeat lock, the protections and the password rules; for the other settings (language, display, connections) only that they changed. The chain of the log shows changes and gaps, but it is not signed: whoever could rewrite the whole log and recalculate the chain could not be told from the original on the log alone - so the archived and exported copies (with the seal) are the evidence. The integrity code of a report alone detects changes of the file but anybody can compute one - the UP↑SCALE SEAL is what shows that a file comes unchanged from a certain scale; the drawn signature on the report is an image; a signature on a saved report file is checked against the log of the scale and its archive and sealed by the scale, but it is not a qualified electronic signature (no certificate of a trust service - labs that need one sign the PDF with their own qualified signature tool); seals made with a copied key have no fitting entry in the archived log; the seal does not show that the firmware of the scale is unchanged; recipe approval covers recipes only (not THINGS or settings); the operator role is enforced for the actions listed above, not for every single setting; passwords are optional unless you use strict audit or rely on the protections. Reports are HTML files or sealed PDF files (the PDF shows the report as text and tables - charts are left out - and carries the HTML report as an attachment). The time of every record comes from the connected app (the scale has no clock): lock the clocks of your devices and record it with the device qualification - the scale logs every big clock jump (clock guard).
📝 🗂 All logs and records at a glance
UP↑SCALE keeps several logs, each for its own job. They differ in what they hold (single weighings, measured values, events), where they are kept (in the open app, in your browser, on the scale) and how long. You reach all of them from the menu: ☰ → 📝 LOG & STATISTICS (📝 WEIGHING LOG, 📊 STATISTICS & ℮) and ☰ → 🗂 RECORDS (one entry per tab of the records page).
📝 Weighing log (all versions)
Every weighing or counting you log with ∑, with its name, units, prices and nutrition values, plus the results of the programs (coffee, factory, sorting, tracking, recipes). It is your working list: labels, the 📒 diary, a new THING, baker's percentages and the batch report are made from it. It only lives in the open app: closing or reloading the app clears it.
📊 Statistics & ℮ (statistics on the screen free; ℮, reports and exports with PRO)
This is not a log of its own, it evaluates the others. It takes the single values of a running check-weighing, factory or sorting program (the scale keeps up to 2,000 of them in its memory until it restarts, so every connected app gets the same series) and the lines of the weighing log. From them it works out the mean, the standard deviation, Cp/Cpk and the ℮ lot check (more).
Kept on the scale: the last 10,000 events (who changed or did what and when: settings, calibration, logins, sales, stock, programs, tests, reports, signatures, refusals), linked by a SHA-256 chain. The app archives it in your browser and saves it as record files. Single weighings are not in it, and they don't need to be: they would push the important events out of the 10,000 entries within hours. On the 🗂 RECORDS page, tab 🛡 ACTIVITY (more, how it is protected).
☑ Test history (ENTERPRISE & COMPLIANCE)
Kept on the scale: a short result of each of the last 40 test runs of the test suite (date, test, result, worst error in d, who), with a trend chart. The full measurements are in the reports you save. On the 🗂 RECORDS page, tab ☑ TESTS.
✓ Routine checks (all versions, exports with ENTERPRISE & COMPLIANCE)
Every routine check before use and every removed warning, kept in this browser (the last 1,000 per scale). The scale itself keeps the last passed check, an open failure and the interval. On the 🗂 RECORDS page, tab ✓ ROUTINE.
👥 User access register (ENTERPRISE & COMPLIANCE, the blank form is free)
The users of the scale with their permanent user numbers and roles (on the scale), the people behind them and the access review (your QA, in the browser), and the access history from the archived activity log. On the 🗂 RECORDS page, tab 👥 ACCESS (more).
🖥 Device register (ENTERPRISE & COMPLIANCE, the form is free)
Kept on the scale: every phone, tablet and PC that runs it, with its qualification (qualified, failed, due, retired), its record number and the due date; the sealed, signed records are your files. On the 🗂 RECORDS page, tab 🖥 DEVICES (more).
💶 Sales log (PRO)
Kept on the scale until you delete it: every sale (a booking of the 🛍️ sales mode) with its time, pieces, weight (with the d of the scale), revenue, cost, user, lot number and member ID - numbered and chained by the scale, so a changed, inserted or deleted record shows up (sealed report). The scale keeps 366 days with sales and warns 30 days before the oldest ones get deleted. It opens from the 🗂 RECORDS page, tab 💶 SALES, or from the 📦 stock.
Other records
The tracking recordings and the last 20 stock checks per item are kept on the scale. The 📒 nutrition diary is kept in your browser. Saved reports and record files are the files on your disk.
Why are they separate? They have different rules. The weighing log is meant to be changed, cleared and started again. The activity log must stay complete, may never be edited and has room for events only. Measured values need a long series in the memory of the scale. They are linked where it matters: the activity log notes the start and end of every program, the sales and stock changes with their numbers, every report number and every signature, and the statistics values carry the user who weighed them.
Tamper-evident records: audit trail, hash chain and digital seal in plain words
For audits (ISO/IEC 17025, GLP/GMP, 21 CFR Part 11, EU Annex 11, ALCOA+ data integrity) your records must show that nothing was changed or removed afterwards. UP↑SCALE does this in four layers:
Audit trail in the scale. Every change (calibration, settings, users, recipes, test results, reports) becomes an entry with who, when, what and - in strict audit - why. Entries cannot be edited or deleted in the app or on the scale; the log only grows (the last 10,000 are kept).
Hash chain. Each entry contains the SHA-256 fingerprint of the entry before it, like links of a chain. If anybody changes, removes or inserts one entry, all following fingerprints no longer fit and the check shows it. This is what "tamper-evident" means.
Record files. The app copies new entries into an archive and you can SAVE FOR THE RECORDS: a file with a printout of its fingerprint. Each file names the one before it, so a missing or changed file is found by Check exported archive files (🗂 RECORDS → 🛡 ACTIVITY, archive card: choose several files at once). Strict audit stops work before entries could be overwritten unarchived.
Numbered sales records. The sales log is a chain of its own: every record gets a number of the scale and the fingerprint of the record before it, and the scale checks the weight against its own reading. The sealed report shows gaps and breaks.
Digital seal. Reports, PDF, JSON and CSV exports are sealed: the scale signs the fingerprint of the file with its own key (ECDSA P-256), and our licence key vouches for that scale. Anybody can check a file on the public page Check a sealed file, in the browser, without account or upload.
Is this WORM (write once, read many)? Not as hardware: the memory of the scale can be rewritten, and a file on your disk can be deleted. What we give you is tamper-evidence: changes and gaps cannot be hidden unnoticed. It is made to go together with a procedure of yours: the record files go to write-once storage or a read-only, backed-up share, you keep the printed fingerprint of the newest file separately, and you save the files regularly. The time comes from the device that runs the app (the scale has no clock). Both are covered by the tools of the app: the DEVICE QUALIFICATION form with templates for your IT, the storage settings, the device register and the clock guard of the scale. The key of the scale is kept in its flash memory, so it does not prove genuine firmware. Please take this into your validation; this is not a certification.
🖥 Devices, time and record storage for audits: solved with the app
Two questions come up in every FDA, GMP or ISO audit of a system like this one. Where does the time come from? The scale has no clock of its own: the time of every record comes from the phone, tablet or computer that runs the app, which we call the device; the scale checks it against its own running time and, if you use it, against the signed exact time from up-scales.com (clock guard). A device is what runs the app, a scale is what you weigh on: each has its own register (the device register and the equipment register). Can the records be deleted? The record files are saved on the disk of the device. Our records are tamper-evident (a change or a gap shows up), but a file on a disk can still be deleted. Both points are closed the same way everywhere, with a short procedure (SOP) of your own: every device takes its time from a time server and the operators can't change it, and the record files go to write-protected storage. The app has the tools built in, so this is an afternoon per site, not a project:
Templates for your IT (free): ready-made settings, filled with your time server and your records folder: a Windows .reg file (browser policy, time server, the hidden Date & time page), Linux (chrony / timesyncd and the Chrome policy), an Android MDM policy, macOS, and the browser policy as JSON. The app never changes your systems: your IT checks them and applies them, best by Group Policy or MDM.
The DEVICE QUALIFICATION form (the form and the blank paper version are free, the sealed PDF needs ENTERPRISE & COMPLIANCE): one form per device, with the tests for its system. Each test has a COPY button for its command and a box for the output. Paste what the command printed, and the app proposes PASS or FAIL. For example, it finds "Operation not permitted" or "synchronized: yes", or it sees that the time source is not the local CMOS clock (it also knows German and Spanish system messages). The tester decides.
Filled in by the app: everything it knows goes into the record by itself, so nobody has to type it:
the device: its ID, browser, time zone and app version
the scale: firmware, Device ID, serial number, licence, certificate data, last calibration, event counters and the last routine check
the data-integrity setup of the scale: users and roles (and whether all have a password), strict audit, the protections, the device register, heartbeat lock, tare guard, EXACT WEIGHT, the seal key and its certificate
the clock: the difference between this device and the scale, and every clock jump the scale logged since the last qualification
the archive: the state of the activity-log archive, the newest record file with its SHA-256 fingerprint, and whether the browser keeps its storage
The record, sealed and signed:SAVE SEALED PDF makes a numbered PDF that is sealed by the scale. It has the device data, the data read by the app, every test with its command, the expected and the observed result, the deviations (every failed test gets a row for the action taken), the result with the date of the next qualification, and lines for the printed name, signature and date of the tester and the QA reviewer. If the scale has users, you pick both from the user list. The pasted outputs and a JSON of the form are attached to the PDF. PRINT prints the same record, CSV / JSON exports it for your QA system, and BLANK FORM prints the paper version with all systems and tick boxes. Sign it electronically right after saving it (✍ SIGN REPORT …): the tester signs it as performed, and the QA reviewer (a QM) as reviewed or approved, on this device after logging in or with SIGN A REPORT FILE on their own one. The scale checks the password with every signature and writes it to the activity log. The signed report is saved as an HTML file, and every signature is added to it. The PDF is the sealed copy for printing, with lines for signatures by hand.
The device register on the scale (ENTERPRISE & COMPLIANCE, below): every qualified device, with its due date, the number of its record, and a list for every app.
Where to find it:
the ☑ TEST page, tab 🖥 DEVICES (card DEVICE QUALIFICATION; keys 📋 DEVICE REGISTER, ⋯ → 🖨 BLANK FORM (PAPER) and ⬇ TEMPLATES FOR YOUR IT, and 🖥 DEVICE QUALIFICATION FORM)
the warning line of a device that is not qualified (QUALIFY NOW) at the bottom of the weighing page, and the line ⏱ when the clock of the device differs from the scale (DETAILS): yellow, ✕ hides it until the app is opened again; red without ✕ when the strict audit refuses changes and sales on this device
The form keeps its entries in the browser of the device until you start a NEW RUN (behind ⋯, asked first), so a re-qualification only needs the tests again. The overview of the tests lists the device qualification too.
The form window: the tabs Device (the fields and what the app reads), Tests (a card per group, a block per test with its command, the expected result, your observation and the result - its edge green for ✓, red for ✗), Templates for your IT and Result and deviations (the deviations with an action each, tester and QA reviewer, remarks, the result, the format of the export). The key row: ⋯ (⬇ CSV / JSON, 🖨 BLANK FORM, ↺ NEW RUN) | 🖨 PRINT | 📄 SAVE SEALED PDF (ENTERPRISE & COMPLIANCE; without it PRINT is the main key); right after saving, ✍ SIGN REPORT n ELECTRONICALLY joins them. Without an asset ID the save goes to the tab Station and marks the field.
Who does what: run every test as the operator of the device, not as an administrator, because the point is that the operator can't change the time. The QA reviewer checks the record and signs it.
Time integrity of the devices: Windows, Linux, Android, macOS
The scale takes the time from the app when it connects and counts on from there. Your procedure makes sure that this time is right and that nobody can change it. These are the settings, and the form has a test for each of them:
Windows:
gpedit.msc → Computer Configuration → Windows Settings → Security Settings → Local Policies → User Rights Assignment: Change the system time and Change the time zone only for Administrators (and LOCAL SERVICE for the system time). Standard users may change the time zone by default, so remove them there.
Administrative Templates → System → Windows Time Service → Time Providers: Configure Windows NTP Client Enabled, NtpServer your-server,0x9 (or time.windows.com,0x9), Type NTP, and Enable Windows NTP Client Enabled.
Administrative Templates → Control Panel → Settings Page Visibility: hide:dateandtime hides the Date & time page.
Tests: time 12:00:00 must answer "A required privilege is not held by the client", tzutil /s "UTC" must be refused, and w32tm /query /source must name your time server, not "Local CMOS Clock" or "Free-running System Clock".
Linux:
Operators get a normal account that is not in the groups sudo / wheel / admin.
Use chrony or systemd-timesyncd with your time server.
Tests: date -s "12:00:00" must answer "Operation not permitted", timedatectl set-timezone and set-ntp false must ask for an administrator password, and timedatectl status must show "System clock synchronized: yes" and "NTP service: active".
Android (tablets at the weighing station):
Use an MDM (Android Enterprise) with kiosk mode or a work profile without the Settings app.
Set automatic date and time zone to enforced and switch the developer options off.
Android takes its time from the mobile network (NITZ) or from time.android.com, not from the time server of your Wi-Fi. A company time server needs an option of your MDM or of the maker, such as Samsung Knox.
Test: Wi-Fi off, restart, Wi-Fi on. The time must correct itself.
macOS: operators get a standard account; "Set time and date automatically" on, with your time server; date 1200 must answer "Operation not permitted".
The browser (all systems):
The developer tools must be off (policy DeveloperToolsAvailability = 2). With them, a page could be given another time.
No incognito, guest or new browser profiles. They would lose the archive and get a new device ID.
No extensions, except the ones you approve.
The download folder is the records folder (see below).
Test: paste the list of chrome://policy into the form.
BIOS / UEFI password on PCs, so the hardware clock can't be set before the system starts.
The clock checks of the app:
The app compares this device with the time of the scale (at most 5 s).
A live clock lets the tester compare the device with a reference, such as a radio clock or your time server (limit 2 s by default).
The app shows the time zone.
It counts the clock jumps since the last qualification.
Record storage: write-protected (WORM) storage for the record files
The app saves every report, record file and export as a download. Real WORM (write once, read many) is a feature of the storage, not of an app. So the files have to land on storage where they can be added, but not changed or deleted:
Best: a NAS or server with a WORM folder or immutable snapshots, for example Synology WriteOnce, a QNAP WORM folder, or ZFS / btrfs read-only snapshots. Cloud storage with an object lock in compliance mode also works, for example S3 Object Lock or Azure immutable blob storage. Keep a second copy as well.
Make the download folder the records folder on every device with the browser policies DownloadDirectory (your records share) and PromptForDownloadLocation = false. Every record then goes there by itself, without a save dialog. AutomaticDownloadsAllowedForUrls (the address of the app) lets the archive device save its daily record files without asking. The Windows and Linux templates contain this.
Back up the records share (and keep a copy elsewhere): the daily record files of the activity log and of the LEGAL MODE store are the only complete copy of what the scale overwrites. The archive inside the browser is a working copy, not a backup.
Not WORM, but write-protected:
Windows: a folder where the operators may only create files (NTFS rights: create files / write data on the folder, files read-only, no delete).
Linux: a folder with chattr +a (files can be added, not deleted or renamed).
Together with our hash chain and seal this is tamper-evident, but an administrator can change it. Say so in your validation, or add the WORM storage.
Android: the tablet is never the archive. A sync app or the download folder of the MDM brings the files to the records storage.
The tests of the form:
Save a test file. It must land in the records folder without a dialog.
As operator, try to delete it, change it, rename it or overwrite it. Each must be refused.
Show the retention setting and the second copy.
Write the fingerprint of the newest record file into your binder. The app shows it.
The device register of the scale (ENTERPRISE & COMPLIANCE)
A scale is often run from several devices: the PC in the lab, a tablet at the bench, the phone of the QA. Each of them is a device and needs its own qualification. The scale keeps the device register, the same list for every app (up to 50 devices):
the asset ID, the system, the state (qualified, failed, due, retired), the date, the due date, the report number and who saved it
devices that are connected now are marked ●
How it works:
Add a device: run the form on it. SAVE SEALED PDF enters the result in the register.
Re-qualification: after 3, 6, 12 or 24 months, as you choose on the form. A due device is shown as due.
Remove a device:RETIRE it (⋯ at the end of its row), with a reason chosen in the window (replaced, broken, lost, not used any more, other). It stays in the register, because older records were made on it. A device is not deleted at once. When all 50 places are taken, the oldest retired device is replaced.
Keep the list clean: a retired device can be taken off the list with 🗑 at the end of its row (REMOVE, asked first). Without users on the scale that works at once (for example for a test entry from the setup). With users, only an ADMIN or a QM can do it, with a reason in strict audit, and not before 12 months after the retirement, so a device that was in use during the last year stays on the list for the auditor. A device that is still qualified, failed or due has to be retired first. The removal is an entry in the chained activity log (device, record number, who, when), so the log still shows that the device existed - the log itself is never cleaned.
A scale that is out of service: in the box QUALITY RECORDS & USERS, the list All scales seen by this browser has a 🗑 (REMOVE) at the end of the row of each scale. It cleans this browser only: the register line, the archive of the activity log of that scale, the access-register notes and the remembered checks. It works only when a saved record file covers the whole archive (otherwise the app asks you to save the record first), and not for the connected scale. The scale itself and the record files on your disk are not touched - keep those as long as your retention rules say.
Who: with users, only an ADMIN or a QM can save or retire a device. With strict audit, a reason is asked like for every other change. Every change is written to the activity log.
Enforcement: once at least one device is qualified and strict audit is on, a device that is not qualified (or is due) can still weigh and print, but the scale refuses its changes and sales. A sale is kept in the app and sent again after the qualification. Without strict audit, the app shows a warning bar, and the scale writes a log entry when such a device connects.
📄 PDF of the register window (or 📄 REGISTER AS SEALED PDF on the 🗂 RECORDS tab 🖥 DEVICES) makes a sealed PDF of all devices, also the retired ones, with the device events and clock jumps from the archived activity log. ⋯ exports the register as CSV or JSON; QUALIFY THIS DEVICE (the main key) opens the form.
What a device is: a browser profile with a random ID that the app makes once. It is not a hardware ID. The asset ID on the record and your SOP tie it to the device. A new browser profile is a new device, so block new profiles with the browser policy.
The clock guard of the scale
Every app sends its time when it connects and every hour while it stays connected. In between, the scale counts on with its own timer, which does not overflow. A scale that runs for months without a restart therefore keeps the time of your synchronized devices; its chip alone would drift a few seconds a day. A small correction backwards never makes the time of the log go back. Summer and winter time change nothing: the scale counts in UTC, and every record is shown in the local time that applied at that moment. Sales also keep the UTC offset of the app, so the hour that repeats in autumn stays in the right order. The scale compares every time it gets with its own running time.
A difference of more than 2 minutes is a clock jump. The scale writes it to its activity log with the user, the seconds and the time that was sent. The scale allows a little more after long running times, because the clock of the chip drifts slightly.
With strict audit on, the scale keeps its own time and refuses the jump. Without strict audit, it takes the new time and keeps the entry.
Every app shows a warning when its own clock differs from the scale by more than 2 minutes.
The form lists the clock jumps since the last qualification.
A time source of its own: 🕒 EXACT TIME FROM UP-SCALES.COM (card LICENCE on the scale info page, row 🕒 Exact time → FROM UP-SCALES.COM, every edition). One tap, and after your OK the app fetches the exact time once, taken from several public time servers and signed for this scale (it is only valid for this scale and only once). The scale sets its clock to it - also back - and writes it to the activity log as exact time signed by up-scales.com, with how far its clock was off. From then on every device is checked against this time, not only against the one before it: a device that is more than 2 minutes off is a clock jump (see above). Tick automatically at every connect, and after every restart of the scale the first online device gets the signed time again a few seconds after it connected. A device without internet goes on as before. Only the Device ID and a one-time random number are sent.
This shows a wrong or changed clock. It does not replace the time server of your devices: the time of a record still comes from the device that runs the app, and differences below 2 minutes are not logged. Without the signed time, the first app that connects after a restart of the scale sets its time again.
All of this is a record of your own qualification for your validation binder, not a certificate. We describe what the tools do. Your QA decides whether they fit your procedures (21 CFR Part 11, EU GMP Annex 11, ISO/IEC 17025, ALCOA+).
Validation of the calculations (IQ/OQ)
The ENTERPRISE & COMPLIANCE edition has a box VALIDATION OF THE CALCULATIONS (IQ/OQ) in the tab 📑 VALIDATION of the test page: RUN VALIDATION computes reference cases worked out by hand from OIML R 76-1 (the routine check from USP <41> / Ph. Eur. 2.1.7, the air buoyancy correction from OIML D 28 / R 111-1, the pipettes from ISO 8655-6:2022 / ISO/TR 20461:2023 - water and air density, Z factor, volumes (also against the published examples of EURAMET cg-19 and the model of NIST SOP 14), uncertainty, decision rule, the limits (secondary sources of ISO 8655-2, and your own table), the balance table and the deviations from ISO 8655-6:2022, CMC, the amount of a sale (price x weight to the cent), the lot check of ℮ packs) with the code that judges your tests - the cases of the tests by the UP↑SCALE firmware (connect your UP↑SCALE first), the others by the app - and makes a printable record with the fingerprints of the app code and of the UP↑SCALE firmware for your validation file (the record, its HTML file and the PDF: the keys of the box and its ⋯). ⋯ → EXPORT DATA (CSV / JSON) saves the same run as raw data for your QMS - a table with the IQ data and one row per case (expected, calculated, pass/fail), in the format of the export picker and sealed by the scale like every export; the LIMS forwarding can also send every run to your server by itself. The description, what it does not cover and the full list of the cases with their working are on the public page Check & validation for officials and auditors.
Validation pack (IQ / OQ / PQ)
The ENTERPRISE & COMPLIANCE edition makes one document for the validation file of your laboratory: the box Validation pack (IQ / OQ / PQ) in the tab 📑 VALIDATION of the test page (and 📑 VALIDATION PACK behind ⋯ in the key row of the pipette register). Pick the chapters - always the UP↑SCALE system with its weighing data, users, audit trail, signatures and records; optionally the pipette check and calibration, the compounding assistant with the solution mode, the routine check with the minimum weight and the air buoyancy correction, the fill weighing with the check of prepackages, the density mode and the LIMS / ELN forwarding (the app remembers your choice) - give the system a name and press SAVE AS PDF (numbered, sealed, signable) or ⋯ → PRINT / HTML. Run the validation of the calculations first in the same session (▶ RUN VALIDATION of the box), else its chapter stays empty.
Purpose, scope and intended use - and what the pack does not cover (the metrology of the balance itself, your SOPs and IT, the pipetting technique).
System description, filled in from this device: app version and code fingerprint, firmware and its fingerprint, Device ID, edition, weighing range, certificate data, users, qualified devices; for the pipettes the instruments, their uncertainties, the run settings, the decision rule, the source of the limits with the SHA-256 of your limits table, the CMC version.
Validation approach and acceptance criteria (risk-based, GAMP 5 and ISO/IEC 17025 7.11.2).
References with their editions: ISO/IEC 17025, 21 CFR Part 11, EU GMP Annex 11, GAMP 5, OECD GLP No. 17, PIC/S PI 041, OIML R 76-1, USP <41> / <1251>, EURAMET cg-18; for the pipettes ISO 8655-1/-2/-6:2022, ISO/TR 20461:2023, GUM, ILAC P14 and G8, EURAMET cg-19, NIST SOP 14, Tanaka 2001, OIML R 111-1; for the other chapters e.g. Ph. Eur. 2.9.5 and 2.1.7, USP <795>, PIC/S PE 010, OIML D 28, 76/211/EEC, NIST HB 133, OIML R 87, RFC 8259 / 9110, FIPS 180-4 and the MHRA data integrity guidance.
Requirements and their tests (traceability matrix) and a risk assessment pre-filled for you to confirm.
IQ: each check with what the app found and met / not met where it can judge (e.g. the instruments within ISO 8655-6 table 2, the d of the balance for every pipette of the register).
OQ part A: the result of the reference cases of this device by area, every case in annex A; OQ part B: test scripts with steps and expected results and blank fields for the tester - the weighing data path, audit trail, roles, lockout, signature, seal, export, time, power cut; for the pipettes a delivery with a calibrated weight, Z, limits, implausible values, evaporation, balance and deviations, CMC, records, import, multichannel, a lost app; and the scripts of each chapter you ticked (e.g. lots and identity check, QM release, a tolerance narrower than d; a failed routine check on every device; the ℮ limits and a run that goes on after a power cut; the water density; a server that is not reachable).
PQ: the test suite of the balance; for the pipettes the comparison with an external calibration (En ≤ 1), pre-filled with your last calibrations; for every other chapter a PQ in routine use with its pass criterion and fields to fill in.
21 CFR Part 11 / Annex 11 checklist (what the system brings, what stays with your procedures), supplier information, change control (when to repeat what), and the pages for deviations, conclusion and approval.
The VSR takes the signed pack as its IQ/OQ document. The validation is the work and the decision of your laboratory; the pack gives the structure, the data of the system and the tests - no certification.
Validation Summary Report and release for use (VSR)
The ENTERPRISE & COMPLIANCE edition has a box VALIDATION SUMMARY REPORT & RELEASE in the tab 📑 VALIDATION of the ☑ TEST page (below the validation pack) - the route for the final sign-off of your QA team:
Pick the two reports you saved and signed before: the IQ/OQ record (validation of the calculations, or the signed validation pack) and the PQ baseline report (the final report of the test suite), each as HTML or as our PDF. The app checks each one at once: integrity code, seal of the scale, its signatures, and that it belongs to this scale; it shows the SHA-256 of the file.
✍ ROUTE FOR FINAL SIGN-OFF makes the VSR: purpose and scope, the two reports with report number, integrity code, SHA-256, check result and signatures, a hardware and software inventory made at that moment (scale, Device ID, edition, Max / d, firmware, certificate data, calibration and configuration counter, register entry, routine check, minimum weight, the qualified devices, app and browser), the open points the app found (a missing, changed or unsigned report, an open routine check failure), the release decision and three lines to sign on paper. It is saved as a sealed PDF with a report number; the two reports are attached to it unchanged. Open points are listed, they do not stop you.
Sign it: a QM / approver signs the VSR approved - right away, or later with ✍ SIGN A SAVED VSR - with the password, which the scale checks again (never the person who made the VSR: four eyes). This signature releases the scale: it keeps "released by VSR no. X, by QM, when" together with its event counters of that moment, writes it into the activity log and every phone and computer shows RELEASED FOR USE. You can also print ⋯ → 📄 VSR WITH LINES TO SIGN (PDF) and sign on paper - that is a record for your binder, the status on the scale needs the electronic signature.
After a calibration, a change of the configuration or a new firmware (the counters of the scale go up; a new firmware build is written into the activity log as Firmware changed), the box shows RELEASE TO BE CONFIRMED: make and sign a new VSR. When the firmware or the app is not the one of the attached IQ/OQ document (their fingerprints differ), the VSR lists it as an open point: run the validation again. Only the UP↑SCALE 420, 710 and HDS calibrate themselves (they read their own load cell). The RS-232 adapter, the DIY and the virtual scale show only the configuration counter in their reports and in the VSR - the balance on the cable is calibrated by itself, which the UP↑SCALE does not see: after a calibration, adjustment or exchange of the balance start a new release by your procedure. A QM can ⋯ → WITHDRAW RELEASE (logged). An RS-232 adapter keeps one release per scale profile. Nothing is blocked - the release is a status for your QA and its procedures. It is your own release decision, no certification of the device by UP↑SCALE.
The editions of the tests
One rule for the scale tests and the pipettes:
FREE: the simple checks with their results on the screen - the routine check, the calibration check, repeatability, eccentricity and the weighing test with the default loads - and the choice of the standard.
PRO (also PRO VIRTUAL): the extended checks - tare, zero-setting, discrimination, creep & zero return, warm-up, stable equilibrium, the strict procedure with its checklist - and a plain print of a single test (not numbered, not sealed); the pipette check as before.
ENTERPRISE & COMPLIANCE: the scale calibration along EURAMET cg-18 with its uncertainty, the pipette calibration, the final report, numbered and sealed reports with sign-offs (print / PDF, HTML, CSV), the certificates, the service documents, the registers of the test weights and of the customers' scales, the due list and the history.
The standard: OIML R 76 / EN 45501 or NIST HB 44 (free)
The block ⚖ STANDARD of the tab 🔍 SELF-CHECK sets the rules the UP↑SCALE judges every test by - free in every edition. Region: EU / EEA, Germany, Switzerland and the United Kingdom use OIML R 76-1 / EN 45501; the USA uses NIST Handbook 44, Section 2.20. The region also shows the usual legal verification interval (Germany: 2 years; the EU: set by each country; the USA: set by each state or county). Judged for:verification (the limits of a first verification; HB 44: acceptance tolerance) or in-service inspection (twice these limits; HB 44: maintenance tolerance).
Where both standards give the same result (classes I and II, class III up to 4000 e, class IIII up to 400 e, for the weighing test) the options and the report say "OIML R 76 / NIST HB 44". Where HB 44 differs - a fourth tolerance step for classes III and IIII, the range of the repeatability and shift (corner) tests, discrimination with 2 d, zero return 0.83 e - the UP↑SCALE judges by the standard of the region, and the report names the standard and the purpose it applied. Not covered: class III L, the class III footnote for prescription scales (e from 0.01 g), wheel-load and axle-load weighers and crane and hopper scales; class IIII may have up to 1 200 e. It stays a self-check: no verification, no legal value. The extended tests judged by HB 44 stay PRO, like the tests themselves.
Typed readings: scales without a link
A scale with no connection - any maker's scale - can be tested too: type the readings you see on its display into the test form. The results are worked out on a connected UP↑SCALE, or, without one, on the free virtual scale in the app - the same maths as on the scale. Typed readings follow the edition of the test: a PRO test with typed readings needs PRO. The report marks the readings as typed. With the strict procedure, rounded readings need the changeover step ΔL for every row; a run that cannot be judged that way is refused with the reason.
⚖ Scale calibration along EURAMET cg-18 (ENTERPRISE & COMPLIANCE)
Where: ☑ TEST, tab ⚖ CALIBRATION, card SCALE CALIBRATION (EURAMET cg-18). It measures the error of indication of a scale at several loads with its expanded uncertainty, and the uncertainty of a weighing result in use - for the calibration certificate of your lab. It does not adjust the scale (that is the calibration in the settings). Six steps, one form each (← / ▶), and the draft is kept when you reload:
Error of indication: the loads - the app picks the weights for every load; where the weights do not reach, substitution loads (substitute material, counted along cg-18 in the conservative way).
Conditions and the uncertainty in use: temperatures, air (for the air buoyancy) and the use of the scale.
Result and certificate: the errors E with their expanded uncertainty U, the certificate data, the checklist of the document rules, PRINT DRAFT or ISSUE.
Readings come from the stored test results of the connected scale or are typed. All the maths runs on the UP↑SCALE, the report shows the values it used. U is the expanded uncertainty with the coverage factor k from the effective degrees of freedom (about 95 %), rounded up to two significant digits. The uncertainty in use (a straight line over the range, the global value and the minimum weight) is only stated when the fit to the errors passes; otherwise the certificate gives the errors and U per load only. The maths follows the formulas and worked examples of the guide; the reference cases are public on the check page.
🧱 Test weights (ENTERPRISE & COMPLIANCE): weights and sets are entries of the instrument register - class (OIML R 111 E1 to M3, ASTM E617 classes 000 to 7, NIST class F), certificate, its issuer (lab and accreditation), calibration date and due date, k of the certificate, and per piece the nominal value, the correction and U. Only the OIML classes have a table of limits in the app; a weight of an ASTM class or class F needs the correction and U of every piece from its certificate.
⚖ Scales of the customers (☰ → SCALES; ENTERPRISE & COMPLIANCE): the scales you calibrate - of any maker, connected or not: customer, site, location, Max, d and e of up to three ranges, class, interval and due date. One customer list for scales and pipettes. Every change goes into a chained history; a scale is retired, never deleted. The due list shows everything that is due - scales, pipettes, instruments and test weights - and saves it as a calendar file.
📜 Calibration certificates of scales and pipettes (ENTERPRISE & COMPLIANCE)
The certificate is always the certificate of your lab; the app fills it in and checks it. Two types:
Standard certificate: the certificate of a calibration by your lab or service, with your company data.
Accredited certificate: only for a lab with its own accreditation. You enter your accreditation body, your number and upload your own accreditation symbol as an image (the app ships none), and your calibration and measurement capabilities (CMC). Results outside your accredited scope or outside the CMC are marked with *. An accredited certificate stays pending until an authorised signatory approves it with an electronic signature.
Completeness: before a certificate is issued, the app checks its content against the listed rules (ISO/IEC 17025 7.8 and the rules of the accreditation bodies and the guides); every rule shows ✓ or ✗, and a missing required item stops the issue. The app checks completeness against the listed rules - the lab stays responsible for the content. Procedure register: your calibration procedures with title and version (changed or retired, never deleted). Signatories: who may approve which certificate. Certificate register: every issued certificate with its number, a chained history and the link to the one before; a correction is a new certificate as an amendment that replaces the old one (which stays), and a certificate can be withdrawn with a reason. A connected ENTERPRISE UP↑SCALE writes each of these steps into its activity log. Pipette certificates (ISO 8655) go through the same path, with the same lab data, CMC and registers.
📋 Service documents of the legal metrology (ENTERPRISE & COMPLIANCE)
With the results of the TEST MODE (connected or typed runs, judged on the UP↑SCALE with the standard and purpose it applied) the app fills in a service document - the button 📋 SERVICE DOCUMENT in the report row of the tab 🔍 SELF-CHECK:
Placed in Service Report (USA): for a registered service agency, with the fields of NIST HB 130 (the forms of the states differ: a generic layout).
Repair and test protocol for the verification authority (Germany): for a repairer or service company; the authority may take it into account.
Verification certificate of a private body (EU: Austria, Spain, France, Italy, Belgium, Sweden): for a body that may verify scales in use in its country.
The document is the document of the agency, the body or the company that issues it; the legal verification stays with the authority. Seals removed and applied, the type approval, the software and the event counter go into the document and into the scale's register. The content is checked against the listed rules like a certificate, and the document is numbered, sealed and registered the same way.
📦 The data file: move your registers to another browser
Where: ☰ → DATA → 📦 DATA FILE. One file (UPSCALE-DATA-1) for the registers and settings of your lab or service - to move them to another browser or to keep a copy. Tick the parts you want, one, several or all: company, tester and lab settings; customers; scales of the customers; test weights; instruments and sensors; pipettes, models and CMC; runs; certificates; scales seen by this browser; scale profiles of a connected adapter. A part can be cut to one customer. The file is sealed with ENTERPRISE & COMPLIANCE; the parts follow the editions of their registers.
IMPORT merges by default (MERGE): the newer version of an entry wins and both histories are kept. CLEAN OVERWRITE makes the state of the file the state here, for the ticked parts only. A history is never shortened or rewritten: an overwritten entry keeps its history with one more entry, an entry that is only here is retired (not deleted), and runs and records are only added. An entry with a broken history chain is not taken. The activity log, the audit archive, seals, users and devices are never in the file.
For inspectors, verification offices and service companies
The test suite is also a work tool for people who test scales for others: verification officers and their helpers, calibration and service technicians, quality managers. It does not replace their reference weights, their procedures or their decision - it takes the paperwork and the arithmetic off their hands and makes the record error-proof:
Every test of OIML R 76-1 Annex A.4 that can be done at the scale, in one guided suite: repeatability, accuracy & linearity (weighing up and down with the changeover step ΔL), eccentricity, tare, accuracy of the zero-setting and the zero indicator, discrimination, creep (30 minutes, with the 4 hour alternative) and zero return, warm-up time (after 8 hours off), stable equilibrium - plus the calibration check. The strict procedure fills in the loads, preload, legal display and steps of the standard.
The checklist shows what is done, with which result, and which test is asked for a verification / conformity assessment, an inspection in use or a calibration - so no test is forgotten.
The limits are calculated by the app: the maximum permissible errors of the class that fits d, e and Max, the uncertainty of the readings, the test weight tolerance (R 111, 1/3 of the mpe), the room temperature change and the warm-up are checked, and every rule shows ✓ met, ✗ not met or ? not known.
Your data on every printout: name, organisation, inspector ID, place, report number, remarks and a signature that you draw once on the screen (saved only in this browser).
The FINAL REPORT puts the scale data and its settings, the checklist, a summary per procedure, all result tables with the single readings, the class overview and the lines for date, signature, stamp and remarks on certificate-style pages. Print it or save it as PDF; the CSV has all readings for your own records or software. The inspector reads the legal display of the scale while testing, ticks "values checked against the display" and signs.
With the RS-232 adapter the same suite runs on an existing verified scale: with EXACT WEIGHT the readings are the values of that scale, unchanged (the report names the mode). The adapter is no part of the legal measuring chain; the display of the verified scale stays the legal reading.
What it is not: the report is a test record that the signing person is responsible for. Whether it is accepted as part of a verification, a certificate or an inspection is the decision of the responsible authority or accredited body; UP↑SCALE is not affiliated with OIML, EURAMET or any authority, and the report uses none of their forms or logos. Which tests an inspection in use asks for differs from country to country - the checklist shows a usual set; please use the rules of your own authority.
The report
All tests of this session are collected in one report: the scale (name, firmware, serial number, max. load, d, e, classes), the settings during the test, the date of the last calibration, every value with its error and limit, and the status of each class. OPEN REPORT (PRINT / PDF) opens it in a new window - use the print dialog of your browser to print it or to save it as PDF. DOWNLOAD HTML saves the report as a file that opens and prints on any device, DOWNLOAD CSV saves the raw values for your spreadsheet. The report says that it is a self-check and not a verification. With PRO a single test prints as a plain page (not numbered, not sealed); the numbered, sealed reports, the HTML and the final report need ENTERPRISE & COMPLIANCE. The report names the standard and the purpose the UP↑SCALE judged by.
⚖ Config+++
Weighing things is not as easy as it maybe looks ;-) Measuring weight digitally in real life is quite interesting actually, if you break it down. To explain it fastly, our digital scales is made from 3 main components, a Microcomputer, a analog-to-digital converter and a so-called load-cell. Inside the load cell - which is this is a transducer whose function in the scale is to convert the force exerted by the object of measurement into an electrical signal - is a set of strain gauges arranged like the resistors in a Wheatstone bridge. When a load is placed on these strain gauges, they compress. This compression changes the resistance of the gauge, just like in a Wheatstone bridge. Now there will be a voltage output corresponding to the change from the stressed gauges - which we will transform to a digital signal, with the help of a Sigma-Delta analog to digital converter, to be exact. This digital signal we can use now to calculate the weight.
But - and here it gets complicated - there are many variables! There is nothing like an absolute and always same weight of a thing, it depends how, how long and where you are weighing it, how the conditions are around, if the thing is maybe moving and and and... Because of that there are many variables in the background which define how the weight will be read from the sensor, how the data will get stabilized and displayed, when the weight is getting "stable", what you define as stable weight, and so on and so on... The output at the end should not jump around too much, should be reliable and repeatable - but also need to be shown as fast as possible. While normal scales do everything automatic for you, and normally need to find some kind of compromise, professional scales will let you adjust some-, and our scale will let adjust you all of this settings! Maybe you want your scale faster reacting for filling things - or you need a long average weighing time because you need to weigh some moving animals or maybe you need the scale a bit less reactive because you are working outside or people are bumping on your table all the time! With our scale this is no problem at all, you can adjust every single variable - or just leave it in the automatic mode!
This Settings you will find within your scale-app, at the Scale-Settings page. Here you can define how fast the sensor should be, how it will calculate the averages, how long it will take to define the weight as stable and if you want to ignore, for example, small bumps on the display - and how big they can be.
PART 5 Connections & Hardware
⚖ 1, 2, Many Scales!
Using Multiple Scales
Yes, you can connect more than one scale at the same time!! Just connect an additional scale within the app and weigh on all of them simultaneously! Maybe you are using a big platform scale together with a smaller tabletop one? Just connect them both and track all ingredients at the same time and in the same program!
Weigh on all connected scales separately - but also get the combined weight of all of them.
To connect with an additional device, open ☰ → 🤝 CONNECTION → ⚖ ADD MORE SCALES, select your scale, and pair with it (the next one: ⚖ EVEN MORE SCALES).
All scales are shown in the weight display above the (now combined) weight, each with its own tare key ⚞T⚟. The standard TARE key works for all scales now.
To disconnect one of the additional scales again, tap its ✕ in that row. The scale goes back to the normal weighing mode.
If an additional scale loses its connection (out of range, switched off), it is removed automatically and its weight isn't counted anymore. Just add it again.
The weights are combined in the classic weighing- and in the recipe mode, and in the 🛍️ SALES, 🥦 CCLUB, ☕ ROASTERY and 🏪 DISPENSARY MODE (also in 🛍 fill weighing): the big weight, the booking and the price take the sum of all scales, stable only when all of them are - e.g. a platform scale for the sacks and a fine scale for the bags, both on RS-232 adapters, in one app at the same time.
Devices in counting, check-weighing and sorting: here the additional scales are not added up - every scale is a device of the program on the main scale. In counting each scale (also the main scale as ⚖1) shows its weight and pieces small on top, with the piece weight of the main scale - the big number in the middle is the sum of all scales, like in the weighing mode, and so are the average and total below it, the units, the label and the ∑ log (also in the takeaway mode), in the check-weighing PERFECT, OK, TOO LESS or TOO MUCH (with the ranges of the setup in this app - otherwise the % of the setpoint), in the sorting mode the category of the item on it. Only the main scale counts, logs and fills the statistics.
Two-scale reference counting: count the sample on a fine scale (e.g. a 710) and the bulk on a big one (e.g. a platform scale with the 232 UP↑SCALER) - see piece weight from an other scale.
A connected additional scale can also exchange its THINGS, RECIPES and MOLDS with your main scale: see ⇄ sync between scales.
Scales of other makers:⚖ ADD MORE SCALES also lists Bluetooth scales of other makers: scales with the Decent Scale protocol (also the Half Decent Scale with its own firmware, EspressiScale, ButtsHaus Scale and the DIY Bluetooth boards "Tencent Scales"), BOOKOO (Themis Mini, Themis Ultra), Acaia (Lunar, Pearl, Pearl S, Pyxis, Cinco, Umbra), Felicita (Arc, Parallel), Timemore (Black Mirror 2 / DUO, Basic 3, Basic 3.0 Link, Dot - the Dot asks to pair once), Varia AKU, DiFluid Microbalance (and Ti), Hiroia Jimmy, Skale 2, Eureka Precisa / Solo Barista, Atomheart Eclair, WeighMaster / MantaBrew, WeighMyBru, Wilfa Svart Uni, IKAPE (IK-Cafe), INSMART 863A, coffee scales with a DSD TECH module, the Etekcity ESN00 nutrition scale, many no-name kitchen scales (the FFB0 type of FitDays / ICOMON: MY_SCALE, blackcoffee, NSCALE, Nutridays; Chipsea "smartchef"; Lefu / Futula / LFSmart; QN-KS (Yolanda / Arboleaf CK10A); Sencor), the DIY coffee scales mpy-coffee / coffee-scale, the old Perfect Company scales and scales with the standard Bluetooth weight profile (mostly body scales, 5 g steps). Scales without a tare command (Chipsea, NSCALE, QN-KS, Sencor, Perfect Company, DIY, standard profile) are tared in the app. The IKAPE scale sends no weight below zero (it shows 0 g then). A scale whose maker or model the app does not know says so when it is chosen. The maker's app is not needed: our app talks to the scale directly - no account, no sign-up, no maker cloud, nothing phones home and nothing tracks you; the weight goes over Bluetooth straight from the scale into the app on your device. They join like an additional scale of ours: their weight is shown on top and added to the sum, the ⚞T⚟ button and the big TARE tare them on the scale itself. They have no THINGS, recipes or programs of their own, so they are no sync or backup scale. With an RS-232 adapter (232 UP↑SCALER, DIY board) as your main scale the app asks whether the other scale shall be the weight source of the adapter instead: then the adapter gets its weight like that of a scale on its cable and does everything with it - tare, filters, all programs, the sales log, its licence and its data, for every app connected to it. The app passes the weight on, so keep it open; without it the adapter gets no weight. A sale booked this way is marked RELAY in the sales log. Press ⚖ ADD MORE SCALES again to stop it. Where the scale sends no stable mark (e.g. Decent Scale, BOOKOO, Acaia, Felicita), the app calls the weight stable when it stays within 2 digits for a second. The first time such a scale is added, the app loads the part that talks to it (a small download, offline afterwards). Bluetooth serial adapters on the RS-232 port of a scale (u-blox like the OHAUS Scout kit, Nordic UART, Microchip, HM-10 / JDY / HC-08 / BT05 type) are listed too: choose one with CONNECT and the virtual scale reads the RS-232 text of the scale (all presets, auto-detect, requests); or make it the weight source of an RS-232 adapter. As an extra scale they do not work (they need the RS-232 reader of a firmware). Please tell us how it works with yours. All ways to connect in one place: connectivity. The names of these scales and their makers belong to their owners; UP↑SCALE is not affiliated with or endorsed by them.
The virtual scale (without UP↑SCALE hardware): no UP↑SCALE scale or adapter at hand? The app runs our scale firmware inside your browser - UP↑SCALE (VIR) - and puts your own scale in front of it, with the same programs and settings. Bluetooth scale of another maker (see scales of other makers): press CONNECT and choose it. Serial or USB scale (or a USB-to-RS-232 cable) on a computer with Chrome, Edge, Opera or Firefox (from version 151): press CONNECT in the row 🔌 USB cable and choose its port (🔌 USB cable → CONNECT on the start page; Chrome / Edge also list the classic Bluetooth serial adapters (SPP: Mettler ADP-BT-S, KERN YKI-02, HC-05) once they are paired in the system). Bluetooth serial adapter (Bluetooth LE) on a scale: CONNECT and choose it. The app then looks for the settings of the scale by itself (LOOKING FOR YOUR SCALE, up to 40 seconds: protocol, baud rate, data bits, unit and d, see auto-detect) - press USE IT, done. Not found? SET BY HAND leaves the RS-232 settings open. The browser remembers the scale: next time 🔌 USB cable connects to it at once, without the port list (as long as no UP↑SCALE is plugged in too). Only on a port of an ESP32 board (the USB of our adapters and DIY boards) that does not answer, the app asks first. ⌨ A scale that types like a keyboard (balances in their USB keyboard mode: A&D Quick USB, Sartorius / Mettler PC-Direct, Shimadzu WindowsDirect - and the USB or Bluetooth dongles of other makers that type the weight into Excel): press CONNECT in the row ⌨ Keyboard scale on the start page (the app asks once in its own window, OK starts it). The virtual scale takes every line the scale types (up to Enter or Tab, or a short pause) like a line of its RS-232 port - press PRINT on the scale or let it send by itself. Works on computers and, with a Bluetooth keyboard dongle, on phones and tablets. Keep the app in front and click no text field meanwhile: a text field gets the typed text itself (barcode readers keep working there). If numbers or the unit come out wrong, set the keyboard layout of the scale or dongle to US/English or to the layout of your device; a scale that types in ounces or pounds: set the unit in the RS-232 settings. The app can not ask such a scale for its weight or tare it: TARE tares in the app. A line of only digits (8 or more) is taken for a barcode and left out. 📡 A scale that only broadcasts (Medisana kitchen scales and others that send their weight without a connection): press CONNECT in the row 📡 Broadcast scale on the start page. Browsers listen to such broadcasts only with a test switch: Chrome, Edge or Opera on a computer or an Android phone, chrome://flags → Experimental Web Platform features on, restart the browser (not on iPhone / iPad, not in Firefox) - the row is gray then and its i explains it. The window lists the known scales with their weight: USE IT and the virtual scale takes that weight (TARE tares in the app). Your broadcasting scale is not known yet? RECORD FOR US: put a few weights on and off, STOP + SAVE saves a text file - send it to us with the maker and model, and if we make it work, you get a free lifetime licence. 🧰 No scale at all: test weights - to try the app: press TRY IT in the row 🧰 NO SCALE? TRY IT WITH TEST WEIGHTS on the start page (the last of the other ways to connect) - it works in every current browser, also in Firefox and Safari (the ways to connect your browser can not use are gray on the start page: a tap tells you why and what to do). The app keeps the language of your browser. The app opens on the virtual scale with a wooden box of test weights at the bottom of the page. Click or tap a weight: it flies onto the scale; right click, a long press or its red badge (×2 −) takes one off again. A moment after a weight left its cavity, the cavity is full again - so you can put on several of the same weight. Small weights that make up a bigger one are swapped for it, as you would do on a real scale (6 × 1 g become 1 g + 5 g; one smaller weight stays). Three sets: LAB (1 mg to 100 g), METRIC (0.1 g to 1 kg) and OZ (1/16 oz to 16 oz) - the weights of another set stay on the scale; the labels follow the unit of the app (mg, g, kg). ALL OFF takes all weights off, ✕ takes all off, disconnects the virtual scale and goes back to the start page. The weight goes through the firmware like the weight of a real scale: tare, the stability light, ∑ LOG IT (take the weight off afterwards), the programs and modes work as with a scale. Meanwhile the virtual scale shows steps like a lab balance (0.001 g up to 4 kg, 0.01 g up to 40 kg, 0.1 g above - so every milligram shows, also with heavy weights on); its own setting comes back when it gets its weight from another scale. Keep the app in front: in a hidden tab the browser slows the box down. The licence is the one of the virtual scale (below): FREE with 1 THING, PRO for the first 3 days, PRO VIRTUAL to buy, no ENTERPRISE & COMPLIANCE; it belongs to this browser on this device (cleared website data, a private window or another browser = a new virtual scale). The first time the app loads the virtual scale (a small download, offline afterwards). Free: the free version of the app with no time limit (all units also in volumes, prices and nutrition values, nutrition diary, counting, simple check-weighing, coffee timer, cocktails, recipes, labels, the calculators, 1 extra scale, tracking 15 minutes) with 1 THING and 1 recipe saved - and the full PRO for the first 3 days, automatically. PRO VIRTUAL: the PRO functions at the own price of the virtual scale (24.90 EUR a year or 89.90 EUR once, incl. VAT; the four versions), one licence for 2 browsers (for example your phone and your laptop: each has its own virtual scale and Device ID and gets its own key); in the shop choose PRO VIRTUAL as the licence (the card 🛒 BUY in the scale info of the virtual scale opens the form with its Device ID, add the second one below it). APP ONLY or UP↑SCALE hardware - all differences (the most important: no ENTERPRISE & COMPLIANCE, the data lives in this browser - back up your THINGS and recipes -, every browser is its own virtual scale):
Why the difference? With UP↑SCALE hardware (our RS-232 adapter, a DIY board or our scales) the UP↑SCALE firmware runs on the device itself, so it can do it all: your data stays on the device, all your phones and computers work with it at the same time, the device has its own connections, with PRO the Bluetooth link is encrypted, and ENTERPRISE & COMPLIANCE adds tamper-evident records. The virtual scale is the simpler way: the same firmware runs only inside your browser, with the scale you already have. Some advanced functions can not work there - that is why its licence costs less.green better with UP↑SCALE hardware · red a limit of the virtual scale
APP ONLY the virtual scale, licence PRO VIRTUAL
UP↑SCALE hardware adapter, DIY board or our scale
What you need
your own scale of another maker and a browser - nothing to buy, nothing to install
the supported Bluetooth scales; serial and USB scales on a computer (Chrome, Edge, Opera, Firefox from 151); Bluetooth serial adapters; scales that type like a keyboard - no USB postal scales
every scale with an RS-232 port (protocol presets and auto-detect), USB scales through an ESP32-S3 adapter, our own scales
Where the scale runs
in the app - it works only while the app is open
on the device - it keeps working without the app
Your data
in this browser - another browser or cleared site data gives a new, empty virtual scale (back up your THINGS and recipes; we move the licence on request)
on the device, for every phone and computer
Several phones and computers
no - every browser is its own virtual scale, no sync
yes, as many as you like, all at once on the same device
FREE
the free version with 1 THING and 1 recipe - and the full PRO for the first 3 days, by itself
the free version with 3 THINGS and 1 recipe - and 14 days of PRO and ENTERPRISE
PRO
PRO VIRTUAL: 24.90 EUR a year or 89.90 EUR once - one licence for 2 browsers (e.g. phone + laptop)
PRO: from 4.43 EUR a month, 46.90 EUR a year or 119.90 EUR once - per UP↑SCALE device, with as many phones, tablets and computers as you like
none of its own: no USB serial output, no Bluetooth broadcast of the weight, no ESP-NOW devices or machine control, no Decent Scale protocol for coffee machines
the USB serial output for other programs and scale software (live, on request or EXTENDED++, ASCII as HEX or plain text, commands like TARE back), the weight as a Bluetooth broadcast, ESP-NOW devices and machine control, the Decent Scale protocol for coffee machines
Calibration
done by your scale itself
our scales: in the app; an adapter: your scale calibrates itself
Trade use
not verified for trade, like every consumer scale
a verified scale stays verified: EXACT WEIGHT, only key commands over its protective interface (or LISTEN ONLY) - the app itself is no verified display
Thermometers, hygrometers, barometers and espresso pressure sensors of other makers work next to the UP↑SCALE: the app connects them over Bluetooth (or a USB cable) from the same browser - the UP↑SCALE stays connected. Their values fill the fields of the modes; a value you type always wins, and every mode works without a sensor as before. Open ☰ → 🤝 CONNECTION → 📡 SENSORS (or the 🌡 chip in the head of the menu and on the lab pages, or the 📡 SENSORS block of the settings). The window has four tabs under its title - 📡 DEVICES, 🔗 SOURCES, 🌡 ROOM, 📒 REGISTER - and the keys of the tab at its bottom (more behind ⋯); ✕ or Esc closes it. The form of a source and of a register entry open in the same window with ←; leaving one with changes asks first.
Connect:+ SENSOR (tab DEVICES) offers + ROOM SENSOR (air temperature, humidity, air pressure), + THERMOMETER (a liquid: the water of a pipette test, the brew water, the water of the density mode), + PRESSURE SENSOR (espresso), another Bluetooth device … (every device - a DIY sensor with its own name), 📡 broadcast sensors … and 🔌 serial sensor … with its baud rate; then the browser asks for the device. Each device is a card: every value live (right), its role, the battery; ✕ in its head disconnects. A link that drops is tried again by itself.
Roles: every value of a device has a role - room air, liquid, humidity, air pressure, pressure - one device per role. A probe thermometer used as the room thermometer, or the second probe for the air: choose its role in the list; the app keeps it for that device.
Where the values go: a 📡 next to a field takes the value of now, an empty field fills itself, and the records say for every value whether it came from a sensor (which one) or was typed: pipettes and glassware (water, air, pressure and humidity at the start and the end, the room live during the run), the air buoyancy correction, the batch record and the solution report (the room at the start and the end), the routine check, the temperatures of the strict procedure of the tests, the water of the density mode, the environment test, the weighing log (🌡 on every line), the LIMS forwarding (field env), IPC steps in °C, %RH or hPa, and the espresso timer.
PRO: connect and use the live values. ENTERPRISE & COMPLIANCE: the register, the room limits and the room log, and the full source (register entry, certificate, correction) in the records.
📒 Register of the sensors and instruments (ENTERPRISE): for each device its name or asset number, serial number, calibration certificate, its date and the due date, and for each value the expanded uncertainty U (k = 2) and the correction from the certificate (reading + correction = the value used). Also instruments without Bluetooth - a glass thermometer, a hand-held barometer or hygrometer: pick one next to a typed value and the record and the report name it. The uncertainty budget of the pipettes takes U / 2 of the instrument when that is bigger than the lab's own value; a due date that has passed shows ⚠ and counts as a deviation of a pipette or glassware calibration. The thermometer, barometer and hygrometer of the pipette lab settings move into the register once. Only an ADMIN or a QM changes it; every change keeps who, when, the old and the new value and, under the strict audit, the reason, and with an ENTERPRISE UP↑SCALE connected it is also an entry in the activity log of the UP↑SCALE. An entry that is no longer used is retired (⋯ → ⏏ RETIRE in its form, asked first), never deleted; a retired entry is shown read only. Tab REGISTER: a row per entry (the due date right, red ⚠ when it has passed, retired ones gray at the end; a tap opens it), the keys ⋯ (PDF, ⬇ CSV, ⬇ EXPORT, ⬆ IMPORT), 🖨 REPORT and + INSTRUMENT WITHOUT BLUETOOTH. Kept in this browser; ⬇ EXPORT and ⬆ IMPORT move it to another one (the message counts the entries really taken over). Whether a value is traceable depends on the calibration certificate of your sensor.
🌡 Room: the values of now (PRO) - and with ENTERPRISE limits (air temperature, humidity, the change of the air within an hour; presets for pipette labs after ISO 8655 at 20 or 27 °C) with a warning when the room leaves them (once; the chip turns red), and the room log: one mean value a minute, kept in this browser, with the chart of a day (the day in the head of the card), CSV and a report (print / PDF). Tab ROOM: the live line (red over a limit), the cards ⚑ LIMITS (kept in this browser; a value typed by hand turns the preset back to own values) and 📈 ROOM LOG; the keys ⋯ (⬇ CSV), PDF, 🖨 REPORT - with changed limits ⋯ (⬇ CSV, PDF), 🖨 REPORT, 💾 SAVE. It logs only while this app is open and a room sensor is connected - it is no continuous monitoring of the room. A sensor that stops sending leaves a gap in the chart; the room log notes when it stopped (no values since 14:32, a grey dot) and when its values came back. Values are used for 5 minutes (a liquid: 30 s), then they count as none - a record then has no room conditions instead of old ones.
📡 Broadcast sensors: sensors that only send their values in their advertising, without a connection - their coin cell lasts long, the best choice for the room log: BTHome v2 (not encrypted, e.g. Shelly BLU H&T, Xiaomi / Qingping sensors with the pvvx firmware), the pvvx format, Govee H5072 / H5075 / H5101 / H5102 / H5177, RuuviTag. Browsers listen to them only with a test switch: Chrome or Edge, chrome://flags → Experimental Web Platform features on, restart the browser (the app explains it). 📡 broadcast sensors … (in + SENSOR) shows a card above the devices with the known ones and their values: USE IT; ⏹ STOP ends the scan. An unknown one: ⋯ → RECORD FOR US, note what its display showed, ⋯ → SAVE the file and send it to us.
🔌 Serial sensors: a thermometer, hygrometer or barometer on a USB or RS-232 cable that sends lines of text (Chrome or Edge on a computer): choose its baud rate, then 🔌 serial sensor … and its port. Every number with its unit is taken: °C / °F, %RH, hPa / mbar / kPa / inHg / Pa - also labelled lines like T= 21.3 'C RH= 45.2 %RH P= 1013.2 hPa.
🔗 Other sources - the ways labs connect their instruments:MODBUS RTU - temperature, humidity and pressure transmitters on RS-485 (for example Vaisala HMP60 / HMP110, preset included) through a USB-RS485 adapter, Chrome or Edge on a computer; a preset or the register map from the manual of the maker (register, int16 / uint16 / int32 / uint32 / float32, word order). HTTP JSON - the app reads an address every n seconds and takes each value by its path in the JSON (e.g. room.temp); the address must allow the app by CORS (see CORS setup) - for example a small bridge of your lab (n8n, Node-RED, a script) in front of your monitoring system (viewLinc, Saveris and others offer REST or OPC UA). MQTT - the topics of a broker of your lab over WebSockets with TLS (wss://): Home Assistant, Shelly, ESPHome, Zigbee2MQTT; a plain number or the value at its path in a JSON payload. Each value = reading x factor + offset (e.g. Pa -> hPa: factor 0.01). Set up in the tab 🔗 SOURCES of the 📡 window (+ SOURCE, then the kind; kept in this browser): a row per source with its state (red while the link is lost), CONNECT / DISCONNECT and 🗑 at its end (asked first); a tap on the row opens its form. CONNECT makes a source a device like a Bluetooth sensor (roles, register, room log).
Known Bluetooth sensors: RuuviTag (air temperature, humidity and air pressure), Xiaomi LYWSD03MMC, Nordic Thingy:52, Aranet4 (it may ask to pair once with the PIN on its display), every device with the Bluetooth service Environmental Sensing or Health Thermometer (also Xiaomi / Qingping sensors with the ATC / pvvx firmware and DIY boards), ETI BlueTherm (Thermapen Blue, ThermaQ Blue), MEATER, and the espresso pressure sensors (Bookoo Espresso Monitor, Pressensor PRS, ESPROFILE, Popsicle, Transducer Direct). A connected sensor uses its battery faster - disconnect it when the work is done. Written from the published protocols; tell us how it works with yours. The names of these devices and their makers belong to their owners; UP↑SCALE is not affiliated with or endorsed by them.
⚖ Connectivity
This scale can send data and receive commands via USB, BLE and ESP-NOW to ensure connectivity to old and new software as like as to own programs and microcomputers!
All this options you can switch on/off and setup in your App, on the 🛠︎ SCALE SETTINGS page, tab LINK, card USB, BLUETOOTH & ESP-NOW.
Here some more detailed info about the possibilities and data formats for...
You can control the scale from your mobile/computer via BLE with your app.
Normally its only allowed to be connected with one Device at the same time, but you also can use multiple devices at the same time if you need so!
Switch on the Multiple masters option to do so, but take care in public with that!
Off, it counts the app on the USB cable too: one app in all. While a Bluetooth app is connected, 🔌 USB cable is refused with a note, and while the cable app is connected, the scale is not visible for Bluetooth apps. (With the Decent Scale protocol on: the machine and one app, as over Bluetooth.)
The scale can also speak the public wireless protocol of the Decent Scale. Machines and apps that support the Decent Scale can then use your scale, for example Beanconqueror, GaggiMate (brew by weight) and the Decent DE1 app. The option is off by default; switch it on in the settings card USB, BLUETOOTH & ESP-NOW (tab LINK, row "Decent Scale mode") and save. The scale restarts.
The name
These machines and apps only find scales whose name begins with "Decent Scale", so the scale is called "Decent Scale UP-420 #12345" then (with your scale type and number). Our app finds it under this name too: connect it again after the restart. Everything else works as always, also with an UP↑SCALE app connected at the same time.
Weight & tare
The machine or app gets the weight 10 times per second (stable or moving), in 0.1 g steps (a 710 sends 0.1 g too) up to ±3276 g, and can tare the scale.
Shot timer
When the machine or app starts its timer, the scale switches from the weighing page to the Espresso Timer and records the shot - without its own tare, the machine tares by itself. Its timer stop ends the shot (the auto-stop waits for it), and our app saves it in the log ("Started and stopped by the machine/app"). The coffee timer shows if a machine or app is connected.
Login
Machines and apps can't log in as a user: on a scale with users they get nothing until an UP↑SCALE app has logged in (until all connections are gone again).
Limits
Only weight, tare and timer: battery is always reported as USB power, LED and power-off commands are ignored, and the connection is never dropped (no heartbeat needed).
Decent, Decent Scale and DE1 are names of Decent Espresso International Ltd. UP↑SCALE is not affiliated with or endorsed by Decent Espresso; the scale is only compatible with the published Decent Scale protocol.
Wireless for all
If this option is on, the scale will send weight data continuously to everybody! This is working like a FindMe Device, the data will be packed in the public readable manufacturer data part where it can get read out from everybody nearby! This can be super-useful if you want to read out this data from multiple devices without connecting/pairing first! This will allow some additional options, like showing the weight of multiple scales together at the same time - or you could read it out with your smart-watch or integrate it into your home automation systems on that way! The data by self will be transmitted as ASCII in HEX - check out all info about the exact format below!
While your app is connected and no further app may connect (MULTIPLE MASTERS off), the scale keeps sending the weight to everybody - but nobody else can connect then. Switch the option off and the weight leaves the air at once. For a watch or a widget that should also send TARE, use the WATCH LINK.
Example Code for Arduino / ESPs
We've prepared a small and simple example "sketch" for Arduino.
This will scan for scales around you and show the live weight on it - without being connected/paired to a scale first!
Check this out to learn how to scan for the scales and how to read in and use the weight data.
Click here for the ARDUINO BLE WEIGHT MONITOR, copy, paste and there you go :)
PS.: If you also want to get more (or faster) data, and if you also want to be able to send commands to the scale, check out the ESP-NOW options for ESP Microcomputers instead!
Html/JavaScript Example Code
Same is working directly in a Web-Browser! We've prepared an example just with html and javascript which is reading out - and summing up - all scales around you!
Click here to check it out (and to test it live and direct).
Just copy the code from there and build something own on it :)
⌚ WATCH LINK
⭐ PRO - gray in the free version; try it with the free trial.
A small, open door of the scale for watch and widget apps, displays and your own devices: they connect over Bluetooth and get the live weight as the scale shows it, the state of the running program (TOO LESS, OK, PERFECT, TOO MUCH) and its signals (a pack is counted, a mold is full, an alarm) - and they can press TARE, OK and RESET, like the ESP-NOW devices. Switch it on in the settings card USB, BLUETOOTH & ESP-NOW (tab LINK).
Open on purpose: the WATCH LINK is not encrypted, also with the Encrypted Link on - like WIRELESS for all, you switch it on only if you want it. Everybody nearby can read its weight.
The buttons only on an open scale: with a password or users, TARE, OK and RESET of a watch only work while somebody is logged in with the app (a watch can not log in). A TARE of a watch is a tare of a person: the tare guard waits for a still weight.
One more connection: with MULTIPLE MASTERS off, one more device may connect next to your app. Note: that place can also take a second app.
Switching it on or off restarts the scale - the scale only has the WATCH LINK while it is on.
An app on the watch is needed: no watch or phone shows it by itself. The packets are open (below), so anybody can write one - for Wear OS, a Garmin watch (the packets fit its limit of 20 bytes), an ESP32 display or a phone widget.
For developers: the packets
Service 00000010-1312-420b-b710-d90000000420, one characteristic 00000011-1312-420b-b710-d90000000420 (notify, read, write). Switch the notifications on; numbers are little endian. A watch finds the scale by its name, which begins with UP↑SCALE (.
Packet
Bytes
LIVE (on every change, max. every 100 ms, and every second)
[0] 0xA1 | [1] flags: 1 stable, 2 net (tared), 4 center of zero, 8 overload, 16 taring, 32 a tare was refused, 64 no valid weight (not a weighing page or no value from the RS-232 scale), 128 an alarm is open | [2] mode × 16 + state | [3] the number of the last SIGNAL | [4-7] float: the value as shown | [8-16] the value as text (9 characters, right-aligned) | [17-19] the unit (3 characters, e.g. g, PC)
0 none, 1 TOO LESS, 2 OK (low), 3 PERFECT, 4 OK (high), 5 TOO MUCH, 6 waiting / take it off
SIGNAL (once, when it happens)
[0] 0xA2 | [1] its number | [2] kind (+ 128: an alarm - vibrate) | [3] detail | [4-7] float | [8-19] a text (UTF-8)
kinds
1 check-weighing: a pack is counted (detail: its state, float: its weight) · 2 factory: 1 a cavity filled, 2 a mold full, 3 the filling stopped · 3 sorting: an item is counted (detail: the category, text: its label) · 4 tracking alarm (detail: which alarm) · 5 coffee alarm · 6 recipe: 1 an ingredient done, 2 the recipe done · 7 fill weighing: 4 a fill passed, 16 the run ended · 8 overload · 9 a tare was refused: 0 a key, 1 a program's tare (held), 2 a step, 3 a ZERO with a load on the scale · 10 no value from the RS-232 scale · 11 a button was refused (closed scale)
Buttons (write)
the first byte: t TARE, o OK, r RESET (on every page as for the ESP-NOW devices), zZERO (load cells), a the alarm is seen (clears flag 128; a tracking alarm is acknowledged as in the app)
USB Connectivity
🔌 The App on the USB Cable
On a computer the app can reach your scale or adapter over its USB cable instead of Bluetooth - with everything it does over Bluetooth: weighing, all modes, settings, THINGS, recipes, sales, users and logins. Plug the scale or the adapter into the computer (the same cable powers it), open the app in Chrome, Edge, Opera or Firefox (from version 151) and press CONNECT in the row 🔌 USB cable on the start page, then choose the port of the scale. It needs no Bluetooth on the computer - also for computers without it. Bluetooth apps (phones, tablets) can stay connected at the same time; changes reach all of them. The cable link is not encrypted (the encrypted link is for the radio): whoever has the cable in hand has the scale in hand. While the app uses the cable, the scale sends no serial weight lines on it; they come back when the app is closed. Phones and tablets: iPhone and iPad cannot use this. Chrome on Android shows the button from version 148, but lists a USB port only on devices whose Android (17 and newer) passes it on - on most phones the list stays empty (not tested by us yet).
Bluetooth Power
The transmit power of the scale, in the settings card USB, BLUETOOTH & ESP-NOW (tab LINK). The standard is +9 dBm: what most Bluetooth devices use, enough for a scale on your desk; +15 dBm gives a steadier connection through a wall or across a room (the heartbeat lock likes a steady link), +18 dBm (maximum) reaches farthest, for long distances or many walls. Higher means more range and a more stable connection, but also a bit more current: a scale on a powerbank or battery runs a little shorter. Lower means less current, but less range and more gaps. Go down to +6 dBm if the scale stands right next to your phone or computer or you want less radio around you. The lowest choice is +6 dBm: below that the connection already breaks up at 2-3 metres (measured: 0 dBm gave about 12 dB less signal than +15 dBm). The battery of your phone doesn't change. The scale also asks every phone or computer for a steady link (a connection event every 15-30 ms, and a short gap of up to 4 seconds is no disconnect - the app still locks the weight after 1.25 seconds without data).
Encrypted Link
On PRO and ENTERPRISE scales the scale and the app encrypt what they send each other over Bluetooth: the weight packets, the settings, the sales, everything on the three data channels of the scale (AES-128 with a 128-bit tag). For every connection the scale makes a one-time key, signs it with its own key, and agrees a new key with the app by ECDH P-256; the app checks that signature, so nobody can sit in between. A listener sees only random bytes. Because the one-time key is thrown away, a recording of a connection stays closed even if somebody reads out the scale later. It is on by default; the switch is in the settings card USB, BLUETOOTH & ESP-NOW (tab LINK) and is greyed out on a free scale.
WIRELESS for all is off on a new scale and goes off when you switch the encrypted link on, because it sends the weight openly to every device nearby. You can switch it on again if you need it.
Not covered by the encryption: the weight that WIRELESS for all sends on purpose (if you switch it on), the Decent Scale service for coffee machines and apps (switch it off in the settings if you do not use it), and USB serial (ESP-NOW has an encryption of its own). Your own Bluetooth programs (like the example sketches for developers) cannot read an encrypted link: switch it off for them.
The key of your scale: on an ENTERPRISE scale with a seal certificate the app checks the key against the certificate from UP↑SCALE. Otherwise the app remembers the key at the first connection and asks you if it ever changes - say no if you did not reset the scale.
Registered apps only (ENTERPRISE & COMPLIANCE, off by default, in the same box): only the browsers and phones you registered can connect. Each one keeps its own key that cannot be copied out of the browser; the scale keeps up to 8.
On every device you want to use: REGISTER THIS APP (with users: logged in as an ADMIN).
Switch Registered apps only to ON (only a registered device can do that).
A new device later: on a registered device tap LET A NEW APP IN (2 MIN.), then register the new device within 2 minutes.
All registered devices lost: connect the scale by USB, open a serial terminal (115200 baud) and send LINKOPEN. That switches it off.
Users and passwords stay as they are; this is one more lock in front of them. Every change is written to the activity log.
The limits: a sniffer still sees that a connection exists and how big and how often the packets are (so it can tell that the weight changes). A connection that sends 16 packets that do not open is dropped. The key of the scale sits in its memory without a secure chip: somebody with the scale in their hands and the right tools can copy it and then pose as the scale for new connections. We say "encrypted", not "certified". All of it in plain words, with the algorithms for your IT: Security.
USB Serial Output
If you switch on the USB SERIAL OUTPUT at your scale settings, then you are able to send live data from the scale via the USB port - and also to receive commands (like TARE) via USB! For the USB output you have several different options:
The weight will be sent as ASCII in HEX (or plain text), as soon as requested via a command from the user. (S for the next stable- or Q for direct output.)
Multiple data-points will be sent continuously as human readable text!
Configuration
We are sending with: 115200 Bits/s; 8 Data Bits; NO PARITY; ONE STOP BIT
LIVE- & ON REQUEST MODE:
In this 2 modes, the data will be delivered as ASCII in HEX! This is for the old-school computers and scale software. Optionally you also can switch to a plain text ASCII output! Check out the EXTENDED++ Mode for exhaustive outputs as readable data for modern data-loggers etc.!
Request Data (only REQUEST MODE)
In bidirectional command mode (ON REQUEST), the weight can be immediately retrieved by sending the command "Q" followed by Carriage Return and Line Feed to the scale. The next stable weight can be requested with "S" and will be received once after stabilization in this case. If you send "T" you can tare the scale.
Nr.:
1
2
3
1
2
3
1
2
3
ASCII:
Q
CR
LF
S
CR
LF
T
CR
LF
Hex:
51
0D
0A
53
0D
0A
54
0D
0A
Serial Data Formats:
The Scale always sends 19 bytes for every measurement. This are ASCII characters as HEX (or optionally as plain ASCII characters as decimals) in the following format:
Nr.:
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
ASCII:
U
S
,
-
0
0
0
0
4
2
5
.
9
8
g
CR
LF
HEX:
55
53
2C
2D
30
30
30
30
34
32
35
2E
39
38
20
20
67
0D
0A
DEC:
85
83
44
45
48
48
48
48
52
50
53
46
57
56
32
32
103
13
10
The first 2 characters are the header
1
2
S
T
Stable Weight (Weighing Mode)
U
S
Unstable Weight (Weighing Mode)
S
Q
Stable Quantity (Counting Mode)
U
Q
Unstable Quantity (Counting Mode)
The header is separated from the measured value by a comma.
The measured value consists of 11 characters and always begins with a sign (+/-; + is used for a value of zero). The measured value is transmitted right-aligned, unused positions are filled with zeros. The decimal separator is a point by default.
After the measured value, 3 characters for the unit are transmitted. Also right-aligned, with spaces for unused characters. The last 2 characters are always a CR/LF (Carriage Return/Line Feed) and indicate the end-of-line.
Example Output:
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
S
T
,
+
0
0
0
0
2
3
.
4
2
0
o
z
CR
LF
23.420 oz (stable weight)
S
Q
,
+
0
0
0
0
0
0
0
4
2
0
P
C
CR
LF
420 pcs (stable count)
U
S
,
-
0
0
0
0
0
0
3
.
9
1
g
CR
LF
-3.91 g (unstable weight)
We've also prepared some code examples for html/javascript and Arduino which will read in this data wireless! Use this scripts to see how to receive and process this data via BLE!
EXTENDED++ MODE:
This mode will blast out weight data 40 times each second via the USB Serial Port (1 line each 25 milliseconds)!
Its possible to select between- and show up to 7 different weight-data-points from the scale at the same time, starting from the raw sensor data, via the unfiltered and filtered averages, up to the final weight on the display! (Also a raw gross weight is available!). Additionally/Optionally you can add a 3-line-summery every second about the amount of grabbed data-points within the last second, including their deviations! Otherwise just leave this untouched.
The scale only sends this output if a program on the computer reads the USB port. Without one (or with the port closed), the lines are skipped, so the scale never waits for the computer.
Use this mode to connect to data-loggers etc. to output the weight as live graph or so, or to monitor the scale with a serial monitor - for example to fine-tune it with the extended scale settings. Or, if you want, build your own software with our scale! Btw.: This data will NOT come as HEX, like the simple output, but directly as human-readable ASCII text!
Example Output for 1 sec.
ESP-NOW Connectivity
⭐ PRO: ESP-NOW devices (a free scale keeps ESP-NOW off; the saved setting comes back with PRO; free on the UPCENT↑420 for its pump relay) - gray in the free version; try it with the free trial (free and PRO).
ESP-Now Peripherals
You are able to connect up to 5 devices via ESP-NOW.
This standard will transmit data by 2.4 GHz radio, but without the need of internet, a router or a login.
Add the MAC address of the device to the scale once, paste its key into the sketch of the device - and everything works as soon as both are powered.
The page: tab LINK of the settings → 📡 ESP-NOW devices ▸ (← back to the settings). THIS SCALE shows the MAC address of your scale, BASIC how often it receives and broadcasts (or OFF), the radio channel and the inactivity stop, Connected Devices the MAC addresses of up to 5 devices: 6 fields of 2 characters each (0-9, A-F) - or all 6 empty for no device. A device whose MAC address is saved gets 🔑 KEY (shows the line for its sketch: 📋 COPY, HIDE) and ↻ NEW KEY (asks first). 💾 SAVE checks every address first - a MAC address that is not complete turns red and is not dropped - then it asks and the scale restarts.
Connected devices will get live data from the scale (5, 10 or 20 times a second)
Connected devices also can send commands (tare, ok, reset). The scale only accepts commands from devices whose MAC address you have added - and only when they are encrypted with the key of that device.
5 devices can be connected at the same time!
🔒 Encrypted: every device has its own key
Every ESP-NOW packet between the scale and a device is encrypted and protected against changes (AES-128-GCM) - always, there is no open mode. Nobody nearby can read your weights, and nobody can send the scale a TARE, OK or RESET without the key.
The key: when you save the MAC address of a device, the scale makes a random key for it (128 bits). On the ESP-NOW settings page, 🔑 KEY below the device shows it as one line for the sketch: const uint8_t UPSCALE_KEY[16] = {...}; - 📋 COPY it and paste it into the sketch of that device in place of the UPSCALE_KEY line, then upload the sketch. The app never saves the key.
The key never goes over ESP-NOW. The app gets it only over the encrypted Bluetooth link (PRO). Without it (e.g. the UPCENT↑420 in the free version): connect the scale by USB, open a serial monitor (115200 baud) and type NOWKEY - the scale prints the lines of all its devices.
Its own key for every device: a lost or stolen device opens nothing else. ↻ NEW KEY makes a new one (the old one stops working at once - paste the new line and upload again). A new MAC address in a slot gets a new key too. With users, only who may change the settings sees the keys.
No replays: every packet carries a counter that only goes up, also over restarts. A recorded packet sent again later (an old TARE, an old "fill now") is refused.
Limits: the key is kept unencrypted in the flash memory of the scale and of your device - keep the device where nobody can read it out. Anybody nearby can still see that packets are sent (how big, how often), and nobody can stop a radio jammer: a filling machine should stop when no data comes (the example sketch closes the valve after 500 ms).
To connect ESP-NOW devices, which could be our peripherals or your own microcomputers, simply switch on the ESP-NOW options at your scale settings page.
Select a send interval and the radio channel, add at least one MAC address - and ready you are!
Check out the following example code if you want to program your own device!
Example Code
If you like to create your own device based on an ESP32 or ESP8266 microcomputer, you're welcome! Receive live data from the scale and simply show it on a display - and/or transmit it to some other devices, monitors, relays, home-automation-systems or to your website! Here you can find a small example sketch, the [UP↑SCALE MasterBlaster], made for an ESP32 at the Arduino IDE. This small script receives live data and will send commands on button clicks. Check out how (and what kind of) data is incoming from the scale, and how simple it is, to send commands back. Use this as inspiration and example/base for your own programs :)
Example: ESP-NOW in Sorting Mode
While a sorting program is running, the scale sends everything you need to let other machines react on the sorted items:
Current weight and the category of the item on the scale (number, label, its count and total weight). Lighter than the first category is category 0 (UNDER), heavier than the last one is the number after the last category (OVER).
Status (like in factory mode): weighing (item on the scale, not counted yet), counted (remove the item now) and taring. You get it as three flags and as one status number.
Ping: in the moment an item gets counted, the scale sends one extra packet with the ping flag set, the counted category (number and label) and the counted weight. Use this signal to trigger an other machine, a gate, a conveyor or a light. A ping number goes up with every count, so your device can find out if it has missed one.
Counted amounts: total items and total weight, plus the whole category table: one row (label, max. weight, count, total weight, incl. UNDER and OVER) in every packet, row by row.
Active ingredient (if the program has one): title, unit and the amount of the weight on the scale, of all counted items and of the last counted item.
Commands: TARE tares the scale, OK counts the item on the scale now and RESET resets all counts of the running program.
Example: ESP-NOW in Tracking Mode
During a tracking run every packet contains its state (0 no run, 1 recording, 2 stopped, 3 waiting for the start), the amount of points, the last point and the start weight, the target and the interval - and the alarms (1 target reached, 2 constant weight, 4 jump, 8 restart, 16 peak, 32 above / 64 below the average band) until they are seen in the app. The packets go out 5 to 20 times a second, also while the app shows another page, so a receiver sees every new point (the amount goes up by one, the last point is the new one) - even at 0.25 s per point. A new alarm comes at once in an extra packet (with a number that goes up with every alarm). The example code switches its ALARM_PIN on while there is an alarm.
All weights are sent in the weight unit selected on the scale. The MasterBlaster example sketch contains the complete data structure and shows the ping on the serial port. (The sorting data is added at the end of the data structure, so your own devices stay compatible.)
Example: ESP-NOW in Factory Mode (filling machines)
With the predictive cut-off switched on, the scale controls a filling machine: fill_run is true while the machine should fill, fill_stop comes early by flow rate × lag (in an extra packet at once), plus a stop number, the flow rate per second and the lag. With the coarse / fine dosing set, fill_slow is true from the slow-down point until the next cavity/container (also in an extra packet at once). fill_state tells the step of the program (0 none, 1 START - put a mold on, 2 filling, 3 slow, 4 stopped, 5 pause, 6 mold full) - the MasterBlaster example prints every change. (Also added at the end of the data structure.)
Example: ESP-NOW in the Espresso Timer (pump relay)
With the early stop switched on, the same values control the pump of an espresso machine: coffee_shot is true (the fill_ values belong to the shot), coffee_target is the target yield, fill_run is true with the cup on the scale and while the shot runs, and fill_stop is STOP NOW (until the next shot). The MasterBlaster example switches the relay on FILL_VALVE_PIN with them. Lever or switch machines: switch the brew switch off after the shot. (Added at the end of the data structure.)
Energy Saving Mode
If there are no connected devices around you (any more), then the radio module of your scale will automatically shut down to save energy. In that case you would need to restart the scale to start transmitting ESP-NOW data again (just unplug and plug the scale once). You can set up this standby/shutdown time by self (standard is 5 minutes) or you can switch to "Always ON", if you don't care at all!
⚖ UP↑SCALE DIY on the ESP32-C3 RS232 Adapter
Our RS-232 firmware also runs on the small ESP32-C3 RS232 Adapter (sold by TinyTronics): buy the board, install our free firmware, try it - and unlock PRO with a licence. More on the UP↑SCALE DIY page.
First install
You need a computer (Windows, macOS, Linux, Chromebook) with Chrome, Edge or Opera and a USB-C data cable (many charger cables carry no data). No driver, no program: it runs in the browser.
Open the update page (UP↑SCALE DIY) - the DIY firmware starts with the pins of this adapter, nothing to choose - and click INSTALL at the newest version (on top).
Choose the port of the board in the window of the browser (the one that appears when you plug it in, often "USB JTAG/serial debug unit", a COM port on Windows, usbmodem/ttyACM on macOS/Linux) → Connect.
No port, or "failed to connect"? Download mode: hold BOOT, press and release RESET (RST), release BOOT - then INSTALL again.
"Erase device?" - on a new board you may tick it. Next → Install, wait for "Installation complete" (about 1 minute, don't unplug).
Press RESET once (or unplug and plug in the cable) so the firmware starts.
Open the app, CONNECT, choose "UP↑SCALE (DIY) #..." - the free version runs. Then connect it to your scale.
Problems: no port → another (data) cable or USB port, then download mode; on Linux add your user to the group dialout (sudo usermod -a -G dialout $USER, log out and in). "Port busy" → close other programs that use it (Arduino IDE, serial monitor).
Activate PRO
In the app (adapter connected): tap the licence in the head of the ☰ menu (💎 licences & prices) or open ✰ SCALE INFO → both show the Device ID.
ORDER at a PRO row (a month, a year or for ever; on the licence page or in the card 🛒 BUY of the scale info) opens the order form with the Device ID filled in. After the payment you get a link and a QR code by email - after a card payment at once, also on the page.
With the adapter connected: open the link on the same device, or scan the QR code with 📷 SCAN (✰ SCALE INFO, card 🔑 LICENCE, row 🔑 Licence key), or PASTE KEY (⋯ next to it). The adapter checks the key and switches to PRO - forever.
Updates
The app tells you when a new firmware is out ("UPDATE YOUR SCALE"). Updates are free.
Board by USB-C to the computer → update page → INSTALL at the newest version → choose the port (download mode is usually not needed).
"Erase device?" - do NOT tick it: an erase deletes the RS-232 settings, THINGS and recipes on the board. (The licence comes back: the app sends it again, or offers to look it up online - see below.)
"Installation complete" → press RESET → connect in the app.
Connect it to your scale
The adapter has a DB9 male plug wired like a PC (pin 2 RX, pin 3 TX, pin 5 GND; RTS on pin 7 is switched on). Plug it in where the PC cable of your scale would go - with the scale's own RS-232/PC cable or a gender changer.
Power: USB-C (charger or power bank) - it gets no power from the scale.
No signal at all (AUTO-DETECT says so): switch on the interface of your scale (continuous output, or send with the print key). If it is on, the wires are crossed the wrong way: use a null-modem adapter (or a straight cable instead of a null-modem cable) - the direction is fixed in the RS-232 chip of the board and can not be switched in the firmware.
Data, but unreadable: choose the preset of your scale, or set baud rate, data bits and parity by hand.
UP↑SCALE is not affiliated with TinyTronics or LEOTRO. Product names belong to their owners.
⚖ UP↑SCALE DIY: build your own RS-232 adapter
For makers: the same RS-232 firmware runs on almost every ESP32-S3 or ESP32-C3 board with at least 4 MB flash, together with a 3.3 V RS-232 board. Our recommended set: a Seeed XIAO ESP32S3 (pre-soldered) + the Waveshare RS232 Board + 4 female-female jumper wires - about 15 EUR of parts, no soldering. Shopping list with shops, wiring and help: UP↑SCALE DIY page. We support the firmware and the app, not the hardware you build.
Install: board by USB-C to the computer → update page for the DIY → INSTALL: the installer picks the S3 or C3 version by itself. Then connect "UP↑SCALE (DIY) #..." in the app.
Pins: a XIAO ESP32S3 wired like on the DIY page works at once. XIAO ESP32C3 and other boards: BOARD PINS in the RS-232 settings.
Connect to the scale: the Waveshare board has a DB9 male plug wired like a device (pin 2 sends, pin 3 receives - the other way round than the ESP32-C3 RS232 Adapter): with the scale's PC cable you usually need a null-modem adapter. Then set it up like every RS-232 scale (preset or AUTO-DETECT).
PRO: the PRO licence (from 4.43 EUR a month, 46.90 EUR a year or 119.90 EUR once, incl. VAT) is made for the Device ID of your board, exactly like on the ESP32-C3 RS232 Adapter.
ENTERPRISE & COMPLIANCE: the third level (from 13.00 EUR a month, 144.00 EUR a year or 599.00 EUR once, incl. VAT, includes everything of PRO; 14 days free to try) adds the sealed, checkable test reports (PDF/HTML) and the final report of the scale test suite, the equipment register with test history, users with ADMIN, login and an activity log, report extras (company header, report number, integrity code) and the table exports in your format, sealed by the scale (plain reports, your logo and plain CSV exports are already in PRO). Order it with the Device ID of the same board at the order form (see month, year or once). The tests themselves and their results on the screen stay free for everybody. Buttons of this level look greyed out without it; a click tells you how to get it.
UP↑SCALE is not affiliated with Seeed Studio, Adafruit, SparkFun or Waveshare. Product names belong to their owners.
⚖ Legal for Trade: your verified scale + the 232 UP↑SCALER
You sell by weight or pack ℮ prepackages? Then you need a verified (legal for trade) scale - and you probably have one already. Our 420 and 710 are not verified. But plug the 232 UP↑SCALER (our own box is a prototype - the same firmware runs on the ESP32-C3 RS232 Adapter) into the RS-232 port of your verified scale and it gets the whole app - with the weight of your verified scale.
Set it up in 5 steps
Connect the 232 UP↑SCALER to the RS-232 port of your scale (null-modem cable) and power it with USB-C.
In the menu of your scale: set the serial output to continuous (also called stream, auto or always send) and, if you can, the weight unit to grams.
In the app: ☰ → 🛠︎ SCALE SETTINGS → 🔌 RS-232 SETUP (SCALE & CONNECTION). Select the maker of your scale (or "RS232 WEIGHING SCALE"), set capacity and precision like on the label of your scale, or press 🔍 AUTO-DETECT (with "RS232 WEIGHING SCALE" it finds the type, baud rate, data bits, unit and d).
Switch on EXACT WEIGHT (the app shows exactly the weight and the stable mark of your scale, without our filters) - the dongle then sends only the key commands of your scale (weight request, tare, print) over its protective interface (rückwirkungsfreie Schnittstelle). Only if your scale has none, also switch on LISTEN ONLY (the dongle never sends anything to your scale). 💾 SAVE (at the right end of the key row).
Check: put a weight on the scale - the app must show the same value as the display of your scale.
What you get
USA package check along NIST Handbook 133 - packs labeled by weight, count or volume (MAV, sample plans, moisture allowances like cannabis 3 %, the tare sample) with a lot report.
℮ prepackage control (EU 76/211/EEC, German FPackV and OIML R 87): mean, T1 and T2, the sample plan for your lot size and a printable lot report with your company header - on a 232 with EXACT WEIGHT it states: "Weighed on a verified scale in exact-weight mode" and whether the dongle sent only key commands or nothing (listen only) - the line an inspector wants to read.
Statistics with run chart, histogram and Cp/Cpk, check-weighing with tolerances, counting (also with a second scale for the piece weight), sorting, factory filling with predictive cut-off.
Recipes with batch reports and lot numbers, the inventory and the sales log, labels and label printers, the log - and everything else of the app, on your phone, tablet or laptop.
🔏 LEGAL MODE (ENTERPRISE): an alibi memory for balances that have none - every used value is stored on the dongle first, with the line of the balance and its own number (UP-000123), sealed and checkable.
The interfaces that scale makers sell for their verified scales (USB, network, wireless) often cost a lot and come without software - the 232 UP↑SCALER comes with the app.
What stays legal - and what you have to check
Your scale stays as it is: the dongle sends only what the keys of the scale do (weight request, tare, print) over its protective interface - almost every verified scale has one (check the manual of your scale); never an adjustment or a setting. A scale without one: with LISTEN ONLY the dongle only reads, it sends no requests and no tare signals.
The display of the verified scale stays the legal reading. The app shows its exact weight (minus your software tare), but the app itself is not verified. When you tare in the app (or use a container) and Net/Gross is switched on in the basic settings (off by default), the line below the weight shows "NET: tare ... gross ..." as soon as more than 2 d are tared (also on a load-cell scale, however small the tare) - the tare and the gross weight (what is really on the plate - on an RS-232 scale that is what its own display shows). So nobody mixes up the two numbers. A virtual weight is named on its own ("virtual weight included"): it is no weighing result. With EXACT WEIGHT the tare is the last value of your scale, so the net weight stays on the steps of your scale.
℮ packs: the label shows the nominal quantity (like "500 g ℮"), the filling is checked with your verified scale and the app keeps your records. Inspectors still check their own samples.
Selling at the counter (shop, market stall, club, dispensary - the customer is there): in the EU every display and every printer for the seller and the customer must be verified together with the scale (directive 2014/31/EU, Annex I). An alibi memory alone does not change that: in Germany an extra, unverified display of weight and price is only allowed if the verified scale itself prints or stores the weight, the unit price and the price unchanged and the buyer can see them (a price-computing verified scale, MessEV § 4) - ask your office. So the customer must see the display of the verified scale, and the weight or price on the bill comes from the verified scale or your verified scale system - not from a label, receipt or screen of the app. For these sales switch off ⚖ LIVE WEIGHT AND PRICE ON THE LABEL in the label setup (your choice). The indicator view is no verified customer display.
Labels with a measured weight or price away from the counter (variable-weight packs, price tags, delivery notes): these are legally relevant too. PC software can be allowed if a verified alibi memory keeps every result and its number is on the label (EXTRA LINE). The German verification offices also ask for a note on the label or receipt that the value can be traced in the alibi memory, records kept for at least 3 months, and a person on site who can show them - ask your verification office (in Germany the Eichamt) before you use it. For a verified scale without an alibi memory, 🔏 LEGAL MODE is planned for this too, with our 232 UP↑SCALER.
Pharmacies and the making of medicines (Germany): a device or software that documents the weighings of a verified balance - like the batch record of the compounding assistant - is only allowed if it has a certificate of a conformity assessment body, or if the balance (or a verified device of it) keeps every value unchanged in an alibi memory or prints it on an alibi printer, and every value in the record can be traced to it. This works today with a balance that has an alibi memory: switch on EXTRA LINE and switch on ALIBI ON TAKE (or press its print key before you take each ingredient) - the batch record then shows the alibi number below every ingredient, so every value is traced to the alibi memory, which stays the legal record. For a balance without an alibi memory, 🔏 LEGAL MODE stores every value on the UP↑SCALE with its own number - its certificate for this use is planned for our 232 UP↑SCALER, so until then ask your verification office.
No problem: labels without a measured weight (the nominal quantity "500 g ℮", lot numbers, dates, THC / CBD, ingredients) and all your own records.
Connecting can end a verification: verification offices warn that a cash system connected to a scale after its verification can end the verification. The key commands of a protective interface (tare, print) do no more than the keys of the scale, and LISTEN ONLY sends nothing at all - still, ask your office before you use the setup in trade.
USA: a scale in trade needs an NTEP certificate (or a state type approval) and is registered and inspected by the weights and measures office of your state or county; dispensaries mostly need accuracy class II. The app and the adapter have no NTEP certificate - the scale and its display stay the device for the sale. The weights in the Metrc files of the DISPENSARY 🏪 MODE come from that NTEP scale; the app warns about weights typed in by hand or weighed without NTEP data.
Check the rules of your country. This is our help for your own work, not a legal certificate. Not sure? Ask your verification office yourself - here is an example text of what the adapter does.
⚖ UPCENT↑420 (experimental)
You have a Half Decent Scale? Try running it with our firmware and app: you get all the options of an UP↑SCALE on your coffee scale - counting, recipes with baker's percentages, the nutrition diary, THINGS with units and prices, check-weighing, labels, statistics, the coffee timer, tracking and much more. And your coffee machine or coffee app still works with it. UPCENT↑420 is our name for the UP↑SCALE firmware on a Half Decent Scale. It is free; the app then shows a small donation note at the bottom of the screen.
Still experimental: it is new and not tested on many scales yet. Please write us if something does not work, or if you need an extra option!
What exactly happens
The install takes over the full hardware of your scale (sensor, display, buttons, battery), installs a new operating system (our firmware) and deletes everything that was on the chip before: the Decent firmware, its settings and its calibration.
Make a backup first - if you can ;-) Write down your settings of the Decent firmware; nothing of it can be taken over.
Afterwards the scale works only with our app, not with its original software anymore. (To go back, you would have to flash the Decent firmware again yourself with their updater.)
Coffee machines and apps still work: the Decent Scale protocol is on by default, so the DE1, GaggiMate and Beanconqueror find the scale as "Decent Scale UP-HDS #...".
Install
Open the update page on a computer (Chrome, Edge or Opera) and choose UPCENT↑420.
Connect the scale with USB-C. No button is needed: the scale shows up as "USB Serial", "wchusbserial" or "Half Decent Scale" (on a Mac you may need the CH34x USB driver).
Click INSTALL. If the scale does not restart afterwards, press the reset hole under the scale with a paperclip.
Connect the app: pick "Decent Scale UP-HDS #..." in the list of the browser. The app asks for the PCB version of your scale: choose it in the settings (tab LINK, card BOARD HARDWARE, "Scale hardware": V8 (auto), V8.0 or V8.1 - only V8 boards so far) and 💾 SAVE.
Calibrate the scale with a known weight: the app opens the calibration by itself as long as the scale is not calibrated (see Calibration).
Display, buttons & battery
Display
The weight (or the count in the counting mode) in the unit of the scale, "STABLE", the running time of the coffee timer, the connection (BLE, +DECENT when a machine/app gets the weight) and the battery.
○ button
Click: tare. Hold (1.5 s): switch off.
□ button
Click on the weighing page: start the coffee timer; in the coffee modes: start/stop/OK; in other modes: OK. Hold: back to the weighing page.
Display & buzzer
In the settings: "Display" turned (like the Decent firmware) or standard, if the weight is upside down. "Buzzer" (V8.0 boards): off by default, on, or on (inverted) - if it beeps all the time or never.
Switch on
Hold a button for half a second (a short touch is ignored, so it does not switch on by accident in the bag).
Battery & auto-off
The app shows the battery in the head of the page and in the menu (tap it for the voltage) and in the settings (card BOARD HARDWARE: "Battery", or "USB: charging"). The scale switches itself off after the "Auto-off" time of the settings (5 - 120 minutes or never; 10 by default) without weight changes, clicks or a connected app/machine - never while charging. At an empty battery it switches off too. Switch it on with a button.
Decent, Decent Scale, Half Decent Scale and DE1 are names of Decent Espresso International Ltd. UP↑SCALE is not affiliated with or endorsed by Decent Espresso. The firmware is our own code for this open hardware.
Start with the free version - the board costs under 15 EUR and you may never need more. If you do, unlock PRO from 4.43 EUR a month, 46.90 EUR a year or once for 119.90 EUR incl. VAT - or ENTERPRISE & COMPLIANCE from 13.00 EUR a month (once: 599.00 EUR). The lifetime licences are once and forever, no subscription needed; no advertisements, everything works offline, updates stay free.