I’m waiting on the load cells to arrive, so just did a simple experiment on the OpenScale, and it has me perplexed. Thought one of you may be able to help:
I put two series 1K resistors between E+ / E- , then tied their midpoint to A+ / A- . This sims the loadcell’s outputs being perfectly balanced with no load.
But OpenScale is outputting < -8000 lbs. This makes no sense to me. I’m using the default values in OpenScale (reset them just to be sure). Listed below, you can see that there is a huge tare value as a default. Is this reasonable?
So I did a ‘Tare scale to zero’ and it decided that the tare value should be -1853, and then started reporting about -1.9 lbs… why not 0 ?
Does any of this make sense? Seems most of the forum commentary is people having problems. Starting to think the OpenScale’s firmware has issues…
Since I (somewhat) slandered OpenScale’s firmware, I thought I’d repost a retraction.
I mostly figured out my problems and now have it working quite well. Initially when I put the same voltage in both S+/S-, I expected a near-zero raw value, so was surprised at the large raw number. Turns out it was actually fairly close to 0, but having a 24-bit ADC makes small differences have big numbers still ! The number was simply the difference in arrival voltage, of S+/S-, at the ADC, after likely going through some resistors and amplifier circuits which would not be perfect gain matches on the two signals. So that makes sense now, and you can use the TARE function to compensate that value to be ‘0’.
You’ll also want to update to v1.2 firmware, I think solely to get the better calibration UI. Works much much nicer.
After TARE and calibrating, my little hacked bathroom scale ($9 on Ebay) now reads nearly perfectly.
Sorry little OpenScale guy… you’re a nice little kit that works well !!!