I’ve been working with the qwiic scale for a little bit and it works great except for one thing. I’m using a bathroom scale at the moment but also have a custom built scale. Sometimes the chip seems to freeze or at least stops completing conversions. In other words myScale.available() will never return true again. This is not predictable but it always happens after I add and remove weight from the scale. In particular weights in the 10-20kg range but it might happen with smaller weights. Thanks.
I’m wondering if static electricity could be the cause. A bit of static discharge when you place an object on your scale could be confusing the IC.
You might try setting a non conductive platform on the scale and then placing your weights on that to see if the problem goes away. If it does, grounding the platform on your scale might fix the issue.
Sounds reasonable. Thanks. In any event I will also try implementing a watchdog method because I will be measuring live weight and having the user reset the measuring system is not an option. If myscale.available returns false for some period then I will (assuming the chip is still responsive) issue a reset then do a calibration then reload the cal/offset. If the chip is not responsive then I will need to cycle the power.
I am putting together a “falling chain” demonstration for a physics class I teach. I am noticing as well at random times the nau7802 stops converting and the myScale.available() in the loop() function stops returning “true”. One pattern … there are usually 3 reads with large negative weights (I use myScale.getWeight(true)) and then nothing.
While the times are random, usually it’s more than an hour before the nau7802 locks up. Plenty of time to do a class demo of a falling chain.
I haven’t encountered this problem since switching to load sensors which were purchased from Aliexpress (ie not the ones from Sparkfun) Not dissing the Sparkfun sensors but there have been other strange anomalies that have disappeared since switching sensors. I’m not running a science lab and have a tendency to change lots of things at the same time so… But I do think the problem with lockup is probably due to putting in too much voltage on the differential input for too long.