General • Re: ADC creapage on the RP235X

I don't understand it and I have never observed this behaviour with any other ADC I have used.

I do have one theory. I don't know how credible it is but feels plausible to me; that the ADC convertor is seeing two voltages mixed together, is converting the highest of the two.

I am thinking there may be some sort of internal capacitor which gets discharged to 0V whenever the ADC is read and slowly self-charges thereafter. While that capacitor voltage is below the ADC0 input voltage that input voltage takes precedence. After 13 minutes or so the capacitor voltage has risen to above the input voltage and that is what gets seen by the ADC convertor.

I did some more testing and it seems to be around 12 to 14 minutes when this weird behaviour starts to manifest itself.
I guess that if we could find an exact timing for when this presumed capacitor voltage starts to exceed the input voltage we could better characterise it, determine its capacitance and even charge rate - though that might not be easy if it's not a fixed charge voltage or rate.

The trouble is these tests take ages and are inconvenient when wanting to use the device for other things. I'll put further testing on hold until I have more down-time to do that in, will look at logging to file again so I can leave tests running overnight.

Statistics: Posted by hippy — Wed Nov 20, 2024 11:02 am

---