PPK2 showing incorrect measurements at specific voltages

[Yoonoo]

I'm using the Power Profiler Kit II to measure current draw of an ESP32 board at idle. The PPK2 is connected to the battery terminals of the board (XIAO ESP32-S3)  and whether it's set to source meter or ampere meter, when the voltage is at 3.55V it shows a current draw of about 300mA. When the voltage is set to 3.4v or 3.7v it shows the correct draw of 36mA (As verified with a Fluke digital multimeter).

When nothing is attached to the board this issue happens around 3.55V, but with some components attached it happens around 3.7V, so it doesn't always happen around the same voltage. I did try another XIAO ESP32-S3 board with the same result.

Is there something I need to do differently? With the same exact circuit, other meters show the correct current no matter what the voltage.

[Kenneth]

Hello,

I suspect the problem is likely that it's hopping between two measurement ranges and somehow this cause an issue:
https://docs.nordicsemi.com/bundle/ug_ppk2/page/UG/ppk/ppk_measure_resolution.html 

Are you able to experiment with adding some decoupling capacitors between the board and the PPK 2? Just to check how much decoupling would be needed.

For test, can you try the firmware Audun shared here to see if it affects the problem in any way?
https://devzone.nordicsemi.com/f/nordic-q-a/120653/power-profiler-kit-ii-measurement-error 

Based on your feedback I might try to ask a bit more internally.

Kenneth

[Yoonoo]

Thanks for your suggestion! I tried a few decoupling capacitors and settled on 33uF. This made the measurements steady across the entire voltage range (2.8V to 4.2V), they are still higher by about 10% from my other multimeters but this is insignificant.

What I don't fully understand is if this capacitor would have any side effects on the measurements of other circuits.

I will also test with the firmware you recommended and see if that makes a difference.

[Kenneth]

Yoonoo said:

What I don't fully understand is if this capacitor would have any side effects on the measurements of other circuits.

There is no negative side effects by adding capacitance, other then it will "smooth" out the current consumption, so it will not be as transparent to see the exact timing of short transients.

Kenneth