Power Profiler Kit II: incorrect measurement

[Peter49]

Hi,

I am using the PPK-II to measure our system board current (our system uses the nRF9160 and nRF52840) over a long period (many hours).

System is run from a Li-Po battery, so PPK-II is connected between battery and system board.

PPK-II is in datalogger mode. 100 samples/second, sample for 6 hours.

At the moment, the initial baseline current is measured at an average of 1.52mA. See screenshot.

But after an LTE call-in the average baseline current increases to ~2.00mA (excludes the call-in current). 

In the screenshots, the >50mA current spikes are LTE call-in each hour.

In investigating to the reason for this, i have additionally connected a DMM to simultaneously measure the current with the PPK-II.
The DMM measurement remains at 1.52mA while the PPK-II is measuring higher at ~2.00mA.

This seems to be reproducible after a long runtime > 2 hours, and after an LTE call-in where there is an increase in current possibly when the current reaches >150mA (poor signal).

Interestingly, the current measurement on the PPK-II returns to the correct measurement after the datalogger is stopped and restarted again from the PC Power Profiler App (v3.5.5).   

Power not is removed from the PPK-II (USB to PC).  Only button press to stop and start the datalogger.

Note. when stopping the datalogger, the power to the system is not turned off, so the system remains ON.  The system is still running normally and has no change when the datalogger is halted.

The screenshots below is a zoomed version of the above where you can see the increased baseline current after the LTE connection has completed.

Are you able to provide a reason why this could be occurring?
 

[Håkon Alseth]

Hi,

 

Thanks for providing such in-depth explanation, in addition to screenshots of the measurements.

In investigating to the reason for this, i have additionally connected a DMM to simultaneously measure the current with the PPK-II.
The DMM measurement remains at 1.52mA while the PPK-II is measuring higher at ~2.00mA.

This seems to be reproducible after a long runtime > 2 hours, and after an LTE call-in where there is an increase in current possibly when the current reaches >150mA (poor signal).

Interestingly, the current measurement on the PPK-II returns to the correct measurement after the datalogger is stopped and restarted again from the PC Power Profiler App (v3.5.5).   

This is very interesting. If the PC software sampling is cycled, ie. Stop/Start sampling, without actually cutting the power; you observe the correct 1.5 mA consumption, even in the GUI software.

I checked the ppk2 fw, and it is not much that is actually "stopped" when you hit the stop sampling button, which can point towards a race-condition in the PC tool.

 

Do you happen to have the .ppk2 file saved, so I can share this with the PPK2-team for further analysis?

 

Kind regards,

Håkon

[Peter49]

Hello Håkon,

Unfortunately, I don't seemed to have the saved the ppk file for the first 3 screenshots above, but i do have the file for the last screenshot (this was from another test).


In this data, you can see the behaviour change from average baseline 1.521mA to 2.09mA at the ~3 hour mark after the LTE has completed and terminated the connection.

ppk-20231027T192251-data-logger_BORB03.ppk

Here is a second data file with similar results

ppk-20231024T230950-data-logge_BORB03.ppk


I also ran another test today, where the baseline current increased after 1 hour (after an LTE connection).
After 3.5 hours i stopped the logging and saved the file. And restarted for another 5 minutes to log the lower baseline after the restart.  The time between the stop and start was less than 20 seconds.

Here are the 2 files for this test if it helps.

ppk-20231030T200812-data-logger_BORB03.ppk

ppk-20231030T201325-data-logger_BORB03.ppk

[Håkon Alseth]

Hello,

 

Thank you for sharing the files. I have reported this internally to the PPK2 team and shared your files with them.

 

Kind regards,

Håkon

[Peter49]

Hello Håkon,

Was your team able to provide a reason to the above issue?

Thanks.

[Håkon Alseth]

Hello,

 

Peter49 said:

Was your team able to provide a reason to the above issue?

My apologies, but it seems that this has been a bit of a tough scenario to reproduce at our end.

We have not been able to reproduce this yet, unfortunately.

 

Could I ask about your physical setup?

Q1: Is the PPK2 connected through a USB hub? If so, is there any change if you connected directly to a USB 2.0 port instead?

Q2: Are you running on a AMD or Intel Platform?

 

Kind regards,

Håkon