PPK2 strange current reporting after DUT goes into low power state

RE: PPK2 strange current reporting after DUT goes into low power state

DevinCallahan — Mon, 09 Aug 2021 15:27:49 GMT

Thank you so much for looking into it! Its not a problem for us at the end of the day and I'm relieved I don't have to go track down whatever could be consuming current like that! The PPK2 is still doing a fantastic job for us over all.

Cheers!

Devin

RE: PPK2 strange current reporting after DUT goes into low power state

Simonr — Fri, 06 Aug 2021 07:28:41 GMT

Thank you. I've reported that DevZone is not able to upload .ppk files. Thank you for providing the error message. I have reported this issue internally.

As for the PPK trace, I had one of the PPK developers take a look at it, and he does indeed seem to think that this is one of the switches causing this current consumption. You can also see that the measurements between these 700uA spikes are all the way to 0, indicating that there are some negative currents as well. These samples are not to be trusted I'm afraid. Our recommendation is to ignore these measurements, as the sleep current seems to run as expected after this "noisy period". Unfortunately, this is the greatest weakness of the PPK2, as the bandwidth does not support the switching occurring here.

Best regards,

Simon

RE: PPK2 strange current reporting after DUT goes into low power state

DevinCallahan — Thu, 05 Aug 2021 15:26:26 GMT

Attaching the ppk of the same capture here now!

I'm not sure why but every time I tried to upload the file with the .ppk extension it would fail with this error:

{"jsonrpc":"2.0", "error": {"code":101, "message":"The file's name could not be determined.  The file cannot be accepted."}, "id":"id"}

As such I am uploading it with the .csv extension which DOES seem to be working. When you download it you will need to rename it with the proper .ppk extension.

devzone.nordicsemi.com/.../MeasurementWithJunk_5F00_ppk.csv

Thank you!

RE: PPK2 strange current reporting after DUT goes into low power state

Simonr — Thu, 05 Aug 2021 08:54:54 GMT

Sorry, could you save it as a ".ppk" file instead, as I'd like to take a look at it in the Power Profiler app on my end.

Best regards,

Simon

RE: PPK2 strange current reporting after DUT goes into low power state

DevinCallahan — Wed, 04 Aug 2021 22:30:51 GMT

I have a .csv showing off the capture here now. I am not 100% convinced the excess consumption isn't from something else waking up on our device, but the signature of the "spiky" current is very different from everything else I have seen.

Let me know if I can try to capture something else!

devzone.nordicsemi.com/.../MeasurementWithJunk.csv

RE: PPK2 strange current reporting after DUT goes into low power state

Simonr — Mon, 02 Aug 2021 11:19:47 GMT

Hi again

Thank you! We've not seen this issue on our side before, there might be one of the switches that are struggling indeed. With the PPK2 it's possible to export the Power Profiler data as a file (see the video in my last reply). Can you do so and upload it here so we could analyze this "noisy part" on our side? I know there is a switch at ~50µA, but if the MSP430 is steady below this, it should be fine.

Best regards,

Simon

RE: PPK2 strange current reporting after DUT goes into low power state

DevinCallahan — Fri, 30 Jul 2021 15:52:09 GMT

Hey Simon,

Thank you for the response!

[quote userid="75734" url="~/f/nordic-q-a/78014/ppk2-strange-current-reporting-after-dut-goes-into-low-power-state/322578#322578"]First off, are you measuring in Source meter mode or Ampere meter mode?[/quote]

The PPK is configured as an ampere meter for this setup.

[quote userid="75734" url="~/f/nordic-q-a/78014/ppk2-strange-current-reporting-after-dut-goes-into-low-power-state/322578#322578"]Can you also provide some details on how exactly you go into "low power mode"?[/quote]

The supply being measured in this instance is a 3V regulated supply that we use as VDD_GPIO for the nrf9160, as well as VCC for the other microcontrollers we have on our design. The microcontroller turning on in this instance is an MSP430 which is being put into its low power mode. That is done manually in its firmware and is indicated by the digital pin 7 going "low" in the capture.

RE: PPK2 strange current reporting after DUT goes into low power state

Simonr — Fri, 30 Jul 2021 06:28:02 GMT

Hi Devin

Due to the summer vacation period here in Norway we are short on staff and delayed replies must be expected. I'm sorry for any inconvenience this might cause.

This behavior seems strange indeed. First off, are you measuring in Source meter mode or Ampere meter mode? If you're in Source meter mode, make sure that the source output voltage is set to a value suitable for your device. Can you also provide some details on how exactly you go into "low power mode"? Are you going into System OFF mode, or in an idle system ON mode of some kind?

Depending on the sampling rate you'd like, you can increase the sample duration up to 500 days as explained in this YouTube video. We can't guarantee the behavior if you start modifying the JS yourself, but there shouldn't be any restrictions in the HW keeping you from setting this sample period yourself.

Best regards,

Simon