Issues with the new PPK2

[navman]

Hi nRF

I just bought the new PPK2 tool which I think is awesome. However I think there are some issues with it.

Issue 1:

First of all thy X axis scaling is wrong by a factor of 6, that is the time measurements performed/showed in the PPK2 Power Profiler tool reports timing to be 6 times shorter than real life. See attached pictures. The "Power profiler snaphot" file is a snapshot of the current consumption of 869 Mhz transmitter in active transmitting state. As can be seen the selected area is reported to be ~60 ms. However taking the measurement with an oscilloscope (voltage drop over a 5 ohm series resistor) it is reported to be ~370 ms (time between the cursors in the "oscilloscope snaphot" file). Are these observations correct? I also noticed that when the X scale is set to a span of 3 sec I would expect a current change to take 3 sec to travel from right to left when live view is turned on. However it takes 18 sec. Am I missing something here?

Issue 2:

Again looking at the Power profiler snaphot" file, in the selected area there are a lot of measurement points reporting almost 0 uA current drawn by the DUT, but in the "oscilloscope snapshot" file, it doesn't reveal that (which is correct). Is this due to auto-scaling of the measurement circuitry within the PKK2 tool? At other points the PKK2 tool reports up to 27 mA but the DUT only consumes ~5 mA in theses states, I guess this is the same issue just sort of reversed (or could it by RF from my 10mW transmitter entering the sensitive measurement circuitry in the PKK2?)

Can you confirm there are issues with the tool or do I have a faulty unit?

0675.Power Profile Snapshot.pdf

7522.Oscilloscope Snapshot.pdf

br

Kim

[Øyvind]

Hi Kim, 

We are aware some issues with the new PPK2. I have reported back to our R&D team, and will get back to you next week.

Kind regards,
Øyvind

[PVitek]

Hi Øyvind,

I have the same problem. I'm using PPK1 kit and it works very well. Now I bought the PPK2 kit and the measurement is not applicable. There was a lot of noise and sample inaccuracy in the measured data.

I measure on my golden device, where I know current profile. The measurement was created on the PPK1 and PPK2 kit.

I use a desktop computer without a USB hub. nRF Connect v3.6.1, Power profiler v3.0.3

Start up device sequency:
PPK1_StartAppCurrent
PPK2_StartAppCurrent

Transmitting on the Sigfox:
PPK1_TxSigfoxCurrent
PPK2_TxSigfoxCurrent

Sleep mode with 100 ms wake-up periode:
PPK1_SleepModeCurrent
PPK2_SleepModeCurrent

Flashing LED in sleep mode:
PPK1_FlashLedCurrent
PPK2_FlashLedCurrent

Can you confirm that there is a problem with the tool or that I have a defective unit?

Thanks,

Petr

[PVitek]

Hi Øyvind,

I am using a custom IOT device where the MCU is STM32G030xx and the Sigfox SFM10R1 module.

Measure configuration:

Measure_setup

Destop PC -> 1 m USB cable -> PPK1/PPK2 - source meter 3.3V -> 15 cm cable -> device.

Today I measure other tests in an ampere meter with a laboratory power supply HAMEG HM7042-5 and a battery. The result is better. I think it's a hardware problem in insufficient power filtering.

HAMEG have configuration 3.3V and battery 3.6V.

Start up device sequency:
PPK2_StartUpCurrent_HAMEG
PPK2_StartUpCurrent_3V6_Battery

Transmitting on the Sigfox:
PPK2_TxSigfoxCurrent_HAMEG
PPK2_TxSigfoxCurrent_3V6_Battery

Sleep mode with 100 ms wake-up periode:
PPK2_SleepModeCurrent_HAMEG
PPK2_SleepModeCurrent_3V6_Battery

Flashing LED in sleep mode:
PPK2_FlashLedCurrent_HAMEG
PPK2_FlashLedCurrent_3V6_Battery

I can try create some code for STM32 nucleo dev kit for reproduction.

Best regard,

Petr  

[Øyvind]

Thank you Petr.

Due to Christmas holidays next week, I need to ask for some patience with answers as most of our PPK team will be out of office. 

Kind regards,
Øyvind

[navman]

No problem, merry Christmas to all of you at Nordic.

br

Kim

[Øyvind]

Thank you! Merry Christmas to you too.

-Øyvind

[Øyvind]

Hello Kim, 

Happy new year! Have you had success in getting the right/better results? The PPK2 is more advanced / powerful device with a higher sampling rate. This yields more details around power spikes etc. 

Kind regards,
Øyvind

[navman]

Unfortunately my issues still persist. However using the PPK2 connected to my laptop seems to produce better results (but not correct). As mentioned earlier timing (X-scale is way off on my desktop but better on laptop) and there is a lot of "sampling noise" again most prominent on my desktop computer. The noise is not present using an oscilloscope to measure the current consumption as a voltage drop over a small series resistor and timing is of course correct.

Looking forward for a solution.

Br

Kim

[PVitek]

My problems also persist. Measurement with PPK2 is full of noisy signal, see. previous pictures. In source meter mode, the signal is the worst. In ampere meter mode with battery source, the measurement is almost good. I think the problem is in hardware power supply filtering and usb power filtering. I still haven't been able to test the device on a laptop powered by an internal battery. If the problem isn´t be solved, then for me PPK2 isn´t usable for measuring IOT devices compared with PPK1.

Regards
Peter

[Øyvind]

PVitek, can you please register a new ticket, as the issues you have do not seem related to what navman is seeing.

Please provide images with time stamps, and include the sidebar showing SMU voltage and other settings. Our PPK team are concerned about the antenna close to the PPK2 cables as well, are you able to increase the distance between (move the antenna further away)? 

With regards to the PPK1 vs PPK2 results, the PPK1 did have more noise reduction and averaging on the signal, and a lower frequency range compared to the PPK2.