NRF52832 advertising current too high

[devotronix]

I'm developing a bluetooth device that only advertises data, it's set to BLE_GAP_ADV_FLAGS_LE_ONLY_GENERAL_DISC_MODE and BLE_GAP_ADV_TYPE_NONCONNECTABLE_SCANNABLE_UNDIRECTED. I'm advertising 12 bytes of custom data and reading a pin voltage every 5 minutes. For reading the pin voltage, I'm using a timer with NRF_SDH_CLOCK_LF_ACCURACY set to 7 (20 ppm). My advertising interval is set to 1000ms, and tx power is -12. DCDC regulator is enabled. Clock source is internal RC. Input voltage is 3.3V from a bench supply. As far as I can tell, and according to the online power profiler, this should be an extremely low power device. Current peaks should be around 5mA, and average should be around 10uA. However, I'm seeing an average of about 85uA, with my peak current topping out at 34mA. I can't figure out what is causing the high current peaks. Any ideas what's going on? I've included a screenshot of the power profile for 4dbm tx power and for -12dbm tx power for reference.

4dbM:

-12 dbM:

[Kenneth]

I think you may have damaged your PPK, you may check out this thread for some debug steps:

https://devzone.nordicsemi.com/f/nordic-q-a/41088/ppk-gives-wrong-value 

https://devzone.nordicsemi.com/f/nordic-q-a/47560/power-profiler-kit-gives-wrong-current-measurments 

[devotronix]

I don't think I damaged my PPK. I just tried with a different board and received more expected results. Here is what I've tried. I have 3 boards I'm testing, I changed the TX power to 4dbM:

1. nRF52832 - average 80 uA, peak 39 mA

2. TI TLV61225 - average 9 uA

3. nRF52832 + TLV61225 (voltage input 3V) - average 33 uA, peak 10mA

4. nRF52832 + TLV61225 (voltage input 2.4V) - average 44 uA, peak 16mA

It's possible the board that I removed the boost converter from has a damaged bluetooth chip, which could explain why the board with the boost converter looks fine. What I REALLY want to do is test the board with both the nRF52832 and the TLV61225 together with a voltage range from 0.8 to 1.3 volts. However, the lowest input voltage the PPK will let me use is 2.1 V. Anything below that and the power profiler app starts wigging out. Any ideas how I could go about testing the lower voltage range? I don't have an oscilloscope, just a Fluke 87V, which doesn't agree too well with the PPK. For case 3 it gives an average current of 21 uA, and case 4 an average current of about 80 uA, and then about 120 uA at 1.4 V. Any ideas short of buying an oscilloscope?

[Kenneth]

devotronix said:

Any ideas short of buying an oscilloscope?

Sorry no, there are also other more expensive power analyzers from third party such as Agilent. But these are typically more expensive than an oscilloscope.