Power consumption - DWM3001C

[Braxatoris]

Hello guys,

Im developing RTLS with DWM3001C and right now Im working on personal tag powered by CR2450. So naturally, power consumption should be as low as possible.

So far I managed it to these levels - idle not moving - around 15-20 uA, after moving tag UWB TX is around 22-30 mA spike, and periodically there are every 100ms spikes of 60 uA +/- from accelerometer sampling.

Over past year, i tried everything possible but this is as low as it gets... from what i read it should be able to go 5 uA +/- ? 

Any help will be appreciated :) 

IDLE consuption 

Pinging

[Simonr]

Hi

The idle power consumption here looks very noisy on your end, so can you share some details on how exactly you measure current here. Are you able to recreate the same consumption on a development kit with a similar application on it. 

What device are you using for current consumption measurement here, and what is the accuracy of this device? Do you by chance have a Power Profiler Kit II to do current consumption measurements on? You are correct that an idle nRF52833 should be <5µA, depending on what settings you have enabled.

Best regards,

Simon

[Braxatoris]

Hello Simonr, thanks for reply !

Device for measuring is of our own making, and at its heart is sitting  INA229 chip. Here is measurement with nothing hooked in. 

Here noise is < 2 uA so it shouldnt be problem of measuring method right ? There shouldnt be anything else on board consuming power.. maybe DWM module which is in deep-sleep (+ few nA)... 

Regretably, I dont have PPK 2 at this time, so this is all i get... I didnt go to much detail of my settings and config because for past year i tried almost anything which I was able to find on internet... I have disabled everything i could... Im hoping we find some Kconfig which i didnt try or tried but didnt work ? 

[Simonr]

Hi

Okay, do you have an nRF52 series development kit you can measure current on with this device just to make sure we have the same point of reference? For example the system OFF sample here running on a DK can make sure the measuring device is correct: https://docs.nordicsemi.com/bundle/ncs-latest/page/zephyr/samples/boards/nordic/system_off/README.html#nrf_system_off 

Please upload what current consumption you see and what DK you're testing on. If this is expected/similar to what we see on the PPK we can move on to your board.

Best regards,

Simon

[Braxatoris]

Firstly, my apologies for late replay...

Secondly, I have DK only for DWM30001C, I could test it there but I would need to inlcude drivers for DWM and deepsleep it too... maybe later when I found some time..

But meanwhile... we bought PPK II :) so here are results...

This is when device finishes TX and goes to deep sleep... it stabilizes on 13-18 uA ticking... 

This is deep sleep interval... peaks to 28 uA are IMU (LIS2DH12) readings...

DWM should consume around few nA in this state and only thing which should burn power now is LIS sensor

but it is ODR 10hz and LOW POWER mode... in this mode it should consume on average 3-5 uA ? So it still appears to me that NRF is doing something that is consuming power.. or if average calculation of PPK2 is anything to go by... could this be correct ? 5 +/- uA of nrf sleep + 5+/- uA of LIS2DH12 working could average on 12 uA as in picture... 

[Simonr]

Hi

Can you share more details on what SDK version you're using and what else is in use on the board? The floor current seems to be around 9-10µA, so there's definitely something else drawing current from the VDD net here (assuming you measure current consumption from the VDD supply on the board. That could be a leakage current from another GPIO or from another onboard device/sensor. Are you using other external components than the LIS2DH here? Are you sure the LIS2DH goes to sleep correctly?

Best regards,

Simon