Help debugging power consumption in low-power modes on an stm32wle target

[AAlat.1]

Hi!

I have a custom board built around an STM32WLE5CCU7.

My problem is that it consumes way to much power (100 times expected).

110uA in Shutdown and Standby

190uA in STOP2 WFI.

CubeMX power tool estimates these two modes to at most 0.640uA and 1.3uA respectively.

I have created a completely blank project (see attached files) which does nothing but enter various sleep-modes. It is using this project I measure the figures mentioned.

It is of course possible that I have leakage currents on my custom board, but

1. I have de-soldered everything I can think of that can consume power on the board (and all stm32 pins as default i.e. analog input).
2. It doesn't explain the 80uA difference between shutdown and STOP2

The board has a SMPS for generating 3v3 but I measure the current on the 3v3 after the SMPS so I dont think it should matter...

The current-diference between stop2 and shutdown is very puzzling to me, and indicates that I have (atleast one) software issues.

I'm at a complete loss here and very grateful for any suggestions on how to proceed the fault finding..

Edit: I have the debugger physically disconnected during measurements (and powercycle the chip after disconnectintg the debugger)

Best Regards

Axel

[Louis AUDOLY · Accepted solution]

Hello @AAlat.1​ ,

Have you try the PWR_STANDBY single core example provide in the FW package STM32Cube_FW_WL_V1.2.0 ?

This could help you to determine if this come from your software or if you still have an issue with your hardware.

Also you can find here an article from a member of the community that explain how to enter in standby.

I hope this help you, let me know if you succeed.

Best regards

Louis

[AAlat.1]

Thanks for your reply, and sorry for my late response!

I keept comparing my code to your links and eventually found the issue.

The issue itself was twoffold.

Problem 1

Was hardware, the RF-switch (taken from the reference design) seems to consume 110uA. We had missed that during schematic check ��. After desoldering the switch shutdown mode consumed <2uA (Cannot measure more precise in my current setup)

Problem 2:

STOP2 mode issue remained and consumed 80uA. I erroneously thought all pins were analog-input by default (which some parts of HRM can lead you to believe). It seems the debugger pins (PB3-4, PA13-15) are exceptions that are configured in peripheral mode (which thinking about it makes sense, otherwise the debugger cant control a blank chip after reset).

After manually setting these 5 pins to analog input stop2 also consumes <2uA.

My problem is solved for now.

Now I need to turn things on one at a time and see that I can retain the low power in sleep, but now I have a baseline to work from!

[AD1]

Sorry...I am Vietnamese

Can you give me your <2uA sample project... when I use it, it is at 290uA

i tried a lot of different ways but couldn't get it down to even <10uA

[AD1]

sorry.....I don't know where to ask, so I have to ask here

i can't sleep at <10uA, i'm using RAK3172 with arduino firmware then i can sleep at <10uA when i use cubemx i can't deep sleep specifically at 290uA

[STTwo-32]

Hello @AD1 and welcome to the ST Community .

It will be better if you create a new Post on the STM32 MCUs wireless Forum and give more details about your problem. 

Best Regards.

STTwo-32

B

[AD1]

I can't post because I'm new to this... is there any way to help me get back to deep sleep as quickly as possible?

[AD1]

Can you help me via email? i am vietnamese

my email: vi.trinhdinh@meiko.vn

thank you so much