cancel
Showing results for 
Search instead for 
Did you mean: 

STM32WB55 STOP2 + BLE advertising current stuck at ~1.1–1.4 mA (WeAct WB55CGU board)

Araleb
Associate II

Hello,

I’m working with STM32WB55CGU (UQFN68) on a WeAct WB55 board and trying to achieve ultra-low power during BLE idle / non-advertising periods using the STM32 WPAN framework.

Expected behavior

According to ST documentation and examples:

  • STOP2 mode with BLE stack idle should be in the tens of µA range

  • I previously verified 10–20 µA STOP2 current without BLE enabled, so the board itself is capable of low power

Actual behavior

With BLE enabled (advertising):

  • Idle (non-advertising period): ~1.1–1.4 mA

  • Advertising peak: ~8 mA

  • BLE communication works correctly

  • PB6 debug pin toggles, confirming STOP2 entry

  • Removing ST-LINK reduces only ~100–200 µA

Hardware

  • MCU: STM32WB55CGU

  • Board: WeAct WB55

  • External LSE present

  • Measured using shunt resistor + scope (differential measurement)

Software setup

  • Based on BLE_p2pServer example

  • Cube FW: STM32Cube FW_WB V1.23.0

  • IDE: STM32CubeIDE

  • RF firmware loaded correctly (same as NUCLEO examples)

  • USB, ADC, PWM, DMA disabled for low-power test

Main loop

 

 
while (1) { MX_APPE_Process(); // WPAN scheduler }
 

Low-power flow (confirmed)

  • UTIL_SEQ_Idle() is called

  • UTIL_LPM_EnterLowPower() is called

  • PWR_EnterStopMode() → HAL_PWREx_EnterSTOP2Mode()

  • PB6 toggles inside STOP2 hooks

What I already checked

  • STOP2 works perfectly without BLE

  • No busy loops or delays

  • No active timers except BLE stack

  • No GPIO leakage

  • HSEM / IPCC initialized as per ST examples

  • Same behavior with and without ST-LINK

  • RF wakeups are expected, but average current seems far too high

Question

  1. Is ~1–1.4 mA expected for STM32WB55 during BLE idle when using the WPAN framework?

  2. Is there any mandatory low-power configuration specific to the WB dual-core / IPCC / RF subsystem that is not obvious from the examples?

  3. Are there known differences between:

    • NUCLEO-WB55 boards

    • Custom boards (like WeAct WB55)
      regarding default power domains, SMPS/LDO configuration, or RF clocking that could explain this?

  4. Is there a reference project (p2pServer or similar) that demonstrably reaches <100 µA average during advertising intervals on WB55?

Any guidance, checklist, or clarification would be highly appreciated.

Thanks in advance,
Aharon

0 REPLIES 0