cancel
Showing results for 
Search instead for 
Did you mean: 

Bootloader over SPI using a STM32U5A5 and STLINK-V3SET

hiimron
Associate

Hello all,

We are trying to get into the bootloader of the NUCLEO-U5A5ZJ-Q over SPI using a STLINK-V3SET. But are having issues with connecting the V3SET in the STM32Programmer and through the CLI, which is getting this error.

Screenshot 2025-08-18 174316.png

So far, we are seeing this happen on the SPI lines through a logic analyzer:

Screenshot 2025-08-18 164757.png

Screenshot 2025-08-18 161822.png

Screenshot 2025-08-18 164916.png

And we are not sure why...

We followed these two ST posts but neither post seem to be faring any success for our board combination:

Solved: STM32H563 bootloader command over SPI - STMicroelectronics Community

How to use STM32CubeProgrammer and the STLINK-V3SE... - STMicroelectronics Community

We have correctly applied the correct pattern from the AN2606 document, which is Pattern 12 for the STM32U595 and have this pinout between the boards:

SignalNucleo U5A5ST-Link V3

GND

CN7

CN6 pin 3

T_VCC

CN8 - 15 (VIN)

CN6 pin 1

SPI1_MOSI

CN7 pin 12 (SPI1 MISO)

CN9 pin 2

SPI1_MISO

CN7 pin 14 (SPI1 MOSI)

CN9 pin 3

SPI1_SCK

CN7 pin 10 (SPI1 SCK)

CN9 pin 4

SPI1_NSS

CN7 pin 16 (SPI1 CS)

CN9 pin

Are we missing something crucial or is this even possible for this particular Nucleo board?

3 REPLIES 3
MHoll.2
Senior III

T_VCC must be the voltage used by the MCU, so You should not connect VIN (5V) but VDD (3.3V or 1.8V).

But in You case T_VCC is not used (The SPI interface is specified only for 3.0V to 3.6V signals).

The MCU is not responding to the STLINK.

- Have You tried to used the VCP interface provided by the integrated STLINK on the NUCLEO board, to verify that the MCU is in Boot mode?

- Have You tried to invert SPI1_MISO with SPI1_MOSI?

Good point, the VCC of the V3SET expects 3.3V so we moved the pin to a 3.3V pin. 

We were able to successfully connect the Nucleo board to the bootloader using the VCP interface and getting this output from the log: 

Screenshot 2025-08-19 090005.png

We did also switch the SPI pins a couple of times and does not seem to change behaviors. 

hiimron
Associate

Anyone have any updates to this or any other suggestions?