cancel
Showing results for 
Search instead for 
Did you mean: 

STM32H745 is unresponsive

Rawcode
Associate

Hi everyone,
I've built a STM32H745ZIT6-based board. The microcontroller uses the LDO to generate VCORE. The idea of this board was to build a standalone solution based on my Nucleo-H745ZI-Q project.

Screenshot from 2025-08-23 18-34-51.png

I did prepare a very simple project, where I toggle a pin with a simple delay, to check that everything works, but sadly nothing seems to work. I'm trying both HSI and HSE clock source, but there is no difference.

The first strange thing that I noticed is that during debug the M4 core is stuck in hard fault, even if the configuration used is the same of my project based on the Nucleo:

Rawcode_3-1755967799864.pngRawcode_1-1755967728129.png

 

Rawcode_4-1755967830586.png

Rawcode_5-1755967842628.png

Here is the debug window, and the fault analyzer output:

Rawcode_6-1755967957096.pngRawcode_7-1755967975848.png

To understand if at least the CM7 works, I commented the "#define DUAL_CORE_BOOT_SYNC_SEQUENCE" to allow it to continue the execution. Nonetheless, the microcontroller seems completely unresponsive.
During debug it seems to go through "HAL_GPIO_TogglePin(CH8_GPIO_Port, CH8_Pin);", but the pin doesn't physically change state (it remains LOW). Furthermore, "HAL_Delay(500);" never returns (uwTick is always at 0);
I noticed that System tick timer NVIC1/2 has Preemption Priority of 15, but reducing it to 0 didn't change anything.

Did I do something wrong? Am I missing something or do I have a faulty unit?

1 REPLY 1
TDK
Super User

Connect with STM32CubeProgrammer. Is there code at both 0x08000000 and 0x08100000? Doesn't seem like the M4 code got uploaded

If M7 core is at line 90, that's the first line in main()--an automatic breakpoint--so it's probably waiting for you to hit resume.

If you feel a post has answered your question, please click "Accept as Solution".