STM32F446 starting squence

[fab04]

Hi everybody,

I've two cards with a STM32F446 on them.

My first one works very well, the µc starts a each time.

On the second card, the µc starts sometimes (I have a LED on my card that blinks at start).

I've checked Vcap (2.2µF), the 3V3 rising (1ms), boot is well connected to GND with a 10k resistor.

I cannot see what happens in debug mode, because in debug mode , it starts everytime.

Is there a mean to check something else ?

I don't understand what happens.

If you have some ideas...

Thank you.

 

[mƎALLEm]

Hello,

---

@fab04 wrote:

Hi everybody,

I've two cards with a STM32F446 on them.

---

Which board? custom board? ST board?

[fab04]

Hi,

 

No, it's a custom board. I think something happens on my power supply. I'll check it.

[Bob S]

> because in debug mode , it starts everytime.

Do you mean while using a debugger (STLink,etc.)?  That is because the debuggers bypass the "power on reset" sequence - manually loading the PC/SP from the reset vector and then starting the CPU.

You don't mention what you have connected to the NRST pin, nor the crystal or oscillator circuit.  Can you post portions of your schematic related to power, reset and external crystal (if you are using one).

[TDK]

Usually this is due to a hardware issue. Perhaps bad design, perhaps manufacturing defect such as a cold solder joint.

[TDK]

Perhaps blink an LED before SystemClock_Config and then again after. Perhaps it's getting stuck in there.

If power is valid and NRST is high, which it looks to be, the chip is running something.

[fab04]

I think I've found the solution. I put two capacitors for LSE (22pF).

But the crystal that I choose needs 12pF. I've made the modification and now it starts each time.