[DONE] The New STM32F429/F439

[Giovanni]

Could you use halPolledDelay(x) and determine the X required?

Giovanni

[trepidacious]

I've had a look at this, and the results are a bit odd.

A halPolledDelay(1) instead of the 1ms sleep will fix things, I've tried it a few times in a row and it seems reliable. However if I run with a delay so that the USB starts normally, then remove the delay and reflash, it will also work. If I just use the reset button, it will start up again with no problem. However if I remove the mini-USB (and hence power down the MCU) then reconnect, it will not start up.

So the problem seems to be only on first power up, and only requires a very short delay in this case.

[Tectu]

This really sounds like some missing pull-ups/pull-downs somewhere... no?


~ Tectu

[trepidacious]

6 of these will also work reliably, any less will only work sometimes:

Code: Select all

asm volatile("mov r0, r0");

[Giovanni]

OK, it seems a bus synchronization issue, it just requires a small delay. I'll search on the OTG manual for the cause of this. I'll add a small delay anyway.

Giovanni

[Giovanni]

Hi,

I added a small delay to the reset procedure, could you please try the change? thanks.

Fixed as bug #436.

Giovanni

[Tectu]

I can confirm that it works with the current master.


~ Tectu

[Tectu]

Just wondering... why is the STM32_VDD in the board file set to 300 instead of 330?


~ Tectu

[Giovanni]

I see 3V on the schematic, it does not make difference anyway, it is the same working range.

Giovanni

[trepidacious]

Sorry for the delay - I didn't have the disco board over the weekend, the new code works fine for me

On the 3V thing, it's due to a protection diode. The user manual says D1 and D2 on 5V and 3V supplies, but I think it's actually D3. The 407 discovery board had the same thing and it really irritated me, so I just replaced the diode with a wire to get 3.3V. As long as you don't try to use the 3V as a power input I think it's safe, but no guarantee