stm32f030CC cannot connect under reset but can connect under hotplug

[Luke Iliffe]

Hello all,

      I've been getting some strange issues occuring with some of our kit out in the field which I would like some input on. We have multiple boards that use a  stm32f030CC and stm32f091 MCU. The boards have been reported a faulty due to the stm32f030CC being non-functional. When the field engineer tried to connect to the stm32f091 using a stlink he was succesfully able to do so in connecting under reset. When he then tried connecting to the stm32f030CC he was unable to do so and had to instead connect using HOTPLUG. Given that the inital fault was failure of the stm32f030CC to run I figured that this might well be related. Once he was connected he could erase and reprogram the stm32f030CC and it then successfulffy ran. This didn't fix the "connect under reset" however and im worried that there may be some underlying damage to the stm32f030CC.

Does anyone know what kind of current draw the pulse generator on the NRST pin of the stm32f030CC is meant to handle.

as far as i can tell its 5ma

The reason I ask is that our circuit on the NRST pin allows the stm32f091 to reset the stm32f030CC using a GPIO set to PP_OUTPUT through a 1k2 resistor, with an additional 10k pullup and im wondering if i might be damaging the pulse generator on some IC's. An example of what we have done is shown below

I dont think that this would draw enough current to cause any issues when the NRST pin is pulled down internally but am interested in what everyone thinks.

Thanks for your time

Luke

[AScha.3]

Hi,

i dont think, you damage something with driving nrst by 1k2.

But for the connected st-link : it has to fight against this, to get nrst low.

Just try: setting the F091 pin to open drain output; this way you can reset the F030 as before, 

but only the 10k is active as pullup. I would do it this way.... 🙂

[Peter BENSCH]

@Luke Iliffe You also don’t need a pull-up of 10k (even if some Asian board manufacturers keep adding this) as there is an internal one behind NRST. However, I would still add a 100nF capacitor to NRST to suppress interference pulses.

[MNapi]

just remove those 2 resistors and see if you can connect now. my guess is that the chip is under reset all the time