[STM32G431] Programming issue

[BBute.1]

Hello,

I am facing big issue to program STM32G431 on custom boards. The issue appears on two different designs.

I am using ST-Link V3 through SWD to program the device as I do usually but it is really hard to get the connection.I have tested 3 different boards to avoid any soldering issue with 2 different schematics.

The device does not seem to respond as shown below (blue is SWCLK red is SWDIO)

The low voltage level pulled by the device seems to stay at 0.7V. All board have the same behaviour.

SWDIO and SWCLK are directly connected to the microcontroller. NRST pin is also directly connected with a 100nF capacitor to the ground. VDD is generated externally @3.3V to ensure thee is no power supply issue.

BOOT0 pin is hard wired to ground and is not accessible.

On a very few occasion, I managed to get a proper connection.

The issue seems to be only with this particular microcontroller as it is the second design with the same issue.

I never had this kind of problem with our programming circuit on another microcontroller.

NRST is rightly pulled-up to VDD when not pulled low by ST-Link.

VDD Capacitor are 2x10µF + 2x100nF. VDDA is connected to VDD through a ferrite bead and then decoupled by 1µF + 100nF + 10nF

Any help would be appreciated.

 

[Andrew Neil]

Please show schematics

[Tesla DeLorean]

Schematics + Connectivity

Some grounding issue? Double check net-lists, net names, that there aren't any odd disconnected islands.

>>I never had this kind of problem with our programming circuit on another microcontroller.

There's some basic viablity issue here. Grounding BOOT0 makes alternate signs-of-life tests difficult. Although you can probably probe via System Loader pins (AN2606) when FLASH is blank. Check there's no conflicts of AN2606 expectations.

[BBute.1]

Hello,

I understand your concerns about BOOT0 being grounded but we are using bespoke bootloader on the app side and always keep programming pins (SWCLK and SWDIO) without any other functionality, so I think that in this case it is correct to wire it like this, don't you think?

Regarding my issue, I made some other tests and I saw that the device is programming like a charm (like we use to have for other mic) at VDD = 3V. As soon as I increase external power supply to a bit more than 3.00V it is failing.

I still think there is something wrong on our end, but I don't know what. Note that it is working at any speed and on all boards now with VDD=3V. Any insight?

[MM..1]

Dont worry, G4 is little sensitive , i for any other MCUs use clone STLink, but here do exactly same as your trouble. Then i cut STLink from Nucleo board and all is ok. Recommend try more programmers and choice one working.