Short circuit on STM32F051 after programming

[AreaSX]

Hello everybody,

I'm a newbie on ST processors although I have a deep experience on several other MCU (Microchip on top).

I'm using an STM32F051K8T6 with minimal connection (VCC,GND,NRST,SWDIO,SWCLK and a Led) to start developing a porting from an existing Microchip firmware that I wrote. The final board will use the STM32F051K8U6 (so the same MCU with a different package) but for an easier soldering I chose the T version (LQFP package).

The problem that I'm facing is that when I program the MCU (with a firmware that simply blinks the led) the MCU goes in short circuit (I read 3 ohms between pin 1 (VDD) and 32 (VSS) ) and there is no way to recover the situation.

I have already 3 MCU "bricked".

In my experience there is no way to short circuit permanently an MCU just by firmware, even if the firmware is compiled for another package (that in my particular case has an IO pin where I have the GND connection).

I'm missing something important? Now I ordered some other MCUs to test a new project compiled for the exact chip type, but in the meantime I would like to know if there is an explanation for this behavior.

Thank you!

[Tesla DeLorean]

>>I'm missing something important?

Hard to know, present complete connectivity as a schematic.

All the supply pins need connecting. Analogue supplies needed beyond ADC/DAC as they power the POR,PLL,VCO

Supply paths shouldn't be run through the parts, need to have wire on the PCB

PC13,14,15 are in a low current/power domain. Not suitable for LEDs, for the most part.

I'm not aware of a failure mode like this. You can lock yourself out of parts by disabling debug pins, or get LDO/SMPS or VOS settings wrong. Odd behaviour if VCAP pins not got the right capacitance, and FLASH failing to program, or with incorrect wait states.

Failures as described sound like creating inappropriate current paths, and then much of it flowing.

[AreaSX]

Thank you for your answer and clarifications.

I'm currently working on an adapter PCB (from LQFP to 2.54mm pitch connectors) and some soldered wires, so obviously it is'nt the best solution. I will double check wirings and I will wait for the new MCUs to arrive to investigate further; the only very strange thing is that all electrical measurements before the programming shows "normal" values (by ex. the 3.3V on power supply) then after programming the voltage drops to 0.7V and the current consumption rises above 2A. I think I'm making some big mistake but I cannot find it!

I will provide you the electrical schematic as soon as I return to office on Monday. 

Thank you again!