2015-02-28 11:37 AM
Hello,
I am facing very strange behaviour with uC mentioned in the topic. After connecting the microcontroller to the 3V3 supply and the SWD (IO, CLK, NRST and GND) on the Nucleo board, I get this error in ST-LINK Utility: ''Internal command error''. After a lot of trials I've found out that after lowering the supply voltage below 3V, the debugger is connecting properly.I've tried different board as a programmer (F4 Discovery), even microcontrollers from 2 different suppliers, without luck. Of course supply pins are properly connected and with decoupling capacitors. Pulling up or down BOOT0 and NRST didn't change a thing.What is cause of this behavior? I need to use 3.3V supply on my board, <3V is a no-go for me. #stm32f0302015-03-01 04:31 AM
Does the target board have it's own supply? If so only common the ground.
The error is indicative that the part isn't working. Do other signs of life tests, like the System Loader via USART respond?2015-03-02 05:28 AM
Thank you for response,
All parts I've tested so far (5 pieces or more, on 3 different boards) behave the same way. They indeed work (when programmed at lower supply voltage). The bootloader works at 3.3V, but ST-Link Utility is giving those errors. I have even tested other debugger - J-Link, still not working at 3.3V. I've found more threads on the web with this problem but no solution at all. Is there some bug in the silicon? Maybe I have some early revision chips?EDIT:The board is supplied by DC/DC converter, of course only GND and SWD signals are connected to Nucleo. Testing with 3.3V from Nucleo gave the same result - ''Unexpected error''What I've found is that the threshold when the problems start is around 3V.2015-03-02 06:16 AM
I've found more threads on the web with this problem but no solution at all. Is there some bug in the silicon? Maybe I have some early revision chips?
This is a user forum, if you think you have chip problems you should discuss these with your local ST rep or FAE. Bug and revision data that's documented will be in the Errata for the chip.