stm32g0b1 swd interface not working

[prekop]

Hello community,

we made a nice project with TouchGFX and STM32G071C8. We needed increase flash memory. We change mcu to STM32G0B1CE. 

I cannot this mcu programming over SWD. I use various STM ARMs, has never been such a problem. I test two new mcu with the same result. I checked waveform signals, reseting modes, speed, programmer FW, i have read a test all advice from forums. I decode protokol - without response from MCU. - No STM targed found.

Response on SWDIO from mcu is nothing (it is still HI). The mcu is only on power supply, all GPIO are free, but consuption is interesting: 12mA, by holding NRESET  was 1,5mA. Both pieces.

Please can You me help?

[Peter BENSCH]

Welcome @prekop, to the community!

Which hardware tool do you use to program the STM32G0B1CE?

Regards
/Peter

[prekop]

Thanks, i use original ST-LINK/V2, then also ST-LINK/V2 M89. The waveforms  look good.

Data flow:  60 SWCLK pulse, then inicialization message (switch JTAG to SWD 0b0111100111100111) then 60 pulse, and then message READID without reaction. I compare on same board with STM32G071C8.

 

[Peter BENSCH]

If these are not the counterfeits in colourful tin cans, can you please post a photo (front and back) of your ST-LINK/V2.

[Peter BENSCH]

The ST-LINK/V2 in the tin can is a counterfeit that cannot be supported by the software tools from STMicroelectronics because it is built differently and there is a fake STM32 in there, usually even with different or no marking.

Try updating the firmware of the ST-LINK/V2 in the white housing with the current STM32CubeProg. Does it show a serial number when you try to connect?

[prekop]

Thank You for reply. I update one more time firmware to last version V2J41S7 without results. I test also debuger from Nucleo board. All three debugers have unique ID. All three debugers have the same start serial sequence (SWD). 

I don't know how to proceed further, I'm desperate.

[Peter BENSCH]

The most recent firmware version is V2J42S7 for both ST-LINK/V2 (Stand-alone) and ST-LINK/V2-1 (NUCLEO64).

[prekop]

I test V2J42S7 also V2J43S7 without success. Without response on SWD sequence.

[Peter BENSCH]

Please try to tackle the problem step by step:

1. you can check with your NUCLEO with ST-LINK/V2-1 and connected jumpers on CN2 if the STM32CubeProg can connect to it
2. then remove the two jumpers on CN2 on the NUCLEO, connect the ST-LINK/V2 to the target MCU of the NUCLEO (SWCLK is accessible at CN2.1, SWDIO at CN2.4, NRST at CN6.3, VDD for T_VCC at CN7.5 and GND e.g. at CN6.7) and check the same also in STM32CubeProg (if T_VCC is not connected, you should work with Software Reset)

If an ST-LINK can be connected to STM32CubeProg via Connect, it is initially functional in principle. The next step would be to check the software (version numbers?) with which you are having problems.

[prekop]

Hello after a long time i tested programing over SWD these STM32G0:

3x STM32G0B1CET6 (TQFP48) from different suppliers - without succes

STM32G0B0CET6 (TQFP48), STM32G0B1CEU6 (QFN48), STM32G0B1RET6(TQFP64), STM32G0C1CET6 (TQFP48) - everyone worked.

So I was unlucky. Also we tested hundreds STM devices in machines. We have experience that when testing in 60°C for 24°C, then it breaks down, for no reason, some periphery (ADC, GPIO..) to 1% MCUs.

Thank You for Your help!

 

[Peter BENSCH]

Well, you shouldn't put several different problems in one thread because you can't answer them together and later readers get confused.

To stay with the difficulties with programming via SWD: there is no logical connection to the programmability with different STM32G0B/C if you have used an original ST-LINK. The only other idea I currently have is that RDP Level 2 is already set on the ones that don't work.

Regards
/Peter