cancel
Showing results for 
Search instead for 
Did you mean: 

STM32H723xx won't program at 3.3 volts but will program at 3.2 volts

CClar.3
Associate II

I am using a STM32H723ZGT6 on a small project. The project fails to program when I am using 3.3 volts. It programs just fine at 3.2 volts. Always gives me a fail in final launch sequence. However, it always programs when I apply a small amount of heat to the MCU. Here are the schematics of the MCU and the fail message I receive.

I am asking if anyone knows why it works at 3.2V but not at 3.3V. And also why it works with heat but not at room temperature.

0693W00000aHsufQAC.jpg0693W00000aHsuaQAC.jpg

2 REPLIES 2

VCAP is supposed to have a pair of low ESR 2u2F capacitors, or one 4u7F

What model ST-LINK are you using?

How is it wired?

Do you have similar difficulty when using the GUI STM32 Cube Programmer Directly?

Tips, Buy me a coffee, or three.. PayPal Venmo
Up vote any posts that you find helpful, it shows what's working..
CClar.3
Associate II

These are good questions.

1-I have changed out the VCAP capacitors for 2.2uF with no effect. I have changed out every component for different values to no effect. It always performs the same. I can even program using a 10uF on the NRST line if the voltage is 3.2 volts. But it won't recognize the chip at all if I don't use a capacitor there.

2-I am using the STLink V3Set for programming. I also have a V2 but I think it's a knock-off. Regardless, using either programmer, it exhibits the same problem.

3-I have a custom made PCB that I designed to attach directly to the V3Set. This gives me the ability to use a Tag-Connect TC2050 connector. Everyone in the office is using this setup. We have never had any problems with anything we've programmed until now. This also allows me to use the UART of the device. Here is a picture:

0693W00000aHt67QAC.jpg4-Regarding STM32CubeProgrammer... I hadn't tried that until you mentioned it. I attempted to program the device but it never got past the flash erase. I restarted the programmer but unchecked 'skip flash erase before programming' and it worked. This means that there is a problem with the flash erase somewhere along the way.

As you can see, the PCB is extremely simple. I have a similar design for another project which is exactly like this one. And I've never had a problem with that other project PCB. I cloned the project and changed a few pinouts for this one. I'm now working on the second revision of the PCB as the first one had the same problem. This is weird. I wish I knew what to do. In the end, I will likely just run the board at 3.2 volts but I feel like that is a band-aid to a greater problem.