MCU resets after couple hours

[Adistrict]

Hi, everyone

I have a weird problem, my stm32f103c8 chip goes to reset mode(nRST pin voltage goes below 1.5 v) after working for couple hours. The MCU is switching the GPIOB1(frequency is 500KHz). Do you have any suggestion about this issue? Is it a power up problem(could this be because of wrong voltage rise sequence in VDD,VDDA and nRST pines)?

Thanks in advance,

Amir

[Tesla DeLorean]

Sounds like a design issue.

Does VDDA fall below the POR Threshold?

[TDK]

What happens to VDD before/as NRST goes low?

What does the RCC->CSR register indicate as the reset source?

[Adistrict]

I'm attaching my hardware document. What seems to be wrong in the design?

[Tesla DeLorean]

That's really a job for you and your colleagues.

Would suggest you observed the failure, whilst measuring the temperature and voltage out of the regulator, and look at the potentials on VDDA/VSSA with respect to the ground elsewhere, and against the NRST. Look also at the input current, and that consumed by the MCU and attached logic

[Piranha]

What are we supposed to do with a 3D model? And you didn't answer any of the questions asked by Tesla and TDK.

[Adistrict]

Ok Piranha, sorry about this!! I was testing the board whole time.

And I just attached the CAD file so you can check the file too.

[Adistrict]

I changed the clock source to HSI(PLL MUX : 6, Core frequency : 24MHz).

So far so good, but I'm intended to use an external crystal resonator.

Do you think this is a clock malfunction?(I doubt that GND plane affects crystal waveform due to its capacitive effect)

[TDK]

I don't see any answers to the questions above. Attaching a schematic is going to be more useful than the 3D model.

[Adistrict]

Here you are, I attached all the documents that you need to investigate the project.