Why is the STM32 serial port 2 burned out, but other places are not burned out

shijizai

Why is the STM32 serial port 2 burned out, but other places are not burned out [Copy link]

 

Why is the STM32 serial port 2 burnt out, but other parts are not burnt out? The phenomenon is that the serial port 2 can be used at first, but later only the serial port 2 cannot be used, while other serial ports and other functions can be used. If it is burnt out, shouldn't the whole chip be burnt out?

ienglgge

If it burns out, shouldn't the entire chip be burned out? That's not necessarily the case. It's possible for a single pin to burn out. For example, a pin may be over-voltage, damaging a pin. If there's a problem with the power supply, the entire chip can be damaged.

shijizai

ienglgge posted on 2018-11-5 20:10 If it is burned out, shouldn't the entire chip be burned out? That's not necessarily the case. It is possible that a single pin is burned out. For example, a certain pin may be over-voltage and broken...

Oh, that's it, thank you very much

heningbo

Can you explain how it burned out? Without describing the phenomenon, how can others know whether it is correct?

elvike

It's winter now, and static electricity may discharge to the port

shijizai

heningbo posted on 2018-11-6 09:27 Can you explain how it burned out? Without saying the phenomenon, how can others know if it is correct.

There is no phenomenon. The phenomenon is that serial port 2 does not work. Move the program of serial port 2 to another serial port for transmission, and the transmission is successful.

shijizai

elvike posted on 2018-11-6 11:48 It's winter now, and it's possible for static electricity to discharge to the port

Yours is unlikely

shijizai

heningbo posted on 2018-11-6 09:27 Can you explain how it burned out? Without explaining the phenomenon, how can others know if it is correct?

The phenomenon is that the communication of serial port 2 was originally working, but then it kept failing. I moved the transmission program of serial port 2 to another serial port of the same microcontroller, and the transmission was successful.

wsmysyn

Not necessarily. It is possible to burn out a few IOs. You can imagine the chip as a PCB board. Some parts are broken, but others may be good. If all are broken, it would take a lot of violence to cause all of them to break down You can test the IO of the serial port to see if there are any changes. It is easy to test.

shijizai

wsmysyn posted on 2018-11-10 10:07 Not necessarily. It is possible to burn out a few IOs. You can think of the chip as a PCB board. If some parts are broken, others may be good. If all are broken, then...

Hmm

ketose

Use a logic analyzer to look at the signal of serial port 2 to determine what the problem is.

heningbo

Shijizai published on 2018-11-10 09:29 The phenomenon is that the communication of serial port 2 was originally good, but later it kept failing. Move the transmission program of serial port 2 to another serial port of the same microcontroller, and the result is...

Not necessarily, you have to make sure which one it is. If you use serial port 2, is there a chip similar to 232 outside, or directly use TTL level, which one is broken. Don't just say that the main chip is burned out and that's it.

heningbo

You can use an oscilloscope to see if TX has an output signal, and then do a loopback test. Then use something else to read it out.

shijizai

heningbo posted on 2018-11-12 16:14 You can use an oscilloscope to see if the TX has output signals, and then you can do a loopback test. Then use something else to read it out

Hmm

wsdymg

Check whether the chip pin is broken or the chip peripheral circuit is broken. You can use an oscilloscope to look at the waveform on the pin.

hjl2832

I have experience with this. I used to have a product that used F103 and 485 communication. Because the shell of the peripheral product was directly grounded, the TVS on the 485AB line directly formed a loop with the ground and leaked electricity to the IC in the reverse direction, burning the 485 chip (checking the burned chips, you will find that they are all seriously bulging and even the plastic packaging shell of the chip bursts directly); the reasons for the formation will be analyzed later (for outdoor industrial equipment, I strictly follow the national standard requirements that the shell is virtual ground (connected to the earth), the ground of the TVS is connected to the shell, and it is connected to the circuit ground through a balancing resistor). The peripheral equipment is powered by 24V and shares the same power supply. The peripheral equipment is connected to a power switch. At the moment the power switch is closed, if the negative pole lags behind the positive pole, the 24V will pass through the 485 TVS to the AB port of the 485 chip to form a loop with the ground (negative pole). The high current formed by the instantaneous high voltage will break down and burn the IO port of the 485, and the IO of the microcontroller is also likely to be damaged. The entire board is measured except for the abnormal function of the damaged serial port, and all other functions are normal, including the power output of the PWM control. And I have at least dozens of such broken chips in my hands (the equipment installers at the time did not understand the cause of the damage. When they found that the board was damaged by the switch, they replaced the board and tried again, which resulted in dozens of boards being damaged).