I am using STM8S003F3 & attached is circuit.
I have connected some switches in pull-up configuration as shown.
After some correct operations, I have seen that some of boards stopped working.
It was problem 1 as below in damaged boards & problem 2 in one of damaged board.

1. On debugging, I had checked that voltage at reset pin is around 2V.
This keep the device in reset state & won't allow running.
I don't understand why this is happening. All the voltages that I have applied are within range.
I have also removed the RC network across reset pin, but still pin is showing 2V approx.
What could be possible reason?

2. On one of switch pin, either I give 0V or 5V, software always read zero.
However when I configure same pin as output, I am able to toggle the pin.
But as soon as configured as input, it always read zero.
All other switches are working fine i.e can read them properly.
What could be possible reason?

3. Board works fine for some operations but suddenly reset pin problem occur after that.

 

If the reset pin is being physically damaged, you might need a schottky diode across R1. Sometimes when powering down, a capacitor on a pin (C4 in this case) holds the pin voltage up while Vdd is going down exceeding the maximum rating of the pin voltage vs. Vdd and damaging the input. Some uCs have protection for this, others don't - particularly ones that use the RESET line during programming.

Your device specifies a max Vin>Vdd of .3Vmax with limits on how much current can be injected into the pin without breaking it (+/- 4 with no units??!). You might consider a resistor in series with the RESET pin to keep that under control.

Good luck.

 

@JohnInTX , I guess that was issue.
I had removed 10K resistor across reset, now module is working from last day.
However I will check for 3-4 days before conclusion.

That was 4mA that you had mentioned. I will also put a series resistor of 10K to check.

 

@JohnInTX , I guess that was issue.
I had removed 10K resistor across reset, now module is working from last day.
However I will check for 3-4 days before conclusion.

That was 4mA that you had mentioned. I will also put a series resistor of 10K to check.

I am not sure you understood me - or I am not understanding you. To be sure we are on the same page:
R1 and C4 are OK in the circuit.
I would ADD a 1N5817 or similar shottkey diode in parallel with R1 with the anode (banded end) to +5V. That will take care of the first problem of the pin > Vdd during power down.
I would also consider adding something on the order of 1.5K in series from R1 || 1N5817 - C4 to the RESET pin to address the injection current issue.
The NRST pin on the programming header should connect directly to the RESET pin to avoid possible problems with the programmer caused by the reset network.

Hope that does it.

 

1. I think you were talking about ckt as I have attached.

2. Currently I have removed all reset circuit. Only C4 remain i.e decoupling cap to reset pin, as reset pin internally pull-up by around 40K in MCU.

 

1) Yes. That is what I had in mind.
2) See how it works..