This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

  • Resolved

MSP430F6736A: unexplained MCU resetting

Expert 1900 points

Replies: 2

Views: 59

Part Number: MSP430F6736A

To any who read this post & have a similar issue....

I am developing a residential power meter.

Last year i had an issue with my custom board resetting whenever i had a series of high frequency EMF interference due to the power supply relay opening & closing.

At the time, i associated this issue with the fact that the voltage at the reset pin had dropped below 2V for over 200uS (observed via a scope).

Whilst a few suggestions were put forward, none resolved the issue & much effort went into reducing the effects of EMF from the relay.

This year i found a similar issue when communicating to a new i2c memory chip at higher i2c speeds (observed via a logic analyser).

As a consequence of solving the i2c issue, i have now determined that the cause of the resetting problem with the relay was the EMF affecting communication on the i2c bus...specifically when setting the address to read/write...& which resulted in timeouts that caused the WDT to fire.

The following cases all relate to the above determination...

https://e2e.ti.com/support/microcontrollers/msp430/f/166/p/796606/2951720#2951720
https://e2e.ti.com/support/microcontrollers/msp430/f/166/p/795637/2943237#2943237

https://e2e.ti.com/support/microcontrollers/msp430/f/166/p/934027/3452561#3452561

  • Hey Moshe,

    Very glad you were able to find the source of your issue, Good job! 

    Thanks for providing an update and closing the loop with the forum.  Debugging system-level issues like this are usually very difficult.  

    Regards,

    JD

    _____________________________________________________________________________


  • In reply to JD Crutchfield:

    Hi JD

    Yes, the problem of the i2c being the cause of the timeout was not, at the time, contemplated as the degradation of Vcc on the reset pin shouted out from the scope & hence 2+2= 5...i think thats commonly referred to as being a cognitive miser.

    However, now armed with my new found knowledge, should anyone in the future mention such issues, naturally i2c coms will come to the fore & it may indeed prove to be wrong for that situation.

    :)

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.