• State Resolved
  • Locked Locked
  • Replies 2 replies
  • Subscribers 62 subscribers
  • Views 265 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

Original question:

How can I catch all fault-states when using dpwmfaultctrl ?

CCS/UCD3138: The problem with the Fault register. FAULTMUXINTSTAT not react

Rutskoy Andrey
Intellectual 465 points
Part Number: UCD3138

Tool/software: Code Composer Studio

Hello!

Have a problem with flags register FaultMuxRegs.FAULTMUXINTSTAT.all (UCD3138A) . It does't indicate the events. I know that it clean-read register. I already commented out all lines where I read this register (just use in init).

If I use FAULTMUXRAW all is fine, all events detected and programm works properly. For testing I use LEDs also and the result the same as from GUI. Always zero

The signal on the inputs of the comparators is stable, I use DC reference level now for testing. 

Why is register always empty and no reaction to events? How it could be cleared if I don't read it anymore?

  • 0
    TI__Mastermind 18370 points

    The memory debugger (GUI) read from FAULTMUXINTSTAT can also clear it.

    You may want to use DPWMFLTSTAT as an alternative for detecting your fault and respond to it. And, of course, drive the fast interrupt with the DPWM interrupt, rather than the Fault Mux one.

    DPWMFLTSTAT is latched until you clear the DPWM fault and is not clear on read.

    The fault logic FAULTMUXRAWSTAT only samples every 32 nsec, while the DPWMFLTSTAT samples every 4 ns. So a short pulse may be missed by the fault interrupt, but not by the DPWM. 

    Also if an interrupt arrives at the same time as a read, the clear on read has priority, and you will never see the interrupt. 

    If you have two fault sources, and you need to determine which one has caused the fault.

    What we do is to run one fault source to FAULT_AB, and run the other one to FAULT_A and FAULT_B of another DPWM. This way both DPWMs get shut down, and you can still tell which source caused the fault. That way the slower detection and clear on read for the FAULTMUXINTSTAT are completely avoided.

    Regards,

  • 0 in reply to Yitzhak Bolurian
    Intellectual 465 points

    Dear Yitzhak,

    Thank you for your support! With DPWMFLSTAT works correct.

    I think in manual wrong. It is not clear on read bits as I see.