3 Replies Latest reply on Nov 30, 2015 2:33 PM by user_1377889

    Timer vs USB_UART?

    neal.o'hara

      Does anyone have any idea why a 16 bit fixed function timer, that interrupts about every 4s,  would interfere with USB_UART operation?  

         

      I'm debugging my current boards  (with a CY8C5667LTI-LP041) ,  which show up on Device Manager on an appropriate COM port, but will not connect to putty (and yes, I've double checked my serial settings) .

         

      I've turned everything in code off one-by-one until USB_UART is permanently on.  

         

      With the Timer_Start(),  the USB will not connect, but if I comment that out, USB will connect to putty again.

         

      The odd thing is that I have another timer, a 32B UDB timer, that does not interfere with the USB_UART at all.

         

      Both timers are software enabled, have dynamic periods based on software, and cause external interrupts. 

         

       

         

      Any advice would be appreciated.

         

      Thank you

        • 1. Re: Timer vs USB_UART?
          user_1377889

          I've never seen something similar.

             

          How is the connection from Rx/Tx to your PC made? Level shifter?

             

          Is it possible for you to post your complete project, so that we all can have a look at all of your settings? To do so, use
          Creator->File->Create Workspace Bundle (minimal)
          and attach the resulting file.

             

           

             

          Bob

          • 2. Re: Timer vs USB_UART?
            neal.o'hara

            Further investigation shows it is the 25ms worth of blocking adc/eeprom action right after the Timer Interrupt that breaks the UART communication.
            Also, my exampe USB_UART code was written polling, so mine is polling instead of interrupt driven.
            I'll need to investigate changing that.
             

            • 3. Re: Timer vs USB_UART?
              user_1377889

              I am not quite sure, but I remember that during eeprom write clocks are changed.

                 

              25ms in an interrupt handler is a very long time

                 

               

                 

              Bob