Odd signal from PIC chip

Discussion in 'The Projects Forum' started by mikekehrli, Jun 30, 2015.

  1. mikekehrli

    Thread Starter New Member

    Oct 2, 2012
    24
    3
    Hi Guys,

    I have been doing my nut over an odd signal I'm getting on a new board I've designed. Basically the design is (supposed to be) an improvement of an existing design. Part of the design is communication with another PIC chip across an RJ-45/Cat-5 cable. The communication is working fine on the new design, except for on one channel. I'm getting these spikes in the signal that are causing errors in comm. I can't figure out where they are coming from. I was hoping someone here had seen the phenomena before or could otherwise, through deep magic, point me in the right direction.

    See the 2 photos. Good signal and bad signal.

    There are 4 data channels and one clock channel. On the old design, all data channels and the clock channel look like the good signal. On the new design, the clock and 3 of the 4 data channels look like the good signal and one of the data channels is the bad signal picture. I'm using everything the same when testing both the new design and the old design. The only difference is which board is plugged in. I've pretty well ruled out any possibility of anything else causing the problem other than the new board. It appears that the board matches in every respect, the old board. Yet these pulses appear.

    I'm basically clutching at straws here. Can anyone tell me anything about that bad signal and hazard a guess as to where to find the problem?

    Good Signal:
    good_signal [1024x768].jpg

    Bad Signal
    bad_signal [1024x768].jpg
     
  2. mikekehrli

    Thread Starter New Member

    Oct 2, 2012
    24
    3
    Guys, stand by a few minutes on this. I think I already solved it. This is not the first time I've figured a problem out after assembling the data for a forum post.

    This is cross talk. And now I've found there is a difference in the new design. There's a 4.7K resistor on the input line, allowing the capacitance in the wires to cause the spiking. The old design doesn't have that resistor. I'm going to lower the value to 100 ohms. I think that will solve it completely. I'll re-post to verify.
     
  3. mikekehrli

    Thread Starter New Member

    Oct 2, 2012
    24
    3
    That question is solved. This was cross talk, and it's virtually gone now. Sorry for the premature post.
     
  4. atferrari

    AAC Fanatic!

    Jan 6, 2004
    2,648
    764
    And what is that new resistor doing that seemed not necessary in the previous board?
     
  5. GopherT

    AAC Fanatic!

    Nov 23, 2012
    6,058
    3,820
    Resisting his efforts. What fun is a success without a few challenges along the way.
     
Loading...