Triggering for an Eye Diagram

Discussion in 'General Electronics Chat' started by Sparky49, Apr 28, 2016.

  1. Sparky49

    Thread Starter Active Member

    Jul 16, 2011
    834
    417
    Hi all,

    I'm trying to setup my oscilloscope to construct an eye diagram of some UART connection, however I am unsure as to how to trigger it properly without a clock. Is it possible to trigger my oscilloscope (HP54542A) so as to produce a diagram without a clock source? If so, how? I have tried playing with the pattern, state, delay triggers, etc, but if I'm honest, I am simply guessing.

    I have managed to at least produce a portion of an eye diagram, see below. :oops:
    eye-ish.gif

    Many thanks for your time,

    Sparky
     
  2. kubeek

    AAC Fanatic!

    Sep 20, 2005
    4,670
    804
    Try triggering at midpoint, and triggering on both edge polarities.
     
    Sparky49 likes this.
  3. Sparky49

    Thread Starter Active Member

    Jul 16, 2011
    834
    417
    Hi Kubeek,

    Thanks for your reply. Would you be able to give a bit more detail, please? If it's scope specific, don't worry, but, for example, which type of trigger should I be using?

    Many thanks for your time.
     
  4. kubeek

    AAC Fanatic!

    Sep 20, 2005
    4,670
    804
    I would use triggering on edge, and set it to trigger on both rising and falling edges. Set the trigger level in the middle between low and high logic level, and move the trace a bit to the left so that you can see the whole eye.
     
    Sparky49 likes this.
  5. Sparky49

    Thread Starter Active Member

    Jul 16, 2011
    834
    417
    Many thanks for your reply again. However, I cannot seem to see a way of both rising and falling edge triggering on the HP54542A, so would it be fair to say that it is impossible to perform on this instrument?
     
  6. Sparky49

    Thread Starter Active Member

    Jul 16, 2011
    834
    417
    This was the best I could figure out.
    eye-ish2.gif
     
  7. kubeek

    AAC Fanatic!

    Sep 20, 2005
    4,670
    804
    First you don´t want it to trigger automatically, but only at the trigger condition. Second I am not sure if it is truly triggering on both edges, can you somehow verify that?

    Edit: yes, it looks like that the scope cannot do triggering on both edges.
     
    Sparky49 likes this.
  8. Sparky49

    Thread Starter Active Member

    Jul 16, 2011
    834
    417
    Thanks for your time though.
     
  9. OBW0549

    Well-Known Member

    Mar 2, 2015
    1,328
    890
    No, it is simply not possible to get a meaningful eye diagram by triggering on the waveform you are trying to evaluate, regardless of the scope's triggering mode or threshold settings.

    Think about it: the scope triggers on a specific voltage level, and the point in time when the signal crosses that threshold is always displayed at a specific horizontal position on the scope screen, no matter the actual time that has elapsed since the last threshold crossing. Your signal could actually be jittering like crazy, and you would never know it because the scope will neatly line up all the individual traces on top of one another at the threshold voltage crossing points.

    Without access to a true UART bit clock signal to trigger on, your best approach will probably be to trigger on the signal at the transmitting UART's TxD pin, before the line driver chip. While that won't give you a complete "eye" diagram, it will at least give you some useable waveform quality information.

    To get a real eye diagram, you'll have to disable the UART on the transmitting end and feed the line driver chip with a pseudorandom binary sequence (such as from a LFSR) clocked at the same rate as the UART's baud rate, and trigger your scope on the LFSR clock signal.
     
Loading...