3 Replies Latest reply on Mar 5, 2018 7:33 AM by weston_beal

    Eye diagram before + after rx equalisation

    pjs

      I'm trying to examine the eye diagram of a signal before and after the rx equalisation. The main reason is to ensure that  we meet the minimum eye opening at the receive pins as specified in the datasheet. This is usually specified before the receiver equalisation is applied and so I'm trying to confirm that with our channel design that this will be possible.

       

      I can't see how to do this in Hyperlynx as I'm finding that regardless of where I place the probes on the transmission line, the equalisation parameters in the receiver always affect the eye very substantially.

       

      Is there a way in the AMI Channel tool to see what the eye diagram looks like before the Rx equalisation is applied?

       

      Thanks

        • 1. Re: Eye diagram before + after rx equalisation
          weston_beal

          PJS,

           

          When you run the AMI Channel Analyzer, you can request a set of additional waveform files for debugging. In the View Analysis Results page of the wizard, find the _debug_info_ setting at the bottom of the page. Change this setting to Extended.

           

           

          Then when you run the AMI analysis, the tool will write a bunch of LIS files in the design folder. I think that the file named BeforeRxGetWave.lis is the one that you are looking for. You can use either waveform viewer (Digital Oscilloscope or EZ-wave) to open this LIS file and show the waveform as an eye diagram.

           

          Regards,

          Weston

          • 2. Re: Eye diagram before + after rx equalisation
            pjs

            Thanks Weston - very helpful.

             

            Just to confirm the BeforeRxGetWave.lis will presumably taken at the probe point? Does this waveform capture include the effects of the transmitter AMI settings - e.g. pre-emphasis?

            • 3. Re: Eye diagram before + after rx equalisation
              weston_beal

              PJS,

               

              I'm not certain of that, but it makes sense. You should confirm by experiment. I would run a few simulations with different TX settings and observe the specified waveform. This way you can see if the TX behavior is captured in that waveform.

               

              Regards,

              Weston