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.

How to decide using retimer or redriver?

Hi TI experts,

In design concept stage (at this time, it may be no layout placement and routing), which key design constrain (or condition) is for selection of retimer or redriver? (trace length, or?) Thanks!

Jack

  • Hi Ladisch,

    Thank for the information! It needs depending on insertion loss, right? Seems we still need have draft PCB routing to estimate it.

    In Figure.5 of application report, it separated 3 portions (approaching limit, marginally over limit and well beyond limit) to consider retimer or redriver. How to define ">20dB", "<35db" and ">35dB"? Many thanks!

    Jack

  • Hi Jack,

    Apologies for the delay.  In terms of defining what these different categories, this will in part depend on your application and the signal conditioning devices available for your application.

    In the application note, 20 dB denotes the receiver equalization capabilities of a typical Ethernet ASIC.  This figure will vary on the ASIC in your application.  35 dB is generally around the maximum equalization that a redriver can offer.  Beyond this, a retimer should be used.  A retimer also is capable of reducing jitter in the signal chain, while a redriver can only compensate for ISI.  Retimers are often considered for insertion losses <35 dB if meeting jitter specifications is challenging.  In general, simulation can typically be done with IBIS-AMI models for redrivers/retimers with a model of the channel in your system.  This can help ensure that the correct decision is made when selecting a signal conditioning device.

    Thanks,

    Drew