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.

DS90UB960-Q1: Tradeoffs and design differences between Twisted Pair and Coax

Part Number: DS90UB960-Q1

Hey FPD-Link team,

I have a customer evaluating the DS90UB960-Q1 and we are hoping to get some guidance from your team on things to consider when evaluating twisted pair versus coax for FPD-Link III using UB960 <--> 4xUB953’s:

  • Do you have any advice or information on the tradeoffs involved with twisted pair versus coax and what the benefits/tradeoffs are for each?
  • Currently there is a board designed for PoC with UB953/UB960; what would be involved with modifying the design to test a twisted pair setup?

Any advice or resources you can provide would be greatly appreciated!

Best regards,

Matt Calvo

  • Hi Matt,

    Q1: Do you have any advice or information on the tradeoffs involved with twisted pair versus coax and what the benefits/tradeoffs are for each?

    • Most customers design using coax. With coax configurations the N trace of a differential pair is terminated by a 50 ohm resistor and the P trace is routed all the way to the connector. In twisted pair configurations both the P & N traces of the differential pair are routed to the connector. This would require different connectors to be used in STP vs coax configurations. Another difference between the two configurations is STP would require double the PoC networks because you would no longer be terminating the N trace of the differential pair.

    Q2: Currently there is a board designed for PoC with UB953/UB960; what would be involved with modifying the design to test a twisted pair setup?

    • To modify the design to test a twisted pair setup you would follow the typical applications STP diagram from the UB953 & UB960 datasheets. 

    Best regards,

    Anthony

  • Thanks for the reply Anthony! 

    I will go ahead and relay this info to my customer and have them comment back on this thread directly with any additional follow-up questions they may have.

    One additional item we were hoping you could help provide advice on is types of twisted pair cables that your team recommends using in place of the coax that is currently being used. Do you have any specific types or versions of twisted pair cables that are used in your testing which you recommend customers use as well?

    -Matt Calvo

  • Hi Matt,

    Our team recommends using for the DACAR686-3 STP cable.

    Best regards,

    Anthony

  • Thanks for the info Anthony!

    As a follow-up question, we wanted to check with your team to see if it was theoretically possible to take a design based on PoC and change it to STP without re-spinning or redesigning the PCB? Is there a twisted-pair cable in the market today that can be used in place of coax without needing to redesign the board?

    My assumption here is that this is likely not possible but wanted to run it by you to see if there are any workarounds or alternative solutions that I'm not aware of.

    -Matt Calvo

  • Hi Matt,

    It is not theoretically possible for the customer to take their PoC designed PCB and use it for an STP design. If the customer wishes to use STP instead of PoC they will need to redesign their board following the STP typical applications diagrams in the datasheets. Most of the PoC designed PCB will remain the same in the STP designed PCB.

    The main reason the customer cannot just reuse their PoC designed PCB is because with an STP design you are no longer terminating the N trace with a 50 ohm resistor. Instead you are sending the N trace through a PoC filter as you would with the P trace in a PoC design.

    I hope this helps.

    Best regards,

    Anthony

  • This makes perfect sense Anthony and I appreciate the added explanation.

    I figured that would be the case so thank you for confirming! 

    -Matt

  • Hi Matt,

    Of course!

    Best regards,

    Anthony