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.

SN65HVD11: SN65HVD11QD Star Connection

Part Number: SN65HVD11
Other Parts Discussed in Thread: THVD1450, THVD1410

Hello to TI interface support team ,

           I use SN65HVD11QD part in  connections between PCB's and in to PCB of my project , according to figure attached ( Star connection with short distances In figure 50cm to each PCB , 10cm into PCB distance ) . Can you please recommend places locate termination resistors , there is 13 units of  SN65HVD11QD i use connect together. 

         1.  Mode  operation #1 : 4 units connected together in one PCB , 3 PCB connected together on parallel connetion on main PCB ( #4 on figure ) without PC . 

         2.  Mode Operation #2 : PC connection to the system of 13 X SN65HVD11QD ( according to figure below) .



Alexander Burtakov


  • Hello,

    The location will not be so critical for such a short distance network. The delay through the connections will just be a few ns, which is shorter than the transition time of the differential signal. I would recommend trying the termination at convenient location at the ends of the cables. Generally termination would be placed at two different end points of the network.


  • Hello Max ,

        Thanks for your quick response ,

    1. I make re design now and my core question is in so relative  small  485 network distance of  13 X IC's  for this bus is it importants Star network connection (Backbone with stars) as in my figure above or Daisy Chain Bus connection : from PCB 4 to PCB3 after it PCB2 and PCB1 and back to PCB main #4 ( ~2.5m of chain ) ? .

    2. IS SN65HVD11QD most suit to this network connection ?


    Alex Burrtakov

  • Alex,

    I don't think you would see much difference between either approach.  The daisy chain approach would result in no stubs at all and so reflections could be totally mitigated with termination at the far ends.  But, the reflections generated from stubs of just 50 cm would be fairly minor as well and so if the star connection is more convenient in your application then I don't think you need to change it.  Is there a way you can try it out and measure the RS-485 signals to verify they look OK?

    By the way, how much of an impact a stub reflection has on the signal depends on the how the propagation delay down the stub compares with the rise/fall time of the signal.  For this reason, RS-485 transceivers often come in families with different rise/fall times on the driver outputs and, correspondingly, different supported maximum bit rates.  Immunity to stub effects is typically greatest when choosing a transceiver only a little "faster" than the application requires.

    That said, to help recommend a best part for you it would be helpful to know the intended operating bit rate.

    In general, though, I would recommend a newer device like THVD1410 or THVD1450 over the earlier-generation SN65HVD1x devices.  They make some improvements in VCC range supported, operating voltage ranges on A/B, ESD immunity, and package availability.


  • Hi Max,

          Thanks for your response , 

    The daisy chain approach would result in no stubs at all  --> This configuration possible to get an example connection like that .  Chain connection in to each PCB and between PCB's overall  . 2 Resistors  termination on far sides of the BUS . no stubs at all.

       Figure i atached above is combined topology  : Star as each PCB , 3 rays of star as 3 PCB's and daisy chain into each PCB .





  • Hi Alex,

    Thanks for sharing the updated diagram.  What you've shown is an ideal signal routing since it is completely daisy-chained between each node, giving a simple linear topology.  I wouldn't expect any signal integrity issues with this connection.

    If this wiring is difficult in practice then I think you would still have a working system even with a non-ideal set-up (since the distances are short), but you would be best to test it out to confirm.