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.

SN65HVD75: The maximum stub lenght in 115200bps

Part Number: SN65HVD75
Other Parts Discussed in Thread: SN65HVD72, THVD1410

Hi,

    I have a question about the stub length of SN65HVD75. In an application I will connect 8 devices through RS485 bus,and using SN65HVD75 transceiver. Baud rate is 115200bps in the  application. The stub length is 0.4m. I saw the requirment of the SN65HVD75 in datasheet is 0.05m, so I want to know the maximum stub length 0.05m if is  the essential condition when  transceiver runs on 20Mbps or any slower baud rate?

  • Hi Harvy,

    The datasheet defines maximum stub length based on the device's rise/fall time (tr, tf). Because this value is independent of the data-rate that is being driven, changing the baud-rate will not allow for a longer sub length. In order to increase the suitable stub length for your system, a driver with larger rise/fall times may be used. The SN65HVD72 is listed in the same datasheet with a max stub length of 7 meters. The THVD1410 is a recommended alternative part for the SN65HVD72 which will have similar rise/fall time characteristics. 

    I hope this helps. Let me know if you have any more questions.

    Regards,

    Eric

  • Hi Eric,

         Thanks for your answer. On the other hand, if a lower rate is used, does the reflection or deterioration of the signal have little effect on the integrity of the signal? Because in some applications, I would also increase the baud rate to 921600bps, but using point-to-point connections.Therefore, I do not want to replace the transceiver to simplify manufacture.Could you give me some suggestions?

    Regards.

    Harvy

  • Hi Harvy,

    You are correct regarding the relative effect of these reflections on a low-frequency signal. The reflections and ringing on the bus due to the stub may cause glitches in the receiver near the transition. Depending on the characteristics of the controller, these glitches could be filtered or ignorged. As long as extra edges of potential glitches near the bit transition would not pose an issue to your system, I don't think that having a stub of this legnth will be a problem.

    The calculations for Lstub on the datasheet are also pretty conservative. You may also want to include some high-z termination on the stub nodes to decrease the magnitude of these reflections.

    Regards,

    Eric

  • Hi Eric,

            Thank you for your reply, I got it.

    Regards,

    Harvy