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.

Pl. refer following document "TI Home » TI E2E Community » Blogs » Analog Wire » Commandments of RS-485"

Other Parts Discussed in Thread: SN65HVD82

Pl. explain "local ground", If a cable from node1 (point no.2) to node2 (point no.9) uses only 2 pair, 2 pairs are free
then, where to connect 2 free pair ?  at node1 ground OR node2 ground ?

5, 6 should be on master side ?

is there any chart (as like as no.7) for determine stub length ?

Any substitute for SM712

How to calculate RC low pass filter value

  • Hi Kishor,

    It is recommended not to connect the free pair to ground on both modules (both ends of the wire). This is due to the fact that this can create ground loops. You should connect the pairs to ground on one module only.

    Point 5 & 6 refer to the fail safe biasing termination style. We are recommending that you use this type of termination on one side of the bus. The other side should have standard termination.

    The maximum stub length is based on the rise time of the driver. The faster the drive time, the shorter the stub lengths you can have before reflections start to cause an issue with signal integrity. The rule of thumb is that the propagation delay of the stub needs to be less that 1/10 of the signal rise time. As you can see you will have to decide on a transceiver before this be any use to you.

    What is wrong with the recommended SM712? This device is designed to be used with RS485 systems.

    The corner frequency with the lumped RC value of the system is 1/(2*pi*R*C).

    Thanks,

    John

  • Hi john,

    There is no problem with SM712, only issue is availability.

    Instead of calculating of stub length, If i create double side PCB track as like as twisted pair then ?

    will it work ?

  • Hi Kishor,

    Where are you located I found thousands of the SM712s available on DigiKey, Mouser and Newark (click them for the links).

    In terms of the stub length, if you are daisy chaining the nodes, then you do not need to worry about this. As I said before the rule of thumb is that the propagation delay on the stub should be less than 1/10 the rise time of the signal. If you estimate the propagation of a signal to be around 5ns/meter, any normal PCB trace length would not cause and issue. It should be negligable.

    To give an example the SN65HVD82 is one of our RS485 transceivers, and it has a minimum rise time of 400ns. Therefore, 1/10 of that is 40ns, so a stub should be less than 8 meters long at 5ns/m propagation delay.

    Thanks,

    John