• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 34 subscribers
  • Views 728 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

SN75176B help

Oleh Kernytskyy1
Prodigy 55 points
Other Parts Discussed in Thread: SN75176B, SN65176B

We have problem with RS-485 communication for the new design.It appearthat previous design did not have this problem. It is electrical motor control system with MAX3483EESA on the console/computer side and SN75176B device on the motor controller board.MAX3483EESA is powered from 3.3 V voltage regulator, and SN75176B is powered from 5.0 V voltage regulator. Communication line is 100" ling cable (non-twisted pair for RS485 +ground wire). Communication data rate is 38.4 kb/s. Some products work fine, another - does not work, because RS-485 communication dropps. It was determined that 10kOhm pull-down resistor from RS-485- to ground wire fixes this problem. To make any remedy, i would like to determine why we have this problem. It seams that we did not experience this problem with the prvious design. I still did not have time to completely compare previous design to the existing one to find, if we made any significan changes. What is the best way to move forward and to determine possible reason for this RS-485 failure? Any other suggestions to debugg it? I disconnected communicaiton cable between MAX3484EESA and SN75176B and measured voltages on SN75176B RS485- and RS-485+. They are very repeatable, about the same for the working and non-working products, and are approximately 2.4 V.  I also measured differential voltages between RS485- and RS-485+, and it is approximately 6 V, when communication cable connect both tranceivers. What other torubleshooting steps can I take?:

  • 0
    TI__Mastermind 28265 points

    Oleh , despite your cable is only 100" long (100 inches = 2.5m) you might have large difference in gorund potentials between driver and receiver. Using a ground wire can cause high loop currents then. Any external noise couplying into ground will appear on your signal lines and corrupt your data transmission.

    1) Please disconnect your ground wire on one side (which one does not matter to break the ground loop). 2) Also remove the 10k pull-down from ground. 3) Start data transmission and measure the signals at the A and B bus terminals of the MAX3484 with regards to ground of the MAX3484. Then calculate the differential signal or use the Math function of your digital scope if you have one. 4) Do the same measurements at the A dn B bus terminals of the SN65176B with regards to ground of the SN65176B and determine the differential signal.

    If no communication possible, try conneting a 100 ohms resistors into the ground wire and connect each end to its transceiver. This might limit the ground current.

    Please provide me with your scope shots and a schematic if possible. It is much easier to see what you are doing.

    Thank you, Thomas

  • 0 in reply to Thomas Kugelstadt
    Prodigy 55 points

    Thomas,

     

    Thank you!

    Disconnection ground on one side or connection a 100 Ohm resistors into the ground wire is not an option, because ground wire is not just for RS-485 communicaiton, but also for 12 V power supply. Using RS485 communication ground as a ground for the power supply is the possible reason of failure.

     

     I spoke with design engineer today, and he mentioned that differential voltage in the Idle state may changed from - 1V to + 3 V.

    Because of this, we are loosing the first communication bit.

    He informed that he made a lot of calulcations and determined that 10 k pull-down resistor (and the same pull-up resistor in the future design) will help to eliminate this problem.

    I am not sure if he determined correcly root cause of failure and proposed an appropriate remedy.

    I will try to make separate ground wire for RS-485 communication only, if it will be possible with the existing PCB layout.

     

    Oleh

     

  • 0 in reply to Oleh Kernytskyy1
    TI__Guru**** 165036 points

    Oleh,

    If the line is idle with no drivers enabled then the line can float.
    A pull up resistor on one line and a pull down resistor on the other line will hold the un-driven bus in a known state.

    See The RS-485 Design Guide: Page 3 Fail safe section.
    http://e2e.ti.com/support/interface/industrial_interface/m/videos__files/264978.aspx

  • 0 in reply to Thomas Kugelstadt
    Prodigy 55 points

    Thomas,

    I am new for this forum. Originally, I contacted TI to get Technical Support, and TI passed me to this forum.

    How I can E-mail to you?

    Thanks,

    Oleh