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.

RE: RS-485 Compatibility Between 3.3V-supply and 5V-supply Devices

Other Parts Discussed in Thread: SN65HVD08

HI, 

We are facing issue while we communicate to 5V Slave devices. Our Master is 3.3V Device And part No SN65HVD08 if we give request is not responding and the same slave device is responding it for the 5V operated Master Device. Could you give corrective action on this issue  it will be great full to you. Thanks 

Reg

Sakthi 

  • Hi Sakthi,

    I've split your question into a new forum thread, since the original topic that you responded to is fairly old. For this issue, are you able to observe the RS-485 waveforms on an oscilloscope and show us the results? It would be good to look into the communication both from master-to-slave (HVD08 transmitting) and slave-to-master (other device transmitting) if possible.

    Another thing to check is the DE control pin. This needs to be high only on the master when the master is transmitting and high only on the slave when the slave is transmitting. If multiple drivers are enabled on the bus at the same time it will cause communication issues.

    If you could provide further details on the cabling used (length/type) and termination scheme as well it may help us to better understand what could be going wrong.

    Regards,
    Max
  • Dear Max, 

    We are doing this DE Pin High, The same hardware is doing the request and response for the another hardware which is 3.3V Device Slave. Only not responding it for the 5V Device Slave. Can we try some 680 E Resistor to boost up the Current to match the 5V Device ?? please advise. Thanks 

    Reg

    Sakthi ivetel

  • Sakthi,

    What exactly do you have in mind regarding the 680-Ohm resistor? Would you use this as a termination resistor instead of 120 Ohms? That would successfully increase the differential voltage on the bus and so it may be worth a try, but I doubt that the signal amplitude is what is causing an issue. Any RS-485 driver (whether 3.3-V or 5-V) will output a differential signal of at least 1.5 V (across a 54-Ohm load), and any receiver should be able to properly detect high/low levels with as little as 200 mV of signal.

    Are you able to observe the bus waveforms? That would be a good way to see what exactly is happening that is preventing communication from working.

    Regards,
    Max
  • Max, 

    680E is for Pull up in A and Pull Down in B Line. This Pull up and Down will required? I Read in some of your Forum ,if need to match the 5 V Slave device master should have Pull Up in A and Pull Down in B for 3.3v Device 680 E and 5 V Device 4.7K . Let assume 3.3V Operated Master and 5 V Operated Slave without putting any pull up and Down will it work?? 120 E is for impedance matching if the cable distance is 1KM here we are using shielded cable approx 1.5 Mtr. 

    485 Means it has to work  on either 5V Device or 3.3V Operated Device. Please clarify in details reg this issue. Thanks

    Reg

    Sakthi ivetel 

  • Sakthi,

    Thanks for clarifying - now I understand what you mean. Pull-up and pull-down resistors are not required for RS-485 operation and are not needed for communication between devices with different supply voltages. The reason these resistors are used are just to provide some bias voltage to the bus for the case in which no drivers are active. (Otherwise, the differential bus voltage would be close to 0 V during this condition, leading to some RS-485 receivers having indeterminate outputs.) When these biasing resistors are used, it is best to only include them on one node of the network. In the case you mentioned, it is not an issue to have the resistors on both nodes since one node uses very high resistances which will not have much impact on the bus.

    Whether or not biasing resistors are used, it is still important to use a 120-Ohm terminating resistor at either end of the transmission line in order to provide a matched load impedance (especially in the case of a longer cable).

    Earlier you confirmed that the DE pin was high for the node that was transmitting data. Can you also please confirm that the DE pin is low on the node that is receiving data? For half-duplex communication to be successful, only one driver can ever be enabled on a shared bus. I just want to double-check, since this is a common misunderstanding that people have.

    Do you have an oscilloscope? If so, that would make a great tool for debugging this issue since it would allow us to see directly which portion of the system is not acting as expected.

    Regards,
    Max