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.

J-1708 set up recommended

Other Parts Discussed in Thread: DS36277, SN65HVD1785, SN65LBC184

Gentlemen, I hope you can provide me an educated advise on the following:

I am currently working in the design of a telematic system to remotely perform a number of simple actions on a vehicle. To make things easier and not get into a CAN or LIN interface, I  would like to implement a low speed serial bus (physical layer of SAE J-1708 standard, used under a custom simple protocol of my own) to communicate my main control module with the remote actuators. It will be fine working at 9.6 kbps, as stablished in J-1708, so a super high speed is not required on the transceiver.

Being common in an automotive electrical environment heavy transients, etc, what set up would you consider robust enough for this?

My initial approach includes the DS36277 choice, protected on the outside by two MOVs, two TBUs and a dual TVS, as shown on TI literature.

Thanks in advance for your comments,

Norb.

  • Hi Norb,

    Yes, using DS36277 along with external protection circuitry could be a good solution for you.  I also wanted to point out a couple of other robust RS-485 transceivers that you may want to consider as well:

    • The SN65HVD1785 features +/- 70 V bus fault protection.  This means that the device will not be damaged even when shorted to these voltages.  This could be useful in an automotive environment in which you would need to protect against shorts to the battery voltage (especially for higher battery voltages like 24 V or 48 V).  This feature could also make it easier to choose external transient protection circuitry - for example, a TVS that clamps at much higher voltages could still be used to protect this transceiver.

    • The SN65LBC184 has a TVS diode built in for transient protection.  This allows it to support IEC ESD transients up to +/- 30 kV (IEC 61000-4-2 contact discharge) as well as 400-W surge transients without requiring external circuitry.

    One thing to note with these devices is that their "Driver Enable" control is active-high (as is common in RS-485) rather than active-low like the DS36277.

    Please let me know if you have any questions.

    Regards,
    Max

  • Max,

    Thank you for such a solid advice!

    One extra question on the same subject:

    Considering tha I will be using a cable with more than one twisted pair and that power requirements on each node won't be high in fact, would you consider a good practice to supply power to each node by means of one of those extra wires running parallel to the data pair? Let's say: I will be having one twisted pair for data, and a second twisted pair for ground and power.


    Thanks once again,

    Norb

  • Hi Norb,

    It is generally OK to use adjacent wires to transfer power along with the signals, especially if the power consumption is low per node (otherwise, with enough supply current you may start to get an appreciable voltage drop due to the cabling resistance).  You may end up with some higher-frequency noise coupling from the data lines to the power lines.  If this is a concern, you could slow the edge transitions of the data down using capacitors on the bus or just filter the power supply using ferrite beads/decoupling capacitances.  If the power rail will be used for any especially noise-sensitive analog components, you could also choose to distribute a higher-voltage power supply and then regulate it down to a lower voltage (e.g., using an LDO).

    Best regards,
    Max