• TI Thinks Resolved

ISO1042: CANH/CANL signal oscillated

Expert 3005 points

Replies: 6

Views: 100

Part Number: ISO1042

Hi Team,

  My customer reported an issue while using ISO1042:  the CANH/CANL signal oscillated while connected to CAN BUS.

  Could you kindly help give comments on below 2 items:

 

  1. Comments on how the CANH/CANL signal oscillated with customer’s schematic

  2. How to fix this issue?ISO1042B CANH_CANL oscillated issue.docx

More details for the schematic and test waveform shown as attached.

  • Benjamin,

    This looks like it could be an issue with the termination. Can the customer remove the 120ohm resistor from wherever it is in the CAN bus and replace R739 on both CAN nodes with a 120ohm resistor? Alternatively, the customer can populate R742 and R743 with 60ohm resistors and populate C753 with a 1nF to 4.7nF capacitor.

    Respectfully,

    Lucas

  • In reply to Lucas Schulte9:

    Benjamin,

    I haven't heard from you in a while. I am marking this as resolved for now but please feel free to ask more questions if you need to.

    Respectfully,

    Lucas

  • In reply to Lucas Schulte9:

    Hi Lucas,

      Sorry to reply so later. Have tried your suggestion, the oscillated rating was decreased but still existed.

      In customer real products,  the power module is added to the CAN bus, the terminal resisters were installed in other modules, can't be changed in the power module, so this issue still pending in customer side.

      Also noticed that while using ISO1042,  the resister from CANH/L to GND is about 25kohm while other modules in CAN bus aren't powered up.     And there is products test standard that the  CANH/CANL rising/falling timing should be less than 200ns,  the oscillated common noise couldn't fulfill the requirement;

      Could you kindly give comments on below items?

      1) Reasons for 25Kohm resister from CANH/L to GND while ISO1042 installed in the CAN bus and module powered up?

      2) As the terminal resister couldn't be modified in customer's power module,  could you kindly give suggestions on how to remove the oscillated common noise?

    Expect for your reply, thanks.

    Best Regards

    Benjamin

  • In reply to Benjamin Zhou:

      1) Reasons for 25Kohm resister from CANH/L to GND while ISO1042 installed in the CAN bus and module powered up?

    The 25k that is being measured is the expected input resistance for either CANH or CANL that the datasheet describes. 

      2) As the terminal resister couldn't be modified in customer's power module,  could you kindly give suggestions on how to remove the oscillated common noise?

    The datasheet calls for proper termination of the CAN bus in order to reduce reflections that can cause the ringing that customer is seeing. One other thing is to make sure the stub lengths of the CAN bus are the proper length.

    https://www.ti.com/lit/an/slla270/slla270.pdf?ts=1593028334024

    These documents contain good design practices for CAN network and cover proper stub length based on the system.

    Respectfully,
    Lucas Schulte

  • In reply to Lucas Schulte9:

    Hi Lucas,

       The below figure shown that some unexpected happened while the CANH/CANL in recessive level(marked as RED arrow),  could you kindly give come comments that why this situation happened, and how to remove the issue?     while in dominant level, it's oscillated, based on previous documents you provided, which could be in optimized by the choke.

      Expect for your reply, thanks.

    Best Regards

    Benjamin

  • In reply to Benjamin Zhou:

    Benjamin,

    This is due to some amount resistance on the CANH or CANL lines to ground external to the transceiver. Removing this resistance and properly terminating the CAN bus as I've stated before should remove this issue.

    Repsectfully,

    Lucas