• State
  • Locked Locked
  • Replies 4 replies
  • Subscribers 34 subscribers
  • Views 789 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

TCAN332G: weird ack waveform for the remote node TCAN332G

Frank Xiao
Genius 12055 points
Part Number: TCAN332G

Hi team,

really thanks for your help before with the post https://e2e.ti.com/support/interface/f/138/p/867957/3211181#3211181.

We have tested, our assumption is correct. The total system has 10 nodes. Each node ACK is ok. But we find the remote node CAN waveform is a little weird, could you please help check the risk and whether we can improve it?(We test the CAN_H/CAN_L)

  • 0
    TI__Guru 62910 points

    Hi Frank,

    We are out of the office today but can look into this in more detail later this week once we return.

    In the meantime, though, can you please tell us more about the conditions during this measurement? How many nodes are on the network, and how many are sending an ACK bit? Is this waveform measured at the remote node directly or on one end of a cable? How long is the cable connecting the remote node? How is termination implemented?

    Regards,

    Max

  • 0 in reply to Max Robertson
    TI__Guru 62910 points

    Looking at the waveforms, it looks like the remote node's ACK delimiter may be delayed by ~1 us compared to the other nodes.  This may be due to the propagation delay through the cable.  For example, a common estimate for cable delay is 5 ns per meter.  If there were a 100-m cable connecting to the remote node, then it would take 500 ns before that node sees the CAN frame, then once it ACKs that ACK delimiter would take another 500 ns to propagate back to the sending node.  That would explain why the amplitude increases halfway through the ACK bit.  Then, the other "local" nodes go recessive while the remote node is still driving dominant (for another ~1 us, since the bit duration is 2 us).  The amplitude of this portion of the bit may be smaller, though, if it is only coming from one node at the end of a long cable (since the cable could introduce loss).

    Regards,
    Max

  • 0 in reply to Max Robertson
    TI__Genius 12055 points

    Hi Max,

    The phenomenon will happen when 3 more nodes conditions(cable <40cm), not happen all the time, 50% of the time happen).

    When two nodes, doesn't have the phenomenon.

    Thanks.

  • 0 in reply to Frank Xiao
    TI__Guru 62910 points

    Hi Frank,

    Do all three nodes have identical hardware?  And, is there no significant length of cable used between any of them?  How is termination implemented?  Have you figured out any pattern to when the phenomenon occurs (e.g., does a certain node need to be transmitting)?

    It may be helpful to try probing the TXD input of each of the three transceivers at the same time you are looking at CANH/CANL.  That way we can better understand which node is driving the bus dominant and when.

    Regards,
    Max