• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 33 subscribers
  • Views 471 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.

TCAN332: CAN bus spacing

Timothy Luca
Intellectual 790 points
Part Number: TCAN332

Application note SLLA279A provides a formula for spacing CAN nodes to get reliable transmission. The article seems to indicate that this would occur if you had a really long bus length (as in the example provided), but the resultant equation for CAN node spacing is independent of the overall bus length. The equation d> CL/0.98C meters, where d=node spacing, C is in pf/m and CL is pf. Is there some additional condition for this to be applicable? As it stands, it appears that if the nodes were separated by 1m vs. 100m, it wouldn't matter => the same equation would apply. 

  • 0
    TI__Guru 62920 points
    Hi Timothy,

    The equation in the app note is intended to describe the minimum spacing for a cluster of nodes such that the capacitance is distributed well enough to not significantly degrade the effective impedance of the transmission line. (Nodes spaced very close to one another could appear as a large lumped capacitive load, which would present an impedance discontinuity to an incident wavefront.) The effects of degraded impedance are typically most apparent when the cluster connects to a longer transmission line, but this isn't what the equation is intending to cover.

    In other words, the equation tells you if a given spacing results in a significant impedance change. You would then need to determine whether that impedance change would present communication issues if placed at a given position in a network.

    Please let me know if that doesn't make sense.

    Regards,
    Max
  • 0 in reply to Miguel Robertson
    Intellectual 790 points
    Hi Max,
    Thanks very much for your response. I do understand your comments - but it would be clearer to me if you could take me through an analysis of a system comparable to the example except that the 100 meters is only 1 meter. As in the example, each of the nodes in the cluster is separated by 5 inches. Let's use the same capacitance figures as in the example. Could you take me through timing analysis to see whether a communication issue potentially exists?
    Regards,
    Tim
  • 0 in reply to Timothy Luca
    TI__Guru 62920 points
    Tim,

    Sure. Let's assume that you have small enough spacing and high enough node capacitance that a cluster of nodes generates a reflection of significant amplitude to be problematic. Note from the app note that the duration of the distortion period introduced by the reflection effect is equal to the delay through the cable (see Figure 7). Let's assume the propagation delay of signals over the cable is 5 ns/m. With a 100-m cable, the length of the distortion would be 500 ns, which could be problematic especially at higher CAN data rates. (At 1 Mbps, for instance, this would occupy half of a bit period.) With a 1-m cable, the distortion would only last 5 ns. This would be unlikely to cause a noticeable effect since this is smaller than the transition time of most CAN drivers.

    Max
  • 0 in reply to Miguel Robertson
    Intellectual 790 points

    Hi Max,

           OK, that all makes sense. Thanks very much.

    Regards,
    Tim