1. This forum section is a read-only archive which contains old newsgroup posts. If you wish to post a query, please do so in one of our main forum sections (here). This way you will get a faster, better response from the members on Motherboard Point.

Why can CAN bus only have 110 nodes?

Discussion in 'Embedded' started by Robert Scott, Apr 5, 2004.

  1. Robert Scott

    Robert Scott Guest

    The non-destructive priority resolution used by CAN requires that two
    contending transmitters be able to sense each other's transmission
    within a fraction of a bit-cell time. The propagation delay along 40m
    of cable is just too long to ensure this timing requirement at 1M/sec.
    On the other hand, protocols like ethernet are designed to allow two
    transmitters to occasionally step on each other's transmissions. They
    use backoff and retransmission, but they sacrifice real-time
    determinancy.


    -Robert Scott
    Ypsilanti, Michigan
    (Reply through this forum, not by direct e-mail to me, as automatic reply address is fake.)
     
    Robert Scott, Apr 5, 2004
    #1
    1. Advertisements

  2. Robert Scott

    lynch Guest

    Why can CAN bus only have 110 nodes?

    Why cann't the transmitting distance exceed 40m when the maximum
    transmitting speed is 1M/s?

    How can we get thses data?
     
    lynch, Apr 5, 2004
    #2
    1. Advertisements

  3. That's not strictly a limitation of CAN bus, but of the transceiver
    chips you're using. It's a fan-out relation, i.e. it's controlled by
    how many receivers a single transmitter can feed.
    Because signal speed on cables is limited (typically 2/3 of the speed
    of light), and because of the way CAN bus handles arbitration and
    acknowledge bits.

    I suggest you get yourself a textbook on CAN --- these are very basic
    questions you shouldn't have to revert to Usenet to for answering.
     
    Hans-Bernhard Broeker, Apr 5, 2004
    #3
  4. The fact that Ethernet uses collision detection rather than
    collision avoidance doesn't remove the segment length
    restriction based on propogation delay. It means that the
    propogation delay must be shorter than the message length
    rather than the address length.
    True, but it has nothing to do with segment length restrictions
    imposed by propogation delay in both CAN and Ethernet.
     
    Grant Edwards, Apr 5, 2004
    #4
  5. Care to recommend a good textook on CAN? I coughed up $90 for
    Etschberger's book and was pretty dissappointed.
     
    Grant Edwards, Apr 5, 2004
    #5
  6. I got myself the 4th German edition of the one by Lawrenz. It's a
    collective work, which means there's the occasional duplication, but
    that doesn't exactly hurt. Some things actually benefit from being
    explained twice, by different authors. It doesn't handle protocols
    implemented on top of CAN (CanOpen, DeviceNet, ...) in heavy detail,
    but since I didn't use those yet, that didn't bother me.
     
    Hans-Bernhard Broeker, Apr 5, 2004
    #6
  7. Glad you asked ;-)
    however, I might be slightly biased...

    http://www.CANopenBook.com

    Current Amazon price (US): $34.97

    Note that from the 537 pages only about 50 deal with such
    basic/generic CAN issues. So it definitely does not cover the level of
    hardware details as Lawrenz or Etschberger do.

    Focus of the book is: what to do if you need more than the possibility
    to exchange a few bytes here and there...

    Olaf
    Tutor at ESAcademy
    www.canbus.us
    www.canopen.us
     
    Tutors of ESAcademy, Apr 6, 2004
    #7
    1. Advertisements

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.