Capacitance on a bus network

Discussion in 'Embedded Systems and Microcontrollers' started by David Fowler, Mar 7, 2016.

  1. David Fowler

    Thread Starter New Member

    Feb 11, 2016
    25
    0
    A pretty basic question for you all but I just want to check my understanding on this.

    I've got a bus network, the bus itself is short (contained on a PCB) and until now the stubs have also been no more than few cm each. But I'm looking at scaling this out and want to check my understanding of how capacitance is going to affect things. Would the capacitance be uniform across the whole network or is it going to vary from stub to stub depending on the length of the run? As far as I understand things, it should be uniform but just want to check that I'm understanding it correctly?

    Cheers all!
     
  2. ErnieM

    AAC Fanatic!

    Apr 24, 2011
    7,388
    1,605
    Different things are different. If the run length carries then so does the capacitance.

    If the max length is acceptable then the others should be also.
     
  3. Papabravo

    Expert

    Feb 24, 2006
    10,142
    1,790
    If stubs are "short" with respect to wavelength then there is generally no problem. As a rule of thumb you can use 300/f(MHz.) to approximate wavelength in meters. So 300 MHz. has a wavelength of 1 meter.
     
  4. David Fowler

    Thread Starter New Member

    Feb 11, 2016
    25
    0
    Cheers guys, from that equation I don't think that, with the speeds that I'm transmitting, I'm ever likely to hit a problem.

    Everything I've worked on up until now has all been confined to a single PCB or very short wired connections so I'm just trying to figure out what I can and can't get away with when expanding things out.
     
  5. Papabravo

    Expert

    Feb 24, 2006
    10,142
    1,790
    It is not the datarate you need to worry about but the edgerate of your square waves. The faster the rising and falling edges, the higher the effective frequency you need to deal with.

    http://www.edn.com/electronics-blog...-The-bandwidth-of-a-signal-from-its-rise-time
     
    Robin Mitchell likes this.
Loading...