Maximum Distance for Pulse Output to ODI?

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts
  • Paul_B
    Automated Home Legend
    • Jul 2006
    • 608

    Maximum Distance for Pulse Output to ODI?

    Subject pretty much says it all really. What is the maximum distance for a pulse signal and an ODI? If the answer is, it depends, then what are the tolerances and how can these be measured?

    Paul
  • Viv
    Automated Home Ninja
    • Dec 2004
    • 284

    #2
    Originally posted by Paul_B View Post
    Subject pretty much says it all really. What is the maximum distance for a pulse signal and an ODI? If the answer is, it depends, then what are the tolerances and how can these be measured? Paul
    The ODI (Octal [8], Digital Input) is used to detect switching signals such as door contacts, PIR outputs etc. These are slow rate of change signals. When using a ODI to count pulses the pulses themselves may be of short duration and would be considered fast rate of change.

    As you suspected in your question the answer is.. it depends and there is no definitive answer.
    It depends on
    The signal source, what pulse width it is sending. Fast pulse width - short distance, long pulse width - long distance. Also its output impedance, but typically not to influential.
    The cable. High capacitance cable – shorter distance. High resistance cable, shorter distance.
    However at the risk of getting it wrong I would say with the correct cable, perhaps an additional pull-up resistor on the ODI several hundred meters should be possible.
    (We have installations where and the Idranet bus to an MFP is over 150m away).

    Viv.

    Comment

    • Paul_B
      Automated Home Legend
      • Jul 2006
      • 608

      #3
      Thanks for the explanation Viv, it gives me something to work with

      Paul

      Comment

      • Karam
        Automated Home Legend
        • Mar 2005
        • 863

        #4
        Just to add to Viv's reply - you have to consider that the digital input has to be able push around 1mA through the 'closed' connection in order to sense it as closed. Considering the internal electronics this equates to the closed contact acting effectively as a resistor of value up to around 1000 ohms (though I'd aim for 500 ohms max to be sure). Assuming the contact was a perfect switch then the only additional significant resistance is that of the wires connecting it to the digital input. At around 20 ohms/100m you'd have to have a pretty long CAT5 pair to break that particular condition for a steady signal. The next problem is capacitance/inductance of the cable which has an effect on the dynamic nature of a signal. This is possible to quantify if you have a good description of the signal but I'll not attempt to go down that path only to say that basically the sharper the pulse the more easily attenuated and smeared. However bearing in mind that typically cat5 pair capacitance is of the the order of 5nF/100m and inductance about 50uH/100m, simplistically speaking the former at least is quite small relative to the 100nF already at the digital input. So what this implies is that if the digital input is able to detect the pulses robustly with a short cable then an extra 5nF relative to the 100 is not going to hugely affect the result (assuming you had sufficient margin to begin with). I don't think there would be too much concern either about the effect of 50uH (say) on the kind of pulses typically seen.

        In my opinion I think the bigger problem with long cables and lightweight signals is susceptibility to noise - hence the use of a packetised network like IDRANet to carry the data from remote sensors . The susceptibility both to noise and effects of capacitance and inductance can be improved at the expense of sensitivity by decreasing the 'pullup resistance' as Viv suggested (though you'd need to make modifications to an ODI in order to achieve this), or noise susceptibility can be improved (at expense of increased capacitance) by using screened cable and connecting screen to IDRANet screen at module end or if not available then module 0V.

        Bet you wished you never asked now...

        Comment

        • Paul_B
          Automated Home Legend
          • Jul 2006
          • 608

          #5
          Viv and Karam,

          Thanks for the very detailed explanation and their I was thinking you might say "...no more than 2m...". The takeaway is it depends but can be quite long depending on environmental factors, especially influence of noise from mains. I think I have the info I need so appreciate the background education (Physics A-level was quite a long time ago)

          Paul

          Comment

          Working...
          X