On Thu, Sep 10, 2009 at 10:17 PM, David Duffy (AVD) wrote: > It got me thinking that since there is still a spare pair in the cable, > should I double up on the comms as well? What are the implications of > having two balanced pairs in parallel, bearing in mind the low data rate? I expect it would work in most situations, but you might be chasing very odd bugs due to doubling them up. Keep in mind that each pair has a different twist rate in the cable, and slight differences in connectors, cable length, etc may result in reflections and ringing that is worse than with one cable. Even at the slow rate of 2400, it can cause problems. Imagine the signal on one wire arriving slightly ahead of the signal on the other wire, so it reaches the single termination point and travels back up the second wire. Further, you may experience interesting magnetic noise as you are forming a loop. Lastly, if one wire becomes damaged the signal will still go through (yay redundancy!) but the two pieces of long wires that are hanging off the terminals at each end will act as great antennas. You will have a very difficult time understanding where certain noise is coming from because the units 'mostly' work. Might be better to have a unit fail and understand the problem immediately than to have it work and try to trace down a bad redundant line. These concerns, however, should be very minor in most slow setups, but they bear thought and testing. Doing this in a very electrically noisy place, for instance, might exhibit worse problems than an office or greenhouse, for instance. You might consider designing it to wire everything straight through, and then use jumpers (or cut traces) to select which pair (or both) any given transceiver is on. Would allow testing with little additional effort should it not work out. If you do try it, let us know how it goes! -Adam -- http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist