Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Clock synchronization problem

  1. Apr 18, 2007 #1
    Is it possible to realize a situation in which all the synchronized and ticking clocks of a given inertial reference frame read a zero time?
  2. jcsd
  3. Apr 19, 2007 #2
    Why not? There shouldn't be any problem at all, if you synchronize them with a spherical light pulse starting from a fixed point.
  4. Apr 19, 2007 #3

    Thanks. At which time should I emit the synchronizing light signal?
  5. Apr 20, 2007 #4
    -00. (So you have to wait a lot of time to record all "0" times!)
  6. Apr 21, 2007 #5
    Yes it is, for a finite number of clocks. Having established from a given clock by pairwise signalling, the round trip light signal time to every other clock, a series of synchronising signals is sent to each specific clock at zero minus [half the round trip time for that clock], upon receipt of which each clock is zeroed. Thus all the clocks in the inertial system are [Einstein] synchronised at zero.
    Of course the assumption that the clock be synchronised at half the to-and-fro time is the crucial assumption of Einstein synchronisation that puts the "relativity" into special relativity. The clocks will be "synchronised" only for observers actually in the inertial system concerned.
  7. Apr 21, 2007 #6
    clock synchronization

    Thank you for your help. Please let me know what do you mean by "pairwise signaling" (outgoing and reflected light signals?) .
    Considering the clocks C(x) located at the different points of the OX axis to be synchronized with clock C(0) located at the origin O, clock C(x) is stoped and fixed to read -x/c. A light signal emitted from O at t=0 arriving at C(x) starts it and so all the C(x) clocks read the same running time.
    In order to obtain the situation when all the clocks C(x) read a zero time,each of them is stoped, read t=0 being synchronized by different light signals emitted at -x/c respectivelly. Of course your proposal is more in the spirit of Einstein's special relativity. I think that this simple fact is worth to be mentioned in a relativity lecture: In the first case a single light signal synchronizes all the clocks, whereas in the second case different clocks are synchronized by different light signals.
  8. Apr 22, 2007 #7
    Er, no. I think you mean to say C(x) is stopped at +x/c, otherwise the clocks would not synchronise.
    In any case your method requires that x be already known for each clock, and the only practicable way of establishing this is with a to-and-fro light signal between already synchronised clocks !
    So one comes back anyway to the method I described, which is not merely "in the spirit of" SR but is the very method on which SR is based.
  9. Apr 24, 2007 #8

    Could you tell me which is the geometric locus of the simultaneous events which read t'=0 in I' when detected from I?
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook