Clock Synchronization: How to Measure Photon Velocity

[juju]

Hi,

I have been thinking that in order to actually prove Einstein's second postulate of SR about the speed of light (and SR in general) you need to make a one-way measurement of the velocity of individual photons. To do this the clocks at the two ends of the path have to be synchronized. My question is how do you synchronize these clocks without assuming the second postulate (or other parts of SR) is true in the first place.

juju

 

[selfAdjoint]

Think of each clock as a a twin, in the famous twin "paradox". The clocks keep proper time in their own rest frames. And the point of the twin story is that even if they are synchronized at one moment in the same frame, their subsequent travels over curved world lines (i.e.accelerations) will cause them to read differently when they are brought back together. And when they are apart, you can't have them synchronized at all, either they are matched at different times and relative speeds (by radio or light signals) and so they tell different times by Lorentz transformations, or else they are spacelike related in which case the concept of simultenaity, necessary for synchronization, simply does not exist.

 

[juju]

Hi SelfAdjoint,

The conclusion I am tempted to deduce is that many aspects of special relativity cannot actually be tested adequately because they require the assumption of certain other aspects of relativity for their test.

juju

 

[selfAdjoint]

Hi SelfAdjoint,

The conclusion I am tempted to deduce is that many aspects of special relativity cannot actually be tested adequately because they require the assumption of certain other aspects of relativity for their test.

juju

Only if you define naive experiments

There have been thousands of experiments supporting the Lorentz transformations, and from the Lorentz transformations you can deduce the rest of relativity.

 

[juju]

Hi,

OK, then. How do I synchronize two clocks, 10 miles apart, in order to measure the one way speed of light on the surface of the earth. Do this without assuming any part of relativity.

juju

 

[4newton]

At noon time you take two clocks 10 miles apart. At some point in between the two clocks near the middle you send a signal to both clocks and reset them. The clocks then send signals to you and arrive at your location together after the reset. As the Earth turns the pulse of one clock should arrive before the other and reach a maximum difference around midnight. As the Earth returns to noon at your location the clocks should again agree. If you see a difference then the one way speed of light is different.

 

[juju]

Hi 4Newton,

Doesn't sending the original signal to reset the clocks assume that the speed of light in both directions is the same, and is not the same true of the first signal that you get from the clocks.

juju

 

[4newton]

Hi juju:

Doesn't sending the original signal to reset the clocks assume that the speed of light in both directions is the same, and is not the same true of the first signal that you get from the clocks.

The reason to send the first sync pulse is to remove all differences in the system at the start. This gives you a starting point of zero difference for the velocity you are moving on the Earth at the start. As your relative velocity changes as the Earth rotates the pulses should change if the one way path of light changes.

To have higher resolution you can start a counter time base when one pulse is received and stop counting when the other pulse is received. You may accumulate this count over any time period to achieve the resolution desired.