[jeffreyH (OP)]

The setup: Two long sheets of photographic paper on two separate conveyor systems. These move at the same rate, one clockwise and the other counter clockwise. Photons are fired at them from opposite directions. One set of photons each at each film. The spacing between photons is identical in each case. The marks on the paper should differ in spacing if the one way speed of light differs. Of course the whole apparatus must be long enough and moving fast enough for there to be a detectable effect. If this shows no difference then the one way speed of light idea is dead and buried.

[wolfekeeper]

This is dead and buried. The one way speed of light is always 'c'.

You can explain this result in various ways. The way I prefer is to point out that if Newtonian mechanics is correct, and Maxwell's equations are correct as well, then given that matter is held together with electromagnetic forces, with extreme care you can derive special relativity.

The way it works is that you can show that clocks must run slow when they start to move, and rulers get Lorentz contracted, and further, due to the time dilation that clocks have there's inevitably a systematic difference in time along any length in the direction of travel ("lack of simulteneity").

If you put those three effects together you can show that the one-way speed of light is completely independent of speed, as measured from any frame of reference, moving or not.

[kpvats]

If you mean practical measurement of one way speed of light, then it is going to require synchronized clocks, otherwise, it is not possible.

[PmbPhy]

Hi Jeff- Great question as always. Not only can it be done but it has been done. Its required knowledge for very long baseline interferometry (VLBI) in radio astronomy. It requires having two synchronized clocks at each location of each station, Dave Cooper thought it was impossible but to me it merely seems that he lacks a full understanding of clock synchronization, There are more than one way to synchronize clocks, People always seem to think that it its not done the way Einstein explained i then its not a real synchronization. On the contrary. Two atomic clocks can be synchronized at one place and then each clock is transported at a sloe speed (driving a car perhaps) and the clocks will still read the same time within experimental limits. If the one way speed of light was different depending on direction then inconsistencies with be found in the data, None were found,

[jeffreyH (OP)]

Thanks Pete. Do you have any references to papers on the subject?

[PmbPhy]

Thanks Pete. Do you have any references to papers on the subject?

No. But I an scan that portion of Ohanian's text and put them in a PDF file and upload it here. Isn't the 21st century marvelous!!

[David Cooper]

The setup: Two long sheets of photographic paper on two separate conveyor systems. These move at the same rate, one clockwise and the other counter clockwise. Photons are fired at them from opposite directions. One set of photons each at each film. The spacing between photons is identical in each case. The marks on the paper should differ in spacing if the one way speed of light differs. Of course the whole apparatus must be long enough and moving fast enough for there to be a detectable effect. If this shows no difference then the one way speed of light idea is dead and buried.

I can't work out from that what your alignments are for the paths of the photons relative to these conveyor systems. Should I assume that its a shallow angle from the end aimed with high precision half way along sheet of photographic paper? I'm also imagining that the whole apparatus is moving along through space in the same direction as the conveyor systems are aligned, so one sheet of photographic paper is moving forwards more quickly than the other.

If that's what you had in mind, then you have to length-contract your photographic paper appropriately for its speed of travel through space, and once you've applied the right amount of contraction, you'll find that the light pulses hit the paper in such a way that when you stop the experiment and measure the distances between them carefully, they'll have identical spacings between them on both sheets of paper (i.e. the same separation distance on both sheets). [If you're shooting both sheets with photons from both ends, you'll have two lots of dots recorded on each sheet with different separations, but these two separation values will be the same for the patterns of dots on both sheets of paper.] You can't cheat relativity. Incidentally, both conveyor belts will length-contract equally, but the contraction applying to the paper on different sides will not match - there will be more paper on one side than the other at any moment in time while the system is moving through space.

[David Cooper]

Dave Cooper thought it was impossible but to me it merely seems that he lacks a full understanding of clock synchronization,

What did I think was impossible? If you're going to name someone and attribute a failing to them, spell out clearly what that error is and link to some evidence to back your claim. There is no difficulty in synchronising clocks (using a wide variety of methods), except for one little problem: they are only genuinely synchronised if you're synchronising them for the absolute frame (which you can't identify).

[yor_on]

Well David, Don't know a 'absolute frame', but if you by it might mean being in a 'absolute' same frame of reference then I think even a slow moving car will fail that requirement, unless both are in it at a 'same frame of reference' at all times. Then again, the question reminds me a lot of the idea of what I think is called 'weak experiments', where you expect something intermediate to have no discernible effect on the main process/system/interaction (whatever that might imply) you want to observe.

[Bill S]

Yor_on; I'm confused (happens often).  Are you saying that the experiment to which Pete refers would be classed as a "weak experiment"?

[yor_on]

Yes, that's my view of it. Doesn't mean it's 'wrong'. It's about you defining your 'system', and from there draw conclusions, as light actually 'moving', not only 'propagating'.

[David Cooper]

Well David, Don't know a 'absolute frame', but if you by it might mean being in a 'absolute' same frame of reference then I think even a slow moving car will fail that requirement, unless both are in it at a 'same frame of reference' at all times. Then again, the question reminds me a lot of the idea of what I think is called 'weak experiments', where you expect something intermediate to have no discernible effect on the main process/system/interaction (whatever that might imply) you want to observe.

"Absolute frame of reference" has a very specific meaning, being the frame in which light travels at c in all directions relative to a stationary object. Clock ticks are only genuinely simultaneous if the clocks synchronised for that frame, but because we can't identify that frame, we just have to pick a convenient frame and synchronise them for that instead. I took part in a long discussion about this in another thread somewhere (which I can't now find), and I assume I'm being accused of making some kind of error there, but what I said there stands, so the error is an imaginary one either based on someone's misunderstanding of what was said there or on a mangled recollection. This accusation of an error comes from someone who ended his post by saying:-

If the one way speed of light was different depending on direction then inconsistencies with be found in the data, None were found,

That shows a complete failure to understand what LET and the MMX say about how the maths of this works, because no inconsistencies should be found in the data. This has been understood for a very long time, but still some people have not caught up with the facts.

[jeffreyH (OP)]

David makes an excellent point above about the absolute frame of reference. It is subtle but necessary to understand. I will have more to say when I have some free time.

[Bill S]

I’m obviously going to have to look again at the idea of an absolute frame of reference. 

My understanding is that an absolute frame of reference is some fixed reference frame that every observer in the universe would agree is at rest, and that relativity ruled this out.

"Absolute frame of reference" has a very specific meaning, being the frame in which light travels at c in all directions relative to a stationary object.

I thought that light travelled at c (in a vacuum) relative to any stationary (or relatively moving) object.  Wouldn’t that mean that the whole Universe was an absolute frame of reference?

[jeffreyH (OP)]

Don't let what I say sway you. Challenge me!

[alancalverd]

If V_AB ≠ V_BA then the value measured by any experiment will depend on the direction of AB.

This is not found by experiment.

Therefore the "one-way" speed of light  is the same as the "two-way" speed.

[jeffreyH (OP)]

In an inertial frame, and without looking for external indicators, we cannot tell if we are moving or in what direction that motion may be. Galaxies that recede from us will shift their light into the red end of the spectrum. They do tell us which way they are traveling. Would an object moving away inertially also red shift its light or is acceleration a necessity?

[PmbPhy]

Jeff - Sorry for the delay but here's what I promised:
http://www.newenglandphysics.org/other/Ohanian.pdf

Suggestion: Don't confused something being tested with something being proved. Physics is not about proving things: See: http://www.newenglandphysics.org/common_misconceptions/Alan_Guth_04.mp4

[David Cooper]

My understanding is that an absolute frame of reference is some fixed reference frame that every observer in the universe would agree is at rest, and that relativity ruled this out.

There is no such universal absolute frame - the expansion of the space fabric means that the absolute frame of reference at one location isn't necessarily the same frame as the absolute frame of reference at another location far away. The absolute frame is the frame at any point where a stationary object is stationary relative to the local fabric and where light passes that object in all directions at c relative to it.

I thought that light travelled at c (in a vacuum) relative to any stationary (or relatively moving) object.  Wouldn’t that mean that the whole Universe was an absolute frame of reference?

Light travels at c (in a vacuum and outside of any gravity well) relative to any stationary object, but two stationary objects a long way apart can be moving relative to each other due to the expansion of space, so the whole universe cannot have a single absolute frame of reference other than within an exterior space within which our space fabric is expanding, but it's beyond the ability of physics to access that at the moment, so it moves into mere speculation. Many people will tell you that discussing an absolute frame is also a step into mere speculation, but a correct interpretation of the Michelson-Gale-Pearson experiment says otherwise.

[David Cooper]

If V_AB ≠ V_BA then the value measured by any experiment will depend on the direction of AB.

This is not found by experiment.

Therefore the "one-way" speed of light  is the same as the "two-way" speed.

Wow - you're still clinging to that old mistake after all this time! Have you still not bothered to crunch the numbers in the way LET tells you you should so that you take into account length contraction (caused by the phenomenon of relativistic mass) and clocks being slowed by their movement through space (due to the lengthened round-trip communication distances)? When you do that, you find that no experiment will be capable of detecting any difference no matter how widely the one-way speed of light varies relative to the apparatus in opposite directions.