Is it feasible to measure one way speed of light this way?

[Dale]

What exactly is impossible?

Measuring the one-way speed of light is impossible without adopting a simultaneity convention.

 

[bob012345]

Measuring the one-way speed of light is impossible without adopting a simultaneity convention.

Is the synchronicity requirement implicitly doing that? Thanks.

 

[Ibix]

Is the synchronicity requirement implicitly doing that? Thanks.

You mean the clock synchronisation? Yes, choosing your methodology for that is the operational aspect of choosing a simultaneity criterion.

 

[.Scott]

From 1905 to this day we have not experimentally measured the one way speed of light between a source to the detector only the roundtrip from the source to the detector and back again. We just assume that the speed of light is the average speed and it’s the same at all directions equal to c.

The speed of light is the same in both directions - and this is not simply assumed.
The Michelson-Morley Experiment was first performed in 1887 - and has certainly been repeated many times since 1905. The experiment compares the speed of light in two perpendicular directions. The entire interferometer can be rotated while the comparison is made, so if light traveled at a different velocity in one direction than another, the interference pattern would change as the apparatus was rotated.

When the experiment is conducted in an inertial frame, no such shifting interference pattern is observed.

 

[Ibix]

The speed of light is the same in both directions - and this is not simply assumed.
The Michelson-Morley Experiment was first performed in 1887 - and has certainly been repeated many times since 1905. The experiment compares the speed of light in two perpendicular directions. The entire interferometer can be rotated while the comparison is made, so if light traveled at a different velocity in one direction than another, the interference pattern would change as the apparatus was rotated.

When the experiment is conducted in an inertial frame, no such shifting interference pattern is observed.

No, the Michelson-Morley experiment confirms that the two-way speed of light is isotropic. It doesn't (and cannot) say anything about the one-way speed of light, which was the question in this thread.

 

[Vanadium 50]

While the OP is to be commended for the complexity of his device, people who say they have found a way to measure the one-way speed of light should be treated the same as people who claim they can solve one equation in two unknowns.

Specifically, the one-way speed of anything (not just light) going from A to B is:

$$v = \frac{x_B-x_A}{(t_B - \delta t)-t_A}$$

where δt is the difference in clock syncronization from A to B. In Newtonian physics, δt = 0. In Relativity, you tell me what one-way speed you want, and I'll tell you the clock syncronization convention to use. It's one equation in two - count 'em, two - unknowns.

 

[.Scott]

No, the Michelson-Morley experiment confirms that the two-way speed of light is isotropic. It doesn't (and cannot) say anything about the one-way speed of light, which was the question in this thread.

Let me work this through...

Let's call the 2-way (or all way) length of one arm L1 and the corresponding length of the other (perpendicular) arm L2. Then we have the speed of light in the North, South, East, and West direction will be cN, cS, cE, and cW respectively.

And we will apply this additional constraint: L1 and L2 will not be precisely equal. They will be off by at least a few wavelengths - enough to be noticeable as the experiment is conducted.

If the measurement is first made with L1 oriented North/South, the interference pattern will reflect the difference L1(1/cN+1/cS)−L2(1/cE+1/cW). Then as the device is rotated clockwise, the interference pattern reflects a range of other cX vs. c(X+π/2) 2-way combinations, and since the interference pattern is not shifting, all of these also yield the same value.
That is: T=L1(1/cX+1/(cX+π))−L2(1/(cX+π/2)+1/(cX−π/2)) is a constant.

So after 90 degrees, you will have:
T=L1(1/cN+1/cS)−L2(1/cE+1/cW)=L2(1/cN+1/cS)−L1(1/cE+1/cW).
0=(L1−L2)((1/cN+1/cS)−(1/cE+1/cW))
(1/cN+1/cS)=(1/cE+1/cW)

More generally:
(1/cX+1/(cX+π))=(1/(cX+π/2)+1/(cX−π/2))

At this point, we cannot even claim that the 2-way path is isotropic - there could be some pattern (like a square) that repeats every 90 degrees. But if we repeat the experiment with (for example) the arms at 1.5 radians to each other instead of 90 degrees, we demonstrate that the two-way is isotropic.

This only says that the sum of the two time periods remains a constant as the direction changes.
So it seems you (@Ibix ) are right.

Letting C=T/(L1−L2)=1/cX+1/(cX+π) (a constant).
The the average time for each opposing direction would then always be proportional to C/2.
We can define a periodic function f(X) that indicates how much the transit time changes from the average:
f(X)=1/cX−(C/2)

So C=1/cX+1/(cX+π)=(C/2+f(X))+(C/2+f(X+π/2))
Thus f(x)=−f(X+π/2).

So if we add one more constraint, that f(X) is a continuous function, we can demonstrate that there is at least one direction X0 such that f(X0)=−f(X0+π/2)=0. For direction X0, the speed of light is the same in both directions.

If there really is no way to determine f(X) for an inertial reference frame, then by Occam's Razor, we should take f(X)=0.

 

[Nugatory]

then by Occam's Razor, we should take f(X)=0.

To be precise here, by Occam's razor we should assume that $f(x)=0$... and that is exactly the point.

We are assuming that the one-way speed, which we cannot measure, is equal to the two-way speed, which we can measure. Occam's razor says that's a good assumption, when we consider the consequences we find that it is a very helpful assumption, so helpful that it would be perverse to reject it, no one will seriously tell you that this assumption is invalid and that you shouldn't trust results based on it (as long as you don't erroneously misinterpret a coordinate velocity, which is a different problem)... but it is still an assumption, not an observational fact confirmed by an experiment.

 

[darth boozer]

How do you propose to ensure that your measurements are the same in all directions, given the definition of the metre is based on a fixed value for the speed of light?

The metre, symbol m, is the SI unit of length. It is defined by taking the fixed numerical value of the speed of light in vacuum c to be 299 792 458 when expressed in the unit m s–1, where the second is defined in terms of the caesium frequency Cs.

 

[John Kovach]

Sorry for dragging this on. But what is bugging me is no one has mentioned entanglement. Is it possible to synchronize a clock at the destination using an entangled photon at the moment a test photon is released at the source to measure the oneway speed of that test photon?

 

[Vanadium 50]

Is it possible to synchronize a clock at the destination using an entangled photon at the moment a test photon is released at the source to measure the oneway speed of that test photon?

No.

 

[Nugatory]

Sorry for dragging this on. But what is bugging me is no one has mentioned entanglement. Is it possible to synchronize a clock at the destination using an entangled photon at the moment a test photon is released at the source to measure the oneway speed of that test photon?

No. This idea doesn’t work for the same reasons that you can’t use entanglement to send any message. Google for “no communication theorem”, or check out some of the many threads over in the QM section.

 

[kinsler33]

Didn't Dr Michelson settle this argument with the ether drift experiment? He did the first one in my home town of Cleveland, but you can buy the stuff to do it in your own physics class. I wonder if all this interest in the 'one-way' speed of light is perhaps an effort to rehabilitate the luminiferous ether still so beloved by pseudoscience.

 

[Ibix]

Didn't Dr Michelson settle this argument with the ether drift experiment?

No - this is a different topic. See post #40 and responses to that.

 

[kinsler33]

Didn't someone calculate a reasonable value for c by astronomical observation?

Procedure (I think.) :

Note the time T1 o'clock when a star peeks out from behind some heavenly body (a planet, maybe) on a night when that h.b. is at distance D1 from the earth. On another night, six months later, note the time T2 o'clock when we see the same star emerge from behind the same h.b., but now that h.b. lies at a distance D2 from the earth. c = (D1-D2)/(T1-T2). The light always goes the same direction, and I think that the speed of light came out to be about 3e8 m/s.

 

[Ibix]

Didn't someone calculate a reasonable value for c by astronomical observation?

Ole Rømer studied the moons of Jupiter. He used them as a distant clock and ascribed their apparent tick rate variation solely to changing distance, from which he could deduce a speed of light. In a relativistic analysis of his work he was assuming slow clock transport, which is equivalent to assuming an isotropic one-way speed of light. So this is not a true one-way speed measurement either.

 

[Vanadium 50]

No - this is a different topic. See post #40 and responses to that.

c = (D1-D2)/(T1-T2)

No. See post #41 and responses to that.

 

[OCR]

Well, I sees. . .

See post #40 and responses to that.

See post #41 and responses to that.

But, now look, see. . .
If you linked to your post numbers, it'd save a bunch of scrolling. . . .

.