One Way Speed Of Light - Definition & Experiments

[whosapopstar?]

Referring to:
http://en.wikipedia.org/wiki/One-way_speed_of_light

I am about to quote 2 sentences that obviously have different meanings. The problem is that I don't understand, why they have different meanings, and yet at the same time they appear to be understood as not negating each other:

1. "...they show that, when measured in an inertial frame, the one-way speed of light is independent of the motion of the source within the limits of experimental accuracy. In such experiments the clocks may be synchronized in any convenient way, since it is only a change of speed that is being measured..." (Under the heading: "Experiments that can be done on the one-way speed of light")

2."Although the average speed over a two-way path can be measured, the one-way speed in one direction or the other is undefined (and not simply unknown), unless one can define what is "the same time" in two different locations. To measure the time that the light has taken to travel from one place to another it is necessary to know the start and finish times as measured on the same time scale..." (Under the heading: "The one-way speed")


So in quote no.1, they say that only the change in the speed of light is measured, and that is the reason for light speed being independent of the motion of the source? What does it mean? And if it is has nothing to do at all with the source, you may also say that a decisions to put a tank of water with 5 gold fish and not 8 gold fish, in the way of a light beam (the fish are always above the beam), is a human decision, independent of the motion of the source… so is that sentence, actually just playing with words? If so, why is this sentence included as part of the term "The one way speed of light"?

 

[jtbell]

1. "...they show that, when measured in an inertial frame, the one-way speed of light is independent of the motion of the source within the limits of experimental accuracy. In such experiments the clocks may be synchronized in any convenient way, since it is only a change of speed that is being measured..." (Under the heading: "Experiments that can be done on the one-way speed of light")

Allow me to quote the entire paragraph:

Although experiments cannot be done in which the one-way speed of light is measured independently of any clock synchronization scheme, it is possible to carry out experiments that measure a change in the one-way speed of light due, for example, to the motion of the source. Such experiments are the De Sitter double star experiment (1913), conclusively repeated in the x-ray spectrum by K. Brecher in 1977;[15] or the terrestrial experiment by Alväger, et al. (1963);[16] they show that, when measured in an inertial frame, the one-way speed of light is independent of the motion of the source within the limits of experimental accuracy. In such experiments the clocks may be synchronized in any convenient way, since it is only a change of speed that is being measured.

(I added the boldface for emphasis.)

The one-way speed depends on your clock-synchronization convention. However, provided that you do not switch to a different clock-synchronization convention in the middle of your experiment, you can nevertheless measure changes or differences in the one-way speed during the experiment, between the same source and detector, if such changes really exist.

 

[whosapopstar?]

OK JT,
Can you please try to help me to express in words, what i am trying to ask in return to your response?

If so, then when having a source in motion, you are able to have more than one p.o.v (clock-synchronization convention ), and then if you look at this experiment (the full no1.qoute) from several p.o.v, you see the same results, right? e.g. that there is no change in the speed of light as a result of the source motion. But this is the same one beam of light! There are no several beams of light! So these several unchanged speeds are properties of the same beam of light! So it is the same speed that is unchanged! Where are my words misleading me here?

 

[ghwellsjr]

The referenced article cites the De Sitter double star experiment in which there are two beams of light, one coming from each of the two stars in mutual orbit. The issue is whether the light coming from the approaching star would arrive earlier due to its supposed increased speed compared to the light coming from the receding star due to its supposed decrease speed resulting in the apparent motion of the stars being out of sequence. Since this didn't happen in the many cases De Sitter examined, he concluded that there was no difference in the speed of light due to motion of the source.

Does that help you understand what the article is saying?

 

[LaurieAG]

The issue is whether the light coming from the approaching star would arrive earlier due to its supposed increased speed compared to the light coming from the receding star due to its supposed decrease speed resulting in the apparent motion of the stars being out of sequence. Since this didn't happen in the many cases De Sitter examined, he concluded that there was no difference in the speed of light due to motion of the source.

If you look at the De Sitter double star experiment from the perspective of a stationary observer equidistant from the pair, would you only get a different answer when the distance to the observer was < one galactic year away from the centre of rotation of the pair?

If the pair were equidistant from the observer at the start point, there was no interruption of the light over the galactic year the light took to travel to the observation position and the pair were at the same equidistant position after the completion of the galactic year cycle (i.e. at the observation time), there is no reasonable way to expect that the answer would be different.

 

[whosapopstar?]

ghwellsjr,
I was not referring to the De-Sitter experiment, but to an imagined experiment, where only one beam is considered and not two. Consider a one way single source of light that changes rapidly its speed and also changes its emitted color in an unnatural way (the colors change not in a natural order) and never returns to the same color. This beam of light speed is measured using different conventions of clock synchronization that all refer to the same color (actually 'segment' of color- there is no question here as to 'what' is the color only that it is the same color segment measured by different conventions). Now it is considered that all the different sync conventions, agree upon the constancy of the one way speed of light - of the same one way single beam of light (color segment), which according to all of the conventions' results never changed it speed between source and target detectors, therefore it is the same speed that was measured - they are independent because they call the same light speed convention to the same single unchanged phenomenon. Maybe they even don't agree regarding the speed of the source of light but they do agree regarding the speed of the light beam itself, so regarding the light beam they are independent. In other words there is no change in the speed of light as a result of the fact that the source is moving, but there is also no change of the speed of light as a result of using different conventions. If it was a different speed of light, then it had to change its speed when a certain convention was 'dominant', but this change could then be detected by other 'non-dominant' conventions and it was not detected. Where is the flaw in such a thought experiment?

 

[Dale]

I was not referring to the De-Sitter experiment, but to an imagined experiment, where only one beam is considered and not two. Consider a one way single source of light ...

Then the first quote cited in the OP is probably not relevant to the case you are interested in. So, no need to worry about any perceived conflict between the quotes.

 

[whosapopstar?]

D.S,
There is.

First of all let me clarify what is my problem with the relation between the two quotes: It is not clear from the text, if the one way constancy of the speed of light, is a matter of definition, or something that was proved by experiment (taking into consideration that there is an agreement on clock sync).

In other words, in my opinion, it should be emphasized in the text, that 'a matter of definition' relates to the question of how to sync the clocks, and not to the fact that light speed is constant - that can be only proved, by experiment, and not through definition. If I am wrong here let me know.

The only question that has to be resolved, please correct me, in order to make a connection between the 'imaginary' (thought) experiment i asked about, and the De-Sitter experiment, Is if it is possible to construct an experiment (under a defined clock sync system -or rather in this case, multiple clock sync systems, that measure the same light beam) that proves that the one way speed of light is constant - Using only one beam of light and not two beams of light! if it is possible to construct such an experiment, then the question regarding the need to define a clock sync, becomes irrelevant when measuring light speed (and still relevant to any other moving object) - granted this experiment multiple sync definitions, will all agree that the same beam of light has a constant speed (although the light source is moving). please correct me.

Are you saying that under a defined sync clock system, **with less than two beams of light**, it is not possible to construct an experiment that will prove equivalently to De-Sitter experiment, that the constancy of the one way speed of light exists? Why is the minimum requirement, two beams of light? and if so why does the text header say: 'Experiments that can be done on the one-way speed of light'? it should say 'more experiments that can be done on the two-way speed of light', isn't that so?

 

[Dale]

My point is only that you were being overly dismissive of the De-Sitter experiment. Why does it matter if one or two beams is used? If it is really absolutely critical that only one beam be used then why do you care what someone says about an experiment with two beams?

As said above, it is not possible to measure the one-way speed of light independent of your clock synchronization convention. It is possible to measure changes in the one way speed. In other words, say we have a theory where the speed of light is infinite in one direction and 1/2 c in the other direction, with no change in speed as a function of the speed of the emitter. This theory would be consistent with the experimental data, but would have a different synchronization convention than Einstein's.

I suspect that in theory you could do a "change in speed" experiment with a single beam, but it would probably require more accurate knowledge of the acceleration of the emitter than you can obtain practically. So a two beam experiment will allow much higher accuracy. Simply dismissing the two-beam data out-of-hand seems unreasonable to me.

 

[PAllen]

I'm sure this has been thought of before, but I'm curious to see description of how a theory with non-isotropic lightspeed would explain the following thought experiment (not sure if such a thing has been done or is practical):

Imagine a triangle with a light source and clock at one vertex and mirrors at the others, such that a flash of light is reflected back to the source. The triangle sides are measured in some directly way not involving light. There is one clock involved - the one at the vertex. We perform this experiment in many possible orientations and triangle shapes. We always get the same speed for c.

You can also add finesse that each mirror is a special type of mirror that both reflects light and instantly emits some light as well when hit by light. So light along a path consists of both emitted and reflected light.

I am probing here what I see as a basic problem of distinguishing 'away versus back' in proposals that one way speed of light is not known.

 

[ghwellsjr]

D.S,
There is.

First of all let me clarify what is my problem with the relation between the two quotes: It is not clear from the text, if the one way constancy of the speed of light, is a matter of definition, or something that was proved by experiment (taking into consideration that there is an agreement on clock sync).

In other words, in my opinion, it should be emphasized in the text, that 'a matter of definition' relates to the question of how to sync the clocks, and not to the fact that light speed is constant - that can be only proved, by experiment, and not through definition. If I am wrong here let me know.

The only question that has to be resolved, please correct me, in order to make a connection between the 'imaginary' (thought) experiment i asked about, and the De-Sitter experiment, Is if it is possible to construct an experiment (under a defined clock sync system -or rather in this case, multiple clock sync systems, that measure the same light beam) that proves that the one way speed of light is constant - Using only one beam of light and not two beams of light! if it is possible to construct such an experiment, then the question regarding the need to define a clock sync, becomes irrelevant when measuring light speed (and still relevant to any other moving object) - granted this experiment multiple sync definitions, will all agree that the same beam of light has a constant speed (although the light source is moving). please correct me.

Are you saying that under a defined sync clock system, **with less than two beams of light**, it is not possible to construct an experiment that will prove equivalently to De-Sitter experiment, that the constancy of the one way speed of light exists? Why is the minimum requirement, two beams of light? and if so why does the text header say: 'Experiments that can be done on the one-way speed of light'? it should say 'more experiments that can be done on the two-way speed of light', isn't that so?

All the answers to your questions are in the article you referenced. Why don't you read it carefully and not take any parts out of context?

The article is basically reiterating what Einstein said in his 1905 paper introducing SR and in his subsequent writings. You can measure the round-trip speed of light using a source, a mirror located some measured distance away from the source and a detector located at the source and a timing device located at the source. Any inertial observer who carries out this experiment will get the same value c for the total round trip speed of light by taking double the measured distance divided by the measured time interval.

But the unanswerable question is: when, during that time interval, did the light hit the mirror? Or to phrase the question another way: how do we want to partition the time interval between the two portions of the trip? Do we want to put all of it on the outgoing part and none on the incoming? Do we want to put all of it on the incoming part and none of it out the outgoing? Do we want to divide it in some other unequal way? Or do we want to make the two parts equal? That's what Einstein said to do in his second postulate. If we do that, then we are saying that the propagation of light is the same in all directions according to the definition of time or according to his convention for synchronizing remote clocks.

So now we have defined the one-way speed of light and from that we have defined what time means at remote locations. So any measurement that we make of the one way speed of light using our definition of remote time will, of necessity give us the value c.

Note that if another observer, moving with respect to the first one, uses the first one's definition of time, his measurement of the one way speed of light will not, in general be c. He has to use his own definition if he wants it to always be c.

So if you understand all that, can you figure out if your imaginary experiment adds anything more to the issue? (Quite frankly, I can't follow your experiment or what is the issue you are concerned with or what you think might be a flaw in it.)

 

[ghwellsjr]

I'm sure this has been thought of before, but I'm curious to see description of how a theory with non-isotropic lightspeed would explain the following thought experiment (not sure if such a thing has been done or is practical):

Imagine a triangle with a light source and clock at one vertex and mirrors at the others, such that a flash of light is reflected back to the source. The triangle sides are measured in some directly way not involving light. There is one clock involved - the one at the vertex. We perform this experiment in many possible orientations and triangle shapes. We always get the same speed for c.

You can also add finesse that each mirror is a special type of mirror that both reflects light and instantly emits some light as well when hit by light. So light along a path consists of both emitted and reflected light.

I am probing here what I see as a basic problem of distinguishing 'away versus back' in proposals that one way speed of light is not known.

Did you read in the article the part under the heading "The two-way speed" where it says, "Any measurement in which the light follows a closed path is considered a two-way speed measurement"?

 

[PAllen]

Did you read in the article the part under the heading "The two-way speed" where it says, "Any measurement in which the light follows a closed path is considered a two-way speed measurement"?

Yes, but I am asking 'can someone demonstrate a mathematical model non-istropic lightspeed that is consistent with such an experiment', all done in one frame with one clock?

And to my mind, LET is rather strange in this regard: it would really treat this a measurement of isotropy of light speed in the aether frame; in any other frame it would be a measurement of effective isotropy by virtue of assuming that motion through the aether exactly mimics SR for all possible phenomena.

So, I guess I am saying, is there any other possible theory of non-isotropic lightspeed that could be consistent with this experiment? If not, then I would say: you can measure one way speed of light, but can't thereby rule out LET which makes all differences from SR unobservable.

[EDIT: I find Edward's theory absurd on its face because you need to reverse q based on knowledge of the complete experimental setup. If this is the only possibility besides LET, then again, I would say we can measure isotropy of light speed, in general. But there happen to be a couple of tailored theories with implausible assumptions which still can't be ruled out, and have unobservable non-isotropic lightspeed. This would be like quibbling over: you can't really compare the weight of two objects because you can never way them in the same place at the same time.]

 

[ghwellsjr]

Yes, but I am asking 'can someone demonstrate a mathematical model non-istropic lightspeed that is consisent with such an experiment'. The only example I know of is the unique form of LET

The article discusses LET and others. Your experiment sounds a little bit like the Trimmer experiment. Do a search on this forum for Trimmer and see if it helps.

 

[PAllen]

Let me add one more point about Edward's type theories in the context of experiments like my proposal: the exact same radiation moving the same direction and essentially the same place will need to travel at different speeds depending the what experimental set up it is part of.

To me, the existence of tailored, implausible theory that makes x unobservable is not normally taken to mean x is actually unobservable. Instead, one would say x has been measured so as to rule out any plausible theory that predicts something different from x. On that score, I believe both isotropy and one way speed of light are measurable.

 

[whosapopstar?]

It might be an easy way to refer someone to read just this one more time a 'holy' text, when you don't understand what they are asking.

I on the contrary will make things simple for you and simplify my question to a non-possibility not to understand:

Is there or isn't there an experiment that proves what De-Sitter experiment proves, using only one beam (ray, line, burst or whatever word you chose here) of light in only one direction? and not two beams in any configuration of directions.

This must be clear enough not to get any pedagogic retaliation but just a clear cut yes/no and what is the experiment called answer.

 

[Dale]

Yes, but I am asking 'can someone demonstrate a mathematical model non-istropic lightspeed that is consistent with such an experiment', all done in one frame with one clock?

Yes, but the math is easier if you do a square rather than a triangle. Suppose the speed of light is infinite in +x, 1/2 c in -x, and c in +y and -y. Suppose that we are using mirrors arranged in a square that is 1 light second on each side, aligned with the axes. Then a pulse of light will go as follows emitted at t=0, reaches first mirror at t=0 (+x direction), reaches second mirror at t=1 (+y direction), reaches third mirror at t=3 (-x direction), and completes the loop at t=4 (-y direction). Total time is 4 s and total distance is 4 light-seconds.

 

[PAllen]

Yes, but the math is easier if you do a square rather than a triangle. Suppose the speed of light is infinite in +x, 1/2 c in -x, and c in +y and -y. Suppose that we are using mirrors arranged in a square that is 1 light second on each side, aligned with the axes. Then a pulse of light will go as follows emitted at t=0, reaches first mirror at t=0 (+x direction), reaches second mirror at t=1 (+y direction), reaches third mirror at t=3 (-x direction), and completes the loop at t=4 (-y direction). Total time is 4 s and total distance is 4 light-seconds.

Fine, but I'm supposing you do the experiment with all sizes and shapes of triangle in all orientations.

 

[Dale]

Yes, the derivation is left as an exercise for the interested reader

Basically, you just determine the speed of light in each direction of whatever polygon path it traverses and then add the time up all the way around.

 

[LaurieAG]

Is there or isn't there an experiment that proves what De-Sitter experiment proves, using only one beam (ray, line, burst or whatever word you chose here) of light in only one direction? and not two beams in any configuration of directions.

This must be clear enough not to get any pedagogic retaliation but just a clear cut yes/no and what is the experiment called answer.

Yes but nobody has done the experiment.

De Sitters double star experiment involves observing binary star pairs from a relatively fixed (during the observation) position.

http://en.wikipedia.org/wiki/De_Sitter_double_star_experiment

Since the total flight-time difference between "fast" and "slow" lightsignals would be expected to scale linearly with distance in simple emission theory, and the study would (statistically) have included stars with a reasonable spread of distances and orbital speeds and orientations, deSitter concluded that the effect should have been seen if the model was correct, and its absence meant that the emission theory was almost certainly wrong.

The Twins paradox refers to a similar experiment that could be considered consistent (by a minority of physicists) with De Sitters results when looked at in absolute terms.

http://en.wikipedia.org/wiki/Twin_paradox

The time differential explanation in absolute terms
...
For example, A.P. French writes, regarding the twin paradox: "Note, though, that we are appealing to the reality of A's acceleration, and to the observability of the inertial forces associated with it. Would such effects as the twin paradox exist if the framework of fixed stars and distant galaxies were not there? Most physicists would say no. Our ultimate definition of an inertial frame may indeed be that it is a frame having zero acceleration with respect to the matter of the universe at large."

If the twins experiment was conducted, where the observer was at a stationary point on the Hubble flow (zero acceleration with respect to the matter of the universe at large), you would also expect to get the same results as the De Sitter double star experiment for single sources due to the reasons outlined in my previous post in this thread.

 

[whosapopstar?]

I don't get it.
Does De-Sitter experiment prove the one way speed of light constancy, under certain limitations (having a defined clock sync sys) or does it actually prove something else?
If the goal of De-Sitter experiment is only to prove the one way speed of light under these defined conditions, then why go so complex and far away as to measure the light coming from a binary star sys and all that. Why not just construct some kind of moving light source experiment here on Earth or in near Earth space, that proves the same thing?

 

[ghwellsjr]

According to the article that you linked to in your first post, it is not possible to measure the value of the one-way speed of light but you can determine that it is not changing. When you ask if the De-Sitter experiment proves that the one way speed of light is constant, it implies that you are asking if it proves that it has the value of the constant c, which is not what the article is saying. It is saying that whatever the value of the one way speed of light is in the De-Sitter experiment (c or <c or >c, who knows?), it wasn't changing.

You also don't seem to understand that in the type of experiment the article was saying could be done on the one-way speed of light, you don't even have to "synchronize" your clocks, you just have to make sure they are running at the same rate, because all you are looking for are changes in the differences between the clocks under different conditions of motion of the light source.

If you ask why nobody has done a better experiment, it's probably because they don't believe they would learn anything from it, and so they are not motivated. I certainly would not be motivated but apparently you are. I wish I could understand what your experiment was so that I could offer you some help but I asked you what it is you think the flaw is and you didn't answer. What is it that you think such an experiment will prove?

 

[whosapopstar?]

gh,
I am trying through this suggested thought experiment to understand what exactly there is to be defined if you want to conclude the constancy of the one way speed of light. Because if you have several different defined sync clock systems, that are all measuring together at once the same one light beam, and all of them agree that it (the light beam) did not change its speed (although the light source did change its speed), then the speed is the same for all the different clock sync systems, isn't it? So what is there to be defined per only one sync clock system? I guess this is not clear enough or I am not using the terms in a customary sense, Am I? If one sync clock sys would say that it measures c+0.00001 and another c-0.00001, then this is a problem, but if I understood correctly, experiments, no matter how you sync the clocks, always show exactly c. They all show c that never changed although the source was moving. Therefore it is the same c. Why should it be a different c if they all show c regarding the same exact beam?

 

[ghwellsjr]

gh,
I am trying through this suggested thought experiment to understand what exactly there is to be defined if you want to conclude the constancy of the one way speed of light.

Again, I don't know if when you use the phrase, "constancy of the one way speed of light", you mean that it is equal to the constant c, or just equal to some arbitrary value that could be greater than or less than c as long as it is constant for any given setup.

Because if you have several different defined sync clock systems, that are all measuring together at once the same one light beam, and all of them agree that it (the light beam) did not change its speed (although the light source did change its speed), then the speed is the same for all the different clock sync systems, isn't it?

No, one of those systems may have defined the one-way speed of that light source to be one half of c, another might say that it is twice c and a third might say it is three times c.

So what is there to be defined per only one sync clock system?

You're defining the times on the remotely located clocks in relation to one local clock that is considered the master clock.

I guess this is not clear enough or I am not using the terms in a customary sense, Am I? If one sync clock sys would say that it measures c+0.00001 and another c-0.00001, then this is a problem,

It's not a problem if that's how the clocks were defined.

but if I understood correctly, experiments, no matter how you sync the clocks, always show exactly c.

No, they show what you define them to show.

They all show c that never changed although the source was moving. Therefore it is the same c. Why should it be a different c if they all show c regarding the same exact beam?

I've tried to explain this from the standpoint of an observer who has synchronized his clocks according to Einstein's convention so that in a Frame of Reference in which he is at rest, the one-way speed of light is defined to be c, and then to think about another observer moving in that frame and seeing what the one-way speed of light is according to that same FoR, and it will not be c for many directions but whatever it is for a given direction, it will still be a constant, even if the source is moving with respect to the second observer.

But if that is not convincing, consider what Lorentz Ether Theory says. It says that the one way speed of light is equal in all directions to c relative only to the ether. If a light source is moving through the ether, the speed of the light emanating from it will be independent of its motion and will continue to be constant with respect to the ether. Now consider an observer who is also moving through the ether along a path in line with the motion of the light source. Although the speed of the light relative to him is not c, it is still a constant value, as long as he is inertial, and he can measure that it is constant even if he cannot measure the actual value of that constant. All of this would be true, even if relativity were not true or if we were to consider some other medium propagating a slower wave such as water or sound. If it were not true, we would truly live in a bizarre world.