Consistency of the speed of light

[gregory_]

With such an embarassing biased statement all you'll get is the disproof for your Gagnon disproof.

I am always willing to consider that I may be wrong. But your claims are tantamount to saying "we measured the REAL/CORRECT coordinate system of the universe".

We can use whatever coordinate system we want to describe the universe. Yes, I strongly agree with you that some coordinate systems are easier to use than others. But the fact remains that one-way velocities do not have a coordinate free meaning. That is why I feel such statements are reasonable here.

If this offended you, I appologize.
Since you agreed that one-way velocities are a coordinant system dependent quantity, and you even agreed that SR and GGT predict the same results for experiments ... I am baffled as to why we still need to discuss details. It is as if you agree to these and then deny their consequences.

This is where our disagreement is: at the basics. So I feel we should discuss there. I am sorry if suggesting that offended you. I have taken much time to explain my point of view and back it up... I feel this discussion is still very much alive. I don't appreciate you trying to kill it because you don't like my statement of what the mainstream view is.

 

[Aether]

The correct statement is that each experiment measures one way light speed isotropy within the adopted clock synchro scheme. Makes sense?

Please explain this in more detail. Experiments do not directly measure one-way speeds, period. One-way speeds are mathematical artifacts of the overlay of a coordinate system (including a clock synchro scheme \epsilon) onto an experiment/measurement. Krisher et al. stipulate in note 14 that "the relative synchronization of the two clocks at an initial time is completely arbitrary", and immediately following Eq. (4) they say "\epsilon is the vector determined by the procedure adopted for the global syncrhonization of clocks in S". I could write a program to show that Eqs. (3) through (8) of Krisher et al. yield substantially isotropic one-way speeds when \epsilon=-v/c_0^2, and substantially anisotropic one-way speeds when \epsilon=0 if you wish. That should settle it, wouldn't you agree? This has been a good exercise, but it would be nice to resolve the issue at some point.

 

[clj4]

The Mansouri-Sexl "test theory" gives you a set of parametrized transforms. In its most general form the parameters are obviously not fixed.
The \epsilon parameter is tied to the clock synchro scheme. For
\epsilon=0 the MS test theory reduces to GGT theory (and the transforms reduce to the so-called "Tangerlini" transforms) that imply absolute simulataneity.
For \epsilon=-v/c_0^2, you recover the Lorentz transforms and the absence of absolute simultaneity.
In between these two values lies an infinite number of values for
\epsilon , an infinite number of clock synchronization schemes and an infinite number of theories different from SR.
It is common practice to use these fully parametrized theories (BTW, there are two more parameters \alpha and \beta) as "test theories" of SR. The Krisher paper is an example of application of such a test theory. There are many more papers , especially in particle physics, that employ the fully parametrized MS theory as a means of testing SR. They employ more or less the same mechanism:
-an experiment is outlined
-the fully parametrized MS theory is used to make a prediction for the experiment outcome that will differ from SR
-a set of expressions in the \alpha,\beta,\epsilon parameters is being obtained
-the theoretical data is compared with the experimental data and the parameters are constrained to values that are very close to zero

The experiments do not need to measure the one way speed of light, they measure its anisotropy (for example, in both the Gagnon and the Krisher experiments they measure a phase difference) and compare it against the prediction of the "test theory" (in both the Gagnon and the Krisher case this is GGT, the Krisher paper uses the more sophisticated parametrized form while the gagnon paper doesn't). It is interesting to note that Gagnon/Torr come back with a parametrized GGT a little later. Since the experiments invariably come back with experimental values that disagree from the predictions of the test theory, the conclusion is invariably that there is no one way light speed anisotropy and that SR has it over the test theory in cause.
This is why no one in mainstream relativity supports any of the "aether" theories (preferential reference frame). Everyone understands the class of MS theories for what they are , a very valuable tool to test SR to higher and higher levels of precision, never as a viable rival to SR.

As an interesting aside, the experiment proposed by Hans might be a good candidate for separating MS from GR, it needs more work in terms of casting it in the MS formalism.

 

[gregory_]

Since the experiments invariably come back with experimental values that disagree from the predictions of the test theory, the conclusion is invariably that there is no one way light speed anisotropy

No. Again, the Krisher experiment does not (and CAN NOT) constrain the \bf{\epsilon} value. They even admit this themselves. This experiment DOES NOT distinguish between a GGT theory and SR.

The authors admit this, I don't understand why you refuse to admit it.
Just admit it so we can focus on Gagnon.

Please answer this question:
Do you agree that the Krisher experiment can not distinguish between a GGT theory and SR?

 

[clj4]

Read again:

...the Krisher paper uses the more sophisticated parametrized form while the gagnon paper doesn't)

Krisher-Will constrain \alpha,\beta. This is what the declared intention is, this is what they do. As per the explanation at post 303 there are THREE parameters to work with.

 

[Aether]

This is why no one in mainstream relativity supports any of the "aether" theories (preferential reference frame). Everyone understands the class of MS theories for what they are , a very valuable tool to test SR to higher and higher levels of precision, never as a viable rival to SR.

Gagnon himself said that "both Newton's purely absolute view and Einstein's purely relative view seem incomplete. We suggest here that spacetime has dual properties: both absolute and relative. The defintion of physical time is not unique" (see post #262).

Read again:



Krisher-Will constrain \alpha,\beta. This is what the declared intention is, this is what they do. As per the explanation at post 303 there are THREE parameters to work with.

They claim to constrain only \alpha and (1\2+\delta-\beta), and explicitly state that "synchronization of the two clocks at an initial time is completely arbitrary". The synchronization of the two clocks at an initial time, embodied in the parameter \epsilon, is the only difference between SR and GGT.

 

[clj4]

They claim to constrain only \alpha and (1\2+\delta-\beta) , and explicitly state that "synchronization of the two clocks at an initial time is completely arbitrary". The synchronization of the two clocks at an initial time, embodied in the parameter \epsilon, is the only difference between SR and GGT.

Good, so at least you admit that the authors have a valid experiment and that they constrain two parameters.
You asked me to explain how test theories work and I did that for you. A MS violation is a violation by any parameter you measure it by (in this case \alpha AND \beta).
1. So the MS theory used by Krisher et. al produces a violation as per formula (8).
2. This violation is infirmed by experiment.
So light speed IS isotropic (look at the paper title).

 

[gregory_]

So light speed IS isotropic

No. One-way velocities are a coordinant system dependent quantity (as you admitted yourself). Many qualifications need to be added to any statement where an experiment claims to have measured or constrained a one-way velocity. You continue to deny this no matter how much explanation and evidence is shown to you. It appears you have some kind of metaphysical belief that you cannot bear to let go of. You really want the Lorentz transformations to be the ONE REAL transformations between coordinant systems. There is no such thing and your claims are only metaphysical nonsense (you are beginning to sound like rfnorgan).

We can choose ANY coordinant system we wish to describe the universe. Do you deny this?

Let me take the time to explain more explicitly what some common added requirements are used to be able to make such "one way velocity constraints".

#1 - first these usually restrict themselves to inertial coordinate systems (which I'll define as one in which the velocity of a free moving object is constant), and consider gravity effects negligible

#2 - The next most common qualification is that the coordinate t correspond to the time as measured by a "standard" stationary clock. This allows us to eliminate the ordinary galilean transformations and many others.

Note: we still have GGT and Lorentz transformations.

#3 - We add the requirement that a) space is described isotropically or b) physical laws have the same form in all frames.

Ignoring the quantum gravity mess, it appears Lorentz transformations satisfy this.


Only by going through many successive qualifications to the statement of "one-way velocity measurement" can we even discuss such things. Usually we stop at #2, because going to #3 is so restrictive that we have just DEFINED what the velocity of light is, not measured it.

clj4,
Krisher even takes the time to mention in the paper that they don't restrict the value of \epsilon. Do you deny this? This allows the one-way speed of light to be any value one chooses. Do you deny this as well?


Also, can you give us an update on your Gagnon "disproof" of the fact that they forgot to use the GGT form of the Lorentz force?

 

[clj4]

Ah, coming from you:

No. One way speed measurements are not valid.

...how can it have any credibility?

As to the scientific process of "proof" via selective quoting:

Krisher even takes the time to mention in the paper that they don't restrict the value of \epsilon

You left out the immediately following sentence in reference to the derivation of (7) and (8):

"In SR, the w-dependent terms vanish identically."

The disproof of your irrelevant disproof is coming. I suggest that you use this time to recheck your "disproof".

 

[gregory_]

The disproof of your irrelevant disproof is coming. I suggest that you use this time to recheck your "disproof".

What makes my disproof irrelevant?
I seem to remember someone claiming that they could actually admit Gagnon was wrong if I showed why mathematically ... seems I had good cause to be skeptical. You seem incapable of considering the possibility that you are wrong. You must first consider this possibility if you ever hope to learn anything.

Until then, I look forward to your "disproof" that Gagnon forgot to use the GGT form of the Lorentz force.

 

[Aether]

Good, so at least you admit that the authors have a valid experiment and that they constrain two parameters.

I have said nothing about their experiment. I am only talking about their theoretical interpretation.

You asked me to explain how test theories work and I did that for you.

No I didn't, I asked you to explain a statement that you made.

A MS violation is a violation by any parameter you measure it by (in this case \alpha AND \beta).
1. So the MS theory used by Krisher et. al produces a violation as per formula (8).
2. This violation is infirmed by experiment.

What violation? Eq. (8) is identically zero for both SR and GGT.

So light speed IS isotropic (look at the paper title).

For both SR and GGT the isotropy or anisotropy of all one-way speeds is determined by \epsilon. I offered to show that to you using Eqs. (3) through (8). The paper title is not justified.

 

[clj4]

What violation? Eq. (8) is identically zero for both SR and GGT.

You sure about this one? If this were true, then the whole construction of the paper would be irrelevant. Why would the authors bother to write such a paper and why would the reviewers accept it?

For both SR and GGT the isotropy or anisotropy of all one-way speeds is determined by \epsilon. I offered to show that to you using Eqs. (3) through (8). The paper title is not justified.

I think that you should re-read the paragraph 2 of the third MS paper. What they state is something entirely different see the italics below):

The mean velocity of light along a closed path has been calculate in equation (36) of Paper I and it is independent of the synchronization coefficient \epsilon.

We are not talking mean, we are talking one way

OK. Why don't you prove this to us? While you are at it, please explain why
\alpha(w) and \beta(w) have no effect (see above). And please do not pick \epsilon=-v/c_0^2 for your exercise.

Before you launch into your calculations, please take a second pass at the third of the MS papers, the one that deals with second order effects.
Please note that this paper is very different from the first two:1. The authors state clearly (top of page 810) that "No assuptions concerning synchronization will be made in the analysis of these experiments" (Michelson Morley, Kennedy Thorndike).

2. The authors proceed with the analysis and, in stark contrast with the other two papers make no assertion of equivalence of SR and their theory. (they do that in the paper on first order effects)

3. Moreover , the authors make an analysis of the possible violations as expressed in the "remaining parameters" \beta and \delta (top of page 805)

5. Furthermore they (MS) lobby for a higher precision Kennedy Thorndike experiment for the purpose of constraining \beta (bottom of 813 and top of 814)Interestingly enough, both the Gagnon paper and the Krisher one are...higher precision forms of Kennedy Thorndike (I said this many times in this thread). Do you really think that the authors, reviewers and editors didn't know what they were doing? What are the odds of this being true?
And if things fell through the cracks, don't you think that MS would have come back to point out the incorrectness of the two papers? They (MS) seem quite capable in analyzing experiments...

 

[Aether]

You sure about this one? If this were true, then the whole construction of the paper would be irrelevant. Why would the authors bother to write such a paper and why would the reviewers accept it?

Yes, I'm sure. Constraining (1+2\alpha) and (1/2+\delta-\beta) by experiment is a fine goal, it's just not accurate to say that's a test of the isotropy of the one-way speed of light.

I think that you should re-read the paragraph 2 of the third MS paper. What they state is something entirely different see the italics below):

The mean velocity of light along a closed path has been calculate in equation (36) of Paper I and it is independent of the synchronization coefficient \epsilon.

We are not talking mean, we are talking one way

I don't disagree with that (see post #94 and others). I'm not sure what you're getting at here.

OK. Why don't you prove this to us? While you are at it, please explain why
\alpha(w) and \beta(w) have no effect (see above). And please do not pick \epsilon=-v/c_0^2 for your exercise.

Show you that Eq. (8) is identically zero for SR, GGT, and any arbitrary value of \epsilon?

Here's Eq. (8):

\delta \phi/\phi_0=w(1+2\alpha)cos\theta+w^2(1/2+\delta-\beta)cos^2\theta​

For both SR and GGT: \alpha=-1/2, \beta=1/2, and \delta=0.

Variations of these parameters from SR and GGT would indicate violations of both (e.g., a violation of local Lorentz symmetry); we're assuming perfect Lorentz symmetry in this discussion.

(1+2\alpha)=0

(1/2+\delta-\beta)=0

\delta \phi/\phi_0=0​

Before you launch into your calculations, please take a second pass at the third of the MS papers, the one that deals with second order effects.
Please note that this paper is very different from the first two:

1. The authors state clearly (top of page 810) that "No assuptions concerning synchronization will be made in the analysis of these experiments" (Michelson Morley, Kennedy Thorndike).

2. The authors proceed with the analysis and, in stark contrast with the other two papers make no assertion of equivalence of SR and their theory. (they do that in the paper on first order effects)

3. Moreover , the authors make an analysis of the possible violations as expressed in the "remaining parameters" \beta and \delta (top of page 805)

5. Furthermore they (MS) lobby for a higher precision Kennedy Thorndike experiment for the purpose of constraining \beta (bottom of 813 and top of 814)

OK, but I would just like to know exactly what it is that you want me to show by calculations first. I said from the beginning (see post #1 here: http://www.bautforum.com/showthread.php?t=38765) that: "There are two completely separate issues here: 1) that the various "test theories" of special relativity are typically applied to guide the design of experiments that probe for violations of Lorentz symmetry, and 2) that the interpretation of any one-way speed measurement is always coordinate-system dependent. Although I am personally interested in designing experiments that probe for unique violations of Lorentz symmetry, whether or not any such violation exists is not the primary issue here. For the purposes of this thread, we are only concerned with understanding the coordinate-system dependent (or otherwise) nature of one-way speed of light measurements.".

Interestingly enough, both the Gagnon paper and the Krisher one are...higher precision forms of Kennedy Thorndike (I said this many times in this thread). Do you really think that the authors, reviewers and editors didn't know what they were doing?

What are the odds of this being true?
And if things fell through the cracks, don't you think that MS would have come back to point out the incorrectness of the two papers? They (MS) seem quite capable in analyzing experiments...

I think that they made some mistakes.

 

[clj4]

Show you that Eq. (8) is identically zero for SR, GGT, and any arbitrary value of \epsilon?

Here's Eq. (8):

\delta \phi/\phi_0=w(1+2\alpha)cos\theta+w^2(1/2+\delta-\beta)cos^2\theta​

For both SR and GGT: \alpha=-1/2, \beta=1/2, and \delta=0.

Variations of these parameters from SR and GGT would indicate violations of both (e.g., a violation of local Lorentz symmetry); we're assuming perfect Lorentz symmetry in this discussion.

(1+2\alpha)=0

(1/2+\delta-\beta)=0

\delta \phi/\phi_0=0​

We are still talking the Krisher paper, right? The Krisher paper is not GGT.
See again posting 303:

In between these two values lies an infinite number of values for
\epsilon , an infinite number of clock synchronization schemes and an infinite number of theories different from SR.
It is common practice to use these fully parametrized theories (BTW, there are two more parameters \alpha and \beta) as "test theories" of SR. The Krisher paper is an example of application of such a test theory. There are many more papers , especially in particle physics, that employ the fully parametrized MS theory as a means of testing SR. They employ more or less the same mechanism:
-an experiment is outlined
-the fully parametrized MS theory is used to make a prediction for the experiment outcome that will differ from SR
-a set of expressions in the \alpha,\beta,\epsilon parameters is being obtained
-the theoretical data is compared with the experimental data and the parameters are constrained to values that are very close to zero

I think that they made some mistakes.

duh...

Look, there are 5 papers, all written to the standards of Physical Reviews A and D. You would need to refute all of them (if one is correct, your whole theory collapses). Try to give the authors the respect and refute the papers in a way that is tractable (i.e. with mathematical formulas, not with words, we have gone thru 300 posts and there is no such mathematical refutation in sight). Imagine that you were going to submit your refutation to Phys Rev A or D and you were hoping to have them publish it. Until you do this, the papers stand.

 

[Aether]

Look, there are 5 papers, all written to the standards of Physical Reviews A and D. You would need to refute all of them (if one is correct, your whole theory collapses). Try to give the authors the respect and refute the papers in a way that is tractable (i.e. with mathematical formulas, not with words, we have gone thru 300 posts and there is no such mathematical refutation in sight). Imagine that you were going to submit your refutation to Phys Rev A or D and you were hoping to have them publish it. Until you do this, the papers stand.

Until you fulfill your commitment to show us a "disproof" of gregory's argument that Gagnon forgot to use the GGT form of the Lorentz force, then Gagnon (Phys Rev A) stands both as recanted by the authors and thoroughly refuted by gregory and myself for the purposes of this discussion. Does anyone besides clj4 disagree? Gregory and I have shown that for the purposes of this discussion the title of Krisher's paper is not justified. Does anyone besides clj4 disagree?

What are the other three papers that you are referring to?

 

[clj4]

Until you fulfill your commitment to show us a "disproof" of gregory's argument that Gagnon forgot to use the GGT form of the Lorentz force, then Gagnon (Phys Rev A) stands both as recanted by the authors and thoroughly refuted by gregory and myself for the purposes of this discussion. Does anyone besides clj4 disagree? Gregory and I have shown that for the purposes of this discussion the title of Krisher's paper is not justified. Does anyone besides clj4 disagree?

What are the other three papers that you are referring to?

1. I will disprove gregory, don't worry. His so-called disproof is mathematically incorrect at best, most likely irrelevant. I just wanted to give him time to recheck his calculations and reconsider.
But where is your promised disproof? Got stuck in partial differential equations? Or is it waveguide theory that got you stumped?

2. This is not a political poll, until you get the papers disproved, you have nothing (the other two papers are the two Gagnon papers and the one by Spavieri http://prola.aps.org/abstract/PRA/v34/i3/p1708_1 ). So you have 5 to work with. I bet that the members of this forum may be able to come up with more. And you need to refute all of them in order to prove your "aether" theory right :-)

3. You haven't shown anything on the Krisher paper either. Do you think that declaring your work done means that you are done?

4. You seem to forget that we are under the BAUT rules: if you advance an "Against the Mainstream Theory" you need to prove your point. We moved the thread here because of the LatEx capabilities such that you could prove your assertions mathematically, so far all we got from you was a collection of arithmetic errors.

 

[Aether]

1. I will disprove gregory, don't worry. His so-called disproof is mathematically incorrect at best, most likely irrelevant. I just wanted to give him time to recheck his calculations and reconsider.
But where is your promised disproof? Got stuck in partial differential equations? Or is it waveguide theory that got you stumped?

When you disprove greogry, then we'll have something to talk about.

2. This is not a political poll, until you get the papers disproved, you have nothing (the other two papers are the two Gagnon papers and the extra one from 1986-you'll have to find it on this thread).

Yes, it is a political poll. At least for the purposes of this discussion it is. Sure, we can take it to Physical Review later with a more rigorous paper.

3. You haven't shown anything on the Krisher paper either. Do you think that declaring your work done means that you are done?

Unless someone besides you disagrees, then for the purposes of this discussion, yes, my work is done. A rigorous paper on this subject could be submitted to Physical Review at some point, but what has been said seems sufficient for the purposes of this discussion.

4. You seem to forget that we are under the BAUT rules: if you advance an "Against the Mainstream Theory" you need to prove your point. We moved the thread here because of the LatEx capabilities such that you could prove your assertions mathematically, so far all we got from you was a collection of arithmetic errors.

Let's see if anyone here agrees with you that Gagnon (Phys Rev A) or Krisher's title is still viable for the purposes of this discussion. If so, then we'll continue discussing it. I'm not denying at all that we could do a more thorough and concise job in a Physical Review article, but you have not been fairly representing the mainstream position on this in any event.

 

[clj4]

OK, what the heck-we aren't going to get anything meaningful in terms of math out of you beyond of what you already did.
Let's make it a poll, this should be fun.
I will give the gregory disproof, this is sure going to be some fun as well.
In the meanwhile, why don't you send an email to C.M.Will , tell him that he doesn't "get" the MS theory and that his paper is all wrong? This should be even greater fun, maybe he weighs in.
You know, this guy:

http://arxiv.org/PS_cache/gr-qc/pdf/9811/9811036.pdf

 

[Aether]

OK, what the heck.
Let's make it a poll, this should be fun.
I will give the gregory disproof, this is sure going to be some fun as well.

OK, great.

 

[Aether]

In the meanwhile, why don't you send an email to C.M.Will , tell him that he doesn't "get" the MS theory and that his paper is all wrong? This should be even greater fun, maybe he weighs in.
You know, this guy:

http://arxiv.org/PS_cache/gr-qc/pdf/9811/9811036.pdf

Where does it say in that paper that SR and GGT are not empirically equivalent (e.g., predict different outcomes for any experiment)? That is what this discussion is about.

"Provided that one deals with observable quantities, the outcome of physical experiments of this type is unique in all cases and is independent of synchronization; thus the TPA and other such one-way experiments do provide valid tests of possible violations of SRT. It is important to emphasize that those violations are embodied in functional forms of a, b, and d that could differ from those quoted above not in the form of \epsilon, which is arbitrary and irrelevant." -- C.M. Will, Clock synchronization and isotropy of the one-way speed of light, Physical Review D 45(2), p.404 (1992).

 

[clj4]

Where does it say in that paper that SR and GGT are not empirically equivalent (e.g., predict different outcomes for any experiment)? That is what this discussion is about.

The paper is listed simply to introduce you to the author, you know, the one that "made some mistakes" in the Krisher paper (in your opinion, listed at the bottom of post 313)

"...thus the TPA and other such one-way experiments do provide valid tests of possible violations of SRT. It is important to emphasize that those violations are embodied in functional forms of a, b, and d that could differ from those quoted above not in the form of \epsilon, which is arbitrary and irrelevant." -- C.M. Will, Clock synchronization and isotropy of the one-way speed of light, Physical Review D 45(2), p.404 (1992).

Yes, so you just added another one that you need to refute? You are simply reinforcing the point that the MS theory is a valuable tool for testing violations induced by the assumption that one way light speed may is considered to be anisotropic under MS. C.M.Will is practically saying the same thing he's saying in the Krisher paper.
You have been great in providing quotes, you have been very generous in buying and sharing the papers, for all this : thank you. Would you care to share this one with all of us?

 

[clj4]

OK, let's see the popular votes. In order to prevent fraud :

1. To vote, you have to have been a registered member on 3/7/06 when this thread was started:

https://www.physicsforums.com/showpost.php?p=930823&postcount=107

This precludes "ballot stuffing"

2. People that registered after 3/7 (and especially after 3/27) should abstain from voting

In other words don't bring your colleagues, friends and family into this. I will not bring mine.

 

[clj4]

"gregory"'s incorrect and irrelevant disproof of Gagnon

I will divide this into two parts:

1. In the first part I will show that "gregory's" disproof of Gagnon is both incorrect and irrelevant.

2. Once we agree that point 1 has been demonstrated , I will show that Gagnon et. al did their calculations for the boundary conditions of their partial differential equation correctly.

Here goes part 1:

No. One way speed measurements are not valid.

B] The Lorentz force law is different. I have shown that.

1. While this is true, it will turn out to be irrelevant. Which nullifies your attack from the start.
The main reason is that the Lorentz force mixes E and H (B) in the same formula which turms out to mess up your proof.

The boundary conditions on a metal are such that: there is no force on charges in the material, there can only be a force on the surface charges perpendicular to the surface. Because the Lorentz force law normally looks like \vec{F} = q(\vec{E} + \vec{v} \times \vec{B}) this is equivalent to the boundary condition on the fields of \bf{E}_\parallel=0, \bf{B}_\perp=0.

2. This is both irrelevant AND incorrect.
The mode you are describing is TEM: \bf{E}_\parallel=0, \bf{B}_\perp=0, a mode that is NOT used by the paper and a mode that is most never used in practice. At this point your counter is NULLIFIED as IRRELEVANT.

The waveguide mode used by Gagnon is TE (page 1768 bottom). This mode has ONLY the boundary condition \bf{E}_\parallel=0, there is no boundary condition for \bf{B}_\perp
The TE mode DOES not have any boundary condition on B (this is why it is called TE). Another mode, TM has the boundary condition \bf{B}_\perp=0. This mode is NOT used in the experiment.
For detailed explanations on wave guide theory see [1]

Because the force law is not the same in GGT frames, the boundary condition is not the same either.

3. This statement is incorrect and irrelevant as well. First off, since E and H are mixed together in the expression of the Lorentz force, you cannot draw the conclusion that E,H do not conserve form under transformation just because the Lorentz force doesn't.

Either way, the Lorentz force is not the issue. Conservation of the condition \bf{E}_\parallel=0 AT THE BOUNDARY is the ONLY ONE. And Gagnon et. al treated the boundary condition \bf{E}_\parallel=0 correctly.
This will be demonstrated in a separate post once we get agreement on this post.

Summary:

The calculations in the Gagnon experiment have been shown to be wrong on their starting assumptions. They are wrong.

Turns out that "gregory"'s disproof is both incorrect and irrelevant.
In the second post, I will show that Gagnon did his calculations correctly - you need to respect him and the reviewers of Phys Rev D more.

[1] Fundamentals of Electromagnetics with Engineering Applications – Stuart M. Wentworth (p338-355)

 

[clj4]

Gagnon did his calculations right

This is part 2.
Turns out that the boundary conditions conserve, Gagnon et al. had it right.
Please look at the attached file.
This by itself means that the Gagnon paper is right.
The consequences are clear. M-S were very careful in their paper when they asked for higher precision KT experiments. Gagnon gave them exactly this.

Hans, there is a chance that you may have a very good second order experiment in the spirit of what MS are asking for. If you couple it with a theoretical explanation (GGT would be the simplest, the parametrized MS use by Krisher would be the next more complex one) you could have a very nice paper worth publishing in Phys Rev.

 

[reilly]

Special relativity is empirically equivalent to "...an ether theory taking into account time dilation and length contraction but maintaining absolute simultaneity..." R. Mansouri & R.U. Sexl, A Test Theory of Special Relativity: I. Simultaneity and Clock Synchronization, General Relativity and Gravitation, Vol. 8, No. 7 (1977), pp. 497-513. In such an empirically equivalent ether theory, the speed of light varies with direction.

This is not just some cherry-picked statement from an article that I pulled out of thin a- er, um air, it is the very crux of a famous paper referenced by most if not all of the experiments published over the past 30 years which measure local Lorentz invariance.

If what you say is true, why have most of us never heard of this theory? If I'm mistaken about the breadth of knowledge, has anyone done quantum theory, the equivalent to the Dirac eq., or demonstrated full equivalence to verified SR results?

Regards,
Reilly Atkinson

 

[NotForYou]

Because I am jumping in here, please have patience if I bring up points already agreed upon.

If what you say is true, why have most of us never heard of this theory? If I'm mistaken about the breadth of knowledge, has anyone done quantum theory, the equivalent to the Dirac eq., or demonstrated full equivalence to verified SR results?

The reason you haven't heard of it is because as a theory it has no predictive power (actually, due to the messiness, it probably hinders search for new physics if one constrains themselves to that theory). It is defined to have the same physical laws in one coordinate system as SR. In any other frame, you must transform back into this special frame, do the calculations, and transform back. So by definition the predictions for all experiments will agree with SR.

However it doesn't provide any predictive power for new physics, and makes calculations messy. So it has no real benefit except to remind us that if we really wanted to, we could describe the universe with any coordinate system we choose.

The mode you are describing is TEM: , a mode that is NOT used by the paper and a mode that is most never used in practice. At this point your counter is NULLIFIED as IRRELEVANT.

I am sorry, but you appear to be misunderstanding what a boundary condition even is. The boundary conditions gregory_ listed are true for any electromagnetic fields by a conductor (this includes all of TE, TM, and TEM waves in a waveguide). Just because B_\perp=0 on the boundary, does not require this to be true IN the waveguide.

So to repeat:
The boundary conditions on a metal are such that: there is no force on charges in the material, there can only be a force on the surface charges perpendicular to the surface. Because the Lorentz force law normally looks like \vec{F}=q(\vec{E}+\vec{v} \times \vec{B}) this is equivalent to the boundary condition on the fields of E_\parallel =0, B_\perp=0.

So clj4's misunderstanding of E&M lead him to incorrectly conclude that Gagnon's mistake didn't matter.


Also, clj4, please note that the boundary conditions come from the force on charges in the metal. Because the Lorentz force has a different form according to GGT, the boundary conditions on E and B are different. That is the point you seem to keep missing.

 

[clj4]

I am sorry, but you appear to be misunderstanding what a boundary condition even is. The boundary conditions gregory_ listed are true for any electromagnetic fields by a conductor (this includes all of TE, TM, and TEM waves in a waveguide). Just because B_\perp=0 on the boundary, does not require this to be true IN the waveguide.

I understand what a boundary condition is very well. The point is that TE mode does NOT have B_\perp=0 as a boundary condition. Clear?

The other points are that you or gregory_ or whoever cannot infer boundary condition mismatches from Lorentz force mismatches

Because the Lorentz force has a different form according to GGT, the boundary conditions on E and B are different. That is the point you seem to keep missing.

You cannot infer that. E and H are mixed together in the expression of force. You cannot even use the force mismatch in order to infer the boundary condition mismatch under transformation. You need to do it from base principles. And when you do that you find that E as a boundary condition is invariant (zero transforms into a zero)

 

[quantumdude]

Let's stick to physics. I've just edited out the recent extracurricular activity, which has no place in this thread.

 

[clj4]

Let's stick to physics. I've just edited out the recent extracurricular activity, which has no place in this thread.

Thank you, sounds good to me. Let's stick to physics.

 

[NotForYou]

I understand what a boundary condition is very well. The point is that TE mode does NOT have B_\perp=0 as a boundary condition. Clear?

Sorry. Your understanding of electrodynamics is wrong.

- Reread that book you listed.
- Or, read this: http://maxwell.byu.edu/~spencerr/phys442/node6.html
- Or read Griffith's "Introduction to Electrodynamics" (page 405 in the 3rd edition).

Until you learn what a boundary conditions is and what the boundary conditions are here, it will be difficult to discuss this topic with you. Please take the time to read up on this topic.