Silvertooth's Experiment Cannot See the Ether
Tom Roberts
Author: Tom Roberts tjro...@lucent.com
Date: August 13, 1999
Several frequent contributors to this newsgroup tout Silvertooth's
experiment [1] [2] as a direct measurement of the ether. In a previous
article [3] I discussed several instrumentation effects which affect his
experiment and which he did not address in the articles I referenced. In
this article I show that very general assumptions about the properties
of light imply that in addition to SR and LET, _every_ reasonable ether
theory predicts a null result for Silvertooth's experiment. By
"reasonable ether theory" I mean a theory which is not already refuted
by other experiments. So while he reported a non-null result, it cannot
correspond to the ether as described by any reasonable ether theory.
The properties used are:
(A) The round-trip speed of light is isotropically c in any lab on earth
for a table-top experiment.
Note that this is supported by numerous experiments, including
the MMX, Brillet and Hall, and all of the speed-of-light
measurements in all the standards bodies in the world. This is
so well established that the meter is now defined in terms of
the round-trip speed of light.
(B) Light can be accurately described as a traveling wave, and when
multiple waves are present their amplitudes add linearly.
I do not require that light actually _be_ a traveling wave,
merely that it can be accurately described as one. This is the
physical optics approximation to the Maxwell's equations, and
also the corresponding large-photon-number limit of QED (in
which light most definitely is not such a wave, but can be
described by one to an incredibly-accurate approximation for
experiments like this). Note that this is also supported by
numerous experiments, including Wiener's experiments [4],
Young's two-slit experiment, and practically any diffraction
or interference experiment.
My discussion here applies to the entire _class_ of theories which
satisfy these two assumptions, regardless of whether or not they
include an ether. This class includes SR, LET, LPET, and every ether
theory seriously advocated in this newsgoup during the past 2 years
(it's pretty difficult to deny either (A) or (B) seriously).
OK, there are some additional assumptions, such as the speed
of light in any given direction is constant, independent of
either time or position, etc.
Basically I will show that the properties of Wiener fringes (and the
interference pattern in general for counter-propagating light rays)
are _independent_ of any assumed anisotropy in the one-way speed of
light. This is consistent with the general observation that no first-
order experiment is possible -- synchronization of clocks is an
arbitrary choice of the experimenter, and the one-way velocity of light
depends directly upon this choice as do all first-order experiments;
such an arbitrary choice _cannot_ affect any physical phenomenon (at
most it only affects how we use those clocks to describe it).
Some people claim that since Silvertooth's experiment uses
no clocks that clock synchronization is not an issue for it.
That is true of the experiment itself, but NOT for any
analysis of it. As soon as one claims "the wavelength of this
light ray is X", one has implicitly assumed a simultaneity
convention, because that is saying that the phase of the wave
is the same at any two points separated by X _AT_THE_SAME_
_TIME_. Similarly for "the speed of this light ray is c1".
This article presents a theoretical analysis of an idealized version of
Silvertooth's experiment. So I will assume that the lasers emit a
perfectly monochromatic light ray which is unaffected by any feedback
into the laser, the ray has no divergence or diffraction and is a
perfect plane-wave front, the apparatus is perfectly rigid, all optical
components are perfect, etc. As is the usual practice, it is up to the
experimenter to discuss differences between the actual experiment and
the idealized one. See [3] for a discussion of important differences
which Silvertooth failed to mention.
I. Wiener Fringes
-----------------
Consider the following physical situation:
--------- light path | |
| laser |>---------<->-----------------|----<->----| Mirror
--------- | |
Detector |-> x axis
Let the period of the light be T, and let w = 2pi/T. Let the right-going
speed of light be c1, the left-going speed of light be c2, and the round-
trip speed of light be c. Define the origin of the x axis at the mirror,
with +x to the right, as shown.
From (A) above we know that the round-trip speed of light is c; we just
assume an arbitrary 1-way anisotropy by letting c1 and c2 differ from c.
Using this, (B) above, and the assumed monochromaticity of the laser,
the emitted (right-going) light wave can be described as:
emitted = A sin((w/c1) x - w t) A=constant {1}
Note that I have arbitrarily selected t=0 to correspond to an instant
when the incident wave is 0 at the mirror x=0. No physical consequence
can arise from this choice; it is merely convenient.
Because there is no light at all to the right of the (perfect) mirror,
and because a lightwave is continuous, the wave must always be 0 at x=0.
So the reflected (left-going) wave is:
reflected = -A sin(-(w/c2) x - w t) {2}
Clearly the reflected wave has the same frequency as the incident wave.
From (A) above, c1, c2, and c are related by:
1/c1 + 1/c2 = 2/c {3}
1/c2 = 2/c - 1/c1 {3a}
Let us now consider the wave intensities at the set of points
{x = -mcT/2 = -m pi c/w, m integer >= 0}:
emitted = A sin(-m pi c/c1 - w t) {4}
reflected = -A sin(+m pi c/c2 - w t) {5}
= -A sin(-m pi c/c1 - w t + 2 pi m) {6}
= -emitted
Clearly the net intensity (emitted+reflected) is always 0 at these
points, and we have found the nodes of the standing wave. Note that
their spacing is _independent_ of c1 and c2, and depends only on the
round-trip speed of light and the frequency of the wave we are using
(cT = wavelength (round-trip), and these nodes are spaced at half-
wavelength intervals).
Let's examine the midpoints {x = -mcT/2-cT/4, m integer >= 0}:
emitted = A sin(-m pi c/c1 - pi c/(2 c1) - w t) {7}
reflected = -A sin(+m pi c/c2 + pi c/(2 c2) - w t) {8}
= -A sin(-m pi c/c1 - pi c/(2 c1) - w t + 2 pi m + pi)
= emitted {9}
And we have indeed found the antinodes, with total amplitude 2A.
Consider them at time t=0:
total = 2A sin(-m pi c/c1 - pi c/(2 c1)) {10}
This snapshot is clearly a set of antinodes spaced at cT/2 (half
wavelength) with approximately alternating signs and varying heights
(because c/c1 ~ 1). The phase difference between antinodes (which
accumulates to cause this variation in amplitude of the antinodes in
this snapshot) is clearly pi(c/c1-1). If we plot the total wave as a
function of x at t=0, it is a modulated sinewave passing through
these nodes and antinodes.
Let delta be the distance between the successive largest antinodes
(independent of sign), so the accumulated phase difference must add
up to pi:
pi = m pi (c/c1 - 1) (determines m) {11}
delta = mcT/2 {12}
If we follow Silvertooth (Eqn. (5) of [2]) and write (v = velocity
of lab wrt the ether):
c1 = c / (1 + v/c) {13}
we get:
delta = (cT/2)(c/v) {14}
which is easily obtained from Silvertooth's eqns. (8) and (9) of [2]
(my cT is his \lambda, my delta is his _capital_ \Delta).
Note that even though this snapshot at t=0 contains a modulated
standing wave corresponding to Silvertooth's description, it is
_NOT_ what the detector measures, and is ultimately unobservable.
Silvertooth's detector measures the time-average of the square of
the wave [1] [4]. From {7} and {9} we see that at each antinode the
total wave is:
total = 2A sin(w t + P) {15}
Where P is a _constant_ phase (which differs for each antinode).
The time-average of total^2 is clearly the same for each antinode.
What is happening is that the different antinodes reach their maximum
values at different times in this coordinate system. These coordinates
were determined by the assumption that the one-way speed of light is
c1 and c2 in the two directions, and the various antinodes "just happen"
to not reach their maxima simultaneously in these coordinates (unless
c1=c2=c). But they all reach the same maximum value and all have the
same time average, and THAT'S what the detector measures. So as the
detector is moved along x in the figure above, it will display a sin^2
pattern with nodes equally spaced at cT/2 (wavelength/2), and with
_EQUAL-HEIGHT_ antinodes centered between them. This _is_ what real
experiments measuring Wiener fringes actually observe.
Remember I said that modulated standing wave is unobservable. Imagine
a different kind of detector which responds to the instantaneous value
of the wave, and is very thin (perhaps using a vastly lower-frequency
wave). Put two of them in the above figure, and align them at different
antinodes. If you want to observe the time difference between their
antinodes' maxima, you will need to bring the two signals together, and
in so doing you will introduce anisotropic delays which exactly cancel
the difference in times at which the antinodes reach their maxima --
the maxima will appear simultaneous at the comparator. This is
independent of which antinodes you select, of where you place the
comparator wrt them, and of the speed and path of the signals (as long
as they are electrical or optical and identical-length paths). Note
that this is independent of c1 and c2 (or v), and is exactly the same
conclusion one obtains for an isotropic one-way speed of light. Just
one more instance of the one-way speed of light being unobservable.
II. Silvertooth's Experiment
----------------------------
Here is a diagram of the important parts of Silvertooth's experiment [2].
Note that he has published descriptions of other similar experiments,
but this is the one I am discussing. In this diagram, "o" represents a
bidirectional light beam. This is not quite his actual experimental
setup, but is a conceptual drawing sufficient to describe and analyze
his experiment:
/ \
M1 /oooooooooooooooooooooooo\ M2
/ o o \
o o
o ........... o
o . --- . o
o . | | . o
o . |L| . o
Detector --- . | | . o
o . --- . o
o . o . o
o . o . o
o . o . o
o .BS \o . o/
o . \ooooooooooo/
o . o\ . / M5
o . o .
\o . o/ .
\oooooooooo/ .
M3 \ . / M4 .
. . <---->
........... Movable table
The components enclosed in the dotted line are mounted on a table which
can move left and right by precisely-controlled small amounts (~mm max).
There is a Michelson interferometer (not shown) with one mirror mounted
on the table, which is used to monitor its position. The He-Ne laser L
generates light which is injected into the ring interferometer
consisting of BS-M4-M3-Detector-M1-M2-M5-BS. The detector is a custom
phototube with a thin photocathode 500 A (~1/10 wavelength) thick.
The basic operation of the experiment is to move the table an integral
number of fringes in the Michelson interferometer and observe the
effect on the signal from the phototube detector in the ring
interferometer. The Michelson interferometer is powered by another
He-Ne laser with approximately the same frequency as the one shown.
As this is an analysis of an idealized experiment, I will
assume that the two lasers are _exactly_ the same frequency.
And I will ignore the difficulties of aligning the ring
interferometer to have a circumference _exactly_ an integral
number of half-wavelengths. It seems that later versions
of this experiment included using a single laser for both
interferometers, and pinhole isolators used to make each
light path laser->detector unidirectional (thus preventing
the ring from being an interferometer all the way around), so
these assumptions are quite reasonable, and apparently
correspond to later versions of the experiment.
Silvertooth claims that the standing wave in the detector depends upon
the table's position as in eqns. {10}-{14} above. But the discussion
around eqn. {15} shows that this is in error -- he erroneously claims
the standing-wave at a snapshot (e.g. t=0) corresponds to the signal in
his detector (he doesn't explicitly claim this, but his usage of {14}
directly implies it). In actuality the amplitudes of the different
antinodes are all equal, as is their signal into his detector.
The derivation {1} - {15} assumed a mirror at x=0. That was only needed
to specify the position of a node; it is a general derivation for any
pair of counter-propagating waves subject to the assumptions given,
one merely finds a node and uses its position to define x=0. If one
tunes the apparatus to have a given phase difference between the two
rays at the detector (and therefore a given signal output), then if
one moves the table left or right by an amount
m cT/2 (m integer) {16}
Then the initial phase relationship at the detector will be preserved,
as will the signal output. But {16} is also the spacing of the fringes
in the Michelson interferometer. So Silvertooth's technique of moving
the table an integral number of Michelson fringes will keep the detector
output constant. That is a null result. That is, one can assume _ANY_
velocity v in {13} (|v|<c), and one obtains the same prediction: the
detector output won't change when the table is moved by an integral
number of Michelson fringes.
Note that moving the table introduces an overall phase
offset in each wave, but does not affect the difference in
their phases or the signal in the detector. This is equivalent
to an overall time offset in the t coordinate, which is
unobservable (the discussion after {15} applies).
Silvertooth's publication of a positive result is puzzling, given that
there is no reasonable theory which predicts a positive result. It is
also puzzling in that his paper [2] shows that the Michelson fringes
are independent of any assumed anisotropy in the one-way speed of light
(i.e. of any choice of v) -- apparently he never realized that his
demonstration also applies to his ring interferometer, and for the
same reasons. In [3] I discuss two different instrumentation effects
which clearly plagued his experiment [2]. It seems his later
publications deal with some but not all of these effects.
In view of the disconnect between his experiment and the class of
theories discussed here, it is desirable that his experiment be
repeated. But if such a repetition is to be believed, at a minimum
it must deal with all of the instrumentation effects described in [3].
References
----------
[1] Silvertooth and Jacobs, "Standing wave sensor", Applied Optics
_22_#9, p1274 (1983).
[2] Silvertooth, "Experimental detection of the ether", Speculations in
Science and Technology _10_#1, p3 (1986).
[3] Tom Roberts, article posted to sci.physics.relativity 7/28/1999,
Subject: Silvertooth Measured an Instrumentation Effect.
[4] Jenkins and White, _Fundamentals_of_Optics_. Different editions have
different page numbers; look up "Wiener's experiments" in the index.
Tom Roberts tjro...@lucent.com
Angel Garcia
Everything is passable up to this point. Here you are adding a new
and FALSE assumption; namely, that there is no phase-shift at
the 'perfect mirror'. Without yhis false assumption your (2) becomes
correctly:
reflected = .... (...... -wt + g)
where g is the phase-shift.
More later.
--
Angel, secretary of Universitas Americae (UNIAM). His proof of ETI at
Cydonia and index of book "TETET-98: Generacion del Hombre en Marte" by Prof.
Dr. D.G. Lahoz (leader on ETI and Cosmogony) can be studied at URL:
http://www.ncf.carleton.ca/~bp887 ***************************
Angel Garcia
> reflected = .... (...... -wt + g) ..... (2')
>
> where g is the phase-shift.
>
> More later.
The time was off. Let me give a hint to such error of Tom, although
possibly is not relevant for the Silvertooth experiment.
This is a very general and widespread error in all currently published
books and papers, to our knowledge. Such error is consequence of the
assumption that SR is valid even in 'its principle'. Since such
principle is NOT valid as shown as a fact by Silvertooth, the more
refined theory (replacing SR) has to be used to analize equation (2).
A 'perfect mirror' DOES NOT behave equally in the ether-frame as in
any other inertial frame. Equation (2') above with g=0 is ONLY valid
in the Ether-frame; in all other inertial frames moving at speed v
relative to Ether the phase-shift g is extremely minute but NOT ZERO
and depends essentially on orientation with respect to stars.
In fact a new (triggered by Silvertooth Effect) experiment (also
very difficult as it now seems) is possible by using such mere prediction
of the correct formula for the Wiener fringes. That is why Lahoz says
that "Wiener fringes hide the secret of the Ether". Lahoz will
explain these ideas in detail in his planned book TETET-99.
Rediscovery of the correct formula (2') above is not difficult and
very much welcome; Lahoz surely will mention it and include it in his
planned book.
......
>> The basic operation of the experiment is to move the table an integral
>> number of fringes in the Michelson interferometer and observe the
>> effect on the signal from the phototube detector in the ring
>> interferometer. The Michelson interferometer is powered by another
>> He-Ne laser with approximately the same frequency as the one shown.
.....
That is correct.
......
>> Silvertooth claims that the standing wave in the detector depends upon
>> the table's position as in eqns. {10}-{14} above. But the discussion
>> around eqn. {15} shows that this is in error -- he erroneously claims
>> the standing-wave at a snapshot (e.g. t=0) corresponds to the signal in
>> his detector (he doesn't explicitly claim this, but his usage of {14}
>> directly implies it). In actuality the amplitudes of the different
>> antinodes are all equal, as is their signal into his detector.
Yes. There are errors of interpretation of the experiment in the
published papers; but that is totally irrelevant for the experimental
FACT as Dr. Silvertooth explicitly says in his 1989 paper.
>> In view of the disconnect between his experiment and the class of
>> theories discussed here, it is desirable that his experiment be
>> repeated. But if such a repetition is to be believed, at a minimum
>> it must deal with all of the instrumentation effects described in [3].
OK. This is a nice (and not offensive) contribution. Of course
it has to be repeated... details left to be described by the new
experimenter which most probably will be Mr. Steve Silvertooth.
We all are waiting for new data. Instrumentation effects have been
already posted and dismissed by myself taking accurate account of
what Mr. Steve kindly communicated to us on behalf of his father.
Angel Garcia
Angel Garcia (bp...@FreeNet.Carleton.CA) writes:
> Tom Roberts (tjro...@lucent.com) writes:
>> Subject: Silvertooth's Experiment Cannot See the Ether
>> Author: Tom Roberts tjro...@lucent.com
>> Date: August 13, 1999
>>
>> Because there is no light at all to the right of the (perfect) mirror,
>> and because a lightwave is continuous, the wave must always be 0 at x=0.
>> So the reflected (left-going) wave is:
>>
>> reflected = -A sin(-(w/c2) x - w t) {2}
>>
>> Clearly the reflected wave has the same frequency as the incident wave.
>
> Everything is passable up to this point. Here you are adding a new
> and FALSE assumption; namely, that there is no phase-shift at
> the 'perfect mirror'. Without yhis false assumption your (2) becomes
> correctly:
> reflected = .... (...... -wt + g)
>
> where g is the phase-shift.
>
> More later.
>
>
be g NOT EQUAL to zero in the case of the compound net Wiener fringe:
emitted + reflected = .... (... wt + g)
with g depending on orientation.
......
>> m cT/2 (m integer) {16}
>>
>> Then the initial phase relationship at the detector will be preserved,
>> as will the signal output. But {16} is also the spacing of the fringes
>> in the Michelson interferometer. So Silvertooth's technique of moving
>> the table an integral number of Michelson fringes will keep the detector
>> output constant. That is a null result.
What kind of proof is this ?. A proof by DOGMA ?. IF your (16)
is spacing for both Wiener and Michelson fringes then you are right,
but that is precisely what has to be proven. Foolish proof which
amounts to nothing in your paragraph above !. No more to it !.
C.J. Luke
>Subject: Silvertooth's Experiment Cannot See the Ether
>Author: Tom Roberts tjro...@lucent.com
>Date: August 13, 1999
>
>
>Several frequent contributors to this newsgroup tout Silvertooth's
>experiment [1] [2] as a direct measurement of the ether. In a previous
>article [3] I discussed several instrumentation effects which affect his
>experiment and which he did not address in the articles I referenced. In
>this article I show that very general assumptions about the properties
>of light imply that in addition to SR and LET, _every_ reasonable ether
>theory predicts a null result for Silvertooth's experiment. By
>"reasonable ether theory" I mean a theory which is not already refuted
>by other experiments. So while he reported a non-null result, it cannot
>correspond to the ether as described by any reasonable ether theory.
>The properties used are:
>
> (A) The round-trip speed of light is isotropically c in any lab on earth
> for a table-top experiment.
>
> Note that this is supported by numerous experiments, including
> the MMX, Brillet and Hall, and all of the speed-of-light
> measurements in all the standards bodies in the world. This is
> so well established that the meter is now defined in terms of
> the round-trip speed of light.
Though I am not an 'aetherist' or an 'SRist' I take exception to the
above:
Since the day that Physics 'forsook' a physical measure for distance
and replaced it with a 'light based reference' (first as a measure of
the number of wavelengths of red light emitted from an isotope of
Kripton gas then as simply a calculation based on the defined speed of
light) all of the experimental data that you refer to above is
tainted. Tainted in the sense that any experiment that uses an
'optical based' method of measuring distance will mask any anisotropy
due to 'movement' through any 'aether medium'.
>
> (B) Light can be accurately described as a traveling wave, and when
> multiple waves are present their amplitudes add linearly.
>
> I do not require that light actually _be_ a traveling wave,
> merely that it can be accurately described as one. This is the
> physical optics approximation to the Maxwell's equations, and
> also the corresponding large-photon-number limit of QED (in
> which light most definitely is not such a wave, but can be
> described by one to an incredibly-accurate approximation for
> experiments like this). Note that this is also supported by
> numerous experiments, including Wiener's experiments [4],
> Young's two-slit experiment, and practically any diffraction
> or interference experiment.
>
Since 'A' is not a given, the rest is pointless.
"The lack of reason is overcome by the passion of belief"
< c...@totcon.com >
Angel Garcia
C.J. Luke (c...@totcon.com) writes:
> Tom Roberts <tjro...@lucent.com> wrote:
>
>> (A) The round-trip speed of light is isotropically c in any lab on earth
>> for a table-top experiment.
>>
>> Note that this is supported by numerous experiments, including
>> the MMX, Brillet and Hall, and all of the speed-of-light
>> measurements in all the standards bodies in the world. This is
>> so well established that the meter is now defined in terms of
>> the round-trip speed of light.
>
> Though I am not an 'aetherist' or an 'SRist' I take exception to the
> above:
> Since the day that Physics 'forsook' a physical measure for distance
> and replaced it with a 'light based reference' (first as a measure of
> the number of wavelengths of red light emitted from an isotope of
> Kripton gas then as simply a calculation based on the defined speed of
> light) all of the experimental data that you refer to above is
> tainted. Tainted in the sense that any experiment that uses an
> 'optical based' method of measuring distance will mask any anisotropy
> due to 'movement' through any 'aether medium'.
No, Luke, no. Tom is absolutely right on A). True that an 'apparent'
tainting seems to be at hand when one replaces a pure length by a
pure velocity c to measure a length. But realize that Tom is speaking
about 'roundtrip' and in such case the Ether theory gives accurately
a net value c EXACTLY identical as the speed of light in Ether in
which (and only in which) it is isotropic.
Thus using such ROUNTRIP value of c in any inertial frame to measure
a distance is accurate (except, one may doubt, as for 2nd. order
length contraction of actual rods and distances; but for Silvertooth
effect this is irrelevant because we are dealing with genuine
1st. order in v/c).
Tom Roberts
> Though I am not an 'aetherist' or an 'SRist' I take exception to the
> Since the day that Physics 'forsook' a physical measure for distance
> and replaced it with a 'light based reference' (first as a measure of
> the number of wavelengths of red light emitted from an isotope of
> Kripton gas then as simply a calculation based on the defined speed of
> light) all of the experimental data that you refer to above is
> tainted. Tainted in the sense that any experiment that uses an
> 'optical based' method of measuring distance will mask any anisotropy
> due to 'movement' through any 'aether medium'.
Most experiments, Silvertooth's included, base actual distance
measurements on physical meter sticks and such -- no use of "light-
based reference" for distance is actually used. Your condemnation
here is completely irrelevant to actual experimental physics _AS_
_IT_IS_ACTUALLY_PERFORMED_, in most cases and this one in particular.
Silvertooth gives no measurements of his apparatus, and no
dimensions are really necessary -- such measurements as you
condemn are neither made nor needed.
In your claim of "masking" of "anisotropy due to 'movement' through any
'aether medium'", the word "any" is a gross overstatement -- it is
"masked" only for aether theories for which the aether is in principle
undetectable (i.e. for theories in which the round-trip speed of light
is isotropic).
Note that as I said, the definition is really in terms of the round-trip
speed of light, not the one-way speed, and your criticism falls apart.
Measurements of the _anisotropy_ in the round-trip speed of light need
not make any distance measurement at all.
> Since 'A' is not a given, the rest is pointless.
'A' is the claim that the round-trip speed of light is isotropically c.
A good experimental measurement of this is Brillet and Hall. In their
experiment the frequency of a laser was observed to be independent of
orientation to a few parts in 10^14 (IIRC). There is no use of any
length measurement at all in this experiment -- the relevant lengths
are the length of their reference laser and the length of their Fabry-
Perot interferometer. Neither of their measurements contribute to the
result; the experiment looked for _variations_ in them.
Your argument is completely bogus.
Tom Roberts tjro...@lucent.com
Paul B. Andersen
>
> Angel Garcia (bp...@FreeNet.Carleton.CA) writes:
> > Tom Roberts (tjro...@lucent.com) writes:
For EM-waves a "perfect mirror" is the surface of a perfect
conductor. The boundary condition which must be met on such
a surface is that the E-field i zero. That means that the E-field
in the incident and reflected wave always must be of the same
magnitude and of opposite direction, as in Tom's equation above.
(If we interprete A as the electric field of an EM-wave.)
> Equation (2') above with g=0 is ONLY valid
> in the Ether-frame; in all other inertial frames moving at speed v
> relative to Ether the phase-shift g is extremely minute but NOT ZERO
> and depends essentially on orientation with respect to stars.
But this is nonsense. There is no way this phase can depend on
the state of motion or orientation of the mirror - ether or no ether.
The boundary condition is that the E-field is zero. Thus the phase
difference must be pi. (The minus in Tom's equation.)
This is extremely well verified through experiments and
innumerable practical applications.
If we had got this wrong, the medium through which we now
are communicating would not work.
> In fact a new (triggered by Silvertooth Effect) experiment (also
> very difficult as it now seems) is possible by using such mere prediction
> of the correct formula for the Wiener fringes. That is why Lahoz says
> that "Wiener fringes hide the secret of the Ether". Lahoz will
> explain these ideas in detail in his planned book TETET-99.
> Rediscovery of the correct formula (2') above is not difficult and
> very much welcome; Lahoz surely will mention it and include it in his
> planned book.
He will have a hard time if he is going to make the boundary
conditions of a reflecting surface depend on it's orientation
in the ether. :-)
Paul
Frank Wappler
> Most experiments [...] base actual distance measurements on
> physical meter sticks and such --
> no use of "light-based reference" for distance is actually used.
> [...] is completely irrelevant to actual experimental physics
> _AS_IT_IS_ACTUALLY_PERFORMED_
While that may or may not be so for any experimental observations
themselves, it is _not_ so for any measurements derived from that.
By containing marks that are labelled with
"<rational number> <order_of_magnitude prefix> m",
the "meter sticks" used for actual distance measurements
in many experiments imply _that_
the corresponding "light-based" measurement would obtain
the same numerical value in units of "meter"
(with accuracy depending on the individual "meter stick").
Actual distance measurements on various individual physical
meter sticks, in various trials, can only be reproduced and
meaningfully put in relation to each other _through reference_
to that same unit and the reproducible measurement procedure
by which it is defined, i.e. the SR distance definition.
> Silvertooth gives no measurements of his apparatus, and no
> dimensions are really necessary
Does Silvertooth require or imply certain pairwise distances
to be "constant" throughout the various stages of
the experiment, or others to "vary in a definite way"?
Did he measure any of those relations?
Can those requirements, results, and experimental procedure
be reproduced using the "light-based measurements"?
> [...] the claim that the round-trip speed of light is isotropically c.
> A good experimental measurement of this is Brillet and Hall.
... "Improved Laser Test of the Isotropy of Space", PRL42(9), 549, 1979 ...
> In their experiment the frequency of a laser was observed to be
> independent of orientation to a few parts in 10^14 (IIRC).
> There is no use of any length measurement at all in this experiment
> -- the relevant lengths are the length of their reference laser
> and the length of their Fabry-Perot interferometer.
> Neither of their measurements contribute to the result;
> the experiment looked for _variations_ in them.
AFAIU, Brillet and Hall search for variations in the relative
_frequencies_ of a Fabry-Perot stabilized HeNe laser wrt. a
CH4 stabilized reference;
and derive a fractional frequency shift (consistent with zero) of
"delta_f / f = (1.5 +/- 2.5) * 10^(-15)".
Then they go on to state their result as a measurement of
directional anisotropy of space (consistent with zero):
"fractional length change delta_l / l = (1.5 +/- 2.5) * 10^(-15)".
How'd they do that?
Regards, Frank W ~@) R
and...@ibm.net
>
[snip]
> Though I am not an 'aetherist' or an 'SRist' I take exception to the
> above:
> Since the day that Physics 'forsook' a physical measure for distance
> and replaced it with a 'light based reference' (first as a measure of
> the number of wavelengths of red light emitted from an isotope of
> Kripton gas then as simply a calculation based on the defined speed of
> light) all of the experimental data that you refer to above is
> tainted. Tainted in the sense that any experiment that uses an
> 'optical based' method of measuring distance will mask any anisotropy
> due to 'movement' through any 'aether medium'.
>
The observed experimental variation in c using the old definition
of the meter was much smaller than any ether theory can explain.
In fact, it's so small that with the current precision of the standard
for the second, it's MORE precise to define the meter in terms of
c than the other way around which is why it's done this way now.
John Anderson
Tom Roberts
> you are adding a new
> and FALSE assumption; namely, that there is no phase-shift at
> the 'perfect mirror'. Without yhis false assumption your (2) becomes
> correctly:
> reflected = .... (...... -wt + g)
>
> where g is the phase-shift.
This g must be constant. It only offsets the reflected wave by a constant interval
in x. It has no effect on the overall analysis -- merely offset my x coordinate
by the appropriate amount.
For the experiment itself, there could presumably be such an offset for each
mirror, and these could change for each orientation. Thay cannot induce a
variation in the detector output as a function of table position. In other words:
they cannot induce a non-null result.
Tom Roberts tjro...@lucent.com
Tom Roberts
> A 'perfect mirror' DOES NOT behave equally in the ether-frame as in
> any other inertial frame. Equation (2') above with g=0 is ONLY valid
> in the Ether-frame; in all other inertial frames moving at speed v
> relative to Ether the phase-shift g is extremely minute but NOT ZERO
> and depends essentially on orientation with respect to stars.
Such a phase offset cannot vary with position of the table, and cannot
induce a variation in the detector output as a function of table position.
It cannot induce a non-null result.
Note that Jenkins and White ([4] in my original article) state that the node
is not at the surface of the mirror. For silver it is 0.043\lambda under
the surface. Again, all this does is introduce an offset in the x coordinate.
Tom Roberts tjro...@lucent.com
Tom Roberts
> IF your (16)
> is spacing for both Wiener and Michelson fringes then you are right,
> but that is precisely what has to be proven. Foolish proof which
> amounts to nothing in your paragraph above !. No more to it !.
The spacing of fringes in a Michelson interferometer is well knoan, and I
saw no reason to repeat the obvious (and a virtual repeat of my earlier
calculations). Silvertooth did it in his Specl. Sci. Tech. article (eqn. 7
of [2] in my original article). This is a simple and direct consequence of
the isotropy of the round-trip speed of light.
Tom Roberts tjro...@lucent.com
StevJensen
> Note that moving the table introduces an overall phase
> offset in each wave, but does not affect the difference in
> their phases or the signal in the detector. This is equivalent
> to an overall time offset in the t coordinate, which is
> unobservable (the discussion after {15} applies).
>
>
>Silvertooth's publication of a positive result is puzzling, given that
>there is no reasonable theory which predicts a positive result. It is
>also puzzling in that his paper [2] shows that the Michelson fringes
>are independent of any assumed anisotropy in the one-way speed of light
>(i.e. of any choice of v) -- apparently he never realized that his
>demonstration also applies to his ring interferometer, and for the
>same reasons. In [3] I discuss two different instrumentation effects
>which clearly plagued his experiment [2]. It seems his later
>publications deal with some but not all of these effects.
>
Why can't there be a GR effect similar to the clock tower experiment?
Nearly all points on the surface of the Earth rise and fall in the Sun's
gravity well. Recording an effect would seem to require a path length
difference in BS_M5_M2_M1_D to BS_M4_M3_D.
This would require that the effect be synchronized with the Earth's
rotation. Don't know if this is the case for Silvertooth, but other
MM type experiments show a local time synchronized artifact
for example Brillet & Hall.
Angel Garcia
"Paul B. Andersen" (paul.b....@hia.no) writes:
> Angel Garcia wrote:
>>
>> Angel Garcia (bp...@FreeNet.Carleton.CA) writes:
>> > Tom Roberts (tjro...@lucent.com) writes:
>> >> From (A) above we know that the round-trip speed of light is c; we just
>> >> assume an arbitrary 1-way anisotropy by letting c1 and c2 differ from c.
>> >> Clearly the reflected wave has the same frequency as the incident wave.
>>
>> >
>> > Everything is passable up to this point. Here you are adding a new
>> > and FALSE assumption; namely, that there is no phase-shift at
>> > the 'perfect mirror'. Without yhis false assumption your (2) becomes
>> > correctly:
>> > reflected = .... (...... -wt + g) ..... (2')
>> >
>> > where g is the phase-shift.
>> >
>> > More later.
>>
>> The time was off. Let me give a hint to such error of Tom, although
>> possibly is not relevant for the Silvertooth experiment.
>>
>> This is a very general and widespread error in all currently published
>> books and papers, to our knowledge. Such error is consequence of the
>
> But this is nonsense. There is no way this phase can depend on
> the state of motion or orientation of the mirror - ether or no ether.
> The boundary condition is that the E-field is zero. Thus the phase
> difference must be pi. (The minus in Tom's equation.)
>
> This is extremely well verified through experiments and
> innumerable practical applications.
> If we had got this wrong, the medium through which we now
> are communicating would not work.
.....
> He will have a hard time if he is going to make the boundary
> conditions of a reflecting surface depend on it's orientation
> in the ether. :-)
Yes, Paul. I committed an error in those two impromptu posts;
I should have applied my comments to the TOTAL = emitted + reflected
wave, which is the actual Wiener-fringes. It is only in the TOTAL
where the g phase-shift (depending on orientation) REALLY appears.
See details in my immediately (long) previous post answering to Tom.
Angel Garcia
Tom Roberts (tjro...@lucent.com) writes:
> Angel Garcia wrote:
>> you are adding a new
>> and FALSE assumption; namely, that there is no phase-shift at
>> the 'perfect mirror'. Without yhis false assumption your (2) becomes
>> correctly:
>> reflected = .... (...... -wt + g)
>>
>> where g is the phase-shift.
>
OK. I made an error about this g and here goes the meditated correction:
(As preliminary introduction:)
------------------------------------------------------------------------
From: Angel Garcia // Newsgroups: sci.physics.relativity
Subject: Re: Standing Waves /// Date: Aug 11 1999
Jackie (ja...@netcom.ca) writes:
> I was in a boat floating on the river.
> There were waves on the rocks. I measured the speed of all these waves
...
That has nothing to do with relativity. Surface-waves in canals
or rivers is a well known item of Hydrodinamics; very complicated, though.
See the classical textbook by Lamb.
Standing-waves of EM waves and light is an entirely different matter
which, Lahoz says, NEVER has been so far studied in a thorough maner.
These belong indeed to this relativity group; the reason being that
the medium where such standing-waves are 'seen' is not the same inertial
frame as the laboratories on Earth. Thus it is wrong to assume that
such 'standing-waves' of light (Wiener and Lippmann) are of the form:
[1] cos(a(x-ct)) + cos(a(x+ct))
as currently it is assumed in all published books and papers; such
form is ONLY valid in the Ether-frame (or, by analogy, in sound
within a pipe and even your mentioned ones in stagnant water when they
are really stationary).
-----------------------------------------------------------------------
From: Angel Garcia /// Newsgroups: sci.physics.relativity
Subject: Re: Silvertooth's Experiment Cannot See the Ether //Date: Aug 13 1999
Tom Roberts (tjro...@lucent.com) writes:
> Subject: Silvertooth's Experiment Cannot See the Ether
> Author: Tom Roberts tjro...@lucent.com
> Date: August 13, 1999
>
...
> The properties used are:
>
> (A) The round-trip speed of light is isotropically c in any lab on earth
> for a table-top experiment.
....
> (B) Light can be accurately described as a traveling wave, and when
> multiple waves are present their amplitudes add linearly.
....
> OK, there are some additional assumptions, such as the speed
> of light in any given direction is constant, independent of
> either time or position, etc.
Angel:
Yes. 'in any given direction..' means that 'for a given orientation
wrt to stars the speed of light is either c1 = c + v*cos(s) or
c2 = c - v*cos(s) '. Where s is the angle wrt ether-wind
> Basically I will show that the properties of Wiener fringes (and the
> interference pattern in general for counter-propagating light rays)
> are _independent_ of any assumed anisotropy in the one-way speed of
> light.
.....
> I. Wiener Fringes
> -----------------
>
> Consider the following physical situation:
>
>
> --------- light path | |
> | laser |>----------<->----------------|----<->----| Mirror
> --------- | |
> Detector |-> x axis
>
Angel:
It should be without the double arrow, namely:
--------- light path | |
| laser |>------------>----------------|----<->----| Mirror
--------- | |
Detector |-> x axis
as already discussed, to avoid coupling to the laser and to have pure
and idealized Wiener-fringes.
> Let the period of the light be T, and let w = 2pi/T. Let the right-going
> speed of light be c1, the left-going speed of light be c2, and the round-
> trip speed of light be c. Define the origin of the x axis at the mirror,
> with +x to the right, as shown.
>
> From (A) above we know that the round-trip speed of light is c; we just
> assume an arbitrary 1-way anisotropy by letting c1 and c2 differ from c.
> Using this, (B) above, and the assumed monochromaticity of the laser,
> the emitted (right-going) light wave can be described as:
>
> emitted = A sin((w/c1) x - w t) A=constant {1}
...
> Because there is no light at all to the right of the (perfect) mirror,
> and because a lightwave is continuous, the wave must always be 0 at x=0.
> So the reflected (left-going) wave is:
>
> reflected = -A sin(-(w/c2) x - w t) {2}
>
> Clearly the reflected wave has the same frequency as the incident wave.
Angel:
Boyo !. A consultation with Dr. Lahoz has put me in trouble because I
really made an error in following statements written just few hours
after Tom's post. Impromptus in Physics are usually bad as shown here.
I said:
[[1]] "Everything is passable up to this point. Here you are adding a new
and FALSE assumption; namely, that there is no phase-shift at
the 'perfect mirror'. Without yhis false assumption your (2) becomes
correctly:
reflected = .... (...... -wt + g)
where g is the phase-shift."
-----
and minutes later:
[[2]] "Let me give a hint to such error of Tom, although
possibly is not relevant for the Silvertooth experiment.
This is a very general and widespread error in all currently published
books and papers, to our knowledge. Such error is consequence of the
assumption that SR is valid even in 'its principle'. Since such
principle is NOT valid as shown as a fact by Silvertooth, the more
refined theory (replacing SR) has to be used to analize equation (2).
A 'perfect mirror' DOES NOT behave equally in the ether-frame as in
any other inertial frame. Equation (2') above with g=0 is ONLY valid
....
--------
These [[1]] and [[2]] above are my wrong impromptus;
fortunately less than 3 hours later I added:
"I will consult later with Dr. Lahoz about this g. May be he meant to
be g NOT EQUAL to zero in the case of the compound net Wiener fringes:
emitted + reflected = .... (... wt + g)
with g depending on orientation."
Lahoz:
Yes; of course it is only in the total Wiener-fringes and not in
the partial emitted or reflected waves where the g phase-shift
occurs.
Angel:
So what do you think of Tom's discussion of the Wiener-fringes ?
Lahoz:
It is good. No basic error in his statements about them. He has
really rediscovered my ideas about them and he will be honorifically
mentioned in the book, despite of being hypothetical issue in
his treatment.
Angel:
And bla, bla, bla.. Lahoz told me to post his general ideas about
Wiener-fringes (and not about correct interpretation of Silvertooth
experiment because Tom is still dennying it).
They go as follows:
"""Let Tot' = (1/2)cos(wx'/c - wt') + (1/2)cos(-wx'/c - wt') ...... (a)
be the total Wiener-fringes in the Ether-frame with mirror at rest in
the Ether, where the coordinates x' and t' are here used; (this is totally
equivalent to my preliminary post above in formula [1], but now with Tom's
notation to compare ideas). Trivially (a) above can be rewritten:
Tot' = cos(wx'/c) * cos(wt') ...... (b)
Listen to this !:
The wave equation |~| f = 0 (D'alambertian of f = 0) is
invariant by the Lorentz-Einstein-Poincare transformation:
x' = gamma ( x + vt) |
t' = gamma ( t + vx/cc) | ...... (c)
(where we are assuming that the Wiener-mirror is moving towards right
with velocity v relative to Ether; and x,t are the same coordinates as
in Tom's figure and formulae above)
it follows that, always in the Ether-frame, a NEW type of wave can
be 'created' which is GLUED to the moving mirror:
Tot~ = cos{ w[gamma(x'+vt')]/c } * cos{ w[gamma(t'+vx'/cc) } ....... (d)
That is merely by replacing (c) into (b) and still using x',t' as
coordinates for the Ether-frame; and now we have the EXACT
formula (d) of the Wiener-fringes as seen in the Ether. These waves (d)
are as real, but very different, as the original (b) when the mirror
was at rest in the Ether-frame: truly a new 'creation' of the Ether as
soon as the mirror started to move with velocity v relative to Ether.
They look, obviously, as a system (or packet) of travelling waves in
the Ether with nodes at x'+vt' = n*pi/2 * c/(w gamma)
(namely GLUED to the mirror) and maxima which are varying in time for
any given point x' in the Ether.
How these Wiener-fringes look-like in the moving coordinates attached
to the moving mirror?
Here we are going to drop 2nd. order terms in v/c to simplify an
analysis which is unnecessary for Silvertooth effect; and we get:
x = x' - vt' |
t = t' | ...... (e)
(galilean transformation which is correct to 1st. order in v/c). Or:
Tot ~= cos(wx/c) * cos[wt + wv(x+vt)/cc]
Tot ~= cos(wx/c) * cos(wt + wvx/cc) ...... (f)
And that is it !. Formula (f) gives the actual Wiener-fringes,
with correct approximation to 1st. order in v/c as needed.
In any general orientation, that v has to be replaced by v*cos(s), of
course. And now we see why IN FACT there is a phase-shift
g = wvx/cc ...... (g)
to be ALWAYS added to the temporal portion of the classical
standing-waves in the case of Wiener-fringes.
Such g is obviously dependent on orientation wrt stars.
This NEVER has been published before because, according to relativity,
formulae (e) have been replaced always by the Lorentz-Einstein
transformation (which is incorrect in pure kinematics as here)
yielding the usual standing-waves without the phase-shift (g) and
without ever explaining why they are glued to the mirror in the lab
which is moving relative to the Ether medium.
In summary: The Wiener-fringes are truly standing-waves with nodal
planes truly independent of orientation (against incorrect assumption
in Silvertooth's papers). But the temporal part wt of such unique
standing-waves has a systemic phase-shift (g) which varies linearly
with distance x to the mirror and is indeed dependent on orientation.
However the time-average of their energy (and, consequently, the
output in the Silvertooth's sensor) is totally independent of such
curious AND REAL phase shift (g); thus such output does not depend on
orientation.
Which gives a clue to why never before such Wiener-fringes have been
properly written and studied. """
Angel Garcia
>> is spacing for both Wiener and Michelson fringes then you are right,
>> but that is precisely what has to be proven. Foolish proof which
>> amounts to nothing in your paragraph above !. No more to it !.
>
> The spacing of fringes in a Michelson interferometer is well knoan, and I
> saw no reason to repeat the obvious (and a virtual repeat of my earlier
> calculations). Silvertooth did it in his Specl. Sci. Tech. article (eqn. 7
> of [2] in my original article). This is a simple and direct consequence of
We already know and said many (too many !) times that nor Silvertooth
nor Whitney got the correct and convincing 'theory' for such genial
experiment. They assume incorrectly that the Wiener fringes have
maxima dependent on orientation. You showed correctly that such assumption
is incorrect and now I have posted (long, immediately before this) how
and why all this is so for Wiener-fringes.
Of course Silvertooth HAS to assume that Michelson pattern is independent
of orientation in order to give some sense to his 'theory'.
And now you are following 'blindly' to him and everybody-else regarding
Michelson; thus you conclude incorrectly that Silvertooth's experiment
should give a 'nul result'. I am glad that you say by now that
published experimental results are 'puzzling'. That is the proper
way to say it in your case of 'theoretical denial'; and not by brutally
condemning the full 20-years experiment.
Actually such results are correctly predicted by the 'theory' and
therefore are fact: both experimentally and in theory.
It is up to you or others to re-study the actual situation in the experiment.
If you change your mind and re-discover why (as you have already
done with the Wiener-fringes issue), then Lahoz will be compelled to
disclose his complete argument even before the book is published;
exactly as he has already done with the Wiener-fringes item.
Paul B. Andersen
>
> Lahoz told me to post his general ideas about
> Wiener-fringes (and not about correct interpretation of Silvertooth
> experiment because Tom is still dennying it).
> They go as follows:
>
> """Let Tot' = (1/2)cos(wx'/c - wt') + (1/2)cos(-wx'/c - wt') ...... (a)
> be the total Wiener-fringes in the Ether-frame with mirror at rest in
> the Ether, where the coordinates x' and t' are here used; (this is totally
> equivalent to my preliminary post above in formula [1], but now with Tom's
> notation to compare ideas). Trivially (a) above can be rewritten:
> Tot' = cos(wx'/c) * cos(wt') ...... (b)
OK.
(I regard the minus sign in front of wt' in the inital equation as a typo.)
> Listen to this !:
> The wave equation |~| f = 0 (D'alambertian of f = 0) is
> invariant by the Lorentz-Einstein-Poincare transformation:
> x' = gamma ( x + vt) |
> t' = gamma ( t + vx/cc) | ...... (c)
> (where we are assuming that the Wiener-mirror is moving towards right
> with velocity v relative to Ether; and x,t are the same coordinates as
> in Tom's figure and formulae above)
> it follows that, always in the Ether-frame, a NEW type of wave can
> be 'created' which is GLUED to the moving mirror:
>
> Tot~ = cos{ w[gamma(x'+vt')]/c } * cos{ w[gamma(t'+vx'/cc) } ....... (d)
Right. This is the waves in the mirror frame:
Tot = (1/2)cos(wx/c + wt) + (1/2)cos(-wx/c + wt)
or
Tot = cos(wx/c) * cos(wt)
transformed to the "ether frame" by the Lorentz transform.
> That is merely by replacing (c) into (b) and still using x',t' as
> coordinates for the Ether-frame; and now we have the EXACT
> formula (d) of the Wiener-fringes as seen in the Ether. These waves (d)
> are as real, but very different, as the original (b) when the mirror
> was at rest in the Ether-frame: truly a new 'creation' of the Ether as
> soon as the mirror started to move with velocity v relative to Ether.
> They look, obviously, as a system (or packet) of travelling waves in
> the Ether with nodes at x'+vt' = n*pi/2 * c/(w gamma)
> (namely GLUED to the mirror) and maxima which are varying in time for
> any given point x' in the Ether.
Sure. The standing waves are moving along with the mirror.
The nodes are moving with the speed v.
Not very surprising. :-)
> How these Wiener-fringes look-like in the moving coordinates attached
> to the moving mirror?
Ah. That's a really easy one. That was where he started, wasn't it?
So:
Tot = (1/2)cos(wx/c + wt) + (1/2)cos(-wx/c + wt)
or
Tot = cos(wx/c) * cos(wt)
> Here we are going to drop 2nd. order terms in v/c to simplify an
> analysis which is unnecessary for Silvertooth effect; and we get:
> x = x' - vt' |
> t = t' | ...... (e)
> (galilean transformation which is correct to 1st. order in v/c).
HEY!!! :-)
This is ridiculous! Isnt't that blatantly obvious?
See what is done:
1. Starting with the equations for the waves in the mirror frame.
(But being careful not to write them down, though.)
2. Transform the equations to the "ether frame" using the Lorents transform.
3. Transform them back to the mirror frame **using the Galilean transform.**
which yields:
> Or:
>
> Tot ~= cos(wx/c) * cos[wt + wv(x+vt)/cc]
> Tot ~= cos(wx/c) * cos(wt + wvx/cc) ...... (f)
>
> And that is it !.
That's it indeed. :-)
A transformation from the mirror frame to the mirror frame
using the transformation:
x = x
t = t + vx/cc
> Formula (f) gives the actual Wiener-fringes,
> with correct approximation to 1st. order in v/c as needed.
> In any general orientation, that v has to be replaced by v*cos(s), of
> course. And now we see why IN FACT there is a phase-shift
> g = wvx/cc ...... (g)
> to be ALWAYS added to the temporal portion of the classical
> standing-waves in the case of Wiener-fringes.
> Such g is obviously dependent on orientation wrt stars.
> This NEVER has been published before because, according to relativity,
> formulae (e) have been replaced always by the Lorentz-Einstein
> transformation (which is incorrect in pure kinematics as here)
> yielding the usual standing-waves without the phase-shift (g) and
> without ever explaining why they are glued to the mirror in the lab
> which is moving relative to the Ether medium.
So this the "Silvertooth effect". :-)
Simply an arbitrary addition of the factor vx/cc to t.
> In summary: The Wiener-fringes are truly standing-waves with nodal
> planes truly independent of orientation (against incorrect assumption
> in Silvertooth's papers). But the temporal part wt of such unique
> standing-waves has a systemic phase-shift (g) which varies linearly
> with distance x to the mirror and is indeed dependent on orientation.
> However the time-average of their energy (and, consequently, the
> output in the Silvertooth's sensor) is totally independent of such
> curious AND REAL phase shift (g); thus such output does not depend on
> orientation.
> Which gives a clue to why never before such Wiener-fringes have been
> properly written and studied. """
This was - pathetic?
Do I dare to suggest that Prof. Dr. D.G. Lahoz should stick to
his fantasies about Cydonia, and not try to do physics?
No. I don't dare, so I did not say that.
But I did think it. :-)
Paul
Angel Garcia
>> """Let Tot' = (1/2)cos(wx'/c - wt') + (1/2)cos(-wx'/c - wt') ...... (a)
>> be the total Wiener-fringes in the Ether-frame with mirror at rest in
>> the Ether, where the coordinates x' and t' are here used; (this is totally
>> equivalent to my preliminary post above in formula [1], but now with Tom's
>> notation to compare ideas). Trivially (a) above can be rewritten:
>> Tot' = cos(wx'/c) * cos(wt') ...... (b)
>
> OK.
> (I regard the minus sign in front of wt' in the inital equation as a typo.)
>> Listen to this !:
>> The wave equation |~| f = 0 (D'alambertian of f = 0) is
>> invariant by the Lorentz-Einstein-Poincare transformation:
>> x' = gamma ( x + vt) |
>> t' = gamma ( t + vx/cc) | ...... (c)
>> (where we are assuming that the Wiener-mirror is moving towards right
>> with velocity v relative to Ether; and x,t are the same coordinates as
>> in Tom's figure and formulae above)
>> it follows that, always in the Ether-frame, a NEW type of wave can
>> be 'created' which is GLUED to the moving mirror:
>>
>> Tot~ = cos{ w[gamma(x'+vt')]/c } * cos{ w[gamma(t'+vx'/cc) } ....... (d)
>
> Right. This is the waves in the mirror frame:
> Tot = (1/2)cos(wx/c + wt) + (1/2)cos(-wx/c + wt)
> or
> Tot = cos(wx/c) * cos(wt)
> transformed to the "ether frame" by the Lorentz transform.
Very childish comment. We have not made any transform of coordinates
at all. We never spoke of your cos(wx/c)*cos(wt) but of my formula (b)
which is written in the Ether-coordinates x',t': it represents the
REAL Wiener-fringes in the Ether-frame AND when the mirror was at REST
in the Ether-frame.
Then we used a PROPERTY of the famous Lorentz-Einstein EQUATIONS
(not transform of coordinates) in order to CREATE a NEW wave packet
in the Ether and still using Ether-coordinates x',t' in the Ether-frame;
that new thing is formula (d) above which is the mathematically EXACT
expression of the ordinary Wiener-fringes observed in laboratory when
mirror is moving relative to Ether. But (d) gives such REAL waves still
in the Ether-coordinates x',t' and as seen by observer in the Ether-frame.
I hope this helps.
Don't insult me or I will drop you to your childish level.
Tom Roberts
> Why can't there be a GR effect similar to the clock tower experiment?
There cannot be one due to the earth's gravity, because all components of
the experiment are supported against gravity and are on a horizontal table.
> Nearly all points on the surface of the Earth rise and fall in the Sun's
> gravity well. Recording an effect would seem to require a path length
> difference in BS_M5_M2_M1_D to BS_M4_M3_D.
Yes, but I am virtually certain that such an effect is far too small to detect.
I have not done any such calculations, however. My point is basically that his
entire approach is flawed, and there never was any hope at all that such
an experiment could ever observe the ether.
> This would require that the effect be synchronized with the Earth's
> rotation.
Yes. He claimed his effect was synchronized to Leo being on the horizon,
which over the span of a few days is synchronized with the earth's rotation.
I have no idea if his observations extended over a long period of time (he
made no mention of such in the only papers of his which I have).
> Don't know if this is the case for Silvertooth, but other
> MM type experiments show a local time synchronized artifact
> for example Brillet & Hall.
Brillet and Hall's residual 17 Hz signal was synchronized to orientation
of their rotating laser wrt their lab. It has been ascribed by others
(H. Aspden, IIRC) as due to the acceleration of the apparatus due to the
rotation of the earth.
Tom Roberts tjro...@lucent.com
Paul B. Andersen
> >> The wave equation |~| f = 0 (D'alambertian of f = 0) is
> >> invariant by the Lorentz-Einstein-Poincare transformation:
> >> x' = gamma ( x + vt) |
> >> t' = gamma ( t + vx/cc) | ...... (c)
> >> (where we are assuming that the Wiener-mirror is moving towards right
> >> with velocity v relative to Ether; and x,t are the same coordinates as
> >> in Tom's figure and formulae above)
Note that the following is clearly stated above:
The mirror is moving to the right with the speed v relative to the ether.
x,t are the co-ordinates in the mirror frame.
> >> it follows that, always in the Ether-frame, a NEW type of wave can
> >> be 'created' which is GLUED to the moving mirror:
> >>
> >> Tot~ = cos{ w[gamma(x'+vt')]/c } * cos{ w[gamma(t'+vx'/cc) } ....... (d)
And this is the wave Lorentz transformed to the ether frame.
> > Right. This is the waves in the mirror frame:
> > Tot = (1/2)cos(wx/c + wt) + (1/2)cos(-wx/c + wt)
> > or
> > Tot = cos(wx/c) * cos(wt)
> > transformed to the "ether frame" by the Lorentz transform.
Obviously.
> Very childish comment. We have not made any transform of coordinates
> at all. We never spoke of your cos(wx/c)*cos(wt) but of my formula (b)
> which is written in the Ether-coordinates x',t': it represents the
> REAL Wiener-fringes in the Ether-frame AND when the mirror was at REST
> in the Ether-frame.
No.
Your equation b) was in the case where the mirror was stationary
in the ether. This is a different case. The mirror is moving
relative to the ether, as you clearly stated above.
> Then we used a PROPERTY of the famous Lorentz-Einstein EQUATIONS
> (not transform of coordinates) in order to CREATE a NEW wave packet
> in the Ether and still using Ether-coordinates x',t' in the Ether-frame;
> that new thing is formula (d) above which is the mathematically EXACT
> expression of the ordinary Wiener-fringes observed in laboratory when
> mirror is moving relative to Ether. But (d) gives such REAL waves still
> in the Ether-coordinates x',t' and as seen by observer in the Ether-frame.
> I hope this helps.
> Don't insult me or I will drop you to your childish level.
Sure, I am childish.
But that is rather irrelevant to the fact that what you have done, is:
1. Starting with the equations for the waves in the mirror frame.
(But being careful not to write them down, though.)
2. Transform the equations to the "ether frame" using the Lorents transform.
3. Transform them back to the mirror frame **using the Galilean transform.**
Which yields nonsense, of course.
Paul
Tom Roberts
> Of course Silvertooth HAS to assume that Michelson pattern is independent
> of orientation in order to give some sense to his 'theory'.
> And now you are following 'blindly' to him and everybody-else regarding
> Michelson;
So why don't you present your argument that the Michelson fringes are
not evenly spaced independent of orientation?
> Actually such results are correctly predicted by the 'theory'
So you claim, but keep failing to demonstrate.
Tom Roberts tjro...@lucent.com
Tom Roberts
> Tot ~= cos(wx/c) * cos[wt + wv(x+vt)/cc]
> Tot ~= cos(wx/c) * cos(wt + wvx/cc) ...... (f)
> And that is it !. Formula (f) gives the actual Wiener-fringes,
> with correct approximation to 1st. order in v/c as needed.
I will not comment at this time on the methods used to obtain this
equation....
I will remark that the first-order approximation you use is
known to be risky (if not invalid). In particular the second-
order terms you neglect can conspire to cancel an ostensibly
first-order term (note the c SQUARED in its denominator). I
have not checked to see if this disease affects your
derivation.
I will point out that this term is completely unobservable in
Silvertooth's apparatus, because his detector measures the time-average
of the square of the wave, and that is independent of any phase offset
like this. His detector will see that "cos(wx/c)" factor only, squared.
Again: at this time I am commenting _strictly_ on what I quoted above,
and not on how it was obtained (except for the warning above)....
> In any general orientation, that v has to be replaced by v*cos(s), of
> course.
It does not work that way -- lengths are contracted along the direction
of travel, but not perpendicularly. The usual vector equations don't
work in a theory with an anisotropic speed of light.
> according to relativity,
> formulae (e) have been replaced always by the Lorentz-Einstein
> transformation (which is incorrect in pure kinematics as here)
> yielding the usual standing-waves without the phase-shift (g) and
> without ever explaining why they are glued to the mirror in the lab
> which is moving relative to the Ether medium.
In SR and classical electrodynamics the fringes are "glued" to the
mirror by the boundary conditions defined by the conductor which is
the mirror. This is just elementary E&M.
Your phase shift is unobservable by any time-averaging detector.
I suspect it is unobservable by any direct wave-sampling
detector either -- you need a reference phase, and I
suspect that it will vary such that this phase shift
is cancelled. That certainly happens for the related
experiment of Torr and Kolen....
And naturally, an SR discussion does not include any ether. Why you
think that is an omission in an SR analysis is beyond me....
Why do you think the "Lorentz-Einstein transformation" is "is
incorrect in pure kinematics"? After all, the Lorentz transforms
essentially _define_ kinematics in SR.
> However the time-average of their energy (and, consequently, the
> output in the Silvertooth's sensor) is totally independent of such
> curious AND REAL phase shift (g); thus such output does not depend on
> orientation.
If you already knew this, why did you go off on what you now admit
is a completely irrelevant tangent?
Tom Roberts tjro...@lucent.com
Angel Garcia
Silvertooth's experiment has two interferometers: an ordinary Michelson
one mounted with another showing Wiener-fringes.
....
Although Dr. Silvertooth will eventually be recognized as a genial
experimentalist for his discovery of anisotropy of light, the theory
that he presented in his first report of 1987 has two basic errors:
A) One regarding the Michelson interferometer.
B) Another regarding the Wiener fringes.
The correct theory for such Wiener fringes goes as follows:
....
output in the Silvertooth's sensor) is totally independent of such
curious AND REAL phase shift (g); thus such output does not depend on
orientation.
properly written and studied. """
In passing from one node to the next of Wiener fringes via his
genial sensor, orientation is irrelevant because the only
detail in such fringes which depends on orientation is a mere
phase-shift g buried in the 'temporal' part of such standing waves.
And the output of the sensor ignores it (time average).
Given such theoretical (final) datum it follows that Silvertooth
had to committ a second error of theoretical interpretation in
order to 'cancel' the first and, therefore, be consistent with
the experimental results that he 'repeatedly' observed and measured.
Angel Garcia wrote:
> Of course Silvertooth HAS to assume that Michelson pattern is independent
> of orientation in order to give some sense to his 'theory'.
> And now you are following 'blindly' to him and everybody-else regarding
> Michelson;
Tom:
So why don't you present your argument that the Michelson fringes are
not evenly spaced independent of orientation?
Angel:
I already excused myself for such unsuported announcement and asked for some
patience.
Lahoz is writing a full short book about...
and I cannor short-cut his theory in this board. See below. It is much better
to see 'many' ideas related to it by various posters and then collect them
and discuss the merits of each one. If I put the theory then many people
will be biassed (in favor or against) and we loose original works as yours
about Wiener fringes. As it is now we have two different and original
arguments which are fully concordant as to why Silvertooth made basic error
B) above.
Angel:
> Actually such results are correctly predicted by the 'theory'
Tom:
So you claim, but keep failing to demonstrate.
Angel:
Thus error A) has to be found and remains as a challenge for
everybody in this board. Lahoz found such A) error long ago and
consequently the correct 'theory' of the Silvertooth Effect; and
he will publish it, Deo volente, in planned book TETET-99.
The effects of such 2 errors in Silvertooth's papers are important
regarding his basic interpretation about actual orientation of the
observed ether-wind. With Lahoz's correct theory the cosmic line
of anisotropy Leo-to-Aquarius is merely a projection exactly perpendicular
to ether-wind; and not the actual orientation of ether-wind as Silvertooth
repeatedly assumes.
But this is totally irrelevant regarding the experimental fact of
anisotropy.
Angel Garcia
From: Tom Roberts <tjro...@lucent.com>
Date: Mon, 16 Aug 1999 12:41:08 -0500
Angel Garcia wrote:
> Tot ~= cos(wx/c) * cos[wt + wv(x+vt)/cc]
> Tot ~= cos(wx/c) * cos(wt + wvx/cc) ...... (f)
> And that is it !. Formula (f) gives the actual Wiener-fringes,
> with correct approximation to 1st. order in v/c as needed.
Tom:
I will not comment at this time on the methods used to obtain this
equation....
I will remark that the first-order approximation you use is
known to be risky (if not invalid). In particular the second-
order terms you neglect can conspire to cancel an ostensibly
first-order term (note the c SQUARED in its denominator). I
have not checked to see if this disease affects your
derivation.
Angel:
Those methods are now (after ether-wind detection) the logical ones:
to use always the Ether-frame for intial and final expressions of
the corresponding wave-packets; and then to use the many times
published 'transference' from ether to laboratory frame:
x' = gamma(x-vt)
t' = t/gamma
These transforms are accurate and differ from SR in the temporal
coordinate only. It implies a syncronization using classical
methods: as if signals were at infinite speed; the practical way
to do it is letting every clock at any point in laboratory to
be (at t=0) synchronous with the ether-one at same instantatenous point.
Or equivalently to synchronize using light signals of velocity c+v or
c-v with corresponding length contractions. Not final... just my
impromptu vision of such NEW and difficult situation.
No. No fear about problems arising from using 'galileo' in the case
of theory for Silvertooth as I posted. About the phase-shift
wvx/cc it can be written 2pi*(x/lambda)*(v/c)
thus it is a first order in v/c using the wave-length as unit for
distances to the mirror. It comprises all details that you
found in your post !: congratulations.
Angel:
> In any general orientation, that v has to be replaced by v*cos(s), of
> course.
Tom:
It does not work that way -- lengths are contracted along the direction
of travel, but not perpendicularly. The usual vector equations don't
work in a theory with an anisotropic speed of light.
Angel:
contractions are 2nd. order: irrelevant for Silvertooth Effect; which
is why I presented the CORRECT expression for Wiener-fringes as seen
in the laboratory with approximation of ONLY 1st. order; and with
such proviso my v*cos(s) statement is fully correct.
> according to relativity,
> formulae (e) have been replaced always by the Lorentz-Einstein
> transformation (which is incorrect in pure kinematics as here)
> yielding the usual standing-waves without the phase-shift (g) and
> without ever explaining why they are glued to the mirror in the lab
> which is moving relative to the Ether medium.
In SR and classical electrodynamics the fringes are "glued" to the
mirror by the boundary conditions defined by the conductor which is
the mirror. This is just elementary E&M.
No. The WAVE-EQUATION is ONLY valid (strictly speaking) in the
coordinates fixed in the MEDIUM where light propagates.
As it is the case with sound, with water-waves, etc.
To say that such Wave-equation for light or for EM fields or
for atomic wave-packets is also valid in the lab is the 'mistake'
of SR: a false postulate too good to be true as now we have
learned from Silvertooth giving a big 'fiasco' to SR.
The only 'safe way to proceede' is to use the fact that the
wave-equation is strictly valid in the Ether AND to build
NEW solutions strictly valid in the ether via the
famous Lorentz-Einstein equations USED as mere mathematical tool
in identical coordinate system for original and final wave-packet,
namely the ether-coordinates.
So, from cos(x/lambda)*cos(wt) ... (a nice solution ONLY in the Ether)
we derive another mathematically exact solution (and new wave-packet)
also valid in the same Ether-frame (my (d) formula in previous posts).
And we do this using the Property of those famous Lorentz-Einstein
equations ( my (c) in previous posts with x',t' and x,t being
mere mathematical variables IN THE UNIQUE ETHER-frame; both sets
of variables representing coordinates in one and only one inertial
frame, the Ether.
Now... once we got a NEW exact solution which is therefore a possibly
existent and real wave-packet in the Ether-frame, we can examine
the ways to 'transfer' the formula to a set of coordinates in the
laboratory, but that cannot be done by using SR transform because
we know such SR to be a mere (and gross) approximation only.
Tom:
Your phase shift is unobservable by any time-averaging detector.
I suspect it is unobservable by any direct wave-sampling
detector either -- you need a reference phase, and I
suspect that it will vary such that this phase shift
is cancelled. That certainly happens for the related
experiment of Torr and Kolen....
Angel:
Right, of course. You said that in your analysis; the oessential
difference with Lahoz's ideas is that you 'assume' it phenomenologically
whereas Lahoz 'afirms' it giving the deep cause of it using the Ether-frame
where CERTAINLY the wave-equation is valid.
Tom:
And naturally, an SR discussion does not include any ether. Why you
think that is an omission in an SR analysis is beyond me....
Angel:
Of course is not an omission in SR; I said it that clearly: that
if one uses SR to transform the Wiener-fringes in Ether-coordinates
(my formula (d)) to corresponding expression in lab-coordinates then
one gets the idiocy pointed out by Paul; namely 'identical' expression
for standing-waves in Ether as in the lab . "All inertial frames
describe equally the equations of Physics".
The point is that, once ANISOTROPY of light has been experimentally
shown (Silvertooth), the SR is dead; and one has to start 'da capo'
and in a 'safe' way. All equations of fundamental Physics are strictly
valid in the ether-frame and, via 'relativistic invariance', wave
packets in the ether can be built which very-closely mimic situations
in the lab. How closely remains to be studied in each case; and we
have done it for the Wiener fringes:
the expression
(1) .... cos(x/lambda)* cos(wt)
is NOT an existent wave packet when x,t are coordinates of the
laboratory; whereas
(2) ... cos(x/lambda)* cos[wt + 2pi(v/c)(x/lambda)]
it IS indeed an existent wave-packet with such lab-coordinates in
a good approximation of first order in v/c; sufficient to
theorize the Silvertooth Effect.
Angel Garcia
Just correcting a typo in last lines of previous post:
Angel Garcia (bp...@FreeNet.Carleton.CA) writes:
> is NOT an existent wave packet when x,t are coordinates of the
> laboratory; whereas
> (2) ... cos(x/lambda)* cos[wt + 2pi(v/c)(x/lambda)]
> it IS indeed an existent wave-packet with such lab-coordinates in
> a good approximation of first order in v/c; sufficient to
> theorize the Silvertooth Effect.
Clearly there is a missing 2pi in formulas (1) and (2):
(2) ... cos(2pi* x/lambda)*cos[wt + 2pi(v/c)(x/lambda)]
To avoid such typos Heisenberg was wearing a tie with the symbol
|/
/|
|~~|
| | or h/(2pi)
and similarly it is constantly used for lambda/(2pi).
Tom Roberts
> The point is that, once ANISOTROPY of light has been experimentally
> shown (Silvertooth), the SR is dead;
Again: you are assuming what you should be trying to prove.
As I have said several times:
(1) Silvertooth's experiment has significant and unresolved
instrumentation effects.
(2) there is no reasonable theory which predicts a non-null result
for his measurements.
And then you claim that _Silvertooth's_ experiment "shows" an
"ANISOTROPY of light" (!).
It should be obvious that this claim is of no use in your attempt to
disprove _either_ of my claims (1) or (2) above. The veracity of
Silvertooth's experiment is precisely the issue.
What you need to do to make headway against my claim (1) is to
get someone to re-do Silvertooth's measurements, taking care not
to permit any instrumentation effects to interfere with the
measurement. That implies they must obtain a clear sin^2 pattern
on their oscilloscope plotting piezo-voltage vs detector-output
(etc.). That is a worthwhile effort, but one which is very likely
just resolving instrumentation effects; when they are removed it
is quite likely that a null result will be obtained (because of my
point (2) above).
What you need to do to make headway against my claim (2) is to
present a theory which avoids at least one of the two conditions
of my article "Subject: Silvertooth's Experiment Cannot See the
Ether". To be useful this theory must predict a non-null result,
and must also correctly describe other optical experiments, such
as Wiener's experiments and Young's experiment. I _seriously_
doubt you can do this, but would be interested to see it....
So far you have made no headway on either point.
Tom Roberts tjro...@lucent.com
Angel Garcia
>> shown (Silvertooth), the SR is dead;
>
>
experimental fact; he repeated the results many times with
different conficurations and got consistency as published in 3 papers.
> (2) there is no reasonable theory which predicts a non-null result
> for his measurements.
Yes. There is. The challenge remains until Lahoz publishes TETET-99
...
> Ether". To be useful this theory must predict a non-null result,
> and must also correctly describe other optical experiments, such
> as Wiener's experiments and Young's experiment. I _seriously_
> doubt you can do this, but would be interested to see it....
About the Wiener's fringes ?. There is no other theory except
what I posted in intro. of present thread.
Note that what you derived assuming speeds c+v and c-v which
is mere 'galilean' addition of velocities in case of an ether
becomes in PERFECT agreement with what Lahoz derived and I posted.
So we got TWO totally independent and perfectly consistent (if
not identical) theoretical proofs of what the Wiener fringes
role is in the Silvertooth experiment. Both clearly show why the
OUTPUT in his sensor is totally independent of orientation.
Consequently the Wiener fringes arm is NOT the measuring arm, but
the 'reference' arm; (whereas, sadly, Dr. Silvertooth and Dr. Whitney
published 'theory' which contradicts OUR theoretical result: the first
basic mistake in their theory).
What was the 2nd. mistake in 'their theory' so that the experimental
FACT holds true ?.
Tom Roberts
> I merely TRUST what Silvertooth has published as purely
> experimental fact; he repeated the results many times with
> different conficurations and got consistency as published in 3 papers.
Your trust seems misplaced to me, but it does not break the circularity of
your argument.
> Note that what you derived assuming speeds c+v and c-v which
> is mere 'galilean' addition of velocities
Apparently you did not read my derivation very carefully. I made an _arbitrary_
Ansatz that the 1-way speed of light is c1 in one direction and c2 in the
opposite direction, subject to the constraint that the round-trip speed of light
be isotropically c (i.e. 1/c1 + 1/c2 = 2/c). There's nothing "Galilean" about
that -- it is general enough to include every reasonable ether theory I know
of, including all the ones advocated in this newsgroup.
In fact the Galilean values c1=c+v and c2=c-v do not satisfy that
constraint except for v=0.
> What was the 2nd. mistake in 'their theory' so that the experimental
> FACT holds true ?.
You keep claiming there is a "second mistake", but never actually discuss or
describe it. Please do so.
I know of no mistake in my analysis, which shows that the fringes of Silvertooth's
"ring interferometer" are equally spaced regardless of orientation, with exactly
the same spacing as that of the fringes in his Michelson interferometer. So his
experiment is predicted to yield a null result using _any_ theory belonging to the
class I discussed.
Tom Roberts tjro...@lucent.com
Angel Garcia
>> experimental fact; he repeated the results many times with
>> different conficurations and got consistency as published in 3 papers.
>
> Your trust seems misplaced to me, but it does not break the circularity of
> your argument.
the experiment and its result is right. SECOND it follows that SR is gone.
THIRD I never use anymore an incorrect theory as SR in cases when
a more refined theory is needed. No more to it.
>
>> Note that what you derived assuming speeds c+v and c-v which
>> is mere 'galilean' addition of velocities
>
> Apparently you did not read my derivation very carefully. I made an _arbitrary_
> Ansatz that the 1-way speed of light is c1 in one direction and c2 in the
> opposite direction, subject to the constraint that the round-trip speed of light
> be isotropically c (i.e. 1/c1 + 1/c2 = 2/c). There's nothing "Galilean" about
> that -- it is general enough to include every reasonable ether theory I know
> of, including all the ones advocated in this newsgroup.
Your equation (13) says:
c1 = c/(1+v/c) = c -v - terms in c(v/c)^2 and higher
which is essentially 'galilean' and not SR. Apparently you don't know
what you have done. If you drop higher terms than 1st. for speed of
light you are in strict galilean assuming speeds ADDED to c (which is
a blame to SR), namely c-v and c+v. And yet with such assumption
for your c1 and c2 you get EXACTLY the Lahoz formula for Wiener
fringes (just complete your analysis and see !); in any case the
phase-shift and modulation that you correctly describe with any
c1 and c2 is correct and becomes identical to what Lahoz's formula
says once one uses the Silvertooth's c1 ~= c-v and c2 ~= c+v which
is what you do in your (13).
>
> In fact the Galilean values c1=c+v and c2=c-v do not satisfy that
> constraint except for v=0.
That's stupid !. See above.
(The rest is a parrot repetition which I have answered many times).
Jim Carr
}
} Though I am not an 'aetherist' or an 'SRist' I take exception to the
You might read the rationale for making the change, which is
contained in an article in Nature 303, 373 (1983). Precision
measurement essentially demanded it. Note that at each point
the old standard was included within the new, since the new
one was "traceable" to the old one to within measurement error.
} Since the day that Physics 'forsook' a physical measure for distance
} and replaced it with a 'light based reference' (first as a measure of
} the number of wavelengths of red light emitted from an isotope of
} Kripton gas then as simply a calculation based on the defined speed of
} light) all of the experimental data that you refer to above is
} tainted. Tainted in the sense that any experiment that uses an
} 'optical based' method of measuring distance will mask any anisotropy
} due to 'movement' through any 'aether medium'.
In article <37B45982...@lucent.com>
Tom Roberts <tjro...@lucent.com> writes:
>
>Most experiments, Silvertooth's included, base actual distance
>measurements on physical meter sticks and such -- no use of "light-
>based reference" for distance is actually used. Your condemnation
>here is completely irrelevant to actual experimental physics _AS_
>_IT_IS_ACTUALLY_PERFORMED_, in most cases and this one in particular.
>
> Silvertooth gives no measurements of his apparatus, and no
> dimensions are really necessary -- such measurements as you
> condemn are neither made nor needed.
>
>In your claim of "masking" of "anisotropy due to 'movement' through any
>'aether medium'", the word "any" is a gross overstatement -- it is
>"masked" only for aether theories for which the aether is in principle
>undetectable (i.e. for theories in which the round-trip speed of light
>is isotropic).
Correct. In any other, what you would observe is the apparent
change in the length of your apparatus rather than a change in
the speed of light -- which would suffice to make the point you
wish to make experimentally.
However, you *will* be limited to the precision with which stable
lengths can be maintained with material methods such as an etched
line on a metal bar unless you make some additional assumptions
about the stability of your apparatus. Then you can just used
a change in the light-measured distance as the equivalent of the
previous experiment looking for a change in light speed.
--
James A. Carr <j...@scri.fsu.edu> | Commercial e-mail is _NOT_
http://www.scri.fsu.edu/~jac/ | desired to this or any address
Supercomputer Computations Res. Inst. | that resolves to my account
Florida State, Tallahassee FL 32306 | for any reason at any time.
Tom Roberts
Author: Tom Roberts tjro...@lucent.com
Date: August 28, 1999
Since posting the original article in this thread, I have realized there
is a much simpler and shorter proof that any reasonable ether theory
predicts this experiment will have a null result. It starts with the
lemma I presented in the article "Subject: Torr and Kolen's Experiment
Cannot See the Ether".
Consider the class of theories for which the round-trip speed of light is
isotropically c in any inertial frame. All reasonable ether theories
belong to this class, where "reasonable" means not already refuted by
other experiments. Note that SR is a member of this class, as is every
ether theory seriously advocated in this newsgroup. The different
theories differ in the 1-way speed of light and the corresponding clock
synchronization.
Lemma. In any theory of this class and for any assumed anisotropy
in the 1-way speed of light, the time delay for a signal to make a
round trip around a path of fixed shape and size is independent of
the orientation of the path in an inertial frame.
The time delay for such a path is measured by a single clock, so
differences in clock synchronization cannot affect the delay for a round-
trip path. This lemma is clearly true for one theory of the class (SR),
and the only difference between the theories of this class is how clocks
are synchronized [1], so it is true for all of them.
Corollary. For any experiment which has a single source and a single
detector which measures the time delay between two signals from the
source propagating by different paths of fixed shape and size, the
detector output will be independent of the orientation of the paths
in an inertial frame. This is true for any theory of this class and
for any assumed anisotropy in the 1-way speed of light.
Proof: move the source so it is collocated with the detector, and
transfer its output to where it was originally located by a path of fixed
shape and size. Now the signals are round-trip and the lemma applies; as
the time delay from emission to detection is independent of orientation
for each signal, it must also be true for their difference. As the
additional path contributes identically to both paths, this must be true
for the original configuration.
The corollary applies to each of Silvertooth's interferometers. So for
a given position of his moveable table the output of both detectors is
independent of orientation. This directly implies that the relationship
between the fringes in the two interferometers is independent of
orientation. That is a null result.
The striking difference in length between this demonstration and that in
my original article is an instance of the power of invariance principles
like the lemma and corollary above. If you can find an applicable
invariance principle it can greatly simplify an argument or proof.
Tom Roberts tjro...@lucent.com
References
----------
[1] Zhang, _Special_Relativity_and_its_Experimental_Foundations_.
My class is what he calls "Edwards Frames".