Re: An Explanation of Dayton Miller’s Anomalous “Ether Drift” Result


Tom Roberts писал(а):

Sergey Karavashkin wrote:
Tom Roberts писал(а):
....
An upper limit on “absolute
motion” of 6 km/sec is derived from his raw data, fully consistent with
similar experimental results and the prediction of Special Relativity.

Dear Tom, I’m so pleased to see that you already confirm Miller’s
result not zero but 6 km/s

Obviously you did not read the paper. The 6 km is an UPPER BOUND AT THE
90% CONFIDENCE LEVEL. Indeed, the actual value obtained by the analysis
is PRECISELY zero, but there is an errorbar on that result leading to
the upper bound.


But you are wrong, of course, that SRT predicts it.

Not true; in the limit in which the rotations of earth and
interferometer can be neglected, SR predicts a truly null result. Taking
them into account one can show their effects are enormously less than
the errorbar of 0.015 fringe. Indeed, even vastly more sensitive
measurements like Brillet and Hall's cannot see the predicted effects of
the rotations.


Unfortunately, in
the space-time contraction, according to the Fitzgerald hypothesis
which Lorentz and Einstein used in their interpretations, the result
must be strongly zero.

Hmmm. You seem to be saying two contradictory things. In any case, it is
SR on which modern physics is based, not the Lorentz-Fitzgerald hypothesis.


Predicting some, even little, speed with respect
to aether, relativists have to admit the motion with respect to some
absolute reference frame which is strongly rejected by Relativity.

You need to learn what relativity actually ways, instead of your clearly
wrong guesses.


Though I much disagree with your manner to analyse, but if even with
all your manipulations with data you had to admit the same value of
speed as Miller claimed, this in the best way evidences Miller’s
results stable.

NOT TRUE AT ALL. Miller reported non-null results, and my re-analysis
CLEARLY shows there is no signal at all, and the values of Figure 11 for
the reasonably stable runs are IDENTICALLY ZERO.

You need to actually READ THE PAPER.


Tom Roberts

I have looked through your paper yesterday. It is written vaguely, no
one to be able to understand, what you have subtracted from what:

<< Therefore if the data from the first turn is subtracted
marker-by-marker from the data of every turn, the result is completely
independent of any orientation dependence, and contains only
systematic(time). >>

The matter is, the data of first turn contained errors and masking
effects the same as others. Subtracting, you on one hand as if selected
the systematic error, but on the other you introduced the error of the
first turn into the data for all next, which could sufficiently change
your result.

Similarly, suggesting

<< It is convenient to take advantage of the 180 degree symmetry of the
apparatus, and combine the data for markers 180 degrees apart. This
gives 8 orientations, and 8 independent measurements of differences in
systematic(time), shown in Fig. 10 >>,

you proceed from the idea that Miller had to obtain ideal
characteristics. Factually such symmetry is basically impossible, the
more that the degrees of turn have been measured not with the accuracy
you would need. This is also a systematic error which you involved in
your calculation, adding it from row to row.

I could continue but will not.

Miller was in this sense more correct than you are. He handled the data
of each line separately in supposition that the full-turn effect is
linear (which is seen in your Fig. 10). So he avoided to transfer the
systematic errors from line to line. After such handling he already
might average the results without systematic errors which you
introduced to your computation.

Though I can say, you have extracted the systematic related to the
full-turn effect with an outward elegance. But Miller also selected the
systematic – though outwardly not so elegant but more correctly than
you did.

I would also mark, you could blame Miller that he flawed his
experiment, should he, like Michelson, simply handle raw data as they
are. But Miller accounted the full-turn effect – maybe in his own
way, and you account the full-turn effect – in your own (and both
ways give comparable results). So from this point it hardly is correct
when you say, Miller flawed. As I showed above, your technique also can
be criticised: you well know, all techniques of data handling are
imperfect, especially if the signal is at the level of confidential
interval. And spectral methods can introduce here no less error than a
simple arithmetic averaging. Because the unavoidable errors of
experiment, as well as temperature variations of which you and
Shankland say, naturally introduce their spectral distortions that do
not relate to the full-turn effect. Relying namely on the spectral
technique, one can introduce more error than in usual arithmetic
average where such effects are known to have the property of effective
mutual compensation.

As to your

<< The 6 km is an UPPER BOUND AT THE 90% CONFIDENCE LEVEL >>

I would say, first, you have undervalued the data. Miller’s speed –
not your 6 but 9,3 km/s – has been derived from same raw data which
you used by way of compensation of the full-turn effect. These are all
100 % of the confidence interval and even more. Second, even your
confidence level of 90 % is even too good from the view of relativistic
experiments. :) LIGO expected to detect the gravity waves at the 10-15
% of confidence, and no claims from relativists. The same in other
relativistic experiments. So these 90 % are too much from the point of
relativists, and it is not worthy to doubt, Miller has obtained
positive result. But I said you in that old discussion and am repeating
now: we may not consider Miller’s results quantitatively – only
qualitatively, as the fact of aethereal wind existence. To study
aethereal wind, the more to obtain trustworthy data of its value and
direction, we need other experiments that have nothing in common with
the interferometric scheme and would distinctively register the
phenomenon. These have to be the first-order experiments without
masking effects inherent in the interferometric scheme.

Concerning, whether SRT predicts zero or non-zero effect, could you
relativists first agree among yourselves, and re-read Einstein where he
referred to the negative result of MMX. :)

And as to zero in average. If we sum any sinusoid over a period, you
know it will give zero. Thus, is it worthy to average in Michelson’s
style ‘over the hospital’. When the chief doctor is asked, how his
patients are, he answers: a half lies in fever and another is already
dead and cold, so the average temperature over the hospital is 36,6
Celsius degrees and they in average are ok. :) Your handled curves show
a clear periodic regularity. It is not worthy to turn things inside
out.

Sergey

.