Hi,
Imagine a square with 300,000 km long sides. Diagonals are around
423,000 km long. A light is flashed at one corner. One light beam
travels the diagonal and directly reaches the observer at the opposite
corner, while the other light beam is reflected off the mirror at the
other corner and again reaches the observer.

http://i95.photobucket.com/albums/l1.../untitled3.gif

In the first figure, the delay between observed direct light and
reflected light is 0.59 seconds ( (1+1) - 1.41 ). In figure two,
suppose that one way speed of light is only 150,000 km/second in the
opposite horizontal direction (but same in the vertical direction). I
calculated that the delay would be 1.2 seconds ( (2+1) - 1.80 ). So,
can this setup detect one way light speed anisotropy, or where is my
error?

(I assumed that in the first figure the x and y speed components of
light are both 210,000 km/sec. In the second figure they would be
105,000 and 210,000 so the diagonal speed is around 234,000 km/sec.
423,000 divided by diagonal speed yields the 1.80)

That will not detect anisotropy.

There are so many errors in your description that I do not know where
to start.

wrote:
That will not detect anisotropy.

There are so many errors in your description that I do not know where
to start.

Thanks for the reply, but please do tell me the first major error. I
suppose you have relativistic effects in mind, however note that there
is only one frame of reference and there is no clock synchronization
involved, so I don't think they apply. I think the only way to find out
the result of this experiment is to perform it.

Thanks for the reply, but please do tell me the first major error.

"suppose that one way speed of light is only 150,000 km/second..."

Suppose 0=1. You can get anything from that.

And when one invokes owls vs twls, the synch and measurement procedures
must be specified.

And if you are thinking of ether theories (you are talking about owls),
lengths of objects are contracted by gamma.

aso...aso..aso...

wrote:
Thanks for the reply, but please do tell me the first major error.

"suppose that one way speed of light is only 150,000 km/second..."

Suppose 0=1. You can get anything from that.

Well, imagine a person's running speed forward and backwards, they can
be different. This is just an experiment setup like MM experiment, to
detect the difference.


And when one invokes owls vs twls, the synch and measurement procedures
must be specified.

No sync is needed here, because there's only one clock, we're only
measuring the time delay that light takes by travelling two different
paths. Same in the second figure, only the orientation is changed.


And if you are thinking of ether theories (you are talking about owls),
lengths of objects are contracted by gamma.


As I mentioned, there's only one reference frame involved here. The
contraction doesn't apply locally.

aso...aso..aso...

Thanks anyway.

I think you've been too hard on him.

On Mon, 16 Oct 2006 12:07:06 -0700, wrote:

Thanks for the reply, but please do tell me the first major error.

"suppose that one way speed of light is only 150,000 km/second..."

Suppose 0=1. You can get anything from that.

Sure, sure, we _know_ SOL is 150,000 km/sec. But his point is to
assume SOL is different in two directions, and then determine whether the
proposed experiment would actually detect the difference. A "real"
difference would be many orders of magnitude smaller than the proposed
factor of two here, of course; the use of a gross difference is just to
point out that, in that gross case, the proposed experiment would show a
positive result.


And when one invokes owls vs twls, the synch and measurement procedures
must be specified.

Not in this case: He's proposed measuring a time difference at a single
observer, using a single clock. Then he repeats the experiment with
everything flipped, and again measures the time difference between two
events using a single clock. As long as the clock _rates_ can be said
with some assurance to be the same, there's no need for syncing between
distance clocks here.


And if you are thinking of ether theories (you are talking about owls),
lengths of objects are contracted by gamma.

Now you get to the heart of the problem: He's proposed that SOL is
different L-R versus R-L by hasn't said why, nor explained how SOL might
vary as the angle changes, and before setting up any real experiment we
would want to have some idea of what we thought we were testing.

Determining what you might expect for SOL at angles other than 0 and
pi/2 will determine how large the measured difference between the two
proposed cases will be, and if there will even be a difference in the two
cases.

In particular, once one has made clearer assumptions about exactly how SOL
will vary with angle, one needs to determine if the measured effect should
be first order, second order, or higher order in the difference in OWLS
and in the size of the apparatus.

I haven't done any math on this one (and don't intend to) but my guess
would be that the measured "difference in the differences" in the two
cases would be, at best, second order in both the size of the apparatus
and in the difference in OWLS.


aso...aso..aso...

--
Nospam becomes physicsinsights to fix the email
I can be also contacted through http://www.physicsinsights.org

Well, imagine a person's running speed forward and backwards, they can
be different. This is just an experiment setup like MM experiment, to
detect the difference.

Correct. Just like the mmx. Just like the mmx, just as predicted by sr
and ether theories, there will b no difference.

No sync is needed here, because there's only one clock, we're only
measuring the time delay that light takes by travelling two different
paths. Same in the second figure, only the orientation is changed.

Ok, no synch is necesary, just as in the mmx. You can use one clock or
use none with interference patterns.

And if you are thinking of ether theories (you are talking about owls),
lengths of objects are contracted by gamma.


As I mentioned, there's only one reference frame involved here. The
contraction doesn't apply locally.

In ether theories it does apply to all frames. Wrt the ether frame, the
length of one of the sides of ur setup is "really" contracted and thus
you must use the apropriate equations.

Try to be more explicative of your setup...

I haven't done any math on this one (and don't intend to) but my guess
would be that the measured "difference in the differences" in the two
cases would be, at best, second order in both the size of the apparatus
and in the difference in OWLS.

There would be NO measured difference, exactly.

I would try to explain him further but I will need a more explicative
setup. How (and especially why) he did his calculations the way he did.

In the first figure, the delay between observed direct light and
reflected light is 0.59 seconds ( (1+1) - 1.41 ).

Correct. According to SR, the time interval indicated on the clock at
the observer's location would be 1+1 - 1.41.... Here why do you
(implicitly) assume that the speed of light on the straight segments
are c where as in fig 2, you do not assume that? In fig 1, why do you
assume that the speed of light is c and not c+v or c-v like in fig
2?...

And anyhow, SR predicts the same time interval for the two figures.
Both will give 2-1.41 as predicted by sr, as predicted by ethers.

wrote:
In the first figure, the delay between observed direct light and
reflected light is 0.59 seconds ( (1+1) - 1.41 ).

Correct. According to SR, the time interval indicated on the clock at
the observer's location would be 1+1 - 1.41.... Here why do you
(implicitly) assume that the speed of light on the straight segments
are c where as in fig 2, you do not assume that? In fig 1, why do you
assume that the speed of light is c and not c+v or c-v like in fig
2?...

yes, you are right, it should have been 450000 km/sec and 150000 km/sec
in the two directions. This is because I assumed the setup is in a
spaceship travelling at 150000 km/sec, with respect to another
reference frame. I tried to avoid this other frame.


And anyhow, SR predicts the same time interval for the two figures.
Both will give 2-1.41 as predicted by sr, as predicted by ethers.

I will try to calculate how this is predicted by sr. Thanks.