Relative Motion and Time Dependance on Velocity of Light

[DewaldS]

The relative motion and the relative time of any inertial system to another one is in literature available (to me) subject to multiplication by a factor lambda = (1-v^2/c^2)^-1/2. It is found in the Lorentz transformation as well as STR. Is it possible to arrive at the same result and formulae without the 2nd postulate of STR? I note that some physicists do not go with the 2nd postulate of STR, (Lorentz contraction for Michelson Morley device arm) but use the Lorentz transformation (with speed of light as a max inside these formulae) in their arguments - how can this make sense?

Where can I find a derivation of the Lorentz transformation with the arguments that are put forward?

Is the progression of physics not hampered by the 'maximum possible speed of information transmission' being the speed of light? I mean - if I do not see an object traveling at c via it's light reflection -it does not mean that it is not there. Or if I go faster than the speed of light - even directly into the light - why am I going 'back into time'? (Even in relation to the light wave or 'light particle's' clock)? I do not get this - please help!

 

[JesseM]

The relative motion and the relative time of any inertial system to another one is in literature available (to me) subject to multiplication by a factor lambda = (1-v^2/c^2)^-1/2. It is found in the Lorentz transformation as well as STR. Is it possible to arrive at the same result and formulae without the 2nd postulate of STR? I note that some physicists do not go with the 2nd postulate of STR, (Lorentz contraction for Michelson Morley device arm)

That last comment is a little confusing, the 2nd postulate doesn't say anything about Lorentz contraction, rather it says that the speed of light is the same in all inertial frames. From the two postulates of SR you can derive the Lorentz transformation for relating the coordinates of one frame to another, and from that you can derive Lorentz contraction and time dilation.

but use the Lorentz transformation (with speed of light as a max inside these formulae) in their arguments - how can this make sense?

"Use the Lorentz transformation" to do what, exactly? To derive the length contraction and time dilation equations? Like I said, the usual sequence is to start from the two postulates of SR (that the laws of physics are the same in all inertial frames, and that the speed of light is c in all inertial frames), use them to derive the Lorentz transformation, and use the Lorentz transformation to derive the length contraction and time dilation equations.

Where can I find a derivation of the Lorentz transformation with the arguments that are put forward?

What do you mean "the arguments that are put forward"? Do you want a derivation of the Lorentz transformation from the two postulates?

Is the progression of physics not hampered by the 'maximum possible speed of information transmission' being the speed of light? I mean - if I do not see an object traveling at c via it's light reflection -it does not mean that it is not there.

The coordinates of events in a given inertial frame don't depend on when you see them, if that's what you mean. Normally inertial frames are defined in terms of a hypothetical lattice of rulers and synchronized clocks at rest in that frame, with the coordinates of each event depending on local readings on this lattice so there is no problem with light delays. For example, if I see an explosion through my telescope when my clock reads t=15 seconds, and I see it happened right next to the 10 light-second marking on my ruler, and the clock sitting at that marking read t=5 seconds at the moment it happened, then I assign the event a time-coordinate of t=5 seconds, not t=15 seconds when I actually saw it.

The one tricky part about using local readings on rulers and synchronized clocks is defining what it means for two clocks at different locations to be "synchronized"--Einstein suggested the convention that each observer defines the meaning of "synchronization" using the assumption that light travels at the same speed in all directions in their own frame, so that if I set off a flash at the midpoint of two clocks, I define them to be synchronized if they both read exactly the same time at the moment the light from the flash hits them. It is true that this is just a synchronization convention--one could define what it means for clocks to be "synchronized" in other ways--but what Einstein postulated was that the laws of physics would have the interesting property that they would obey exactly the same equations in the coordinate systems of different inertial observers who each synchronize their own clocks in this way, a property of the laws of physics known as "Lorentz-invariance" (because the coordinate systems of different observers are related to one another by the Lorentz transformation). So far, all investigation into the fundamental laws of physics has backed up the hypothesis that the laws of nature are always Lorentz-invariant.

Or if I go faster than the speed of light - even directly into the light - why am I going 'back into time'? (Even in relation to the light wave or 'light particle's' clock)? I do not get this - please help!

Different inertial frames define simultaneity differently, so that if you could send a message that was faster than light in one frame, there would be some other frames where the message would actually be received at an earlier time than it was sent! And since the first postulate of relativity is that the laws of physics work the same in all inertial frames, if it were possible to have messages arrive before they're sent in one frame, then this would have to be possible in all frames.

The reason different frames disagree about simultaneity has to do with the synchronization convention I mentioned earlier. Suppose I am in a ship with clocks at the front and back, so I synchronize them in the ship's rest frame by setting off a flash at the midpoint of the ship, and setting the clocks to read the same time at the moment the light from the flash reaches each one. But if in your frame you observe the ship to be moving forward, then if you assume the light moves at the same speed in both directions in your frame, naturally that means that in your frame the light must reach the back clock before the front clock, since the back clock was moving towards the point where the flash was set off while the front clock was moving away from that point, so the light will take longer to catch up to the front clock. Here is a little youtube movie which illustrates this point. And if you want to know more about the relation between FTL and time travel, see this recent thread:

https://www.physicsforums.com/showthread.php?t=252523

 

[DewaldS]

Hi JesseM,

Thanks for your reply.

My question should have been : What does the Lorentz transformation look like when the 2nd postulate of STR is not used to derive it? All my available resources only provide it in that form. Or does it revert back to the Galilean Transformation?.

I am just curious about other possible explanations for the MM experiment before I start working with all the outcomes of the 2nd postulate.

 

[atyy]

My question should have been : What does the Lorentz transformation look like when the 2nd postulate of STR is not used to derive it? All my available resources only provide it in that form. Or does it revert back to the Galilean Transformation?.

If the second postulate is not stated, and you require that the laws of physics be the same in all reference frames moving at constant velocity relative to each other, then you will find that there are two possibilities: the Galilean transform and the Lorentz transform. The two possibilities are distinguished by stating whether or not there is an upper "speed limit". (JesseM has probably posted the details on some other thread.)

Is the progression of physics not hampered by the 'maximum possible speed of information transmission' being the speed of light? I mean - if I do not see an object traveling at c via it's light reflection -it does not mean that it is not there.

It means that if you see an object, it was there some time ago, but it is not necessarily still there. It means that when we see galaxies that are far away, we are actually seeing them as they were a long time ago. So we can use that to infer what the universe was like in the past! (Those statements are roughly right, and when want to be precise, we need to be careful about what we mean by "time".)

 

[JesseM]

My question should have been : What does the Lorentz transformation look like when the 2nd postulate of STR is not used to derive it? All my available resources only provide it in that form. Or does it revert back to the Galilean Transformation?

If the only assumption you make is the first postulate that the laws of physics are the same in all inertial frames, then this would be compatible with either the Galilei transform or the Lorentz transform, so it isn't enough to derive a unique coordinate transformation.

I am just curious about other possible explanations for the MM experiment before I start working with all the outcomes of the 2nd postulate.

Well, technically all you need to explain the results of the MM experiment is the idea that there is a frame where light travels at the same speed in all directions, and the idea that any object moving at speed v in this frame will be shrunk by a factor of $$\sqrt{1 - v^2/c^2}$$ (as measured in this frame) in their direction of motion, you don't need time dilation at all.

 

[DewaldS]

I have know systematically worked my problem with this theory back to my basic concern. I am now starting to ask the questions I should have asked when I was at school. Please bear with me!

When we 'measure' the speed of light - how do we know it is actually a constant? How could we distinguish between making a relative or real measurement? If we define a coordinate system with spatial dimensions then that is we have defined. But our measuring sticks do not necessarily conform to the 'true' spatial dimensions - depending on what is happening to them. For this argument, let us also work just with a time basis that is constant (the clock used for the measurement is in the same 'physical' location on earth).

Has the speed of light been physically tested vertically (radially with respect to the Earth's centre)?

 

[atyy]

When we 'measure' the speed of light - how do we know it is actually a constant? How could we distinguish between making a relative or real measurement? If we define a coordinate system with spatial dimensions then that is we have defined. But our measuring sticks do not necessarily conform to the 'true' spatial dimensions - depending on what is happening to them.

Yes, if two people do two different experiments and get the same value for the speed of light, that of course doesn't mean the speed of light is constant. How do we even know that both of them used the same definition of length? Let's suppose they made their rulers of the same material, and compared the lengths at the same location, so they know they had the same definition of length. What if one of them did his experiment at a colder temperature, and actually his rulers were all shorter. But he miraculously got the same result because he used a pendulum clock and the length of the pendulum shrunk by exactly the right amount, and all the errors canceled out. Then, of course, I wouldn't believe that the speed of light is constant.

So to make sure that silly things like that don't happen, we should get the result using the same apparatus, and vary that velocity of that apparatus. That is essentially what Michelson and Morley did (or so I am told) - I don't know if they ensured that the temperature of their apparatus was the same when it was moving at different velocities!

Anyway, there are lots of details in doing a good experiment, so one should indeed be skeptical. I haven't been skeptical enough, so I can't tell you the details. One good place to start is:

http://www.math.ucr.edu/home/baez/physics/Relativity/SpeedOfLight/speed_of_light.html
http://www.math.ucr.edu/home/baez/physics/Relativity/SpeedOfLight/measure_c.html
http://www.math.ucr.edu/home/baez/physics/Relativity/SR/experiments.html

 

[DewaldS]

Hi Atyy,

MM was designed for detecting aether drift. See http://www.glafreniere.com/sa_Michelson.htm for a beautiful visual representation.

I was asking about the basic measurement of speed in m/s.

 

[atyy]

Hi Atyy,

MM was designed for detecting aether drift. See http://www.glafreniere.com/sa_Michelson.htm for a beautiful visual representation.

I was asking about the basic measurement of speed in m/s.

Yes, there are two questions.

1) How do you measure the speed of light
2) If you measure the speed of light with your apparatus moving at different velocities, do you get different answers? If an "aether" existed, then the answer would be yes. If the answer is no, and the "aether" is not detected, then we say the speed of light is constant.

I thought you were asking about the constancy of the speed of light?

 

[DewaldS]

Hi Atyy,

Let's be playful here.

I am putting a new theory forward:

Light travels faster in outer space than closer to Earth or any other planet. I am also suggesting that the closer it gets to a planet and the more mass the planet holds, the slower it will go.

The first consequence is then that the planets are further than what we think.

Please prove me wrong (solid thought is fine with me)!

 

[DewaldS]

I forgot to add that we are considering speed of light in a vacuum - planets in our thought game do not have atmosphere.

 

[atyy]

Hi Atyy,

Let's be playful here.

I am putting a new theory forward:

Light travels faster in outer space than closer to Earth or any other planet. I am also suggesting that the closer it gets to a planet and the more mass the planet holds, the slower it will go.

When you drop a ball to earth, the ball goes faster the closer it gets to earth. Why do you think light is different from a ball?

 

[atyy]

The first consequence is then that the planets are further than what we think.

Please prove me wrong (solid thought is fine with me)!

That's presumptuous! What if your theory is right?

 

[DewaldS]

I am really wondering about this - there must be a simple logic explanation to prove me wrong (I do not even know myself but will think about it tonight).

Or how about we turn it around and say that light moves faster as it approaches bodies and slower when further.

The gravity equations are not going to help us here, because either G could be somewhat 'out' or the 'mass' that is calculated for the other planets could be somewhat 'out'?

 

[DewaldS]

If my 'theory' is right, then a lot of our physics is obviously wrong. But I am sure that someone has checked this - anybody that can help or refer me to the correct site?

 

[atyy]

Or how about we turn it around and say that light moves faster as it approaches bodies and slower when further.

Actually, no theory can prove you wrong. Only an experiment can (but those experiments have already been done).

Anyway, why do you say light moves faster as it approaches a massive body? Are you really using the ball analogy?

 

[DewaldS]

You are right - theory cannot be wrong - only less useful.

I am sincerely worried that us scientists are taking some values for granted without making sure that they were actually tested. I have studied the Focault and Fizaut technique, as well as the improved Michelson version and I think the values are useful. My sources does not say if the speed was tested in all the directions (including vertical relative to earth).

I am trying to get someone to prove to me that the speed of light was either measured or deduced in a proper way to be constant through space (also far from a body with a big mass).

I am not putting forward a theory - I thought the 'theory' might prompt someone to refer me to proper data.

 

[JesseM]

You can read about a large number of experiments to test various predictions of relativity, including the speed of light, on this page. And the GPS navigation system depends on receivers on Earth very precisely calculating their position by using the arrival time of time-stamped signals from the satellites to calculate their distance to each satellite (and thus 'triangulate' the position of the receiver) under the assumption that the signals traveled at the speed of light (the time stamps which the satellites transmit are based on precise clocks on board the satellites which are designed to compensate for the effects of relativistic time dilation in both SR and GR so that the clocks remain synchronized in an Earth-centered frame of reference).

 

[atyy]

Let's be playful here.

I am putting a new theory forward:

I am not putting forward a theory - I thought the 'theory' might prompt someone to refer me to proper data.

Oops, sorry - I thought you were being playful (which is good) - it's definitely worth taking all the possibilities to their logical conclusion.

I am sincerely worried that us scientists are taking some values for granted without making sure that they were actually tested. I have studied the Focault and Fizaut technique, as well as the improved Michelson version and I think the values are useful. My sources does not say if the speed was tested in all the directions (including vertical relative to earth).

I don't know if the speed has been tested vertically directly.

However, we do have a theory that incorporates the constancy of the speed of light into a framework called Lorentz covariance. We also have an alternative theory in which the speed of light is not constant and uses a different framework called Galilean covariance. Lorentz and Galilean covariance have consequences for all particles, not just light, and both have been tested in all directions, and are continually being tested by high energy experiments at places like CERN and Fermi Lab. All the experiments so far have been consistent with Lorentz covariance, and have shown Galilean covariance to be wrong by many orders of magnitude.
http://www2.slac.stanford.edu/vvc/

We also have a theory called General Relativity which incorporates the constancy of the speed of light and the force of gravity. In this theory, light has weight, and can be bent by gravity. An introduction to the experiments supporting this theory is given by Rainer Weiss (one of our finest skeptics of General Relativity):
http://www.aapt-doorway.org/TGRUTalks/Weiss/WeissTalk1of9.htm

Another place to look at the use of relativity is the Global Positioning System (GPS), which uses the constancy of the speed of light as a principle (and which would certainly count as a vertical test). However, I don't know what the error bounds on the constancy of the speed of light are from GPS.
http://www.emis.de/journals/LRG/Articles/lrr-2003-1/index.html

However, learning General Relativity is difficult, and I'd suggest you just take a quick look at the articles on General Relativity, and concentrate on learning Special Relativity.

 

[russ_watters]

Hi Atyy,

Let's be playful here.

I am putting a new theory forward:

Light travels faster in outer space than closer to Earth or any other planet. I am also suggesting that the closer it gets to a planet and the more mass the planet holds, the slower it will go.

The first consequence is then that the planets are further than what we think.

Please prove me wrong (solid thought is fine with me)!

It's pretty simple: we have spacecraft out there in the outer solar system and we know how much fuel they used to get there and calculated their trajectories with exquisite precision. We'd notice anything more than a tiny (a few parts per billion perhaps?) deviation in the speed of light.

 

[DewaldS]

I have now done the calculations for the experimental setup as was done in the MM experiment. As I understand it, a phase shift would cause the interference pattern i.e the two light beams arriving at different times back at the eyepiece . As I have it, yellow light was used with a wavelength of 570nm.

Questions:

i) Am I right in saying that the interference pattern would also be introduced by moving any of the reflecting mirrors a small distance? It will actually disappear and reappear as a mirror is moved.
ii) Did the experimenters actually 'fine tune' the distances to get an initial image free of interference patterns and
iii) WERE ANY INTERFERENCE PATTERNS EVER OBSERVED? (could this be proved from literature?)
iv) My calculations indicate that two arms of exact equal length would already be 'out of phase' - I used 29km/s for speed of machine and 300 000 km/s for speed of light to get a feel.
v) What type of mechanical fine tuning mechanism would be used that could shift a mirror in fractions of a micrometer?
vi) How would one avoid 'standing waves'?

 

[JesseM]

iv) My calculations indicate that two arms of exact equal length would already be 'out of phase' - I used 29km/s for speed of machine and 300 000 km/s for speed of light to get a feel.

Others may know more about the historical details of the MM experiment, but as for this question, did you take into account that in the frame where the apparatus is moving at 29 km/s, the arm that's parallel to the direction of motion will be shrunk due to Lorentz contraction? See my example in post #51 of this thread to see how this ensures that the light will take the same time to return to the point of origin along both arms. Of course, Michelson-Morley themselves did not anticipate the Lorentz contraction effect, which is why they expected to see interference when the device was moving relative to the aether!

 

[DewaldS]

iv) My calculations indicate that two arms of exact equal length would already be 'out of phase' - I used 29km/s for speed of machine and 300 000 km/s for speed of light to get a feel.

This is actually not really relevant - unless we are trying to catch the 'exact same wavefront'?

 

[DewaldS]

Jesse

Yes, I understand the Lorentz contraction. I think it is a much more useful theory and the possible reasons for this contractions should be investigated by serious scientists.

It is just difficult for me to understand that the results from the MM machine could have been used to arrive at a theory like GTR and STR. The machine had to be stable and mechanically exact within 6 e-7 m. Anybody spoke to an engineer about this? Are physicists willing to come to far reaching conclusions like the one's implied by GTR and STR on the grounds of the results of a machine like this?

Question:

If we view an atom (the standard atom model). The elctron is moving around the nucleus in some orbitary fashion and determines the size of the atom. Let' say now the atom is moving with constant velocity. The electron is now racing in a circular fashion around the nucleus (which is moving itself). So we end up with some contraction. If we accept the model of the atom as useful, then already we have an explanation for contraction.

 

[DewaldS]

Yes, I understand the Lorentz contraction. I think it is a much more useful theory and the possible reasons for this contractions should be investigated by serious scientists.

I am referring to Lorentz contraction without the STR part in it, Remember - without the result from MM that implies that light has a constant relative velocity to the observer regardless of the motion of the observers's inertial system we have NO STR

 

[russ_watters]

I am referring to Lorentz contraction without the STR part in it, Remember - without the result from MM that implies that light has a constant relative velocity to the observer regardless of the motion of the observers's inertial system we have NO STR

As was already pointed out, there is a mountain of experimental verification of STR. You're really barking up the wrong tree here. Note also: it isn't clear if Einstein knew about the MMX and regardless, it isn't needed for the formulation of the theory (but does provide good evidence to support it).

And also, please note that our guidelines expressedly forbid free-form idle speculation and unverified personal theories. This is a place to learn physics, not a place to indulge your own personal speculations.

 

[JesseM]

This is actually not really relevant - unless we are trying to catch the 'exact same wavefront'?

What do you mean by "catch"? The idea is that if the peaks of the wave were lined up when the beams were departing (because they were created from a single beam using a beam splitter), they still need to be lined up when they merge again in order for there to be no interference observed. If you assume light moves at c in the rest frame of the apparatus, this would mean that if there were any difference in the length of the arms, it would have to be some integer multiple of the wavelength of the light in order to avoid interference. Since MM designed the arms to have the same length, obviously they were thinking that if the device were at rest relative to the ether, you would have exactly the same peaks lined up at the end as were lined up at the beginning. And in relativity, it's true that the same peaks that were lined up at the beginning will reunite at the end if the arms are equal length in the device's own frame, regardless of what inertial frame the device happens to be at rest in (it's no longer required that it be at rest in some preferred ether frame for this to be true). See the animations on this page, where the peaks going along one arm are shown in green and the peaks going along the other are shown in red (the page is from a somewhat crackpot site that advocates a Lorentz ether theory where objects objectively contract when moving relative to the ether, but the animations are helpful anyway).

edit: it seems I was incorrect that MM were trying to make sure the arms had exactly the same length, I was misled by schematic diagrams which show the device this way; see the link in my next post, which makes clear they just adjusted the arms so that no interference was seen initially, then rotated the device to see if this would change the pattern.

 

[russ_watters]

It is just difficult for me to understand that the results from the MM machine could have been used to arrive at a theory like GTR and STR. The machine had to be stable and mechanically exact within 6 e-7 m. Anybody spoke to an engineer about this? Are physicists willing to come to far reaching conclusions like the one's implied by GTR and STR on the grounds of the results of a machine like this?

While I've never personally done it (I'm an engineer, not a physicist), the MMX has been duplicated thousands of times. It isn't like they did it once and everyone just assumes what they said was right. It has been repeated and refined. Precision was, indeed, an issue, but a quick look at the Wiki for it shows that they did take that into account: the apparatus did have the required precision to find what they were looking for, had it existed. Speculation of flaws in the experimental setup is a non-starter.

 

[JesseM]

It is just difficult for me to understand that the results from the MM machine could have been used to arrive at a theory like GTR and STR.

Uh, in what way do you imagine the MM was used to arrive at GTR? You understand there has been a whole lot of subsequent evidence for the STR, and that completely different types of evidence are needed to verify the predictions of the GTR, right?

As for the original MM experiment, subsequent engineers and physicists have not questioned that it was sufficiently precise to get meaningful results, although of course much more precise versions of the experiment have been performed since. Here is a page which gives details of the setup if you're interested. It appears that they did not actually try to ensure that the arms were of equal length, just that the lengths were such that no interference pattern was observed at one orientation, and then they changed the orientation of the apparatus to see if this would change the interference pattern as would be expected if the device were in motion relative to the ether.

 

[DewaldS]

What do you mean by "catch"? The idea is that if the peaks of the wave were lined up when the beams were departing (because they were created from a single beam using a beam splitter), they still need to be lined up when they merge again in order for there to be no interference observed. If you assume light moves at c in the rest frame of the apparatus, this would mean that if there were any difference in the length of the arms, it would have to be some integer multiple of the wavelength of the light in order to avoid interference. Since MM designed the arms to have the same length, obviously they were thinking that if the device were at rest relative to the ether, you would have exactly the same peaks lined up at the end as were lined up at the beginning. And in relativity, it's true that the same peaks that were lined up at the beginning will reunite at the end if the arms are equal length in the device's own frame, regardless of what inertial frame the device happens to be at rest in (it's no longer required that it be at rest in some preferred ether frame for this to be true). See the animations on this page, where the peaks going along one arm are shown in green and the peaks going along the other are shown in red (the page is from a somewhat crackpot site that advocates a Lorentz ether theory where objects objectively contract when moving relative to the ether, but the animations are helpful anyway).

My calculations show that same length arms does not necessarily mean that the same wave front arrives at the same time. I set up formulaes to calculate the time it takes to meet the 'moving target' in the direction of the motion , then come back and also formulaes to calculate the time it takes to meet the 'moving target' on the right hand arm and then come back. Am I making an error?

I will not overstress the problems associated with making and measuring anything mechanical that is so exact - but it could be enlightening for anyone to visit a precision engineering workshop and ask them to show you how a micron is machined and or measured.

 

[DewaldS]

As was already pointed out, there is a mountain of experimental verification of STR. You're really barking up the wrong tree here. Note also: it isn't clear if Einstein knew about the MMX and regardless, it isn't needed for the formulation of the theory (but does provide good evidence to support it).

And also, please note that our guidelines expressedly forbid free-form idle speculation and unverified personal theories. This is a place to learn physics, not a place to indulge your own personal speculations.

The basis of the theory has to be verified otherwise it is useless. It is like division with zero - all that follows after that is worth nothing. I will look into the other experimental verifications.

I apologize for my own 'unverified' speculations.

 

[JesseM]

My calculations show that same length arms does not necessarily mean that the same wave front arrives at the same time.

It does if they are the same length in the device's rest frame, and likewise it does if you assume that in a frame where they are moving, the arm which is parallel to the direction of motion shrinks by a factor of $$\sqrt{1 - v^2/c^2}$$ (and in both cases you assume light moves at c in whatever frame you're calculating the time for the wave fronts to travel to the end of the arm and back).

I set up formulaes to calculate the time it takes to meet the 'moving target' in the direction of the motion , then come back and also formulaes to calculate the time it takes to meet the 'moving target' on the right hand arm and then come back. Am I making an error?

If your formulas disagree with what I say above, then you are making an error. Try applying these formulas to the specific numerical example I provided in post #51 of this thread.

I will not overstress the problems associated with making and measuring anything mechanical that is so exact - but it could be enlightening for anyone to visit a precision engineering workshop and ask them to show you how a micron is machined and or measured.

All I can tell you is that you will find no mainstream physicists or engineers who doubt that the equipment that MM were using had sufficient precision for their results to be meaningful. I have no idea if the techniques used then to get micron-level precision would be the same as the techniques used today.

 

[DewaldS]

Question: The 'ether' for which the MM device was actually designed. I would suppose the ether was up to some point in history used by some scientist to explain how light could 'move' through space? Why would the relative movement of the light through the ether be met by any restriction? I would suppose this is an energy consideration, but I would like to know, please.

 

[atyy]

It is just difficult for me to understand that the results from the MM machine could have been used to arrive at a theory like GTR and STR. The machine had to be stable and mechanically exact within 6 e-7 m.

Yes, the machine had to be stable to a high precision (ie. the measurements at different times had to be with the same machine, with all components at the same pressure,temperature etc). However, the key to the Michelson-Morley experiment is that it is measuring the difference in the speed of light due to the movement of the Earth in space - which is about 30 km/s. Michelson and Morley's conclusion was "the measured velocity was approximately one-sixth of the expected velocity of the Earth’s motion in orbit and “certainly less than one-fourth.” Michelson and Morley, as model experimentalists, were among the greatest skeptics of their own result, and sought repeatedly to prove themselves wrong.

The textbook caveat to the Michelson-Morley experiment is that it places a limit on the two-way speed of light, and not the one-way speed of light. Apparently, experiments since then have shown that even the one-way speed of light is independent of the Earth's movement, but I only know that as a textbook statement and don't know enough to refer you to an exact experiment.

As an example of an experiment since the MM experiment that gives us high confidence that the speed of light is constant is the experimental measurement of electron magnetic moment to greater than 1 part in 1000,000,000, AND the fact that this measurement matches the theoretical prediction incorporating the constancy of the speed of light as an assumption.

The experiment:
http://hussle.harvard.edu/~gabrielse/gabrielse/papers/2006/NewElectronMagneticMoment.pdf

The theoretical prediction using the constancy of the speed of light:
http://hussle.harvard.edu/~gabrielse/gabrielse/papers/2006/NewFineStructureConstant.pdf

You can see that the best physicists are actually the greatest skeptics. Before Gabrielse made these latest measurements, it was already known that experimental and theoretical the electron moments matched. He obviously wanted to see if there would be a mismatch with greater experimental accuracy. His more precise measurement forced a new theoretical calculation (using the same theory, but extracting consequences from it with higher precision). A discovery that the any of our present theories do not match experiment and need to be improved would actually be very exciting!

Edit: The erratum at the end of the second paper is very interesting to read, if you're interested in details.

 

[DewaldS]

It does if they are the same length in the device's rest frame, and likewise it does if you assume that in a frame where they are moving, the arm which is parallel to the direction of motion shrinks by a factor of $$\sqrt{1 - v^2/c^2}$$ (and in both cases you assume light moves at c in whatever frame you're calculating the time for the wave fronts to travel to the end of the arm and back).

If your formulas disagree with what I say above, then you are making an error. Try applying these formulas to the specific numerical example I provided in post #51 of this thread.

All I can tell you is that you will find no mainstream physicists or engineers who doubt that the equipment that MM were using had sufficient precision for their results to be meaningful. I have no idea if the techniques used then to get micron-level precision would be the same as the techniques used today.

Yes, the formulae do agree with what you are saying. I am a slow mover - I will only use any STR derived formulaes once I am convinced that the theory is sound.