Is There a Flaw in the Michelson-Morley Experiment?

[mtworkowski@o]

I think I read somewhere that there's a flaw in the Michaelson Morley experiment. Is this true?

 

[russ_watters]

The Michelson Morley experiment is over a hundred years old. It has quite a few flaws! And it has, of course, been refined considerably since then. But at the time that it was done, it was done correctly and as well as could be expected and provided groundbreaking evidence of the constancy of the speed of light and lack of existence of an aether.

 

[Nickelodeon]

The Michelson Morley experiment is over a hundred years old. It has quite a few flaws! And it has, of course, been refined considerably since then. But at the time that it was done, it was done correctly and as well as could be expected and provided groundbreaking evidence of the constancy of the speed of light and lack of existence of an aether.

Did they test for an aether traveling vertically or just horizontally?

 

[cristo]

Did they test for an aether traveling vertically or just horizontally?

What, you mean into the earth?

 

[maverick_starstrider]

What's 'vertically' vs. 'horizontally' on a rotating planet rotating a star rotating in a galaxy, etc.

 

[mtworkowski@o]

when I think of IR frames, the one thing I think of is this. Is the speed of light dependent on the speed of the source. Here's the thought. Tell me if this is amusing. "Which is the worst of the collisions? A: 2 cars head on at 60 mph each. (both same mass). B: 1 car, 60 mph hits a bridge abutment. Oh and can I just add this? Somebody told me that because the MME used light going in two directions that it averages out the delta v. What do you all think?

 

[russ_watters]

when I think of IR frames, the one thing I think of is this. Is the speed of light dependent on the speed of the source. Here's the thought. Tell me if this is amusing. "Which is the worst of the collisions? A: 2 cars head on at 60 mph each. (both same mass). B: 1 car, 60 mph hits a bridge abutment.

I'm not sure that example works because cars are designed to be crushed while bridge abutments are not. How about this: a head-on collision between two cars going at 60mph is the same energywise as a collision between a car moving at 120mph and a stationary car. In any case, yes, kinetic energy is relative to the reference frame. I'm not sure what that has to do with the speed of light, though.

Oh and can I just add this? Somebody told me that because the MME used light going in two directions that it averages out the delta v. What do you all think?

That's a very common misconception about the MM experiment, and I can never understand why: If you read any halfway decent description, of it, the usual analogy is to compare the MM experiment to measuring the speed of a boat moving sideways across a river -- not up and down the river.

So if the point of your initial question was to ask 'was the MM experiment capable of measuring what it claimed to measure?' - The answer is yes.

 

[HallsofIvy]

when I think of IR frames, the one thing I think of is this. Is the speed of light dependent on the speed of the source.

Yes, the speed of light is independent of both the speed of the source and the speed of the "observer". That's the whole point of the Michaelson-Morley experiment (and its many refinements).

Here's the thought. Tell me if this is amusing. "Which is the worst of the collisions? A: 2 cars head on at 60 mph each. (both same mass). B: 1 car, 60 mph hits a bridge abutment.

If two cars, each going 60 mph, hit head on, that is equivalent to a car hitting a parked car at (approximately) 120 mph. While it's not quite equivalent to hitting a bridge abutment (which isn't going to "give") at 120 mph, I think it would be much worse than hitting an abutment at only 60 mph.

Oh and can I just add this? Somebody told me that because the MME used light going in two directions that it averages out the delta v. What do you all think?

Yes, the only way to measure the "speed of light" in two different directions (at right angles to one another) at the same time and same place was to use light that goes "out and back". In his original paper on Special Relativity, Einstein clearly stated that he was only assuming that average speed of light "here to there and back to here" was the same in all reference frames.

 

[russ_watters]

Clarification on the last point - I read an extra piece into that question that wasn't actually there. Usually the question is meant to ask: 'because the speed of light is averaged, doesn't that mean it can't really measure the speed of the ether?' The answer is no.

 

[mtworkowski@o]

Yes, the speed of light is independent of both the speed of the source and the speed of the "observer". That's the whole point of the Michaelson-Morley experiment (and its many refinements).

If two cars, each going 60 mph, hit head on, that is equivalent to a car hitting a parked car at (approximately) 120 mph. While it's not quite equivalent to hitting a bridge abutment (which isn't going to "give") at 120 mph, I think it would be much worse than hitting an abutment at only 60 mph.


The negative acceleration in the case of the two cars is exactly the same as that of the one car. A high speed photo will show that the center of impact does not move in either case. 60 mph to 0 takes the same time in both cases.
Now for the model with the light. The way we're measuring time is arrival time. We don't have to vary the speed and direction of the source if we can do the same for ourselves. We're sailing along in space and hit the wave front of an arriving light. At the same instant and location our stationary twin is just receiving the same wave front. We pass out twin at exactly the instant that the wave front reaches our measuring equipment. Two reference frames; one for us and one for our twin. The light reaches us at the same time.
Where have I gone wrong? That is light being measured at the same speed in two different IR frames.

 

[Nickelodeon]

What, you mean into the earth?

Yes - if you wanted to test for an aether then it would be worth checking that direction too.

 

[mtworkowski@o]

No Russ I don't know if that's true. If you say it is I'll believe you but it doesn't help the analogy.

 

[mtworkowski@o]

[That's a very common misconception about the MM experiment, and I can never understand why: If you read any halfway decent description, of it, the usual analogy is to compare the MM experiment to measuring the speed of a boat moving sideways across a river -- not up and down the river.


I'll draw the analogy the way I see it and you can clear it up. wind is west to east at 50 mph. Plane is traveling at 150 air speed west to destination and then back. Ground speed out is 200 and back is 100. Average is what? That's the problem these people are seeing.

 

[mtworkowski@o]

Sorry, I was distracted. Fix the directions. you know what I mean.

 

[Nickelodeon]

I'll draw the analogy the way I see it and you can clear it up. wind is west to east at 50 mph. Plane is traveling at 150 air speed west to destination and then back. Ground speed out is 200 and back is 100. Average is what? That's the problem these people are seeing.

I don't think they were expecting the average speed of light in the respective directions to be different. I imagine they were hoping to see a wavelength variation. With your plane analogy above, the interference pattern would be apparent. I think that is the general idea.

 

[Borek]

How about this: a head-on collision between two cars going at 60mph is the same energywise as a collision between a car moving at 120mph and a stationary car.

And not stationary plus one moving at about 85 mph? Or have I misunderstood "energywise"?

 

[paw]

Yes - if you wanted to test for an aether then it would be worth checking that direction too.

Why? Just wait 6hrs and the Earth plus apparatus will rotate 90 degrees.

Furthermore the Earth couldn't be moving through the ether in two orthoganal directions at the same time so the two arms of the interferometer are enough to detect an ether if it was there. A third arm is completely unnecessary.

 

[mtworkowski@o]

Borek, what did you say?

 

[mtworkowski@o]

I don't think they were expecting the average speed of light in the respective directions to be different. I imagine they were hoping to see a wavelength variation. With your plane analogy above, the interference pattern would be apparent. I think that is the general idea.

Doesn't an interference pattern indicate a slight change in speed? I'm not getting this I apologize. I think if the leg that's parallel to the aether is containing the light going out and back we have an average. Is there a fringe on this? Oh the leg parallel is reading only incoming. What a dope I am...still?

 

[mtworkowski@o]

You know, this still sounds fishy.

 

[paw]

You know, this still sounds fishy.

Think of it this way. If one arm of the interferometer were parallel to the ether the effective path length would be 2L. But then the other arm would be orthoganal to the ether and the effective path length would be >2L. How much greater would depend on the Earths velocity through the ether but it would be greater.

It's this difference in path length that would cause the interferance pattern in the detector. Ultimately there were no interferance patterns and no ether detected.

 

[mtworkowski@o]

Think of it this way. If one arm of the interferometer were parallel to the ether the effective path length would be 2L. But then the other arm would be orthoganal to the ether and the effective path length would be >2L. How much greater would depend on the Earths velocity through the ether but it would be greater.

It's this difference in path length that would cause the interferance pattern in the detector. Ultimately there were no interferance patterns and no ether detected.

You mean the average of 2L? And why would the one at right angles to the aether be >2L? I'm reading larger.

 

[Nickelodeon]

Why? Just wait 6hrs and the Earth plus apparatus will rotate 90 degrees.

Furthermore the Earth couldn't be moving through the ether in two orthoganal directions at the same time so the two arms of the interferometer are enough to detect an ether if it was there. A third arm is completely unnecessary.

If there is an aether then the Earth isn't moving through it, as shown by the MM experiments. However, it would be worth checking that the aether isn't traveling towards the Earth's centre. You would need to put one of the mirrors at the top of a cliff and the other mirror an equal distance but horizontal. The time the light takes to complete the two paths should be the same but the wavelength of the merging beams should be different.

 

[ZapperZ]

I think I read somewhere that there's a flaw in the Michaelson Morley experiment. Is this true?

This is a fishing expedition.

We require that when someone says "I read somewhere" or "I heard somewhere", that the exact valid references is given. Or else, it is impossible to know if what you read is correct or valid, if you interpreted it correctly, or if you are reading some crackpot information. So from now on, please provide the exact source if you wish to understand if what you read or heard is correct.

However, please note that, per the https://www.physicsforums.com/showthread.php?t=5374", dubious sources are not allowed to be referenced here.

when I think of IR frames, the one thing I think of is this. Is the speed of light dependent on the speed of the source. Here's the thought. Tell me if this is amusing. "Which is the worst of the collisions? A: 2 cars head on at 60 mph each. (both same mass). B: 1 car, 60 mph hits a bridge abutment. Oh and can I just add this? Somebody told me that because the MME used light going in two directions that it averages out the delta v. What do you all think?

I think you need to double check if you have understood relativistic velocity addition, because simply using Galilean velocity addition to negates the MM experiment just doesn't work.

Zz.

 

[russ_watters]

I'll draw the analogy the way I see it and you can clear it up. wind is west to east at 50 mph. Plane is traveling at 150 air speed west to destination and then back. Ground speed out is 200 and back is 100. Average is what? That's the problem these people are seeing.

Again, you have it wrong. In this airplane analogy, if the plane is traveling east-west, the wind must be traveling north-south. You'll never understand why the MM exp works if you can't comprehend why traveling perpendicular to the ether takes longer the faster the ether is moving.

Btw, when applied to a plane or a boat, this is called crabbing: http://en.wikipedia.org/wiki/Crab_landing

 

[russ_watters]

And not stationary plus one moving at about 85 mph? Or have I misunderstood "energywise"?

When two cars collide at the same speed or when one car hits a bridge, the final energy of the system is zero (everything is stationary after the collision) and all of the kinetic energy went into crushing the car. When a moving car hits a parked car, both cars are moving after the collision, and there is a remaining kinetic energy that has to be subtracted from the collision.

 

[Fredrik]

When two cars collide at the same speed or when one car hits a bridge, the final energy of the system is zero (everything is stationary after the collision) and all of the kinetic energy went into crushing the car. When a moving car hits a parked car, both cars are moving after the collision, and there is a remaining kinetic energy that has to be subtracted from the collision.

Consider that last scenario (a car hitting a parked car at 120 mph) viewed from a frame that's moving at 60 mph relative to the ground. It's going to look exactly the same as a collision between two cars both doing 60 mph in the frame where the ground is stationary (neglecting friction).

 

[zonde]

If there is an aether then the Earth isn't moving through it, as shown by the MM experiments. However, it would be worth checking that the aether isn't traveling towards the Earth's centre. You would need to put one of the mirrors at the top of a cliff and the other mirror an equal distance but horizontal. The time the light takes to complete the two paths should be the same but the wavelength of the merging beams should be different.

As I see idea that aether is traveling towards the Earth is most logical taking into account knowledge that light is bent near massive objects.
However I doubt that it is technically feasible to perform interferometer experiment vertically. You will have material deformations due to gravity and probably a lot of vibrations when rotation of apparatus is performed. And predicted outcome of experiment is not obvious to me as well. To make a prediction one needs good understanding of material deformations in accelerated frame.

And it is not obvious that wavelength of the merging beams should be different. Why should they differ?

 

[MikeLizzi]

I see a lot of confusion here. At the risk of adding more, allow me to suggest this discussion I have posted on my web site.

http://mysite.verizon.net/mikelizzi/MichaelsonMorelyAnalogy.htm

 

[mtworkowski@o]

This is a fishing expedition.

"We require that when someone says "I read somewhere" or "I heard somewhere", that the exact valid references is given. Or else, it is impossible to know if what you read is correct or valid, if you interpreted it correctly, or if you are reading some crackpot information. So from now on, please provide the exact source if you wish to understand if what you read or heard is correct.

Who is "We"? You're the only one who has complained about this. It's just a Question.

"However, please note that, per the https://www.physicsforums.com/showthread.php?t=5374", dubious sources are not allowed to be referenced here."

I'm not asking if the material I read is correct. I'm asking about the MME. Why are you reprimanding me for asking a question. Would it have been better to leave out the " I read somewhere"? We all get our ideas from somewhere.

"I think you need to double check if you have understood relativistic velocity addition, because simply using Galilean velocity addition to negates the MM experiment just doesn't work."

vi+vii = vi+vii/[1+(vi*vii/c^2)]. I think that's correct excuse the vii as v2. Whatever. Now can we get on or is there more that you wanted to say.

 

[mtworkowski@o]

Consider that last scenario (a car hitting a parked car at 120 mph) viewed from a frame that's moving at 60 mph relative to the ground. It's going to look exactly the same as a collision between two cars both doing 60 mph in the frame where the ground is stationary (neglecting friction).

I think that might be correct.

 

[mtworkowski@o]

I see a lot of confusion here. At the risk of adding more, allow me to suggest this discussion I have posted on my web site.

http://mysite.verizon.net/mikelizzi/MichaelsonMorelyAnalogy.htm

That's a great link...But does, and I not challenging the logic here, the MM machine have the capacity to turn the end mirrors so that the cross current beam comes back to it's source. Or is that too small an amount to be measured? Remember the one boat had to come back fighting a current and was coming back on a diagonal path.

 

[mtworkowski@o]

Again, you have it wrong. In this airplane analogy, if the plane is traveling east-west, the wind must be traveling north-south. You'll never understand why the MM exp works if you can't comprehend why traveling perpendicular to the ether takes longer the faster the ether is moving.

Btw, when applied to a plane or a boat, this is called crabbing: http://en.wikipedia.org/wiki/Crab_landing

Gee thanks. I can't comprehend what your not saying. I didn't know it was the perpendicular leg that was the one we were looking at. Sorry. I just saw where you said it. Double sorry.

 

[PhilDSP]

Just curious: has anyone performed a related experiment where both group velocity and phase velocity were recorded at the same time?

 

[mtworkowski@o]

Just curious: has anyone performed a related experiment where both group velocity and phase velocity were recorded at the same time?

What does that mean...exactly? I'm a lay person. Sorry.

 

[ZapperZ]

This is a fishing expedition.

"We require that when someone says "I read somewhere" or "I heard somewhere", that the exact valid references is given. Or else, it is impossible to know if what you read is correct or valid, if you interpreted it correctly, or if you are reading some crackpot information. So from now on, please provide the exact source if you wish to understand if what you read or heard is correct."

Who is "We"? You're the only one who has complained about this. It's just a Question.

"However, please note that, per the https://www.physicsforums.com/showthread.php?t=5374", dubious sources are not allowed to be referenced here."

I'm not asking if the material I read is correct. I'm asking about the MME. Why are you reprimanding me for asking a question. Would it have been better to leave out the " I read somewhere"? We all get our ideas from somewhere.

It isn't a reprimand (which would involve an official warning). I ask this from everyone who comes here and mentions the same thing. It is the only way to make sure what you read is the same thing as what you interpret. You don't think having a proper citation is at all important? It also forces you to pay attention to your sources, which is something we hope people will take away from being on here.

While PF is still an open forum, we are trying to impose a stricter standard than other forums that do not demand such quality in their discussion, and one of such standards is the use of clear citations of sources.

Zz.

 

[MikeLizzi]

That's a great link...But does, and I not challenging the logic here, the MM machine have the capacity to turn the end mirrors so that the cross current beam comes back to it's source. Or is that too small an amount to be measured? Remember the one boat had to come back fighting a current and was coming back on a diagonal path.

Insightfull response. Following my analogy would suggest that the MM apparatus was reflecting the beams back into the source (the flame). No, it didn't do that. Both cross current and parallel current beams were ultimately reflected toward an observing screen upon which the constructive or destructive interference would be seen.

I should revise my analogy to suit. That will add more complexity, though.

Oh, and yes, the position and angle of the mirrors were adjustable to the maximum precision of the day. I'd hate to be one turning set screws to move something 10 nanometers.

 

[PhilDSP]

What does that mean...exactly? I'm a lay person. Sorry.

Here is a good explanation: http://en.wikipedia.org/wiki/Group_velocity

 

[mtworkowski@o]

http://en.wikipedia.org/wiki/Group_velocity

 

[mtworkowski@o]

Here is a good explanation: http://en.wikipedia.org/wiki/Group_velocity

Thanks, that's exactly where I was when you posted.

 

[mtworkowski@o]

Since the 1980s, various experiments have verified that it is possible for the group velocity of laser light pulses sent through specially prepared materials to significantly exceed the speed of light in vacuum.

 

[ZapperZ]

Since the 1980s, various experiments have verified that it is possible for the group velocity of laser light pulses sent through specially prepared materials to significantly exceed the speed of light in vacuum.

.. and as in the NEC experiment from many years ago (have been discussed many times in here), you need to closely look at what they did and why no part of the wave actually traveled faster than c (another example where knowing the source and reading the actual source can make a difference). When the pulse is reshaped due to the anomalous dispersive material, you can get the appearance of a group velocity moving faster than c. But nothing here violated SR. Such effect doesn't happen in a typical MM experiment.

http://physicsworld.com/cws/article/print/482

Zz.

 

[mtworkowski@o]

.. and as in the NEC experiment from many years ago (have been discussed many times in here), you need to closely look at what they did and why no part of the wave actually traveled faster than c (another example where knowing the source and reading the actual source can make a difference). When the pulse is reshaped due to the anomalous dispersive material, you can get the appearance of a group velocity moving faster than c. But nothing here violated SR. Such effect doesn't happen in a typical MM experiment.

Zz.

My only reason for pointing it out was that I had read it and that the mathematics had been ambiguous...Excellent link, by the way.

 

[cristo]

My only reason for pointing it out was that I had read it and that the mathematics had been ambiguous.

Doesn't this go back to Zz's point though: where did you read about the experiment? Do you have a source showing the 'ambiguities' in the mathematics?

By the way, I've cleaned up the quote tags in your post on the previous page. It gets pretty confusing telling who said what if you either mess the tags up, or don't put them in. For future reference, to break up a quote so that you can write inside use [ /quote ] and [ quote ] (without the spaces in the brackets). You can hit "quote" on your post above to see the input, and see what I mean.

 

[pmb_phy]

I think I read somewhere that there's a flaw in the Michaelson Morley experiment. Is this true?

Yes. It is true. The flaw has to do with what is known as the extinction theorem. This is mentioned in Special Relativity, by A.P. French, Norton Press, (1968). pages 127-128. The author writes

As mentioned in Chapter 3, this result, although an essential feature of Einstein's formulation of special relativity, did not receive a convincing demonstration until much later. One crucial reason is that the propagation of light through a medium (even a transparent one) involves a continual process of absorption of the incident light and its reemission as secondary radiation by the medium - it takes only a very small thickness of matter to bring about this replacement. Thus, for example, with visible light, a thickness of 10^-8cm of glass or 0.1 mm of air at atmospheric pressure is almost enough to erase any possible memory, as it were, of the motion of the original source. This phenomena, known as extinction (even though it may not involve any appreciable loss of intensity in the light beam), has invalidated some of the observations (e.g. the apparent motions of binary stars, already referred to in Chapter 3) that were at first believed to provide confirmation of Einstein's second postulate - the invariance of c.

I'm sad to see that you had problems asking this question. Its a shame that this kind of thing happens here. Rest assured that not all of us think like that. I admire it when people ask questions like this. Its obvious that if you had more than a mere "heard of" then you probably wouldn't have had to ask this question. Please don't let this discourage ypou from asking questions in the future. If you are given a hard time again then feel free to ask me in PM.

Best wishes and keep asking these probing questions.

Pete

 

[mtworkowski@o]

Thank you Pete.

 

[pmb_phy]

Thank you Pete.

You're welcome.

Do you think that this may be what you were referring to?

Pete

ps - Check your PM.

 

[Nickelodeon]

As I see idea that aether is traveling towards the Earth is most logical taking into account knowledge that light is bent near massive objects.
However I doubt that it is technically feasible to perform interferometer experiment vertically. You will have material deformations due to gravity and probably a lot of vibrations when rotation of apparatus is performed. And predicted outcome of experiment is not obvious to me as well. To make a prediction one needs good understanding of material deformations in accelerated frame.

And it is not obvious that wavelength of the merging beams should be different. Why should they differ?

I think the the light beam going skywards against gravity would be blue shifted and red shifted on its return. The horizontal beam would be unaffected.

 

[mtworkowski@o]

Do you think that this may be what you were referring to?

Something along those lines.

 

[russ_watters]

Yes. It is true. The flaw has to do with what is known as the extinction theorem. This is mentioned in Special Relativity, by A.P. French, Norton Press, (1968). pages 127-128. The author writes...
Pete

Pete, I'd just like to clarify that that passage was actually referring to the MM experiment. It doesn't actually say that in the passage. Is the "this result" it is referring to (from the previous sentence?) a reference to the MM experiment? And if so, how, exactly does it point to a flaw in the experiment?

I am unable to find any confirmation of that (merely googling "michelson morley experiment extinction theorem" turns up only this thread!) and I don't have that book handy...