Question about Michelson Morley experiment

[Kalidor]

Hi everyone. I'd like you to help me make sense of a question I found about the Michelson Morley experiment. This is it:
"Explain how and why the uncollimated light rays form interference fringes?
Can you say what would happen if the rays were collimated?"

Does this question even make sense? How would you answer in that case? I study math, not physics, so I tried to read a description of the experiment but this question doesn't quite add up with what I read.

Thanks in advance.

 

[Simon Bridge]

Start from basics:
1. how do fringes form at all?
2. what does it mean for the rays to be "collimated"?

... note: this is more a physics question than a math one.

 

[cr7einstein]

you will get the answer in feynman's sum over history theorem.i found it to be way too unrealistic in the beginning, but when you really understand it its beautiful. it describes how an object goes from a point to another in every possible path. you have to be comfortable with wave particle duality to understand it. you think of the object as a wave, and then assign numbers to its positions in the wave cycles eg crests or troughs. then you find the probability of each path, and get your answer by estabilishing a relationship with the wave's amplitude. its not that complicated, trust me.

 

[Kalidor]

Start from basics:
1. how do fringes form at all?
2. what does it mean for the rays to be "collimated"?

... note: this is more a physics question than a math one.

I expect fringes to form because of a difference in the path of the two rays, one of them getting to the screen earlier than the other. The rays are collimated if they do not disperse much with distance, but tend to stay coherent and "point" in the same direction. I don't even understand what collimated and uncollimated refer to here. The rays of light at the source? The rays after they have gone through the mirrors/glasses?

 

[Kalidor]

you will get the answer in feynman's sum over history theorem.i found it to be way too unrealistic in the beginning, but when you really understand it its beautiful. it describes how an object goes from a point to another in every possible path. you have to be comfortable with wave particle duality to understand it. you think of the object as a wave, and then assign numbers to its positions in the wave cycles eg crests or troughs. then you find the probability of each path, and get your answer by estabilishing a relationship with the wave's amplitude. its not that complicated, trust me.

Thanks. Do you have any reference for this? I don't think I can work out the answer from these hints.

 

[Simon Bridge]

I expect fringes to form because of a difference in the path of the two rays, one of them getting to the screen earlier than the other. The rays are collimated if they do not disperse much with distance, but tend to stay coherent and "point" in the same direction. I don't even understand what collimated and uncollimated refer to here. The rays of light at the source? The rays after they have gone through the mirrors/glasses?

See: http://en.wikipedia.org/wiki/Collimated_light

These days students meet diffraction and interference in the context of lasers - or tightly collimated beams. Such a student may come to think the fringes will only form when you have a tight beam. So the question is to get you to think about it more closely. After all, these guys didn't have lasers - what did they use instead? Did they use a collimator in the apparatus for example?
See: http://en.wikipedia.org/wiki/Michelson–Morley_experiment

To get interference fringes you need some sort of path difference - how does that occur in the apparatus? What are the paths that are different? Do you know how the apparatus is supposed to work?

I'll agree the question is poorly written - where did you see it?

 

[Kalidor]

Hmmm no I suppose they did not use collimated light back in the day, just white light I suppose. I also know that of course there was no path difference at all since the aether doesn't exist and light has the same speed in all inertial frames of reference, but the only thing I could think of qualitatively is that interference fringes will be less evident on the screen, as light disperses and less photons will reach the screen. Do the things that cr7einstein wrote have to do with the "solution"?

P.S.
The question was formulated by a professor, it is in my class notes.

 

[Simon Bridge]

"I expect..." (post #4)
"I suppose..." (post #7)
... did you go and find out so you wouldn't have to rely on supposition and expectation?
This experiment is well documented.

Feynman's sum-over-many-paths is actually the solution...
See Feynman explain the method in freshman-friendly terms:
http://vega.org.uk/video/subseries/8

...but most people don't meet it formally before post-grad.
You should be able to answer the questions simply just using your understanding of interference and diffraction in ray-optics and of the experiment in question.

i.e. do you get fringes formed from light where the rays are at random angles?
If you don't know, suppose, or expect something, then you should check through your notes and online.
It will probably help to view the question in the context of the particular lecture - what was the prof talking about, what was he expecting at the time? If you don't remember, maybe you should ask?