Aligning a Michelson interferometer

[johng23]

Are there any tricks for getting the two arms of a Michelson interferometer exactly equal, if all I have is a CW HeNe laser to align? The real experiment will be using the interferometer on short pulses, so I will only see interference when the path lengths are equal to within 30 microns or so. I can't think of any way, since the coherence length is probably 10's of centimeters.

I saw some website that was saying that as you scan the delay of one arm, the fringes will move in one direction, and when you pass the equal length point, they will move in the opposite direction. But that makes no sense to me. Maybe I'm just not seeing it, but if I visualize scanning two beams past each other, the fringes continually move in one direction, and the absolute difference in phase is not discernable.

 

[Cthugha]

Just a question to understand your problem better: You have already aligned the two beams such that they are perfectly parallel and are now looking for the exact position of zero relative delay, right?

If so, how exactly are you moving your movable mirror? Coarse movable delay line, piezo-driven fine delay or both? And what exactly is the highest possible distance you can move it?

 

[johng23]

Just a question to understand your problem better: You have already aligned the two beams such that they are perfectly parallel and are now looking for the exact position of zero relative delay, right?

If so, how exactly are you moving your movable mirror? Coarse movable delay line, piezo-driven fine delay or both? And what exactly is the highest possible distance you can move it?

The two beams probably are as close to parallel as I can make them. My movable mirror is on a coarse translation stage, with a total travel of 50 mm. The smallest step size is 1 um, and I'm not even sure how repeatable that is.

 

[Andy Resnick]

Are there any tricks for getting the two arms of a Michelson interferometer exactly equal, if all I have is a CW HeNe laser to align? The real experiment will be using the interferometer on short pulses, so I will only see interference when the path lengths are equal to within 30 microns or so. I can't think of any way, since the coherence length is probably 10's of centimeters.

Tricky problem- the best I could think of, if you have the equipment, is to use filtered broadband light. For example, I have some filters that pass about 20nm wide spectra centered on green, blue, red, etc. The coherence length approaches 30 microns in some cases (500 nm +/- 10 nm passband)

 

[Cthugha]

A step size of 1 micron might make measurements complicated. Unless of course you are not interested in visible light, but far infrared or the terahertz range. In that case it would not pose a problem.

Regarding the problem itself, Andy Resnick is right. Filtering broadband white light or something similar is a good idea. As an alternative, there are some common LEDs with coherence lengths in the micron range. Those should also not be that expensive.

 

[johng23]

Ok thanks everyone. I ran out of time, I guess I'll just try to measure it with a tape measure and look for my signal the hard way.