Best way to separate almost parallel light rays?

[andresB]

It is probably a basic question, but here I go.

Suppose a source emit two almost parallel light rays. The rays travel very close to each other so at short distances (the ones in laboratory) they are indistinguishable and any measurement device will detect them as one.

So, the question is, using reflections, refractions and any other tool from geometric optics, what good ways ways are there to maximize the separation between the two rays?

 

[Charles Link]

It is probably a basic question, but here I go.

Suppose a source emit two almost parallel light rays. The rays travel very close to each other so at short distances (the ones in laboratory) they are indistinguishable and any measurement device will detect them as one.

So, the question is, using reflections, refractions and any other tool from geometric optics, what good ways ways are there to maximize the separation between the two rays?

One way that might work is to use a paraboidal mirror, e.g. an off-axis paraboloid and separate them as they come to a focus in the focal plane using a small aperture to select the one of interest. A spherical reflector might work almost as well. They may even come to a focus somewhere outside of the focal plane if they originate from point sources and are slightly divergent, but in any case, that is one possible method.

 

[mfb]

What is "very close"? If diffraction is not relevant, it is easy to separate them. If it becomes relevant, it can be impossible because the beams actually overlap.

 

[andresB]

Suppose that diffraction between the two rays is not relevant at all.

 

[nasu]

The rays travel very close to each other so at short distances (the ones in laboratory) they are indistinguishable and any measurement device will detect them as one.

If they are indistinguishable to any device, what is the meaning to even say that you have two beams and not just one?

 

[andresB]

If they are indistinguishable to any device, what is the meaning to even say that you have two beams and not just one?

I meant indistinguishable for standard laboratory setups. In principle, I we were able to put the measurement device (really) far away from the source the two rays will be able to separate one from each other and be distinguishable by the measurement.

 

[mfb]

Standard laboratory setups can distinguish between everything where talking about two different beams is meaningful. If you just want to separate them optically, use a diverging lens or mirror. If they happen to have a well-defined and opposite polarization, use a polarizing beam splitter. If they have a well-defined but different frequency range, you can use a prism or diffraction grating.

 

[blue_leaf77]

As mfb and Charles have suggested, there are many options to realize what you want to do, depending on the properties (frequency/color, polarizations, etc) of the two beams. The question for you is, do you know how these properties of the beam are?

 

[andresB]

As mfb and Charles have suggested, there are many options to realize what you want to do, depending on the properties (frequency/color, polarizations, etc) of the two beams. The question for you is, do you know how these properties of the beam are?

Yes. same frequency, same polarization, almost parallel beams that originate from the same point, and they don't have to be emitted at the same time (that is why there is no need to consider interference between them)

The problem is that the angle between them is very tiny and a single reflection or refraction don't do the trick. For example, before posting the question here I considered an idealized (of course) setup with two parallel mirrors, one ray can be directed so it moves back and forth between the same points of the mirrors while the other will start diverging but with that setup I would need more than 20 reflections for the separation to be measurable.

I have not thought about diverging lenses, they look promising, I will think about them to see what I can do with them.

 

[mfb]

If they are not emitted at the same time: a rotating mirror, similar to early speed of light measurements.

The problem is that the angle between them is very tiny and a single reflection or refraction don't do the trick. For example, before posting the question here I considered an idealized (of course) setup with two parallel mirrors, one ray can be directed so it moves back and forth between the same points of the mirrors while the other will start diverging but with that setup I would need more than 20 reflections for the separation to be measurable.

There is no setup with that property where diffraction would not be relevant.

If you are thinking about specific examples, give them, otherwise this thread continues with way too much speculation about what you want to do.

 

[blue_leaf77]

originate from the same point

Earlier you said that the beam are propagating side by side and are very close to each other, but now you say that the two beams actually originate from the same point. If your last statement is true, then these two beams are actually one single beam and hence there is absolutely no way to separate them.

 

[Tom.G]

Using lenses, think microscope.