sizes of cylindrical lenses, PBS cubes and waveplates, advice needed

[Laser57]

Thanks, left message for him with what I need.

Does anyone here have experience crunching numbers to determine the amount of expansion I can get when using an IR diode with a 100um emitter size and 30 degree beam (actually, 100x1 and 8/30) straight into a PCX collimator with a 250mm FL? I could use the help to verify my own figures.

I've read there is a general rule of thumb for astronomy telescopes of no more than 60X magnification per inch of refractor lens diameter, but I don't know if this also applies to expanding a laser beam too, but since a BE is essentially a telescope, I'm guessing the same applies, does it? I've tried googling searching for the limitations of the amount of expansion I can achieve using a refracting BE, but unable to find the information yet.

[Laser57]

20 dollars via PPal to who ever can correctly solve this, keep the money or donate.

Chris

[kecked]

I read your post wrong. Keep your money. You can only do one polarization combine. The result is a beam with p and s polarization so you can not add two such beams with a pbs cube. You can as steve suggested use dichroic as is done in every projector. You want a dichroic that passes 520 but reflects anything above say 490. Just make sure the reflect is good to 20nm or so beyond your highest blue. Other possibility is to knife edge. You really only use one beam at a time so why not. That seems easiest. Or mount one behind the other so the beam just passes the barrel of the other. Could also be done with fiber but that is using a jack hammer to open soup cans.

Last point I am assuming you already combined the beams with pbs to get mixed blue. But you can only use one pair. Knife the pairs before the pbs to combine two of each wave then pbs then dichro

Steve.....axicon!

[Laser57]

Thanks, the information is helpful and will be used. I am unable to try anything now because I am out of country, but I am researching this now to order parts to have them sent home in the ready. I had considered knife edging but threw the idea out believing one of the beams being slightly off axis would cause a problem. Perhaps a knife edge of just two diodes will have so little spacial difference between them that won't be a factor if one of the two isn't in the exact center of the PCX lens, but if I do use a KE, due to wanting to expand the beam to several inches diameter, I would then need to add an expander lens. I wanted to avoid using another lens to reduce the complexity as well as total loss of energy. The 140-150mm diameter lenses I found to collimate the beam are over 5 inches diameter with AR coating for 1064nm, usable at 520nm too, and were manufactured for use with a 1064nm YAG, bought four of them for what I thought was a good price.

While there is no reason to use both diodes at the same time I have concern the focus for each will be different, so I need to figure out a way to compensate for the two widely different wavelengths so I can set the pointer to infinity focus at both wavelengths, any suggestions? Maybe I will need to add another lens after all?

The offer for a 20 dollar reward for info was for the divergence question with a 100x1um emitter and PCX lens with a FL of 250mm, anyone?

[buffo]

You want a dichroic that passes 520 but reflects anything above say 490.

Actually, he needs just the opposite. He wants to combine 520 nm green with 1064 IR, so ideally he wants a reflect-green, pass everything LOWER.

Chris: These are sometimes marketed as "cold mirrors", as they reflect the visible light and allow the IR to pass through them. But yeah, most decent optics suppliers will have something off-the-shelf that will work. (Semrock, Edmunds, etc)

Other possibility is to knife edge. You really only use one beam at a time so why not.

My understanding is that this is for precise aiming of a laser cutter. So you would always be a few mm off if you knife-edged them.

Adam

[Laser57]

This won't be used for cutting or the green as a guide, just a simple laser pointer with the beam expanded to about 5 inches using a large YAG PCX lens I found on ebay. I think I have my answer now, except for the question in regard to how much expansion would occur with a 100x1 emitter size but I've decided to use a much lower power single mode laser diode with a 5X1um emitter size. Does anyone know if the output from an emitter that small can be expanded to 5 inches without reaching some kind of magnification limit? The expansion would be by simply letting the beam expand that wide before collimation.

Here's something I bought on ebay cheap for parts, photo below:



The last photo of the bunch is much higher resolution than the first few photo's.

The seller had no clue what wavelength it is (Edit: New photo's received, appears everything is 633nm, but I still don't know what this configuration does but it appears to have a mounting slot for a gas laser). Anyone here care to speculate on what this thing might have been for? I see a lot of waveplate mounts in there.

EDIT: I believe this unit is a interferometer, but what purpose it would have with a single 633nm source I don't know yet.

[Laser57]

I guess I'm now hijacking this thread, now the original op isn't adding to it, but I have a question about PBS cube combining too.

Easy question for those in the know; is there any reason the beam from a LD must be collimated prior to passing through a PBS cube? I want to use a cube to combine the power of two diodes, but let the beam expand from the diodes through them, as long as they are large enough cubes, why not?

[planters]

If you want to be a perfectionist, the PBS is coated to minimize/maximize reflection at a given angle. The expansion adds another vector that will add to increase the net angle. Otherwise, it will work.

The other consideration is that good PBS cubes are expensive and the surfaces will have to accommodate the largest dimensions of your beams.

Half wave plates are another matter. As the beam enters off-parallel the distance between the layers increases and the plate becomes less optimally tuned for the design wavelength.

Why does it matter if it is used for cutting?

[Laser57]

Thank you for your help. I won't use it for cutting, I want to use it for signaling tests using IR and need low divergence. Regarding the half wave plates on that unit, I won't be using those for that project.