I need an variable attenuator, and was wondering if anyone has bought one of these:
http://cgi.ebay.com/Circular-Variable-N ... 999.c0.m14
Hopefully that link works. This circular attenuator looks similar to some that used to be available from Surplus Shed, but unfortunately are no longer available.
And since I also need a new variable beamsplitter, I was wondering how much light is reflected back. It would be lossy I'm sure, but the Edmund variable beamsplitter has a lot of losses also (I think it's an inconel coating).
For beamsplitting, a cube might be better, but two half wave-plates would be needed also (and be good for both 532 and 633), and that system would be quite expensive. I might want to just put up with the losses of metallized beamsplitter.
variable attenuators
-
Ed Wesly
variable attenuators
Don't give up on the polarizing approach! As far as waveplates go, mica is cheaper than quartz and surprisingly broadband or panchromatic! Of course then you need a broadband polarizing beamsplitting cube, but you might be able to get by with some beamcombining cubes found in 3 LCD projectors or video cameras.
-
Joe Farina
variable attenuators
Thanks Ed. I think you saved me considerable frustration, again 
I checked my half-wave film sheets, and they worked well enough for 532 and 633. The design wavelengths are 560 and 640, but they do a decent job, and can withstand my beams without being damaged (it's the Edmund half-wave film, but I don't know if it's available anymore). So the way's open to use a broadband cube -- hopefully Bob still has one that he can sell me.
I checked my half-wave film sheets, and they worked well enough for 532 and 633. The design wavelengths are 560 and 640, but they do a decent job, and can withstand my beams without being damaged (it's the Edmund half-wave film, but I don't know if it's available anymore). So the way's open to use a broadband cube -- hopefully Bob still has one that he can sell me.-
dave battin
variable attenuators
I too was suprised to find my how well my cube beamsplitter worked in the red as well as green laser light, i actually puchased an additonal waveplate and polarising prism, thinking i would need it for my green laser, if your still interested ill check to see how well it works for both colors ..............
-
Jeffrey Weil
variable attenuators
Hello Everyone,
If your going to use a 1/2 wave plate at a freq other than what its designed for here's a neat trick.
A 1/2 wave plate is an even multiple of its design wavelength in thickness. If you tilt it relative to the beam your changing its effective thickness and therefore the freq that it works on.
Put your waveplate in a holder and a polarizer after that. Rotate the polarizer until you've cut off as much of the beam as possible. Now tilt your waveplate and rotate it along with the polarizer until you find the point where the light is most extinguished.
Now if you do this by rotating the mag holder or whatever your using you have to do this everytime you move the optic. That would be like rotating it about its vertical axis. If you tilt it using the holder itself so its rotated about its horizontal axis it'll be set forever.
I used to use 670 plates for 632.8 with perfect results.
Jeff Weil
NorthBeach Holography Inc.
If your going to use a 1/2 wave plate at a freq other than what its designed for here's a neat trick.
A 1/2 wave plate is an even multiple of its design wavelength in thickness. If you tilt it relative to the beam your changing its effective thickness and therefore the freq that it works on.
Put your waveplate in a holder and a polarizer after that. Rotate the polarizer until you've cut off as much of the beam as possible. Now tilt your waveplate and rotate it along with the polarizer until you find the point where the light is most extinguished.
Now if you do this by rotating the mag holder or whatever your using you have to do this everytime you move the optic. That would be like rotating it about its vertical axis. If you tilt it using the holder itself so its rotated about its horizontal axis it'll be set forever.
I used to use 670 plates for 632.8 with perfect results.
Jeff Weil
NorthBeach Holography Inc.
-
Joe Farina
variable attenuators
Thanks for the tip, Jeffrey, I will keep that in mind. Since I will be using a combined 532/633 beam, polarization and beamsplitting options seem to have suddenly become complicated and expensive
Fortunately BobH gave me an excellent deal on a broadband polarizing cube. Next comes the issue of rotating the polarization of the combined beam (for beamsplitting) and then rotating one of the exit beams (to get Brewster's angle) for the reference beam. Unfortunately, efficiency is an issue because my HeNe power is so limited.
Broadband half-wave plates are incredibly expensive, as are fresnel rhombs (and other schemes as well). I was rather surprised to find little on eBay.
Fortunately BobH gave me an excellent deal on a broadband polarizing cube. Next comes the issue of rotating the polarization of the combined beam (for beamsplitting) and then rotating one of the exit beams (to get Brewster's angle) for the reference beam. Unfortunately, efficiency is an issue because my HeNe power is so limited.
Broadband half-wave plates are incredibly expensive, as are fresnel rhombs (and other schemes as well). I was rather surprised to find little on eBay.
-
Jeffrey Weil
variable attenuators
Hey Joe,
Heres something to keep in the back of your mind when you design you rig. It might come in handy.
Whatever you put into the cube doesn't really matter as far as polarization goes, your going to get perfect stuff out. Pure S and P. Within reason of course.
So, if you have two freqs like 532/633 you could try a plate in the middle of those numbers, or tilt something to get to the middle, and use that. The two beams will not be fully rotated or be a bit off as far as purity of polarization goes but the cube will clean that up. All you will lose is the ends of your beam splitting range. You won't be able to go from 10 to 90 percent on the beams, maybe just 30 to 70 (made up numbers). If that still falls within the beam powers you want you won't miss a thing!
I've never tried this so I don't know how far apart the two freq's can be and how much you will lose, etc...but it sounds good
For after the cube you could use that double vertical mirror trick to rotate one set of beams to match the other.
I need to give this more thought to figure out if it would work. It just came to me but you might be able to make a rig that will swing that double vertical mirror trick about some axis to rotate the polarization to any angle you want with NO wave plates.
All of the above is just brain storming. I have no idea if any of it will work.
Jeff Weil
NorthBeach Holography Inc.
Heres something to keep in the back of your mind when you design you rig. It might come in handy.
Whatever you put into the cube doesn't really matter as far as polarization goes, your going to get perfect stuff out. Pure S and P. Within reason of course.
So, if you have two freqs like 532/633 you could try a plate in the middle of those numbers, or tilt something to get to the middle, and use that. The two beams will not be fully rotated or be a bit off as far as purity of polarization goes but the cube will clean that up. All you will lose is the ends of your beam splitting range. You won't be able to go from 10 to 90 percent on the beams, maybe just 30 to 70 (made up numbers). If that still falls within the beam powers you want you won't miss a thing!
I've never tried this so I don't know how far apart the two freq's can be and how much you will lose, etc...but it sounds good
For after the cube you could use that double vertical mirror trick to rotate one set of beams to match the other.
I need to give this more thought to figure out if it would work. It just came to me but you might be able to make a rig that will swing that double vertical mirror trick about some axis to rotate the polarization to any angle you want with NO wave plates.
All of the above is just brain storming. I have no idea if any of it will work.
Jeff Weil
NorthBeach Holography Inc.
-
Joe Farina
variable attenuators
Thanks for your thoughts on this Jeffrey. I'm new to polarizing cubes, so I appreciate your advice.
I might have a simple solution, but will have to try it. While my 633nm light is precious, my 532nm light isn't. In fact, I will have to attenuate the 532nm beam by 50% or more, so that the power balance will be right. So I can afford to squander a lot of 532nm light. I have two half-wave rotators that have good efficiency at 633nm. So I can use these (at the correct angle for 633nm) on either side of the cube. Then I will just put the combined beam through the system, and the 633nm light should be conserved as much as possible. The 532nm light should be attenuated, but not by anything close to 50%.
I will just have to make sure my beams are color-balanced correctly by taking measurements after going through the system, instead of taking the measurements before going through it.
I might have a simple solution, but will have to try it. While my 633nm light is precious, my 532nm light isn't. In fact, I will have to attenuate the 532nm beam by 50% or more, so that the power balance will be right. So I can afford to squander a lot of 532nm light. I have two half-wave rotators that have good efficiency at 633nm. So I can use these (at the correct angle for 633nm) on either side of the cube. Then I will just put the combined beam through the system, and the 633nm light should be conserved as much as possible. The 532nm light should be attenuated, but not by anything close to 50%.
I will just have to make sure my beams are color-balanced correctly by taking measurements after going through the system, instead of taking the measurements before going through it.