Anamorphics and waveplate help?

Hey all,

It's been awhile since I did some laser building so I just ordered two sets anamorphics prisms (coated 445 and 650nm) and a waveplate.

I intend to combine two M140s with a PBS and then try fix the divergence. (the other prisms are for the 635s which I'll play with at some other stage)

I've always hated the + beam from my current setup hence this time I wanted to use a waveplate so I get a - profile of around 4 watts. The part I'm trying to figure out is, the anamorphic prisms are polarized. Does this mean I won't be able to shape one of the beams after it exits the PBS? Because each beam has a different polarization?
Ideally I want to use one set of anamorphics after both beams exit the PBS.

I know I'm being a little impatient and could find out by experimenting when the arrive....but it'll only get in 2-3 weeks time. Thanks

Put one through the waveplate. This means both beams will be "vertical" as they enter the PBS. The output from the PBS will be just like a single beam, but twice the power. Since both beams are overlapped with the same beam profile, you correct them both at the same time with the prisms.

If the the prisms are indeed polarized, you'll need to use the prism prior to the beam entering the polarized beam cube. Otherwise, what'll happen is that the prism will block/reflect the output of the perpendicularly polarized beam.

Even if your prisms aren't polarized you don't want to be correcting for both at the same time. To work with a PBS, the polarizations of the beams need to be perpendicular to each other:



So say you want to use your prisms to correct an axis that is wider in the X-axis (looking into the beam). When you rotate one of those beams to work with the PBS, now one of the beams' Y-axis is now oriented in the X-axis in the combined beam. The end result is that you'll end up correcting the wrong axis of one of the beams.

Finally, you don't want to be passing beams that don't need to be corrected through extra correction options -- even if those correction optics won't affect that beam. Each set of optics you pass your beam through will cause losses in power. So you'll want to perform all the corrections on the specific beams you need to correct.


As you'll be experimenting, what you might want to see is if combining those two beams in the PBS will effectively balance out the mismatched axis divergences in the final beam because now each axis in the final beam has both the X and Y axis of any single beam.

Another thing to try is knife-edging and then performing the beam correction on the knife-edged set of beams.

Bionic-Badger said:

If the the prisms are indeed polarized...

Click to expand...

Let me stop you right there. What in rod's name are you talking about? :can:

I'm just going by what was said in the OP. I'm not particularly sure what was meant by "anamorphic prisms are polarized" myself, though I have encounter X-cubes with polarized film on them. If what was meant was that it affects one axis, the "anamorphic" part of the descriptions should have been sufficient.

Thanks for the replies. So I'm going to have to buy another set of prisms to correct the each beam before they enter the PBS?
Regarding your question about the anamorphics being polarized, herewith is the link and description of the prisms:
Anamorphic Prism Pair blue and red

Hmm, yeah it seems like you'll need to ensure that you've correctly aligned your beams before passing them through those. Before buying another pair, make sure the polarization the prisms utilize is compatible with the axis you're planning on correcting.

On the subject of a second set of prisms: if you don't mind having the output coming out at an angle, you can try using a single prism per laser. Utilizing a pair just doubles the amount of correction as a single prism while at the same time having the beam exit in the same direction that it entered.

Bionic-Badger said:

Hmm, yeah it seems like you'll need to ensure that you've correctly aligned your beams before passing them through those. Before buying another pair, make sure the polarization the prisms utilize is compatible with the axis you're planning on correcting.

On the subject of a second set of prisms: if you don't mind having the output coming out at an angle, you can try using a single prism per laser. Utilizing a pair just doubles the amount of correction as a single prism while at the same time having the beam exit in the same direction that it entered.

Click to expand...

So I can use one prism to narrow the fast axis with the only complication being that the incident beam will be at a different angle when it exists the prism?

I see what they're saying. It just means the losses will be slightly higher on one of the polarization components. You can take the hit and loose 3% (wild guess) of the power, or you can correct each beam individually before the PBS/waveplate.

I'd recommend taking the loss. It will be impossible to tell the difference without a power meter anyway. It also means less aligning and 2 fewer components to buy.

blrock said:

So I can use one prism to narrow the fast axis with the only complication being that the incident beam will be at a different angle when it exists the prism?

Click to expand...

Yes you can do that. If you look at anamorphic prism pair ray diagrams you can see that the beam is narrowed after it exists the first prism. The only caveat is that the amount of beam size reduction will be reduced (halved) as compared to having both prisms. I'd perform some tests and see if a single prism meets your needs, and then if you need an improvement, you can go buy another pair.

Cyparagon said:

I see what they're saying. It just means the losses will be slightly higher on one of the polarization components. You can take the hit and lose 3% (wild guess) of the power, or you can correct each beam individually before the PBS/waveplate.

Click to expand...

After reading a bit more about anamorphic prism pairs, the losses due to incorrect polarization (s-polarization) can be substantial, on the order of 30-70% depending on the angle of incidence; it's probably more with a special coating. The prisms are designed to operate at the Brewster angle and their low-losses are only applicable for p-polarized light. I do think it'd be best to do the correction prior to entering the PBS to minimize losses.

Thanks Bionic-Badger. +1 to you.
You've put me on the right path. I've now done much on these prisms and it's interesting that the guys seem to EXPAND to slow axis rather the SHRINK the fast axis. I was all for shrinking the fast axis. Why would you want a big dim square rather than a tight, brighter flat beam?

Thanks Bionic-Badger. +1 to you.
You've put me on the right path. I've now read much on these prisms and it's interesting that the guys seem to EXPAND to slow axis rather the SHRINK the fast axis. I was all for shrinking the fast axis. Why would you want a big dim square rather than a tight, brighter flat beam?

Thanks Bionic-Badger. +1 to you.
You've put me on the right path. I've now done much on these prisms and it's interesting that the guys seem to EXPAND to slow axis rather the SHRINK the fast axis. I was all for shrinking the fast axis. Why would you want a big dim square rather than a tight, brighter flat beam?

***SORRY FOR THE TRIPLE POST**** Don't know why it did that?!?

blrock said:

it's interesting that the guys seem to EXPAND to slow axis rather the SHRINK the fast axis.

Click to expand...

Neither, actually. You expand the fast axis, so it diverges less as a consequence.

Cyparagon said:

Neither, actually. You expand the fast axis, so it diverges less as a consequence.

Click to expand...

Please excuse my ignorance, but wouldn't expanding the fast axis result in the width being the same after x meters. ie if it was 3 inches after 10 meters, now it would be 10 inches and also fatter ?
Hope it makes sense...