445nm specific PBS cube GB

[The_Doctor]

More like 2mm x 4mm at the aperture with roughly 1 mrad divergence. But others may have had better luck.

I suggested an unfair contest. Single mode diode and CAY046.. That you got as close as you did is amazing. At the expense of half the light lost, would that 4mm be reducible to 2mm? And 1.5W plus out, total with both diodes and PBS? If so this is a standard I ought to aim for. 1mr is good. Plenty of single mode diodes with that CAY046 wouldn't be that good.

Now, the single-mode diodes are an entirely different matter. I haven't actually seen the beam from one in person, but I've been told that they're worlds better than the high-power 445 diodes. They might actually hit your specs of 2mm square and less than 1 mrad...

I'll try that too. My fast pitch-mod code is working out, I think, and my keen-ness to risk diversion into laser experiments is growing fast now. Those diodes sound like ideal matches for the reds and lenses I was using.

Well, the single mode blues aren't really "high power". They top out at something like 120 mw. By comparison, you can get nearly 3 watts out of a single 445 nm diode in a 9mm can, but it's multi-mode and will definitely need secondary optics to correct the fast axis.

I was slightly bewildered with all the options out there, but it sounds like you and others have found a good consensus on parts and methods. Makes things a LOT easier for me now. Those small PBS's are especially nice, it looks like a match with the tiny adjustable mounts I made. VERY tiny, maybe a record for small, even here. I never did set that up in a proper demo. Now, just maybe, I might.

The single-mode units sell for under $50, and the high-power multi-mode diodes are between $50 and $90, depending on which one you get. (A140, M140, or the 9mm can units)

5.6 mm full can for me. Ideally... If long chipped inside for narrow divergence, all the better. That might let me use a lens slightly further away for easier fine tuning of collimation, etc.. I already have 4 of the earlier projector diodes left untouched. A few single-mode diodes and a couple of those PBS cubes are something I should get ASAP. I don't like PayPal, quit using them over 2 years ago, so that might be harder to manage. I'll have to find out what payment methods any sellers are willing to take.. (If any sellers are reading, feel free to PM me about that, along with specs and prices).

I believe the wavelength shift per degree C of temperature change is very small, like .1 nm or something, so unlike the reds, cooling doesn't really change the color very much. (And in any case, cooling would tend to make the wavelength shorter, which is the wrong way to go if you're starting at 445. We want a blue that's closer to 460 nm, not 430...)

Indeed we do. I don't know why I had the vague idea that blue diodes might go long-wave with cooling. Probably just a strange notion of symmetry. I never had mine set up to test that. I pretty much decided I wasn't going to cool them as such, just to make sure the waste heat got pumped away fast.

[buffo]

I was testing much lower losses than 5% with that particular cube - you have rotated diodes for max thruput?

Hi Dave;

No, not yet. I only have the one test rig running (so just two diodes so far), and they're aligned vertically so they'll work with the secondary optics for fast-axis correction. (I'm actually still playing around with the 4X lenses you sent me earlier...) You may recognize the mount that is holding the cubes in that picture above... It's one of yours! Normally it's for anamorphic prisms, and the spacing is too close for the cylindricals, so I've been working with other mounting solutions for those. But it holds the cubes nicely!

Anyway, I did try rotating the cube slightly (maybe 15-20 degrees each way) by shimming the mount on either side, and I didn't notice any significant change in throughput. When I get some time later this week, I'll dial the power back down below 1 watt and check the losses again using my calibrated lasercheck. (The initial power measurements I took were with a home-made power meter using one of those Ophir heads. They seem pretty accurate, but I don't have a calibration certificate to prove it.)

Also, this weekend I should have time to loosen the back plates on the diode mounts and rotate them slightly just to be sure.

Still waiting to get the second set of diodes running so I can do a proper test with all 4 running at the same time. So far though the cube is holding up just fine. (And I can certainly live with 5% loss.)

I was slightly bewildered with all the options out there, but it sounds like you and others have found a good consensus on parts and methods. Makes things a LOT easier for me now.

I've actually been working on a "PL-Spec Projector" document for some time now. When it's complete, it should be a basic "how to" manual that will list parts, prices, and vendors, along with detailed assembly instructions on how to make your own 3 watt RGB projector from scratch. Green will be off-the shelf DPSS, but red and blue will be home-made modules with secondary optics. Most of the stuff is pulled straight from the forums here, but I've been working on a cleaned-up version that should be easier to follow. No idea when it will be finished though.

Adam

[planters]

To get the very best beam out of the high power blues I would drive a pair mounted identically and rotate the polarization of one with a wave plate prior to combining them with a PBS. Beam correct with either a cylinder pair or an anamorphic prism pair and then pass the roughly square beam through a pair of positive lenses to create a convenient near field focus. This is where you can trim the beam down to your maximum acceptable dimensions. In addition to cleaning up the beam the spatial filter placed here takes advantage of the near Gaussian energy distribution so that a 50% reduction of beam area here may reduce your power throughput by only 10-20%.

[buffo]

To get the very best beam out of the high power blues I would drive a pair mounted identically and rotate the polarization of one with a wave plate prior to combining them with a PBS.

Long term, that is exactly the plan. Well, except that there will be a provision for a quad arrangement as well. But Dave already has the waveplates in stock, and I'm testing one of the cubes now. (At one point he was looking into different PBS cubes that had the waveplate built-in to one of the faces, but it's better if it's mounted separately so you can adjust the rotation slightly for best throughput.)

Beam correct with either a cylinder pair or an anamorphic prism pair

Tests underway at the moment to determine which performs better. So far the cylindricals (4X) seem to be winning.

Regarding the Keplerian collimator with spacial filter, that's something that might be be incorporated as an add-on, but it's probably more effective on quads, hex-builds and "octo-moms" than on a standard PBS dual. For the optional high power multi-diode red design, it will definitely help though.

And yeah, a lot of this borrows heavily from your earlier work (and videos) on spacial filters...

Adam

[planters]

Don't pass on the spatial filter until you have tried it. They actually work best with a dual because the knife edging turns the beam from a simple Gaussian into a bi-lobed or extended rectangle. Also, by varying the FL of the two lenses you can adjust the beam size to just fill, but not overfill the scanners. The contrast improvement in the far field is remarkable. Try it out on your set up now just to get a feel for what the result looks like before you decide if you should design in the room for this in the projector.

[Bradfo69]

Are we getting any closer to getting these cubes? It's been quiet for awhile and I thought we were close to enough people being committed.

[platinum]

Are we getting any closer to getting these cubes? It's been quiet for awhile and I thought we were close to enough people being committed.

Partially my fault. I never got around to testing the cube I had and when I finally found time that worked, I couldn't get my current LPM to work in a reliable manner. Fortunately, dsli_jon has volunteered to test the one I had so it was shipped to him last week.


- Jonathan

[polishedball]

Any word on if he tested it out yet?

John

[dave]

Tests have been carried out by Buffo

OK, I've had a chance to do some more testing. With the power level down around 700 mw, the losses are pretty low - in the 2% range. When I bump the power up to a full watt, the loss increases to around 4 %, and when I hit the cube with the full 3 watts, the loss ends up at around 9%.

No damage apparent

One is with Jon, just waiting on his results

[buffo]

To expand on what Dave already posted:

I own three power meters, but only one of them (my Coherent Lasercheck) is actually calibrated. The other two are home-made meters, and as such the calibration is suspect on each of them. (One is based on an Ophir head, and the other is one of the Kenom-meter style units based on the surplus coherent heads that were available back around 2009 or so.)

On the lower end of the power spectrum, I checked the power with all three meters, and the results matched up really well, which would seem to prove that all three meters are reasonably accurate. However, once I get above 1 watt the Lasercheck is useless, so I can only go by the readings from the other two meters. It is possible that those meters are less accurate at higher powers, as I have no way to verify their calibration at anything above 1 watt.

Nevertheless, if the calibration was off, it's likely to be off on the low side. (That is, power might read lower once you get up into the multi-watt range.) So if that's the case, then the cubes may actually be performing better than my numbers suggest.

However, even if the readings I got were completely accurate, you're still talking about less than 10% loss. I let the cube cook for a day and a half with 3.3 watts hitting each face, and it never showed any signs of optical damage. The losses were consistent from start to finish, so I felt pretty good about the durability of the coatings.

Bottom line - based on my numbers and the low cost of these cubes, I can't see any reason why they would be a problem.

Adam