Mitsubishi ML520G71...Red Holy Grail or Flashlight Fail ??

[andythemechanic]

Actually from the datasheet it is more like 5.6mm mrad. In principle IF you are willing to accept this beam quality you can stack 6-7 of these diodes in the fast axis without sacrificing beam quality. You however will have to stack along the long axis after the collimation lens, and do a beamshaping afterwards to reach a small spot. Together with polarization coupling this can give you over 3W of beam power, even when staying in spec of the diode.

Andreas

[dave]

I just had a run of x/y adjustable housings arrive.

See here:

http://www.photonlexicon.com/forums/...961#post227961

[kiyoukan]

that is a very interesting way to adjust the XY took me a second to see how it worked.
So the plate holes that hold the diode are a bit bigger than the bolts, so then you just tap it a bit and tighten it down.
Not the most accurate but a way that works.

[dave]

Yes, that's how the design works. Plenty accurate for me with the right adjustment technique.

that is a very interesting way to adjust the XY took me a second to see how it worked.
So the plate holes that hold the diode are a bit bigger than the bolts, so then you just tap it a bit and tighten it down.
Not the most accurate but a way that works.

[planters]

Well....
An interim report. I did not have all the components to do a rigorous test of combining the G71s between the telescope optics in order to see if there might be a knife edging advantage, however I did a thorough test with a pair of 445 diodes which have about 1/4 the divergence. Stability and precision were not an issue. I used micrometer mounts on an optical bread board, custom lens mounts and MANY repeated tests. It was a wash. Everything gained from the proximity of the smaller spot on the knife was lost from the telescopic magnification of the knife process. At smaller scales the flaws of the knife will become more significant and so I would call the process a net loss.

The much larger divergence of the G71 might yet allow this method to be useful. I have my doubts, but will run another series of tests when all the pieces come together. We'll see

[andyf97]

Just received a test batch of these from Dave, i must say they are perfect and i am happy to see he took my advice to thin the base down a bit.

I just had a run of x/y adjustable housings arrive.

See here:

http://www.photonlexicon.com/forums/...961#post227961

[Solarfire]

I was just working on my quads I'm putting together and I was prealigning diodes in the mounts to get the best possible beam profile. I did a little experimenting with beam cleanup; mainly the vertical wings and I got some pretty good results with slightly clipping the top of the beam which strangely enough also cleaned up underneath the beam. In the pics I just roughly placed a razorblade to clip the top of the beam at the diode mount aperture this gave amazingly good results with reasonable power loss. The power measured before cleanup was 191mW and after 178mW, the collimator was daves 2mm lens.

[andyf97]

Try it with the sharp part of the blade, I think it will work better.

[Solarfire]

Yeah, I tried that at first, but I couldn’t place the magnet on the back side to hold the blade for testing because it would be in the open part of the center of the blade. I’m going to cut some 15mm pieces of the sharp edge of the blade and just glue them into position on the front of the mounts. I may test placing the blade at the module aperture after the beam combining/correction, the results may be even better, less loss, cleaner?

[badger1666]

got my 3 banger up and running today tested @1.534mw , installed in my projector got everything alighned, fired it up down the hallway
and said to myself ohh f**k.
man what a difference that extra 500mw of red made to my setup, now she bangs 3.9w after optics with 4.5w raw power
need some better dichros to reduce the losses and here's the pics taken in daylight with fog