How hard to find a NUBM44 that hits 455nm or higher?

Alaskan said:

Telescoping the beams together makes them look more like one beam, but it won't help the divergence, for this diode, I'd have to expand the beam quite a bit, it's already high enough, telescoping the beams together smaller would just increase the divergence even more. I don't think I am telling you anything you aren't probably more schooled in than I am, but thought I'd mention it for the others who might consider this approach. This diode is almost as bad as the output of a infrared FAP laser, the light becomes a rake very fast, unless focused down to a spot, or expanded.

Click to expand...

Yes it is. I have recently measured the NUBM44's useful output. I have worked out that the divergence is 11.06x1.06mRad. Very high indeed.

My measurements showed a bit higher, but they weren't done as precise as I would have liked, so I believe your numbers. Edit: Let me make a disclaimer though, it all depends upon the diameter of the beam when making the measurement, if using a three element lens the divergence will be a little less compared to a G2 with its shorter focal length producing a slightly smaller beam width. But then a three element lens will cut some of the beam off too, problematic to compare the two.

Should use a cylindrical lens pair to correct the long axis before the beam expander.

I focused mine as best I could, and used verniers. I only measured the useful output, not the extra wings etc. I measured at 780mm away, and then at 1580mm. I got 8.88x0.54mm @ 0.78m and then 17.73x1.39mm @ 1.58m. Just for anyone to reference or correct.

Let me make a disclaimer though, it all depends upon the diameter of the beam when making the measurement, if using a three element lens the divergence will be a little less compared to a G2 with its shorter focal length producing a slightly smaller beam width. But then a three element lens will cut some of the beam off too, problematic to compare the two.

Oh yes, I forgot. Mine was measured with a G2 lens. :beer:

AaronT said:

Should use a cylindrical lens pair to correct the long axis before the beam expander.

Click to expand...

RedCowBoy has used cylinder lenses with this diode and is very happy with the result. I'm different in my approach, I like to take the raw output of the diode and use a large diameter long focal length lens for my pointer with this diode, up to 3 inches in diameter seems a good choice to me, even larger. It works great for me, I don't care if the beam is squared up so the fast and slow axis outputs of the diode are better matched, because the divergence with a lens that big is so much better for both axis', their divergences are reduced so much that it doesn't matter to me, and if focused to a point who cares, a point won't show a rectangle, it will be too small to see and still burn like hell. Although I will agree correcting the beam can make it burn hotter, I just don't think it really matters, for burning it still toasts everything it touches when focused into a spot.

Alaskan said:

RedCowBoy has used cylinder lenses with this diode and is very happy with the result. I'm different in my approach, I like to take the raw output of the diode and use a large diameter long focal length lens for my pointer with this diode, up to 3 inches in diameter seems a good choice to me, even larger. It works great for me, I don't care if the beam is squared up so the fast and slow axis outputs of the diode are better matched, because the divergence with a lens that big is so much better for both axis', their divergences are reduced so much that it doesn't matter to me, and if focused to a point who cares, a point won't show a rectangle, it will be too small to see and still burn like hell. Although I will agree correcting the beam can make it burn hotter, I just don't think it really matters, for burning it still toasts everything it touches when focused into a spot.

Click to expand...

That works great for a single diode and I love the simplicity.

Each of the NUBM44s in this array needs it's own G2 lens before polarizing and combining so the correction has to come after.

Or am I missing something and the raw output could be combined first?

I suppose the diodes closest to the PBS cubes could be moved back so the raw output divergence of all diodes align.

You are right, if the beam is not fully collimated before correction, or shooting through a cylinder pair, there can be a problem. I tried to shoot a beam which was a bit out of focus through a cylinder pair and the expanded beam was expanding too much before it reached the 2nd collimating cylinder lens, over shooting it too much.

Moving the diodes back so their path lengths between the PBS and dichro are identical would be a must, if both collimating and correcting after those optics, but not sure it will work out for you due to the problem of the beam expanding too much prior to reaching the 2nd cylinder lens, that and the focal lengths might cause another problem.

The convention has been to correct each beam individually before doing more with them such as PBS etc, but I don't see that it is absolutely required, or that work arounds can't be made, perhaps compromises accepted so that the path lengths are made to match one another to keep the individual divergences from each diode the same. However, the geometries of the beam shapes will cause problems, unless wave plates are used. I suppose it just depends upon whether you want your beam footprints to all land on top of one another, or allow them to be in a cross pattern. If you don't care, then wave plates won't be used but then you cannot correct a crossed beam shape.

Am I seeing this correctly guys?