40 Watt 638nm module question

Browsing the interwebz, found this:

Attachment 45240

Now, 1st of all I realize that this little module will probably cost as much as a house, or at least a new car. But in the interest of scientific curiosity, if you screwed in a fiber optic cable (SMA end, 400um, multimode (I'm assuming), what would be coming out of the other end of the fiber? I'm sure that it couldn't possibly be as simple as laying out a few collimating and corrective optics and having a 40W usable beam, would it?

They say integrated FA and SA collimation:

"Through the optimization of semiconductor chip structures and optical parameters, DILAS’ T-Bar architecture delivers high beam quality and high power using standard micro-optic fast-axis collimators (FAC) and slow-axis collimators (SAC), all assembled with automated processes. The T-Bar is a monolithic, multi-emitter source, allowing the handling of multiple emitters during each manufacturing step, to lower complexity and ease manufacturing. The result is enhanced reproducibility, beam quality and fiber-coupling efficiency."

More info:
http://www.dilas.com/pages/products....y=25&series=47

looks ace, ill take two please

Ignoring the potential cost for now, I think that you could get a usable beam out of this, but not a good one. I suspect that the far field quality would be similar to an 808nm fiber pigtailed diode equipped with a 400um fiber with a NA of .22. Compare this to a 60um x 5um 638nm Mitsu diode. The effective divergence of a 0.22 NA is 13 degrees or 780 mrad which is similar to the Mitsu diode before it is collimated, but the Mitsu diode source is many times smaller and so will be collimated many times more effectively.

Ok, thanks Eric. I'm not surprised that it wouldn't be easy. Can't make a silk purse .. etc .. But. let's say for a moment that there is no concern with scanner mirror sizes, and the aperture beam could be up to 1 inch in diameter. Do you think that it would it be possible to have a decent divergence with this scenario, or would the same limiting factors still apply?

400 micron fiber sucks. Take it from some one who has been there, got the Tee shirt, done that, and ignore this. Not with a nice Dilas, but one of their inferior competitors. 25mm mirrors barely work as the X mirror. In fact you will have vignetting (clipping) as the mirror moves. The Y mirror has to be even bigger. 400u is a flashlight. You have to mask off over 40% of the beam diameter to fit the low divergence portion on the mirror.

That laser cluster is for video projectors...

Show producers demand fast scanning these days. Useful fat beams are provided by cheaper MSR series lamp sources.

200u is a bear and is marginally useful at best for beam shows. Even then you still end up masking about 30% power to make it useful.

Steve

Quote:

---

Originally Posted by planters

Ignoring the potential cost for now

---

I've found you all one for around $3,000. :cool:
http://www.ebay.co.uk/itm/Dilas-D4FS...item2ed78f0f6d

I estimate you'll get around 12mm beam, 4-5 mRad from a 400um; 70-80% coupling efficiency
(Do not take as an exact spec; figures from my other fibre experiments - I've not tested this exact diode stack).


Re: SMA Connectors... you'll need connectors specially rated for high power. Usually they are limited to 300mW or so, high power connectors have air gaps and don't use epoxy to bond the parts together.

" 15 characters"

Thanks for the info, you all :) I wonder if you cryo these, they'd turn orange? :p Nice link Daniel. I wonder if those have had a hard life.. ?
I (if I could afford one now) wouldn't scan in the conventional sense, because it wouldn't produce good results as you guys have pointed out. The show use would be more for : expanded/mirrorball, static beam, rotating mirror-wheel for liquid sky, diff grating, part of a rainbow etc .. several options without using conventional x-y scanners .. Just looking at a 40W beam of 638-640 would be entertainment enough imho for me :cool:

Dilas is amazing. If it does have some DILAS optics in there you should have ZERO trouble getting a good beam out of it. I have several DILAS units up to 50W and use between a 400-600um fiber based on model. A simple collimator or beam expander will give you a clean, tight beam. This is why DILAS gets to charge a huge premium over other manufactures. They don't use simple fiber bundles, but precise optics to help clean up the beam. I also found the fiber output to be somewhat polarized vs say a Coherent FAP.
ADAM

Sounds good. Now I just need spare $3K and a guarantee that those "used" modules are not end-of-life ;)