combining knife edging with DPSS?

[johnjack]

The 5W 808nm diodes seem to be in the $100 range, the 10W ones in the $200 range.
If one was to combine the beams of several of these with a knife edge setup then run only the final (corrected with a beam expander) beam through crystals to get 532nm green beam or other visible wavelengths, wouldn't that be a far more cost effective way to get super bright beams?
I'll admit I'm still learning about DPSS but the only difficulty I see is better temperature control required and <50% power loss. That's still crazy power from 4 such diodes.
But if it was a no brainer, everybody wouldn't be using visible diodes for their knife edge setup.
What am I missing?

 

[Gadget]

For one, crystals taking that kind of power input would need some serious active cooling.

-G

 

[NightExplorer96]

Yeah, that would get hot pretty quick. Also after you get 532nm light, how do you plan on combining the beams?

Im not exactly sure what knife edging looks like but would it be as follows?
808nm~~~~~|1064nm~~~~~|532nm~~~~~
808nm~~~~~|1064nm~~~~~|532nm~~~~~
808nm~~~~~|1064nm~~~~~|532nm~~~~~

I mean im sure this is possible if you set your mind to it, but realize there are many challenges which stand in your way.

-Alex

 

[johnjack]

Not really what I meant.

If one was to combine the beams of several of these with a knife edge setup then run only the final (corrected with a beam expander) beam through crystals to get 532nm green beam

I mean combine several 808nm beams with knife edging and beam expander together, then do the 808->1064->532 crystal magic on the final combined beam only.
And the question is not only if it's possible, but whether is makes sense for cost-saving. I don't know of a commercial DPSS module which does this, but then again I don't know the internals of most commercial modules above 1W.

(butchered an image I found, no copyright infringing intended)

 

[NightExplorer96]

Hum...

I don't think that would be very efficient both for you & your wallet. I don't know much about the DPSS process but it looks like a much more complicated process to obtain higher 532nm light then instead using a powerful 808nm diode(if they exist).

Also as said by Gadget, you would need some pretty good heat-sinking for that and you might not be able to have it on for long even then.

-Alex

 

[Alaskan]

Consider looking for a higher power C-Mount 808nm laser diode, I know of some which can put out up to 15 watts of output power with a FAC lens which can be had on ebay for 200 dollars or less from time to time from one seller. As far as how much power a give size of KTP crystal can take and the kind of cooling required, I'd be interested in knowing too.

 

[johnjack]

Consider looking for a higher power C-Mount 808nm laser diode, I know of some which can put out up to 15 watts of output power with a FAC lens which can be had on ebay for 200 dollars or less from time to time from one seller. As far as how much power a give size of KTP crystal can take and the kind of cooling required, I'd be interested in knowing too.

Yes, exactly what I have in mind. There are >10Watt 808nm C-mount diodes relatively cheap (<$200).
This is way beyond my knowledge, that's why I'm asking. This is the idea and I just want to know if it is possible and if yes, is it cost efficient enough to consider doing.

Combine say 2 of these 10W 808nm beams with knife edging, and even with 70% power loss through the conversion process you have a whooping 6Watt green beam. That's around $400 + the cost of crystals + cost of beam expander + advanced cooling.

With the NDG7475 1W green diodes the cost is $300 for 1 1W green diode only, so to get 6W it would cost around $1800 for the diodes only + cost of beam expander + heatsinking.

Definitely with both not something you can fit in your pocket, but as a module I wouldn't mind a 6W green beast. Keep in mind 532nm appears several times brighter than same power 445/450nm blue.

https://www.youtube.com/watch?v=4wzM0RLjOSE

And not like green is the only color that can be produced with crystals, right?

Aside from actually building such a thing, I'm also just curious what's inside those 10-30W commercial modules. The "pure-diode" ones I'm guessing it's just a very large knife edge setup, but with DPSS modules I don't know.

 

[Gadget]

Consider looking for a higher power C-Mount 808nm laser diode, I know of some which can put out up to 15 watts of output power with a FAC lens which can be had on ebay for 200 dollars or less from time to time from one seller. As far as how much power a give size of KTP crystal can take and the kind of cooling required, I'd be interested in knowing too.

Ask, and ye shall receive. This is just power ratings tho, no cooling, and you'll have to do some math 'cause the list is for Nd:YAG Q-switched lasers...
http://www.quantumtech.com/apps/908.pdf

And, here's one from Sams Laser FAQ....
http://www.repairfaq.org/sam/lasercds.htm#cdssups

-G

 

[crazyspaz]

It certainly can be done- I know Coherent made a multi-cmount knife edge module...for what purpose I have no Idea. I would assume it would be for DPSS, but IIRC it output through a lens, not into fiber (which you would expect if it was for a solid state system). I will see if I can find a pic for you.

 

[NightExplorer96]

And not like green is the only color that can be produced with crystals, right?

Nope, but 532nm DPSS is the most common. Other wavelengths which are produced using crystals are:

UV wavelengths/473nm/488nm/543nm/589nm/593.5nm/660nm/671nm/IR wavelengths.

-Alex

 

[johnjack]

It certainly can be done- I know Coherent made a multi-cmount knife edge module...for what purpose I have no Idea

I have no particular purpose in mind, I just want one.

Waiting for some pics/links.

 

[crazyspaz]

Can't find any, sorry. Not much to see though, if you can picture what knife edging is then that's pretty much it (just with c-mounts).

 

[Sigurthr]

Y'all are only missing one key factor; SHG and SFG crystals are polarization dependent! You'd have to align the individual beams polarizations to a plane of axis, and then orient the crystal set to that plane. This of course, greatly enhances the difficulty of such an undertaking. Also, precludes the use of further beam combining techniques (such as a PBSC) without advanced polarization rotating optics (AOM/PCAOM).

 

[H2Oxide]

I get the feeling that some people think you can cool this with heatsinks, IMHO I don't think you could passively cool such a monster, diodes or DPSS.