basic question about "back-reflection"

[bbshamsa]

Hi all.

This seems like a very basic question considering the level of some of the threads I've seen so far, but all other sources I've tried haven't managed to help.

Anyway, my question is: can a beam that's partially reflected back into a 532nm diode cause any damage?

Say, from experimenting with non-AR coated lenses?

Also, does anyone know a good source of cheap beam expanders, lenses, prisms, mirrors etc... ?

Cheers.

[stanwax]

Its unlikely
Back refelcetion is a phenomena associated with laser diodes and though many 532nm lasers are often refered to as 'diodes' they are not. They have a diode in them but the green laser light is made by a different process so back reflection is not a problem.

Rob

[bbshamsa]

Ah, so even if the beam is partially reflected back into the module, it wouldn't pass through the two crystals between the collimating lens and 808nm diode itself?

Presumably, it boils down to (a) if the 532nm beam can reach the diode, and if so, (b) if the interaction between the 532nm light input and the 808nm light output could adversly affect the diode itself, in some way.

[stanwax]

Back reflection can happen to any diode but visible diodes tend to be operated much closer to their limit so they naturally have a tendancy to be easily damaged as only a few mW of BR will tip them over the edge and destroy the reflective surfaces within the cavity. 808 pump diodes can be damage in the same way but are not pushed as hard so have more compliance to a bit of BR. If you see a dpss cavity in operation, despite coated optics you will see that the pump chip is already bathed in green light but it may be that it is not focussed correctly so much of it does not enter the diode cavity in any case.

Rob

[buffo]

Ah, so even if the beam is partially reflected back into the module, it wouldn't pass through the two crystals between the collimating lens and 808nm diode itself?

More like it wouldn't make it through the optical coatings on those crystals, but yeah, you've got the general idea.

(a) if the 532nm beam can reach the diode,

Which it can't....

(b) if the interaction between the 532nm light input and the 808nm light output could adversly affect the diode itself, in some way.

Unlikely. Back-reflection kills diodes by overloading the face of the diode. Literally the optical power density becomes so high that the face is damaged. Now, on most red diodes you're running the thing very close to it's maximum rated power, so a little extra power reflected back at the face can be enough to cook it. On the other hand, 808nm pump diodes are normally not run anywhere close to their absolute maximum rating, so you have lots of headroom there.

Also, the pump diode in a DPSS laser normally puts out many times more power than the laser's visible output rating. So for a 100 mw laser, the pump diode might be making as much as 500 mw to a full watt. As such, even if you could get the 532 nm light back to the diode face, you're only increasing the power density by 10-20%.

Bottom line: you won't kill a DPSS laser with back-reflection. With a direct injection diode, however, it's possible - especially if you're already operating very close to the max rated output. (This is why it's advisable to dial the power down whenever you're adjusting the optics in your projector. If you accidentally send the beam back to the red diode, you want to be sure the power level is very low so you don't kill it.)

Adam

[bbshamsa]

That's great, thanks guys.

Talk about a "master class"!

Well, I have a 1.2W 532nm module with less than 0.5% residual IR, that's specced to put out 125-140mW of 532nm, constant out, (the last 9 minute+ test averaged 140mW+/- 6%) so based on what you said about the overall power density on the diode faces, I guess it would be within limits in the worst-case scenarios.

I'm intending on building my own housing for it, that can accomodate varying sizes of lenses to experiment with beam expansion. I'm focusing the design on thin-as-possible wall thicknesses and as much surface area as is practical

Just need to research for lens sources now; while I'm learning, I don't want to outlay the extortionate prices for properly coated optics. Once I get my ideal design nailed, I'll think about getting the proper kit.

Thanks again for the info.

Shammo.

[Jem]

O.K., i'll just add my twopenneth (or two cents worth )...

You may not be able to kill a DPSS laser with back reflections but you can temporarily de-stabilise it. This doesn't appear to do any harm but you may not get constant output and the beam may flicker etc.

Certain applications require that a laser beam be absolutely stable without any mode hopping etc. (holography for instance) any back reflections will potentially kill the stability of your beam.

As I said, just my twopenneth

Jem

[stanwax]

Good point Jem and I believe this phenomena could occur in most/all lasers - can it cause the beam to seemingly flash on and off as the reflected beam and incident beam interfere with each other - as I have seen with my he-ne and argons

Rob

[Jem]

Good point Jem and I believe this phenomena could occur in most/all lasers - can it cause the beam to seemingly flash on and off as the reflected beam and incident beam interfere with each other - as I have seen with my he-ne and argons

Rob

I have to be very honest here and say i've not noticed it on my HeNe or my single line argon. That's mainly because I don't use the argon for holography and i've virtually retired the HeNe (which was only 10mW, so it may not have affected that as bad). I now use the Coherent C315m for holography and that's probably one of the most stable single line DPSS lasers you can get. Direct back reflections on this absolutely kill the stability. It's easy to avoid though by just offsetting all the optics in the beam path slightly.

I do suspect you're correct though and it will affect all lasers.

Jem

[allthatwhichis]

Ahh, can you damage a gas laser with back reflection? I just got my multiline Argon last night and as I was playing with a grating I wondered about that...