US Patent 5,940,419 by XIE pretty much covers the history of internal cavity CW frequency doubling and much of the things that can go wrong. He covers most of the solutions that are used.
I have one very nice little ~150 mW DPSS "L" cavity laser on my bench right now that I detuned the KTP as part of an experiment to enhance the normally undesired, lasing of the 1084 line to get a tiny bit more of 543 which was lasing in the background. I'm sort of regretting detuning it as I will now have to align it from scratch. Along the way I saw the instability caused by 1064 and 1064.1 lines mixing on a photodiode,, and I saw the tiny but stable peak of 537 nm from 1084 self mixing with 1064 in the cavity on the spectrometer. Turns out the designer tuned the KTP length to act as a filter. I've got to find the doubling crystal's sweet spot, because I can see lots of 532 being generated for about a second and things become unstable as I move the rear mirror to try and peak it. Right now it wants to lase single longitudnal mode at just a few mW of 532, and it "giant pulses" one big flash of green every once and a while as I move the mirror.
In other words, the detuned crystal is acting as a huge loss and the way it acts changes as I move it because that changes how the electric field in the crystal acts, which causes it to heat up a different way, which changes the pattern of modes formed in the cavity, which causes it to have unstable longitudinal modes, which causes spatial hole burning, which causes even more instabilty.
Yet I'm happy because what I am seeing matches the theory, I've learned how to tell when things are getting unstable, and I scored the 1084 and 1073 nm lines along the way, at least on the spectrometer.
I have a feeling realigning this one will take a few hours more.
This is part of a project to learn how to get "other" colors from Nd:YAG
543 is pretty promising if you pump with 888 instead of 808 and can create optics that kill 1064 some how. Its a pretty color, too.
Problem is dielectric mirrors that favor 1084 and kill 1064 are very, very, very pricy to have coated. I ended up sticking a crude piece of microscope slide into the cavity to act as an etalon to favor the 1084. I saw strong flashes of 543, if you consider a few hundred microwatts of 543 as strong.
Yet Casix makes a 543, so they paid for a coating run of some really difficult to make mirrors.
Point of the matter, this aint easy some times. All I wanted was a photo with a grating showing both 532 and 543 spots on the wall.
While trying to force the issue with parts designed to prevent that from happening.
Steve
Last edited by mixedgas; 01-25-2020 at 19:35.
Qui habet Christos, habet Vitam!
I should have rented the space under my name for advertising.
When I still could have...