Building DPSS from scratch. Trouble getting YV04 to produce any spontaneous emission

[brianpe]

I am not yet temperature controlling the KTP -- I didn't forget to plug it in I just haven't done it yet. My understanding (and I've never done this, so my "understanding" comes from the internet so take that for what it's worth) is that the angle is very temperature dependent, but you can pick any temperature as long as it's stable and that decides the angle. I was hoping to fix it at about 25ºc using a TEC. That's not connected yet -- I am waiting on some new machine work to arrive next week that will hold the crystal better before I glue in the thermistor. I also have the end of the crystal holder bored out to hold a small cartridge heater if I end up having to do non-critical phase matching at higher temperatures or if I can't get good enough control from the TEC.

My thought was that I'd have a few moments of good conversion efficiency that I'd keep having to "hunt" for by tuning the angle as the crystal heated up. But I'm not seeing any hints of decent efficiency. There is definitely a sweet spot where it converts, but it's nowhere near as precise as the literature make out. Maybe my angle tuning is just too coarse. When I have the TEC running my intention is to get the angle mostly right by hand and then fine tune the temperature for the angle.

Thanks for the insight about the TEC placement. I am a little worried about this too and not sure the thermal contact is good enough for decent control. I didn't want to go with water cooling and wanted to use the mass of the base plate as a sink for the TEC but there isn't great thermal contact between the mount and the cylinder that a holds the KTP. I may need to revisit this design.

The crystal is AR coated on each end for 1064 and 532. I have not tried it outside the cavity.

[kecked]

I am not yet temperature controlling the KTP -- I didn't forget to plug it in I just haven't done it yet. My understanding (and I've never done this, so my "understanding" comes from the internet so take that for what it's worth) is that the angle is very temperature dependent, but you can pick any temperature as long as it's stable and that decides the angle. I was hoping to fix it at about 25ºc using a TEC. That's not connected yet -- I am waiting on some new machine work to arrive next week that will hold the crystal better before I glue in the thermistor. I also have the end of the crystal holder bored out to hold a small cartridge heater if I end up having to do non-critical phase matching at higher temperatures or if I can't get good enough control from the TEC.

My thought was that I'd have a few moments of good conversion efficiency that I'd keep having to "hunt" for by tuning the angle as the crystal heated up. But I'm not seeing any hints of decent efficiency. There is definitely a sweet spot where it converts, but it's nowhere near as precise as the literature make out. Maybe my angle tuning is just too coarse. When I have the TEC running my intention is to get the angle mostly right by hand and then fine tune the temperature for the angle.

Thanks for the insight about the TEC placement. I am a little worried about this too and not sure the thermal contact is good enough for decent control. I didn't want to go with water cooling and wanted to use the mass of the base plate as a sink for the TEC but there isn't great thermal contact between the mount and the cylinder that a holds the KTP. I may need to revisit this design.

The crystal is AR coated on each end for 1064 and 532. I have not tried it outside the cavity.

I was kidding sorry if I offended.

the idea is to raise the temp above room temp so as to keep it stable. It’s easier to heat than cool. I had a weird crystal that only worked one way. Try flipping it over.

[brianpe]

I was kidding sorry if I offended.

the idea is to raise the temp above room temp so as to keep it stable. It’s easier to heat than cool. I had a weird crystal that only worked one way. Try flipping it over.

Certainly not offended. I will try that. Will also try bringing the temp up a bit. New mount arrives tomorrow

[brianpe]

I've got the KTP temperature pretty well regulated. I set it to 30ºC and coarse adjusted the angle to maximize green, then I tweaked the temperature to maximize output. Wow, two degrees is the difference between max output and almost no output. The TEC layout seems to be working OK, at least for these lower powers. I'm regulating to +- .1 degrees or so.

I've been tweaking the cavity design to get a good waist into the KTP. I should be at about 80 microns. I found a research paper that had SHG conversion efficiency measurements for a crystal that's very close to what I'm using. The way I've been thinking about this is, assuming I want 10 - 20% conversion rate (to act as an output coupler), there will be an optimal pump power and beam waist into the KTP I need. I should choose the final output power I want, compute the intracavity power for the optimum output coupling, and design the cavity to focus to a waist that gives the needed power density in the KTP to hit that conversion efficiency. Does that sound like the right plan?

One thing I was thinking about is my output coupler transmissivity is now variable with pump power, and this makes linear control of the laser output hard. Go below the optimum power, conversion efficiency drops way off and output drops with the square of the pump power. Pump higher than the design goal and all you get is more waste heat and little added output because the transmissivity increases (decreasing efficiency). Am I thinking about that the right way?

At 80 microns the power density is not enough to get decent conversion, but I did try replacing the 20% OC with HR 1064 HT 532 and I carefully monitored temperature of the vanadate as I slowly ramped up the pump current. I can get OK power out of this - at 22 amps pump current (the max I've been comfortable trying) I see about 250mw of pure green. Compare this to IR however (where I get a little over 2 watts) and the efficiency is way off. When I plot power input to output for both green and IR the slope of the graph looks the same -- if I was hitting a tighter waist I would expect the green to have a steeper slope.

I'm also using a PCX lens to focus into the KTP. I would rather do this with mirrors for less loss, but I haven't found a mirror combination that gets the the waist I need. The lens I'm using has a N-BK7 substrate, which adds about 8% loss to the cavity. Thor Labs has lenses made from MgF2 that have a transmissivity of 95% so I'll go for one of those once I get this dialed in. Still like to reclaim that 5% though so I haven't totally given up on a mirror solution.

Beam Shot

[kecked]

I've got the KTP temperature pretty well regulated. I set it to 30ºC and coarse adjusted the angle to maximize green, then I tweaked the temperature to maximize output. Wow, two degrees is the difference between max output and almost no output. The TEC layout seems to be working OK, at least for these lower powers. I'm regulating to +- .1 degrees or so.

I've been tweaking the cavity design to get a good waist into the KTP. I should be at about 80 microns. I found a research paper that had SHG conversion efficiency measurements for a crystal that's very close to what I'm using. The way I've been thinking about this is, assuming I want 10 - 20% conversion rate (to act as an output coupler), there will be an optimal pump power and beam waist into the KTP I need. I should choose the final output power I want, compute the intracavity power for the optimum output coupling, and design the cavity to focus to a waist that gives the needed power density in the KTP to hit that conversion efficiency. Does that sound like the right plan?

One thing I was thinking about is my output coupler transmissivity is now variable with pump power, and this makes linear control of the laser output hard. Go below the optimum power, conversion efficiency drops way off and output drops with the square of the pump power. Pump higher than the design goal and all you get is more waste heat and little added output because the transmissivity increases (decreasing efficiency). Am I thinking about that the right way?

At 80 microns the power density is not enough to get decent conversion, but I did try replacing the 20% OC with HR 1064 HT 532 and I carefully monitored temperature of the vanadate as I slowly ramped up the pump current. I can get OK power out of this - at 22 amps pump current (the max I've been comfortable trying) I see about 250mw of pure green. Compare this to IR however (where I get a little over 2 watts) and the efficiency is way off. When I plot power input to output for both green and IR the slope of the graph looks the same -- if I was hitting a tighter waist I would expect the green to have a steeper slope.

I'm also using a PCX lens to focus into the KTP. I would rather do this with mirrors for less loss, but I haven't found a mirror combination that gets the the waist I need. The lens I'm using has a N-BK7 substrate, which adds about 8% loss to the cavity. Thor Labs has lenses made from MgF2 that have a transmissivity of 95% so I'll go for one of those once I get this dialed in. Still like to reclaim that 5% though so I haven't totally given up on a mirror solution.

Beam Shot

You should see at least 20% conversion if you get it right. Don’t forget if you put too much energy to fast into the crystal it will change the focus and Kerr you as the refractive index changes so there are limits. What does the divergence look like? Is the beam good quality? If it is I’d be looking for some losses in optics. 0.1C is good for temp control.

Next step. Mode lock it and you’ll see much higher peak power. I think my verde runs 80mhz. Looking good! Serious balls doing optics in a garage! Careful cleaning as you’ll have nasty hard small particles to deal with.

22 amps for 250mw is really low. I have a pointer that does 100mw on two aa batteries for a good hour. I would figure you’d see some like couple watts.

[brianpe]

Divergence is not great. I haven't measured it, but the beam is about 3/8" five feet across the garage. Starting beam diameter inside the KTP is 80um. But, I don't have any collimating optics or a beam expander after the resonator yet. The output mirror is concave and the output is also passing through that focusing lens, so I expect a bit of a mess coming out right now. I still need to figure out how to do the right collimation. I think I treat the middle of the KTP as a point source and use a lens with a EFL back to the KTP. I've been more focused in what's going on inside the cavity.

Speaking of which, for giggles I tried removing the focusing lens to see what happens. This puts the beam waist in the KTP at about 300um, which is pretty huge. But oddly enough, this greatly improved my output. I can hit close to half a watt now. I guess 8% loss in the cavity matters more than I thought. I have been thinking that a larger beam waist / lower energy density in the KTP is just causing a higher circulating power. I don't want that power too high though.

When lasing it mostly looks like TEM00 but when I watch the beam expanded on the wall I can see it mode hop a bit. I read a little about mode locking tonight; I'll have to think more about how to do it.

The garage...well, you use what you have. I keep everything covered when I'm not working on it but dust / dirt is always a risk. I do check optics before starting but I should be more rigorous here.

[kecked]

Interesting. thoughts. You are being Kerr, the lens is messing up the phase, you are not using enough of the crystal.

the angle of the cone is too deep. Try the lens and rack it in and out of focus. If the output goes up your getting Kerr.

take a second lens and build a telescope and expand the ir beam to make it less divergent. Put a pinhole in the middle to make a spatial filter. While the beam will be slightly larger at aperture, at distance this should improve the end result and it might give you more power. The telescope should expand the beam but try it both ways. The spatial will clean up the wings so if there is any weird stuff happening from refraction and interference this should trim it out. The idea is good beam in better beam out and if you are Kerr this will spread out the beam and focus n more of the crystal.

steve I get that right?

[brianpe]

I did some modeling on this in my cavity software:


Here I'm using a 2x beam expander. The dip is the expander focal point where the pinhole goes. Small pinhole -- software says that's 24um. KTP waist is a big 650um but the divergence looks great. I also tried flipping this in software and the KTP waist was about 160um, so I can easily try both ways. I have the parts here to approximate this with a lens tube and 50/38mm lenses. The pinhole will take new parts and some thinking -- I doubt my setup is center-accurate enough for such a small aperture so I'll likely need pinhole I can size down and translate x-y.

But good tips and things to try!

[kecked]

Try burning the pinhole in place. I do this all the time. I use thin plastic. Just let the IR do the work. Couple pulses is all it takes and the hole exactly matches the power core of the beam. The trick is finding a plastic that doesn’t shrink. Capton tape or nylon works but I’m not being very picky. It’s just to clean up diodes.

[brianpe]

Try burning the pinhole in place. I do this all the time. I use thin plastic. Just let the IR do the work. Couple pulses is all it takes and the hole exactly matches the power core of the beam. The trick is finding a plastic that doesn’t shrink. Capton tape or nylon works but I’m not being very picky. It’s just to clean up diodes.

I’m running this CW, not Q-switched. Will I still see that kind of build up? Intracavity seems really fragile for me - anything in there seems to kill the beam.