Trying to build an insane 532 nm laser

[Laserbuilder]

I've got a couple of relatively high power Nd:YAG lasers, so I had a regular wish to frequency double one of them.

My expectations were like: "OK, I'll take a 125W pumping chamber, insert it between two HR mirrors, place a frequency doubling crystal between the chamber and one of the mirrors, blast 50 amperes in the lamp and get 50W of green" LOL. All my expectations epically crashed and shattered about the cruel reality.

So, I had bought a couple of SHG crystals including a NOS 6*6*15 mm LiNbO3 crystal which was considered for 532 nm generation and a shard of KTP with the size of 3*3*5 mm from a DPSS laser. I also have a BaNaNb crystal in its fixture with a heating element. So, I started my experiments. At first, I put together a YAG laser with the 20W pumping chamber and tried to focus the beam into the KTP with a short focus lens. The result was quite disappointing -- there was very little amount of green, possibly 5 mW. Out of 20W! I tried to heat the krystal to 50-60 degrees C, but it made everything even worse.

My second experiment was placing the LiNbO3 crystal in the laser cavity. Both mirrors were HR for 1064 nm. This was a little bit better, at optimal temperature the green beam was brighter, but nothing really interesting. Next, I tried to pump the BaNaNb crystal with the same pumping chamber inside the laser cavity -- the result was bad too, even at optimal crystal temperature. With no hope I replaced the K301A chamber with the 125W K301V one. The BaNaNb crytal totally refused to work with it -- it suppressed lasing at all. The LiNbO3 one gave the best result of about 50 mW of green at its peak. Just like the TYLaser303 chinese pointer(((((

Where could I have mistaken? I've seen the Laserscope KTP532 machine that can shoot out 40W of green. It has a Z-fold cavity, and the laser from which I took the BaNaNb crystal had a Z cavity too. At the other hand, DPSS lasers very often work with a linear cavity with all elements placed very tightly inside.

Are L- or Z-folded cavities really that neccesary? Why does it not work in a linear cavity with flat HR mirrors?

Now I'm looking for a driver for an acoustooptic Q-switch, this is the only thing I didn't try.

I'll post some photos a bit later.

 

[Vision]

Looking forward to this. I gotta get into some DPSS and qswitching too.

 

[diachi]

I've got a couple of relatively high power Nd:YAG lasers, so I had a regular wish to frequency double one of them.

My expectations were like: "OK, I'll take a 125W pumping chamber, insert it between two HR mirrors, place a frequency doubling crystal between the chamber and one of the mirrors, blast 50 amperes in the lamp and get 50W of green" LOL. All my expectations epically crashed and shattered about the cruel reality.

So, I had bought a couple of SHG crystals including a NOS 6*6*15 mm LiNbO3 crystal which was considered for 532 nm generation and a shard of KTP with the size of 3*3*5 mm from a DPSS laser. I also have a BaNaNb crystal in its fixture with a heating element. So, I started my experiments. At first, I put together a YAG laser with the 20W pumping chamber and tried to focus the beam into the KTP with a short focus lens. The result was quite disappointing -- there was very little amount of green, possibly 5 mW. Out of 20W! I tried to heat the krystal to 50-60 degrees C, but it made everything even worse.

My second experiment was placing the LiNbO3 crystal in the laser cavity. Both mirrors were HR for 1064 nm. This was a little bit better, at optimal temperature the green beam was brighter, but nothing really interesting. Next, I tried to pump the BaNaNb crystal with the same pumping chamber inside the laser cavity -- the result was bad too, even at optimal crystal temperature. With no hope I replaced the K301A chamber with the 125W K301V one. The BaNaNb crytal totally refused to work with it -- it suppressed lasing at all. The LiNbO3 one gave the best result of about 50 mW of green at its peak. Just like the TYLaser303 chinese pointer(((((

Where could I have mistaken? I've seen the Laserscope KTP532 machine that can shoot out 40W of green. It has a Z-fold cavity, and the laser from which I took the BaNaNb crystal had a Z cavity too. At the other hand, DPSS lasers very often work with a linear cavity with all elements placed very tightly inside.

Are L- or Z-folded cavities really that neccesary? Why does it not work in a linear cavity with flat HR mirrors?

Now I'm looking for a driver for an acoustooptic Q-switch, this is the only thing I didn't try.

I'll post some photos a bit later.

A Z-fold or L-fold cavity lets you focus the intracavity beam into the middle of your frequency doubler crystal. The small beam waist and resultant higher power density results in considerably better efficiency.

Is your laser polarized? Is your doubler phase matched? More peak power, such as achieved with a Q-switch will make the process more efficient, but not if everything else isn't right. The Laserscopes you're referring to can only achieve that 40W when Q-switched and properly maintained/adjusted. Without the Q-switch the output is closer to 10W 532nm.

As you've discovered, simply throwing a KTP into the cavity and expecting 20W of green out isn't quite how it works.

 

[Laserbuilder]

Is your laser polarized?

Probably not. As far as I know, there are no conditions in the flat mirror cavity for the beam to get certain polarization.

Is your doubler phase matched?

All best results I got with the phase matched crystal at the optimal temperature which was quire hard to maintain without a proper thermocontroller. When it is properly aligned and at optimal temperature -- the green spot at the wall is more or less round and at maximum brightness. When it is nor phase matched -- the green spot is weak and surrounded by a ring of green light. Ususally, when the crystal is not phase matched the beam looks like a weak green spot in the center of a ring, but my LiNbO3 crystal didn't show this, the spot was not in the center. Also at optimal conditions the green beam from it was not perfectly round. Maybe this crystal is cut for another wavelength? There was no word about it in the datasheet for the crystal.

The first two photos show my very first experiment with a KTP ourside from the cavity, with the beam focused in it with a lens. The next four -- a LiNbO3 crystal is placed inside the cavity, and the spot when the crystal is phase matched (bright) and not (weak with a ring around).

 

[Alaskan]

CVI Yag Laser 1064nm Polarizer | eBay

The above link: Built for far lower power YAG's than you are using. I see them on ebay from time to time, this one is cheap, you could experiment at lower power using this, if it isn't too small. I believe this might have been built for a eye laser.

These are cheap too, but the seller is not sure what wavelength, my guess is 1064 nm:

http://www.ebay.com/itm/set-of-2-Ellex-Laser-Systems-Polarising-optic-plates-13mm-/272538212256


This is more like what you need, but I already bought them:

http://www.ebay.com/itm/361864815442

 

[Laserbuilder]

Does The plate from the first link have to be placed under Brewster angle?

 

[Alaskan]

If you want the beam to stay straight I believe so, every YAG cavity I have seen with a polarizer in it prior to an intracavity SHG crystal has them at an angle like this.

 

[diachi]

Probably not. As far as I know, there are no conditions in the flat mirror cavity for the beam to get certain polarization.



All best results I got with the phase matched crystal at the optimal temperature which was quire hard to maintain without a proper thermocontroller. When it is properly aligned and at optimal temperature -- the green spot at the wall is more or less round and at maximum brightness. When it is nor phase matched -- the green spot is weak and surrounded by a ring of green light. Ususally, when the crystal is not phase matched the beam looks like a weak green spot in the center of a ring, but my LiNbO3 crystal didn't show this, the spot was not in the center. Also at optimal conditions the green beam from it was not perfectly round. Maybe this crystal is cut for another wavelength? There was no word about it in the datasheet for the crystal.

The first two photos show my very first experiment with a KTP ourside from the cavity, with the beam focused in it with a lens. The next four -- a LiNbO3 crystal is placed inside the cavity, and the spot when the crystal is phase matched (bright) and not (weak with a ring around).

Have you verified that the YAG is actually aligned and producing the expected intracavity power? Did you realign the cavity after placing the doubler inside of the cavity?

Did you try the KTP inside of the resonator? Saying that ... it's a little small, may have a hard time fitting the intracavity beam without a folded resonator or at the least a hemispherical linear resonator.

You're using flat mirrors right? May have better luck with a hemispherical configuration, with doubler placed near the beam waist.

Both HR@1064nm but what are they at 532nm?

A non-round output beam could suggest misalignment or astigmatism/aberrations inside of the cavity, although with that resonator I wouldn't expect a particularly nice beam anyway...

Throwing a Brewster plate polarizer in there may result in improvements due to the polarization sensitive nature of KTP/LiNBO3.

Proper mounting of your doubler with good temperature control and pan/tilt alignment may produce better results too.

Any idea of the coatings on your KTP/LiNBO3? Likely for 1064/532 on the KTP, but could be for something else on the LiNBO3.

May be an idea to start with something a little smaller if possible, with a shorter cavity, diode pump and vanadate instead of YAG.

If you want the beam to stay straight I believe so, every YAG cavity I have seen with a polarizer in it prior to an intracavity SHG crystal has them at an angle like this.

Yeah, if that's supposed to be a Brewster plate and not a waveplate then it'll need to be at Brewsters angle, which will of course result in some refraction.

 

[Laserbuilder]

Have you verified that the YAG is actually aligned and producing the expected intracavity power?

Yes, I did. The OC mirror of my laser is hardly mounted, the HR and the YAG are adjustable. At first, I aligned the cavity to get maximum output power. Then I changed the OC on a second HR and verified that the laser worked. After that I inserted the doubling crystal in the cavity and made adjustments of its position. then I heated the crystal. When adjusting it I could adjust the YAG and the adjustable HR to get the brightest output.

Did you try the KTP inside of the resonator?

No, I didn't. I don't have an appropriate mount for it and it is very hard to handle it because the crystal is very small, only 3*3 mm aperture, and the YAG diameter could be at least 4 mm in K301A or 6.3 mm in K301V chamber.

Both HR@1064nm but what are they at 532nm?

I can agree that they probably are not intended for use in frequency doubled lasers, because they were taken out from a non-doubled 1064 nm laser.

Any idea of the coatings on your KTP/LiNBO3? Likely for 1064/532 on the KTP, but could be for something else on the LiNBO3

Honestly, I don't really know. the LiNbO3 crystal seems not to have any coatings at all, more over it heats up itself in the multiwatt IR beam, it looks like it has noticeable absorbtion of 1064nm wavelength. On the contrary, the BaNaNb crystal with the proper mount has an AR coating possibly for 1064 nm and it doesn't heat up. I think, I'll restart my experiments again with this particular crystal, and with a Brewster plate in the cavity, as I know for sure that it is specially for 532 nm generation. Here are some photos of my final experiments, both with the BaNaNb and LiNbO3.

BaNaNb crystal worked with the 20W YAG chamber and it generated a nice round beam, but very weak one. LiNbO3 was much brighter, but the beam was much worse, the spot shape can be seen at the pictures.

By the way, what is the power of Laserscope's pumping chamber at 1064 nm? 60W? 100w?

 

[Alaskan]

Diachi is the guy to answer these questions, almost don't even want to stick my nose into this thread because of that, but I wanted to say that these crystals are not very efficient unless you use a polarized input, that is why you need to rotate the crystal to the correct angle, because the input is supposed to be fully polarized. Not having a polarized input you are losing 50 percent of its ability to SHG right there. Also, unless the beam is focused into a small enough point with the waist of the beam inside the crystal, it won't be very efficient. In addition to that, the temperature of the KTP should be very closely regulated to its "sweet spot", or it won't be very efficient. KTP crystals are more efficient than many other crystals, but the problem with them is they are much easier to damage with high power, but they become far more efficient with Q switched pulses from a YAG. I have read accounts of others who have done exactly what you are working on and until they Q switched the output, the crystal wasn't very efficient when outside of the cavity. For non-pulsed, intracavity is a must.

 

[diachi]

Diachi is the guy to answer these questions, almost don't even want to stick my nose into this thread because of that, but I wanted to say that these crystals are not very efficient unless you use a polarized input, that is why you need to rotate the crystal to the correct angle, because the input is supposed to be fully polarized. Not having a polarized input you are losing 50 percent of its ability to SHG right there. Also, unless the beam is focused into a small enough point with the waist of the beam inside the crystal, it won't be very efficient. In addition to that, the temperature of the KTP should be very closely regulated to its "sweet spot", or it won't be very efficient. KTP crystals are more efficient than many other crystals, but the problem with them is they are much easier to damage with high power, but they become far more efficient with Q switched pulses from a YAG. I have read accounts of others who have done exactly what you are working on and until they Q switched the output, the crystal wasn't very efficient when outside of the cavity. For non-pulsed, intracavity is a must.

Yep, you want the polarizer in there for sure. Or get a YAG rod with brewster cut ends. The beam is going to be randomly polarized without, so a whole bunch of power is going to be at the wrong polarization for efficient SHG.

Higher powers being more efficient apples to other non-linear optics too, not just KTP. Q-switch will result in efficiency gains either way, as will a cavity designed to have the beam waist inside of the doubler.

To answer your question Laserbuilder, IIRC the max output you can get with a Laserscope 800 (properly adjusted and maintained that is!) is 100W@1064nm Q-switched. 40W@532nm Q-switched is typically the maximum you'll see with these (again, properly adjusted and maintained). A small few people have managed >50W@532nm, but that's with a sort of custom built version designed for show use.

This sort of high power with 532nm is a whole different ball game.

 

[ultimatekaiser]

KTP isn't the greatest for high power stuff, it gets damaged easily. usually KDP or LBO, or even niobates are used for wide beam/high power 532s. They need to be actively held at their proper temps with good thermal conductivity, and needs to be for the correct wavelength. they are all cut along their optical axis differently for different purposes. Also, not all crystals are also not made to be in the cavity at a perfectly parallel angle-many are, but not all.

A intracavity polarizer is a must, and needs to be both aligned, and again, for the right wavelength. polarization rotates as it makes passes through both the crystals and reflecting off surfaces. it needs to be set up so that losses are controlled properly.

L folds or Z folds are far better for this, and usually a very high speed repeating Q switch and or a pockels cell are needed to get good output. also make sure that you minimize losses through lenses and mirrors. lenses in particular can ruin a good cavity. usually mirrors are of curvatures and spacings used to control the beam waist. don't get stuff that is too highly reflective/ not reflective enough or you get poor results because the light can't oscillate properly. you could even burn your optics. you can literally blow coatings off optics that aren't perfectly aligned.

Not a small or cheap project and definitely not an easy one. lots of complex physics at work here.

 

[diachi]

KTP isn't the greatest for high power stuff, it gets damaged easily. usually KDP or LBO, or even niobates are used for wide beam/high power 532s. They need to be actively held at their proper temps with good thermal conductivity, and needs to be for the correct wavelength. they are all cut along their optical axis differently for different purposes. Also, not all crystals are also not made to be in the cavity at a perfectly parallel angle-many are, but not all.

A intracavity polarizer is a must, and needs to be both aligned, and again, for the right wavelength. polarization rotates as it makes passes through both the crystals and reflecting off surfaces. it needs to be set up so that losses are controlled properly.

L folds or Z folds are far better for this, and usually a very high speed repeating Q switch and or a pockels cell are needed to get good output. also make sure that you minimize losses through lenses and mirrors. lenses in particular can ruin a good cavity. usually mirrors are of curvatures and spacings used to control the beam waist. don't get stuff that is too highly reflective/ not reflective enough or you get poor results because the light can't oscillate properly. you could even burn your optics. you can literally blow coatings off optics that aren't perfectly aligned.

Not a small or cheap project and definitely not an easy one. lots of complex physics at work here.

Was hoping you'd chime in! Thanks for the confirmation on the points made previously! Kaiser makes a good point about blasting coatings off of optics... Procedure for aligning these sort of things is usually lamp current turned down, Q-switch off, that reduces the risk of cooking your optics. If you throw a Q-switch in there make sure you have first pulse suppression, the first pulse from the Q-switch can have significantly more power than under normal operating conditions, resulting in various optics being toasted.

Normally I'd be worried about most people even attempting this sort of thing, but Laserbuilder clearly has experience with these sorts of systems and powers, so I don't have the same concerns I'd normally have.

Laserbuilder, keep us updated on your progress. :beer:

 

[Alaskan]

Online advice from Sam: https://www.repairfaq.org/sam/lasercps.htm

Converting a Lamp Pumped IR YAG Laser to Green?

Multiwatt Halogen, arc lamp, or flashlamp pumped solid state lasers are readily available at reasonable prices from laser surplus outfits, eBay, and elsewhere. What about adding a KTP (or other SHG crystal) and getting them to output nice 532 nm green? (Similar comments apply to higher harmonics but the difficulty level increases significantly.) While this would appear at first to be an easy project, the realities may be quite different. There are a number of issues all relating to peak power density available and shape of the beam inside the crystal. Maximum harmonic power will be generated when a narrow polarized collimated beam is properly phase matched to the KTP (Type-II critical phase matching). If the original laser isn't polarized (e.g., Nd:YAG without Brewster angled rod ends), adding a polarizing element will significantly increase performance. Since harmonic generation is a non-linear process, it is desirable to maximize power density so long as it is safely less than the damage threshold of the KTP.

Lamp pumped SS lasers often use flat-flat resonators to prevent hot spots inside the YAG rod. The intra and output beams both have a large diameter (almost equal to the rod diameter) and are massively multi (transverse) mode. Just inserting a KTP crystal into the beam (inside or outside the cavity) without taking into these factors and peak power density of the beam will likely result in either poor performance or damage to the KTP. With the multimode beam, it isn't sufficient to just consider the average power density in the beam cross-section - peak power in the hot spots due to the sub-beams must be taken into account.

CW SS lasers: External doubling will probably be very inefficient even with careful beam shaping due to the limited maximum power density available. Intracavity doubling may be possible but will probably require additional optics to achieve a narrow beam waist inside the KTP.
Non-Q-switched pulsed SS lasers: Intracavity or external doubling may be possible depending on peak power and beam structure. It may just be a matter of inserting a KTP crystal in the appropriate location.

Q-switched pulsed or quasi-CW SS lasers: Due to the extremely high peak power of Q-switched laser pulses, external doubling will probably be quite efficient but selection of the proper beam waist is most critical to avoiding destruction of the KTP.
In short, this may be possible but don't expect miracles, at least not unless the conversion is carefully designed and executed.

----------------------------------

Edit: Note he references Type II SHG crystals as being critical for polarized phase matching.

More on temperature: https://www.rp-photonics.com/phase_matching.html

 

[Laserbuilder]

Great thanks for your comprehensive answers. :thanks: After finishing my current CVL project I'll probably restart this one. +rep.

 

[diachi]

Online advice from Sam: https://www.repairfaq.org/sam/lasercps.htm

Converting a Lamp Pumped IR YAG Laser to Green?

Multiwatt Halogen, arc lamp, or flashlamp pumped solid state lasers are readily available at reasonable prices from laser surplus outfits, eBay, and elsewhere. What about adding a KTP (or other SHG crystal) and getting them to output nice 532 nm green? (Similar comments apply to higher harmonics but the difficulty level increases significantly.) While this would appear at first to be an easy project, the realities may be quite different. There are a number of issues all relating to peak power density available and shape of the beam inside the crystal. Maximum harmonic power will be generated when a narrow polarized collimated beam is properly phase matched to the KTP (Type-II critical phase matching). If the original laser isn't polarized (e.g., Nd:YAG without Brewster angled rod ends), adding a polarizing element will significantly increase performance. Since harmonic generation is a non-linear process, it is desirable to maximize power density so long as it is safely less than the damage threshold of the KTP.

Lamp pumped SS lasers often use flat-flat resonators to prevent hot spots inside the YAG rod. The intra and output beams both have a large diameter (almost equal to the rod diameter) and are massively multi (transverse) mode. Just inserting a KTP crystal into the beam (inside or outside the cavity) without taking into these factors and peak power density of the beam will likely result in either poor performance or damage to the KTP. With the multimode beam, it isn't sufficient to just consider the average power density in the beam cross-section - peak power in the hot spots due to the sub-beams must be taken into account.

CW SS lasers: External doubling will probably be very inefficient even with careful beam shaping due to the limited maximum power density available. Intracavity doubling may be possible but will probably require additional optics to achieve a narrow beam waist inside the KTP.
Non-Q-switched pulsed SS lasers: Intracavity or external doubling may be possible depending on peak power and beam structure. It may just be a matter of inserting a KTP crystal in the appropriate location.

Q-switched pulsed or quasi-CW SS lasers: Due to the extremely high peak power of Q-switched laser pulses, external doubling will probably be quite efficient but selection of the proper beam waist is most critical to avoiding destruction of the KTP.
In short, this may be possible but don't expect miracles, at least not unless the conversion is carefully designed and executed.

----------------------------------

Edit: Note he references Type II SHG crystals as being critical for polarized phase matching.

More on temperature: https://www.rp-photonics.com/phase_matching.html

Great information! :beer:

The FAQ is always usefull. A large chunk of my knowledge comes from the FAQ. Sam and Steve are the people to talk to when you're stuck and can't find the answers. There aren't many (if any) laser related questions they won't be able to answer. Of course, they are often very busy and as a result don't always reply quickly. But their knowledge is invaluable so it's always worth the wait.