Welcome to Laser Pointer Forums - discuss green laser pointers, blue laser pointers, and all types of lasers

Thanks for supporting LPF!

Avery's Instagram | Open Source Cybersecurity Software by Avery | Considering selling LPF, DM if interested

Do it Yourself Complete Green DPSS with O-like Crystals

Jun 1, 2008
Hello ! a rapid thread showing an homemade DPSS system :

ORIGINAL PICTURE of the discovery are IN THIS THREAD : http://laserpointerforums.com/f52/o-like-crystal-40671-2.html

First you need to know a little about the working of a green dpss, this picture come from sam laser FAQ :

In the picture above, in middle, note the concave face of the OC (output coupler) mirror in front of KTP crystal . We will use this setup.
Site of Rp-photonics :

Vanadate crystals are naturally birefringent, which eliminates thermally induced depolarization loss in high-power lasers. Also, the laser gain is strongly polarization dependent, the highest gain is usually achieved for polarization along the c axis. The pump absorption is also strongly polarization-dependent (except at special wavelengths), which can cause problems e.g. when using a fiber-coupled pump source with drifting polarization.

List Of Material (clik on material to have link)

OC Mirror (Output Coupler)


R=50 mm, Transmission = 3%@1064nm
Size :10mm(dia)*3mm(thickness)
Coating : Concave side:HR1064/ Ht532; plane side: AR532

Nd:YVO4 and KTP
In the picture in background, a typical KTP and Nd:YVO4 crystal from a newish green pointer to show the size comparison

808 NM HIGH POWER DIODE you can buy any 808nm diode,9mm case or cmount (which are easier to made a custom heatsink) on ebay

A LM317 driver set at 1.5A, with a strong power supply (a power supply from a computer for example)

A TEC element (which you can find on ebay). The power supply will supply the high current. Mine is set a 5V with 3 or 4A. It work very well only if the heatsink on the hot face is big enough

2 cooling Fan, one will cool the heatsink with the TEC, the second the LM317, which have also a heatsink (it's getting hot very rapidly!)

First, you must build a driver which is solded. Don't take risk with non solded connection which can fried your Cmount diode.
You must then build an heatsink for the diode. Use a part of the TEC to cool the heatsink which will maintain the temperature of the diode.

Next, you can test rapidly which face of the OC mirror is the good one (both side are differently coated) and the vanadate too.

Finding the concave face of the OC :

For the side of vanadate, try your luck. I can't tell you what side is good. Don't worry with the side of KTP, both side are coated so any face will work.

PRECAUTION : BE EXTREMELY careful with the coating of the crystal and output coupler ! no water, no bad clothes to clean them (or don't clean them!) they degrade very fast in greasy environment (smoke, solder smoke..).

Another Precaution : DON'T LOOK at the HIGH powered diode even uncollimated ! i see it for second and i loose the feeling of colour of one eye ! (the left eye seen objet with normal colour, the right seen object with very few colour, it's like a big cloud of fog in your eye that you can't remove ! mine become normal after few hours, i got luck, so no mistake !!)

You can use a lens (aixiz plastic lens) in front of the laser diode to collimate the beam near the Vanadate, (not on the surface). If you want to make a direct die setup , you must decan your diode and place the vanadate just in front of the semiconductor without touching it.
I use a Cmount Diode put just in front of the vanadate crystal.

For a rapid setup, I found that if you mount the OC on a moving support is more accurate (don't manipulate the OC itself with your hand, but the support which is sticked the OC) .


To check if this work, put the KTP and vanadate just close each other, and the vanadate in front of the diode (in the picture above, the diode is a Cmount one.)
Place your OC on a mount, at 3 to 5 cm far from the face of vanadate. It's easier to find the right cavity with a far mirror.


Of course your vanadate is big, but you will only use a small portion of the surface. Try to obtain at least one photon of green, by rotating the OC up and down, right and left. Be patient. The precision need to be far below in the millimeter range ! Don't put your crystal on a fixed location, move it far or closer of the KTP.
If you have no photon of green after several minute, try the other face of vanadate.
If you have some green, you can check different position and rotation of the KTP and Vanadate to obtain the maximal output. I used a bolt with a holed screw for the vanadate, but it's not a clever setup, try to find an other mount for vanadate. The vanadate must be cooled with the other part of the tec.
Don't fix the fan to the Tec heatsink, the vibration will blow your cavity.
Once you have one photon of green, move the KTP! you will get more power. You will see some mod (TEM00 is almost rare), and some other strange thing (like double ray of 532 nm see original post or cavity ray path :


Some tips from GoeeyGus :
another tip, experiment with placing the OC closer and further away. You should get to one point where you get substantially more green. Also, I know it may be a bit hard with your setup, but if you can get your beam centered on the OC, that is optimal.

In this type of setup your vanadate and KTP really need to be in two separate mounts. The KTP doesn't need any cooling, it actually does much better when it is warm (so a cold KTP is probably killing quite a bit of power). But you really need the KTP in the center of the cavity. You will obtain much more green if it is further out from the vanadate. Also with two separate mounts you will be able to adjust the KTP to find the point in which you get the most power.

Try placing the OC and the vanadate exactly 50mm apart. Then, move the KTP until you get the most power. The reason is that the OC actually focuses the 1064nm light, meaning that the 1064nm light will focus to a pinpoint at some point in the cavity. You want the KTP to be within this pinpoint. As long as you are lasing (which you obviously are) try moving the ktp around a bit. It's position isn't super critical, and I bet if you move it forward a bit, closer to the OC, you will gain quite a bit of power.

I'm interested in how to have more power, my setup can be upgraded: I use a 808nm pump diode which have some dark line (death emmiter), maybe the mount of the vanadate could be better, maybe use a cleaned OC mirror (it goes dirty fast), be more clever to use more of the vanadate surface.....

I built a "portable" setup, it's really a good experience, I seriously doubt that we can have power in the 400 mW range as o-like.com sell, they are so well made, this experiment show us how it's difficult to obtain some 532 nm, and a TEM00 ! For the final lens (after the output coupler) , I had problem, the one that worked is from a 532 nm 50 mw tunable aixiz labby.
Honnestly, my power.... is weak ! between 5 or 10 mw, no more, no less.
But i could be ameliored im sure, I remenber one time, my hand burned hardly when it was really lasing, but I did'nt find this output again.
Some picture, from original post, and newer :






(+water fog, notice the other beam)





After some improvement with a nice optical mount :

beamshot (no fog, only time exposure)


The DSS Built is on the right (it goes invisible after), on the left a 20/30 mw one from susie. It 's less visible because it's not completely collimated : at 15 m, spot size is more than 30 cm, whereas spot of susie'sone is few cm. To collimate more, I have to put closer the lens to the OC, but I could'nt make that without destroying my cavity.


50 sec exposure, 200iso, no fog.

In conclusion : It's fascinating to find the cavity and have the generation of 532nm, we are in the pure physics, it's worth a try ! it's AWESOME !
I would like also thanks Susie which sell vanadate, KTP and OC (which are DIFFICULT to find!!) at a really good price !! but honneslty the price don't matter for this experiment

rendezvous in one year for 473 nm ! edit 2012 : have all the part but never had the time :/
Last edited:

May 4, 2009
Awesome, its always nice to see someone who has taken time to document their project with plenty of pics and and a thorough write up. :gj:
May 10, 2009
Great DIY, I enjoyed the bit on identifing the face of the OC using the reflection of a light bulb. I have worked in the field of coatings and paint for over 10 years and a light source in a reflection can tell you many things about the detail of the surface of an object. The diagram with the light bulbs is a great way to explain that, great post!


Jun 7, 2009
Wow, this is very detailed. It clears a lot of things up for a noob like me, although there will always be questions to be asked. But what are forums for anyway? :) Thanks mauswiesel!
Jun 14, 2009
Should be made a sticky of how green lasers work. Thanks I've never seen pictures like this


Sep 21, 2007
Really great drawings and pictures to explain everything!

Feb 9, 2008
Awesome job. Seeing how a green dpss laser works in action really clarifies things. That must have taken a lot of time and patience to set that up and not bump the table for enough time to take the pictures. Very cool.



Feb 9, 2009
Frickin' Awesome! :gj:

Those pics remind me of that guy that grew his own homemade crystals, and managed to get some green (But he was pumping it with a YAG laser, so he had SLIGHTLY more horsepower to work with!;))

You might want to try to put a small TEC near the KTP crystal (perhaps between the mount and that big brass-nut looking thing on the right side of the later pics?) and reverse the current, to warm the KTP crystal slightly, as I think this could increase your output power. I remember seeing a design for a homemade DPSS labby that did this, as I recall, adjustments to the KTP temp had a significant impact on output power! (I'll see if I can find that link for you).

The reason is that the OC actually focuses the 1064nm light, meaning that the 1064nm light will focus to a pinpoint at some point in the cavity. You want the KTP to be within this pinpoint. As long as you are lasing (which you obviously are) try moving the ktp around a bit. It's position isn't super critical, and I bet if you move it forward a bit, closer to the OC, you will gain quite a bit of power.

This makes sense, as if I recall correctly, conversion efficiency of the doubling crystal goes up dramatically with higher power densities.

Of course, trying to focus an invisible IR beam can be a bit of a problem! LOL Here's an idea on that - perhaps using the right kind of digicam or similar that can "see" IR, to help achieve optimum focus of the 808 beam in the center of the KTP?

(I also seem to remember reading something about using a lens to "narrow" the beam width, b4 it enters the KTP?)

As I recall, on the guy using the YAG, he was also able to dramatically increase green output by Q-switching his YAG pump laser. The idea was to increase the instantaneous energy density inside of the KTP crystal. This actually worked quite well - the increase in green output power when he did this was VERY noticable in his pics!

I don't know that it's possible to Q-switch with an end-pumped DPSS design like you are using - perhaps one of the experts in the DPSS arena here can chime-in to confirm that?

However, something that perhaps might work instead would be to "pulse" your IR diode? Most LD's are capable of higher pulse power than CW - and the increased power density inside the KTP might tend to further magnify the increase in output!

(In fact, wasn't this was one of the "tricks" that some of the first DPSS pens used???)

Anyways... :topic:

+reps for innovation!
Nov 7, 2008
^Due to the nature of a q-switch, they won't work with a end-pumped design. You can, like you said, modulate the system and get similar results, but often at the cost of pump lifetime if you increase the current..

The beam waist from the pump diode should be just inside the YAG/YVO4 but the KTP is not so critical. It does help efficiency to have a second waist (in the 1064nm output from the lasing crystal) inside the KTP, but this requires extra optics or a different cavity design. For simplicity it is no problem to take the output directly from the lasing crystal and feed it straight into the KTP.

Also, you are correct that KTP likes to be warm. In smaller systems (a couple hundred milliwatts or less) the efficiency gains from temp. tuning the KTP are negligible and not worth the extra work. With LBO it's a different story altogether. LBO has a very narrow temperature range in which it is efficient at doubling. Outside this range, you'll be lucky to get any doubled output. Temperature tuning is a must on these systems.
Jun 1, 2008
Apr 29, 2007
VERY well done, this is what I like to see people learn to do. You learn so much more by building a laser on your own then just buying one.

Good Job! I propose that this is stickied.
Jun 1, 2008
im losing a good amount of 532 nm (maybe one third) when i put some final lens to collimate the beam. Honnestly, compared to some commercial one, i would like to think that the output range is between 30...50 mW. I need to find some coated lens to make a final build.