Page 1 of 2 12 LastLast
Results 1 to 10 of 20

Thread: Mounting KTP and Nd:YVO4 crystals -- some questions

  1. #1
    Join Date
    May 2017
    Posts
    22

    Default Mounting KTP and Nd:YVO4 crystals -- some questions

    Hi!

    I am planning on building a green DPSS laser and have been thinking about crystal mounting over the past few weeks.

    The only information I've read with regards to mounting vanadate comes from Sam's Laser FAQ:

    The following suggestion would provide good thermal contact between a vanadate crystal and copper plates. At least one of these (the plate in contact with the pump-side of the vanadate is most important) could be attached to a heatsink or TEC.(From: Christoph Bollig (laserpower@gmx.net).)

    1. Take two copper plates, 3 to 5 mm thick. Screw them tightly together.
    2. Drill a 1 mm (0.04 inch) hole threw them, best if countersunk on the pump side to avoid clipping the pump beam.
    3. Take apart, put indium foil on both sides, put paper in between and screw together again (not too tight).
    4. Use the back of the 1 mm drill bit to push through. This will make a hole in the indium foil.
    5. Separate again. In theory, one piece of indium foil should stick to each metal piece. Take the paper off.
    6. Put the crystal in between and screw together again. Make sure, it is not skewed. Don't overtighten - vanadate isn't very strong!
    If I understand correctly, this mounting setup has copper plates lined with indium foil making contact with the front face (laser entry) and back face (laser exit), with a small aperature for the beam.

    This makes sense, as most of the heat produced comes from the first atoms struck by the 808nm pump beam.

    However, I have major reservations when it comes to touching indium foil to the delicate optical coating of the crystal.

    Anyway, the questions:

    - Are the dichroic optical coatings resilient enough to survive indium foil being pressed against them?

    - Is this a one shot deal? i.e., can the crystal be removed from its enclosure and still be in usable condition?

    - How necessary is thermal contact with the front face if I'm using a ~2 watt pump? Is thermal lensing inevitable otherwise?

    I've also got a question about mounting the KTP crystal:

    - Since there is a correlation between KTP temperature and its theta, can I mount the crystal at a fixed 45 degree angle and just use a TEC to keep it at the whatever temperature produces a theta of 45 degrees in the KTP?

    - I mocked up a possible KTP mount:

    Click image for larger version. 

Name:	wCOnIMr.png 
Views:	8 
Size:	202.6 KB 
ID:	52065

    The inner cylinder would be machined a little bit too small to allow the application of thermal paste. Once the optimal angle is found, it is kept in place by a set screw (ultimately probably three set screws).

    I know I'm asking to be spoon fed here, but these crystals (especially the vanadate) are ridiculously expensive... I really don't want to make a $400 mistake.

    Thanks!

  2. #2
    mixedgas's Avatar
    mixedgas is offline Creaky Old Award Winning Bastard Technologist
    Infinitus Excellentia Ion Laser Dominatus
    Join Date
    May 2007
    Location
    A lab with some dripping water on the floor.
    Posts
    9,890

    Default

    I'll try to get back to you after work today.
    Your mount design is missing a few things and we need to discuss the properties of how indium sticks.
    ~
    The screw coming down from the top will place measurable stress on the crystal, possibly causing some nasty birefringence or power loss. Think split halves that clamp the crystal mount with low stress. You also need to heat the ktp in most cases.
    ~
    Your mount also needs to be align-able within .05 degree or better. As well as to align the crystal along the beam axis in rotation (A pilot would call it Roll) very precisely. Yes, you will temperature tune the KTP, but odds are your crystal has the 45' cut already to minimize drift at near room temperature. Type I SHG cuts are more or less temperature tuned, more so then type II.
    ~
    Steve
    Last edited by mixedgas; 05-05-2017 at 08:32.
    Qui habet Christos, habet Vitam!
    I should have rented the space under my name for advertising.
    When I still could have...

  3. #3
    Join Date
    Jun 2008
    Location
    Yellowknife, NT, Canada
    Posts
    2,147,484,113

    Default

    Subscribing...

    15 chars

  4. #4
    Join Date
    Jan 2006
    Location
    Charleston, SC
    Posts
    2,147,489,446

    Default

    Eric (Planters) has done some experimenting in this area... Might want to ask him what his research into "the big green" project yielded in terms of KTP mounting options.

    Adam

  5. #5
    Join Date
    May 2017
    Posts
    22

    Default

    Quote Originally Posted by mixedgas View Post
    I'll try to get back to you after work today.
    Your mount design is missing a few things and we need to discuss the properties of how indium sticks.
    ~
    The screw coming down from the top will place measurable stress on the crystal, possibly causing some nasty birefringence or power loss. Think split halves that clamp the crystal mount with low stress. You also need to heat the ktp in most cases.
    ~
    Your mount also needs to be align-able within .05 degree or better. As well as to align the crystal along the beam axis in rotation (A pilot would call it Roll) very precisely. Yes, you will temperature tune the KTP, but odds are your crystal has the 45' cut already to minimize drift at near room temperature. Type I SHG cuts are more or less temperature tuned, more so then type II.
    ~
    Steve
    After thinking about it some more, I don't think my design is even close to a tenable solution.

    I've looked at some lab-quality SLM DPSS lasers, and exactly none of them use adjustable mounts for the KTP -- it always seems to be soldered.

    However, there's a YouTube channel called TechIngredients where the guy shows the cavity of his home-built DPSS laser.



    I've looked at this many times, but don't understand exactly how it works. I can't make out how the triangular prism shaped KTP mount is attached to the rest of the apparatus.

    I heard him mention that he's a member of these forums, but he didn't mention his username. Does anyone know his username? I'd love to bug him with some questions

    EDIT: I think I neglected to mention that my goal for this laser is to achieve single longitudinal mode operation, the ultimate purpose being holography.

    And while we're in this area... does anyone happen to know how stable the KTP and vanadate temperatures need to be to achieve SLM? I've read some sources saying they need to be kept stable to within one thousandth of a degree :|
    Last edited by lasersareuseful; 05-06-2017 at 16:24.

  6. #6
    Join Date
    Apr 2009
    Location
    North of Jackson Michigan USA
    Posts
    363

    Default

    PL member Planters did this.

    ...Mike
    Runs with Lasers

  7. #7
    mixedgas's Avatar
    mixedgas is offline Creaky Old Award Winning Bastard Technologist
    Infinitus Excellentia Ion Laser Dominatus
    Join Date
    May 2007
    Location
    A lab with some dripping water on the floor.
    Posts
    9,890

    Default

    It sounds like your trying to build a SLM laser for holography or something? What is your end goal for this laser?
    ~
    Look, I've serviced DPSS for about four years now. Including SLM Compass, and forget soldering the crystal at home. My current record is 32 Joules of 532 at 700 picoseconds, so I might just have an idea of what I'm talking about. If you have something bizzare, like an older Coherent compass, they have been known to apply a weak electric field to the crystal for fine control, having soldered it in place with heat pads built into the glass ceramic baseplate and a low melting point solder. Then use the heat pad at a much lower wattage to course stabilize the laser. That is way beyond the scope of a first time build. Its also overkill for most uses. I've also worked on Chinese lasers that literally had a glued KTP crystal shimmed to alignment with a piece of chewing gum wrapper. What exactly are you trying to do?
    ~
    This is simple. Take a copper or aluminum cylinder. Split it down the middle. Mill a slot into both halves. The Vertical dimension in the slot is milled to the thickness of the crystal plus the twice the thickness of the indium foil minus a few thousandths of an inch. You place indium on both sides of the crystal. Indium has near zero yield stress, and as you compress the halves together around the crystal it will one, bond to the crystal, two extrude outwards. The patch of indium typically is 3/4 the overall all length of the crystal.
    ~
    Mill the horizontal dimension so you can get a mylar or plastic shim stock spacer between one side of the crystal and the slot, or experiment with the milling so that you can wrap the crystal with indium on all four sides. As the non optical faces of the crystal are rarely parallel and perpendicular, I tend to go with the mylar. Otherwise you might crack a few crystals if your using 3x3x5 to learn on. The more expensive the crystal, the better the side faces. My previous employer used very thin Teflon pads, but their crystals were 15x15x30 mm.
    and gripped the crystal on two faces.
    ~
    At the end of the cylinder, either knurl it, put a wheel on it, or drill holes that you can engage with a allen wrench to rotate about the roll axis. This matches the crystal to the intra-cavity polarization, and is quite critical, especially with Vandate hosts. My guess is Planters used the triangular shape to make it easy to set the roll axis with his fingers.
    ~
    There are no bolts or screws holding the cylinder together, once the indium sticks, it sticks, and you will nearly have to destroy the crystal or gently melt the indium to remove it.
    ~
    To prevent shear stress on the crystal, its mount is clamped between two halves of the heater block. Unless you want to suffer, put the crystal on some form of X-Y tilt mount the first time you build one of these, to figure out what adjustments you need. Once you have the design nailed down, and have figured out how your crystal manufacturer marks the polarization axis on the crystal, you can make a second mount with more stability. That white dot of paint or pencil "X" on the top of the ktp can have different definitions at different factories. You may need a few millimeters of "Z" translate to find the sweet spot the first time you build a cavity.
    ~
    Most times the crystal is heated to about 20 to 30''C over room temperature, unless a specific cut is ordered, my former employer liked a 70'C cut and orientation . This only needs a say five to twenty watts of heat to achieve, depending on the crystal size and how much heat conduction, which is mount dependent. Usually controlled by a thermistor and a PWM loop. Often, just a NPN Darlington transistor in a TO220 case is used as the heater, for very small crystals. In most cases, 0.1' to 0.3' C is just fine, and most Chinese laser show lasers I've seen hit say 1 to 3' C accuracy. So for SLM, try for 0.1' Deg C or better.
    ~
    My former employer used just two op-amps, one for the oscillator for the PWM ramp and one for the comparator. There was a precision voltage reference on the board, which had all of say 20 components (Sorry, don't have the diagram for you.) We then had a 20 turn trim pot for fine T tuning, after a course alignment. Generally the factory set that once....
    ~

    Some lasers use the KTP as a self quarter wave plate to rotate the intra cavity polarization to reduce "green noise" or select a specific mode, if using a YAG host. Other lasers use Vandate, which is self polarizing, so you need to pay attention to the host crystal orientation as well. So it is best to start with a scientific paper that publishes a known design. After all, you need the crystal setting on a intra-avity beam waist in low power lasers to get enough energy density for conversion to happen. Think Quadratic equation with the first term a Cube (^3) and the second term is a square, (^2), so frequency doubling needs a very high field strength to occur.
    `
    So for a beginner, its wise to design a set of small, flexible mounts for your first laser with a lot of adjustments.
    ~
    KTP alignment is easy if you have the crystal even remotely close, you'll usually see some green even at far off misalignment or not at temperature. You may even find two or more angular regions where it will convert IR to green, however only one region will be stable and efficient. You'll then fine tune the temperature. Temperature and angle interact.
    `
    The big companies have the time and money to design something ultra stable. A beginner is going to have to reach that thru trial and error.
    ~
    0.001 C, NO, never, not that tight. If you need something that precise, one cavity mirror will be on a Piezo element to translate its length and lock the cavity mode for either peak amplitude, or Pound Drevor Hall locking to an etalon. SLM lasers usually have a light loop, a current loop, and a few temperature loops, including heat/cool the host crystal as needed.

    ~
    What is on Youtube, and what is inside most SLM or high power green lasers are two different things. Most people wont crack open a 10,000 to 30,000$ laser for filming. Think monolithic aluminum block, hollowed out on a mill, with the block floating from the optical table, setting on top of an active thermal control pad, and then the diode, ktp, and laser crystal having heating or cooling Peltier elements as needed. (or in the case of older Verdi, a water cooling pad for heat sinking the massive diodes)
    ~
    Planters on here = Tech Ingredients..
    ~

    If you just want a few hundred mW of stable SLM light, there is always the other way, the tiny aluminum monolithic miniature laser, such as that described in:
    !
    Compact and efficient Nd:YVO4 laser that generates a tunable single-frequency green output
    Graham J. Friel, Alan J. Kemp, Tanya K. Lake, and Bruce D. Sinclair
    Applied Optics Vol. 39, Issue 24, pp. 4333-4337 (2000)
    !
    The above is tuned with one Piezo for the whole assembly, and one Piezo for the fiber coupled pump diode. It uses the KTP as a birefringent filter to ensure SLM.
    Tear down a Ebay JDSU micro-green and you will see a similar structure.
    !
    Steve
    Attached Thumbnails Attached Thumbnails Lexel KTP  Mount..png  

    Last edited by mixedgas; 05-08-2017 at 08:27.
    Qui habet Christos, habet Vitam!
    I should have rented the space under my name for advertising.
    When I still could have...

  8. #8
    Join Date
    Jun 2008
    Location
    Yellowknife, NT, Canada
    Posts
    2,147,484,113

    Default

    Glad I subscribed, thanks Steve.

    On a side note... 32J at 700ps works out to a very nice number when you calculate the wattage, 45,714,285.285714285714285714285714285...W ad infinitum... Also... WOW.

  9. #9
    mixedgas's Avatar
    mixedgas is offline Creaky Old Award Winning Bastard Technologist
    Infinitus Excellentia Ion Laser Dominatus
    Join Date
    May 2007
    Location
    A lab with some dripping water on the floor.
    Posts
    9,890

    Default

    Diachi,

    That was considered a small one, and it was designed for a very high beam quality, in fact, the beam was large and collimated, so it would barely damage black spray paint on a 2x4. If you focused it, however, things would get interesting. No, We never tried the air snap thing, least the snap plasma do something non-linear and reflect beam back into the amplifiers, which is a possibility at very high peak powers.
    !
    Yes it was impressive, but after you were around a few shots, it became nothing more then a big oscillator-amplifier-amplifier-amplifier YAG with a few interesting side effects. Woe betide you if you did not respect its NOHD and any scattered light, however. The rep rate was very, very, low.
    !
    I just helped tune and install it, and handle relations with the customer. Install consisted of bolting all the parts to an optical table, tuning the master oscillator, and aligning everything else. The designer was a shy, but very nice fellow(RIP) and taught me a lot about beam profiling. I did also get to see it at the factory before it was shipped over here. I spent much of my time pulling cables for the flash-lamps and keeping people away from the pulse.
    !
    Before anyone asks, you aim it with just the master oscillator on and the rest of the lamps carefully locked out. A very tiny pulse of green comes out, then.
    !
    Steve
    Last edited by mixedgas; 05-09-2017 at 14:11.
    Qui habet Christos, habet Vitam!
    I should have rented the space under my name for advertising.
    When I still could have...

  10. #10
    Join Date
    May 2017
    Posts
    22

    Default

    Quote Originally Posted by mixedgas View Post
    What exactly are you trying to do?
    Indeed, I'm trying to build an SLM laser that operates in the hundreds of milliwatts, and yes it's for the purpose of holography. If this project goes well I'd like to ultimately build an SLM laser in range of seve

    Quote Originally Posted by mixedgas View Post
    This is simple. Take a copper or aluminum cylinder. Split it down the middle. Mill a slot into both halves. The Vertical dimension in the slot is milled to the thickness of the crystal plus the twice the thickness of the indium foil minus a few thousandths of an inch. You place indium on both sides of the crystal. Indium has near zero yield stress, and as you compress the halves together around the crystal it will one, bond to the crystal, two extrude outwards. The patch of indium typically is 3/4 the overall all length of the crystal.
    So it's OK if I'm not making thermal contact with the front and back faces? I was worried about thermal lensing, but I'm not sure if that's an issue in the 100-200 milliwatt range.

    Quote Originally Posted by mixedgas View Post
    Most times the crystal is heated to about 20 to 30''C over room temperature, unless a specific cut is ordered, my former employer liked a 70'C cut and orientation
    I am very unfamiliar with the different cuts of KTP and vanadate -- when I convinced Castech to sell me a crystal (got them down from $450 to $150 with a bit of sweet talk ) they asked what kind of cut I needed. I did some very quick research and responded a-cut (also because of how they phrased the question, "you want a-cut, right?")... what does this mean?

    Quote Originally Posted by mixedgas View Post
    At the end of the cylinder, either knurl it, put a wheel on it, or drill holes that you can engage with a allen wrench to rotate about the roll axis.
    Would the cylinder fit into a block like the design I posted? If so, would it be reasonable to apply arctic silver thermal paste between the cylinder and hole?

    Quote Originally Posted by mixedgas View Post
    My former employer used just two op-amps, one for the oscillator for the PWM ramp and one for the comparator. There was a precision voltage reference on the board, which had all of say 20 components (Sorry, don't have the diagram for you.) We then had a 20 turn trim pot for fine T tuning, after a course alignment. Generally the factory set that once....
    The voltage reference had 20 components? I ask because there are integrated circuits which advertise the function of "precision voltage reference" -- if I could make a controller with just a couple of op amps, a transistor and a precision voltage reference and get 0.01c temperature control accuracy that would be great. I'm handy with electronics but a lot of the temperature control circuits I've seen are way over my head.

    Quote Originally Posted by mixedgas View Post
    SLM lasers usually have a light loop, a current loop, and a few temperature loops, including heat/cool the host crystal as needed.
    Can you explain the light loop? Is this just the photodiode you see in commercial lasers for measuring the actual power of the laser output? I'm not familiar with the current loop either -- is this the constant current circuit for the laser diode driver?

    Quote Originally Posted by mixedgas View Post
    Think monolithic aluminum block, hollowed out on a mill, with the block floating from the optical table, setting on top of an active thermal control pad
    Active thermal control pad... is this the two layers of TECs you see in some Coherent lasers? I assume the purpose is to get rid of any potential temperature gradients and also reduce the load on the component-specific TECs?

    Thanks for taking the time to reply, and forgive me if I fail to pick up some of the wisdom you're putting down. Just trying to get a crystal clear (heyo) picture of what I need to do before I start. I will be mocking up the laser cavity in Autodesk Fusion 360 once I'm close to feeling competent.

    That paper you mentioned is fantastic by the way! For anybody interested in reading it:

    http://www.osapublishing.org.secure....=ao-39-24-4333

    This is a startlingly simple cavity -- the paper states its length as 15mm! If this is the case, why don't we see lasers in the 200mW range this simple and compact? Also, I don't completely understand the Brewster plates role in mode selection. The randomly polarized light from the fibre-coupled pump should be polarized once it exits vanadate, no? The paper mentions it in conjunction with "birefringence" which I have a very shallow understanding of It also seems that the far side of the KTP is functioning as the OC mirror?

    My vanadate has the following properties:

    Dimension 3x3x2mm and 1% doping.
    S1:HR@1064&532 nm,HT808 nm, R>99.8%@1064&532nm,T>90%@808nm
    S2:AR@1064&532 nm, R<0.2%@1064nm,R<0.5%@532nm

    My KTP is 3x3x5mm with AR coatings for 532 and 1064nm.

    Compared to this design:

    Click image for larger version. 

Name:	dpss1asm.gif 
Views:	27 
Size:	8.8 KB 
ID:	52079

    It would then be resonable to omit the HR mirror as its present on the backside of the vanadate, and add a brewsters plate between the vanadate and KTP?

    By the way, I actually did buy a cheap 10mW JDSU uGreen and take it apart for educational reasons. I'm wondering if it would be possible to use the OC and possible the brewsters plate in my laser, but the power would be 10-20 times higher.




    As you can probably tell, I have no formal education in optics/photonics (or anything really)

    Thanks again, it really means a lot to me. Hopefully this thread will help out a fledgling laser hobbyist like myself in the future

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •