Results 1 to 6 of 6

Thread: KTP Crystal Mount: How Many Degrees of Freedom?

  1. #1
    Join Date
    Jun 2017
    Location
    Seattle, WA
    Posts
    47

    Default KTP Crystal Mount: How Many Degrees of Freedom?

    Most of the designs I've seen for a KTP crystal mount allow X / Y angle along with rotation. I've built such a mount and in my testing I didn't find the X / Y angles to make much of a difference. Do they under certain circumstances? Crystals I've seen are cut to work parallel to the optical axis. Is flexibility here just done to account for imperfect beam alignment since the acceptance angle is quite narrow? My mount used the back end of a mirror mount for X/Y angle and adjusting it didn't make much of a difference unless I was way off parallel.

    I'm asking because I'm in the process of re-designing this crystal mount and it would be much easier to temperature control if I had a nice rigid connection to the optical table.

    Thanks all, Brian

  2. #2
    Join Date
    Oct 2010
    Posts
    41

    Default

    I think all 6 dofs are important.

    Having translation makes it possible to use a different spot of the crystal if the coating ever gets damaged.
    Z-translation might be almost useless.
    The crystal is often short compared to the rayleigh length having it not precisely centered should make almost no difference in crystal volume usage.

    Rotation (X/Y) does not need to be accurate but can be helpful to align or remove "ghost" spots that get generated by not perfect AR coatings on the crystal faces.
    Having a "ghost spot" close to the real output would bother me.

    Rotation (Z) is needed for polarization matching of the pump beam (depends on if your laser medium is actually sensitive to this).

    Using a L-bracket with oblong screw holes can give you 4DOF (X/Y rot+trans) already.
    Z Rotation could be accomplished by using a precision fit hole/bore.

    That gives reasonable thermal performance (with copper) while allowing all DOFs and being small at the same time.

  3. #3
    Join Date
    Jun 2017
    Location
    Seattle, WA
    Posts
    47

    Default

    Thanks for the insight -- I think I'll go with that. The current design mounts to a x/y flat mirror mount and the idea was to sandwich a TEC at the bottom of it with a flexible thermal pad to the base that allows for Y rotation. But I didn't think about the pressure needed to make good thermal contact with the pad. If I apply enough pressure to make good thermal contact the mirror mount shifts out of its spring-loaded ball bearings. I had to use a much thinner pad with much less pressure (and therefore worse thermal contact). It works, but takes a while to reach temperature and then overshoots. Laird makes TECs with holes in the middle so the new design will sandwich the TEC directly behind the crystal.

    Ghost spots...like these?

    Click image for larger version. 

Name:	IMG_1097.jpeg 
Views:	2 
Size:	697.7 KB 
ID:	59778

    Those are on my list too...just a bit further down and I haven't put too much time into tracking them down. But it would be interesting to pitch/yaw the current mount a bit and see if they shift. Bottom mirror is the OC and the KTP is just behind it.

  4. #4
    Join Date
    Oct 2010
    Posts
    41

    Default

    These are ghost spots that I was talking about. They get produced by all optical element's that should pass a beam.
    The AR coating isn't completely perfect and still reflects some light back.
    This light gets reflected by cavity mirrors end ends up in the main spot. OR next to it if the elements are not perfectly aligned.


    Squishy thermal pads as a thermal interface and compliance layer for alignment?
    I think you might be thinking way to complicated. (and sounds like a nightmare to keep aligned and thermally controlled)
    The temperature drop across the TIM would be very high and therefore only make a loose coupling with your thermal load possible. A nightmare to control (slow-overshoot-unstable in unsteady conditions)

    Have a look here:
    Attachment 59779

    The tec is on the left side.
    The holder got all DOFs.

    The tec interfaces directly with thermal compound to the L bracket part. The L-bracket is adjustable on the tec in a up and down and rotational fashion - "pitch" because of the slots for the screws.
    Same goes for the actual mount on the L-bracket. It offers translation and "yaw" rotation.

    Z- translation and rotation "roll" would be managed by the precision bore which gives you 360° rotation.

    Granted its not possible to fine adjust them but as you noticed, fine isn't really needed.
    and the holder should be optimized for your application, how much movement is necessary, how big is your crystal and so on.

    The L-bracket is quite lossy (thermal loss) - around 1,5°C at 5W, so making that bigger would be better. Copper is needed, Al would be way to high thermal resistance.

    Pro tip: don't place the screws where you can't access them without blocking the beam path. If you do that, there is no way to adjust something and watch what it actually does.

  5. #5
    Join Date
    Jun 2017
    Location
    Seattle, WA
    Posts
    47

    Default

    Thanks. The attachment doesn't work for me, but I'd love to see it. This is the design I currently have (not the new one I'm designing):

    Click image for larger version. 

Name:	Screenshot 2022-12-23 at 2.43.09 PM.png 
Views:	1 
Size:	1.17 MB 
ID:	59781

    Blue is the TEC. The crystal mounts solidly and is rotatable. the aluminum block on the right is screwed to an X-Y (pitch yaw) mirror mount and uses nylon stand-offs so it doesn't transfer much heat. The thermal pad is under the TEC to connect it to the base. The crystal stays in solid alignment and temp control regulates pretty well once it settles. But it takes a long time to settle due to the poor thermal connection. And I don't know how much heat the KTP produces itself but I doubt this will remain regulated if I run for a while at higher powers. I have the MCU shut everything down if the KTP temp ever gets .5ºC outside of the set point and haven't tripped that safety yet, but I don't run at high powers for long.

    The cylinder that holds the KTP was originally bored to hold a cartridge heater (large hole -- small hole is for the temp sensor), but I want to keep the TEC's ability to regulate in both directions. New design will use one of these sandwiched against the back:

    https://lairdthermal.com/products/th...annular-series.

  6. #6
    Join Date
    Jun 2017
    Location
    Seattle, WA
    Posts
    47

    Default

    Here is the redesigned mount. I don't have much play in the Y direction to keep the KTP centered on the hole in the TEC, but I have a good degree of freedom for X. The TEC is sandwiched behind the copper block using nylon screws to limit heat transfer. This will be mounted on a rail so I can get it in about the right position easily. I'm using a flush rail so I should get OK heat transfer back to the base breadboard. Feedback welcome.

    Click image for larger version. 

Name:	Screenshot 2023-01-02 at 11.13.46 AM.png 
Views:	1 
Size:	648.2 KB 
ID:	59812

Posting Permissions

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