Mirror alignment. Don’t touch!

The new OAPs in protected aluminium have arrived. The rotating CCD-mount is complete. In other words, it’s time to play.

Figuring out how to align a plane mirror isn’t so bad. It’s still challenging to carry out though, not least because my flat is from 1936, two years before the 90° angle and the level floor-board was invented.

Figuring out how to align two off-axis parabolic mirrors (OAP)  was less easy, and I’m still not convinced I’ve done it right. But judge for yourself (and let me know if you find it wanting).

The spectrograph’s geometry is something like this:


The two OAPs are furthest away. The three other mirrors are plane elliptical mirrors. Without the two elliptical mirrors close to the OAP, the OAPs are supposed to have coincidental focal points. And the – what’s it called? – the other end is supposed to point the exact same area on the grating (which will sit where the two lines from the OAPs cross).

The plane elliptical mirrors are “easy”. They just have to be perpendicular to the floor, and intersect the optical axes at 45°. The OAPs have to tilt inwards towards the grating at an angle of 5 point something degrees, and be turned – also 5 point something degress, so their focal points coincide. And here’s how I did:

  1. I started with just the OAP mirror holders. I’ll probably do a technical drawing of everything next time we have another “interesting” meeting at work:


    We’re looking at the bottom of the box (Galaxy 388). The box is made of 2 mm aluminium and rather flimsy, so the actual floor of the spectrograph is the 8 mm aluminium plate bolted on top of it. In the far right corner is the OAP-holder.

  2. The OAP mirror holders were adjusted to be perpendicular to the floor, using my most best piece of metal that I trust to be almost 90° at every angle. It looks like this:

    It may not look super aligned, but I also had to hold the camera. If I were to start over, the first thing I’d do would be to buy the CNC-machine and cut and drill the aluminium with proper precision.

  3. Then I mounted an LED-torch on a tripod 7 meters from the OAP, and using lasers and magic, I did my best to make sure, that the torch was on a line straight from the OAPs’ ..other end. The one that doesn’t focus.
    Then I mounted the lower OAP, and turned it so it’s focal point would be 32 mm above the 8 mm plate, or 40 mm over the bottom of the box:


    It’s a bit hard to see, but there’s a small spot of light on the aluminium target. Right where it’s supposed to be: 151,8 mm (ish) in the x-direction, 13,5 mm in the z-direction. That’s 152,4 mm in space, the focal lenght of the OAP.

    Then I mounted the upper OAP and did the exact same thing. No photo of that, sorry.

  4. Then I mounted the first plane mirror, and to make sure it’s perpendicular to the floor and 45° in the optical path, I moved the target and adjusted the mirror.
  5. I then installed the second plane mirror. I placed the sides of the entire box in their right position, and adjusted the mirror until the optical fiber attached to the fiber port showed the greatest intensity.

    From the right: The two OAPs stacked. Plane mirror 1, and plane mirror 2. You can just barely see the fiber port pointing at plane mirror 2.


  6. Finally plane mirror 3 was inserted. It was adjusted so the focused spot fell on the CCD-mount 40 mm above the bottom of the box, that’s 32 mm above the 8 mm thick floor of the spectrograph

    We see the two OAPs. The rotating CCD-mount. The custom STM32F405-board with ADC-conditioning circuit etc. And in the bottom of the photo, we can juust see the corner of the 3rd plane mirrors 3D-printed kinematic mount.

    Now I’m not 100% sure that 3rd mirror is 45° on the optical path, but it is perpendicular to the floor.

So now I won’t ever forget, and if I do, I can just come back here and see what past-Esben did. Of course the grating is still missing, and the second part of the alignment of the OAPs is missing. Remember their “infinity ends” have to cross, but at the moment they’re parallel. To do this last bit, I’ll have to remove plane mirror 1, install the grating, and tilt the two OAPs toward the same point on the grating. Then reinsert mirror 1 and hope I can position it properly.

Of course there’s also tilting the grating, so the angular dispersion matches the CCD. So not quite done yet. Thank you, and good night.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.