Collimating with a holographic laser
In the spring of 2004, at a star party, I had the opportunity to try collimation of Roger Persson’s 20 inch f/5 Obsession telescope, with a Howie Glatter laser collimator. I have long suspected that the holographic grid would be quite useful for centering the secondary, but I had not before had an opportunity to try it out.
What we did was to collimate the secondary with the “plain” laser, and then the primary with the Barlow attachment. After that, we put in the holographic grid attachment on to the laser collimator and held a paper below the secondary, picking up the reflection of the grid.
(Photos by Philip Noack, by permission. They are "color balanced" towards the blue, to bring out details better)
The focuser with the laser is “below” the photo, and the laser is turned to get the grid roughly aligned with the beam direction.What you see is the square grid reflected by the secondary, its corners cropped. At least in the second photo, you see a diffuse red background, outlining the full circular reflection. You will notice that the grid is offset to the left, and slightly upwards in the photo.
The common practice with Newtonians is to use a secondary that is so small that it crops the light from the primary, and at least with low-power wide-field eyepieces, the outer parts of the field will not receive light from the whole primary. This light loss is considered acceptable, but one would like to have the fully illuminated part of the field centered in the focuser and the eyepiece. This means that the secondary should be offset – it will be, when seen centered in a sight tube, but also when the laser grid is centered in the reflection in the secondary.
To move the grid down in the picture, we had to loosen the nuts on the center bolt and move the secondary holder down the tube a little (when these pictures were taken, we had already moved the bolt a good deal, but a little remains, as you can see). To shift the grid to the right, we need to shim the focuser to tilt it somewhat to the right, enough to center the grid (of course, you should check the centering of the secondary within the cage, to make sure it is not the secondary that is displaced to the left – if so, center it first before you decide if the focuser needs shimming! A focuser with an adjustable base could have been even more easily adjusted).
This done, the secondary is optically offset, and you can go on to fine-tune the tilt of the secondary, centering the beam on the primary’s spot, and last, collimating the primary using the Barlow attachment.
To get the optical axis exactly centered in the cage, the secondary should also be offset a little away from the focuser, but this is usually of little or no importance – for Roger’s 20” f/5 Obsession with (IIRC) 3.1” (~80 mm) secondary, the offset is approximately 4 mm – if the secondary is centered, the optical axis is offset some 4 mm towards the focuser instead, too little to matter much.
However, you can take a sheet of semi-opaque material (e.g. thin paper) and cover the tube or cage opening, and use the holographic grid again. Now you will see how well the grid is centered in the opening, and you can measure the distance from the rim to the grid at 4 points to see accurately how the centering is. (As an aside, you will see the shadow of the secondary within the grid, offset away from the focuser. Do not try to center it!) If you think you would like to improve the centering, what you need to do is move the secondary the required amount (the position of the collimated optical axis is fixed to the optical [offset]!), then re-collimate. This operation is rather pointless, of course, if you haven’t also adjusted the sling to center the primary just as accurately within the mirror box!
With a “faster” telescope than the f/5 one used here, the grid will appear smaller inside the reflected circle. I haven’t tried, but I expect it will work with faster telescopes – if necessary, rack the focuser as far out as you can.
After this was written, I received my own Howie Glatter laser setup, similar to Roger's. So here is what I tried, in order to get the positioning and adjusting of the secondary in one single operation - this may be of interest to some. The secondary cage is low-profile (the light cover is wrapped around the outside during use), and the secondary is not adjustable at the center - instead, the 3 vanes are clamped to the cage at their outer ends, and the clamps are loosened and the vanes are pushed into place before the clamps are tightened again (perhaps a bit tricky to adjust but I have full control of positioning, and the secondary is very stable when adjusted).
I made a strip of white cardboard and shaped its ends to fit precisely the openings in the cage, just below the secondary. I measured out and marked the very center of the cage with a cross on the strip, cutting a hole at the center of the cross. Next I inserted the laser with the holographic grid generator and turned it on, watching the grid pattern on the cardboard strip - and rotating the laser to align the grid.
(These photos are mine.)
Now, the idea is to center the grid (and the beam, of course) on the cross and hole on the strip, and also center the "cut-out" grid. To center the beam/grid "sideways" on the cross/hole, I have to shim the focuser or focuser board. To center it towards/from the focuser, I need to move the secondary along the tube axis, in the direction toward/from the primary, respectively. If I then adjust the tilt of the secondary to center the beam on the primary's spot, I have the focuser axis accurately aligned and reflected, and as the hole/cross marks the center of the cage, it is centered here, too.
This however does not ensure that the secondary is positioned correctly. I will have to move it sideways to get the grid cut out symmetrically in this direction, and move the secondary along its own plane in the other direction - that is, equally towards the focuser and away from the primary (without changing the tilt) or vice versa.
This may sound a bit messy, but since you see exactly what you have accomplished, the whole thing is easier than it may seem before you try. This method, of course, assumes that you can freely adjust the secondary in all directions needed to both get it centered on the line of sight and the optical axis centered in the tube - this may not be the case with non-adjustable vanes, and anyway if you haven't ensured that the primary is centered in its end of the tube, centering the optical axis at the other end of the tube is not much to strive for - it won't affect the performance if not accurately centered.
What you get this way is fully offset collimation.
Nils Olof Carlin, updated version June 12, 2004