Telescope Reviews: Catseye vs. Howie Glatter and Blug??

Permit me a few comments, though coming late into this thread.

I am quite sure that whichever tools you use, "passive" with AC or "active" with laser, it is possible to collimate your telescope far better than you can expect to tell visually.

You might consider what collimation errors do: perfect collimation of the primary mirror axis will put the spot of zero coma accurately in the center of the field of view - an error will put that spot a little off center. You can try: if you can find one spot within the FOV but a bit off center and get a clearer, more contrasty view, then the primary collimation is off by just that offset (this could happen - most likely, the marker isn't exactly centered, or else but much less likely, the optical center is not in the true center of the blank. If this is the case, star collimate accurately and move the center spot accordingly. This is the only reason I can think of for star collimation!).
You could always try and see how much error it takes before you can see it - collimate the secondary, then "carefully" miscollimate the primary using a Cheshire or barlowed laser (the true miscollimation is half the shift of the spot! - or if you use the AC, 1/8 of the distance between the most separated spots).

Miscollimation of the focuser axis will affect only the outer parts of the FOV - try to detect a difference in star images around the field edge - and see if you can move the focuser to get the best sharpness at either of the "opposite" edges of the field.

The tolerance I estimate for focuser axis error is < 1/30 of the primary diameter - this will make the tilt defocusing error < 10% of the total error (dominated by coma) anywhere within the field (this isn't even dependent on F/ ratio!). Given that the Paracorr reduces the coma of a paraboloid mirror by a factor of 6, it can be meaningful to reduce the focuser axis error by 6 too - which is likely rather over-ambitious considering the inherent errors of real eyepieces. But since it is easy to achieve, there is no harm, and little time lost, to do even a little better - like going for less than 0.1" centering error for a 18" mirror (that is, less than 0.6" for the most separated images in the AC, given the primary is exact!!!).

This is less important with visual than with photo - if you do wide-field photo with a fast Newtonian, you will use a coma corrector such as the Paracorr or even one of the even more sophisticated coma correctors available, anyway, and tweak the camera mount to get the most uniform star images across the field.

You may strive to get collimation as close to perfect as possible, or else just collimate close enough that errors will be assuredly invisible. This may affect the approach - using a (good) laser collimator with some type of barlow attachment (such as the Blug) will get you all the way in two steps (after centering the secondary of course) - first adjust the secondary to center the beam at the primary, then use the barlow attachment to adjust the primary - and you're done (with the 1mm aperture, it can be even simpler!).

While the AC will magnify the apparent errors (4 times the displacement in the Cheshire or barlowed laser, 6 times the misalignment of the focuser axis - or with the CDP, twice), the disadvantage is that what you see is the mixture of errors and you have to sort out which is what. But with proper tools of either type, you won't be able to tell the difference looking at celestial objects.

Nils Olof