Alignment and Collimation of Fast Newtonians

[MarkS]

There is a lot of literature on the subject of Newtonian collimation. There is very little on initial alignment of the secondary mirror, and even less on emphasizing its critical importance to performance of fast astrographs with large camera sensors.

 

Whilst axial collimation alone may be fine for observing and perhaps for very small photographic fields of view, it is a necessary BUT INSUFFICIENT condition for, say, an ASA10N and 16200 or 16803 class camera.

 

Such systems have very low tolerance to errors of co-planarity between the telescope focal plane and the camera sensor plane. Exact co-planarity is achieved only when three conditions are met:

 

      1. The primary mirror optical axis must be at exactly 90 degrees to the focuser optical axis;

      2. The primary and focuser optical axes must intersect on the secondary mirror optical centre;

      3. The elliptical secondary mirror major axis must be at exactly 45 degrees to both the primary and focuser axes.

 

Using any of the axial collimation tools, 'perfect' axial collimation can be achieved without ANY of these conditions being met.

 

The critical starting point is condition 2. This is effectively about the position of the secondary mirror along the primary axis. If this condition is not met, the telescope plane of focus will not be the same as the camera sensor plane, no matter how good the axial collimation is.

 

Yet I have seen articles which tell me that the secondary initial alignment is not too critical! On the other hand, ASA documentation gives the secondary offset figure for the telescope, but it doesn't really stress why it is so critical to get it right.

 

The best way to get it right that I have found is to block the primary reflection with a card, and using the Cat's Eye adjust the secondary so that its projected circular edge in the view is EXACTLY concentric with the edge of the CAT's Eye in the focuser barrel. Then I collimate the system as usual, and recheck the concentricity at the end of that process. The only final check on how perfect it all is, is to take a set of images of populous star fields and check the corners (I use Pixinsight's Aberration Inspector script.)

 

How critical is it? According to a rough calculation for an ASA10N and a 16200 camera, a 0.6 mm error in the primary mirror to secondary mirror distance will cause the focal plane to be non-aligned with the sensor plane by ~ 35 micron across the sensor diagonal. This will 'eat up' the total depth of focus at F/3.6 of +/- 17 micron.

 

Without trying too hard I have achieved beautiful axial collimation using Jim Fly XLK and Cat's Eye for secondary distances differing by 2 mm.

 

All this becomes a matter for attention if for any reason secondary mirror removal is required, or if the central secondary mounting bolt has been moved.

 

Hopefully, this is 'too much information' for everyone, but it may help any desperados who can't get round stars across their images even when the collimation is good!

 

Season's greetings all!

 

Mark

Edited December 17, 2017 by MarkS