Date: Fri, 25 Sep 1998 11:44:40 -0600
From: Patrick Waddell
Subject: Re: 2.5m obslog 980924
I may be incorrect, but if "assymetric in the scan direction" means
"leading" vs "trailing" parts of donuts are of different brightnesses, we
may have been experimenting with the wrong axis.
When equator drift scanning, the "scan" direction, along CCD columns, is
parallel to the X axis of the primary.  To remove X axis coma, such as
induced by a translation in the X direction, we need to tilt about the Y
axis.
-p
>The rotator value reflects a slight change once tracking tests were done.
>Between about 21:30 and 22:00 we attempted to improve image quality by
>tilting the primary.  As the donuts were assymetric in the scan direction,
>an x-tilt change was indicated.  We tried moving in the positive direction
>but images were obviously worse.  We then tried tilting in the negative
>direction, however little improvement was seen in a change as much as 25
>arcseconds.