ahh- isn't it fun
very much FWIW - a few ideas that might possibly help.
1. get a laser collimator and align it precisely - as Alexander showed
http://www.iceinspace.com.au/forum/s...ad.php?t=75601
2. carefully check the central spot ring on the primary and get it exactly right.
3. do a laser collimation through the coma corrector
4. check that collimation holds reasonably well by moving the scope around the sky with the laser running - if possible tighten anything that allows movement of the collimation spot. pay particular attention to the focuser and make sure that the shaft is tight enough that you do not get any rocking of the draw tube
5. if you still have corner-to-corner variability, then you have tilt, either in the camera or in the CC. make a cardboard spacer shim so that you can screw on the camera at a different rotation angle on the CC. If the tilt has rotated with the different camera position, it is in the CC, if it does not change, it is in the camera. if the corners all look the roughly the same, but are not fully corrected, check the CC spacing.
also FWIW, I had to add strengthening to the CF OTA around the focuser to stop flex and had to put new side shims on the primary to stop it from sliding too far sideways. I also added two nylon locking screws on the side of the focuser draw tube to ensure that the CC tube is locked solidly to the draw tube (it could wobble slightly when only locked at the top end). Also shimmed the camera to minimise very slight tilt in both the CC and the camera - found the rotation angle at which the two slight tilts came closest to cancelling each other. The optics still tends to drift a bit and recollimation is needed maybe every month or two - not a big price to pay for a fast scope.
cheers Ray