So I think I pretty much know my imaging train isn't exactly aligned to my optical train. This conclusion is arrived after having attending a talk about using your flats to diagnose imaging issues. During that talk it was discussed that having a well aligned imaging train would eliminate the need to take flats for any specific rotation of the imaging train, which is what I have always thought. The reasoning behind this is that the dust that causes the out of focus doughnuts in our images are all from the cover slip or surface of the CCD, filters, and other close proximity surfaces such as AO lens surfaces. In my case using an RC the next closest optical surface would be the secondary mirror which would be too far away to affect the flat. That said, the camera, filter wheel, AO or pickoff mirror would all rotate at the same time so also would the effects causing the flat to be the same. I use a FlatMan XL
http://www.optecinc.com/astronomy/catalog/alnitak/flatmanxl.htm and have it mounted so that when the telescope is parked it is facing the El panel and take my flats accordingly. I have re-read this article many times
SkyandTelescope.com/reprint/610 and always seem to miss something that I must be doing wrong. I take 72 flats for each PA and filter for my images. 36 are at the PA of the east side of the Meridian and 36 for 180 degrees opposite for the west side. So for my current M51 project I have my object and guide star set at a PA of 90 degrees to get the guide star in the guider chip of my SBIG STL-11002M dual chip camera. When the object crosses the Meridian to the west side my rotator rotates to 270 degrees to pickup the same guide star. This is all automated using ACPExpert (ACP + Scheduler) and MaxImDl.
Any thoughts on this would be appreciated. I have been wondering if I could use image calibration or Pixel Math to check for differences in the flats to try and determine where the differences are coming from. In other words use the west flat to flat the east flat and see what differences there are. Is there any logic to this? Or should I be looking elsewhere? The only thing I can think that could make a difference is if the filters are loose but then using the same flat on multiple images would have different results and that's not happening. A repeatable stop for the filters might also make a difference but then again I would expect the dust doughnuts to become elongated instead of round when stacked. You also need to understand the process of how the images are usually taken. The color series is usually done by taking a number of red, green, blue images such as r,g,b, r,g,b,r,g,b, and so on. Not say 10 red, 10 green, and 10 blue. So the filter wheel is always moving between filters and back to the luminance filter to plates solve and focus.
Thanks for any thoughts.
