Hi guys,

George, I don't do a library with flats (although I do with darks/bias). I take flats for each target.

John, I'm 99% sure MaxIm doesn't recognize PA as a criteria to determine which flat to use. I don't think there's a "standard" for the .fts header with respect to PA for MaxIm to use. As it is now, the PA captured by AutoFlat is the physical rotator PA since it doesn't know for sure what the actual sky PA is. (If the offset in RotatorInfo.txt is right, it could be calculated, but there's no guarantee it's right when taking flats.) The sky PA is what's captured for light images.

Regarding the "aging" of flats, here's my experience. Nothing in my setup changes from night to night, so there's not much opportunity for new dust to get on the various surfaces. With respect to rotator accuracy, while the Pyxis is probably only accurate to about 1 degree, I don't find that to be a problem at all. The real issue is the CFW8 accuracy. Normally, it's fine, but once in a while, apparently the CFW8 isn't in the exact same place for the flats and images, so the flats don't "work" as well as they should (meaning dust donuts aren't completely eliminated). But, I don't think the accuracy of the CFW8 is affected by the time between when flats are taken and when the images are taken. The issue is far more random - and doesn't occur all that often.

With my setup, I've worked pretty hard to eliminate as much reflected light as possible. I always image with a lined dew shield, I've flocked my scope, and I've painted all the exposed surfaces throughout the imaging train. Having said that, reflected light still causes problems with flat effectivness. The brighter the sky (like when imaging when the moon's out), the more likely that optical problems aren't completely eliminated by my flats.

Yes, it's kinda ugly dealing with flats when using a rotator. While you might figure out a better way, here's what I do:

1) Once I select a target, I'll do the flats for it as part of my imaging session. Often I'll do dusk flats for one target, and dawn flats for another one. I don't see any difference in "performance" between dusk and dawn flats. My "standard" is 12 exposures for each filter, 6 at the specified PA and 6 at PA + 180 (done automatically by AutoFlat.vbs).

2) I save these flats along with the flat plan to document for sure what my sky PA intent is on my image processing computer.

3) My normal imaging mode is to image Luminance over 1 (or more nights); the implication is I need to use both east and west flats. Then, I image the RGB, in that order; the implication is the R requires east flats, G requires both east and west flats, and B requires west flats.

6) In MaxIm, My darks/biases are already setup as masters. Then I'll add the clear flats manually - "Clear1" are the east flats; "Clear2" are the west flats. Check "Clear1" flats (uncheck "Clear2" flats)
and calibrate the east luminance images. (You have to review the ACP logfile to determine when the flip occurred.) Check "Clear2" (uncheck "Clear1") and calibrate the west luminace images.

7) Do the same for the RGB images. If you do what I do in step 5, you only have to add east R flats and west B flats. You need to deal with the G flats the same as with the Clear flats.

8) The closer you do the calibration to the actual imaging, the better - if only because you deal with the flats once, then forget them.

None of this is particularly hard, but I'll admit it does add extra work and requires some extra organization. However, I know I'm happier with my results when using an internal guider than when using an external one. So, it's worth it for me.