Calibration frames, meridian flip and DSS

[Delboy_Hog]

I'm really struggling here, and hoping you can help!  In the past I've not had any trouble with DSS coping with the meridian flip and calibration frames, but something's gone very wrong this time, and you'll see from the image I've got weird artifacts appearing all over the image (I've circled a few of the weirder ones).

On this occasion, I took about 40 light frames (M81 / M82 with ED80 and Canon 6d), then as the forecasted cloud rolled in for an hour or so, I took my flats and a few darks, then as the cloud had cleared, did the Meridian flip and took another 40 or so light frames, then took the bulk of the darks. 

Is it because I took flat frames before flipping, rather than at the end of the session?  Or because some of my darks were before and some were after the flip (but I wouldn't have thought darks could be responsible for these artifacts)?  Or has something else crept in here that I haven't thought about?  My bias frames are from another evening, but they worked fine last time (or at least, didn't cause any obvious issues), so can't believe they're responsible.  I tried rotating the RAW files, but have just learned you cannot apparently do this without converting the file type to something else.

I'm about to spend the next few hours (or days, with the speed of my computer!) trying to re-stack various combinations of files to try and route out the problem, but thought I'd see if anyone here knew what might be going on.

I'd be very grateful for your thoughts!  Image straight out of DSS...

[Delboy_Hog]

Hmmm, no, I'm having no luck with this....

The artefacts seem to be coming from the flats, and the issue is being compounded by the flip, duplicating some of the artifacts.  I take two sets of flats at different exposures in case one of them doesn't work...and this time neither does, regardless of whether I apply them to the pre or post-meridian flip light frames, or the whole stack.  I just don't understand how these flats can seemingly generate distinct artifacts?

It's a shame because I think there's some reasonable data here, and about 3 hours of it.  Unfortunately one of the perils of working with a 6d on an ED80 is that if your flats don't work, you're left with a largely unworkable light drop-off.  The good thing about working with a 6d is that it's a reasonable camera, so the data that falls right in the center of the image isn't awful, even if your heart is breaking at having to crop off 80% of the overall image!  Quick levels process in photoshop...

I might try and download one of those auto flat-generator software thingies to see if that can salvage anything from the wider-field data...

[david_taurus83]

Darks have no effect on artifacts in the image. I have a 6D and I don't bother with darks as the dark current is so low anyway. I just use a master bias to calibrate the flats and lights.

Meridian flip also has no bearing on the flats. If you have dust bunny in the top right of your image it will still be in the top right of the image after the flip. What can happen is the dust can move between images. I'm guessing the shutter action doesn't help. Are those dust bunnies in the individual subs? Try stacking the lights without calibrating with flats and see if they match the master flat.

[Delboy_Hog]

17 hours ago, david_taurus83 said:

Darks have no effect on artifacts in the image. I have a 6D and I don't bother with darks as the dark current is so low anyway. I just use a master bias to calibrate the flats and lights.

Meridian flip also has no bearing on the flats. If you have dust bunny in the top right of your image it will still be in the top right of the image after the flip. What can happen is the dust can move between images. I'm guessing the shutter action doesn't help. Are those dust bunnies in the individual subs? Try stacking the lights without calibrating with flats and see if they match the master flat.

Thanks David, yes I suspected it wasn't the darks - and as you say, I rarely use darks with the 6d unless my exposures are 10+mins - a good dither between frames more or less does the trick there, especially in this cold weather!

I would agree with you that the flip shouldn't affect the flats - it hasn't for me in the past.  But this time there are artefacts that seem to be generated through the flat frames (i.e. they're not there when I just stack the lights) that are being repeated across the frame (e.g. see the light and dark smudges in the top right and bottom left of the full frame).  I'm not sure if there's a light frame that the flats are correlating to that isn't stacking in the same orientation, or what's going on...

The only things I can think of that would make the flats not correlate to the lights is that the white t-shirt flats technique either added some dust to the front of the scope lens (which must have blown away when I took the t-shirt off, as the lights I took after that point don't work with the flats either!) or possibly that there was some crinkle or mark on the t-shirt, or an issue with the even lighting of the screen I was using.

Looking back over the frames, I also wonder if I didn't help myself by not re-framing precisely enough after I'd moved away to a brighter star to check focus.  There's also some shift in the orientation (field rotation?) in the stacked frames, which you can see in the full field image above, which may not have helped my cause, though again, neither of those have been a problem for DSS before...