Flats undercorrecting in LRGB [Deep Sky] Processing techniques · Adam Drake · ... · 9 · 160 · 3

andrea tasselli:
Normally, it would be a light leak of some sort (by light leak we mean an additive source of illumination, not necessarily a factual leak). If the only constant is the filters then one possibility is that the edges of the filters reflect some light back into the optical train (and it would do it differently between flats and light exposures, saying just to preempt confusion).

John Hayes:
Flat correction requires a number of conditions to work well.  First, you can’t have additive stray light—as Andrea pointed out.  Second, you have to be operating with linear signals.  I have never taken flats with such a short exposure and I have to wonder if that might be at least a part of the problem.   Have you tried a longer exposure?   The shortest that I ever used is about 3s and I try to get the peak of the histogram to be between 66% and 75% of saturation.  Third, you cannot have too much vignetting.  Flats can correct up to about 50% vignetting but beyond that, you can start to have serious problems.  You can tell if you have way too much vignetting if the flat histogram is fairly broad.  A properly configured system will produce a flat with a pretty narrow histogram.  Just based on what you’ve shown, it doesn’t look like you’ve got too much vignetting so I doubt that is the problem.

Beyond that, your code may not be configured with the right parameters to work properly.  For example, PI has a number of options that if checked, can really screw things up.  How are you performing the calibration?

John

Adam Drake:
I've used the default Calibration settings in AstroPixelProcessor for years without any issues on most of my other scope/camera/filter combinations.  I'm leaning towards trying PI for stacking, just to take the app out of the troubleshooting equation.  However I've stacked the same data with DSS and gotten similar results.

As for taking flats, I agree that the next step is to try longer exposures.  I recall from my days struggling with getting my light source  both dim and bright enough for my 294 series cameras' need for 3-second-plus exposures, and I'm not looking forward to dealing with that again if that is indeed the issue.

Also, my actual vignetting is quite small; as I'm sure you know the stretching greatly exaggerates it.  The scope used in this capture is made to cover full-frame cameras, so one as relatively small as the ASI533 poses no challenge there.