After months of trying I cannot calibrate my RGB images!

[fmeschia]

The reflections are likely still there, they should scale in the same proportion as the “wanted” signal.

 

[cloudbait]

No it was much worse at 32000 ADU. This got rid of some of the reflections going to 12000 ADU

That right there points to something wrong with either the flat frames or the calibration process. The histogram for the flat just looks odd to me. Maybe that's because of the extreme vignetting (that and the distortion show that this sensor is too large for the telescope). My flats are much more symmetrical (and better exposed). I think the suggestion to compare sky and panel flats is good. I could never make a panel flat work on my SCT. If the two don't look the same, it means your whole procedure for taking flats needs to be examined.

What is certain beyond reasonable doubt is that the shadows cast by the dust on the filters is changing. Either because the filter wheel position isn't repeatable, or because the mirror is moving and shifting the light path, or some other flexure. That doesn't fully explain the failure of the flats, but it's the only explanation for that embossed appearance. I suspect the mirror, since the effect is greater on shadows cast from a greater distance from the sensor.

 

[tomb18]

Thanks, will do the panel flat tomorrow. It is supposed to be nice out.

 

[tomb18]

what happens if you calibrate a master sky flat with a master panel flat?

rob

Hi
I took my light panel flats yesterday and they make no difference. In fact, they introduce more reflections than before.
How exactly would I do this and what would I expect?

 

[pfile]

just calibrate the panel flat subs with a master sky flat, or the other way around. when you integrate the flat-calibrated flats, you should get an almost perfectly flat image. if the illumination differs between the panel flat and the sky flat, you'll see circular artficats in the integrated image.