Flats don't appear to be working properly

[kirkster501]

See the arcs in this integrated M51 luminanmce image?  This is using pre-processing script in PixInsight to create the registered and calibrated subs and then I integrated them manually with Windsorised sigma clipping.  I don't understand why I have these arcs in place in the integrated image, reducing low-sigma on integration does not help.  Maybe I might try doing this entirely manually and skipping the script.

Meridian flip occurred but this should not matter?

Any thoughts or theories folks please?

[vlaiv]

Can you identify similar arc structures in your flats? If there are such structures, regardless if they are bright or dark (and it seems that one arc in the image is bright and the other is dark) it would be helpful to pinpoint problem to flats.

I'm not sure if flats are to blame, as you say you've done meridian flip. Any sort of problem with flats would manifest it self as symmetric in relation to subs center (same position on flats, but some light subs would be rotated after calibration due to align so artifact would end up in position rotated 180 degrees).

Also, flats tend to over correct or under correct (depending on problem causing artifacts), but rarely both under correct and over correct in same image (it can happen that it looks like that if for example there is problem with under correction and dust particle settles before taking flats which would make shadow of that particle over correct because it was not present in original light subs).