Help please! Some circle is messing with my images

The only time I had a similar problem it was due to a bad flat caused by a light leak. Show us your setup and we might be of help.

andrea tasselli:
The only time I had a similar problem it was due to a bad flat due to a light leak. Show us your steup and we might be of help.

*** Hello Andrea, thanks for you reply. I uploaded a picture of the camera installed in the focuser with all the accesories. Tell me if you need something else. Here is the link: https://drive.google.com/file/d/1pactX6t9ztromg1lBbzG5U0LsLwcAn1s/view?usp=sharing***

You will always have cirular patterns in your flat, or better: your flat is basically a circular pattern. If the flat was taken with a slightly different focus than the light, it will lead to circular patterns in the corrected light as well.

Could it be dust?  I have seen this before with dust and once I cleaned my sensor all was much better.

andrea tasselli:
Ok. Looked at them files. From the imaging train pic there isn't any clue that the light leek might be there. Seems a pretty solid setup. I'd like to see the rear of the scope which is possibly another source of light leeks. Looking at dark/flat-dark nothing to say about them either. Looking at the light frame and the flat frame it seems the flat is overcorrecting the light. So the question is how do you take your flats and in what conditions (day/night/dusk, with what and with the telescope mounted in situ or dismounted and so on).

*** Hello Andrea. Here is the link to a photo of the rear of the telescope: https://drive.google.com/file/d/1voE40M0TKhiRf_uhwZRSzOKcr4dQa2Dy/view?usp=sharing
As you can see, is covered with a paperboard (i painted it black). I took the flats during the sunset when the light is enough, pointing to the sky with a white t-shirt  I use the program called NINA and the Flatwizard feature (in this case the flat is 0.73 s of duration with the histogram at 50%, i have a tolerance of 10% also). The telescope is in situ always. I was thinking that something weird happened the last time that i was taking photos. I pointed to the south and the circle almost dissapears, but when i point to the southeast or east, the circle becomes very visible. I´ve only tried this once, so i´m not so sure (i'm going to try again definitely). Could it be related with the Baader Coma Corrector? I have the MKIII version and some people in others forum said that there is some problem with the coated optics. Thanks for your help!***

Hey, Fernando.  My work computer can't open TIF files, so I couldn't look at the files you shared.  Based on your descriptions of the setup, I'm going to go with either a light leak when taking your darks (there are too many joints in your shadowbox assembly, although I do love that idea!) or changing illumination when taking your flats.  I used to get very frustrated with trying to take flats at dawn/dusk because of how fast the ADU levels changed, so I just gave up and took them when the sun was fully up.  Just aim the scope well away from the sun, make sure there were no clouds, and use more layers of t-shirt.

Hope some of my ramblings can help out, please let us know!

After processing the final image I'm leaning toward an issue with processing rather than fundamentally wrong flat-fielding. Here is the stretched final image once cleaned up in PI. What program was used to process it?:

And, finally, unprocessed but auto-stretched in PI:

Hello everyone! i read all your replies and definitely i'm going to try all your suggestions. An stunning processing from Andrea and Andy, congratulations both of you. Wich tools did you use in Pixinsight to get rid of the circle? Some of you left questions. I'll try to answer them:

1-. Was the moon visible at that time (when you took the flats)? No, it was new moon. i have to add that there are some street lights (LED Lights) at 50 m from my imaging position. But, as you already saw, i've tried to cover all possible light leaks in my OTA. I've also added a dewcap in the front of the scope.

2-. What program was used to process it? I've integrated the files with Deep Space Stacker. After that, i process the image with PixInsight. Now i'm trying to do all with Pix.

3-. Did you change the rotation of your camera between flats and lights? No, the camera is always fixed. Every time that i select a new target, i renew all my calibration frames (Flats, Darks, Darkflats). I don't use Bias Frames.

Recently some people in other forums told me that maybe the Skywatcher 200PDS could have a smaller secondary mirror, and this can cause problems when you have to illuminate the whole sensor (my sensor size is 4/3"). Could this be related with my issue?

The rest of the week i'm going to have some clear nights. As i said before, i'll try all your suggestions. I hope i can point the problem soon, because i'm struggling with this for some time. Thanks to everyone!

Ref a couple of your points:

*  I used ABE with a very large number of small sample sqaures to extract the background;  I tried DBE first on it, but found ABE did a better job in this instance.
*  I also have the 200PDS/4/3" sensor combo and do not have the circle issue.  I do, however have some pretty significant vignetting, but that is easily corrected with flats.

One last thought as someone who transitioned from DSS to Pixinsight:  the Weighted Batch Preprocessor script was a very natural way to move from DSS to Pixinsight as it covers most of the same functions as DSS.  You will clearly want to learn the individual steps involved downstream, but it would be a quick way to start.