Can't get flats correction to work properly

[iva]

Hi there, I've been a user of PI since a while already and appreciate all the help coming from this forum, even if I never posted a question directly, thank you all!

Ok, here's the problem:
since I changed my setup adding a Barlow lens, the flats became uglier and now PI doesn't seem to be able to correct the subs properly.

I haven't been taking pictures for 8 months or so, meanwhile PI has been updated and so I'm not sure the issue is only cause by the quality of the subs or if I'm missing some new "knobs" in PI compared to the past.
I've also noticed warnings all the time now regarding Darks, these were not appearing last year and I haven't investigated yet why that is the case.

Now to the pictures!
First you can see how the old flats looked like, there is a main reflection which is repeated a few times.
Now the new flats have the same issue but with smaller size (probably due to the added Barlow) but spread all over the shop, plus one uglier area which wasn't present before.
Even with 20 flats PI doesn't seem to be able to correct the subs, third picture is a session from last night, stack of 25 light of 3 minutes each plus, 20 flats, 10 darks (taken from a master) and 20 bias (also taken from a master).
where I see the "wavy" pattern that can be taken care of only by moving the black point very high and eating a lot of information (which I wouldn't want to do), I've exaggerated the stretching to highlight the problem.
I'm planning to go to the basics and start taking subs and flats simplifying the image train, but if you have any suggestion how to process the current subs or where to look specifically that will be more than welcome!

My imaging setup is a Skywatcher 750/150 PDS newtonian, the 2.5x 2" barlow lens, an Atik Filter wheel with Baader RGB/NB filter and an Atik 414 EX.

PS I tried to stretch the images to highlight the problems
PPS I'm tempted to get rid of the Barlow and move to an 8" RC telescope to get the a bit more light and same focal length, but before making the purchase I'd like to understand if I'm asking too much of the good old newtonian!

[TobiasLindemann]

Hi, the donuts are caused by dust on an element which is near to the camera. As you had not the problem before, I would suspect the barlow. Nevertheless the flat should correct this at your calibrated image. It seems like it will be over-corrected. This is sometimes an issue of missing, or wrong biases. Do you use the preprosessing-script? Maybe you can try use all raw files and not the masters. Maybe the script will try to correct the already calibrated masterflat with the bias again, which is certainly contra productive.

Tobias

[John_Gill]

Hi,

I think you have some dust/fingerprint/smudge/watermark on your optics.  Perhaps give your optics everything a gentle clean.

Look up
John

[iva]

Hi, thanks for the replies.
I tried without the masters, also taking new dark and bias (and also new flats) in all possible combinations in the post-processing script, but the result of the stacked image was always similar to what I posted above.
I could give it a go doing the steps manually but I need to read a bit more about it, until now I always used the script (with pretty good results, I have to say).

Certainly the optical train needs a good clean up, the point though is understanding up to what point a good pre-processing with flats can get rid of the issue.
At this level of "ugliness" in the flats it seems that the effect can't be taken away.
So I ended up moving the black point very high to remove the waves and as a result I lost tons of details of the fainter stars, see attachment.

Regarding the cleaning, should I check also the primary/secondary or only start from the filter to the camera?

[TobiasLindemann]

It is strange that you have overcorreted images. Is it possible you can upload one image of a light, dark, flat and bias?

[AstroDude]

hi everyone, me and my friend are also having the same problem, of overcorrecting flats in our final image...
maybe we should forgo the Bias frames and try again. my flats are from teh T3i @ 22500 ADU or 1/3 of the histogram in NEB4.
any suggestions?       and even after fixing our DSLR workspace our image has no color (shot in CR2)

 mitch/Gregg

[TobiasLindemann]

Hi Mitch/Gregg,
what do you mean by "forgo" the Bias frames?
I think 22500ADU will do, but I would go to a higher level of about 35000 ADU, but I think this is not the problem. It would help, if you can upload some raw-files (light, dark, bias, flat) somewhere.
Concerning the greyscale: do you have debayered your image after calibrating? Do you have used the preprocessing-script, or do you have done it manually?

Tobias

[AstroDude]

Tobias, i can upload the CR2 frames )light, dark,flat,bias) in dropbox for you and put the link here..

hope this will help you to find the problem with overcorrecting flats.

FORGO= do not use them ...

Mitch;.

[TobiasLindemann]

that would be great :-)
The problem with the greyscale image you have solved, yes?

[AstroDude]

not sure if the post worked.

https://www.dropbox.com/sh/5ijtydmstcdkz58/AAB4nwL4boLnNa-OBNPV23tNa?dl=0

mitch