Flat causing rings ?

[thibaut]

Hi everybody !

First of all, I would like to precise I'm a beginner in all of this : stack, DOF, etc. But I've read a lot, and I think I'm rather aware of basic concepts. I'm shooting with a D850 unmodded, with a Tamron 35mm f1.4 @ f2 ; 2min ; ISO 400 for this example.


I'm facing a problem, and I cannot find its origin. After DBE, I have a kind of ring in my picture. And I cannot remove it. It's "subtle" but increasing with further post processing. Here with a huge stretch : https://drive.google.com/file/d/1kDfD5VtCXcA8sZnK-dKf5t6tgLzifM8j/view?usp=sharing


It may come from the flat, so I did a loooot of tests : expose to the left, to the right, centered, on the sky, using a teeshirt, high ISO, low ISO, fast flat, low flats... I used 1, 15 and 30 flats to calibrate. With other post-processing software, the same occurs. I try to stack uncalibrated raw files, and the ring seems to be here as well.


I've read this : https://www.cloudynights.com/topic/557931-concentric-coloured-rings-caused-by-flats/ ; and this https://www.cloudynights.com/topic/746131-nikon-coloured-concentric-rings/ but not sure if related.

Or, maybe I'm not good at using light pollution removal tools. Or I failed, each time, doing proper flats. I'm currently stack an other object, with another lens (90mm), from another place... I will see if I observe the same, to check if it comes from my camera.


Do you have other ideas of what I could check ? Here is a flat, a bias and a light (no dark as I was at ISO 400 and my sensor's camera is really powerfull) :
It's the first time I ask on a forum a solution, normally I always find it by myself. ?


PS : just an idea by writing, the focal could be the problem ? 35mm on a full frame sensor is quite wide.

 

[ngc1535]

Is this image a single image (not integrated result correct)?
If so, I can see faintly the ring and how it could impact things after image integration.
However, it is a small enough error that a number of things could be at play. Typically the viability of darks or the use of optimization... or a handful of things are at play.

However, you mention you did not even use a dark? I do not understand the comment about not having a dark because you were at ISO 400.

My suggestion is to start from calibration. The number of flats isn't the issue. How the flats are being applied likely is- and darks may be involved.
Get real darks next time you have a session with matching data and see how it goes!

-adam

 

[thibaut]

Thank you for your answer.

The first image is the integrated result after crop, DBE and huge stretch.

I did not use a masterdark because my data are very low noise. Am I doing wrong ? What is optimization supposed to do ?

I took darks, so I am gonna calibrate again, but I don't understand how it could handle this problem.

 

[thibaut]

Is this image a single image (not integrated result correct)?
If so, I can see faintly the ring and how it could impact things after image integration.
However, it is a small enough error that a number of things could be at play. Typically the viability of darks or the use of optimization... or a handful of things are at play.

However, you mention you did not even use a dark? I do not understand the comment about not having a dark because you were at ISO 400.

My suggestion is to start from calibration. The number of flats isn't the issue. How the flats are being applied likely is- and darks may be involved.
Get real darks next time you have a session with matching data and see how it goes!

-adam

I re-integrate with darks, I obtain the same result.

 

[ngc1535]

I would be curious to see if you have the same problem with another lens.
I have experienced difficulties like this with large pieces of glass near to the sensor- and not flattening well.
(This is particularly true for focal reducers.)
If a pedestal (like a bias/dark) is not removed from the flat- it may not work well and give this problem..that is why I mentioned it.
(Maybe I should ask..did you calibrate your Flats with a Flat dark or bias?)

It is a fairly low level error based on your stretch. I don't think I can figure it out without looking at the dark from scratch (and even then... not certain I would know).

-adam

 

[thibaut]

I calibrated flat frames with a superbias. I strictly followed this : https://pixinsight.com/forum/index.php?threads/for-beginners-guide-to-pis-imagecalibration.11547/

I read again this topic : https://www.cloudynights.com/topic/670370-nikon-z6-testing/ ; and I'm almost sure I have the same problem, which seems to be related to my camera.

I took a look to the generated slope during integration of my masterflat, and to the blue channel divided by the green. Look what I have obtained.

According to Mark from cloudynight, to solve this, I have to overexposed both my flat and lights (as those rings could be present in my lights as well...). I'm not 100% sure it is the same issue, but... What do you think ?