weird vignetting - assuming it is user error

[joshtb86]

hi i am getting these weird vignetting after gradient correction, even though i took appropriate flats with NINA - all my images are acquired with NINA



used the skyflat wizard to take these



this is a stacked image with flats darks and biases

I am using the qhy268c

ive attached a screenshot of the extreme auto stretch after stacking, then after gradient correction with default settings and simplified gradient set to 1



below is a link to the stacked tiff. i am re stacking - i noticed i still have some artifacts so i changed settings and am restacking



if anyone can assist taht would be great



did i screw up on the flats?

ive noticed sometimes this happened on older sets of data where i didnt take correct flats, but i did this time so i am not sure what happened

also in the workflow - i may be doing something outdated when researching it
i am using bias - which ive read is unnecessary with a modern CMOS
i am optimizing the darks when i calibrate my light exposures - this may be incorrect since they are the same exposure time and temperature of the lights
the flats were calibrated with darks and biases - darks optimized in this

i think i im subtracting bias 3 times with this and read i probably do not need bias frames with a modern cmos sensor like the imx571


thanks so much



https://drive.google...bL6?usp=sharing

 

[joshtb86]

so i reprocessed without bias and recalibrated all my images and still get it

i did not use bias - only flats and darks



im noticing im getting it after stacking - it really isnt noticeable in individual light frames



does this help troubleshoot?

 

[fredvanner]

Exactly how are you calibrating your flats?

 

[cloudbait]

so i reprocessed without bias and recalibrated all my images and still get it

i did not use bias - only flats and darks



im noticing im getting it after stacking - it really isnt noticeable in individual light frames



does this help troubleshoot?

Flats have to be calibrated. That is usually with bias frames, but sometimes with darks. Have to use one or the other, or the offset in the resulting flat will produce just the sort of result you're seeing.

 

[joshtb86]

Exactly how are you calibrating your flats?

With darks only, optimized

 

[cloudbait]

With darks only, optimized

Would suggest you use bias frames, unless you have one of the fairly uncommon sensors that require darks, in which case I'd strongly suggest proper darks matched to the flats.

 

[fredvanner]

With darks only, optimized

What exposure were the darks? What exposure were the flats? If the difference is more than about 10% you need either flat darks or bias frames (if your flats are short enough exposure).

 

[joshtb86]

Flats have to be calibrated. That is usually with bias frames, but sometimes with darks. Have to use one or the other, or the offset in the resulting flat will produce just the sort of result you're seeing.

Flats have to be calibrated. That is usually with bias frames, but sometimes with darks. Have to use one or the other, or the offset in the resulting flat will produce just the sort of result you're seeing.

I had been calibrating flats with darks and bias but notice this the last few times

Maybe I will need to retake the bias frames

They also need to be taken in relative darkness correct?

 

[fredvanner]

... dark optimisation is something to be used as a last resort.

 

[joshtb86]

What exposure were the darks? What exposure were the flats? If the difference is more than about 10% you need either flat darks or bias frames (if your flats are short enough exposure).

Darks were 120 seconds flats were around .07 seconds, qhy268c

 

[fredvanner]

I had been calibrating flats with darks and bias but notice this the last few times

You need one or the other, not both.

 

[fredvanner]

Darks were 120 seconds flats were around .07 seconds, qhy268c

Don't even think of optimisation with a difference like this.

 

[joshtb86]

Don't even think of optimisation with a difference like this.

So I should recalibrate the flats with the darks without optimization?

 

[joshtb86]

Could a full moon have done something like this?

 

[fredvanner]

So I should recalibrate the flats with the darks without optimization?

Calibrate 0.07s flats with 120s darks??? Absolutely not.
These darks are useless for calibrating your flats. Calibrate them with bias only. But you should also in future attempt to capture flats with longer exposure (at least 1 - 2 s).

 

[fredvanner]

Could a full moon have done something like this?

With such dramatically unsuccessful flat calibration it is hard to tell.

 

[joshtb86]

Calibrate 0.07s flats with 120s darks??? Absolutely not.
These darks are useless for calibrating your flats. Calibrate them with bias only. But you should also in future attempt to capture flats with longer exposure (at least 1 - 2 s).

Thanks for your help. I'm using the sky flat wizard in nina and that's what it's exposing for the appropriate adu

I wasn't having issues before until the last couple of recent targets

 

[fredvanner]

I'm using the sky flat wizard in nina

Are you using true (night-time) sky flats or "twilight" sky flats? If it is the latter, perhaps you need to take your flats when the sky is a bit darker.

 

[joshtb86]

Are you using true (night-time) sky flats or "twilight" sky flats? If it is the latter, perhaps you need to take your flats when the sky is a bit darker.

it was around 4pm imaging away from the sun if i remember correctly sunset was at 6. is this too bright?

would it be easier to just use a flat panel for them?

 

[joshtb86]

im uploading a single light, dark and my masterbias/superbias frames here if anyone can take a look

shared for cloudy nights - Google Drive

drive.google.com