Back to forums

Flats wrong orientation during calibration creating dust motes?

Pleiades Astrophoto PixInsight ZWO ASI6200MM Pro Processing Calibration Deep Sky
Dark Matters Astrophotographyandrea tasselliAshraf AbuSaraJon Rista
•25 replies•575 views
Ashraf AbuSara avatar
Ashraf AbuSara avatar
I have run into this very bizarre flat calibration issue in Pixinsight. 

Just some background, I imaged a target with my Luminance filter on my 6200mm pro last night, 60s each frame, for a total of 250 frames, at gain 100, default ASIair offset (50), -5c, using the AG Optical Convergent FA12. I captured 30 flats using a Primalucelab Gitto 430mm flat field frame generator. I set the brightness of the flat panel to take 1.4s Luminance Flat frames to achieve just over 20000 ADUs. 



I calibrate my flat frame with bias frames (being the shortest exposure possible for my 6200mm pro). 

So after calibrating and stacking all 250 light frames in WBPP, I get this:



So multiple issues with this frame. There was undercalibration of the flat frame patterns. The second more strange issue is that not only was the dust mote on the right not calibrated out, but the dust mote on the left is something that did not exist at all in the light frames or the flat frame. 

I really was sctratching my head, thinking maybe I got that dust later in the imaging session, but blinking through the images did not reveal that much. And then I realized, if I flip flat frame 180 degrees, the same dust mote on the right falls exactly in position of that *new* non existent dust mote on the left. 

I then further looked into this. My imaging session of 250 frames, had 184 frames before the meridian flip, and 66 frames after the meridian flip. My theory was that the flat frame calibrated in the wrong orientation for the second half of the light frames, or something to that effect. To test that out, I just stacked the calibrated first 184 frames before the meridian flip seprately. This was the result:



The image is still under corrected. The large ring artifacts are still present, but the dust mote that was created to the left upper corner completely disappeared. 

I have now taken flat darks to match my L 1.4s flats, and will try to stack and calibrate with WBPP again to see if it will fix the undercorrection, but at this point I do not have any faith in what is going on with my flats. 

Any idea what the problem is here?
Helpful Insightful Engaging
andrea tasselli avatar
andrea tasselli avatar
Darkflats are, normally, the answer for this sort of issues. And master darks with no bias subtraction and no correction in PI (make sure you unchecked that in WBPP). Weird what it did with that flat-fielding since you shouldn't flip the flat as there is no reason for it. OTOH I won't use WBPP even under torture so there you are…
Ashraf AbuSara avatar
Ashraf AbuSara avatar
andrea tasselli:
Darkflats are, normally, the answer for this sort of issues. And master darks with no bias subtraction and no correction in PI (make sure you unchecked that in WBPP). Weird what it did with that flat-fielding since you shouldn't flip the flat as there is no reason for it. OTOH I won't use WBPP even under torture so there you are...

I have used PI exclusively since I started AP. It has worked so well accross so many setups. This is a different beast though. I am really not sure what is going on. I tried Flat darks instead of bias frames to calibrate the flats and to no avail. Still the same problem. It almost looks like those two specific dust motes are worse in the calibrated image than the unflattened one. 

I will start from scratch, and maybe post some master files to the PI forums and try to pick their brains.
andrea tasselli avatar
andrea tasselli avatar
Just be sure you got the same offset and gain across the whole thing. I know it is unlikely but sometime things slip…
Ashraf AbuSara avatar
Ashraf AbuSara avatar
andrea tasselli:
Just be sure you got the same offset and gain across the whole thing. I know it is unlikely but sometime things slip...

Will do. I am pretty sure both are the same (gain 100, offset 50) but I will tripple check later today.
José Carlos Rodríguez avatar
José Carlos Rodríguez avatar
Ashraf AbuSara:
I have run into this very bizarre flat calibration issue in Pixinsight. 

Just some background, I imaged a target with my Luminance filter on my 6200mm pro last night, 60s each frame, for a total of 250 frames, at gain 100, default ASIair offset (50), -5c, using the AG Optical Convergent FA12. I captured 30 flats using a Primalucelab Gitto 430mm flat field frame generator. I set the brightness of the flat panel to take 1.4s Luminance Flat frames to achieve just over 20000 ADUs. 



I calibrate my flat frame with bias frames (being the shortest exposure possible for my 6200mm pro). 

So after calibrating and stacking all 250 light frames in WBPP, I get this:



So multiple issues with this frame. There was undercalibration of the flat frame patterns. The second more strange issue is that not only was the dust mote on the right not calibrated out, but the dust mote on the left is something that did not exist at all in the light frames or the flat frame. 

I really was sctratching my head, thinking maybe I got that dust later in the imaging session, but blinking through the images did not reveal that much. And then I realized, if I flip flat frame 180 degrees, the same dust mote on the right falls exactly in position of that *new* non existent dust mote on the left. 

I then further looked into this. My imaging session of 250 frames, had 184 frames before the meridian flip, and 66 frames after the meridian flip. My theory was that the flat frame calibrated in the wrong orientation for the second half of the light frames, or something to that effect. To test that out, I just stacked the calibrated first 184 frames before the meridian flip seprately. This was the result:



The image is still under corrected. The large ring artifacts are still present, but the dust mote that was created to the left upper corner completely disappeared. 

I have now taken flat darks to match my L 1.4s flats, and will try to stack and calibrate with WBPP again to see if it will fix the undercorrection, but at this point I do not have any faith in what is going on with my flats. 

Any idea what the problem is here?

Hello, I have the same problem as you regarding halos. I don't understand what you mean by saying that the image is not sufficiently corrected. Can you explain it to me? How do you correct it?

I attach a shot of 600s Ha of the Dolphin Head Nebula (SH2-308) taken from my house, in Malaga Capital, with a bortle of 8-9.
You can see the kind of donut in the photo that I cannot remove with processing.

Equipment:
Tube: OS RH200 AT
ESATTO ROBOTIC FOCUS 3.5” LP
QHY600 PH-M SBFL CAMERA
Ultra high speed filters 3.5/4 nm f/3: optimized for CMOS (H-alpha/O-III/S-II)
FILTER HOLDER WHEEL: QHYCFW3XL
Capture program: NINA
To say that I have the problem with all the filters.


Thank you

Link to full photo in fits
Jon Rista avatar
Jon Rista avatar
This looks like faint, diffuse internal reflections. The concentric rings, could be things within your image train, potentially even within the scope, that are reflecting light onto the sensor. It could either be oblique angles and non-blackened edges somewhere, or might be some reflection deeper in the image train, bouncing back out the scope, then off of something else and back to the sensor. 

If you take the camera off, and look through the scope from the camera's perspective, with the pupil brightly illuminated…do you see anything reflecting. It may not be bright, but if anything shows up, it might be contributing to the problem. Someone recently had a similar problem and started a thread on CN. They did some blackening of some of their image train, and resolved some of the issues. They may still have some reflections from within the scope.
Helpful
Dark Matters Astrophotography avatar
Dark Matters Astrophotography avatar
You are all wrong. smile 

OP and other affected people, do me a favor. Take sky flats. The ring pattern will go away. The problem isn't with your optics, it is with the panel you use to take flats being unable to illuminate properly to capture the effect you see there. 

If you need a method to do this, both NINA and Voyager have automated sky flat routines. You do not need to go outside with T-shirts or any of that. Just set it up to capture the sky flats either before or after your imaging session.

Once you do this and test this, come back and let me know. Also, stop using panels. They are nowhere near as good as the sky. Trust me on this one.

-Bill
Helpful
Jon Rista avatar
Jon Rista avatar
You are all wrong.

Dark Matters Astrophotography avatar
Dark Matters Astrophotography avatar
@José Carlos Rodríguez is affected by improper or double bias/dark flats though, or something else in the calibration process.

The OP needs sky flats and to test the efficacy of them on new frames to rule out the dust motes moving around. My solution here was not for the motes, it was for the rings.
Jon Rista avatar
Jon Rista avatar
@José Carlos Rodríguez is affected by improper or double bias/dark flats though, or something else in the calibration process.

The OP needs sky flats and to test the efficacy of them on new frames to rule out the dust motes moving around. My solution here was not for the motes, it was for the rings.

Do sky flats always correct those kinds of rings? I had a similar problem with my RC and I did switch to sky flats, but was only able to entirely correct my issue once I blackened reflecting parts (mostly spacers) in my image train.
Well Written Insightful Respectful Concise Engaging
Dark Matters Astrophotography avatar
Dark Matters Astrophotography avatar
Jon Rista:
@José Carlos Rodríguez is affected by improper or double bias/dark flats though, or something else in the calibration process.

The OP needs sky flats and to test the efficacy of them on new frames to rule out the dust motes moving around. My solution here was not for the motes, it was for the rings.

Do sky flats always correct those kinds of rings? I had a similar problem with my RC and I did switch to sky flats, but was only able to entirely correct my issue once I blackened reflecting parts (mostly spacers) in my image train.



Yes, on mirror systems. The E160ED is notorious for folks getting one and using a panel for flats. I never used a panel with mine, as when I got it I was already on the automated sky flats train. Many people contacted me with rings in their data after calibration and stacking and as soon as they used a DBE or ABE they immediately saw rings. I told them to use the sky for flats and the problem went away immediately.

Now some RC's, especially the older GSO ones did have a baffle issue. If the sky flats do not completely correct the problem that was going to be my next piece of advice. I did not look up the scope the OP was using, but the first line of defense here is sky flats. Usually adapters will put one ring in an image, not more than one. When I saw it was multiple rings, that screams flats issue to me.

If it was one, very likely a shiny adapter.

-Bill
Helpful Insightful Engaging Supportive
Dark Matters Astrophotography avatar
Dark Matters Astrophotography avatar
I should explain the one ring thing. The one ring would not be spaced as well and far apart as the ones the OP shared. You could have multiple shiny adapters and they would all impact the image, they just would do so fairly close together. If you look you will see the rings are very far apart in terms of their overall size and placement.

That is why I would try the sky flats first.
Ashraf AbuSara avatar
Ashraf AbuSara avatar
Thanks Bill. I will give sky flats a try. It would be disappointing if such an expensive flat panel ended up being the problem all along.
Well Written
andrea tasselli avatar
andrea tasselli avatar
Nothing beats sky flats, followed by dome flats…
Rodrigo Roesch avatar
Rodrigo Roesch avatar
Hello
where did you take your flats? I have had the same problem before and the issue was I was taken my flats in a room with a lot of light, so there was light leak in the camera and telescope creating the rings and  the bright halos. 
So try to retake the flats in a dark room
Helpful Concise
Dark Matters Astrophotography avatar
Dark Matters Astrophotography avatar
Rodrigo Roesch:
Hello
where did you take your flats? I have had the same problem before and the issue was I was taken my flats in a room with a lot of light, so there was light leak in the camera and telescope creating the rings and  the bright halos. 
So try to retake the flats in a dark room



Possible, but using the sky would eliminate this entirely, and requires no physical gear to do. Just software to set the exposure time correctly and snap/dither the flats. Be sure it dithers/slews a little between each flat so if stars get in them, they will be removed when you stack the calibrated flats.

-Bill
Concise
andrea tasselli avatar
andrea tasselli avatar
Sky flats are supposed to be taken at dusk, so no chance of getting stars if properly pointing the scope.
Dark Matters Astrophotography avatar
Dark Matters Astrophotography avatar
andrea tasselli:
Sky flats are supposed to be taken at dusk, so no chance of getting stars if properly pointing the scope.



They can be taken at dawn as well. I have done this for almost a year now at the observatory with zero issues.
Well Written
andrea tasselli avatar
andrea tasselli avatar
I'm never up at dawn. Too tired to take flats anyway.
Dark Matters Astrophotography avatar
Dark Matters Astrophotography avatar
andrea tasselli:
I'm never up at dawn. Too tired to take flats anyway.



I don't get up to take mine. Voyager does it for me, every night the scope runs it takes flats before it runs its goodnight routine and restarts itself for the next night. I could disappear for months and the scope would continue to work on the stuff I have loaded without me doing anything.
andrea tasselli avatar
andrea tasselli avatar
To each his/her own…
José Carlos Rodríguez avatar
José Carlos Rodríguez avatar
Thank you all very much for your interest and contributions.
I'll try the Sky Flats and I'll let you know.
Thank you
Well Written Respectful
Dark Matters Astrophotography avatar
Dark Matters Astrophotography avatar
José Carlos Rodríguez:
Thank you all very much for your interest and contributions.
I'll try the Sky Flats and I'll let you know.
Thank you



The error you showed looked more like double bias in the calibration process.. Nonetheless you should use sky flats at well. 

Everyone should. You take lights against the sky. Taking flats against the same sky, is the best way.
AdrianC. avatar
AdrianC. avatar
Never took sky flats, only panel flats. I usually reduce the illumination so the flat will have up to 5-6 sec exposure. Don't need bright source that could cause reflections, but even illumination. Perhaps that is why some report better flats with sky flats …