Jump to content

Banner.jpg.b89429c566825f6ab32bcafbada449c9.jpg

Help! - Flats causing circular concentric rings


Recommended Posts

Something's up with my flats.  I'm struggling with circular concentric rings in the image after stacking and it's not just happening on one scope.  I've experienced this over my last two set ups, with the 150P and now my EdgeHD 8.  To try and remedy this I have taken flats with artificial light at different intensities and sky flats  but following stacking they all have the same result.  I'm a bit puzzled as I haven't had this issue previously on my RedCat51 or 200P.

I take my flats with the ASIAir Pro and the auto exposure routine and that has worked well in the past.  Recent exposures with a flat panel, white t-shirt, the L-eXtreme and the 2600MC at Gain 100 were 8.3s in length.  It's really got me scratching my head.  If I stack without flats the image is fine, I've tried lots of combo's in DSS and with Siril too all with the same result, as soon as flats are added the rings are back. 

I am probably doing something daft but I keep going around in circles.  I've attached an example image with flats and without and my master flat file.  Any thoughts or similar experiences would be greatly appreciated.

2022-03-26T23.32.03.png

2022-03-26T23.41.22.png

2022-03-26T23.56.01.png

Link to comment
Share on other sites

3 hours ago, bottletopburly said:

What calibration frames are you taking?

I've tried all combinations.  I have a library of darks an bias frames and have substituted them in and out of different stacks in a trial and error way to see if there are any changes.  Each time the flats are the cause of the rings.  I've also stretched and checked my darks and bias and there are no unusual artefacts.

I've shot some more flats this morning without the filter to see if that makes a difference.  I'm wondering if I am getting a strange reflection from something in the imaging train.  Trial and error is the only way I guess.

The rings on the blue channel and red are particularly prominent which is why I am wondering if it's related to the L-eXtreme.

Blue.png

Red.png

Link to comment
Share on other sites

I had this recently! Vlaiv helped me work out that it was caused by using Median-kappa-sigma averaging in my lights. I suggest what he suggested- use "average" for all your files, light, dark, bias, flat etc. Then try other methods. I find normal Kappa-sigma (as opposed to median kappa sigma) works quite nicely.

 

Link to comment
Share on other sites

On 27/03/2022 at 12:42, pipnina said:

I had this recently! Vlaiv helped me work out that it was caused by using Median-kappa-sigma averaging in my lights. I suggest what he suggested- use "average" for all your files, light, dark, bias, flat etc. Then try other methods. I find normal Kappa-sigma (as opposed to median kappa sigma) works quite nicely.

 

Unfortunately this hasn't resolved my issue.  A bit of internet hunting (cloudy nights) and I've found similar issues related to reflections in the imaging train.  Potentially related to the reducer, so took the reducer apart and DIY flocked the internals.  I'll take some more flats and see if it improves things.

Link to comment
Share on other sites

7 hours ago, Zummerzet_Leveller said:

Unfortunately this hasn't resolved my issue.  A bit of internet hunting (cloudy nights) and I've found similar issues related to reflections in the imaging train.  Potentially related to the reducer, so took the reducer apart and DIY flocked the internals.  I'll take some more flats and see if it improves things.

Also be aware that DSS will use the already calibrated master files once they've been made, so if you re-stack flats or bias or dark frames it may not actually recreate the master file it uses to stack, but just use the one it made previously.

Try deleting your old master files and re-stacking with all settings set to average (no median)

Link to comment
Share on other sites

5 hours ago, pipnina said:

Also be aware that DSS will use the already calibrated master files once they've been made, so if you re-stack flats or bias or dark frames it may not actually recreate the master file it uses to stack, but just use the one it made previously.

Try deleting your old master files and re-stacking with all settings set to average (no median)

Yep. Definitely my flats, flocking has not helped either.  I'm beginning to wonder if the ASIAir is calculating too long an exposure or something, I did double white t-shirt sky flats to no avail.  Deleted all master files and stacked in Average, the result being the bullseye image (tried all different stacking iterations).  Stacked without flats and get the second image.

I think I am going to have to create some synthetic flats.

All Calib_Ave_Stack.png

Ave_Stack_No_Flats.png

Link to comment
Share on other sites

I had a very similar issue when I started using my ASI2600mc and stacking in APP. Mabula fairly quickly worked out that it was because APP was saving the master flats as 16bit and for this camera it must be 32bit. Now I dont know anything about DSS but maybe check it out.

The APP forum thread is here- https://www.astropixelprocessor.com/community/main-forum/integration-causing-nasty-rings/#post-16488

Take alook and you'll see that my flats looked very similar (infact i'd say identical) to yours with concentric rings and the stacked image looked very similar to yours too. Saving masters as 32bit files immediately resolved it and Mabula then made 32bit masters the default in the next APP uprev.

Hope this helps.

 

Edited by mackiedlm
Link to comment
Share on other sites

58 minutes ago, mackiedlm said:

I had a very similar issue when I started using my ASI2600mc and stacking in APP. Mabula fairly quickly worked out that it was because APP was saving the master flats as 16bit and for this camera it must be 32bit. Now I dont know anything about DSS but maybe check it out.

The APP forum thread is here- https://www.astropixelprocessor.com/community/main-forum/integration-causing-nasty-rings/#post-16488

Take alook and you'll see that my flats looked very similar (infact i'd say identical) to yours with concentric rings and the stacked image looked very similar to yours too. Saving masters as 32bit files immediately resolved it and Mabula then made 32bit masters the default in the next APP uprev.

Hope this helps.

 

Thanks for this David and a very interesting thread.  Indeed the master files are saved as 16bit not 32bit.  I have Siril which now supports 32bit so will try to stack in this and see if it improves things, just have to remember how to stack in Siril with the scripts, it's not quite as user friendly as DSS.

What is confusing me is that I have had many successful stacks up until now using the 2600MC on my Redcat51 and 200P using DSS.  I'm trying to figure out what may have changed.  I'm going to retake some flats on the RedCat and restack some older data to make sure that something hasn't gone awry in the capture process.  I've just finished retaking a set of darks so I can hopefully eliminate them from the mix.

I do have some slightly reflective elements in the tube from the corrector to reducer on the HD8 that I may have to paint with some stove paint to eliminate them too.  I'm keen to try the software route first before I put my fine art skills to the test on a pricey piece of kit!

It's all rather annoying as I took advantage of 5 (yes really) clear nights and have captured a bucket load of data.

  • Like 1
Link to comment
Share on other sites

Yes I was exactly the same - 3 months into the 2600mc without an issue then suddenly ..... garbage! Mabula was never able to explain that to me but from the day I changed to 32bit masters all was good..  My own, totally ignorant and probably nonsense, idea is that something was different in the skys when it showed up. Too bright, more LP, lower down higher up????? but I never bothered to try to do the analysis to work it out as it was resolved. 

Thats a bummer, but if you have the raw flats you should be able to rebuilt working masters.

Good luck and please do post back here on how you get on.  

  • Like 1
Link to comment
Share on other sites

On 31/03/2022 at 19:07, mackiedlm said:

Good luck and please do post back here on how you get on

Well it turns out keeping things 32bit in Siril has worked.  I've got to redo my flats as they are not correcting properly, probably me playing too much to try different options.  I have taken flat darks so might switch them over in the processing instead.  But no more bullseye!

It seems to have worked well, the Siril process is a bit more convoluted and it created a mass of temporary files (132Gb...).  I am at a loss to why things suddenly changed.  I wonder whether it is to do with the focal length and oversampling.  Who knows.  Now to work out whether you can software bin X2 in Siril.

 

2022-04-02T13.20.46.png

  • Like 2
Link to comment
Share on other sites

3 hours ago, Zummerzet_Leveller said:

Well it turns out keeping things 32bit in Siril has worked.  I've got to redo my flats as they are not correcting properly, probably me playing too much to try different options.  I have taken flat darks so might switch them over in the processing instead.  But no more bullseye!

It seems to have worked well, the Siril process is a bit more convoluted and it created a mass of temporary files (132Gb...).  I am at a loss to why things suddenly changed.  I wonder whether it is to do with the focal length and oversampling.  Who knows.  Now to work out whether you can software bin X2 in Siril.

 

2022-04-02T13.20.46.png

Thats fantastic! I'm delighted it worked. Yes I dont understand why the sudden change - happened to me too. Anyway, I look forward to seeing the final image.

  • Thanks 1
Link to comment
Share on other sites

On 02/04/2022 at 18:02, mackiedlm said:

I look forward to seeing the final image.

Thanks for the help David.  My image of Lowers nebula was a bit soft so I'm not happy with it but I'm pleased with this Thor's helmet considering it was taken low down in the murk and is only 110mins of exposures.  Looks like it calibrated well, starting to get the hang of Siril stacking too.  It's actually not that bad once you get used to it, and delete all the temporary files!

Final_V2_low.jpg

  • Like 1
Link to comment
Share on other sites

  • 1 month later...

I'm having the problem reported in this thread. See the image below. The frames were stacked in the DSS. Is the problem really in the flat frames?

716763636_ngc3201-crculos.jpg.f360de903618765f91218ffe3f756a2f.jpg

Thank you very much in advance.

Link to comment
Share on other sites

2 hours ago, Adriano said:

I'm having the problem reported in this thread. See the image below. The frames were stacked in the DSS. Is the problem really in the flat frames?

716763636_ngc3201-crculos.jpg.f360de903618765f91218ffe3f756a2f.jpg

Thank you very much in advance.

I'd definitely try stacking in Siril instead. What camera do you have?

Link to comment
Share on other sites

3 hours ago, Zummerzet_Leveller said:

I'd definitely try stacking in Siril instead. What camera do you have?

I used a QHY8L to make the image above. It was stacked in DSS. But I tried stacking in Siril too and the result was not good. The image below was stacked in Siril.

resultado_3-siril.thumb.jpg.759619d98f411511d38fa017b79d6608.jpg

 

Link to comment
Share on other sites

13 hours ago, Adriano said:

I'm having the problem reported in this thread. See the image below. The frames were stacked in the DSS. Is the problem really in the flat frames?

716763636_ngc3201-crculos.jpg.f360de903618765f91218ffe3f756a2f.jpg

Thank you very much in advance.

Try just stacking only the light just to see if the effect appears , the re stack with darks until you can pinpoint which frames are causing issue 

  • Like 1
Link to comment
Share on other sites

7 hours ago, bottletopburly said:

Try just stacking only the light just to see if the effect appears , the re stack with darks until you can pinpoint which frames are causing issue 

2nd this. Also check your lights to see if there are any frames with thin cloud or dewing of anything in your imaging set up that cause a misty frame. DSS is not very good at checking frame quality. Siril has the edge here with the histogram stretch function. Really highlights poor frames.

Link to comment
Share on other sites

On 14/05/2022 at 06:39, bottletopburly said:

Try just stacking only the light just to see if the effect appears , the re stack with darks until you can pinpoint which frames are causing issue 

 

On 14/05/2022 at 13:56, Zummerzet_Leveller said:

2nd this. Also check your lights to see if there are any frames with thin cloud or dewing of anything in your imaging set up that cause a misty frame. DSS is not very good at checking frame quality. Siril has the edge here with the histogram stretch function. Really highlights poor frames.

Thanks for your advice. I will try to do what you said.

  • Like 1
Link to comment
Share on other sites

  • 3 weeks later...
On 13/05/2022 at 21:38, Adriano said:

I'm having the problem reported in this thread. See the image below. The frames were stacked in the DSS. Is the problem really in the flat frames?

716763636_ngc3201-crculos.jpg.f360de903618765f91218ffe3f756a2f.jpg

Thank you very much in advance.

Here, even more than in the OP's post, the rings are not only distinguished by brightness but by colour.  Not only that: the colours seem to be organized somewhat like an opened-up white light spectrum. (Or perhaps a sequence of RGB,RGB etc?)

Could this be connected with the debayering process or some other aspect of colour processing? Or is it in the captured data due to some optical effect?

What would happen if you took your debayered master flat and discarded its colour information then restored it to RGB format? It would be monochrome and identical in all channels but, since it would be in RGB format, the stacking software should accept it.

I certainly think that this looks, above all, like a colour problem but is it optical or electronic in origin? No idea.

Olly

Edit: I'd certainly want to look at the master flat colour-by-colour to compare the rings.

 

Edited by ollypenrice
  • Like 1
Link to comment
Share on other sites

On 02/04/2022 at 14:31, Zummerzet_Leveller said:

Well it turns out keeping things 32bit in Siril has worked.  I've got to redo my flats as they are not correcting properly, probably me playing too much to try different options.

...

Still some vignetting so the flats definitely need redoing IMO.

Link to comment
Share on other sites

On 01/06/2022 at 08:11, ollypenrice said:

Here, even more than in the OP's post, the rings are not only distinguished by brightness but by colour.  Not only that: the colours seem to be organized somewhat like an opened-up white light spectrum. (Or perhaps a sequence of RGB,RGB etc?)

Some might regard this proposal as heretical but here goes. If it is the only data you have there may be no alternative.

Three things (at least) are obvious in your image. The rings are coloured, they are wide and they are circular.

That they are coloured indicates that if you process them, you should treat the R, G and B channels separately: split the image into three, process independently, and recombine at the end.

That the rings are circular indicates that the two-dimensional Fourier transform of the images will show strong symmetrical structure.

That the rings are wide indicates that there will be strong structure at low frequencies in the Fourier transform.

Therefore, try taking the Fourier transform of each image and then zero out all but the strong low-frequency structure. If when you inverse transform the result you end up with broad circles in the same position as the originals, and not much else, you know you have selected the correct Fourier components.  Use that inverse FT image as a flat.

Completely screws any photometric accuracy and may add other artifacts as well, but it will get rid of the aesthetically displeasing rings.

 

 

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. By using this site, you agree to our Terms of Use.