Jump to content

Vignetting or a stray light gradient?


Kris N
 Share

Recommended Posts

Hi guys! I am having an issue with my last image, which happens to be the second ever picture I have taken with a dedicated astro camera. I have not collected all the data yet, but after initial stacking I have noticed vignetting or stray light gradient which I can't calibrate out. I have ran 3 different stacks using 3 different sets of flat frames, each one of them exposed differently. I used a tablet as a light source for one of the sets, and a light box for the other 2 sets (t-shirt method for all of them).

 

Set 1: 3.8s 21658 ADU (light box)
Set 2: 4.7s 30000 ADU (tablet)
Set 3: 2.95s 43000 ADU (light box)

 

Here is the image stacked using Set 3 (which appeared to look the best out of 3, also generated in auto mode). It is at an early stage of processing. This is all I have done in PixInsight: background neutralization, colour calibration, scnr green noise removal, dynamic background extraction (subtraction) and also ran the ez denoise script. All at linear stage.

 

1728232152_Integrationvignette43k2.95s(lightbox).thumb.jpg.5531fe8cb1d86ce8dafb883f58389d76.jpg

 

Here are the frames that I used for stacking in PixInsight (all frames taken at gain 100, -5 degrees):
75 light frames (300s)
25 dark frames (300s)
35 flat frames (2.95s - main image)
35 dark flat frames (2.95s - main image)

 

This is the picture of all 3 final stacks for comparison, all at the same stage of processing. The order is from left to right using flats set 1 / 2 / 3. The top row shows a comparison of the left side of the image (dark) and the bottom one shows a comparison of the right side of the image (bright). It appears taking flats at a different exposure / ADU had little effect.

 

1526441500_Vignettecombined.thumb.jpg.726b59cf7e9d97ca2be4090134abe475.jpg

 

This is what the master flat looks like:
 

1236442384_Masterflatstretched43k-lightbox.thumb.jpg.cf1ddf3f1519567f037f4e5636763a9e.jpg

 

My imaging train:
Altair Astro Starwave 102ED-R F7, 714mm
Altair Astro Starwave v3 0.8x Flattener/Reducer (recommended backfocus 61mm)
Altair spacers and Altair Filter Holder Version 2 (actual backfocus 62mm)
Antlia ALP-T Dual Band 5nm Filter- 2"
ASI 2600mc

 

Image acquisition: 
ASIAir

 

My main question is: can this be caused by stray light from sources in the direction of which the telescope is pointing? The reason why I am asking this question is that I did not have such problem when I took my first picture. My optical train was the same (including camera rotation), except from a filter. I used the Altair TriBand instead of Antlia Dual band 5nm. There was only a slight vignetting / gradient in one corner of the image, but nowhere close to this. When I took the first picture the object was overhead meaning the telescope would not be pointing towards any significant light sources. The Horsehead Nebula, on the other hand, would be between 20 - 35 degrees, and there would be some light sources in the direction the telescope was pointing. The picture here shows the path of Horsehead when I am imaging it (not the actual night, just an example) and all the light sources below that. 

 

20220112_174501.thumb.jpg.51048eac84dc846a17d0243f2388abaf.jpg

 

Could this explain why I can't calibrate it out, or is there any other reason for it?

 

Thank you for your advice!

 

Kris

Link to comment
Share on other sites

I've never seen this with my refractor but it does look to me like a Flat issue. Maybe the focuser position has moved from infinity. Do you still have all the optical elements in the train (FR +filter holder with the same filter)?

Link to comment
Share on other sites

Thanks guys! I also thought it was a flat issue, but it may as well be a gradient which the flat is not accounting for. Someone suggested it in Altair fb group.

 

I have only used PixInsight with it's 'standard' settings to calibrate a master flat. I have not used DSS yet. I may give it a go to see if there is any difference. 

 

There is something I did not mention... I am based in London, Zone 3 and light pollution here is quite severe, hard to imagine worse than this to be honest. 😄 The Horsehead's path is in the direction of the city centre's skyglow, I have not taken pictures of that part of sky yet. I guess this could add to the whole problem? It's the first time I am this 5nm Dual Band filter by the way.

Link to comment
Share on other sites

4 minutes ago, Kris N said:

I have only used PixInsight with it's 'standard' settings to calibrate a master flat. I have not used DSS yet. I may give it a go to see if there is any difference. 

Calibrating the flats might not be the problem, but the integration settings. These are the settings I use to make the master flat.

 

Untitled-1.thumb.jpg.08fc13c52e9b86f22a969532f473d5e0.jpg

Link to comment
Share on other sites

1 hour ago, AstronomyUkraine said:

Calibrating the flats might not be the problem, but the integration settings. These are the settings I use to make the master flat.

 

Untitled-1.thumb.jpg.08fc13c52e9b86f22a969532f473d5e0.jpg

 

I have been using Weighted Batch Preprocessing Script to generate Master Darks and Flats (data from 4 different nights). I only used WBPP to calibrate the light frames. I then then measured them, star alligned and used Image Integration only to generate final stack.

 

I have not changed any WBPP settings relating to image integration of flats. They looked like this:

 

image.thumb.png.dd22e6e1c68499c7e3027c4db0589bc6.png

 

 

Link to comment
Share on other sites

Ok... I am trying something new. I wrapped a tight around the flattener, spacers and the filter holder to eliminate any light leaks. Will give it a go tonight if the weather is good.

 

20220114_115533.jpg.thumb.jpg.70996bab953dc068afb4dddf387a04f0.jpg

 

Here is a picture from last night. Have a look at a street light behind the scope... there is a 'slight' chance of a leak, this could explain the gradient. There are other light sources around. I will see how it goes.

 

20220113_194537.jpg.thumb.jpg.ce4625080fbed87596f98fe634e7ffd2.jpg

 

 

 

Link to comment
Share on other sites

16 hours ago, Greg M said:

What do your individual subs look like prior to calibration, just auto-stretched?

 

Greg, here is a single not calibrated sub.

 

1330279439_singlesub-notcalibrated.thumb.jpg.886c893e5f66fc30a7820201652917e5.jpg

 

And autostretched in Pixinsight:

 

image.png.5de84a722b33bd3d438a00e1bad14fc4.png

 

I have just ran 2 stacks (one with darks and one without) from last night's data and the result was pretty bad. I had the reducer, spacers, filter drawer and the front of camera wrapped up in a few layers of a thick tight so it does not seem to have much to do with a potential light leak coming from that area... The Moon was closer to the target so it probably did not help.

 

Here is a comparison, picture on the left stacked with darks, and picture on the right with no darks, 2hr 15min exposure.

 

image.thumb.png.3c6d0be8ed8f82ba790d77e866a33763.png

 

I am going to take new dark and dark flat frames and make sure there is no light leak when I do it. The red led in the ASIAir is quite strong and I have it mounted not far from the filter drawer, it could possibly cause a leak.

 

 

  • Thanks 1
Link to comment
Share on other sites

Your master Flat does look like it has a red cast coming in from the left. If the light is coming in from the street then as your mount rotates you would expect to see the light move about from frame to frame. If the light is coming from the equipment on the scope then it will stay in a fixed position on each frame regardless of the orientation of the scope. 

 

(I have heard of some cameras getting light pollution from their own power LEDs) 

 

Have you tried stacking without flats or without dark flats. The first should show clear vignetting, the second should fix the vignetting but with a slight black point offset but will show you if the flat darks are influencing the result. 

 

 

  • Thanks 1
Link to comment
Share on other sites

7 hours ago, paul said:

Your master Flat does look like it has a red cast coming in from the left. If the light is coming in from the street then as your mount rotates you would expect to see the light move about from frame to frame. If the light is coming from the equipment on the scope then it will stay in a fixed position on each frame regardless of the orientation of the scope. 

 

(I have heard of some cameras getting light pollution from their own power LEDs) 

 

Have you tried stacking without flats or without dark flats. The first should show clear vignetting, the second should fix the vignetting but with a slight black point offset but will show you if the flat darks are influencing the result. 

 

 

 

Thanks Paul! Shall I include darks in those stacks or discard them and only play with flats / dark flats?

 

I stretched the master flats, dark / dark flat and noticed a similar cast on them, although inverted. The red cast appears to be coming from the right. Should this even be the case, or should the dark files be completely uniform?

 

Here is the master dark (300s):

 

475599590_masterDark_BIN_1_EXPOSURE_300_00s_integration_RGB_VNG1(debayered).thumb.jpg.748cbeb09a625dbbd778b5760d779d19.jpg

 

Here is the master dark flat (2.95s):

 

1678756930_masterDark_BIN_1_EXPOSURE_2_95s_integration_RGB_VNG(debayered).thumb.jpg.83bf391d7a9f4748c893fd9bf2478e60.jpg

 

Link to comment
Share on other sites

I am convinced it is how the flats are produced. IMO your flat master seems to have particularly strong vignette - stronger than your stretched sub image. I would expect to "squint" at the sub image and see how the flat maps too it.

 

Did you do a median flip during the image capture? Which way round the image is rendered will depend on which side of the meridian the reference image is.

 

Your darks don't look like the source of the issue. The glow to the right looks like barely detectable off-sensor "amp" glow which I would not worry about. Darks normally have to be stretched a lot to see any features and since it is a colour camera I can understand a colour cast due to  varying response to the infra red glow. (Perhaps start a new topic if you want 2600MC specific advice).

 

 

  • Thanks 1
Link to comment
Share on other sites

10 hours ago, paul said:

I am convinced it is how the flats are produced. IMO your flat master seems to have particularly strong vignette - stronger than your stretched sub image. I would expect to "squint" at the sub image and see how the flat maps too it.

 

Did you do a median flip during the image capture? Which way round the image is rendered will depend on which side of the meridian the reference image is.

 

Your darks don't look like the source of the issue. The glow to the right looks like barely detectable off-sensor "amp" glow which I would not worry about. Darks normally have to be stretched a lot to see any features and since it is a colour camera I can understand a colour cast due to  varying response to the infra red glow. (Perhaps start a new topic if you want 2600MC specific advice).

 

 

 

Paul, I did not do the flip, it was all photographed on the same side of meridian. I created the whole new set of calibration frames to ensure zero light leak, flats were taken in auto mode. The new master flat now looked like this and vignette looks more like the one in the individual sub:

 

891898441_masterflat2_86s.thumb.jpg.28bfb71ed5b96a45b8c365135507b316.jpg

 

I stacked the same data using the new set of calibration frames, slightly changed the setting in DBE in Pixinsight and the result was better. Below there is a comparison of the old stack (left) and the new one (right), both 6hrs15mins of exposure taken with no Moon. I can still see a slight outline of the gradient, but the image looks better.

 

1621907029_integrationnights1-4comparison.thumb.jpg.79412445120094e916b8309db007dc8e.jpg

 

 

I then added another 2hrs45mins taken when the Moon was out (on the same side as the object) and the gradient got worse again. Here is another comparison. The image on the left is the 6hrs15mins stack, the image on the right is a 9hr stack (including the 2hrs45mins when the Moon was out), I used the same new set of calibration frames for both.

 

773385732_integrationnights1-4vs1-6(withMoon).thumb.jpg.f23285cd611700d4804986931a90e293.jpg

 

 

Whatever is happening here it almost feels like 'unwanted' light is a big contributor to the problem.

 

Please have a look at this picture, I took it over 3 different nights when the object was close to the zenith. The setup was all the same apart from the filter. The telescope was pointed overhead, away from streetlights etc. There was only a slight vignette in one corner, I had to crop it just a bit. It was nowhere close to what I am getting now.

 

1372603132_2ndprocessHLVG.thumb.jpg.2f5dfd62c9f83f6e664dd54d8e7da79a.jpg

 

I have now ordered a dew shield extension to see what effect it will have on stray light and if it helps. 

 

Link to comment
Share on other sites

Your new flat certainly looks improved. If the Flat is good and matches your original imaging conditions you should not be able to see evidence of the vignetting.  If the flat was created under different circumstances, if the focus has shifted between imaging and when the flat is produced then the fall off of light will change and you will see it in the image. Also, is it possible the camera rotated?

 

When taking the flat images do you retract the dew shield? I suspect you may need to do that.

 

Regarding the light leakage, perhaps you are getting some light across primary from the street - a longer dew shield might help.

 

If this is the first time of using the filter - is it in the right way, some have different coatings on each side. Perhaps try the filler on a target away from the LP and see if the problem goes away?

Link to comment
Share on other sites

14 hours ago, paul said:

Your new flat certainly looks improved. If the Flat is good and matches your original imaging conditions you should not be able to see evidence of the vignetting.  If the flat was created under different circumstances, if the focus has shifted between imaging and when the flat is produced then the fall off of light will change and you will see it in the image. Also, is it possible the camera rotated?

 

When taking the flat images do you retract the dew shield? I suspect you may need to do that.

 

Regarding the light leakage, perhaps you are getting some light across primary from the street - a longer dew shield might help.

 

If this is the first time of using the filter - is it in the right way, some have different coatings on each side. Perhaps try the filler on a target away from the LP and see if the problem goes away?

 

Paul, I made sure imaging train was all intact when I took the new set of flats. Camera rotation and focus were the same.

 

I always extend the dew shield for flats, the same as when I take images. Should this be retracted? I did not know...

 

I can only mount a filter one way (filter drawer) which I believe it's how it should be used. I will try a dew shield extension soon, I wonder if it helps. It would be great.

 

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...