Reverse vignetting?

[Larry 1969]

I'm not even sure there is such a thing but I've been having trouble getting my flats to correct properly. The trouble seems random.
Some of my images are WAY over-corrected and I think I found out why by accident. I recently made a little animation of comet 67P out of 17 light frames stitched together with Google Photos so I stretched each image separately, with no calibration frames and I notices "reverse vignetting" on the left of the image. It's a bit easier to see in the animation but that's too large a file to upload here so here's a single frame that shows the problem (you have to look closely). The left upper and lower corners are brighter than the center and the right corners are darker (normal vignetting)...

I can't imagine what could cause that. Light entering my camera sensor from somewhere other than my scope?
How else could light hit my sensor outside my imaging circle???

Thanks for looking!

Larry

[metastable]

Is there a street light / outdoor light near your telescope?

[bobharmony]

Larry - when I looked at the image on the PC I use for processing (and I know how unreliable that is) it looked like there was a ring of "lightness" going all around the edges of the frame, with a darker ring appears in all the corners, more noticeable in the right hand corners but also present on the left side. I looked at the image on my tablet to check what I was seeing and it showed me the same thing. It makes me think that what you have been considering as over-corrected flats may actually be under-correcting in that they aren't removing all of the light ring you see in your uncalibrated comet subs. This is a assuming that there wasn't some external light source that caused the issue you see here.

To explore this further it would be helpful to know more about how you make your flats, light source, exposure, gain etc. I looked up the 26C and read it is a 16-bit camera. That being said the peak ADU for your flats should be about 32K to work most effectively. If it is significantly less, you could be undercorrecting, and if higher, you could be overcorrecting.

Of course someone may come along in 5 minutes and say "I've seen that before, it is caused by (insert your best guess here), get rid of the cause and you will be good to go".

These issues are some of the peskiest we deal with in the AP world, but getting them sorted out is well worth the effort in getting better future results and much less aggravation!

Bob

[Larry 1969]

Is there a street light / outdoor light near your telescope?

Not really. There is a second story inside light on at times but not in the direction of imaging. Thanks.

Larry

[Larry 1969]

Larry - when I looked at the image on the PC I use for processing (and I know how unreliable that is) it looked like there was a ring of "lightness" going all around the edges of the frame, with a darker ring appears in all the corners, more noticeable in the right hand corners but also present on the left side. I looked at the image on my tablet to check what I was seeing and it showed me the same thing. It makes me think that what you have been considering as over-corrected flats may actually be under-correcting in that they aren't removing all of the light ring you see in your uncalibrated comet subs. This is a assuming that there wasn't some external light source that caused the issue you see here.

To explore this further it would be helpful to know more about how you make your flats, light source, exposure, gain etc. I looked up the 26C and read it is a 16-bit camera. That being said the peak ADU for your flats should be about 32K to work most effectively. If it is significantly less, you could be undercorrecting, and if higher, you could be overcorrecting.

Of course someone may come along in 5 minutes and say "I've seen that before, it is caused by (insert your best guess here), get rid of the cause and you will be good to go".

These issues are some of the peskiest we deal with in the AP world, but getting them sorted out is well worth the effort in getting better future results and much less aggravation!

Bob

I see that now. I use a Pegasus flat panel and APT's flat aid to capture 20,000 ADU flats because I was under the impression that approximately 1/3 peak ADU was most desirable. I can certainly take 32.000 ADU flats easy enough and try them. I'll see if I can find one of my corrected images and post it here for comparison.

Thanks Bob!

Larry

[bobharmony]

I see that now. I use a Pegasus flat panel and APT's flat aid to capture 20,000 ADU flats because I was under the impression that approximately 1/3 peak ADU was most desirable. I can certainly take 32.000 ADU flats easy enough and try them. I'll see if I can find one of my corrected images and post it here for comparison.

Thanks Bob!

Larry

Hi Larry, yes 20,000 sounds a bit low to me. Over the years I have read conflicting information about how bright flats need to be to work properly, everything form "it doesn't matter at all" to "make sure flats are multisecond exposures and peak in the middle of the range". My experience tells me that the latter idea works well for me. Member @Stuart was very helpful in sorting out issues with flats for several folks about a year ago or maybe a little further back.

I am currently using 5 second flats that peak in the middle of the range and corresponding dark flats to correct the camera generated bias. When I started out a few years ago I would set the DSLR to A/V mode and let them rip, but over time I have found that running longer and exposing to the middle of the histogram makes post-process easier and has improved my final images. Of course every camera is different and you will need to experiment to see what works best for you and the Altair.

Bob

[metastable]

Was the moon out that night? I've seen very similar effects when either the moon is out or there is a street light somewhere nearby (the telescope doesn't have to be pointed towards the moon or the street light for this to happen).

[metastable]

[dcrowson]

Larry,

are you subtracting bias or a matching-duration dark (if using a CMOS camera) from the flats?

Dan

[Larry 1969]

Larry,

are you subtracting bias or a matching-duration dark (if using a CMOS camera) from the flats?

Dan

Yes. I shoot dark flats and bias frames. Mainly because APP gives me a warning if I don't. The image I posted above is completely uncalibrated though.

Larry

[dcrowson]

Sorry, I just saw flats and not working and went from there. It is odd to get bright vignettes versus dark. The only thing I can think of is possibly light getting around the edges of the filters but the camera you have listed is a OSC. Maybe try pointing at something without the coma corrector and just see if it gives the same kind of result.

[OhNo]

The Quattro series of scopes are terrible for light leakage. Next to impossible to do darks in the daylight unless you wrap the whole thing in a dark light eliminating blanket.

[Larry 1969]

The Quattro series of scopes are terrible for light leakage. Next to impossible to do darks in the daylight unless you wrap the whole thing in a dark light eliminating blanket.

I'm pretty sure I still have a little light leakage but I'm having a hard time understanding what could cause illumination OUTSIDE my imaging circle. It would almost have to be light entering the camera and hitting the sensor right?

Let me know if I have this right:
I have a rectangular (APS-C) sensor and a round imaging circle. The corners of the sensor get dark (vignetting). This is normal and can be calibrated out with flats. All light entering my scope, including any "light leakage" should be projected onto the sensor INSIDE the imaging circle. I don't understand where light could enter and hit the sensor in the corners...

Larry

[Larry 1969]

Was the moon out that night? I've seen very similar effects when either the moon is out or there is a street light somewhere nearby (the telescope doesn't have to be pointed towards the moon or the street light for this to happen).

It was actually under a full moon...

Larry

[OhNo]

Light can enter around the primary mirror. At least mine does. Shooting away from the moon would expose the primary end to moonlight.....

[Larry 1969]

Light can enter around the primary mirror. At least mine does. Shooting away from the moon would expose the primary end to moonlight.....

Yeah. I get that, but let's say light did get into the scope from around the primary. I would expect it to show INSIDE my imaging circle. I would expect normal vignetting with the corners being dark. I don't get a whole lot of clear nights (I'm sure you know how it goes) so I'm going to try NGC 1333 tonight and take flats at 32,000 ADU and see what happens.

Larry

[metastable]

How is the baffling around your focuser? Stray light could be coming in (pictured below as a black cone around the secondary mirror)

[Larry 1969]

How is the baffling around your focuser? Stray light could be coming in (pictured below as a black cone around the secondary mirror)

The tube is baffled from the focuser down to the primary.
I was able to get about 4 hours on NGC 1333 last night and I took 32,000 ADU flats. Stacking now.

Thanks!

Larry

[metastable]

I meant you might find that you need baffling in the focusing tube itself to prevent light from the moon reflecting off the side of the main tube from entering the focusing tube as shown by the black cone in the diagram.

[Larry 1969]

I meant you might find that you need baffling in the focusing tube itself to prevent light from the moon reflecting off the side of the main tube from entering the focusing tube as shown by the black cone in the diagram.

I understand now. Thanks. I'm not sure how I would do that though. My CC inserts pretty far into the draw tube. I only have about 1 inch unused.

Larry