Click here if you are having trouble logging into the forums
Privacy Policy |
Please read our Terms
of Service | Signup and
Troubleshooting FAQ | Problems? PM a Red or a Green Gu.... uh, User
RoundStars
member
Reged: 07/09/08
Posts: 52
Loc: Hutchinson, KS
|
|
The attached flat appears to show that something in my imaging train is skewed. How will this affect my images? Thanks!
-Joe
|
Jared
Carpal Tunnel
Reged: 10/11/05
Posts: 2527
Loc: Piedmont, California, U.S.
|
|
Can't say for sure based on this flat. We need a bit more information...
The most critical question is how significant is the difference in vignetting from edge to edge? In other words, prior to stretching (or at least after a purely linear stretch--setting of the white point only) what was the numeric difference in ADU between the four corners? If it is small, I wouldn't worry about it at all--I'd just use flats to remove the vignetting and call it "fixed." If the difference is significant--more than just a few ADU--you probably need to figure out what is going on.
Second, if you are using a refractor, you probably won't see this making much of a difference in image quality even if the vignetting is noticeably different between corners. Most refractors are aplanatic (or nearly so) and provide substantially similar image quality from edge to edge. I'd judge this based on the quality of the star images in the corners.
Third, make sure you consider the possibility that the issue is with your flat procedure/equipment rather than with the actual optical train. Was this a twilight flat or a light box? Are you sure the light box is evenly illuminated? Are you sure there wasn't a sky gradient causing the problem? Not trying to knock your technique, just want to make sure you consider all possibilities rather than focusing in on the imaging train itself.
What was the scope/camera/filter wheel combination in this image? That might help narrow things down as well.
--------------------
- Jared Willson
- TMB 152 f/8 Apochromat
- Fluorostar FLT-110 w/ TEC optics
- Stellarvue SV80S
- Astro-Physics Mach1 GTO
- Takahashi Teegul SP Mount
- STL-11000
|
RoundStars
member
Reged: 07/09/08
Posts: 52
Loc: Hutchinson, KS
|
|
Jared:
Thanks for your comments. My question is more concerned about how a skewed imaging train will affect image quality, i.e. field flatness, aberrations, coma, and whatever else.
This was an AT66 refractor with an SCT 6.3 focal reducer/flattener and a Canon 350D, no filter wheel. This is a daylight tee-shirt flat, no levels or stretching, only a 7-pixel blur of the original image.
I need to study your question about the ADU numeric difference between the four corners. I don't know. I can tell you that applying the master flat to the lights more-or-less eliminates the vingetting, although I can sometimes still see a gradient in the calibrated images corresponding to the vingetting.
-Joe
Thanks.
-Joe
|
HunterofPhotons
sage
Reged: 04/26/08
Posts: 265
Loc: Rhode Island, USA
|
|
Joe,
Comments are below:
Quote:
.... My question is more concerned about how a skewed imaging train will affect image quality, i.e. field flatness, aberrations, coma, and whatever else.
*** A camera's chip will not necessarily be centered on the optical axis, though it's nice if it is. Obviously, the more off-axis that you are, the more the off-axis aberrations will show.
Quote:
This was an AT66 refractor with an SCT 6.3 focal reducer/flattener and a Canon 350D, no filter wheel.
*** That's a heavy camera to hang off the back of that scope. You might be seeing some tilt.
Quote:
This is a daylight tee-shirt flat, no levels or stretching, only a 7-pixel blur of the original image.
*** Are you saying that you blurred your flat? If you are, then don't. A blurred flat won't work properly.
Quote:
..... I can tell you that applying the master flat to the lights more-or-less eliminates the vingetting, although I can sometimes still see a gradient in the calibrated images corresponding to the vingetting.
*** If your daylight flat wasn't taken at the correct angle from the sun, then, yes, there will be a gradient in the final image. A flat only works properly if it's taken of an evenly illuminated area. Most areas of the sky are not evenly illuminated.
dan
|
Jared
Carpal Tunnel
Reged: 10/11/05
Posts: 2527
Loc: Piedmont, California, U.S.
|
|
Quote:
Jared:
Thanks for your comments. My question is more concerned about how a skewed imaging train will affect image quality, i.e. field flatness, aberrations, coma, and whatever else.
Most refractors are relatively aplanatic, so there shouldn't be significant astigmatism or coma off axis. Field flatness, of course, is another matter. I suspect you will find that if your optical train is not perfectly in alignment with the optical axis that stars in one or more corner are a bit softer--perhaps even radially elongated--than stars in the remaining corners. The effect is likely going to be quite minor, though. The easy thing to do, of course, is just to take a well focused, short exposure of a star field and look for yourself.
Quote:
This was an AT66 refractor with an SCT 6.3 focal reducer/flattener and a Canon 350D, no filter wheel. This is a daylight tee-shirt flat, no levels or stretching, only a 7-pixel blur of the original image.
Wow, that's actually a fair amount of vignetting for an unstretched image--more than a full stop. It's hard to figure out the exact amount since the white point is clipped, but it's at least 1.3 stops from my measurement. Of course, flats will help significantly. When taking flats you'll want to make sure you don't clip the white point or you run the risk of being left with a funky gradient. Also, rather than blurring the flat you would be better off bias subtracting and then averaging several flats together. That is going to get you a good signal to noise ratio without reducing the effectiveness of the flat for addressing dust donuts--especially the spots on the CCD cover slip/IR filter itself.
Hard to know whether the lack of evenness in the flat is caused by a gradient in the sky (since you'd be imaging a pretty wide swath of sky--even with the 'T' shirt acting as a diffuser) or by the weight of the camera and focal reducer pulling something out of alignment.
Quote:
I need to study your question about the ADU numeric difference between the four corners. I don't know. I can tell you that applying the master flat to the lights more-or-less eliminates the vingetting, although I can sometimes still see a gradient in the calibrated images corresponding to the vingetting.
-Joe
Thanks.
-Joe
In your place, I believe what I would do is take a picture of a star field to try and determine whether stars in all four corners are equally sharp. If they are, I would ignore the lack of symmetry in the flats. This scope/reducer combination is (obviously) going to be imaging a huge swath of sky, so you are likely to see light pollution gradients even with proper flats. As long as the flats are addressing the vignetting and the dust donuts I would continue to use them. I would definitely recommend doing some research to see whether there are field flatteners out there for the 66 that don't vignette so badly with an APS-C sized sensor. Of course, this may be a limitation in the scope itself--it uses an SCT threaded visual back, yes?
--------------------
- Jared Willson
- TMB 152 f/8 Apochromat
- Fluorostar FLT-110 w/ TEC optics
- Stellarvue SV80S
- Astro-Physics Mach1 GTO
- Takahashi Teegul SP Mount
- STL-11000
|
RoundStars
member
Reged: 07/09/08
Posts: 52
Loc: Hutchinson, KS
|
|
Dan and Jared:
I'm going to assume that the camera chip is not centered. Reason: I notice that the vingetting is similarly off-center when I use the same camera on my SCT. That would seem to indicate that it's not caused by flex in the refractor's drawtube.
Yes, the AT66's drawtube terminates in an SCT-style thread.
It could be that the SCT-style focal reducer is causing the severe vingetting. When using the same focal reducer on my SCT, the vignetting looks very much the same. So it looks like the vingetting is caused by the focal reducer and not the telescope.
Astro-Tech makes a field flattener for refractors in this f-stop range, that Jerry Lodriguss has reviewed favorably. I think my next step is to try that.
Examining the star images at the four corners shows that there's no radial distortion. Instead, it's -- I don't know the proper term -- "concentric" or circular. From using the SCT-style focal reducer? -- I don't know. I've attached a composite of the four corners below, at 200%. From this composite, it looks like the upper left corner is not illuminated as well as the other three corners.
When doing my tee shirt flats, I'm careful to use an evenly-illuminated swath of sky. I do the flats by photographing a cloudless sky in full daylight, making sure that the sun is not striking the tee shirt itself. I'll take your suggestion of not blurring the flat, and avoiding clipping the white point.
-Joe
|
Jared
Carpal Tunnel
Reged: 10/11/05
Posts: 2527
Loc: Piedmont, California, U.S.
|
|
What you are seeing in the four corners is ordinary field curvature--only reversed! The SCT flattener/reducer is over-correcting the curvature produced by your telescope. It is possible that you could improve things with different spacing, but I would take a look at a flattener made for your your focal length refractor rather than for an SCT. The distortion in the four corners is very similar, so I wouldn't worry about the lack of collimation in the imaging train.
--------------------
- Jared Willson
- TMB 152 f/8 Apochromat
- Fluorostar FLT-110 w/ TEC optics
- Stellarvue SV80S
- Astro-Physics Mach1 GTO
- Takahashi Teegul SP Mount
- STL-11000
|
RoundStars
member
Reged: 07/09/08
Posts: 52
Loc: Hutchinson, KS
|
|
Here's some feedback on your comments:
I've found that the severe vingetting and odd star shapes at the corners of images were in fact caused by the SCT focal reducer. Removing the FR gives good illumination all the way to the corners of my APS sensor, with typical radial distortion of star shapes. Even though the sensor appears to be slightly off-center, the corner star shapes are similar.
So, I'll try using the proper field flattener for the telescope, which should give better results. Thanks for everyone's comments!
-Joe
|
|
4 registered and 10 anonymous users are browsing this forum.
Moderator: Charlie Hein, knuklhdastnmr
Print Thread
|
Forum Permissions
You cannot start new topics
You cannot reply to topics
HTML is disabled
UBBCode is enabled
|
Thread views: 98
|
|
|
|
|
|
|
Previous
Index
Next
Threaded
Edit
Reply
Quote
