Imaging help please.

[Tony_]

I seem to be having trouble getting a decent image lately (9.25 SCT. CGEM mount, Canon 550d). The attachment show the problem I am having with stars. (Don't worry about the quality - I haven't processed it much) I know SCTs don't give stars as sharp as a good refractor - but I rarely used to get stars like this now I get it every time.
I use a Bhatinov mask to focus - and I am sure I get this quite close.
It's not mirror flop. Collimation is quite good.
I suspect it has something to do with seeing conditions? It is cloudy most of the time with just a few clear "gaps" (1 night) - perhaps this problem is caused by disturbances?
The problem is that the stars are either a little circle or a star with a circle around it.
Does anyone know what causes this and how to fix it? I used to use a CG-5 mount now I have a CGEM which tracks better - I haven't made any other changes.

Regards,
Tony.

[Nico13]

Hi Tony,
That's a strange look you have there, I recently had been trying to do some imaging and by the way I have a similar setup to you and while trying to focus my guide scope and camera was getting a similar sort of halo around the stars in the PHD window, not as defined a circle as you have there but it turned out it was just the humidity level was so high, I mean 100% humidity according to the weather website for our location.
So my stars looked to be very soft on focus but no ring like that.

Maybe someone else has seen something that more resembles it.
Good luck sorting it out

[Tony_]

Thanks Ken,

I now suspect it may be caused by dust/grime on the corrector plate - but I don't know if this causes this problem. Anyway I will clean it. It doesn't look that bad - until you look at it closely.
I have had issues with atmospheric moisture - but I think that usually causes the stars to be big and fuzzy?

Regards,
Tony.

[rat156]

Hi Tony,

I have never seen anything like this before. I know that this probably doesn't help much, but I've done just about everything wrong whilst taking images and I've never seen this before.

The brighter stars look a bit like the image I use for collimation, i.e. just out of focus, though I don't think I have ever gotten such a bright central spot.

What sort of camera are you using? Have you got another, even a guide camera to take an image?

First things first you have to isolate the cause, is it optical or is it an artefact of the imaging system? Chuck an eyepiece in, put the biggest magnification eyepiece in, with a barlow if you've got one and just defocus the image. Do the stars look like you've imaged? If so it's a simple focus problem. If not, then maybe it's an out of focus reflection off some part of the imaging system.

Don't trust focus masks for wide field imaging, there great for high magnification work, but for wide field, isolate a star, put the camera in focus mode and move your focuser back and forth until the star image is the smallest possible, or if you've got software that does FWHM calculations, use that.

When the weather clears next in Melbourne (I'm about to spend four weeks OS, that should help), report back on the results.

Cheers
Stuart

[Poita]

I've had a similar effect when there has been a tiny amount of condensation and dust on the corrector plate. Try cleaning it and use a long dewshield and see if the problem goes away. Check that the CCD window is nice and clean whilst you have the cleaning gear out.

[gregbradley]

Dew on the corrector or you imaged through a cloud.

Greg.

[Tony_]

Thanks for your help everyone - but I still am not sure how to fix it.
I cleaned the corrector plate and took some pictures tonight - same result.
Seeing conditions weren't that good - but I have taken shots through light clouds/moisture before and the result didn't look like this (stars were fuzzy).
I didn't have this problem same time last year - now I have it every time.

Maybe it is as simple as focus - but surely I would get the focus close enough occasionally? Also the stars with the dot in the middle and the circle around it isn't what an out of focus star looks like? The camera sensor is clean, the primary and secondary mirror look clean.

Maybe it is just the weather? - but I guess other people in Victoria with SCTs aren't having the same problem. If the secondary mirror isn't in the correct rotational position - could this cause this problem (I doubt it ... but I'm getting desperate)

What else can I try, apart from focus? Any ideas.

Regards,
Tony.

[silv]

hey tony,

for troubleshooting take everything out of the imaging path that isn't absolutely necessary. take test photos of stars only and put back stuff one by one.
if the "feature" reappears at any stage you know where to look closer.
if it doesn't reappear then yay!

good luck :)

[atalas]

Can you put up a bigger image?its hard to evaluate at this scale....with what program did you process this shot?

[gregbradley]

The details are a bit sketchy. Your first post implied you were imaging in cloudy weather. Was this the case?

Are you able to try a different camera? Could it be the camera? Is there something on the sensor window or on your filters?

What you posted looks a bit like brighter stars go when you image when there is high thin cloud. Not the case?

Greg.

[Tony_]

Thanks once again for your replies - everything helps to learn more.

I only have one camera - the sensor looks very clean, no smudges.

The seeing conditions have been poor - I can hardly remember when there were 2 consecutive clear nights! Possibly was high level (virtually invisible) clouds. I have taken photos in moisture ridden skies - usually the stars get big and fuzzy? I haven't seen a deep blue sky for a long time - always a slight haze - but I do get some half decent images - but nothing like most of you guys.

I looked more closely at the individual subs - I can't see this ring affect on them. However I did notice that a lot of the images had a lot of elongation from drift. My mount has (or rather had) a lot of play in the RA gear - I have now tightened it. This possibly gave poor polar alignment (the celestron all star polar alignment is usually very good - I usually get less than 1 arcminute drift/hour. I know you probably get better - but I have to do a full set up every time.) Last night the drift was more like 1 arcminute/4 minutes which is hopeless without guiding.

Possibly (?)- deepskystacker caused these star rings by trying to stack/process poor images - probably not, but at least now I have removed this. I also uninstalled and re-downloaded DSS - just in case there was a bug, also unlikely but I have to eliminate as many things as I can think of.

Next time I will concentrate on manual focusing - maybe my bhatinov mask is giving an inaccurate diffraction pattern? Also, now I have tightened the gear I will get the autoguider back in action.

I'll keep posting until I resolve the problem - maybe what I learn will help others.

Regards,
Tony.

[silv]

Great that you were able to fix the play in RA!


:thumbsup:


I had misunderstood that we were looking at a single frame in your first post.

what stacking methods in the Result tab and Light tab do you choose in DSS?

I produce "a lot of elongation" (which is quite a relative term), too, but with
- "Mosaic" in the 'Result' tab
- and "median" in the 'Light' tab
- and an "only the best 80%" setting, DSS doesn't produce those circle stars in my otherwise very humble results.



It already does.

[Poita]

Ah, if that first image is a stack and not a single sub then it will be in the processing.

Well, even if the problem wasn't with the 'hardware' you got the scope cleaned up and other problems solved as a result. Win-win!

[Tony_]

Yes it is a stack - about 90% of 80x 30 second subs. I used to get reasonable results with this set-up. I have a guidescope but I had trouble with it recently (probably due to excessive play in the RA gears). Also the attached picture is a crop.

I reloaded dss in case there was a bug. I also deleted the worst 40 subs and restacked @ 50% so only the best 20 subs were used. The stars still looked like rings.

Next clear night (which seem rare) I'll experiment with the position of the centre spike on the bhatinov pattern and see if the centre position is the best focus or not. I'll also try some visual focusing. Eventually I'll eliminate all controllable factors - can't do much about the seeing conditions though.

Hopefully I'm getting closer to a solution (and learning more). How do I attach bigger pictures - the limiting size is 200kb?

Regards,
Tony.

[silv]

If the rings are not on the subs then they must be caused by a processing "feature" of DSS. :)

Which is a good thing.

Re-installing the program might not get rid off the user's previously used saved settings.

So which settings have been used, Tony? Just open DSS and pretend to stack another set. DSS will show you the last used settings per default.

[Tony_]

I don't think the settings have changed, these are the main ones:

Standard Mode (no drizzle)
Light: Kappa-Sigma clipping, Kappa 2.00, iterations 5.
Dark: Median, Dark Optimization
Flat: Median
Automatic alignment
Intermediate Files: FITS files
Cosmetic: Detect and clean Hot pixels: 3px, 50%
Cold Pixels: 2px,50%

I think these are the default settings, I actually don't know what the optimum settings are - but if someone can suggest better settings I'll try them.

Regards,
Tony.

[silv]

yeah,

instead of Standard, try Mosaic.
instead of Lights-kappa sigma, try Median.

leave the rest as is.

and let DSS choose your 50-80% best pictures for the stack in that main window.

will you post your resulting image? that would be great.

***

When you click on "recommended settings" - what does DSS suggest in blue colour?

thanks. :)

[silv]

when you inspected the single light frames, did you enhance exposure/luminance to see what is actually captured?
just thinking that, maybe, the single frames in fact do show the circle but you just didn't notice them...

[Tony_]

I haven't solved the problem yet - still not sure if I'm even getting closer.

Took a photo last night - same result. I took a closer look at some of my past pictures - they actually do have the same problem but not as bad. The tracking (unguided) was near perfect - virtually 0 drift just PE. So at least tightening the RA gears has fixed the polar alignment issue.

I also took a closer look at last nights subs - I adjusted exposure etc. but I couldn't see any sign of the circles around the stars. If this is correct then it must be processing? (but I still haven't ruled out focusing or weather). If it is weather then other people in Melbourne with SCTs would have the same issue?

The green selections in DSS recommended settings are "Use median combination method" - for dark files and flat files (I don't use dark flats or offset bias). and "Use RGB background calibration".

I have the trial version of nebulosity - I'll do some stacking with that and see if it is better, I also have AIP but I haven't used the stacking in this - it seems more complicated.

I use an unmodded canon 550d - attached directly with an adapter/t-ring, no filters or reducer.

Any more ideas?

Regards,
Tony.

[Tony_]

The DSS blue recommendations are:

If you are processing narrowband images - Use per channel background calibration

If the resulting image looks too grey - Use perchannel background calibration.

I don't think either of these are relevant?

Regards,
Tony.

[Tony_]

I just did a stack with nebulosity 2 - the stars were a little fuzzy (this is probably weather or focus) - but no rings around stars!
So maybe it is a processing problem - all I need to do now is find the correct parameters (I hope).
What should I use for an unmodded DSLR?

Is there better software (preferably free) than DSS that does auto-stacking etc.

Regards,
Tony.

[Tony_]

Thank-you to everyone who offered suggestions. Everything helps.

I have now solved this problem - it was a processing issue. I reset DSS to the default settings and it worked. My Grus Trio picture wasn't too bad for a 30 minute total exposure time. Focus was good - with some structure in the galaxies. I haven't looked in detail at the new parameters - and I don't know when/why I originally changed the settings.

I can now reprocess my old images and improve them a lot + I have several that I thought were no good that may be okay after all.

I just have to work on better guiding, longer subs and increase my processing knowledge.

I have a learned a few things - I now know that tightening the worm gear on a cgem is easy. My bahtinov mask seems to give a good focus. It pays to learn more about processing etc. I think I need to start including biases too. The hazy sky doesn't prevent pictures being taken - but obviously isn't good either.

Regards,
Tony.