Iris and Ghost re-process

[Carole]

I wasn't happy with my original process of this object done at Ollys and have now done a completely new re-stack and process and feel happy with this revision.

Details as before total 5h 10mins LRGB though I didn't use all the Lum data (forgot about the flipped subs), so it's actually 1 hour less 4h 10mins.

I had to use DSS to stack (with cosmetic reduction of HP) as Astroart was leaving lots of hot pixels.  

Larger file:
http://cdn.astrobin.com/images/a4a21ec5-c833-4524-87c2-9d8da65040e6.png



Larger file:
http://cdn.astrobin.com/images/34385173-c2ec-44a3-ae93-1b5051011193.png

[MarkS]

They're looking very good indeed Carole.  Especially with the hot pixels gone. Good star colours and the dust looks excellent without lots of noise.

I did notice that on the Iris that the stars look slightly elongated but I don't notice it on the Ghost.

[Carole]

Thanks Mark,

I did notice that on the Iris that the stars look slightly elongated but I don't notice it on the Ghost.

Yes I noticed that too Mark, the Ghost is from the centre of the image, but the Iris is to one side and so I guess there must have been a bit of coma, I did not notice this when I was processing first time around, guess I was too concerned about other problems, like the hot pixels.

Carole

[Rocket Pooch]

Was you using Olly's Tak FSQ, if so that flat as a pancake with no comma.  Or were you trying something else?

[MarkS]

Was you using Olly's Tak FSQ, if so that flat as a pancake with no comma. 

Oh no!!  Not another scope that's been squashed by a car?

[Carole]

Olly's Tak FSQ,

Yes I was, I've just examined the 4 corners of the luminance before full processing.  There is definitely some elongation in the bottom right of the image.  



Carole

[Rocket Pooch]

Who's camera?

[Carole]

It was all Olly's kit.  It was his Atik11000.

Be glad of an explanation if any-one knows what the cause is.

Carole

[MarkS]

Be glad of an explanation if any-one knows what the cause is.

My best guess is that there was a small amount of CCD tilt.

[mickw]

I wouldn't have expected Olly to have duff kit.

Carole did you assemble camera, scope etc. or did Olly set it up ? 

Possibly camera not sitting squarely to the scope.

[Carole]

Carole did you assemble camera, scope etc. or did Olly set it up ? 

It was all set up by Olly. 

I think I am going to E Mail him and ask him what he thinks.

Possibly camera not sitting squarely to the scope.

Could it be those screws that hold the camera onto the drawtube shoving it over slightly, but then surely it would have the opposite effect on the opposite side, and seems to me like every-one would have the same problem if this were the case.  Otherwise I can't see how a camera can not be sitting square to the scope.

Why don't they put 3 screws in so you can screw evenly from all directions.  They have a centering gadget for collimating Newtonians for exactly this reason. 

Carole

[MarkS]

Was a focal reducer being used?

[Carole]

Was a focal reducer being used?

I don't think so Mark, but I'll re-check with Olly who incidentally can't offer any explanation.  He asked me to check back with the original fits files in case it was something the software might have done, but it's on the original fits files too.

Carole

[Carole]

NOW THIS IS INTERESTING.

You may recall that in the first post of this thread I mentioned that I forgot to use the flipped luminance subs.
there is no such elongation in the flipped luminance subs nor the RGB subs, so why would it only appear on the pre-flipped luminance subs and only in one corner and not appear after the flip and not on any of the RGB subs? 

Luminance were all done the same evening, but RGB done over several evenings but can't remember if they were done were after the object passed the meridian, certainly some were as I definitely remember green being done VERY late. 

Could it possibly be the weight of the filterwheel having an effect only in certain positions? 

Carole

[Rocket Pooch]

I would have checked it at the start of each image, although its Olly's kit its still subject to the normal issues.

[Mike]

Sounds like the camera has shifted under its own weight towards one of the corners to me.

[Carole]

That's the conclusion I came to.  I must take a look at the subs done with the same set up also done pre-flip.

Carole

[MarkS]

Sounds like the camera has shifted under its own weight towards one of the corners to me.

I think it's slightly more complex than that.  I think the top left and top right corners are equally affected by the "tilt" (if that's what it is - as you say, probably shifting under its own weight).  However it seems there is also a very slight elongation of stars over the complete field (possibly a slight guiding issue).  This elongation is is the same direction as the elongation of the stars in the top right and so the effect is additive making it more pronounced.  But this universal elongation acts in a different direction to the "tilt" elongation of the stars in the top left thus causing them to be elongated in 2 different directions and therefore they become almost circular again.  An analysis shows that the stars in the top left have a higher FWHM (or bloat) than the stars in the rest of the field of view.

I wish I hadn't started this!

[Mike]

That Atik weights nearly 1kg, with the weight of the filter wheel as well it will be well over 1kg so it will need a good strong focus unit to hold it perfectly still all night long.

Like Chris says, it's best to check the subs from time to time as you go along

[Carole]

Olly was having trouble with guiding at some points during the week as PHD kept pinging but most of the time the stars appeared to be round so we decided to ignore it as we couldn't fathom out what was going on.  This was the 2nd guiding set up we used as the Lodestar was having connection problems and we had to swap the whole guiding set up + use an extra computer as there were no drivers for the Atik guide camera we were now using.

I normally do check my subs (when I am at home and networked) as I go along and any dodgy ones get removed.
I am not sure how I missed this, I think I generally tend to look at the stars in the centre of the image rather than the edges, and thinking they were all OK I didn't re-check when I did my re-stack and process.

I guess I'll have to re-stack and process it again when I get time using the post-flip luminance subs.

Carole