M63 Sunflower Galaxy

[MartinB]

I imaged M63 back in 2009 with my 10"LX200ACF and QSI532 along with SX active optics.  Looking back I felt it was worth adding some more data and doing a better job.  In the end the data gathered in 2009 with the AO kit was better than this time around without AO except for one night when the seeing was pretty good

Scope: 10" LX200ACF with AP 0.68 reducer - approx 1900mm fl 

Camera: QSI 532 wsg

Filters: 2009 astronomic CLS and 13nm Ha (I didn't use 2009 colour data)  2016 IDAS + Baader LRGB

2009 16x 300secs CLS  No sure of the Ha details

2016 12x360 secs IDAS Lum  20x200 secs RGB

All bodged together in Maxim, lum deconvolved in CCD sharp.  processed in PS

It's been a bit of a struggle and I ended up throwing away a lot of lum data on account of poor seeing.  Lots of processing challenges esp colour balancing, and deciding how hard to work on pulling out the outer halo.  

 

[swag72]

Nice one Martin - That looks very nicely processed and very natural.

On a side note - You say that you threw away data - Well I've just done a little experiment and adding 12 bad exposures into a stack of 30 made a big difference for noise and made no discernible negative difference. I always go by the adage that its worth binning poor data - But as long as the guiding is OK, now I'm really not so sure.

[laudropb]

Very nice image.

[opticalpath]

I really like this one, Martin.  Nicely done.

Adrian

[johnrt]

That's a fine M63! Lots of glorious detail in those arms!

[MartinB]

Thanks everyone

• 17 hours ago, swag72 said:

  On a side note - You say that you threw away data - Well I've just done a little experiment and adding 12 bad exposures into a stack of 30 made a big difference for noise and made no discernible negative difference. I always go by the adage that its worth binning poor data - But as long as the guiding is OK, now I'm really not so sure.

  Actually Sara, my comments were a bit of an over condensation of what I did!  I used the images with the best definition/lowest FWHM using CCDinspector, maxim and a mark 1 eyeball.  I used these for the bright galaxy because s/n wasn't an issue.  I then threw in the rest of the data and used this for the halo and background.  Unfortunately, owing to a schoolboy error some of the frames didn't match up properly and I wanted the maximum field possible so there was a bit more pruning!  To be honest, in the end I lost track of how many subs I used.  My M63 processing folder is chocker block with different luminence stacks.

[alan4908]

Very impressive Martin, I do like your processing, particularly of the dust lanes and outer halo.

On the throwing away subs theme:  before I recently watched many hours of PS processing tutorials by Adam Block (Cosmic Canvas series) I used to throw away any sub that wasn't near perfect. What surprised me about the videos was some of the (relatively) poor subs Adam kept due to poor seeing, guiding errors etc. The reason he kept these was that he found that the stacking program he used (CCDstack), when used correctly, was very good at rejecting the poor parts of the image and keeping the good parts. 

Alan

[MartinB]

Thanks a lot Alan.  

I have CCDstack and also Adam Blocks DVDs for PS and CCDstack.  US imagers seem to love CCDstack whereas Europeans seem to go for PI!  Have you ever tried comparing the results with CCDstack and Maxim.  When I did I found CCDstack more labour intensive but couldn't tell the end results apart?  I might have to have another look!

[alan4908]

Martin - Yes, I've done a few comparisons between CCDstack and MaximDL and I would suggest you have another look....

On the data rejection front, the main feature I like about CCDstack is that you can see which parts of a sub are being rejected for a given overall rejection setting, so if you think more or less rejection is required for a particular sub you can increase/decrease the rejection without impacting the data rejection of the other subs. I also like the statistical weighting that CCDstack quantifies for each sub, so if I have anything below 0.5 I discard. With my Trius SX26C, I've also found that CCDstack gives a much better results at the frame edges, in particular, if have a stack of subs which have been captured before and after a meridian flip, stacking in MaximDL often resulted in a "double star" effect at the edges. I presume this is due to field curvature with the median flip amplifying the effect. On the plus points with MaximDL: I do like the manual alignment process, which is much easier than CCDstack, the astrometric stack feature is also invaluable when you have subs which have large mis-alignments and CCDstack either will not stack them or will stack them incorrectly. 

(As an aside, the reason I bought CCDstack was not for any of the above, it was because I found it gave very good results with respect to deconvolution. With MaximDL I found it very difficult to get acceptable results).

Alan 

[Barry-Wilson]

Lovely and natural M63.  A really great image.

I think you have the colour balance spot on from other images I've seen.  You also have defined well the two prominent and unusual linear ejecta on the right of the disc - fabulous.

[MartinB]

Thanks a lot Barry, that's encouraging

Alan, I shall definitely have try another comparison, I have used it quite a bit in the past but gradually moved over to Maxim, I might be being a bit lazy.   Recently I've just been using it for deconvolution.