Resolving the core of the catseye. Lucky imaging maybe?

A place to share images that you have taken with SharpCap.
Forum rules
Please upload large images to photo sharing sites (flickr, etc) rather than trying to upload them as forum attachments.

Please share the equipment used and if possible camera settings to help others.
Post Reply
timh
Posts: 515
Joined: Mon Aug 26, 2019 5:50 pm

Resolving the core of the catseye. Lucky imaging maybe?

#1

Post by timh »

The core of the catseye nebula NGC6543 in Draco is compact - 16-20 arcsec - and quite bright (Mag 8.2) and so seemed planet-like enough that it might just be susceptible to lucky imaging techniques? Here I had a go and had some success but concluded that getting it to work convincingly would probably require something bigger --maybe a 16 inch Dob at F 4 or so?

200 mm PDS SW Newtonian plus Baader flattener at F 5.0 on an Ioptron CEM70 mount. Bortle 6. Moon full and one day before full

ASI 294 MC camera at -10C - oixel size 4.63 uM
ASI 294 MM camera at - 10C at BIN 1. - pixel size 2.315 uM

Expt 1. Was to use conventional deepsky imaging. 93 x 10s ASI 294MC exposures at gain 124 (near unity) were selected in SC and analyzed in PIxInsight indicating an average FWHM resolution score of < than 2.5 arc sec (i.e all the frames were sharp). Preprocessed , stacked and processed in Pixinsight - using a high dynamic range transformation to bring out detail in the bright core yielded image A. The image scale here is 0.95 arcsec/ pixel.

Expt 2. was a 320 x 240 pixel SC capture of an 8 bit RGB AVI file at maximum gain 570 and 2396 frames of exposure length 100 ms using the ASI294 MC. These were debayered in PIPP and 270 frames selected, stacked and processed in Registax to produce image B (same image scale as above).

Expt 3. was a 320 x 240 SC capture of an 8 bit mono AVI file at maximum gain 570 using the ASI 294MM. From an initial 17000 frames only a hundred or so were stacked to produce image C. The image quality was very poor using the BIN1 smaller pixels (image scale 0.425 arcsec/ pixel)

The conventionally obtained image A was not at all bad - the best that I have had of the catseye core anyway. In principle though, lucky imaging at exposures < 100 ms should be able to resolve better than the seeing limit of 2.5 arc sec.

However, even with a sensitive MONO camera at maximum gain this object appeared simply not to be bright enough (or uniformly bright enough) to capture adequately exposed frames at 100ms for the method to work at a pixel size and image scale (0.425) that could, in principle have improved resolution down to about an arcsec (Image C.)

Neverheless lucky imaging did work - and reasonably well in image B at an image scale of about 1 arc sec/ pixel - but then providing only marginally if any improvement in resolution over conventional imaging -- and with a total imaging time of only 27s - at a cost of decreased detection of light.

Possibly a quixotic quest but it seems to me that in principle this approach could still work for the catseye but that it would just take maybe 4 x more light per pixel and a larger image scale - maybe a 16 inch dob or a C11 might be up to the task if any one maybe has already tried it?

TimH
Attachments
catseye core.JPG
catseye core.JPG (20.63 KiB) Viewed 1892 times
User avatar
oopfan
Posts: 1320
Joined: Sat Jul 08, 2017 2:37 pm
Location: New York
Contact:

Re: Resolving the core of the catseye. Lucky imaging maybe?

#2

Post by oopfan »

Hi Tim,

A very interesting experiment! I like the detail of the inner core in image B, but I am at a loss to explain why the central star is so large. It makes me believe that it was dancing around from frame to frame. What does a typical single frame look like? I am surprised that Registax couldn't tame it. But then again this is not a typical lunar landscape.

Brian
User avatar
Menno555
Posts: 1051
Joined: Mon Apr 20, 2020 2:19 pm
Location: The Netherlands
Contact:

Re: Resolving the core of the catseye. Lucky imaging maybe?

#3

Post by Menno555 »

Tim

I did some testing. Meade LX200 8" f/10 ACF with a Zwo ASI071MC Pro.
Just a quick and dirty test. I did try 100ms but that became too noisy. Maybe also the full moon is not helping now :)
On the left captured in full resolution of 4944 x 3284 and on the right with a ROI of 1920 x 1080.
Full resolution was saved as FITS and stacked in DeepSkyStacker and the ROI I saved as SER, debayerd in PIPP and stacked in PlanetarySystemStacker (this gave a much sharper image then AstroStakkert did)
Cropped and doubled the image size.
Both with the same settings: 600 x 250ms / Gain 400 / Brightness 10

Menno
Lucky_Cat.jpg
Lucky_Cat.jpg (278.09 KiB) Viewed 1883 times
timh
Posts: 515
Joined: Mon Aug 26, 2019 5:50 pm

Re: Resolving the core of the catseye. Lucky imaging maybe?

#4

Post by timh »

Thanks folks,

It is certainly fun comparing different methodologies and also improves my understanding of quite where the limits lie

Brian, yes indeed the image was never stable and the problem I had in stacking that there was not a single image that could act as a good reference - and provide a decent basis for alignment points ---hence that bloated core.

Menno, yes maybe one area to experiment in would be to go longer than 100ms? The problem is that the proportion of 'good seeing' frames will drop and so the number of frames will have to increase?

I am not sure whether or not there is a higher surface brightness planetary core than the catseye to try out on?

I think what probably will work with the equipment I have available - and probably better for your set up Menno? - - is at the core of one or other of the globular clusters- say M13? Plenty of light - some structure to resolve - might be interesting to try down at 50ms or so?

TimH
MikeHuerto
Posts: 75
Joined: Thu Jan 21, 2021 12:05 pm
Location: Spain

Re: Resolving the core of the catseye. Lucky imaging maybe?

#5

Post by MikeHuerto »

Hi Tim


Have you seen this image of the cats eye recently posted? Albeit with a 23 inch Dob!. Pretty impressive, and shows what can be done with lucky imaging.

https://www.astrobin.com/7hazju/

As the owner of a 14 inch Dob, I just noticed some of your other posts on DSO imaging using a Dob and short exposures - very interesting and informative, and supports my trial and error experience that 1-4 sec exposures over 10-15 mins can produce some pretty nice DSO images.

Best
Mike
Bortle 5
Skywatcher GOTO Synscan 14" Dobson
Skywatcher 130PDS on HEQ5
ZWOASI224MC and ASI294MCPro
Laptop = Lenovo Legion 5 FullHD 144Hz (Intel Core i7-10750H, 16GB RAM, 512GB SSD, NVIDIA RTX2060-6GB, Windows 10
timh
Posts: 515
Joined: Mon Aug 26, 2019 5:50 pm

Re: Resolving the core of the catseye. Lucky imaging maybe?

#6

Post by timh »

Hi Mike,

Yes indeed many thanks for that link. 23 is really going some!

Above was my first foray into this interesting area. Since I have been working systematically to define the window that works best in terms of sampling rate, gain, F number and subframe exposure time for my particular 8 and 10 inch Newts. Sub-frames that are too short leads to SNR being too low with too much read noise in the final stack - too long on the other hand leads to less of an advantage in image sharpness. And so I reached the same conclusion as you I think -- i.e frames in the 1-3s range seem to hit a sweet spot.

Just as an example. With the 8 inch and sampling at 0.475 arcsec/ pixel I measured the FWHM of some final stacked images (about 15 min) of M57 comparing 1s subframes with 10s. The 1s subframes took measured FWHM down to ~ 1.9 whereas the 10s were at about 2.45. Not a huge decrease but an improvement worth having and a noticeable sharpening up when the FWHM 1.9 luminance was added in to improve the sharpness of the 10s RGB stacks. i.e .Below is the comparison pair with first the 10s stacked image with and then with the 1s frame image luminance added in.

One other important comment to make which seems obvious now but that didn't occur to me at first. Not only the subframe exposure time but also the sampling rate is very important. Usual Nyquist rule - it works if you sample at 0.5 arcsec/ pixel or less but not so much at 1 arcsec.


So yes I completely agree that the technique really does work and is useful for getting some good DSO images. I am surprised in a way that there isn't more out there on it because it seems a great niche for more and better astrophotography using bigger unguided telescopes...saves a lot on big mounts. Once galaxy season comes around I am looking forward to trying to get more detail on the cores of some of the brighter galaxies using a 12 inch -- but maybe they will be just too faint - we shall see? Your 14 should be ideal.

Tim
Attachments
M57pairCapture.JPG
M57pairCapture.JPG (56.14 KiB) Viewed 1663 times
MikeHuerto
Posts: 75
Joined: Thu Jan 21, 2021 12:05 pm
Location: Spain

Re: Resolving the core of the catseye. Lucky imaging maybe?

#7

Post by MikeHuerto »

Hi Tim

I actually think there are quite a few large Dob owners (12" to 23") routinely using their scope for DSOs. Here's a link to some of my efforts in the past year.
https://www.astrobin.com/users/MikeHuerto/

All are 1-4 sec exposures, but not Lucky Imaging per se (except for use of SC FWHM filter to throw out baddies). I livestack in SC for PNG EAA images, then use the stacked FITS 16 bit file of SC to beautify images in Startools. A nice routine for getting the instant joy of EAA images with the ability to greatly improve the best images with later processing SC stacks (when I have time!) - and also avoids accumulating gobs of RAW files. I have a large back log of FITS-stack files to process from the past few months. I know, I might get better stacks if I saved and restacked the RAW files in specialized stacking program, but for the moment, Im pretty happy with SCs stacked FITS output.

I have also found that camera pixel size makes a huge difference in terms of resolution (though I guess thats your arcsecs/pixel point). I am about to switch over from DSOs using a ASI 294MC to imaging planets using a ASI 224 MC + Barlow for the next few months. I will also try the 224 on various small planetary nebula, (as well as true planets).

Stay tuned

Mike
Bortle 5
Skywatcher GOTO Synscan 14" Dobson
Skywatcher 130PDS on HEQ5
ZWOASI224MC and ASI294MCPro
Laptop = Lenovo Legion 5 FullHD 144Hz (Intel Core i7-10750H, 16GB RAM, 512GB SSD, NVIDIA RTX2060-6GB, Windows 10
timh
Posts: 515
Joined: Mon Aug 26, 2019 5:50 pm

Re: Resolving the core of the catseye. Lucky imaging maybe?

#8

Post by timh »

Hi Mike,

Nice pictures at astrobin. M27 especially.

Yes you are right of course. Lots of very good Dob astrophotos out there. By my comment I meant only that - as a newbie starting on astrophotography a couple of years ago - the stuff that I would read in magazines and in expert online articles led me to the belief that for deep sky work there was really no point in big aperture and neither any point sampling at below 1 arcsec/ pixel because the sky seeing and the mount etc would defeat you anyway. So therefore astrophotography was mainly about scopes smaller than D < 130mm (Dawes limit 1 arcsec) , decent mounts and good guiding etc. It came as something of a revelation to find out that, at least for quite a few classes of DSO's, the big Dob approach is in some ways actually a better way forward. As it happens I started out with a Dob (my preferred visual instrument) and, by default (because of movement) used very short subs (FWHM-filtered by SC) and was always surprised to find that the images so simply obtained as SC autostacks were in fact sometimes consideraby sharper than images from longer guided (< 0.9arcsec RMS) subs obtained using my triplet refractor (although in theory I might have expected the seeing to about equally blur the images from both of them).

So that is just to explain the journey that I have been on --- from 3s subs to much longer guided images and then back to 3s subs again :-)

In truth I think that the Dob short sub approach is only good for certain targets and certain (bright) features of targets. Read noise obviously limits SNR. For galaxies, globular clusters etc which have bright centres I intend to experiment more with the dual approach of using short subs to capture detailed luminance at the centre and then incorporate this into high dynamic range compilations with long exposures to capture faint details at the periphery. Will see how that goes this coming year.

best wishes
Tim
MikeHuerto
Posts: 75
Joined: Thu Jan 21, 2021 12:05 pm
Location: Spain

Re: Resolving the core of the catseye. Lucky imaging maybe?

#9

Post by MikeHuerto »

Tim,

Interesting - exactly my path as well - the 14" Dob was my first scope (bought as a novice last June from an expat in Benidorm), and I was told and read its only good for planets and visual etc etc. The key step for me was switching form an unmodified DSLR (Canon 450D) to an astrocam (ASI294MCpro)... all of a sudden I was able to pick up feint nebulosity etc etc. A couple of months ago I acquired a used HEQ5 pro mod from a guy in Slovinia, but have only played around with it using the 294 and and a Sigma 100-400 telephoto. Have held off on purchasing a decent small refreactor for it. Still keep coming back to the Dob - which meanwhile is completely remote controlled from the comfort of my office.... but thats another story!

Ive found that galaxies also work quite well with the Dob- still need to process and post on Astrobin - stay tuned.

My biggest advance in the past couple of months has been improving my flats to eliminate vignetting- Robin's tip somewhere in this forum of 0 (zero) gain and sub 1-2 secs, has worked very well for me. Every day's a discovery!

Best
Mike
Bortle 5
Skywatcher GOTO Synscan 14" Dobson
Skywatcher 130PDS on HEQ5
ZWOASI224MC and ASI294MCPro
Laptop = Lenovo Legion 5 FullHD 144Hz (Intel Core i7-10750H, 16GB RAM, 512GB SSD, NVIDIA RTX2060-6GB, Windows 10
Post Reply