M51 HaLRGB project

Discussion of using SharpCap for Deep Sky Imaging
Post Reply
User avatar
turfpit
Posts: 1779
Joined: Mon Feb 13, 2017 8:13 pm
Location: UK
Contact:

M51 HaLRGB project

#1

Post by turfpit »

I am embarking on an HaLRGB capture of M51. I thought I would share my experiences as the project evolves. This will be my first serious attempt at guiding using PHD2.

The intention is to capture L @ 100s, RGB @ 300s and Ha @ 600s. The finish will hopefully produce an image with 4 hours of integration time. The first phase was to try out the equipment with guided 300s exposures. The night chosen had thin/intermittent cloud so nothing would be lost if difficulties were encountered. This is the result of a 12x300s imaging session.

M51_11x300s_filtered.png
M51_11x300s_filtered.png (287.87 KiB) Viewed 1072 times


Before commencing the image run, a single frame was captured and analysed in FITS Liberator.

FITS-frame-unstretched.JPG
FITS-frame-unstretched.JPG (130.92 KiB) Viewed 1072 times


Note that the linear histogram has clearance of 11,000 at the left hand side (no faint data lost) and the histogram does not touch the right hand side (no over exposed data). As a check a stretch using the function ArcSinh(ArcSinh(x)) was applied to the data. A healthy looking histogram (classic shape for a deep sky object) suggested the settings were good for a data run. See above image for results.

FITS-frame-stretched.JPG
FITS-frame-stretched.JPG (132.13 KiB) Viewed 1072 times


The capture plan was devised using @oopfan's (thanks Brian) SNR calculator.


plan.JPG
plan.JPG (43.09 KiB) Viewed 1072 times


The tool allows for RGB white balance using number of frames rather than torturing the resultant stacks in some astro-imaging software (which can result in loss of colour and increased noise). This work can be attributed to Al Kelly's article White Balancing RGB Filters with a G2V Starhttp://www.kellysky.net/White%20Balanci ... ilters.pdf . The way the Atik 314E and my filters behave with red, I need twice as many red as green frames to achieve colour balance.

Also note that the SNR for each of the RGB stacks is roughly equal which will avoid saturation of stars in a particular colour. The L and R+G+B SNR are roughly equal as most of the detail will come from the Luminance frames.

I summary the project will entail captures as follows:

L: 40x100s
R: 8x300s
G: 4x300s
B: 7x300s
Ha: 6x600s

bias frames 50
flat frames 50 for each filter
dark frames 20 for each exposure

I will probably capture 2 or 3 extra RGB frames to allow for planes, satellites and meteors.

The darks are 20x600s, 20x300s, 20x100s which totals ~5.5 hours. Tonight is cloudy but the darks are being collected with the scope and camera on a bench in the greenhouse. I will forgo temperature matching perfection for this project.

There are a couple of imaging opportunities over the next week albeit with a full moon around.

This is the result of the trial run last week (best 11 of 12 frames, stacked in APP, black point adjusted in GIMP and denoised in NEAT) :


M51-11x300s_filtered.jpg
M51-11x300s_filtered.jpg (373.24 KiB) Viewed 1072 times


I will add to this thread as the project evolves.

Looking at Canes Venatici there are some interesting objects to capture over the next month or so. All except the M3 globular cluster will probably work with the above exposure settings.

canes_venatici.JPG
canes_venatici.JPG (44.38 KiB) Viewed 1072 times


Dave
User avatar
oopfan
Posts: 1321
Joined: Sat Jul 08, 2017 2:37 pm
Location: New York
Contact:

Re: M51 HaLRGB project

#2

Post by oopfan »

Dave,

Thanks for the H/T. Those exposures I recommended were for my Bortle 5 skies, so your SNR values will be lower under Bortle 6. Also, given my latest research, I recommend that you run with an "SNR per sub" that is greater than 2. Like this:
Dave M51.jpg
Dave M51.jpg (58.82 KiB) Viewed 1065 times
Brian
User avatar
turfpit
Posts: 1779
Joined: Mon Feb 13, 2017 8:13 pm
Location: UK
Contact:

Re: M51 HaLRGB project

#3

Post by turfpit »

Thanks Brian. I will run with the amended figures. That saved me 20x100s of dark frame capture.

Project update: I now have 20x600s & 20x300s darks. Good use of the greenhouse on a rainy & windy night. Now I need 2 clear nights.

Dave
User avatar
turfpit
Posts: 1779
Joined: Mon Feb 13, 2017 8:13 pm
Location: UK
Contact:

Re: M51 HaLRGB project

#4

Post by turfpit »

It took 3 nights of cloud dodging but at last ...

M51 - summary.png
M51 - summary.png (754.84 KiB) Viewed 971 times

  • After performing an 'RGB Colour Combine' in Astro Pixel Processor, there was no need to carry out any colour balancing in software. Colour balance was achieved by stacking appropriate numbers of RGB frames.
  • At each session, the frame was centred on the core of the larger galaxy. The full frame above has not been cropped, indicating the PHD2 guiding was working satisfactorily over the 2.7 hours and 3 nights of capture.
  • An extra 2 frames were captured for each of RGB, giving the flexibility to cope with planes, meteors, satellites and intermittent cloud.
  • The L frame count was supposed to be 40 but I had to discard 23 frames (I captured 50) due to intermittent cloud forming/clearing.
  • After exporting the 32-bit TIFF from APP, GIMP was used to adjust the black point and contrast with a gentle denoise applied using NEAT Image. All APP settings were defaults.
  • The total capture to date is 930Mb.

Next tasks will be to capture all 300s frames with the RGB binned 2x2 at capture and 1 hour of Ha (6x600s).

Time to reflect on my first M51, captured December 2016. No darks, no flats, poor polar alignment & bad focus.

M51-December-2016.jpg
M51-December-2016.jpg (10.64 KiB) Viewed 971 times


Dave
User avatar
turfpit
Posts: 1779
Joined: Mon Feb 13, 2017 8:13 pm
Location: UK
Contact:

Re: M51 HaLRGB project

#5

Post by turfpit »

Captured with L @ bin1x1 and RGB @ bin2x2.

summary.png
summary.png (654.33 KiB) Viewed 924 times

  • 2.5h of integration, I am tempted to see what visual improvement doubling the integration would make.
  • Only the Ha frames to capture now.
Dave
Post Reply