M31 (feat. M110) - Andromeda Galaxy - first AP experience

[Alex E]

So I promised I'd come back with something, here I am. The two shots below are my first two experiences with astrophotography. I was happy but a little bit underwhelmed with my first one (as often happens I guess).. and then I gathered more light or my second and boy was I mindblown when I finished processing the result.

 

Some first lessons learned:

- take dark/flats/etc

- some star trailing in the second session which I only saw when I started processing so I think I need to polar align more precisely and pay more attention to test shots

- probably need to get balancing figured out a bit more, not sure if that was the cause for the above

- get my tripod anchored down (I shoot on wooden decking so not exactly the most stable and vibration free surface though I expect at this level not a massive issue yet)

- I need to either buy a right-angle viewfinder for the polar scope OR a new neck

- spotting scope and guided tracking at some point when I can afford more expense would be _very_ nice

 

Equipment: William Optics Redcat 51 APO, Canon EOS 700D, SkyWatcher Star Adventurer Pro

Software: APT, Topaz DeNoise AI, Adobe Photoshop CS, DeepSkyStacker

 

First session - 20 x 30s lights (17 used), no flats/darks/bias, no noise removal

 

 

Second session - 30 x 2min lights (23 used), 15 darks / 50 flats / 100 bias

 

Edited September 7, 2021 by Alex E

[vlaiv]

Slightly blown core - maybe it was over exposed in 2 minute image? If not - try to tame it a bit.

Background is too dark and clipping. Try to avoid histogram clipping to the left when processing the image.

Otherwise very good start.

[Alex E]

2 minutes ago, vlaiv said:

Slightly blown core - maybe it was over exposed in 2 minute image? If not - try to tame it a bit.

Background is too dark and clipping. Try to avoid histogram clipping to the left when processing the image.

Otherwise very good start.

Absolutely agreed on the core, I might have another go at it to see if I can even that out, but I might need to layer mask some shorter exposures on my next go at it as it does look overexposed in the lights.

Also, appreciate the comment on the background, you're absolutely right there as well, I need to learn to pay more attention to the histogram. Working on sorting that out now.

[Alex E]

Right so on above solid advice (thank you!), here's a revised version of session 2. Obviously some shorter exposures to use for the core masking would probably be far better but I think I was able to tone it down decently by just masking with the original stacked image for the core region. Resolved the clipping as well and of course some more lovely detail is visible now. Flipped for better perspective.

[vlaiv]

5 minutes ago, Alex E said:

Right so on above solid advice (thank you!), here's a revised version of session 2. Obviously some shorter exposures to use for the core masking would probably be far better but I think I was able to tone it down decently by just masking with the original stacked image for the core region. Resolved the clipping as well and of course some more lovely detail is visible now. Flipped for better perspective.

If your core is not blown in stack - then there is no need to do layers / masking. Simple stretch can preserve core and reveal faint structure.

Yes, background is now showing nice histogram in each channel - only tiny issue - it is too red. Red channel has histogram peak at 20 while blue and green at 17.

Did you use some sort of synthetic flat fielding to correct for vignetting? It looks like red channel has some sort of "dip" around galaxy?

[Alex E]

Yeah so quite obviously I am still learning the processing bits and have fiddled with the Astronomy Tools action set for PS CS on it. There's a few things in the that I ran based on a checklist from astrobackyard, but again, lots of the stuff I'm doing is half guesswork half bits picked up from various places. I'm guessing there isn't a pre-set list of things you would go through since different DSOs would require different processing but do you have any preferred guides or how-tos in terms of processing post-stacking?

I reduced the red channel down once but I'm guessing it needs a bit more.

[Alex E]

There we go, we're at 17/17/16 now.

Edited September 7, 2021 by Alex E

[Grant93]

Definitely an impressive improvement from the first to the second attempt! I'd be well happy!

[irtuk]

That's looking pretty good, a smidge on the green side maybe.  Looks like we are getting another clear night down here and one of the joys of astro is you can keep adding to your data year on year and still stack it together so long as you are using the same gear.

Big up the East Kent massive btw

 

Ed.

 

[vlaiv]

3 minutes ago, Alex E said:

Yeah so quite obviously I am still learning the processing bits and have fiddled with the Astronomy Tools action set for PS CS on it. There's a few things in the that I ran based on a checklist from astrobackyard, but again, lots of the stuff I'm doing is half guesswork half bits picked up from various places. I'm guessing there isn't a pre-set list of things you would go through since different DSOs would require different processing but do you have any preferred guides or how-tos in terms of processing post-stacking?

I reduced the red channel down once but I'm guessing it needs a bit more.

Yes, I understand, it is quite normal to follow a tutorial for some time at start and later you'll develop workflow that you prefer.

I personally base my processing on data and math rather than prescribed set of actions and I don't want to confuse you at this stage with that sort of thing. I can give you some useful advice though.

Try to perform proper calibration of your data - it yields the best results. Do actual flat frames rather than relying on synthetic flats (that won't help in this case since you already captured the data - but for future). Do darks as well.

General workflow that I would recommend as beginner workflow would be:

1. wipe background. Unfortunately I can't recommend any particular tool for this purpose - I use plugin for ImageJ that I wrote for this purpose. I think that background should be wiped at linear stage

2. Do any color corrections you need next.

3. Color compose and do simple three step levels stretch. First step is to move right slider left until you are about to saturate brightest parts of the image. In this case - you would look at core of M31 and stop while it is still point like without blowing it.  Apply levels. Next round is to move middle slider to the left until you show faint bits properly. This will expose background quite a bit - but don't worry about it now. Apply again. Last step is to move left slider at the foot of histogram without clipping it. This will return background to normal.

4. Do selective noise suppression.

Here is post I made some time ago that demonstrates some of the steps in processing:

 

[Alex E]

Thanks for the suggestions, very much appreciated, it's a learning curve but it's also great fun.

I did have darks and flats for the second session, but I saw some variable horizontal banding on my flats when I looked at them this morning so I think I need to work a bit more on taking good flats as, like you say, that's much preferred to synthetic methods of removing vignetting and other gradients.

 

[Alex E]

26 minutes ago, irtuk said:

That's looking pretty good, a smidge on the green side maybe.  Looks like we are getting another clear night down here and one of the joys of astro is you can keep adding to your data year on year and still stack it together so long as you are using the same gear.

Big up the East Kent massive btw

 

Ed.

 

My ClearOutside says it's going to go dark after 8PM today unfortunately, but not sure if that's just the coastal weather out here and Canterbury might be spared. Are there any local AP groups/gatherings that sort of thing?

Big up indeed

[Alex E]

Right I think I managed to get the greenish tinge out so I shall leave it here and return with other (mis)adventures and questions when clear skies avail next. Thank you everyone for your input.

[irtuk]

rats. ClearOutside is indeed reporting clouds coming in at about 8. it's changed it's mind since lunch so, there's always hope!

[Alex E]

So I decided to have another run at this as the last result just didn't sit right and I think ended up on a slightly better result - core looks better and the balance is better color-wise, managed to remove more of that left gradient and it just looks closer in color to the standard. At the very least it gave me a bit more practice with stretching and playing around with my curves. Detail-wise there's not much more there but I'm still happy with it until I get another clear night and more exposure time.

[pete_81]

Nice job Alex!