help Help - I Don't Know What I Did Wrong :( - Andromeda

[Chuleton08]

Hi Guys,

Last night, I attempted to take a series of 520 photos with a exposure time of 2 seconds of the Andromeda Galaxy, but got this(Click on it) when done stacking and post processing. I was using a Canon 400D with a 17-200mm sigma lens, with the length being at 200mm, and a f/5.6 (The lowest I could go). I took all the other frames(eg. light, dark bias...) and used a remote to take the photos. I think it could be because of the light pollution, or I just need more frames? I took the picture in West Sussex, England.

If anyone could help me find out what's wrong, or needs more information, please comment on it, and I will get to you ASAP.

Thanks!

[Swillis]

Hi,
@Chuleton08 it might help others answer your question if you could give more info on which software you used for stacking and post processing?  Also what steps you took with post processing would be helpful to know.

Also if you could share the output file from you stacking software that would also be great. Others will then be more able to identify where any issues are.

Thanks

Simon

[Chuleton08]

Hi Simon,

Thanks for replying. I used deep sky stacker, and I was following all the steps in the video: 

 

 

I think this is the file you are looking for:

m31.tif

[Elp]

As part of your PP have you done extensive levels stretching on the linear stack, it doesn't seem like it as the stars are still very dull.

2s is also very short at F5.6, though I understand your max exposure time may be limited without tracking. When I first did Andromeda on an alt az mount I took 9s images and over 300, the galaxy was there but it took a lot of PP work to get it out of the linear data. I think you do need a lot more time, luckily Andromeda is quite bright, but also not in the best position in the sky at the moment as it rises quite late (compared to winter when it's already high in the sky at dark).

[Chuleton08]

Thanks Elp for the reply,

This is my first time doing astrophotography, so I'm still learning. I'm currently trying stacking again with Siril, hoping that the PP is going to be a bit better. Do you think its worth sacrificing the 2sec exposure time for something a little longer? If so what, and why.

Thanks!

[Elp]

What mount?

Something you can do in Siril straight away with the DSS stack is open it, then at the bottom of the preview screen is a drop down menu, it'll be set on linear initially so the image will be black unless it has been pre stretched, change the mode to histogram which will show in extreme stretch what data is present so you'll know if there's more to bring out. The preview mode doesn't alter the data, it just previews it different, auto stretch is also akin to a finished image stretch.

I used to use DSS but moved to Siril as it also has a suite of PP functions which are essential to getting good use out of your data like background extraction which is essential to enable you to push the stretching when working on starless data.

[Chuleton08]

This is what I got. What does this mean then? I still want to try from scratch with the raw files, and process it with Siril though, I think there will be better results right?

[Elp]

If that's the histogram preview there's not enough data, youve only got the core at the moment. You need more total time, maybe at least 2 hours on M31.

This is my first attempt at F5.9 through my Z61 with the galaxy high in th sky (so you get better separation from LP), 51 minutes:

And this is 2 hours with a fast lens, equivalent at F5.6 might be quadruple the time needed at least:

Id still continue with the data you've got, if only to familiarise yourself with Siril, it also has a banding reduction feature which you can apply to the stack to help with the horizontal lines you have typical from a DSLR, I sometimes also use it on astro camera data if I've got the banding. You can then combine each "master" stack from multiple sessions to add up all your imaging time, restacking everything will give better results but I usually stack the stacks and have never had an issue with it.

 

[Chuleton08]

Wow, those photos look amazing!

I don't have a sky tracker, do you think I can get away with something like a 8-9 sec exposure time? Also so now is not a good time to be taking photos as its too low? Or could i just go somewhere darker than my back garden and still get a decent photo?

[Chuleton08]

Got this at the end when using Siril. Does this mean my photos were bad to begin with? 

[Elp]

Thanks, it's a continual learning process, the first one is when I just bought my azgti and was still using it in altaz mode, the second by then I was using the mount autoguided in EQ mode so could image per sub for longer.

Do some tests, I think at 200mm you'll struggle to go past what you're already doing without star elongation. When I was using a 50mm last I don't think I could even do 10s on a fixed mount. You really need a star tracker or tracking mount for deep sky work.

Darker skies help tremendously, but from memory when it's dark ATM (remember it's not astronomical dark at present, have to wait until September or later for that to come back if in UK), Andromeda is quite low near 10 degrees or so, so you're imaging through a lot of atmosphere which will also cause your images to blur more. It's not ideal to image at present. With your setup Id try open star clusters because stars are bright.

[Elp]

4 minutes ago, Chuleton08 said:

Got this at the end when using Siril. Does this mean my photos were bad to begin with? 

Might be your registration algorithm is wrong, needs to be deep sky global if I remember correctly, you can look at your images if you've put the lights into a sequence via the file list button (bottom right) and you can down arrow key through them all whilst in histogram preview to see if the stars are round, which they need to be, some ovality will also register. Whilst you're in the file list make sure the registration image is good otherwise they won't register, by default the first image is the master registration file, you can simply scroll down to another file you like the look of and check the use as register check box at the top of the dialogue box.

Edited May 20 by Elp

[TiffsAndAstro]

31 minutes ago, Chuleton08 said:

Thanks Elp for the reply,

This is my first time doing astrophotography, so I'm still learning. I'm currently trying stacking again with Siril, hoping that the PP is going to be a bit better. Do you think its worth sacrificing the 2sec exposure time for something a little longer? If so what, and why.

Thanks!

Try it and see. Take a 5 sec at a high iso then zoom in look for star trails/elongated stars. If none use 5 sec. If you do see trails etc drop to 4 sec, and look again.

I'm very new to this but I think I saw trails at 50mm crop frame at about 5 seconds or more. You're at 200mm so if you don't see trails at 2 sec I say you're doing well I think.

I think iso800 might be best when taking your subs

[TiffsAndAstro]

27 minutes ago, Chuleton08 said:

This is what I got. What does this mean then? I still want to try from scratch with the raw files, and process it with Siril though, I think there will be better results right?

Check out deeps space Astros siril tutorials on YouTube. Quite amazing

[Chuleton08]

1 minute ago, TiffsAndAstro said:

Try it and see. Take a 5 sec at a high iso then zoom in look for star trails/elongated stars. If none use 5 sec. If you do see trails etc drop to 4 sec, and look again.

I'm very new to this but I think I saw trails at 50mm crop frame at about 5 seconds or more. You're at 200mm so if you don't see trails at 2 sec I say you're doing well I think.

I think iso800 might be best when taking your subs

That's a good idea. I will hopefully do some testing 

[Chuleton08]

7 minutes ago, Elp said:

Thanks, it's a continual learning process, the first one is when I just bought my azgti and was still using it in altaz mode, the second by then I was using the mount autoguided in EQ mode so could image per sub for longer.

Do some tests, I think at 200mm you'll struggle to go past what you're already doing without star elongation. When I was using a 50mm last I don't think I could even do 10s on a fixed mount. You really need a star tracker or tracking mount for deep sky work.

Darker skies help tremendously, but from memory when it's dark ATM (remember it's not astronomical dark at present, have to wait until September or later for that to come back if in UK), Andromeda is quite low near 10 degrees or so, so you're imaging through a lot of atmosphere which will also cause your images to blur more. It's not ideal to image at present. With your setup Id try open star clusters because stars are bright.

I will hopefully be able to take my camera in August to the south of Spain, and in the middle of nowhere with little to none light pollution. Hopefully by then, I would have learnt how to take the photos

[Chuleton08]

7 minutes ago, Elp said:

Might be your registration algorithm is wrong, needs to be deep sky global if I remember correctly, you can look at your images if you've put the lights into a sequence via the file list button (bottom right) and you can down arrow key through them all whilst in histogram preview to see if the stars are round, which they need to be, some ovality will also register. Whilst you're in the file list make sure the registration image is good otherwise they won't register, by default the first image is the master registration file, you can simply scroll down to another file you like the look of and check the use as register check box at the top of the dialogue box.

Thanks I will have a look

[TiffsAndAstro]

6 minutes ago, Chuleton08 said:

That's a good idea. I will hopefully do some testing 

Btw m31 gets higher in the sky over the next month or two, so you're fighting a lot more horrible atmosphere to image right now. Also the moon and technically no proper darkness.

I'm new, and desperate to image m31 with my new gear, but it's still too low for me right now.

Having said that, it's probably ideal to practice and get data to practice with

Edited May 20 by TiffsAndAstro

[Chuleton08]

Just now, TiffsAndAstro said:

Btw m31 gets higher in the sky over the next month or two, so you're fighting a lot more horrible atmosphere to image.

I'm new, and desperate to image m31 with my new gear, but it's still too low for me right now.

Having said that, it's probably ideal to practice and get data to practice with

Isn't the higher it is the less atmosphere there is, and the lower it is, the more atmosphere? Its always good to practice.

[Elp]

Yes, that's what they said.

[Chuleton08]

silly me 😅

[Chuleton08]

Thanks for everything everyone !!!

[900SL]

The following may help:

2s at 200mm fl on an APS-C sensor is both too long and too short. What do I mean?

Too short at f5.6, you are not getting above the noise floor (data is clipped).  I don't know if you can check the histogram on that camera on a single unstretched sub, but i suspect it is hard left.

Too long at 2s, rule of thumb untracked for APS-C is 300/200 so 1.5s max, and your target at 40 degrees Dec so not stationary and also low on the horizon, so shooting through the murk

The lens and camera are not the best for untracked astrophotography. The aperture is too narrow at f5.6, and you are likely to have a lot of lens artifacts like coma, astigmatism etc

Options are:

1: Use a tracker, although 200mm fl may require guiding unless you have a good quality tracker like the Fornax LT2 (but costs more than your camera and lens)

2: Go to a shorter focal length wider aperture, like the Samyand 135mm f2, and shoot from darker site. This will capture data 8 times faster at 3 stops down, and then you can also increase the exposure duration

 

What ISO were you using? Canon DSLR are not ISO invariant

 

 

[900SL]

This is what I got from a Bortle 3 site (not much light pollution) with a Nikon D5600 and 135mm Samyang at f2.8, needed 100s exposures at ISO 800, back in my beginner days.

So f2 would halve this, and ISO 6400 would get you down to around 6s exposures, so you might be able to get something at 3s exposures from a darker site untracked, at higher elevation and not too close to the celestial equator

Dealing with noise will be the big issue

[900SL]

This is the best I can get using your stack. The data is not there to bring out, it was buried in the noise floor (although one of the processing whizzes here might be able to draw something out..)

 

Edited May 21 by 900SL