Eagle Nebula Test and Failure?

[spyonthesky]

I am coming up to speed with Astrophotography, image stacking and using StarTools. The good news is I have accomplished a lot, given where I was a few months ago at square one! Here is what I have going on:

Gear:

• Celestron HD800 with CGEM Mount
• Canon 50D doing Prime Photography
• Astronomik CLS Clip-in for Canon
• Backyard EOS
• PHD2

Accomplishments:

• Successful polar alignment and reasonably precise tracking (no streaks after 10 minute exposure)
• Successful with Dumbbell and Surveyor Galaxy (see below)
• Success capturing darks, flats and bias frames
• Used DeepSkyStacker 3.3.4 to stack raw .cr2 files with darks, flat and bias
• Used StarTools 4.1 64-bit to resolve Dumbbell to a decent quality (see below)

I am shooting from my yard in Old Town, Maine and the ambient light is low relative to a big city, but Bangor is not far away and I get a lot of local light pollution from various sources. Here are the images that I am pretty happy with:

Survey Galaxy with 8 exposures ranging from 150 to 600 seconds at ISO 800 with camera noise reduction turned on and darks used. No flats and no bias.

SurveyGalaxy_s.jpg (350.62 KiB) Viewed 5481 times

Dumbbell Nebula with 19 exposures ranging from 60 to 400 seconds. Four at ISO 1600, twelve at ISO 800, two at ISO 400 and one at ISO 640 with camera noise reduction turned on and darks used. No flats and no bias.

DumbellNebula_s.jpg (128.04 KiB) Viewed 5481 times

The problem I am seeing is that some of my images are noisy with a lot of red noise/lines/bands. I tried a few different things with the Eagle Nebula and the images just seem impossible to get anything useful from. I want to understand why and what I can do to fix it. Even the Survey Galaxy should be a lot better I know.

As a test, I shot a Galaxy (not sure which one) with 30, 90 second exposures. Camera noise reduction off at ISO 2000. I shot bias, dark and flats. I thought higher ISO and shorter exposure would solve any noise problems. But I saw the same noisy red mess in my shots as if the signal is too low which suggests I need longer exposure.

Here is what I see in StarTools for my Eagle Nebula I am trying to resolve:

noisyEagle.jpg (509.12 KiB) Viewed 5481 times

My workflow to this point was autoDev, wipe with vignette, autoDev redo-stretch (darks, bias, flats)

Here is the Eagle Nebula .FTS file that came out of DeepSkyStacker with darks, bias and flats applied (linear, auto-wb):

http://www.oakleafs.com/startools/eagle_nebula_fts.zip

I do not know where to begin solving this problem. In fact, I am not even sure there is a problem! I would appreciate your help telling me if something I am doing during the capture phase is causing my raw material to be bad from the start. Or, maybe what I capture is okay, and the problem is in the stacking or in my StarTools workflow. The score values I get from DeepSkyTracker are between 15-180. I never get any score higher than 200 (not sure if this is important or not).

Here is a zip file with 4 images I shot at different ISOs and different duration in an attempt to narrow in on the problem. For these shots, I had camera noise reduction turned on. I have since turned that off, but either way, the photos end up looking the same.

http://www.oakleafs.com/startools/EagleNebula.zip

Just for companion, here are my raw shots of the Surveyor Galaxy which turned out better as you saw above. This was done with 8 exposures and a duration between 150 seconds and 600 seconds, all at ISO 800. Camera noise reduction was turned on. I had darks, but no bias or flat.

http://www.oakleafs.com/startools/SurveyGalaxy.zip

So to summarize:

• What do you suggest I use for exposure duration, ISO and number of exposures for my setup?
• What should I do In StarTools? I have tried several of the length workflows you have shared to help others. But thoes do not help me. Just looking at the first 3 steps in the workflow, I see the following:
• ---> From YouTube, I see autodev, wipe, autodev (redo stretch). This always leads to a useless image for me
• ---> Manual Develop, digital development to 80, wipe, mess a lot with colors, sharpen (to get my dumbBell)
• Any other tips or pointers to get me on the right track for making my long nights more effective?

Thank you

Bryan

[almcl]

Bryan

I am far from an expert (as numerous queries here will attest) nor am I familiar with your particular camera, but I think ISO above 800 will add noise. To test this I have stacked your two raw file that were shot at ISO 800 and processed them in ST. I think the result (below - it's cropped quite a bit) may be better, but see what you think? The red level in the images is quite high and makes me wonder if the clip filter was fitted for both images?

The other thought (and I don't know your location, so this may not be relevant) but the Eagle is fairly low down in Europe at the moment and any upper cloud plus city lights will tend to produce really noisy images.

Flat frames and some bias may also help. The EXIF data for both of your files I used shows Noise reduction on. Not sure if you can turn that off? It might help.

Have you tried shooting at ISO400? I've read suggestions elsewhere that earlier Canons were not as noise resistant to high ISOs as their current offerings, and at least one writer experimented to find out where unity gain occurred and concluded it ws between 400 and 800. Anything above this is merely amplifying noise as well as signal, to no advantage.

2 ISO 800 raw files reworked

AS.jpg (450.92 KiB) Viewed 5466 times

[admin]

Hi Bryan,

What do you suggest I use for exposure duration, ISO and number of exposures for my setup?

Get the longest possible exposure times you can get away with without blowing out the image. Roughly speaking (there are some other subtleties to take into account), beyond a 'native ISO' for your camera increasing ISO will simply multiply the signal, only causing the image to blow out quicker, but you're not gathering more signal. For your particular model I probably wouldn't go higher than 800.
In-camera noise reduction is the same thing as shooting darks, so it's not needed if you have a dark library already. As a matter of fact, if you do in-camera noise reduction and apply darks, your are introducing noise! (as your effectively applying darks twice)

In the future, please refrain from mixing different exposure times; they do nothing for astrophotography but cause noise and artefacts, except in the extreme case of M42 where the core may blow out. In any case, they will make StarTools' noise propagation tracking impossible as noise and signal quality will vary all over the place.

The lines you're seeing are called 'banding'. Bias frames will help with countering them to a degree. You may also benefit from using the Band module which was created to mitigate exactly the problem you're experiencing.

What should I do In StarTools? I have tried several of the length workflows you have shared to help others. But thoes do not help me. Just looking at the first 3 steps in the workflow, I see the following:
---> From YouTube, I see autodev, wipe, autodev (redo stretch). This always leads to a useless image for me
---> Manual Develop, digital development to 80, wipe, mess a lot with colors, sharpen (to get my dumbBell)
Any other tips or pointers to get me on the right track for making my long nights more effective?

Have you seen the quick start tutorial?
They will also show you the correct settings for DSS; your current settings also exacerbate noise and make noise tracking harder for StarTools.

The Eagle stack is exceptionally noisy and I'm hoping it's a stacking issue (how many frames is this for lights, darks, flats and or bias, and what are the exposure times?)

Cheers,

Ivo

[Rowland]

CLS filters usually require longer exposure times. Astronomik, the last time I spoke to them on this subject, recommended sufficient exposure time to saturate the background. With short exposures the background will return a lot of noise. I recall exposures of 4 - 5 minutes at least, more if needed, in a semi rural setting- on the fringes of the suburbs.

Backyard EOS has dithering. It's a good idea to dither DSLR images. You will find it reduces noise and fixes a number of other issues.

Your Eagle image is dominated by background noise (try series of exposures at iso800 of 3 - 4 minutes with the CLS filter removed and compare the background with the images above). If you are getting 10 minutes tracking, bump all your exposures up to 6 minutes with the filter installed and see if it makes a difference.

You are right to turn off the in-camera noise reduction, but if you do use it you will need flats and a set of bias frames with which to calibrate the flats.

[admin]

Very good point Rowland.