Eta Carinae Nebula (NGC 3372)

[Flugel88]

My first decent image with my atik 383L taken 30/12/14

Now that i'm using a larger chip than my Asi 120mm i've had an introduction to the dreaded coma :(.
I think i need to invest in an MPCC for my Meade Schmidt Newtonian.

All up i'm pretty happy with the atik 383 my only wish would be for faster download speeds to make image centering and focusing easier i tend to spend way to much time trying to get that all right.

http://www.astrobin.com/144827/

LRGB
40 40 40 40 (10min subs)
Darks Bias Flats

[Amaranthus]

Very nice Michael - it's a spectacular object, but surprisingly difficult to get to look 'nice' in photographs, I think due to the inherent graininess of the nebulosity/stellar juxtapositions. This is a really fine balance that you've achieved really well. Isn't it fun imaging in mono? :)

[Flugel88]

Yeah not wrong Barry I made 4 versions on this all with efferent levels curves.I think this was the best one.

It is hard to get a nice balance of detail without blowing it out with to much intense colour.

Certainly is a very Bright target i might have over done it a little with 10 min subs at F4.

[Amaranthus]

Michael, have you checked out Ray's advice here on optimal sub length?
http://www.iceinspace.com.au/forum/showthread.php?t=117010

You might be surprised how short the ideal subexposure length can be for LRGB, especially given that you are imaging at f/4 with a 10" aperture. I suspect it might be only 2-3 minutes for you.

[Flugel88]

Thanks for the link Barry i just roughly worked out from that formula to aim for 2981ADU.


targetADU = Bias + 10*RN*RN/camera_gain
Not sure if this is correct!

351+ (10x11x11)
______
0.45
=2981 ADU

and here is what averaged for each channel in the image at 10min subs

L=15,000
R= 3,500
G= 3,800
B= 4,100

So does that mean i have overexposed the light frames with around 5x to much exposure?
Does this just mean that i'm not getting any extra data with the longer subs (wasting time) or is it more about over blooming stars and colour balance?

Thanks for helping me out with this BTW :D

[Amaranthus]

Assuming a linear accumulation of photons, yes, it means your subs are unnecessarily long, and about 5 times too long for your Lum.

Now although you are not losing data (faint stuff) by going longer, but you will be increasing the fraction of sky background in your total ADU (I assume that figures you quoted above are for the darkest parts of your image), and you won't be getting any significant benefits in read-noise compensation since you're way above the floor at this point.

Yes, you are also risking blooming (or just plain saturation of brighter objects, since your camera has anti-blooming) and making your life more difficult in terms of tracking/guiding. More subs also help in terms of statistical calibration and drizzling etc.

For colour balance, you need to map exposure length against your filter transmissibility etc. - different issue. But you can set each colour exposure length against the target ADU (2981) and then adjust the number of subs you take with each channel to get the colour balance you desire. Or simply tweak it in post-processing (or both!)

BTW, have you tried binning the 383L+ when you are centering, and putting it in 'preview' mode, for faster DLs?

[Flugel88]

Not to sure about that i'm just reading from Sequence generator image stats as shown in attachment.

So for my next target ill be aiming for more subs less exposure specially with luminance.

Wow i didn't know that i really am a novice so much to learn.

I've learned a tonne from you today barry thank you
I cant wait to get out there and try some of these better methods.

[Chris85]

Great image, Michael!
I use my 383 with SGP like you and I find using bin 3x3 with 0.3s exposures, sub-framing, then zooming in on a single star a great way to get focus right. Here's an example from tonight :)
Getting a motorised focuser helps too
Cheers!
Chris

[Amaranthus]

Michael, looking at your image stats, you can see that the min ADU is 5,879. This is an unreliable figure for your background, however, because it might reflect a relatively insensitive pixel. However, it will be closer to the background than the mean figure of 15,050. I'd say your background is probably around 6000, but hunt around the image and check that out. This suggests that an exposure for L in the range of 4-6 min might be suitable.

Do you image at a dark-sky location? I know Cessnock is in a rural area, so if you're away from the town a bit, then your sky background could be quite low, except in the direction of the Newcastle light dome.

[Flugel88]

Thanks for the tips Chris :thumbsup:
I had no idea that binning sped up the download process until Barry mentioned it.
It sure will take out allot of the frustration i have with focusing and centering image.
I do have an old motorized JMI focuser that has a shoestring USB connecter does make it much easier.



Barry i took a better look at my Lums.
I found most background areas hovering around the 12k to 14k Mean in the fluorescing gas parts and around the 9k for the darkest areas.




Actually pointing towards Newcastle has the darkest skies for me as i live close to a prison so the North and North west i don't even bother trying to image near there.