Should I Lower My Expectations?

[niteman1946]

Here are a couple of images of NGC6946.  Both are the same image, with the first un-stretched and the second stretched.  This represents 180 minutes (3 hrs) of integration at 10 minutes per sub with the Atik 383L+ mono camera.  The scope is a 12”LX200 classic with a F7.0 FR/FF and an Orion Skyglow filter.  The image was calibrated and integrated using PixInsight.

Integrated and un-stretched,
http://astrob.in/19070/

Integrated and stretched,
http://astrob.in/19072/

I recognize that NGC6946 is a dim object, but the Atik is looking through the same setup as was my modded 450D Canon, and that camera captured the image as well, if not better, and in color.
I was really expecting more from 10 minute exposures with the Atik.  Maybe over-exposed stars and the galaxy jumping out at me.  Instead, this is the best I can get with a hard stretch.
There is also something peculiar going on in the Histogram curve.  Both individual subs and the integrated version show a 24% feature at the far left side of the curve.  That should not be there, and may be the cause of this anemic performance (IMHO).

Histogram curve,
http://astrob.in/19074/

Any suggestions or recommendations would be welcome.

Mark

[andysea]

Hi Mark,
Here is my image with the 383L+ color. Same object.
http://www.flickr.com/photos/andyinsea/7835112552/in/photostream
The states are exactly the same as your image. 3 hours of integration time at 10 min subs. Shot at f8.
I would expect the mono camera to be way more sensitive than my color one.
I can't really assess your result but this will give you a comparison.

Andy

[niteman1946]

Thanks for the response Andy.  I’ve been following your saga in the various fora and hope you get your issues worked out.   I also hope I’ve not stepped in it with the 383L+.  Nice image BTW.

Just a couple of questions.

1. Do you find the Atik does gather much more light than the DSLR?  Mine does not appear to.

2. There is something wrong with my images evidenced by the histogram.  There should not be the data over on the far left.  Having two humps is not right.

Histogram curve,
http://astrob.in/19074/

Mark

[joelshort]

Mark,
Did you take flats, and if so how long were your flat exposures?  In looking at your stretched image, I noticed in the upper left corner that there is a lighter area that extends into the center of the image for about 1/4 of the overall image.  I get the exact same pattern in my images.  For a  long time I've simply assumed that this is caused by light pollution gradient.  However, recently I've begun to think that this is actually caused by improper flats.  I haven't tested this yet, but take a look at the attached LUM flat that I took recently (it is over-stretched to show a darker area in the upper left corner).  I think this darker area might be shutter shadow caused by too short of flat exposures.  What happens is that when the flat gets applied to the light frame, the improper flat "overcompensates" and shows up as a light area in the image.  I plan to take longer flat exposures (like 2-3sec) and see if that dark corner still shows up in the flats.  If it doesn't, then the dark corner was caused by shutter shadow and would have adversely affected the final image after calibration.

Now about your image of the fireworks galaxy, this is a very faint object so I am not really surprised at your image.  If you skies and mount will allow it, I would push the exposure to 15min and you will also need longer over-all integration time to increase the signal to noise ratio.  For LRGB images I usually try to get a minimum of 5hrs LUM and 2hrs each of RGB. 

HTH,
joel

[andysea]

Thanks for the response Andy.  I’ve been following your saga in the various fora and hope you get your issues worked out.   I also hope I’ve not stepped in it with the 383L+.  Nice image BTW.

Just a couple of questions.

1. Do you find the Atik does gather much more light than the DSLR?  Mine does not appear to.

2. There is something wrong with my images evidenced by the histogram.  There should not be the data over on the far left.  Having two humps is not right.

Histogram curve,
http://astrob.in/19074/

Mark

Hi Mark
Yeah I've had a few issues with mt 383. The most annoying one is definitely the unexpected noise due to  the supplied power being less than 12.5v.
I'm not sure that the atik actually gathers more light than any of my dslr's. My Ccd however is OSC, the mono version should be more sensitive. The biggest advantages for me are the regulated cooling and the absence of pattern noise.
I looked at your histogram, my first reaction is to agree with you that there shouldn't be two bumps, however I will look at mine and let you know if I get the same type of curve. I don't have access to my images now but I will be able to get to them in a couple of weeks.
Joel's point is well taken. I have not done flats so far but I have a flat panel on order and I will start doing that. The shadow that I see in his flat definitely looks like its' generated by the shutter.

I have a friend who uses the qsi 683 mono and his camera seems to be way more sensitive than mine, I would expect the 383 mono to be on par as far as sensitivity goes.

Andy

[niteman1946]

Did you take flats, and if so how long were your flat exposures?  In looking at your stretched image, I noticed in the upper left corner that there is a lighter area that extends into the center of the image for about 1/4 of the overall image.  I get the exact same pattern in my images.  For a  long time I've simply assumed that this is caused by light pollution gradient.  However, recently I've begun to think that this is actually caused by improper flats.  I haven't tested this yet, but take a look at the attached LUM flat that I took recently (it is over-stretched to show a darker area in the upper left corner).  I think this darker area might be shutter shadow caused by too short of flat exposures.  What happens is that when the flat gets applied to the light frame, the improper flat "overcompensates" and shows up as a light area in the image.  I plan to take longer flat exposures (like 2-3sec) and see if that dark corner still shows up in the flats.  If it doesn't, then the dark corner was caused by shutter shadow and would have adversely affected the final image after calibration.

Hi Joel and Andy,

Yes I did take flats and the gradient you referred to did show up in my integrated image.  However, PixInsight has an excellent tool for this called Dynamic Background Extraction.  And this fixed the problem.  My guess is this was a light gradient from the city lights I was shooting into.

With this new camera I am inventing technique as I go along.  My first Flat Master was built from images taken at dusk and for exposure periods from 1 to 7 seconds.  I simply (quickly) reviewed the image properties after each shot and tried to stay around 35,000 ADU.  Admittedly, this was very crude with individual values bouncing between 20,000 and 50,000.

In review of the Master Flat, I may be getting some OAG prism shadow, but it is subtle.

My issue with the weird histogram was the result of my registering the calibrated subs to a previous DSLR sub of the same target.  Differences in FOV caused a large black border around the registered subs.  This caused the histogram to go nuts, and kept stretching from operating properly.  I fixed that by registering with one of the Atik subs.  All went well after that.

I'm getting very clean images from the Atik with good dynamic range,  and will get better at composing and focusing.

Here's my work in progress:
http://www.astrobin.com/full/19326/?mod=none

Thanks for your help,

Mark