What is causing these color pixels - 383L OSC

[eoverstreet]

I have had this camera for a year plus!  Can anyone help me with "what could be causing these red green and blue pixels.  This is not the noise I am accustom to.    This is just 10 subs, no darks, flats, or bias frames but it doesn 't matter if I process with supporting files, I always get these pixels.   

Thanks for your help!

[Chris]

My first thought is that they are simply hot pixels and a dark frame subtraction would handle them.  The pixels are coloured because it's a colour camera.

Buy you say that it's the same with dark subtraction, I would expect the same pixels to be hot on the dark frames and so subtracting the dark frame would remove them.

Could there be some misunderstanding in how you are subtracting the darks?  The simplest way is to take dark frames with the same exposure and at the same temperature as the lights.  Average the raw dark frames then subtract the master dark from each raw light frame before you do anything to it.  Any decent astronomical imaging processing package will do this for you.

Chris

[eoverstreet]

I understand the concept of  darks, flats, bias and I routinely take these when using my Nikon D800e modified, or D810a.    These images on this link  http://www.astrobin.com/users/Nikonshooter/  I took this year (I just started doing astrophotography 15 months ago and I am struggling tring to figure things out - thanks to forums and tutorials online for my education).   

When I bought the CCD camera (wish I had chosen mono now) I gave it a go and was disappointed with the camera's image quality and wrote it off as human error.  Being new to astronomy and specifically astro-imaging....I was sure it was my failure to know how to properly use it.  That may still be the case.  I hope so.  At any rate, I packed it back in the box and thought of selling it.

Recently while searching through Astrobin images taken with the Atik 383l color I was blown away by some of the outstanding imaging.

I have attached this effort which is M81, and is comprised of 50, 400 seconds subs,  25, 400 second darks (taken after taken subs, closed shutter and covered scope) 50 Bias taken at the fasted shutter speed ota covered and 50 flats, taken next morning shooting sky before sunrise covered with softbox cover.    I used both DSS to calibrate, register, and debayer......and PixInsight's Batch processing script.    The results of both DSS and Pixinsight were pitiful.  I would sell it but I am not too sure I have a good camera or a bad one....and I wouldn't want to sell a defective one.

The image turned out so awful I came to the conclusion it is not worth fooling with and perhaps I have bought a lemon.    The image below is the result of the Pixinsight batch and the only thing I did was a histogram stretch.   I didn't bother to bring up the banding.

I have tried reading from these forums as much as I could before posting my issues.

1) I use AC power (from the house - I ordered the power chord using Atik's recommendation.  It may all be my problems are stemming from electrical issues.....but I used a battery pack (Celestron) when I first got the camera and had the same RGB pixels and banding.

2) I am really unsure of how to set the temp correctly.  So I use a target about 10 degrees below ambient - here in may lie my problem.  I really need a tutorial (easy to understand one) on how to set the temp......I have no clue what terms like Delta are....easy peazy for me.

3) I use APT - Astro Photography Software to image. (I plan to use Artemis next time)

If you have any suggestions I would jump at trying them.

At age 71 I hate to waste a night of imaging trying to figure it out.   I read reviews from may 383l owners that they are getting great images right out of the box.

Not me!

[Chris]

There's something seriously wrong here, but I'm not sure what.

There are no stars and I'd expect to see quite a lot.

There are dust doughnuts, which indicate that the flat correction hasn't been done properly.

The black vertical lines look like some sort of processing artefact, as if lines of hot pixels, which these cameras have, especially if not cooled, haven't been compensated for properly.

If you think the camera is bad then contact Atik support.

A look at the raw data may reveal something, try posting a sample of a few frames of the light, dark, bias and flat data.  The FITS files as generated by the camera image acquisition software, no processing.

Or perhaps someone else can suggest something.  I only have a mono 383L+.

Chris

[PAB64]

Just a thought.

Is the colored noise "Colour Mottle"

https://www.youtube.com/watch?v=PZoCJBLAYEs

Astro imager Tony Hallas describes it, and shows you how to remove these artifacts.

[eoverstreet]

Thanks for sharing that link!!!

[eoverstreet]

I believe in my case....my problem is just flat human error.    I have been experimenting with the camera (daytime) and using both Artemis and Dawn to process the image.  Although I am not seeing any color...and cannot seem to bring any out, I believe I can figure that out.  The important thing is that the camera is NOT exhibiting any of the RGB hot pixels and the banding and artifacts are gone.    As soon as I get a clear sky....I will again put it to the test but this time I hope to minimise my errors using Artemis and Dawn.  I will let you know what happens!

[CCD-Freak]

Are you subtracting the Darks before doing the color conversion?

[eoverstreet]

Yes.....darks are being subtracted which looks like the problem to me as well.   I have all but given up on the camera.  I am sure there are those who have had success using this cam but my struggles have wasted far too many nights. 

I use a Nikon D800e modified at Spencers with a heat reduction system, Nikon D810a, and I now have  a QHY9 mono.  I have had zero problems processing images using PixInsight with either of these cameras. 

I may pull it out and give it a try in the future but for now it is in it's box. 

and before it is suggested, I wouldn't dare try to sell it as I really think I bought a bad cam.   

[Jo]

Have you been in touch with our support team about any of this? If you send them some of your original fits files on support@atik-cameras.com they'll be able to take a look and work out whether the problem is with the camera as you think, or if there's something else at play.

Jo

[CCD-Freak]

Looking at the  images you posted it looks like something is not right with the calibration process. The camera may have a cooling problem or something but I am sure Atik can figure out the issue.  Don't mark the 383L+OSC a bad camera since they are capable of great images.  This image of M20 is 5 x 15 minute subs, captured, calibrated and stacked with Astro-Art V4. 

You will get this figured out.

John
CCD-Freak
WD5IKX

[Jo]

Great filaments in that Trifid, John, and a great example of what the camera is capable of. Whether you decide to keep or sell it, eoverstreet, I'm sure our support team can get to the bottom of the issue and it'll make a great camera for someone someday!

[RQKimball]

The Atik 383L is a great camera.  I hope you can find where your processing is going wrong.  I suggest you get a free DropBox account and upload your raw files so we can look at them.

[Testgear]

I don't know your knowledge and capabilities so sorry if you have tried the following but,

Can I suggest you debayer all your images/flats etc before stacking/processing etc.
Also, check the images you are taking are 1x1 binning.
Check the Cooler is on as heat can cause a lot of noise.

Sorry in advance if you have done these.