Star Alignment take 2

[Emanuele]

Alright, I am really not having much success with Calibrating, Alignment and Integration.

I have some data taken remotely from New Mexico. Bias, Darks, Flats and Luminance. There's really not much to do while calibrating: just select the bias, darks and flats and the files to have those applied to. Problem is that the files come out with huge gradients as if calibration failed. Am not sure if there are settings I should be playing with?

also, when i Star Align: I used the Average Sigma Clipping (have 8 lights) and used 2 for Hot Pixel Removal.

Below is a screen shot of the pattern i am getting: look at the background - the stars are aligned but the background has some strange pattern. And also look at the vertical streaks. Seems to me that the Lights were not calibrated correctly?

[georg.viehoever]

Is this with a Canon EOS camera?
Looks similar to what I sometimes see in my data.

Georg

[Emanuele]

Georg,

no, this is taken with an Apogee U42 camera.

[georg.viehoever]

Maybe you rotate the image by 90 degrees and try this script http://pixinsight.com/forum/index.php?topic=1159.msg9587#msg9587 . It helps for the problems in my Canon EOS40D

Georg

[Emanuele]

Georg,

thanks for that! I dont know how to use the scritps though. What do I do with them?

[Emanuele]

If anyone wants, I can send the master dark, Master bias and master flat, and a single raw FIT and see what is happening?

I have dropbox in case someone wants!

[georg.viehoever]

thanks for that! I dont know how to use the scritps though. What do I do with them?

1. Download the script CanonBanding6.js
2. Load it into the script editor of PI
3. Run it on your Window with Compile&Execute in the script editor

As an alternative, it is possible to install scripts into the menu system using Script/Feature Scripts... Replace the existing Canon Banding Removal script by the new one.

The will remove horizontal bands from an image by adjusting the average brightness. This works nicely for the kind of banding frequently seen in Canon EOS Cameras, I dont know if it works on the type of banding that you see. Fora more detailed discussion of the script, please read the thread http://pixinsight.com/forum/index.php?topic=1159 .

Georg

[Jack Harvey]

We used to run Apogee U 42s at CTIO and also had background issues.  Originally we we told it was IR fringing from using a back illuminated chip, but later we found that if we turned off the power to the Ace filter wheel the pattern went away.  So we felt it was noise introduced by the Ace filter wheel.  As I remember it was a hardware bug that required Ace to swap out a resistor in the CFW.  So you might try shooting some images with and without the filter wheel powered up.  Of course you may not be running a Ace filter wheel???  But worth a try anyway???

You can get rid of the pattern with the Background Neutralization tool, but will have to raise the Lower Limit slider to around 2.5 to 3 or a bit more?  Let me know if this gives you what you rate looking for in terms of back ground.

You might also be able to get rid of it using the Integration tool and in Pixel rejection use Windsorized Clipping and check both the high and low clipping boxes. If you have the rejection masks checked you will see the rejected pattern.

If your camera control program is Maxim why not try calibrating with Maxim and see if you have the same issue.  If not then somehow you have the Cal tool set wrong?

Another thought is to try to use superflats in your calibrating  http://www.astro.uni-bonn.de/~mischa/datareduction/fringing.html

Good Luck

[Emanuele]

Jack,
Thank you so much for your reply and invaluable information. I will try your suggestions and report back with the outcome!

[Emanuele]

okay, I have determined that there's something wrong with the way that PI is calibrating my files.
Maybe I have a setting wrong but I have not touched anything.
IN Maxim the files look great once calibrated, but in PI they look like this:

[Emanuele]

And this are the calibration settings:

[Emanuele]

okay, I have discovered that by unchecking "Optimize" in the Master Dark section, the vertical banding is gone. Problem is that it is not scaling my dark with this option unchecked .... am I correct?

Still the flat did not calibrate well. I don't see any settings for flat calibration, so what do I have to do? Here's a screenshot:

[pfile]

are these files from lightbuckets? i have never gotten to the bottom of it, but there's something wrong with the master flats provided by them. their calibration masters seem to work fine in maxim, and only maxim...

[Emanuele]

Uh oh. Yes pfile, those are from them! How is it possible that they work only in Maxim?

[Niall Saunders]

Uh oh. Yes pfile, those are from them! How is it possible that they work only in Maxim?

If memory serves me right, I think we decided that it was because the data was saved in 32-bit Float mode, or something similar. I don't have time to <search> the Forum for the relevant thread, but it should be easy enough to find.

Cheers,