evaluation of noise and gain for QHY6Pro

Ask your questions, show your results
Martin Dubs
Posts: 141
Joined: Fri Jan 06, 2012 9:16 pm
Location: Maienfeld, Switzerland

Re: evaluation of noise and gain for QHY6Pro

Post by Martin Dubs »

Hello Dong Li,

I guess Maxim has a somewhat special display of the histogram with a logarithmic scale for the vertical axis. A Gaussian has the following shape, taken from a bias of my camera (SBIG ST8300):
bias from ST8300 in screen stretch
bias from ST8300 in screen stretch
bias histogram Maxim.png (152.21 KiB) Viewed 6164 times
The logarithmic scale changes the Gaussian shape into a characteristic sugar loaf shape.
The logarithmic scale you can see if you open process - levels and stretch the intensity values so you can see the principal intensity range:
bias in Levels diagram, stretched
bias in Levels diagram, stretched
bias levels Maxim.png (10.42 KiB) Viewed 6164 times
In this display the vertical scale is plotted, down to 1, so you see also rare warm or hot pixels. The pedestal is cut off in the previous screen stretch histogram.
The pedestal in your histogram indicates that apart from the gaussian noise you have some pixels with a wider distribution of intensities. This might indicate some additional noise from readout, I have seen this with poor quality cables, if the cable picks up some noise or ripple on the power supply. With a linear scale this hardly shows up so you can work with your CCD.
If the pedestal comes from a constant nonuniformity of the pixels it might be subtracted out during processing. For measuring readout noise you take the difference between two bias images. If you have some fixed pattern noise the diffference bias should show lower noise, if the noise is completely random, the difference noise is larger by a factor sqare root 2.
Unfortunately the documentation of Maxim does not say exactly what is displayed in the histogram, you normally don't bother about it for making nice images.
Do not use Histogram - levels, this destroys the linearity of your camera response and makes the images useless!
I hope this clarifies somewhat what you measure. It is a good idea to check your camera before using it, readout noise, dark current and linearity are essential for good quality spectra.

Regards, Martin
Firefly
Posts: 51
Joined: Wed Sep 28, 2011 8:03 am
Location: Tianjin

Re: evaluation of noise and gain for QHY6Pro

Post by Firefly »

Hello, Martin:

Thank you very much for your detailed explanation! You know, I am so bad at both computer technology and maths that sometimes I really don't know much and well about software or something about them.

You are right, now I have been making my QHY6Pro CCD a test, and I want to calibrate it to a standard & normalized level, esp. the initial setup on both GAIN and OFFSET. But now I have something strange on this, and I am to go to its factory in Beijing this week to ask the designer how to handle... And I want to change to a bigger frame CCD, at least can include 3 lines for good linear calibration, such as a CCD with Sony 285AL, mine is 429AL, a bit little and only have 2 lines. BUT I have read Jeff's PDF on alpha H calibration with 2400 LHIRES III, his CCD has 2 lines,too. :( I always worry that if I take spectrum with this, perhaps my results are not scientific enough, maybe uselessful... sad...

Anyway Martin, thank you again for your help~~

Dong .Li
Robin Leadbeater
Posts: 1926
Joined: Mon Sep 26, 2011 4:41 pm
Contact:

Re: evaluation of noise and gain for QHY6Pro

Post by Robin Leadbeater »

Firefly wrote: mine is 429AL, a bit little and only have 2 lines. BUT I have read Jeff's PDF on alpha H calibration with 2400 LHIRES III, his CCD has 2 lines,too. :( I always worry that if I take spectrum with this, perhaps my results are not scientific enough, maybe uselessful... sad...
You can certainly make good scientific measurements with your small size CCD. It is true that you can only get a linear calibration at H alpha with 2 neon lines but unless you live on a mountain top or in a very dry region, you can make a more accurate non linear calibration using telluric H2O lines from the atmosphere which can usually be seen in the spectrum. Many amateurs use them instead of the neon lamp calibration as they are very accurate, even when they have larger CCD .

Cheers
Robin
LHIRES III #29 ATIK314 ALPY 600/200 ATIK428 Star Analyser 100/200 C11 EQ6
http://www.threehillsobservatory.co.uk
Firefly
Posts: 51
Joined: Wed Sep 28, 2011 8:03 am
Location: Tianjin

Re: evaluation of noise and gain for QHY6Pro

Post by Firefly »

Dear Robin:
Thank you for your help! I'v read here----http://www.astrosurf.com/buil/isis/tuto ... ion_us.htm. Is this text the method you mentioned? I feel that this is used for different gratings setup.

Could you please tell me the normal and STANDARD non linear calibration procedure by H2O? It will be better that procedures are step by step. I can not find the correct lessons on this point, could you send me some informaion or a link where I can read and learn more?

Looking forward to your help...

lots of thanks

Dong .Li
Robin Leadbeater
Posts: 1926
Joined: Mon Sep 26, 2011 4:41 pm
Contact:

Re: evaluation of noise and gain for QHY6Pro

Post by Robin Leadbeater »

I use Visual Spec to do the calibration but I am sure you can do the same thing in ISIS. Here is a screen shot showing the H2O telluric lines I normally use. This was taken using a LHIRES III with a 2400 l/mm grating and a CCD of similar size to yours.

Cheers
Robin
H2O tellurics used for non linear calibration
H2O tellurics used for non linear calibration
LHIRES III #29 ATIK314 ALPY 600/200 ATIK428 Star Analyser 100/200 C11 EQ6
http://www.threehillsobservatory.co.uk
Firefly
Posts: 51
Joined: Wed Sep 28, 2011 8:03 am
Location: Tianjin

Re: evaluation of noise and gain for QHY6Pro

Post by Firefly »

Dear Robin:
Maybe my appendix is too large to upload here, then I'v sent a letter to your personal e-mail box, please check it.

Dong
Post Reply