What could cause a discrepancy in the Bias of a CCD

[starstruck_]

This is assignment related, however it’s not really a problem solving question.

So we were to take darks from our school’s CCD. The manual for the CCD says there should be ~1000 ADUs/pixel for the bias, my graph (as well as many other people- they had similar numbers) says that the bias should be about 520 ADUs/pixel.


I need to explain this discrepancy between the bias I found and the manual.

Here are some ideas I had, and I have no clue if they’re right or not:

The bias in the manual is the maximum possible

Maybe the binning might effect it? Although I don’t see why- the binning was set to 3x3


The temperature that the plots for my graph were taken at was at -10 degrees Celsius, so when the line of best fit and it’s function was used to determine the bias, it was in a linear relationship with data points in -10 degrees temperature. Perhaps it could be that the actual bias doesn’t have a linear relationship to this plot as it would have been taken at a different temperature?

 

[mfb]

Does the manual specify a temperature for the measurements?

 

[Drakkith]

Without knowing which camera you were using and the steps you took to calculate the bias it's difficult to say where the discrepancy comes from. Did you do any processing, such as dark subtraction or apply any flats?

Perhaps it could be that the actual bias doesn’t have a linear relationship to this plot as it would have been taken at a different temperature?

My understanding was that bias was unaffected by temperature, but I'm not sure about binning.

 

[starstruck_]

Does the manual specify a temperature for the measurements?

The manual does not, it only specifies a temperature for the dark current.

 

[starstruck_]

Without knowing which camera you were using and the steps you took to calculate the bias it's difficult to say where the discrepancy comes from. Did you do any processing, such as dark subtraction or apply any flats?

We only took 0.4 s exposure pics and did not process any data- the telescope was capped, we were only taking darks. All we did was plot the average of ADUs/pixel that was given in the side panel of the software we used for every pic.

The camera is the SBIG STLX-6303E

I unfortunately did have to hand in the assignment so I’m hoping for the best, so you don’t have to answer, but if you do, thank you!

I use the same camera for the research I do and haven’t ever been put in a position where I had to think about the bias discrepancy.

 

[Drakkith]

Try taking some bias frames unbinned and seeing what the bias is. If it's about 1000, then I would guess that the binning is the cause of the problem. Other than that I don't have any tips for you.

 

[Tom.G]

This turned out to have more questions than answers, but here goes.

The general rule-of-thumb is that both chemical reactions and electronic device leakage currents double for every 10°C rise in temperature. That's why imaging chips are often cooled, as is yours. And the spec sheet for the camera lists values at 0°C, so that could account for the factor of two discrepancy.

I just checked the operating manual on-line (http://diffractionlimited.com/wp-content/uploads/2016/12/operating_manual_for_stx.pdf) and it does not have the word "bias" in it. So the above may not be applicable. Where did you find the word "bias"? Can you at least define it as used in the manual? What are the variables you are graphing?

Obviously, I'm not very familiar with telescope cameras but I hope the above is a little bit of help.

 

[andrew s 1905]

In a CCD the bias or offset is a feature of the output electronics to ensure you don't get negative values from the A to D converter. It is commonly not the exact value specified in the manual as as long as it not too high or too low it does not matter. I

It should not vary significantly with temperature unlike the dark current which is quite sensitive to temperature and is a feature of the CC'D light sensitive pixels.
Regards Andrew

 

[starstruck_]

I just checked the operating manual on-line (http://diffractionlimited.com/wp-content/uploads/2016/12/operating_manual_for_stx.pdf) and it does not have the word "bias" in it. So the above may not be applicable. Where did you find the word "bias"? Can you at least define it as used in the manual? What are the variables you are graphing?

Obviously, I'm not very familiar with telescope cameras but I hope the above is a little bit of help.

For the project we’re required to purchase Project Manuals (specific to my Uni) which include all of our assignments and important stuff pertaining to the project and one of the things it includes is the Manual for the CCD which lists its Bias. As defined in the manual, the Bias is an offset in the signal per pixel. It exists because charge exists before the CCD is turned on or being used.

 

[Drakkith]

Like I mentioned before, I would find the bias level for the CCD when it is unbinned and see if it is different from what you are getting when the camera is binned.