Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Learning about CCD

  1. Jan 30, 2009 #1
    I am learning about CCD s, and I always get the term BIAS, but I can't find any clear definition of what that is, can anybody help please?

    Thanks,
    Randa
     
  2. jcsd
  3. Feb 2, 2009 #2
    Re: Bias

    Nobody answered this question, I still don't know what it means, any help would be really appreciated.
    I looked online and in manysources, there is a long description about how to fix it...etc,,, but I don't understand what it really represents.
     
  4. Feb 4, 2009 #3
    Re: Bias

    I am assuming you mean CCD 'bias frames'

    A CCD bias frame is taken with zero length exposure setting and the shutter closed. It essentially brings up camera noise. The bias frame will read noise of the CCD and computer system.

    I think most modern CCD are well compensated in the noise area, both internal and external.

    I have a lower end CCD on my scope and I don’t bother with bias frames. However, I will take dark frames.

    As opposed to bias frames a dark frame is taken a with time exposure element and shutter open, however the scope is absent of light or covered to not allow any light.

    If the CCD camera will extrapolate the noise of the bias frame the dark frame has only the thermal noise.

    Here is a link the will explain. To cover it all here would be quite a long post.
    Hope this helps.

    AAVSO link
     
  5. Feb 5, 2009 #4
    Re: Bias

    Yes, actually this is what I always read about bias frames, but what does 0 s exposure time mean? I mean, it sounds like we're not doing anything... how can we get any data in 0 s? this is really confusing ...
     
  6. Feb 5, 2009 #5
    Re: Bias

    Some cameras / computers software are equipped to set to zero exposure. If not available setting of 1/8000 or 1/10000 of a second will suffice or what ever is available on the CCD.

    In an analogious way you are right, you are not really doing anything. When you take a bias frame or dark frame there is no subject. You are not taking an exposure of anything. However, in a bias frame you will not get zero data, or at least you shouldn’t as strange as it sounds.

    If you were to take a CCD sensor, cover it to totally block all light, take an exposure you would expect a totally uniform black screen. However, this is not the case. You will have pixels in varying degrees of intensity. The brightest of these pixels being the ‘hot pixels’.

    This varying intensity is caused by ‘noise’ in the system. Both from the CCD and computer. The remaining pixel intensity being caused by thermal differentials.
    So, this pattern is not caused by light or any exposure to light.

    If you take a CCD exposure and this ‘noise ‘ is left in, it will degrade the quality. So, by taking bias frames and dark frames this extraneous noise can be extrapolated.

    Dark frames and bias frames become more useful as the ‘actual’ exposures are taken at longer exposures.

    Look at it this way,

    You have a 35mm digital SLR camera. Let's say the camera has a defect, every time you take an exposure a small wire somewhere inside shorts out and glows momentarily.
    If you cover the camera completely and set it to the fastest exposure capable then trip the shutter.
    When you view the picture some of the pixels will be alive or hot, even though the exposure was zero an no light let into the camera because of the glow from the wire.



    If you take an exposure of tree, with the defective wire present.
    If the exposure is relatively short the quality will be minimal even with the internal glow

    However, if you take the picture at night the exposure will have to be increased, thus the glow from the wire will have a much more pronounced effect.

    If you cover the camera lens allowing no light and set to zero exposure or the shortest possible, then take a frame, which will be a dark or bias frame. The frame will still have data because of the internal glow

    Take the long exposure of the tree then remove the bias and dark frames the picture will have improved.



    Now, replace the DSLR with a CCD and replace the defective wire with internal noise and thermal excitation of the pixels.

    You can see the reason for bias and dark frames and why a bias or dark frame will still have data even though the CCD is covered and set to zero or the lowest exposure possible.

    Its obvious the difference a bias and dark frame make when doing long exposures of very dim objects.

    The dark frame deals more with hot pixels, where as the bias frame deals with the overall average of the pixelated frame.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?



Similar Discussions: Learning about CCD
  1. Producing a CCD image (Replies: 2)

Loading...