10D sensor dust visible only at large F numbers (small aperture)

Discussion in 'Digital Photography' started by fake name, Jan 21, 2004.

  1. fake name

    fake name Guest

    My collegue thought it's a good idea to blow over the sensor and some
    saliva got there. He tried to clean it but it ain't going away. Could it
    be damage to the coating of the glass covering the spatial filter/CMOS

    Anyway, the real question is why this appers on the picture only at
    small apertures (large F numbers). He says that at F/3.5 or so it's not
    visible, but it's very visible at f/22 for example.

    One reason could be that the image is sharper at F/22 compared to f/3.5,
    but could the lens (Canon 28-135 IS) be that "unsharp" at f/3.5? I doubt
    it. he says that most people complaining about dust on the 10D sensor
    (and maybe other DSLR?) agree that the dust is more visible al small
    aperture. WHY???

    fake name, Jan 21, 2004
    1. Advertisements

  2. I think it's because there are various things in front of the sensor: IR
    blocking filter and antialiasing filter. Thus the distance from the sensor
    itself of the front of the sensor assembly is quite large. At wider f stops,
    the cone of light from the lens for each pixel is only partially blocked,
    but at narrow f stops, the dust blocks the whole cone.

    Aha! Here's a way to test if Olympus is lying (as I suspect they are). Put
    some dust on the E-1 sensor, and see if it's more visible at f/22 than at
    f/2.8. If it is, they're lying about the "parallel rays" designs they claim
    to use.

    David J. Littleboy
    Tokyo, Japan
    David J. Littleboy, Jan 21, 2004
    1. Advertisements

  3. fake name

    fake name Guest

    Yes, you are right... now I remember reading about this Olympus claim a
    while back, but didn't think much about it. It is probably a gimmick,
    since if it was indeed "parallel rays" (physically impossible, any good
    physicist understands this) the the position of the sensor in repect to
    the lens would not be critical, in other words everything would be in
    focus. Another way to think of it: if you move the lens a few mm back
    and forth the image should stay the same ("parallel" rays, remember?)
    but this is exactly how the focus is achieved in most cameras...

    back to the original topic: anyone knows the distance from the first
    surface of the sensor (whatever that is, glass, coating, etc) to the
    CMOS itself? then I could run some back of the envelope calculations...
    fake name, Jan 21, 2004
  4. Hardly parallel, but their early C-series cameras had problems with 5 to
    10 pixels of radial blur at wide angles. That's a fat blur for a 2 or 3
    MP camera. Later lenses don't project the light with such steep angles.
    The C4040 has about 1/5 the radial blur of a C3030 or C2020.
    Kevin McMurtrie, Jan 21, 2004
  5. fake name

    JPS Guest

    In message
    It's not about the sharpness of the lens. It's about the width of the
    light beams at that distance from the sensor, which converge to a point
    at the sensor. The larger the aperture, the wider the cone of light
    that converges to a point on the sensor, giving softer shadows.
    JPS, Jan 22, 2004
  6. fake name

    Paolo Pizzi Guest

    Use CO2 instead of air or, even worse, your breath. It won't
    leave any residue on the sensor (actually what you are cleaning
    is the filter.)

    You can remove anything stuck to the sensor with pec pads
    (you will need to fabricate an applicator, a blunted plastic
    knife will do) and a couple of drops of methanol (or Eclipse,
    if it makes you feel better.)
    Paolo Pizzi, Jan 23, 2004
  7. fake name

    fake name Guest

    He already tried some cleaning paper and cleaning solution from B&H. I
    also helped him prepare a nice pad from KODAK lens paper and 99.999%
    methanol (the beauty of wirking in a lab, you have access to all this
    stuff). But it didn't help. I looked at it and my opinion is that he
    damaged the coating on the filter. That is not really the filter, is it?
    it? I hope there is a glass slide that protects the spatial filter,
    bayer filter and CMOS sensor.
    fake name, Jan 23, 2004
    1. Advertisements

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.