DUST: Does this rule out buying DSLR's?

Discussion in 'Digital Photography' started by Jim F B, Feb 16, 2006.

  1. Well, the 50 f/1.4 moves the entire lens barrel during focus so it pumps
    air in and out of the mirror chamber every time it focusses - if the
    lens was sealed then air would be pumped in and out through the unsealed
    5D body, which would probably be even worse.

    Also, unlike the L series lenses, the 50 f/1.4 doesn't have a rubber
    seal around the lens mounting flange, so any moisture getting onto that
    interface will carry dirt inside due to capillary action, which will
    turn to dust as the moisture dries out.

    Day to day atmospheric pressure changes, walking up and down or climbing
    hills or mountains, flying in aircraft etc. all result in air,
    potentially contaminated with dust, gets pumped in and out of the camera
    through the various openings - and you wouldn't want it to be perfectly
    sealed or the body and optics would distort.

    Believe me, it is a real problem designing lightweight cameras for
    unpressurised flight whilst maintaining dust and moisture integrity.
    The normal solution is a desiccated filtered breathing port, but that
    requires regular maintenance and filter/dessicant replacement.
    Kennedy McEwen, Feb 18, 2006
    1. Advertisements

  2. Two whole pages in the Canon 5D manual - p42 & 43. This is 33% more
    than there is on flash operation and 450% more than there is on Canon's
    ETTL-II flash!
    In fairness to Canon, that is only the last line after "If you cannot
    remove all of the dust..." ;-)
    Kennedy McEwen, Feb 18, 2006
    1. Advertisements

  3. Your time management is way off, but no more so than most in this

    So, you clean your dSLR once every 4 months, in an operation taking 5
    minutes. But the dust builds up almost every time you use the camera
    and/or change the lens. So, unless you don't use the camera between
    cleaning operations there is a significant amount of retouching time for
    all of those images which have dust present that you have failed to
    account for.

    By contrast, the Olympus approach means no retouching because every time
    the camera is switched on or the lens is changed the sensor is cleaned.
    No manual cleaning time, no manual retouching time.

    Olympus marketing may well be over-hyping the advantage, but you are
    also guilty of undervaluing it.

    IME, it is simply impossible to eliminate dust with cleaning in a
    typical domestic environment, and requires laminar airflow. Some people
    may be a little less critical of it than others, but cleaning in a
    domestic environment will usually introduce more dust than it removes.

    It is just a pity that Olympus have themselves locked into an
    intrinsically inferior format, or they could be a serious dSLR
    contender. They made a decision long before the digital era to move out
    of the camera premier league and concentrate on the mass consumer
    market, so the inferior format isn't an issue in their business model,
    because they see their competition as being cameras with even more
    inferior formats. Olympus have, in their time, come up with market
    changing features that all other manufacturers eventually adopt (eg. TTL
    flash metering) and this could be another example.
    Kennedy McEwen, Feb 18, 2006
  4. Well that is a non-sequitur since the sensors on *all* digital cameras
    use sealed sensors, yet most dSLRs still get dust problems. The sensors
    are all in hermetic packages with at least a cover glass window and
    usually an AA filter as well. Not one single camera on the market
    permits access to the sensor itself, since its surface is much more
    fragile than you could imagine. Those colour filters in the Bayer
    array, for example, are extremely soft and would be wiped off by most of
    the recommended cleaning techniques and dissolved by many of the
    cleaning solvents used! I couldn't imagine running a Pecpad across a
    dSLR *sensor* without damaging it.

    A typical sensor package is quite clear in this internal shot of the
    Canon D30 - the sensor is inside a sealed front windowed ceramic
    You can also see the AA filter in the camera body, which sits just in
    front of the sealed sensor package when the assembly is closed. This is
    what you actually get access to when the mirror is raised and shutter
    opened for cleaning.

    What Sony have probably done is something similar to the CCD package of
    older video cameras - the window of the sealed package was a
    considerable distance in front of the focal plane, not the 1-2mm of most
    dSLR packages. On the Panasonic and Sony video cameras I have
    disassembled the CCD package has a glass window about 1-1.5cm in front
    of the focal plane. It is some time since I looked inside a typical
    video camera, but the size reduction since then suggests that this
    approach has generally been abandoned.
    Kennedy McEwen, Feb 19, 2006
  5. Skip, I have to question the value of those lens flange seals on the L
    series - the lens flange is only one of many ways for air and dust to
    get into the camera. Only one of the L lenses I have uses internal
    focus and zoom yet none of the movements have any appreciable
    resistance, indicating that there are no pressure seals in the internal
    lens mechanisms.

    The 100-400 f/4.5 IS USM, for example, more than doubles its internal
    air volume between extremes of its range of focus and zoom. Canon
    should advertise its other use - emergency water pump in a leaking ship.

    But many other L lenses are just the same - at least the 16-35mm f/2.8
    is an internal zoom & focus design, so it probably justifies the flange
    Kennedy McEwen, Feb 19, 2006
  6. SNIP
    What does seem to help though is sensor cleaning in a (presumably)
    relatively low dust environment like a bathroom after moisturizing the
    air by running a warm shower for a while.

    While I'm not necessarily advocating the buck-naked approach, it could
    at least help with reducing clothing generated dust ;-) However, skin
    flakes are present everywhere, especially in the bathroom ;-(.
    On a more serious note, using a plant water sprayer (set to fine mist)
    will help in reducing static and air borne dust particles.

    Eventually there will be a point where additional cleaning attempts
    will only displace dust, rather than remove more. That's when I stop
    obsessing with periodic cleaning, Photoshop CS2's Spot healing brush
    will take care of the residual issues quite well.

    Bart van der Wolf, Feb 19, 2006
  7. Agreed - I am sure that the shower cubicle is the most dust free area in
    my home. However, I would be more than a little concerned about running
    a warm shower first - humidity is just as likely to stick dust to the
    sensor as anything. And if you happen to live in a hard water area, as
    I do, then that humidity is likely to contain limescale which will do a
    lot more harm to the sensor while trying to remove it than mere dust
    Shower caps and the like are mandatory apparel in even the lowest grade
    clean rooms - for good reason! ;-)
    True, but only a small fraction of dSLR owners have access to PS - and
    even fewer want to use it to edit every single frame they take.

    Remember how much time you used to spend spotting film scans before ICE
    came along?

    Now that gives me an idea... a white LED in the camera to illuminate the
    sensor after the shutter closes and indicate the presence of dust,
    perhaps a white back to the shutter blind and a couple of LEDs in each
    corner illuminating it. Then any dust present on the AA filter can
    automatically be cloned out in the camera. Using the EXIF data it would
    be relatively trivial to determine the radius of the clone tool
    required, based on the aperture and focal length used to take the
    original shot.

    Public domain release - if Canon want to patent this, its too late!
    Kennedy McEwen, Feb 19, 2006
  8. Jim F B

    Skip M Guest

    Except it doesn't on a consistent basis. Fast Zuiko zooms are larger and
    heavier than their full frame equivalents in the Canon line. I've detailed
    a couple, but, of course, you've ignored that to continue to claim something
    that isn't consistently true. And, then, too, many of the Nikkor lenses
    were designed with full frame in mind, and they are smaller and lighter than
    the Zuikos, too.
    That's not true. The reason the Oly avoids light fall off on the corners is
    that the lenses are as large, or larger, in diameter than the Canon full
    frame equivalents, using only a crop of the image circle of the lens. It's
    simple. And the 5D hardly has "poor" results with wide angle lenses.
    Skip M, Feb 19, 2006
  9. Kennedy,
    I agree with the other stuff you said. But there is an easier
    way right now than the above approach. The problem with the
    above approach is that dust is on the filters above the sensor
    so is not in focus. The apparent size and intensity of a dust
    spec on the image is dependent on aperture. So one must
    correct the image data based on the the lens and aperture
    in use for the image. That is easily done as a flat field.
    (I'm sure you know about flat fields, so this is for others.)
    Simply take a white sheet of paper and hold it so it appears
    uniformly illuminated (e.g. by the sun), and point the camera
    at the paper, still focused as you did the subject image,
    put the camera up close so the paper is out of focus,
    and take a few frames. Later, neutralize the color, average
    several frames, and ratio real images by the flat field image.
    This corrects for light fall-off as well as dust spots and any
    other irregularities in the transmission of the system.
    Programs like ImagesPlus are designed to do this easily.

    I have had very few problems with my DSLRs (Canon 1D Mark II, 10D, and D60),
    and simply blow off the dust. I used canned air (mostly depleted
    can). I first blow out dust from the camera with the shutter
    closed. Again, this is a light stream so as not to damage the
    shutter. I hold the can stationary, start the flow and move the
    camera held upside down into the flow. I then put the camera
    in clean mode and do the air again. It has always worked
    well enough for me, except once on a one dust spec.

    On another subject (previous iso 20,000 thread) I have obtained some
    great low light data down to 0.00016 lux. I'm analyzing the
    data and will post results in a few days. I had to do
    a better characterization of my camera (1D Mark II) and
    just completed that. See my draft at:
    Procedures for Evaluating Digital Camera Noise and Full Well
    Capacities; Canon 1D Mark II Analysis

    Roger N. Clark (change username to rnclark), Feb 19, 2006
  10. Jim F B

    Skip M Guest

    Ironically, the 100-400 is the only "L" I have that doesn't have a gasket on
    the mount. Odd, eh? But the main vent for the air seem to be around the
    zoom ring, there is a little perceptible air movement from the rear
    element/mount area, but quite a bit from the sides and front element.
    (Unscientific test: aim rear of lens at pile of cigarette ashes and see if
    any move...) The piston like movement of the rear element may account for
    some of that movement, too. Not all of the air could be coming from inside
    the lens.
    The 70-200 IS is an internal focus/zoom lens, too. The 24-70 moves the
    front element to zoom wider (counterintuitive, but effective use of the lens
    hood.) Both of them have gaskets on the mount, too, as does my old 20-35
    f2.8L, which doesn't extend when either focusing or zooming.
    Skip M, Feb 19, 2006
  11. Jim F B

    Jim F B Guest

    You say that most dSLRs still get dust problems. So, by inference, the
    conclusion in the review of the permanently mounted lense camera, the Sony
    DSC-R1, that it has no dust problems, would be a fair one? I guess that
    extremely few (if any) digital cameras that have permanently mounted lenses
    would get dust on their sensors?


    What do you think of the Olympus solution to reducing / eliminating the dust
    problem in their DSLRs? I know of a few people who won't buy DSLRs (other
    than an Olympus) until the dust on the sensor problem has been properly
    addressed by the DSLR manufacturers.

    Regards, Jim
    Jim F B, Feb 19, 2006
  12. You can, but it is not a fair comparison.

    Here is revised text from my f/ratio myth thread
    from a couple of weeks ago, based on discussion in that
    thread including corrections and some clarifications based on
    the discussion.


    There are often questions about smaller versus larger pixels
    and the corresponding camera size. Many of us would like smaller
    cameras that did just as good a job as larger ones. Is that
    possible? No, at least in terms of signal-to-noise that can
    be recorded. Here is why.

    There is a common idea in photography that exposure doesn't
    change between different size cameras when at the same f/ratio.
    For example, the sunny f/16 rule says a good exposure for a daylight
    scene is 1/ISO at f/16. Thus for ISO 100 film, you use a 1/100
    second exposure on an 8x10 camera at f/16, a 4x5 camera at f/16,
    a 35mm camera at f/16, an APS-C digital camera at f/16, down to the
    smallest point and shoot camera at f/16 (assuming the small camera
    goes to f/16). This leads people to think cameras scale
    easily and still give the same image. But there is a fallacy
    in this idea, and that is the spatial resolution on the subject.
    The smaller camera, even at the same f/ratio, has a smaller lens
    which collects a smaller number of photons per unit time.
    The smaller camera gets the same exposure time because the UNIT
    AREA in the focal plane represents a larger angular size on the

    The rate of arrival of photons in the focal plane of a lens
    per unit area per unit time is proportional to the square of
    the f-ratio. Corollary: if you keep f/ratio constant, and change
    focal length then the photons per unit area in the focal plane is
    constant but spatial resolution changes.

    So how does this apply to making smaller cameras?

    The problem is that if you scale a camera down, say 2x, the
    aperture drops by 2x, the focal length drops by 2x (to give the
    same field of view), the sensor size drops by 2x, and the pixel
    size drops by 2x (to give the same spatial resolution
    on the subject). The aperture has collected only
    1/4 the number of photons. If we kept the same sensor, then
    each pixel would collect the same number of photons because each
    pixel now sees a larger angular area. But we want the same
    resolution, so the pixels are 2 times smaller. The smaller
    pixels each collect 1/4 less photons since their area is
    divided by 4 to keep spatial resolution constant.

    Another way to look at the problem is aperture collects light, the
    focal length spreads out the light, and the pixels are buckets that
    collect the light in the focal plane. BUT THE TOTAL NUMBER OF PHOTONS
    (ignoring transmission losses of the optics). Thus, photons
    delivered to a pixel for a given resolution on the subject goes
    as the square of the aperture (and camera size)! Decreasing your
    camera by 2x means 4x less photons per pixel if you want to
    maintain field of view and megapixel count!

    This is just what we observe with small cameras: their
    smaller sensors have smaller full well capacities, that get filled
    for a given exposure time with a smaller number of photons.
    That in turn means higher noise because there are fewer

    A good example is the Canon 20D with 6.4 micron pixels and a
    maximum signal at ISO 100 of 50,000 electrons, compared to
    the Canon S60 with 2.8 micron pixels with a maximum signal
    of about 11,000 electrons at ISO 100. The pixel size is
    (6.42 *6.42) / (2.82 * 2.82) = 5.2x scaling, similar to
    the 50000/11000 = 4.5 scaling of maximum recorded signal.

    Then, for photon noise limited systems, signal-to-noise ratio
    achievable in an image is the square root of the number of
    photons collected, so signal-to-noise ratio scales linearly
    with the camera pixel size.

    If you tried to make a smaller camera that collects the same
    number of photons as a larger camera, you must keep the aperture
    constant. Given a camera, for example, with a 50 mm f/1.4 lens
    then to shrink the camera 2x, you would need a 25 mm f/0.7
    lens that had double the resolution if you wanted to keep
    the same detail in the image. That means the smaller camera
    would not be much smaller, and might be more expensive due
    to the lens specifications.

    Check out this web page for more info on this subject:

    Roger Clark
    Photos, other digital info at: http://www.clarkvision.com
    Roger N. Clark (change username to rnclark), Feb 19, 2006
  13. Not necessarily - all cats are four legged mammals, but all four legged
    mammals are not cats. Just because most dSLRs get dust problems doesn't
    mean that most fixed lens cameras don't. They are certainly less likely
    to get it but, as others have pointed out in the thread, its much harder
    to get rid of when they do.
    Well there are several ways around the dust problem, and Olympus
    certainly appear to have one of them. Another is a totally sealed
    mirror/shutter box that Sigma(?) use. That is a variant of the older
    video-ccd design where the sensor had a window at least 1cm in front of
    the sensor - easy to implement in a fixed lens camera, less so in an
    Kennedy McEwen, Feb 19, 2006
  14. Compare zooming on smooth setting with the rear cap on and off - the
    lens only vents around the zoom ring when the rear cap is fitted and it
    cannot vent through the rear. Consequently it blows air through the
    camera body during zooming - and that isn't filtered.
    I don't think I'll be getting the inside of my lens close to airborne
    ash, but you can feel the draught in the palm of your hand when the zoom
    is moved.
    That is the point - that rear mechanism is pumping air into the camera
    body, some of which has come in through the lens (and some of which may
    have come in through the camera body, depending on the model).
    Where else can it be coming from? It is inside the lens at 400mm (how
    it got there from the outside environment doesn't matter) and it is
    pushed into the camera body as you go to 100mm. If that air has dust in
    it then there is a high probability that it will still be airborne when
    the shutter opens, and hence get onto the sensor.
    The 24-105 f/4 has both gasket and pumping action with focus and zoom.
    Kennedy McEwen, Feb 19, 2006
  15. Yes, that is what I meant with using the EXIF data - so the camera knows
    what aperture the shot was taken with. Of course, the close proximity
    of an illuminated white shutter blind will mean that the image of dust
    on the sensor is less contrasty and sharp than the effect produced with
    say, f/22, however that should be relatively straight forward to filter
    and threshold. All that is necessary to achieve is a correction mask,
    with the position of the defect.
    Yes, but a flat field requires the user to implement it later, when the
    dust may be in a different position or be worse. Added to which, there
    isn't a simple ICE-like filter available for removing the effect of dust
    based on the flat field data. You can correct for vignetting and the
    like with layers, but it will just replace the dust specs with noise, as
    opposed to cloning them from low noise adjacent pixels.
    I haven't had to clean my D5 yet but, given the other thread about dust
    being blown under the SI screen when you do, I am dreading the time when
    I need to and will probably go for a brush option rather than air for
    that reason.
    I'll read that later - though I thought you already had data at that
    light level, albeit with longer exposures than would be required for ISO
    Kennedy McEwen, Feb 19, 2006
  16. Jim F B

    Skip M Guest

    Since the rear element is in a cylinder, it could just be moving the air
    that is in that cylinder, but I've no way of showing/proving that. Without
    my hand on the mount (I felt that gave a better seal than a lens cap, since
    there's no gasket on this lens) I could feel it venting around the front
    element. With my hand there, I could feel it venting around the zoom ring
    and front element, even more. But, I'm sure that some air does vent into
    the mount area, and thus into the camera. I wasn't really trying to defend
    the lens' rear sealing, although it may have seemed so, just pointing out
    that it may not be as bad as we all guess.
    The 24-70 vents from the rear mount area on zoom (but not focus,) so I guess
    the gasket merely assures that any dust captured by the lens is transferred
    directly to the camera... :-/
    Skip M, Feb 19, 2006
  17. Jim F B

    Mxsmanic Guest

    No--the problem is masked, but not gone.
    You have to do that with any brand if you want the pixel fixed.

    Of course, if sensors were interchangeable in the way film is, that
    wouldn't be a problem, but I don't expect vendors to come up with
    something as economical as standardized and interchangeable sensors
    any time soon.
    Mxsmanic, Feb 19, 2006
    Roger N. Clark (change username to rnclark), Feb 19, 2006
  19. Jim F B

    Larry Lynch Guest

    I dont know of a brand that "FIXES" a pixel..they all re-map unless the
    problem is huge, and the camera is in warantee

    L Lynch
    Mystic CT
    Larry Lynch, Feb 19, 2006
  20. Cloning dust doesn't need to be that accurate - the size of the dust
    spec can't change by more than a pixel or so with different iris shapes
    of the same aperture.
    But that is much less convenient for the average user.
    That is true, but have you ever looked at the IR output of an ICE
    equipped scanner? It is just the same on film, but with appropriate
    high pass spatial filtering the position of the dust can be readily
    What problem? Cleaning or dust getting under the SI screen? The latter
    appears to be a unique problem that Canon have introduced on the 5D,
    according to the reports from those who have suffered it.

    Almost on cue, today I noticed an extremely sharp spot in the viewfinder
    of my 5D, which I suspect may be one of these dust spots on the SI
    screen - much blacker and sharper than anything I have ever seen on a
    focus screen. As it is it isn't too much of a distraction, so I will
    avoid trying to clean it for the moment - it might disappear as
    mysteriously as it arrived. :-(
    Kennedy McEwen, Feb 19, 2006
    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.