Stroboscopic snowflakes?

Discussion in 'Digital Photography' started by darryl.luscombe, Dec 2, 2005.

  1. It doesn't snow very often here in Vancouver, BC - unlike the rest of
    Canada - so when it started snowing heavily a couple of nights ago I
    decided to take some time exposure shots of the falling snow,
    illuminated by a nearby street lamp.

    The results were a little intriguing. Some of the images showed a very
    distinct banding effect, while others, the ones where the snow was
    falling less rapidly, showed no banding in the streaks on the image.
    Example:
    http://photos1.blogger.com/blogger/3267/635/320/IMG_3661_st.jpg

    The shot was taken with my Canon 300D, 70-200mm f4L lens @ f5.6 and a
    half second exposure at ISO 400.

    There are more images available on my blog here:
    http://dazza101.blogspot.com/2005/11/stroboscopic-snowflakes.html

    I'm guessing that this was caused by some sort of strobe effect from
    the sodium street lamp, but wonder if anyone has a more definitive
    explanation of what caused this?
     
    darryl.luscombe, Dec 2, 2005
    #1
    1. Advertisements

  2. darryl.luscombe

    ASAAR Guest

    Before checking your pictures I figured that if the sodium lamp
    flickered at 50hz or 60hz, a half second exposure of the flakes
    would display a punctuated trail containing up to 25 or 30 segments.
    I only counted approx. 18, but there were probably more because the
    beginning and ending of the trails were too faint and indistinct.
    But it seems to confirm that the blips in the trails were caused by
    the strobe effect of the flickering street light.
     
    ASAAR, Dec 2, 2005
    #2
    1. Advertisements

  3. Yes, that sounds like what I found when I was hunting for an
    explanation on this. But if you check out the other photos you will see
    that the effect was only seen in some images. When the flakes were
    falling slower, there was no banding visible in the streaks. Perhaps
    the flakes had to be spinning as they fell? Or perhaps the direction of
    the flakes was also important?
     
    darryl.luscombe, Dec 2, 2005
    #3
  4. darryl.luscombe

    U-Know-Who Guest

    High Pressure Sodium bulbs are "arc tube" type, and do not flicker.
     
    U-Know-Who, Dec 2, 2005
    #4
  5. From my reading, Vancouver City uses low pressure sodium lamps and
    these would use a ballast that could lead to the strobe effect seen
    here?
     
    darryl.luscombe, Dec 2, 2005
    #5
  6. darryl.luscombe

    ASAAR Guest

    Ok. But one, we don't really know if they really were HPS, and
    two, even if the light output is continuous, is there no
    sinusoidal-like component to the light intensity, even with HPS
    bulbs that would produce any noticeable flicker? Even incandescent
    bulbs have a slight flicker when powered by A.C. The photo doesn't
    appear to show alternating streaks of light and the complete absence
    of light. Whatever type of light was shown in the photo, it wasn't
    producing a steady, uniform light output. It's variation clearly
    was at the approximate rate that would be expected from common AC
    power lines, given the length of the exposure. What type of light
    source do you think was used? In a sense though, the actual type of
    light used is almost irrelevant unless you or someone else can come
    up with a more reasonable explanation for what produced the pulsed
    streaks. The use of A.C. power seems to be of more significance
    that the type of bulb that was used.
     
    ASAAR, Dec 2, 2005
    #6
  7. darryl.luscombe

    Martin Brown Guest

    They just need to be moving fast enough that their minimum brightness is
    not swamped by an adjacent bright patch. The street lamp is on AC power
    and light output is varying at 120Hz USA, 100Hz UK (twice the mains
    frequency).

    Regards,
    Martin Brown
     
    Martin Brown, Dec 2, 2005
    #7
  8. darryl.luscombe

    Matt Ion Guest

    It's all about the timing. Example: say a flake is 2mm wide, and within
    the space of one flicker falls 3mm, then you'll get a 1mm gap between
    each image. If it covers less than 2mm over that time, the exposures
    will blend smoothly together. Of course, the drop rate can easily vary
    with wind currents, so the effect would not necessarily be uniform even
    within the same picture.


    ---
    avast! Antivirus: Outbound message clean.
    Virus Database (VPS): 0548-1, 12/01/2005
    Tested on: 12/2/2005 7:54:03 AM
    avast! - copyright (c) 1988-2005 ALWIL Software.
    http://www.avast.com
     
    Matt Ion, Dec 2, 2005
    #8
  9. Bingo! This explanation accounts for the pictures on the blog. Those
    where the snowflakes were moving faster exhibited the "banding" while
    those that were moving slower showed a continuous streak.
     
    Barry L. Wallis, Dec 3, 2005
    #9
    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.
Similar Threads
Loading...