Loading ...
Sorry, an error occurred while loading the content.

Re: nice stary panorama

Expand Messages
  • JD Smith
    ... This is nothing like a normal panographers workflow, since their goal is to estimate automatically night sky brightness. They are using an Apogee and SBIG
    Message 1 of 4 , May 10 4:20 PM
      On Thu, 10 May 2007 13:53:19 +0000, hmmsomethingelse wrote:

      > If
      > anybodies library gives them access to:
      > "Title Measuring Night-Sky Brightness with a Wide-Field CCD Camera
      > Author(s) Dan M. Duriscoe, Christian B. Luginbuhl, and Chadwick A. Moore

      This is nothing like a normal panographers workflow, since their goal
      is to estimate automatically night sky brightness.

      They are using an Apogee and SBIG CCD devices on a robotic telescope
      mount, with Nikon 35mm & 50mm lenses, and a single V-band filter.
      These are low read noise, megapixel astronomical CCDs. They operate
      the setup at f/2, with reasonable pixel scales (~100-200
      arcsec/pixel). Since the lens outresolved the detectors, they
      defocussed slightly to avoid photometric issues. Exposures were
      10-12s. An all-sky exposure sequence requires 45 images for the large
      format CCDs, with a number of bias (zero exposure) frames
      interspersed. It takes 22min to complete, and is entirely automated,
      including software post-processing. Frame positioning was based on an
      astrometric solution of the stars (the ultimate control points!). The
      images were then mosaicked using custom software.

      If you hope to create something like this yourself with your normal
      equipment, the first issue is sensitivity. These are professional
      grade CCD's with big 24 micron pixels (vs 6-10 micron in DSLR's).
      They have high fill-factor, high quantum efficiency, very low
      read-noise, and are generally going to clean up on DSLR sensors. The
      main and unavoidable issue is light gathering power: 10-20 times that
      of the "puny" DSLRs. You can't win here by cranking up ISO either,
      since that adds noise near the "floor" where detection is an issue.
      You could bin, say 3x3 pixels together, but you've paid the read noise
      penalty for each of them, so you won't gain as much back.

      The second (related) issue you'd have to contend with is the moving
      sky. These frames were obtained by tracking with the Earth's rotation
      via the telescope controls. If you can get by with 10s exposures (at
      low f/#), you could almost neglect this (in 10s, a star moves at most
      150 arcseconds, about 3 times the pixel scale at 35mm), but you
      probably need minute long exposures. The relative motion of
      foreground and background will also not yield so easily. One option
      is to take your row of images with sky and ground at once, then move
      onto the sky. Using stars as control points, and anchoring on one of
      your "star + sky" frames, you can achieve reasonable alignment in

      "Star trail" images, with multi-minute or multiple averaged exposures,
      are much easier. Composites, where you track for the sky, but then
      composite in the foreground separately, are easier as well. Here's a
      stunning example, another APOD:


      Click through to read how he made it (30s exposure at ISO 3200, Canon
      5D). I can recommend viewing his entire gallery for inspiration.

    Your message has been successfully submitted and would be delivered to recipients shortly.