[PLUG-TALK] CCD cameras (was "Art institute...")

[Keith Lofstrom]

On Sun, Jun 26, 2005 at 08:54:09PM -0700, Russell Senior wrote:
> [1] Is it possible to do interesting long-exposure stuff with digital
>     cameras?  Or do the stupid embedded computers and their
>     programmers get in your way?

No.  Digital cameras contain largish chips of silicon called "Charged
Coupled Devices".  These CCDs have areas of "depleted" semiconductor
in them, where an arriving photon may get absorbed.  The result of the
absorption is the generation of an electron and a hole,  one of which
gets driven by the electric field in the depletion region to add to
the charge stored in the CCD "channel" Theoretically, you could just
sit there and accumulate charge for many minutes, and the CCDs in
telescopes do just that.

Unfortunately, Herr Doktor Boltzmann rears his bearded head.  
Hole-electron pairs are generated thermally as well, so gradually the
CCD channel accumulates random hole-electron pairs.  Defects in the
silicon also generate pairs.  These cause something called "dark current",
which gradually paints the picture a noisy gray.  Telescope CCDs avoid
this (partly) by being operated in liquid nitrogen, and by cherry-picking
CCDs with few defects, then characterizing the heck out of those defects.

You can see dark current on your camera - just take the longest time
exposure it permits, in a dark area.  Everything will look grainy.
So the exposure limit for digital cameras is usually counted in seconds.
There are probably very good cameras that do "many seconds", but few
to none that can do astronomical grade long-time exposures.

You can try taking pictures while immersed in liquid nitrogen, but it
is hard to operate the shutter release when you are frozen solid.

Keith

-- 
Keith Lofstrom          keithl at keithl.com         Voice (503)-520-1993
KLIC --- Keith Lofstrom Integrated Circuits --- "Your Ideas in Silicon"
Design Contracting in Bipolar and CMOS - Analog, Digital, and Scan ICs