[vsnet-chat 883] Re: Magnitudes, sequences, etc.

[aah@nofs.navy.mil (Arne A. Henden)]

Stan,
  Brian has already answered reasonably well (both he and I are observing
tonight at different observatories in Flagstaff), but I'll put my two
cents' worth of comments.
  While our CCD is about as good as they get for blue response, the
CCD QE in the U filter is ~ 1/3 that at the peak red response.  This
translates into rough exposure factors of 4:2:1:0.8:0.8 for UBVRI on
F-type stars.  In other words, you have to expose forever or else
use a large telescope or limit yourself to bright stars to do (U-B)
photometry with a CCD.  Don't do it...rarely worth the effort.  Restrict
yourself to BVRI photometry.  I won't even mention the transformation
problems and detail to filter design for a good U filter.
  As Brian said, standard stars used to calibrate your photometry are
always brighter than the stars in your sequences.  This is to decrease
the time spent measuring standards.  You usually expose the standard
stars to the same peak DN as you will your sequence stars, just with
shorter exposures.  Device linearity does not come into play until you
need the dynamic range to measure stars of different brightness on the
same CCD image.
  Note that you might be able to detect V=16 stars, but that is at
S/N = 3 or so, and you need S/N=100 to do good photometry.  In other
words, you may not go as faint as you think with such a small telescope
and shortish exposure times.  We do decent photometry at R with the
0.2m FASTT telescope down to R=15 with 3minute exposures.  I'd count
on about 0.7mag brighter limit with the typical amateur CCD and a
typical V filter...scale for your telescope and exposure time.
  The Tycho catalog is pretty good for B & V measures of every star
in the sky down to about V=10, falling apart fainter than that.  The
upcoming TASS survey should do a good job of covering the sky in
BVRI down to V=14.  It is fainter than that, or in crowded fields
where confusion will limit TASS, that you can make the biggest impact.
(Of course, until TASS or other surveys are actually underway, any
sequence is helpful!).  I've been talking with the AAVSO as to how
to handle their thousands of chart updates with modern CCD sequences,
and it is a massive effort.  In the end, the ideal is a whole-sky
survey since 1-2 percent of all stars are variable, and as you get
fainter you reach a limit where every CCD field includes something
variable.
  Miras are difficult because of their red color.  No standard-star
set, whether Cousins or Landolt or Johnson, goes that red and you
have to extrapolate.  Stay clear of them at first, and stick with
CVs or SNe or something simple like that.
  More later, after I finish sky flats!
Arne Henden

VSNET Home Page

Return to Daisaku Nogami

---

vsnet-adm@kusastro.kyoto-u.ac.jp