Re: Suspected Variables in TASS data

[aah@nofs.navy.mil]

Period98 is another good program, as Michael S. suggests,
so give it a look.  If Michael K. has trouble understanding AVE,
though, he will have the same problems with Period98.  You need a
basic understanding of the principles of period determination
  There are at least three people on this list who
prefer to be called "Michael", so if I may suggest that
you at least include their last name-initial, as in
Michael K., to differentiate.  There used to be three
professional astronomers named Arne (I worked with one
of them at OSU for 8 years, talk about coincidence),
so I have on occasion had that experience you Bob's,
Jim's etc. have all the time. :-)
  Regarding a tutorial on phase plots, ephemerides, etc.,
I know that several textbooks have sections regarding this
stuff, but I haven't found a good web-based reference.
Perhaps someone has a useful URL.
  Tom, does your latest processing include new data for
the 25 stars posted by John?
  John, you once wrote:
>I've found 3 possible eclipsing binaries with periods of 1.0 days, or
>possibly a multiple on CD 23.

>The difference between exactly a day and the value doesn't seem to
>amount to the expected roughly 4 minutes sidereal day - solar day
>difference.

>Lots have stars have been observed by the TASS data, but I'd more expect
>something like 1.1 or 0.9 days to turn up. No surprise that there should
>be unknown eclipsers around these periods because the are usually biased
>against by selection effects.
  Actually, 1.00d periods should be selected *for* in the dataset, for
two reasons:  the data was taken at roughly the same time window every night,
and any variability index is going to be maximum for variables that show
the most variation each observing session; 1.0d variables will do it
every session.
  FASTT had a spatial filter:  it was just great at finding 4arcsec doubles.
These stars would blend or split, depending on the seeing, and so the
photometry was quite variable.  I can give you a list of several hundred
such pairs.  We added an extra astrometric test to remove these false
variables.  I expect TASS to have many such outliers, so I'd prefer that
we look at the newly discovered 'variables' quite closely, with different
equipment, to understand the frequency of false detections.
Arne