Re: Star 07 26 04.2 +07 47 12 GSC 0764 2517 0.1734d dSct?

[jg <jg@jgws.freeisp.co.uk>]

Michael Sallman wrote:

> I was getting periods of .1717 and .2081 (and also a few others), but I
> wasn't sure which one (or both) was the correct period.
> I'm still trying to figure out CLEANEST. I read the paper, but I can't
> get the program to do what I think it is supposed to.

That's probably why I never use TS.  It's loitering somewhere on me hard
disk.  A couple of URLs were posted earlier that told of AVE.  It takes
some getting used to, but it is a good package.  MOST IMPORTANTLY, if
allows you to plot phase diagrams, or folded light curves, or whatever
the jargon of the month in vs analysis is ;)

How to do a phase plot without programs like AVE.  Sorry if you already
know this, but just in case.

Best if your data is sorted on date, or at least the earliest date is
first in the list, then do this:-

((date minus first_date) divided by test_period) - integer((date minus
first_date) divided by test_period)

and you do this for each date, remembering to keep the matching
magnitude for each date.

or in Excel, if you use it :- =((A1-$A$1)/$F$1)-INT((A1-$A$1)/$F$1)
where in this instance the test period is placed in cell F1.  You copy
that down for every cell/JD date.

Sent separate is a GIF of the data for this star as folded on the
periods you mentioned.  Mag is Y axis, phase is X axis, 0.2081 dataset
is offset by 0.2 magnitudes for illustrative purposes.

With data of short runs that is seriously gapped, as the TASS data is,
this simple sort of test can make all the difference, because lots of
"aliasing" is going to occur, "clean" or not.  If I remember rightly,
there is nothing to stop you following up an alias instead of a true
period with clean, it's all a matter of experience I'm afraid.  That's
why I often reckon it's extra complexity ain't worth the effort unless
you've got dense datasets that well sample the entire lightcurve, but
the periodicity is well hidden.
 
> I haven't had a chance to investigate the other star on the page, #64362
> (TASSJ72038.7+64055.6)

Interestingly the eclipse shown on the url you gave was better
illustrated in that plot than the plot I managed for myself earlier, but
it still looks like an RS CVn star to me... ...don't ask me why, I don't
really know ;)

Cheers

John