[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: GSC 0279 0321 is probably not an eclipser (also note on colour index use)




Cheers for all that, Dirk!

I've certainly learnt a few things.

The case again is better shown with actual examples than with tied down
book definitions.

(I like eclipsers too, and am quite happy sometimes just to stand around
and eyeball Algol undergoing one, for instance.  But when it comes to
analysis  I have to admit to never really being able to get any data to
look at, as so few folk seem to measure them, so it's nice to hear a
voice of experience).

Sorry to Michael K if I confused you with it all ;)


Cheers

John


Dirk Terrell wrote:

> Your statement that because the
> color curve is flat, then you don't have an EA system is simply wrong.
> A binary with two very similar stars will have little to no color
> variation through the eclipses. There are plenty of systems like this
> such as CW Cep and AG Per.
> 
> See, here is an example of what I meant about using light curve
> morphology to conclude too much about the stars. The prototype for EB
> curves is Beta Lyrae and it has a 13d period! The two stars there are
> _nothing_ like each other. One is a normal B star and the other is
> buried under a geometrically thick accretion disk. All you can really
> say with an EB curve is that the surface brightnesses are only slightly
> different.
>
> That is not true. To say that a star is too red to be an EB system is
> like saying that a car is too green to be a Ford. Color has nothing to
> do with whether a binary can have an EB light curve. Here are three
> systems of the top of my head that are "too red to have EB curves": W
> Crv (B-V=0.71), RW PsA (B-V=075), AK Her (B-V=0.53). And there are
> plenty of others.
>
> Wrong again. AO Cas is a typical early-type system with an EW light
> curve and has P=3.52d.
> 
> But AO Cas is an O-type. Again the fallacy of relating light curve
> morphology too tightly to the intrinsic properties of the stars rears
> its head. O and B types with EB/EW light curves are relatively common.
> (In absolute numbers they are more rare, of course, compared to later
> types simply because the number of high mass stars is smaller than low
> mass stars.) Overcontact and near-contact binaries exhibiting EW and EB
> light curves exist at all spectral types.
>
> And some people find them more interesting. :-) One person's noise is
> another person's gold mine of data. EA stars fill a critical niche in
> stellar astrophysics by giving us absolute parameters like masses and
> radii for relatively undisturbed stars. They are no more or less
> important than pulsators.
> 
> Not true yet again. One night observing mid-way between the occurrence
> of eclipses with a longer wavelength filter like Ic will usually do the
> trick. Now, if it has an eccentric orbit, then more work might be
> involved, but binaries with large eccentricities in the period range
> we're likely to be looking at are pretty rare.

yep, that's why I originally got 67 days for delta Velorum's period, it
confused me.  Fortunately Sebastian Otero himself and Chris Lloyd
managed to get the proper 45 day period out of the quite sparse data.
 
> We shouldn't dismiss possibilities out of hand especially when basing
> it on a flawed idea of light curve morphology and colors. That might
> work ok for pulsators but it certainly doesn't for eclipsing systems.
> The answer is not to sit around and argue about what's impossible.
> Craft observational tests and perform them. Last night I got three
> hours on this star, following it as far over as I could. Hopefully when
> I reduce the data, they will help us decide between EB and pulsator
> since I'm a bit further west than Michael K. and should have gotten
> data farther along the curve. I'll reduce them tonight and let everyone
> know how it turned out.
> 
> Dirk