Skip to main content

New Eclipsing Binary?

6 posts / 0 new
Last post
Ken Hose
Ken Hose's picture
New Eclipsing Binary?

How do I find out if an eclipsing binary I “discovered” is a known binary system? I have checked the AAVSO eclipsing binary ephemeris but it is not listed there.

I was imaging the transit of HAT-P-8 b and when I extracted the light curve it was really messed up. I had successfully used the comparison ensemble before with good results. I tracked down the offending comp star and was able to extract a partial eclipse. After several nights of observing I was able to capture the primary and secondary eclipse and determine the period. The star in question is GSC 02757-01212 in Pegasus. The period is 4.117 days and the V magnitude of the primary eclipse changes from about 10.9 to about 11.2.  I was able to create an ephemeris for the next couple of months. From the light curve it looks like a fairly uninteresting detached binary.

Is this a new discovery or is it known and listed somewhere I don’t know about? If it is new, I plan on getting more light curves to better characterize the system.


Sebastian Otero
Sebastian Otero's picture
A new discovery

Hi, Ken,

VSX is the place to go:

The star is not in VSX so it's a new discovery.

You can double-check in VizieR to see what else can you find.
VizieR lists data for the star found in every catalogue so it is useful to get its magnitude, position and other information if available (e.g. spectral type, some other publication).
VSX is included in VizieR as one of the catalogues listed so you can go directly there.

Nothing. This is a new variable.

You have NSVS data for this star with several eclipses recorded:

Combining this data with yours (this data is unfiltered, just shift the magnitudes of each dataset independently to your scale) will allow you to derive an accurate period, since these observations are from 1999, so you'll get a time baseline of 13 years!.
Keep in mind that NSVS dates are MJD so you need to add 2450000.5 to them to get JD and then apply the heliocentric correction.

Then consider submitting it to VSX (or publishing somewhere else). If you decided to do so, you should attach a phase plot with the star's light curve folded to your erived period and add the information needed for the VSX form (Epoch, range, eclipse duration, other identifications (from VizieR), etc.).
Here you have a guidelines to read about submitting new variables:


Ken Hose
Ken Hose's picture
Eclipsing Binary


Thank you for your response. It looks like nobody has given this binary a good look. I am kind of new to this business and I was not aware of all the resources available. I looked over the NSVS data. Even with a maximum cadence of 4 per day the survey was able to capture some eclipses. The eclipse duration (my data) is about 0.2 days and the uncertainty in the minimum (NSVS data) is probably about +/-0.1 day or so. In spite of the uncertainty, over the long baseline I was able to refine my period calculations by 0.001 day. I read the guidelines for VSX submission and I am in the process of gathering the documentation. I have a light curve and phase plot. Hopefully this will prevent some hapless amateur from using this binary as a comp star in the HAT-P-8 b field. This EB is one of the brighter stars in the field and it is tempting to use and there is no indication that it is variable.


GTN's picture
Eclipsing Binary


    Congrats on your discovery of a new eclipser, Ken!  And a pretty cool one, at that - The light curve shows virtually equal depths, so it is likely to be a double-line spectroscopic binary, too.  Combining the light curve with the radial velocity curve of truly detached eclipsing binaries is the most accurate way of determining stellar masses, temperatures, etc.

     If your eclipser is semi-detached, there is still lots of neat info you can get from the light curve, including the mass.  The light curve shows interesting "blips" on either side of egress, so I think it is more likely semi-detached or possible spotted.  Can you measure B and V light curves?  That will give a decent idea of the stars's temperatures.  Masses for hot stars - say, spectral types earlier than about A2 - are needed most, as the error for the measurement of the mass of, say, a B5 star is still about 25-30%.

    This is potentially a very interesting system, and could make a great collaborative project with some of the amateur spectroscopists who are measuring radial velocities with good accuracy.  Since the eclipse duration is only ~4 hrs, a densely covered RV curve during minima would be very interesting and relatively easy to observe.


Thom Gandet (GTN)

Ken Hose
Ken Hose's picture
Eclipsing Binary


Thanks for the encouraging words. I think it would be great to collaborate with some spectroscopists. I think we would need resolution of a few km/s or so, which should be doable. I will have to do some research to find out who could help.

I am submitting my documentation to VSX today. Rainy weather has ended the observing season here in Oregon so it will probably be spring before I can do any more photometry. The EB is in Pegasus and it is getting lower in the sky, too.

I am attaching an ephemeris in case anyone is interested in having a look at the EB.


TCB168's picture
Spectra of EB

35 deg north is not doable for me. There are plenty of northern hemisphere people that might be able to help. Ask on the ARAS forum



Log in to post comments
AAVSO 49 Bay State Rd. Cambridge, MA 02138 617-354-0484