Possible Trinary Star System KID 10091110

Possible Trinary Star system with great analysis by Talk members at Planet Hunters for KID  10091110 / APH51954360:

The Talk thread is well worth your time. Here is Kian Jek's comment:

"I don't think we're disputing that it's a circumbinary, but rather that it's a trinary, the third object isn't planetary, but most likely stellar - and from the 2:1 resonance seen, we just think it's very unlikely to be a blended BG EB. The Skyview doesn't show any elongated object, so if there is a blend, the 3rd light contaminant has to be very close, within 1-2" of the target.

Anyways, I've done a TTV analysis of the 3rd, 8.53d period eclipse. What I did was to take the first observed eclipse, treat this as T0, compute future eclipse timings from this and compare with each succeeding observed eclipse extremum. i.e. a straightforward O-C analysis. Here's my preliminary results. I think the picture tells a thousand words, no?

It looks like the minimum of each eclipse is arriving successively slower and then later. And it's not a trivial difference either - they are on the order of about 14 minutes earlier and later. After every tenth eclipse or so, it changes by double this amount and then reverts to the 14 minute interval again. I'm guessing the source of this third eclipse is therefore physically bound to this system.
Unless of course, it is from a blended EB, which itself has another 3rd object that is causing the TTVs of the blended eclipse. I think a certain William of Ockham would have some objections to that hypothesis..."


Jerry Orosz, Associate Professor, Department of Astronomy San Diego State University makes this comment:
 
"I think this is a blend of two unrelated EBs. There are two distinct periods in the light curve. There is an Algol-type light curve with a period of 4.21850952 days. The primary eclipse is a little over 2% deep, and the secondary eclipse (which appears at phase 0.5) is about 0.2% deep. The second signal has a period of 8.52998265 days, and only one eclipse is evident when phased at this period. The primary eclipse is nearly the same depth as the primary eclipse in the other light curve.

We often see light curves with only one eclipse per cycle. If the orbit is eccentric, the secondary eclipse can be missed. In the case of the longer period EB, it could be eccentric. If not, then we have a binary with twice that period where the two stars are equal, and the orbit is circular. One would need radial velocity measurements to be sure. I don't think these EBs are related since the respective O-C diagrams don't show any signals."