Ignoring, for the moment, all the problems and questions about the observing run on the night of November 26, did we actually get enough information to do some real science?
Summarizing what we have:
- A total of 511 images, each exposed for 30 seconds
- The final 148 images of those 511 are generally totally unusable (lights from a neighbor’s garage are visible at one point)
- The first 270 images are generally good, except for several passing clouds (around images 66, 107, and 270).
First, let’s extract brightness information for KIC9832227 from those first 270 images. Plotting those points… (Each point is a measurement from one picture.)
Observations:
- Data scatter generally gets worse as the hours pass. The scatter around 1-1/3 hours corresponds to the clouds that passed around image 107 or so. By hour 4, data scatter is too big to trust anything from that point on.
- There is a clear minimum in the graph around 2.9 hours after the observing run started. This marks the center of an eclipse of the two stars that make up KIC 9832227.
- There might be a maximum around 0.1 hours. Or maybe not.
It’s worthwhile comparing this to a typical eclipsing binary star. Here’s the graph (lightcurve) for another eclipsing binary star that I’ve been following since the start of the summer. (There are parts of the lightcurve that are missing, but they are boring parts.)
Two characteristics of a “normal” eclipsing binary star light curve are notable: there’s an extended period of time in between eclipses when the light is constant and at its brightest, and there are two kinds of eclipses (named primary and secondary) when the dimmer star passes in front of the brighter and when the brighter star passes in front of the dimmer one. But looking at the light curve that we captured on November 26, there doesn’t seem to be an extended period of time when the light is maximum and constant.
What does this mean? Simply that there isn’t an extended period of time when the two stars aren’t interfering with each other. How can that be? That happens when the two stars are touching. This is what KIC 9832227 seems to be: a contact binary, with the stars so close that they are actually touching. Here’s a computer model of KIC 9832227:
All this from a lightcurve obtained from a rather messed up observing session!
Exploring the binar star KIC 9832227 
Mark, is it possible that the lack of an extended brightness is due to the system actually being well separated and that the secondary is in fact an outlier and moving relatively slowly? In that case how much observation time would be needed to see a full cycle, and is that time available over the course of a night?
Mike
LikeLike
Mike: A full cycle takes 11 hours. Although it’s unlikely that this star will be observable for 11 continuous hours any night soon, we can piece together a graph of its brightness over a full cycle by combining “chunks” of the total cycle obtained on different nights. Enough data to do that should be available with just one or two more observing runs, and then we can begin to see how repeatable the lightcurve is. A very repeatable lightcurve is an indication that these two stars really are in close orbit to each other.
LikeLike