The eclipsing binary star NN Serpentis

11.3.2002, M. Rudolf

Imagine a star which suddenly disappears from view, and reappears 9 minutes later? And this disappearance/reappearance cycle is repeated every three hours?

Such an unusual star exists indeed: the star NN Serpentis is an eclipsing binary star with the above characteristics, located in the constellation Serpens (Caput). Uneclipsed, NN Ser is a 17-magnitude star, during eclipse its brightness drops down to 23 mag, and it is then invisible except in the world's largest telescopes.

My interest in NN Ser was raised some years ago (1994), by a brochure "Die Ausstellung Astronomie im Deutschen Museum" issued by Deutsches Museum Munich. This brochure contains a picture sequence showing the disappearing star as an example of an eclipsing binary. Further investigations about NN Ser however were, at that time, fruitless and only resulted in contradictory information about its period, amplitude etc; the General Catalogue of Variable Stars (GCSV4) for example, a reference work for variable stars, indicates a brightness range from 16.6-17.2 mag.

So the wish remained to see the eclipses of NN Ser - if they actually existed. In the meanwhile I had upgraded my astro equipment with a CCD camera (ST8). With the CCD camera, a successful observation of NN Ser should be within reach.

The night of 11.3.2002 was particularly clear in Munich. In the morning hours I tried, for the first time, to observe NN Ser from our garden in Munich Trudering. I just wanted to know whether NN Ser can actually be seen with my amateur equipment (a 6" f/6 Maksutov telescope and the ST8 CCD camera) and whether it makes sense to try to record its eclipses.

NN Ser is located at right ascension=15h 53', declination=+12°55', and thus conveniently observable in spring. Finding the region of NN Ser was quite easy, starting from alpha Ser. The sky chart (from Guide 7) shows the region around NN Ser. The blue frame indicates the field of view of the ST8 CCD chip.

An overview CCD image (2x2 binning, 3 min exposure time) indeed shows a tiny star triangle composed of 17-magnitude stars, with NN Ser being the star just below the marker line. (Note also the galaxies UGC 10068 and PGC 56241 in the field).
In the VLT image to the right, NN Ser is the encircled star.

I was already satisfied with the results so far: it was definitely possible to record NN Ser with my equipment with a reasonable exposure time; I had learned to find NN Ser and meanwhile I got somehow acquainted with the aspect of the sky region around NN Ser on the screen.

I shot a series of CCD images, just as I do usually, for further processing. I did not expect to accidentally record an eclipse of NN Ser during my first observing session, since I had absolutely no idea about the minimum epoch of NN Ser, i.e. the time when an eclipse should take place.

Suddenly the unexpected happened: the star field on the screen looked somehow different, I was puzzled that I could not find the small star triangle on its usual place: I was indeed observing the disappearance of NN Ser - an eclipse has happened just after only ca. 20 minutes of observation! That was pure luck!

I eagerly awaited the next image: it also showed nothing where NN Ser was expected to be, whereas on the consecutive frames, NN Ser was present again - quite as expected after two 2-minute exposures, plus some time for downloading the images and some delay for the autoguider to resume its task.

The images below show NN Ser disappearing and reappearing, the time (UT) refer to the begin of each 2-minute exposure.
NN Ser is directly below the marker line:

The true nature of NN Ser, its light curve, and the question whether the eclipses are partial or total, were resolved in 1999 only, with observations with the Very Large Telescope (VLT) at Paranal/Chile. See the ESO press release about NN Ser which describes the NN Ser system in detail.