We analyze 37 months of Kepler photometry of 2M 1938+4603, a binary system with a pulsating hot subdwarf primary and an M-dwarf companion that shows strong reflection effect. We measured the eclipse timings from more than 16 000 primary and secondary eclipses and discovered a periodic variation in the timing signal that we ascribe to a third body in the system. We also discovered a significant long-term trend that may be an evolutionary effect or a hint of more bodies. Upon the assumption that the third body is orbiting in the same plane as the primary, we establish that it must be a Jupiter-mass object orbiting with a period of 416 days at a distance of 0.92 AU. This mass is the lowest among all tertiary components detected in similar systems.

Context. We investigate the validity of the claim that invokes two extreme exoplanetary system candidates around the pulsating B-type subdwarfs KIC 10001893 and KIC 5807616 from the primary Kepler field. Aims. Our goal was to find characteristics and the source of weak signals that are observed in these subdwarf light curves. Methods. To achieve this, we analyzed short-and long-cadence Kepler data of the two stars by means of a Fourier transform and compared the results to Fourier transforms of simulated light curves to which we added exoplanetary signals. The long-cadence data of KIC 10001893 were extracted from CCD images of a nearby star, KIC 10001898, using a point spread function reduction technique.Results. It appears that the amplitudes of the Fourier transform signals that were found in the low-frequency region depend on the methods that are used to extract and prepare Kepler data. We demonstrate that using a comparison star for space telescope data can significantly reduce artifacts. Our simulations also show that a weak signal of constant amplitude and frequency, added to a stellar light curve, conserves its frequency in Fourier transform amplitude spectra to within 0.03 µHz. Conclusions. Based on our simulations, we conclude that the two low-frequency Fourier transform signals found in KIC 5807616 are likely the combined frequencies of the lower amplitude pulsating modes of the star. In the case of KIC 10001893, the signal amplitudes that are visible in the light curve depend on the data set and reduction methods. The strongest signal decreases significantly in amplitude when KIC 10001898 is used as a comparison star. Finally, we recommend that the signal detection threshold is increased to 5 σ (or higher) for a Fourier transform analysis of Kepler data in low-frequency regions.

Context. In this study, we independently test the presence of an exoplanet around the binary KIC 9472174, which is composed of a red dwarf and a pulsating type B subdwarf. We also present the results of our search for Jupiter-mass objects orbiting near to the eclipsing binary KIC 7975824, which is composed of a white dwarf and type B subdwarf, and the pulsating white dwarf KIC 8626021. Aims. The goal is to test analytical techniques and prepare the ground for a larger search for possible substellar survivors on tight orbits around post-common envelope binaries and stars at the end of their evolution, that is, extended horizontal branch stars and white dwarfs. We, therefore, mainly focus on substellar bodies orbiting these stars within the range of the host’s former red-giant or asymptotic-giant phase envelopes. Due to the methods we use, the quest is restricted to single-pulsating type B subdwarf and white dwarf stars and short-period eclipsing binaries containing a white dwarf or a subdwarf component. Methods. Our methods rely on the detection of exoplanetary signals hidden in photometric time series data from the Kepler space telescope, and they are based on natural clocks within the data itself, such as stellar pulsations and eclipse times. The light curves are analyzed using Fourier transforms, time-delays, and eclipse timing variations. Results. Based on the three objects studied in this paper, we demonstrate that these methods can be used to detect giant exoplanets orbiting around pulsating white dwarf or type B subdwarf stars as well as short-period binary systems, at distances which fall within the range of the former red-giant envelope of a single star or the common envelope of a binary. Using our analysis techniques, we reject the existence of a Jupiter-mass exoplanet around the binary KIC 9472174 at the distance and orbital period previously suggested in the literature. We also found that the eclipse timing variations observed in the binary might depend on the reduction and processing of the Kepler data. The other two objects analyzed in this work do not have Jupiter mass exoplanets orbiting within 0.7–1.4 AU from them, or larger-mass objects on closer orbits (the given mass limits are minimum masses). Conclusions. Depending on the detection threshold of the time-delay method and the inclination of the exoplanet orbit toward the observer, data from the primary Kepler mission allows for the detection of bodies with a minimum of ~1 Jupiter-mass orbiting these stars at ~1 AU, while data from the K2 mission extends the detection of objects with a minimum mass of ~7 Jupiter-mass on ~0.1 AU orbits. The exoplanet mass and orbital distance limits depend on the length of the available photometric time series.

The existence of exoplanets around evolved objects is one of the most interesting subjects from the viewpoint of planetary system evolution and its fate. What happens to the exoplanets engulfed in the host star envelope during red giant branch (RGB) phase? Can planets survive this evolutionary stage of the host star? Here, we are showing that at least some of the exoplanetary candidates recently found around a couple of sdBV stars, KIC 5807616 and KIC 10001893, might not be exoplanets after all. One “exoplanetary signal” visible in the light curve FT of KIC 10001893 can be just a frequency combination of stellar pulsation modes, while others are likely artifacts. Similarly, low frequency signals found in KIC 5807616 light curve FT, are beating frequencies of stellar oscillations, rather than resulting from the exoplanetary radiation. We also analyzed frequency and amplitude changes of the signal around 0.256 c/d (∼3.9 day) visible in the light curve FT of the KIC 10449976 sdO star. Our simulations show that it is difficult to reproduce the observed signal frequency variations by the weather changes in the exoplanet atmosphere.

Abstract. We present a preliminary analysis of the 2M 1938+4603 star. It is an eclipsing binary system consisting of a primary sdB component and a secondary M dwarf. The photometric data are dominated by mutual eclipses and a very strong reflection effect. The primary has a fairly rich pulsation spectrum which can be used to study its interior. On the other hand, the pulsations affect the binary trend and vice versa what makes the analysis very difficult. Therefore, we attempted at proper modeling of the light variation due to eclipses and reflection and their removal from the data so it does not affect the Fourier analysis of stellar pulsations. We focus on mid-times of over 16 000 primary and secondary minima, which were used to verify stability of the orbital period. The O-C diagram indicates possible parabolic and sinusoidal variations, commonly explained by period changes caused by evolution and a presence of a third body, respectively.