3. B-type pulsators
From our point of view, we regard the discovery of three pulsating B-type stars as of the greatest interest. As it is shown in Table 1, there are two Cephei stars and one slowly pulsating B-type (SPB) star in the observed field. We shall describe them now in detail.
3.1. Oo 692 = BD +56 o501
The MK spectral type of Oo 692 (B0 V) was derived by C. Boehm and R. Stalio as cited by Franco et al. (1985). Shortly after the beginning of our observations, the star was identified as a possible Cephei variable, pulsating with a period of about 0.172 d (Krzesinski 1995). The analysis of the whole 1994-1997 data set confirmed the presence of the 0.172 d signal. Using all epochs of maximum light calculated from the seasonal data in all filters, we derived the following ephemeris:
In Eq. (1), denotes the number of elapsed cycles and the numbers in parentheses are the errors with the leading zeroes omitted.
As we already pointed out in Paper I, our data suffer from systematic effects resulting in considerable power at low (2 d-1) frequencies. This effect is even increased in the case of Oo 692 since the comparison star, Oo 843, is quite far from it. Therefore, we decided to decrease low frequency effects by removing linear trends from individual runs. Consequently, only long runs were used. The phase diagram of the resulting differential photomery of Oo 692 in all four filters is shown in Fig. 2, and the results of sine-curve fitting are given in Table 2.
Table 2. The results of the sine curve fitting to the data of the three pulsating B stars found in h Persei. is the number of individual observations and RSD is the residual standard deviation. Fitted parameters have the same meaning as in Eq. (1) of Paper I with = HJD 2 449 600.
Although it is possible that Oo 692 does exhibit low-frequency variations, we think that the signal we removed is mostly spurious, because of the systematic effects we mentioned, and the fact that no single low-frequency peak appears in the periodograms of a few separate subsets of our uncorrected data. The Fourier spectrum of the corrected data is shown in Fig. 3a. After prewhitening the data with the 5.824295 d-1 frequency, no other signals with amplitude exceeding the noise level (2 mmag) are seen (Fig. 3b). In the spectral window (Fig. 3c and d) both the daily and yearly aliases appear to be of similar height; many other peaks of intermediate frequencies are also seen. All these peaks, however, do not exceed 75% of the height of the main peak. This means that the ambiguity in deriving periods for Oo 692 and other short-period variables from our data is rather small.
3.2. Oo 992 = BD +56 o520
Schild (1965) gives the spectral type of Oo 992 as B1 Vn. The star is another Cephei variable belonging to h Persei. The amplitude of the only period we have found above the noise level (Fig. 4) is extremely small (see Table 2 and Fig. 5). The I-filter data for this star were not used because of the large scatter. Despite the small amplitude we are sure that the variations seen in Fig. 5 are real because the same 7.539 d-1 signal was found in a few separate subsets of our data. This time, the ephemeris for the time of maximum light was derived from the V-filter data only:
The differential magnitudes, shown in the phase diagram (Fig. 5), were determined using five comparison stars (Oo 936, 963, 977, 980, and 1004). Because they are all relatively close to Oo 992, the residual standard deviation as well as the noise level in periodogram are considerably smaller than in the case of Oo 692. Like for Oo 692, we used only long runs corrected for linear trends.
3.3. Oo 893
This star appeared to be variable with a period of about 1.2 d. No spectral type is available for Oo 893, but its magnitude and colours are consistent with a mid-B type star lying on the cluster main sequence. As can be seen in Fig. 6, the photometric amplitude in the band is almost twice as large as in the three remaining ones. This feature is typical for all SPB stars observed by Waelkens (1991). On the other hand, as is argued by Jerzykiewicz & Sterken (1993), it would be very difficult to explain such amplitude behaviour by means of variations in an ellipsoidal system. The period we found for Oo 893 is also characteristic for Eri variables. However, because of the small scatter in the light curve and the fact that our photometric H observations (Pigulski et al., in preparation) have not revealed any H emission in the star, this explanation seems to be improbable. We therefore conclude that, despite its evident monoperiodicity, Oo 893 is an SPB variable. This is the first such a star found in this cluster. As can be seen in Fig. 7, after removing the main frequency (f = 0.83778 d-1) and its harmonic, 2, there is no significant signal above the noise level (2-3 mmag).
© European Southern Observatory (ESO) 1999
Online publication: April 19, 1999