Astron. Astrophys. 319, 515-524 (1997)

6. Conclusion

In this work, we model the angular velocity time series of OAO 1657-415 and characterize the type of accretion (wind or disk). The noise power spectrum and the structure function analysis showed that the angular velocity time series history of OAO 1657-415 is consistent with a random walk model,(with steps at order of several days: days) and that all the observed fluctuations are associated with external torques. The random walk strength rad² /sec³ is consistent with that found by Baykal & Ögelman (1993), rad² /sec³, using the data given by Nagase (1989). This noise level is a decade higher than Cen X-3 and GX 1+4 (Finger et al., 1994; Chakrabarty 1995) and times higher than Her X-1, Vela X-1, 4U 1626-67 (Boynton 1981; Deeter et al., 1989; Chakrabarty 1995). These are the systems in which the noise power density has been studied in detail (for the other sources' estimates see Baykal & Ögelman 1993). The magnitude of the angular accelerations in the spin-up/down episodes are consistent with systems spining-up/down rapidly such as SMC X-1 (secular spin-up) and GX 1+4 (secular spin-up before 1980, and secular spin-down after 1984). The high level of angular accelerations in the spin-up/down episodes suggest that the Keplerian disks are present in these episodes. We have found almost no correlation between angular acceleration and mass accretion rate. The strong correlation between angular acceleration and the specific angular momentum implies that the specific angular momentum is changing its sign randomly, as is seen in hydrodynamical calculations of wind accretion (Blondin et al., 1990). Assuming that mass enters the magnetosphere at the corotation radius ( cm), where the angular momentum is added (or subtracted) according to sign of specific angular momentum (), and using the mass accretion rate gm sec^-1 (Chakrabarty et al., 1993), and the approximate values obtained for the torque event rate days and duration of events days), we obtained the approximate value of the random walk strength as,

This is close to our random walk noise strength value, indicating the possible formation of episodic Keplerian disks. Note however that the mass accretion rate used here is rather uncertain, since it is based on a luminosity which is calculated indirectly using a value for the pulsed fraction from earlier observations, and a very uncertain distance of   10 kpc based mainly on a high interstellar absorption and a low galactic latitude (Kamata et al. 1990). Eq. 15 also shows that the level of noise strength increases with mass accretion rate.

Our findings are strongly sugesting that OAO 1657-415 has an OB type companion. In the case of a Be type of companion, the accretion disk forms from the low velocity equatorial stellar wind. In this case the orbital velocity of the system is higher than the wind velocity. Therefore the flow has enough initial specific angular momentum, and hence the formed disk should be in the same sign with orbital motion. In this case, it is unlikely to expect disk reversals (or flip flop instabilities) and one should see positive correlations between angular accelerations and mass accretion rates (Ghosh & Lamb 1979a, b).

One of the important observations to test the idea of formation of disk from the stellar wind is to obtain spectral information and examine the column density (or filaments) during the orbital phases. The distribution of ionization lines and column density during the orbital phases will give additional information about accretion process (Blondin et al., 1990).

© European Southern Observatory (ESO) 1997

Online publication: July 3, 1998
helpdesk@link.springer.de