AN ANALYSIS OF THE LIGHT CURVES OF THE EARLY-TYPE BINARY BF AURIGAE

The UBV observations of the massive binary BF Aur were made at the Ank ara University Observatory during 1988, 1989 and 1996. Asymmetry of th e light curves, arising from unequal height of successive maxima, indi cates that the system is active. By analysing these observations in th e framework of the Roche model (including the presence of bright regio ns on the components) one obtains a semidetached configuration of the system, with the cooler secondary component filling its Roche lobe. Th e analysis of the light curves yields consistent solutions for mass ra tio q = m(2)/m(1) somewhat less than one. The influence of the mass tr ansfer on the change of the system-orbital-period is relatively small. The upward parabolic character of the O-C diagram (Zhang et al., 1993 ) indicates a mass transfer from the less massive secondary to the mon massive primary. This inturn requires the less massive secondary to f ill its Roche lobe. This is consistent with our solution. Based on the se facts we introduced the following working hypothesis. At the place where the gas stream from the secondary falls on the primary, relative ly small in size but a high temperature contrast active hot-spot (hs) region is formed. As a result of the heating effect caused by the irra diation of the hot-spot region, on the secondary's side facing the hot spot a bright-spot (bs) region is formed. The bright-spot region is l arger in size but with significantly lower temperature than the hot sp ot. This region can be treated as a 'reflection cap'. By analysing the light curves in the framework of this working hypothesis the basic pa rameters of the system and the active regions are estimated. The probl em is solved in two stages: by obtaining a synthetic light curve in th e case when the parameters of the corresponding Close Binary (CB) Roch e model (Djurasevic, 1992a) are given a priori (the direct problem) an d by determining the parameters of the given model for which the best fit between the synthetic light curve and the observations is achieved (the inverse problem) (Djurasevic, 1992b).