Accuracy of the component mass estimation methods in LMXB
There is a mismatch between modelled and observed distributions of optical stars masses in BH LMXB. Companion masses in BH LMXB are found in the mass range 0.1 - 1.6 $M_{\odot}$ with the peak at 0.6 $M_{\odot}$. The standard evolutionary scenarios require the donor mass distribution peaks to be at least $\sim 1\, M_{\odot}$ to eject a massive envelope of the black hole progenitor. Imperfect of the methods of optical stars masses determination may cause this difference. Two common used approximations of real Roche lobe filling star as distortion sources have been tested. On the one hand, there is the a approximation of real Roche lobe filling optical stars as sphere. We tested rotational broadening of absorption lines based on an exact calculation of CaI $λ6439.075 \, Å$ absorption profiles in the spectra of Roche lobe filling optical stars. There is overestimation of projected equatorial rotational velocity $V_{rot} \sin i$ and, accordingly, underestimation of mass ratio $q=M_x/M_v$ in the spherical star model. On the other hand, the approximation of a real Roche lobe filling star as disk with uniform local profile and linear limb darkening law is more rough. In this case overestimation of projected equatorial rotational velocity $V_{rot} \sin i$ is $ \sim 20\%$. Such overestimation of $V_{rot} \sin i$ can result in significant underestimation of mass ratio $q=M_x/M_v$ at hight value of $q=M_x/M_v$. Refined value of $q$ does not affect the mass of a black hole, but the mass of an optical star has shrunk 1.5 times. Hence, the masses of optical stars in BH LMXB must be corrected downward, despite the contradictions to the standard evolutionary scenarios.