Abstract Non-equilibrium losses in flash evaporation account for the superheating of the brine at the flash-chamber exit in relation to the state of the vapor phase. For a proper MSF design, these non-equilibrium losses must be known. A number of correlations are proposed in the literature differing with respect to the considered (implemented) physical parameters and, unfortunately, leading to different results for a given orifice configuration. It is shown, that six of nine correlations from literature give physically wrong results. Based on a discussion of the relevant parameters and the analysis of experimental data reported in literature an improved semi-empirical correlation is derived for 2 orifice configurations - the simple “rectangular” orifice and the combination orifice/weir.