Numerical assessment of a two-phase Tesla turbine: Parametric analysis

•Tesla turbine performance analysis in two-phase condition via CFD and analytical model.•The phase change model is adapted to R404a based on saturation temperature.•1D-homogeneous analytical model estimation is comparable with 3D nonhomogeneous model.•The surface roughness effect is more pronounced at lower gaps between the disks.•The two-phase tesla turbine produced ~ 0.8 W and torque of 3.6 mN-m at 2000 RPM. The scenarios on the future energy systems invariably point to heat pumps as an emerging technology to reach efficiency goals alongside energy and CO2 reduction targets. Moreover, they may progressively foster the use of chillers and refrigeration units. In this study, a technology that could enhance the efficiency of systems based on inverse cycles is analysed. Particularly, the two-phase flow behaviour of a Tesla turbine is numerically investigated. The main objectives are to clarify the role of the second phase, the actual operating range, and examine the flow mixing. Two computational approaches are developed, relying on a customised home-built mathematical model and CFD analyses carried out with a commercial software. The calculations are performed for two-phase R404a fluid over significant ranges of rotational speed, plate gap size, and plate roughness. All the critical liquid–vapour interactions are determined and discussed utilising the CFD solution of the Eulerian-Eulerian approach with a frame motion technique. The other model is a homogeneous finite-difference solution, and the mass transfer is directly determined based on the phase diagram of the fluid neglecting other phase-interaction parameters. The two approaches are compared to some available experimental results from the literature, revealing an excellent agreement. The results show the average power output of 0.8 W with a delivered torque of 3.6 mN-m at a rotational speed of about 2000 RPM. The numerical analysis explains the different effects of the two-phase conditions on the turbine efficiency, paving the way to an accurate design of boundary layer turboexpanders operating under these conditions.