Numerical prediction of velocity coefficient for a radial-inflow turbine stator using R123 as working fluid

Organic Rankine cycle (ORC) is a reliable technology for converting low-grade heat into electricity. The accurate design of its expander is the key to ensure an expected cycle efficiency. This work numerically investigates the stator velocity coefficient for the radial ORC turbine using R123 as working fluid. The effects of outlet blade angle, solidity, blade height, expansion ratio, and surface roughness, on the stator velocity coefficient are evaluated by a verified 3-D viscous numerical modeling. A modified 1-D model is also derived within a wide range of various parameters. The results show that the velocity coefficient is relatively sensitive to both blade height and surface roughness while almost independent of expansion ratio, mainly due to the high Reynolds number and the small flow boundary layer when using refrigerant vapor. Since the existing semi-empirical formula fails to well capture the velocity coefficient for the stator with rough wall, a modified model considering the surface roughness successfully resolve this issue with a maximum deviation around 3.5%. Its applicability for the stators with different stator inlet conditions (380–410 K) and working fluid (R245fa) is further explored, and the results also exhibits satisfactory accuracy.