Numerical and experimental investigations of diffusion-induced boundary layer separation on aero-engine nacelles

Aero-engine nacelles have to fulfill design requirements at both cruise and off-design conditions. Under engine windmilling conditions the ingested streamtube massflow is relatively low. A key off-design condition is take-off, which, in conjunction with an engine windmilling scenario, results in the stagnation point of the ingested streamtube being located significantly inside the intake. The combination of high angle of attack and low engine massflow rates leads to a strong flow acceleration and subsequent diffusion of the boundary layer on the upper quadrants of the external nacelle cowl, which can terminate with subsonic separation from the leading-edge. Under this condition, Reynolds number effects can play a dominant role on the separation onset and characteristics and 3D-annular wind tunnel tests cannot always achieve Reynolds’ number equivalent to full scale. A novel quasi-2D rig configuration representative of the aerodynamics of a full-size aero-engine nacelle under windmilling end of runway conditions examined in detail the characteristics of the boundary layer on the external cowl of a nacelle prior to diffusion-induced separation. Separation of the boundary layer was independently promoted through changes to represent different engine massflow rates and freestream Mach number on the rig to determine the limits of steady Reynolds Averaged Navier Stokes (RANS) methods to discern the onset of boundary layer separation. For the conditions and geometry investigated, the combined experimental and computational results showed that there was laminar to turbulent transition of the boundary layer ahead of the subsonic diffusion. The work showed that steady RANS can predict the onset of boundary layer separation with an uncertainty of approximately 10% on notional engine massflow rate and 0.05 on freestream Mach number relative to a nominal operating freestream Mach number of 0.25. This provides guidance for the industrial design and optimization of future windmilling-tolerant nacelles for large ultra-high bypass ratio turbofan engines. • CFD validation for aero-engine nacelles under end-of-runway windmilling conditions. • Full-size engine Reynolds number representative conditions. • Laminar to turbulent transition of the nacelle boundary layer identified. • Steady RANS able to discern onset of flow separation on nacelle external cowl. • Guidance for the aerodynamic design optimisation of windmilling-tolerant nacelles.