Abstract
The NACA 65-series blade family remains a reference geometry for subsonic axial compressor design, and the systematic low-speed cascade measurements of NACA Report 1368 provide an open benchmark for computational validation. This study establishes and validates an incompressible Reynolds-averaged Navier-Stokes methodology for the NACA 65-(12)10 cascade using the open-source solver SU2. The blade section is generated analytically from the tabulated thickness form and the a=1.0 mean line, and a periodic single-passage mesh is built with a quadrilateral boundary layer. Turning angle and total-pressure-loss coefficient are extracted at an inlet-flow angle of 45 degrees and a solidity of 1.0 across an incidence sweep from minus 0.9 to 25.1 degrees, and the results are compared with the data of Report 1368. The Spalart-Allmaras model reproduces the measured turning angle to within one to two degrees across the design incidence range of 5 to 16 degrees. A systematic over-prediction develops at high incidence, where the measured turning angle flattens toward stall while the steady computation continues to rise; the limitation is identified with the inability of the fully turbulent model to represent the onset of boundary-layer separation. The validated methodology and its documented incidence-dependent accuracy provide a foundation for subsequent parametric cascade studies.