Answers are grounded in this paper and cite their sources.
The new versions are intended for applications in which predictions of separation by the initial Spalart-Allmaras model (SA92) are thought to be consistently inaccurate in either direction; SA92 remains the default version. They require only a different set of constants, besides a minor modification to the [Formula: see text] function and a new simple function [Formula: see text]. They are constrained by coordinating the constants to return the same result in the flat-plate boundary layer, which is an essential component of many flows. This follows the policy of Menter in his Generalized [Formula: see text]-Omega (GEKO) model. Important in Aerodynamics is the fact that the stall of the NACA 0012 airfoil happens at significantly different angles of attack, with SA92 providing the best predictions. The versions giving, respectively, Very Early, Early, Standard, Late, and Very Late separation are denoted by VESSA, ESSA, SA92, LSSA, and VLSSA. Key airfoil results are discussed. Examples of VESSA advantage include a thick wind-turbine airfoil and the NACA 4412 airfoil. Conversely, the VLSSA model improves predictions on an airliner high-lift system, in the Fifth High-Lift Prediction Workshop. In both cases the results are favorable while leaving room for improvement, and the choice between versions adds a burden to the user.
No takes yet. Share an insight, caveat, or question.
Spalart et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: