For a three-dimensional blunt body in high-speed flow, the laminar boundary layer on the symmetry plane is governed by equations that contain the spanwise gradient of the crossflow velocity at the boundary-layer edge, (∂w/∂z)e(\partial w/\partial z)_e, as a required outer boundary quantity. Wang, Li & Gu (2024) derived symmetry-plane boundary-layer equations for blunt bodies and closed this quantity using Newtonian impact theory, an approximation valid only under hypersonic conditions; they state that no theoretical formula exists for (∂w/∂z)e(\partial w/\partial z)_e on the symmetry plane of a general three-dimensional blunt body in subsonic or supersonic flow, that extracting it from three-dimensional CFD is numerically ill-conditioned (a 0/0 limit at the symmetry plane, sensitive to grid resolution and shock-capturing dissipation), and that their simple edge-parameter estimation neglects entropy-layer effects. The open problem is to obtain a general, regime-independent determination of (∂w/∂z)e(\partial w/\partial z)_e — a theoretical expression or an equivalently general validated closure or extraction procedure with quantified accuracy — so that the symmetry-plane boundary-layer theory becomes applicable across subsonic, supersonic and hypersonic regimes without routine dependence on full three-dimensional viscous CFD.

Background

A boundary layer is the thin viscous layer adjacent to a body surface in high-Reynolds-number flow, described by Prandtl-type boundary-layer equations rather than the full Navier–Stokes equations. For a three-dimensional body (for example a hypersonic vehicle nose at angle of attack), the flow in a plane of geometric and flow symmetry — such as the windward meridian plane — is special: the crossflow velocity ww (the velocity component perpendicular to the symmetry plane along the circumferential or spanwise coordinate zz) vanishes identically on that plane, but its gradient ∂w/∂z\partial w/\partial z does not, and this gradient enters the symmetry-plane boundary-layer equations as an effective transverse strain term. Solving these equations therefore requires the edge value (∂w/∂z)e(\partial w/\partial z)_e — the value of ∂w/∂z\partial w/\partial z at the outer edge of the boundary layer, where the viscous solution must match the inviscid outer flow — as a boundary condition. Classical studies of symmetry-plane boundary layers (Wang 1970 for a spheroid at incidence; Roux 1972 for a supersonic cone at incidence; Degani, Smith & Walker 1992 for the turbulent case) obtained this quantity by extraction from numerical inviscid solutions for specific body shapes. For axisymmetric bodies, (∂w/∂z)e(\partial w/\partial z)_e follows from the kinematics of axisymmetric flow alone, without any pressure-distribution input. Wang, Li & Gu (2024) derived symmetry-plane boundary-layer equations in a body-fitted orthogonal curvilinear coordinate system tied to the principal curvatures of the surface, showed they degenerate correctly to the standard flat-plate, cone and sphere equations, and — for hypersonic flow — derived the first theoretical expression for (∂w/∂z)e(\partial w/\partial z)_e on the symmetry plane of general three-dimensional blunt bodies, by combining the edge zz-momentum relation linking (∂w/∂z)e(\partial w/\partial z)_e to the circumferential pressure curvature ∂2pe/∂z2\partial^2 p_e/\partial z^2 with a Newtonian pressure model evaluated on an equivalent cone — a fictitious cone tangent to the body surface at the symmetry-plane point of interest, whose semi-angle is set by the local surface slope and whose geometry carries the local principal radius of surface curvature, so that the effect of surface geometry on the pressure curvature is captured. The Newtonian (impact) flow model assumes hypersonic surface pressure depends only on local surface inclination, Cp≈2sin⁡2θC_p \approx 2\sin^2\theta; it has no counterpart local law in subsonic or moderate-supersonic flow, where the pressure curvature at the symmetry plane is a nonlocal quantity determined by the surrounding flow field and body geometry. The remaining edge parameters in their examples were obtained from a normal-shock plus isentropic-expansion route that neglects the entropy layer — the layer of high-entropy gas produced by the curved bow shock ahead of a blunt nose, which the growing boundary layer progressively swallows, altering the effective edge conditions and hence wall heating. The equations were solved by the difference–differential method, a marching scheme that reduces the boundary-layer PDEs to ODEs by finite-differencing the streamwise direction, initiated from the stagnation point (the point of zero surface velocity at the nose). The authors explicitly state the resulting theory is applicable only under hypersonic (Newtonian) conditions and that they found no general method to determine (∂w/∂z)e(\partial w/\partial z)_e in subsonic or supersonic regimes; the standard reference formulation of the symmetry-plane equations is due to Hirschel, Cousteix & Kordulla (2014).

Problem Statement

Obtain a general, regime-independent determination of the crossflow velocity gradient at the boundary-layer edge on the symmetry plane of a general three-dimensional blunt body. Concretely: derive a theoretical expression — or an equivalently general, validated closure or extraction procedure with quantified accuracy — for (∂w/∂z)e(\partial w/\partial z)_e on the symmetry plane of general three-dimensional blunt bodies that is valid in subsonic and moderate-supersonic flow, where the Newtonian impact-pressure law underlying the existing hypersonic closure of Wang, Li & Gu (2024) fails, and that is well-conditioned where naive direct CFD evaluation is not (the quantity is a 0/0 limit at the symmetry plane), so that the symmetry-plane boundary-layer equations admit a self-contained solution across flow regimes without routine dependence on full three-dimensional viscous CFD; the determination sought is of quantified accuracy and validated against three-dimensional Navier–Stokes solutions (and, where available, experimental heat-flux data) over a defensible range of Mach regimes and representative blunt-body geometries. The source paper additionally records the selection of a starting station for difference–differential marching on non-blunt bodies as a separate open task that 'needs additional consideration', and notes that turbulence and high-temperature reacting-flow models can be incorporated into the symmetry-plane equations along the lines of Anderson (2006), though this is not carried out in the paper; neither item forms part of the closure target above.

References

  1. Q. Wang, J. Li, S. Gu, 'Hypersonic boundary layer theory in the symmetry plane of blunt bodies', Journal of Fluid Mechanics 997, A21 (2024). DOI: 10.1017/jfm.2024.708, https://doi.org/10.1017/jfm.2024.708
  2. S. Gu, 'Tangential-Velocity Gradient at Stagnation-Point Boundary-Layer Edge for Hypersonic Spheres', AIAA Journal 64(6), 3040–3060 (2026). DOI: 10.2514/1.J066315, https://doi.org/10.2514/1.J066315
  3. S. Gu, 'Universality of the Heat-Flux Distribution over a Hypersonic Sphere', AIAA Journal, 1–16 (2026). DOI: 10.2514/1.J066971, https://doi.org/10.2514/1.J066971
  4. K. C. Wang, 'Three-dimensional boundary layer near the plane of symmetry of a spheroid at incidence', Journal of Fluid Mechanics 43, 187–209 (1970). DOI: 10.1017/S0022112070002318, https://doi.org/10.1017/S0022112070002318
  5. B. Roux, 'Supersonic laminar boundary layer near the plane of symmetry of a cone at incidence', Journal of Fluid Mechanics 51, 1–14 (1972). DOI: 10.1017/S002211207200103x, https://doi.org/10.1017/S002211207200103x
  6. A. T. Degani, F. T. Smith, J. D. A. Walker, 'The three-dimensional turbulent boundary layer near a plane of symmetry', Journal of Fluid Mechanics 234, 329–360 (1992). DOI: 10.1017/S0022112092000818, https://doi.org/10.1017/S0022112092000818
  7. E. H. Hirschel, J. Cousteix, W. Kordulla, 'Three-Dimensional Attached Viscous Flow', Springer (2014). DOI: 10.1007/978-3-642-41378-0, https://doi.org/10.1007/978-3-642-41378-0