This paper presents a framework for extending the height-function technique for the calculation of interface normals and curvatures to unstructured non-convex polyhedral meshes with application to the piecewise-linear interface calculation volume-of-fluid method. The methodology is developed with reference to a collocated node-based finitevolume two-phase flow solver that utilizes the median-dual mesh, requiring a set of data structures and algorithms for non-convex polyhedral operations: truncation of a polyhedron by a plane, intersection of two polyhedra, joining of two convex polyhedra, volume enforcement of a polyhedron by a plane, and volume fraction initialization by a signed-distance function. By leveraging these geometric tools, a geometric interpolation strategy for embedding structured height-function stencils in unstructured meshes is developed. The embedded height-function technique is tested on surfaces with known interface normals and curvatures, namely cylinder, sphere, and ellipsoid. Tests are performed on the median duals of a uniform cartesian mesh, a wedge mesh, and a tetrahedral mesh, and comparisons are made with conventional methods. Across the tests, the embedded height-function technique outperforms contemporary methods and its accuracy approaches the accuracy that the traditional height-function technique exemplifies on uniform cartesian meshes.

Nitric oxide (NO) planar laser-induced fluorescence (PLIF) has been use to investigate the hypersonic flow over a flat plate with and without a 2-mm (0.08-in) radius hemispherical trip. In the absence of the trip, for all angles of attack and two different Reynolds numbers, the flow was observed to be laminar and mostly steady. Boundary layer thicknesses based on the observed PLIF intensity were measured and compared with a CFD computation, showing agreement. The PLIF boundary layer thickness remained constant while the NO flowrate was varied by a factor of 3, indicating non-perturbative seeding of NO. With the hemispherical trip in place, the flow was observed to be laminar but unsteady at the shallowest angle of attack and lowest Reynolds number and appeared vigorously turbulent at the steepest angle of attack and highest Reynolds number. Laminar corkscrew-shaped vortices oriented in the streamwise direction were frequently observed to transition the flow to more turbulent structures.

No abstract

Nitric oxide planar laser-induced fluorescence (NO PLIF) imaging at repetition rates as high as 1 MHz is demonstrated in the NASA Langley 31 in. Mach 10 hypersonic wind tunnel. Approximately 200 timecorrelated image sequences of between 10 and 20 individual frames were obtained over eight days of wind tunnel testing spanning two entries in March and September of 2009. The image sequences presented were obtained from the boundary layer of a 20° flat plate model, in which transition was induced using a variety of different shaped protuberances, including a cylinder and a triangle. The high-speed image sequences captured a variety of laminar and transitional flow phenomena, ranging from mostly laminar flow, typically at a lower Reynolds number and/or in the near wall region of the model, to highly transitional flow in which the temporal evolution and progression of characteristic streak instabilities and/or corkscrew-shaped vortices could be clearly identified. RightsThis paper was published in Applied Optics and is made available as an electronic reprint with the permission of OSA. The paper can be found at the following URL on the OSA website: https://www.osapublishing.org/ ao/abstract.cfm?uri=ao-50-4-A20. Systematic or multiple reproduction or distribution to multiple locations via electronic or other means is prohibited and is subject to penalties under law. were obtained from the boundary layer of a 20°flat plate model, in which transition was induced using a variety of different shaped protuberances, including a cylinder and a triangle. The high-speed image sequences captured a variety of laminar and transitional flow phenomena, ranging from mostly laminar flow, typically at a lower Reynolds number and/or in the near wall region of the model, to highly transitional flow in which the temporal evolution and progression of characteristic streak instabilities and/or corkscrew-shaped vortices could be clearly identified.