Dynamics of strong swept-shock/turbulent-boundary-layer interactions

Influence of separation structure on the dynamics of shock/turbulent-boundary-layer interactions

“…These parameters are chosen to facilitate comparisons, documented in previous work, with concurrent or archival experiments, depending on availability. Specifically, impinging shock simulations [3] were performed in coordination with experiments by Webb et al [84], swept-compression-ramp interactions [2,7] with experiments by Vanstone et al [81,82], Vanstone and Clemens [80], and sharp-fin interactions [4,7] with experiments by Arora et al [11], Baldwin et al [15], Mears et al [49], Arora et al [12], Mears et al [50], Jones et al [42]. Since recent double-fin interactions are not available, these are performed based on conditions of archival experiments by Garrison et al [36], Garrison and Settles [35] and archival RANS calculations by Gaitonde and Shang [27].…”

“…The mechanism underlying this frequency band relates to mean flow gradients in the post-separation shear layers, and the development of convective, vortical, coherent fluctuations in these shear layers, which have been described by several experimental [74,75] and computational [3,7,10,13,55] investigations. Therefore, this middle frequency band actually scales with local properties of the separated shear layer (see, e.g., Helm et al [41], Adler and Gaitonde [7]), with St δ ∼ 0.1 providing a crude estimate based on global interaction properties. Extending these concepts from 2-D to 3-D interactions presents significant challenges.…”

“…Concerning the lowest frequency band of unsteadiness, while a constant global (span independent) separation length scale (L) exists for spanwise-homogeneous interactions, the same is not true for simple swept interactions, due to their quasi-conical symmetry [5]; the separation length becomes a local (not global) property of the flow. Therefore, the mechanism underlying the low-frequency band of unsteadiness in 2-D interactions, scaling with St L , must change, and it appears to be significantly muted in simple swept interactions [7]. Concerning the middle frequency band of unsteadiness, unlike spanwise-homogeneous interactions, which exhibit 2-D freeinteraction scaling [18,38], simple swept interactions exhibit a mixture of 2-D and 3-D [72] free-interaction scaling laws for the separated shear layer [5], which result in a mixture of 2-D and 3-D scaling laws for the resulting middle frequency band of unsteadiness [7].…”

“…Therefore, the mechanism underlying the low-frequency band of unsteadiness in 2-D interactions, scaling with St L , must change, and it appears to be significantly muted in simple swept interactions [7]. Concerning the middle frequency band of unsteadiness, unlike spanwise-homogeneous interactions, which exhibit 2-D freeinteraction scaling [18,38], simple swept interactions exhibit a mixture of 2-D and 3-D [72] free-interaction scaling laws for the separated shear layer [5], which result in a mixture of 2-D and 3-D scaling laws for the resulting middle frequency band of unsteadiness [7]. The high-frequency band largely retains the St δ ∼ 1 scaling of the incoming turbulent boundary layer; however, this band is also slightly altered locally due to extreme compression or expansion and separation or attachment events.…”

“…Some previous results from these campaigns have been described in a variety of forms, including experimental measurements of the swept-compression-ramp (SCR) [80][81][82] and sharp-fin (SF) [11,12,15,42,49,50] interactions, along with high-fidelity large-eddy simulations [2,[4][5][6][7] of both configurations. The double-fin flowfield fosters a better understanding of the separation structure-unsteadiness connection, since it combines elements of swept 3-D interactions with those of 2-D interactions-the symmetry plane, for example, shows the influence of both-and thus aids in the establishment of a more comprehensive perspective.…”