Title page for ETD etd-111697-152512

Type of Document Dissertation
Author Cox-Stouffer, Susan K. Jr.
Author's Email Address coxstosk@antares.appl.wpafb.af.mil
URN etd-111697-152512
Title Numerical Simulation of Injection and Mixing in Supersonic Flow
Degree PhD
Department Aerospace and Ocean Engineering
Advisory Committee
Advisor Name Title
Grossman, Bernard M.
Ng, W. F.
Walters, Robert W.
Yates, Charlie L.
Schetz, Joseph A. Committee Chair
  • scramjet
  • RNG
  • CFD
  • ramp injector
Date of Defense 1997-11-20
Availability unrestricted
Numerical Simulation of Injection and Mixing in Supersonic Flow

Susan K. Cox-Stouffer


A numerical investigation of the performance of two candidate designs for injection into supersonic flow, including a comparison of two renormalized group theory (RNG) based k-epsilon turbulence models with a more conventional k-epsilon model. The chosen designs were an unswept ramp injector with four injection ports and a novel nine-hole injector array. The objectives of the investigation were to provide reliable computational solutions to the flowfields in question using both RNG and standard k-epsilon turbulence models and to compare the solutions to experiment, thereby to judge the relative performance of the turbulence models. A second objective of the investigation was to use the computed data to provide design insights for the nine-hole injector array.

This investigation made use of GASP(tm) version 2.2, a commercial computational fluid dynamics code that was augmented by the addition of one RNG-based k-epsilon turbulence model derived by Zhou, et. al. and one variant of Zhou's model, which was derived by the author. Mesh sequencing studies were performed to measure solution quality, with the fine mesh for the injector array containing roughly one million grid nodes and the fine mesh for the ramp injector containing more than six million grid nodes. Results of these studies indicated that the injector-array solution was significantly under-resolved in the farfield, though the quality was better in the vicinity of the injector itself. The ramp-injector solution, while not perfectly grid-resolved, showed much better grid convergence in both the nearfield and farfield. Accordingly, comparison with experiment was better for the ramp injector than for the injector array. For both injectors, the differences between solutions generated with RNG-based k-epsilon and standard k-epsilon turbulence models were negligibly small.

Despite inadequate grid resolution in the farfield, the computational investigation of the nine-hole injector array did yield several important design insights. Particularly, the significance to mixing and losses of the placement of the outer injectors of the second and third rows was determined.

  Filename       Size       Approximate Download Time (Hours:Minutes:Seconds) 
 28.8 Modem   56K Modem   ISDN (64 Kb)   ISDN (128 Kb)   Higher-speed Access 
  back.pdf 68.04 Kb 00:00:18 00:00:09 00:00:08 00:00:04 < 00:00:01
  c1.pdf 33.61 Kb 00:00:09 00:00:04 00:00:04 00:00:02 < 00:00:01
  c2.pdf 72.11 Kb 00:00:20 00:00:10 00:00:09 00:00:04 < 00:00:01
  c3.pdf 107.41 Kb 00:00:29 00:00:15 00:00:13 00:00:06 < 00:00:01
  c4.pdf 236.53 Kb 00:01:05 00:00:33 00:00:29 00:00:14 00:00:01
  c5.pdf 348.91 Kb 00:01:36 00:00:49 00:00:43 00:00:21 00:00:01
  c6part1.pdf 320.13 Kb 00:01:28 00:00:45 00:00:40 00:00:20 00:00:01
  c6part2.pdf 684.95 Kb 00:03:10 00:01:37 00:01:25 00:00:42 00:00:03
  c6part3.pdf 511.55 Kb 00:02:22 00:01:13 00:01:03 00:00:31 00:00:02
  c7.pdf 815.53 Kb 00:03:46 00:01:56 00:01:41 00:00:50 00:00:04
  c8.pdf 78.34 Kb 00:00:21 00:00:11 00:00:09 00:00:04 < 00:00:01
  front.pdf 44.29 Kb 00:00:12 00:00:06 00:00:05 00:00:02 < 00:00:01

Browse All Available ETDs by ( Author | Department )

dla home
etds imagebase journals news ereserve special collections
virgnia tech home contact dla university libraries

If you have questions or technical problems, please Contact DLA.