The noise generation by a main landing gear door

Experimental measurements and numerical simulations were conducted on a simplified main landing gear model that consists of a leg-door, and a main strut in a parallel configuration. The effects of varying the leg-door angle of attack, and the gap distance between the two elements, were initially stu...

Full description

Saved in:
Bibliographic Details
Main Author Fattah, Ryu
Format Dissertation
LanguageEnglish
Published University of Southampton 2016
Online AccessGet full text

Cover

Loading…
More Information
Summary:Experimental measurements and numerical simulations were conducted on a simplified main landing gear model that consists of a leg-door, and a main strut in a parallel configuration. The effects of varying the leg-door angle of attack, and the gap distance between the two elements, were initially studied by two-dimensional and low-order numerical simulations, using the unsteady Reynolds-Averaged Navier-Stokes equations. The strut diameter was specified to the same diameter as a full-scale main landing gear, and simulated under a free-stream Mach number of 0.2, and a Reynolds number based on the cylinder diameter of 1:7 x 106. Further three-dimensional and high-order numerical simulations were conducted on models with a constant gap distance of 8.7% of the cylinder diameter. The high-order solver evaluates the three-dimensional Navier-Stokes equations in the full-conservation form, with the Zonal Detached-Eddy Simulation model. The fidelity of the numerical solver was improved in two parts. Firstly, an Eigenvalue analysis for a multiple-block environment was developed to optimise the combination of spatial and filtering schemes for maximum grid resolution that is numerically stable. Secondly, a grid quality metric, which correlates strongly to the solution accuracy, was developed. A validation database of experimental measurements on a tripped 26% scale interaction model, at a free-stream Mach number of 0.09, and a Reynolds number based on the cylinder diameter of 2 x 105, was developed at the 2:1 m x 1:5 m wind tunnel at the University of Southampton. The experimental and numerical results show that the wake generated by the interaction model is dominated by low frequencies that correspond to the vortex shedding modes of the cylinder, and the door. As the door angle is increased from 0 to 10.7 degrees, the intensity of the cylinder shedding mode decreased. The sound pressure levels of the radiated noise were calculated using the FW-H method. The dominant noise source is a compact dipole, which reduced in strength as the door angle was increased.
Bibliography:0000000459147883