Date of Award

Summer 2013

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Aerospace Engineering

Committee Director

Colin P. Britcher

Committee Member

Robert Ash

Committee Member

Drew Landman

Committee Member

Stargel Doane

Abstract

Wind tunnel testing continues to play an important role in vehicle aerodynamic development. Accurate results are strongly associated with whether the wind tunnel can closely simulate the on-road conditions, including the Reynolds number and all boundary conditions. Rolling road systems (or moving belts) have been a successful tool for many auto makers and racing teams to simulate the relative motion between the stationary vehicle model and the floor in the test section. The mechanism of the rolling road system is simple, but how it affects the adjacent flow field and how this flow interacts with the flow underneath the vehicle model are still topics for research. The flow analysis within this gap may become more complicated if the crosswind conditions need to be simulated. This research computationally and experimentally investigates the issue of simulation of crosswind driving conditions in a conventional wind tunnel equipped with a rolling road.

A rolling road system designed and constructed specifically for this task was installed in the ODU 3 by 4 foot low speed wind tunnel. Computational results were calculated using Fluent™. Three different model/belt configurations (belt aligned with the model, 5 degrees inboard of the model, and 5 degrees outboard) with model yaw angles up to 15 degrees were studied using CFD and experimentally. For crosswind simulation, a fixed (non-yawed) belt with the model yawed was studied in CFD as well. Mass flow rate data from CFD was collected from four different planes under the model and was compared in order to seek an optimum configuration(s). The results suggest that the configuration with model and belt aligned with each other may be the proper choice when conducting crosswind simulations. Excessive misalignment between the model and the belt should be avoided.

DOI

10.25777/vng1-yf84

ISBN

9781303512575

Share

COinS