Date of Award

9-2005

Document Type

Thesis

Degree Name

Master of Science

Department

Department of Aeronautics and Astronautics

First Advisor

Paul I. King, PhD

Abstract

The Air Force Research Laboratory, Propulsion Directorate at Wright Patterson Air Force Base has studied the performance of turbine blade geometries utilizing a large scale, low speed, drawdown wind tunnel in an effort to better understand gas turbine blade aerodynamics. Currently, the Air Force's Unmanned Aerial Vehicle (UAV) Global Hawk has been operated primarily at flight conditions other than the design point of its Allison AE3007H turbofan engine. This off design condition decreased the Reynolds number at the low pressure turbine causing losses in efficiency and loading. Two different blades were studied to maximize performance of the Global Hawk turbine. The first was an experimental, high turning angle blade designated the Pak-B and the second was based on the two dimensional mean diameter section of the first stage blade of the low pressure turbine used in the Global Hawk (GH1R). The Pak-B blade has been the subject of past research. The primary goals of this study were to validate the wind tunnel after previous upgrades, physically modify the test section to accept the Global Hawk blades, and to characterize the GH1R blades. A Reynolds number sweep was performed from 10K to 100K by 5K increments on a linear turbine cascade of 8 first rotor test blades. Measurements of wake velocity, total pressure losses, and boundary layer velocity were made to examine the flow. These measurements resulted in an averaged integrated total pressure loss profile for the Global Hawk first rotor. The operational primary mission inlet Reynolds number for the GH1R blade was calculated to be 13,500 and compared to the loss profile. It was shown that the Global Hawk first rotor has much lower losses than that of the Pak-B blade, and shows no signs of mid-line separation.

AFIT Designator

AFIT-GAE-ENY-05-S02

DTIC Accession Number

ADA440261

Share

COinS