M.S. Thesis Presentation by Timothy James Remley
Friday, March 12, 1999

(Dr. Said Abdel-Khalik, advisor)

"Single-Phase Heat Transfer in a Trapezoidal Channel"

Abstract

The purpose of this investigation was to experimentally determine the distribution of local heat transfer coefficients around the perimeter of a non-uniformly heated, trapezoidal channel under single phase turbulent forced convection conditions; friction factor data were also obtained over a wide range of Reynolds numbers. The work was motivated by the need for such data in the design and safety analysis of the lateral light-water-cooled blanket for the Accelerator Production of Tritium (APT) project currently under development at Los Alamos National Laboratory.

The test section had a trapezoidal cross section with rounded corners, a hydraulic diameter of 0.45 inches, a heated length of 24 inches, and a total length of 54 inches. Twenty eight, individually-controlled, cartridge heaters were placed around the periphery of the test section in order to control the azimuthal heat flux profile. A large-scale thermal-hydraulic test facility was constructed. Water at controlled flow rates, and inlet temperatures was delivered to the test section. Experiments were conducted for wide ranges of Reynolds and Prandtl numbers, 1.35x104 to 8.63x104, and 2.3 to 5.0, respectively, which span the expected operating conditions for the APT blanket. For a given set of operating conditions, the local heat transfer coefficients were determined using wall temperature measurements provided by seven thermocouples positioned around the channel exit perimeter. Contrary to reported data for uniformly-heated channels, the highly non-uniform heat flux applied to the perimeter of the trapezoidal channel produced higher local heat transfer coefficients in the corners. The data were corroborated using CFD calculations performed using the FLUENT code. Average Nusselt number data were compared with predictions of widely-used circular tube correlations using the hydraulic diameter approximation, with poor agreement.