Utilizing Transitional CFD to Analyze the Performance of a Racing Sailplane
Open Access
- Author:
- Axten, Christopher Joseph
- Graduate Program:
- Aerospace Engineering
- Degree:
- Master of Science
- Document Type:
- Master Thesis
- Date of Defense:
- December 05, 2019
- Committee Members:
- Mark David Maughmer, Thesis Advisor/Co-Advisor
Sven Schmitz, Committee Member
Amy Ruth Pritchett, Program Head/Chair - Keywords:
- computational fluid dynamics
sailplane
boundary layer transition
aircraft performance
aircraft design - Abstract:
- Traditional computational fluid dynamics solvers either model the flow as laminar or with assuming the presence of turbulence. If the flow is modeled with turbulence the initial influence of turbulence is minimal, and the flow can be considered laminar-like, but as the flow develops the amount of turbulence grows until it acts as a fully turbulent boundary layer. Neither approach properly models flow dynamics for the flight regime of a sailplane. To demonstrate the potential of using computational fluid dynamics for sailplane design a racing sailplane is analyzed with computational fluid dynamics using a recently developed transition model to accurately model viscous effects. The results of the analysis are validated against a conventional sailplane analysis program and are found to agree well. Regions with complex flows, such as the wing-fuselage juncture and the empennage juncture, are examined to highlight the potential for utilizing computational fluid dynamics to refine junctures in ways not possible with conventional design methods. Practical uses for computational fluid dynamics in sailplane analysis, such as investigating the stall characteristics and evaluating the tailwheel and pushrod fairing drags, are also discussed along with notable gains in aircraft performance. Two computational fluid dynamics transition models are compared and found to predict similar lift and drag characteristics but determine conflicting transition locations at high-speed, with the recently developed model more closely matching the predictions of the conventional analysis program.