Project Director

Sreenivas, Kidambi

Department Examiner

Arabshahi, Abdollah


Dept. of Mechanical Engineering


University of Tennessee at Chattanooga

Place of Publication

Chattanooga (Tenn.)


The research was based on the 4th Propulsion Aerodynamics Workshop which was organized by the American Institute of Aeronautics and Astronautics (AIAA) Air Breathing Propulsion System Integration Technical Committee and its objective was to assess the accuracy and numerical prediction capabilities of Computational Fluid Dynamics (CFD). It focused on the work of Behrouzi and McGuirk where they tested two different nozzle configurations under set boundary conditions. The nozzle configurations included a high aspect ratio rectangular nozzle with a clean exit and the same nozzle with an aft-deck attached to the exit. Three different unstructured grids were generated for the clean nozzle using Pointwise that varied in refinement and number of points, but only one unstructured grid was generated for the aft-deck. The solver that was used was Tenasi and Fieldview was used for post-processing. The first assessment involved testing if the solutions for the clean nozzle were grid independent by comparing Total Pressure values along the axial direction. It was found that the values for the three solutions were between 0.995% and 13.03% from each other. Because of this, the values for the medium definition grid (Grid 2) were compared against the experimental values. This showed that along the centerline, the values for Grid 2 were as accurate as 0.06% from the experimental data up to an x/Dh of 4.5, but beyond that, it went as far as 24.28% from the experimental data. At different cross-sectional planes along the axial direction, the values were between 0.06% and 27%. This meant that the data from the CFD analysis from Grid 2 was good enough to use, especially in the viscous layer at the exit. For the aft-deck configuration, the CFD values matched the experimental data along the centerline up to around x/Dh of 5 but deviates from the experimental data after this point by up to 26.15%. The final test was to compare the static pressure along the surface of the aft-deck. The comparison showed that the surface domain for the aft-deck was viscous enough that the solution converged into the experimental solution.


Throughout the time I have spent completing my research, I have received an enormous amount of support and guidance. First and foremost, I would like to thank my family, especially my daughter Ariana and her mother Brenda, for always standing behind me as I completed my research, as well as giving me motivation and reason to work hard daily. I would like to thank my thesis director, Dr. Kidambi Sreenivas, for guiding me and giving me the resources that allowed me to finish my research and complete my thesis. This also includes everyone at the UTC Sim Center, such as Dr. Abdollah Arabshahi, Dr. Ethan Hereth, and Kim Sapp, who were there every step of the way providing assistance anytime that was needed. Finally, I would like to thank the faculty of the UTC Honors College for having the confidence in me and allowing me to be part of the Honors College and take part in completing a Departmental Honors Thesis. Also, thank you to all my previous instructors from the UTC and Chattanooga State Mechanical Engineering Departments for giving me the knowledge and experience that allowed me to further understand and analyze my area of research.


B. S.; An honors thesis submitted to the faculty of the University of Tennessee at Chattanooga in partial fulfillment of the requirements of the degree of Bachelor of Science.




Computational fluid dynamics -- Computer simulation; Propulsion systems


CFD Computational Fuild Dynamics; Mechanical engineering; AIAA; Aerospace; Pointwise grid generation; Nozzle comparison


Mechanical Engineering

Document Type



32 leaves




Under copyright.