Location
Room # EB302
Start Date
5-6-2019 9:30 AM
Description
Rapid acceleration of rocket engine turbopumps during start-up imparts significant transient effects to the resulting flow field, causing pump performance to vary widely when compared to quasi-steady operation. To improve turbopump design in response to the transient effects of start-up this paper presents a method to simulate turbopump startup using CFD. Cavitating pump performance is initially evaluated using a simulation with a constant outlet pressure boundary condition. Based on the difference between simulation inlet pressure and target inlet pressure, the defined pressure on the outlet boundary condition is modified. This process is repeated until simulation inlet pressure is essentially constant during start-up. Using this simulation method, the performance of a centrifugal turbopump during start-up is simulated. Reasonable solution convergence is reached in one single phase and four cavitating simulation iterations. After these five simulation iterations, the average error between inlet pressure and inlet target pressure is 10%. Cavitating simulation iterations 3 and 4 agree within 11% on average for inlet total pressure during startup, 0:1% on average for head coefficient, 13% on average for cavitation volume, 20% on average for flow coefficient, and 2% on average for RMS force on the impeller. The agreement between simulation iterations 3 and 4 suggests that a reasonable solution has been reached.
Cavitation Inception and Performance of a Centrifugal Impeller During Startup
Room # EB302
Rapid acceleration of rocket engine turbopumps during start-up imparts significant transient effects to the resulting flow field, causing pump performance to vary widely when compared to quasi-steady operation. To improve turbopump design in response to the transient effects of start-up this paper presents a method to simulate turbopump startup using CFD. Cavitating pump performance is initially evaluated using a simulation with a constant outlet pressure boundary condition. Based on the difference between simulation inlet pressure and target inlet pressure, the defined pressure on the outlet boundary condition is modified. This process is repeated until simulation inlet pressure is essentially constant during start-up. Using this simulation method, the performance of a centrifugal turbopump during start-up is simulated. Reasonable solution convergence is reached in one single phase and four cavitating simulation iterations. After these five simulation iterations, the average error between inlet pressure and inlet target pressure is 10%. Cavitating simulation iterations 3 and 4 agree within 11% on average for inlet total pressure during startup, 0:1% on average for head coefficient, 13% on average for cavitation volume, 20% on average for flow coefficient, and 2% on average for RMS force on the impeller. The agreement between simulation iterations 3 and 4 suggests that a reasonable solution has been reached.
Comments
Session 2