Transient flow simulation of a compressed truncated perfect nozzle

Abstract

Transient flow characteristics in compressed truncated perfect (CTP) nozzles with cold gaseous nitrogen were studied using axisymmetric Navier-Stokes computation. The numerical results are compared with the experimental results for validation. A mechanism which possibly causes serious side-load during the start-up transient of the rocket engine nozzle is discussed. In case of the CTP nozzle, the transition from the free shock separation (FSS) to the restricted shock separation (RSS) is more likely to occur in a long nozzle with a highly compressed contour. Fast movement of the Mach disk and the separation point as well as a rapid change of the wall pressure distribution in the nozzle were observed during the transition process between FSS and RSS and during the discharge process of the high-pressure regions in the reattached flow from the nozzle exit. As a result, a sudden change of forces imposed on the nozzle wall occurred. Under the same condition as those under which the sudden change of the forces occurred, remarkable side-load peaks were observed in the experiments. Therefore, a possible cause of side-loads during start-up transient is considered to be the asymmetry of the flow field in the circumferential direction during these sudden changes of the flow structure. INTRODUCTION Elimination of serious side-loads during a start-up transient is one of the most important issues in the design of nozzles for the first-stage rocket engines. Extensive works have been done to investigate the cause of side-loads. Most such studies have dealt with parabolic nozzles or thrust-optimized nozzles. A side-load is typically observed as an impulsive load that occurs at some * Senior researcher, Kakuda Space Propulsion Laboratory, NAL, Member AIAA. t Researcher, Kakuda Space Propulsion Laboratory, NAL, Member AIAA. J Group leader, Kakuda Space Propulsion Laboratory, NAL, Member AIAA. H Associate senior engineer, NASDA. Copylight©2001 The American Institute of Aeronautics and Astronautics Inc. All rights reserved. particular pressure ratios between the nozzle plenum pressure and the ambient pressure. In case of the thrust-optimized nozzle, the side-loads are closely related to a particular separation pattern, called restricted shock separation (RSS) , which is characterized by a reattachment of a separated boundary layer on a nozzle surface with an over-expanded condition. However, the mechanism, which causes the side-loads, has not been fully understood. Currently, transient flow characteristics in the compressed truncated perfect ( CTP) nozzle proposed by Hoffman are being investigated under collaboration between NAL and NASDA, in which both cold gas flow experiments and numerical simulations have been carried out. In the present paper, the results of the axisymmetric Navier-Stokes computation of the transient flow characteristics in the CTP nozzles are reported. The results are compared with the experimental results for validation. The transient flow characteristics of the nozzle flow as well as a possible cause of serious side-loads at the start-up transient are discussed. NUMERICAL PROCEDURE A axisymmetric Navier-Stokes computation was applied to simulate the nozzle flow during the start-up transient. The test gas was nitrogen and was assumed to be a perfect gas. The molecular weight and the specific heat ratio were 28 and 1.4, respectively. The ambient gas around the nozzle was also assumed to be nitrogen, instead of air. The flow field was assumed to be fully turbulent and the one-equation turbulence model of Spalart and Allmaras was employed in the present computation. The convection term and the viscous term were discretized by the AUSM-DV scheme with the third-order MUSCL interpolation and the second-order centered scheme, respectively. Time integration was carried out using an implicit method based on a point-Jacobi relaxation at each time step to reduce the computation time required for computation of high Reynolds number flow. It should be noted that the implicit time integration method used here is not a time accurate procedure for unsteady computation. However, major phenomena during the start-up transient in the cold flow experiment, such as movement of the Mach disk and the separation point, were likely in a 1 American Institute of Aeronautics and Astronautics c)2001 American Institute of Aeronautics & Astronautics or Published with Permission of Author(s) and/or Author(s)' Sponsoring Organization.