Date of Award

Summer 1995

Document Type


Degree Name

Doctor of Philosophy (PhD)


Mechanical Engineering

Committee Director

Surendra N. Tiwari

Committee Director

Ajay Kumar

Committee Member

Sushil K. Chaturvedi

Committee Member

Arthur Taylor


A numerical study is conducted to simulate shock-induced combustion in premixed hydrogen-air mixtures at various free-stream conditions and parameters. Two-dimensional axisymmetric, reacting viscous flow over blunt projectiles is computed to study shock-induced combustion at Mach 5.11 and Mach 6.46 in hydrogen-air mixture. A seven-species, seven reactions finite rate hydrogen-air chemical reaction mechanism is used combined with a finite-difference, shock-fitting method to solve the complete set of Navier-Stokes and species conservation equations. In this approach, the bow shock represents a boundary of the computational domain and is treated as a discontinuity across which Rankine-Hugoniot conditions are applied. All interior details of the flow such as compression waves, reaction front, and the wall boundary layer are captured automatically in the solution. Since shock-fitting approach reduces the amount of artificial dissipation, all the intricate details of the flow are captured much more clearly than has been possible with the shock-capturing approach. This has allowed an improved understanding of the physics of shock-induced combustion over blunt projectiles and the numerical results can now be explained more readily with one-dimensional wave-interaction model than before. For Mach 5.11 the flow field is found to be unsteady with regular periodic oscillations of the reaction front. There is a progression of higher frequency and lower amplitude oscillations as the Mach number is increased with a steady flow observed at some point above the C-J velocity. Numerical results show good qualitative agreement with the ballistic range shadowgraphs. In addition, the frequency of oscillations, determined by using the Fourier power spectrum is found to be in good agreement with the experiment.

Various parameters for the triggering of the instabilities have been identified. Projectile diameter is one of the parameter and an unstable reaction front can be made stable by choosing an appropriate small diameter projectile. The other parameter is the heat release rate which, in turn, depends upon the free-stream pressure. A number of simulations of shock-induced combustion past blunt projectiles in regular and large-disturbance regimes are also made at a Mach number of approximately 5 and pressures in the range of 0.1 to 0.5 atm. For a free-stream pressure of 0.1 atm, the reaction front is steady; at a pressure of 0.25 atm, the reaction front develops regular, periodic oscillations. As the pressure is increased to 0.5 atm, the oscillations become highly pronounced and irregular. Combustion with periodic oscillations has been classified as a regular regime and combustion with large, irregular oscillations has been classified as a large-disturbance regime. These calculations are in agreement with the experimental observations from ballistic-range tests. The transition from steady reaction front to regular, periodic oscillations, and then to large-disturbance regime is explained by a one-dimensional wave-interaction model.