Modified multiflame model for AP-HTPB composite propellant combustion

Among composite solid propellants, AP/HTPB is the most widely used. In spite of being studied for the last few decades, modeling for predicting the burn rate of these heterogeneous propellants still remains an active research problem, and new models for the same have been proposed in the recent literature. Amongst them, BDP and Cohen and Strand formulation are mostly used, which uses phenological models to predict the burn rate and parameter sensitivity in a quick time. A modified model is proposed in the present work. The modifications include (1) Three flame structure along with separate surface temperature for oxidizer and binder, (2) Detailed energy balance equations at surfaces, (3) Modified surface heat release term, (4) Flame temperature as a function of initial temperature, pressure, and fuel/oxidizer ratio, and (5) Modification of diffusion flame stand-off distance formulation. The predicted burn rates from the modified model are compared against experimental and other theoretical results for AP monopropellants, AP-HTPB monomodal, and multi-modal configurations, including packs. The new modified model gives better predictions than the existing ones in all the cases considered in this study. The temperature sensitivity of the AP monopropellant model comes well within the range of experimental values available in the literature, and the value drops slightly as chamber pressure increases. For composite AP/ HTPB, the temperature sensitivity lies within the experimental range. However, it does not give any particular trend and is highly dependent on the type of flame dominance for that particular composite configuration and pressure. The current modified model is highly robust as there is excellent agreement with experimental data from different research groups incorporating a wide range of formulations covering monopropellant to composite multi-modal packs accounting for a range of particle size and solid loading variation. The model gives the results quickly, making the model very efficacious in terms of time, effort, and computational resources.