A CFD-model of the Fluid Flow in a Hydrogen Peroxide Monopropellant Rocket Engine in ANSYS Fluent 16.2
MetadataShow full item record
The use of hydrazine in monopropellant thrusters has dominated the market for several decades. Due to the toxic nature of hydrazine, the interest in developing thrusters based on green propellants has increased. A common issue found during the development of these thrusters is the optimisation of the catalyst used to decompose the liquid propellant into a hot gas. The work presented in this thesis was therefore performed in order to evaluate if it is possible to simulate the fluid flow of hydrogen a hydrogen peroxide thruster with the commercially available CFD software ANSYS Fluent 16.2. A successful simulation would make it possible to investigate the fluid flow inside the catalyst, which could lead to an optimised design without the large costs connected to the optimisation process through extensive tests. Through an extensive research on the different models available in the commercial CFD software ANSYS Fluent 16.2 it was found that the software should be able to model this problem. As the phases in a HTP thruster were found not to be in an equilibrium, the fluid flow had to be modelled by the complex Eulerian multiphase model. The different chemical substances were modelled through the species model. The pressure drop over the catalyst was modelled as a porous media. The heterogeneous reaction mechanism for HTP and vaporisation of water was modelled through a volumetric reaction model and two-resistance model respectively, as part of the phase interaction methods. The turbulent flow in the thruster was modelled with the $k-omega$ turbulence model. The simulations performed with the Eulerian multiphase model did not converge despite taking several measures to stabilise the solver. The simulation was therefore split into different cases, where one sub-case built on the previous case results, to find the root cause of the stability issues. Through these simulations it was found that the Eulerian multiphase model is not capable of handling the compressible flow in this simulation. An attempt to model the flow with the mixture multiphase model was therefore attempted. By comparing the simulation results obtained for mixture model case 1 and 2 with experimental test data the results in case 1 were verified and validated. The porous media model was concluded to be modelled incorrectly as the pressure drop far exceeded results found during experimental test. The y+ values obtained in the solutions were used to conclude that the $k-\omega$ turbulence model used in this simulation was accurately modelled for the overall goal of the simulation. Instability issues were found when the reaction mechanism was enabled. Fluent reported that divergence in the x-momentum equation had occurred. An error in this equation could have many root causes, as the equations in the simulation is closely linked to each other. The results obtained prior to divergence in this case displayed a large pressure in the thruster catalyst region. To remove the porous media as a source for divergence, the results obtained from the pure nozzle flow (case 1/3) was used as an initial field for the reaction case. The solution diverged in this case as well, with the same results as the previous case. The issue connected to this problem was not found. Through these simulations it was concluded that with the setup used in this simulation it is not possible to simulate the fluid flow in a monopropellant rocket engine in ANSYS Fluent 16.2 due to stability issues connected to the Eulerian multiphase model and by enabling the heterogeneous reaction model modelled in the mixture multiphase model.