Simulative Prediction of Leakage for Seat Valves and Bio-Hybrid Fuels

Seat valves play a critical role in various technical applications, such as automotive injectors. Here, predicting leakage is vital, as it can lead to poor combustion behavior or complete system failure. To address this, a simulation model originally designed for predicting leakage of air and hydrogen in ball seat valves was adapted for the use with bio-hybrid fuels. The simulation model uses a modified flow equation based on the Hagen–Poiseuille equation and the effective medium approach to calculate leakage flow. The corresponding input parameters are calculated according the Persson’s contact mechanics considering the surface roughness at the sealing zone. This method considers the elasto-plastic deformations of the surface asperities on a microscopic level, leading to a description of the contact pressure relation. This relation is used in the calculation of the elastic deformations on the macroscopic level. Good agreement between the model and experimental results was found for air and liquid water. Since many bio-hybrid fuels differ significantly in their fluid-mechanical properties from conventional fuels and gases like air and hydrogen, the simulation model cannot be used for these liquids without any adjustments. For instance, leakage is influenced by their low viscosity, low density, and high vapor pressure of these fuels. Furthermore, deviating interfacial interactions between liquid and solid bodies within the valves is of great importance when investigating leakage. For this reason, additional experiments and simulations have been conducted for bio-hybrid fuels. This study aims to investigate the influence of deviating fluid properties of bio-hybrid fuels on the prediction of leakage in seat valves in comparison to conventional fuels. For this reason, the relevant fluid properties are determined for selected bio-hybrid fuels and implemented in the simulation model. Subsequently, the obtained results are compared with experimental data.