Turbocharger systems enhance the engine power and efficiency, reduce its pollution, and downsize the engine volume. As the significance of exploiting turbochargers in gasoline engines is surging among automobile manufacturers, the necessity of improvement in the system components becomes more critical. This paper investigates the impacts of a turbocharger turbine bypass and wastegate geometry alterations on the performance and the flow field characteristics via numerical simulations. The numerical method verification and mesh independence study are performed for the original geometry. The simulation results of the altered geometries indicate that better alignment between the bypassed and the bulk flows leads to higher efficiency of the turbocharging system. In addition, the bent or elbow-shaped and protruded turbine walls immediately downstream of the wheel are found to be unfavorable. It is also uncovered that if the wastegate shape and the housing walls are designed in such a way that the effects of ensuing vortices are minimized, it improves the stage efficiency, which is desirable for two-stage turbochargers. Furthermore, a novel manufacturable design is proposed in this study, which increases the efficiency and useful power by 19. 9% and 6. 7%, respectively.

The gas flow estimation is crucial for the proper operation and monitoring of turbocharged (TC) engines with a torque structured engine control unit (ECU). This paper presents mean value models developed for modeling gas flow over the compressor, the turbine and the wastegate (WG) in a TC gasoline engine not equipped with a hot film air-mass flow meter (HFM). The turbine flow was modeled by an isentropic nozzle with an effective pressure ratio. A physically sensible model was developed for estimating the WG flow and its position which estimates them over a wide range of the engine operating conditions. Finally, a novel model for estimating the compressor air flow was proposed, using coupling between powers generated by the turbine and used by the compressor. All models are simple and accurate enough to be utilized for engine real-time control applications. Experimental results from a modern gasoline TC engine were used to validate all the models. Prediction results from models were in very good agreement with the experimental measurements.