An investigation into the thermal field associated with typical automobile engine-bay flow fields

Computational Fluid Dynamics (CFD) modelling has been widely used within the automotive environment for areas such as the enhancement of the external aerodynamic performance of automobiles and the optimisation of the combustion process. More recently, however, the study of vehicle thermal management systems (VTMS), an area which encompasses the cooling, air conditioning and underhood airflow of vehicles, has embraced the use of CFD in an attempt to refine designs. Additionally, with continuing emphasis being placed on noise, vibration and harshness (NVH), engine design has tended towards neater, more integrated packages with a minimum of mechanical ancillaries; such as brackets. One of the significant aspects of this design philosophy is that there is now a proliferation of electronics in modern vehicles. One of the solutions to increasingly complex wiring harnesses has been to decentralize and migrate control units to the engine and the powertrain. This development has had consequent effects on environmental factors related to the powertrain control unit (PCU) itself as increasingly strict legislation limits the kind of design solutions which may be employed to overcome these effects, the underhood area has had to be researched thoroughly. For the work described herein, a CFD package and associated pre- and post-processors was used to assess the underhood flow characteristics with a view to determine the optimum positioning of these component packages. A further goal arose from the opportunity to check if the CFD package would be suitable for a non-specialist operator to use. To achieve these goals, the underhood volume of a vehicle was modelled both numerically and experimentally. The experiments were carried out using a physical instrumented model in the smooth controlled flow of the Department's Handley Page wind tunnel. This provided the analogue model for a generic car, against which the numerical results could be compared. The numerical modelling was performed using the commercial CFD package FLUENT 5.5. It was envisaged that the data gained from the two models could be used to enhance current design procedures and influence future design methodologies in engine compartment layout, with regard to the placement of electronic devices. It may also serve to validate the integration of data provided by simulation and experiment. It must be noted that the work was essentially an assessment of the usefulness of the methodology. Albeit the comparisons between CFD and experiment appear to possess significant differences, this early work illustrates the potential of the procedure and suggests the "follow-on" research that is required