CFD study and method development for air intake systems using OpenFOAM

Abstract: This report is the result of a joint collaboration between the two Swedish companies AFRY and Scania AB. The need of CFD analyses for internal flow purposes are continuously increasing in companies like Scania. However the use of CFD software is limited to license costs. The objective and idea of this thesis is to esablish whether OpenFOAM, an open-source CFD software, could be a suitable alternative to the currently used CFD commercial software StarCCM+ when performing total pressure drops analysis on air intake systems. Furthermore, a second objective is to develop a standard method to analyze this class of systems using OpenFOAM. Air intake systems play a significant role in the efficiency of the engine of a bus or a truck, being responsible to drive clean fresh air to the engine with the minimum amount of pressure drops. After an illustrative introduction of the topic and theory principles, the following technical analysis focuses first on a CFD study of the simple Buice and Eaton 2D diffuser, to primarily benchmark OpenFOAM, StarCCM+ and the experimental data in a turbulence models and a mesh independence analyses. For both software the turbulence model RNG k − ε fails almost entirely to detect the separation and wide re-circulation bubble formed in the diffuser, while k − ω SST, being not much more computationally expensive, gives very accurate results. For these reasons, even the velocity profiles, pressure coefficient and skin friction coefficient plots analyzed with RNG k − ε turbulence model show larger discrepancy compared to the experimental data.  Secondly, a total pressure drop CFD study of the Scania air intake system follows. Performing multiple simulations for a range of different mass flow rates, both OpenFOAM and StarCCM+ show very good agreement with the experimental data. This time, due to the lack of big separation zones in the system, k − ω SST and RNG k − ε perform quite similarly, the therefore k − ε relative inexpensiveness in this more complex and heavy system makes the latter the most convenient turbulence model for this kind of study. Looking at different parameters such as user-friendliness, time required, software costs and versatility, OpenFOAM proves itself to be a convenient, accurate and valid alternative to its expensive commercial counterpart. Finally, following these conclusions a user-guide method to analyze not only the single but the entire class of air intake systems for Scania internal purposes is developed and reported.