Numerical evaluation of the applicability of steady test bench swirl ratios to diesel engine dynamic conditions

Engine coherent flow structures such as swirl and tumble motions are key factors for the combustion process due to their capability to rise turbulence levels and enhance mixing which, in turns, severely influence both fuel efficiency and pollutant emissions. Automotive industry has therefore put great efforts over the last decades in evaluating air flow during induction stroke and air flow within the cylinder. Nowadays swirl and tumble motion characterizing a specific cylinder head are evaluated experimentally at design stage mainly using stationary flow benches. Such tests allow characterizing each head prototype using non-dimensional parameters like swirl and tumble ratios and, finally, to compare the different designs. In the present work the authors focused their attention on the swirl ratio characterization, firstly reviewing the two main methodologies for evaluating such parameter and more precisely the AVL and the Ricardo ones. A numerical method is then proposed in order to reproduce the stationary test bench with the final goal to develop a fast and accurate virtual test bench for cylinder head design. Simulations have been carried out on different VM Motori engine heads for which experimental data were available. The comparison between computational and experimental swirl ratios allowed to evaluate the suitability of using a virtual test bench as alternative or complementary to experiments. These results widened the understanding of the swirl fluid-dynamics and suggested that care must be taken when comparing duct designs having no geometrical similarity. Finally dynamic simulations have been performed for the head prototypes in order to compute the engine swirl in realistic conditions and to compare it with the steady bench results. This allowed evaluating the capability of the two different "static" swirl ratio definition (AVL/Ricardo) in correctly estimating real engine swirl. \ua9 2015 The Authors. Published by Elsevier Ltd