EFFECTS OF OPERATING SPEED ON 3-D MEAN FLOWS MEASURED AT THE END OF INTAKE STROKE IN AN IC ENGINE

Measurements oi the instantaneous in-cylinder flow fields at the end i ntake stroke at bottom dead center (BDC) were carried out in an FloDyn e(TM) water analog engine simulation rig using 3-D particle tracking v elocimetry (3-D PTV). The measurements were performed in a 4-valve per cylinder, typical pent-roof type combustion chamber with both intake valves activated generating mainly tumble vortex structures. The exper iments were repeated for two different operating points simulating an idle condition (600 PRM) and a low load condition (1200 RPM). These tw o operating points represent almost a factor of two in effective Reyno lds (Re) numbers. For each case, 100 cycles of data were acquired. Eff orts were made to optimize the particle seeding density (and the resul ting number of 3-D velocity vectors) to yield 500 to 600 instantaneous vectors at each cycle. The raw data coosisted of sets (between 50 000 and 60 000 velocity vectors each) of 3-D instantaneous (but phase con ditioned at BDC) velocity vectors, randomly (but relatively uniformly) distributed over the entire cylinder volume. Sophisticated statistica l tools were applied to these data sets to evaluate the ensemble avera ged mean flow field and total fluctuation fields in the root mean squa re (r.m.s.) sense. In addition, ''conventional'' integral measures suc h as tumble, cross-tumble and swirl ratios were computed from the data by integrating the angular momentum components over the cylinder volu me. The results indicate that the Re number effects are relatively sma ll between the two conditions investigated, with a tendency of the sma ller scale structures at BDC for the higher Re number case. This paper illustrates the tremendous potential of the water analog engine simul ation used in conjunction with 3-D PTV as a rapid engine flow field ev aluation tool. This approach allows to conveniently obtain 3-D maps of the mean and total r.m.s. fluctuation levels of each velocity compone nts as well as integral parameters such as tumble and ratios (under re alistic unsteady conditions). Furthermore, this method also allows var ious configurations or operating points to be conveniently compared wi th respect to many of their flow characteristics.