Size-scaling effect on the velocity field of an internal combustion engine, part II: Turbulence characteristics

Abstract

Top dead center turbulence characteristics in two geometrically scaled, two-valve, single-cylinder research engines were investigated to study engine size scaling relationships. Velocity data were acquired using high-magnification particle image velocimetry with different port geometries (two), different port orientations (two), and with both shrouded and nonshrouded intake valves. The data were analyzed using both an ensemble-average method and spatial-average method of defining the mean velocity field. The fluctuating velocity fields were analyzed by calculating the spatial two-point correlation functions and integrating them to find the integral length scales L 11 and L 22 and also by fitting the turbulent kinetic energy spectrum to a model energy spectrum to find the integral length scale £ and the turbulent Reynolds number Re £ . An extensive comparison between the different analysis methods was undertaken; the spatial-mean method provides more internally consistent results, in part due to the elimination of large, anisotropic structures from the analysis. The integral length scales obtained in the two engines normalized by their respective clearance heights were on average 20% larger in the large engine than in the small engine when using the ensemble-mean method, significantly less than the scale ratio (1.69), and were independent of intake port geometry. This scaling with clearance height has been anticipated, but never experimentally demonstrated. The turbulence intensity in the two engines was found to directly correlate to the mean piston speed, thus the effect of engine size on turbulence intensity is limited to its linear relation to piston speed. A given intake geometry produced the same scaling factor between turbulence intensity and mean piston speed for the two engine sizes. The turbulence intensity was found to be collapsed to a single linear relationship when plotted against a modified Mach index, which normalizes the mean piston speed with a mass-weighted flow coefficient.

Development and Application of Bivariate 2D-EMD for the Analysis of Instantaneous Flow Structures and Cycle-to-Cycle Variations of In-cylinder Flow

“…However the decomposition is based on the harmonic functions, in addition like most of turbulent flows, the choice of a cut-off frequency/length remains a complex subject [8,26]. In addition, various spatial filter techniques such as Gaussian kernel exist, which distinguish CCV from turbulent fluctuations [19,38]. However, the results depend again on the filter size.…”

Scaling liquid penetration in evaporating sprays for different size diesel engines

“…Although the similarity of engine performance and emissions can be achieved to some extent, as shown in the above literature review, the existing three scaling laws exhibit several weaknesses that should be further addressed. For example, the brake thermal efficiency (BTE) is difficult to be completely scaled due to the increased heat transfer loss 18,19 and friction loss 20,21 of small engine caused by the larger surface to volume ratio. In the previous study, the intake temperatures of the small engines were usually increased to compensate for the effect of larger heat transfer losses.…”

Similarity of split-injected fuel sprays for different size diesel engines

“…They found that the S rule is more preferable in the LTC regimes, due to the identical timescale for chemical reactions. Recently, Heim and colleagues 12,13 investigated the similarity of turbulence properties in different size engines. They found that the swirl center locations are in nearly the same location between the large and small engines.…”