INTAKE-PORT FLOW BEHAVIOR IN A MOTORED AND FIRED 2-STROKE RESEARCH ENGINE

Velocity information was collected from an intake port of a single-cyl inder piston-ported two-stroke engine by a laser-Doppler velocimetric (LDV) system to better understand and quantify the behavior of intake flow exiting into the cylinder during the scavenging process. Ten meas urement locations were chosen along a vertical line through the center of the port exit area. Motored, fired (skipfired), and steady (statio nary engine) radial velocity measurements were recorded at engine spee ds of 600, 900, and 1200 rpm along with cylinder, intake (two position s), and exhaust pressure data. The ensemble-averaged mean radial veloc ities during motoring, when plotted versus crank angle, are generally flat over most of the port area and influenced by changes in pressure differential between the intake and exhaust ports. During the initial phase of the port opening, a high-velocity jet exits from the port, ca using a peak in the mean radial velocity profile. As the port opening area increases, the reattachment point within the port passage apparen tly moves toward and beyond the exit plane of the port, resulting in n ear-zero velocities for the measurement points close to the top of the port. Mean velocities measured after the firing cycles are strongly a ffected by the expansion waves in the intake system, which induce back flow of cylinder gases at 600 and 900 rpm. At 1200 rpm, velocities for the motoring and firing cycles are very similar since the pressure di fferential between the intake and exhaust ports during the intake peri od are similar for the two cases. The root mean square (rms) velocity fluctuations are generally higher in the fired engine over all the mea sured engine speeds. Properly normalized ensemble-averaged mean veloci ty profiles along the measured direction are not shaped like a top hat but are very similar for motored, fired, and steady engine conditions , particularly at higher engine speeds. The volumetric flow rate is es timated using LDV results, compared with the actual measured value, an d discussed.