Poppet Valve Dynamics

Abstract

Measurements of the motion of a pushrod-operated poppet valve at normal operating speed showed considerable deviation from the motion at low speed. Jumping was observed at a speed very much lower than the speed which would be predicted by the usual analysis. The observed discrepancy is attributed to flexibility in the operating mechanism. A mathematical analysis is presented which takes account of flexibility and gives a reasonable check on the observed motion. The effect of speed, valve spring pressure and cam form was investigated by means of a mechanical model. Special cam forms for eliminating jumping at a single speed are suggested. T lift curve of a poppet valve in operation is usually assumed to be identical with the lift curve at a very low speed. Few actual measurements of lift curves under operating conditions have been published; among these was an article written in 1929 by Jehle and Spiller. In this article, the authors dealt mainly with the effect of cam form on valve spring vibration, mentioning the rigidity of the parts of the valve gear as a contributing cause of faulty action, but offering no computations to show the effect of this variable. The experiments and analysis presented herein are for the purpose of supplying this deficiency in the literature on poppet valve mechanisms. The result of flexibility in the valve operating mechanism is to cause the valve motion to differ from that prescribed by the cam. In such cases the actual valve lift curve determines the motion of the springs, and so its harmonic components, and not those determined from the cam form are significant for spring surge predictions. Along with the effect on spring surge, flexibility may give rise to unexpectedly high stresses in the valve gear, with consequent pounding out of valve seats, distortion or fracture of valve stems and heads, and pitting or battering of cams and cam followers. In addition to these mechanical difficulties, the change in lift may cause a change in breathing capacity with a consequent change in output. EXPERIMENTAL PROCEDURE The valve gear of a dummy engine with pistons, rods, and extraneous parts removed, was arranged to be operated by an electric motor. Measurements of valve position were made by a telescope equipped with a crosshair and micrometer adjustment as shown in Fig. 1 This illustration also shows the means for stroboscope illumination, controlled by a movable contact on the crankshaft. For each measurement, the contact was STROBOSCOPE MICROMETER SLIDE/ STABLE CONTACT CRANKSHAFT Presented at the Power Plants Session, Seventh Annual Meeting, I. Ae. S., January 26, 1939. F I G . 1. Measuring apparatus. set to flash the stroboscope at an arbitrary crank angle while the corresponding position of the valve was observed through the telescope. Valve lift curves determined from these measurements, both with and without gas pressure loading on the valve, are presented in Figs. 8 and 9. Note in Fig. 8 the appearance of jumping at a speed very much lower than the normal operating speed (1900 r.p.m.). By comparing the measurements with a relatively simple mathematical analysis, it was found that the valve motion could be predicted with reasonable accuracy. In addition, the analysis gives an understanding of the basic nature of the problem.