Statistical Correlation Between the Crankshaft’s Speed Variation and Engine Performance—Part II: Detection of Deficient Cylinders and Mean Indicated Pressure Calculation

Abstract

The goal of this two-part paper is to develop a methodology using the variation of the measured crankshaft speed to calculate the mean indicated pressure (MIP) of a multicylinder engine and to detect cylinders that are lower contributors to the total engine output. The statistical model of a harmonic component of the engine torque developed in the first part of the paper is used to achieve this goal. The analysis of the half-order components of the gas pressure torque permits to identify distinct phase angle domains of the resultant torque vector that are specific for the deficiencies of given cylinders. Based on the rigid-body model of the crankshaft, these phase angle domains are correlated to the phase angle domains of the half-order component of the crankshaft speed. Then, the phase angle of the half-order component of the measured crankshaft speed will identify the deficient cylinder. The amplitude of the first major harmonic component of the measured crankshaft speed is correlated to the corresponding harmonic order of the gas pressure torque and is used to calculate the MIP of the engine. The accuracy limits of this “software dynamometer” are also presented.