COMBUSTION ANALYSIS AND CYCLE-BY-CYCLE VARIATIONS IN SPARK-IGNITION ENGINE COMBUSTION - PART 1 - AN EVALUATION OF COMBUSTION ANALYSIS ROUTINES BY REFERENCE TO MODEL DATA
Jk. Ball et al., COMBUSTION ANALYSIS AND CYCLE-BY-CYCLE VARIATIONS IN SPARK-IGNITION ENGINE COMBUSTION - PART 1 - AN EVALUATION OF COMBUSTION ANALYSIS ROUTINES BY REFERENCE TO MODEL DATA, PROCEEDINGS OF THE INSTITUTION OF MECHANICAL ENGINEERS PART D-JOURNALOF AUTOMOBILE ENGINEERING, 212(D5), 1998, pp. 381-399
This paper investigates a number of new approaches to mass fraction bu
rned analysis for application to spark ignition engine combustion proc
esses. The rationale for these new approaches is to retain the simplic
ity and computational efficiency of the technique of Rassweiler and Wi
throw, but to overcome some of the limitations inherent in the standar
d methods. The approaches investigated include the use of a two-zone m
odel to determine information about the burned and unburned gas temper
atures and the modification of the standard Rassweiler and Withrow exp
ression to incorporate polytropic indices for compression and expansio
n. The investigation makes use of an engine simulation model, which wa
s used to generate the pressure data. This technique provided a 'known
' burn rate against which to evaluate the methods investigated. Some e
xperimental data taken from a Rover K4 optical access engine is also p
resented. The successful aspects of this investigation are the use of
the Rassweiler and Withrow equation with different compression and exp
ansion indices, and the application of this analysis to investigate cr
evice gas burn-up. What was found to be not successful was the use of
a two-zone model for calculating gas temperatures. The results indicat
e that the model is, in general, not reliable for computing temperatur
es and this is due to the temperature gradient in the burnt zone and t
he disproportionately high rate of heat transfer from the mixture that
burns first during combustion. The mass fraction burned calculations
using this model were also found to be not as accurate as those based
on simpler models.