Page 86 - Methods For Monitoring And Diagnosing The Efficiency Of Catalytic Converters A Patent - oriented Survey
P. 86
68 Methods for Monitoring and Diagnosing the Efficiency of Catalytic Converters
6) determining the catalytic converter efficiency from a predetermined catalytic converter
efficiency characterized as function of a transport time delay and time constants
The equations of step 5 are formed as follows: The exhaust is described as a transport delay
time T, (due to time from fuel delivery to engine cylinder till exhaust gas reaches the EGO
sensor) and a set of first order low pass filters connected in series (due to catalytic converter
oxygen storage damping of exhaust fluctuations, response of EGO, physical mixing and
chemical reactions of the exhaust gases in the exhaust pipe).
In a form of Laplace transform, the transfer function lV,,(s) of the system is
where To are the low pass filter constants.
1” and 7d are unknown parameters which are to be determined during a test
A controller with transfer function in Laplace form
can be used, where H is the calibratable gain (jumpback) and G is the controller ramp
The limit cycle period 7;, of the controller can be calculated from the following system of n+l
equations with (n+l) unknown parameters
where:
c, is a known function of the controller jumpback H and the time constants T, .
The set of equations is formed by changing G, H or both of them n times. Additionally, (n+l)
limit cycles Ti corresponding to each controller parameter setting should be measured. Then
the set of (n+l) algebraic equations may be solved using any known numerical method. The
catalytic converter efficiency is then determined from a predetermined catalytic converter
efficiency characterized as a function of the calculated transport time delay and the calculated
time constants.
In US5159810 (1992) the downstream EGO sensor is used for the engine closed loop airkel
control during monitoring of the catalytic converter. The measured limit cycle frequency at
various engine speednoad conditions is compared to stored benchmark limit cycle frequencies
