Catalyst Monitoring

General description of Catalyst Monitoring

Catalyst monitoring is based on monitoring its oxygen storage capability.

The engine closed loop feedback control generates lambda (air/fuel ratio) oscillations in the exhaust gas. These oscillations are dampened by the oxygen storage activity of the catalyst. The amplitude of the remaining lambda oscillations downstream of the catalyst indicates the storage capability.

In order to determine catalyst efficiency the oscillation of the upstream sensor is need to calculate the "02-in and-output (catalyst)" by engine air mass and lambda-deviation. The downstream sensor signal for a "threshold catalyst" then is derived (model) from this basic value.

Any time the real sensor signal oscillation (downstream) corresponds to the model an defective catalyst is recognized.





Monitoring structure of Catalyst

Monitoring Cycle

Computation of the efficiency of the Catalyst

The alternating-current component of the voltage of the Oxygen Sensors before and after catalyst is determined, rectified and averaged.

The actual quotient ARact of the voltage of the oxygen sensor before catalyst and the voltage of the oxygen sensor after catalyst is determined. Simultaneously the theoretical quotient ARtheo of this voltages of the oxygen sensors is computed in dependency of the operating point of the engine. The respective operating point is determined using the parameters load and engine speed.





For Diagnosis the Difference AR between the actual quotient ARact and the theoretical quotient ARtheo is made.

The Difference AR expresses the efficiency of the Catalyst.

Within the time of Diagnosis an adaptation matrix is filled in dependency of the actual operating point of the engine.

Fault evaluation

At the end of the diagnose period the number of stored values in the adaptation matrix exceeding a limit is determined. If this number of stored values itself exceeds a threshold a defective catalyst is evaluated.

Check of monitoring conditions

The monitoring principle is based on the detection of relevant oscillations of the downstream sensor signal during regular lambda control. It is necessary to check the driving conditions for exceptions where no regular lambda control is possible, e.g. trailing fuel cut-off.

During such periods, and for a certain time afterwards, the computations of the amplitude values and the following processing is interrupted. Thus, a distortion of the monitoring information is avoided.