The development of the On-Board Diagnostic Systems (OBD) requires the determination of the critical level of the poisoning of the exhaust aftertreatment components.
In order to have an accurate analysis, one way is to separate the effect of the poisoning of the two components: oxygen sensor and three way catalyst [ 1 ].
Our experimentation has been concentrated on poisoning of lambda sensor.
An apparatus fed with model gases was realized to evaluate the poisoning of oxygen sensors coming from a fleet test of 6 taxis after a mileage of 100000 km. Our apparatus has been able to differentiate the poisoning level of the sensors by measuring their “switchability”.
To confirm these measurements the lambda sensor switchability has been verified with a bench test using the Mercedes- Benz M111E engine. Exposing the aged lambda sensor to the engine exhaust gas, its response was coherent with its behavior on the apparatus fed with model gases.
On the same engine, each aged lambda sensor was coupled with a UEGO (Universal Exhaust Gas Oxygen) sensor, which is able to detect the air-fuel ratio value of the burned mixture. Analyzing the statistic distribution of Air Fuel-Ratio (AFR) it has been verified a correlation between the different lambda sensor switchability with the width and the symmetry of AFR excursion in respect of stoichiometric value.
The switchability performance of aged lambda sensors was able to explain the different Three Way Catalyst (TWC) efficiency of the six taxi vehicles.
Running the six taxi with their aged catalysts (100000 km) on chassis dyno, ECE cycle EUDC has been performed to measure the effect of the replacement of aged lambda sensors with a new lambda sensor.