Mysterious oxygen sensor fault - why does the engine light turn on, and why not?
The car in question is a Lancer 1.6 16 v 113 hp SOHC petrol, engine code 4G92, model year 2000, driven 142,000 km (88,200 miles) - but I suppose similar problems could be possible also on other cars / models.
When the engine light (CE) lights up, it gives the fault code 11, which is for the front oxygen sensor (before the catalyst). The light will go on or off in the following situations:
1) Driving over 70 km/h (43 mph, 2,000 rpm) for about 6 km (3.7 miles) lights the lamp. When I disconnected the sensor's cable (ie, the connection between it and the ECU), the lamp was lit in exactly the same way, with the same delay. At lower speeds / turns, the lamp will not light (with the cable connected).
2) At longer driving (up to 100 km/62 miles) at 100-120 km/h, the lamp can be switched off. Same after shorter driving when accelerating with full throttle eg 90-120 km/h. If the lamp does not turn off on the first attempt, it will probably turn off in a succeeding attempt. When the light goes out, it remains off also at lower speeds and as long as the engine is running. When the engine is switched off and restarted, the lamp lights up as above.
3) At cooler outside tempertures than minus 7 centigrades (19,4 F), the lamp does not light up, no matter how you drive.
Could anyone explain these phenomena, even in part? What happens during acceleration that causes the lamp to go off? Is there some part in the car getting warmer than usual (which?), and why does that turn off the light? Why, on the other hand, does outside temperatures below a certain level keep the light off? The best thing to know would of course be how fix this problem.
The following research and actions has been done so far:
1) The oxygen sensor has been changed three times, the first two times at the dealership (to a Denso sensor) and the last time I changed it myself to a chinese noname one (still with the correct type of contact). After each change, the problems have remained exactly the same.
2) The heater circuit of the Denso sensor has a resistance of 13.8 ohms, the chinese one has 6.0.
3) The heater is fed with 12 volts from the car when it is running, measured at the connector connecting to the sensor.
4) The alternator-regulator gives a steady voltage.
5) The resistance of the sensor's signal cable on the car side (= the cable between the ECU and the connector to the sensor) is 1.0 ohm - which should indicate a cable in good condition. The resistance of the corresponding grounding cable is 1.1 ohms, as of the heating cable ground cable - so even these seems to be ok.
6) When I learned that also the mass air flow sensor could cause the same fault code, I checked the 8 cables to that one, which also were ok. The inside of the sensor itself is clean.
7) The capacitors of the ECU has been renewed, considering the age of the car.
8) Possible leaks on the intake and exhaust side has been checked.
Regarding the error history, the engine light was first lit when the car was only a couple of years old, but this was sporadic and was not repeated after a fault code reset. Several years later, the light started getting on at a regular basis, but after a longer and faster drive than now (highway for about 30 km). After that the renewals of the oxygen sensors started, without any impact on the problem.
I'm not an electrician, but a detail I'm wondering about is that the circuit diagram shows a ground shield around the signal cable (I don't know what that shield is called officially, but it's due to the weak signal of the cable of typically 0.5-1 volts). However, I can't see such a thing, which I imagine should be some kind of cover or extra layer around the cable in question. The cable looks exactly the same as the other three ones to the sensor, or the other about 60 cables connecting to the ECU. Does anyone know how such a protective shield should look like? If it's lacking, could it explain the problems?