Wideband giving strange readings

[mike_980]

Hi all, I have a Colt/Mirage which I have done a gen 2 GSR conversion on (4g93T). I have an AEM wideband fitted just before the cat which does nothing except for give me a reading inside the car. Up until yesterday I hadn't had any problems and it would happily sit between around 14.5 and 14.8 at idle then drop to between 12 to 12.5 under full load. Yesterday in 3rd gear under hard acceleration I got some hesitation. I looked down to see what my AFR was and I was maxed out lean - not even reading a value. Worse than that, even steady state cruising the AFR seemed to be mostly lean with the occational flick to 10 then back to lean. Today when idling at temperature the AFR bounces really quickly between 12-18 and just doesn't seem to be able to find a place it is happy although the engine feels, sounds and drives perfectly fine.

The plugs don't look too bad but they do have some evidence of lean combustion, they certainly don't look perfect and they have covered about 3000 miles.
Plenty of fuel is getting to the rail - not done an exact measurement but I have pumped into a bottle and the flow seems what I would expect.
My EvoScan cable doesn't work at the moment for some reason, so I checked the 02 voltage with a multimeter and it just seems to sit at about 80mV. I've only checked this at idle and after checking I looked at what the wideband was reading and it was maxed out lean.

The 02 sensor has 2 wires to it so presumably a ground and a signal. I did notice that when I checked the ground, I had 150Ohms of resistance. Is this value correct? The fact it was such an exact value made me think that it was going through a pull up/down resister in the ECU.

I haven't yet checked continuity between the sensor and ECU - does anybody know what pins I should be checking?

 

[Benzo4gT]

Being that your plugs look lean I'd be looking at fuel.
Do you have a way of checking fuel pressure? What fuel pump are you running at the moment? When was the last time the fuel filter was changed? What state is the fuel tank in, i.e. is it full of crap that may have blocked the foot filter of the pump?

The position you have your WB in is a safe spot as far as heat goes, mine is in a similar spot. Test you wideband first with this simple test and see what the results are.
https://www.youtube.com/watch?v=rmnxKLjM25g

 

[Benzo4gT]

Also, my sensor had 5 wires from memory..............

 

[bazeng]

Sensor sounds like it is bad.
Either try and recalibrate or compare with a second sensor.

This has happened to me a few times and itr usually gets resolved worth a free to air recal or worse case replacement of sensor.

What type of wideband?

4 or 5 wire?

Remember sone rules for wideband;
1. Place sensor in pipe between 9 o'clock and 3 o'clock to prevent moisture damaging the sensor.
2. Do not run engine without having the wideband on
3. Excessive Heat can kill the sensor. Before cat is far enough from turbo so that should be ok.

Alternatively, check that the gauge is showing what the software is showing. I know most gauges run from 0-5v and some wideband modules can often default depending on the output you have connected to. Example. If your wb02 device output 1 is 0-1 v and output 2 is 0-5v on default and for any reason it has defaulted the settings, the gauge connected to output 1 if using custom settings will show the incorrect afr if it is needs to be defined in the software as a 0-5v output to match the gauge.

How this makes sense.

I've upgraded my wideband to an ntk type as they are known to be more reliable and less prone to faults like these.

Gets annoying to constantly recalibrate.

O2 sensors are unfortunately a consumable item. Most neglect this fact and are happy to run a sensor that may potentially be providing bad readings.

 

[Parabola]

I had the exact symptoms with my aem wideband.
Replaced the wideband o2, problem went away.
The only thing I could see wrong with the sensor, was part of the heat shielding around the wires had been stripped off.

 

[mike_980]

Thanks for the replies. My wideband is an AEM UEGO gauge with the Bosch 4.9 LSU sensor. It is mounted in the exhaust at about 10 I clock to prevent moisture from killing the sensor. It was brand new when I fitted it about 6 months and 3000 miles ago and worked faultlessly until yesterday. I am inclined to believe the wideband at the moment until I get some more information I think. The only thing which I have doubts with about the sensor is after I installed it, I later realised that the top of the sensor and the wires coming out of it were pressed against the heat shield which could have caused damage to the wires - but since I sorted that out it has been fine for a couple of thousand miles.

I don't have any way of checking the fuel pressure other than measuring the flow to the fuel rail by measuring how much comes through in a fixed amount of time. It seems fine though. The fuel pump is the 4g93T GSR fuel pump and the ECU is standard. Are there any electrical tests I can do on the switching sensor in the turbo elbow? I was curious as to whether a resistance of 150Ohms between the harness connector (earth pin) and the battery earth was correct?

I have had to do some modifications and repairs to the engine harness so it is possible that I have made a mistake somewhere which has caused a problem to show after some time.

 

[mike_980]

Attached is a picture of a plug - they all look the same