Diagnose me..

Wish they didn't put the battery in the boot, on a nearly 5m car I'm gonna need to bridge the alternator/voltmeter/battery gap with some jump leads :p

Sounds like switching plugs and sparks might be worth a shot then, I'll give that a whirl. Only thing that confuses me there is - why would going open circuit improve driveability if there were a coil or plug issue? And why would it even go open circuit, unless it decided it was no longer believing what the O2 sensor told it?

Cheers again for the help, I do appreciate it.
 
I would hazaard a guess as to the engine running a little rich under open loop, a weak coil would struggle more under closed loop as the mixture is generally leaner.
 
Sleeping on it.... I'm not 100% convinced that the original O2 sensor issue was fixed tbh. I'm suggesting a few tests here and there that you can perform that cover common issues rather than an in depth study so to speak.

It's hard to suggest the way to attack the problem without knowing someone's ability and the equipment they have to hand, not suggesting that you're not capable btw, but there are things which could potentially occur during testing that would seriously damage the car.

I suppose I could walk you through checking the integrity of the O2 sensor circuit, and then how to check the actual sensor response. This would require than you know which wire is the signal from the O2 sensor back to the ECU - you have a scan tool already.
 
MatteH Oxford said:
Sleeping on it.... I'm not 100% convinced that the original O2 sensor issue was fixed tbh. I'm suggesting a few tests here and there that you can perform that cover common issues rather than an in depth study so to speak.

It's hard to suggest the way to attack the problem without knowing someone's ability and the equipment they have to hand, not suggesting that you're not capable btw, but there are things which could potentially occur during testing that would seriously damage the car.

I suppose I could walk you through checking the integrity of the O2 sensor circuit, and then how to check the actual sensor response. This would require than you know which wire is the signal from the O2 sensor back to the ECU - you have a scan tool already.
Click to expand...

This was my suspicion also, that there's an unresolved O2 sensor issue (or something that caused the O2 sensor issue/vice versa). I'm currently struggling to see how the LTFT on bank 1 would be staying at 0.0 - the most logical thing to me is that it's not getting a signal from the O2 sensor to adjust the LTFT but that being the case why would the STFT be changing and why would the O2 sensor not be throwing a code.

In order to check the O2 sensor properly I'll obviously need to access the ECM which is buried beneath a load of stuff (mostly trim, but also wiper arms which need a special tool to remove). This is rather awkward and why I opted to change the O2 sensor in the first place!
 
You don't need to remove any trim to test the sensor or wiring if you have a scan tool, back probe(stick a paper-clip in to the back of the O2 wiring plug so it contacts the pin, being careful that you don't short it) the signal wire on the O2 sensor while looking at the sensor data on a scan tool.

First up is to test ECU response to voltage changes, this test will confirm that the wiring & the ECU is good - not the actual sensor, we'll test that later:

1. Ground the signal wire by placing a jumper wire connected to earth on the back probe pin, this should pull the signal to 0v.

2. If that's good, then lick you fingers and while holding the end of the jumper wire (still connected to your back probe pin) in one hand, touch your other hand on the battery positive - basically using you own body as a resistor so you don't send 12v back to the ecu. This should raise the O2 signal by a few volts.

O2 sensor response.

1. Introduce propane gas into the intake using a plumbers torch or similar. This should increase the voltage on the O2 and drop the ST fuel trim immediately as you've introduced unmetered fuel.

2. Create a vacuum leak after the MAF sensor, this will lower the signal and up the ST fuel trim but it's hard to get it to drop as low as it would be by actually grounding the signal wire (as in the previous test)

If you don't see a response on the scan tool to any of these tests then you will have to remove the trim so you can first check the wiring for continuity or shorts. If that's o.k. then the ECU is suspect but then are several tests before you condemn one i.e. check all the ECU powers, grounds and the complete 5v reference circuit(s).
 
Thanks, I'll look into doing this when I have a spare day, probably the weekend! Using my body as a resistor eh, bold move, might just check the voltage with the meter beforehand :p
Am I right in thinking the ECU shouldn't be expecting anything more than 1V back from the sensor?
 
miniyazz said:
Am I right in thinking the ECU shouldn't be expecting anything more than 1V back from the sensor?
Click to expand...

It should be higher than that as you have wideband O2 sensors? The precat sensors on my Alfa go up to at least 3.5v although I can't say how high they should go on an Jag. If they are wideband, normally higher than 1v.

Using your body as a resistor is safe enough, I doubt it will let more than 2-3v through.
 
OK, a little update. Looks like my scan tool doesn't actually give me the O2 sensor output voltage. Additionally, according to here the wideband O2 sensors report back via current rather than voltage - in any case, I've not been able to do the above :)
Instead I've pulled off trim (bending the cowl vent screen instead of removing it allowed me just enough access) and checked the connections to the ECM. No significant resistance between the O2 sensor side and the ECM side, no short to battery, no short to ground. Incidentally the O2 sensor harness on the right bank was displaying the same voltage on all pins as on the left bank harness but obviously not been able to compare with the O2 sensors both plugged in..

I've also swapped over all spark plugs and coils to the opposite side just in case there's an issue there. Some of the rubber on the right bank coils looked slightly discoloured compared to the left bank (see pic).

I'm planning on driving it again, see if there's any difference, then try introducing some butane gas. Is it safe to just put it in through the air filter? (Or take out the air filter and put it through there?) Or could it damage the MAF?



Cheers
 
It sounds as if your ECU does the O2 voltage calculation, a different method than my car.

In that case, one of the wires should supply a reference voltage, somewhere around 3-3.5v?. Backprobe that pin instead of the signal wire and by doing the short to earth/body-as-resistor test you'll change the reference voltage and hence check the ECU response that way. Oh, I think your car may also see higher voltage as lean and lower as rich, this is "back to front" from a older "switching" type o2 sensor - something to note when doing the vacuum leak & introducing propane into the system.

Re butane, I can't say if that would be safe? propane is basically LPG afaik
 
Just to update on this.

I ended up taking this to a specialist which I'm pretty sure was the right decision.

They found ongoing inconsistent readings when removing the oxygen sensors from the exhausts with intermittent resolution on O2 sensor replacement. Further investigation revealed a potentially (again, intermittently) faulty relay that was on for too long and burning out the O2 sensor heater elements.
In addition to this they ended up stripping back and reseating the connectors from the wiring loom to the ECM and currently (touch wood) the issue seems to be resolved..

So no air leaks were responsible!

Thanks for your help.
 
You must log in or register to reply here.
Back
Top Bottom