13: O2 Sensor Circuit

[Fidhle007]

I keep getting an O2 sensor code that I can't figure out. I'm on my third O2 sensor and I can't find an issue with the wiring anywhere so far. When running the car with the scantool hooked up it gives me two voltage readings for the O2 sensor: one holds steady at -.15 and the other fluctuates between -.2 and -.9. The negatives lead me to believe I have something hooked up backwards in my wiring adaptor but the wiring diagrams I've found online (AKG, RM European) are conflicting and based on the E36 Bentley I can't find where the O2 sensor circuit even goes through the main body plug. The car is a '90 325is and the motor and harness are '95 M3. I mainly followed RM's diagram but it basically seemed like I was matching like color wires once I got down to it. Any help or advice would be appreciated, I'm pretty sure I'm using a lot more fuel than I need to be at this point!

[equate975]

You have something hooked up backwards, check it with a multimeter. It should ideally fluctuate between .1-1v. Not "-" voltage. .2-.9 is close enough, as long as its changing really fast.

Just bypass all the wiring and hook it right into the ECU like it should just to see if the code goes away.

[Fidhle007]

Does it run through the c101/x20 plug at all though?

[Fidhle007]

AKG lists Pin 6 (Oxygen sensor, heater relay, e30 side) as going to Pin 13 (Oxygen sensor, fuel pump relay, E36 side) but RM lists Pin 6 as Diagnostic Connector on the e30 side and has it going to Pin 22 on the e36 side. (According to AKG, pin 22 is the reverse light switch?) I went with the latter option as the wire colors matched and I was eager to get the swap done but I guess that could be the route of my issue?

From what I can see, it looks like the E30 has two separate wires for the o2 sensor (heater relay maybe?) and the fuel pump relay but the E36 runs two things on one wire?

For reference, here's the charts I'm using:
http://www.rmeuropean.com/Articles.asp?ID=135
http://www.akgmotorsport.com/e30convpins.html

Thanks!
Brendan

[equate975]

I'm looking at the e30 layout and I don't see anything about it going through the c101. But my diagrams suck.

There should be 4 wires for the o2 sensor (s5x anyways) 2 heater wires, ground, and signal.

It looks like the m20 uses a 3 wire one? I haven't looked at one in a long time but thats what the diagram showed.

Regardless..., if you are getting power to the o2 then the heater relay is working so thats hooked up right. My guess is the power and ground is just backwards. But I can't fucking tell where they go through lol.


Edit:

Ah I see, thats just for the relay. I knew it would be weird to have the signal go through the c101. Anyways, if you have power to the o2 then thats not the issue.

Do this, take a volt meter and stick the ground on a good ground somewhere. Then back probe 13 and then 40 on the ecu and see what you get.

I think 40 should be (*edit*ground) and 13 should fluctuate for the o2.

[Fidhle007]

You mean at the ECU plug? Or just the wire, while the engine is running?

[equate975]

You mean at the ECU plug? Or just the wire, while the engine is running?

Pull the black plastic cover off the back of the connector for the ECU. Find pins 13 and 40.

Start the engine up then jam the red wire from your multimeter into either 13 or 40 from behind, where the wire goes in. Stick the black one on a good ground. Repeat for the pin you didn't do.

Don't unplug anything, and make sure the engine is warmed up so the o2 is closed loop.

If you can't fit the end of the multimeter in, jam a paper clip in there. Or buy some backprobe tools ;)

http://us.fluke.com/usen/products/AccessoryDetail.htm?cs_id=34639(FlukeProducts)&catalog_name=FlukeUnitedStates

Paper clip does the same job though.

[Fidhle007]

Wouldn't that have the same effect as the O2 voltage readings I get from my scantool? I have a snapon (not sure of the model) at the shop and that's where I'm getting the two O2 voltage readings from, both of which seem to be kooky.

I've been pouring over this among many other threads but still can't find anything solid, unless my car has also fallen victim to the mysterious fuel pump backfeeding issue Jason89i talked about in the aforementioned thread...

[equate975]

If the ground shows that it has voltage then we know the two are mixed up. If it shows up as a normal ground then we know that the ground is ok and the signal is mixed up elsewhere.

Edit:

Also sometimes snapon and the OTC Genesis I use on my jetta is wonky. Unless you know exactly which data stream to look at it will show me erroneous information about sensors that is just wrong. For instance, the OTC one I last used showed o2 sensor voltages 4 times. 3 of the 4 was just random information, one was correct.

Mainly I am trying to rule out (or in) the o2 ground. Its probably going to be those are backwards. If it was crossed with a heater line you would just have +12v all the time and your ecu would probably be dead. Since its jumping the appropriate voltages it has to just be backwards. Why? No idea, just gonna have to trace the wires back and see what got fucked up.

Easiest way just hook test for continuity between 13 and the signal wire on the o2 sensor.

Edit: 2
1. I need to read everything before I post.
2. I doubt its feed back, if it was I would look for something going over the 1v range or just acting weird. Like I said, yours moves like it should, its just "-" instead of "+"

[Fidhle007]

So after edit 2, where does that leave us? :-)

It was fine in the M3, and as far as I can tell the only thing that's changed on the wiring where my adapter connects to the E30. I guess it could be that backfeed that's throwing the code then?

[equate975]

Oh shit, dyslexia strikes again... I thought it was feedback, not backfeed lol....

Yeah I would do what jason said, see how that works. Looks like a "common" enough problem. Otherwise the only other idea I have is your + and - are backwards, which would cause the same thing.

[Fidhle007]

Ok, so I bypassed fuse 11 as per the thread I posted above but I still get the strange negative readings. Hmmm...

[Fidhle007]

Well, I was unable to get a tester on the pins at the ECU, without the plastic cover it kept unplugging itself and killing the car and throwing all kinds of crazy codes. I did try punching it into the code reader as a '94 325 just for kicks and it only showed me one O2 sensor voltage which stayed flat at -.15 if that tells anyone anything...?

[equate975]

What ecu are you running? As far as I was lead to believe the m50/s50 used one o2 sensor. But when I look at the pinouts I see theres supposed to be 2. I'm using a silver 413 and there is only provision for 1 o2 on the harness. This might explain why one o2 sensor is -.15 because its technically not connected.

I would say just to rule the wiring out, run a jumper wire from each pin on the ecu for the o2 and check for continuity on the corresponding wire on the o2. Just to make sure everything is in place as it should be before going any further.

[So Live]

I would do the test with the voltmeter instead of the scan tool just to be sure. One person said the tool sometimes acts a little funny.