weird o2 sensor problems

[Project_Legacy]

first off, my car has basically an ej22t, but with an ej20g shortblock. i believe it runs a bit rich as there is less cylinder space in there. basically, for the last 8 or 9 or so months, i have been running this car with an unplugged o2 sensor.

i just got an o2 sensor (used but looks fine) and tried running it on the car.

for about 5 minutes the car drives good. after that, it starts to bog, and not want to go above 30mph unless i step on it. after about 10 mins or so of this, it starts smoking. the smoke is blue smoke, like oil.

i pull over, unplug the o2, and everything is fine again. has this happened to anyone? im thinkin the bog is from running rich. but the blue smoke is totally random. it never smokes blue, unless i get on it.

anyone have any ideas??
thanks in advance!

Don

[Matt Monson]

Blue smoke? There's oil in your system. If it was a rich running issue, it would be black, not blue...

[Project_Legacy]

yea, but the weird thing is, without the o2 sensor hooked up, there wouldnt be any blue smoke at all, unless i actually hit like 5psi or more. thats why i thought it was so weird. any other suggestions out there??

[my3awds]

Perhaps the PCV is allowing more oil then usual in your intake track? If there is oil in there, perhaps you can either run the pcv line to the ground or put a filter inline that goes to the intake track and see if that helps any? Just a suggestion.

I use to have a 16g on my old dsm that would spout out black smoke under boost because it was leaking oil into the housing. There was a noticable amount of shaftplay as well. You may want to check that as well.

[sammydafish]

bluish/black smoke can be just running rich.... sounds like the o2 you got is bad. Get a brand new one (they are cheap anyway). Resent the ECU and run it. I bet it will be all better after a few drive cycles and you'll get better fuel economy too.

[Project_Legacy]

you think so? i always thought blue meant burning oil. well, the o2 sensors i see online are like 80 dollars and up brand new... i cant afford one at the moment. i just bought another car, and am scrounging(sp?) to even pay off for the registration, title transer, and tax on it...


the car runs fine without the o2 sensor. any way the ecu learned to run without one? it was like that for probably ALMOST a year.


also, i am running stainless steel headers, the o2 location is right after the 4 to 1 part. could that also be a contributing factor?


there is oil in my intake tract, i know that. i believe my turbo, and maybe even the pcv valve, could be contributing to that. i only see the bluish smoke when i accelerate into boost (without the o2 sensor).

so i guess my best bet would be to get a new o2 sensor and see if that helps.

[Project_Legacy]

anything guys?

[vrg3]

The ECU can run fine without the sensor, but you will occasionally get trouble codes.

As you've found, used oxygen sensors are usually not worth it.

So your sensor is upstream of the turbo? I suppose it's possible that it could be overheating. It's also possible that the stainless steel header is providing an insufficient ground for the sensor. If that's it, you might be able to add a ground wire with a hose clamp around the sensor.

[Project_Legacy]

ooo. it uses the header itself as a ground huh? i dont know what the deal is, but the cheapest solution for me right now is to try the ground wire dealie.

so, usually the o2 sensor on the legacy turbo would be, on the downpipe? this could also be the main problem of it.

im going for emissions soon, maybe i can trim the maf signal with a potentiometer(sp?) so i dont run too rich. its ok to do that with a non working o2 sensor?

[Brian Mau]

Isn't the O2 sensor only operates in closed-loop mode (idling / light load)? Without the O2 sensor your car's probably already running at its richest possible. As I know the O2 sensor has an output of 0-1V in which 0.5V indicates a 14.7:1 stoichiometric AFR. Anything below that means the car's running too lean (exhaust gas isn't hot enough) and anything above that means running too rich (exhaust gas is too hot). If you've disconnected it = floating output to the ECU = 0V = indicating it's lean and the ECU will try to compensate = running rich. When you're accelerating (open-loop) the ECU basically will ignore the signal from the O2 sensor as it's always rich - unless on cars like Toyota Prius which has a wide-band O2 sensor and it's always in closed-loop mode, regardless if you're accelerating or not. The ECU will operate the injectors with the pre-determined fuel map according to the engine load (MAF/MAP sensors, TPS etc) and not the O2 sensors when in open-loop mode.

Hope that helps.

[Project_Legacy]

ahh. ic how that works now. yea im sure i am running pretty rich.

[vrg3]

Yes, 3-wire sensors like the stock sensor grounds through the body. And yes, the stock sensor is at the top of the downpipe:

http://www.graphics.cornell.edu/~v/pics ... xhaust.png

I don't know if I'd ever say it's okay to tweak the MAF signal with a
potentiometer.

When the sensor is disconnected, the ECU knows it (it applies a bias voltage that an operational sensor overcomes) and does not try to compute fuel trims with it.

But, yes, without the sensor you may well run a bit rich, depending on things like the state of your injectors, your MAF sensor, your fuel pressure regulator, et cetera. One of my cars tends to run a little lean without the sensor, as measured using a wideband.

Why are you sure you're running pretty rich? The "less cylinder space" concern isn't as big a concern as you might think because fueling is primarily determined by directly-measured airflow.

Have you reset the ECU since disconnecting the sensor?

Incidentally, I don't believe the Prius has a wideband oxygen sensor, but it does run closed-loop all the time. It just keeps a stoichiometric ratio even at full load.

[Project_Legacy]

well, i think it runs a little rich because i have the 2.0 bottom end on the stock 2.2t fueling. i thought it would make me run rich because wouldnt the ecu calculate the fuel based off of the car being a 2.2 liter?

i just wanted to use the potentiometer because someone suggested using it, when i go to take emissions, so i can guarantee a pass. or something of that sort... i have 3 inch catback, with a 3 inch high flow cat, the o2 sensor is disconnected, and i think it misses just a little bit after i drive it hard (i rarely do that anymore).

im just hoping to pass emissions is all...

[vrg3]

Like I said, it calculates fuel primarily based on airflow. 14.7 grams of air means roughly one gram of fuel, no matter how big the engine is.

If you want to try screwing with the MAF signal, look for a post I recently made about how you might approach that.