o2 sensor wont cycle

[apedog]

My o2 sensor wont cycle when the car is starting to warm up. However when I start it completely cold it wont cycle (which is normal) and then cycle for a about a minute then completely flatline at zero. The car is running REALLY lean when that happens. The log shows when the car was kinda warmed up and me pressing the gas to see if it will cycle. The o2 is a bosch and its a few thousand miles old

 

[gofer]

I can't look at your log because I'm at work but I thought I'd ask, why do you have a narrowband o2 sensor when you have an AEM UEGO wideband simulating it?

I just read through your post about it here ---> http://www.dsmtuners.com/forums/tuning-engine-management/423359-aem-datalogging-wire-where-does-go.html

There could be more to it and it might even answer my question after looking in the log, so I'll apologize now if its easily answered.

 

[snowborder714]

I'm not sure how he successfully got the narrowband sim working considering he doesn't have v3 (as people were instructing him to do in the other thread). If you did have narrowband sim, the signal would have to come straight from the AEM unit (not sure if this is possible).

 

[apedog]

okay I forgot to mention that I dont have the wideband hooked up to the o2 wire anymore. My wideband is hooked up to the egr wire so I cant simulate my wideband to cycle as a narrowband. I am running DSMlink v2 through v3

 

[snowborder714]

So do you have a factory o2 sensor in then? If so, make sure the signal wire is hooked up well (after splicing the wideband in). If it is, then the sensor might be bad if it's not cycling.

 

[apedog]

So do you have a factory o2 sensor in then? If so, make sure the signal wire is hooked up well (after splicing the wideband in). If it is, then the sensor might be bad if it's not cycling.

Well a bosch o2 sensor is what I have. I dont wanna hook up the wideband to it though because my tuner needs it to be on the egr wire so the wideband is accurate in link.

 

[gofer]

You're understanding Brian wrong. If you previously CUT the front o2 sensor wire, white wire at pin 4 of your ECU, then spliced it back together to run the WB off the EGR then you need to make sure its making a good connection at the ECU.

 

[apedog]

Yep it has a good connection. Like I said the o2 will cycle for like a minute and then drop off and it will go into open loop and run really lean for some reason. I have no idea why it would do that

 

[snowborder714]

Your log shows it's constantly in open loop and never goes to closed loop, so it's never looking for the front o2 feedback.

I'm not sure if simulating the idle switch from the TPS would cause this, although as a 1g owner, you shouldn't need to do this with the factory idle switch being external from the TPS.

Here are items that can cause the car to be in open loop (only one has to be true).

• Throttle position too high (varies by RPM)
• Airflow too low (units are tricky, but it's very low)
• Coolant temp < 87F
• The O2 sensor did not switch around 0.5v for over 20 seconds while running in closed loop mode. In that case, the ECU locks itself in open loop mode.

For the ECU to go into learn mode (closed loop), all of the following have to be true.

• Coolant temp >= 190F
• Intake air temp < 123F
• Baro >= 22.9 inHg
• Baro <= 31.6 inHg

 

[apedog]

Could a bad ecu be the problem?? I took it apart and found there is some corrosion on some of board.
Which does make sense because if I were to put my other ecu in it will run perfectly fine, just rich because it dosent have dsmlink

 

[gofer]

Absolutely not.

Your front o2 sensor is cycling, then it goes straight to 0.00v unless you give it gas, which means its going EXTREMELY lean. Its a 0-1v sensor, 0v is lean - .5v is stoich - 1v is rich. This shows signs of a really bad vacuum leak, when was the last time you did a good BLT?

 

[apedog]

Absolutely not.

Your front o2 sensor is cycling, then it goes straight to 0.00v unless you give it gas, which means its going EXTREMELY lean. Its a 0-1v sensor, 0v is lean - .5v is stoich - 1v is rich. This shows signs of a really bad vacuum leak, when was the last time you did a good BLT?

I havent done a good BLT in about a year now. I know for a fact I have a little one at my exhaust manifold because theres a broken off bolt but thats been like that before I bought this ecu with dsmlink and it ran fine.
I guess it wouldnt hurt to do the test anyway maybe im missing something

 

[gofer]

Well, first of a a BLT doesn't check for the exhaust manifold or exhaust for boost leaks because those are exhaust leaks. However, an exhaust leak BEFORE the o2 sensor like you described on the exhaust manifold would do show the same result in ECMLink, a lean o2 sensor reading.

Fix the exhaust manifold leak first, then give it a good BLT and odds are you won't have this problem.

 

[apedog]

Well after all this time I found out that I had a 90 ecu and are in fact different from 91-94 ecu's. All's I had to do was switch pin 6 and 14 and now im good to go.