I agree with Jason4x4, either the the sensor is shorted internally or it has a ground issue that is backfeeding battery voltage through the signal (input to PCM) wire. Try a new sensor, clean your O2 ground or go to a 4 wire sensor.
CK5
Register an account today to become a member! Once signed in, you'll be able to participate on this site by adding your own topics and posts, as well as connect with other members.
That damn TBI idle variance - FIXED
- Thread starter Cricket
- Start date
Perhaps your issue is that the heated O2 is doing it's job, and that when you disconnect the heater wire, the system simply goes open loop and THAT is what the problem is...something is wrong with closed loop and the engine only "sees" the problem when the sensor is heated and trying to keep the system in closed loop.
Open loop runs off of a bunch of pre-programmed/assumed values, many times if something else is going on with a sensor, vacuum leak, etc., taking most of the sensors out of the equation (again, open loop) the system runs better, or at least more consistent. I believe with the TBI setups, disconnecting the MAP sensor forces open loop, which is something you already did, and stated the engine idled better like that, correct?
Could do the same test by disconnecting (or depinning) the O2 feed wire to the ECM and leave the power wire hooked up, see if the results are the same. But again, I believe disconnecting the O2 sensor puts the setup into open loop just like disconnecting the MAP sensor (if not going into a limp mode) so you'd likely be doing the same test different ways. But at least it would prove the power wire is or is not the problem.
Open loop runs off of a bunch of pre-programmed/assumed values, many times if something else is going on with a sensor, vacuum leak, etc., taking most of the sensors out of the equation (again, open loop) the system runs better, or at least more consistent. I believe with the TBI setups, disconnecting the MAP sensor forces open loop, which is something you already did, and stated the engine idled better like that, correct?
Could do the same test by disconnecting (or depinning) the O2 feed wire to the ECM and leave the power wire hooked up, see if the results are the same. But again, I believe disconnecting the O2 sensor puts the setup into open loop just like disconnecting the MAP sensor (if not going into a limp mode) so you'd likely be doing the same test different ways. But at least it would prove the power wire is or is not the problem.
It's very unlikely the O2 would cool down enough this time of year to make it go into open loop, if it did the idle would go all over the place like it was prior to removing power then go into open loop and then retry closed loop and repeat again and again. My blazer uses a single wire sensor and this is the cycle it goes through when it's way cold outside, works fine as long as it's above 30 degrees outside though.
The simple way to tell is connect a scan tool or similar and watch what the O2 voltage does and the loop status is. My bet is with the heater wire connected the voltage goes 1 volt or higher and the computer tries cutting fuel to get the voltage down under 400 milivolts which causes the engine to run poorly because it's so lean.
Diconnecting the MAP will put it into limp mode which is different than disconnecting the O2 sensor. No O2 voltage will cause it stay in open loop and run off of the learned adaptives or the base program if the computer has been disconnected and the memory lost.
The simple way to tell is connect a scan tool or similar and watch what the O2 voltage does and the loop status is. My bet is with the heater wire connected the voltage goes 1 volt or higher and the computer tries cutting fuel to get the voltage down under 400 milivolts which causes the engine to run poorly because it's so lean.
Diconnecting the MAP will put it into limp mode which is different than disconnecting the O2 sensor. No O2 voltage will cause it stay in open loop and run off of the learned adaptives or the base program if the computer has been disconnected and the memory lost.
You are right, scanner watching loop status is the easy test, but you've got to have one handy to watch this.
Watching millivolts though, could be bad wiring, bad sensor, or the O2 actually reacting to a real problem further upstream.
Exhaust gas is what, 600*+? Obviously ambient temp variances can have an effect, you've seen it, but from what I've seen with MY setup, ambient temp has little to do with the sensor operation, while engine output changes instantaneously changes O2 operation.
As long as the sensor is hot enough, even a heated O2 sensor will work without voltage going into it, right? I can't see why it wouldn't, it creates voltage, the heater is there solely to get the sensor up to a temperature where it can actually "see" whats in the exhaust stream.
Watching millivolts though, could be bad wiring, bad sensor, or the O2 actually reacting to a real problem further upstream.
Exhaust gas is what, 600*+? Obviously ambient temp variances can have an effect, you've seen it, but from what I've seen with MY setup, ambient temp has little to do with the sensor operation, while engine output changes instantaneously changes O2 operation.
As long as the sensor is hot enough, even a heated O2 sensor will work without voltage going into it, right? I can't see why it wouldn't, it creates voltage, the heater is there solely to get the sensor up to a temperature where it can actually "see" whats in the exhaust stream.
With exhaust manifolds ambient temperature has no effect on sensor temperature. Long tube headers are another story entirely, and this is where I get in trouble with a non heated sensor when temps drop below 30 or so and add in some idle time and things get a little flaky.
I'm obviously just guessing/speculating on the sensor heater issue with a little information and symptoms.
Scanner or winaldl information will go a long way towards giving a concrete answer. So take whatever I type as a best guess suggestion and nothing more.
I'm obviously just guessing/speculating on the sensor heater issue with a little information and symptoms.
Scanner or winaldl information will go a long way towards giving a concrete answer. So take whatever I type as a best guess suggestion and nothing more.
It's all best guess of course, all sorts of variables to every single application/installation out there, without being there first hand, pretty hard to diagnose correctly. 
Things like checking open/closed loop with a scanner is easy, anyone can do it, and may or may not help in diagnosis, but certainly can't hurt!
Things like checking open/closed loop with a scanner is easy, anyone can do it, and may or may not help in diagnosis, but certainly can't hurt!
Thanks guys, there's a lot of good info being posted. I believe there's been a change in the sensor because I ran a heated 3 wire for at least a year without the weird idle in park/neutral. My belief is that I have an Air Fuel ratio spike/oscillation when it's plugged in and reaches closed loop. Closed loop is triggered at 600 degrees when the ECM begins reading 02 information. According to the Manual in open loop the ECM uses other sensors and a pre-programmed fueling table to maintain idle.
I think the fact that the idle gets weird at closed loop when 02 information comes into play says a lot about what's happening. For the time being I have left the heater wire unplugged so it defaults back to a single wire with an extra ground, or 2 wire. It was 40 degrees out this morning and it took around 3 minutes to hit closed loop where it ran smooth as can be. No adverse driving issues, hesitation, codes, et.
If anything changes I will post up. If I had my own WINALDL set-up I would take readings plugged and unplugged and post them. Unfortunately it will have to wait until after BB.
I think the fact that the idle gets weird at closed loop when 02 information comes into play says a lot about what's happening. For the time being I have left the heater wire unplugged so it defaults back to a single wire with an extra ground, or 2 wire. It was 40 degrees out this morning and it took around 3 minutes to hit closed loop where it ran smooth as can be. No adverse driving issues, hesitation, codes, et.
If anything changes I will post up. If I had my own WINALDL set-up I would take readings plugged and unplugged and post them. Unfortunately it will have to wait until after BB.
Did you ever figure this out because I'm having the EXACT same problem. I have Win ALDL but I dont know how to read it. If it would help I can post some datalog info?
I will try that but I just replaced my heated one with another one and the same thing is happening. I think I still have my one wire one laying around.
Ok so I did the same thing and disconnected the switch power to the 3 wire o2 and it made it idle better but once I start driving it around the same surging comes back. Its throwing up a Closed loop flag and a Old High Gear Flag on Winaldl. If I shut the truck off it will start back up and idle fine until I start driving it again (both flags will come on again). This is starting to kill me. I have checked everything I can think of plus all of the sensors are either new or have been tested or I have tried hooking up other sensors that I know are working properly off my other tbi engine. I have also hooked it up to my buddies scanner and cant figure anything out but it does run fine in limp mode.
