What's new
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.

Heated O2 install ... engine mising like crazy, what did I do wrong?

Jessie James

Jessie James

1/2 ton status
Joined
Oct 16, 2009
Posts
2,514
Reaction score
8
Location
VA
So I have a 91 K5, TBI, CA truck that has the emissions pump removed, and headers installed. The O2 sensor was not upgraded with the headers, so it always ran way too rich.

I bought a heated O2 sensor (can't remember which one, but it was the same as someone else here bought) and installed it carefully. No silicon got on the sensor. I wired up to 12v+ on the fuel pump relay. I used an OEM style quick-disconnect connector, and a wire tap to connect the O2 sensor signal wire to the harness sensor wire (not soldered).

When I first started my truck, it immediately started up, unlike previously where it would take 3-5 seconds, so that was good. It idles absolutely perfectly. And for the first 5 minutes of driving, it's absolutely perfect.

However ... once it gets warmed up the engine starts missing like crazy. It has to be missing on 4 or more cylinders. I am pretty sure it's running too lean because the smell coming out of the exhaust is a scorched burning smell.

What did I do wrong? Is there a specific heated O2 sensor I should be using? Have to reset the ECM?

Any help would be greatly appreciated.
 
When your engine is cold, it is in "open-loop" operation, meaning that the O2 sensor has no control over how the engine runs. Once the engine has warmed up and the O2 sensor is hot enough to operate (meaning the computer reads varying voltage off the sensor) then the system goes into "closed loop", in which it adjusts based off input from the O2 sensor. Since your engine runs fine when cold, it would appear you have an issue with your closed loop operation.

Did you change anything other than adding the O2 sensor?

Did you attempt to pull the codes? If the sensor is reading lean it should set a code 44

I used this sensor because that's what my Howell kit called for
http://howellefi.com/misc.-components/oxygen-sensor-heated-4-wire-gm-product/

If the O2 sensor was the only thing you changed, I would bet that there is either a communication problem with the ECU due to wiring, or a compatibility problem. There are ways to test the sensor to ensure it's functioning properly. If the engine is running that bad though, there definitely should be some codes to narrow down the issue.
 
I made no other changes, only the O2 sensor.

I hadn't thought to pull codes, thanks. I'll check that and update you.

I'm pretty sure the wiring is solid.

My new O2 sensor is a 3-wire unit. How do I wire a 4-wire unit if I were to update that? My original sensor was only 1-wire.
 
The AFS-74 is what is normally recommended to be used. I suspect much of the part number differences have to do with wire length, or connector style, however there might be some difference in how they operate. The overall voltage range will be no different however. I just stuck with the -74 because people way smarter than me say run it.

It's too bad Mark is no longer with us, he probably knew for certain which, but many of the truck applications are open loop idle. Heated or not shouldn't make a difference at that point. If you have a problem at idle, something else might be going on. Startup doesn't depend on the O2 sensor either.
 
You have to be careful these days. Time marches on, and oxygen sensor tech does too.
I have seen several articles about new sensors that are completely different from what was used before. For instance, there are some out there now that don't need an outside reference any more.

Also almost all of then we deal with output a voltage according to the oxygen levels. But there are some that require that the ecm put out a voltage, and change resistance according to the levels.

I would double check and make sure its the same number as what has been used here before. And, most important, make sure that the wiring is correct.
Don't want the power lead bleeding into the sensor output.
 
Agree with you completely. A few years back, the Bosch O2 sensors that were supposed to be direct replacements for the Delco's were found to not work the same as the Delco units, thus could cause issues.

The AFS-74 has been proven to work with the TBI/TPI units for years, and I would strongly suggest using nothing but that one due to potentially different construction of other units.

Edit: Good timing lol.

I'd concur with double checking the wiring first things first. But I'd keep my mind open, the other problems you describe do not seem to be O2 related.
 
I broke down 4 days in a row on the exact same stretch of highway next to the exact same Advanced due to the exact same coil being replaced 4 times, I finally limped to the junkyard and bought a coil, and its been fine for a year. Sadness. But look at the bright side, now we are giving millions of jobs away manufacturing these parts to 3rd world countries, all to save 2$ and replace it 14 times. Now that's using your dipstick America.
 
Ok, pulled codes and got nada (just code 12).

I guess the next step is to replace the O2 sensor?
 
Did the headers already have a provision for the O2 sensor or was it added and is it slightly above horizontal?
 
They were on the truck when I got it, so all I can tell you is there is a O2 sensor hole. It is nearly vertical, facing towards the rear of the truck, and is probably about 1' (?) away from the nearest cylinder on the head.
 
Update:

I just replaced the O2 sensor with a brand new Delco sensor, checked the wiring, and took it for a spin. The exact same issues are occurring, with the truck missing like crazy once it warms up. The engine is running very rich.

Here's where the O2 sensor is mounted, drivers side, at the collector for the rear two cylinders:

http://i.imgur.com/q0mMsEM.jpg

The headers are (I believe) Doug Throley 385Y's, and are no longer made.

I have confirmed the O2 sensor is getting 12V when the engine is running (it's connected to the fuel pump relay).

This is a CA truck, but someone removed the air pump and connectors to the headers, so I don't know if that makes a difference.

Is there any way to read the voltage being sent from the O2 sensor to the ECM? Can I read it from the ODB1 port? Tap into a wire at the ECM? Something smarter?

Any ideas would be appreciated!
 
Last edited:
You need an ALDL cable and Tunerpro RT to read/record the sensor data.

That O2 is only reading two cylinders, you sure both of those cylinders are operating correctly?
 
that O2 should be down at the main collector, where all 4 meet.. that's the purpose in having a heated O2 in headers, so you can run them down at the collector...
 
ryoken said:
that O2 should be down at the main collector, where all 4 meet.. that's the purpose in having a heated O2 in headers, so you can run them down at the collector...
Click to expand...
Doh, that makes complete sense! The original O2 sensor is mounted in the collector 9as I recall), and mine is not. That explains the bad readings and running rich. A quick Google search shows everyone else has their sensor in the collectors.

I'll take it to a shop and have them relocate this. Thanks!
 
That isn't likely to fix the problem.

It doesn't matter if the O2 sensor is at the exhaust port or at the end of the tailpipe, as long as it's hot enough to operate.

If one cylinder is running rich or lean, as long as the O2 sensor is in the exhaust stream of THAT cylinder, it will see the issue. 4 cylinders, 1 cylinder, doesn't matter. If all cylinders are burning the same, then the *volume* of exhaust gas matters not. If you have problems on one or both of those cylinders, then the whole engine will run bad because fueling is off based on the "input" the O2 provides the ECM from those two cylinders.

It's unlikely, however if the engine has problems on other cylinders and those two happen to be running good, you'd not see the issue with the other cylinders. On even a standard TBI (or TPI for that matter) setup, the ECM only sees one cylinder banks output...if a cylinder on the other side is bad, O2 would give no indication.

I'd be more inclined to find out if both of those cylinders are getting spark and have compression.
 
dyeager535 said:
It doesn't matter if the O2 sensor is at the exhaust port or at the end of the tailpipe, as long as it's hot enough to operate.

.
Click to expand...



bull.... relying on 1 cylinders stoich, as opposed to the average of 8, or 4, would be foolish....
 
ryoken said:
bull.... relying on 1 cylinders stoich, as opposed to the average of 8, or 4, would be foolish....
Click to expand...

I didn't say it was the best way to do things, I said it isn't likely to fix the problem.

1 cylinder or 4, if one isn't combusting right, as long is it's in *that* exhaust stream, the O2 sensor is going to see it.

How is the engine going to run poorly if the O2 was reading one cylinder, and that cylinder was operating correctly? O2 sensor isn't measuring volume.
 
You must log in or register to reply here.

Latest Posts

Top Bottom