Okay. ZZ4TPI MAP SLP 1 5/8 headers, auto tranny.
My problem: The O2 sensor will not function properly to allow the car to go into closed loop. I just put in brand new dead soft aluminum gaskets. Headers are tight as can be. New HEATED 3 wire O2 sensor. I had my Y-pipe cut and rewelded to fix a slight misalignment of the collector gasket to y-pipe connection on the driver (O2 sensor) side. I really felt I had fresh air invading the header tubes, throwing off the O2 sensor.
Some history, the car is doing this so I replace the copper Mr Gasket header gasket. Obvious leaks. Works for about a week then it does the same thing. So I then pull the bolts I had tapped down my old air tube holes and have them welded, works great for another couple days. Starts doing it again. I find faulty welds on two of the holes, get them re-welded, works great for another day or two then the same thing. No closed loop, 540mv max on the O2 sensor. So I then get the Y-pipe re-welded to fix the collector gasket. Same thing. Then as Yoda seez, Unlearn what you have learned. The best I can get is like 640mv on the O2 sensor when the car will go into closed loopw. Now it's back to it's normal, 540mv and the O2 sensor reporting "not ready". My IAC is at 0 counts at idle but I cannot find any vacuum leak (used some good starter fluid to screw up the idle, nothing).
I am seriously at a loss here. Is there anything that could be fooling the computer into learning stupid stuff? Any thoughts appreciated.

 

your problem lies elsewhere besides exaust leaks. have you tried another o2 sensor? maybe yours is faulty or you possibly the 3 wire is hooked up wrong?

 

I'm pretty sure the O2 sensor is function properly. One hot, one ground, one to the computer. The O2 sensor appears to be doing what it is supposed to, reads 408mv key on but engine not running, climbs as the car gets warmer, getting cross counts on the scanner etc.
But I agree, something else is affecting the exhaust gas, inturn affecting the 02 sensor.
I may pull the single wire out of my 92 Z28 and wire that into the car temporarily and see what readings that one gives.

 

ttt

 

Gee, Rob, I hate weird problems like that. Actually, I like them when they belong to someone else! <grin>

You seem to know which directions to look in and have tried *most* of the obvious with the exhaust fixes and stuff. Right now I don't have any good ideas but I've read your message about 4 times. I guess trying another o2 sensor would at least eliminate that as a factor.

What confuses me most is how it ran ok for a couple of day, then screwed up again. Just some question though. . .

With that combo, what PROM are you using. Store bought, custom from tuner, one you did you self?

You say it's a speed density system; what's your vacuum at idle?
(wondering because if your PROM calibration is close you should be able to get it to idle with about 20-30 IAC counts.)

Have you checked the MAP sensor too?

Do you have another ECM to try out? 40 buck, salvage yard.

Have you checked the fuel pressure?

I did have a problem similar to that once and found the grounds loose on the rear of the passenger head as I was adding a heated o2 sensor to try to solve the problem. (Don't ask how they got loose - brain fart in the middle of rewiring my MSD box and forgot to retighten the bolt back there) Now you don't have to ask.

I'll be thinking about it though and let us know if you fix it!

Good luck!

BTW, although I've read how to set the IAC valve several times. I've had best luck with just adjusting the screw while idleing, in gear, and reading the scanner. I adjust the screw until the counts are around 30. I have no idle hunting problems this way and never did seem to get it right the 450 RPM with it unhooked and all that jazz.

 

Oddly enough tonight I also wanted to check the collector gasket bolts after the exhaust work. I found the driver side a little loose. I got a few turns out of the two bolts. When I checked the scanner, the O2mv actually DROPPED from around 515 to low 400mv! The car is freakin possessed I tell ya!:lala:

 

Well I checked the O2 sensor. I pulled the one from my 92 Z28 and it hung around 380-400mv at idle. So no difference (unless I hooked it up wrong which I don't think I did).
I checked vacuum at idle 17 in/hg at idle. I believe that is also fine.
Oh, O2 cross counts like to stay frozen and not count. This causes the Rich/Lean to stay "rich".
Any suggestions now?
Thanks

 

Rob,

Promise not to crucify me if I'm mistaken, but since the 3-wire O² sensors were only out for a couple of years, I don't have a lot of exposure to those systems.

IIRC, the 3-wire HO2S has a power wire for the heater element, a ground for the heater element, and a signal wire (same as a single-wire unit). The "ground" for the heater is not necessarily a ground for the sensor signal or sensor case.

A FOUR-wire HO2s actually has a case ground wire for the sensor signal, but not a three-wire:



Given all the gasketting issues and fitment problems you have experienced with your headers, it is possible that the headers themselves don't have a solid ground connection. Have you tried wrapping a bare copper wire around the sensor shell or O² bung and grounding it to the engine case?

Just a thought.

 

And just to test this theory, you could meter resistance from the HO2S leads to ground - both at the sensor shell and the engine ground.

The other thing that comes to mind is poor ECM grounding, but that would probably cause other issues.

 

Vader,
If I leave the key "on" but the engine not running, the heated O2 sensor physically gets "hot" to the touch. Would that prove the heated portion is working?
Tks

 

something that may have been overlooked but did you use any non sensor safe silicone anywhere? that stuff screws up O2 readings in a hurry and you end up having to replace the sensor. and it dosent seem to matter where you used it, It seems to pick up portions of silicone or whatever odors it lets off. When I worked at the parts store this guy rebuilt his engine using that stuff and couldnt ever get the computer to read right, had to tear the motor apart completely to clean it all out and start over, needless to say he wasnt too happy.

 

Monte,
Last time I used anything was when I added my Edelbrock base, many moons ago. Like I mentioned I tried my 92 Z28's O2sensor last night and no difference. I put it back in my 92 Z28 and it worked fine. So I can't think of any sealers I used. I was aware of this problem and ask the parts counter guy or read the label for O2 sensor safe.
Update:
I drove it to Key West again today. What I notice happens is at first it hangs around 500-510mv then as the car get warmer and warmer the mv's start to climb to just below 600mv. The car runs really crappy here, at idle, injector PW climbs to like 3.1. Today after the ride home the car dropped into closed loop as I headed down my road at 35-40mph. It climbed to the 570-590 mv threshold and ran really rough, then I hit the magic 600 +mv and closed loop hit. However I was not in Learn Mode long enough to go from the regular 128blm to the 108 the car loves. I only made it to 118 blm before I pulled in my driveway. After sitting at idle the car dropped out of closed loop mode, but will stay at a blm of 118 until I shut it off.

 

ttt

 

Rob,

I understand that the heater element is working, but am questioning whether the heater element ground connection is actually the case ground for the sensor. If this were not the case, the sensor may not have a good ground reference, thus the question about resistance between the sensor shell and a known good engine ground.

 

Ahhhh, I get what you are talking about. I will take the ground from the back of the driver side head and mount it directly to the header, should provide the same effect.
tks.

 

I pulled out the service manual and here is what it says about Code 45 (which I don't get) but is for a rich exhaust indicated:

HEI shielding. An open ground CKT 453 (distibutor ground , reference Low) may result in electromagnetic interference or induced electrical noise. The ECM looks at this noise as reference pulees. The additional pulses result in a higher than actual engine speed signal. The ECM then delivers too much fuel, causing system to go rich. Engine Tach will also show higher than actual engine speed, which can help in diagnosing problem.

My thoughts and question #1 - My tach showed several hundred RPM higher than my scanner. And now the tach finally died. How do I tell if I have bad HEI shielding?

Canister Purge. Check for fuel saturation. If full of fuel, check conister control and hoses.

Question #2 - How do I check the canister for saturation? Shake it and listen for a slosh? Turn upside down and see if gas comes out?

Map Sensor. An output that causes the ECM to sense a lower than normal vacuum can cause the system to go rich. Disconnecting the MAP will allow the ecm to set a fixed value for the sensor. Substitue MAP if rich condition goes while sensor is disconnected.
No question on this, I'll try this tomorrow though I doubt the MAP is the problem. It's new and seems to be functioning properly but I'm not ruling anything out.

TPS. An intermittent TPS out will cause the system to go rich. My TPS seems fine. Linear and not intermittent.

EGR. An egr staying open, especially at idle will cause the o2 sensor to indicate a rich exhaust. However we have EGR off in the PROM and I have a block off plate over the egr port.

Now with that being said perhaps the car is seeing a lean condition and dumping fuel (which I think is happening). This is what the book says about Code 44 (which I also don't get)

MAP. A map sensor output that causes the ECM to sense a higher than normal vacuum will cause the system to go lean.
Probably not this since my system is running rich, but the ECM could be compensating with move PW to the injectors.

Exhaust Leaks. If there is an exhaust leak avove the O2 sensor, outside air can be pulled into the exhaust stream and flow past the sensor causing a lean condition. Vacuum or crankcase leaks can cause a lean condition.
My original theory. I am gonna pull the driver side header and check the gaskets for leaks. I checked out the Felpro exhaust collector gasket, same as the one that I got from NAPA. Others are using the Felpro with no problems.

Also the book shows that the engine ground comes from wire 413 Tan/Wht to GE15. When I did Bernards 12volt check directly to the sensor it read close to 1000mv and with it open it dropped very low. That should prove a good ground.