Ok, I finally got a DMM today, which I should've had ages ago. Boy did it expedite my troubleshooting My original posts on my problems are https://www.thirdgen.org/techbb2/sho...hreadid=115015 and https://www.thirdgen.org/techbb2/sho...readid=120631. To date, my observations are as follows:

1. I get SES light every 10-15 minutes when I'm cruising on a light highway (ie: the kinds with lights every so often). It stays on for approx 30-60 seconds and goes off. Retrieving codes yields a code 13 and nothing else.

2. When cruising at constant speed at light throttle (enough to hold RPMs), I get a bad lack of spark (the exhaust starts smelling REAL rich). Under acceleration (not necessarily WOT though) the problem is non-existent. This may or may not be related to 1, since sometimes the two happen at about the same time, and sometimes they don't.

3. I got WinALDL going, and did some datalogging. This is here:
https://www.thirdgen.org/techbb2/sho...readid=117066. Most things looked fine to me, except the ECM constantly sees .451v from the O2 sensor. Occasionally it changes +/- .002v, but it's pretty much the same all the time.

4. I bought a new O2 sensor, even though the one in there is only about 6 months old (~ 1200 miles or so). No change after replacement. I took the "old" one and did the torch test, and it acted as a O2 sensor should - started producing varying voltage at varying levels of heat. So I know it's not completely dead. I also had the casing grounded while doing this, and it was producing stuff in the .100-.900v range.

5. I tested the resitance from the tip of the ECM connector to the O2 harness, and yielded .02ohm. This doesn't seem particularly high to me, and infers proper continuity on the wire.

6. After driving about 45 minutes home and getting things plenty warm, I unhooked the O2 sensor, and took a measurement. I was getting varying voltage from -.040v to .020v, which seems really low. I was grounding the DMM to myself at the time because of lack of hands, but that was the same thing I did when I performed the torch test. I should be enough ground for a 1v circuit I'd assume. I then took a wire, dangled it onto the O2 casing and held it in my hand. The reading then jumped to ~1.2v, but constantly changing +/- .4v or so. I was sharing the ground with the DMM, which I'm not sure if that affects the readout at all.


Based on this, I have a couple suspicions. One is that my O2 sensor may not be grounding properly to my exhaust manifold. The other is that my ECM may be on it's way out the door (assuming my in-car O2 test was just wrong), since it's not seeing much at all from the sensor, and the circuit itself seems fine. Would it be a bad thing to hook the DMM up to the O2 lead at the ECM while it's running to measure the voltage that far up the line ? Or am I way off, and in need of further guidance ? Thanks a ton!


Kyle

 

The O2 sensor can be fooled by several things. A fouling plug, an air leak at the exhaust manifold allowing excess air into the system, and possibly a leaky intake manifold. With an intake problem sometimes it sucks oil and causes a plug or plugs to foul. Also a faulty EGR valve can cause a code 13. If it doesn't open and close properly it will allow exhaust gas into the F/A mixture causing an O2 code. I would inspect each plug for color. Keep in mind that the sensor looks for oxygen in the exhaust, and any way that you can get extra air into the system will cause it to malfunction. The ground strap is tough to get to. You can get some ground braid and add a ground in a convienient place to see if it helps. Grounding through your body is not the way to go. With a one volt swing as a sensor has, a good ground is more critical. Also the input resistance of the DVM is important. You must have at least 10 megohm input resistance or your readings will be inaccurate and you risk damage to the sensor. A normal swing is .1 volts(lean mix) to .9 volts(rich). At ideal 14.7:1 stochiometric ratio it should be about.45/.5 volts and fluctuating. Always use chassis ground for the DVM, you know how many ohms you are? At 30kv. you'll be a decent ground, at 1 it's an open circuit.

 

Well, I finally solved the problem. Looks like my engine <-> chassis ground wasn't hooked up. I can't seem to figure out exactly where it goes, but in the mean time (as a test) I extended it a bit and slipped it under the nearest bolt with good chassis contact. Anyone have an '91 LO3 who might know exactly where the sucker is supposed to bolt to the firewall ?



Kyle