Running poorly/misfiring, code 33

[mecanicus]

according to the second post of this thread:http://www.gearhead-efi.com/Fuel-Inj...6197427+wiring
i would check the ground connection of the lambda heater circuit.As well it should be possible to attach a four wire lambda probe by connecting the sensor curcuit ground to the ecu ground.I would prefere this.

[JeepsAndGuns]

Sorry for the late response. Weather has not been on my side. Seems like it has been raining here every weekend, the only time I have to work on it.

I checked for proper power and ground to the sensor and it was good. I checked the wire between it and the ecu and it was good.
I got the new sensor. While it the connector looked like the same weatherpack connector, it would not plug in. Turns out the little tabs around the terminals are slightly different. So no idea what the o2 sensor was from that I was running.
So I just simply depinned the old connector and used it on the new sensor. The pinout was the same.
Hooked it up and tested it.
It fired up and ran fine, but the AFR was still not adjusting. I hook up the computer and the value for the o2 sensor was still stuck at 451.
I pulled the memcal, read it, compared it and everything was right. I double checked the signal wire and it was still testing good.
In some frustration, I grabbed my spare ecu and plugged it up, and it now works! O2 sensor voltage was now swinging and the AFR was now adjusting.
I took it for a drive and it ran great, just like it always has. AFR was normal and it seemed like everything was back to normal.
So my guess was a bad pickup coil originally, then a mismatch of ignition coil and pickup coil when I tried swapping in the junkyard parts. Then the o2 sensor issue, who knows?

So I unbolt the old ecu from the dash and install the spare one. I park it for the day, thinking all was well again.

I go out the next day and me and the misses are going to go out for lunch. I decide to take the Jeep since it was finally fixed.
I get it in, it fires up just fine, but then sounds a little off. As soon as I start driving it, it is back to the original issue of missfiring like crazy, not on any 1 cylinder, and feeling like it is only hitting on half the cylinders.

I feel like I have just made a 5 month long complete circle right back to where I started. I just want to bang my head against the wall. I feel like I complete failure of a mechanic.

[steveo]

oh man we have all been there. it took me 6 months one time to fix a car where the brakes didn't work worth a crap, and it ran super lean
after replacing every single brake component, the maf and map sensor, and resealing the intake manifold chasing a phantom vacuum leak, i found a hole in the *(#$t brake booster
why don't you take a break and work on the WJ? i have one of those man it's my favorite side project

[MO LS Noobie]

I also feel your pain, it is frustrating. I also agree on stepping back from and playing with a fun project to clear your mind.

Then I would start over as if it were a new problem. Since it ran good with this combination for a little while I would first check the feedback side of the system, the oxygen sensor. Try unplugging it, doing a hard reset on the ECM, then start it up and see what happens. This should put it in open loop and if your base programming and wiring is mostly correct it should run pretty good. If not you are chasing a ghost can be difficult to find. It runs good you are looking at something in the feedback either the wiring or your new oxygen sensors.

Remove the fan belt, hard reset, then retest. This will eliminate alternator AC voltage and, believe it or not, problems with static electricity building up the belt itself (I have only seen this issue one time in 40 years being a professional mechanic).

You could have a bad sensor failing when hot, ie. CTS, TPS, or even PCM. Test, don't just replace. I have seen a lot of new parts, counterfeit factory parts, and even new factory parts are bad right out of the box. Remember "NEW" means Never Ever Worked.

If it still runs poorly, I would then look at the wiring, doing a load test on each individual wire, since it seemed to run okay until you mounted the PCM. Wires can pass an ohmmeter test and still not be able to carry the amps necessary to do their job. I use a regular halogen headlights bulb (5 amps) that I have wires attached to it to check small circuits up to 16 gauge wire.. DISCONNECT PCM. Disconnect both ends of the circuit you are working on, this includes every connection that branches off of the circuit. One lead goes to negative battery terminal and the other goes to one end of the circuit you are testing. You then take the other end of the wire and touch it to the positive battery terminal. It should light up the headlight bulb brightly. Repeat on the other branches of the same circuit.

If any circuit fails the above test, then you're going to have to inspect the harness going as far as stripping off the loom and tape and checking each individual wire physically, even tugging on the wire, I have seen the conductor broken inside the insulation.

Good luck, Jeff

[dave w]

I'm thinking that swapping the ECM is a clue to what is happening.

Something electrical is back-feeding the ECM to cause the problem? Possibly a voltage spike?

I'm always suspect of old ECM / Engine wiring, because the copper will corrode inside the insulation.

Possibly try installing another ECM / Engine harness. Send me a PM, I have a loaner harness (deposit required).

[In-Tech]

^^^^^ This!
Just moving an old harness can "break" it :( So many hot/cold cycles and it will live a very long time, till it moves.

[steveo]

with a simple homemade load bank (volt meter to measure drop at one end of the wire, and apply a few amps of load to it on a button) you can suss out an entire disconnected engine harness in an hour max.

i like a load bank with like a 1 amp, 5 amp, and 10 amp load each with its own button. big stuff like injector supply gets a 5 amp load, little stuff like a maf sensor signal gets a 1 amp load.

if a thin old GM sensor wire is even remotely compromised (like even if it's wiggled itself back into place) a few amps usually cooks the damn thing and there ya go.

i use paper clips as my probes, seem to fit inside the female ecm pins fairly well.

[steveo]

oh another useful addition to your kit is a megger (megaohmmeter) they are fairly cheap on amazon. a gazillion volts will identify compromised insulation instantly. hell if you have the lights off you can see the spark.

i know it seems like a lot of work but in reality you're only testing a few dozen wires

if you own this old junk you really need to be set up to test your wiring all the time

[In-Tech]

Good call steveo :)

[JeepsAndGuns]

Life and lots of crappy weather have been giving me fits.
It seems like it has been raining every weekend, then we had that big cold snap.
Then when I had a sunny weekend, I had so much other stuff going on, the jeep had to take the back burner.

I did have a little time a couple weeks ago to start it and let it run. It had been setting since my last post.
It started up ok, but I could hear the random missfire thing it does (sounds like it is not running on all cylinders, but the miss can not be traced to any specific cylinder).
I just let it set there and idle for 10-15 min while I was starting up a bunch of other equipment to let them run (lawn mower, JD gator, chain saws, weed eaters, leaf blowers, etc...)
After running all the other equipment for a bit and putting them all back up, I decide to drive the jeep.
It ran perfectly. At some point, it decided to clear back up. It ran like nothing was ever wrong.

However I know it is still not right, it is just intermittent. I wiggle wires and the harness everywhere I could and it just ran normally.

Rain dang near every weekend since has prevented any further work. And now, my actual job has started picking up and I have working 7-5 during the week, and 7-12 on saturday.

However after work today, it was sunny and somewhat bearable temperature (mid to high 50's).
I brought home a headlight wiring harness from a lawn mower. It has two single filament light bulbs the same as what is used in a lot of older vehicles for brand or signal lights (1156 bulbs).

I still think the issue is ignition related. So I disconnect the ecu, disconnect the distributor connector, then using a paper clip in the ecu connector, I use a jumper wire and ground each of the 4 wires one at a time.
I then connect the other end to the mower headlight harness and then connect that to the battery.
This put the load of two of those lightbulbs on the wires. I tested each of the 4 ignition wires, one at a time.
Both bulbs were very bright with no dimming and all wires were the same brightness. While I do not know the exact amount of amp draw that was putting on the wires, I would have thought if there was a issue, that would have been enough of a load to bring it to light (no pun intended).

I put it all back together and crank it up. It fired off fine and kinda sounds like it was doing the missfire thing, but I did not rev it up. I just let it set and idle while I put up the tools and wires.
After running for about 5 min, I take it for a drive and it is running perfectly fine. I drive it up the road and back a couple miles with no issues.

So now it seems like it is going to do it intermittently, which makes it even more fun to find.

[MO LS Noobie]

Keep at it, you were fighting several issues. Start over as if it is a new situation. Still sounds like a ground issue to me. Voltage dropping during cranking, resets adaptive values, runs rough till it relearns.

[TriumphR3]

Possibly a ground, but I keep thinking back about an issue on my C1500 where a wire in one of the connectors at the ecm had worked itself loose and became intermittent. The truck ran terrible. This circuit went to one of the fuel injectors on the tbi. It was just dumb luck that I found it. I tugged on wires all over that truck. Replaced the ecm a couple times. Checked wires with a dvm. Finally tugged on the right wire and it reared its ugly head. Reterminated the particular contact and issue resolved.

Makes me wonder, if it’s consistent it’s usually ignition related. But when it randomly misses it might suggest fuel related.

[MO LS Noobie]

See my post #49 above. I've also had issues with customers wiring harnesses where it flexes and breaks the conductor inside but believes the outer insulation. I put about 5 pounds of stretching force on each individual wires, especially near the connectors. If it stretches the conductor inside is faulty. Regarding grounds they can look good but don't pass the high amperage needed. Voltage drop is the way to test them, but unbolting them and cleaning them with a wire brush, and putting dielectric grease on the connection is good preventative maintenance anyway.

[TriumphR3]

Fair enough Mo, I was just sharing my experience, thought it might be helpful. Shed some light at a different angle. No disrespect intended.

[MO LS Noobie]

Fair enough Mo, I was just sharing my experience, thought it might be helpful. Shed some light at a different angle. No disrespect intended.

None taken, we are all here trying to help each other. It can be very difficult to remotely diagnose the vehicle, especially one that is over 40 years old.