shadowspawn2

Heyya;

I have a stock 1998 straight 6 4L and something strange has been happening for just over a month.

You are familiar with the mindset that when you know your engine and your engine is running perfectly, one misfire will make your neck snap if you are outside and it's at idle. You'll interrupt whatever conversation you were having to just listen for it again. One time the engine would misfire, maybe once 10 minutes. Drove me nuts until I stared at it at night, heard the click, popped the hood, stared at it, saw a spark jump out of a sparkplug boot, new wires and plugs were procured and installed the next day. (About a year, 10k miles ago) She always idles perfectly.

Flash forward to the present:

She starts fine, under a full rotation of the engine, almost instantly. She purrs like a kitten. Some days heading to work I'd just let the engine change to it's non-cold RPM's, head to work before it's fully warmed up. I live on a long hill, so it's idle and brakes for about a mile, stop and go for about 5 minutes still downhill, then on the main road to work so by then it's warmed up enough. Did this morning ritual one day in about 2 minutes I lost power and the throttle was barely responsive.

She was like I was running out of gas so I went into the old carburetor mode: foot off the throttle, feather it use the throttle pump, hunt around and feathering to find the sweet spot to get at least far enough to pull over, too much throttle no power, let off on it then goose it and she'd go for a bit, and damned if she was acting like it was a carb with a clogged fuel line filter or clogged jet, but mostly it felt like it was loss of fuel. Thing is, this is FI not a carb. So I'm stumped.

She all of a sudden acted fine. I still pulled over, raced the engine a few times, kinda was rough with her in the parking lot I admit, but no problems at all. So now the worrying started. After work she ran perfect on the way home.

The next day I went out about 10 minutes early. Started her up, still ran perfect. But then I heard it: she started to misfire. Not really bad, but bad. I got in and was goosing the engine, revving it, raising the idle, the engine is shaking. Ok (again falling back to old carb ways, let's blow out carbon etc.) held her steady, misfiring about 3k, 4k... poof. Check engine light came on. "GREAT!" I thought, this is a good thing. At least I have something to go by. Hooked up a diagnostic tool, cylinder 2 misfiring (pretty much an O2 sensor). Now the "DAMN!". Revving her so high whatever was wrong triggered the error threshold. Unburned gas, way too less or too much, but no black smoke or smell of unburned or rich. She started idling fine again. Shut her off, cleared the code, started her up, and she was fine.

So I repeat this process the next morning but instead of running the throttle up when she misfires, I let her go through her morning cough. Idles and runs fine for about 2 minutes. Then starts misfiring for about a minute or two. RPM drops, obviously. Then the engine revs up to about 1.5k for a second (sometimes she'll go back to misfire, rev for a second time) and the problem simply goes away. Like a smoker's cough in the morning.

I've eliminated fuel pump pressure., pressure is not dropping off when the coughing starts. It's pretty steady. Cut her off about a minute into the fit she's throwing and yanked plugs but no wet plugs (usual look) and it's frustrating because by the time she starts to misfire the plugs are already hot enough to evaporate anything that would be left on them. It's too difficult to figure out if it's a particular cylinder because by the time I'm starting to yank wires as she's running, the problem goes away. If there's an offending cylinder I can't find it. The only way I can reproduce it, when it's warm out or -10f it doesn't matter, is by letting her sit for about 12 hours. doesn't matter if it's overnight or 4 days. Acts the same. I know this is back-woods troubleshooting but it's worked for me before.

Starts fine, idles fine, starts misfiring, a surge in RPMs, then runs fine for short distance runs or long hauls. I have my guesses but something should have failed by now (or the engine and parts are really built well and hold up when they start to fail).

What the heck would cause this? It's like part of the morning warm-up process for me. No drop in gas usage on highway or back roads, no loss in power going up big hills, no problems at all after that morning sputter. It's been doing this for over a month now. Except for me beating the heck out of her with racing the engine during her fit there's nothing in storage as far as error codes. I just am trying to narrow this down before a catastrophic failure if I'm too far from home.

After a prolonged rest: Cold start fine, cold idle fine for a bit, cold sputter for a bit, then cold up to hot smooth operation no matter the ambient temperature or the block's starting temperature... as long as she's been sitting for a while. After a long period of rest. No error codes.

-edit- spelling and gonna go take a video of this and post it on youtube so check back in a bit.

cruiser54

How long since plugs, cap and rotor? They should be replaced every 30,000 miles along with the plug wires.

shadowspawn2

The plugs and wires were last year. Didn't do the cap and rotor in a while, but there's no corrosion. That was my first thought too. I could try it but it doesn't make sense why it only does it for a minute or two after idling cold after a minute or two then goes away.

I clean it from time to time (like when I am bored or just want to clean the grime off of things) with electronics cleaner, emery cloth, coat it with a small fillm of dielectric grease for good measure. There's never been moisture or oil inside of it and no blackening. The contact on the rotor isn't pitted and is the same length as a stock one. In my experience failure of the cap/rotor would usually cause misfires under load.

The rotor doesn't jiggle or is loose and I know it's getting a strong spark, nice and blue. The skip on a cylinder that I had before, just random, and I mean like every 5 minutes hot or cold is when the spark would literally jump out of the sparkplug boot randomly from time to time (like just once and a while it'd just miss on that cylinder). Jiggled it, coated everything with dielectric grease inside the boots, but it'd eventually start doing it about 3 months later. So good wires and new, properly gapped plugs fixed that.

This is different. This is a coughing fit that goes away. Still gotta take that video. Waiting for a bit. I could try it but I'm not sure that it'd fix it. The problem isn't constant, it just happens after resting. I checked for moisture inside of it trying to figure out what it is, nothing inside of it but shiny metal.

The last rotor and cap replacement was about 5 years ago, less than 30k. Plugs and wires just over one year ago. Plugs looked fine when I just yanked them to see if it was an offending cylinder, cutting the engine off during the latter part of the misfire.

-edit- she made a liar out of me this morning. Sat there recording from cold start to full heat-up and it was fine. It could be because I drove her less than 12 hours ago. I will be vigilant, but it does happen. I'm not insane.

gat

You mentioned that the fuel pressure was steady. What is the steady pressure? The cap, rotor, wires and plugs are new. Is the filter for gasoline and air, as well?

Do you think that the fuel injectors could be just a hair clogged and the engine operates better after a couple of minutes simply because there is more heat in the combustion chamber to combust a slightly leaner mixture?

jburg318

I didn't read your full description, but those are the same symptoms I had when my upstream O2S was going bad. I cleaned the O2S heater relay contacts and that didn't work so I replaced the upstream and it was fine afterward.

jburg318

If I remember correctly, my voltage was staying low when the engine entered into closed loop. Caused a 'rich' condition on STFT as the ECU was trying to remove fuel. Bogged and sputtered a minute or two from time to time and then ran fine once, what I though was the heater, reaching temp and getting an accurate reading.

Idunno

Yep. #1 O2 sensor symptom. Replacing mine was the cure.

shadowspawn2

I thought it could be that, but it doesn't fluctuate much once running, it stays steady at around above 47 psi. On just turning the ignition on without starting it hangs and stops at 40. Turning it off it doesn't really drop that quickly. It doesn't dip when the misfiring starts. I'm using an old gauge It vibrates a bit but it seems steady (it doesn't have one of those "balloon"/pressure tank attachments to steady out the pulses. Need to get one of those one of these days.)

shadowspawn2

Wouldn't that cause an error code in the ECU's memory? The only time I tripped one was gunning it and keeping it at a high rev.

I'm about to get a new exhaust system so might as well. Someone offhand mentioned "it's always the oxygen sensor" when I asked at work, he said no matter what the car, it's the O2 sensor. Now three people on this thread says it could be same thing. I like it when I see people agree on things I don't understand, AFAIK I figured if the O2 sensor went it'd trip something.

Wow, it does start once it goes into closed loop. That first stage when the idle starts to drop a tad bit, like a tiny bit, right? It's not really noticeable and I'm used to carbs, not FI on Jeeps, but I can hear it and see it. On any other car the transition from open to closed is transparent. On the jeep I don't ever put it in gear until I notice that little tiny drop. I guess my problem that's happening is when the ECU is "hunting" and the O2 sensor isn't hot (or hotter than, am unsure as to which way they fail) as it should expected to be yet, starving the fuel.

Then when it all of a sudden reports for duty, that's why the surge in RPM before it calms down?

jburg318

I am pretty sure that the surge corresponds to the ECU correcting the fuel pulse width once the heater is at temp and the voltage across the sensor is reading correctly than the ECU realizes it's been backing way off on fuel pulse width due to the bad reading. It should throw a code, but I think in order to set it, it has to be truly dead and not be sending back data for a set amount of time (short to volt or something along that line) I have to look through the manual. Mine did not throw a code for the issue. It would run fine, switch into closed loop and run like crap and bog out, the surge and rev for about 30 sec and be fine. I only figured it out when I took a look at the real time data and the voltage was staying low (indicating a rich condition - corresponding to the ECU removing fuel and reducing the length of the fuel pulse width). The voltage across the sensor should change within 45 sec to a min. Mine stays low in closed loop, switches on and leans the pulse width out (~.9 volt), and then starts fluctuating.

​​​​​​ I've never really noticed when mine switches to closed loop, it does happen surprisingly quickly though. I have the option to monitor it on my scanner.

Just go with NTK, they are OEM for '98.

shadowspawn2

Got the video. The "knocking" sound is just the cat rattling around. Like I said above I'm getting a whole new exhaust system from the cat back, so I might as well do the O2 sensor.

https://www.youtube.com/watch?v=YvXh...ature=youtu.be

I didn't even touch the pedal. It just went from idling fine to rough, revved up, then went smooth as silk.

Thanks guys.

Waynerd