Hard starts when cold, rough idle, plethora of codes

[gogmorgo]

I've been noticing the symptoms for a few months now, but only when it's really cold out, approaching -30°F or so, and the block heater wasn't plugged in. Basically it'll crank over fine but won't fire, maybe the odd cough but not really trying unless I open the throttle to 1/4 or so, at which point it sounds like it'll go but just takes a while before it actually fires up.

I figured this wasn't unusual for an old vehicle in the cold, but it didn't do this last winter or any other winter from my memory. My guess was something was causing it to dump in too much fuel due to the cold, maybe a failing coolant temp sensor that was getting fuzzy toward the bottom end of its range or something.

But this morning it took a hell of a time to get it started. Nothing if I wasn't on the throttle, and only the odd cough if I was. Then when it did eventually start, it didn't want to idle, unless I had the throttle open a touch, until it was warm. When I took my foot off, it would drop down to a very rough 500rpm and then stall. The second or third time it died on me I checked trouble codes and just got 12 and 55, so not really significant, the battery would have been disconnected two weeks ago when I put a new starter on it. (Starter was replaced because the old one would spin but not engage the flywheel, when I pulled it there was about 1/4" of play in the drive gear).

At lunch I had to run an errand and it still didn't want to start. It had warmed up almost to 0°F by then. It idled rough and wanted to stall again but after 30sec or so of holding it at ~1200 rpm it held a 750rpm idle. Needless to say I didn't shut it off till I was back at work.

Leaving work it was the same story, wouldn't idle till it warmed up a touch. I went and got the mail, leaving it running the whole time, and when I got home I shut it off and started it again a few times and it was fine, so I checked the codes, and got three new ones, in this order:

12-battery disconnected; I'd expect that to have cleared by now but NBD.

53- PCM fault, requires specialist diagnostics tools/diagnostics

76- fuel pump bypass relay circuit, requires specialist tools/diagnostics

42- Auto-shutdown relay, check relay and circuit, requires specialist tools/diagnostics

55- end of codes.

After pulling the codes, I let it cool down for 20 minutes of so, and then ran to the hardware store. It started fine both at home and at the store.

I picked up a new cord end so I can actually plug it in tonight, it's supposed to get down to -30°F again so I'm not optimistic about it starting without the block heater. So we'll see how it goes. I'll definitely be heading out early for work in case I end up walking.

 

I haven't messed around with anything yet, but beyond pulling relays and checking them I'm not sure what to do really. I've got a working-when-pulled '93 ECU I could swap in I suppose, but it runs fine and I can't really see the ECU warming up much at all in 30 seconds.

The other thing is that none of those codes triggered seems to have much to do with something opening the throttle would cure, or have any effect on. If it was starving for fuel because the pump wasn't getting proper power, you'd think opening the throttle and leaning out the mixture more would just make the problem worse. I can't see any way the throttle would affect either the ASD relay or the ECU, unless maybe the TPS is sending wonky readings but I can't see that going away after running for under a minute. The only thing I really can think of the throttle helping is if it's dumping in too much fuel. I was also wondering about the idle air control, but I can't see that triggering any of those fault codes.

Any ideas?

This isn't the best time for this to happen... Found out yesterday I'm moving across two provinces in a couple weeks, got tons of crap to deal with and don't need more!

[HOrnbrod]

The code 53 comes up when the ECU senses an internal fault. The other codes could be be caused/effected by the ECM fault code. First thing I'd do disco the cable and eyeball the 60-pin ECU connector for obvious bent and/or recessed pins. There's probably gunk on the pins too that becomes brittle from the extreme cold overnight. Clean it up real good with some contact cleaner and a tooth brush. Reconnect and try again. If the same symptoms and codes persist, I'd try another ECU. I'm not sure a 93 ECU will play on a 91 because some of the pin assignments are not the same, but it wouldn't hurt anything to try. Even the 91 and 92 ECUs have some pin assignments............

[gogmorgo]

0°F is far from extreme... we don't get extreme cold warnings until -40, so I suppose -30 is getting there. But I guess that's down to personal opinion. It's pretty annoying that the cold is cause me problems this year because it hasn't in the past.

 

I cleaned the ECU connectors and everything off when I was having issues this summer (turns out I was out of gas... oops) but I'll give it a shot again if it acts up tomorrow. I suppose I could be low on fuel again now, but it isn't behaving like it is. At least I haven't noticed any problems during driving maneuvers that implement full throttle and fuel sloshing...

[HOrnbrod]

Well excuse me for saying the temp was extreme.   :shake:   

[DirtyComanche]

Well excuse me for saying the temp was extreme.   :shake:   

 

I would probably find the temperature of an average day in 'Bama to be extreme. :rotfl2:

 

 

If you previously cleaned up the connector for the ECM (or other ones) I would start there again.  They're supposed to have a light film of corrosion preventative grease on them (typically of the dielectric type) and removing it can cause troubles down the road.  Give the IAC and throttle body a cleaning too, for the few minutes it takes it rules out what can turn into some tail chasing.

[gogmorgo]

Any sort of lubricant I should use on the IAC and throttle body after I clean it?

It was ridiculously hot in 'Bama when I was there a month ago. Seriously, it must have been around 50 degrees. I was down to short sleeves, windows down, sunroof open, hair blowing everywhere, it sucked. Don't know how you can stand to live there.  :teehee:

[gogmorgo]

So it was -35°F this morning when I tried to start my truck, wound up walking to work. After a while moving the throttle did nothing, but before then it would only cough and try to fire when my foot was in it a bit. Filmed my attempt to start it:

After that I went inside, got dressed, and reattempted about five minutes after the above and couldn't get it to even cough.

Not two hours later I had to stop in from work to grab my OBD2 reader, and since it was in the MJ I attempted to start it. I did end up touching the gas but it fired up, and while it didn't want to idle nicely at first, it was only about five seconds or so of me holding it at 1000 rpm until it caught about an 850rpm idle on its own. I'll point out for reference that it had warmed up to about -10°F by then. So I drove to work, left it idling for about 10 minutes there so it was up to operating temperature (it's not a very long drive, less than a mile) and shut it off. It started up just fine after work at right around 0°F. I did fill up on the way home, and even though it was only a 55L fill, that's still getting down towards where I've had it start to have running-out-of fuel symptoms ( I really need to adress the Delphi-fuel-pump-adapters-suck issue where it's pulling fuel from the middle of the tank) and while I know fuel volume isn't affect all that much by temperature, it wouldn't surprise me that a drop in volume due to temperature is in fact causing problems. But it's getting down to -20°F again tonight (and it's clear enough it wouldn't surprise me if it goes colder) so I've left the block heater unplugged and I'll see if it starts tomorrow to try to eliminate the low-fuel possibility. It surprises me that the fuel tank would have changed temperature as quickly as the air though, so I'm pretty skeptical. But I've got too much other crap to do what with moving and all that I don't know when I'll get time to actually pull the throttle body off and clean it up.

I'm also inclined to wait until it's warmer out, haha. The weather prophets have been staring at the cat's vomit long enough they've come up with a prediction of highs in the 40's by the middle of next week.

[Ωhm]

Not to be some kind of smartazz here, but what about your spark plugs? Sounds like that vehicle should have started. I like that -35° 'stay in start' ignition switch and how you pumped that pedal (accelerate pump style). Spark plug mileage, gap (.032), and rounded electrode, would be where I'd start first.

[gogmorgo]

Plugs and wires were new in August, maybe 5000 miles on them. I've thought about pulling them to have a look, maybe I will tomorrow morning if it doesn't start. And yeah, that's why it's frustrating. It'll sound like it's trying to go until it quits trying and just cranks away.

Yeah, I've been using white lithium grease in the lock cylinder, which was recommended for cold weather by a locksmith but I'm not positive his cold was quite mine. It's good down to -10°F or so but below that it doesn't want to pull itself back. Whether it's a tired old spring or the grease solidifying I'm not sure. It's kind of annoying because until this winter I never had cold-related issues with anything.

[Ωhm]

Battery heat blanket might help for a longer crank time. Sound like voltage is dropping below 7vdc and thats when the ECU will act a little funny. 5 volt regulators need at least 7 volts to stay alive.

[gogmorgo]

I'll do a load test on it before starting it tomorrow. Battery is admittedly a cheap one but it's only 18 months old and has never been discharged out in the cold. It hasn't seen much more than a couple weeks of heat above 90° either, and the longest time it's spent without the Jeep starting has been a couple weeks at most. If the voltmeter on the dash is to be believed in the above video it doesn't look like it dropped below 10 volts under cranking, and if it's cranking over that hard without losing speed at -35, it's not anywhere close to seven volts. The starter is only a few weeks old, so it's not like it's drawing more than it's supposed to be, either.

[gogmorgo]

It dropped to about -25°F last night, although by the time I was out messing with things it had warmed up to around -5.

Load test on the battery didn't drop below 600CCA. The battery is fairly new, has never been abused and I've noticed nothing to suggest it's bad. No offence but in my mind a battery blanket is a bandaid solution for a dying battery that will leave you stranded when you inevitably park somewhere you can't plug it in. It's not unusual for me to be parked at a trailhead with no cell service for 15 miles in any direction, coming back to the truck around sunset, and I'd rather not walk out in the dark.

But yeah. Left the block heater unplugged, and again it wouldn't fire without throttle. Once it did it took me a couple minutes of nursing the pedal until it would idle by itself. Having a full fuel tank doesn't seem to make a huge difference.

I've got to run to my parents' to drop off a bunch of my stuff today so I unfortunately don't have time to mess around with it a whole bunch more.

[HOrnbrod]

The ECU code 53 can be caused by a bad O2 sensor (no signal to the ECU), so for grins make sure the harness hasn't fried or something.

 

Where are you moving to Gog? Hopefully it will be warmer.  :rotfl2:

[gogmorgo]

Little bit warmer but not by much. Jasper, AB. Think Banff but a little further north.

 

And about that O2 sensor thing... Going down the road just now I hit a bump and my exhaust got a TON louder. Pulled over and discovered one of the nuts had gone AWOL off the manifold to downpipe studs and the other one's working its way off too. If that's been leaking for a while I'm guessing that could be messing with the O2 readings. The wires looked okay. What didn't look okay was the obvious evidence of the downpipe self-clearancing on the driveshaft. The O2 itself is an NTK that's 18 months old, 25,000km on it (so 15,000 miles maybe? I didn't do math for that) Could still be having issues I suppose :dunno:

I don't know how a major exhaust leak would affect starting, given the O2 is ignored until closed loop. Maybe the donut washer thing in there expanded and sealed the gap better when it's warm? I have been noticing that my fuel economy's been pretty low, although I've been blaming it on more time spent idling and also getting sideways around many corners not entirely unintentionally cause I gotta do something out here for fun.

[Ωhm]

Found something you might be interested in. Article on cold starting, relay to ICM pg 6. Just thinking about where you are makes me cold.

 

https://files.acrobat.com/a/preview/7cab64fd-fa7d-45be-b2da-a8209a4fbef3

[HOrnbrod]

He doesn't have a Renix.

[gogmorgo]

:yeah that:

Also it was pretty warm today. Got up to around 10°F.

[gogmorgo]

Okay I take that back. -28°F again this morning. Now I'm wondering about my exhaust system, cause it started okay-ish with the massive exhaust leak. Didn't have to touch the throttle and it only cranked for ten seconds or so before turning over. Not bad considering I left it unplugged cause I didn't check the forecast.

Fortunately my neighbours are only neighbours for the next week and a bit cause I'm pretty sure they'll hate me if I leave this thing idling outside for too long. Although on the plus side they won't need an alarm clock...