Dead at side of road

[superbad]

Currently writing this waiting for a tow truck.

1. Had a rattling cat starting last week, car started running like absolute s**t (it was fine prior). Figured it was excessive back pressure from broken cat.
2. Had the cat replaced this morning. Car was better (made boost at least), but still very rough and obviously not right
3. Has been very difficult to start since cat problem came up, especially when warm. Started poorly leaving the exhaust shop.
4. Petrol smell while driving home just now, not sure if that was me though.
5. After taking highway off ramp, car immediately died and I coasted to a stop in present location.
6. Attempts to start are futile and produce considerable whitish smoke.
7. CEL came on for a couple minutes today, both before and after new cat.

Likely causes?

[superbad]

No smoke now on cranking, still won't start though. Was that smoke unburned fuel going through the exhaust and burning off on the hot cat?

[djcar]

not getting fuel maybe? fuel pump? fuel filter clogged?

[superbad]

Plugs are black and wet with fuel (not surprisingly). Definitely getting fuel. The conductor on plug #4 came apart when I pulled it off. Cause or coincidence? Entirely possible it happened when I pulled off the wire. Definitely felt like an intermittent single cylinder misfire driving home, but it seems like there must be something else going on as well. This thing should (and has) run all day on three cylinders.

photo (Medium).JPG

[superbad]

Only code stored is 45 (rich exhaust). Thanks for that. Can't run ElanScan with only three plug wires.

So... I have ordered new plug wires. I still can't believe that is the cause of these issues though. Any other ideas?

[chris.t]

White smoke > sounds like head gasket

[Fredjohn]

White smoke could be burning brake fluid which has been sucked in by vacuum from servo, where seals may have gone.

[superbad]

Don't think it's either of those. Wasn't really coolant fog-like. Only had a gasoline smell. Coolant level is unchanged, and I checked compression a few weeks ago, all cylinders were fine.

There was a standing lake of fuel in the cylinders when I pulled the plugs. O2 sensor was wet as well. But that was after a lot of cranking trying to get it to fire so I could limp it home. It died 1/2 mile from my house (cost me about $200/mile to get it home on the tow truck).

Could the O2 sensor be bad? It was (correctly) reading rich though, and should have leaned it out to compensate. What could cause rich running that can't be corrected by the ECU?

[chris.t]

A duff coolant sensor can cause massive over fuelling. But that should give a fault code. Still try a sniffer test to check for combustion products in the cooling system.

[cliff]

Would have thought the O2 sensor should give a fault code, not sure what the ECU does if it goes down.

I had one fail on the Grand Cherokee, the ECU is programmed to over fuel to stop one bank running lean. A 4.7 liter V8 overfuelling is not good for the pocket!

[Brit-Car-Nut]

As I understand it, the ECM doesn't go into "closed loop" if the O2 sensor doesn't report correctly. You are basically running on the "limp home" program. If the temp sender doesn't work, the ECM thinks it is still cold and increases the pulse width for every injector burst. It basically has a stuck choke which is also very rich and will soot up the engine/O2 sensor quickly.

Excess fuel also washes the cylinders and thins out the oil so that might cause the white smoke but you should also check the PCV and the power brake booster hose for oil in the line.

If you are using an antifreeze with ethylene glycol, it will be part of the burn cycle if the head gasket is bad or something is cracked. You should be able to "taste/smell" the sweet aroma at the tail pipe. I have been known to use the Polish "tongue test" with the tail pipe residue (just a pinch). Antifreeze has a bitter sweet taste and it doesn't take much to test. It is NOT HEALTHY to consume it but a finger to the inside of the pipe and a tiny touch to the tongue will let you know if antifreeze is exiting through the exhaust. Using an emissions sniffer will also let you know if exhaust is getting into the coolant, but coolant often gets into the combustion chamber first. If you try this, have a couple of clean paper towels handy and some water to "clean" your tongue. Don't swallow until you are sure things are clean.

If you have ever been in a car where the heater core has leaked and the engine got to full temperature, you will know the taste/smell I am talking about.

Check the engine oil for petrol contamination. If in doubt, change it. I would use a 10W "normal" oil for the short test process.

Let the cylinders dry out, replace the O2 sensor, fresh plugs & plug wire, hook up the Elan Scan. Start recording and start the car.Watch the coolant temperature. It should go up slowly and reach a point at 185 degrees where it will oscillate up and down until the entire cooling system reaches 185 - the thermostat is opening and closing as the cooler water knocks the temp down when it opens. Eventually, the temp equalizes above 185 and the thermostat stays open.

If everything is good, rechange the oil to the exotic synthetic of your choice.

[Andy_J]

Couls it be just a simple spark problem, duff coils?

[rip]

A duff coolant sensor can cause massive over fuelling. But that should give a fault code.

Only if it is out of range. (I believe it can cope with climates as cold as -40c).
I ran Elanscan once with a non-runner when warm to find that the coolant sensor was reporting -37c IIRC & I had just driven there with my roof down! The CEL was not on. Engine ran for about 1 minute from cold before cutting out.

[HJ2]

A duff coolant sensor can cause massive over fuelling. But that should give a fault code.

Only if it is out of range. (I believe it can cope with climates as cold as -40c).
I ran Elanscan once with a non-runner when warm to find that the coolant sensor was reporting -37c IIRC & I had just driven there with my roof down! The CEL was not on. Engine ran for about 1 minute from cold before cutting out.

Don't rule out the air temperature sensor in the manifold as well as a potential cause of overfueling...
Elanscan would help.

Good luck!

[chris.t]

Over fuelling would give rise to black smoke too.

[superbad]

Thanks for the help. I'll give that sequence a try once the new plug wires get in and get some ElanScan data. I am unfortunately well acquainted with the smell of coolant, having had it spewed all over the engine bay on multiple occasions. Really don't think this is HGF, although there could be considerable wishful thinking in that assessment. Coils are about a month old, so they should be fine.

Wouldn't limp home mode throw a code though? My CEL came on for maybe a minute on the drive home, and went off by itself. On every other car I've had, the CEL stays on until you either clear it manually, or start the car a few times without a recurrence of whatever caused the error.

[Brit-Car-Nut]

I assure you I am not hoping it is a HGF. Just suggesting things to check to get a better understanding of what took place. I know the coils are new, but didn't they get soaked when the header tank burst a few weeks ago? Any chance they need to be removed and all connections including the base assembly get a good cleaning?

I don't trust the error code storage feature of the ECM. I read way too many posts where people have huge issues and no codes are stored.

I think the designers wanted to keep it more like a hunt and seek game...

[norman lovie]

I don't trust the error code storage feature of the ECM. I read way too many posts where people have huge issues and no codes are stored.

I think the designers wanted to keep it more like a hunt and seek game...

This has been the best money making engine aspect the car industry has invented
take a spotty blob of a yound apprentice - give him a widget plug in box and he can tell you instantly whats wrong - then off to the parts dept. to buy the grossly overpriced bit the computer say's is malfunctioning, and - you know! the computer never lies
A week later, customer returns with the engine code flashing again
repeat performance from spotty blob - and another load of dosh for another sensor - cash ching ching
I,m sure we all know stories of the the car we or a friend bought very cheaply because some ridiculously priced electronic widget is faulty - -to find a loose wire

its a tool - a good tool, but it ain't the orical

[superbad]

I assure you I am not hoping it is a HGF. Just suggesting things to check to get a better understanding of what took place. I know the coils are new, but didn't they get soaked when the header tank burst a few weeks ago? Any chance they need to be removed and all connections including the base assembly get a good cleaning?

I don't trust the error code storage feature of the ECM. I read way too many posts where people have huge issues and no codes are stored.

I think the designers wanted to keep it more like a hunt and seek game...

This set of coils has never been wet, they were replaced after the header tank fiasco. Anyway, here's hoping it's just a bad O2 sensor or something similarly easy. I'll report back when I'm able to collect some data.

[Brit-Car-Nut]

I assure you I am not hoping it is a HGF. Just suggesting things to check to get a better understanding of what took place. I know the coils are new, but didn't they get soaked when the header tank burst a few weeks ago? Any chance they need to be removed and all connections including the base assembly get a good cleaning?

I don't trust the error code storage feature of the ECM. I read way too many posts where people have huge issues and no codes are stored.

I think the designers wanted to keep it more like a hunt and seek game...

This set of coils has never been wet, they were replaced after the header tank fiasco. Anyway, here's hoping it's just a bad O2 sensor or something similarly easy. I'll report back when I'm able to collect some data.

OK, good to know. I hope it turns out to be something trivial.