Car on fire: plausible?

December 18, 2007 7:52 PM

Plausibility-filter: Recently my car engine wouldn’t stop running, and eventually caught fire. Damage was pretty minimal, but how it started is a bit of a mystery. Does this sound plausible? Backstory (long) and the mechanic's explanation below:

I drive a 1991 Honda Civic hatchback. On monday I was driving to work when I noticed that every time I took my foot off the brake the engine would rev up and the car would speed up dramatically, without putting my foot on the accelorator. After a few blocks of being puzzled by this I noticed the temperature guage was over the top, so I immediately pulled over to the side of the road and turned off the engine...except the engine kept running, clunking even, and I had removed the keys from the ignition. I opened the bonnet to see if anything was obviously going awry, but I don't have much of a clue what I'm looking at. I called roadside assistance, they said they'd be there asap.

While waiting for them more and more smoke was billowing from the back of my car's engine, until eventually it caught fire. I called the fire dept, and a nice old man came out of his house with a bucket of water and doused it. It flared up again, but thankfully the fire dept. arrived quickly and put the hose on it. The engine kept running throughout the whole drama, only ceasing when one of the fire guys pulled out the plugs on the side of the engine.

Anyway, the fire guys thought it was the catalytic convertor, the tow truck guy thought it was electrical, and my mechanic was baffled. He calls me the next day to say he reckons it's the carburretor - flooding the engine with fuel, which was being burned by the engine, causing it to accelerate. Excess fuel was being spat out the exhaust, which heated up and caught fire. So now I need a new carburretor and a new handbrake cable (burnt through)

Does this scenario sound plausible? Can a carburettor fail like that and effectively set the engine on fire? Has anyone ever had this happen to them, or heard of something similar happening ever?

posted by robotot to travel & transportation (16 comments total)

Your first comment about every time I took my foot off the brake the engine would rev up made me think it was that wire that connects your throttle and carby was getting stuck. Effectively the same guess as the carby failing. Entirely plausible.
posted by pompomtom at 7:59 PM on December 18, 2007

So now I need a new carburretor and a new handbrake cable (burnt through)
You need a new car. If it's got into such a state that it can do something like this, god knows what condition the rest of it is in. Car fires are terrifying, and incredibly lethal. You got my sympathies.

As for the plausibility of it: I've seen carburettors fail in interesting ways, yes. It's where the air and fuel meet. You control the air to an extent with the choke, you control the fuel with the accelerator. Any bit of that goes run, and you have an unhappy engine. I'm not sure about the whole ignition-off scenario, especially on a Honda.
posted by bonaldi at 8:03 PM on December 18, 2007

Is the electrical system stock, or is there any kind of spark booster on it? When I put an MSD box on my Impala one night, I couldn't shut it off - the tiny bit of current that powers the "generator" idiot light was enough to keep the spark on the car. I finally killed it by physically removing the carb while the car still ran.

More to your point, the throttle could have been stuck, but that wouldn't explain why the car overheated. Yeah, probably something in the carb. BUT - did they still have a "carb" on a '91? My 90 CRX (same car, basically) has fuel injection.
posted by notsnot at 8:10 PM on December 18, 2007

I had an old '89 Ford Festiva (actually a Kia something-or-other) that had a problem with the "Secondary butterfly valve". (it's part of the carb.) sometimes it would get stuck open just a crack, and let too much air into the engine. kinda scary, but i was able to stop it by shutting the ignition off...

anyway, sounds more or less plausible to me.
posted by ArgentCorvid at 8:23 PM on December 18, 2007

Sounds about right to me. I once drove a car (an 84 Civic hatchback, as it happens) that would diesel for a while after you shut it off, but the carb wasn't stuck open so far as to dump enough fuel into the engine that it would get blown out the exhaust. It was a small enough amount that it would quit on its own after a few seconds.

I'm surprised that a '91 Civic came without fuel injection, though. Honda tends to be pretty standard across their line, and the accords that year all had fuel injection.

Unlike bonaldi, I wouldn't be too concerned once you have the repairs completed, as most carbureted gas engines will happily diesel as long as they stay warm. It's not good for the engine, though, as it causes uneven burning, which can crack a piston, crack the head, foul your valves, and a whole host of other things. The car I referred to earlier had the head crack about 20k miles after the dieseling started. The owner temporarily "fixed" it with some JB Weld and drove it for a few more months before he came up with another beater to replace it.
posted by wierdo at 9:47 PM on December 18, 2007

Could the engine have become so hot that it was running itself just from the heat? This happened to me a week ago in an old car I have, the head gasket blew, vented all the water from the water jacket and then the car got hot VERY quickly. When I turned it off, it kept running rather roughly, until I dropped it into gear to stall it.

Another thing is that my dad had an old car that had an air flow valve break off and jam so that when I took my foot off the accelerator, it wouldn't idle but instead revved pretty hard.

Combine these two things and maybe you've got a car that will open up the throttle, get extremely hot then continue to run by itself with a wide open throttle until a catastrophic failure.
posted by tomble at 9:54 PM on December 18, 2007

Second the new car recommendation. Anything newer will have fuel injection, no more weird carb related fires. And that car is likely no longer worth the value of repairing it. (It's recently been ON FIRE, and it's 15 years old).

Ok, maybe not a new car, but a car that's new to you. Like a '93 or something. ;)
posted by TeatimeGrommit at 10:19 PM on December 18, 2007

Carb on a Honda? That seems unlikely to me, too. You might want a second mechanic to look at it, unless you know for certain the car isn't fuel injected.
posted by Goofyy at 12:13 AM on December 19, 2007

Amazing - my ex-gf had that same model car... she actually blew up two engines before they found the problem - the carburretor.
posted by puddpunk at 12:31 AM on December 19, 2007

Tell your mechanic to take a look at your cooling system, (radiator, water pump, etc.). What you experienced sounds like run-on, which occurs when the surfaces inside of the combustion chamber of the engine get so hot that they ignite the fuel-air mixture without an ignition source, ie. spark plug. This is why the engine continued to run even after you turned the ignition off. It's a relatively rare occurrance, but it happens mainly in older engines with larger amounts of carbon deposits in the engine, because the deposits will get hotter than the metal surfaces in the engine.

It might be possible that the throttle was stuck wide open which would explain the overreving when you let off the brake, but that would not explain the run-on. It might be a combination of a stuck throttle and something wrong in the cooling system. But I can almost guarantee that this was not the carburetor alone.

I would put my money on a burst coolant hose somewhere, it's something that you would not notice until the car overheated, and in conjunction with a stuck throttle it explains this situation perfectly.

By the way I am a senior mechanical engineering student who has just completed a course on internal combustion engines, and have been working on cars since I was a kid.
posted by goHermGO at 12:52 AM on December 19, 2007

btw i should have mentioned that I had this happen to a small-block 305 that I was working on, and the problem was what I just explained above.
posted by goHermGO at 1:01 AM on December 19, 2007

A similar thing happened to me while I was driving a friend/customer's car.

It was a '92 Saab, and no it's not carburated either.

A major vacuum line blew out (it's a turbo, but it happens on non-turbos too) and the engine started revving like mad. Fuel management was kinda confused; generally the engine runs lean when this happens. Lean = hot. Hot enough means there's sources of ignition besides the sparky plugs = can't shut it off.

Did the brakes feel different while this was happening? Harder to press?

This is actually not a rare occurrence.
posted by KenManiac at 6:44 AM on December 19, 2007

I had an old Accord (87) that did this. The engine would idle very high in neutral sometimes, and just blipping the gas pedal would cause it to drop down to a more normal 800rpms. If the engine was in the higher idle when I shut it down, it would diesel (keep turning over without the car's ignition) until I stalled it with the clutch. So, you could simply avoid it by tapping the throttle to get the correct idle before shutting the engine off. Maybe this will work for you.
posted by knave at 7:06 AM on December 19, 2007

My old US-model 1991 Civic 1.5L hatchback was definitely fuel injected, not carbureted. OP's profile says Australia -- the version sold in the southern hemisphere may have had a carburetor, I don't know. puddpunk's post suggests that it did.

For future reference, if the car has a manual transmission one way to stop a runaway engine is to stall it intentionally. Pull over to a safe place with nothing in front of you, then put the transmission into a high gear and let the clutch out with your foot on the brake. Sure, it'll take a few thousand miles of the clutch's life but that beats an engine fire IMO. (Can't help you if you've got an automatic transmission, I avoid those things like the plague.)
posted by harkin banks at 10:23 AM on December 19, 2007

as has been mentioned above the problem is classic dieseling ... perfect description ... my old VW Bug used to do something similar.

traditional fix is to clean up the pistons, cylinders, valves and exhaust ports of all the gunked up carbon deposits.
posted by jannw at 12:03 PM on December 19, 2007

Hmm, sounds like there is another issue at work here.

You said your car kept running for awhile after you turned off the ignition, that means that the insides of your cylinders were so hot that the gas was igniting without any spark from the spark plugs. Also, your temp gage said the car was extremely hot.

I'd take a look at the water pump, heater hoses and radiator. Your cooling system could have failed.

That doesn't explain the open throttle though, but that could be a weird coincidence.
posted by sideshow at 2:57 PM on December 19, 2007