Drove 91 Merc Capri through puddle, rough engine, gas smelling smoke, fuel loss


Yesterday evening I drove through a flooded area of the road, probably about 1.5 feet deep. I did not realize it was quite that deep. The water came through the floor, and probably soaked my ECU.

The engine died, but will idle very rough. I doubt I got water in the cylinders, not much at least, because the engine turned over no problem. I also have an air filter that would be difficult to get too much water into from the way it is situated.

It is putting out white smoke that smells like gas (a lot initially, now not as much), and the engine is smoking a little as well…white, gas odor…I think it is coming out around the exhaust manifold.4

It is running very rough, sounds like from 2 cylinders, and after letting it idle in the hopes of drying everything out, I noticed that the fuel gauge is at empty…it was at 3/4 earlier.

What are your thoughts? What should I start by doing?

The fuel gauge reading is likely in error due to water in a connection or wires. The motor issues could be multiple. Water contacting a hot exhaust manifold could crack it for instance.

It smells like I am dumping a lot of gas though… could it be water in a connection causing one or more of the injectors to stay open? Or water in the ECU causing some other fuel problem?

When you drove thru that deep puddle it submerged the catalytic converter, and this may have caused the substrate in the converter to break apart from thermal shock.

If this is what happened, the broken substrate may now be causing an exhaust restriction.This would cause the white smoke and a gas smell out the exhaust because the engine can’t take in enough air to burn all the gas in the cylinders, it would cause the engine to run rough, and if the restriction is high enough it can blow out the exhaust manifold gasket.

Try rapping on the catalytic converter with a rubber mallet and listen for something rattling around inside the converter.


I think the exhaust system is very likely a big part of your problem(s). As @Tester stated the cat could be damaged and restricting flow. On the flip side the manifold maybe cracked or come apart which would be noisy and reduce back pressure in the exhaust. Both restricted flow and excessive flow (to little back pressure) would make the motor run “rough”.

Most of the electrical and fuel injection components are mounted high on the motor and are less likely to get wet, but if you had water spraying all over the engine compartment that means everything got wet and puts more potential problems into play.

I let it dry out, and it idles fine now. However, when I drive up hills, the engine starts fluttering and loses power. Sometimes it will catch randomly and run fine for a few seconds at full power.

I changed the fuel filter, but that made no difference. I am not blowing any smoke out the exhaust, but there seems to be an oil leak onto the exhaust pipe, because the exhaust pipe starts smoking pretty badly directly under the engine after it gets hot.

Any idea on where to start? Would a change of plugs and wires be a good starting point? Or would that not possibly be a cause?


Pull the distributor cap off and look for moisture. Moisture in the cap can allow for cross-firing and misfires.

My suggestion would be to replace the plugs as a start and check the plug wires and coil wire on both ends for any signs of moisture or corrosion on the wire ends. It might be a good idea to remove the distributor cap and check for moisture inside the cap.also.

I’ve seen some Capris get water down into the holes where the spark plugs reside. One of them only ran on two cylinders. Can you blow some compressed air into the plug holes to see if water blows out? You need to do that before pulling a plug out anyway, to keep any possible water out of the cylinders.

If , as Tester suggested, youc catalyst grid inside your converter has cracked, it will fall back against the outlet pipe and plug it when you go uphill.

Even if the distributor cap is dry now, if you got any moisture in it it might have formed carbon tracks that can act like high resistance circuits. I’d just go ahead and change the distributor cap, rotor, wires & plugs.

From there you’re going to be looking for two things initially;
(1) erratic spark, which can be tested for with a scope (ideally) or even just with a timing light (not the best way, but a gross inconsistency will show up).
(2) the ability of the engine to breath out… through that cat converter. The ceramic core does not stand up well to thermal shocking. That can be tested for with a vacuum gage. If the cylinders are having a hard time purging themselves of exhaust gasses, they’ll be unable to develop any real vacuum. A kit should come with instructions for use.

Those cars are low, and if the intake snorkel is typical it’s taking air from the front of the car at a point low enough to have ingested some moisture, although fortunately not in enough volume to cause hydrolocking. Ingesting moisture could also have thermally shocked the mass airflow sensor in the same manner as the cat converter. The MAF sensor contains a heating element, and shocking it with cold moisture might have damaged it.

Another possibility is the crank position sensor. That probably got submerged. It isn’t subject to thermal shock damage, but also isn’t designed to be submerged.

The good news is that
(1) it didn’t hydrolocck, and
(2) the ECU I think is located above your knees, high enough that it shouldn’t have gotten wet. I could be wrong on this, but I don’t think so.

That means the possibilities are all repairable.

Oh, and if it were mine, once I got it running I’d change all the fluids and filters too.

If any issues crop up with wheel bearings or steering/suspension components keep in mind that driving through deep water could have played a part in those items going south.

Thanks a lot for the advice! I will replace the plugs and check the distributor as well. Could someone explain why spark issues would only appear under load, and not when revving the engine?

Because your engine needs a good strong spark when under load. It’s a consequence of packing more air/fuel mix into the cylinder, which affects the amount of voltage necessary to create a spark (an arc) through the compressed fuel.

Honestly, rather than checking the distributor cap and rotor, I’d just change them. A carbon track can be hard to see on the black plastic they use for the caps. Unless, of course, yours is that red brick color they often use.