Car shuts off when warm, white smoke from exhaust

Maintenance/Repairs
21

A bad crank sensor will stop your engine from running like turning off a switch. However, this will happen at any RPM, not just when you take your foot off the gas. Although admittedly, a weak sensor might tend to work passably at higher RPM. But the engine would probably run pretty poorly at all RPMs. I tend to doubt this is your problem.

When you say a “good amount of white smoke”, how much are we talking about? What does it smell like? Do you live in a cold climate? If so, is it possible what you’re seeing is just normal steam from the exhaust that you will always see on a cold day? Or is it a lot more than that?

What you can do is rev it until you see the clouds of smoke or steam coming out, then shut it off immediately and pull the spark plugs. If one is very clean and shiny, may even be wet, you do indeed have a bad head gasket. A bad head gasket can cause the rough idle too. Your car is not getting vapor lock. Fuel injected cars do not suffer from this, especially not in winter.

22

Hard to describe how much smoke or steam. When idling none really. When I hold throttle down there is definitely smoke present. How about this - if my head was cracked - would it cause the exact symptoms I am seeing? Specifically, car will seemingly without overheating until you take off throttle when it’s at operating temp, in which case it will instantly die and cannot be started for 10-15 minutes (until it get’s a bit cooler)?

I don’t need to wait until it completely cools down. If temp gauge reads midway, just letting it dip slightly below that will allow me to turn the car over fine. Also, I am just not seeing any oil in my coolant nor coolant in my oil and I’ve been looking religiously as I am testing things.

The throttle body was cleaned, IAC cleaned and PCV valve cleaned and the bearing moves easily.

What could this realistically be if not a head gasket? Mainly, the running until warm stall.

Coolant temp sensor, spark plugs, plug wires, ignition coil… what else?

23

Head gaskets can fail in many different ways. You can even have one that holds pressure at idle but leaks when under more stress, such as when accelerating. I would suspect that you do have a bad head gasket. A cracked head as you suspect could also cause these problems. The TPS should generate a code when unplugged–at least it has on every car I’ve tried it on, though I’m not familiar with how Hyundai’s system works.

24

You are not loosing any coolant though? If the white smoke is caused by coolant, then the coolant level must be dropping. Unless oil or transmission fluid is replacing it. But I’m assuming you’ve looked at the coolant and there’s no obvious oil or transmission fluid contamination. Over how long of period have you monitored the coolant level? One week? One month? If there’s any path from the cylinder to the coolant, there major flow will be from high pressure to low pressure areas. That means there should be exhaust gasses getting into the coolant. Are there any bubbles in the coolant when you look in the radiator while the engine is running? After you “juice” the engine a bit, see any bubbles in the coolant? There are inexpensive chemical test kits you can buy at auto parts stores that will give you a definite “yes”/“no” to whether there is exhaust gas getting into the coolant. Might be worth doing in your case. The critical issue here appears to be whether there’s a cracked head or a head gasket leak or not.

25

My intake manifold gasket is leaking in several spots and code P0121 (TPS sensor, voltage differential between it and MAP sensor code). That is what I found. Jive with my symptoms?

26

Mainly the car refusing to start when warm after it stalls. What would a leaky intake manifold gasket do to cause that?

27

I would think it would cause it to run like crap. Get it sealed and try again. Clear your codes and see if you get the same codes.

28

I will be relacing the intake gasket tonight. I will report back to bring closure to this thread. I have found anecdotal reports of others with my vehicle getting the same TPS code with a leaking intaking manifold gasket. Also, this gasket is not just leaking - it is blown out on both sides.

29

Now-a-days an engine is supposed to be a totally sealed system. Air is supposed to get into the combustion chamber only certain paths. Those paths are typically the main air intake through the throttle plates, the IAC path, and, this part varies from car to car, various paths for increasing the idle speed under certain conditions, like turning the steering wheel or turning on the headlights or engine fan or rear window defroster. If air gets into the engine in other ways, this really confuses the ECM (computer). It can’t figure out how much gasoline to inject to maintain the desired fuel to air mixture. Poor idling, surging, poor or no-starts, and generally poor drivability result.

30

MAP means Manifold Air Pressure sensor. With a manifold leak you will get a MAP code.

31

The “A” in MAP stands for “absolute” rather than “air”. A car’s MAP sensor uses a silicon sensor to measure the intake manifold’s absolute air pressure relative to a total vacuum. This makes it possible to use the MAP as an altimeter, so the car’s computer can compensate the fuel/air mixture as altitude changes, like when driving in mountains.

If you happen to come across a discarded MAP sensor, check to see if it works before throwing it away. If it works you could use it as a DIY altimeter, or as a barometer to predict the weather.

32

Fixed gasket, idles much better, still shuts off when warmed up and won’t turn back on. If you don’t let it cool down a bit it will stall rather quickly when you finally get it to start.

33

Also, no codes.

34

Couple of ideas for when engine run poorly or stalls when warm.

  1. The EGR isn’t allowed to function when the engine is cold. Either the computer blocks it, or there is some coolant sensor specific to the EGR, sometimes it works by blocking a EGR vacuum port when the engine is cold. But as soon as the engine warms up, then the EGR is allowed to function. But even when it is allowed to function, it is only supposed to open up and allow exhaust gas into the intake manifold in certain specific situations, like during rapid accelerations. That’s b/c that is a situation in which nitrogen pollutants can happen. It isn’t supposed to open up at all during idle. In your case there may be something wrong w/all that, and it s opening up as soon as the engine warms up, even during idle, which would likely cause the engine to stall. If there’s a way to temporarily completely disable the EGR from functioning, try that. Could provide a clue.

  2. As mentioned above, the O2 sensors come on-line only when the engine is warm. They could be failing, or their signal mis-interpreted by the computer. The car-specific scan tool is probably needed to diagnose this.

35

You need to test the TPS. You either have a TPS problem or wiring problem. Testing a TPS is a very simple thing to do even if you just have an electrical meter. (If you have access to a scantool that’s even easier). Most are 3 wire. One is ground, one is a 5V reference signal from the PCM, the third is the signal wire that returns voltage corresponding to throttle position back to the PCM. Backprobe the connector. Check for the 5V ref voltage. If that’s good, check the signal wire. It should be in the neighborhood of .5V with the throttle closed. As you open the throttle voltage it should steadily climb to about 4.5V at WOT.

36

I stand corrected. If I could only keep all the acronyms straight I might have a chance. Car, computers, aircraft, flight, it just goes on and on.

37

Car runs much better with the new manifold gasket but still the same problem. Here is my best description. The car is not being driven at the moment. It is stationary.

Car starts cold, idle is good @ 1200 rpm and slowly drops to a steady 750 rpm. I give it throttle @ 2500 rpm to heat it up faster. The car runs fine idling or giving throttle UNTIL the temp gauge in car goes to half. It will not act up or shut off until it reaches mid way. If I still give it throttle it will not stall but at this point the car will drop rpm even if I am steady on throttle. I.e. it will pulse from 2500 down to 2200. At this point I know the issue is underway and if I let the throttle off even a little the car will instantly drop like a rock from 2500 rpm down to zero and shut off.

When you try to start it it will crank and crank but not start no matter what. Give it 10 minutes for the temp gauge to drop a bit and it starts right up but will stall much faster as it’s still warm.

I took the crankshaft sensor out and put it back in. I was able to get the car to throw a P0336 (cks) code soon after reinstalling it but not sure if that is the problem or just a symptom of me removing and reinstalling the sensor.

I am no longer getting the TPS code since replacing the intake manifold gasket and perhaps was getting it to begin with because I was holding throttle down and when this problem starts the car goes from high rpm to a dead stall instantly. Unsure on that.

I am going to begin by replacing the crankshaft sensor then coolant sensor if that does not fix the problem. I have new plugs, plug wires and ignition coil on the way. I do not want to “throw parts” at it but in all honesty this problem is costing me money being out of work and if I don’t fix it I will end up at a garage and will be fleeced for more $ than the $150 I’ll be spending on parts.

Searching google I get tons of the same issue with many solutions from crankshaft sensor to fuel pump issue when warmed up.

Any other ideas welcome.

2003 hyundai doesn’t have an EGR.

38

The fact that this problem can be reproduced and it temperature sensitive is a major clue which should make it relatively easy to diagnose. Maybe a little more info would be helpful …

Which engine do you have on your 2003? 1.5, 1.6, 2.0, 2.4, 2.7, or 3.5?

39

This is the 1.6 DOHC.

40

Let’s see what it says here about the 1.6/L4 … hmmm… it appears you have a PCV, but no EGR (as you already said), a single 3-way cat (for CO, HC, and N), EI/ECITS spark control, and 2 heated oxygen sensors. Have you replaced the O2 sensors yet? Not suggesting that you do, but wondering if you have already.

You say you aren’t getting any codes, but that may be b/c you’ve cleared them by disconnecting the battery, and they haven’t had a chance to return yet. Did you in fact clear the codes or disconnect the battery? The problem you are having is so severe that it’s hard to imagine there wouldn’t be at least one code thrown.

Ok, well not all the data available yet, with no codes reported at all, so time for some guess-work. It sounds to me like one of these four would be where I’d start with my suspicions for the guilty party

  1. O2 sensors on the fritz
  2. Fuel pressure problem
  3. Exhaust leak prior to the O2 sensors
  4. Spark problem, quite possibly the crank sensor, but could be the ignition module too

The reason I say these 4 is that all four typically exhibit temperature related symptoms. While it could be other things, for example I wouldn’t expect such extreme temp sensitivity for the MAP or TPS.

Unfortunately none of these are particularly easy to test without the proper testing equipment & shop manual info. hmmmm … well , you could get a length of garden hose I suppose and use it to probe for a small exhaust leak prior to the O2 sensors. It would be making a putt-putt or pop-pop sound near where the leak is emitted. A cracked exhaust manifold isn’t at all an uncommon thing, esp with these smaller engines. The crack can be closed when the engine is cold, then open up as the engine warms.

& you could test for spark at the spark plugs when this is occurring. There are in-spark-plug-line gadgets you can buy at auto parts stores that make this relatively simple.

Checking the PCV is simple so you might as well try that too.

This kind of drivability problem is often better solved by someone experienced and with the car’s scan tool. Like the dealership shop. I sense you don’t want to go that route. But you might could be dollars ahead by considering this, at least have someone with the proper tools and experience do the diagnosis, then you do the parts replacing.

Sorry, that’s the best this amateur DIY’er can offer up at this point. Best of luck.