Mechanics

WHY DOES MY TRUCK START, THEN IMMEDIATELY DIE?

2000 Dodge Dakota • 150 miles

The fuel pump can be heard going on, so I really don't think its that. I will start my truck, it will turn over, sputter for a second, and then immediately die. The only way I can get the engine to stay firing is by pressing the accelerator, but once I let go for it to idle, it is dead. I'm thinking this could be the MFA or O2 Sensor, but I'm not sure. What are your thoughts?
Avatar
Bdanehey
January 1, 2012.




Sounds like the battery might have been disconnected or run dead recently. If so, the Engine Computer lost its memory and has to relearn "minimum throttle" before it will know when it must be in control of idle speed. It also might not give you the normal "idle flare-up" to 1500 rpm when you start the engine. To meet the conditions for the relearn to take place, drive at highway speed with the engine warmed up, then coast for at least seven seconds without touching the brake or gas pedals.

Caradiodoc
Jan 1, 2012.
The oxygen sensors aren't even in the picture until the engine is warmed up. Chrysler never used the troublesome mass air flow sensor on their domestic vehicles. They are the only manufacturer that could make their engines run right without one.
AD

Caradiodoc
Jan 1, 2012.
How am I supposed to drive the vehicle when it won't start? It has to be a constant depression on the accelerator for the engine to keep going. The second I let go, the vehicle stalls. Theoretically, I could keep it pressed for the duration of my journey, but this would put a heavy toll on the transmission when reversing and stopping.

Tiny
Bdanehey
Jan 1, 2012.
You're asking how to do a standard part of any test drive after your mechanic had to disconnect the battery. You only have to hold the gas pedal down 1/8" to keep the engine running. The intent is to keep the engine at normal idle speed, that's all. The transmission is designed to take that.

The Engine Computer needs to know when your foot is off the gas pedal. At that time it memorizes the voltage from the throttle position sensor. Any time from then on that it sees that voltage, it will know it must be in control of idle speed. It knows your foot is off the gas pedal by the high manifold vacuum of coasting, and it knows you're coasting when that high vacuum exists for at least seven seconds. That's why simply snapping the throttle won't do it. That will give a burst of high vacuum but only for one or two seconds.

I post this procedure at least a dozen times per week, and no one has ever said they couldn't get 'er done. Most people are surprised the fix was so simple. If you can't control engine speed to a safe level and drive normally at highway speed, you may have some other issue such as a misfire condition. Even with misfires, the automatic idle speed motor has enough control of idle speed to keep a V-8 engine running with 6 cylinders disabled! I watched a Chrysler instructor do it. It obviously won't run well, but that shows how much control it has. It can easily correct idle speed that is too low due to one or two misfiring cylinders. For that reason, idle speed that is too low is always due to a need for the relearn procedure or the idle air passage is plugged with carbon. Carbon hasn't been a problem for the last 15 years with better fuels.

When My '88 Grand Caravan developed a problem with the idle speed motor, the engine stalled at every stop light unless I held the gas pedal down just a little. I ignored that problem and drove it like that with two feet for two years and over 15,000 miles. I'm pretty sure you can do one test drive. On the off chance it doesn't solve the problem, you'll need a scanner that can display live data to analyze the operation of the idle speed motor.

As a final note, you should be grateful to Chrysler that they made this fix so simple. Had you disconnected the battery or let it run dead on a Volkswagen, we'd be discussing an engine that might not start at all, and if it did, it wouldn't come out of park, and engine speed would not increase even when you pressed the accelerator all the way to the floor. The car would have to be dragged onto a flatbed truck for a trip to the dealer for them to unlock numerous computers and reprogram "minimum throttle". Volkswagen and General Motors are among the least customer-friendly manufacturers in the world and have designed in many ways to separate owners from their money. In my opinion, Hyundai, Toyota, and Chrysler are the most customer-friendly.

Caradiodoc
Jan 1, 2012.
I'm sorry but the trick didn't work. Surprisingly, when I went to start it. It actually started. I went on with the test, driving at 50-65 mph and then releasing my foot and not touching anything for 5-10 seconds. When I went back to my garage, it did not start. I can't figure this damn thing out! Thanks for your help though, much appreciated.

Tiny
Bdanehey
Jan 2, 2012.
The first thing to look at is whether you get a nice "idle flare-up" to at least 1500 rpm when it starts. If it does, and then comes right back down to around 800 rpm, the idle control system is working. From your original description, I suspect this is not the case.

The next thing is to connect a scanner that can display live data. Watch the "AIS Steps". The Engine Computer can command the automatic idle speed motor to any of 256 positions called steps. For a properly running engine, step 32 is about normal. If you find it at step 0 before starting the engine, the computer has not learned minimum throttle. You must coast for a minimum of seven seconds for that to take place. Five seconds won't do it.

If your truck has cruise control, the brake switch will have up to three sections. Even though the brake lights are not stuck on, a pitted contact can block the relearn from taking place. My Chrysler DRB3 scanner shows the status of the brake light switch so it's real easy to see if it's working properly. On some vehicles the part of the brake switch that kicks the cruise control off can block the minimum throttle relearn from taking place. You might get a clue if the cruise control won't set too.

If you find the AIS is at around step 50, it has been set to the starting point and should provide the idle flare-up. If you still have to press the gas pedal to get the engine started, one of two things is blocking the air flow through the passage around the throttle blade. That passage is blocked with carbon or the motor is not responding. Carbon buildup used to be real common on the 3.0L in the early '90s, but with better additives in fuel today, I haven't had to clean one in a long time. Other engines didn't seem to have that much trouble.

The Engine Computer constantly monitors the AIS motor and its wiring. It will set a diagnostic fault code in memory and turn on the Check Engine light if it sees a problem. Besides that, the armature can become tight and fail to turn. The motor is not a regular motor that spins rapidly. It is a metal core that is placed in various positions by pulsing current through four electromagnetic coils. As it turns to those positions, a screw thread extends or retracts a pintle valve to expose more or less of the air passage. At the same time as it opens that valve to let more air in, it also lengthens the on-time of the injectors' pulses. More air and fuel increases the idle speed.

Many scanners have a test mode for testing the AIS motor operation. By pressing the right sequence of buttons, you can tell the Engine Computer to bump the idle speed up to 2000 rpm in 200 rpm increments. If that fails to occur, suspect the AIS motor itself.

Another thing you can do without a scanner is to remove the AIS motor from the throttle body, then watch what it does when a helper turns the ignition switch to "run". You should see the valve retract. If it does not move, squeeze the valve to force it in. It's going to take both hands and lots of force, but it will move. The total length of its travel is about 1/4". If you leave the ignition switch on while you reinstall the motor, then turn it to "crank" without turning it off first, the engine should start and have a really high idle speed. If the idle speed doesn't come down within about five seconds, suspect the motor is defective. There are four coils of wire inside that are all connected together. If only one has a broken connection, electrically the entire motor will still appear to be okay to the computer, but it won't respond properly to the computer's commands. You can't test the AIS motor with an ohm meter because all four coils are interconnected. If one coil is open, the meter will just read through the other three. Also be aware the AIS motor is held on with two Torx bolts with thread lock on the threads. Your Torx bit must be the proper size and it must be fully seated to prevent rounding the heads off. New motors come with new bolts with that thread lock already on them.

If you turn the ignition switch off, then back on to start the engine, the computer will run the AIS motor more than enough to completely close the valve, then it will back it off to around step 50 in preparation for starting. That's why if you retract the valve by hand, then reinstall the motor, you must not turn the ignition switch off before cranking the engine.

It's always possible one of the drivers in the computer has failed so it won't adjust the AIS motor. That is extremely rare. I've never solved a low idle complaint by replacing the computer. About one out of twenty complaints were solved by replacing the AIS motor. The rest just needed a test drive and nice long coast.

Caradiodoc
Jan 2, 2012.
Alright I'll see what I can do. Maybe it did work, just not right away because it seems to start ( a little sluggishly with several cranks) all the time now, with the idle flare up to 1500 rpm. I'm probably going to take it in to the shop to have it looked at, but I'm happy it worked in the mean time. Thanks again, great advice!

Tiny
Bdanehey
Jan 2, 2012.
Here's something else to consider, but this only applies if the idle speed is remaining steady AFTER the engine has been running.

Fuel pressure will normally hold steady for weeks when the engine is off. In case it does bleed down a little, the fuel pump always runs for one second each time you turn on the ignition switch to be sure the pressure is up high enough for starting. That one second isn't enough time to build pressure if it had bled all the way down to 0 psi. Also, since the battery voltage is drawn down a lot while cranking the engine, the fuel pump will run slowly during that time and take even longer to build pressure. That can result in a long crank time, perhaps as much as five seconds. Any time you suspect that might be happening, turn the ignition switch on, then pause for one or two seconds before turning it the rest of the way to "crank". I do that out of habit with my '88 Grand Caravan, and it always fires right up. That pause gives the pump time to build the pressure needed for the injectors to spray fuel.

Three things can cause the fuel pressure to bleed down. A leaking fuel pressure regulator and a leaking check valve in the pump are the least common possibilities. A leaking injector is fairly common. None of those are serious, just irritating. Fuel leaking from the pump's check valve or the pressure regulator goes right back into the tank. Fuel that leaks from an injector vaporizes and will usually be burned in the engine when it starts.

This problem has nothing to do with maintaining idle speed but it can seem like it's related because the engine often finally starts about the time the driver tries pressing the gas pedal. If you don't press the gas pedal, that idle flare-up time has passed by the time fuel pressure builds sufficiently, so engine speed ramps up very slowly as that pressure builds. Giving it a little gas gets the engine running quicker. Remember though that once the engine finally starts, it will continue to idle properly.

I don't mean to add another layer of confusion, but you could very well have had two different things going on.

Caradiodoc
Jan 3, 2012.

AD