I have a 01 GrandAM 3.4. I have a reoccurring problem with this car. The car will randomly not start, and die while driving. Some days I can drive the car with no problems, and some days it will only have starting problems. These problems are very random. When the car doesn’t start it will crank and turn over however the fuel pump will not kick in. You can turn the key on and off several times, and it will fire up. This DOES NOT seem to be a problem with the passlock theft system. The security light has never came on, and you do not have to wait any amount of time for the car to start, you only have to flip the key back and forth several times. The car after running will occasionally die, after the car dies it will usually start back up on the first try. I am very puzzled by this problem not sure if it is the ignition switch, fuel pump, or something in between. Any help greatly appreciated.
Sounds like a potential fuel pump…but as you probably know, it could be many things.
One way to test the fuel pump is to bang on the bottom of the fuel tank, causing it to operate, or operate better. You might try attaching a fuel pressure gauge, and if it reads low, try the banking on the tank to see if that changes pressure.
You can sometimes tell if an engine dying is related to ignition or fuel by taking note of how it dies when it does. If suddenly, without a sputter, as it does when the ignition key it turned off, then it may more likely be ignition related. If it sputters, looses power, bogs down, etc, either just before it shuts off or even in general; this may indicate a starvation for fuel.
If a car has like 200,000 miles on it, it is kind of nice to just replace the fuel pump. The pump is a wearable item, and if you consider that it will need to be replaced at one point in the cars life (assuming the car won’t drive 400k), it is nice to just do it on your own terms…if you care for my two cents…
Take Care,
Jeffrey
I have tested the fuel pressure, the pump has perfect pressure when it is running. When the pressure tester was hooked up while running the pressure did not drop until shut off, don’t remember the complete bleed off time, however I remember that it did not seem like there were any issues with it. The fuel pump is around 200$, and not a fun job so replacing the pump will be last resort. When the car dies it seems as though it was shut off with the key, not as though it is being fuel deprived.
Oh, and also, to determine whether it is the fuel pump or the power getting to the fuel pump; if the car doesn’t start/fuel pump doesn’t turn on, WHILE this condition exists, you can test the wire the enters the gas tank to see whether or not it has 12 volts going to it…and/or disconnect the wires going to the fuel pump and apply your own power to it via a jumper wire from the battery.
If the pump works with your power applied to it, then the issue is somewhere in the circuit leading to the pump.
Car repairs in general are often diagnosed by isolating the issue by breaking it down into components; in this case, if the issue were determined to be in the circuit that powers the fuel pump, then that circuit would be broken down into components such as wiring (eg wiring from key to computer, from computer to relay, from relay to pump), or certain components such as the relays (relays are easily tested because there is usually a similar one elsewhere on the vehicle which can be swapped, then if the issue “follows” the relay, it is a faulty relay causing the issue).
Always start with the easiest things first…unless you are “unencumbered by the thought process”… in which case you would proceed down that path. 
-Jeffrey
This problem can also be caused by a faulty crankshaft position sensor.
If the computer doesn’t receive a signal from the crank sensor when attempting to start the engine or while the engine is running, the computer see’s no reason to operate the fuel/ignition systems so either the engine won’t start or shuts down at random times.
Tester
Thanks for the info, if I am not mistaken there are 2 crankshaft position sensors on this vehicle. Do you know which of these would cause this problem, or would it be both?
I’ve never heard of what Tester said. The fuel pump on many vehicles initially turns on with the key in the “ON” position, before the engine starts and the CPS has given an signal to the computer.
A fuel pump can be totally intermittent, showing full pressure then failing…so a low pressure result may be indicative of a poor pump, but a good test doesn’t necessarily eliminate the possibility of a failed-but-intermittent pump.
The pump for this car list at 229$ is there any way to test the pump to decide if it is or is not the pump?
There’s one crankshaft position sensor and one camshaft position sensor.
The crankshaft position sensor informs the computer whether or not the engine is rotating and also detects for misfires.
The camshaft position sensor informs the computer when the fuel injectors/ignition coils should fire.
If the computer doesn’t detect crankshaft rotation the engine won’t operate.
If the computer doesn’t detect camshaft rotation, the computer will operate on a default value for those lost signals and go into a limp mode. The engine will still run, but very poorly.
Tester
+1 to Tester, this sounds like a crankshaft sensor going bad
Thanks for the help. This is the sensor that goes into the back of the block correct? When I go to order the part 2 come up a Crankshaft Position Sensor, and Engine Crankshaft Position Sensor. Part # : CSS23 and Part # : CSS16 are the two part numbers.
Hard to say what’s the actual cause, but getting back to the fuel pump idea, the fact that the fuel pressure tests ok at idle doesn’t necessarily imply the pump is working correctly. If there’s significant leak back into the gas tank through a faulty check valve for example, you might find the fuel pressure is ok then, but the engine sputters or dies when accelerating from a stop, as the fuel pressure could drop in that situation. Not enough fuel going towards the engine in other words, it’s leaking backward. And fuel pressure needs to be tested both at the pump and the fuel rail. The more critical measurement for a well-running engine is the fuel rail pressure.
There should be a spec for how much loss of pressure decline is allowed in the rail fuel pressure after the engine is shut off. Did your measurement comply with the spec? If not, that would be a clue that the pump check valve is faulty.
One quick thing a DIY’er could try is to remove the vacuum hose connected to the fuel pressure regulator. Engine off, follow all working-with-gasoline safety procedures. There should be no gasoline at all inside that hose. If there is, the diaphragm inside the regulator has probably sprung a leak. That could cause a symptom like this.
The pressure was tested at the fuel rail, and the pressure drop off was within specification. The pressure has been tested multiple times. The car does not sputter or show any signs at all of loss in fuel pressure, or loss of fuel. When the car dies it dies, as though you flipped off the key. Fuel regulator was the first item checked no leaks or signs of problems.
At any of those times that the fuel pressure was tested…did the engine die. If it never died while doing the pressure test…the test is void and useless as eliminating the pump as the problem
Had the engine died and you noted that the pressure was still within spec’s…then you could eliminate fuel pressure as the problem.
All that said, I think @Tester is right and it’s as faulty crankshaft sensor that is affected by heat.
Yosemite
Two crankshaft position sensors are listed for this engine. One mounts behind the harmonic balancer. The other mounts in the side of the engine block.
I doubt if it uses both. I’m guessing it’s just a variation in engines. You would have to look and see which location your engine uses.
I am still pretty leery of this CPS (Crank Sensor) idea. I am pretty sure (without knowing the specifics of that model) that the CPS has nothing to do with the operation of the fuel pump. As noted in his post, the sensors control the spark/fuel injection, but not the operation of the fuel pump. Again, the fuel pump on most cars turns on with the key in the “ON” position, BEFORE the key ever turns to the “START” position to crank the engine–meaning that if those sensors detect engine rotation, they are not active when the key is in the “ON” position. Most/all fuel injected vehicles prime the system for about 3-5 seconds when the key is turned to “ON”, for easier starting. In fact, Triumph motorcycles will not start for about 3 seconds after the key is turned on, while the fuel pump primes the system to build pressure for startup. It is common practice for some persons to do this manually; turn the key to “ON” for a moment before cranking the engine just to help it start. I can all but guarantee that the issue is not a CPS.
This isn’t relevant to your car, but for reference to other readers; there are some primitive motorcycle electronic ignitions which need the engine to rotate at least once before starting, sending a signal to the computer to let it know that it should turn on, before they fire on the next cycle.
Other than testing the pressure as you have done, the fuel pump can be tested by simply listening to see if it operates when the key is turned to “ON”. There should be a buzzing noise for a few seconds as is the case with most cars. Alternately, since you are understandably avoiding replacing the pump if not absolutely needed, and if you have more time than money to invest in diagnosing the issue, you could run a jumper wire (fused) from the battery directly to the fuel pump. The fuel pump wiring is probably accessible through the opening in the trunk and if you were to run the wire it should be disconnected from the car’s wiring before applying your own power to the pump so that it doesn’t charge the rest of the car with your jumper wire power by feeding back through the system.
If you cared to deal with it, and since the issue is intermittent, you might keep that jumper wire attached while you drive the car for a while. You could put a toggle or relay switch in the circuit and turn it on while the car is running, verifying that the pump has power going to it at all times when the car is in operation. If the pump stopped buzzing/working at any time and you know that it has power going to it, then you can be sure that the issue is the pump (or the off chance that the short wire in the gas tank).
An intermittently failing pump could be the problem. The pump works fine while you do the testing,then later on the road it dies again. I’d hgook up the pressure tester and take it out for a ride…hoping that it would die. If it did and the pressure showed -0- psi as it died you would know it the fuel pump/ wiring or relay.
But @Husqv, an intermittently failing CPS could also cause the engine to die…if the pressure is really good all the time.
Yosemite
The CPS may well affect the fuel pump. On my late 70’s VW Rabbit for example there was circuitry so the fuel pump would only run if the engine was rotating and the points were opening and closing. No engine rotation detected, no enabling of the fuel pump. Don’t know if it applies to your vehicle, but I’d be surprised if something like that wasn’t in the design for most vehicles. It’s a safety thing, to prevent the fuel pump from running and pumping gas all over the place in the event of a crash.
I know a failing CPS could cause the engine to stop. A fuel pump could cause an engine to stop instantly as well, just as an electrical issue would. The reason I asked that question about sputtering when it stopped, is that if it sputtered then it might indicate a fuel issue, while instant stop could be both fuel and electrical. …Like an oak is a tree, but not all trees are oaks; we could ask certain questions to see if it like had leaves, indicating a deciduous tree (not an evergreen), but not necessarily an oak.
The reason I dismissed the CPS is that the fuel pump didn’t operate when the key was turned to “ON” (but not cranking in the “START” position). I don’t know the specifics of every application, but as noted, the CPS doesn’t affect the fuel pump when the engine is turned “ON” and the fuel pump should pump.
If a fuel gauge is handy and can be left hooked up, I had thought of that idea of keeping it on when driving to see if it drops when the engine stops…but that isn’t very reasonable, even if the best way to check if the fuel pump is the issue.
Please let us know if/when it is fixed, what the cause was.
I got under the car yesterday near the rear end and found a wire hanging down I started the car and pressed it to the body, and the car died. I then locate the f/p injr 15amp fuse and it was blown switched it and the car started. I could not locate what this wire goes to so I taped the bare end and taped it up to not drag or short out. I drove the car for about 20 miles, after the car sat for about 15 minutes I got in to drive and the car died again this time I was driving straight, at about 40 mph with my foot on the gas. When it died it was as though I had turned the key off. The car then re-started, and I drove it another 40 miles with no problems.
Any ideas as to what this wire could go to?