Cranks and cranks and cranks then finally starts

[MWR]

Here’s the problem with my ’04 Grand Cherokee SE 4.0L: About a month or so ago it started getting difficult to start. Instead of firing immediately, it would turn over half a dozen or more cranks before firing. Then last Tuesday it wouldn’t start at all. That day I replaced the plugs, checked for spark (which was good), and assumed no fuel so I changed the regulated fuel filter; started right up. Next morning, it wouldn’t start again. I discovered that after priming multiple times (can hear the fuel pump prime when the key is turned on) and turning over for about a minute or so, it would start. It would run rough for a few seconds, then idle fine. At this point let me say that it never stalled or shut off while driving, never ran rough, and showed no loss of power.

Still assuming no fuel (and doing some on-line research read that it could be the check valve in the fuel pump), I replaced the fuel pump. A very expensive proposition, as you know. It ran fine when I got it back, but the next morning it was back to the same thing; crank and crank and crank to get it started. A fuel pressure gauge on the fuel rail shows in the mid 40s for psi once primed, and bounces around 60 psi while running. Thinking a possible bad injector, I pulled the plugs again to check for a fouled plug; nothing. After replacing the plugs it started a little easier the next few times, than back to the same thing. After a few tries it sounds like it wants to fire, does that a few times, and then eventually sputters a few times, then finally actually fires and starts.

I checked my battery with a multi-meter. It was showing about 12.1 volts a few minutes after turning off. Since that is a little low and the battery was 3 ½ years old and showing a little swelling around the positive terminal. I just replaced it because I had read that Jeeps don’t like low voltage and it was on its way out anyway; still no change. Even though it doesn’t stall, run rough, or lose power, could it be the Crank Position Sensor? From what I am reading, this sensor can cause multiple problems when going bad.

Anybody have any ideas or suggestions?

 

[SteelHeadz]

Recheck your fuel pressure. With the gauge installed on the fuel rail, turn the key on(dont start the engine) and see where the pressure holds. It should be around 48-49psi average. Turn the key off and monitor the fuel pressure gauge for a bleed. If the gauge drops within 5mins, this would indicate a faulty pressure valve within the fuel pump module. The symptoms you have described along with this test will verify a faulty fuel pump module. I recommend replacing the entire module instead of just the pump within the module. The main reason for this is that the fuel pump module as a unit has integral components that regulates fuel pressure to the fuel rail as it is gettting pumped and holds the pressure in the rail as you power down the ignition.

 

[MWR]

I've already checked fuel pressure with new fuel pump installed. Pressure is good.

I've already changed spark plugs, fuel filter, entire fuel pump, battery (only holding 12.1 volts, needed a new battery anyway) and CPS. Still no resolution to the problem and I'm into it for almost $400 (not including the plugs and battery). At this point I'm thinking I should have just gone to the dealership. I probably have more invested than they would have charged me to repair it. Of course, hindsight is 20/20.

What I observed this morning, after changing CPS last night, is the temperature gauge going to about 175 degrees as soon as I turned the key on. The outside temperature here is 23 degrees. After sitting 12 hours, the block should have been very cold.

I'm thinking that my actual problem is simply the temperature sensor in the block has gone bad and is reading too high or too low resistance, depending on what the computer is expecting to see, and the computer thinks the engine is already warm. Therefore, the computer tells the throttle body where to set the choke for proper air/fuel mix and in this case is setting it too lean. Then it takes a good bit of turning over in order to build up enough fuel in the cylinders to start.

The reason this makes sense to me is that it is always hardest to start when the engine is cold. The colder it is, the harder to start.

The question is, what kind of resistance reading should I be looking for across the temperature sensor? I’m guessing that it should have a low resistance when it’s cold, and the resistance increases as it heats up. It this assumption correct?

Any thoughts on these observations?

Of course, if I had put a multimeter across the CPS, I could have determined that there was nothing wrong with it and saved a few bucks in the process. Again with the hindsight.

 

[MWR]

And the winner is, drumroll please, COOLANT TEMPERATURE SENSOR!!!

The indication from the temperature gauge (180 degrees before turning the engine over when cold) was my best clue. As it turns out, this sensor is a NTC (negative temperature coefficient) type, which means that a low temperature equals a high resistance. The ECU reads this resistance and determines from this data how much to choke the throttle body to make the fuel mix correct for the engine to start at the current temperature. My ECU was seeing that the engine was already warm, and therefore was making the mix leaner than necessary to start in the cold conditions that actually existed. The easiest way to verify this was to simply unplug the sensor, which creates an open in the circuit (this equals infinite high resistance) and telling the ECU that the engine is extremely cold. Because the ECU needs to see some resistance, even if it's very high, it sees a failure of the sensor and sets a code. Cleared the code and replaced the sensor.

The resistance of the new sensor at 70 degrees F room temperature is approximately 12k Ohms.

I cannot tell you what the resistance of the old sensor is because it actually fell apart when removing it. The probe end was smashed flat and perforated. The RTD inside was exposed directly to the coolant and the contacts were corroded inside. I would like to know how the end got smashed and perforated. This should not be able to happen when installing. Even if you screw in the sensor to the bottom of the threads it should not hit anything inside the housing. I can only assume that it was damaged before it was installed, which is before I purchased the vehicle.

The temperature gauge now shows the proper reading when cold, and the jeep starts like it should.

To all who contributed ideas, thanks.

 

[Bounty__Hunter]

Glad you got it figured out. Unfortunately the expensive trouble-shooting happens all too often.

I'm sure your descriptive follow-up with help fellow Jeepers with similar problems. Thanks.