need help from auto mechanic

[Ether]

Are there any auto mechanics in CD-land who would be willing to toss out some ideas on what could be the cause of this perplexing problem with my daughter's 1998 Chevy Lumina with 3100 SFI V6 engine?

The starter motor fails to crank under the following sequence of steps:

1) engine dead cold, then
2) start the engine (starts OK), then
3) turn the engine off after running approx 1 minute, then
4) try to start engine again and it will not crank. No clicking, no buzzing, nothing.

I slapped a voltmeter on the battery to see what was happening in step 4, and there is no drop in voltage when turning the key to the cranking position.

I tried two different keys (the keys have a security chip in them) and the problem exists with both keys.

Once the engine is warm (drive, say, 10 miles) the engine will repeatedly re-start with no issues.

The engine also always starts the first time when cold. The problem occurs only when trying to re-start as described above.

Once the engine fails to crank (as described above), I have to wait for a while* and then it will crank (and start) again.

I am at a loss for a working hypothesis. Any ideas what could be wrong?


* I haven't got a handle on the minimum required wait time yet. Three hours seems to be sufficient, but it may be shorter.

[dsherw00d]

Seems security or VATS related. The wiring and starter must be OK if it works most of the time other then that scenario. Maybe a bad sensor, but I don't know what sensor would cause a no start. Maybe oil pressure or something, but no crank seem extreme. I also think you would get a code.

When this happens, the security light is not on or flashing and there are no OBDII computer codes_

[Garrett.d.w]

I had something very similar happen to my car and it turned out to be the starter motor. That being said, the difference between a 66 mustang and a 1998 chevy is astronomical.

I'd recommend checking for an OBDII code. If there isn't one, look at the starter solenoid and starter. Work your way out from there (check related sensors, etc).

Good luck

[tuyauxtu]

Our Legacy's starter had a similar but likely not identical issue. A combination of brush carbon and cold Canadian winters occasionally resulted in inadequate armature movement, so starter engagement was intermittent but temperature sensitive. As a test, I'd suggest a couple of smart impulse inputs with a medium size hammer to the side of the starter frame. If the problem goes away that may be the problem. For us, this temporary condition "fixed" the starter for about one month. When I repaired it, there was carbon build up and worn brushes, one of which was slightly misaligned. Your mileage may vary. Now back to the kick off channel!

[Mr. Lim]

Originally Posted by tuyauxtu

Our Legacy's starter had a similar but likely not identical issue. A combination of brush carbon and cold Canadian winters occasionally resulted in inadequate armature movement, so starter engagement was intermittent but temperature sensitive. As a test, I'd suggest a couple of smart impulse inputs with a medium size hammer to the side of the starter frame. If the problem goes away that may be the problem. For us, this temporary condition "fixed" the starter for about one month. When I repaired it, there was carbon build up and worn brushes, one of which was slightly misaligned. Your mileage may vary. Now back to the kick off channel!

I was about to suggest this exact thing.

[Ether]

Originally Posted by dsherw00d

The wiring and starter must be OK if it works most of the time other then that scenario.

That was my thought process. I couldn't come up with a reasonable explanation why just that specific scenario - and no other - caused a failure to crank... and why waiting three hours would allow it to crank again.

[MrForbes]

I second the hammer suggestion. Worn brushes.

[Ether]

Originally Posted by MrForbes

I second the hammer suggestion. Worn brushes.

I hope you're right. Better that than an engine computer problem.

I'll give it a try tomorrow afternoon after the kickoff.

But in the meantime, can you think of (an even remotely plausible) reason why this happens only under the conditions described in the OP ? That's the part that's stumping me. I've repeated this about half a dozen times over the course of the past week and it always follows the described pattern.

[DonRotolo]

Let's approach the problem systematically. Sorry for the long post.

We'll discuss the time-related aspect of this in a moment.

We know* the ignition switch, when turned, sends a 12 volt signal that eventually ends at the starter solenoid, powering it and engaging the starter. In a '66 mustang this is a bigger wire that leads through the neutral safety switch (NSS) and then to the starter. In a modern car with an 'immobilizer' system, it is more likely like this: (from Autozone.com)

The starting, or cranking system consists of the battery, starter motor, ignition switch and related wiring. These components are connected electrically. When the ignition switch is turned to the START position (and the theft protection module recognizes the key code, as equipped) battery voltage is applied to the starter solenoid (through the theft deterrent relay, as equipped) S terminal and the solenoid windings are energized. This causes the plunger to move the shift lever, which engages the pinion with the engine flywheel ring gear. The plunger also closes the solenoid contacts, applying battery voltage to the starter motor, which cranks the engine.

We then see this info for testing the starter:

Disable the ignition system by unplugging the coil pack. Verify that the vehicle will not start.

Connect a voltmeter between the positive terminal of the battery and the starter B+ circuit.

Turn the ignition key to the START position and note the voltage on the meter.

If voltage reads 0.5 volts or more, there is high resistance in the starter cables or the cable ground, repair as necessary. If the voltage reading is ok proceed to the next step.

Connect a voltmeter between the positive terminal of the battery and the starter M circuit.

Turn the ignition key to the START position and note the voltage on the meter.

If voltage reads 0.5 volts or more, there is high resistance in the starter. Repair or replace the starter as necessary.

OK, if the starter and solenoid are OK, you should then trace the starter solenoid wire from the ignition switch, through the NSS, to the theft-deterrent relay. Then from there to the starter, after checking the relay itself (just the electromechanical relay, not the electronics).

OK, what's one minute have to do with it? Heat. Something is warming up to a critical point and failing. A can of freeze spray might be valuable here. My personal bet is either the NSS, or maybe the relay itself. Jumper wires can help isolate it...

In one car I owned, the ring gear sensor (a magnetic sensor that detects the starter teeth on the ring gear at the back of the engine, to allow the motor controller know the engine's RPMs) was flaky. At a certain temperature, it would fail - shutting the engine off as if the key was switched off. Waiting about 1 hour allowed it to cool down, and the engine ran again. For a while. Replacing the sensor solved the problem.

Really, you need a wiring diagram of the starter circuit to properly trace the start signal from the switch to the starter. Do that and you'll find the problem....when it is happening only, of course. I found a diagram on the AutoZone.com web site, it is attached below.

As you can see, the theft-deterrent module supplies a ground to the relay.

OK, get out the multimeter and have fun!


*maybe

[DonRotolo]

If the above gets you nowhere, how many miles on the car?

[Ether]

Originally Posted by DonRotolo

If the above gets you nowhere, how many miles on the car?

It's a southern car with no rust and only 158,000 miles. It's still a pup. My daughter put 30,000 miles on it last year; no problems.

Not time to scrap it out yet.

[Mr. Lim]

Originally Posted by Ether

I hope you're right. Better that than an engine computer problem.

I'll give it a try tomorrow afternoon after the kickoff.

But in the meantime, can you think of (an even remotely plausible) reason why this happens only under the conditions described in the OP ? That's the part that's stumping me. I've repeated this about half a dozen times over the course of the past week and it always follows the described pattern.

When the starter heats up and cools down, parts of it expand and contract. With worn brushes/armature/contacts/carbon deposits, that expanding and contracting could be creating just enough clearance between contacts such that your armature loses electrical contact. Letting it cool allows things to contract, and the clearance disappears. That's why whacking it with a hammer is an effective test.

[sanddrag]

Had an electric radiator fan with similar symptoms and cause as above. It's worth a good thwak to find out.

[dsherw00d]

A quick test would be to check the 12v energizing wire going to the starter during the no start condition. In older Chevy's it's purple. If this isn't getting 12v when the key is turned, then the computer isn't sending 12v out. Another quick test to see if it's the starter locked up is to turn the headlights on and try to crank. If the juice is going to the starter and not turning it, the lights will dim. If they don't dim, then I would say no power is getting to it.

[MrForbes]

The computer is not involved in the starter circuit.

It does cut off the injectors when it wants to disable the engine.