techman1

Hello all, and thanks for reading.

1995 coupe, no modifications to the electrical or charging. Pretty stock everywhere else.
Newish battery, and ground cable.

I have had a strange issue for a few years, and it has become more pronounced, needs to be dealt with.

The car starts normally when cool, no immobilizer issues, hot or cold.
When the car gets to normal temp, and I stop somewhere for 5 ish minutes, heat soak gets to the starter, or something.
Immobilizer button, turn key, crank, and the starter spins without the bendix kicking it into the flywheel. zweee.
There will be a little clink after I release the key when the starter stops spinning. Happens once or twice then finally engages properly.
Now it happens 3-4 times, I wait for the car to cool a little, and it will work normally. Embarrassing at the gas station after a fillup.

Actions tried so far:
Removed starter, disassembled, cleaned, replaced bendix. Clean ground cable at starter.
Replaced ground cable at battery, good ground there.
Checked relay #61, opened, no overheat there, thinking it was not providing enough current. looks new.

There are a few occurrences of this issue in the archives, but they did not post any resolution.

Need someone with the electrical "Path" that happens during start to fill in the blanks:
Starter connections:
Starter is grounded. Thick cable goes directly to battery. Additional wire provides go signal to starter.
Turn key to crank, it energizes relay 61?
Current goes from #61 directly to starter?
Does the starter spin as the bendix kicks the gear forward? Or....?

I recall shop class discussing old starters, the bendix kicked, and when it made contact internally, current was applied to spin the starter motor.

pp000830

Did you also replace the solenoid that kicks the Bendix into position to engage the gear on one flywheel?
Andy

techman1

Yep, and due to the symptoms I thought it was a sure thing. Unfortunately, no.

pp000830

These are some educated guesses:
The smaller wire connection to the solenoid has hidden corrosion in or near its termination adding resistance so it is not getting power consistently to insert the Bendix fully- most likely
The moving parts in the Bendix/solenoid are not smooth or have developed burrs and need to be inspected for hanging up. Sanding the cast housing parts smooth with 600-1000 grit sandpaper and using a little high-temperature grease may help.
The ring gear engagement is not good due to ring gear damage - unlikely.
Just some thoughts,
Andy

jay@EZimmoblock.com

Yes. The immobilizer turns R61 on when disarmed and ignition is ON (immo Pin 16 to R61 term 86).
Yes. When energized, R61 passes the ignition "START" signal directly to the starter solenoid (R61 term 87 to starter control).

The ignition switch, immo and R61 appear to be doing their job. I've never been inside that starter but sounds mechanical to me.
Have you tried try bench-testing the starter? That'll tell you whether it's in the starter or the vehicle. Plus you can test your high temperature hypothesis.

pp000830

I think this validates that the relay & ignition switch are good-to-go:
"Immobilizer button, turn key, crank, and the starter spins without the Bendix kicking it into the flywheel."
Suggests the solenoid is sticking due to friction internal to it, friction in the Bendix, a bad solenoid out-of-the-box or a corroded control wire to the Bendix reducing the force the internal coil applies to the plunger.

pp000830

After some additional thought, the solenoid's plunger is moving far enough to cause it to engage its internal switch powering the starter's motor. Could it be that the starter is not fully seated to the bell housing and so doesn't reliably engage with the ring gear?

jay@EZimmoblock.com

In some starters the pinion is designed to engage the flywheel before the contactor closes and applies power to the starter. If the 993 starter is of that type, the fact that the contactor has closed (OP notes starter is spinning) implies the solenoid has traveled far enough that it should have already engaged the flywheel. Why the pinion isn't is the mystery. If anything were binding inside the starter, the solenoid wouldn't be able to move far enough to close the contactor, but it is. Two possibilities: 1) the actuator arm connection to the solenoid is flaky or 2) the pinion is being driven outward ok but it's slipping on the starter shaft. Either of those would match the symptoms OP noted.

Excellent starter motor animation

Dan Riley

Low current circuit is pulling in the "motor disk" in solenoid allowing high current circuit to actuate motor. Motor runs fine. Sounds like the part of the solenoid for the mechanical actuation of the bendix lever is not completing full travel. May appear to be a mechanical bind in the lever that worsens when hot. I would think the lever should operate the bendix with little or no resistance at all times.

notabot

I had a warm/hot start issue last year. Held me up at the Costco gas station right before a PCA driving tour.

Turned out to be the ISV and a parasitic battery drain. Replaced both and the car fires up instantly all the time, even after running numerous errands in our Texas summer.

techman1

Dan and Jay,
Good food for thought. Bendix replaced, it is good.
The actual lever it pulls when actuated might be binding when hot.

I am getting good at starter removal.
I will wait and see if anyone has battled this exact challenge and had a fix.
If not , I will remove and disassemble the starter. Again.

3healey

Just curious: do you have the stock flywheel, or an aftermarket mwf/lwf?

jay@EZimmoblock.com

If you end up taking the starter out again, do a quick bench test to see if the pinion is pushing out (fully). All you need is the battery and a couple jumpers. Do it hot and cold. Do it a bunch of times. Listen carefully. I haven't seen the insides of the 993 starter but if it's constructed like the one in the video, if the motor spins, the solenoid is stroking, at least enough to pull in the contactor. Possibly it's the type where the contactor pulls in before the pinion is pushed out, in which case something could be binding after the contactor pulls in but before the pinion is pushed all the way out.

Post some pics of the innards if you can.

techman1

Stock flywheel.

Jay, planning on a removal this weekend. I will update with pix.

71-3.0-911

Considering what you've done already, it may be an easy check to grab a new battery and verify the operation with it. Low voltages in these cars do weird things. I'd start with the simplest first before tearing that thing out again.