911SC crazy electrical issue- can't turn off the engine

[mradovan]

Fellow 911SC owners,

The other day I put the engine backing my car after a very long DIY transaxle rebuild. Before removing the engine the car was running perfectly. I rebuilt the engine a few years ago and have barely logged any miles which is a bad thing and one I am tying to correct.

The Car has an MSD 6A ignition module and is running a wide band O2 sensor that includes an output to run the CIS lambda system. All this has worked flawlessly in the past.

Here is the interesting part. Before attempting to start the car I disconnected the ignition and cranked it to make sure the engine had oil pressure. That went well. Next, I tried to start in and the start would not shut off even when I pulled the connector off the electrical module behind the key. I replaced the starter with another unit I had and I was finally able to get the car to start after I found the missing vacuum line to the air box that I forget to connect. Due to the long storage, I fogged the engine with oil so it was running crappy and smoking which is normal but once the smoke cleared I was still hearing a strange noise which I think was the starter running so I attempted to shut the car off and that's when it got weird. The car kept running even when I pulled the plug to the electrical module so I removed the battery lead.

Somehow the ignition module and the starter are getting power with the key off and the starter is staying energized when the car is running. I have swapped the existing ignition module for another unit and the problem is still present.

I have been looking at the electrical schematic and wonder if this is an alternate problem or possibly an issue with the rev limiter since it is common to both systems.

Any ideas? Has anyone else experienced much strange failure?

[john walker's workshop]

Typical starter solenoid sticking. Get another starter.

[mradovan]

Thanks. The replacement starter was a new rebuilt Bosch unit. I can see how the MSD gets energized if the starter selenoid sticks. The first starter that I used to prime the engine was probably the original unit. I guess I will pull it out and take it back but I am suspicious that something else is going on.

[Tea Tray]

I was presented this problem years ago on a custom Norton Commando. The shop that built the bike wired the rectifier/zener output wire to the ignition side of the switch.
So whenever the engine speed was above idle the ignition switch wouldn’t kill the engine. The shop claimed his new ignition switch was bad and since he supplied it, he was on his own.
A simple move of one wire cured the issue.
I wonder if your situation could be similar.
Maybe isolate the questionable wires and try to duplicate the issue.

[Superman]

To cut the engine, you disconnected the battery in the trunk. This is good. Never try to cut the engine by disconnecting a secondary (high voltage) ignition wire (such as the one at the top of the distributor).

[john walker's workshop]

Another thing that can keep the starter engaged is the bendix teeth getting stuck in the ring gear teeth due to chewed ring gear teeth. When that happens, the solenoid is kept in it's full contact position and keeps the atarter activated.

[T77911S]

did you remove the 14 pin connector on the relay panel. if so make sure it is connected properly.

[JSV798]

I thought that disconnecting the battery with the engine running killed the alternator? Pulling the fuel pump fuse or relay would have done the trick or just stall it.

[Walt Fricke]

JSV - those of us who went racing learned from the instructions for the Merit (German company) red key battery disconnect that the 3d pole on that switch was to ground the lead from the alternator through a supplied resistor with the key in the off position to avoid damaging one or more diodes in the alternator. Didn't say if not wiring this up caused back EMF or voltage spikes or whatnot to cause damage, but most of us believed the instructions.

On the other hand, my big high Watt low Ohm resistors, over the years, have sometimes broken off at a lead, and for one reason or another I have shut off the engine with the disconnect without first turning it off in the normal way (it is mounted roughly in the center of the dash, with a cable out to the cowl). And I can't see that I had alternator damage as a result. But maybe if you shut the engine off in this way often, or at higher RPMs than usual, that sort of damage could result?

[mradovan]

Thanks for all the suggestions. I haven't been able to trouble shoot the problem but I am confident the starter gear is fine. The starter is a new remanudactured unit but it could be bad. This is the most obvious and simpliest solution. I think I am going to disconnect the alternator belt and see what happens to rule out something strange happening with it. I also need to unplug the Rev limiter to be sure that is not causing some problem. Worse case is that the Mechanical Engineer (me) waits for my son who is a senior in colleger studying electrical Engieering to figure it out!

I will report back when I learn something.

[T77911S]

engineer?
dont over think it.

you have connected something wrong if it worked fine before putting the motor back in.
check the 14pin connector and check the wiring to the starter.
i remember someone putting the battery cable on the wrong teminal on the starter and the starter motor running all the time.

[mradovan]

I found the problem which was self-inflicted. For reasons I don’t fully understand the starter solenoid was getting 7 volts when the car was running due to my hidden kill switch being in the kill position. When I installed the switch a million years ago it worked as intended but I now have a MSD ignition unit that gets power from the starter. The kill switch grounds the fuel pump relay. On a SC this ground goes away once the air plate rises. Needless to say the kill switch is no longer functioning as intended. Once I figure it out and closed the switch all my problems went away.

Thanks to everyone who offered ideas for possible solutions to my problem.

[Walt Fricke]

Well, who would have guessed this, not knowing you had previously made a modification which had worked. Maybe the guys who suspected an anti-theft device?

Gave me, anyway, a reason once again to run through the systems Porsche used on these models to a) prevent the fuel pump from running with the key in the on position but the engine not running (gee, how can opening a switch to ground change what a relay does?), and b) basically to do the same for the CDI box, along with allowing the car to start and run.

[T77911S]

#11 had it closest