Help - My 911 only starts when it wants to...
 

My '71 911 T has been doing its best to test my nerves as of late.

Out of the blue, it randomly won't start even though the battery is charged. When I turn the key the booger does one of two things: 1) fires up, or 2) nothing.

And by nothing, I mean nothing...no clicking of a dead battery or anything at all. In fact, during this I can turn on the headlights, radio or anything else powered by the battery, except the engine of course.

Interestingly, if I turn the key several times it every now or then will (or may not) turn over.

One more thing: this seemingly only happens when the car is driven often, i.e. when I daily drive it. When it sits for days or weeks, it fires up when asked.

My guess is that somewhere between the ignition and the starter there is an issue (the starter isn't grounded properly??), but I am by no means an electrical expert so that is a guess at best.

Any thoughts or suggestions? If I'm right with my guess, what can I do to pinpoint the issue?

Thanks in advance for your help!

Two possibilities spring to mind: Starter/solenoid, or ignition switch.

Next time it fails to turn over, try this. Look in the trunk, behind the cardboard that covers the back of the instruments, you will see a wire loom coming through the bulkhead just in front of the tach. There should be a yellow wire that comes out of the loom for a bit, has a connector in the middle, and reenters the loom. This is the wire from the ignition start switch to the solenoid. Disconnect the connector and jump a hot wire to the part that is farthest from the dash. The starter should activate (obviously, be sure car is out of gear.) If you get nothing, then you have a problem with the solenoid/starter. If it activates, the problem is likely with the ignition switch.

Put a multimeter or test light on the end of the wire that is closest to the tach. Have someone turn the ignition switch to "start". There should be a reading or lighting of the test light. If not, the switch or connections to it are suspect.

Being that the problem is intermittent and only after the car is warm, I'd suspect the solenoid/strarter first, but the wire test is easy to perform and you won't get dirty.

I would check/clean all the ground connections before I went diggng in to anything more complicated.

Get a Fluke and measure voltage at the starter. It's a two man job.

But it's most likely the switch or solenoid....have you tried tapping the solenoid when it happens?

+100 - very good call.

My problem went from once-every-year to "any time the car was cold". Starter had been replaced anyway, new ground straps made no difference (but needed replacing anyway).

Eventually turned out to be that either the 14-pin engine bay plug or the plug on the dash bulkhead read 100% continuity - but with enough induced voltage drop that the solenoid couldn't kick in, most of the time didn't even click.

Cleaned the sockets and pins thoroughly and hasn't happened since 2 years later.

But go after the most likely causes first, solenoid or switch. It's usually those.

(Although I wonder how many people replace the ignition switch and "fix" the problem simply by disturbing & remaking the oxidized bulkhead connections - because in retrospect, that's exactly what I did at first, and it fixed it for me for a few months)

Had exactly the same problem a few months ago. Replaced the electric cassette on the ignition switch, it did nothing. Replaced the starter solenoid and the problem seems to have gone away.

So far this is all I have done and so far, so good...there was quite a bit of corrosion in several places, so I'm cautiously optimistic that will do the trick.

Of course, we both know I'll be back here in two or three weeks to report that it was the starter/solenoid.

Thanks for all of the responses. Really appreciate it.

the simplest is usually the best. bad connections account for a lot of electrical gremlins, and not just the connections at the battery, the other end of the cable is important too.

i was planing on a new post for this, and even though u may have it fixed, this is for your entertainment.

http://hyperphysics.phy-astr.gsu.edu/HBASE/electric/autbat.html#c1

we measure battery term. resistance at work, well not me, but someone does, it is suppose to be in the micro ohms, for those that are math challenged, thats, .000001 equals 1 micro ohm.