Carpokes Porsche Forum

Update! I put the new switch in but unfortunately it didn’t fix the problem. Spent a few days with it just to see how it would behave and here’s what I’ve noticed:

It only happens when cold as I’ve mentioned before. After driving for a bit to get her warm I can turn it off and it cranks right up.

The fuel pump seems to play into this as well because the fuel pump priming is intermittent as well.

I did recently have a fuel pump relay burn up which I replaced and this problem might or might not have started after that. I pulled the fuse panel to inspect the back and while out cleaned all the connections and ground points but it looks good.

Not sure where to go from here

Add a 'telltale" light from that starter relay socket jumper to ground, and confirm that the bulb goes and stays full-bright every time you turn the key to the start position III. You'll be working through the troubleshooting procedure I described above. The test light will sub for your DMM this time. If the lamp stays bright regardless of the no-crank, you'll work downstream to find where the light doesn't stay bright while it doesn't crank. If the bulb doesn't stay bright, the problem is between the relay socket and the ignition switch.

There's always the annoying possibility that the starter itself is "the problem". A sticking solenoid will present exactly the symptom you are describing. Starts OK cold, but is intermittent when hot. The easy diagnostic step used to be to just try another starter, but those are getting a bit spendy to purchase just for a test. The reported 'key gets hot while trying to crank' symptom you described points to solenoid engaged, even if it doesn't actuall pull in and engage the starter motor.

Because it's not difficult at all and the car is just outside my office, I took a quick measurement for solenoid coil resistance in my later car, which measures somewhere around 0.5 Ohms ****. A little lower than I expected, and that pegs the current through the solenoid at around 25 Amps. So using the amps function on a common DMM isn't an easy option for testing for trigger current.. Most max out at about 10A in direct-reading mode. But the idea for the diagnosis method lets you test current flow as you exercise the key to start position, to see if full current is going to the solenoid. You could replace that CE panel relay socket jumper with a wire jumper, and use a clamp-on DC ammeter for the measurement. Guessing with voltage drop when cranking would make that current about 20A, so looking for 18-25A would be pretty reliable. If there's that level of current flowing, but no crank and no 'clunk' from the solenoid pulling in, there's a case for removing the starter and testing it on the floor with jumper cables. Then carefully disassembling the front of the starter motor, the solenoid, and the drive mechanism. This is a clean-and-lubricate exercise, to make sure that the solenoid has full travel available. There are plenty of generic starter disassembly and diagnosis procedures on youtube for guidance on that part. You can test on the bench with the started grounded negative plus positive to the 50 terminal on the solenoid. It should pull in and push the starter drive out, but this way the motor itself doesn't run. Perhaps this will be easy enough to try if the other electrical diagnostic steps aren't conclusive.

**** To test, you can measure resistance to ground at the 14-pin connector by the jump post under the hood. The top half of that connector can be gently pried from the fixed bottom half, exposing the pins under the top half. The pins are numbered on the mating face of the connector, and pin 14 carries the 50 circuit directly to the solenoid trigger 50 terminal on the solenoid. My ohm meter from pin 14 to chassis ground offers the solenoid and the connecting KS-2 conductor, without crawling under the car. At least so far in the process. I'll confess that I'm lazy, and would rather test and diagnose than throw money, time, and actual real work into figuring stuff out. Given the choice anyway.


The fuel pump 'priming' function happens when the key goes to run position II, when it energizes the 15 circuit to the fuel pump relay. The 'prime' function is supposed to happen only once per starting effort even if the car doesn't start on the first try. So it will prime once, it may not do that again unless you wait a few minutes between start attempts with key off. The actual function is part of the fuel pump relay smarts, so it's hard to say how long it may take to get the function back.

Cheapo imitation FP relays depend on the ignition pulse detection function to decide how long and how frequently the priming function will activate. This isn't really critical to the 'won't crank' problem.



Let us know what you find please.

You can also take your starter to any national auto parts chain for testing. They test for free. I wouldn't take the results as gospel (at all), but sometimes it can help shed a little light.

Thanks Tom and Dr Bob.

My mechanic recommended jumping the starter directly, but I didn’t have it lifted so decided to try jumping it off the jump post and 14 pin connector. Starter came to life right away. When I got in the car it cranked on the first turn.

I am not sure I’m measuring resistance at the yellow wire in the connector properly. I set the DMM to ohms and put the black to ground and red to the yellow wire. Car and ignition off and I get no reading. Am I supposed to get the reading with the key turned to run?

I am not sure I’m measuring resistance at the yellow wire in the connector properly. I set the DMM to ohms and put the black to ground and red to the yellow wire. Car and ignition off and I get no reading. Am I supposed to get the reading with the key turned to run?

No -- The resistance is measured with the key off. But it really doesn't make any difference whether it's in run or off, so long as you aren't trying to crank the starter when you measure. Do make sure your meter is connected to an actual body ground. I've made the mistake of thinking that the handy right-there top radiator bracket bolt is ground, but it isn't a reliable ground. Better to use a bolt for the crossbar. Remember also that you are measuring through the engine harness half of the connector. the one with pins. Not the 'stationary' half on the bottom 'female' half.

** With everything under the hood assembled, you should be able to read that solenoid coil resistance at the jumper in the starter relay socket. It would include the contact resistance at the 14-pin under the hood, and if the resistance is more than the coil, you'd be looking at both ends of that yellow wire in the KS-2 harness anyway.

Electrical troubleshooting for me uses a 'divide and conquer' method. Test at the middle of a circuit, and decide from that whether you move towards the source or towards the load. Decision is biased based on how easy it is to measure, and how much community history we have on prior symptoms.


My mechanic recommended jumping the starter directly, but I didn’t have it lifted so decided to try jumping it off the jump post and 14 pin connector. Starter came to life right away. When I got in the car it cranked on the first turn.

This is interesting. The pin 14 yellow-wire connection by the jump post goes to the starter solenoid engage connection, and is the one you measured with the meter. Was the connector top half lifted/disconnected when you did this, or did you lift the cover off to access the back of the pin with the connector still assembled? Regardless, this may be a clue. The design of the connector is not 'self-wiping' contacts, although just lifting and reconnecting may be enough to clear a little corrosion from the pins. The 50 circuit to the solenoid is by far the highest current flow that passes through that connector, and as such is usually the least sensitive to resistance in a connector contact. That might explain why you can't see the solenoid coil resistance with a meter, yet the higher load of the solenoid will sometimes be enough to 'burn through' an oxidation layer.

Do share back what you find. There's a decent chance that just 'disturbing' your 14-pin will make starting more reliable. I'm not often that lucky, so end up doing the real cleaning after I get a hint like yours. Probably worthwhile brass-brushing the connectors and connections at the starter solenoid too. There are two smaller connections to clean there. One is the 50 circuit we've been discussing, and the other is an aux contact that bypasses one of the load resistors in your ignition coil supply. Do the two large lugs from the battery and forward to alternator and the jump post too. And the jump post connections since you'll be in the area and have the tools out.


TL;DR - loosely related ramblings --

Continuing on that theme, I like to think I'm pretty religious about keeping electrical things well-maintained or at least well-inspected. The last time I had that connector apart on mine was a a couple years ago now with the last timing belt replacement. I honestly don't remember inspecting or cleaning that connector at the time. But I struggle to remember even half of the things I've forgotten... Anyway, I pulled that connector open to measure the resistance through the solenoid for this thread, and decided it needed a bit of attention now. So out came a brass detailer's toothbrush, and I brightened up the pins (with their tin plating) and the little brass sleeves in the bottom half that pins slide in to. The real telltale for that was a water and soap stain in the mating faces of the connectors. I have the plastic cover on it, the one that's supposed to keep rain and car wash detergent out of the connector and jump post, but apparently a little snucked past it and into the connector. Everything still functioning perfectly, just an interesting little hint that it might need more protection. I don't drive it in the rain on purpose, so the most likely cause is wash water. Anyway, if you've read this far, the hint is to take a careful look at your connector and decide if a little more attention is needed. There are a few contact and connector chems like Deoxit that a few folks swear by. Almost any good contact clear might help if you can't get the connecting pieces looking great again. But the basic piece of this is the brass brush to remove oxidation. My local Harbor Freight store sells a kit of steel, brass, and plastic detailers brushes. Typically a couple dollars at most on sale, and handy enough that I keep a couple sets. They were all out yesterday to do a service on a neighbor's electric golf cart, so handy for Other Things besides the 928.

----

I like using the soft brass brushes for cleaning electrical connections, particularly the tin-plated copper or brass terminals. The tin plating is a 'sacrificial' treatment on the copper or brass, and is particularly effective at avoiding chloride contamination. Using the soft brass brush offers a chance to remove any contamination or oxidation from the base metal, so much better/safer than using a steel brush or sandpaper. There are plenty of opportunities to use them during your regular electrical maintenance efforts like ground points and the ring connectors used there. Porsche used a plated beryllium copper crimp terminal for those. I do an annual electrical maintenance that includes inspecting and cleaning those, avoiding a slew of stooooped and hard to diagnose symptoms. I also started using a little bit of Vaseline on those after they are clean and snug. It melts and flows into any voids and protects the metal from casual corrosion. Wipes up easily with a paper towel, doesn't stain clothes but will stain leather and carpet though. But the focus is on the brass brushes and all the connection cleaning.

Bob you should get someone to compile your posts into a book or something.

I jumped the starter by pulling the connector out with the cover still on and going straight from the pin to the jump post. I got some sparks first then the starter engaged. It’s started right up cold several times ever since. I did use a test light the next time I cold started and it lit up and the car fired up right away.

I have had issues with the connector causing start problems before, so the first thing I did when this started was pull the plug and spray some contact cleaner, but it didn’t make a difference. The last bit of your post leads me to think I could take it a little further - especially cleaning the solenoid contacts.

Took delivery of a nice set of 1982 workshop manuals today!

Good on you, both for the success and for grabbing a set of the manuals. Lessons I learned long ago, I can't imagine owning one of these cars without the manuals. Shortly after the car came to me, a then-local Cali dealer advertised a full manual set for a great price. I figured they would be used, but they had a pallet of them new in boxes. Publisher and PCNA were clearing them out, and no more printing was planned. Porsche then sold the publishing rights to house in Germany who told us they would not be printing any more either. Offered to buy the rights and they had no interest in selling. A small group ended up scanning and indexing the manual set on CD, along with a then-extensive collection of user support posts. For $15 including shipping back then. They may still be available from a couple 928-centric parts vendors. Worth looking for. Even though I have the printed set, it's still easier to search the disks, print the pertinent pages, and don't worry about fingerprints or pencil notes fouling the still-pristine 928 'bible'.

Glad you were able to get it going again!