Intermittent swinging tach needle caused by ... ?

[billjam]

Just lately, I have been starting to get a dose of intermittent swinging tach needle in my 930.
The needle swings wildly up and down for a couple of seconds, the car misfires a bit at the same time, and then it all returns to normal. It is quite intermittent and usually happens with moderate throttle.
It's obviously a juvenile electrical gremlin - I'd just like some pointers to nip it in the bud before it turns into a full on meltdown in the middle of the freeway

I've wiggled and removed/replaced all accessible relays and plugs, but no joy.
Ideas?

[jwasbury]

Hi Bill (and Toni)

Tach weirdness is a classic sign of alternator problems...may want to check that out. I helped a local buddy with his tach problems on a '78 SC. He had taken the car to two different independent shops, neither of them diagnosed the problem correctly. One of them replaced his plug wires which didn't solve the issue. I put my voltmeter on his battery with the car running and noted high voltage spiking, and also noted the battery was hot. Told him to get a new alt, and he said: "so that's why this car has been eating batteries"

Since the tach is driven by the ignition system, I suppose a malfunctioning CDI may also be possible.

[wwest]

Yes, pull the headlights on when this happens and if things settle down look into a new Alternator. Run it this way too long and you'll also need a new/rebuilt CDI ignition.

The DC-to-DC upconverter, 12 volts to 300, in the CDI box is not regulated or limited so if the 12 volts rises to ~16 the 300 volts increases to 400. 400 volts turns out to be a bit too much for the breakdown/breakover rating of the CDI's SCR, Silicon Controlled Rectifier.

So the CDI "fires" randomly, engine runs rough, and the tach bounces since it is "driven" by the CDI "sparks".

PS: Since the battery is all the way up front and the voltage regulator at the rear you must be sure of good electrical connections, both positive side and ground/body otherwise you might get HUGE voltage spiking that cannot be seen with a simple voltmeter, takes an O-scope.

[pdx930]

I had similar symptoms that progressed to the car losing all power on while on the road, only to restart after the car rolled off of the flat bed into my garage. Like you I replaced lots of relays, cleaned lots of connectors and all of the grounds straps, but it didn't go away until I replaced the CDI unit and the coil. I had Loren check mine, but he was not able to find an issue. Still, that is what fixed it.

[mark houghton]

Originally Posted by wwest

Yes, pull the headlights on when this happens and if things settle down look into a new Alternator. Run it this way too long and you'll also need a new/rebuilt CDI ignition.

The DC-to-DC upconverter, 12 volts to 300, in the CDI box is not regulated or limited so if the 12 volts rises to ~16 the 300 volts increases to 400. 400 volts turns out to be a bit too much for the breakdown/breakover rating of the CDI's SCR, Silicon Controlled Rectifier.

So the CDI "fires" randomly, engine runs rough, and the tach bounces since it is "driven" by the CDI "sparks".

PS: Since the battery is all the way up front and the voltage regulator at the rear you must be sure of good electrical connections, both positive side and ground/body otherwise you might get HUGE voltage spiking that cannot be seen with a simple voltmeter, takes an O-scope.

That's the best description I've yet heard on how voltage regulator/alternator problems affect spark. Excellent stuff.

[A930Rocket]

That was good. Copy and paste it to the "my car won't start" sticky above.

Originally Posted by wwest

Yes, pull the headlights on when this happens and if things settle down look into a new Alternator. Run it this way too long and you'll also need a new/rebuilt CDI ignition.

The DC-to-DC upconverter, 12 volts to 300, in the CDI box is not regulated or limited so if the 12 volts rises to ~16 the 300 volts increases to 400. 400 volts turns out to be a bit too much for the breakdown/breakover rating of the CDI's SCR, Silicon Controlled Rectifier.

So the CDI "fires" randomly, engine runs rough, and the tach bounces since it is "driven" by the CDI "sparks".

PS: Since the battery is all the way up front and the voltage regulator at the rear you must be sure of good electrical connections, both positive side and ground/body otherwise you might get HUGE voltage spiking that cannot be seen with a simple voltmeter, takes an O-scope.

[wwest]

Originally Posted by pdx930

I had similar symptoms that progressed to the car losing all power on while on the road, only to restart after the car rolled off of the flat bed into my garage. Like you I replaced lots of relays, cleaned lots of connectors and all of the grounds straps, but it didn't go away until I replaced the CDI unit and the coil. I had Loren check mine, but he was not able to find an issue. Still, that is what fixed it.

What often happens is that the random sparks occur so rapidly that the engine overspeed, RPM limit module, triggers, actually LATCHES up in the overspeed state, and shuts down the engine entirely via disabling the fuel pump.

Turn the engine ignition switch COMPLETELY off, count to 10 slowly, and now the engine will restart. Switching the ignition completely off removes 12 volt power from the overspeed detection module and with power it will not remain in the "latched" state.

Sorry I didn't bring this up earlier but the factory overspeed detection module seems to also be subject to the battery overcharge/overvoltage situation. I spent hours troubleshooting this engine overspeed shutdown issue thinking had something to do with the "upside down" fuel flow/pump disabling microswitch in the engine intake.

[mark houghton]

Originally Posted by wwest

Sorry I didn't bring this up earlier but the factory overspeed detection module seems to also be subject to the battery overcharge/overvoltage situation. I spent hours troubleshooting this engine overspeed shutdown issue thinking had something to do with the "upside down" fuel flow/pump disabling microswitch in the engine intake.

Are you talking the overspeed relay beneath the drivers seat? That little dude seems to be linked to a lot of transitory gremlins, including fuel pump(ing) issues that many of us have spent an inordinate amount of time discussing and troubleshooting.

[JFairman]

I wouldn't be surprised if the factory alarm also uses the "speed relay" relay to disable the fuel pumps, as does the overboost sensor white wire when it is not grounded.

[equality72521]

I had those symptoms and it turned out one of the VR wires backed out of the plug and was losing it's connection to the VR. Fixed it and problem went away.

[billjam]

I'll recheck all my wires and plugs again before I tackle the alternator - getting that thing out scares me

If I find that voltage spiking is occuring, what's the fix? New alternator, or can the voltage regulator be replaced separately?

I guess the next question might be - anyone got one?
I could look it up in the parts catalog, but I'll ask while I'm here ... anyone know off the top if same parts are used on similar year 911s? (1986)

[wwest]

In my case neither the alternator or the regulator had failed, exactly. When I took the removed alternator apart I discovered multiple points where the internal wiring insulation had degraded over the years, '78 to '09. My conclusion was that under the vibration of driving one of the alternator phases would short out, the regulator would "pump-up" the rotor current to compensate, and now when the short went away, was "removed", too much voltage would be generating in accordance with the inductive time constant of the rotor windings/etc.

We had noted that hitting a bump often resulted in the engine dying.

I purchased my Alternator from our host but you should look at Ebay.

I don't know which year Porsche went to the Alternator/regulator combined but I would suggest you follow their lead regardless.