Bouncing Tach when Cold, and Zap Noises

[OsoMoore]

Spring is (sort of) here and my 911 SC is back out for the daily commute.
When I star the car, it idles high (as it should) for about 30 seconds to a minute. After that, the weirdness begins.

For the next 10 minutes or so, the tach regularly takes little bounces. It will idle at 900, and then skip up to 1500 or even 2000, and immediately drop back down. Sort of like someone flicked the needle with their finger. Once the car is up to full operating temperature, the bouncing is little more than +200 RPM, although it is still noticeable.

I read that the voltage regulator is a prime candidate, so yesterday I hooked up my meter and tried to watch for correlation. The car was not totally "cold", so the bouncing wasn't more than about 500 RPM. I had readings between 14.25 and 14.29 volts. But mostly, it only varied by 0.02 volts or less.

While checking for vacuum leaks, I noticed a semi-regular "zapping" noise in the engine compartment. It seems to be coming from the area of the distributor. It sounds like a little arc jumping intermittently. I opened up the distributor and I think it might have some wear. See the pictures below.

In these shots, I am concerned that the golden coating is missing from part of the contact.



The scratches on the plastic here are from my sandpaper attempt to clean up the contact. I don't think it is from rubbing on the cap.


In these shots, I am not sure if the black is bad.



The engine is firing smoothly, but the little noises seem unusual and I'd like to be sure something isn't failing.

[Uncle Enzo]

How do your points look? I had a similar situation but mine would drop off all together. I found the points were shot.

[wwest]

The distributor cap/rotor does not contact physically, you're looking at carbon deposits from the short ARCs.


The carbon arc "trail" on the distributor cap near the "contact" point have a different tale to tell...

The coil is sometimes being fired without the rotor being in the proper rotation position.

A CDI will do that with over-voltage spiking, multiple firings.

Puzzle is, the tach trigger/drive comes from the CDI's distributor signal input, not the CDI's firing of the plugs.

If you have an external VR, unplug it long enough to discern.

You need to check the voltage near the alternator output, the battery serves as a LARGE capacitor filtering any spikes.

[wwest]

Hmmm...

1979 911 SC...

You have a magnetic pickup distributor, no points, and if factory original, an external VR.

Highly likely you have intermittent shorting inside the alternator, resulting HUGE magnetic field, RFI/EMI spike, "talks" to the distributor PU, out of sequence CDI firings.

[ThomO]

wwest is probably correct. your meter is most likely not fast enough to catch the spikes. The bouncing tach is the give away.

[wwest]

Really interested in knowing if unplugging the VR works.

[OsoMoore]

Originally Posted by wwest

Really interested in knowing if unplugging the VR works.

I'll try that tomorrow morning! Should I worry about causing problems with the unregulated voltage?

EDIT:
I'm getting conflicting results searching for where my VR is located. Some say it is on the inside of the fan housing attached to the alternator, some say it is off to the left inside the engine bay. I'll take a look when I get off work.

[wwest]

Originally Posted by OsoMoore

I'll try that tomorrow morning! Should I worry about causing problems with the unregulated voltage?

EDIT:
I'm getting conflicting results searching for where my VR is located. Some say it is on the inside of the fan housing attached to the alternator, some say it is off to the left inside the engine bay. I'll take a look when I get off work.

Unplugging the VR will result in running only on battery.

To be certain the alternator doesn't produce power remove the VR and then start the car.

The factory VR is on the left side of the engine bay looking forward.

If the factory alternator has been replaced at some point it most likely has an internal VR. In which case don't bother, 'way to much trouble.

The factory VR may be still mounted on the panel in either case.

[OsoMoore]

Unfortunately, it appears mine is internal. there is nothing mounted between the defogger relay and ignition discharge control unit.

[wwest]

You can probably accomplish the same task, disable the alternator, by removing the battery charge indicator/bulb. A bit more trouble but better than possibly pulling a good alternator.

The bulb supplies the initial rotor excitation, magnetization.

[OsoMoore]

I removed the oil/temperature gauge and removed the spade connector on one side. The tach is still doing it's bouncing thing. Does this mean its not the alternator or VR?

(The bulb had a double spade connector on one side, and a single on the other side. I removed the single wire (which was blue)).

I also did some more listening and I am fairly certain the faint sparking noise in the engine bay is coming from the distributor.

[OsoMoore]

Video of the test.

[wwest]

'77 diagram shows blue (Bu) wire on one side, black/red (Bk/R) on the other.

Yes, that seems to rule out the alternator/VR

The carbon arcing traces inside the distributor cap does seem to indicate random ignition timing and/or multiple firings.

That seems to leave the CDI, the magnetic distributor pickup, or possibly intermittent wiring between the 2.

Solid state electronic components are more likely to exhibit intermittent symptoms with rising heat.

Only thing I might try at this point is some coolspray (Radio Shack) (***) on the CDI after the car warms up and the ignition timing settles down.

*** fine spray from a garden water hose....

[wwest]

My own experience with the bouncing tach symptom was that it was definitely tied to battery overcharging. In my case the tach would sometimes bounce so high as to "hit" the rev limiter, which would then latch up and shut down the engine, oftentimes while underway.

The rev limiter circuit would then remain latched until the ignition switch was turned to off for a slow count to 10. There was also a definite correlation between hitting a bump in the road and the engine shutting down, rev limiter latch-up.

New battery was installed at least twice during year so episode.

Installing a new VR (external, solid state) was of no help.

Finally removed and replaced the alternator(host) which cured the bouncing tach, but now the CDI had been "cooked" and it would shut down once the engine compartment warmed it. Would restart immediately with a minute or so of fine water spray. No Radio Shack in McCall Id.

Loren repaired the CDI...

Opened up the alternator and found multiple points of wire insulators burned away.

Road bounce, stator phase shorts(my assumption), high voltage spike, engine stops.

Oh, one other battery over-charging indication, seemingly, the e-brake light would latch on, had to disconnect the battery to clear.. until the next time.

[OsoMoore]

It sounds like you had quite an adventure chasing that one down.

I'll do some searching and figure out where the CDI and magnetic distributor pickup live. You mention they may exhibit issues as the heat rises. Mostly mine seem to decrease as the engine warms up. Still, maybe it is a thermal thing.

[wwest]

What meant to say is that solid state components are more likely to exhibit impending failure symptoms as the temperature rises.

Item 7 in your post #9 is the CDI. The magnetic pickup has replaced the points in the distributor.

Let's not overlook the possibility that the tach itself is failing since in the video the tach was jumping but the engine idle seemed normal.

On the other hand SOMETHING is causing those carbon paths/traces with the rotor out of position.

Check the rotor alignment notch is tight to the shaft?

Worth cooling the CDI once things warm up and settle down?

[wwest]

Adventure...

Yes, the car was driven from Moscow Id to McCall by alternating the VR in/out/in/out.. ETC...

Probably what cooked the CDI.

[OsoMoore]

There is some play in how the cap aligns - you can rotate it back and forth a little even when it is clipped on. I looked carefully inside the distributor cap. I noticed the notch which keeps it fixed in place appears damaged on one edge. That is about the amount of play I'm seeing. Could this be related? I tried rotating it back and forth in it's play amount while the car was running, but there was no noticeably effect.

I also listened and watched very carefully, and I am 90% certain that the little spark noises correlate with the tachometer jumps. When they come in a burst together, it results in a bounce. The "normally", they occur more spread out. Their frequency is like that of popcorn in the microwave.

[wwest]

Buy a new rotor and cap for certain. I think I burned carbon pinholes in the cap sides near the plug "contacts".

Not only are there carbon track but also an indication that something has been rubbing/scratching along the same lines.

Someone else needs to chime in but it seems to me that those black carbon tracks on the copper contacts should be closer to the top of the cap. Cap not fitting on the distributor housing or rotor to far down on the shaft.

Still doesn't explain the popcorn..

[wwest]

Under that black epoxy on the rotor is a resistor.

Do I not see a wire coming out of the epoxy that should be spot welded to the copper rotor tip?

Came loose and is n/flailing the sides of the cap?