My tachometer all of a sudden acting strange
 

I know there could be many causes for tachometer acting strangely, but maybe someone had these exact symptoms. My tach is fine at high RPMs, but when I'm idling it has started to bounce around slightly. The engine is running smooth (its not hunting idle or anything), but needle starts moving slightly. It has never done this before. Today was the first fairly cold day I've driven it in some time, but not sure if that has anything to do with it. Any place to look first?

Could be instrument related, but I would doublecheck what the alternator/VR is doing through voltage measurements.

Doyle

Check condition of ign points and ign points gap - if you still have old-school points. My tach starts to goes crazy when the points are about to expire.

Grimlins;-(
 

Yes you should check the alt/reg but it could also be a small air leak somewhere in the CIS system. Do you have a high idle when warm?

It is a classic symptom of a failing alternator/regulator.

Alt....

Alt, VR, or ground wire. If you are over charging you can do some serious damage to your electrical system so don't drive it until you get it checked.

Yes my 74 911 used to do that also, change the points and re-gap if you don't want to call the tow truck.

A coupe of years ago I had the same jumping tach needle with my 82 SC. It would happen only when the car was warmer or on hot days. I check the system with a VM and the readings were jumping with the needle. I replaced the VR and that helped but it would return every blue moon.

The problem has recently returned in a very different way. Over the past few weeks of fall on my weekend drives, I noticed that the tach was much more jumpy and that the car was running a little rough around 2800 rpm. Then, after I had been traveling and the car was left out in the rain (usually in garage but we had home repairs under way and didn't want the GC carrying ladders along the car), the battery was dead and when I replaced it, the car was hard to start and will hardly idle. When it is running and I try to give it some gas, it bogs down and starts backfiring until it gets to about 3500 rpm and evens out. I'm about to pull out my tools starting with the VM to check the volts but could this be the VR or the alternator? How do I check the points on my car (I think it doesn't have points).

Any guidance would be great. Thanks.

The capacitive discharge SCR (Silicon Controlled Rectifier) in the ignition system is of relatively poor design or selection and will "self-fire" randomly if/when the capacitor charge voltage exceeds ~350 volts, which will more often happen at lower RPM when there is more time for charging between normal firings.

The over-charging of the capacitor, in turn, is the direct result of lack of regulation in the DC-to-DC, 12 volts to 350 volts, CDI inverter. That, voltage above 350 volts, of course, results from the 12 volts supply voltage itself going into "orbit", >16 volts. Keep in mind that the alternator output voltage "spikes" might well be >16 volts while the battery terminal voltage, all the way up there in the front of the 911, might never be above ~14-15 volts.

Once the problem becomes more extreme the extra sparks will result in triggering the engine over-rev limit circuit, which is also defective in design and tends to lock up with that alternator output over-voltage. Dead engine results, fuel pump circuit relay OPEN, and now you have to shut the ignition off, count to 10 slowly and restart.

The ACTUAL problem.

Defective alternator, stator winding insulation aged/burned, windings shorting intermittently, VR tries to make up the difference pumping up the rotor current, short opens, alternator output goes into orbit briefly. Bumps, bumpy road makes these "things" happen more often.

Thanks wwest. Sorry for my ignorance. Are you saying that I need to replace the alternator? I checked the dist cap and it looks fine...there was very minor pitting on the rotor contact...what I would consider normal.

Something similar happened to me about a year ago. Then I noticed the fasten seat belt light coming on at different times while driving. No apparent pattern. Some drives nothing at all. I made an adapter to plug my voltmeter into the cigarette lighter. Took a long drive and sure enough the symptom matched a voltage spike. Replaced the alternator and it went away.

Cool. I'll be ordering one on Monday. Will report results. Thank you.

Today I bought the voltmeter that goes into the cigarette lighter. The voltage seems fine, it is always between 13.5-14V while engine is running, while the tach needle is bouncing slightly at the lower RPMS. So I'm thinking my charging system is OK. I was going to actually take out the Tach to see if the wires on the back are OK. Anything else to check? Car seems to be running fairly well as always.

+1 on VR, what is the voltage at idle? 13.5? What is your battery voltage with the car off?
If you have an old-school 'mechanical" voltage regulator you can adjust it a bit, if it is off.

+2 on make sure your points are clean and adjusted well, if you have points.

Voltage at idle is ~13.6. If there was something wrong with the points, wouldn't the car run like crap? So that's why I've kind of eliminated the charging system and distributor as a cause.

I agree with you then that you're probably ok as far as Alternator and VR goes.

If you have points, it sure doesn't hurt to check the gap/dwell and make sure they are cleaned up on the contact surface. Points are a 'wear item' and can change even in a few thousand miles.

Based on the year of my engine (77) and from taking the distributor cap off, I don't believe I have points. I also checked the wiring on the back of the tachometer, and everything was plugged in tight. So now I've ruled out charging system, points, and wiring on the tach.

Grimlins again.:-(
 

So like I was saying can you(cheaper)or a shop check the CIS system for leaks. I realize this is not for the faint of heart(or wallet). I have a genuine interest as I have a similar thing going on with my SC. As you have found out this condition is temperature related as the rubber hoses expand and contract.

I don't understand why air leaks would only cause the needle to move? As I said in my initial posting, the engine turns at constant speed. It's not like I'm hearing actual RPMs change. If the idle was actually changing, you could hear and feel that. Is that what you think is happening? This has to be an electrical issue. The needle doesn't know about air leaks. It only knows when RPMs change.