82 911SC Intermittent Loss of Power

[JMH82SC]

Over the last month I have experienced a strange loss of power that happens for under 60 seconds or so that I hoping someone could help me diagnose. The vehicle has the stock ignition and CIS system. Vehicle runs fantastic but this problem is so intermittent I'm finding it hard to locate problem.

The loss of power feels like the lambda relay is unplugged and has a few symptoms

1 Happens most when car is idling but has happened when driving as well. Vehicle temp has no correlation as it has done it hot and cold. I have had it happen 1 mile from my house then clear up and not happen for days then yesterday I drove 25 miles and it happen on the way home.

2 Tachometer goes straight to 4 to 5 thousand RPM and real RPM would be well below 1K then out of the blue tachometer reads normal and vehicle is fine.

3 Feels like it is going to stall but you can get RPMs up and pull off the road just has no power like lambda relay and again tach is bouncing 4 to 5K.

4 I noticed yesterday when it happened the radio cut out also but I have never noticed this before.

Any direction would be a huge help.

Jon

[tirwin]

When I first started reading your post I thought fuel delivery problem like fuel pump or maybe debris in the tank. Then as I read more I started thinking electrical. A couple of easy things to check -- fuel pump relay and O2 relay. You should keep one of each in the glove box anyway so no harm in buying them if they are not the problem.

What kind of CDI box do you have? Bosch or PermaTune or maybe MSD? My CDI died last weekend and it was an intermittent failure. One reason I'm thinking the CDI is the tach symptom. One wire in the CDI harness goes to the tach. It sends a square wave at a frequency that the tach reads. If the CDI is going bad it could be sending a bad signal to the tach. Your best bet to test the CDI is to find a local friend with a known good CDI and coil to swap out.

You should also check your battery to see if it is perhaps undercharged or overcharged. Problems with the VR can cause a wild bouncing tach. Does it normally crank strong?

Edit: My CDI died last weekend and I had to get AAA to flatbed the car home. I replaced it with an MSD Street Fire and Blaster II coil on Friday as described in this thread.

[GaryR]

It's electrical, the tach going whacko and radio cut out says that. First thing to do is examine and clean all your grounds, start with the transmission strap-to-body and the battery ground. If that doesn't do it start looking at the circuits common to the symptoms. The CDI could certainly be the culprit, just not sure how that would effect the radio..

[tirwin]

Gary,

About a week before my CDI died I was listening to music from my iPhone through the stereo and the music paused suddenly. When I looked down the lights on the radio flashed a few times. I turned it off (well it only has a standby mode). I feared that it might be an over-charging problem but the battery voltage was fine. I have had no luck finding a bad ground. I'm not sure how it could be related to the CDI either. I am planning an engine drop soon and I'm replacing the front and rear fuse panels with Fred Cook's ATO panels. Maybe I will turn up something else then. My Optima red top is 2008 vintage so I'll probably replace it then too.

[GaryR]

If the radio is on the CDI circuit it could be affected by it, so I wouldn't discount your idea at all! I always start with the simple stuff, like grounds, as they are normally ignored over the years..

[tirwin]

Jon,

Curiosity question. Do you happen to have a stereo amp? I do and it is wired directly to the battery. Just asking because that is a difference from stock in my setup.

Gary has me thinking. I'm going to go look at the electrical diagrams to see if there is anything common between the CDI and radio circuit. In any case, you should take Gary's advice and check for bad grounds.

[tirwin]

I just looked at the Bentley. It looks like to me that the switched power from the ignition switch comes in on fuse 12 and is bridged to fuse 11 where it connects to the CDI on circuit 15. The tach also has a connection to circuit 15. And the tach is also directly connected to the CDI via the black/violet wire. If the switched power for the radio coming off the ignition switch is also circuit 15, then I think we have found a commonality. I just can't confirm the radio yet.

[Bob Kontak]

Originally Posted by GaryR

The CDI could certainly be the culprit, just not sure how that would effect the radio..

Agree, pull the CID connector and look for cooties.

The original radio power is a direct feed from the ignition switch and appears to have no interface with the Lambda/CDI.

However, if someone were to have tapped their radio into an un-switched power circuit, like the O2 relay circuit (which feeds the glove box light and is right there), maybe issues with the radio could do some weird things to the juice feeding the Lambda/frequency valve/02 relay?

The CDI is fed power through fuse 22, a 25 amp fuse that also feeds the rear engine electric blower. Now, could weird things like a tired blower motor sap some of the juice from the line causing the CDI to not have enough? I doubt it but I don't know. I would think it would deliver the 12V and asked for amp draw till a fuse blew.

The tech info center has an 82 electrical diagram.

[JMH82SC]

No amp and grounds looked good but cleaned none the less. I will trace radio wiring to make sure it is still in stock section of harness. Thank you for the help! When this started to happen my first thought was CDI box ( 6 Pin Bosch ) but I was under the impression they would work if good and not work if bad.

[EdT82SC]

My understanding is this could be a bad voltage regulator. The tach will bounce around, and the engine run rough when the voltage spikes.

[JMH82SC]

Voltage is normal at battery with engine running but have never tested when rough running condition is present. I can install my volt meter to monitor voltage when it happens next time.

Jon

[Bob Kontak]

Originally Posted by EdT82SC

My understanding is this could be a bad voltage regulator. The tach will bounce around, and the engine run rough when the voltage spikes.

It's a good thought. Mine never ran rough when the V reg was tired but it did vary the voltage at the battery all the time. Tach jumping was intermittent. Maybe mine was just tired and had not gone totally nutso. Maybe if it allows high spikes of voltage?

On a side note, the tach wire goes through the 14 pin connector - #12. Worth a peek there but doubt it would make the engine run bad at the same time??

[tirwin]

When this started to happen my first thought was CDI box ( 6 Pin Bosch ) but I was under the impression they would work if good and not work if bad.

Can't say for sure with the Bosch - mine had PermaDoom. I drove it from the house, it was fine. Went to leave a couple of hours later and no start. Swapped fuel pump relay and the car started. Got almost home and it died driving. It started a couple more times once I got it home but wouldn't stay running for long.

Edit: A couple of weeks before the PermaTune joined the choir invisible I was chasing a problem that seemed to be a result of installing a wideband O2 sensor with a narrowband output fed into the ECU. I had a loss of power similar to what you described but no bouncing tach needle. I ditched the aftermarket O2 and installed a regular Bosch. Adjusted the mixture after that (with help from Jim, Bob, JR and others) and thought all was well. A week later, dead CDI. I'm not saying it's the same thing for sure but I find it suspicious that we had 2 similar symptoms (radio quit and intermittent loss of power). Maybe you and I should at least go buy lottery tickets.

The VR idea is interesting. Could a faulty VR affect or damage the CDI or cause premature failure? Might be something to ponder further.

[ischmitz]

Bosch CDI boxes can fail with intermittent symptoms. Definitely get a voltmeter and check the reading while the issue happens. Over-voltage is really bad for the CDI box. What is the condition of your battery?

[tirwin]

Hey, Jon -

Well this thread got me troubled. I have driven the car 3 times since Friday when I got the new CDI installed. I didn't notice anything bad but I thought maybe I was missing something. So tonight I backed the car out of the garage and closed the garage door. With the headlights on and the car running I rev'd the engine. With the car just 3-4 feet from the garage door I noticed the headlights brightened with the RPMs. I'm not sure I would've noticed it driving but it was definitely noticeable once I was looking for it. I do have the headlight relays installed.

Then I went for a quick drive. About halfway through I tried to turn the radio on and it wouldn't. I had wired it directly to the fuse panel with an inline fuse. Pulled the fuse and it was fine. Started the car again and this time the radio came on. Weird. I decided to check battery voltage:

12.61 off
13.78 idle
13.70 2k rpm

Checked fan belt and it seems fine.

I am thinking I might have a bad VR or bad ground somewhere. (Just because I can't find it doesn't mean it's not there.)

Just wanted you to know what I had found. Don't want you to fry your CDI if you can avoid it!

I'm running out of time to do more troubleshooting tonight. Have to leave tomorrow morning for a business trip this week so unfortunately I won't be any more help for a few days.

Hope this helps...

Edit: Check out the posted dated 01-27-2012 | 11:55 AM by user Perma-Tune in this thread:

911 SC no start Perma Tune question

I am on my phone so I can't see the post number. Looks like a good write-up.

[wwest]

Voltage measurement, and/or voltage SPIKE detection, back in the engine compartment.

Battery represents one HUGE "sink" for voltage spikes, has a very definite "smoothing" effect on voltage at the battery itself vs all the way back there in the engine/(alternator) compartment. Always clean/burnish battery terminals, battery chassis connection, engine/chassis connection, battery/alternator interconnect point at the starter 's battery supply terminal.

My experience with random over voltages in my '78 Targa, ~90,000 miles, began with the e-brake warning light latching on.

Then the engine started cutting out (rarely) seemingly randomly.

New battery, problem abated for a goodly amount of time...Targa rarely driven.

Engine cutting out randomly turned out to be related to bouncing tach.

Just a guess, but the guess is that whatever was causing the tach to bounce would sometimes "hit" the rev-limiter, which would then latch up, keeping the fuel pump disabled.

Car would not restart unless/until ignition was completely switched off, removing power from the rev-limiter, for a slow count of 10.

At the end of the day I determined that the voltage regulator was not faulty.

It was the alternator itself, internal stator insulation was mostly burned away.

Hit a bump in the roadbed, one of the 3 stator windings shorts out momentarily, VR increases rotor current to maintain AVERAGE 12 volt battery level, huge voltage spikes, spiking, back at the alternator.

CDI is "subject" to engine compartment voltages, fails shortly thereafter...

So, why does the tach bounce? The rev-limiter is a complete separate module, only commonality is battery supply voltage and tach input signal derived indirectly from the distributor magnetic pickup via CDI circuitry.

My theory...HUGE magnetic field arising, caused by, shorted stator field, coupled into the nearby magnetic distributor pickup, results in MULTIPLE firing pulses to the CDI. While the CDI itself cannot possibly respond to the burst of pulses, the tach and rev-limiter most certainly can, it appears to me anyway.

If you encounter this problem have a look at the alternator stator winding insulation.

In any case, you need a new alternator, get one with the internal solid state regulator circuits.

My CDI failed due to the extended period the internal 80 volt zener was forced to endure 14-16 volts.

[tirwin]

Originally Posted by wwest

If you encounter this problem have a look at the alternator stator winding insulation.

In any case, you need a new alternator, get one with the internal solid state regulator circuits.

My CDI failed due to the extended period the internal 80 volt zener was forced to endure 14-16 volts.

Williard,

I agree this points to a alternator/voltage regulator kind of problem. You mentioned the e-brake light being on. That happened to me one day too. It didn't stay on long and went out shortly after I noticed it so I dismissed it. What seems to be happening in my case is that these events are brief which makes it hard to catch it in the act. Another possibility is that the "wrong" ground path is being used. Maybe this wrong path is only intermittently active.

At the end of the day, checking the alternator, VR and grounds is worth doing.

A follow-up question: Would the diagnostic tests on the charging system performed at most FLAPS (friendly local auto parts store) be able to identify such a problem with the alternator/VR without visual inspection of the stator windings? I suspect not, but I would be happy to be wrong because it would be easier to know something was wrong before tear into the alternator.

[JMH82SC]

I just wanted to update that my condition has been resolved with a new ignition switch.

[JV911SYDNEY]

Good news but keep an eye on the voltage. "wacko tacho" usually means a buggered voltage regulator. Been there, done that. Didn't end well.