ammonman

All,

My alternator is giving symptoms of either a bad voltage regulator or bad diode(s). The alternator light/exciter circuit passes the warning lamp test and and the dash voltmeter indicates approximately 13.5 volts at idle with low to moderate load (stereo, headlights, fog lamps,etc.) I have verified this voltage at the jump post with a DMM. In normal driving with minimal load on the electrical system the dash meter indicates 13-14VDC consistently. Once you apply a large load from say the HVAC blower or the electric defroster, the dash voltmeter drops to 12-12.5VDC when above idle in the 2000 rpm range dropping to 10VDC when at idle. My thought is that either I have a diode or diodes going bad or the voltage regulator is starting to fail. Any way to test these two components without removing the alternator?

Thanks

Mike

UPDATE- The battery was toast. Hydrometer test per Dr. Bob showed 1.200 S.G. which is barely half charged. The battery had been on the charger in manual mode for six hours when the readings were taken. In automatic it would simply stop charging as if the battery was fully charged. Voltage after charging was 12.7VDC. Parts store electronic load test showed good battery. I pulled the alternator and had it tested by the rebuilder that did the clean-up work back in March or April. He ran the load all the way up to 160A and the voltage never dropped below 12, even though the rated output is only 140A. He also checked the diodes and said all was well.

I dropped a Duralast Gold battery in today and with the same heavy load as the original test (headlights, fog lights, blower, and de-icer) the jump post voltage never dropped below 12.75VDC. I am planting the victory flag on this one. Thanks to all for the help, especially Dr. Bob and Alan. Lesson learned about checking with the hydrometer for the final word on battery health.

Mike

Alan

No not really for most users (with an oscilloscope yes) - the symptoms you describe are most likely related to:

1) Belt not tight enough and slipping at idle
2) Voltage regulator failing
3) Possibly 1 phase diode failure but this is accompanied by a reduced best case voltage (you don't seem to have this).

Except for 1) you need to remove the alternator anyway and you can have a good alternator shop (but only a GOOD alternator shop that specialized in such work) determine what needs to be fixed - likely the regulator unlesss them belt in insufficiently tight (not loose - just not really-really tight).

Alan

ammonman

The belt is tight as that was the first thing I checked. Sounds like the most likely candidate is the voltage regulator. In the past I've always experienced sudden failure of the regulator. This slow loss of regulation is unexpected.

Guess I'll pull the alt this weekend and get it reworked while I'm out of town next week.

Mike

Alan

It is not so uncommon on 928's and note the alternator belt needs to be tighter than you might expect...

see if its getting unduly hot at high load idle...

Alan

ammonman

I'll check that Alan. When I first got the car back in the spring, I pulled the alt and had it cleaned by a local shop. The PS leaks had coated it in oil and then dirt, sand etc. The duct was missing as well so there was all sorts of cr@p that had been sucked into the case. The shop cleaned it and put new bearings in it while they had it open. Everything else checked ok at the time. When I installed it I also replaced the missing duct so it should be getting good cooling air.

Mike

76FJ55

My first guess would be brushes, which you may be able to get to with out removing the entire alternator. I haven't been into my Alt yet but it is is similar to the other Bosch units I've worked on you can remove the brush assembly which is integral with the regulater but first removing the the cooling housing from the rear of the alternator to expose the back side. then the reg/brush assembly is typically held in by two screws. If this is the same I would pull it and check the condition of the brushes first. If they look good then you will probably need to pull the Allernator.

Mrmerlin

when you drop the alternator pull the regulator from the unit 2 screws and then see if it need new brushes this usually is the case when the charge stops under load, if the brushes are good then as Alan says you might have a bad diode.
If the system wont charge higher than 13.4 then this would indicate a possible bad diode

dr bob

Check the voltage at the battery terminals and compare with the jump post and alternator terminal voltages under the same conditions. That will help identify where voltage is dropping. I have an extra-long test wire with alligator clips that I can use to check voltage drop directly too. A bit of sleuthing with a meter will tell you a lot about what things have built up resistance over the years.

FWIW, I've watched the das gauge move around between 12.5 and almost 14 indicated volts, while a DMM attached to the jump post sat at 13.8 v the whole time. Loads vary with fans and ac, headlights and cooling fans, jump post steady as a rock. Cleaned all the grounds and that helped some. Biggest improvement came with a new battery, one that didn't suck so much current just to stay charged. So do a little connection cleaning, and do a quick check of the fluid in the battery. There are cheap (<$2) little floating ***** testers for specific gravity that can quickly tell you a lot about your battery's health. That tool should be in toolbox, but in a plastic bag of course.

WallyP

Mike,

As Alan says, the alternator belt has to be tighter than you expect. The old "half-inch with thumb pressure" doesn't cut it on the 928. If you can twist the alternator belt more than fifteen degrees by hand, it isn't tight.

ammonman

I went back through my receipts and the local shop that cleaned the alt in the spring also put in new brushes and new bearings.

I am going to pull the battery and get it load tested this weekend to eliminate it as a source of trouble. I have cleaned all the grounds except the those under the pod and the two on the back of the block (can't fin the little buggers.) I took some readings last night at the jump post and compared then with the apparent dash gage readings. There appears to be about a 1VDC difference. Pulling the pod to clean the grounds, re-lamp the cluster, clean the edge connectors and re-finish the cracks is on my to do list I just haven't made it that far yet.

I may need a new belt as the adjuster is nearly bottomed out and I don't think I'm quite that tight. However it is already much tighter than the 1/2"-with-the-thumb setting.

I have also cleaned and checked the 14-pin connector, the blower resistance group, blower connections, and recently had the CE out to repair a burnt blower fuse. Until I found and fixed the blower fuse I was sure my voltage drop was all due to the blower as it seemed to be the only thing that caused the voltage to droop. Now there is a new 30A in-line fuse pulling blower duty, but I am still seeing this voltage droop under heavy load. I suspect that after I load up the alt and the voltage drops the alt gets hot, because just shutting the loads off does not allow the voltage to rise back to pre-load levels. Get in the car the next day and it fires right up and the dash gage goes right up to 13.5 or so. If my alt belt is stretched too far it could be that as the belt gets hot it starts slipping. Guess I'll see if I can get one locally and check the battery before I drop the alternator out.

Thanks for all the help folks. I love my 928, but without my automotive skills and you all here on the list it would quickly turn in to an exercise in frustration and a short trip to the poor house.

Mike

Mike Frye

QFE

Before you go pulling things out and replacing them re-read this post. You said you confirmed voltage at the jump post at idle with light load but then you said with the HVAC fan on it drops. Did you go back and confirm the voltage at the jump post and the battery?

The gauges in these cars are good to tell you something is going on, but don't consider it gospel until you confirm what it's telling you with some other test. My voltage gauge dips when I use the power windows or sunroof, it dips when the directionals are on, it dips and stays down when the blower is on, but I've confirmed at the jump post that the dip is barely perceptible. It's like the difference between 13.5 and 13.8 vdc or something like that.

Always confirm your diagnosis before treating the root problem. You may find that the symptom is not telling you what you think it is. It's what Dr. Bob would do.

ammonman

Voltage confirmed.
I should have included "under all observed load conditions...: At times the voltage at idle will drop low enough that the dash lights get extremely dim and then brighten some with increased revs. If this symptom was not present I would not be looking quite so hard.

Mike

ammonman

Well, I put some more tension on the alternator belt today and took a bunch of voltage readings with the DMM. I also took the battery and had it load tested. It checked as good.

With the car running at idle and no additional electrical load I had the following readings:
Battery=13.83
CE Panel=13.73
Jump Post=13.89

I then started loading up the electrical system in the following order:

HVAC Blower, Electric Defroster (high), head lights, fog lights.

I took readings at each stage but the final readings at idle with all of that load on the alternator were:

Battery=12.08
CE Panel=11.65
Jump Post=11.95

When I was done taking the readings and with the load still on the alternator I checked how hot the alternator had got. It was too hot to keep my hand on the case.

Based on the readings I know I need to do some work on the connections to the CE panel. I cleaned the connections at the battery and at the Jump Post back in the spring. I guess I need to take a closer look at the terminal ends and the wire condition at the Jump Post and see if I can get better voltage to the CE panel. I just had the panel out and cleaned all the connections while fixing my burnt blower fuse problem. Guess tomorrow I'll pull the alternator. What do y'all think? Is it the regulator or the diodes? I can replace the regulator my ownself, diodes too if I knew where to get them. Anybody got a source for the diodes or the diode assembly?

On a positive note my ride height setting is coming along nicely!

Mike

dr bob

Mike--

Your symptoms point to a tired battery. Before you trust the parts store 'load tester' consider that it's the same tester they use to avoid replacing batteries under warranty. Grab one of the little testers for specific gravity of the fluid in the cells. Maybe a dollar at the battery display at Wal-Mart, less than $2 at Autozone. NAPA Store has a real fancy one with calibrated float for less than $10. The cheap one with the little ***** tells you the same thing as the fancy one though, and takes less space in the toolbox. If the battery is charged, the acid in the cells should not be depleted. If the tester indicates it's depleted, the alternator will work its heart out trying to charge the battery, but the battery will not maintan voltage under serious load. If the tester shows less than good in any cell, replace the battery.

ammonman

Dang it, I knew there was something I was forgetting while at the parts store. Thanks for the reminder. I'll pick one up tomorrow as I need to return the belts I bought today as none of them fit. Figured it was worth a shot to eliminate a slipping belt as part of the problem.

The battery has a sticker indicating April 2008 as the warranty date but who knows how long it sat on the shelf before that.