dagersh

Holy Cow! The mystery deepens....

So, we replaced what I thought was a faulty relay, but the same symptoms occurred. So we started looking elsewhere. As it turns out, there is something going on with the alternator! We disconnected it, and the lights worked fine at all RPM's. Great, easy fix, my alternator was in decent shape, but we were able to source a nicely rebuilt Bosch at a fair price. We then installed the new unit... same cut out of the lights at 3200 RPM, again disconnect the alternator, the lights work fine.

We are at a loss....

worf928

Have you checked all of your grounds as I suggested?

dagersh

100%, all over the car!

Mrmerlin

Hi Mike,
Can you please tell me what the output of the alternator is while its running and when the lights shut off .

Take the reading from the hot post,
and from the battery terminals both on positive terminal and negative,
then switch the ground DVM lead to the chassis.
this will also give more info on voltage drop.

NOTE that you may need to replace one or more of the ground straps,
since the engine rebuild it would be good to post pictures of the bottom of your engine installed in your car.

worf928

What Stan wrote --^

It would also be useful to set up your volt meter on the hot post so that you can see it from the driver's seat. Then you can monitor the reading as you rev past the 'trouble spot.' A digital meter may not be 'fast' enough. A cheap-o analog meter with a needle might be better.

This test will help determine if you have a power problem or a ground problem.

One other question I have is: does the operating condition of the two electric cooling fans modulate the behavior? In other words, does this happen when both fans are not running? If so, then do the above test with the A/C button on - thus forcing the fans to operate - and see if the behavior changes (for example occurs at a lower rpm.)

worf928

How are you disconnecting the alternator?

Kiln_Red

^^^This.^^^

Simply checking resistance isn't enough information. It's critical to know the ability of a given circuit to sustain load. The process outlined by Stan will help you to do that.

I suggest to inspect the headlight switch plug connector, too. Has there been recent instrument pod work? The battery should always be disconnected when working in the pod. There is a wire on the headlight switch that has constant power. If that wire shorts against the pod cover, it can affect many circuits in a hurry! It can especially affect headlight operation obviously. If an event such as this has transpired, there will be evidence. I consider this a less likely scenario, but it is worth the investigation as it is quick to visually verify the integrity of that wiring.

I suspect there may be a visible arcing show going underneath your car at the specified RPM range. Do you have a helper close by? What has been the behavior of your volt gauge once the lights start misbehaving?

Landseer

Battery groundstrap has failed in at least 15 cars here on Rennlist.

Corrosion of the flat braid occurs beneath the plastic shroud.

Continuity check shows no resistance. But not enough current can pass when under load.

Before troubleshooting any further, temporarily install a replacement. Get a long round one at Advance Auto for $10.

That would eliminate neg bat ground as a factor. Also ensure red connector is clean and little red wire at battery is clean and tight, if GTS has that setup.

dagersh

Hey Guys,

Turns out it was a faulty "Wittrin" replacement relay. A lot of work for nothing, but my electrics are really tidy!!!! I got a tested used relay out of a 928 S4 from 928 International, and it immediately solved the problem. Putting the car back together now and she will be back on the road!

We also somehow solved a draw problem along the way and the car now has a minimal draw on the battery - so I guess a lot of work and research did pay off!

Gersh

GregBBRD

.....almost a year later.
Ouch.

Mark Anderson

I didn't think Wittrin made one for the 87-95 cars. Maybe you had an early one installed

dagersh

I was wrong, it was not a Wittrin. The relay was actually made in W. Germany and it was the second one I sourced from Vertex, exchanges the first one as it did not work either. Yes, its been almost a year, and things got fuzzy. I will reach out to the guys at Vertex, I know they will take the relay back.

Now to finish buttoning her up and taking it for a nice long rip through the mountains!

Gersh

SwayBar

Nice job, and thanks for updating the thread.

dagersh

Took her out, ran through the gears... Dreamy!