Battery light comes on during acceleration for a brief period.

[Jeff K]

On my 81 Auto, I changed the alternator to a brand new DMC Delco 105amp unit. I cleaned all the contacts before installing.

During acceleration, not at idle, the battery light comes on briefly and goes away. My idle voltage is not great with the fans on and ac - but the light comes on only during acceleration.

My first thought was loose belt. The belt is brand new and tight. Unless it has to be crazy tight.

Any suggestions?

Jeff
PS - I bought the Hervey ground strap but have not installed it yet.

[Michael]

If the belt is tight, then check how it's sitting in the pulley. Better yet, snap a pic and post

[Ron]

Probably the alternator. The causes for this symptom can be tested for on the care at most auto parts stores (free).

[Jeff K]

Probably the alternator. The causes for this symptom can be tested for on the care at most auto parts stores (free).

Ok gotcha .. I hope not. It's brand new, but it could be defective.

[Bitsyncmaster]

What does your voltmeter show at idle and accelerating?

[sdg3205]

Defective happens. My new DMC alt lasted 100 miles.

[jawn101]

I second the slipping belt notion, I have the same issue right now and the alternator is known good. It clocks a steady 14.5-ish volts at all times, but between 3000-4000 RPM accelerating on the highway the light flickers until upshift. My belt is on as tight as it will go (literally the alt is all the way to the end of the bracket) but I can still deflect the belt at rest more than I'd expect to be able to. I bought a new belt in a slightly shorter length yesterday and will put it in this week. I am hoping it corrects the issue.

[WelmoedJ]

New belts have to be run in, just like new cars.
After some period of time the belt has stretched and needs retension.

[Jeff K]

1019310194

Ok, this are at Idle -
1st Pic is no ac on, daytime
2nd pic is ac on, daytime

Running with no ac about 13
With AC about 10.5 (I'm guessing)

Jeff

[ccurzio]

That's the lowest reading I've ever seen on that gauge. Definitely seems to say alternator.

[WelmoedJ]

+1
After seeing the pictures of the Voltage.
Needle should be at or over the first mark above 13V.

[dmc6960]

Whoa now. Before condemning the alternator lets verify with a good voltmeter on the jump start posts what the REAL voltage is. If it still shows as low as the dash gauge then you got problems.

[Jeff K]

I forgot to mention.. I replaced the previous alternator which exhibited exactly the same results.

The alternator looked pretty beat so I figured it was a good idea.

I'm my 83 the voltage is rock solid steady about 14, with everything on.

[jawn101]

If an alternator change produces the same results in the same car, it does sort of point to a wiring or a gauge issue. You should definitely do like Jim said and test your actual voltage with a known good multimeter before leaping to the conclusion that it's the alternator.

[Jeff K]

If an alternator change produces the same results in the same car, it does sort of point to a wiring or a gauge issue. You should definitely do like Jim said and test your actual voltage with a known good multimeter before leaping to the conclusion that it's the alternator.

I agree. Where should I connect a meter for a good reading?

Thanks

[dmc6960]

I agree. Where should I connect a meter for a good reading?

Thanks

Jump start post in the engine compartment for the 12V. Most metal objects on the engine should give you a suitable grounding point. If your coil cover is removed you can use the bulkhead ground lug.

[Jeff K]

Jump start post in the engine compartment for the 12V. Most metal objects on the engine should give you a suitable grounding point. If your coil cover is removed you can use the bulkhead ground lug.

Great thanks! I will do that and post my finding.

[EarlHickey]

I second the slipping belt notion, I have the same issue right now and the alternator is known good. It clocks a steady 14.5-ish volts at all times, but between 3000-4000 RPM accelerating on the highway the light flickers until upshift. My belt is on as tight as it will go (literally the alt is all the way to the end of the bracket) but I can still deflect the belt at rest more than I'd expect to be able to. I bought a new belt in a slightly shorter length yesterday and will put it in this week. I am hoping it corrects the issue.

The puny belts, had this car remained in production, would have been upgraded to multi-ribbed belts and likely a single serpentine with a spring loaded idler tensioner.

There is room for such a setup and some enterprising and energentic young person should market such a much needed upgrade.


pssstt... the serpentine can be done with reasonably common junkyard parts unless my calibrated and lucky eye is out of whack

[Chris4099]

Another cause of the light flickering at higher RPM is a bad positive connection to the alternator and battery. I had this and it went away after cleaning and tightening the positive connection to the alternator.

[ramblinmike]

Remember the cluster grounds that Bill found under the dash? Could that cause the poor voltage reading?

[dmc6960]

Remember the cluster grounds that Bill found under the dash? Could that cause the poor voltage reading?

Yes, that, and poor contacts on fuse #5. Only relevant though once it is confirmed if the actual car voltage is much higher.

[Jeff K]

Remember the cluster grounds that Bill found under the dash? Could that cause the poor voltage reading?

I don't.. can you send me a link to a post or pic if you'd be so kind.

Thanks
Jeff

[jawn101]

Remember the cluster grounds that Bill found under the dash? Could that cause the poor voltage reading?

I can't remember where those were - they were seperate from the ones at the HVAC panel, weren't they?

[dmc6960]

I don't.. can you send me a link to a post or pic if you'd be so kind.

Thanks
Jeff

Bills research and diagram shows the following things sharing a ground with the instrument cluster...

Door Plunger
Steering Collumn
Inertia Switch
Turn Signal
Mounting bracket for wiper delay, flasher, and door buzzer
Windshield Wiper Motor

From that list, the two big powerhogs are the Inertia switch (which ground the fuel pump) and the wiper motor. Since the wipers are not going all the time, a good test would be to unplug the inertia switch, and ground the fuel pump through a temporary jumper. That will eliminate the one big load on that ground and tell you if the ground is a problem.

Also remove and clean or replace fuse #5.

One final note, it is possible that every connection in the car is good, except for the connection at the instrument cluster itself. Those can get very bad, and are a very poor design (fine when new, not with 30 years of age though).

Here is Bill's ground diagram...

http://dmctalk.org/attachment.php?attachmentid=4441&d=1314237798

[jawn101]

Oh, interesting. So the ground bundle that includes the inertia switch isn't a direct ground point, but rather one of those big taped up bundles that eventually feeds back to the bulkhead? That isn't at all what I was expecting to see but it's interesting and makes sense. The fuel pump really does deserve better than that, I guess that's as good an argument as any for the ground bus idea.

[dmc6960]

The fuel pump really does deserve better than that, I guess that's as good an argument as any for the ground bus idea.

I dont see that alone as reason for a whole ground bus. While the idea has some merits I'm not sold on every aspect of it. If done like Bill's it introduces ground loops can be a MAJOR electronics issue. Of course since Bill has ditched nearly every electronic component from his car there aren't any left to get affected. What I do agree with without question though is isolating the inertia switch's ground to its own attachment point.

[jawn101]

I dont see that alone as reason for a whole ground bus. While the idea has some merits I'm not sold on every aspect of it. If done like Bill's it introduces ground loops can be a MAJOR electronics issue. Of course since Bill has ditched nearly every electronic component from his car there aren't any left to get affected. What I do agree with without question though is isolating the inertia switch's ground to its own attachment point.

Interesting. I never considered a ground loop. I suppose it could be avoided with careful planning but you're right, the important thing is isolating that pump well.

Anyone given any thoughts to the wisest place for an attachment point for it? If you run it straight to the frame then you've got one more thing to remember should you ever seperate body and frame. Just thinking out loud :)

[dmc6960]

Anyone given any thoughts to the wisest place for an attachment point for it? If you run it straight to the frame then you've got one more thing to remember should you ever seperate body and frame. Just thinking out loud :)

Easiest and simplest place would be the front frame ground point. Its on the drivers side where the upper radiator bracket bolts to the frame. There is already a ground there for a few of the devices in the front of the car. You could easily run the new wire through the grommet for the speedometer cable.

My fuel pump is currently grounded here, but it does so directly, no longer going through the inertia switch (my car never had one when I bought it). After a ground failure to the fuel pump a few years ago I ditched the inertia switch bypass I had in place and grounded it directly to the frame.

I suppose my own car could have a very bad ground at this point (junction behind A/C duct) since my fuel pump can no longer operate from it (I originally thought it may be a bad connection in the washer bottle area), my dash voltmeter is constantly low, and my wipers never go fast enough for my tastes.

[Bitsyncmaster]

http://dmctalk.org/showthread.php?1010-Fuel-pump-voltage/page2&highlight=frame+ground

Look at post #19 for my solution. I know your voltmeter will read higher if you get the pump circuit off that ground juntion.

[Jeff K]

Ok, I took a voltmeter to the engine compartment.

1036210363

Both are at Idle, second (and lower) is with AC on at idle.

It seems the Voltage is correct in the engine compartment, but not the cabin.

I though it might be the dash only, but I used a cigarette lighter voltage tester, it was the same as the dash gauge. Also the headlights are pretty dim too.

I replaced fuse #5..

Any ideas where to go from here?

Thanks
Jeff

[dmc6960]

Remove your rear wall and coil cover. Disassemble and clean all contacts on the grounding bulkhead bolt on both sides.

[Jeff K]

Remove your rear wall and coil cover. Disassemble and clean all contacts on the grounding bulkhead bolt on both sides.

Funny, I had the rear wall out today to tighten the engine hatch cover bolts, they were loose.

Thanks - I'll do that tomorrow.

Jeff