Voltage woes

The wife's new 1998 GMC Safari van has developed a voltage issue. It started with the battery occasionally being too low to start the engine after sitting overnight. The past few days the voltage has started dropping while driving, and in two cases the engine has started to loose power, probably because it was too low to properly run the computer. After a second or two it will go back up to the normal 14 ish volts and everything is fine.

I cleaned the battery terminals and grounds last night, but it's still having issues. This afternoon the kids fell asleep in the van, so I sat it it with the engine idling when I got home. The battery voltage dropped from normal down to as low as 10. Revving the engine did not bring it back up, but a few seconds after letting it return to idle it went back up for a few minutes before falling again. When I shut it off it was back at 10, and when I went to restart it a few minutes later it barely started.

Since it has been hot the ac has been on full blast, and the van has running lights so there is a fair load on the electrical system when this has been happening. What says the hive mind...alternator? Battery? Both? Something else?

V belt, or serpentine? I'd make sure the alternator belt isn't slipping.

No squealing, but I will double check.

Have the alternator tested, its probably lost too much smoke.

Nope, not slipping. I just drove it around the block. 10 volts with the key on, barely enough to start it. Went up to 13.5 with the engine running. I cranked on every fan, light etc and it dropped to 12 or so. Turn them off, back to 13.5.

The alternator needs to be pulled to test, right? Being a van that is easier said than done. If I pull it I should probably just replace it.

Nothing to add other than I thought this was going to be about the chevy volt not holding up well, but that would be 'volt age woes'.

Sounds like you're pulling an alternator; those numbers are pretty poor.

+1 for bad alternator, the numbers pretty much confirm it.

What type of battery does it have? I've found that maintenance free batteries do stuff like that right before they fail for good.

Not sure what type of battery, but I believe it has removable cell covers so I don't think it's maintenance free.

The thing that makes me question my alternator diagnosis is the rapid voltage drops when the engine is running. If the alternator stops charging shouldn't the battery be able to keep the voltage up for a little while?

If the alternator stops charging the voltage will immediately drop to whatever voltage the battery can supply (~12.5v) or less.

But in this case it was dropping down to 9 or 10 volts which is why I am questioning myself.

Now that I think about it, when I cleaned the battery terminals the negative lead may have been slightly loose. It's possible that it was randomly giving no contact to the battery. That would explain the large and sudden voltage drops. Could this have fried the alternator?

Usually what happens with this sort of symptom is a failure of the regulator/rectifier section of the charging system. You may have kind of okay output when cold but as the components of the system gets hot the output voltage goes off a cliff and the battery has to hold things up. Eventually the battery's bucket-o-watts gets depleted and things like fuel pumps start dropping off.

My personal thought is that an otherwise healthy charge circuit with good wiring and connectors and all that should be able to throw voltage at a bad battery like it was meant to and the battery itself just won't take charge. Perhaps an internally shorted battery could fail in a way that sends a lot of power to ground but if you take your battery out of the circuit and use a second known good battery and it acts the same way? Something is up with the charge circuit. Get a battery on the ground, pull the neg cable off the in-vehicle battery, and "wire around it" using jumper cables so the installed battery isn't actually part of the circuit.

Since your alternator contains both the stator and the regulator/rectifier, pulling the alt and having it tested should tell you a lot after you've used the second battery to make sure the battery isn't the issue. And you won't have the fun that I had a few weeks ago with my VFR; regulator was actually good but the system acted like it was bad... and it turned out that the stator was fried but functional when cold. Either way a bench test should show this up. If it tests good on the bench you'll have to start chasing through the wiring.

In reply to bgkast:

I think intermittent connectivity of the main cables could make life hard on the reg/rectifier so yes, in effect, the alternator could be damaged from bad connectivity.

bgkast wrote: Not sure what type of battery, but I believe it has removable cell covers so I don't think it's maintenance free.

Its probably not maintenance free then, but it would still be a good idea to load test it before pulling the alternator.

As gameboy said, when the alternator stops working, the voltage will drop to battery volts, and will actually go a bit lower still due the system electrical loads. But easily dropping below 10v makes it sound like an unhealthy battery. It may just be severely discharged from a weak alternator, or it could be a problem in and of itself.

Also check the condition of the cables from the alternator to the battery. Check the connections and look for corrosion and rotten spots in the cables too.

Thanks. I will try the battery on the ground idea. The alternator needs to come out to be able to access and check the wiring to it, so it sounds like it's coming out.

I ran it off the Miata's battery last night. It idled perfectly at 14 volts for half an hour or so. I then hooked the van's battery back up. It idled perfectly at 14 volts. Doh! It's going into the shop for AC work at the end of the week, maybe I'll have them look at this too.

Do you have a toaster or carbon pile battery tester? They're handy for this sort of things, diagnosing funky batteries and alternators.

No, just a toaster. I'll let the shop look at it tomorrow since it's going in anyway

If you have a toaster and a DC amp clamp, you can hook the toaster to the battery, amp clamp over the alternator wire, and with it running, turn on the load and note how the alternator reacts.

This is assuming the battery alone can pass a few toaster cycles in a row.

It really sounds like the alt is shot.

The big named parts stores usually have a machine that can test the charging system and battery on the car. May not be as accurate and precise as individual testing but can narrow down possibilities.

Another easy test to do if you have a multimeter with a AC voltage setting, check and see if there's any significant AC voltage on the line. If there is, you've got a blown diode in the rectifier portion of the alternator.

In reply to Karacticus:

Not all meters will do that correctly, test on a battery or other trusted DC supply first.