Hi all,
I've got a '95 Chevy G20 conversion van. I replaced the alternator a few weeks ago because the voltage gauge was indicating low voltage - appox. 11.5V. The auto parts store confirmed the alternator diodes were defective and could not sustain a charging voltage.

Here's the problem: With the new alternator installed, the voltage gauge reads @ 14.7V when the engine is first started. After a short time, the voltage decreases to approx. 12V or less (confirmed with a digital volt meter and read at the battery terminals). This happens at idle, with accessories on. I know the voltage should not decrease below 13.3V or so to the battery.

The voltage gauge will also fluxuate while driving. Sometimes it reads as normal (@14V), and then it will dip down.

Since the alternator is new, where else can I look for a possible problem? I've checked the water in the battery and all cells are full. Not sure where to go from here.

Thanks in advance to all.

Chris

 

Make sure your battery terminals are clean by removing them from the battery and cleaning as needed.Then have your battery tested by load or conductance.You could maybe see if the battery is bad yourself by measuring the cell to cell voltage with your voltmeter.If the voltage varies by more that 2/10 of 1 volt the battery is likely bad.On a fully charged battery the voltage should be 13.2 volts or 2.2 volts per cell.

 

Run the vevicle at 2000rpm and check votage output. Now turn on all acc. and recheck at 2000 rpm. If votage is within .5 volts of first reading, there may be no problem. If votage drops more than .5 volt, alternator is to small or is defective.

 

Thanks for the responses. I did a few more tests today with a DVM. The ground is good between the negative battery terminal and the vehicle body as well as the engine block. Still have to do the 2000 RPM voltage test. Trying to figure out how to regulate the RPMs without a tachometer. Assuming a good guess is close enough.

I did notice that the side-mount battery + connection looks to be rust colored though it is nice and tight. I'll take it off tomorrow, clean it up really well and reinstall. The negative battery terminal connector looks nice and aluminum colored. Will post back with results.

Chris

 

Do not assume a battery connection is corrosion free because it looks good on the outside. I had a side post terminal the other day that looked fine. But the terminal was corroded so hard to the battery that I could hardly get it off.

 

Davo/Car Nut – The side terminal connections to the battery were dirty. I had to use baking soda and water to clean them prior to reinstalling. Tightening all connections again seemed to improve my voltage reading a bit.

TopTech – I did the 2000 RPM tests you suggested and the voltage fluctuation did not appear to exceed 0.5 volts.

One thing I did see was that the voltage read 14.00 volts with all accessories off (@ 2000 RPM). When I loaded the battery with all accessories and high-beam lights, the voltage dropped to less than 13 volts (at idle). When I got back into the car to rev up to 2000 RPM, the voltage on my digital meter began to climb steadily until it topped out around 13.50 volts (@ 2000 RPM).

I’m making an assumption that this is a normal scenario, and that the engine has to be running at a certain speed to “turn on” the charging circuit of the alternator. Is that correct?

Also, I took a few amperage readings using an inductive clamp-style amp meter. Not sure how to interpret the data, though. Here’s what I get with the engine at idle:

Accessories off:
Battery Positive (heavy gauge to starter) = 0 (zero) amps DC
Battery Positive (small gauge to junction block) = 1.2 to 2.0 amps
Battery Negative (heavy gauge to ground) = @ 5 amps DC

Accessories on:
Battery Positive (heavy gauge to starter) = 0 (zero) amps DC
Battery Positive (small gauge to junction block) = 1.2 to 2.0 amps
Battery Negative (heavy gauge to ground) = @ 39 amps DC

Could not check the amp readings at 2000 rpm, as my wife wasn’t around to lend a hand.

 

when you checked your ground from battery to engine/body...how exactly did you perform this test? voltage drop is WAY more accurate than an ohm meter. so...operate the engine...load the charging system by turning on some accessories and the place positive lead on the negative battery terminal and the negative meter lead on the engine...or even the housing of the generator. 100 millivolts or less would be nice...this same test can be performed on the positive side of the circuit...positive meter lead on the gen output terminal...negative lead on the positive battery terminal. again...less that 100 millivolts would be desirable. this will tell you if you have any resistance that you've missed.

as for low output from the generator at idle...that's pretty normal. ac generators need to spin pretty fast to achieve output potential...this is why when you sped the engine up during the load test of the charging system, you saw the voltage climb. now...a declining battery will make this worse...so as davo stated earlier, have it properly tested by conductance or load method. the battery acts as an electrical "shock absorber" so the better the battery condition, the better the stability in the system overall.

 

Thanks, Carguyinva.

Initially, I checked the ground via resistance (ohm meter). As you suggested, I did two voltage drop tests:

1> from the alternator casing to battery negative terminal = .35 millivolts
2> from the alternator BAT + terminal to the battery positive terminal = .35 milivolts.

All looks good in that regard. After reading your comments about the battery acting like a shock absorber, I suspect the battery may be the cause of the voltage fluxuations. Heck, I've tested the $#@& out of everything else

Here's the specific gravity measurements I got for each cell:

Cell 1: 100%
Cell 2: 100%
Cell 3: 75%
Cell 4: 75%
Cell 5: 100%
Cell 6: @ 65%-75%

The water level in each cell is perfect, so my thoughts are the battery should be replaced. Would you agree?

Thanks again,
Chris

 

altho specific gravity doesn't tell the complete story on a battery's ability to provide electrical power. SG tells you the condition of the electrolyte but you can also have internal resistance which prevents the transfer of electrons. conductance is the most accurate determiner of internal resistance...and more and more shops are using this evaluation method. pretty much all the dealers have to test this way now for warranty purposes. based on your specific gravity measurements, i agree with your conclusion...the battery is on it's way out. there is also a TSB from GM that talks about voltage fluctuation at idle with load...causing "flickering" of lights, etc...it's considered normal as long as the battery and charging system are ok. so...if a new battery doesn't completely cure your problem...it may just be the way it is. let me know if you want the TSB#...i'll get it for you.

 

I'd like to have the TSB for my records, if it's not too much trouble for you to get.

Also, I'm going to have Sears do a full load test this weekend, but 99.99% going to replace the battery anyhow just for peace of mind. If that doesn't clear things up, then the TSB will be the final explanation.

Thanks for providing so much valuable information. I've learned quite a bit during this troubleshooting process, which makes doing it very enjoyable and rewarding.

Chris