Electrical charging system question, Alternator? Options

[Dave_Darling]

plugging in the alternator directly into the voltage regulator. (this works, right?)

Nope. You need the warning light in the circuit for the alternator to charge. You won't have the light unless you go through the relay board, so plugging the VR directly into the alternator wiring harness will not cause the alternator to charge the battery.

Also, the light does not show all possible problems. For instance, if the VR gets unplugged, the light will not come on. Which sucks if you're dumb enough not to use the screws that hold the VR onto the relay board... Uhh, not that I know from personal experience or anything. (IMG:style_emoticons/default/headbang.gif) And, as another example, if you have the wrong type of light in the warning light socket, you can have charging problems that don't necessarily show on the light... As mentioned in James Thorusen's article, you want the stock light bulb in there.

I'm sure there are other possibilities as well.

--DD

[rjames]

Nope. You need the warning light in the circuit for the alternator to charge. You won't have the light unless you go through the relay board, so plugging the VR directly into the alternator wiring harness will not cause the alternator to charge the battery.

Also, the light does not show all possible problems. For instance, if the VR gets unplugged, the light will not come on. Which sucks if you're dumb enough not to use the screws that hold the VR onto the relay board... Uhh, not that I know from personal experience or anything. And, as another example, if you have the wrong type of light in the warning light socket, you can have charging problems that don't necessarily show on the light... As mentioned in James Thorusen's article, you want the stock light bulb in there.

Ok, so plugging in the alternator directly into the VR won't charge the battery, but will it allow me to test the alternator at all?

How about the method of jumping the D+ and the D terminal on the relay board as suggested by someone else? Good thing to check next? The belt on the alt seems tight (I've read that I should be able to depress the belt 1/2 to 1", and mine seems more in the 1/2" range).


I've been taking the car out for 30+ minutes at a time the last few days, and the battery seems to stay charged to around 12.5 volts. Could the alternator be 'kind of' working? I'm getting close to just putting one in to eliminate it as the culprit. Also read something somewhere about checking fuse 8 and 9 (or was it 7 and 8?) on another charging issue thread. Maybe I'll search for that thread again too and look up what those fuses are for while I'm at it!

[Dave_Darling]

The light is needed for the alternator to charge anything. Without it in the circuit, the pulley just spins, but no current gets generated. I think the jumpering test will tell you if the alt is working, but I haven't tried it myself.

--DD

[rjames]

OK, now I'm totally confused. I switched the relay board for another one that I was given awhile back. As soon as I plug it in I start the car, and measure 13.7 at the battery terminals. I think my problem is solved! After driving around for a few minutes at fairly high revs I check the voltage again, and now I'm back down to 12.5-12.7 and the turn signals don't light as bright verifying that yes, I have dropped back down to a lower voltage.

Relay board? Maybe? Could the alternator work well when the car starts, but gets 'tired' and can't keep up the charge? This would happen with the other relay board at times where it would show 13.7ish right when starting the car (after it had sat for awhile) and then drop down to 12.something after a couple of minutes.

Should I just change the friggin' alternator already? Could the alternator be working correctly for a minute or two when cold and then lose some of it's charging ability soon after?


Jeez, I'd rather be installing my sway bar then messing with this electrical crapola.

[davep]

It sounds like you have heat related resistance. Where is the big question. A high current through a dodgy connection will generate a lot of heat that will cause a higher resistance. The higher resistance will cause a voltage drop and thus less current, but the temperature of the connection remains high. Realy boards are well known for having poor contacts due to corrosion. You have to remove the tar, clean, then reseal. On one car I had the stud on the engine wall for the battery ground was solid, but didn't conduct worth beans. A battery cable from the ground post to the fan shroud solved that problem.

[Dave_Darling]

Did you try the jumper test yet?

--DD

[rjames]

It sounds like you have heat related resistance. Realy boards are well known for having poor contacts due to corrosion. You have to remove the tar, clean, then reseal.

Hmm seems kind of odd that 2 of them would exhibit the exact same behavior, but I guess it wouldn't surprise me. The tar is in pretty good shape on both of the boards. I could see how in removing the tar I could possible damage the board. Would love to test with a known good board before I go that route. (Anyone? I'll gladly pay shipping).

On one car I had the stud on the engine wall for the battery ground was solid, but didn't conduct worth beans. A battery cable from the ground post to the fan shroud solved that problem.

I cleaned and checked all grounds for resistance, and they're all good. (at least according to my DVM/DOM.)

Did you try the jumper test yet?

--DD

I haven't yet. I am kind of leary of trying this (afraid to fry something else like a battery). But I guess I will barring any warnings from other people.

I'm still wondering if someone can tell me whether or not the alternator could just be 'working enough' to keep things at 12.whatever volts?

[ClayPerrine]

The only risk you have from "full fielding" (aka jumpering the D+ to the DF contacts for the regulator) is if you run it for a long time like that. It makes the alternator run wide open at 16 volts or so, and if it is run for a long time it can boil the acid out of the battery. A 5 minute test is NOT going to harm anything.

Do the test like this:

1. Put a DVOM on the battery posts.
2. Remove the voltage regulator from the relay board.
3. Start the 914.
4. Using a jumper wire, connect the D+ and DF contacts.
5. You should see 16 volts or so on the DVOM. If you do, your alternator is good.


From recent threads, it sounds like I need to do a pictorial, step by step alternator troubleshooting thread.

[rjames]

Do the test like this:

1. Put a DVOM on the battery posts.
2. Remove the voltage regulator from the relay board.
3. Start the 914.
4. Using a jumper wire, connect the D+ and DF contacts.
5. You should see 16 volts or so on the DVOM. If you do, your alternator is good

Clay, thanks for the step-step walkthrough. I learn best that way. (IMG:style_emoticons/default/beer.gif)
And thanks to all who've chimed in, thanks for checking into this thread on an ongoing basis.
I really appreciate all of the help!

Will post results later tonight.

[rjames]

Do the test like this:

1. Put a DVOM on the battery posts.
2. Remove the voltage regulator from the relay board.
3. Start the 914.
4. Using a jumper wire, connect the D+ and DF contacts.
5. You should see 16 volts or so on the DVOM. If you do, your alternator is good

Did the test, and it showed 16+ volts. Hmm... so after doing this test I plug the voltage regulator back in, drive around a bit and what-do-ya-know, it's showing 13.3 at Idle. Nice.

The next day, I drive around and I'm back to showing 12.5 or so at idle, and when the car is off the battery is reading 12.3ish.

The one thing I haven't checked for is when it's idling at 12.3 or whatever, to short the D+ and DF contacts to see if the alternator kicks right in. Although, my guess is that it will. So I'm back to my original question- is it possible for the alternator to work correctly sometimes, and other times not? Can both of my relaby boards be causing problems? Could both my voltage regulators? Is it worth it at ths point to check the alt bulb in the dash to make sure it's the correct one? Are these still available new?

[Thack]

Why don't you take a poll to see how many volts at idle that everyone measures. I thought 12.5 was normal at idle. You should be more concerned what hapens while driving unless you like to watch your car idle. (IMG:style_emoticons/default/smile.gif)

[jk76.914]

I know the regulator is brand new, but I have personally bought "bad" new regulators. The light will go out on the dash and everything too, when the regulator is out. Trade it in and see what happens.

There's something about today and Quality Control which I don't understand.

b

I also got a bad regulator brand new. I got a solid state one from NAPA, and while it regulated, it regulated to a low voltage. I took it back, and they have a machine they plug it into, and he agreed with me. He took another new one, plugged it in, tested it, and I gained almost a volt....

Also, what voltmeter are you using? The one in the console? Or a handheld? The one in the console is subject to all the drops in the wiring.... and these drops increase with load.

[wilchek]

ditch the bosch regulator it is crap IMHO. get a solid state, sealed unit from NAPA or elsewhere. They are rock solid and work at 13.8 volts as they should.

[Dr. Roger]

when you observe 12.3 V at idle and then you put a load on the system like high beams and bring the engine up to 1500 RPM's you should observe that the battery never drops below 12.3V. you should see a voltage fluxuation up and down and up ...

the voltage regulator senses a particular voltage and kicks on (a low voltage state) until it senses another voltage ( a fully charged state) then stops charging...

voltage regulators vary their trigger voltages and some are even adjustable.

12.3V could be you voltage regulators low voltage trigger point. does it ever get so drained that you cant start your car? i ask because i bet it never goes below 12.3V as this is your regulators low voltage trigger point.

[rjames]

when you observe 12.3 V at idle and then you put a load on the system like high beams and bring the engine up to 1500 RPM's you should observe that the battery never drops below 12.3V. you should see a voltage fluxuation up and down and up ...

the voltage regulator senses a particular voltage and kicks on (a low voltage state) until it senses another voltage ( a fully charged state) then stops charging...

voltage regulators vary their trigger voltages and some are even adjustable.

12.3V could be you voltage regulators low voltage trigger point. does it ever get so drained that you cant start your car? i ask because i bet it never goes below 12.3V as this is your regulators low voltage trigger point.

Dr. Roger- thanks for this posting. I didn't realize that it the system could fluxuate that much at idle depending on the voltage trigger point. It hasn't ever gotten so low that I couldn't start the car- maybe it's been charging ok the whole time, and I've just happen to take voltage measurements when it was near the VRs trigger point?

(And to those who've asked if I'm using the volt meter in the car- no, read back from the beginning, please)

I will look to see what the lowest voltage reading is. The reason I started checking this in the first place is because I noticed that my blinkers weren't as bright sometimes- which coincided with when the voltage read 12.5 (or thereabouts) at idle. When I was getting a 13.3ish at idle my signals (and headlights) were brighter. Making me think that things weren't charging correctly. So while the VR might be allowing the voltage to drop a bit more than I'd like before kicking the alternator in, things still don't seem right if I'm getting dim signals while racing around at 3000 rpms. Mayby I should try a non-bosch VR as suggested by Wilchek?

As always, thanks for the suggestions!

[jk76.914]

a light bulb just went off in my head... no pun intended.

I had the identical symptom a long time ago. Kept driving for over a year, but I tried to run with the headlights off for a while to top off the battery before shutting down.

Anyway, I tried everything you've tried, and finally pulled the alternator. It had a bad diode. Apparently, when it was cold, it operated more-or-less OK, but as it heated up the reverse-bias current leakage increased and one of the phases of the alternator contributed less and less to the total power output of the alternator. Eventually, the diode failed completely. A failed bridge rectifer will decrease the total power output for the alternator by a third? I think. So under light load, it could regulate to a higher voltage, but it dropped with load.

When it failed entirely, it was only delivering about 10.5 volts under load, though still almost 13 volts with everything turned off.

So that's one more input to think about....

In my own car, I now have the voltmeter connected directly to the + battery terminal, through a relay that drops out when the ignition is off, so I get a pretty good indication of actual charging voltage. Between high load- headlights, fogs, and front blower- and minimum load, I only see about a 0.3 volt difference, with the lowest being about 13.3.

I'm also using a NAPA solid state regulator...

[rjames]

I had the identical symptom a long time ago. Kept driving for over a year, but I tried to run with the headlights off for a while to top off the battery before shutting down.

Anyway, I tried everything you've tried, and finally pulled the alternator. It had a bad diode. Apparently, when it was cold, it operated more-or-less OK, but as it heated up the reverse-bias current leakage increased and one of the phases of the alternator contributed less and less to the total power output of the alternator. Eventually, the diode failed completely. A failed bridge rectifer will decrease the total power output for the alternator by a third? I think. So under light load, it could regulate to a higher voltage, but it dropped with load.

When it failed entirely, it was only delivering about 10.5 volts under load, though still almost 13 volts with everything turned off.

So that's one more input to think about....

In my own car, I now have the voltmeter connected directly to the + battery terminal, through a relay that drops out when the ignition is off, so I get a pretty good indication of actual charging voltage. Between high load- headlights, fogs, and front blower- and minimum load, I only see about a 0.3 volt difference, with the lowest being about 13.3.

I'm also using a NAPA solid state regulator...

Hmmm thanks for this posting jk76. I will start with the sollid state VRegulator first as that's an easy one, but may just try pulling the alternator. Not being electrically enclined I'd be interested in knowing just how to wire the volt meter the same way you did so I could get accurate readings. If you're able, I'd love a step-by-step on how to do this- keeping in mind I'm a novice at electrical stuff.

[jk76.914]

Hi Robert,
Well, I finally got a chance to get some pictures for you on the relay installation on the volt meter. I'm afraid the pic doesn't show much in the way of step-by-step, but it's pretty straightforward.

1. You can basically pick ANY relay. Just make sure it has a 12v coil. Radio Shack has a couple that would work for short money. The contact rating is unimportant, as you'll only be switching a few milliamps of current at most- just enough to energize the coil in the meter itself. The one I got was actually too small to handle easily, and I would go for a bigger one it I had it to do again.

2. You take the two wires that originally connected to your voltmeter and connect them across the relay coil. If you can dig up a relay that has 1/4" tab connections, they would plug on directly! (RadioShack #275-226 ($6.49) would work, even though it is rediculous overkill in terms of contact rating, which is OK.) Now, when the ignition is turned on, it energizes the coil and closes the relay's contacts to complete the circuit to the volt meter. And, when the ignition is off, the circuit to the meter breaks, so you don't drain your battery. See schematic below...

3. The hardest part of the whole job is running a new wire from the battery + terminal to the console. I ran mine before my interior was in. I drilled an oversize hole in the firewall, and ran a wire through it, protected by a rubber grommet that I got at the hardware store. You MUST run a new wire and dedicate it to the voltmeter, or the meter readings will be affected by voltage drops caused by whatever other load you have hanging off the same wire. In the console, this wire connects to the switch inside the relay. All relays have pictures on them somewhere showing which terminals are for the coil, and which go to the switch.....

4. Easier, but still important, is running a new ground wire. I ran mine from the forward-most 6mm bolt that holds the console's heater lever to the top of the tunnel. First I removed the lever and polished up the surfaces on the tunnel and the bottom of the lever with a wire brush. This also will help make sure you don't have any grounding problems with the rear blower fan..... Then I reinstalled the lever and the rear bolt, and finally installed the front bolt- with its new ring lug sandwiched between two internal-tooth lockwashers under the bolt head- and that was it. The other end of the new ground connects to the ground terminal on the voltmeter.

5. Last step- connect a wire from the other switch connection on the relay to the + terminal of the voltmeter.

And that's it! You will now measure voltage directly at the battery terminal. When you turn on your direction signals, you won't see the needle dancing anymore! And, you can see the effect of load and RPM on charging and the battery. Much more useful than measuring voltage drop across all the wiring and connectors between the battery and the load.

I'm rebuilding my tach right now, but when I get it in and I'm driving again, I'll take some photos of the voltmeter under various load and RPM conditions, and add them in here as well.

Have you made any progress on yours?

regards,
Jim

[SGB]

Hmmmm.
I also experienced the "charging v. temperature" situation. I chenged many components en-route to changing the alternator for the third time. Similarly to the rebuilt voltage regulators, apparently a 49% failure rate is OK. I somehow found a NOS VR, and got a rebuilt alternator from a place in DC called Arlington Armeture. It isn't a Bosche, it was lighter, put out a little more juice, and most importantly, did NOT fail within 6 months. Replacing the thing isn't that hard, but it is dirty. Wear eye protection. The biggest challenge is weaving the heating duct stuff out and in.