What ground do I check?

[JHZR2]

I have old European cars. The car of interest in this post is a 91 BMW 318i, but it also relates similarly to my 82 MB 300CD.

Both cars have new, OE, 14.1V voltage regulators, and good alternators as far as I know.

I didn't drive the 318 last week, and yesterday I went and checked the battery - 11.9 at the cig lighter.

Pretty low, and it was about 43F out.

Recharged, drove around today, probably 20-25mi.

Stopped, checked voltage of battery, 12.3V.

Alternator puts out 13.8v at this temperature (currently 23F out), and when loaded with lights and AC/fan is at 13.4-13.5v.

Even when warmer out, I get about the same voltage, and I think the battery isn't charging fully. I'd like to get the voltage up a bit.

My MB is the same but worse - it's 55A alt will get down to 12.6v in warm weather with things on.

Now the kicker - the chassis ground in the BMW is in the trunk - clean and looks like new (NB is under the hood but I could say ditto). Current draw on both cars is minimal, like 35 mA or so. This happens even with brand new batteries in the cars so it isn't a battery. I think it is grounds and these old, low output Bosch alternators.

I've seen the mods to add a diode to the regulator. I'm not keen on that because I can't imagine that there isn't a root cause I can't fix, and a mod like that can be unreliable, and I take my cars very long distances.

I've never had a flat battery - but I want to get to the root of the problem. How do I best diagnose and attack the problem?

Thanks!

 

[chad8]

It might take a bit , but you can test the resistance in each ground wire if you can find them all. A good multi-meter is necessary.
The battery wires are a good place to start. Check both. You might have a wire that is internally corroded or one of the butt end splices is bad. I would also add a ground from the alternator body to the frame. That would assure a good ground.

 

[JHZR2]

Is there a way to check the wires without removing them, like from lug to lug attached? Or is that an invalid way?

 

[spackard]

At very low resistance you need a different type of multimeter probe, one that takes into account the resistance of the probe wires. http://en.wikipedia.org/wiki/Four-terminal_sensing

I have a lot of grounding problems on my BMW motorcycle. I've been able to find the problem spots, take bolts off and scrape oxide from aluminum a little. Any place the alternator attaches to the its mount points can be a place for oxides, and I guess in a car any block to frame metal-metal mounts. You have to chase all those things. It's not just a negative battery cable problem.

Also, it seems the alternator should be putting out more than 13.8@24F. Battery resistance increases as temperature decreases, and in my experience the regulator has been built to compensate for that.

You can probably get by without the special multimeter. I have. You scrape then look for change. Also, perhaps your alternator is weak. Test it for full output to make sure you're chasing the right problem.

 

[MrHorspwer]

Originally Posted By: chad8
It might take a bit , but you can test the resistance in each ground wire if you can find them all. A good multi-meter is necessary.
The battery wires are a good place to start. Check both. You might have a wire that is internally corroded or one of the butt end splices is bad. I would also add a ground from the alternator body to the frame. That would assure a good ground.


Testing ground resistance using a high impedence multimeter is a nearly pointless test when trying to located a marginal ground. This is especially true with a high current ground. There's simply not enough current from the multimeter to reliably locate anything but the most deteriorated ground and/or circuit.

If you want to find a marginal ground, espeically one that you suspect is causing issue during high load, do a voltage drop test while simulating the electrical load as much as possible.

I can bug a vehicle again and again with a bad ground that will test less than 1 ohm when doing a continuity check. You'll never catch it unless you test with a load applied. It drives students nuts.

 

[JHZR2]

I have a nice fluke meter. What is the recommended type that I get to diagnose?

I cannot imagine there are many high current grounds to check, the alternator and a few accessories are the only high current devices. Perhaps Id be better just removing all of them and cleaning well?

If I was to add an alternator ground, what is the best way to do this?

 

[SLCraig]

I would first start by removing all grounds, and cleaning them well. It may just be that simple. I've been down that road before, and a loose wire (or dirty) is all it takes.

 

[HerrStig]

On the old Saab 8 valves where the alternator bracket was about three feet long owners would slip a ring terminal with 14 or 12 guage wire attached under one of the alternator mounting bolts and run the other end right to the negative battery terminal bolt. Helped under high load conditions. High resistance in a charging curcuit can fool an internal, alternator mounted regulator into thinking the battery is fully charged when it is not. You have TWO grounds to worry about in every car circuit, one is the battery negative terminal to car body ground, and the other is the individual ground for each electrical device.

 

[HerrStig]

One suggestion.... Google "high output alternators" see what's out there for the "Dieter Vagons". Outfits like "Magamaniacs" sell rebuilt units at a fraction of OEM prices with much higher output. I bought one for my Toyota, 110 amps output for 120 bucks or so. Got 50K on the thing now . It is a Denso OEM core with thicker windings.

 

[JHZR2]

Update,

This morning i started up when it was cold soaked at about 18F outside. I was getting 14.3V at the socket, which isnt bad, IMO.

So perhaps in the BMW at least (the MB may be another story, because a 55A alternator is pretty small no matter what), the grounds and regulator are OK... Its just a matter of not charging the battery enough.

Which would still come back to the alternator and grounds, but in a different way, not because the grounds are bad and the voltage is too low.

or is it?

 

[The_Jeff]

My volvo says right in the owners manual that one should charge the battery occasionally as regular maintenance. It's calibrated to put out around 13.6 volts or so, kind of maddening. It's got lots of Bosch parts (and electrical engineering) making it tick.

Sounds like your temperature compensation is working; wonder if there's a hack to that varistor to make it "look" a little colder.

BTW, batteries with removable caps handle lower charging voltage better than maintenance free ones, something about the chemistry.

 

[mechtech2]

German cars use as little wire as possible.
Check and clean all grounds and power connections, and the resistance of the cables.
Add a ground if necessary, or larger cables.

But your voltages are not that bad a all.

 

[JHZR2]

Originally Posted By: mechtech2
But your voltages are not that bad a all.


Youre right, but my battery (has happened before on other batteries too) sits at like 12.3V at 40F. That's a bit too low, and an indicator of low SOC. It is the case even after a long highway drive.

 

[meep]

MUCH EASIER WAY TO CHECK GROUNDS.

engine off, key on.

put extension leads (scrap wire) on the multimeter. Set it to DC volts. Probe between battery post (-) and different points on the chassis. Try to run a healthy load, headlights and brights. Maybe the heater fan. Work out the draw of these devices. Say 8 amps for fan, 10 amps for lights, and 5 for engine electronics. That's 23 amps coming from battery (-).

now, V=IR ... volts = amps x resistance. volts displayed/23 amps = resistance. but you don't need to even worry about the math. You shouldn't see more than .5V lost over ground, preferably much less.

Now start the car and let it run for 10 minutes. Do the same test between ALT frame and battery (-) and ALT and chassis. anywhere you get a higher V reading, there is resistance in that path.

in the bimmer, your 11.9-12.3 V battery may be toast.

13.8 and 14.1V is very healthy for a vehicle, unless the battery temp is cold, like 40F or less-- when it needs a higher voltage to charge quickly.... BUT this only works if the voltage is lowered as the battery warms up... otherwise you overcharge and burn off fluid.

So--- you may not have a problem with the regulator, in other words.

The higher the voltage displayed, the more resistance that exists between those two points.

If you are seeing 12.3 V at battery after a long drive, you may have a shorted cell. 12.6 V = full charge. 13.2V is possible with full charge + surface charge.

Telltale will be the volt drop test between alt frame and batt (-). if you see anything approaching .5V, you need to clean grounds, replace grounds, or run supplemental ground.

 

[mgm13]

I would just buy some 8 or even 10 gauge wire and start adding additional grounds. Lay them out with string and make your wires up with ring terminal and if your fussy solder and shrink and get to it.
Inexpensive way to go at it buy a set of jumper cables, 12 foot for ten bucks at my local Walmart, cut clamps off and go from there.

 

[edwardh1]

add mo grounds.
some 80s vw rabbits had only one ground engine to chasiss and when this flexing wire broke the only remaining path was via the speedo cable which of course would melt. And melt real quick.
My 99 nissan maxima has lots of extra grounds, many engine to frame.

A whole nother topic but if the germans design it, they think its perfect.

 

[andyd]

how is the belt tension? Do the Bosch alts have a combination voltage regulator/ brush pack. Maybe the brushes are getting worn out? The 14.1 VDC out put is barely enough to keep the battery topped off. If the battery is low, charge it.

 

[JHZR2]

Originally Posted By: andyd
how is the belt tension? Do the Bosch alts have a combination voltage regulator/ brush pack. Maybe the brushes are getting worn out? The 14.1 VDC out put is barely enough to keep the battery topped off. If the battery is low, charge it.


Tension is ok though the pulley alignment is off. Belt still looks like new so seems like no biggie.

It is a combo reg/brush, but I replaced it as pm when I learned that the Bosch version was 14.0v and the hella was 14.1. Still where I'm at...

 

[The_Eric]

JHZR2, my memory could be a little fuzzy, but wasn't it you that changed your starter and hooked the wires up wrong causing your alternator to short to ground? I remember the poster saying that they could hear a whining sound (the alt. working extremely hard). This could well have done it in.

An alternator can still put out sufficient voltage, but not enough charging current. To test, apply a load (carbon pile load tester) to the battery with the engine running and measure the current coming from the alternator. It should be in the ball park of the applied load.

Here's a good example:

 

[JHZR2]

no, I never did anything of the sort.

An alt test should be in the works. Problem is on these older cars, by the time I pull the alt to do a test (I dont have a carbon pile tester so I need to take it off and to a shop that can do a test), I might as well buy a rebuild and put something newer on.

Of course if that does not fix it, then I wasted $, creating a chicken and egg.