90 Amps load test at alternator post

[harvgwen]

And, I’m probably stating the obvious, when checking for voltage drops across each segment (as in your above diagram) it needs to be done while under the 90A load to be meaningful.
_________________
1986 Westfalia Vanagon, 2.2 Subaru
Okanagan Valley, British Columbia

[Sodo]

[harvgwen]

I agree, a load tester is much better than using the starter. In my case, I don’t have one.
The voltage drop test does need to be done under load. If a contact is quite corroded for example, the minuscule amount of current that the volt meter uses might not notice the corrosion at all, while 90A will have trouble passing through it. Maybe envision a thick ground cable that has a section so worn through that there are only a few wires still intact. A small dash light bulb may still work just fine and your volt meter may read almost no voltage drop, but if you try turning your starter through that worn cable, your voltage drop will be huge.
_________________
1986 Westfalia Vanagon, 2.2 Subaru
Okanagan Valley, British Columbia

[fxr]

Personally I'd use a separate voltmeter while using the '90A' load tester - there may well be some voltage drop across its crocodile clips which you need to discount. That voltage drop could be variable as well, even if the 90A current is fairly consistent.
_________________
Jim Crowther
1984 1.9l EJ22 Westy Wolfsburg Edition
Vespa GTS 300

[Sodo]

[harvgwen]

Oh wow! I would not have guessed that short nice looking ground cable would have caused that much loss. My question here is if your seat rail (where your red lead is connected) is a good ground. Or maybe we all need heavier battery ground cables.
_________________
1986 Westfalia Vanagon, 2.2 Subaru
Okanagan Valley, British Columbia

[AlfromNH]

What if you put the red lead on the shiny metal right next the ground lug?
_________________
‘88 2wd Auto Westy project

[Sodo]

[E1]

Some of us still do that, even if without the garage.

Seeing more clarity a comin’ in this thread, Tom… Well Done.

Are you thinking of making a larger ground?
_________________
If ever twice as rich, we’re gettin’ a double-wide
’84 “Westfailure”/2.1 Digijet/5.43 Ring & Pinion/Peloquin/D-rated BFG KO2s

AI has spoken to further illiteracy, to steal, to cheat, and to replace humans
The caveman’s first question to the headhunter: “Hey, you got any coffee?”

[crazyvwvanman]

Instead of assuming why not measure some more?
Put the tip of both meter probes into the strands of the negative cable, as close to each end of the cable without touching the terminal ends of the cable.
That should tell the drop due to that 7 inches of cable.
How much?
Then with a probe still in the end of the strands near the battery put the other probe at the battery post center.
How much?
Then the other end of the strands and the body metal right next the the bolt.
How much?

On and on to see if there is a bad crimped connection or bolted connection, or what.

Mark

[MarkWard]

The battery end in that picture looks very distorted and it appears to not be fully down on the terminal. In my years, I have seen many voltage drops through the battery clamps. The posts get oxidized and in this case disimilar metals. I have a battery brush tool for this purpose. You would not believe how many vehicles over the years I have fixed.

If I had a 40 year old vanagon with the starting battery under the front seat, I'd probably just run a long ground cable like the positve cable right to the engine starter. That's your biggest load. Everything else can with in reason work with some drop.

I was helping someone at the VW GTG with a slow crank on a new battery they purchased on the way to the GTG. I found that there was a voltage drop through the actual chassis. I guess over the years, the spot welded panels just weren't in the mood to pass current any more. So they could get home, they had a long set of jumper cables and I showed them how to temporarily connect the starting ground to the engine case. I suggested when they got home that they have their mechanic run a new cable as above.

[E1]

Great info, Mark!

So… instead of removing the transaxle ground and replacing that with a longer starter-to-frame ground, is what you’re saying?:
— Still remove transaxle-to-frame ground?
— *And* run a long, single (battery gauge-sized) ground cable directly from starter to battery negative?

This makes massive sense… eliminates two ground points... and still supports Sodo’s adept theory that using the tranny as a ground is a really bad idea.

Is that additional length at all detrimental in terms of increasing resistance from the longer overall path?

Or is that more than compensated for by using a larger-gauge cable as logic might indicate?
_________________
If ever twice as rich, we’re gettin’ a double-wide
’84 “Westfailure”/2.1 Digijet/5.43 Ring & Pinion/Peloquin/D-rated BFG KO2s

AI has spoken to further illiteracy, to steal, to cheat, and to replace humans
The caveman’s first question to the headhunter: “Hey, you got any coffee?”

[MarkWard]

The chassis still needs a ground for everything else, so the transmission ground would remain. You could also add an 8 gauge or so wire from the battery negative to the original ground location for the normal chassis loads. I'm not saying this is necessary, only sharing something I came across this year.

At a VW GTG, you can imagine how many experts looked at this problem. I usually hang back in those situations. After the crowd dispersed, I got involved. I carry a spare transmission ground that I added, not replaced. I also redid the battery ground. Still I was getting about a volt or so drop across the chassis. It was an 80 air cooled with new paint. My assumption was the spot welds just couldn't pass the current any more.

[E1]

Awesome.

Our battery ground is the “belt style,” which is probably okay, but I don’t know.

My factual knowledge is a bit scarce on this point, but I personally try to have my grounds touching sanded and brushed metal, with a spot of dielectric grease to delay future rust — as opposed to wondering what effect paint has on contacts.

THANKS!
_________________
If ever twice as rich, we’re gettin’ a double-wide
’84 “Westfailure”/2.1 Digijet/5.43 Ring & Pinion/Peloquin/D-rated BFG KO2s

AI has spoken to further illiteracy, to steal, to cheat, and to replace humans
The caveman’s first question to the headhunter: “Hey, you got any coffee?”

[Sodo]

Paint is an insulator.
With paint, the current goes

from the lug to the bolt
then
bolt to the (welded to chassis) thread.

That's a series connection (a daisy-chain) of two jumps, and the resistance is "additive" thus the vDrop is higher.
Thass' why I say eliminate all those bolt-to-case-to-bolt-to-case-to-bolt etc across the gearbox and bypass ALL of them with one copper cable.

You want the electricity to go from the lug direct to the chassis (one jump).
Soldering the lug to the chassis could be better, but somewhat difficult.
If you MINIMIZE the count of jumps, it can be "real GOOD".

======= ========

[Sodo]

[fxr]

[Sodo]

[Sodo]

[MarkWard]

I would have agreed about the chassis not being the resistance in the ground circuit, but I did the starter voltage drop test a few different ways. Same results.

In fact, I honestly did not think it would start. The only way it would start was if a jumper cable was connected from the battery negative post and the engine case. Easiest access. Once it started you could remove the jumper cable.

Anyhow, I don't have these issues with our van. The battery is where it should be, next to the starter and alternator. Battery is grounded right to the block and the alternator has an additional ground added.

Also, how is it 40 year old spot welds can't be the problem, but somehow grounding the engine using a transmission case bolt causes the current to flow through oily bearing surfaces easier? Asking for a friend.