Weird ground issue

[seiplej]

I just fired up my freshly rebuilt motor. While the motor was out I cleaned a lot wiring up and made new battery cables. I made a new 2gauge negative off a new battery terminal to the engine ground point and regrounded both the brown harness wires on the back of the engine where they were. The starter barley cranks till I add another ground wire from battery down to below the bell housing and I vise grip it near the starter. I also made a new ground jumper to the body in the battery box with a 6gauge I think. Any ideas before I just run another ground to the the bell housing?

[Tom]

I just fired up my freshly rebuilt motor. While the motor was out I cleaned a lot wiring up and made new battery cables. I made a new 2gauge negative off a new battery terminal to the engine ground point and regrounded both the brown harness wires on the back of the engine where they were. The starter barley cranks till I add another ground wire from battery down to below the bell housing and I vise grip it near the starter. I also made a new ground jumper to the body in the battery box with a 6gauge I think. Any ideas before I just run another ground to the the bell housing?

Where did you bolt the main battery neg wire? Not sure what car you have, but on a 951 (and pretty sure on an n/a) it should be on the bellhousing with one of the brown wires that come out of the engine harness. If you have a 2 gauge wire from the battery to that location, and still need to clamp a supplemental ground to the bellhousing in order for the starter to work right, then there must be something wrong with the battery cable you made, or it's connection, no? If you have a two gauge wire bolted to the neg battery terminal on one end, and to the bellhousing on the other end, an extra ground should be totally redundant.

[seiplej]

I have a 2 gauge wire from the battery terminal to the ground point on the back of the block. Brand new wire with a crimped and soldered connection. The starter drags unless I supplement with another ground to the bottom of the bell housing. Maybe a bad starter? The positive to the starter is brand new also same 2 gauge wire. I wire -48vdc power plants and such for a living so this is not new to me. Its really strange behavior.

[Tom]

Back of the block? On my car, the big negative cable goes to the bell housing. Is that what you mean by the back of the block...? I'd agree the starter might be bad, except I can't imagine a 2 gauge wire not carrying enough current? But maybe? Or Maybe experiment a little with the wires -- see if the supplemental alone works without the battery cable?

[walfreyydo]

Make sure where the starter connects to the bellhousing is cleaned up/free of corrosion. Since the starter doesnt have its own ground wire thats how it grounds itself so needs a solid connection where it mounts to the bellhousing.

Also, as others have mentioned, make sure your negative battery terminal is grounded to the main post on the bellhousing, near where the speed/reference sensors are as it is stock. If its grounded somewhere else, that may be your problem.

[seiplej]

Heres a few pics. I may take the starter back out and clean the surfaces.

[dr bob]

On a similar car, the battery negative cable and the separate cable connecting the block to the chassis are both critical. The chassis to block cable in the 928 is an uninsulated copper braid with crimped lugs. The 'uninsulated' is a common feature among many other makes too. As others report, the copper braid is a known weak point as it ages and corrodes, and I'll share some personal experience chasing issues in ground cables that looked and tested great but couldn't keep up. The Good News is that those cables are relatively cheap. I hunted down some marine-rated cables with tinned copper strands in the braiding, less subject to corrosion I hope/plan.

[Tom]

Does the ground wire get hot if you crank the starter without an extra ground strap? It's safe to assume the bell housing is as conductive as any other, so the possibilities seem limited -- either the starter is all but shorted out inside, or there's too much resistance somewhere between the battery post and the starter casing. Might be worth getting out the multimeter and checking the ground at the starter to see if the resistance is obvious, and if so, backing up at every contact point toward the battery to see if you can isolate where it is.

Is the supplemental ground strap made of the same wire as the wire bolted to the flywheel?

[dr bob]

Too often, the first symptom folks discover with a missing or failing engine-to-chassis ground is some smoke from other wiring in the engine bay. If you have jumper cables handy, use them to connect engine to chassis to battery negative, and see if the symptoms go away. Do this before the smoke leaks out of those other wires if you can.

[icb]

Since you get better performance when you improvise a ground near the starter, I think you're on the right track to look at the starter to bellhousing interface. Make sure it's all clean
shiny metal where they meet and tighten to the torque spec, I don't remember exactly but it's something like 46Nm.
If the connections are clean and tight you won't need anything but the factory connections and wire gauges.
If you can get a multimeter across the various connections while having someone crank the starter, including starter case to bellhousing, you will find the problem area(s) when you find the biggest voltage drops.
This is assuming your fresh rebuild isn't too tight.