I need to find what the electrical resistance is of the stock A-body (w/ ammeter) standard-duty wiring harness as measured between the alternator BAT stud and the starting relay stud.

In other words, how much voltage drop is normal for a given heavy current between those 2 points?

Our existing configuration melted somewhat and I'd like to avoid a repeat, but keep the ammeter fully functional.

You should measure no significant resistance at all (less than 1Ω) across the alternator B+ stud and the starter relay stud.

Sounds like you might have a very high output alternator. If you want to keep the ammeter working, but avoid a repeat of the meltdown, you'll need to remove the ammeter's internal shunt, make a new external shunt, and put in a new cable from Alt B+ directly to the starter's large terminal or the battery's positive terminal.

While the shunt-wiring will prevent any fire hazards associated with full-flow ammeters, it basically renders the ammeter useless. Electricity will always follow the path of least resistance, which, in this case, will be through the shunt wire instead of the gauge.

Just like '81 up trucks with a shunt setup, the ammeter will basically be a decoration.

$0.02

I know the resistance should be very small (<<1ohm), but I'd like to know what's the nominal value is for a system in good shape. I estimate it should be something like 20 milliohms.

I plan to install a nonlinear shunt, but to do that I need a good measure the actual resistance of a good system. The stock alternator put out something like 35A, the new alternator something like 120A, so it's clear something must be done.

Any simple linear shunt becomes very problematic; it is very difficult to maintain the relative ratio of very small resistances over time. Essentially, the 1975-up B-bodies did what you describe, but it it tended to not work well (IMHO). In most of them that I've seen, the ammeter no longer indicated anything.

I tried recalibrating my '75 by using a little steel to shunt some of the magnetic field in the ammeter to increasing its sensitivity, but the field then was so weak that the g-force in turns made the needle read inaccurately.

It would be handy if someone with a good stock system could put a known (~10-20A) load at the alt's BAT connection and measure the voltage drop, so I can compare my system with it.

Thanks, Kevin

The shunt ammeter in my 1987 truck has been working perfect for years. I had to clean and tighten all the connections in the circuit first.

Years ago I knew someone who had a 1982 Omni/Horizon and the ammeter did not work. I cleaned and tightened all the connections for it on the instrument cluster and then it worked great. Later I found a service bulletin for that car and the "fix" was to splice on some special pigtails to the amp gauge so all the the circuit on the printed circuit board on the back of the instrument cluster was no longer being used for the shunt amp gauge.

Now my 1976 Valiant with shunt ammeter.... (the only year A-bodies used a shunt ammeter)... I can't get that one to work at all. Everything tests fine, I've been through it several times and each time it will barely noticeabley work and after a couple months it does not work at all. Next time I'm gonna have to measure and study how the one in my truck works, then compare it to my Valiant to see if I can determine the exact cause of it not working.

Craig, will you keep us up to date on what you find? I could use that info for my '89 D100...

Ya sure, don't know when I'll get to it. I live in an apartment and have no workshop. I have lots of projects to do and don't know how to get them done.

Thanks for the input; I've seen how delicate the balence is to get the ammeter working in the later Mopars, which is one of the reasons I don't plan to go that route. Any *linear* shunt scheme will reduce the ammeter deflection unless the ammeter is replaced or modified.

The only factory ammeters I've taken apart were the full current (early) type and a '75 B-body shunted type. Both types are pretty simple inside and use the vector addition of magnetic field of the current and that of a perm. magnet to orient the needle. Neither offers an easy way to increase their sensitivity w/o decreasing the restoring magnetic field or adding turns to increase the deflection field.

My nolinear shunt scheme is to run a heavy cable direct from the starter relay to the alternator through a pair of high current, low voltage, diodes connected in antiparallel. At normal currents (say, <~25A), the resistance of the stock ammeter system (say, ~.020ohm) would give rise to a voltage drop of <~0.5V, and the diodes would essentially be nonconducting.

As the current rises to say ~60A, the voltage drop would have risen to ~1.2V w/o the diodes (~70W under the dash!), but the diodes will essentially look like shorts (after their internal voltage drop is exceeded). The ammeter would still work normally at normal currents, but would read highly nolinearly at currents well above stock (basically, the ammeter wouldn't peg and melt).

The heart of this, of course, is ordering and mounting just the right diodes, which is why I'd like to know the actual nominal resistance of the stock system. You don't find diodes like this at Radio Shack.

I will measure it myself on our car, but the car is currently difficult to access and it would be helpful to compare our car with a stock system in any case.

That sounds like a wicked idea.

Couple questions though; first, what is antiparallel? Versus series or parallel?

Second, why the huge alternator? Are you running a sound system or hydraulics or something?

_________________
I've been calling it as i see it for my entire life and that's not about to change. Take it or leave it.

Thanks. Antiparallel means they're in parallel pointed opposite ways - the pair will conduct in either direction after the voltage drop through the stock harness builds to ~0.7V or so in either direction...

Our club's goal is to build a very nice, but not necc. stock, car to raffle off for charity. We put in lots of goodies that we'd like in a car, but of course we don't always agree on what those might be. You can see the cars we've done at http://www.tidewatermoparclub.com .
I generally handle all the electronic and electrical issues.

The club replaced the stock 35A alternator with a very pretty ~120A? chrome one mainly for looks; there are no unusually large electrical loads in the car.

Unfortunately, the battery had been run nearly dead due to prolonged cranking, and when the car started and the new alternator put out its full current through the stock harness, something had to give.

I now need to do some repairs and upgrades. I have a choice of using a power resistor in series with the field winding to limit the output current at full load, or to come up with a more elegant solution.

I remember seeing people complain on this site that the ammeter design was weak. I've found overheating damage in the ammeter and associated wiring in many, but not all, of the old Mopars I've worked on that had ammeters.

I've measured 0.5-1.6V drops on my '71 B-body; it is very close to stock and all the wiring is in good shape as far as I can see, so I wouldn't be surprised those were typical for the stock A-body system.

It seemed to me that a big heavy shunt cable that bypasses all the underdash wiring and bulkhead connectors and that only turns on when you need it is the way to go.
I may have to run to the starter terminal rather than the starter relay, depending on the gauge of the battery to relay cable. but as I intend to put a fusible link in the shunt for safety, that may be more trouble than it's worth.

The spec sheets for many ~100A diodes are availabe online, so I'd like to order the ones best matched to this application. To know that, I need to know the resistance of a stock system in excellent shape.

I also need to consider the heat disappation of the diodes, the connections, etc. At 100A, even 0.7V is 70W. In practice, though, even a dead 80A-hr battery won't accept 100A for very long, and the current will drop rapidly.

I should have it fixed in a month or so; I'll let you know how well it works.