Weird alternator problem

[alctel]

So my alternator should be putting out 35-40 amps (I run it at 75% capacity, it being a 55 amp hitachi alternator modified to use external regulation) but it's only putting out 10 or so.

What is more puzzling is that it was working fine, then started only putting out 10 amps for a few days and then started working fine again for a while, now it's back down to 10.

It's not the regulator - it's an external regulator and I tried plugging the alt field wire into 12 volts and it still only gave out 10 amps. I've checked and even re-crimped the connections on the back of the alt.

I took it to an alternator shop to run on their machine and they said it was fine as I guess it decided to behave that day.

Any ideas? Is it worth be taking it to pieces to have a look?

[bigdogandy]

Since alternator current output is based on ththe battery's state of charge and acceptance rate, could it be that the batteries are fully charged so the regulator is calling for the alternator to ramp down output?

Also - does the external regulator do 3- stage charging, and if so do you know what stage it is in when putting out 10 amps?

Have you measured what voltage the 10 aMP current is being generated at?

[Dsanduril]

Sounds like high winding resistance. At full field the current for typical alternators is 6A. Can you measure the field current when 12V is directly connected to field? It should be pretty close to 6A. If it is then the problem is with the alternator/rectifier. If the field current is significantly less than 6A then the winding resistance is too high.

Does your alternator have a separate negative cable connection? Or is it grounded through case, engine block....?

Winding resistance should be 2.4 Ohm from the field connector to the ground connection. If you have a negative cable connected to the alternator then it should be measured there, otherwise to the cable connection to the engine block. It is not unusual for the negative connection, especially at the engine block, to get corroded/dirty and increase in resistance. This in turn increases the total resistance in the winding circuit, which in turn cuts down alternator output. Also the alternator hinge bolt and tensioning bolt should be checked, as these are part of the ground circuit when a block ground is used.

Since the alternator tests OK at the shop it seems likely to be a connection on the engine, and the main ground connection is usually the most suspect.

[kenbo]

Originally Posted by Dsanduril

Sounds like high winding resistance. At full field the current for typical alternators is 6A. Can you measure the field current when 12V is directly connected to field? It should be pretty close to 6A. If it is then the problem is with the alternator/rectifier. If the field current is significantly less than 6A then the winding resistance is too high.

Does your alternator have a separate negative cable connection? Or is it grounded through case, engine block....?

Winding resistance should be 2.4 Ohm from the field connector to the ground connection. If you have a negative cable connected to the alternator then it should be measured there, otherwise to the cable connection to the engine block. It is not unusual for the negative connection, especially at the engine block, to get corroded/dirty and increase in resistance. This in turn increases the total resistance in the winding circuit, which in turn cuts down alternator output. Also the alternator hinge bolt and tensioning bolt should be checked, as these are part of the ground circuit when a block ground is used.

Since the alternator tests OK at the shop it seems likely to be a connection on the engine, and the main ground connection is usually the most suspect.

Agree with your analysis however rather than general winding resistance my first thought was worn brushes/slip ring pitting.

I believe the Hitachi alts are case ground but checking the resistance from the case to battery negative is an easy check.

[Dsanduril]

Yeah, the problem with the case ground units is that the field current has to go from the field input, through the winding, out to the case, through the alternator mounting bolts, through the engine block to the negative cable connection, through the negative cable up to the negative battery post. Lots of places there for a bad connection to happen. And I suspect it is somewhere in that chain because the OP says the alternator tests fine on a bench.

Since the winding resistance is only 2.4 Ohm, even a 1 to 2 Ohm resistance in some bad connection can result in a significant drop in output. And in wet environments bad connections can come and go, making chasing them down truly difficult.

[kenbo]

Agreed but the solution if the case ground is the problem is installing a cable of equal size to the B+ cable. Putting a 1/4" or 5/16" negative post on the back of the Hitachi case is challenging.

[alctel]

There is actually a negative post off the back of the alt which I ran a separate ground lead from to the common ground

Thanks for all the suggestions so far, I'll measure the field current as suggested

[alctel]

I'm wondering if it's a loose connections INSIDE the alternator itself - is that possible?

[kenbo]

Sure it's possible. But before you split the case you can do a lot of diagnostic work.

Did you do the conversation from IR, A type to ER, B type or did someone else?

[alctel]

an alt shop did it a few months ago

[kenbo]

Here's the way if approach the issue:

Discharge the battery to at least 70%SOC.

Full field it again and check the current going through the field circuit, should be 5-6 amps.

Check the resistance of the entire field circuit from field wire in to the negative battery terminal, should be 2.5 to 3 ohms max.

If higher, break that circuit down into it's components and check the resistance of each component.

If all checks well, split the case and have at it.

One last thought, is your tach signal from the alternator stator or flywheel ring gear proximity sensor?

[Dsanduril]

Can easily be a loose connection inside the alternator. Especially for a unit converted from internal to external regulation, there are a couple of connections there that can be difficult to make correctly. Could also be brushes, or brush connections. Since the field goes through the rotor there are lots of options of places that things can go wrong. You may or may not be able to chase them by checking resistance on a de-energized unit, but with intermittent problems that can be a PITA. I'd start with resistance from the outside field connection to ground terminal, and also check the resistance from ground terminal back to battery post (just to make sure).

[alctel]

This is now fixed - pretty unbelievable.

Somehow a split washer from I guess the top mount had found it's way INSIDE the alternator and was resting on the stator pack (I think it's called). Removed the washer, fired up the alt and it works fine.

Wow. Always track down the washers you drop I guess.