Testing a generator armature

[Will N]

Question for you electrical gurus out there. Long story, but my generator stopped charging. Not going to go into how it happened. There was what looks like melted wax coming out the bottom. So I pulled the armature out and put it on a growler. No indication of a short (the hacksaw blade has no vibration), nice continuity between the commutator segments, no continuity between the commutator and the shaft. So no shorts, right? But in testing the voltage output between the commutator segments, it's only reading a fraction of a volt. Shouldn't it be reading 6+ volts? What might be wrong with the armature if there are no shorts but it isn't putting out any voltage?

[Willie Krash]

How it happened is important.
Given what you say I would put the armature back into service. BTW I use 110V to test between the com and shaft. What are the F.C drawing? Are any of the segments darker or black? 180° apart?

[Tom Wesenberg]

Did you by chance have your meter set on DC rather than AC?

The growler and armature act like a transformer with the armature being the secondary. Transformers work on AC, so you must set the meter to AC when checking the output. I always like to check the output as a final step in growler testing. You must also hold the meter leads horizontal.

The common long generator armature on my Snap On growler puts out about 12.5 volts AC.
The powerhouse armature on the same check puts out about 5 volts.

BTW, generator armatures in the car also produce AC, but the commutator and brushes changes it to DC.
Generators are great. They don't need no stinkin' diodes to change AC to DC.

[Tom Wesenberg]

Here's a better picture of how I hold the leads when checking armature voltage on the growler.

[montanafordman]

Tom, that almost looks like the armature I sent you! If it is I hope it checks out. Good news is I was only into it for $5. Thanks for your informative post. I tried using a friends growler and from what I could tell it seemed ok and I THOUGHT I was doing the right things, but I have yet to assemble any generators myself and know for sure.

[Tom Wesenberg]

Sponsored Links (Register now to hide all advertisements)

Aaron, you have a good eye. That is your armature and it checked out fine. I'd like to find a few good armatures for only $5 each.

[montanafordman]

Originally Posted by Tom Wesenberg

Aaron, you have a good eye. That is your armature and it checked out fine. I'd like to find a few good armatures for only $5 each.

I was happy to get lucky on that one. Slightly offsets getting burned by my "professionally rebuilt generator" that I got off ebay.

[Tom Wesenberg]

Will, just wondering if you found out any more about your armature?

[Will N]

Thanks for the replies folks. How it happened was I put one of those new fangled voltage regulators that look like a cutout on the generator. I set the third brush as instructed so that it was not putting out more than 15 amps. Something must have gone wrong with the regulator, because the generator stopped charging, melted wax leaked out the bottom of it. It also shorted the armature in my horn and fried the turn signal flasher.

The commutator segments are darkened, but they are not black. The growler I have doesn't have an AC/DC switch. The voltmeter is graduated from 0 to 15 only(or something like that, I don't recall exactly as I don't have it handly at the moment), it certainly does not go to 110, so I am pretty sure it's a DC setup only.

The growler has a two prong probe to measure the voltage between adjacent segments. I haven't measured the voltage 180 apart yet. I'll try that with my multimeter.

[Tom Wesenberg]

I also have an antique growler with a volt meter and two probes as you describe. The meter is only used for certain old car armatures that can't be tested the common way on the growler.
Dyke's or the old Motor's Manual I think tells which cars need the meter to check the armature.

I would just ignore the meter and use your multimeter to measure, as I described above.

[tbirdtbird]

Tom, that was the most lucid description of how a growler works I have ever seen. I have used one for 50 yrs and never fully understood the electrics of it until now, and it makes perfect sense. Of course now you have me wondering why certain armatures can't be tested the usual way....I might dig out my Dykes and see if he explains that, because that doesn't make any sense

Will, that 'wax' had to come from somewhere....something def. overheated.

[Joe K]

I think that certain armatures are wound "differently." Something about the start and end points and how the coils overlap.

I bought a growler on ebay which I have yet to really use (a "Marquette") and it has the separate meter and two fingered probe. (aimed at adjacent commutator segments) Also a light bulb and two probes and the ability to use 120V to do a continuity check between segment and shaft.

But the meter was trashed. It seems to have been an AC voltmeter but for me it's not possible to tell offhand. (The two types of voltmeters, and also Ammeters IIRC, show almost identical d'arsonval type movements - only shunting or types of coils give a clue and I'm not current enough (heh) to know which is which at quick glance.)

But I wasn't worried since I have a separate (and identical) Harbor Freight meter like Tom's above. Good to see how it is applied in real life.

Heh. One of the websites online describing the use and theory of the growler says something to the effect of... "An old school diagnostic tool not often used today as components are usually diagnosed on overall functionality and replaced outright if found not working."

Yuh. Hitch a computer to it and see if it works.

Is this what Americans really do now? Let someone else determine through software electrical component functionality?

Pretty piss poor. (sorry!)

Joe K

[Tom Wesenberg]

Yes, what Joe said about some armatures are wound differently and may check bad using a saw blade on the growler, so you skip that and use the volt meter on adjacent segments.

Joe, I don't think H/F sells Sperry SP-152A multimeters. I bought this one at Menards when they closed them out. I really like this meter because it has a continuity buzzer, battery test, and 20 amp capability. This is great for checking anything on the Model A and most other cars, except for the high amp starter draw. At only $10 each I should have bought every one they had.

[Joe K]

Sorry on that meter. Not to impugne since the HF meters are um, functional - but that is about all.

But yours looked identical to my yellow HF meter. The newer red HF meters are "better" (ha, ha) and FREE if you get the coupon. I had three coupons so over a period of days (one customer coupon per day) I got three red meters.

Edit: looking back I see your meter is ANALOG - so put this one up to a quick look/information input error. To err is - what is that again?

My thought is to eventually get back to that Marquette armature tester, and use the HF or other meter and measure segment to segment voltage using two probes on adjacent segments. Once I have an idea of the voltage range developed, I'll get a replacement plastic panel meter from Ebay (the cutout in the Marquette case is a standard size) and use that in place of the trashed meter with the Marquette combined probe .

I just don't like having an open hole in the front of my tools.

Joe K

[Kurt in NJ]

The meter on 2 of my growlers are scaled in AC amperes, one has the 2 separate probes, the other has them set for bars next to each other, many armatures are wound that testing adjacent bars checks 1 winding

[Joe K]

Originally Posted by Kurt in NJ

The meter on 2 of my growlers are scaled in AC amperes, one has the 2 separate probes, the other has them set for bars next to each other, many armatures are wound that testing adjacent bars checks 1 winding

This is good to know. The meter from the Marquette tester was marked 0-10. No indication of volts, amps, dc or ac or anything. Do you know your scale markings on your meters?

Joe K

[Will N]

OK, so I measured the voltage 180 degrees across the commutator as Tom described, and I'm getting 8.8-9 volts AC all around the commutator. So I have no short in the armature windings, good voltage output, continuity through the field coils, but it wasn't putting out any charge. Why not? Should the field coils still have a magnetic charge when everything is disconnected from a voltage source or are they magnetized only when attached to the battery? I put a screwdriver against the field coils, and there was no magnetic attraction.

The actual sequence of events is this. I suspected the voltage regulator was going south because I was loosing power intermittently, just like when a condenser starts to go. It finally stalled altogether, and I was only getting a fraction of a volt at the points. I gave the voltage regulator a rap, and suddenly got 6 V at the points. I changed the voltage regulator back to the cutout, and the car ran with the generator charging at 6 amps after that. But then later, while running at at 45 mph, the car just stalled. The fuse had blown. Put a new fuse in, and the horn started smoking. Disconnected the horn, and at that point the generator stopped charging. So when the fuse blew with the generator spinning at high speed, there was no longer a battery in the circuit to regulate the generator voltage, and the voltage must have spiked. I imagine the horn armature (which is shorted out) must have been impacted by whatever the voltage regulator was doing while it was failing which caused it to short out later on.

[Joe K]

The field coils retain what is known as "residual magnetism." The field windings only add to this in a snowball effect between more field voltage yields more field magnetism yields more armature generated voltage yields more field voltage sort of thing.

The lack of any residual magnetism on a new generator is why typically you have to "flash" (or motor) your generator before it can be a generator - to set up this residual magnetism to "seed" the snowball.

Your description of the failure mode sort of agrees with my interpretation. But I think there might have been something going on at the first indication when the VR failed - like you had a shorted ignition cable or something. 0 volts at the points is sounding like something is pulling it down. The battery should have given you 6ish volts there.

Some minor vibration or random short while traveling at 45 mph might have taken it out - until then it was intermittent but enough to fry the VR?

I'll let Tom comment.

Joe K

[Tom Wesenberg]

I agree with what Joe just said.
High voltage or not should have no effect on the horn. If the horn started smoking, then that was a seperate problem, unless you kept pushing on the horn button while the generator was putting out very high voltage.

On a good working generator, I disconnected the field wire to see how much voltage the generator would put out with only the residual magnetism. At about 1500 generator RPM the output was only 1/2 volt. I did this just as a reference to help with generator problems.

Any time the generator is worked on, or the wires disconnected, it's good practice to flash the fields by jumping the 2 cutout terminals together for 1 second. This simply assures correct polarity so you don't try reverse charging the battery

[Joe K]

Any time the generator is worked on, or the wires disconnected, it's good practice to flash the fields by jumping the 2 cutout terminals together for 1 second. This simply assures correct polarity so you don't try reverse charging the battery

I like this thought. Enough magnetism to establish residual - but not so much to fry your armature on "locked rotor."

I will add to be sure in reassembly of a generator to have the field poles oriented correctly - you have a 50-50 chance of connecting them "opposing" - which will give you a generator that CAN'T be flashed.

Joe K