Need Help!- Open circuit in my generator rotor winding

[keebler303]

I bought an old craftsman 5kW generator real cheap off of craigslist knowing that the engine ran well but it didn't put out any juice. I traced the problem to an open circuit in the rotor winding. The resistance of the winding was above the range of my DMM when I first measured it. After about 30 minutes with the bearing puller and screw jacks I managed to remove the rotor from the engine crankshaft. After I had removed it, the resistance between the 2 slip rings had dropped to around 85 ohms! I must have beat on the right thing. Before I was gonna put it back together, I tried connecting it to a 12V power supply and measuring the current. The current was about 200mA for a 30 seconds or so and then dropped to nothing. I then measured the resistance again and found that it had climbed to 80 kohms now. I scraped bits of the varnish off the wire to check where the problem was. I have narrowed the problem down to somewhere in the last 100 turns or so on one side of the winding. The resistance from there to the "good end" of the coil is about 56 ohms. I measured as well where the wire goes from one winding to the other and it measures about 29 ohms there.

Possible options might include rewinding the rotor, isolating the bad section of coil as well as possible and then jumper around the problem, or just junk the thing and say the heck with it. Of those options I'd prefer to just jumper out the bad part of the winding. I am a little concerned that this would create an imbalance between the positive going and negative going sine wave voltages. Would an imbalance here be a big problem?

Couple pictures for reference
Thanks
Matt

[neilw20]

There is capacitance between the windings and the core/shaft/laminations.

When it is is the open circuit state you can use this characteristic to help narrow down where the open circuit is.

Measure the capacitance of each ring to the shaft.
The ratio of the capacitances will give some guide as to how close to one end the open circuit is.
If you don't have a capacitance meter, apply say 24vAC in series with an AC current meter, on the most sensitive range and see what you get.

You are are applying the voltage between each slip ring and the shaft.
Not ring to ring.

Using a signal generator at between 1 and 10KHz and an old sensitive analogue multimeter (if you can find one) on AC current should work fine.

Some low cost digital meters might give you misleading results.

[keebler303]

Not a bad idea at all, but I think I have already found approximately where it is just by rubbing off the varnish and checking the resistance at various places. I have measured 29, 31, 39, 51, 56 ohms from one ring to the place along the winding I have scratched off the varnish, all the while the other slip ring is still reading 70-80 kilo-ohms. I measured where the magnet wire goes from one winding to the other and the resistance is about 29 ohms there. That should mean that the ring to ring resistance should be near 58 ohms. Since I have found a spot at 56 ohms where the winding is good, I was thinking I may be able to just jumper from that known good spot to the ring connection. I am quite sure that would run a lightbulb with no problem, the question is how would it affect the waveform it put out and would it matter for common household appliances?

Matt

[Al_The_Man]

The total amplitude output is going to be slightly less depending on how much winding you have bypassed, which will affect the total Kw capacity somewhat.
At a 5Kw original the wire gauge is going to be fairly large, can you tell what gauge it is wound with?
Al.

[keebler303]

Al

As this is only the field winding, it is quite small, I'd guess in the 20-24 AWG range.

Matt

[Al_The_Man]

Of course I should have known it was wound stator output, in theory then, I would think the field current will be slightly higher for a given field voltage.
In a 3ph alternator this would also cause some imbalance between phases.
1ph, the amplitude will be affected.
It also probably is going to depend on how the field voltage regulator works on it?
In an alternator the output is dependent on field current, raise the current and the output voltage goes up, by bypassing some of the winding, you are going to increase the field for a given voltage.
At this stage I would think you do not have alot to lose, repair the winding the best you can and try it.
Al.

[neilw20]

Because the winding has DC on it, bypassing the fault and even causing a shorted turn will probably cause negligible problems.
Just make sure bypass wire is well secured against centrifugal force.

[Al_The_Man]

Although DC, the modern alternator regulator is PWM in nature so a shorted turn could have some effect on mean field current?
Al.

[keebler303]

This being a "Sears and Roebuck" era alternator (70's??), there is no PWM anything. To be honest I have no idea how the regulator even works. There is a single diode that goes from one stator wire to the chassis and then the normal bridge rectifier with a small capacitor to supply the field. There is also a small transformer looking deal included in the circuit. Not sure how it all plays together. In any case, there is very little to go wrong. To be honest I only bought it because I figured it was old and hardy and whatever problem it had was minor. Its actually more of a pain than I thought it would be but it should still work out ok. I'll zero in on the open turn and jumper as little of the winding as possible.

Thanks for all the suggestions
Matt

[Al_The_Man]

The very simplest early kinds monitored the voltage level output and just varied the field either in an analogue fashion or more often with just with a mechanical (or electrical) on/off switch (relay) on the field supply.
Al.

[neilw20]

Well, the sparky contact type once used was actually PWM. Just a little crude.
If it uses a POT then inductance doesn't matter a tinker.

[keebler303]

Well I think I got it fixed. Turned out the wire is aluminum. After fiddling unsuccessfully for about a half hour trying to solder to the aluminum, I gave up, cut the wire and spliced it with a crimp on connector. I would have preferred a more solid connection but it seems fine. I then covered the whole splice in fiber reinforced epoxy. I put around 40V on it for a half hour and it held up fine, current stable around 700mA. I ended up with around 56 or 57 ohms total resistance. I think the original was closer to 58. WIll tyr to get it back together bu the weekend and give it a shot.

Matt