To the electrically minded

How long did you charge the new battery before testing? Normally you have to put a new battery on a slow charge for about 8 hours before it's considered fully charged. If it wasn't fully charged, it would take a while for the battery to show full voltage on it in the bike. However, 12.9v is probably too low, so there must be something still going on.

It was a new, dry charged battery (acid came with spec grav of 1.28). I filled to top lines and put on my "smart charger" at 2.5 amps. When at 2.5 for about an hour and 1.2 for twenty minutes. I let it sit an hour, tried the cycle again, quit out full in 3 minutes.

Yeah, 8 hours is probably overkill, but I was talking about a trickle charge. The smart charge a 2 amps/1.2 should do it, like you said.

When the battery is getting charged on the smart charger,what is the voltage as measured on the battery?

If it's getting 2 amps on the bike, it should have about the same voltage as when getting 2 amps from the charger.

On the smart charger, after about the first 10 minutes, it also shows upwards of 14.3-14.7 volts. Definitely going to ride with voltmeter tonight a a good long one to the suburbs and back. I'll post results when back.

This bike is seriously testing my belief in the universal consistency of physical law.

Alright, got home and strapped the small voltmeter to the tank and ran two wires to the battery terminals, and took off. At idle, I'm at about 12.5V.

Immediately getting onto lake shore, I get 13.6 V at 5,000 rpm. Bike sounds smoother than I have ever heard it. As I go down lakeshore, I stay consistently at 5,000 rpm, but the voltage is slowly dropping from 13.6 to about 13.4 by the time I do the 6 mile trip to downtown.

Get on another highway and run 15 miles at 5,000 rpm again. Voltage slowly but steadily dropping the whole ride out to about 12.95 V by the time I turn around.

Highway trip back, another 15 miles at 5,000 rpm, doesn't break 12.75V

Go back north on lake shore drive, it was running about 3,500 to 4,000 rpm and the voltage dropped to about 12.4 running. At idle I'm at 12.1V and it can't break 12.5 even revving to nearly 6,500~7,000 rpm momentarily.

Run upstairs to grab amp-clamp. All circuits pulling normally. Break even now about 3,000 rpm due to lower voltage across battery. Bike now idling VERY roughly again.

Oddly, I let it run at 4,000 for about 2 minutes with the amp-clamp on. Starts at positive 1.89A, but ticks down a .01 A every 10 seconds or so. Definitely weird at steady rpm that should be charging, but it could also just be fluctuations within the accuracy of the amp-clamp.

Battery out on charger in apartment now, but I'm absolutely at a loss.

Man, that is strange.

Alright, here's what you have so far:

With a capacitor, the bike works fine. That means the alternator has enough power to power the entire bike, with a little to spare (because it hit the regulator's limiter). That means it needs no power from the battery while running. That means there should be no drain on the battery, and actually a little bit of current into the battery.

When the battery absorbs around 1 amp, it floats to 14 volts.

The stator is new, the rec/reg is new, and the battery is new.

That right?
Then yes. The laws of physics don't apply to your bike.

Actually, there may be a voltage drop somewhere in the wiring which only shows up when the battery is taking current. Remember, the cap charges almost instantly so it won't create a heat-related voltage drop in the wiring. The battery takes much longer to get up in voltage. During that time, the connections may heat up and cause a voltage drop. The only problem with this hypothesis is that something must be getting pretty hot somewhere between the reg/rec and the battery. I think you would have seen it by now.

I guess you should look for voltage drops in the wiring.
Check the voltage at the output of the rec/reg. If it matches that of the battery, check the voltage on the brown wire as close to the reg/rec as possible. If that matches close to the battery voltage, then you have to check voltages on the three yellow wires (unless you have replaced them). That would be check the AC voltage on the yellow wires, one at a time, from one end of a wire to the other. This is just to see if there is a significant drop along the wire.

It's almost acting as if one phase is missing. You could try a phase-to-phase AC test. That is put a 55watt headlamp on each pair of yellow wires to see if it lights up bright on each of the three phases. You would put the AC voltmeter in parallel with the lamp to keep an eye on the voltage. You shouldn't let it exceed 15v (AC RMS, which is what the meter is reading).

I guess there is also a chance that the magnets are just weak enough that they will power the entire bike, but just won't produce a single amp more to re-charge the battery. It's such a slim margin to fall into, but at this point all possibilities must be considered. (So far, I have never seen a KZ from the 1980's with weak magnets, or any KZ for that matter, but I've HEARD of some bikes from the 1970's losing magnetism.)

Post edited by: loudhvx, at: 2007/06/11 23:15

Alright, voltage gettin' eaten up in wiring I know is a possibility, but I feel like i don't know a good way to find it besides amp draw. Is there something I'm missing on that? How can I effectly test subcircuits?

Nothing seems to be getting crazy hot. Three yellow wires are long and new with the new stator. Yes, could be dropping a phase, but with a brand new stator, I doubt it. Will test nonetheless.

Last year the bike DID get layed on it's left side that the stator cover got scratched up. I did think it could be the rotor loosing magnetism. Got a different rotor off ebay, but it's about 8 or so mm larger in diameter than the one on there. Can I replace with larger one, or will air gap kill any possible gains in magnetism?

Super fresh charged battery with no fuse but main in should be near cutoff at idle, or certainly by 2 or 3k, yes?

Stator has OD of 93mm, rotor i got off ebay has ID of 105 mm

Yes with no lights, I find my 550's get to the 14.4v cutoff just barely above idle (1200 rpm or so).

If the wiring is losing voltage, that means you should measure at least 14v at the red wire on the reg/rec. Just trace the path to the battery and measure voltages (always using battery ground as the reference).

Which reminds me, you also need to verify that the reg/rec ground is the same as battery ground. When you measure the voltage from battery ground to reg/rec ground, there should be less than 1 volt, ideally.

If the rotor is not identical, I don't think I'd try it. Is it possible the rotor is slipping on the crankshaft? Does it wobble at all?
Maybe that little extra load from the battery can cause it to slip?

Post edited by: loudhvx, at: 2007/06/11 23:59

yes, it has separate lines, think lines running directly from reg +/- output directly to batt +/- posts.

I edited my previous post. Any chance the rotor is slipping after the crash?

I have to go for now.

Is possible. I'll check it out ...tomorrow.

I just plopped the freshly charged battery in the bike and ran on only main fuse. 15.20 V at about 1500 very nice.

So it has to be a subcircuit!!!

It seems to be especially problematic when it warms up. Anyone have suggestions on how to narrow down the suspicious subcircuit? I can't run without fuses on the street, but aircooled in the city, no chance to use a fan.

Ideas? So close, so very very close.