To the electrically minded

dnpurdy wrote:

...
I assume the regulator getting hot had something to do with seeing this behavior. Before I buy I replacement, just wanting to see what the consensus opinion of the weird test results were.

I think you're right on. Let me know if you want to try a used one.

Post edited by: loudhvx, at: 2007/05/31 10:51

Alright, I broke down and bought a new regulator/rectifier from z1enterprises. When it came, I removed the connector, and tried my 2 headlamp test on this new regulator. I hooked up the stator wires into the regulator, positive/negative to 2 parallel 55W lamps, and left the sense wire unconnected.

Again, good numbers. At idle 12 something, with rapid voltage increase as rpms go up. So, I decided to connect the sense wire up to the positive side, as the diagram below shows. Here, again, I got the weird results that it worked fine until about 14V, then bulbs oscillation, then voltage dropped to 10V.

Lacking answers, I just decided to hook the reg/rect up to the bike hoping it was just a weird thing I wasn't getting right. I went for two long 40 mile rides, and it didn't seem to be charging. Wasn't draining badly, but wasn't chargin.

So I am left with two questions. Since this occurred on two different regulators, the only systems that could be wrong are the sense circuit and how the regulator is shunting to ground.

so:

1. I have replaced the stator to the regulator wiring, and the regulator to battery +/- lines. Only wire not replaced is sense wire. Is there a way to temporarily replace this for testing purpose? Just a wire from positive batt terminal, or should I put a resistor in line to limit current?

2. Anyone have ideas as to which the switching, sensing functions are working so erratically? It seems to be related for the "shunting to ground" -- seems like once it shunts to ground, it either shunts too much or shunts badly...

So close to finally fixing this issue.

You are connecting the sense wire to the lamps, but not the battery? If so, it won't work right. The sense line expects the voltage to be smoothed out by a battery.

The high peaks will trick the regulator into thinking the average-voltage is much higher than it actually is. You could put a big capacitor in parallel with the 2 lamps to simulate a battery. A 5000 uF, 50v cap should work. Or you could put in another battery (the bike still need the original one to run on) with the lamps.

But really, at this point I would just hook up the bike like normal and read your voltages on the battery. You know the alternator is putting out more than enough power. The only thing left to replace is the battery.

I had replaced the battery about 3 to 4 weeks ago. When the bike runs at idle, I get 12.5V, revved up to 5K I got about 13.

GRRR

I've verified generator output, replaced regulator and battery. What else is there????

Is there a way I can run a replacement sense wire so see if that's the issue?

Yes, but in the post above, you mention a diagram, but I don't see it.

Your post leads me to believe you have the output of the reg/rec connected to the lamps, and the sense wire connected to the positive side of the lamps. Is that correct? If so, it won't work like it should and the readings will be as you described... much lower than 14v after shunting begins.

No, what you described makes perfect sense. I've attached the diagram now for reference, but your analysis about the average shunted voltage is totally right.

I do know that since i never get about 13.7V potential, I never break the electrochemical potential needed to charge the battery.

Since I can generally verify the generator and regulator, any other explanations for lack of charging voltage??

I ran an ammeter on each subcircuit on the fuse box, and as I remember, they seemed in line with the expected loads. Would the voltage drop be explained by a leaky load?

Circuit attached

dnpurdy wrote:

Circuit attached

In this circuit, the voltage will stay very low because the peaks fool the reg into shunting when it shouldn't.

If you put a battery parallel with the lamps, it should work like normal.

Have you tried hooking everything up normally and check the voltage with the new reg/rec?

dnpurdy,

What are the results of your ammeter test readings?

It looks like you have repeatedly tested your stator and regulator and determined them to be good. You are also sure of the connection from regulator to battery and ground, right?

If you are still not charging the battery, the logical conclusion is that the charging system is always at full output and the rest of your bike is using the current and not letting it get to the battery. A system that is using a lot of current will pull current from the battery and the reg/rec, pulling the whole system voltage down as you observe.

Times like these, a DC Clamp Meter comes in real handy. I used one this week to condemn my Dyna III (see other post).

inline --

definitely going to take your advice and do current draw tests again across all circuits to insure there isn't any atypical loads. I can figure out the expected current load on each circuit by roughly adding up the load wattages and dividing by 14V to obtain expected current loads. This works for all circuits except the main coil circuit.

How can I tell what a reasonable current draw is across the iginition coils? I vaguely remember seeing 1 amp per 1,000 rpms somewhere, but don't know if that applies to my coils.

Per manual and measurement, my coils are 2.8 ohms primary resistance coils. I know in the circuit diagram they are hooked up in parallel making an effective resistive load of 1.4 ohms. However, since they are coils, i know they are inductive loads and not totally resistive loads, so using 14/1.4 = 10 amps is not accurate expected load due to back emf and the fact that they aren't drawing constant current.

Anyone got this tidbit of data for me to verify my coil load?

Ignition typically uses less power as the RPMs increase. This is assuming a stock electronic ignition. At idle the dwell angle is about 100 to 120 degrees. The coil is about 2.5 ohms. At 13v, that's about 5 amps, but we have to correct for the duty cycle. The dwell angle is 120 degrees so the duty cycle is about 1/3. So the average current draw is about 5/3 which is about 1.7 amps. Times two coils, that gives about 3.3 amps.

As the RPMs increase, the coils don't reach their maximum current (because they take time to charge), so the average current drops somewhat. The Kawasaki system increases dwell slightly as the RPMs increase so exact numbers are hard to come up with theoretically.

But I think you are skipping ahead. You still need to verify the regulator operation. Problems with the sense line almost always causethe regulator to put out too much voltage. Not too little.

Your strange readings on your test circuit are because the regulator is sensing too much AC and is being tricked into shunting too often, and too early.

Have you hooked up the new regulator to the bike and restored everything back to normal and taken a voltage reading?

Inline, he hasn't really verified the regulator yet, other than replacing it. But I don't think we know a problem exists right now since he hasn't really tried the new regulator in the bike normally yet. With the headlight off, and the regulator disabled, but rectifier working, we know he gets 15v and possibly higher.

The next step is to restore the wiring and the bike and see if it actually is not charging. The test rig he setup won't actually test the regulator, so it needs to go into the bike with a charged battery.

Post edited by: loudhvx, at: 2007/06/04 13:33