Regulator/Rectifier 101

[Dale_W]

Hi All,
I see some of you do some work on ATVs. This question may be applicable to any 12 volt charging system.
My 2008 Polaris 800 seems to not be charging. The battery light in the dash started flashing on an 18 month old battery, and by the time I got home from my ride, the battery did not have enough juice to crank the engine.

I put in a new battery, and the light went away, uh, for about an hour.

That new battery had 13.0 volts before the install.
Once hooked up, at idle, 13.0 volts.
revving the engine: 13.0 volts.

The manual shows 3 yellow wires coming off the stator. With engine off, resistance from y1 to y2, y1 to y3, and y2 to y3 should all be .19 ohm. They are. There should be an open circuit between each leg and ground, there is.

The next stator test is to measure AC volts between y1 to y2, y1 to y3, and y2 to y3. At 1300 rpm there should be 18 Vac for each of the 3 combinations. There is.
At 3000 rpm, 42 Vac, and at 5000 rpm 64 Vac. All good. So I am thinking the stator is OK.

Next I throw my ancient analog volt meter on the regulator output. Of course there are 3 yellow inputs from the stator. Then there are 2 black output and 2 red output. The diagram in the manual shows the 2 black wires going to ground. The two red wires going to the starter solenoid. The Battery is connected to the same terminal on the solenoid.

Here is my question, How can I tell If the regulator is shot? At the battery, remember, 13 volts even when revving engine. Not a good sign, but could be a pinched wire or something.

If I disconnect the voltage regulator, engine running, I get about 1 or 2 volts out of each red output wire. I think disconnected is not a valid test.

So I re-connect the regulator and stick the probes into the connector. Now each red wire give me 4 volts at idle, and 10 volts revving the engine. Is this a valid test telling me the voltage regulator is only give up 10 volts to charge the battery? I think I should be getting between 13 and 14.3 out of the regulator.

Thanks

Dale

[Arkie]

IMHO:
Just from a arm chair analysis. (and not seeing a wiring diagram)

Appears maybe your AC stator ouput is ok, but the DC output from the reg is low or none at the bat.

Check the DC voltage right at the regulator output terminal, if higher than the bat voltage while reving the engine you have a bad fuse or connection between the reg and solenoid terminal. If same as bat voltage the reg might be at fault. If a solid state reg lightly peck on the reg while reving the engine and also check the reg terminals for corrosion.

Your regulator should have a higher than bat voltage even with it's output charge wire disconnected from the solenoid term. (normally)

I prefer reading charging amps using a DC clamp on digital meter instead of charging volts, but your analog is ok for now.

Main thing is the analog should read higher than the static battery voltage when reving the engine so as to indicate some charging.

[bgsengine]

KE4 will probably have better info but it sounds about right to me - I suspect you have a bad regulator if you have correct inputs to the regulator and your battery is not showing at least 14.8v when engine is revved up I would strongly suspect the regulator , or possibly a bad regulator ground (check the ohms to ground from regulator black wires to battery negative? ) might also do a voltage drop test on ground side (measure voltage between backprobed regulator ground and battery - , if you see more than 0.2v with engine running then you have a bad ground somewhere between those points, I would think)

I believe the regulator would be "switched off" if battery voltage is above 12.8v (fully charged battery) so you shouldn't see voltage output when battery voltage is above that - hows it look if your battery is discharged a bit (below 12.8v) - thing I wonder is at your test meter if it is showing a battery at 13v then your meter may be off since a fully charged battery should go to 12.8 (engine off) with an accurate meter.

[Dale_W]

Thanks for the quick replies. I am at work now, but can do a little more testing tonight and let you know.

Funny you should mention wondering about the meter, bgsengine, as you have seen in the tools forum, I am looking for for a new one.


Dale

[KE4AVB]

Normal Charge voltage is 13.0 to 14.6 VDC.

Do the following tests.

1. Key off drain test. Remove negative battery lead and install amp meter. Max current drain should 0.01 or less negative amps. If there is larger drain start disconnecting things until you find the source of the drain.
2. Break even current test. Shift transmission into Park and start the engine. With the engine running at idle, observe meter readings. Increase engine RPM while observing ammeter and tachometer. Note RPM at which the battery starts to charge (ammeter indication is positive). With lights and other electrical loads off, the “break even” point should occur at approximately 1500 RPM or lower.
3. With the engine running, turn the lights on and engage parking brake lock to keep brake light on. Repeat test, observing ammeter and tachometer. With lights on, charging should occur at or below 2000 RPM.

Note for the voltage regulator to work both 12v lines (Red) need to powered by the battery. A bad connection to either will cause the voltage regulator not to work.

[38racing]

......

Note for the voltage regulator to work both 12v lines (Red) need to powered by the battery. A bad connection to either will cause the voltage regulator not to work.

Meaning regulator output cannot be measured unless connected to battery? Like some battery chargers that will not output unless voltage detected.

[KE4AVB]

Correct. A good example would be the Kawasaki 210662070 voltage regulator used on the FD750D-GS02. No output unless battery sense lines are connected to the battery.

Had one back in April that was bad (badly burned) but the new replacement would not work until I repaired the wiring at the ignition switch. F56 connector completely melted from years of being loose.

[Dale_W]

Thank you for the replies. These are the kinds of testing I was looking for. I hope to get ahold of a clamp style amp meter soon.
The key off draw is less than .01 of an amp.

I suspect when I get the amp meter, I will see that the engine/charging system never reaches break even.

Also,Black wires from regulator to ground: good connection with ohm meter. Good test here! How many times is a bad ground an issue?!? (answer: lots)

I disconnected the positive lead from the battery, and put the ohm meter from each of the red wires near the regulator, and the positive lead for the battery.
I did this because each of the two wires off the regulator have a fusible link near the starter. This proved the fusible links are good.

You are correct in this regulator needing to be plugged in. If I unplug it, I get 1 volt dc at idle, 4 volts when revving. Plugged in, I get 10 volts dc, idle or revving.

I have a little more testing to do. I will let you know the results. I am pretty confident the regulator is bad, but I wanted to learn a little more before simply ordering a new one.

Thanks again.

[KE4AVB]

Question...What is the voltage between the red leads and the battery negative terminal or chassis ground with the switch off? Should be the same as the battery voltage when it is measure across the battery. Shouldn't be less a 1/10 of volt drop due wire resistance. Sounds suspiciously like a bad terminal. Try jumping from the solenoid post to regulator with new wire. I have had a few solderless terminals to be bad under load. Even had one that the voltage was where it should be until the load was applied then the connection would drop out completely.

[Dale_W]

KE4AVB
Should I do that first test with the VR unplugged, and checking the volts from the red output and battery neg? Or on the other side of the plug? Or with it plugged together?

I will also run a temporary wire to the starter terminal from the VR output.
(There are two red wires coming out of the VR. Each has a fusible link, and show near zero ohms resistance. I wanted to be sure the fuses were not blown)
Could this part of the circuit be failing under load? Sure it could. And I NEVER would have thought about that.

I will let you know what I find after work on Thursday.

Thanks again.