'82 won't charge with yellow wire connected

[aldo5468]

I went all through the charging guide a couple times - battery, rotor, brushes, stator, reg/rect all good, all connections clean and tight, wiring in good shape. Finally discovered that it charged just fine with the yellow wire from the stator disconnected and stopped charging as soon as I re-connected it. Does this indicate a failure in the safety relay (Omron G2M-AS1)? Is there a way to test the safety relay? Or is the fix to just plug another one in and see what happens? Has anyone run into this before? Thanx in advance!

 

[Grinder]

Do your lights come on after the bike starts? The other function of the safety relay is to not allow the start button to activate the starter after the bike is started. If those functions work the relay is probably doing its job.

 

[pamcopete]

Well, just a guess, but if you have stock wiring, then the safety relay also turns on the headlight on that model, so I would guess that the charging system is not capable of sustaining the current for the headlight meaning that there is no surplus current to charge the battery. The other possibility is that your battery is not fully charged and you are getting some current to charge the battery, but with the headlight on, there is not enough surplus current to actually show an increase in battery voltage, so charge up your battery overnight on a trickle charger and try again in the morning.

 

[retiredgentleman]

As a test, unplug the Safety Relay, and measure the resistance from the yellow wire terminal to the black wire terminal. It should read about 21 to 26 ohms. If you find a very low ohms reading, then the relay coil has shorted out.

 

[aldo5468]

Thanx for speedy input. Safety relay yellow-black resistance is 24+ ohm; relay does not seem bad. I did start with my battery fully charged off the bike; it showed 13.2 when I put it back in. Voltage at brown-wire brush terminal is battery minus 0.3 VDC with engine running or off. "Slap" test is just a kiss, not a smart slap. One year-old Racetech rotor ohms 5.0-5.4 at multiple points around the inner and outer rings, infinite from either ring to ground. Stator: all 3 pairs show 0.5-0.6 ohm white-to-white, taken across pins of unplugged connector and infinite resistance to ground for each white wire. VAC across all 3 pairs of whites at 1200-1300 rpm idle is 8.4-8.5 max with headlight on, 9.9-10 with headlight removed. Analog voltmeter installed on bike reads same as my new VOM when testing across battery terminals - I get 14+ VDC with headlight removed; 12-12.5 with headlight on at 2000-3000 rpm. No obvious shorts in any visible wiring. All connectors involved are clean and tight. Fuse holders are all clean and tight - I did the "Radio Shack" fix several years ago when I first got the bike to replace all the OEM copper terminals when one of them broke. No abnormal electrical load - I have stock incandescent twin-filament bulbs in both front and rear signals for running lights, but my taillight is a single-bulb Superbright LED in a Lucas-style re-pop unit. Headlight is stock OEM unit, I believe 55 watts hi beam/40 low beam- "light-on" voltage given above was taken with low beam on. Rect/reg is new. All this seems to indicate that the magnetic field produced by the rotor is weak but not absent, but I can't identify a clear culprit. What am I missing? Any other test I can do to clarify? Thanx again for any input/suggestions.

 

[retiredgentleman]

Thanx for speedy input. Safety relay yellow-black resistance is 24+ ohm; relay does not seem bad. I did start with my battery fully charged off the bike; it showed 13.2 when I put it back in. Voltage at brown-wire brush terminal is battery minus 0.3 VDC with engine running or off. "Slap" test is just a kiss, not a smart slap. One year-old Racetech rotor ohms 5.0-5.4 at multiple points around the inner and outer rings, infinite from either ring to ground. Stator: all 3 pairs show 0.5-0.6 ohm white-to-white, taken across pins of unplugged connector and infinite resistance to ground for each white wire. VAC across all 3 pairs of whites at 1200-1300 rpm idle is 8.4-8.5 max with headlight on, 9.9-10 with headlight removed. Analog voltmeter installed on bike reads same as my new VOM when testing across battery terminals - I get 14+ VDC with headlight removed; 12-12.5 with headlight on at 2000-3000 rpm. No obvious shorts in any visible wiring. All connectors involved are clean and tight. Fuse holders are all clean and tight - I did the "Radio Shack" fix several years ago when I first got the bike to replace all the OEM copper terminals when one of them broke. No abnormal electrical load - I have stock incandescent twin-filament bulbs in both front and rear signals for running lights, but my taillight is a single-bulb Superbright LED in a Lucas-style re-pop unit. Headlight is stock OEM unit, I believe 55 watts hi beam/40 low beam- "light-on" voltage given above was taken with low beam on. Rect/reg is new. All this seems to indicate that the magnetic field produced by the rotor is weak but not absent, but I can't identify a clear culprit. What am I missing? Any other test I can do to clarify? Thanx again for any input/suggestions.

If you are using stock #1156 bulbs (single filament) in all 4 signal lights, they consume a lot of power. Each bulb draws about 2 amps, so that would be an extra 8 amp load if you have then on at all times as "running lights". That is way to much current for the stock charging system to handle.

You say you have twin filament bulbs for running lights. If you mean they are the #1157 bulbs, that are normally used for the tail/brake light, those bulbs draw about 0.6 amps as a tail light. That would be an extra 2.4 amps, which is still a large extra load.

Disconnect the power to the "running lights" and then see what battery voltage you get at 3500 rpm.

 

[aldo5468]

Thanx, RG - I'll try it out. All 4 signal bulbs are 1157s. Seems like net load increase with the taillight change to a single LED would offset quite a bit of the 2.4 amp gross. Until a week ago, I had been reading 13.8-13.9 on my VDO voltmeter with this combination for quite some time, so something went south recently.

 

[pamcopete]

aldo5468,

"something went south recently"

Well, I have had the experience of a headlight filament partially shorting before it finally burned out. The headlight seems to work normally except that it draws much more current which fits your yellow wire theory. A standard replacement automotive Halogen sealed beam is only about $15.

The other possibility, among many, is that both the high beam and low beam filaments are on at the same time.

 

[retiredgentleman]

Thanx, RG - I'll try it out. All 4 signal bulbs are 1157s. Seems like net load increase with the taillight change to a single LED would offset quite a bit of the 2.4 amp gross. Until a week ago, I had been reading 13.8-13.9 on my VDO voltmeter with this combination for quite some time, so something went south recently.

That's a good move to replace the tail lights with LEDs. You have removed 1.2 amps from the tail light, but added 2.4 amps with the running lights, for a net gain of 1.2 amps. The LED current is tiny in comparison.

At 1200 rpm, you should be getting about 12 volts AC from the stator leads, so your measurement of AC voltage is low.

Are the brushes long enough? Maybe you're not getting full rotor current. Another test you couild do is to measure the current that the rotor is drawing. Connect you VOM as an ammeter, in series with one of the brush wires. Should be around 2 amps at 1200 rpm, dropping to about 1 amp at 3000 to 3500 rpm.

 

[aldo5468]

I followed RG's suggestion to disconnect the running lights. First I unplugged the rear ones - no change; then disconnected the front ones and saw normal charging, 14+VDC above about 2500 rpm, with low beam on. Plugged the rears back in - still normal charging. Took out and inspected the pigtail that I had made to power the front lights, checked continuity, all good. Re-installed it, plugged the front lights back in, and saw normal charging with all four running lights on. Rode it about 50 miles - all good - 14 VDC and over anywhere north of 2000 rpm! Dunno what I did to "fix" it - must have been a wiring gremlin of some kind up front! One more experience that shows how much value there is in having an onboard voltmeter. Thanks again to all who responded - the pool of knowledge on this forum is just incredible.

 

[gggGary]

Aldo I hereby award you this months "Best charging system question" honors!

I have never seen a post with all the charging system troubleshooting info, voltages at various RPMs including the AC lead volts, parts inspected and replaced, that you provided. Two thumbs up.

I am sure glad you found the "problem". I'll guess there was a short that was just enough to cause a heavy draw, but not quite enough to blow a fuse.

 

[retiredgentleman]

Yes, there must have been a bare wire or connector touching ground somewhere up front. With your voltmeter on board, you're prepared to catch it, if it should appear again.

 

[aldo5468]

Well, I figured that providing all those data would do two things: most importantly, give you guys the best chance to offer some insight that escaped me, and secondly, show that I'd tried to do my homework first. In the end, it was some "abnormal" load, but like Gary said, not enough to blow a fuse. I can't even blame it on a PO, because I did the wiring to power the running lights. A learning experience - I still don't entirely understand how the headlight relay works in detail, but now I know where it is, and that the one mounted above it is the starter cutout relay.

 

[Grinder]

Doesn't make sense to me why unplugging the yellow wire gives normal charging. Wouldn't the coil sides and the load sides of the safety relay be independent of each other? I suppose unplugging the yellow just shut off the lights and really had nothing to do with the actual problem.

 

[pamcopete]

Grinder,

With the yellow wire unplugged, the headlight is not powered, so that leaves more amps available to power through the partial short. If he has the running lights connected to the headlight circuit, then without the yellow wire there is no power to the suspected area of the partial short. In both cases the alternator is free to produce 14.5 Volts.

 

[aldo5468]

I think you nailed it, Pete - one output of the headlight relay is power to the tachometer and speedometer lights via the blue wiring. That's the circuit that I tapped into to power my front running lights (and the circuit in which the problem occurred), so unplugging the yellow wire to the headlight relay turned off both the headlight and meter light circuits. The blue wiring that feeds the taillights and my rear running lights is a totally different circuit, so unplugging the latter had no effect on the "not charging" issue.