1982 XJ 550 won't charge after it warms up

I have an 82 XJ550 seca. The voltage meter on the bike will indicate it charges at 14 volts when cold (when the engine is at 2000+rpm). As the bike warms up and becomes hot, it eventually will drop to 12v or lower when over 2000+rpm. The battery is new. Do you guys have thoughts of what the problem could be? Or a posting of the same problem.
Thank you!

 

How many miles on the bike?

Possibilities: Alternator brushes; wiring; the big connector where the regulator/rectifier plugs into the harness; fusebox; ground straps between the engine and frame.

Knowing more about the bike would help us give you better answers.

 

Sounds like a loose ground or connection somewhere.

It doesnt make sense because metal expands when it becomes warm.

 

As temperatures rise, there is more resistance in the wires. I agree that there is probably a shorted ground or other wire. As the bike warms up, it's harder for the power to get through. Check your alternator brushes first, though.

 

heat has been to known to make electronics stop working when the electronics get old.

i'm thinking the regulator/rectifier.

fuel pumps work when cold and stop when hot.
some electronic ignition modules also work when cold and stop when hot.

 

I guess it would depend if others are having the same thing happen at cold startup - is this a normal characteristic of your typle of bike ? - will let you know if mine does when finished in a year or so

 

i just noticed he said volt meter on the bike. LOL.

try an actual volt meter to see if it corresponds with the dash meter. it could be just the dash meter going funky and not the charging system itself.

 

I'll back Snowwy up on that one, I've little faith in the factory voltmeter. I've seen +2 volt differences between a real meter and the dash instrument.

 

-it's not a "normal characteristic" of the 550 Seca or any other XJ.

-the in-dash voltmeter is NOT the most accurate instrument in the world; I would not trust it for any serious troubleshooting, it's more of a go/no-go kind of thing; however in this case I believe it's at least providing a clue.

By all means use a "real" DVOM to troubleshoot the charging/regulating circuit; you'll need to anyway.

If it turns out to be only the meter itself, PM me; I have a few spares.