Yup, and the starter switch too ( not the handle bar switch but the solenoid) .
corrosion kills them good and makes this symptom.

 

Mine was doing the same thing. I pulled the odometer fuse permanently, i don't know what the deal is with that draw other than memory, so I don't have that right now. For some reason, it seems to be drawing more than it should for memory.
Anyway, i put a new battery in from Advance Auto(stranded on the road) and it wasn't able to charge it longer than an hour and it failed to operate properly(same symptoms). I bought the OEM Yuasa and charged it according to instructions, put it in and it fried the rebuilt RR i just put in a week before. I bought an new RR from honda shop put it in and no problem anymore. Screwy **** i don't understand(electrical eludes me), but proper initial charge is probably crucial for long term.

As far as gel vs OEM, orig lasted 9yrs. on battery tender. However, maybe letting battery go that long wastes the RR prematurely???

 

Its actually very simple, lol.

Listen people ... Battery chargers are voltage jackhammers. Even the cheap little trickle chargers.

Not all but 99% are not filtered at all. All they needed to add was a fat capacitor or even some chokes to take the spike off the wave. But whats hitting your battery ( and the battery likes too ) is momentary DC. But look at the wave on a scope and the cheap battery chargers pulse has a sharp spike on the leading edges.
OK for a tank like a battery, but not the crystaline little diodes (20 amps max) in your stock R/R. While charging with many types of battery chargers, your R/R rectifier diodes heat up in the wrong way. Different than if you were running the bike, this damage to the R/R was very well documented back in the 80's I know of.
Honda warned all dealer service departments about charging batteries on bike.
We were instructed to remove the positive terminals from batteries getting a charge.
This obviously isolates the battery from the rest.
In reality you should do this on a car as well if you use one of those big *** welder sized chargers.

The only safe charger I personally know of (Im not trying hard) is the Yuasa gell cell charger, the autosensing one. About 50 bucks, and I promise you can bake your batteries with this for years and not hurt your R/R diodes.

 

If your charger was putting out 14V DC unfiltered , then the peak voltage in the wave will be 14x1.414=19.8V. The peak inverse voltage for most diodes starts at about 50V so they can easily handle 20V peaks and will not heat up at all. The 20A rating you speak of, is the current rating in the FORWARD direction i.e stator --> R/R --> battery.

 

of course, but we arent hurting the diodes with reverse / inverse voltage.
Thats that part that will make you scratch your head.
Just disconnect the positive battery terminal when on-bike charging or run the risk.

Plus, the diodes in your stock honda R/R SUCK. Obviously engineered this way on purpose...

 

Because of the way a R/R works, you will send voltage forward through some of the diodes... IE 20V forward (or more depending on actual charge voltage)... They do NOT like that... (and no I'm not confusing V vs A here...) And yes, some of them will get the backwards spike and shrug it off...

And yes, Circuit_Burner is absolutely correct... The engineers at Honda used the cheapest possible components that didn't blow up on the test bench to build those R/R's... And then they sealed them in with minimal heat dissipation in a solid clump of resin... There is a reason they fail regularly...

It's not hard doing stuff the right way... Building a R/R with good components and a sink is easy... Sometimes it's even easier than spending the effort doing it the wrong way... So if it wasn't on purpose, at least it was done by an incompetent designer/engineer...

 

I would have it load tested to determine cold cranking output.