Charging system odd behaviour

[CookMySock]

Tried to start bike normally and battery appears flat. Terminal voltage 12V. Cranking volts 4V. Hrm it does this from time to time.. Why?

Put on the charger for a couple of hours and the bike starts really well. Terminal voltage 13V. Cranking voltage not checked.

Bike idling at 1500rpm, headlight flickers slightly, terminal voltage 14V. Raise engine rpm to 3500 and battery voltage drops under 13V and headlight dims. This makes sense - lots of open-road running and the battery goes flat. Idling around and it stays charged.

Test stator continuity check to ground, high ohms - pass. Test stator AC volts to ground, 0V AC - pass, stator is isolated from ground. Stator output voltage 40V++ across any combination of stator outputs.

So why does the battery terminal voltage drop when I rev it above idle?


Steve

[Hopeful Bastard]

Im not too sure.. But i would love to know why the ol' battery dies when you are doing long rides.. I thought the longer the ride was, The more charge the battery got.. Not the other way around..

[pete376403]

Extra load from the ignition - making a lot more sparks/minute.
Battery will almost always show 12v, even when it's flat as a wet turd. You need a load tester (which I appreciate not everyone has)
40 volts AC stator to stator seems a bit low - my GS1100 does about 70VAC at around 5K revs.
Good fault finding chart - http://www.thegsresources.com/garage/gs_statorfault.htm

[jonbuoy]

An ammeter between battery and rectifier/reg will give you a few more clues.

[CookMySock]

An ammeter between battery and rectifier/reg will give you a few more clues.

Yeah looks like I gotta pull some covers off tomorrow and I'll put the clamp meter over it.

I had previously checked the stator output, continuity, and isolation, and it was good. I'm thinking the regulator is poked. Maybe I should replace it before it fucks the stator.

Steve

[Tink]

Tried to start bike normally and battery appears flat.?


Steve

GO FIGURE... get off the bloody net ya nut...and ride it.. hehe

[neels]

Check continuity between the 3 outputs of the stator to each other, may be an open circuit winding? Or another suspect would be a faulty rec/reg unit.

A quick search on the net found a comet 650 manual, I'm going to leap to conclusions and assume the electrical checks will work, try these, it agree's with pete376403 that 70V is about normal.

[Paul in NZ]

Yup - def should have about 70V AC across each phase of the alternator output.

The two simplest most easily tested items are battery (unlikely) or rect / reg. However, on nearly any bike I'd chuck a few extra earth wires around the place as age / dirt etc play havoc with chassis mounted rect / regs 9eventually)...

[notme]

This doesn't really help the OP, but I have always had a battery monitor on my bikes - a single discretely mounted LED can save a lot of blood sweat and tears - when I had the CBR600RR the light alerted me to a charging system problem and the first thing I did is think that the battery monitor itself was on the fritz! over the next 2 months the batttery slowly got worse and proved me wrong and the battery monitor correct.... :-(

DB, what happens as the revs increase further? You say the charging volts drop as the revs increase, does this continue or does the voltage drop then increase again?

[CookMySock]

DB, what happens as the revs increase further? You say the charging volts drop as the revs increase, does this continue or does the voltage drop then increase again?

Thanks allun. Ok, from scratch ;

From 1400rpm idle, the headlight (ordinary incandescent 55/60W H4) flickers at what appears to be 30-40Hz (by eye), battery voltage constant 14V ish. Raising revs to 2000, headlight flicker goes away and light dims ever so slightly and battery terminal voltage drops a full volt to 13V ish. This is confirmed by the attached battery charger as its' output current meter raises from 2A to nearly 3A. Raising revs above 2000rpm does nothing perceptible to any of the meters or headlight behaviour.

I'm pretty sure it's reg/rect failure. Will pull covers off it as soon as I can and put the meter over the reg/rect. I've already checked the stator and it's fine. I'm worried now about burning the stator by running it will a stuffed reg/rect, as it will short the stator hard and force it to disapate its' entire output as heat in its' own windings. Contrary to popular belief, I doubt the stator windings are cooled by oil - there could not possibly be oil in there with the rotor running at engine rpm.


Steve

[howdamnhard]

I would bet on the Reg/Rect being the guilty party . I presume Battery is holding charge. As others have said 70V ac is about normal output from stator. If you have a manual it will have the specs.
Goodluck and let us know how you got on.

[pete376403]

Contrary to popular belief, I doubt the stator windings are cooled by oil - there could not possibly be oil in there with the rotor running at engine rpm.Steve

Suppose it depends what sort of engine you have but GS Suzukis are definitely oil cooled, and will cook their windings (due to particularly shitty voltage regulation system) if the oil level is too low.

[notme]

Hmmmm stator V is ~40V - is this with the RR connected? Sounds like 1/3 of the RR bridge rectifier is shorted.

[CookMySock]

I'm tempted to build a switch-mode reg/rect, but there isn't a 400 watt design available to pinch. I understand the technology, but not well enough for my design to assemble and work first time.

It would put much much less stress on the stator and consume less engine power as well. This idea of burning excess energy as heat in the stator is rather outdated.

Any electrical engineers want to offer a circuit?

Steve

[CookMySock]

Hmmmm stator V is ~40V - is this with the RR connected? Sounds like 1/3 of the RR bridge rectifier is shorted.

Thats with it open and the revs quite low.

It's been a while since I did that test. Will pull its covers off and redo all tests when I can get to it.


Steve