charge SLA battery from magneto

Thanks for the suggestion. More info is in order. I'm using a full wave bridge rectifier. It's wired to both ends of the coil providing the AC, so it's full wave. Your right though, I'm measuring the AC w/a cheap multimeter, certainly not RMS. I guess the DC coming out of the rectifier is noisy, but I would still think that it would light a bulb. Thanks again.

In article , martin griffith wrote: [....]

Perhaps we should consider this:

C1 D2 -----!!---+--->!-------+----- To lights ! ! ---D1 --- C2 ^ --- ! ! GND GND

A voltage doubler will take power from magneto on both halves of the cycle. The capacitor C1 has to be huge if the frequency from the magneto is low. C2 is semi optional. Light bulbs don't mind weird pulse like power in place of DC.

This circuit has a disadvantage in that when you rev the engine up the lights may burn out. We also need to consider the battery and charging.

D2 nicely prevents current from flowing back into the magneto if the battery is providing power to the lights. This would let you do something like this:

From D2 ----------+---------------+---------- ! ! V --- --- ----- ^ ! !Char ! ! --! ger !------+ ! ! ! ----- Battery ! GND

If you go this way, C2 in the first circuit is no longer optional. You need it to prevent the battery from having to fill in the dips between the pulses.

This still doesn't address the need to regulate the voltage from the magneto to prevent it from going too high. I think a fairly dumb circuit could do a good enough job.

If D2 had a switching device in series with it or D2 was an SCR, a circuit could be made to disconnect D2 when the output voltage goes too high. I would need to know more about the real situation to suggest an actual circuit for this but the idea is like this:

SCR D3 -----+---->!------------ ! \\ ! ! ! \\ ! / ! \\ R2 ! / ! ! -/\\/\\--+ ! R1 /-/ Zener at maybe 15V ^ ! GND

--
--
kensmith@rahul.net   forging knowledge

Is it full bridge or half bridge rectifier? How is it plugged?

argh, nasty auto electronics

1) probably irrelevant, but what are you measuring the Vac with, a DVM, the cheaper ones are certainly not rms responding. A possible source of error. 2)AC mag, I assume one side is grounded to chassis 3)Rectifier, do you mean a diode? If so you are loosing half the waveform

if you isolate the lights from from the chassis, it may be possible to use a full wave rectifier to get the missing half cycle. Bit of a wiring nightmare. But you can float the battery from the + and - rectifier and power the lights from there.

martin

After the first death, there is no other. (Dylan Thomas)

w/a

the

bulb.

The Vs you give indicate somethings not right there. I suspect maybe one side of the light is grounded to chassis, and so is the dynamo, and when you connect bridge rec output to light youre shorting things out.

NT

check the rectifier, you may have damaged it

martin

After the first death, there is no other. (Dylan Thomas)

I think you're right, or at least it's some variance of this. The coil has two wires. Between them, they measure about 11VAC. Between each of them and ground I measure about 5.5V. Almost sounds like they're out of phase (like 220 in my house). But, and here's the weird part for my non AC mind, one wire lights a bulb just fine with it's 5.5V. The other will not. This wire will only light a bulb if the first one is also lighting a bulb. I measured the resistance combinations, and it's like this:

between the two wires, 7 Ohm

between wire which lights bulb on its own and ground, 0 Ohm

between wire which only lights a bulb when the other wire lights a bulb and ground 7 Ohm.

There are more properties I've found, but I'm hoping someone might have an idea of how these two wires are connected to the coil so maybe I can figure this out. Thanks again.

Please sit down with a piece of paper and pencil next to your scooter, and draw little boxes and circles labeled, "magneto", "motor", "frame", "Light bulb 1", "light bulb 2", and so on, with wires between them represented by lines, labeled, "red", "brown" "[whatever]" and so on.

Believe it or not, this will help immensely. Even knowing which component you've been referring to as "coil" so far will help - there are several components that _could_ be called "coil."

Thanks, Rich