So orginally I posted this on a motorcycle forum, with hopes to get some help, but it occured to me that I would probably have better luck on an eletronics forum.
I'm going to start off with the wiring diagram for my motorcycle, it's a yamaha xt350.



Okay, and here is some background to my problem.

So recently I bought an XT350 which pretty much starts and runs perfectly. With only one small major problem, its eletrical system, surprise!



Okay so, the headlights never worked, that's problem one. Problem two, the battery never charged. Oh, it didn't come with a battery.



Naturally, I bought a battery, plugged it in, with a bit of wiring, fixing the connectors, getting some rust off the switches(she was neglected for a good year). I managed to get the turn singnals, odometer lights, tail light and horn to run off the battery, as it was wired originally. Though, the battery never got recharged from the magneto, I jumped past the rectifier/regualtor because the headlight wasn't working either.



I opened up the flywheel, found the charging coil disconnected from one of the poles(It's very thin wire) and the lighting coil has a small cut across it shorting the connections together, as well as a few fried wires. I unwound the lighting coil, just until the cuts, soldered on a fresh piece and rewound, then fixed the charging coils connector, and any fried wires. I put it all back together, and it worked, somewhat. So the lighting coil should've given me 0.39 ohms, I get 0.8 ohms. I used the same wire, don't judge me, I tried without breaking it further.



Okay, I've explained myself, here's my problem. With the regulator/rectifier plugged in, I only get 3v to my lights, yeah, an entire 3volts, which isn't enough, this is AC. So even with the rectifier/reg plugged in, guess what, still not charging my brand new battery. So I figured that's broken.


I decided to unplug the rectificer/reg, and I did, once I did, the wire to the front headlight gives off a solid 30volts. This doesn't affect the engine running at all. So here's my question, instead of buying a new regulator replacing it, and having a solid 3v going to my lights(LEDS maybe?). Could I build a working rectifier(full cycle), get a decent regulator, to run off this 30Volts, convert, say, enough for 12volts to charge my dear battery and have the rest as AC for my headlight(Its a fairly constant 30v, not much flux with rpm).


Would I run into any problems? Could anyone think of an alternative system I could use?


Do take into consideration, I'm a poor student, I'm looking for the cheapest possible outcome here vs best results.


Also, on the magneto, 4 wires lead to the cdi, and 2 not, the voltage between those 2 measures a solid 12volts, thought I'd try and connect the headlight here, just for kicks, it glows beautifully, I didn't test the difference when disconnecting these two(12v) and the headlights wires 30v output, I probably should have, I know.

So if I could do this. Say I built a fullcycle rectifier to convert enough AC to DC for the battery system to charge, could I use a LM317T voltage regualtor to regulate the DC to charge my 12v 3ah battery, would I run into any problems?

 

Regulators in a flywheel alternator like in your case are called 'shunt' regulators. they put the excess voltage/current to ground. You say, "I'm a poor student", but in reality buying a new original equipment regulator is the least expensive way out for this case. Building your own regulator, will take a many tries and lots of burned out components and time. And you still might not get one to work. Not wanting to discourage you, just being practical.

 

Not wanting to discourage you, just being practical.

I was looking at this at 6 AM. I find the drawing impossible to read and I didn't know this was a, "dump the excess" system, so I couldn't help. (I thought it was a field coil excitement regulator.) TS (Reese) is talking about a linear regulator. The more I thought, the more holes I punched in that theory...until I gave up.

I'm glad there are people here way better than I am with motors and generators!

 

Regulators in a flywheel alternator like in your case are called 'shunt' regulators. they put the excess voltage/current to ground. You say, "I'm a poor student", but in reality buying a new original equipment regulator is the least expensive way out for this case. Building your own regulator, will take a many tries and lots of burned out components and time. And you still might not get one to work. Not wanting to discourage you, just being practical.

Time I have. Just to put this into context, in my country, buying a new rectifer/regulator is the same I'd pay for the components to build 15 rectifiers and regualtors, and that's not with buying in bulk. Importing is out of the question because of the incredible shipment fees, and postal takes 90 days. So, building it myself, is probably the best option.

I get to learn something new, and at the end of the day, walk away with a sense of accomplishment.
So lets put any discouragement, and being practical aside.

I want to build this, would it work as imagined? Where would I run into problems? As mentioned, I want to use a fullcycle diode setup for the recitifer, and a LM317T voltage regualtor. You mentioned a shunt regulator.
I have 30volts AC to split between my headlight and my charging system, should I keep the headlight running on AC? or should I just rectify it all to DC and then use that for the headlight and charging system to make things easier.

Also, if I were to convert it all to DC, should I run the regulator for the battery's charge in series or parallel to the headlight?

 

lets put any being practical aside.

I can work with that.

A 317 isn't a shunt regulator. That's part of what threw me off.
I need to know how the thing that makes electricity is wound. Is it two identical coils that are wired in parallel to increase current? (I already said I'm not good with motors and generators.)

I would start out with, "Leave everything AC except the shunt regulator."
Then I wonder how I'm going to make DC to charge the battery.

Anyway...there are a buncha ways to make a shunt regulator. TL431 makes adjustment easy, a zener with some transistors is more immune to voltage spikes, that kind of stuff. But first, how is the electricity maker wired? How much current can it make? How is it wired to send AC to the headlights? I need that to design the rectifiers, and I can't understand the connection blocks in the drawing.

 

Here are a couple of methods:
Which immediately brings the question: Can you buy parts, like fat transistors and resistors and IC's?

 

I share #12's views here. Without actually seeing those connection blocks it's impossible to know what connects where. I hate car 'schematics' . If we knew what the 6 wires from the magneto and the 4 wires of the rectifier/regulator did it would help. Do you have a link to either of those?

Its a fairly constant 30v, not much flux with rpm

What current can it provide at 30V? I suspect the volts will drop dramatically as you draw current. Typically the increasing impedance of the generator coils with rpm limits current available from the generator output.

 

I can't give you the exact current readings as of yet, waiting on a fuse replacement for my multimeter. As for how the coil is wired to power headlights and rectifier, this diagram should help.
As for parts, my local eletronics store, any components that I have access to, can be found here: http://www.communica.co.za

 

So orginally I posted this on a motorcycle forum, with hopes to get some help, but it occured to me that I would probably have better luck on an eletronics forum.
I'm going to start off with the wiring diagram for my motorcycle, it's a yamaha xt350.



Okay, and here is some background to my problem.

So recently I bought an XT350 which pretty much starts and runs perfectly. With only one small major problem, its eletrical system, surprise!



Okay so, the headlights never worked, that's problem one. Problem two, the battery never charged. Oh, it didn't come with a battery.



Naturally, I bought a battery, plugged it in, with a bit of wiring, fixing the connectors, getting some rust off the switches(she was neglected for a good year). I managed to get the turn singnals, odometer lights, tail light and horn to run off the battery, as it was wired originally. Though, the battery never got recharged from the magneto, I jumped past the rectifier/regualtor because the headlight wasn't working either.



I opened up the flywheel, found the charging coil disconnected from one of the poles(It's very thin wire) and the lighting coil has a small cut across it shorting the connections together, as well as a few fried wires. I unwound the lighting coil, just until the cuts, soldered on a fresh piece and rewound, then fixed the charging coils connector, and any fried wires. I put it all back together, and it worked, somewhat. So the lighting coil should've given me 0.39 ohms, I get 0.8 ohms. I used the same wire, don't judge me, I tried without breaking it further.



Okay, I've explained myself, here's my problem. With the regulator/rectifier plugged in, I only get 3v to my lights, yeah, an entire 3volts, which isn't enough, this is AC. So even with the rectifier/reg plugged in, guess what, still not charging my brand new battery. So I figured that's broken.


I decided to unplug the rectificer/reg, and I did, once I did, the wire to the front headlight gives off a solid 30volts. This doesn't affect the engine running at all. So here's my question, instead of buying a new regulator replacing it, and having a solid 3v going to my lights(LEDS maybe?). Could I build a working rectifier(full cycle), get a decent regulator, to run off this 30Volts, convert, say, enough for 12volts to charge my dear battery and have the rest as AC for my headlight(Its a fairly constant 30v, not much flux with rpm).


Would I run into any problems? Could anyone think of an alternative system I could use?


Do take into consideration, I'm a poor student, I'm looking for the cheapest possible outcome here vs best results.


Also, on the magneto, 4 wires lead to the cdi, and 2 not, the voltage between those 2 measures a solid 12volts, thought I'd try and connect the headlight here, just for kicks, it glows beautifully, I didn't test the difference when disconnecting these two(12v) and the headlights wires 30v output, I probably should have, I know.

So if I could do this. Say I built a fullcycle rectifier to convert enough AC to DC for the battery system to charge, could I use a LM317T voltage regualtor to regulate the DC to charge my 12v 3ah battery, would I run into any problems?

Basically - any Yamaha 4-stroke with model number starting with an "X"; sell it on and change address ASAP.

Mostly, Yamaha electrics are pretty good - but you need a good grounding in electrical engineering to make any sense of the schematics in workshop manuals.

 

As for how the coil is wired to power headlights and rectifier, this diagram should help.

That looks compatible with the drawings I made.
Right now, the design limit is the amount of current the magneto can produce. The regulator has to dump that current so I need to know how big to make the transistors.

 

http://www.communica.co.za

Holy smoke! 25.65 ZAR somethings for a 2N3055 that costs $2.34 here!
What's the conversion factor to American dollars?

http://www.communica.co.za/Catalog/Browse?cId=d6bb2a2817a4484baab3b39dc6225d22
http://www.mouser.com/_/?Keyword=2N3055

I think I'll investigate a mosfet for the regulator, if you can buy mosfets...

 

Regulators in a flywheel alternator like in your case are called 'shunt' regulators.

@tcmtech
Please look at the .png in post #8 and tell me about that partial winding tap in the magneto.
I can't understand why one winding is tapped and fed separately to the regulator.
It looks like W is going to have more voltage than Y.
It looks like the magneto is wired with some shorted turns!
Am I supposed to dump excess current from the Y/R wire to ground?
The way this is wound allows several methods.

4 instance, the voltage difference between W and Y will be proportional to the peak voltage, so that difference could be used to turn on a transistor, but it looks obvious that just dumping amps from Y/R should work.
I'm not sure. That's why I'm asking.

 

I think I'll investigate a mosfet for the regulator, if you can buy mosfets...

No. Gate voltage drops 20% at 100C. That would cause a major under-charge to the battery when the transistor gets hot.

 

That looks compatible with the drawings I made.
Right now, the design limit is the amount of current the magneto can produce. The regulator has to dump that current so I need to know how big to make the transistors.

Usually - motorcycle dump regulators use thyristors, they operate on the AC bit so commutation is no problem. That generator is probably somewhere around 130 - 150W - lots of 3055s and a big finned heatsink.

In most cases; the thyristors actually form part of the bridge rectifier - but if that's a flywheel generator, I'm not so certain.

Flywheel generators often include high voltage windings for no inverter CDI.

 

I can see how to do the thermodynamics. It's going to require a slab of aluminum, and some air flow won't hurt at all!

I don't know how to do this with an SCR. If I fire it on the rising voltage, it will dump the whole rest of the current. Got link?

 

The diagram is color coded which makes it easy to follow. Ignore the blocks. Wires are connected color to color. Headlights are ac. Must be an older model than mine. On new ones lights are always on with key.
The tap on the other coil may be part of the regulation circuit.
I believe the ac from lighting coil only goes to regulator to clip the ac peaks, saving the headlight from overvoltage.
Some of the early models just had a crowbar scr circuit to ground on the headlight circuit. When it failed lamps burned out.
It was a sealed unit, but may have been as simple as an scr and zener. Commutated each cycle. That may be one part of regulator. Other part is low current charging circuit.
That should be enough information to get headlights working, by following diagram.

 

I was looking at this at 6 AM. I find the drawing impossible to read and I didn't know this was a, "dump the excess" system, so I couldn't help. (I thought it was a field coil excitement regulator.) TS (Reese) is talking about a linear regulator. The more I thought, the more holes I punched in that theory...until I gave up.

I'm glad there are people here way better than I am with motors and generators!

The regular modern motorcycle alternator is usually a three phase wired stator, and the rotor is permanent magnets mounted to the crankshaft and spins around the stator. Don't know about this model though, it could be different.

 

I get to learn something new, and at the end of the day, walk away with a sense of accomplishment.
So lets put any discouragement, and being practical aside.

I really wasn't trying to discourage, just thought you could buy a replacement where you live. Many people come here thinking something like this can be done cheaper than buying a new OEM part, not always the case, good luck with your project. And please keep us informed on how you do.

 

It should be safe to run headlights at idle direct from lighting coil, while designing regulator. Orange/yellow it looks like.

 

Basically - any Yamaha 4-stroke with model number starting with an "X"; sell it on and change address ASAP.

Mostly, Yamaha electrics are pretty good - but you need a good grounding in electrical engineering to make any sense of the schematics in workshop manuals.

It's a great bike, it requires very little, if you use it mainly for offroad, you'd only really need the headlight.

What's the conversion factor to American dollars?

It's about R13.80 to $1, so it's about right.

The full service manual can be found HERE, page 21 and onwards is all the eletrical.

And yeah, the Y/R wire goes to the headlight and the regulator, plugged in to the regulator I have on the bike, the Y/R only has 3volts AC for the headlight, which isn't enough to light the H4 globe, and the regulator doesn't output anything, my guess is that it's fried. It's also a sealed unit.