# Dynamo Regulation Under Low Loads

Joined Jul 26, 2022
12
Greetings,

I have a 1960s vintage motorcycle with a motor from East Germany. It has a Bosch-type (field coil wired to armature) 6v/60w. Originally, this motor was designed for a motorcycle that had lights, battery, and a regulator. The motor was transplanted into a racing frame and successfully raced without lights or a battery. Unfortunately the wiring harness was lost and a mouse chewed up the dynamo wiring. I repaired the dynamo to spec but have no clue how it was wired to run.

The dynamo has sat a very long time so no residual magnitism. I created a wiring diagram of the unit, not showing a regulator. The dynamo as an integrated ignition circuit (points/cam/condenser) and the power contacts are D+, DF, and D-. The field coil is around 1.7 ohm and there's a 4.4 ohm "field resistor" between D+ and DF. The mechanical regulator operated by shorting DF to ground at a voltage designed to charge a battery. I've looked at just about every commercial dynamo regulator out there and none can control at this low of a field coil resistance, most require a battery.

My guess is that this bike was operated with a small battery pack to energize the field if it sat awhile. Past that, I see three options: 1) a small 6v regulator hidden someplace; 2) a big zener diode circuit to absorb the output of the dynamo at higher revs; or 3) they ditched the dynamo and just fired the ignition off a portable battery. I think #3 is unlikely because there is no place to mount anything on this bike and it raced in enduro races that stretched up to 500 miles over rugged terrain.

The machine is fully restored - regulating the dynamo is my last challenge. Firing the ignition requires only a modest amount of dynamo power. The ignition requirement is a 'nearly' fixed load. Depending on whether I use a standard coil (3ohm primary) or high output coil (1.8ohm) the current requirement would be, accounting for dwell, about 1.2A-1.5A or 2.7A-3.0A depending on the revs. I'm guessing that the ignition circuit can handle a pretty noisy signal. My simple electrical brain says 'keep strict control of the current through the field coil to prevent the dynamo from outputing any more power than needed'. If I can do that I'll avoid having to dump a lot of heat.

I understand that dynamo output depends upon revs and field current. There's very little performance data on this dynamo but I do know that voltage builds quickly once the motor fires. Though I'd prefer avoiding it, I think that a zener circuit is the simplest way to do this, but that might be totally wrong. If I have to use a zener option I would hope to control things sufficiently to minimize the heatsink site since there's little real estate to mount things on this machine.

I could really use your thoughts on this and how to design a regulator circuit that would work in this application. Thanks in advance for the help!

#### Attachments

• 249.5 KB Views: 11

Joined Jul 26, 2022
12

In mentioning a 50w or so power zener, as were used often in 60s, I don't think that zeners and dynamos work well together - another reason I've not been keen on that option. I don't see how a zener could keep the dynamo's current at a safe level, so either the zener or dynamo would fail. Also, I realize that my challenge is voltage, not current, regulation since the ignition coils won't demand more current, only what they need.

#### Marley

Joined Apr 4, 2016
489
To start the dynamo generating it either needs residual magnetism or a battery.
Yes, a battery could be temporarily connected using the quick disconnect to get it going.
Originally, the dynamo probably had a simple electromechanical regulator controlling the field current.
It might be possible to make something using a relay. Possibly a relay and a zener diode might work. Or better: a power transistor and a voltage sensing circuit.
Operation: If the dynamo output voltage gets over a certain value, the relay energises and shorts out the field winding. The field current continues to flow (due to back EMF) for a short time and dies away relatively slowly. The output voltage reduces, relay drops out, field current builds back up. In practice what happens is that the relay "buzzes" and the output is relatively stable.
The relay method is crude but has advantages in that it is robust.
An electronic solution is nicer but due to the un-supressed ignition circuit there will be lots of voltage spikes and reliability might be an issue!
You will need some residual magnetism in the field coil to get it started if there is no battery.

Joined Jul 26, 2022
12
Yep - from what I can glean the motor could sit a few weeks and still kick start or get going with a bump start; after that you need to battery to bring the field back to life. Normally this motor was used in a bike with lights and a battery with a pretty large mechanical regulator with three states: 1) no control to bootstrap the dynamo up to voltage, 2) connecting the Field Resistor to drop the voltage flowing to the field coil, and 3) short the field entirely and no output. So, relays did an an effective job and the battery would help smooth things out. In my case, the ignition coil is a more or less fixed load that will only pull the current it needs. I'm leaning towards the electronic solution with a voltage sensing/control circuit and a capacitor to smooth things out.

#### LowQCab

Joined Nov 6, 2012
2,628
It looks like the Circuit that you have drawn may work "as-is".
If so, the Coils don't really care how high the Voltage gets.

Have You tried to run it in this current configuration ?
.
.
.

Joined Jul 26, 2022
12
Nope. I know the dynamo is safe to run without the regulator hooked up. I have some of the engine manuals (in German) and they suggest testing dynamo output with the regulator but with a warning not to rev the motor or leave the regulator out of the loop for too long. However, the manual was written for a machine with lights and a battery and they were most likely concerned about the battery drawing the dynamo to produce too much current frying it or the battery. To me, the coil will deliver what the coil needs and there is no overcurrent potential. I don't think a coil is too fussy about how clean the DC power is. To be on the safe side I thought it best to put in some simple voltage control circuit and a capacitor to clean some of the spikes and noise. Kicking the motor with the dynamo out of the loop and powered off a DC power supply I get a good spark; nice and bright when I run a high output (1.8ohm) coil.

#### LowQCab

Joined Nov 6, 2012
2,628
Why does the Bike have 2- Ignition-Coils with a Selector-Switch ??????
-----------------------------------------------------------------------------------------------------------

Batteries and Coils do not "draw power".

A Lead-Acid-Battery can be destroyed by being left in a discharged-state for too long.
A Lead-Acid-Battery can also be destroyed by excessive Charging-Voltage.

An Ignition-Coil has a low enough Primary-DC-Resistance to create
more Load on the Dynamo than it can overcome.
The Ignition-Coil(s), most likely, are running at around ~75% of their rated Primary-Current,
so there is probably no need to be concerned about over-heating the Coil(s).

Proper Regulation of the Dynamo can get You higher, and more consistent,
Power from the Dynamo, ( Power = Volts X Amps ),
it can also keep the Voltage very consistent,
but it is questionable whether these qualities will provide any advantage to You,
considering the fact that You don't want to install a Box for a Battery and Regulator-Circuitry.

If You decide that a small-Box would be acceptable,
I can show You how to run a small Li-Ion-Battery, with proper Regulating-Circuitry,
and a stout Capacitive-Discharge Ignition-System from a Chinese-Scooter.
Parts cost will be around ~$250.oo, ( mostly for the special High-Performance-Battery ). This will also eliminate Points-Adjustment, and make starting much more reliable. ( It will also fire fouled-Spark-Plugs ). . . . Thread Starter #### traderyoda Joined Jul 26, 2022 12 The dual coil setup was meant to provide redundancy should a coil fail. This machine was designed for desert racing and would take quite a pounding. And you could easily get stranded in the middle of nowhere. Coil failures were not uncommon. So a twin coil/plug arrangement was fitted with a toggle switch mounted in front of the seat. Because this machine is the only one built I've been trying to keep it original (as I received it). I have a magneto conversion kit for this motor I picked up on eBay made in the Czech Republic but I've resisted using it thusfar. I take your point that improving the regulation of dynamo power might be gilding a lily - very good advice. I'm being careful because my dynamo experience is zero and replacing it, if I screw up, would be very hard. I have another race bike (same period) with a Lucas permanent magnet alternator with a regulator/rectifier plus large computer capacitor and no battery - it starts and runs beautifully. But dynamos are not alternators and it's new territory for me. I was confused by this statement: "An Ignition-Coil has a low enough Primary-DC-Resistance to create more Load on the Dynamo than it can overcome." The bike was fitted with twin 3ohm coils and they were pretty beat up. I have a pair of new 1.8 high output coils. How could either of these coils present a significant load to the dynamo? What am I missing? The case housing the dynamo has a fair amount of extra space so I thought I could squeeze in a regulator if it was small enough. I've heard from some people that running without a regulator is inviting the dynamo to burn up. Yet I can't see how that would happen with a fixed load that's well below the rating of the alternator. #### LowQCab Joined Nov 6, 2012 2,628 Using 2 CDI-Ignition-Modules from a Chinese-Scooter would allow You to run BOTH Plugs at the same time, they are far more efficient, and powerful, than the old Points and Automotive-Coils setup. It would also eliminate the change-over-Switch. Having a "Dynamo", ( DC-Generator ), with a Field-Coil is actually a bonus, the Field-Coil makes Voltage-Regulation relatively easy, and without a lot of unnecessary Heat-Dissipation. Heat is the enemy. Cramming Electronic-Parts under an Engine Side-Cover is just asking for the Evil-Blue-Smoke to leak out of the parts, it's WAY too Hot in there for good reliability. On the other hand, there's usually some room up-under the Gas-Tank that is out of site, not to mention the space taken-up by those huge Automotive-style Ignition-Coils, that must look really kludgy ............ A High-Current Li-Po Battery, including the desired Voltage-Regulator-Circuitry, will easily fit in the space that those Coils consume. "An Ignition-Coil has a low enough Primary-DC-Resistance to create more Load on the Dynamo than it can overcome." With the setup "as-drawn", it is very likely that the Ignition-Coil(s) present enough Load, ( with the Points "closed" ) that the actual Voltage at the Coil(s) is probably only around ~4-Volts. This is because the way the "Dynamo", ( DC Generator ), is currently set up, it probably can't supply enough Current to maintain ~6-Volts with the Points "closed". With the Points "open", the Voltage may go MUCH Higher, maybe ~20 to 30-Volts or more. With a 6-Volt system, and an Ignition-Coil DC-Resistance of 1.8-Ohms, the required Current would be 6 / 1.8 = 3.33-Amps. 3-Amps is probably the maximum amount of Current that the Dynamo can produce with a proper Voltage-Regulator connected. And You currently don't have a proper Voltage-Regulator, only a cheezy Resistor hacked in place to "make it work", ( kinda-sorta work anyway ). Currently, it is not possible to apply "Full-Field" Voltage to the Dynamo Field-Windings, therefore, it is probably operating with very little stress, and a "barely good enough" Output-Voltage. Unless You are running a "Shunt-Regulator" style Voltage-Regulator, the chances of "burning-up" the Dynamo are very slim, and running a Shunt-style-Regulator on anything other than a "Permanent-Magnet" Alternator would be really dumb. ( Most modern Motorcycles have a "Permanent-Magnet-Alternator" because it's much cheaper to manufacture ). You have a readily available, "adjustable", Field-Winding, just like a Car-Alternator, to not use it as it was designed would be silly. It's very likely that, with no other changes other than a properly designed Voltage-Regulator, your "Dynamo" will quite comfortably produce a steady ~14.5-Volts, especially if You change over to 2- CDI Scooter-Ignitions, which draw much less "Average-Current" than the Automotive-style-Coils. . . . . Thread Starter #### traderyoda Joined Jul 26, 2022 12 Terrific stuff! I hear you on trying to stuff a regulator inside the case - not a great idea. The fuel tank area is certainly filled by side-by-side coils, even though they are small 40mm diameter units. There's precious little performance data on these MZ dynamos but I asked an expert in Germany to provide some and he posted me a few files that I'll have to translate (always a headache - the dyname, "Lichtmaschine", translates as "alternator" which threw me for awhile). It's rated for 10A (60w). I think you're spot on that without a regulator the output will be quite limited but enough to fire the plugs. When I purchased this project it came with an extra set of coils. The ones fitted to the bike were 3ohm coils; the extra set were 1.8 ohm. Maybe the 3ohm coils were a better match to the dynamo's output. I can put one of each in the system and quickly tell which works best. The bike would be started and then run flat out. I'm guessing it would run rough at idle and improve at higher revs - and that would be fine for a racing machine. Focusing on that 4.4ohm Field Coil Resistor for a moment - it's a factory part mounted on the side of the dynamo. Originally (with the old mechanical regulator), it came into play between low voltage (current flowing directly to the field) and high voltage (field shorted); the 'suspended position' between 6.6v-7.2V for battery charging. When this resistor was in play, about 1A would be fed back into the field. Right now that resistor sits between D+ and DF. In operating without a regulator I was concerned that this path would allow a small amount of current to flow back into the field. Snce the dynamo's wiring was destroyed I can no longer tell how they wired it for operation without a regulator - this resistor might have been disconnected. #### LowQCab Joined Nov 6, 2012 2,628 The Resistor, ( if original ), energizes the Field-Windings just enough to "get by". This takes maybe 25% of the Current required for the "average" Load, off of the mechanical-contacts of the original mechanical-Regulator, making it last much longer, and reducing "Electrical-Noise", ( sparking between the contacts ). The Engine will only run rough at idle if a seriously radical Camshaft has been installed. ( which wouldn't be too surprising ) If You want to keep everything "as-is", and the Generator will not put out enough Current to run the lower Impedance Coils, You can simply experiment with different sized resistors ( lower Ohms ) until You get enough output to work acceptably. But, this means that You run the risk of over-heating the Coil(s) during extended high-rpm operation. Building an excellent Voltage-Regulator is actually fairly simple. Do You have any experience with building your own "Dead-Bug-style" Circuits ?, ( no Printed-Circuit-Board required, only ~6 to ~8 components soldered directly to each other ). You could also easily make a "Transistorized-Ignition" to take the Electrical-Load off of the Points and make the Ignition much more reliable, but I still like the idea of the Chinese-Scooter CDI-Ignition, ( which would require proper Voltage-Regulation, and "could" get somewhat complex because of the need to provide a "Flying-Magnet" and Pick-Up-Coil-Sensor, although it "might" also work with the existing Points ). All of this depends on what You intend to do with the Bike. Do You want it to be reliable ?, or original ?, You can't have both. . . . #### shortbus Joined Sep 30, 2009 9,755 Would the old trick they used back in the day when making a "chopper" from a Triumph or BSA work for you, the battery eliminator? Basically a capacitor that keep a charge to keep the generator field magnetized. Like this one, there are others - https://www.thebonnevilleshop.com/p...-54170009-m-54483156-99-0737-82-8319-03-1075/ I had built a chopper using one of these caps back in the day and they do work with a kickstart only bike. Or now they have come up with a totally electronic type that eliminates the battery and all of the other things except the generator to do the same thing. Again just one to illustrate there are other brand available if you look. https://www.thebonnevilleshop.com/p...-200w-battery-eliminator-chopper-pn-tbs-4030/ Thread Starter #### traderyoda Joined Jul 26, 2022 12 Thanks for sticking with me on this - it's helping me get focused - much appreciated. Last question first: my goal is 'as original as possible'. I won't ride this much (not terribly worried about damage the points) but I can't stand restorations where the engine is never started. The bike needed a good bit of TLC but no parts were missing, but the electrical side has been a mystery tour. The only thing electrical I got was the switch for the coils, two sets of coils, and a trashed dynamo. I'm feeling my way slowly forward to reconstruct how they made this bike work electrically. The engine is a two-stroke, by the way. That Field Resistor has long confused me. Dumb question #1: the dynamo's output is determined by field current and RPM - as the revs climb the voltage climbs and more current will flow to the field. The voltage drop across the Field Coil resistor does protect the field coil somewhat, but how do I know how much current I can push through the field coil without frying it? Dumb question #2: The way I've drawn the circuit the Field Resistor is in parallel with the ignition coil, so the coil sees the same voltage as the Field Resistor. As the bike revs won't the ignition coil see current flow increase linearly with dynamo voltage and eventually overload the ignition coil? These are the thoughts that made me think a simple regulator was a wise consideration. With having to worry about a battery, a regulator in this application would only have to monitor voltage at D+ and short the field coil if voltage climbed above 6v or so. Make sense? Yep - I've built a bunch of circuit stuff using PCBs and just soldering things together - always with the help of electrical engineering friends. I've been repeatedly told it very hard to control a dynamo without a battery, but no one has yet told me why. Cheers! #### LowQCab Joined Nov 6, 2012 2,628 "" but how do I know how much current I can push through the field coil without frying it? "" The Field-Winding is normally designed for the maximum Generator-Output, in this case, a nominal ~6-Volts. --------------------------------------------------------------------------------------------------------------- "" As the bike revs won't the ignition coil see current flow increase linearly with dynamo voltage and eventually overload the ignition coil ? "" The Coils are not connected across the Field-Resistor in the Schematic You provided. But, Yes, the Output-Voltage will increase with RPM. It won't be enough to be reliable at Cranking-Speeds, and it will be too much Voltage at high RPMs. ----------------------------------------------------------------------------------------------------------- Has it occurred to You that the guy who put this Bike together may not have had the slightest clue as to what he was doing ?, or that, he may not have had the Money or resources to do it right ?, or that, he just kept removing more and more parts until it quit running, and then put the last part removed back in place and called it perfect ?, ( fewer parts to fail ). ---------------------------------------------------------------------------------------------------------- "" this application would only have to monitor voltage at D+ and short the field coil if voltage climbed above 6v or so. Make sense? "" Shorting-out the Field-Coil is not a good plan ........ The way to do it is to replace the Resistor with 2 heavy FET Transistors wired "back-to-back" to create what is effectively an "electronically adjustable Resistor" that will automatically control the Power flowing to the Coil, based on the Generator Output-Voltage. The Transistors may get "warm", and keeping them cool will insure a longer service life, so it's not a good plan to mount them inside the Engine. ------------------------------------------------------------------------------------------------------------------ "" I've been repeatedly told it very hard to control a dynamo without a battery, but no one has yet told me why. "" That's because they've never had to deal with the "tricks" that are used by manufacturers to reduce production costs. Some of the designs are literal "blasphemy" to someone who has been working with complex, sensitive Electronic-Circuitry all their life. NO, NO, WRONG, WRONG, NO WAY, YOU CAN"T DO THAT !!!!! Well ......... Yes ....... You can, and sometimes You can get away with it for decades without any problems. But it breaks "the-Rules", it breaks "convention", and I was taught never to do that !!! ....... Why ?, 'cause it's cheaper to make, that's why .............. It's not complicated once You figure-out why certain things were done in such a strange manner. Generators and Alternators generally don't "require" a Battery to work, unless having a Battery connected to it was part of some design compromise, and therefore it won't "self-energize", or maybe the "Capacitance" of the Battery provides a function that allows the funky design to to work reasonably well. This is why You may hear of replacing a Battery with a large Capacitor, but if the "system" was better designed, ( usually meaning more expensive ), it wouldn't need a big Capacitor, or a Battery, to work. --------------------------------------------------------------------------------------------------------------- Is this Bike actually "Famous", ( documented, record-setting, recognized as the first ever, etc ), or, do You just think it's really cool ? . . . Thread Starter #### traderyoda Joined Jul 26, 2022 12 I have a BSA equipped with a similar setup (Tympanium) that runs for years without a hiccup. Uses a large computer capacitor to keep the lights happy and noise down. But that's a permanent magnet alternator that needs a rectifier/regulator and doesn't have to deal with excitation. This motor is an old-style soft iron dynamo. There are 6v dynamo regulators out there. The closest I came to a commercial solution was the DVR2 or the Boyer dynamo regulators. The DVR2 can control without a battery in the circuit but won't handle the low resistance of this field coil. Being UK firms their products are aimed at motors equipped with Lucas-type dynamos and this machine is based on Bosch designs. VAPE makes a small dynamo regulator for Bosch dynamos, but its circuit is powered off a battery - back to square one. Thread Starter #### traderyoda Joined Jul 26, 2022 12 The Field-Winding is normally designed for the maximum Generator-Output, in this case, a nominal ~6-Volts. Sort of implies staying away from running unregulated. The Coils are not connected across the Field-Resistor in the Schematic You provided. But, Yes, the Output-Voltage will increase with RPM. It won't be enough to be reliable at Cranking-Speeds, and it will be too much Voltage at high RPMs. Without a temp DC power source, or the bike having a goodly amount of residual magnetism starting would be tough. I guessed they just jumped the bike with a battery to get it started. From what I've been told by the MZ crowd, without a battery the bike will stall at 900 rpm, so once it's running you keep up the revs. Not a problem for a race bike. Has it occurred to You that the guy who put this Bike together may not have had the slightest clue as to what he was doing ?, or that, he may not have had the Money or resources to do it right ?, or that he just kept removing more and more parts until it quit running, and then put the last part removed back in place and called it perfect ?, ( fewer parts to fail ). Yep! All could be true. The motor was shipped to England, and existing frame type was modified for the motor and built up, then it was shipped off to the U.S. It was running because it was immediately tested. Things went south when the motorcycle was sold at auction. At that point cluelessness took hold. They couldn't get it running nor sort out why. In the process they tossed out whatever had been done. Shorting-out the Field-Coil is not a good plan ........ The way to do it is to replace the Resistor with 2 heavy FET Transistors wired "back-to-back" to create what is effectively an "electronically adjustable Resistor" that will automatically control the Power flowing to the Coil, based on the Generator Output-Voltage. The Transistors may get "warm", and keeping them cool will insure a longer service life, so it's not a good plan to mount them inside the Engine. Speaking of FETs... take a look at this MOSFET circuit: https://tinyurl.com/5955rkzh I took this from a discussion on StackExchange about regulating dynamos: https://electronics.stackexchange.c...regulator-to-replace-an-old-mec/593824#593824. I tweaked one of the circuits proposed that used two MOSFETs switched by transistors with a zener and discrete parts. All parts are small and cheap. That's because they've never had to deal with the "tricks" that are used by manufacturers to reduce production costs. Some of the designs are literal "blasphemy" to someone who has been working with complex, sensitive Electronic-Circuitry all their life. NO, NO, WRONG, WRONG, NO WAY, YOU CAN"T DO THAT !!!!! Well ......... Yes ....... You can, and sometimes You can get away with it for decades without any problems. But it breaks "the-Rules", it breaks "convention", and I was taught never to do that !!! ....... Why ?, 'cause it's cheaper to make, that's why .............. It's not complicated once You figure-out why certain things were done in such a strange manner. I've run into "can't" so many times I simply ignore it and go about doing it anyway. I've found that many highly complex systems, things we rely on and trust, contain lousy trades and risky dependencies. Even massive projects costing$Bs are riddled with some truly asinine engineering. Sometimes such systems fail spectacularly, othertimes a bad designs lays dormant and pops out with tragic consequences and finding it is a challenge... and often too expensive to fix so it gets a patch.

Generators and Alternators generally don't "require" a Battery to work, unless having a Battery connected to it was part of some design compromise, and therefore it won't "self-energize", or maybe the "Capacitance" of the Battery provides a function that allows the funky design to to work reasonably well. This is why You may hear of replacing a Battery with a large Capacitor, but if the "system" was better designed, ( usually meaning more expensive ), it wouldn't need a big Capacitor, or a Battery, to work.

Many of the dynamo regulators I've looked at can handle only very narrow performance windows and/or need a battery source to run the regulator. That doesn't inspire confidence!

Is this Bike actually "Famous", ( documented, record-setting, recognized as the first ever, etc ),
or, do You just think it's really cool ?

It's certainly really cool and it has legendary status as does the chap who owned it. When this was built, folks in the UK saw two-strokes taking over the racing circuit and they wanted their line to offer 2T power. They were aiming at the huge U.S. market where motocross was really taking off. They made multiple tries and failed. The idea of using the MZ motor came from the guy this machine was built for. He had raced in the famous ISDT races in Europe. MZ had dominated that competition for a decade using this motor. It was built like a tank and known for being reliable, powerful, and fast. He had a relationship with the MZ motorworks and talked the folks in England into trying the motor. In one famous desert race it won all classes against machines far more powerful. He himself was more than a motorcycle racer, he was a business man and a writer. He invented the terms "hot rod", "mini-bike", and "go-kart" writing for HotRod Magazine. He and I got along well because we were both airplane mechanics and loved to talk shop - he was a great friend to have. He owned many motorcycles and this was his favorite so I want to keep it as he owned it. Unfortunately, the MZ motor never worked out, but not because of any limitations of the motor. Russia's invasion of Czechoslovakia in 1968 put a pause on business dealings with Soviet Bloc countries, so getting motors from MZ became a pipe dream. It'll be great to hear this thing fire up again!

#### LowQCab

Joined Nov 6, 2012
2,628
I wouldn't want to find myself 20 miles out in the Desert with this current setup.

The reminiscing, story-telling, and even the ( maybe slightly embellished ) history,
is all great entertainment, ( I've got a few stories of my own ),
but,
I'm waiting to see if You are going to decide to do what it takes to make it Crank and Run reliably.

Do You have two big buddies who are willing to push you up and down your street until it cranks ?
.
.
.

Joined Jul 26, 2022
12
Do You have two big buddies who are willing to push you up and down your street until it cranks ?
Yes, actually. But I'm a real engineer who's too resourceful to ever have need of burdening them so.

#### shortbus

Joined Sep 30, 2009
9,755
The system used in your generator was at one time the got to way in 6V generators, 3(or third) brush generators. Used in motorcycles and cars of most makes. And in modern times they have regulator for them, solid state ones. http://www.splitdorfreg.com/
https://www.themagnetoguys.co.uk/three-brush-dynamos

What I haven't figured out is why you're thinking the generator is losing or has lost it magnetism? I've put old generators on cars that sat on a shelf for years that still had their magnetism still in them. but back in the day you still needed to "flash" a generator to make sure it had the correct polarity to charge the battery.