On the old farm tractors and American made vehicles there were two lugs on the generator body. One is the main power output from the armature and the second one is the field coil.
They are either set up in a field to case or a field to armature configuration. That is in the field to case set up if you connect the field and armature lugs together to get its full output (power to field lug). In the other design you tie the field lug to the case to get the full output (ground field lug).

One other problem is that back then they did not have diodes so they used a voltage sensing relay that would connect the armature to the battery if the armature voltage was high enough and then disconnect it when it dropped below a specific voltage That way the battery would not feed back to the generator if its output voltage was to low.
On the few solid state ones I built I just used a 20 amp dual diode type rectifier in a TO-247 case to replace the relay. Given that the old tractors I was working with only had about 15 -20 amps peak output just using the metal case of the old regulator for a heat sink was good enough.

I am not sure if current regulation is absolutely necessary. Many of the American made mechanical regulators didn't have it built into them. If the regulator is a square one it typically only has the isolation relay and the voltage regulator relay. Some had a current bias winding over the voltage control relay and used a sort of weak current bias to drop the output voltage if a high enough current was being drawn. The longer rectangular ones are typically the type with the adjustable current limiting built in.
Although many old charging systems didn't even have actual voltage regulators. Rather they just had the isolation relay and a specific value resistor inside the regulator box that just kept the generator at a proportional current output . The faster the generator turned the more amps it delivered. It was crude but still millions of them were used.

You are in fact right that they used a rather crude mechanical PWM of sorts. Thats probably why a basic comparator design works rather well. It uses the same basic effect as the mechanical system but is much faster and more accurate plus uses actual diodes and provids far better spike suppression than the buzzing contact regulation systems ever had.

Just some things to think about.

ok I think I start to get it, so is this a generator with a stator coil and a rotor coil, the third terminal is to magnetise one (in place of permanent magnets)

Originally Posted by tcmtech

On the old farm tractors and American made vehicles there were two lugs on the generator body. One is the main power output from the armature and the second one is the field coil.
They are either set up in a field to case or a field to armature configuration. That is in the field to case set up if you connect the field and armature lugs together to get its full output (power to field lug). In the other design you tie the field lug to the case to get the full output (ground field lug).

One other problem is that back then they did not have diodes so they used a voltage sensing relay that would connect the armature to the battery if the armature voltage was high enough and then disconnect it when it dropped below a specific voltage That way the battery would not feed back to the generator if its output voltage was to low.
On the few solid state ones I built I just used a 20 amp dual diode type rectifier in a TO-247 case to replace the relay. Given that the old tractors I was working with only had about 15 -20 amps peak output just using the metal case of the old regulator for a heat sink was good enough.

I am not sure if current regulation is absolutely necessary. Many of the American made mechanical regulators didn't have it built into them. If the regulator is a square one it typically only has the isolation relay and the voltage regulator relay. Some had a current bias winding over the voltage control relay and used a sort of weak current bias to drop the output voltage if a high enough current was being drawn. The longer rectangular ones are typically the type with the adjustable current limiting built in.
Although many old charging systems didn't even have actual voltage regulators. Rather they just had the isolation relay and a specific value resistor inside the regulator box that just kept the generator at a proportional current output . The faster the generator turned the more amps it delivered. It was crude but still millions of them were used.

You are in fact right that they used a rather crude mechanical PWM of sorts. Thats probably why a basic comparator design works rather well. It uses the same basic effect as the mechanical system but is much faster and more accurate plus uses actual diodes and provids far better spike suppression than the buzzing contact regulation systems ever had.

Just some things to think about.

thanks great help, hopefully I can get my hands on his car soon and see what working conditions I'm under

yes that's what I have said to him, I need to know about the generator he has

Originally Posted by tcmtech

I made a number of homemade refits of those old buzzer regulators using a simple op amp voltage comparator circuit and a large transistor for the field control many years ago. They fit inside the original regulator housing and always worked well. I still have my original one from about 20 years ago and its still in service!
It just basically a slightly modified version of a stock external regulator the older Ford alternator systems use.

I used them on our old farm tractors until I got around to electrical system upgrades and installed high output alternators in order to keep up with modern lighting loads that were added then.

how are you controlling the switching speed ? surely a comparator driving a power transistor will just ran as fast as the comparator and/or transistor can switch ?

That basically all it was. The actual field coil inductance will limit the actual speed to about the same realistic rate the buzzing contact system used. Or at least it seemed to have the same basic output characteristics only without the spiking and RF hash. As I said before it was rather crude but did prove it self effective at holding the system voltage more stable.

ah yes because the coil wil hold a magnetic field for a bit so provide a delay

Well I tried out the attached circuit and things have been a bit hectic. now firstly at some point I had trouble around Q1 as nothing i did would make it switch full on and hence deliver a clean signal to the mosfet gate. anyhow at some point after some proding we got it to work but only when the field coil was powered via the battery did it work proporly, when we put the field coil supply onto the armature output it just ran until we pulled it off at 16.5 v battery voltage.

to remove all doubt I want to make the circuit on a soldered board rather than breadboard in case of faulty connections as at 1 KHz there should not have been a problem.

for some reason Q1 was always at 10ish volts on the collector instead of 0.2 volts when on

Originally Posted by Thunderchild

yes that's what I have said to him, I need to know about the generator he has

All the dynamos we used to play with as kids (on old cars and motorbikes) were of the same type. A set of rotating coils connected via a commutator to give DC output (just like a normal DC motor). But instead of a permanent magnet, it was surrounded by an electromagnet, and this is switched ON and OFF to alter the charging current.

The residual magnetism in the magnet allowed such dynamos to start working without any external power, and the output polarity was governed by the polarity of the electro magnet.

It was simple to change the polarity (for anyone who remembers positive and negative earth radios), you just needed to touch a wire from the field coil to the live side of the battery, this would set the residual magnetism the correct way. You could do the exact same thing by manually pressing one of the relay contacts in the regulator, assuming it was an open one, and you could get to it.

We once put the battery on backwards by accident and it worked perfectly fine, except as you went faster the lights got dimmer

well nigel sounds like you had fun, good old days aye ?

I made R1 a bit high maybe, I have changed this for a 1 K resistor which will give (at average hfe) a current capability of 300 mA on the collector, I've soldered it all together so hopefully will work better tonight !

Originally Posted by Thunderchild

well nigel sounds like you had fun, good old day aye ?

Oh yes, great fun - trying to scrounge money togther to buy a gallon of petrol at five bob!

Makes you wonder how we ever survived, the crazy things we used to do - I'd have had a fit if my daughter had done the things we used to.

For example, as kids (probably six years old onwards) we used to climb this:

Cratcliffe Rocks on Flickr - Photo Sharing!

BTW, for those of you who can remember the film "The Princess Bride", much of it was filmed round there.

well nigel if you survived that your certainly fit for anything, trouble is these days we are so mamby pamby its a miracle evolution does not take its natural course and eliminate us through natural selection. If only much of the helath and safety stuff did not exist, if your that stupid that your going to get yourself killed because you just can't manage well what can be done about it ?

I've seen the princess bride great film, pity some of his later stuff was not as good.

Hi, Thunderchild

I Digged out from my archives a complete scheme + calculations +...+...+....+++ ...

that was published in ...1983 by Radio Plans magazine ... issue 430.

Of course, Everything written in French ...

It's a 12v regulator, but no PIC aboard ... ( ROFL )

Interested in ???

Alain