What happens when shunt regulating a 3 phase stator?

Thread Starter

Rich:-)

Joined Oct 14, 2016
24
When looking at a motorcycle power system (commonly referred to as charging system), the basic setup is to have a permanent magnet generator composed of magnets in the flywheel revolving around a stator, often 3 phase. This then feeds to a Reg/ReC (regulator & rectifier in one package) before powering the bike.

Here is diagram of the setup.



The simplistic understanding of what's happening here is that the stator produces the max power is can at a given revs, the load from the bike will hold the voltage level down, at some point the stator will start to output more than is used and the voltage peaks will start to get too high. This is when the regulator steps in and starts to fire the SCR's which short each phase of the stator to the other phases.

It's easy to assume that it is just grounding each phase and thus dumping the unwanted power back into itself, turning it into heat.
Another thought is that it is actually dumping back through the other 2 phases and thus using them a dump resistors..
I'm not sure either of these are correct.

I came across a post on a bike forum where it was suggested that shorting the stator back to itself has a magnetic effect, and that doing so actually prevents the stator from generating - thus no excess heat / wasted power.

My electronics understanding is ok but I'm not so good on magnetism & generators.
What do you think about this please?

Rich.:)
 

tedstruk

Joined Oct 8, 2016
12
First, back in the days when motorcycles were run by magnetos that didn't need a battery or regulator or rectifier, when the engine turned over it induced current which was pumped through a coil that stored up current until it fired the plug and ran the motor. A good one took about 15 strong stomps to build up enough current in the coil.
Today, capacitors are used to store the current and some motorcycles don't have batteries anymore. Regulators were installed in vintage bikes because the engine generators made to much power that blew up the battery. A while after that, designers decided that a motorcycle that had lots of lights and accessories would require a better powerplant and installed an AC generator that required a rectifier to fire the coils(ac has to short a phase to build high voltage coil power.) The only other way to make an AC system charge a coil, is to shunt the generator directly into the coil. This burns up coils, and was proven to be to much of a hassle to do...to hard to fix when it broke down and caused alot of other problems when it failed. So, rectifiers take the AC (short phase) in one end, and run it through a set of diodes that are each.... part of a circuit cycle that grounds, increasng the phase as it runs around in there... Each time the diodes keep the ground from running directly back into the system, it gains a bit of phase and pretty soon... walla.. DC! So I guess what you want to know is whether the shunt is capable of handling the high voltage??? Well capacitors can handle the DC, but AC s generally controlled by specially made capacitors that handle a certain voltage. If you can stabilize the volts....install a fixed cap in the starting system that is the right size, and regulate the wiring volts in the system so it doesn't melt... Why not !!!!!?!
 

shortbus

Joined Sep 30, 2009
7,088
First, back in the days when motorcycles were run by magnetos that didn't need a battery or regulator or rectifier, when the engine turned over it induced current which was pumped through a coil that stored up current until it fired the plug and ran the motor. A good one took about 15 strong stomps to build up enough current in the coil.
Today, capacitors are used to store the current and some motorcycles don't have batteries anymore. Regulators were installed in vintage bikes because the engine generators made to much power that blew up the battery. A while after that, designers decided that a motorcycle that had lots of lights and accessories would require a better powerplant and installed an AC generator that required a rectifier to fire the coils(ac has to short a phase to build high voltage coil power.) The only other way to make an AC system charge a coil, is to shunt the generator directly into the coil. This burns up coils, and was proven to be to much of a hassle to do...to hard to fix when it broke down and caused alot of other problems when it failed. So, rectifiers take the AC (short phase) in one end, and run it through a set of diodes that are each.... part of a circuit cycle that grounds, increasng the phase as it runs around in there... Each time the diodes keep the ground from running directly back into the system, it gains a bit of phase and pretty soon... walla.. DC! So I guess what you want to know is whether the shunt is capable of handling the high voltage??? Well capacitors can handle the DC, but AC s generally controlled by specially made capacitors that handle a certain voltage. If you can stabilize the volts....install a fixed cap in the starting system that is the right size, and regulate the wiring volts in the system so it doesn't melt... Why not !!!!!?!
HUH?
 

crutschow

Joined Mar 14, 2008
23,536
It's quite simple, notwithstanding ts's explanation. :rolleyes:
The SCR's short-circuit the windings (through their respective diodes) when the voltage gets too high.
The current, under these conditions, is intentionally limited by the magnetic saturation of the iron.
The power dissipated than equals the square of this short circuit current times the winding resistance of the coils.
This dissipated power will heat up the coils some but it is much less than the maximum output power the generator can provide at it's normal 12v output.

This does also generate a fair amount of heat in the SCR's and diodes, so the regulator case is typically finned to help get rid of the heat, although they can still get very hot.
(I've had three different units fail on my motorcycle due apparently to their overheating.)

To minimize this, some of the new regulators use MOSFETs rather than SCR's to provide the short circuit, since MOSFETs can have a fraction of a volt drop when ON as compared to the nearly 2V of an SCR.
The MOSFETs can also do double duty as synchronous rectifiers, replacing three of the diodes, which further reduces dissipation.
Here's a short article I wrote on such a MOSFET design.
 
Last edited:

MrSoftware

Joined Oct 29, 2013
1,576
My understanding is more high level than @crutschow, but to over simplify my understanding; shunt type regulators limit output power by basically shunting (shorting) the coils as necessary. It's very electrically noisy, wasteful and creates heat. But I'm guessing they're cheap and easy to produce and reasonably reliable, though the heat doesn't do any favors for the coils. On my personal bike, after a cold start the cover over the stator gets warm faster than the rest of the motor, I'm assuming it's due to the stator heating up.

In this thread (click) I posted some oscilloscope screen shots from my motorcycle while kick starting, with and without a battery. You can see how noisy the power output is without the battery to clean it up. Some bikes use capacitors in place of batteries.
 
It's quite simple, notwithstanding ts's explanation. :rolleyes:
The SCR's short-circuit the windings (through their respective diodes) when the voltage gets too high.
The current, under these conditions, is intentionally limited by the magnetic saturation of the iron.
The power dissipated than equals the square of this short circuit current times the winding resistance of the coils.
This dissipated power will heat up the coils some but it is much less than the maximum output power the generator can provide at it's normal 12v output.

This does also generate a fair amount of heat in the SCR's and diodes, so the regulator case is typically finned to help get rid of the heat, although they can still get very hot.
(I've had three different units fail on my motorcycle due apparently to their overheating.)

To minimize this, some of the new regulators use MOSFETs rather than SCR's to provide the short circuit, since MOSFETs can have a fraction of a volt drop when ON as compared to the nearly 2V of an SCR.
The MOSFETs can also do double duty as synchronous rectifiers, replacing three of the diodes, which further reduces dissipation.
Here's a short article I wrote on such a MOSFET design.
So I'm new to this forum. I'm poking around a bit trying to figure out how a stator works and what not. I have a mechanical engineering background... not electric. So anyway, I have a motorcycle that is known to burn up stators. There are mixed opinions on how to limit this on our bikes. Some people say to keep ALL electronics on to limit the heat output of the stator. Others in the thread say limiting electronic usage will help keep the stator cool. There is no data supporting either assumption, just theory.

Based on this post, it seems like pulling more power/current from the stator reduces the short circuit current, which would reduce the power/amperage running back into the stator wires, thus reducing the heat. So I should leave many electronics on.

Anyway.. you guys seem smart. Any thoughts? Should I keep my lights on?
 

Thread Starter

Rich:-)

Joined Oct 14, 2016
24
I would say that the thinking of "use the power rather than shunting the power" is still the best assumption.

I use to ride without my front light on when off tarmac (trail riding), after thinking about this more I decided I'd prefer that 55 Watts to be dissipated in my head light rather than in my stator. You can be smart about this too, like if your doing lots of low speed offroad riding (and low engine revs), then switching the headlight off might maximise battery charge, especially if your doing frequent re-starts. If your doing lots of continuous medium to high revs riding then switch everything on (where possible).:)

Another thought on premature stator death is oil quality. Some think that premium fully synth engine oil does a better job of moving heat from the stator, or maybe just runs cooler in the engine. Either way it was noted that for my bikes engine type, most of the stator failures were from bikes running cheaper semi-synth oils, and this had burnt onto some of the stator winding thus heat insulating those coils.
 

Thread Starter

Rich:-)

Joined Oct 14, 2016
24
So, from the above responses I think we all agree that there is no weird magnetic self balancing effects going on.
The stator just makes as much power as it can all the time, once it goes over a predetermined voltage level the short circuit SCR for each phase is fired and then part of each sign wave gets shorted back across the stator cut the output to the bike.
 
Yea, so the change in magnetic field is what's confusing me. People explain that is why electric motors heat up so much when running at low RPM. There is some kind of mag field flux that happens at lower RPMs which increases heat. I was wondering if that was somehow connected to the power demand of the stator when electronics are on/off assuming constant RPM (or maybe should not assume constant RPM?).

Also, I've measured the AC voltage output from the stator when my lights are on/off at constant RPM. The voltage increased when I turned the lights on. I couldn't get a good amperage reading though.
 

Thread Starter

Rich:-)

Joined Oct 14, 2016
24
I think it depends on what type of motor it is, but generally speaking if a motor is forced to run slower it will push harder & draw more current, this is not what a stator is doing btw. The stator runs at full demand all the time.

To see what is happening with the output of a stator you really need to view it with an oscilloscope. The pulses are a mess as it is 3 phases, full wave rectified (so 6 pulses) overlaid on top of the DC from the battery, and as the revs climb they get partially chopped by the regulator.
That is until you switch the lights on which increases the load, thus lowers the voltage output which can in turn stop the regulator from chopping the pulses.
A DMM will try to average out what it's "seeing", which will be increased voltage output voltage from the stator.
 
Much appreciated. I'll keep the lights on I think. I already have a spare ricky stator for the bike. My stocker still works, just have the spare for when the stocker blows out.
 

ian field

Joined Oct 27, 2012
6,539
When looking at a motorcycle power system (commonly referred to as charging system), the basic setup is to have a permanent magnet generator composed of magnets in the flywheel revolving around a stator, often 3 phase. This then feeds to a Reg/ReC (regulator & rectifier in one package) before powering the bike.

Here is diagram of the setup.



The simplistic understanding of what's happening here is that the stator produces the max power is can at a given revs, the load from the bike will hold the voltage level down, at some point the stator will start to output more than is used and the voltage peaks will start to get too high. This is when the regulator steps in and starts to fire the SCR's which short each phase of the stator to the other phases.

It's easy to assume that it is just grounding each phase and thus dumping the unwanted power back into itself, turning it into heat.
Another thought is that it is actually dumping back through the other 2 phases and thus using them a dump resistors..
I'm not sure either of these are correct.

I came across a post on a bike forum where it was suggested that shorting the stator back to itself has a magnetic effect, and that doing so actually prevents the stator from generating - thus no excess heat / wasted power.

My electronics understanding is ok but I'm not so good on magnetism & generators.
What do you think about this please?

Rich.:)
The generator makes at least some pretence of being constant current.

At low RPM there just isn't enough energy to damage anything - as RPM increases; inductive reactance becomes significant.

Most Honda generators peak at 5000RPM, output current tails off above that.
 

rudy93

Joined Nov 30, 2016
8
I know it's an old thread, but why don't leave the circuit open?
All the rectifiers short the stator to ground, but why don't open the circuit that there can't flow any current??
 

Thread Starter

Rich:-)

Joined Oct 14, 2016
24
You can... motorcycles just seem to come with shunt rectifiers but there are now FET based series types that do exactly that, they open circuit rather than shorting.

They have been tested on many motorcycles, on various forums, and have so far been found to work great. They run cooler as does the stator itself when being controlled in this manner.

The only concerns I have is that when you let a stator run without a load the voltage is allowed to peak, so the output wires will have about 80 VAC on them, this might weaken the insulation over time (but it doesn't seem to have so far).
Back EMF clamping - I haven't seen a circuit diagram of what is in these Series R/R's but I would hope that there is some spike damping for when the circuit is broken, because if not there will be large voltage spikes generated and I'd suspect those could harm the stator.
 

rudy93

Joined Nov 30, 2016
8
What do you mean with a high voltage(few hondred/thousand?) You do the same with the old method you short the winding to ground. And will it be spikes or high voltages with high currents. If it only will be some spikes just put a few TSV diodes in parallel.
 

ian field

Joined Oct 27, 2012
6,539
It generates a high voltage at the generator output which can short the windings.
You beat me to it - you can even get a painful belt off an unloaded bicycle hub "dynamo". The bottle dynamos are probably worse - but I didn't bother sticking my fingers on one of those.
 

Thread Starter

Rich:-)

Joined Oct 14, 2016
24
What do you mean with a high voltage(few hondred/thousand?) You do the same with the old method you short the winding to ground. And will it be spikes or high voltages with high currents. If it only will be some spikes just put a few TSV diodes in parallel.
I would expect it to be in the region of a few hundred but have never tested it.

No, shorting the windings to ground does not create the high voltage spikes as the magnetic field is not collapsed, which is what happens if you open the circuit.

I expect the Series regulators will have some form of TSV protection, either that or it's FET's are rated to survive the spikes, the problem with this is the spikes are still created and the stators' own insulation will not be rated to that voltage level.
 
Top