Found this.
Original article.

Facts about shunt-based regulators

Facts about shunt-based regulators

I feel that in all the debate over burnt stators, etc..., that a lot of misinformation keeps floating around in regards to regulator/rectifiers (RR), thus giving shunt-based RR’s a bad reputation in the process. So I come-forth in defense of the poor little but mysterious shunty, by starting this thread in an effort to discuss, dispel, and clarify a few myths that seem to run rampant in these ongoing debates. Please feel free to link-back to this thread anytime or contribute, as I’ll standby (Pepsi-challenge) the following:

Myth: Shunt-based regulators cause the stator to run at max power output (100% of the time) regardless of the electrical demands on the bike.
i.e., if you have a 500 watt stator, and you’re bike is only consuming 100 watts currently, the stator is still producing the 500 watts by dumping (shorting) the excess 400 watts to ground!

Sub-Myth: Because the stator is held at max power 100% of the time, excess energy causes the stator coils to work harder, overheat, and harder on the wiring & regulator.

Sub-Myth: Because the stator is held at max power 100% of the time, this translates into additional loss in horsepower, poor gas mileage due to the constant magnetic "drag" of generating wasted electricity.

False, false, & false – if you literally consider the terms Watts and Power in the context of "wasted" energy.
Typically, during low electrical demands, aka regulation, shunt (SCR-based) regulators continuously waste the equivalent of about a 60 watt light bulb, which then must be dissipated by the regulator. It is true, however in that the stator does in-fact produce (move) 100% of the max rated AMPs being generated, but as I’ll explain below, this does NOT translate into wasted power/energy, and is as by design. Given that said, there is credence to the additional heat and horsepower loss argument, BUT only in the context of the 60 watt loss and not a 400 watt loss in the above example.

I find it a great marketing ploy to crap all over shunt regulators. The biggest being that that shunting the excess current is somehow bad for the stator coils and/or generates excess heat. While these events do occur (I’ll get back to this), they don’t occur to the huge extent that the average layman would equate to dumping 500 watts equivalent energy of a toaster oven – it simply doesn't/can't happen. Some people suspect that shunting is some how related to or causing premature stator burn-out, thus somehow swaping with a "series" regulator is better. While pre-mature fried stators are and have been a problem, I caution anyone that is quick to single-out shunt regulation as a considerable contributory factor. I attribute much of the myths to a general lack of understanding about the unique aspects of PMA’s, which is acceptible given the complexities of that subject. So let me start there breifly:

PMA’s (permanent magnet alternators) affectionately but loosely referred to as the Stator, unlike field-driven alternators (automotive), are alternating but fixed (think quantity) current electrical generators. What that means is that they "pump" the same amount of electrical current regardless of voltage state, and spinning them faster doesn't produce more current proportionately as it will voltage. While shunting/shorting a battery or your home’s electrical socket may be a very bad thing, properly shorting a PMA field coil IS NOT! The reason lies much in it being a close-loop paradigm, coupled with the phenomenon of inductors (coils), impedance, etc... The best analogy I can think of is a closed-loop hydraulic system where one has identical cylinders/rams on both ends respectively (1:1 power ratio). It's not the best analogy – I know - but it does do a decent job at highlighting the core fundamentals of current, voltage, how these both translate into power, and more on-topic, how a short behaves in this type of closed-loop induction system.

If one pushes/pulls the ram on one end (aka master), so equal is the reaction of the ram on the other end (aka slave). Vice-versa, if there is resistance in the slave moving, then equal resistance is felt at the master. Now, if you monitor the PSI in the hydraulic lines connecting the two rams, you will observe momentary fluid pressure Fluctuations as both rams desire to equalize – especially when there is resistance. This is akin to voltage pressure, as it takes pressure to overcome and do any real work (power). The slave ram basically translates fluid pressures into mechanical motion energy again which moves the ram. As a result of this energy translation, less pressure is returned in the hydraulic return line going back to the master ram (aka closed loop system). In the electronics world we call the differences in these supply/return pressure a voltage-drop. All loads cause voltage-drops to occur because it means work being done aka power/energy consumption. The more work (resistance) the slave ram has to overcome, the more pressure occurs in the supply lines at any given time, the greater the voltage-drop on the return line, AND MORE IMPORTANTLY equally is the resistance felt on the master ram! Again, mechanical energy is only "transmitted" from point A to point B in the form of voltage (pressure) - not current. But that doesn't mean current isn't important! Current acts as a reservoir to maintain these voltage differentials for a quantitative duration. Current can be represented as the quantity/volume of hydraulic fluid moving at any given time around the loop, therefore the more fluid you have, the bigger rams you can move, etc...
Important note: no matter how much pressure you require, and how great the pressure-drop, an equal amount of fluid flows to and from of the master/slave rams.
Side note: Did you know that every AMP you get from your power company is returned to the power company?? Yup, it goes right through your lamp and back out to the electric co. So what you are actually paying for then is the pressurization of these amps, thus the difference in the amount of pressure you returned (delta) is actually being measured over time (aka Kilowatt hours) and billed to you.

Back on-topic, so, how does this dispel the popular myths regarding shunt regulators??
Simple... If I were to "short-circuit" the lines in my hydraulic analogy (bridge the supply & return lines between the master and slave rams), the fluid (current) would flow freely back to the master ram with little-to-no resistance (pressure). Since there is no pressure, there is no voltage-drop situation, therefore impossible to transform mechanical energy from one state to the another, i.e. no way to transmit the mechanical energy being inputed into the master ram. Instead, the power I put into the master ram cycles immediately back around the loop of hydraulic lines via the short, and into the opposite side of same master ram piston! This is Neutral work-effort - aside from possible friction and other transient resistance in the lines themselves, or even the bypass/shorted junction (remember that point). This is what fundamentally occurs when a regulator shunts excess current to ground. It is simply diverting excess current directly back to the stator with minimal voltage-drop (back-pressure). If the regulator failed to do this, current if not consumed (voltage-dropped), would otherwise translate into increased voltages (ohms law). The result may sound crazy, but during the shorted duration your stator is actually magnetically unloaded thus less/no drag on your engine (there goes the gas mileage and lost horsepower myths). One again, let me just say t shunt hat shunt regulators DO NOT keep your stator continuously loaded (100% duty), whether the bike's electrical demands need it or not! This is misinformation I see all too often, parroted from one person to another, perhaps also perpetuated by vendors looking to sell the next best "snake oil" product. I will however, admit that MOSFET regulators are marginally better, but not by the leaps and bounds some vendors claim over OEM regulators. Shunts do not "waste" as much power as some may try and make you believe. Furthermore, due to the constant-current nature of PMAs, true-series regulation is not the ideal method for regulating power from these devices. While series regulation does and can work fine, it is simply not complementary to the dynamics associated in permanent magnet inductor generation systems.

 

Continued:

Where I will concede is on a technical-level (hence the snake-oil comment). And that is, remember my point about the efficiency of the bypass junction – aka the short? Well, *most* OEM shunt regulators use SCR's to facilitate the actual shorting. The big problem here is that the electrical characteristics of SCRs do not offer an ideal short, because they still cause a voltage drop to occur across them. Albeit nominal, it’s still a drop nonetheless. On average, this equates to about 1.4-2.0 volts in totality, which translates into about 45-70 watts of power being consumed. The result is heat dissipation by the regulator (SCR’s), think 60 Watt light bulb and now you know why these regs get so hot! That said, some folks may wonder how then can the newer MOSFET-based regulators, which BTW are still shunt-based units, run cooler?? Again, this has much to do with voltage-drops (or lack thereof) as MOSFETs have significantly less conductive resistance, offering a much better short than SCR’s during the shunted cycles. Again, these newer MOSFET regs only further solidify my points that shunting isn't a bad thing, wastes power, or causes your stator to "see" 100% power load continuously. The truth is, in fact, that shunt-method regulation of PMA sources is the preferred/ideal method due to the inherent characteristics of PMAs being an inductive (coiled) constant-current closed-loop type of animal.

In summary:
* Shunt regulation is not a bad word. But if you still want to save a lightbulb's-worth of "wasted" energy demands off your stator, going with a MOSFET-based shunt regulator isn't a bad idea at all, given your OEM RR is SCR-based (most are).
* Shorting stator coils yields the same behavior of magnetic drag (magnetic reluctance) as open circuiting them, thus no Power is consumed and no drag is felt.
* Be mindful about the term 'series' being thrown around loosely. Series in terms of regulation describes the electrical circuit relationship/architecture as it is physically connected to the main circuit it is trying to regulate. In this case, ALL shunt regulators are classified as parallel regulators. Linear and switch-mode regulators are classified as series regulators. MOSFET’s can be implemented in either parallel or series applications, thus should not be assumed the classification determining factor.
* Are shunt RR's a factor in pre-mature stator burn-out? Sure, maybe, could be, BUT, so can be said with all methods of electrical regulation as they all put some degree of stress on the stator. We just can go around saying the shunt regulation is yesteryear, "cave-man" technology and the only reason the manufacture implements it is because it's simple & cheap, and of that, might be only slightly true when considering SCR’s –vs- MOSFETs.
* No I don't have a personal theory on the notorious burnt stator issues. However, I will say it is most likely a culmination of many factors including engine heat, # coil turns, gauge wire size, connectors, epoxy resin, strength of magnets, battery condition, SCR’s.

This thread is open for discussion, questions...

References:
Voltage regulation
http://www.tpub.com/content/neets/14.../14178_146.htm
Permanent-Magnet Moving-Coil movement
http://www.azsolarcenter.org/images/...teries/ch6.pdf
continuousWave: Whaler: Reference: Permanent Magnet Alternators
http://www.takisnet.org/~abayko/appnotes/vreg.pdf
http://home.comcast.net/~loudgpz/GPZ...gnetField.html
http://www.zefox.net/~bob/mc/vfr/alternator.html