Possible Regulator / Rectifier upgrade.

[rt66john]

Before I go any further:  I am not an electrical engineer, electronics wizard, automotive designer, etc.  I'm just a guy who stumbled around outside the box and bumped into a possible improvement.

Many here have had to replace a stator during their time with an X.  Some have replaced more than one.  Others have gone very long periods of time / lots of miles with the original.  I recently replaced mine at 75xx miles.

Efforts by many in our family to lengthen the life of our stators revolved around removing heat from the stator.  Fan forced oil coolers, full synthetic oil, reducing alternator load, richer tune to keep overall engine temps down, and even the controversial removal of a restrictive orifice were all good efforts by good people who were just trying to keep our bikes healthy after the factory closed.

After getting a replacement stator from Atlantic EH, I set about finding someone to rewind the burned out one, since Marty didn't ask for it as a core.  During my travels on the Interwebs, I discovered a couple of things.  First, it seems our alternators are very similar to, if not exactly like the alternators used in Triumph 995 triples from the same era.  This kind of makes sense, seeing as many of the electronic bits on our bikes were also on Triumphs.  Second, many of these Triumphs were having regulator failures, or a burned up connector between the stator and regulator.

I happened across this thread on a Triumph forum concerning upgrading the R/R unit to something with more modern technology. http://www.triumphrat.net/speed-triple-forum/104504-charging-system-diagnostics-rectifier-regulator-upgrade.html  The original thread is from 2009 but the owner, DEcosse, has updated it throughout the years as he's found even newer technology.  I'll summarize below.

In essence, the stator heats because of the regulator.  Our bikes, as well as many others, uses a series style regulator.  When it sees voltage above a set reference in the diodes, it shunts the excess voltage back into the stator, creating heat.  In the original thread referenced above, DEcosse recommended a MOSFET style regulator to replace the SCR "shunt" type typically found.  While more efficient, it's still a "shunt" style regulator, which dumps excess voltage into the stator.  Another technology, called a "series" style regulator, actually opens the circuit when an excess voltage is seen.  This style of regulator seems to help lengthen the life of stators in bikes that have a history of stator failures, and it's what I decided to try with #2016.  Interestingly, there was a short discussion over on the Resource Board about this very thing, with someone even seeing if a manufacturer would work with us, but the discussion was short and I haven't found anything to indicate other research had been done.

First, I obtained a Shindengen SH775BA from eBay.  They're pretty easy to find as they're the OEM R/R for lots of Polaris Utes.  I then ordered wiring kits from a company named Eastern Beaver.  They have wiring kits with high-grade wire and the correct connectors for the Shindengen R/R.  The kit for the battery uses 12AWG high-grade wire.  I also bought a 30A ATC fuse holder, terminals, and heat shrink tubing.  My plan was to hardwire the stator side to the replacement stator from Atlantic and run the battery kit directly to the battery, bypassing the original wiring.

The battery side of the wiring went in pretty easily.  There's actually plenty of space to run the new wiring along the frame next to the original, up the positive side of the battery box and over to the battery.  I crimped brass terminals to the wires and screwed them down on top of the original battery wires, putting a 30A fuse in the positive side.  Likewise, the stator wires were connected and run near the originals.

The Shindengen R/R is a touch smaller than the original on our bikes but needs to be turned 90 degrees because the left front motor mount interferes with the connectors.  Fortunately, I was able to turn it and use the original regulator bracket by drilling two new holes and using nylon locking nuts with  1/4-20 cap screws.

Initial tests with this setup appear positive.  Before I started this upgrade, I was (again!) getting an intermittent battery light after starting the bike.  On one ride home, I got stuck at a pretty long light and saw the light after idling for only a few minutes.  Tests to the stator were encouraging, but I was getting only 12.05V at the battery with the engine running.  The voltage measured between the legs was good and there was no short to ground on any of the legs.  After the upgrade, I now see 13.95V at the battery with the engine idling after warm-up.

Is this retrofit going to help our stators?  I don't know, but the evidence from others is very encouraging.  Besides Triumphs, many others with Suzuki GS, V-Stroms, Aprilias, and others have made similar updates to the charging circuit, most with good results.  Based on my tests, my regulator was not right and I managed to pull one of the stator wires out of it, and I can't get the stator side connector apart without destroying something, so I chose to experiment with this upgrade on my bike.  This will be a long-term test.  I will provide updates as necessary.

[Jumper]

Keep us in the loop John!!!

Jumper
 

[Donkey Hotey]

OK John, I'm in for the discussion! 

I thoroughly applaud you for taking this on. You also took great care in installing the new regulator. Very nice. Now I'll share what I know about the charging system and my understanding of three-phase alternators.

First off: lets define the wires coming out of the stock regulator and harness:

• Three yellow wires: the A, B and C legs coming from the stator
• Black / White: two wires, identical in function--they're chassis ground
• Black / Red: two wires, also identical / redundant--they go to full-time hot, directly to the battery

That's it! If you really think about that, the regulator effectively only has two wires to connect it to the charging system (ground and hot) and the three wires going to the stator.

Anybody familiar with other common motorcycle charging systems knows that there is usually a switched power wire that tells the regulator when the bike is "on." There is also usually a feed wire to illuminate the idiot light on the dash.

It's important to define these functions ahead of time because our bikes operate uniquely from other motorcycles because of these differences.

• The charging idiot light on the dash doesn't really tell you if it's charging or not. All that light does is tell you if the voltage is below a threshold or not (I think it's below 12V). So if you turn the key on, have a strong battery and maybe just took it off of a charger, that light might not illuminate, even if the engine isn't running and isn't generating a single watt of charging power. The light will illuminate when you start cranking and may stay on after that if the bike doesn't start. In summary: idiot light is nothing but a basic voltmeter.
• The voltage regulator is ALWAYS in the circuit. That means that whether the key is on or off, the regulator is regulating. The first place I noticed this was with the use of a Battery Tender. Ever notice that a Battery Tender LED NEVER gets to solid green? Mine is cycling between red and flashing green right this second. The reason? The regulator is sitting there, wondering where the heck the extra electrons keep coming from and it's regulating the battery back down below where the Battery Tender wants to take it (13.8V before it ramps down to 12.6V). This behavior DOES support the theory that the regulator is shunting the extra current to ground, rather than operating as a Pulse Width Modulator.

So what did you do about the extra wires? I would imagine that the new regulator must have its own idiot light feed and must want a switched hot wire to tell it when things are operating. No?

With that out of the way, lets discuss how things operate. I'm also no expert on this but, have read many theories on the web. I once believed this style regulator shunted the extra current to ground. I was later told that the SCRs are just switching the current on and off (pulse width modulation) to control the output. Before we get to that, lets talk about the stator.

Our charging system is dumb. Like most motorcycle charging systems, it's got fixed magnets, spinning inside a stator. With this arrangement, the amount of energy generated is proportional to RPM. If you unplugged the regulator and measured the yellow wires, you'd get all kinds of wild voltages as you revved the engine. When it's connected to the battery and rest of the electrical system is when it gets anchored to something and becomes meaningful.

When our bikes need wattage to charge the battery and run the rest of the system, that presents a load. The stator pours its heart out to keep up with that demand. If the demand is approximately equal to the supply, it goes unregulated and just charges.
When there is a surplus of wattage is where the regulator has to step in. If the regulator were forced to shunt the extra current to ground, the act of disconnecting loads, turning off lights, etc, would not help the stator at all. The regulator would just shunt the extra current to ground.

If the stator were capable of 400W, the bike would use it, or the regulator would use it. Nothing would go unused. The load on the stator would forever remain constant. What this also means is that if you turned off your headlight during the day, it wouldn't do a thing to help the stator but, it would sure put an extra load on the regulator (needing to burn another 55-60W of power because the headlight was off).

On the other hand, if it were being switched in and out of circuit (PWM) to regulate the current, the unused current would remain in the stator. This surplus of wattage would actually RAISE the AC voltage on the yellow wires, when in a non-charging state. In theory, if you started the bike and measured the voltage on the yellow wires, you might see 14VAC between any pair.

Now if you pulled the fuse on the headlight, you might see that voltage go up to 15VAC, even though the DC side of the system remained constant. Keep in mind that all of this is with a fully charged battery and might not behave like this until the battery is back above the cutoff voltage.


I don't have a variable AC power supply to test this but, that's my theory on how the system should behave. I could be totally wrong but, I'm glad you're tearing into this. Maybe we can collectively come up with some solid upgrades to the longevity of the system?

[wytfut]

John....
I find all of this very intriguing...   as I've always felt our charging system was less than satisfactory (yeah I know, many other bikes out there operate the very same way). .....   

I'm crossing my fingers that you maybe onto something....      let us all know...

Bruce

[rt66john]

I was hoping you'd respond, Don (why do I want to sing "Lord of la Mancha"? ).  You're a bright dude.

I'll respond as best I can based on what I've read about m/c charging systems while researching this.  Once again, I am not an expert and I may be really off base here.

With a few exceptions, m/c alternators consist of a stationary stator bolted to the inside of a cover (the primary in our case) and a rotor with permanent magnets bolted to the crankshaft.  The only difference with what you described is the rotor in most m/c alternators is outside the stator, not inside.  I imagine this is because the rotor has to be large enough to hold enough permanent magnets of sufficient strength to generate enough current to meet demand.  At any rate, each phase of the alternator produces a bunch of alternating current (AC) electricity and passes that electricity over the 3 yellow wires from the stator to the regulator / rectifier (r/r).  

The r/r has two jobs.  The first is to change AC to DC (direct current).  There are diodes in the r/r that allow electricity to flow in only one direction.  Once that is done, it's on to the diodes that provide voltage regulation.  Those vary with design, but if I understand what I read properly the shunt style regulators, like the OEM, will actually dump excess voltage back into the stator, not to ground.  This can generate more heat in the stator windings.  The series type of regulator, like the unit I installed, will actually open the circuit when voltage is over a reference point in the diodes.  This can help reduce heat in the stator.  In fact, one of the Aprilia forums I read had a thread where one guy aimed a thermal camera at the stator cover for his bike.  The Shindengen r/r recorded a temperature about 20 degrees F lower at the cover compared to the OEM r/r.

You are right; our alternators are dumb.  The amount of current and the voltage will vary with engine RPM to a point.  They produce at full capacity somewhere around 3000 rpm.  The service manual says the OEM alternator should produce between 60 and 70 VAC each leg at ~3000 rpm.  After a certain voltage, the OEM regulator will begin to shunt current back into the stator (not ground).  Adding a load will reduce the amount that is shunted, if I read the theory correctly.  The r/r I installed opens the circuit above a threshold voltage.  Adding a load reduces the time the circuit is held open.

As far as wires for the idiot light and a switched wire to tell the regulator when the engine is running, I don't see either wire on either the OEM or the after-market r/r.  The OEM has 7 wires total; 3 AC in wires from the stator, two 12 VDC positive wires out, and two 12 VDC negative wires out.  The positive wires run under the electronics cover and disappear into the harness.  If the wiring schematic in the service manual is accurate, they connect to one side of the 50A circuit breaker, which feeds the battery, and the fuse bank and relays to power the bike.  The negative wires go to the chassis ground.  The r/r I installed has 5 wires.  Three are the AC input wires, but only two 12 VDC output wires.  I ran both of those directly to the battery, with a 30A fuse in the positive side.  I left the OEM wiring in place so the rest of the bike could have power.  As far as I know, the r/r doesn't really know if the engine is running or not; it just waits to see AC power from the alternator.  The idiot light, like you said, is just a poor man's voltage meter.  The ECM will switch it on if voltage at the battery falls below a threshold.

My Deltran brand battery tender has never behaved as you describe yours.  Mine will light the red lamp solid if the battery is below a certain voltage, flash one lamp when battery voltage rises to 80% of capacity, and finally light the green lamp solid when battery voltage is at a point where the charger will no longer charge.  I have always run the battery tender leads directly to the battery posts.

[Donkey Hotey]

Ahhh...you caught me! I said spinning inside, rather than outside. My brain was thinking of three phase motors. Yes, you are absolutely correct--the magnets usually spin around the outside, though the result is the same.

As for shunting power back into the stator, I'm not sure how it would really do that, unless they were shorting the coils to one another, which would make no sense. Again, a voltmeter is the best indicator as to what's happening inside. If you tapped a pair of the yellow wires and measured the voltage of the stator, while the loads were changing on the battery side, that would tell you what was happening.

I did a little searching yesterday and from what I could find, the terms we would be looking for would be "shunt regulator" and "switching regulator." I couldn't find a single motorcycle, ATV type regulator that was a switching type (PWM). That suggests that they're ALL shunt type. The SCR to MOSFET change is a worthy one but, it sounds like it just improves the durability of the regulator. I don't know if that would cause any change in the life of the stator.

Has anybody ever put a couple of thermocouples on the primary cover to monitor the temperature of the stator area vs somewhere else? Maybe one on the screw head right over the centerline of the crank and another on the clutch cover? It would be interesting to see if that area is hotter than the rest of the cover, or if the aluminum is effectively conducting it away. I'd try my laser thermometer but, I'm not sure they work very well on chromed surfaces.

On the Battery Tender: yes, all three of my EHs behave the same way and it's a genuine Battery Tender Plus. None of my other bikes behave that way. My EHs started out with the connections directly at the battery, though I moved them to the starter relay as soon as I was into the electrical box for other reasons. I prefer to keep extra connections away from under our chromed battery covers. Too risky for a short.

I have to wonder if our stator life is actually shorter than other bikes or we're just sensitive to it. Look up nearly any motorcycle and charging problems are very common. As we've discussed, the components are the same as other bikes. I've never been a fan of fixed output alternators for al these reasons. My 1986 Gold Wing ate a stator and regulator like clockwork every 60K. I put 190K on that bike and did exactly three alternator / regulator swaps. That was in a water cooled bike that was engineered to death. They still failed.

You're still on a good path. One of the things that I learned with the Gold Wing was that I suspected the regulator was the first part to fail and that the extra load from a partial internal short was what would ultimately fry the stator. I got rid of it before being able to test that theory beyond the 240K stator.

One last thing before I head to work: it's important to keep a healthy battery in a motorcycle, to reduce the load on the charging system. That sounds like a platitude but, it's really not. Some people will have a battery with a weak cell and keep running it. The charging system is dumb and it will work its ass off trying to get that weak battery back up to 14V. All that happens is the battery absorbs the charging current in heat and it kills the stator. For want of a $100 battery, they end up tearing into the engine.

[2CoolWheels]

  Great information guys. Much of what you've discussed confirms some of my own observations. I remember when I first got into EH's, another owner told me to keep my highway lights on because the charging system would stay cooler and last longer. It seemed counter-intuitive to me at the time, but now that I understand the regulator shunts back into the stator, it actually makes sense.
  I've replaced the stator twice in 741, which has no accessories on it at all. It has just under 30K miles on it now. I have never replaced the stator in 901 which has highway lights,  and added  set of oil cooler fans as well as an electric oil temperature gauge. It has just over 30K miles on it. Apparently the load on 901's charging systems minimizes the shunt time compared to 741, and hence a cooler stator. However, I'm sure there are other factors involved.