How to fix common regulator/Stator failures

ummm candyred....he did already. stay current please

 

sorry I will try

So is this a Chinese RR or a modern MOSFET Shindengen RR?

Sent from my A0001 using Tapatalk

 

As far as I am aware, it is the later of the two OEM types that Honda fitted to the 800Fi.

I have had RR failure before, both on early factory-fit RRs and upgraded aftermarket units, and really thought I had it sussed. I'm just confused as to how the unit is passing a diode test but failing to output what it should.

I've ordered an R1 RR, a mod' I did on my first bike - a '91 VFR750. Problem is, I don't really feel confident about the RR being the source of the problem. And at least until it shows up I have an unusable motorcycle sat outside my door when I desperately need it running! :grey:

 

there is two stages in that animal, the part that converts ac to dc, and the part that maintains the voltage. the diodes change ac to dc in one of two ways, 1/2 wave or full wave rectification

 

The insulation is abraded from oil contaminants, and may be thinner than it was when new. Also, breakdown voltage decreases with dielectric thickness.

CandyRed does this open up the possibility for rjgti's solution? You mentioned 100 V. That is not "high voltage" and 60V is on the same scale, so I would think so.

If yes, and you had this stator in your hand, how would you test and validate which of these two problems the stator exhibits?

 

Yeah I see how rjgtis point can have some merit as the dielectric thickness is approaching zero, Voltage/thickness, even with voltage relatively low, would go to infinity. But the whole whole reason that thickness would be approaching zero, would be because of thermal expansion. That's what I was getting at.

 

But as covered earlier, I would not go about testing the stator in hand, out of bike. The only stator testing that would have any value, would be while in bike and at operating temperature. This is where a stator could short to ground, if faulty.

Sent from my A0001 using Tapatalk

 

VFRnewbie, you've done the tests needed. Your R/R is internally shorted, the rectifier diodes are OK, but the regulator part is shorted on one phase. What I'd do is get a new FET-type regulator which will run cooler, and mount it someplace cool. I put mine on a passenger peg bracket in the airflow. Your stator measured OK with the R/R disconnected. It's just barely possible that the stator is shorted when hot, but in that case, it'd surely look burnt, and it looked good.

 

If the stator were taken out of the bike then I would use a standardized test method while baking the stator at 210°F.

You have lost a bit of conciseness that you originally started with. You are talking about a crack which exposes itself due to expansion, which is different from there being a layer of insulation that is thin, worn, but which the voltage travels right through under certain conditions. Yes we are all splitting hairs just for the interest of it.

I just came up with the test between these two solutions: Get the stator up to temperature. If the problem is thermal expansion the problem will now occur at all RPMs here forward. If it is related to dielectric breakdown, it most likely will start at some higher rpm, and go away when the rpm comes back down. :congratulatory:

 

Will a short in the regulator short the output of a stator phase the way these are built? I am presuming his measurement of the stator output was done on the stator side. If yes this makes sense.

If true, continuing to run this bike will increase the likelihood that the one phase of the stator will fry.

 

If I understand you correctly, you're saying it makes sense because the stator outputs a to b, a to c, c to b all read healthy when measured with the stator disconnected, but one of these readings drops to almost nothing when measured with the wires connected to the RR?

 

it could if the diode was forward biased and the gate of the mosfet had expierenced punch through and shorted to the drain in reg portion of r/r

 

anyhow ya put it the r/r has let out the magic smoke

 

My observations about the V four charging system (I'm a retired electronics engineer) for whatever they're worth.
The basic issue comes down to cooling. A permanent magnet alternator is always on, generating its maximum power which increases with RPM up to about 4000 or 5000 RPM, after which it stabilizes due to the inductance of the coils. That power has to be regulated or it will overcharge the battery and burn out lights, etc. That means burning off the excess someplace, which generates heat. The easiest solution is an electronic rectifier which turns the AC from the alternator to DC for the battery, combined with a regulator which shorts out the alternator stator when the battery voltage is high enough. The alternator is always putting out maximum current, on my bike about 25 amps above 2500 RPM, but the regulator clamps the voltage so that brief bursts go to the battery, enough to maintain the system's voltage. As you can see, the battery's job is to smooth out the current flow, take these bursts and feed them out at a constant 12-14 volts.

A weak battery will cause any measurements to be off, or wander around, depending on just where the voltmeter took a snapshot. If you get a chance, look at your system with an oscilloscope at various points. You'll see all sorts of spikes and garbage on top of the DC. A low resistance direct connection from the R/R to the battery will help stabilize the output voltage.

The stator has to dissipate considerable heat due to the current flowing through its wire resistance. It gets rid of the heat via conduction to the alternator cover and passing air and by an oil spray. If you replace a stator, be sure the stator and the cover are clean and in good thermal contact before buttoning them up. 6th gen VFRs with VTEC apparently have a pipe which blocks the oil spray to the stator, so they run hotter than earlier ones. Air cooling is pretty marginal down in the lower fairing.

The regulator also has to get rid of the heat due to current passing through the diodes and the regulating devices. Diode heat is proportional to load current, so if you're running extra lights and a heated suit, the diodes will run hotter. The regulator part generates heat as it shorts out excess current, so it runs hotter as load goes down. FETs have a lower voltage drop than SCRs used in OEM R/Rs, so FET regulators run cooler. The R/R is also small, so it's possible to move it somewhere cooler or with better airflow than where Honda put it. My 5th gen VFR has an extra terminal on the R/R which is the voltage sensing input. That input goes off into the wiring loom to someplace in the instrument area. If that wire opens up, the regulator will go to full output and boil the battery. If there is a bad connection in that sensing circuit, the voltage will be too high at the battery, again shortening its life. Most replacement R/Rs don't have the remote sensing wire; one less thing to go wrong.

The best solution is a "series" regulator which opens the circuit between stator and rectifier / battery when power isn't needed. There are some aftermarket series regulators, but they haven't worked very well on high RPM engines so far. A series regulator doesn't burn off excess energy, it cuts off the alternator, reducing the heat load on all parts. Suzuki is starting to switch to series regulators, so maybe they'll come up with one for a GSXR or other high revving engine which we can use.

My somewhat educated opinions. I haven't tried aftermarket wiring kits, etc. so I have no idea how well they work. I have written about what I have tried. At the moment, I'm running cheapo Chinese MOSFET regulators which are of a different design than Shindengen. So far, they are working fine, but it's an experiment, and I wouldn't necessarily recommend them to anyone else. I have wired my R/R directly to the battery through a fuse and soldered as well as crimped all the connectors involved, since the connectors seemed to heat up at the crimp joints. I still have connectors to the R/R so I can swap in a spare at the roadside (As I said, I'm experimenting.) I have removed the ground block / diagnostic connector and soldered all the grounds together. At this point I'm pretty comfortable with the way the charging system works. If something burns up, I'll let everyone know.

 

That is how I read H3nry's analysis, that a short in the R/R, yes creates continuity from the one phase of the stator, to the R/R, to ground.

Grounds do affect the system from start to finish. For example a dead battery acts as a ground. Run the bike and it demands a huge current from the stator, a current on par with the rated current of the system. Then anything in between is at risk of frying from the stress.

We are rooting for you!

 

It would seem so, indeed.

 

Thanks H3nry, useful information.

 

Thanks, I'll let you know what the result is once my wait for a new RR is over! Right now, the sun is shining here for the first time in days and my lack-of-functional-bike is starting to feel like a missing limb...

 

I would have thought that with all these smart electrical engineers here they would have been able to come up with a R/R that worked so we wouldn't have this continual meltdown. We can understand these things are built to a price so it is a crap shoot but if something decent was available I'm sure people would go for it

 

Engineers report to accountants, yes. If the engineer were in charge, the R/R would cost $750, you could use it on every brand of bike, it would last forever, and it would massage your back for you.

I do not see the R/R quality as a problem. Take the price and depreciate it over 40,000 miles or whatever life it would have if everything in the system is working correctly. That is reasonable.

The core problem is that the motorcycle system is not closed loop and fault tolerant. This means handling errors or at least warning you of them. Such would include, when the voltage at the important junctions is high/low, when current is high, when the bike is off, if there is a drain larger than 3mA warn you about the short, and if any of the ground blocks are no longer grounded throw a warning.

If all of these variances were caught immediately, there would not be time for the failure to fry the rest of the components they are attached to. Replace the aged part, or clean a connection, and you're done.

The problem is systemic.