regulator failure

Just replaced my voltage regulator again after less than a year's
service.

This is an automotive type 3 wire regulator, energizing the alternator
rotor through slip rings.

Is this second failure pure coincidence and bad luck or could there
be a problem such as a high resistance or bad diodes taking out
the regulators ?

Any insights or suggested diagnostics appreciated.

 

Assuming your battery's okay, get your alternator checked out by someone
good. There are various things a failing alternator can do that will fry a
voltage regulator. Worse yet, the obvious symptom is a failed regulator, and
the guilty alternator usually continues to put out some power, so many will
just replace the regulator and look no further. (Had a sports car years ago
that was eating regulators. After replacing the third regulator, my mechanic
figured there had to be more, and found my alternator had a loose wire that
*sometimes* flopped over where it didn't belong, zapping the regulator.)

 

Come to think of it, I fried a regulator years ago with a bad
path to ground.

I've already cleaned up the battery terminals which had gotten
crapped up in the humid weather but I should look deeper into
the guts of the charging system.

Thanks

 

What bike is it? Are we talking a separate regulator here or a combined
reg/rec?

 

This problem has been discussed in detail for the past 10 years on the
VFR list.

http://tinyurl.com/ystg5k

Here are the Cliff's Notes on solutions.
Run a large diameter ground wire directly to your battery from the RR.
Make sure it is getting a good supply of cool air. I mounted a
computer fan to mine.
Clean the back side of the RR, the mounting area and apply dielectric
grease to the mating surfaces.
Clean all connections. And if you are really detail oriented, change
crimped connections to soldered connections.

HTH,

JohnA in LA

 

Is it actually a car voltage regulator, and not the OEM regulator, in
an attempt to save a bit of money?

http://www.xmission.com/~wendell/GS/node3.html
http://www.xmission.com/~wendell/GS/node27.html
http://www.xmission.com/~wendell/GS/node3.html

Is your GS wired like the diagram? I see three AC phase wires, a
ground wire, the red DC output to the battery, and a blue sensing wire
going to the regulator. The brown wire would be the field excitation
current and the black wire would be the excitation current return to
the voltage regulator.

Does your GS have a generator warning light in the speedo?

A pass transistor voltage regulator would simply apply 12 battery
volts to the transistor, and the current through the transistor would
be reduced by its internal resistance.

In the diagram, the transistor's gate gets about 14.5 volts from the
rectifier, and that shuts the transistor off until the voltage drops.

So the transistor is a switch that has to cycle continuously while the
engine is running and changing RPM.

The resistance of your alternator rotor might be low, or one of the
brushes may be shorting to the brush rigging, and that might be
blowing out the transistor in the regulator.

 

R100GS with separate diode board.

 

Right now, I'm running the OEM Wehrle unit. Prior units
have included Transo and Hella, all of which have been
used regulaly and sucessfully by other owners and shops.

All of the above are basically automotive 3 wire units,

Probably minor differences in wiring but not in charging,

Yes, Appears to work fine.

Had it open yesterday and nothing real obvious. Diodes
appear to be OK. Might have been a marginal wire to the
brushes and the battery terminals were crapped up a little.
All of these would be high resistance conditions though
not shorts.

There may be some other intermittant heavy draw though.
Had it fail to start once with low battery symptoms followed
by some perfect starts and charging. I should take another
look at the cables and grounds again too.

Today, I spilled in the rain while trying to sort out the frikking
electrics, so I'm nursing a very sore shoulder and contemplating
the mystery. Probably hold off on wrenching for a couple days
until I've got both arms and shoulders working well again.

 

Oh blimey.

It's usually the diode board on ShiteldBoxers that gives trouble.
Funnily enough, I was looking at an R100RS recently with just that
problem.

And IIRC, a diode board and associated bits is something like £300
($600). Can't remember if that's with the regulator or not.

My money would be on the diode board being faulty.

 

I've heard that from others but can't quite figure out
how that could cause the symptoms I'v seeing.

Measured with an ohmmeter, the diode board looks
OK, at least the three large ones for rectifying the
AC output. I've got a spare I can look at and compare.

Motorrad Elektrik in the U.S. sells a well thought of
aftermarket diode board for $90. The BMW price you
quote is ridiculous and probably accurate.

http://www.motoelekt.com/charging.htm

 

The usual 3-phase full-wave rectifier bridge has SIX diodes and the
three AC input phases are hooked up between the anodes and cathodes
of THREE pairs of diodes.

http://www.electrosport.com/Images/instr.diode.test1.pdf

Voltage on each phase rises in rotation, A phase, B phase, and C
phase,
and power from each phase has to return to a different phase in that
rotation.

All SIX diodes get used, as the current changes polarity at varying
frequency.

If ONE diode is OPEN, AC power from one phase obviously cannot return
to the
other phase for half a cycle, and you lose half of the alternator's
output power for half a cycle.

If ONE diode is SHORTED to ground, you also lose half the alternator's
output.

I disagree with Electrosport's statement that you can't test a diode
with an analog meter. The main thring that happens with the analog
meter is that it puts so much current through the diode, you get a
reverse polarity reading, IOW, the diode reads continuity AGAINST the
arrow, instead of with the arrow.

No matter. If you put your positive ohmmeter lead on any of the yellow
AC inputs, and the negative lead on the rectifier bridge's black
output wire, and you don't get a reading on the Rx1 scale, switch the
leads and check again, because you may just be over-biasing the diode
with the analog meter.

If you get a reading from any yellow wire to the black wire, you
should get the SAME reading between each of the other two yellow wires
and the black wire.

It might be around 15 ohms, but the exact reading is dependant on the
internal resistance of your analog meter. That tests THREE of the SIX
diodes.

Then put the negative ohmmeter lead on the red wire and run your
continuity check to each of the three AC input leads. Again, you
should get the SAME reading on each of the three phases. If they they
all read the same, those THREE diodes are OK.

If you only have THREE power diodes, you have a HALF-WAVE rectifier
bridge. I cannot imagine any sane electrical engineer designing such a
device that ignores half the power of the alternator.

Electrosport makes a modern rectifier unit housed in a heat sink. The
price is $129.

http://www.electrosport.com/Images/fitting.esr450.pdf

 

I uncovered a second problem in the starter. One bad spot
on the starter seems to have been shorted out and creating
a serious path to ground when the starter was positioned
just right.

Battery would drop below 9 volts without cranking while
at some other times, it would crank just fine. The starter
was just over a year old and the shop I bought it from
replaced it without question when I reported the symptoms.

I'm not really clear what the exact failure was or how a
bad short would take out a 3 wire regulator, but having
replaced both of them everything seems to be working
fine again.

The tests I should have made immediately were to
charge and test the battery, then monitor battery voltage
while attempting to crank the bike over. The voltage
drop was an obvious clue that I should have caught
earlier. Instead, I let the bad starter trick me into thinking
there was a bad connection causing it to not work.

Thanks to all for the suggestions and advice.