Cushman’s new Permalite lighting system was
introduced for the 1949 model year to replace the separate AC generator that had
formerly been used. The new system
used two X6675 lighting coils located on the magneto plate and wired in
parallel. When Cushman first introduced the Permalite system coils ware large,
measuring approximately 1.75 inches across and had approximately 150 turns of
wire. The headlight and the taillight were both connected to these coils. There
was no stoplight in those days. The
coils provided adequate voltage and the lights were fairly bright.
In a simple electrical generating circuit like Cushman
used the voltage developed is determined by the magnetic field strength, the
number of turns of wire, and the speed at which the magnetic field passes the
coils. It can easily be seen that if the voltage is to be limited to about 7 .7
volts maximum at near 4000 RPM to avoid burning out the lamps, the voltage
developed at an idle speed of 1200 RPM is only about 2.31 volts.
Actually, it is not quite this low at idle for reasons that will be
Perhaps Cushman received complaints about the lamps burning or perhaps they just
wanted to save wire, but for whatever reason the number of turns in the coil was
reduced to about 100 sometime around 1952 or 1953. The new coils, part number
X8804, measured approximately 1.375 inches across. With a 33% reduction in the
number of turns the lights were dim at low speeds but lamp life would have been
greatly increased. In my opinion Cushman went overboard in reducing the number
of turns on the coils. I have found that adding about 30 turns of wire to the
X8804 coil improves the lighting while still providing acceptable lamp life. If
you try this use number 15 or 16 wire and very carefully wind 30 or more turns
on top of the original winding in several rows. You must be sure to continue to
wind in the same direction or you will reduce the voltage instead of increase
it. Glue or tack down each row to keep it in place before starting a new row.
You cannot wind as neatly as the factory did but if you are very careful you
should get close to 30 extra turns. You
can coat the completed coils with epoxy to keep the turns in place. Be sure that
you do not make them so large that they no longer fit on the magneto plate. If
you plan on keeping it a 6 volt system you must wind exactly the same number of
turns on each coil. If you would
like to change it to a 12 volt system it will not matter if the two coils have a
slightly different number of turns. To
convert it 12 volts put the two
coils in series by removing the wire that goes to ground (the inside of the
winding) from either coil and insulating it. Now remove the opposite coil’s
outside lead from the output post and reconnect it to the wire you removed from
ground. Change your lamps to the ones used with the 12 volt system. 12 volt
lamps require half as much current as the former 6 volt lamps so with the
voltage doubled the amount of power generated remains the same. The brilliance
of the new 12 volt lamps will be about the same as the former 6 volt lamps. You
may have a little less voltage loss in your light wiring because of the reduced
current but it is not significant. The main reason for changing to 12 volts is
that you can now use a half wave rectifier to steal a little current and use it
to charge a 12 volt motorcycle battery. There are more efficient ways to charge
batteries but the complexity make it impractical in most cases
In paragraph two above I explained that the voltage generated is a determined by
the magnetic field strength, the number of turns on the coil and the speed at
which the magnetic force passes the coil. There are other factors that can
modify this equation. The cross sectional area of the coil's core can purposely be
made small so that it will saturate and limit the current to any desired value.
This principal is well known and is used to limit the current in all
modern car alternators. In the 6 volt Cushman lighting system the core cross
section is relatively large for the current required and only a small amount of
core saturation occurs. A second factor that tends to limit the current
delivered to the lamps is the internal resistance of the coil windings and the
wiring harness and this is most likely what helps the six volt system limit to
some degree. Without core saturation and with only limited current reduction
from the coil resistance, the lighting voltage vs engine RPM is close to linear.
I have experimented with a larger number of turns combined with an increased
load. More core saturation and resistance loss occurs and the voltage will
change much less from idle to wide open. There is a limit to how far you can go
because the coils will overheat from the increased current .
Using coils with 30 added turns you can connect a simple half wave
silicon diode to charge a motorcycle battery at about 200
milliamps while still maintaining acceptable lights.You can also connect
your stop light directly to the battery for increased brilliance and to avoid
dimming the headlight when you brake. The Cushman lighting system generates a
very high peak to average AC voltage and for this reason the battery will charge
on voltage peaks even though your RMS or average reading voltmeter may indicate
less than 12 volts AC is present.
When Cushman introduced the 12 volt lighting system in 1958, intentional or not,
they took full advantage of core saturation.
If you compare the core area in the 12 volt system to the older 6 volt
system you will find that it was reduced from a cross sectional area of about
.36 square inches to about .16 square inches. A direct comparison cannot be made
because, unlike the 6 volt system, the 12 volt system uses a separate coil for
the tail light. Another separate coil is provided for the brake light. The
operation of the brake and tail lamp coils are not discussed because they
operate very similar to the headlight coils. The two 12 volt headlight coils
have approximately 350 turns each. As in the 6 volt system they are in parallel.
Relating this to the older system would mean 175 turns would have been used (1/2
times 350) at six volts, which, without any core saturation, would have blown
the lamps out almost instantly at high engine RPM. As a result of core
saturation, the headlight voltage on the 12 volt system at an idle is normally
about 10 volts and as the engine RPM increases the cores go into saturation and
limit the voltage and current to a safe value for the lamps.
Some of you may have noticed that the wire on the 12 volt coil is smaller than
wire on the six volt coil. Going from 6 to to 12 volts doubles the required
number of turns while allowing the circular area of the wire to be cut in half.
In this case it means that we can go from number 15 (larger) to number 18
(smaller) wire, a decrease in cross sectional area of one half.
I have not measured the size of the wire used in the 12 volt coil because
I would have to ruin a coil to do so but it appears to be even smaller than
number 18, possibly 19. This
paragraph applies only if the coils remain connected in parallel.
Recently aftermarket vendors have introduced 12 volt retrofit coils for the old
6 volt Wico system. Like the coils for the later 12 volt system, these coils
have twice the number of turns and use wire approximately half the size of the
original 6 volt coils and they remain connected in parallel. Their output
voltage and performance will be the same as wiring the two original X8804 coils
in series to obtain 12 volts.
As a footnote, perhaps someone would like to experiment by cutting down the area
of the core in a 6 volt system and winding new coils with 150 to 175 turns and
see how it works.