Electrical Insanity with '87 CBR600

There are two power transistors inside the spark box, usually one PNP
and one NPN device.

They are in the ground leg of the negative side of the ignition coil
wiring.

They are a solid state version of a set of ignition points.

The transistors are turned ON when you turn the key to the RUN
position...

When one of the pulser coils sends its pulse to the spark box, it
turns one of the two transistors OFF, but the transistor turns back on
almost immediately to be ready for the next pulse.

Then the next pulser coil sends its pulse and the other transistor
turns OFF, etcetera, etcetera, ad infinitum nauseamque...

There are also some little coils and capacitors and resistors in the
spark box, arranged in what is called an RLC network. RLC stands for
resistor, coil, capacitor.

An RLC circuit slows down the passage of current.

They act as a timer to RETARD the spark. As more and more pulses come
from the pulser coils, there is less retardation of the spark, so the
spark comes earlier and that's how the so-called ignition "advance"
curve works.

Simpler electronic ignition boxes didn't even have the RLC circuit,
they had a mechanical advancer unit on the end of the crankshaft, with
springs and flyweights.

 

There are two power transistors inside the spark box, usually one PNP
and one NPN device.

<snip>

Usually the 2 channels in the ignition pack are identical so generally both
transistors are NPN (some units use low power PNP transistors in the
pre-amplifier stages) - but high current PNP transistors that can handle the
500V or so LT back EMF are a bit rare and expensive.

 

Aha. I've finally found someone who "Sees All! Knows All!! Tells All!!!"

Pray tell, kind sir, given that the "pulser coils" are solid-state:

1.) Why they fail at all?

2.) Why they fail intermittently?

Note that the yellow unit that failed back in '96 (and took a long,
long time to diagnose) *still* tests (~500 ohms) OK sometimes. Other times
it's dead as the proverbial flogging doornail.

P

"I Ain't Blind, I Just Don't Wanna See"
- the title of a tune by Little Joe Blue, maybe 1966

 

Also a few shorted turns won't make any noticeable difference to a DC
resistance test, an AC test on the other hand would give a large reduction
in measured resistance (reactance).

 

The pulser coils aren't solid state, they are just copper wires
wrapped around a core, and that technology has existed for about 200
years.

The power transistors that take the place of ignition points are the
solid state components I referred to.

They either short out between coil windings are they short out to the
the mounting plate as the extremely thin enamel or plastic insulation
fails.

The insulation is either sprayed on or the wire is dipped in it during
manufacturer.

Heat and vibration are the usual suspects.

There's a bad connection where the leads are attached or a broken wire
in the coil that occasionally touches.

 

I'm fairly sure that Suzuki used one NPN and one PNP in my GS1100,
because the Chiltons manual tells home mechanics to hook up a
flashlight battery one way to test one transistor and then to reverse
polarity to check the other transistor.

 

Then each is just a "relay" per http://en.wikipedia.org/wiki/Relay ?
There is an armature inside the nasty little sealed bugger?

Why the Hull do they call 'em "Pulse Generators"?

When (i.e. what year) did solid-state take hold for bike ignitions?

"they are just copper wires". OK. Question still stands.

What difference between such copper wires and those found in, say,
the 1920 Emerson table fan:

http://www.vintagefans.com/gallery/desk_fans/emerson/1920em12osc.html

Note that such fans, given minimal maintenance (few drops of oil),
are likely to outlive both us and our children.

Note also that there were tremendous differences between the
technology, etc available to the Emerson folk in 1920 and that in
possession of the nice, nice Nipponese fellers in the 1980's (when
my bike was designed, etc.).

...

Y'all will forgive me if I take your explanation as less than 100%
of the story. I will assume (until I have evidence to the contrary)
that "little errors" of design and/or manufacture best explain the
problem.

Cheers,
Puddin'

"I Ain't Blind, I Just Don't Wanna See"
- the title of a tune by Little Joe Blue, maybe 1966

 

There is NO moving armature inside the pulser coil.

A relay is an electromechanical device for using a smaller amount of
electrical current to control a larger amount of electrical current by
moving electrical contacts.

There are no movable electrical contacts inside the pulse generator
coil and they don't switch current on and off, they *create*
electrical current when the magnetic field around them is disturbed...

They sense a disturbance in The Force, and generate a very short, low
voltage electrical pulse, just half of a sine wave.

In order to mechanically generate electricity, you need a coil of
wire, a magnetic field, and *something* has to move.

You could attach a magnet to the end of the crankshaft and move that
magnet past the coil of wire to get a pulse, but the magnet would
probably get weaker and weaker from being bolted to the steel shaft.

So the designers bolt a four-toothed non-magnetized steel rotor called
a reluctor to the end of the crankshaft and spin it past the coil,
which is wrapped around a magnet or mounted in close proximity to a
magnet.

As the steel reluctor passes the magnet, the magnetic lines of force
go through the reluctor and are disturbed momentarily and a pulse of
electricity is generated as the magnetic lines of force move in
relation to the coil of wire.

About 1980.

The first copper wire was formed in the 13th century when two Scots
found a penny at the same time and neither would let go of it...

Copper is copper, but if Nippon Denso or Kogyo or whoever made your
pulser coils wanted to improve them, they would use a better
insulation material.

http://en.wikipedia.org/wiki/Magnet_wire

"Older insulation materials consisted of cotton, paper, or silk and
were useful for lower temperature applications (up to 105C). Modern
magnet wire typically uses one to three layers of polymer film
insulation, often of two different compositions, to provide a tough,
continuous insulating layer. Magnet wire insulating films use (in
order of increasing temperature range) polyurethane, polyamide,
polyester, polyester-polyimide, polyamide-polyimide (or amide-imide),
and polyimide. Polyimide insulated magnet wire is capable of operation
at up to 250C."

 

I believe the pickups (pulse gen) are only available as part of the stator
from Honda. I believe this is also the case with the stator I suggested. If
you have had 2 failures and the components are as poorly made as it seems,
isn't it a fools errand to install a failure prone piece that you know has a
lot of miles/ hrs on it? How many miles are on your bike? About the same
mileage between failures? Has it been overheated? Does it get all
recommended oil changes? Just some random thoughts.
Hank

 

As I originally thought.

An electrical device, designed to be an integral part of a larger
(ignition) system which is "solid-state", the device, a sealed unit,
having no electro-mechanical components. And you maintain that the
little buggers are not "solid-state"? You must be a transistor-guy.

And it's OK they should have a material failure rate. Last about as
long as 2-3 good batteries. Eh?

Interesting. Compliments on the description.

What are they using for this function in newer/newest bikes? Obviously
I don't keep up with such stuff. Other fishes to fry ...

Does this mean you admit to differential characteristics of electrical
devices designed/mnfgrd 2 decades ago when compared to those from
centuries and centuries ago?

The little buggars have this mutant-stylized emblem on them that looks
like "TEI" or "TEL", but I couldn't swear to it.

Prost,
Puddin'

"I Ain't Blind, I Just Don't Wanna See"
- the title of a tune by Little Joe Blue, maybe 1966

 

Some after market units for upgrading from points used optical or Hall
effect sensors, the optical types were less successful because dirt or oil
contamination could block the light beam, the Hall sensors are a special
semiconductor in which the path of current flow is diverted by a magnetic
field and are immune to dirty environment.

 

Newark Electronics sells a replacement sensor,
Newark part # 96f1986, Honeywell part # 2AV54
for about $14 USD + shipping. Seems to match
most euro Bosch systems.

I'm attempting to rehabilitate a junk BMW unit to include
two separate sensors, the second being a plug-in spare.
Got everything ready to go except for locating the stupid
male plug to mate with the stock harness.

 

Transistorization of the electronics industry started about 1960, when
I was attending Air Force tech school. The instructor was raving about
"poles" and "holes" in the electron orbits of certain semiconductor
crystals, and that sounded weird...

I never wanted to get very deep into solid state electronics, but I do
confess to having breadboarded a solid state 80-meter amateur radio
receiver with hand wound toroid coils and transistors on perf board
during a manic period...

Solid-state physics

Solid-state physics, the largest branch of condensed matter physics,
is the study of rigid matter, or solids. The bulk of solid-state
physics theory and research is focused on crystals, largely because
the periodicity of atoms in a crystal — its defining characteristic —
facilitates mathematical modeling, and also because crystalline
materials often have electrical, magnetic, optical, or mechanical
properties that can be exploited for engineering purposes.

More at http://en.wikipedia.org/wiki/Solid-state_physics

Solid state (electronics)

For other uses, see Solid state.
Solid-state electronic components, devices, and systems are based
entirely on the semiconductor, such as transistors, microprocessor
chips, and the bubble memory. In solid-state components, there is no
use of the electrical properties of a vacuum and no mechanical action,
no moving parts, although a considerable amount of electromagnetic and
quantum-mechanical action takes place within. The expression became
prevalent in the 1950s and the 1960s, during the transition from
vacuum tube technology to semiconductor diodes and transistors. More
recently, the integrated circuit (IC), the light-emitting diode (LED),
and the liquid-crystal display (LCD) have evolved as further examples
of solid-state devices.

More at http://en.wikipedia.org/wiki/Solid_state_(electronics)

I have never had a pulser coil failure on any vehicle that I have ever
owned. I did have two automobile distributor mounted ignition modules
*gradually* stop working, but I suspect the problem was environmental,
since both vehicles had
cooling system problems that released hot water vapor under the hood.

 

Actually, the Hall sender that I've looked at is an integrated
magnet and chip with a slot between them. What spins is
a slotted aluminum cup. The slotted sides of the cup pass
between the magnet and sender, intermittently blocking
the magnetic field.

 

I spoke in error. It is in fact ferromagnetic.

 

AFAIK the hall sensors require a mechanical advancer, pulser coils can be
loaded with shunt filter networks to produce the advance/retard curve
electronically.

 

That would be the case with an ECU containing a EPROM full of digital
advance curves - and it'd still use some kind of sensor to determine speed
and position of the crankshaft, IIRC this thread was about the old type with
a pulser coil driving a transistor ignition box.

Although I wouldn't be surprised if some manufacturers had come up with some
clever trick with variable reluctors for hall sensors that did away with
mechanical advancers.