Date: Tue, 17 Oct 2000 00:33:52 -0400
Reply-To: David Beierl <dbeierl@ATTGLOBAL.NET>
Sender: Vanagon Mailing List <vanagon@gerry.vanagon.com>
From: David Beierl <dbeierl@ATTGLOBAL.NET>
Subject: Re: Coil Fires, 7.5KOhms
In-Reply-To: <001a01c037e8$5ef79c80$14a6fea9@900b>
Content-Type: text/plain; charset="us-ascii"; format=flowed
At 23:14 10/16/2000, pensioner wrote:
>The coil works as a dynamic resistance or impedance if you like. The
>secondary DC resistance of 7.5K has virtually no effect on the primary
>current draw under steady state (DC) operation.
I think it's safe to say "no effect at all" under those
circumstances...however that's now how this circuit operates, as you
obviously know.
>The basics are: Run current through the primary until the core is
>"saturated" then abruptly stop the current flow in the primary.
<snip>
I posted a lengthier explanation of how this works in practice back in May
-- I'll append it in case someone wants a refresher.
> From Kettering..."the only parts with an infinite service life are those
>which are never installed" Kettering invented this type of ignition system
>and it bears his name.
I think you could argue that the modern electronic ignition such as used on
waterboxers is no longer really a Kettering system, in that it no longer
uses a points+capacitor system to generate the primary coil pulse. In the
modern systems the distributor generates timing info only which is passed
to an active electronic circuit that generates a pulse shaped to suit the
designer of the system.
>So a primary fire cound be a DC issue with a short to ground causing the
>points when closed to fuse and direct battery current through the primary of
>the coil.
Sorry, no points -- the waterboxer uses a Hall-effect detector with a
castellated encoder wheel excited by a permanent magnet on the opposite
side of the distributor...the output drive transistor in the ignition
module could perhaps fuse, but the ignition module wouldn't work after that.
>been my experience that you can short the primary to ground with 12VDC
>applied without inflammatory results. But hey, I'm just a dwEEb.
Most point-type systems use a series resistor in the coil circuit (they
used to fail regularly in Chrysler vehicles) which is switched out of
circuit during cranking. Using such a system without the resistor leads to
rapid coil failure. Not sure what rapid is -- a few weeks? One long drive
in hot weather? In any case, the situation of an overheated coil which
oozes tar before failing or catching fire is well known, and I'm quite sure
it results from turns shorting together in either primary or secondary
winding. I consider the primary a likely candidate, but wouldn't rule out
the secondary on the basis of what I know (or think I do).
>A better answer might be a primary feed wire grounding without going through
>the primary coil. This would seriously smoke the insulation on said wire
>and maybe with enough inflammable grunge on the wiring cause a "fire".
It would do worse than that -- it would instantly smoke the harness all the
way back to the ignition switch. The coil feed comes straight from the 15
circuit in the fuse/relay panel, driven by the 15 terminal on the ignition
switch. And if a fire resulted I don't think it would need quotes around it...
>Donning my nomex and awaiting the flames ....
No flames...welcome to Vanagonland, where engine fires are taken *very*
seriously.
david
(note from May quoted below)
At 22:07 5/13/2000, Tom and Dana Cates wrote:
>Well, I think that the Pertronics must not be just a switch off/on. I think
>that as the leading edge of the magnetic field encounters whatever the
>switching device is, they 'Open' just like points and when the trailing edge
>of the field exits whatever the switching device is, they 'Close', just like
>a set of points.
mmm...not really. The rotor/sensor (magnetic, optical, whatever) gives
timing info to the control module, but it doesn't actually impose any
particular behavior. It's entirely up to the module what it does --
whether the output, for example, has a variable delay from the input pulse
depending on rpm, or whether the output duration varies ditto.
>So, the dwell would still be a factor when retaining the rest of the stock,
>point type ignition. I think.
The only thing left of the stock ignition would be the coil...it's
certainly possible that a different coil would be optimum.
If you don't care about theory, stop here... :)
I don't know exactly how the Pertronix or other solid-state ignition module
generates the coil pulse. In a point-type ignition it's quite simple, yet
also quite subtle: The ignition coil is simply a step-up transformer. If
you apply a small ac voltage to the low-voltage side you'll get a large ac
voltage from the other side, and vice versa. However, a transformer works
because the varying magnetic field generated by the AC input
"primary" cuts across the output "secondary" windings and induces a tiny
voltage in each one (the voltages add together -- in fact the ratio of
input to output voltage is the ratio of primary turns to secondary
turns. The current ratio is the inverse of the voltage ratio -- you're
trading one against the other just as with a block and tackle or a gear
train. The total energy remains the same). All this means that if you
apply a DC voltage to the primary, you'll get no output as long as the
voltage is steady.
In the Kettering ignition system you have an apparent conflict -- you need
a changing voltage to make the coil work, but you have only DC to work
with. Also, you need many thousands of volts at the output, but you have
only 12, or maybe even six at the input. Charles Kettering, the smart SOB,
came up with a wonderful answer (in between inventing starters and sundry
such odds and ends...).
He took advantage of another property of the transformer -- each winding is
an inductor, a fundamental electronic component with effect of resisting
changes in the current passing through it. In mechanical terms, it acts
like a large mass. So he hooked it up in series with the battery and a set
of contact points -- and he hung a capacitor (acts like an electrical
spring) across the points. Here's how it works:
When the points close, current starts flowing in the circuit, but because
of the inertial effect of the coil primary, it rises comparatively slowly,
and stabilizes at some maximum value. Then, at the instant when the spark
is needed, the points open. Suddenly the current has nowhere to go except
the capacitor, and with the inertia from the coil behind it, it very
rapidly charges the capacitor, storing energy in the "spring." The
current drops extremely rapidly, and since the power in the system is the
product of voltage and current, as the current drops, so does the voltage
rise -- up to let's say 100 volts. Also as the current drops, the magnetic
field in the primary coil collapses and the windings of the secondary,
which absorbs some of the energy and turns it into a spark.
Now the capacitor starts to discharge and shove a current in the other
direction, -- the coil generates a spark as the current increases, and
again as it decreases. This continues, with the current bouncing back and
forth, charging the capacitor in opposite directions, until all the energy
is absorbed by the spark generation, and by resistance in the primary
circuit. And a long time later, the points close again and build up energy
for the next round.
If the capacitor is missing, two problems result -- first is that the spark
is of very short duration and may not ignite as well. The second is that
when the points open the inertia of the current forces the voltage to rise
until it generates an arc between the points. This arc carries the current
by actually detaching metal from one point and depositing it on the other
one, rapidly destroying the points.
Now this all worked remarkably well. It was light-years better than
anything else around at the time, and in fact seems to be about as good as
you can get using strictly passive components. It survived with no
competition until the '60s, when various transistor circuits were developed
to more actively control the primary pulse. But it has a major problem,
which is that the faster you spin the engine, the less time is available to
load up the coil with energy for the next set of sparks. And if you jack
up the current so it works well at high speeds, you tend to melt the coil
at low speeds. Big compromise here, and the ultimate answer has been to
shift to active electronics to get control over the pulse shape, duration
etc., to get proper ignition at high speed without destroying the system at
low speed. And once you've done that, there's no need for a set of
current-carrying points, all they're doing now is generating a timing
signal. So why not dump them entirely and use a non-contact system that
doesn't wear? And here we are.
Now, children, be sure to put on your rubbers before you go home, as the
dew has fallen. And say goodnight to old Uncle Wiggily.
david
David Beierl - Providence, RI
http://pws.prserv.net/synergy/Vanagon/
'84 Westy "Dutiful Passage"
'85 GL "Poor Relation"