Plasma Jet Ignition

[Raymond C Drouillard]

>Conventional coil type ignitions deliver 30 to 50 MJ, but a plasma jet
>typically requires 1 joule or more stored energy and this is released in
 < 20
>microseconds.   

At one joule per zap, we're talking about feeding 400 watts into the
plasma plugs of a V8 running at 6000 RPM.

6000 Rev/min * (1 min/60 sec) * 4 sparks/rev * 1 joule/spark = 400 W
(or about 25 W / cylinder)

How efficient can we make the circuit that charges the capaciter?  Even
if it's VERY efficient, it's going to suck down a significent fraction of
the output of a typical alternater.  Of course, it'll only do that at
high RPMs.

I think it would be best to design it so that each cylinder has its own
capaciter, charger, etc.  I wouldn't try to feed all of this energy
through the dizzy gap.

Lesee... we'll assume that we're going to charge the cap to 200 V.  E =
.5 * C * V^2 (if I remember correctly <:-)  )  That gives use a
capacitence of 50 microfarads.  A flash capaciter ought to do the trick.

How fast do we want to rev the engine?  Let's say, for example, 12,000
RPM (and we can downgrade it for more ordinary applications).  That's one
zap every 0.01 sec.  If we want to charge the capaciter in eight
milliseconds with one joule of energy, we'll need a converter that can
feed the capaciter at an average rate of about 125 Watts.  Remember,
though, that the output of this circuit will be high current/low voltage
at first.  The voltage will go up and the current will go down as the
capaciter charges.


I'm not trying to throw bricks at the idea.  I would like to try it.  I
just wanted to rough out some design parameters so that we'll know what
we're dealing with.  For one thing, we need to consider how much spare
capacity is in the vehicle's alternater.  This thing might end up sucking
down 30 A or so at cruising RPM.  Will the increase in BSHP make up for
the increased drag on the engine?

We essentially need an ordinary ignition system coupled with the
circuitry from a high speed xenon strobe.  Anyone wanna try taking apart
a high-speed timing light?  (JK)

Maybe we can feed the arc current through the secondary.  The main
problems will be the inductance of the coil and the coil's ability to
handle the high current.

Running it through an HV diode stack will also cause trouble.  The
forward breakdown voltage of an HV stack is in the neighborhood of 100 V.
 That'll dissapate a lot of the power that we want to go to the arc.

A more practical solution might be to have seperate HV and LV electrodes
that are very close to each other.  This could be tricky.  How can we
design the electrodes so that the LV circuitry won't absorb the energy
from the HV trigger.  I'll have to ruminate on this one for a while.

Ray Drouillard


___________________________________________________________________
You don't need to buy Internet access to use free Internet e-mail.
Get completely free e-mail from Juno at http://www.juno.com/getjuno.html
or call Juno at (800) 654-JUNO [654-5866]