Ignition only

Chris Conlon synchris at ricochet.net
Fri Nov 10 23:18:57 GMT 2000


At 04:29 PM 11/10/00 -0500, Bruce Plecan wrote:

>From: "James Ballenger" <vtjballeng at yifan.net>

>> As an fictional example:
>> Combustion side.  40,000V = 10000ohms*4amps
>> Waste side 5,000V = 1,250ohms*4amps
>> Same current passing through both of them... it is just like a
>> series circuit with the voltage source in the center as I see it.  If this
>> is wrong, please tell me why.
>
>Ya, well meaning and good but.
>Your trying to expalin things as a DCv,  and it's just not that way.
>This has to "balance" on both sides, doesn't apply.

No, he's right. The *current* in a simple series circuit, which is
a fancy word for "loop", is the same at all points. (If there is
leakage, that constitutes a 2nd, parallel circuit, and it's not
a simple series circuit anymore.) This has nothing to do with AC
or DC, it's true either way. Imagine oil pumped in a closed loop,
with oil filters, oil coolers, etc. The same *amount* of oil
(current) passes through each, the pressure drop (voltage) may
not be the same in each.

A spark plug can be modeled, pretty closely, as a resistance. It's
inductance and capacitance are low by comparison. The trick is it's
a variable resistance, also it has what's called a "negative
resistance" characteristic. It's like a little neon lamp. It's an
open circuit (roughly) when not lit, but put enough V across it
and current through it, and it becomes a low resistance device.
If you could put an ohmmeter to it when firing (somehow), you'd
see that the R has become pretty low.


>Does any one know of a on-line picture or scan of a scope pattern
>of a plug firing?.

No pics, but I had a couple ignition systems on the bench scope for a
few weeks a while back, and I'm afraid I have to totally disagree
with you about the AC/DC thing. The main part of the spark on the
systems I looked at was a single consistent polarity, in the single
digit milliseconds range. If the spark actually fired, there was
very little ringdown, and no spark during ringdown, since the first
spark heavily damped the RL oscillation on the secondary side. If
there was no spark during the cycle then ringdown was impressive
indeed... but as I said there was no spark in this condition, and
certainly not multiple alternating polarity sparks.

Further, I put inductive pickups (attached to a scope) on the
plug wires of a couple different systems. These pickups are
sensitive to the direction of current flow in the plug wire as
well as rate of change. They showed the same thing, one single
spark in the expected direction.

Maybe there is some confusion between the use of "A/C" to include
a varying signal with the same polarity (or the A/C components of
a signal) vs. a varying signal that regularly switches polarity
(i.e. 60Hz power line signal).


If you're looking at the coil primary (I had the scope on that too),
things may get a bit confusing to say the least. The ringing there
may look severe because (if the ignition driver is fully off) that
half of the circuit is now open. The voltage will go higher because
of this, but there is virtually no power being lost out that side,
and it does not really correlate directly to how the actual spark
looks.


>If it was a DC spark then one electrode would really wear.
>If DIS truely fired both plugs equally all the time they would wear out 2x
>as fast, hmm, and with the EPA tuneup reguirements that would be a definite

Well, it doesn't fire them equally, not as far as arc voltage anyway,
which someone just pointed out. plug erosion is very dependent on the
breakover voltage (spark current being equal), and not necessarily
linear, so the easy spark on the wasted side is much less erosive than
on the live side.

Also see below about platinum plugs in 1/2 of some engines.


>It's an EE thing.
>The spark is A/C so nothing is backwards

Disagree strongly. I even heard lately of some OEM DIS system that
required platinum plugs on 1 of 2 cylinders fired by each coil. Wish
I could remember now which it was... late model Ford v-6 IIRC.

>There is a 50-50 chance on which way the spark lights, and for the first

Disagree strongly. In my circuit the direction is 100.0% predictable,
and had to be because part of the setup was a spark monitoring
circuit, for which the polarity had to be correct. Also the Toyota
systems (which are part of what I was replacing) have a similar
circuit with the same requirement.


>It's the repeated colapsing on the magnetic fields that generate the current
>to maintain the firing line on a scope.

Disagree strongly. The inductance on most coils secondaries is so
high the field *cannot* ring quickly. Big inductors charge and discharge
slowly. The coil *does* ring, but if things are right it fires on the
first segment of ringdown. If the spark does jump at that time it
draws off most of the coil energy, and subsequent phases of the
ringdown do not generate a spark. This is a millisecond thing, not
a microsecond thing.




Don wrote:

> I'm having trouble understanding the concept of firing the second plug
> backwards. After the spark jumps the gap on the first plug, isn't the
> "path of least resistance" going to be head-block-chassis instead of
> trying to jump the airgap of the second plug??

No, because the spark needs to get back to the *other terminal of
the coil*, *not* to ground or Vbatt. In a typical dizzy ignition
the coil secondary is wired to Vbatt or ground, so the spark path
*is* back through the block, chassis, battery. A waste fire system
has to use floating coils (not grounded or connected to Vbatt on the
secondary side). One coil HV terminal (when firing) might be at -30kV,
relative to the ground/head/block potential voltage, the other HV
terminal might be at +15kV. This lets the engine block/chassis stay
grounded, but the spark plugs still all fire.


Anyway... I'm not trying to be negative or confuse people; I am
trying to report what I've seen and help sort things out. Obviously
I have not had every ignition system on the bench, but the few I
did have were very consistent in the details that I've reported.

   Chris C.

----------------------------------------------------------------------------
To unsubscribe from diy_efi, send "unsubscribe diy_efi" (without the quotes)
in the body of a message (not the subject) to majordomo at lists.diy-efi.org




More information about the Diy_efi mailing list