Coils for Ion

garfield at pilgrimhouse.com garfield at pilgrimhouse.com
Wed Jun 3 06:31:17 GMT 1998


On Tue, 2 Jun 1998 23:32:52 -0400, cosmic.ray at juno.com (Raymond C
Drouillard) wrote:

>I'd be glad to... Um... Where did I put that pesky inductance meter? 
>Oops, I forgot.  I don't have one!  Does that mean that I'm not allowed
>to ask questions?

EXACTLY, you can't ASK a question you're not prepared to also answer.
Hee. Geez, talk about self-defeating!

Actually, the reason for my response was it's also a tad more
complicated than JUST what you were thinking outloud about. That's why I
said try to work out some numbers, instead of just asking vague
questions. I assumed you had an idea of the inductance ranges, but even
with ANY guessed number for that, just as soon as you sat down to do
your first calculation, you'da realized THERE IS SOMETHING MISSING
STILL!! The ringing frequency in the coil's secondary, that you see on a
scope trace, is a function of the inductance of the primary/secondary
(depending on which one you wanna focus on), but ALSO, would there be
ANY ringing if it was JUST L and R? Nope. So there's a C in the equation
somewhere. Ooooops. Now you're gonna need more than an inductance meter.
You're gonna need something to measure the distributed parasitic
capacitance in the primary/secondary windings. Ouch. You're life as a
theorist just got alot harder!

Practically speaking, you don't measure the LRC components and do a calc
to get the resonant freq, you get your equipment out, bang the coil, and
MEASURE the actual resultant frequency. Cuz it ain't easy to CALCULATE,
cuz it ain't so easy to measure the distributed internal coil
capacitance.

>What is the typical inductance of a spark coil?  Is it on the order of a
>henry or so?  0.1 henries?  10 henries?

On the primary, it's like say 2-10 milliHenries, and the secondarys can
be pretty big, say between 25-50 Henries.

>What is the inupt impedance of the current sensing circuit?  What does
>the waveform look like?  I'm sure that that information is available
>somewhere on the web, but I haven't seen it.  If someone who has that
>information at their fingertips can pass it to me, I'll be glad to do
>some calculations and post the results.

If you'd only READ the previous posts, you'd have that info or something
better.

>The nagging though that the reactance of the coil might require some
>special considerations has been bugging me.  I didn't have the specific
>information I need to do the calculations, so I tossed it out to the list
>for consideration.  I believe that "brainstorming" is the term used for
>this.

Yeah, but here again, consider what you've said. Tossing out broad
questions may be well and good if that's the ONLY sources you have for
investigation, but I've repeatedly posted the source for the schematics
(the patents), there's been posted a couple times that I can recollect
well, a link to a diagram showing both real and modeled coil AND
ionization current waveforms (Lars Eriksson's & Mercel's paper,
"Interpretation of Ionization Currents), and suggesting that knowing the
inductance of the coils is gonna give you what you need to calculate
results is kinda naive. If you'd looked at that waveform, you'd be able
to see in an instant that the frequency of the coil ringing is WAY above
the components in the measured ion currents (here's the path, ONCE
AGAIN; go take a look and see what I mean:
	http://www.vehicular.isy.liu.se/Research/activity96/node5.html

And that would also tell you all you needed to know about the input
impedance of the measurement circuit, cuz it's gotta be low enough freq.
to pass them ionization waveforms, and they are slower by probly a
factor of 10 compared to the ringing, but since they don't both occur
simultaneously, as you can see from the waveform, the whole issue is a
red herring, anyway. See, so you DID have all the info you needed to
answer your question!

Brainstorm by all means, but with an emphasis on the "brain", not the
"storm", or maybe in this case the "fog". B)

There's yet another interesting phenom (well, it was to me when I heard
it) involved here, that you can't actually see in those waveforms on
that web page, but you CAN see clearly on a good scope trace of any IGN
secondary waveform. The PLASMA itself has a resonant frequency! And the
frequency is dependent solely on the gas constituents. I can only pass
along what a collegue at work who's knowledgable about these arcane
sorts of "physics" things, told me, but I gather from what he described,
that the VERY high frequency what-looks-like-noise fuzz on top of what
is usually refered to as the "arc line", the portion of the waveform
just after the HV spike has ignited the charge, if you look at it on a
scope, it will always look like it's a THICK line segment, suggesting
there's a small very high frequency component superimposed. That's
apparently the resonance of the plasma itself. Now, in none of the
patents do I see any mention of using or trying to detect the presence
of the plasma from it's resonant frequency signature, probably because
it's much easier to tell if there's a plasma at all (all you really care
about) from the conductivity of it, but I thot this was an interesting
insight into why the typical HV IGN trace looks the way it does in that
"arc line" portion. I'd always wondered why it looked so THICK there. I
guess this is why. Curious, no?

See, with that lil tidbit he shared with me, and I simply pass along to
you, you are now prepared to explain EVERY single feature of the HV IGN
trace on a scope, should you ever be asked to do so at a cocktail party
or beer bust. The peaks, the ringing, the arc line, all make perfect
sense; and now the fuzz on the arc line does too!  B)

Gar




More information about the Diy_efi mailing list