ION knock detection

[garfield at pilgrimhouse.com]

On Wed, 20 May 1998 11:14:31 -0400, cosmic.ray at juno.com (Raymond C
Drouillard) wrote:

>When you design the boards, you might want to consider adding some
>optional circuitry.  For instance, I plan on taking the 0-5V output from
>EGOR and running it through a couple of op-amps to get a 0-1V output that
>spans only the enrichment levels that I'll be using.  The Holley 4Di
>computer has a table that allows the user to specify an O2 voltage set
>point for various operating conditions.  I plan on running it lean at
>cruise and rich at WOT.  The whole idea is a joke with a standard sensor,
>but should work fine with EGOR.  I'll hook the 0-5V output to a dash
>gauge just for kicks.

If you read the previous posts carefully, this whole subject was
discussed, and my conclusion at the moment is that there's NO way we
could/should include any "optional circuitry" to cover all the diverse
apps that EGOR will be used in, so we'll wanna treat EGOR as a small
assembly or "chip/module" if you will, that you mount on YOUR board
where you do whatever interfacing is peculiar to your app. There are
just too many variations on what people need to do with the raw output
to cover them all onboard EGOR. Hence the plan to make it a
small-as-possible DIP module instead. Same seems to apply for ION.

>Are you planning on using HV diodes for ION, or are you going to put a
>bias voltage at the low voltage end of the coil?

Definitely the latter, IF your IGN is gonna be coil-per-plug, cuz the
"Saab way" is certainly the most elegant of the circuitry I've seen to
date. BUT, having said that, I also wanna at least sketch out and get
working a version of the technique that can be used on dizzied IGNs, and
also crankfired/wasteFire systems. As far as I can see at this point in
my understanding, HV diodes MUST be used to do the former, and maybe
even the latter, since most/all wasteFire coils are 3-terminal devices,
meaning you don't have free access to the low side of the secondary;
you've got all this other crap dangling from it.

There are many things we don't know that will be quickly and easily
resolved once eXperimentation begins. Give ya an example or two to give
ya a flavor for what I mean by that.

The only current known working deployment is in the Saabs, right? OK,
well they use two things, coil-per-plug or "direct" IGN, AND CDI. These
two attributes have some distinctives that may affect the success of
their deployment. For example, the spark duration on inductive IGNs is
bloody HUGE compared to CDI, leaving a much wider and sharply defined
measurement window for the ionization detection stuff with CDI. Another
issue is that in Saab's (and most any) CDI setup, the HV transformers
are actually small pulse transformers, and their secondaries are MUCH
lower impedance than an inductive IGN coil. Right off the bat, I don't
SEE any issue there per se, but it's a distinct diff. that MIGHT hide
gremlins from my eyes at the moment. Like I said, won't know some of
these things until we bench it.

One thing I AM confident of is IF you wanna build a CDI system (Dick
Smith in Oz sells a pretty decent looking kit for one, #K3307 if my
notes are right), and IF you use 4-terminal per-plug coils, I see NO
hitches to getting a system just as good as Saab's. How many of these
ifs/ands/butts we can roll away and bring the tech into our more normal
world of 3-terminal coils and inductive IGN, only time will tell. The
patents say it's doable, but them guys ain't DONE it, just talked about
it, whereas Saab HAS done it. So we'll have to grope around and see if
these other guys really know their stuff like Saab obviously does.

That's part of what makes this fun; it's gonna be a grand adventure and
eXpedition. B)

Gar