Knock Question

[xxalexx at ix.netcom.com]

> From:          "M&D" <91coupe at bellsouth.net>
> To:            <diy_efi at efi332.eng.ohio-state.edu>
> Subject:       Re: Knock Question
> Date:          Sat, 14 Mar 1998 23:59:33 -0500
> Reply-to:      diy_efi at efi332.eng.ohio-state.edu

There must be over 100 papers on this subject.
All manufactures are looking at measuring  ion current of combustion
for individual cylinder a/f, pressure(knock), and misfire detection.
It seems the simplest method is to run a constant DC voltage across
spark plug after firing. A current will be generated depending on 
ions in gas.   I was looking into this last year, I will try to give 
a literature review later.  I think SGS just came out with a chip.
alex



> >Whilst re-reading my Heywood trying to figure out how a slower burning open
> >chamber is going to make more power without knocking to destruction, I ran
> >across something obscure I had forgotten about.
> >
> >Seems that when "knock" occurs, a high intensity flash is observed and it
> is
> >accompanied by a sharp increase in ionization.  He then says that optical
> >probes and ionization detectors can be used to measure this.  In fact, the
> >spark plug itself.
> >
> >Now my engine builder is a little reluctant to cut a hole through the
> >cooling jacket and put a quartz window in for an optical detector, and I am
> >rather reluctant to undertake a total redesign of my electronic ignition
> >system.  Which leaves sticking a probe into the exhaust and measuring the
> >amount of ionization.
> >
> >Heywood while referring to the measurement of ionization, leaves the
> >methodology to do so as an exercise for the student.  So what I am asking
> >for help on is any references to how to measure the ionization - SAE
> papers,
> >Reference books not leaving it as an exercise for the student etc.  Circuit
> >Ideas for modifying the ignition would work. One idea that springs to mind
> >(I am absolutely positive it has been done before by someone!) is to use a
> >separate spark plug in the exhaust as an ionization detector.
> >
> >The reason for this interest is that acoustic knock sensors do work.
> >However, to work well they must be closely tuned to the acoustics of a
> >particular engine and speed range and without an excellent processor can
> >give false reading (mainly knock when there ain't - fail-safe). If an
> >Ionization Detector could be built that resides in the exhaust or a
> modified
> >ignition, then it could be coupled to the engine management system
> >electronically and be virtually platform independent. Self tuning even.  No
> >muss fuss dyno time thingee.
> >
> >Any thoughts, info or references would be appreciated. TIA
> >
> >
> >
> 
> 
>