Ignition module

[Paul E. Campbell]

PATTEEUW at etcv01.eld.ford.com writes:
> >b. How does one detect knock? And what do you do about it from an ignition
> >point of view?
> 
> Knock is usually detected with a resonator type sensor.  The specific mounting
> location that will give the best results is very engine dependent.  Spark is
> usually retarded on an individual cylinder basis.

Can you list an example?

> >c. What is the optimum spark voltage? The higher the better or is there some
> >curve? How do you measure it? What are the dependent variables?
> 
> This is not my field of expertise, but I do know that the current Ford ignition
> systems generate lower voltages than what we had 15+ years ago.  That high a
> voltage just wasn't necessary.  Voltage is measured directly with special high
> voltage probes (like those for measuring CRT voltages).  The bigger the gap on
> the plug, the more voltage required to break it down.

The way that you normally measure these is either the suggestion given on the
list before (measure the voltage across a very small resistor) or by placing
a couple turns of wire around the wire you want to measure. The coil acts
like a transformer and lets you tap the signal. This is also the operating
principle of those "clamp ammeters" for power lines that have a jaw that
fits around the line.

> >d. How long should you maintain a spark? <snip>
> 
> In general, the longer the better.  The big reason is to prevent possible
> "flame out" on air/fuel mixtures with poor distribution.  There are two
> approaches: high engery (long duration) or multiple restrike.

I can build a circuit that uses the coil like an inductor to initiate the
spark, then leave it connected to the battery (with appropriate resistance)
to continue maintaining the spark for as long as I want acting as a
transformer rather than an energy storage device (high energy).

I can also use capacitors which are cheaper and often easier to work with to
generate as many sparks as you want (multiple restrike).

The duration on these is pretty much infinite. Start when I want and stop
when I want. So the question becomes WHEN to start and stop and what to
measure to determine this.

Is there somewhere that a comparison of multiple restrike vs. high energy
has been done?

I would conjecture that you could couple the coil directly to the spark plug.
Keep ramping up the charge (feeding the primary) until it goes to discharge,
then maintain the juice to the spark plug for the required amount of time at
the lower spark maintaining voltage. In this case, the spark will ignite when
there is enough potential to achieve breakdown, unless there is a leakage
problem since the dielectric constant is going to be low with the fuel/air
mixture in the chamber?

Anyone who has ever done arc welding will understand this example intuitively.

After discharge occurs, there is an ionized trail inside the chamber. Enough
turbulence will move the ions around that multiple restrike probably has
problems following the original ionization trail, so it would have to reionize
the same spark gap each time. The advantage is that the charge does not have
to be maintained over the entire ignition interval.

The only way that a high energy system would have to restrike is if the
spark went out due to the "spark maintaining voltage" being too low.

As far as how long, there should be SOME way to tell when to quit. I know
that Ford usually specifies that somewhere between about 65 degrees BTDC to
10 degrees ATDC is the right range to start (as far as the TFI module in
the EEC-IV system goes) and the amount of time to spark ranges up to 1500
milliseconds.

Now, the start time should probably occur so that the force from expanding
gasses hits right when the piston is at TDC so that it does not interfere
with the compression stroke and maximum power is achieved during the downward
stroke. The time for this to occur should be a fixed time in a static
condition, but the effect of compressing gasses (from the piston compression)
probably modifies the idea of a "fixed reaction rate". The compression will
be identical in every stroke at TDC, but the rate of compression may or may
not be a factor (is it?). Aside from calculating the TIME necessary for
expansion to become significant after ignition, there is the fact that as
the engine runs faster, this "constant" time (assuming rate of compression
is not a factor) causes the point where spark should occur to be further and
further back from TDC since TDC comes sooner. Two other factors are the fact
that the EFI module will be varying the fuel mixture, which causes the
spark ignition point to change, and the time necessary to ramp up to the
ignition voltage itself (dwell time) in a high energy system.

Once you have started the spark on time, the time to stop the spark is either
when the piston reaches the bottom of the stroke (or the exhaust valve
closes) or when the fuel has been burned either completely or as far as you're
going to get it. For the latter, it seems to me that the conductivity of the
exhaust will be significantly lower than the fuel/air mixture and that you
should be able to measure the appropriate shutoff time by measuring the
conductivity of the spark gap in the ignition module.

I may be shooting blanks here. If so, I invite someone to rip me to shreds.

> >How do you determine when to start the spark? I suppose this is related to
> >the previous question.
> 
> There are lots of papers on MBT (bean best timing) and IMEP (indicated mean
> effect pressure).  Higher IMEP = more power.

I will look for them. I guess I was really asking for a GOOD example, such as
a nice review/summary or one of those "breakthrough" type papers that reviews
everyone else's work before presenting the new technique.

> I have two small (<20 pg) catalogs from SAE containing many different books 
> and/or collections of papers.  The first is called "SAE Publication on
> Electronics" (catalog number 2884).  Lots of good books in here, such as
> "Automotive Electric/Electronics Systems", "Automotive Handbook" (both from
> Bosch), "Automotive Sensory Systems", "1995 Electronic Engine Controls" (a
> collection of 16 paper), etc.  The other catalog (2894) has more general
> automotive titles.  SAE phone number is 412-776-4970.

Most societies have multiple journals broken down by category. Usually, there
are about 10-20 major journals. I was not being clear, but I was trying to
find out if SAE has such things. I've found a couple minor ones, but I didn't
know if SAE publishes a regular journal related to say spark ignition engines
or fuel injection or combustion engines.

The nice thing about the journals is that you can pretty rapidly learn a lot
about the "traditional ways of doing things" and what books in the industry
are considered "bibles", even though the value of most of the journal articles
may be very low since research papers usually deal with obscure or special
circumstances or attempts at characterizing known qualitative observations.

Since you mentioned mean best timing and indicated mean effect pressure, I
will check the papers and books here for those key words and read up on it
(essentially doing a search on it using journals as indices).