Using PC HW (& Ignition timing reference points)

[tom cloud]

Y'know, I've been reading with the greatest of anticipatory delight
the posts on various efi schemes.  But ... one thing keeps coming to
my muddled mind (I've got old-timers disease).

One of the 'cute' things about the original Apple that's sorta found
its way down to the MAC is the mentality to save a few bucks by makin'
the processor do all the work (used to have to scan the keyboard,
be the disk controller, the video I/O, etc.)  Saved money, didn't have
to make any hardware changes, so was faster (as long as the processor
could do it).  Problem is, that it has disadvantages also.  Processor
is busier.  Interrupt service routines become more impotent (no, not
a slip'a da tongue).  Timing becomes critical and an Al-Gore-ythm error
can cause missed events.

Back when I was a young tech (before monolithic op-amps), one watched
their pennies.  You didn't use too many transistors or zeners, etc as
they cost money.  Then along came the 709 op-amp.  Quickly followed
better versions and then the 741.  One day I criticized a design that
used an op-amp where a simple transistor would do, only to realize that
the op-amp was less expensive and worked better.

Pleez: what'm I trying to say?

Am I mistaken, or do many of the posts here seem to want to make the cpu
do everything?  Take an A/D and read a noisy signal and let the software
filter it (when a simple integrating lo-pass filter would save so much
trouble -- and add, maybe, 32 cents to the cost).  Or, try to do all
the ignition timing via internal counters/timers and interrupts, when
maybe a little PLL in the front end could process some of that and make
the cpu / software task lots simpler.

For example, I have a friend (who is a brilliant engineer), that designed
a speed density system using blowers for racing jet skis.  He didn't use
a computer at all, but combined analog and digital techniques.  He used
a programmable up/down counter driven by a clock to produce the injector
pulse.  The counter was loaded with a number that corresponded to how
long the injector stayed open, and was triggered by RPM.  If that
were used by a computer controlled system, that number would only be
changed when the injector timing needed changing (and some filtering
could spare lots of unnecessary changes due to slight variations).

A similar system could be devised for ignition timing.  Let a PLL take
a few timing points from the crank or cam and produce any degree (pun)
of resolution desired.  Load a counter with the advance desired and let
that peripheral circuit control the timing.  Only change the advance
when needed.

Now, understand, I really haven't thought about perzactly how to do all
these things.  Just wondering if anyone else has.  It is my experience
that being too much the purist (i.e. a hacker / fanatic trying to make
the confuser doo it all) puts you way up on the diminishing returns curve,
when the most expeditious (isn't that a new Ford product?) approach would
seem to be to use any and all tricks to get what we want.  [Now, if what
you want is to play wid da cornfuser, so be it.  As I've said before, I
wanna D-R-I-V-E !]


Tom Cloud <cloud at hagar.ph.utexas>