Hi Douglas....
fellow electronics engineer here :-)
A simple Hexfet would drive the coil just fine.
The Pertronix Ignitor (tiny...fits totally inside the dist)
is nothing more then a low ohm (<.028 ohm)
Mosfet/Hexfet transistor and a hall effect sensor
driving it.
We use to use 6-8 IRFZ40 (Mfr = International Rectifier)
hexfets in parallel in our R/C car electronic speed controllers.
With such low on state resistance, there is next to nil
power dissipation across the Hexfet. (then again
we were switching 200+ amps) you'd only need to be able to
handle around 4-6 amps to fire the hottest of ign. coils.
National Semiconductor use to make a series of
absolute and gauge pressure transducers in a variety of
scales.
...just pulled an old (1977) NS spec book off the shelf....
LX16xx and LX17xx series sensors come in a variety of
rugidized housing as well as PC board mountable hybrid
modules.
the best I found (similar use years and years ago)
was the LX1604G/LX1704G ...scaled at -15 to +15 psig
putting out 333+/-6 mV/psi
- 25 Hg/in is roughly -12.3 lbs./in so you're well within usable scaling.
Another readily avail source is almost any GM vehicle built
after say...1978 that had elec ign. or a computer. They all use
vacuum sensors.
Most are probably still under $20. I
I remember 10-20+ years ago Popular Electronics Magazine
had a whole series of articles on 'build it yourself' car electronics
back when EFI was first coming out.
My 1974 Midget...back in 1980... ended up with a complete
custom 'homemade' digital dashboard. All the gauges fit
within visual site inside the diameter of the steering wheel.
It had a
bargraph and number tach display,
speedo (w/trip) display,
switch selectable volts/amps display,
switch selectable oil temp/press display,
gas gauge that read out in 10 percent increments,
vac gauge,
switch selectable water temp/head temp.
Back then no vac sensors were within $$ range I could afford
just to play with (1976 = $250)
So I ended up building my own sensor from a stock mechanical
gauge. As the bourbon tube would collapse and curl, it would
slide a long bodied ten turn pot along its length (the end of the pot
was busted out so the screw could just slide in and out of the case)
It was good for about 1psi repeatability.
...enough ranting... aaahhh the good old days of time on your hands.....
Best of Luck in you endeavor
Paul Tegler ptegler@gouldfo.com www.teglerizer.com
----- Original Message -----
From: "Douglas Braun & Nadia Papakonstantinou" <dougnad@bellatlantic.net>
To: "Ptegler" <ptegler@gouldfo.com>; "Jeff McNeal" <jmcneal@ohms.com>;
<spitfires@autox.team.net>
Sent: Monday, February 26, 2001 12:51 PM
Subject: Re: stumble resolved
This reminds me:
I have been thinking of building a microcontroller-based electronic
ignition that would have a reasonably sophisticated table-based
algorithm for controlling spark advance based on RPM and manifold
vacuum. It would also have a control input that would let me
adjust the advance while driving, for optimizing the curves.
This would take maybe $25 in parts, and would do most of what
those fancy $300 aftermarket electronic ignition systems do.
There are a couple of open issues I need to figure out:
1: What is a good circuit for actually driving a coil?
Can I simply have the microcontroller drive a high-current,
high-voltage power transistor? Do I need some equivalent
of the condenser? Has anyone ever seen a schematic diagram
for a Crane or similar aftermarket electronic ignition?
2: Where can I find a manifold vacuum sensor that would interface
to the analog-digital convertor on the microcontroller? I'm
sure that any parts store would have something suitable on the shelf,
but where would I find the electrical specs?
If anyone has any links, references, or even answers, I'd love
to see them.
Doug Braun
'72 Spit
At 11:55 AM 2/26/01 , Ptegler wrote:
>It turned out to be a combination of which vac. advance curve,
>(where it starts, ends, and total advance supplied) which
>source of vac is being used, (ported or manifold)
>and the mechanical advance curves and engine/dist
>mech. timing.
|