I just ran across this data on the 2nd gen RX7 mass airflow sensor.
 Bosch wrote it up as part of the L jetronic injection system manual.
 I suspect it is made by Denso under license and installed
 in 1986 to 1991 RX7's. I don't think it is a computer.
 I think it is similar to a 555 system. All hardware and no
 software.

 Perry Mick has been flying this system for years in his Long EZ
 with hundreds of hours.

 It is very reliable. He has tricked the ECU into thinking it is
 still in car. He has even installed an RX7 instrument panel
 in the cockpit.

 There are probably over 100,000
 of these systems out there. They are real cheap I suspect.

 Paul Lamar

 Paul,
 This may not be doable for Aviation and I have no doubt that someone will
 tell me why, however all ideas should be explored if only to eliminate
 them
 IMHO. Now some motorcycle carbs have floating slides whereby the density
 of
 the incoming air varies the slide compared to the throttle position - for
 altitude compensation. I can't remember exactly how this is accomplished
 by
 lifting the slide with the incoming air. There is a sloped face which is
 supposed to have a critical angle to accomplish this task. It may be
there
 is a calibrated spring that air density is pushing against.

 This is the simplest solution air density adjustment I have heard about.
 For your consideration.
 George ( down under)

 George,
 My motorcyle has these. Tuned for sea level, it runs rich when going over
 the Sierra's.
 If you tune it for Denver, pings like crazy at sea level.....
 The slide is lifted pulling the mixture needle out of the jet by the low
 pressure area created by the air flowing by an orfice in the bottom of
the
 piston.
 The top of the slide has a larger piston (or diaphram) and the vac on the
 top side pulls the slide up. But they are not density sensitive.
 Sorry.....
 Larry

 Larry,
 I disagree, there is a degree of self adjustment when crusing. There will
 obviously be a POWER loss as I'm sure you are aware. There are limits,
 because the diaphragm is frequently over sized so the slide will respond
 quickly, many sudden throttle openings, (like driving in traffic), will
 show rich on the plugs. Steady state crusing will lean after you reach
 steady state. With respect to aircraft, the carbs will work better than
 the
 motorcycle use since throttle openings don't change very quickly. For
 leaning that is

 Bill Jepson

Thanks Bill, Larry, Matt & Vlad.
There certainly are some interesting points, food for thought. Not as
simple
as I first thought.
Never heard about lowering the jet instead of raising the needle Matt, my
initial thoughts are that it may raise some sealing issues as the fuel
goes
through the jet - probably some O-rings involved. My idea of raising the
needle may well need some O-rings as well
I guess I will have to keep thinking.
George (down under)

Bill,
you are correct in there is some degree of self adjusting, being a rung up
the ladder from the typical booster venturi carb design.
hot wire mass air is just further up the ladder.
Larry

Larry, Paul, and group,
I think there are distinct ways to go on the Rotary. Simple: Carburetor
induction and simple ignition, even coil and points will work. Moderate
technology: Keep as many of the systems mechanical as possible. Bendix or
airflow technology mechanical injection. Proven reliable electronicignition.
Lastly full modern system: EMS or at leastfull electronicFI. I do agree
withPaul that the modern MAF sensor is the way to go. These are very
reliable devices and are not as pulse sensitive as the map sensor. For the
rotary the EMS should be capable of being a "single lever" system with
complete altitude compensation but for best results LOP operation should be
possible.

The modern system is where we have been headed, with some diversions. The
rotary really begs for LOP operation to improve BSFC. The real problem that
we have is that often people expect a system to be "bolt on" and that is a
reasonable thing to want. The problem is that there are so may variables on
amateur-built aircraft that it is almost impossible for an EMS to be
programmed properly "out of the box". I read a very interesting article
written by one of the recent Corvette engineers about all-electronic engine
management systems. His comment was that to make one work (an EMS) took
every bit as much tuning as a carburetor and mechanical ignition system, and
sometimes more. It was a very honest evaluation, with the insight of a
manufacturer to back it up. The comment he made was that every time you
change induction, exhaust orengine internals a new complete engine map has
to be made. The system will get the engine close to begin with and is
somewhat self adjusting but fine tuning needs to be done for the particular
combination of induction exhaust and engine timing. (cams and valves in a
piston engine, porting and compression in the rotary)

My point is that the
guys that buy Tracy's system shouldn't be supprised that some tuning skills
will be needed and the tuning will take some time. This is going to be true
OF ANY SYSTEM from the simplest to the most complex. An EMS gives you more
tools, and a degree of self-adjustment but you will still need to know what
you are doing, there is no "silver bullet" system. When Paul was working on
the dyno He didn't have the time, (read $$$), to tune Tracy's system and
changed over and ran a carb system that the Mazdatrix had already spent the
time tuning. A wise choice considering the cost of Dyno time. It seems like
we cycle through the desire to keep everything a simple as possible to a
full EMS and back. Any system used will take time to get exactly right. For
what it's worth.
Bill Jepson


Paul,

I apologize if my reply is hashed up..  I haven't got gmail fixed yet,
but can't resist replying..  :)

Not arguing with Bill..  I agree with his points in general.  But:

The airplane operating requirements are relatively simple compared to
that of a performance car like a Corvette.  In the Corvette, they want
good throttle response at all RPM, and good fuel economy with very
minimal driver involvement - one lever.  On the other hand, the
airplane really has only a couple of requirements..  It can't quit.
It can't burn too much fuel in cruise.  It has to make the peak
horsepower afforded by the mechanical limits of the engine.  That's
not too bad by comparison.  The airplane engine doesn't have to idle
all that well as long as it doesn't actually quit.  It can have a
separate lever for controlling power vs. economy.  All of this makes
the tunability of the airplane easier to accomplish by comparison.

As far as running the engine on the dyno, I think that was a p-port,
and it was being run with the manifold pressure tap located at the
wrong place on the engine for good readings.  It would be interesting
to hop on the dyno with the mp tap installed in the housing and see
how the tuning goes.


Regards,

Matt-

Looks fixed to me. This looks fine.

Paul Lamar