Paul,  the following statement was posted on the
       Tailwind discussion
       group.  Do you think it is true and how would you
       explain such
       performance?   Jerry

       Just found this: lsecorp.com <http://lsecorp.com>, then click on
       news on
       left side. Klaus
       Savier flew the Vari-eze nonstop from Santa Paula,
       CA to Panama City,
       FL, 1985 statute miles in 8 hrs 58 min on 25.8
       gallons. Slightly under
       80 miles per gallon.

       Its true. Klaus had a strong tail wind. As I recall
       he also had oxygen :)
       He leans his engine severely and advances the
       ignition. He also uses
       a high energy multi-spark capacitor discharge
       ignition system to
       fire the lean mixture. Another factor is the
       aircraft is highly
       refined aerodynamically. Yet another factor is he
       uses solid state fuel
       injection but it is not computer controlled. Cheap
       analog circuits (cost $1 each)
       called 555's or something similar are used to change
       the pulse width
       on the electric injectors. Each cylinder gets the
       same amount of fuel. He may be
       able to adjust the fuel on each cylinder
       individually according to the EGT. You have
       plenty of time up there to diddle around.


       Paul Lamar ...No rotor no motor.



       Paul,
       How do the chips know how much fuel to deliver does
       it work on RPM or throttle position or something
       like that?
       George ( down under)

       I am not exactly sure about all the details but with
       a fixed pitch prop you could probably
       get away with RPM alone. You only really have two
       RPM points where you need precise
       mixtures, climb and cruise. You could manually
       adjust for those two conditions.
       Paul Lipps probably designed the system. The
       attached schematic diagram for a rotary
       uses a throttle pot. I have attached another 3D
       showing a way to make a throttle pot device.
       Freescale manifold pressure sensors could also be
       used.

       Paul Lamar ...No rotor no motor.

       Paul,

       You did it again!  (I am referring to a correction
       sent about 6 months
       ago.)

       Check your 555 graphic in above email.

       Don't put TWO tabs on your "timing wheel" - use just
       ONE.  The ONE tab
       needs to rotate around (at e-shaft speed) to trigger
       the two separate
       injector circuits 180* apart.

       As shown in your graphic, you will fire both
       injector sets
       simultaneously (for rotor 1 and rotor 2) twice per
       revolution.

       Normally I would just lurk, but the Wankel confuses
       enough people
       already.  (It shouldn't - it's simpler than a piston
       engine.)

       FWIW.

       -Vince Orton

       True. Sorry. Is that your diagram or mine Vince?
       Well I had it lying around so I used it. The subject
       only comes up once a
       year or so. I probably never bothered to redraw the
       diagram.

       How's this?

       There is another mistake in that two injectors are
       fired for each rotor
       while only one power transistor is used. I would use
       one power transistor for
       each of four injectors instead.


   ----------------------------------------------------------------------


       Here is some more info on the little jewels.

       --
       Paul Lamar ...No rotor no motor.

       And for a negative opinion:

       ------------------------------------------------
       Subject: Super simple 555 based fuel injection
       system
       From: Rotary Engine <rotaryeng@earthlink.net
       <mailto:rotaryeng@earthlink.net>>
       Date: Tue, 02 May 2006 16:14:19


           I have actually built an electronic fuel
       injection system. It is a lot
           more complicated than this.

           You have to change the duty cycle with manifold
       pressure. The engine
           won't run unless you do this, at least in a
       simple fashion.

           You have to change the duty cycle with RPM.
       Again, the engine won't run
           unless you do this at least in a simple
       fashion.

           The throttle position alone isn't enough to set
       the duty cycle. The
           engine won't run.

           In theory, you could sense the intake air mass
       flow and set the duty
           cycle based on that alone. (Manifold pressure
       and RPM together allow you
           to estimate mass flow.)

           I think you would be better off with a
       hand-controlled needle valve than
           you would be with a 555 based EFI.

           Bill Dube'

       ---------------------------------------------------

       Bill overlooked the fact that the pulse rate is
       directional proportional
       to RPM as there is a shaft sensor that triggers the
       555.
       He may have a point about throttle position. Perhaps
       it is easier
       and cheaper to use manifold pressure to change the
       pulse width.
       I'll look into to it and see if I can come up with a
       circuit that works.
       I don't want to use mass air flow alone as that
       implies mass airflow sensors
       that restrict the intake flow.

       --
       Paul Lamar ...No rotor no motor.



   paul,

   i'm no fuel injection expert, but a reference i have
   calls your system an "alpha-n" system.

   gary in midland
   --------------------------------------------------
   page 9, fuel injection basics, petersen publishing

   "alpha-n

   "there is one other efi control system that is
   generally used only in racing called an alpha-n
   system. this control system's major inputs are
   throttle position and rpm. this system was developed
   because race engines often operate at idle and part
   throttle with very little manifold vacuum. this makes
   using a map sensor difficult. this system is less
   precise than a speed density or maf and is therefore
   generally only found in racing or heavily modified
   street engines with big camshafts. a map sensor can
   still be used with alpha-n, but it is generally
   employed as a barometric pressure sensor to detect
   altitude changes."


   The mass airflow sensor is  probably the easiest way to control
   a 555 system. You do need the e-shaft trigger but the mass air flow
   sensor can control the pulse width. It is possible to have high
   RPM with low fuel flow if there is no load on the engine however you
   don't see that in aircraft engines with fixed pitch props.
   Load is probably proportional to RPM squared.
   Just about everybody has one of these Mazda mass airflow sensors
   lying around.

   --
   Paul Lamar ...No rotor no motor.



   Paul,
   A better mass air sensor to use is the unit found in ford V8 trucks
   from 97
   and up. Many of them in salvage yards. They were for engines of 220
   to 260
   hp putting them in the flow range we need. Also very light and reliable.
   Just don't get one from the supercharged trucks, as they are 90mm diam.
   Larry

   Thanks for the tip Larry. Are they mechanical like the Mazda or
   electronic?
   The electronic are harder to use as they are tougher to interface with
   the 555. The Mazda unit is just a trap door connected to a pot.

   --
   Paul Lamar ...No rotor no motor.


I believe Ford uses "heated wire" sensors.  Air flowing over the heated
wire cools it, and hence changes its resistance and hence voltage drop.
I believe the system has to compensate for inlet temperature for it to
work properly.

Does the Mazda unit depend on the mass of the trap door.  Or a spring?
If it's the mass, then it's kinda like the old K-Jetronic Bosch
injection used on lots of European cars - some of those are almost
completely mechanical.  A plate attached to the fuel distributor piston
rides on air rising up through in intake chamber.  A bit bulky, but
works great - though the throttle response is a little doggy because
there's no throttle position/accel circuit.  It appears that the shape
of the chamber controls the mixture curve to some extent.  The ones I
have seen are a cut off conic section.  At idle the piston sits in the
bottom - where the small end of the cone was cut off.

I think a similar system could be designed that was completely
mechanical - including the engine driven pump..  The Bocsh systems run a
high pressure electric pump - though with all-mechanical injectors
(orifice).  Except for the bulk, it would probably be possible to adapt
the Bosch to a 13b airplane engine.  I don't know if there's a unit
capable of supporting 250hp, but if not, you could use two side-side
units...  As I said, the downside of these systems is they require 1g
gravity to work precisely.  Low to negative g's goes rich (the plate
rises disproportionately to air flow).  High positive g's goes lean.
Maybe the system could be slightly redesigned to use spring pressure -
is that what Mazda does?  I like the idea of _all_ mechanical (even
though I am an Electrical Engineer)..

This page has some decent diagrams that show how the system works.

http://www.auto-solve.com/mech_inj.htm


Matt-

You are right about the Ford units. I had one in my Cougar and I have one
out of a Ford econo car.

The Mazda units use a spring to keep the door closed.

I think something like that could be made to work.
I have three of the Mazda units lying around however so I am going to build
up a 555 system using one of those to test on the RX8 p-port engine. It looks
like close to the cheapest and simplest  EFI system yet.