Paul,

   An earlier post indicated that both mass air flow and rpm were needed to
   determine the timer "on" duration.  It also appears that way to me.
   Here's my
    thinking:

   The timer "on" time determines the fuel flow per revolution, which
   needs
   to be proportional to the quantity of air per revolution.

   But the mass air flow sensor provides the quantity of air per minute.

   To determine the quantity of air per revolution, we need to divide the
   quantity of air per minute by the revolutions per minute.

   Would a PIC chip be a simpler way to achieve this calculation than an op
   amp circuit?

   Philip

   Not quite that way. The injector fires once per rev. With a fixed pulse width
   (say 2 ms) that makes the amount fuel injected directly proportional to RPM.
   Now it depends on how much air is entering the engine at any RPM. That depends
   on throttle opening. The engine can rev to 9000 RPM in the RX8 car if the
   engine
   is not in gear (no load) with very little throttle opening. However the
   aircraft
   prop load is increasing with RPM so the engine is going to
    need more and more
   fuel per revolution as RPM increases. It may need more than 2 ms worth of fuel.

   The mass airflow sensor now comes into play. At a fixed RPM the mass air flow
   sensor measures the amount of air entering the engine and adjust the pulse
   width
   and hence the mixture up or down by controlling the 555 through pin 5 which is
   what pin 5 is designed for. Probably up.  This is called "trim" in
   EFI and OBD
   II parlance.

   It might work on the mass airflow sensor signal alone pulsing the injector at
   a constant rate but that would not be "timed" injection.

   Here is an OBD II mixture chart for an actual acceleration run on an RX8.
   "Lamda" is shown  in percentages of 14.7:1 air fuel ratio. In other
   words Lambda
   1 (one) is 14.7:1. At 67 MPH and 6359 RPM Lambda is .76.  So .76 is full rich
   for acceleration. .76 times 14.7 is 11.71:1 air fuel. This is why you want to
   keep your foot out of
    the throttle in an RX8 :) The throttle position and load
   value are also shown but this must be taken with a grain of salt as the PC OBD
   II monitoring computer is not really fast enough to handle rapidly changing
   engine conditions.

   We would set up the 555 pulse width in an aircraft engine to be rich all the
   time for engine acceleration purposes. No need for an accelerator pump. Many
   aircraft engine carbs don't have them. Once you reach cruise you lean the
   mixture to as little as 18:1 for best BSFC.

   No need for a $1 pic chip with arcane and complex software  however it will
   surely work.

   Paul Lamar


I agree with your thinking - as long as we are dealing with a fixed
rpm.  When the airflow drops to half at a fixed rpm the on-time can be
cut to half to maintain a lambda of 1.0.

But I'm running a fixed-pitch prop on my Tailwind and the rpm drops to
80% on climb out but I'd still need the the max on-time fuel per
revolution.  This behavior is common to other aircraft in this
performance category, such as the Van's RV series.

I find I do quite a bit of leisurely flying at reduced power settings -
for example 80% rpm and about 50% MAP.  In this case, the airflow would
be about 40%.  If the airflow sensor cut the fuel to 40% the engine
would be running quite lean, with a lambda of about 1.25.

As I see it, the fuel on-time calculation becomes more challenging when
I'm in the pattern setting up for landing.  I cut the rpm to about 50%
and the air flow becomes quite small.  The power is greater than idle
and I want the engine to be very controllable and very responsive.

I'm doubtful that there's a simple air flow vs. on-time relationship
that will address these different operating conditions.  I hope there is
because I much prefer simple control systems.

Philip

The 555 can change the amount of fuel over a nine to one range at 6000 RPM.
More at lower RPM's.

Most people use a manifold pressure sensor but I think the mass air flow sensor
will do a better job job of sensing the amount of fuel required. Certainly
far better than a carburetor.

Well we will see. On the prop dyno we can change the engine load at
any RPM by changing the pitch of the prop. If that does not give us sufficient
control over the load we can change the diameter.

Paul Lamar