Subject: Measuring the amount of air going into the engine
From: Rotary Engine
Date: 6/1/2007, 3:41 PM
To: AARotary Engine



   Historically the amount of air going in the engine
   is measured with a venturi. The vacuum generated was then
   used to draw air out of a float bowl. Very simple. The drawback here
   is the venturi
   causes a reduction of air going into the engine due to aerodynamic drag.
   There is a trade off between the amount of vacuum you get verses the
   aerodynamic
   drag of the venturi. If you reduce the vacuum signal the carburetors
   tend to get
   more complicated and expensive This one of the reasons that
   carburetors have
   disappeared on cars. For a good discussion on this and the effect of
   right angle
   intake manifolds read the attached pdf.

   Aircraft engines assume the amount of air going is is directly
   proportional to the intake manifold pressure. This assumes there is
   some sort of
   valve that controls the intake manifold pressure. Be it slide
   throttle or butterfly.
   This is OK until you get near WOT and then changes in air flow
   result in very
   small changes in intake manifold vacuum.

   Another way is to measure the throttle valve opening and assume the
   amount of fuel
   is proportional to the area of the valve opening. This too is fraught
   with draw backs and suffers from some of the same problems with
   measuring
   the intake manifold vacuum.

   Modern cars use a hot wire in the air stream. The more the
   air the more the cooling effect on the hot wire. Measure the cooling
   effect
   and you know the mass air. Cold air is more dense than warmer air so
   this
   works pretty good. They are called mass air flow sensors and are an
   improvement over measuring intake manifold alone. They too have some
   aerodynamic
   drag and are rather expensive and bulky. The stock versions are cheap as
   they can be obtained in junkyards. The drawback is most come with a
   butterfly
   valve that add a lot aerodynamic drag. Even at WOT. A third
   disadvantage is you
   need to know what voltages they put out for X amount of air flow at
   Y temperature
   so they can be interfaced with the required computer. This
   information is not
   readily available. It is not a show stopper however as some experimental
   investigation can solve the problem.

   Another way of measuring the air flow is use a pito tube. These
   can be very small made from hypodermic needle tubing and
   consequently can
   have very low aerodynamic drag. They can be mounted anywhere in the
   intake system.
   They are cheap and easy to make. Almost free. IMHO The best place to
   mount them
   is near the beginning of the intake runners just after the trumpet
   mouth. This is
   where the air is least likely to be turbulent. What pito tubes do is
   measure
   the force of number of high speed air molecules impacting a flat
   surface.
   The force generated is proportional to speed squared so they are very
   sensitive to changes in air velocity (volume in a close tube) at all
   throttle openings.

   A crude form of this was a mass airflow sensor widely used on early
   EFI systems that consisted of a spring loaded trap door in the intake
   manifold. The air pressure would tend to open the door. How far it
   opened was measure and sent to the computer. Bosch built the early
   systems and they were widely used on many different cars including
   the second gen RX7. The draw back again was the spring resisted
   the door opening so they also caused aerodynamic drag. Also
   the signal was not very sensitive due to the geometry near WOT
   much like measuring the intake manifold vacuum.

   Solid state pressure sensors are cheap so a pito tube feeding data to
   a cheap PICAXE microcomputer controlling the amount of fuel going with
   pulse width modulated electric fuel pump sounds like a cheap and simple
   way to go. Measuring the temperature of the air will also
   be necessary but that is cheap and easy.

   The use of an air filter near the ground would be mandatory.

   Paul Lamar ...No rotor no motor.


   Nice overview of the subject, Paul.

   Just wondering about that picaxe -- what's the temperature range?

   Megasquirt claims their chip is automotive grade, which is -40 to
   125 C, I believe. That range is not really enough for fliers in cold
   areas. On a 0 F day, the temperature at 15,000 feet will be -54 F,
   or -47 C.


   Regards,

   Gordon.

Same as the chips Tracy is using. You put the computer in the cockpit
anyway. You are not going to fly for long with out a heater at
-54 F. I was stationed at Ramore radar station for a couple of years. And flew
the Beaver around up there once or twice. -54 F and die cast car door handles
shatter :)


Paul Lamar ...No rotor no motor.

The Rotary Engine NewsLetter. Powered by Linux.
ACRE NL web site. http://www.rotaryeng.net
Copyright 1998-2006 All world wide rights reserved.