Read this first.
http://www.rotaryeng.net/simple-cheap-555.html


just had another idea.

Use a cheap CPU with a look up table who's only function
is to modify the voltage sent to the 555 as a function
of manifold pressure. This correction or "trim" in EFI parlance can
be edited in a spread sheet and downloaded using the USB.
Henry Nee's idea.

Atmel chips have a similar asm instruction set to a 6502 or 68K.
ZBasic sell's them on boards with USB software drivers.

http://www.zbasic.net/

The program would be so simple it would be easy and quick to do and
maintain.

Main loop:
read the manifold pressure a to d
read the rpm.
Look up the correction.
Apply the correction to the d to a output pin
connected to the 555 pin 5, pulse width control pin.
Do it over again.

------------------------------******--------------
The main reason EFI system use computers to trim the mixture is the air
actually getting into the engine is not  a linear function of RPM as one
would
expect. The culprit is the intake manifold. How well it works is affected
by all the different resonant fluid conditions excited at different RPM's.
The intake air pulse frequency changes with RPM.

The other thing a 555 system does not do is compensate for air temperature.
It is a linear inverse function of temperature so it can be mathematically
achieved by a simple multiplication or division.

That can be done by the simple computer mentioned above but it could
also be done in hardware as Paul Lipps demonstrated in Klaus's early
analog fuel injection system. I just don't know how to do that yet :)

Here is a 3D surface that illustrates these points. A straight 555
system would have no hills and valleys. Just a simple smooth
surface.

BTW this can be plotted in a spread sheet program.
The Megasquirt tuning software does it for you.


Paul Lamar


I recently did a Subaru Wrx engine swap into an earlier Impreza. The owner
did not want to spend money on parts or the time it would take to install a
late model fuel tank assembly that matched what came with the car the
engine came out of.

When it came time to deal with the fuel temp sensor, I added a resistor in
the place of where the fuel temp sensor should have been.

With the main swap completed, we took the car to a local smog test station
to check if it would pass before taking the car to the Calif. BAR smog
station to legitimize the engine change in the eyes of mama pelosi.

The car easily passed the local tail pipe test with the resistor in place.
Upon taking the car to the official state exam station, the car failed tail
pipe.
The reason? The temperature on the first test was about 90F; the second
test the temp was about 65F. I was therefor paid to install a later model
fuel tank system.

Presumably, we will have to address the same issue, and other similar
issues with the 555 system. Particularly because altitude temp differences
can be far greater than what is regularly encountered while driving your
car.

Ed Carver

Use a mass airflow sensor instead of a manifold pressure sensor.
Since a mass airflow sensor has intake air compensation built in we can
use the computer that holds the trim values to linearize the non linear
mass airflow sensor.


Paul Lamar



Do you think using the maf will effectively cancel out variations in fuel
viscosity? I can see how variations in air will be compensated for. The
oems compute for fuel temp as well.

Ed Carver

I think that is probably not needed as Tracy's system is used by
dozens of people flying and there is no compensation for fuel
temp. The mass airflow sensor does not do that.

Paul Lamar


Engines ran fine for x years without fuel temp data. But now, we can tweek
for peak efficiency. So it's worth a try. If leaving it out meant failing
on HCs at the smog station, greater than 50 ppm, then retesting with it in
the loop meant passing with about 5ppm HCs; then it should be worth the
effort.
Fuel temp sensors are cheap. They are a variable resistors.

Ed Carver

5 parts per million is a drop in the ocean :)
A speck in the universe.

Paul Lamar