This article is written by Jim Roal and describes how to properly modify
your EFI fuel system for performance applications.
Details:
Performance EFI
Since the mid-1980's, carburetors have basically disappeared from production
cars and trucks. The main factor driving this was emissions compliance
and fuel economy standards but everything has improved because of it.
Today you can build a 10 second daily driver that can get good fuel economy
and your grandmother could drive it. You no longer have to battle
with constant carburetor problems, points that wear out, poor fuel economy,
and all kinds of driveability problems. Today you can drive to the
drags (and get well over 20mpg getting there), run consistent 12 second
quarter mile passes all night (or even quicker), and drive home.
You probably don't even need to bring tools. Most of this has come
about because of electronics. Modifying a late model EFI system
for performance is quite different from the old carburetors of yesteryear
however. In this article I will attempt to explain how a typical
EFI system (Bosch type, for Otto cycle engines) works and how you can
modify it to support more power.
Basics
The main components of most EFI systems includes fuel injectors, an engine
control module (ECM), an electric fuel pump, a fuel pressure regulator,
and several sensors. The typical EFI fuel system uses 40psi fuel pressure
across the injectors. The reason I say across the injectors is because
it uses a pressure regulator to maintain 40psi pressure difference between
the fuel pressure in the fuel rail and the air pressure at the injector
tip. This is done by connecting a vacuum line between the air side
of the regulator diaphragm and the intake manifold. When the manifold
is in a vacuum, the fuel pressure in the rail (what you would measure on
a gauge) will be 40psi plus manifold gauge pressure. For instance,
if you have 20 inches of Mercury vacuum in the manifold (approximately negative
10psig) you will have 40psig -10psig = 30psig fuel pressure in the fuel
rail. If the intake manifold has boost, say 6psig, you will have 40psig
+ 6psig = 46psig fuel rail pressure. The reason for this is to have
a constant pressure across the injector so the flow rates will be stable
for a given injector pulse width. This simplifies programming for
the engine control and eliminates the need for a fuel pressure sensor.
Many new EFI systems however, have eliminated the fuel pressure regulator
and now have a fuel pressure sensor. These systems control fuel pressure
by modifying the power to the fuel pump.
The ECM reads engine data from several sensors and uses this information
to calculate and deliver what is called a pulse width to the injectors.
The injectors have battery power to one of the 2 electrical pins whenever
the key is ON. The other electrical pin is connected to the ECM.
The ECM grounds this pin for a certain period of time which will deliver
a specific amount of fuel through the injector. The injector is
pulsed (or fired) relative to engine speed. Older systems fired
injectors in banks (4 at a time for a V8) but most newer systems fire
each injector independently. On a bank-fired system, each injector
will will fire every other crank revolution. Fuel flow is directly
related to pulse width. A longer pulse width will deliver more fuel.
A pulse width is the time in milliseconds the injectors is ON (fired).
The duty cycle is the percent of ON time relative to the maximum amount
of available ON time. As the engine speeds up, there is less time
between injector firing events because they are happening quicker.
100% duty cycle means the injector is ON the maximum possible time.
Injectors are rated by fuel flow at 100% duty cycle, generally pounds
mass of fuel per hour.
The ECM needs to know how much air, by mass, is entering the engine so
it can calculate the proper amount of fuel to achieve the proper air fuel
ratio. There are several ways to do this. You can measure
air flow and density (temperature and pressure)(this is a vane air flow,
or VAF, system), mass air flow (mass air flow or MAF system), calculate
air flow based on engine characteristics, speed, and air density (often
called speed/density), or other less common methods. Most early
port fuel injection systems used the VAF approach. It had a simple
air flap door that was pulled open by incoming air. There was a
temperature sensor in the system to measure inlet air temperature, and
most systems used an atmospheric pressure sensor as well. Air density
was calculated using the air temperature and pressure. Flow was
then measured using the VAF meter and the mass of air entering the engine
was then calculated. The speed density system required data from
a dynamometer to determine how much air entered the engine based off of
air density, engine speed, and intake manifold pressure. These systems
did not tolerate internal engine modifications because the air flow would
no longer be accurate based off of the manifold pressure. MAF systems
actually measure mass air flow directly, eliminating the need for more
complicated calculations. They are the most tolerant of internal
engine modifications.