ECI System Fundamentals Explained (STB-87-14-001)
Reference number: STB-87-14-001
ECI SYSTEM FUNDAMENTALS EXPLAINED
TECHNICAL SERVICE BULLETIN
ECI SYSTEM FUNDAMENTALS
Model 1983-87 Mitsubishi Starion (EFI); 1984-87 Mitsubishi Cordia/Tredia Turbo (EFI); 1985-87 Mitsubishi Galant/Mirage Turbo (EFI)
Group Engine
Bulletin No. STB-87-14-001
Date February, 1987
INTRODUCTION
The electronically-controlled fuel injection system in a Mitsubishi car is an example of advanced electronic technology applied to a popular consumer product. The system is made up of a computer, several sensors, and a set of injectors to meter fuel to the engine. Yet despite the system's complexity, it is actually as easy to diagnose as a mechanical system, and once the technician is acquainted with the characteristics of the system when it is functioning normally, he can successfully diagnose and correct even the most bewildering condition quickly and effectively. This bulletin describes the fuel injection systems used in current Mitsubishi cars and tells how they work. Also included is a diagnostic chart to help in determining when sensors are responsible for system malfunctions.
SYSTEM DESCRIPTION
Viewed in its simplest form, the electronic fuel injection system does nothing more than deliver to each cylinder the exact amount of fuel that is needed under a given set of conditions. Think for a moment of the conditions that would call for a change in the amount of fuel needed. Begin with the engine fully warmed up and idling. This can be called the "base line" because it represents the smallest fuel flow that the engine will need. Closed-throttle deceleration is an exception, because no fuel flow at all is desired at this time. Any other operating condition will call for more fuel. The sensors tell the computer how much additional fuel to order. They measure and report:
- Air flow through the intake system
- Coolant temperature
- Intake air temperature
- Throttle position
- Exhaust gas composition (O2 Sensor)
- Intake manifold pressure
Together, these sensors tell the computer to enrich the mixture so the engine can start and run when it is cold, or to provide an extra shot of fuel when the accelerator is depressed for acceleration. They tell it to lean-out the mixture as manifold pressure decreases or when the accelerator is released. In response to the O2 sensor, the computer changes the composition of the exhaust gases to minimize emissions. There are other sensors as well, such as the detonation sensor, but since these sensors do not control fuel flow they are not covered in this bulletin.
CLOSED-LOOP AND OPEN-LOOP OPERATION
The ECI system, like any other system that uses an oxygen sensor, has two modes of operation - "Closed Loop" and "Open Loop."
CLOSED LOOP
In "Closed-Loop" mode, the air-fuel ratio is varied by the ECU based on information supplied by the Oxygen Sensor.
OPEN LOOP
In "Open Loop" mode, the air-fuel ratio is controlled by information programmed into the ECU at manufacture. Fig 1 compares the two modes graphically. Both modes of operation are found in all MMSA ECI systems.
TYPES OF INJECTORS
Four different types of injectors are used on Mitsubishi cars. A single-point injector is used on the 1985 Galant. The 1986-87 Galant has a multi-point port injector system. Two types of twin-injector systems are used on all others.
SINGLE INJECTOR - 1985 GALANT
The 1985 Galant uses a single injector in the throttle body. This injector is capable of delivering fuel in response to a variety of requirements from idle to maximum power.
TWIN INJECTORS - BEFORE 1987
On pre-1987 cars with twin injectors, (Starion, Cordia/Tredia, and Mirage) the injectors pulse alternately - first one and then the other. See Fig 2 . Thus the total output of the two injectors is twice that of each one individually.
TWIN INJECTORS - 1987
On 1987 Starions and Mirages, twin injectors are also used, but on these cars two injectors of different flow rates are used. At idle speeds, fuel is injected through the smaller injector only. As load increases, fuel is injected by both the smaller injector and a larger second one. See Fig 3 .
MULTI-POINT INJECTORS
Cars with multi-point injection systems, such as the 1986 Galant, have one injector per cylinder. Thus the quantity of fuel delivered by each 1986 Galant injector is only approximately one-quarter as much as the single-point injector of the 1985 Galant. The pulse width is correspondingly less. Fig 4 shows the injector pulse pattern graphically. Note that in the Galant system, the pulses are timed by ignition system firing pulses.
INJECTION PULSE WIDTH AND FREQUENCY
In every Mitsubishi fuel injection system, injector nozzles open and close very rapidly in a series of pulses. The amount of fuel delivered per unit of time is controlled by varying the duration, or width, and frequency of the pulses while fuel pressure is held constant. Thus fuel delivery can be increased by increasing the width of each pulse, or by timing the pulses closer together i.e., increasing their frequency. The ECI computer controls this.
Pulse Width
The basic injection pulse Tb occurs at idle when the engine is warm. The pulse width is modulated by the computer in response to signals from the various sensors. See Fig 5 . Each logic circuit and memory in the computer connected with each sensor has its own modulation factor k, to change T. These relationships are explained mathematically later in this bulletin.
Pulse Frequency
In normal operation, each injection is synchronized to the signal pulse from the air flow sensor. The computer divides the frequency of the pulse of the air flow sensor signal frequency by N (2, 3, or more), as shown in Fig 6 below. An exception to this rule is the multi-point system in the 198 and 1987 Galant. In the Galant system, the fuel injection pulses are synchronized to the ignition pulses.
The chart below shows the pulse width at idle, medium load, and heavy load conditions. Note that the modulating ratio (1/N) is also given to give an idea of the frequency of pulsation.
SYSTEM DIAGNOSIS
To diagnose trouble in the system, it is useful to know how each sensor influences the pulse length and/or frequency. As you may remember, the air flow sensor is the only sensor whose signal is not a variable voltage. It is instead a pulsating constant voltage of varying FREQUENCY. The higher the air flow, the taster the air flow sensor pulses. Other sensors - throttle position sensor, air and engine temperature sensors, manifold pressure sensor, and throttle position sensor - put out variable voltages. The O2 sensor acts almost like an ON/OFF device, its output fluctuating back and forth between about 0.9V when the mixture is rich and about 0.1V when the mixture is lean.
The illustrations in Fig 7 through Fig 11 show the output characteristics of the various sensors under normal conditions.
Air Flow Sensor
The air flow sensor output cannot be measured with your voltmeter. An oscilloscope would be the best way to "see" the signal of this sensor. Fig 7 graphs the frequency range of this signal.
Coolant and Air Temperature Sensor
The coolant sensor and air temperature sensors both have the same characteristics: their output voltages range between about 1/2 V hot and 5V cold. Fig. 8 shows signal voltages and sensor resistances at five different temperatures.
O2 Sensor (Oxygen Sensor)
The O2 sensor sends a voltage signal that is high when the air/fuel ratio is richer than 14.8 to 1 and drops off sharply when the air/fuel ratio becomes leaner than 14.8 to 1. See Fig 9 . Because the computer changes the mixture instantly in response to the O2 sensor signals, the sensor output will normally fluctuate rapidly between 0.9 and 0.1V.
Pressure Sensor
The manifold pressure sensor voltage output can range from 0V to 4V. When the engine is stopped, the output should be about 2.3V. See Fig 10 .
Throttle Position Sensor
Output voltage of the Throttle Position Sensor should be familiar to every Mitsubishi technician because the initial setting is 0.48V. This is the voltage output of the sensor when the throttle is closed. At wide open throttle, the output voltage increases to about 5V. See Fig 11 .
SYSTEM DIAGNOSIS
The output graphs shown in Fig 7 through are for sensors in normal operating condition. A sensor whose output is not within the ranges on the graphs is suspect.
To help in diagnosing the ECI system, the following diagnostic chart should be helpful.
INJECTION PULSE WIDTH - MATHEMATICAL DESCRIPTION
As mentioned earlier, the basic injection pulse occurs at idle when the engine is warm, and the pulse width is changed by the computer to provide more fuel as the engine's requirements increase with speed and load.
The computer modulates the basic pulse width by a multiplier that IS composed of the inputs of:
- the air flow sensor
- the air temperature sensor
- the coolant temperature sensor
- the oxygen sensor
- the manifold pressure sensor
- the throttle position sensor for acceleration
- the throttle position sensor for deceleration
Stated in mathematical terms, the relationship is:
T = Tb x K , where K = Kwt x Kat x Kta x Kas x Klb x Kap x Kaci x Kbci
T = working injection pulse width
Tb = Basic injection pulse width
Kwt = Coolant temperature sensor factor
Kat = Factor for air flow sensor or engine speed
Kat = Intake air temperature sensor factor
Kas = Timer circuit and coolant temp. sensor factor
Klb = Oxygen sensor factor
Kap = Pressure sensor factor
Kaci = Factor for throttle position sensor - accelerating
Kaci = Factor for throttle position sensor - decelerating