Tech2Tech: EFI/DFI Series (Part II): EFI/DFI Theory

By Dave Worden

In part two of this series, I am going to cover the basic theory that two engine manufacturers use in DFI/EFI. Kawasaki uses Digital Fuel Injection (DFI) and Kohler Co. uses Electronic Fuel Injection (EFI). I will then begin breaking down components in regard to troubleshooting. While my strength will be with EFI because it is where I spent time in training and development, I believe the majority of troubleshooting techniques will be very similar.
DFI Theory

Kawasaki has been manufacturing fuel-injection systems for approximately 20 years. The Kawasaki DFI is a proven system in terms of simplicity, reliability, ease of maintenance, performance, economy and reduced emissions. The system uses an Engine/Electronic Control Unit (ECU), which monitors engine speed, load, atmospheric pressure, engine and air temperature. This system only uses three sensors and the ignition coils to determine the precise fuel needed. The Kawasaki DFI does not use an oxygen (O2) sensor, throttle position sensor or crank position sensor. This is called an “open-loop” system. It also uses a single ground system unlike other manufacturers. I will try and simplify the system and how it works:

1) When the key switch is on, the fuel pump is activated and the ECU reads the atmospheric pressure by using the vacuum pressure sensor. The ECU stores this value while the engine runs.

2) When the engine is started, the ECU uses the voltage from each ignition coil and uses the voltage spikes to calculate the rpm and position of the crankshaft relative to the pistons. It’s important to understand the ECU does not control the ignition coils.

3) The vacuum pressure sensor is read continuously while the engine is cranking or running.

4) The electric fuel pump is activated when the key is switched on. The fuel pump is passively controlled by the ECU via the OEM fuel pump relay. For safety reasons, if the engine does not crank three seconds, the pump will turn off. When the engine is cranking or running, the fuel pump operates continuously. Its location varies by OEM.

5) The fuel pressure regulator is fully mechanical and responds to the changes in the intake manifold pressure. Its function is to regulate fuel pressure to the fuel injectors as manifold pressure changes. As vacuum in the intake manifold increases, fuel pressure decreases. The fuel pressure regulator controls fuel pressure by changing the resistance of the fuel flow from the regulator back to the fuel tank.

6) The coolant temperature senses the engine temperature, and the ECU uses a denser mixture when the engine is cold. Conversely, on a warm engine, the ECU uses a leaner mixture. The coolant temperature sensor acts basically like a choke in a carbureted engine.

7) The fuel injectors are ground-controlled components. While the engine is cranking or running, the ECU intermittently grounds the fuel injector solenoid as determined by the ECU.

8) After receiving information from sensors, the ECU determines the fuel delivery to determine the fuel requirements. The ECU calculates when and how long to energize the fuel injector solenoids and grounds the control terminals. If the ECU detects a failed circuit, a fault code will be displayed. 


  Figure 1 — A fuel-injection system

EFI Theory

The EFI system used on Kohler Engines is designed to provide peak performance and optimum fuel efficiency. One of the major differences in this system from DFI is that it is a “closed-loop” system,” meaning it monitors itself and tries to continually operate to its maximum efficiency.

The central component in both systems is the Engine Control Unit (ECU). This component manages the majority of the components and their functions.

1) When the key switch is turned on, the relay sends a signal to the fuel pump and begins to pressurize the fuel system. Early units would turn on and build pressure up to 39 psi, and any excess fuel would return to the tank. On newer units, the electric fuel pump goes through an auto prime function if the unit is not started immediately. This also assists if the unit was run out of fuel and helps keep the fuel pump from burning out. Once pressurized, the fuel is fed to the rail and the injectors, and injected into the intake ports as needed. The length and duration will be measured in milliseconds and occurs each crankshaft revolution or twice for each 4-stroke cycle (see Figure 1).

2) Once the engine is started and running, it is in “open-loop.” The ECU then uses the speed sensor to determine the load and initial fuel mixture necessary. The primary signals are compared to pre-programmed “maps” and adjust as needed.

3) The ECU also monitors the load by using the throttle position sensor (TPS). This is similar to a Rheostat in that it sends a small amount of voltage to the ECU, which uses this data to again help determine the fuel air mixture.

4) Another sensor used is the oil temperature sensor, which helps determine how much fuel is needed when starting cold and also assists in sending the unit into “closed-loop.”

5) The last sensor used is the oxygen (O2) sensor. This unit provides an electrical voltage on initial start-up, and then when the exhaust temperature reaches 709 degrees Fahrenheit, it starts a cyclical voltage and sends the unit into the “closed-loop” stage. During “closed-loop,” the ECU has the ability to re-adjust temporary and learned adaptive controls, and provides the necessary changes to try and always run at its best fuel-to-air ratio.

This is a very simplified version of the theory, and it should make more sense when you start to troubleshoot the engines. As we break down the components for each unit and what can be done to service or repair both units, everything should become easier to follow. I would strongly suggest that all technicians attend any and all seminars on these types of fuel systems as they will make their jobs much easier and more efficient.

When it comes to troubleshooting EFI systems, there needs to be a systematic approach. Using flow charts and/or manufacturers’ diagnostic tools will always make things quicker, more efficient, and more importantly profitable. Too many technicians will become parts replacers and eventually fix the problem, but at what cost and do they know what part was causing the issue? I heard that the definition of insanity is to have a problem, apply the same solution over and over, and expect a different result! That is what can happen when working on EFI systems. Take your time to troubleshoot correctly. Do not cut corners. If you are having an issue, get assistance or step back from the problem, and look at it with someone’s assistance.


 Figure 2 —  Digital voltmeter

Let the troubleshooting begin

When a fuel-injection system is working properly, the signal lamp or MIL light will light up upon starting and should go out. If the light stays on, it is an indication that there is a fault. To access the manufacturers’ fault codes that will vary, the DFI will flash when the unit is running. Turn off the unit and turn it back on, and after a two-second delay, count the number of long and short blinks. On the EFI, the MIL will stay on after the engine is running. To access the codes, turn the key switch on-off-on-off-on, leaving it “on” in the third sequence. Any stored fault codes will then be displayed as a series of MIL blinks (from 2 to 6) representing the first digit, followed by a pause, and another series of blinks (from 1 to 4) for the second digit. In either case, once you have the blink code, you will then begin the transition to test and repair the fault.


 Figure 3 — A poor contact can cause interference.


  Figure 4 — Engine/Electronic Control UnitAs was stated in the first article of this series in July 2009 OPE, one of the most important tools needed will be a Digital VOA Meter (see Figure 2). In some cases, you will be taking readings in millivolts, and it is critical that your meter is capable of reading these small voltages. You must also be careful not to bend or damage the pin connectors; anything that could cause a poor contact between the wire leads and the ECU could cause false signals (see Figure 3). You should also realize that most of the electrical connections are moisture resistant. It does not mean they are pressure washer proof! Too many times units are pressure washed, which can cause water to enter areas and lead to connection issues that will again send a technician on some very drawn-out checks. Grease, spray, chemicals and water can all lead to interference of electrical connections and incorrect readings, so make sure all electrical connections are tight and that the proper sealants and/or cleaners have been used.

Certain components can be tested but not repaired

There are some components that can be tested, but due to cost or practicality, it is better to replace them. This is not meant to guess your way through and replace components haphazardly! Testing the component is necessary to do a proper repair. You can test the following items, but it would be best to replace versus trying to repair them:


 Figure 5 — Electrical relay


 Figure 6 — Fuel injector


 Figure 7 — Throttle position sensor


 Figure 8 — Oxygen sensor


 Figure 9 — MAP sensor1) Engine/electronic control unit (ECU) (see Figure 4)
2) Electrical relays (see Figure 5)
3) Fuel injector (see Figure 6)
4) Throttle position sensors (TPS) (see Figure 7)
5) Oxygen (O2) sensors (see Figure 8)
6) Manifold absolute pressure (MAP) sensors (see Figure 9)*
*A manifold absolute pressure sensor (MAP) is one of the sensors used in an internal combustion engine’s electronic control system. Engines that use a MAP sensor are typically fuel injected. The MAP sensor provides instantaneous manifold pressure information to the engine’s electronic control unit (ECU). This is necessary to calculate air density and determine the engine’s air mass flow rate, which in turn is used to calculate the appropriate fuel flow.

With this being said, you’re probably asking what can be serviced or repaired? There are still the other mechanical components that will need service due to natural wear. However, a unit running with EFI/DFI fuel systems will run cleaner and should have fewer issues in these areas as long as MDD has not set in. “MDD” stands for “Maintenance Deficiency Disorder” or improper maintenance being done by the owner of the equipment.

In the third article of this series on EFI/DFI engines, I will cover some basic troubleshooting codes and tests.

 Dave Worden has 40 years of extensive experience in the outdoor power equipment industry at the dealer, distributor and manufacturer levels. After beginning his career as a service technician for a dealership, he made the jump to a Central Distributor. There, he continued to work in the service department before he was promoted to educational director, representing Briggs & Stratton, Kohler, MTD and Tecumseh. He then moved up to the manufacturer level, serving as a territory manager for McCulloch Corp., a training specialist for Kohler Co. Engine Division, and a general manager for a manufacturer-owned dealership. In addition to being a contributing writer for OPE, Worden is currently operating a dealer consulting company called DJW Consultants, as well as serving as vice president of the Equipment & Engine Training Council, a committee chair for National SkillsUSA and an at-large representative for SkillsUSA Massachusetts.


Leave a Reply

Your email address will not be published. Required fields are marked *