Mario Farrugia

ECU Development for a Formula SAE Engine

Motivated by experiences in the Formula SAE ® competition, an engine control unit (ECU) was designed, developed and tested at Oakland University. A systems approach was taken in which the designs of the electronic architecture and software were driven by the mechanical requirements and operational needs of the engine, and by the need for dynamometer testing and tuning functions. An ECU, powered by a 68HC12 microcontroller was developed, including a four-layer circuit board designed for EMC. A GUI was written with Visual C++ ® for communication with a personal computer (PC). The ECU was systematically tested with an engine simulator, a 2L Ford engine and a 600cc Honda engine, and finally in Oakland’s 2004 FSAE vehicle.

Development of a programmable engine control unit to multi-fuel capabilities, progression to diesel

Electronic engine control is now an integral part of any research work on internal combustion engines. Electronic engine control provides versatility in engine manipulation for both performance and emissions objectives. The development of a versatile Engine Control Unit (ECU) for use on a variety of engines and engine configurations is discussed in this paper. The ECU handles gasoline (petrol), Liquefied Petroleum Gas (LPG, also referred to as AutoGas for vehicles) and recently also diesel. The architecture of the ECU using an embedded microcontroller and FPGA gives the ECU the capability to handle engines at 17000 revolution per minute. ECU embedded software and GUI provide ease of use for calibration and engine research. Communication via CAN bus, Bluetooth® and WiFi give high speed and wireless connectivity respectively. This paper gives an overview of the overall development of the ECU and details the most recent specifics to handle common rail diesel engines.

Mechatronics for water injection in SI engine

Internal combustion engines are generally speaking mechanical devices but have benefitted tremendously from electronic controls. The electronic implementation of fuel injection and spark ignition on an older single-cylinder JAP engine through the use of a programmable Engine Control Unit (ECU) is discussed in this work. Furthermore the additional electronics required for water injection and boost control that were not supported by the programmable ECU are detailed. The injection of water into the Spark Ignition (SI) engine was the main objective of this study but the solution for this task through the use of electronics and LabView provided a robust and flexible system whereby the whole range of parameters could be explored. The engine was electronically controlled by the ECU through the sensing of 36 teeth that were machined in the dynamometer coupling and by having the cam sensor triggered by the maximum lift of the tappet of one of the valves. Water injection electronic circuitry was designed and built from scratch. The variable duration water injection pulse was generated by a 555 timer IC in Monostable (configuration) mode that was triggered every two engine revolutions by the fuel pulse. The duration of water injection per cycle in milliseconds was recorded by the ECU through the use of a dual potentiometer that controlled both the RC value for the 555 timer and also generated a voltage that varied linearly with water pulse duration for data logging by the ECU. Boost control of the engine from the laboratory compressed air supply was first tried to be solved by mechanical means namely manual control, an external wastegate (used as a proportional valve) and also different sized pressure regulators, however these attempts were unsuccessful. A very straightforward but ultimately functional solution was the use of an ON/OFF solenoid valve that opened compressed air (6 bar nominal) to a large (approx. 1m3) tank that fed the SI engine. The ON/OFF control was performed through LabView and a specifically built solenoid driver circuit. The control of air temperature into the SI engine was also electronically controlled through ON/OFF control from LabView which controlled the water temperature of a charge-air heat exchanger. These systems were integrated together and performed as desired to test all desired water injection durations, boost levels, air inlet temperatures, spark timing and fuel quantity.

Transient surface-heat-flux measurements in the exhaust of a SI engine

Surface-heat-flux measurements were performed at two locations in a straight-pipe extension of the exhaust-port of a SI engine. The measured surface-heat-flux history was characterised by a double peak, during the exhaust valve-open period. The largest peak resulted from the action of high-velocity, blowdown gases exiting the combustion chamber at the beginning of the exhaust process. The significantly lower second peak was due to the slower piston-driven gas motion that occurred during the displacement phase. During the closed-valve period, the local surface-heat-flux was found to be relatively low (10-45 kW/m 2 ) in comparison to the peak heat flux (80-288 kW/m 2 ), and to exhibit minimal decay. The predicted heat-flux histories from an engine-simulation analysis, which employed a heat transfer correlation developed by the authors, were in good agreement with the corresponding measured histories during blowdown, however, predictions were relatively poor during the displacement phase, both in magnitude and behaviour.

Common rail diesel engine, fuel pressure control scheme and the use of speed — Density control

Diesel engines in both the automotive and industrial sectors are nowadays electronically controlled. Electronic control provides accurate real time processing of the relevant sensory data on which fuelling is determined. Fuel economy as well as emissions are very dependent on the diesel fuel quantity, timing and injection pressure, which in turn determines the shape of the injection rate diagram, the spray distribution and the fuel droplet size. NOx is also known to be very dependent on flame temperature which can be controlled by the quantity of Exhaust Gas Recirculation (EGR). Research on the real time control of a common rail diesel engine was undertaken at the University of Malta. A Peugeot 2.0HDi common rail engine was controlled with a custom made electronic Engine Control Unit (ECU). A control strategy for the diesel rail pressure was developed that is based on simple 2D look up tables rather than more complex 3D maps. This allowed the ECU Graphical User Interface (GUI) and ECU firmware to be very similar to the gasoline (petrol) version of the ECU. The successfully developed fuel rail pressure scheme based on RPM, Mass Air Flow and Torque Requirement is given in this paper. Furthermore the values found experimentally for the speed-density control of the engines are also given. Top Dead Center (TDC) determination based on in-cylinder pressure measurements was conducted to better calculate the Start of Injection (SOI) and Indicated Mean effective Pressure (IMEP).

Integration of force sensing for robotic generic assembly

A force sensor was integrated into a robotic cell and the sensor data was made available in real-time for the robotic arm. Integration of the extra sensory data was done through an external supervisory computer system that was developed as a generic controller to the robotic arm. The supervisory computer system was implemented using transputer modules running parallel processes and programmed in C programming language. The generic assembly meant that the robot arm used, an ADEPT using Val II as a programming language, was used in a very basic form whereby the required task was split up into more fundamental operations of feed transport and mate as proposed by Selke [1]. Manipulators often have programming languages which apply to only a specific brand or type of manipulator, in fact this is the problem that has led to the need of a generic controller for assembly. The generic controller was designed to achieve the necessary assembly task irrespective of the specific manipulator programming language. Testing of the developed hardware and software was performed on both a toy car assembly system and a peg in hole setup. Both testing setups were successfully assembled using real time force data. The insertion force traces for the peg in hole tests is reported here. Further development in intelligent assembly using statistical methods is also proposed.

The Effects of Thermocouple Materials and Insulating Mica in an Erodable Surface Thermocouple

A finite-difference model of a surface thermocouple (erodable-ribbon type) of a heat flux sensor was built to analyze the transient response of the thermal junction and the two-dimensional effects created by the insulation between the thermocouple materials and the body material of the sensor. Such transient heat flux sensors have previously been used for measurements in internal combustion engines. It is commonly assumed that the heat transfer within these devices is one-dimensional even though the sensors are constructed from at least two different materials. It is common practice to calculate the transient heat flux using properties of body material and this leads to a substantial error as demonstrated by the model. With these sensors, low thermal capacity thermocouple junctions are formed near the surface by abrasion and response times as low as 30μs have been reported. Experiments were performed on an E type surface thermocouple heated at 11W by means of a copper vapor laser pulsating at 10kHz. Measurement of surface thermocouple temperature was performed at a 100kHz sampling rate. A finite-difference model was used to analyze the response of these sensors to the pulsed laser heating operating at 10 kHz. The insulation between the thermocouples and the body material was mica and the body material was AISI 316 stainless steel. The experimental measurements and simulation results are reported in this work. The analysis and comparison of experimental and simulation results showed that for such thermocouples two-dimensional effects exist due to the presence of mica sheets. The temperature decay between pulses was better matched using thermal properties of mica sheets rather than the thermal properties of the body material. However the body material still dominates the temperature swing of the thermocouple junction.Copyright