Optimization of FreeValve´s fully variable valve control system for a four-cylinder engine
Robin Hamberg
Master of Science Thesis Stockholm, Sweden 2017
Optimization of FreeValve´s fully variable valve control system for a four-cylinder engine
Robin Hamberg
Master of Science Thesis MMK 2017:36 MFM 168 KTH Industrial Engineering and Management
Machine Design SE-100 44 STOCKHOLM
Examensarbete MMK 2017:36 MFM 168
Optimering av FreeValves fullt variabla ventilstyrningssystem för en fyrcylindrig motor
Robin Hamberg
Godkänt
2017-05-16
Examinator
Anders Hultqvist
Handledare
Andreas Cronhjort
Uppdragsgivare
FreeValve AB
Kontaktperson
Urban Carlson
Sammanfattning
Bilindustrins avgaskrav är striktare än någonsin och tillverkare spenderar allt större resurser på att sänka bränsleförbrukning och avgasutsläpp från sina bilar. Något de flesta inte provar är att ändra en av grundprinciperna i förbränningsmotorn. Kamaxlar används för att styra ventilerna i gasväxlingsprocessen och har trots många variationer sett mer eller mindre likadana ut det senaste århundradet. Den största nackdelen med kamaxeln är att den optimala gasväxlingsprocessen varierar med motorvarvtal och last, men det gör inte kamaxeln. Variabel ventilstyrning som många tillverkare använder sig av minskar problemen men för att få ut mesta möjliga prestanda krävs en fullt variabel ventilstyrning. FreeValve AB är ett litet svenskt företag som utvecklar ett sådant system som använder en Pneumatisk Hydraulisk Elektrisk Aktuator för att styra varje ventil.
Syftet med examensarbetet var att designa om FreeValves elektriska ventilstyrsystem för att passa nya krav på att styra en fyrcylindrig motor i en bil. Designprocessen av kretslösningarna presenteras tillsammans med en litteraturstudie som identifierar designaspekter till en kommande prototyp av ventilstyrsystemet som ska användas i ett motorrums krävande miljö.
En kravspecifikation sammanställdes och det verifierades att delarna av systemet som rymdes i projektets omfång var designade och fungerande enligt ställda krav och enligt litteraturstudiens förslag, med huvudsyfte att styra aktuatorerna med lämplig strömprofil.
Master of Science Thesis MMK 2017:36 MFM 168
Optimization of FreeValve´s fully variable valve control system for a four-cylinder engine
Robin Hamberg
Approved
2017-05-16
Examiner
Anders Hultqvist
Supervisor
Andreas Cronhjort
Commissioner
FreeValve AB
Contact person
Urban Carlson
Abstract
Automotive exhaust legislation is stricter than ever and manufacturers are spending increasing amounts of resources on reducing fuel consumption and emissions from their vehicles. What most do not try is to change one of the most fundamental concepts of the internal combustion engine. The camshaft is used to control the poppet valves in the gas exchange process and has, despite many variations, stayed more or less the same the last century. The largest disadvantage with the camshaft is that the optimum gas exchange process is dynamic with engine speed and load, where the camshaft is not. Variable valve timing as many manufacturers explored reduces the problem but in order to get the real advantages a fully variable valve timing system is needed. FreeValve AB is a small Swedish company that develop such system which is based on a Pneumatic Hydraulic Electric Actuator to operate each valve.
The aim of this Master’s thesis was to re-design the FreeValve electric valve control system to suit new and updated requirements of running a four-cylinder engine in a car. The circuit design process of the system is presented along with a literature study to identify design considerations for a future valve control system prototype to operate in the demanding environment of an engine compartment. A requirement specification was established and it was verified that the parts of the system within the scope of the project were successfully designed according to the literature review and specification, with the main task to deliver the correct current profile to the valve actuators.
FOREWORD
I would like to thank the following persons for making this thesis project possible:
My supervisor Urban Carlson for giving me the opportunity to perform an interesting thesis project on a hot topic at FreeValve AB and providing help and advice when needed.
Lars Ivarsson for giving me guidance, ideas and advise in all technical aspects.
My academic supervisors Andreas Cronhjort and Anders Christiansen Erlandsson for help with all administrative practicalities and for keeping the project on track.
My friends and fellow MSc students Goutham and Paul for providing feedback during planning and closure of the thesis to improve the work.
At last thank you to all the others that contributed in different ways, you know who you are.
Robin Hamberg Solna, September 2016
NOMENCLATURE
Abbreviations
A-D Conversion Analog-Digital Conversion BGA Ball Grid Array
BOM Bill Of Materials
CAN Controller Area Network
CTE Coefficient of Thermal Expansion DEP Divided Exhaust Port
DTM Dynamic Thermal Management
DVFS Dynamic Voltage and Frequency Scaling ECU Engine Control Unit
EMC Electromagnetic Compatibility EMI Electromagnetic Interference ESD Electro Static Discharge FEM Finite Element Methods FVVT Fully Variable Valve Train ICE Internal Combustion Engine IGBT Insulated-Gate Bipolar Transistor IMS Insulated Metallic Substrates
MOSFET Metal Oxide Semiconductor Field Effect Transistor NEDC New European Driving Cycle
PCB Printed Circuit Board
PHEA Pneumatic Hydraulic Electric Actuator PQFP Plastic Quad Flat Pack
PTH Plated Through Hole VCS Valve Control System
WLTC Worldwide harmonized Light duty driving Test Cycle
TABLE OF CONTENTS
Sammanfattning ... 3
Abstract ... 5
FOREWORD ... 7
NOMENCLATURE ... 9
TABLE OF CONTENTS ... 11
1 INTRODUCTION ... 13
1.1 Background ... 13
1.2 Purpose ... 13
1.3 Delimitations ... 14
1.4 Method ... 14
2 FRAME OF REFERENCE ... 15
2.1 FreeValve´s PHEA ... 15
2.2 Electronics integration in cylinder head ... 18
2.2.1 Elevated temperatures ... 18
2.2.2 Thermal cycling ... 19
2.2.3 Vibrations ... 20
2.2.4 Sealing ... 21
2.2.5 Lubrication oil ... 21
2.2.6 Surface treatment ... 21
2.2.7 EMC design ... 22
2.2.8 Expected lifetime ... 23
2.3 Previous valve control system ... 23
2.4 Valve timing concepts ... 24
3 IMPLEMENTATION ... 25
3.1 Requirement specification ... 25
3.2 Prototype design ... 26
3.2.1 Boost voltage controller ... 26
3.2.2 Actuator controller ... 29
3.2.3 Diagnostics ... 32
3.2.4 Power input protection ... 32
3.2.5 Fuel injectors and ignition coils ... 33
3.2.6 Microcontroller ... 34
3.2.7 New microcontroller... 34
4 RESULTS ... 35
4.1 Literature study ... 35
4.2 Prototype ... 36
4.2.1 Boost power supply ... 37
4.3 Validation ... 38
5 DISCUSSION ... 43
5.1 Valve control system ... 43
5.2 Boost supply ... 43
5.3 Ethics ... 44
5.4 Scope of thesis ... 44
6 Conclusions ... 45
7 FUTURE WORK ... 47
8 REFERENCES ... 49
List of references ... 49
List of data sheets and application notes ... 50
APPENDIX: SUPPLEMENTARY INFORMATION... 53
Appendix A ... 53
... 56
... 57
... 58
Appendix B ... 61
1 INTRODUCTION
This chapter will describe the background and motivation of the project. The purpose of the project will be presented along with the method that was followed and the delimitations.
1.1 Background
One of the largest challenges for the automotive industry is to meet current and future exhaust legislation. The transportation sector accounts for a large part of the global greenhouse gas emissions and of which a major part is from the internal combustion engines. Even though electric propulsion is growing fast among passenger transportation the energy density of the liquid fuels, short refueling time and well established infrastructure for an ICE (Internal Combustion Engine) is still far ahead of alternative technologies. The development of the ICE has made small progress the last century and is still relying on old principles with different modern variations and add-ons. The ICE is basically an air pump and by adding fuel with an often stoichiometric mixing ratio, power can be extracted. The important gas exchange process is controlled by the camshafts and poppet valves inside the engine. A well designed camshaft together with efficient flow paths for the air is necessary for good overall engine efficiency. A major drawback with the camshaft is that there is no one design that fulfills all needs (Trajkovic, 2010). The optimum cam profile differs with intended use, engine load and engine speed. There are several systems to alter the cam profile and timing, for example BMWs Vanos or Hondas V-TEC, but there is still the dependence on the camshaft and full flexibility is not possible. One solution to get full flexibility is to remove the camshaft and actuate all the valves individually.
FreeValve is a small Swedish company that is developing a valve control system (VCS) for valve actuation in internal combustion engines without the use of a traditional camshaft. Each valve is controlled by an PHEA (Pneumatic Hydraulic Electric Actuator) and the valve actuators are housed in two rails in the cylinder head, where the camshafts normally sits. As a sister company to Koenigsegg the development at FreeValve has progressed for more than a decade to change one of the most basic principles of combustion engines. The FreeValve technology offers a Fully Variable Valve Train actuation (FVVT) which opens new possibilities for higher fuel efficiency, higher power and lower emissions.
1.2 Purpose
The aim of this thesis is to convert the electrical part of an existing valve control system (VCS) from modular laboratory use to a prototype to use in a car on road. The current VCS is a development friendly modular system and has been tested thoroughly, with the next step being to adapt it for use on a four-cylinder engine in a car. The VCS is to be optimized to control 8 valve actuators, read 8 valve position sensors and control 4 coils, either fuel injectors or ignition coils. The VCS should handle up to 10 000 rpm engine speed. The deliverable of this thesis is a circuit design of the VCS that connects between the engine ECU (Engine Control Unit) and the actuator rail.
The possibility to integrate the VCS circuits in the cylinder head will be investigated in a literature review and the prototype VCS circuits should be adapted for this already now to as large extent as possible.
1.3 Delimitations
The deliverable in this thesis is based on an existing system, which means the solutions will not be developed from scratch. The system will within the time frame be re-designed to support operation in a car, but will not be put into series production. The system will be fully designed except for physical PCB (Printed Circuit Board) design and manufacturing.
The implementation of the VCS electronics inside the actuator rails will be investigated in a literature review. No such prototype will be built but design considerations obtained during the literature review will be used in the development of the deliverable where applicable to shorten the lead time in future rail integration.
1.4 Method
The thesis will follow a common product development methodology starting with establishing a requirements specification of what requirements the deliverable must satisfy. Quantitative requirements will be given an interval that prototype validation measurements should operate within. A literature review of related work will be conducted to identify design considerations as well as the current state of the art within the company and in the industry. Later the existing circuits will be broken down into suitable size parts prior to the re-design phase. The completed parts will be put together and form the hardware for the prototype. At last the prototype will be validated against the product requirements.
Following these steps ensure that the prototype is designed with the latest technology and knowledge in the field.
2 FRAME OF REFERENCE
In this chapter, the state of the art within the company and in the industry, will be presented as a foundation to the system design.
2.1 FreeValve´s PHEA
The FreeValve system consists of a number of valve actuators as depicted on the cover page which are mounted on a rail, one actuator for each valve as seen in Figure 1.
The abbreviation PHEA stands for Pneumatic Hydraulic Electric Actuator. Pneumatic which is the acting force when a valve is actuated. Hydraulics act as a dampener to the pneumatic force and to lock the actuator in open position. Electrics control the pneumatics and hydraulics and do also monitor the movement of the valve. The symbiosis of those three techniques results in a system that is reliable, fast, quiet, compact, light and reasonably cheap. All of which necessary in a production car.
A typical peak efficiency for a conventional gasoline engine is 37,6%. However, since the engine is not optimized to run on part load, when it’s rather throttled to lower power output the efficiency drops to about 20% on part load (FreeValve 1, 2016). The PHEA enable the engine to be optimized for different loads and to run closer to peak efficiency for a larger part of the driving cycle. This can be done in a number of ways, among others by removing the throttle for higher volumetric efficiency, deactivate cylinders to reduce part load losses, deactivate valves to alter the charge flow in the cylinder, and utilize different residual gas scavenging techniques to prevent knock and reduce exhaust emissions. An over expanded combustion cycle called the Miller cycle, with longer expansion than compression stroke, can be utilized to decrease the engine power density and thermal efficiency in order to reduce the pumping losses. A trade off which can increase the overall efficiency and reduce fuel consumption at low load conditions
Figure 1: FreeValve rail with actuators and valves (www.FreeValve.com)
(Wan and Du, 2013). With a fully variable valve train (FVVT) it is easy to alter the effective compression ratio, hence the engine can be built with a higher compression ratio, admitting a higher peak efficiency (Trajkovic, 2010).
The power consumption of the PHEA is roughly 2% of the engine power output and around 10% lower than a conventional valve train during normal driving (FreeValve 2, 2016). During higher engine speeds, the PHEA will experience higher losses than a conventional system, but the average driving cycle on lower average engine speed will still benefit from the FreeValve system. An estimation on an engine currently in development is an improvement in part load efficiency by around 5% and a fuel consumption reduction by 12-17% compared to an equivalent cam shaft engine (FreeValve 3,2016).
The FreeValve system offer the valve lift profiles in Figure 2, where normal camshafts engines are stuck with the regular cam lift profile with heavily rounded corners as depicted in Figure 3, with some variation. The FreeValve lift profiles offer the possibility to change height and change width from cycle to cycle which allow great responsiveness and high engine performance for any operating condition.
Figure 2: FreeValve valve lift profile
The larger area under the FreeValve profile compared to a regular cam profile means that more air-fuel mixture will pass with each valve opening. The benefits of this combined with throttle less operation and other improvements possible to implement is shown in Figure 4.
Another benefit of the individual valve actuation compared to a camshaft engine is higher fault tolerance, despite the higher complexity. The FreeValve engine can operate when one or a few valves are defective, with reduced output power but without damaging the engine. Up to 75%
of the actuators can fail before the engine can no longer provide limp home operation, but with an otherwise undamaged engine. This is when a broken timing belt normally has catastrophic consequences in damaged parts and large costs.
Figure 4: Engine torque improvement example with FreeValve system.
Figure 3: Camshaft generical valve lift profile
2.2 Electronics integration in cylinder head
The current VCS is located in an external box in the engine compartment of the car and is communicating with the engine ECU to control the actuators inside the rails in the cylinder head. Four wires to each actuator implies that a lot of wires have to be routed from the 16 actuators through the engine compartment to the VCS. To reduce the wire harnessing and get better communication with less disturbances it is favorable to move the VCS inside the actuator rails. This will reduce wiring cost and complexity and will simplify assembly.
Housing electronics in the harsh environment inside the engine compartment that should comply with requirements on longevity and reliability is a difficult task, and to integrate electronics inside the engine itself pose an even bigger task. Higher levels of temperatures and vibration, exposure to oil and high pressure are the downside of having a system this highly integrated.
This section will identify design considerations that apply to the future rail integration in particular, but also to some extent for the prototype of this theses. To design according to these considerations is important in order to meet the requirements for this prototype and to save development time in future rail integration.
2.2.1 Elevated temperatures
The operating temperature of the electronics depend on the mounting location in the vehicle. A possible temperature range according to (Wayne Johnson et al, 2004) is -40ºC to +175ºC for electronics mounted on the engine, but can reach 200ºC depending on location and operating point. Figure 5 show a FreeValve rail operating on an engine under load and 25ºC ambient temperature. The rail is shielded from the glowing turbo and exhaust manifold to the left and has a surface temperature close to 80ºC where the electronics would be located.
Even in this cold environment the headroom to the maximum temperature limit of the electronics have shrunk significantly and the system will have to be carefully designed for sufficient thermal management. Too high operating temperature reduces the operating life, reliability and efficiency in an electronic system. Moreover, a power device operating at its maximum rated temperature can sink zero power without violation of its temperature rating due to the internal heating. This gives very small headroom for devices operating in the engine compartment with temperatures close to their maximum rating. Higher power density and the tendency to mount electronics on hotter locations require efficient thermal paths to maintain the operating temperature within specified range. Adding a traditional heat sink on components to reduce operating temperature is well proven but add a costly step in assembly (Lall et al, 2006).
One alternative is to use plated connection vias through the PCB to transport heat to other heat conducting layers, for example thick copper layers. Worth noting is that only filled vias can be used underneath components to prevent solder soaking and faulty solder joints. Heat can effectively be dissipated in a PCB metal backing, called IMS (Insulated Metallic Substrates).
Metal backed PCBs provide excellent thermal dissipation and mechanical stability against vibrations, but due to higher CTE (coefficient of thermal expansion) than traditional PCBs the mechanical strain on components is higher and the solder joint reliability is lowered. A thermal cycling experiment conducted by (Lall et al, 2006) showed a solder joint crack propagation rate 1,7 times higher for a BGA (Ball Grid Array) package mounted on a IMS board compared to a standard board. Hence a tradeoff has to be made between thermal dissipation and reliability.
The impact of mechanical strain can be reduced using BGA underfill, but this adds cost and steps in assembly.
The estimated power dissipation from the electrical components will have to be calculated and suitable thermal paths will have to be designed. The system can either be cooled actively by the pressurized air, oil and cooling fans, or cooled passively by a heat sink to ambient air. The active cooling need a cooler for the cooling medium and a guarantee that the temperature will not rise to unaccepted levels when the engine is stopped. For example, with a delayed stop of the active cooling. A passive cooler will have to be properly located and be large enough to dissipate enough heat in worst possible operating condition.
2.2.2 Thermal cycling
Any component mounted on an ICE will be subject to thermal cycling and thermal shocks.
Some of which spanning through the entire specified temperature range in a short time, during a cold start in winter for example.
Different CTE for different components will cause strain in the joints when heat is applied (Li, 2004). In electronics, the solder joint between components and the PCB is subject to this strain and hence fatigues during thermal cycling. By selecting components and PCB assemblies with matching CTE the strain on joints are reduced, otherwise the stiffness in the solder joint must be low enough to meet the lifetime requirements. Leaded components are less sensitive to thermal cycling than leadless components due to the compliance in the gull wing leads (Li, 2004). Experiments also showed that the BGA ball fatigue performance differed with type of packaging, PCB material and layout. (Li, 2004) showed that in a BGA package the inner rows
solder joints are more prone to crack than the corner joints since the thermal stresses are amplified by the rigid silicon die. The opposite was shown when (Lall et al, 2006) performed thermal cycling on a IMS PCB and showed that the corner BGA solder balls experienced the highest deformation. This concludes that several factors play a role in the solder joint fatigue and it’s difficult to predict where the largest stresses are occurring. Thermal shock experiments by (Pippola et al, 2014) showed that a common type of PCB failure was so called barrel crack failures in plated through hole vias due to the difference in CTE of the PCB material FR-4 and the PTH (Plated Through Hole). This can be limited by matching the PTH with the PCB properties.
An electronic unit mounted on the engine is directly impacted by the engine temperature and its variations with different driving conditions. For example, city driving will have intermittent changes in air speed and engine load. Apart from ensuring high thermal conductivity to the ambient to avoid overheating DTM (dynamic thermal management) can be applied to the electronics (Park et al, 2011). DVFS (dynamic voltage and frequency scaling) is a popular method to limit the power dissipation, where operating voltage and frequency of an integrated circuit changes with operating condition to reduce power loss. The engine temperature measurements by the ECU and the engines thermal capacity between the heat source and the electronics can be used to estimate the temperature in the electronic unit to ensure maximum performance without overheating, and still not using any extra temperature sensors.
2.2.3 Vibrations
Despite the use of different engine concepts and balance shafts the engine will still be a major source of vibration, which everything mounted on it will have to endure. Vibrations will cause the PCB to bend and cause component and solder joint fatigue.
Experiments by (Li, 2004) showed that for BGA packages there was no solder joint failures for packages located close to the PCB mounting location. PQFP (plastic quad flat pack) packages showed a 5-10 times longer fatigue life when mounted close to the PCB mounting location opposed to the board center. The PQFP package broke off the leads close to the package corners rather than crack the solder joint. Similarly, the highest ball stresses for the BGA package was near the package corners. FEM (Finite Element Methods) analysis confirmed that the bending of the PCB due to vibrations is the dominant cause of joint failure. This low cycle vibrations with large amplitudes and plastic deformation on the solder joints should be limited. A stiffer PCB with better clamping will instead yield high cycle vibrations with lower amplitudes where the deformation in the joints remain elastic.
2.2.4 Sealing
A proper sealing of the system will prevent any of the oil or the pressurized air to escape the system or let any moisture or ambient matter in. A conductive EMI-sealing gasket (Electromagnetic Interference) will help in reduce EMI escaping or intruding the system by forming a continuously electrically conducting housing environment. If the system varies between under- and over pressure in relation to ambient, air will try to pump through all the seals with possible moisture ingression as a result. The sealing must be designed for the necessary over pressure but also some under pressure inside the system. Any moisture inside the system will condense after engine operation on all contacting surfaces. Sensitive surfaces like electronics will need some sort of surface treatment to avoid damage and ensure good reliability.
Connectors must be properly sealed to prevent any leakage and rated for the expected temperature and vibration levels. No cables can be routed directly from the rail since pressurized air can escape through them, between the strands and the insulation. The rail must be sealed off with a connector.
2.2.5 Lubrication oil
The electronic unit will be exposed to lubrication oil which can cause complications. The oil needs to be dielectric enough to not cause any short circuits in the electronics. At the same time, it has to be conductive enough to not build up electrostatic charge. A phenomenon known as tribocharging is known to occur in flowing oil when the contact between the oil and a solid surface is sheared, generating free electrostatic charges (Harvey et al, 2002). The charge builds up is influenced by the base oil, additives, age of oil and occurrence of moisture, contaminates and oxidation among other things. An ESD (Electro Static Discharge) can seriously damage the electronics or burn the lubrication oil, forming sludge. The conductivity should be higher than 400 pS/m to keep charge build up at a reasonable level (Ölchecker, 2012). The same source state that charge build up can be reduced by using proper sized earthed oil piping, have moderate flow speeds, avoid dissolved air and use low friction filter elements.
Lubrication oil might as well be corrosive or carry conductive debris like metal wear particles to the electronics. Good filtering and PCB coating will prevent corrosion and short circuits.
2.2.6 Surface treatment
Surface treatment can be used to protect the electronics from the harsh environments inside the actuator rail. Different kinds of polymer coatings or moldings can protect against particles, corrosive oil, pollutants and humidity along with reducing thermal and mechanical stress (Pippola et al, 2014). Such polymer moldings and coatings can be epoxy, polyurethane, silicone, acryl or polyamide, all with different properties. When choosing suitable surface treatment properties like CTE, elastic modulus, resistivity to chemicals, moisture absorption, adhesion and thermal and electric conductivity should be considered. Some parameters reduce the reliability of the systems and have to be carefully balanced to the benefits of the surface
treatment. For instance (Pippola et al, 2014) identified several threats to reliability caused by conformal coating and moulding material. Potential thermal shock failures included delamination and cracked solder joints under polyurethane moulding caused by the rigidity and different CTE of the moulding in relation to the PCB and component interconnections. Results by (Goth et al, 2012) and (Pippola et al, 2014) showed that silicone based moulding material is well suited for thermal cycling due to the low elastic modulus is not putting strain on component interconnections. The silicone moulding did also decrease the heating rate of the PCB and caused a more constant heat distribution in ambient temperature cycles. How this affects the internally heated components is not discussed, but it can be assumed to change the heat dissipation to air so the thermal conductivity of the moulding must be considered for sufficient cooling.
One observation made by (Pippola et al, 2014) was that the conformal coating not only was protecting the components, but in some cases penetrated into the component and put excess strain on internal copper wires causing them to fail. A different risk with conformal coating is potential ingression of water or other corrosives in case of a coating defect which will damage the electronics (Goth et al, 2012). A more expensive conformal coating is parylene with excellent dielectric, thermal, flexibility and penetration properties. But the application process is complicated and not very suited for large volumes.
A good method to protect a circuit where serviceability is not of concern is potting encapsulation where the PCB is embedded in potting material. This gives excellent protection against mechanical shocks and vibrations. Depending on the thermal conductivity of the potting material good thermal properties can be achieved which improves reliability in thermal cycling (Goth et al, 2012). The distribution of heat in the potting material helps in reducing hotspots and the impact of differences in CTE. By choosing a potting material resistant to oil all the issues in the earlier oil section will be of no concern.
2.2.7 EMC design
The VCS will be enclosed inside the cylinder head which will act as an EMI shield if sealed properly according to the sealing section. Hence the largest considerations for EMC (Electromagnetic Compatibility) will be the internal radiated EMI between different parts of the PCB and also the EMI transmitted though the cabling. To keep the sourcing of EMI low the first step is a proper and robust design of the PCB and cabling. For example, by keeping current paths and current return paths close to each other and to keep current return paths separated between power and signals. Also use proper shielding on cables to reduce stray inductance
2.2.8 Expected lifetime
The expected lifetime of an electronic unit depends on the environment it will be subjected to.
Often a combination of the discussion in the sections above. Adding to that is the lifetime of single components like capacitors expressed in hours or cycles and is dependent on operating conditions. Power cycling models can be used to estimate the lifetime of an electronic system (de Brito et al, 2015). The power cycling includes the Joule effects internal heating due to internal resistance of the components and may give a more accurate estimation on lifetime, but is difficult to model. Simulations of the PCB lifetime including solder joints and such can be done with software, for example with Sherlock Automated Design Analysis™ provided by DfR Solutions or with The Anand Viscoplasticity model in ANSYS. Component and PCB specifications is used with profiles for the environment the system will operate in. Such profiles include information of for example expected lifetime and operating temperatures depending on on-time schedules (Hu and Garfinkel, 1998). One of such profile is the NEDC (New European Driving Cycle) that is used in Europe to assess the fuel consumption and exhaust emission certification. Despite the name the cycle is old and is due to slow transients and long idling periods difficult for a consumer to mimic. A newer profile is the Worldwide harmonized Light duty driving Test Cycle (WLTC) which better reflect an average driving cycle (Tutuianu et al, 2013). A guess from such a driving cycle, where speed profiles and average speeds are given, is that the average engine speed is around 2500 rpm. The total number of lifetime cycles for each component depends on the application, for example a capacitor in a switching circuit at 250 kHz will endure more cycles than a valve actuator at less than 100 Hz during a driving cycle.
During its lifetime, the rail and VCS have to be possible to remove from the engine for maintenance. The rail must be robust enough to not break during handling, including accidental mechanical shocks. If the rail would break down it’s favorable to be able to disassemble it to change the broken parts. If the PCB or an electrical component fails the safest thing is to replace the entire PCB to preserve an intact surface treatment.
2.3 Previous valve control system
The previous VCS is a modular system with add on cards controlling 4 actuators each. New add on cards can be added to extend the VCS capability up to 24 actuators, corresponding to a six-cylinder engine. The system is easy to use with different engine configurations and is well suited for measurements and development of the VCS. The system includes all functionality necessary including actuator control, ECU communication through CAN (Controller Area Network), programming port to a computer and diagnostics. Due to the number of add on ports some of them are left unused when operating a four-cylinder engine. This leads to an unnecessary bulky and costly system for further testing in an engine compartment of a car. The add on cards and connectors also have a doubtful robustness since they are not designed to operate in the harsh environments of the car.
2.4 Valve timing concepts
In order to design a Fully Variable Valve Timing system care has to be taken to not design any unintended limitations in either hardware or software. Here follow some valve timing concepts to keep in mind during the design process.
Regular four-stroke operation is the simplest operation in the sense that the timing and ignition order of a regular camshaft can be duplicated. The only difference is the possibility of longer valve opening due to the quick opening and closing time of the valve. This requires that the valves can be operated in pairs. The next step is different timing for the two intake valves in one cylinder to enable precise control of air charge mixing, tumble and swirl. Different exhaust valve timings are also useful for DEP (Divided Exhaust Port) with several benefits among low exhaust back pressure and quick catalyst warm up. This operation requires that all the valves can be operated individually.
Running in two-stroke operation is possible to get a higher torque output. By operating one pair of valves for each of two cylinders firing in parallel in an actuator rail the torque output can be increased roughly by 20% in a gasoline engine and 50% in a diesel engine. Since the diesel engine is not lambda regulated a higher torque is easier to achieve compared to the gasoline engine. This two-stroke operation require that the VCS manage four valves at the time.
A simple way of increase the efficiency of the engine is to deactivate cylinders to increase the loading of firing cylinders. Even though the engine is not throttled there is some room for improvement in thermal efficiency. Cylinder deactivation is possible by simply shutting off the valves to the cylinders which are to be deactivated. By deactivating the cylinders instead of only shutting off the fuel no air is pumped through the engine, which ensures proper three-way catalyst operation (Wilcutts et al, 2013). Instead of one or more designated cylinders to shut off they can all be running eight-stroke, like a four-stroke engine with every second cycle deactivated, or in any other similar way. This ensure that all cylinders keep operating temperature and wear evenly. This operation is no different from regular four-stroke operation in terms of hardware, just an omitted signal in software.
These concepts lay the foundation to endless variations of valve timings. Although not all valves can be actuated at the same time most realistic timing concepts will be covered by actuating up to half of the valves individually and simultaneously.
3 IMPLEMENTATION
The implementation describes the design of each part of the system in the following chapter.
3.1 Requirement specification
The valve control system should:
Priority 1
• Control 8 valve actuators
• Read 8 valve position sensors
• Control 4 auxiliary coils, either fuel injectors or ignition coils
• Allow 10 000 rpm engine speed
• Allow -40ºC to +125ºC operation temperature
• Allow 6V-30V battery voltage
• Lifespan of 10 000 hours at an average of 2500 rpm
• Allow full flexibility for valve opening
• Be robust enough to use in a car
• Be possible to diagnose
• Use CAN communication Priority 2
• Be optimized for low cost
• Be optimized for small packaging dimensions
• Be optimized for low energy consumption Priority 3
• Be prepared for cylinder head integration
3.2 Prototype design
This section will describe the different parts of the valve control system and how they were developed. The parts are illustrated in Figure 6, with the flow of power to the left and signals to the right.
3.2.1 Boost voltage controller
A boost voltage controller is a device that is used to amplify a DC voltage to a higher voltage.
A simplified schematic is shown in Figure 7. The basic operation is that the MOSFET (metal oxide semiconductor field effect transistor) in the middle is closed to build up an electromagnetic field in the inductor. When the MOSFET is opened the sudden change of current through the inductor will produce a voltage spike that is harvested through the diode to the load. The boost controller will operate the MOSFET to achieve desired output voltage.
Figure 6: Valve Control System overview
The maximum engine speed in a non-performance engine, around 6 000 rpm will allow enough time for the actuator to open, but when running at 10 000 rpm all possible delays should be reduced. To shorten the opening time of the actuator an elevated voltage of 50V is applied to the actuator during opening. The boost voltage is supplied by a Texas Instruments LM5122 boost controller. It uses peak current mode control to regulate the output voltage to a level programmed with a simple voltage divider. The LM5122 was chosen because it is a high efficiency boost controller that allows a wide input voltage and automotive operating temperature. It also required relatively few external components.
The output voltage is fully adjustable and the controller can be tuned for good stability and high efficiency at light loads. The external components were calculated and chosen with help of the LM5122 data sheet to ensure proper operation with good stability and not violating any pin or controller limitations. The switching frequency was set to 250 kHz to have a good compromise between the smaller component size of a high frequency and the efficiency of a low frequency.
The efficiency is of special concern at light load operation since the boost controller is designed to keep up at maximum engine speed but will operate at lower load most of the time. Due to switching and bias current losses at low load different strategies can be implemented to reduce the losses. A diode emulation mode of the high side MOSFET is chosen together with pulse skipping operation that skips pulses, or effectively pauses the boost controller during low load operation.
The actuator boost current consumption and duration per actuator in the existing system were measured and were later extrapolated to 5A at 410μs to accommodate headroom for a future raise in current. The highest mean current is during four-stroke operation at 10 000 rpm. This will set the continuous output current capability for the boost supply, according to equation 1 and with the result of equation 2.
𝑂𝑝𝑒𝑛𝑖𝑛𝑔𝐷𝑢𝑟𝑎𝑡𝑖𝑜𝑛
𝑃𝑒𝑟𝑖𝑜𝑑𝑇𝑖𝑚𝑒 × 𝑁𝑜𝑂𝑓𝐶𝑦𝑙𝑖𝑛𝑑𝑒𝑟𝑠
𝑅𝑒𝑣𝑜𝑙𝑢𝑡𝑖𝑜𝑛𝑠𝑃𝑒𝑟𝑂𝑝𝑒𝑛𝑖𝑛𝑔×𝑁𝑜𝑂𝑓 𝐴𝑐𝑡𝑢𝑎𝑡𝑜𝑟𝑠 𝑐𝑦𝑙⁄ ×𝑐𝑢𝑟𝑟𝑒𝑛𝑡 = 𝐼 (1)
Figure 7: Simplified schematic of a boost converter
0,41𝑚𝑠 12𝑚𝑠 ×4
2×2×5𝐴 = 683𝑚𝐴 (2)
The highest momentary current is during two-stroke operation at 20A for 410μs, according to equation 3.
𝑐𝑢𝑟𝑟𝑒𝑛𝑡×𝑁𝑜𝑂𝑓 𝐴𝑐𝑡𝑢𝑎𝑡𝑜𝑟𝑠 𝑐𝑦𝑙⁄ ×𝑁𝑜𝑂𝑓𝐶𝑦𝑙𝑖𝑛𝑑𝑒𝑟𝑠𝑃𝑒𝑟𝑃𝑢𝑙𝑠𝑒
= 5𝐴×2×2×2 = 20𝐴 (3)
The two-stroke operation is only used with lower engine speeds and will have a lower mean current output than four-stroke operation. Hence the four-stroke operation sets the size of the bulk output capacitors. The boost circuit was built on an experiment card, as seen in Figure 8, to verify proper operation. After problems with instability caused by long signal paths and insufficient grounding an evaluation PCB was instead modified with the new components, Figure 9, with a satisfying result. The schematic for the boost circuit can be found in Appendix B.
Figure 8: Boost controller circuit on experiment card
3.2.2 Actuator controller
Each valve actuator is driven by a current from the actuator controller. The controller is responsible for sourcing the correct current level to every actuator with the correct timing. An opening of a valve actuator is requested by the VCS microcontroller to the NXP Semiconductors MC33816 solenoid controller. The controller receives a signal labeled Startx for each actuator when to open and will keep it open for the duration of the signal. The profile of the current is illustrated in Figure 10.
Figure 9: Boost controller circuit on a modified evaluation board
Figure 10 illustrates how the boost voltage is applied to the actuator solenoid during the peak phase to quickly open the actuator with a high current. Later during the hold phase the battery voltage is used to hold the actuator open with a lower current. The switching between voltages, the current regulation and all the timing is governed by the actuator controller. The peak phase time is dependent on the dynamic properties of the actuator with the same duration for all the openings. The remaining hold phase time is varying with the duration of the Startx signal.
The MC33816 actuator controller is commonly used to control up to six solenoids, connected in three pairs and has an integrated boost voltage supply. One of the pairs lack the boost voltage functionality and neither of the solenoids in a pair can be used at the same time since they share the same current sensing resistor. The current sensing resistor is used by the actuator controller to source the correct current level to the actuator. In this case it will be necessary to operate all actuators individually in order to run different valve timing concepts and use different timing
Figure 10: Typical actuator current profile. Source: NXP, Four Injector and Fuel Pump Drive, 2014
This enables four actuators to be individually controlled by each MC33816 controller chip, hence two controllers are needed to run a complete actuator rail. Due to an excess of low side MOSFET drivers, some of them were used with a small N-channel MOSFET to switch a P- channel MOSFET to allow high side switching. As a result, no MOSFET driver was left unused.
To test the rearranged design an KIT33816AEEVM evaluation board was rebuilt and extended with all the new connections and components to fulfill the requirements. The design is shown in Figure 12 and the schematics can be found in Appendix A.
Figure 12: Actuator controller Figure 11: Actuator voltage switching
The green card is the modified evaluation board with the actuator controller chip and the remaining components. To the left is the cabling to the microcontroller and to the right is the connectors to the actuators. The yellow experiment card contains all the components that did not find room on the evaluation board, such as the boost voltage switches at the top and a few extra components. After all components were fitted the assembly was tested to verify proper operation.
3.2.3 Diagnostics
The diagnostics is based on connections of the source pin of the high side MOSFETs and the drain pin of the low side MOSFETs to the microcontroller. With other words, the two connections to the load. By measuring the upper and lower voltage and comparing them to the boost or battery voltage and the ground respectively it can be concluded if each MOSFET is switching as intended. If not, proper action must be taken, for example shut off the errant actuator. A small current, too small to trigger the actuator, can be sourced through the load and depending on the voltage in each measurement it can be concluded if the load is connected properly without short or open circuits.
3.2.4 Power input protection
To protect the VCS from damage due to over voltage or reverse battery voltage a circuit based on the Maxim Integrated MAX6496 over voltage limiter controller was implemented. This device has an adjustable threshold for the input voltage over which it regulates the output voltage like a simple buck controller to not exceed the threshold. This ensures safe system functionality in the event of an over voltage transient, for example a load dump when the battery is disconnected with the generator running. In the event of a reverse battery voltage the load will be isolated to prevent damage. The input voltage range is expressed in Figure 13 by the referenced application note by Littelfuse.
capacity at maximum ambient temperature. The evaluation board with the MAX6496 is shown in Figure 14.
The use of a MOSFET for reverse polarity protection instead of a regular diode admits a lower voltage drop and better system efficiency. The speed and precision of the active over voltage protection is superior any passive component.
3.2.5 Fuel injectors and ignition coils
In addition to the valve actuator drivers it would be convenient to house the drivers to the fuel injectors and ignition coils in the VCS according to the requirement specification. Even though each system will control either of fuel or ignition, the drivers to both has to be implemented in order to have identical systems packaging for lower cost and fewer unique parts. The solenoid in the fuel injector is similar to the solenoid in the valve actuator and is controlled with the same peak and hold current control and uses the same diagnostics. Hence one actuator controller can be used for the fuel injectors with minor changes in timing to account for the dynamic properties of the fuel injector solenoid. This means there will be three MC33816 in the VCS including the two for the valve actuators.
The ignition system control depends on what kind of ignition coils are used. Either with a built- in driver or without. In the case of a built-in driver there has to be a small circuit interfacing the driver with the microcontroller. The other case without built in driver is a bit more complicated where an IGBT (insulated-gate bipolar transistor) with suitable voltage rating and sufficient cooling is required for all ignition coils.
Figure 14: Power input protection
3.2.6 Microcontroller
The currently used microcontroller NXP MPC5644A communicates with the ECU via CAN and controls the VCS according to the demands from the ECU. The position of the valve is constantly monitored by the microcontroller through the valve position sensor. The reading of the sensor is done digitally without A-D conversion which allows real time monitoring with an accuracy of 0,1 mm. The processor communicates with the actuator controller via a pin to each actuator that is pulled high when an actuator opening is requested and pulled low when the actuator closing is requested. The processor manages the diagnostics for the actuators to detect short circuits and open circuits and will send a flag to the ECU if any errors occur. The microcontroller is also flashing the software to the actuator controller upon start up in order to start the controller properly. The operation of the microcontroller and its peripherals in detail is not within the scope of this thesis.
3.2.7 New microcontroller
During the project, a new microcontroller NXP MPC5746R was discovered that was similar to the existing microcontroller but included most of the functionality of the actuator controller.
This results in that the actuator controller can be removed by adding some extra functionality in the microcontroller, for example the diagnostics of the actuators. This will give a simpler system without the actuator controller and the separate software associated. Instead the software functions can be included in the microprocessor software. There will also be room to operate the fuel injectors and ignition coils directly from the microcontroller. The benefits of using the new microcontroller were so many that the remainder of the project focused on preparations for a transition to the new processor. Since the scope of the thesis does not cover the microcontroller this will instead be of high priority in the future work section.
4 RESULTS
In this section, the result of the literature study is presented along with the developed prototype with the knowledge from the study implemented where applicable. The prototype is also validated against the requirement specifications.
4.1 Literature study
Design of an electric system intended to operate in the harsh environment of an engine compartment was as expected a non-trivial task. Luckily a lot of research has been done and the industry has many solutions to the problems in this type of environment.
Even if all components used allow -40ºC to +125ºC there should be carefully designed heat flow paths to keep the temperature rise from internal heating low to avoid violation of the temperature limits. This will allow high engine compartment temperatures. Efficient heat flow paths are essentially a low thermal resistance path, a combination of high thermal conductivity and a large conduction area in every transition. Large copper planes, many copper plated vias between planes, a PCB coating with low thermal resistance and external cooling are elements to a low and evenly spread circuit temperature. The temperature will differ with different driving conditions so low thermal resistance is important to reduce hotspots from a sudden increase in load and power dissipation. When temperature is cycled, sufficient compliance is needed in component joints to accommodate movement due to different CTE. Too stiff joints will cause damage on joints or components.
To prevent damage from vibrations a stiff and properly fastened PCB is needed and that any cabling or large component is properly secured to keep deflection amplitudes small. Large deflections will shake loose components or crack joints. By mounting large components close to PCB mounting locations and make use of moulding or potting will help reduce deflection of the PCB. The system must be properly sealed with an electrically conductive casing in order to keep moisture and contamination on the outside and to avoid EMC problems. Internal EMC is already considered when designing the PCB but some shielding is useful in order to not interfere with external sources or victims outside of the system causing any unwanted behavior. If the system is properly designed to emit low levels of EMI and robust enough to not be affected by other EMI sources the conducting layer may be omitted.
The lubrication oil must, from an electric perspective, be slightly electrically conductive. Not too much to cause any shortcuts but slightly to not build up any electrostatic charge. A discharge of such can potentially damage the electronic system causing very unpredictable errors that is hard to diagnose. To protect the circuits from oil, corrosives and other pollutants surface coating is widely used. Silicone based moulding material, conformal coating or potting encapsulation would be well suited for the VCS with its flexibility and good thermal cycling properties. It is easy to apply and also possible to remove in order to change components. If potting of the system is used only EMC is of larger concern since the potting provides good mechanical and thermal stability.
The lifetime of the system is equal to the lowest lifetime of either any of the components in the system or the PCB assembly. The lifetime of components is often stated by the manufacturer but the assembly should be simulated to ensure that sufficient lifetime can be expected. This can be done with different simulation methods with different accuracy and level of detail. If the system would fail preterm the diagnostics must be designed well enough to take early and accurate actions to limit hazardous consequences. The system should also be possible to remove for repair or replacement.
4.2 Prototype
The finished prototype is shown in Figure 15.
The parts of the VCS designed within the scope of the thesis is the two actuator controller cards
Figure 15: The Valve Control System with test actuators connected. microcontroller not shown
frame of reference. A software limit in the actuator controller to ensure this is suggested in future work.
4.2.1 Boost power supply
The benefits of pulse skipping operation can be seen in Figure 16 where the efficiency increase at low load compared to a typical regulator without it as exemplified in Figure 17.
Figure 16: LM5122 Boost voltage supply efficiency
Figure 17: Typical boost regulator efficiency
The output current is proportional to the engine speed and for the design case of an actuator consuming 5A for 410μs the lowest efficiency of 78% correspond to 1500rpm and the maximum efficiency is 89% at 940rpm and 5690rpm.
4.3 Validation
The boost power supply regulates the 50V boost voltage in a proper way throughout the input voltage range as expected according to Figure 18.
Figure 18: Boost supply, 50V red output voltage graph, 6V-30V blue input voltage graph
The boost output voltage is properly regulated when the load is changed as seen in Figure 19
Figure 19: Boost supply, 50V blue output voltage graph, 0-0,8A red load current graph [0,2A/div]
During an experiment in room temperature at full load the temperature of the hottest components on the experiment card is depicted in Figure 20.
The operation of the input voltage protection is seen in Figure 21. The output voltage in red followed the input voltage in blue within the limits of 0V to 17V. The load got isolated when a
Figure 20: LM5122 Boost voltage supply temperature
reverse voltage was applied and an input voltage above the threshold was regulated down to the threshold voltage.
The actuator controller is verified that the boost and battery voltage is properly switched to the valve actuators to achieve the correct current profile as described by Figure 10. The actuators are smoothly following different engine speeds according to signals from the microcontroller.
The current to the actuators is shown in Figure 22.
Figure 21: Input voltage protection, red output voltage, blue input voltage
The actuator current without the boost voltage supply connected is shown in Figure 23, for reference. Notice the lower maximum current and the slower increase in current, in the beginning, and decrease in current at the end.
Figure 22: Actuator current [100mA/div]
Figure 23: Actuator current without boost voltage applied [100mA/div]
Since this system is a prototype and not the final product for operation in a car the requirements of robustness were not verified. The operating temperature requirement was used for design but was not verified for practical reasons.
5 DISCUSSION
A discussion is presented below which mention the outcome of the thesis, its implementation and potential hazards involved with the result.
5.1 Valve control system
The parts of the valve control system within the scope of the project were successfully designed.
As stated in the requirement specification the system is with an external microcontroller able to control 8 valve actuators at a speed corresponding to an engine speed of 10 000 rpm. The components in the system are chosen to withstand temperatures from -40ºC to +125ºC and a battery voltage of 6V to 30V including high voltage transients that may occur in the automotive environment. The design lifetime is at least 10 000h that applies to all components and the ability to diagnose the system is implemented. The hardware circuits admit full flexibility to open the valves for all useful operating cycles. The drivers to the auxiliary coils for fuel and ignition are discussed but only the circuits to the fuel injectors are designed since the type of ignition coils were not yet decided.
Since the deliverable is still a prototype that is not ready to be put in a car some design considerations from the frame of reference do not apply yet. For example, considerations of robustness, surface treatment and sealing. However, those are important to consider early in the design process to reduce the risk of run into problems later on. Many design considerations do apply to the prototype and special care has been taken that all components fitted are rated for the temperatures they might be subject to. The lowest temperature range used is -40ºC to +125ºC junction temperature. Extra care was shown to the electrolytic capacitors where long useful life in high temperatures are rare.
5.2 Boost supply
Although the high and low side MOSFET switches are the hottest components in the configuration of the experiment in Figure 10 the boost controller chip has the smallest headroom to its maximum junction temperature. In this configuration, this roughly translates to a maximum engine compartment temperature of 100ºC but a PCB design with more efficient cooling will allow higher ambient temperatures.
This application puts high demands on the boost power supply with a maximum voltage amplification of 8 times and a wide output current range. It is unfortunate that the lowest efficiency, despite the pulse skipping strategy, is in the range of frequent operation but it is necessary in order to cover the whole operating range. This is in perfect analogy with the cam shaft issue the FreeValve system is aiming to solve.
5.3 Ethics
The required lifetime in this case is 10000 hours. If a system would fail, regardless of within or beyond the expected lifetime the hazards of a failure must be assessed. For a VCS, there is a potential risk of lethal consequences if the vehicle behaves unpredictably at high speed. Hence any failure should be detected by the diagnostics, identified and take a fast and accurate action.
At the same time alert the driver. The main motivation to the FreeValve system is to reduce emissions and improve performance in cars, but also the redundancy of many individual valve actuators and the high fault tolerance it yields when simply switching off defective ones without damaging the engine is possible. The positive long term environmental impact and reliability will balance the potential risks.
5.4 Scope of thesis
During the planning phase of the thesis there were many uncertainties as to what parts would need to be included in the system and how long the design of each part would take. The recent and earlier achievements within the company were studied along with the direction of current projects in order to formulate a requirement specification for the VCS which was up to date and heading into the future in the right direction. Since the thesis was part of an ongoing project at the company some flexibility in the project plan was needed in order to contribute in a meaningful way. Early in the project it was realized that the scope of the thesis was too wide for the time available and was narrowed down in discussion with the supervisors according to the project plan. Since priority 2 and 3 in the requirement specification apply to all parts in the system rather than being individual deliverables a few tasks were left out from priority 1. The microcontroller and its peripherals such as ECU communication and valve position sensor reading circuits were moved outside the scope of the thesis and are subjects for future work.
This also concluded that no final prototype would be built to operate in a car. Later during the thesis when the new microcontroller was discovered the need for the actuator controller were changed. The decision was made to finish the rebuilt design in preparation for a smooth transition to the new microcontroller. No software was written for the old microcontroller, which anyway would end up unused.
6 Conclusions
The parts of the FreeValve valve control system in the scope of this thesis has been designed to operate under the conditions given in the requirement specifications with the main task to deliver the correct current profile to the valve actuators. The profile shape and timing varies with operating condition of the engine and the interspacing of profiles change with the engine speed. The electrical circuits of the design are well functioning but from the literature study it is empathized that the hardware design and protection from the automotive environment is of huge importance for a reliable operation throughout the expected lifetime of the system.
Suggestions are made on design considerations to ensure reliable operation. Of special concern is the thermal cycling which tend to strain components and solder joints. Efficient thermal paths in the PCB and protective coating help reduce the problem. The importance of a mechanically stable PCB and assembly to resist vibration and the internal and external issues of EMC are also discussed.
7 FUTURE WORK
Since the scope of this thesis was narrowed down, the first task for future work is to finish the design of the VCS and manufacture the hardware for installation in a car. The parts that remain are to move the actuator controller circuits from the old solenoid controller to the new microcontroller and design of the sensor reading circuits and ignition system interface. The software has to be written to control all parts of the microcontroller including the valve actuators, fuel injection and ignition, ECU communication, diagnostics and valve position sensor reading.
When designing the software to the actuator controllers, care must be taken to not program the actuation of more than half of the valve actuators simultaneously for any kind of operation.
Preferably a safety function would be implemented to avoid this from happening.
When the VCS is mounted in the car the exciting work begins to tune all engine parameters for optimum performance, low fuel consumption, low emissions and a smooth and enjoyable driving experience.
8 REFERENCES
List of references
Sasa Trajkovic, “The Pneumatic Hybrid Vehicle”, Lund University, Lund, 2010
FreeValve AB, “Fuel Consumption Reduction”, 2016,
http://www.freevalve.com/technology/fuel-consumption-reduction/, Accessed 2016-09-01 Yu Wan, Aimin Du, “Reducing Part Load Pumping Loss and Improving Thermal Efficiency through High Compression Ratio Over-Expanded Cycle”, SAE International, 2013
FreeValve AB, “Questions and Answers”, 2016,
http://www.freevalve.com/technology/questions-and-answers/, Accessed 2016-09-01
FreeValve AB, “FreeValve Technology”, 2016,
http://www.freevalve.com/technology/freevalve-technology/, Accessed 2016-09-04
R. Wayne Johnson, John L. Evans, Peter Jacobsen, James R. Thompson, Mark Christopher,
“The Changing Automotive Environment: High-Temperature Electronics”, IEEE Transactions on Electronics Packaging Manufacturing, vol 27, no 3, 2004, pp 164-176
Pradeep Lall, Mohd Nokibul Islam, John Evans, Jeffrey C. Suhling, Tushar Shete, “Damage Mechanics of Electronics on Metal-Backed Substrates in Harsh Environments”, IEEE Transactions on Components and Packaging Technologies, vol 29, no 1, 2006, pp 204-212 Ron S. Li, “Failure Mechanisms of Ball Grid Array Packages Under Vibration and Thermal Loading”, SAE International, 2004
J. Pippola, T. Marttila, L. Frisk, “Protective coatings of electronics under harsh thermal shock”, Microelectronics Reliability, vol 54, no 9-10, 2014, pp 2048-2052
Sangyoung Park, Soohee Han, Naehyuck Chang, “Control-Theoretic Dynamic Thermal Management of Automotive Electronics Control Units”, IEEE Journal on Emerging and Selected Topics in Circuits and Systems, vol 1, no 2, 2011, pp 102-108
Ehsan Parsa, Hao Huang, Abhijit Dasgupta, “Multi-physics Simulations for Combined Temperature/Humidity Cycling of Potted Electronic Assemblies”, 14th International Conference on Thermal, Mechanical and Multi-Physics Simulation and Experiments in Microelectronics and Microsystems, EuroSimE 2013, 2013, pp 1-6
T.J. Harvey, R.J.K. Wood, G. Denuault, H.E.G. Powrie, “Investigation of electrostatic charging mechanisms in oil lubricated tribo-contacts”, Tribology International, vol 35, no 9, 2002, pp 605-614
Ölchecker, “Electrostatic discharges in hydraulic oils and lubricants”, Oelcheck, Spring 2012, pp. 6-7, 2012
Christian Goth, Jörg Franke, Andreas Reinhardt, Philipp Widemann, “Reliability of Molded Interconnect Devices (MID) Protected by Encapsulation Methods Overmolding, Potting and Coating”, 2012 7th International Microsystems, Packaging, Assembly and Circuits Technology Conference, 2012, pp 137-140
Tim Williams, “EMC for Product Designers”, Elsevier Ltd, ISBN: 978-0-7506-8170-4, chapter 13, pp. 341-374, 2007
Alirio Cavalcanti de Brito, Marcelo Lopes de Oliveira e Souza, “A Discussion on the Methods of Thermal Cycling and Power Cycling for Reliability Prediction of Solder Joints of Electronic Components”, SAE International, 2015
Jimmy M. Hu, George A. Garfinkel, “Correlation of Thermal Cycle Tests to Field Usage Profiles for Solder Joints in Automotive Electronics”, SAE International, 1998
Monica Tutuianu, Alessandro Marotta, Heinz Steven, Eva Ericsson, Takahiro Haniu, Noriyuki Ichikawa, Hajime Ishii, “Development of a World-wide Worldwide harmonized Light duty driving Test Cycle (WLTC)”, Transportation Research Part D, vol 40, pp 61-75, 2013
Mark Wilcutts, Joshua Switkes, Mark Shost, Adya Tripathi, “Design and Benefits of Dynamic Skip Fire Strategies for Cylinder Deactivated Engines”, SAE International Journal of Engines, vol 6, no 1, 2013, pp 278-288
List of data sheets and application notes
Chipcon Products, “LC filter with improved high-frequency attenuation”, 2003 Diodes Incorporated, “BSS123Q-7 Datasheet”, 2016
EDN Network, “Controlling switch-node ringing in DC/DC converters”, 2016 Freescale Semiconductor, “KIT33816AEEVM Evaluation Board Datasheet”, 2014 Freescale Semiconductor, “MC33816 Datasheet”, 2015
