You are on page 1of 29

VISVESVARAYA TECHNOLOGICAL UNIVERSITY, BELGAUM

A technical seminar report on

kCAMLESS ENGINEl
Submitted in partial fulfillment as per VTU curriculum for VIII semester

Bachelor of Engineering In Mechanical Engineering Of Visveswaraiah Technological University, Belgaum


Submitted by

RISHI RAJ TOMAR [1DS06ME088]

2009-2010
Department of Mechanical Engineering DAYANANDA SAGAR COLLEGE OF ENGINEERING BANGALORE-560078

VISVESVARAYA TECHNOLOGICAL UNIVERSITY

DAYANANDA SAGAR COLLEGE OF ENGINEERING Department of Mechanical Engineering Bangalore 560078

CERTIFICATE
This is to certify that the Technical Seminar report entitled CAMLESS ENGINE has been presented by Mr. RISHI RAJ TOMAR, USN: 1DS06ME088 in partial fulfillment for the award of Bachelor of Engineering in Mechanical Engineering of the Visvesvaraya Technological Universit y, Belgaum during the year 2009-10. It is certified that all correction/suggestions indicated for internal assessment have been incorporated in the report deposited in the department library.

Date:

Sign of HOD/Coordinator

Internal Examiners: Name & sign. 1. 2.

ACKNOWLE

E ENT

It is with great satisfaction and euphoria that I am submitting the Technical ENGINE. I have completed it as a part of the seminar report on CAMLE curriculum of our university. I would like to thank H.O.D Dr. K.J.Sudhakar, Prof. C P S Prakash for his support and guidance in assistance provided at every stage. I would extremely thank my Principal Dr. Netaji Ganesan, DSCE, Bangalore who has always been a great source of inspiration & has encouraged me all through.

RISHI RAJ TOMAR

ABSTRACT
The cam has been an integral part of the IC engine from its invention. The cam controls the breathing channels of the IC engines, that is, the valves through which the fuel air mi ture (in SI engines) or air (in CI engines) is supplied and exhaust driven out. Beside by demands for better fuel economy, more power, and less pollution, motor engineers around the world are pursuing a radical camless design that promises to deliver the internal combustion engine s biggest efficiency improvement in years. The aim of all this effort is liberation from a constraint that has handcuffed performance since the birth of the internal combustion engine more than a century ago. Camless engine technology is soon to be a reality for commercial vehicles. In the camless valvetrain, the valve motion is controlled directly by a valve actuator - there s no camshaft or connecting mechanisms. Precise electronic circuit controls the operation of the mechanism, thus bringing in more flexibility and accuracy in opening and closing the valves. The seminar looks at the working of the electronically controlled camless engine with electro-mechanical valve actuator, its general features and benefits over conventional engine.

The engines powering today s vehicles, whether they burn gasoline or diesel fuel, rely on a system of valves to admit fuel and air to the cylinders and to let exhaust gases escape after combustion. Rotating steel camshafts with precision-machined eggshaped lobes, or cams, are the hard-tooled brains of the system. They push open the valves at the proper time and guide their closure, typically through an arrangement of pushrods, rocker arms, and other hardware. Stiff springs return the valves to their closed position.

CONTENTS
j INTRODUCTION j WORKING OF PUSH ROD ENGINE

5 6

j Cranshaft j Camshaft j Working j AN OVERVIEW OF CAMLESS ENGINE


Camless Valve Train Hydraulic Pendulum Valve Openinig & Closing Valve Motion Control Unequal Lift Modifier Cylinder Head

j COMPONENTS OF CAMLESS ENGINE j Engine Valve j Hydraulic Pump


Soleniod Valve High Pressure pump Low Pressure Pump Cool Down Accumulator

20

j ADVANTAGES

2
j CONCLUSION

31
j BIBLOGRAPHY

32

CAMLESS ENGINE
INTRODUCTION
The cam has been an integral part of the IC engine from its invention. The cam controls the breathing channels of the IC engines, that is, the valves through which the fuel air mixture (in SI engines) or air (in CI engines) is supplied and exhaust driven out. Besieged by demands for better fuel economy, more power, and less pollution, motor engineers around the world are pursuing a radical camless design that promises to deliver the internal combustion engines biggest efficiency improvement in years. The aim of all this effort is liberation from a constraint that has handcuffed performance since the birth of the internal-combustion engine more than a century ago. Camless engine technology is soon to be a reality for commercial vehicles. In the camless valve train, the valve motion is controlled directly by a valve actuator theres no camshaft or connecting mechanisms .Precise electrohydraulic camless valve train controls the valve operations, opening, closing etc. The seminar looks at the working of the electrohydraulic camless engine, its general features and benefits over conventional engines. The engines powering todays vehicles, whether they burn gasoline or diesel fuel, rely on a system of valves to admit fuel and air to the cylinders and let exhaust gases escape after combustion. Rotating steel camshafts with precision -machined eggshaped lobes, or cams, are the hard-tooled brains of the system. They push open the valves at the proper time and guide their closure, typically through an arrangement of pushrods, rocker arms, and other hardware. Stiff springs return the valves to their closed position. In an overhead camshaft engine, a chain or belt driven by the crankshaft turns one or two camshafts located atop the cylinder head. A single overhead camshaft (SOHC) design uses one camshaft to move rockers that open both inlet and exhaust valves. The double overhead camshaft (DOHC), or twin-cam, setup does away with the rockers and devotes one camshaft to the inlet valves and the other to the exhaust valves.

WORKING OF PUSH ROD ENGINE


Pushrod engines have been installed in cars since the dawn of the horseless carriage. A pushrod is exactly what its name implies. It is a rod that goes from the camshaft to the top of the cylinder head which push open the valves for the passage of fuel air mixture and exhaust gases. Each cylinder of a pushrod engine has one arm (rocker arm) that operates the valves to bring the fuel air mixture and another arm to control the

valve that lets exhaust gas escape after the engine fires. There are several valve train arrangements for a pushrod.

Crankshaft
Crankshaft is the engine component from which the power is taken. It receives the power from the connecting rods in the designated sequence for onward transmission to the clutch and subsequently to the wheels. The crankshaft assembly includes the crankshaft and bearings, the flywheel, vibration damper, sprocket or gear to drive camshaft and oil seals at the front and rear.

Camshaft
The camshaft provides a means of actuating the opening and controlling the period before closing, both for the inlet as well as the exhaust valves, it also provides a drive for the ignition distributor and the mechanical fuel pump. The camshaft consists of a number of cams at suitable angular positions for operating the valves at approximate timings relative to the piston movement and in the sequence according to the selected firin g order. There are two lobes on the camshaft for each cylinder of the engine; one to operate the intake valve and the other to operate the exhaust valve.

Working
When the crank shat turn the cam shaft the cam lobs come up under the valve lifter and cause the lifter to move upwards. The upward push is carried by the pushrods through the rocker arm. The rocker arm is pushed by the pushrod, the other end moves down. This pushes down on the valve stem and cause it to move down thus opening the port. When the cam lobe moves out from under the valve lifter, the valve spring pulls the valve back upon its seat. At the same time stem pushes up on the rocker arm, forcing it to rock back. This pushes the push rods and the valve lifter down, thus closing the valve. The figure-1,2 shows cam-valve arrangement in conventional engines

the timing of the engine is dependent on the shape of the cam lobes and the rotational velocity of the camshaft, engineers m st make decisions early in the automobile development process that affect the engines performance. The resulting design represents a compromise between fuel efficiency and engine power. Since maximum efficiency and maximum power require unique timing characteristics, the cam design must compromise between the two extremes. This compromise is a prime consideration when consumers purchase automobiles. Some individuals value power and lean toward the purchase of a high performance sports car or towing capable trucks, while others value fuel economy and vehicles that will provide more miles per gallon. Recognizing this compromise, automobile manufacturers have been attempting to provide vehicles capable of cylinder deactivation, variable valve timing (VVT), or variable camshaft timing (VCT). These new designs are mostly mechanical in nature. Although they do provide an increased level of sophistication, most are still limited to discrete valve timing changes over a limited range.

Sinc

Figure-1 Single cam and valve mechanism

Figure-2 conven ional valve train

AN OVERVIEW OF CAMLESS ENGINE


To eliminate the cam, camshaft and other connected mechanisms, the Camless engine makes use of three vital components the sensors, the electronic control unit and the actuator

SENSORS

ELECTRONIC CONTROL UNIT

ACTUATORS

Mainly five sensors are used in connection with the valve operation. One for sensing the speed of the engine, one for sensing the load on the engine, exhaust gas sensor, valve position sensor and current sensor. The sensors will send signals to the electronic control unit. The electronic control unit consists of a microprocessor, which is provided with a software algorithm. The microprocessor issues signals to the solid-state circuitry based on this algorithm, which in turn controls the actuator, to function according to the requirements.

Camless valve train


In the past, electro hydraulic camless systems were created primarily as research tools permitting quick simulatio n of a wide variety of cam profiles. For example, systems with precise modulation of a hydraulic actuator position in order to obtain a desired engine valve lift versus time characteristic, thus simulating the output of different camshafts. In such systems the issue of energy consumption is often unimportant. The system described here has been conceived for use in production engines. It was, therefore, very important to minimize the hydraulic energy consumption.

1. Electromechanical Poppet Valves


This type of system uses an armature attached to the valve stem.The outside casing contains a magnetic coil of some sort that can be used to either attract or repel the armature, hence opening or closing the valve.

Most early systems employed solenoid and magnetic attraction/repulsion actuating principals using an iron or ferromagnetic armature. These types of armatures limited the performance of the actuator because they resulted in a variable air gap. As the air gap becomes larger (ie when the distance between the moving and stationary magnets or electromagnets increases), there is a reduction in the force. To maintain high forces on the armature as the size of the air gap increases, a higher current is employed in the coils of such devices. This increased current leads to higher energy losses in the system, not to mention non-linear behaviour that makes it difficult to obtain adequate performance. The result of this is that most such designs have high seating velocities (ie the valves slam open and shut hard!) and the system cannot vary the amount of valve lift. The electromechanical valve actuators of the latest poppet valve design eliminate the iron or ferromagnetic armature. Instead it is replaced with a current-carrying armature coil. A magnetic field is generated by a magnetic field generator and is directed across the fixed air gap. An armature having a current-carrying armature coil is exposed to the magnetic field in the air gap. When a current is passed through the armature coil and that current is perpendicular to the magnetic field, a force is exerted on the armature.When a current runs through the armature coil in either direction and perpendicular to the magnetic field, an electromagnetic vector force, known as a Lorentz force, is exerted on the armature coil. The force generated on the armature coil drives the armature coil linearly in the air gap in a direction parallel with the valve stem. Depending on the direction of the current supplied to the armature coil, the valve will be driven toward an open or closed position. These latest electromechanical valve actuators develop higher and bettercontrolled forces than those designs mentioned previously. These forces are constant along the distance of travel of the armature because the size of the air gap does not change.

The key component of the Siemens-developed infinitely variable electromechanical valve train is an armature-position sensor. This sensor ensures the exact position of the armature is known to the ECU at all times and allows the magnetic coil current to be adjusted to obtain the desired valve motion.

Referring now to Figures 1 to 4, an electromechanical valve actuator of the poppet valve variety is illustrated in conjunction with an intake or exhaust valve (22). The valve (22) includes a valve closure member 28 having a cylindrical valve stem (30) and a cylindrical valve head (32) attached to the end of the stem (30). The valve actuator (20) of the poppet valve system generally includes a housing assembly (34) consisting of upper and lower tubular housing members (36) and (42), a magnetic field generator consisting of upper and lower field coils (48) and (52), a core (56) consisting of upper and lower core member (58) and (68), and an armature (78) suitably connected to the valve stem (30). The armature coil is preferably made from aluminium wire or other electrically conductive lightweight material, which is highly conductive for its mass. Minimising the armature mass is especially important in view of the rapid acceleration forces placed on it in both directions. The ability of the electromechanical valve actuator to generate force in either direction and to vary the amount of force applied to the armature in either direction is an important advantage of this design. For instance, varying the value of the current through the armature coil and/or changing the intensity of the magnetic field can control the speed of opening and closing of the valve. This method can also be used to slow the valve closure member to reduce the seating velocity, thereby lessening wear as well as reducing the resulting noise.

This system is able to operate without valve springs as shown in Figure 1 or can equally be equipped with them as shown in Figures 6 & 7. Siemens report that a special software algorithm is used to control the actuator coil currents such that the valves are decelerated to a speed near zero as they land - in conjunction with a switching time of barely three milliseconds. For the valves this means minimal wear and minimum noise generation. The 16-valve four cylinder engine that is currently undergoing tests in Germany, by Siemens, is equipped with 16 valve actuators and the corresponding armature-position sensors. A Siemens ECU is used and two cable rails connect the actuators to it. A 42-volt starter-generator provides the power. An alternative to the conventional poppet valve for use in camless valve trains is a ball valve. This type of electromechanical valve system consists of a ball through which a passage passes. If the ball is rotated such that the passage lines up with other openings in the valve assembly, gas can pass through it. (Exactly like the ball valves many of us use valve is accomplished by electromagnets positioned around its exterior. to control our boost.

2. Electromechanical Ball Valves

Opening and closing the Referring to Figure 10, the valve housing (7) is shown in two pieces. Ball valve (8) has two rigidly attached pivots (12). The disc (10) is permanently attached and indexed to the ball valve and contains permanent magnets around its perimeter. The electromagnets (11) are situated on both sides of the ball valve (8) and they are fixed to the valve housing.

The electromagnets are controlled through the ECU. A crank trigger sensor on the crankshaft provides information about the position of the pistons relative to top dead centre. Thus, at top dead centre of the power stroke, the ECM could be used to fix the polarity of both electromagnets so that they are of opposite polarity to the magnets in the ball valve, rotating the ball valve to the closed position. The substitution of a simple, efficient ball valve and valve housing arrangement in a a four stroke reciprocation piston engine eliminates all the independent moving parts in the valve train. This may even be an improvement over the poppet valve camless system - the ball valve needs only to rotate on its axis to achieve the desired flow conditions, rather than be accelerated up and down in a linear fashion. A partially open ball valve state may also be able to be used to create more turbulence.

Electromechanical valve train implementation would not be possible with a normal 12V electrical system. As has been covered previously in AutoSpeed ("Goodbye 12 volts... hello 42 volts!"), the automotive industry has chosen a 42V electrical system as the next automotive standard. Consequently, the energy demand of EMVT can be optimally matched by a crankshaft-mounted starter-generator (KSG - in Siemens speak) operating at 42V; it is integrated in the flywheel and designed for the starting process as well as generator operation.

Electrohydraulic Poppet Valves


In general terms, present designs of electrohydraulic valves comprise poppet valves moveable between a first and second position. Used is a source of pressurised hydraulic fluid and a hydraulic actuator coupled to the poppet valve. The motion between a first and second position is responsive to the flow of the pressurised hydraulic fluid. An electrically

operated hydraulic valve controls the flow of the pressurised hydraulic fluid to the hydraulic actuator. In one design, the provision is made for a three-way electrically operated valve to control the flow of the pressurised hydraulic fluid to the actuator. This supplies pressure when electrically pulsed open, and dumps actuator oil to the engine oil sump when the valve is electrically pulsed to close. The use of engine oil as the hydraulic fluid simplifies and lowers the cost of the design by removing the need for a separate hydraulic system.

The basic design of the electrohydraulic valvetrain hardware is illustrated in Figure 11. The engine poppet valves (22) and the valve springs (24) that are used to reset them are shown. The poppet valves are driven by hydraulic actuators (26), which are controlled by electrically operated electro-hydraulic valves (28) supplying hydraulic fluid to the actuators via conduit (29). The preferred hydraulic fluid is engine oil, supplied to the electro-hydraulic valves by the pressure rail (30). An engine-driven hydraulic pump (32) supplies the oil pressure, receiving the oil from the engine oil sump (34). The pump output pressure is also limited by an unloader valve (36), as controlled by an accumulator (38) connected to the oil pressure rail. With this design the hydraulic pump could be periodically disconnected, such as under braking, so that the valve train would run off the stored accumulator hydraulic pressure. As is the trend with all modern engine systems, the camless engine has an even greater reliance on sensors. The valve actuation and control system typically needs a manifold pressure sensor, a manifold temperature sensor, a mass flow sensor, a coolant temperature sensor, a throttle position sensor, an exhaust gas sensor, a high resolution engine position encoder, a valve/ignition timing decoder controller, injection driver electronics, valve coil driver electronics, ignition coil driver electronics, air idle speed control driver electronics and power down control electronics.

A valve developed by Sturman Industries is said to be about six times faster than conventional hydraulic valves. To achieve such speeds, it uses a tiny spool sandwiched between two electrical coils. By passing current back and forth between the coils, a microprocessor -based controller can quickly move the spool back and forth, thereby actuating the engine valves in accordance. However, electrohydraulic systems are mostly being developed for diesel truck use because it is currently not clear whether the technology will have the speed needed for higher revving passenger car engines.

Hy raulic pen ulum


The Electro hydraulic Camless alve train, (EC ) provides continuously variable control of engine valve timing, lift, and velocity. It uses neither cams nor springs. It exploits the elastic properties of a compressed hydraulic fluid, which, acting as a liquid spring, accelerates and decelerates each engine valve during its opening and closing motions. This is the principle of the hydraulic pendulum. Like a mechanical pendulum, the hydraulic pendulum involves conversion of potential energy into kinetic energy and, then, back into potential energy with minimal energy loss . During acceleration, potential energy of the fluid is converted into kinetic energy of the valve. During deceleration, the energy of the valve motion is returned to the fluid. This takes place both during valve opening and closing. Recuperation of kinetic energy is the key to the low energy consumption of this system.. Fig ure 7 illustrates the hydraulic pendulum concept. The system incorporates high and low pressure reservoirs. A small double-acting piston is fixed to the top of the engine valve that rides in a sleeve. The volume above the piston can be connected either to a high- or a low-pressure source. The volume below the piston is constantly connected to the high-pressure source. The pressure area above the piston is significantly larger than the pressure area below the piston. The engine valve opening is controlled by a highpressure solenoid valve that is open during the engine valve acceleration and closed during deceleration. Opening and closing of a low -pressure solenoid valve controls the valve closing. The system also includes high and low-pressure check valves.

Figure 7. Hydraulic Pendulum. During the valve opening, the high-pressure solenoid valve is open, and the net pressure force pushing on the double-acting piston accelerates the engine valve downward. When the solenoid valve closes, pressure above the piston drops, and the piston decelerates pushing the fluid from the lower volume back into the high-pressure reservoir. Low-pressure fluid flowing through the low-pressure check valve fills the volume above the piston during deceleration. When the downward motion of the valve stops, the check valve closes, and the engine valve remains locked in open position. The process of the valve closing is similar, in principle, to that of the valve opening. The low-pressure solenoid valve opens, the pressure above the piston drops to the level in the low pressure reservoir, and the net pressure force acting on the piston accelerates the engine valve upward. Then the solenoid valve closes, pressure above the piston rises, and the piston decelerates pushing the fluid from the volume above it through the high-pressure check valve back into the high-pressure reservoir. The hydraulic pendulum is a spring less system. Figure 8 shows idealized graphs of acceleration, velocity and valve lift versus time for the hydraulic pendulum system. Thanks to the absence of springs, the valve moves with constant acceleration and deceleration. This permits to perform the required valve motion with much smaller net driving force, than in systems which use springs. The advantage is further amplified by the fact that in the spring less system the engine valve is the only moving mechanical mass. To minimize the constant driving force in the hydraulic pendulum the opening and closing accelerations and decelerations must be equal (symmetric pendulum).

Figure pendulum.

8.

Dynamic

characteristics

of

hydraulic

Valve opening and closing


A more detailed step-by-step illustration of the valve opening and closing process is given in Figure 9. It is a six-step diagram, and in each step an analogy to a mechanical pendulum is shown. In Step 1 the opening (highpressure) solenoid valve is opened, and the high-pressure fluid enters the volume above the valve piston. The pressure above and below the piston become equal, but, because of the difference in the pressure areas, the constant net hydraulic force is directed downward. It opens the valve and accelerates it in the direction of opening. The other solenoid valve and the two check valves remain closed. In Step 2 the opening solenoid valve closes and the pressure above the piston drops, but the engine valve continues its downward movement due to its momentum. The lowpressure check valve opens and the volume above the piston is filled with the low-pressure fluid. The downward motion of the piston pumps the high-pressure fluid from the volume below the piston back into the highpressure rail. This recovers some of the energy that was previously spent to accelerate the valve. The ratio of the high and low-pressures is selected so, that the net pressure force is directed upward and the valve decelerates until it exhausts its kinetic energy and its motion stops. At this point, the opening check valve closes, and the fluid above the piston is trapped. This prevents the return motion of the piston, and the engine valve remains fixed in its open position trapped by hydraulic pressures on both sides of the piston. This situation is illustrated in Step 3, which is the open dwell position. The engine valve remains in the open dwell position as long as necessary. Step 4 illustrates the beginning of the valve closing. The closing (low-pressure) solenoid valve opens and connects the volume above the piston with the low-pressure rail. The net pressure force is directed upward and the engine valve accelerates in the direction of closing, pumping the fluid from the upper volume back into the lowpressure reservoir. The other solenoid valve and both check valves remain closed during acceleration. In Step 5 the closing solenoid valve closes and the upper volume is disconnected from the low-pressure rail, but the engine valve continues its upward motion due to its momentum. Rising pressure in the upper volume opens the high-pressure check valve

that connects this volume with the high-pressure reservoir. The upward motion of the valve piston pumps the fluid from the volume above the piston into the high-pressure reservoir, while the increasing volume below the piston is filled with fluid from the same reservoir. Since the change of volume below the piston is only a fraction of that above the piston, the net flow of the fluid is into the high-pressure reservoir. Again, as it was the case during the valve opening, energy recovery takes place. Thus, in this system the energy recovery takes place twice each valve event. When the valve exhausts its kineti c energy, its motion stops, and the check valve closes. Ideally, this should always coincide with the valve seating on its seat. This, however, is difficult to accomplish. A more practical solution is to bring the valve to a complete stop a fraction of a millimeter before it reaches the valve seat and then, briefly open the closing solenoid valve again. This again connects the upper volume with the however, is difficult to accomplish. A more practical solution is to bring the valve to a complete stop a frac tion of a millimeter before it reaches the valve seat and then, briefly open the closing solenoid valve again. This again connects the upper volume with the low-pressure reservoir, and the high pressure in the lower volume brings the valve to its fully closed position. Step 6 illustrates the valve seating. After that, the closing solenoid valve is deactivated again. For the rest of the cycle both solenoid valves and both check valves are closed, the pressure above the valve piston is equal to the pressure i n the low-pressure reservoir, and the high pressure below the piston keeps the engine valve firmly closed.

Valve motion control


Varying the activation timing of both solenoids varies the timing of the engine valve opening and closing. This, of course, also vanes the valve event duration. Valve lift can be controlled by varying the duration of the solenoid voltage pulse. Changing the high pressure permits control of the valve acceleration, velocity, and travel time. The valve can be deactivated during engine operation by simply deactivating the pair of solenoids which control it. Deactivation can last any number of cycles and be as short, as one cycle. Increasing the number of valves in each cylinder does not require a corresponding increase in the number of solenoid valves. The same pair of solenoid valves, which controls a single valve, can also control several valves running in-parallel. Thus, in a four-valve engine a pair of solenoid valves operates two synchronously running intake valves, and another pair runs the two exhaust valves. UNEQUAL LIFT MODIFIER - In a four-valve engine an actuator set consisting of two solenoid valves and two check valves controls the operation of a pair of intake or a pair of exhaust valves. Solenoids and check valves are connected to a common control chamber serving both valves (Figure 10). In a four-cylinder engine there is a total of eight control chambers connected to eight pairs of valves. For each pair, the

volumes below the hydraulic pistons are connected to the high pressure reservoir via a device called the lift modifier. In a neutral position the modifier does not affect the motion of the valves, and activation of the solenoid valves moves both engine valves in unison

. Figure 10. Paired valves with unequal lift control. To enhance the ability to vary the intake air motion in the engine cylinder, it is often desirable to have unequal lift of the two intake valves, or even to keep one of the two valves closed while the other opens. In some cases it may also be used for paired exhaust valves. The lift modifier is then used to restrict the opening of one the paired valves. The modifier is shown schematically in Figure 11 as a Rotating rod with its axis of rotation perpendicular to the plane of the drawing. The rod is installed in the cylinder head between the two intake valves. A cutout in the rod forms a communication chamber connected to the volumes below the hydraulic pistons of both intake valves. The communication chamber is always connected to the high pressure reservoir. In the case A the modifier is in the neutral position, and both valves operate in unison. In the case B the modifier rod is shown turned 90 degrees clockwise. The exit of oil from the volume below the hydraulic piston in the valve No. 1 is blocked and the valve cannot move in the direction of opening. However, the entry of oil into the volume below the hydraulic piston is permitted by a one-way valve installed in the modifier rod. This guarantees that, whenever deactivation takes place, the valve No. 1 will close and remain closed, while the valve No.2 continues its normal operation. If the modifier rod is turned 90 degrees counter-clockwise (from the position shown in the case A), the valve No.2 is deactivated, while the valve No. 1 would continue normal operation. In the case C the lift of one of the valves is reduced relative to the second one. The rod is turned a smaller angle so that the exit of oil from the valve No. 1 into the communication chamber is not completely blocked, but the flow is significantly throttled. As a result, the motion of the valve No. 1 is slowed down and its lift is less than that of the valve No.2. Varying the angular position of the modifier rod 26 varies the degree of oil throttling, thus varying the lift of the valve No. 1.

Figure 11. Unequal lift control.

DESIGN APPROACH FOR CAMLESS ENGINE


The camless engine was created on the basis of an existing fourcylinder, four-valve engine. The original cylinder head with all the valves, springs, camshafts, etc. was replaced by a new cylinder head assembly fully integrated with the camless valvetrain. The camshaft drive was eliminated, and a belt-driven hydraulic pump was added. There was no need for lubrication, and the access for engine oil from the engine block to the cylinder head was closed off. No other changes to the engine have been made.

Cylinder head
Two cross sections of the cylinder head are shown in Figure 12. The aluminum casting is within the original confines and contains all hydraulic passages connecting the system components. The high- and low-pressure hydraulic reservoirs are integrated into the casting. The reservoirs and the passages occupy the upper levels of the cylinder head and are part of the hydraulic system. The hydraulic fluid is completely separated from the engine oil system. A finite element analysis was used to assure the cylinder head integrity for fluid pressures of up to 9 M a. The lower level of the head contains the engine coolant.

Figure 12. Cross sections of cylinder head. The engine valves, the check valves and the modifiers are completely buried in the body of the head. The solenoid valves are installed on the top of the cylinder head and are kept in their proper locations by a cylinder head cover. Hydraulic and electric connections leading to the hydraulic pump and the electronic controller, respectively, are at the back end of the cylinder head. The height of the head assembly is approximately 50 mm lower than the height of the base engine head. Figure 13 is a photograph of the head on the engine with the head cover removed. 27

COMPONENTS OF CAMLESS ENGINE


Main components of a camless engine are-Engine valve, solenoid valve, high pressure pump, low pressure pump, cool down accumulator, etc. A cross section of the engine valve assembly is shown in Figure 14. The valve piston is attached to the top of the valve, and both the valve and the piston can slide inside a sleeve. The sleeve openings above and below the valve piston allow the hydraulic fluid to enter and exit. A seal in the lower part of the sleeve prevents leakage of fluid into the intake or exhaust port. A leak-off (not shown) unloads the seal from excessive pressure, which otherwise increases friction. There is a tight hydraulic clearance between the valve and the sleeve. However, the clearance between the sleeve and the cylinder head is relatively large,

Engine valve

which improves the centering of the valve in its seat Circulation of hydraulic fluid through the chambers above and below the valve piston cools and lubricates the valve. All the forces acting on the valve are directed along its axis. Absence of side forces reduces stresses, friction and wear.

Figure 14. Engine valve

Solenoid valve Figure 15 shows a cross section of the solenoid valve. The solenoid has conically shaped magnetic poles. This reduces the air gap at a given stroke. The normally-closed valve is hydraulically balanced during its movement. Only a slight unbalance exists in the fullyopen and the fully-closed positions. A strong spring is needed to obtain quick closing time and low leakage between activations. The hydraulic energy loss is the greatest during the closing of either the high- or the low-pressure solenoid, because it occurs during the highest piston velocity. Thus, the faster the solenoid closure, the better the energy recovery. The valve lift and the seat diameter are selected to minimize the hydraulic loss with a large volume of fluid delivered during each opening. Both high-pressure and low-pressure solenoid valves are of the same design.

Figure 15. Solenoid valve.

Lift modifier - The design of the lift modifier permits a simultaneous


hydraulic control of a group of modifiers with a single pulse-width modulated solenoid-valve that adjusts the pressure in a control gallery.

Hydraulic system
A diagram of the hydraulic system is shown in Figure 16. An enginedriven variable-displacement pump automatically adjusts its output to maintain the required pressure. The high-pressure and the low-pressure reservoirs are connected to all corresponding locations serving the highand the low-pressure solenoids and the check valves.

Figure 16. Hydraulic System. the quantity of fluid delivered by the high pressure pump with the actual needs of the system at various engine speeds and loads is critical to assuring low energy consumption. To conserve the mechanical power needed to drive the pump, its hydraulic output should closely match the needs. A variable displacement, high efficiency, axial plunger-type pump was initially selected for that reason. Taking into account the prohibitively high cost of such pump for automotive applications, a low-cost variable capacity pump was conceived. A cross section of the pump is shown in Figure 17. The pump has a single eccentric-driven plunger and a single normally-open solenoid valve. During each down stroke of the plunger the solenoid valve is open, and the plunger barrel is filled with hydraulic fluid from the low pressure branch of the system. During the upstroke of the plunger, the fluid is pushed back into the low pressure branch, as long as the solenoid valve remains open. Closing the solenoid valve causes the plunger to pump the fluid through a check valve into the high pressure branch of the system. Varying the duration of the solenoid voltage pulse varies the quantity of the high-pressure fluid delivered by the pump during each revolution.

High pressure pump:

Figure 17. High pressure pump. - A small electrically driven pump picks up oil from the sump and delivers it to the inlet of the main pump. Only a small quantity of oil is required to compensate for the leakage through the leak-off passage, and to assure an adequate inlet pressure for the main pump. Any excess oil pumped by the small pump returns to the sump through a low-pressure regulator. A check valve 1 assures that the inlet to the main pump is not subjected to pressure fluctuations that occur in the low-pressure reservoir. - The system also includes a cooldown accumulator that, during normal operation, is fully charged with oil under the same pressure as in the inlet to the main pump. When the engine stops running, the oil in both the high- and the low-pressure branches cools off and shrinks. As the system pressure drops, the accumulator discharges oil into the system, thus compensating for the shrinkage and preventing formation of pockets of oil vapor. The high pressure branch is fed from the accumulator through a check valve 2 that is installed in-parallel to the main pump. The low-pressure branch is fed through an orifice that is installed in-parallel to the check valve 1. The orifice is small enough to prevent pressure wave propagation through it during each engine cycle, but sufficient to permit slow flow of oil from the accumulator to the reservoir. In some applications, the orifice can be incorporated directly in the check valve. After the oil in the system has cooled off, the accumulator maintains the system at above atmospheric pressure by continuously replenishing the oil that slowly leaks out through the leak-off passage. When the engine is restarted, the accumulator is recharged again. If the engine is not restarted for a very long time, as it is the case when a vehicle is left in a long-term parking, the accumulator will eventually become fully discharged. In that case, the pressure in the accumulator drops to an unacceptable level, and a pressure sensor, that monitors the accumulator pressure, sends a signal to the engine control system which reactivates the electric pump for a short period of time to recharge the accumulator. This process can be repeated many times, thus

Low pressure pump

Cool down accumulator

maintaining the system under a low level of pressure until the engine is restarted. After the engine restarts it takes less than one revolution of the main pump to restore the high pressure. Operating the hydraulic system in a closed loop contributes to low energy consumption. The amount of hydraulic power consumed by the system is determined by the flow of fluid from the high- to the low-pressure reservoir times the pressure differential between the outlet from and the inlet to the high pressure pump. A small loss is also associated with leakage. There are good reasons to use high hydraulic pressure in the system, one of them being the need to maintain a high value of the bulk modulus of the oil. In a closed-loop system the pressure in the low-pressure reservoir can also be quite high, although lower than in the high-pressure reservoir (thus the pressure in the low-pressure rail is low only in relative terms). Hence, the system can operate with very high hydraulic pressure, and yet the energy consumption remains modest due to a relatively low pressure differential. The ratio of high pressure to low pressure must be sufficiently higher than the ratios of the pressure areas above and below the valve piston to assure reliable engine valve closure.

ADVANTAGES OF CAMLESS ENGINE


` Electro hydraulic camless valve train offers a continuously variable and independent control of all aspects of valve motion. This is a significant advancement over the conventional mechanical valve train. It brings about a system that allows independent scheduling of valve lift, valve open duration, and placement of the event in the engine cycle, thus creating an engine with a totally uncompromised operation. Additionally, the ECV system is capable of controlling the valve velocity, perform selective valve deactivation, and vary the activation frequency. It also offers advantages in packaging. Freedom to optimize all parameters of valve motion for each engine operating condition without compromise is expected to result in better fuel economy, higher torque and power, improved idle stability, lower exhaust emissions and a number of other benefits and possibilities. Camless engines have a number of advantages over conventional engines.

In a conventional engine, the camshaft controls intake and exhaust valves. alve timing, valve lift, and event duration are all fixed values specific to the camshaft design. The cams always open and close the valves at the same precise moment in each cylinder s constantly repeated cycle of fuel-air intake, compression, combustion, and exhaust. They do so regardless of whether the engine is idling or spinning at maximum rpm. As a result, engine designers can achieve optimum performance at only one speed. Thus, t he camshaft limits engine performance in that timing, lift, and duration cannot be varied. The improvement in the speed of operation valve actuation and control system can be readily appreciated with reference to Figure 12. It shows a comparison between va lve speeds of a mechanical camshaft engine and the camless engine valve actuation. The length of the valve stroke in inches versus degrees of rotation of a mechanical camshaft is illustrated. When graphed, the cycle of opening and closing of a valve driven by a mechanical camshaft will display a shape similar to a sine curve. The opening period (as measured in crankshaft degrees) remains constant for any engine load or rpm. However, the cycle of opening and closing of valves driven by the electromechanical valve actuators operates much faster. Designed to match valve-opening rates at the maximum engine rpm, the electromechanical valve actuators open the valve at this same rate regardless of engine operating conditions. Because of this improved speed, greater flexibility in programming valve events is possible, allowing for improved low-end torque, lower emissions and improved fuel economy. The massive opening period for the electromechanically driven valve can also be seen But in a cam less engine, any engine valve can be opened at any time to any lift position and held for any duration, optimizing engine performance. The valve timing and lift is controlled 100 percent by a microprocessor, which means lift and duration can be changed almost infinitely to suit changing loads and driving 0conditions. The promise is less pollution, better fuel economy and performance. Another potential benefit is the cam less engines fuel savings. Compared to conventional ones, the cam less design can provide a fuel economy of almost 7- 10% by proper and efficient controlling of the valve lifting and valve timing. The implementation of camless design will result in considerable reduction in the engine size and weight. This is achieved by the elimination of conventional camshafts, cams and other mechanical linkages. The elimination of the conventional camshafts, cams and other mechanical linkages in the camless design will result in increased power output. The better breathing that a camless valve train promotes at low engine speeds can yield 10% to 15% more torque. Camless engines can slash nitrogen oxide, or O x, pollution by about 30% by trapping some of the exhaust gases in the cylinders before they can escape. Substantially

reduced exhaust gas HC emissi ons during cold start and warm -up operation.

CONCLUSIONS
1. An electro hydraulic camless valve train was developed for a camless engine. Initial development confirmed its functional ability to control the valve timing, lift, velocity, and event duration, as well as to perform selectively variable deactivation in a four -valve multicylinder engine. 2. The system employs the hydraulic pendulum principle, which contributes to low hydraulic energy consumption. 3. The electro hydraulic valve train is integral with the cylinder head, which lowers the head height and improves the engine packaging. 4. Review of the benefits expected from a camless engine points to substantial improvements in performance, fuel economy, and emissions over and above what is achievable in engines with camshaftbased valve trains. 5. The development of a camless engine with an electro hydraulic valve train described in this report is only a first step towards a complete engine optimization. Further research and development are needed to take full advantage of this system exceptional flexibility.

BIBLIOGRAPHY : Michael M.Schechter and Michael B.Levin Camless Engine SAE paper 960581  P. reuter, P. Heuser, and M. Schebitz, Strategies to Impove SIEngine Performance by Means of ariable Intake Lift, Timing and Duration , SAE paper 920449.  K. Hatano, k. Lida, H. Higashi, and S. Murata, Development of a ew Multi-Mode ariable alve Timing Engine,SAE paper930878  J-C. Lee, C-W. Lee, and J. itkiewitz, The Application of a Lost T System to a DOHC SI Engine,SAE paper 950816 Motion  John B. Heywood, Internal combustion engine fundamentals  William H. crouse. Automotive mechanics.  John Steven Brader ,A Thesis on Development of a Piezoelectric Controlled Hydraulic Actuator for a Camless Engine  mwww.machinedesign.com  www.halfbakery.com  www.deiselnet.com  www.greendieseltechnology.co  www.me.sc.edu