Woodyard D. (ed.) Pounders Marine diesel engines and Gas Turbines

Подождите немного. Документ загружается.

260 Fuel Injection

timing and injection rate shaping: capabilities achieved only with the aid of

electronic control systems.

Conventional injection systems with mechanical actuation include in-line

pumps, unit pumps with short HP fuel lines and unit injectors. A cam controls

the injection pressure and timing while the fuel volume is determined by the

fuel rack position. MTU asserts that for future engines with high injection

pressures, however, the in-line pump system can be ignored because it would

be hydraulically too ‘soft’ due to the long HP lines.

A comparison has been made between unit pump and unit injector systems,

assuming the unit injector drive adopts the typical camshaft/push rod/rocker

arm principle. Using simulation calculations, the relative behaviour of the two

systems was investigated for a speciﬁed mean injection pressure of 1150 bar in

the injector sac. MTU explains that this time-averaged sac pressure is a deter-

mining factor in fuel mixture preparation, whereas the frequently used maxi-

mum injection pressure is less meaningful. The pressure in a unit pump has

been found to be lower than that in a unit injector but, because of the dynamic

pressure increase in the HP fuel line, the same mean injection pressure of

1150 bar is achieved with less stress in the unit pump.

With the unit injector, the maximum sac pressure was 1670 bar (some

60 bar higher than that in the unit pump). To generate 1150 bar, the unit injector

needed 3.5 kW, some 6 per cent more power. During the ignition delay period,

12.5 per cent of the cycle-related amount of fuel was injected by the unit pump

compared with 9.8 per cent by unit injector. The former is, therefore, overall

the stiffer system. Translating the pressure differential at the nozzle oriﬁce and

the volume ﬂow into the mechanical energy absorbed, the result was a higher

efﬁciency of 28 per cent for the unit pump compared with 26 per cent for the

unit injector.

MTU reports that from the hydraulic aspect, the unit pump offers beneﬁts

in that there is no transfer of mechanical forces from the push rod drive to the

cylinder head, and less space is required for the fuel injector (yielding better

design possibilities for inlet and exhaust systems). With conventional sys-

tems, the volume of fuel injected is controlled by the fuel rack; and matching

the individual cylinders dictates appropriate engineering effort. The effort is

increased signiﬁcantly if injection timing is effected mechanically.

The engineering complexity involved in enabling fuel injection and timing

to be freely selected can be reduced considerably by using a solenoid valve to

effect time-orientated control of fuel quantity. To produce minimum fuel injec-

tion quantity, extremely short shift periods must be possible to ensure good

engine speed control. Activation of the individual solenoid valves and other

prime functions, such as engine speed control and fuel injection limitation, is

executed by a microprocessor-controlled engine control unit (ECU). Optional

adjustment of individual cylinder fuel injection calibration and injection timing

is thus possible with the injection period being newly speciﬁed and realized for

each injection phase. Individual cylinder cut-out control is only a question of

the software incorporated in the ECU.

With cam-controlled injection systems, the injection pressure is dependent

on the pump speed and the amount of fuel injected. For engines with high mean

effective pressures in the lower speed and low load ranges, this characteristic is

detrimental to the atomization process as the injection pressure drops rapidly.

Adjusting the injection timing also inﬂuences the in-system pressure build-

up. For example, if timing is advanced, the solenoid valve closes earlier, fuel

compression starts at a lower speed and thus leads to lower injection pressures,

which, in turn, undermines mixture preparation.

To achieve higher injection pressures, extremely steep cam conﬁgurations

are required. As a result, high torque peaks are induced into the camshaft,

which dictates a compensating degree of engineering effort on the dimension-

ing of the camshaft and gear train, and may even call for a vibration damper.

MTU argues that while the solenoid valve-controlled system has a number

of advantages, it retains the disadvantages of the conventional systems; and in

the search for a new ﬂexible injection system, it represents only half a step.

A full step is achieved only when the pressure generation, fuel quantity and

injection timing functions can be varied independently by exploiting the CR

injection system.

In the CR system conﬁguration for a V16-cylinder engine (Figure 8.22)

the HP pump delivers fuel to the rail, which is common to all cylinders. Each

injector is actuated in sequence by the ECU as a function of the crankshaft

angle. The injector opens when energized and closes when de-energized. The

amount of fuel injected per cycle is determined by the time differential and the

in-system pressure. The actual in-system pressure is transmitted to the control

unit via a pressure sensor, and the rail pressure is regulated by the ECU via the

actuator in the fuel supply to the HP pump.

mtu common rail system for high-speed engines 261

Crank angle

top dead centre

Rail

Pressure

sensor

Fuel supply

Fuel stop valve

Injecto

Electronic

control unit

Throttle valve

High-pressure

supply pump

FIgure 8.22 Cr injection system arranged for a v16-cylinder high-speed

engine (mtu)

262 Fuel Injection

In such a system the injector incorporates several functions. The nozzle

needle is relieved by a solenoid valve and thus opened by the fuel pressure.

The amount of fuel injected during the ignition delay period is regulated by

the nozzle opening speed. After the control valve is de-energized, an additional

hydraulic valve is activated to secure rapid closure of the nozzle needle and,

therefore, a minimum smoke index. This servo-assisted injector allows the

opening and closing characteristics to be adjusted individually and effected

very precisely. MTU claims that it is capable of extremely high reaction speeds

for controlling minimum fuel quantities during idle operation or pilot injection.

Compared with a conventional injection system, the pumping force is

considerably lower: pressure generation is accomplished by a multi-cylinder,

radial piston pump driven by an eccentric cam. Pressure control is realized by

restricting the supply ﬂow. Locating the HP pump on the crankcase presents no

problems, while the deletion of the fuel injection control cams from the cam-

shaft allows that component to be dimensioned accordingly.

The rail system is required to supply all injectors with fuel at an identical

pressure, and its design is based on criteria such as minimum fuel injection

quantity deviations between cylinders, faster pressure build-up after start and

minimum pressure loss from the rail to the nozzle sac. Simulation exercises

have shown that pressure ﬂuctuations can be very low.

The ECU determines the engine speed and calculates the amount of fuel to

be injected, based on the difference between actual and preset engine speeds.

The individual injectors are energized as a function of crankshaft angle and ﬁr-

ing order.

Control of the CR in-system pressure is also carried out by the ECU. The

advantages inherent in the solenoid valve-controlled conventional injection

system (regarding speed control due to the improved actuator dynamic) can

also be applied to the same extent to the CR system: at the moment of each

injection, the governor-generated data can be immediately processed. MTU

says that pressure regulation is highly effective. Load shedding (required, e.g.,

when a waterjet propulsion unit emerges from the sea during rough conditions)

can be effected from full to zero load in just 10 ms.

While this leads to a rapid rise in engine speed, the governor reduces fuel

injection to zero, which, in turn, causes a pressure increase in the rail system

as the individual injectors become inactive. The pressure regulator reacts and

restricts the HP pumps to a maximum of 1330 bar (a pressure excess of 130 bar

over the speciﬁed value of 1200 bar). Additionally, the ECU assumes the con-

trol and monitoring functions that are standard for MTU engines, including

sequential turbocharging control.

With CR injection, the complete system is permanently subjected to

extremely HPs. In the event of failure of a single injector, the engine is pro-

tected in that the injector is cut off from the fuel supply by the fuel stop valve:

no fuel can enter the combustion chamber. The engine can, however, still be

operated at reduced (get-you-home) power.

A mechanical pressure relief valve is actuated if, due to a pressure regulat-

ing system malfunction, the pressure rises to an unacceptable level. Leaks in

the system are identiﬁed by the pressure monitoring system.

MTU reports that during tests the unit pump system displayed the best

characteristics due to the high speciﬁed injection pressures. Greater ﬂexibil-

ity, improved dynamics, reduced design effort and, as a result, lower costs are

offered by the unit pump system with solenoid valve control. MTU asserts that

only a CR injection system, however, makes it possible to achieve injection rate

shaping over the complete operating range and thus to underwrite the speciﬁca-

tions for future diesel engines.

The HP pump for the Series 4000 fuel system was changed from a four- or

eight-cylinder radial piston pump to a four-cylinder inline unit to increase its per-

formance and achieve a higher delivery rate for the V20-cylinder engine model.

mtu series 8000 engine Cr system

MTU subsequently applied an electronic CR system to the group’s largest-ever

high-speed engine, the 265-mm bore Series 8000, introduced in 2000 (see Chapter

30), whose V20-cylinder version develops up to 9000 kW at 1150 rev/min. As on

the Series 4000 engine, the system enables inﬁnite variation of injection timing,

volume and pressure over the entire engine performance curve. The supply of

pressurized fuel to the injectors, however, is achieved by different means. On the

Series 4000 engine there is a rigid fuel accumulator (CR) for each row of cylin-

ders, while the Series 8000 engine has a separate accumulator for each cylinder

located in the vicinity of the injector and with a capacity of 0.4 litres.

This arrangement facilitates a smaller overall engine width and optimum

progression of combustion, and reduces the interplay between the injectors and

the pressure ﬂuctuations in the fuel pipes. Only two HP pumps are required to

generate the substantial pressure and the necessary ﬂow volume for the entire

engine.

The L’Orange system for the Series 8000 engine is based on two 3-cylinder

HP fuel pumps producing a maximum rail pressure of 1800 bar. These pumps

are integrated in the gear drive at the front side of the engine and driven by the

crankshaft with a gear ratio of 1.65 at speeds of up to 1800 rev/min. The HP

part of the injection system, with the functions of HP generation and accumu-

lation as well as fuel metering, comprises the following:

l Two HP fuel pumps with integrated pump accumulators.

l Up to 20 accumulators, each equipped with a ﬂow limiting valve.

l High-pressure fuel lines.

l Up to 20 electro-magnetically controlled injectors.

Double-walled HP fuel lines distribute the fuel from the pumps along the

engine to each cylinder. An accumulator serving each cylinder acts as a stor-

age space to maintain an almost constant fuel pressure during injection, despite

the relatively small cross-section in the fuel supply lines. The compressibility

of the fuel ensures an almost constant injection pressure during injection. The

mtu common rail system for high-speed engines 263

264 Fuel Injection

intervals between injections are used to replace the injected fuel mass in the

accumulators and build up the fuel pressure again. The accumulator incorpo-

rates a so-called fuel-stop valve, which interrupts the fuel supply to the rel-

evant injector in cases of excessive fuel ﬂow rates (e.g., caused by leakages

or continuous injection). The engine management system will then adjust the

power output of the engine accordingly.

Rounded-off spray hole intake geometries on the electronically control-

led injector ensure an optimally sprayed injection of fuel into the combustion

chamber. The start of injection and the quantity injected are accurately control-

led to the microsecond by the engine control system in line with the engine

operating condition.

The system fuel pressure is available at the nozzle seat as well as in the

control unit above the plunger for needle lift control (the so-called connect-

ing rod). Compared with the usual two-cycle systems, this offers the advantage

of a faster operating response. The different-sized pressurized surfaces of con-

necting rod and needle result in a hydraulic closing force. When the solenoid

is energized by an electrical signal, the armature is lifted. This opens an accu-

rately machined cross-section of the outlet throttle in the control unit, causing

a rapid drop in the hydraulic pressure on the connecting rod. The nozzle needle

lifts off the seat and starts the injection process, which is stopped when the

solenoid is de-energized. The closing ﬂank of the injection characteristics can

be shaped very rapidly through an intermediate valve after the control unit; it

opens additional supply cross-sections during closing of the solenoid.

Fuel is injected in the precise quantity and at exactly the moment required

for the conditions under which the engine is operating. The pump’s intake

ﬂow throttle ensures that only the precise amount of fuel required is pres-

surized, resulting in a very high pump efﬁciency and consequently low fuel

consumption.

The advantages of the CR injection system are fully exploited in conjunc-

tion with the engine control system. The ﬂexibility of the injection system with

respect to injection pressure and timing allows the option to access and opti-

mize the engine operating parameters throughout the performance map (e.g.

addressing fuel consumption and emissions).

For each operating condition of the engine, the engine control system

analyses the sensor information and computes them into a setting for system

fuel pressure, injection start and timing. The computed data take into account

engine speed, power demand and transient operating conditions such as engine

warm-up after cold starting. The scope also embraces surrounding conditions

and possible faults in the drive system, with appropriate countermeasures auto-

matically being initiated. In addition, the engine control system monitors the

limits to certain operating parameters (e.g., maximum pressure in the fuel sys-

tem) to avoid component damages.

WärtsIlä Cr sYstem For medIum-sPeed engInes

A CR fuel system developed by Wärtsilä in conjunction with Delphi Bryce

(now part of the Woodward Governor group) for application to the group’s

heavy fuel-burning medium-speed engines is modular, with common com-

ponents designed to serve engines with cylinder numbers from 4 to 20.

Development was stimulated by increasing market pressure, especially from

passenger tonnage, for no visible smoke emissions under any circumstances.

Wärtsilä Cr system for medium-speed engines 265

(1) Injector

(2) Accumulator

(5) Drive cam

(4) Shielded HP pipes

(3) HP pump

FIgure 8.23 Cr fuel system for Wärtsilä medium-speed engines

Rail feed

FIgure 8.24 schematic diagram showing the function of the Wärtsilä Cr fuel

injector

266 Fuel Injection

The Wärtsilä system (Figures 8.23 and 8.24) is based on a pressurizing

pump and a fuel accumulator for every two cylinders, the series of accumulators

interconnected by small-bore piping. Connections from every accumulator feed

the injectors of two cylinders (in engines with an odd number of cylinders, an

extra accumulator is installed with a connection to one cylinder).

Functionally, the main reason for choosing such a system is that by split-

ting up the fuel volumes in several accumulators, the risk of pressure waves in

the CR is avoided. From a safety viewpoint, an advantage is that HP fuel exists

only in the same area as in a conventional engine (in the hot box). A third merit

cited is economic, the system using a camshaft that is present anyway to drive

the valves.

The computer-controlled fuel injection system opens up possibilities to

optimize engine performance, thanks to the freedom to choose key parameters:

injection timing, injection pressure independent of engine speed, maximum

injection quantity dependent on charge air pressure and rotational speed, opti-

mized injection during starting and optimized injection timing at load pick-up.

Each CR fuel pump is driven by a two-lobe cam, whose pumping pro-

ﬁle can be considerably slower than the conventional jerk pump type that it

replaces. In a CR system there is no longer the requirement to pump fuel to

the injector directly within the typical range of 6–10 s; this fosters a reduc-

tion in mechanical drive noise and drive energy consumed. The interconnected

accumulator arrangement reduces the pulsation migration ability along the rail.

Additionally, the rail components are rigidly ﬁxed to the engine block, reduc-

ing the risks of piping leakage. Each accumulator is connected directly to two

injectors, and each pipeline contains a hydraulic ﬂow fuse for safety reasons;

each accumulator also contains a pressure sensor.

Wärtsilä’s engine control system (WECS) regulates the CR system, moni-

toring several system parameters; it also controls fuel pressure in the rail and

injection quantities and timings according to speciﬁed maps. The solenoids

are not responsible for all control; pressurized engine oil is used to achieve the

necessary forces in the injectors and stop solenoids (this oil system is termed

‘control oil’).

The CR pump is constructed with a tappet and barrel element integrated in

one housing. The use of heavy fuel oil makes it preferable to divide the pump

internally; this places the pumping element in a fuel-only environment, which

eliminates the risk of lacquer accumulation from the fuel and chemical interac-

tion with the engine oil. The pump can be driven by two- or three-cam lobes.

The two-lobe alternative is preferred both for the roller/roller pin oil ﬁlm duty

cycle and because it creates fewer pumping pulses to settle into the rail prior to

injections.

The volumetric efﬁciency of the pump is typically above 80 per cent. This

level is generally higher than that of jerk pumps with a plunger spill helix,

where the end of the plunger stroke marks the rejection of fuel into the return

circuit. Smaller plunger sizes are thus allowed, which in turn signiﬁcantly

reduces the maximum camshaft drive torque for an equivalent pumping pres-

sure target.

Charging of the CR is controlled by regulating the volume of fuel allowed

to enter the pumping chamber; this is effected by a ﬂow control valve installed

in either the barrel or the pump housing. The ﬂow control valve itself is rotated

by a rotary solenoid, which has a position feedback sensor for fault detec-

tion prior to engine start or during engine operation. Another detection device

is incorporated in the barrel. A thermocouple sensor is installed close to the

pumping chamber to detect delivery valve leakage; if the delivery valve were

to leak, fuel from the rail would be re-pumped, causing a temperature increase

in the pumping chamber.

Each accumulator, mounted rigidly on the engine block, receives fuel from

the HP pump, stores it under HP and distributes it to the injectors. The accu-

mulator itself is an empty space. Flow fuses located at outlets from the accu-

mulator to the injector limit the quantity of fuel that can be transferred from

the accumulator to the injector if the latter malfunctions. The accumulator is

also ﬁtted with a device that makes it easy to ﬁnd a leaking pipe if the leak

alarm is activated.

At least one of the accumulators is also equipped with a start-up and safety

valve (SSV), which opens the ﬂow path from the accumulator and is control-

led by the WECS. The SSV is opened when the engine is in the pre-warming

phase or when the engine is to be stopped. Incorporated in the SSV is a

mechanical safety valve to protect the system against excessive rail pressure,

this valve operating completely independently of the main valve. Connected

to the SSV is an empty space in the fuel outlet channel, whose function is to

dampen the pressure waves when the SSV is opened.

The fuel injector was designed for efﬁciency and safe operation, with four

main elements: nozzle, shuttle valve, solenoid valve and piston acting on the

nozzle needle. The injector is ﬁtted with a conventional nozzle that is linked

to the CR fuel source via the shuttle valve. Two key functions are performed

by the shuttle valve: it controls injections and de-pressurizes the injector nee-

dle. In its closed (static) position, the valve isolates the injector passages and

nozzle from the pressure source, so that pressure is not applied to the nozzle seat

between injections. The shuttle is opened by applying a hydraulic load using the

control oil source. The control oil is switched on and off using a fast-acting sole-

noid valve (a reliable automotive component produced in high volumes).

Wärtsilä explains that such an approach fosters full ﬂexibility of the start of

injection and fuelling quantity. The piston above the nozzle, together with any

pressure on it, acts to hold the nozzle closed; pressure is normally applied only

to the piston at the start and end of injections. The piston arrangement has the

following three functions:

1. The pressure on the piston at the start of injection slightly delays

opening of the nozzle, allowing time for the shuttle to complete its

travel—even for the smallest injections. This promotes fuelling consist-

Wärtsilä Cr system for medium-speed engines 267

268 Fuel Injection

ency because the need for partial shuttle travel at the lowest outputs is

avoided.

2. At the end of injection the pressure on the piston helps to close the

nozzle. The fuel pressure at the nozzle seat at the end of injection is

much higher than that in conventional systems, and thus each injection

is terminated very rapidly.

3. A safety beneﬁt: if the shuttle becomes stuck in a partially open posi-

tion, the pressure is continuously applied to the piston, this force keep-

ing the nozzle closed in this fault condition.

These design features resulted in an injector with a high mean operating

pressure, capable of working up to 2000 bar. Injections start crisply and termi-

nate with the nozzle being closed while injecting a nearly full pressure. The

injector was also designed for robust safety in the event of a fault condition:

l Rail pressure is not applied to the nozzle seat between injections; this

ensures that the fuel is cut off if the nozzle needle is prevented from

fully closing.

l The piston acting on the nozzle needle prevents injection if the shuttle

seizes or is prevented from fully closing.

Engine operation on heavy fuel oil is addressed by the following features:

l During the engine’s pre-start warm-up phase, the pump ﬂow control

valves are shaken and opened separately every 5 s. Heated HFO ﬂows

through each pump in turn and then onward to the rail and back to the

tank via the bypass SSV valve system.

l The fuel injectors do not have HFO present in the solenoid armature

area; a servo oil system is used.

l In the fuel injectors the rail pressure does not act directly on the nozzle

needle, reducing the risk of needle sticking; and the rail pressure is

released from the nozzle needle between injections to improve safety.

The Wärtsilä CR system was designed for retroﬁtting to the group’s exist-

ing engines by simply removing the original conventional injection pumps

and replacing them with CR fuel pumps and accumulators that ﬁt exactly into

current engine blocks.

Acknowledgements to Duap, Fiedler Motoren Consulting, Delphi Bryce

(Woodward Governor), MTU, L’Orange and Wärtsilä Corporation.

See also chapters on Caterpillar Marine Power Systems (MaK), MAN

Diesel, MTU and Wärtsilä four-stroke engines. Two-stroke engine fuel injec-

tion systems, including electronically controlled conﬁgurations, are discussed

in Chapter 9 and the MAN Diesel, Mitsubishi and Wärtsilä low-speed engine

chapters.

269

C h a p t e r | n i n e

Low-Speed Engines—

Introduction

Low-speed two-stroke engine designers have invested heavily to sustain their

dominance of the mainstream deep sea propulsion sector formed by tank-

ers, bulk carriers and container ships, and recently extended market oppor-

tunities to large twin-screw LNG carriers. The long-established supremacy

reﬂects the perceived overall operational economy, simplicity and reliability

of single, direct-coupled crosshead engine plants. Other factors are the con-

tinual evolution of engine programmes by the designer/licensors in response

to or anticipation of changing market requirements, and the extensive network

of enginebuilding licensees in key shipbuilding regions, notably Korea, Japan

and China. Many of the standard ship designs of the leading yards, particularly

in Asia, are based on low-speed engines. Chinese low-speed engine licensees

have proliferated in recent years to support an expanding domestic shipbuild-

ing industry, and licences have also been arranged in Vietnam.

The necessary investment in R&D, production and overseas infrastructure

dictated to stay competitive, however, took its toll over the decades. Only three

low-speed engine designer/licensors—MAN B&W Diesel, Mitsubishi and

Sulzer—survived into the 1990s to contest the international market. MAN B&W

Diesel was superseded in 2007 by MAN Diesel, which retains MAN B&W as a

brand name for its low-speed engines. The Sulzer diesel engine business became

part of the Wärtsilä Corporation in 1997 and the Sulzer brand is no longer used

in the Wärtsilä portfolio.

The roll call of past contenders in two-stroke engine design and construc-

tion includes names either long forgotten or living on only in other engineer-

ing sectors: AEG-Hesselman, Deutsche Werft, Fullagar, Krupp, McIntosh and

Seymour, Neptune, Nobel, North British, Polar, Richardsons Westgarth, Still,

Tosi, Vickers, Werkspoor and Worthington. The last casualties were Doxford,

Götaverken and Stork, some of whose distinctive engines remain at sea in

diminishing numbers. The pioneering designs displayed individual ﬂair within

generic classiﬁcations, which offered two- or four-stroke, single- or double-

acting, and single- or opposed-piston conﬁgurations. The Still concept even

combined the Diesel principle with a steam engine: heat in the exhaust gases