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

Подождите немного. Документ загружается.

more than half and insoluble deposits in the oil reduced dramatically, signiﬁ-

cantly extending oil ﬁlter life. Carbon-cutting rings can be retroﬁtted to deliver

their beneﬁts to engines in service. Removal prior to piston withdrawal is sim-

ply effected with a special tool.

Designers now also favour a ‘hot box’ arrangement for the fuel injection

system to secure cleaner engine lines and improve the working environment

in the machinery room thanks to reduced temperatures; additionally, any fuel

leakage from the injection system components is retained within the box.

The major medium-speed enginebuilders have long offered 500-mm-

bore-plus designs in their portfolios. MAN Diesel still ﬁelds its L58/64 series

but MaK’s 580-mm bore M601 and the Sulzer ZA50S engines have been

phased out, as was Stork-Wärtsilä’s TM620 engine in the mid-1990s. In the

1970s MAN and Sulzer jointly developed a V-cylinder 650-mm bore/stroke

design (developing 1325 kW/cylinder at 400 rev/min) that did not proceed

beyond prototype testing.

Wärtsilä’s 64 series, launched in 1996, took the medium-speed engine

into a higher power and efﬁciency territory, the 640-mm bore/900-mm stroke

design now offering an output of 2010 kW/cylinder at 333 rev/min. A V12-

cylinder model delivers 23 280 kW at 400 rev/min. The range can therefore

meet the propulsive power demands of virtually all merchant ship tonnage

types with either single- or multi-engine installations. The key introductory

parameters were 10 m/s mean piston speed, 25 bar mean effective pressure,

and 190 bar maximum cylinder pressure. The Finnish designers claimed the

64 series to be the ﬁrst medium-speed engine to exceed the 50 per cent ther-

mal efﬁciency barrier, and suggested that overall plant efﬁciencies of 57–58

per cent are possible from a combi-cycle exploiting waste heat to generate

steam for a turbo-alternator.

At the other end of the medium-speed engine power spectrum, the early

1990s saw the introduction of a number of 200-mm bore long stroke designs

from leading builders such as Daihatsu, MaK and Wärtsilä Diesel, contesting

a sector already targeted by Sulzer’s S20 model. These heavy fuel-burning

engines (typically with a 1.5:1 stroke–bore ratio) were evolved for small-ship

propulsion and genset drive duties, the development goals addressing over-

all operating economy, reliability, component durability, simplicity of main-

tenance and reduced production costs. Low and short overall conﬁgurations

gave more freedom to naval architects in planning machinery room layouts and

eased installation procedures (Figure 16.4).

The ~320-mm bore sector is ﬁercely contested by designers serving a high-

volume market created by propulsion and genset drive demands. A number of

new designs—including Caterpillar/MaK’s M32 and MAN Diesel’s L32/40—

emerged to challenge upgraded established models such as the Wärtsilä 32.

A Japanese challenger in a medium-speed arena traditionally dominated by

European designer/licensors arrived in the mid-1990s after several years’ R&D by

the Tokyo-based Advanced Diesel Engine Development Company. The joint ven-

ture embraced Hitachi Zosen, Kawasaki Heavy Industries and Mitsui Engineering

Medium-speed engines—introduction 513

514 Medium-Speed Engines—Introduction

and Shipbuilding. The 300-mm bore/480-mm stroke ADD30V design, in a V50-

degree conﬁguration, developed up to 735 kW/cylinder at 750 rev/min, signiﬁ-

cantly more powerful than contemporary medium-speed engines of equivalent

bore size. A mean effective pressure of around 25 bar and a mean piston speed of

11.5 m/s were exploited in tests with a six-cylinder prototype, although an mep

approaching 35 bar and a mean piston speed of 12 m/s are reportedly possible.

In addition to a high speciﬁc output, the developers sought a design,

which was also over 30 per cent lighter in weight, 10–15 per cent more fuel

economical and with a better part-load performance than established engines.

Underwriting these advances in mep and mean piston speed ratings are an

FIgurE 16.4 MaK’s M20 engine represented a new breed of 200-mm bore designs

anti-wear ceramic coating for the sliding surfaces of the cylinder liners and

piston rings, applied by a plasma coating method. A porous ceramic heat shield

was also developed for the combustion chamber to reduce heat transfer to the

base metal of the piston crown.

A key feature is the single-valve air intake and exhaust gas exchange

(Figure 16.5), contrasting with the four-valve (two inlet and two exhaust)

heads of other medium-speed engines. The greatly enlarged overall dynamic

valve area and the reduction in pressure losses during the gas exchange period

promote a higher thermal efﬁciency. The system is based on a heat-resistant

alloy main poppet valve located over the centre of the cylinder and a control

valve placed co-axially against this main valve. The control valve switches the

air intake and exhaust channels in the cylinder cover. Both main and control

valves are driven hydraulically.

Medium-speed engines—introduction 515

FIgurE 16.5 The Japanese ADD 30V engine is distinguished by a single-valve gas

exchange system comprising a main valve located at the centre of the cylinder and a

control valve placed co-axially against the main valve. The control valve switches the

air intake and exhaust channels in the cylinder cover. Both valves are driven hydrauli-

cally. Side-mounted fuel injectors are arranged around the cylinder periphery

516 Medium-Speed Engines—Introduction

Fuel injection is executed from the side through multiple injectors arranged

around the cylinder periphery instead of a conventional top-mounted central

injection system. Combustion characteristics were optimized by raising the

fuel injection pressure to around 2000 bar, thus enhancing the fuel–air mixture

formation and fostering low NOx emissions without sacriﬁcing fuel economy.

A computer-based mechatronics system automatically controls the timing of

fuel injection and valve opening/closing to match the operating conditions.

The ﬁrst 6ADD30V production marine engines, built by Mitsui, were speciﬁed

as the prime movers for the diesel-electric propulsion plant of a large Japanese

survey vessel.

Offshore vessel and LNG carrier propulsion market opportunities—and

the potential of wider mainstream shipping interest as environmental curbs

tighten—have encouraged a number of medium-speed enginebuilders to

develop dual-fuel and gas–diesel designs offering true multi-fuel capabilities

with high efﬁciency and reliability, and low carbon dioxide emissions. The

Hydraulic oil

for main valve

Hydraulic oil

for sub valve

Swirler

Sub valve

Air intake

Exhaust

Spring air

Main valve

Sub valve

T.D.C.

B.D.C.

T.D.C.

Exhaust

Air intake

T.D.C.

B.D.C.

FIgurE 16.5 Functioning of the ADD 30 V engine’s single valve gas exchange system

engines can run on gas (with a small percentage of liquid pilot fuel) or entirely

on liquid fuel (marine diesel oil, heavy fuel or even crude oil). Switching from

one fuel to another is possible without interrupting power generation (see

Chapter 2).

The high cost of R&D to maintain a competitive programme and continu-

ing investment in production resources and global support services have stimu-

lated a number of joint ventures and takeovers in the four-stroke engine sector

in recent years. Most notable have been Wärtsilä’s acquisition of the former

New Sulzer Diesel and Caterpillar’s takeover of MaK. Earlier, Wärtsilä had

acquired another leading medium-speed enginebuilder, the Netherlands-based

Stork-Werkspoor Diesel. This trend towards an industry comprising a small

number of major multi-national players contesting the world market has con-

tinued with the absorption of the British companies Mirrlees Blackstone,

Paxman and Ruston (formerly part of Alstom Engines) into the MAN Diesel

group. Rolls-Royce inherited the Bergen Diesel interests in Norway through

the takeover of Vickers-Ulstein.

Considerable potential remains for further developing the power ratings

of medium-speed engines, whose cylinder technology has beneﬁted in recent

years from an anti-polishing ring at the top of the liner, water distribution

rings, chrome–ceramic piston rings, pressurized skirt lubrication and nodular

cast iron/low friction skirt designs.

The pressure-lubricated skirt elevated the scufﬁng limit originally obtain-

ing by more than 50 bar, reduced piston slap force by 75 per cent and doubled

the lifetime of piston rings and grooves. Furthermore, it facilitated a reduc-

tion in lube oil consumption and, along with the simultaneous introduction

of the nodular cast iron skirt, practically eliminated the risk of piston seizure.

The anti-polishing ring dramatically improved cylinder liner lifetime beyond

100 000 h, and lube oil consumption became controllable and stable over

time, most engines today running at rates between 0.1 g/kW h and 0.5 g/kW h.

A further reduction in piston ring and groove wear was also achieved, and the

time-between-overhauls extended to 18 000–20 000 h. The ring itself is a wear

part but is turnable in four positions in a four-stroke engine, fostering a lengthy

lifetime for the component.

Such elements underwrite a capability to support a maximum cylinder

pressure of 250 bar of which 210–230 bar is already exploited in some engines

today. Leading designers such as Wärtsilä suggest that it may be possible to

work up towards 300 bar with the same basic technology for the cylinder unit,

although some areas need to be developed: bearing technology, for example,

where there is potential in both geometry and materials. A steel piston skirt

may become the most cost effective, and cooling of the piston top will probably

change from direct oil cooling to indirect. The higher maximum cylinder pres-

sure can be exploited for increasing the maximum effective pressure or improv-

ing the thermal efﬁciency of the process. Continuing to mould the development

of the medium-speed engine will be NOx and carbon dioxide emissions, fuel

ﬂexibility, mean time-between-failures and reduced maintenance.

Medium-speed engines—introduction 517

518 Medium-Speed Engines—Introduction

DESIgNINg MEDIuM-SpEED ENgINES

Investment in the development of a new medium-speed engine may be com-

mitted for a number of reasons. The enginebuilder may need to extend its port-

folio with larger or smaller designs to complete the available range, to exploit

recently developed equipment and systems, or to enhance the reliability, availa-

bility and economy of the programme. As an alternative to a new design, it may

be possible to upgrade an existing engine having the potential for improved

power ratings, lower operational costs, reduced emissions and weight. An

example is the upgrading by MAN Diesel of its 48/60 engine, whose B-version

offers a 14 per cent higher speciﬁc output with considerably reduced fuel con-

sumption and exhaust emission rates, a lower weight, and the same length and

height but a narrower overall width than its predecessor (Figure 16.6).

Having set the output range of the new engine, a decision must be made

on whether it should be built only as an in-line cylinder version or as a V-type

engine as well. In the case of the largest and heaviest medium-speed engines,

the efforts and costs involved in adding a V-type to the existing portfolio

might not be justiﬁed if there is a limited potential market for engines above

20 000 kW. Other restraining factors could be the capacity of the foundry for

casting very large crankcases or difﬁculties in transporting engines weighing

400 tonnes or more overland.

Once the power per cylinder of the proposed new engine is known, the ﬁrst

indication of its bore size is determined by the piston load, which relates the

speciﬁc output to the circular piston surface. Piston loads have increased with

FIgurE 16.6 MAN Diesel’s 48/60 engine, shown here in V14-cylinder form, beneted

from a redesign that raised specic output by 14 per cent and reduced fuel con-

sumption and emissions

the development of better materials: the original 48/60 engine has a piston load

of 58 kW/cm

, while the 48/60B engine achieves 66.4 kW/cm

The bore and stroke of an engine are interrelated by the stroke–bore ratio,

which in turn is based on the designer’s experiences with earlier engines. Some

25 years ago it was not uncommon to design medium-speed engines with very

similar bore and stroke dimensions (so-called ‘square’ engines); more recently,

the trend has been towards longer stroke designs, which offer clear advantages

in optimizing the combustion space geometry to achieve lower NOx and soot

emission rates. A longer stroke can reduce NOx emissions with almost no fuel

economy penalty and without changing the maximum combustion pressure.

The compression ratio can also be increased more easily and, together with

a higher ﬁring pressure, fuel consumption rates will decrease. Finally, long-

stroke engines yield an improved combustion quality, a better charge renewal

process inside the cylinder and higher mechanical efﬁciency.

A good compromise between an optimum stroke–bore ratio and the costs

involved, however, is an important factor: it is a general rule that the longer the

engine stroke, the higher the costs for an engine per kilowatt of output. The trend

towards longer stroke medium-speed engine designs is indicated by the follow-

ing table showing the current MAN Diesel family; the ﬁrst four models were

launched between 1984 and 1995, and the remaining smaller models after 1996.

Model Stroke–Bore Ratio

58/64 1.1

48/60 1.25

40/54 1.35

32/40 1.25

27/38 1.41

21/31 1.47

16/24 1.5

This table is a simpliﬁed outline of the initial design process and the real-

ity is much more complex. The ﬁnal conﬁguration results from considering a

combination of choices, which may have different effects on fuel consumption

rates and emissions. Nevertheless, the usual trade-off between fuel economy

and NOx emissions can be eliminated. The use of high-efﬁciency turbocharg-

ers is also essential.

A decision on engine speed is the next step. The mean piston speed in

metres per second can be calculated from the bore and speed of an engine

using the following formula: mean piston speed (m/s)  bore (m)  speed

(rev/min)/30. The upper limit of the mean piston speed is primarily given by

the size and mass of the piston and the high forces acting on the connecting

rod and crankshaft during engine running. A mean speed between 9.5 m/s and

Designing medium-speed engines 519

520 Medium-Speed Engines—Introduction

slightly above 10 m/s is quite common for modern large bore medium-speed

engines. Any substantial increase above 10 m/s will reduce operational safety

and hence reliability. Since medium-speed engines may be speciﬁed to drive

propellers and/or alternators, the selection of engine speed has to satisfy the

interrelation between the frequency of an alternator (50 Hz or 60 Hz) and

the number of pole pairs.

In designing the individual engine components, two main criteria are

addressed: reducing manufacturing costs and reducing the number of overall

engine parts to ease maintenance.

A valuable summary of the design process is also provided by Rolls-Royce,

based on its development of the Allen 5000 series medium-speed engine in

conjunction with Ricardo Consulting Engineers of the United Kingdom. The

initial functional speciﬁcation deﬁned a target power of 500 kW/cylinder and

a wide application proﬁle embracing industrial and standby power genera-

tion, pumping, propulsion and auxiliary power. It was considered essential to

design-in sufﬁcient ﬂexibility to accommodate all key parameters necessary

to ensure superior performance in these sectors well into the 21st century. The

following basic concepts were adopted by the design team:

l A power range of 3000–10 000 kW in steps of 1000 kW

l An output of 525 kWb per cylinder  10 per cent for 1 h in 12 h required

for power generation (equivalent to 500 kWe energy at both 720/750 rev/

min synchronous speeds)

l An output of 500 kWb per cylinder maximum continuous rating  10

per cent for 1 h in 12 h required for propulsion and pumping over a

speed range of 375–750 rev/min  15 per cent overspeed

l Fuel injection equipment to be of advanced technology, with the ﬂexibil-

ity to satisfy future emissions legislation and speciﬁed fuel consumption

requirements

l Fuel ﬂexibility covering heavy fuel oil and distillates

l High standards of reliability, durability and maintainability

l Condition monitoring capability essential

l Low number of parts.

The choice of brake mean effective pressure (bmep) was crucial to the

competitiveness of the design. It was essential to choose a bmep that would

allow the engine to be sold at a competitive price in pounds per kilowatt output.

Before commencing design work, therefore, a survey of competitive engines

was carried out. The survey showed that bmep levels increased on average by

0.25/0.33 bar per year, and predicted that in 1998 the highest rated engines

would have a rating of 25/25.3 bar. On this basis, coupled with engine econom-

ics, a rating of 26 bar bmep was selected for the new design. Cycle simulation

studies showed that this was achievable with a boost pressure ratio of 4.5:1,

provided that the combustion and injection systems were properly matched.

Any signiﬁcant increase in bmep would dictate the use of boost pressures of

5:1 and, while turbochargers were available to deliver this level of performance,

the special features incorporated in them to achieve it commanded a premium

price and increased the cost of the engine.

To secure the speciﬁed power per cylinder of 525 kW at a speed of

720/750 rev/min (60 Hz and 50 Hz synchronous speeds), a cylinder bore between

310 mm and 330 mm was required. It was decided at the onset of the design

process to limit the maximum piston speed to 10.5 m/s, which equates to a

stroke of 420 mm. The bore–stroke ratio therefore lay between 1.2 and 1.4,

typical for an engine in the anticipated market sector.

Combustion system design is governed by the customer’s need for compli-

ance with exhaust emissions legislation and excellent fuel economy. This dic-

tates that the engine should have a high air–fuel ratio, a shallow open bowl

combustion chamber and a very high compression ratio. The fuel injection sys-

tem must be capable of generating at least 1600 bar line pressure; an injection

system with built-in capability to vary the injection timing was a key feature of

its speciﬁcation.

In order to achieve a high trapped air–fuel ratio, the valve area should be

high and the ports designed for maximum efﬁciency with a small amount of

swirl (around 0.2–0.4 swirl ratio). The bowl shape should be wide and open

with no valve recesses, and the top ring should be placed reasonably high so

that the crevice volume is minimized to secure a high air utilization factor.

A maximum cylinder pressure target of 210 bar was chosen with the aim of

achieving excellent fuel economy; this in turn allows a high compression ratio

to be used (in the range 15.5–16.5:1). An injector nozzle with eight to ten holes

was speciﬁed. The combustion bowl design was optimized to ensure excellent

mixing of the fuel sprays to keep the air–fuel ratio as homogeneous as possible

and minimize the peak cycle temperatures and hence NOx formation.

The choice of bore size was inﬂuenced by considering the following: power

requirement (525 kW/cylinder), bmep (30 bar maximum), mean piston speed

(10.5 m/s), overall length/width/height, speciﬁc weight and speciﬁc cost.

To fulﬁl the design concept, the power requirement for the engine was

deﬁned as 500 kWe/cylinder continuous rating with a 10 per cent overload

capacity. The maximum continuous bmep is 26.6 bar, a limit set by the availa-

ble boost pressure level from single-stage turbocharging. The limit of 10.25 m/s

mean piston speed was selected to allow reliable operation at the maximum

continuous speed without signiﬁcant risk of piston, ring and liner scufﬁng

and wear problems. By selecting high values for rated bmep and mean piston

speed, the size of the cylinder components was minimized; this kept the overall

engine size small, with consequent beneﬁts in production costs.

Having set the piston speed and bmep at rated speed, it is possible to

derive the piston area and hence the cylinder bore: 320 mm. Having limited the

mean piston speed to a maximum of 10.25 m/s at the maximum rated synchro-

nous speed of 750 rev/min, the maximum allowable stroke was determined as

410 mm. Supporting these decisions, simulation work using the WAVE pro-

gram was carried out to examine bore sizes from 315 mm to 330 mm and stroke

sizes from 380 mm to 420 mm. The results conﬁrmed the selected values.

Designing medium-speed engines 521

522 Medium-Speed Engines—Introduction

Factors inﬂuencing the choice of cylinder conﬁguration (in-line or V-form)

are weight, cost, installation limitations, stress limitations (crank), vibration

levels, production tooling and market acceptance. From manufacturing, weight,

cost and installation considerations, a V-form family appeared attractive.

V6- and V8-cylinder designs, however, have inherently poor vibration char-

acteristics and attempts to reduce this to acceptable levels would add cost and

complexity to the engines. The following conﬁgurations for the new engine

were therefore initially selected: in-line six and eight, and V10, 12, 14 and 16-

cylinders. (V18 and V20 models were subsequently added to the production

programme.)

The range of V-angles for competitive engines was from 45° to 60°, with

a compromise angle of 50° favoured by several manufacturers. Angles 45°

were considered impractical due to close proximity of the lower end of the left

and right bank cylinder liners. For even ﬁring intervals and low torque ﬂuctua-

tions to give acceptable cyclic speed variations, optimum V-angles vary from

72° for a V10 to 45° for a V16 engine. An investigation of the effects of

V-angle on cyclic torque ﬂuctuations concluded that a common angle of 52°

was suitable for V12, 14 and 16 engines, while the V10 required an angle of

72° to optimize torque ﬂuctuation and balance.

Simulating the performance of the Allen 5000 series engine was conducted

using a WAVE engine performance simulation model based on the in-line six-

cylinder design. For the concept design stage, the basic engine design para-

meters, such as fuel injection characteristics and valve events, were chosen and

then ﬁxed. These were based on previous experience and data from engines of

a similar type with the aim of achieving good performance and emission char-

acteristics. The dimensions of the manifolding and air chest were typical for

this size of engine.

For the baseline simulation, an inlet cam period of 248° and an exhaust

period of 278° were chosen, with an injection period of 30° and start of injec-

tion at 13° before top dead centre. The port ﬂow data were typical of a current

port design with a good ﬂow performance. The trapped air–fuel ratio was set at

31:1 and a compression ratio of 14.5:1 applied.

The initial simulation model used a bore of 320 mm and a stroke of

390 mm. The results of the simulation showed that at the 10 per cent overload

condition, the predicted P

max

was 227 bar and the boost pressure 4.7 bar, which

were considered to be excessive. Increasing the stroke to 400 mm reduced the

max

to 202 bar and the boost pressure to 4.2 bar. Based on this initial bore and

stroke investigation, larger bore and stroke sizes were examined. This was con-

sidered necessary since there was a requirement for the engine to operate at

720 rev/min at the same rating of 525 kW/cylinder and also at the 10 per cent

overload condition, both of which result in higher cylinder pressures and boost

pressure requirements. Combinations of bore sizes of 320 mm, 325 mm and

330 mm and strokes of 410 mm, 415 mm and 420 mm were assessed for boost

pressure, P

max

and fuel consumption predictions.