Dragsted History Booklet 2013

The first 50 years of Turbocharged 2-stroke, Crosshead, Marine Diesel Engines
Jrn Dragsted
The First 50 Years

of Turbocharged
2-Stroke, Crosshead,
Marine Diesel Engines
Fig. 2 Scavenging systems in turbocharged, crosshead diesel engines (Author)
Year
Sulzer
Stork-Werkspoor
Mitsubishi
MAN
Fiat /GMT
Gtaverken
Doxford
Burmeister & Wain
1950
-P
SAD
RD
SAD72
RSAD
HOTLo75/150
KEBS68/125
UEC-A
UEC75A
KZ-C
KZ78/140C
...S
750S
VG-U / VGS-U
520/900VGS-U
LBD BDS
600LBD
-J
K98FF
K-EF/FF
1960
1060S
1970
RND105
RND
UEC-D
KSZ105/180
KZ-F KSZ
SW90/170
UEC-C
KZ-E
#
UEC-B
SW85/180
KZ-D
900S
B...S
K-GF
L-GF
KSZ-A
RNDM
Cross scavenging
two stage turbocharging
Electric driven blowers
RTA
1980
RLB
UEC-L
RTA-M
RTA-2
RTA-C
1990
RTA84T
RTA-2U
-LSE
RTA84TD
RTA-2UB
RTA-8T -TB
RTA96C
RTA60C
UECLSII UEC85LSC
UEC75LSII
UEC-LS
# impulse charging for 6,9 and 12 cylinders only
UEC-HA UEC-LA
KSZ-C/CL
UEC-H
RLA
Seahorse engine only
*with seperate exhaust camshaft
CC600
C...
58JS3C
-JS-C
K/L-GFC
K/L/S-MC/MCEmark 3,4,5,
K/L-GFCA
L-MC/MCE
K98MC
L-GB/GBE
K-MC-C
S-MC-C
KSZ-BL
UEC52E
UEC-E
KSZ-B
B...
-J-C
SEAHORSE project
750/1600 & 630/1400 VGS-U*

520/900VGS-U*
850/1700VGA-U
850/1700VGS-U*
VTBF
VT2BF
74VTBF160
VTBF opposed piston
Loop scavenging
Constant press. charging
Underpiston pump(s)
TURBOCHARGED TWO STROKE CROSSHEAD MARINE DIESEL ENGINES
Scav.principle
Uniflow scavenging.
Charging system
Impulse charging
Assisting air supply Seperate scav. pumps
Imprint
CIMAC
Central Secretariat
Lyoner Str. 18
60528 Frankfurt am Main
Germany
Contact:
Phone +49 69 66 03 13 55
Fax
+49 69 66 03 23 55
email cimac@vdma.org
Jrn Dragsted
The First 50 Years

of Turbocharged
2-Stroke, Crosshead,
Marine Diesel Engines
Jrn Dragsted
CONTENTS
INTRODUCTION 4
SETTING THE SCENE 5
I
1.
2.
3.
4.
5.
6.
7.
8.
POWER & RELIABILITY 6

The first turbocharged engines and some later generations 6
B&W 74VTBF160, 75VTBF150/50, VTBF, VT2BF, K-EF
Werkspoor-Lugt KEBS68/125
Stork HOTLo75/150, SW85/180, SW85/170
Gtaverken 520/900VGS-U, 850/1700VGA-U
Mitsubishi UEC75/150, UEC-A.D
Doxford 600LBD, BDS, -P, -J
Fiat 680S, ...S, B...S
MAN KZ78/140, KZC...F
SULZER SAD72, RSAD, RD, RND
Large bore engines 27
B&W 84VT2BF180, Doxford 76J, Gtaverken 850/1700VGA-U, Fiat 900S, MAN KZ84/160C,
MHI UEC85/160A, Stork/SWD SW85/180, SULZER RD90.
Super large bore engines 28
B&W K98FF, MAN KSZ105, Fiat 1060S. SULZER RND105
The last Stork-Werkspoor 2-stroke crosshead engines 32
SW85/170, SW90/170
Gtaverken stops 1972 33
VGS-U
Engine modernisation 35
B&W K-GF, GMT B, MAN KSZ-A, SULZER RND-M, Doxford Seahorse,
2-stage turbocharging 39
B&W L94NF, MHI UEC-E, MAN KSZC/CL-H
Fuel quality versus engine reliability 41
II FUEL CONSUMPTION, RELIABILITY & MULTIPLICITY 42

9.
Long stroke and slow running engines, engines for geared application 42
B&W L-GF, MHI UEC-E, MAN KSZB/BL, KSZC/CL, Sulzer RL, GMT C.., CC600, Doxford JC, JSC
10. The fuel consumption race 47

B&W, GFC, GFCA, SULZER RLA, RLB, MAN KSZC/CL, KEZC/CL , MHI UEC-H
11. Three survivors 53

MAN-B&W GB/GBE, SULZER RTA, MHI UEC-HA, MAN-B&W K/LMC/MCE, MHI UEC-L,UEC-LA,
UEC-LS, SULZER RTA-2, RTA-M,
Fuel consumption development and comparison
Engine failures
12. Engines for the big Container Ships 62

MAN-B&W K90MC/ MC-C, K98MC-C/ MC, SULZER RTA84, RTA84C,RTA96C, MHI: UEC85LSC
13. Engines for Very Large Crude Carriers, VLCCs 69

MAN-B&W S80/90MC, MHI UEC75LSII & UEC85LSII, SULZER RTA84-T/TB/TD
14. Medium and small bore engines up to year 2000 72

MAN-B&W MC-C, SULZER RTA-2U, RTA2U-B, RTA-8T, RTA-8TB, RTA60C, UEC-LSII, UEC-LSE
15. The future of 2-stroke crosshead engines 76

III DESIGN TIMELINES 77
REFERENCES 92
INTRODUCTION
This story of diesel engine development is, apart from personal experience, very much based on information found in CIMAC
congress papers, papers read in Engineering institutions and articles in the trade magazines, mainly The Motor Ship and not
least the many publications from the engine designers in question. It hurts a little to cut the story down to the present extent,
as so many valuable engineering ideas and in depth research results cannot be given credit.
It may be considered easy to judge the importance of ideas and the designs introduced in the past, but when going through this
story it should be remembered that any technical solution is only a compromise of certain advantages and disadvantages,
the original standard and tradition being of great influence as said by MANs Professor Srensen and Dr. Smidt.
The selection of the content is made to cover both interested engine designers and manufacturers but also the engine users,
who in great numbers, have served in the commercial fleets, down below.
My thanks to MAN-Museum and Historical Archive of MAN in Augsburg, The Werkspoor Museum in Amsterdam and to
good colleagues and friends Mr. Gran Dahlbring, Mr. K. Tayama, Mr. S. Yoshihara, Mr. Keith Wilson and especially Mr.
N. Hansen who as chairman of the CIMAC working group Engine Users have encouraged and assisted me in this work.
Furthermore, thanks to Dr. Thomas S. Knudsen, Mr. Christian Ltzen, Mr. T. Yamada and Mr. David Brown for taking time
to read and comment on the manuscript. And last but not least thanks to CIMAC Central Secretariat for their organisational
support and special thanks to the sponsoring companies Mitsubishi Heavy Industries and Wrtsil for financing the printing
of this book.
Setting The Scene

In October 1952 the M/T Dorthe Mrsk went to sea propelled by the first turbo-charged 2-stroke, crosshead diesel
engine. For years, and even during the war engine designers
had been calculating and testing the possibility of turbocharging 2-stroke engines to obtain increased power intensity, but
Burmeister and Wain (B&W) came first with their uniflow
scavenged crosshead engine, the 74VTBF160.
At that time no less than 9 designers of large 2-stroke engines
were fiercely competing in the shipbuilding market to supply the yards with powerful and reliable engines, dominated
by B&W in Copenhagen, Denmark, MAN in Augsburg,
Germany and Sulzer in Winterthur, Switzerland. The other
players were: Doxford in Sunderland, Great Britain, Stork in
Hengelo and Werkspoor in Amsterdam, Holland, Gtaverken in Gothenburg, Sweden, Fiat in Turin, Italy and far from
the European scene Mitsubishi in Nagasaki, Japan.
Apart from production taking place at the companys own

premises a substantial number of engines were produced under licence by big, mainly shipbuilding companies all over
Europe and in Japan. This meant, that the companies that
had strong allies serving the shipbuilding industry were well
situated to gain from an increased shipping activity. The demand for ocean transport after the war was rising by about
10% per year, and so was demand for ships and engines rising. Vessels dwt. was growing as well, and consequently the
need for higher power for propulsion.
Turbocharging of 4-stroke engines was already well accepted
by the market in the mid 1930s, but it is more complicated
to turbocharge the 2-stroke engines, as there is no separate
exhaust and air intake stroke, but merely a short period of
time around the bottom dead centre to empty/ push out the
exhaust gas and charge the cylinder with fresh air.
Fig. 1 The three scavenging systems: L Uniflow scavenging, Q Cross scavenging, U Loop scavenging. Time cross-sections (below) for inlet E,
outlet A and possible slide valves S. Right attainable degree of purity in the cylinder plotted against scavenging air excess s.(ABB [6])
The scavenging process of the engines designed by the mentioned companies were quite different, B&W, Doxford,
Gtaverken, Mitsubishi, Stork and Werkspoor used the
uniflow principle, Fiat and Sulzer used cross scavenging and
MAN preferred loop scavenging, Fig. 1, so it was not an easy
common solution, that could be found.
This story is just about 2-stroke low speed crosshead diesel
engines. Trunk engines made in parallel by the same designers
are not included, as they disappear from the market early in
the period considered, i.e. in the 1960s.
The history of turbocharger development has already been
published by two highly professional writers and is consequently not included either.
The BBC/ABB turbocharger development is described in a

comprehensive book The BBC Turbocharger by Dr.Ernst
Jenny [6]. A more widely covering book including and comparing most of the used chargers for 2-stroke engines is written by Mr. K. Imakire of Mitsubishi [86].
In Fig. 2 (inside cover page) an overview is shown of the development of the scavenging systems used in the 2-stroke
crosshead engines from the first engine types at sea and the
following 50 years. Further the figure shows, when the various engines types developed by the designers were introduced
(given by the position of the first letter), and when the companies for various reasons decided to cease operation. So, here
is given the answer to what and when hopefully the following will enlighten the reader in respect to why.
Setting The Scene5
I POWER & RELIABILITY

1. The first turbocharged engines and some later generations
Burmeister and Wain
Poppet valve engines
Supercharging of B&W engines was previously done by
rotating blowers according to the Roots principle, and the
74VTF160 long stroke engine was the first engine to be turbocharged. These single acting machines were designed to
ease maintenance, that was rather time consuming on previous double acting engines, and had a diaphragm between
crankcase and scavenging air belt to avoid contamination of
the lubrication oil, which was important for the use of heavy
fuel oil instead of diesel oil.
It was B&W tradition to apply uniflow scavenging, as the requirement to air supply was less demanding than in other systems. The crosshead engine with a poppet valve in the cylinder
cover was thus a relatively easy starting point for a conversion
from chain driven Roots blowers to turbocharging. The difficulty was the part load performance, or so it was generally believed. Actually B&W engineers calculations as well as BBCs
showed, that utilisation of the exhaust pulse when opening the
valve would make the system work even at part load.
The 6 cylinder 74VTF160 shown in Fig. 3 was thus able to
run just with the turbo blowers supplying air at all loads, a
considerable advantage compared to many competing systems,
as will be demonstrated later. The relative simplicity of the conversion is demonstrated in Fig. 4, however missing the increase
of exhaust valves and scavenging air ports as well as the capacity of the fuel pumps. BBC turbochargers were applied to a
start, but later Rateau, Napier chargers and chargers of B&Ws
own design were available. It was unacceptable for B&W to
rely on just one supplier for such an important engine component, and furthermore to risk that the knowhow gained by sub
suppliers would be spread to their other costumers i.e. B&Ws
competitors.
The engine structure shown on fig. 3 is all welded without staybolts. Alternatively the engines were designed in cast versions
with staybolts and this more traditional design was chosen
for subsequent engine types having higher ratings and as a consequence of crack formation in the welded structures.
Engine power was increased from 920 to 1250 BHP/cyl by an
increase of the mean effective pressure, (mep). And with the
few modifications this was a win win situation for manufacturers, shipyards and engine users, as the fuel consumption was
reduced from 162 g/BHPh to 158 g/BHPh. B&W earned a
fortune on the turbocharged engines, but sadly invested a lot of
it in the shipyard rather than in the engine facilities.
Fig. 3 Cross section of the first turbo-charged 2-stroke engine

74VTBF160, bore 740 mm, stroke 1600 mm, yielding 1250
BHP/cyl (MCR) at 115 rpm and Mep = 7 bar. (B&W)
only at the introduction of the later engine series K-EF and

K-FF was a reliable solution found (see K98FF).
Reliability and maintenance are closely related, but especially
the maintenance burden has undergone considerable development. Sren Hansen, B&W chief development engineer, reported at the CIMAC congress in 1955 [7], that the
cylinder wear was just as good as on the non turbocharged
engines, i.e. 0.2 to 0.4 mm/1000h. In those days seagoing engineers were used to frequent maintenance, so the result was
acceptable. Todays acceptance level would be less than 1/5!
The users comment of that time was generally very positive
it was a good engine.
B&W could, in 1959, present a high pressure turbocharged
84-bore engine as a start to a general uprating of the VTBFengines to high pressure turbocharging the VT2BF-engine series. These engines followed the same concept as the
74VTBF160 engine that definitely was B&Ws bread and
butter design. It was easy in those days for the sales department of B&W, so maybe they fell asleep and at the end
of 1964 Sulzer managed to beat B&W on market share. It
hurt the pride of B&W engineers, but more seriously it hurt
the top management, that was well aware of the advantage
of being number one. So, in 1965 it was decided to launch
uprated versions of the VT2BF series of engines the K-EF
types. This decision was an economical necessity, but B&W
were technically unprepared, so the design was nothing to be
especially proud of.
For turbocharged engines cooling of the compressed air is a
necessity meaning that air coolers are required after the turbocharger. However, the ever increasing scavenging pressure
means that condensed water need to be removed after the air
cooler as otherwise water droplets acting as projectiles when
entering the cylinders through the ports can lead to scuffing.
In the first turbocharged engines condensation is avoided by
increasing the scavenging temperature, later highly rated engines are equipped with so-called water mist catchers.
Fig. 4 The required modification of the standard unturbocharged

74VTF160-engine to provide turbocharging. (B&W)
The engine had two camshafts to simplify the reversing arrangement, which adds to cost and the exhaust valve and
valve gear was also relatively costly to manufacture, but who
cared knowing the overall gain. It was only with the following
engine series, that the design was changed to encompass just
one common camshaft. Of the other components the piston
design with a screwed on piston crown was causing trouble by
leaking oil and was later replaced with a shell type design with
support of the crown by the ring belt, however this piston,
which was also used in the following VT2BF engines, showed
cracks in the transition between crown and the top land, and
Opposed piston engines

B&W had, in parallel with their poppet valve engines, also a
series of opposed piston (or coverless) engines in the portfolio.
During the war Harland & Wolf in Belfast had independently developed the opposed piston type under C.C.Pounders
leadership, and it obtained just as Doxfords opposed piston
engine success in the UK. Consequently H&W put heavy
pressure on B&W to design opposed piston engines.
The first turbocharged opposed piston engine to be delivered
from B&W in 1955 was a 5 cylinder 50VTBF110/40.
A cross section of the lager sister 75VTBF150/50 is shown in
Fig. 5. The total stroke is 2000 mm, but the upper exhaust
stroke only 500 mm. A later execution had a combined stroke
of 2300 mm. The forces from the upper piston are transferred
to eccentrics on the crank webs. This resulted in a very stiff
crankshaft allowing B&W to build the engines with 5 to
10 cylinders, whereas the competing opposed piston design
The first turbocharged engines and some later generations7
could only build their engines

up to 6 cylinders (see later). The
engine is pulse pressure charged
and has no requirement for airblower or pump to assist at low
loads.
One different aspect of the Harland and Wolf built engines was
that they used an Archaoulofftype fuel injection pump. In
this pump compression pressure
from the engine cylinder was
connected to the top of the fuel
pump where there was a piston,
approximately 10 times larger in
cross-sectional area than the fuel
pump attached to it. Although
this meant that a separate supply
of cylinder oil had to be fed to
the upper part of the pump, it
negated the need for a camshaft
so saving on building costs.
The B&W opposed piston engine is gradually phased out
during the 60s as it was more
costly to produce than the poppet valve engine. A longer upper
stroke might have justified the
expensive running gear of the
exhaust piston to a certain degree.
Burmeister & Wains competitors were taken by surprise and
Fig.2 really illustrates this, as it
takes more than two years for
most of the competitors, before
they launch their new turbocharged engines. At B&W there
are speculations on the sort of
solution that they sooner or later will meet in the market place.
Technical director Haakon Andresen goes as far as to present
the expected moves by all the
competitors at meeting in the
Norwegian Engineering Society
in Oslo in April 1953, it could
be wishful thinking, but he did
not mention anybody copying the B&W system: Uniflow,
pulse pressure charging without additional air supply. He
forgets to include Mitsubishi,
and he is mistaken as far as Stork
is concerned [8].
Fig. 5 B&W opposed piston engine 75VTBF150/50, bore 750 mm, combined stroke 2000 mm,
yielding 1650 BHP/cyl (MCR) at 120 rpm and Mep = 7 bar. (B&W)
Werkspoor
Werkspoor are the first to follow B&W. The Werkspoor-Lugt
12-cylinder KEBS68/125 is not their first turbocharged engine, but the first real crosshead engine with turbocharging
[9]. The engine was installed in the passenger-cargo liner
Prins Willem van Oranje in 1953. The inventor of the
basic engine concept Mr. G.J.Lugt had passed away before
the turbocharging was introduced, but many of his ideas are
still maintained in the turbocharged version of the Lugt engine, a cross section of which is shown in Fig.6, with the turbocharger arrangement shown in Fig. 7.
Fig. 6 Cross section of the first turbocharged, crosshead WerkspoorLugt engine, KEBS68/125, bore 680 mm, stroke 1250 mm,
yielding 800 BHP/cyl at 125 rpm. Constant pressure charged
with engine driven double-acting pumps and a special exhaust
valve gear. (MS [10])
Fig. 7 12 cylinder KEBS68/125, main dimensions and turbocharging arrangement. (MS [10])
The 12-cylinder engine developed 9600 BHP at 125 rpm

and had a bore of 680 mm and a stroke of 1250 mm. The
supercharging arrangement is a constant pressure system
and consist of two BBC VTR630 chargers arranged in series
with the upper sections of the 12 engine-driven reciprocating
double-acting scavenge pumps, but at full load the chargers
will deliver directly to the scavenging receiver, the lower sides
of the pumps deliver directly to the scavenging receiver and
draw air from the engine room, see Fig. 8 [10].
Fig. 9 KEBS68/125 engine top. Note the special exhaust valve

gear. (MS [10])
The engine structure is of cast iron and the bedplate carries

short columns supporting the scavenging pump beam which
in turn carry the cylinder frame. The cylinders are in two
parts to facilitate exchange of piston rings without pulling
the piston. This feature adds to the height and weight of the
engine, but with the wear rates of the piston rings normally
ten times as high as the cylinder liner wear, it was perhaps a
sensible solution.
Fig. 8 Diagram of the Werkspoor-Lugt constant pressure scavenging

principle. (CIMAC [ 9])
The engine has 4 exhaust valves in cages and a very special

actuating gear operated from an extension of the scavenging
pump piston rod by means of cams and rollers and pull rods,
each rod operating a pair of exhaust valves, as seen in Fig. 6
and 9. A reporter from The Engineer found: the working
of this gear to be very smooth and practically silent, not unlike the valve gear of a modern horizontal steam engine [11].
There is no need for a camshaft, the reason for this is also
attributed to the fact that instead of having one fuel pump
for each cylinder, the fuel pump is contained in a completely
separate unit driven from the crankshaft through gearing between cylinder 6 and 7.
10
Like B&W Werkspoor focus in their first published service

records on cylinder liner wear. The average liner wear running
on heavy fuel is between 0.3 and 0.45 mm/1000h and of
corrosive nature, very much the same as found in the B&W
74VTBF160 engine. Werkspoor also found some occasional
burnings of the valve faces after 1.200 running hours, but
were able to improve the lifetime by the introduction of stellited valve faces.
In 1954 Werkspoor and Stork merge and new Stork-Werkspoor engines are now developed based mainly on the Stork
HOTLo concept [12] (see next chapter). The poque of the
turbocharged Werkspoor-Lugt cross-head engine was just
about one year and the Werkspoor-Lugt engine design
gradually disappeared.
Stork
The Stork engine is in scavenging respect like the B&W engine, but as Werkspoor, it also applies 4 exhaust valves in the
cylinder cover. Indeed the obvious system choice for Storks
turbocharged engines is also Uniflow, pulse pressure charging
without additional air supply at part load. Stork presents in
1954 their 8 cylinder HOTLo75/150, Fig. 10, where they
chose to apply one BBC VTR500-charger per two cylinders
in order to obtain as short exhaust pipe connections as possible and in this way to get the most out of the kinetic energy
in the exhaust pulses Fig. 11. This system is supported by
arranging a crankshaft with adjacent pairs of opposed cranks
so that the pulses for one turbocharger follow at equal time
intervals. Mr. Wieberdink mentioned in his CIMAC paper
at the congress in den Haag 1955 [13] that they are aware
that the intervals could be minimised to 120 as wanted for
a 6 cylinder engine, and further the fact that they have two
exhaust ducts per cylinder has been a helpful contribution
for the 5 and 7 cylinder engines, Fig. 12. The mean effective
pressure of the engine was 6 bar at 100% load.
Fig. 11 8 cylinder HOTLo75/150 engine with 4 turbochargers.

(CIMAC [13])
Fig. 12 Cylinder Cover with two outlets per cylinder, HOTLo75/150.

(CIMAC [13])
Fig. 10 Cross section of the first Stork turbocharged crosshead

engine, the HOTLo75/150, bore 750 mm, stroke 1500 mm,
1100 BHP/cyl at115 rpm and Mep = 6 bar. The engine is
impulse pressure charged. (CIMAC [13])
Why the choice of 4 exhaust valves? Today one valve is

standard, but at the time there are four expected advantages
that promote this solution: 1) a large time cross-area, facilitating scavenging 2) only one fuel valve is required, and this
valve can be placed in the cylinder cover in the centre of the
cylinder, 3) the thermal load of the components of the piston
crown and cylinder liner becomes more symmetrical and easier to handle and 4) the bosses for the exhaust valves supports
the cover bottom which then can be made relatively thin. The
clear disadvantage is the more complicated (costly) and main-
tenance requiring valve gear, Fig. 13. In fact the valve spindles
themselves becomes an extra burden because these are not
uniformly heated, however the valve seats are water cooled
in cages relatively easy to handle, Fig. 14, a feature that more
than 15 years later is seen on other engine makes. Mitsubishi
chooses, as it will be shown, three exhaust valves, and only
much later, in 1981 they solve their reliability problem with
the exhaust valves by changing to just one.
Fig. 14 Exhaust valves with water cooled valve seats, HOTLo75/150.

(CIMAC [13])
Fig. 13
Cylinder coover with exhaust valve gear, HOTLo75/150.
(CIMAC [13])
The engine structure is with welded bedplate, welded columns and cast iron cylinder blocks, all held together with
through-going staybolts. Thermal loaded components are
simple, Fig. 15 and the cylinder liner is protected from excessive heat load by the dish shaped cylinder cover allowing a
low position of the liner collar.
The firing order of the 8 cylinder engine is different from the
practise used by others, but Stork also comes back to usual
practise, when they later introduce their large bore engines
SW85/180, Fig. 16 or SW85/170 [14] and apply the same
firing order as used in B&W engines with 8 cylinders. Now
Stork also realize that the efficiency on 6 cylinder engines becomes higher if instead of using one charger per two cylinders
one use the 3-group with 120 between firing. So, the engine
mean effective pressure becomes 0.7 bar higher for 6, 9 and
12 cylinder engines than for the 8 cylinder engines (B&W
differentiate in the same manner when it comes to their KEF-engines), whereas the 7 cylinder engines is missing in the
programme.
12
Gtaverken
In the late thirties Gtaverken (GV) started design and
production of their own engines. Previously from 1915 to
1948 GV have had a license agreement with B&W, but due
to B&W in 1937 not being willing to design a single acting 2-stroke crosshead engine for them, they took on the job
themselves. GV were indeed so tired of their clients complaints about the workload of the B&W double acting engines (trouble acting as the engineers said) that they got the
push to realise some of their own ideas. Ironically B&W designed the crosshead engine that GV was asking for in 1939,
and this engine was the one converted to impulse turbocharging. The reason for this strange behaviour of B&W is not
known, it was certainly mismanagement, a lost licensee and
future loss of market share.
So, GV were actually introducing a single acting uniflow
crosshead engine ahead of B&W, whereas most people usually thought, that it was the other way round. The first turbocharged engine was a small 6 cylinder 520/900VGS-U,
Fig. 17, which went to sea in 1955 installed in M/S Axel
Gorthon, 3 years later than B&Ws 74VTBF160. It is not
possible to explain this hesitation.
Fig. 15 Piston mounted on rod connected to the crosshead,

HOTLo75/125. (CIMAC [13])
The 520/900VGS-engine is the first all welded engine (the

S stands for welding), and it was tested as early as in 1944.
When turbocharged it was named VGS-U (where U stands
for Turbocharged). The experience gained from the welded
type was so satisfactory, that new engines with one exception,
the 850/1700VGA-U, were all of welded design without
staybolts. VG-engines without further letter attachments are
just engines with cast bedplate and -entablature.
The GV engine was constant pressure charged with assistance of one or two crosshead driven, double acting scavenge air pumps for each cylinder and maintained this feature
throughout GVs existence. All other uniflow designs at that
time chose pulse pressure charging. The reason was to ensure
a high degree of operational reliability in service at sea with
complete combustion at all loads, while the pumps would
provide ample air for acceleration and manoeuvring.
Professor Collin tries in 1962 at the CIMAC congress in Copenhagen to justify the constant pressure system, Fig. 18, versus the impulse system [15], but realistically Gtaverken were
choosing a too complicated system at the time later with increased turbocharger efficiency available the priority changes.
The GV engines have various design features that deviate from
other brands and several will be mentioned later. However,
regarding the mechanism driving the exhaust valve mention
should be made. It was a unique characteristic of Gtaverken
engines retained since the birth of the first engine in 1939.
Fig. 16 Cross section of SW85/180, bore 850 mm, stroke 1800 mm,
2100 BHP/cyl at 115 rpm and Mep = 8.1 bar. (MS [14])
Each exhaust valve is operated by cam segments mounted on

the respective crank webs through rollers, levers and pull rods
to the yoke into which the valve stem is connected, clearly
seen in Fig. 19, and the valve motion is retarded by a nest
of four helical springs surrounding each pull rod being exerted between the cover and the yoke. The advantage of this
design is clear, the valve stem is actuated vertically without
horizontal side forces, and troubles with worn out steering
bushes and leaks of exhaust gas through these worn bushes
as in engines with rocker arm actuation has been avoided.
The disadvantage is that it conflicts with an optimal design of
the crankshaft and make overhaul of main bearings more difficult. Further, and with reference to the discussion following
a paper delivered by Professor Collin to the North East Coast
of Shipbuilders and Engineers in 1962, there was apparently
another disadvantage associated with exhaust valve gear: It
was noisy. The opponent said: I would prefer to see a poppet
valve operated by cam and push rod and as there is a camshaft
already on this engine in the right position that would represent a considerable improvement.
In 1960 a large bore engine is ready, the 850/1700VGA-U,
Fig. 19 and front cover. It was to a start moderately rated
with a mean effective pressure of 7.3 bar, but dimensioned
for a higher rating after sufficient service experience had been
accumulated, and in 1964 the mean effective pressure is increased to 8.7 bar, which is more on line with competition
(see table later). This large bore engine is a sort of a bastard
engine in the sense that it does not maintain the fully welded
features of the entablature. It has quite conventional A-frames
and staybolts. The reason was, that classi-fication societies at
the time of designing the engine were issuing new rules for
welded structures. The bedplate, however, remain welded as is
has been proved trouble free, and not a single crack has been
detected in the bedplate construction [3]. The engine does
otherwise maintain typical GV features:
the pioneering full width cross head bearing, but now with
oil channels milled in the pin instead of in the bearing
metal
Bosch type fuel pumps supplying two valves per cylinder,
two part cylinder cover with a cast iron section facing the
combustion chamber and a steel cast part as the strong
back
the unique exhaust valve actuation system previously
mentioned and
scavenging system with constant pressure turbocharging
Fig. 17 First turbocharged Gtaverken 2-stroke engine on test bed,

520/900VGS-U, bore 520 mm, stroke 900 mm, constant
pressure charged. (GV [3])
Regarding the latter it has been mentioned by former GV

experts, that the most delicate parts of the system were the
many non return valves in scavenge air system. These valves
should be exchanged when damaged in order not to spoil the
scavenging efficiency. This maintenance was often neglected
and caused increased thermal load on the engine components, especially hurting the large bore engines.
Referring to maintenance; Gtaverken took pride in designing for easy main-tenance and quick overhaul operations on
the test bed GV demonstrated that a piston overhaul could
be done in 58 minutes from stopping the engine to starting
up again with 4 men working on the top and two inside. This
was quite remarkable and is not done faster on the following
engines to be described.
14
Fig. 18 Diagram of the Gtaverken constant pressure charging

system. (CIMAC [15])
Fig. 19 Cross section 850/1700VGA-U, bore 850 mm, stroke 1700 mm, yielding 1850 BHP/cyl at 115 rpm and Mep = 7.3 bar. Note the special
cam segment for the exhaust valve drive on the crankshaft. (GV [3])
Mitsubishi, MHI
Mitsubishi, just like Stork resembles the B&W engine in respect of scavenging, but has 3 exhaust valves.
Mr. Fujita reports at the CIMAC congress in 1955 [16], that
MHI after many basic tests decided first to build a small,
turbocharged 220 mm bore experimental engine and later in
November 1952 a large size 3 cylinder, turbocharged experimental engine, 3UEC72/150. The first full size turbocharged
crosshead engine, 9UEC75/150 was manufactured in September 1954 and after the confirmation tests installed in M/S
Sanuki Maru of Nippon Yusen Co, that went in service in
May 1955.
In contrast to all other engine designers, that have involved
BBC when launching their first turbocharged 2-stroke engines, MHI courageously develops their own water cooled
chargers for the UEC-engines. They could not benefit from
BBCs growing experience with similar scavenged engines obtained at B&W and Stork. Not that MHI were unknown
with BBC chargers, also being a Sulzer licensee they should
soon test the first turbocharged Sulzer RSA76-engine in Kobe
in 1955/56 equipped with BBC VTR630 chargers [6].
The 75-engine is shown in Fig.20 in a side view as well as a
cross section. It features a cast bedplate of several pieces bolted together, A-frames, or columns connected to a box section
and the upper part consisting of the cylinder block cast in
several pieces bolted together and separate cylinder jackets.
Tie rods secure that it all functions as one block.
The hot components are all water cooled. Only the piston
crown is made of forged steel all other components are of cast
iron. With the 3 exhaust valves MHI obtain the same advantage as Stork with the four valves: The fuel valve gets an optimal position in the centre of the cylinder. The valve gear
is with Mr. Fujitas words ingenious and attractive and is
shown in Fig. 21, but maintenance of the many bearings and
bushings guiding the spindles is as in the Stork engines labour
demanding. 25 years has to pass before MHI with the introduction of the UEC-H-engines settle on a single exhaust valve.
Fig. 21 Engine top with exhaust valve gear, 3UEC72/150. (CIMAC

[16])
Fig. 20 The first Mitsubishi turbocharged 2-stroke crosshead engine, 9UEC75/150, bore 750 mm, stroke 1500 mm, 1333 BHP/cyl at 120 rpm
and Mep 7.4 bar. (MHI [2])
16
The camshaft to drive the exhaust valves is located along the

side of the engine crankcase near its top. Another camshaft
for fuel injection and starting air control is placed above the
exhaust cam shaft. Both cam shafts are driven by a train of
gear wheels from the crankshaft. As in B&Ws case the functions of the two shafts are soon combined in one.
Until 1960 the engine type is developed with 7 bore sizes,
and the series is finalized with the UEC85/160. The first
UEC engines have hereafter an A attached, and the following
engines of higher rating or of new design will have the next
letter in the alphabet attached.
A major improvement of the MHI engines is when in 1964
the MET uncooled turbocharger is introduced. Hereby expensive replacements of corroded gas inlet and outlet casings,
caused by sulphuric acids of the exhaust gasses, are done away
with. Whereas the MAN turbochargers from the outset were
uncooled, the other competitors introduced this feature at a
much later stage.
The UEC85/160 is in 1970 converted to a longer stroke as

UEC85/180D, Fig. 22, an easy design change with a uniflow
engine, but with an output of 2900 BHP/cyl at 118 rpm the
mean effective pressure is increased to 10.7 bar. This requires
in turn an increased turbocharger efficiency and compressor pressure ratio and leads to the introduction of the High
Pressure MET turbocharger. The increased thermal load also
necessitates introduction of water cooled exhaust valve seats.
Mitsubishi did not invest in development of a super large
bore en-gine (see later), but for their own demands they had
the Sulzer license and actually also a B&W license, that by
the way only was used to build a single 8K98FF engine at the
engine factory in Nagasaki.
Fig. 22 12UEC85/180D on testbed, bore 850 mm, stroke 1800 mm, 2900 BHP/cyl at 118 rpm and Mep 10.7 bar. (MHI)
Doxford
Doxford is synominous with opposed piston engines. They
designed their first engine in 1913 and continued the manufacture of such engines ever since. Doxford made significant
contribution to Britains survival during the war and the post
war rebuilding of the fleet. Even though someone considered
it a somewhat antediluvian engine at the time when Doxford
were contemplating turbocharging, it was certainly in high
favour by British shipyards and ship-owners, and the engines
were produced by a handful of British workshops on license.
The opposed piston concept have the inherent advantage
of being very well balanced and that the force flow due to
combustion pressure is transferred to the crankshaft as pure
torque and not transmitted to the engine frame, Fig. 23 [17].
The all-welded structure must, however, accept the torque reaction of the engine in the form of the horizontal thrust from
the guide shoes.
Actually Doxford were the first to order a large turbocharger
from Brown Boveri for their experimental 600LBD-engine,
but as mentioned they were not the first to get the design
ready for sale and being sent to sea. However, in 1954 the 3
cylinder 600LBD-test-engine with a total stroke of 2000 mm
was installed in M/V British Escort, where it could produce
3580 BHP at 120 rpm.
At the CIMAC congress in Den Haag in 1955 Mr. P. Jackson [18] is very optimistic about the power to be achieved
by turbocharging and concludes that powers up to 15.000
BHP in the largest six cylinder engine will (in future) be undertaken on a single shaft. The question about additional air
supply during low load and starting is more difficult than for
the poppet valve engines, as the lead of the exhaust crank is
limited for reversible engines. Doxford therefore decides to
arrange the turbo blower in series with engine driven reciprocating pumps rather than reducing the port heights and thus
the air amount and power.
The LBD engine design has one demerit that prevents Doxford in obtaining more power for a given bore and that is the
crankshaft, it is long and heavy and very flexible. Torsional
vibration conditions will not allow more than 6 cylinders and
bearings have to be mounted on spherical seats to prevent
edge contact. Experience with the LBD-engine and further
development activities resulted in the appearance of the BDS
range, Fig. 24 which incorporated many improvements and
dispensed with lever-driven scavenge air pumps, yet still
maintaining the same crankshaft design.
The P-type in 1960 does not solve the problem either, even
though the crankshaft has been shortened, lighter and stiffer,
a six cylinder engine is still the maximum.
Doxford are at this time accused of not producing an engine of large power, but they defend themselves by saying,
that from the commencement of the 2. World War to 1958
the companies engine works were fully occupied in producing engines for cargo ships and tankers and in this period of
18
Fig. 23 Pictorial arrangement of running gear and gas forces.

(Doxford [17])
intense production activity it was not possible to do much

development. Doxfords trade is in ships up to 18.000 tons
requiring about 7.000 BHP. But British shipbuilding is seeing mergers between yards, and bigger ships are planned just
at Doxfords doorsteps in Sunderland.
It became apparent that a design had to be found that would
permit more than six cylinders. This leads to the J-series of
Fig. 24 Cut away view of the 6 cylinder BDS engine with 1833 BHP/cyl. (MS [19])
Fig. 25 Comparison of crankshafts for 4 cylinder engines. (Doxford [17])
engines, not exceeding present bore sizes, but with a crankshaft design that allows up to 10 cylinders.
The development of the crankshaft design is shown in Fig.
25, where the J-type is on line with principles used in the
B&W opposed piston engine design promoted by Harland &
Wolff. Here eccentrics on the main crank webs are connected
to the side rods and the side crank webs form the main journals. Somewhat strange that Doxford not readily had adopted
this concept, as it was used for Harland & Wolff double acting engines since the war, and at a time where Doxford in the
beginning of 1950s had tremendous problems with cracked
crankshafts. The broken crankshafts turned the attention of
many of Doxfords former licensees and customers to other
makes of engines [19].
The 76J-engine is tested in 1963 and an elevation is shown in
Fig. 26. Now Doxford find themselves in the same league as
the continental companies and publish a table of comparison with all existing 20.000 BHP engines showing that the
J-type has a specific weight of 29 kg/BHP compared to 36
to 45 kg/BHP for the continental competitors, and that the
length of the engine is about 3 m shorter [20]. The combined
stroke of the engine is 2180 mm, so it is indeed a long stroke
engine with accompanying, attractive, low revolutions 120
rpm. On the elevation it is not possible to study the cylinder
liner design, but in Fig. 27 it is seen that it consist of 3 parts,
the upper and lower liner of cast iron and a steel cast combustion belt with bosses for two fuel valves, starting air valve and
safety valve. On the figure it is also shown, that drillings for
water cooling of the upper liner above the ports have been introduced this was a necessity to prevent the synthetic rubber
ring to harden due to high temperatures. Further cooling
has been intensified in the combustion zone of the upper and
lower liner with a large number of axial ribs that also serves to
transmit the combustion load to the steel jackets [21].
Fig. 27 Cylinder liner in 3 parts for 76J engines (MS [21])
20
Fig. 26 Elevation through 76J, 760mm, stroke 2184 mm, 3000 BHP/
cyl, at 123 rpm and Mep 10.9 bar. (Doxford [17])
This engine is pioneering the application of loose bearing

shells for the crosshead bearings more than 10 years ahead
of any competitor.
Fiat/GMT
Fig. 28 750S on test bed in Turin. (CIMAC [22])
Dr. Gregoretti mention at the CIMAC congress in Zrich

in 1957 [22] that Fiat started testing turbocharged engines
in 1952 and in December 1955 the first marine application,
a converted 8 cylinder 680S engine installed in M/S Sestrire, was sent to sea. It is a cross scavenged, constant pressure charged engine with scavenging pumps working in series
with the turbochargers, and this system was maintained in
all Fiat designed turbocharged, 2-stroke cross head engines.
However, the first engine specifically designed for turbocharging is the 750S with a bore of 750 mm and a stroke
of 1320 mm developing 1100 BHP/cyl. at 125 rpm with a
mean pressure of 6.7 bar, Fig. 28 and 29.
The scavenging principle is illustrated in Fig. 30 with its two

stage air compression system; the first stage is obtained from
the turbo blower and the second through the engine driven
air pumps, each with its corresponding cooling stage. To cope
with the increased thermal load and to maintain the same
time between overhauls as the non turbocharged engines i.e.
6000 to 7000 hours great care is invested in the design of the
hot components. The piston is a new design having material
temperatures matching the non turbocharged engines, but
with a 40 % higher mean effective pressure. Meticulous thermocouple measurements on the piston crown were carried
out and the outcome is illustrated in Fig.31, indeed a very
acceptable temperature level.
Fig. 29 Cross section of the first purpose built turbocharged Fiat engine, the 750S. Cross
scavenged concept with constant pressure charging and cylinder driven assisting
punps. Bore 750 mm, stroke 1320 mm, 1100 BHP/cyl at 125 rpm and Mep = 6.7
bar. (CIMAC [22])
Fig. 30 Diagram of the Fiat cross

scavenging system with
engine driven scavenge pumps
(CIMAC [22])
This piston design does not look like any of the competitors
design it is oil cooled, yet the heat transfer is just accomplished by circulation of the oil through oil channels without
any splash effect this design wont stand for higher mean
effective pressure as pointed out later. Quite unique is also
the number of piston rings 7, hereof 3 with chrome plating
of the running surface and 4 made of spheroid graphite
absolutely avant-garde. The cylinder liner design is also very
special, as will be illustrated later.
In 1959 Fiat launch a large bore 900S engine [23] after having
produced a 2-cylinder prototype for tests. The concept remain the same as for the 750S, but dimensions require some
modifications, e.g. of the cylinder cover, now made in two
parts as seen on a cross section of the combustion chamber,
Fig.32. The liner design consists of an upper part with a cast
iron bushing in a steel cylinder bolted to a lower part in cast
iron containing the exhaust and scavenging ports. The intention of the designers were to allow cheap retrofitting of the
cast iron bush when worn out or cracked and save the costly
lower part, but the price paid for this feature must have been
considerable. The two parted liner remains a design feature of
the Fiat engines as will be reported in later sections.
Fig. 31 Thermal load of the oil cooled piston, S750S. (CIMAC [22])
22
Fiat presents in 1964 an extremely short stroke engine, the

B600S with a bore of 600 mm and a stroke of 800 mm i.e.
MAN
MANs first turbocharged engine ready for sales was a
KZ78/140 [25]. MAN were late in introducing the turbocharged 2-stroke engine, but they were of course quite familiar
with the challenge, as they during the war had made various
tests on double acting 2-stroke engines and even produced
some. However, the war put an end to that, but it took until
1949 before MAN could manage to provide a new engine programme, and with quite an effort they were in 1953/54 ready
to board turbo charging. They felt forced by B&Ws success in
1952 having the M/S Dorthe Mrsk at sea with the 74-engine to get started and many test were accomplished, mainly
on the K6Z70/120-engine, with nearly all possible combinations of impulse/constant pressure exhaust systems combined
with under piston pumps in parallel, in series or in series-parallel with the turbocharger compressor.
MAN were sure that the loop scavenged engine was suitable for
turbocharging, whereas their licensees would have preferred a
change to uniflow scavenging, but MAN saw, as Sulzer did, a
lot of mechanical complications in production and service of
poppet valve engines the loop scavenged engine was in their
eyes preferable, because of its extraordinary simplicity as Professor Srensen and Dr. Schmidt
argued in a paper read in The Institute of Marine Engineers in
1964 [26]. They did however also say that: On the other hand
it should, however, be pointed out, that any technical solution
is only a compromise of certain advantages and disadvantages,
the original standard and tradition being of great influence.
They refer to facts, it is hard to leave the known tracks and do
something completely different.
Fig. 32 Combustion chamber 900S-engine. (CIMAC [23])
a stroke/bore ratio of 1.3 in comparison to the ratio 1.8 on

the larger engines the effective mean pressure is 7.7 bar.
This engine is mainly intended, without reduction gear, for
use in ferries.
In the late 60s Fiat presents a 780S engine with 2000BHP/
cyl and a stroke of 1800 mm [24] and a really new and interesting engine the 1060S superlarge bore engine, mentioned
in chapter 3.
The MAN turbo charging concept was also very pragmatic,

they did not go for a uniform solution. As a start they select
the impulse-parallel system for 6, 9 & 12 cylinder engines,
whereas they preferred impulse-series-parallel system for the
other cylinder numbers. Later, with a higher degree of turbo
charging they chose inpulse charging were it was preferable, i.e.
for 6, 9 & 12 cylinder engines and constant pressure charging
for the other cylinder numbers and until the introduction of
the KSZ series of engines (starting with the KSZ105/180) they
preferred the series-parallel supercharging system.
The turbocharging systems are schematically shown in Fig.33.
A cross section of the KZ78/140 is shown in Fig. 34. The mean
effective pressure is 6.4 bar compared to the 5.1 bar for the normally aspirated engine a power increase of 25 %. The rigidity
of the engine is accomplished through the welded bedplate and
the line of cylinder blocks bolted together. Between bedplate
and cylinder blocks A-frames and a two part entablature is arranged. Tierods hold the components together. The bottom of
the entablature is designed as diaphragms with stuffing boxes
for the piston rods, which permits the use of the piston undersides as scavenging air pumps. The piston is fresh water cooled
through telescopic pipes arranged outside the actual crankcase
in the smaller engine versions the pistons are oil cooled. The
cylinder cover is a strong back design with fresh water cooling
of the bottom part. Fuel injection pumps have plungers of the

Bosch type and are driven by a camshaft which in turn is driven
by a train of gears from the crankshaft. The crosshead bearing
is in two parts, and lubrication is secured by an attached high
pressure pump on each cross head.
The race for power continues, in Augsburg they are successful
and obtain already in 1956 a bigger market share than Sulzer.
MAN increases the power range and designs the 74-type with
an increased stroke of 1550 mm. To increase the stroke of the
loop scavenged engine is not so easy, and actually MAN made
64 tests of possible port configurations to find the optimal solution. Soon however the competition, and in particular the
demands from the Japanese licensees and shipyards, forces
MAN to develop a large bore engine, the KZ84/160, but within a year this engine is replaced with KZ86/160 in fact the
old engine with 20 mm bigger bore.
Until 1970 MAN launches 8 different engines of the KZ-type,
that concept wise are very similar to the KZ74/140C, with a
mean effective pressure ranging from 6.5 to 9.5 bar, each class
of engines are according to their design stage named C, D, E
and F.
Fig. 33 Impulse-parallel, Impulse-series parallel and constant

pressure-series-parallel turbocharging system (MAN)
24
Fig.34 Cross section of the KZ78/140, the first turbocharged MAN

crosshead engine to be sent to sea, bore 780 mm, stroke 1400
mm, 1140BHP/cyl at 118 rpm. Loop scavenged with Mep =
6.4 bar. (MAN)
Sulzer
Sulzers start on turbocharching was based on tests with the
2-stroke 6TA48 trunk engine already in 1941, and with the
RS58 crosshead engine from 1953 [4]. In the 50s Sulzer have
made trials with cross-scavenged engines applying pulse as
well as constant pressure charging and various under-piston
arrangements. They also compared performance with a uniflow scavenged engine featuring 4 exhaust valves in the cylinder cover just as in the Stork engines. This concept was
however discarded presumably because of the complicated
cylinder cover and valve gear, and probably the safe route
applying the SD-engine with small modifications, now time
was running out, was quite attractive. The first turbocharged
engine, the SAD72, Fig. 35 has its turboblower charging in
series with the scavenge pumps in a constant pressure system
as in the case of Fiat and Gtaverken.
As it can be seen in Fig. 2 the SAD72 engine is after one year

followed by the RSAD76, which also is a sort of fast adaptation to the existing cross scavenged RSD-type. The constant
pressure system used in the SAD-engines is now replaced
with a pulse pressure system, and the separate scavenging
pumps are discarded. This is possible because the engine has
a diaphragm so that piston under-sides could be used to supply the scavenge air. A series-bypass system was applied where
only a part of the air supplied by the turbochargers was led to
the piston uder-side pumps.
Fig. 36 RD90 first loop scavenged engine design from 1959, bore
900 mm, stroke 1550 mm, 2100 BHP at 115 rpm and Mep
8.1 bar. (Sulzer [27])
In 1957 the RD-series was introduced [27]. It was the first

engine types, designed for turbocharging from the outset. A
concept was established that characterised Sulzer engines the
coming nearly 25 years, Fig. 36. The engines were loop scavenged, impulse charged with exhaust rotary valves to omit the
long piston skirts, and the uderpiston pumps were working in
series with the blower. The cylinder porting geometry of the
loop system cause a reduction of flow resistance as the air flow
dont need to be deflected as shown in Fig, 37.
Fig. 35 First turbocharged Sulzer crosshead engine, SAD72. Cross

scavenged constant pressure charged with scavenge pumps,
Mep = 6.2 bar. (Sulzer [27])
What would have happened if Sulzer had chosen differently?

A contra factual story with great possibilities, as it will appear later. Be it as it may it was certainly a wise decision to
stick to a cross scavenged engine, soon converted to the loop
scavenging concept, and keep production costs down. Later
in 1964 the Sulzer loop scavenged engine should conquer the
leading position in the market place.
Fig. 37 Scavenging efficiency with different porting arrangements.

(Sulzer [27])
The basic outcome is that the RD engine has an output of

150 % to 170 % compared to previous unturbocharged engines of the similar size.
A marked design change is the water cooled pistons, where
the water is led into and out from the pistons through telescopic pipes placed inside a watertight enclosure to avoid
water contamination of the lubrication oil and consequent
corrosion of bearing journals. Sulzer claims that the increased
heat transfer to the combustion chamber components, a consequence of the increased mean pressure, necessitates water
cooling with superior heat transfer, and water cooling on the
other hand makes it possible to design the piston with radial
ribs supporting the piston crown. The new design is shown in
Fig. 38, where it is compared to the not very well designed,
oil cooled piston used previously.
The other thermal loaded components follow basically the
previous concepts.
The engine has welded A-frames, and the bedplate is welded
as well. Despite the application of pretensioned tie rods the
bedplate caused troubles as cracks developed in imperfect
welds under the main bearing saddle. The design was consequently modified with insertion of a forged or cast steel
part as bearing saddle instead of the 100 % welded design. It
should however be mentioned that this precaution was not
a guaranty against cracks poor castings has in several cases
been the origin of cracks in modern engines.
BHP/cyl
Fig. 38 Piston development for RD-engines, from oil cooling to water

cooling. (Sulzer [27])
Sulzer strengthen in 1967 their market position by the introduction of the RND series of engines [28]. These are now
constant pressure charged engines with under piston pumps
in series with the turbo blower. Hereby it is possible to get
rid of the rotary exhaust valve (a very welcome simplification), as the exhaust pressure in the receiver prevents the cylinder charge to escape. The first engine type is the RND105
(see later), and this engine is rapidly followed by 90, 76 and
68 bores, all of which gives 25 % more output than their
equivalent predecessors, which improves the market position.
The engine programme for the two arch rivals, B&W and
Sulzer, is compared in the table below and shows, that Sulzer
has chosen their bore sizes smartly, always being able to offer
more power with a comparable engine size.
RND105
K98FF
RND90
K84EF
RND76
K74EF
RND68
K62EF
4000
3800
2900
2540
2000
1900
1650
1360
The simplicity of the RND design is attractive for the engine

users and the popularity of this engine type, also by the yards
because of costs, makes it possible for Sulzer to consolidate
their market leadership, conquered from B&W in 1964.
26
2. Large Bore Engines

The technical competition between the eight engine designers is interesting to follow and will perused throughout this
history, but at the time when turbocharging was introduced,
diesel propulsion had a total market share below 50 %, the
majority was still in the hands of the steam turbines. The development of sea-going transportation also favoured steam
turbine plants, as the ship types became bigger, among others
triggered by the first closing of the Suez Canal in 1956
1957 and even more so when the Canal in 1967 was closed
for 8 years as a consequence of the Arab/Israel conflict. For
the ships of 23.000 tons deadweight and upwards the market
share of diesel ships were only around 20 % in 1954, but
due to introduction of larger engine bores and high-pressure
turbocharging the market share increased to 60% already in
1960 [29].
It was obvious, that the engine designers were keen to develop even more powerful engines, as they all would like to
grab a big share of the steam market. So in the beginning of
the 1960s the engine designers had their engines uprated to
higher mean effective pressure, mep and bore, as set out in a
table of large bore engines below:
Designer
Engine type
B&W
The challenge of the diesel engineer is always to design the

larger bores. This is primarily because the thermal load of the
combustion room becomes more and more delicate to handle, as the wall thickness to cater for the increased combustion forces has to be increased to maintain unchanged safety
against fatigue cracks in the design. The increased wall thicknesses mean on the other hand, that the surface temperatures
of the combustion chamber components will raise everything
equal. So, new designs are the only way to preserve reliability,
however, this was being learned the hard way.
Most of the components in the combustion chamber were
more or less just scaled to the bigger dimensions, and consequently there were generally many failures in service due to:
Too hot surfaces, leading to material burn down of piston crowns and exhaust valve bottoms and especially valve
seats
Fatigue cracks in pistons, cylinder covers and cylinder liners
Thermal cracks of cylinder liner surfaces and around ports
(loop & cross scavenged engines)
Bore mm
Stroke mm
RPM
Mep, bar
BHP/cyl CSR
Year in service
84VT2BF180
840
1800
110
8.5
2100
1960
Doxford
76J
760
2180
115
8.6
2222
1965
Gtaverken
850/1700VGA-U
850
1700
115
7.3
1850
1961
Fiat/GMT
900S
900
1600
110
8.3
2100
1961
MAN
KZ84/160C
840
1600
115
7.7
1800
1961
Mitsubishi
UEC85/160A
850
1600
120
8.1
2000
1961
Stork/SWD
SW85/180
850
1800
110
7.3
1875
1962
Sulzer
RD90
900
1550
115
8.1
2100
1960
The shipping society was not too happy with the engine development and in particular the Norwegian shipping circles
they organised a so called Large Bore Project with DNV
and NTH involvement, where cylinder component temperatures were to be measured at sea in selected engines running
on heavy fuel, which by the way was believed to be a major
influential factor for the high temperatures. Several of the engine designers were more or less forced to participate in the
project not to lose face.
Large Bore Engines27
3. Super Large Bore Engines

Despite the mentioned difficulties the power race was continued. 4 super large bore designs were presented in the late
Designer
Engine type
B&W
60s and in the beginning of the 70s, and are shown in the
following table:
Bore, mm
Stroke,mm
RPM
Mep, bar
BHP/cyl
Year in service
K98FF
980
2000
103
10.8
3800
1968
Fiat/GMT
1060S
1060
1900
106
10.0
4000
1971
MAN
KSZ105/180
1050
1800
106
10.7
4000
1969
Sulzer
RND105
1050
1800
108
10.,5
4000
1969
Cross sections of these engines are shown in Fig. 41, in the

same scale.
At the first look the loop and cross scavenged engines looks
somewhat simpler than the uniflow engine with its complex
exhaust valve drive, push rods and rocker arms. It is cheaper
built, but it is not so simple, the complexity is hidden in the
scavenging system with its many small valves and a somewhat
complex cylinder liner, and as an outstanding example the
1060S liner is shown in Fig. 39 a cross section as well as an
exploded view [30].
Sulzer decides to change their scavenging system, and the

RND105 is the first engine to have constant pressure turbocharging [31].
MAN also, finally decides for all their engines irrespective of
cylinder number to use a constant pressure parallel system
with a limited number of under-piston pumps in action plus
injectors, that only are used within the lower power range,
Fig. 40, [32].
Fig. 40 MAN, Constant-pressure parallel system. (MS [32])
Fig. 39 FIAT 1060S cylinder liner design, cross section and exploded
view. (CIMAC [30])
28
Fig. 41 K98FF, 1060S, RND105, KSZ105 (B&W [33], MS [24],

Wrtsil [31], MAN)
Super Large Bore Engines29
However, something had to be done to improve the reliability

of the engines in general, but especially of the hot components.
The K98FF is the first large 2-stroke engine where bore cooling is applied in the collar of the cylinder liner, and the
RND105 follows although the design differs.
Three of the engines get new piston designs:
The oil cooled piston of K98FF, Fig. 42 is a shell design with
inner support and the top clamped to the piston rod through
the piston skirt by means of a belleville spring this design
is used in B&W engines for the next 15 years. It replaces the
VT2BF-design with outer support of the crown, also shown
in Fig. 42 [34].
MAN also introduces a new piston Fig. 43, water cooled
as usual, but this time with a double backing of the piston
crown, by elastic pillars supporting at two diameters a clever way to avoid too high thermal stresses and temperatures of
the crown surface. Even the best concepts may, however suffer from some childrens diseases and MANs piston is no exception slight modifications of the supports were required
to eliminate fatigue cracks.
Fiat, who like most of the others, have had difficulties with
their piston design, made a new support of the piston crown,
where the channels for cooling oil were machined in the support element rather than in the crown behind the piston ring
belt. This reduced stresses in the area concerned. This measure should however soon become obsolete. Test were carried
out using water cooling instead of oil and despite previous
conclusions that the oil cooled wall temperatures were very
acceptable, Fiat decides to introduce water cooling even on
engines with 4000 BHP/cyl.
Fig. 43 Piston design development for the KSZ-engines. (MS [32])
Fiat argues that this is done to be prepared for a higher rating

of the engine due to normal technical progress.
Sulzer continues with water cooling. They are pretty happy
with their design that after modifications of the ribs, function well, without cracks. Actually they are so happy that they
announce in trade magazines: Sooner or later everyone will
apply water cooled pistons and they are exposing pictures of
the different temperature levels ref. Fig. 38. It was an unusual
announcement that indirectly told the readers that B&Ws
and Fiats piston designs were obsolete later in this report it
will be clear that Sulzer proved themselves wrong.
All the other thermally loaded components were more traditional and some created trouble in service. The B&W cylinder cover was a disaster-skilled welders had to be flown
around the world to repair cracked covers and spare covers
were in high demand. A new design was prepared and the
cover was split into two cast steel parts. Sulzer were also having difficulties with cracked covers.
The wall temperatures of these super large bore engines were
naturally a case of great interest to the users, and the temperatures was published in various technical bulletins as shown
below, Fig. 44, 45, 46, 47.
Looking at the temperature levels there seems to be a certain
agreement as to the desired maximum temperature of the pistons, and it is a bit difficult to understand why Fiat changes
to water cooling instead of a well-designed oil cooled piston.
Cylinder liner temperatures at the point of the upper piston
ring in top dead centre is between 140 and 170 C, low by
todays standard (see later), but adequate considering the firing pressure and fuel quality. B&W had a problem in keeping
the exhaust valve temperatures low enough to avoid thermal
corrosion, burning of the valve bottom.
Fig. 42 Combustion Chamber and fuel injection arrangements

comparison between K98FF (left) and the 84VT2BF180
engine (right). (MS [34])
30
Fig. 47 Combustion chamber temperatures RND105 at 4000 BHP/

cyl. (MS [28])
Fig. 44 Combustion Chamber temperature measurement, 7K98FF
and comparison of liner and ring groove temperatures of
84VTBF180 and K74EF engines. (B&W)
Fig. 46 Combustion chamber temperatures, 1060S with an oil cooled

(left) as well as a water cooled piston (CIMAC [30])
Fig. 45
Piston crown temperatures KSZ105. (MS [32])
Super Large Bore Engines31
4. The last Stork-Werkspoor 2-stroke crosshead engines

Stork did not present a super large bore engine, but they
preferred a different way to obtain high power by modifying
their SW 85/170 (the successor of the SW85/180). The new
engine, the SW90/170 was presented in 1966 [36,37] and
was in fact the engine with the highest cylinder output at the
time 2800 BHP per cylinder at 115 RPM. The high power
was a consequence of Stork being able to increase the bore
of the 84 engines without altering the cylinder distance and
increase the mean pres effective mean pressure to 10.2 bar.
After some turbocharger and fuel injection system modifications some of the first engines were indeed tested with up to
3333 BHP/cyl at a mep of 12.1 bar, a 6-cylinder engine is
here shown on the test bed at Hengelo, Fig. 48. Following
this success Chief Designer van der Molen declared that 3166
BHP/cyl should be regarded as the day in, day out rating.
Like the SW84/170 the 90-bore engine series did not contain
a 7 cylinder engine, and the 8 cylinder version had a lower
rating than the others. It was the intention that the 6 and 9
cylinder engine of the 84/170 series should fill the gap in the
power range. Stork had hoped that two stage turbocharging
would provide a means of giving equal cylinder outputs irrespective of the cylinder groupings. Experiments were in 1965
done in co-operation with BBC on their 3 cylinder 75/160
test-engine; however, this was of course a too costly way of
solving the equal cylinder rating question.
In general the engine characteristics were following the previous uniflow scavenged Stork engine, but with further emphasis on easy maintenance and prolonged time between overhaul. To that end the 4 exhaust valves were provided with
rotators (just as on the medium speed engines), a feature that
is generally introduced in the 80s for all other designs.
Furthermore Stork-Werkspoor made use of a full width
cross head bearing, pioneered by Gtaverken, which made
the previous cross head driven high pressure lube oil pumps
superfluous. Turbocharging was affected by BBC VTR750
chargers with non standard, plain bearings, the same that
Gtaverken previously had forced upon BBC, and not the
usual ball bearings.
Stork-Werkspoor was a relatively small company and could
not financially support the development that was required
to follow the leaders in the market place. Consequently they
decided to stop the development of low speed marine diesel
engines in 1969 and instead use their resources on the groving market for production of medium speed engines the
market was right at their doorsteps.
Fig. 48 SW90/170 at testbed in Hengelo, bore 900 mm, stroke 1700 mm, 3300 BHP/cyl at 115 rpm and
Mep = 12 bar. (MS [36])
32
5. Gtaverken stops in 1972

In 1964 Gtaverken launch a new series of VGS-U-engines
with bores of 850, 750, 630 mm, later to be followed by the
520 mm bore engine. This is indeed a modernisation and uprating programme. Now lessons learned in the past and new
possibilities are combined in this all welded design shown in
Fig. 49.
higher position of the valve spindle will together with the reduced material thickness contribute to a reduction of the spindle temperatures. Further better water circulation is arranged
in both the cylinder cover and liner, where the temperature
on the liner surface at position of the upper piston ring in top
dead centre is as low as 170C. GV contemplate water cooling
of the pistons, but refrain because of the risks of contaminating the crankcase due to water leaks. Instead GV decides to
do work on modification of the present piston design. The
result is a change of the contour of the inner part of the piston
as seen in Fig. 51, which reduced the maximum temperature
facing the combustion chamber with 120 C to 455 C, a very
acceptable level to avoid hot corrosion (burning) [38].
Fig. 50 Combustion chamber development for 850/1700VGS-U. (GV)
Fig. 49 Cross section 850/1700VGS-U, all welded structure and

separate camshaft, bore 850 mm, stroke 1700 mm, yielding
2400 BHP/cyl at 119 rpm and Mep = 10.8 bar. (GV)
The engines are as a start rated to a mean effective pressure of

9.2 bar, but to get any further thermal loaded components
had to be scrutinized to reduce temperatures and stresses.
In Fig.50 the modification done to the combustion chamber
is shown, most remarkable is the reduction of the exhaust
valve disk from 58 to 51 per cent of the cylinder bore, but the
Most remarkable is the modification of the characteristic

GV exhaust valve actuating gear. The pull rods are no longer
driven from a cam segment on the crankshaft web, but are
now connected to a fork shaped lever actuated by cams on
a separate camshaft. The actuating gear is contained entirely
in the crankcase which at least has some lubrication merits;
however the vertical actuation of the exhaust valve spindle,
which is the key parameter of the design, is still maintained.
Much less modification has been done to the crosshead design, Fig.52. The pin diameter has been increased to cope
with the increased firing pressure a consequence of the increased mean pressure, and the connecting rod ration has
been reduced from 1:4 to 1:3.6, which provides a higher periphery velocity, hence increased oil film thickness.
Gtaverken stops in 197233
The welded engines are now Gtaverken standard. The concept is interesting and different from what is used by all other
designers in the field. The entire entablature comprise one
cylinder only and staybolts are not used, Fig. 53. Instead a set
of bolts are securing the connection to the bedplate.
Fig. 51 Piston crown development for the 850/1700VGS-U engine.

(GV)
Fig. 53 Engine columns ready for assembly. (GV [3])
The units require minimum machined surfaces and further

the cylinder units can be completely assembled before being
placed on the welded bedplate. In the same way the units are
used when shipping the engine to the shipyards. GVs design
witness that they have faith in their welding procedures and
quality, and not to forget, they have gained a lot of experience since the first welded 520/900VGS engine was produced
in 1944. All other engine designers apply through-going stay
bolts to compress the welded areas of their bedplate and entablature and thus reduce tensile stresses, but this is not necessary on the GV engine as the design has a very modest
varying stress level.
This new engine series will more or less entirely be built at
Gtaverkens own premises; unfortunately the licensees only
build a couple and no 850 bore engine. Of the total 539 turbocharged GV engines to be built only one quarter is made
by the licensees and in the late 60s the market share of GVengines is pendling around 3 %. No engine development can
be paid by so few produced engines. To make things worse
Gtaverken have in co-operation with other Swedish engine
builders embarked on the development of a big medium speed
engine the UDAB project that never was commercialised.
Fig. 52 Crosshead design for VGS-U engines. (MS [39])
The reduced connecting rod ratio serves of course to reduce

the engine building height as a main purpose [39].
34
The most logical thing to do for Gtaverken was to continue

engine production in their factory based on a license, and
in 1971 they re-entered in an agreement with Burmeister &
Wain. The technical development staff and access to test facilities in Gothenburg was taken over by B&W for a period
of time, which actually helped B&W in their modernisation
programme the K-GF series of engines as mentioned in the
next chapter.
6. Engine modernisation
B&W
Several of the engine designers had in the beginning of the
70s engine series that needed a brush up for one reason or
another.
As mentioned previously B&Ws K-EF engines were just an
uprating of the VT2BF types. A thorough revision of many
components was required. With Carsten Andersen as new
technical director the design process was carefully planned
and followed up. The basic scavenging principles were maintained, but compared to previous designs new emphasis were
put on optimising production and mounting work. The engines were thus equipped with fully welded frame boxes with
mounted on guide planes. This was indeed also complying
with the users wishes for long time the B&W engines had
been compared with robust Danish agricultural machines,
whereas the Sulzer engines were compared to Swiss clockwork
to say it, B&W engines were not tight, and oil was seeping
out in many places.
Fig. 55 Slim Fuel Valve for all B&W 2-stroke engines (B&W)
Fig. 54 Bore cooled Cylinder Cover, principal calculation of

temperature distribution at Mep = 12 bar (below section
A-A, above section B-B) (MTZ [40])
The K98FF engine experience made it also quite clear that

some of the thermal loaded components needed upgrading.
The previously mentioned cylinder cover difficulties were
solved by the pioneering design of a bore cooled cylinder cover, now used in all modern engines irrespective of make. The
idea behind this design is briefly to control varying stresses
due to ignition forces by the height of the cover more or less
independent of the thermal stresses that in principle are proportional to differences between the mean tempe-rature of
the block and actual temperature [40], Fig 54. Even though
the basic concept was sound, there were once again some childrens diseases to cure such as cracks in welded-on cooling
jackets etc.
Slim, elegant, light, uncooled fuel valves were designed to fit

into these covers Fig, 55, a design really appreciated by sea
going engineers.
Further the engines were provided with a hydraulic actuation
system for the exhaust valve, Fig. 56, never seen before on
a two-stroke crosshead engine, however previously tested at
sea on a K84EF engine [41]. A major advantage apart from
lower first costs is the elimination of the transverse force acting on the valve spindle. This is an unavoidable consequence
of the rocker arm movement that causes high wear rates of
the guiding spindle-bushing something that Gtaverken,
as mentioned previously, solved in a different way. Today all
2-stroke engines use hydraulic activation.
The new engines series was named K-GF and consisted of
4 bores, a K90GF as the first followed by K80, K67 and
K45GF, where the old Gtaverken staff designed the 80 bore
engine. All engines with a stroke to bore ratio ~2 and a mep
of ~ 11.5 bar.
Engine modernisation35
GMT
In the beginning of the 70s Fiat Grandi Motori in Turin and
state owned Ansaldo S.P.A. in Genova, previously building
B&W engines on license, are combined in Grandi Motori
Trieste, GMT and move to a big, brand new diesel engine
factory in Trieste. Five years after the beginning of the GMT
production, modernisation and uprating of the previously
mentioned engines is prepared, and a new B600 long stroke
engine with a stroke to bore ratio of 2.08 is developed. The
B-series of engines then consists of: B600, B780, B900 and
B1060 [42].
The new engine series offers 15 % higher power at lower specific fuel consumption, about 150 g/BHP. The mean effective
pressure is close to 12 bar and the corresponding maximum
pressure 95 bar. The most important changes required to accommodate the increased thermal load and maximum pressure are:
1. Improved efficiency of the scavenging air coolers
2. Increased capacity of the injection system and reduced injection duration
3. Reinforcement of the engine structure and improvement
of oil tightness
4. Improved oil film in cross head bearings
5. New cylinder covers with reduced varying stresses
Fig. 56 Hydraulically activated exhaust valve for the K90GF engine

series with the gear consisting of a piston pump driven from
the roller guide by a cam disc on the camshaft and delivering
pressure oil to the working cylinder on top of the valve
spindle. (B&W)
Most interesting is the new crosshead design the so called

eccentric-type bearing where the pin and the matching
bearing shells are divided into two zones eccentric to each
other. The design is shown in an exploded view in Fig. 57.
The movement of the connecting rod will cause a lift of the
pin relative to the bearing
counterpart and allow oil
to fill the clearance while
the other part of the bearing is carrying the load.
The two bearing zones are
not having the same size
and the wider, inner part
takes the load when the
forces are biggest from just
before the top dead centre to somewhat after the
bottom dead centre. This
concept is applied in all following GMT-engines even
in engines with a full width
bearing.
The cylinder cover is now
made of a forged block
and bore cooled following
the same concept as in the
B&W and Sulzer engines.
Fig. 57 Crosshead design for the GMT B-type engines with eccentric
journals. (MTZ [42])
36
MAN
Sulzer
In 1968 MAN are, after discussions with licensees, ready with

a development program for modernized KZ machines to be
named KSZ [43] where the S stands for simplified Service.
The engine series consist of the 105-engine (to be updated)
and new KSZ 900/160, 78/155 and 70/125 engines. The effective mean pressure lies between 10.3 and 10.7 bar with a
firing pressure about 80 bar. These engines are all constant
pressure charged as mentioned in the Super large Bore section.
As just shown in the previous table Sulzer need to catch up in

order not to lose their leading market position, and in 1975
they are ready with an all metric version of the RND the
RNDM series of engines.
The essential design upgrades are:

1. Crosshead with full width bearing
2. Piston as for the KSZ105/180 for the two biggest bores
but only with a single annular support for the small engines.
3. Reduced connecting rod ratio (1:3.6) for the 3 smallest
engines, to reduce engine building height.
In 1973 the KSZ engines are replaced with KSZ-A types
where the mean effective pressure of the engines is increased
to 11.2 to 11.8 bar. Now a higher degree of turbocharging
is required and an exhaust diffuser is introduced after each
cylinder as also seen later. The supercharging system is further
simplified; the under-piston pumps are deleted and replaced
with electric blowers in series with the turbocharger, but only
for use at part load and during acceleration, Fig 58. Pistons
are once again modified and now a build-up type is preferred
to get control of the quality of the inner surfaces and thus
increase the fatigue strength of the crown. So in 1973 MAN
are well prepared for fighting their main competitors.
Their 90-bore engine is now on line with competition; however, something happens in parallel to this introduction, the
long stroke engines are marching in, as reported later.
Sulzer RND90M
BHP/cyl, MCR
RPM
Mep, bar
Pmax, bar
3350
122
12.3
92
The RNDM-types have the same bore, stroke and rpm as

the RND counterpart, however the higher overall efficiency
of turbochargers have facilitated an 15 % uprating with but
few design modifications: The cylinder cover is now as on
the B&W engines bore cooled and also with a lower split
between cover and liner to protect the liner from the highest thermal load. Actually the only really new modification is
found in the crosshead bearings that now are equipped with
thin tin-aluminium alloy lined steel shells. The fatigue limit
of this bearing type and its load carrying capacity is considerably higher than that of a white metal bearing the question
is if the bearing sensitivity to dirt is adequate white metal
has an unsurpassed imbedability to dirt particles. Sulzer make
use of well proved 4-stroke technology, but many 2-stroke
protagonists believe that it is essential to preserve 2-stroke
low speed features, where robustness is the keyword. However, no serious accidents happened caused by this new bearing design.
Additionally the RNDM had a new accumulator-type of cylinder lubrication system the CLU-1.
Fig. 58 Constant pressure system with electric blower in series for

use during low load. (MAN)
90-bore engines, 1973
BHP/cyl, MCR RPM Mep, bar Pmax, bar
MAN KSZ90/160A
3330 122
11.8
90
B&W K90GF
3410 114
11.6
90
Sulzer RND90
2900 122
10.4
84
However, the low number of revolutions of the uniflow scavenged B&W engine is attractive for the shipbuilder, as it
means higher propulsion efficiency.
Engine modernisation37
Doxford
Fig. 59 The Seahorse engine. Projected and tested to 2500BHP/cyl at 300 rpm and Mep 10.7 bar. (CIMAC [45])
In 1969, under the inspiration of Mr. E.P.Crowdy technical

director of Hawthorn Leslie & Co (a licensee of Sulzer and
Doxford), it was announced that Doxford and Hawthorn had
joined forces in a research programme on an engine to augment the Doxford J-series and provide powers up to 70.000
BHP. The concept was based on the view that: the optimum
marine engine must be able to burn the heaviest fuels with
complete reliability, have only one cylinder size to cover the
complete range, thus making possible world-wide provision
of interchangeable spares and must permit propeller speed
to suit vessels of all sizes [44]. Of course they were choosing an opposed piston crosshead engine, and not surprisingly
they choose a 580 mm bore engine with a combined stroke
of 1300 mm, developing 2500 BHP/cyl. at 300 rpm with
a mean effective pressure of 10.7 bar. A cross section of the
Seahorse, as the new engine was called, is shown in Fig. 59
[45]. The development work including testing of a prototype
engine was finalised in 1976 after many interruptions due to
financial difficulties.
38
The opposed piston engine is in principle very suited as a

medium speed engine, but the seahorse concept was actually positioned between two chairs slow speed and medium
speed, and the latter engine type was very cost competitive
with its higher number of revolutions. Further and unfortunately the Seahorse became a marketable product at a time
where shipbuilding was hit by depression, and the engine was
never sold.
Mitsubishi
Mitsubishi modernise their engine programme in a way that
many have talked about for quite some time, as reported in
the next chapter.
7. Two-stage Turbocharging
I the first half of the 70s engine designers sees an end to
further exploring the power potential of the engines as the
turbochargers not have the efficiency to yield a higher pressure ratio in a single stage. Two-stage turbocharging becomes
more realistic, than it was for Stork in the 60s.
Sulzer first undertook tests with two-stage turbocharging in
the mid 1960s and continued in 1975 on three cylinders of
an 8RND90, with a mean effective pressure up to 16 to 17
bar. However, the added complications and costs prevents realisation of new 2-stage turbocharged designs.
B&W have as a consequence of the increased demand for
high powered super tankers where a 300.000400.000 tdw
tanker would require 30.000 to 45.000 BHP at 80 to 70 rpm
drafted a project, the L94NF with 5500 BHP/cyl at 83 rpm
and a mean effective pressure of 17 bar. 2-stage turbocharging
was the only way to reach this high rating [1]. The new license
agreement with Gtaverken included as mentioned the possibility to utilise GVs test facilities, and 2-stage trials were run
on the 3 cylinder 750VGS-U in Gothenburg, Fig.60, where
even Mitsubishi were invited to the testbed. The total collapse
of the tanker market during 197374 made B&W burry the
project.
Fig. 60 Two-stage turbocharging test at Gtaverkens premises in

Gothenburg performed on a 3-cylinder 750VGS-U engine
(B&W)
Mitsubishi finds in 1975, that the best way to improve cost

performance of their engines is to increase the power rate and
introduce 2-stage turbocharging their E-type engines [46].
The E-type obtain abt. 30 % higher power output within the
same frame size as the predecessor D-type with single stage
turbocharging, meaning also a substantial saving in engine
room space requirements.
The impressive potential in increasing the mean effective
pressure without compromising thermal load was demonstrated by Mitsubishi, who so far are the only company to
commercialise 2-stage turbocharging.
The first production unit is an 8UEC52/105E, and the arrangement of the 2-stage turbocharging system is shown in
Fig. 61, and the engine cross section in Fig. 62. To accommodate the increased firing pressure in the cylinders the engine
frame, bearings and combustion chamber components are all
reinforced relatively to the D-type and the liner is bore cooled
in the upper part as in the B&W KG-F-engines.
Fig. 61 Two-stage turbocharging system applied to 8 cylinder

52-engine (MS [46])
Two-stage Turbocharging39
supplemented with a 52/125E and a 60/150E long stroke

engine with a stroke to bore ratio of ~2.5, see later.
The service result of these engines in respect of cylinder liner
wear is 0,01 to 0,04 mm/1000hours, which must be considered very satisfactory and the ring wear is about 10 times
higher, which must be considered normal.
MAN announces [47] in February 1980 that they have adopted 2-stage turbocharging in their KSZC/CL-type of engines
(see later). They have developed turbochargers where both
stages are accommodated in one casing. In the 1980 production programme these engines are indentified with the suffix
H which stands for high efficiency turbocharging.
Fig. 62 The worlds first 2-stage turbocharged 2-stroke engine,

MHIs 8UEC52/105E, bore 520 mm, stroke 1050 mm, 1330
BHP/cyl (980 kW/cyl) at 175 rpm and Mep 15.1 bar. (MS
[46])
Fig. 64 Comparison of combustion chamber wall temperatures

between the single stage and the two stage turbocharged
UEC52-engine. (MS [46])
Fig. 63 Comparison of engine performance between the single stage

and the two stage turbocharged UEC52 engine. (MS [46])
It is interesting to compare the performance of the D and

E-type and Fig. 63 shows clearly that the increased power is
not bought for free; the maximum pressure is now 110 bar in
order to maintain the specific fuel consumption.
The comparison on thermal load is shown in Fig. 64, and the
temperatures are all moderate maybe not surprising as the
cylinder diameter relatively small, but the remarkable thing
is, that the highly rated 2-stage turbocharged engine actually
comes out much better than the predecessor. In addition to
the normal stroke to bore ratio of ~2 the engine series is
40
The engine power remains the same, but the fuel consumption between 80 and 85 % MCR is reduced by 3 g/BHPh,
however, the engine price will be 4 to 5 % higher. To quote
Mr. Schiff, the technical director: Every ship-owner can
work for himself whether, at the present-day and future prices
of heavy fuel oil, the saving in fuel will offset this higher capital expenditure. The H-engines never went to sea.
8. Fuel quality versus engine reliability

It was an opportunity as well as an associated problem that all
the just mentioned crosshead engines were running on some
kind of heavy fuel.
The opportunity is related to the fact that the trunk engines
at that time did not run safely on heavy fuel, but the competition in form of steam plants could make use of these cheap
bunkers now the crosshead engines could do that as well.
It was, however, not an easy task for the engineers to make it
happen, mainly because heavy fuel is just a common name for
what in fact is a variety of products with very, very wide specifications and also containing a lot of unspecified material.
Normally the bunkers are classified according to the fuel viscosity in order to handle it, but additionally there are several
handling parameters [48], such as specific gravity, pour point
and flash point. As far as the engine condition is concerned,
the conradson carbon and asphaltenes are influencing fouling
of gas-ways and compatibility with other oils, sulphur, ash,
cat-fines and water has an impact on corrosive wear, mechanical wear, thermal load and fouling, and lastly sodium and
vanadium could be harmful to exhaust valve corrosion and
turbocharger deposits.
Burning of heavy fuels has been and is still a subject of great

interest to designers as well as engine users, but to get into a
discussion of this subject is beyond the scope of this report.
One can, however, notice that many, many discussions between designers and users has been related to the fuel influence if this or that component failed because of the fuel
used or, if it was a question of poor design.
The early use of heavy fuel showed that something had to
done to save the cylinder condition. Something beyond the
capability of the designers. The helping hand came from the
oil industry with their invention of highly alkaline cylinder
oils to combat the corrosive cylinder liner wear.
Fuel quality versus engine reliability41
II FUEL CONSUMPTION, RELIABILITY &

MULTIPLICITY
9. Long stroke and slow running engines, engines for geared application
B&W
The oil crisis 1973-74 increased the interest in fuel economy.
At B&W the development department had for some time
been considering to design a long stroke version of the K-GF
series of engines. The mean piston speed and effective pressure should be retained to maintain engine reliability, but the
reduced engine rpm would provide a substantial improveEngine
ment of the propeller efficiency provided the yards also agreed

to modify the hull design. In 1976 the first L67GF engine
was ready for test at Mitsui it was a relatively simple job to
design the new series, as the majority of the components and/
or concepts were reused. A comparison of the K and L-series
of engines is shown in the table below and in Fig. 65.
K90GF
L90GF
K80GF
L80GF
K67GF
L67GF
L55GF
K45GF
L45GF
Stroke,mm
1800
2180
1600
1950
1400
1700
1380
900
1200
BHP/cyl
3410
3410
2640
2640
1870
1870
1340
880
850
114
94
126
103
145
119
150
228
164
rpm
It was obvious, that the long stroke engine would put the
loop and cross scavenged engines under heavy pressure as
these concepts require a relatively short stroke because of the
scavenging process in order to obtain high air purity in the
cylinders and low fuel consumption. B&W had hoped that
the long stroke engines would bring back some of the lost
market-share, but they were surprised of the ingenuity of
their main competitors to cope with the new situation, as will
be seen in the following.
Mitsubishi
At the same time as B&W Mitsubishi had, as previously
mentioned, included two long stroke engines in their 2-stage
turbocharged engine programme; the UEC52/125 and
UEC60/150. These engines are, as the B&W engines having a stroke to bore ratio of ~2.5 and retain by and large the
mean piston speed and pressure as on the standard engines,
from which they design-wise are hard to distinguish. The
long stroke programme is later completed with the addition
of two smaller bore, constant pressure charged engines. This
so called H-type will be dealt with later.
MAN
Fig. 65 Comparison of short and long stroke engines, K/L67GF
(B&W)
B&W claim that the 18 percent rpm reduction will result in

approximately 5 % reduction of fuel consumption, actually
3 to 5 % would have been more realistic, and on top of this
about 2 g/BHPh reduction in fuel consumption because of
the more favourable stroke to bore ratio.
42
MAN has from 1975 replaced the KSZ-A with an uprated

KSZ-B-type. The B engines are more or less a copy of the
A-type having the same stroke to bore ratio, only the mean
effective pressure is increased and so the power in the range
between 9 and 15 % dependent on the engine type.
More interesting is the addition, the BL-type that is found
on the sales list in 1977. The BL version is basically of the
same design as the B version, however being slow running
yet maintaining identical maximum power. MAN has simply

reduced engine rpm by 10 % and at the same time increased
mep by the same percentage, and claim that the thermal load
of combustion chamber components is proportional to the
product of mean piston speed and effective pressure [49].
This is based on an empirical formula, however, the measured
temperatures as shown in Fig. 66 clearly shows that wall temperatures are inclined to follow mep. Furthermore the fuel
consumption is increased with ~2 g/BHPh relatively to the
B-version. The BL-engine is an uprated B-type in disguise
and has no added value to offer in terms of fuel consumption
at full load. However, at part load a new variable injection
timing system provides for a consumption reduction of ~ 3 g/
BHPh at the most, around 75 % load.
Fig. 66 Comparison of combustion chamber temperatures between

KSZ70/125B and KSZ70/125BL. (MAN)
Fig. 67 Cross section of KSZ90BL, bore 900 mm, stroke 1600 mm,
3670 BHP/cyl (2700 kW/cyl) at 110 rpm and Mep 14.5bar.
(MAN)
The B-Engine design is much more attractive, as shown in

Fig. 67. The construction of the engine frame now consists
of a welded box (two for the 90-bore engine) instead of the
A-frames. The smaller engines have got bore cooled pistons,
with finger bores as in the Sulzer 56RLA design (see later),
but with a quite different way of directing the coolant into
the bores as illustrated in Fig. 68. This cooling method is also
applied in the outer cylindrical region of the cylinder cover,
Fig. 68, where it elegantly is combined with the traditional
design. The 52-engine has even a bore cooled cylinder liner
collar, and all engines are provided with cast-in cooling pipes
between the exhaust ports.
For all the KSZ-B-type engines the cylinder cover is still
a strong back design, where the bottom is clamped to the
top by several studs thus prestressing the bottom plate. It is
however difficult to avoid relative micro movements between
the two parts and many covers of the big engines suffer from
cracks in the bottom, steel part. It is difficult to understand
why MAN not have adopted a fully bore cooled design, as
most of the competitors, and as they have done in the piston
design.
Fig. 68 Cylinder Cover and Piston, KSZ52/105. (MTZ [50])
Long stroke and slow running engines, engines for geared application43
Sulzer
Sulzer has just finalised their RND-M series of engines, but
they have prepared themselves for the new era of low revolutions, as they in 1976 announce a 56 bore engine with a
stroke to bore ratio of 2.05 the RLA56, to be followed in
1978 by RLA90, and in 1979 the RLA66 with a stroke to
bore ratio 2.102.12. This stroke to bore ratio seems to be
the limit for the loop scavenged engines, and MAN reach the
same conclusion.
GMT
In 1979 after 3 years of intensive study and development GMT
introduces the CC600-engine as their answer to the increasing
demand for low fuel costs. Whereas B&W have chosen to design a range of engines with long strokes and low revolutions in
addition to their standard programme GMT have, as Doxford
with their Seahorse engine, opted for a short stroke (stroke/
bore = 1.3) engine ideal for geared installations. This gives of
course the freedom to select optimum propeller speed, with a
limited number of engine designs and provides also ideal conditions for use in vessels with limited head room. It is indeed a
competitor to 4-stroke medium speed engines, as the 2-stroke
characteristics of simplicity and robustness is maintained, but
costs are relatively high.
The introduction rating is 1650 BHP/cyl at 250 RPM and
12.9 bar mean effective pressure and the fuel consumption is
146 g/BHPh.
The concept retains all the traditional GMT features, cross
scavenging, constant pressure turbocharging, scavenge air
pumps and the open under-piston space, but some new features are of interest.
All the components surrounding the combustion room are
now planned to be of bore cooled design and water cooled,
Fig. 69. The first engines will however have the normal strong
back liner design, but when it comes to ratings exceeding the
introduction figures it is all bore cooling. The liner design is
especially interesting because the bores ends blindly which
means that high stress concentrations are avoided in the top of
the liner collar. This concept is later copied by MAN-B&W in
their MC-engines.
The crankshaft is now fully forged, and the crosshead is having
a full width bearing, however still maintaining the eccentric
pin feature introduced in the B-type engines.
44
Fig. 69 Combustion chamber with finger bore cooled liners for the
CC600 engine. (GMT)
The final 2-stroke crosshead engine programme consists in

1980 of:
Engine type
Bore, mm Stroke,mm
Power, BHP RPM Mep, bar
B1060
1060
1900
4600 106
11.5
C900
900
1600
3700 110
13.1
C780
780
1600
2800 122
13.2
C600
600
1250
1650 160
12.9
CC600
600
800
1650 250
12.9
In mid 80s GMT decides to take a license from New Sulzer

Diesel and that stops the development of some of the most
interesting 2-stroke crosshead engines, unique in their concept and in many components .
Doxford
The J-engine was until 1975 designed with a 4 cylinder configuration as the smallest in the engine series having cylinder
diameters of 580, 670 and 760 mm. After the experiments
with the Seahorse engine it was found that constant pressure turbocharging instead of impulse charging would provide several advantages; one being that the side cranks which
are connected to the upper piston, could be given a crank
lead of 180 relative to the centre crank, which is driven by
the lower piston whereas the impulse charged engines would
require 188. This in turn means that it now was possible to
design engines with just 3 cylinders and obtain an even more
Engine type
excellent mechanical balancing. As compactness is one of the

competitive edges of the opposed piston engines the addition
of the three cylinder configuration was an advantage ad par
with the reduced number of components to overhaul and the
increased reliability. An elevation of the 76J-C engine, as the
constant pressure charged engine is called, is shown in Fig.
26, and the engine is seen on the test bed in Fig. 70.
In 1978 Doxford introduce short stroke versions of the J-C
series of engines named JS-C, and the total engine programme in 1978 is seen below:
BHP/cyl
RPM
Stroke mm
MEP, bar
No of Cyl.
58J-C
1750
150
1850
11.52
58JS-C
1833
220
1220
11.42
36
67J-C
2500
127
2140
11.52
36
67JS-C
2500
160
1700
11.51
36
76J-C
3000
123
2184
10.87
39
76JS-C
3000
155
1870
10.07
37
Especially the short stroke version of the 58 bore engine, Fig.

71 is seen as a competitor to medium speed engines, but just
as the Seahorse engine this engine was not a success, maybe
because British shipbuilding at that time was shrinking. The
fuel consumption of the listed engines was quoted to 145.5
g/BHPh which was too much compared to contemporary engines of other makes, but Doxford have apparently not the
tools at hand to make use of the inherent advantages of their
super long stroke engines.
Comparing the 58J-C to B&Ws L55GFCA (see next section) as in the table above it is seen that the specific fuel
consumption of the 58J-C engine is 6 g/BHPh higher than
the B&W L55GFCAs. The influential parameter S/D is in
favour of Doxford, whereas the Pmax/MEP relation is in favour of B&W in this case they more or less outweigh each
other. The question then remains, why is the 58J-C engine
having such a high consumption? The main reason is that the
maximum pressure is too low, but it cant be higher because
the pressure influenced components are not designed to the
higher values maybe as a consequence of Doxfords limited
experience with the use of modern calculation tools. Looking
at the design of the cylinder liners, cast in one piece and bore
cooled, Fig. 72, it is clear that the potential for higher cylinder pressure is present.
Fig. 70 J-type on testbed at the Pallion works in Sunderland.

(Doxford)
Other areas could influence the consumption, and looking

at the piston design, Fig.73 suggests that the very low positioned piston rings will lead to big pressure losses in the scavenging ports and the strange wave cut top land would require
a clearance that is higher than with a straight cut surface. The
temperature level in the piston crown of this small bore size
is low enough to allow a higher position of the piston rings.
Long stroke and slow running engines, engines for geared application45
Fig. 71 Doxford 58JS3C engine, bore 580 mm, stroke 880/340 mm, 1833 BHP/cyl (1349kW/cyl) at 220 rpm and Mep 11.4 bar. (Doxford [17])
Engine type
BHP/cyl
RPM
Bore mm
Stroke mm
S/D
MEP bar
Pmax bar
Pm /MEP
SFOC* g/BHPh
58J-C
1750
150
580
1850
3.2
11.52
85
7.4
145.5
L55GFCA
1495
155
550
1380
2.5
13.00
100
7.7
139.5
* ISO ambient conditions, and fuel oil calorific value: 10.200 kcal/kg
A pity that Doxford chose to
modernise their engine programme with the Seahorse
engine instead of modernising the design of the super
long stroke engines. Doxford
is closed down in beginning
of the 80s despite Sir Robert
Atkinsons (Chairman of British Shipbuilders) desperate attempts to give the Sunderland
Works a new chance.
Fig. 72 Cylinder liner for J-C type engines. (Doxford [17])
46
Fig. 73 Piston for 58JS3C. (Doxford [17])
10. The fuel consumption race

B&W
It took the maritime world by surprise when B&W in the
summer 1978 suddenly, but well planned introduced a completely new engine series with 7% lower specific fuel consumption. During many years B&W had made their propaganda in relation to the benefits of Uniflow scavenging and
inpulse turbocharging without assistance of part load blowers. The new engines were constant pressure charged! The
gradual increase of turbocharger efficiency had made it possible to apply this system with just small auxiliary blower capacity at part load. B&W may have become a victim of their
own propaganda, as it definitely would have been possible to
introduce constant pressure earlier. The gain in engine efficiency is illustrated in Fig. 74, the P-V diagram of the engine
cylinders.
The new charging system made it possible to offer engines
with identical rating irrespective of cylinder number, what
Stork so desperately wished in 1965, and even with better reliability as the exhaust temperature during the opening of the
exhaust valves were lower which reduced the material temperatures of the valve spindles and the valve seat a sore point
for B&W in the competition. In November 1978 the new
concept were tested on a 7L67GFC-engine in B&Ws workshop in Copenhagen and the expectations were fulfilled [52].
Apart from the scavenging system the constant pressure engine is a copy of the impulse engine, the K/LGF, meaning
that it is a fairly simple and cheap new development.
At the CIMAC congress in the spring 1979 Mr. A. stergaard (B&Ws chief design engineer) presented a paper [53],
that by technical and economical argumentation should clear
the way for a further power increase of the constant charged
engines and document the large layout flexibility, of course
with great emphasis on the fine, low specific fuel consumption. So, from spring 1979 the engines were rebaptised and
offered with a 15 % increased power and unchanged specific
fuel consumption the CFCA-type was born. A cross section
of the 12L90GFCA-engine is shown in Fig. 75; this engine
has a specific fuel consumption of 142 g/BHPh at 95 bar
maximum pressure.
Fig. 75 Cross section of the L90GFCA engine, bore 900 mm, stroke
2180 mm, yielding 3940 BHP/cyl (2900 kW/cyl) (MCR) at
97 rpm and Mep 12.9 bar. (B&W)
Fig. 74 Arrangement of exhaust pipes for 6 cylinder impluse- and
constant pressure charged engines, and comparison of the
lower part of the engine working diagram for impulse- and
constant pressure charged engines (MS [52])
The fuel consumption race47
Sulzer
The B&W constant pressure engines put pressure on Sulzer
to make a counterattack. They have already in September
1977 released that the RLA90 was under development. In
the autumn 1978 at the SMM exhibition in Hamburg Sulzer
announce the new RLA90. Peter Sulzer, vice president, had
previously told the press [54], that it was Sulzers aim to maintaining their position as market leader by providing total
economy it was obvious that the threat from B&W was
recognized.
Except from the increased stroke to bore ratio and following
reduced revolutions the RLA engines differ from the RNDM
engines with a 3 g/BHPh lower consumption i.e. 146 g/
BHPh.
Already a year after the introduction of the RLA engines,
Fig.76 a RLB series of engines are offered with two ratings. A
full rated engine (10 % higher rating than the RLA) as usual
and a derated engine, a so called economy rated engine, with
lower BHP and rpm, which provides for a fuel consumption
reduction of 3 g/BHPh, Fig 77.
Simultaneously an optimisation of combustion and scavenging combined with use of the new VTR4-series turbo-chargers, with higher efficiency and pressure ratios, Fig. 78 & 79, is
respon-sible for further reductions in fuel consumption. The
development of turbochargers comes in very conveniently for
Sulzer, but is of course generally a prerequisite for the continuous uprating of the engines.
In the performance diagram, Fig. 80, one will note rather
low fuel consumption over the normal service range. This is
due to a new feature a simple mechanism that permits
automatic control of the permissible maximum combustion pressure. The flat maximum pressure curve from
110 % down to about 85 %
engine load is a consequence
of this variable injection timing, the VIT. Unfortunately
it is not possible to maintain
the high maximum pressure
further down in the low load
range due to bearing loads
being influenced both by the
forces from combustion and
the oscillating masses, and
piston rings influenced by
the pressure rise.
The combustion optimisation is also a consequence
of an increase in maximum
pressure, and the RLB
version of 1980 has now
reached 120 bar versus the
RLA56 of 1976 with 98 bar.
48
Fig. 76 Sulzer 6RLA90 on testbed. (Wrtsil)
Fig. 77 Engine rating principle, RLB-engines and Fuel consumption RLB90. (SULZER)
Fig. 78
ABBs turbocharger, VTR454. (ABB)
Fig. 80 RLB90 engine performance with the VIT mechanism.

(Sulzer)
Fig. 79 Turbocharger development at ABB in pressure ratio and
maximum overall efficiency (BBC). (ABB)
This require great skill if reliability shall be maintained and

is among others seen in the combustion chamber design, Fig.
81 a masterpiece with bore cooled components. Especially
the piston design is outstanding, the RLB version is water
cooled, as the Sulzer tradition dictates. No wonder that Sulzer
feel content, when the in trade magazine advertisments they
present a simplified version of the combustion chamber with
the text: Why bother with exhaust valves, burn heavy fuel
in our simple furnace, Fig. 82. Indeed a daring text as it will
appear in the following.
MAN
Mr. A. Schiff, of MAN announces on a press conference in
Tokyo in May 1979 [56] new long stroke diesel engines with
a stroke to bore ratio between 2 and 2,2, the C/CL range, but
only covering the following three types: 90/190, 70/150 and
52/105. The design principles are identical to the B/BL series and the design features have only been modified in cases
where the increase in stroke made this necessary, the moderate increase in maximum pressure of 5 bar is apparently not
considered important.
Mr. Schiff compares openly the consumption figures with
B&W and express his discontent, as B&W are using different
ambient conditions in their published figures and the comparison shows: that the loop scavenging system is not inferior to the uniflow scavenging system he says and continues:
With the same specific fuel consumption, the valveless engine is simply better than the uniflow scavenged unit with its
high expenditure on the exhaust valves or valves per cylinder
and the connected maintenance etc. That remains to be seen.
Fig. 81 Combustion chamber RLA56. (Sulzer [55])
At the same occasion MAN explains that they since April have
been testing an electronically controlled injection system developed jointly with Bosch on a 3 cylinder KEZ52/105 C/CL
test engine. The system, Fig. 83 has many interesting features:
It operates with constant pressure injection based on feed
pumps driven off the camshaft as in the Doxford common
rail system.
Regardless of pump delivery it is possible to vary timing,
duration and pressure.
Due to the load independent pressure an optimal atomization is secured at part load with markedly lower consumption rates and smoke free combustion.
The E-engine is publically demonstrated in October 1979,
where it is reported, that MAN hope to get one or two plants
to sea in 1980. Unfortunately the E-type never got a break
through at sea.
Fig.82 Simplified drawing of the combustion chamber. (Sulzer)
Fig. 83 Electronically controlled fuel system for KEZ engines. (MAN

[57])
50
The three main competitors

The very tense competition between the just mentioned designers is summed up in the table below. New engines are
introduced at a rate never experienced before, and before anybody could gain sufficient service experience, even though
they would always claim the opposite.
From the table it is clearly seen that the two loop scavenged
designs are squeezed on maximum pressure compared to the
uniflow engines with the longer stroke. B&W have here a
possibility to do something relatively fast with limited effort.
An overview of the development in SFOC will be demonstrated later in greater detail.
Engines with 900 mm bore from B&W, SULZER and MAN

Date of announcement Engine type
Stroke, mm Power/cyl BHP RPM Pmax bar MEP, bar SFOC, g/BHPh *
Oct. 1977
RLA90
1900
3600
98
100
13.4
149
June 1978
K90GFC
1800
3410
114
84
11.6
141.5
June 1978
L90GFC
2180
3410
94
84
11.6
140.5
Aug. 1978
RLA 90
1900
3600
98
100
13.4
146
May 1979
KSZ90C
1900
3740
105
110
13.0
139.5
May 1979
KSZ90CL
1900
3740
95
110
14.4
142.5
June 1979
K90GFCA
1800
3945
117
95
13.0
142.5
June 1979
L90GFCA
2180
3945
97
95
12.9
139.5
Dec 1979
RLB90
1900
4000
102
118
14.3
142
Mar. 1980
KSZ90CH
1900
3945
105
115
13.7
139
Mar. 1980
KSZ90CLH
1900
3945
95
115
15.2
142
Mar. 1980
KEZ90C
1900
3740
105
110
13.0
140
Mar. 1980
KEZ90CL
1900
3740
95
110
14.4
142.6
Sep. 1980
RLB90
1900
4000
102
14.3
137
* At MCR, ISO conditions and fuel oil calorific value: 10.200 kcal/kg
Mitsubishi
Already in 1979 MHI are ready
with a series of constant pressure charged engines following
the same concept as the B&W
engines [58]. The UEC52H, Fig.
84 and UEC60H engines are
more or less a copy of their counterparts with 2-stage turbocharging and have identical power and
revolutions, however the specific
fuel consumption is reduced
~9 g/BHPh and the maximum
pressure is now 108 bar. After just 4 years Mitsubishi have
abandoned the two-stage turbocharged engines, and they can do
that, because the development of
their MET turbocharger now
has reached the required efficiency level to safeguard scavenging
providing high air purity and low
Fig. 84 6UEC52H on test bed at Kobe Diesel. (Kobe Diesel)
thermal load. The Super MET, that is used, has the features
that also later on will prove successful: Non water cooled casing, plain bearings and sturdy, wide cord turbine blades without damping wire, Fig. 85.
Instead of following their big European competitors MHI

chose to concentrate on the lower power range, and they introduce two new engines; the UEC37H & UEC45H [59].
No other company can offer direct drive low speed crosshead
engines in this power range, but MHI will soon get company by B&W and Sulzer, as the Japanese market is quite
attractive in this power range, Fig. 86. The design of the small
engines differ from the big engines and has just one central
exhaust valve, and now it is also possible to introduce a bore
cooled cylinder cover, something that would have been very
complicated with the three exhaust valves. Previously the fuel
valve was positioned in the centre of the cylinder, now two
fuel valves are applied and mounted in the periphery of the
cylinder. So, two features so far synonymous with Mitsubishi engines have disappeared. Gradually the new concept is
transferred to the other engines in the series.
The specific fuel consumption is very low compared to the
B&W uniflow and constant pressure charged engine of similar size, as shown in the table below.
Fig. 85 Mitsubishi, MET SC turbocharger, pressure ratio 3.5. (MHI)
Engine type
Bore, mm
Stroke, mm
Power, BHP/cyl
RPM
MEP, bar
Pmax, bar
SFOC, g/BHPh*
UEC45H
450
1150
1000
165
14.6
108
139
L45GFCA
450
1200
985
175
13.0
95
142.5
UEC37H
370
880
650
210
14.4
108
141
52
11. Three survivors
Fig. 86 UEC37H (MHI), L35GB (B&W), RTA38 (Sulzer).
B&W
In 1979 the majority owner of Burmeister & Wain is facing
bankruptcy, and in an attempt to overcome this threat he decides to sell the engine divisions of the company to MAN. So,
on the verge of B&Ws greatest triumph since the introduction of the turbocharged engines, at a time where the probability of re-establishing market leadership is almost certain
the competitor, MAN takes over. How does this influence
the development in Copenhagen and Augsburg? At the time
there were many speculations, but in 1980 MAN have their
plan ready: The designers in Copenhagen get the responsibility for 2-stroke engines while the people in Augsburg get
the responsibility for 4-stroke design. The B&W name disappear and is transformed to MAN-B&W, and so it will remain during the rest of this story. Further development of the
loop-scavenged engine is stopped, but previous MAN licensees sees an advantage in getting a MAN-B&W license, and
B&W licensees accept the new condition, which provides a
good starting point for the new MAN-B&W.
L-GFCA whereas the L-GBE has the same mean pressure as

the L-GFCA, 13 bar. The maximum pressure has been increased from 89 bar to 105 bar on both models. The fuel consumption is respectively 3 and 6 g/BHPh lower for the two
models than for the L-GFCA engines i.e. down to 133 and
130 g/BHP for the 90-bore engines. This is somewhat strange
as the ratio Pmax to mep for the L-GB engine is practically
the same as for the L-GFCA, however increased turbocharger
efficiency is referred to as the decisive parameter. The fuel
system is modernised in the sense that variable injection timing now also is a part of the MAN-B&W concept, as it was
on the MAN engines in the past. On the hydraulic actuated
exhaust valve the set of springs arranged around the spindle
to close the valve are now replaced with an air spring, which
remains standard on future engines. Interesting is also that
MAN-B&W now are offering an economy version, the Eversion as an alternative to the full rated engines fuel prices
will decide what is the most economical selection.
MAN-B&W are quick in improving their market position

and already in March 1981 they are presenting a new engine
programme the LGB/LGBE series of engines. The new series is basically uprated L-GFCA-engines, the mean effective
pressure of the L-GB-engine is 15 bar as against 13 bar in the
New to the programme is a small bore, the L35GB/GBE engine, the first competitor to Mitsubishis UEC37-engine
also to be produced in Japan at Mitsuis sublicense, Makita.
The comparison is shown below:
Engine type
Bore, mm
Stroke, mm
Power, BHP/cyl
RPM
MEP, bar
Pmax, bar
SFOC, g/BHPh*
UEC37H
370
880
650
210
14.4
108
141
L35GB
350
1050
680
200
14.8
120
136
Three survivors53
This engine is a real long stroke engine with a stroke to bore

ratio of 3. The design is not just a down scaling of the bigger
engines in the series it is really designed as a small engine.
The entablature and cylinder block is made in one cast iron
part and with the chain drive at the front end. The crankshaft
is fully forged in one piece and the crosshead bearings are for
the first time in MAN-B&W engines of full width, a neat
little engine as seen in Fig. 86 also showing the competitors
engines in the same scale.
The previously mentioned race on fuel consumption continues mainly between Sulzer and MAN-B&W but also with
Mitsubishi on the small and medium size bores, especially in
the Japanese market.
Sulzer
Sulzer have realistically seen that their loop scavenged concept
could not be developed further in direction of longer stroke,
and they also realise, that this is the way the demand for enEngine type
gine design go. If they want to stay in business there is only

one way to go to change the concept and design a uniflow
2-stroke engine series. To a certain extent they are familiar
with the uniflow concept, as they once have used this system
in their 2-stroke Z40 medium-speed engines (from 1963).
In December 1981 Sulzer announced a brand new engine
series, the RTA-engines. It takes the maritime world by surprise, but nearly nobody doubts that Sulzer also will master
this concept, and the decision is much respected by influential
customers.
The new RTAs are announced as the Superlongstroke engines,
and indeed they have a stroke to bore ratio of 2.9 on line
with the small MAN-B&Ws
L35GB-engine and what certainly must be expected from the
competition in the future. The engine programme consists of
the following, here shown with data related to the highest rating (R1):
Bore, mm
Stroke, mm
Power, BHP/cyl
RPM
MEP, bar
Pmax, bar SFOC, g/BHPh*
RTA84
840
2400
4030
87
15.35
125
124
RTA76
760
2200
3290
95
15.31
125
127
RTA68
680
2000
2650
105
15.34
125
128
RTA58
580
1700
1920
123
15.31
125
129
RTA48
480
1400
1320
150
13.62
125
131
RTA38
380
1100
830
190
13.61
125
133
Sulzer are offering the new engines with four different standard ratings R1, R2, R3 and R4, where the minimum rating R4
represents abt. 70 % full power at 80 % speed. By retaining
maximum and mean pressure, the R4 rating offers abt. 4 g/
BHPh reduction in fuel consumption compared to the R1
level.
metal. The Sulzer valve controlled fuel injection pumps are

also retained in the RTA series as is the variable injection timing concept.
Mr. G. Wolf, in charge of the RTA development, has decided

to split the design between large bore and small bore engines.
Mr. M. Briner to be responsible for the big engines and Dr. G.
Lustgarten to be responsible for the small bore engines with
Dr. M. Eberle in charge of general research and thermodynamics/engine performance [60].
The large bore engines are where most of the old Sulzer design features are carried along. Most markedly is the retaining
of the water cooled piston and as a consequence the split crosshead bearings. Also the tin-aluminium lined bearing shells are
kept as in the loop scavenged engines for the crosshead bearings a consequence of the water cooled piston concept. The
oil supply to the bearings is delivered from a special high pressure system as usual in Sulzer engines.
The engine structure is by and large as in the RL-series of engines. Crankshafts are semibuilt and main bearings and connecting rod bottom end bearings are shells lined with white
54
Fig. 87 RTA combustion chamber and hydraulically activated exhaust

valve. (MS [60])
A consequence hereof is that full width crosshead bearings

are introduced, however still with use of tin-aluminium lined
shell bearings. The crosshead assembly is provided with two
oil supplies through swinging link connections. One to supply
the oil for piston cooling and guide lubrication another with
high pressure to supply the crosshead bearings in accordance
with Sulzer design practice. Another new feature is that the
water transfer from cylinder liner to cylinder cover is taking
place inside the engine, Fig. 89. The camshaft is positioned
near the engine top and the multi cylinder jacket enable optimal accommodation in an integral camshaft box as on
4-stroke engines.
New is also the fuel pump. The traditional Sulzer valve controlled pump is replaced with a helix type as on the 4-stroke
engines and given a VIT-like shape derived from the variable
injection timing concept on previous engines.
Fig. 88 The first RTA to be built in Europe, the 6 cylinder RTA58,

11520 BHP (8460 kW) at 123 rpm at testbed at Sulzers
Winterthur works. (Wrtsil)
Of course the great question is related to the new Sulzer feature: Design and reliability of the exhaust valve. Sulzer have to
make the exhaust valve a success because why bother with
exhaust valves ... So the message is clear: Here is no room
for mistakes. And Sulzer make no mistakes, the valve is hydraulically operated with a rotating spindle in Nimonic 80A
material, will in the future set records, and be unsurpassed as
far as time between overhauls is concerned. The exhaust valve
assembly and the combustion space are illustrated in Fig. 87.
The camshaft is positioned at engine mid height and is driven
by a gear train from the crankshaft with just one intermediate
gear wheel. The other new components are the fuel valves, and
here Sulzer are very cautious and introduce 4 uncooled valves
in the 84-engine i.e. one more than MAN-B&W in order to
obtain as even a temperature distribution in the combustion
chamber as possible. This was maybe a bit overdone, and later
the number is reduced to three as in MAN-B&W engines of
similar size. The first RTA-engine was tested in 1983. Fig. 88
show the engine on test bed in Winterthur.
The small bore engines are of a different breed. The low costs
and compactness required for installation in smaller ships dictates different design initiatives compared to the bigger brothers, see Fig. 86. The engine structure is thus cast instead of fabricated. On the RTA38 the crankshaft is forged in one piece,
whereas the 48-engine maintains the semibuilt design because
of weight considerations.
The specific requirements of the small bore engines make it
possible to provide a fair compromise between low temperatures and low mechanical stresses in the piston crown with
application of oil cooling, the system to become standard in
later RTA engines (page 66).
Fig. 89 RTA38 combustion chamber. (Wrtsil)
Three survivors55
Mitsubishi
Following the introduction of the GB and RTA engines Mitsubishi decides to uprate the H-series, now named HA. The
power is 11 to 14 % higher than the corresponding models of
the H series, and the engines are now offered with an economy rating as well. The fuel consumption is reduced by 8 % by
adoption of the latest MET-SB turbocharger, improvements
of the fuel system and a substantial increase of the cylinder
maximum pressure.
MAN-B&W
The superlongstroke engines from Sulzer will soon get company, as MAN-B&W already in February 1982 announces
their ultra longstroke engine, the L-MC/MCE line of engines. In December the same year all details are published
[61] in The Motor Ship. The new line consists of 6 engines,
as Sulzers, this time however always with some cm oversize
in bore compared to the competitor and now with an even
higher stroke to bore ratio 3.24:1. In addition MAN-B&W
maintain the E-versions with lower rating and specific fuel
consumption. The L35GB, previously mentioned, is renamed to L35MC.
The maximum pressure is now 125 bar, as in the Sulzer RTA
engines and the fuel consumption is consequently on line
with that of the RTAs. More about fuel consumption later.
Structure wise the MC-engines are like the GB type, but there
are several interesting new components. Most remarkable is
the welded crankshaft produced by using narrow gap submerged arc techniques, Fig. 90. The crankshaft is introduced
as an alternative to the semibuilt shafts, but is announced to
become the only standard for the MC/MCE engines. The
idea behind this production method was to get the freedom
to allow overlap of the journals which would be required in
short stroke engines, but with the present long strokes the
purpose of introducing the new technology was merely a
matter of costs, or so it was thought.
Long last MAN-B&W got a full width crosshead bearing, as
in the MAN engines.
The high maximum pressure required reinforcement of the
piston crown to keep wall thicknesses reasonably thin and
temperatures low and with a similar purpose the exhaust
valve is now provided with water cooled seats.
Also from MAN a new technique in casting cylinder liners
was adopted; instead of drilling the cooling passages in the
collar of the liner for the large engines, cooling was now provided with cast-in cooling pipes. MAN used this technique
for cooling between the exhaust ports, ref. section III.
56
Fig. 90 Welded Crankshaft, using narrow gap submerged arc

techniques (MS [61])
Mitsubishi
Already a year after the HA engines MHI are ready to challenge the competition with a new engine series, the L-type,
super long stroke engines [62], however still based on the concepts of the H and HA types. Apart from the longer stroke
the L-type has increased maximum pressure , 130 bar, but
to get to a fuel consumption as low as 125 g/BHPh for the
largest UEC60L engine Mitsubishi have invented an entirely
new scavenging principle, the so called controlled swirl scavenging system, CSS. It is all about changes of the scavenging
port design as illustrated in Fig. 91, and thereby achieving a
cleaner and thus colder air charge in the cylinder at compression start, hence reduced fuel consumption. In addition the
exhaust valve temperature is decreased.
Fig. 91 CSS scavenging air porting (right) compared to previous conventional porting (left). (MHI)
MCR performance data of the UEC60LS is compared to MAN-B&W and Sulzer engines of similar size in the following table:
Engine type
Bore, mm
Stroke, mm
Power, BHP/cyl
RPM
MEP, bar
UEC60L
600
1900
2100
110
15.6
127
125
L60MC
600
1944
2080
111
13.0
125
128
RTA58
580
1700
1920
123
15.31
125
129
Because of the increased maximum pressure the
piston crown is redesigned with an inner annular
ring support, and previous radial ribs have been
done away with. Further the cylinder liner collar
is increased in height and thickness and is now
more intensely cooled. The exhaust valve housing is now just cooled around the valve spindle
and insulation has been applied around the duct
in order to increase the exhaust temperature for
more effective use of the exhaust energy.
The fuel injection system is with Bosch-type
pumps and VIT, but an interesting fuel valve
feature with automatic control of the opening
pressure of the valve based on load improves low
load fuel consumption and reduces smoke emissions, Fig. 92.
Fig. 92 Fuel valve with automatic control of opening pressure dependent on engine
load shown in water cooled as well as non cooled versions. (MHI)
Three survivors57
The L-types are developed into a LA-type and further a LS-type for the 52 and 60 bore engines to be introduced in 1985 and
1987 respectively. A comparison of the engines with 600 mm bore is shown below:
Engine type
Year
Stroke, mm
Power, BHP/cyl
RPM
MEP, bar
Pmax, bar
SFOC, g/BHPh*
UEC60L
1983
1900
2100
110
15.6
127
125
UEC60LA
1985
1900
2100
110
15.6
127
122
UEC60LS
1987
2200
2400
100
17.0
130
122
stroke to bore ratio of 3.47 against the 3.24 of MAN-B&W
and 3.16 for the Mitsubishi L and LA engines. Furthermore
a RTA84M tanker-engine based to a great extent on the
RTA84 is added to the programme [63]. There is no end to
the attraction of low revolutions and higher propeller efficiency.
The reduced consumption is claimed to be a consequence of

a modification of the scavenging timing and hereby extending the effective stroke, which is possible due to the higher
efficiency of the MET-SC turbocharger, Fig. 85 now applied
on these engines.
In addition to the LS programme of 1987 a LSII engine is
developed, as will be mentioned later.
The new engine design follow the previous except for one
important feature: The pistons are now oil cooled not exactly the prophecy of Sulzer in the late sixties: Sooner or
later everybody will apply water cooled pistons, but maybe
the prophecy can become right sometime. The oil supply to
the crosshead is arranged as in the two small engine types. As
a conse-quence of oil cooled pistons the cross-heads are now
designed with full-width lower bearings.
Sulzer
To compete with MAN-B&Ws L-MC engines Sulzer introduces in 1984 the so called 2-engines, RTA52, RTA62
and later (1986) RTA72 all 2 cm larger in bore than the
corresponding MAN-B&W engines and furthermore with a
155
K84EF, K98FF/GF
RD90, RND105
150
L90GF
90RND-M
g/BHPh
145
140
L90GFC
L90GFCA
135
RLB90
L90GB
130
RTA84, RTA84C, RTA96C

K90MC, K98MC
125
120
1965
RLA90
B&W
1970
1975
1980
1985 Year 1990
Fig. 93 Development of the specific fuel consumption shown for large bore B&W and Sulzer engines. (Author)
58
SULZER
1995
2000
Fuel consumption development and comparison

The very intensive competition on fuel consumption between
the two left main competitors is illustrated in Fig. 93, where
the red line shows the dramatic development over 10 years.
However, the heading official fuel consumption refers to the
fact that during this period of time, the major engine designers are introducing rather unfair methods, when they present
the consumption figures and that goes for Mitsubishi too.
Before 1980 the tolerance on consumption was 3%. Anders
stergaard, Head of research and development at B&W publish [64] in Schiff & Hafen the outcome of measurements
from delivery tests with many L-GF engines, Fig. 94, and it is
noteworthy that the tolerance margin of 3 % is ample to cover
the spread in the measured values. Gradually the consumption
is given with + 3 % margin and in the late 90s in connection with IMO NOx certification it becomes + 5 %. What
happens? There is no more any tolerance and with modern
measurement techniques at the time one dont need 3 %. As
however, very effective calculation tool have been developed
for estimation of the fuel consumption at the design stage,
the tolerance is eaten up, and actual measurements on test
bed will show values very close to 3 to 5 g/BHP higher than
shown on the figure. A closer analysis of the engine performance data also clearly reveals that the ratio between maximum and mean pressure, being the most decisive parameter
in relation to fuel consumption, not has been increased corresponding to the dramatic reduction in fuel consumption.
The IMO NOx regulations [65] are more or less forcing the
engine designers, by engine internal measures, to reduce
the NOx emission as the cheapest and fastest solution. This
means that the beginning of the combustion in the cylinder
is retarded (the maximum pressure reduced) so, that the combustion is taking place at lower temperatures, but unfortu-
Fig. 94 Actual fuel consumption measurements on B&Ws testbed

compared to guaranteed figures (S&H [64])
nately also that the fuel consumption is deteriorated. Instead

of increasing the quoted SFOC figures the designers just add
extra 2 % to the margin. Not too smart, and certainly not
smart either that IMO regulations concentrate so one-sided
on NOx emissions and ignore, that the CO2 emissions are
increased as a consequence of the regulation.
Comparing all 3 survivors fuel consumption on engines of
similar size as done in Fig. 95, clearly shows that MHI typically are ~3g/BHP lower than the others, so the mentioned
special features of the Mitsubishi engines as the CSS ports
and others must certainly be very important. The competitors
have tried to imitate the CSS ports without success.
The market share development reflects the importance
of being first in offering low
consumption and low engine Revolutions. This can be
seen in Fig. 96, where MANB&W after introduction of
the long stroke, constant
pressure turbocharged engines recapture the position
as market leader lost to Sulzer
in 1964.
127
126
g/BHPh
125
124
123
122
B&W
SULZER
MITSUBISHI
121
120
119
400
500
600
700
800
900
Engine bore, mm
Fig. 95 Typical fuel consumption figures from MAN-B&W, Sulzer and Mitsubishi for engines of similar
bore sizes. (Author)
Fig. 93 also illustrates the

rapidness of engines being
introduced. Compared to
the past, at the start of turbocharging, there were 6 to
7 years between new engine
types. Here in the 70-80s
there are only 2 to 3 years.
There are many good explanations, how this was
Three survivors59
ossible, but one thing for sure is missing; the implementap

tion of service experience of the previous design to new designs. It is simply not possible. Engine users are often a bit
annoyed and the industry also see through their co-operation
in CIMAC, that it would be appropriate to include the users
more actively, so in 1993 the CIMAC Engine Users working group is established. This working group starts to collect
experience data from their fleet and provides documentation
that will be use in the following.
60
Market share, %
50
B&W
40
SULZER
30
MAN
Mitsubishi
20
10
0
1965
1970
1975 Year 1980
1985
1990
Fig. 96 Market share development. (Author)
Engine failures
Due to the very size of the 2-stroke engines it is not possible to run endurance tests and sort out failures at the test
bed. The usual practice is therefore to carry out tests on new
components on existing engines in service. However, even the
most well designed new engine will have some failures that
need to be solved, and sometimes the time factor is important, as off service of the vessels could be a costly affair. Many
ship-owners have as a consequence of this acted as the test
field for the engine designers.
The engine users experience concerning the period from
1975 to 1999 has been reported at several CIMAC congresses
[66,67,68], and the following general observations are:
About 50 % of all engine failures are related to the thermal
loaded components, hence the focus in this report on this
specific item.
Poor cylinder condition including cylinder liner cracks
represents nearly half of the failures on the thermal loaded
components, meaning that overhauls of cylinders were
much more frequent than foreseen by the designers, Fig.
97.
Turbocharger failures are frequent Fig. 98 and complete
break downs sometimes disastrous, Fig. 99.
Main bearing failures are significant and found in most
engine types.
60
More specifically the engine users analysed the distribution of

failure cases related to all the engines out of guarantee of the
three designers. The failures are categorised in main sections
of the engine as shown in the following table, Fig. 100. It is
clearly seen that hot components as mentioned is a hot issue, and dominating for the two major makers. The reason
why so few failure cases on MHI-engines are found in Hot
Components section could be, apart from a well done design, because the engines generally are designed for relatively
slow ships with low revolutions and consequently have a high
stroke to bore ratio and an amply sized combustion chamber.
The economical consequences of failures are included in engine maintenance cost. For the years 19802000 the CIMAC
WG found among others that the specific operation costs (excluding fuel, lubrication oil, crew and financial costs) of 14
different engine types and makes was larger for the small bore
engines than for the large bore engines, Fig. 101, and just as
surprising that the cost/kWh were fairly low. The small figures of the specific costs could be misleading but the actual
yearly cost for a big large bore engine could be in the region
of 200.000 US$ per year, which is more tangible.
Fig. 97 Cracked cylinder liner, KSZ90BL. (Author)
Fig. 99 Blower wheel of complete wrecked turbocharger (Author)
Fig. 101 Engine operational costs versus engine bore. (CIMAC [68])
Fig. 98 Fatigue crack in turbine blade (Author)
Fig. 100 Distribution of failure cases on main engine sections. (CIMAC [67])
Three survivors61
12. Engines for the big Container Ships

MAN-B&W
It must be noted that contrary to previous MAN-B&W practice all new engines will maintain the name MC or MCE,
meaning that is becomes difficult to distinguish between the
different designs. Later in time MAN-B&W add a mark number, e.g. MC6 which make identification easy. Where needed a mark number will be used in this text.
In 1986 MAN-B&W are ready to extend their portfolio of engines and update the existing L-types. The programme include
two new classes, the S-type (see later) and a K-class intended
for rather big and fast container vessels and the entire programme now totals 14 different engines with a cylinder power
from 395 to 5360 BHP to cover the entire propulsion requirements. The multiplicity of this 2-stroke crosshead engine programme is impressive and of course decisive for maintaining a
large market share.
are highest, while the running conditions for the piston rings
are maintained as before.
Years later improved production technology made room for
the use of cast-in cooling pipes.
The thermal load of the piston crowns also became a problem,
as burn down of the material and large thermal cracks reduced
the lifetime considerably. At the same time the exhaust valve
spindles had an unsatisfactory time between overhaul and
lifetime. Some users made a change to Nimonic material, as
the Sulzer standard solution, and finally in 1994 MAN-B&W
also introduced Nimonic material as standard for valve spindles of large bore engines.
In 1989 a K90MC-C is added to the programme and the
K80MC is converted to K80MC-C (the additional C to
indicate the suitability for Containerships). The K90MC-C
has a shorter stroke and higher revolutions than the MC counterpart and is consequently cheaper to produce.
The two K-engines, K90MC3 and K80MC3 are having a

relatively short stroke (stroke to bore ratio 2.8:1) and consequently the combustion space is relatively flat making it more
difficult to avoid high thermal load on the components, now
the mean pressure is increased to 16.2 bar. Most of the design When B&W in 1978 decided to stop their turbocharger profeatures introduced in the L-MC series are maintained in the duction they made an agreement with BBC to use their chargK-class also the cast-in cooling pipes in the cylinder liners ers. So BBC became more or less absolute on B&W engines.
mentioned earlier. This turned out to be a disaster as most of The BBC chargers had the efficiency that was required, but
the liners cracked in the cast-in tubes because of loss of fatigue were facing a draw back because of expensive maintenance;
strength of the tubes, as these were decarbonised during the water-cooled cases were corroded and the lifetime of the
casting process and the steel transformed to ferrite having low roller bearings too short. So with MAN as the owner it befatigue strength. A design change was required and that was came possible to have a competitive product specified. MANs
a return to bore cooling, but
this time MAN-B&W used
the GMT principle applied
in the CC600 with bores
that ends blindly, as shown
in Fig.69, a very effective and
good solution. However, a solution had to be found for the
vessels in service, and here the
licensee Mitsui provided an
exciting dual cast liner that
could be mounted without
any major modifications of
the existing engine [69]. The
dual cast liner is produced
by centrifugal casting where
the outer layer that first is
poured into the machine is
plain carbon steel, while the
inner layer, the running surface is the standard cast iron
for cylinder liners. In this
way high tensile strength is
achieved at the cooling surface, where the tensile stresses Fig. 102 MAN, NA-TO turbocharger (MAN-B&W)
62
turbochargers were uncooled

and had plain bearings that
functioned well, Fig. 102.
However, it is difficult to
make things right, and on
the large K90MC3 engines
the MAN NA-turbochargers
with long, slim turbine blades
suffered in several cases from
breakage because of resonance problems between the
turbine nozzles and blades
with accompanying fatigue
cracks, Fig. 98. MAN-B&W
were consequently pressed by
users to also make the Mitsubishi MET turbocharger
standard on their engines, as
this charger had short wide
chord blades with a very
good service record. Now
there was full competition.
Solutions to the problems
were found, and in 1991
MAN-B&W
are
ready
with the first uprating, the
K90MC5, where the mean
pressure is increased from
16.2 to 17 bar and the revolutions from 90 to 94 rpm.
The demand for large containerships were however
growing as the transportation
market growth was about 10
% per year and that meant
also an increased demand
for higher engine power for
larger vessels offering lower
transportation costs. Post
panmax container ships were
being built at the Odense
Shipyard, and that forces Fig. 103 Cross section of K90MC (6), bore 900 mm, stroke 2550 mm, yielding 6220 BHP/cyl (4570 kW/
cyl) at 94 rpm and Mep = 18 bar. (MAN-B&W)
MAN-B&W already in 1993
to offer a new K90MC6,
now with 18 bar mean effective pressure. A cross section of work solutions, among others introduction of a so called piston cleaning ring integrated in the top of the liner where it
K90MC6 is shown in Fig. 103.
scrapes off deposits built up on the piston topland, the right
As mentioned before the time between introductions of new solution was found i.e. to eliminate the soot formation and
designs is too short to accumulate service experience, and that secure clean combustion. An old idea of Dr. Sunn Pedersen
is also the case of the mark 6 engine. The combustion in the (MAN-B&Ws technical director) to eliminate the oil sack in
uprated engine was not satisfactory, actually it was not too the fuel valve was introduced. This so called slide valve, Fig.
clean either in the old mark 3 engine, and coke formation was 104 had a decisive effect on the cleanliness in the engine cylbuilding up on the piston top land and grew to a size, where inders as seen in Fig. 105, where the condition of the piston
it wiped off the lubrication oil on the cylinder running sur- rings is seen with and without slide valve. In addition the
face. The consequence was heavy scuffing of piston ring and fuel consumption was reduced as the oil left in the atomizer
cylinder liner surfaces. After introduction of several patch- tip now was burned effectively.
Engines for the big Container Ships63
Fig. 104 Standard and Slide Valve (MAN-B&W)
For some time there had been focus on the lubrication oil
consumption as the cost of lube oil were of the same of magnitude as the maintenance cost of the entire engine. The
cleaner ring pack was a good starting point for introduction
of more a modern lubrication system that made lube oil consumption reduction possible in a controlled way.
Fig. 105 Engine cylinder condition showing a vast improvement

when using slide fuel valves instead of standard fuel valves
(Author)
It was also necessary to find a better piston design, as the temperatures in this highly rated, short stroke engine were too
high with risk of burning and crack formation of the piston
crown. By use of advanced computer simulation tools now
available a completely new design of the piston crown with a
small mountain in the centre instead of the traditional bowl
reduced heat transmission so much, that temperatures could
be reduced about 100 C to 409 C, Fig. 106 and Fig. 107
[70]. A high margin for future use in uprated engines were
hereby established, and a further development potential in
applying an entirely bore cooled design were at hand.
In Fig. 106 it is also demonstrated how technology is transferred from MAN to MAN-B&W in the adaptation of the
finger bore cooling system for the cylinder cover, introduced
for the KSZB/BL-engines, Fig. 68.
64
Fig. 106 Combustion chamber temperatures K90MC (6), Mep = 18

bar. (MAN-B&W)
In 1994 MAN-B&W introduced again a 98-bore engine,

the K98MC-C intended for use in the large post panmax
container vessels, that were built in a drastic increasing number all over the world. The specific power of the engines was
7760 BHP/cyl at 104 rpm and with a mep of 18.2 bar as the
K90MC6. The engine did not become popular at the yards
because of the high number of revolutions, and in 1998 a
K98MC with increased stroke and lower revolutions was designed and immediately sold. The considerable differences
between engine parameters are illustrated in the table below.
Fig. 107 Comparison of the effect on piston and exhaust valve
temperatures when using a standard piston crown versus a
design having a small top (Oros) at the centre of the crown.
100% load, Mep = 18 bar. (CIMAC [70])
Engine type
Charging principle
Year
Stroke mm
Power, BHP/cyl
rpm
MEP bar
Pmax, bar
SFOC, g/BHPh*
K98FF
impulse
1967
2000
3800
103
10.8
80
157
K98MC6
constant p.
1998
2660
7780
94
18.2
130
126
Sulzer
The development of the prestigious, high output engines for
the large, fast containerships is the most challenging engineering task. Sulzer, although losing market share generally,
were matching MAN-B&Ws share on the large engines for
containerships. Particularly the RTA84 and to some degree

RTA84M were ordered for many containerships. However, in
late 80s there were increasing indications towards the need
for higher speed and power, so, on the basis of the RTA84 the
new RTA84C was ready in
1988 with a cylinder power
of 5200 BHP at 100 rpm and
a mean effective pressure of
17.2 bar [71].
The RTA84C is shown in a
cut-away view in Fig. 108.
It has basically the same design characteristics as the
RTA84M and the 2-series,
which means that engine
structure, combustion chamber components, running
gear with white metal crosshead bearings are having
known track records. The
most recent improvement
of these engines are introduction of a so called Core
Package to improve piston
running conditions and this
feature is also a part of the
RTA84C concept.
Fig. 108 Cut away view of the RTA84C, bore 840 mm, stroke 2400 mm, 5200 BHP/cyl at 100 rpm and Mep
17.2 bar. (Wrtsil)
The engine cylinder condition is a multi disciplinary

engineering task and as such
involving the entire design
team to find a reliable solu-
tion. Sulzers Core Package [72, 73] consisted of three major

improvements:
1. Better cylinder liner material with higher hard phase content
2. Increased temperatures of the cylinder liner running surface
3. Introduction of multilevel lubrication (seen on Fig. 108)
This Core Package was based on a lot of research and experiments, and is exemplary described in the references. However,
one thing is theory another is practice, and the cylinder condition was not in all cases acceptable and reliable, as reported
later.
A disturbing element of company politics at this time must
be reported: Sulzer Brothers in Winterthur had little interest
in maintaining the heavy industry in their portfolio of enterprises and in 1989 they discuss with MAN in Augsburg to sell
out the diesel division. Apparently the two parties reach a conEngine type
clusion, but the acquisition was vetoed by the West Germany

Economics Ministry in January 1990 on the grounds that it
would create a monopoly. This action was indeed a Sulzer top
management blunder, because who would buy engines designed by somebody whos destiny was in the hands of the biggest competitor, so Sulzers market share was hurt. Now as it
was common news, Sulzer had to find other solutions fast. In
July 1990 it was reported that the majority of shares in Sulzer
diesel was sold to Fincantieri and a consortium of West Germanys Bremer Vulcan and Deutsche Maschinen und Schiffbau in East Germany and a small management stake. On the
15 November 1990 New Sulzer Diesel was born, still producing and designing engines bearing the Sulzer name.
In 1993 New Sulzer Diesel uprates the RTA84C on both revolutions and mean effective pressure and in December 1994
a super large bore engine the RTA96C is added to the programme of container ship engines [74], ref. the table below:
Year
Bore mm
Stroke, mm
Power, BHP/cyl
RPM
MEP, bar
Pmax, bar
SFOC, g/BHPh*
RTA84C
1988
840
2400
5200
100
17.2
135
126
RTA84C
1994
840
2400
5510
102
17.9
140
126
RTA96C
1994
960
2500
7470
100
18.2
142
126
Fig. 109 Cross sections of the RTA84C (left) and the RTA96C (right) to the same scale. (Wrtsil [74])
66
A cross section of the two engines to the same scale is shown

in Fig. 109.
scavenging air pressure, which allows to lower the piston position somewhat and further to improve the cooling effect
by researching the opportunities of the combined effect of
shaker and oil jet measures, Fig. 112 [75]. The cooling system
inside the piston crown is shown in Fig. 113.
Fig. 110 The first 12-cylinder Sulzer RTA96C engine on test at Diesel
United in Aioi, Japan. (Wrtsil)
The first RTA96C engine completed tests in May 1997, Fig.

110. The general performance and the temperatures of the
combustion chamber components, Fig. 111 were measured
and found in line with precalculations.
Fig. 112 Operating principle of jet-shaker piston cooling system

(drawing based on RTA-U engine). (Wrtsil [74])
Fig. 113 Cooling inserts for pistons, RTA96C. (Wrtsil [74])

Fig. 111 Combustion chamber surface temperatures, 11RTA96C at
Mep = 18.2 bar. (Wrtsil [74])
The wall temperatures were expected to be the main hurdle

to pass, as this engine has a very short stroke to bore ratio
of about 2.6 and consequently a rather shallow combustion
chamber. The difficulty in overcoming high thermal load of
the components were overcome by selecting a relatively high
The service experience with these container ship engines was

generally acceptable. The Core Package, was now replaced
with a more comprehensive Tribo Pack [76] containing
among others an Anti-Polishing Ring as in the MAN-B&W
engines and ceramic chrome plating of the upper piston ring,
as has been used for quite some years in medium speed engines. After some years in service the cylinder condition was
still not satisfactory there were too many scuffing incidents.
Much was done to find the root cause of these incidents, and
some of the conclusions were interesting: The liner temperature designed to prevent acid corrosion of the liner surface
was so high that there not were margins left to allow for deterioration of engine conditions (aging), and the hard phase
content of the liner material already introduced in the Core
Package had to be reduced.
Mitsubishi
Another dominant contribution to higher reliability proved

to be the introduction of so-called zig-zag oil grooves. These
groves replaced the former cylinder standard with grooves
only slightly inclined (Fig. 108).
The engine is developed based on the UECLSII concept with

the 85 bore as the largest, as will be reported later.
Strangely enough the prescribed remedies of the Core Package or Tribo Pack worked well on most other engine types,
even the big tanker engines.
Engine type
Mitsubishi had not been very active to provide solutions for

the large container ships, and it took until 1993 before they
had a design, the UEC85LSC, to challenge the virtual duopoly in this sector MAN-B&W and New Sulzer Diesel. Size
wise it matches the RTA84C as seen in the table:
Mitsubishi were also keen to introduce measures to reduce

the expensive lubrication oil. The Japanese Shipping companies had traditionally used a rather high lube oil dosage in
their engines, but with MHIs introduction of the Danish SIP
lubrication system in the UEC85LSC engine it was possible
to reduce the consumption by 40 % and even maintain a low
wear rate of piston rings and cylinder liners [77]. In the late
90s lube oil consumption had become an issue, as the price
was 10 times that of fuel oil and MHI were very active to
promote their new system as retrofit on existing engines. Also
MAN-B&W and NSD were keen to introduce new systems,
the Alpha system and the CLU-3 load dependent accumulator system respectively.
Bore, mm
Stroke, mm
Power, BHP/cyl
RPM
MEP, bar
UEC85LSC
850
2360
5300
102
17.1
135
121
RTA84C
840
2400
5510
102
17.9
140
126
68
13. Engines for Very Large Crude Carriers, VLCCs

The term Superlongstroke engine was used, when Sulzer introduced their RTA series of engines. Since then the stroke to bore
ratio has been stepwise increased to what now may be termed Mega long stroke engines. These mega long stroke engines are
intended for big tankers having relatively low speed and slow running propellers.
MAN-B&W
In 1986 the S-series is introduced and consists of 5 bores, the
S80MC3 being the largest and most powerful. It develops
4560 BHP/cyl at 77 rpm and has a mean effective pressure
of 17 bar. The stroke is no less than 3056 mm which gives a
stroke to bore ratio of 3.82 so far a new record. The engine
is gradually uprated and in year 2000 it develops 5220 BHP/
cyl at a mep of 19 bar.
The design concept follows the K90MC3 already mentioned
and the S80MC3 also shares some of the same reliability difficulties among others cracks in cylinder liners due to the
cast-in cooling pipes. Additionally the cylinder condition was
extremely poor with high liner and piston ring wear rates, so
the time between piston overhauls were short, around one
year. The means to remove the difficulties were also the same
as for the big containership engines.
Engine type
Bore, mm
Stroke,mm Stroke/Bore
In 1998 a S80MC-C, a S90MC-C, and a L90MC-C is added

to the programme (the C here stands for Compact, but
more about this in the following section) being more cost
effective than the S80MC. A comparison between the engine
types offered in year 2000 and the old K98FF primarily used
in supertankers in the 70s is shown in the table below.
3200 mm stroke is now the record and is expected to stand
for some years still quite formidable, just as the stroke to
bore ratio of 4, nearly the double of the K98FF engine, and
the specific weight is nearly reduced by 50 %.
The extremely long cylinder liner and the relatively high
mean piston speed, Cm compared to the old K98FF engine
is a challenge in terms of lubrication, but with the new remedies at hand, firstly introduced on the containership engines,
the cylinder condition has been very good.
Mep,bar
Power,BHP
rpm
Cm,m/s
BHP/m
Kg/BHP
6S90MC-C
900
3188
3,54
19
39900
76
8.1
3301
27.7
6L90MC-C
900
2916
3,24
19
39840
83
8.1
3213
27.0
6S80MC-C
800
3200
4,00
19
31680
76
8.1
2907
25.4
6S80MC
800
3056
3,82
19
31320
79
8.0
2874
27.6
6K98FF
980
2000
2,04
10.8
22800
103
6.9
1351
45.2
Mitsubishi
Mitsubishi are ready with their super-tanker engine in 1988,
the UEC75LSII.
The cut away view of the engine is shown in Fig. 114, from
where it is seen that the engine construction now features
monoblock, welded bedplate and entablature, but the other
major engine parts are following previous MHI concepts
adapted as needed to the increased engine size. Here is also
seen that the cylinder lubrication takes place at two levels in
the liner as in the Sulzer RTA engines. The combustion chamber is well cooled and component temperatures are demonstrated in Fig 115. New, but not seen, is the system that uses
part of the energy of the hydraulic oil to rotate the exhaust
valve spindle [78].
The most interesting feature is the option to install MET66SRVG turbochargers with a radial-flow turbine and VG (Variable Geometry) nozzles instead of the standard MET SCcharger with an axial flow turbine [79]. The first 8-cylinder
engine was actually equipped with the VG unit, and when
inspected after two years at the ships guarantee docking the

actuating device and the nozzles were free and operational
much to the surprise of users often seeing turbines fouled as
a consequence of the heavy fuel burning. Basically it was the
intention that VG should serve to
reduce the fuel consumption at part load,
increase the engine overload ability up to 120 %,
maintain the surge margin in rough sea.
The idea was not adopted by customers, as it was considered
too expensive, but there is no doubt that VG is an interesting
feature, when total economy of the engine plant, making use
of the waste heat, is a must.
The UEC75LSII had one sore point the main bearings. For
some ships in service white metal separation was observed on
the side edge of the lower bearing shell. This phenomenon
was analysed and the results were reported at the CIMAC
congress in Interlagen in 1995 [80] by Mr. S. Yoshihara. It
appeared that due to vertical forces in some of the bearings
Engines for Very Large Crude Carriers, VLCCs 69
Fig. 114 Cut away view of the tanker engine 8UEC75LSII, bore 750 mm, stroke 2800 mm, 4000 BHP/cyl at 84 rpm and Mep 17 bar. (MHI)
it was necessary to reinforce the bearing caps in question to

solve the problem. The research required and the effectiveness
of MHI to find a fast solution was appreciated by the Users
of the engine type.
In 1991 a larger brother, the UEC85LSII, is launched with
very much the same characteristics as the smaller engine.
70
This engine has more or less the same performance data as

the Sulzer RTA84T introduced at the same time. The funny
thing is that while MHI have given up selling the low load,
VG feature of the 75LSII, Sulzer have included features to the
same effect in their new engine, as will be told later. Another
funny thing is that MHI with their Sulzer license actually
have the possibility of offering both engines to their clients.
The RTA84T is introduced with some very interesting and

brand new features not seen before in 2-stroke engine design
[81]. Sulzer anticipate that VLCCs will slow steam for quite
some time, and that it consequently would be an advantage
to provide the engine with features that makes this possible
without sacrificing specific fuel consumption nor putting cylinder running conditions at risk.
To that end the well known VIT (variable injection timing) is combined with VEC (variable exhaust closing). The
VEC works between 100 % and 80% of engine maximum
load, and its function is to keep the compression pressure
and therefore the maximum pressure constant by early closing of the exhaust valve at part load. The VIT takes over from
80 % to 65 % engine maximum power, meaning that the
maximum pressure of 140 bar is constant down to 65 % load
indeed requiring some considerations regarding the bearing
design; however the bearing pressure of the RTA84T engine
is even slightly lower than on previous engines.
Fig. 115 Combustion Chamber temperatures for UEC75LSII at 4000

BHP/cyl and Mep = 17 bar. (MHI [78])
Sulzer
The RTA84M introduced in 1984 has been the preferred
Sulzer engine for super tankers.
In the beginning of the 90s there is a strong competition from
the MAN-B&Ws S80MC and Mitsubishis UEC75LSII and
the new UEC85LSII. Especially the Mitsubishi engines are
both more compact and not so heavy i.e. cheaper to produce
than the M-type. So, a new engine is required to be competitive in this important market segment, and the RTA84T is
introduced in 1991. The performance data of the RTA84T
and M engine is shown in the table below and compared with
1991 versions of MAN-B&W and Mitsubishi engines.
Even though the stroke is higher than on the M-type the total
engine height is lower as the connection rod ratio (half the
stroke/the length of the connecting rod) has been reduced
from 1:2.4 to 1:2.1, as also has been the case in some previous
RTA engines.
Engine type
The other new feature was a load dependent cylinder liner

cooling. This was introduced, as it is important to avoid liner
temperatures below the dew point in the cylinder in order
to avoid cold corrosion over the whole load range. The idea
is logical and has furthermore the additional advantage that
the increased liner temperatures at low load will lead to increased mechanical efficiency, as the viscous friction losses are
reduced. It is strange that this feature disappears from later
new designs.
Despite the new cooling system the parameters were not set
quite right from the outset, and the consequence was heavy
corrosive liner wear, and soon after the engine came into service it became necessary to introduce an improvement package [82], where water ingress from condensation after the air
coolers were reduced and/or drained, and the liner temperature level was increased.
In 1996 the engine was upgraded or modified on some points
to improve the competitiveness in design for manufacturing
and named RT84T-B; however the performance parameters
remained unchanged. The cylinder condition of the B-type
was very good, as it was on the next following version the
D-type announced in 1998, where the previous mentioned
Tribo Pack cylinder running solution was implemented.
However, the cylinder oil consumption was pretty high.
Bore mm
Stroke, mm
Stroke/Bore
Mep, bar
Power, BHP
rpm
BHP/m
Kg/BHP
6RTA84M
840
2900
3.45
17,2
30240
81
2535
32.2
6RTA84T
840
3150
3.75
18
31680
74
2830
27.1
6S80MC
800
3056
3.82
18
29700
79
2521
29.6
6UEC85LSII
850
3150
3.70
18
31500
76
2876
27.2
Engines for Very Large Crude Carriers, VLCCs 71
14. Medium and small bore engines up to year 2000

The medium and small bore engines has to cover a broad
variation of ships applications above 2000 tdw, such as Bulk
carriers, Product tankers, Container feeders, Gas carriers, Car
carriers, and many specialised ships. This area constitutes
the biggest market for diesel power, but to be preferred as
a 2-stroke low speed designer you need a good coverage in
power and engine revolutions to cope with the competition
from the medium speed engines, where the required propeller
revolutions always can be matched.
Multiplicity is the key word to success in this power market,
and comparing engine programmes of the three designers
one really sees who is the market leader:
MAN-B&W have 15, Wrtsil (Sulzer) have 6 and MHI
have 6 engines with different bore or stroke to cover the market (bores above 800 mm not included).
Fig. 116 Broken valve spindle forged through piston crown. (Author)
Most of these engines are design-wise very close to the larger

bores and from that aspect not so interesting, but they are
generally characterised with a very high degree of reliability.
Large engines big problems, small engines small problems.
In comparison with medium speed engines the medium size
slow speed engines are very robust and even big casualties
will rather seldom result in total disasters. An example may
explain what this could mean: A fatigue crack in a exhaust
valve spindle caused that the valve to fall into the cylinder,
but apart from some damages to the piston top and cylinder
cover bottom there were no other damages on a medium
speed engine the result would most probably have been, that
the connecting rod was passing through the side of the crank
casing. Two such failure cases are shown in Fig. 116 & 117.
Fig. 117 Broken valve spindle curled-up in the compression space.
(Author)
MAN-B&W
MAN-B&W were as mentioned having a vast engine programme. One small engine deserves specific mention, the
S26MC included in the programme in 1986. It has a cylinder
diameter of 260 mm and a stroke of 980 mm yielding 495
BHP/cyl at 250 rpm and at a mep of 16.8 bar. The maximum
pressure was higher than in any other engine at the time
160 bar. A cross section of the small engine is shown in Fig.
118. The small engine was demanding component solutions
with different design than the larger engines, as the camshaft
integrated in the engine structure as in medium speed engines, and the running gear with round guideshoes Fig. 119,
as in some of the Werkspoor-Lugt engines. Crosshead bearing
shells are lined with tin aluminium.
The most important development at MAN-B&W was however the new MC-C series of engines (C for compact) with
bores of 46, 50, 60 and 70 cm and of very different design
compared to the predecessors of the same bore [83]. Concentrating on the 50-bore a com-parison between the MC and
the MC-C is shown in Fig. 120.
72
As the main development target has been to provide lighter

and shorter engines the designers have focussed on removal of
material, where it was unnecessary and adding material where
required by strength and stiffness considerations.
The new engine features twin staybolts secured in threads in
the top of the bedplate compared to the traditional through
going single staybolt. This is the most radical design change in
B&W structural design practise. The through going staybolt
was intended to provide compression in the welding seams
of the bedplate, now welding quality is considered so perfect,
that a higher degree of tensile stress can be accepted actually
this was already done in the Gtaverken VGS-engines more
than 50 years ago. Details of the frame box and the bedplate
are shown in Fig. 121. There are several advantages of this
new design compared to the traditional design: 1) Deformation of the main bearing bore is much less than previously,
2) The crosshead guides are less deformed, 3) The welding
process has been much simplified, and the casting of the main
bearing saddle has been simplified too. The MC-C engine has
now a cylinder frame with integrated camshaft housing as on
the small bore engines.
Fig. 118 Cross section of S26MC, bore 260 mm, stroke 980mm,
yielding 495 BHP/cyl (365 kW/cyl) at250 rpm and Mep =
16.8 bar. Maximum pressure 160 bar. (MAN-B&W)
Fig. 119 S26MC running gear with a relatively short connecting rod.
(MAN-B&W)
Fig. 120 Comparison of the compact S50MC-C and the standard

S50MC engine. The MC-C with 10 % more power per
cylinder than the MC but with about 16 tons less dry mass
irrespective of cylinder number. (CIMAC [83])
Fig. 121 Comparison on engine frame design between S50MC-C and

S50MC engines. (CIMAC [83])
Medium and small bore engines up to year 200073
The weight of the 6S50MC-C is 210 tons; a reduction of 28

tons compared to MC and has now reached a specific weight
of only 16, 3 kg/BHP.
The length of a 6S50MC-C-engine has been reduced by
more than 700mm compared to the MC counterpart due
to a shorter cylinder distance and a shorter thrust bearing/

chain drive section. This means also that the width of bearing is reduced and the diameter increased to maintain reliability, however with an additional reinforcement of bearing
shell design using Aluminium-tin to increase material fatigue
strength another break in B&W bearing design practise.
Sulzer
The RTA-2 type engines launched in 1984 was in 1992 upgraded in view of adaptation for simplified production and
assembly, and the power output was increased by 3 %, now
designated RTA-2U. Further in 1997 another 3% power
increase was added together with a modernisation package,
based on the latest engine development of the T line of engines launched the year before. The designation was now
RTA-2U-B [82].
The mentioned T line of engines has bores of 48, Fig 122, 58
and 68 cm, and are collectively termed RTA-8T [84]. These
engines are introduced to cover the propulsion demand for
the greater standardisation of ship designs for bulkers and
tankers, the so called Handysize, Handymax and Panamax
types. The 8T-line is characterised with the very high stroke to
bore ratio ranging from 4 to 4.17 (as the MAN-B&W MC-C
engines) to accommodate low revolutions required by the
large propellers of the mentioned vessels with sailing speed
typically between 14 and 16 knots. Furthermore the engine
length is extremely short.
The introduction rating is on level with the other RTA engines, but in 1997 an updated version; RTA-8TB was introduced with a mean effective pressure of 19 bar and a mean
piston speed of 8.5 m/s. Even so, the thermal load of the combustion chamber is very moderate compared to the large bore
engines, as shown in Fig. 123.
In 1997 New Sulzer Diesel is taken over by Finlands big engine builder Wrtsil and the company is now Wrtsil NSD
Corporation. The 2-stroke engine development is maintained
in Winterthur, whereas New Sulzer Diesels 4-stroke medium
speed engines are phased out and replaced with a Wrtsil developed 4-stroke programme.
The first brand new engine to be developed under the new
ownership is the RTA60C, which is ready mid 1999. This engine is oriented towards faster vessels such as container ships,
container feeders, car carriers and so on. The trend towards
designing the engines to be more economical to manufacture
is continued in this engine, and the specific engine weight
is now down to 17.1 kg/BHP, fairly low for that size of an
engine.
The engine is designed in two versions, 1) the standard mechanically controlled version and 2) a RT-flex version which
has a fully electronically control-led common-rail systems for
fuel injection and exhaust valve ope-ration. Actually the first
engines to be sold are of the RT-flex version; how-ever this is
outside the scope of this story.
Fig. 122 The RTA48T on testbed in Aioi. (WNSD [84])
Fig. 123 Combustion chamber RTA58T, surface temperatures at full

load i.e. 2720 BHP/cyl (2000kW/cyl). (Wrtsil [84])
74
Mitsubishi
Following the development of the big engines for supertankers the UEC-LSII series is completed during 1993 to 1997
with 5 smaller engines and these are rapidly followed by 4
LSE-engines as the next generation is named. The LSII engines are very much like the bigger brothers apart from the
engine structure, which for the smallest engine still is made of
cast iron assembled monoblocks. The LSE-engines are getting
Engine Type
a power boost, and the engine performance parameters are

now very much on line with competition as seen in the table
below. The only novelty is application of the newly developed
A-ECL cylinder lubrication system being modified from the
SIP system mentio-ned previously and the application of Nimonic material for the exhaust valve spindles, just like the
competition.
Bore mm
Stroke, mm
Stroke/ Bore
Mep, bar
Power, BHP
rpm
BHP/m
Kg/BHP
6UEC50LSII
500
1950
3.90
17.4
11220
124
1736
16.9
6UEC52LSE
520
2000
3.85
19.0
13920
127
2034
18.9
6RTA48T-B
480
2000
4.17
19.0
11880
127
1936
16.5
6S50MC-C
500
2000
4.00
19.0
12870
127
1998
16.1
Medium and small bore engines up to year 200075
15. The future of 2-stroke crosshead engines

In the 1970s there were many that thought that diesel technology was outdated, and that the large engines would be replaced with advanced steam turbine plants and later on with
gas turbines. It did not happen that way, at contrary, what was
believed to be a threat for the diesel engines is today nearly
expelled from the market, because of the diesel engines unsurpassed thermal efficiency.
It is indeed an impressive development that has taken place
during the 50 years that this report covers. The specific power
of the largest engines at any time, have been increased from
about 1000 BHP/cylinder (735 kW/cylinder) to 8000 BHP/
cylinder (6000 kW/cylinder) on the basis of an increase in
mean effective pressure from about 6 bar to 19 bar, a maximum pressure from 50 bar to 150 bar and an increase in mean
piston speed from 5.5 m/s to 8.5 m/s, as shown in Fig. 124.
The future will also see the diesel engine as the preferred propulsion machinery. Apart from the superior fuel economy this
is due to the great possibilities at hand to make use of the exhaust gas waste heat, where consumption reductions from 10
to 15 % are possible. This will of course require considerable
additional investments in the ship machinery plant, but the
pay pack time will only be 3 to 4 years under the prevailing
economical conditions, and the inherent reductions in CO2,
NOx, SOx and particulates will be an attractive extra free
bonus.
The engine development, as demonstrated in this report, has

brought along concepts that now are so close to each other
that it is nearly impossible to distinguish the three present
makes. The new development will take place on electronic
control systems and hydraulic activation systems. Sulzer, in
1981, was the first to make experiments with an engine having
a single cylinder activating concept. This was abandoned, but
MAN-B&W have taken over the idea and could in 2000 as
the fist designer install the new concept on a 6L60MC engine
in service. Mitsubishi has adopted a similar principle. Sulzer
have instead chosen an electronically controlled common rail
system, the RT-flex system mentioned previously, and the first
RT-flex common rail engine entered service in 2001. So, now
again there will be a marked difference between the makes,
however, it is outside the intentions of this report to get into
the details of this development.
The low-speed two-stroke engine has proved to be the most efficient prime mover for ship propulsion and, with turbocharging, has been developed to even greater efficiency with longer
times between overhauls. The power output has been developed by an order of magnitude from the first turbocharged
engines in the 1950s to the most powerful engines today. It
has also been developed to have the flexibility of burning really poor quality fuel oils and even natural gas. Besides the
drastic reductions in fuel consumption, there have also been
dramatic reductions in exhaust emissions. With such a past,
todays turbocharged two-stroke crosshead engines have an
exciting future.
7000
B&W
SULZER
5000
MHI
Stork
4000
Gtaverken
Fiat/GMT
3000
Mean effective pressure,

MEP bar
kW/cyl.
6000
MAN
Doxford
2000
1955
1960
1965
1970
1975
1980
1985
1990
1995
2000
MAN
Doxford
1955
1960
1965
1970
1975
1980
1985
1990
1995
2000
140
120
8,00
Pmax, bar
Mean piston speed,

m/s
Fiat/GMT
160
7,50
7,00
B&W
SULZER
6,50
B&W
SULZER
100
80
MHI
Stork
Gtaverken
60
MHI
6,00
Stork
Gtaverken
5,50
Fiat/GMT
MAN
1955
1960
1965
1970
1975
1980
1985
1990
1995
2000
Fig. 124 development in performance parameters over 50 years (Author)
76
Stork
Gtaverken
5
Year 1950
8,50
5,00
1950
MHI
11
10
9
9,00
Year
B&W
SULZER
8
7
6
1000
Year 1950
20
19
18
17
16
15
14
13
12
Year
40
Fiat/GMT
MAN
20
Doxford
0
1950
1955
1960
1965
1970
1975
1980
1985
1990
1995
2000
III Design timelines

In this section the development of some of the most important engine component characteristics are shown for all the
designers during the 50 years. The somewhat simplified characterisation is clarified by examples of actual designs and by
references to figures presented in the text of the previous two
sections.
The survey makes it possible to follow where and when new

ideas were introduced and by whom, and how they possibly
were adopted by competition later on. The design characteristics are in certain cases not identical for small and large bore
engines and differences have to a certain degree been noted
by additional text.
Components for B&W engines from VTBF to K/LGF are

very clearly shown in Half Century of Diesel Engine Progress at Mitsui [85].
In the horizontal lines the engine types of the different designs are included with the first name letter positioned at the
year of introduction. The name position on one of the three
horizontal lines belonging to each designer has no specific
meaning it is just a matter of finding space.
The systems or components listed are:

Scavenging systems (inside cover)
Crossheads
Cylinder covers
Cylinder liners
Pistons
Exhaust valves
Engine frames
Crankshafts.
Design timelines77
78
Year
Sulzer
1950
SAD
RD
SAD72
RSAD
HOTLo75
KEBS68/125
UEC-A
UEC75A
Mitsubishi
Werkspoor
Stork-Werkspoor
Stork
KZ-C
KZ78/140C
MAN
(A)
...S
750S
Fiat /GMT
-P

VGU / VGSU
520/900VGS-U
LBD BDS
600LBD
-J

K98FF
K-EF/FF
UEC-B
1060S
1970
RND105
RND
UEC-D
KSZ105/180
KZ-F KSZ
SW90/170
UEC-C
KZ-E
SW85/180
KZ-D
B...S
L-GF
KSZ-A
RNDM
RTA
1980
(E)
RLB
UEC-L
RTA-M
RTA-2
(F)
RTA-C
1990
RTA84T
RTA-2U
-LSE
RTA84TD
RTA-2UB
RTA-8T -TB
RTA96C
RTA60C
UECLSII UEC85LSC
UEC75LSII
UEC-LS
* eccentric X-head pin
UEC-HA UEC-LA
KSZ-C/CL
C
* * *
CC600
UEC-H
RLA
with seperated central pad support
58JS3C
-JS-C
KSZ-BL
UEC52E
UEC-E
KSZ-B
(C)
from attached high pressure pumps
steel shells with Al Sn lining
K/L-GFC
K/L-GFCA
L-MC/MCE
K98MC
L-GB/GBE
K-MC-C
S-MC-C
(D)
-J-C
SEAHORSE project
K-GF
thin steel shells, white metal lined

full width bearing
from high pressure system
750/1600 & 630/1400 VGS-U*

520/900VGS-U*
(B)
850/1700VGA-U
850/1700VGS-U*
1960
VTBF
VT2BF
74VTBF160
VTBF opposed piston
Gtaverken
Doxford
Burmeister & Wain
Crosshead
bearing material
cast-in white metal
design
2 seperate bearings
bearing lubrication from common lube oil system
Design timelines79
80
Year
Sulzer
Werkspoor
Stork-Werkspoor
Stork
Mitsubishi
MAN
Fiat /GMT
Gtaverken
Doxford
Burmeister & Wain
Cylinder liner
design
1950
(A)
VT2BF
BDS
-P
RSAD
SAD72
SAD
RD
STORK HOTLo75
1960
SW85/180
UEC-B
cast-in cooling tubes
(C)
1970
RND105
RND
(E)
RNDM
RLA
1980
RLB
RTA
(G)
(I)
UEC-L
(J)
K-MC-C
RTA-2
RTA-M
(H)
RTA-C
1990
RTA-2U
RTA84T
K98MC
-TB
RTA60C
RTA96C
RTA-8T
RTA-2UB
RTA84TD
-LSE
S-MC-C *
(K)
UECLSII UEC85LSC
UEC75LSII
UEC-LS
#liner in two pieces
UEC-HA UEC-LA
KSZ-C/CL
UEC-H
KSZ-BL
UEC52E
UEC-E
KSZ-B
2 part liner
UEC-D
KSZ105/180
SW90/170
KSZ-A
#
CC600
1060S
KZ-F KSZ
L-MC/MCE
L-GB/GBE
3 piece liner
2 piece liner with strong back
58JS3C
-JS-C
(F)
-J-C
L-GF
K/L-GFCA
K/L-GFC
(D)
850/1700VGS-U*
BS
K-GF
SEAHORSE project
(A)
520/900VGS-U*
UEC-C
KZ-E
K98FF
K-EF/FF
(B)
850/1700VGA-U
KZ-D
strong back
*bore/cast-in tubes cooling for large bore engines only
750/1600 & 630/1400 VGS-U*
KEBS68/125
UEC75A
UEC-A
KZ78/140C
KZ-C
750S
520/900VGS-U
VGU / VGSU
600LBD
LBD
-J
bore cooled (collar)
VTBF opposed piston
74VTBF160
VTBF
simpel with collar
Design timelines81
82
oil cooling
inner support
Year
Sulzer
Werkspoor
Stork-Werkspoor
Stork
Mitsubishi
MAN
1950
(B)
SAD
RD
SAD72
RSAD
STORK HOTLo75
KEBS68/125
UEC-A
UEC75A
KZ-C
(A)
KZ78/140C
S
750S
Fiat /GMT
-P
VGU / VGSU
520/900VGS-U
lower piston LBD BDS

600LBD
upper piston
UEC-B
(E)
1970
RND105
RND
SW90/170
UEC-D
RNDM
strong back
bore cooled
semi bore cooled
(I)
RLA
1980
RLB
(J)
UEC-L
38 & 48 bore only

RTA-M
RTA-C
(K)
RTA-2
1990
RTA84T
RTA-2U
(M)
-LSE
RTA84TD
RTA-2UB
RTA-8T -TB
RTA96C
RTA60C
UECLSII UEC85LSC
57 & 60 bore only
UEC75LSII
UEC-LS
90/160B/BL only
# with crown stainless steel coated

1060 engine
UEC-HA UEC-LA
RTA
(H)
UEC-H
KSZ-C/CL
CC600
58JS3C
-JS-C
K/L-GFC
(L)
K/L-GFCA
L-MC/MCE
with inconel cladding, large bore..
L-GB/GBE
K-MC-C
S-MC-C
KSZ-BL *
UEC52E
UEC-E
KSZ-B
-J-C
L-GF
# #
KSZ-A
# # # #
1060S
KZ105/180
(G)
K-GF
cast, radially ribbed
SEAHORSE project
KZ-F KSZ
BS
UEC-C
KZ-E
SW85/180
KZ-D
(D)
1960
-J
K98FF
(C)
K-EF/FF
outer & inner support
750/1600 & 630/1400 VGS-U*

(F) 520/900VGS-U*
850/1700VGA-U
850/1700VGS-U*
VTBF
VT2BF
74VTBF160
(C)
VTBF opposed piston
water cooling
shell, outer support
Gtaverken
Doxford
Burmeister & Wain
Piston
piston cooling
piston crown
Design timelines83
RTA84C piston withdrawal. (Wrtsil)
84
Year
Sulzer
Werkspoor
Stork-Werkspoor
Stork
Mitsubishi
MAN
Fiat /GMT
Gtaverken
Burmeister & Wain
Cylindercover
design
design
1950
(B)
VT2BF
(A)
RSAD
SAD72
SAD
RD
STORK HOTLo75
KEBS68/125
UEC75A
UEC-A
KZ78/140C
KZ-C
750S
520/900VGS-U
VGU / VGSU
UEC-B
1960
(E)
(G)
1060S
KSZ-A
(F)
1970
RND105
RND
SW90/170
(I)
RNDM
UEC52E
UEC-E
(J)
(K)
RLA
1980
RLB
RTA
(L)
RTA-2
RTA-M
RTA-C
1990
RTA-2U
RTA84T
RTA84TD
-LSE
RTA60C
RTA96C
-TB
RTA-2UB
RTA-8T
UECLSII UEC85LSC
UEC75LSII
UEC-LS
UEC-HA UEC-LA
UEC-L
KSZ52 only
KSZ-C/CL
CC600
S-MC-C
K-MC-C
K98MC
strong back, semi bore cooled

forged, bore cooled
L-MC/MCE
L-GB/GBE
UEC-H
KSZ-BL
KSZ-B
# small bore, 60 and below
UEC-D
KSZ105/180
KZ-F KSZ
BS
L-GF
K/L-GFCA
K/L-GFC
# large bore, 850 & 750 only
K-GF
(H)
850/1700VGS-U*
520/900VGS-U*
UEC-C
KZ-E
SW85/180
(C)
(D)
KZ-D
K98FF
( B)
K-EF/FF
strong back, steel/cast iron

single piece, cast iron
750/1600 & 630/1400 VGS-U*

850/1700VGA-U
VTBF opposed piston
74VTBF160
VTBF
single piece, steel cast, shell type

steel cast, 2 concentric pieces
Design timelines85
86
Exhaust valves
Year
Sulzer
Werkspoor
Stork-Werkspoor
Stork
Mitsubishi
Gtaverken
Burmeister & Wain
Spindles
Seat
Exh. valve(s), general
1950
(A)
(B)
( C)
(D)
RSAD
SAD72
SAD
RD
STORK HOTLo75
KEBS68/125
UEC75A
UEC-A
520/900VGS-U
VGU / VGSU
74VTBF160
VTBF
single valve, rocker arm

austenitic steel, stellite
uncooled
UEC-B
SW85/180
1960
K98FF
K-EF/FF
1970
RND105
RND
SW90/170
UEC-C
UEC-D
850/1700VGS-U*
520/900VGS-U*
K-GF
RNDM
UEC52E
UEC-E
L-GF
(F)
L-MC/MCE
K-MC-C
with rotator
1980
RLB
RLA
UEC-L
(H)
RTA-2
(G)
RTA-M
K98MC
-LSE
S-MC-C
(I)
RTA-C
1990
RTA-2U
RTA84T
-TB
RTA60C
RTA96C
RTA-8T
RTA-2UB
RTA84TD
with hydraulic rotator.
UECLSII UEC85LSC
UEC75LSII
UEC-LS
UEC-HA UEC-LA
RTA
UEC-H
L-GB/GBE
K/L-GFCA
1 valve, hydraulic actuation
1 valve, pullrods-yoke
chrome steel, no facing
K/L-GFC
4 valves, rocker arm/pullrod

austenitic steel no facing
750/1600 & 630/1400 VGS-U*

850/1700VGA-U
VT2BF
3 valves, rocker arm

nimonic, with rotator
water cooled
(Wrtsil)
Design timelines87
88
UEC-A
UEC75A
Mitsubishi
Year
Sulzer
SAD
RD
SAD72
RSAD
STORK HOTLo75
KEBS68/125
KZ-C
KZ78/140C
MAN
Werkspoor
Stork-Werkspoor
Stork
S
750S
1950
-P
VGU / VGSU
520/900VGS-U
#
#
LBD BDS
600LBD
UEC-B
1060S
1970
RND105
RND
UEC-D
KSZ105/180
SW90/170
K-GF
KSZ-A
(C)
RNDM
welded multi cyl box

welded multi cyl box
cast iron multi cylinder box
welded, one or two parts (bolted together)
(D)
RLA
1980
RLB
UEC-L
UEC-LA
38 type only
RTA-C
1990
RTA84T
RTA-2U
-LSE
RTA84TD
RTA-2UB
RTA-8T -TB
RTA96C
RTA60C
UECLSII UEC85LSC
UEC75LSII
small bores
UEC-LS
RTA-M
RTA-2
UEC-HA
RTA
UEC-H
KSZ-C/CL
CC600

# VGU/VGA-engines only
58JS3C
K/L-GFC
K/L-GFCA
L-MC/MCE
(E)
L-GB/GBE
K-MC-C
S-MC-C
small bores
-JS-C
KSZ-BL
UEC52E
UEC-E
KSZ-B
-J-C
L-GF
SEAHORSE project
KZ-F KSZ
BS
UEC-C
KZ-E
SW85/180
KZ-D
1960
-J
K98FF
K-EF/FF
welded single cylinder units

welded single cylinder units
# 750/1600 & 630/1400 VGS-U*

520/900VGS-U*
#
(B)
850/1700VGA-U
850/1700VGS-U*
VTBF
VT2BF
74VTBF160
(A) VTBF opposed piston
cast per 1 or more cylinders (boltet together)

A frames, cast or welded
cast iron, one or more parts (bolted together)
Fiat /GMT
Gtaverken
Doxford
Burmeister & Wain
Engine frame
cylinder frame
crank case frame
bedplate
Design timelines89
90
UEC-A
UEC75A
Mitsubishi
1950
KZ-C
KZ78/140C
MAN
Year
Sulzer
SAD
RD
SAD72
RSAD
STORK HOTLo75
KEBS68/125
S
750S
Fiat /GMT
Werkspoor
Stork-Werkspoor
Stork
VGU / VGSU
520/900VGS-U
LBD BDS -P
600LBD (B)
K98FF
K-EF/FF
UEC-B
1960
1060S
1970
RND105
RND
UEC-D
KSZ105/180
KZ-F KSZ
SW90/170
UEC-C
KZ-E
SW85/181
KZ-D
BS
K-GF
KSZ-A
RNDM
RLA
1980
RLB
UEC-L
(C)
(D)
RTA-C
38-type only
RTA-2
RTA-M
1990
RTA84T
RTA-2U
RTA84TD
RTA-2UB
RTA-8T -TB
RTA96C
RTA60C
UECLSII UEC85LSC
-LSE
52 and smaller bores
UEC75LSII
UEC-LS
*52/105 only
UEC-HA UEC-LA
RTA
UEC-H
KSZ-C/CL
CC600 only
58JS3C
-JS-C
small engines
welded shaft
K/L-GFC
(E) K/L/S-MC/MCEmark 3,4,5,
K/L-GFCA
L-MC/MCE
K98MC
L-GB/GBE
K-MC-C
S-MC-C
single piece, forged
KSZ-BL
UEC52E
UEC-E
KSZ-B
(B)
-J-C
L-GF
SEAHORSE project
750/1600 & 630/1400 VGS-U*
520/900VGS-U*
850/1700VGA-U
850/1700VGS-U*
(B)
-J
up to 9 cyl
(A)
semi-built cast or forged cranks, forged pins
VTBF
VT2BF
74VTBF160
VTBF opposed piston
fully built
Gtaverken
Doxford
Burmeister & Wain
design
Crankshaft
Design timelines91
References
Reference to Motor Ship in figure texts are in agreement with engine builder/designer in question.
General references:
1. MAN Diesel, Ole Srensen: B&W Dieselmotorens Historie 1898 2008 (in Danish).
2. Keijiro Tayama: Historical Development of Two Stroke Slow Speed Marine Diesel Engine, National Science Museum Japan, vol. 8 March 2007. (in Japanese).
3. Ture Rinman: The story of an engine, Gtaverken publication, 1964
4. David T. Brown: A History of the Sulzer low-speed marine diesel engine, Sulzer publication 1984
5. MAN internal documents from 1941 to 1984, MAN-Museum and Historical Archive of MAN Augburg (in German)
6. Ernst Jenny: The BBC Turbocharger, Birkhuser Verlag AG 1993
Specific references:
7. Sren Hansen: 2-stroke turbocharged engines for outputs above 10.000 BHP per shaft experience with the first plants of this type. CIMAC Den Hague 1955
8. H. Andresen: The Present Day Marine Engine (Paper read at Norsk Ingenirforening 1953), B&W publication
9. F.G. van Asperen: Development and service results of a high powered turbocharged two cycle marine diesel engine, CIMAC Den Haag 1955
10. The Motor Ship: A 9600-BHP Werkspoor-Lugt Engine, July 1953
11. The Engineer: 1800 b.h.p. Werkspoor Lugt Marine Oil Engine, February 1951
12. K.K.van Hoffen: Werkspoors Stake in Marine Diesel Engine Development, The Motor Ship April 1970
13. G. Wieberdink & A. Hootsen: Supercharging by means of turboblowers applied to two-stroke diesel engines of large output, e.g. More than 10.000 BHP, for
ships propulsion, CIMAC Den Haag 1955
14. E. A. Van der Molen: Test-bed Research on the first Stork Large Bore Engine, The Motor Ship January 1963
15. L-T Collin: The constant pressure charged two stroke engine, its present state and development possibilities, CIMAC Copenhagen 1962
16. Hideo Fujita: High Powered UEC Marine Diesel Engine, CIMAC Den Haag 1955
17. G.G.Jackson: Progress on the Development of the Doxford Oil Engine, Doxford Engines 1979
18. P. Jackson: Development of the Opposed Piston Engine to 10.000 H.P. and over, CIMAC Den Haag 1955
19. T.A.Greenwell: 60 Years of Doxford Opposed-piston Engine Development, The Motor Ship, April 1970
20. The Motor Ship June 1962: A Doxford Opposed-piston Engine of 30.000 b.h.p. Envisaged
21. The Motor Ship October 1966: Design changes in the Doxford J Engine
22. Dr. Ing Antonio Gregoretti: La suralimentation des moteurs 2 temps balayage transversal ayant particulirement gard aux moteurs Fiat, CIMAC Zrich
1957
23. Dr. Ing. Antonio Gregoretti: Considerations sur le Projet dun Moteur Diesel de la Puissance maximum de 3.000 CH/CYL., CIMAC Copenhagen 1962
24. P L Scioldo: A marine unit to give more than 40.000 bhp at 102 rev/min, The Motor Ship, Special Survey, August 1967
25. Internal MAN document by Dr. Ing. D. Von Lassberg 1974/75: Die entwicklung der Mittel-Gross-Dieselmotoren im Werk Augsburg der MAN 19411971
26. Prof. Dr. Ing. E. Srensen & Dr. Ing. F. Schmidt: Recent Development of the MAN Marine Diesel Engine, The Institute of Marine Engineers 1964
27. W. Kilchenmann: 75 Years of SULZER Diesel Engines, Sulzer Technical Review 4/1974
28. G. Wolf: The design and development of the RND engine, The Motor Ship, Special Survey, October 1969
29. Henry Klintorp: Burmeister & Wain engine installations in tankers, B&W 1962
30. Dr. Ing. G. Ciliberto & Dr. Ing. L. Mariani: Considerations sur le projet et les essais de mise au point du moteur Fiat 1060. CIMAC Stockholm 1971
31. G. Wolf: Konstruktion und Entwicklung der neuen Sulzer RND-Motoren, MTZ November 1970, Courtesy of Wrtsil
32. Dr. Hugo Schobel: Development and operating results of MAN KZ105/180 engine, The Motor Ship, August 1967
92
33. Carsten Andersen: Development of a Super Large Bore Engine, North East Coast Institution of Engineers and Shipbuilders, 1968
34. The Motor Ship December 1967: Burmeister & Wains New K98FF Engine
35. J. Schmidt-Sorensen: Development of B&W KGF-Engines
36. The Motor Ship April 1970: New Stork 900 mm-bore Engine Capable of 3.500 bhp/cylinder
37. E. A. Van der Molen: The Development of Stork Engines in 1966, The Motor Ship February 1967
38. L.G. Thulin & G. Dahlbring: Gtaverken turbo-charged crosshead diesel engines, GV, 1969
39. Sven Landberg: A Review of Gtaverken Diesel Engine Progress, The Motor Ship, Special Survey, November 1964
40. J. Dragsted: Untersuchung von Zylinderdeckeln fr Schiffs-Zweitakt-Dieselmotoren, Motortechnische Zeitschrift, November 1970
41. J. Dragsted: Considerations on hydraulic operation of exhaust valves, The Motor Ship, Special Survey, November 1972
42. Dr. Ing. Ruggro Maciotta: Uber den gegenwrtigen Entwicklungsstand der GMT-Dieselmotoren 5 Jahre nach Beginn ihrer Produktion, Motortechnische
Zeitschrift 38, 1977
43. Internal MAN document by K Luther: Die Viertakt- und Zweitaktmotoren der M.A.N.Augsburg 1970 1975
44. J.F.Clarke: Power on Land & Sea, Hawthorn Leslie Ltd.
45. J.F. Butler & E-P Crowdy: Development of a high output medium speed two cycle opposed piston engine, CIMAC Washington 1973
46. R. Tsnuneya, The two-stage turbocharged E-type engine, The Motor Ship, special survey , May 1979
47. A. Schiff: Hand-out at a press conference in Tokyo 20. February 1980
48. Ole Grne & Per Lindhardt: New problems in burning todays residual fuel, B&W 1979.
49. Folker Bhm: Die MAN Zweitaktmotoren des konstruktionsstandes B/BL, Motortechnische Zeitschrift 39 , 1978
50. R. Hfner & H. Lindner: Strahlgekhlte Waben ein neues MAN-konstruktionselement fr Brennraumwende, MTZ 1977
51. GMT, March 1979
52. A. stergaard & R Tjnneland: Turbocharging of two-stroke diesel engines, The Motor Ship November 1978, Special survey.
53. A. Oestergaard & R. Tjoenneland: Further development of constant pressure turbocharged, poppet valve type low-speed diesel engines. CIMAC Vienna 1979.
54. Schiff & Hafen/ komandobrcke, Heft 9 1979
55. G. Wolf, A. Bitterli & A. Marti: Bore-cooled Combustion Chamber Components on Sulzer Diesel Engines, Sulzer Technical Review 1/1979.
56. A. Schiff: Hand-out at a press conference in Tokyo 29. May 1979
57. Folker Bhm: MANs Two-Stroke Programme with particular reference to C/CL long-stroke engines and electronic injection, MAN publication 1979. (Historical Archive of MAN Augsburg)
58. K. Ujita: The constant-pressure turbocharged H-type engine, The Motor Ship, May 1979
59. The Motor Ship July 1980: Mini-bore low speed crosshead engines launched by Mitsubishi.
60. The Motor Ship, Special Supplement, August 1983
61. The Motor Ship, Special Supplement, December 1982
62. Y. Hirayama: Design and development of the superlongstroke L series engines, The Motor Ship, supplement November 1983
63. M. Briner, N. Mikulicic and P. Baumgartner: Sulzer RTA engines for economical ship propulsion, Sulzer paper April 1986
64. A. stergaard: L-GF exceeding expectations, Schiff & Hafen, Heft 9 1978
65. Protocol of 1997 to amend MARPOL73/78, Annex VI of Marpol 73/78 Regulations for the Prevention of Air Pollution from Ships. International Maritime
Organisation 1997
66. Y. Okumura: Comprehensive countermeasures to enhance reliability and durability of marine diesel plant, CIMAC Interlagen 1995
67. K. Wilson, N. Hansen, P Storegaard Jensen & J. Dragsted: Service and operational experience as discussed and compiled by CIMAC WG Engine Users,
CIMAC Copenhagen 1998
68. J. Dragsted, N. Hansen, & K. Wilson: Service experience with diesel engines as compiled by the CIMAC Users WG, CIMAC Hanburg 2001
69. T. Biwa, T. Fukuda, T. Uyama & T. Tanaka: Development of double layered cylinder liner, CIMAC Florence 1991.
70. C-E Egeberg & A. stergaard: The MC Engine and its future development, CIMAC Hamburg 2001.
References93
71. G.A. Lustgarten & K Aeberli: Sulzer RTA84C The Modern Containership Engine, Motor Ship Conference London 1990
72. K. Aeberli & G.A. Lustgarten: Improved running performance of low-speed diesel engines, New Sulzer publication 1987
73. K. Aeberli: Results of the application of High technology research methods and their transformation into the design of simple and reliable two stroke engines,
CIMAC Florence 1991.
74. R. Demmerle: Sulzer RTA84C and RTA96C engines: The reliable driving forces for large, fast containerships, Wartsil NSD October 1997
75. K Aeberli & M Geist: The Recent Development in the Sulzer RTA Low Speed Engine Programme; Highlights and Results of Operation, CIMAC Copenhagen
1998.
76. R. Demmerle et al.: New insights into the Piston running behaviour of Sulzer large bore Diesel Engines, CIMAC Hamburg 2001
77. H. Sakabe & K. Sakaguchi: The Uec engine program and its latest development, CIMAC Kyoto 2004
78. Y. Hirayama et al: Development of the Mitsubishi UEC75LSII Type Diesel Engine, MHI July 1988.
79. Imakire et al: Development of MET-SR-VG Turbocharger Driven by Radial-Flow Turbine with Variable Geometry Nozzle; CIMAC Tianjin 1989
80. S. Yoshihara et al: The main bearing behaviour of UEC type large bore diesels, CIMAC Interlagen 1995
81. H Brunner: The Sulzer RTA84T engine design, New Sulzer Diesel publication, August 1992
82. K Aeberli & M. Geist: The Recent Development in the Sulzer RTA Low Speed Engine Programme; Highlights and Results of Operation, CIMAC Copenhagen
1998
83. C. E. Egeberg; The MAN B&W two stroke diesel engine and its prospects, CIMAC Copenhagen 1998
84. Markus Geist: Sulzer RTA-8T engines: Compact two strokes for tankers and bulk carriers, Wrtsil NSD April 1998
85. Half Century of Diesel Engine Progress at Mitsui, Mitsui publication 1976
86. Koichiro Imakire: Historic Development of Exhaust Gas Turbocharger for Low Speed Engine in Japan, National Museum of Nature and Science 2011
94
VLCC
Fuel savings: ~15 t/day
Ultra-large container vessel

Suezmax tanker
Fuel savings: ~4.5 t/day
Capesize bulker
THE NEW
GENERATION X ENGINES
At Wrtsil, we are passionate about optimising lifecycle value by delivering whatever
you need from the most complete marine offering. Wrtsils new Generation X
Engine family includes Wrtsil X62, Wrtsil X72, Wrtsil X82 and Wrtsil X92
low-speed engines which are ideal for most merchant vessels. This includes such
ship types as Suezmax tankers, Panamax/Capesize bulkers, VLCC/VLOC, ultralarge container vessels, Panamax container vessels, as well as feeder container
vessels. High propulsion efficiency, low fuel consumption and full compliance
with emission legislation are key features of Wrtsils new Generation X Engines.
In some cases, the fuel savings can be over ten tonnes per day compared to
previous solutions. LETS FIND YOUR SHORTER ROUTE TO BIGGER PROFITS.
www.wartsila.com

Dragsted History Booklet 2013

Uploaded by

Copyright:

Available Formats

Dragsted History Booklet 2013

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Dragsted History Booklet 2013

Uploaded by

Copyright:

Available Formats

The first 50 years of Turbocharged 2-stroke, Crosshead, Marine Diesel Engines

The First 50 Years

Fig. 2 Scavenging systems in turbocharged, crosshead diesel engines (Author)

Burmeister & Wain

# impulse charging for 6,9 and 12 cylinders only

Seahorse engine only

*with seperate exhaust camshaft

750/1600 & 630/1400 VGS-U*

TURBOCHARGED TWO STROKE CROSSHEAD MARINE DIESEL ENGINES

The First 50 Years

POWER & RELIABILITY  6

Large bore engines  27

Super large bore engines  28

B&W K98FF, MAN KSZ105, Fiat 1060S. SULZER RND105

The last Stork-Werkspoor 2-stroke crosshead engines  32

Gtaverken stops 1972  33

B&W K-GF, GMT B, MAN KSZ-A, SULZER RND-M, Doxford Seahorse,

B&W L94NF, MHI UEC-E, MAN KSZC/CL-H

Fuel quality versus engine reliability  41

II FUEL CONSUMPTION, RELIABILITY & MULTIPLICITY 42

10. The fuel consumption race 47

11. Three survivors  53

12. Engines for the big Container Ships  62

13. Engines for Very Large Crude Carriers, VLCCs  69

14. Medium and small bore engines up to year 2000 72

15. The future of 2-stroke crosshead engines 76

Setting The Scene

Apart from production taking place at the companys own

The BBC/ABB turbocharger development is described in a

Setting The Scene5

I POWER & RELIABILITY

Fig. 3 Cross section of the first turbo-charged 2-stroke engine

only at the introduction of the later engine series K-EF and

Fig. 4 The required modification of the standard unturbocharged

Opposed piston engines

The first turbocharged engines and some later generations7

could only build their engines

The first turbocharged engines and some later generations9

The 12-cylinder engine developed 9600 BHP at 125 rpm

Fig. 9 KEBS68/125 engine top. Note the special exhaust valve

The engine structure is of cast iron and the bedplate carries

Fig. 8 Diagram of the Werkspoor-Lugt constant pressure scavenging

The engine has 4 exhaust valves in cages and a very special

Like B&W Werkspoor focus in their first published service

Fig. 11 8 cylinder HOTLo75/150 engine with 4 turbochargers.

Fig. 12 Cylinder Cover with two outlets per cylinder, HOTLo75/150.

Fig. 10 Cross section of the first Stork turbocharged crosshead

Why the choice of 4 exhaust valves? Today one valve is

The first turbocharged engines and some later generations11

Fig. 14 Exhaust valves with water cooled valve seats, HOTLo75/150.

Fig. 15 Piston mounted on rod connected to the crosshead,

The 520/900VGS-engine is the first all welded engine (the

Each exhaust valve is operated by cam segments mounted on

The first turbocharged engines and some later generations13

Fig. 17 First turbocharged Gtaverken 2-stroke engine on test bed,

Regarding the latter it has been mentioned by former GV

Fig. 18 Diagram of the Gtaverken constant pressure charging

The first turbocharged engines and some later generations15

Fig. 21 Engine top with exhaust valve gear, 3UEC72/150. (CIMAC

The camshaft to drive the exhaust valves is located along the

POWER & RELIABILITY 6

Large bore engines 27

Super large bore engines 28

The last Stork-Werkspoor 2-stroke crosshead engines 32

Gtaverken stops 1972 33

Fuel quality versus engine reliability 41

II FUEL CONSUMPTION, RELIABILITY & MULTIPLICITY 42

10. The fuel consumption race 47

11. Three survivors 53

12. Engines for the big Container Ships 62

13. Engines for Very Large Crude Carriers, VLCCs 69

14. Medium and small bore engines up to year 2000 72

15. The future of 2-stroke crosshead engines 76

Fig. 3 Cross section of the first turbo-charged 2-stroke engine

Fig. 4 The required modification of the standard unturbocharged

Fig. 9 KEBS68/125 engine top. Note the special exhaust valve

Fig. 8 Diagram of the Werkspoor-Lugt constant pressure scavenging

Fig. 11 8 cylinder HOTLo75/150 engine with 4 turbochargers.

Fig. 12 Cylinder Cover with two outlets per cylinder, HOTLo75/150.

Fig. 10 Cross section of the first Stork turbocharged crosshead

Fig. 14 Exhaust valves with water cooled valve seats, HOTLo75/150.

Fig. 15 Piston mounted on rod connected to the crosshead,

Fig. 17 First turbocharged Gtaverken 2-stroke engine on test bed,

Fig. 18 Diagram of the Gtaverken constant pressure charging

Fig. 21 Engine top with exhaust valve gear, 3UEC72/150. (CIMAC

Fig. 23 Pictorial arrangement of running gear and gas forces.

Fig. 25 Comparison of crankshafts for 4 cylinder engines. (Doxford [17])

Fig. 27 Cylinder liner in 3 parts for 76J engines (MS [21])

Fig. 28 750S on test bed in Turin. (CIMAC [22])

Fig. 30 Diagram of the Fiat cross

Fig. 33 Impulse-parallel, Impulse-series parallel and constant

Fig.34 Cross section of the KZ78/140, the first turbocharged MAN

Fig. 35 First turbocharged Sulzer crosshead engine, SAD72. Cross

Fig. 37 Scavenging efficiency with different porting arrangements.

Fig. 38 Piston development for RD-engines, from oil cooling to water

2. Large Bore Engines

3. Super Large Bore Engines

Fig. 40 MAN, Constant-pressure parallel system. (MS [32])

Fig. 41 K98FF, 1060S, RND105, KSZ105 (B&W [33], MS [24],

Fig. 43 Piston design development for the KSZ-engines. (MS [32])

Fig. 42 Combustion Chamber and fuel injection arrangements

Fig. 47 Combustion chamber temperatures RND105 at 4000 BHP/

Fig. 46 Combustion chamber temperatures, 1060S with an oil cooled

4. The last Stork-Werkspoor 2-stroke crosshead engines

5. Gtaverken stops in 1972

Fig. 50 Combustion chamber development for 850/1700VGS-U. (GV)

Fig. 49 Cross section 850/1700VGS-U, all welded structure and

Fig. 51 Piston crown development for the 850/1700VGS-U engine.

Fig. 54 Bore cooled Cylinder Cover, principal calculation of

Fig. 56 Hydraulically activated exhaust valve for the K90GF engine

Fig. 58 Constant pressure system with electric blower in series for