by Peter Yule
class. Consequently, the Swedish navy was no longer involved
in the improvement of Hedemora engines. Roine Carlsson,
the Swedish Minister for Defence at the time, recalls telling
Oscar Hughes that the next generation of Swedish submarines
would have German diesels, but the Australian project stayed with
Hedemora for the design advantages and to avoid delays.7
S E T T I N G T O W O R K 1 9 8 7 – 8 9
127
The $60 million contract for the submarines’ batteries illus-
trates both the international nature of the project and the depth of
Australian industry involvement. During project definition it was
widely assumed that the Swedish company Tudor would provide
the batteries, but the tender was won by Pacific Marine Batter-
ies, a partnership between Varta Batteries of Germany and Pacific
Dunlop Batteries of Australia. Ove Gustafsson was the technical
director of the Scandinavian branch of Varta, which built subma-
rine batteries for the Swedish navy, and helped put together a bid
to provide the Australian batteries. Varta won the contract and
Gustafsson went to Australia to set up the joint venture company
Pacific Marine Batteries with Pacific Dunlop. He couldn’t find a
manager and was eventually told: ‘You’ve caused the problem so
you’ll have to fix it.’ He moved to Australia in 1990 to establish
the business.
Originally, the batteries were to be built in Sydney, where
Pacific Dunlop assembled the Oberon batteries. However,
Gustafsson believed that it would be better to build a new fac-
tory close to ASC in Adelaide, largely to save on the transport
costs of each 400-tonne battery. The South Australian government
assisted by giving land for the factory and clearing away polluted
soil. The challenge of starting a manufacturing business ensured
‘a couple of very tough years – the learning curve was nearly ver-
tical’ – but the company provided six batteries to ASC, and now
has a long-term contract to provide replacement batteries.8
Below the major contractors were 426 sub-contractors and
many more sub-sub-contracts. Of the sub-contractors, 306 were
companies operating in Australia (including both Australian com-
panies and Australian subsidiaries of foreign companies), with the
remaining 120 companies coming from Sweden, Germany, France,
Britain, the United States, Canada, Italy, Holland, Belgium,
Norway and Denmark.9 Almost half of these contracts (56) were
with British companies, with 21 American, 15 Swedish and 17
German companies also involved.
The ASC site on the banks of the Port River at Osborne was
always referred to as a ‘greenfields site’, but this conveys a very dif-
ferent image from the reality of the desolate, polluted swamp seen
by the notables who came to watch Prime Minister Bob Hawke
drive the first peg on 29 June 1987. Jeff Rubython from Wormald
moved quickly in planning the new yard. The essential feature
128
T H E C O L L I N S C L A S S S U B M A R I N E S T O R Y
of the design was that it was not a traditional shipyard – ships
were not constructed as a single entity but assembled from mod-
ules brought from elsewhere. Parts and equipment were manufac-
tured in every state of Australia and in many countries around the
world, and brought to Osborne at the appropriate time to install
in a submarine.
Concrete Constructions Pty Ltd was the construction manager
for building the shipyard and employed up to 400 men on the site.
Significantly, the greatest problems came from wind, sand and the
unstable soil rather than from labour unrest. Oscar Hughes recalls
that the promise to keep the unions under control had been one of
the main attractions of the South Australian bid. In the early days
of building the yard a strike occurred over a demarcation dispute.
Hughes was in Beazley’s office and had him ring Ralph Willis,
the Minister for Industrial Relations. The message was that if the
strike was not over within 24 hours construction would move to
Western Australia. Almost immediately John Bannon rang and
asked if Hughes was serious. Hughes said he was and within
24 hours the whole issue was dead. For Hughes, the $20 million
already spent was nothing compared with the hundreds of mil-
lions it could have cost to continue the project with poor industrial
relations.
On 22 December 1987 ASC reached an agreement with three
key unions on a national agreement covering the submarine
project. Among other things this limited the number of unions
on the Osborne site, greatly reducing the likelihood of demarca-
tion disputes. Industrial harmony was one of the main reasons
why the facilities at Osborne were completed on time and under
budget. Karl Bertil Stein recalls that Kockums was surprised by
this, as: ‘We had heard that nothing was done on time in Australia
because of the construction workers. We were delighted that there
were so many fewer days lost on strikes at Adelaide than we
expected.’10
All through the process of getting the project up and running,
Oscar Hughes was always involved – bullying, cajoling, com-
manding and persuading. His view was that ‘if you don’t behave
like a bulldozer then the project is going to get into trouble’. As
the industrial relations episode showed, anywhere there was an
obstacle that threatened to hold up the project, there would be
Oscar Hughes with his bulldozer. One of his project staff, Ken
S E T T I N G T O W O R K 1 9 8 7 – 8 9
129
Grieg, recalls that: ‘He was not the easiest person to work with,
but I guess he wasn’t trying to be easy.’
The Swedes, both at Kockums and in the Ministry of Defence,
respected and admired Oscar Hughes. Major General Kurt Blixt,
who chaired the committee supervising the memorandum of
understanding between the Swedish and Australian governments,
says: ‘We had no problems getting on with Oscar Hughes – he
had a role to take the project forward. He was rather hard when
it came to discussions but he was always fair.’ Similarly Tomy
Hjorth, managing director of Kockums and chairman of ASC from
1990 to 2000, thought Hughes an excellent project director: ‘He
made many decisions at a time when decisions were needed on a
continuous basis; he made deals, most of which were reasonable.’
However, Hjorth also observed that: ‘Oscar Hughes was running
the project outside the rest of the navy. This was necessary but it
created enemies for the project.’
Others shared this view that the project had become dissoci-
ated from the navy. Paddy Hodgman, who was the navy’s public
information officer from 1990–93, observed that:
In the late 1980s there was a substantially independent and
powerful submarine project. They still saw the chief of the
navy as being in charge but there were centrifugal forces and
divergences of view and approach between the project and
&
nbsp; stakeholders such as the submarine community, the rest of
the navy and others in defence and government. This meant
that ASC was always able to find folk to play off against each
other. Defence did not have it in one sock – it did not have
unity in what it wanted, what was wrong, what the fix might
be – effectively defence had disempowered itself.11
The consequences of this were not evident at the time, but became
clear when the submarine project ran into strife later in the 1990s.
Generally, the feeling around the project – at ASC, Kockums and
the project office – was that the project’s early stages had been
well managed and that all was going well.
C H A P T E R 12
Designing the Collins class1
Sweden is a country with few tall buildings. There are none in
central Stockholm and few in other cities. Malm ö in the south
is a slight exception to the general pattern. Today the main fea-
ture of the skyline is the new 54-storey ‘Turning Torso’ building,
built on the old shipyard, but for the previous 25 years the city
was dominated by the huge gantry crane and the 11-storey head-
quarters of its largest employer, Kockums. It was in this solid,
cylindrical building overlooking the increasingly derelict shipyard
that Australia’s new submarine was designed.
Traditionally Sweden’s submarine force was made up of several
classes, with each class being an evolution from the previous class
rather than a completely new design. Kockums’ designers did not
sit down with a blank sheet of paper, but rather had an evolving
submarine design that they were always testing and modifying.
The process of designing the new Australian submarine began
even before the navy’s operational requirements were released.
Olle Holmdahl, Kockums’ head of systems engineering in the
late 1970s, recalls that when Kockums’ designers first heard of
a possible submarine project in Australia they pooled their expe-
rience and knowledge of Australia’s Oberon operations, and then
130
D E S I G N I N G T H E C O L L I N S C L A S S
131
put together a design to meet the anticipated requirement. The
emphasis was on long range and endurance, with the battery size
paramount. Roger Sprimont and Olle Holmdahl discussed this
conceptual design with an Australian navy team led by Bill Rourke
as early as December 1981.
The conceptual design became a more detailed proposal dur-
ing 1983 in response to the Australian request for tenders. From
the start the primary aim of Kockums’ designers was to meet the
Australian requirements and avoid presenting a generic design
with the message ‘This is what you want’. The shape of the Aus-
tralian submarine was based on the V ästergotland, which entered
Swedish naval service in 1986, but was enlarged and redesigned
to meet Australian range and endurance requirements.
Once Kockums was selected for the project development phase
in May 1985 a far more detailed process began. A core design team
headed by Olle Holmdahl worked exclusively on the Australian
project, with many more involved when their particular exper-
tise was needed. The Kockums designers worked closely with the
Australian navy team at Malm ö, taking their guidance and occa-
sionally arguing for the revision of some specification, either to
reduce risk or improve results.
One area where Kockums disagreed with the navy team was
over the proposed requirement for automation, seen by Kock-
ums as excessive and potentially dangerous. Frequently struggling
to crew submarines, the Australian navy sought to automate as
many functions as possible to reduce crew numbers in the new
submarines. The initial requirement proposed an automatic sur-
facing system autonomously blowing the tanks and surfacing if
a hatch or valve was not closed. The Kockums designers argued
that this could be extremely dangerous if the submarine popped
up like a cork without knowing what was above. The requirement
was modified.
The use of flexible hoses was another element that the Aus-
tralian navy opposed, but these were central to the Kockums
design philosophy. Modular construction allowed Kockums to
place all machinery on rigid, heavy platforms that were then slid
into the hull as completed units and mounted on rubber isolating
blocks. This resulted in less machinery noise being transmitted into
the water through the hull and reduced the chances of damage to
the machinery from explosive shock. The benefits of isolating the
132
T H E C O L L I N S C L A S S S U B M A R I N E S T O R Y
submarines’ mechanical components would have been compro-
mised if machinery was then connected to the hull by rigid pipes,
so flexible hoses were used.
Conventional submarines have diesel engines to charge their
batteries. Needing air to operate, they can only run when the
submarine is surfaced or close enough to the surface to draw air
through a raised pipe in an operation known as ‘snorting’. On or
near the surface and running hot and noisy engines, submarines
are at their most vulnerable while charging their batteries. Conse-
quently, the diesels need to be powerful to recharge the batteries
as quickly as possible, and when snorting are subjected to extraor-
dinary stresses, as when the air supply is suddenly cut by waves
washing over the snort tube – salt water and diesel engines are nat-
ural enemies. Submarine diesels are far more complex and operate
under far more difficult conditions than any other diesel engines.
During his navy service, Roger Sprimont had seen turbo-
charged diesels aboard a Soviet Foxtrot submarine. Such systems
use a turbine driven by engine exhaust gases to pre-compress air
entering a cylinder and increase power output. They are more
efficient than the mechanical superchargers that have been used
in most designs and Sprimont thought they would provide the
performance sought by the Australians. He approached MTU,
one of the largest builders of diesels in the world, but they
would not adapt their design to use a turbo-charger. Designing a
turbo-charged diesel to operate while snorting is difficult – engine
operating pressure varies through the troughs and crests of the
waves and varying exhaust back-pressure applies stresses through
the turbine. Kockums went back to Hedemora, the company
that had supplied reliable diesel engines for most of the Swedish
submarines.
In any case, the Hedemora engine had some advantages. It
was rugged and able to operate with some components disabled,
and the crankshaft could be removed horizontally without hav-
ing to cut open the submarine. The company was also expe-
rienced in turbo-charged engines. Importantly, the Hedemoras
could be built in a country like Australia with no established
diesel engine manufacture, and this would help achieve a 70 per
cent Australian industrial content.
However, the engine for the
Australian boat was bigger, with more cylinders, than any subma-
rine engine Hedemora had built previously; it had an additional
starter motor added for operational safety reasons and, following
D E S I G N I N G T H E C O L L I N S C L A S S
133
a design review, had the flywheel removed to save weight and
improve connection to the generator unit.
The large amount of fuel oil needed meant that the designers
‘had to look for tank space everywhere’.2 Consequently the fuel
supply system was complex, with 15 separate tanks each able to
supply the engines either separately or through the complete sys-
tem. Pelle Stenberg comments that this made it necessary to follow
strict sequences in emptying the tanks to keep the boat in trim.
The detailed requirement for the new submarines included
many specifications that were based closely on the Oberons,
extending features that had worked well. Kockums thought this
left too many things based on what, by the late 1980s, was a very
old design. They proposed alternative solutions based on the latest
Swedish submarines, such as a trimming system based on that of
the V ästerg ötland. Such differences were generally resolved ami-
cably and both parties felt that the design improved as a result.
A major concern during the design phase was ensuring that
the components to be supplied by the major sub-contractors met
the Australian requirements and were compatible with the overall
design. The need for close liaison between designers and suppliers
is illustrated by the design of the periscope masts. The original
requirements specified that these should be faired to minimise their
wake, but somehow this requirement was lost along the way. Ian
Hill, who was a member of the Australian navy team at Malm ö,
thinks that this was due to excessive confidence on the part of the
periscope company that their design would not need modifications
to suit the Collins’ ambitious requirements. Consequently they
had not studied the benefits of streamlining the periscope mast.
The fundamentals of the design were settled before the contract
between ASC and the Commonwealth was signed. Nonetheless,
the design task facing Kockums after June 1987 was daunting.
Gunnar Öhlund recalls that:
The design of Collins was a huge challenge. It is a big
