by Peter Yule
prudence might have suggested a more predictable counterbal-
ance. The British and Dutch camps expected their more mature
designs would put them in this role.2
Yet the Swedish Type 471 appeared even riskier than its Ger-
man competitor: it was more than twice as large as any of
Kockums’ previous submarines, the company had no experi-
ence in managing overseas building programs and Sweden’s non-
membership of NATO posed security problems. There was a
widespread feeling that Kockums was chosen as a straw man to
be tried, found inadequate and thus ensure the selection of the
German submarine.3
If the bets were on IKL/HDW to be accepted as the submarine
builder, no odds would have been posted against Rockwell to
provide the combat system. Cabinet insisted on a competition,
but it was clear that Rockwell was the preferred tenderer and
most people thought that Signaal was in the competition simply
to entice a better offer from Rockwell.4
To assist the designers to understand the navy’s expectations,
the project appointed liaison teams to work with each company.
These were made up of submariners, navy and civilian engineers
and naval architects and, on the combat system side, some of the
software engineers from the SWSC who had designed the ambi-
tious combat system requirements. Mick Millington from the cen-
tre went to Signaal in Holland with submariner Rod Fayle from the
project office. Rick Neilson went to the Rockwell headquarters
in Anaheim together with submariner John Dikkenberg who,
because he spoke Dutch, had expected that his posting would
be with Signaal.5
Engineer Greg Stuart and Commander Rick Canham went
to Malm ö to work with Kockums, while engineer Peter Bull
and Commander Tony Parkin were sent to L übeck and Kiel as
the liaison team with IKL/HDW. Greg Stuart had been working
with submarines since 1972 and wondered if going to Sweden
T H E P R O J E C T D E F I N I T I O N S T U D Y 1 9 8 5 – 8 6
91
was a good career move as he thought that Kockums would
not win.
In August 1985 Oscar Hughes issued instructions for the over-
seas liaison teams covering their relationships with the contractors
and foreign navies and communications with the project office in
Canberra. There were several prohibitions: no documents were to
be altered nor any ‘interpretation, explanation or guidance’ pro-
vided that might result in changes to the design, its cost or sched-
ule, or that might constrain the contractors’ initiative or actions
and thus affect the competitive nature of the definition studies.6
The teams were instructed not to communicate with each other,
and information concerning the sensitive issue of the integration
of the submarines and the combat systems was only to be passed
through the office of the project director.
A system of ‘Chinese walls’ was put in place and the partic-
ipants agree that it worked. There were no accusations of the
liaison teams leaking information to competitors during the defi-
nition study and the accounts of the participants show that none
had an accurate idea of the experience of their counterparts. How-
ever, none of the overseas liaison teams, except perhaps the team
in Germany, followed the dispassionate objectivity mandated by
the directive.
At Signaal Mick Millington and Rod Fayle continually pushed
the company to design a system around distributed architecture,
although this was not Signaal’s preferred system architecture. The
Signaal engineers were sceptical, with one of them saying after
a meeting with Millington, ‘That man has no boundaries to his
fantasies’. Signaal’s chief computer engineer would look at the
Australians’ suggestions and agree Signaal could probably do it
but always with ‘there’ll be two problems’ so that addressing the
problems created an exponential problem curve for the project
team.
The Signaal proposal was based on the system designed for the
Taiwanese Sea Dragons, with dual central processors ‘very similar
to what the Rockwell solution ended up being’, says Millington.
Then, about the time of the mid-term review Signaal suddenly
reconsidered. Millington had visited Philips Electronics at Jarfala
in Sweden, which was developing a distributed architecture com-
bat system for the Danish Navy that used the common consoles
and processors sought by the Australians. At a meeting following
92
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
his return to Signaal, Millington hammered the table and said:
‘You’ve fundamentally missed the basics about what this combat
system is all about – it’s not dual processors we want, we want
distributed processors.’ He told them they were building a sys-
tem just like everyone else and this would not win the job – they
needed something special to meet the true objectives of the combat
system.
Signaal replaced its chief system design engineer and com-
pletely revised their design. Millington thinks Signaal’s revised
system was brilliant. He has been involved in designing many
combat systems and studied many more and still considers that
Signaal’s proposal for the Australian submarines was ‘the best
combat system architecture I’ve ever seen – it looks simple but it
is bloody awesome’.
The enthusiasm of Millington and Fayle was at odds with the
advice Signaal was receiving from senior sources in Canberra that
it was important to contain costs. If cost was to be a central con-
sideration, the Dutch company could not afford to invest in devel-
oping new approaches.
Similarly, in the United States Neilson and Dikkenberg were
encouraging Rockwell to adopt an adventurous approach to the
combat system’s functional specifications. From Neilson’s point
of view the liaison teams were there to help the company win,
because everything the teams did helped the company propose
something better for Australia.
In Malm ö, Greg Stuart insisted that his team avoid getting too
close to Kockums but found that it was a fertile working environ-
ment. The Australians got on well with Swedish naval personnel,
who were open and friendly and pleased to show them design
concepts at work in operational submarines. Roger Sprimont had
deliberately located the Australians on the same floor as Kockums’
design office and, allowing for national security arrangements,
in such close proximity the liaison team and the designers met
constantly and inevitably became close. Tore Svensson, one of
Kockums’ structural designers, saw the relationship as formal
but friendly, with the Australians guiding Kockums through the
requirements while familiarising themselves with the Swedish pro-
posals. Sprimont thought that Kockums gave the Australians more
access to the company’s practice and technology than their com-
patrio
ts received at IKL/HDW.
T H E P R O J E C T D E F I N I T I O N S T U D Y 1 9 8 5 – 8 6
93
Greg Stuart recalls that:
The team in Malm ö saw their role as ensuring the required
ship’s characteristics were met as closely as possible and to
fully understand the Swedish design philosophy, as my
previous work with the USN (Naval Sea Systems Command)
had highlighted how two designers could produce designs to
the same requirement but actually produce two vessels of
very different capability. I learnt, while working within the
USN, that a specification has to clearly state the conditions
under which a requirement is to be tested as well as stating
the requirement. Without these two being stated together a
requirement becomes very subjective. This later proved true
regarding the Collins’ noise signature.7
The general perception of those involved was that the liaison team
with IKL/HDW was less effective in guiding the designers to meet
the Australian requirements. The liaison team in Germany was
housed in separate buildings from the designers and they never
became close to the IKL and HDW staff. Later, after Australian
Maritime Systems had lost the competition, John White observed
that there seemed to have been bias in the operation of the liaison
teams, with that in Sweden more closely involved than the team at
IKL/HDW. However, for their part, the German members of the
consortium felt that relations with the liaison team assigned to
them were efficient and useful, although they now think perhaps
not as harmonious as in Sweden.8
While the project teams were packing their bags and head-
ing to Europe or America late in 1985, Oscar Hughes and John
Batten at the project office were developing a new management
control system to compare the costs of every element of the
different proposals under study. When in charge of naval pro-
duction, Hughes had concluded that contractual measures for
evaluating shipbuilders’ progress and authorising payment were
unsatisfactory. The system of ‘milestone’ payments used on
projects in the 1970s and early 1980s proved difficult to moni-
tor as it was hard to check that everything necessary had been
done for each milestone. For example, if the milestone was that
the funnels were to be completed, there had to be a way to check
that all the necessary piping and fittings had also been completed,
94
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
otherwise the contractor could just put up a funnel and claim
payment.
Hughes combined the American navy system of dividing
the shipbuilding process into thousands of small blocks with a
BHP system of contracts and sub-contracts, to develop the con-
tract monitoring and control system (CMACS). In Hughes’ view
CMACS gave the Commonwealth great control and visibility of
payments; it was used during the project definition study as the
basis for making financial comparisons between the German and
Swedish bids, and again during the construction phase as the basis
for making payments.
It was five months into the definition study before CMACS was
ready, but Hughes mandated that responses should be in the new
format. The contractors had already started cost and schedule
calculations on the basis of the American CS2 system and, not
surprisingly, were reluctant to change. But Hughes insisted and
CMACS was in the first draft of the contract issued in January
1986.9 The companies adopted it grudgingly – Hans Saeger of
HDW thought CMACS ‘a perfect intellectual task but far too
complicated for commercial use’. Nevertheless, during the post-
definition study negotiations, he was able to use CMACS data
to respond quite easily to the Commonwealth’s calls for changed
responses. He did, however, think that it had an inherent fault in
that ‘money flowed regardless of what was achieved’.10 From the
project’s point of view, Oscar Hughes found that ‘CMACS gave
us visibility and control particularly when it came to payments
for work done’, and he noted that ‘ASC embraced CMACS much
more openly and willingly than AMS’.11
In March 1986 the four companies met with the project team
in Canberra to discuss progress and future directions at a ‘prelim-
inary design baseline review’. Rockwell came fully prepared with
company executives and public relations team. They brought a
balsa wood and polystyrene mock-up of both submarine control
rooms equipped with the Rockwell combat system. The Australian
participants were convinced that Rockwell was on the right track.
In contrast, the Signaal team had little impact, with most Aus-
tralians feeling that Ron Dicker played a virtual lone hand.
At the review the Australian submariners were entranced
by Juergen Ritterhoff’s encyclopaedic knowledge of submarine
design, considering Kockums’ approach somewhat stiff and
T H E P R O J E C T D E F I N I T I O N S T U D Y 1 9 8 5 – 8 6
95
formal. However, Kockums provided a glimpse of the future when
they programmed a 3-D computer model of the Type 471 that
enabled the viewer to ‘walk’ through the submarine.
It was at this meeting that the German consortium made judg-
ments that effectively determined the result of the competition.
Ritterhoff was in Australia in the 1970s with Krupp Atlas, pro-
viding sonars for the Oberon upgrade, and he knew that the Aus-
tralian navy’s technical experts were deeply attached to British
submarine design. Consequently, he read the ship’s characteris-
tics for the new submarine as a requirement that fitted the British
approach, and was confident that IKL could exceed all criteria.
In his view the requirements represented an average, not overly
ambitious approach. The company’s best-selling Type 209 sub-
marine would not be adequate, but Ritterhoff was confident that
IKL could produce a design fully meeting the Australian require-
ment without having to ‘push the boundaries’ with experimental
ideas.12
At the preliminary design review AMS was anxious to seek
assurances that its design met the navy’s ideas of the ship’s char-
acteristics. If the navy agreed, it would then supplement its offer
with the possible inclusion of air-independent propulsion based
on fuel cell technology and the incorporation of other developing
technology.
However, the principal reason AMS wanted the assurance was
that it intended to freeze the performance levels of the design
at the point it thought the Australian performance expectations
would be met, and concentrate on winning the contract on price.
The submarine project was always aware that a cost blow-out
could kill the project and continually emphasised the importance
of keeping the price down. Rod Fayle had informed companies
they could offer reduced performance if forgoing that last degree
could significantly reduce cost
s. He knew that German subma-
rine designers were particularly good at identifying such ‘cost
drivers’.13
The Germans left the review convinced that the navy had
encouraged it to produce a ‘submarine as small as possible, consis-
tent with the requirements’.14 The navy, however, was not going
to give up the option of achieving the design best suited to meeting
operational requirements, nor of accepting unexpected technical
excellence.15 Neither would the Commonwealth, at so early a
96
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
stage in a competitive project, allow categorical assurances to be
given on the validity of the AMS contention that its proposals met
the ship’s characteristics.16
Oscar Hughes recalls emphasising at the review that he was
looking for the best possible submarine for the money and making
it clear to all the companies that the project team was looking for
maximum performance. If the designers felt they could exceed the
requirement they could ‘auction capability against cost’. Yet he
had the feeling that the Germans were not really listening and
that ‘the Germans thought they had a winner with their escape
sphere and were taken aback when this was ruled out’.17 Hughes
believes that the Germans did not take the Australian requirements
seriously, whereas the Swedes were keen on doing what Australia
wanted and were more flexible in responding.18
The consequence, in Hans Ohff’s view, was that the design
developed by IKL/HDW during the definition study was far too
conservative. It is hard not to conclude that the Germans left the
preliminary design review poorly advised and with an inadequate
knowledge of the Australian purchasing process. Whatever the
reason, after the preliminary design review the two contending
submarines took different paths. The AMS boat became shorter
and lighter while the Kockums design increased in weight, a few
metres in length and from 7.5 to 7.8 metres in diameter.19
The review also marked a turning point for the project team as
many of the original members departed. After four and a half years
as director Graham White was posted off the project. Although
Rod Fayle retained his position supervising operational require-
ments until 1987, he was at this time with the liaison team in
Holland. With other members of the original team leaving, there
