by Peter Yule
navy, but Robertson felt that once he had accepted a job out of the
navy, it would not be honest to stand up on the platform during
the launch ceremony.
Robertson was succeeded as commanding officer of Collins by
Peter Sinclair, another experienced Oberon captain. After attend-
ing the launch, he and the crew spent most of the next year training
in the simulators in Western Australia and Sydney while Collins
was prepared for the contractor’s sea trials. Sinclair recalls:
The first launch was purely political. Most of the submarine’s
systems were incomplete and nearly all hull valves were
blanked. ASC needed another year to complete fit out and it
was during this period that the real training for the crew
started and they were able to get involved in systems set to
work programs.
Many of the problems raised in the preceding years were
fixed after the submarine’s launch. Collins was taken out of
210
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
the water a week after launch and her fit out completed. This
gave the crew the opportunity to be heavily involved in the
systems set to work program and the extensive licensing
process. System task books were produced that ensured each
member of the crew was competent in their specific ‘part of
ship’. Notably in the 12 months before the submarine was
launched for the second time there were no resignations and
the crew went to sea for the first time with a high team
spirit.
The trials process was an exhaustive and comprehensive process
to ensure that each new submarine was safe and met the contract
requirements before its delivery to the navy. For the first of class
in particular it was also the time to discover and fix flaws in the
design and construction of the submarines. In retrospect it seems
extraordinary that the general expectation appears to have been
that Collins would roll off the production line perfect in every
respect, when every other nation that has ever built submarines
has had problems with the first submarine of every new class.
Oscar Hughes noted, in his ‘haul down’ report presented to the
Chief of Navy at the time of his retirement in 1993, that it was
probable that the submarines ‘would experience a range of defects
and shortfalls in performance during the trials phase’ and ‘that this
outcome was to be expected for any new major capability, partic-
ularly one based on the development of a new design’. However,
this view had not permeated the project and the project’s reports
did not mention the possibility of ‘first of class’ problems until
1995.
There were strong arguments against building a prototype for
a class of only six submarines – how do you retain the skilled
workforce while a prototype is being evaluated? – but the relent-
less demands of the production schedule, itself determined by the
schedule for taking the Oberons out of service, made it difficult
even to slow the trials process to enable the results to be evaluated
and applied to the later submarines.4
The trials were managed by a trials board consisting of the sub-
marine’s commanding officer, the ASC trials manager, the project
director’s representative and the head of the sea training group.
This was a contractual requirement, and was established to care-
fully control the trials process and ensure the highest levels of
T H E T R I A L S O F C O L L I N S
211
safety. Each of the members had power of veto and could stop the
submarine from sailing for trials or stop the trials program and
bring the boat home.
The trials process involved achieving a staged sequence of
licences, with the first allowing the submarine to run on the sur-
face, followed by shallow static dives and gradually progressing
to unaccompanied deep dives. The delivery of a submarine to the
navy could not take place until the trials were completed and all
licences achieved. The licensing program was developed so that
the competence of the crew and the material state of the submarine
matched the increasing levels of hazard in the trials program.
The licensing program also included licensing the crew by
putting them through countless emergency drills in the simulator
and on the submarine, alongside and at sea. Peter Sinclair recalls:
This was an exhausting period: constantly practising
emergency drills and standard operating procedures, floods,
fires, hydraulic bursts. It often meant changing procedures on
the spot. With the excellent support of Mike Gallagher and
his sea training group and the professionalism of the crew we
became more confident and competent each day . . . What we
weren’t prepared for was the number of intermediate or
docking level defects that occurred in those early months.
Our inability to fix these problems stemmed from a lack of
in-depth training for our engineers. In hindsight though I
must give praise to them, because they handled every
circumstance with utter professionalism.
Originally Collins’ trials were due to begin early in 1994, but this
was too optimistic and trials did not begin until 31 October 1994.
The combat system was the source of greatest frustration. The
submarine could not go to sea without at least a rudimentary com-
bat system, as the radar, sonars and periscopes are required for safe
navigation. During 1994 there were constant disputes between
Rockwell and the project on the minimum combat system perfor-
mance needed to allow the sea trials to begin. The major problem
was that the whole system was extremely slow, constantly crashed
(taking up to eight hours to re-boot) and could not perform many
tasks. Another serious difficulty at this stage was integrating the
French-made sonars with the Australian tactical data handling
system.
212
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
Ironically, one of the main reasons why the project office had
refused to countenance defaulting Rockwell in September 1993
was that the combat system would be needed for Collins’ sea trials,
yet the performance in mid-1994 was so poor that Collins actually
went to sea and did most of its trials using little of Rockwell’s
system. In its place they used ‘stand alone’ equipment to run the
navigation, with specific software for the navigation functions and
a stand-alone navigation display to get Collins to sea. For example, the combat system was unable to process data from the sonars,
so the project had to pay the sonar sub-contractor, Thomson, to
install a back-up sonar with a separate sonar display.5
On 8 August 1994 ‘basin trials’ began and the main motor was
turned for the first time. During September Collins was fuelled and
stored and ‘she proceeded to sea under her own power at 10.00
on Monday 31st of October to surface sea trials in the Gulf of
St Vincent’.6 Mike Gallagher recalls that: �
��Peter Sinclair disap-
peared at a great rate of knots . . . there was something of a sense
of sheer delight in actually getting the boat to sea.’ But at the same
time Marcos Alfonso and his engineering team were beginning a
long struggle with the diesel engines. By the time Collins passed
the Young Endeavour (which was coming into Port Adelaide) two
of the engines had broken down.
Collins’ first static dive was on 9 November and the submarine
remained submerged for 12 hours carrying out trim and inclin-
ing trials. Trials continued throughout November, after which the
project director reported that:
Despite appalling weather Collins performed extremely well
in the first phase contractor sea trials in November exceeding
the contracted requirements in many areas. Dive trials will
commence early in 1995 and Collins remains on schedule for
delivery and commissioning in November 1995.
However, there was a note of warning:
The combat system software remains the area of greatest
concern. Software integration activities for the
commencement of dive trials are progressing.
Following a break over summer, when Collins was taken out of
the water for ‘completion of outstanding work and some minor
repair work to the hull’, surface sea trials and static dive trials
T H E T R I A L S O F C O L L I N S
213
resumed in early 1995. However, further dive trials were delayed
when the next release of the combat system software ‘failed its
performance verification in a number of key areas’, and in an omi-
nous note the project’s quarterly report noted that: ‘The number
of items on the design and defect list has reached the stage where
it would be an advantage in lifting Collins out of the water to clear them.’
Learning from the early sea trials of Collins, the project office
saw the need ‘to review aspects of the design’, although at this
stage the problems cited were relatively minor such as ‘hydrogen
clearance during gas charging of the batteries’, the circuit breakers
in the propulsion motor and hydraulic system sealing.7
By early June 1995 it was agreed that the combat system had
achieved the minimum level of performance required for Collins
to begin its under-way dive trials. Due to the shallow water in
the Gulf of St Vincent off Adelaide, the diving trials were held
in the deeper water of the Southern Ocean near Port Lincoln,
although the open ocean was not the ideal location for trials. The
first dive took place on 9 June. The project office’s report said that:
‘After carrying out post diving and system checks at periscope
depth, Collins proceeded to 60 metres to conduct further trials.
The scheduled trials met or exceeded the contracted requirement
in every area.’
However, the following day the weather was rough and
Collins’ anchor cable parted. Peter Sinclair and Graham White
wanted to stay at sea and continue the trials, but Paul Greenfield,
the project director’s representative on the trials board, exercised
his veto, forcing the submarine to return to ASC to replace the
anchor. One of the problems that had become apparent during
the trials was that the propulsion motor would trip out and stop
when doing manoeuvres, and this occurred just as Collins was
approaching the wharf. With no power, the submarine drifted
helplessly into ASC’s ship lift at two or three knots, damaging the
sonar dome in the bow, before the emergency propulsion came
on and the boat shot backwards, damaging the trials support ship
HMAS Protector.
The propulsion system is software driven, and it was eventu-
ally discovered that the problem was due to a programming error
in the propulsion switchboard. The sub-contractors, Jeumont-
Schneider, accepted responsibility for the problem and it was
214
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
quickly remedied. The damage to Collins was soon repaired, with
the sonar dome being replaced at a cost of about $250 000.8
There were two significant issues raised by this incident. The
first was the reaction of ASC and the project team. Collins had
significant damage to the glass fibre sonar dome, but the amount
of damage was publicly denied – even though it was quite visible
from across the river. This naturally made the media unwilling
to accept that ASC and the project office were telling the truth
when they denied later stories. The second issue was that both
ASC and the project tended to blame the crews for problems like
the propulsion system failures.9 This left the crew in an invidious
position arising from the nature of the trials. ASC and the project
office were trialling a first of class submarine, but as they were
using a navy crew, every incident had to be reported according to
naval regulations. The crew felt that ASC and the project team
used them like ‘crash test dummies’, while the navy required them
to operate under strict navy rules.
Marcos Alfonso recalls that the trials period was
indeed ‘tough days’ as the crew was only operating based on
the training provided by ASC and the project office, who
were then quick to raise the operator error issue when in fact
it was either faulty equipment, poor training or lack of
training. Equipment did break down due to crew not
operating correctly, however this was due to the crew never
being taught how to operate it correctly and in most cases
training themselves with the assistance of ASC production
staff who through their set to work expertise knew how to
operate the systems.
During 1995 there was a distinct change in the tone of the
project office’s reports. The expectation that the submarines
would be perfect was fast fading and the project was taking the
more realistic attitude that faults would be found in a first of
class submarine. Thus the June report noted that, ‘A number of
equipment and/or system deficiencies have been revealed during
sea trials but no more than expected for a first of class trial.’
The crew members also had begun to accept that trialling a first
of class submarine was not like driving off in a new car, although
their perspective was rather different to that of ASC or the project
office. They found that there were many minor mechanical faults
T H E T R I A L S O F C O L L I N S
215
and they rarely went to sea without something breaking down.
However, it was not the faults themselves that most concerned
them, rather the fact that they found it hard to diagnose and fix
the problems. On the Oberons the crews prided themselves on
knowing what spares to take and being able to fix anything that
broke, but with the new submarines the allocation of spares was
based on the idea that few would be needed because there would
be no breakdowns. Similarly, the manuals were simplistic, as it
was thought they would be rarely needed. Marcos Alfonso recalls
being surprised that when ASC technicians were on board they
were often able to fix faults that his team was unable even to
diagnose. It took him some time to realise that the ASC techni-
cians were not using the manuals provided for the crews but the
source data from the suppliers, and would arrive on the subma-
rine with support and test equipment or diagnostic tools that were
not available to the crew on board. When he asked them to diag-
nose and fix the problems with the tools and reference materials
carried on board, they were no more successful than the crew
had been. In his view the worst problems for the crew were the
training and documentation rather than the actual faults with the
submarine.
While it became the project’s approach to problems with
Collins to say that they were normal for the first of class, it is
clear that neither the project nor ASC had planned for dealing
with these problems. In June 1995, for example, the project report
said:
A number of equipment and/or system deficiencies have been
revealed during sea trials but no more than expected for a
first of class trial. These are being progressively rectified but
are diverting effort from the follow-on submarines.
If the first of class ‘deficiencies’ had been expected, resources
should have been allowed for this in the project planning without
diverting effort from later submarines.
The sea trials for Collins continued throughout 1995 and into
1996, with the date for delivery being regularly delayed, primar-
ily because the combat system could not achieve the minimum
levels agreed for acceptance, but also because the submarine was
plagued by frequent mechanical failures. Significantly, the prob-
lems appeared to worsen rather than improve during this period,
216
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
even though the submarine spent many months undergoing repairs
and modifications at ASC. The list of defects was long, but the
recurring problems centred on the diesel engines, the stern shaft
seals, the communications mast, the periscopes and the hydraulic
couplings. Noise only emerged as an issue fairly late in the tri-
als program, both because of problems with the noise range and
because noise was not regarded as a potentially serious problem
before the submarine began its high speed trials. But overrid-