by Peter Yule
should not cavitate’. But the boat did cavitate . . . I had one
of my famous blow-ups with Hans who said: ‘That’s only
because you don’t know how to handle the boat.’ And I said:
‘I don’t care how quiet the boat is [at slow speeds], but if it’s
detected it has to use speed to evade and this boat’s going to
cavitate and that’s death to submariners.’
It is difficult to design a submarine propeller that will be powerful,
efficient and quiet, with the power required for speed and accel-
eration always a likely cause of cavitation. Thus the Australian
propeller was seven-bladed for power and efficiency, while a five-
bladed propeller might have been quieter.16 Cavitation was not a
problem with Swedish propellers in the 1970s and 1980s, but the
propeller for the Australian submarine had to drive a submarine
that was much larger than any Swedish submarine. Greg Stuart
notes that the final propeller design was not selected until well
into the design process. The boat had become bigger and longer
but the motor had not changed, so the propeller needed to be
changed, but they did not go back and do the testing again.
Both Kockums and ASC blame manufacturing faults rather
than the design for the problems with the propellers. According
to Hans Peder Loid, SSPA’s testing of the propeller showed no
cavitation problems, and he is convinced that any problems that
later developed must have been caused by faulty manufacture or
changes to the submarine design. The faulty manufacture argu-
ment is supported by the fact that each submarine had different
cavitation characteristics, which would not have been the case
if the design was the problem. Hans Ohff is insistent that ‘there
was nothing wrong with the design of the propellers’ and argues
that the ‘problems were in manufacture’, both in the castings in
England and the hand grinding in Western Australia. After the
problems with grinding were identified, ASC reworked the pro-
pellers on their five-axis boring and milling machine, and the end
product was much more successful than hand-finished work. In
the end, however, only one or two propellers could be improved
because too much material had already been removed and on at
least one propeller the pitch had been incorrectly cast.
While admitting to manufacturing faults in the propellers,
ASC and Kockums felt that the problems with cavitation were
‘ T H E Y W E R E P R O B L E M S W E D I D N ’ T E X P E C T ’
231
exacerbated by the way the new submarines were handled. Ron
Dicker, who had had experience of cavitation problems with
Dutch submarines, thinks that the links between the operators,
the project and the builder were weak and the operators were not
fully aware of the techniques required to operate the new pro-
pellers. The authors have found that submariners react strongly
against suggestions that any of the problems with the submarines
were due to their inexperience, but surely it would not be surpris-
ing if it took some time to learn how best to operate a new class of
submarine?17
In 1998 a further problem appeared when some of the pro-
pellers began to develop fatigue cracks. This was seemingly the
result of the ‘root’ of the propeller being too thin and manufac-
turing techniques that were inadequate when working with Sonos-
ton. The resolution of this issue is a story for the next era of the
project.
During Collins’ trials in May and June of 1996 another unex-
pected issue emerged when it was found that the periscopes tended
to vibrate in certain conditions, and had dangerous optical char-
acteristics including difficulty focusing, duplication and what is
called a ‘double dove’ effect (caused by internal reflections).
When the attack periscope was raised as the submarine was
coming to periscope depth it would begin to vibrate so much that
the vibrations would transmit through the hull and things would
shake all through the boat. Paul Greenfield recalls Peter Sinclair
saying that using Collins’ periscope was ‘like being shaken by the
hand of God’. The vibration was caused because the periscope
was not streamlined, leading to turbulence and water resistance.
During 1997 this issue was a cause of bitter dispute between ASC
and the navy. John Dikkenberg recalls being furious when ASC
told him that it was not its responsibility because the contract did
not say that the periscope should not vibrate. Dikkenberg took
the view that a submarine builder should know that this would be
unacceptable. ASC’s perspective was that the periscopes are part
of the combat system and the problem was therefore Rockwell’s
responsibility. Nonetheless, early in the project ASC had queried
Rockwell and periscope maker Barr & Stroud on vibration risk
but was assured it would not be a problem. As nobody wanted
to pay for a fairing (and there were space constraints as well)
Kockums decided to put the top bearing as high as possible to
232
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
minimise potential problems, and Rockwell and Barr & Stroud
signed off on the design.18
The vibration problem was eventually fixed quite simply after a
defence scientist suggested putting spiral wraps on the periscopes
to alter the water flow, but the issue was a revealing example of the
poisonous relationship developing between ASC and the navy.19
The optical problems were not so easily resolved. Mike
Gallagher of Farncomb recalls that both periscopes were diffi-
cult to focus when changing magnification; the search periscope
had a ‘double-dove’ effect, with a grey band appearing across
the field of vision; the attack periscope had duplication; and they
could sometimes see ‘two suns’ through the periscope even when
the elevation was zero. Barr & Stroud sent a Scottish technician
to investigate the problems. When he looked through the search
periscope the sun nearly cooked his eyes and he said, ‘You bas-
tards, you’ve set me up’, thinking that the crew had deliberately
aimed the periscope at the sun – but when he checked he saw the
elevation was zero.
Eoin Asker saw the problems with the periscopes as coming
largely from the navy’s excessively ambitious requirements. This
view is supported by Ron Dicker, who recalls that:
At the first or second design review, Barr & Stroud presented
their initial design for the search periscope. At the top of the
mast, going down the tube, they had the infra-red sensor then
the optical sensor then the single-pulse radar. The navy said:
‘This is not what we want – we want the optical view first
when the periscope goes up.’ Barr & Stroud said this was
silly and they should have the infra-red first, but the navy
insisted so the infra-red had to be below the optical. This led
to big technical problems as the optical signals had to be
bounced around the infra-r
ed sensor . . . This was the basic
cause of the periscope’s optical problems.
As with the cracking propellers, the optical issues with the
periscopes were not fully resolved until the next phase of the
project.
During the trials of Collins and Farncomb the project office listed all the current technical problems in its quarterly reports. A
large number of issues were listed, but most of them appeared
only once or twice, indicating they were quickly resolved.
‘ T H E Y W E R E P R O B L E M S W E D I D N ’ T E X P E C T ’
233
Further, a majority of the issues that appeared repeatedly in the
early months gradually disappeared. For all the criticism of ASC,
the project office and the navy and the difficulties they had in
working together, most of the submarines’ problems were worked
through and fixed.
A good example of this was the successful resolution of the
leaking shaft seals that plagued Collins and Farncomb. The specification for the shaft seals allowed for a leak of up to 10 litres an
hour, but in the early trials hundreds of litres were coming into the
boats. Paul Greenfield recalls that ASC’s reaction was ‘She’ll be
right – just get the sailors to tighten the glands manually’, but this
meant that someone would have had to tighten the glands with a
spanner when descending and loosen them when ascending, which
went against the small crew philosophy of the new submarines.
On Collins the crew worked out that if they went astern at
100 revolutions it would throw the shaft seal back into align-
ment and keep the water under control. However, on one occasion
during deep diving trials this did not work. Peter Sinclair recalls
that he asked the engineers to check the water levels and got a
very muddled answer, so he told Marcos Alfonso to go aft and
check the problem. Alfonso soon called Sinclair on the command
line and said: ‘Captain, you need to make your depth shallow as
quickly as possible while keeping the submarine level.’ The sub-
marine was taking on 1000 litres of water a minute and the aft
main bilge pump was barely keeping up. However, Alfonso did
not pipe ‘flood’, as the response to this would have been to blow
ballast and accelerate towards the surface at a steep angle. A sailor,
Gary ‘Chook’ Fowler, was underneath the propeller shaft trying
to identify the leak and tighten the gland, and if the submarine
had gone to the surface at a bow-up angle he could have drowned.
Sinclair was able to bring the submarine to periscope depth while
maintaining an even keel and they were then able to throw the
seal back into alignment with the ‘astern trick’.
In the adversarial mood that developed in the mid-1990s ASC
claimed that the leak specification was not achievable, and the
situation was complicated because the original German supplier
of the shaft seals had gone out of business.20 A British com-
pany that produced similar seals for Royal Navy submarines took
over the sub-contract and fitted modified seals to Collins and
Farncomb which alleviated the problem. In December 1996 the
234
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
project office reported that the leak problems in Collins and Farncomb were not finally resolved, but the modified seals allowed sea
trials to continue while they waited for re-designed seals. Signif-
icantly, the report in December 1996 noted that: ‘Helpful advice
has been provided by US Navy sources where similar seals and
problems have been encountered.’ This is possibly the first men-
tion in the whole history of the project of assistance received from
the American navy. In December 1997 the project office reported
that a re-designed shaft seal from a new supplier had been fitted
on Farncomb and successfully tested to deep diving depth.
Many defects found in Collins and Farncomb were fixed by
ASC and its sub-contractors, and the lessons learnt were able to be
applied to the later submarines. In March 1998 the project office
recorded: ‘The first review of Waller’s form TI 338 A [the formal
record of shortcomings on delivery] indicates a large number of
defects overall but with significantly fewer causing concern than
for the first two submarines.’ For each successive submarine the
list was shorter and the defects less serious.
The early submarines suffered from numerous mechanical and
technical defects. These would all have been seen as normal ‘first
of class’ issues if not for three things. Firstly, the breakdown in the
relationships between ASC, Kockums, the project office and the
navy meant that many simple problems were not simply solved
but became subject to bitter dispute. Secondly, the change of gov-
ernment after the federal election of March 1996 made the project
the subject of political controversy. Thirdly, and most importantly,
the combat system still did not work at anything like the level that
had been hoped for. If the combat system had worked to expecta-
tions the other problems with the submarines would have faded
into insignificance and the new submarine project would never
have been seen as anything but an outstanding success.
C H A P T E R 20
The role of Defence Science: noise
and diesels
In the early years of the new submarine project, scientists from
DSTO were deeply involved in several key areas, notably steel
and welding, the sonars and the development of anechoic tiles.
The second phase of DSTO involvement centred on resolving the
problems shown up during the trials of the early submarines.
When the results of the noise range testing of HMAS Collins
showed that the new submarine was noisier than expected and
there appeared to be propeller cavitation, generating more com-
plex noise characteristics, the project team called in the Ship Noise
and Vibration Group from DSTO to assess the situation and help
the navy argue its case against ASC and Kockums.1
This was the beginning of a concerted defence science effort
to understand and resolve the problems of the new submarines.
Many of DSTO’s divisions and experts were to be pulled into the
effort but it was the Aeronautical and Maritime Research Labo-
ratories (now Platform Systems Laboratory) in Melbourne under
Dr Bill Schofield2 that was central to research on the submarines,
and the Maritime Operations Division in Adelaide that took the
scientific lead on the combat system.
235
236
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 Ship Noise and Vibration Group was set up at DSTO’s
Maribyrnong Laboratory in 1989. Oscar Hughes had approached
his adviser on steel and welding, John Ritter, after a Friday
evening’s ‘backslapping wash-up meeting’ in Canberra to suggest
that DSTO should establish a group to research acoustic signa-
tures. The chief of the Materials Division, Maurice de Morton,
supported the idea and so a group of scientists studying the char-
<
br /> acteristics of machinery and hull noise in water came to join the
metallurgists and others in the division. Some came from inside
DSTO, but among those recruited from outside was Dr Chris Nor-
wood, who led the investigations into the acoustic characteristics
of the Collins submarines.
The propagation of sound under water is a complex matter.
Generally, sound moves four times faster under water than in air
and can travel vast distances. A sound will travel twice the distance
for an energy increase of only three decibels – an important fact,
as submarines are hunted through their noise signature. Sound
travels so well under water that the sea is a very noisy place. In
the past this favoured submarine operations because a competent
submarine designer would aim to produce radiated noise levels
approaching the ocean’s background level. With ever-improving
sonar software technology, this advantage has become harder to
sustain. Passive sonar can be ‘tuned’ to discriminate between more
variable natural background noise and those frequencies that are
typically produced by machinery. Consequently, submarines must
be ‘quiet’ but also have certain sound characteristics minimised
or eliminated.
For five years, Norwood’s acoustics group studied problems
with surface ships and had developed enough expertise to have
credibility when the project office approached it with concerns
that the Collins class had an inherent noise problem. A major
difficulty was that a solution would not be found simply by iden-
tifying the source of the noise being radiated. The critical issue was
how noise was transmitted into the water, as this was the path that
propagated the sound tones being emitted by the submarine.3 The
DSTO group simply had to persist with a series of experiments
aimed at successively excluding possible causes until a consensus
emerged about how the submarines were generating the unwanted
noise. The most obvious approach was to go to where the prob-
lem seemed to manifest itself and try to directly measure the forces
T H E R O L E O F D E F E N C E S C I E N C E : N O I S E A N D D I E S E L S
237
at play. The project arranged a simple means to test Norwood’s
hypothesis and this proved that the tones, thought by the designer
to be coming from elsewhere, in fact came from the propulsion
