Brent Clark, now former CEO of Naval Group Australia, has joined the debate ^[1] on the pump jet propulsion system. He follows Gerard Autret and Sean Costello ^[2], Naval Group’s Herve Guillou ^[3], Rear Admiral Greg Sammut ^[4], Andrew Davies ^[5] and myself making comments on the pump jet on these pages, which has since been further taken up in Senate ^[6] committees.

Clark comes to the pump jet’s defence with two new arguments. The first is that some of the research papers that the pump jet’s critics rely on are old. The second is that the pump jet would not be a suitable choice for a small submarine, but the large size of Australia’s future submarine now makes the pump jet viable.

In my earlier post, I pointed out that there were fundamental reasons why pump jets tended to have much lower efficiency at lower speeds ^[7], and how this an inevitable consequence of the design that allows them to have better acoustic performance at higher speeds.

An important conclusion is that it’s the speed at which a vessel operates—not the size of the vessel—that is crucial in choosing the propulsion system. Moreover, I found that the efficiency impact was probably quite significant. A switch back to a propeller could very plausibly increase dived range and endurance by about 60%, which could amount to additional days, or hundreds of nautical miles. Both in the post and in a longer research paper ^[8], I made reference to papers from the 1960s and 70s like this one ^[9].

Clark says that some critics have used old studies to justify their argument. In my research, I have also referenced plenty of modern papers (see here ^[10] and here ^[11]). But most striking in this debate is the ignorance of how the science of marine propulsion and turbomachinery (pumps) has progressed in recent decades. Professional engineers and scientists don’t lightly disregard works based on their age. Many of the best older works provide the most foundational, enduring principles on which subsequent works depend. In nuclear physics, Einstein still matters.

In marine propulsion and turbomachinery, momentum theory and Euler’s equations (both centuries old) provide the backbone of the science, and enduring bounds on what is possible. The likes of GF Wislicenus ^[12] were responsible for extensively applying these principles in the context of pump‑jet and torpedo or submarine propulsion in the 1960s, 70s and 80s. To say they ‘wrote the book’ on these topics is only a slight exaggeration, as can be seen in the references of modern texts ^[13] (large pdf) on relevant topics.

Today most scientific papers no longer contest what has long-since been established and accepted. A perfect example would be a 2008 study ^[14] into using a fully-submerged waterjet for an anti-submarine frigate in order to reduce the acoustic signature of the ship (just as would be intended for a submarine). It summarises the obvious in one sentence: ‘It is known that waterjet technology is beneficial at higher speeds and not as efficient at lower speeds.’ The results of the research confirm as much, and never attempt to prove otherwise.

The study concludes that the ship designed with jets would require an additional 250 tonnes of fuel. The operating profile of that ship required 90% of its time spend at 10kt or above, where the efficiency penalty was significantly more moderate. A submarine would spend much more of its time below 10kt, where the penalty is more extreme. The consequences for fuel and battery requirements would probably be even more substantial.

Brent Clark’s second defence of the pump jet neatly connects with this concept. He concedes that a pump jet wouldn’t be viable for a ‘1,500‑tonne coastal submarine’, and says that the uniquely large size of the submarine we’re building is what makes the pump jet viable, for the first time ever, on a conventional submarine.

This is a crucial admission because there is no credible argument that pump jets don’t work well at a smaller scale. The original science was pioneered very much in the context of torpedoes as well as submarines, and they’re still used extensively on both today despite the gulf between their relative sizes. In the early foundational literature ^[15], the same theory, examples, diagrams and equations for pump jets ‘on the after end of a body of revolution’ were used to refer to submarine and torpedo applications interchangeably.

Rather, Clark is pointing out that it’s harder to fit hundreds of additional tonnes of fuel (and batteries) in a smaller submarine. Or equivalently, since making a submarine larger inherently improves its range, it’s hard to impose an efficiency handicap on a small submarine without growing it considerably larger to meet the same range and endurance specification.

No one from Naval Group has ever denied the consequences of this efficiency penalty. Instead, it was acknowledged ^[16] some time ago by Jean-Michel Billig when he said in October 2017 that the submarine ‘may end up with conventional propellers as well as air independent propulsion, which helps increase underwater endurance’. Now Clark has confirmed that the efficiency penalty will not be overcome by any ingenious feat of engineering, nor is it so small as to not matter.

Rather, the viability of the pump jet relies on (or necessitates) the vast scale of the boat to accommodate the extra fuel and batteries. This might explain why the current designs, at around 5,000 tonnes submerged, are as large as a Collins class (3,400 tonnes) and Sweden’s Gotland Class (1,600 tonnes) combined. Surprisingly, the design only needs to offer similar or equivalent range to the Collins according to the Program Chief ^[4], or the 2016 Defence White Paper ^[17] (para 4.27).

This is a substantial concern for the stealth of Australia’s future submarine. Bigger isn’t better when playing hide-and-seek. A larger submarine reflects more signal from active sonar sources, making it more susceptible to dipping sonars and to any other active sensor deployed from aircraft or surface ships, or even on the ocean floor.

Nuclear submarines enjoy the luxury of sustained high speed, which allows them to escape a threat area more quickly than most ships could follow. This ability to ‘run’ takes a lot of pressure off their ability to ‘hide’. But an overly large conventional submarine fighting far from home, lacking air cover, and without the speed needed to escape a surface warship will actually be far more vulnerable than a smaller one.

How such a submarine will be regionally superior in crowded Asian waters is extremely unclear.