SSN-AUKUS: a ‘mature’ design…just with many defining things not settled.

With SSN AUKUS construction to start before 2030 & Aust's 1st boat in 2032, it seems detailed design hasn't begun.

Written by

Marcus Hellyer

What is SSN-AUKUS? This is not a metaphysical question along the lines of what is the meaning of life, or even that other perplexing one, namely what is AUKUS? It’s a straightforward question about what the design of the nuclear-powered submarine that Australia and the United Kingdom intend to build and operate actually is.

Despite the gallons of ink spilt on all things related to AUKUS, there’s actually very little real information in the public domain about the design of the submarine, even in official government publications, for example in the booklet published by the three governments outlining the ‘optimal pathway’ last year.

And closely related to the question of what SSN-AUKUS is meant to be when it is a real submarine in service, there’s equally important question of what it is today. Answering that second question allows us to have a better understand of when an actual SSN-AUKUS submarine can reasonably be delivered. But there’s very little public information on that issue either.

AUKUS leaders have stressed the maturity of the design. Vice Admiral Jonathon Mead, Head of the Australian Submarine Agency has said that ‘SSN-AUKUS is actually quite mature in design, it’s about 70 per cent mature.’

That might suggest we’re nearly at the start of construction. But what does it mean? No term is more frequently misused in the defence acquisition space than ‘mature’, as yet another report into the ridiculous process that resulted in Australia’s selection of the Hunter-class frigate has recently confirmed.

Mead has also said that SSN-AUKUS will be a “bigger, better, faster and bolder”, evolution of Britain’s Astute-class submarines. The term evolution is obviously being used here to emphasise maturity and minimise scope of work required along with the resultant cost, schedule and risk. But evolution is an imprecise term; how much evolution—from amoeba to human or from one Galapagos Island finch into another?

Even a modest evolution from a mature design does not necessarily go well. The UK’s current Astute-class SSN was originally intended to be a modest improvement on the Trafalgar class, but it has ended up being a very different and much larger boat and the design and construction process went off the rails.

The absence of information on what the SSN-AUKUS design is and its maturity is regrettable for several reasons. The first is that it would be nice for Australian taxpayers to know what they are getting for their $268-368 billion (and for British taxpayers to know what they are getting for their money). The second is that without knowing more about the design, it’s difficult to assess whether the high-level schedule for AUKUS Pillar 1 is viable or not.

Recall that the UK’s construction of SSN-AUKUS is meant to start this decade, which is at most five years away, with Australian construction to start several years later around 2032. We don’t have unbounded amounts of time, particularly with the Collins-class submarine already showing its age. If SSN-AUKUS is not going to hit its schedule, we need to know that as soon as possible.

The design itself

Let’s recap what we do and don’t know about the final design. Numerous commentators have suggested that one reason Australia is acquiring a British design is that British SSNs are smaller (and potentially cheaper) than US ones. But that is not the case with SSN-AUKUS.

Mead stated in May this year that SSN-AUKUS will be 10,000 tonnes. That makes it considerably larger than the first four tranches of the US Navy’s Virginia class SSN (Block I to IV) at 7,900 tonnes and similar in size to the Block V Virginia which has grown to 10,200 tonnes through the insertion of a module containing four vertical launch systems carrying a total of 28 cruise missiles. So SSN-AUKUS will be a very large attack submarine. The French Suffren-class is around 5,000 tonnes, for example.

There’re three key parameters of size: weight/displacement, length and beam/width. If you know two, you can work out the third. While we have a broad indication of tonnage, I’m not aware of any definitive statements from either the UK or Australian governments on length or width/beam. The Australian Submarine Agency’s Senate estimates brief from October 2022 is silent on the submarine’s dimensions. So we only have one of the three.

Beam is crucial; you can lengthen a submarine by adding plugs (as was done with Sweden’s existing submarines in a mid-life upgrade, and with the design of the Block V Virginia), but you can’t alter its beam once construction starts. Moreover, the more you ‘evolve’ a design by increasing the beam, the more design work you have to do. That was one of the reasons the Australian Department of Defence decided not to pursue an ‘evolved’ Collins for its future submarine requirements; to meet Defence’s requirements the Collins’ beam would need to be increased and that, in Defence’s assessment, would result in as much design work as a new design.

We should note that some informed commentators had assumed that SSN-AUKUS would have the same diameter as the UK’s Dreadnought SSBN design that is currently under construction. SSBNs have larger diameters than SSNs to accommodate the ICBM  launch tubes (Dreadnought is 12.8m compared to Astute’s 11.3m—and Astute is quite a fat SSN). One British diplomat told me at a dinner that SSN-AUKUS would essentially be a ‘cut down’ version of Dreadnought with the ICBM tubes removed.

However, Mead’s statement that SSN-AUKUS is ‘mature’ and an ‘evolved’ Astute could be taken to mean it will retain Astute’s diameter. A 10,000 tonne submarine with an 11.3m diameter would be a very long SSN. We won’t get into discussion of the advantages and disadvantages of different length to diameter ratios, (Section 2.3.3 of Edward Dawson’s thesis provides a useful overview of the issues) other than to say this would result in a high length to diameter ratio for an SSN.

But the bottom line is, we don’t know either the length or beam. What we can say is that if the beam has not been finalised, the design is by no means mature. And if the beam has been finalised but has changed from the Astute’s, we are talking an evolutionary leap, not a modest step.

In addition to dimensions, we also lack clear government statements on key components, leading commentators to make assumptions. For example, the UK and Australian governments have said that SSN-AUKUS will be powered by a reactor made by Rolls-Royce, but they haven’t stated which reactor. Considering that the PWR-2 that powers the Royal Navy’s Astute-class is no longer being manufactured, that would only leave the PWR-3 that Rolls-Royce is developing to power the Royal Navy’s Dreadnought class. That’s a reasonable assumption, but there is no confirmation from the governments.

However, at around 17,200 tonnes the Dreadnought is a much bigger submarine the SSN-AUKUS. So that’s a lot of reactor for SSN-AUKUS. Will SSN-AUKUS use the same version of PWR-3 or a modified version of it? If so, that would require redesign effort. Moreover, government statements have indicated that US technologies will be used in the propulsion system; does that mean further design work is required for PWR-3 or for a modified or evolved version? As we’ll discuss below, the UK government awarded Roll-Royce a contract in 2021 for early design work on the UK’s future attack submarine. What was that for?

That’s not the only key component where we don’t have definitive information. It seems to be the case that SSN-AUKUS will also have vertical launch cells using the US system employed on the US’ Virginia class that has seven tubes in a launcher. That’s potentially one factor driving the size of the submarine, but there’s no clear statements from any government on how many tubes or launchers. Again, if that hasn’t been settled, the design isn’t ‘mature’.

In fact, the selection of many components still appears to be outstanding. The governments have stated ‘the Virginia class SSNs will have a high degree of commonality with the SSN-AUKUS submarines, including sharing elements of the propulsion plant, combat system and weapons. This commonality in design, components and performance will help with the transition to SSN-AUKUS, while enhancing interoperability among the AUKUS partners.’ There may well be benefits to commonality, but it has not been stated what those common elements are.

There’s little information on what components will be shared between the US and UK designs. But the more US elements are included in the SSN-AUKUS, the more that will require replacing components that were used in the Astute, lengthening the evolutionary leap required.

Dan Packer, the AUKUS director for the Commander of Naval Submarine Forces, is reported in the media as saying SSN-AUKUS is a UK-designed boat with US technology, but “we are still arm-wrestling over what US technology that means.” DefenseNews paraphrased Packer saying ‘SSN-A will have some American, some British and some Australian components in the nuclear reactor plant — but the parties are now considering the implications of whose technology to use, even down to whose turbine generator might be compatible with whose associated lube oil, for example.’ Packer stated that ‘right now, we are in a game of trying to put metrics to which system is better.’

Putting aside the metaphor of arm-wrestling which does not convey a sense of collegial cooperation, the admission that the technologies that will be included in SSN-AUKUS have not been decided again indicates this is far from a mature design.

How mature is the design?

How much design work is required?

So if SSN-AUKUS isn’t mature, where exactly is it in the design process. First, as an aside, designing a submarine is a lot of work. Here’s an illustration of the amount of work that went into designing the Virginia. According to the RAND Corporation, ‘‘The Virginia-class design effort took an estimated 35 million man-hours over the approximately 15-year design period.’ [page 32]

Source: John F. Schank et al., Sustaining U.S. Nuclear Submarine Design Capabilities, RAND, 2007, online.

While don’t know how much design work is involved in SSN-AUKUS, we can understand why AUKUS leaders want to describe the SSN-AUKUS design as an evolution rather than a new design—it suggests that its Mount Design may not be quite so high or steep or as difficult to conquer. But wherever SSN-AUKUS sits on the evolved vs new spectrum, it will still be many million man-hours. They don’t all need to be performed before the start of construction, but what we should see is a dramatic ramp up in design hours as we approach the start of construction as detailed design takes a lot more work than the earlier phases.

Design phases

So where is SSN-AUKUS on the design path? It’s possible to use much more precise terms than ‘mature’. In US terminology, a naval ship or submarine goes ‘progresses through four basic phases, with each successive phase adding more detail to the evolving design products’. RAND Corporation describe them as follow:

  • Conceptual design phase: ‘future missions and threats are evaluated and weighed against the availability of future technologies suitable to accomplish the required missions. Various concepts are explored and defined, and trade-offs are made among military effectiveness, affordability, and producibility.’
  • Preliminary design phase: ‘The preferred concept is matured and top-level requirements are established in greater detail. Subsystems are defined, and alternatives are evaluated for military effectiveness, affordability, and producibility. Detailed analysis of structures, hydrodynamics, acoustics, and combat system performance is also conducted.’
  • Contract design phase: ‘The top-level requirements are transformed into contract specifications for detailed design and construction of the submarine. Subsystems are defined, initial analyses and testing are completed, projected costs are established….’
  • Detailed design phase: ‘The contract drawings and ship specifications of the contract design phase are transformed into the documents and drawings necessary to construct, outfit, and test the submarine.’

However, not all phases are equal. The first three combined are shorter than detailed design, as this RAND Corporation graphic indicates:

Source: John F. Schank et al., Sustaining U.S. Nuclear Submarine Design Capabilities, RAND, 2007, online.

The UK may not use identical terminology, but it will have an analogous process. Where is SSN-AUKUS in that process? Again, the AUKUS governments have not made any clear statements, but the evidence is clear that we have not yet started detailed design.

Vice Admiral Martin Connell, Second Sea Lord of the Royal Navy, is another AUKUS leader who has defaulted to the word ‘mature’. He stated at a convention in April this year that design efforts led by the U.K. were ‘relatively mature’. But he then stated that ‘discussions with the industry partners now centre around “here’s what we think the future operating environment is going to look like, here are the capabilities that we think we’re going to need, and here is the timeline we’re going to need to deploy it. We need to start building this pretty soon. So there’s going to be more of that over the next 12 to 24 months as we lock in the final build design”’.

Discussions on requirements don’t indicate a mature design. It simply can’t be called a mature design if we haven’t locked in the ‘final build design’ (a term that suggests the detailed design phase). If we are still assessing the capabilities we need to address the future operating environment, then that sounds like we are still in the conceptual design phase. Packer’s comments provided above, namely that subsystems were still being evaluated, suggests we may at best be in the preliminary design phase.

On 1 October 2023 the UK Ministry of Defence announced it had signed the Detailed Design and Long Leads (D2L2) contracts valued at 4 billion pounds with BAE Systems, Rolls-Royce and Babcock. That doesn’t necessarily mean detailed design has started.

Incidentally, considering the difference between Australia and UK’s operating environments, it seems unlikely there will be complete unanimity on the requirements those will generate or the solutions to those requirements. How those differences will be resolved is not clear, but since the detailed design contract is between BAE and the British government, with the Australian government not a signatory, we can easily see which government has the greater say in the matter.

Design milestones

Another way to assess maturity is to state which milestones the program has passed and how far it is from future ones. A standard form of milestone in systems engineering are technical or design reviews. Again, these are not fuzzy concepts like ‘mature’ but are set out in international design standards. SEA 1000, the Attack-class submarine project, used IEEE 15288.2 – IEEE Standard for Technical Reviews and Audits on Defense Programs. It states ‘Technical reviews and audits are a foundation element of an effective systems engineering (SE) approach and form the backbone of a robust technical assessment process. Technical reviews and audits provide a venue for baselining technical requirements, evaluating the system’s technical maturity, and identifying and assessing risks to system performance, cost, and schedule.’

Here is SEA 1000 progress against its design review milestones (as well as its schedule performance in achieving those milestones) just before it was cancelled, published in the Australian National Audit Office’s Major Projects Report. So it’s quite possible to state publicly where a project is against those milestones.

Source: Australian National Audit Office, 2020-21 Major Projects Report.

Again, the SSN-AUKUS may use a different standard and requirements reviews (IEEE 15288.2 was originally an American standard). But so far we have no statements from the AUKUS governments on what standards are being used to set milestones or where the SSN-AUKUS program sits against those milestones. Without them, terms like ‘relatively mature’ or ‘70% mature’ mean little.

Money spent, work done

Another way of assessing maturity is the amount of money you’ve spent and the amount of work you’ve done (assuming the money has been spent on useful work). But not a lot of money has been spent on SSN-AUKUS design work to date.

On 17 September 2021 (i.e., the date of the initial AUKUS announcement), the UK Ministry of Defence announced that ‘design work for the next-generation of Royal Navy submarines is underway following the award of two contracts to UK industry, the Defence Secretary Ben Wallace has today announced. Two contracts worth £85-million each had been awarded to BAE Systems and Rolls-Royce to deliver design and concept work for a future Class of Royal Navy submarine.’

In the same media release, UK Submarine Delivery Agency CEO Ian Booth was quoted as saying ‘Designing and building submarines is one of the most complex and challenging feats of engineering that the maritime industry undertakes. It is essential that work on the next generation underwater capability commences as early as possible. This relies on some of the nation’s most experienced defence nuclear experts from the very beginning of the design phase.’

These statements indicate initial design work was only just starting less than three years ago.

Moreover the amount of money involved was a relatively small A$320 million. The media release said that funding would support around 250 workers at BAE Systems in Barrow and another 100 Rolls-Royce in Derby. 350 workers doing somewhat less than three years of work is around 600,000 man-hours. That may sound like a lot, but recall the Virginia required 35 million man-hours over 15 years.

As a point of comparison, Australia’s SEA 1000 was spending over $500 million per year before it was cancelled, and it was still around three years from the start of construction. So the expenditure to date on SSN-AUKUS is by comparison very small, indicating there is still a lot of work to do.  

Concluding remarks

Despite AUKUS leaders’ fondness for the term ‘mature’, there is little evidence that suggests that SSN-AUKUS is mature. The statement that SSN-AUKUS is ‘70% mature’ means little and is open to interpretation. Perhaps it means there is 70% commonality between the Astute and SSN-AUKUS’s components. But those components still need to be designed into the SSN-AUKUS. Considering the amount of man-hours required for detailed design, it’s very hard to conclude based on the small amount of work already completed that the program has completed 70% of detailed design work or that it is 70% of the way through the design schedule.

Does that mean SSN-AUKUS won’t start construction this decade, as set out in the ‘optimal pathway’s’ high level schedule? It is very difficult to say based on the public information. The Virginia took around eight years from the start of design work and four years from the start of detailed design to start construction, so through herculean effort it may still be possible.

SSN-AUKUS likely hasn’t started detailed design. Moreover, detailed design will take a massive ramp up in design resources from where the program has been over the past few years. We are still a long way from retiring the schedule risk around achieving the start of construction.

The AUKUS governments could, of course, provide more transparency around this issue. The ANAO’s reports show that it is possible inform the public where a project is against its key milestones.

The fact that the governments have not disclosed basic information such as the dimensions of SSN-AUKUS could have two causes. The first is that they don’t like transparency. That’s concerning but can be easily fixed—and must be if they want to bring their publics along on this journey. The second is they don’t yet know what the dimensions of the submarine are. That is also concerning but would indicate that we are a very long way from having a ‘mature’ design.