May 24, 2019

So You Want to Build a Modern Navy - Aviation Part 4

This is the last part in "So You Want to Build a Modern Navy". I have a bit more in the archives, but it's probably not worth publishing. The following is based on discussions in Aviation Part 2.

John Schilling: I hadn’t thought about my proposal as a recreation of the Light Fleet Carrier, but given the incredible success of the Light Fleet Carrier program over half a century or so, I’ll own it. I’m proposing the Light Fleet Carrier of the 21st century; how can you possibly refuse something with that winning track record.

To address your more specific points.

Saab may not have built carrier-based fighters, but they have a great deal of experience building fighters for short, rough, austere field operations. And as you note they have Boeing. And we’ll be working with them. The technical risk of Gripen Maritime is low. The market risk is also low, because the Gripen family has a large and proven export base and our variant – even if we are the sole customer – will share probably 90+% of the production line and logistics base. By comparison, the technical risk of our developing our own fourth-generation V/STOL fighter fighter from scratch is going to be Not Low. The market risk is also Not Low. So unless you are certain we can afford a pair of Queen Elizabeths, this is the only low-risk path we have to being able to deploy 21st century aircraft at sea in a crisis.

You can’t be confident we can afford a pair of Queen Elizabeths, or anything like them. Steel is cheap, ish, but engineering is hard. Military procurement is also hard, and you cannot simply wave your hand and assert we will not make France’s mistakes. France, after all, made those mistakes, and France is not run by fools. But they are subject to the same economic realities as we are, and more importantly the same political and bureaucratic realities. Every democratic nation is (and the non-democratic nations have their own problems). We will face cost overruns on what would be an expensive proposition to begin with.

For that matter, where are you going to build them? Drydocks and shipyards that can handle a 20,000-ton warship are reasonably common; I count eleven in nine nations which have done so successfully in the last twenty years. By 50,000 tons, you’re down to four facilities in three nations, none of which really do export sales. Or you’re planning on having a commercial shipyard do the work and hoping it has more combat readiness and survivability than a cruise liner. Or you’re expecting our parliament to pay for naval shipyard infrastructure that will build maybe three ships every thirty years (and with no potential for export sales). At 30,000 tons, we’re probably still in the low risk category for shipbuilding, but it’s going to increase rapidly beyond that. So will cost, unless you are making use of otherwise-idle capacity of a shipyard sized for very large warships.

The Royal Navy was able to fly a handful of Buccaneers from the Hermes, but tried and failed to integrate the F-4K, which is probably a closer match to an F-18. The French were able to demonstrate Rafales on the Foch, but AFIK didn’t use them operationally. Any plan to operate F-18s, especially F-18Es, from anything much under 50,000 tons is going to be at least medium risk.

From the deck area and logistic footprint, we should be able to fit three Gripens in the space of two F-18s. That’s a wash in terms of net payload delivery and a clear win in operational flexibility. Plus, the Gripen NG has a better strike radius even before you consider tankers.

This has given me an excuse to go break out my old aircraft-design textbooks and do a preliminary design for a twin-turboprop common support aircraft, aka Mini-Viking. 18,000 kg MTOW with a modular 3,000 kg payload capacity, 2000nm range at 300 kts or 9 hours endurance at 200 knots. Should work just fine on a 30,000 ton carrier, should be able to carry something like the Erieye radar in an AEW variant. Or a Growler’s worth of EW gear, or a Viking’s worth of surveillance and ordnance, and the tanker variant can transfer 4000 kg of fuel at 600 nm. Which is enough to boost the strike radius of a pair of Gripens by more than 50%. This also should be a low development risk, especially compared to a V/STOL fighter, and the export market won’t be limited to just carrier operators.

The export potential for the Light Fleet Carrier 21 is not guaranteed, but consider its namesake. By comparison, the only export sales for anything over 30,000 tons have been cold-war surplus Russian salvage. A carrier program for a nation like ours is going to be expensive, perhaps prohibitively so (see e.g. France). Anything that offers the possibility of cost-sharing with foreign partners needs to be considered very closely.

Light Fleet Carrier 21 combines low cost and technical risk for the ship with low cost and technical risk for the air wing, doesn't require us to deal with the USA's perverse export laws, and is the only certain way to operate modern combat aircraft at sea for a nation that isn’t richer than France or the UK. And even the UK isn’t entirely happy with the price they are paying for that capability. They were, however, more than pleased with the results of the original Light Fleet Carrier, and my vote is for following in their footsteps.

Bean: That's an attractive campaign speech, but I see a couple of problems. First, I think you fundamentally misunderstand the conditions that lead to the success of the Light Fleets. I see two big problems: markets and cost. WWII radically changed the nature of naval power, which meant that navies were scrambling for carriers. The Light Fleets were what was available at the time. The US did hand out three of its CVLs to France and Spain. I'd assume the UK was more generous because of its postwar economic problems. Today, everyone has settled into a set of choices for the current environment. They may not have the budget or the manpower to take on a new carrier or three. It may not fit their doctrine or strategic choices.

On the cost front, I believe all of the Light Fleets that were exported were sold at a net loss, which makes exporting a lot easier. The Colossus class were built using war funds, and after the war, when Britain needed hard currency, selling them for however much they could get made a lot of sense. The Majestics weren't complete by the end of the war, but the last was launched in September of 45, so a lot of the work was already done and paid for before the Australians, Canadians, and Indians got involved. We'd have to sell at a profit for this to make any sense. And for a big project like a carrier, it's really hard to compete with the domestic jobs lobby. We might be able to sell the design and let them build locally, but even that's hard to do, and we'd need to show a major cost advantage.

I'm of several minds on aircraft selection. On one hand, the navalized Gripen does look to be a good compromise, light and easy to operate, but with relatively low risk. On the other hand, we have to sell both the carrier and the airplane. Brazil is the only country that flies the Gripen that might realistically want a carrier, although I agree that they're at the top of the list of possible partners. If we design for Super Hornets and Rafales, even if we go with the Gripen ourselves, we open up a lot of options. France, India, and Australia in particular operate one or the other of those planes, and it would make it a lot easier to sell to them.

On the other hand, if we go for the Expeditionary Fighter, we only have to sell the plane. The target markets already have their own ships, and we're offering them an upgraded capability. I think the case for CATOBAR is strong enough we don't want to go that way, but I think that it's got better export prospects than Light Fleet 21 and Sea Gripen.

As for not having the same problems France did on the Charles de Gaulle, that's not impossible. We don't yet have a huge parasitic organism attached to our procurement process, which makes working efficiently feasible. Note that when western militaries can protect programs from said organism, they're very effective. The B-21 is a good example.

You make a good point on size and building facilities, although we might be able to get around that. I'm thinking that our best option might be to partner with France on this, actually. They're still interested in more carriers, and have the facilities to build them. And they're by far the most export-focused of the major powers. If we offer to split the development costs on a conventional CDG/QE class carrier, they might bite. If they won't, India has been making noises about going to CATOBAR. In either case, we'd need to make it big enough to fly Rafales, and I expect we'd end up flying them too if we do a deal with France.

Hermes was retired as a CATOBAR ship just about the time that the Phantoms entered service. I've checked several books, and wasn't able to get a clear answer on why Hermes was transitioned to a commando carrier, but none of the mentioned the problems with the Phantom that wiki talks about. A refit would have been needed, but that was well-known. In Fighters Over the Fleet, Friedman goes into some detail on the Phantom performance off Hermes. It did impose some tradeoffs, for instance in CAP endurance, but it was only serious in the tropics in a dead calm.

I'm not certain that Light Fleet 21 is a bad idea by any means. If it turns out that we can't get a partner who wants a bigger carrier, it's probably the way to go. But I'm less certain of the export prospects of a smaller design with a nearly unique aircraft than you are.


  1. May 24, 2019redRover said...

    Or you’re planning on having a commercial shipyard do the work and hoping it has more combat readiness and survivability than a cruise liner.

    I don't think this is as true for front line carrier/destroyer type ships, but haven't other nations done basically this for second line ships like amphibious assault vessels, replenishment vessels, and so on? (I.e. essentially built to commercial standards than painted gray, rather than built to naval standards)

    Not that you want your carrier built by some Chinese freighter company for bottom dollar, but I think cruise ship builders might be close enough. France's Chantiers d'Atlantique (sp?) for instance, builds both cruise ships and naval vessels, so if we can build a cruise ship construction industry perhaps that will support some of the industrial base.

    Though of course, given the French experience maybe that's a strike against it.

  2. May 24, 2019bean said...

    Commercial standards are kind of a complicated issue. They're not necessarily worse than warship standards in a lot of cases, just different. For instance, they have thicker skins, but a low fewer frames. This is bad if you happen to be dealing with a major shockwave (say, from a nuclear weapon), but good if you run into something. Likewise, things like firefighting are based more on accidents than on weapon damage. It's somewhat cheaper to build to them because most commercial shipyards know a lot more about how to build that way, nothing more.

    Also, remember that ships don't make money when they aren't moving, and that the sea is unforgiving, so commercial standards are pretty high. There's very little to choose from between, say, an oil exploration vessel and a military oceanographic vessel.

    And it's also a rather flexible term, because you can ask for more in certain areas. I'm sure the typical auxiliary bought by a first-rate navy, even if "built to civilian standards" is more survivable than a typical oil tanker or cruise ship. "Yes, the commercial book says we only need two emergency generators, but we'd like three instead. And yes, we'll add a few more bulkheads in the liquid tanks, too."

  3. May 26, 2019IsANobody said...

    For the hypothetical Mini-Viking, would a navalized AT-802L work?

  4. May 26, 2019wubbles said...

    If we're going to be building a nuclear submarine, we might as well order another pair of naval reactors for the carrier. The benefits of nuclear propulsion aren't as extreme as for a submarine, since carriers are operating on the surface, but depending on fuel consumption it could simplify the logistics, at the cost of more crew. Also electrical launch has some benefits for airframes.

    Canada considered using an inherently safe conduction cooled reactor on their submarines, but decided not to. If we want to take less of a technical risk, we can stick with PWR.

  5. May 26, 2019cassander said...


    the charles de gaulle class shows the folly of that approach. the CDG is very small for a carrier, and has been woefully under powered its whole life because of its reliance on submarine reactors. You'd have to building a very large submarine (something columbia sized at least, if not much more) to get a reactors powerful enough that 2 could power a decent carrier.

  6. May 27, 2019redRover said...

    I don't think the AT-802 would work, as the flight envelope isn't really suitable.

    Depending on how large the carrier is, I would think a C-2 or a V-22 variant would provide a lot of capability, especially re in flight refueling, as well as conversion to being an AEW platform. The C-2 is about 25% larger than the platform identified by Schilling, but it also exists so there is no marginal development costs.

  7. May 27, 2019redRover said...


    I know you're generally skeptical of sub powered carriers, but it seems to me that as a small nation with a small production run of carriers and subs, the R&D versus operating cost trade-off is a bit different than if you're the US or Russia, as you have less units to amortize the R&D over. (Especially if there is no export market.) Obviously you would need to run a more in depth analysis, but it seems like you broadly have the option of 3x or 4x sub reactors, 1-2x special surface combatant reactors, or conventional dead dinosaur propulsion of some variety. While in peacetime I bet conventional comes out better, during actual combat operations it seems like nuclear could be more attractive, given the increase in logistical tail required in terms of extra oilers and so on, and their vulnerabilities.

  8. May 27, 2019bean said...

    I'm with cassander on small CVNs. They just don't make sense. You can cram a lot more power into a dinosaur-burner plant than a nuke, and a big ship takes proportionately less power to drive it at a given speed. It's not really a matter of sub reactors vs non-sub reactors.

    (As an aside, I think part of the problem with the CDG is that she was designed to run on low-enriched Uranium because of a complicated nonproliferation scheme, which did all sorts of bad things to her.)

    That said, I'm almost wondering if the Kirov solution doesn't make some sense here. A sub reactor or two for base load/cruise/electricity, and a boost plant burning fossil fuels for when you really need to move. It's probably a bad idea (although integrated electric makes it more attractive than it otherwise would be) but it does have advantages.

  9. May 28, 2019bean said...

    Also, I just found out that Norman Friedman is about to publish a new book on British submarines through about 1945. I'm not particularly surprised by this, as it was the obvious gap in his series on British warships. (I keep hoping for United States Auxiliary Ships: An Illustrated Design History, but that was always a longshot.) What I am surprised by is that USNI apparently has no inkling of this, and it's coming out from Pen & Sword in the UK in a month, at least according to Amazon UK. I'm debating whether to order it from there, or wait until the US edition comes out.

    In other book news, I'm still waiting for the new Dulin & Garzke and RA Burt books that should release in the next few weeks. Get here soon!

  10. May 31, 2019doctorpat said...

    On the subject of

    Drydocks and shipyards that can handle a 20,000-ton warship are reasonably common; I count eleven in nine nations which have done so successfully in the last twenty years. By 50,000 tons, you’re down to four facilities in three nations, none of which really do export sales.

    Does this include the Ukraine? Because they've got the old USSR docks, and I imagine they aren't too opposed to export earnings at this point. Ukraine dockyard

  11. May 31, 2019bean said...

    First, I'd really rather not rely on Ukraine for something like this. They're just not stable enough for me to wager our navy on their remaining friendly. Second, there might be legal issues. China bought what is now Liaoning using the cover story that it was to be a casino and hotel in Macau at least in part because there was some concern about the legal issues of trying to move a foreign warship through the straits. And reading the Montreaux Convention, there's definitely no provision for what we'd want to do. And I really, really don't want to hold our carrier program hostage to Turkey, either. Third, a shipyard isn't just a big drydock. You need skilled workers and expertise, and I'm not sure the Ukranians have that necessary to build an acceptable warship of any size at this point, much less a carrier. I'd rather we convinced a Japanese or Korean civilian yard to do the job.

  12. June 01, 2019Inky said...

    That said, I’m almost wondering if the Kirov solution doesn’t make some sense here. A sub reactor or two for base load/cruise/electricity, and a boost plant burning fossil fuels for when you really need to move.

    Does it make sense from the logistic standpoint? I mean, a ship of the size of Kirov or a carrier is likely to be a flagship of a battle group, which is likely to include a supply ship. Left on it's own it's an easy prey anyway. So what's the point in increased range anyway, since the limiting factor will probably be the range of smaller ships?

    Also, I second bean on the (dis)ability of Ukrainian shipyards to build pretty much anything at this point. Most of the skilled workforce was lost as early as first half of 90-s. The shipyard that doctorpat probably had in mind now looks like this.

  13. June 01, 2019bean said...

    Does it make sense from the logistic standpoint? I mean, a ship of the size of Kirov or a carrier is likely to be a flagship of a battle group, which is likely to include a supply ship. Left on it’s own it’s an easy prey anyway. So what’s the point in increased range anyway, since the limiting factor will probably be the range of smaller ships?

    I think it might. Big ships take a lot of fuel, and carriers in particular tend to operate in ways that burn huge quantities. And the last few knots are particularly expensive. You're often looking at doubling power to get 4-6 knots. So a carrier capable of 30 kts with a 20 kt nuclear plant will have a nuclear plant a quarter of the power of a 30 kt nuclear carrier, the balance made up by gas turbines, which are quite light. But we spend most of our time running around at 20 kts, and only hit 30 kts occasionally. This change has probably cut our fuel consumption by 75%, which means we get a lot more utility out of that oiler. On the other hand, this has a lot of the other drawbacks of nuclear, without all of the benefits. I'd guess that there are significant economies of scale in manning and operating costs on this. Not a decision I would want to make without actual numbers.

  14. June 03, 2019John Schilling said...

    Late to the comments on this one, but I do like the idea of CONAG propulsion for a navy that has submarine reactors already. The carrier is not going to operate without an escort group, which will be burning oil even at 15-20 knots, so as bean notes it is helpful that the carrier isn't competing with its own escorts for the resources of the replenishment ship. And for that matter, there is precedent for aircraft carriers refueling their own escorts at sea, which is more practical if the carrier doesn't absolutely need fuel for its own use.

    The problem is that it would likely limit export sales due to nonproliferation concerns; submarine reactors typically use the sort of highly-enriched uranium that can easily be made into atom bombs, and IAEA safeguards can't readily be applied to the reactor room of a warship. Unless a careful study shows that there really is no export market, I'd rather have that cost-sharing than the marginal advantage of CONAG propulsion.

    Regarding a AT-802 derivative as the support aircraft, it's really too small. Not just in terms of gross weight, which is perhaps negotiable, but volume as well. We want something that can carry at least 8-10 passengers or equivalent bulk cargo in a COD role, or two pilots and two systems operators and a bunch of accessible rack-mount electronics in most of the operational roles. Internal carriage for the basic weapons load is a must to meet the range and endurance requirements.

    A navalized King Air 350 would be a possible candidate, if you could do it. But I'm skeptical that airframe can be made compatible with carrier operations, and the gross weight is too small for the tanker missions.

Comments from SlateStarCodex:

Leave a comment

All comments are reviewed before being displayed.

Name (required):

E-mail (required, will not be published):


You can use Markdown in comments!

Enter value: Captcha