From a post to davidweber.net on 9/5/2009

The nature of Manticore's battlecruisers

    Okay, someone's asked me to put in my two cents worth on the entire nature of the Manticoran Navy's view of battlecruisers, about whether or not Nikes can be fitted with Keyhole-Two, how practical it is to refit from Keyhole-One to Keyhole-Two, etc., etc. I understand that people are sort of heaping speculation on top of speculation, and then proceeding to argue with one another over it.

    The sad truth is that I do not have time to deal with this by chasing down all of the threads. For that matter, at the moment my Internet connection is what one might call less than fully reliable, which makes trying to read comments on line slow, frustrating, and… less than effective. I am up to my hip pockets in deadline pressure, and I have a convention coming up in a couple of weekends which is going to blow yet another hole in my writing schedule. Accordingly, what I'm going to do is offer a brief thumbnail about Keyhole, Apollo, Royal Manticoran Navy battlecruiser policy, and design decisions. I do not have time to go back and thoroughly peruse my notes before writing this, so I do not guarantee that I will get every single detail correct. Trust me, there's enough technology floating around in the Honorverse by now that even I need to refer to the Tech Bible if I'm going to be positive that I've got all the details right. Also, do not forget that this is ongoing, evolving technology, that the storyline is also ongoing and evolving, and that I may choose to go in a different direction.

    Now then.

    Keyhole-One was originally envisioned solely as a telemetry relay platform. Think of it being something along the lines of a submarine raising its radio mast to transmit. The idea was to get a control platform outside the boundaries of the wedge in order to allow a ship's fire control to establish and maintain telemetry links around the interfering barrier of the wedge. Moreover, the original concept concentrated almost entirely on considerations of improved offensive fire control, which did include the idea of giving greater flexibility to target management. In particular, one of the very early concepts was to facilitate "hand-off" between ships, allowing the ship with the best "visibility" to manage fire from her consorts, but did not include any great concern with managing counter-missile fire, extending sensor reach, or making any direct contribution to the mounting ship's close-in defenses.

    As the concept began working its way through development, however, it began to evolve. Initially, the idea had been that each ship would carry a large number of relatively small, cheap, expendable communications platforms. They would be outside the wedge and the side walls, and thus vulnerable, and what they were expected to do was relatively simple. The RMN's remote sensor platform capability was already good enough that the emphasis was on a cheap platform designed purely to communicate with outgoing missiles.

    In the development process, BuWeaps came to consider additional missions Keyhole might be expanded to include. One of the very first was to include additional telemetry links for counter-missiles. Another early contender was to use the new system to expand a ship's "onboard" sensor perimeter, giving it better "situational awareness" in its own area, regardless of where the remote sensor platforms might be deployed. As its mission and capacity grew, Keyhole became a steadily more sophisticated and capable — and thus larger and less expendable — platform. As it incorporated its own sensor suite and expanded its communications, it became increasingly valuable (in both the tactical and the logistical senses), which made it only logical to fit it with its own point defense. It was made as stealthy as something radiating as powerfully as it did could be made, and it was equipped with its own rudimentary ECM in order to make it more difficult to localize it and destroy it. And, of course, each incremental increase in capability brought with it its own incremental increase in size and cost. You can, if you will, think of this as setting out to design the F-16, or even the A-10, and ending up with the F-15. Every step along the way made absolute, demonstrable, unquestionable good military sense, and the final product was worth every penny of investment, and yet what emerged at the end of the developmental process had changed so much in degree that it had ended up changed in kind, as well. It was, effectively, a completely different animal from the initial concept.

    So, at the end of the development process (I'm speaking here of Keyhole-One development), the original cheap, expendable, single-function telemetry link had evolved into a highly capable platform which was an integral part of the mounting ship's sensor suite, provided a much more capable communications node then had originally been envisioned, was stealthy and hard for any opponent to lock up for offensive fire control, and which possessed sufficient onboard point defense capability to not simply defend itself but offer a significant increase in the mounting ship's close-in defenses, as well. The platform itself is stuffed full of essential equipment and hardware, but probably at least a quarter of Keyhole-One's capabilities depend on computer support aboard, and (especially) power generation from, the mounting ship.

    The original Keyhole-One platform was about the size of a LAC. The more fully developed Keyhole-One platform carried aboard units like the Nike-class battlecruisers is substantially larger, and fitting a ship to carry it costs quite a bit of potential broadside armament space. It also presents some armoring difficulties, since the platform itself has to be armored when it is tractored into its bay on the exterior of the mounting ship, and the bay itself has to be armored in order to protect the ship when the platform is deployed. Because of those considerations, at the moment, no Keyhole-capable ship currently carries more than one platform in each broadside. This would give a squadron of six ships 12 Keyholes, and, especially given the platform's elusiveness and self-defending capability, the RMN regards this as sufficient to guarantee reasonable survivability through redundancy.

    Keyhole-Two is another can of worms entirely. First, the platforms themselves are substantially larger. While the final (or, at least, currently final) generation of Keyhole-One is somewhere around 65,000 tons (or darned near the size of a prewar destroyer), Keyhole-Two is even larger. This is because in addition to the requirement that it must retain its light-speed telemetry links for counter-missiles and non-Apollo shipkillers, it must also fit in the dedicated FTL coms used to communicate with the Apollo control missiles. In other words, a Keyhole-Two platform has to be "bilingual," with the capability to perform its Apollo control function in addition to all of the standard Keyhole-One functions, and this inevitably drives size upward. It is also even more heavily defended, since each platform is individually bigger (and more expensive), represents a larger increment of the mounting ship's capabilities, and (because of its size and emission signature) is a less elusive target. The power budget is also substantially greater. A very large percentage of the computer support carried on board by Keyhole-One has to be located inside the mounting ship, which eats into the ship's internal volume. Additional power generation and transmitting equipment is also necessary, which eats even further into internal volume.

    A superdreadnought fitted with Keyhole-One can be re-fitted with Keyhole-Two fairly quickly. Note the use of the word "fairly." Essentially, the superdreadnought has sufficient volume inside its protected core hull that fitting in the additional equipment — while not especially easy — is much simpler than it would be in, say, a Keyhole-One-equipped battlecruiser. Every superdreadnought so far fitted with Keyhole has been a pod-layer, and the quickest and simplest way to accommodate Keyhole-Two is to partition off one end of the central missile core, armor it thoroughly, and then mount the required equipment in the protected space thus created. This somewhat reduces ammunition stowage, does not give you ideal access for servicing and routine maintenance, and leaves the critical new components at least marginally more vulnerable than they would be if they were located inside the core hull proper. On the other hand, any hit which got to the onboard end of the Keyhole-Two installation would almost certainly have to come up the missile core from aft. In such a case, they hit would already have done so much damage that the ship's true main battery — it's missile pods — would already have been mission-killed. Note, however, that a ship need not be able to launch its own pods in order to control someone else's pods through its Keyhole platforms, so theoretically, at least, even a superdreadnought whose missile core had been completely gutted could still be combat-effective. For example, a consort whose own fire control had been crippled might well roll pods for an SD(P) which had lost its own missile core but still had Keyhole-Two capability.

    It is far, far more difficult to upgrade Keyhole-One to Keyhole-Two in a ship below the wall. Something like an Agamemnon, with its own missile core, could theoretically retrofit the same way a superdreadnought does, but the mass penalty is exactly the same for both ships, and the much smaller BC(P) has far less ammunition volume to sacrifice than a SD(P). In other words, while the actual volume and mass is identical, the penalty paid is proportionately much higher for the BC(P) than for the SD(P). Moreover, the installation would be far more weakly protected in a BC(P), simply because the BC(P)'s much lower total mass is less capable of absorbing damage in the first place, which doesn't even consider the fact that its general armoring scheme is so much flimsier that of any superdreadnought.

    Retrofitting a Nike would be even more difficult. There is very little empty volume in a Nike. In order to shoehorn in all of the necessary ancillary computer support, power generation and transmission, etc., hardware a Nike would have to (a) locate it somewhere outside her core hull and (b) give up additional broadside weaponry. She's already sacrificed a greater proportion of potential armament in order to fit in Keyhole-One than a superdreadnought with Keyhole-One, because the same size of platform represents a proportionately greater surface area of her hull. If additional weapons are removed, her broadside gets still weaker, and the shipyard has to get into cutting, moving, and relocating armored bulkheads between the skin of the ship and the core hull in order to create compartments in which to place the shipboard end of the system, which is a nightmarish task. Moreover, even after the modifications are made, the critical equipment would be outside the core hull, which means that it would be extraordinarily vulnerable.

    It might well be possible to build an entirely new class of battlecruiser and include Keyhole-Two capability from the beginning. That would be a considerably more practical solution than trying to squeeze a system even ships-of-the-wall have trouble accommodating (as a refit) into a relatively small hull. In other words, don't expect to see the current generation of Keyhole-Two refitted to a Nike anytime soon.

    Now, even if it were feasible to refit Keyhole-Two to an Agamemnon or a Nike, it's unlikely that the RMN would do so. The Manticorans are not interested in building the hyper-capable equivalent of the Graf von Spee. (For those not familiar with the reference, looking it up is left as an exercise for the student. [G]) They've already come perilously close to doing that with the Agamemnon, and as a consequence they've had quite a few BC(P)s blown out of space fighting true ships-of-the-wall. This is not and never has been (as far as Manticore is concerned) proper battlecruiser doctrine. They are not going to build ships which will encourage this misuse of the type still farther.

    The entire reason the Manticorans built the new Nike-class in the first place was to resurrect the classic battlecruiser — something which can destroy (or at least reasonably encounter) anything below the wall, which is an ideal commerce-raider, which provides a powerful convoy escort (when needed), which can be built in sufficient numbers to carry out power and influence projection, flag-showing, and similar roles without diverting ships-of-the-wall, and which can function effectively as a space-control vessel, usually by itself or in company with only the rest of its own division. Note that in none of these cases is the ship expected to take on and defeat superdreadnoughts. Obviously, situations can (and have, and undoubtedly will continue to) arise in which battlecruisers find themselves forced to "fight above their own weight," anyway. That is, unfortunately, the nature of the beast. (Or, as Eric Flint's and David Drake's Bellisarius is fond of saying, "Things always get screwed up when the enemy arrives; that's why he's called 'the enemy'.") However, BuShips has no intention of designing vessels which will encourage fleet commanders to do anything of the sort.

    The Nike is much, much, much tougher than any BC(P), has better active defenses and far heavier armor (and Keyhole-One) than the Saganami-C, is designed to have more combat endurance (defined in terms of sustained engagement time rather than total throw weight) than either an Agamemnon or a Saganami-C, can run away from anything that can defeat her, has the acceleration to catch anyone else's battlecruisers or cruisers, out-ranges any potential enemy accept an MDM-equipped Havenite SD(P), and has designed accommodations capable of supporting a much heavier Marine complement than is normally embarked. In addition, every Nike is equipped to function as a flagship. In other words, she was designed from the keel out to be a battlecruiser on steroids, and the RMN is very pleased with the way the design has worked out in practice.

    As soon as possible, BuShips plans to phase the Agamemnon out of production entirely. There is no battlecruiser role which a Nike cannot perform as well as or better than an Agamemnon. The fleet's concerns with the type's fragility and vulnerability are increasing, not decreasing, and the decreased manning requirements of current-generation SD(P)s make the heavier type even more attractive where true MDM capability is required. I understand there have been some suggestions that an Agamemnon equipped with Apollo should be able to devastate hostile SDs from far beyond their effective range. BuShips and BuWeaps are not strangers to that same argument. However, just carrying standard MDMs aboard an Agamemnon already puts a severe squeeze on her ammunition capacity. Carrying Apollo pods would make that still worse. Besides, economically, industrially, and in manpower terms, it costs substantially more on a per-missile/per-pod basis to carry even standard MDMs onboard an Agamemnon than it does to carry them aboard an Invictus or a Harrington II. Not only that, but an Agamemnon is substantially smaller than a Nike. Sacrificing volume and tonnage to refit with Keyhole-Two (which would be necessary to create an Apollo-capable Agamemnon) would take a much greater proportional bite out of the ship's weapons capability than would be the case for properly redesigned a Nike.

    At the present, there is no Apollo-capable version of the Mark 16 dual-drive missile in prospect. That's not to say that one might not be developed at some future time. Should that happen, however, the RMN is virtually certain to field the missile aboard a Nike variant rather than aboard an Agamemnon-derived design. It is probable (and I stress that all of this is purely hypothetical and that the hardware to make it work is not on anyone's radar scope at this time) that an Apollo-capable derivation of the Nike would still be equipped with broadside tubes, but would mount perhaps a pair of MDM-capable tubes in each broadside for the specific purpose of launching the Apollo control missiles. Again, I am most assuredly not saying that any such class is on the horizon. I'm simply saying that current Manticoran thinking is such that a pod-layer Apollo-capable BC is unlikely to emerge.

    I have no doubt that someone is busy suggesting that the Manticorans ought to be designing some sort of "command battlecruiser" whose sole function would be to manage Apollo broadsides. This is not going to happen. The Royal Manticoran Navy has learned the hard way (both from observation of others and from its own wargames) that it does not want such special-function units, and that it especially does not want them in its cruiser or battlecruiser force. The very nature of Manticoran doctrine regards the ships as multifunction, multipurpose units — generalists — and the Manticoran Navy has never been really happy with the BC(P). The BC(P) has been viewed (by the vast majority of the Navy) as an interim, improvised platform designed to get additional MDM-capable platforms into space in the shortest possible time period. Note also that the Agamemnons (and the Graysons' Courvoisiers) were both pre-Mark 16 designs. In other words, the only way to give a battlecruiser a range advantage over standard single-drive missiles, at the time the design was conceived, was to use all-up MDMs. Neither Grayson nor Manticore was ever really happy about packing such volume-intensive weapons into a battlecruiser-sized hull; it was simply the only way they could get them deployed. This is one reason why the Agamemnons switched over to pods of Mark 16s as soon as they became available — it let them regain some of the missile capacity they'd had to sacrifice with the larger MDMs. I'm not trying to say that there aren't still factions in both the RMN and the GSN which favor the original BC(P) concept. I'm simply saying that officers who hold that opinion are in a distinct minority and that the design policies of both navies have set rather firmly against the continuation of the type.

    Someone has mentioned to me that there has also been discussion of an "escort" type — presumably an escort cruiser or even an escort battlecruiser. Please tell me that no one has been suggesting an escort superdreadnought! [G] I can't absolutely vouch for what some other Navy might do, but there's no way that Manticore is going to field anything along those lines. First, they aren't going to give up the tonnage and the platform capability for what would be, effectively, a purely defensive design. Secondly, the RMN firmly believes in spreading it's anti-missile capability about as broadly as possible in order to ensure its survivability through redundancy. Thirdly, the proliferation of increasingly sophisticated long-range surveillance platforms — including, especially, the Manties' own FTL-capable platforms — is making it increasingly straightforward to identify individual units in an enemy formation by analyzing their fire patterns and their emissions signatures at close range. This means that it will become increasingly feasible for the other side to identify a specialized design — like an anti-missile escort or a Keyhole-Two-equipped battlecruiser — and give special attention to killing it as soon as possible. Fourth, anything lighter than a Nike has become increasingly less survivable in a fleet engagement, and the RMN declines to put its personnel in harm's way aboard identifiable, fragile, easily killed vessels. Fifth, Manticore has concluded that the most effective way to thicken the wall of battle's anti-missile defenses is to deploy CLACs and use their LACs in the antimissile role. LACs are much more difficult targets than any hyper-capable starship. They are far less likely to be targeted in their own right, and it is much less likely that a missile which has lost its original target will acquire a LAC in its place. Its own active defenses are very nearly as good as those of a destroyer (although the exact mix of active systems, stealth, ECM, and maneuverability is different), and the antimissile capability per crewman exposed to hostile fire is actually much greater with something like a Katana or even a late-generation Shrike equipped with Vipers or Mark 31 counter-missiles.

    All right, that's roughly 3,000 words, which is approximately 60 percent of my programmed output for a day spent in front of the computer. That's why I have not tried to read all the posts in this thread, why I haven't been turning up on the website or Baen's Bar more often, and why a whole bunch of things haven't been happening. I've been concentrating on getting the books I'm committed to deliver done, and I'm going to go back and do that some more now. In the meantime, though, I hope that this has helped to clear up some of the confusion and ambiguity.

    Take care, all.