February 25, 2019

Open Thread 20

It's time for our biweekly Open Thread. Talk about whatever you want.

Book Review: The Yard

The Yard profiles the construction of the USS Donald Cook at Bath Iron Works in Maine. It's slightly dated (Cook commissioned in 1998) but it's well-written, and I couldn't find more than a few minor technical details to take umbrage with, which is really good in a book of this type. The author, Michael Sanders, paints a good picture of not only the process of shipbuilding but also the people who do it. It's definitely written more on a popular level than a scholarly one, but I learned quite a bit from it, too. I'd say it should be easy to find in libraries (it certainly was 10 years ago), but I suspect that the reason it's so cheap at Amazon is that they're getting rid of their copies. Overall, highly recommended.

Overhauled posts since last time are Amphibious Warfare Part 4, Classes, my discussion of Dreadnought, Strike Warfare and the first two parts on battleship propulsion.


  1. February 26, 2019AlphaGamma said...

    News: The name for the RN's third Dreadnought-class SSBN has been announced. Warspite joins Dreadnought and Valiant (these were also the names used for the RN's first three SSNs).

    This means it is likely that the fourth boat in the class will be HMS Resolution.

  2. February 26, 2019bean said...

    I’m not so sure of that. Resolution was an SSBN, not an SSN. My top three picks are Churchill, Conqueror (if they decide to skip Churchill for some reason, plus it gives a nice thumb in the eye to Argentina) or Barham if they want to complete the set of QE-class names they can actually use (Malaya being out for obvious reasons).

  3. February 27, 2019Aula said...

    Somewhat old news by now, but I haven't seen this mentioned here: the wreck of USS Hornet (CV-8) has been discovered.

  4. February 27, 2019bean said...

    I actually checked, and was frankly staggered by the number of ships Paul Allen's team has found. Anyone interested in naval history owes him a great debt, and I have half a mind to start a lobbying campaign to get the Navy to name an oceanography ship after him. There can be few more deserving recipients of such an honor.

  5. February 27, 2019bean said...

    It looks like there might be some good news on the readiness front. The commanders of PacFlt and Fleet Forces Command (which, by the way, is a much worse name than LantFlt) are claiming that they've held ships back for extra training when they weren't ready to deploy. I'm slightly worried because 7th Fleet wasn't mentioned in the article, and that's where the problem has been worse.

    They're also trying to increase manning on the DDGs, hoping to get it up to 318, which was the design number, in 2023. However, they're still short several thousand sailors.

    On a related note, The Yard mentioned that there was some skepticism around the manning of the DDGs even then, although that seemed to center around the ability of the crew to conduct multiple warfare missions at the same time, because a lot of people have multiple jobs.

  6. February 27, 2019DampOctopus said...

    I came across an interesting topic while reading Air Power Australia. This site seems to have been set up primarily to lobby for Australia to buy in to the F-22 programme, back when that was a possibility, but the particular article I'd like to link is this one, written in 2000, which discusses defence against anti-ship cruise missiles (ASCMs). It raised a few questions for me that I'd like to run past you.

    The scenario described in the article (search for "Defending against Supersonic ASCMs") is of a salvo of sea-skimming ASCMs being detected as they clear the horizon, 20-25 nm or 40-60 seconds away. The target vessel immediately launches a semi-active radar homing SAM, and illuminates one of the ASCMs long enough for the SAM to intercept and kill it. The illuminator then slews to a second ASCM in the salvo (which is now much closer), a second SAM is launched, etc. It calculates that a maximum of three ASCMs can be stopped this way: by the time the fourth SAM is launched, the ASCM salvo is too close for the SAM to accelerate to an effective intercept speed. As the typical Australian ship - an Anzac-class frigate - lacks any further point defence (e.g. CIWS), the fourth ASCM will hit it, unless it is decoyed somehow.

    So my first question is whether this picture seems reasonable: the detection of sea-skimming ASCMs at <60 seconds' warning; the strict limit of one SAM per illuminator; the effectively-100% hit rate with the SAMs; and the conclusion that a salvo of four ASCMs is sufficiently to score a hit on a representative surface combatant.

    My second question is about how to improve the defence. The article mentions increasing the number of illuminators, which would (I expect) linearly increase the size of a survivable salvo. Does the upgrade of the RIM-162 ESSM - the most likely SAM in this scenario - to block 2, with an onboard active radar, get around the illuminator dependence? What about a phased-array radar illuminating multiple ASCMs, flicking between them on millisecond timescales to guide many SAMs simultaneously? Do these raise the maximum survivable ASCM salvo to about the same number of missiles as there are SAMs on board?

    Doing a bit of further reading I see that, since this article was published, the Royal Australian Navy has acquired the Hobart-class destroyer. These are similar to the Arleigh Burkes, with fewer VLS cells but with the same SPY-1 radar, and of course Aegis. I assume this is a response to the vulnerability to ASCMs highlighted in the article. Are Hobarts and Burkes better able to defend against ASCMs than in the scenario described above? Can the phased-array SPY-1 be used for illumination and guiding SAMs, or is it just a search-and-track radar?

  7. February 27, 2019bean said...

    I have questions about several aspects of this. First, I'm not sure about detection ranges without doing a bit more research. I suspect that you could do stuff like infrared plume detection to push the range out more (launching before you have a good track), and you obviously have to fly higher if you're moving faster. Second, I think he underrates the speed of SAMs. ESSM can do Mach 4+. Third, you're very much right about the effects of active missile seekers. No illuminator limits, no problem. And most ships carry more than one illuminator anyway.

    As an aside, the illuminator isn't done by the SPY-1. The dishes on top of the deckhouse are the illuminators. The SPY-1 is useful for guidance, in a manner I've described elsewhere.

    So yes, Aegis ships are vastly less vulnerable to this kind of saturation than older vessels. I may have to dig into this more, but I have stuff to do tonight.

  8. February 27, 2019bean said...

    Another assumption I hadn't thought to question was on timing. He's assuming that every single missile arrives on target at the exact same time, which is not particularly realistic. To avoid missiles hitting each other, any launcher is going to need a second or two between launches, which changes the defense picture dramatically.

    That said, I ran this in CMANO with both an Adelaide and a Hobart facing a salvo of 8 3M54T (SS-N-27) with the Mach 2.9 terminal attack. The Adeliade nailed the first one (they were actually very lucky to pull this off, as it was a 15% PH) and got a second missile off, targeted at said first Vampire, but it went blind when that one was destroyed. The second ASM malfunctioned and the third ASM killed the frigate.

    Hobart ate all 8 for breakfast. PK was somewhere around .55, and it was able to retarget missiles if their target was killed. No issues at all. It might have been harder if they'd all been right on top of one another, but they weren't.

    I should also point out that I had any offensive ECM shut down to take that off the table.

    So overall, a 4-ASM salvo might have been adequate against something like an Adelaide (which, for those playing at home, was a Perry variant), but not against a modern Aegis combatant.

    Just for fun, I also ran the scenario again with a 2015 Adelaide, instead of the 1999 variant I used the first time. It picked them up at 17.2 nm, and managed to get off at least 4 ESSMs (added sometime in the intervening years) and 2 SM-2s, all of which missed despite hit probabilities in the 50s. I think Phalanx nailed one missile and Nulka drew off another. I re-ran it, and the frigate nailed the first three missiles before the fourth one hit. It couldn't do retargeting, and picked them up just a little bit closer in, which limited how many missiles they could get off.

  9. February 27, 2019bean said...

    I tried one more thing, just for laughs. Took a modern Burke, set all the jammers going, and had a Bear doing radar sweeps instead of just giving away the location of the target. Fired 16 missiles at her, then set her on her merry way. Turned out that none of them actually attempted to hit. I think the jamming was just too much, and they mostly passed aft, to the point where I'm pretty sure I fired a lot more missiles (at really unfavorable engagement geometry) than Aegis actually would have in this situation.

  10. February 28, 2019DampOctopus said...

    Thanks for the thorough answer! I didn't know about CMANO; it looks like a perfect tool to test this sort of scenario.

    The article understated the SAM speed (Mach 3 vs Mach 4+ for an ESSM), but it assumed instantaneous acceleration to top speed, which I think balances out on the optimistic side.

    I did notice the implied assumption that the salvo is perfectly synchronised, but I figured it doesn't make a difference unless the spacing between missiles becomes significant relative to the ~60-second engagement time. Which it may, especially if you're co-ordinating multiple launch platforms.

    If the illuminators are still physically-steered antennas they'll be limited by slew speed, which means there's an advantage to attacking from multiple directions simultaneously, assuming you can co-ordinate the salvo. Unless the SAMs use active radar exclusively, and don't need illuminators at all.

    Does infrared plume detection mean that you detect the exhaust of the missile while its body is still below the horizon?

    I guess the take-home message is that Aegis really is that good. Overwhelming an Aegis-equipped ship with a simultaneous salvo of ASMs doesn't work: you need to fire enough ASMs, at whatever cadence, for it to run out of SAMs. (At a PK of .55, that means you'll .55 ASMs for each SAM it has onboard.)

    Or, more fundamentally: ECM is at least as important as SAMs. The idealised scenario of an inbound salvo ASMs can only occur if the attacker knows exactly where the ship is - and your last test, if CMANO is even slightly accurate, indicates that this is unlikely.

  11. February 28, 2019bean said...

    Koop's model is a rather weird one. It makes some assumptions that seem incredibly optimistic (100% Pk) and others that seem incredibly pessimistic (7 seconds between engagements). Normally, you'd expect one or the other, but it's rather GIGO here.

    (Also, it bears pointing out that Air Power Australia is very much a lobbying blog, and everything on it should be taken with many grains of salt.)

    ESSM accelerates really fast. It's an incredible piece of technology, and has really changed the game on missile defense.

    Salvo spread makes a big difference to this specific model. If you see the sort of launch spacings I was getting in CMANO, you essentially go from 1 engagement every 7 seconds to 1 every 5 seconds, a nearly 30% increase. In other words, an extra missile shot down.

    Coordinating salvoes from multiple directions is at a minimum very difficult. If I have some indication that you've just launched at me (say that I pick up the radio calls of the launch), I'm likely to maneuver in an attempt to either (a) leave your missiles staring at a blank patch of sea or (b) change the geometry so that one salvo arrives well before the other does.

    Does infrared plume detection mean that you detect the exhaust of the missile while its body is still below the horizon?


    One area that I think CMANO falls down is in the OODA loop modeling. It requires something like 8 seconds to begin the engagement after the missiles come over the horizon, and I know Aegis is capable of working faster than that when it sees something small come over the horizon at 30' above the surface, to say nothing of the IR signature, which wasn't picked up.

    I won't say it's completely impossible to saturate Aegis with enough missiles, but it's very, very difficult. As for the last test, I have no idea exactly why they didn't go for the ship, as they should have been able to see the ship.

    Actually, doing a more through investigation of missile defense in CMANO might make a good post of its own.

  12. February 28, 2019DampOctopus said...

    Yeah, I did catch that the intended message of the APA article was "Ships are too vulnerable, so we should spend all our money on the Air Force instead.".

    I'm having trouble finding stats for just how fast ESSM accelerates - but given the dates, the SAM would probably actually be the older RIM-7.

    I understand that a back-of-the-envelope model like Kopp's won't be perfectly accurate, but it has the advantage of being more transparent than the CMANO code: easier to see what's going on.

    That said, I think I misinterpreted the model: I thought the interval between SAM launches was the flight time plus 2 seconds for slewing; but re-reading it, it seems he assumed a minimum interval of 7 seconds.

  13. March 01, 2019bean said...

    I’m having trouble finding stats for just how fast ESSM accelerates - but given the dates, the SAM would probably actually be the older RIM-7.

    I'm really not sure, because the mashup of numbers he gives doesn't correspond to any real missile. In 2000, ESSM wasn't in service anywhere, but it was clearly coming. Australia had both Standard and regular Sea Sparrow in use. And I can't find any specific data on the ESSM rocket motor, which would give a good idea of acceleration. That said, it seems to be through Mach 2 very quickly in CMANO, and those people have definitely done their homework.

  14. March 01, 2019John Schilling said...

    ESSM launches vertically, and takes ~3 seconds to pitch over to engage a surface or sea-skimming target. That's probably more important than acceleration, and it might be a weak argument for trainable launchers like the old Mark 29.

    But only a weak argument, because missile flight time is only a factor for the first engagement in a salvo. The spacing between engagements is limited by the time required for the director to shift between and illuminate targets for the terminal phase, and you can launch against targets N+1, N+2, etc, before the director has finished dealing with target N.

    Using Kopp's numbers (which seem approximately right here), you figure out when the first ESSM is going to hit the first target, then schedule your subsequent ESSM launches to hit their respective targets at 7-second intervals. If that means two or three or four missiles in flight at the same time, and only one director, that's what inertial and midcourse guidance are for.

  15. March 01, 2019bean said...

    That seems like a long time for pitch-over for what is a point-defense missile. I'd totally buy that for something like a Tomahawk or even a Standard, but this is neither of those things. Pictures of Burkes firing ESSMs are impossible to find for some reason, but I did find an image of an ESSM launch by, weirdly, an Anzac. Some eyeballing suggests that the missile is going sideways at ~300'. I'd have to back out what that implied about missile acceleration, and I don't feel like doing the math right now. (I also need to reread the article on ESSM in World Naval Weapons Systems, but that's another issue.)

    And I really question the 7 second time interval. 5 seconds seems like a long time for terminal homing, and my estimate would be more like 3 seconds. Which, when combined with a 2-second interval between ASMs, means that you're now looking at over twice the engagements he estimates. (Other factors drag this down in practice, I think, but that's a different issue.)

    He does actually assume that you're limited by illuminators and not by flight time in his math on extra engagements, but I still have doubts. He completely ignores the bit where ships move and can do other things to make the bad guy's life hard.

    Last night, I ran my standard scenario against a pair of Anzacs, one in 1999 trim, the other in 2015 fit. The 2015 one survived unscathed, although they were really lucky, both in missile hits and in decoying off two missiles. The 1999 one, still carrying RIM-7s, died after nailing either zero or one missiles (can't remember which).

  16. March 01, 2019Johan Larson said...

    Have some video propaganda from one of the smaller navies of the world, the Finnish Merivoimat or "sea forces".

  17. March 02, 2019bean said...

    First test of large-scale missile defense in CMANO. I picked 1985 as my year, because I have lots of references for strength of both US and Soviet navies, and then gamed out an attack on a CVBG using a regiment with 24 Backfires each carrying a single AS-4 missile. None of the escorts were Aegis or even NTU. And they didn't get a single hit. Four missiles did try to hit, but all were spoofed. Worse than that, this was basically a matter of terrible missile allocation. The ship that was actually targeted (and that I could tell was targeted) wasn't shooting at those missiles, even though it could, due to a quirk in CMANO's algorithm. I'm sure there's a way to fix this, but it will take work. Overall, I'm fine-tuning my scenarios on this. (It's worth pointing out that the AS-4 isn't sea-skimming, so they were being engaged at long range.)

    I then ran a second test, and fixed the problem with the Backfires shooting at the escorts. Weird things happened Several of the escorts were ineffective because they were slightly offset from the threat axis and their missiles ended up behind the target so they couldn't see the illuminator. I suspect this is one of the things that NTU/Aegis changed, and it almost certainly would have taken the engagement from needing to draw a couple missiles off via SRBOC to being able to keep them comfortably outside the point-defense perimeter. The missiles are still coming in rather staggered, because of limitations on my automated mission-planning skills.

  18. March 02, 2019Johan Larson said...

    In 2003, a Boeing 727 was stolen from an airport in Angola. The aircraft was never seen again.

    I'm having trouble understanding what the thieves were trying to do. Are there parts of the world so remote that one can operate a stolen jetliner without questions being asked? Or can used aircraft parts be sold without extensive documentation?

  19. March 02, 2019Tuna-Fish said...

    Are there parts of the world so remote that one can operate a stolen jetliner without questions being asked?

    Not really.

    Or can used aircraft parts be sold without extensive documentation?

    No. But the people who stole the plane were working to make it airworthy, so they would have access to the documentation. They just needed an unscrupulous buyer who is willing to purchase parts they know to be stolen.

    ... Of course, dealing with such buyers is a bad idea, especially when you basically need to fly the plane to them first. We have already established that they are unscrupulous, why would they pay you in money instead of bullets?

    Which is probably why they thieves disappeared. Assuming they didn't just crash the plane they were not rated to fly.

  20. March 02, 2019bean said...

    First, given that nothing at all has been seen since, I'd guess they went down either in the ocean or in some really remote patch of jungle. And I can't see anywhere allowing them to fly a stolen airliner.

    That said, stripping for parts isn't implausible. Parts with falsified certifications were a serious problem even in the US a few years ago, until a big FAA crackdown, and I suspect that the situation in Africa would be significantly worse.

  21. March 02, 2019bean said...

    Also, I finally got the CMANO scenario working right. 23 AS-4 Kitchen missiles were fired at the carrier within a plausibly short space of time even if the planes weren't right on top of one another. (The 24th was fired before they found the carrier, for complicated reasons.) The group stopped all the missiles, but only just. A tighter salvo would have done better, but the missile defense was far from perfect, too. This is a good base to work from.

    (The group, for those who are interested, is built around CV-66, with two FFG-7s, two Spru-cans, a Farragut-class destroyer, a Leahy and a Belknap.)

  22. March 03, 2019DampOctopus said...

    If the ESSM reaches an altitude of 300' in 3 seconds, that implies an acceleration of 3.1 gees (or a bit more, if some of the thrust is going into horizontal movement). But at that acceleration it would take 45 seconds (and 30 km) to reach Mach 4, which seems like an awfully long time.

    I found another image (from this page) showing two ESSM launches from the same ship. The second one has tipped over at about the same altitude as in Bean's picture, while the first is about 500 metres further along. We might be able to get the acceleration from this if we knew the interval between the launches.

    Also, the ship in this case is another Anzac. Maybe these pictures aren't meant to be published because they give away too much about the ESSM's capabilities, and the Australians are just sloppier than the USN about security.

    Bean, after the previous test I was wondering whether (a) CMANO was too optimistic about defensive ECM or (b) the USN was really more-or-less immune to Soviet Naval Aviation throughout the Cold War.

  23. March 03, 2019bean said...

    Actually, I have a better quantification of acceleration based on this video clip of a Japanese ESSM launch. From the time it starts moving, it's out of the launcher in two frames, and it moves its own length in the third frame. The video is at 30 fps (because it's Japanese, probably) and the missile is 12' long. This gives me an acceleration of 168G (169 if we take gravity into account) and the missile should reach 300' in a third of a second, not 3 seconds.

    Also, the ship in this case is another Anzac. Maybe these pictures aren’t meant to be published because they give away too much about the ESSM’s capabilities, and the Australians are just sloppier than the USN about security.

    I would not be at all surprised if the USN was considerably more paranoid about this than other navies were.

    Bean, after the previous test I was wondering whether (a) CMANO was too optimistic about defensive ECM or (b) the USN was really more-or-less immune to Soviet Naval Aviation throughout the Cold War.

    I'm not 100% sure. There was a 20-year difference between the missile seekers and the Burke's ECM, which probably made a big difference. More recent tests have shown it to be close. The 24-missile salvo was my approximation of a naval aviation regiment, and based on my reading, they wanted 1-2 per carrier. Concentrating three regiments against 2 carriers with the sort of escorts I've implemented would probably result in some missiles getting through. But on the other hand, I'm also neglecting the Outer Air Battle with F-14s and the like. That could make a big difference.

  24. March 08, 2019ADifferentAnonymous said...

    Wondering about a sort of generally fact-check and credibility-evaluation on this piece:


    The headline is IMO a pretty strained reading even if all of the assertions within are true.

    Some key claims I suspect this thread can answer satisfyingly:

    a) Without American trade (all trade, not just arms), Britain would have been starved into surrender.

    b) Stalin's purges weren't so crazy--they hurt the army a lot, but the loyalty of the purgees to the USSR was legitimately questionable.

    c) If Britain surrendered and the US didn't support the Soviets either, Germany would have had a good chance to win on the Eastern Front.

    d) The alternative to the nuclear attacks on Japan was not an invasion but starvation-into-surrender. The main benefit to the US from taking a faster approach was preventing the Soviets from getting a piece of the pie like they did in Korea.

    Also this one, which I'm pretty sure is just false:

    e) "The result of WWII was the redirection of Britain and France’s economic output to America and Germany."

    (This obviously doesn't describe the Marshall Plan, which was USA -> Britain, France, and Germany, with Britain and France receiving more. It sounds like it's supposed to describe the Bretton Woods system, but I don't really see how that adds up)

  25. March 08, 2019bean said...

    I'm not a specialist in the diplomatic history of the war, but it strikes me as about 60% truth interpreted in the most cynical possible way, and 35% stretching the truth as far as you can go without plunging into outright falsehoods. The other 5% is outright falsehoods. Particularly glaring:

    Remember that one of the sore points of interwar Germany was that the German surrender came with the field armies in decent and even advancing strategic position, under pressure from hunger-stoked socialist rebellion on the homefront - the “stab-in-the-back” or Dolchstoß. Of course today you see that often dismissed as a myth, the “Dolchstoßlegende”. Dismissed, of course, by the mainstream historiographers aligned with the regimes which legitimized themselves against the regime that legitimized itself on it.

    I'd hope that in 2019, with the centenary of the war just behind us, we could be free of this sort of nonsense. Germany was beaten in the field, and falling back on the Reich about as fast as the allies could push them back. Austria-Hungary and the Ottomans had signed separate peace accords already. Any attempt to salvage the "stab in the back" is utter nonsense.

    America wanted Japan to surrender while they were still the only ones around to surrender to, rather than face a division with the Soviets like Korea and Vietnam.

    Vietnam wasn't partitioned for a decade, so there was absolutely no fear of that happening. I don't believe the division of Korea was formally agreed upon until December.

    The Russian Navy, by contrast, spent the early 1900s occasionally rebelling against everyone because their institutional continuity had been shattered by the near-total wipeout of the Russo-Japanese war.

    Not quite so glaring, but this is a naval blog that has spent some time dealing with the Tsarist navy, so I'll nitpick it here. Only half of the fleet was wiped out. The Black Sea Fleet, including Potemkin, was untouched by the war, as was most of the high command/staff structure. They had to do work to regain public trust after the war, but it wasn't the degree of discontinuity this implies.

    WRT your points, a) is probably true. Without US support, Britain would have been in a much worse position, and their staying in the war was pretty close-run as it was. b) is not something I can comment on with any degree of authority. c) is questionable. Russians are crazy where the safety of their homeland is involved. I'd guess you'd end up with a long-running low-grade war somewhere near the Urals. I usually answer statements like d) with "Why do you hate the Chinese?" Or, for that matter, why do you hate the Japanese citizenry? Truman had every reason and every responsibility to bring the war to an end as quickly as possible. Today, there seems to be this view that being killed by a nuke is worse than starving to death because the Japanese stole all the rice, a view I can only ascribe to anti-nuclear propaganda and sloppy thinking. By dropping the bombs as he did, he managed to shock the Japanese into accepting Allied terms then and there. (Claims that they were ready to do so and it was allied foot-dragging are utter nonsense of the worst kind.) The alternative is the slaughter throughout the remains of the Japanese Empire continuing. And, yes, more of it falling under Soviet control, which was also bad from a humanitarian standpoint. A surprisingly large fraction of the Japanese captured in Manchuria never made it home.

    Re e), I'm not really the one to ask about that, but I'd guess we were under "most cynical possible interpretation of true facts" again.

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