April 15, 2018

Anti-Submarine Warfare - WWII Weapons

Last time, we discussed detecting submarines, but that was only the start of the process. Once a submarine had been detected, it still needed to be destroyed. At the start of the war, the only weapon was the venerable depth charge, essentially an explosive thrown over the side and fuzed to go off at a specific depth. However, as the war went on, the traditional depth charge, unchanged over the previous 20 years, became increasingly inadequate, and new and improved weapons were developed.


K-Gun with traditional "ashcan" depth charge

One of the biggest problems, even during WWI, with the traditional depth charge rack was that it only produced a narrow pattern. Submarines could estimate when a ship would lose contact, and then maneuver violently, hoping to dodge the pattern. During WWI, the USN developed the Y-gun, a device mounted on the centerline which threw two depth charges between 50 and 80 yards on either side of the ship. The biggest problem was that it consumed valuable centerline space, which was also in demand for guns and torpedo tubes. To solve this problem, the USN introduced the K-gun, which was essentially half a Y-gun, mounted on the railings. These were mounted in great numbers on destroyers and destroyer escorts during WWII.


Mk 9 depth charge

As submarine technology advanced, another problem arose. Depth charges sunk relatively slowly (8-9 ft/sec), which meant that a deep submarine had plenty of time to dodge, even if the pattern was expanded by the use of K-guns. The first solution was to raise this about 50% by reducing the warhead by a third and adding lead weights. However, this was not enough, and the USN soon developed the Mk 9, which had a teardrop shape and a sink rate of 22 ft/sec. Most of the improved depth charges also had their hydrostatic pistols modified to include deeper settings, and by the end of the war, a 600 ft maximum depth was standard, twice what it had been in 1939.


Hedgehog

Of course, the ideal weapon is one that would attack targets ahead of the ship, before they passed into the sonar's blind spot. These were primarily developed by the British, and the first to reach active service was the Hedgehog. Hedgehog was a spigot mortar, firing a total of 24 65-lb projectiles at a fixed angle ahead of the ship. The 35 lb of explosives in each projectile would have been inadequate as a depth charge, but the projectiles were fused to go off on contact with the target, where it was more than adequate. A second bonus of using contact fuses is that, unlike conventional depth charges, a failed Hedgehog attack did not cloud the water with spurious echoes from clouds of bubbles. The US developed a rocket-powered version, known as Mousetrap, for ships which could not withstand the substantial forces generated by a Hedgehog firing. In British service, Hedgehog sank 47 submarines in 268 attacks, nearly 10 times the success rate of depth charges.


A Squid on HMCS Haida

The British also developed Squid, a more advanced weapon that threw a trio of 390-lb depth charges 275 yards ahead of the launcher, where they formed a triangle 40 yards on a side. They were fused like depth charges, but the sonar system automatically updated the depth settings until the moment of launch, reducing the risk of getting the charges at the wrong depth.1 Ultimately, Squid sunk 13 submarines in 50 attacks. The British preferred the Squid, the Americans the Hedgehog, probably as a result of the Americans having generally better sonar operating practices.


Mk 24 FIDO

The most influential ASW weapon developed during the war was very different. It was an aircraft weapon, never used from surface ships: The Mark 24 mine. This weapon, also known as FIDO, was not a mine of any sort, and was only called that as a cover. It was instead an air-dropped homing torpedo, a truly remarkable weapon for the time. 19" in diameter and weighing 680 lbs, it was a viable substitute for the depth charges which had previously been the only option for aircraft attacking underwater targets. After being dropped into the water, it would circle, four hydrophones in the nose listening for the target. If it detected propeller sounds, it would steer towards them, attempting to equalize the signal between both the left-right hydrophone pair and the up-down hydrophone pair. The 340 torpedoes dropped by American and British aircraft were responsible for sinking 68 submarines and damaging 33 more, an incredible success rate for the time.


Rockets being loaded onto a Fairey Swordfish aboard an escort carrier

Depth charges remained an important part of the aircraft weapon inventory, although a serious limitation was the inability for most aircraft to set their fuses after takeoff.2 A submarine on the surface might be able to ride out a depth-charge attack, as the charges would be fused against a diving submarine. The primary weapon used against surfaced submarines was the 30-lb rocket, usually fitted with a solid-steel "warhead" to pierce the heavy pressure hulls of submarines. These would be fired just short of the target, hopefully punching holes below the waterline. Rockets were also used on the "retrobombs" fitted to MAD-equipped aircraft. MAD only detects targets directly below the airplane, which is a problem because bombs continue to travel forward with the plane's momentum after they're dropped. A retrobomb was fitted with a forward-facing rocket to more or less stop it in place, allowing it to hit MAD contacts.


Depth charge track containing Mk 9 depth charges3

So far, we've covered the material factors of the anti-submarine war. However, these are only half of the story, and other, less tangible factors played an equally large part. Next time, we'll look at the methods the British came up with to make most efficient use of their forces.


1 The Japanese famously set their depth charges too shallow until mid-1943, when an idiot in Congress revealed this in a press conference, and other idiots published it.

2 I believe the Short Sunderland was an exception, as the bombs were stored internally and cranked out onto the wings by the crew.

3 My photo, from USS Cassin Young in Boston.


For more on WWII ASW, see Hunters and Killers, Volume 1 and Volume 2.

Comments

  1. April 16, 2018ADifferentAnonymous said...

    Wait, what? This is the first I'm hearing about homing torpedoes in WWII. Why wasn't the same system used for ship-to-ship torpedoes? Or, if it was, why wasn't it paradigm-shiftingly effective?

  2. April 16, 2018Directrix Gazer said...

    @ ADifferentAnonymous

    The speed of the Mk 24 was only 12 knots. That's not a problem for a weapon designed to be dropped relatively close to a submerged diesel-electric submarine, but it pretty much eliminates the possibility of being used against surface targets. There was a (slightly) modified version of the Mk 24, designated Mk 27, made for use against surface vessels by submarines, but it wasn't any faster and was thus useful only against slow moving, un-alerted targets at close range.

    The Germans also developed a series of acoustic homing variants of their G7e electric torpedoes. These were faster at around 20 knots (IIRC), and had some success when used against escorts, but it was... far from a perfect weapon, especially in the guidance department. The Allies, of course, rapidly introduced a towed decoy called Foxer which was pretty successful. There are some fun stories about Foxer's development, but I'll let Bean tell them.

    The main reason the early homing torpedoes were relatively slow was to reduce flow noise, which otherwise would have deafened their hydrophones.

    Incidentally, the US and German acoustic homing torpedoes were introduced within a few months of each other in 1943.

  3. April 16, 2018bean said...

    Directrix Gazer pretty much nailed this one. They didn’t have the tech to home at high speed, which limited the utility of the torpedoes quite a bit. They were also more complex. The Mk 27 Cutie’s average success rate is pretty much identical to that of contemporary conventional torpedoes. I suspect that some of this was the increased complexity of the equipment. They also weren’t particularly easy to use, as you had to get in really close, and I suspect there were reliability problems. I don’t know that much about G7e.

    By the end of the war, they were getting close to 28-kt acoustic torpedoes, but the tech wasn’t quite there yet.

    @Directrix Gazer

    Feel free to tell the stories of Foxer. I have memory of some vague hijinks, but nothing specific, and I don’t see Naval Gazing covering that any time soon.

  4. April 17, 2018ADifferentAnonymous said...

    Ah, got it.

    Also, I assume the Mark 24's electronics were way too delicate to fire out of a gun, but I kind of love the idea of Iowa firing a salvo of those at a crash-diving sub 30000 yards away.

  5. April 17, 2018bean said...

    Also, I assume the Mark 24′s electronics were way too delicate to fire out of a gun, but I kind of love the idea of Iowa firing a salvo of those at a crash-diving sub 30000 yards away.

    I’m not sure if I’m really amused or horrified at that image. Besides the obvious problem that it would be pulped, it’s also 19″ in diameter, which means that you couldn’t fit it into a 16″ gun. The obvious solution is to use rockets.

  6. April 17, 2018ADifferentAnonymous said...

    Bah, don't pretend you're any kind of ambivalent about turning battleships into ASW powerhouses :p

  7. April 17, 2018bean said...

    You're right. I'm not. Hunting submarines is beneath the dignity of a battleship. That's what destroyers are for. A noblewoman does not go around hunting dishonorable rogues, but leaves it to her servants.

Comments from SlateStarCodex:

  • bean says:

    Naval Gazing wraps up the current look at WWII Anti-submarine warfare with weapons. I’ll return to look at some of the operational aspects at some point down the road. (But it’s going to be quite a while, because I have lots of other things to do first.)

    • John Schilling says:

      You mention the inability to change fuze settings on aircraft depth charges. This lead to an early victory for Operations Research, aka using lots of rationalist-type math to solve real-world problems. The obvious depth setting for an aircraft depth charge was 100 feet, on the grounds that an attacking aircraft would be spotted an average of four miles away and would have time to crash-dive to an average depth of 100′ while the aircraft made its attack run. But almost no submarines were ever sunk or seriously damaged that way.

      Enter Dr. E.J. Williams, Fellow of the Royal Society and previously known as a top boffin in particle physics. Told do think about something useful for a change, he analyzed every reported aerial depth charge attack for which data was available, applied math, and had the depth settings changed to a ridiculously shallow 25 feet. The rate at which submarines were sunk by aerial depth charge attack (normalized by number of attacks), increased by a factor of seven.

      Exercise for the student: Why, aside from “because the math said so”, did this work?

      • albatross11 says:

        It sounds like the airplanes that were successfully sunk by aerial depth charges weren’t trying to evade–maybe when the sub detects the airplane four miles away, it has enough time to evade that there’s little chance of hitting it even if you get the right depth setting. But a sub that doesn’t see the airplane will stay in place and be easier to hit.

        This is just a guess, though–I know nothing at all about naval stuff.

      • bean says:

        Blast it, John. I was saving that one for later. (Seriously, I was planning to use exactly that for the operations research in WWII ASW post, and will this refrain from answering.)

      • sfoil says:

        I don’t know much about this subject, but my guess is: 25′ is deep enough to accomplish the sub’s purpose of not being seen from above unless the water is ridiculously clear, so instead of diving as deep as possible (to 100′), they only dove to 25′. Going out on a limb here since I know nothing about performance characteristics, perhaps subs preferred to dive to 25′ and maneuver than to dive 100′ and not maneuver. Perhaps Dr. Williams’ math showed that submarines could not be spotted at much shallower depths in non-ideal circumstances.

        • bean says:

          No. Diving submarines is surprisingly tricky, and if you want to go down fast, you’re going down a long ways. It’s basically impossible to dive quickly to 25′. (Which, incidentally, would leave you visible. The submarine has a height dimension, and 25′ would leave parts of it very close to the surface, definitely close enough to give it away.

      • SolveIt says:

        Presumably submarines that managed to get to 100ft were well-nigh impossible to sink, hence the

        But almost no submarines were ever sunk or seriously damaged that way.

        So the only sinkable targets were those that, for whatever reason, didn’t manage to get that deep, but by setting the depth of the charges to 100ft, the aircraft wouldn’t be able to sink those either. Changing the settings to 25 ft fixed that problem.

        tl;dr although most subs are at 100 ft, the subs at 25 ft are so much easier to sink that weighting for sinkability, most of the sinkable mass is at 25 ft.

        • bean says:

          That’s half of it. Yes, the problem was that a target at 100′ was basically immune, and the charges set for 100′ were too deep to kill a 25′ target. The next half is why this was.

          • yossarian says:

            Maybe at 100′ the submarine is too deep to be noticed from the plane, so the fliers would basically waste the bombs dropping them at the guessed last location they saw, while at 25′ the submarine can still be detected, so the fliers, knowing they have their depth meters set to 25′, would only bomb still-visible submarines and refrain from wasting ammo when the sub is not visible?

          • bean says:

            That’s closer, but there’s still a key component that you’re missing.

          • Andrew Hunter says:

            I’d expect the CEP to go up dramatically higher with depth, as a coupled function of two facts:

            – depth charges presumably drift due to currents, etc, and have more time to do so as they go deep.
            – a deeper sub is a sub that submerged longer ago, and thus has had more time to get away from the plane’s best estimate of location.

            We’re trying to max \int_{D} P(sub at depth D) * P(sub killed at depth D | we set charge at depth C.) The prior for sub depth isn’t under our control, and John’s story doesn’t seem to deny the fact that most subs would be deep, so you seem to be asking for why P(sub killed at 100 | charge at 100) << P(sub killed at 25 | charge at 25), and unless e.g. water pressure matters here (doubtful) that basically has to cash out in CEP, no?

          • christhenottopher says:

            Total guess here:

            At depths of 100 feet the pressure of the water contained the blast radius of the depth charges more, shrinking the area they could threaten and making the sub harder to sink.

            Other guess:

            Visibility is the same but the area the sub can be is larger making a hit harder. At 25 ft the sub has only been diving for 1/4th as long, thus having less time to manuever away from the airplane’s path.

          • bean says:

            Andrew, your second one is the correct insight. Target errors dominate weapon errors in most ASW.

          • bean says:

            @Chris

            At depths of 100 feet the pressure of the water contained the blast radius of the depth charges more, shrinking the area they could threaten and making the sub harder to sink.

            This is not a thing. Depth charges probably get more effective at depth, as the hull is under higher stress to start with. At reasonable depths (not at the bottom of the Marianas Trench) the depth charge is unaffected by pressure.

            Visibility is the same but the area the sub can be is larger making a hit harder. At 25 ft the sub has only been diving for 1/4th as long, thus having less time to manuever away from the airplane’s path.

            This one is exactly right. Submarines would hit the rudder when they went under, which meant that in 4 minutes/100′, they could be anywhere in a very large area. After only 30 seconds-1 minute (I’d have to look up the exact times) the submarine’s position was still known well enough for a strike, but it was too shallow for the 100′ charges to be effective.

      • johan_larson says:

        Airplanes only find submarines that are on the surface. Also, aircraft are hard to spot, so submarines that get found typically have little or no time to dive. It therefore makes sense to set depth charges shallow.

    • Montfort says:

      I think there is a distinction between “traitor” and “extremely reckless with sensitive information,” and unless you have evidence that either May or the newspapers intended for Japan to catch wind of the information on depth charges, the latter seems more appropriate. His war-profiteering seems much more disloyal than his indiscretion.

      • bean says:

        At the very best, it was depraved indifference to the lives of our submariners, and everyone involved should have ended up in jail for a very long time. Except the censors who approved publication, who should have been shot. But I probably should fix the wording.

    • gbdub says:

      I’m not quite sure what you meant by this: “The British preferred the Squid, the Americans the Hedgehog, probably as a result of the Americans having generally better sonar operating practices.”

      Why would “generally better sonar practices” lead one to prefer one system over the other?

      • bean says:

        AIUI (and this is a case where several sources said the same thing, but didn’t really explain it) the US managed to get their sonar working well with Hedgehog before the British did. I don’t know how much was procedural and how much was technical, at least not offhand. The US evaluation of Squid was that it wasn’t any better than Hedgehog, but I’d guess the British got their operational methods/gear right for it, in a way they didn’t for Hedgehog.

    • Urstoff says:

      I saw a picture of the USS Lexington (CV-2) recently and thought its superstructure looked kind of odd. Why are there two distinct superstructure towers rather than just the one as on later carriers?

      • bean says:

        I’m not entirely sure. One aspect is that Lex and Sara were really high-powered compared to later carriers, which means getting rid of lots of gas (The aft one is just a funnel). Another is that they had existing infrastructure to work around, which may have constrained what they could do.
        If I remember, I’ll check the relevant Friedman, but do remember that Lex and Sara were still somewhat experimental. Wildly successful experiments, but a lot wasn’t known when they were built.

      • cassander says:

        There were some connections between the two towers on the lower levels. Bean is right to point out that they were experimental ships, converted from partially complete battlecruisers, so they might not have been able to line everything up exactly. But If I had to guess, I suspect the spacing came about as a way to improve visibility around the large exhaust stack. Possibly for AA fire as well, but I since the lexington wasn’t originally equipped with light AA, that seems somewhat unlikely. Light AA was added almost immediately after completion, though, so maybe someone had thought about it.

      • bean says:

        Speculation: Lexington, unlike other US carriers, was designed with a substantial surface armament. She thus had a full heavy-gun fire control suite. Those are mounted high up to ensure maximum range. But that often causes trouble with funnel gasses, particularly in a ship as powerful as Lady Lex. So they had to separate the bridge (with mounted director) from the funnel. Later carriers had only 5″ guns, which can have lower directors.

        • Urstoff says:

          Thanks for answering! Was its being so powerful an artifact of it starting as a cruiser?

          • cassander says:

            They were battlecruisers, not mere cruisers. And being so powerful was was why they selected for conversion to carriers. Speed is very important for carriers, because it makes both landing and taking off easier. That said, there was not a small amount of resistance to carriers that were so large in the navy. The lexingtons were about twice the size of any existing carrier in the world, far bigger than was thought helpful, especially since total carrier tonnage was limited by the treaty that resulted in their conversion. Teddy Roosevelt junior (who was an assistant Navy secretary) had a clause slipped into the treaty against the wishes of some admirals that allowed for their conversion knowing that congress would insist on the conversion if it were an option.

            As it turns out, though, bigger was basically better for carriers. Larger carriers were more stable in rough seas, were better able accommodate aircraft that grew rapidly in size and weight (which meant longer takeoff runs), and the power of a strike grew geometrically with air group size.

          • bean says:

            Pretty much what cassander said. They were some of the most powerful ships of their day, and absolutely enormous by contemporary carrier standards. Which turned out to be a really good thing, because they could keep up with growing airplanes, and their contemporaries couldn’t.

            the power of a strike grew geometrically with air group size.

            That’s not quite right. They were a bit too big for the airplanes of the early/mid-20s. Once you pass 90-100 airplanes, you can’t operate them effectively, and a smaller ship would have been more useful. The Midways had the same issue in the late 40s. But the planes grew into both ships.

          • cassander says:

            That’s not quite right. They were a bit too big for the airplanes of the early/mid-20s. Once you pass 90-100 airplanes, you can’t operate them effectively, and a smaller ship would have been more useful. The Midways had the same issue in the late 40s. But the planes grew into both ships.

            Hah! I actually wrote a sentence about this problem referencing the midways, but couldn’t remember if it had been a problem for lex and sara too or if I was just conflating them with the midways.

          • Eric Rall says:

            Speed is very important for carriers, because it makes both landing and taking off easier.

            I imagine speed is also important because Fisher’s dictum that “Speed is armour” applies to carriers. A carrier isn’t designed to stand in the line of battle against large surface warships(*), so unless her air wing can reliably sink fleets from outside of gun range, a carrier needs to be able to hold the range open against anything her escorts can’t handle. And until well into WW2, the effectiveness against carrier air wings against heavy surface fleets was an open question.

            (*) I suppose you could probably design a carrier to stand in the line of battle, but that would be silly because adding armor makes a worse carrier (more tons of armor == less tons of aircraft, etc), and because sticking around while a surface combatant closes on you gives up the huge advantage that carriers can hit ships at ranges where they can’t hit back.

          • bean says:

            I imagine speed is also important because Fisher’s dictum that “Speed is armour” applies to carriers.

            That’s not quite what “speed is armor” meant. It was specifically a matter of beliefs about fire control, and the ability of speed to thwart enemy solutions. That was obsolete by WWII.

          • Eric Rall says:

            That’s not quite what “speed is armor” meant. It was specifically a matter of beliefs about fire control, and the ability of speed to thwart enemy solutions.

            I remember hearing about that (probably from you), but I thought “speed is armour” also referred to the ability of the vessel or fleet with a substantial speed advantage to dictate the terms of the engagement: choosing the range, choosing when to break off or continue, etc; but a slower ship would need to be tough enough to fight an engagement on disadvantageous terms.

            I’m pretty sure Fisher saw dictating the terms of the engagement as a key benefit of faster ships, but I might be mistaken in conflating it with the “speed is armour” slogan.

          • bean says:

            I’m pretty sure Fisher saw dictating the terms of the engagement as a key benefit of faster ships, but I might be mistaken in conflating it with the “speed is armour” slogan.

            That was part of his thinking, yes. One problem with trying to figure out Fisher is that he changed his mind every few years, and even when he was being consistent, he certainly wasn’t above saying what he thought people wanted to hear.

      • AlphaGamma says:

        Note that the new British carriers are also built with two islands- bridge in the forward island, flight control (and a backup conning position for emergencies) in the aft one.

        Various reasons have been given for this, including the ability to site both in the ideal place for the best view without having to compromise, better routing for exhaust trunking (they have two fully independent gas-turbine systems, each with uptake and downtake in one of the islands), reduced turbulence over the flight deck, and better separation of radars.

        Of course, there are also disadvantages like more reliance on intercom to coordinate between OOW and flight controllers.

        • bean says:

          David Hobbs, who should know if anyone does, says it was primarily trunking, which makes sense. Gas turbines require a lot of air.

    • Andrew Hunter says:

      A random question: is it not possible to build AESA sonar, and if so why not? Too much frequency dispersion in water waves as compared to radio (my best guess, but I don’t know any of the realities.)

      Every discussion of sub warfare points out that active sonar reveals your location. But we can build active (so to speak) LPIR systems by–well, as more of an information theorist than a radar guy, what I’d call variants of CDMA.

      In principle wave superpositions should work in audio to make interesting phased array chirps too, no? The only reason I can think of this isn’t a thing are a) too much decoherence somehow to maintain the desired interference patterns b) inability to build the transmitters, but I don’t actually know anything here.

      • cassander says:

        My understanding is that existing sonar systems work more like a PISA, but have transmitter/receivers that are individually wired and thus could be individually controlled given sufficient software and processing technology, and that the latest generation sonars are headed somewhat in this direction, but the sub community is a lot more tight lipped than the air force is, and “advanced signal processing” could really mean a lot of things.

      • bean says:

        Phased arrays have been standard in sonar since the end of WWII. I’m not sure that an AESA/PESA distinction makes much sense in a sonar context, but I’m not an electrical guy. So far as it does, sound is easier to play games with than radio, so they’ve probably been there for a couple of decades. I’d guess that they’ve done what they could on LPI, but water is a lot messier to sonar than air usually is to radar. Probably coherence plays a major part.

  • Leave a comment

    All comments are reviewed before being displayed.


    Name (required):


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

    Website:

    You can use Markdown in comments!


    Enter value: Captcha