Tuesday, July 4, 2017

Torpedo Lethality Myth

This post is going to ruffle a few feathers.  It should be fun!

Let's see a show of hands.  How many of you think a torpedo kills a ship by breaking its back due to suspending the ship over a giant bubble of air?  Most of you raised your hands and the rest started to but hesitated because they sense a trap coming.

There is a widespread school of thought that a single torpedo hit spells instant doom for any ship in the world, no matter its size.  Thus, proponents say, there’s no sense applying armor to a ship – it would be pointless.  In fact, many of these people believe that adding armor increases the weight of the ship and, due to the greater weight, makes the ship more prone to breaking its back when suspended over the gas bubble formed by a torpedo explosion.  Presumably, these same people would advocate the thinnest sheet metal covering on a hull that is sufficient to keep out the ocean during normal sailing.  By logical extension, one would also have to assume that these people see no point in damage control measures because a single torpedo is an absolute guarantee of a sunk ship.

Well, this concept could not be more wrong and it’s time to learn why.  First, we’ll take a look at the characteristics of underwater explosions.  Then, we’ll examine the damage mechanisms associated with underwater explosions.  With that base of understanding, we’ll look at the specific case of a “broken back” by a ship suspended over an explosion created bubble.  Lastly, we’ll examine torpedo damage mitigation measures.

The foundation of this post is a review of scientific papers on the subject of underwater explosions.  Note that this is a blog post not a thorough and comprehensive review of every piece of scientific data out there – that would require a book length piece of writing.  That said, I’ve reviewed numerous papers and selected and referenced the ones that best illustrate the various relevant concepts.  I am also forced to summarize and, to a degree, simplify the scientific data for the sake of brevity and clarity.  For example, few of us are trained to understand the advanced mathematics contained within these types of papers nor do we care.  We are interested in the results – hence, my summations.  I’ve cited the references so you can peruse them if you are so inclined.

Underwater Explosion Characteristics and Behavior

An underwater explosion manifests two major effects:  an initial shock wave and a gas bubble. 

A gas bubble is created due to the formation of hot gaseous byproducts of the explosive chemical reaction.

“The underwater detonation of an explosive charge can best be described as an exothermic chemical reaction that is self-sustaining after initiation. Forming throughout the detonation process are gaseous reactive components that are at an extremely high temperature (approximately 3000 degrees Celsius) and pressure (approximately 50000 atmospheres). The entire detonation process represents a rapidly propagating reaction, with propagation speeds in the neighborhood of 25000 feet per second.”

As the gas bubble slowly forms (slow, on a scale of seconds), a shock wave propagates outward in all directions through the surrounding water.  The shock wave propagates quickly (fast, on a scale of milliseconds), relative to the gas bubble formation.

Shock wave pressure begins at a peak value and decays exponentially over time.  For example, a 250 lb HBX-1 explosive charge detonated at a distance of 50 ft from the target measurement point has a peak value of about 2500 psi and decays exponentially down to a value of about 850 psi in 0.62 milliseconds. (1)

After the shock wave passes, the gas bubble forms, expands due to the temperature and pressure of the enclosed gases, overexpands due to momentum, and then collapses back in on itself.  Similar to the overexpansion, the bubble over-collapses (over compresses the gases) and reforms and re-expands.  This cycle of expansion and collapse of the bubble occurs several times, each time less energetically, until either the entire bubble reaches the surface (it rises vertically the entire time since, like any bubble, it is less dense than the surrounding water) and vents or, if the explosion was deep enough, the bubble’s energy is dissipated and the bubble collapses a final time.

Each expansion/contraction cycle of the bubble generates an additional pressure wave (as distinct from a shock wave), the first, and largest, of which can be 10-15% of the peak pressure of the initial shock wave. (4)

There are secondary effects, as well, such as surface layer shock wave reflection, ocean bottom shock wave reflection, bubble-rigid surface jet effects, internal bubble reflective shock waves, etc., but from a ship damage perspective, these are usually of lesser import.

Keil presents a nomograph of maximum bubble radius as a function of explosive charge weight and depth of the explosion.  For explosions typical of a torpedo, say 500-1500 lbs charge and 30-50 ft depth, the resulting maximum bubble is 50-60 ft diameter. (5)  The bubble size is relatively insensitive to charge weight and depth within the range of expected torpedo charges and depths.  The most common torpedo charges and depths tend to produce a bubble around 50 ft diameter or a bit less.

Underwater Explosion Damage Mechanisms

Understanding the basics of an underwater explosion, we can now ask, what is the damage mechanism towards a ship?  According to Wardlaw and Mair (2),

“The 1D [ed.- one dimensional; a modeling technique] explosion exhibits two important damage mechanisms: the initial shock, and subsequent pressure pulses from bubble collapse and rebound.” (2)

Best (3) discusses the possibility of cavitation damage from high speed liquid jets that form on the side of a bubble opposite a solid surface and compress the bubble to a non-spherical form and eventually contact the solid surface.  The magnitude of this effect, if it holds in large underwater explosions, is unknown and it should be noted that this mechanism only applies if the bubble is in direct contact with a solid surface (ship’s hull).

Keil notes several damage mechanisms: (5) 

  1. Initial direct blast damage from the explosion itself if the explosion occurs in contact with the ship (torpedo or mine contacting the hull).  The size of the resulting hole and extent and degree of damage is a straightforward comparison of the explosive kinetics (crudely, charge size) and the various yield, tensile, sheer, and other properties of the ship’s plating (generally steel).
  2. Damage from the initial shock wave.
  3. Damage from the subsequent bubble pulses (cyclical expansions and contractions).
  4. Damage from the bubble water jet.

The magnitude and relative contribution of each type of damage is dependent on the location and depth of the explosive charge.

Keil (5) goes on to describe the mechanism of failure of the ship’s structural members.  The mechanism is one of sequential elastic flexing and relaxation of the strength members of the ship (bulkheads and longitudinal members) in response to the various shock and pressure waves.  If certain structural properties are exceeded, a permanent deformation of the hull structure will occur.  If those properties are exceeded by a sufficient amount, the deformation (flexing) cannot be recovered (relaxation) and the structural members tear – the iconic broken back scenario.  This is conceptually identical to the phenomenon of scoring a piece of metal and flexing it back and forth until it cleanly snaps.  Thus, the broken back is seen to be the result of repeated flexing of the structure.

Broken Back Scenario

Now that we understand the basics of underwater explosions and the associated damage mechanisms, let’s look at the widespread notion of a torpedo breaking a ship’s back by suspending the ship on a bubble of air.  For ease of typing, let’s hereafter call this the air break phenomenon.  Here is a conceptual illustration of the phenomenon.

Torpedo Back Breaking Myth

We’ve already noted that the damage mechanisms are direct explosive effects and shock waves of various origins.  I have not found any mention in any scientific examination of underwater explosions and damage mechanisms of the air break phenomenon.  Instead, the broken back phenomenon is explained by the rapid, repeated, elastic deformation (flexing and relaxation) of the ship’s structural members or the instantaneous application of shock wave pressure that far exceeds the structure’s various strength properties.

Still not convinced?  Let’s apply basic logic and see where that takes us.

First, a vessel must be just the right size and construction to even be susceptible to the air break phenomenon.  For example, a canoe can be lifted at each end and suspended in air indefinitely with no ill effect.  A Cyclone class PC (180 ft long) can be suspended and moved on slings near the ends of the vessel with no ill effect.  A super tanker or super carrier is too heavy to be “lifted” by a bubble of air.  So, in order for the air break phenomenon to occur, the ship must be bigger than a PC and smaller than a large tanker or aircraft carrier.  That would seem to limit the phenomenon to a destroyer size ship.

Let’s consider further the concept of suspending a ship over a bubble of air and breaking its back.  Intuitively, we all recognize that a one inch bubble of air under the hull of a ship isn’t going to break the ship’s back.  Why is that?  Why do we intuitively believe that a ship is immune to breaking its back over a one inch bubble of air?  It’s not just intuitive, either.  Ship’s encounter bubbles of that size under their hulls all the time from wave action, wake effects, etc. and don’t sink.  So, not only do we intuitively know a one inch bubble can’t break a ship’s back, empirical evidence proves it. 

Back to the question – why do we intuitively know a one inch bubble can’t break a ship’s back?  It’s because we understand, without needing any engineering calculations to back it up, that a one inch bubble doesn’t “suspend” enough of the ship’s hull to cause a problem.  The ship’s structure is sufficiently strong enough to withstand the stress of being “suspended” over a one inch bubble.  So, bubble size must be important in this purported phenomenon.  What about a one foot bubble?  No, that won’t break a ship’s back.  What about a ten foot bubble?  Hmmm …  No, that doesn’t seem likely.  Well, what size would cause a problem, then?  A hundred foot bubble?  Two hundred feet? 

Hey, while we’re speculating about the bubble size, I wonder how long a destroyer size ship is?  Well, a Burke is a touch over 500 ft and an LCS is around 380 ft.

Wait a minute!  If we’re going to “suspend” a destroyer size ship over a bubble and break its back, we need a bubble that nearly spans the length of the ship, right?  That means we need a near 500 ft bubble to break a Burke and a near 380 ft bubble to break even an LCS.  Wait …….  Wait …… I’m vaguely recalling a key piece of information from earlier ….

Didn’t we note earlier that for typical torpedoes (charge and depth) the resulting maximum bubble size was on the order of 50 ft?  Yes!  Yes, we did.  Is 50 ft enough to beak a ship?  Well, 50 ft is only 10% of a Burke’s length.  Does suspending 10% of a ship’s hull seem like it would cause instant, fatal damage?  No, that doesn’t seem believable.  Even for the LCS, a 50 ft bubble is only 13% of the ship’s length.  For a one thousand foot carrier, a 50 ft bubble is only 5% of the ship’s length.

I’m starting to think that the air break phenomenon may be a misconception.  The utter lack of scientific mention and the failure of the logical analysis suggest that the widespread belief that a ship’s back is broken by a bubble of air is false – a myth.

The final piece of the logical analysis is the videos we’ve seen of ship’s breaking in two during a torpedo test.  Going back over those videos, what we’ve failed to note is that the ships are thrust up, out of the water with their backs already structurally broken in an inverted “v” shape.  In other words, the structural back of the ship was deformed by upward pressure (or direct blast effects or the initial shock wave) not downward pressure as the air break phenomenon would mandate.  The broken back was not due to suspension over a bubble but by weak structural members deforming and snapping due to initial pressures, most likely the initial shock wave or direct blast effects.

There is no such thing as an air break phenomenon.  A torpedo cannot break a ship’s back by suspending the ship over a bubble.  A torpedo can certainly break a ship’s back but it’s not by suspending the ship over a bubble!  It’s from simple pressure effects causing deformation to the structural members that exceed the structure’s ability to resist or recover.

Having settled that question, let’s now look at the corollary.  Many people believe that a single torpedo is instant, unstoppable death to any ship of any size. 

Lethality and Mitigation

We’ve debunked the air break myth but there’s no denying that torpedoes are powerful and, often, deadly but are they instant death for any ship?  The answer to this is that the degree of lethality is almost wholly dependent on the size of the ship – the bigger the ship, the more resistant it is.  History bears this out irrefutably and I’m not going to waste much time on it.  The interested reader can peruse the various histories of ship sinkings to ascertain this for themselves.  A large tanker or super carrier cannot be sunk by a single torpedo or even a few.  It would take several, at least, to do the job.  Conversely, a destroyer (Burke) size ship might sink from one but would likely require two hits.  Smaller ships are likely single hit sinkers.

The more interesting and relevant question is whether anything can be done to mitigate torpedo damage.  Again, the “torpedoes can’t be stopped and are instantly fatal” crowd believes there is nothing that can be done to mitigate torpedo damage, so why even try?  Of course, nothing could be further from the truth.

Now that we understand the torpedo damage mechanism and the structural failure mode of the target ships, we can begin to develop torpedo resistant ship designs.  Note that this is not the same as “torpedo proof”.  It merely means that we can lesson the resultant damage from a torpedo hit just as we can lesson the damage from any other kind of weapon on land, sea, or air.  There’s nothing uniquely unstoppable about torpedoes.

Setting aside the active and passive torpedo defenses, there are design modifications that could and do impart inherent torpedo resistance.

Keil noted the use of bubble curtains to mitigate the effect of shock waves (5).  US Navy ships already have bubble curtains of a sort in the form of the Prairie/Masker quieting system.  Adapting Prairie/Masker to torpedo defense would not seem terribly difficult.

Keil also suggests that considerable underwater explosion damage resistance can be achieved by designing in a large degree of elasticity into the ship’s structure and plating as opposed to attempting to resist damage via increased hardening (5).  This illustrates the concept of designing the ship to absorb torpedo damage rather than trying to resist it.

On a related note, in hull panel testing, Rarnajeyathilagam and Vendhan (4) noted that concave panels offer better resistance to shock loads.  Thus, varying the geometry of the ship’s hull (round versus flat versus v-shape, etc.) offers a possibility of mitigating underwater explosion damage.  A concave v-shape hull may, then, mitigate underwater explosion damage.

A reasonable extrapolation of the concave geometry induced variation in shock resistance is the expectation that the degree of shock wave induced damage is dependent on the angle the shock wave strikes the target.  Just as a shell is more likely to ricochet from an angled hit or a radar wave scatters and reflects when hitting an angled surface, so too, does it appear that a shock wave is scattered and mitigated when striking an angled surface relative to the incident direction of the wave.  Thus, a flat bottomed ship design would seem to be the worst possible design for resisting under-hull shock waves.  Again, a curved or v-hull of some sort would seem to offer a degree of mitigation.

To belabor the point, a v-shaped hull on land vehicles is proven to mitigate underbody explosive effects.  An underwater explosion follows the same laws of physics as an explosion on land/air.  Yes, some properties are different, notably the density of air versus water, but the behavior is still governed by physical laws.  Just as v-hulls deflect land/air explosive forces, so too have underwater explosive forces been proven to be mitigated by properly shaped hull plates.  Thus, there is every reason to believe that a v-shaped ship’s hull would offer a degree of protection from underwater explosions.  Whether the degree of protection is sufficient to warrant any adverse effects on the ship’s overall seakeeping is unknown.

Void spaces, fluid filled tanks, and collapsible spaces have long been known to mitigate torpedo damage and ought to be a designed-in aspect of every warship.

Increasing the number and strength of the longitudinal structural members of a ship would greatly increase the overall resistance to shock.  Each longitudinal member acts as a mini-keel, tying the length of the ship together and transmitting the shock loads across the length of the ship rather than trying to resist the shock in just a few, localized spots.  Thus, even if the torpedo punches a hole in the hull, damage and flooding would be localized as opposed to breaking the ship’s back and outright sinking.

Armor belts and armored decks act as keels in that they are longitudinal structural members.  Large ships like battleships, carriers with armored flight decks, and heavy cruisers with armor belts and armored decks essentially have multiple keels.  Thus, the “loss” of the traditional keel (broken due to a torpedo) on the bottom of the ship is not even remotely a fatal event.  The remaining “keels” bind the ship together and each is capable of maintaining the structural integrity of the ship.  Of course, this only applies to larger ships.  Smaller ships do not have sufficiently strong and heavy enough belts and decks to constitute “keels”.

It is obvious, then, that torpedoes are not instant death to a ship and by understanding torpedo damage mechanisms we can design resistant ships of all types and sizes.  Again, this does not mean that ships can laugh off torpedoes.  What it means is that the degree of damage can be mitigated and offer the target ship a better chance to survive and continue fighting.


This post also illustrates the danger in accepting truisms.  Not all are actually true!  We need to continually question our assumptions rather than blindly repeating them.  This also illustrates that conventional wisdom, even that "documented" on the Internet, may well be wrong.  


(1)Naval Surface Warfare Center, “Underwater Explosion Phenomena and Shock Physics”, Frederick A. Costanzo, Feb 2010,

(2)Shock and Vibration, 5, ” Spherical solutions of an underwater explosion bubble”, Andrew B. Wardlaw, Jr. and Hans U. Mair, Feb 1998, p.89-102,

(3)“The Dynamics of Underwater Explosions”, John Philip Best, University of Wollongong, 1991

(4)Defence Science Journal, Vol. 53, No. 4, October 2003, pp. 393-402, “Underwater Explosion Damage of Ship Hull Panels”, K. Rarnajeyathilagam and C.P. Vendhan,

(5)“The Response of Ships to Underwater Explosions”, A.H. Keil, Presented at the Annual Meeting, New York, N. Y., November 16-17, 1961, of The Society Of Naval Architects and Marine Engineers,


  1. Nice write up.

    We got Torpex from the Brits in '43. It was optimized as a higher density explosive than standard TNT, resulting in roughly a 20% faster shock-wave and great brisance. This was done by adding Aluminum powder to the mix.

    The downside of Torpex (45%rdx/37%tnt/17%al) was that a bullet or shell fragment could set it off, thus it was relegated to undersea use. HBX (67/11/17) was its immediate successor. It should be mentioned that as the percentage of aluminum is increased (to a point), all the "good" qualities of an explosive increase while the gas produced is actually decreased (1).

    The fact remains that aluminum found in modern explosives is designed to increase explosive force as well as the speed of the reaction, both of which are critical to causing the most damage for the least amount of material.

    While the evolution of explosives leaves us to conclude that the air bubble kill is a myth based on the changing characteristics of the explosives, its really hard to argue that even a single torpedo against a modern warship would result in a mission kill. Without wanting to go out on a limb to describe modern warfare, a mission kill is just as bad as a catastrophic kill because we don't fight wars of attrition anymore.

    Your post comparing range and speed of WWII ships to modern was eye opening. We have sacrificed a lot with our modern ships to have the latest geegaw which always seems to be more fragile and sensitive to what came before. We don't even like to shock test our ships anymore (3).

    1. http://www.dtic.mil/dtic/tr/fulltext/u2/135175.pdf

    2. https://maritime.org/doc/ordnance/index.htm

    3. http://www.businessinsider.com/uss-gerald-r-ford-shock-trials-2017-6

    1. "hard to argue that even a single torpedo against a modern warship would result in a mission kill."

      I assume you meant to say "would NOT result in a mission kill"?

      Mission kills are a different topic but, yes, modern ships are very fragile and susceptible to mission kills. That fragility is due to a combination of sensitive electronics and poor warship design practice which has lead to no armor, little redundancy, poor separation of critical equipment, general dropping of the requirement for component shock hardening, etc.

      You were aware that we've all but dropped the requirement for individual component shock hardening, right? Similarly, we've all but dropped the requirement that all electrical items be capable of EMCON mode. We really have abandoned good warship design!

    2. "You were aware that we've all but dropped the requirement for individual component shock hardening, right?"

      Really? Somehow I missed that. If the Navy is willing to build 3 gagillion dollar uber destroyers with guns that don't work, and the LCS, it would be fascinating for me if they, or a manufacturer, built a test ship like SeaShadow, but this would be a modern day ship built out of HY-80 and/or HY-100, a couple of mk 45 mod 4's, a SeaRAM, modest VLS pack, isolated powertrain of whatever type you want, and shock resistant, maybe some redundant systems. Go for simple and fixable but rugged.

      It doesn't need Aegis. It does need the latest SEWIP and some form of towed array.

      Then sail it around the world and test it.

      This is a test ship, so it would do test things. ASW work with its sound reduction and towed array. Shock test the devil out of it. See what sailors can do with it. Try gunnery with it. Play with the VLS. Shoot NLS, Tomahawk, and ESSM out of it on test ranges and test as realistically as possible (you might need to operate it remotely if you are doing this).

      In the end, do a sinkex next to an old 'Burke or 'Tico and see how they work.

      Then maybe we can have data to test simple and tough vs. gold plated.

      Yes, I'm having a bit of a snit.

    3. "we don't fight wars of attrition anymore."

      We may not but our enemies almost certainly will and it's very difficult, if not impossible, to avoid getting sucked into a war of attrition. For example, when faced with a human wave charge by infantry, you're in a war of attrition whether you want to be or not.

      Similarly, if your enemy is willing to trade ship sinkings just to deplete your supply of ships, it's very difficult to avoid. You can either hold your ships back from combat, which is a "win" for the enemy, or accept the war of attrition which is also a win for the enemy if they happen to have more ships than you do.

      In our pursuit of technology, we've consciously ceded the advantage of numbers to China and, in many areas, Russia. Knowing this, I suspect China and Russia will happily force us to engage in attrition warfare as one of their preferred tactics.

      WE may not want attrition warfare but our enemies get a vote and they likely will want it. Then what do we do?

  2. Could it be that the idea of "there's no sense applying armor to a ship to protect against torpedos" has at least part of its origin in cold war thinking where armouring your ships against torpedos would invite your opponent to put nuclear warheads on its torpedos ?

    I guess that a lot of ship design might change depending on the likelyhood of your opponent using nuclear weapons.
    If nobody uses nuclear warheads on its torpedos/AShM armour makes a lot more sense - but once you have armoured your ships your opponents might decide that they need nuclear weapons to seriously harm your ships.

    1. Using that reasoning, we shouldn't give our soldiers rifles because it might invite the enemy to use nuclear weapons. In fact, we shouldn't have an army because it could lead to nuclear war!

    2. I was thinking more along the line of "it might be more expensive to armour your ships than your opponent needs to spend to defeat your armour".
      This might have been a part of the thinking about ship's armour during the cold war.

    3. I don't know what misguided thinking took place during the Cold War but the value of armor is not purely its stopping power - it's the mitigation power. Armor limits whatever damage occurs. Instead of a single rifle bullet sinking a ship, it requires a much bigger weapon and more of them. Armor limits damage.

      I really get tired of making this same comment over and over (not just to you). Go read the previous post, Armor For Dummies

      It tells you what purpose armor serves and why it is needed even against nuclear weapons (you might also want to read about the Bikini tests - the damage from nuclear weapons was nowhere near as severe as most people think).

  3. Another excellent article. The submarine sailors would constantly refer to our surface ships as "targets". I think most of them believed in the keel breaking, one torpedo can sink any ship hype. I still have a lot of respect for the submarine crews considering the job they do and the conditions they endure, but they tend to get full of themselves. How do modern torpedoes compare to WW2 torpedoes in terms of destructive power? Armored ships with good damage control capabilities where frequently able to sustain torpedo damage and not sink or be mission-killed. Are there any cases of a sinkex where a ship suffered catastrophic damage from a single modern torpedo?


    1. I don't think the basic explosiveness of torpedoes has changed drastically since WWII. Yes, we've developed newer, better explosives but they're not overwhelmingly better. What has changed is the guidance and fusing systems. We can now cause a torpedo to explode under a ship rather than just hitting the side.

      SinkEx results are not published so I have no idea what has been learned from them. I know a supercarrier was SinkEx'ed in a classified experiment. Also, SinkEx target ships are prepared in various ways which may include leaving watertight compartments open. And, of course, there is no damage control or flooding control measures.

      Also, SinkEx'es are designed to give as many ships/weapons a shot as possible. It appears that the targets are subjected to a series of hits, each of which is more powerful than the preceding one, so that by the time a torpedo is fired the target has absorbed many hits of various types.

      So, it's tough to really draw any useful conclusions from SinkEx'es unless you have access to the classified or non-public data.

    2. Something essential has changed: warhead design.

      Modern warheads are no longer a plain bunch of explosives. They're specifically designed to create a particular set of shockwaves that are especially destructive to the presumed target's structural integrity by going above and beyond its stress limits in several seesaw fast pulses. While (also especially designed) armor and design can mitigate the effect, it's going to be a rough ride. Traditional armor won't do the job, I don't mind how many feet of old armor do you have if I can transmit the appropriate shockwaves to your internal structure ---and I can even use your huge armor to do it. I don't mind either is there's nothing holding it together anymore.

      You're right about the air break phenomenon as a primary destructive mechanism, but if I've managed to compromise your structural integrity before it arrives, or it's already close to the breaking point, it will be a massive damage multiplier, if not the last straw.

      Any country with a good knowledge of hydrodynamics, a decent lab and computers to model explosive effects underwater and its transmission to a vessel's structure can dramatically increase its warheads' destructive capability without increasing the amount of explosive. Nowadays, that's pretty much every country into the torpedo business.

      And that's why modern warships and subs are no longer designed like the old ones. Show me the most badass WW2 battleship, give me a couple 21st century torpedoes, and I can swear to you that the poor battleship is going to have a real bad day.


    3. I have heard/read nothing along those lines. Please provide a reference.

  4. Excellent Article!

    Anyone have an idea where this myth originated from?

    When did it first show up?

    I ask only so we if we understand how it came about, we can maybe not repeat the logic.

    1. Great question. I have no idea where the idea came from or how it came to be. I suspect that people witnessed sinkings (or videos) and saw a ship thrust upward (not noting that the ship's back was already broken) and then dropped back onto the "bubble" (in reality, not the bubble but the large area of disturbed "white" water) and then breaking in half. Their seemingly reasonable, though incorrect, conclusion was that the ship broke while suspended over the bubble - not realizing how small the bubble was relative to the ship. Just speculation on my part.

  5. Yes, a 324mm Lightweight Torpedo might not sink a carrier, but lets look at the Blegrano it had a displacement similar to the latest Arleigh Burke types .. we all know how that turned out, and yes it was sunk by unguided torpedoes

    1. What's your point? Is it that a 44 year old ship, poorly maintained and served by an ill-trained crew, sailing with all watertight doors open, could be sunk by three 21" (533 mm - not 324 mm) torpedoes (reports vary about whether the third torp exploded or not)? I think most of us would assume a ship in that condition was lucky to not spontaneously sink! Do you have a larger point to make?

    2. My point is that any modern large surface combatant can be sunk by a few modern torpedoes. To put it simple, look at the comparison between tanks and ATGM's no matter how more they up-armour them and what passive and active protection systems they use they get more heavier and more expensive, you can knock out a 10 mil$ MBT with one or two 50.000$ modern ATGM,s. Same with torpedoes or ASM for that matter

    3. "you can knock out a 10 mil$ MBT with one or two 50.000$ modern ATGM,s. Same with torpedoes or ASM for that matter"

      No. While I understand the monetary point you're trying to make, you're only looking at it from a very skewed perspective. That cheap torpedo that sinks an expensive ship - the point you're making - is not really cheap. The torpedo doesn't just magically appear in the water under a target. The torpedo launch mechanism, without which the torpedo is useless, must be factored into the cost equation. The torpedo launch mechanism is a submarine that costs a half to a couple of billions of dollars. Thus, the torpedo cost is the cost of the sub plus the cost of the torpedo.

      If an enemy has to build a two billion dollar sub to deliver a one million dollar torpedo, the cost to the enemy is enormous.

      Similarly, if an anti-ship missile has to be delivered by a four billion dollar ship or a hundred thousand dollar aircraft or whatever, the true cost of the weapon is much higher than just the cost of the weapon which is non-functional by itself.

      Consider the Soviets who had to construct entire regiments of long range bombers and whole fleets of nuclear submarines to have a chance to deliver missiles that had any hope of sinking a US carrier. Thus, the cost of the Soviet's anti-carrier weapon was enormous.

      You have to maintain an overall perspective of the entire weapon-counter weapon cost curve.

    4. Now lets look at that cost curve as you say :

      The best western diesel subs cost around 500mil$

      Russian ones cost even less, Vietnam bought six improved Kilos for less than two billion dollars.

      Not to mention mini submarines witch cost even less

      So a sub will be able to do more damage before being taken out, say my improved Kilo cost me 300mil$ but before being taken out i was able to hit a Burke and a LCS.
      Point being the "torpedo myth" might be real but if you fire off a few more than one against a big target you might probably very well sink it.

      Say, a modern SSK manages to launch six torpedoes against a carrier , what do you guys think will it survive ;)

    5. "Say, a modern SSK manages to launch six torpedoes against a carrier , what do you guys think will it survive ;)"

      Considering that a Nimitz is arguably rated to absorb between 8 and 'more than can be carried by a single Sturgeon-class submarine', depending on who you talk to, direct hits from torpedoes?
      I'd give the carrier a good 50/50 chance of limping away if in a major battle, 75% if otherwise, and 95% if it was a one-shot surprise attack that the SSK couldn't repeat for whatever reason.
      I don't give it higher than 50/50 for actually being repaired, however.

      The Iowas, for example, were rated as 'immune' to the MK48s by NAVSEA (or '5-6 impossibly accurate torpedoes that can somehow see in the dark without using sonar', and they were a lot smaller than the Nimitzs with a very similar under-keel TDS.
      The Nimitzs are actually not badly protected.

      - Ray D.

    6. Interesting RayD.
      Sow how much hits would you give to a Arleigh Burke class destroyer ( or any modern equivalent of similar displacement ) i'd say between two or three 533mm torpedo hits would be enough.
      P.S. i just recalled that some soviet/russian subs are able to use 650mm with a 500kg warhead, now imagine being hit by a few of those, they'll blow a LCS sky high

    7. "Sow how much hits would you give to a Arleigh Burke class destroyer (or any modern equivalent of similar displacement)"

      2 to 4, depending on where they hit and how damage control was 'feeling that day'.
      Really, it just depends on the damage control and the enemy's targeting - a Burke could logically eat 8 or more torpedoes under ideal conditions.
      Under-Keel ('backbreaking')?
      One excellent hit, two to four normal hits.

      "now imagine being hit by a few of those, they'll blow a LCS sky high "

      Not really, it would just cause a bigger hole. Unless that hole opens more lengthwise sections than the smaller hole, it's not going to make much of a difference.
      The only particular reasons why the LCS would be doomed from a single torpedo hit is a near complete lack of damage control and their aluminum construction, it has little to do with the torpedo. An LCS may well be doomed if a single Terrorist shot it with an AK while in port, or at the least stuck there in port for weeks while specialist technicians are flown out to repair the internal mechanical damage...

      - Ray D.

    8. Allright, and what about a direct hit by a supercavitating torpedo, are there any calculations on the sheer kinetic impact it would do wit a direct hit at a speed of 200knots, it would probably penetrate inside the hull and if it has a very limited time delayed fuse blow up inside the ship, that would be terrible damage.
      In the future more and more players will have them , south korea recently tested one indigenous design .

    9. And something on topic, now in this video fast forward to 4:23 min and tell me is that a "bubble effect' or what , the ship broke up in half


    10. "Allright, and what about a direct hit by a supercavitating torpedo, are there any calculations on the sheer kinetic impact it would do wit a direct hit at a speed of 200knots"

      A direct hit, literally contact? It would cause the same effects as an underwater shell hit of comparable effective explosive weight and impact velocity.
      Russia's VA-111 Shkval would be roughly comparable to a 9-11in shell, for instance; which, while no laughing matter, is not the most severe damage you could cause with a torpedo.
      Supercavitating Torpedoes' major advantage/lure is that they are much easier to actually hit their targets with (traveling at several times the speed of conventional torpedoes), acting in many ways like a slow underwater missile moreso than a torpedo proper.

      "And something on topic, now in this video fast forward to 4:23 min and tell me is that a "bubble effect' or what , the ship broke up in half"

      That looks like a pretty standard material stress failure of a ship with little TDS and that isn't properly braced to deal with that type of shock damage. A strong keel would have prevented the ship from breaking in half, but the flooding would have still been devastating for a ship of that size.
      In other words, yes, that is the legendary 'backbreaking' effect of the Mk48 torpedo - largely conventional damage given an easier path.

      - Ray D.

    11. "a Nimitz is arguably rated to absorb between 8 and 'more than can be carried by a single Sturgeon-class submarine', depending on who you talk to, direct hits from torpedoes"

      That is not common knowledge. I've never heard that, for instance, other than indirectly in the form of some generalized data Hughes presented (taken from another source which I can't recall, at the moment). If you're going to make claims that are not general knowledge, please offer references. I'm not disputing your claim (neither am I accepting it), just asking that you document uncommon knowledge.

    12. "is that a "bubble effect' or what , the ship broke up in half"

      I'm assuming you read the post. This was addressed. What is seen in all these types of videos is that the ship's back is broken at the very start, before any bubble suspension effect occurs. The damage is being done by the initial blast effects and the initial shock wave. There is no air bubble suspension effect.

      Further, what you "see" as an air bubble is mostly disturbed, agitated, frothing water - not the actual bubble. The actual bubble is under the ship and not visible.

    13. Ok, so am i correct to assume that a single 533mm torpedo hit can successfully take out/destroy a corvette or a smaller frigate. And for a destroyer with a displacement of around 8000/9000 tons it would take two or three?

    14. As a general statement, that's about right. Hughes presented data somewhat relevant to this subject in his littoral warfare book if you care to delve into it. His data described the number of major weapon hits a ship could take as a function of ship size.

    15. "Allright, and what about a direct hit by a supercavitating torpedo, are there any calculations on the sheer kinetic impact it would do wit a direct hit at a speed of 200knots"

      That's easily estimated. Here's the calculation.

      k.e. = 0.5 * m * v2

      m = 2700 kg; Russian Shkval torp

      v = 100 m/s; 200 kts


      k.e. = 0.5 * 2700 kg * (100 m/s)*(100 m/s)
      k.e. = 13,500,000 (kg*m2)/s2 =13,500,000 J

      By comparison, a kg of TNT releases 4,184,000 J. Thus, the k.e. of the supercavitating torpedo is equivalent to around 3 kg of TNT. To put that into context, a U.S. Navy lightweight Mk54 torpedo has a warhead weight of 44 kg (we'll assume it's TNT even though it isn't). That means the supercavitating torp would have kinetic energy equal to 7% of the explosive energy of a Mk54 lightweight torpedo - not enough to even be noticed, by comparison.

      Double check my math but I think I'm good on this.

  6. Brilliant Article very very well done.

    The geometry of the modern surface attack mode on Mk48 torps is critical. It all comes down to the positioning, and that is very complex. if its wrong the whole mechanism is nearly worthless.

    Obviously this kind of attack is significantly better then simply running your torp into the side of the ship and blowing up. Although that in of itself was devastating.

    I can only assume the myth came from "the best shot you have for a single hit kill is a heavy weight torpedo"

    or of course the Russians used \ "use" nuclear torpedoes. That IS a single shot kill. if you don't mind killing your own sub to do it.


    1. "nuclear torpedoes. That IS a single shot kill."

      Beware of unproven statements. For example, the Bikini tests showed that ships could stand up to nuclear explosions far better than was anticipated. Further, a nuclear torpedo is a "small" nuclear weapon (that's a relative comparison!) and might not prove structurally fatal to a large ship like a carrier (ignoring radiation or other ultimate effects) depending on the exact location and depth of the explosion - or it might be instant death. Again, the Bikini test results were eye-opening and enlightening using much smaller ships than a modern supercarrier. This is just wild speculation on my part.

      As I cautioned in the post, don't accept "conventional wisdom" without proof.

    2. I think you're thinking of the air drop test from Op. Crossroads, when the 'damage' to the ships was largely the result of radiation and fallout. The second shot--when a 23kT nuclear weapon (large at the time but well within the range of modern 'Tactical' nuclear weapons) was set off at a depth of 90 ft under a landing ship--did not go so well for the ships. No identifiable part of the landing ship was ever found.

      The USS Arkansas (a 27,000 ton Wyoming class battleship) was 170 yards from surface zero and was lifted almost entirely out the water and the damage was described as:

      "The underwater shock wave crushed the starboard side of her hull, which faced the bomb, and rolled the battleship over onto her port side. It also ripped off the two starboard propellers and their shafts, along with the rudder and part of the stern, shortening the hull by 25 feet (7.6 m)"

      The Saratoga (450 yards from surface zero) sank within 8 hours.

      The Nagato (770 yard from surface zero) sank several days after the tests (and as you say, would probably have been able to survive with damage control, assuming the crew didn't die of radiation poisoning).

      While a modern super carrier is indeed heavier and larger than the ships involved, if a 10kT warhead went off under the keel (the proposed nuclear torpedo scenario) it would almost certainly vaporize a large hole in the middle of the ship of vaguely LSM size. I assume they don't bother to build a Nimitz to survive having a 100-200 ft spherical void appear in the middle of it in ~.1 a second while having 2 million tons of water forced up into the hull at 700 m/s.

    3. Both tests caused less immediate damage than might be imagined. Of course, damage is a relative term! Saratoga, 450 yds from the explosion, probably would not have sunk with a crew available to provide damage control. She sank from slow flooding over several hours. Modern dive reports suggest that her watertight doors were not closed. A modern carrier would certainly have survived. Radiation is, of course, another issue.

      I note that Saratoga had already survived one atomic explosion. It is unknown what underlying structural damage and weakening, if any, she suffered from that prior to the second explosion.

      It should be noted that the test was conducted in a confined lagoon with a much shallower bottom than would be encountered at sea. Presumably, this magnified the effects.

      I also don't know what the yield is for modern nuclear torpedoes. One report cites a 5 kT yield for an older Soviet torpedo. Other reports suggest yields up to hundreds of kT.

      I'm not suggesting that a modern carrier would laugh off a nuclear explosion directly under the ship. I'm pointing out that danger in assumptions that are not based on actual data. Modern naval warfare discussion seems riddled with incorrect assumptions as I've been pointing out in my posts. No one knows what the effect of a 5-10 kT (if that's the size of nuclear torpedo) explosion on a 100,000 t ship with watertight integrity, damage control, and no shallow sea bottom to magnify the effects, would be. I'm actually surprised the Navy hasn't tested that scenario using one of the retired supercarriers. Or, perhaps they already know that the result is a foregone, fatal conclusion.

      I suspect that even if the ship physically survived, the crew would be rendered unconscious or dead from being slammed around by the shock and vibrations. And, again, this ignores radiation effects.

    4. Im aware of those tests, but they were an air bourn explosion. The propogation of shockwaves through an essencially incompressible medium ( water ) is totally different, and would ensure a massively more efficient energy transfer.

      Here's a video of a nuclear depth charge. 8Kton


      ( an oiler is there for scale )

      If you can imagine a battle ship resting on top on that explosion, in the "break back" mode, even inacuratly, you can see nobody would survive.

      The shockwave would supersonically first go through all the bulkheads ( near enough instantly ) and puree the crew, even before the ship hull started moving.

      The drawback is that any submerged vessel would experience this fate even faster. Any implication that a Russian sub crew would survive their attack is fantasy.


    5. Ben, there were two tests - one air and one subsurface.

  7. Here's a link to a video of the sinking of the former HMAS Torrens (DE-53) by a MK-48 torpedo. Something appears to have exploded under the ship, causing the ship to bend or flex, then break in two.


  8. Here is the USS Honolulu in '44.


  9. Reffering to a few old books on Warship design by well known writers ( Friedman DK Brown). Breaking back by underwater explosion is mentioned when discussing torpedo effects.
    eg This from Friedman- Modern Warship Design and development 1979, discussing a Torpex on a Cleveland class cruiser.
    "Light Cruiser Wilkes Barre was destroyed by under keel the keel explosion, which produced an oscillating rising bubble of gas. On the first expansion the bubble lifts the ship out of the water, straining the hull girder. generally the second expansion suffices to produce the result shown [ in pictures] , the hull simply breaks. Conventional hull armour merely makes matters worse as it makes the hull girder more rigid."

    I dont think a physicist and highly knowledgeable author like Freidman got it wrong.

    1. I don't think all the scientific sources I cited got it wrong.

    2. To my thinking the description by Freidman agrees with your expert'
      "Keil (5) goes on to describe the mechanism of failure of the ship’s structural members. The mechanism is one of sequential elastic flexing and relaxation of the strength members of the ship (bulkheads and longitudinal members) in response to the various shock and pressure waves. If certain structural properties are exceeded, a permanent deformation of the hull structure will occur. If those properties are exceeded by a sufficient amount, the deformation (flexing) cannot be recovered (relaxation) and the structural members tear – the iconic broken back scenario. "

      The process seems to be the same, a couple of expansion/collapses of the high energy bubble is enough to cause structural failure.
      I suppose how much damage is from the shock wave and the from various bubble effects .
      This technical analysis ( Reid)seems to suggest high energy water jets from the bubble can be responsible for significant damage. Also to be considered is the period of oscillation of the bubble matching the natural frequency of the hull girder, a fairly small air bubble, like you mentioned, if repeated have have much larger effects. But only in the right circumstances.

      "The response of surface ships to underwater explosions"- Warren Reid

      Having looked into the details, the bubble of air effect is as you said simplistic and the damage is likely from a range of effects from the explosion.

  10. CNO, the 'myth' is actually a well known, well understood phenomena in Naval Engineering.
    Mind you, 'common knowledge' of it and the internet in particular have blown it out of proportion, apparently, so it may as well be myth and legend at this point.
    I couldn't help but laugh at the bubble graphic in this write up, the bubble is way over-sized for a torpedo (more fitting a multiple kiloton level nuclear bomb). I will assume in good faith that it was created for the consumption of the average layman and therefore comically exaggerated to make a point. I disagree with the practice on a personal level, and institutionally, but I understand why they do it.
    Now that I have seen this comedy, I believe we are closer to the same page, please allow me to explain myself from prior comments.

    The 'gas bubble' effect is determined by 'beam of keel support', the overall beam that is integrated into the 'spine' of the ship, not a ship's length - length doesn't actually strongly factor into the equation beyond the standard length of sections.
    That being said, Length is commonly stated instead even by Naval Architects simply because certain Lengths assume certain beams on ratio for its design intent (1:8-9.5 Beam-Length ratios are expected for gun cruisers, for example), the long a ship is the wider it generally is as well (speed builds like the Iowas not withstanding).
    What is happening is that the weight of the hull/superstructure cracks (JUST cracks, no magical disintegration) the main keel on the tiny section that isn't suspended on the water (only a small area of unsupported weight is needed unless properly braced) purely as an issue of material failure, which drastically drops the Ultimate Tensile Strength of the material at that specific spot (due to introducing faults), which in turn gives the warping properties of the conventional damage effects a much easier time shattering the keel along that entire area, effectively snapping the ship in half with a much smaller warhead than would otherwise be needed to pull off that stunt. Thus, 'backbreaking torpedoes' (properly, 'under-keel' torpedoes, as even 'contact' torpedoes detonate under keels today).
    Proper shipbuilding practice and design characteristics can, at cost, completely negate this effect on ships of any beam simply by thickening the main keel, or (much more expensively) using multiple keels, or (if you throw cost concerns out the window) both (such as with the US' Fast Battleships). If the Keel is yield capable of supporting the raw weight (not displacement) of 2 sections length, then the ship will shrug off the effect and move along the bubble. Alternatively, you can simply build the ship to beam roughly 1.15 times the expected diameter of the under-keel torpedoes' bubbles (about 55.3', for the MK48 ADCAP), assuming standard WW2 cruiser type hull construction, and call it a day.
    Ironically, this means the Burkes are better protected against the backbreaking effect than the Ticos. However, modern ship building/design practices have resulted in massively lower area tensile strengths, resulting in a wider required beam, which the Burkes do not meet either (around 80' was an estimate I had jotted down from a while back, meaning the Zumwalts meet this).
    This could, and should, be fixed in any new design ship and therefore the 'Torpedoes > Everything' is - and I agree with you here - a myth that I venomously detest, yet it is an unfortunate fact that (Real) Destroyers are simply not worth the expense to properly armor against torpedoes in general. Not that that would be a problem, a Real Destroyer design would actually be buildable in mass and still more durable than our current 'Destroyers'.

    - Ray D.

    Still laughing at that bubble graphic, I had actually forgotten that existed. Thank you, I needed this laugh.

    1. If I understand you correctly (and I'm not sure I do!), you seem to be suggesting that the air break phenomenon from the post DOES occur (your description of "cracking"). I've disproved that via scientific papers and referenced them. If you believe this is not correct then you need to offer references supporting your view. Otherwise, what you're claiming is simply not true.

      As I said, it is not clear to me exactly what you're claiming so I may well be misinterpreting your comment. If that's the case, try again to explain what it is that you're trying to describe.

      More generally, you're making a lot of statements that are not common knowledge. Such statements should be supported with references of some sort.

    2. "More generally, you're making a lot of statements that are not common knowledge."

      Ah yes, "common knowledge." While I admire this blog's emphasis on rigor and logical thinking, it's important to keep in mind that cross-disciplinary discussions will inevitably run into issues with respect to what is, and is not, "common knowledge." While ComNavOps's comment policy does establish the expectations of this blog, I humbly ask that ComNavOps give some consideration to the fact that those of us with different backgrounds may not think to provide citations for assertions that are considered "common knowledge" within our fields.

      Having a BS in materials science (although currently working in the legal field), I think I can speak for Ray to explain that "cracking," as Ray refers to it, relates to crack initiation at a microscopic level (i.e., with respect to the hull materials microstructure). Essentially, the initial shockwave will, almost certainly within some area of the hull, rapidly increase stress well beyond the materials instantaneous fatigue limit and initiate crack formation. Whether or not the initial shock wave is powerful enough to propagate these cracks through enough of the hull material/members to "crack" a ship on a "macroscopic" level will depend on many factors. Based purely on my personal knowledge, I'll wager that the advantage of forming a bubble beneath the hull is that it increases the time period over which the effects of the explosion are applied to hull (i.e., able to do more "work" on the hull through effects 3 and 4 noted in the original blog post). It's kind of like the difference between a high explosive and a low explosive. The initial shock wave MIGHT weaken the hull enough that subsequent shock waves due to bubble expansion/collapse and the force of the bubble jet MAY sheer structural members.

      This paper from my alma mater explains some of the physics of crack initiation and propagation: http://fcp.mechse.illinois.edu/files/2014/07/2-Mechanisms.pdf.

      Ray also does NOT seem to disagree that the "air-break" phenomenon is blown WAY out of proportion. He's merely trying to point out that although "real," there are plenty of ways to design, at a cost, a warship to survive the phenomenon on a realistic scale.

      To put things in perspective, we can design ships to do this at a reasonable mass and price: http://www.heavyliftspecialist.com/wp-content/uploads/2013/01/Pacific-Orca-Jack-up02-1024x767.jpg

    3. "The final piece of the logical analysis is the videos we’ve seen of ship’s breaking in two during a torpedo test. Going back over those videos, what we’ve failed to note is that the ships are thrust up, out of the water with their backs already structurally broken in an inverted “v” shape. In other words, the structural back of the ship was deformed by upward pressure (or direct blast effects or the initial shock wave) not downward pressure as the air break phenomenon would mandate. The broken back was not due to suspension over a bubble but by weak structural members deforming and snapping due to initial pressures, most likely the initial shock wave or direct blast effects."

      I agree and disagree with this conclusion. While I agree that, "the broken back was not due to SUSPENSION over a bubble," emphasis mine, I don't think that we can conclude that "3. Damage from the subsequent bubble pulses (cyclical expansions and contractions)" and "4. Damage from the bubble water jet" are not significant "factors." My hypothesis is that these additional factors due to bubble formation and collapse may, in some cases, be sufficient to cause a "broken back" if the initial blast effects, including the initial shock wave, merely weaken the ship's structure. I believe this because it's well understood in metallurgy that, in general, it takes more energy to initiate a crack (i.e., a dislocation) than it does to propagate one for a variety of reasons. Note that this pertains to brittle failure mechanisms and that there are also plastic deformation failure mechanisms to consider. It's also worth noting that it's for this reason that older/more fatigued ships will likely be much more susceptible to a "broken-back" scenario due to metallurgical defects incurred over their service lives.

    4. Gripen, very useful and informative comments and you provided some references. Well done.

      Nothing you said contradicts anything in the post, as far as I can tell. I described the breaking of a ship's structural members, in general terms, as being due to direct blast effects and/or shock and pressure waves. You went further and described the phenomenon on a microscopic level. Nothing wrong with that. You need to understand, however, that a blog post is constrained by both length (attention spans!) and complexity. The audience as a whole is not trained to grasp the minute details of these mechanisms at a microscopic and doctoral thesis level nor do they have any interest in such. Therefore, I have to summarize and simplify in order to covey the basic concepts. I stated exactly this in the post.

      The problem with summarizing and simplifying is that, inevitably, someone, who has more advanced training, will respond and "critique" (I'm putting it politely) the writing, finding fault with it rather than recognizing the conscious simplification that had to be done to make it generally understandable.

      You responded appropriately by offering additional detail for those interested and offering references to support your statements. I would like to take everyone's statements at face value but I too often encounter people who make statements and claims that are false (whether they are made intentionally or are simply repeating misinformation is another question and irrelevant for this discussion). Thus, I have to request references for statements that I DO NOT KNOW TO BE TRUE. That way, I can verify the information and either remove it, if incorrect, or allow it to stand as a useful and welcome addition to the discussion, if true.

      You might note that the entire post was my response to commenters repeatedly citing the "fact" that ship's backs are broken by suspension over an air bubble. I'll assume they all meant well and were just repeating misinformation but it illustrates that even "common knowledge" may not be correct! I suspected that this "common knowledge" was not correct but I did not know for sure so I spent several months researching the topic, leading to this post. The hardest part of the endeavor was deciding how much to summarize and to what degree to simplify.

    5. "I don't think that we can conclude that "3. Damage from the subsequent bubble pulses (cyclical expansions and contractions)" and "4. Damage from the bubble water jet" are not significant "factors."

      I did not say that they are NEVER significant factors - just that they are not the predominant factors. My research indicted that under some conditions, those factors may well be significant but in the general, common case (if there is such a thing, they are lesser factors.

      This also goes to the complexity issue. It is very easy to cloud and confuse the overall concept by presenting too much data and too many qualifiers to an audience that does not have the background to properly interpret it. Thus, I make it my responsibility to evaluate and summarize/simplify in order to best serve the general audience. You, with more training, have "pounced" on this as an incorrect presentation. I would ask that you recognize it for what it is - a simplification for the purposes of a blog post.

      Consider : You have written as much as the entire post on just one tiny aspect of the overall issue!!! Had I taken that approach and applied that level of detail to the overall post, I would have written a book that would have been incomprehensible to the general audience.

      The hardest part of writing for a blog is not deciding what to write but deciding what to leave out - knowing that some commenter is waiting to pounce on any slight simplification!

      So, again, I ask that you recognize the post for what it is and what it is not.

      If I touch on a subject that you feel I have covered insufficiently, and that you feel you have superior knowledge about, feel free to offer a guest post. I mean that sincerely. A guest post from a subject matter expert is always welcome. Of course, then you'll have to decide what to summarize and simplify!

    6. Thank you for explaining more fully the complexity of creating a post. And no, I don't think my comments contradict the content of your post. I merely thought I'd jump in to explain that Ray's comments with respect to cracking do have a metallurgical basis. I can't speak to his comments with respect to warship design.

      I appreciate the consistent effort you put into your posts and your responses to commentators. Thank you for making this blog one of the few places I've found online where people can have reasoned discussions about warfare technology and strategy. I just wish to add a bit more context where I can.

    7. I welcome informative comments that are supported by data and logic!

      Here's a question that you might be able to offer some insight on - from your perspective and knowledge base, is it possible to design/construct a structure (the ship's frame, structural members, and hull plating) that is more resistant (or, better able to absorb) to underwater explosions? Maybe a different structural framework design instead of the traditional transverse and longitudinal members? Maybe different structural materials like composites or some such? Maybe increased diagonal supporting structures like Humphrey's Constitution? Layered armor plates? Floating structural members rather than rigidly attached? More void spaces? Any thoughts from your perspective?

    8. "with respect to cracking do have a metallurgical basis. "

      If metallurgy is your specialty, would you have any interest in guest authoring a post on the common phenomenon of cracking in the LCS and other aluminum constructed ships. I have a Navy report on the initial LCS cracking (a bit disturbing and more prevalent than the Navy publicly acknowledged) and I'm sure it contains some substantial insights but I don't have the training to draw worthwhile conclusions from it. You might be able to make more sense of it.

    9. CNO, Gripen has my intention and point correct.

      I was not so much disagreeing with your main point (in fact, largely I was agreeing with you) as I was saying 'the backbreaking effect is real, common assumption is just wrong about how and why it happens'.
      The 'bubble effect' (to use a common term) actually boils down to material failure due to the difference between 'shock yield' and 'sustained yield'; now, Gripen or others may have to correct me on my terms, Naval Engineering only teaches a simplified form of Materials Science and in numerous cases use internal nomenclature for various aspects that are not used elsewhere (what he called 'instantaneous fatigue limit', I just called 'shock yield', for instance).
      Simply put, it's wave suspension dynamics on a micro-scale causing 'cracking' in the keel (again, micro-scale); the initial shockwave (before the explosion hits) slams against the hull (pushing the section it hits up) while at the same time instantaneously applying the entire weight of the gas-suspended section to the keel (and likewise other lengthwise frame supports). The combined properties of these two effects cause the keel (et all) to nearly instantaneously warp, causing micro-cracks to form unless the keel's shock yield is higher than the combined effect, which then allows the conventional effects (your entire 4 point summary) to tear through the ship at much less resistance.

      I also have to comment on one thing, asking for pardon.
      As Gripen stated, the backgrounds of people involved very drastically influence what is or is not considered 'common knowledge' by various types of people.
      For myself, my background in Naval Architecture (specifically, Naval Engineering) has me viewing much of this as 'common knowledge', so common in fact that some aspects have become 'unthinkable fact' and 'unspoken truths'.
      The above comment on the Nimitz and Torpedoes is an example of this. Your reply to it actually gave me pause as my mind was trying to get past the 'But, it's obvious' stage of thought, hardly being able to grasp that, no, it's not. I merely read the question and answered it out of second nature, it didn't even occur to me to explain how or why. (The Torpedo endurance of the Nimitz is a function of its size, heavy conventional TDS, the sheer number of watertight compartments, and stability characteristics under load - it would take writing a book to actually explain it in depth).
      This is the unfortunate circumstances of the field, where 'you don't ask why, you just know it does'.
      Take Armor, for example. I can tell you off the top of my head that the effectiveness of Secondary Armor Plates in a Spaced Armor scheme are valued at 0.71684, but I can't even begin to tell you why exactly that is nor point you to a resource to verify that (okay, Nathan Okun's papers). I'm an engineer, not a physicist, I just take the factors that I am given and apply them to get the desired results.
      With respects to the Materials properties, I cannot tell you why X or Y happens exactly, I am not a Metallurgist, I can just tell you 'this happens if X, Y is how you stop it'. If you stop me and ask 'why?', I and most of my colleagues would probably just stand there and stare at you dumbfounded, as we really just don't know.
      This allows us to have people do what they specialize at, and can get great results, but then budget cuts and they let the ordnance guy go, so the team runs around screaming 'who has any idea on this guy's job?'. Eventually some schmuck ends up moonlighting in the role, until they let the Hydrodynamic Physicist go and half the team walks out because they all know nobody could replace him.
      There is a reason I gave up on that field and never completed my apprenticeship, nobody has any idea what they are doing anymore.
      Regardless, I apologize for my comments that have been out of form.

      - Ray D.

    10. Ray, you are quite correct about taking one's area of specialized knowledge for granted and assuming that everyone shares it. The reality is that we all have specialized knowledge that is not "common". Hey, we went to school for years to acquire our knowledge so it must be somewhat rare, right? Otherwise, it wouldn't take years of training to master!

      That means, however, that someone - me in this case - has to try to ensure that information is "generalized" enough to be understandable by the broad audience. I also have to verify, as best I can, that information in comments is correct when I encounter something that is outside my realm of expertise - hence, my request for references.

      Regarding references, I completely understand that you may have knowledge that is true but that you can't readily produce a reference for. That's fine and in that case it is sufficient to simply state that you are a trained (fill in the blank) and that you are presenting knowledge known to your field of expertise. That isn't quite as good as citing an actual, specific reference but it's quite acceptable.

      There are good and bad aspects to being a subject matter expert and not citing a specific reference. On the plus side, we can all benefit from the expert level knowledge the expert has. On the minus side, the lack of a reference prevents anyone from disputing (or further learning about) the presented information unless they are a fellow expert. So, it cuts both ways but I find that type of information useful and have no problem accepting "expertise" as the reference as long as the degree of expertise (experience, education, etc.) is spelled out for the reader to understand.

      Your information was fine (and interesting and helpful) and simply needed a specific reference or an indication that you were presenting information based on your training and experience along with a brief explanation of what that training and experience was.

      For example, as a Ph.D Anti-gravitational engineer with 15 years designing anti-gravity vehicles I can say that ........

      Well, that was a very long-winded way of saying, thanks for information and contribution and, in the future, just please include a reference (of whatever sort) so I/we have a basis on which to assess the authoritativeness (?) of the information.

      Again, thanks for the information and contributions! Let me know if some aspect of naval engineering intrigues you enough to want to write a guest post.

    11. Ray D
      Interesting information. The new Chinese destroyer 055 has a 66ft beam. If 55.3 ft is all that is needed, the new destroyer seems to be designed to be resistant to the mk48 torpedo

  11. Ray....

    "What is happening is that the weight of the hull/superstructure cracks (JUST cracks, no magical disintegration) the main keel on the tiny section that isn't suspended on the water (only a small area of unsupported weight is needed unless properly braced) purely as an issue of material failure, which drastically drops the Ultimate Tensile Strength of the material at that specific spot (due to introducing faults), "

    Couldn't the same thing happen by a hull girder merely having its length suspended between waves?

    I know that was at least a concern when the Edmund Fitzgerald went down, given her length. But it would seem at least possible on a ocean going ship of 500 odd feet given the right wave conditions.

    1. "Couldn't the same thing happen by a hull girder merely having its length suspended between waves?"

      Technically? Yes, there are a number of circumstances that could cause this to happen along the length, particularly if it is a much older ship in question built to lax codes or with experimental procedures that are not well understood.
      Realistically? It's unreasonable if the ship is built to a solid set of standards. This is why we shifted to Longitudinal Framing for most ships, after all - it provides much greater structural stability along the length than Transverse Framing.

      You ask specifically about waves, but the thing you have to realize about waves - even in high sea states - is that they operate over a length of seconds, allowing the materials in question a generous amount of time to gently bend with the wave pressures and ride it out.
      The gas bubble, like any torpedo concussion, operates over the length of milliseconds, which doesn't allow the materials in question any time at all to adjust to the pressures involved. No gradual shift of weights, just sudden 'boom' and there it all is.
      To give a simple analogy... a reasonably strong man may be able to lift and hold a ~58lb weight with one arm and not suffer any real damage, but you suddenly drop that same weight on the same man's arm when he wasn't expecting it and that arm is going to break.

      The Edmund Fitzgerald was, from my understanding of the case, a literal perfect storm of problems and faults that led to everything possible to go wrong going wrong.
      Being a lake ship, she was never built to withstand the extremely high sea states that she ended up experiencing when she sunk (waves of over 45ft, which would have and did sink WW2 ships, see Typhoon Cobra), on top of this her hatch doors blew off, on top of this she wasn't being maintained properly and her hull had deteriorated to the point of not being seaworthy, etc.

      - Ray D.

  12. Okay. Thank you... so one further question. Sorry if it seems overly simplistic....

    Could you build a ship made to 'bend' instead of break?

    I'm thinking of how with armor you had face hardened armor built to decap, but behind that softer steel built to absorb the blow. Or, to borrow your 50lbs weight analogy, use something like oil filled voids to act as a 'pad' to absorb the initial punch while a softer spine bends with the lessened blow.

    Just thinking outside the box. So I realize it might be crazy.

    1. In the post I cited Keil who suggested that elasticity could be designed in to absorb rather than resist shock and pressure. Further, we routinely design skyscrapers and bridges, among other structures, that are designed to flex and bend. Some of the videos of bridges undulating are simply stunning!

      So, to answer your question, yes, it is conceivable that ships could be designed to absorb shock and pressure better than they do now.

  13. "Keil also suggests that considerable underwater explosion damage resistance can be achieved by designing in a large degree of elasticity into the ship’s structure and plating as opposed to attempting to resist damage via increased hardening (5). This illustrates the concept of designing the ship to absorb torpedo damage rather than trying to resist it."

    Duh. ID10T error on my part. Read it but didn't absorbe it. Sorry.

    1. No problem!

      The Navy has, inexplicably, stopped pursuing warship protection since WWII. There have been no advances in warship protection. Contrast this with the myriad advances in active and passive protection schemes for tanks and vehicles.

      Worse, this has lowered the standard. We now think a Burke is a well designed warship. Imagine if a WWII warship designer had proposed a cruiser size ship (which is what a Burke is) with no armor, thinner than normal hull and deck plating, and weaker than normal steel? He would have been tossed out on his rear end. Yet, we now consider that to be the gold standard of the Navy and most people think the Burke is a well designed and built ship. That's absurdly and disturbingly hilarious!

  14. P.S a little suggestion, you should run Open Comment post from time to time where we can share interesting naval stuff , videos pics e.t.c.
    today i found out the Borei class has a sauna :D

    at 16-30 min
    still not as good as the swimming pool in the Akula class

    1. That's a good suggestion. I'll give it some thought and maybe try it on an experimental basis.

  15. The late British Naval Constructor David K. Brown did some analysis into WW2 battle damage from a variety of weapons, including back-breaking from underwater weapons. He agrees with you that size/construction play a significant role in ship survival or loss. The most vulnerable ships were smaller ones, although the large light cruiser HMS Belfast suffered severe damage from a German magnetic mines that warped her her/wrecked machinery and other systems much like USS Princeton in her Gulf War mine damage.

    1. On working on a related post which examines several warships that took torpedo/mine hits and what kind of damage they sustained as a result.

      One of the aspects I've already noted from the research (yes, I actually research these posts!) is that mines, even modern ones, tend to explode offset to the side of a ship (sometimes contact fused against the side) rather than under the hull as with a torpedo. I'm trying to find an example of a side exploding mine (or any mine) that broke a ship's back along the lines of a torpedo. I've haven't found one yet although there likely is an example somewhere. I'm also trying to find an example of a cruiser or larger ship that had its back broken by a torpedo. Again, I haven't found one yet. Still early in the research process.

    2. Just a reminder - I'd love to have a guest post on the historical role of frigates, going back to the age of sail, and how that role translates to modern navies and combat (who/what has taken on the traditional frigate role/function?) and how that relates to the modern "frigate" which is actually, in most navies, the new capital ship.

      Let me know if you're interested.

  16. Torpedoes in the past were set for vary depths according to the target . Optimum was exploding the torpedo 10 foot or less below the hull. When you talk of a 5o foot blast bubble, if that starts 10 feet from the hull/keel, the incomprehensibility and weight of water is going to force that expanding gas bubble at 10 feet to hit that point of contact with the hull like a shaped charge , expanding much more against the hull in a radiating area around the initial point of contact . The remaining 40 foot of bubble is going to largely act against that area of the hull rather than expanding into the water. It does not break the keel hull by flexing, rather it simply punches though and destroys that area of the keel/ hull , the rise/flex of the hull is incidental , the keel is already broken by the initial blast bubble it got caught in. Sure a bubble exploding too deep and rising up though the water to hull , shrinking in effect and pressure the whole time might just raise a hull some and not break the keel, but in that case the torpedo was obviously set too low or a deep runner , that is not how torpedoes are depth set to operate . 10 feet or closer to the hull they are shaped blast demolition charges and a 1/2 ton or ton of explosives at that range against steel with say 30 feet of weight of water to "shape" it the steel does stand a chance.

    I hope you understand my point from "layman's terms" and pleae don't take offence but they set torps to blow up close to hulls, without touching them as the optimal blast bubble is more effective a few feet away rather than contact , but on the flip side iMO you are more describing a torpedoes effects when it detonates too far away. The effectiveness now of torpedoes is even greater now than when they manually set them before launch because torpedoes can either be guided or guide themselves to that "sweet spot" 10 feet or whatever it is for that torpedo to knock a 10-50 hole in the bottom of a ship,no need to to lift it at all , to inflict fatal damage.


  17. Note: As an addenda, I suggest watch film of U-Boat's torpedoing ships in WWII. Many of those are from their "Magnetic influenced torpedoes" detonating a few feet under ships. You see the blast just tear thought the bottom of the ship, some blown in half , but you dont see much lifting because the blast bubble goes through ship from the bottom rather than expanding anymore in the water after it contacts the ship hull ,at the point it acts increasing against against the ship hull in those first milliseconds. this is what made magnetic torpedoes so deadly, driving the need for degaussing efforts, which neutralized those torps , however torpedoes are back to and far past that level of deadliness, given technology advances.

  18. A good argument for torpedoes.. The issue however is.
    That lack of armor actually has much more to do with Missiles then Torpedo's.

    And there is a pretty simple reason for this, nations have made Missiles with armor penetration values of over 1600mm of armor. And that was in the 1970's. They can also be programmed to simple ignore belt armor by 'popping up' and then hitting deck armor.

    The fact that we don't armor ships vs Torpedoes anymore is just kind of a side effect of this. And all that is ignoring also that people have stopped making missiles to penetrate large amounts of armor because there are no ships with armor anymore. The truth is that if the USA started to make a heavy armor ship, other nations would have a heavy armor penetration missile made before the ship was launched.

    1. Utterly absurd! There is no missile with a demonstrated penetration of over 5 ft of armor! Russia has made ridiculous claims along those lines but they make ridiculous claims all the time.

      The US Navy test Harpoons against battleship armor and they didn't even dent the armor. That's about the only actual test that I'm aware of. Please share details if you know of any other actual testing.

      Penetrating armor requires a massive weight and shaping to the projectile and no modern missile has that. Recall the 2000 lb 16" AP shell. That was barely rated to be able to penetrate 16" or so of armor. You need to read up on Okun's writings armor and shells before you talk about AP missiles.

      You also completely fail to understand the function of armor. You should read, "Armor for Dummies"

  19. I have recently been reading articles and research papers on underwater explosions, particularly the bubble jet effect. That is how I ended up here. The “air break” myth would certainly be a misguided notion. For the sake of this argument, the bubble could be considered air with the same density. It’s not of course, but close enough. If a bubble was able to apply any upward force to a ship, air would be able to as well. In other words, the ship would be buoyant in air. Ships don’t fly and a bubble will not lift a ship. It will disperse and flow around the hull like any other gas pocket underwater. The damage from a torpedo is caused by three effects. First, the shock wave lifts the ship. This stresses the structural members. Second, the cyclic pressure waves of the collapsing bubble flex these members until plastic deformation occurs. Some articles refer to this as a “plastic hinge” which then allows the ship to sag into the bubble below. The third and most interesting blow is from the bubble jet. Using the magnetic signature of the ship, torpedoes are detonated at a specified depth to achieve a contact bubble. As the bubble collapses, it collapses faster at the bottom and will fire a very high speed jet of water through the hull. It does not take much imagination to visualize how these forces in combination would break the back (keel) of a ship. All of these forces are applied in a matter of milliseconds.

    I tried a small experiment at home to observe the bubble jet effect. On a bowl of water, I started a drip from the faucet. I made the drop as big as possible to maximize the effect. Then I used the slo-mo video function on my iPhone to record it. The drop forms a hemispherical depression. As the half bubble collapses, a perfect jet forms every time. The jet was approximately three times the radius of the bubble. The jet that strikes the ship moves at approximately 900 fps and has a diameter of several meters. In a near surface detonation, this jet can rise 100 meters into the air.

    1. Nice to see someone else investigating this phenomenon! You've essentially repeated the main points of the post. From my investigation, the water jet appears to be potentially quite damaging but also quite hit or miss in actually making contact with the target. In order for the jet to have an effect, the bubble must, essentially, be in contact with the target and this is nowhere near a certainty. While you note that torpedoes are designed to explode at an optimum point below the target, the reality is that there is quite a bit of variability in the explosion location. This is analogous to the anti-ship missile that is designed to hit a particular spot but the reality, due to a host of factors, is that missile's strike location is almost random. Similarly, the torpedo may explode too deep and the jet won't contact the target or it may explode a second too early or late and the bubble/jet form to the side of the ship and the jet doesn't make contact.

      As you know from your own investigations, there is not a lot of direct data on actual torpedo explosion damage effects due to the jet phenomenon. My conclusion is that the jet is powerful but, due to the variability in the torpedo explosion location, not a major factor (not a common factor might be a better way of stating it) in torpedo damage effects. It's a hit (pun not intended) or miss occurrence.

    2. "a bubble will not lift a ship."

      Yes and no. You're correct that a bubble, alone, will not lift a ship, however, the bubble expands and displaces the water around it and it is the displaced water that lifts the ship. The magnitude of the lift depends on the weight of the ship as well as a variety of other factors related to the bubble. Films clearly show the lifting (rising on the upswell of the displaced water is a better description?) of the hull.

    3. If you haven't already seen it, you might want to check out the two part post on historical torpedo and mine damage. Here's a link to Part 1

  20. You are incorrect. Ships are held afloat by their displacement of water. If there is a large pocket of air under a ship, there is no buoyancy. In exact terms, hydrostatic pressure is created by the seawater: formula p = rho x g x z where p = hydrostatic pressure, rho = density of seawater, g = acceleration of gravity and z = depth. The ship sinks to a depth where the hydrostatic pressure on the hull imparts an upward force equal to the force created by the acceleration of gravity: F = ma where F = force, m = mass, a = acceleration.
    When any portion of the seawater is replaced by air, this section of the ship would be supported by the same equation (rho x g x z) but rho is now the density of air. Expressed as a ratio of the differing densities, the air bubble has .124% the density of water. If a 10 ton (displacement in seawater) section of the ship is supported by the bubble, air will support 24.8 pounds for the same displacement. That section of the ship will sag into the bubble. Bubbles of gas, whether air or expanding gasses from an explosion, will not support a ship nor will it impart an upward force.
    The upward motion or hogging of the ship’s hull is imparted by the shockwave, not a bubble or displaced water. A shockwave is produced by any wave traveling faster than the local speed of sound; in air, 343 m/s (767 mph), in seawater, 1,550 m/s (3,467 mph.) The detonation velocity of RDX, the main ingredient of Torpex, has a detonation velocity of 8,650 m/s (19,349 mph.)
    The following link is a very good lab experiment performed at the College of Shipbuilding Engineering, Harbin Engineering University, Harbin, China.
    The reading and equations are a bit dense, but the attached tables and media are very interesting. They largely ignore the shock wave as the introduction states, this is widely researched. This experiment focuses on the cyclic pressure loads of the bubble and the bubble jet. It mentions the negative bending moment specifically as the bubble makes contact with the hull.
    There is a photo of a scaled ship hogging due to the bubble. This is not due to buoyancy, but the cyclic pressure waves of the bubble as it collapses and expands.

  21. Your article was excellent! Then, I found myself following the comments with equal intensity, with the experts that you drew in being icing on the cake. Just WOW!


Comments will be moderated for posts older than 7 days in order to reduce spam.