Multi-hull warships

[Starglider]

I see that military use of multi-hulls seems to be continuing to expand beyond the well known Independence-class Littoral Combat Ship and various earlier experiments (even the Royal Navy is thinking about it. The advantages of drag reduction, higher stability and large deck area to waterplane ratio seem generally applicable though even to earlier eras.

Ignoring construction cost, risk and yard availability for the moment, how much sense would a catamaran and/or trimaran hull form make for;
a) modern supercarriers?
b) cold-war era non-VLS missile cruisers?
c) WW2 era carriers?
d) WW2 era destroyers and torpedo boats?
e) Battleships (e.g. do outrigger hulls provide sacrificial torpedo defense)?
f) Clipper ships (e.g. better cago volume/speed ratio)?
g) Age of sail ships of the line (might have to assume iron structural components to make this possible)?

Is there any plausible alternate history in which we might have seen such ships?

[Elheru Aran]

It's not a great concept for planked wooden ships like ships of the line and clippers. Too much drag versus sail area, never mind the inherent flexibility of wooden ships. Now if you could have converted, say, giant sequoia tree trunks into massive dug-outs, then that might have worked. More rigidity that way. And yes, you could have two sets of sails, but trying to sail into the wind would get a lot more tricky as one set of sails would tend to block the other; if you can only sail before the wind, that removes a lot of your tactical and navigational flexibility.

Not sure WW2 era materials science could make a robust enough trimaran or catamaran in battleship or carrier size. And that's about the extent of my knowledge so I'll leave it at that and watch what else ensues...

[Purple]

There really is absolutely no reason to build a twin hulled ship when you could for marginally more cost build two single hulled ones.

[Simon_Jester]

Can you prove that assertion, Purple?

[The Duchess of Zeon]

Starglider, I promise you detailed explanations when I am less tired, but suffice to say the main reason multihulls are becoming so common is that we have the computing power to optimize the design of each hull in relation to the others to obtain favourable rather than unfavourable hydrodynamic interactions between the hulls. Most of this would have held true a long time ago, and you could design a trimaran battleship with sacrificial protection (whether or not this would be weight effective due to the stability reserves, however, is questionable), say, but the computing ability to make the design functional wasn't there. It isn't very great, though, some of it was also experimental knowledge, so a trimaran battleship might work with say, Babbage's Analytical Engine having been evolved upon until we got absurd molybdenum hardened gear steam computers circa 1910 that could crunch the maths and we'd respected Polynesians enough to do the tow tests.

There are at least anecdotal evidences of catamarans in use in Hellenistic times as very large mobile riverine barges, but whether the structural strength to use them as warships on the open sea could have been obtained is questionable.

[Purple]

Can you prove that assertion, Purple?

Well, it's two assertions actually. So I'll take them in order.

1. You can build two ships for a marginally greater price than one double hulled one.
- A multi hull ship is going to require all the same facilities two separate ships do if you want it to have double the capacity of the single hull. It needs Engines and steering for both hulls lest you end up in an imbalance. It needs to have double, or nearly double the armament, armor and other equipment of two separate ships. And it generally takes as much effort to coordinate and control as those two ships. Finally, it needs quarters for the nearly double crew. So you basically have the case of there being two ships in one. And yes, it should be as symmetrical as you can make it as you want the two hulls to be balanced in the water. You do not want to have your battleship or better yet aircraft carrier listing heavily to one side.

2. It's better to pay extra and build two.
- Two ships, even though they cost more have the benefit of being in two places at once. Something which for the most part regardless of the ship class gives you a lot of strategic and tactical options. More importantly however, for many ship types having a twin hull would really not make a lot of sense because hull volume and deck space are not their true limitations. For example, the broadside of an Age of sail ships of the line is limited by the surface area of the sides of its hull, so all the deck space and internal volume is wasted if you use double the wood to make a ship with the same broadside. Dreadnaught battleships, and anything else with gun turrets is going to have problems firing one side of turrets from the other. And the limit to an aircraft carriers air wing are not space but command and control facilities and equipment.

Designs that emphasize hull space whilst making for a small profile such as these would be better suited to some sort of tanker or cargo ship that does not need to be fast, but needs to be stable and have a lot of volume without tipping over. And even than I'd rather have two ships because they will have an easier time finding suitable port facilities.

[Jub]

Can you prove that assertion, Purple?

Well, it's two assertions actually. So I'll take them in order.

1. You can build two ships for a marginally greater price than one double hulled one.
- A multi hull ship is going to require all the same facilities two separate ships do if you want it to have double the capacity of the single hull. It needs Engines and steering for both hulls lest you end up in an imbalance. It needs to have double, or nearly double the armament, armor and other equipment of two separate ships. And it generally takes as much effort to coordinate and control as those two ships. Finally, it needs quarters for the nearly double crew. So you basically have the case of there being two ships in one. And yes, it should be as symmetrical as you can make it as you want the two hulls to be balanced in the water. You do not want to have your battleship or better yet aircraft carrier listing heavily to one side.

2. It's better to pay extra and build two.
- Two ships, even though they cost more have the benefit of being in two places at once. Something which for the most part regardless of the ship class gives you a lot of strategic and tactical options. More importantly however, for many ship types having a twin hull would really not make a lot of sense because hull volume and deck space are not their true limitations. For example, the broadside of an Age of sail ships of the line is limited by the surface area of the sides of its hull, so all the deck space and internal volume is wasted if you use double the wood to make a ship with the same broadside. Dreadnaught battleships, and anything else with gun turrets is going to have problems firing one side of turrets from the other. And the limit to an aircraft carriers air wing are not space but command and control facilities and equipment.

Designs that emphasize hull space whilst making for a small profile such as these would be better suited to some sort of tanker or cargo ship that does not need to be fast, but needs to be stable and have a lot of volume without tipping over. And even than I'd rather have two ships because they will have an easier time finding suitable port facilities.

Purple, did you actually look at any of the double/multiple hull designs that Starglider linked to before typing this? They're not two complete ships that happen to meet in the middle, but catamaran designs that have outriggers enabling more speed and stability than could be had for the same tonnage with a conventional design. Of course things won't work correctly if you do things in such a simple minded way.

[Starglider]

Starglider, I promise you detailed explanations when I am less tired, but suffice to say the main reason multihulls are becoming so common is that we have the computing power to optimize the design of each hull in relation to the others to obtain favourable rather than unfavourable hydrodynamic interactions between the hulls.

I didn't realise this was so difficult. Is it really impractical to do this with tank tests?

A multi hull ship is going to require all the same facilities two separate ships do if you want it to have double the capacity of the single hull. It needs Engines and steering for both hulls lest you end up in an imbalance.

This is only a problem for catamarans, not outrigger trimarans, and even for catamarans it is not an issue for warships above destroyer size as they all have two or more engines and screws anyway, for steering and redundancy. Using a catamaran form inherently gives you compartmentalisation between the two engines.

It needs to have double, or nearly double the armament, armor and other equipment of two separate ships. And it generally takes as much effort to coordinate and control as those two ships

Side armour would not be increased because you don't need to armour the inside faces of the hulls. Armament would not be increased at all since it it payload above the waterline; you are using two or three narrower hulls to support the same amount of topweight as a single wider hull. That said while mounting centerline turrets on a trimaran should be easy enough (and maybe more stable than a monohull for broadsides), it might be structurally impractical for a catamaran. For a PT boat though you could just drop torpedoes into the forward inter-hull space to launch them instead of having tubes in the bow.

I don't know why you think it takes any more effort to control. Above the waterline, a catamaran ferry or sailing yacht operates just like a monohull. The wider separation of screws makes engine steering more effective, but the wider beam can make it harder to dock. I admitt the dual rudder linkage might be challenging for early age of sail technology.

And the limit to an aircraft carriers air wing are not space but command and control facilities and equipment.

This is at odds with everything I've read about WW2 carriers, where the limitations are primarily deck and hanger space, and secondarily fuel and armament storage. A wide multi-hull fleet carrier could potentially solve the issue of separate launch and recovery areas before angled decks were invented (via a flight deck on each side; prop planes could use narrower ones than current jets). Even if you did need more 'command and control' facilities, the larger above-waterline volume to hull volume that you get in a multi-hull design would help.

For example, the broadside of an Age of sail ships of the line is limited by the surface area of the sides of its hull, so all the deck space and internal volume is wasted if you use double the wood to make a ship with the same broadside.

AFAIK the limitation on age of sail broadsides was stability preventing heavier guns from being placed on anything but the lowest gun deck; three-deckers in particular had to have light guns on the top gun deck. Secondly being able to place the guns further above the waterline was an inherent advantage when artillery ranges were so low. So I was thinking that perhaps a multi-hull ship of the line (configured like a modern jumbo sailing yacht) could mount heavy guns on the second and even third deck, due to the stability advantage. The lower hull drag would preserve speed over a traditional first-rate that would get bogged down by all the weight. If Duchess is correct then the drag advantage wouldn't exist due to unoptimised design, but it should still have the stability advantage.

[Simon_Jester]

Starglider, I promise you detailed explanations when I am less tired, but suffice to say the main reason multihulls are becoming so common is that we have the computing power to optimize the design of each hull in relation to the others to obtain favourable rather than unfavourable hydrodynamic interactions between the hulls.

I didn't realise this was so difficult. Is it really impractical to do this with tank tests?

The problem with a tank test, I speculate, is that if it turns out your catamaran hull isn't performing satisfactorily, you can't necessarily figure out how to do a better job without computer simulation.

[Elheru Aran]

Tank tests are useful but they can't do everything. At some point you would need to test it full-size IRL (expensive if it fails). Using computer modeling can help eliminate much of the need for IRL testing by simulating elements that you can't get in a tank such as precise input on how hull variances affect water flow and what not; instead of having to do the math and possibly erring, the computer does the math for you.

[Sidewinder]

It needs to have double, or nearly double the armament, armor and other equipment of two separate ships. And it generally takes as much effort to coordinate and control as those two ships

Side armour would not be increased because you don't need to armour the inside faces of the hulls.

Yes, you do, for torpedo defense.

The torpedo is designed to detonate under the keel of a surface ship, breaking the ship's back and destroying its structural integrity.

[Simon_Jester]

On a modern warship you do. On a World War II battleship, you don't, because nobody at that time had a reliable under-the-keel torpedo that could tell the difference between blowing up under a battleship's outrigger and blowing up under her keel.

[Titan Uranus]

If you are assuming iron structural components for the age of sail vessels, then the iron would probably be better used for ironclads. In addition, it is my understanding that these Bulls are simply harder th make, even ignoring the design difficulties, depending on how much more difficult they are to make, that might cause them to be ignored during WW2 even in a universe where they were common before the war.

In addition, aren't multihulls much more affected by bad weather (particularly in their speed) than monohull ships?

[Elheru Aran]

They aren't that much harder to make than regular ships; merely extra working time and materials required. The trick is structural strength-- connecting multiple hulls rigidly in an efficient manner while preserving strength for seaworthiness-- and power efficiency-- do you have strong enough engines (or sailpower) to compensate for the extra drag caused by two or three hulls? So the building isn't an issue as long as the builders know their job and have the appropriate materials and resources.

[Purple]

Yea, my mistake. The images would not load to me and I just had a heated debate with a guy the other day about the idea of basically stitching two aircraft carriers together at the hip to get double the deck space where he pointed at some cold war concept art. So I figured this was the same thing. My mistake.

[Thanas]

There are at least anecdotal evidences of catamarans in use in Hellenistic times as very large mobile riverine barges, but whether the structural strength to use them as warships on the open sea could have been obtained is questionable.

Also the largest warships were supposed to be catamarans, especially the Egyptian and very large Phoenician ships, though these were by all accounts either unsuited for combat or large mobile artillery platforms.

[Titan Uranus]

They aren't that much harder to make than regular ships; merely extra working time and materials required. The trick is structural strength-- connecting multiple hulls rigidly in an efficient manner while preserving strength for seaworthiness-- and power efficiency-- do you have strong enough engines (or sailpower) to compensate for the extra drag caused by two or three hulls? So the building isn't an issue as long as the builders know their job and have the appropriate materials and resources.

Yes, but in a total war even a small difference in material useage is liable to be important, and a ship which requires even 10% more structural steel or 10% more powerful engines than a more or less equivalent tradional design would not get built. The extra construction time is also a major problem, as long as it is in any way relevant.

If the extra construction time is relevent, then that precludes it's use for almost all of the war-expedient ships, CVE's, destroyer escorts, freighters, most of the destroyers, etc.
In addition, if I recall correctly multihulls are only faster if they are light ships moving at high speeds, if true, that would mean that you would have to calculate the benefits of better stability against the benefits of higher speed for vessels of any significant size. If that prewar calculus comes out on the side of higher speed for the battleships specifically, then the carriers, cruisers, and fleet destroyers would all have to match that speed.

That would essentially leave only the torpedo boats to take advantage of multihull designs.

Although I honestly do not know, and right now I'm pretty sure that this entire thread's only purpose is to come up with questions to ask Dutchess whenever she feels up to it.

[Terralthra]

Starglider, I promise you detailed explanations when I am less tired, but suffice to say the main reason multihulls are becoming so common is that we have the computing power to optimize the design of each hull in relation to the others to obtain favourable rather than unfavourable hydrodynamic interactions between the hulls.

I didn't realise this was so difficult. Is it really impractical to do this with tank tests?

The problem with a tank test, I speculate, is that if it turns out your catamaran hull isn't performing satisfactorily, you can't necessarily figure out how to do a better job without computer simulation.

It's still worse than that. Not meaning to step on the Duchess's toes, but turbulent flow in fluid dynamics is still not even close to a solved problem. Simulations are ever improving, but they still aren't anywhere close to e.g. solid kinematic simulation.

[Sea Skimmer]

Side armour would not be increased because you don't need to armour the inside faces of the hulls.
Yes, you do, for torpedo defense.

The torpedo is designed to detonate under the keel of a surface ship, breaking the ship's back and destroying its structural integrity.

Hmm that isn't that a big deal because, the normal TDS armor won't help with underbottom attacks like that. The gas bubble more or less MUST reach the surface, and the ship is in the way and full of air, which is highly lacking in resisting power. The hull bottom is going to be blown in.

Only Yamato among battleships actually built had any armor on the hull bottom at all, and only then under her magazines only. She was awesome like that, and also had the magazines raised well off the hull bottom. The counterside of that is it made her magazines more vulnerable to shellfire, higher placement relative to the waterline increases the chance a penetration would actually reach one and explode within it. Yamato was so huge she could make this work, its one of the reasons why you can also have a reasonable looking 9x18in gun battleship on about 50,000 tons.

US tests and some war experience indicated that neither armor or a triple hull bottom actually improved resistance to underbelly attack any more so then simply building a normal double bottom stronger. And doing that will only help so much without just building a bigger warship.

Vertical anti torpedo armor wouldn't be used on outrigger hulls because unless we are talking about a huge ship those outriggers won't be big enough to hold an effective bulkhead based TDS. They'd just be one compartment wide, though secondary gun magazines might have patches of armor purely to resist detonation of the magazine by fragments from the hull and torpedo warhead itself, not to stop flooding.

Shear size and strength of hull will let a ship start to resist under the keel blasts but you have to get to be something like a CVN before that's going to make much difference over a random battleship. The fuel or ballast water in the double bottom will help spread the gas bubble slightly but your really just terminally screwed if more then a couple hundred pounds of torpedo warhead detonates very closely. Under the keel rounds usually go off some depth below the target and not always with exact precision of placement under keel though; modern active, as opposed to WW2 passive magnetic fusing remains imperfect. Passive magnetic fusing was only useful if set to local conditions. Active fuses generate their own magnetic field to sense, which takes considerable power to be effective, but will work about the same anywhere on the planet.

Multi hull designs do have an advantage in that they have more distributed hull strength, making them less vulnerable to any one hit in theory provided it can maintain stability. The problem is if we are talking about a ship for a gunnery action the span to the outriggers is probably going to require a huge amount of topside armor. You can use the outriggers to shrink the ships length and shield the inner hull from torpedoes, but the minimal armored length is set by the boilers and armament already. And unless this is before WW1, at least some deck armor has to be present everywhere. Even 1-2 inch armor adds up rapidly.

Any torpedo hit under the span area also will penetrate UP into the underside of the structure, the water has more then enough force for that, though I would presume armoring against this is possible if you really wanted.

If you don't armor the topside and ends of the outriggers meanwhile they'd be incredibly vulnerable to raking fire and bombs tearing out whole chunks of girder. I really don't see a way around this for a heavily armored ship, and I question it on a cruiser because of how vulnerable they already were structurally to heavy aircraft HE bombing, and likely to be shot numerous times by smaller weapons. Bombs of 1000lb and over are simply 'big' relative to a 10kt cruiser in the same way a battleship can be over matched by a ship with much bigger guns.

[madd0ct0r]

Starglider, I promise you detailed explanations when I am less tired, but suffice to say the main reason multihulls are becoming so common is that we have the computing power to optimize the design of each hull in relation to the others to obtain favourable rather than unfavourable hydrodynamic interactions between the hulls.

I didn't realise this was so difficult. Is it really impractical to do this with tank tests?

The problem with a tank test, I speculate, is that if it turns out your catamaran hull isn't performing satisfactorily, you can't necessarily figure out how to do a better job without computer simulation.

It's still worse than that.

Not meaning to step on the Duchess's toes, but turbulent flow in fluid dynamics is still not even close to a solved problem. Simulations are ever improving, but they still aren't anywhere close to e.g. solid kinematic simulation.

Water molecules and viscosity don't scale with your tank model. Things like Reynolds numbers become not your friends. It is possible to scale model with water, but you are always sacrificing something. A large area flood map model, for example, needs exaggerated vertical relief to achieve proportionally close to the same flows.

I am intrigued by the possibility of positive hydrodynamic interactions between hulls, I wasn't aware that was even possible. Is this the wash from one hull reducing friction further back on the second one?

[LaCroix]

I am intrigued by the possibility of positive hydrodynamic interactions between hulls, I wasn't aware that was even possible. Is this the wash from one hull reducing friction further back on the second one?

I'm thinking it might be related to hull speed, an important thing when designing your boat.

For a normal hull, maximum hull speed is determined by the length of the waterline. You are generating a bow wave and a stern wave. At a certain speed, you are basically riding in the valley of a standing wave that is created by the bow and stern wave meeting up. At this point would need to ride UP onto your bow wave to increase speed, thus your energy requirement to increase speed starts to rise exponentially. For sail boats, this is usually where things hit the virtual brick wall. (Unless you use tricks like planing - design the hull so it basically "pops" out of the water at high speed ans starts gliding on top of the water instead if in it. But that is only possible for quite light boats.)

There are ways to counter this effect - make your ship very long, make the bow as sharp as you can to minimize the bow wave, or use a bulbous bow. A bulbous bow disrupts this wave by creating another wave that cancels this effect, to a degree, allowing for higher hull speed than a normal bow. A correctly designed multi-hull ship could use the overlap of bow waves to disrupt the pattern in the same way.

[madd0ct0r]

I am intrigued by the possibility of positive hydrodynamic interactions between hulls, I wasn't aware that was even possible. Is this the wash from one hull reducing friction further back on the second one?

I'm thinking it might be related to hull speed, an important thing when designing your boat.

For a normal hull, maximum hull speed is determined by the length of the waterline. . (Unless you use tricks like planing - design the hull so it basically "pops" out of the water at high speed ans starts gliding on top of the water instead if in it. But that is only possible for quite light boats.)

define light: http://upload.wikimedia.org/wikipedia/c ... oyager.JPG - img tags broke formatting

[LaCroix]

define light: http://upload.wikimedia.org/wikipedia/c ... oyager.JPG - img tags broke formatting

Applying 100000 shaft horsepower and a catamaran configuration is kind of cheating...

[madd0ct0r]

well, we're talking about multi-hulls in a naval context - I can see the value in rapid response smaller support boats.

Not sure about the need for a hydroplaning super carrier. How fast a circumnavigation do you need?

[Irbis]

Water molecules and viscosity don't scale with your tank model. Things like Reynolds numbers become not your friends. It is possible to scale model with water, but you are always sacrificing something. A large area flood map model, for example, needs exaggerated vertical relief to achieve proportionally close to the same flows.

Yes. Tank testing becomes useless for a lot of cases, since water is a lot more viscous than air and tends to behave very differently with scale, non-linearly, at that. Testing single hulled models is generally much better understood than innovative shapes, so you can compensate here a bit, when you try something new often what you get has no semblance to reality.

Case in point - Imperial Russian battleship Novgorod that promised superior performance to conventional ship in a lot of ways during small scale testing turned out to be unstable, wobbly turd jumping in unpredictable ways when main guns were fired:

http://englishrussia.com/2012/07/22/rou ... n-admiral/