The conventional sailing ship has a single hull. However, multihulls—two or more hulls joined together by a deck or poles—can be found in the seventeenth century in both the Indian Ocean and the South Pacific.
The chief advantage of the multihull is extreme lateral stability, which in turn means that it can carry more sail over narrower hulls, and can forego ballast. Thus, it can be built for high speed.
The twentieth century debate over whether multihulls are better than monohulls was about as passionate as the seventeenth century theological ones of the Catholics and the Protestants. (Fortunately, monohullers aren’t allowed to burn multihull advocates at the stake.)
First, I am going to try to whet your appetite for learning more about multihulls by talking about the missions that they conceivably could be designed to perform. After that, I will talk about the types of multihulls, their history, and the nitty-gritty of designing them. Expect this section to be optimistic in tone; I will talk about problems later.
Courier. The racing catamaran designs could be adapted for use as military couriers. Since these carry information, not goods, they can be designed with the light displacement necessary for planing. They will rely on their speed to evade enemy warships.
Landing Craft. Small Hawaiian catamarans were able to reach land despite heavy surf, when conventional ship’s boats were forced to stay offshore. (Morton 63).
Passenger/Vehicle/Fast Freight Transport. In 1997, 43% of the high-speed ferry fleet were catamarans. (Sahoo). The benefits of the catamaran design include large deck areas, shallow draft, high stability, and energy-efficient high-speed power cruising (Van Leer). It’s not a very large step from a passenger ferry to a troop transport. A multihull could be used, like a nineteenth century clipper, to carry perishable or high value/low density cargo.
Tankers. The high initial stability of multihulls, especially catamarans, gives them an advantage in carrying liquids, which can slop around and destabilize an ordinary ship.
Warships. Multihulls should be more stable firing platforms, and if adequately powered, they are faster than monohulls. And the cross-structure deck area provides room for additional armament. (See “Fighting Ability” below.)
Offshore Drilling. The open deck of a catamaran could be a drilling platform for exploratory drilling in lakes, rivers, and shallow coastal waters. Examples would be Caddo Lake, Louisiana; the delta region in Nigeria; Lake Maracaibo, and the waters off Long Beach, California. The catamaran could have a “well” at its center, where motion is minimized, for lowering a drill. (Van Leer).
Aircraft Carrier. The open deck of a multihull might be handy for launching a tethered balloon. This could be used for both observation and communication. In 2003, the trimaran RV Triton was used as the launch vessel for a manned balloon. (See “Air Support” below.)
Types of Multihulls
The catamaran has two equal-size “demihulls,” whereas the proa has a main hull and a single, narrower outrigger. In the standard catamaran, the hulls are side-by-side, but hulls can be staggered (bow of one forward of the bow of the other) to reduce wave resistance. In the extreme variant, the “Weinblum,” the stern of one hull is forward of the bow of the other. There has also been experimentation with asymmetric hull shapes. (Zaraphonitis).
The trimaran has three hulls. These could be equal-sized, but more usually they are a main hull and two narrower outriggers. The outrigger hulls can be shorter than the main hull, and they can also be set in an arrow configuration with their midsections aft of the main hull midsection. In an extreme version, their bows are aft of the stern of the main hull, as in Cousteau’s Alcyone, so what you have is a “monocat,” with a single hull fore and two hulls aft.
The four-hulled tetramaran has been the subject of some theoretical analysis. You could have, side-by-side, either four identical hulls, or two main hulls like a catamaran as well as two outriggers like a trimaran. Or a two-by-two, with two leading hulls and two trailing hulls. Or modify that into a blunt arrow (“slice”) configuration, by placing the trailing hulls further apart. Or make a “diamond”: four identical hulls, with the two central hulls virtually nose-to-tail, and the other two hulls flanking them.
A pentamaran has five hulls. In the Nigel Gee designs, it has one central hull and four short outriggers (“sponsons”). One pair, kayak-shaped, flanks the midsection of the main hull, and the other pair, rowboat-shaped, follows behind it. When the ship is upright, the aft sponsons kiss the water, but the forward sponsons are above it (Gee). I have seen theoretical analysis of a pentamaran with one large, two medium, and two small hulls, all side-by-side.
Pre-Ring of Fire Asia
In 1521, Magellan encountered the single outrigger (proa) canoes of Guam, which a Spanish chronicler likened to “dolphins, jumping from wave to wave.” (Levinson, 84). In 1616, Le Maire and Schouten saw a Tongan double canoe (catamaran), and another was seen by Abel Tasman in 1643. As for double outrigger canoes (trimarans), Drake saw them in 1579 in the Caroline Islands (Morton 75), and Tasman saw them near New Ireland (59). There is also an Easter Island petroglyph depicting a double canoe, possibly with one mast per hull. (Kane).
It is likely that even in this early period there were numerous variations, from one Pacific island to another, in hull shape and crossbeam design (Harvey 5; Morton 59–76).
The proas were fast; Anson thought they could make twenty knots (Morton 72). And Polynesian multhulls weren’t just fishing boats; some were over 100 feet long (63), and could carry several hundred men (63). Eighteenth- and nineteenth-century pirates, from Borneo, Malaya and the Philippines, used proas much as the Barbary Coast corsairs used xebecs; to overwhelm sailing ships that were becalmed.(Morton 76; Warren, 170).
OTL Europe and America
The first European proa was built in 1860. Proas are capable of planing; Munroe’s thirty-foot 1898 proa could travel at eighteen knots (2.5 times hull speed).
European experiments with catamaran designs date back at least to the 1660s, when Sir William Petty (1623–87) built three single-masted, double-bottomed catamarans, Invention I (1662; 30’LOA, 1.75 tons), Invention II (1662; 30 tons, carries 30 men; hulls 20′ x 2′), and The Experiment (60’L, 16 guns). The Invention II won a race with the Holyhead-Dublin packet boat, and The Experiment held its own with three fast vessels of similar size. (IWHR; McMullen 24). The Experiment was able to “come within less than five points of the compass, some say very much less.” (A square rigger typically couldn’t sail closer than six points of the wind.)
In the 1780s, Patrick Miller constructed a large catamaran (235 tons; 100’L, 31’B, five masts with square sails and five paddle wheels between the hulls). Miller also built the first European trimaran.
Robert Fulton built a peculiar 20-odd gun steam battery, the Demologos (1814). While it has been called a “catamaran” by many popular sources, it would be more accurate to think of it as a monohull divided longitudinally into three compartments; the middle one, which was 15 feet wide, holding the paddlewheels. This was partially flooded as a result of water welling up through a 66 feet long “race” at the bottom. (Photos NH74702, NH65481, NH61883; Bauer 53). The point of the design was to protect the paddlewheels from enemy fire. However, the middle compartment would have greatly reduced the lateral stability provided by the large beam; if the ship heeled, the water would surge, shifting the center of gravity in the destabilizing direction.
Colonel Stevens used a trimaran “horseboat” on the Hoboken ferry line in 1814. In essence, the horses were on a treadmill and this turned the paddle wheels between the center and side hulls. (Baxter. 19). The same line also employed a double-ended catamaran steamboat. The hulls were 80 feet long, 10 feet across, and 5 feet deep, with a hull separation of 10 feet (so the deck across the hulls was 20 feet wide). The waterwheel was between the hulls, and there was a cabin 50 feet long and 10 feet wide. (20).
In 1862, a catamaran snagboat was converted into the ironclad river gunboat USS Benton, and served as Commodore Foote’s flagship. But it appears that the builder used the old catamaran hulls “as a pair of bracers,” connecting them at both top and bottom. (Konstam 10; Slagle 190).
A very large twin-hulled ship (290 feet long, 60 feet wide), the steamer Castalia, was launched in 1874. She had a draft of only six feet, which let her freely enter “tide-controlled” ports. Nonetheless, she was resistant to rolling. (Rogers 65–7).
In 1876, Nathaniel Herreshoff’s Amaryllis (24’L) catamaran won the New York Yacht Club’s Centennial race against over thirty monohulls, ranging up to 40′ long. The racing officials disqualified it, and later barred all catamarans.
However, multihulls made a serious comeback after World War II, thanks to “new” materials such as aluminum, plywood, fiberglass and most recently carbon fiber composite. They are popular for private racing and cruising, and dominate the passenger ferry market. Nonetheless, few warships and no dedicated cargo ships are multihulls.
Will their fate be different in the new time line?
Multihulls in Canon
According to canon (1633, Chap. 4), the timberclads used in the Baltic War campaign are catamarans, with paddle wheels positioned in-between their hulls. Unfortunately, we know nothing about their dimensions. Their success will give the catamaran design a certain degree of credibility that it would not have possessed previously.
As for the ironclads, Simpson ” . . . built them around what was effectively a double hull. Each of the propulsive pumps—and the tunnel in which it worked—occupied its own individual “pod,” separated from the rest of the hull (and from one another) in order to prevent them from being disabled by a single hit or hull breach. It was almost a catamaran effect . . . .” (1634:The Baltic War, Chap. 44). This sounds much like the Demologos or Benton.
The ironclads are “low, squat,” “slab-sided,” looking something like the CSS Virginia (although probably lacking a ram). The dimensions are 500 tons displacement, about 120 feet long, and minimum draft (with trim tanks empty and centerboard up) of 4 or 5 feet. (1633, Chap. 28; 1634: The Baltic War, Chaps. 31, 52). I would guess a maximum beam of 30–40 feet, because of river travel limitations. I suspect that they are smaller but more heavily armored versions of the Union “City-Class” ironclads.
The basic parameters of the catamaran are its length, depth of hold, and hull width and spacing. (For a trimaran there is also the size ratio of the hulls.) Its effective beam, for stability purposes, is the sum of the hull widths and the clear space between them. The weight of the ship and its contents will determine its displacement and thus its draft.
Modern pleasure catamarans have length/overall beam ratios about 1.5–2:1 (with racers up to 3:1), and trimarans 1–1.5:1 (Shuttleworth; McMullen 56). A wide beam is advantageous from the standpoint of stability and wave interaction, and disadvantageous in terms of stress on the cross-structure. Generally speaking, the trimaran will have a greater overall beam than a catamaran of equal length, because the stresses are divided over two connecting structures. (Harvey 41, 60).
Pleasure catamaran demihulls are slim, with length/width ratios in the range of 8–16:1, with 10–13:1 being most common. For trimaran main hulls, that ratio is 8–10:1. (Harvey 58ff; White 53ff).
According to a 1990 commercial catamaran survey (two-thirds fast ferries), the ships were mainly 10–40 meters, with length/hull width 6–12:1, depth/draft 1.5–3.5:1, and centerline hull separation of 20–45% demihull length and 150–240% demihull width. (Insel 13ff). Catamaran fast ferries constructed 1994–2003 had waterline lengths of 37–106m, and beams of 9–30m. They carried 150–1200 passengers and 10–312 cars. (Soare 1184).
An open-deck (OD) multihull provides a minimal cross structure; a network of rods, springs or cables that connect the hulls, and some kind of net or trampoline surface so that the crew can scramble from one hull to another. This cross-structure is very vulnerable to the forces that drive the hulls apart (see “Structural Integrity”). Herreshoff provided joints which allowed for a certain amount of independent rolling and pitching of the hulls, thereby reducing the stress on the connections (USP189459; Kemp 353).
The next step up is a flat solid-deck (SD). This offers greater strength, and can carry deck cargo or armament. If the ship rolls over enough to “fly” a hull out of the water, the exposed underside increases the windage and thus the risk of a capsize
The most elaborate cross-structure is a higher profile “bridge-deck” (BD); it provides an enclosed area for crew or cargo and thus has at least two decks (floor and ceiling). Hence, both the roof and floor (“wet deck”) of the bridge are holding the hulls together. A high profile deck adds weight and increases windage but, if it remains watertight, provides reserve buoyancy.
A full bridge-deck is one that runs the entire length of the hull and is common in catamaran ferries. A pleasure boat is more likely to have a short, highly streamlined bridge. Usually, there is just netting in the forward third of the ship, so a wave can’t push the bow up.
The so-called SWATH (small waterplane area twin hull) design, invented in 1938 (as Frederick Creed’s proposal for an aircraft carrier!), combines an elaborate cross-structure with underwater hulls. This isolates the hulls from wave motion (if the hulls are deep enough), but of course it also increases draft, and you need one long (Duplus-type) or two shorter (Kaimalino-type) wing-shaped vertical struts to connect each hull to the cross-structure. In 2000, about fifty SWATH ships were in operation or construction (Dinsmore). A variable draft SWATH has ballast control, like a submarine, so it can either rise up to enter shallow harbors or sink down to minimize wave response.
If the side hulls of a trimaran are the same height as the main hull, then a deck can be built across all three.
If the side hulls are shorter, there are three options. First, the side hulls may be suspended directly from the crossdeck, drawing less water than the main hull. (See USS Independence, 2008). Secondly, they can be mounted on vertical or curved struts that come down from the main deck so the keels are even. Finally, they can be mounted on wings that extend from one of the lower decks of the main hull.
You can also build SWA trimarans; you can put just the main hull underwater, or just the outriggers, or all three hulls. (Dubrovsky).
Stability and Seaworthiness
A ship can heel over (tilt to one side) as a result of crosswind pressure on sails, waves coming against the beam, recoil from firing a broadside, and turning. A self-powered catamaran doesn’t have to worry about the first of these, but the others still apply.
The great advantage of the multihull is lateral stability (resistance to heeling). Imagine a monohull ship in the form of a rectangular block that floats in water. If you cut the original hull in half lengthwise, and connect the halves with some lightweight crossbeam, the resulting catamaran has the same length, displacement and draft as the original multihull, yet the metacentric height (initial stability is metacentric height times displacement) is increased. If the space between the halves is equal to the total widths of the hulls, thus doubling the effective beam, the increase in metacentric radius is 4.8-fold (Biran 65), and, if depth is half-breadth, density 0.5, and center of gravity at half-draft, the increase in metacentric height is seven-fold. The improvement in stability is such that a multihull doesn’t have to carry ballast.
The Navy conducted a study of the effect of prolonged wave-induced rolling. Only 10 degrees roll cut crew efficiency by 50%; 20 degrees by 80%. (White 23).
On the other hand, a multihull does have some seakeeping weaknesses. If the freeboard (height of main deck above water) is unchanged, then simple geometry dictates that because of the greater beam, the deck is immersed at a lesser angle, and stability decreases after that immersion angle is passed. The heel angle of maximum stability might be 6 degrees for a catamaran, 20 for a trimaran, and 60 for a monohull (Shuttleworth).
Likewise, multihulls are likely to have a lower angle of vanishing stability (heel angle at which the ship no longer has a tendency to right itself) than a corresponding monohull. Nonetheless, the dynamic stability (total work which a wind or wave must do in order to capsize the ship might be 50% more for a multihull than a monohull.
Monohull warships have a relatively high center of gravity because of armor and armament. This would reduce their range of stability if they weren’t given a large metacentric height to compensate. This makes them “stiff”; if they heel over, they will snap back too fast and this can cause discomfort for those on board, or even dismast a sailing ship. Having a short roll period also means that they are more likely to encounter sea conditions in which the wave period matches the roll period, and that exacerbates the rolling motion. (Atwood, 69ff).
Multihulls naturally have a large metacentric height. However, the roll period is lengthened if there’s a lot of weight distant from the roll axis. That’s certainly true of an open-deck catamaran, which has almost all its weight in the hulls (Shuttleworth), and it’s true to a lesser degree of trimarans and catamarans with elaborate crossdecks. (A tall mast helps, too. White 194).
A ship pitches as well as rolls. Because ships are longer than they are broad, they have more resistance to pitching. Because multihulls are squatter than monohulls, their pitch period is closer to their roll period, and this can result in a very unpleasant motion called “corkscrewing.” This can be alleviated by redistribution of weight.
A multihull running with too much sail up can pitchpole (somersault forwards). Pitchpoling can also be caused by the bow being buried in rough seas, which is why you don’t want to have a simple solid crossdeck extend all the way to the ends.
Monohull proponents complain that if a multihull capsizes, its multihull’s lateral stability becomes a disadvantage, as it’s just as stable upside-down as rightside-up. Multihull fans retort that a monohull is just as stable on the ocean floor as on the surface.
SWATH designs will have a larger beam than a conventional catamaran of the same displacement to compensate for their relatively small waterplane area, which determines their initial resistance to wind-induced heeling. (swath.com). However, because the SWATH hulls are deep underwater, they are virtually immune to wave action.
Multihulls can be powered by the wind (as captured by sails) or be self-powered by combustion of fossil fuel.
Trimarans are essentially monohulls with outriggers, and the mast(s), if any, will be placed on the main hull.
For sailing catamarans, there are two choices. First, to place the mast in the center, i.e., on the cross-structure. If the multihull is open-deck, the main beam, on which the mast rests, must be strong enough to support it. Indeed, the Hawaiians, working with wood, stepped the mast on a longitudinal beam that distributed the downward thrust over three or more crossbeams. (Kane).
The second option is to put the mast(s) on the hulls. In the “biplane” (parallel) mast array (DUO 425), there’s one mast on each main hull. The problem is that the masts can only be stayed on the “inboard” side. A variant on this are A-shaped “bipod” masts (SMG50), in which the masts are mounted on the hulls but meet above the crossdeck.
It is worth noting that by putting two short masts on the hulls, rather one long one on the cross-deck, you lower the center of effort and therefore reduce the “heeling” action of the wind (see Stability).
The first transatlantic record-setting self-powered ships used a steam engine to drive a paddle wheel (Great Western 1838, 14 knots); later ones powered a propeller (Napoleon 1857, 20 knots). The steam engine was replaced, first with a steam turbine (Mauretania 1907, 25 knots), then with a diesel engine (Normandie 1935, 30 knots). Large warships in which the was driven by a gas turbine (which has a high power to weight ratio) appeared on the scene in the Sixties. Finally, the propeller was replaced with a hydrojet, powered by a diesel engine (catamaran Hoverspeed Great Britain 1990, 37 knots) or a gas turbine (monohull Destriero 1992, 53 knots). (Pinder 143).
Paddlewheels can be placed between the hulls for protection, but propellers and hydrojets would probably be mounted on the hulls themselves. This is problematic in the case of SWATH designs; “until very recently it has been very difficult to package much power in the submerged bodies.” (Friedman).
Simpson’s timberclads had paddlewheels, whereas the ironclads “used powerful diesel-driven pumps scavenged from the Grantville coal mine to provide hydro-jet propulsion.”
On a hybrid (sail/steam) ship, it’s advantageous to be able to lift the paddlewheel or propeller out of the water when sails are in use, to reduce resistance. Patrick Miller’s Edinburgh trimaran had paddlewheels whose immersion could be varied. And the CSS Alabama had liftable screws.
On a hydrofoil-supported catamaran (“hysucat”), foils generate lift, like airplane wings, once the ship is in motion (Hoppe). Foils can be mounted below more conventional underwater hulls (see sailing biplane catamarans Techniques Avancées and Spitfire), or suspended (“midfoil”) from the cross-structure. To maintain pitch stability, two foils are better than one (Loveday 5).
In light winds, sailing multihulls are slower than monohulls of equal displacement because the two hulls experience greater frictional resistance. As winds pick up, multihulls can surpass the monohulls, because they can safely carry more sail (if the hulls are spaced far enough apart). Because of the great beam, the trimaran can carry more sail without capsizing, and thus has a higher maximum speed than a catamaran of equal length. However, three hulls cost more than two, and on a fixed budget, you may need to choose between a longer catamaran and a shorter trimaran.
For racing sail, the trimarans are superior in light winds (the outriggers are barely in the water and resistance is like that of a monohull) and the catamarans beat them when winds strengthen. (Howard 5).
In 1905, the three masted 56m schooner Atlantic (300 tonnes, 41.18 m LWL, 8.85m beam, 1720m2 sail; Ancko) established a record run of 12 days, four hours (10 knots) for the 2950 mile run from New York to Falmouth. The record, first broken by a trimaran in 1980, has been held exclusively by catamarans since 1988, and is now (2006) held by the catamaran Orange II, which made the passage in 4 days, 8 hours (28 knots). The same ship also holds the round-the-world record (50 days, 16 hours) and the 24 hour record (706 miles). Orange II has length 36.8m, beam 18m, mast height 45m, and 700 m2 upwind and 1000 m2 downwind sail area. (multiplast; Bernard Gallay Yacht Brokerage).
Of course, Orange II is strictly a racer. Her displacement was 28 tons light, 30 tons fully loaded, for her 2005 round-the-world record journey (50 days, 16 hours). And to achieve that low displacement, her hull, deck and sparring are all carbon composite. A catamaran intended to carry cannon or cargo would have a much lower sail power-to-displacement ratio, and wouldn’t have this turn of speed.
That said, any monohull whose maximum speed is limited by its ability to safely carry sail rather than its hull speed could be improved on by “slicing it in half” and redesigning it as a multihull.
The problem of frictional resistance can be alleviated by using a hull of semicircular cross-section (McMullen 35), the circle being the shape that provides the lowest ratio of perimeter to area and thus the least wetted surface for a given carrying capacity. (For stability reasons, a monohull can’t have that shape unless it’s heavily ballasted.) Even then, two hulls will experience about 40% more frictional resistance than a monohull of the same displacement and length (Stevens). The world’s largest sailing catamaran had about 50% more wetted surface area than an equal length monohull, despite having a smaller draft. (Lawless).
The disparity is smaller with low-displacement racing designs. If you take into account both elimination of ballast and beam/draft ratios, a catamaran equivalent of a YD40 yacht (ballast is 40% displacement) might have an increase in wetted surface area of only 8%.
On the other hand, the hulls of a catamaran might be made very narrow, to reduce wave resistance further, and this increases the ratio of surface to enclosed volume. This is taken to an extreme in the SWATH designs, where the wavemaking is by very narrow struts, reducing wavemaking by perhaps one-third (Dubrovsky), but the wetted surface area is perhaps 2.3 times that of an equal-length monohull. (Stevens). For this reason, they are usually shorter than a monohull of equal displacement (swath.com).
Of course, you can give the multihull an engine so it doesn’t have to worry about light winds. However, it will still need to put up with greater frictional resistance than would the equivalent powered monohull. The trick is to give it enough power so that its cruising speed is such that its total resistance is less than that of the monohull at that speed. And remember that reduced resistance, for a self-powered ship, translates into greater fuel efficiency. Modern “power cats” typically have fuel efficiencies of 10–20% at 6–10 knots and 40–70% at 15–25 knots (BYA).
Wavemaking resistance is small at low speeds, but more important at high ones. If we ignore interaction among the hulls, the resistance is proportional to the sum of the squares of the widths of the hulls. (Yeung). Hence “slicing a monohull in half” reduces that resistance by 50%.
The hull interaction can reduce or increase resistance; if you can control your speed (as you can with a self-powered multihull) you can choose one that benefits from the reduction. The hull separation/length ratio determines which diverging waves are reduced, and which strengthened, at a given speed by hull interaction. For a catamaran, Tuck and Lazauskas recommended a centerline-to-centerline separation of 20–30% hull length. They have also said that “at low and high speeds, we should keep the hulls close together, at intermediate speeds the hulls should be widely spaced.” (Lazauskas/Solar). Other sources recommend separations of 35% (Tasaki) or 40% (I&M) hull length. Interference falls off if your speed is high enough (above Froude number 0.6; “hull speed” is 0.4)(Loveday 16; I&M). If the hulls are asymmetric, there are additional complications, but interference was minimal with “tunnel” width 40% length. (Zaraphonitis).
The ship’s motion also creates transverse waves. As speeds increase, transverse wave resistance peaks first, and then it falls off as diverging wave resistance continues to worsen. Staggered hulls flatten out transverse waves by hull-hull interference. (Tuck/OHS). Transverse wave formation is also dependent on the slenderness of the hulls, and White (57) says that it’s significant only when the demihull’s length/width ratio is less than 8:1.
It’s conceivable that a trimaran could have the ability to vary the longitudinal and lateral outrigger spacing, much like a swing-wing aircraft. (Lazauskas/Solar). Variable width catamarans, with telescoping or folding cross-beams, have also been proposed.
Tuck/OHS mathematically compares a large number of multihull configurations with the same displacement (31.25 tonnes), length (19.1 meters) and draft (1.25) meters. Looking first at side-by-side hulls, the wave resistance was about the same up to 7m/s, but at higher speeds the tetramarans beat the trimarans which beat the catamarans. Up to 10 m/s staggered hulls were superior to side-by-sides. Among the staggered hulls, the diamond and slice were superior to the arrow.
If you looked at total drag, two hulls were better than three, and three better than four. If Tuck doubled the hull widths, thus doubling the displacement (and increasing wavemaking resistance), this shifted the advantage to the trimaran and tetramaran for speeds of 7 m/s up.
If you keep the displacement low, the multihull is likely to draw less water than the equivalent monohull, because it doesn’t need ballast for stability. Consequently, it will be able to enter shallower waters.
The low draft of a multihull is a disadvantage when it comes to sailing “upwind”; the multihull offers less “lateral resistance” to the side force exerted by the wind. A ship can increase lateral resistance permanently by having a keel that extends below the main body of the hull (which also permanently increases draft), or by dropping a vertical centerboard or daggerboards through a well in the bottom of the hull. Usually, a catamaran will have one on each hull, and a trimaran just one on the main hull. The latter is more effective and hence a trimaran is usually faster to windward.
A low displacement is also a disadvantage when changing tack; the multihull doesn’t have much momentum and therefore is prone to stopping when it’s pointing directly downwind. One solution is to build up speed before tacking; another is to gybe, that is, turn the stern rather than the bow through the wind.
Multihulls typically turn more slowly than monohulls because of the wider beam. Catamarans are slower than equivalent trimarans because the weight is further from the turning axis.
The greater beam is also a problem in constricted waters, such as rivers and small harbors. And docks may not be large enough to accommodate a large-beamed multihull.
In nineteenth-century river catamarans, central paddlewheels acted as unintentional “ice-catchers.” (Baxter, 21). This could become a problem with the USE’s timberclads.
Our ships are more than just racing shells, they have work to do. That means we need space to put passengers, cargo or armament. With a multihull, we have two choices: the interior of the hulls proper, or the crossdeck (if any).
If speed were our only concern, we would build narrow hulls, to minimize frictional and form resistance. However, a hull that is too narrow can’t be used for anything but to provide buoyancy. Figure that a large wine or tobacco barrel will have a diameter of about three feet. But of course you can’t fill the hull with cargo, you need some empty space for buoyancy. So for the hull to be used for cargo, it probably has to be at least six feet wide, and ten would be better. Frankly, you can probably stow more in the hold of a monohull with a ten foot beam than in two six foot demihulls.
An open deck catamaran with two ten foot wide hulls spaced twenty feet apart, a depth of hold of five feet, and a length of eighty feet, would have a cargo capacity in the hulls of something like eighty tons. So loaded, it wouldn’t sail like a racer, of course.
A closed deck greatly increases the cargo capacity of a multihull, since the multihull has a much wider beam than a normal ship. We can increase the cargo capacity even further by providing one or more covered decks. However, covered decks will perhaps be less important once containerization of freight transport takes place.
With a SWATH design, the space in the struts and lower hull is rather inaccessible, and best reserved for machinery and infrequently used equipment. The main living and cargo space will be in the cross-structure.
Displacement and Draft
So, yes, the big-beamed multihulls have lots of empty space. The problem for a multihull comes when you start making use of that space, thereby increasing the displacement. First, the wavemaking resistance is proportional to the displacement. Secondly, increasing the displacement increases the density of the ship and thus the draft, and if a multihull has skinny individual hulls, that rapidly increases the wetted surface area (Sailmagazine) and reduces the clearance of the cross-structure above the water. Also, adding decks, or storing cargo on the cross-structure, naturally raises the center of gravity. Increases in both that and density impair stability.
For a ship with vertical sides above the original waterline, there is a linear relationship between the added weight and the increased draft. For Lulu, a catamaran with elliptical hulls 96 feet long and 14 feet wide, 6 tons weight translated to one inch immersion. (Vandiver 41).
The problem is particularly acute for those full bridge-deck multihulls. Remember, when the specific gravity (density relative to water) of the loaded ship is 1.0, the ship is completely submerged. If the bridge-deck volume is twice the total hull volume, then when the specific gravity is just 0.33, the waterline is at the underside of the bridge-deck, which obviously isn’t acceptable. Acceptable loading is probably to a specific gravity of at most 0.2. For typical steel ships, a passenger ship is about 0.43 and a cargo ship 0.68–0.8. (Arnott 12).
Because of the sensitivity of performance to weight, designing a multihull is more like designing an aircraft than a conventional ship. (Van Leer).
Structural weight is roughly proportional to surface area, and so a multihull is going to weigh more than a comparable monohull. The structural weight of the world’s largest sailing catamaran was 85% higher than for the equal length NG380 monohull. (Lawless). A Naval Surface Warfare comparison of proposed 700 ton high-speed vessels had the structural weight of the catamaran being about 21% more, and the trimaran 33% more, than the monohull. For the monohull, structural weight was 20% of the full displacement.. (Nguyen). The structural weight/enclosed volume ratio for a steel hull is 5–6 pounds/cubic foot for a monohull and about 7.5 for a catamaran. (Stevens 25); water is 64 pounds/cubic foot.
Bear in mind that the fuel for self-powered ships contributes significantly to the displacement. For a steamship with a lightship displacement of 5500 tons that was 55.5 days at sea, it was about 1400 tons, leaving it with a cargo capacity of 11,400 tons. (SNAME).
Curiously, cannon for warships do not result in a huge increase in displacement. A 36-pounder might weigh 3 tons and require 10′ lengthwise and 12′ inboard of gun deck (MurrayS 16). If the gundeck were 5′ high, that would come to 600 cubic feet. That is a “stowage factor” (SF) of 200 cubic feet/ton. (The real SF is somewhat less, perhaps, as one must take into account ammunition, gunpowder and gun crew, but it should be close.) Cars, which are regularly transported by modern catamaran ferries, have an SF of 150—as compared to 18 for iron ore or 35 for water. The effective SF for passengers on those ferries is probably around 600, taking into account the standard deck area allowance per passenger. You can see why multihulls can profitably carry passengers and cars, but not bulk goods.
For some cargos—such as passengers—deck area is more important than volume. A catamaran typically offers 2–4 times the deck area of a monohull of equal length, and a SWATH design perhaps half that. (Dubrovsky).
The cross-structure must be strong enough to keep the hulls connected, despite a variety of forces that work against this.
Gravity. The first of those forces is gravity. The cross-structure can be modeled as a beam supported at its ends; it suffers a load as a result of its own weight and that of whatever it’s carrying. It will tend to sag, causing the upper surface to be compressed and the lower surface stretched. These stresses increase with an increase in the load, the distance between those surfaces, or the length of the beam (the clear hull separation). If they exceed the ability of the material to resist (mild steel has about four times the tensile strength of wood), the cross-structure ruptures and the hulls go their separate ways (briefly).
If we have a SWATH design, then the end-supports are the struts, which behave like slender columns, which are compressed by the weight of the cross-structure and can fail by buckling. The load which will cause buckling is inversely proportional to the square of the strut height. Unfortunately, it has to be high enough so the cross-structure isn’t slammed too often by wave action between the struts. Another approach is to increase the thickness; the buckling resistance increases as its cube. However, the thicker the struts, the more waves they make, slowing down the ship. The buckling resistance is linearly proportional to the “length” (parallel to centerline of the ship) of the strut, so longer SWATHs are less susceptible than smaller ones. Some materials resist buckling better than others; the resistance is measured by Young’s Modulus; iron and steel have a value perhaps fifteen times that of wood and three times that of aluminum; fiber-reinforced plastic is perhaps 75% as good as steel. If buckling occurs, then one side is compressed and the other stretched, and whether rupture occurs then depends on the factors from our beam analysis.
If the struts are too stubby to fail by buckling, they can still fail by crushing, which is dependent on the compressive strength of the materials of which they are made.
Wavemaking. Once the ship starts moving, new forces come into play. The flow about the hulls isn’t symmetric and side forces are created that usually push the hulls apart. (If the separation is small, there can be a Venturi effect that sucks them together). (Loveday).
Wave Action. In anything but still water, we must consider wave action. First, consider what happens under the cross-structure. Ocean waves, especially in a “lumpy” sea, can pass between the hulls. Also, the bow waves come together under the cross-structure. Either can pound the cross-structure from below if it’s too low. Then there’s slamming from above. A wave can break and come down on the deck. Or the ship can bury its bow in the water.
According to the second commander of the USNS Hayes, a catamaran research ship, her cross-structure cleared calm water by seventeen feet. That wasn’t enough to avoid pounding. To make matters worse, the cross-structure was “just plain housing structure designed to take no more than five-pounds per square inch of pressure.” Tests later showed that the pounding pressure in a heavy sea was 200 psi. However, he was able to ease the situation by going to full speed and turning the bow into the sea. (Slowbell).
The wider the catamaran, the lower the hulls will ride on the shoulders of a wave in a “beam sea,” bringing the crest closer to the underside. A general rule of thumb is that the clearance should be the larger of 6% of the waterline length or 20% of the span between hulls. (Currie; Sail Magazine; Van Leer). Of course, the problem with a higher bridge-deck is that you have a higher center of gravity.
The independent movement of the hulls can cause a variety of problems. Remember, each hull can move on three axes (heaving, surging, swaying) and can also rotate about three different axes (pitching, rolling, yawing). The two hulls will “feel” different parts of the wave system and will react differently. Which opposed motions will be worst will depend on whether the waves are coming from in front (head sea), the side (beam sea), behind (following sea), or somewhere in-between (quartering sea). Estimating the load that this will place on the structure requires a knowledge of the “wave spectrum” (the frequency of waves of different heights, wavelengths and directions) in the waters you are planning to sail in.
For example, for a catamaran, the worst transverse racking (one hull higher than another) occurs in a “beam sea” when the wavelength of the waves is twice the centerline separation of the hulls, since one hull is on the crest and the other in the trough (SSC 29).
Quartering seas promote pitching in opposition, causing twisting. The larger the cross-sectional area of the cross-structure, the greater is its resistance to twisting. The twisting causes shear stress. The stress is reduced by increasing the cross-sectional area of the structural lattice and the thickness of its members. The ability of a material to resist it depends on its shear strength (steel is 12–50 times better than live oak, depending on whether the shear is perpendicular or parallel to the grain).
The trimaran can achieve a given overall beam with two short spans rather than the single long one of the catamaran. If the required structural weight of the cross-structure is more than linearly proportional to the beam, that will give the trimaran the advantage. (Paine 97).
The various staggered hull configurations require more complex cross-structures, which is why I wouldn’t expect them to crop up in the first few generations of multihulls.
For modern pleasure ships, catamarans cost about twice as much as monohulls of equal length at the low (35′) and high (60′) ends of the spectrum, and perhaps 50% more for a 49-footer. However, the costs per cubic foot capacity are about the same. (BYA). The world’s largest sailing catamaran (44m length overall) cost about 35% more than the most comparable monohull, but there was a similar difference in displacement. (Lawless). Estimated hull structural costs for a 700-ton naval vessel had the catamaran 50%, and the trimaran 100%, more expensive than the monohull. This was based on assumption of costs per pound which were 20% more for the catamaran and 50% more for the trimaran, presumably related to the strength requirements. (Nguyen).
It may seem as though you can save money by converting an existing sailing ship to a trimaran by attaching outriggers. While that will help with stability and therefore the ability to carry sail under strong crosswinds, the converted hull would have a length/width ratio that is considerably smaller (usually 3–4:1) than that typical of the main hull of a trimaran designed as such. That in turn means that it will have more wavemaking resistance at high speed.
A better candidate for conversion would be a hybrid sail-oar galleass, as those had length/ beam ratios of perhaps 6:1.
When the market demands high-speed—for example, for delivery of passengers or time-sensitive cargo—a powered multihull is more fuel-efficient than its monohull equivalent. For bulk cargo delivery, the monohull is likely to remain dominant (White xi). Fuel bills are lowered by operating at low speeds, and at that speeds, the multihull suffers a frictional resistance penalty and hence consumes more fuel than the monohull. While fuel is not relevant for sailing ships, a heavily laden multihull will not be as fast as its monohull counterpart since it’s unlikely to capture enough wind to travel at the speeds at which it’s favored.
Offense. One advantage of the multihull for combat is its stability. This has several consequences. First, there is no need to time a broadside to match the extreme of a roll, in order to maximize accuracy. Secondly, the multihull is less likely to have the problem, when sailing in a crosswind, that as a result of heeling (being tilted by wind action) its leeward guns are pointing too low or that its windward guns are pointing too high. It can also carry its guns higher, where the ports are less likely to be closed because of a high sea state. And rolling interferes with the reloading of the guns.
Minimizing rolling also has defensive advantages; the ship doesn’t expose the area below the normal waterline to an opponent on its windward side.
For a sailing multihull, the cannon probably would be mounted in the hulls. However, these would each only need to be wide enough to accommodate the working of a single broadside if the rigging were on the cross-structure between the hulls.
Another possible advantage of the multihull is that the breadth of the cross-structure would permit the warship to have greater fore-and-aft armament, an advantage in both pursuit and flight.
Taking that a step further, the multihull can enjoy a very large deck area. That means that it can hold a lot of mortars for use in bombarding a fort or city. (Curiously, deck area is becoming more important in 21st-century warships, but to accommodate launch cells for cruise missiles.)
In 1682, the French Navy introduced the first bomb vessel, a galiot-type vessel with a ketch rig with forward-pointing mortars (Goodwin, the bomb vessel Granado 1742, p. 7). Mortars, please note, are artillery which fire explosive shells in a high arc, thereby circumventing defensive walls. In 1726, the British placed the mortars on turntables (primitive turrets). Ideally, the multihull would be self-powered, so that the rigging didn’t restrict the firing arcs.
The main constraint on putting cannon, especially mortars, on the cross-structure is that the deck must be strong enough to support the weight of the weapons (4 tons for a 13″ mortar) and more importantly withstand the shock of their recoil (the mortar’s powder charge was 12–15 pounds and propelled a 10 pound shell) (Goodwin 16). This is probably the greatest impediment to the use of multihulls for shore bombardment. Siege mortar-bearing “bomb vessels” had to be extremely heavily built in order to survive their own fire; the mortars were supported on multiple “cross-hatched” layers of beams, with coiled rope as a shock absorber under all. (Ware 10).
For it to be practical to use the cross-structure as a platform for mortars, it would have to be a multiple deck structure, with the crossbeams extending deep into the flanking hulls, in order to spread out the load more effectively. Or better yet, build it in steel. You can’t expect to take two conventionally constructed hulls, build a single connecting wooden deck, and plop a siege mortar on top of that. Unless you want to make a big hole in the deck after you fire it a few times . . . .
However, you could put the mortars in the hulls, and still put fairly heavy cannon on a closed deck cross-structure. Even though they were lightly built ships, a Mediterranean war galley’s centerline armament could be as nasty as a 40–55 pound full cannon on a sliding, recoil-absorbing mount, flanked by two pairs of lighter weapons. (Guilmartin 323).
Defense. In wartime, we have to worry about vulnerability to enemy action. A pole connection could be severed by a single cannonball, but it is a rather small target, so hitting it, save at close range, would be a matter of chance. Closed decks are easier to hit, harder to destroy. Still, the cross-structure is likely to be a relatively light structure, vulnerable to raking fire.
Another hazard is that an enemy could try to row or sail a small boat, filled with explosives or incendiaries, into the “tunnel” under the cross-structure, and blow it up. Vigilance, bow and stern armament, and perhaps some sort of netting, are the best defense.
If a sailing catamaran had its masts on the hulls, it might be possible to operate the balloon from the crossdeck. Otherwise, to avoid fouling of the balloon tether by masts and rigging, the multihull would have to be self-powered.
Dirigibles have such a huge cruising range that they can usually make do with land bases. However, multihull dirigible tenders may come in handy as forward bases when we are using military dirigibles in operations against distant enemy territory. They would need a well-anchored mast to which the dirigible could be moored, and they would also have to carry enough fuel to get the dirigible to the next base.
A multihull can certainly provide a good landing area for a helicopter (the Pigeon and Ortolan had helipads, as did the 1972 SSP Kailamino); what I am not sure about is whether we can build helicopters any time soon.
Lastly, there is the possibility of supporting airplane operations with a multihull. The simplest case would be one in which the multihull merely carries fuel for a seaplane that takes off from and lands on the water, but then there is no advantage to the multihull design. The next step up would be to catapult the seaplane off the multihull’s deck. Finally, we could provide a landing strip for an airplane, but that would require a very long multihull and carrier-type arresting gear. In 1926 Italian naval designer Guidoni proposed a 3500 ton catamaran aircraft carrier! (Preston 67).
Ship motion is a major cause of carrier landing accidents and hence there is reason to hope that a catamaran, especially of the SWATH type, would provide a more stable air support platform. (Stevens 31).
It makes a significant difference whether the ships you are designing are wind- or self-powered, and whether they are built out of wood, steel, or high tech materials (fiberglass, carbon fiber composite, etc.).
A multihull can safely carry more sail in strong winds, because of its enhanced stability. The conventional warships which are mostly likely to benefit from a multihull design are the shorter ships, because the wind heeling “moment” would be proportional to the fourth power of the length, and the resisting moment to the third power. Also, the shorter warships were most likely to be given a relatively high (>=4:1) length/beam ratio, both to increase their speed and to maximize the broadside armament they could carry. You could replace any of these with two demihulls of half the width, and separate them by a “tunnel” of one or two hull widths. At the low end—two ten foot demihulls separated by a ten foot tunnel—this is comparable to the nineteenth-century catamaran river steamers, and so it should be possible to construct an adequate cross-structure out of wood. After that, we’ll see . . .
Still, my prediction would be that sailing multihulls are likely to be limited to light-displacement dispatch vessels and landing craft. While the heightened stability of multihulls allows them to carry more sail than a monohull of equal displacement, they can’t count on getting strong winds, and in light winds they will be slower.
Steam power is going to make multihulls much more attractive, as they can be driven to the speeds at which reduced wavemaking resistance is more important than increased frictional resistance. While they’re too displacement sensitive to be suitable for cargos like iron ore, they should work fine for carrying passengers, low density “fast” freight and naval artillery.
Steel and other new shipbuilding materials with high strength/weight ratios will allow ships to be longer, have wider crossdecks, and carry more payload. Bear in mind that in wooden ships, the hull accounts for about half the displacement, whereas in steel ships, it’s as low as one-fifth. (Reed 72; Arnott 12).
Jonathan Swift made fun of Petty’s late seventeenth century catamarans, writing of “our crazy double-bottomed Realm.” With steam and steel, multihulls may not be so crazy after all.
Author’s Note: The “Multihull Addendum” posted to http://www.1632.org/gazetteextras/ provides a bibliography, tables of ship data for selected multihulls, and some stability calculations.