Multihull Structure Thoughts

A word about designing for impact (shock loads). I used to be involved in designing conveyors for sawmills and we always used 2 x the mass of the moving object when calculating for shock loads. for example a log stop: to provide a fixed length for a saw to cut a log. The log would be driven on to the stop (maybe a retractable arm say). We would design the stop to with stand 2 x the mass of the heaviest log moving at the speed of the conveyor. We never had any failures so that must have been sufficient....
same principle can be applied to a boat.
Tb

 

The Le Breton Yachts SIG45 is an interesting catamaran. It is 45 x 27.5 foot weighs 12000 lbs and in the initial design had a max payload of 3000 lbs. The rig carries a 75 foot carbon wing mast has 92 square foot of area with a 1000 square foot mainsail, 460 square foot self tacking jib and a 1500 square foot gennaker. The hulls length to beam is 13:1. The underwing clearance is 3.8 foot. This is not a floating caravan. The SIG45 can sail faster than wind speed having sailed at 22 knots in 16 knots of wind in tests and the maximum speed seen is 28 knots. The hulls show a wave-piercing type of bow, and there is zero overhang in either the bow or the stern. In plan form the hulls are slightly asymmetrical flaring out inboard amidships to gain some internal volume.

This cruising catamaran (yes, it is a cruiser) comes from the LeBreton yard in Amsterdam. The naval architecture was done by Van Peteghem Lauriot Prevost in France. Unlike most cruising catamarans, this cat does not use the bridgedeck area for accommodations. The flared-out sections for maximum hull beam provide enough accommodation to fit a practical galley, seating double berth cabins and loo etc. The accommodation is comfortable and works.

The daggerboards are carbon and outboard in each hull, and give a board-down draft of 9 feet and a board up draft of 3 feet 10 inches due to rudder depth. The rudders are also carbon fiber. On deck the mainsheet traveller spans the connecting beam at the stern. The boat is steered with surprisingly short tillers. Forward is a soft trampoline but aft the connecting deck is rigid and appears to have a teak planked veneer over at least part of it with the rest being nonskid paint. Halyard winches are clustered at the mast base and I suspect the self-tacking jib sheet is controlled from this bank of winches also. There are three other sheet winches port and starboard aft at the steering stations. These winches also control the broad traveller. The sheet loads of the mainsail can top 9000 lbs.

There SIG45 was originally designed in 2002 and actually finished in 2009. The first boat was to be built with a foam e glass expoxy sandwich with carbon used in cross beams and other high load points. The cross beams are a D section with the aft full width bulkhead has top and bottom unidirectional carbon flanges with the forward sections acting as an aerodynamic shape that handles torque loads. The designers had analysed the structure and concluded that the weight savings of all carbon fibre would not be sufficient to justify the additional cost. Especially when carbon fibre hull skins would be to thin to handle knock resistance etc. The second SIG45 was built in the US for an American client who wanted an all carbon foam epoxy cat. Result was a slightly lighter boat but I do not know what later builds were done in. The original SIG45’s cost in excess of $1 mill. The jpegs give the idea. PS you can charter a SIG45 in Greece. Those who have sailed on it say its fast, fun and really annoys Lagoon owners who take twice as long to get somewhere.

 

The following paper is some research on reinforcing timber beams with a carbon fibre pultrusion to find out how much extra force the beam could handle and what effect it had on stiffness. The results were perplexing but is a big warning to all that applying carbon fibre to a timber beam does not always solve a problem.

The timber beams were from 140 mm to 220 mm high in 20 mm increments by 100 mm wide. The beams were either 9.75 or 13.1 foot long. The timber was carefully selected to minimise any knots or fractures etc. A lamella (pultrusion) of carbon fibre of 50 x 1 mm or 50 x 2 mm was glued to the bottom of the 100 mm wide beam. Strain gauges were placed on the timber beside the pultrusion’s to understand the effects of load throughout the loading process.

The beams were then subject to loads on a test rig to the point of failure. Now we will give you the conclusion first.

“According to the charts and the tables it is clear that increase of the strength influenced by the reinforcing is significant in the case of cross sections of 120x100, 180x100 and 200x100 whereas in the case of cross sections of 140x100, 160x100 and 220x100 slight increase or even no increase occurred. This means that there is no unequivocal trend of increase of the strength depending on the dimensions of a cross section which is caused by conditions of experimental verification (e. g. small number of tests in connection with a variability of material characteristics of timber) and also by random phenomena (e. g. defects within material).”

The results are as follows: 140 x 100 mm no difference between CF reinforced and unreinforced timber under load: 160 x 100 a 6% increase in load capability of CF reinforced versus unreinforced timber: 180 x 100 a 23% increase in load capability of CF reinforced versus unreinforced timber: 200 x 100 a 23% increase in load capability of CF reinforced versus unreinforced timber: 220 x 100 a 5% increase in load capability of CF reinforced versus unreinforced timber. Please remember these results are based on a short series of tests and are averages. There were several failures due to timber failures eg knots or unseen fractures in the timber.

The other real issue was failure of the glue lines between the timber and the pultrusion. “They calculated the theoretical value of ratio of modulus of elasticity of timber to modulus of elasticity of carbon n. In the case of timber the modulus of 10·10 to the 9 [Pa] was considered and in the case of carbon the modulus of 155·10 to the 9 [Pa] was considered.

This difference causes the shear stress between layers and that is why these values are important. The shear stress acts straight to the layer of the glue which needs to stand this loading. During the experiments we observed breaking the glue layer very often. This fact probably occurred because of wrong preparing and manufacturing of the reinforcement. It is necessary to say that perhaps the problem was caused by smooth surface of the lamellas. It could be recommended to the manufacturer to try to change it.”

Summary. Imagine you have a Wharram with solid timber cross beams or a tri with solid timber cross beams. You say, I will apply carbon fibre to the beam to strengthen it. This paper basically says depending on the timber cross section, the timbers condition including moisture content, the carbon fibre you use and the epoxy glue and surface preparation the beam may be strengthened. Later I will talk about another paper that shows how you place your CF on timber can effect timbers strength.

 

Interesting paper. I am not sure that gluing a pre-prepared section of carbon is a good idea. The shear strength required is very high, as evidenced by the glue breaking. Which makes me think - how is this useful to us when we usually laminate uni glass with wide strips of carbon. But the strange results would seem to suggest that gluing carbon strips on is a low value method of re-inforcing.

What is also strange is that the beams failed in tension. Most crossbeams made from wood have larger compression members than tension members. It could be that the team did not use scarf joints and the laminations failed in tension where the different lengths joined. So it looks like an interesting paper on how NOT to build beams for boats.

cheers

Phil

 

The Pacifico PS 153 is a catamaran built by Pacifico Yachts, a Russian company. The cat is a cruiser that is 50 x 24.6 foot weighing 33,000 lbs and carrying a 60 foot mast and about 1200 square foot of sail. The cat is designed by Albatross Marine. This boat is about accommodation with reasonable performance.

It has 900 square foot of accommodation space with 6.5 foot ceilings through out the boat. As standard it is available with 3 (owner) or 4 cabins (charter) and can be further customized to special requirements. The main deck saloon has a functional galley, a spacious dining area, a chart table (second control post).

Built to the highest standards with extensive use of high modulus fabrics, carbon fibre and epoxy resin (for higher longevity and no risk of osmosis) the boat has full ISO Certification in Cat A. It is a catamaran designed and built to sail across oceans in any weather conditions, in comfort and safety.

Why the interest? The price is $453,000 US ex factory. Relatively cheap for a boat this size and type. 5 to 10 year old Lagoon 50’s sell for more money. A new Lagoon 50 will cost about double. Unfortunately Pacifico does not specify what is included in the standard price and I suspect it will add a bit to the base. If the boats structure is built as specified and the pricing is real this cat could be a good cruising cat. The designers are experienced and have produced good boats previously.

 

For information only. Carbon Fiber Structural Beams are becoming interesting. Many load-bearing structures could use customized carbon fiber structural beams. Element 6 Composites specializes in designing and fabricating custom carbon fiber structural beams for a wide array of industries and users. Although these beams are typically more expensive than a similar metal component, the substantial weight savings that can be achieved through advanced composites often outweighs this upfront investment. EG The first jpegs are of a beam weighing in at only 9 pounds, this 8ft curved beam held 2500 pounds without breaking.

The second jpegs are of a 24-foot long arched box beam was designed for an application that required a lightweight structure to support a heavy distributed load. Several rounds of analytical and finite element analyses drove optimization. A combination of selective carbon fiber placement and hole pattern provided a minimum weight design that met all strength specifications. The web site is at: Carbon Fiber Building Material | Element 6 Composites https://element6composites.com/carbon-fiber-applications/carbon-fiber-trusses-beams/

 

There is a competition in the US called the SAMPE student bridge competition. The idea is to build a 2 foot long structural I beam of maximum cross section dimensions of 102 x 102 mm with a 15 mm max width vertical I beam section that weighs under 600 grams (1.3 lbs). The winner of the contest is the I beam that can support the most weight in the centre when the 2 ends are supported. This group students managed to support a 3000 lbs load force in their best design. I have seen other groups 2 foot long bridges support over 3500 lbs.

The following PDF paper is quite detailed in the bridges design, fabrication and testing of 5 iterations of the bridge design. The design stage compares a carbon fibre composite with an aluminium, a steel and a titanium version of the bridge which gives an insite into why carbon fibre composite is preferred. The analysis and build of the 5 different structural versions of the carbon fibre composite versions of the bridge gives an insite to the types of failures that can occur with differing core materials and disposition of fabrics within the bridge.

The build method(s) include a detailed description of vacuum bagging and resin infusion used to build the 5 versions of the bridge.

The testing showed the lowest performing bridge failed at 1675 lbs load and the best performing bridge failed at a maximum load of 3100 lbs. For basically the same set of materials of CF, epoxy, aluminium honeycomb or foam used in slightly different combinations this is a large range of load capability.

This PDF paper is 50 pages and 5 meg and is written by students who are teaching/ learning along the way. It is detailed but gives some good insites that are applicable to cross beam design. You really do have to understand the material capability, how the material combinations can be brought together in the best design and how to fabricate the structure to get the best results.

Again, I say, a designer needs to have a lot of experience and/or excellent technical skills to achieve a good design. But a bit of testing goes a long way. I can name 5 big name multi designers who have a broken cross beam on an ocean going design. PS a bridge made of bamboo held a 1700 lbs load in the 2011 competition.

 

The C-60 is the first multihull design by German Frers (very famous large mono designer). The C60 is 60.7 x 29.8 foot weighing 44000 lbs and carrying a 85 foot spreaderless full carbon Lorima wing mast with a 1474 square foot main and a 677 square foot self tacking jib with a 1355 square foot code zero. The rig is controlled by a number of motorised winches to ensure a crew of just two can sail the boat. Most ropes are run under deck to ensure safety and simplicity with no tangling.

The C-60 is built as a semi custom by Ocean Quality Systems, a company founded in 2010 by the owners of four companies that do carpentry, engineering and electrics for Nautor, Swan and Baltic Yachts. The companies still continue to supply parts and services to the other yards. The owners ambition was to create a Finnish yard building high quality multihulls that matched Nautor, Swan and Baltic Yachts.

The C-60 is on the edge between comfort and performance. Any more hull beam would affect performance. The bow is reversed and sharp to reduce pitching in waves, as a true wave piercer. As a hidden safety feature the bows have crash boxes protecting the hull and structure in case of a collision. Each hull has 3 watertight sections. The cat has curved carbon dagger boards. They are lifted by a push button B&G system indicating what depth they are.

The hulls and decks are vacuum infused rigid foam glass with a full carbon foam option. The deck house was designed and built in Titanium. The strength to weight and stiffness of titanium has major benefits and allows the fitting of large areas of windows of 12mm laminated and tempered glass.

Bow thrusters enable fast and efficient manoeuvring. The engine rooms are located midships to help centralise weights. Frers is applying high quality mono techniques to multi’s.

The boat accelerates well, is very stable and is fast with speeds of over 20 knots recorded. The boat is also “comfortable” with an owners wife claiming “I can actually sleep in the forward cabin” whilst heading upwind at above 12 knots. Money will not be one of your worries if you can afford this boat as the several million will buy a very high quality product that sails well. We can all dream.

 

The following PDF is about various carbon fibre epoxy I beams, glass fibre epoxy I beams and a few other beams tested for resistance to “notch” failure over a life cycle. The notch failure is the equivalent to drilling a hole in the web in one series of tests and in a second series of tests drilling a small hole on the flange of the I beam.

The test was to put an undamaged I beam through a series of load cycles (between 1 and 8 million cycles) to test its resistance to failure. Then the same beam (a replica) with a hole in some part was tested again through a load cycle.

The general results are in the conclusion: “The present paper is the third in a series which is concerned with the static and cyclic fatigue failure of composite I-beams. Both carbon-fibre/epoxy-matrix and glass-fibre/epoxy-matrix composites have been used to manufacture the I-beams, which had a multidirectional stacking sequence consisting of a balanced layup of 0, +45 and -45 plies. Both unnotched and notched I-beams have been studied. A four-point flexural configuration has been used to test the I-beams and in no cases were the beams fatigued above the loads at which buckling of the compression flanges occurred during static testing. The carbon/epoxy and glass/epoxy I-beams which were unnotched did not exhibit any detectable damage within 1.2x10 power 6 and 8x10 power 6 fatigue cycles, respectively. At these numbers of cycles the fatigue tests were halted. The excellent fatigue behaviour of these unnotched composite I-beams is most noteworthy, especially since the maximum fatigue loads which were applied represented typically about 75 to 100% of the loads needed to cause buckling of the compression flange during static testing. Hence, the I-beams were notched in order to induce failure under the fatigue loads. The most damaging type of notch that was introduced was a 60mm diameter hole in the web section of the I-beam (see Fig. 2). The notched I-beams now did indeed fail under the fatigue loads. For example, in the case of the carbon-fibre/ epoxy-matrix I-beams which were fatigue loaded at 5Hz from 5kN to 50kN (which represented about 9 to 90% of the load at which buckling of the compression flanges occurred during static testing) failed after 4.78x10 power 6 cycles. Fatigue failure was due to various types of damage, including delamination, matrix microcracking and fibre fracture, occurring around the 60mm diameter web notch. The damage mechanisms in the notched carbon-fibre/ epoxy-matrix I-beams were studied in detail. The most severe of such damage was caused by the tensile stresses which were present around the web notch. The principal mode of damage was matrix cracking, in plies oriented at 90to the local direct tensile stress. A significant proportion of this damage occurred within the first 0.5x10 power 6 cycles. The matrix cracking led eventually to delamination and fibre fracture, the latter being the final cause of structural failure of the I-beam.” The PDF gives the details, it is a bit of a read but tells a lot about not only notch failure but gives a few clues on material choices and fabrication techniques.

Translation. The undamaged beams can handle the test load cycles very well. When a hole is drilled in either the web or the flange it may appear to initially handle the test load but after a much reduced number of load cycles the beam structure started to fail mainly with crack propagation around the holes but in some cases delamination of EG a flange. In short don’t drill holes in any part of a beam unless the beam is designed to have those holes. Second the beam may appear to handle a load after damage but its life may be considerable shorten due to load cycle eventually propagating the damage.

PS Further investigation of SAMPE bridge building competition found the 2018 winners from Chengdu university built a 500 gram (1.1 lbs) bridge that supported 11,060 lbs in the centre. This is 3 times more than the 2011 competitors I reported on above. For the same amount of materials utilized in a different design structure you can achieve a 3 times better result says a lot about what can done with good design and build.

 

The Cross 46 is a popular cruising trimaran from the late 60’s. The tri is 46 x 25.25 foot weighing 19,000 lbs and displacing 27,000 lbs. The ketch rig carries a 165 ft mizzen, 320 ft 14 oz main, 220 ft 12 oz self tacking jib, 400 ft genoa and a 1070 ft 1.5 oz spinnaker. It has aluminium masts with 5/16 inch stainless steel rigging with 3/8 inch SS caps.

This boat is really about accommodation. I walked over a Cross 46 in Cairns owned by 2 South African couples. They had no problems living on the same boat after 2 years. Literally one couple had the starboard side of the boat and the other had the port side of the boat. They treated it like 2 apartments. They only shared the main saloon. Very few boats I know could achieve this level of privacy. The jpegs give an idea of the level of comfort.


The trimarans were often home built from cold moulded Marine-ply / Epoxy sheeted and timber. These are large boats that take a lot of time to build but were designed for home building. There specifications were often slightly oversized to allow for the material quality variations etc. EG the hulls were double diagonal 6 mm ply. The hull curves have a big radius that helps minimise build problems etc. Several of these boats had the original mini-keel running about half the length of the main hull removed and daggerboards were installed in each float. The installation of the boards has allowed a Cross 46 to point up another five degrees and increased her windward ability immensely. The Cross 46 when cruising could average 135 miles/day on ocean passages. There have been reports of peaks of 15 knots in really good conditions (like surfing down seas in strong winds).

The modern interpretation of these boats are either eg a NEEL 50 or a Lagoon 50 both of which could average higher speeds across an ocean but unless you have electric winches you will be working a lot harder to achieve that performance. A Cross 46 suited its time and if you can find a well maintained one would be a good buy, but remember you are buying a very big boat which will require continuous maintenance.

 

This is a story of what can happen to a reasonable design when business is the driver. Carino started life as a 40 x 23.6 foot cat designed by John Morgan. It weighs 8800 lbs and was designed to be a charter boat in NZ for day trips in the Bay of Islands and could carry 40 passengers which means its displacement was about 17000 lbs. The mast was about 200 x 150 mm carrying about 700 square foot fractional sloop rig could carry full sail in 30 knots of wind to meet charter regulations. The cat was basically two 6 foot wide hulls at gunnel level and an open bridge deck with seating for passengers. When I sailed on the boat it performed quite well doing about 7 knots upwind in 10 knots of breeze with a max of 12 knots reaching in 15 knots of wind. The day I sailed it only had 12 people on board.

The original construction was ply timber with 9 mm Gaboon ply hulls and hull decks with stringers at about 1 foot centre lines and ring frames every 3.5 foot. The bridge deck is 12 mm ply on top, 18 mm honeycomb core and 6 mm ply on the water side (remember this has to take up to 7000 lbs of passengers). The crossbeams are 2 layers of 12 mm ply either side of a timber top and bottom flange and uprights. The fore beam is an aluminium section. The bridge deck had a lightly framed canvas spay hood for guest comfort. A nice boat that could be turned into a minimalist cruiser for fun.

But the charter business is not so kind. First Carino magically it is extended 10 foot to 50 foot long (cut here and insert and additional piece in each hull and bridgedeck). Next add a substantial cabin to accommodate those who don’t like wind or water, add some grunty desil engines to replace your two 10 horse power outboards and now you have a better money earner that motors as much as it sails. The original Carino spent most of its life sailing. Not that the current version is bad, its just the original 40 foot version was a very nice boat. Money often wins over nice. The first jpegs are of the old boat the last jpeg is of the 50 foot version.

 

Excalibur 1 is about one thing, speed. Dave Knaggs is a man who does not like being beaten, so in the early 90’s he designed and built a 31.66 x 20.1 foot open bridge deck racer “cruiser” catamaran who hulls, beams and pods weigh 1760 lbs. The cat total displacement is 2670 lbs. Working sail area is 600 square foot while it can set 1270 square foot reaching on a rotating dual spreader aluminium mast. The mast started as a 150 mm round tube but was rolled down to 180 x 115 mm oval section. The big jib or gennaker are used up to 10 knots of true wind, the small jib is used up to 15 knots of true wind and then the small jib is reefed for winds above 15 knots. The fully battened mainsail is controlled by a hydraulically controlled mainsheet system. A 4.25 foot ram can be pumped up to tighten up or release the mainsail. The pressure on the hydraulic ram is about 1800 lbs in “normal” sailing. Each cockpit has a control unit in it. Combined with a full width traveller system the main sail (which is the power house of this boat) the boat is very controllable. How controllable? This boat can be controlled and is often sailed single handed whilst sailing at 27 knots in a 20 knot sea breeze. At 20 knot boat speed Excalibur is just cruising. Hard on the wind the boat will do 15 knots. This boat is so fast, a broad reach ends up being a close reach as the apparent wind is chased forward.

Excalibur 1 is home built will 9 mm Western Red Cedar with 280 gsm unidirectional e glass on either side in epoxy. The glass layer is doubled below the waterline. 3 foam ring frames (one under each cross beam) is also in the hulls with bunks and shelves acting as stiffeners. The hulls are faired with microballons and painted. The cross beams are round tubes with inserts at gunnels and under the mast. The rear beam is 150 mm, the main beam is 140 mm with a dolphin striker and the forebeam is 125 mm diameter. All are 6351 T6 aluminium.

This is a seriously fast boat that was well engineered (dave did his own work), designed and built that did not cost a lot of money. But as per usual Dave Knaggs decided he needed more speed and later developed a 35 foot foiler. Well done dave, a brilliant boat.

 

The following 19 page PDF describes the design development of a sailing hydrofoil in about 1983. The reason for showing this paper is shows many aspects of the thought and analysis process required for each aspect of the design. It would help anyone who would like to develop or add foils for a design. FORCE 8 is 20 x 16 foot with an all up weight of 500 lbs including crew. The sail area 140 square foot. The rig is interesting in its development although I think there are better options.

FORCE 8 also has a 3 point submerged foil system with height and roll control. The detail of these systems is good. The boats performance is equal to Mayfly, a cat that was considered very fast and the record holder at the time in its class in speed week. There are only 2 jpegs I could find but the 3.5 meg PDF has many photo’s and drawings.

 

A quick story of an evolving cruising catamaran. Kurt Hughes 45 bridge deck cat is one of his more popular designs. The revised 45 is 45 x 25 foot that weighs 12,000 and displaces18,300 lbs. The aluminium or the composite mast carries a 495 square foot main, 232 square foot blade, 368 square foot genoa and a 700 square foot spinnaker. So far there is not much changed from previous versions. The previous versions had a cylinder moulded ply or fiberglass composites hulls with ply or composite beams decks etc. The mast can be aluminium or carbon etc. Request the build option you like and Kurt Hughes will probably be able to provide it. As Kurt says he has chosen to allow the builder to make as many of their own parts as much as possible. EG The bow tube is composite, as is the compression tube, chainplates, mast and boom etc.

But the real difference is in the hull shape. The previous versions of the 45 just had a simple U hull shape without steps or flares and an old “normal” bow. Now the hulls have reverse bows to dampen pitch in waves, and a spray deflector that slides back into hull flare. The hull flare is the best way to get good room inside the hulls and still have a skinny waterplane so it can be fast. What appears to be a small change requires quite a bit of design rework of bulkheads etc especially when you are also providing build your own options of masts chain plates etc.

Basically the 45 has about the same internal layout options, but the boat has a suncover over the cockpit and transom steps for easy water access added.

I have only spoken to one 45 owner of the older non flared hull model and he was very happy with his boat saying it was a great cruiser with reasonable speed and a simple rig. He could get 8 knot averages which is all he wanted. Top speed was irrelevant to him.

The first 3 jpegs are of the older model but the layout and rig are similar. The following jpegs are of the new shape done by a guy who paid for the design and did an approximate 3D version of the hulls. The final jpeg is his plan for the foam core, orange is 100 kgs/meter and blue is 80 kgs/meter.

 

There would be no logical reason for selecting this alloy. It is rare, no doubt not easy to find = expensive.

One wonders why this alloy was selected in the first place. Since all modern alloys, such as 6082 is/are more than equal to this alloy and is readily available.
The only "possible" reason, is that many suppliers of this alloy have varying amounts of copper, but many indicate the amount of copper is much less than the, at the time go to alloy of 6061.
Thus, 6082 falls into this category, much less copper content for improved corrosion resistance.