Multihull Structure Thoughts

Cirrostratus tris is interesting because its interior room is is quite surprising for such a small tri. The accommodation is also liveable mainly due to the 2 separate cockpits either side of the saloon seating on the wingdecks. This tri is 32.8 x 26.3 foot weighing 3100 lbs and displacing 5500 lbs carrying 590 square foot in the main and foretriangle. The tri sails well having done many sea miles. Many have been built some in ply timber others with foam glass hulls and ply decks. The ply versions are mainly built with 6 or 7 mm ply on 36 x 19 mm stringers at 150 mm centrelines. Frames are 75 x 19 mm backed with 6 mm ply in many instances. The crossbeam structures are fully integrated with the wingdeck structure which provide excellent rigidity for the overall boat. This is a well thought out design, structure and accommodation layout in a tri that sails really well. Only one boat has had any real problem when a float fell off. The boat and float were recovered, repaired and is still sailing. No clue what the problem was. 2 people could really live in this boat and it would be a more comfortable cruiser than a Buccaneer 33 due to the extra internal space.

 

A follow up on the Seaclipper 28 with a more detailed description and some build photo’s. Another item will be coming in an hour or so.

 

This is the first of 2 items about one of my favorite sailing boats. Juniper is a 52 x 30 trimaran that was designed to weigh 13500 lbs and carry in its original free standing ketch rig 975 square foot. Chris White the designer commented that “Juniper is long in order to provide comfort sailing in the ocean, wide for stability in extreme conditions, but simple in construction, accommodation and systems. She’s an excellent performer, and has a very comfortable motion in waves”.

The concept of the Juniper 2 is a new design incorporating several refinements and improvements from her predecessor. She is wider overall, has a longer and finer bow, and uses a balanced spade rudder instead an outboard hung rudder. Aesthetic details such as the crossbeams and their fairings and the sheer line were refined. The hull underwater shape has less wetted surface than the original and the fin and rudder sections were refined for less drag. The interior layout was refined to increase elbow room and now incorporates a dedicated shower stall. Incorporating a ketch rig that is an absolute joy to handle, tacking is a matter of just turning the wheel. The crossbeams are very high above the water and there are no cabin projections beyond the hull sides. These features eliminate pounding in rough seas, contribute to her comfort underway, and was designed and built for safety and survivability offshore. Her interior and exterior finish is excellent. All systems and equipment were installed and have been carefully maintained and upgraded with an eye for simplicity, reliability and serviceability.

The original Juniper was built using constant camber approach which is a thick hull panel with minimal hull framing. How thick, 25 mm of moulded cedar (more details in next item). This produces strong long lived hulls. Juniper was built in 78/79 and circumnavigated the globe 6 times without any major or minor repair. Its rig was replaced to improve its performance not because of any issues. Again Chris White comments "Two different rigs were tried on Juniper 1. An unstayed cat-ketch rig and a more powerful stayed ketch rig with rotating spars. Each has its advantages, but in a nutshell, the stayed rigs can be more powerful although they are more expensive and require more attention to sail".

When Jim Brown and Mark Hassall were invited on Juniper's maiden sail, here is what Mark wrote about his experience with the flexible, unstayed rig: "Though we spent the better part of the day trying to find wind. Late in the afternoon, the wind found us. The free-standing masts suddenly bent at an alarming angle and then twanged upright again. Juniper shot forward at such an incredible speed that we all grabbed at whatever was handy to keep from falling overboard! "17 knots!" yelled Chris. "My God, she went from 0 to 17 in a couple of seconds! Have I got a boat or have I got a boat?" ..........."It's no boat," I yelled. "It's a goddamned slingshot!"

Juniper is a sensible fast boat that is relatively easy to sail but it is a very big boat with limited accommodation. It would be a good live aboard for 2 people. Chris White owned Juniper for 17 years, lived aboard for years and sailed it far and wide.

 

Carisa is the second version of Juniper known as Juniper 2. It is 54.5 x 31.5 weighing 14000 lbs and carrying 900 square foot of sail. As one person said ”Carisa is HUGE on the outside & small on the inside. They never got it all painted at once, it was so large. They would be back in the yard doing more painting, every other year. Also, the free standing mast were Chris White's originals. They required sock sails that were VERY hard to raise, and the mast required reefing early due to bend. (a spinnaker was risky too for this reason). Carisa easily sails 9-10 knots with wind speeds of 15-16 knots. With decent sail conditions she will easily make 200+ mile days. In a flotilla she will out sail every boat except performance catamarans and 65'+ racing mono-hulls.”

Hulls and decks are vacuum bagged triple diagonal Alaska Yellow Cedar 6 mm, 12 mm, 6 mm constant camber using WEST System® epoxy resin with glass/epoxy exterior skin. Crossbeams are epoxy laminated Douglas Fir and marine plywood. Freestanding ketch rig provides effortless tacking and superb offshore reliability. Tapered hollow wood/carbon fiber epoxy masts 55 foot high of deck, 310 mm diameter at deck tapering to 105 mm at the mast head. Mainsail and mizzen (450 sq ft) by Haarstick Sails, 2008, Dacron cloth. 3 x reef points, fiberglass battens, rugged construction details, sleeve luff. Mizzen staysail (500 sq ft).

More details of the original build of Juniper is in the attached jpegs. The most important aspects of this sort of building is it does not require quality timber just light timber as the vacuum bagging and epoxy will fill any voids in the moulded timber layup. Second it leaves a clean interior without any stringers and minimal frames. Thirdly once set up the process of building a large boat hull structure is relatively fast. But the rest of the fit out electric’s etc will take the same time. The photo’s will give an idea of how clean the interior is. Finally this boat is light for its size, the reason it is fast with such a small sail area.

 

A proa is a controversial boat not helped by the difference of opinion between Atlantic and Pacific proa types. Rob Denny has heavily promoted his version of Atlantic proa’s, the Harryproa. They are fast, light and effective. The major voice for modern Pacific Proa’s is Russell Brown son of Jim Brown. Russell built 4 Pacific proa’s. His first was a 30 foot $1000 (yes $1000 for the complete boat) Pacific proa called Jzero. According to Russell it was a 1000 lbs displacement pile of junk built from 6 mm ply, stringers and frames lightly filleted and taped together at chines and joints in pods etc. The crossbeams are 5 layers of 19 mm x 90 mm fir tapering at the float end. It has a second hand rig etc. Not much room but a fast proof of concept. This “pile of junk” is now 30 plus years old with improved cross beams and as still sailing a few years ago.

Next came Jzerro a 36 foot ocean crossing (USA to Australia and return some single handed) Pacific proa. Russell Brown comments on the 3250 lbs weight Jzerro. “One major attraction for me of the Pacific proa is that the concept is, in my mind, as close to structurally perfect as a multihull sailboat can get. In flat water, a proa can be sailed as hard as possible and the structure stays very lightly loaded. Yes, there’s a bit of mast compression (but not much, because of the wide staying base), and a small amount of bending load on the beams, but most of the sailing loads go to the shroud lifting the outrigger. Adding water ballast in the outrigger doesn’t load up the structure very much, and because it is so far to windward, it doesn’t take very much ballast to keep sailing flat.” Russell rarely used the water ballast option.

Jzerro was built in Port Townsend, WA and launched in 1994. This boat has many similarities with my previous proas, but was lightly and carefully built. In the fall of the year it was launched, Jzerro sailed a sustained 22 knots on the way to a race.

Jzerro will reach at 17 or 18 knots in 12 knots of wind. On an Australia to New Zealand trip Jzerro did 1050 miles in 6 days with 750 miles of it to windward. When going downwind in lots of wind, the main is down and just a jib or genoa up. This makes the steering easy (autopilot) and plenty of power. I averaged better than 10 knots this way from the Cook islands to Fiji with just a working jib.

Jzerro is now available for charter in New Orleans and its current owner on May 8, 2018 raced in a 555 mile effort along the coast of the Gulf of Mexico. Starting in Pensacola, Florida to Isla Mujeres, Mexico, which is just off the coast of Cancun. After being neck in neck throughout the race, he beat a Corsair C31R Trimaran winning line honors and his class (MultiHull). He was the first to finish with a finishing time of 71:43:08.

The only real weakness for the average sailor is its rig. The rig is powerful BUT if the boat is caught aback due to its minimal staying on one side of the boat the rig can be blown of the boat. This is a problem with most wire rigged Pacific proa’s. This is the reason why Harry proa’s have free standing rigs. So why didn’t Russell use a freestanding rig, his words “Yes, a cat ketch rig with free-standing masts has appeal. A rig like this would be easier to tack (especially with wishbone booms), and less vulnerable to being caught aback, but the compromise in performance, especially upwind, is not attractive. Jzero raced against Azulao, a 42’ Atlantic racing proa, and was about 18 hours ahead at the finish. Azulao was faster reaching in breeze, but didn’t come close upwind.”

After Jzerro Russell built a 37 foot proa Kauri and a 38 foot proa called Cimba for friends. Russell is not interested in doing any designs of proa’s and leaves that to others. We will discuss their construction in the next item. Jpegs give idea’s of the boats.

 

Jzero the 30 x 19 foot 400 lbs weight, displacing 1000 lbs, carrying 175 square foot of sail proa was Russell Browns first “modern Pacific proa” built in 1978. This boat survived three 100 knot hurricanes during Russell’s ownership and at one point they used a pair of pants weighed down to slow the boat in heavy seas. Not bad for a junk boat. There is a Jpeg of its “interior” enough for a 17 year old when Russell built the boat and a jpeg of the boat in 2000.

Next came Kauri a 37 x 21 foot boat built in about 1985, weighed 3600 lbs. Kauri’s hulls were 9.5 mm sheet plywood tortured into shape over stringers and frames. The internal room of this boat was a significant improvement over the 30 foot Jzero.

A friend Lew McGregor asked Russell to design and build a proa for him. Russell in about 1989 designed and built Cimba. Cimba is 38 x 26.25 foot weighing 4350 lbs carrying 600 square foot in a main and genoa. The main hull length to beam is 13.5 to 1 and is symmetrical. The hull halves were built of 4 layers of 3 mm WRC covered with 200 gsm 90/0 cloth in epoxy. The hull halves were slightly tortured into shape. The keel line is carbon powder/epoxy fillet glassed over inside and out. Gunnels are placed in the hulls. There are no stringers or intermediate frames, only bulkheads at crossarm points and fore and aft crash bulkheads. Cross arms are timber plywood box structures. According to Russell if Kauri and Cimba have very similar performance. Cimba was sold and renamed Pacific Bee. The new owner, who is an exotic racing monohull builder, was planning to design and build a 53 foot version from 5 layers of 3 mm WRC.

Russell then in about 1994 decided he would build what was his final version of a Pacific proa Jzerro. Jzerro is 36 x 21 foot weighing 3250 lbs. The main hull is slightly asymmetrical and more rounded than Cimba’s requiring a wooden keel structure. It carries a 36 foot rotating aluminium wing section mast 95 mm x 190 mm (another report says 200 mm), and weighs 160 pounds with all standing and running rigging. The sail area of main and genoa is 600 square foot. Jzerro was built from 2 layers of 4 mm ply covered with 200 gsm cloth in epoxy over 19 x 32 mm stringers at 150 mm centrelines, frames and bulkheads at point of cross beams and crash bulkheads. The pods on all his 36 foot plus designs carry a double berth and serve the very important function of providing stability to 90 degrees if the proa is caught aback. The cross arms are timber plywood boxes. The cross arms were strengthened with carbon fibre in Australia after Jzerro broached on a bar entrance flexing the cross arms as it was pushed down the waves float first.

These boats are very fast in the right conditions (22 knots plus is not unknown) but need good seamen to sail them offshore. The jpegs give an idea of the structures and internal space.

 

Old multi, you don't always get everything right, but this thread on multihull structures (now 38 pages long) is brilliant!
It's like a textbook on the structural history of multihulls. I'm not sure how you find all this stuff, but I'm sure that it's a ton of work. I have been wanting to voice my appreciation, but seeing the familiar proas prompted me to apply for a new password so that I could post.
I agree about the cirrostratus at the top of the page. I saw a few of them in Australia and they are super cool.

 

Yes to the above; oldmulti has created a wonderful resource that is open to everyone.

 

Russell thank you for the kind words. One small question, what was the overall beam of Kauri and Jzerro, I could not find that detail anywhere. And for everyone, if there are any glaring errors in what I have written please advise the thread. I prefer everyone has facts, my minor ego is unimportant.

A small discussion about the Mainecat 22. There were only 16 of these boats built in the mid 1990’s. The trailable cat is 22 x 13 foot capable of being folded to 8.5 foot. The cat weighed 1750 lbs and was capable of taking a maximum 1200 lbs payload for a total maximum displacement of 3000 lbs. The rig is a sloop with 270 square foot of sail. The headsail is on a roller furler and tacks to a wire bridle. The jib is self-tacking. The designer of the boat is Richard Newick and was assisted by the builder Dick Vermeulen of Mainecat. Richard Newick also built the plug for the Mainecat 22.

The Mainecat 22 has both pivoting centerboards and rudders. Draft, with all appendages up and the boat loaded, is 11 inches. There are four watertight compartments accessible for stowage. There is a Porta Potti located in the port hull, exposed but convenient. The biggest feature that separates this cat from other small cats is the rigid bridgedeck, which gives the 22 the feel of a bigger boat. The cockpit extends down into the hulls, providing comfortable seating for daysailing. The cuddy cabins are ideal for storing the camping gear and keeping it dry.

The Mainecat 22 has a light helm and is capable of holding 15 knot speeds for short periods. It is an allrounder according to various owner reports. The cat can be setup or retrieved in under an hour according to owners. The trailers for carrying the boat require special beds to support the boat but they can be created at home. I do not have any build detail beyond it is foam glass and is structurally good. The folding hinges are on the bridgedeck and look very strong. A very nice idea that provides a great daysailer.

 

The next cat is a design study that I think was never built, hence no jpegs. Jim Brown was requested by a Philippine’s furniture manufacture to design a working boat for diving expeditions etc. Jim worked through the design process and came up with Lazy Susan an 80 x 40 foot cat weighing about 56,000 lbs and displaced about 90,000 lbs. The mast is 70 foot high carrying about 2000 square foot of sail in the main and foretriangle. This was at the limits of known multihulls at the time and Jim thought laterally about how to design the cat. Result was an X structure configuration. The X structure was integrated into the boats shell and accommodation allowing a very strong overall structure that was very resistant to torsional loads and used the shell structure especially the main deck cabin to help spread a lot of the forces being placed on the boat. EG the mast base is directly on top of the centre of the X. The hull shell was 2 plies of 6 mm hardwood and 2 plies of 9.5 mm softwood with a fiberglass layer on the outside. Total thickness about 32 mm. The X beams are made of the same layup as the hulls but the wood plies are at 60 degrees. This boat would be a massive undertaking even for a professional team but very doable with the simplicity of constant camber panels and some very good heavy lifting equipment. Some of the completed panels would literally weigh in the 1000’s of lbs. Jim also indicates that the design could be done in metal or as a composite if required. A thought provoking design study.

 

The attached document about composite sandwich mainly using Nomex honeycomb cores is heavy on the maths of test panels and their results. But a few basics to start with before we do some simplified summaries.

A single skin panel has a Relative Bending Stiffness, Relative Bending Strength and Relative Weight of 1 for each.

A composite panel twice as thick (each skin is half of the single skin panel) with a Nomex core has Relative Bending Stiffness of 7, Relative Bending Strength of 3.5 and Relative Weight 1.03.

A composite panel four times as thick (each skin is half of the single skin panel) with a Nomex core has Relative Bending Stiffness of 37, Relative Bending Strength of 9.2 and Relative Weight of 1.06.

Translation, for a minimum weight gain just making a composite panel 4 times thicker by the use of a core dramatically increases bending stiffness and bending strength for virtually no weight gain. The core material can be foam, balsa, WRC or some form of honeycomb structure. Many others have investigated composite panels with ductile foam cores making appropriate assumptions about the elastic and plastic behaviour of the core and skin. However there has been little work on failure of panels with honeycomb cores whose shear anisotropy can reveal the important role of core shear in the bending of sandwich beams. Furthermore, resistance to indentation failure is rarely considered as an important factor although it can be important in determining the durability and service life of the structure. Nomex is one such honeycomb core.

Nomex is constructed from ribbons of aramid paper running longitudinally. These are glued together at intervals along the ribbon and the stack of ribbons is then expanded into a honeycomb by pulling in the transverse direction. The paper substrate is finally dipped into phenolic resin to build up the walls of the honeycomb.

The general results of the documents investigations.

Foam glass composite panels are more flexible than Nomex glass panels. This means indentations can be more easily absorbed by foam glass panels with less total damage. Nomex glass panels tend to be stiffer and are more prone to localised fractures if the core density is high enough.

Now the really interesting finding which effects the above. The Nomex honeycomb cell size really effects the performance of a Nomex glass panel. A panel with 3 mm cells is significantly “stronger”, has less core shear and is more resistant to indentation than a panel with 12 mm cell size. Also the glass skins are more resistant to wrinkling and or torsional flexing etc. with smaller cell size. To give a relative feel for the numbers, with equal glass and core thickness’s and core densities the 13 mm Nomex cell size can only take half the impact load over longer spans than 3 mm Nomex cell panels of the same span length.

For a given cell size the density of the Nomex core effects its strength. The 29 kg/cubic meter Nomex core has half the impact strength of a 64 kg/cubic meter Nomex core. This is similar to foam cores where increased density equals increased strength.

The mode of panel failure changed depending on cell size. 3 mm cell panels generally failed due to glass face wrinkling. The 13 mm cell size exhibited elastic intracell buckling. Translation panels with smaller cell size the glass skins failed, larger cell sizes the Nomex core failed.

This is a very generalized simplified version of some of the results, details are in the PDF. Basically, the choice of core materials needs to be carefully considered. As Nigel Irens found on some of his 60 foot racing tris in the early 2000’s some tris were fracturing at the hull crossbeam intersection because the core’s chosen were to stiff. More flexible cores solved the problem.

 

Schionning design provides a nice selection chart that provides a lot of information about the characteristics of the boats, time to build and cost of the boat in basic specification. The build time would include skilled, experienced builders who are working full time and would be capable of finishing a boat in half the time of a first time builder doing it part time building job in the open. Also some people are very skilled at negotiating good bulk buy deals with minimum wastage versus others who are buying bit by bit and paying for small quantities.

Result, this is a reasonable guide but not an accurate assessment of the reality of a home build versus custom build boat. I would not be surprised at some people taking 50% longer. I have watched a part time 32 footer take 5 plus years, a full time flat panel 36 footer take 4 years and a 48 footer has so far taken over 5 years part time and the man buys every kit component (eg internal furniture) he can. My guess is he has another 3 plus years to go. All of them have been built inside sheds.

Also Schionning designs are generally fairly light weight with smaller payloads than EG French cats. The boats require good building materials and practices to build down to the specified weights or the available payload capability will be reduced. After watching several Schionning cats being built from 32 to 48 foot of strip plank and/or flat panel construction I can assure you at times you are pushing the limits of some of the materials. EG the chamfer panel connecting the wing to the inner hull side is a flat Duracore panel that needs a lot of persuasion to fit in his 36 foot flat panel cats. It has a twist in the panel as it goes forward and once is in place and glassed it’s OK. It just takes time and effort. Schionning boats are fundamentally well designed but they depend on you following the plans for EG internal furniture closely as the furniture often forms a structural part of these boats. The attached PDF provides the chart.

 

Friend of mine took 5 years - every day in the shed apart from about 8 days off a year. Nice 1320 Waterline. That works out to about 14 000 hours. He had fun and produced a nice boat so it was worth it for him. He had outfitted a bare hull before so he wasn't a neophyte.

 

Various fabric, types, styles and weights is the core of most modern multihulls. The following document provides a snapshot of a variety of cloths weights and thicknesses. There are many companies providing similar documents which can be very useful. The fabric type of E glass, Kevlar, Carbon and combination fabrics hybrids gives a clue of options. The material column tells in the case of carbon is an indicator of the number filaments used in one roving. 1K stands for 1000 filaments, 3K is 3000 filaments etc. For unidirectional fabrics the filament count number does not matter. For woven fabrics it can have an effect. In the hybrid section the % of carbon versus aramid can make a real difference as the two materials have different structural characteristics. The E glass multiaxial fabric 0 45 90 -45 degree orientation will give a clue as to the real strength direction of a fabric. The thickness of most fabrics helps in calculating the total thickness and weight of a layup. Useful, but as I said most suppliers have these charts and they can tell you a lot about the product you may be buying.

 

The cat for discussion is a 50 x 27 foot weight 20,000 lbs displacing 30,000 lbs carrying 1500 square foot of sail. Its real but I will not name the design. This boat is a “typical” modern high performance bridge deck catamaran. We will focus on the hull structure today, then decks, then interior etc.

The hulls are round bilge high length to beam. The bridge deck is 3 foot of the water. The hull bottoms are 1000 gsm e glass quadax (0 90 45 45), 20 mm pvc foam minimum 5 lbs/cubic foot, 850 gsm e glass quadax (0 90 45 45). There is an additional layer of glass along the keel line for abrasion resistance. The hulls are topside 1125 gsm e glass triax, 25 mm PVC foam of a minimum of 4.7 lbs/cubic foot, 751 gsm e glass triax. Normal bulkheads are 751 gsm 45/45 or triax, 15 mm pvc foam minimum 4.7 lbs/cubic foot, 751 gsm 45/45 or triax. The underwing is 1125 gsm e glass triax 40 mm PVC foam of a minimum of 4.7 lbs/cubic foot, 1125 gsm e glass triax. The underwing support beam is of 2 layers of 600 gsm 45/45 over a foam core. Dagger cases are 751 gsm 45/45 or triax, 15 mm pvc foam minimum 4.7 lbs/cubic foot, 751 gsm 45/45 or triax. All of the above is done with vinylester or epoxy resin.

Next we will discuss the cross beam structures and decks. The attached jpegs ARE NOT the cat being described here there just to give a feel for the type of cat being discussed.