Multihull Structure Thoughts

Whilst we are talking about Shotover, the original builder/owner Adrian Rogers then designed/built a 40 x 25 foot light racing cat Shotover 2. This interesting part of this boat was its schooner rig. Shotover 2 has 2 equal high masts with fully battened mainsails. The 2 mainsails had big roaches when they were not normal and were placed close together. Before the boat hit the water many detractors predicted the rig would be slow due to interference between the sails. Shotover 2 was fast on the race track. Upwind was meant to be a weakness but proved to be good because the forward mainsail, when sheeted correctly, smoothed air flow onto the rear wing mast mainsail unit. On a reach this boat was fast. Another smaller home built version designed by Ross Turner the Jarcat 7 23 x 15 foot weighting 900 lbs carrying 300 square foot of sail (designed to be trailed) had similar characteristics to Shotover 2. Jarcat 7's can sail quite well across all points of sail. Wharram also was surprised at how well his 31 foot schooner rigged cats sailed. Again good design can surprise us all. The only problem is you have to build 2 masts, 2 rigs and deal with 2 mainsails. It is probably cheaper than multiple headsails winches etc but looking after 2 wingmasts whilst cruising would not be relaxing.

 

A photo of Shotover 2 sailing.

 

I will start a small series about cross beams. I will start with the older style of cruising catamaran cross beams and work forward. The first crossbeams were plywood and timber (many designers still use them today), then they evolved to combination of timber, plywood and often unidirectional glass. Later versions had some timber, plywood, foam and glass. Later versions are mainly foam glass with some compression timber. Finally some designers are using carbon fibre and foam. As cross beams evolved they became lighter, often easier to build and more integrated into the structure. Now I need to specify the type of boat I will be taking about, They will be bridge deck cruising catamarans that have a main mast supporting cross beam and an integrated rear beam. The forward beam will have a forward wing deck which acts as a torque box to take care of torsional twisting loads through the wing frames and associated upper and lower decks. I will do something later on small tube cat cross beam calculations. I will give details about several cross beams, some I will not identify the boat as the design is still being sold commercially. On the first plan the forward cross beam structure can take about 4 times the load of the rear cross beam which was normal for the spindrift series of 37, 40 and 45 foot cats. All the beams were built with the same concepts just different sizes of timber and ply. The first plan is the forward cross beam.

 

Now we will jump ahead to how to construct a foam glass cruising catamaran cross beam structure. The following web site story_e1 http://www.saunalahti.fi/pekkajlh/boat/story_e1.htm is about the building of a Cosmos 1100 bridgedeck catamaran 36 x 22 foot displacing 9000 lbs. I will be giving structural details of other cross beams in a later post but please understand the overall process. The following web site Autumn01 http://www.saunalahti.fi/pekkajlh/boat/Autumn01.htm gives a very good idea how to build a strip plank forebeam. It also gives some details of the building of the main cross beams on Winter01_02 http://www.saunalahti.fi/pekkajlh/boat/Winter01_02.htm The unidirectional glass is simply compressed easy on one edge but will require a reverse mold for the reverse curve. The installation of the bulkheads is shown on this page spring_summer02 http://www.saunalahti.fi/pekkajlh/boat/spring_summer02.htm Please understand the process and the details shown, it will help later.

 

The following Coastal Passage magazine contains a "full" cat plan for a 30 foot cruising cat that was built for $21,000 Australian. The boat uses a lot of simple build tricks and cheap materials but if you want to get on the water quickly it can be done. The main interest is the wooden cross beam described as it shows how it can be done effectively and cheap. It will weigh a bit due to the 18 mm thick ply on both faces but it will work. The web address is attached and a PDF copy of the magazine. The design starts on page 22. https://www.thecoastalpassage.com/papers/tcp75.pdf If you go to the web site look around at the logs of other boats built as they also contain some useful information. Look at this excellent sanding tool. Building a cat https://buildacat.com/bbsplit-sander.html

 

The following 2 plans show the rear cross beams of a Spindrift cat. Similar rear cross beams plans but different features as there is an original plan and a mark 3 plan. Again its a timber ply cross beam and and the structure has less strength than Spindrift main cross beam structure shown earlier in the thread. Remember this only applies to bridgedeck cruising catamarans. Small tube cats tend to have the same strength beams for and aft because of the torsional loads and crew weight. Most multihulls tend to have there centre of pitch around 60% aft of the bow which puts greater loads on the aft beams as the sterns move more. The main beam is often placed around 45% aft of the bow so the mainbeam wing box handles most of the diagonal twisting (torque). In small cats if there is no forward wingdeck attached to forward beam, the rear beam handles most of the diagonal twisting (torque) control. This is a simplified explanation of the forces involved in a cat as it goes over waves.

 

After timber and ply crossbeams the second phase of cruising catamaran crossbeams evolved to combination of timber, plywood and often unidirectional glass. This is one sample of a bridge deck catamaran 36 x 22 foot displacing 9000 lbs forward and rear crossbeams. The main mast carrying forward cross beam is a plywood box beam with a uni directional glass top and bottom flange. The beam and associated hull bulkheads run the full width of the boat and are 950 mm high in the mid beam. The main beam "box" is 6 mm ply web faces 48 mm apart with 48 x 30 mm timber verticals in between the web faces at 750 mm centrelines. Diagonals of 48 x 30 mm run between the vertical timbers. Along the top and bottom edges of the bulkhead is a 48 x 25 mm timber strip. The web faces have 45/45 glass epoxy on the faces. The plywood extends above and below the timber internal frame to create a channel for the Unidirectional glass. A 48 x 40 mm unidirectional glass/epoxy strip is built up in the top channel to form the top flange. In the bottom of the bulkhead across the bridgedeck and down the sides of the inside hulls is a 48 x 35 mm unidirectional glass epoxy flange tapering to 20 mm near ends. The rear cross beam is the same basic timber ply basic structure but only 40 mm between the ply faces and approximately 700 mm high in the mid crossbeam. The top and bottom unidirectional glass/epoxy flanges are 40 x 35 mm tapering to 40 x 20 mm at the ends. The layup of the glass flanges needs to be done over about 12 to 24 hours to prevent excessive heat build up as the unidirectionals cure which will cause fracturing of the unidirectional layup. the mid bulkhead (cabin cockpit bulkhead) has 30 x 15 mm glass edging on it.

 

The third phase of cruising cat crossbeam evolution was the glass face webs foam core with unidirectional glass epoxy flanges on top and bottom of the bulkhead. The attached C106 1 PDF shows the design of one boats cross beam structure is done. Many designers/designs have a similar approach to crossbeam construction which allows us to do some interesting reverse engineering. Some designs will not be named due to the plans being still sold. I will focus on the top flange of the main mast supporting main beam in each boat. Righting moment is half the distance between hull centrelines by the weight or displacement of the boat. Sorry I am going to mix up mm with square inches for my convenience. Older Windspeed 36 x 23 foot cat 5000 lbs weight = righting moment = 42500 f00t/lbs. Top flange 19 x 50 mm = 1.75 square inches. Divide top flange into righting moment. Top flange stress 42500 divided by 1.75 = 24,300 lbs/square inch. C106 is a 40 x 23 foot cat 12000 lbs weight = righting moment 114,000 foot/lbs. Top flange is 72 x 40 mm = 4.5 square inches. Top flange stress 114,000 divided by 4.5 = 25,333 lbs/square inch. A 46 x 23 foot cat 12000 lbs weight = righting moment =120,000 foot/lbs. Top flange 80 x 50 mm = 6.2 sq inches. Top flange stress = 120000 divided by 6.2 = 19400 lbs/square inch. A crowther 60 x 23 foot cat weighting 27,000 lbs = 324,000 ft lbs righting moment. The top flange 80 x 100 mm = 12.1 square inch. The top flange stress is 26,800 lbs/square inch. Notice the consistency of the top flanges lbs/square inch stress which in this set of examples is between 19400 lbs/sq inch to 26,800 lbs/sq inch.
Now we talk about resins Polyester = 9400 lbs/square inch tensile strength. Vinylester = 11800 lbs/square inch tensile strength. Epoxy = 12500 lbs/square inch tensile strength. Fiberglass unidirectional E glass epoxy laminate ultimate tensile strength = 257,000 lbs/square inch. Fiberglass unidirectional S glass epoxy laminate ultimate tensile strength = 342,000 lbs/square inch. The ultimate tensile strength is assuming a perfect layup with a high glass to resin ratio. The effective maximum working load is about half the ultimate tensile strength. So assuming an average E glass unidirectional epoxy layup the safe maximum working load will be about 110,000 lbs/square inch. Now put in a Safety factor of 4, the realistic maximum top flange breaking strain should be 27,500 lbs/square inch. Notice that the cat crossbeams top flange lbs/square inch is between 19,400 lbs to 26,800 lbs/square inch. So put simply get the righting moment of your cat divide by EG 22,000 and you will get approximately how many square inches cross section your top flange on your main beam should be. WARNING this is reverse engineering based on my database not proper calculation that is (hopefully) done by designers.

 

Whilst we are on beams I will tell you about a Wharram Tiki 31 builder who had 2 sets of wooden crossbeams rot out so he decided to replace them with a set of foam glass composite beams. This is a warning. The new beams were made of Divinycell foam and glass. The divinycell is solid and is the same dimensions as the original wooden beams. The tops and bottoms have two layers of 3/4 oz CSM, a layer of 24 oz unidirectional glass fibers and a layer of 10 oz biaxial carbon on them. The fore and aft faces of the beams have 4 layers of 3/4 oz CSM, 2 layers of 24 oz unidirectional glass fibers and 2 layers of 10 oz biaxial carbon on them. Some additional glass was wrapped around the ends that lay on hull gunnels. He did not get them engineered by a designer. There are only a few problems with this approach for a tube cat. Beams "strength" needs to be about twice as strong vertically as it is fore and aft. This beam is twice as strong fore and aft as it is vertically. Next CSM has very little strength especially in a beam situation. Next mixing Carbon fibre and fiberglass together is a real danger. Fiberglass elongates more than carbon fibre. So if real stress goes on the beam the carbon fibre breaks first then the fiberglass loads up. Do either carbon fibre or unidirectional glass but not both on the same face. Finally the entire beam does not have to be solid divinycell. It can be a "box" with foam glass walls and unidirectional glass or carbon on the top and bottom. Pay for a designer when it comes to beams. The money you will save on materials will probably pay for the designer. The attached is photo's of the beams.

 

We have mentioned Ray Kendrick as a trimaran designer who does a series of 16 to 32 foot tri's and 2 cats. Each of his designs cost $150 irrespective of size. Good value. His folding tri's have been built by many but the best home builder I have seen is a canadian who knows about Finite Element Analysis (FEA). He did FEA on the Scarab 22 design as he built it, strengthening and simplifying it as he went. His blog is Scarab 22 construction http://www.voile.org/trimaran/ Look thru the blog it will provide many hints on structure, building, vacuum bagging, resin infusion etc. The complete boat took 1500 hours to build. The cross beams pages give a good idea of how much work can be involved in building folding cross beams. The following page is the main beam page 4 cross beams! http://www.voile.org/trimaran/progress/Mars05/mars_10_2005.htm The guy sold his boat after a few years then looked around for another design to build. Designers were very interested in having him as a client as one designer wrote "he asked if any FEA had been done on the main cross beam of the design. I said no and welcomed his interest." The pictures are of the scarab 22.

 

The next trimaran is a home design home build. www.trinardo.com http://www.trinardo.com/home The designer produced a reasonable design shape for a 20 foot day sailor but I cannot make any comments on its sailing ability. The interesting part of this boat was the testing the guy did on the foam used in the foam glass construction. He tested extruded polystyrene foam against Divinycell H60. MaterialTest - www.trinardo.com http://www.trinardo.com/y.a.t-yetanothertrinadofoamcompare3 The results were that a Finish high density extruded polystyrene foam was, after testing, his final choice for the foam component of his small tri. I know a 29 foot racing tri that has blue polystyrene foam plus thick glass in its cross beam structure and has survived 20 years. I also know of low density polystyrene foam glass furniture in a cat that started to crumble after 2 years. I also know other smaller (under 20 feet) mono's, tri's, proa's that have survived with extruded polystyrene in their structures with glass coverings. In short if you want a cheaper day boat test and try a higher density (above 35 kg/cubic meter) extruded polystyrene foam (hopefully with closed cell and low water take up under 5%). I am not suggesting this material for any out of sight of land sailing, just cheap fun day sailor or experiments. But if you want real fun with blue foam you could buy R Parker's conch 32 foot monohull plan that has blue foam as a core in its deck and cabin structure. Its light, has a radical rig and Parker normally designs a good boat. Again no guarantees about the structure. I will get back to more sane deck structures next.

 

Before more sane deck structures I will mention a cross beam approach used on some large catamarans if you cannot find suitable timber or aluminum tubes. First is a 46 foot classic Wharram builders used steel rsj I beams as cross beams. Old Sailor sailed on the boat in Canada and found it solid. It did thousands of trouble free miles. Hugo Myers (designer of 44 foot seabird) designed a racing 50 x 25 foot version of seabird called Manta. It weighed 9000 lbs with 1400 square foot of sail. The cat had a main, forward and rear beam of aluminum I beams. The main beam was 300 mm high with 175 mm top and bottom flanges of 12 mm aluminum. The forward and rear beams were 250 mm high with 150 mm wide top and bottom flanges that were 12 mm thick. The beams were bolted to the hulls with 19 mm thick stainless steel bolts. The main beam has a dolphin striker on it. I dont know if the design was ever built but I knew an Australian guy who brought a set of plans and then immediately said he was going to build a deck cabin on it.

 

Oldmulti et al, if this is the right place and time for the question, Sven Yrvind posted:

‘‘ . . . I use carbon on the inside. There it contributes very much more to strength than if I had placed it on the outside, because on the inside the fibers takes up tension forces distributed by the Divinycell over a large area. The tension forces has its origin in much smaller local compression on the outside skin. . . . ’’

And he posted the next day some about 15 years older hamer tests he did on that on plywood.

- Sept. 30, 2018: work on the inside (about his current build 'Exlex Minor' aka 'new Exlex')

- Oct. 1, 2018: more about carbon on outside or inside (about the older tests on plywood)

His theory and test looks logical to me, but I've never heard of it elsewhere, what's your opinion on this matter . . ?

 

Angélique
Any structure and its arrangement needs to be analysed and the load paths identified and materials sought to overcome the loads it is expected to experience.
What may be true for the case you cite, may not be true for another arrangement.

Every design has its own unique way of doing things and cannot be assumed to be the case for another vessel. JEH sums it up nicely here - what he said.

 

Well, the loads besides getting smashed on the rocks or a collision aren't like puncture blows with a sledge hammer, but those things could happen. So I'm wondering if it makes sense what Sven says there as I can't analise it myself. His previous 4 boats were complete failures (thread Yrvind post #560), but not structural, besides always being far more heavy than planned. His 1971 20' plywood Bris brought him fame, but for that it needed a complete rebuild and re rigging and a keel after a first sail from Sweden to the Netherlands, from which she needed to be trailered back home to do the reconstruction works. His next boat the 1976 19½' aluminum Bris brought him Cape Horn success and fame, but its aluminum was rotting away as he was pulled a leg on what was sold to him to built her. So in general I'm a bit wary about Sven's theories, although a lot seems to make sense when he explains it beforehand.