Snelson’s Canoe
In my last version of Palindrox, I went unstayed based on my choice of a cambered junk rig. But it bothered me deep down inside that I was not taking advantage of the natural tensegrity of the proa. So this design is about making tensegrity the design driver.
What we have here is a 22’ proa “beach cat”. If the vaka split into two pieces, this whole boat could go on top of a mid sized car. He has almost 110 sq feet of sail in his balanced and boomed jib. He’d slab reef. But his neatest feature is the forked tensegrity mast.
He’s named after tensegrity sculptor and Bucky Fuller student Kenneth Snelson. You can check Ken’s work out here:
http://www.kennethsnelson.net/
Rudders, the stuff that connects the beam to the hulls, a trampoline and other doo-dads are coming soon.
Chris
What an interesting concept! Beautiful and magic. I wonder if it is possible to fix the beam to the hulls in the same style.
Alexander
That’s very nice. I’ll be interested to see how you do the rudders. I did a design once that used the tensegrity idea (before I knew what a tensegrity structure was, so it’s not a pure tensegrity structure) to create a pantographing ama proa with a roller reefing ‘vortex’ sail. The rudders were suspended on wires as well.
Alexander:
Thank you! I love your idea of mounting the hulls to the crossbeam/fig structure through tensegrity. You have any ideas how to do that?
Mal:
Wow, that’s cool! It’s very Bernard Smith (thank you, Michael for turning me onto that!). Snelson’s canoe is also deeply influenced by his tetrahedral concept for a sailboat. BTW, my boat is not a pure tensegrity structure either. The forked mast breaks the pure tension vs. compression rule. As does the (as yet undrawn) clamping of the hulls to the aka. But the more I think about the virtual mast step the more I like it.
Ok here are a few more views. I modified the mast so its less flat, built a more tensegric (is that a word?) boom and closed up the stems a little.
Here’s a question. Is this a new proa rig? What’s its closest neighbor? I was trying to avoid canting the mast, and this solution came to me. Does the shunting make sense to folks?
chris
Nice, very nice.
Thanks for the link to Snelson’s website, some of the content could be classified as geometric porn IMHO.
One of my first thoughts after looking at the design was you could eliminate the fork in the mast by using two beams but that is counter to the whole concept of one hull, float, beam, sail, mast….... Another alternative could be to have a hole in the beam for the mast to pass thru but that’s probably no better than the forked mast. For the semi-tensegric (?) Bionic Broomstick I took the cheap and easy way out and landed the mast on the beam, everything else is fairly pure.
The sail rig should work well and shunt simply. The only real downside to a clubbed jib that I came across is that rig forces are large to provide luff tension. P52’s clubbed jib was the best of several rigs tried over the years.
http://www.youtube.com/watch?v=iVoVP5ozj4I&list=FLalt_wPBV70Iv9dYXHu6xEg&index=1&feature=plpp_video
cheers,
Skip
Thank you, Skip. I’m a big fan of your work!
Correct me if I’m wrong here; your “forestay” on your club jib gets cranked down towards the new bow on each shunt, right? Was the problem getting enough tension on because of mechanical advantage, or because of flex in the structure?
My hope for this is that by putting a forestay and backstay on my club, and having that whole system in direct opposition to the windward shroud, I can crank on the tension, and leave it be. The rig would then pivot on swivels top and bottom. By creating a closed system between the aka, mast, rig and windward shroud, I’m hoping for a relatively stiff structure. The two stays that go from the mast to the vaka bows are really just for keeping the whole thing from falling over front to back.
Did you find that a lack of forestay tension hurt your windward performance?
Where do you come down on the alignment of the drive and drag question? I’ve tried to bring the center of effort a fair bit to windward. My lower swivel is about the same distance to windward that madnesss mast is, proportionally. Your boats (like crab claw proas) have the rig pretty far to leeward. Any thoughts on how that matters or doesn’t?
I think the forked mast might be ok. Especially if I put another tension stay between the forks.
Thanks for your input.
Chris
You’re welcome Chris,
And thank you also, though I’m not sure how much credence should be given to an off and on old hobbiest like yours truly.
My first reply got lost in cyberspace when I spent so much time on it that the forum decided I was a spambot.
So a little quicker this time.
Staysails are almost by definition high tension devices. Straight luff = infinite tension, tension doesn’t fall off that much as the luff sags. James Brett’s nice ‘Free Radical’ switched from a staysail system to a cambered junk for that reason (I believe).
P52’s weakest link was the 5/8” braided polyester bridle on windward shroud I left in place when I switched from 5/16” double braid (OK for crab claw) to 3/16 & 1/4” amsteel blue for standing rigging. Flex there was what allowed mast to go out of column and snap during first try at the Texas 200.
http://www.texas200.com/2009/stories/johnson/index.htm
There wasn’t that much lack of forestay tension, P52 could go to windward, not exceptionally, but it could do it in 16” of water or less. Biggest problem for windward work was a full cut polytarp sail.
Regarding balance of drag and drive, I’m not sure.Having rig to leeward doesn’t seem to adversely affect windward or reaching performance. Downwind sucks. Hence my effort to have a rig that can do both, time will tell.
The joy and the agony of contemporary proadom is there isn’t a well established body of knowledge and rules of thumb to guide us, so we have the joy of discovery tempered by disasters (at least in my experience).
cheers,
Skip
By creating a closed system between the aka, mast, rig and windward shroud, I’m hoping for a relatively stiff structure.
I once did a few rough calculations for the structure Bernard Smith proposed to form the tetrahedral structure for his fliptackers. I was surprised at the compression load on the spars. I’ll try something similar for your proposal. More knowledgeable people can improve on it (or just shoot it down in flames; this is only an approximation).
Just from eyeballing your drawings, I am guessing the sail’s centre of effort would be about 4m above the centre of lateral resistance. Say the crew rides 3m to windward of the lee hull’s centre of buoyancy and that the weight of crew and ama is 800N (Newton, the force exerted by gravity on a mass of 102g at sea level, or the force needed to accelerate 1kg at 1m/sec^2). Given the lever arms, that means 600N at the sail’s centre of effort. If I model the force exerted on the sail as a point force on a wire at the centre of effort, I can calculate the force on the wire if I decide how far the wire is allowed to sag. Say 20cm.
I think modeling sail force a a single point overestimates the force on the wire compared with force distributed along the wire, but I think it will do for a ball park figure, especially because there I presently ignore another factor that will increase the forces.
If the sail has 7m span with centre of effort at 3m from the boom, then 20cm sag perpendicular to a 3m distance means the force component along the line from boom to mast top has to be 600N * 300cm/20cm = 9000N.
The angle between mast and windward stay looks to be 22 degrees (ratio horizontal to vertical = 1 to 5), the angle between mast and sail looks like 14 degrees (ratio 1 to 8). The vertical load on the mast from the sail is then 9000N *cos(14degrees) = 8930. The horizontal force at the mast top is 9000N * sin(14 degrees) = 1116N. That has to be balanced by an equal horizontal force from the windward stay. I can simplify the calculation by looking at the vector diagram, and simply take that 1 to 5 ratio. The additional vertical load going into the mast is 5*1116N = 5580N. That adds up to 14510N compression on the mast, not counting whatever tension you put in so that the stays to the bows don’t slacken.
In the frontal view, the depth to beam ratio of the wire bridle is about 1 to 8. Each Newton vertical force generates 8 newtons horizontal force. The compression on the beam then is 116080N. If you stood the beam vertically, that’s the force generates by a weight of 11840.16kg or 26137.55lb.
If the allowable sag is 10cm, double that.
The thing about such structures is that compression forces can multiply with successive steps, like compound gearing. There is an additional step I have left out. The sail’s profile is partly determined by forces from the foot to the head, but more by forces along the shorter path between luff and leach. To prevent the sail profile from getting too deep, you must have enough tension to prevent luff and leach from moving too close together. That force in the sail fabric is another force multiplier of the same type as before, only more complicated to model than I have done here. The sag you can allow will be limited. I expect more in the range of 10cm than 20cm. My 1.45 tons compression on the mast and 11 tons compression on the beam are likely to be an underestimate, I guess by a factor 2 or 3.
Add dynamic loads and a safety factor for fatigue. If it were my design, I would look up the dimensions for beams that can take at least 50 tons compression load. The strengths of materials can be quite surprising, so perhaps such a beam would not be too heavy.
If you use rigid battens as in the junk rig, you don’t need the vertical tension to keep luff and leach apart. If you sew in camber into each panel, you almost abolish vertical loads on the sail from aerodynamic loads. Basically, you allow a lot of sag over only short distances, and some of the aerodynamic load even contributes to keeping the battens apart. Having removed two of your force multipliers, the loads on the structure will be correspondingly less.
Of course, there may be a fundamental flaw in my line of reasoning. Perhaps someone who knows more about tensegrity can chip in?
The various versions of Exoplane that Dider Costes brought to Weymouth Speed Week on the 70s and 80s had some elements of a tensegrity structure, though he used a sail design that could be allowed to sag. I read he would spend most of the week assembling the boat. Onlookers speculated that he knew he was finished if a pigeon released in the middle of the rig could not escape through the wires. Then Costes would go very fast until something snapped. As a consequence, he had few completed runs, and the measured speeds were not often impressive.
His experience with a later Exoplane design using a tetrahedral tensegrity structure much like Bernard Smith’s (though I understand that Costes’ development was independent) was that the boat was too wobbly. I am not sure this is the way to get a stiff structure. The one and only direct comment on stiffness I remember reading was that such structures were not stiff, that the point was their ability to give and distribute loads. If my reasoning above is correct, you have to take care to avoid that compound gearing effect that I think you have.
Then again, Bernard Smith’s last monomaran, the one shown at http://www.geocities.com/aerohydro/designframeset.htm had a tensegrity tetrahedral frame, and he reported it working fine. I don’t remember an explicit comment on the stiffness, but at least it did not seem to be a problem.
Regards
Robert Biegler
If you use rigid battens as in the junk rig, you don’t need the vertical tension to keep luff and leach apart. If you sew in camber into each panel, you almost abolish vertical loads on the sail from aerodynamic loads. Basically, you allow a lot of sag over only short distances, and some of the aerodynamic load even contributes to keeping the battens apart. Having removed two of your force multipliers, the loads on the structure will be correspondingly less.
Robert Biegler
I had been of the opinion that a junk or battened type sail would benefit significantly stresswise from some form of parrells to fix the lateral location of the battens, now I’m not so sure. Only way to really find out is to finish the boat I’ve started and try it both ways.
Thanks,
Skip
Thanks very much for your thoughtful reply, Robert.
What is the fundamental difference between my forked mast, pseudo tensegrity mast step compared to stepping the mast on the beam and having a dolphin striker under the beam? Dopesn’t that load up just the same way?
Another flaw in my system is that it is not in pure tensegrity. There are not going to be pure compression/tension loads—but bending and torsional loads imparted by the beam being rigidly attached to the vaka and ama. The sea forces have a terrific opportunity to knock the aka out of column. That would not be good when its loaded up 33 tons of compression. Plus, its hard to imagine how the stays could align the compression forces with the centers of the columns, once all of the clamps and stuff are in place.
Yet another flaw is that I’ve adopted the compression load issues that catamarans have—putting them between the hulls rather than mostly on the vaka.
But it sure is pretty! I love the look and simplicity of the balanced jib. It solves a lot of problems…while creating a whole slew of new ones…
ah, design…
chris
But it sure is pretty! I love the look and simplicity of the balanced jib. It solves a lot of problems…while creating a whole slew of new ones…
ah, design…
chris
Chris,
Ah indeed.
Snelson’s Canoe is one of the slickest things I’ve seen in a while, I hope you keep on developing the concept. I fiddled with your balanced club on Nomad (imitation is the sincerest form of flattery) but results had a certain “camelish” tinge, suspect once again designs need to be developed in toto. Don’t be surprised if your balanced jib shows up down the road, probably battened and cambered.
One detail, Phil Bolger claimed that a balanced jib was very easy to overbalance and the pivot needed to be at one fifth point down to one eight.
Skip.
I fiddled with your balanced club on Nomad (imitation is the sincerest form of flattery) but results had a certain “camelish” tinge
COOL! You didn’t happen to take any pictures did you? What is a camelish tinge? Did you have forestay tension issues?
I have a really good feeling about this solution. It has a lot of holly grail qualities for proa rigs. Its balanced, for low sheet loads. The CE is both to windward and very close to the mast, fore and aft. No raising or lowering sails for shunting. It’s feathered during the shunt. Its supremely reefable, either with hanks or furling, but I think I’d go for hanks. But the rigidity to get all that tension is a toughie!
I like the idea of battens and camber. But full battens don’t allow me to have a big old roach that can pass behind the “backstay” when shunting. Good comment on the balanced club jib. I have the pivot point 1/3 of the foot back. Part of the reason I did that was to move the CE forward. But with the pivot 1/5th, that should improve forestay tension. The smaller the ratio, the less “backstay” it takes to tension the “forestay”.
I think that the next version will have a more conventional mast step on the aka, with a (still forked) dolphin striker. Maybe I’ll move the balance point to 1/5th, which will bring the mast to leeward a bit, and maybe reduce my compression issues. Robert Biegler, any thoughts?
I’d love to see what you’re envisioning for the cambered junk version. Does it raise and lower on a single stay where the mast would normally be?
best,
chris
I fiddled with your balanced club on Nomad (imitation is the sincerest form of flattery) but results had a certain “camelish” tinge
COOL! You didn’t happen to take any pictures did you? What is a camelish tinge? Did you have forestay tension issues?
best,
chris
Nope, no pictures, just doodling in 2D Autocad. Camelish tinge refers to camel being designed by a committee, just picking up a good idea and slapping it on an existing design idea is not the way to truth and beauty. No idea about forestay tension as of yet.
I think you are right about there being a lot to like about the rig. The real downside is all that rig tension but Robert B may be on to something about the cambered panel sail and straight stiff battens drastically decreasing the amount of tension required. Best way to determine that is to try it out so I’m already noodling a second rig to try out on the broomstick.
If the balanced club, cambered panel, stiff battened rig works there, Nomad may well end up with the same in a schooner rig. Such an iteration would probably roller furl on the basis that the battens would provide ‘catch up’ points for sails tendency to bag up. Masts would step on windward side of hulls adjacent to pod shear web, a good structural location. Raising and lowering rig up in the air a bit (?), I don’t see how to do it at the moment except to set flying with the boom held in place with a couple of jack stays, still a lot of goodness is possible.
Course it all depends on finishing the broomstick and trialing things out. If it does work there’s a lot to recommend such a setup. I’d always avoided a shunting schooner because the rigs I’m enamored of all require too much fiddling to shunt. But a fixed tack, shunting, balanced club, cambered panel, roller furling (whew!) schooner wouldn’t be that onerous. On the plus side, cockpit’s completely open, lots of steering and trim options, will probably look pretty good.
Thanks to you and Robert Biegler for pointing me off in this direction, don’t know exactly where it’ll lead but it’s a fun journey.
Skip
As usual, I’d love to see some pics of your doodling, however crude or camelish.
And you’ve inspired me to bang out another drawing this weekend….
Chris
Sorry, I simply missed your reply.
What is the fundamental difference between my forked mast, pseudo tensegrity mast step compared to stepping the mast on the beam and having a dolphin striker under the beam? Doesn’t that load up just the same way?
At that point there is no difference I can see. But then, I remember reading in the 80s that the mast of a racing catamaran weighing about 7 tons put 50 to 60 tons vertical compression load on the beam. I have no idea whether that boat had a dolphin striker, and if yes, how much horizontal compression that put on the beam. I expect part of the load is the multiplication factor imposed by the angles of the shrouds, part is main sheet and forestay tension. The higher aspect ratio rigs on the ORMA 60s and the new MOD 70s probably impose even higher loads. The 40 multihulls that flourished for a few years in the late 80s became too expensive because the loads imposed by increasingly high aspect ratio rigs demanded high tech solutions.
Have a look at the video of Geant in rough weather: http://www.youtube.com/watch?v=Lc3Mei46PdY. If you watch carefully between 1:18 and 1:19, as the boat pitches up again you can see the downwind ama shiver briefly, at a frequency I estimate about 2 - 4 Hz. Think of the size of the structure. For something that big to have such a fast resonance frequency takes incredible stiffness. Though I admit that stiffness is not the same as strength.
Getting back on topic, there is one potential difference elsewhere in the structure. I have a vague notion that tensegrity structures are designed to take their loads at the vertices where spars and stays meet. One reason why the forces get so high once you have a staysail in there is that the force is applied perpendicular to a tension member. Someone even used that principle to design a nutcracker that delivers high compression loads for modest pull on a wire. If you then want to keep deflection of that tension member low, you need enormous strength and stiffness.
Regarding the balance point of a jib, I don’t know what data are available regarding the centre of effort of thin wings such as single surface sails. What I have seen stated that the centre of pressure moves further forward in low aspect ratio wings. The normal rule of thumb for conventional wings is that at best lift to drag, the centre of pressure is at about 25% of chord length. 20% should be safe enough, and still give you reasonable control over the leach. Give the boom several points between 20% and 25% where you can attach the downhaul, so that you can try out some variations. Design the clearance between boom and mast for the 25% point.
You should reduce compression load a bit by moving the mast further from the centre of the beam and increasing the angle between mast and windward shroud. I haven’t tried calculating how much.
Regards
Robert Biegler
I really like the idea of moving the rig to leeward and canting the mast to windward. It moves the compression load closer to the vaka but keeps the CE of the rig well to windward. I also changed the dolphin striker so that it stretches from the windward end of the aka, under the mast and then to the rig pivot. This looks like it closes the tension loop better.
I played around with stepping the mast on the aka, but then I have to make a compliant mast step—like a ball joint. It just didn’t feel as cool and minimal as the forked, floating mast.
I’m still pondering a battened cambered version, but I"m having a hard time getting it to balance. Either it goes up and down on the forestay (threaded right through the forward end of the battens), and then you need a backstay to balance the club, and everything has to stay inside the triangle. It makes it pretty sparse on the top Or the stay that takes the place of the mast goes through the pivot axis of the rig. I don’t like the stay breaking up the camber—though folks say it doesn’t make a difference. Maybe the sail splits along the pivot axis so that the forward half of the panel acts like a jib to the back half. Has that been investigated by anybody?
The next version is going to have a T2-esque hull (to eliminate dagger boards) and pivoting, surface piercing rudders on the leeward side; a lagoon racer.
Thanks for your comments, its helping the evolution!
chris