A Game of Efficiency
Future Sailing: Imagining a Single-Skin Wing
Inspiring as it is, and yes, it is inspiring, to see the one-design AC45s racing in US waters, it stirs my appetite for the custom AC72s yet to come.
Whenever I do a public program about the America’s Cup, I run people slapdash through the history, because it’s a brilliant history. Consider how many times in its first 132 years—the longest winning streak in sports—the America’s Cup was defended against a faster boat. Consider how unlikely it was that Dennis Conner and company would push their losing, 1983 match to seven races against a much faster boat and almost keep that winning streak alive. Switch boats and...
How do you build a legend? I ask that, and then I run a show of AC45s on the fly.
and then I cut to a comparison rendering of one-design 45-foot catamarans versus custom 72-foot catamarans and —
— there’s always a gasp.
I remind the audience that the hulls, wings and foils of the AC45s are not cutting edge. They represent “safe” technology, because the engineers had to get it right the first time, to get this fleet sailing. And they got it right. Demonstrating that this much of the envelope is known.
Under the design rule for the 72-foot catamarans that will actually compete for the America’s Cup, engineers are accorded dimension limits, and within those dimensions, they have a clean sheet of paper.
More-complicated wings are a given. Curved daggerboards, supporting most of the displacement, are a given. There’s an invitation to over-achieve, because the time frame for trial and error is short, and it is easy to imagine creating something that is faster, but not manageably faster.
In that lies the “art” of engineering.
Oracle Racing Team Coordinator Ian Burns puts it this way: “I was involved with writing the rule for the AC72s, and when we addressed the wing, we started with a relatively complicated rule, to limit what a designer could do. We added more and more pieces as we thought of more and more outcomes, and we came to a point where it was so complicated—and it was still going to be hard to control, because the more rules you write the more loopholes you create – that we reverted to a simple principle. Limit the area very accurately, and make it a game of efficiency.”
C-class racing has already demonstrated that a three-element wing is faster than a two-element wing. That is, a three-element wing with two vertical slots accelerating and re-attaching flow over the leeward side. I take it as gospel that we will see three-element wings when AC72s hit the water.
But if a three-element wing is faster than a two-element wing, what about a four-element wing?
“One of beauties of a wing is you can set it up however you want it,” Burns says. “Our flaps [on Oracle Racing] are somewhat continuous, with an even curve up the back edge, but I know that some of the teams have been trialing discontinuous flaps, disconnected so they can set flaps at different angles, which I think is quite interesting. Depending on how many flaps you have on your 72, you can set up the wing a thousand different ways.”
But here’s the deal —
“Under the ACC rule for monohulls, the boats that we raced in Valencia in 2007 went through a whole iteration of complicated keels, and tandem keels, and we all came back to the simple configuration of a single fin with a couple of wings, and you couldn’t do better than that. We may see the same cycle again with wings and lots of clever multiple elements, but people probably will come back to simple solutions.”
Unless someone makes a leap beyond.
Last summer, when Jimmy Spithill, John Kostecki and company took me for a ride on San Francisco Bay on Oracle 4, I looked up and observed that the inside, concave surface of the wing—picture yourself looking up from the deck to the windward, inside surface of a sail—was formed by a grouping of convex surfaces. A soft sail, or the underside of a bird’s wing, would be a continuous, smooth surface.
Even to somebody with my Engineering-for-Dinosaurs eye, that cried out as an opportunity, however complex. Some time later, I found myself at Pier 80 in San Francisco. This is Oracle Racing headquarters until 2013, and I put the question to Burns and to Oracle Racing wing designer Scott Ferguson: Considering that bird wings don’t have convex surfaces on the lower, concave portion of their wings, aren’t we already seeing a built-in opportunity for performance enhancement?
The answer was yes, sorta.
Fair warning, we’re not coming to any conclusions here. No earthquake revelations. But for me it was a fascinating conversation as Burns went on to say—
“The C-class guys had a crack at making a wing out of a solid carbon skin rather than this 1930s airplane technology [meaning plastic film stretched over Nomex-cored carbon tubing, conceptually the same as canvas stretched over a wood frame] and probably a single skin is where we end up, some day. But for that we will need incredibly light laminates, materials not currently available. This plastic skin we’re stretching over frames, if I put my finger through it, we can fix with tape in a second. But with a carbon fiber skin it’s not so simple. Maybe not in this Cup, or maybe at the end of this Cup or the next, people will end up with a solid skin.”
Now let’s get Scott Ferguson, incidentally a Laser Master champion, in on this. Here’s Scott: “Achieving the perfect shape is hard using membrane-over-frame, but perhaps you can design around that problem by bowing frames to make the shape more of what you want. Then there’s the problem that you want minimal frames, because frames have weight, and meanwhile you have to have a strong leading edge because, if the leading edge goes soft, the wing loses shape. [AC45 wings have a Kevlar leading edge.] That level of strength becomes less critical moving aft. We’ll probably be seeing some attempts at single-skin wings, but the laminate skin we are using now is light at 60 grams per square meter. If you go with a [heavier] carbon fiber skin, it’s still going to show some dimpling, and you’re still going to have to have a core. At that point, with available materials, you’ve built a heavy wing.”
Meanwhile, materials technology is on the move. For example, here’s Ferguson again: “For its 3DL sails, North has started making super-thin-ply carbon fiber laminates. They’re incredibly thin tapes, and they create lighter laminations than you get from pre-preg.”
We take it for granted that weight aloft is a bad thing, but when you are designing an AC72 wing, keeping it light turns around and creates its own issues. And still, you have to do it. “The AC72 rule limits the wing by weight, and that’s a good thing,” Burns says. “As we designed the rule, there was a lot of work done on weights by different groups, and they all came back with similar numbers. You can’t do anything too complicated.”
The designers are up against a total-weight limit for an AC72. If the wing comes out heavy because there’s a lot going on, then the hulls have to come out lighter, and that’s not an easy trick. You really, really, really don’t want the hulls-structure to be fragile. “If you build a six-element wing [or a single-skin wing?] with a lot of fancy controls, you would struggle to not be overweight,” Burns said. “If you’re overweight in total, you don’t have an AC 72. You can’t sail.”
Compared to everything that went before, does that make the designers’ job easier or harder?
Ferguson says: “It’s just a different problem.”
And get these extra complications, per Burns: “The wing designer’s job is to create a big, light structure, and that is a good thing to achieve. Until you want to attach something to it. Then, if you’re trying to attach a shroud to a thin piece of carbon fiber, you suddenly have a complicated problem. In our 2007 monohulls, the masts were 6-8 millimeters thick, and you could just drill a hole and screw something in. By comparison, the forces were high, the solutions were brute-force, and almost any component failure would bring the whole rig down. Now everything is very light. Attaching anything is a challenge, but the beauty is that very few of the components are individually critical to keeping the wing upright.”
And as I run out of quotes, I realize that I left a rather large question on the table. If a three-element (two slot) wing is faster than a two-element wing, why would a one-element wing be faster yet?
By doggies, I think we’ll have to come back to this.