AC34: The Anatomy of an AC72
Secrecy is as much a part of the America’s Cup as the “Auld Mug” itself, and AC34 has been no exception. That said, it’s hard to hide what you’re up to aboard a full-foiling catamaran; no more hiding your underwater appendages behind a skirt as you take the boat in and out of the water.
Here, we take a closer look at the AC72, in an effort to demystify this amazing boat. Use the numbers on the image and compare them to the numbered design elements listed below to learn what, exactly, makes this boat fly.
1. PEDESTALS shift between pumping oil to the hydraulics and driving winches via mechanical linkages
2. MULTI-SPEED WINCHES with a half-dozen or more “gears” allow trimmers to sheet in fast during maneuverers and then fine-tune trim under load
3. BACKSTAYS are critical to providing forestay tension for jibs and luff tension for reaching sails
4. CROSSBEAMS alone support rig and hull loads aboard Oracle: the configuration is prone to twisting, or “racking,” under high loads but also lowers windage
5. WING POD extending beneath Oracle’s crossbeams restricts airflow under the wing and may produce additional lift sailing to windward
6. DIGITAL PERFORMANCE INDICATORS are displayed on the wing for easy viewing by trimmers and helmsmen
7. SOFT HEADSAILS are cut and built to routinely operate in gale-force apparent winds
8. WING-CONTROL CABLES run through a master-control quadrant at the base of the wing
9. THREE SEPARATE "COCKPITS" line each hull aboard Oracle, but there is a single, continuous cockpit in the hulls of the challengers
10. BUTTONS ON WHEELS are used to fine-tune daggerboard trim
11. FIXED WINGLETS on rudders provide a balance point when full foiling
12. PDAS allow trimmers and tacticians to see performance data at a glance
13. DOZENS OF OPTICAL SENSORS in the hulls, daggerboards and rig measure loads on different points of sail in different conditions—critical to design and training, but not as vital when racing
14. LIFTING foils, or winglets, on the daggerboards are angled upward to help stabilize the boats when full-foiling
15. DAGGERBOARDS are hydraulically canted and/or raked to achieve optimal lift, in contrast to foiling Moths aboard which the winglets themselves are adjustable: although it was meant to prevent foiling entirely, the rule forbidding adjustable winglets has only served to make foiling more awkward and dangerous
16. WINDWARD DAGGERBOARDS must be fully retracted at all times—one of a number of regulations that were supposed to prevent full foiling
17. AERODYNAMIC FAIRINGS on crossbeams and other structures reduce windage, which is critical aboard boats going this fast
18. HULLS need to be slim for straight-line speed, but include enough volume and buoyancy forward to prevent the wave-piercing bows from submarining, especially during bearaways
The wings in AC34 probably won’t change too much from AC33, although the first Artemis wing was a bit different. Wing design science has been developing for over 100 years, and the most important methods for calculating flow around an airfoil and drag due to lift—also called induced drag—were discovered before the 1920s.
All AC72 wings have two large aerofoil elements separated by a small gap and a smaller third element attached to the trailing edge of the first one, like a trimtab on a keel. (Note: the multiple large flaps along the wing’s trailing edge comprise a single element and serve to provide twist.) The smaller third element is only activated when the trimmers want to increase camber for maximum lift in light air and downwind. AC45s use a two-element wing because it provides most of the benefit of a wing, but is simpler to build.
Emirates Team New Zealand’s wing #1 has a twisting leading element as well as twisting flaps, whereas Oracle Wing #1 has a rigid leading element. The advantage of leading-element twist comes when sailing downwind, because there is more twist in the apparent wind angle (AWA) than upwind. It is for this reason that the feature has long been essential aboard C-class cats, which don’t have the benefit of headsails to provide a header to the wing. That having been said, AC72s are much faster downwind with less AWA twist, so it is unlikely to provide much benefit, especially when foiling.
ETNZ’s leading element has a straight taper, while Oracle’s is “curved,” as is evident in the “kinks” in the wing’s leading and trailing edges as it narrows toward the top. Curved edges more closely follow the ideal lift distribution, but straight-taper molds can be cheaper and faster to build. Twist is also easier to control with a straight hinge line, and a straight hinge line makes it easier to seal gaps between the fore and aft elements.
Ultimately, the different wings will probably have quite similar performance upwind; although when full-foiling downwind, the ability to twist the leading edge may be more important as the gennakers become smaller or even non-existent in heavier air.
Artemis’s first wing stood apart with a number of horizontal gaps between the flaps. It seems the team valued controlling the center of effort over the losses that come from air leaking through the gaps. They have since returned to conventional twisted flaps, so perhaps the gaps added too much drag.
The first thing you notice when sailing with a wing is that it’s hard to judge twist and depth, because unlike the concave windward surface of a conventional sail, the windward side of each wing element is convex. There are two other reasons: 1) The nose of the fat leading element is not visible because it is obscured by the rest of the structure, and 2) the centerline of the nose at deck level moves to windward as the leading element is rotated, because the rotation point of the wing is aft of the nose. Therefore wing trimmers rely heavily on instrumentation to repeat the settings for camber and twist. The art of trimming still exists, but it is now much more focused on numbers and boat performance than keeping a stretchy sail and bendy mast in shape.
Unfortunately for designers and sailors, measuring wing shape is not as easy as it first appears. Although all the teams likely have angle sensors on the flaps and leading elements, these wings are so large and lightly built that they deflect under their own weight in the shed, which makes calibrating the sensors a challenge.
All teams will have also made an intensive effort to both measure and display the data in an easily readable form. Look for the trimmers and tacticians, in particular, to have small, wireless personal data assistants (PDAs) strapped to their wrists. The speed of these boats means that something as simple as the way the performance data is formatted can have a noticeable effect on the trimmers’ ability to keep the wing near its optimum shape.
The flaps are controlled by hydraulically powered cables running from a master control quadrant at the base of the wing up to each hinge line. The length of the cables will be a source of additional calibration errors due to stretch. Among the advantages of the cable system, which was first developed by Dave Hubbard in the 1970s, is the fact that the settings are replicated when tacking. I don’t think ETNZ, Luna Rossa or Oracle have any hydraulics up the wing. Not so sure about Artemis—early on they mentioned 38 hydraulic cylinders in wing #1, but a self-tacking system would be difficult.
The most dramatic advance in this Cup so far is the stable foiling configuration pioneered by Emirates Team New Zealand, which includes an upward angle on the winglet of the daggerboard, like a tick, or a checkmark. Due to this upward angle, the vertical lift is coupled with a side force. As the boat rises there is less daggerboard area in the water, so the leeway angle increases. This greater leeway angle, in turn, reduces the vertical lift generated by the winglet, so the boat lowers until the forces balance again.
Upwind boatspeed is limited by the righting moment (RM) to around 20 knots, and this RM is the same for all the yachts due to the class rule restricting beam and weight. Because the AC72s are so sleek, the reduction in hull drag from foiling won’t be greater than the drag of the foil itself until the boats are sailing in the high 20s, at which point full foiling becomes faster. In other words, no foiling when sailing to windward.
ETNZ, Luna Rossa and Artemis have all used a combination of a Y-span type structure and trusses to support their boats’ loads. Oracle relies on unsupported cross beams.
Because catamarans have wide platforms with small hulls, torsional deflections are much greater than on monohulls. The windward shrouds are also led well aft in the boats to provide forestay tension. The torsion comes from that offset load.
The Y-type being used by the three challengers is similar to the one used by the Alinghi 5 catamaran in AC33. This type of structure is very stiff in torsion because the loads are transferred by members in pure tension or compression. It also follows the shortest path between the load points.
Oracle, on the other hand, carries the torsion by twisting its crossbeams, and the loads must follow a longer path from the chainplate to the mast step via the hull and then along the beam. Materials in torsion are less stiff for the same weight as pure tension or compression, which means the structures have to be made larger and/or heavier. However, Oracle still comes out ahead in terms of drag because the beams are streamlined and there is less rigging, which is difficult to streamline effectively.
Aerodynamic drag is not just turbulence from the under-rigging but also from the losses under the wing. Team NZ and the other challengers have a large gap between the wing and water, whereas Oracle’s centerline pod carries the pressure closer to the water. This will make a much bigger difference in light airs and upwind. Full foiling will diminish the advantage of the pod because the gap to the water is increased. It’s interesting to see that Oracle 2 has a reduced pod size compared to #1—probably to spend the weight elsewhere.
In many ways, hull form is less crucial in AC34 than things like foil shape, control systems, etc. However, hull design is still important, albeit with less emphasis on pure drag and more emphasis on stability. In a slow-speed bearaway, for example, the foils do not contribute much lift, so the resistance to pitchpoling comes from volume in the bow. A full bow will have more drag upwind, but perhaps not being able to defend in a prestart will be more decisive?
Now that stable(-ish) foiling has arrived, hull choices can be slanted to speeds below 20 knots. Certainly above 30 knots the hulls should be clear of the water on foils.
Return to CupWatch coverage here
Photos courtesy of Gullian Grenier/AC34 and Nigel Marpel/Luna Rossa and Chris Cameron/ETNZ