Australia II Wing Keel Controversy – Part 1

Daniel Forster

Revelers wanted to touch the wing keel of Australia II following the 12-Meter yacht’s victory over the U.S.
America’s Cup defender, Liberty, in 1983. The controversial appendage was credited with making the performance difference that ended the trophy’s 132-year possession by the United States.

[F]ew sporting events have led to more publicity, debate, and discussion than the 1983 edition of the America’s Cup, when Australia II, representing the Royal Perth Yacht Club, ended the 132-year possession of the “Auld Mug” by the New York Yacht Club (NYYC); a winning streak started when owners of the schooner America won the R.Y.S. 100 Sovereign Cup (as it was then called) by beating all other yachts in a race around the Isle of Wight in 1851. The owners of the victorious yacht gave the Cup to the New York Yacht Club in 1857 under the terms of a deed of gift that specified it was to be preserved as a perpetual challenge cup for friendly competition between countries. Despite the predictability of the American winner in the following decades, the boating press always closely covered the competition for the Cup, with yacht design, refinement, and building of particular interest to naval architects. But in 1983 Australia’s upset victory fueled a frenzy of reporting and writing.

I have found some 20 books and innumerable newspaper and magazine articles about the events that took place the summer of 1983 in Newport, Rhode Island. Most of the debate focused on the legality of Australia II and whether the Australian team had contravened the nationality rules that the NYYC had imposed, one of which required that a participating yacht be designed by a national of the represented country.

Guy Gurney

Liberty and Australia II trade tacks off Newport, Rhode Island, during America’s Cup competition in September 1983. With tank-testing help from Dutch naval architect Peter van Oossanen, the Australian 12-Meter design had been optimized for the prevailing moderate wind conditions.

The legality question was focused on the feature that as Australia II heeled, the maximum draft as defined in the “International Twelve Metre Class Rating Rule and Measurement Instructions” was exceeded by the winglets attached to the bottom of the keel. This became a nonissue on August 10, 1983, when the chairman of the measurement committee of the International Yacht Racing Union (IYRU), Tony Watts—then guardian of the Rule and the organization responsible for the measurement of participating yachts—announced that the measurement committee unanimously reaffirmed the legality of Australia II’s keel. On August 20 Peter de Savary, chairman of the British Syndicate, announced a confidential ruling from the IYRU to the effect that winglets are permitted on 12-Meter keels if the winglets are fixed and not capable of being adjusted in trim or retracted. Subsequently, the NYYC halted efforts to have Australia II disqualified on the basis that the winglets were a “peculiarity” not permitted under the Rule.

The nationality issue surfaced on August 24. It wasn’t resolved, and the America’s Cup Committee of the NYYC met to decide whether they should cancel the event just before the first race between Liberty, the U.S. defender, and Australia II. It was reported that three members were for abandoning the event, while four were for pressing on—primarily because of the loss of face the abandonment would cause.

Peter van Oossanen

Comparison of body plans for the original 12-Meter Australia (as raced in 1980), above, and Australia II,
below, reveals differences in the hull shape as well as the dramatic addition of winglets to a new keel.

Peter van Oossanen

The legality and nationality issues had been raised by the NYYC or by people on its behalf. This was widely seen as unfair since the questions were raised only after it became clear that Australia II would become the yacht to challenge for the Cup. It had already convincingly beaten six challenging yachts, and the NYYC officials appeared concerned that Liberty didn’t possess the speed required to win.

The account below chronicles my involvement in the design of Australia II. It is excerpted from Chapter 16, Volume 5 of my work The Science of Sailing (, which includes a more detailed account of the Australia II project and the Dutch contribution to it. Summarized here is what I was responsible for in 1981, when Australia II was designed and its performance evaluated (see the sidebar “Winged Victory,” by Dan Spurr in Professional BoatBuilder No. 121, page 52).

Guy Gurney

Australia II’s winged keel was unsuccessfully challenged by the New York Yacht Club on grounds that it was a “peculiarity” not permitted under the 12-Meter rule. Note the trim tab on the keel’s trailing edge.

The project was carried out at the Netherlands Ship Model Basin (NSMB), now called the Maritime Research Institute Netherlands (MARIN), by my staff and me, except for calculations performed by Joop Slooff, an aerodynamicist from the National Aerospace Center (NLR) in Amsterdam. NLR was subcontracted by NSMB to make comparative lift and drag calculations for different keel arrangements.

Although the events took place 40 years ago, I had kept meticulously detailed records in diary notes of everything that happened, and I retained copies of correspondence (letters and telex messages) and other relevant documents.

Bob Fisher Archive/PPL

The client, Alan Bond, was an English-born Australian businessman who bankrolled four America’s Cup challenges, culminating in the successful 1983 campaign.

Quest for a Client

The lead-up to the project starts with my conviction that a 12-Meter class yacht, intended for racing for the America’s Cup in Newport’s prevailing wind conditions from July through September, should be given the minimum length on the flotation waterline in measurement condition (referred to as the MWL waterline in the Class Rule, the length of which is denoted by LWL) allowed by the America’s Cup Deed of Gift (44 /13.41m)—taking advantage of the greater sail area and lesser displacement that accompany this length. This strategic conclusion was the result of a study of the attainable speeds of 12-Meter class yachts utilizing my first velocity prediction program (VPP), developed in 1977–78 as a hobby project. It was essential that by adopting this minimum waterline length, the yacht’s ability to withstand heel must not suffer. That was difficult to attain, because when adopting the minimum displacement associated with the 44 waterline, the amount of ballast that can be placed in the keel is less than in a yacht with a longer waterline. I realized that this limitation required another type of keel—one in which the lead ballast could be located lower.

I met Ben Lexcen, the Australian sailor and naval architect, and Alan Bond, an Australian businessman and head of the Australia II Syndicate, for the first time at the Sebel Townhouse Hotel in Sydney on January 19, 1978. I had written to Bond two weeks earlier, introducing myself and explaining that I had just completed a study of how to optimize the design of 12-Meter class yachts for the Newport, Rhode Island, summer wind conditions. I hinted at the possibility of improving the performance of Australia, Bond’s 12-Meter, which had competed in the America’s Cup four months earlier. I also pointed out I believed in using large-scale models to accurately determine performance differences between designs, explaining that I worked at the NSMB, had grown up in Sydney, and was keenly interested in yacht design.

My motives for approaching Bond were: first, a longing to become involved in experimental sailing yacht research; and second, to bring to the Australian effort the knowledge and capabilities of one of the foremost hydrodynamic research institutes in the world.

Daniel Forster

Bond built his 1983 effort on what he had learned in the previous three Cup series, the last two sailed in the 12-Meter Australia, seen here trailing the successful U.S. defender, Courageous, in 1977.

The first thing Bond wanted to know was the result of my analysis of Australia’s loss against Courageous in the 1977 America’s Cup. I explained that the keel needed improvement because Australia lost ground upwind against Courageous. Downwind, Australia’s performance was fine. He agreed, becoming very amicable. Explaining more fully who I was, what tank-testing at NSMB would entail—and that the secret of model-testing when small speed differences between designs are important is the use of large models—I proposed that we adopt 1⁄3-scale models, which had never been used before.

Bond asked me about the costs of model-testing, and I showed him an example test program I had readied. He also asked me if I was prepared to work with Lexcen on the project. I said I would be happy to work with him or anyone else. We then discussed in detail what we could bring to his effort in terms of experimental testing and design.

The three of us met again the following morning, at Bond’s suggestion, “to finalize arrangements.” I was again quizzed about how to improve Australia and how I would design a new yacht. I described the results of my mathematical study and pinpointed that, for sailing in Newport, the yacht must have the shortest possible waterline (44 ), the associated smaller displacement and larger sail area, and some means to improve stability beyond the value normally associated with a 12-Meter 1 to 2 (305mm to 610mm) shorter (on the MWL waterline) than recently designed America’s Cup yachts.

Bond said he was considering selling Australia to Keith Turner, a business associate in Perth who was also interested in competing for the Cup. That would allow him to build a new yacht, he said. He wanted to trial Australia against the new yacht to allow fine-tuning and test different sails. He would agree with Turner to take only the fastest yacht to Newport. Finally, Bond said he would be ready in three to four weeks to discuss details. He asked Lexcen and me to further discuss design during the next few days before I returned to The Netherlands.

Bob Fisher/PPL

Ben Lexcen was the Australian naval architect responsible for designing all Bond’s 12-Meters, starting with
Southern Cross for the 1974 America’s Cup and ending with Australia IV in 1987.

I didn’t know what to make of this situation. I appreciated that Bond had taken me into his confidence and explained the Turner deal. On the other hand, I was disappointed that the meetings had not led to a firmer commitment. I explained that I was intent on getting my concept accepted one way or another, indicating that if the arrangements with Bond were to fall through, I would approach other individuals or organizations.

At Bond’s suggestion, I accepted Lexcen’s invitation to visit his home in Seaforth the following weekend. We drove to the Royal Prince Alfred Yacht Club to take a close look at the 5.5-Meter class yachts there and discussed many details of yacht design. We talked about endplates on keels and rudders in great detail, and I remember outlining the possibility of fitting endplates to the top and bottom of the trim tab to reduce the component of drag associated with side force, which we refer to as induced drag. We agreed it was a topic that required further investigation.

On February 22, one month later, I still hadn’t heard from Lexcen or Bond. I wrote a second letter to the latter, asking his plans for the 1980 America’s Cup. A few weeks later, with still no answer and my telephone messages to no avail, I decided to approach other syndicates. I believed my proposal for the design and tank-testing of large models for the America’s Cup challengers would find a positive reception somewhere, as this approach had not been tried before.

At that time, most sailing yacht research and design programs were carried out in small towing tanks with small models, a practice that had already led to disaster in the case of the Mariner project for the 1974 Cup. In that instance, the lack of proper simulation of the full-scale boundary layer on the model had led Britton Chance, the designer, and the staff of the Davidson Laboratory of the Stevens Institute of Technology in Hoboken, New Jersey, to truncate the underwater aft body of the hull, believing that this led to less drag and therefore greater speed. This was not the case, and when the yacht was built accordingly, it was found to be off the pace. Thereafter, model-testing fell into disrepute, particularly in the U.S., and Independence, the Ted Hood–designed 12-Meter for the 1977 Cup, was not tank-tested at all. Rather than spend money on testing models in a towing tank, Hood spent time and money on full-scale research. He was quoted as saying, “Since 1967 and Olin Stephens’s Intrepid, we seem to have gone backward in 12-Meter design,” and, “I don’t think we know enough about tank-testing, and those who have done it have gotten into trouble.”

My offers to various other America’s Cup organizations were either not answered or were turned down. I now know that when an organization has been put in place to design a yacht for the Cup, the designers rarely acknowledge the possible benefit of an idea or assistance offered by outsiders. To protect their position in the organization, they inevitably turn down the idea or offer of assistance, unless it is of such a nature that in turning it down, the opposition would be bound to benefit. I have since learned it is better to deal with the syndicate’s management, who are more likely to view an offer of technical assistance differently. But by approaching designers in 1978 and offering them the assistance of a large towing-tank organization like NSMB, then the largest non-navy-owned towing tank foundation in the world with more than 450 people on staff, I was doomed to fail. By mid-1978, I was no further than when I spoke with Bond and Lexcen in Sydney in January.

I received a letter dated November 2, 1978, from Warren Jones, executive director of Bond’s America’s Cup Challenge ’80 Ltd. Three weeks earlier I had telexed Bond asking him to meet while I was in Perth on November 15 and 16. The reply I had received from his secretary on November 1 stated that Bond was overseas and not available on those days. In his letter, Jones thanked me for my offer to help design a new yacht and to carry out the tank-testing work. He explained that Bond had been extremely busy with his business, and the 12-Meter campaign we had discussed had been deferred. He also wrote, “In view of the lateness of our decision, we will be campaigning Australia after carrying out minor modifications.” On November 7, 1978, I called Lexcen, who confirmed he would only be modifying Australia for the 1980 campaign. There were insufficient funds available to build a new yacht.

In view of the disappointing reactions from those I had approached about my 12-Meter design ideas, I decided to take results of my study public. I presented a paper at the Fourth Chesapeake Sailing Yacht Symposium in Annapolis, Maryland, in January 1979, demonstrating how the performance of 12-Meter class yachts could be further optimized. I revealed some results of my calculations to support the main conclusion of this study, which was then contrary to the prevailing opinion that the optimum flotation waterline length of 12-Meter class yachts for Newport was between 45 and 46 (13.7m and 14m). The leading 12-Meter designer of the day, Olin Stephens, questioned me about my conclusions after my presentation, stating that they were at odds with the practice found on the water. I pointed out that the calculations had shown that a prerequisite for the 44 waterline length is that the yacht be given a greater ability to withstand heel than had been the case until then. My inquisitor replied that within the bounds set by the value of the waterline beam, as required by hydrodynamic considerations, that would not be possible. He obviously hadn’t thought of the possibility of adopting another type of keel, allowing for positioning the lead ballast differently.

The managing director of NSMB at the time, Dick van Manen, supported me in every way in attracting an America’s Cup team to carry out research at NSMB. He and Rinus Oosterveld, head of research and development and my immediate superior, knew that we’d need to set aside significant resources to update the NSMB test setup for towing models at varying angles of heel and leeway, for analyses of the resulting test data, and for extrapolation of those data to full-scale to derive performance predictions for the actual yacht.


Then, out of the blue, Lexcen called me on January 31, 1981. He was in Amsterdam and proposed driving to Wageningen to see the NSMB facilities. When we walked through the seven laboratories, I could tell that he was impressed. He told me Bond was now convinced that to continue his quest for the America’s Cup, he needed to build a new 12-Meter. He asked me to prepare a formal proposal that he would forward to Warren Jones. Due to travel abroad, I couldn’t draw up the proposal until February 17. I sent it by regular mail, not believing that it would lead to an order, but when Lexcen sent me a telex about not having received anything on February 20, I concluded that this time my hunt for an America’s Cup project was not a wild goose chase.

My proposal was based on testing a model of Courageous that Otto Scherer at Hydronautics Inc. in Laurel, Maryland, was prepared to lend us in exchange for the test results. The data would serve as a baseline to compare to the results of a new design. I had met Scherer when presenting in Annapolis in 1979 and again at a conference in Washington, D.C., in January 1981. We became friends. I telexed him on February 23, 1981, to confirm my wish to organize transport of the model to Wageningen as soon as we received an order from the Australians. Three days later he said he needed to get permission from the party that paid for the model—just in case. I received a letter, dated March 10, 1981, from Karl Kirkman at Hydronautics stating: “Regarding the Courageous model, our client is unwilling to allow the loan of this model. Based upon my discussions, I do not expect to see the model circulated [to various testing institutions] because of the obvious proprietary status of its configuration and the performance data.”

Lexcen ordered the work I had described in my telex proposal on March 3, 1981, on behalf of Jones. I informed him the next day of my plans to travel to Sydney to discuss the project. I met with Lexcen and Jones on March 13. Not having the Courageous model available as a “trial horse,” we agreed to build a 1⁄3-scale model of Australia as it raced for the Cup the previous year. Lexcen provided a hand-drawn lines plan of the yacht on a scale of 1:13.333 (9⁄10 equal to 1 )—the scale adopted by Olin Stephens for all his 12-Meter designs—which suggested that the respective lines were related to the Courageous lines plan that Johan Valentijn would have had access to when he worked for Sparkman & Stephens, before he worked together with Lexcen on the original Australia.

Peter van Oossanen

Complex instrumentation at the top of the tow mast at the Netherlands Ship Model Basin (NSMB) accommodated the free-running test method developed by the National Research Council Canada.

Back in Wageningen we computer-faired the lines Lexcen supplied, resulting in a set of lines on a 1⁄3 scale, which allowed us to manufacture the test model. I had Bonjean curves, hydrostatics, and stability data calculated from these computer-faired lines.

A formal order for the program discussed in Sydney on March 13 arrived by letter dated March 17. Jones gave us authority to modify the program as required, allowing the cost to vary accordingly. Payment was to be in three installments. He demanded “maximum security of all information compiled for us.” Finally, he asked that NSMB not perform any similar work for other syndicates associated with the 1983 challenge.

Peter van Oossanen

This NSMB test apparatus diagram indicates the adjustments and inputs possible for testing as well as the measurable performance indicators.

A meeting at NSMB about the towing method and the design of the associated dynamometer took place at my request on March 26, 1981, with experts from the towing tank at Delft University of Technology attending. We decided to adopt the so-called free-running test method developed by the National Research Council Canada, as described by David Murdey in 1978. This allowed for the greatest freedom of the model in adopting heel and running trim. It required towing the model at the center of effort of the resultant sail force and the setting of only leeway, trim tab, and rudder angles before each run. This distinguishes it from the so-called captive test method, in which the heel angle is also set before each run. Every run in the towing basin using the free-running test method simulates an actual sailing condition, requiring no more than about 20 to 25 runs in the tank to establish the model’s performance over a range of speeds (wind conditions) and apparent wind angles. A disadvantage of this method is that the model must be fitted with lead ballast, so it has the same ability to withstand heel as the actual yacht.

I took a few days off from my duties in Newport, and together with Charles Gommers, head of the instrumentation department at NSMB, I presented a paper on this towing method at the meeting of the American Towing Tank Conference held at the Davidson Laboratory of the Stevens Institute of Technology on August 2–4, 1983.

In Part 2 we’ll detail the tank-testing and the hull and keel refinement it in-formed for the designers of Australia II.

About the Author: Naval architect Peter van Oossanen was a principal scientist at the Maritime Research Institute Netherlands (MARIN) in 1969–1991. He wrote the definitive text on the resistance and propulsion of ships in 1988 and founded the Van Oossanen Group in 1992. Since retiring in 2013, he has written five books about the science of sailing.