A Brief Overview of the Design Journey Over 25 Years of Developing RC Racing Yachts
I’ve referred to RC yachts as my first love. They certainly were an early way for me to learn about design and engineering through hands-on experimentation. They taught me to test ideas in competition and learn their validity from observation. They are a microcosm of rapid iteration reflecting the ‘fail fast’ approach so championed by modern start-ups.
Though I worked (and had some commercial and racing success) in International One Metres and 10 Raters, I always predilected the Marblehead (AKA M) Class. The simplicity of the M Class concept and rule has allowed it to endure since 1932. Until the recent maturation of CFD, it was a relevant way to experiment at scale and was used to pioneer some significant features now commonplace in ‘full size’ yachts.
Below is a sampling of key designs (with descriptions of the thinking they embody) representing milestones in my development journey.
They are presented in chronological order (oldest first) stretching back to the mid-1990s!
Maverick - 1995
Heavily influenced by what I was seeing in the IMOCA 60 world, my initial approach was to reduce ballast mass by seeking righting moment through a wide ‘skimming dish’ hull.
Part of validating the concept was to build a control hull with minimal wetted area (nicknamed ‘Goose’, pictured on the right below). The control boat had semi-circular sections and some overhangs to really stake out that extreme of the design space.
The contrast was significant: Both boats achieved the same righting moment but with 2.6Kg and 3.8Kg of ballast respectively!
Structured experimentation also extended to construction techniques. Both designs were prototyped in strip-planked balsa with a light fibreglass skin. The prototype hulls were then used as plugs to create moulds. From the moulds I made a a series of composite hulls with various layups and joining techniques. Including iterations using then-exotic Kevlar honeycomb cores.
There was a lot of learning! The principal lesson being that the nature of the rule has far-reaching effects. Specifically, the M Class restricts overall length AND maximum sail area. This means that one side of the sail area to wetted area ratio is fixed. Thus wetted area is at a premium far more than is the case in the IMOCA (where sail area can be added to compensate for increased wetted area).
Coupled with the structural weight saving of a narrow hull and the practical advantage of higher displacement when manoeuvring (greater inertia), exacerbated by the small scale (low Rn), the narrow boat was measurably superior. Thus the initial vision of little Open 60s skimming around the pond was proven to be flawed.
Development then shifted to optimising the narrow ‘piercing’ concept.
Nexus - 1998
Part of the exercise for me was proving I could carry out the relevant design calculations independently. Down to drawing out the sections and applying Simposn’s Rule to measure the volume and centres. This gave me a fundamental understanding and feel for the key design parameters.
Taking the lessons learned with Maverick (and looking at the exceedingly elegant 1995 generation of IACC yachts with their narrow beam and slab sides), the next logical step was to take the minimal wetted area concept to the extreme.
To do this I came up with a simple process to define the canoe body: Set a waterline length, displacement, and prismatic coefficient (how full the ends are). Then find the semicircle with area such that it would give the desired displacement at my postulated length and prismatic coefficient…
The radius of that semicircle determined the half-beam at the waterline and canoe body draught!
I made a template of that semicircle out of 3mm MDF sheet, set it in a block of foam, and sculpted the hull from there. The maximum cross-section was set at 55% length so the keel fin quarter-chord line could be placed there with the mast exactly amidships. The topsides were vertical, joining the semicircle template as a tangent at the waterline.
The slab topsides were carried forward to a high-freeboard but ‘piercing’ bow (incidentally giving a classically handsome looking sheerline). The idea being that resistance to nosediving is mainly driven by the stern sections (which were dishy and rose reasonably steeply for both energy recovery upwind and pulling the stern down when running in excess of hull speed).
Once I was happy with the sculpted shape, I float-tested it to make sure displacement was correct (given I’d sculpted the prismatic coefficient by eye) and then proceeded to digitise it and fair it on an early version of MaxSurf.
Next I made a plug (pattern) and experimented with a male-moulding technique where the hull shell is laid over the plug and wrapped around the gunnels. The shell being flexible enough when cured to be sprung off the plug. This actually worked well but proved too labour-intensive for production as the final surface finish was heavily dependent on very careful application of peel ply to the wet-layup skin.
Though the build process reverted to female moulding, by retaining the rolled gunnels I gained sufficient stiffness to delete all core materials. The deck consisting of a single moulding around the mast/keel that incorporated the RC and battery tray. Fore and aft of this moulding, only sticky back cloth kept water out of the hull.
Before practical 3D printing and widely available CNC machining, minimising part count was absolutely critical. Nexus could be built from its tooling set with virtually no metal fittings required. The sheeting system was internal, giving a very clean deck, and the access hatch for the batteries was the lid from a Nutella jar (the hatch collar being moulded in carbon over the glass jar itself).
This was also a time of great experimentation and progress in rigs and aerodynamics. Having committed to ‘swing rigs’ for their simplicity and ability to always project maximum sail area when reaching and running, I developed a system of ‘wingmasts’ (or, more accurately, rotating streamlined masts) stayed to outriggers mounted to the boom junction block. Once the weight penalty was minimised (by improving lamination techniques and using higher modulus carbon), the aerodynamic advantage proved significant.
To the highly streamlined swing rig was added a ‘gizmo’ (the original term coined by legendary designer and racer Graham Bantock) that used a series of levers and purchases to flatten the rig automatically when sheeted hard in.
The picture below right shows a further experiment with canting the rig to windward. This was a way to gain righting moment, analogous to canting the keel, but with actuation that could be lighter and less complex. Though this showed promise in terms of straight line speed, the handling compromises involved made it dubious as an obvious line to continue to pursue. Canting the rig moved the centre of effort which changed the helm balance and made tacking very difficult. More significantly, realising the gain required losing over 1Kg of ballast which further affected manoeuvrability.
Nexus proved competitive, was sold in reasonable quantities, and obtained some good results.
It took fully four years to gain enough knowledge to warrant a fresh design.
Ajax - 2002
Nexus was an exercise in extremes. Minimal wetted area, absolute minimal part count – basically the smallest simplest boat under the lowest-drag possible rig.
Over years of racing under the control of different skippers in venues all over the world, some limitations began to emerge consistently. These were mainly to do with handling. The long straight bow sections and relatively aft centre of gravity made the boat tricky to turn in waves. And limited volume made it vulnerable to burying if pressed hard when conditions got very fresh.
Ajax attempted to address this by slightly increasing above-water volume and moving the centre of gravity forward to the middle of the hull. Advances in modelling software meant this could all be achieved with very little additional wetted area. Adding some curvature to the topsides also improved structural stiffness, thus offsetting some of the weight gain resulting from the boat having grown a bit bigger.
Perhaps the most significant advance was in the appendages. A better understanding of laminar flow allowing optimised sections and improved manufacturing techniques making the keel strut significantly stiffer. This was the beginning of longer-chord fins with sub-6% thickness to chord ratios. Note also the LERX (Leading Edge Root Extension) visible in the middle pic below.
As can be seen in the pic at left below (number 18 streaking away from the pack is an early Ajax), Ajax was built with the option of conventional rigs for the smaller suits. This also makes handling more forgiving when pressed because the jib helps to lift the bow and the booms have more clearance over the water.
Ajax was the first of my designs to be built commercially by a third party. It went on to win various events at state and national level as well as placing respectably at international regattas.
Though I personally had less time to devote to developing RC yachts during this period (I did regularly sail my IOM design Light Ice in Auckland while I lived there as an America’s Cup team member), Ajax was actively developed, based on feedback from the builder and customers, for a good six years – longevity being an indication of the soundness of a design!
Octave - 2008
Octave was a foray into some radical extremes of the design space. It was intended to answer specific questions about wave-piercing hulls, unconventional appendages, and really pushing the limits of structural weight.
I’ve included it here as an example of interim designs that, though not successful on the race track, served as test labs for the following generation of design thinking.
Two boats were built from temporary tooling and were campaigned with very specific intent and clear lines of communication in place.
Though some of the compromises were known at the outset, quantifying how such an extreme boat actually behaves was very valuable (including some surprising areas of promise that were not obvious going in).
The boats were also used to test fittings that increasingly took advantage of commercially available CNC machining and 3D printing.
Katana - 2020
Katana incorporates all the lessons learned from my previous M designs but also learnings from other applications of aeroelastics, prepreg manufacturing techniques (including spread-tow fibres for the appendages), and process optimisation.
The hull concept can be traced back to Nexus but attempts to subtly improve righting moment, handling, manoeuvrability, and response to waves. The rocker line dips in the forefoot, pulling volume under the bow whilst the diagonals aft are very even to encourage the flow to stay attached…
The bulb is area-ruled and made of bronze (an interesting compromise trading some density for stiffness and better surface finish). This allows an extreme slenderness ratio, thin beaver tail, and optional machined wings…
Perhaps the most visible unique feature is the peaked deck (inspired by MarGo which, together with Paradox, is one of my favourite designs of all time). The peaked deck helps to seal the rig aerodynamically, shed water, and improve structural efficiency.
Katana displays the influence of very powerful CAD tools in optimising surfaces, as well as machining and 3D printing as manufacturing options. I can proudly say it represents the state of the art and incorporates decades worth of knowledge in the pursuit of excellence in this little niche sport.
IOMs and R10Rs
In closing, here are some images of my LightIce IOM and Rubicon R10R.
These were both in production at various times but never benefited from the time commitment I reserved for the Marblehead (though lessons learned in the M were still applied).
Now with the advent of the RG65 and increased popularity of multihull and foiling RC yachts, as well as the radical evolution of RC gear and batteries, there are some new design spaces to explore!