Above: the transom flap alternative favoured by designers Owen-Clarke was the Interceptor – employed on their Imoca designs for Mike Golding’s Ecover (here) and Dee Caffari’s Aviva. The Interceptor is a vertical carbon blade running the full width at the bottom of the transom that is raised or lowered into a narrow slot. Designed originally for Russian high-speed naval vessels, when it is lowered it locks in a wedge of water acting as a virtual trim tab. This is Ecover sailing at speed with the blade raised (top) showing displacement sailing flow off the stern and with the blade lowered and stern lifted (above left) with a clean wake more like a planing dinghy. Golding found an upper speed limit cut-off at around 17kt, at which point he needed to lift the bow again to go any faster – and maintain control. Right: examples of the numeric calculation and CFD modelling employed in the course of the ORC’s new research into transom effects
based on its definition in terms of these chosen parameters. Choosing the parameters is the key element to this AI process.
In understanding the aero factors the training set parameters need to be influential on sail performance. To illustrate with an analogy, imagine that we are using this approach to make a perfect-tasting pancake: we can see how the ratio of eggs to flour to water and the pan temperature and cooking time are likely to affect the taste, and if we made 1,000 random pancakes we should identify the best pancake recipe. If, however, we decided that the important parameters were the colour of the mixing bowl, or whether or not we use an electric mixer we wouldn’t learn much. And so it is with ‘parametric’ force models: if you don’t characterise your training set with sensible parameters no amount of AI can rescue you. Using this approach we now have an improved aero model, and
that helps not only because we can predict effects like heeling force and thrust more accurately, but also how this affects other forces. For example, one of the vexing questions we have had since adoption of this method has been why does the VPP tend to overestimate the benefit of stability? There’s no question that having more righting moment helps performance, why else would people hike? Yet depow- ering is also beneficial: as the wind strengthens to keep a boat on its feet you have to depower the sails, reduce the heeling force, and get the centre of effort lower. Aerodynamic theory and wind tunnel tests both show that as you
depower with increased windspeed the overall sailplan efficiency is reduced because the effective span is reduced. Therefore a stiff boat enters this zone of reduced efficiency less quickly as the wind rises. Turns out that we have been over-estimating this loss of efficiency a little bit: our AI analysis of the virtual wind tunnel data showed that there are sail trimming strategies that can mitigate the loss of efficiency that our limited wind tunnel test database couldn’t identify. When that was introduced into the VPP the bias towards favouring stability was reduced. Now we have moved our force model a step closer to reality and
the predictions all work better. Talking of reality, it is entirely possible that the top sail trimmers have reached the same conclusion as the AI thanks to their years of trimming experience and feedback from skilled helmsmen – it’s come full circle. This just goes to show that a degree in engineering is not required when trimming sails. On the hydrodynamic side we are engaged in a magnum opus,
a world that is not as comfortable as easily defined sail shapes and an airflow that is undisturbed by the water. Now to be accurate the VPP must also deal with hull drag which is in two parts: the hull
34 SEAHORSE
friction, which for now is under control because there is a wealth of published data about which coefficients to use and, knowing the wetted surface area and the boat speed, it’s an easy calculation. The other component, however, is wave-making (or residuary resis-
tance, so called because it’s the bit left over after you have taken away the friction resistance) – and this is more problematic. It depends not only on how heavy the boat is but also how long and wide it is, how the hull volume is distributed towards the ends of the boat, and crucially how much transom area is immersed in the hull wave system… this is a distinct feature on most modern fast designs. There is no doubt that moving from 20 tank tests analysed with
a slide rule to 1,000 tests processed with a modern computer’s neural network has made a big improvement. And this means it’s no longer as easy for the yacht designer to look at the rule and decide which parameters he can exploit for a favourable handicap. But as hull shapes have moved towards shorter aft overhangs
this ‘transom drag’ is a factor that must be tackled. What’s needed is a characteristic ‘wave-making length’ that is sensitive to where the aft end of the boat is and a method to capture the drag of the recirculating flow behind the transom. In the course of 300 CFD runs, undertaken by Jason Ker and his
associate Marcus Mauleverer, the position of the running waterline along the hull was captured and analysed to build an algorithm that can predict this running waterline position around any hull. This in turn means that the position of the waterline at the transom can be determined. Also the wave pattern aft of the transom can be calculated and used to define an aft end of the effective waterline. Knowing the wave pattern and the immersed depth of the
transom, another algorithm was devised to predict how the water behind the transom behaves. For a boat with an immersed transom at low speeds the water recirculates behind the transom, but as speed increases the wetted transom area reduces and finally at top speeds the transom runs clean. Given this effect it’s crucial to calculate the extent of transom wetting and the associated pressures on the transom – because when it is wet the water is ‘leaning’ on it and thus pushing the boat forwards. Said quickly this all sounds do-able, but spare a thought for the
poor engineer who sits bleeding from the ears to try to make this work. We’re fortunate to have such dedicated experts on our ORC team, and this pause in racing for a few months may very well help us get more out of them this year on VPP progress than in otherwise frantic times. Andy Claughton, International Technical Committee
q
JAMES BOYD
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72 |
Page 73 |
Page 74 |
Page 75 |
Page 76 |
Page 77 |
Page 78 |
Page 79 |
Page 80 |
Page 81 |
Page 82 |
Page 83 |
Page 84 |
Page 85 |
Page 86 |
Page 87 |
Page 88 |
Page 89 |
Page 90 |
Page 91 |
Page 92 |
Page 93 |
Page 94 |
Page 95 |
Page 96 |
Page 97 |
Page 98 |
Page 99 |
Page 100 |
Page 101 |
Page 102 |
Page 103 |
Page 104 |
Page 105 |
Page 106 |
Page 107 |
Page 108 |
Page 109 |
Page 110 |
Page 111 |
Page 112 |
Page 113 |
Page 114 |
Page 115 |
Page 116 |
Page 117 |
Page 118
