Windage
The following is a bit of a rant, so I apologise in advance.
When designing proas, there is always the temptation to align the cabin along the transverse centreline, locating it between the akas. I often see cabin structures that are streamlined transversely, but which present large flat substantially vertical surfaces fore and aft. Many are modelled on the traditional Tepuke proa. I believe that this is a mistake.
Proas by their nature tend to have small sail plans and rely on efficiency rather than brute force for their speed. Therefore I think it is particularly important on a proa to reduce windage to the minimum. My belief is that wind pressure on a fore and aft facing surface can be somewhere in the order of 10 times more detrimental to performance than wind pressure on a similar sized side facing surface. My justification for this statement follows.
Wind pressure on a forward or aft facing surface is directly opposed by the trust driving the vessel forward, so the loss in thrust (drag) is equal to:
(wind pressure / frontal area) x drag coefficient
Wind pressure on a side facing surface is directly opposed by the lateral the vessel forward. This does not directly effect the drag, but it adds to the lateral thrust (lift) which must be provided by the hull or foils and these generate drag in the process of providing lift. So the drag due to side windage is equal to:
((wind pressure / side area) x drag coefficient) / (L/D of lateral resistance surfaces).
Typically the L/D of a lateral resistance foil may be in the order of 10. Note the drag coefficient mentioned can be reduced by raking the surfaces away from the wind direction (streamlining).
What this means in practice is that any forward or aft facing surfaces should be as small as possible and or raked aft as much as possible. However it is not nearly as important to streamline side facing surface. So if you must have a large flat surface anywhere (and you do need then to gain internal space) put them on the sides of the boat.
It might be mentioned that the drag is more accurately Velocity2 x drag coefficient, so it increases exponentially. More simply, twice the speed means four times the drag.
Malcolm, I believe this is a take on a subject that I’ve never heard before, and that’s saying something! 😉 I’ve often sketched up deck structures that are more streamlined from the sides than from the front… oops. Looking at Russell Brown’s proas, I see they follow your guidelines pretty well, with gently sloping surfaces from the front and the only real flats appearing on the sides of the pod.
I should also mention akas. It is the nature of beams that they are stiffer if they are deep (distance between top and bottom flanges) and they are lighter if they are narrow (width of flanges). However, a nice deep, narrow aka will present a large flat surface which will be beautifully aligned perpendicular to the direction of travel. Hence mechanically efficient akas will create a lot of wind drag.
Again it is advantageous to consider either streamlining the akas in the fore and aft direction and/or reducing the depth of the akas as much as possible. Both of these options require a compromise on the strength and/or weight of the structure. Recently I have been experimenting with designing akas using an open truss like structure, separating the top and bottom flanges, to reduce the frontal area. You can see an example of this on the Arcsail model, where both the aka and the sail boom (or beam) have used this type of structure.
I should also mention akas. It is the nature of beams that they are stiffer if they are deep (distance between top and bottom flanges) and they are lighter if they are narrow (width of flanges). However, a nice deep, narrow aka will present a large flat surface which will be beautifully aligned perpendicular to the direction of travel. Hence mechanically efficient akas will create a lot of wind drag.
I have been thinking a lot about this when designing my akas for my AS-Proa. My solution is box-beams that is standing on the corner of the box. This will have the added benefit of aligning the strength of the boxbeam with the loads they have to take. I think its quite rare that they get loaded straight up or down. The biggest loads are probably at 45 degrees upwards or downwards, when the ama hits a wave. One can align the flanges to take that kinds of load better that way.
Johannes.
Thanks for starting this thread, Mal.
I had been thinking about windage, but in a static way only. As in, if I stop sailing, what wind forces are acting on the hull that isn’t moving, and has minimal steerage? This happens every time a proa shunts. Does it make sense to minimize cross windage at the hull ends so that gusts don’t blow the resting proa off course as it moves out of the shunt? Or does having significant windage at the hull ends serve to stabilize the hull until it regains steerage? I suppose a schooner rig could be used to correct course at low speed, so maybe hull cross windage is moot?
It seems that Mbuli is a good example of designing for minimal drag is the direction of travel, while still using slab sides.
Cross windage might also influence whether a design uses a hard chine or a round-bottomed hull. The hard chine would contribute to lateral resistance, while distributing the force over the hull, whereas the foils needed by a round-buttomed hull would take the load in a small area.
John Shuttleworth makes a similar argument in his article on seaworthiness, in the section on windward ability: http://www.john-shuttleworth.com/Articles/NESTalk.html
It might be mentioned that the drag is more accurately Velocity2 x drag coefficient
Agreed.
so it increases exponentially.
I am aware this is accepted colloquial English, but an exponential function would have speed in the exponent, not the constant 2. Makes a big difference, and when making quantitative statements, it’s worth not introducing a contradiction between the equation and its verbal description.
(And yes, I also get upset by “All that glitters is not gold”, because it is logically equivalent to “Nothing that glitters is gold”, which is clearly not what is meant. The widespread use and adaptation of the phrase is a victory of style over content. Where Shapespeare can’t get the logic straight, he doesn’t deserve to be quoted.)
Regards
Robert Biegler
Thanks for starting this thread, Mal.
I had been thinking about windage, but in a static way only. As in, if I stop sailing, what wind forces are acting on the hull that isn’t moving, and has minimal steerage? This happens every time a proa shunts. Does it make sense to minimize cross windage at the hull ends so that gusts don’t blow the resting proa off course as it moves out of the shunt? Or does having significant windage at the hull ends serve to stabilize the hull until it regains steerage?
Generally a boat left to its own devices will end up beam on to the wind. This is because the centre of effort (CE) will always move towards the leading edge as a flat plate is rotated in a moving fluid. Equilibrium is acheived when the plate is at 90 dgrees to the fluid and the CE is located at the centre of area of the plate. I would expect that having more freeboard at the ends would cause the CE to shift more, but not by a really significant amount.
It might be worth thinking about how much windage actually effects performance. Firstly, windage is only detrimental to windward performance. Downwind you actually want more windage!
The higher the boat points into the wind, the more damaging windage is. The more you rotate a foward facing flat surface into the wind, the more the wind pressure it experiences, roughly in proportion to the cosine of the heading relative to the apparent wind. At the same time, the available forward thrust from the sail decreases the higher you point into the wind.
For a simple case of a boat sailing at 45 degrees apparent wind angle, about 70% of the total lift from the sail acts as forward thrust. At the same time, the drag of a forward facing flat plate is about 70% of that of the same plate at 90 degrees to then wind. Now the drag coefficient of a flat plate is roughly similar to the lift coefficient of a sail, so at 45 degrees to the wind, for every square metre of forward facing windage surface, you need an equivalent area of sail to overcome the drag.
An efficient proa should be able to sail significantly closer than 45 dgrees to the apparent wind, in which case you need increasingly more sail area than the equivalent area of forward facing windage surface to overcome the drag, for instance, for boat at 30 degrees to the apparent wind you need about 1.7 times as much sail to balance the windage drag.
All these figures are very rubbery, but it might help to get a sense of the magnitude of the forces involved.
While this is all true - there are reasons to compromise - in hot climates - vertical sides are much cooler than angled sides. That and that it’s easier to fit people and all their “stuff” into a bigger/more vertical space….. or fit more into a smaller footprint.
For performance, it makes sense.
How about a formula 1 proa that uses these shapes to channel the wind onto the sails or act as end plates?
This leads to an interesting idea:
If the pod makes a curve from end to end, view from above the hull has much of an aerofoil shape.
It therefore could act as a sail itself and may give some useful lift to windward.
Mark
Hi Mark,
may an idea I posted some years ago in proa_file group will fit your requests.