USA46 said..
I was thinking will I post this link or not, because I knew that this old theory is inside text.I hope no one will read all text.
Sailquik you are now open new most complicated topic "ELIPTICAL VS BELL spanload distribution"[/b]

Many thanks. I really appreciate that you did decide to publish these links. I have spent the whole morning going down that rabbit hole and I am beggining to see the light at the end of the tunnel. That is exactly the sort of stuff I was hoping for.

USA46 said..
our sails has bell spanload,especially with high downhaul trim, page 32 ,figure 19 [/b]

Can you tell us where this is and what it refers to please?

USA46 said..

I didn't understand you well,how do you mean where this is?

I put link below this text in my post..maybe you didn't see it..open page 32,figure 19

link:
mauiultrafins.com/technology-2/windsurfing-mechanics/


figure 19 show lift distribution over height of sail.You can see in lower part is max lift(power zone),lift is droping with height as consequnece of sail twist and taper,if you trim sail with high downhaul like on slalom sail,head will be floppy ,so lift line in graph will touch zero point even before 5m height on this picture...etc

Ahhh, gotcha now. I didnt notice that link has pages.

decrepit said..
It's also obvious that the bell curve has a much lower CoE, just what windsurfers need to control the power, and reduce the down force on the nose.

So now we go back to Andrew's original question, how much twist is needed for the optimum bell curve?

Well, I have not finished reading though and digesting all the links and papers yet, but my initial feeling is that we are not getting it yet, even with the most twisted sails.

The reason is that the sail is a flexible structure (although so is the birds wing) and we cant actually get enough twist to get a reversal of lift at the top (negative lift). But I am thinking birds can't either, so more thinking and study required......

If it is possible to remove the tip vortex from the actual top/tip of the sail to some place lower on the span, we would have it. Maybe we do have it now in some circumstances??? Maybe not. Maybe it would requite us to go back to a more Higher Aspect sail???

I am wondering if the Little AC designers tried to achieve this. If so, there should be evidence of significant twist in their wings.

decrepit said..
I'm not sure you can shift the tip vortex?
I think it's either at the tip or not there at all.
It's at the tip because the high pressure can get to the low pressure there. It can't happen on the trailing edge because the airflow stops that happening.

Perhaps I should have said the 'vortex' As I understand it, In the Bell Curve situation, there is still a vortex off the wing, but it is not at the tip, it is where the lift crossed from positive to negative, and I assume it is a lot smaller. And, as I understand it so far, this results in a higher lift to drag ratio, caused by a big decrease in induced drag (from the tip vortex).

Pacey said..
..I believe that both Oracle and Team NZ played with negative lift in the head of their wings for the 2013 America's Cup. This was possible as the wings were a slotted flap design, and it was possible to set the angle of the rear flap (and its twist) independently for the lower, middle and upper sections of the wing. However, my understanding is that this was done for the purpose of increasing righting moment when overpowered as opposed to minimising drag.

Been aware of this concept of aerodynamic ballast for several decades now. It is quite interesting but in the past difficult to impliment. The AC cats would be the perfect platform to try it.

^^ ... presumably our backwards raked sails will exaggerate this effect by encouraging more airflow up the sail.

(I forgot it to write word "local" in discription of AoA2(my picture in previous post), AoA2 is angle of attack between local apparent wind(upwash) and sail head.)




Picture 3a shows downwash distribution behind wing.Arrows pointed down represent downwash and arrows pointed up represent upwash behind wing.We can see that at some point of wing downwash shift in upwash.This is point where induce drag become negative/this wing part produce small thrust component.


People allways ask me how wing can produce positive lift(lift vector poitned up) at outborad wing part if arflow behind this wing section goes upward(upwash)?
Yes it can, because outborad wing part "feel" oncoming local apparent wind("upwash") is coming from different angle to wing...

-Upwash is upward airflow movemet in relation to freestream wind
-Downwash is downward airflow movement in relation to freestream wind
-Freestream airflow/wind is wind that is not disturb by any object or phenomen,in windsurfing it will be apparent wind(vector sum of true and head wind)

In general:
Whenever wing produce lift, upwash is happening in front of wing and downwash behind wing.Because lower pressure on top of wing has influence on oncoming airflow, it " sucks" airflow upward.And behind wing is going down because wing is tilted down at angle(AoA).
But here we have on outboard wing upwash infront and behind wing , BUT both in RELATION TO FREESTREAM WIND.
So this is relative term,If we look at outboard part of wing separately,when freestream wind is not our reference point,than again airflow infront is bend up(upwash) and ariflow behind wing is bend down(downwash) as it should be whenever every wing produce lift.
Beta is angle between freestream and local upwash airflow,this part of wing "feel" local upwash airflow not freestream wind.
I know that all this is confusing but I'm doing my best to make it clearer..

Yes, spot on Decrepit.

Thankyou USA46. you are indeed making it much clearer.

So it seems that we indeed may be benefiting from an overall reduction in induce drag with a highly twisted, flat camber head, sail.

So back to my initial question. How much twist do we need to take full advatage of this?

These are the questions that come to mind:

Is there a way to calculate the 'Beta' (Upwash) angle for a given sail? If we assume a certain angle of sail and angle of apparent wind?

I can see that the camber of the sail and the agle of attack will influence this, along with the velocity of the airflow. Can we get close enought on what these things would be in a speed sail to get a number for twist?

Also, would it be correct that the deeper the camber/draft of the sail in the lower section, the greater the upwash Beta angle would be at the top, and that we would need more twist?

This also seems to mean that on an upwind sail, which has a finer leading edge angle and shallower draft, the upwash angle beta would be less and we would want less twist for the same benefit.

Boy, a lot of interesting stuff here. As has been shown you can see how lift is generated, what happens at the ends of wings as lift tries to escape the easiest way possible, and what can be done to control it if need be.One thing not really mentioned in here is that windsurf sails are hand held, alot that goes into windsurf sail design is not so much searching for ultimate lift to drag numbers but it's how to be able to get going, once going being able to keep accelerating and ultimately be able to hang on to it in the end. The advent of heaps of twist coming into sail design (along with strong draft forward wherever it exists) enabled more sail area to be carried, flat but twisted off it generates drive forward (because its rotated what lift it does generate is more in the direction you want to go) but once underway and apparent comes forward that power turns off (the head turns into a wind vane) as the rest of the lower sail powers up with speed so you can still hang on. As the head is now more of a wind vane the flow on each side is equal and it now acts like a fence and keeps vortices in check to some degree. So the twists job changes through the range. In building twist in the other effect is that the mast becomes shorter, the head is no longer at the top of the mast but in a sense about a third of the way down, that's where your leach tension becomes directed to. The mast then effectively becomes shorter and stiffer which reduces mast movement and the changing of draft position that usually comes with it, the sail is more stable and once again you can hang on to it better.
What's the right amount of twist? ,depends on what angles you are sailing and how the head matches how the rest of the sail is being set. Keep in mind that downhauling the crap out of your sails is to increases all the above effects and to overpower the material makeup of the sail, to increase its flexibility to deal with mast movement, and all lead to control which gives you speed. Control is often turning things off, not on, you can't hang on to certain kinds of efficiency. To be fast you sometimes have to work with inefficiencies, so don't get carried away crunching numbers looking for magic, good sails have range because they are tuneable to what's happening on the day and you have to set the tune.

Mr Fussy, thats a Great easy to understand comprehensive explanation. Agreed there is no one magic number. As with all windsurfing gear its about what suits the individual on the day. But id guess theres a certain scope/amount of twist thatll work for most.
Bring out the smoke flares Sailquik. Lets see how different sails look in reality.

sailquik said..
I am sure you will agree that more knowledge of the actual physics can help to focus the direction of further, more rapid advancements.

The best way to demonstrate a complete understanding of the physics is to write a numerical model that simulates measured results. The simulations for the first moon landing were so good that the astronauts felt during the landing that they had been there before.

sailquik said..
Oh come on Barney. Who's is going to do that?

Perhaps one of the internet experts who understands the physics or knows how to use google.

I'm with Daffy on thinking that more experimental data could lead to a better understanding and therefore better sails. I also agree with Barney831 that a successful simulation would be a very good way to "demonstrate a complete understanding". The funny thing is, though, that to call a simulation successful, you need to show that it matches actual measured data (unless you're the marketing guy).

I think numerical simulations have helped in fin design in some cases- the Tectonics Weed Demon comes to mind, and there must be others, too. But these are comparatively easy simulations with a fixed fin shape and constant speeds and angles, as opposed to a constantly changing sail shape in variable wind - and they still required tons of (super-)computing power. Adding all these variables pretty much introduces a combinatorial explosion - the typical "brute force" approaches to numerical simulations don't stand a chance.

This reminds me a lot of computer approaches to playing chess and go. Chess is comparatively simply, and enough CPUs will allow brute-force solutions that beat the best human players. The same approach in Go completely failed - at the time where any computer was able to play chess at the level of a grand master, even the best go programs on high end clusters had no chance against top level amateur go players.

The outcome changed dramatically just a couple years when a new neural network approach succeeded in beating the best human go players. Most amazingly, the software learned to play better than any human only by playing against itself - the only "knowledge" from humans it needed where the rules of the game (which, importantly, include how to determine the winner).

I think a similar black-box, neural-net based approach might be the only feasible approach to simulating the air flow around and forces on a flexible windsurf sail. Since China is leading in both AI research and in windsurfing competition, maybe there is hope! But as Daffy explained, we need data! Smoke generators in front of the mast to visualize the air flow should work beautifully with GoPros to capture the data at 120 fps in real-world conditions. Much cheaper than wind tunnels, too! This could be a great use for all those safety flares required for slalom racing that expire unused. Might be best for places like Shark Bay, though, to minimize the chances that the data gathering is cut short by an early "rescue".

I was talking to Mal Wright about his fin design analysis software (Finmaker) yesterday. I wonder of it could handle the shape of a sail?

sailquik said..
Why is it that all these great films demonstrating this stuff, were all made 40-50 years ago It was the same with all the hyrofoil stuff that Fangy dug up.

Maybe because all the best sailors were made 50 years ago too?