Question about wingsail foil shape

If you read the Marine Technology article, you realize that induced drag far outweighs profile drag, even in fairly fast boats. That's why the efforts to reduce drag with things like stayless masts are a waste of time, especially if they are heavy. It was eye opening to me.

Lohring Miller

 

For fun, I tried to make basic calculations for an A-Cat*, and I could see that sail section drag is around 7N if you assume 1% section drag or 21N if you assume 3% which seems a lot

Induced drag with a footsail @ 30cm from trampolin will score 44N, if you close the gap you decrease induced drag to 27N for a 200 Newtons total drag (Aero + Hydro)

* A-Cat windward with 9knts TWS and around 16knts AWS

Of course it is hand made calculations and not CFD, just to get a rought idea.
In this case the point is for a teardrop mast + sail, the section drag coef if probably a bit more difficult to appraise than for a NACA wing section, that is why I make guess between 1 and 3%

 

Found this old Naca video some my find interesting.

https://www.youtube.com/watch?v=q_eMQvDoDWk&list=PL_hX5wLdhf_K3mK1TZNCkmdD-JMZYGew1

 

That was tremendous,red-thanks!

 

Question to the hive mind.

Wing masts, at what percentage of sail area does a slot start to show real benefit ? 30% 50% ?

 

There's not a lot of difference between sections that have a main element/flap ratio in the 35%/65% to 60%/40% range. It depends on what your requirements are for the rig. If the emphasis is on high lift, you probably want to go more toward the smaller main element & larger flap. If the emphasis is on low drag, you might want to go the other way. It also depends on whether or not you believe you can achieve significant amounts of laminar flow or not.

And just to clear up a possible misconception, the purpose of the slot is not so much to vent from the windward side to the leeward side. It's much better to think of the main element and flap as being two wings close together such that they interfere in a purposeful way. For example, the flow moving at right angles to the chord around the leading edge of the flap is used to change the boundary condition at the trailing edge of the main element, increasing its lift.

 

Thanks Tom,
The inevitable supplementary questions,
How large should the slot be ?
Is it important to have the leading edge of the rear element slightly to windward or does inline work just as well ?
Inline is easier to construct and control as it's the same on each tack, for what I am thinking the rear element would be a sail rather than a profiled element. For simplicity.

 

The leading edge of the flap definitely needs to be to windward of the main element trailing edge. Inline misses the whole point of the exercise.

Read AMO Smith's paper on high lift aerodynamics. He talks about the 5 ways that a slotted section increases maximum lift. It's important to locate the trailing edge of the main element near the high velocity region on the flap leading edge to get a high "dumping velocity" for the main element. You can't do that with an inline flap.

 

The way I have been thinking about it is that the slot needs to create a nozzle converting pressure from the windward side into air velocity in a favorable direction. This air jet can cut a thick boundary layer down at the tail end of main foil. I think the video supports this line of thought. With this in mind I have a bunch of inferences that I would like to run by you.

1 Slots are a high angle of attack tool -pointless below say 10 deg
2 Slots are better suited to large main little flap wings (like 70/30 or more) where thinning the boundary layer of the main yields more
3 The optimal tail of the main and forward edge of the flap in a slot design would differ significantly from free stream designs because the interaction is what counts and you have such a high degree of control from above and below. It's a nozzle, not a head and a tail.

I have read that slots in two element wings are problematic, difficult, not worth it but could it be that they didn't consider the above? Could the 'trouble' be that their design was a couple of NACA profiles that don't make a good slot?

4 The last thought I have is that because small variation in the slot makes big differences in the boundary layer ->lift wouldn't it be a great low energy control compared to the high power required to control main angle of attack?

 

Unfortunately, that's not the way it works.

There can be uses for slots at low angles of attack, too. For example, Whitcomb's original supercritical airfoil design had a slot in it and it was designed for high-speed cruise.

If you measure the angle of attack from the main element chord, wingsails actually operate at a negative angle of attack much of the time.

Not so. They are quite helpful for the 50% chord flaps used on USA 17. The ETNZ AC72 wing had a 60% chord flap. So I wouldn't say that slots are best for small flaps.

That's not true at all. It's definitely a trailing edge and a leading edge. If you put a tab on the main element, it acts like a plain flap. The trailing edge gets designed the same way it does for a single element wing, but taking into account the interference effects of the flap. In both cases, one takes into account the amount of adverse pressure gradient the boundary layer can withstand without separation.

It doesn't work out that way. Here's an example. The first figure shows the cross section shape from an AC72 with 30 deg of flap deflection and three different tab angles, which change the width of the slot.

The second figure shows the pressure distribution at an angle of attack of -10 deg (relative to the main element chord). The arrows show the direction of change with decreasing tab angle (opening the slot). The lift on the main element is decreased because the tab is acting like a plain flap on the main element. The lift on the flap is increased because there is more air flowing through the slot, and more circulation around the flap. As a result, the flap develops a leading edge suction peak.

The third figure shows the effect on lift, drag, and moment from opening the slot with the tab. There is almost no change in lift or drag in the linear range for the first change in tab angle. For the second change, the leading edge suction peak on the flap results in an increase in drag, but there is only a modest decrease in lift. Maximum lift is predicted to increase somewhat as the slot is opened.

But the big change is in the moment curves. There is a clear shift in the moment due to changing the tab. This is because the changes to the main element and the flap are offsetting each other. The decrease in lift on the main element is offset by the increase in lift on the flap, so there isn't much net change in the total lift. However, the production of the lift has shifted from the main element to the flap, so there's a change in the moment, moving the center of effort aft. (Which was useful on a boat that had lee helm.)

You don't get these effects from thinking of the slot as a nozzle that blows on the boundary layer. They make a lot more sense if you think of the main element and flap as separate wings in their own right that are interacting closely. Opening or closing the slot with a tab is not an effective means of depowering the wing.

 

Thanks Tom. This is how we learn.

To follow up on your comments, my interest in aerodynamics (and thus my comment) is all for sailboats ,subsonic, incompressible.

Thanks in particular for the graphic explanation of the AC72 wing. I love this stuff but it leads me to a few more questions.
Lee helm and the value of shifting the center of effort -is lee helm intrinsic to the AC72, or foiling catamarans (for stability), or is the value of shifting CE true of upwind/downwind races?

Looking at the pressure distributions I can't help extrapolating to close the slot completely. The result I would expect is a slight increase in lift and a slight decrease in drag. If I sent a sailor out with that profile I would not be surprised if that is exactly what they would do and come back with the opinion that the slot doesn't help and the boat needs to be re-balanced.

 

Do more research.

 

Any particular recommendations? I found great classes on EdX but they are focused on airplanes.

 

There is still a question about the reason of use the multi-element slotted wing. Tom presented the performance graphs for such a wing. Was it at Re=1370000, like in the first your graph, Tom?

If so, here are the performance graphs (obtained by XFLR5) of a single, symmetric foil at the same Reynolds number (as well as Mach number and ncrit. BTW, why did you use ncrit=1? Do you expect so turbulent air?). It took me about an hour to design the section, it may not be optimal, but pretty good.

As you can see, the single section is not much worse then the slotted system.
Minimum Cd is the same, maximum L/D is almost the same, and occurs at almost the same Cl.
The maximum Cl is about 2.1, compared with 2.7 for the slotted wing (for ncrit=5 it grows to 2.25). So, if you need the same maximum lift, you have to make the wing 28% bigger - unless you raise an additional sail.

Even so, I think the bigger wing without moving parts will be still simpler, lighter, stronger than the slotted one.

regards

krzys

 

C class cats.