# Foiler Design

Discussion in 'Sailboats' started by tspeer, Nov 12, 2003.

1. Joined: Feb 2002
Posts: 2,316
Likes: 290, Points: 83, Legacy Rep: 1673
Location: Port Gamble, Washington, USA

### tspeerSenior Member

There area a lot of misconceptions about ladder foils. An abundance of tips is not the problem. Nor is junction drag, for the most part.

The number one problem with ladder foils is people tend to use a ladder to reduce the span of the foil. This increases the drag because induced drag is inversely proportional to span squared.

The other problem with ladder foils is interference between the foil elements. This tends to load up the upper elements more than the lower elements, and since the elements closer to the surface have more induced drag, this also increases the net drag.

A T foil with tips near the surface would experience an increase in induced drag because the surface becomes deformed by the foil. In fact, biplane theory predicts a doubling of induced drag for a shallow-running T foil, and this has been confirmed by tank tests.

2. Joined: Feb 2002
Posts: 2,316
Likes: 290, Points: 83, Legacy Rep: 1673
Location: Port Gamble, Washington, USA

### tspeerSenior Member

This is only partially true. The profile drag is proportional to area and increases as speed squared.

But the induced drag is independent of the area (or the aspect ratio). Instead, it's inversely proportional to square of the span and decreases with the square of the speed.

It is true that profile drag dominates at high speed. But the induced drag is what's likely to determine if you can take off or not. Whether you produce the lift with angle of attack or flap deflection is immaterial to the induced drag - what matters is the shape of the downwash distribution. You want a uniform downwash for minimum induced drag. But a flap can help to optimize the profile drag by centering the low-drag region ("drag bucket") about the operating lift coefficient.

A T foil will have a U-shaped drag characteristic - high drag at low speed due to induced drag, high drag at high speed due to profile and other parasite drags, and a minimum drag speed in between. You can size the foil to place the optimum speed according to the performance requirements.

It's possible to design a T foil that will outperform a V foil at any given single speed. However, careful design of the V foil may allow it to sacrifice some minimum drag for better off-design performance compared to the T foil. But this takes a combination of foil sizing, planform shape, and operating angle of attack. It's particularly important that the pilot of a V-foil craft be able to adjust the pitch trim for best performance.

3. Joined: Feb 2002
Posts: 2,316
Likes: 290, Points: 83, Legacy Rep: 1673
Location: Port Gamble, Washington, USA

### tspeerSenior Member

It depends on the range of angles of attack you are looking to cover. When cruising at a constant speed and using the flap to regulate height, there are three conditions to consider. All three have the same lift coefficient because the lift has to equal the weight and the speed is the same. The first is the design angle of attack at zero flap deflection. The second condition is negative angle of attack with positive flap deflection, and the third is positive angle of attack with negative flap deflection. The range of angles of attack at which a constant lift can be maintained with flap deflection determines the size of the waves that can be perfectly platformed with this arrangement.

The low-drag region of the profile drag is generally going to be centered on the design lift coefficient. Positive flap deflection will move the low drag region to higher lift coefficients. Since the operating lift coefficient is constant this means the limit of optimum performance will be when flap moves the bottom of the drag bucket up to the operating point. Likewise for negative flap deflections moving the drag bucket to position the upper end over the operating point.

If you rotate the whole foil instead of deflecting the flap, then the drag bucket stays fixed and by rotating the foil to mantain the operating lift coefficient, you stay within the drag bucket. So the two techniques are pretty much equivalent, depending on how wide the drag bucket is and how it's affected by flap deflection.

For full span flaps, the spanwise lift distribution is comparable for the two approaches, so the induced drag is the same.

The greatest range of off-design performance is to use a combination of incidence and flap, increasing or decreasing both together as required. Naturally, this makes for a much more complicated mechanism.

4. Joined: Feb 2005
Posts: 5
Likes: 0, Points: 0, Legacy Rep: 10
Location: Australia

### jimbJunior Member

Thanks for the discussion on foil design. My original question was about the best trade off between stability and performance. If we have a fixed high lift foil to get lift out at 6 knots then fly at 20 knots, presumably large amounts of flap or a bow down attitude will be required to reduce lift to that required to support the boat at maximum speed. The bow down attitude will be pretty scary as variations in the vertical component of speed ie running through waves, will result in disproportionate changes in lift which may be difficult to recover from even with sensor control of a flap.

To give a sense of security, many sailors will run at positive angles of attack well outside Tom Speer’s parameters and rely on large flap deflections to destroy the lift. While this will still be running within the drag bucket for this arrangement, the absolute drag will presumably be higher than running with no flap and hence the boat will be slower.

What layout will be least sensitive to changes in vertical speed? This is the critical question to gaining acceptance of foiling by small boat sailors. Moths have very low mass inertia and will respond quickly to variations in pressure on the foils. If a low lift foil is an answer to allow running at positive angles of attack with no flap at speed, will a reasonable size foil be able to lift out at similar speeds to high lift foils without stalling?

5. Joined: Nov 2003
Posts: 161
Likes: 1, Points: 0, Legacy Rep: 15
Location: Sydney

### WardiSenior Member

Why not use a fully articulated lifting foil which will always find its own least drag angle of attack for the required lift?.

It would also enable level flight, not requiring bow up attitude to get the required angle of attack.

6. Joined: Apr 2004
Posts: 36
Likes: 0, Points: 0, Legacy Rep: 10
Location: Stockhom, Sweden

### NiklasLStudent member

A picture says more...

Wich configuration, I'm not shure, do you mean ward? Do you mean the main foil?
Im not 100% English talking...
/NL

#### Attached Files:

• ###### Config 123.JPG
File size:
24.9 KB
Views:
538
7. Joined: Nov 2003
Posts: 161
Likes: 1, Points: 0, Legacy Rep: 15
Location: Sydney

### WardiSenior Member

Hello Niklas,
I was meaning and am using your type 1.

8. Joined: Apr 2004
Posts: 36
Likes: 0, Points: 0, Legacy Rep: 10
Location: Stockhom, Sweden

### NiklasLStudent member

I see three diffrent ways to confront the wave problem, if normal riding control is solved so it's not disturbed by this:

A. The foil is fully articulated (or articulated as 1 or 2 above) and is set to dampen the vertical motion with spring-dampening system set at sensitive frequencies.

B. The whole strut is hung on a feather-dampening system ("like a cars wheel but light weight") wich eliminates the sensitive frequencies for vertical motion. The foil remains at a constant angle of attack and is perhaps less sensitive to trig bubbles?

C. The high aspect foil is so wide that it flexes and cancels the vertical motions out at the sensitive frequencies (constant angle of attack).

/Niklas

9. Joined: Apr 2005
Posts: 1
Likes: 0, Points: 0, Legacy Rep: 10
Location: glasgow/minorca

### chrismillerNew Member

hi,

this question is really directed towards Tom Speer, but if you know the answer feel free to enlighten me. In refrence to your Generic hydrofoil study paper, you say that a surface piercing V foil will have an increasing induced drag as more of the foil comes out the water, ie as the velocity of the boat increases. This makes sense due to the span decreasing, but thinking about it further, why is there induced drag from the foil at all? since both ends of the foil are out the water theres no circulation round the end and so no vortex is produced.

chris

10. Joined: Feb 2002
Posts: 2,316
Likes: 290, Points: 83, Legacy Rep: 1673
Location: Port Gamble, Washington, USA

### tspeerSenior Member

The notion that induced drag comes from the flow around the tips is an over-simplification that leads to a lot of mis-directed designs.

Induced drag, in general, comes from the conservation of momentum - in order to produce a given amount of lift with a finite span you need to deflect a finite amount of fluid - and conservation of mass - the deflected fluid, in turn, deflects and is replaced by other fluid. It's the same thing you see when you drag a canoe paddle through the water - you deflect a "chunk" of water and vortices are formed at the edges of that volume as the water moves out of its way and rushes in to take its place.

When you apply a constant force to the canoe paddle, it starts off slowly but then you end up dragging it through the water a constant rate, because you've actually set up a local current that is pushing the paddle along and relieving some of the force on it. This local current is the analog of the induced velocity or "downwash" (to use an aeronautical term) at the hydrofoil.

You get the same sort of vortices forming at the edge wherever you have one body of fluid moving pas another, whether it is the downwash from a business jet

or the downwash from a crop duster hitting the ground (note the progression of the smoke between the two frames and how the smoke from the bomb on the ground is going straight sideways)
.
or the the fluid following behind a rising air bubble

or the wake behind an island

Wherever you have a slug of fluid moving through another fluid, you will get a vortex structure at the boundary between the two. In the middle, forming the stalk of the mushroom-shaped structure is the moving volume of fluid, and surrounding it is the flow induced in the rest of the flowfield through the conservation of mass.

You still have induced drag with a surface-piercing foil because you are still deflecting water down to produce lift and that deflection still affects the local angle of attack along the span of the foil. True, water is not flowing directly around the tip, but the water surface changes height, allowing the flow to twist downstream of the tip. At the surface, the situation is actually worse than it is at the tip of a fully submerged hydrofoil, because instead of dense water pushing back on the water displaced from the pressure side of the hydrofoil, there's only air.

Induced drag, in detail, depends on the amount of lift generated at a given spanwise section times the induced velocity at that section. The lift and induced velocity don't change in the same way across the span because when you increase the lift and increase the deflection of the fluid at one station, you also change the induced velocities and affect the lift at all the other stations. And if you're keeping the total lift the same, when you increase the lift in one place you have to decrease it elsewhere to maintain the same lift.

It turns out that when you add up each section's lift times the induced velocity at that section, you get the least total if you've arranged things to that the induced drag is uniform across the span. So the aim of the designer iin reducing induced drag has to be #1 to make the span as great as possible, and #2 to minimize variations in the induced velocity across the span.

11. Joined: Jun 2004
Posts: 307
Likes: 29, Points: 28, Legacy Rep: 324
Location: MIT Aero & Astro

### markdrelaSenior Member

Not quite. The local induced-drag/lift ratio must be uniform across the span. This is equivalent to having the induced velocity uniform across the span like you said.

For non-planar wings, the requirement is that the vertical component of the induced velocity should be uniform. "Vertical" here is defined as parallel to the overall lift force vector.

Yep.
It's also useful to note that the induced drag penalty for a non-uniform induced velocity varies as the *square* of the nonuniformities. Lets say we have three wings of the same span and lift, but with different induced velocity distributions wi(y) across the span:

Wind A has a uniform wi(y)
Wing B has +/-10% wi(y) deviation from uniform
Wing C has +/-20% wi(y) deviation from uniform

And lets assume the wi(y) deviations for B and C have the same shape. The relative induced drags might look something like

Wing A: Di = 1.0
Wing B: Di = 1.02
Wing C: Di = 1.08

The conclusion is that it's not worth removing the last bit of wi(y) nonuniformity, since the resulting Di penalty will likely be very small.

12. Joined: Feb 2004
Posts: 15
Likes: 1, Points: 0, Legacy Rep: 10
Location: LA JOLLA, CALIFORNIA

### RVELLJunior Member

Ogival or plano-convex hydrofoil

Originally posted by Ian Ward
...I would like to propose we begin a completely new thread, to encourage a collaborative group to actually design a new boat (a bit like Linux) where we split the contributions into Hull, Rig, Foils, class rules etc and discuss each separately but in detail and with a common aim in mind. ...

To Tom Speer and Ian Ward:

Many of the subjects proposed in 11/12/2000 have been well discussed. Some discussions, like how to control flying height and pitch, are sure to continue well into the future. A subject I have been hoping to see is fabrication of hydrofoil wings.

I see two classes of fabrication problems. Some builder/designers are hoping to win races and set records. Others, like myself, are interested in putting hydrofoil ideas to the test. I find one of the biggest obstacles to building hydrofoil boats is creating the foils themselves. Builder/designers like me do not need record-breaking foils, but foil sections of any kind are hard to come by.

The group I work with here in San Diego has created functioning human powered and motor powered hydrofoils by purchasing foil sections commercially available. We have also had success in creating custom carbon-fiber-epoxy foils. Creating foils of this type is difficult and time consuming. Both commercial and hand made foils of this type are expensive and difficult to join together or join to supporting struts. Incidentally, I also use extruded aluminum foil sections, but my supply is limited and they are really hard to obtain.

With all this considered, I will cast out an idea on how to build a foil from common materials using non-exotic fabrication techniques. This idea has its roots in the methods used to construct Sabrefoil, a motor powered one or two man hydrofoil I created in 1968.

With Sabrefoil, I used a profile called Ogival or plano-convex hydrofoil. It consists of a section of a tube forming the top of the foil and a flat plate forming the bottom. In 1968 TIG and MIG welding were not common as they are today, so I was content to roll a plate of black iron to form the topside. To that I welded a flat plate to form the underside. That sounds heavy, but it was hollow. The big advantage is that a strut can be welded directly to the upper surface for a good junction. The strut is made by welding two sections of a tube together to create a symmetrical foil/strut.

With improved welding techniques available today, I plan to change the material to aluminum, For example, aluminum 6061 pipe 150mm in diameter with a wall thickness of 4mm is locally available. This will be cut into 70mm strips and a 40mm plate will be welded inside the curved section. The long welds will be at the leading and trailing edges. To improve the performance characteristics I plan to grind the leading edge to a 3mm radius. All dimensions are approximate.

Attached is information showing the characteristics of the unmodified plano-convex form. It is compared to the airfoil numbered 63412 to give some perspective. My thanks to R.B.Wade who’s report No. E-79-6 provided the plano-convex numbers.

Tom Speer, your opinion on this hydrofoil profile would be welcomed. I know no records will be set with this foil section, but is does fly and it can be fabricated with a little aluminum and some argon gas.

#### Attached Files:

File size:
103.3 KB
Views:
1,090
File size:
21.6 KB
Views:
1,228
• ###### Ogival 1 Cl & Drag table.doc
File size:
57 KB
Views:
356
13. Joined: Feb 2004
Posts: 79
Likes: 0, Points: 0, Legacy Rep: 10
Location: Windy Wellington, New Zealand

### boogieMember

composite foil manufacture

hi guys,

a very cost effective way of making foils with faily good accuracy is to make throw- away moulds [not cores] from styrofoam.

with a home made hotwire bow powered by a car battery charger you can quite easily cut sections of wings 50-60cm long. they can either be paralell or straight tapered even with a flow of sections and twist built into them.
instead cutting the wings out of the foam and using them as cores [like i do for RC model planes] you can quite easily use the outside of the cut wings as two female half moulds.
just join up as many of the pieces of mould to form the size/shape required.

you can either laminate straight into the styrofoam with epoxy resin and your choice of fibre reinforcement, then break the mould away after curing or prep the mould for easy release by putting the foam mould itself into a vacuum bag. no resin will migrate into the foam and the release is very good with a pretty good surface finish. you can even vacuum the laminate over the vacuumed mould for better consolidation of the laminate.
the good thing about using moulds is that you can pile up the reinforcements in the mould without distorting the shape of the section.

some styrofoam suppliers use CNC hotwire machines to cut all forms of shapes like extruded sections. they can cut very long sections too. like 4-5m as far as i can remember from my work at StyrotechCNC.
we mainly used 40kg/m^3 styrofoam for CNC routing of yacht superstructure plugs and keel bulbs.

if you are interested in more details about this technique let me know and i'll put some pictures together.
here are two of many links how to cut foam wings:
http://members.fortunecity.co.uk/slmohr/rcinterest2.htm
http://webpages.charter.net/rcfu/ConstGuide/FoamWing.html

cheers
boogie

14. Joined: Feb 2002
Posts: 2,316
Likes: 290, Points: 83, Legacy Rep: 1673
Location: Port Gamble, Washington, USA

### tspeerSenior Member

One of the reasons ogival sections have been popular in the past is you can make them easily out of wood by attaching wood blanks to a polygonal cylinder. Then turn the whole thing on a lathe to create a circular cylinder and remove the finished shapes. However, I believe the sharp leading edges on such sections were responsible for the trouble these yachts had with ventilation.

But today I think you can find someone that can numerically machine a sophisticated shape at a reasonable price. Or you can buy prefabricated foil stock from companies like Fastacraft. Maybe not as low-cost as you had in mind, but when you consider the time and money you have in the rest of the boat, I think it's money well spent.

Burt Rutan used an interesting method to create the wing skins for Voyager. He cut out female templates of the shape, and Bondo'ed them to the shop floor. Then he glued sheet metal to the templates to form a smooth female mold. Depending on the size of your foil, a similar technique might be used. Possibly with a leading edge piece routed to shape.

15. Joined: Feb 2004
Posts: 15
Likes: 1, Points: 0, Legacy Rep: 10
Location: LA JOLLA, CALIFORNIA

### RVELLJunior Member

Foil fabrication, more

Good response, Tom & Boogie. Actually within our little group of hydrofoil fans here in San Diego we have purchased Fastacraft foil sections, and Jake Free’s carbon fiber “Shutt strut”. All have performed as advertised.

We have also done hotwire Styrofoam core making. Steve Ball has used this and a similar technique: First, using a bench grinder, he hand grinds a blade to duplicate a foil’s profile. The blade is used in a “shaper” machine to scoop out wood taken from the poplar tree. This creates a female mold. In general, the mold is packed with carbon fibers and resin. The cost of carbon fibers and resin is significant.

I love Bert Rutan’s idea. I interpreted this as making “female” ribs that create the shape of the female mold after they are glued lineally and then lined with sheet metal. Before bondo-ing these to the floor, I need to check with my wife.

Incidentally, try this: In your computer, describe the section of a foil numerically in the conventional way. That is, assign thickness dimensions to stations along a cord. The dimensions go into a simple table. Use Microsoft Excel to make a line graph of the points. Fool around with the scaling until the line on the graph accurately depicts the foil section. You can smooth the line with an Excel command (which I have forgotten) or you can mathematically interpolate between the numbers to create a large number of stations—more than the conventional ten stations. This also smoothes the line.

Once you have the excel file you can scale the foil. In Excel use “File/Print Preview/setup/page/adjust to (percent size)” to change the size of the printout in 1% increments. When happy with the form, print it onto conventional paper or cardboard--if your printer can handle cardbooard. The paper/cardboard is then trimmed and you retain the male or female part, as needed. You then glue it to the end of wood stock or Styrofoam. Conventional methods are used to cut away the material to conform to the foil’s shape and bingo, you have a rib, mold, foil core, etc.

Like all computer things this procedure may sound lengthy and convoluted, but once you have the foil’s profile in memory, it can be reproduced, re-scaled, gender changed, whatever. You can also email the profile to a friend.

Two sticky points need addressing: How does one create a strong, compact joint between the foil and its strut, and how does one create and attach flaps and its linkage to the flaps?

These problems make aluminum attractive. The joints between foil and strut are welded, no problem. As for flaps, the material can be grooved, drilled and threaded. This makes the attachment of stainless steel piano hinges possible. These are commercially available.

I favor this technique, yet to be tried: Create a flat trailing edge, perpendicular to the bottom surface of the foil. Along the trailing edge weld short sections of aluminum tube one inch long and each separated by a one inch gap, for example. During welding the tube sections are held in line with a stainless steel rod running through the sections. After welding, the tube holes are re-drilled using a long drill bit. The SS rod is repositioned as a hinge pin.

To fabricate the flap, weld a thin bar of aluminum on a tangent to the same type hollow tube used above. The flap tubes are spaced to dove tail to the tube segments described above. The underside of the assembly is then filled with Bondo to create a streamlined shape. An attachment fixture is welded to the trailing edge of the flap to receive the control rod. Grind & file smooth and you are ready to fly.

Forum posts represent the experience, opinion, and view of individual users. Boat Design Net does not necessarily endorse nor share the view of each individual post.
When making potentially dangerous or financial decisions, always employ and consult appropriate professionals. Your circumstances or experience may be different.