Foil section? For foil board

Discussion in 'Hydrodynamics and Aerodynamics' started by Slingshot, Dec 24, 2019.

  1. Slingshot
    Joined: Aug 2019
    Posts: 22
    Likes: 3, Points: 3
    Location: South pacific

    Slingshot Junior Member

    i am seeking input on what foil sections are being used for foil boards or any suggestions of a section to try. What I note so far is the area of the foil is adjusted depending on the use of the foil. Ie surfing=large area, low aspect, slow speed 5knots and up. Kite racing = small area, high aspect, high speed 15-35 knots.

    I am planning to make a few foils with my son to get started and have fun building a project with him. At this point I am thinking of using an asymetrical NACA series profile, but would welcome any suggestions or discussion.
     
  2. BlueBell
    Joined: May 2017
    Posts: 1,216
    Likes: 224, Points: 63
    Location: Victoria BC Canada

    BlueBell Ahhhhh...

    Good plan. Lots of calculations ahead.
    Have you done any research?
    What materials will you build with?
    Or will you purchase foils?

    A profile with just enough inherent lift for your weight and span would be a good starting place.
     
  3. Slingshot
    Joined: Aug 2019
    Posts: 22
    Likes: 3, Points: 3
    Location: South pacific

    Slingshot Junior Member

    I plan to make the tower and foils out of foam and carbon. I have found reference to people using an Eppler 817 profile for the foils. The foil will be foam with 2 layers bi ax both sides, one layer uni on under side of main foil.
    For the 1 meter tower my laminate schedule from one side to other will be 2 layers 200 gram bi ax, 3 layers uni carbon, foam to fit NACA 0009 section, 3 layers uni carbon, 2 layer 200 gram bi ax. All wet lay up and vacuum bagged.
     
  4. BlueBell
    Joined: May 2017
    Posts: 1,216
    Likes: 224, Points: 63
    Location: Victoria BC Canada

    BlueBell Ahhhhh...

    So, my knowledge is pretty limited but I wouldn't go any thicker on the 0009 strut
    and I have every confidence your research on the Eppler 817 is correct.
    Choice(s) of foam(s)?
     
  5. tspeer
    Joined: Feb 2002
    Posts: 2,273
    Likes: 247, Points: 63, Legacy Rep: 1673
    Location: Port Gamble, Washington, USA

    tspeer Senior Member

    I don't think the E817 is a good choice. It is intended to operate at high speeds without cavitation - much higher than you will be sailing. At the kinds of speeds with the size of foil you'll be using, it can form a leading edge suction peak that can lead to stall or ventilation. Here is a section that is designed more for the kinds of speeds you'll be encountering. I've uploaded some Xfoil data at a variety of Reynolds numbers. Cl=0.3 is in the middle of its design range.
    h105shape.jpg
    Of course, the best choice is a section that is designed for your specific requirements using a tool like Xfoil.

    More important than the section shape is getting the area and the span sized right.
     

    Attached Files:

  6. tspeer
    Joined: Feb 2002
    Posts: 2,273
    Likes: 247, Points: 63, Legacy Rep: 1673
    Location: Port Gamble, Washington, USA

    tspeer Senior Member

    This chart shows some of the design tradeoffs vs the E817. Your choice of takeoff speed will be very important, because that will size the foil and determine the foil loading. I'm guessing you'll want the foil loading to be fairly low, perhaps around 200 lb/ft^2/10 kN/m^2 to 400 lb/ft^2/20 kN/m^2. In that range, the H105 actually has a higher incipient cavitation speed than the E817. More importantly, it has a lower takeoff speed.
    [​IMG]
     
  7. Erwan
    Joined: Oct 2005
    Posts: 384
    Likes: 17, Points: 18, Legacy Rep: 97
    Location: France

    Erwan Senior Member

    A candid question

    Are coordinates of this H105 section available on the UIUC database, or elswhere ?

    Thanks and have a nice year ending party

    EK
     
  8. Doug Halsey
    Joined: Feb 2007
    Posts: 440
    Likes: 121, Points: 53, Legacy Rep: 160
    Location: California, USA

    Doug Halsey Senior Member

    They're in the files that Tom attached two posts up.
     
    BlueBell likes this.
  9. BobJohansson
    Joined: Jan 2020
    Posts: 2
    Likes: 1, Points: 3
    Location: Seattle, WA

    BobJohansson New Member

    Is there a "Remedial Hydrofoils 055" tutorial where a total n00b might learn how to read that chart? I just want to tinker around with them for a small craft, say, 1000lbs gross weight including 1 pax. So far I've found no textbooks on the subject and only bare mentions in aerodynamics textbooks. Thanks!
     
    BlueBell likes this.
  10. tspeer
    Joined: Feb 2002
    Posts: 2,273
    Likes: 247, Points: 63, Legacy Rep: 1673
    Location: Port Gamble, Washington, USA

    tspeer Senior Member

    There's a lot of information on that chart, so I'll break it down.

    Cavitation occurs when the local pressure falls below the vapor pressure of water. So if you can keep the local pressure above that threshold, then cavitation cannot occur. Most text books present the cavitation bucket in terms of the minimum pressure coefficient, Cpmin, which is the nondimensional version of the lowest pressure to occur anywhere on the surface of the section. The pressure coefficient varies with the square of the speed, and this tends to compress the bottom of the chart. I prefer to change the pressure coefficient to the velocity ratio (local velocity/freestream velocity) for the Y axis of the chart because it expands the scale of the high-speed portion.

    The faster the boat goes, the less is the increase in velocity that can be tolerated without exceeding the vapor pressure threshold. This is represented by the horizontal lines on the chart, which show the allowable velocity ratio for different freestream velocities.

    The horizontal axis is the lift coefficient at which the foil is operating. If you know the lift coefficient - given by the X axis - and you know the velocity - given by the horizontal lines - then you know the loading on the section (lift per unit area). You can connect the dots at each combination of lift coefficient and speed to create the locus of points that all correspond to the same loading. This is shown by the fine black and dashed lines, with the solid lines being English units and the dashed lines being metric units. This grid makes it convenient to look up the lift coefficient for any combination of loading and speed. When the hydrofoil is flying straight and level, the lift has to equal the weight, so a fully submerged foil tends to operate along one of the loading lines. You can divide the weight by the planform area to get a quick idea of where the foil will be operating.

    The heavy colored lines correspond to the characteristics of different foil sections. When you sweep the section through a range of angles of attack, each angle of attack will produce a lift coefficient and there will be some point on the section that has the minimum pressure. The minimum pressure point is also the point of maximum velocity ratio. The section curves plot the combination of lift coefficient and maximum velocity ratio over the angle of attack range.

    The section curves are bucket-shaped. Near the design operating condition, the maximum velocity point will typically be near the middle of the chord on the suction (upper) side. As the angle of attack increases, the local velocity increases more at the leading edge, while the velocity at the trailing edge hardly changes at all. This causes the point of maximum velocity to move forward as the lift coefficient increases. The magnitude of the maximum velocity also increases, which causes the curve to bend upward. Eventually, a narrow pressure peak forms at the leading edge, which causes the curve to bend sharply upward. Although cavitation is generally associated with high speeds, the leading edge can also cavitate at low speed and high lift. This can put a lower limit on the takeoff speed.

    As the speed increases from the design condition, the pressure distribution will form a suction peak on the underside of the leading edge. This is what causes the section curve to bend sharply upward on the left side of the diagram. So at a given operating speed, there are typically two lift coefficients that will produce a local pressure low enough to cause cavitation.

    Bottom line is the operating point needs to be inside the bucket to avoid cavitation. At a given speed, as long as the lift coefficient is between these two extremes, cavitation cannot occur. At a moderate speed, say 25 - 35 kt, there is a wide range of lift coefficients available for cavitation-free operation. But at high speed, the two extremes narrow until there is only one loading that will give the highest possible cavitation-free speed. For the NACA 4412, this is approximately 1000 lb/f^2 or 50 kN/m^2 at 32 kt. For the E817, it is about the same loading at 40 kt. If you follow the loading curve up and to the right, you'll find the E817 experiences leading edge cavitation at approximately 22 kt. The NACA 4412, however, is off the chart and may experience stall before the leading edge cavitates. So the NACA 4412 would be a better choice than the E817 for a low-speed hydrofoil that doesn't need to go much over 30 kt.

    The takeoff speed for the E817 can be lowered by increasing the planform area, thereby lowering the loading. While a loading of 600 lb/ft^2 or 30 kN/m^2 would lower the takeoff speed to the mid teens, at high speed it will start cavitating on the underside of the leading edge, limiting its high-speed operation to around 38 kt.

    At the same loading, the H105 would have a slightly lower takeoff speed, but its big advantage would be that it can go almost 42 kt without cavitating. At first glance it may appear the E817 has a much bigger cavitation envelope than the H105, but the bottom right corner of the E817 cavitation envelope is not really useable. It corresponds to loadings that are so high that it doesn't take much of a speed change to cavitate on either the upper or lower side of the leading edge.

    The blue dashed and brown lines show what happens if you scale down the E817 in hopes of achieving really high subcavitating speeds. Reducing the thickness from 12% to 7% gets you 55 kt, and 5% thick will get you 60 kt. But the leading edge cavitation problem becomes severe. Only a narrow range of lift coefficients are available, and the takeoff speeds are quite high. It's unlikely that a fully submerged hydrofoil could be operated like this. However, it might be feasible to consider a surface piercing hydrofoil. Surface piercing would allow the immersed area to be varied with speed so the foil didn't operate at a single loading. If the area were four times greater at takeoff than at high speed, then such a thin section could be kept in its narrow operating range.

    That's a long explanation, but I hope it shows how this chart brings a lot of system design information together in a way that lets one compare the implications of different section shapes.
     
    gigi, Dolfiman, The Q and 4 others like this.
  11. BobJohansson
    Joined: Jan 2020
    Posts: 2
    Likes: 1, Points: 3
    Location: Seattle, WA

    BobJohansson New Member

    Thanks! I didn't expect a personally-written explanation.

    Freestream velocity is the overall boat-vs-water velocity? Local velocity (used in the ratio) is around the foil itself? Takeoff velocity is the point at which the foil begins to provide measurable (or perhaps "useful") lift?

    Is takeoff velocity shown on the chart, perhaps related to the inflection point of the foils' curves? or can it be estimated from them?
     
  12. tspeer
    Joined: Feb 2002
    Posts: 2,273
    Likes: 247, Points: 63, Legacy Rep: 1673
    Location: Port Gamble, Washington, USA

    tspeer Senior Member

    Yes, freestream velocity is the boat speed and local velocity is the speed of the water relative to the foil at a specific point on the foil contour. The local velocity will be zero at the point where the flow divides between flowing around the top vs under the bottom. From there it accelerates as it moves around the foil and then decelerates toward the trailing edge before leaving at a speed that is close to the freestream velocity.

    Technically, we're talking about the local velocity just outside the boundary layer. The velocity (relative to the boat) of the water that actually wets the foil surface is zero everywhere! As you inspect the flow while moving out from the surface, you'll find the flow very near the surface is moving at a slow rate; the flow just outside that moving a bit faster; the flow outside of that flowing a bit faster yet; and so on until you reach the outer flow. This near-surface region is known as the boundary layer. The variation in velocity with distance from the surface is a shearing of the flow that constitutes the skin friction.

    Another strange feature of fluid physics is although the fluid in the outer flow may be traveling along a curved path, it does not rotate! If you were floating with it, you'd always be facing the same direction even though your direction of travel would be changing. And there are no losses in this part of the flow - the total energy is the same everywhere. But in the boundary layer, the flow is rotating because of the shearing going on in the boundary layer. All the losses that make up the profile drag occur in this rotating flow in the boundary layer.

    There isn't a distinct division between the boundary layer and the outer flow - things just fade away. The boundary layer grows in thickness as it is swept down the surface. Imagine that you coated the entire foil with perfume and then turned on the flow. At the leading edge, you'd have to get very close to the surface before you detected the perfume diffusing out into the flow because it would be quickly swept away before it could be replaced by more perfume diffusing outward. But further down the foil, you'd be able to detect the perfume much farther away from the surface because at that location not only would the perfume be diffusing out from the surface, but you'd also be detecting the perfume that had been swept back earlier and had more time to diffuse outward.

    The pressure doesn't change across the boundary layer, so the pressure at the surface is the same as the pressure outside the boundary layer. And the amount of shearing at the surface depends on the velocity outside the boundary layer. This is why we talk in terms of the outer flow velocity and pressure when we're really interested in the pressure acting on the surface.

    If the pressure is decreasing in the streamwise direction, then the boundary layer gets sucked along and doesn't grow as quickly or even becomes thinner. But if the pressure is increasing in the streamwise direction, then a given slice through the boundary layer has higher pressure ahead of it and lower pressure behind it, so it gets slowed. But the velocity isn't the same across the height of the boundary layer - it is low near the surface and faster farther out. So it can get the point where the low velocity near the surface gets slowed to a stop and then pushed backwards. Where the velocity next to the surface is brought to a halt, it can't go upstream because of the momentum of the flow coming at it. It can't go downstream because of the increasing pressure and the flow being pushed back upstream. It can't flow into the surface because the surface is solid. There is only one direction it can go, and that is out away from the surface. This is what causes flow separation. The boundary layer flow that is being forced out from the surface pushes other flow out of the way, and the flow is no longer following the surface contour. This is what causes stall - a loss of lift and an increase in drag with a turbulent dead zone being left in the wake.

    It turns out the whole art of section design has to do with manipulating the distribution of pressure over the surface in order to control the development of the boundary layer. And avoiding constraints like cavitation. If it weren't for the boundary layer, every section shape would behave just the same and there wouldn't be any profile drag at all.
     
    gigi, Dolfiman and Erwan like this.
  13. tspeer
    Joined: Feb 2002
    Posts: 2,273
    Likes: 247, Points: 63, Legacy Rep: 1673
    Location: Port Gamble, Washington, USA

    tspeer Senior Member

    With respect to takeoff speed, it could be anything. It depends on your design - you can take off at any speed as long as you make the foil big enough! But a big foil has excessive drag at high speed because of all the wetted area. So there's a strong motivation to make the foil chord as small as possible.

    But there are limits to how narrow you can make the chord. A small foil has to support just as much weight as a large foil. At low speed, the small foil needs higher angle of attack, and if the size is made too small it will stall. Even before the foil stalls, the profile drag will be increasing rapidly, and this will require more power. And if the foil does not stall at takeoff speed, it may experience leading edge cavitation which will reduce the lift and increase the drag.

    Before takeoff, the craft has the drag of the hull, the parasite drag of the foils, and the drag due to lift on the foils. Compared to the boat alone, the hydrofoils are adding drag and hurting the performance. As the boat starts to lift out, the wetted area of the hull decreases and the immersed volume decreases, so the skin friction and wave drag start to drop. Drag due to lift decreases with the square of the speed, so as the boat lifts out and accelerates, this drag is reduced, too. The parasite drag of the foils will be increasing, but that should be a minor contributor to the total drag. So starting at rest, you have the drag increasing, then increasing less rapidly until you reach a drag peak (know as the "hump drag"), and then decreasing with speed. Eventually, the parasite drag will increase until all it soaks up all the power available at the boat's maximum speed.

    If the boat does not have the power to get past the hump drag, then it will never be able to take off, even if it could sustain itself once foilborne. This is exactly the same issue as a boat that doesn't have the oomph to get onto the plane.

    Small foils mean the boat has to accelerate to a higher speed, incurring more hull drag, before it can lift out and the hump drag will be comparatively high. Large foils mean lifting out earlier, but the foils themselves have more parasite drag - contributing to hump drag - and they will limit the maximum speed as well as requiring more power during cruise. So sizing the foils becomes a balancing act, taking into account the lift at low speed, hump speed and drag, cruise drag, and maximum speed.

    A useful starting point is to look at the limiting factors. The drag polar will show what the maximum lift coefficient is (keeping in mind that 2D computer programs tend to over-predict maximum lift and on a 3D foil the lift coefficient is not constant along the span, so the 3D foil will start to stall earlier). The cavitation diagram will show the speed at which leading edge cavitation can begin. These factors set a minimum value for the takeoff speed. The cavitation diagram will also indicate the maximum speed that can be obtained before the buffeting, drag rise, and surface erosion from cavitation can set in.

    These cavitation speeds are indicated by the intersection of the section's curve with the intended loading line. There will typically be an intersection at a low lift coefficient that corresponds to the onset of high speed cavitation. If there is an intersection on the high lift coefficient side of things, that indicates the onset of leading edge cavitation. But if there is no intersection at lift coefficients above the maximum lift coefficient set by stall, then leading edge cavitation is not a factor at all. You pick the loading line by sizing the foil. If you need less loading, then make the foil bigger.
     
    Dolfiman, Erwan and revintage like this.
  14. masrapido
    Joined: May 2005
    Posts: 263
    Likes: 35, Points: 28, Legacy Rep: 330
    Location: Chile

    masrapido Junior forever

    Good to see you are still around and as informative as always, Tom.
     
    BlueBell likes this.

  15. revintage
    Joined: Nov 2016
    Posts: 254
    Likes: 46, Points: 28, Legacy Rep: 10
    Location: Sweden

    revintage Senior Member

    Hi Tom,
    Found the chart and the explanation to it very interesting. As I havenĀ“t learned XFoil, I took Cp-crit data from JavaFoil and plotted H105 and 63-412 to see how far off they where from XFoil. I then plotted my flat bottomed, easy to build, N12FLAT designed from using NACA0012 and manipulating it in Excel. Above CL=0.8 I was unsure about the kt scale, so it is more of guesswork. Just wanted to get an indication and doublecheck how it compared against wellknown and more sophisticated foil sections in this respect.
    speerplot.png N12FLAT.png
     
    Last edited: Jan 23, 2020
Loading...
Similar Threads
  1. Sherkin
    Replies:
    10
    Views:
    5,369
  2. TTTP
    Replies:
    7
    Views:
    2,539
  3. revintage
    Replies:
    3
    Views:
    1,223
  4. revintage
    Replies:
    11
    Views:
    1,026
  5. Niclas Vestman
    Replies:
    25
    Views:
    1,530
  6. Iridian
    Replies:
    11
    Views:
    1,186
  7. jmf11
    Replies:
    3
    Views:
    908
  8. Kenneth Dodd
    Replies:
    26
    Views:
    2,027
  9. container
    Replies:
    7
    Views:
    1,572
  10. CocoonCruisers
    Replies:
    5
    Views:
    984
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.