Question about wingsail foil shape

Discussion in 'Hydrodynamics and Aerodynamics' started by satrams, May 6, 2014.

  1. satrams
    Joined: May 2014
    Posts: 13
    Likes: 0, Points: 0, Legacy Rep: 10
    Location: Rochester, NY

    satrams Junior Member

    Could someone help me to determining what foil shapes the AC45 wingsail has? I am going to try to build a sailboat style (hike-able with tiller) land-yacht and want to build a wingsail modeled after the AC45 (about 1/4 scale).

    I have read different generalized descriptions that talk about NACA foils- maybe an 0018 to 20 on the front element and like a 0012 on the flap, but from what I see and have tried in the simple programs like Foilsim, this isn't right.

    I got closer doing a modified 00, ie, 0019-43, but it's still too rounded in the latter half. Their foils look to go towards a much straighter taper for the back 1/2 to 2/3.

    I am going to include a close up of the very generalized rough representations of the shapes that I got from one of the plan line drawings- I know they're not meant to be exact, but they look to me to be pretty similar.
    [​IMG]
     
  2. tspeer
    Joined: Feb 2002
    Posts: 2,319
    Likes: 303, Points: 83, Legacy Rep: 1673
    Location: Port Gamble, Washington, USA

    tspeer Senior Member

    The sections weren't NACA sections, but were quite similar. The scale effects between the AC45 and your model are such that you'd be better off picking sections that are appropriate for your model's conditions.

    The sections used on the AC45 wingsail were actually two airfoil shapes hinged together at approximately 90% of the forward section's chord, so that when the flap (second section) was deflected, it formed a slot between the two shapes. Each element was 50% of the total chord.

    Foilsim is not capable of analyzing such a configuration that includes a slot. You need to use something like Javafoil. You'll get a pressure distribution that looks like the attached plot. However, you shouldn't expect to get the same lift at model scale, because the wing will stall earlier.

    For your model a NACA 0012 would work OK for the forward element. You will want something thinner, like maybe a NACA 0009 or less, for the flap section. A NACA 0018 is probably too thick at model scale - it proved too thick when some friends and I built a full scale landyacht using NACA 0018 sections in the wingsail, although that wing had a much smaller flap chord. There's no point in going beyond 25 - 30 deg with the flap deflection. For a model, I'd expect you'll get your best performance around 15 - 20 deg.

    Twist is important in getting the most out of a wingsail in high winds. If the yacht is overpowered, you'll want to twist off the flap at the head so you can keep the bottom of the wing powered up. It may be difficult to build a model with a flexible flap like that, however, and it'll depend on how complicated you want to make your control system.

    If you only have one servo for wing trim, then you'll want to have either a string from an arm at the foot that limits the flap travel (which is how the real AC45 does it) or you'll want adjustable stops. The wing should be free to hinge so it pops through to being cambered the opposite way when you tack, with the amount of camber/flap deflection being set by the stops. The servo should control the rotation of the whole wing with the sheet attached to the clew of the flap.
     

    Attached Files:

  3. tspeer
    Joined: Feb 2002
    Posts: 2,319
    Likes: 303, Points: 83, Legacy Rep: 1673
    Location: Port Gamble, Washington, USA

    tspeer Senior Member

    BTW, at the speeds landyachts sail, you don't want to be hiking. Your landyacht will easily be capable of exceeding the speeds of the real AC45. There will be less windage if you are in an enclosed cockpit. And you need to be wearing a seat belt and a helmet.
     
  4. lohring
    Joined: Nov 2006
    Posts: 57
    Likes: 3, Points: 0, Legacy Rep: 32
    Location: Eugene, OR

    lohring Junior Member

    There is a lot of information on the Yahoo wingboats forum. Below is a generic wing sail plan and control system. The estimated co ordinates are also posted for use in java foil.

    Lohring Miller
     

    Attached Files:

  5. satrams
    Joined: May 2014
    Posts: 13
    Likes: 0, Points: 0, Legacy Rep: 10
    Location: Rochester, NY

    satrams Junior Member

    Thanks very much for responding guys. I mis-spoke- I do know about and have experimented with JavaFoil using the front element and flap of 50% (equal front and back as the AC45 is). I understand how the AC45 wingsail (and others) work, including the slot and the way the flap is hinged.
    I also know from reading info you guys have posted in the past about how the twist system works and is rigged.
    I am not building a model, but a relatively simple and lo-tech 3 wheeled vehicle with approximately a 15 to 17' wing that I was planning to pretty much copy from the AC45 and scale down. I was going to try to use the twist system.
    I have a nice 2 piece lightweight aluminum mast from a sailing dinghy we used to have that I plan to use as my internal 'spar'.
    I realize that a land-yacht can potentially go much faster, but that is not my primary goal. I am not interested in building a 'fuselage' type body and sitting in the cockpit. I am a sailor. I am after somewhat of the experience of a sailboat on land- and am unconcerned about attaining high speed. As I said, I plan to steer using a tiller and want to at least have the possibility of heel and hiking, and will sit attop a basic platform or partial tramp arrangement perpendicular to the vehicle.
    It may be the case that this wing would be too much power- that I am not that sure about at this point and am partially guessing and going on instinct.
    I will not be sailing this in gigantic open desert or pavement areas but local parking lots including a not-so-big one behind my house.

    Tom- I don't know if you were able to see the drawing I posted of the AC45 shapes- but as I said they seem to have a straighter curve towards the latter half or even 2/3 of the foils- defenitely different than a standard NACA foil, which I can't seem to re-produce in JavaSim, and I'm assuming they must have had some pretty good research in on the topic given their huge budget. :)
    So what do you think?
     
  6. lohring
    Joined: Nov 2006
    Posts: 57
    Likes: 3, Points: 0, Legacy Rep: 32
    Location: Eugene, OR

    lohring Junior Member

    Again, look at the Yahoo forum. They have complete plans for simple land yachts with wing masts as well as tests and construction information.

    Lohring Miller
     
  7. satrams
    Joined: May 2014
    Posts: 13
    Likes: 0, Points: 0, Legacy Rep: 10
    Location: Rochester, NY

    satrams Junior Member

    OK, I'll check it out. Thanks.

    Actually, I just checked out the attachments- thank you Loring, sorry I didn't look at them before responding. They are very helpful, especially the Open Wing plan (I've seen some of the Patient Lady info but not this overall one). (I would still like to know the exact shapes of the AC45 foils, but it doesn't matter that much- and these are definitely closer to what I was looking for.)
     
    Last edited: May 12, 2014
  8. satrams
    Joined: May 2014
    Posts: 13
    Likes: 0, Points: 0, Legacy Rep: 10
    Location: Rochester, NY

    satrams Junior Member

    One more question to Tom. Sorry, Tom, I am new to this stuff, but the thumbnail that you posted- is that something that was generated by you, or is this taken from somewhere regarding AC45 info? (because they do look like the actual shapes)
    I guess what I'm asking is- do you know what the foils are in this particular thumbnail and/or does the title description contain some info regarding this, or possibly none of the above? :)
     
  9. tspeer
    Joined: Feb 2002
    Posts: 2,319
    Likes: 303, Points: 83, Legacy Rep: 1673
    Location: Port Gamble, Washington, USA

    tspeer Senior Member

    The sections for the trimaran USA 17 were designed with flat contours in 80% of the flap and 40% of the main element. This was intended to reduce the cost of building the wing.

    It turns out that flat is a pretty decent choice for the pressure recovery portion of a section. Convex can result in too rapid an increase in pressure at the trailing edge, leading to premature separation. Concave can enable high lift, but tends to separate all at once, and can have large amounts of separation prematurely if the forward parts of the section aren't performing as predicted. A flat contour is a good compromise between these two extremes, as it results in a concave pressure distribution for the recovery but still promotes a more gradual progression of separation that starts at the trailing edge and moves forward with angle of attack.

    I don't know what the design approach was for the AC45 sections. I think they were based on a combination of NACA sections and experience with the trimaran.

    The sections for the AC72 were intended to maintain longer runs of laminar flow, if laminar flow could be obtained, without incurring any significant penalties in drag or maximum lift over a fully turbulent design if laminar flow could not be attained. Although the aft portions of the AC72 sections were not intentionally designed to be flat, they turned out to be fairly flat.

    The construction of the wing surface is important, too. A rigid surface (like Aethon's wingsail) can maintain its shape, but a film-covered surface will not. For film, the spanwise tension tends to maintain the section shape of the ribs across the bay, but the chordwise tension tries to pull the film into a straight line between the trailing edge and the end of the D tube. The resulting shape is a blend between the two, tending to the rib shape at top and bottom of the bay and being flatter in the middle of the bay.

    Then there's the distortion caused by the pressure loads. What was seen on the AC72 was considerable dimpling of the film on the windward side, but the section contours were remarkably smooth on the lee side. It turned out that the pillowing effect of the lee side suction was just about offset by the chordwise tension, and the light had to be just right to see the small dimpling of the film at the ribs. The end effect was additional camber in the aft part of the section, which was actually helpful. The effect could have been enhanced if the aft spar had been designed to avoid distortion of the dimpled windward side film and the edges of D tube and trailing edge structure designed to provide a smoother transition between the hard shell and film contours.

    In any event, the precise shape of the aft part of the sections is not a big driver to the performance of the section. The leading edges see a big increase in local velocity that greatly magnifies the skin friction there, to the extent that the leading edges dominate the total profile drag. As long as the flow is attached, you're not likely to experience any noticeable difference in performance for different section shapes.

    I'd still give some thought to the safety aspects of a hiking posture for a landyacht. 30 kt road rash is painful.
     
  10. satrams
    Joined: May 2014
    Posts: 13
    Likes: 0, Points: 0, Legacy Rep: 10
    Location: Rochester, NY

    satrams Junior Member

    Yes, you're right, it certainly could be dangerous. I would certainly take it in baby steps. And I also had the thought of possibly attaching 'training wheels' during testing- maybe longer arms that come off the side that either have rollers or skids on them.

    This info you just imparted is very interesting. So that thumbnail of a pressure distribution you gave me that is labeled AC45 is not necessarily the AC45 shapes? They actually look pretty close.

    Yes, I am really am mostly a complete novice about this stuff- I am a good sailor and have been doing a lot of reading about all these wing aerodynamic topics this winter, (plus watching all the America's Cup and studying up on any pictures, articles, diagrams, etc. I could find), but my physics and math skills are pretty limited. But, I have a super healthy interest, curiosity, and desire to learn about it, at least enough to try to build the wing, know how it works, and the decisions to make about construction as regarding real-world performance- just as you have talked about above.

    Thanks for steering me to worry most about the leading edge. I was figuring that was probably the case. This is going to be the trickiest part for me- as I don't have the materials or construction techniques that most of you guys use to apply to this stuff.

    You guys will probably roll your eyes and think I'm crazy as I go along here, but this is going to be a very unconventional project as compared maybe to what is usually seen among your peers. (Probably no fiberglass, or the like- I have almost no budget, but I do have a lot of potential materials around our family house and workshop to pull from)

    As of now- (and it could be a complete failure) I am going to try using cardboard and rigid foam sandwiches to make ribs out of (probably side edge-surfaced with thin plastic), and possibly thin wood as bottom plates and such. As I said, the leading edge section will be the toughest to pull off and where I am going to have to experiment the most.
    I have a concept right now where I may attempt to use a type of aluminum sheeting- like flashing or gutter material, to form a shell in the leading edge shape. Attaching it and properly forming it into the correct shape are the challenges that I do not yet know if I can be successful with. Obviously, weight is an issue.

    Just how smooth does the surface have to be where the front sections transition into where you would use the shrink plastic? I guess they used double-sided tape to attach the plastic on the AC45s (and maybe 72s)? Can you have a bump of tape where that is or does everything have to be flush? For instance- could I have tape bridging the spot where the leading edge piece stops and the the sheathing begins? These are all the more nitty-gritty mysterious details. :)
     
  11. satrams
    Joined: May 2014
    Posts: 13
    Likes: 0, Points: 0, Legacy Rep: 10
    Location: Rochester, NY

    satrams Junior Member

    P.S.- sorry, I will stop asking you specifically about the AC45-72 shapes. Obviously you worked with Oracle and helped design them!- and I know they are proprietary. The tips and ideas that you have given me are great and are much appreciated. How awesome that someone like me can go online and ask questions of a super-experienced aeronautical engineer. :)
     
  12. satrams
    Joined: May 2014
    Posts: 13
    Likes: 0, Points: 0, Legacy Rep: 10
    Location: Rochester, NY

    satrams Junior Member

    Re-reading your last post reminds me of another question I have had ever since I learned about the twist system of the AC45s and others, and then realized the flap sections themselves had to twist from top to bottom to allow the twist to happen. So that must have been (and would be) quite tricky to get the right balance between rigidity and flexibility.

    So thinking about this, the main thing that puzzles me besides, is how can the sections twist without the film getting deformed or pulled off? I guess the obvious answer has to be that it has to be able to stretch enough to accommodate that twist? Because generically, when the section twists, the plane of the one side wants to move down and the plane of the other wants to move up, if I am not mistaken? Do they actually allow the ribs to twist in the design? (or I guess they really have to)
     
  13. Ben G
    Joined: Jun 2013
    Posts: 49
    Likes: 0, Points: 6, Legacy Rep: 10
    Location: Brisbane

    Ben G Junior Member

    Hi Mr Satrams,
    while I love crazy inventions, be careful with rear wheel steering. It's not inherently stable; As the tyres start to scrub sideways, the rotation centre of your kart will move ahead of the front axle causing oversteer. Front wheel steering with a nice understeer will be the safe way to go

    Cheers,
    Ben
     
  14. tspeer
    Joined: Feb 2002
    Posts: 2,319
    Likes: 303, Points: 83, Legacy Rep: 1673
    Location: Port Gamble, Washington, USA

    tspeer Senior Member

    It was from the head of the AC45 wing. The sections were thicker lower down.
    The thickness of the tape and film was not important because the flow would have transitioned to a turbulent boundary layer there and it was not a region that had a large adverse pressure gradient. So the boundary layer characteristics at the D/film junction were not critical.

    The thickness of the lettering was right at the limit of what could be tolerated by a laminar boundary layer. I don't have any data on what the transition locations really were.

    There's some good information on the practical aspects of designing for laminar flow in The Leading Edge.
     

  15. satrams
    Joined: May 2014
    Posts: 13
    Likes: 0, Points: 0, Legacy Rep: 10
    Location: Rochester, NY

    satrams Junior Member

    To Ben G: No, Ben, I will be using a tiller but it will be linked by cables to the front wheel- so no worries, I will be using front wheel steering. :)
     
Loading...
Similar Threads
  1. Erwan
    Replies:
    6
    Views:
    3,650
  2. dustman
    Replies:
    25
    Views:
    3,832
  3. Alexanov
    Replies:
    2
    Views:
    887
  4. tropostudio
    Replies:
    30
    Views:
    5,376
  5. nzl51
    Replies:
    5
    Views:
    1,858
  6. Brian Needham
    Replies:
    21
    Views:
    4,507
  7. massandspace
    Replies:
    11
    Views:
    2,203
  8. Southern Cross
    Replies:
    32
    Views:
    5,597
  9. laukejas
    Replies:
    40
    Views:
    8,535
  10. cristianon
    Replies:
    5
    Views:
    5,431
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.