Ekranoplan sections

Discussion in 'Hydrodynamics and Aerodynamics' started by sigurd, Nov 11, 2017.

  1. sigurd
    Joined: Jun 2004
    Posts: 827
    Likes: 8, Points: 18, Legacy Rep: 65
    Location: norway

    sigurd Pompuous Pangolin

    I would like to investigate WIG sections, but have no idea where to find some. I would especially like to see DHMTU (fig 1) coordinates (or how to generate them), and any sections you think would be a good starting point. Would it be possible to do something clever with Xfoil, even if it can't model ground effect?

    From literature, I've noted the following desireable attributes of Ekranoplan main wing sections:

    1) Max thickness at 20% [1]
    2) Flat underside [5], or
    3) A weak reflex tail or S shape on the underside (decreasing CP movement) [2], [3]
    4) A thin section (increased efficiency, especially at low h/C) [4]

    And these are desireable traits

    1) Continuous increase in CL as h/C decrease [3]
    2) Little change in CM with h/C (IE minimum movement of CP)
    3) Foils are sensitive to L.E. separation at low h/C, pay attention to that

    Also, can anyone explain what the thickness at 20% rule comes from? It seems the DHMTU sections I've seen have their thickness much further back.

    Attaching the only DHMTU coordinates I have found [6].
    More recently some people used a genetic algorithm to generate WIG foils, and they came out without any reflex. (fig 2) [7]

    [2] Rozhdestvensky. "Wing-in-ground effect vehicles"
    [4] Michael Halloran, Sean O'Meara. "Wing in Ground Effect Craft Review"
    [5] Nikhil, Anil, Aravind, Rahul, Sudheesh, Zahir, Antony, Manojkumar. "Investigation on airfoil operating in Ground Effect region"
    [6] Rhodesa, Sayers "Experimental Investigation: Stability Criteria of an Uncambered Airfoil in Ground Effect"
    [7] Kyoungwoo Park, Byeong Sam Kim, Juhee Lee, Kwang Soo Kim. "Aerodynamics and Optimization of Airfoil Under Ground Effect"

    fig 1a) DHMTU 10-40-2-10-2-60-21-5 XFLR5.png fig 1b DHMTU 10-40-2-10-2-60-21-5 jfoil.jpg fig 2) WIG Pareto.jpg

    Attached Files:

  2. Ad Hoc
    Joined: Oct 2008
    Posts: 5,397
    Likes: 210, Points: 63, Legacy Rep: 2488
    Location: Japan

    Ad Hoc Naval Architect

    Firstly, I admire your interest being so focused.

    However, what you're doing is no different to all those that say "what is the best hull form for....XXX" choose your poison.

    The design of a vessel/boat is greater than the sum of its individual parts, why? Because 'design' in the true sense is a series of many compromises to make the vessel/boat work, at its given SOR. It is thus very easy to highlight one such design constraint and say...hey this is not optimised, or it produces too much drag, or is structurally inefficient etc etc. And in 'isolation' those arguments are difficult to counter. But in the context of a design and satisfying an SOR, the accusation of such is pure ignorance of what is required to make a design work.

    Thus if you said you wish investigate aerodynamics of flight alone and wing sections and that was it, no reference to WIGs - fine. Plenty to review and search. But your title suggests that there is some "magical" section suited for a WIG. It is small beer in the contact of what is required to make the vessel(plane) fly..just one of many variables that constantly change, merely to satisfy an SOR.
  3. sigurd
    Joined: Jun 2004
    Posts: 827
    Likes: 8, Points: 18, Legacy Rep: 65
    Location: norway

    sigurd Pompuous Pangolin

    I have no idea what you mean, sorry. Ground effect sections have different requirements from airplane sections.
    There is movement of CP when you get close to the ground, and it normally goes the wrong way, so you get nose down pitch the lower you get. This has been solved by other vehicle geometry, such as a tail of half the main wing size, or a reverse delta planform, but also by section design. Another peculiarity is that if you have too much rocker in the the foil, you can get sucked down by venturi effect.

    NACA 4412 and 4415 seems to be the 'old school reference' WIG section, and then there are the ones I mentioned above which aren't available to me (except the one section).

    So given this, do the questions make sense now, Ad Hoc?
    A more specific objective: I want to find/make a main wing section that lets me reduce the size of the tail of an Ekranoplan as much as possible, while keeping the section L/D good.
    Feel free to contribute anything related.
  4. Ad Hoc
    Joined: Oct 2008
    Posts: 5,397
    Likes: 210, Points: 63, Legacy Rep: 2488
    Location: Japan

    Ad Hoc Naval Architect

    Of course - it was never in question.


    ...aaahh, and that goes back to my whole point!!

    You need to design your WIG in its entirety before you can answer those questions. Since how do you know you have a problem with the tail section or a bad L/D etc etc? What's the SOR?
  5. sigurd
    Joined: Jun 2004
    Posts: 827
    Likes: 8, Points: 18, Legacy Rep: 65
    Location: norway

    sigurd Pompuous Pangolin

    I want to investigate the foils I find and make in Javafoil, but there is a problem.
    JF needs you to offset the foil in order to use the Y=0 line as ground. But it still measures Cm from Y=0, not from the offset Y position. So the Cm slopes gets muddled up. What to do? Should I tell Martin Hepperle about it, or is there a reason this is the best way to calculate things?

    Fig 1 is a polar of the same foil in the same conditions, in unbounded flow field (UFF), but the green plot is offset 20% chord length upwards.

    I attach two examples of what I'm trying to do, to compare the sections at different h/C (height/chord) (fig 2 a and 2 b).
    The flattest Cm curve is the unbounded flow field. The others are 0.2, 0.1 and 0.05.

    Also Xfoil doesn't agree with Javafoil's unbounded flow field Cm (fig 3). It calculates a more positive slope for the foils I've tested so far, especially the assymetric ones - how come?

    Attached Files:

  6. HJS
    Joined: Nov 2008
    Posts: 274
    Likes: 20, Points: 18, Legacy Rep: 288
    Location: 59 45 51 N 019 02 15 E

    HJS Member

  7. sigurd
    Joined: Jun 2004
    Posts: 827
    Likes: 8, Points: 18, Legacy Rep: 65
    Location: norway

    sigurd Pompuous Pangolin

    Move Cm.jpg I'm probably explaining myself poorly. How can I translate a Cm that is taken from 25, 0 to a different X, Y coordinate?
    Last edited: Nov 17, 2017
  8. daiquiri
    Joined: May 2004
    Posts: 5,368
    Likes: 235, Points: 73, Legacy Rep: 3380
    Location: Italy (Garda Lake) and Croatia (Istria)

    daiquiri Engineering and Design

    Hi Sigurd,
    it is not difficult, if you understand the basic relationships between Cm, Cl, M and L, where:
    - L is the lift: L = 1/2 rho V^2 c Cl
    - M,x is the moment at the position x: M = 1/2 rho V^2 c^2 Cm,x
    - Cl and Cm,x are the lift and moment coefficient (at x), respectively.​

    Then consider this pic:
    Airfoil moment.gif
    When x=0.25 you have the familiar values of M,0.25 and Cm,0.25. L and Cl are independent of x. Cm,0.25 and Cl are both assumed to be known for a given AoA.

    To calculate the Cm,x (Cm at an arbitrary x position) you have to consider the equivalent moment around the leading edge:
    M,0 = M,x - x*L​
    when x = 0.25*c, the relationship is still valid and becomes:
    M,0 = M,0.25 - 0.25*c*L​
    M,0 is the same in both cases, so the two equations combine into:
    M,x - x*L = M,0.25 - 0.25*c*L​

    In terms of coefficients, the latter equation becomes:
    c^2 Cm,x - x c Cl = c^2 Cm,0.25 - 0.25 c Cl​
    which simplifies into:
    Cm,x - x/c Cl = Cm,0.25 - 0.25/c Cl​

    And there you are, the Cm,x is then:
    Cm,x = Cm,0.25 - Cl*(0.25 - x)/c​

  9. sigurd
    Joined: Jun 2004
    Posts: 827
    Likes: 8, Points: 18, Legacy Rep: 65
    Location: norway

    sigurd Pompuous Pangolin

    Thanks. Anyway, in this case it's not the X, it's the Y offsetting that screws up Cm 0.25.
    The following file is the same foil (NACA 0020) at Y=0, Y=100% and Y=-100%.
    No ground effect, just the offset. Alpha 0-4'. I don't know where it pivots the foil to change alpha. It does not use the "Modify" pivot, like I perhaps insinuated above. It's annoying because it changes the h/C per alpha in the polar, when using ground effect.

    Attached Files:

    Last edited: Dec 12, 2017
  10. sigurd
    Joined: Jun 2004
    Posts: 827
    Likes: 8, Points: 18, Legacy Rep: 65
    Location: norway

    sigurd Pompuous Pangolin

    I understand what's happening now, finally. The pivot point for polar analyzing is probably fixed at x=25, y=0. So when it increases the angle of attack of a foil that is offset upwards, it moves the foil back too. Thanks for the help, Daiquiri.

    Attached Files:

  11. Zulu40
    Joined: May 2015
    Posts: 32
    Likes: 2, Points: 8, Legacy Rep: 18
    Location: Australia

    Zulu40 Junior Member

    I think this is among the most eloquent synopsis on the nature of the problem at hand.

    Much research has been committed to WIG sections as though seeking some magic formula. When the truth is the most successful machines to date has among them the crudest and most common of sections.

    Where they agree is more a problem of of how they satisfy the geometry required, sufficient to balance the aircraft when the nature of balance is faced with such torrid circumstances as to be almost ever changing in a given sea state.

    It would be my advice to forget the magic aerofoil and focus more on the problem of balance via wing geometry of the aircraft whole, and why particular machines have succeeded and failed.

    For it is here that the human mind can develop what computers and computations fail at, ... innovation.
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.