Keel Section

Discussion in 'Hydrodynamics and Aerodynamics' started by motorbike, Feb 21, 2015.

  1. motorbike
    Joined: Mar 2011
    Posts: 165
    Likes: 10, Points: 18, Legacy Rep: 87
    Location: Beam Reach

    motorbike Senior Member

    How does one go about selecting an appropriate section for an existing keel?

    My keel is a semi bulb shark fin from the early 1960's, it has a long chord at the base and at the tip, raked forward with a cut away at the aft middle section. From my internet searches it seems that the naca 00 series gives good all round performance, ideally from 9 to 12% and up to 15% maximum width. Assuming that I end up using that section or even a 6X series, is there is a problem the front the keel is an hourglass and from the side the position of maximum width forms a curve. Would that set up areas of pressure differential that are counter productive? Would I be better to have the maximum width parallel to the front edge and adjust the sections to suit?
     
  2. Sailplan
    Joined: Jan 2012
    Posts: 16
    Likes: 1, Points: 0, Legacy Rep: 16
    Location: Dublin

    Sailplan Junior Member

    Would it be possible to see a simple sketch of what you are describing?

    Also what kind of boat is the keel from?

    Thanks
     
  3. Karsten
    Joined: Jun 2004
    Posts: 184
    Likes: 8, Points: 0, Legacy Rep: 33
    Location: Sydney

    Karsten Senior Member

    I wouldn't worry about the section. NACA00XX will be fine and any further improvements will be minuscule. Spent the effort in optimising the plan form (side view) instead. It sounds like there is room for improvement and the gains will be much larger than getting the last 1% performance out of the section.
     
  4. motorbike
    Joined: Mar 2011
    Posts: 165
    Likes: 10, Points: 18, Legacy Rep: 87
    Location: Beam Reach

    motorbike Senior Member

    [​IMG]

    Thanks for your help, this is the keel in question. I havent got much room to change the shape shown here I could alter the fairing into front and rear somewhat, but want to make sure the foil section is at least something modern. At present its got a hollow last third and a max width seems about 20% of the chord. Lots of sanding or filling to do!
     

    Attached Files:

    • keel.jpg
      keel.jpg
      File size:
      23.3 KB
      Views:
      1,477
  5. PAR
    Joined: Nov 2003
    Posts: 19,126
    Likes: 498, Points: 93, Legacy Rep: 3967
    Location: Eustis, FL

    PAR Yacht Designer/Builder

    The foil sections used on that fin are like the same as we'd suggest. The NACA 00 series shapes where developed in the 1930's and are the same shapes employed today. There are some more modern shapes, but considering the age and likely shape of your boat (CCA era?), no new foil or plan form changes, are going to make much of a difference, in the performance envelop on you boat. It would be helpful if you told us the make, model and year of the boat.
     
  6. Canracer
    Joined: Aug 2009
    Posts: 620
    Likes: 9, Points: 18, Legacy Rep: 47
    Location: Florida

    Canracer Senior Member

    A few photographs would be awesome. Semi-bulb and a somewhat trapezoidal shape; I'd love to see some pictures.
     
  7. motorbike
    Joined: Mar 2011
    Posts: 165
    Likes: 10, Points: 18, Legacy Rep: 87
    Location: Beam Reach

    motorbike Senior Member

    I'll get some pictures next time its out, but the plan shape of the keel must be retained, I am working of an original set of plans to get as close as possible. However there is some leeway to refine the foil section provided the ballast is not altered. I was curious how one would attack it, I thought drawing horizontal lines parallel with the waterline and templating the existing section first.

    What is the effect of having the an 0012 section along the lead at the bottom and an 0009 section midway up or should the section be the same from root to tip?
     
  8. myszek
    Joined: Jan 2013
    Posts: 86
    Likes: 37, Points: 18, Legacy Rep: 22
    Location: Lodz, Poland

    myszek Junior Member

    Another, more general question. I recently worked on a new keel design for Laser 28. I analyzed various sections with XFLR5.

    With a chord about 0.8m and average speed 3m/s, the Reynolds number is 3,000,000. The keel of a normally used yacht is not very smooth, and the turbulence level near the water surface is rather high, so I assumed Ncrit=1.

    With such parameters, it proved to be very hard to obtain any laminar flow. There was no "drag bucket" for any NACA6x series foil. I also failed when tried to design a laminar or semi-laminar foil for these conditions, until it was 22% thick.

    For 22% semi-laminar foils (transition at 40-60%) there was a significant drag reduction for AoA up to 4deg. According to XFoil, they are superior over the thin NACA00 foils (while NACA6x laminar ones proved to be worse).

    Using a 22% foil section is, however, unusual, and may be risky. So, the question is: are the results by XFoil credible?
    If it is actually impossible to obtain laminar flow over the first 40% of the foil chord, it would be better to use NACA0012 or similar.
    If my pessimistic assumption about turbulence are wrong, it could be better to use NACA6x series.

    What do you think?

    regards

    krzys
     
  9. Karsten
    Joined: Jun 2004
    Posts: 184
    Likes: 8, Points: 0, Legacy Rep: 33
    Location: Sydney

    Karsten Senior Member

    To achieve laminar flow you will have to achieve three things:

    1. The surface has to be smooth enough so that the small bumps do not create turbulence. I'm not sure what surface roughness is acceptable but from memory it was not much and pretty much required a polished surface.

    2. The geometry has to be accurate. If the section geometry has dips and bumps and is different to the ideal geometry, the flow is going to go from laminar to turbulent. I would think that some form of CNC machining would be required to achieve an accurate surface geometry. If you shape the surface by hand you will need a lot of CNC machined metal templates for checking and a lot of patience.

    3. You actually have to operate the foil in the low drag bucket which means a relatively low Cl. If the foil is actually supposed to work as a lift surface it is possible to achieve a higher Cl / Cd ratio at higher Cls. Most foils (even laminar foils) are most efficient near stall. This means the most efficient foil is not going to operate in the low drag bucket when going upwind where high Cls are required. If it is operating in the low drag bucket you could improve the foil by making it smaller.

    Because of this, laminar foils are only going to operate in the laminar region when running or sailing downwind, the surface is very smooth and the foil geometry is very accurate. Upwind a regular NACA00XX is going to be more efficient because of the superior Cl/Cd ratio at optimum operating point.

    Unless you are designing a downwind flyer that is taken out of the water before each sail to polish the foils don't worry about laminar profiles.

    Laminar foils are more relevant to aircraft because aircraft have to land at low speed. To still remain in the air at low speeds, the size of the wing has to be larger than the optimum. Therefore the Cl is relatively low at higher cruising speeds and the foil can operate in the low drag region when cruising.
     
  10. tspeer
    Joined: Feb 2002
    Posts: 2,319
    Likes: 303, Points: 83, Legacy Rep: 1673
    Location: Port Gamble, Washington, USA

    tspeer Senior Member

    I think you've found out why people use NACA 4-digit sections for keels instead of NACA 6-series sections. Your results are correct. When you assume the boundary layer is essentially fully turbulent (a very realistic assumption), there's very little difference in the drag of different section shapes, so long as the flow is attached.

    I think you'll find there's a definite correlation between minimum section drag and thickness. Plot the drag from different families of sections vs thickness. You'll find the different shapes pretty much fall on the same line.

    This is why there are so many section shapes to choose from. It's because the section shape is not all that important when the flow is turbulent and the section is not being pushed to its maximum lift. And even at high lift, you can get more lift with less drag by simply increasing the area instead of going to maximum lift. If the section shape was really important, designers would have settled on a limited number of shapes long ago.
     
  11. Petros
    Joined: Oct 2007
    Posts: 2,934
    Likes: 148, Points: 63, Legacy Rep: 1593
    Location: Arlington, WA-USA

    Petros Senior Member

    if the surface is not always perfectly clean and you stay within the drag bucket range, than you will end up with more drag, less lift and poor performance.

    Unless you intend to get the last little bit of drag off the keel, and are skilled enough skipper to know how to stay within the drag bucket, I would not use anything other than the proven and reliable NACA 4 digit serirs.

    Your proposed spread of staring with 12 percent and narrowing to 10 or 8 as you get closer to the hull is a good idea and is not uncommon way to get more weight down low on the keel, improving heel angle.
     
  12. motorbike
    Joined: Mar 2011
    Posts: 165
    Likes: 10, Points: 18, Legacy Rep: 87
    Location: Beam Reach

    motorbike Senior Member

    Thanks Petros, from what I have gleaned so far for a boat such as mine the best strategy is a simple proven 00XX section, careful attention to symmetry and fairness, especially the nose to maximum width section, then ensure the tail is 3mm or less.
     
  13. myszek
    Joined: Jan 2013
    Posts: 86
    Likes: 37, Points: 18, Legacy Rep: 22
    Location: Lodz, Poland

    myszek Junior Member

    Not quite. If the keel is not adjustable (usually it'not), its AoA is also the drift angle of the hull. This produces much additional drag, so it's better to have more keel area, and less AoA - even if the wetted surface is slightly bigger.

    Another reason to have the keel area reserve is an occalosional lost of speed due to waves.

    So, the AoA of the keel rarely exceeds 3-4deg.

    Of course I did. If the boundary layer is really fully turbulent, the choice is obvious: as thin 4-digit NACA as possible.

    However, there are advantages of the thick foils. First more structural strength. Next, more space for ballast. In the case of thin foil you will need some kind of bulb to contain the necessary ballast. My collegues' experience with bulbs on the sea water is not good, they observed significantly more drag.

    So, the idea of a thick foil is temptating - if only there is any chance to obtain some laminar flow.

    I attach my results, by XFLR5, for Re=3,000,000, and Ncrit=1.

    My semi-laminar SL22 seems to be the best at these conditions, but becomes the worst with forced transition at 5%. NACA0020, of the same volume, is rather poor. NACA63-015 and -018 were the only foils with some drag bucket (although rather unstable), and they are not bad for the fully turbulent flow - but they will need a bulb. And there is NACA0012 as a reference...

    regards

    krzys
     

    Attached Files:

  14. PAR
    Joined: Nov 2003
    Posts: 19,126
    Likes: 498, Points: 93, Legacy Rep: 3967
    Location: Eustis, FL

    PAR Yacht Designer/Builder

    What is the make, model and year of this boat? Again, if she's what I think she is, foil section changes will make little difference in her abilities, simply because of hull form, wetted surface, drag, etc. associated with these era yachts.

    The best thing to do is get her smooth and symmetrical, which will provide the most bang for your effort's buck.
     

  15. myszek
    Joined: Jan 2013
    Posts: 86
    Likes: 37, Points: 18, Legacy Rep: 22
    Location: Lodz, Poland

    myszek Junior Member

    It's not so simple. Her original keel is a shoal draft one, which seriously limits her performance. Despite this, her owner achieved some good results in ORC races.
    Now he is going to improve her keel and rudder blade.

    According to XFLR5, there is noticeable difference between various keel solutions. I compared the whole hull+keel+rudder systems. Of course, XFLR5 did not calculate the wave making drag, but the friction and induced drag can be estimated.

    The difference between a sophisticated keel with the thick foil section and a rectangular one with NACA0012 section and a bulb was about 9% of the total drag (without wave drag). That means about 4% speed difference at light winds. Very important for the racer.

    Anyway, I tried to design a more reliable section for the keel. It is a modified NACA63(3)-018, 19% thick, laminar flow on 30% of the chord. In the fully turbulent flow its performance is also accaptable.

    regards

    krzys
     

    Attached Files:

Loading...
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.