Replace a thin low aspect Centerboard while keeping low speed manouverability

Discussion in 'Hydrodynamics and Aerodynamics' started by JohanH, Nov 3, 2016.

  1. JohanH
    Joined: Nov 2016
    Posts: 7
    Likes: 0, Points: 1, Legacy Rep: 10
    Location: Sweden

    JohanH Junior Member

    Hi, I am new here and have been searching for some input regarding redesign of my trimaran centerboard. I have a Dragonfly 28 with the original centerboard used by Quorning from 2008 to 2011. The board is very low aspect/thin section and seems far from ideal for my usage. The plan shape is almost square (1.2 m draft and 1.1 m width) with the lower aft corner rounded off to be able to swing into the case. Thickness is only 44 mm giving around 4%.
    In 2012 Quorning introduced a new centerboard with less surface but higher aspect ratio. The width was reduced to around 0.8 m. The profile was also changed to a laminar profile with maximum thickness around 50% and small radius leading edge.
    I have a friend with the new centerboard and he consider it faster when going above 7 knots. No noise and a smooth ride. When maneuvering in harbors and tacking at low speed he missed the surface area of the old board. Nothing strange with that I guess. He thinks the new design also stalls due to the extreme laminar profile with a small radius leading edge. So much that he had the profile modified last winter by the Swedish multihull designer Stefan Tornblom.

    Since the boat will be on dry land during the winter I am considering building a new centerboard. For fun and performance improvement.

    Now I am considering the planform of a new centerboard knowing that the thin section of today will be impossible to improve regarding stalling. A better thickness ratio of 8% will on the other hand require the width and area to be reduced by 50% while keeping the draft and the ability to fold the centerboard.
    Either I go for the Quorning like compromise at around 6% thickness but with a more forgiving profile ( hard to do perhaps with that thin section). Another thought I have is to make some kind of tandem or slotted design with two foils on the same board. The forward foil could be a laminar section followed by a slot of say 0.1 to 0.2 m followed by a classic NACA turbulent section. This would keep the area up while the two thicker sections with decent leading edge radius will be less prone to stall. The surface drag will of course be larger but how will the lift and induced drag compare?

    Some usage background is that we use the boat around 30% for inshore archipelago racing and the rest for cruising and just beating most other yachts around :). Speed is usually around 7 knots going upwind in low power conditions. As soon as the wind increases above around 12 knots the boat jumps up to 8-13 knots on most angles. 13-16 knots on flat water is not unusual in a breeze.

    To summarize my questions:
    1. How does a less area planform of say 50-80% compare with a 90%-ish slotted design regarding lift and drag?
    2. Are there any other radical options to consider keeping the centerboard within the 44 mm thickness and 1.2 m draft? Or should I just go conservative and live with the less area and compromised maneuvering ability?

    All input is welcome. I am a mechanical design engineer and have some experience using cad and CFD for other applications but I haven't started that part of this project yet. Thanks,
  2. tspeer
    Joined: Feb 2002
    Posts: 2,319
    Likes: 302, Points: 83, Legacy Rep: 1673
    Location: Port Gamble, Washington, USA

    tspeer Senior Member

    1. I don't see how putting slots in a symmetrical board will improve anything. A slotted section works because of favorable interference between the deflected flap and the forward element. The slot is worse than useless if the two elements are aligned in tandem because of symmetry constraints.

    2. I would go with the conservative approach. With regard to laminar flow sections, there's a lot of doubt as to whether you would be achieving much laminar flow in the first place. If this was a race boat that was dry-sailed, the board was accurately constructed to the designed shape and polished before each outing, then you might realize some laminar drag reduction. But I wouldn't expect a significant amount of laminar flow on a board that had antifouling paint and was kept in the water.

    No section profiling will restore the loss of lift at the same low-speed leeway angle that you'll lose from cutting the area. Multiply the nondimensional coefficients by the chord to get the lift and drag areas, and then compare the sections on the basis of drag area vs lift area. You'll see the smaller board has less drag at zero lift, but there will be a crossover point at which the larger board has less drag. So the question becomes, "Are you operating below or above the crossover point?"

    The idea chord size will have the board operating at the point on the two-dimensional drag polar where the lift/drag ratio is the highest. A smaller board will operate at higher lift coefficients where the profile drag is increasing rapidly. A smaller board will operate a lower lift coefficients and will have excessive wetted area. But you can only get this ideal match at one operating condition. The penalty for a too-large board is a lot less than the penalty for a too-small board.

    If you want to go radical, then you should consider that the biggest source of drag is not the profile drag but the induced drag from lift on the board. And induced drag is inversely proportional to the span. So if you really want to improve the performance, you need to go with a longer board.

    A longer board won't fit into the trunk. So what if you don't retract it all the way? You could have a longer board that looks like a skeg on the bottom when retracted, as it extends past the aft end of the board trunk. The non-retractable skeg area will require modification of your trailer, if you trailer your boat. It will mean that you can't beach it - but do you beach it now? And the longer board will put more stress on the board trunk because it will have more leverage while producing the same side force. But it's a much more likely way to improve the performance upwind and in light winds.
  3. JohanH
    Joined: Nov 2016
    Posts: 7
    Likes: 0, Points: 1, Legacy Rep: 10
    Location: Sweden

    JohanH Junior Member

    A lot of insights, thank you Tom. Glad to have that strange design eliminated by facts. When thinking about the slotted design I mostly considered it working with an angle of attack as we never go directly downwind but I guess the distance between the two foils need to be increased and the drawbacks of the design are hard to compensate. The only improvement compared with the current board would be a fuller cross section and larger radius leading edge but maybe that can be achieved with a different profile instead.
    I have had the thought of building a longer board that sits like a skeg when folded. I do store the boat on a trailer but can remove the board by diving before lifting the boat. I have yet to try beaching as we mostly have rocks here in the Stockholm archipelago. A longer board will affect my rating so that has to be considered if going that route.
    Thanks again.
  4. Petros
    Joined: Oct 2007
    Posts: 2,936
    Likes: 148, Points: 63, Legacy Rep: 1593
    Location: Arlington, WA-USA

    Petros Senior Member

    I agree with Tom, a laminar foil is usually not a good idea unless you are maintaining at every race a pure racing boat. seldom do you get much laminar flow even in ideal conditions, only the forward 3-7 percent, with an ill-advised slot, not much.

    A high aspect surface stalls sooner, low aspect ratio surface is more stall resistant. that is likely the reason for the loss of pointing ability. However the high aspect raito surface has a better L/D, low aspect surface lower L/D.

    the small leading edge usually does not so much affect stall angle, but makes for a more rapid stall with little warning.

    thinner sections actually have more stall resistance, a thicker section of otherwise identical shape will stall at a lower angle of attack. thinner sections have less drag at all angles of attack. So You always want to run the thinnest section you can structurally tolerate. IOW, the maximum load on the surface, and strength of the material determines thickness.

    Larger surface will improve low speed pointing, but will have increased drag at higher speeds, so size is always a trade off between low speed ability and high speed drag.

    slots are general misunderstood, they increase drag, lower L/D, but delays the stall to angles of attack higher than with out the slot, and why they are only used for low speed extra lift, as in landing an aircraft with a smaller low drag wing at lower speeds. not efficient, but it is just for landing.

    pick a good performing non-laminar foil section, as thin a section as the material will tolerate, and I like high aspect ratio surfaces, but you must choose between stall resistance and L/D (higher speed performance). Same with size, just big enough to point high in a low speed tack, but no larger than necessary.

    good luck.
  5. SukiSolo
    Joined: Dec 2012
    Posts: 1,270
    Likes: 27, Points: 0, Legacy Rep: 271
    Location: Hampshire UK

    SukiSolo Senior Member

    I concur, having altered a few boards to have a larger lateral area. The main symptom of too small a board is stalling coming out of a tack at low speed, and pointing fine but not actually achieving any VMG close to the point angle. The latter is most noticeable when racing against similar craft even if in a restricted (not one design) class.

    It actually does not seem to matter too much in practice if you widen the board if you cannot make it longer. So an elliptical (side profile) can be made somewhat more 'square' and it does work. Also just widening (increasing the side profile size) the board if it will still fit in the case. There are some very high a/r boards that stall from being too extreme. Again, especially with rudders these stall too easily coming out of a tack. If you can get a bit more length than that is generally preferable, but more total area, in practice seems to work perfectly well.

    As little as 50 X 50mm can make the difference and certainly 100 X 100mm so 4" X 4" in imperial. Check the total lateral area of old and new boards. Personally I don't think a 50% aft chord max thickness with a very fine nose is ideal. You need to do a little more hunting at profiles that might work in your speed range. My 'gut' feel would be closer to between 38 to 42% with a fine but not extreme parabolic 'nose' but I would go through the profiles and test it on X foil too. Again although I have read a lot of information about the 8% and near thickness, I have used thinner % profiles that work OK as long as they have sufficient stiffness, not quite 4% but certainly 6 - 6.5%. Pretty hard to find low speed data for thin foils though...

    I'm sure Tom is right though, so I'm just adding what I have found from a combination of going through available data and emperical trial in the real world.
  6. JohanH
    Joined: Nov 2016
    Posts: 7
    Likes: 0, Points: 1, Legacy Rep: 10
    Location: Sweden

    JohanH Junior Member

    Thanks again for your input. There are so much to learn and things to optimize now when we have half a meter snow here in Stockholm.

    I got a photo with the new original board on top of the old original DF28 board. It is about half the area as can be seen on the photo. direction of travel is to the right. The max span is 1,3 m and max thickness is 44 mm.

    I have now also done some simulations in xflr5 with different potential foils and planforms but I am just starting. Very interesting this area of engineering.

    As the tip vortex is not calculated in xflr5 what shape do you suggest for the tip? The larger centerboard on the photo indicates the maximum area available and still be able to fold the board.
    Intuitively the tip vortex for the smaller shape will be higher up than the max span of the board even though it is far better than the large board. Is there a way of optimizing this further?
    What I am aiming for is a slightly larger area planform than the new original board in order to get some better low speed lift. Any more thoughts?

    Attached Files:

    Joined: Jun 2014
    Posts: 166
    Likes: 1, Points: 18, Legacy Rep: 31
    Location: Land O' the Great Lakes

    UNCIVILIZED DIY Junkyard MadScientist

    If you wanted to keep drag from the slot down, you could make the board more L-shaped. So that when it's fully down, the trunk is fully filled.
    And while I know only the most basic things about foils, aren't fatter foils more stall resistant? In addition to being easier to reattach the flow to?

    Also, the shape of the new board is more in keeping with the trend towards a higher aspect ratio, so as to better generate lift. Though as to where the discussed crossover point of lift vs. board area is, I haven't any idea. How does one figure this?

    One thing which comes to mind when experimenting with new boards is something from keelboat racing. Where the trimmer(s), primarily the headsail trimmer, needs to be continually communicating with the helmsman after coming out of a tack. Letting him know where boat speed is at, ditto on how far out the jib is. And to keep feeding him info as speed rises & the jib is trimmed in. Since such boats are slow to accelerate after a tack. And if you try & come up too fast, you make a lot of leeway, along with poor boat speed.

    So good instruments would help you to determine how well a new board does when tacking. As well as upwind. Since they'll also assist you in measuring leeway, whether your pointing is good or bad. Along with, of course, how well the boat points once she's fully in upwind mode after a tack.
    You'll also want to track performance both through, & after tacking. As well as leeway, & performance in different conditions: wind, & wave. So that you know how it does relative to the old board.

    I'm assuming that changing them with the boat in the water is possible?

    The other thing which comes to mind is that if you did build a board that's too long to fit into the trunk when retracted. Could you perhaps build it so that the portion of the tip which sticks out of the trunk when it's up is sacrificial? Or that at least part of the board is? Obviously the shape of both parts wouldn't be as ideal as if it were a long daggerboard, which with the loss of it's tip leaves it with a decent plan form. But maybe it would work?

    Why on the new board is the tip cropped at an angle like that instead of perhaps having an eliptical, or squared off tip?

    Ah, I almost forgot. If the board is easy to remove while the boat's in the water, you could build a "cassette" for the trunk that houses a daggerboard, if you wanted to experiment with such profiles. That way you'd just raise & lower the daggerboard as needed when beaching, or launching. And it would also be easy to experiment with different boards that way.
  8. markdrela
    Joined: Jun 2004
    Posts: 307
    Likes: 30, Points: 28, Legacy Rep: 324
    Location: MIT Aero & Astro

    markdrela Senior Member

    Why did you cut the bottom end at that angle??? That looks like a terrible tip design. The angled aft face will act as a massive blunt trailing edge, and increase the profile drag. The lost lifting area will also reduce the effective span, increasing the induced drag.

    On a low-AR surface, a simple square tip will give close to the optimum spanwise loading -- no fancy shapes are needed. You can also add a tip "endcap" which is the tip airfoil formed into a half-body of revolution. This eliminates the sharp edges and any separation they might produce at an angle of attack.
  9. JohanH
    Joined: Nov 2016
    Posts: 7
    Likes: 0, Points: 1, Legacy Rep: 10
    Location: Sweden

    JohanH Junior Member

    I have thought about that and it lookes like most of the slot except the last 0.2 m can be covered with a L-shape.

    I think it is the compromise between light wind and strong wind in my case. The larger board clearly had better L/D when racing in 1-2 knots this summer but is hopelessly noncompetitive in 7-15 knots, boatspeed that is.

    Unfortunately not. Will need to dive and have a lot of lead attached to the centerboard to be able to change in the water. Need a crane when on land which is rather expensive.

    I think it is to get as much area as possible from the lower part of the board. The reason for the angle is that that is the largest possible and still being able to fold into the trunk.
  10. JohanH
    Joined: Nov 2016
    Posts: 7
    Likes: 0, Points: 1, Legacy Rep: 10
    Location: Sweden

    JohanH Junior Member

    It is not me it is the Quorning boatyard that designed both the large and the small board on the photo.
    I am going to design and build (in carbon to save some weight) my own board. I am curious about the angle of the tip and how it can be improved. I think the backward angle is to keep the span as large as possible with the given geometry of the current board/trunk. The angled aft face of the original large board is to just clear the aft edge of the trunk when folding in the board.
    Would a square tip design improve the performance even if the span has to be reduced by 80 mm or 6%?

    This is a similar boat by the way:
  11. markdrela
    Joined: Jun 2004
    Posts: 307
    Likes: 30, Points: 28, Legacy Rep: 324
    Location: MIT Aero & Astro

    markdrela Senior Member

    The outermost part of the span doesn't help much if it can't efficiently carry lift. Because of the angled tip cut, the airfoil at the extreme tip is basically a 2x4 with a nice leading edge -- that's not a good lifting shape. But it's surely better to fix the tip region than to cut it off.

    The trunk clearance constraint is a severe design bind in this case. Looking at the photo... possibly the best compromise is to use the LE outline of the original black board, and the TE outline of the new brown board. That will leave sufficient chord at the very tip to give a reasonable spanwise loading shape. I'd also taper the airfoil thickness rapidly over the bottommost part of the span, so the angled cut can be made into a normal sharp trailing edge.

  12. JohanH
    Joined: Nov 2016
    Posts: 7
    Likes: 0, Points: 1, Legacy Rep: 10
    Location: Sweden

    JohanH Junior Member

    Thank you! That confirmed my thoughts about the tip design. I will make sure all planform also have a decent profile even in a short region of the board.

    I am considering something like that. I only have to take the center of effort into account as well. The boat is very well balanced today and moving the CE forward might result in windward helm. I might have to move the design backward a bit loosing some of the span from the LE part. Any thoughts on that?
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.