Telescopic Wingsail - Why Not?

Discussion in 'Sailboats' started by David Cooper, Apr 7, 2018.

  1. David Cooper
    Joined: Jan 2015
    Posts: 153
    Likes: 10, Points: 18, Legacy Rep: 42
    Location: Scotland

    David Cooper Senior Member

    I've seen a set of colourful toy cups which fit into each other like Russian dolls. What's to stop a wingsail being designed the same way? Change the shape of those cups from round to aerofoil shape (and I'll continue to refer to them as cups for the sake of brevity), then imagine pulling the first one up until its base reaches the top of the next smallest cup, and repeat this for the second cup with the third, and so on all the rest of the way down. You end up with a full-height wingsail, only it's much better than that because it can be reefed.

    If there's a hole through each lid/floor for the mast to go through, the cups will be able to slide up and down it, but an alternative would be to have as many mast sections as there are cups, each section of which would be locked to the top of one of the cups, and there are advantages to this, particularly when it comes to fitting and removing the mast (no big crane needed), and for transportation. Reefing can reduce the wing size right down to the height of the lowest cup, making it easy to continue sailing in extreme conditions, and a telescopic mast gains strength as you reef due to internal reinforcement as the sections telescope together, allowing you to continue sailing in stormy conditions without needing stays to hold the mast up.

    For a proper wing sail, you need two sets of these "cups", but that's not impossible to arrange - the linkages are built around the top of each adjacent pair of cups and simply stack together when the cups are telescoped in, and it's easy to shape them so that the spacing between front and rear cups remains constant all the way up even though the higher cups are smaller than the ones below them. Flexibility in the telescoping of the aft set of cups makes it easy to support twist in the back half of the sail. At the top end of the unreefed sailing range, the top section of wing can be set to push air the wrong way to improve RM (reducing stress on the mast in the process), although it may be more efficient to reef and carry on with a less flattened sail, in which case this ability to adjust RM with the top of the rig would simply be used to avoid shifting weight low down during gusts.

    The base of each cup (in the front set) would have holes in it for control lines to run through, these being used to raise and lower the masts and sail, and to control each link between front and rear cups. Sensor wires could simply coil up in loops as the sail is lowered, and the same would apply to power lines if you build photo voltaics into the sail - I can imagine this being used for ocean cruising as well as racing because it's very robust when fully reefed and it's simpler to reef and unreef too (quite apart from any performance gains).

    Control systems: the mast would rotate to control the front half of the wing, and all you need to do is let the wind push it until it hits lock positions which it can't pass - the lock positions would be changes for upwind and downwind, while an emergency button would release them to dump all power in an instant. All the other control is done at the connections between front and rear cups, and with these it's a matter of pulling them against the force of the wind and being able to press a button to let them back out a bit (or a lot) to reduce (or dump) power. Perhaps other controls are needed for sideways movements at these connectors to adjust the slot - that's beyond my knowledge because most of what I know of how wingsails work comes from watching them on TV, so feel free to step in and fill in the gaps for me. The big question is, is this viable? If not, what are the problems with it that I haven't addressed?
     
  2. Doug Lord
    Joined: May 2009
    Posts: 15,374
    Likes: 200, Points: 73, Legacy Rep: 1362
    Location: Melbourne, Florida

    Doug Lord Flight Ready

    Already been done......

    Bene-


    "budjawee"-
     
  3. David Cooper
    Joined: Jan 2015
    Posts: 153
    Likes: 10, Points: 18, Legacy Rep: 42
    Location: Scotland

    David Cooper Senior Member

    Thanks Doug, those would certainly suit cruising, but are they also fast on a race course? The second one of those has to work with the air flowing the opposite way across it for each tack and can't change shape in the way I've discussed, but the first one (the soft wing) could be adapted to provide more shape control, so it might be able to get close to the same level of performance. I don't know how important the slot is between the front and rear sections of wingsails though. Aiming for something that duplicates the shape of the wingsails used on AC boats is likely the best approach for performance, so my aim was to think about how to make a telescopic version of that.
     
  4. markdrela
    Joined: Jun 2004
    Posts: 305
    Likes: 28, Points: 28, Legacy Rep: 324
    Location: MIT Aero & Astro

    markdrela Senior Member

    The main problem I see is that this is a horribly inefficient structure for carrying the bending moment imparted by the airload.

    The overall bending moment is zero at the top, and increases rapidly but smoothly to the bottom. But the bending moment on each segment must have a "tent" distribution, such that all these overlapping "tents" superimpose to make up the smooth total curve. The very rapid variation of bending moment on each segment then results in immense shear loads on the segment --- far larger than the shear loads on a single spar. If you have N segments then the shear loads are approximately N times greater.

    An airfoil-shaped hollow shell is terribly inefficient at carrying both bending moment and shear loads, since it is prone to various instabilities. Two examples...

    1) The bending load will try to collapse the airfoil shape of the shell, like a drinking straw collapses into a "hinge" when you bend it. On a normal spar this collapse is prevented by the spar web, but this wing can't have a web and still retract. Only the topmost segment can have a web, but there it's needed the least.

    2) The shear load will try to wrinkle the shell, like a paper tube will wrinkle and collapse if twisted. On a normal spar the shear web efficiently takes the shear load, and its shear load is much smaller to begin with.

    The overall consequence is that such a wing will require massively more structural material than a conventional wingsail, just to control the various buckling and wrinkling instabilities. And that's before the mechanical issues with sliding joints are considered.
     
  5. David Cooper
    Joined: Jan 2015
    Posts: 153
    Likes: 10, Points: 18, Legacy Rep: 42
    Location: Scotland

    David Cooper Senior Member

    I had thought that if you have a decent amount of overlap between two sections of mast, the strength would be the same while the cost would be extra weight. Clearly it isn't possible to have decent overlaps if the wing sections are severely short in height (as in the second video Doug linked to), but if you were to make the wing out of just five sections (which is what I had in mind), the mast sections would still be long enough to have a substantial overlaps when fully extended (although when they're fully down they would still stick up above the deck a long way, but probably not enough to cause trouble on a mooring in a gale in the way that a wing mast does, and it would also be possible to take the entire thing down without needing a crane, and to store it in an ordinary shed rather than having to build an aircraft hangar for it).

    "An airfoil-shaped hollow shell is terribly inefficient at carrying both bending moment and shear loads, since it is prone to various instabilities."

    The mast would take those loadings. The airfoil sections would only need to support the forces acting directly on them from the wind blowing over their surface, and the tops and bottoms of sections would be well supported by horizontal plates, though other horizontal plates/frames could be present in between those, stacking together when the sail is down, but pulled out at even spacings when the sail is raised. In the first video Doug linked to, it appears as if a soft sail can hold the required kind of shape with very little structural support from widely spaced battens.

    "1) The bending load will try to collapse the airfoil shape of the shell, like a drinking straw collapses into a "hinge" when you bend it. On a normal spar this collapse is prevented by the spar web, but this wing can't have a web and still retract. Only the topmost segment can have a web, but there it's needed the least."

    Any bend in the mast would lead to a section of wing slightly telescoping inside the one below it on one side while being pulled slightly further out on the other. Each wing section would be a fair bit narrower than the one below, and the fewer there are, the thicker each can be, so there is an option to provide some reinforcement on the inside (such as thicker vertical parts needed when one end of a section doesn't have a plate inside it at exactly the same altitude).

    "2) The shear load will try to wrinkle the shell, like a paper tube will wrinkle and collapse if twisted. On a normal spar the shear web efficiently takes the shear load, and its shear load is much smaller to begin with. "

    Again, it's the mast that takes these forces rather than the wing that it carries, which leads me to wonder if you originally missed the bit about there being a mast involved. What I'm envisaging would work in large measure with a soft sail too, but there would be no need to roll it up, so it could be made from rigid material.

    "The overall consequence is that such a wing will require massively more structural material than a conventional wingsail, just to control the various buckling and wrinkling instabilities. And that's before the mechanical issues with sliding joints are considered."

    It shouldn't need much more structural material than a conventional wing - it's primarily the mast where that would be needed. The inflated wingsail thread in this subforum shows a boat with a telescopic mast that clearly can do the job (at least in light winds), and it looks as if it uses a similar number of mast sections to the number I'm talking about using (link below to video). You can see it bend quite a lot, but it would also be possible to use stays with a telescopic mast, so long as its height is only changed when the sail isn't under high load, and if a system can be devised to keep an inner support plate level with each end of each section at all times, it might even be possible for the stays to attach at 3/5 way up the mast rather than at the top. (It would also be possible for the stays to be released in an emergency so that the sail can swing to any angle, and then to tighten them up again after coming head to wind.)

    VIDEO: Is This The Future? >> Scuttlebutt Sailing News http://www.sailingscuttlebutt.com/2017/08/22/video-is-this-the-future/?utm_medium=email&utm_campaign=Scuttlebutt%204902%20-%20August%2023%202017&utm_content=Scuttlebutt%204902%20-%20August%2023%202017+CID_408c7bd787eff9f777ca6f3a5c64b72c&utm_source=Email%20Newsletter

    One bit I got wrong before is with the linkages between the front and rear halves of the sail - the upper ones will need to contract to narrow the slot higher up as the sail extends, and the slot will have no option other than to widen there as the sail is reefed, so slot-width controls must be added.
     
  6. markdrela
    Joined: Jun 2004
    Posts: 305
    Likes: 28, Points: 28, Legacy Rep: 324
    Location: MIT Aero & Astro

    markdrela Senior Member

    I skimmed through the video, and that didn't show a mast.

    You could put a rib at the bottom of each section, with each rib sliding up a fixed untapered mast. But this still has the fixed mast.

    A segmented telescoping mast would need to overlap at least 3-4 diameters at each joint. Otherwise the shear loads and "crowbar" hoop-burst loads in the mast would get so big that the associated material would weigh more than the primary bending material.

    I'm not saying this can't be done. It just seems very heavy and very complicated mechanically.
     
  7. PAR
    Joined: Nov 2003
    Posts: 19,133
    Likes: 467, Points: 93, Legacy Rep: 3967
    Location: Eustis, FL

    PAR Yacht Designer/Builder

    Weight and contrivance alone should discount this sufficiently enough, to knock it off the what if list.
     
  8. David Cooper
    Joined: Jan 2015
    Posts: 153
    Likes: 10, Points: 18, Legacy Rep: 42
    Location: Scotland

    David Cooper Senior Member

    If you mean the inflated wing video, they don't show it directly, but you can see what its minimum height must be at the start - they make no attempt to lower it completely to deck level.

    "You could put a rib at the bottom of each section, with each rib sliding up a fixed untapered mast. But this still has the fixed mast."

    I originally had ideas about a two-section mast with 8 wing sections, the top ones sliding up the wider section and onto rings (that never go lower than the top of the lower mast section) which they'd lock into to slide up the narrower section, so there are many options. A big advantage of using several telescoping mast sections though would be that the top part of the wing sail can be substantially narrower than the lower parts of the mast - you want each mast section to be as wide as the wing section that it supports to maximise strength, and this allows greater strength low down where it's most needed.

    "A segmented telescoping mast would need to overlap at least 3-4 diameters at each joint. Otherwise the shear loads and "crowbar" hoop-burst loads in the mast would get so big that the associated material would weigh more than the primary bending material."

    I started this thread primarily to guard against someone getting a patent for any of the ideas in it so that I'm not blocked from using the idea on my own boat designs. I'm thinking about using two small wingsails, one on each hull of a 12ft foiling cat. There would be five mast sections 7ft in length each to carry five wing sections, each 5ft tall, adding up to 25ft of vertical wing. With the sail fully reefed, it would be left standing to the height of a person. This allows for 2ft overlaps where the mast sections join.

    "I'm not saying this can't be done. It just seems very heavy and very complicated mechanically."

    It could make the C class more affordable, and if the AC boats were required to have a telescopic wingsail, that could drive the development of such technology and could reduce their costs too over time, as well as increasing the wind range in which they can race. I also started thinking about this because of Randy Smyth's wingsail which, even though it's free to spin in a gale, is still vulnerable to gusts going in different directions at different heights - with that design already going onto ocean-crossing boats, it's time to start looking for rival systems.

    Making control of the sail practical for a crew of one or two people is another issue, of course, but I have ideas which I'm still exploring (such as a telescopic rotating rod to transfer power to whatever's locked onto it at the time, while the power would be supplied by pedalling - it's then just a matter of selecting what you're adjusting with it, using that to move parts of the wing against a force and pressing buttons to release them to let them move back).
     
  9. David Cooper
    Joined: Jan 2015
    Posts: 153
    Likes: 10, Points: 18, Legacy Rep: 42
    Location: Scotland

    David Cooper Senior Member

    The wingsail sections would be similar in total weight to a normal wingsail. The weight of the mast needn't be ruinous either - the stats at the end of in the first video Doug linked to show that the free-standing rig used there is 2/3 the weight of the original for similar sail area, but it doesn't take a lot of innovation to use stays with a telescopic mast and also have their length adjust slightly as the mast rotates so that they can attach to the outside of the wing. Well-designed and well-engineered systems with a little bit more complexity than conventional ones aren't contrivance, but useful innovation.
     
  10. upchurchmr
    Joined: Feb 2011
    Posts: 2,701
    Likes: 66, Points: 48, Legacy Rep: 579
    Location: Ft. Worth, Tx, USA

    upchurchmr Senior Member

    Please go ahead and build this ---- thing.
    Then report back on the weight and performance.
    And Cost.

    Talk is cheap.
     
    OzFred likes this.
  11. David Cooper
    Joined: Jan 2015
    Posts: 153
    Likes: 10, Points: 18, Legacy Rep: 42
    Location: Scotland

    David Cooper Senior Member

    Some people like to be overly negative; others just get on with designing the future by overcoming the problems that stand in the way. Talk should always be the first step - one should first check with other people to see if an idea is already known to be fundamentally flawed, and that comes out of the conversation, you save a lot of money by not building. I don't yet see a fundamental flaw - what I see instead is problems simplifying. At the moment, all I want to do is explore the idea and ensure that I can't be blocked from using it by someone else slapping patents on it. If it continues to look viable, I will build it once I have the money to do so (which depends on the success of a programming project).

    The weight shouldn't be substantially more than for a non-telescopic wing, and the performance should be near identical if the extended wing is the same shape as proven AC designs.

    I've resolved some more of the problems with it. The stays will attach to the top of the middle section (3/5 of the way up the mast when fully extended), and the first two stages of reefing will involve lowering the top two wing sections into the middle one, thereby not requiring a change in stay lengths. Reefing beyond that will reduce the height to the point where stays are no longer needed, so that's a considerable simplification. I had originally thought the sections would all concertina into each other at the same rate rather than doing each section in succession, but that would make twist impractical when reefed, so it's a good thing that the correct way of doing it also simplifies the way the stays are used, and it also allows reefing to be carried out with the wing under load (except when going between using two and three wing sections). [However, to reduce cost, I should also consider using four sections instead of five, which means the stays would attach 3/4 of the way up.]

    [Correction - it can't reef the top sections under load as the support inside the top of the middle section would not be properly supported internally in the intermediate positions.]

    For controlling the rear wing section angles, the power will be supplied by a rotating telescopic rod which will pull each rear section of wing to one side or other depending on the rotation direction of the rod. When a section reaches a certain angle, it disengages and all the energy goes into the other sections that are still free to move, so the shape of the wing is controlled by six angle limits which could be adjusted electronically. This means that during a tack or gybe, all that's required to set the shape to the mirror image of the previous shape is to press a release button and then turn the telescopic rod the opposite way from the previous time. Fine adjustments can then be made by changing the limits and winding more or letting sections out. That makes control by one person practical (even if there are two of these sails). Two buttons can be used to set the most recently used shapes for upwind and downwind at a single press.

    Additional controls are needed for the top section(s) to balance the boat by allowing them to be eased in isolation from the lower ones (and pulled the wrong way), but again all the power can be tapped from the rotating telescopic rod, so it's a lot simpler than I'd originally envisaged - controlling it should be fully practical for a crew of two or even just one.
     
  12. upchurchmr
    Joined: Feb 2011
    Posts: 2,701
    Likes: 66, Points: 48, Legacy Rep: 579
    Location: Ft. Worth, Tx, USA

    upchurchmr Senior Member

    The weight of this will be horrendous.
    The cost will be similar.
    Performance is a serious question. Any sentence starting with "XXX should be similar to YYY" has so much uncertainty as to scare the living daylights out of me.

    Please refer me to a rotating telescopic control rod. My concerns are cost, weight, and slop in the mechanism.

    Please provide a sketch for your ideas so we can have a useful discussion.
     
  13. David Cooper
    Joined: Jan 2015
    Posts: 153
    Likes: 10, Points: 18, Legacy Rep: 42
    Location: Scotland

    David Cooper Senior Member

    In which case the weight of existing wingsails must also be horrendous, or you're imagining great additions of extra weight which aren't necessary. The horizontal structures in the wing are much the same whether it's a single-piece vertically or split into multiple sections, while the weight of the surface material is practically the same whether it telescopes or not. A five-section mast with overlaps of the proportions I outlined adds 23% extra weight there, but once you look at that in the context of the total weight of the rig, I'd expect it to be less than 10% (though I'm having to guess the relative weights of mast and the rest as I've no actual stats on existing wingsails to go by).

    "The cost will be similar."

    The cost will be high, of course - that's only to be expected with any wingsail, but other costs associated with wingsails will plummet due to the reduction in size when in transit and storage, so this leads to much greater affordability.

    "Performance is a serious question. Any sentence starting with "XXX should be similar to YYY" has so much uncertainty as to scare the living daylights out of me."

    If the only significant difference is 10% in weight, does that sound like a scary addition? However, the ability to reef also opens the door to weight reduction by making it less robust when unreefed, instead generating more power from the lower sections in strong winds with a reduction in overall drag. I suspect the performance could end up being just as good as a non-telescoping wingsail in most conditions and infinitely superior in conditions which the non-telescoping kind can't handle.

    "Please refer me to a rotating telescopic control rod. My concerns are cost, weight, and slop in the mechanism."

    The telescopic rod is for power transfer (like a drive shaft), so slop is unimportant. Its weight is also trivial if the rotation speed is high (which can be achieved through the gearing). Picture a telescopic aerial on a radio - the telescopic rod is like that kind of aerial, but with a square section to transmit the rotation through each part, and at the top of each section you need to imagine a rounded part (bush needed to enable rotation where it's being held) and a cog (to transfer power) which remain at all times within the horizontal plate at the bottom of a wing section. That's trivial to engineer and weighs very little. And a cog might be overkill - it could potentially just use a belt.

    "Please provide a sketch for your ideas so we can have a useful discussion."

    It's really a moving diagram that's needed, and that's beyond my ability to produce - anything else becomes a complex mess that's hard to read because too many things are happening in a tight space to show clearly. Indeed, I'm struggling to visualise it all properly without a model. There's a part of the design that I got wrong before - it would work with some systems, but not the one I'm currently exploring. Earlier I said the following:-

    "A big advantage of using several telescoping mast sections though would be that the top part of the wing sail can be substantially narrower than the lower parts of the mast - you want each mast section to be as wide as the wing section that it supports to maximise strength, and this allows greater strength low down where it's most needed."

    That's clearly not possible if the top section of wing is to fit around the bottom section of mast when fully lowered. I've worked through dozens of configurations along the way and was mixing up the current idea with older ones in which the wing sections sit above the top of the mast sections in the fully reefed configuration, which either requires far too many mast sections or leaves you unable to lower the top sections into the lowest one. However, the addition of stays to minimise weight means that a thinner, lighter mast is practical, so this is not a major issue - ideally I'd like to do away with stays to avoid the whining noise they make at speed, but that's probably asking too much.

    There are other problems which I thought I'd solved earlier too, but hadn't, although I can see a variety of viable solutions which I'm trying to improve. For example, the top of each mast section has to pull up the bottom of its corresponding section of wing rather than the top, which is awkward (losing a simple way to prevent the wing rotating against the mast unless the mast is non-circular or has vertical grooves in it), and any connections from the power transmitting rotating rod inside the wing has to be picked up inside the bottom plate of a wing section and then be passed up to the top of that section before it can reach one of the outer structures that supports the adjustment of the rear part of the wingsail, so this would be best done ahead of the mast with each section sticking far enough ahead of the one above it to provide room to hold this mechanism (though that happens to be a better shape for the wingsail in any case as the joins between front and rear halves of the wing should run vertically, which means the front part should slope).

    It's easier to follow what I'm describing if you get hold of a radio and use its telescopic aerial to visualise everything else against. Start with the fattest section exposed and all the rest inside it, then pull out the next section while keeping the inner ones inside it. Imagine one or more wires going down from the top of the second mast section to the bottom of the second wing section and you should be able to visualise it being pulled up with it, but these can only link together by having a gap in the bottom plates of the inner wing sections for these wires to run through right next to the mast. Repeat this for the next ones, but you can see that we're back to the problem that the base of each wing section has a hole in it for the mast to pass through, and the mast sections narrow as it moves up them, so the hole in the base either has to narrow in some way to stop its wing section wobbling around the mast, or it must run onto a ring that slides on up the narrower part of the mast (and we'd have rings locking onto inner rings too). I had thought I could avoid that by attaching the top of each mast section to the top of its corresponding wing section, and there may yet be ways to do that, so I'm still exploring that. There are certainly some solutions that will work, so it's just a matter of working out which one's best.
     
  14. upchurchmr
    Joined: Feb 2011
    Posts: 2,701
    Likes: 66, Points: 48, Legacy Rep: 579
    Location: Ft. Worth, Tx, USA

    upchurchmr Senior Member

    Sketch. Vague glowing promises and descriptions don't get it when trying to understand what it takes for structure.
    The overlap joints placed in bending will have to be much heavier than a continuous mast.
    Your overlapping joints in a radio antenna don't carry the force of a sail in bending.

    I'm going to quit talking until you produce a sketch. There is nothing to talk about.
    I'm sure you will be glad I'm gone.
     

  15. OzFred
    Joined: Nov 2015
    Posts: 334
    Likes: 24, Points: 18, Legacy Rep: 16
    Location: Earth

    OzFred Senior Member

    Well, get on with design then. Until you have that, any talk of strength or weight is irrelevant.

    One purpose of building a model is to discover problems with the design, but at this stage you don't even have that so both problems and resolutions are just thought bubbles. For example, claiming that it will be able to be installed or removed without a crane can only be substantiated if you have a realistic estimate of weight, which will require a detailed design and materials list. Guessing that it won't be much more than a solid wing sail is just that, guessing.

    A detailed design will also lead to other estimates, such as construction methods and cost. Those are the easy bits, performance, longevity, convenience, etc. can only be determined once you have a sufficiently detailed design to be confident that it can actually be built and tested. You might then proceed to test elements of the design at some reduced scale on a platform other than an actual boat. Models don't have to be functionally complete to be useful, they just need to be sufficient to test a particular concept (i.e. genuine prototyping).

    If you're into software development, you should already know the software development lifecycle, which is much the same as any product development lifecycle. It seems you're in the concept phase, time to get into design or you'll suffer paralysis by analysis. ;-)
     
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.