Wing mast section, truncated NACA 63015?

Discussion in 'Hydrodynamics and Aerodynamics' started by MichaelRoberts, Sep 6, 2015.

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

    tspeer Senior Member

    I've not had the tools to analyze a mast section with a backward facing step. Frank Bethwaite sent me some tracings of Tasar mast sections some years ago, but the shape was too extreme for XFOIL or MSES to handle. Now that I'm learning to use RANS with OpenFOAM, stepped wingmasts are on my list to tackle.

    I suppose a truncated NACA 63015 would be a reasonable starting point. As with any other type of section, the best solution will be a shape that is specifically designed for your requirements.

    I suspect the key to optimizing a stepped mast is to get the right balance between skin friction and base drag. There's a drag penalty associated with the separation bubbles behind the mast. The bigger the step, the more the drag is going to be. But you can have a more favorable pressure gradient, and more laminar flow, with a wider base. It doesn't take much base drag, however, to wipe out any gains you get from laminar vs turbulent skin friction. Base drag is notoriously tricky to predict accurately, and if you have to make assumptions as to what the base pressure is, you can get any answer you want to name.

    With the teardrop sections, I found that the best performance was obtained not when the lee side was smooth, but when there was a modest separation bubble on both sides. This made the teardrop sections behave much like a stepped section.

    In principle, you could cut off the mast at the separation point and probably not change the aerodynamics by very much. So that might be a way to design sections with existing tools like XFOIL. Design a teardrop mast and then truncate it at the separation point, especially if you could rotate the mast so the flow separated at the same chordwise location on both the windward and leeward sides.
     
  2. MichaelRoberts
    Joined: Sep 2015
    Posts: 71
    Likes: 16, Points: 8, Legacy Rep: 16
    Location: Australia

    MichaelRoberts Archimedes

    Good morning Tom,

    Many, many times have I read your paper on rotating wing masts. You are generous to have done that research and posted it up for all of us, thank you.

    Your paper sits on my altar to boat design along with Marchaj, Gougeon, Dave Gerr, Chris White and a very old, second hand copy of Elements of Yacht Design by Norman Skene which seems to have a hand written note by L Francis Herreshoff dated 1944, saying that Norman was good at sums but knew nothing about cruising: "I have noted of late ... that subject would be like the blind leading the lame"?!

    Very happy to have your participation in this discussion.

    Anyway back to business. I could revert to something like NACA0020 as Petros suggests and that would narrow the tail and so reduce that aft suction area but something about this 630018 seems like it wants to knife into the wind and trick the flow into thinking the chord is longer. Maybe I'll just add another 50 mm to the tail and reduce the chord just a touch.

    Petros if I promise to polish the leading edge every morning can I keep this section?

    To clarify matters and ensure we are all on the same page I just made this little diagram, the step generates vortices - hopefully they are "Critically bound" and constrained to remain in their little pockets as suggested by J E Hardiman.

    This is a great conversation, thanks all

    Michael
     

    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

    It's not clear how you intend to attach the sail to the mast. I think just mounting a track on the centerline of the aft shear web (which may have to be strengthened for the purpose) for the batten cars would be the way to go. Just by having a rotating mast you've got the majority of the benefits.

    There are a lot of practical issues to consider, like how to get your fingers to the pins on the batten cars and sail slides to attach and remove the mainsail. On a 60 ft boat, I'd aim for the 80% or 90% aerodynamic solution and then put the emphasis on handling the boat in all conditions. The sail has to go up and down the mast and be reefed, and having some mechanism to let it slide to one side of the mast is just going to add weight and complexity for little (if any) gain.
     
  4. John Perry
    Joined: Nov 2003
    Posts: 308
    Likes: 53, Points: 28, Legacy Rep: 129
    Location: South West UK

    John Perry Senior Member

    Having read what Tom Speer has to say in his last message I think I would want to see some further analysis before committing to this concept for the mast for a 60 foot catamaran. That analysis should presumably cover three cases - a full teardrop section, a truncated teardrop section with mainsail luff on the centreline and a truncated teardrop section with the mainsail luff offset to leeward as proposed here. As Michael Roberts says, one of the problems with a conventional large cord wing mast is excessive windage in storm conditions with no sail hoisted. If little performance would be lost by simply truncating the section while still leaving the mainsail track on the centreline of the section, I would have thought that would be a very attractive saving in windage under 'bare poles' compared with a typical wing mast, and also a weight saving.

    I am sure others here are better qualified than myself to analyse such options. I had intended it to be this winter's project to investigate the possibilities of OpenFoam, but so far haven't even started. A new year resolution perhaps. Can anyone give me an idea just how difficult is it, or is that a silly question?

    I couldnt help thinking about how the original proposal could be implemented mechanically. Small transverse tracks each mounted on a slider that runs on a track vertically up the back of the mast, these transverse tracks needing to be stabilised against rotation about a vertical axis. So possibly two vertical tracks or a single track and two guide surfaces. All doable but would need custom made parts, I think it would be cumbersome if implemented using only standard yacht hardware. I just sketched another possibility below, a bit of a flight of fancy. The sketch, which is diagramatic, shows the inner ends of the battens free to bounce around in the cavity in the back of the mast, which would need to be reinforced to withstand such abuse. The inner end of each batten is linked by a short cord (the wiggley line in the sketch) to a slider that runs on/in a lightweight vertical track, perhaps even for a big boat this would only need to be the kind of track you would see on a dinghy mast. The length of this short cord is such that it stops the inner ends of the battens escaping when hoisting/lowering the sail. In addition to this, the inner end of each batten is linked by a cord to the slider for the batten below it and by another cord to the slider for the batten above it, so that when the sail is hoisted/lowered the sliders keep pace with the battens and hopefully nothing jams.

    Michael Roberts wrote:
    "Do you want to see this boat? Here are two pictures.
    Thank you so much for your responses, please keep up the advice. Now I had better get to the boat shed."

    The design is aesthetically beautiful, in the modern idiom (IMHO) Presumably, since there is no structure near the centreline forward it will be moored/anchored from a bridal attached to the two bows and with an anchor windless somewhere back aft. I did wonder what is the thinking behind four rudders - shallower draft? I hope that boat shed you mention is a big one!
     

    Attached Files:

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

    tspeer Senior Member

    For me, it's difficult, because I've been away from coding for a while. OpenFOAM is not a CFD program. It's a system for programming CFD programs. And there's next to no documentation - you have to rummage through the source code to find anything you want to know. I took a 4-day course on it, and I don't think I'd have a prayer of getting it going without that. There are a number of OpenFOAM spin-offs that make it easier to use, such as Helyx-OS and Caelus that can make it easier to use, but I've not tried them myself. It is also reputed to be very difficult to build proper boundary layer meshes with OpenFOAM's native mesher, SnappyHexMesh.
     
  6. MichaelRoberts
    Joined: Sep 2015
    Posts: 71
    Likes: 16, Points: 8, Legacy Rep: 16
    Location: Australia

    MichaelRoberts Archimedes

    Hi John, Tom and all members of the advisory board,

    Thanks again for the comments. I'm getting the message that this cut off mast is a bit adventurous - maybe I should make a foam section and blow some smoke past it.

    John your idea for the car with wiggly bits is OK but it's even easier to have a lateral slide made of 6 mm rod with an slightly oversized eye bolt riding back and forth. The eye bolts screw into, and tension, the end of the tubular battens. The car is a 100 mm length of aluminium U extrusion with four nylon wheels running inside that hollow aft channel of the mast - the heavy and expensive extruded track is eliminated. The back stop for those wheels is not shown on my sketches.

    May I ask what is Open Foam and what does CFD mean?

    Are we talking finite element analysis? If so I can suggest some wonderful analysis software that I use with Rhino. It is simple to use and produces excellent visual and numerical results. It is called Scan and Solve, developed over 25 years by Professor Vadim Shapiro. You simply draw the solid with Rhino, type the command SnS, specify the loads and restraints and the properties of the material and within a few minutes out comes gloriously coloured analyses of stress, strain, displacement and many other parameters. Here is one of many links http://v5.rhino3d.com/group/scan-solve.

    Yes John the shed is biggish - 20 m x 6.5 m made of second hand beams. It was OK for the hulls but it's too tight for the cabin top which I'm now having to build in two halves.

    And yes, planning a bridle from the bows for the anchor. There are two anchor compartments under the foredeck. All decks are wide and level so you can zip around in your zimmer-frame without falling in the drink.

    Here's a pic of the temporary framing for one of the hulls - the flare out is where the bridge deck joins on.

    Have a good day

    Michael
     

    Attached Files:

  7. John Perry
    Joined: Nov 2003
    Posts: 308
    Likes: 53, Points: 28, Legacy Rep: 129
    Location: South West UK

    John Perry Senior Member

    Goodness Michael - that looks a fantastic project - wish you all the best with it. I take it that what the photo shows in your shed is the plug for a foam sandwich hull. You mentioned the hulls having a shallow forebody and the consequent need to position the daggerboards further forward. I can imagine that the depth of the forebody would have a small effect on the optimum position of the daggerboard but I would not have thought it would be a huge effect, although difficult to quantify without sophisticated analysis. Would not just setting a different foresail have more effect than that? And of course, with retractable daggerboards you do have scope for adjustment. Any special reason to have an unusually shallow forebody in the first place?

    Dont bother to answer my questions if you are too busy - I can see that you have a lot of boat building to get on with!

    The Scan and Solve software you mention would appear to be FEA software for structural analysis. I have made much use of comparable software over several decades and at one time was a member of the team writing the code for one of the early FEA software packages, but I have not used Scan and Solve, nor Rhino for that matter.

    I wont try to explain what CFD is since I am sure that others here are better qualified to do that than I am.

    Thank you Tom for your comments re. OpenFoam. I better do some initial investigation through the internet to see if it is something I might be able to cope with, but it does not sound easy.
     
  8. tspeer
    Joined: Feb 2002
    Posts: 2,319
    Likes: 303, Points: 83, Legacy Rep: 1673
    Location: Port Gamble, Washington, USA

    tspeer Senior Member

    I think the cut-off mast is a reasonable approach. Like anything in yacht design, it just needs to consider how it fits with the mission of the boat and the rest of the design. You will probably want to limit the chord based on the windage you can stand when hove to under bare poles in a gale. How easy it is to raise the mainsail and reef are important, too.

    In general the profile drag of the mast and sail is not a big driver in the aerodynamics. The induced drag due to lift is much larger, so the height of the rig and the ability to control twist is more important than details of the mast shape. But since you have to give the mast some shape, it's worth doing some homework to give it a decent shape.

    CFD means computational fluid dynamics. The problem is simple aerodynamic methods leave out much of the flow physics. They often assume the flow is fully attached in order to calculate the flow using just elements placed on the surface of the mast & sail. If you're dealing with separated flows, such as behind the truncated mast, then you need to use something like a finite volume code, in which the entire flowfield is broken into millions of little boxes, and a massive bookkeeping exercise is used to make the input and outputs to each box consistent throughout the flowfield. This takes a lot of computation, and sophisticated software.

    Commercial CFD codes cost tens of thousands of dollars per year for the licenses. OpenFOAM is an open-source CFD code that is freely available, so it's getting a lot of attention these days. It's not as efficient as the commercial codes, and it takes a lot of effort to learn to use it. But the price is right.

    Finite element methods do for structural analysis what CFD does for fluid dynamic analysis. Some CFD codes use a finite element formulation, but finite volume formulations are more popular these days. CFD is concerned with what's happening outside the mold line, and finite element methods are concerned with what is happening inside the mold line. You would use a CFD code to come up with the loadings that you'd apply to a finite element model to get the stresses in the structure.

    From the Scan&Solve FAQ:
    "Are orthotropic materials, such as wood or composites, supported in Scan&Solve™?
    Scan&Solve™ presently works with isotropic materials, such as metals and unreinforced concrete, that do not have properties which vary with respect to orientation or position in space. The ability to model orthopropic materials is a high priority for future development.

    Can I use Scan&Solve™ to analyze objects consisting of multiple (or composite) materials and components?
    Analyzing assemblages of multiple materials and components is a high priority for future versions of Scan&Solve™."

    These are serious limitations for using Scan&Solve to analyze boats. Unless you're building in steel or aluminum, the materials will be orthotropic. And while a purely linear code is quite useful, the inability to predict buckling is another serious limitation.

    I think this illustrates the current gulf in engineering software. The user interfaces are getting easier to use, so that you don't have to be a practitioner to get results. But the easy-to-use codes are sometimes simplified to the point that the results can be misleading. On the other hand, the professional codes are too expensive and require too much expertise for the amateur or casual user.
     
  9. Petros
    Joined: Oct 2007
    Posts: 2,934
    Likes: 148, Points: 63, Legacy Rep: 1593
    Location: Arlington, WA-USA

    Petros Senior Member

    lots of things about fluid mechanics are not intuitively obvious, one of the reason that a viable aircraft (or even a sail plane) was not possible until we had the analytical tools to study it in detail. and why all sail boats from ancient times up until only about 100-120 years ago could never make better than about 90 degrees to the wind. Now is is common for sailboats to be able to make headway against the wind.

    vortex generators work by putting energy into the air stream, to keep the flow attached to poorly shaped surfaces. the idea is the drag it "cost" to generate the vortex is less than the drag the large flow separation causes down stream of the the vortex generator.

    It is actually not a laminar flow question, it is attached flow (to the surface) vs. detached flow. Large areas of flow separation on a hull or fuselage will cause a lot of drag, or loss of control on rudders, ailerons, etc, the vortexes generators energize the boundary layer so it stays attached, reducing or eliminating the flow separation. It is especially important on a rudder or aileron on aircraft at low speeds.

    lots of common sailing terms carry these misunderstanding from earlier times, I often find some old sailor telling me that you want more curvature or camber on a sail in light wind conditions so it will "catch more wind". Sorry to inform you all that it "catches" the same amount of wind be it curved like a spinnaker or flat. the reason it works is that more curvature means it accelerates the mass of the air moving over the surface of the sail more than with less curvature, so providing more lift or thrust.
     
  10. MichaelRoberts
    Joined: Sep 2015
    Posts: 71
    Likes: 16, Points: 8, Legacy Rep: 16
    Location: Australia

    MichaelRoberts Archimedes

    Good morning all,

    Can't believe it - all this typing instead of glueing and screwing - maybe social media ain't so bad after all.

    John thanks for your reply. I also spent many years writing code - in assembler - about your forebody and dagger board comments:

    a shallow forebody and the consequent need to position the daggerboards further forward.
    I tried to understand why some cats hobby horse and it seems to happen when the hydrodynamic lift under the forebody is overcome by the forward driving wind force acting on the centre of effort half way up the mast. Using Maxsurf I was able to calculate the driving force (=resistance) at 15kn and the longitudinal overturning moment. The long ramp of the forebody is designed to keep the nose up even at this speed.

    optimum position of the daggerboard
    By projecting the underwater side view of the hull onto a plane (Rhino) and finding the centroid of that shape (Rhino again), one can approximate the centre of lateral resistance. Similarly for the centre of effort - project sail shape onto a plane and find centroid. Then I placed the dagger boards to balance them. And as you say, I can always use a different area of headsail.

    Attached is a picture of the hulls underwater - 20 kn here we come?!

    Petros you are firing up my imagination: vortex generators work by putting energy into the air stream. Are these like the ridges at the rim of a frisbee, or the little scoops on the back of some SUVs? If so that's exactly what I'm thinking about the chopped mast. I feel that the trailing edge must be crisp and maybe some mechanism to induce a favourable direction of rotation. Which would be clockwise for a left to right flow?

    And Tom thank you for your erudition.
    the materials will be orthotropic. And while a purely linear code is quite useful, the inability to predict buckling is another serious limitation
    I made a sample foam sandwich beam, tested it, calculated the effective modulus of elasticity (a pretty pathetic 9GPa) and used those parameters in Scan and Solve to get some idea of stress concentration. Then I made the real aft beam, bought some hydraulic jacking stuff on eBay, clamped it's ends to the floor and jacked it's middle. All the while plotting deflection. Watched on by one of my old colleagues from Sydney Uni who knows a thing or two about structures. The concrete floor burst open as we passed ten tonnes. Up to that point the beam deflection was nicely linear - no where near any plastic deformation - deflection at 10 tonnes point load in centre of beam was 9 mm.

    As for buckling, I wish I had an answer. For the mast design I used both the Euler and the Rankine Gordon criteria, then averaged the result and twiddled the second moment of area until I got 26 tonnes, twice the max compression of 13 tonnes. Unsure about impulse loads in slamming conditions.

    Well the boat shed is calling, this has become a fascinating discussion, looking forward to more of your valuable comments

    Thanks again

    Michael

    PS this is an old picture when I only had single fin rudders and no elliptical leading edge on dagger board foils
     

    Attached Files:

  11. schakel
    Joined: Jul 2008
    Posts: 386
    Likes: 16, Points: 18, Legacy Rep: 48
    Location: the netherlands

    schakel environmental project Msc

    Is there a naca profile bible?

    Hello fellow designers,

    The more I dig into Naca profiles, the more encyclopedic it gets.
    Is there a site, book, article that gives an overview in names, formula's, drag?
     
  12. DCockey
    Joined: Oct 2009
    Posts: 5,229
    Likes: 634, Points: 113, Legacy Rep: 1485
    Location: Midcoast Maine

    DCockey Senior Member

    Theory of Wing Sections: Including a Summary of Airfoil Data Ira H. Abbott, A. E. von Doenhoff, 1949 http://store.doverpublications.com/0486605868.html

    Theory of Wing Sections is a standard reference work on NACA profiles. While "Theory" is the first word in the title there are descriptions of the NACA series and formulas for the shapes of the series which can be generated directly by formulas, and an overview of the methods used for generating the shapes of the series which are based on theoretical pressure distributions. An appendix of over 350 pages has tables and graphs of Basic Thickness Forms, Mean Lines, Airfoil Ordinates, and Aerodynamic Characteristics of Wing Sections.

    Theory of Wing Sections has been available for many decades in an inexpensive paperbound version published by Dover.
     
    Last edited: Jan 7, 2016
  13. DCockey
    Joined: Oct 2009
    Posts: 5,229
    Likes: 634, Points: 113, Legacy Rep: 1485
    Location: Midcoast Maine

    DCockey Senior Member

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

    tspeer Senior Member

    Another good book is Airfoil Design and Data, by Richard Eppler. It is three things - a discussion of airfoil design philosophies, an expanded user-guide for the Eppler airfoil design code, and a compendium of data on Eppler's section designs. It's a post-NACA Theory of Airfoil Sections.

    Summary of Low Speed Airfoil Data, Vol. 1,
    Summary of Low Speed Airfoil Data, Vol. 2,
    Summary of Low-Speed Airfoil Data, Vol. 3,
    Wind Tunnel Aerodynamic Tests of Six Airfoils for Use on Small Wind Turbines,
    Summary of Low-Speed Airfoil Data, Vol. 5
    by Michael Selig are good sources, too.

    In general, look for papers by Michael Selig, such as High-Lift Low Reynolds Number Design and Systematic Airfoil Design Studies at Low Reynolds Numbers.

    If you're looking for the coordinates of different sections, the place to go is the UIUC Airfoil Database.

    Since the 1980's, the approach of selecting an airfoil from a catalog of section data has largely been supplanted by the approach of designing a section for the specific application using programs like the Eppler code (user's guide) which itself has been supplanted by programs like XFOIL and Javafoil. You can learn a lot by just experimenting with these tools to see what effect changes to a section have. There are too many subtleties about a section's performance to be reduced to simple formulas. These programs are as close as you're going to get to that, and both are free. Javafoil is easier to learn to use, but I trust the results from XFOIL more than Javafoil.
     

  15. schakel
    Joined: Jul 2008
    Posts: 386
    Likes: 16, Points: 18, Legacy Rep: 48
    Location: the netherlands

    schakel environmental project Msc

    Hooray, Thanks a lot.

    I am busy for the next few months.
     
Loading...
Similar Threads
  1. pmudesign
    Replies:
    14
    Views:
    2,885
  2. Will Fraser
    Replies:
    12
    Views:
    3,509
  3. Maarten88
    Replies:
    5
    Views:
    998
  4. lunatic
    Replies:
    7
    Views:
    2,152
  5. Forecaddie
    Replies:
    9
    Views:
    1,631
  6. Forecaddie
    Replies:
    14
    Views:
    1,952
  7. lunatic
    Replies:
    15
    Views:
    2,508
  8. dustman
    Replies:
    25
    Views:
    3,832
  9. Jhomer
    Replies:
    4
    Views:
    2,209
  10. mfly
    Replies:
    2
    Views:
    2,048
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.