Carbon wing mast for Archimedes 60

Discussion in 'Multihulls' started by MichaelRoberts, Dec 19, 2020.

  1. MichaelRoberts
    Joined: Sep 2015
    Posts: 40
    Likes: 5, Points: 8, Legacy Rep: 16
    Location: Australia

    MichaelRoberts Archimedes

    This thread was suggested by another member - I hope it will be helpful and interesting

    • We built a 20 metre long wing mast using carbon fibre and epoxy.
    • It has to take 15 tonnes of compression - load gets up near this in 33 knots with full mainsail broadside to a gust
    • The cross section is a modified Clarke Y - a D section advocated by A J Marchaj and F Bethwaite

    • The mast sits on a spherical roller thrust bearing.
    • The halyards pass through the bearing into the cross beam underneath
    • There are two failure modes to worry about - long wave and short wave buckling. Long wave (Eular buckling) means bowing under compression and short wave means crinkling up.
    • The buckling problem is difficult unless you use diamond stays because Eular buckling load goes down fast as a square of length. Need adequate cross section - second moment of area
    • Diamond stays are out because they bang into the shrouds when mast rotates

    • A form was made with CNC cut ribs and stringers
    • Lots of conduits through ribs for halyards and electrics lead to sockets for sheaves
    • Over the form thin aircraft ply was bent - sheets scarfed together - a bit of fairing
    • The form is permanent - it adds considerable strength and weight is acceptable
    • A 21 m x 1.2 m polythene sheet was placed under the form ready to be sealed along the top
    • Then laminations over the ply - bi-axial and uniaxial carbon - moving fast
    • Then seal the polythene vacuum bag and pull a medium vacuum - don't want to crush the fragile form - release vacuum after one hour - so form can resume its shape
    • Finished with a bi-axial layer of carbon
    • Build a big hardwood end on the bottom to house the bearing

    • Bolt thick reo rods from fixing points for the top shrouds to a hydraulic RAM on the base
    • Micro-meter and pointer to measure compression deflection at the top end
    • Pump up to 10 tonnes
    • Barely measurable linear compression up to 3.5 mm

    • Mast was supported at the base and the top shroud point - supports 16 m apart
    • Load was applied at the centre of the span
    • Plotted nice linear deflection with load steadily increasing to 3000 N (300 kg = 660 pounds)
    • Calculated the Eular critical load from the deflection slope - 24 tonnes - plenty strong

    • The method, the calculations and the number of laminations seems to work
    • Cost of materials was around $5,000
    • Actual weight of complete mast 265 kg - 583 pounds - a bit heavy?
    • A lot of hours - but fun - so that's ok
    • Now to see if the aerodynamics work - 20+ knots here we come
    Mast ribs assembly enhanced and cropped.jpg Mast white enhanced March 2020.jpg
    Corley likes this.
  2. Russell Brown
    Joined: Jul 2012
    Posts: 127
    Likes: 32, Points: 28, Legacy Rep: 25
    Location: washington state

    Russell Brown Senior Member

    That looks like a good spar. Why the flat trailing edge? I really like the plywood & lumber masts. How much laminate went on yours?
    I built a 54' plywood & lumber wing mast for a racing tri that had only light glass over it. It was designed by someone who had built lots of similar masts for the Caribbean charter cats. It seemed lightly built and a bit bendy, but it's still going 25 years later.
  3. Gary Baigent
    Joined: Jul 2005
    Posts: 3,000
    Likes: 118, Points: 63, Legacy Rep: 509
    Location: auckland nz

    Gary Baigent Senior Member

    Is it a D shaped mast (like in the image) for double mains airfoil shape?
    If not and built for a single main ... very draggy and inefficient.

    Attached Files:

  4. MichaelRoberts
    Joined: Sep 2015
    Posts: 40
    Likes: 5, Points: 8, Legacy Rep: 16
    Location: Australia

    MichaelRoberts Archimedes

    Russell your experience with the glass and wood spar is reassuring. Did it take 35 + knots ok?

    Garry, yes it is D shaped and the mainsail is single sided

    However the cars have lateral slides.

    So that the lee side of the mast to sail transition is smooth. Almost no step. Lee side air flow should be laminar.

    Air on the windward side of the mast D will vortex into the step then hit the luff. Will this form a sort of virtual sock? Hope so.

    If it works I'm expecting substantial improvement in lift.

    There is more... angle of attack stuff. Needs diagram to explain. Next post maybe.
    Gary Baigent likes this.
  5. catsketcher
    Joined: Mar 2006
    Posts: 1,303
    Likes: 144, Points: 63, Legacy Rep: 790
    Location: Australia

    catsketcher Senior Member

    Thanks for the detail Michael. The sharp D mast was developed by Frank Bethwaite and used on the Tasar. Frank found in tunnel testing that the sharp aft edges energised the boundary layer and was lower drag than the more normal wing section. The Tasar certainly is fast for its sail area. The Reynolds number will be different on Michael's boat so there may be some differences in flow. An easy way to do the lateral slides may be to use longer webbing strips to the track slides on a centreline track. The mainsail luff then sits on the aft edge of the mast as the battens are under compression. My non rotating mast has webbing and slides and the luff "tacks" each tack. The luff sits on the flat aft section of the mast section - it is a really nice, low tech and very low chafe idea. It may not work with some mainsail profiles that pull aft on the slides but I have used it for twenty years now with very little chafe or hassle.

    I made a cedar strip mast for a Newick tri. It was pretty fun to make and didn't need spreaders. Fat airfoil section - 50% IIRC

  6. MichaelRoberts
    Joined: Sep 2015
    Posts: 40
    Likes: 5, Points: 8, Legacy Rep: 16
    Location: Australia

    MichaelRoberts Archimedes

    Hi Catsketcher

    What does 50% IIRC mean? Did you have to glass over the cedar.
    Good term "energised"!
    Thanks for the strap idea but I've already designed cars with lateral slides and the wide track is already bonded in. Fingers crossed!

    I found reconciling the fatness and the buckling problem rapidly became very difficult as the panel length went past 7 m. The D section helped a lot, much stiffer than a teardrop shape. The standing rigging (Dyneema) consists of upper shrouds to 16 m above the bearing and lowers to 8m. Panel length 8 m.
    Analysis with six layers of carbon.jpg Analysis with six layers of carbon 1.jpg
    The D section extends aft, shielding the track
    The aim is to minimise air leaking from windward to lee side by closing the gap between the mast and the sail.
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.