Transverse frame calculation

Discussion in 'Class Societies' started by DUCRUY Jacques, May 1, 2010.

Thread Status:
Not open for further replies.
  1. Brent Swain
    Joined: Mar 2002
    Posts: 951
    Likes: 35, Points: 0, Legacy Rep: -12
    Location: British Columbia

    Brent Swain Member

    Boats which have survived as much ,and have done as much over 30 years, tens of thousaned of miles of open ocean, and several hundred hulls, speak for themselves. If you say the sun rises in the east and sets in the west ,can you show me mathematical calculations to prove it? If not, then it must not be true? It couldn't possibly have been true before the invention of mathematics. Musta rose and set some other way!Duhhh
    I gave you the name of several experts, Kinny, the experts who design the keels that fall off in most round the world races, the experts at the Davidson lab wher e my first disasterous design was tested. Don't know their names, so they must not exist.
    I designed my twin keels the same shape as those designed by the designer of bluebird of thorne and by Laurent Giles . They were a bad idea when it comes to fouling anchor line when they pass between them, a total design screwup which needed modifying by me. Mathematical calculations wouldn't tell you that. Only practical experience would. Numbers don't do much sailing, nor do they get their hands dirty building .

    If there were any schadenfreude, it would have happened in the last 30 years , dozens of very successful boats ago. Hasn't happened ? Then it aint gonna happen, or it would have by now..
  2. tazmann
    Joined: Aug 2005
    Posts: 329
    Likes: 17, Points: 18, Legacy Rep: 215
    Location: California

    tazmann Senior Member

    Little better explanation please

    Hello Adhoc
    Your analagy does make sence when your talking about taking smallish plates and testing them in the way you describe but seen as how there are so many frameless or socalled framless boats and designs out there it cant mean much Vandestadt, Roberts , and Brents to name a few. Now if a guy was talking about building a 35' framed design with 1/8" skins and decided to get rid of the frames and built it framless using the 1/8" skins then your bukling analagy would be so true but if the skins were upped to say 3/16" with the correct hull shape, longs and bulkheads then the bukling problem goes away, Is this not true ?
    I'm not arguing framed or framless or whitch one is better just trying to get a better understanding of the bukling problem.
  3. Ad Hoc
    Joined: Oct 2008
    Posts: 7,323
    Likes: 1,216, Points: 113, Legacy Rep: 2488
    Location: Japan

    Ad Hoc Naval Architect


    Hmmm, ok, i'll wade in once more, finally.

    You need to differentiate between a build method and a structural philosophy. A structural design/philosophy is independent of how something is built. This simple fact is lost on non-engineers wishing to punt their method as being the saviour of the world as the world is wrong. I can build this way, ergo it works!

    (For example, if I design an I-beam, you can build it from 3 plates, or 2 FBs and a web plate, or an extrusion TEE and a FB or even an extrusion etc. The way you build/construct this I-beam, all are different, but the I-beam, ie the structural design of the I-beam and the loads/forces it can take, remains the same).

    This is also highlighted here:

    when another non-naval architect structural engineer “thinks” or “feels” or “believes” their way is better, without any understanding of structural principals. There is no point constantly discussing structural mechanics with people that will not alter their preconceived opinions based upon ignorance, faith and anecdotes, despite the voluminous evidence to the contrary of their assertions. The person making the claim that flies in the face of world wide acceptance of theories must provide qualitative evidence of THEIR claim, not the other way around!

    Plate size is irrelevant. Buckling is driven basically by two main parameters

    1) thickness of plate
    2) panel aspect ratio (length by breadth or a/b)

    The stress at which a panel of plating will buckle is given by:

    Sigma = K (E/1-v^2) x (t/b)^2

    K = a constant. This changes from 22.2 with a low a/b to 3.29 for infinitely long plate.
    v = Poisson’s ratio
    E = modulus of elasticity
    t = thickness.

    So, one can deuce that the buckling stress is directly proportional the thickness (and/or breadth, depending how you look at it). There are many different formulae depending upon the edge constraints, but basically follow a similar pattern.

    Ergo, the thicker the plate the higher the critical stress to cause buckling. Again, no magic there.

    So, upping from 1/8” plate to 3/16” plate, increases the critical buckling stress. This is independent of the size that is, small or large, it relates to the panel aspect ratio. So a 1.0 x 4.0m plate and a 100 x 400mm plate, the aspect ratio is the same!

    The factor which affects the different size, as you like to call it, is “b”, in the formula.

    Therefore changing from a panel that is say 4mm over 100mm wide has the term [(t/b)^2] = 0.0016 as its product. Whereas a 4mm plate of 1.0m wide is 0.000016.

    This means the larger the plate (not panel size) decreases the stress at which it shall buckle.

    Again, nothing to do with build method, just simple structural engineering.

    So, take a thick plate, say 100mm x 400mm that is 4mm thick. Push the short ends together, as it lays flat…eventually it will bend and buckle.

    Now, make the same 4mm plate but 1000x4000m, same aspect ratio, but large size…push the ends. The force required to push these two ends together is now much much less. It has become what is called slender. This is noted by the buckling stress has reduce by a factor of 100.

    So, does this mean you can’t build a frameless boat? No. Does this mean the boat will buckle much sooner than if the size of plate for its thickness (ie more slender) is now less, yes. How much, you need to determine by calculation. None of this has been demonstrated by any of the Religious zealots above.

    Everything is about understanding the structural philosophy of the implications of your method to build.

    As an extreme example, take say a super tanker, made from solid steel. It will sink. Ok, slowly carve out a hollow until it floats, ie satisfies Archimedes principal. The thickness of the plate will be thick, but the boat still floats. Is this rocket science, no. Bend the boat what will happen….it shall resist until the buckling stress or section modulus reaches its limit. Is this limit less than if the boat is made with frames and longitudinals…yes.

    How much less, is for you to determine and as such, what are the safe limits of that vessel.
  4. LyndonJ
    Joined: May 2008
    Posts: 295
    Likes: 19, Points: 0, Legacy Rep: 233
    Location: Australia

    LyndonJ Senior Member

    Well he's saying endlessly that everyone who advocates transverse framing is basically loco. Yet when we do buckling analysis we have to adjust grillage spacing and that includes some form of transverse.

    The alternative is all longitudinal but then you up against the old length squared curse.

    Then the rabbit Brent pulls from the hat is his version of origami, he tried to 'prove' it by the way and got all his engineering horribly wrong. So the logic doesn't scale to 6o feet.

    Look at his typical post above, what can you actually take away from that as valid information ? That it works for 36 foot boats !


    Your expert Mr Kinney wasa civil engineer like me ! The tank would have tank tested it for resistance, that's not a vindication of the design by the way .
    So you had a bad experience with a yacht designed by a civil engineer in a style that was a poor design anyway. That turns into a lambasting attack on Naval architects who try to point out that you are wrong in your structural arguments.

    If you actually worked with them you'd get a lot more respect and achieve a decent engineering approach. Can you admit that your suggested engineering reasoning is wrong?

    Would you like some help to put it right?

    Great post by AdHoc and it's nice to read posts from people who actually understand buckling.
    1 person likes this.
  5. RAraujo
    Joined: Apr 2010
    Posts: 136
    Likes: 28, Points: 28, Legacy Rep: 114
    Location: Singapore

    RAraujo Senior Member - Naval Architect

    Usually in (steel) boats built by traditional methods the plating and stiffeners are taken to their shape by plastic deformation (by either cold or hot work).

    In Brent's method there will be always a part of the deformation, which may be quantified, that will be elastic.

    In general there will be internal stresses which are tension stresses in the outer shell and compression stresses in the longitudinals.

    When external load is applied, perpendicular to the plating, it will generate internal stresses wich are compression stresses in the outer shell and tension stresses in the longitudinals.

    The super-imposition of both effects, from the deformed pannel under elastic deformation and the external loading, will lead to smaller stresses on the structure (like concrete pre-stressed structures).

    Therefore we can expect that larger transverse frame spacing is permissible (on the limit no transverse framing at all). How large it can be? It will depend on the size, shape and other elements that can have the same supporting effect as the frames (bulkheads, either partial or complete).

    It would be interesting to have some FEA simulations with the pre-stressed approach.

    I think this forum is supposed to be a place for open discussion on something we all love - boats. We all have things to learn so, please, let's try to keep the discussion on an adequate level without insulting or non constructive arguments.
    3 people like this.
  6. LyndonJ
    Joined: May 2008
    Posts: 295
    Likes: 19, Points: 0, Legacy Rep: 233
    Location: Australia

    LyndonJ Senior Member

    Trouble is that engineering dialogue just gets sidelined by anecdote and then everyone gets frustrated .
    Washed over... pounded on... hit by... collided with....
    and anecdote is not a reliable witness to the actual loads imposed in the situation.

    The pre-stress from the 'elastic bending' is relative to the force you put into it . Brent What force is required to pull the panel into shape ?

    You'll have some pre-existing compresiion in the longitudinal frame and some tension in the plating. But not a lot when you consider it's over the whole side .

    When the plating is in tension its already part way towards yield from a neutral position and a member in compression is already part way toward buckling. This could actually reduce collision strength in some circumstances like the blow to the plate I mentioned earlier.
  7. apex1

    apex1 Guest

    You should have followed these religious "arguments" during the past years, you would understand that no one here is willing to discuss with Brent on a "adequate" level. Because that would mean discussing anecdotes!

    Btw. that is not "Brents" system, it was done long before he entered the scene.

    Ad Hoc

    thanks for taking the time again and again. Probably it may persuade some of the novices. sure it will not convince Brent!

  8. bhnautika
    Joined: Feb 2006
    Posts: 852
    Likes: 57, Points: 38, Legacy Rep: 571
    Location: australia

    bhnautika Senior Member

    Finally we make some steps to enlightenment but I do have some queries.
    Ad Hoc if the structural design/philosophy is independent of how something is built, how would, say in your example, part of the design philosophy was cost and speed of construction, would not that limit your choice on how it was built? So guiding the out comes.
    Within their peer group do naval architect structural engineer “thinks” or “feels” or “believes” their way is better? I would think it was only human.
    What changes and allowances would you make to your calculations for bending /buckling for the compound curvature of the plate. Also assuming the edges are fixed to some degree and they have been dragged kicking and screaming to their final position.
    How much extra stiffness/support could be added, by the size, shape and number of the longitudinal members if things were scaled up to reduce buckling.

    LyndonJ thankyou for bringing up “actual loads” this I would think is an important factor in the failure or other wise of any structure. Just what are they? Washed over... pounded on... hit by... collided with.

    Apex1 my religion is I confess, “I am a curious *******” we practice zen and saying why.

    RAraujo welcome
  9. tazmann
    Joined: Aug 2005
    Posts: 329
    Likes: 17, Points: 18, Legacy Rep: 215
    Location: California

    tazmann Senior Member

    Thankyou Adhoc
    That did help
  10. Northman
    Joined: Mar 2007
    Posts: 87
    Likes: 9, Points: 8, Legacy Rep: 137
    Location: Norway

    Northman Junior Member


    Finally some real information!
    I suppose there are more than just me following this and similar discussions who are much more interested in why something does or does not work instead of hearing categoric statements that it works or not! AdHoc, I trully appreciate you taking the time! As an amateur it is for this kind of information that I follow the discussions on this forum!
  11. Brent Swain
    Joined: Mar 2002
    Posts: 951
    Likes: 35, Points: 0, Legacy Rep: -12
    Location: British Columbia

    Brent Swain Member

    A three ton come along can get quite heavily loaded in pulling the hull together. I recommend renting a big one for the first couple of days to get the hull together, then you won't need it any more.
    When determining panel size, consider the space between longitudinals as panel size between supports. Then consider the longitudinal curves as transverse support. The entire longitudinal panel is not unsupported for it's entire length, in curved topsides.
    I have never claimed to have invented origami metal working . The ancient Egyptians used the same methods, and it has been standard sheet metal working technique for as long as sheet metal has been around. Any sheet matal worker can confirm this . It simply hasn't been used as much for boat building, as when they started building boats out of metal , instead of asking for advice from sheet metal workers, they made the mistake of asking wooden boat builders ,and adapted methods suitable for wood, a material which only has strength in one direction , and not necessarily the best for a material which has equal strength in all directions.
    Galilaeo, Columbus, Archimedes, Copernicus etc etc all went against the accepted beliefs of their time . If no one ever did that, we would still be in the stone age. All technological progress comes from someone questioning, and not blindly accepting , the standard practises of their time. Those who don't do nothing to advance technology
    More math
    Here is some of the math used by origami critics.
    Fibreglass at a tensile strength of around 15,000 PSI ,and molded fir at a tensile strength of 1500 PSI doesn't need frames, as it is strong enough without them. Steel at a tensile strengths of 60,000 PSI , 40 times the strength to weight ratio of wood , needs frames ,as it is not strong enough. Curved shapes make frames unneccesary on molded fir at 1500 PSI and fibreglass at 15,000 psi ,but the same curves don't make steel strong enough, at 60,000 PSI !
    Fibreglass and wood, which will break when sharply folded 90 degrees, is sure tough, but steel ,which can be folded 180 degrees without breaking, is not as tough.
    Fibreglass and wood, which can be stretched very little, without breaking, is tough enough , but steel ,which can stretch to 1.4 times it's original length before breaking , needs additional reinforcement to be tough enough.
    If a fibreglass boat gets blown ashore in 8 ft surf , it will break up in 20 minutes or less, and wood in far less time . If it is Lloyds approved, it is strong enough, but if an origami steel boat blows ashore in the same conditions, and is pulled of thru 8 ft surf 16 days later , with minimal damage, but is not Lloyds approved, then it is not strong enough. A cruiser would be much wiser and safer in the boat which breaks up in 20 minutes or far less, in the same conditions, that in one that doesn't break up in 16 days , if the later is not Lloyds approved.
    If a boat blows onto a Fijian coral reef in 8 ft surf and is pounded across 300 meters of coral before stopping , without serious damage, it is not strong enough, if it is not Lloyds approved. If a fibreglass boat blows ashore in the same conditions, and breaks up in minutes , it is far stronger than the former , if it is Lloyds approved.
    Structulal shortcomings can hide for 30 years in extreme torture tests in dozen of boats doing everything from circumnavigations , to pounding on lee shores in huge surf and single season passage thru the NW passage , without showing themselves in any way. Only on the beginning day of year 31 will the entire boat suddenly fall apart , and all others like her, like the wonderful one horse shay~!
  12. LyndonJ
    Joined: May 2008
    Posts: 295
    Likes: 19, Points: 0, Legacy Rep: 233
    Location: Australia

    LyndonJ Senior Member

    Keh ???

    Says who? In what sized vessel ?

    Transverse support comes from bulheads and partial bulkheads too, these count as transverese framing.

    Also it might help the discussion if you use the right terms, material toughness, stiffness and strength are all quite different.

    These circular arguments always come back to 30 something foot boats, but you were advocating up to 60 foot boats without transverse support.

    Do you know any basic beam theory? It helps understand why the length of an unsupported frame loses capacity to the length squared.
    A long framed hull uses bulkheads to limit the frame length. This is a really important concept. Half the support length and it is 4 times stronger.
  13. bhnautika
    Joined: Feb 2006
    Posts: 852
    Likes: 57, Points: 38, Legacy Rep: 571
    Location: australia

    bhnautika Senior Member

    LyndonJ are you saying it is not possible engineer a cored composite boat without transverse frames support. Excepting of the weight and thickness penalties.
    In “basic beam theory” length is not the only thing, material properties, sectional shape and so on play a roll in their resistance to force and that is Brent’s point about materials, are the properties of the material in question being used to full advantage?
  14. Brent Swain
    Joined: Mar 2002
    Posts: 951
    Likes: 35, Points: 0, Legacy Rep: -12
    Location: British Columbia

    Brent Swain Member

    If you use slightly tapered bricks to build a Roman arch, it stays in place, and is extremely resistant to any weight put on it. Even if you don't use any mortar, it still stays in place. If , as according to your theory, the inside of the arch is under tension, what keeps it from opening up? Faith? Lloyds, Allah and his brother Daniel all work in strange ways.
    You can save yourself from scratching a hole in your head, by simply making up a sheet metal, framless model, and see for yourself the principles involved. Until you do, it is obvious some of you will never be capable of comprehending the principles involved, and how things like shape, and the connection between the various shapes which make up a sailing hull and decks, support one another.

  15. Morgan99
    Joined: May 2010
    Posts: 1
    Likes: 0, Points: 0, Legacy Rep: 10
    Location: United Kingdom

    Morgan99 New Member

    Sorry don't have any idea about this.So can't help you.Hope some one help you the best...

    london electrician
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.
Thread Status:
Not open for further replies.