Transverse frame calculation

[MikeJohns]

The pre-stressed shape is considerably weaker than one built without any 'inbuilt' stress.

The stress adds to the pre-stress and it will fail under a considerably lower point load.

This method of construction is a poor choice for a structure and the longitudinals and plate should be pre bent before welding. Doing this will make it a lot stronger for the same amount of material. Particulalry for impacts.


In simple terms, the residual stress produces a structure that wants to straighten out, the only thing stopping it is the restriant of being welded in place. When a load is applied the side gives more readily undergoes a local stress reversal and transfers the stress additionally to the already stressed adjacent structure. The result is that it dents more easily and those dents will be harder to straighten, it will also collapses more easily and under a much lower impact. By pre-stressing the curve you lose the structural advatages of the curve. It also has a much higher and larger area of elastic deflection from impact.

This is quite significant, boats built this way are arguably a little bit stronger for an evenly distributed load ( like a hydrostatic load ) , but for any impact they are weaker by a surprising factor .

This is why class societies are shy of 'novel' construction methods unless you can show them an analysis. They are conservative by necessity.

Bad news for you Brent I'm sorry.

[bhnautika]

Thanks for the input mike, Your statement “This is quite significant, boats built this way are arguably a little bit stronger for an evenly distributed load (like a hydrostatic load), but for any impact they are weaker by a surprising factor” I found interesting and made me think that very few if any pleasure craft are designed for impact resistance so are we getting hung up on all the coral bouncing. In your beam model of the longitudinal stresses what added strength is add by the athwart ship/sectional shape i.e. smaller panel size, curvature of plate and deck and chine connection?

[RAraujo]

Sorry Mike but I didn't understood your statement "The stress adds to the pre-stress...".

The pre-stress induces compression in the inner fibres and tension in the outer fibers.

The outside load induces compression on the outer fibres and compression in the inner fibers (assuming the load is acting on the convex side).

They surelly add but since they have opposite signs a one will cancel (at least partially) the other...

Lets take, as an experiment, a plastic ruler. Bent it slightly and then apply a force on the convex side.

Now let it be more or less straight (without applying the pre-stress) and apply a load.

Which is more difficult to deform? I know that the shape also contributes and that the results are not immediatley comparable as one is straight and the other slightly curved but since the curvature is small...

Or maybe we not seeing it the same way...

[MikeJohns]

Originally Posted by bhnautika

Thanks for the input mike, Your statement “This is quite significant, boats built this way are arguably a little bit stronger for an evenly distributed load (like a hydrostatic load), but for any impact they are weaker by a surprising factor” I found interesting and made me think that very few if any pleasure craft are designed for impact resistance so are we getting hung up on all the coral bouncing. In your beam model of the longitudinal stresses what added strength is add by the athwart ship/sectional shape i.e. smaller panel size, curvature of plate and deck and chine connection?

Specifically it puts some measure of comparisonof strength between metal building methods. Also the anecdotes just become anecdotes for metal construction rather than the framing method.

Pleae note I’m not designing a boat here, just illustrating some fallacies presented by an advocate of the method. [
Impact...., well significantly impact can be slamming loads in heavy weather and not just collision. Although I think it’s important to consider collision in a blue water or high latitude cruiser.
We always want the most efficient use of material structurally and either you can argue that it makes the panels fairer and it’s quicker, or you go slower and get stronger. Not the message so often presented that you can have your cake and eat it
Not only is it weaker but in failure it incorporates the whole structure without some division such as bulkheads or transverse frames.

The panel connections always stiffen and restrain by altering the panel aspect ratio. Analysis is indicative of what happens and where the structural arguments go awry and simple beam theory is clear enough to indicate this.

I haven’t modeled the static even load case yet, it is possible that elastic instability will occur there depending on the variation of radii of curvature, I was just reasoning that the neutral axis will shift as the load increased and the pre-load would need to be overcome first but it really depends on the overall shape whether it is ‘stable’. It is possible it will fail under a lower static load than the pre-frame post plate method with the same framing. And if we add wracking hogging and sagging stresses it alters again and this is where large versions made to this method are highly inadvisable

In effect you have a structure which if you load locally it will not shift the stress into a stronger more rigid member but will just add to the existing pre-stress in another part. If you like; the whole concept has an in-built failure mechanism. Not an in built enhancement as advertised.

This is not a damning of Brents 36 footers and I’m not saying they are unsafe or not strong enough, but they could be built stronger by not pre-attaching the framing before pulling the hull together.


cheers

[MikeJohns]

Originally Posted by RAraujo

Sorry Mike but I didn't understood your statement "The stress adds to the pre-stress..."...

The framing is in pre-existing compression, the whole pre-stressed curved structure is less stiff and yields elastically more than the conventional bent frame. There is a considerable transfer of stress towards the restrianed ends of the structure.
The compressive stress is significantly increased here and they buckle much earlier than in the pre bent frame structure without pre-residual-stress.

Go to your ruler, analogy: probably better to use a mylar ruler or strip so you can see the effect scaled.

Hold the ends well supported in your fists, curve the middle up a little and get someone to push on it anywhere along the span with a finger. You'll se it change shape as you depress flattenning in the middle and bulge in a tighter radii towards the ends in response.

[LyndonJ]

Originally Posted by Brent Swain

Curvature is a huge factor in adding to the strength and stiffeness of a hull, such that I have found transverse frames totally redundant, and useless on boats under 60 feet. The longitudinals are far more relavant in resisting distortion on the highly curved hull panels of a sailing hull, as they are arcs under compression.
Any rules that dont take this into account are grossly inacurate and irrelevant.

So Brent this is just a beleif system that's being preached endlessly and also promoted in "The book" ?
The only problem with this beleif is that it's about the mechanics of solids , and preaching that to a company of engineers and boatbuilders has seen the structural collapse of your soap box so to speak.


If you want to know just how much weaker your method is then give the radius of curvature, length and the force required to pull the panel in.

Are you interested?

[bhnautika]

LyndonJ count me in! What ya got.
Mike thanks again for your reasoned response. To me looking at the construction of the hull shell it seems that the ends of the boat have the most pre stress built in, where the mid section, because of the chines, some of that stress has been taken out of the shell in the top sides and transferred to the longitudinal stiffeners. My question would be as it would seem that structures under load are on the road to failure would it be fair to say that what they are stressed too along that route would be something that you factor in as to the overall strength of the part and its resistance to bending to a given load. I would agree after what you have said that it would be stronger adding the frames after (we are talking cut frames?), and would not add that much to the cost and in time. I would really like to see a hull section amidship to get a look the spans and the general layout of other structure.

[Brent Swain]

Originally Posted by MikeJohns

Heres a T section beam in pre-arched pre-stressed then point loaded mid way with a moderate load . all well below failure.

I pre-stressed the beam as generously as possible before the equivalent of brents longintudinal would collapse from buckling.

The displacments from the load are scaled to show 5 times the actual displacement to help illustrate the concept.

The actual structure would be a lot more slender than my choice here but I wanted to show the points of contraflexure, where the stress reversal occurs.
Negative stress is compression, look for the neutral axis.

In reality with a more slender longitudinal the contraflecure points will be much closer to the beam ends.

It should be clear enough to show that the longitudinal frame becomes a tension member under the load and that it has no analogy to an arch at all.

As you compress the middle the ends must move outward. Where do they go, while being welded to the plate?In order for them to move the plate they are welded to has to stretch diagonally at 60,000 psi tensile, and from chine to chine. For the ends to bulge , the plate has to stretch between the chine and deck line at 60,000 PSI. This clarifies the difference between calculating the strength of a longitudinal in isolation and that welded to a hull, showing the fallacy of calculating the compression strength of a longitudinal or frame in isolation. In a hull it is anything but in isolation...
Again, the problem with most of your calculations is it calculates the strength of the individual member, without even remotely considering the extreme complexity of its attachment to the rest of a hull , giving structural strength far beyond what your simplistic calculations are capable of comprehending.

[Guest62110524]

Originally Posted by Brent Swain

As you compress the middle the ends must move outward. Where do they go, while being welded to the plate?In order for them to move the plate they are welded to has to stretch diagonally at 60,000 psi tensile, and from chine to chine. For the ends to bulge , the plate has to stretch between the chine and deck line at 60,000 PSI. This clarifies the difference between calculating the strength of a longitudinal in isolation and that welded to a hull, showing the fallacy of calculating the compression strength of a longitudinal or frame in isolation. In a hull it is anything but in isolation...
Again, the problem with most of your calculations is it calculates the strength of the individual member, without even remotely considering the extreme complexity of its attachment to the rest of a hull , giving structural strength far beyond what your simplistic calculations are capable of comprehending.

i did not read the whole thread and so forgive if I missed
From what I have seen Brent you yachts are very modestly canvassed
A 60 foot yacht may have a main of 1000 sq/ft According to your logic, there is no need to have a ring frame by way of Chain pl. mast area I can see the deck arcing up massively and pulling the sides in so as to lose any tension you may have on the shrouds Which leans me to ask, how can you tension the shrouds, answer is you can not

[Jeff]

< mod note: I've tried to clean this thread up a bit. Please, let's keep the discussion about design without personal attacks >

[LyndonJ]

Originally Posted by Brent Swain

As you compress the middle the ends must move outward. Where do they go, .............. considering the extreme complexity of its attachment to the rest of a hull , giving structural strength far beyond what your simplistic calculations are capable of comprehending.

Brent

You were wrong on the arch analogy and that should be abundantly clear to everyone by now.

You are implying that extreme complexities in your structure make engineering analysis incorrect............................ that's rubbish.

Now the FEA is 'indicative' There's a big difference between an FEA simulation and a calculation. FEA is very accurate and allows complex structures to be analysed that could not have been calculated. Thats how you got away with this is the past, the tools were not there to show how your thinking was wrong. Bit now they are.

Don't you think it was already considered that the ends are restrained?

This 'bulging' would be hardly noticable while adding some very real stress , you are probably not going to notice it until the structure actually buckles.
Your argument about it being restrianed by the stem and the deck stand for every boat regardless so indicative comparisson is quite valid.
Also you need to think about were all that stress actually goes ( aka structural design). There is no magic, just cold hard mechanics.



Whoosh makes a good point about rig - righting moment-keel loads too on something as large as 60 feet.

So really to convince you now you are going to have to see the entire side of the hull modelled. Even just the topsides should be convincing enough. So can you could furnish say a mid topside waterline to get the curve, the framing detail and an estimate of the force required to pull it in ( although that I can calculate ) oh and the plating thickness.

Have you ever thought that the arrangement you end up with produces a shape that wants to straighten out and is assisited in doing just that from an external load ? [edit added ] the fact that the ends are restrianed acts to resist it straightening out but that 'resistance' aka stress is additional to the pre-stress: Thats what this is all about.

[tazmann]

I'm not getting in the middle of this, I'm just a farm code engineer- welder but my 26 is built the same as the 31 or 36 just lighter material. Stu to give you visual on the chain plates my hull sides are projected up an 1-1/2" from the deck, 3/4" sch 40 pipe is fully welded to the top of that. my chain plates are 4" wide 3/8" thick by about 3" tall stainless welded to the top of the 3/4" pipe 2 per side. forward lowers attach to trunk cabin corners, bridge under mast step 3/8" x 5" wide carried around to cabin sides with supports running down to the chines. Beleave me there is no flex or bulging up the decks there. I have a second pair farther back at the balance point I lift the boat with , no flex there eather.
Tom

[Guest62110524]

Originally Posted by tazmann

I'm not getting in the middle of this, I'm just a farm code engineer- welder but my 26 is built the same as the 31 or 36 just lighter material. Stu to give you visual on the chain plates my hull sides are projected up an 1-1/2" from the deck, 3/4" sch 40 pipe is fully welded to the top of that. my chain plates are 4" wide 3/8" thick by about 3" tall stainless welded to the top of the 3/4" pipe 2 per side. forward lowers attach to trunk cabin corners, bridge under mast step 3/8" x 5" wide carried around to cabin sides with supports running down to the chines. Beleave me there is no flex or bulging up the decks there. I have a second pair farther back at the balance point I lift the boat with , no flex there eather.
Tom

Mate, comparing a 60 footer to your boat is like comparing a farmall cub wheel tractor with a D11 quarry cat, the loads are enourmous, how are you?

[tazmann]

Originally Posted by whoosh

Mate, comparing a 60 footer to your boat is like comparing a farmall cub wheel tractor with a D11 quarry cat, the loads are enourmous, how are you?

Hello Stu, still hanging in there . You buiding yet?
I wasnt comparing to a 60' , just the others that Brent builds and sells plans for. If it works on his 36' I don't see why it wouldn't work going larger provided you upped the shell plate and longs to compensate but it does seem to me that you would get to a point the weight would be a killer vs going to a framed boat. After doing my 26 I could only emagine trying to pull together 60 to 70' hull using 1/2" plate LOL not me no thanks
Tom

[bhnautika]

LyndonJ I think you are jumping ahead a bit far saying that the FEA analyses of the pre stressed curved beam some how relates to the globe structure of the stresses in the hull, I think it’s a bit more complicated than that. You talk of stress but we haven’t ascertained what loads we want to be dealing with or how they react with in a structure we haven’t defined. If you think about it there are a few examples of pre stress in boats already which are excepted, cold moulded timber hulls as RAraujo mentioned, both in small and large boats, some even have bent frames put in after. The steel plate on a framed chine boat has some, and then there is the standing rig of a yacht. So far we have some proof, some opinion but no context! Yes there is pre stress but where does it sit in the good, bad scale. Just how much pre stress can it live with?