Transverse frame calculation

This is because the forces from gas are very uniform, so that the metal skin is essentially all in tension. Unfortunately this doesn't work as well for unsupported structures experiencing external pressure http://www.youtube.com/watch?v=Zz95_VvTxZM and http://www.youtube.com/watch?v=Uy-SN5j1ogk

Similarly if you bend a plastic straw everything is fine and dandy up until the point of failure.

Now to get back to your geometry problem...

A cube is MUCH stronger at the corners and the edges than a sphere to point loads. Simple geometry shows that if I put the force on a corner it is transfered into the edges/faces in a pure compression mode. Similarly if one of the faces fails, the other walls will still support it. So while the failure strength of the cube may be less than the sphere, the cube will fail more gracefully. But again, it really comes down to "it depends".

Much of the strength from curvature you speak of really has nothing at all to do with curvature, but translating the forces that act on the structure from torsional or sheer forces into compressional or tensional. This sacrifices the strength in those other modes. I think it was Apex1 who pointed out earlier in the thread or in another thread that if you bent a steel member and compressed it endwise, it is essentially partially prefailed. In one direction, though strengthened in the other.

Again if we look at your propane tank. You mention the 100 PSI. I can engineer plastic bags enclosed in cordura in a variety of shapes that can take thousands of internal PSI. But it all the other modes it is floppy. Unless it is internally pressurized (think an infinite number of internal structural supports). If I puncture the skin, those internal supports fly out thru the hole. And it responds to puncturing very poorly. Similarly most large rocket engines are thin (relative to weight/volume) and that works for them, as the forces are mostly collinear with the skin/tube structure, the internal pressure form the tube helps, but if it tweaks a little the **** hits the fan real fast as the thing cumples and unzippers itself. Welcome to static vs. dynamic loads!

 

I suggest that you do, it is worth the read. You have missed out on many issues which have been raised and very succinctly shown to be a gross misrepresentation, false or otherwise.

No strictly correct. The local loads are driven by panel aspect ratio, ie length by breadth and the span of the member. So a transverse frame spacing of say 1.5m with a long.t stiffener spacing of 300mm, is identical to a panel where the trans frame spacing is 300mm and the long.t stiffeners are spaced at 1.50m, ie longitudinally framed. The only difference is the span, between supports of the structural member.

There is also an error in LR SSC rules in the way it calculates these members (based upon aspect ratios) and its internal assumptions when the algorithm kicks in to compute. Thus giving erroneous results.

The same for deck girders/frames. The computation in LR SSC rules is stress based and makes several assumptions. Again this is incorrect and results in a far more flexible structure than should be expected, despite the stress levels checks being ok. If based upon deflection, as should be especially with aluminium, the resulting stiffness requirements are greater than actually given by the SSC rules. It incorrectly applies 1/2 load not full laod.

Being over 80m, global loads also come into play, not just the local pressures/laods.

I have highlighted both these issues to LR several times; since i am on LRs technical committee (we review and formulate or amend new rules). This has been on-going for several years…progress is slow!

There have been so many gross assumptions made by the authors to arrive at an answer that this is somewhat meaningless, they even in a round about way say this. Without showing how they arrived and what criteria they applied to calculate these “differences” no value can be placed on any result nor conclusion.


Finally, the report is based upon an 80m motor yacht, which is built to satisfy Class rules. Brent’s boats do not pass Class rules nor are they 80m in length.

Therefore despite the “minor” errors and assumptions made in the report, it is comparing chalk and cheese again I’m afraid, and maybe shall just confuse the issues more?

 

Thanks for the insight. I am not convinced that some flexibility is a bad thing. The common practice is to build as rigid as possible, but FEA lets us design stuff with much more confidence. You are the NA, so I'll take your word for it (without me doing the math myself [because I am lazy]) in regards to the strength of the various configurations, however with the caveat that it may be more important to align your strengths in certain directions ala in the composites world this can be make or break. If you think of Unidirectional s2 glass as being on one end of the continuum and "stringerlets" on the other they are "sort of" the same idea (minus the direction normal to the surface).
Similarly in a composite sandwich construction, the foam core is "sort of" like having an infinite number of longitudinal and transverse cross members (of relatively small cross section). I think for many composite boats, esp catamarans, there is some serious skimping on the cross sectional supports, and it may help to have some more bulkheads or "stringer like" structures. even at the expense of some internal useable volume, or style points.

As I noted in my other post, scale is definitely a major factor.

80m boats are not the same as 20m boats or 10m boats. But unless I am mistaken most of the variation is global stuff. Like the size of waves relative to the size of the boat, etc. Things like the wind pressure on a small boat may be irrelevant except for making speed against a headwind. But on a large structure, that could be a critical factor. A few PSI can add up to thousands or even tens or hundreds of thousands of pounds of force over a structure.

 

Nice of you to say so, thank you…. But don’t just take my word for it, you should convince yourself too, by using the rules and ascertaining the limitations to understand them better. Unless my précis, and also given by others in this thread, is suffice for your needs.

Well aeroplane wings flex, they are ok. As with everything in design, it is all about understanding the loads how to apply the loads where to apply the loads and then ascertain the fail-safe criteria to arrive at a satisfactory structural design.

 

Ok, so without doing the math (again laziness), but using your previous example.

1#
Transverse Frames 300mm
Longitudinal Frame 1500mm

2#
Transverse Frames 300mm
Longitudinal Frame 1500mm

Identical?

I guess in once sense they are exactly identical in proportion, weight, strength etc, except for one thing. Direction.

I think if the supports were the same thickness, and depth it would seem unlikely that they would be just as strong beam to beam and bow to stern, or even diagonally in torsion.

Of course you could mix it up, but I think that is what the article was saying... Though I am going to fall back to my FEA and "it depends" statements.

Suppose these configurations involving deep supports and shallow supports. (assuming a boat of fairly high L/D aspect)

Assuming no other internal structural support and uniform geometry. All are half assed guestimations based on conjecture. More meant for inquiry. Structure assumes that the skin is a structural member and is fully enclosed, ie, not a U shape but an O or something similar (boat shaped). As always, devil is in the details, assume any other details are identical.

1# Strongest from bow/stern collision?
Transverse Frames 300mmx50mm
Longitudinal Frame 1500mmx300mm

2# Strongest versus compression?
Transverse Frames 1500mmx50mm
Longitudinal Frame 300mmx300mm

3# strongest in twisting/torsion?
Transverse Frames 300mmx300mm
Longitudinal Frame 1500mmx50mm

4# Strongest in bending bow to stern? Strongest from beam side impact?
Transverse Frames 1500mmx300mm
Longitudinal Frame 300mmx50mm

 

I used to work doing large event management. If we do this lets make it an event!

 

Direction has no part in local scantlings. It is simply a load being applied locally. As such how you arrange the structure will dictate what scantlings are required.

No.

The direction does not play a part in local scantlings, in the sense you mean. The aspect ratio, is unchanged, regardless of the orientation of the panel. So whether transverse or longitudinally orientated, a 1.5 x 0.3m is the same panel size and aspect ratio.

The span, the span of the structural member that is supporting the panel is the influencing factor. (Save for a pure buckling load of course...but the critical buckling load may be higher than a localised high shear laod).

For example. If the frame spacing is 1.50m and stiffeners at 300mm, what is supporting the stiffeners??...the frames. What is supporting the frames??....that is for you to decide. Thus where the frame is supported dictates the span. If the only 'support' is the keelson and chine...and if that span is 4.0m, that shall yeild a very different stiffness requirment from a span of 2.0m. Yet the aspect ratio and panel size are unchanged!

For any given pressure loading, double the span and the stiffness requirements goes up by the square, that is, it is proportional to L^2. So, a beam of span with 2.0m has 4 times the stiffness of a beam of 1.0m. Thus, how you arrange the structural supports influences the stiffness and hence the weight.

So, the span, or supports, affects your design, not the direction. Since the ‘direction’ is implicit in the ‘local’ analysis.

As for which arrangements ie the best?….that needs to be investigated. No two boats are the same. The worst case for a 10m yacht shall be very different from an 80m yacht. That is the role of the naval architect. To establish the load cases, and then apply the loads accordingly. This of course is greatly influenced by the sea state that the vessels speed at which she shall be sailing.

So, 40 knots in 4m Hs, is very different from 20knots in 4m Hs and is different from 10knots in 4.0m Hs, the loads the vessel is subjected to, in each case, yields different structure.

You set the limits of the vessel, and then you design to those limits. And it is your role to investigate all the known loadings and then and only then can you say where the structure will fail first and why. Beit buckling, bending, shear or whatever... Reducing it to just a panel size debate between various failure modes, does not address what structural design is, and, with respect, misses the whole point of this thread.

 

Note taken, typical misinformation direct from Wiki, should have checked it. Boy, your feet, toes, pints and stones are sometimes too much for a finnish engineer

 

Like boats, there's hoses and hoses. I've installed lots of firewater systems over the years, mostly offshore and in yards, and most systems are held at a lower pressure with a jockey pump until someone discharges the pressure faster than the jockey pump will deliver. When the main firewater pump kicks in it will run at as much as 15-20 Bar.

Anyone who has ever been on a fire-fighting course will know that it is a two-man job to hold a 2" or 1-1/2" branch aimed at a fire at this pressure. Any higher pressure would bowl them over.

The reasons why hoses are rated for 30 bar wp is probably down to the shock load that is delivered when a large fire-pump kicks in, It has been known to blow flanges off the end of the lines. In the earlier North Sea platforms the firemains were installed by domestic fire-fighting companies, but after a few accomodations got flushed out the work has been handled by the process piping contractors who generally install welded cunifer systems.

 

Hello and welcome to the forum

I see AdHoc has already posted on this but I'll add that a lot of confusion occurs over the nomenclature of framing systems.
For example what is termed a longitudinally framed vessel usually has large transverse web frames and bulkheads supporting a lighter more closely spaced system of longitudinals.So the presence of transverse frames is common to both types of framing.
Small transverse framed vessels may have no longitudinals at all, but nearly all longitudinal framed vessels will have some transverse framing or support such as bulkheads or partial bulkheads or web frames.

That can be confusing to many people for a start.

That paper is idealistic. The optimal framing method will always depend on the design of the vessel, its hull form the number and spacing of bulkheads and the requirements of each part of the hull.

 

Thank you,

Even though Ad Hoc has addressed this already, I think that the gist of the paper wasn't either or, but "more or less" and the figures themselves had both longitudinal and transverse framing in varying amounts. Similarly the question of scale, aspect, function, material etc. are all driving conditions.

Over simplistically.... The argument in this thread is sort of like I have a vacuum hose, and the discussion is about how many (if any) rigid rings should there be per unit length, so that the hose doesn't collapse. Essentially the sides are:

The professionals:
A qualified "It depends" this **** is complicated, one needs to know the design parameters before an answer can be given. All things being equal its best to err on the side of not getting killed.

Brent:
I've been using XYZ type hose for 30 years and it don't have any rings, I even stepped on it a few times!

I can see both sides of this. There is something to be said for Brent's attitude in re: backyard engineering. One of my favorite activities is offroad truck racing, and that whole industry is built around people building their own vehicles in their garages. Most people aren't engineers, but there are racing classes and sanctioning bodies that have minimum safety regulations. Are the safety regulations overkill? Well, maybe you could get by with 0.083" 1.5" CRW tubing instead of .120 2.0" DOM? But the difference between a 30 mph, and a 45, or 60mph let alone a 100mph crash is a different universe. Let alone just the forces on the suspension, drivechain etc. Keeping in mind that most racers aren't engineers, or even professional welders, the classes have some safety margin so that your booger-welds are well supported if your day job isn't welding pipe professionally. So, if you ask me, its better to be safe than a little faster and lighter, besides, stronger may be faster, or at least more reliably faster.

However, there is an important implicit point in this... That is for automotive "racing". I think if Brent were building racing yachts meant to "push the envelope" and he was trying to shave off as many kilograms as possible from the gross displacement weight in order to compete in the Route de Rhum, you might hazard him to think it over, have better escape hatches, rafts, radios etc etc. How many pounds of structure is worth sacrificing for safety? Then again, small boat racing in the modern age with a heavy steel monohull seems a bit silly anyway.

But there is another thing to consider, these aren't day sailers, or racers. People live on these boats. And as has been said. Monohulls also have two points of stability.


Floating upright, or at the bottom of the drink.

 

Some propane bottles have a concave bottom which gives the equivalent pressure from the ouside of the dome, as you point out. Plastic bag won't take presure form that direction, either , but the concave bottom of some propane tanks will.
As you pouint out with pre stressed longitudinals , under pressure, it is weaker on one direction and stronger in another , compression, which is what we need.

 

For example. If the frame spacing is 1.50m and stiffeners at 300mm, what is supporting the stiffeners??...the frames.

What is supporting the stiffeners? Longitudinal curves , which are in turn supported by the chines and decks . Flat sided 80 ft powerboats don't have this advantage. Sailing hulls do.

 

3# strongest in twisting/torsion?
Transverse Frames 300mmx300mm
Longitudinal Frame 1500mmx50mm

4# Strongest in bending bow to stern? Strongest from beam side impact?
Transverse Frames 1500mmx300mm
Longitudinal Frame 300mmx50mm[/QUOTE]

Transverse frames have little effect in resisting twisting torsion , or bending from bow to stern. That is done by hull and deck plating resisting being stretched diagnonally . You can easily see this in a model with no decks on. Adding frames has no effect on twisting . Putting the decks on makes a huge difference. To bend fore and aft , hull or deck plates have to stretch longitudinaly, at 60,000 PSI.
Some have spoken about what head of water a hull surface will take. In many Pacific crossings I've never had more than three feet of head against my hull. The hull's buoyancy stops it from geting any higher. She pops up like a cork.That is not the case in much larger hulls which cross several waves at a time , and thus can't rise. Thus such calculations for huge ships are irrelevant for smaller craft.

 

I like Brent he design boat, he write about boat, he build boat and he got the courage to sail his boats which is more than people who write books about fixing boat but never fix anything or design boat but never build anything do. He may not have the superior academics record, the superior brains and the superior tools than other posters in this forum have. He may have just inherited that ardent desire of building boats from Vikings, Phoenician or others would sought to improve on a dugout. Like Herbulot is endeavour of producing affordable craft is noble. Now is the challenge for people with superior capacity: provide a comprehensive study of a frameless boat versus an affordable adequately framed boat bearing in mind that a boat of the size in question (60 ft?)to me is not just a tube or a plate but a complex structural hollow section.
By the way I was in a ship, NA and NE designed the best in his class (oil industry), also the most expensive that capsized and sunk in less than five minutes taking with him 8 men. Obviously nobody is perfect.