Design of Large Lightweight tubes

[lengould]

I'm hoping someone could give me a quick feasibility evaluation of an idea, as I don't have access to a lot of engineering texts or money to hire the work. What I'm looking for is a means to calculate the mechanical load-carrying ability of a large diameter very thin wall composite tube or pipe.

Specifically a pipe fabricated from carbon fiber composite and where the primary design criterion is that the walls be capable of containing a pressure of 165 kpa guage. By my calculations, using eg. Panex PX35FBUDO150 unidir carbon fiber mat, 3.80 gpa, a 4:1 safety factor, a 30 degree winding angle and 60% fiber-to-resin ratio, the wall thickness for a 4.5 meter diameter tube would be 0.75 mm based on the hoop stress. The resulting tube would weigh 19.5 kg/meter.

My question then is, for a 240 meter long tube such as this, what sort of (mechanical loading could safely support)/(additional structural support requirement) as a beam supported at it's ends and the load suspended from the centre of the length, or how can I calculate that. Every formula I've found so far depends on a "given" modulus based on the details of tube diameter and wall thickness of industry standard metal tube sections, which is really of no help in this problem.

Thanks

 

[FredGarvin]

I would think that you would have to start with a standard beam analysis. The problem, as you stated, would be the material constants. How constant would the cross section of the tube be? If it is relatively constant then the area moment of inertia would be easy to calculate and the only question I would have then is what to use as the modulus of elasticity. Since the material is most likely not isotropic, I am not certain as to how this is dealt with.

.75mm is a pretty small wall thickness for such a large pipe diameter. That's impressive.
How does the rigidity of the composite compare to that of a standard steel, say 1018?

 

[PerennialII]

You could introduce the composite aspect by applying theory of laminated structures ... the different plies are then input as having orthotropic properties and when "stacked up" work up to a stiffness matrix for the structure (and you end up with a discontinuous stress distribution across the laminate thickness). This way the composite material properties will be accounted for correctly. FEM analysis is often the easiest way around these, but in this case could probably do with analytical solutions.

A really thin one ... can it stand its own weight without buckling ...

 

[lengould]

The idea of the walls is to have the highest ratio possible of fiber v.s. resin so the cross section should be very constant, if that means uniformly distributed. I was planning multiple layers of a very (low filament count)/(low section diameter) continuous end-to-end fiber wound in alternate directions, the layers wound at a 30 degree angle. I will try to come up with a reasonable statement of the materials stiffness in compression from comparable composites.

I know, it is a really thin one...and that's the basis of my question, e.g. does it need additional structural elements in a truss arrangement for support and if so, how much?

It seems reasonable to me to treat the tube as a series of joined truss tension members comprised of individual fibers which start at the bottom, curve up over the top at a 30 degree angle and return to the bottom at a point 8.16 meters further down the tube, which then has a perfect joint to the next "truss tension member". Remains then to determine if the top wall section of the tube can withstand the resulting longitudinal and radial compression forces without excess distortion / failure? Is that a useful track?

 

[PerennialII]

Would think that the 30 degree angle would be one of the keys in giving it some 2 - axial strength / stiffness. Since the internal loading is plain pressure I'd be pretty sure that a tube essentially made of carbon fibre wouldn't have any sort of principal problems in carrying the load. Essentially if the fibre/resin ratio is really high you could carry out an analysis of it's behavior by simply taking it as a single layer of orthotropic material which is at an angle ... you'll then get the pipe stiffness in the axial and radial directions by a coordinate transformation (of the orthotropic material property matrix) and so on.

Other than that I'd still make a stability analysis (by taking it as a cylindrical shell you can get the local buckling modes involved ... there's quite a bit of engineering material about this topic available) and one of the problems with that sort of a pipe can be how to join it to anything (if need to?) ... composite joints (or supports) of a very thin section can be tricky.

 

[lengould]

Thanks a lot for the input, all. It is at least encouraging to see no-one has yet flatly stated "it's impossible, we tried it", which is one of the things I am looking for.