composite pressure vessel

[gio1]

I need to carry out a FEA on a carbon fibre pressure vessel subject to high pressure.

The vessel consists of an inner aluminum thin liner core which is used as a mandrel for the filament winding process.

I have not much experience of composite FEA, although I am familiar with laminar classical theory. How would you suggest approaching this problem? My first assumption would be to neglect the aluminum and assume the composite is bearing all the load. However due to the manufacturing process the "plies" are not easily identified. How would you recommend modelling the different winding directions? (0,55,90)

Thanks

Gio1

 

[gio1]

I forgot to mention I will be using abaqus

 

[GBor]

I'm not terribly familiar with Abaqus, but if the resin system adheres to the aluminum, then just include it in your composite lay-up. Give it homogeneous directional properties (E1=E2=10e6, v12=v13=v23=0.3, G12=G13=G23=3e6 "off the top of my head").

The difficulty in composite FEA is that you have to orient the stack-up to the global axes and then orient each layer to the global orientation. Typcially, I align the "0" ply along one of the global axes. Then, each layer is usually just an angle input where you input the material properties for the layer.

Garland E. Borowski, PE
Borowski Engineering & Analytical Services, Inc.
Lower Alabama SolidWorks Users Group

 

[rb1957]

i'd have thought that the strength of the composite was in the carbon windings. it's easy to figure out the hoop stress or load per inch length, then figure out how many windings there are.

but definitely use different directions for the windings (as they have to react hoop and endload from the pressure vessel). you know the endload is 1/2 the hoop, so you could have 2 windings at 30deg and one in the hoop (0deg)which would give you an equivalent of 1 endload and sqrt3 +1 in the hoop direction. realistically you probably want 45deg, maybe 60 deg, 30 deg and 0 deg windings.

good luck !