Composite I-beam

General Composites Discussions Engineering Talk
#1

I have been asked to build a composite I-beam for a guy and he sent me his design.

He wants 60mm (web) x 50mm (flange) I beam made from 4 layers of 150gsm carbon, with a 5mm core in the middle.

Now I would usually not care and just make the part but I suspect that the core in an I-beam would only help a bit with buckling and not much else.I just don’t want to make it harder for me to build. It might even be lighter and stronger without the core and a few layers of bi-ax extra.

Is the core worth it? What are your thoughts?

#2

The core greatly helps the buckling strength of the shear web- leave it in. Make sure to interlace your shear layers with your cap layers since there’s a pretty high stress at the intersections.

What is the application?

#3

That makes the construction a bit more complex, but interesting. Im going to use a Nomex comb in the web and test the results.

It is for a solar powered race car that he is building. Fun stuff.

#4

Unless you know how to properly bond in honeycomb I would stick with Divinycell or end-grain balsa.

#5

You’re also loading nomex in shear there more than compression which it is designed for so it is a little pointless surely. So, yes, foam is easier to process, cheaper and may actually be stronger in that application (dont know shear values for both off the top of my head). May be a little heavier but I think the advantages of easier processing outweigh the weight disadvantage

#6

I don’t know how accurate this graph is, but it seems to substantiate what I thought.

But, to be perfectly honest, I prefer not to use prepreg for the first prototype, so I have decided against honeycomb. I think I will use balsa as it has a higher shear strength to weight ratio than Divinycell according to these sites.
http://www.gurit.com/files/documents/balsaflexwev1pdf.pdf
http://dragonplate.com/docs/divinycell.pdf

#7

Hmm, nice graph, I stand corrected! Dont suppose you have a link for that just for some future reading?

#8

Note that if you are using the core in the shear web only, that the loading will be in-plane shear, not interlaminar shear (assuming the I-beam is being loaded in flexure). End-grain balsa wood only really works when the grains are running perpendicular to the shearing force. If you want to use end grain balsa (and get and contribution from its shear properties) for this application I would recommend using 60mm thick core and have it run the height of the web… In reality the shear web carbon skins will provide much more shear stiffness than the balsa, so really all you need is a core that will stabilize the skins (help prevent buckling). You could most likely get away with mid-range density urethane foam.

#9

Thanks, I thought that with the area in shear the core would improve shear strength. I already bought balsa, and have some 75kg/m^3 foam core left. I will make test samples of both and see what works better. If the lighter foam is good enough then I would be happy.

#10

In pure bending your shear will be at 45 and in pure shear it will be at 90. So it depends on the shear/bending loading as to which prevails. Generally you design for 45 when laying up anisotropic materials like carbon fabric.

The carbon will be so much stiffer than the core that the orientation of the core should be optimized to stabilize the beam in buckling and crippling, not to provide shear strength. See Bruhn chapters C5 to C7. So in this case end grain balsa laid with the long transverse grain direction perpendicular to the shear web will be perfect if you don’t mind environmental factors like having to seal for humidity and decay.

#11

Thanks. Reading up as I type this, hopefully it will all come back to me.

There might be some torsion on on one beam, but only in a unlikely situation, but I need to account for that. I thought I would use some 30° biax on the web, would that help? The torsion and would be much less than the bending stress, so I don’t want to change the profile.

#12

Run your shear web layers at 45 and if you really have a lot of torsion then use a thicker core- from a structural efficiency standpoint that will help far more than adding more layers.

Don’t forget to really pay attention to the upper and lower transition areas of the shear web, it’s easy to get delamination there since the stress is high when you combine load cases like this.