Materials testing for T700UD / Solidworks?

[510waffles]

Hi,


I have been trying to do analysis on what can be described as a beam with a constantly changing cross section. It is is difficult to do CLT to get the displacements which I am concerned about.

I have 2 questions:

1.) I used to use ABAQUS and the properties I needed were E1, E2, V12, G12, G13, G23. I moved to a smaller company and we are using solidworks to do analysis. I am aware of the mesh type issue however in the material properties, it asks for E1,E2,E3,Nu12,Nu23,Nu13 instead. What is the difference between the two methods? Do I need to just use micromechanics to get the Nu12/Nu23/Nu13?


2.) T700s UD is fairly common material for composites. At the time of learning composites, we used AS-4 properties instead, but I no longer have those. How can i get the properties? Most of the data sheets I have seen give only E1,E2 and V12



3.)Besides the use of triangles are there any caveats to using the solidworks solver for composites?

4.)I can get E1,E2,V12,G12 with ASTM-3039 and ASTM-3518. I am always lost on how to get the last 2....



I have done this analysis before, but I have forgotten how I got the material data.





Thanks!

 

[ESPcomposites]

For item 1, the important properties (for typical thin laminates) are E1, E2, G12, V12. Beyond this, anything with a 3-component will usually have a minor influence for thin laminates. G13 and G23 affect transverse shear deformation and won't matter much unless the laminate is thick. Nu12 is just V12 so there is nothing new there. You can probably use a micromechanics solutions (or just estimate) N23 and N13 since a more accurate value wouldn't affect the result (as long as they are "reasonable" values). But it is odd that SW does not ask for G12, so are you actually using ply properties combined with a stacking sequence for the laminate?

For item 2, this goes back to item 1. E1, E2, G12, and V12 are probably the driving properties and the others could be estimated accurately enough if really needed (for the most part). If you don't have G12, you can probably estimate it accurately enough (based on a similar matrix). You didn't state the matrix, which is will drive the values for E2 and G12. But for multi-directional laminates, E2 and G12 are not highly influential for carbon fiber laminates. This is because carbon fiber is relatively stiff compared to the matrix (assuming a polymer). Because of this, some "typical" values for E2 and G12 would probably be have a very similar net deflection for the structure as the more accurate data would. By far the most important values is E1 (and E2 if it is a fabric). You could also bound the problem with some higher are lower values and see how much the change is. If there is a minor difference for your accuracy threshold (should be the case for most problems), then you don't need to worry about it.

--

Side note: If the section is deforming in just 2D space, then you can address the taper with the 1D Elements program linked below. You just need to determine the cross section's EI along the taper (and use the documentation on the website as well to address composites). If its a beam and it is short, then you would adjust the transverse shear deformation as well (function of in-plane shear modulus of the laminate if there is a web). This would be closer to a "classical" approach. One advantage is that you can quickly change the section properties and the taper via parameters (also easier to adjust the influence of the layup). If you were to model it with shell elements (instead of beams), that could be time consuming for various sizing configurations and laminates (depending on the FEA program).

https://www.structuralfea.com/software/software.html

Brian

http://www.espcomposites.com