Can cosmosworks analyze anisotropic material? I am trying to analyze a part made of fiberglass, and I want to see if Cosmosworks is worth buying for this.
In CosmosWorks 2003 and 2004 you are able to define your own material. So as long as you know the values of the material properties that you will need, Cosmosworks will do the rest.
Ahh.. but I think hes asking how to take into account the fibre direction in the composite. The properties // to the fibre are completely different than -|.
IE: E(sum //)= f(fibre)*E(fibre)+f(matrix)*E(matrix)
thats the general formula for the law of mixtures which governs the modulus of composites in the // direction.
and the one governing the modulus in the -| direction.
(in the above f is the volume fraction)
I think that the trouble is that cosmos only allows one value of modulus. Since in a composite the modulus can change greatly btw 0 and 90 deg you need to have that taken into account. One way, since the controlling mechanisims are generally linear with applied force direction, (ie 45deg applied force has the general effect of distribution among both directions) may be defining modulus in relation to reference planes would work?
I was thinking the other day that if you created an assembly of thin, long cylinders w/ a solid around them and then used the modulus of the two components you might have a composite structure. If you scaled this thing properly it would take friggin forever to mesh thou.
Just keep in mind that CosmosWorks is only a linear solver.
MadMango "There is no reason anyone would want a computer in their home."
Ken Olson, president, chairman and founder of Digital Equipment Corp., 1977
Have you read faq731-376 to make the best use of Eng-Tips Forums?
Right but even the linear behavior of composites (or more exactly esp. the linear behavior) is highly dependant on direction. Or to be exact again aligned fibre composites.
Do higher end FEA software generally allow anisotropic materials to be modeled accurately? It seems to be that by defining the Modulus as a vector so that it would have direction would work. I've never seen the general expression for the matricies used to compute the strain in the mesh. Is it possible to apply another vector quantity or is a scalar necessary. If only scaler functions can be used can the value be defined in relation to direction?
I just evaluated the CosmosWorks package a couple of weeks ago, and while not certain, am about 90% confident that I remember seeing that it let you enter anisotropic material properties, such as elastic modulus, in the x, y, and z direction.
I'm sorry I didn't explain myself. I meant hollow beams. the round beam would have to have radial properties. I suppose the rectangular beam could be constructed of several parts, each aligned appropiately. But that wouldn't be very accurate would it?
