sorry for delay in answering, but I was busy.
For my last exam I have to do a non linear analysis of a sail yacht rig with the Adina software.
Unlickily I don't have and don't know NX; the only other pre/post processor I have used (but for linear
analysis even if I think that now it can do also nonlinear) is Patran.
So I have to use the original Adina User Interface (or AUI). This is a request (maybe not very smart but that's what he wants)
of the professor.
One of the first problem I face is if it were possible to assign constrain boundary condition, called fixity in Adina,
referring not in the global reference system, but in a local or custom one. I mean, for example, if I have an inclined truss or beam
(in a plane problem), if it is possible to fix just the translation along the beam/truss axis.
Another big problem (but this is due to my still poor understanding of nonlinear analysis) is what exactly buckling means.
I am from Italy and maybe the english definition, that I have noticed is commonly universally used, doesn't help me a lot:
I have found two definitions: linear and non linear buckling.
From what we have studied it seems that the linear buckling is "based" on the well known euler theory of a column when it is subjected ONLY
to an axial compression load that reaches (at that point the equilibrium is indifferent) a critical load and then goes above triggering the instability of equilibrium:
if this were correct I have these big doubts: the theory tells that the used procedure is approximated because the curvature is taken simply equal the second derivative of the
displacement (in the transverse direction referring to the longitudinal axis of the column) neglecting the square root factor appearing at the denominator that the exact curvature expression has: this approximation leads to the known LINEAR differential equation, whose solutions allow to determine critical loads and modal shapes (depending on boundary conditions). The theory states that the approximation of the curvature and so the linear differential equations are valid just if the displacements and the curvature itself are very small.
Anyway, when I perform a linear buckling analysis and then check the results, I can find results where displacements are not very small. Indeed, if I don't change kinematic assumptions of displacements from small to large ones, Adina tells me it is not possible to perform the analysis.
Furthermore, as I stated above, I supposed that linear buckling were associated to slender structures subjected only to axial compression external force (euler theory), but I found the example no. 50 of Adina tutorial (it is easy to find it googling the words Adina tutorial) where the load applied is a transverse force, i.e. a force that already induces flexural moment...I have also tried to apply more than one transversal loads to a beam fixed to one of its extremity (that is the mast of the sailing yacht that is subjected to transversal forces due to the sails) and perform a linear buckling analysis; I got results but this sounds me a little surprising, as I thought that the solution of the eigenvalue problem, that gives the load factors to multiply the original load to reach the critical load, could be given only if only one load were present: so now my doubts are: is linear buckling what I supposed or is it a more general term that describes a slightly different (or more general) phenomena? Are the load factors obtained in a linear buckling analysis, as that I performed (i.e. with more than one load), be intended for every original loads present in the problem studied or they refer just to a particular one (the higher maybe....)??
I know, I started the thread asking help in non linear Adina and maybe I moved a little out, but unlickily the course I attended wasn't clear at all...
The original question, the one of my first post, was generated looking at the already cited Adina tutorial...as a matter of fact I have also doubts on how the nonlinear buckling analysis is performed in that exercise...anyway, at least for now, before proceeding in exposing in detail all my doubts, I think it's better to wait and see if someone can help me here...also because I think the post I have written is very long.
Thanks,
fed
