In analysis, applying complex surface load might be involved, such as tangential surface load or load at a certain angle. Ansys can only apply surface load perpendicular to the surface. This can cause the issue for controlling angles. On the other hand, when the magnitude of the surface load varies with location, it is difficult to apply such surface load.

Now, consider a flat plate composed of two materials, steel on the left side and iron on the right, then they are welded together in the middle with copper. The dimensions of the plate are 1×1×0.2m, with the cross-sectional area as shown in the following figure,

This post conducts a reliability analysis on a wrench in practical use via Ansys. Reliability analysis is a statistical method used to study the ability of a system or component to perform its required functions under stated conditions for a specified period of time. It’s fundamentally concerned with the assessment and prediction of product lifespan and identifying the likelihood of failures. The goal of reliability analysis is to identify areas of improvement in the design, manufacturing, and operational processes to enhance the overall dependability of products or systems.

During analysis, we often encounters complex boundary conditions that vary with time, coordinates, temperature, and other factors. Therefore, it is not feasible to apply a general loading method to impose loads on structures, such as loads that change with spatial location or convective boundaries in thermal convection analysis that vary with coordinate position.

In previous post, the work with Ansys on calculating reinforced concrete beams was overly simplistic. It involved just two rebars modeled as elastic bodies and a rather straightforward set-up for the concrete parameters, which do not require the adjustment of solver convergence criteria to achieve convergence. Hence, I’ve decided to develop a more complex model, based on a model in a book on advanced finite element analysis methods and examples, where I’ve also corrected some minor issues presented.

I was interested in performing finite element analysis of uniformly loaded concrete beams but did not start due to the complexity of choosing appropriate material and load parameters. Recently, there exists an example that seemed perfect for practice. The subject of analysis is a concrete beam with a length of 3m and cross-sectional dimensions of 0.1×0.2m. To simplify the model, the concrete protective layer on the bottom and sides is not considered. The beam is reinforced at the bottom with two 20mm diameter steel rebars, treated as elastic and without accounting for elastoplastic behavior.

When finite element models are large, analysis on a standard computer can become challenging. Typically, the model is divided into coarser meshes in such cases, but this comes at the cost of losing analysis precision. When accurate local results are needed for a model’s specific area, the options are either to refine the mesh locally or to use submodel analysis techniques. The former has a significant drawback, mainly because even coarsely divided large models require considerable computational effort. Refining the mesh locally undoubtedly increases the computational load.

After reviewing numerous posts on Abaqus subroutines, I found the information online to be disorganized. Here, I put a post to summarize some fundamental approaches for future reference.

In Abaqus, the default UMAT subroutines are written in Fortran, with a .for file extension, and are compiled and linked into the simulation during running model. However, if there exists a need to compile the subroutine ahead of time or to protect the source code, the .for file can be compiled into an .obj binary file.

While Abaqus is a great program, its modeling capabilities, like those of many other finite element analysis (FEA) software, can be somewhat limited. Therefore, to enhance the applicability of Abaqus, it is often necessary to utilize external CAD programs for modeling. Once the modeling is complete, the model can be imported into Abaqus for analysis.