• Registered: 2011-11-09
  • Posts: 3

Topic: Step Dependent Material Properties.

Dear All,

I am currently a graduate student at University of Pittsburgh. I have a problem when using ATENA to modelling a bridge. What I concern is, could we define a time dependent material in Atena? Or simpler, step dependent material? I hope to hear from someone have this kind of experience. Thanks a lot!

Chunlin Pan

Quote Post 1

  • dpryl
  • CC software developer
  • Offline
  • Registered: 2007-07-13
  • Posts: 1,107

Re: Step Dependent Material Properties.

Dear Chunlin Pan, yes, this is possible. Please about the material type "3D Variable Non Linear Cementitious" (Engineering 2D, 3D) or "NonLinear Cementitious2Variable" (ATENA Kernel/Theory) in the ATENA Manual(s). Alternatively, you can also search for "Construction process", which is an option to include a part of the structure in a later stage or remove a part during the analysis.

dpryl's Website

Quote Post 2

  • Registered: 2011-11-09
  • Posts: 3

Re: Step Dependent Material Properties.

Concerning this application, could you please teach me how to do a simple simulation? For example, I have a T shaped beam, the top slab at construction Process 1 is defined with zero Young's Modulus, but a finite Young's modulus at construction Process 2. The web is defined with constant Young's modulus (e.g. the same as top slab). The whole simulation consists of two construction processes. Thanks a lot for your time!

Quote Post 3

  • dpryl
  • CC software developer
  • Offline
  • Registered: 2007-07-13
  • Posts: 1,107

Re: Step Dependent Material Properties.

Dear Chunlin Pan, please see the ATENA Science Example Manual, section 2.2 Tutorial for Construction Process for an example of construction Process in GiD. For ATENA 2D, please see the manual "Tutorial for Construction Process Modelling in ATENA 2D".  For ATENA 3D, there is no example, but defining the constructio cases and their use in load steps is pretty straight forward.
There is no example demonstarting the use of the NLCem2Variable material, but you simply define the step history for the elastic modulus (and eventually other material parameters). Of course, don't set the elastic modulus to zero, as this would introduce severe numerical problems (e.g., division by zero). Use something like 1/100-1/1000 of the normal elastic modulus (i.e., of the one used later).

dpryl's Website

Quote Post 4