1 (edited by engpawlo 2013-02-04 05:18:52)

Topic: Confinement, cyclic load, Displacemnt of Nodes, ATENA 5

Hi smile,
First of all I'd like to thank the administrator for the effort being done in this forum; it was very helpful for me to get answers for some questions by browsing this forum.
To give you some heads up, I'm using ATENA 3-D (Static) version 4.3.1.7242 to study the shear capacity of the joint (connection) of the beam-column joints under seismic loads (quasi-static, cyclic-reversed) in which the concrete softening characteristics (descending part in concrete stress-strain curve) and the amount of joint confinement (stirrups) play a significant role in the results. I did experimental testing and need to verify these results by FEM results then start a parametric study. I have some questions that I'd like to discuss as follows:

1-In one of your previous responses (replies to mhaider, PHD_Student, dpenava, … ) you indicated that the confinement effect is normally considered in the NLCem2 material (which I’m using in my model). It is well established, as in the concrete model by Kent & Park and other models, that the confinement provided by means of stirrups increases the concrete strength and ductility (ultimate strain at failure), see references below. I went through ATENA Theory manual (Sections 2.2.4 Plasticity Model for Concrete Crushing, Section 2.2.5 Combination of Plasticity and Fracture model) however, I found that it does not take in to account the reinforcement ratio of transverse reinforcement (stirrups) as a parameter that influences the concrete strength and the ultimate strain (as in Kent & Park concrete model). In this case, do I have to account for the increased ductility of concrete by increasing the value of Wd in the concrete compressive characteristics? I already modeled my specimen using the actual/experimental concrete cylinder strength (with no increase) with Wd= -0.004 m and got results that is close to my experimental tests. Please help and advise.

2- In your response to the member alexandru_fabian (posted on 2011-01-11) to the model cyclic load you advised him to use to 2 steel loading plates, 2 load cases (positive and negative), 2 monitor points; I followed your advice to model my cyclic loading however it took me 5-6 days for the model to finish the solution. Then Instead of the above, I build a stiff steel ring around the point of displacement application which helped me to use only 1 load case and 1 monitor point which reduced the solving time almost by 50% with no change in the results. 2-3 days of solving time is still time consuming for me although that I’m using a powerful computer (Intel core i7 with 8 processors 870@2.93GHz each, 8 Gigs Ram, Windows 7-64 bit, NVIDIA GeForce 310) but I noticed that the program only uses 4 processors of the 8 for the analysis. I understand that my loading history can be intensive (1700 load steps, each is 1 mm) but my question here, is there a way to let the program use the 8 processors together to speed up the analysis (I reviewed the trouble shooting manual, Section 2.1.7).

3- To view the results in the post-processor mode (for example, values of strains and stresses on reinforcement bars) the numbers viewed on the screen are very small and difficult to read and zooming in does not help as the numbers stay small. Please advise me how to increase the font of text of these values. Please help.

4- In the post-processor mode, I can get the nodal coordinates change when I go Files, print text, choose the step #, Nodes, Displacements. Because it is very important for me to relate the displacement of the node to its original position (coordinates), my question here is how I know the original coordinates of each node before applying the load. As far as I understand that the nodes and their coordinates are generated automatically by the program based on the FE Meshing characteristics. Please help.

5- Is there any news about ATENA 5, It was expected to be released at the end of January, 2013.

My apologies for this lengthy post and thank you in advance for your help.

References related to question#1:

==Kent, D. C, and Park, R., “Flexural Members with Confined Concrete,” Journal of the Structural Division, ASCE, V. 97, No. 97, No. ST7, Jul. 1971, pp. 1969-1990.

==Scott, B. S.; Park, R.; and Priestley, M. J. N, “Stress Strain Behavior of Concrete Confined by Overlapping Hoops at Low and High Strain Rates,” ACI Structural Journal, V. 79, No. 1, Jan.-Feb. 1982, pp. 13-27

Re: Confinement, cyclic load, Displacemnt of Nodes, ATENA 5

Hello engpawlo,
1. ad confinement - when looking at the desciption of the plastic material model, please note the influence of stresses in other directions. Under hydrostatic pressure (= identical compression in all 3 directions), the material strength is virtually infinite. Therefore, the transverse reinforcement needs NOT to be considered explicitly - it limits the development of tensile stresses/cracks in the perpendicular direction, or even helps to buil compressive stresses in that direction (due to Poisson's ratio).

2. we are still working on a multithreaded solver.

3. I am afraid there is currently no way to change or size the font used for the node/element numbers etc. in ATENA Engineering 3D. To read stresses and strains, the color scale is typically the fastest.

4. Reference Nodal Coordinates is what you need to output.

5. I have no news about ATENA v5. Please be patient and check our web every few weeks to get any new information.