U.S. Department of Transportation
Federal Highway Administration
1200 New Jersey Avenue, SE
Washington, DC 20590
202-366-4000
Federal Highway Administration Research and Technology
Coordinating, Developing, and Delivering Highway Transportation Innovations
This report is an archived publication and may contain dated technical, contact, and link information |
|
Publication Number: FHWA-HRT-04-095
Date: November 2004 |
Manual for LS-DYNA Soil Material Model 147PDF Version (1.96 MB)
PDF files can be viewed with the Acrobat® Reader® CHAPTER 3. EXAMPLES MANUALThis section presents some examples of simulations that were used during the verification phase of the development of the FHWA soil material model. The first example can be used to check the model accuracy and to familiarize the user with the material model. It consists of a single-element simulation of a triaxial compression experiment. Appendix B contains an example of the input for the triaxial compression single-element simulation. Figure 27 shows the results of the single-element simulation of a triaxial compression test at 3.4 MPa. Figure 27. Z-stress versus time for single-element 3.4-MPa triaxial compression stimulation. The peak strains in this example reach 80 percent. This shows that the material model will successfully analyze problems with large strains (deformations). A second example of the use of the FHWA soil material model is a simulation of a direct shear test. The goal of the tests was to determine the soil properties for the NCHRP Report 350 strong soil using large test specimens. The analysis is of direct shear test 4 (DS-4).(16) Contractors developed the model (see figure 28). The material model input for this simulation is shown in appendix B. Figure 29 shows the comparison between the test and the analysis of shear force versus deflection. The early time test data exhibit questionable trends and the analysis results show the expected trend (i.e., positive stiffness). Figure 30 shows the deformed shape of the cylinder at the end of the analysis. The analysis was terminated at approximately 47 millimeters (mm) of deflection because of the current failure criteria in LS-DYNA. An element fails (i.e., is eliminated from the simulation) when one of the gauss points reaches the failure criteria. For selective reduced integrated elements (8 gauss points), this causes premature failure. This premature failure does not let the internal forces go to zero in the failed elements. In turn, this leads to very large unbalanced forces at the nodes, causing unstable behavior (shooting nodes). figure 28 LS-DYNA model of direct shear test DS-4. The element formulation for this model is selective-reduced (S/R) elements. It is well known (see note 5 in the *SECTION input of the LS-DYNA manual) that poor aspect ratios (highly distorted elements) will cause shear locking. Elements along the shearing surface of the direct shear test simulation experience very large distortions, approximately equal to the element dimensions. Therefore, if severely distorted elements are not eliminated by erosion, the simulation will produce excessively stiff response (shear locking). An obvious way to overcome these problems is to use the standard constant stress (1 gauss point) element. However, time and funding did not allow the exploration of this option. A second option would be to refine the element mesh in the vicinity of the shearing surface to reduce the large deformations of the individual elements. A third option would be to use the Arbitrary Langrangian-Eulerian (ALE) formulation and the constant stress element. However, at this time, the FHWA material model is not available for use with ALE (although the capability of the FHWA soil material model to be used in conjunction with ALE was successfully tested during the early phase of this development effort with the model implemented as a user-defined material model). As shown in figure 30, there are surfaces of the soil, which at the time of the analysis, are in contact with the metal or air. Also, the interface between the two cylinder halves has become a noncontinuous surface (i.e., a slide surface). This behavior cannot be accurately modeled by the continuum mechanics material model; it must be modeled by slide surfaces. Figure 29. Shear stress versus deflection comparison for DS-4 Figure 30. Analysis results for DS-4 deformation As shown in figure 30, there are surfaces of the soil, which at the time of the analysis, are in contact with the metal or air. Also, the interface between the two cylinder halves has become a noncontinuous surface (i.e., a slide surface). This behavior cannot be accurately modeled by the continuum mechanics material model; it must be modeled by slide surfaces. Figure 31 Simple two-material shear model. The input parameters for the soil material were from the direct shear analysis. Figure 32 shows the deformed shape of the 1 gauss point element analysis at 1.75 milliseconds (ms), and figure 33 shows the deformed shape of the 8 gauss point element analysis at 1.75 ms. The analyses were stopped just before the 8 gauss point element becomes unstable because of the failure criteria error in the 8 gauss point element (outside the material subroutine). Figure 34 shows the x-y shear stress in element 115 (shown in the previous figures). Since the soil material has a low cohesion (shear strength at zero normal force) of 6.2 x 10-6 gigapascals (GPa) (6.2 kilopascals (KPa)), the x-y shear stress should not get very large. The 8 gauss point element shows an immediate increase in the x-y shear stress to more than 0.03 GPa (30 MPa); this type of behavior is known as "shear locking." It is caused by the formulation of the 8 gauss point element outside of the material routine. As mentioned previously, this inaccurate behavior is mentioned in the LS-DYNA manual. The 8 gauss point element should not be used for any analysis that involves shearing or failure. Figure 32. Deformed shape of 1 gauss point element analysis. Figure 33. Deformed shape of 8 gauss point element analysis. Figure 34. Comparison of x-y stress at element 115 for 1 gauss point and 8 gauss point elements. The material properties used in the simulations/examples described above were not determined from actual material property test data, but were found by trial and error. Obviously, actual material property data would probably lead to more confident results. |