|This report is an archived publication and may contain dated technical, contact, and link information|
Publication Number: FHWA-HRT-04-094
Date: November 2004
Evaluation of LS-DYNA Soil Material Model 147
PDF Version (2.87 MB)
PDF files can be viewed with the Acrobat® Reader®
CHAPTER 3. BASELINE MODEL: DIRECT SHEAR TEST SIMULATION
In this chapter, a baseline model of the direct shear test is established in order to investigate the effects of the various parameters associated with the soil model.
A finite element model using 10,600 elements was developed based on the large-scale direct shear testing device (see figure 4). This model included solid rigid elements to model the steel, since no deformation had been seen during physical testing, and solid elements to implement the Federal Highway Administration (FHWA) soil model, material type 147.
An overburden pressure of 18.5 kPa was applied and dynamic relaxation was implemented before the soil was sheared in the finite element model. A prescribed motion condition of 1 millimeter per millisecond (mm/ms) was applied to the lower containing cylinder in the device, similar to the quasi-static testing motion condition. Loads and displacements were measured so that simulation results could be directly compared to the physical test results.
Initial parameters for material type 147 were selected from three LS-DYNA models provided by the developer. These models included two triaxial compression tests (cylin.k and txc3-4pr0c.k) and a hydrostatic tension test (hydten1.k). The initial parameters implemented in the developer's models and the comparison to the baseline model used in the parameter study are shown in table 1. The material input parameters in LS-DYNA keyword format are shown in table 2.
E-mail correspondence from the developer later recommended values for the shear modulus, G, and the bulk modulus, K, as 1.3 megapascals (MPa) and 3.25 MPa, respectively. Because of the lateness of obtaining these values, these changes were not reflected in the material parameter study of chapter 4. However, chapter 5 of this report, "Developer -Recommended Parameters," investigates these recommended values.
Table 1 . Developer's and baseline material parameters.
Table 2 . LS-DYNA format: Baseline values for parameter study, model R3.
Deformation of the baseline model results (without the frame structure) is shown in figure 5, the force to perform the direct shear simulation is shown in figure 6, and the internal energy absorbed by the soil during the simulation is shown in figure 7. Unfortunately, it appears that the soil model is only useable in this configuration for about 25 mm of deformation (as shown in figure 6). The force builds up nicely and peaks at 718 kilonewtons (kN) at 18.2 mm of displacement. The soil begins to soften, as it should, until it reaches the valley at 24.6 mm of displacement, with a corresponding force of 385 kN. After that time, the force begins to rise, which never occurs in physical tests. The force should either maintain a relative constant value or begin to drop, as previously shown in figures 2 and 3.
One reason for the rise in forces after 25 mm of displacement is because of the high damage level specification. However, if lower values of damage are used, the model becomes unstable and the code terminates. This will be discussed further in chapter 5. For the parameter study in this report (chapter 4), the main results of the comparison will be the peak force and the valley force, and their corresponding displacements and internal energy absorbed by the soil.
Figure 7. Direct shear soil internal energy: Baseline model.
Topics: research, infrastructure, materials, geotechnical
Keywords: research, safety, soil material model, roadside safety simulation, LS-DYNA
TRT Terms: Soil mechanics–Mathematical models–Evalution, Shear strength of soils–Testing–Computer simulation, Foundation soils, Roadside structures, Soil structure interaction, Finite element method, Impact loads, Computer models