Hazard and Risk Consulting for Critical Facilities

Data Archive

Monterey Sand Bi-Directional Simple Shear Testing
About the database
This testing database was developed by performing bi-directional simple shear testing on sample of dense sand. The objective of this testing program was to develop a 3D dataset useful for developing and validating 3D constitutive sand models intended for use in non-linear time-domain finite element modeling. This program was performed at UC Berkeley as part of my dissertation research.
Understanding the data
The easiest way to understand the resulting database and initial conclusions is to read the 3 papers below, which were intended to form a series of primers.
Summary of key findings
The key findings from these papers includes:

  • There is a range of shear stress values where the balance between driving stress and pore pressure development lead to high strain potential.
  • The assumption that the excess pore pressure within a sample must approach the total vertical stress (resulting in ru ≈1.0) in order for “liquefaction” to be achieve must be reassessed based on results showing that a large number of tests exhibited large strains with relatively low maximum pore pressures (ru,max=0.7or less). This logically follows once it is recognized that (a) there is a relationship between minimum shear stress and maximum excess pore pressure and (b) few multi-directional tests experience points of zero shear stress.
  • The addition of shear stress in a second horizontal direction tends to result in quicker attainment of ru,lim. The biggest change in liquefaction resistance occurs when the load is first applied in the second direction. Further increase in this perpendicular load continues to lower triggering resistance, but with declining effect.
  • After the limiting pore pressure (noted as ru,lim in the papers) has been achieved in a given test, the pore pressures measured at any point during a cycle show a (roughly) linear relationship with the shear stress at that point. This relationship can be estimated from the failure envelope, though some exceptions do apply.
  • The presence of an initial static driving shear stress, as would be found under a sloping ground site or a structure, has a strong influence on shear strain development. Moderate shear strains tend to have the highest strain potential. At sites with a very high initial shear stress, loading may not be large enough to allow for shear strain reversal. When this occurs softening is limited. Unlike in uni-directional tests, however, even a very small load in the uphill direction can have a very large impact on softening.
Support for this research was provided by a National Science Foundation (NSF) Graduate Research Fellowship, an EERI/FEMA NEHRP Graduate Research Fellowship, Pacific Earthquake Engineering Research Center project 2051999, and NSF CAREER award CMS-9623979.

Colleagues who contributed to this research were my dissertations advisors Profs. Ray Seed and Juan Pestana, Prof. Mike Riemer, the Berkeley Geotechnical Laboratory director and testing expert, Dr. Jiaer Wu, who worked on a collaborative project looking at settlement in dense sands, Prof. Giovanna Biscontin, who assisted with rehabilitation of the testing device and the control system, and Prof. Ross Boulanger, who constructed the testing device for his own doctoral work.
Stacks Image 6009
Stress conditions in the bi-directional testing program (also showing the stress conditions from two other studies)
Stacks Image 6038
Example of one cycle of a “figure 8” test. The top two figures and bottom right figure are in plan view.
Stacks Image 6051
Graphic representation of conditions under which large strains are possible in dense sands (see WCEE paper on strains).
New Hampshire Office
35 Alehson Street
Rye, NH 03870

California Office
622 Taylor Avenue
Alameda, CA 94501