LABORATORY MEASURED values of the low strain shear modulus (Gmax) can be used as a basis for design, according to new research by Fugro.
The geotechnical contractor compared laboratory values with insitu measurements from seismic cone tests and those estimated from cone penetrometer tests.
A structure applying cyclic loading was planned for a site underlain by two uniform free-draining sand layers, separated by an inhomogeneous silty sand unit with thin layers of clayey sand. This was considered critical for design, as it was potentially much less resistant to cyclic load ing .
Cyclic triaxial testing was carried out to investigate the build-up of pore water pressure in the unit under cyclic loading.
The silty sand's inhomogeneity meant results were scattered, so the quality of the laboratory tests had to be assessed, says Fugro senior engineer Nick Ramsey. The low strain stiffness of the soil was used as the reference parameter, as stiffness is very sensitive to sample disturbance, he explains.
Standard and seismic cone tests had been carried out during the site investigation.
Results were used to estimate the preinstallation low strain shear stiffness of the soils insitu, directly from results of the seismic cone tests and using correlations for the standard cone.
Ramsey says Gmaxobtained from the two tests agreed closely, indicating the correlation was valid for the site.
Resonant column and triaxial tests with bender element measurements were performed on samples consolidated to take account of the beneficial effect of the dead weight of the proposed structure.
As expected, these led to Gmax values significantly higher than those from the insitu tests.
However, Ramsey says that if the increase in vertical stress imposed by the structure is taken into account when estimating Gmax from standard cone penetration test results, there is actually good agreement between the insitu and laboratory tests.
Results indicated that laboratory sample preparation led to a reasonable estimate of the initial low strain shear stiffness of the soil beneath the proposed structure after construction.
Ramsey says this meant laboratory values of Gmax could be used for design, adding that quality of individual cyclic triaxial test results could be assessed with unrepresentative results removed from the design process.
l The Fugro advanced testing laboratory is offering a training package on the performance and use of advanced laboratory testing.Bender element, resonant column, stress-path, cyclic triaxial and cyclic simple shear tests are described and typical results presented.