A protocol for non-contact measurement technique applied to cyclic triaxial test

Date of Publication


Document Type


Degree Name

Doctor of Philosophy in Civil Engineering


Gokongwei College of Engineering


Civil Engineering

Thesis Adviser

Dungca, Jonathan R.

Defense Panel Chair

Adajar, Mary Ann Q.

Defense Panel Member

Oreta, Andres Winston C.
Garciano, Lessandro O.
Lejano, Bernardo A.
Zarco, Mark Albert H.
Dungca, Jonathan R., dean


Deformation in a conventional cyclic triaxial test is normally measured by using a point-wise measuring device such as a linear variable displacement transducer. This type of measuring device has no capacity to present the behavior of the soil under loading because it measures the average deformation of the soil. In this study, a protocol for non-contact measurement was developed and was applied to a cyclic triaxial test. The protocol was developed to perform full-field monitoring of the behavior of the soil under cyclic loading. The protocol was developed using a mirrorless camera. Lucas and Kanade pyramidal optical flow algorithm (1981) was implemented to track the movement of the particles in the 2- Dimensional space. Calibration was initially performed to eliminate distortion and to ensure that measurements can be properly made. Lighting condition was improved since it greatly affected the quality of images taken. The capacity of the mirrorless camera was investigated by testing a cylindrical rubber dummy. Furthermore, curvature correction was applied to correct the non-linear profile of the sample.
The protocol was also used to estimate the strain-parameter triggering liquefaction such as threshold shear strain (γtv) and cyclic shear strain that would trigger liquefaction (γcl) to assess the triggering of liquefaction. A cyclic strain approach was implemented in an undrained cyclic triaxial test. Based on the results, the estimated values for γtv are affected by the strain amplitude. On the other hand, γcl are affected by relative density, strain amplitude and confining pressure. A range of values was estimated for the parameters this was due to the existence of non-uniform movement. The estimation resulted to experimental bands for porewater pressure ratio and modulus reduction curves considering the local movements of the soil.
The estimated cyclic shear strain (γc) and threshold shear strain (γtv) were predicted using the Modified Hardin-Drnevich model proposed by Darendeli (2001) and Vardenega and Bolton (2011). The model was not able to predict the parameters due to a limited range of curvature coefficient. Equations for curvature coefficients were formulated using Response surface method. D-optimal was used to determine the best data set for the formulation. The equations were formulated as a function of strain amplitude and confining pressure. A good fit was observed based on the formulated equations. The equations are only applicable to the type of soil tested, relative density, strain amplitude and confining pressure implemented.
The protocol for non-contact measurement technique applied to cyclic triaxial test developed was able to show the boundary effect due to the top cap and pedestal. The boundary effect caused the non-uniform deformation of the soil. The protocol developed can monitor the development of the critical zones or zones where large displacement or shear strain are concentrated. Early detection of the critical zones and the shear strain localization can be made using the non-contact measurement technique. Furthermore, analyses of the critical zones and other areas of the sample visible to the mirrorless camera are possible since data from the full-field monitoring can be extracted.

Abstract Format




Accession Number


Shelf Location

Archives, The Learning Commons, 12F Henry Sy Sr. Hall


Soils -- Testing.; Shear strength of soils -- Testing.

This document is currently not available here.