Browsing by Author "Huchzermeyer, Richard Lynn"

Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    Measuring mechanical properties using digital image correlation: extracting tensile and fracture properties from a single sample.
    (Stellenbosch : Stellenbosch University, 2017-12) Huchzermeyer, Richard Lynn; Becker, Thorsten Hermann ; Stellenbosch University. Faculty of Engineering. Dept. of Mechanical and Mechatronic Engineering.
    ENGLISH ABSTRACT: Multiple material properties are required to perform structural integrity assessments and reliability estimates on in-service equipment. Conventional material characterization testing approaches do not cater towards the testing of inservice equipment, and therefore ‘near-non-destructive testing’ approaches, in particular the small punch test (SPT), are preferred. The SPT, while capable of determining multiple material properties from a single small sample, does have limitations both in terms the complexity in analysing the resulting data, and the accuracy of the measured properties. These limitations may be addressed through full-field surface displacement analysis techniques facilitated by digital image correlation (DIC). A combined approach to extracting multiple material properties from in-plane (two-dimensional, 2D) surface displacements, measured on a single sample through DIC, has been developed assuming an isotropic linear elastic material. This approach utilizes the virtual fields method (VFM) to obtain Young’s modulus (E) and Poisson’s ratio (v). These tensile stiffness properties (E and v) are in turn input to a non-linear least squares field fitting approach (FF), which is used to obtain the critical stress intensity factor (𝐾𝑓𝑓) associated with a crack or notch in a material. The VFM and FF are applied to two compressively loaded disk shaped sample geometries (containing central notches) as well as an elongated half compact tension sample geometry (W = 25 mm) manufactured from 6 mm thick polymethyl methacrylate (PMMA). The experimental methodology to obtain suitable two-dimensional surface displacement measurements though DIC is described. Furthermore, the implementation of the VFM and the FF is developed on a sample specific basis. Through a comparison to properties determined using standardized ASTM testing, a relative error for the VFM of -1.5 % to 4.6 % in E and 12.9 % to 40.2 % in v is obtained. A concomitant relative error in the FF is determined to be 33 % to 38 % for 𝐾𝑓𝑓. Experimental errors, in particular out of plane rotation, are identified and the limitations of the assumptions made in applying the techniques are examined. Furthermore, the manner in which the error in E and v obtained through the VFM contributes to the associated error in 𝐾𝑓𝑓 identified through the FF is examined. It is found that the FF approach is less sensitive to error in v provided that the error in E is small. The nominally successful combined application of the VFM and the FF to 2D displacement fields measured with DIC on a single sample (assuming a linear elastic isotropic material), strongly motivates for the further development of this approach. The approach could be extended to accommodate out-of-plane deformations measured through DIC and could be developed to extract properties from ductile metallic materials. It is envisioned that this will be addressed in future work, which could lead to the methodology being applied directly to the SPT. A significant first step towards this is presented in this work, which demonstrates the first successful combined application of the VFM and the FF for extracting stiffness and fracture properties from full-field in-plane 2D displacements measured through DIC on a single sample.