STRAIN EVALUATION IN THE NECK OF A TENSILE TEST SPECIMEN BY ESPI STRAIN RATE MEASUREMENT Bruno Guelorget, Manuel François and Guillaume Montay Université de technologie de Troyes, ICD-CNRS FRE 2848, LASMIS, 12 rue Marie Curie, B.P. 2060, 10010 Troyes Cedex, France ABSTRACT In-plane Electronic Speckle Pattern Interferometry has been used during tensile test of semi-hard copper sheets for measuring the strain rate. Strain was calculated by integrating the strain rate. The width of the strain localization zone has been evaluated by fitting Lorentz curves on the strain rate peaks. Finally, the evolution during the tensile test of the strain localization band width was determined. Introduction Many mechanical properties of materials, such as Young's modulus, yield stress, tensile stress or strain hardening coefficient can be determined by tensile test. Many metals exhibit tensile stress-strain curves whose initial linear portion (elastic region) passes gradually to the elasto-plastic region. In a first stage of this region, strains are homogeneous along the specimen. This stage ends when the diffuse neck appears. Classically, it is considered [1] that the diffuse neck occurs when the force reaches its maximum. Then strains begin to be heterogeneous up to the localized neck. In case of a sheet in uniaxial tensile test, this localized neck is the shear band whose width is of the same order of magnitude as the specimen thickness. As they give a strain map on relatively large areas, optical methods are often used to monitor mechanical tests. Wattrisse and al. [2] have measured strains during a tensile test with a speckle images correlation technique. Kajberg and Lindkvist [3] have determined the post-necking behaviour law of a material through an inverse method, using finite element modeling and strain measurement by image correlation. Cordero et al. [4] have carried out a comprehensive whole-field analysis of a uniaxial tensile test, using 2D-moiré interferometry, extracting strains and stress fields, ratio of anisotropy and maps of shear bands. Thanks to its capability to perform accurate measurements of displacements, Electronic Speckle Pattern Interferometry (ESPI) is used during the tensile test experiments performed in the present paper. Toyooka and Gong [5] have used a setup with one sensitivity vector for recording fringes pattern and a strain gauge for correlating fringes with measured strains. In their next paper [6], they presented a setup with two in-plane sensitivity vectors and they used four strain gauges. A completely different approach has been followed by Panin [7] based on a synergetic methodology of physical mesomechanics, establishing relations between the physics of dislocation-induced deformation, continuum mechanics of solids and fracture mechanics. Several papers compare the fringes pattern obtained with ESPI and Panin's results [8-11]. Shabadi et al. [12] obtained similar results, studying Portevin-Le Châtelier bands, with an ESPI technique. They obtained a large set of results, such as influences of strain, strain rate, thickness of the specimen or ageing on band angle, band width or band velocity. Vial-Edwards et al. [13] have been interested in determining the true stress-true strain curve from a tensile test and ESPI experiment, in order to have a look at the localization onset. The aims of the present paper is to evaluate the strain in the neck of a tensile specimen by an ESPI strain rate measurement and to determine the strain localization band width evolution during the experiment. Experimental The specimen (Fig. 1) was cut in a 0.8 mm thick semi-hard copper sheet, in the rolling direction. The tensile test was performed at a constant crosshead speed equal to 0.5 mm/min. A square grid of 3 mm separation was drawn on the specimen. The experimental setup is presented on Fig. 2. The wavelength λ of the He-Ne laser was equal to 632.8 nm. The beam was expanded, collimated, separated into two beams by a beam-splitter and directed towards the specimen by mirrors. The present setup belongs to an in-plane sensitive configuration, with a sensitivity vector parallel to the tensile direction (Fig. 2). In the present experiment, the incidence angle is about 53.5° and the sensitivity is 0.394 µm/fringe. Total strain can be obtained by measuring the length change of the initial square grid: length initial length measured ln = t ε . (1) The relative displacement between two points A and B is: