Optimized optical setup for DIC in rock mechanics

¹ Laboratoire de Mécanique des Solides, École Polytechnique ParisTech, 91128 Palaiseau, France
² Unité de Recherche Navier, École des Ponts ParisTech, 77544 Marne-la-Vallée, France

^a Email : michel.bornert@enpc.fr

Abstract

In order to investigate the micromechanical mechanisms of argillaceous rocks under combined mechanical load and varying moisture, and observe very low strain rates of about 10^-10s^-1, i.e. about 10^-4 per week, a specific optical setup for Digital Image Correlation (DIC) has been developed at the Solids Mechanics Laboratory, Ecole Polytechnique. This paper mainly presents two procedures to optimize the experimental setup and enhance the full-field strain measurement accuracy. The first one allows us to minimize the systematic DIC errors by choosing an optimal optical setup, including appropriate lens aperture and light source with small wavelength, in combination with a specific grey level interpolation method (bilinear, bicubic, biquintic) in DIC algorithms. The second procedure aims at reducing the errors induced by overall out-of-plane motions of the sample. This procedure makes use of a specifically designed optical microscope, which tracks the few tens of micrometers out-of-plane motions of the sample all along the several months lasting test, to compensate the magnification fluctuations which disturb the macroscopic measurements. The current optimized optical setup gives a better than 10^-5 accuracy at macro scale (3 mm gage length) and 10^-4 accuracy at microscale (about 100 µm gage length).