Advancement of Optical Methods in Experimental Mechanics, Volume 3

distance is 220 mm, and the angle between the two cameras is 22 . The progressive scan CCD chip in the camera is Sony ICX625AL, with 2/300 or 11.016 mm diagonal length, 5.05 million (2,448 by 2,048) square pixels. Two 25 mm lenses (NT63-780 from Edmund Optics) are attached to the cameras. The sample is illuminated by two compact fluorescent light lamps. The Vic-3D software from Correlated Solutions, Inc. was used to analyze the collected images and to calculate the strains on the surface of the SiCf-SiCm sample tube. The correlation algorithm used is based on the normalized square differences criterion which has better tolerance to changes in lighting. The analysis was done using a subset of 29 pixels, and a step of 7 pixels. The image acquisition is made using the Vic-Snap 8 software using a frame rate of 2 per second which is the maximum allowed with the current hardware. 46.3 Results and Discussions A typical hoop and axial strain mapping of the outer surface of the sample obtained by correlating the images of the SiCf-SiCmsample tube is shown in Fig. 46.3, for an internal pressure of 96.5 MPa (14,000 psi). The DIC correlation software reproduces the irregularity of the surface of the sample with great detail. The values of the Lagrange strains on the surface of the sample are shown in different colors, where each color corresponds to a strain value according to the scale bar on the right of Fig. 46.3. The hoop-strain plot shows that the deformation of the sample is not uniform along the surface; the calculated strains have different values depending on the location on the surface of the sample. The non-uniformity is likely due to the texture of the filament wound composite tube which can affect outer surface strain distribution and pressure distribution at inner surface. The foil gage bonded nearby may have also affected the DIC reading. It is important to mention that the typical error for strains when calculated using DIC is about 200 m-strains [11, 12], this must be taken into account when interpreting the values of strains in the results. The axial strains from DIC (shown in Fig. 46.3.) shows that the sample is under very low compression loading, as expected from the rubber bladder approach. The average axial strain is 88m-strains in a range of 400 to25m-strains and strain resolution of the DIC method is ~200 m-strains. The slight negative axial strains at the internal pressure of 96.5 MPa (14,000 psi) is likely due to the Poisson effect of the sample in combination with a small tensile force induced by the aluminum adapters (which tend to shrink) due to its smaller ID. It is believed that the aluminum tube exerted a small tensile force on the SiC sample at the stepped lap joint transmitted through friction. The hoop stress in the SiCf-SiCmsample tube is calculated at the outer diameter using thick-walled cylinder theory for the case of internal pressure only [13] using the equation, sh ¼P 2r2 i r2 o r2 i ð 46:1Þ Fig. 46.3 DIC plotting of hoop and axial Lagrange strains at an internal pressure of 96.5 MPa (14,000 psi) 390 L.H. Alva et al.

RkJQdWJsaXNoZXIy MTMzNzEzMQ==