Chapter 11 Eliminating Air Refraction Issues in DIC by Conducting Experiments in Vacuum P. L. Reu and E. M. C. Jones Abstract A major and often unrecognized error source in digital image correlation (DIC) is the influence of the intervening air between the cameras and sample. Minute differences in air temperature, composition, or both can cause index of refraction changes that act as a lens and cause distortions in the DIC displacement and strain results (Jones and Reu, Exp Mech, 2017). There are limited options to correct this problem as it is both spatial and temporal in nature. One method is to use X-rays for imaging that are not affected by air refraction, but this requires costly equipment. A second method uses a vacuum chamber to minimize the intervening air to remove the distortions, but unfortunately this requires inconvenient setups. Keywords Digital image correlation (DIC) · Full-field · Optical methods · Uncertainty quantification (UQ) 11.1 Introduction Stereo-DIC measurements are subject to many error sources that can create either bias errors or variance errors in the results. There is some discussion as to what the largest remaining error sources are for a typical DIC setup. Aliasing, heat waves and camera motion are all possibilities. Of course, this will be very setup dependent. This study looks at index of refraction changes caused by heat waves that are an often-ignored source of temporal and spatial errors [1]. To help elucidate their effects on DIC, an experiment was setup inside a vacuum chamber with a stationary plate to determine a possible lower bound on the noise floor. 11.2 Experimental Setup To determine the influence of the air on the DIC results we used a stereo-rig observing a stationary speckle pattern with a field-of-view of approximately 100-mm (See Table 11.1 for setup details and Fig. 11.1 for a photo of the setup inside the vacuum chamber). The entire system was placed in a large vacuum chamber that was pumped down to pressures between 3 Torr and atmosphere. The minimum pressure was dictated by the chamber size and leak rate. After shutting off the pumps to minimize vibrations, we acquired images at 5 fps for 3 min at a variety of pressures. Images were also acquired at 140 Hz to look at the frequency response of the system. To better illustrate the influence of heat waves, a hotplate was placed under the speckle pattern at a temperature of 60 ◦C, which simulates the temperature of LED lights. Note that this position of the heat source is a worst-case scenario. Other tests positioned the hotplate either behind the cameras to better simulate a typical experimental lighting setup, or with no hot plate for a best-case scenario. P. L. Reu ( ) · E. Jones Sandia National Laboratories, Albuquerque, NM, USA e-mail: plreu@sandia.gov © The Society for Experimental Mechanics, Inc. 2019 L. Lamberti et al. (eds.), Advancement of Optical Methods & Digital Image Correlation in Experimental Mechanics, Volume 3, Conference Proceedings of the Society for Experimental Mechanics Series, https://doi.org/10.1007/978-3-319-97481-1_11 85
RkJQdWJsaXNoZXIy MTMzNzEzMQ==