Image-based stress and strain measurement of wood in the split-Hopkinson pressure bar

Abstract

The properties of wood must be considered when designing mechanical pulping machinery. The composition of wood within the annual ring is important. This paper proposes a novel image-based method to measure stress and planar strain distribution in soft, heterogeneous materials. The main advantage of this method in comparison to traditional methods that are based on strain gauges is that it captures local strain gradients and not only average strains. Wood samples were subjected to compression at strain rates of 1000–2500 s−1 in an encapsulated split-Hopkinson device. High-speed photography captured images at 50 000–100 000 Hz and different magnifications to achieve spatial resolutions of 2.9 to 9.7 µm pixels−1. The image-based analysis utilized an image correlation technique with a method that was developed for particle image velocimetry. The image analysis gave local strain distribution and average stress as a function of time. Two stress approximations, using the material properties of the split-Hopkinson bars and the displacement of the transmitter bar/sample interface, are presented. Strain gauges on the bars of the split-Hopkinson device give the reference average stress and strain. The most accurate image-based stress approximation differed from the strain gauge result by 5%.