EXPERIMENTAL TECHNIQUES The flow-stress of the material can be measured from tensile experiments on smooth round bars. To account for the influence of hydrostatic stress at the neck of the specimen a Bridgman [2], Siebel [3], and Bakker [4] correction procedure have been followed. For the experimental determination of the material flow stress as a function of strain and strain rate, two separate measurement systems have been applied. The first system was applied for strain rates up to 0.1 s-1, the second system for strain rates beyond 10 s-1. Both systems will be shortly discussed here as well as the techniques that have been applied for image processing. Low Strain Rate Testing For tensile tests at low strain rates a servo-mechanical machine has been used. Load measurement was provided by a strain gauge loadcell. Extension of the specimen has been measured by a strain gauge extensometer up to the strain at which localisation starts. For strains beyond the onset of necking digital camera images have been used to derive the true strain at the centre of the neck. The true strain ε has been taken from: ε= F HGI KJ ln A A0 (1) As the rolled plate material shows a slight anisotropy with respect to plastic flow in the plate thickness direction, initially round tensile specimen develop to an oval shape. Therefore two digital cameras were used to measure both diameters of the oval shape, which area has been estimated using an ellipse relation for the cross sectional area at the centre of the neck. Schematically, the camera set-up has been shown in Figure 1. Figure 1: Top view of the camera set-up for the digital imaging system From the camera images the diameter and radius of curvature of the neck has been derived using imageprocessing techniques. An anisotropy factor has been derived from the strains from both cameras defined as the quotient of two strain values. This factor appears to be independent of the strain level. High Strain Rate Testing The high strain rate testing equipment mainly consists of a hydraulic tensile machine, which can achieve speeds up to 10 m/s. During the experiment a rather constant speed has been maintained. A high-speed camera was used to monitor the deformation in the neck of a specimen. Two markers on the front side of the specimen have been used to measure the elongation during uniform deformation. Images from this conventional rotating drum camera have been developed and scanned with a high-resolution film scanner and subsequently processed with a computer. Image processing techniques will be treated in the next subsection. Load measurement using a piezo-electric loadcell resulted in a vibrating load signal caused by resonance phenomena. Investigations showed that the stresses in the specimen are hardly influenced by these resonances and may be assumed to be equally distributed and without resonance. Smoothing of the
RkJQdWJsaXNoZXIy MjM0NDE=