Design and Development of a New Gripping System and Direct Stress Measurement Method for High Strain Rate Materials Testing, 18-R8473
Kathryn A. Dannemann
Nikki L. Scott
Alexander J. Carpenter
Inclusive Dates: 05/22/14 – 09/22/14
Background — A Hopkinson bar test system has been used extensively in our laboratory for determining the dynamic response of various materials (e.g., metals, ceramics, welds, glass, rocks, composites). This is the most widely used method for evaluating the high strain rate behavior of materials at strain rates ranging from 100 s-1 to 5000 s-1. A schematic of the SwRI direct-tension Hopkinson bar is shown in Figure 1. For dynamic tension, Hopkinson bar testing of threaded cylindrical specimens is an established method. However, the threaded specimen design is not applicable for tensile testing of thin sheet (≤ 3 mm) materials. High strain rate tension testing of sheet materials presents inherent challenges: i) gripping of the specimen, and ii) a weak transmitted signal. The latter can result owing to the small cross-sectional area of the sheet specimen, especially for low sound speed materials. This poses challenges for extracting the stress signal to obtain the desired stress-strain curves. Further signal loss can result due to an impedance mismatch between the grip hardware and the specimen. A functional grip design with proper impedance is critical for high strain rate tensile testing of sheet materials and for obtaining accurate test data.
Approach — The objective of this project was to design and implement new grips for the SwRI direct tension Hopkinson bar that can accommodate sheet specimens for materials with different strengths and thicknesses (≤ 3 mm thick). Grip designs with a mechanical means (i.e., pins, bolts) of securing the specimen were utilized to prevent slippage in the grips. High strain rate experiments were designed and conducted to demonstrate the effectiveness of the new grip and specimen designs, and to confirm the reliability and consistency of the measurements. High-speed cameras were used to photograph the progression of damage during tensile loading. Digital image correlation (DIC) software was used to analyze the images and determine strains in the specimen during testing, and for comparison with strains determined from elastic wave analysis as is typically done in Hopkinson bar testing. Numerical simulations of the experiments were performed to aid with interpreting the experiments and to confirm the accuracy of the test results.
Accomplishments — Two new grip and sheet specimen designs were implemented and proven; schematics of these designs are shown in Figure 2. This was accomplished using carefully designed experiments with aluminum and stainless steel sheet specimens. Materials were selected to determine the effectiveness of the grips for a range of conditions (strength, ductility and thickness: 0.125 to 0.5 mm). A slot grip design that incorporated mechanical restraints (pins, bolts) limited the extent of specimen slippage during testing. The size and placement of the grips holes were engineered to limit stress concentrations and minimize the possibility of crack initiation near the holes. Numerous dynamic tensile experiments, conducted at approximate strain rates of 103 s-1, demonstrated the reliability and consistency of the data. The results show good agreement for: i) strains derived from strain gages on the incident bar using elastic wave analysis (the usual method), ii) direct strain measurements from gages on the bar, and iii) strains maps derived from DIC analysis. Stresses derived from semiconductor gages, located on the transmitter bar approximately 150 mm from the specimen, provide more accurate stress measurements when the transmitted wave signal is weak. Comparison of the experimental data with numerical simulation results for each test condition confirmed the accuracy of the experimental results for ductile materials. For materials with low ductility, numerical analyses are included to ensure accurate interpretation of the experimental results.