69
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2021
L. Lamberson et al. (eds.), Dynamic Behavior of Materials, Volume 1, Conference Proceedings of the Society
for Experimental Mechanics Series, https://doi.org/10.1007/978-3-030-59947-8_13
Chapter 13
Development of a Micro Tensile Kolsky Bar
Daniel T. Casem
Abstract A design for a miniature tensile Kolsky bar is proposed. The design is intended to test metallic samples with typical gage lengths of 500 μm and cross-sectional areas on the order of 10,000 μm
2
. The bars will be steel and 1.5 mm in diameter, although smaller diameter output bars will be used to suit sample needs. The projectile, at least in the initial design, will
be accelerated by a spring to avoid the complexities involved in fabricating a small gas gun. Optical instrumentation will be
used instead of strain- gages; normal displacement interferometers will be used to measure the displacement of the free end
of each bar. This will provide the necessary data to analyze the specimen behavior. The ultimate objective is to use this bar
to study the influence of microstructural defects on failure of ductile metals.
Keywords High rate mechanical testing · Tension · Kolsky bar · Split Hopkinson Pressure bar · Micromechanics
13.1 Introduction
Kolsky (Split Hopkinson Pressure) bars are the standard mechanical test for obtaining data in the strain-rate range of
1000–10,000/s [1, 2]. A number of authors have developed miniature versions with the goal of increasing the maximum
strain-rate that can be achieved [3–5]. Reducing bar diameter allows for higher frequency one-dimensional waves to propagate in the bars, providing a better temporal resolution of the loads and displacements applied to the sample. This leads to
higher quality stress-strain data. Small bars are also better suited to testing small samples, the use of which improves the state
of quasi-static equilibrium as the sample is rapidly deformed. Bars as small as 300 μm diameter have been used, and rates as
high as 700k/s have been obtained [6].
Tensile Kolsky bars are less widely used than compression bars but have seen considerable development. Their operation is very
similar to that of compression bars, with the chief additional complexities arising from the need to introduce a tensile pulse to load
the sample, typically via some sort of impact with a projectile, and the need to grip the sample. The gripping issue is of course not
unique to the high rate method; the same sorts of issues arise in low rate testing with servo-hydraulic load frames, although they are
exacerbated at high rates due to the need to consider the inertia of the gripping components. Nicholas and Bless [7] discuss some
of these issues, and reviews several different tensile bar designs. In the most classic arrangement [8, 9], the input bar passes through
a tubular striker which is propelled by a gas gun at a flange on the impact end of the bar; the flange transmits the tensile wave into
the bar. Several other approaches are described in [7]. More recently, additional designs have been developed. In one case [10], a
pre-tensioned section of the input bar is clamped and the dynamic load is imparted by its sudden release, similar to the approach
taken in most torsion systems [11]. In another design [12], a hollow input bar serves as the gun barrel. The solid cylindrical projectile is launched at the closed end of the hollow bar, opposite the sample, and its impact generates a tensile pulse (the bar is vented
to avoid an air cushion). This permits the use of standard waveshapers to modify the incident pulse. Both of these approaches avoid
one major shortcoming of the classic arrangement: the length of the bar over which the projectile passes cannot be supported by
bushings, and this can lead to poor quality signals. Another approach to avoid this problem is the use of a striker with a U-shaped
cross-section [13]. By leaving the underside of the striker open, the input bar can be supported by bushings at least on the bottom
side, eliminated sag in the bar due to its weight.
D. T. Casem (*)
US Army Combat Capabilities Development Command Army Research Laboratory, Aberdeen Proving Ground, MD, USA
e-mail: daniel.t.casem.civ@mail.mil
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2021
L. Lamberson et al. (eds.), Dynamic Behavior of Materials, Volume 1, Conference Proceedings of the Society
for Experimental Mechanics Series, https://doi.org/10.1007/978-3-030-59947-8_13
Chapter 13
Development of a Micro Tensile Kolsky Bar
Daniel T. Casem
Abstract A design for a miniature tensile Kolsky bar is proposed. The design is intended to test metallic samples with typical gage lengths of 500 μm and cross-sectional areas on the order of 10,000 μm
2
. The bars will be steel and 1.5 mm in diameter, although smaller diameter output bars will be used to suit sample needs. The projectile, at least in the initial design, will
be accelerated by a spring to avoid the complexities involved in fabricating a small gas gun. Optical instrumentation will be
used instead of strain- gages; normal displacement interferometers will be used to measure the displacement of the free end
of each bar. This will provide the necessary data to analyze the specimen behavior. The ultimate objective is to use this bar
to study the influence of microstructural defects on failure of ductile metals.
Keywords High rate mechanical testing · Tension · Kolsky bar · Split Hopkinson Pressure bar · Micromechanics
13.1 Introduction
Kolsky (Split Hopkinson Pressure) bars are the standard mechanical test for obtaining data in the strain-rate range of
1000–10,000/s [1, 2]. A number of authors have developed miniature versions with the goal of increasing the maximum
strain-rate that can be achieved [3–5]. Reducing bar diameter allows for higher frequency one-dimensional waves to propagate in the bars, providing a better temporal resolution of the loads and displacements applied to the sample. This leads to
higher quality stress-strain data. Small bars are also better suited to testing small samples, the use of which improves the state
of quasi-static equilibrium as the sample is rapidly deformed. Bars as small as 300 μm diameter have been used, and rates as
high as 700k/s have been obtained [6].
Tensile Kolsky bars are less widely used than compression bars but have seen considerable development. Their operation is very
similar to that of compression bars, with the chief additional complexities arising from the need to introduce a tensile pulse to load
the sample, typically via some sort of impact with a projectile, and the need to grip the sample. The gripping issue is of course not
unique to the high rate method; the same sorts of issues arise in low rate testing with servo-hydraulic load frames, although they are
exacerbated at high rates due to the need to consider the inertia of the gripping components. Nicholas and Bless [7] discuss some
of these issues, and reviews several different tensile bar designs. In the most classic arrangement [8, 9], the input bar passes through
a tubular striker which is propelled by a gas gun at a flange on the impact end of the bar; the flange transmits the tensile wave into
the bar. Several other approaches are described in [7]. More recently, additional designs have been developed. In one case [10], a
pre-tensioned section of the input bar is clamped and the dynamic load is imparted by its sudden release, similar to the approach
taken in most torsion systems [11]. In another design [12], a hollow input bar serves as the gun barrel. The solid cylindrical projectile is launched at the closed end of the hollow bar, opposite the sample, and its impact generates a tensile pulse (the bar is vented
to avoid an air cushion). This permits the use of standard waveshapers to modify the incident pulse. Both of these approaches avoid
one major shortcoming of the classic arrangement: the length of the bar over which the projectile passes cannot be supported by
bushings, and this can lead to poor quality signals. Another approach to avoid this problem is the use of a striker with a U-shaped
cross-section [13]. By leaving the underside of the striker open, the input bar can be supported by bushings at least on the bottom
side, eliminated sag in the bar due to its weight.
D. T. Casem (*)
US Army Combat Capabilities Development Command Army Research Laboratory, Aberdeen Proving Ground, MD, USA
e-mail: daniel.t.casem.civ@mail.mil