Flyer Plate Experiments Research Articles

Computational simulations of the dynamic compaction of porous media

The goal of this study was to apply and compare different computational compaction models to the dynamic compaction of porous silicon dioxide (SiO 2) powder. Three initial specific volumes were investigated in this study, V 00=1.3, 4 and 10 cm 3/g, where the solid material specific volume is V 0=0.4545 cm 3/g. Two hydrodynamic codes, KO and CTH, were used to simulate the experimental results. Two compaction models, P– α and P– λ were implemented within CTH in conjunction with the Mie–Grüneisen (MG) equation of state. The snowplow (SP) compaction model was implemented within KO. In addition, the MG equation of state based on the experimentally measured Hugoniot was implemented within CTH and was compared to the data as well. One-dimensional flyer plate experiments were conducted with impact velocities ranging from 0.25 to 1.0 km/s, which corresponded to a shock incident pressure range of 0.77–2.25 GPa. The computational simulations were compared to the temporal lateral stress signatures measured with manganin gauges, placed before and after the silica powder. It was found that the MG equation of state (EOS) most accurately reproduce all of the experimental data whereas none of the compaction models accurately reproduced all of the experimental data. However, of the compaction models investigated that the P– α model tended to outperform the other considered.

Effect of shock compression method on the defect substructure in monocrystalline copper

Monocrystalline copper samples with orientations of [0 0 1] and [2 2 1] were shocked at pressures ranging from 20 to 60 GPa using two techniques: direct drive lasers and explosively driven flyer plates. The pulse duration for these techniques differed substantially: 40 ns for the laser experiments at 0.5 mm into the sample and 1.1∼1.4 μs for the flyer-plate experiments at 5 mm into the sample. The residual microstructures were dependent on orientation, pressure, and shocking method. The much shorter pulse duration in the laser driven shock yielded microstructures in recovery samples closer to those generated at the shock front. For the flyer-plate experiments, the longer pulse duration allows shock-generated defects to reorganize into lower energy configurations. Calculations show that the post-shock cooling for the laser driven shock is 10 3 ∼ 10 4 faster than that for plate-impact shock, increasing the amount of annealing and recrystallization in recovery samples for the latter. At the higher pressure level, extensive recrystallization was observed in the plate-impact samples, while it was absent in laser driven shock. An effect that is proposed to contribute significantly to the formation of recrystallized regions is the existence of micro-shear-bands, which increase the local temperature beyond the prediction from adiabatic compression.

Data assimilation with an extended Kalman filter for impact-produced shock-wave dynamics

Model assimilation of data strives to determine optimally the state of an evolving physical system from a limited number of observations. The present study represents the first attempt of applying the extended Kalman filter (EKF) method of data assimilation to shock-wave dynamics induced by a high-speed impact. EKF solves the full nonlinear state evolution and estimates its associated error-covariance matrix in time. The state variables obtained by the blending of past model evolution with currently available data, along with their associated minimized errors (or uncertainties), are then used as initial conditions for further prediction until the next time at which data becomes available. In this study, a one-dimensional (1D) finite-difference code is used along with data measured from a 1D flyer plate experiment. An ensemble simulation suggests that the nonlinearity of the modeled system can be reasonably tracked by EKF. The results demonstrate that the EKF assimilation of a limited amount of pressure data, measured at the middle of the target plate alone, helps track the evolution of all the state variables. The fidelity of EKF is further investigated with numerically generated synthetic data from so-called “identical-twin experiments”, in which the true state is known and various measurement techniques and strategies can be made easily simulated. We find that the EKF method can effectively assimilate the density fields, which are distributed sparsely in time to mimic radiographic data, into the modeled system.

Damage quantification and experimental simulation of momentum trapped flyer plate and tensile split Hopkinson bar incipient failure experiments

Damage quantification and experimental simulation of momentum trapped flyer plate and tensile split Hopkinson bar incipient failure experiments

Damage quantification and experimental simulation of momentum trapped flyer plate and tensile split Hopkinson bar incipient failure experiments

Dynamic and quasi-static failure and dynamic incipient failure experiments were performed on a half-hard 10100 Cu material using several different specimen geometries, including uniaxial stress and several notches. An additional series of uniaxial strain incipient failure experiments were performed using the annealed material. Damage quantification of the incipient failure specimens was performed and statistically reduced for comparison with continuum damage model predictions from explicit simulations of the experiments. Experimentally the test to test variability of the dynamic strain and flow stress measurements greatly exceeded that of the quasi-static measurements. Only the experiments with dynamic strain measurements close to post-test sample measurements were simulated. The ability of the simulations and the dynamic stress-strain measurements to detect the damage induced softening is low, at least until the damage coalesces into a crack. The damage model predictions of porosity evolution are within a factor of two for two experiments that differ in strain rate and stress triaxiality by over an order of magnitude.

Density-functional calculations of the liquid deuterium Hugoniot, reshock, and reverberation timing

The principal Hugoniot of liquid deuterium is calculated with density-functional methods. Particular attention is paid to the convergence of thermodynamic quantities with respect to the plane-wave cutoff energy and other simulation constraints. In contrast to earlier density-functional calculations, it is found that the principal Hugoniot results are in very good agreement with gas-gun data at lower pressures and compression ratios. The results at higher pressures are in very good agreement with data from magnetically launched flyer plates and show slightly less compression than earlier density-functional calculations. In addition to the principal Hugoniot, reshock states from a sapphire anvil and third-shock reverberation timings are also calculated. The latter are found to be in very good agreement with recently published results from magnetically launched flyer-plate experiments.

The Hugoniot and shock initiation threshold of lead azide

The Hugoniot of unreacted dextrinated lead azide has been determined using plate impact techniques in a 2.5 in. bore gas gun. The test specimens were compacted to a density of 3.4 gm/cc. Piezoresistive gauges were used to determine both the impact stress and the stress profile propagated through the sample. Each shot provided transmitted wave data for four sample thicknesses. The Hugoniot is linear up to 10 kbar. It can be represented by the equation σ=41.7 Up (σ in kilobars and Up in mm/μsec). The initiation threshold to long duration (3.5 μsec) shocks was 6.0 kbar based on evidence of reaction in the explosive inferred from increases in stress with propagation distance. Changes of wave speed were found to be noticeable only at higher stresses. Above 6 kbar the delay in initiation decreases with increasing stress levels. The initiation sensitivity to short duration (0.1 μsec) shocks was determined by impacting the lead azide with thin flyers accelerated by exploding foils. The impact velocity was determined by observation with a high-speed framing camera and the transmitted stress profiles were measured by x-cut quartz gauges. The short-pulse initiation threshold for polyvinyl lead azide is 4.2 kbar for a density of 3.6 gm/cc, and 2.1 kbar for dextrinated lead azide for a density of 2.9 gm/cc. Comparison of initiation threshold measurements suggests that there is a minimum thickness, or a run-up distance, before detonation occurs which is independent of pulse width with stresses up to 10 kbar. For long pulse (3.5 μsec) experiments with the gas gun, no evidence of detonation was detected for 1 mm thick samples. Detonation occurs with run distances in the range of 1–2 mm for impact stresses of 8.9 kbar. For stresses greater than 6.0 kbar, evidence of detonation was noted after a 2 mm run. In the thin flyer plate experiments at stresses of 8 kbar dextrinated lead azide displayed a 2 μsec initiation delay. Close to the initiation threshold the stress profiles in the explosive show a gradual transition from an unreactive shock to a stable detonation. It is postulated that reaction occuring behind the shock front produces pressure waves which travel through the explosive interacting with the unreacted explosive ahead of the reaction front causing a nonuniform rate of growth of the shock. This behavior is similar to that observed for heterogenous secondary explosives. The wave speed of the initial wave front propagating in the explosive changes abruptly from that appropriate to unreacted material to the detonation velocity.

On the behavior of stress waves in composite materials—II. Theoretical and experimental studies on the effects of constituent debonding

The theory derived in Part I is compared to data obtained from flyer-plate experiments on laminated composites. The composites were constructed of alternating layers of aluminum and polymethyl methacrylate. Impact occurred on a flat plane oriented perpendicular to the interface planes of the composite constituents. As opposed to the behavior of a fully bonded composite which allows only one longitudinal stress wave to propagate through its interior, two longitudinal stress waves were observed in the debonded composite.Reasonable agreement was achieved in the comparisons of theory to experiment after adjustment was made for the effect of residual bond strength at the constituent interfaces.

A Lattice Model for Stress Wave Propagation in Composite Materials

Geometric dispersion, observed in a wide variety of composite materials, is believed to result mainly from the relatively periodic arrangement of the reinforcing elements in the matrix rather than from the precise shape of each reinforcing element. On the basis of this observation, a lattice model for composite materials which ignores the shape of the reinforcing elements but preserves their periodicity has been developed. For a wide range of engineering applications, this model can be used to predict the behavior of actual engineering composites. In the application of the lattice model to a specific material, consideration of the dispersive characteristics of the composite are set aside, initially, and the composite is treated as a nondispersive homogeneous mixture. The effective or average properties of the mixture are determined either by steady-wave analysis or appropriate experiments. A lattice is then formed by redistributing the mass within the mixture to form a periodic structure of laminated plates. This mass redistribution is carried out in a manner which yields a lattice with theoretical dispersive characteristics that match the measured dispersive characteristics of the composite. The model was applied to composites consisting of a regular array of tungsten fibers in an aluminum matrix and composed of 2.2 and 22.1 percent by volume of tungsten. Two flyer-plate experiments were performed in the plastic range of the composite. The agreement between experiment and calculation for the arrival time and rise time of the wave front and for the frequency of the ringing behind the wave front is good.