Hypervelocity Impact Simulations Research Articles

Numerical simulation of hypervelocity impact on CFRP/Al HC SP spacecraft structures causing penetration and fragment ejection

A representative carbon fiber reinforced plastic/aluminum honeycomb sandwich panel (CFRP/Al HC SP) spacecraft structure has been modeled in the hydrocode AUTODYN using the state-of-the-art ADAMMO material model [Riedel W, Harwick W, White D, Clegg R. Advanced material damage models for numerical simulation codes. ESA CR(P) 4397, 2003] to study the performance of the structure during impact events that cause perforation and fragment ejection. A new procedure combining a series of existing theoretical methods has been developed and applied to derive a full set of coarse material data. The data set has been implemented in AUTODYN, and the results of the numerical simulation have been compared to experimental impact test data. For impact tests performed near the structural ballistic limit, quantitatively accurate results were obtained over a range of impact velocities and angles. A further increase in the projectile size resulted in significant destruction of the sandwich panel front face-sheet and diversion from the experimental damage measurements. Inspection of the numerical model has shown non-localized propagation of inter-laminar delaminations, possibly caused by an under-prediction of the laminate dynamic inter-laminar tensile strength. The effects of the delamination propagation occur over an extended time scale and were not found to affect the state and trends of the fragment cloud ejected into the satellite interior. Accordingly, experimental trends of fragment cloud dispersion have been qualitatively reproduced.

Simulation of orbital debris impact on the Space Shuttle wing leading edge

A series of three dimensional hypervelocity impact simulations has been performed to study the effects of orbital debris impact on the Space Shuttle wing leading edge. The simulations employed an improved hybrid particle-finite element method and an orthotropic elastic-plastic material model recently developed for reinforced carbon–carbon. The simulation results are consistent with the available experimental data, and suggest the use of momentum scaling to estimate damage effects for impact conditions outside the range of current light gas gun technology. Projectile shape and orientation effects appear to be modest for flat plate projectiles at impact velocities above the ballistic limit.

Improving Metallic Thermal Protection System Hypervelocity Impact Resistance Through Numerical Simulations

A design of experiments approach has been implemented using computational hypervelocity impact simulations. The purpose of this study is to determine the most effective place to make minor design changes to an existing metallic thermal protection system to improve hypervelocity impact protection provided to underlying components. Simulations are performed using axisymmetric models in a shock-physics hydrodynamics code and compared with existing experimental data. The axisymmetric models are then used in a statistical sensitivity analysis to determine the influence of design parameters on degree of protection. Several damage metrics are identified and evaluated. Damage metrics related to the extent of substructure damage produce misleading results; however, damage metrics related to the degree of dispersion of the hypervelocity projectile produce results that correspond to physical intuition. Based on analysis of variance results, it is concluded that increasing the spacing between the outer surface and the substructure is the most effective way to increase hypervelocity impact resistance. The second most effective design change is to increase the thickness of the outer foil layer. When design considerations are taken into account for the system in this study, it is explained that increasing the thickness of the outer foil layer is the most practical design change.

Simulation of hypervelocity impact on aluminum-Nextel-Kevlar orbital debris shields

An improved hybrid particle-finite element method has been developed for hypervelocity impact simulation. The method combines the general contact-impact capabilities of particle codes with the true Lagrangian kinematics of large strain finite element formulations. Unlike some alternative schemes which couple Lagrangian finite element models with smooth particle hydrodynamics, the present formulation makes no use of slidelines or penalty forces. The method has been implemented in a parallel, three dimensional computer code. Simulations of three dimensional orbital debris impact problems using this parallel hybrid particle-finite element code, show good agreement with experiment and good speedup in parallel computation. The simulations included single and multi-plate shields as well as aluminum and composite shielding materials. at an impact velocity of eleven kilometers per second.

Extention and validation of a hybrid particle-finite element method for hypervelocity impact simulation

The hybrid particle-finite element method of Fahrenthold and Horban [1], developed for the simulation of hypervelocity impact problems, has been extended to include new formulations of the particle-element kinematics, additional constitutive models, and an improved numerical implementation. The extended formulation has been validated in three dimensional simulations of published impact experiments. The test cases demonstrate good agreement with experiment, good parallel speedup, and numerical convergence of the simulation results.

An improved hybrid particle-element method for hypervelocity impact simulation

An improved hybrid particle-finite element method has been developed for hypervelocity impact simulation. The method combines the general contact-impact capabilities of particle codes with the true Lagrangian kinematics of large strain finite element formulations. Unlike some alternative schemes which couple Lagrangian finite element models with smooth particle hydrodynamics, the present formulation makes no use of slidelines or penalty forces.

Hydrocode modelling of hypervelocity impact on brittle materials: depth of penetration and conchoidal diameter

Summary The Johnson-Holmquist brittle material model has been implemented into the AUTODYN hydrocode and used for Lagrangian simulations of hypervelocity impact of spherical projectiles onto soda-lime glass targets. A second glass model (based on a shock equation of state and the Mohr-Coulomb strength model) has also been used. Hydrocode simulations using these two models were compared with experimental results. At 5 km s−1, the Mohr-Coulomb model under-predicted the depth of penetration, whilst adjustment of the Johnson-Holmquist model bulking parameter was required to match the experimental data to the simulation results. Neither model reproduced the conchoidal diameter; a key measured parameter in the analysis of retrieved solar arrays, so two failure models were used to investigate the tensile failure regime. A principal tensile failure stress model, with crack softening, when used with failure stresses between 100 and 150 MPa and varying bulking parameters, reproduced the conchoidal diameter morphology. Empirically-determined, power-law damage equation predictions for the range 5–15 km s−1 were compared with simulations using both models since no experimental data was available. The powerlaw velocity dependence of the depth of penetration simulations was found to be significantly lower than the 0.67 predicted by the empirically-determined damage equations.

Numerical simulation of oblique impact on orbital debris shielding

Hybrid particle-finite element methods have been proposed as a modeling methodology well suited to the problem of hypervelocity impact simulation. To evaluate the use of such numerical methods for orbital debris shielding design, a series of simulations have been conducted using a three dimensional hybrid particle-finite element code now under development. Two sets of oblique impact simulations, one for a single bumper Whipple shield and one for a dual bumper or stuffed Whipple shield, have been compared to published ballistic limit equations, the latter derived from experiment. The results indicate that hybrid particlefinite element methods can provide an accurate and computationally tractable approach to the simulation of orbital debris shield performance.

Thermodynamics of continuum damage and fragmentation models for hypervelocity impact

General numerical models of hypervelocity impact problems must account for finite strain deformation, isochoric rate dependent plasticity, volumetric and deviatoric damage, and complex energy domain coupling. Incorporating all these effects into current damage and fragmentation models is difficult, given their limited thermodynamic framework. An alternative, systematic approach to the material model formulation process results in a general thermodynamic framework which can incorporate a variety of constitutive assumptions. Application of the method is illustrated by formulation of an elastic-viscoplastic damage model with finite strain kinematics, Grady-Kipp volumetric damage, Johnson-Holmquist deviatoric damage, and thermodynamic compuling through an entropy state. Hypervelocity impact simulations using the developed model show that predictions of fragment size and fracture surface area based on Grady-Kipp fragmentation theory vary significantly with the extent of plastic deformation, over a velocity range of five to ten kilometers per second.

Energy based particle hydrodynamics for hypervelocity impact simulation

Hamiltonian mechanics provides a simple, systematic, energy based approach to the formulation of particle models for hypervelocity impact simulation. The methodology may be applied to particle models based either on mass points which define moving interpolations or on physical particles described by Lagrangian control volumes. Implementation of the method in a three dimensional computer code allows for the simulation of oblique hypervelocity impact on multi-plate orbital debris shield designs.

Three dimensional hypervelocity impact simulation for orbital debris shield design

This paper assesses a Whipple shield impact simulation method which is both accurate and computationally efficient. The paper documents the simulation methodology and results of Whipple shield simulations at an oblique impact angle of 30°. These results are compared with HVI experiments to demonstrate the accuracy of the simulation technique. In addition, simulations of Whipple shields in the velocity regime above 8km/s were completed and the results compared to published ballistic limit equations to demonstrate the reliability of these equations. Finally, the paper documents computational efficiency of the simulation technique.

Oblique hypervelocity impact simulation for whipple shield-protected structures

Three dimensional simulation of hypervelocity impacts on space structures places extreme demands on even supercomputer resources. This is due in large part to the fact that conventional Eulerian and Lagrangian codes are ill-suited to model debris cloud evolution. To reduce the computer time and memory requirements of oblique impact simulations, a three-dimensional elastic-plastic particle dynamics model of debris cloud evolution has been developed. This numerical model has been linked to an Eulerian hydrocode and a Lagrangian structural code, and applied in the simulation of oblique hypervelocity impacts on Whipple shield-protected structures.

Higher acceleration technique of microprojectiles in a magnetoplasmadynamic accelerator

For the simulation of hypervelocity impact of micrometeoroids and microdebris on a spacecraft, an improved acceleration technique is proposed in a magnetoplasmadynamic accelerator. The compression of the plasma flow was enhanced due to modification of the magnetic-compressor-coil geometry in the accelerator. Such enhancement was found to be effective for the improvement of the flight parameters of microprojectiles. Due to the modification of the coil geometry, the velocity of glass beads of 500–600 μm diam was increased from about 3 to 6 km/s which is included in the hypervelocity impact regime of projectiles in space. In addition, the main direction of the projectile flight was improved to become more parallel to the central axis of the accelerator.

Hypervelocity impact tests and simulations of single whipple bumper shield concepts at 10km/s

A series of experiments has been performed to evaluate the effectiveness of a Whipple bumper shield to orbital space debris at impact velocities of ∼ 10 km/s. Upon impact by a 19 mm (0.87 mm thick, L/D ∼0.5) flier plate, the thin aluminum bumper shield disintegrates into a debris cloud. The debris cloud front propagates axially at velocities of ∼14 km/s and expands radially at a velocity of ∼7 km/s. Subsequent loading by the debris on a 3.2 mm thick aluminum substructure placed 114 mm from the bumper penetrates the substructure completely. However, when the diameter of the flier plate is reduced to 12.7 mm, the substructure, although damaged is not perforated. Numerical simulations performed using the multi-dimensional hydrodynamics code CTH also predict complete perforation of the substructure by the subsequent debris cloud for the larger flier plate. The numerical simulation for a 12.7 mm flier plate, however, shows a strong dependence on assumed impact geometry, i. e., a spherical projectile impact geometry does not result in perforation of the substructure by the debris cloud, while the flat plate impact geometry results in perforation.

Advanced algorithms for computer simulation of hypervelocity impact

Presented are several algorithms that allow a Lagrangian hydrodynamic computer code to simulate hypervelocity impact. Such impacts typically involve very large material deformations and ordinarily can be handled only by Eulerian codes, in which the materials flow through a stationary grid. In Lagrangian codes, the mesh is attached to the material and deforms with it; in problems involving severe distortions, the mesh can become tangled, resulting in loss of accuracy or even a breakdown of the computational scheme. However, Lagrangian codes present several advantages in economy, ease of programming and use, and interpretation of results. This paper describes several modifications of the DEFEL code ( Flis, Miller, and Clark, 1984). This code, a descendant of EPIC-2 ( Johnson, 1976), was originally developed to simulate low- to medium-velocity impact and explosive-metal interaction. Modifications include automatic rezoning, a surface-erosion model, and treatment of fracture (cracking) by automatic introduction of sliding lines between elements. These fractures allow DEFEL to handle a variety of problems in the hypervelocity regime. Included are discussions of several example problems.

Sensitivity of hypervelocity impact simulations to equation of State

We investigate the sensitivity of hypervelocity impact simulations to differences in equations of state (EOS's), comparing our hydrocode calculations to recent experiments on lead at G. M. Delco ( Pomykal, 1986). The results suggest that the hydrocode does not correctly predict all the details of debris cloud structure, even when these are significant EOS differences. We used three different equations of state, one of which is new. It was generated just for sensitivity studies using the most recent experimental EOS data.

New two-stage accelerator for hypervelocity impact simulation

The present knowledge of the micrometeoroid environment and the simulation requirements for laboratory devices show that the simulated mass-velocity range must be extended to higher values in both mass and velocity. This is accomplished with the new two-stage accelerator, which consists of a light gas gun as the first stage and a coaxial accelerator with a compressor coil as the second stage. The operation of the new device is described together with a theoretical investigation of the timing of the first and the second stage, which leads to an appropriate adjustment of the two stages. The theoretical and experimental results agree well and show that 600-micron diameter glass beads have been accelerated to velocities near 20 km/sec.

Modelling hypervelocity impact in DYNA3D

This paper presents part of the work on development of tools and modelling techniques for simulation of hypervelocity impact on spacecraft structures using the Lawrence Livermore DYNA3D code. Calculations were performed to evaluate the current capability of DYNA3D to model hypervelocity impact on multiple plate structures. These have shown that improvements need to be made to DYNA3D before it can be used to produce reliable and accurate results. Two important areas that require improvement are the contact algorithm and the equation of state. Impacts with relative velocities of up to 15 kms~* generate pressures of several Mbar. Existing DYNA equations of state (EOS) are not accurate at these high pressures and a new EOS was needed, hence the SESAME equation of state library was selected. This is a tabulated equation of state which covers the range of densities and pressures of interest. The SESAME EOS has been implemented into DYNA3D, and tested. The element erosion contact algorithm in DYNA was used to model bumper shield perforation. The existing erosion algorithm was found to interfere with the formation of shock waves in both bodies. A new erosion criterion solved this problem and gave significantly more accurate results. Additional modification is needed to allow accurate modelling of impact on multi-plate structures.