Elsevier

Engineering Structures

Volume 228, 1 February 2021, 111467
Engineering Structures

Experimental and numerical investigation on the perforation resistance of double-layered metal shields under high-velocity impact of soft-core projectiles

https://doi.org/10.1016/j.engstruct.2020.111467Get rights and content

Highlights

  • Simulation of ballistic impact of soft-core projectile against steel plates.

  • The SPH method was used for the discretization of the projectile.

  • High-strength steel plates of different thicknesses.

  • Study of different configurations of double-layer target.

  • Simulation of the extreme deformation occurring in the projectile.

Abstract

Multi-layered metal shields have been extensively investigated in the past, mainly studying the effect of layering, spacing and order of plates of different materials. However, at present is not feasible to predict the effect of layering and spacing for a particular case but only general pattern can be considered. Some investigations based on numerical simulations showed that the multi-layered configuration made by layers of different metals could be a promising solution in terms of weight reduction of the shield but, different authors obtained contradictory results and comprehensive experimental investigation is still missing. Thus, the scope of this study is to evaluate the perforation resistance of double-layered metal shields constituted by a high-strength steel and an aluminum layer when impacted by a soft-core projectile. It was found that the multi-layered configuration constituted of two layers of different metals, is less efficient than the monolithic configuration. Furthermore, numerical models were developed to increase the understanding of the problem. It was necessary to use an alternative to the commonly used FE method to model extreme deformation occurring in soft-core projectiles in such kinds of scenario. In particular, the SPH method was implemented providing a detailed parametric study of the most relevant numerical parameters. The SPH method was thus exploited to study different configurations of the double-layer target, and it was found that the specific ballistic energy depends on the ratio between the two layers of the target.

Introduction

A ballistic shield is a device aimed to stop or deflects bullets aimed at the carrier. The main requirement of a ballistic shield is the protection level, that is the capability to withstand a specific threat. Different civil and military standards have been developed for the assessment of ballistic shields. For example, EN 1522 [1] is an European standard for the classification of the bullet resistance of windows, doors, shutters and blinds, which defines seven protection levels of increasing menace. Each menace corresponds to a bullet with its own geometrical features, materials, mass and impact velocity. Ballistic shields are assessed experimentally by performing ballistic tests in which a projectile is impacted, usually with a normal incident angle, on the ballistic shield and the impact velocity is measured. An important property of a ballistic shield is the ballistic limit velocity which is the theoretical minimum impact velocity for which the projectile is expected to consistently completely perforate the ballistic shield. Due to the impossibility of precisely controlling the impact velocity and the existence of a zone of mixed results in which the projectile may completely penetrate or only partially penetrate under apparently identical conditions, the ballistic limit velocity is approximated by the V50, that is the impact velocity at which there is 50% probability of complete penetration of the target [2].

High-strength steels is a possible choice for the manufacturing of a ballistic shield due to their excellent mechanical properties and lower cost with respect to other more sophisticated materials such as fiber-reinforced composites and advanced ceramics. To obtain EN 1522 [1] FB6 protection levels a high-strength steel plate with a thickness in the range of 6.5–5 mm is required depending on the hardness of the steel. The higher the hardness of the steel the lower is the required thickness. A plate with a thickness of 5–6.5 mm has got an areal density of approximately 39–51 kg/m2, therefore it considerably increases the weight of the shielded component. Lightweight shields are useful for the protection of vehicles in order to decrease the fuel consumption and increase the payload and maneuverability. Thus, the optimization of ballistic shields, intended as the reduction of weight necessary to obtain the required protection level is of particular interest and many researches were performed in this field. Several authors investigated the performance of multi-layered metal shields [3], [4], [5], [6], [7], [8], [9], [10], [11], [12] either with experimental tests or numerical simulations. Different were the aspects investigated in this field: layering which means replacing a monolithic plate by several plates in contact having the same total thickness; spacing which means apply an air gap between the plates; use layer of different materials. The idea behind a multi-layered metal shield is to obtain an optimal solution exploiting only metallic materials, thus avoiding the high cost and other drawbacks related to composite materials. Recently, Ben-Dor et al. [13] presented a review about the ballistic performance of multi-layered metal shields. The authors stated that at present is not feasible to predict the effect of layering and spacing for a particular case but only general pattern can be considered. In some cases, the effects of layering and of spacing are of the same order of magnitude of the experimental errors, leading to a result that is not very reliable. Taking into account these remarks, the general trends detected for sharp projectiles are: layering and spacing decreases the ballistic resistance; increasing the number of layers decreases the ballistic resistance of space shields with layers of the same width; the order of plates made of different material significantly affects the ballistic resistance of the shield. The general trends for blunt projectiles are instead: in most of the cases layering and spacing has a negative effect on the ballistic resistance but, irrespectively on its sign (improvement or impairment), it is less pronounced than for sharp projectiles; the order for layers manufactured from different material or manufactured from the same material but with different thickness affects the ballistic resistance. Some investigations based on numerical simulations [5], [8], [10] showed that the multi-layered configuration made by layers of different metals could be a promising solution in terms of weight reduction of the shield. However, authors obtained contradictory results. Flores-Johnson et al. [5] investigated the ballistic resistance o monolithic, double- and triple-layered metallic plates made by Weldox 700E and Al7075-T651 or a combination of these materials impacted by 7.62 mm armor piercing projectile in the velocity range of 775–950 m/s. Taking into consideration different multi-layered mixed plate configurations with a maximum thickness of 20 mm and the same areal density the shield with an upper layer of 6.66 mm of Al7075-T651 and a lower layer of 13.33 mm of Weldox 700E proved to have the highest ballistic performance. Rahman et al. [8] presented a numerical investigation on the ballistic limit velocity of multi-layered plates made of high strength steel and Al7075-T6 under impact of 7.62 mm armor piercing projectile with an impact velocity in the range of 700–1200 m/s. The configurations considered were monolithic, double-layered and triple-layered. The best performance was achieved with a triple-layered configuration with a high-strength steel plate of 8 mm as the first and last layer and an Al7075-T6 plate of 9 mm as the middle layer. Thus Flores-Johnson et al. [5] and Rahman et al. [8] arrived to different conclusions. Indeed, results of numerical models which were not validated with experimental test on multi-layered target should be considered with cautiousness. Flores-Johnson et al. [5] themselves stated that their results were obtained with numerical simulations, and experimental validation should be performed. The authors were able to find only few studies in the open literature which experimentally asses the effect of layers of different materials on the ballistic performance of multi-layered metallic shields [6], [7], [12]. Yunfei et al. [7] investigated the ballistic performance of double-layered plates made of different steels impacted by blunt-nosed and ogive-nosed projectiles. The materials considered were 45 steel, with a yield stress of 714 MPa, and Q235 steel with a yield stress of 299 MPa. Both the blunt-nose and ogive-nosed projectile were made of 38CrSi hardened steel, have a diameter of 12.7 mm and a mass of 34.5 g. The double-layered tested configurations have a total thickness of 12 mm and are made of in-contact monolithic plates of 6 mm: upper layer of 45 steel and lower layer of Q235 steel (H + A) or upper layer of Q235 steel and lower layer of 45 steel (A + H). The ballistic limit velocity was dependent both on the target configuration and the projectile nose shape. The ballistic limit velocity increased by 7.2% for the blunt-nosed projectile and by 2.1% for the ogive-nosed projectile passing from the A + H to the H + A configuration. The ballistic limit velocity was 58.6% higher for the blunt-nose projectile. Babaei et al. [6] experimentally investigated the perforation resistance of spaced double-layered targets made by different materials including steel and aluminum impacted by cylindrical steel projectiles with a diameter of 4 mm and a length of 32 mm. The configuration considered were: 1 mm aluminum + 1 mm aluminum (al-al), 1 mm aluminum + 1 mm steel (al-st), 1 mm steel + 1 mm aluminum (st-a) and 1 mm steel + 1 mm steel (st-st). The ballistic resistance was higher for st-st, st-al, al-st and al-al respectively. Comparing the results obtained for al-st with st-al, the ballistic limit velocity slightly increases if the steel plate is situated as the first layer. Rahman et al. [12] studied two double-layered configurations with the same areal density: 15 mm Ar500 steel + 10 mm AA7075-T6 and 10 mm AA7075-T6 + 15 mm Ar500 steel. The experimental tests were limited at a velocity range of 800–850 m/s for which the projectile, which was a 7.62 mm full metal jacket bullet, was not able to completely perforate none of the two configurations. The difference in the penetration depth for the two configurations was quite significant (600%). This study aims at investigating the ballistic performance of a double-layered shield manufactured with two different metallic plates: a steel and an aluminum plate. While experimental test showed that in a double-layered shield with a steel and an aluminum plate, it is better than the opposite configuration [6], [12], the authors were not able to find comparison with a monolithic high-strength steel plate of the same weight.

The scope to this study is to evaluate, either experimentally and by means of numerical models, if a double-layered metallic shield constituted by a steel and aluminum plate performs better than a high-strength steel plate of the same weight, which is nowadays the common solution adopted for the manufacturing of metal ballistic shields. In particular, this study is performed using a soft-core projectile, corresponding to EN 1522 [1] FB6 protection level. This type of projectile is subjected to extreme deformation and fracturing during the impact making the numerical modelling a complex task. Therefore, another aim of this study is to implement the smoothed particle hydrodynamics (SPH) method for the simulation of the projectile extreme deformation, as an alternative to the commonly used finite element (FE) method.

Predictive models play an important role in the reduction of time and cost related to experimental tests. For this reason, several studies in the literature focused on the development of numerical models of high-velocity impact on metal shields usually exploiting the FE method [3], [4], [5], [6], [12], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25]. During high-velocity impact, materials are subjected to large deformation and failure, subsequently a failure criterion must be implemented in the constitutive model. The common FE method is not capable of handling large deformations because it is mesh based and it is limited by issues of mesh tangling and negative volume. For this reason, artificial erosion is usually implemented. This methodology has been successfully used to model steel projectiles since they are subjected to limited plastic deformation during the impact. On the other hand, this is not true for soft-core projectiles. As experimentally observed by Borvik et al. [4], immediately after impact, the soft core starts to mushroom and break down; both the jacket and the core fracture and the materials start to spray backwards forming a cloud of debris. The FE method in not suited for the simulation of this behavior due to its aforementioned intrinsic limitations. Indeed, as experienced by Borvik et al. [4] neither Lagrangian formulation nor multi-material Eulerian formulation could effectively treat this problem. Meshless methods like the discrete element method (DEM), SPH method or element-free Galerkin (EFG) method can also be employed to simulate problems with large deformations and failure [26]. The SPH method is an approach developed by Gingold and Monaghan [27] and Lucy [28] which incorporates a pseudo-particle interpolation approach to compute smoothed field variables. In this method the geometry is discretized by a defined group of pseudo-particles, each pseudo-particle having a mass, a Lagrangian position, Lagrangian velocity and internal energy. Due to the absence of a fixed mesh, this method is well suited for large deformation problems, overcoming the limitation of the FE method. The SPH method was primarily intended for the simulation of fluid flow but it has also been exploited in the literature for the simulation of different scenarios involving impact [29], [30], [31], [32], [33], [34], [35], [36], [37]. However, the authors were able to find only few works about high-velocity impact on metal plates [38], [39], [40], [41]. Xiao et al. [38] developed axisymmetric SPH models for the perforation of aluminum plates. A particle-to-particle contact algorithm was incorporated to treat contact interfaces explicitly and eliminate virtual stresses at contact interfaces. The artificial stress method was extended to remove tensile instability. The numerical model, when compared with experimental results, showed the ability to predict the residual velocity. Swaddiwudhipong et al. [39] developed a numerical model exploiting the SPH method for the simulation of high-velocity impact on Weldox 460E steel plates considering different target thicknesses and projectile shapes. Kılıç and Ekici [40] developed numerical models for the simulation of high-velocity impact of a 7.62 mm armor piercing bullet against high-strength steel plates and compared FE and SPH method for the discretization of the target. They concluded that although the SPH method is expected to predict spall and fragmentation behavior more accurately, there was not a significant difference between the FE and SPH method. Mohotti et al. [41] exploited the SPH method for the simulation of high-velocity impact of a 5.56 mm NATO standard projectile against a monolithic AA5061-H116 plate and performed a parametric study. It was found that the renormalization approach is more accurate in the prediction of the residual velocity. The scale factor of the smoothing length, the number of cycles between particle sorting and the particle spacing instead showed insignificant influence on the residual velocity.

As already mentioned, the scope to this study is to evaluate, either experimentally and by means of numerical models, if a double-layered metallic shield constituted by a high-strength steel and aluminum plate performs better than a high-strength steel plate of the same weight, when impacted by a soft-core projectile. To the authors knowledge, it has not been verified yet, by means of experimental tests, whether this multi-layered configuration is better than the monolithic plate of the same weight. This is an important verification since the possibility to substitutive a monolithic plate with a multi-layered configuration could be an interesting option in the manufacturing of ballistic shields. Furthermore, a soft-core projectile was used in the experimental test, which extremely deforms and fractures during the impact. Thus, it was necessary to use an alternative to the commonly used FE method to handle this complex modelling task. In particular the SPH method was implemented for the simulation of the projectile extreme deformation providing a detailed parametric study of the most relevant numerical parameters. In section 2 the experimental tests are described: high-velocity impact tests were performed, using a 0.308 Winchester bullet against high-strength steel plates of different thicknesses thus providing both the scenario of complete perforation and arrest of the projectile. Furthermore, high-velocity impact tests were performed on double-layered shields constituted by a high strength steel and an aluminum plate. It was thus possible to experimentally compare the ballistic performance of monolithic and multi-layered targets of the approximately the same weight. In section 3 the numerical models are developed and calibrated using the results of high-velocity impact tests on monolithic targets. First of all, the FE method was exploited showing that it is incapable of simulating the projectile extreme deformation; secondly the SPH method was exploited for the discretization of the projectile and a parametric study was performed on different model parameters. Discussion of the results is given in Section 4. Even if it was possible to develop a numerical model of high-velocity impact which accurately predicts the projectile residual velocity in the case of monolithic target, the FE method predicted severe erosion of the projectile and subsequently a low value of the energy ratio. Indeed, the FE method is not capable to realistically predict the projectile deformation and was not validated in the case of multi-layered target. On the other hand, the SPH method is capable to handle the projectile large deformation and was validated in the case of multilayer target. The SPH method was thus exploited to study different configurations of the double-layered target, increasing the understating of the behavior of double-layered targets, when impacted by soft core projectiles. Finally, conclusions are drawn in section 5.

Section snippets

Experimental tests

High-velocity impact tests were performed on monolithic and multi-layered targets. The projectile used was a 0.308 Winchester, which is an ogival bullet with a brass jacket and a lead core. The diameter is 7.81 mm, the length is 29.71 mm, the mass is 9.6 g and the nominal velocity is 830 m/s (standard muzzle velocity). This bullet corresponds to the FB6 protection level of the European standard EN 1522 [1]. The monolithic targets consisted in Ramor 500 plates, which is a high-strength steel

General model description and material modeling

The numerical models were implemented with the software LS-DYNA. The models were three-dimensional and exploited the double-symmetry of the problem by modelling only one quarter of the geometry and applying suitable boundary conditions on the symmetry planes. As shown in Fig. 5, two different numerical models were developed exploiting both the FE and the SPH method for the discretization of the geometry. Model 1 exploited only the FE method for both the target and the projectile; Model 2

Comparison between Model 1 and Model 2

Model 1–02 and Model 2–17 are compared since they are the best configurations for respectively Model 1 and Model 2. Model 2–17 is more accurate in the prediction of the residual velocity of the projectile for target S030 with an error of + 11%, while the error for Model 1–02 is –14%. The energy balance at the end of the simulation for target S030 is shown in Fig. 13.

The kinetic and internal energy of the projectile and the target are reported in terms of energy belonging to the not eroded part

Conclusion

High-velocity impact tests were performed using a 0.308 Winchester bullet on Ramor 500 plates of different thickness, thus covering the scenarios of complete perforation and arrest of the projectile. Furthermore, high-velocity impact tests were performed on double-layered shields constituted by a high strength steel and an aluminum plate. It was thus possible to experimentally compare the ballistic performance of monolithic and multi-layered targets of the approximately the same weight. It was

CRediT authorship contribution statement

Riccardo Scazzosi: Conceptualization, Methodology, Software, Investigation, Validation, Writing - original draft. Marco Giglio: Project administration, Funding acquisition. Andrea Manes: Conceptualization, Writing - review & editing, Supervision, Project administration.

Declaration of Competing Interest

The authors declared that there is no conflict of interest.

Acknowledgment

The Italian Ministry of Education, University and Research is acknowledged for the support provided through the Project “Department of Excellence LIS4.0 - Lightweight and Smart Structures for Industry 4.0”.

References (52)

  • A. Banerjee et al.

    Numerical simulation of ballistic impact of armour steel plate by typical armour piercing projectile

    Procedia Eng

    (2017)
  • K. Senthil et al.

    Experimental and numerical studies on mild steel plates against 7.62 API projectiles

    Procedia Eng

    (2017)
  • M.A. Iqbal et al.

    Oblique impact on single, layered and spaced mild steel targets by 7.62 AP projectiles

    Int J Impact Eng

    (2017)
  • X. Teng et al.

    Ballistic resistance of double-layered armor plates

    Int J Impact Eng

    (2008)
  • M.A. Iqbal et al.

    Influence of projectile shape and incidence angle on the ballistic limit and failure mechanism of thick steel plates

    Theor Appl Fract Mech

    (2012)
  • A. Gilioli et al.

    Predicting ballistic impact failure of aluminium 6061–T6 with the rate-independent Bao-Wierzbicki fracture model

    Int J Impact Eng

    (2015)
  • M.A. Iqbal et al.

    An investigation of the constitutive behavior of Armox 500T steel and armor piercing incendiary projectile material

    Int J Impact Eng

    (2016)
  • D.E. Tria et al.

    Methodology for experimental verification of steel armour impact modelling

    Int J Impact Eng

    (2017)
  • J.K. Holmen et al.

    Influence of fragmentation on the capacity of aluminum alloy plates subjected to ballistic impact

    Eur. J. Mech. - A/Solids

    (2016)
  • V. Mehra et al.

    High velocity impact of metal sphere on thin metallic plates: A comparative smooth particle hydrodynamics study

    J Comput Phys

    (2006)
  • S.K. Buruchenko et al.

    Smooth Particle Hydrodynamics GPU-Acceleration Tool for Asteroid Fragmentation Simulation

    Procedia Eng

    (2017)
  • N. Kılıç et al.

    Ballistic resistance of high hardness armor steels against 7.62mm armor piercing ammunition

    Mater Des

    (2013)
  • J.K. Holmen et al.

    Influence of yield-surface shape in simulation of ballistic impact

    Int J Impact Eng

    (2017)
  • P.K. Gupta et al.

    Energy dissipation in plastic deformation of thin aluminum targets subjected to projectile impact

    Int J Impact Eng

    (2017)
  • P. Sharma et al.

    Quasi-Brittle Fracture of Aluminium Alloy 2014 under Ballistic Impact

    Procedia Eng

    (2017)
  • CEN, “EN 1522 - Windows, Doors, Shutters and Blinds - Bullet Resistance - Requirements and Classification.”...
  • Cited by (0)

    View full text