一種模擬旋轉(zhuǎn)拋射實(shí)驗(yàn)裝置的結(jié)構(gòu)設(shè)計(jì)

一種模擬旋轉(zhuǎn)拋射實(shí)驗(yàn)裝置的結(jié)構(gòu)設(shè)計(jì),一種模擬旋轉(zhuǎn)拋射實(shí)驗(yàn)裝置的結(jié)構(gòu)設(shè)計(jì),一種,模擬,摹擬,旋轉(zhuǎn),拋射,實(shí)驗(yàn),試驗(yàn),裝置,結(jié)構(gòu)設(shè)計(jì) DEFENSE TECHNICAL INFORMATION CENTER InforttLrfUHifor the tkhtut Community DTIC has determined on Jo / f) lr)rJcP that this Technical Document has the Distribution Statement checked below. The current distribution for this document can be found in the DTIC Technical Report Database. S( DISTRIBUTION STATEMENT A. Approved for public release; distribution is unlimited. COPYRIGHTED; U.S. Government or Federal Rights License. All other rights and uses except those permitted by copyright law are reserved by the copyright owner. DISTRIBUTION STATEMENT B. Distribution authorized to U.S. Government agencies only (fill in reason) (date of determination). Other requests for this document shall be referred to (insert controlling DoD office) DISTRIBUTION STATEMENT C. Distribution authorized to U.S. Government Agencies and their contractors (fill in reason) (date of determination). Other requests for this document shall be referred to (insert controlling DoD office) DISTRIBUTION STATEMENT D. Distribution authorized to the Department of Defense and U.S. DoD contractors only (fill in reason) (date of determination). Other requests shall be referred to (insert controlling DoD office). DISTRIBUTION STATEMENT E. Distribution authorized to DoD Components only (fill in reason) (date of determination). Other requests shall be referred to (insert controlling DoD office). DISTRIBUTION STATEMENT F. Further dissemination only as directed by (inserting controlling DoD office) (date of determination) or higher DoD authority. Distribution Statement F is also used when a document does not contain a distribution statement and no distribution statement can be determined. DISTRIBUTION STATEMENT X. Distribution authorized to U.S. Government Agencies and private individuals or enterprises eligible to obtain export-controlled technical data in accordance with DoDD 5230.25; (date of determination). DoD Controlling Office is (insert controlling DoD office). Avtyendoc K ELSEVIER Available online at * ScienceDirect International Journal of Impact Engineering 35 (2008) 920-936 INTERNATIONAL JOURNAL OF IMPACT ENGINEERING locate ijimpeng Impact response of sandwich plates with a pyramidal lattice core Christian J. Yungwirtha, Haydn N.G. Wadleya, John H. OConnor5, Alan J. Zakraysekb, Vikram S. Deshpande0* Department of Material Science accepted 16 July 2007 Available online 20 July 2007 Abstract The ballistic performance edge clamped 304 stainless-steel sandwich panels has been measured by impacting the plates at mid-span with a spherical steel projectile whose impact velocity ranged from 250 to 1300ms-. The sandwich plates comprised two identical face sheets and a pyramidal truss core: the diameter of the impacting spherical projectile was approximately half the 25 mm truss core cell size. The ballistic behavior has been compared with monolithic 304 stainless-steel plates of approximately equal areal mass and with high- strength aluminum alloy (6061-T6) sandwich panels of identical geometry. The ballistic performance is quantified in terms of the entry and exit projectile velocities while high-speed photography is used to investigate the dynamic deformation and failure mechanisms. The stainless-steel sandwich panels were found to have a much higher ballistic resistance than the 6061-T6 aluminum alloy panels on a per volume basis but the ballistic energy absorption of the aluminum structures was slightly higher on a per unit mass basis. The ballistic performance of the monolithic and sandwich panels is almost identical though the failure mechanics of these two types of structures are rather different. At high impact velocities, the monolithic plates fail by ductile hole enlargement. By contrast, only the proximal face sheet of the sandwich plate undergoes this type of failure. The distal face sheet fails by a petalling mode over the entire velocity range investigated here. Given the substantially higher blast resistance of sandwich plates compared to monolithic plates of equal mass, we conclude that sandwich plates display a potential to outperform monolithic plates in multi-functional applications that combine blast resistance and ballistic performance. ) 2007 Elsevier Ltd. All rights reserved. Keywords: Impact; Pyramidal truss; Air shock waves; Energy absorption; Ballistic performance 1. Introduction It is well known that sandwich plates possess a superior bending stiffness and strength to monolithic beams of the same mass under quasi-static loading. Theoretical studies by Fleck and Deshpande 1 and Xue and Hutchinson 2 also predicted that sandwich beams have superior shock resistance to monolithic beams. Subsequently, several experimental studies 3 5 have confirmed these initial theoretical and numerical predictions. Typically, blast events in air are accompanied by high-velocity fragments, but little is known about the ballistic resistance of these sandwich beams. Here we present an experimental in- *Corrcsponding author. Tel.: +44 1223 332664; fax: +44 1223 332662. E-mail address: vsd eng.cam.ac.uk (V.S. Deshpande). 0734-743X/S - see front matter c 2007 Elsevier Ltd. All rights reserve doi:10.1016j.ijimpcng.2007.07.001 vestigation that compares the ballistic performance of sandwich and monolithic plates of equal areal mass. Over the last decade, a number of new core topologies for sandwich panels have emerged. These include metallic foams 6, truss-like lattice materials (Fig. 1), prismatic sandwich cores such as the corrugated and Y-frame cores 7 and various honeycomb cores. The truss-like cores due to their open-cell architecture are ideally suited for multi- functional applications that include combined thermal and structural functionality. In this study, we focus on sandwich plates with the pyramidal truss core. Sandwich panel structures investigated can be thought of as a pair of thin metal plates separated by a lattice of slender trusses. The penetration of thin ductile plates (those where the plate thickness is small compared with the projectile diameter) by spherical tipped projectiles can 20090602244 C.J. Yungwirth et al. / International Journal of Impact Enaineerina 35 (2008) 920-936 b _ c 921 Fig. 1. Examples of truss-like lattice structures configured as the cores of sandwich panel structures: (a) pyramidal, (b) tctrahcdral and (c) double layer Kagomc lattice trusses. occur by either petalling or adiabatic shearing (plugging), sometimes in combination with ductile hole enlargement at high impact velocities 8. Petalling usually occurs in soft materials with high work hardening rates penetrated by low-velocity projectiles. It begins with a dishing deforma- tion of the plate which develops high circumferential strains beneath the impact location. These high strains lead to radial tensile stretching and fractures of the metal plate leading to the formation of typically 4-7 petals that bend away from the incoming projectile. Energy is dissipated by the global dishing and tearing of the plate along with the bending of the petals 9.10. High-velocity penetration of high dynamic strength and low work hardening rate metal plates occurs by shear banding. The shear failure is confined to a thin cylindrical sheath beneath the edge of the projectile 11. If the rise in temperature within the band causes more local softening than the increase in flow stress due to strain and strain-rate hardening, adiabatic shear bands form 12. The metal plug is usually thinner than the original plate thickness because of radial metal flow from beneath the projectile. The penetration of a metal sheet by a normal incidence projectile has been widely studied; see 8,13 for a recent review of the literature. Experimental studies by Almo- handes et al. 14 indicated that distributing the mass of a plate between a pair of identical plates resulted in a lowering of the ballistic resistance of the system compared to a monolithic structure of equal areal mass. However, theoretical studies by Ben-Dor et al. 15 and experimental studies by Radin and Goldsmith 16 indicate little effect. Relatively few experimental studies have investigated the ballistic resistance of metallic sandwich plates. A study by Goldsmith et al. 17 concentrated on aluminum panels with honeycomb cores while Zhao et al. 18 have investigated the perforation of aluminum foam core panels. The ballistic performance of sandwich plates compared to monolithic plates of equal areal mass is as yet not clearly understood and the role of the parent material of the sandwich plates has not been clearly elucidated. The outline of this paper is as follows. First we briefly review the use of penetration mechanism maps to explain the role of projectile mass and velocity on the failure mechanisms of ductile plates and to illustrate the effect of material layering on the ballistic performance. Second, the fabrication of stainless-steel and aluminum sandwich plates with a pyramidal truss core is described along with the ballistic testing procedure. We then summarize the measured ballistic performance of the sandwich and monolithic plates and use high-speed photographs to elucidate the deformation and failure mechanisms. We finally conclude the study by showing that stainless sandwich panels have a similar ballistic performance to monolithic plates of equal areal density even though the mechanisms of penetration are different. 2. Penetration mechanism maps Deshpande et al. 19 have recently developed penetra- tion mechanism maps in order to elucidate the coupling between projectile mass and velocity in determining the ballistic performance of clamped beams impacted by rigid projectiles. Here we briefly review their findings in order to (i) put the experimental observations of this study in context and (ii) help explain some of the observations discussed in Section 4. Consider a clamped beam of span 2L and thickness /; made from a rigid ideally plastic solid of dynamic yield strength (TY and density p. This beam is impacted at mid- span by a rigid projectile of mass G per unit thickness perpendicular to the plane of the beam. Deshpande et al. 19 considered the two critical failure modes: (a) tensile tearing and (b) shear-off. Regimes of dominance of the two failure regimes for an assumed beam material tensile failure strain Ef = 10% and a critical normalized shear displace- ment wcx/h 1.0 are illustrated in the map in Fig. 2a using axes of the non-dimensional projectile impact velocity vp = vp/a/p and non-dimensional projectile mass G = G/(ph). The shear-off regime is essentially insensitive to the beam aspect ratio L/h. However, tensile failure is strongly dependent on the value of L/h: with increasing L/h the regime over which tensile failure is the dominant failure mechanism shrinks and the tensile failure regime marked in Fig. 2a is only valid for the choice L/h = 5. The key features of the failure mechanism map are (i) The critical penetration velocities (or the ballistic limit) decrease with increasing G as both the shear-off and tensile failure modes become more likely. (ii) For the choices of material failure parameters made here, tensile failure is not an operative failure mechanism for G- I 3 o E o 1 1.5 i i i y IX) Tearing / / / 7/ - 03 Xi A- *- Shear-off 0.0 / G/l2 S 4 i 2 i 0.0 0.5 1.0 1.5 2.0 Normalized impact velocity v Vyp Fig. 2. (a) Failure mechanism map for the central impact of a rigid projectile against a clamped beam with an aspect ratio L/h = 5. The map is plotted using axes of the normalized projectile mass G and normalized projectile velocity fp. The beam material is assumed to have a tensile failure strain 6f = 10% and critical normalized shear displacement wf/h = 1.0. (b) Predictions of the corresponding normalized projectile residual or exit velocity fr as a function (iii) For a given value of G8, the failure mechanism transitions from shear-off to tensile failure with decreasing projectile velocity vp. The corresponding predictions of the projectile residual or exit velocity vT = IY/VY/P as a function of the impact velocity tJp are shown in Fig. 2b for four selected values of the normalized projectile mass G and the material failure parameters employed for constructing the map in Fig. 2a. The predicted failure mechanisms are indicated on these residual velocity curves and consistent with the map in Fig. 2a, tensile failure is only operational for high values of G at low impact velocities. Note that when shear-off is the failure mode, a sharp increase in the residual velocity is predicted at the critical penetration velocity. This is rationalized as follows. Just below the critical penetration velocity, the shearing in the beam under the projectile arrests just prior to the shear failure of the beam. The projectile at this instant has some residual velocity and this kinetic energy of the projectile is absorbed by the stretching and bending of the beam. At a projectile velocity just above the critical penetration velocity, shear failure of the beam prevents this additional energy absorption mechanism and the projectile penetrates with a significant proportion of the initial kinetic energy still not being dissipated. This gives the sharp increase in vT just above the ballistic limit. These maps illustrate the effect of the beam material density (and/or thickness and material density) on the ballistic performance of plates. Consider a beam of fixed aspect ratio L/h and areal mass m ph. In order to understand the effect of p, all other parameters are held fixed including the yield strength rjy, material failure parameters (f and vf/h), projectile mass G,and impact velocity t;p. The y-axis of the map in Fig. 2a can then be interpreted as G = Gp/m2. Thus, reducing the material density implies that both the normalized mass G = Gp/m2 and the normalized impact velocity vp are reduced: this would in effect increase the ballistic performance of the beam. A practical realization of such a scenario would be the substitution of high-strength aluminum for a carbon steel beam material. Both these materials have approximately the same yield strength and failure proper- ties but the density of aluminum is about one-third the density of steel. 2.1. Effect of material layering The main focus of this article is to investigate the ballistic resistance of sandwich panels. To motivate this we contrast the ballistic performance of the two systems illustrated in Fig. 3 and made from a material of density p and yield strength rY: (0 a monolithic clamped beam of thickness /; and span 2L and (ii) a sandwich-like configuration comprising two identical but independent monolithic clamped beams each with a span 2L but thickness h/2. Thus, both the systems have equal mass and span and differ only in the way mass is spatially distributed. A C.J. Yungwirth el al. / International Journal of Impact Engineering 35 (2008) 920-936 923 II ? 1 Two beams or thickness (V2 Single beam of thickness h (same total mass) o.o 0.5 1.0 Normalized impact velocity vp- 20 5 I 2L- h/2 Fig. 3. Sketches of the monolithic and bilaycr systems of equal mass and predictions comparing the ballistic performance (in terms of the projectile residual versus entry velocity) of the monolithic and bilaycr systems. The assumed material failure parameters arc f = 10% and t*|;nt/i = 1.0 with predictions shown for selected values of G. projectile of mass G impacts these two configurations at mid-span as shown in Fig. 3. Predictions of the normalized residual velocity vt versus impact velocity vp for the two systems are included in Fig. 3 for three selected values of the normalized projectile mass G = G/(ph). The material failure parameters are the same as those used above, i.e., f = 10% and the critical shearing displacements wTn/h= 1.0 and wcsnt/(h/2) = 1.0 for the monolithic and sandwich-like configurations, respectively. The predictions indicate that the ballistic performance of both the configurations is very similar with the monolithic beam performing slightly better than the sandwich-like configuration. 3. Experimental investigation Clamped sandwich plates with a pyramidal truss core were impacted normally (zero obliquity) and centrally with spherical steel balls. The aims of the experimental investigation are as follows. (i) To compare the ballistic performance of sandwich and monolithic plates of equal mass. (ii) To investigate the effect of material choice on the ballistic performance of structures. Here we compare the ballistic performance of sandwich plates made from 304 stainless steel and a high-strength aluminum alloy. (iii) To elucidate the mechanisms of failure and penetra- tion in the sandwich plates and monolithic plates. 3.1. Construction of lattice core sandwich plates Sandwich plates with a pyramidal truss core were manufactured from 304 stainless steel and an age-hardened 6061-T6 aluminum alloy with densities p = 8000 and 2700 kg m respectively. The sandwich plates comprised two identical face sheets of thickness h = 1.5 mm and a pyramidal core of thickness c = 25.4 mm; see Fig. 4 for detailed dimensions of the sandwich plates. The pyramidal cores had a relative density (ratio of the effective density of the smeared-out core to the density of the solid material from which it is made) p = 2.6% which implies that the net areal mass m = (2/i + cp)p of the 304 stainless-steel and aluminum alloy sandwich plates was 29.3 and 9.88 kg m-2, respectively. The pyramidal lattice cores comprised struts of length 31.75 mm and cross-section 1.9 x 1.9 mm as shown in Fig. 4. The cores were manufactured from 1.9-mm-thick sheets by first punching rhomboidal holes to obtain a perforated sheet and then folding this sheet node row by node row to obtain regular pyramids as shown in Fig. 5. The sandwich plates were then assembled by laser welding rectangular sheets of dimensions 120.7 x 127 x 1.5 mm3 to pyramidal core truss panels comprising 3x3 cells (Fig. 4). Unlike the 304 stainless steel which could be cut, folded and welded in its as-received state, the 6061-T6 aluminum alloy sheets were annealed to the O-condition prior to the perforation and folding operation. The pyramidal Al 6061 trusses were then solution treated and aged to return them to their peak strength condition (T6 condition) and then laser welded to the 6061-T6 aluminum face sheets. 934 C.J. Yungwirth et al. / International Journal of Impact Engineering 35 (2008) 920-936 PreassemWy Assembly Laser oeam spot welding Fig. 4. Illustration of the laser welding process for bonding the pyramidal truss lattice to sandwich plate face sheets. Details of the dimensions of the sandwich plate and the core arc included in the figure. All dimensions arc in mm. Elongated diamond perforated sheet Perforation punch Roll steel Pyramidal lattice structure Waste Fig. 5. Sketch of the punching and folding operation to manufacture the pyramidal truss lattice core. 3.2. Mechanical properties of the parent materials Tensile specimens of dog-bone geometry were cut from each of the as-received steel and aluminum sheets. The uniaxial tensile responses of the 304 stainless-steel and Al 6061-T6 alloys at an applied strain rate of 103s are plotted in Fig. 6 using axes of true stress and logarithmic strain. We note that the key difference between the two alloys is their strain hardening capacity: while the 304 stainless steel displays a linear hardening post-yield response with a tangent modulus Et 1 GPa. the aluminum alloy exhibits an almost ideally plastic response. This enhanced strain hardening capacity of the stainless steel also stabilizes the tensile specimens against necking, resulting in the higher tensile ductility of the stainless steel compared to the 6061-T6 aluminum alloy. 3.3. Ballistic test methodology The ballistic performance of the 304 stainless-steel and aluminum alloy sandwich plates was investigated for projectile impact velocities in the range C.J. Yungwirth el ah / International Journal of Impact Engineering 35 (2008) 920-