Lightweight Stiffening Ribs in Structural Plates.
Friction Stir Processing (FSP) has been used to refine grain structure in sheet metals, and is based on friction stir welding (FSW) principles developed and patented by TWI Ltd, Cambridge, UK in 1991 [1, 6, 7]. In Friction Stir Processing (FSP) a tool generates heat from friction and pressure causing a material to become plastic without melting. The tool then mixes the base material in a circular motion as it traverses laterally through the material (Figure 1).
It is possible to add 2D or 3D Nano particles (Figure 2) to locally alter the material's stiffness (young's modulus). One obtains better material mix when the rotational mixing speed is high relative to the advancement speed. It is possible to process RHA plate up to a depth of 1/4" or 1/2" if processing from both sides.
For example, friction stirred TiB2 particles in cast iron resulted in over 2x hardness and wear resistance by ASTM G35 (Figure 3) [2, 3, 8]. Since FSP is not a forming process, the pattern of the ribs can be any 2D pattern (linear, circular, spiral, etc.). The main focus of the Nano-reinforced FSP is to achieve increased localized stiffness with minimal increase in density of the local material to achieve light weighting. The addition of Nano-particles in addition to grain refinement of the localized microstructure results in improved mechanical properties such as higher yield strength (YS) and elastic modulus 
FSP SIMULATION METHOD
Before assessing the effect of FSP stiffening ribs, it was necessary to establish blast and ballistics performances of the base metals: Aluminum 5182 and 7075 alloys and equivalent weighted Rolled Homogeneous Armor (RHA) steel. Figure 4 shows the 4'x4' blast and ballistics plates used in the simulation. The aluminum 5182 & 7075 plates weighed 282 kg and was 31.8 mm thick. The mass equivalent RHA was 10.9mm thick.
Two FSP stiffening ribs pattern were chosen to analyze the localised stiffness o fibase metalss Al 5182 and 7075. FSP ribs were spaced 4 inch apart with one half inch in thickness. Figure 5 belows shows the two FSP ribs pattern chosen in this study
Simulation matrix & material properties
The material properties for the FSP ribs were estimated as a percent change of the base metal alloy's properties (Table 1). In this particular study, elastic moduluas of an FSP ribs was increased by 100%, YS increased by 50% and density by 10% relative to the base material.
Table 2 shows the matrix of simulation studies conducted. Base metal Al 5182 is the baseline. A mass equivalent amount of RHA was also simulated (equivalent RHA). In order to study which aspect of the nano material had the largest effect, the elastic modulus and yield strength were studied separately. Lastly, in order to gain insight into rib patterns, studies were conducted on ribs in Y direction only, X&Y direction, and completely covering the AL 5182 base material.
Blast simulation was performed by using the well documented Defense Research and Development Canada (DRDC), plate model. The simulation engine was LS-DYNA explicit non-linear solver. Figure 6 shows the simulation set up of the blast model. C4 is used as an explosive material and is buried 2 inches (50 mm) deep inside the soil. Standoff between the plate and the ground surface is at 16 inches as shown. A ballast mass of 10,620 kg is placed on top of the fixture frame to simulate the mass of a light armor vehicle (LAV).
The Arbitrary Lagrange in Euler (ALE) method was used to simulate the blast event. In ALE, a background solid mesh is created and fixed, and material advects through each cell as volume fractions. MAT_HIGH_EXPLOSVE_BURN and EOS_JWL were used to characterize high explosive. High explosive (HE) geometry was defined by using INITIAL_VOLUME_FRACTION_GEOMETRY. HE dimensions used in this study was 254 mm in diameter and 76.2 mm height with density of 1630 kg/[m.sup.3] and detonation velocity of 6930 m/s. Air was represented as MAT NULL with EOS_IDEAL GAS. There are many different soil models available in LS_DYNA, and this study used the MAT_ELASTIC_PLSTIC_HYDRO_SPALL (EPH) with dry density of 1852 kg/[m.sup.3] and bulk modulus (K) of 50.00 GPa.
Base metal plates and FSP ribs were modeled as solid elements with Johnson-Cook strength and failure models and Mie Gruneisen equation of state models (EOS). The LS-DYNA material card with strength and failure parameters for AL 5182 is shown in Figure 7.
Material properties for FSP ribs were derived from the AL5182 model (Figure 8). FSP rib material properties are scaled up from the AL5182 base metal per table 1. Due to limited availability of strain rate effects and damage parameters for FSP ribs, it was decided to use all the strain rate effects, damage parameter of AL5182. Sound speed for FSP ribs were derived from the following relationship.
c = [check] E/[rho] = 6940 m/s
Blast simulation results
Peak vertical displacement of the plate at the center was measured and compared to different FSP ribs patterns from the blast simulation. Figure 9 shows the peak vertical displacement for AL5182 of 287 mm.
Figure 10 shows the pressure distribution of the plate at the center and at 8 inch away from the center. Pressure waves are mostly tensile in and reflective especially away from the center. For the first two milliseconds tensile pressure is dominant and disperses quickly after 2 milliseconds. This can be attributed to the decay of explosive gases immediately after detonation. In general the detonating products of explosives exhibits peak pressures at about 0.5 milliseconds (Figure 11).
Pressure distribution at the center of the plate for the blast simulation is shown in Figure 10. Key takeaway from the displacement plot is that yield strength (YS) has the biggest influence in reducing the peak displacement. Increasing the YS of AL 5182 by 50% reduces the peak vertical displacement of the plate from 287 mm to 235 mm. FSP rib pattern in x directions reduces the peak plate displacement only by 8 mm to 279 mm, where FSP rib pattern in x & y direction reduces the plate vertical displacement from 287 mm to 268 mm. Using FSP over the entire plate produces results equivalent to the 50% increase in yield strength. This means the more FSP is used to modify the plate, i.e., the wider and denser the ribs, the better the performance due to the increase in yield strength. Equivalent RHA plate vertical displacement is much higher than that of AL 5182 at 318 mms. Even though RHA has higher strength, at equivalent mass the plate thickness is much smaller compared to AL 5182. This results in an overall plate that is much weaker and therefore exhibits higher vertical displacement.
Peak displacements of the plate all combinations is shown Figure 12 Figure 13 highlights the pressure distribution on the AL 5182 plate and FSP inserted stiffening ribs for the first 3 milliseconds, where most of the blast load is being absorbed by the plate.
Blast pressures on RHA plate shows higher compressive and tensile pressures as shown in Figure 15 compared to that of AL 5182 and FSP inserted stiffening ribs patterns. This is due to lower thickness of equivalent RHA plate.
Internal energies of the plates are depicted in Figure 15. From Figure 15 it is clear that higher the stiffness of the material lower the internal energy absorption.
Table 3 summarizes the displacements and pressures for the different combinations of AL5182, FSP ribs and equivalent RHA plates. Figure 16 captures the snapshots of deformed plates at different time steps.
Table 4 summarizes the blast results. The peak displacement and energy absorption decreases with the addition of the FSP ribs. This is due to the higher YS of the FSP Nano-material. In fact, the minimum displacement occurs when the FSP Nano-material is mixed throughout  AL 5182 base metal shows lower tensile and compressive pressure, whereas the FSP ribs show high tensile pressure due to localized stiffening.
Next step base plate was replaced from AL5182 to AL7075 with higher yield strength. In this analysis the base plate was modified from a single plate 31.8 mm thick into two sets of layered composites: one with two layers of 15.87 mms and another with 4 layers of 7.9375 mms in a 0/90 and 0/90/0/90 pattern as shown in Figure 17. Johnson Cook strength and damage properties are shown in Figure 18.
Plate center displacements are plotted in Figure 19. The plot clearly shows a reduction in peak displacement from 287 mm to 200 mm for AL7075 due to its significantly higher YS of 548 Mpa compared to AL5182's YS of 250 Mpa.
The ballistic impact performance of AL5182, AL5182 with FSP ribs, and RHA plates were studied. Objective of this study is to understand the ballistics performances of FSP ribs pattern to the base aluminum alloy i.e., AL5182 andAL7075. The same DRDC plates from the blast simulation were cut into small coupon samples, placed vertically, and shot with a mild steel plate proofing projectile for shock testing per MIL-DTL-1225 as shown in Figure 20. This particular simulation set up was successfully correlated to another experimental test. The projectile was 37 mm in diameter and 100 mm in length. Initial velocity of the projectile was set at 414 m/s. The same projectile was used for all experiments. All the simulations and calculation were carried out as Lagrange element formulation, where both nodes and materials move together as one.
Figure 21 above shows the pressure wave distribution of ballistic impact on AL5182 plate. At the center due to bullet penetration pressure wave is zero and at 37 mm away from the center pressure waves are more reflective. At the point of contact shock waves propagates through both the plate and projectile. Pressures and velocities at the contact media must match. One way to estimate the impact conditions is to estimate the impedance matching which is derived from continuity of pressure and velocities across the boundary between the bullet and the target plate. This involves more detailed analysis of shock Hugoniot, beyond the scope of this project. Animated snapshots at different time steps are show in Figure 22. M&S captures very well the dwelling, crater formation and plug on the ballistic plate.
Pressure distribution due to bullet impact and bullet velocities are shown in Figure 23 and Figure 24. The ballistic pressure waves (Figure 25) are much more pronounced than in blast (Figure 13 and Figure 14): the wave lengths are 10 times shorter, and there are many more reflections. It is also interesting to note in the two cases with FSP ribs, the wave inverts almost immediately. This inversion is not seen where the FSP nano-material is distributed throughout (Modulus, Yield Strength & Density scaled). This would imply the pattern and base material boundary have some effect. This is difficult to understand, however, because wave reflection is a function of material density, which only differs by 10%. While this effect has no immediate impact on these results, it is an interesting phenomena to note and should be further investigated.
Ballistic simulations were conducted for AL7075 and FSP ribs pattern derived from AL7075 with 0/90 and 0/90/0/90 layered composite pattern to assess the performances.
Figure 26 shows the projectile kinetic energies and figure 27 projectile velocities. The higher yield strength of AL7075 does stop the projectile completely compared to AL5182. Also note that AL 7075 projectile is stopped and rebounds as shown in figure 27 with positive velocity. In all these ballistics simulation direction of initial velocity of the projectile was in negative X direction. Animated snapshot of deformed projectile and AL7075 plate is captured in Figure 28.
Back of the ballistic plate cracks due to high strength of AL 7075 and stops the projectiles from penetrating through resulting in projectile rebound. Projectile slows down, transfers more kinetic energy to the target plate, expands in diameter and forms a mushroom. These are shown in Figure 28 snapshots.
Table 5 summarizes the ballistic results. All projectiles penetrate the plate. The FSP inserts slow the projectile and absorb more of its kinetic energy due to the insert's increased yield strength.
Summary of projectile kinetic energies, pressures on ballistic plates at 37 mm from center along with velocities are shown in Table 6. It is clear both from AL5182 and AL7075 simulation that yield strength plays a key role in stopping the projectile penetrating through the plate. FSP ribs does show slight benefit in reducing the projectile velocities in both x& pattern.
SUMMARY AND RECOMMENDATIONS
Detailed simulation models were created to evaluate the blast and ballistic performance of FSP reinforced AL 5182 andAL7075. LS-DYNA  nonlinear explicit solver was used to analyze both the blast and the ballistics simulation responses of AL5182, AL7075, RHA and FSP inserted stiffening ribs. The FSP ribs demonstrated effective stiffening. The main benefit was from increased yield strength. For that reason, applying nano-material using the FSP process throughout the base material is preferable to localized ribbing or other stiffness enhancements. However, the ribbing does seem to cause a phase inversion in the ballistic wave. This aspect is not yet well understood.
A recent benefit of the FSP nano-material reported in the literature is increased surface hardness . This, along with the phase changes should be investigated to determine whether FSPNano materials can be utilized in an armor solution against kinetic threats.
This analysis assumed the FSP nano-materials to be isotropic with respect to Johnson-Cook strength and damage properties. However, increased elastic modulus and yield strength typically result in loss of ductility, and the assumption of isotropy may not be applicable in reality. In cases where the material exhibits brittle behavior, the simulation analysis needs to be conducted with brittle material models such as Johnson-Holmquist or Brittle Damage models. In addition, characterizing the high-strain rate properties and EOS variables of such materials should be conducted. It will be beneficial to use these FSP rib inserted materials in non-load bearing components to take advantage of the low density and light weight.
[1.] Mishra R. andMa Z., "Friction Stir Welding and Processing,"Materials Science and Engineering R 50, pp. 1-78, 2005.
[2.] Saumyadeep J., Mishra R. S. andGrant G., Friction Stir Casting Modification for Enhanced Structural Efficiency: A Volume in the Friction Stir Welding and Processing Book Series, Butterworth-Heinemann, 2015.
[3.] Eberhardt J. J., Arbegast B., Stone G., Howard S. andAllen C, "H. Friction Stir Joining and Processing of Advanced Materials Including MMCs.,"Department of Energy, 2005.
[4.] Sudhakar I., Madhu V., Reddy G. Madhusudhan andRao K. Srinivasa, "Enhancement of wear and ballistic resistance of armour grade AA7075 aluminium alloy using friction stir processing," Defence Technology, pp. 11(1):10-17, 2015.
[5.] Komarasamy M., Mishra R. S., Baumann J. A., Grant G. and Hovanski Y., "Processing, Microstructure, and Mechanical Property Correlation oin A1-B4C Surface Composite Produced via Friction Stir Processing," in TiC 440,2013.
[6.] Mahoney M.W., Welding and Joining, (Jan/Febl997):p. 14-16
[7.] Dawes C.J. and Thomas W.M., Welding Journal, 75, (1996):p. 41
[8.] Charit I., Mishra R., and Jata K., TMS Proceedings, Friction Stir Welding and Processing, Nov. 4-8, 2001, pp. 225-234.
[9.] Livermore Software Technology Corp, Livermore CA
For any questions contact the authors Venkatesh Babu at
Dr. Richard Gerth @Richard.firstname.lastname@example.org
FSP - Friction Stir Process
FSW - Friction Stir Welding
RHA - Rolled Homogeneous Armor
PNNL - Pacific Northwest National Laboratory
HE - High Explosive
YS - Yield Strength
LAV - Light Armor Vehicle
DRDC - Defense Research & Development Canada
EOS-Equation of State
US Army TARDEC
Table 1. Material properties. Elastic Modulus Yield Strength Density (Mpa) (Mpa) (kg/m3) Base metal AL 5182 7.14E+04 2.46E+02 2661 FSP Insert 100% of AL 5182 50% of AL 5182 10% of AL 5182 1.43E+05 3.75E+02 2927 RHA 2.10E+05 7.74E+02 7860 Table 2. Simulation matrix for AL 5182 & 7075 and RHA. Thickness (mm) Weight (kg) Base metal (ALS182) 31.8 282 Base metal Elastic Modulus doubled 31.8 282 Base metal Yield Strength increased 50% 31.8 282 Base metal + FSP insert in ? direction 31.8 284 Base metal + FSP insert in X & ? d irection 31.8 286 Base metal replaced by FSP properties 31.8 286 Equivalent RHA 10.9 285 Table 3. Summary of plate displacement & pressures. Peak Thickness Weight displacement (mm) (kg) (mm|) Base metal (AL 5182| 31.8 282 287 Base metal Modulus doubled 31.8 282 287 Base metal Yield strength scaled 50% 31.8 282 231 Modulus, Yield Strength, Density Scaled 31.8 286 220 Base metal + FSP Insert in Y 31.8 284 276 Base metal + FSP in both X & Y 31.8 286 269 Equivalent Weighted RHA 10.9 285 315 Pressure @ 8 inch from center (Spa) Base metal (AL 5182| 0.2405 Base metal Modulus doubled 0.2405 Base metal Yield strength scaled 50% 0.3339 Modulus, Yield Strength, Density Scaled 0.3465 Base metal + FSP Insert in Y 0.3073 Base metal + FSP in both X & Y 0.3282 Equivalent Weighted RHA 0.5195 Table 4. Summary of Blast Simulation Results Case Peak Pressure @ Energy (MJ) Displacement 8" from (mm) center (GPa) Base Al 5182 287 0.241 1.42 AL 5182, FSP in Y 276 0.307 1.39 AL 5182, FSP in 269 0.328 1.37 X&Y AL 5182, FSP 220 0.347 1.31 throughout Base RHA 315 0.520 1.31 Table 5. Summary of Ballistic Simulation Results Velocity Pressure @ 37 Projectile (m/s) mm from KE center (GPa) (Joules) Base Al 5182 279 0.17 34,117 AL 5182, FSP in Y 260 0.20 29,126 AL 5182, FSP in 240 0.38 24,766 X&Y AL 5182, FSP 169 0.35 12,071 throughout Base RHA 175 0.22 13,201 Table 6. Summary of Ballistic Simulation of all combinations Pressure @ Thickness Velocity 37 mm from (mm) (m/s) center (Gpa) Base metal ( AL 5132) 31.8 279 0.17 Base metal Modulus doubled 31.8 279 0.17 Base metal Yield strength scaled 50% 31 8 170 0.34 Modulus, Yield Strength, Density Scaled 31.8 169 0.35 Base metal + FSP insert In Y 31.8 260 0.20 Base metal + FSP in both X & Y 31 8 240 0.38 Equivalent Weighted RHA 10.9 175 0.22 Projectile Kinetic Energy (Joules) Base metal ( AL 5132) 34,117 Base metal Modulus doubled 34,117 Base metal Yield strength scaled 50% 12,160 Modulus, Yield Strength, Density Scaled 12,071 Base metal + FSP insert In Y 29,126 Base metal + FSP in both X & Y 24,766 Equivalent Weighted RHA 13,201
|Printer friendly Cite/link Email Feedback|
|Author:||Babu, Venkatesh; Gerth, Richard|
|Publication:||SAE International Journal of Commercial Vehicles|
|Date:||Oct 1, 2017|
|Previous Article:||Near Automatic Translation of Autonomie-Based Power Train Architectures for Multi-Physics Simulations Using High Performance Computing.|
|Next Article:||Simulating the Mobility of Wheeled Ground Vehicles with Mercury.|