Design of an add-on ceramic composite armour against 14.5 × 114 mm API/B32 projectile for the armoured vehicles and investigation of the ballistic performance of the armour

A ceramic/composite add-on armour system with innovative ceramic geometry (cylindrical) against 14.5 × 114 mm API/B32 projectile was developed and ballistic performance of the armour was investigated both experimentally and numerically. Numerical analysis was used to calculate exit velocities of the projectile after passing through the ceramic/composite layer (before penetrating the Armox 500T which simulates hull structure of an armoured vehicle) and also contributed to the selection of optimum ceramic thickness. The calculated projectile velocity-time curves (from numerical analysis) for three different ceramic thicknesses are given comparatively in the study. The curve characteristics are the same for three different analyses. The duration of the total absorption of the projectile energy is about 0.2 microseconds (ms). There were differences in the transmission of the stress wave and the delamination in the Ultra-High-Molecular-Weight Polyethylene (UHMWPE) layers differed as ceramic thickness increases. The separation between the layers varied with the change in projectile energy. As a result of the ballistic test, the armour prevented 14.5 × 114 mm API/B32 ammunition with desired damage mechanisms. In the x-ray image taken after the shootings, it was seen that the ceramic damage was local which enhanced multi-hit resistance capability and the geometry of the cylindrical alumina played an important role in the localization of the ceramic zone damage during the projectile penetration process. Due to this cylindrical ceramic geometry, the projectile moving on after the moment of impact constantly encounters a curved and new surface, and thus it is deflected and exposed to more wear. The areal density of the armour was also reduced by using the UHMWPE (which is one of the composite material whose fibres have the lowest density and good mechanical properties) composite plate as the backing plate.


Armox 500T
Armor steel brand with a hardness of 500 HBW

Ti6Al4V
An alpha-beta titanium alloy with a high specific strength and excellent corrosion resistance

Introduction
The concept of ballistic protection describes the struggle between materials science and ballistic science between threat and protection level dating back to ancient times.However, from a global point-of-view armed conflicts have increased in the last 25 years intensively.Moreover, the continuous increase in diversity and severity of the threads have led to a debate on the relevance and validity of current ballistic protection standards [1].This situation requires re-evaluation of ammunition types, ammunition impact energies and velocities, and impact angles defined in these ballistic protection standards.Additionally, non-state actors recently have easy access to advanced firearms and ammunition in the inventories of both the NATO and the former Warsaw Pact states.In addition, the exchange of arms and ammunition and technology transfer between terrorist organizations which are active in the Middle East region have expanded the threat spectrum [2].Currently, the most advanced weapons and ammunition, which were previously only available to regular armies, can be easily procured by terrorist organizations and/or crime groups and used outside conventional battlefields; moreover, it is observed that terrorist organizations are able to manufacture several type of ammunition and weapons, indeed [3].These operational environments with highly variable threat levels increase the need for modular ballistic protection of armoured vehicles against high ammunition threat levels.Likewise, the threat of snipers has increased against infantry in residential areas and also there are the cases where sniper rifles use 12.7 × 99 mm, 12.7 × 108 mm and 14.5 × 114 mm diameter ammunition with armour-piercing (AP) bullet cores even with Tungsten Carbide AP core [4].This shooting capacity of snipers with their high caliber and effective bullet cores enables them to make pinpoint shots, especially on armoured vehicles.This situation increases the need for a more localized damage performance and a better multi-hit resistance capability from armour.
In armed conflicts of classical warfare theory, the lines of positions are well defined, and the distance between them can be up to 400 meters.On the contrary, in urban conflicts or asymmetric combats, there are no positions in the form of a single line, nor do the clashes take place on a single plane, and the distance between them varies between 5 meters and 50 meters, which changes the nature of the threat.This means that armoured vehicles can be exposed to high ammunition threats asymmetrically and more frequently [5].Close-range engagements increase the number of bullet cores and shrapnel hitting the target, as well as the amount of kinetic energy at the moment of impact, requiring an increase in the protection limits of ballistic protective materials [6].Ballistic protection requirements have changed in recent years due to the increasing threat level [7][8][9][10].In that sense, armoured combat vehicles need to be protected in the different engagement areas against high-level threats such as 12.7 mm, 14.5 mm, 25 mm etc projectiles.By providing the required protection level, they are also designed according to the limited weight parameter [8].The total weight of an armoured vehicle must remain under this weight limit and meet the required driving performance and fuel economy requirements.Therefore, armour systems should not only provide adequate ballistic protection but also contain lightweight materials [9,11,12] and armour solutions.This makes the selection of the backing plate for lightness important in hybrid ceramic armor system designs.Armour steels can provide sufficient ballistic protection against high-level threats, but this approach highly increases the total weight of the vehicles [13,14].Thus, ceramic based lightweight armour systems come to the forefront for effective ballistic protection in many cases.Ceramic materials, in addition to low density, have other good properties used in armour applications, such as high hardness, compressive strength and multi-hit resistance that can be changed (with ceramic geometry).Due to this facts, many comprehensive studies have been performed on the ceramic materials for ballistic implementations [15][16][17][18][19][20][21][22][23][24][25].
Ceramic materials are used as hard layer (in the strike face) in the armour systems while having a major objective of decelerating the coming projectile by blunting and eroding it and distributing the load over a large area of the composite.A softer backing plate bonded to ceramic layer also absorbs the residual kinetic energy of the projectile and arrests the fragments [7,[26][27][28][29][30][31][32][33][34].Moreover, Al 2 O 3 (Alumina) ceramics are widely used ceramic materials in composite armours as they offer cost-effective solutions when compared to other armours and ceramic-based armour systems [34,35].As this point it is also worth to note that under high stress level ceramics may change their fracture character to a more ductile fashion which is the general idea behind modeling them with viscosity-embedded constitutive formalism [36,37].In particular, beyond a compressive pressure limit, armor ceramics posses a ductile behivour which definitely enhances their energy absorbing limit up to fracture which could also be termed as confinement-induced brittle to ductile transition [38].High caliber ammunition can impact the target at a speed of approximately 900-1000 m s −1 .Strength against variable high-speed dynamic loads is a very important criterion for material selection in armour solutions.
As the more lightweight ballistic protection materials are needed for high-level threats, innovative approaches are necessary for any further improvements.One of the ways for this is the use of ceramic tiles with different topology.In their study, Wang et al [34] used cylindrical shaped Alumina tiles with Ultrahigh-Molecular-Weight Polyethylene (UHMWPE) and two layers of Ti6Al4V (TC4) in the hybrid composite armour against 12.7 mm AP projectile.In this prestigous work, Wang et al [34] provided the significant data regarding design details of the proposed composite armour design however, this work is limited in such a way that, they inspected a unique projectile velocity with a single-hit application (multi-hit performance is not investigated).This work is also stands solely on an experimental basis, i.e., any finite element analysis based numerical analysis was not involved.Similarly, Liu et al [31] proposed a ceramic composite armour system, which consists of ceramic cylinders and the composite backing plate of Ti6Al4V/UHMWPE/Ti6Al4V against 12.7 mm AP projectile.However, in this paper, an innovative lightweight add-on composite armour was proposed with the design details.In particular, cylindrical shaped Alumina tiles were used in the strike face whose geometric details was given and backing plate was composed of UHMWPE layers.Finally, an armour steel (Armox 500T), whose goal is to simulate hull structure of an armoured vehicle was installed to the composite add-on armour.Finite element analysis based numerical analysis was also performed for three armour systems containing ceramics of various thicknesses and ballistic performance was examined.
In the controlled ballistic test setup, armour system was tested and verified against 14.5 × 114 mm API/B32 projectile.This contribution differs from the previous studies by utilizing a higher level kinetic energy ammunition.This study has also a distinguishing character such that the system design details of a proposed armor system which provides a level-4 protection level conforming to NATO STANAG 4569 standard is disclosed in a detailed manner.Furthermore, the provided areal density of the proposed ballistic solution is compared with the possible other alternatives which brings out the high performance of the designed solution.Similar studies in the literature generally include stand-alone armour researchs.Another great advantage of this study is that the armour presented in this paper was designed in a modular (add-on) structure.Modular armour can be easily removed from the vehicle after damage and new armour can be quickly installed.In addition, it is evaluated that the armour system in the study will make a great contribution to the evaluation of the damage mechanism on an armoured vehicle by using Armox 500T steel armour in a way that simulates the vehicle body.

Numerical analysis
The aim of the numerical analysis study is to understand how the armour system (consisting of fiberglass sheet, ceramic and composite layers) affects the projectile velocity depending on different ceramic thickness parameters and to investigate the damage situation in Armox 500T steel, which is widely used solution in combat vehicles.The armour system structure was modelled with LS-DYNA software and the LS-DYNA Explicit solver was used as the solver.The mesh structure in the analysis consists of 3-D hexahedron elements.The Johnson-Holmquist [39] structural model was chosen to characterize the alumina material in the numerical analysis.This plasticity damage model is a useful model for modelling ceramics, glass and other brittle materials.The model consists of three main components.These components are a representation of solid and fractured material strength in the form of a pressure-dependent yield surface, a failure model that gradually transitions the material from a solid state to a fractured state, and an equation of state for characterizing the pressure-density relationship.In LS-DYNA, this model is used via the 'MAT_JOHNSON_HOLMQUIST_CERAMICS' board.
Material strength is written as in equation (1) in terms of equivalent normalized stress; The superscript ' * ' denotes the normalized value.A, B, C, N and M are the material parameters.P * and T * are the normalized pressure normalized by the Hugoniot elastic limit and the normalized maximum tensile stress that the material can withstand.SFMAX is the maximum fracture strength.Damage parameter, D is written as in equation (4); where e p is effective plastic strain and e f p refers to fracture plastic strain.The plastic strain to fracture under a constant pressure is defined as: where D 1 and D 2 are the damage constants.The pressure-density correlation is written as in equation ( 6) and equation (7): are the material constants.r is current density, r 0 is reference density.The additional pressure rise ΔP, models the expansion when the material is damaged and it is written as in equation (8); where β is the rate of elastic energy loss converted to potential hydrostatic energy.ΔU is an increase in elastic energy loss.
The steel projectile and armour steel were modelled using Johnson-Cook flow stress model [40].This model is generally used to study the material behavior of metallic materials subjected to high strain rate during high velocity impact [41].As the level of detail of the material model increases, so does the reality of the results.In addition, material behavior at high velocities is not based on the known theory of plasticity.Impact and penetration problems of metals at these speeds depend on several material behaviors.It has been observed by Johnson and Cook [42] that when the three important material behavior, hardening, strain rate and thermal softening effects are used in a single relationship, the behaviour of metal materials at medium and high speeds gives a realistic response in their simulations.In LS-DYNA, this model is used via the 'MAT_JOHNSON_COOK board'.The Johnson-Cook material model is written as in equation ( 9 and T, is the temperature at which the test is performed, T r , is the room temperature and T m is the melting temperature. In equation ( 9), the first bracket calculates the plastic strain effect on the yield stress, the second bracket calculates the effect of strain rate, and the third bracket calculates the temperature effect.The constants A, B, n, C, m are determine these effects.The Johnson-Cook fracture model based on total damage is used to model the fracture behaviour of the material.It is included in the Johnson-Cook fracture model in LS-DYNA [41].It is written as in equation (10); 5 are the material parameters, P is the mean stress, σ eff is the effective (equivalent) stress.When D = 1, fracture occurs.In equation (10), the first bracket defines the stress in the three axes, the second bracket defines the strain rate, and the third bracket defines the temperature effect.
In the analysis, the solid model was processed using different solid modelling programs and a digital mesh was built on this solid model.Element size of 0.25-1 mm was used for all layers in the local areas where the     The material model parameters for ceramic layer, projectile and armour steel are given in tables 1-3, respectively.Plastic fracture strains for ceramics and composites are defined as 1.5 and 0.5, respectively.This value is accepted as 1 for projectile (100Cr6) and Armox 500T.Since damage or perforation may occur in the armour system, the contact type 'CONTACT_ERODING_SURFACE_TO_SURFACE' was used in the numerical analysis.The contact algorithm 'AUTOMATIC_SURFACE_TO_SURFACE_TIEBREAK' was selected to describe the delamination between the composite layers.In ballistic analysis, the damage criterion in this contact type provides additional advantage by removing damaged elements from the calculation process.Composite material model for UHMWPE with failure criteria for solid element was chosen to model the behavior of homogenized sublaminates, which was the material type #59 within LS-DYNA.The material parameters for composite laminates both UHMWPE and fiberglass are given in the tables 4 and 5.

Experimental study 3.1. Add-on armour configuration
Add-on armour panels are modular structures and they provide additional ballistic protection to vehicle or system base armour such as armour steel, aluminium alloys (Al 5083-H131, Al 7039-T64 etc).Desired ballistic protection is provided with both add-on armour panel and base armour of the system.The add-on armour system used in our study is shown schematically in figure 5, dimensions of the armour panel and armour steel are given in figure 6 and demonstration of the use of the armor panel on the vehicle is given figure 7.There are four layers in the composite armour and an armour steel, which simulates vehicle hull structure, is mounted to the armour panel.While there are cylindrical alumina ceramics on the front face of the armour panel, UHMWPE composite plate is used as the support plate.Ceramic tiles and backing plate are bonded with a special method using polyurethane-based adhesive.The armour plate in the study is coated with a high performance pure polyurea spray elastomer system (LINE-X XS-350) for structural integrity and multi-hit impact resistance of the armour.Armox 500T steel is attached to the armour panel with high-strength bolts to simulate the vehicle structural body.
Ballistic performance can be improved by optimizing the ballistic impact angle in armour design.Even a few degrees of angle between the direction of movement and the axis of rotation of the AP projectile reduces the  penetrating ability of the projectile and shatters the projectile [51].When cylindrical alumina ceramic (with a smooth curvature on the top surface) is used in the armour design, the ammunition takes more damage and loses more energy.This is associated with the oblique impact of the ammunition.Geometrically, the impact of the ammunition on the cylindrical surface will always be oblique, except for the strike at exactly right angles to the surface protruding point.The probability of hitting the surface projection point at a right angle can be statistically assumed to be zero.Oblique impact has the effect of creating large bending stresses on the ammunition and shattering the ammunition [52,53].
3.2.Materials 3.2.1.Projectile 14.5 × 114 mm API/B32 projectile is used in the ballistic test.The images and the technical specifications of the projectile are given in figure 8 and in table 6, respectively.
The strike face of the armour consists of cylindrical alumina tiles.Alumina ceramics are produced by unpressurized sintering production method.It has 98% purity.The porosity value is 0.5% (% by volume).The material properties of the alumina ceramic are given in table 7.

Adhesive
The polyurethane and epoxy-based adhesives are the two commonly used bonding materials [62].Alumina ceramic tiles and backing plate were bonded with ultra high-strenght, hand mix, 100% pure polyuria elastomer with extended pot life by a special method.

UHMWPE backing plate
UHMWPE is a fiber which has high impact damage resistance [63].The use of this fiber in light armour panels (generally as backing plate) and spall liners are the prominent armour application areas.Multi-hit capability, heat resistance and low flammability are other crucial armour features of the fiber.UHMWPE was used in the backing plate of the armour panel in this study and its material properties are given table 4.

Coating
Coating is used to encapsulate the armour system in order to maintain structural integrity and increase multi-hit resistance.Armour system in the study is coated for the structural integrity of the armour and multi-hit impact resistance.It also provides protection against environmental influences.
When the high-speed penetrator hits the ceramic face of the armour, ceramic tiles start to scatter in various directions.Coating plays important role in minimizing scattering of the ceramics thus increasing ballistic performance.The coating (LINE-X XS-350) s a two-component, high performance aromatic pure polyurea spray elastomer system, zero VOCs (Volatile Organic Compounds), 100% solids.LINE-X XS-350 offers elastomeric protective coatings for various substrates.LINE-X XS-350 is designed for moisture insensitive applications because of its pure polyurea chemistry.It produces skin formation for chemical resistance and moisture protection [64].

Armour steel
Armox 500T which simulates an hull body of an armoured vehicle is installed to armour panel with bolts for the integrity of the proposed ballistic solution.The chemical composition and the mechanical properties of this armour steel is given in table 8 are given in table 9.

Ballistic test
The aim of the ballistic tests was to investigate ballistic protection ability and behavior of the armour system with real ammunitions.In the tests, firstly add-on armour panel was tested for its damage behavior and its structural integrity after firing.Secondly, projectile penetration mechanism is observed.Finally, armour steel (simulating vehicle hull structure) damage mechanism (bulging, fracture, perforation etc) is studied.
Ballistic tests are performed for the proposed ballistic solution in a controlled and certificated test centre.Ballistic set up is shown schematically in figure 10.In the test setup, after the time of the projectile passing through the two-sensor curtain is determined, the average speed is calculated by taking into account the distance between the curtains.In this ballistics laboratory, two sensors were placed on each curtain.
And determines the passage time of the projectile between two curtains in μs.The average value is used in calculating the speed.The curtain system consists of two Oehler brand ballistic chronographs to measure projectile velocity and barrel assembly.
Barrels with different diameters from 5.56 mm to 25 mm are located in the test center.A 14.5 mm barrel was used for the testing within the scope of the study.The projectile impact velocity conditions specified in NATO STANAG 4569 are met by the gunpowder allowance setting.
An Aluminum 2024 T3 sheet as a witness plate behind the test sample is connected.If the projectile remains on its target, it means that full penetration does not occur.However, even if a piece broken off from the target penetrates the witness plate, it is considered as full penetration.In the ballistic test, projectile impact points were marked according to multi-hit shot pattern described in NATO AEP-55, Volume 1 (figure11).

Numerical analysis results
The calculated velocity-time curves (from numerical analysis) for different ceramic thicknesses are given comparatively in the graph in figure 12.The horizontal part of the curves shows the exit velocities of the  projectile after passing through the ceramic and composite layer (before penetrating the Armox 500T).From the graph, the exit velocities can be seen as 613 m s −1 , 529 m s −1 and 257 m/s for t, 1.083t and 1.305t ceramic thickness, respectively.Although the curve characteristics are the same for three different analyses, it is understood that the damage effect on Armox 500T material decreases as the ceramic thickness increases, as expected.The duration of the total absorption of the projectile energy is about 0.2 microseconds (ms).The total duration for the specimen with ceramic thickness of t is about 0.17 ms.Especially, the overall curve trend of the specimen with a ceramic thickness of 1.305t is different.
The analysis results in figure 13 show the images of the specimens with representative different ceramic thicknesses t, 1.083t and 1.305t at the moments when the projectile encounters the Armox 500T plate and the projectile energy is absorbed.These images are critical for evaluating the ballistic performance of the armor.The results of the analysis in figure 13 show the effective stress and effective strain outputs.Effective stress represents the internal stress of a material and reflects the actual deformation behavior of the material.Effective strain represents the actual amount of strain of a material.This shows the actual deformation behavior of the material and is important in the analysis of the mechanical properties of the material.The Green-St.Venant stress and strain tensors are mathematical expressions used to describe the behavior of linear elastic materials.These tensors are used to determine the actual behavior and deformation of the material.The Green-St.Venant stress tensor determines the internal stress and stress distribution of the material, while the strain tensor determines the deformation of the material and the amount of strain.In this way, effective stress and effective strain are important parameters for understanding and analyzing the actual behavior of the material.This is because the load carrying capacity, strength and deformation behavior of the material are directly related to these parameters.Therefore, determining these parameters is critical for understanding the performance and durability of the material.The blue colored regions in the graphs represent regions with low deformation and stress values.In these regions, the amount of strain and stress of the material is low.This indicates that the material exhibits a stable behavior under load and usually the material is still intact in the blue regions.Regions in red color represent regions with high deformation and stress values.In these regions, the material deforms and stretches significantly under load.This indicates that the material is heavily stressed and faces high deformations.In the red regions, the material is generally stressed or damaged.These color changes visually represent the deformation and stress levels in different regions of the material on the graph.This analysis is important for understanding how the material behaves under load and which regions are under more stress.In this figure, the colors blue and red are used to visually explain the concepts of effective stress and effective strain.
With increasing ceramic thickness, significant differences in deformation geometry are observed.Thicker ceramic layers tends to distribute the deformation caused by projectile impact more efficiently.As a result, the protective effect of the armor increases and the projectile damage is absorbed more effectively.The delamination observed in UHMWPE layers is related to the ceramic thickness.The use of thick ceramic layers can reduce the effect of delamination between UHMWPE layers while better absorbing the projectile impact.This can contribute to more effective damping of the projectile impact, maintaining the structural integrity of the armor.The increase in ceramic thickness also results in a significant reduction in the amount of deformation in Armox 500T.This indicates that the energy of the projectile impact transmitted to the armor is reduced and therefore the performance of the armor is improved.However, an increase in ceramic thickness also caused changes in the transmission of stress waves.The use of thicker ceramic layers causes more damage to the projectile core, altering the transmission of stress waves.This means that the armor absorbs the impact of the projectile more effectively, while the amount of damage propagating towards the interior increases.In conclusion, the results of the analysis show that ceramic thickness has a significant effect on the ballistic performance of the armor.The use of thicker ceramic layers can help the armor absorb the projectile impact more effectively and maintain its structural integrity.This, in turn, can increase the protective capacity of the armor and better protect the wearer.
Figure 14 shows a cross-sectional view of the stress distribution for different ceramic thicknesses, the back side of the armour structure and deformed projectiles.The stress distribution describes the equivalent stress (von-Mises).These images were studied to further analyze the performance of the armor, its behavior under projectile impact and the delamination mechanism.
Delamination occurs when the bond between different layers of the material is weakened or broken.This can adversely affect the durability and performance of the material.Here is how delamination occurs in real behavior and analysis: In reality, delamination between layers of the material is usually caused by external factors or loads.For example, factors such as impact, thermal fluctuations, high pressure or prolonged loading can cause the connection between the layers of the material to weaken or break.This reduces the strength of the material and compromises its structural integrity.In analyses, delamination is usually realized by examining the mechanical properties of the material and the load distribution.For example, in numerical analyses, factors such as the strength of the connection between the different layers of the material, stress and deformation distribution, stress concentration, and fracture points of the material are evaluated.These analyses can help identify which areas of the material are at high risk of delamination.The material model used allows modeling interlayer delamination.In addition, the manufacturing process and conditions of use of the material should be considered to understand the causes of delamination.Making sure that the material is manufactured and used in an appropriate way can help to prevent delamination.Ultimately, in real life and in analysis, delamination is the result of a weakening of the bond between the layers of the material.This can adversely affect the durability and performance of the material, so the causes and effects of delamination should be considered in the analysis so that the actual behavior can be simulated.
It is clearly seen that as the ceramic thickness increases, the projectile deformation and stress distribution on the backside of the armor structure changes.This indicates that thicker ceramic layers improve the performance of the armor by absorbing the projectile impact more effectively and maintaining the structural integrity of the armor.The importance of the composite plate in absorbing projectile energy is clearly seen and it is noted that delamination successfully absorbs this energy.This shows that composite structures are an effective option for absorbing and dissipating projectile impact.By comparing figure 14, it can be seen that the ceramic layer deforms the steel core projectile, the composite structure reduces the energy and the Armox 500T deforms the projectile to a high degree.This observation indicates that the projectile energy is effectively absorbed and stopped.It is noted that the projectile tip was broken by the ceramic effect and caused further damage after contact with the Armox.This highlights the armor's ability to absorb and dissipate the projectile impact.The high strength and elongation performance of the UHMWPE composite structure indicates that the gap between Armox 500T and the armor structure translates into the energy absorption capability of the composite.From this point of view, layered composite structures are an effective strategy used to enhance the performance of armor.Finally, the results of the numerical analysis indicate that the deformation of the Armox 500T plate decreases with increasing ceramic thickness and at the same time the projectile deformation occurs in a similar way.It is emphasized that on the back side of the Armox 500T plate, stress distribution occurs according to this deformation.This shows that the armor structure effectively absorbs and dissipates the projectile impact.
Figure 15 shows the displacements in the projectile direction for various ceramic thicknesses.In particular, for the Armox 500T, it shows the decrease in deformation values as the ceramic thickness is varied.Significantly, for a ceramic thickness of t mm the deformation is approximately 17 mm, while this value decreases to approximately 15 mm for a ceramic thickness of 1.083t mm.A further reduction is seen with a ceramic thickness of 1.305t mm, where the maximum deformation drops to approximately 10 mm.It is important to note that when the ceramic thickness increases by 8.3%, the deformation of the Armox plate decreases by approximately 12%.Similarly, when the ceramic thickness is increased by 30%, the deflection of the Armox steel shows a large reduction of 41%.These findings highlight the important role that ceramic thickness plays in reducing deflection and increasing the overall structural integrity of the Armox 500T plate.
The results in figure 15 emphasize the importance of preserving the deformation geometry.Preservation of deformation geometry means preserving the shape and structural integrity of a material during the deformation process.This means that as the material changes shape under the applied force, this change occurs homogeneously and in a predictable manner.It can be seen that the deformation in the projectile direction decreases with increasing ceramic thickness.This indicates that the ceramic layer provides a more effective protection against the projectile and controls the deformation of the material.The ceramic layer absorbs the projectile energy while limiting the deformation of the Armox 500T plate, thus maintaining the structural integrity of the material.For example, it is reported that when the ceramic thickness is increased by 30%, the deflection of the Armox steel is reduced by 41%.This indicates a significant reduction in the deformation of the material as a result of the effective bullet protection provided by the ceramic layer.In this way, the preservation of the deformation geometry improves the performance of the armor, providing the user with a higher level of protection.

Ballistic test results
As a result of the ballistic test, partial penetration occurred in the armour system and no penetration occurred in the Armox 500T armour steel plate supporting the armour panel (ceramic and backing plate).Only, bulging-like topology is formed in this armour steel plate without perforation.Thus, the armour system provided protection against 14.5 × 114 mm API/B32 projectile.The views of the ballistic test of the armour system are given in figure16.Shooting results are presented in table 10.
Local damage has occurred in the cylindrical ceramics used in the armour application which was an expected outcome.The ceramic geometry used here also has improved considerably the multi-hit feature of the armour system compared to other systems containing ceramics with classical geometry in the form square, hexagonal,, etc A comparison of cylindrical, square and hexagonal ceramic damages is given in figure 17.In the cylindrical ceramic structure, the damage area is smaller than damage areas in the other ceramic structures based on our experience from our related works.
In the literature, the difference in the fracture behavior of ceramic tiles with square and cylindrical geometries is well-documented.The comparison of damage areas in ceramic structures with these geometries, which was made in another study [66] shows that the damage in the ceramic structure with cylindrical geometry is more localized and the multi-hit resistance ability is better.The similar damage structures were obtained in our studies.This was an important factor in choosing cylindrical ceramics in armor design.
Ballistic performance can be improved by optimization of the angle of ballistic impact in armour design.The inclination of the surfaces to meet the ammunition can increase ballistic performance.When cylindrical ceramic is used, the ammunition is damaged more and loses more energy.This is related to the oblique impact of the projectile.
The UHMWPE composite plate, which is used as a support plate behind the ceramic, has a damage mode that is typically characterized by fibre damage and delamination after ballistic shooting.
It has been seen from the study that cylindrical ceramic-based armour panels are a weight/effective protection solution on combat vehicles against 14.5 × 114 mm API/B32 projectile with their modularity.They can be easily used for critical personnel protection zones in combat vehicles.An example of a cylindrical ceramic-based armour panel usage is given in figure 18.
The areal density comparison of various armour systems that can protect against 14.5 × 114 mm API/B32 ammunition defined in NATO STANAG 4569 as level 4 threat is given in figure 19.The areal density of Armox 500T armour steel (with a nominal hardness of 500 HB) is accepted as one unit and chosen as the reference armour material for comparison in figure 19.
A benchmark study between the proposed armour system and commercially available solutions that provide the same level of protection is seen in figure 19.As can be observed in figure 19, the proposed solution provides an intensive lightweight character compared to other possible alternatives.At this point it is also clear that the monolithic solutions which consists of solely armour steels (like Armox 500T, Armox 600T,) are the heavier solutions whereas the composite solution (like steel -perforated steel and etc) have better performance in terms of areal density.The proposed ballistic solution in this study, has approximately 50% less weight compared to monolithic Armox 500T, and 40% less weight compared to monolithic Armox 600T armour alternatives.In other words, the experimentally verified ballistic solution within the scope of this study provides a level 4 protection level confirming to NATO STANAG 4569 with an efficient areal density value.In this regards, the proposed solution possesses a huge enhancement potential regarding the mobility and/or amphibious capacities of armoured vehicles where the total weight of the vehicle is a key parameter in aforementioned capabilities

Conclusion
An add-on ceramic composite armour system design against 14.5 × 114 mm API/B32 projectile has been developed and verified in this study.For this armour system, both numerical analysis and ballistic testing are carried out, and the following results about the studies and recommendations are provided: • Numerical analyses of armour systems with different ceramic thicknesses are applied.They are used to calculate exit velocities of the projectile after passing through the ceramic/composite layer (before penetrating the Armox 500T) and also contributed to the selection of optimum ceramic thickness.
• In the analysis and alternative studies, it is observed that the armour system using thinnest thickness cylindrical ceramics had tears on the armour steel.Again, it has been seen in the analysis study that the ceramic/composite layer exit velocity (before penetrating armour steel) in the armour system using cylindrical ceramics of this thickness was 613 m s −1 .
• The ceramic geometry used in the armour system within the scope of NATO STANAG 4569 level 4 threat is an innovative structure.Due to the cylindrical ceramic structure, the projectile moving on after the moment of impact constantly encounters a curved and new surface, and thus it is deflected and exposed to more wear.In this case, it increases the ballistic performance of the armour.The use of cylindrical ceramic and UHMWPE plate has considerably lightened the armour design.UHMWPE is one of the composite material whose fibres have the lowest density and good mechanical properties.Thus, the areal density of the armour has also been reduced.
• Projectile damage occurs locally in the cylindrical ceramic-based armour system.This situation changes the multi-hit resistance ability of the armour to other hexagon, square, etc and increases this resistance ability considerably compared to other topological alternatives.
• It has been observed that significant weight gains can be achieved with this armour system developed against NATO STANAG 4569 level 4 threat, when compared to the monolithic solutions that provides equivalent protection levels.This weight advantage gained from the proposed armour systems increases the power/ weight (hp/tonne) ratio of military vehicles, contributes significantly to the road and terrain mobility and maneuverability without compromising the vehicle's survivability and even facilitates transportability of the vehicle and saves fuel.
• The present study also unveils the fact that with the proper and advanced material and contact modeling strategies, the finite element analysis based numerical models are capable of predicting the ballistic test result with a remarkable precision level.Thus, regarding the design optimization and/or limiting the necessary experimental efforts, for any further investigations as like the case of armour solutions dedicated to Level-5 protection level, finite element analysis models would serve as an enpowering tool.
In line with this study, the following recommendations can be made: • Alumina ceramics with different microstructures can be subjected to DOP (depth penetration) test and designs can be made repeatedly.Thus, innovative ceramics with better penetration damage mechanisms against projectiles can be researched and used in armour design.
• Boron carbide material, which is very hard and brittle, is a lightweight ballistic material.However, since the fracture pattern is characterized by sharp and small pieces, that is, it is not preferred due to amorphization problem.In addition, boron carbide material, which is amorphous after impact, empties its place after impact in the armour analysis and may cause the armour system to fail in close shots or in the face of the ammunition hitting the same place.Alumina ceramic material, on the other hand, is not only broken into larger pieces and is still characterized in such a way that it maintains its position, so it can resist the ammunition that will come to the same area.Boron carbide can be used in these studies by improving its properties.It offers a very light armour system solution compared to alumina ceramic (approximately 30%-35% lighter than alumina).
• The adhesive layer, in addition to the task of adhering the ceramic/backing plate, is a layer that increases the multi-hit resistance ability of the armour system and protects the surrounding ceramics from mechanical impacts.However, if the adhesive layer is too thick, it reduces the penetration resistance and limits the local strength of the target.The optimum thickness of the adhesive depends on the damage mechanism and the mechanical properties of the adhesive and bonded materials.Therefore, it is difficult to generalize the optimum adhesive thickness.Density, impedance and shear strength are the prominent features of adhesives in good bonding of ceramic materials.In line with these properties of the adhesive layer, the effect of the adhesive on the ballistic performance can be investigated by changing the adhesives type and thickness.Thus, it can be determined whether different types of polymer-based resin systems can be used as substitutes for each other in such applications.
• The gap between the basic armour and the add-on armour affects the exit velocity and direction of the projectile from the armor panel before entering the basic vehicle body armor.Provided that the design remains constant, in the add-on armour configuration presented in the study, the effect of the this gap on the ballistic performance can be determined again in the analyses and parametric studies.
• The number of shots can be increased by producing larger samples.As the number of shots on the panel increases, an in-depth idea can be gained about the effectiveness of the design.
is the yield stress, ̅ e p is the effective plastic strain, * strain rate and A, B, n, C and m are material model parameters.* T is calculated by the equation * =

Figure 3 .
Figure 3. Aspect ratio of solid elements.

Figure 4 .
Figure 4. Material models of armour layers.

Figure 6 .
Figure 6.The dimensions of the armour panel (a) and Armox 500T steel (b).

Figure 7 .
Figure 7. Use of the add-on armor panel on the vehicle.

Figure 8 .
Figure 8.The images of the core, cartridge case and core section of the 14.5 × 114 mm API/B32 core.

Figure 10 .
Figure 10.Schematic diagram of ballistic test set up.

Figure 11 .
Figure 11.Projectile impact points on the add-on armour panel.

Figure 13 .
Figure 13.Effective strains when the projectile tip meets the Armox 500T plate (a) armour panel with t mm thick ceramic, (b) armour panel with 1.083 mm thick ceramic, (c) armour panel with 1.305t mm thick ceramic.

Figure 14 .
Figure 14.Cross-sectional views of the stress distribution for different ceramic thicknesses: armour panel with t mm thick ceramic, (b) armour panel with 1083t mm thick ceramic, (c) armour panel with 1.305 t mm thick ceramic.

Figure
Figure Armox 500T plate deflection for three different parameters: (a) armour panel with t mm thick ceramic, (b) armour panel with 1083t mm thick ceramic, (c) armour panel with 1.305 t mm thick ceramic.

Figure 16 .
Figure 16.Views of add-on armour system after ballistic test: (a) front side view after the first shooting, (b) front side view after the second shooting, (c) Armox 500T view after the ballistic test, (d) x-ray images of the ceramics.

1 922 m s − 1 Figure 17 .
Figure 17.Comparison of hexagonal and square geometry ceramic damage zones.

Figure 18 .
Figure 18.The use of cylindrical ceramic-based armour panels on combat vehicles (re-created after [67]).

Figure 19 .
Figure 19.Areal density comparisons of armour steel and various armour systems providing protection for NATO STANAG 4569 level 4.
NATO STANAG 4569 NATO Standardization Agreement covering the standards for the 'Protection Levels for Occupants of Logistic and Light Armored Vehicles' NATO AEP-55 Procedures for Evaluating the Protection Level of Logistic and Light Armoured Vehicles LS-DYNA LS-DYNA is a general-purpose multiphysics simulation software package developed by the former Livermore Software Technology Corporation (LSTC).
σi * and σf * are the normalized values related to the Hugoniot elastic limit.σi * is the normalized intact equivalent stress, σf * is the normalized fracture equivalent stress and D is damage variable.σi * and σf * are written as in equations (2) and (3);
projectile came into contact.Projectile core geometry, mathematical model, aspect ratio of solid elements and material models of armour layers are given in figures 1-4, respectively.
Figure 9. Cross-sectional view of cylindrical ceramic body.