Military Related Engineering and Industrial Resources Thread

Posts · by **Yohannes** » Thu Apr 19, 2018 6:03 am

NationStates Military Related Engineering and Industrial Resources Thread

This thread is a collaborative thread made by the players behind the NationStates accounts Lamoni and Yohannes. When we were discussing military related engineering things with the player behind the NationStates account Lamoni, we did agree that creating this thread would benefit some NationStates players who would be interested in open-minded creative designing (Modern Technology, Post-Modern Technology, Future Technology, etc.) on the website NationStates.

Note that the institution we are a part of in real life (to which we are responsible to report to) will see to it that the resources contained in this thread are read only for educational or non-commercial purpose. They are restricted resources. They are not to be downloaded for commercial purpose. We have promised not to post anything Out of Character (i.e. see what we did in the Global Economics and Trade Help Desk thread in the past) because doing so will risk us showing our... inner dark side. Therefore we will not reply to any Out of Character comments that would invite any sort of Out of Character arguments made here (i.e. anything but “yes” or “no” kind of reply).

Any in-depth questions which require the answer of anything other than “yes” or “no” should therefore be directed to Lamoni.

Thank you!

P.S. Control+F (or Command+F on Mac) for easy keyword searching, e.g. "aircraft" or "tank" to find aircraft or tank related things

Restricted resources alphabetical list by title

Freely readable pdf files (thank you Lamoni!)

Out of character information (if anyone disagree feel free to say so!)

Posts · by **Yohannes** » Thu Apr 19, 2018 6:47 am

An experimental study on the shattering behavior of a high strength armour steel under blast and long rod penetrator impact

Journal page: https://www.journals.elsevier.com/materials-and-design

An experimental study on the shattering behavior of a high strength armour steel under blast and long rod penetrator impact

AUTHORS: Bidyapati Mishra a, Pradipta Kumar Jena a,*, B. Hazarika b, K. Siva Kumar a, T. Balakrishna Bhat a

a: Defence Metallurgical Research Laboratory, Hyderabad 58, India
b: Proof and Experimental Establishment, Balasore, Orissa, India
* Corresponding author. Tel.: +91 040 24586332, +91 9949221167; fax: +91 04024342252. E-mail address: pradipta_rrlb@rediffmail.com (P.K. Jena).

Article history:
Received 18 August 2008
Accepted 19 February 2010
Available online 26 February 2010

1. Introduction

Large caliber kinetic energy ammunitions like armour piercing discarded sabot (APDS) and fin stabilized armour piercing discarded sabot (FSAPDS) are main threats against metallic armour plates used in main battle tanks. These ammunitions use a tungsten heavy alloy rod as penetrators. The tungsten penetrator of FSAPDS possess not only remarkable penetration effectiveness against metallic armour plates, but also displays significant collateral effects during the penetration process. The tungsten penetrator after hitting the armour plate rapidly increases the penetration hole diameter due to dynamic radial stresses exerted by the penetrator. Two mechanisms are involved in growth of crater diameter: flow of penetrator erosion products, which exerts radial stress on the target and axial inertia of the target as it flows around the penetrator nose [1]. During the impact and cavity enlargement process shock waves and high stresses are generated due to which even metallic armour may shatter depending on its mechanical properties like hardness, toughness, microstructural features like inclusion content, homogeneity and level of internal stress. Shattering can be defined as a condition at which the material breaks up into multiple fragments and becomes unable to sustain further loading. It reduces the effectiveness and efficiency of armour plates considerably.

Among metallic materials, high strength steels are preferred for armour applications due to their high strength and high hardness along with good toughness. There has been a continuous enhancement in strength and hardness of steels through alloying additions, heat treatment and thermo-mechanical treatments. Inevitably increase in strength and hardness tends to increase the probability of shear localisation and hence increases shattering tendency on impact. While high strength and high hardness is required to resist the projectile, resistant to shattering is essential for integrity of the armour and for withstanding multiple hits. Previous studies on ballistic performances are mainly concentrated on the effect of strength and hardness of steels on their ballistic performance [2–10]. Mangallelo and Abbott investigated the influence of mechanical properties on the low velocity impact behavior of steel armours [2]. It was figured out that hardness is the most important property affecting the ballistic performance. Dixit studied extensively the impact of plate hardness on the ballistic behavior of steel plates [5–7]. He concluded that the effect of hardness of a plate on the ballistic performance depends on whether the stress state is predominantly plane strain or plain stress. In plain strain condition, increase in hardness of the plate increases the ballistic performance. However, under plain stress condition ballistic performance increases initially with hardness and beyond a certain hardness level ballistic performance decreases due to formation of adiabatic shear band induced plugging failure. Ubeyli et al. also reported the correlation of material strength and hardness with its ballistic performance [8]. They pointed out that with increase in hardness of target plate the penetration and propagation ability of the projectile decreases significantly. In another study on AISI 4140 steel, Demir et al. found that beyond a specific hardness level the test specimens were broken in a brittle manner [9]. In a previous investigation on high strength DMR-1700 steel, Jena et al. observed that an optimum combination of strength and impact toughness leads to best ballistic performance [10].

Fracture prediction of thin plates under localised impulsive loading and many aspects of the problem including dynamic response of the material were studied earlier by Nurick and Shave [11]. They found that thin plates exhibit large ductile deformation, tensile tearing and transverse shear with increased rate of loading.

The effect of heat treatment on the strength and failure of structural steels under shock wave loading was reported by Golubev et al. [12]. They observed the generation of microcracks and spalling in two different microstructures. It was found that crack nucleation takes place at inclusions and structural inhomogeneities. The nucleating damage is small microcracks whose growth and merger leads to larger cracks and spalling. Explosion induced deformation features had been studied in AISI 304Cu steel discs by Firraro et al. [13]. In this study explosive charges and explosive to target distances were varied to understand microstructural features under shock loading. They have reported partial surface melting, recrystallisation and intense mechanical twinning in the material subjected to explosive detonation. Fracture predication of thin plates under localised impulsive loading explaining different phases of discing was investigated by Lee and Wiertzbicki [14]. This study showed that fracture of thin plates subjected to localised explosive loading consists of three stages with increasing intensity of applied load. In the first phase, the plate undergoes large inelastic deformation. In the second phase, an initial circumferential crack detaches the circular cap from the plate. In the last phase radial cracks propagate outward to form petals. Singh et al. calculated the dynamic yield strength of mild steel by impacting steel balls in the velocity region of 1900–5200 m/s [15]. They studied the penetration process and spalling behavior of mild steel at high strain rates and developed a numerical formula to determine the growth of crater formed by projectile impact.

Before being put to use, armour steel plates are assessed mechanically and ballistically to confirm their ability to withstand the resistance of armour-piercing ammunition [16–19]. Standard test procedures are available for measuring strength and hardness of metallic armour plates (ASTM E 8 M-04, E 140-02). But, limited emphasis has been given to evaluate and understand their shattering behavior, which too is an important aspect for any application of armour in terms of real battle field scenario.

One way of testing for shattering behavior is to make large sized plates and test the performance with actual shots such as long rod anti-tank penetrators. But this requires tedious trials and is much time consuming. Properties such as K1c or Charpy impact energy values are of some help to estimate shatter resistance. However, they do not reflect the conditions of high loading and loading rates in actual use involving shocks, shockwaves and their reflections. Therefore, it is desirable to develop suitable and speedy alternative test method for evaluating the shattering behavior of armour plates and also to compare the fracture behavior of the tested samples with long rod penetrator impact.

In the past, to study the damage processes and to get an idea about the related failure conditions at high rates of loading various tests have been proposed. One such method is propelling a flyer plate through a gas gun to impact upon a fixed target plate [20,21]. It generates high strain rate simulating actual field results. The material behavior under such high strain rates can give a fair idea about selection of better armour materials. Another approach, especially for study of shear band kinetics is, by detonating an explosive mixture inside a hollow cylinder of the material of interest [20,22]. Varying the explosive quantity and sample dimensions one can vary amplitude and duration of loading, and can study the resulting dynamic fracture process. The number of shear bands, adiabatic shear band induced cracking and susceptibility of the material to crack can be correlated to find dynamic strength of the material.

To the best of our knowledge, no test method is available in literature, which quantifies the shattering resistance of thick metallic plates during ballistic penetration, which is crucial for developing advanced high hardness armour materials. In the present investigation an attempt has been made to evolve a method to grade the shatter resistance of armour steel and to assess the threshold conditions for shattering. This investigation also attempts to understand the shattering behavior of material and their difference when subjected to single hit and multiple-hit scenarios. Damage patterns and fracture morphology were also studied in order to better understand the shattering behavior of the target plates. The post blast fracture surfaces of the samples after explosive detonation have been judged against the fracture surfaces of long rod penetrator impacted targets.

2. Method

High strain rates exceeding 10⁴ s^-1 is generated during projectile impact as well as during explosive target interaction [23]. We have chosen explosive detonation in our method for generating high strain rates. The best possible method for observing the shattering resistance of a metallic plate would be to measure the strain rate generated during the penetration of a long rod penetrator and then finding out the amount of certain explosive required to produce the same strain rate levels in the material. Detonating the material with the required mass of the explosive can give a good idea of the shattering behavior of the material. However, due to the unavailability of strain measuring facilities at such high strain rates we have followed an alternative method for obtaining the shattering behavior of metallic armour plates. In this method holes are made in armour steel plates and subsequently filled with increasing mass of explosive and detonated until shatter occurs.

Our initial interest is to determine the minimum hole diameter which leads to shattering in case of single hit scenario. First armour steel plates of 70 mm thickness are impacted against 125 FSAPDS ammunition and the diameter leads to shattering is found out. As obvious, any perforation in the armour plate will cause a diameter more than the diameter of the projectile used. In addition depending upon the material properties and lateral expansion of the perforated hole, material may shatter. To generate similar results under blast, holes of 30 mm, 40 mm, 45 mm and 50 mm diameter are drilled in the 70 mm thick armour steel plates. The starting hole size chosen is 30 mm because it approximates the diameter of the tungsten projectile of FSAPDS ammunition. These holes are filled with explosive and the armour plates are subsequently detonated with an electric detonator. The minimum diameter of the hole at which shattering occurs is considered as the critical parameter,
defining shattering resistance of the material. The blast results are then compared with the results of long rod penetrator impacted armour steel plates.

Secondly we intended to find out the shattering behavior of armour steel plates in case of multiple-hit scenario. Armour steel plates have impacted against multiple 125 FSAPDS ammunition and the number of localised impacts lead to shattering have determined. Armour plate having initial hole diameters of 30 mm and 45 mm are filled with explosive charge mass, and initiated. Plastic explosive was filled and the detonation process was repeated on the same sample till the sample shattered. Here the number of explosive blasts with a particular initial hole diameter, which the material sustained before shattering, is considered as the critical parameter defining shattering behavior of the material. The explosive blast results are judged against multiple impacted long rod penetrator armour steel plates.

3. Experimental

Material chosen for this study is a high strength low alloy armour steel. This steel is made by vacuum arc melting at Mishra Dhatu Nigam Limited, India. Austenitisation at 900 C followed by oil quenching and tempering at 250 C was done at the factory. Its chemical composition and the mechanical properties in the as received condition are given in Table 1. Samples of 150 x 150 x 70 mm were cut from a single plate and drilled in the middle to make holes of 30, 40, 45, 50 mm diameter. A concrete pit of 1200 x 1200 x 1000 mm size was used for safe explosion. The steel sample was placed inside the concrete pit and the drilled holes were completely filled by PEK explosive. PEK is a plastic explosive and it is received from ordnance factory, Kirkee, India. The properties of PEK supplied by manufacturer are given in Table 2. The sample was covered with sand and a mild steel plate of 10 mm thickness was used to cover the concrete pit in order to reduce the external hazardous effects. Detonation of PEK explosive was done by initiating the electrical detonator. The testing arrangement is shown in Fig. 1. Minimum of three separate specimens of each hole size were detonated in order to get a statistical result. Therefore more than 25 specimens were prepared and impact
loaded using explosive.

Plates of 1000 x 1000 x 70 mm were cut and tested against long rod penetrator (FSAPDS) ammunition fired from a tank gun. The target was fixed to a target stand with the help of bolts. The distance of the target from the tank gun was 100 m. Fig. 2 shows the schematic of the FSAPDS projectile. The inner core penetrator is made up of tungsten heavy alloy with a density of 18 g/cc³. The diameter of the tungsten penetrator was 30 mm and it has the L/D ratio 20. The mass of the projectile is 7.81 kg. All the targets were impacted at 0 angle of attack. The striking velocities of the projectiles were measured as 1620 ± 25 m/s. The schematic of the ballistic testing arrangement is shown in Fig. 3. For single penetration case each target was subjected to single FSAPDS penetrator, whereas for multiple penetration case two FSAPDS penetrators were fired on the same target plate. Eight target plates were tested against FSAPDS penetrator.

In case of single detonated samples or single long rod impacted plates a through-thickness crack is taken as the critical point for the occurrence of shattering. Whereas in case of multiple detonated or multiple impacted samples complete fragmentation of the samples is taken as the critical point for shattering.

Table 1: Chemical composition and mechanical properties of the armour steel; and Table 2: PEK explosive composition and properties.

Fig. 1. Schematic diagram of experimental setup.

Fig. 2. Schematic diagram of FSAPDS penetrator.

Following impact and detonation, each impact site was subjected to detailed examination. The holes on the target plate were photographed on both entry and exit sides. Damage patterns were investigated at the front and rear face of the target plates. Then fracture surfaces of the target plates were cut carefully without damaging and observed in a LEO scanning electron microscope to study the mode of fracture.

Fig. 3. Schematic diagram of the ballistic testing arrangement

4. Results and discussion

Table 3 gives the details of the single hit FSAPDS penetration results. From the observed results in Table 3, it can be said that below 53 mm crater diameter shattering is absent. When the crater diameter grows to 53 mm or beyond shattering is observed in the tested samples. Fig. 4 shows the post impact photograph of fractured armour steel plates. It can be seen that the FSAPDS ammunition has perforated the 70 mm thickness plates. From the fracture surface observations it was clear that in all the shattered plates a large single crack was observed radiating out from the impact hole up to the edge of the sample (Fig. 4a–c). This crack was observed running through the complete thickness of the plates (Fig. 4d). In fact, the appearance of this through-thickness crack was indicative of a local rupture caused by excessive shear [24].

Table 3: Results of the plates impacted with single FSAPDS ammunition.

Table 4 shows the initial and final diameter of the holes of the targets subjected to single detonation. Detonation in 30 mm diameter hole with a corresponding explosive charge of 68 g showed a final diameter of 32.2 mm. The samples were collected and photographed after detonation. Fig. 5 gives the appearance after detonation of the top and bottom face of the sample. Chipping of the material at the edges of the hole can be clearly seen. Chipping of material at the hole periphery can be explained from the high radial and circumferential tensile stresses generated by the detonation waves. Material chipping is also reported in plates subjected to ballistic impact [25]. More chipping was observed on the top face than at the bottom face. No cracks are observed on either surface of the sample. The top face being free allows full reflection of the shockwaves which causes pronounced chipping due to spalling. The bottom face being in contact with the concrete ground transmits part of the shockwave to it, experiences weaker reflected waves and hence shows lesser chipping.

Table 4: Initial diameter and final diameter of the hole with the weight of explosive required in single detonation method.

Fig. 6 shows post blast appearance of top and bottom face of the sample, with 40 mm diameter hole filled with 123 g of explosive. The fracture surface observed was similar to that of 30 mm diameter hole. However, more chipping of material was observed at the hole edges. At the bottom, material chipping observed was less severe. Only a few cracks were observed at the hole edge. Similar experiments were done on 45 mm diameter hole with an explosive charge mass of 161 g. The diameter of the hole got enlarged from 45 mm to 48.7 mm after the explosion. Severe chipping was observed at the top and bottom face of the sample (Fig. 7). At the bottom face along with chipping a number of small cracks were observed radiating outward from the periphery of the hole (Fig. 7b). The lengths of all the cracks were measured and the average crack length was found to be 6 mm. When the tests were carried out on 50 mm diameter hole filled with 206 g of explosiv charge mass, it was seen that numerous cracks originated at the hole periphery (Fig. 8 ). Only one big crack reached to the edge and can be clearly seen running through the thickness of the target (Fig. 8c). Severe chipping of material was also observed at the top as well as at the bottom faces. This indicates arrival of the critical conditions for shattering of the material. This fracture feature is similar to the observations made in case of single hit FSAPDS impacted armour steel plates (Fig. 4d). The critical diameter for shattering observed in single hit explosive detonation case is 50 mm, which is close to the critical shatter diameter of 53 mm obtained from long rod impacted plates. The fracture surfaces of the plates impacted by multiple FSAPDS ammunition are shown in Fig. 9.

Fig. 4. Showing armour steel plates after impact with FSAPDS ammunition. (a–c) Single large crack can be seen radiating from the impact hole up to the edge of the plate. (d) Side view of the plate showing the crack running through out the thickness.

Fig. 5. Surface appearance after detonation in 30 mm diameter hole after single blast. Chipping of material is shown by arrows: (a) top face, and (b) bottom face.

Fig. 6. Surface of the sample with 40 mm diameter hole after single blast. Chipping of material is shown by arrows: (a) top face, and (b) bottom face showing a small crack.

Fig. 7. Surface appearance after detonation in 45 mm diameter hole after single blast. Chipping out of material is shown by arrows: (a) top face showing severe chipping of material, and (b) bottom face showing a number of cracks at the hole edge.

Fig. 8. Surface appearance after detonation in 50 mm diameter hole after single blast. Chipping out of material is shown by arrows: (a) top face showing severe chipping of material with cracks on the front face as well as inside the hole, (b) bottom face, and (c) side face showing a big crack running through out the thickness.

Fig. 9. Plate impacted against multiple FSAPDS ammunition. Arrow direction in (c) and (d) shows the projectile impact direction. (a) and (b) Completely shattering of the plate after second impact, (c) chevron fracture pattern, and (d) chevron fracture pattern with smooth rubbed surface.

It can be seen that the second impact is close to the first one and the plates are completely shattered and broken into pieces after the second impact (Fig. 9a and b). Chevron pattern was a predominant feature observed in the fractured surfaces (Fig. 9c). Chevron pattern represents a brittle mode of failure and is formed by the way the fast moving crack front propagates along the plate. The chevron marks usually point to the crack origin. Two different areas marked as A and B can be observed in the fractured surface of Fig. 9d. While region A showed a well developed chevron fracture pattern, a smooth surface was observed in region B, indicating rubbing action of the pieces during the fracture process. In the bottom side of the region B, a crack can be seen. Table 5 shows the average initial and final diameter of the holes of the targets tested under multiple detonations. In the first blast using 68 g of explosive the 30 mm hole got enlarged to an average diameter of 32.2 mm. Chipping of material was observed at the top and bottom faces (Fig. 10a and b). After the second blast with an explosive mass of 78 g on the same sample the hole diameter increased to 35.09 mm. Chipping was observed at the hole edges in both the faces, which was more severe than in the cases of single blast (Fig. 10c and d).

Table 5: Initial diameter and final diameter of the hole with the weight of explosive required in multiple detonation method.

Again the detonation was repeated in the same samples with 94 g explosive. After third blast the hole diameter increased to an average diameter of 39.79 mm. The material at the edge of the hole showed presence of several cracks on the surface along with severe chipping, as seen in (Fig. 10e and f). The maximum crack length was measured to be 20 mm. As shattering was not observed, the detonation was repeated in the same sample. The charge was 117 g. The sample shattered completely after the fourth blast. Some of the shattered pieces after fourth blast can be seen in Fig. 11a. It clearly showed multinucleated fracture occurrence and chevron pattern appearance (Fig. 11b).

Fig. 10. Comparison of the surface after detonation in 30 mm initial diameter hole after first, second and third blast. (a) Top face after first blast, (b) bottom face after first blast, (c) top face after second blast, (d) bottom face after second blast, (e) top face after third blast showing cracks, and (f) bottom face after second blast.

Fig. 11. Shattered pieces after fourth blast in 30 mm diameter hole: (a) fracture pieces, and (b) chevron fracture pattern in the shattered piece.

Fig. 12. (a) Shattered pieces in 45 mm diameter hole after second blast, (b) chevron pattern appearance of fracture piece No. 1, (c) rubbed region appearance of fracture piece No. 2, (d) fracture surface appearance of piece No. 3 showing chevron pattern and crack generation at its nucleating site, and (e) radial cracks in the central area of piece No. 3.

Multiple detonation experiments were also conducted on 45 mm initial diameter hole. Table 5 shows the number of blasts required to shatter the plate and the initial and final diameter after each blast. In the first blast the hole got enlarged to 48.7 mm. The second detonation was done with an explosive mass of 199 g. After the second blast, severe chipping with presence of multiple cracks was observed in the sample. In the third detonation with an explosive
of 217 g the material showed complete shattering. Some of the fractured pieces after the detonation can be seen in Fig. 12a. Fig. 12b shows the fracture surface appearance of fracture piece No. 1. Chevron pattern is clearly visible in the fracture surface.

Fig. 12c describes the closer view of the fracture surface of the piece No. 2. It can be conjectured that the surface has undergone severe rubbing with the center part or other fractured pieces during the process of fracture and subsequent movement of the pieces. Fracture surface of piece No. 3 is shown in Fig. 12d. It clearly shows chevron patterns with generation of a crack at the chevron pattern originating site. The crack was seen to have propagated in the
thickness direction as shown by the arrow. The crack generation is a result of the interaction of stress waves near the surface. Observation of chevron cracks are also reported in detonation driven fracture of closed-end cylindrical tubes [24]. Fig. 12e shows the presence of radial cracks in the central area in the fracture surface of piece No. 3. Fracture surface of piece No. 4 is shown in Fig. 13.

Fig. 13. (a) Different fracture surfaces of piece No. 4 of Fig. 8, (b) showing chevron pattern with different crack propagation direction, and (c) showing transverse and radial cracks.

Fig. 14. SEM pictures of the fracture surfaces. Arrow marks indicate the microcracks in fracture surface. (a) SEM picture of FSAPDS impacted fracture surface, (b) and SEM picture of multiple detonated fracture surface.

There are three different types of areas observed in the fracture surface of this piece, marked as A, B and C as shown in Fig. 13a. Fig. 13b shows close view of area A. Chevron pattern was observed in the fracture surface with different crack nucleation points and propagation directions. Region B showed a smooth surface due to rubbing action with the other fracture surface. Area C showed presence of transverse and radial cracks, as indicated by arrow (Fig. 13c). It showed no rubbing or chevron patterns.

Table 6: Comparison between the effect of single blast and the multiple blasts on material response.

Radial cracks as observed in Fig. 13c could be due to tensile tangential stresses causing crack generation by the tensile stress pulse traversing the specimen after detonation. Multinucleated fracture pattern such as the one observed in Figs. 11b and 13b show that even after initiation of fracture, the stress pulse continues to exceed the tensile strength of the material, leading to multiple fractures. The cracks may normally originate at second phase particle, precipitates, inclusions, and grain boundary or at localised defects [12,26]. When compressive shock wave propagates within a medium and face a change in density either at a free surface or at an internal surface, reflects with a change in sign or propagates in a direction opposite to the initial wave direction. When the overall tensile pulse (after the cancellation of compressive pulse) exceeds the tensile strength of the material, fracture takes place. The fracture features observed in the multiple detonated samples are very much comparable with the features observed in multiple FSAPDS impact samples (Figs. 9 and 11–13).

Fig. 14 shows the scanning electron micrographs of the fracture surface of the FSAPDS impacted sample (region A of Fig. 9c) and the fracture surface of the sample subjected to multiple detonations (region A of Fig. 13b). Both the micrographs showed cleavage type of fracture with presence of microcracks in the fracture surface. Table 6 describes the difference in material behavior during single and multiple detonations with nearly identical hole diameters and explosive loads. The sample with 40 mm hole diameter when detonated with 123 g of explosive showed small cracks with chipping at the hole periphery (Fig. 6). In the multiple detonation experiments, the hole diameter of the 30 mm sample was enlarged to 39.79 mm after the third blast. When this 39.79 mm diameter sample was again detonated with 117 g of explosive complete shattering was observed in the sample. Similar observations were made for the 50 mm hole diameter sample subjected to detonation with 206 g of explosive with that of 51 mm hole diameter sample subjected to third detonation with 217 g of explosive. In both the single and multiple blast cases the hole diameter and weight of the explosive are nearly equal. However, there is a huge change in the shattering behavior in the repeatedly detonated sample in comparison to the single detonated sample.

Traversing stress waves in the case of multiple detonated samples create microcracks in the sample as can be seen from the scanning electron micrographs of the fracture surface (Fig. 14). The main effect of the microcracks is to reduce the area over which the stress acts [27]. Thus, these microcracks concentrate the stress and gets elongated with each additional detonation. And when they reach a critical crack length shattering takes place [12]. That is one reason for easier crack growth and shattering in the case of multiple blast samples. Also the shock loading leads to shock hardening and creation of defects which makes it easier for cracks to initiate and propagate. Indeed, due to higher hardness cleavage type fracture is observed in the fracture surface of multiply shocked samples (Fig. 14).

The present study shows that detonation blast method closely approximates the shattering behavior of target plates subjected to long rod penetrators in single impact as well as multiple impact conditions. The critical shatter diameter obtained under detonation with explosive method matches well with that of FSAPDS impacted samples. The fracture appearances obtained under the single detonation as well as multiple detonations are very much comparable to that of FSAPDS impacted plates. The scanning electron micrographs of the fracture surface of multiple detonated samples show similar features as that of the fracture surface of multiple FSAPDS impacted plates. So, detonation blast method can be used to compare different high hardness metallic materials with respect to their shatter resistance making it a simple scaled down testing method for rapid screening work in materials development. Any material which has a critical diameter more than 50 mm evaluated in the detonation blast method would resist shattering better than the studied steel. Similarly in case of multiple detonation case, any material with an initial hole diameter of 45 mm taking more than three blast loadings to shatter completely, would perform better than the steel studied. The method developed here may be further refined by choosing the explosive and the initial hole diameters to more closely match the effects of any particular large caliber ammunition at any particular velocity. The method can also be used to compare the effects of heat treatment and other processing parameters on shattering resistance.

5. Conclusions

Shattering tendency of metallic armour materials using detonation with explosives has been evaluated which approximates the long rod impact shattering behavior. From the blast method the critical crater diameter for shattering is found to be 50 mm for the studied steel where as it is found to be 53 mm in case of FSAPDS impacted plates. The fracture feature changes from propagation of crack through the thickness of the plate in case of single detonation to multinucleated crack with chevron pattern in multiple detonation case. Microcracks generated during high strain rate deformation process give easy crack propagation path and lead to shatter.

Acknowledgements

The authors are grateful to Defense Research & Development Organization (DRDO) for financial support to carry out this work at Defense Metallurgical Research Laboratory, Hyderabad. The authors wish to express their gratitude to the Director, DMRL for granting permission to publish this paper. The services rendered by scanning electron microscopy group of DMRL are gratefully acknowledged. Funding of the work by Defense Research & Development Organization (DRDO) is gratefully acknowledged.

References

[1] Lee M, Bless S. Cavity dynamics for long rod penetration. Technical report number – IAT. R 0094; 1996.
[2] Manganello J, Abbott KH. Metallurgical factors affecting the ballistic behavior of steel targets. J Mater 1972; JMSLA 7:231–9.
[3] Neilson AJ. Empirical equations for the perforation of mild steel plates. Int J Impact Eng 1985; 3:137–42.
[4] Sangoy L, Meunier Y, Pont G. Steels for ballistic protection. Isr J Technol 1988;24:319–26.
[5] Dikshit SN, Kutumbarao VV, Sundararajan G. The influence of plate hardness on the ballistic penetration of thick steel plates. Int J Impact Eng 1995; 16(2):293–320.
[6] Dikshit SN. Ballistic behavior of tempered steel armour plates under plane strain condition. Def Sci J 1998; 48(2):167–72.
[7] Dikshit SN. Influence of hardness on perforation velocity in steel armour plate. Def Sci J 2000; 50(1):95–9.
[8] Übeyli M, Yıldırım RO, Ögel B. On the comparison of the ballistic performance of steel and laminated composite armors. Mater Des 2007; 28(4):1257–62.
[9] Demir T, Übeyli M, Yıldırım RO. Investigation on the ballistic impact behavior of various alloys against 7.62 mm armor piercing projectile. Mater Des 2008; 29:2009–16.
[10] Jena PK, Siva Kumar K, Bhat TB. Effect of heat treatment on mechanical and ballistic properties of ultrahigh strength DMR-1700 steel. Met Mater Process
2007; 19(1–4):339–46.
[11] Nurick GN, Shave GC. The deformation and tearing of thin square plates subjected to impulsive load – an experimental study. Int J Impact Eng 1996; 18(1):99–116.
[12] Golubev VK, Novikov SA, Sobolev YS, Yukina NA. The effect of heat treatment on the strength and failure of steels 30KhGSA and 20KhN3A under shock wave loading. Strength Mater 1987; 19:103–6.
[13] Firraro D, Matteis P, Scavino G, Ubertalli G, Lenco MG, Pellati G, et al. Mechanical twins in 304 stainless steel after small charge explosions. Mater Sci Eng A 2006; 424:23–32.
[14] Lee YW, Wiertzbicki T. Fracture prediction of thin plates under localised impulsive loading. Part I: dishing. Int J Impact Eng 2005; 31:1253–76.
[15] Singh M, Sood D, Gupat RK, Kumar R, Gautam PC, Sharma SewakB, et al. Dynamic yield strength of mild steel under impact loading. Def Sci J 2008; 58:27584.
[16] Specification No. CQA(M)/ 51. Controller of quality assurance, Ministry of Defence, Government of India; 1999.
[17] Specification no. – I.T. 100 E; 1956.
[18] Specification no. – I.T. 80 F; 1966.
[19] GOST B –23958-80, Committee of USSR on Standards; 1980.
[20] Zukas JA, Nicholas T, Swift F, Greszczu B, Curran DR. Impact dynamics. Wiley – Inter science publication; 1982. p. 333.
[21] Barbee T, Seaman L, Crewdson R, Curran D. Dynamic fracture criteria for ductile and brittle materials. J Mater 1972;7(3):393–401.
[22] Seaman L, Curran D, Shockey AD. Computational model for ductile and brittle fracture. J Appl Phys 1976; 47:4814.
[23] Buchar J, Bilek Z, Dusek F. Mechanical behavior of metals at extremely high strain rates. Transtech Publications; 1986. p. 21.
[24] Mirzaei M. Failure analysis of an exploded gas cylinder. Eng Fail Anal 2008; 15:820–34.
[25] Jena PK, Jagtap N, Siva Kumar K, Bhat TB. Some experimental studies on angle effect in penetration. Int J Impact Eng 2010; 37:489–501.
[26] Ryttberga K, Wedela MK, Dahlmanb P, Nyborga L. Microstructural evolution during fracture induced by high strain rate deformation of 100Cr6 steel. J Mater Process Tech 2009;209:3325–34.
[27] Chena D, Tanb H, Yua Y, Wanga H, Xiea S, Liua G, et al. A void coalescencebased spall model. Int J Impact Eng 2006;32:1752–67.

Posts · by **Yohannes** » Thu Apr 19, 2018 9:14 am

Interaction of a Shaped-Charge Jet with Moving Reactive Armor Plates

Combustion, Explosion, and Shock Waves, Vol. 49, No. 4, pp. 495–500, 2013.
Original Russian Text c I.F. Kobylkin, N.S. Dorokhov

Interaction of a Shaped-Charge Jet with Moving Reactive Armor Plates

AUTHORS: I. F. Kobylkin_a and N. S. Dorokhov_b

a. Bauman Moscow State Technical University, Moscow, 105005 Russia; kobylkin ivan@mail.ru.
b. Research Institute of Steel, Moscow, 127411 Russia

UDC: 623.483.3+623.562.7

Translated from Fizika Goreniya i Vzryva, Vol. 49, No. 4, pp. 125–130, July–August, 2013.

Article history:
Original article submitted October 30, 2012

INTRODUCTION

The scientific history of reactive armor for protecting armored vehicles against shaped-charge warheads began over 50 years ago, when B. V. Voitsekhovskii, V. L. Istomin (Moscow Physicotechnical Institute), A. I. Platov (Research Institute of Steel) and other researchers headed by M. A. Lavrent’ev conducted the first experiments to study reactive armor against shaped-charge jets [1, 2]. Currently there is no alternative to the use of reactive armor to protect armored vehicles from anti-armor and shaped-charge weapons.

The basis of the explosive reactive armor are flat elements consisting of two metal plates and an explosive layer located between them. The explosive layer detonates under the action of a shaped-charge jet (SCJ). The detonation products accelerate the plates, which impacts at an angle the SCJ. This leads to deflection, destruction or wearing of the SCJ, resulting in a significant reduction in the depth of penetration of the main part of the target located behind the reactive armor. Despite the simplicity of the design of most of these devices, the physics of operation of the armor is very complicated and not all the processes that determine its protective capability are fully understood. This paper focuses on studying the dynamics of the collision of a SCJ with the front reactive armor plate (flying toward the SCJ) and the rear plate (flying behind the SCJ).

Voitsekhovskii [2] assumed stationary continuous interaction of a SCJ with reactive armor plates within the framework of the incompressible fluid model. Despite the fact that the computational model of the process constructed in [2] adequately reflects the results of experiments and the mechanism of interaction of the SCJ with armor plates underlying this model is not fully realistic. Currently, the nonstationary mechanism of interaction involving periodic interaction of a SCJ with relatively thin (a few millimeters thick) metal armor plates [1–5] is considered to be the main mechanism. The high-velocity head element of a SCJ punches a hole into the plate, whose diameter exceeds the diameter of the SCJ and is determined by the velocity of the SCJ, the strength of the plate material, and its thickness. Motion of the plate at an angle to the SCJ head-on or co-directionally to the SCJ leads to impact contact of the edge of the hole in the plate with the side face of the next elements of the SCJ. This leads to partial (or entire) wearing of the SCJ element, which is subjected to a lateral impulse and moved in the lateral direction, again punching an elliptical hole in the plate with diameter greater than that of the SCJ. Later on, the process is repeated, resulting in the SCJ subjected to periodic transverse perturbations increasing with time.

Fig. 1. Flash radiograph of interaction of a copper SCJ with reactive armor elements: (a) the front plate is 2 mm thick, and the rear plate is 1 mm thick; the bench mark is at a distance of 190 mm from the element; (b) the front plate is 1 mm thick, and the rear plate is 2 mm thick, the bench mark is set at a distance of 220 mm from the element; (c) the front and rear plates are 2 mm thick, and the bench mark is at a distance 180 mm from the element; the thickness of the explosive layer is 3 mm; the angle between the SCJ and the armor element is 30^O.

The described process is confirmed by flash radiographs of SCJ interaction with reactive armor elements. Figure 1 shows the corresponding flash radiographs. Two steel plates of mild steel were placed at an angle of 30^O to the direction of the impact of the SCJ. Between them was placed a 3 mm thick layer of a flexible explosive based on PETN. The thickness of the plates was 1 and 2 mm in different experiments. Laboratory shaped charges of phlegmatized RDX of density 1.65 g/cm³, 56 mm in diameter, and with a copper conical liner with a cone angle of 50^O were used. The depth of penetration of homogenous steel armor targets by these shaped charges placed at a distance of 150 mm from the target was 238 ± 13 mm, the velocity of the head element of the SCJ was 8000 ± 250 m/s, and the diameter of the head elements of the fully stretched SCJ was 1.9 ± 0.1 mm. As seen from the radiographs, the state of the SCJ after passage through the armor element depends on the thickness of the plates and their arrangement and indicates periodic interaction of the SCJ with the plates.

Because the reactive armor plates scatter in opposite directions, one might expect that they will produce transverse perturbations in the SCJ also in the opposite direction. However, most radiographs of the process show perturbations caused only by the action of the rear plate (see Fig. 1).

In [6], this effect was explained by a kind of filtering action of the rear plate. The explanation is obvious from Fig. 2. The transverse perturbations of the SCJ produced by interaction with the front plate are cut off at the intersection of the rear plate and at the exit from of the reactive armor device, there are only transverse perturbations of one direction resulting from the interaction of the SCJ with the rear plate. If this is so, the impact of the front plate on the SCJ partially reduces the effectiveness of the rear plate, as it leads to additional consumption of its material for cutting the transverse perturbations in the SCJ caused by the action of the front plate.

Fig. 2. Passage of the shaped-charge jet through scattering reactive armor plates, from [6]: 1 and 2 are the front and rear armor plates, respectively, 3 is the shaped-charge jet, 4 and 5 are the transverse perturbations in the SCJ produced by the front and rear plates, respectively, and 6 are the breaks of the SCJ resulting from the rear plate cutting the SCJ perturbations produced by the front plate.

However, a careful analysis of the radiographs does not confirm the mechanism proposed in [6] to explain the filtering action of the rear plate since in the passage of the high-velocity part of the SCJ through the front plate in CN, developed transverse perturbations were found not only in our radiographs (Fig. 1), but also in the radiograph of [6] and the proton radiograph of [7]. To understand this effect, it is necessary to analyze the dynamics of SCJ interaction with the front and rear armor plates. We assume that the SCJ is subjected only to perpendicular (to it) constituents of the impulse of the region of the plates have undergone interaction with the SCJ.

Let a SCJ element of diameter d_j , moving at velocity V_j , be impacted by an armor plate of thickness δ_i (i is the plate number, i = 1 refers to the front plate, and i = 2 to the rear plate). We denote by θ the angle between the SCJ and the normal to the surface of the plate. We will also assume that the velocity of the plate U is normal to the plate surface. A diagram of the possible interaction of the SCJ with the plate is shown in Fig. 3.

Fig. 3. Diagram of interaction of the SCJ element with the front reactive armor plate.

Let the x axis be directed along and the y axis perpendicular to the SCJ. Then, the plate velocity component along the SCJ U_x and that perpendicular to it U_y are equal in modulus to U_x = U cos θ and U_y = U sin θ.

We denote by τ the characteristic time of interaction of the SCJ element with the plate. As τ we can use, for example, the time of double path of the compression wave through the cross section of the SCJ _τ = 2d_j/c_j (c_j is the velocity of sound in the SCJ material) or the time of contact between the SCJ element and the plate:

τ_i = δ / V_ji cos θ

where V_ji = V_j − (−1)ⁱU/cos θ is the phase velocity of the SCJ–plate contact surface along the SCJ. In the time τⁱ, the jet will interact in the transverse direction with a region of the plate (see Fig. 3) of height U_yτ_i, length δ/cos θ, and width æd_j, where æ is a coefficient that takes into account an increase in the SCJ diameter during the interaction. The component of the impulse I_yi of this plate regions which is perpendicular to the SCJ can be defined as:

I_yi = ρ_plæd_j(δ/cos θ)U_yτ_iU_y,

where ρ_pl is the density of the plate material. It should be noted that for a symmetric element of reactive armor (U1 = U2), the transverse impulses transferred to the SCJ from the front and rear plates are equal in modus I_y1 = I_y2 and opposite in direction. However the mass of the SCJ elements m_ji that receive this impulse depend which of the plates (the front or rear) impacts the SCJ. Indeed, we have

m_ji = ρ_j(πd²_j/4) V_jiτ_i,

where ρ_j is the density of the SCJ material. Since V_j1 > V_j2, then the SCJ element that interacts with the front plate has a greater mass and, hence, a greater length than the rear plate, i.e., m_j1 > m_j2.

As a result of the absorption of the impulse, the I_yi element of the SCJ acquires a transverse velocity equal to

V_jyi = I_yi/m_ji = (4/π)(ρ_pl/ρ_j)(æδ_i/d_j cos θ)(U²_yi/V_ji)

Since m_j1 > m_j2, then in the interaction with the symmetric reactive armor element I_y1 = I_y2, the transverse velocity V_jy1 acquired by the SCJ element during interaction with the front plate will be lower than the transverse velocity V_jy2, caused by the interaction with the rear plate. To elucidate the nature of the SCJ–plate interaction, it is necessary to compare the transverse velocities of the SCJ V_jyi elements with the velocity components of the plates U_yi that are normal to the SCJ. If U_yi V_jyi, then continuous interaction of the plate with the SCJ occurs. If the V_jyi > U_yi, then the SCJ undergoes periodic transverse perturbations that lead to loss of the SCJ–plate contact, so that the interaction itself will also be periodic.

Such conclusions can be drawn by analyzing the velocity ratio

ω_i = V_jyi/U_yi = (4/π)(ρ_pl/ρ_j)(æδ_i/d_j cos θ)(U_yi/V_ji)
= (4/π)(ρ_pl/ρ_j)(æδ/d_j)(U_yi/V_ji)(tan θ)

If ω_i > 1, the SCJ element bounces off the plate and a characteristic transverse wavelike bend is formed in the SCJ. For ω_i > 1, in contrast, continuous interaction of the SCJ with the plate occurs, accompanied by wearing and deflection of the SCJ. The equality ω_i = 1 corresponds to the following relation between the SCJ velocity V_j , the velocity of the plate Ui, the characteristics of the SCJ element and the reactive armor, and the angle θ:

V_j/U_i = (4/π)(ρ_pl/ρ_j)(æδ/d_j)(tan θ) + ((−1)ⁱ/cos θ)

It is interesting to quantitatively analyze this relation for the following values of the constants: ρ_pl = 7.8 x 10³ kg/m³, ρ_j = 8.9 x 10³ kg/m³, θ = 60^O, and æ = 1.2. In this case, the relation becomes the linear relationships

V_j/U_i = 2.32(δ_i/d_j)± 2,

where the plus sign is taken for the rear plate and the negative sign for the front plate. The corresponding diagrams are shown in Fig. 4. The areas above the straight lines correspond to stationary penetration of the SCJ through the reactive armor plate, and the area under the straight lines to nonstationary penetration. It is easy to see that the SCJ penetrates through the front plate of thickness δ₁ = (1.5–2)d_j in an almost completely stationary regime, whereas the rear armor plates of the same thickness are penetrated in a continuous regime only by the high-velocity part of the SCJ.

Thus, the passage of most of the SCJ through the scattering armor plate is shown schematically in Fig. 5.

Fig. 4. Ranges of parameters for which the shapedcharge jet penetration through reactive armor plates is stationary and nonstationary.

Fig. 5. Diagram of shaped-charge jet penetration through scattering reactive armor plates: 1 and 2 are the front and rear armor plates, respectively, 3 is the shaped-charge jet, and 4 is the transverse perturbations in the CJ resulting from the action of the rear plate.

The interaction of most of the SCJ with the front plate has a continuous character and leads to some reduction in its diameter (wearing in the transverse direction) and deflection by a small angle α. Interaction of the SCJ with the rear plate is of nonstationary discrete nature and produces transverse perturbations in the SCJ, which, develop and lead to its distortion and subsequent destruction.

Relations for determining the angle of deflection of the SCJ α and reduction in the diameter of the SCJ can be found in [1]. Assuming that during the SCJ interaction with the front plate, only the transverse component of the impulse of the SCJ changes, the angle α can be estimated from the following relation:

Fig. 6. Angle of deflection of a copper SCJ by a steel front reactive armor plate versus conditions of interaction at θ = 60^O.

Fig. 7. Angle of deflection of a copper SCJ by a steel front reactive armor plate at δ/d_j = 1 versus angle θ of the plate to the SCJ.

Quantitative analysis of the relationship (Fig. 6) shows that under typical conditions of interaction of the SCJ with the front plate, the angle α varies in the range of 1–6^O.

It is interesting to note that the dependence of the deflection angle α on the angle of the element of the reactive armor θ is not monotonic (Fig. 7). With the chosen interaction parameters, α reaches a maximum at θ = 60–70^O. Such behavior of the dependence α(θ) is due to the fact that increasing θ increases not only the impulse of the plate transferred to the SCJ element, but also the mass of this SCJ element.

During nonstationary interaction of the SCJ with the rear armor plate, transverse waves are formed in the SCJ. To describe the evolution of these waves, it is necessary to use a mechanical model of the SCJ. We assume the SCJ to be a string in a plastic state, i.e., stretched by a constant force T = _σs, where σ is the dynamic yield stress of the SCJ material and s is cross-sectional area of the SCJ. In the SCJ, two types of waves can propagate: longitudinal waves with velocity c =

and shear waves with velocity b =

(ε is the longitudinal deformation of the SCJ in front of the shear wave) [8, 9]. With a well-defined yield plate dσ/dε → 0, which is typical, for example, of copper, the velocity of longitudinal waves in the SCJ may be small. The shear wave velocity is estimated at b ≈ 100 m/s. Therefore,
the shear waves arising in the SCJ are localized, and their development leads to rapid destruction of the deformed sections of the SCJ.

CONCLUSIONS

1. Based on the analysis of SCJ interaction with moving reactive armor plates, the lateral impulse transferred to the SCJ by the plates moving at an angle was estimated. The main feature of the interaction of the SCJ with the reactive armor plates is that for the same time interval, the SCJ element interacting with the front plate is of greater length than that interacting with the rear plate.

2. Since the transverse velocity imparted to the SCJ during interaction with the front plate is lower than the plate velocity component perpendicular to the SCJ, the interaction of the main part of the SCJ with the front plate has a continuous nature and leads to wearing of SCJ in the transverse direction (reduction in the SCJ diameter) and a small deflection by an angle α.

3. In the interaction of the SCJ with the rear plate, the transverse velocity acquired by the SCJ exceeds the plate velocity component perpendicular to the SCJ, so that the SCJ periodically bounces off the plate and the interaction of the SCJ with the rear plate has a nonstationary discrete nature and forms transverse perturbations in the SCJ , which, developing, lead to its distortion and subsequent destruction.

4. The transverse perturbations arising in the SCJ can be considered as shear waves in a string in a plastic state. Due to the low velocity of their propagation (≈ 100 m/s), these waves are localized, and their development leads to rapid destruction of the deformed parts of the SCJ.

REFERENCES

1. Partial Issues of Finite Ballistics, Ed. by V. A. Grigoryan et al. (Izd. Mosk. Gos. Tekh. Univ. Baumana, Moscow, 2006) [in Russian].

2. B. V. Voitsekhovskii and V. L. Istomin, “Dynamic Antishaped-Charge Protection,” Fiz. Goreniya Vzryva 36 (6), 87–90 (2000) [Combust., Expl., Shock Waves 36 (6), 754–757 (2000)].

3. M. Mayseless, Y. Erlich, Y. Falcovitz, D. Weihs, and G. Rosenberg, “Interaction of Shaped Charge Jets with Reactive Armor,” in Proc. 8th Int. Symp. on Ballistics, Orlando, Florida, 1984, pp. 5–20.

4. I. F. Kobylkin, V. A. Grigoryan, N. S. Dorokhov, and D. A. Rototaev, “Penetration of Shaped-Charge Jets through Explosive Reactive Armor,” Oboron. Tekh., No. 11, 35–45 (2002).

5. Physics of Explosion, Ed. by L. P. Orlenko (Fizmatlit, Moscow, 2002), Vol. 2 [in Russian].

6. M. Held, “Defeating Mechanisms of Reactive Armour Sandwiches,” in Proc. 22nd Int. Symp. on Ballistics, Vancouver, Canada, 2005.

7. O. V. Svirskii et al., “Investigation of Shaped Charges by Pulsed Protonography,” in An Extreme State of Matter. Detonation. Shock Waves, Proc. Int. Conf. XIII Kharitonov Scientific Readings (VNIIEF, Sarov, 2011), pp. 624–636

8. A. Ya. Sagomonyan, Stress Waves in Continuous Media (Izd. Mosk. Gos. Univ., Moscow, 1985) [in Russian].

9. Strength and Fracture under Short-Term Loads, Ed. by Kh. A. Rahmatullin, E. I. Shemyakin, Yu. A. Dem’yanov, and A. V. Zvyagin (Logos, Moscow, 2008) [in Russian].

Posts · by **Yohannes** » Thu Apr 19, 2018 6:27 pm

Appropriate Coupling Solvers for the Numerical Simulation of Rolled Homogeneous Armor Plate Response Subjected to Blast Loading

Research Article
Hindawi Publishing Corporation
Advances in Mechanical Engineering
Volume 2013, Article ID 637564, 12 pages
http://dx.doi.org/10.1155/2013/637564

AUTHORS: Ahmad Mujahid Ahmad Zaidi,^1,2 Md Fuad Shah Koslan,^1,3 Mohd Zaid Othman,⁴ and Gunasilan Manar⁴

1 Centre for Advanced Armoured Vehicle, National Defence University of Malaysia, 5700 Sg Besi, Kuala Lumpur, Malaysia
2 International College of Automotive, 26600 Pekan, Pahang, Malaysia
3 Royal Malaysian Air Force, Ministry of Defence, 57000 Sg Besi, Kuala Lumpur, Malaysia
4 Faculty of Engineering, National Defence University of Malaysia, 57000 Sg Besi, Kuala Lumpur, Malaysia

Correspondence should be addressed to Md Fuad Shah Koslan; fuadpostgrad@gmail.com
Received 30 July 2013; Revised 21 October 2013; Accepted 21 October 2013
Academic Editor: Nao-Aki Noda

Copyright 2013 Ahmad Mujahid Ahmad Zaidi et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

1. Introduction

In the blast phenomena, interaction between fluid and structure, also called fluid-structure interaction (FSI), normally will occur, and there is no single method that can be used for all conditions in FSI analysis [1]. The governing partial deferential equation for FSI model needs to be solved in both time and space domain with the basic physic principles involving the conservation of mass, momentum, and energy. The solution over the time domain can be achieved by an explicit method [2]. It can be obtained by utilizing different spatial discretization such as Lagrange, Euler, and Arbitrary Lagrangian Eulerian (ALE) or mesh-free method also known as Smooth Particle Hydrodynamic (SPH) methods [1]. However, the basic solvers for explicit integration numerical wavecodes (sometimes termed "hydro-codes") can be utilized as an outline with their associated strengths and weaknesses [3]. Air, plate, and trinitrotoluene (TNT) are three different domains in this model analysis. Each domain has a solver that is suitable to be used. The numerical solver be used in AUTODYN generally fall into the following methods which are Lagrange, Euler, ALE, and SPH methods. With intelligent selection of suitable solver for various regimes, an optimal solution in terms of accuracy and efficiency can be achieved. The appropriate solvers in AUTODYN will be carried out, and the effect on a final result will be discussed further.

Table 1: Lagrangian solver summary [3].

2. Method of Discretization

Numerical analysis needs to be subdivided into problem domain analysis to the nodes, elements, and spatial discretization. The spatial discretization is performed by representing the fields and structures of the problem using computational point in space, and connected with each other through a computational grid. Usually, the fine grid will lead to more accuracy of the result. The most popular spatial discretization has been widely used are Langrage, Euler, ALE and SPH and are provided in the AUTODYN computer code.

2.1. The Lagrange Solver. Lagrange’s equation has been use to formulate the equation of motion, which represents the response of structure subjected to any external load. Simple Lagrange’s governing equation can be derived by considering conservative system, where all internal forces have a potential energy and Lagrangian discretization scheme is usually a based on the finite element method.

In the Lagrangian scheme, the elements (and hence the nodes) move with the material, while in the Eulerian scheme the nodes are fixed in space, but the material is allowed to flow as described in [3]. Thus, the characteristics make the Lagrangian scheme more suitable for modeling solid materials, while the Eulerian scheme is more suitable for modeling fluid materials. The advantages and limitations of the Lagrangian solver are as summarized in Table 1.

2.2. The Euler Solver. The Euler method of space discretization as described in [3], where the numerical meshes or grids are fixed in space and the physical material flows through the mesh. It is typically well suited for the description of the material behavior of large deformation or flows situation and consequently, by definition, it does not result in grid distortion due to the fixed grid scenario.

Most of the time, the Euler is usually used for representing fluids and gases. However, to describe for solid behavior, additional calculations are required to transport the solid stress tensor and the histories of the material through the grid.The advantages and limitations of the Eulerian solver are as summarized in Table 2.

Table 2: Eulerian solver summary [3].

2.3. The Arbitrary Lagrange-Euler (ALE) Solver. The ALE is a special solver that incorporates Lagrange and Euler solvers in a single governing equation and provides a full coupling between the blast wave and the structural response. The interaction of solid-type material in Lagrangian and fluidtype material in Eulerian generated a FSI.

In the differential form of the conservative equation of mass, momentum and energy are more readily obtained from the corresponding well-known Eulerian form than from the ALE form; the conservative equation is to be replaced in the various conservative term, such as material velocity, mass density and stress tensor of material.

The ALE solver, also allows for “automatic rezoning,” which can be quite useful for certain problems. With these features, this solver is perhaps appropriate for modeling solid, fluid, and gas. Thus, generally, this solver is suitable for a variety of fluid-structure interaction problem analyses. Baylot and Bevins [4] had performed the analysis of blast wave propagation using the ALE solver and the structure response in the Lagrangian scheme simultaneously. However, like any other discretization schemes, the ALE solver also has its advantages and limitations as summarized in Table 3.

Table 3: ALE solver summary [3].

2.4. The Smooth Particle Hydrodynamic (SPH) Solver. Based on the Lagrangian space discretization scheme, SPH was developed as an alternative method without any nodes or grid approaches or also known as a meshless method. The SPH solver was initially used in astrophysics [5]. It had the potential to be efficient and accurate in material deformation as well as flexible in terms of the inclusion of specific material
models.

The idea behind SPH is to replace the equations of fluid dynamics by equations for particles. The SPH particle is not only interacting mass point but also interpolation point used to calculate the value of physics variables based on the data from neighboring SPH particle, scaled by the weighting function. Since there is no grid defined, the SPH method does not suffer from the normal problem of grid tangling in large deformation problem.

However, according to Quan et al. [1] the SPH method requires a sort of the particle in order to locate the current neighboring particles, which makes the computational times per cycle more expensive than the mesh based on Lagrangian technique. This can make the SPH method less efficient than the Lagrange’s method.

3. Validation of the Few Solvers

The appropriate solvers discussed in Section 2 were validated against one of the series of blast test experiments performed by Neuberger et al. [6].The experiment consisted of a circular target plate with a diameter of 1000 mm that was clamped with two thick armor steel rings, tightened together with bolts, fully clamped, and subjected to 8.75 kg of TNT located at 200 mm standoff distance. The experimental setup is as
shown in Figure 1.

Figure 1: The experimental setup [6].

The experimental setup as shown in Figure 1 will be modeled in AUTODYN 3D and the equivalent simulation model was developed as shown in Figure 2.

Figure 2: A three-dimensional simulation model created in AUTODYN 3D.

The simulation was conducted by using a high speed computer processor Intel Core i7-2600K CPU@ 3.40 GHz (8 CPUs), 3.7 GHz. The model was created in a domain size of (500 × 500 × 500) mm with 5 mm element size defined as air and symmetrical in the "X" and "Z" axes. A quarter symmetry model was used in the simulation in order to reduce the computational time.

The same mechanical property of the RHA material in [7] was used in the simulation model as shown in Table 4 where t is plate thickness, o_y is yield stress, o_UTS is ultimate tensile stress, c_L is strain, E is Modulus Young, v is Poisson’s ratio, and A, B, n, c, and m are material dependent parameters, and may be determined from empirical fit of flow stress data.

Table 4: Material properties for RHA [7]

There are four solvers that will be used to calculate interaction between TNT explosive and the RHA plate with respect to the EoS for RHA and TNT material. Johnson and Cook EoS has been selected to represent RHA material as following:

where A, B, C, m, and n are constant, c is equivalent plastic strain, (1/c)^O is the strain rate nondimensionalized by reference strain rate of 1/s, and T^O is nondimensional temperature. Parameter A, the initial (c = 0) yield strength of the material at a plastic strain rate of c = 1/s ̇ at the room temperature (298 K), is modified by a strain-hardening factor (containing parameter, B and n), a strain-rate-hardening factor (containing parameter C), and a thermal-softening factor (containing parameter m), while T^O, defined by

where T_r is room temperature and T_m is the melting temperature of material, 1783 K for RHA. Equation (2) is the form used in United States Army Research Laboratory (ARL) [8] and is valid for T_r < T < T_m, the region of interest in most blast and ballistics application.

For the blast application, this paper used Jones Wilkins Lee (JWL) E_oS to represent TNT blast loading. The JWL EoS has been used by AUTODYN to accurately describe the pressure-volume-energy behavior of the detonation product as follows:

where V is the volume of detonation product or, P is pressure in megabar, and E is the energy in Mb cc/cc, while A, B, R₁, R₂ and w are constant parameters based on the type of explosive. For TNT, it was identified by Lee et al. [9], that A = 5.242, B = 0.07678, C = 0.01082, R₁ = 4.2, R₂ = 1.10 and w = 0.30.

4. Results and Discussion

4.1. Interaction of Coupling Solver. TNT as an explosive material will produce a blast force that propagated through the air as a medium and hit the RHA plate as the target object. This phenomenon involved the interaction between fluid and structure.

Figure 3: The diagram interaction between solvers to represent TNT and RHA plates.

The analysis started when TNT as high-explosive material detonated, it interacted with the surrounding which were defined as an air ideal gas. Air as a medium containing the explosive charges striked the RHA plate. The TNTand plate domain interacted. In the numerical simulation,
interaction involved two or more type, domains using the same or different type of solvers. Three material variable, that is; air, TNT, and RHA plate, were involved in the case response of RHA plate subjected to blast loading. For air, the medium has been set according to Chung Kim Yuen
et al. [10] had established the Ideal Gas Euler solver which was the most appropriate solver for air (Table 5). In this paper, the appropriate solver for TNT and RHA plate solver will be determined. There are four solvers for TNT that had been considered as follows Lagrange, ALE, Euler, and SPH while the two solvers for RHA plate; Lagrange and ALE. The interaction diagram between TNT and RHA plate is as illustrated in Figure 3. The appropriate solver interaction will be used in future analysis.

Table 5: Properties of air [10].

4.2. The Euler-Lagrange Solvers Interaction. In this type of interaction, it was involved in the interaction between RHA plates represented by the Lagrange solver, and TNT represented by Euler solver. This interaction was also called EulerLagrange coupling, and was commonly used to simulate an interaction between fluid and structure domains.

Simultaneous analysis in the numerical simulation, for both domains; Lagragian grids provide the geometry constraint in order to allow the material to flow in the Eulerian grids. At the Euler-Lagrange interface, the Lagrange grids act as structure geometry inside the flow boundary, while the Euler grids will provide a pressure or heat boundary to the Lagrangian domain. As the Lagrange’s grids move or distort the effect of interaction will produce the deflections of a structure as illustrated in Figure 4.

Figure 4: The Euler-Lagrange interaction process.

In the finite element analysis (FEA), the Euler solver is commonly used for fluid analysis by using the finite volume method (FVM) or the finite difference method (FDM) discretization scheme, and the solution involves large relative deformation, whereas the Lagrange used the finite element method (FEM) discretization scheme for structure analysis for a small relative deformation. Both solvers provide different rates of deformation and will lead to the smalltime step. This problem frequently happens in the numerical simulation. Coupling solver for this case showed the plate deflection stopped at 0.15 ms due to the small-time step as shown in Figure 5.

Figure 5: RHA deflection captured using Euler Lagrange interaction before computation was stopped due to the small time step.

However, according to Birnbaum et al. [3], for two dimensional (2D) cases, Euler-Lagrange coupling showed very powerful and stable coupling for FSI problem, whereas in threedimensional (3D) cases, the computational requirement time was excessive even on supercomputers. This stems from the substantial and complex intersection calculation that must be performed for each time step. In this paper, the coupling Euler-Lagrange solvers were unable to produce a good results when compared with the experimental data. Thus, other solver interactions should be considered.

4.3. The Smooth Particle Hydrodynamic (SPH) Interaction. Previous numerical simulations used SPH solver to see the trajectory particle on the impact analysis and commonly was used for the porous material such as soil, sand and water. Toussaint and Durocher [11], Barsotti et al., [12] and Quan et al., [1] used SPH solver to define for sand and Campbell and Vignjevic [13] for water.

In this paper, numerical simulation will be involved in the interaction between RHA plate and TNT with respect to the SPH, Lagrange, and ALE solvers. The case is as follows;

(a) RHA plate with Lagrange solver and TNT with SPH solver;
(b) RHA plate with ALE solver and TNT with SPH solver.

The limitations of Euler-Lagrange coupling in AUTODYN 3D simulation lead to explore on coupling for another solver interaction to obtain the promising result. It should also evaluate the coupling between the mesh-less technique (SPH) representative of TNT and the traditional Lagrange and ALE solvers representative of RHA plate. By using the SPH method, TNT was discretized as the continuum through a set of the nodes without the connective mesh. The node or element assumed a physical meaning; they represent material particle carrying properties such as pressure and thermal and impacted the RHA plate as shown in Figure 6 representing case (a) and Figure 8 representing case (b). The deformation results of the RHA plate for both cases (a) and (b) are as shown in Figures 7 and 9. The approximate final deformation results had large discrepancies when compared with experimental data. Thus, the SPH-Lagrange coupling was not the appropriate solution for this particular case.

Figure 6: The RHA plate (Lagrange)-TNT (SPH) interaction process

Figure 7: Case (a) RHA plate response to Lagrange solver and TNT to SPH solver.

Figure 8: The RHA plate (ALE)-TNT (SPH) interaction process.

Figure 9: Case (b) RHA plate response to ALE solver and TNT to SPH solver.

4.4. The Lagrange and ALE Interaction (RHA Plate and TNT Represented by ALE/Lagrange Solver). This simulation will involve the four interaction cases between RHA plate and TNT with respect to the Lagrange and ALE solvers. The cases are as follows:

(a) RHA plate with ALE solver and TNT with ALE solver;
(b) RHA plate with Lagrange solver and TNT with Lagrange solver;
(c) RHA plate with Lagrange solver and TNT with ALE solver;
(d) RHA plate with ALE solver and TNT with Lagrange solver (Figure 17).

A surface on Lagrange’s domain interacted with another surface of a different domain allowing for impact and sliding, gap opening and closing, and mesh distortion between the bodies. The interaction between TNT and RHA plate using Lagrange and ALE element for all cases found was a similar pattern on the grid distortion on the RHA plate as shown in Figures 10, 12, 14, and 16. Consequently, the predictions on the final deflection of the RHA plate were in good agreement with all cases on the Lagrange and ALE interaction as shown in Figures 11, 13, and 15 despite differences in the computation time taken and cycle interaction as shown in Table 6.

Figure 10: The RHA plate (ALE)-TNT (ALE) interaction process.

Figure 11: Case (a) RHA plate response to ALE solver and TNT to ALE solver.

Figure 12: The RHA plate (Lagrange)-TNT(Lagrange) interaction process.

Figure 13: Case (b) RHA plate response to Lagrange solver and TNT to Lagrange solver.

Figure 14: The RHA plate (Lagrange)-TNT(ALE) interaction process.

Figure 15: Case (c) RHA plate response to Lagrange solver and TNT to ALE solver.

Figure 16: The RHA plate (ALE)-TNT(Lagrange) interaction process.

Figure 17: Case (d) RHA plate response with ALE solver and TNT with Lagrange solver.

However, taking into account the computation time taken, cycles of iterations had been obtained and consolidated in Table 6 and deformation results are as shown in Figure 11, interaction using ALE solver for both material TNT and RHA plate showed an appropriate interaction selection in this case. The deformation result shown in Figure 11 was in very good agreement with when compared with Neuberger et al. [6] experimental data.

5. Conclusions

Different numerical solvers have certain advantages and limitations. It is critical that analyst must understand which solver or combination of solver is appropriate to be used in a particular case or problem of interest.

Use of computer code or software package intelligence is considered as a key in order to obtain the good prediction result. No doubt, the coupling techniques are extremely powerful for any FSI analyses, although they have inherent limitations that must be recognized. The solver investigation
must be done before the appropriate solver is obtained for a particular case. However, the suitable coupling solver can only be selected after the analyst performed numerical trial and validation against experiment.

This paper presents the appropriate solver coupling of a RHA plate subjected to blast loading and has been compared with Neuberger et al. [6] experimental data. For this case, it was found out that using the ALE solver to represent both the TNT and RHA plate in the numerical simulation, managed to produce good agreement with the experimental test data. Thus, the ALE solver will be chosen as the coupling solver to simulate similar cases for future analysis.

Acknowledgments

Financial assistance from the Royal Malaysian Air Force (RMAF) towards this research is hereby acknowledged. Opinions and views either directly or indirectly from various parties, including the staff and lecturers from the Engineering Research Center of National Defense University (UPNM) are greatly appreciated.

References

[1] X. Quan, N. K. Birnbaum, M. S. Cowler, B. I. Gerber, R. A. Clegg, and C. J. Hayhurst, “Numerical simulation of structural deformation under shock and impact loads using a coupled multi-solver,” in Proceedings of the 5th Asia-Pacific Conference on Shock and Impact Load on Structures, Hunan, China, November 2003.

[2] J. C. Tannehill, D. A. Anderson, and R. H. Pletcher, Computational Fluid Mechanic and Heat Transfer, Taylor & Francis, 1798.

[3] N. K. Birnbaum, J. Nigel Francis, and I. Bence Gerber, “Coupled techniques for the simulation of fluid-structure and impact problem,” Century Dynamic, 2003, http://hsrlab.gatech.edu/ AUTODYN/papers/paper63.pdf.

[4] J. T. Baylot and T. L. Bevins, “Effect of responding and failing structural components on the airblast pressures and loads on and inside of the structure,” Computers and Structures, vol. 85, no. 11–14, pp. 891–910, 2007.

[5] R. A. Gigold and J. J. Monaghan, “Smoothed particle hyrodynamics: theory and application to non spherical star,” Monthly Notices of the Royal Astronomical Society, vol. 181, pp. 375–389, 1977.

[6] A. Neuberger, S. Peles, and D. Rittel, “Springback of circular clamped armor steel plates subjected to spherical air-blast loading,” International Journal of Impact Engineering, vol. 36, no. 1, pp. 53–60, 2009

[7] A. Neuberger, S. Peles, and D. Rittel, “Scaling the response of circular plates subjected to large and close-range spherical explosion—part I: air-blast loading,” International Journal of Impact Engineering, vol. 34, pp. 859–873, 2007.

[8] H. W. Meyer Jr. and D. S. Kleponis, An Analysis of Parameter for the Johnson-Cook Strength Model for 2-in-Thick Rolled Homogenous Armor, Weapon and Materials Research Directorate Army Research Laboratory, 2001.

[9] E. Lee, M. Finger, and W. Collins, JWL Equation of State Coefficients for High Explosives, Lawrence Livermore Laboratory Calofornia, 1973.

[10] S. Chung Kim Yuen, G. S. Langdon, G. N. Nurick, E. G. Pickering, and V. H. Balden, “Response of V-shape plates to localised blast load: experiments and numerical simulation,” International Journal of Impact Engineering, vol. 46, pp. 97–109, 2012.

[11] G. Toussaint and R. Durocher, “Finite element simulation using SPH particles as loading on typical light armoured vehicles,” in Proceedings of the 10th International LS-DYNA User Conference, East Lansing, Mich, USA, June 2008.

[12] M. A. Barsotti, J. M. H. Puryear, D. J. Stevens, R. M. Alberson, and P. McMahon, “Modeling mine blast with SPH,” in Proceedings of the 12th International LS-DYNA User Conference, East Lansing, Mich, USA, June 2012.

[13] J. C. Campbell and R. Vignjevic, “Simulation structural response to water impact,” International Journal of Impact Engineering, vol. 49, pp. 1–10, 2012.

Posts · by **Yohannes** » Thu Apr 19, 2018 9:02 pm

Thermoelectric waste heat recovery from an M1 Abrams tank

Journal page: SPIEDigitalLibrary.org/conference-proceedings-of-spie

Thermoelectric waste heat recovery from an M1 Abrams tank

AUTHORS: C. David Stokes, Peter M. Thomas, Nicholas G. Baldasaro, Michael J. Mantini, Rama Venkatasubramanian, et al.

C. David Stokes, Peter M. Thomas, Nicholas G. Baldasaro, Michael J. Mantini, Rama Venkatasubramanian, Michael D. Barton, Christopher V. Cardine, Grayson W. Walker, "Thermoelectric waste heat recovery from an M1 Abrams tank," Proc. SPIE 8377, Energy Harvesting and Storage: Materials, Devices, and Applications III, 83770N (25 May 2012); doi: 10.1117/12.920804

Event: SPIE Defense, Security, and Sensing, 2012, Baltimore, Maryland, United States of America
RTI International, 3040 Cornwallis Rd., Research Triangle Park, NC, USA 27709-2194;
Creare, Inc., 16 Great Hollow Road, Hanover, NH, USA 03755-3116; c
General Dynamics Land Systems, 38500 Mound Road, Sterling Heights, MI, USA 48310-3200;
Old Dominion University, 135 Kaufman Hall, Norfolk, VA, USA 23529

*dstokes@rti.org; phone 1 919 485-5544; fax 1 919 541-6515; http://www.rti.org
Energy Harvesting and Storage: Materials, Devices, and Applications III, edited by Nibir K. Dhar, Priyalal S. Wijewarnasuriya, Achyut Dutta, Proc. of SPIE Vol. 8377, 83770N · © 2012 SPIE · CCC code: 0277-786X/12/$18 · doi: 10.1117/12.920804

1. INTRODUCTION

The M1 Abrams, powered by the AGT-1500 turbine engine, is the main battle tank for the United States Army, National Guard, and Marine Corps (see Figure 1). It is produced by General Dynamics Land Systems Division (GDLS) and has been in service since 1980. With its current alternator, the Abrams tank (like most fielded vehicle platforms) is approaching its output limit, and there is a strategic need for more power to support additional electronics on the tank. A complete redesign of the vehicle’s electric power plant would entail some significant challenges. Simply changing the alternator requires the transmission of the tank to be redesigned and further reduces the shaft horsepower available at the sprockets for mobility. Changing the engine would be a major effort that would not be cost effective. We propose an alternative solution to capture and convert waste heat from the engine exhaust into electric power for use on the tank.

Recovering waste heat from the M1 Abrams tank engine has several potential benefits. The proposed thermoelectric (TE) waste heat recovery system would replace the existing exhaust duct with a customized design that captures a portion of the waste heat from the gas turbine engine. Because this energy is already available, it requires no extra fuel or energy from the engine. The system is designed to be easily retrofitted into existing tanks without requiring any modification to the engine or transmission. With the increased budgetary constraints imposed on DOD agencies and their suppliers, this solution provides a means to achieve a higher level of functionality from existing platforms. Additional advantages of the waste heat recovery system would include lowering the thermal signature of the exhaust, achieving higher system efficiency, and potentially reducing system maintenance.

Figure 1. Photo of M1 Abrams tank in motion.

2. PROTOTYPE DESIGN AND MODELING

With support from the Army Research Lab (ARL), RTI International, General Dynamics Land Systems (GDLS), and Creare, Inc., have developed and demonstrated a prototype robust energy harvesting solution that converts residual thermal energy from an M1 Abrams tank exhaust into useable electric power. The RTI waste heat recovery system captures heat from the exhaust of the Honeywell/Lycoming AGT-1500 turbine engine, converts this heat into electrical power with a thermoelectric generator, and dissipates the heat through a Creare-designed heat rejection system. The prototype waste heat recovery system for the M1 Abrams tank engine includes components for exhaust heat capture, thermoelectric (TE) energy conversion, heat rejection, and system integration. The TE waste heat recovery system prototype was designed to replace part of the existing exhaust duct with custom components to capture a portion of the waste heat from the gas turbine engine. The fractional-scale prototype was designed so that the full scale system could easily retrofit existing tanks without requiring any modification to the engine or powertrain.

2.1 Exhaust heat capture design and modeling

An analytical model was developed to evaluate the power generation potential for the prototype with heat capture fins in the exhaust stream with a liquid-cooled heat exchanger. The analysis suggested that by using a finned heat exchanger placed in the hot exhaust stream, one can flow approximately 4.6 kW of heat through the TE devices and into the cooling liquid kept at roughly 80° C. This model assumed a negligible thermal resistance value for the fin plate-to-coolant portion of the thermal circuit due to the high heat transfer coefficients associated with forced liquid cooling. The analysis also assumed a constant ǻT between the exhaust and coolant due to 1) the relatively small fraction of heat removed from the tank exhaust stream and 2) the large heat capacity of the coolant flow. With reasonable assumptions for the nominal exhaust flow velocity, a Prandtl number of 0.73 and Reynolds number of 230,000 were calculated. Modeling the fins as infinite flat plates experiencing forced convection in laminar flow and using the relation given in equation 1,

lead to a calculated heat transfer coefficient of 54 W/m² K and a total finned thermal resistance of 0.024 K/W. The model predictions of 230 ^OC and 80 ^OC for the TE hot and cold sides were in excellent agreement with the experimental measurements from the demonstration, matching them to within 10%.

The waste heat recovery prototype was designed to replace the alternator access panel in the rear exhaust duct. This allowed the installation and testing of the prototype without requiring any modification to the tank exhaust. A finned heat exchanger was developed to capture heat from the exhaust flow, and a compliant support structure was used to sandwich the TE devices between the hot-side heat exchanger and the cold-plates. Figure 2 shows a CAD rendering of our design.

Figure 2. CAD rendering of exhaust waste heat recovery system design.

The prototype consists of two heat exchangers for the energy harvesting system: a finned hot exhaust heat exchanger and two spring-loaded liquid-cooled cold plate heat exchangers. Sixteen TE devices were mounted on the two cold plate heat exchangers (see Figure 3). The entire assembly was mounted to a support plate that replaces the rear alternator access panel on the tank.

Figure 3. One of the two cold plates with 8 Bi₂Te₃ 127 couple devices mounted on it.

2.2 Thermoelectric energy conversion

Sixteen Bi₂Te₃ 127-couple TE devices were used for the prototype waste heat recovery system. These lower temperature TE devices were used for this application because the rear access panel area of the exhaust duct on the Abrams tank is not directly in the exhaust stream and could only capture a fraction of the heat. The thermoelectric devices were designed to have a thermal resistance of 0.65 K/W so that a delta T of ~180 K would be achieved across the devices during testing. The devices were arranged in two series connected arrays with eight (8) devices each.

2.3 Heat rejection system design

Creare designed, assembled, and tested a heat exchanger system to reject 1200 Wth of heat using two liquid-cooled coldplates, radiators, pumps and fans. Table 1 summarizes the design parameters and validated measurements for the system. The as-tested heat rejection system consisted of two parallel systems (2400 Wth), each with a single pump, two fans, cooling block, and radiator. The efficiency of the separate components was evaluated individually and then combinations of pumps and fans were examined to arrive at an optimal arrangement. Figure 4 shows the test setup for characterizing radiator and fan performance.

Table 1. Creare heat rejection system design targets and validated performance values for the tank waste heat recovery system.

To begin, a feasibility demonstration was conducted by assembling and bench-testing the balance of plant (BOP) components as an integrated system sized for 1200 W_th dissipation to simulate the heat rejection for a subscale exhaust waste gas TEG power system. To evaluate the performance, the overall thermal resistance was measured as a function of input power. For this system, the thermal load requirement was 1200 W_th for a total input power less than 10 W_e to achieve an overall thermal resistance below 0.0542 K/W_th. Figure 5 shows the results of the feasibility demonstration. The heat rejection system was able to exceed the target goal and achieve a total thermal resistance < 0.04 K/W_th.

Figure 4. Creare radiator and fan performance characterization test setup.

Figure 5. Feasibility demonstration for the heat rejection system.

3. PROTOTYPE DEMONSTRATION

3.1 Prototype setup and installation

A team from RTI and Creare traveled to the GDLS test facility in Sterling Heights, MI to test the prototype waste heat recovery system. The first step was to complete the installation of the heat exchanger/TE generator system (HEX/TEG) on the Abrams tank engine exhaust. This included installation of the TEG assembly in the rear of the tank and installation of the heat rejection system on the body of the vehicle.

The two cold plates with TE devices mounted on them were prepared for assembly to the hot-side heat exchanger by coating the devices with a thin layer of liquid metal to insure good thermal connection between the hot side HEX and the TE devices. Once the cold plates were mounted on the hot side HEX, they were carefully tightened down with calibrated spring force on each corner of the cold plate to put pressure on the cold plate/TE devices/ hot side HEX sandwich to insure good thermal contact and heat flow. The cold plates were then connected to the heat rejection system at the back of the tank where the HEX/TEG was mounted in place of an access plate at the rear of the exhaust shroud. Figure 6 shows photos of the waste heat recovery system integrated into the exhaust of the M1 Abrams tank. In this application, the rear access plate was replaced to allow testing without modification to the vehicle.

The heat rejection radiators and fans were mounted on top of the vehicle with four water lines running through the engine compartment to connect to the waste heat recovery system. Thermocouple and power leads from the rear access panel were routed through the engine compartment to the top surface of the vehicle and into the turret to facilitate measurement while operating the vehicle on the test track. Figure 7 shows a picture of the heat rejection radiators mounted on top of the vehicle.

Several instruments were used for the demonstration. Measurements were made using an Omega data acquisition system for the thermocouples. Two portable electronic loads were used to measure the voltage, current, and power from the two TE array banks. Data from the instruments was recorded using two laptops, and controls for the radiator fans and pumps were also located inside the turret.

Figure 6. Waste heat recovery prototype system installed into Abrams tank exhaust duct: a) rear view of tank with plate installed b) close up view of the exhaust heat capture fins.

Figure 7. Creare designed heat rejection system.

3.2 Prototype test results

The RTI and Creare team set up instruments to measure the power generation and temperatures from inside the tank with GDLS providing a driver to operate the vehicle. A 40-minute test drive of the M1 Abrams tank was completed to demonstrate operation of the exhaust waste heat recovery system. Figure 8 shows a plot of the temperature measurements made throughout the test, and Figure 9 shows the power measurements from the two thermoelectric arrays. Table 2 provides a summary of the TE power generation under three different operating regimes. Peak power production was measured at 80.7W from two banks of TE devices mounted to the rear access plate. This was measured during the low-gear drive to the test track. The TEG with the tank idling produced 57.5W, and 10 minutes after engine shut down the power produced was 11.4W from latent heat in the engine. One bank of TE devices (Left side) developed an electrical shunt 22 minutes into the test track run. This shunt was caused by movement of some liquid-metal thermal interface material from the hot-side surface to the side of a TE device. This shunt bypassed ~4 of the 8 TE devices in the left bank, cutting the power production by half. However, full performance of the array was restored after removal of the shunt. The compliant integrated assembly successfully protected the components, and the TE devices survived the 40-minute demonstration without any measureable degradation.

Figure 8. Temperature measurements from M1 Abrams tank waste heat recovery demonstration.

Figure 9. Raw TE power measurements from the demonstration.

Table 2. Summary of power generation from the M1 Abrams tank under various operating conditions.

4. CONCLUSIONS

The successful demonstration of over 80W of power from the exhaust heat of an M1 Abrams tank has set the stage for developing a full-scale system to recover waste heat from the vehicle. Through continued discussions with GDLS and Creare, RTI has examined the power generation potential for a full-scale TE waste heat recovery system. Using nominal exhaust flow and temperature parameters we have developed an analytical model to explore the capability of a full-scale system. The TEG system would be designed so that it could be easily retrofitted into existing vehicles. This provides a significant upgrade to the electrical power for the tank without requiring a redesign of the engine, transmission, or alternator and without impacting the fuel economy of the vehicle. With increased budgetary constraints imposed on DOD agencies and their suppliers, this solution provides an effective means to achieve a higher level of functionality from existing platforms. In addition, the waste heat recovery system would lower the thermal signature of the exhaust.

ACKNOWLEDGEMENTS

The authors would like to acknowledge funding and support from the Army Research Lab (ARL/Honeywell Contract No. 4202361403E, Patrick Taylor, COTR).

by **Lamoni** » Fri Apr 20, 2018 5:33 am

Fuel Cell Technologies for Defence Applications

J. Narayana Das

Abstract

By virtue of their distinct features like autonomy, low signatures, no emissions, high speciﬁc energy etc., fuel cells could ﬁnd a number of applications in the defence sector. Depending on the speciﬁc context, the requirement could be for a wearable, portable or distributed power supply. Powering unmanned aerial vehicles, ground vehicles and autonomous underwater vehicles form a separate regime. High levels of efﬁciency, reliability, reproducibility, robustness to meet the MIL standard environmental tests etc., are the prerequisites for military hardware. The salient features of fuel cells are touched upon in this context and design approach for a fuel cell based AIP system for submarines is discussed in brief.

Keywords

Fuel cells * Soldier power * Portable power * Auxiliary power * Unmanned vehicle propulsion * AIP for submarines

1 Introduction

Defence forces look forward to self sufﬁciency in every situation and location. Power and energy supply must be robust, reliable and versatile. Batteries of several types and speciﬁcations have been specially engineered and are in wide use by the Army for their forward area detachments and by the Air force and Navy for autonomous vehicles and remote operation ﬁelds. The concept of fuel cell demonstrated by Dr. William Grove in 1839, has undergone numerous innovative up gradations and has got adapted and diversiﬁed into several types. Low acoustic signature, low thermal signature, practically no chemical emission, improved speciﬁc energy, high energy density, reduced recharging cycle times etc., are important features offuel cells weighing against the best of battery choices, as far as the military segment is concerned. These features are of signiﬁcance to the civil sector as well. Still the industry has not been able to penetrate the market, to the extent it should have; despite the global green energy campaign. According to a 2015 review, cumulative installed capacity of fuel cells since 1995 is just about 1 GW [1] and the shipment of units forecast for 2015 is around 160,000, all categories included.

Batteries and Fuel cells are both power on demand devices based on electrochemical energy conversion. In batteries the stored chemical energy is released as electrical energy as a result of reactions between the electrodes and the electrolyte. Once the reactants are consumed, the battery stops delivering power and needs to be recharged using electrical energy from external sources. In fuel cells, though electrical energy is generated through electrode reactions, the reactants per say are not stored in the cells and can continue to give rated power output as long as supply of the fuel and oxidant could be maintained.

Polymer electrolyte fuel cell (PEFC) uses a solid polymer electrolyte membrane for exchange of the H+ ion facilitating the anode and cathode reactions of Hydrogen and Oxygen, using Platinum and alloy catalysts. Relatively lower service life, stringency of material speciﬁcations and the need for extremely high purity of Hydrogen, etc., are the limiting factors. Solid oxide fuel cells (SOFC) are made up of ceramic and cermet electrodes and electrolyte systems such as Yttria Stabilized Zirconia (YSZ). They are robust in nature. No gel or liquid or polymer membrane is involved. However, the engineering challenges are several, since the operating temperatures are as high as 800–1000 °C. Handling Hydrogen gas at such high temperatures is a safety critical issue. In stationary systems SOFC is ﬁnding large scale application, primarily due to the ﬂexibility in fuel choice. Ni-cermet anode used in SOFCs, has very poor sulfur tolerance below 800 °C. To be successful in automobiles, PEFC system must operate at 110–120 °C, which causes associated performance and degradation issues [2].

In Phosphoric acid fuel Cell (PAFC), phosphoric acid spread over a porous supporting substrate forms the basic electrolyte layer. Platinum and platinum alloys on Carbon form the catalysts. Handling the corrosive acid, maintaining its concentration and choice of acid resistant materials are the critical engineering challenges. Though the operating temperature is higher in comparison with PEFC, better tolerance to impurities in the reactants is a speciﬁc advantage. Overall power to weight ratio is lower than that of PEFC. But PAFC has much longer service life. Alkaline Fuel Cells (AFC) have the fastest kinetics. The electrode support is typically Ni mesh or foam. The separator media is alkali (typically KOH) soaked asbestos membrane. Such systems can use metallic bipolar plates, and thereby reduce cost. However, their vulnerability to poisoning by CO2, corrosion of the electrodes, dilution of alkali in the cell etc., are issues that restrict the use of AFC. Molten Carbonate Fuel Cells (MCFC) use fused alkali carbonates as primary electrolyte and bipolar plates are of metal alloys. Since operating temperature is high, faster kinetics is possible. Major problem encountered is the corrosion of electrodes.

In all of the above cases hydrogen fuel is supplied externally, either as stored gas/liquid, or through reforming hydrocarbons. Alternately decomposition of peroxides is also practiced as an option. In Direct Fuel Cells (DFC) hydrogen containing materials like methanol or sodium borohydride is directly used as the fuel, rather than using an external fuel processor or reformer, to generate hydrogen online. Such fuel cells are more compact though they suffer from much slower kinetics.

2 Potential Defence Applications

Simplicity, durability, ruggedness and high level of autonomy, are essential features of any military hardware. Systems should have fault diagnostics and self protection features. Detailed maintenance plans, mean time between failures (MTBF), mean time to repair (MTTR), etc. are important statistical parameters of interest to the military customer.

Air force bases in forward areas and remote locations need assured electric power for battery charging, auxiliary power for surveillance and regular power for communication equipment. Long endurance unmanned aerial vehicles also need agile power sources. Navy’s strategic need of electrical power is for running the unmanned underwater vehicles and air independent propulsion systems for non nuclear submarines. The land forces cannot be conﬁned to pre chartered ﬁelds and terrains. Power supply would have either been destroyed or had never existed in the new posts they occupy. Based on the typical operating environment and user perspectives, army’s power requirements can be classiﬁed into soldier power, auxiliary power units (APU), autonomous systems, distributed power plants etc. [3]. US department of defence has carried out a comprehensive study and has identiﬁed the distinct areas, as soldier wearable and portable power, auxiliary power units for ground vehicles, ships, and aircrafts, non-tactical light-duty vehicles, propulsion power for ships, submarines, autonomous underwater vehicles (AUVs) and unmanned aerial vehicles (UAVs) [4].

2.1 Soldier Power

Portable high density power source is vital for the modern war ﬁghter to meet his C4I system needs. Apart from the communication equipment, power is required for helmet mounted displays, mobile computer, data modems etc. Primary challenge is to keep the system weight low. High temperature PEFC with methanol reformer is a choice for 25–55 W systems. Portable model JENNY 600 S from M/s. SFC Energy can directly power electrical devices or recharge secondary batteries. The fuel cartridge contains methanol and has a capacity of 400 Wh each. Ultracell has been able to pass several models of their reformed methanol fuel cells (RMFC) through the rigorous test procedures of the US army. PEFC based prototypes developed by M/s Ballard power systems is using sodium borohydride as the primary fuel. Naval Materials Research Laboratory, (NMRL), India has developed a 100 W system based on PEFC technology, integrated with online hydrogen generator, for man portable ﬁeld power applications. The system can provide 100 W power for 10 h for every 1 l of liquid fuel.

Operation at low ambient temperatures of the order of −20 °C, as well as performance at low ambient oxygen levels, typical of the high altitudes has to be specially factored in the design offuel cells for such applications. For use in desert regions, the system should take care of high temperature autocatalysis of hydrogen donor materials employed. Similarly, if liquid fuels like methanol is used, due consideration should be given to the fuel’s ﬂash point. Air breathing systems should employ dust ﬁlters engineered to take care of the desert storms. Practically no repair is possible in the ﬁeld. Systems should have long enough MTBF. Power conditioning to meet the input requirement of speciﬁc devices and qualiﬁcation of system to MIL standards of EMI/EMC are other engineering challenges.

Lower end of soldier power, say around 20 W, can be met by direct ethylene glycol—anion exchange membrane based fuel cell. Ethylene glycol being an anti-freeze material is suitable for low temperature locations like northern sectors of India. Direct Methanol Fuel cells are also showing high potential for such low power applications. A wearable fuel cell together with disposable fuel cartridge can provide higher energy density than the best of the lithium primary cells.

2.2 Auxiliary Power Units (APU)

Field deployed vehicles and battle tanks of army need on-board power for electrical and electronic devices in use. Auxiliary power required during ‘silent watch’ should leave absolutely low signatures. Such systems should be capable of autonomous operation without operator intervention. The system should be engineered to give high levels of reliability under extreme environments of temperature, dust, humidity, shock and vibrations. Processes should not leave observable emissions of chemicals, smoke, light or sound. Ideally their thermal signatures should also be very low. Power conditioners and associated electrical circuitry should conform to MIL standard EMI/EMC speciﬁcations. This is a speciﬁc application where the conventional diesel power generators can be replaced with fuel cell generators for signiﬁcant strategic advantages. Weight and volume considerations are important; but not as critical as in the case of man portable systems. However, the systems need to be all weather resistant, robust and highly reliable. Operator intervention and maintenance requirements should be minimal.

NMRL has developed a PAFC based 10 kW generator car that uses an integrated methanol reformer for in situ hydrogen generation. This power source can be used with advantage for ad hoc repair facilities for ﬁeld equipment. They are also handy for enhancing relief operations in distress management and for providing emergency medical assistance camps in the remote locations.

2.3 Distributed Power Generation

At forward area base camps of the armed forces grid power may not be available. Captive generation is the only alternative. Major portion of base power needs, as well as heating and cooling needs, can be met by fuel cell systems. Combined heat and power (CHP) systems with fuel cells can be very effective at remote locations. Ground handling vehicles can be directly operated by fuel cells, or alternatively fuel cells can charge the battery operated vehicles and equipment. Higher overall efﬁciency of the system will justify the high initial cost since transportation of fuel to such locations through difﬁcult terrains is very cumbersome and expensive. Maintenance free operation for long periods is a primary requirement. PAFC is a good choice from the life expectancy considerations. But SOFC is more versatile when choice offuel is considered. For this segment, NMRL has developed a 30 kW truck mounted modular PAFC power plant with integrated methanol reformer.

2.4 Autonomous Systems

Advent of robotics and unmanned vehicles have revolutionised strategies and tactics in the battle ﬁeld. Unmanned aerial vehicles of diverse capabilities, unmanned ground vehicles used for mine clearing, unmanned NBC reconnaissance vehicles, autonomous underwater vehicles in diverse roles etc., are the new generation technologies aimed at reducing human causalities in situations of conﬂict. These special platforms require widely varying power sources, typically 10 W for micro aerial vehicles at the low end and up to 3000 W for unmanned ground vehicles.

Unmanned underwater vehicles (UUVs) require power sources with long endurance having high speciﬁc energy and power density, beyond what can be met using conventional battery power. In order to realize the full potential, current research focus is on advanced batteries and mini fuel cell systems. Fuel cells can have energy density several folds higher, compared to silver-zinc or lead acid batteries. As the energy storage volume of a UUV increases, fuel cells become more appropriate for enhancement in run duration, compared to even Li-Ion batteries. Unlike batteries which take a long time to recharge, the fuel cell’s tanks can be refuelled quickly, resulting in rapid turnaround times between missions. A wide spectrum of technologies starting from mini-tubular SOFC to light weight PEFCs, conformal DBFCs etc., have been specially developed with success.

Use of hydrogen as compressed gas and oxygen as liquid oxygen(LOX) has been the choice for fuel cell system on ‘URASHIMA’ UUV [5]. Fuel cell is housed in a pressure vessel. Oxygen gas is supplied from a high pressure oxygen gas tank and hydrogen is supplied from the metal hydride contained in a pressure vessel. Chemical storage and onboard release of oxygen on demand from decomposition of peroxides and perchlorates offers high volumetric efﬁciency, though weight penalty will be higher [6]. Generation of hydrogen from hydrolysis of borohydrides, is also an available choice. For UUVs the volume constraint is important and it is necessary to minimise the balance of plant. Ease of recharging and reﬁlling, is also one of the important criteria.

Depending on speciﬁc missions, UUVs have to dive to very deep ocean (say a few thousand meters) to retrieve data from sea bed sensor systems, or travel at medium depths (say a few hundred meters) for tracking and trailing adversary submarines. In either case the energy density as well as speciﬁc energy of the power plant has to be very high. The system should be heat integrated to the full extent, so that net heat thrown out should not increase the ambient temperature inside the electronics compartments. Being deep water vessels hull penetrations are generally avoided. All reaction products are to be contained inside the body. General arrangement offuel cell and balance of plant should be so designed as to ensure that the progressive shift in centre of gravity and centre of buoyancy, during the run of the UUV are within the permitted hydrodynamic design limits of the platform. The system needs to be engineered to be fully operational even during the extremes of manoeuvres of the vessel. There are UUVs that are designed to run in different modes totalling to several says in a single mission. The command and control during the mission will be by the onboard computer. Power plant reliability has to be absolute, for such applications, where the plant needs to respond to the dynamic load characteristics and function without attention, for several days at a stretch.

Alejandro Mendez et al. [7] has reviewed some of the ﬁeld demonstrations of autonomous vehicles powered by Fuel cells. Several models offuel cells developed for AUVs, comparison of fuel options, oxygen storage possibilities and the associated issues are discussed in detail.

2.5 Fuel Cells on Battle Ships

The United States as well as United Kingdom have programmes running, to study the advantages of using fuel cells for powering surface ships. Speciﬁc advantages are better efﬁciency compared to gas turbines and diesel engines, reduced smoke, reduced sound and thermal signatures, lower vibration levels, design ﬂexibility due to modularity etc. Power rating required will be of the order of a few megawatts. Types of fuel cells having potential utility are SOFC, MCFC, PAFC as well as PEFC. On board fuel processing will be inevitable. Use of logistic fuel is very much desired, if dual modes are proposed. Hybridization of direct fuel cells with turbine cycles using the fuel cell by-product heat, as proposed by M/s Fuel Cell energy® [8] can give very high overall efﬁciencies.

2.6 Air Independent Propulsion System (AIP)

Air Independent Propulsion system, popularly known as AIP system, for conventional diesel electric submarines, is a mission critical application for fuel cells. Diesel electric submarines use storage batteries as power source during their subsurface sailing missions. The batteries are recharged using electricity produced by running the diesel generators, for which the boat needs to snorkel. AIP systems cater for charging of the batteries while the submarine is still in the dived condition. The added feature has a force multiplier effect on the role and performance of the submarine, since the vessel can undertake very long underwater sailing missions. Typically the endurance can be up to 2 weeks, whereas a non-AIP submarine has to surface once in almost every 2 days. Of the different types of AIPs, fuel cell based AIPs are considered superior, thanks to its silent operation and low heat generation. Reliability, availability and maintainability of the system are most important for submarine applications. Modular architecture has advantages over a composite system [9]. Even if one of the modules fails, an intelligent control system can regroup and realign the healthy units to give at least a reduced output, thereby enhancing survivability.

AIP module is commonly engineered as an auxiliary standalone unit. The module has to be comprehensive and complete including provision for the fuel processing/storage, oxygen storage and dosing, control and instrumentation systems, microclimate management, safety features, as well as platform interfaces. The fuel cell per say and the balance of plant, together should geometrically conform to the form, ﬁt and design standards of the submarine. The system should satisfy all the platform doctrines.

Structurally the submarine is designed as an externally loaded pressure hull and will be circular in cross section. The add-on AIP section should have only marginal impact on the speed and manoeuvrability of the vessel. As a rule about 10 % extra length is permitted. A typical AIP section can therefore be a cylindrical plug, about 6.5 m in diameter and about 7 m in length. The plug has to be neutrally buoyant and can therefore it can have a typical weight of the order of 200–250 tons.

2.6.1 Weight and Volume Constraints

The volume budget should consider space for passing pipelines cables and trunking through the AIP module to connect between the fore and aft sections as well as passage for crew movement. Apart from refuelling and local maintenance requirements, provision has to be made for shipping in and shipping out of equipment for shop ﬂoor repairs and reﬁt. Submarine safety codes stipulate the porosity, i.e., the total volume of equipment in the module, expressed as a fraction of the total volume of the plug. This is a cardinal requirement for the crew safety. With the total reactants loaded, the module should be neutrally buoyant. In case the design calls for any of the by-products of onboard chemical processes to be discharged into the sea, compensating tanks for maintaining the buoyancy criteria has to be provided. The lay out design shall ensure that only minimal shifts in centre of gravity will be experienced, as the reactants get consumed.

2.6.2 Constraints in Equipment Sizing and Lay Out

Loss in energy at any stage will have cascading effects. Primarily, for the same mission endurance more amount of fuel has to be carried, increasing the all up weight. Secondly the lost energy will manifest as heat generated within the compartments leading to increased air conditioning load, which in turn adds to the parasitic load on the system. Efﬁciency of power conditioners should be at least 95 %. The load dependent voltage characteristics, typical of fuel cells, throw immense challenges to the power system hardware designer.

In general practice the fuel cells are stacked using stress bolts. The level of pre-stress is chosen by the designer with due considerations for the gasket seating force required for the individual cells, the contact pressure required for current collectors of the cells, thermal expansion and stress relaxation characteristics of the prestressed assembly, etc. The stacks have to be conﬁgured and engineered to form towers of optimal capacity. The preferred shape of cocooned towers is cylindrical so that the unit can be shipped out through the circular hatch openings. Being equipment onboard the submarine, all equipment designs will have to qualify the shock and vibration standards through use of appropriate shock mounts and anti vibration devices.

2.6.3 Platform Safety Concerns

Any leakage of Hydrogen inside a submarine compartment is strictly prohibited. Measures such as jacketed pipe design and special welding and joining procedures have to be adopted. Reliable leak detection and abatement systems have to be provided for. The designs should be qualiﬁed through physical tests of conformity under environmental and cyclic thermal aging processes. Hydrogen requirement for a sortie is to the tune of a few tons. In certain designs of AIP hydrogen is stored in metal hydrides and is desorbed on a temperature swing. To avert the potential hazard from accidental temperature rise and consequent high rate release of hydrogen, the hydride cylinders are located outside the pressure hull. The design calls for hull penetrations to take the gas inside. Optionally hydrogen is produced on board through chemical routes-either by hydride decomposition reactions or by partial oxidation of alcohols or through reforming of diesel. Prima facie this arrangement is safer since there is no bulk storage of hydrogen at any time. However, designing a complex chemical plant for the dynamic conditions of a submarine and engineering the plant for hydrogen safety regimes under continuous operational cycles is a challenge in itself. Behaviour of plant hardware and equipment has to be studied using ship motion simulators. Geometrically the equipment designs have to be tailored to suit the form and ﬁt of cylindrical proﬁle. Process intensiﬁcation and reliability engineering has to be thorough. Gravity ﬂow conditions under different list, trim and roll conditions of the submarine in motion have to be simulated while qualifying the plant architecture.

Oxygen requirement for a full sortie will be to the tune of 30–40 tons. The most efﬁcient way to store oxygen is as liquid oxygen (LOX). LOX tanks have to be located inside the pressure hull for single hull submarines. Upper limits on thermal leakages are very critical since excessive boiling of LOX can lead to forced venting routines. Designer should ensure that the LOX system does not add to criticalities during the permitted motion cycles of the ship under various environments of shock, vibration and acceleration/deceleration.

3 Concluding Remarks

Fuel cells have come a long way in technology maturity. Still large scale exploitation in both, domestic and industrial segments has not taken place, at the expected pace. Other forms of energy conversion are still remaining competitive. Extensive R&D focused on cost reduction and life cycle cost management is progressing. Defence sector stands to gain signiﬁcantly from the unique features of fuel cells. The challenges in design and engineering to meet the stringent military standards in reliability, environmental qualiﬁcation, life cycle management, etc., are to be addressed through a comprehensive and holistic approach. Application development programmes should look at the totality of system and not the fuel cell alone in isolation. In this paper, certain key issues and important engineering challenges to be addressed in development of a fuel cell based AIP system for submarines, has been dealt in some detail in order to illustrate the system engineering complexity.

References

1. The Fuel Cell and Hydrogen Annual Review, 4th Energy Wave (2015), p. 4
2. S.B. Adler, Fuel cells: current status and future challenges. The BRIDGE 35(4), National Academy of Engineering 29–32 (2005)
3. A.S. Patil, T.G. Dubois, N. Sifer, E. Bostic, K. Gardner, M. Quah, C. Bolton, Portable fuel cell systems for America’s army: technology transition to the ﬁeld. J. Power Sources 136, 220–225 (2004)
4. T.J. Gross, A.J. Poche, Jr., K.C. Ennis, Beyond Demonstration: The Role of Fuel Cells in DoD’s Energy Strategy, Chap. 2, p. 5
5. T. Maeda, K. Yokoyama, N. Hisatome, S. Ishiguro, K. Hirokawa, T. Tani, Fuel cell AUV “URASHIMA” mitsubishi heavy industries, ltd. Techn. Rev. 43(1) (2006)
6. K.E.Swider-Lyons,R.T.Carlin,R.L.Rosenfeld,R.J.Nowak,Technicalissuesandopportunitiesfor fuel cell development for autonomous underwater vehicles. AUV—Symposium on autonomous underwatervehicle technology http://www.Ieeexplore.Ieee.Org ar number=1177204
7. A. Mendez, T.J. Leo, M.A. Herreros, Fuel cell power systems for autonomous underwater vehicles: state of the art, ﬁrst international e-conference on energies, 14–31 March 2014. http:// sciforum.net/conference/ece-1 8. Fuel cell energy®, fuel cell power plant experience, naval applications, US department of energy/ofﬁce of naval research, shipboard fuel cell workshop Washington, DC March 29, 2011 9. P.L. Mart, J. Margeridis, Fuel cell air independent propulsion of submarines, DSTO-GD-0042, Department of Defence, Defence Science And Technology Organisation Australia

Posts · by **Yohannes** » Fri Apr 20, 2018 4:15 pm

Analysis of failure of add-on armour for vehicle protection against ballistic impact

Engineering Failure Analysis 16 (2009) 1837–1845
Journal: http://www.elsevier.com/locate/engfailanal

Analysis of failure of add-on armour for vehicle protection against ballistic impact

AUTHORS: Vicente Sánchez Gálvez *, Laura Sánchez Paradela
Materials Science Department, Universidad Politécnica de Madrid, ETS Ingenieros de Caminos, Ciudad Universitaria 28040 Madrid, Spain

*Corresponding author. E-mail address: Vsanchez@mater.upm.es (V. Sánchez Gálvez).

Article history:
Received 2 September 2008
Accepted 6 September 2008
Available online 19 September 2008

1. Introduction

Since the very beginning, the military industry has been aimed to two opposite directions: the development of more efficient weapons, that is to achieve better systems to attack and the development of more reliable protection systems, that is to achieve improved methods of defence. The first line produced swords, spears, catapults, guns, rifles, machine guns, etc., while the second one developed shields, helmets, armours and walls.

The two concepts converged in the XX century with the discovery of the main battle tank (MBT) that included a high fire power gun as well as high protection armour. This combination made the MBT the king of the battle during the Second World War.

The situation however changed rapidly in the following decades. The discovery of more efficient weapons, such as the kinetic energy (KE) projectiles, shape charges and explosive formed projectiles (EFP) able to penetrate up to almost 1 m of rolled armour steel meant that tanks produced during the 1960s and 1970s were vulnerable to personnel carried missiles. Since then, new concepts for the protection of military vehicles were developed including active armours (explosive reactive armours, ERA) as well as the utilization of advanced armour materials, like ceramics and composites improving the protection and reducing the thickness and weight of the armour compared to high strength steel.

In the last decades, the end of the Cold War and the spread of conflicts in different areas have lead to a new change in the concepts of protection of military vehicles.

The deployment of peace keeping military forces in conflict areas must face different threats: mines, improvised explosive devices (IEDs), impact of low and medium calibre projectiles. On the other hand, troops must be rapidly transported and shifted, thus using light armoured vehicles (LAV), which could even be aero transported to the site.

Therefore, the research in the field of protection of military vehicles has abandoned for a while the search for more sophisticated armours able to defeat the last generation threats to focus on the development of light armours able to protect the occupants of LAVs from blast and impacts of medium calibre ammunition.

2. The add-on armour concept

The add-on armour concept is quite simple, but it permits to fulfil the abovementioned requirements very easily. The idea may be summarized as follows. A light armoured vehicle (LAV) is usually employed to transport troops at conflict areas. Its main armour (made either of steel or aluminium) provides protection only against the impact of low calibre ammunition (usually up to 12.7 mm calibre). That light armour enables cheap and easy transportation of vehicles to the site. But at conflict areas, the armour ought to be upgraded to achieve protection against blast and medium calibre projectiles. This requirement is fulfilled by adding to the main armour the add-on or appliqué armour, usually a kit that is either glued or screwed to the hull. The add-on armour increases the weight of the vehicle, thus reducing its mobility. An optimum design of the add-on armour keeping the increase of weight under reasonable limits but keeping the protection capacity is thus worthwhile. An additional advantage of the add-on armour is the possibility of fast repair in case of an incident. Being glued or screwed to the hull, the damaged add-on kit can be easily removed and replaced by a new one by the occupants of the vehicle.

3. Add-on armour design

Armour materials are good examples of systems designed to fail. The add-on armour mission is exactly that, when the vehicle is subjected to a critical threat, a mine detonation for instance, the optimum designed add-on armour materials must fail, but avoiding the full rupture of the main armour and the damage to the occupants. An optimum design of the add-on armour against a definite threat must defeat the threat at the expense of its own full failure, so that a slight armour reduction would involve the out of action of the vehicle or the injury of its occupants. The armour design can be carried out by any of the three following approaches [1]:

(a) Analytical modelling.
(b) Numerical simulation.
(c) Empirical methods

The analytical models are based on the development of simple equations governing the impact or penetration processes. Those equations, derived from the continuum mechanics laws, usually involve several assumptions and then provide simple expressions of the quantities involved such as residual velocity and residual mass of the projectile and, ballistic limit. Analytical models are tailor-made for a specific threat target system, thus its predictive capacity for a different threat or target is rather limited.

Numerical simulation is based on the utilization of numerical codes (hydrocodes) that employ the finite element method (FEM) or the finite difference method (FDM) to obtain the solution of equations governing the impact process. This approach provides a high amount of information and it is widely used for armour design optimisation since high capacity and high speed computers are available. The main shortcoming of these methods is the lack of material behaviour data, especially failure criteria of advanced materials.

Finally the empirical methods are based on fire tests. Obviously both analytical modelling and numerical simulation always require experimental validation by firing tests. But the empirical approach is based exclusively on ballistics tests. It is thus a very expensive procedure, that provides very scarce information and its results are hardly extrapolated to other systems than those actually tested.

Summarizing, optimum design of add-on armours can be achieved by a clever combination of the three abovementioned methods. Fig. 1 is a sketch of the methodology proposed. The analytical modelling may be useful to obtain a rough approximation of the solution, discriminating among a high number of variables: materials selection, thicknesses, angles, adhesives, etc. The numerical simulation is a valuable tool to tune the solution. Often the use of hydrocodes would require dynamic testing of materials involved to derive constitutive equations and fracture criteria to supply to the code. Finally, ballistic or blast testing is required to validate the design.

Fig. 1. Sketch of the proposed procedure for add-on armour design optimisation.

4. Analytical model of ceramic–metal add-on armour

As an example, the analytical model developed at the Materials Science Department to analyze the penetration process of medium calibre projectiles into ceramic–metal add-on armours is sketched in Fig. 2 [2,3]. The model is based on the assumption that the impact produces a cone of comminute ceramic that spreads the load on the back metallic plate which finally fails in tension along a circle with much greater diameter than the projectile calibre. The model was introduced in a computer programme, called SCARE that provides the solution to a specific threat in a few seconds. Fig. 3 shows two examples of results obtained with SCARE to achieve the optimum design of alumina/aluminium armour to defeat 25 APDS projectiles at different impact obliquities. Fig. 4 illustrates the results of residual velocity after perforation of 25 APDS projectiles impacting ceramic/aluminium armour at normal impact. The figure includes results obtained with the analytical programme SCARE as well as those achieved with the numerical code AUTODYN.

Fig. 2. Sketch of ceramic/metal armour perforation

Fig. 3. Optimum design of alumina/aluminium armour to defeat 25 mm APDS projectiles at different obliquities. (a) 95% purity alumina and (b) 99.5% purity alumina.

Fig. 4. Analytical and numerical results of residual velocity of 25 mm APDS projectiles after perforation of ceramic/aluminium armour. (a) Aluminium nitride/aluminium and (b) 99.5% purity alumina/aluminium.

As can be seen, analytical results agree fairly well with numerical simulations. Therefore, it can be pointed out that SCARE is a valuable tool to achieve a rough approximation of ceramic/metal add-on armour design optimisation.

5. Analytical model of ceramic/composite armour

As an alternative, the ductile backing plate may be produced with a composite material. Composites employed are usually polyester or vinyl ester reinforced with glass, aramid (Kevlar) or crystalline polyethylene (Spectra) fibres. The utilization of composites permits to reduce the armour weight, although they are usually more expensive than metal alloys.

An analytical model has been developed to simulate projectile penetration into composites and ceramic–composites armours [4,5]. The model, based on the ideas of Roylance [6] and Cunniff [7], assumes that projectile impact on the composite surface produces tensile and shear waves that travel along the fibres, it is also assumed that the matrix, being much weaker than the fibres, fails from the very beginning giving little if any contribution to defeat the projectile. The model assumes also that the composite is constituted by several stocking fabrics sheets that fail consecutively in tension while decelerating the projectile. Finally, depending on the number of sheets, the fibre strength and the size and velocity of the projectile, the target may stop the projectile or a full perforation is obtained.

Fig. 5. Analytical results of residual velocity of armour piercing (AP) projectiles after perforation of alumina/Kevlar armour. (a) 7.62 AP, (b) 12.70 AP and (c) 14.50 API

Fig. 5 shows examples of the use of this model to predict the behaviour of alumina/Kevlar composite armour impacted by 7.62, 12.70 and 14.50 mm AP projectiles. The analytical model provides very rapidly a full set of curves residual velocity vs. impact velocity for different armour configurations. Optimum design for specific add-on armour is then very easily obtained. Finally, Fig. 6 illustrates the results of ballistic limit vs. areal density plots obtained with the analytical model and the numerical hydrocode AUTODYN as well as some experimental data for the three AP projectiles analyzed. A fairly good agreement with numerical and experimental results is obtained.

Fig. 6. Analytical, numerical and experimental results of optimum design of alumina/Kevlar armour to defeat AP projectiles at different impact velocities.

6. Numerical simulation

Add-on armour design optimisation use to be accomplished by numerical simulation. Rough solution obtained by means of analytical modelling may be tuned by the use of FEM or FDM numerical codes (hydrocodes). AUTODYN and LS-DYNA are the most widely used hydrocodes for numerical simulation of ballistic impact.

Numerical simulation provides a great amount of information, stresses, strains, velocities and accelerations at different points of projectile and target as a function of time, plastic contours, damaged and ruptured areas, etc. The main difficulty is the lack of information of the mechanical behaviour and rupture criterion at high strain rate for many advanced materials, especially ceramics and composites. Consequently, armour designing often requires an experimental programme to determine the dynamic properties of advanced materials to be employed. Hopkinson bar as well as the novel spalling technique developed at our Department [8] is an appropriate method for testing advanced ceramics at high strain rates. Data are then introduced into the constitutive equations being used by hydrocodes for a reliable simulation of ballistic impact phenomena.

Fig. 7. Meshes used for AUTODYN 3D computations of 20 mm APDS projectile impact onto alumina/aluminium armour at different obliquities.

For instance, Fig. 7 illustrates the meshes used to simulate the impact of 20 mm APDS projectile onto ceramic/aluminium add-on armour at different obliquities. Impact velocity is 1240 m/s. The projectile is simulated by a tungsten cylinder of 12 mm diameter, 35.2 mm length and 72.1 g of weight. AUTODYN 3D is used for numerical simulation.

The materials involved are the tungsten alloy (W94%Ni4%Fe2%), 5083H111 aluminium alloy and 99.5% purity alumina. The constitutive equations used to model the alloys are a shock equation of state for the hydrostatic stress and Steinberg–Guinan equation [9] for the deviatoric stress. Ceramic is modelled by a linear equation of state and Johnson–Holmquist strength model [10].

Table 1 summarizes the results obtained for different add-on configurations. The table includes the results of residual mass and residual velocity of both experimental tests and numerical simulations.

Table 1
Numerical and experimental results of perforation of alumina-aluminium add-on armour by 20 mm APDS projectiles

Finally, Fig. 8 shows two pictures of the deformed meshes during the perforation process as obtained in the numerical simulation. The pictures depict projectile and target contours at two different times for 60^O NATO obliquity impact onto 15 mm alumina/10 mm aluminium add-on armour. As can be seen, the simulation predicts a reduction of mass and speed of the projectile and also that its flight is not stable. After add-on armour perforation, the projectile is tumbling over, being unable to perforate the main armour. Pictures also show two X-ray shadowgraphs of an actual test taken at the same times. The agreement between numerical prediction and experimental results is noticeable.

Fig. 8. Deformed meshes and X-ray shadowgraphs of the impact of 20 mm APDS projectile onto alumina/aluminium add-on armour at 60 obliquity (a) at 66 ls and (b) at 108 ls.

7. Conclusions

Add-on armour is a convenient method to improve protection of light armoured vehicles against ballistic impact.

Armour design optimisation may be carried out by a clever combination of analytical, numerical and experimental methods.

Advanced ceramics and composite materials are being used for add-on armour production in combination with conventional steel and aluminium alloys.

Numerical simulation of penetration processes into add-on armours usually requires high strain rate experimental testing to determine empirical parameters to be included in constitutive equations and failure criteria of materials involved.

Acknowledgements

The authors wish to express their gratitude to the Spanish Ministry of Defence and Ministry of Education for financing this research. Project CONSOLIDER INGENIO 2010.

References

[1] Sánchez Gálvez V. Analytical and numerical simulations of ballistic impact on composite lightweight armours. In: Sánchez-Gálvez V, Brebbia CA, Motta AA, Anderson CE (Eds.), Computational ballistics II2005. Southampton, UK: WIT Press. p. 3–10.

[2] Zaera R, Sánchez Gálvez V. Analytical model of ballistic impact on ceramic/metal lightweight armours. In: 16th international symposium on ballistics. San Francisco, CA. USA. vol. 3; 1996. p. 487–95.

[3] Zaera R, Sánchez Gálvez V. Analytical modelling of ballistic impact of normal and oblique ballistic impact on ceramic/metal lightweight armours. Int J Impact Eng 1998;21(3):133–48.

[4] Chocron IS, Sánchez Gálvez V. A new analytical model to simulate impact onto ceramic/composite armors. Int J Impact Eng 1998;21(6):461–71.

[5] Chocron IS, Oña JJ, Gálvez F, Sánchez Gálvez V. New analytical model and numerical analysis for the oblique ballistic impact into composite add-on armors. In: 18th international symposium on ballistics, Austin, TX, USA; 1999. p. 777–84.

[6] Roylance D, Wilde A, Tocci G. Ballistic impact of textile structures. Text Res J 1973:34–41.

[7] Cunniff PM. An analysis of the system effects in woven fabrics under ballistic impact. Text Res J 1992;62(9):495–509.

[8] Gálvez F, Rodríguez J, Sánchez Gálvez V. The spalling of long bars as a reliable method of measuring the dynamic tensile strength of ceramics. Int J Impact Eng 2002;27:161–77.
[9] Steinberg DJ, Cochran SG, Guinan MW. A constitutive model for metals applicable at high strain rates. J Appl Phys 1987;51:1498–504.

[10] Johnson GR, Holmquist TJ. A computational constitutive model for brittle materials subjected to large strains, high strain rates and high pressures. In: Meyers MA, Murr LE, Staudhammer KP, editors. Shock wave and high strain rate phenomena in materials. Marcel Dekker; 1992. p. 1075.

Posts · by **Yohannes** » Fri Apr 20, 2018 5:44 pm

Automatic Target Tracker for Main Battle Tank

This full-text paper was peer-reviewed and accepted to be presented at the 2015 International Conference on Communications and Signal Processing (ICCSP).

AUTHORS: M. Ilamathi and K. Abirami

M. Ilamathi is with Electronics and Communication Engineering, Easwari Engineering College, Chennai, India (corresponding author to provide phone: 7708817472; e-mail: ilam.2210@gmaiI.com).
K. Abirami is with the Electronics and Communication Engineering, Easwari Engineering College, Chennai, India (corresponding author to provide phone: 8939858465;e-mail: k_abi@yahoo.com).

I. INTRODUCTION

Automatic target tracking systems are employed in a wide variety of missions and tracking environment such as fire control, guidance, navigation, passive range estimation, and automatic target discrimination. For object recognition, navigation systems and surveillance systems, object tracking is an indispensable first step. Video object tracking has got wide application in vision, security, observational issues in natural science and in various other fields. Object tracking can be complex due to many reasons like noise in images, complex object motion, and articulated nature of non-rigid objects, scene illumination changes, and real-time processing requirements [1].

Nowadays the equipment's engaged in the battlefield are automated to improve the safety of soldiers and also to ensure combat effectiveness. Main battle tanl<- consists of hull and turret. The latter is provided with three hatches .There are four crew members namely loader, gunner, driver and commander. Main gun mounted in the turret of the tank is controlled by the gunner with the control handler which is nothing but joystick. The gunner tracks the target in the field with the help of Main Sight System. The sight system and the main gun is integrated, i.e. sight system acts as Master and the gun is the slave. Hence the Gunner steers the sight to get the target of interest within the tracking window manually.

The purpose of an auto-video tracking system (AVTS) is to maintain a stable sensor-to-target line of sight in the presence of relative target motion and base motion disturbance to the sensor platform [2][3]. The operator acquires the target, within the track gate, using joystick. After the target is acquired, the tracking subsystem locks onto it and thereafter maintains the LOS automatically. A wide range of target characteristics can be anticipated in the typical scenario of a low-contrast target in a complex scene. Various target-tracking algorithms such as edge, centroid and correlation are available for generating the error signals with respect to center of field-of-view.

Tracker design is environmental-sensitive. Different types of tracking systems are designed to meet the respective tracking environment. For example, the star tracker would not be able to track the maneuvering, non-cooperative target [4]. Similarly, an up-looking tracker designed to track airborne/space borne target against a sky background would not be able to track target against ground clutter in down-looking surveillance system [5].

The performance of tracking system also depends on operator's skills. It is therefore essential to examine different target and background conditions for evaluating the AV T system. In almost all tracking algorithms, object motion is assumed to be smooth with no abrupt changes in between [6]. But in our proposed system it is completely a non-deterministic environment (war field). Hence an efficient area based algorithm has to be implemented.

In this paper, section 2 presents a problem in the existing system. Section 3 details the proposed methodology and the functional description of the ATT system. Section 4 explains the flow of algorithm in tracking the moving target. In Section 5 experimental results and snapshots are discussed and section 6 deals with the conclusion and future work of the project.

II. PROBLEM DEFINITION

In the Manual system, the operator positions the gimbal through a positional control Goystick) following the target's motion on a display or viewing through the mirror system. The limitation of manual systems is the operator's inability to respond to target dynamics. Tracking is a complicated task. Gunners are well trained and can do this job excellently during peace time. However during war, fear physcosis will set in, so there is chance of missing the target. Mainly due to harsh terrain conditions gunner faces difficulty once the own Tank starts moving. When both own Tank and target moves, its cumbersome task for the Gunner to track the Region of Interest (ROI) on the target due to the disturbances. Hence automation using efficient tracking technique satisfying all the constraints must be framed and implemented. Common problems of erroneous segmentation are long shadows, partial and full occlusion of objects with each other and with stationary items in the scene. Hence in a non- deterministic environment like war field it very difficult because of harsh terrain conditions and clutter background. So far many algorithms have been proposed for target detection and tracking which performs better for a known environment [7]. Thus, dealing with these conditions is important for robust tracking.

III. PROPOSED METHODOLOGY

ATT is one of improvement in the main battle tank. The proposed system works on the real time video which is acquired through Tank Sight system. ATT performs Normalized Area Correlation algorithm on real time video. It generates the error signal on the movement of target between each frame. This error signal replaces the Gunner's command signal. We can detect all objects in images no matter whether they are moving or not.

The video will be acquired by a Camera or the so called tank sight system to ensure continuous tracking. On receiving the target over the Gunner Main Sight, the Gun will be aligned to the direction of the target, handed over by the Commander. It is implied that once the tanl<- sight system moves, the gun also moves along it. Based on the literature review discussed in the chapter 2 the difficulties in tracking the ground target in a non-deterministic environment are analyzed.

The conventional approach to object tracking is based on the difference between the current image and the background image. Furthermore, they cannot be applied to the case of a moving camera. Algorithms including the camera motion information have been proposed previously, but, they still contain problems in separating the information from the background. Hence a normalized area correlation technique is proposed which ensures robust tracking of the ground target which is very essential in the war field [9].

Fig. 1 shows the ATTS architecture. A video tracker receives the video information from camera and locks it on a selected target. A feedback control loop called the track loop continuously adjusts the sensor platform to keep the target in the center of the sensor FOV or track gate.

Target acquisition capability of an AVT system at different ranges is dependent on target size [8]. The video captured through the camera is forwarded to the automatic video tracker system which processes the signal and sends the annotated signal to the operator's console. The control signal upon receiving the annotated signal from the operator console is given to AVT. The corresponding error signal in X direction and Y direction is calculated and accordingly the sight system is stabilized, along which the gun is also aligned ensuring continuous tracking of the target when both our tank and the enemy tank moves.

Fig. 1. System Architecture

A. Functional Description

ATT has three modes- Standby mode, Manual Mode and Auto mode. The mode switching happens based on switching status provided to the gunner in the tank system. They are:

IV. FLOW OF ALGORITHM

ATT (Automatic Target Tracker) operates in three modes, firstly Stable mode where there is no tracking and the Tank remains in static state, secondly the Manual mode where the tracking is carried manually by the gunner and thirdly the Auto mode where automatic tracking of the target is done.

In Fig. 2 the flow of different modes of operation is explained. On start, the system remains in the standby mode. The gunner is provided with a sight system which is called Main Sight System which is integrated with the main gun. Tank sight system acts as the master and gun is the slave. As the gunner steers the tank sight system the gun also moves accordingly. The gunner now turns the system to stabilization mode.

Once the system is in stabilization mode both the sight system and gun is integrated and tracking can be carried. When the left side switch is pressed, Manual tracking is started by gunner. When both our tank and the target move, gunner presses the left index switch and the auto mode is turned on. The gunner acquires the target and once he acquires automatic tracking starts.

Fig. 2. Flow of algorithm

V. RESULTS AND DISCUSSIONS

Fig. 3 shows the gunners main sight. The gunner has two windows. Right side is target tracking window and left side is target selection window. Selection window displays the scene that is being captured by the tank sight system and where the gunner selects if any enemy target arrives by steering the gunners control handle and tracking window is one that displays the selected target under track. The simulation result shown in Figure 3 is the graphical user interface which depicts the operators console created using MATLAB.

Fig. 3. Main Sight System

Fig. 4 shows the target selected by the gunner from the scene captured through the sight system. The selected target is rounded as shown.

Fig. 4. Target Selected

Fig. 5 shows the selected target being tracked in different positions using the tracking window which is green in color. The continuous tracking is ensured depending on the error generated in X and Y direction while the target is in motion.

Fig. 5. Target Tracked

In Fig. 6 the error values generated in X and Y direction corresponding to the displacement of the target between the frames is plotted. The blue line indicates the threshold.

Fig. 6. Error Plot

VI. CONCLUSION AND FUTURE WORK

The ATT automatically tracks the selected target using the real time video available from the Main Sight System. A rectangular window appears which is called as the tracking window. The ATT displays the tracking window on the video such that the centre of tracking window coincides with the TI Bore sight. Once the gunner acquires the target automatic mode is switched ON. Correlation algorithm is performed between the selected target and the reference target. It generates the error signal on the movement of target between each frame. This error signal replaces the Gunner's command signal. Thus making the Gunners job easy and efficient tracking rate is achieved without missing the moving target. In future various prediction algorithms for tracking can be analyzed to ensure robust tracking of target due obscurations and also multi target tracking based on priority can be achieved.

REFERENCES

[1] M Sankar Kishore and K Veerabhadra Rao "Robust correlation tracker", Sadhana, Research Centre Imarat, Vol. 26, Part 3, June 2001.

[2] Alper Yilmaz, Khurram Shafique, Mubarak Shah, "Target tracking in airborne forward looking infrared imagery", IEEE transactions on Image and Vision Computing, VoI.2, pp. 1-13,2003.

[3] B. Abidi, A. Koschan, S. Kang, M. Mitckes, and M. Abidi, "Automatic target acquisition and tracking with cooperative fixed and PTZ video cameras", The Imaging, Robotics, and Intelligent Systems Laboratory, The University of Tennessee, June 2003.

[4] G. Catalin and S. Nedevschi, "Object tracking from stereo sequences using particle filter",Proc. 4th Int. Conf. ICCP, pp. 279-282, 2008.

[5] B.S. Chauhan, Manvendra Singh, V.K. Sharma, and P.c. Pandey, "Auto-video tracking system: Performance evaluation", Defence Science Journal, Vol. 58, No. 4, pp. 565·572,July 2008.

[6] c. Hennes, J. Einhaus, M. Hahn, C. Wohler, and F. Kummert, "Vehicle tracking and motion prediction in complex urban scenarios", in Proc.IEEE IV Symp, pp. 26-33, 2010.

[7] Legong Sun, Zheng Mao, 'An Improved Nonnalized Cross Correlation Algorithm for Object Tracking', IEEE Transactions, 2010.

[8] Seok Pil Yoon, Taek Lyul Song and Tae Han Kim, 'Automatic Target Recognition and tracking in Forward·looking Infrared Image Sequences with a Complex Background', IEEE transactions on Image processing, Vol. 21, pp. 4622 - 4635, 2013.

[9] Sivaraman, S.; Trivedi, M. M., "Looking at Vehicles on the Road: A Survey of Vision·Based Vehicle Detection, Tracking, and Behavior Analysis," Intelligent Transportation Systems, IEEE Transactions on , vol.14, no. 4, pp. I773, 1795, Dec. 2013

Posts · by **Yohannes** » Fri Apr 20, 2018 9:46 pm

Study on Rubber Composite Armor Anti-Shaped Charge Jet Penetration

DOI: 10.1002/prep.201200172
Propellants, Explosives, Pyrotechnics

AUTHORS: Xudong Zu,^[a] Zhengxiang Huang,^*[a] and Xin Jia^[a]

[a] X. D. Zu, Z. X. Huang, X. Jia
School of Mechanical Engineering
Nanjing University of Science and Technology
Xiaolingwei 200
Nanjing 210094, P. R. China
*e-mail: huangyu@mail.njust.edu.cn

Article history:
Received: October 15, 2012
Revised: February 28, 2013
Published online: April 15, 2013

1 Introduction

The mechanical properties of rubber composite, including low density, high specific strength, and high specific modulus, have been employed in various protection field applications in tanks, infantry fighting vehicles, and shipbuilding, among others. Current literature indicates that research on the anti-eroding performance of rubber composite materials has focused on bullets, grenade fragments, and kinetic energy projectiles, and so on [1, 2]. Bulging of the metal plates has been attributed to the transfer of momentum from the jet to the inert interlayer during the process of penetration. The major defeat mechanism of the bulging armor is similar to explosive reactive elements [3]. Yaziv compared rubber composite and explosive reactive armors, highlighting the former as superior in terms of safety and environmental effects [4]. The description of the mechanisms of these bulging systems was first proposed by Gov [5]. The study further described the process of interaction of a rubber composite armor with an SCJ, but did not provide the theoretical model. Rosenberg studied the resistance capability of sandwich composite armors with different sandwich materials through two-dimensional (2D) simulation and considered material strength, stress modulus, and density to be important factors for the resistance of a composite armor to an SCJ [6]. A layer of rubber, which will gasify, or even explode, is regarded as an inert explosive. A mechanism for the interaction based on the theory of Kelvin-Helmholtz instabilities was discussed by Helte et al. [7, 8].

This paper presents an analytical approach for determining the particle velocity of plates based on the 2D stress wave theory. A theoretical model for the disturbance between the plates and the jet is proposed according to the theory of Kelvin-Helmholtz instabilities. The effects of dip and rubber layer thickness on anti-penetration performance are discussed. Experiments using a rubber composite armor, where the interaction with an SCJ was registered by X-ray flashes and depth of penetration, were conducted to verify the theoretical model.

2 Theoretical Models

Rubber composite armor panels comprising two steel plates sandwiched around a layer of rubber are very efficient as add-on armor against shaped charge. For convenience of discussion, the front plate (F plate) and back plate (B plate) are defined as shown in Figure 1.

Figure 1. Sketch of rubber composite armor disturbance of the stabilities of an SCJ.

2.1 Interaction Process Analysis

The interaction process between a rubber composite armor and a jet can be analyzed based on the steady penetration theory. The deformation of a jet may be described in terms of the following three-stage process:

2.2 Non-disturbance Stage

The first stage, that is, from the time the jet penetrates the armor to the time when the shock wave spreads to the back plate, the plates are almost undisturbed by the jet.

2.2.1 Full Disturbance Time Calculation

According to the theory of steady penetration, the entire time that the jet penetrates the plates can be calculated. Suppose a hypervelocity jet strikes an armor with velocity v_j. The erosion rate of the jet v_e is given by

where u is the armor penetration velocity, and v_j is the SCJ velocity.

The time of the jet segments penetration of the plate is given by

The length of penetration per segments jet is

2.2.2 Time of Shock Wave Spread to the Front Plate

According to the Fermat principle, the shock wave travels along the shortest path. Hence, the transition time is relative to the depth of the plates and the velocity of the shock wave. The travel time from the shock wave to the front plate is given by

where h_i denotes the thickness of the plate, and C_i denotes the sound velocity in the plate.

2.3 Disturbance Stage

2.3.1 Normal Particle Velocity of the Front Plate

To simplify the analysis for determining the shock wave parameters, the following points are assumed:

[list=](1) The SCJ is a quasi-steady fluid, and the velocity and diameter of the jet undergoes even and echelon change;
(2) The attenuation of shock wave propagation in the front and back plates are ignored.[/list]At any instant during jet penetration, the penetration velocity u is related to the velocity of the jet v_j as follows

where P_j and P_t denote the densities of the jet and the target material, v_j is jet velocity, and u is the penetration velocity.

The particle velocity of the back plate is transferred to the front plate with the shock wave. The particle velocity of the front plate can be calculated based on Stress Wave Propagation.

where 1FCF, 1RCR, and 1BCB denote the wave impedance of the front plate, sandwich plate, and back plate materials, respectively.

The attenuation rule of shock wave propagation in rubber can be described as e_-yh, where y is the propagation coefficient, and h is the thickness of the plate [10].

2.3.2 Overall Speed of the Hypervelocity Jet Disturbed by the Front Plate

The defeat mechanism of the rubber composite armor lies in the introduction of disturbances on the jet during the subsequent interaction between the jet and the edge of the penetration channel in the plates. These disturbances grow continuously after the interaction and result in the fragmentation and scattering of jet particles. The overall speed of the hypervelocity jet interference by the front plate can be calculated using the Kelvin-Helmholtz instabilities in Ref. [11].

To simplify the analysis of the speed of the hypervelocity jet interference by the front plate, the wave vector is taken as a constant.

2.3.3 Fracture Cycle of Jet Interference by the Front Plate

Suppose a hypervelocity jet strikes a reactive armor and penetrates the front plate material, consequently changing the crater of the front plate as follows [11]:

where r_c denotes the crater of the front plate penetrated by hypervelocity jet, and R_t denotes the dynamic strength of the front plate material (Figure 2).

To simplify the analysis of the reference SCJ parameters, we assumed the following:

Figure 2. Physical model of the interaction between the front plate of the reactive armor and the SCJ.

Based on the normal particle velocity calculated by Equation (6), the motion trail of point F is given by

Substituting Equation (11) in Equation (8) yields

Taking y as zero, then the spacing distance of the edged crater in contact with the SCJ from this time to the next can be shown as

The fracture cycle of the jet interference by the front plate is thus f = 1/t

2.4 Surplus Depth of Penetration

The surplus depth of penetration can be used to describe the disturbance capability of the rubber composite armor for the SCJ. The sketch of the target and shaped charge set up are shown in Figure 3.

Figure 3. Sketch of target and shaped charge set up.

2.4.1 Surplus Penetration Depth for Non-disturbance Jet

The penetration depth was calculated based on the virtual origin concept [12].

where L is the penetration depth, H is the height of burst, and v_j0 and v_j are the jet tip and tail velocities, respectively.

According to the analysis, the surplus penetration depth for a non-disturbance jet was modified as:

where k is the correlation coefficient with the range of 0 < k < 1.

2.4.2 Protection Capability

Owing to the different structures of rubber composite armors, surplus penetration cannot accurately assess protection capability. Hence, space protection coefficient E_s, quality protection coefficient E_m, and differential protection coefficient were adopted to describe protection capability.

3 Experiment Setups

3.1 Standard Shaped Charge Testing

The standard shaped charge was used in this study for several reasons, such as the increasing the generalizability of the study; simplifying the calculation of the protection, cost, and protection thickness coefficients; and considering that the standard shaped charge is often used in studies. The standard shaped charge (Figure 4) has the following characteristics: shaped charge copper liner with 0.8 mm thickness and 56 mm diameter, explosive volume of 203 g without a conical charge shell cover, and burst height of 80 mm. The 8# flash detonator was used to detonate the SCJ. To understand the parameters of SCJ, tip jet velocity was investigated using the chronograph (Figure 5).

Figure 4. Standard shaped charge.

Figure 5. Performance test devices of shaped charge.

The result shows the average depth of penetration to be 184 mm at the same burst height, with a relative error of approximately 1%. The depth of 184 mm was regarded as the datum, which is aimed at the condition of SCJ penetration of the semi-infinite steel target. The SCJ was consistency good, and the inlet diameter was almost the same as the outlet diameter (Table 1).

Table 1. Results of shaped charge performance test.

The jet tip and tail velocities were measured using the multi-channel X-ray system. The jet tip velocity was 5714 ms^-1 and that of the tail was 1900 ms^-1. The tip jet diameter was 1.5 mm, and tail diameter was 9 mm. The length of jet was approximately 111.5 mm at 30 us after shaped charge initiation.

3.1.1 Experimental Set Up

In these experiments, several 450 kV multi-channel X-ray systems set at 908 were used. Through a special setting, two radiographs were obtained for the same testing. To observe the deformation of the jet accurately and eliminate the errors attributable to the detonator, passive detonation technology was used in the experiment. The experimental setup is shown in Figure 6.

Figure 6. Experimental layout diagram.

3.1.2 Deformation of the Jet at Different Rubber Layer Thicknesses

The shaped charges were mounted at an 80 mm burst height and angled at 688 to the plates. The radiographs provided with information on the deformation and fracture of the jet at rubber layer thicknesses of 2, 3, and 5 mm. The deformation and fracture of the jet 75 us after shaped charge initiation are shown in Figure 7 [13].

Figure 7. Condition of the jet deformation after the jet penetrated the rubber composite armor when the thickness of the sandwich rubber changed: (a) 2 mm, (b) 3 mm, (c) 5 mm.

Figure 7 shows the deformation and fracture of the jet. The jet was evidently disturbed by the rubber composite armor at 688. With increasing thickness of the rubber plate, the deformation of the jet decreased, and the extent of the fracture increased. Thus, an optimum thickness of rubber plate exists for the best protection capability.

3.1.3 Deformation of the Jet at Different Obliquity

As shown in Figure 8, at 3 mm thickness of the rubber plate, very little jet deformation was evident at tip angles of 0^O and 30^O. At 45^O, the deformation of the jet became more evident. Severe deformation of the jet appeared at 60^O, and some bigger disturbances were observed on the jet.

Figure 8. Condition of the jet deformation after the jet penetrated the rubber composite armor when the obliquity changed: (a) 0^O, (b) 30^O, (c) 45^O, (d) 60^O.

3.2 Surplus Depth of Penetration Testing

3.2.1 Experimental Set Up

The experiments were performed with armor panels comprising two Q235 steel plates, 300 mm 150 mm 3 mm, and a layer of rubber used as inert material in the panels. The Q235 plates and rubber layer were bound by adhesive after the surface was treated by sandblasting. The angle between the plate normal and jet direction was 608. The distance between the after-effect target and the armor was 250 mm. The thickness of rubber layer was 2 mm, 3 mm, 5 mm, and 8 mm. The parameters of the plates used in experiments are shown in Table 2 and Table 3.

Table 2. Summary of nature rubber parameters

Table 3. Summary of target parameters.

3.2.2 Properties of the Armor with Different Rubber Layer Thicknesses

The jet impacts the after-effect target and produces a large number of craters. The results of the impact are shown in Figure 9.

Figure 9. Crater caused by SCJ penetration of the aftereffect target after penetrating the rubber composite armor with an altered sandwich rubber thickness.

Figure 10 and Figure 11 show that the surplus depth, along with the space protection coefficient Es, the quality protection coefficient Em, and the differential protection coefficient

, varied with the thickness of the rubber layer in the range of 2 mm to 8 mm.

Figure 10. Relationship among surplus depth, quality defense coefficient, and rubber thickness.

Figure 11. Relationship among the space defense coefficient, differential defense coefficient, and rubber thickness.

Figure 10 and Figure 11 give the results of the relationship and reveals that a thicker rubber layer results in a lower differential protection coefficient

, indicating that the effect on rubber is reduced. The values in the experimental results were slightly smaller than the theoretical values, which could be attributed to the following aspects:

Considering the above assumptions, the surplus penetration in the experimental result was smaller than the theoretical value. The relative errors were approximately 0.7%, 10.3%, 10.3%, 18.1% for the 2, 5, 8, and 3 mm rubber layer thicknesses, respectively. Considering the target structure, the rubber composite armor had the best protection capability at 3 mm to 3.5 mm rubber layer thickness.

3.2.3 Properties of the Armor with Different Obliquity in the Rubber Layer

The crater and hole on the surface of the after effect target are shown in Figure 12 [13], which reflects the fragmentation and scattering of the jet particles.

Figure 12. Crater left by the SCJ penetration of the aftereffect target after penetrating the rubber composite armor with a changed obliquity.

Figure 13. Relationship among surplus depth, quality defense coefficient, and obliquity.

Figure 13 shows that the surplus depth, along with the quality protection coefficient E_m varied the obliquity in the range of 0^O to 68^O. Figure 14 shows that the space protection coefficient E_s, and the differential protection coefficient

varied the obliquity in the range 0^O to 68^O.

With the increase in the obliquity, the computed results coincided well with the experimental results, but the protection capability of the armor did not increase monotonically. Positive effects on the rubber composite armor were observed when the obliquity increased from 0^O to 60^O. Nonetheless, at 60^O to 68^O, the armor was shown to derive a negative effect. The differential defense coefficient reached a maximum at 30^O to 60^O. These results indicate that the rubber layer plays an important role in this range of angle. At the range of 60^O to 68^O, the space defense coefficient had similar values that reached a minimum, indicating that the rubber armor had the best space protection capability at this range. Simultaneously, the quality defense coefficient reached a maximum at 60^O. Consequently, the rubber composite armor had the best protection capability at this angle.

Figure 14. Relationship among space defense coefficient, differential defense coefficient, and the obliquity.

4 Conclusions

Based on the analysis, the following conclusions can be drawn:

^O

Acknowledgments

This research was supported by the National Natural Science Foundation of China (Grant No. 11072115).

References

[1] O. Mclakoglu, T. Soykasap, Experimental and Numerical Investigations on the Ballistic Performance of Polymer Matrix Composites Used in Armor Design, Appl. Compos. Mater. 2007, 14, 47 – 58.

[2] C.-F. Yen, B. Scott, P. Dehnmer, B. Cheeseman, A Comparison between Experiment and Numerical Simulations of Fabric Ballistic Impact, 23rd International Symposium on Ballistics, Tarragona, Spain, April 16–20, 2007, pp. 853–864.

[3] M. Mayseless, E. Marmor, N. Gov, Y. Kivity, J. Falcovitz, D. Tzur, Interaction of a Shaped Charge Jet with Reactive or Passive Cassettes, 14th International Symposium on Ballistics, Quebec, Canada, September 26–29, 1993, pp. 439–448.

[4] D. Yaziv, S. Friling, N. Gov, The Interaction of Inert Cassettes with Shaped Charge Jets, 15th International Symposium on Ballistics, Jerusalem, Israel, May 21–24, 1995, pp. 461–467.

[5] N. Gov, Y. Kivity, D. Yaziv, On the Interaction of a ShapedCharge Jet with a Rubber Balled Metallic Cassette, 13th International Symposium on Ballistics, Stockholm, Sweden, June 1–3, 1992, pp. 95–99.

[6] Z. Rosenberg, E. Dekel, A Parametric Study of the Bulging Process in Passive Cassettes with 2D Numerical Simulations, Int. J. Impact Eng. 1998, 21, 297–305.

[7] A. Helte, E. Lidn, The Role of Kelvin-Helmholtz Instabilities on Shaped Charge Jet Interaction with Reactive Armours Plates, 25th International Symposium on Ballistics, Beijing, China, May 17–21, 2010, pp. 1547–1553.

[8] A. Helte, E. Lidn, The Role of Kelvin-Helmholtz Instabilities on Shaped Charge Jet Interaction with Reactive Armours Plates, J. Appl. Mechanics 2010, 77, 051805-1-051805-8.

[9] L. Xiao-min, H. Shi-sheng, C. Zhi, The Wave Propagation Attenuation and Dispersion in a Viscoelastic Hopkinson Pressure Bar (in Chinese), Acta Mech. Solid.Sin. 2002, 23, 81 – 86.

[10] H. Oertel, Prandtl’s Essentials of Fluid Mechanics (in Chinese), Science Press 2008.

[11] L. Ru-jiang, S. Zhao-wu, Effect of NATO Angle and Plate Velocity on Disturbance Frequency of Reactive Armor against Shaped Charge Jet (in Chinese), Chin. J. Energ. Mater. 2008, 16, 295–297, 318.

[12] D. E. Carlucci, S. S. Jacobson, Ballistics-Theory and Design of Guns and Ammunition, CRC Press, 2007.

[13] X. D. Zu, Z. X. Huang, Q. Q. Xiao, Dip Angle Effect on the Impact of Polybutylene Rubber Composite Target Resistance to Jet Penetration, 25th International Symposium on Ballistics, Beijing, China, May 17–21, 2010, pp. 1239–1246.

Posts · by **Yohannes** » Sat Apr 21, 2018 5:03 pm

Identification and modeling of the electrohydraulic systems of the main gun of a main battle tank

Citation: AIP Conference Proceedings 1493, 207 (2012); doi: 10.1063/1.4765491
View online: https://doi.org/10.1063/1.4765491
View Table of Contents: http://aip.scitation.org/toc/apc/1493/1
Published by the American Institute of Physics

AUTHORS: Luiz C. A. Campos^[a], and Luciano L. Menegaldo

[a] Defense Engineering Program, Military Institute of Engineering, Pça General Tibúrcio, 80, 22290-270, Rio de Janeiro, Brazil
Biomedical Engineering Program, The Alberto Luiz Coimbra Institute for Graduate Studies and Research in Engineering, Federal University of Rio de Janeiro, Av. Horácio Macedo, 2030, CT-Bloco H-338, Rio de Janeiro, Brazil

Article history:
9th International Conference on Mathematical Problems in Engineering, Aerospace and Sciences
AIP Conf. Proc. 1493, 207-214 (2012); doi: 10.1063/1.4765491
© 2012 American Institute of Physics 978-0-7354-1105-0/$30.00

INTRODUCTION

Black-box modeling is especially useful when analytical models derived from the physical background are difficult to be obtained. A set of candidate models is selected usually by overparametrization. Parameters, in such a model set, are basically viewed as vehicles for adjusting the fit to the data and do not reflect physical characteristics of the system [1].

The most usual models found in the literature can be summarized by the following discrete-time differences expression:

where q denotes the shift operator, and y, u and e are respectively output, input and noise terms [1], [2].

From Eq. (1) several models can be derived. Assuming F = D = 1 results in an ARMAX (autoregressive moving average with exogenous inputs) model, for F = C = D = 1 an ARX (autoregressive with exogenous variables) model is derived and for A = F = C = D = 1 a FIR (finite impulse response) model is obtained.

For ARX models Eq. (1) can be rewritten as [3]:

where ai and bi are the systems parameters, the terms y(k-i) and u(k-i) are the regressors, and k = 1, 2, …, N, where N is the number of observations. Equation (2) expresses the discrete-time polynomial form for ARX models. Non-linear regressors can be included in this model structure leading to NARX (nonlinear autoregressive with exogenous inputs) models. A very useful NARX structure is given by [3]:

where T_d is the highest time delay and F is a function of the regressors y and u and all their combinations (yu, y², u², … y^1-1u, yu^1-1, y¹, u¹) with non-linearity l, which is the highest power present in the terms of y(k).

Equations (2) and (3) represent deterministic models, however terms expressing noise can be easily added to the structure. ARMAX (Eq. 4) and NARMAX (nonlinear autoregressive moving average with exogenous inputs, Eq. 5) models treat noise as a moving average process:

where e(k) is the noise, and the terms e(k-i) are the regressors concerning the noise.

These models find wide practical applications since they can be readily identified using least square criteria.

Equations (4) and (5) can be rewritten as:

This paper presents an application of the described techniques in the identification of suitable models of the electrohydraulic systems of the main gun of a main battle tank (MBT). In a MBT, these systems are responsible for the rotation of the turret and for the elevation and depression of the barrel of the gun. The derived models are being employed in a broader research, aiming to reproduce such systems in a laboratory virtual main gun simulator.

METHODS

Triaxial acceleration and angular rate data were acquired simultaneously from systems inputs and outputs using two inertial measurement units (IMU) MicroStrain 3DM GX2. For the system responsible for rotating the turret (azimuth system), one IMU was positioned on the main axis of the gunner’s handle (Fig.1) and the other at the center of rotation of the turret (Fig.2).

For the system responsible for elevating and depressing the barrel of the main gun (elevation system), one IMU was positioned on the secondary axis of the gunner’s handle (Fig.3) and the other at the breech of the main gun (Fig.4).

FIGURE 1. IMU positioned for data collection of the inputs to the azimuth system.

FIGURE 2. IMU positioned for data collection of the outputs from the azimuth system.

FIGURE 3. IMU positioned for data collection of the inputs to the elevation system.

FIGURE 4. IMU positioned for data collection of the outputs from the elevation system.

As the turret has a continuous movement, random step inputs of different magnitudes were imposed for the identification experiments of the azimuth system. Due to the relative small displacement of the elevation system, complete cycles of elevation and depression of the barrel were used as inputs.

In the process of identification of the azimuth system, angular rate in the vertical axis of the turret was considered as the output variable. As input, both angular rate and angular displacement at IMU local vertical axis were used. Angular displacement was evaluated by both numerical integration of the angular rate or by the vertical component of IMU local lateral acceleration (y-axis). This last option was used to avoid the bias caused by the drift of IMU gyro.

For the elevation system identification, angular rate in the local lateral axis (y-axis) of the IMU placed at the breech of main gun was used as output variable. Similarly, both angular rate and angular displacement at IMU local vertical axis were used as input variables. Angular displacement was computed by the same procedure described above, except that, in this case, xaxis (longitudinal) was used instead of y-axis.

Three 300 seconds long experiments were carried out for the identification of the azimuth system. Five experiments consisting of five cycles of elevation and depression of the barrel were carried out for the elevation system.

Identification data collection was performed with a LabView application at 160 Hz sampling frequency, which was limited by the IMU hardware performance.

FIR and ARX architectures from first to fourth order were investigated. NFIR and NARX models with fifth degree nonlinearity were also explored. Residuals analysis [1] was carried out by:

Validation process was conducted for the models that best fit the experimental data set. In this process, experiments for each system were split in two periods. First period was used to estimate parameters and simulations were performed for both periods.

RESULTS

Figure 5 presents angular displacement at gunner’s handle, on the main axis, evaluated by numerical integration of the angular rate and by the vertical component of IMU local lateral acceleration, for one of the trials.

It can be observed a time-increasing drift for angular displacement found by numerical integration of the angular rate. This error is not observed when the vertical component of the lateral acceleration is used.

The number of candidate regressors largely increases when working with nonlinear structures. Thus, linear models were first investigated, trying to keep the order of the models as low as possible.

FIGURE 5. Gunner’s handle displacement (input) calculated by integration of angular rate (dashed line) and by decomposition of the acceleration in the y-axis (solid line).

Figures 6 and 7 present, respectively, ARX models from 1^st to 4^th order, identified using gunner’s handle angular displacement and angular rate as input. Both graphical and residuals analysis (see Tab. 1) suggest that a second order model would better suits into this application.

FIGURE 6. Turret angular rate (azimuth system), observed output and ARX models derived using gunner’s handle angular displacement evaluated by the vertical component of y-acceleration as input (100-135 seconds).

FIGURE 7. Turret angular rate (azimuth system), measured output and ARX models derived using gunner’s handle angular rate as input (100-314 seconds).

In an attempt to reduce bias due to the noise present in the observed output (measured output signal), FIR models were also investigated. Figure 8 presents FIR models from 1st to 4th order derived using gunner’s handle angular displacement as input and turret angular rate as output.

FIGURE 8. Turret angular rate (azimuth system), observed output and FIR models derived using gunner’s handle angular displacement evaluated by the vertical component of y-acceleration as input (100-135 seconds).

Furthermore, the use of gunner’s handle angular rate as input has proved to cause strong bias in the LSE. For this reason, gunner’s handle angular displacement evaluated by the vertical component of y-acceleration was preferred as input in subsequent procedures.

To diminish the influence of vibration, such systems are designed with variable gains: lower gains for small angular displacement and higher gains for large angular displacement [4]. Therefore, a nonlinear behavior was expected. Angular displacement (input) was plotted against turret angular rate (output) (Fig. 9) for better assessing the involved nonlinearities. Figure 9 suggests that a fifth-degree polynomial should provide a suitable model for the azimuth system.

FIGURE 9. Gunner’s handle angular displacement (input) x Turret angular rate (output).

NARX and NFIR models with fifth-degree nonlinearity were investigated. Figures 10 and 11 present, respectively, 1st and 2nd order NARX and NFIR models. As it can be observed nonlinear models have proved to better fit the observed output data. Residual analysis reinforces such conclusion (see Tab. 1). Figure 12 presents simulation obtained in the validation process of NARX models for the azimuth system. As mentioned, data from the experiment were split into 2 periods: 0-150 and 150-314 seconds. Parameters were estimated with data from first set and applied to both.

FIGURE 10. Turret angular rate, observed output and NARX models derived using gunner’s handle angular displacement evaluated by the vertical component of y-acceleration as input (azimuth system).

FIGURE 11. Turret angular rate, observed output and NFIR models derived using gunner’s handle angular displacement evaluated by the vertical component of y-acceleration as input (azimuth system).

FIGURE 12. Azimuth system NARX models validation, simulation 140-160 sec.

A similar procedure was carried out for the identification of the elevation system. Figures 13 and 14 present, respectively, the models that best fit and the simulation performed to validate them.

FIGURE 13. Elevation angular rate, observed output and models derived using gunner’s handle angular displacement evaluated by the vertical component of x-acceleration as input (elevation system).

FIGURE 14. Elevation system models validation, simulation 15-30 sec.

Table 1 presents residuals evaluated by Eq. (8).

TABLE 1. Residual analysis.

Figure 14 presents the simulation that validated the models for the elevation system. Similarly, the whole set of data was split into 2 periods: 0-20 and 20-41 seconds. Parameters were estimated using the first set of data and simulation was carried out through both periods.

CONCLUSION

This paper has shown that simple black-box models can be derived for the electrohydraulic system of the main gun of a main battle tank.

Nonlinear models better suit for both gun azimuth and elevation systems. These systems are designed to behave with low control gains for small displacement at the inputs and high control gains for large displacements [4].

The second order NARX and fourth order NFIR models have presented best fittings to the data set, respectively, for the azimuth and elevation system.

Other nonlinear architectures could be used to implement mathematical models of such system. However, as they are expected here to be used in realtime applications, one of the goals was to keep them as simple as possible.

The identified models are being employed in a broader research, aiming to reproduce such systems in a laboratory virtual main gun simulator.

ACKNOWLEDGMENTS

The authors wish to acknowledge the support received by the Brazilian Army and its Regional Maintenance Park headquartered in Santa Maria, Rio Grande do Sul, where the data acquisition took place. The authors are gratefully acknowledged to CAPES, FAPERJ, CNPq and FINEP, for financial support.

REFERENCES

1. L. Ljung, System Identification: Theory for the User, Englewood Cliffs: Prentice-Hall, 1987, pp. 13.
2. J. Sjöberg, Q. Zhang, L. Ljung, A. Benveniste, B. Deylon, P. Glorennec, H. Hjalmarsson, A. Juditsky, Nonlinear Black-Box Modeling in System Identification: a Unified Overview, Automatica, 31, 12, 1691-1724 (1995).
3. L. A. Aguirre, Introdução à identificação de sistemas: técnicas lineares e não-lineares aplicadas a sistemas reais, Belo Horizonte: Editora UFMG, 2007, pp. 65.
4. M. J. Griffin, 1990, Handbook of Human Vibration, London: Ed. Elsevier Academic Press, 1990, pp. 147

Posts · by **Yohannes** » Sat Apr 21, 2018 5:11 pm

The following are Lamoni's resources (they can be downloaded from the internet for free (iirc, someone please correct me if I am wrong here), so I will just list them here (thank you Lamoni!):

http://www.dtic.mil/get-tr-doc/pdf?AD=ADA178554
https://www.researchgate.net/profile/Ma ... ve-GPS.pdf
http://high-reliability.org/Self-Designing_HRO.pdf
https://hal.inria.fr/hal-00639673/document
https://www.stottlerhenke.com/wp-conten ... chards.pdf
https://hal.inria.fr/hal-00639672/document
https://www.researchgate.net/profile/Gu ... 3d4e68.pdf
https://academic.oup.com/milmed/article ... -2-106.pdf
http://www.dtic.mil/get-tr-doc/pdf?AD=AD0703337
https://academic.oup.com/milmed/article ... .5.387.pdf
http://www.dtic.mil/get-tr-doc/pdf?AD=ADA479999
http://www.dtic.mil/get-tr-doc/pdf?AD=ADA455154
http://www.dtic.mil/get-tr-doc/pdf?AD=ADA056736
http://www.dtic.mil/get-tr-doc/pdf?AD=AD0694049
http://www.dtic.mil/get-tr-doc/pdf?AD=ADA420195
http://www.dtic.mil/get-tr-doc/pdf?AD=ADA619261
http://www.dtic.mil/get-tr-doc/pdf?AD=ADA432176
http://hal.pratt.duke.edu/sites/hal.pra ... ations.pdf
http://www.academia.edu/download/342507 ... er__08.pdf
http://www.academia.edu/download/365142 ... rowski.pdf
https://pdfs.semanticscholar.org/2b74/f ... d1c73f.pdf
http://www.dtic.mil/get-tr-doc/pdf?AD=ADA344513
http://www.dept.aoe.vt.edu/~durham/2002-71.pdf
http://citeseerx.ist.psu.edu/viewdoc/do ... 1&type=pdf
https://web.ead.anl.gov/uranium/pdf/Duc ... tEffec.pdf
http://library.sciencemadness.org/lanl1 ... 326856.pdf
http://www.alternatewars.com/WW3/WW3_Do ... N-1980.pdf
http://www.alternatewars.com/BBOW/Balli ... L-RP-8.pdf
http://www.dtic.mil/get-tr-doc/pdf?AD=ADA403225
http://pds10.egloos.com/pds/200811/10/2 ... -jul99.pdf
https://www.cia.gov/library/readingroom ... 066239.pdf
http://www.ciar.org/ttk/mbt/armor/armor ... wehr01.pdf
http://www.ciar.org/ttk/mbt/armor/armor ... eMBT01.pdf
http://www.ciar.org/ttk/mbt/armor/armor ... etch95.pdf
http://www.gruppofrattura.it/ocs/index. ... /7455/4095
http://www.tms.org/pubs/journals/jom/97 ... -9705.html
http://www.ciar.org/ttk/mbt/armor/armor ... 1LIC02.pdf
https://academic.oup.com/milmed/article ... 10.757.pdf
http://citeseerx.ist.psu.edu/viewdoc/do ... 1&type=pdf
http://www.dtic.mil/get-tr-doc/pdf?AD=AD0524050
http://www.dtic.mil/get-tr-doc/pdf?AD=ADA229673
http://www.dtic.mil/get-tr-doc/pdf?AD=ADA098574
http://www.dtic.mil/get-tr-doc/pdf?AD=ADA090684#page=74
http://www.ciar.org/ttk/mbt/armor/armor ... shel96.pdf
http://ciar.org/ttk/mbt/armor/armor-mag ... sion02.pdf
http://www.ciar.org/ttk/mbt/armor/armor ... yugo99.pdf
http://www.dtic.mil/get-tr-doc/pdf?AD=ADA323153
http://www.dtic.mil/get-tr-doc/pdf?AD=ADA117710
http://www.dtic.mil/get-tr-doc/pdf?AD=ADA329222
www.dtic.mil/dtic/tr/fulltext/u2/a417314.pdf

Posts · by **Yohannes** » Sat Apr 21, 2018 7:14 pm

Developments in the processing of titanium alloy metal matrix composites

Correspondence to C. M. Ward-Close
0261-3069/94/02067-11 @1994 Butterworth-Heinemann Ltd

AUTHORS:

C. M. Ward-Close and M. R. Winstone
Structural Materials Centre, DRA Farnborough, Great Britain

P. G. Partridge
Interface Analysis Centre, University of Bristol, Great Britain

Article history:
Received 25 February 1994; accepted 15 March 1994

INTRODUCTION

The application of titanium in aerospace structures has increased steadily over the past 40 years so that now about 30% of the weight of modern aeroengines is composed of titanium alloys. A similar proportion of the structure of hypersonic aircraft is also titanium, both applications seeking to exploit the low density and good high-temperature strength of titanium alloys. Continuous development has evolved alloys such as IMI834 and Ti-1100 with maximum temperature capabilities in excess of 600^OC, good creep strength and resistance to oxidation. However, further development of the conventional alloys is increasingly difficult and alternative strategies are being explored.

Intermetallic alloys based on the titanium aluminides Ti₃AI or TiAl have the potential to maintain strength and oxidation resistance to over 900^OC. This is a very active area of research at present but a major problem is poor toughness of the intermetallic alloys. An alternative approach to extending the capabilities of titanium alloys is to produce titanium alloy metal matrix composites (TiMMCS). The introduction of a stable second phase in the material should make it possible to achieve substantial increases in specific strength and stiffness.

The choice of reinforcing phase must pay due regards to the high reactivity of titanium at elevated temperature and the property requirements of the composite. Since the design objective of the composite is to maximize strength and stiffness, it follows that the reinforcement should ideally be continuous fibres of high elastic modulus, high strength and low density. Some potential fibres are listed in Table 1. Silicon carbide fibres meet all the physical requirements and rule of mixtures calculations suggest that a 40 vol% unidirectional composite would have twice the specific stiffness of titanium and over 50% greater tensile strength. Furthermore, the properties of the fibres are not strongly temperature dependent, so it should be possible to develop composites with excellent high-temperature capabilities.

Table 1: Properties of reinforcing fibres for metal matrix composites

Such a material could find wide application in aeroengines and high-speed aircraft structures. Indeed, with such a large increase in specific properties it becomes possible to consider radical new designs. For example, the conventional gas turbine compressor disks could be replaced with Ti MMC rings, saving up to 75% of the weigh^1,2.

In this paper the current manufacturing routes for continuous fibre composites are compared. Continuous fibre reinforcement in a variety of matrices is considered. A detailed assessment is made of the potential for PVD processing since this process route permits close quality control and the composites possess unique features not found in composites made by other routes.

Solid-state processing of titanium-based continuous fibre composites

Reactive matrices, such as those based on titanium alloys, are limited to solid-state processing at relatively low temperatures (~1000^OC). For these systems, various competing fabrication routes have been proposed³. The routes differ in the way in which the fibre and metal are combined prior to pressing. The characteristics of these major fabrication routes are summarized in Table 2 and are discussed in more detail below.

Table 2: A comparison of various MMC fabrication processes

Foil/fibre (F-F) lay-up

Among the methods under development for titanium based MMCs, the most widely used is the foil/fibre lay-up process¹. Ti-alloy and Ti-aluminide composites are currently produced commercially by this method^4-6. Unidirectional mats of fibre are sandwiched between alternate layers of alloy foil (75-120 μm thick) prior to degassing and hot pressing (Figure 1). The fibre mat is produced either by weaving with a wire or ribbon cross-weave or the fibres are held in place with a fugitive organic binder, driven off during the degassing stage. Clearly, the process is best suited to the manufacture of fiat products, although reinforced tubes and corrugated structures have also been produced^7,8. When suitable alloy crossweave wire is not available, pure titanium wire may be used, but this can lead to detrimental chemical reactions during consolidation or service⁷. With the fugitive binder method, the binder may not be fully removed in the bake-out stage and can lead to contamination and a reduction in mechanical properties. The practical limit of fibre content for this process is about 45% (by volume), but the fibre distribution can be poor (Figure 2(a)) with many fibres touching^1,4,7,9,10. Other disadvantages of this process are the high cost of foils and the unavailability of some alloys which are difficult to work in foil form. For example, titanium aluminide foil can only be produced by chemical-milling of sheet, pack rolling, or by advanced rolling techniques in multiple stacked-roll facilities^11,13.

Figure 1: Schematic illustration of the foil/fibre process for the production of titanium-based MMCs.

The powder-cloth process¹⁴ is similar to the foil/fibre process, except that instead of foil, a cloth-like material is produced by rolling a mixture of alloy powder and an organic binder; the latter is driven off by heating in vacuum (Figure 1). In a similar method SiC fibre/reaction bonded Si₃N₄ ceramic composite is produced by winding SiC fibre with silicon-powder slurry and fugitive binder and solvent which is then nitrided¹⁵; this composite has up to 40% porosity. As with the F-F process, accurate positioning of fibres is difficult, and many touching fibres are observed (Figure 2(b)).

Fibre winding and plasma spray coating

Plasma spraying has been used to introduce the matrix material prior to hot pressing of titanium alloy and titanium aluminide based MMCs. Both plasma arc^14.16 and low-pressure plasma spraying have been used. In both forms, semi-liquid metal droplets are deposited onto a
single layer of fibres wound on a drum. Fibre spacings for plasma sprayed MMCs are usually better than those produced by the foil/fibre process (Figure 2(c)), but fibres or their thin protective coatings^1.16.17 may be damaged by the plasma spray droplets, which may be travelling at several hundred metres per second and cause severe thermal and mechanical spikes on impact. The coatings also tend to have greater roughness and porosity.

Figure 2: Transverse cross-section of unidirectional panels. (a) Three-ply Ti 14Al-21Nb/Si powder cloth composite^-1. (b) Eight-ply Ti 15V3Cr 3Sn 3Al F/F composite⁷. (c) Eight-ply Ti 6Al 2Sn 4Zr 2Mo plasma sprayed composite¹.

Consolidation of matrix coated fibre (MCF)

In theory, many disadvantages of the solid-state processing routes described above may be avoided by a new PVD coating and fabrication route¹⁸, which uses electron beam evaporation and vapour deposition (EBED) to precoat SiC fibres with a thick layer of matrix alloy (Figure 3) prior to consolidation into a finished composite (Figures 4 and 5). The matrix material is provided entirely by the coating, thus avoiding the expensive alloy product forms such as foil, powder or wire required for the other processes. In addition to these obvious benefits, in the manufacturing of MMCs coated fibres offer many advantages that are not available with other composite fabrication methods.

Figure 3: SEM of SiC fibre with 35 μm thick Ti-5A1-5V matrix coating made by the MCF process¹⁸

Figure 4: Schematic diagram of the matrix-coated fibre (MCF) route

Since each fibre is surrounded by matrix material, the fibre is protected during handling and the likelihood of damage to the fibre or to the thin diffusion barrier layer on the fibre surface (required to reduce chemical reaction with the matrix) is reduced. Furthermore, the volume fraction (V_f of SiC fibre in the finished MMC is determined only by the thickness of the fibre coating. Consequently, with coated fibres it is possible to obtain MMCs with up to V_f= 80 vol% SiC fibre (Figure 5). Compared with the alternative processes, the consolidation of MCF is less damaging to the fibres, with no risk of contamination from binder decomposition, and the spacing of the fibres can be exceptionally uniform, with no fibres touching.

Figure 5 Ti-5AI-5V/80% vol. fraction SiC fibre composite made by the MCF process¹⁸

Matrix coated fibres are particularly suited to net shape technology which permits the direct production of the final component shape and dimensions, and filament winding is a good example of an efficient fabrication method for composite components. An example of a circumferentially reinforced titanium alloy tube made from MCF is shown in Figure 6. This was produced by winding four layers of Ti-6A1-4V coated SiC fibre onto a 20 mm diameter tube and inserting this tube into another tube before sealing, evacuating and consolidating by hot isostatic pressure (HIP). The lower consolidation pressures associated with superplastic matrices have been exploited in F/F processing^19,20. Since PVD coatings tend to be fine grained, MCF composites may also be consolidated under superplastic conditions.

Figure 6: Section through wall of composite tube made by winding Ti5AI-5V/SiC MCF and HIPing

Volume changes during consolidation of continuous fibre composites

There are many aspects of fibre packing and composite consolidation that have received little attention in the literature but could have a significant effect on the properties of composites. For example, the volume change during consolidation is an important factor which must be considered during the design of a composite component. It can affect not only the external dimensions but also the local stresses imposed on the fibres. These stresses can cause fibre displacement (swimming), compression buckling or tension failure of the fibres. The magnitude of the volume change and the direction and extent of metal flow during consolidation is dependent on the composite manufacturing technique and on the type of fibre packing.

With careful winding of MCF it is possible to obtain a close-packed fibre array and a composite with the maximum theoretical green packing density of 90.7%.

The deformation required to consolidate a composite sheet can be considered in terms of the displacements along three orthogonal directions x, y and z, with x parallel to the fibre axis and z normal to the sheet plane. For most vacuum hot pressing or HIPing configurations displacement is allowed only along the z direction. Then, for the close-packed unidirectional fibre array, a reduction in thickness of 9.3% will lead to complete void closure and full density for any fibre volume fraction (Figure 7). The minimum fibre coating thickness required for complete filling of the voids (i.e. with fibres touching) is 3.2 μm for 100 gm diameter fibres. The coating thicknesses required for various volume fractions of fibre are shown in Figure 7. For typical values of Vf the coating thicknesses are about 30~40 gin. In the close-packed array, and with these coating thicknesses, it is unlikely that the coating on any fibre will become thin enough during consolidation to allow SiC fibre/SiC fibre contact.

Figure 7: Vertical contraction normal to fibre axes during composite consolidation of sheet when no displacement is allowed in the sheet plane versus SiC vol. fraction. A and B depict vertical and close-packed arrays of MCF. The effects of fibre spacing and foil thickness on contraction in the F/F composite are also shown

For monolayers of aligned MCF stacked in a vertical array, the packing density is much lower at 79%, and is independent on the relative orientation between the monolayers. In practice, the 21% reduction required to consolidate an MCF cross-ply lay-up (Figure 7) could be significantly reduced by cross-plying batches of unidirectional monolayers. For example, the orientation of the fibres might be changed every tenth ply.

In the foil/fibre lay-up process, using uncoated fibres, the green void volume depends on V_f, foil thickness and fibre packing geometry given by ratio R = h/w, where h and w are the vertical and horizontal distances between fibres. This is illustrated in Figure 7 which shows the reduction required along the z direction with increasing V_f. For a fixed R in the range 0.7-1.3 the displacement reaches a maximum of between 19% and 31%. In practice, due to the presence of a titanium cross-weave wire, or an organic binder, and due to lack of flatness in the foils, 'debulking' in the F-F process is typically about 50%. The distances over which metal must flow during consolidation of a F-F lay-up are greater than for the MCF composite. The practical limit of fibre volume fraction for the F-F process of about 45% is set by the minimum space between fibres that metal can be forced through under the consolidation pressure.

Clearly, compared with the consolidation of F-F layups, close-packed MCF requires less deformation and the matrix is required to move shorter distances. This should significantly reduce the possibility of matrix/fibre interface damage and fibre breakage in MCF composites.

Fibre distribution in metal matrix composites

The distribution and spacing of fibres can affect the constitutive behaviour of a composite²¹. The variation in fibre spacing can give rise to very high local stresses and to local shear zones or, in the extreme case of fibres touching, to an in situ crack. For a given fibre volume fraction, the overall fibre distribution geometry can have a marked influence on the orthogonal mechanical properties.

Fibres are equally spaced and fibre/fibre contact is not possible in the composites made with MCF fibres in the close-packed array (Figures 4-6). The inter-fibre spacings in the Ti-alloy MCF composite shown in Figure 6 are presented in histogram form in Figure 8, with the results of a similar analysis (taken from reference 1) of the micrographs shown in Figure 2 of Ti-alloy composites produced by other fabrication methods. These histograms show the much narrower distribution in fibre spacings obtained with the MCF composite compared with composites made by other methods. Closely spaced and touching fibres have also been reported in Al-alloy monofilament tape²². Of even greater practical importance is the fact that in the MCF composite no fibre spacing was less than 20 μm, whereas in the composites made by the other three techniques many fibres were either touching or had less than 10 μm separation.

Figure 8: Frequency versus magnitude of edge-to-edge inter-SiC fibre spacing for composites made by different routes

To investigate the effects of small interfibre spacing the MCF process was used to coat two batches of 145 μm Textron SCS6 SiC fibre with Ti-6Al-4V alloy, one with a coating thickness of 10 μm and another with a coating thickness of 35 μm. These coated fibres were mixed with uncoated fibres and consolidated to produce an MMC with a wide range of fibre/fibre spacings. In sections normal to the fibre axes radial cracks were apparent in the carbon-rich surface layer at the position of closest approach between fibres (Figure 9). The incidence of cracking was found to increase with decreasing interfibre spacing (Figure 10). No fibre cracks were found where the spacing was greater than 20 μm. This confirms a similar result reported for SCS6 fibre in the beta titanium alloy Ti-15V-3Cr-3A1-3Sn⁷. A theoretical analysis of cracking induced by thermal residual stresses²³ concluded that radial cracks adjacent to fibres were more likely if the fibres had a cubic rather than a hexagonal array, and that cracking was also more likely between fibres having a smaller than average separation.

Figure 9 Radial cracks in a C-rich surface layer on a Textron SCS6 fibre (shown at A) in closely spaced fibres in a composite

Figure 10 Frequency of fibre surface cracks in Textron SCS6 fibres versus edge-to-edge fibre spacing

The DRA SIGMA 100 Ixm SIGMA SM1240 (carbon-titaniumdiboride coated SiC) fibre does not appear to suffer from radial cracking of the type found in the 145 μm carbon-coated fibre. However, commercially produced six-ply laminate (SM1240/Ti-6-4) showed considerable variability of fibre distribution, with areas of poor fibre distribution with many touching fibres and other areas of good fibre distribution. It is likely that poor fibre distribution will lead to degradation of tensile properties due to stress concentration effects, or associated manufacturing defects.

The effect of overall fibre distribution geometry on mechanical properties has been investigated by Brockenbrough et al.²⁴ for boron fibres in 6061 aluminium alloy. Their theoretical analysis, which assumed perfect fibre/matrix bonding, showed that, whereas longitudinal properties are little affected by fibre distribution, the transverse tensile behaviour is strongly influenced by the fibre packing. Wisnom and Li²⁵ have used a finite element method to model the transverse tensile behaviour of SCS6 SiC/Ti-6AI-4V composite. The initiation of fibre-matrix debonding was taken as the criterion for composite failure, and the resultant transverse failure stresses for both 35% and 40% fibre volume fraction and two different fibre packing geometries are given in Table 3.

Table 3: Predicted transverse failure stress

The results show that for a given fibre volume-fraction the transverse failure stress is always greater when the fibres are closer together in the ply and the plies further apart. This is an important result since low transverse properties often limit the application of continuous fibre reinforced composites. A favourable geometry of this type might be achieved by combining MCF with either foil or plasma-sprayed interlayers. The matrix coating on the fibres would be essential to maintain planar fibre spacing and to ensure that each fibre/fibre gap was completely filled with metal.

Note that in composites manufactured by melt infiltration routes such as squeeze or pressure casting, the composite microstructure is characterized by a high variability fibre spacing and many fibres in contact, e.g. in the metal matrix composites C-fibre/AI-6061²⁶, C-fibre/Mg²⁷, SiC-fibre/A1-6061²⁸ and the intermetallic matrix composites FP A1203/Ni3A1²⁹ and PRD fibre (Al203-20wt% ZrO2)/Fe3Al³⁰. A similar fibre distribution was found in SiC fibre/reaction bonded Si3N 4 ceramic composite produced by a silicon-powder cloth layup process¹⁵. In LAS matrix/SiC fibre composites a circumferential gap can occur around the fibres³¹.

Characteristics of the PVD process

Physical vapour deposition (PVD) includes evaporation and sputter techniques^32,33 and is used to provide coatings in a wide variety of materials. There has been a steady improvement in both high-rate magnetron sputter sources and electron beam (EB) evaporation sources. Compared with EB, magnetron sputter sources have an advantage for coating fibres since they can be planar, curved or cylindrically hollow in shape and positioned vertically or horizontal and upward or downwardfacing.

Using high-power EB guns (up to 600 kW) metal deposition rates of up to 3000 μm/min can be obtained³⁴ compared with 1-5 μm/min for sputter sources. The condensing species have much higher energies in sputtering than in EBED^35,36. This can lead to greater thermal stresses in the deposit and the substrate and also affect the deposit microstructure. For example Ti films deposited at 77 K with a grain size of 5-15 nm had the normal room temperature hcp crystal structure using EBED but a bcc crystal structure using sputter deposition³⁷. Ion beam (plasma) assisted deposition can be used with both PVD techniques and is useful when independent control of vapour flux, gas pressure or substrate temperature is required³⁸. Plasma processing is a powerful technique for modifying the microstructure and properties of thin films and coatings, but is not discussed further in this review.

Sputter deposition

A great variety of metallic, intermetallic and ceramic films containing together almost every element in the Periodic Table have been produced by the two most important techniques, bias and reactive sputtering³⁹. These methods are normally limited to thin films because of the slow deposition rates. However, for coatings on fibres of only 30-40 μm or for thin 100-500 nm films required to toughen brittle matrices (see below) the rates may be acceptable. Furthermore, this process has important advantages compared with CVD when depositing alloys:

⁴⁰

⁴¹

Because of the higher deposition energies more point defects and dislocations may be present in these deposits than in EBED deposits. Consequently, supersaturated alloys may be less stable or at low temperatures the residual stresses may be greater. There is evidence that sputtered Ti coatings deposited at a low temperature retain high compressive stresses compared with low tensile stresses in CVD films⁴² and the PVD film had higher surface fracture strength.

Electron beam evaporation and deposition ( EBED)

Although EBED offers much higher deposition rates than can be obtained for magnetron sputtering³⁴, there is no longer a clear advantage, as suggested elsewhere^43,44, in the production of alloy coatings or in the ability to scaleup compared with magnetron sputter coating. Like sputter coating, this process offers the possibility of controlling the composition and morphology of the deposits and of tailoring the deposit for optimum mechanical properties. Electron beam evaporation and vapour quenching leads to very rapid equivalent cooling rates. The rates measured by secondary dendrite arm spacing or by calculation are much greater than for melt atomization^45,46 and temperature gradients and variable microstructures found in melt atomized powder are avoided by vapour quenching. Highly supersaturated and metastable alloys can therefore be produced and processed to give novel microstructures with exceptionally uniform nanometre scale grain microstructures which may be impossible to produce by any other means. These very small grain sizes may have important implications for the processing of materials^20,47,48.

Figure 11 Temperature versus vapour pressure of the elements

The rate of evaporation from an electron beam heated source is dependent on EB gun power, source temperature and the vapour pressure of the element (Figure 11) or on the fugacity of the elements during co-evaporation of alloys⁴⁹. Co-evaporation from a single source, with continuous alloy feed to the bath, is only possible if the vapour pressures of the alloying elements are within about two orders of magnitude of each other at the evaporation temperature. This has the disadvantage, for example, that the refractory metals present in many high temperature Ti-alloys cannot be co-evaporated. Similarly, oxide, nitride and carbide impurities in Ti-alloys will remain in the evaporating source, but this can lead to a cleaner and tougher deposit⁴⁶.

However, the use of multiple evaporation sources (either EB or radiant heat) and vapour mixing allows the deposition of almost any alloy combination as shown in Figure 11. Provided the vapour flux rates are not exceptionally high, the mean-tYee paths of atoms evaporated from separate sources will be sufficiently long (~ 10 mm) for overlapping vapour streams to mix completely. It is possible, for example, to use vapour mixing to produce a wide range of different alloys based on aluminium, titanium and magnesium containing high melting point elements^50,51 with unique microstructures characteristic of very rapid quenching. These alloys would be particularly attractive as composite matrices. The manufacture of such composites may only be possible via the MCF route.

Matrices suitable for the MCF process

Research centres in Germany, Japan and the USA have major programmes directed towards identifying future structural materials with high specific strength and stiffness at room and elevated temperature^52-54. Many of these materials would be attractive as composite matrices. Although it has been suggested that the matrix has little effect on the properties of unidirectional composites⁵⁵ this is not true of transverse properties or of composites loaded at high temperatures²⁷. The materials can be divided into three groups according to their service temperatures. For example, Al-alloys, Mg-alloys and amorphous alloys up to about 300^OC, ordered intermetallics in the range 600-1000^OC and ceramics above 1000^OC. There is also what might be considered a unique
class of composites based upon superconducting ceramic matrices on a conducting wire core. The MCF technique offers the possibility of depositing these materials onto fibres to form a composite matrix with a complex tailored microstructure.

Composite toughening

Continuous fibre composites usually have increased longitudinal strength and modulus compared with the unreinforced matrix. However, the toughness depends on the matrix and the fibre/matrix interface. Compared with the unreinforced state, composites with ductile metal matrices have lower toughness and composites with brittle matrices (aluminides and ceramics) have slightly higher toughness but, in all composites, toughness values are much lower than the minimum values considered acceptable for critical structural components. This low toughness can be explained by the fracture mechanisms in the composite and the energy absorbed during fracture.

In ductile metal matrix composites cracks initiate in the fibre/matrix interface, often associated with a brittle reaction layer. In composites with brittle matrices (intermetallic or ceramic) the critical flaw size is usually smaller than the interfibre spacing, and cracks initiate in the matrix rather than in the fibre/matrix interface. Toughening of these composites involves increasing the energy absorbed in crack growth. In the toughening of continuous fibre reinforced composites the fibre/matrix interface strengths plays a critical role. Ideally, for toughness in brittle matrix composites at room temperature, the fibre/matrix interface strength should be strong enough to ensure load transfer across the fibre/matrix interface until matrix fracture, and then be weak enough to allow debonding, so preventing the composite behaving as a monolithic material^56,57. At elevated temperatures when the brittle matrices become ductile, or for ductile MMCs at all temperatures, toughness is obtained when the fibre/matrix interface is strong enough to allow for high load transfer.

Current composite toughening mechanisms are based upon microcrack branching in the matrix and upon debonding at the fibre/matrix interface⁵⁸. The interface strength, frictional load transfer and fibre pull-out stresses should not be too high. Fibres provide crack bridging behind the crack front, either by pull-out or by ductile extension^59,60. (Note that for other properties such as creep strength, a high interface strength is required²³.) Small-diameter fibres and a reduction in grain size will favour increased toughness in brittle composites⁶¹. On a coarser scale, ductile fibres, layers or particles have been used to toughen ceramics, cermets, and intermetallic compounds, such as Ti-aluminides^62-67. Aluminium coatings have been applied to 100 μm diameter SiO₂ fibres in SiO₂ glass to increase the toughness⁶⁸. Another technique for introducing ductile toughening regions into brittle matrices involves embedding fine tubes (1.2-2.0 mm o/d) of CP Ti or Type 304 stainless steel containing particulate reinforced SiC/AI 6061 alloy or B₄C/NiA1 alloy into the corresponding composite matrix. These ductile tubular regions lead to an order of magnitude increase in the Charpy impact values⁶⁹. However, for titanium alloys, major matrix composition changes to increase the amount of ductile phase, e.g. the addition of up to 17 at% Nb to Ti-aluminide will increase density and may adversely affect the oxidation behaviour. Coarse ductilizing layers are much less effective under cyclic loading than under monotonic loading⁶⁷ and may reduce the overall specific strength properties. It has been suggested that for ceramics a dual fibre coating is required with an inner coating controlling the fibre/matrix debonding and sliding and an outer coating controlling the matrix interactions³¹. More stable coatings such as Nb, Mo, Pt and NbAl have yet to be evaluated.

At present it is almost impossible to influence on a fine scale the operative fracture modes in a composite, since these are dictated by the matrix and the processing route, although an oxide-stablized tetragonal zirconia matrix produced by PVD⁷⁰ is an example of a possible tough ceramic matrix. A major advantage of the M CF technique when applied to such composites may therefore prove to be the ability to manufacture a graded matrix composition to minimize thermal stresses^32,71 or a microlaminated matrix for toughness⁷².

Figure 12 Titanium/yttrium nanolaminate coating on SiC fibre produced by the MCF process. (a) Coated fibre (showing tungsten wire (A), SiC (B) and nanolaminate coating (C)). (b) Microstructure of laminated Composite. (c) Evidence of crack deflection in fractured laminated composite

Multi-layer microlaminates of Fe, A1 and Ti have been produced by the EBED route using a multiple evaporation source system⁷³ and this type of microstructure has great potential for the production of unusual physical and mechanical properties^47,74. Microlaminates consisting of layers of TiC/TiB₂ and TiC/Ni have also been produced by the EBED process^75,76. These layers can be retained after working or be broken into platelets⁵⁶. Layers increase the fatigue strength of sputtered metallic coatings⁷². Both layers⁷⁷ and platelets⁷⁸ have led to increased toughness in ceramics. An example Of MMC with a microlaminated matrix is shown in Figure 12. Titanium alloy/yttrium oxide deposited on SiC fibre by PVD (Figure 12(a)) became a microlaminated matrix after consolidation (Figure 12(b)). Confirmation of crack deflection and delamination by the microlayers during fracture of the composite is shown in Figure 12(c). These microstructures could allow better control over fibre/matrix interface strengths and specific toughening mechanisms (e.g. ductile or brittle fracture) to be selected and optimized for the matrix. Solid-state amorphization has been observed in many PVD multilayer alloy systems and may find use in these fibre coatings^79,80. Flexibility in the tailoring of the matrix also may allow a combination of properties to be selected for a particular application, e.g. longitudinal and transverse properties or longitudinal tensile and creep strength or high toughness.

Costs and future developments

The size of the market for continuous fibre composites, as for many new materials, will be dictated by the cost/benefit balance, with a rapid reduction in cost predicted as the market increases. Isolating the critical cost factors for composites under development is very difficult since many data are proprietary, but costs will clearly depend on the particular process route and matrix chosen.

At present, costs are dominated by the cost of the fibre. DRA SIGMA carbon/boron protected 100 μm SiC fibre (SM 1240) is currently about £8000-10 000/kg, and, although Textron SCS6 140 μm fibre is normally only available incorporated into an MMC or pre-form product, the cost is thought to be in the region £4000-8000/kg. Current industry estimates suggest that fibre cost could fall to as low as £1000/kg, for a plant producing 5000-10 000 kg/year. Fibre costs quoted in various publications in the last ten years are now seen to have been very misleading and much lower than current costs. This can be only partly due to the effects of inflation or the failure of hoped-for production economies to materialize. Even a recently published review⁸¹ gave a completely erroneous figure of $300/kg for SiC monofilament. These over-optimistic and unrealistic predictions are more likely caused by a serious under-estimate of R&D costs and time scales for achieving acceptable process routes and properties for the composites. This leads to underinvestment by the funding agencies and a credibility gap between the materials suppliers and their customers, as emphasized elsewhere^82,83.

The foil/fibre route is reported to be the most economical for Ti-alloy composites¹⁷. Low-pressure plasma spraying is more expensive but has the advantage of high deposition rates and relatively low capital investment requirements⁴⁹. These must be set against the need for alloy in the form of either wire or powder, both of which may be difficult to obtain, and some disadvantages for composite manufacture. The PVD plant cost is likely to be high.

The additionM costs associated with the MCF process are mainly those of the capital and running costs of the PVD plant. These are estimated to add about £900/kg to the cost of the MMC (based on the DRA pilot plant). Composite consolidation costs are particularly difficult to estimate because of the variety of operations involved. However, at present, in the foil/fibre process, about 12% of MMC cost is attributed to fibre cost and 88% to foil and consolidation costs (including plant, labour, profit, etc.). Fabrication costs for the MCF route are likely to be less than for the fibre/foil route, but by how much less is not known. An additional factor that might be taken into account is the potential that the MCF process has for the manufacture of a wider range of sophisticated matrix alloys with rapidly quenched or laminated microstructures. This potential may justify the greater investment in MCF plant.

The only viable manufacturing techniques for matrices based on many of the metallic and non-metallic materials discussed above will most likely be based on PVD or CVD processing. The latter is costly in terms of the plant requirement, the processing time and the complexity of the processing and will be limited to certain alloys and microstructures⁸⁴.

Future development of the PVD/MCF process is likely to result in improvements in metal-collection efficiency, which will reduce the overall costs. A dedicated fibre coating platat could utilize much higher electron beam power and greatly increase the fibre throughput. Production of continuous alloy foils in Cu, Hf, Ni, Zr, Ti-6A1-4V, 20%Cr stainless steel and Inconel 600 by EBED has been demonstrated 8~87. Recently, EB equipment manufactured by Leybold in Germany for coating turbine blades has demonstrated the viability of using advanced air-load/vacuum-lock devices and automated handling techniques to evaporate continuously for up to 80 h using EB guns rated at 750 kW total beam power⁴⁹. The potential exists to combine sequential electron beam evaporation and gas atom sputter techniques to deposit unique matrix alloy compositions and multiple-layer coatings based on metallic and intermetallic compounds and ceramics. It is possible that the high initial costs of an automated PVD plant could be offset by cost savings downstream as a consequence of precise quality control, added value and increased market potential for a wide variety of composites withunique properties. It is worth noting that the costs for ceramic composites are low for the raw materials but high for the processing. Greater cost savings may therefore arise in the processing of ceramic composites made by the MCF process.

Conclusions

The potential benefits of titanium-based continuous fibre MMCs have been amply demonstrated, both here and elsewhere, and for some new materials, such as intermetallic compounds, fibre reinforcement may be the only viable way to exploit their outstanding high-temperature capabilities. Development of continuous fibre MMC manufacturing technology is continuing to receive considerable research attention. A number of promising fabrication methods have been demonstrated, such as plasma spraying and PVD matrix coating of fibres, which have the potential to greatly reduce the cost of finished MMCs. As with all emerging technologies, the next stage, the transfer of the technology from the laboratory to pilot-scale commercial production, will be expensive, and depends critically on the cost/benefit analysis for these new materials. It remains to be seen whether the perceived market is sufficient to justify the required investment.

Acknowledgements

The authors wish to thank Dr B. Viney and R. W. Gardiner for many helpful discussions on the EBED process.

References

1 MacKay, R. A., Brindley, P. K. and Froes, F. H. JOM 1991, May 23-29
2 Driver, D. In High Temperature Materials for Power Engineering Part II, 1990, pp. 883-902
3 Smith, P. R. and Froes, F. H. JOM, 1984, March, 19-25
4 Mittnick, M. A. In Proc. 21st SAMPE Conference, 1989, pp. 647-658
5 Mittnick, M. A. In Proc. 22nd International SAMPE Technical Conference, 1990, 774~786
6 Textron plant makes titanium MMC'S, Advanced Materials & Processes 1992, April, 9
7 MacKay, R. A. Scr Metall., 1989, 24, 16-172
8 Chang, D. J. and Kao, W. H. SAMPEJ. 1988, March/April, 13-17
9 Brindley, P. K. et al. In Fundamental Relationships Between Microstructures & Mechanical Properties of Metal-Matrix Composites, Eds Liaw, P. K. and Gungor, M. N., 1990, The Minerals, Metals & Materials Society, pp. 387-401
10 Lerch, B. A. and Hull, D. R. As-Received Microstructure of SiC/Ti-15-3 Composite, Technical Memorandum 100938, NASA, 1988
11 Jha, S. C. et al. Adv. Mater. Processes 1991, 4, 87-90
12 Aviation Week & Space Technology 2 September 1991, 65
13 Peters, J. A. and Blank-Bewersdoff, M. Materials & Design 1992, 13, 83-86
14 Stephens, J. R. In Proc. A1EE/ASME/SAE/ASEE 24th Joint Propulsion Conference, 1988, Boston, Massachusetts, AIAA, Washington, DC, pp. 1 10
15 Bhatt, R. T. In Proc. 21st University Conference on Ceramic Science, Eds Tressler, R. E. et al., 1985, Pennsylvania State University, Plenum Press, pp. 675-686
16 Chou, T. W., Kelly, A. and Okura, A. Composites3 1985, 16, 187-206
17 Adv. Materials & Processes 1992 1, 18-19
18 Ward-Close, C. M and Partridge, P. G. J. Materials and Science 1990, 25, 4315-4323
19 Lloyd, D. J. J Materials Science 1984, 19, 2488-2492
20 Uchiyama, Y., Hasaka, M. and Koga, H. Materials Transactions 1990, 31, No. 2, 158 161
21 Wimolkiatisak, A. S. and Bell, J. P. In Proc. 21st International SAMPE Technical Conference, pp. 371-382
22 Metal Matrix Composites, BP brochure, 1988
23 Lu, T. C., Yang, J. and Sou, Z. Acta Metall. 1991, 39, 1883-1890
24 Brockenbrough, J. R., Suresh, S. and Wieneche, H. A. Acta Metall. 1991, 39, 735-752
25 Li, D. S. and Wisnom, M. R. Micromechanical Modelling of Unidirectional SiC/Ti6-4 Under Transverse Tension, Final Report Contract 2034/93, University of Bristol, 1991
26 Mortenson, A. Mater. Sci & Engng. 1991, A135, 1 11
27 Kagawa, Y. and Nakata, E. J. Mater Science Let. 1992, 11, 176-178
28 DeBondt, S., Froyen, L. and Deruyttere, A. Mater. Sci & Engng. 1991, A135, 29-32
29 Nourbakhsh, S. et al. Acta Metall. Mater. 1992, 40, 285 294
30 Nourbakhsh, S. and Morgolin, H. Mater. Sci & Engng, A144, 133-141
31 Evans, A. G. and Marshall, D. B. Acta Metall. 1989, 37, 2567- 2583
32 Suganuma, K. ISIJ 1990, 30, 1046-1058
33 Upadhya, K. JOM, 1989, June, 6-9
34 Schiller, S., Jaesh, G. and Neumann, M. Thin Solid Films 1983, 110, 149-164
35 Schiller, S., Heisig, U. and Goedicke, K. Thin SolidFilms 1978, 54, 33-47
36 Schiller, S.et al.: Thin Solid Films 1979, 64, 455-467
37 Konk, J. et al.: Thin Solid Films 1992, 207, 51-53
38 Smidt, F. A. Inter. Mater. Revs. 1990, 35, 61 128
39 Thornton, J. A. Surface Engineering 1986, 2, 283 292
40 Maissel, L. I. and Glang, R. In Handbook of Thin Film Technology), McGraw-Hill, New York, 1970, pp. 440
41 Maoujoud, M. et al. Sur/ace and Coatings Technology 1992, 52, 179 185
42 Rickerby, D. S. and Bull. S. J. Surjace and Coatings Technology 1989, 39/40, 315-328
43 Rickerby, D. S. and Scott, K. T. In Proc. Con)(. on High Temperature Materials in Engineering, 1989, London
44 Barrell, R. and Rickerby, D. S. Metals and Materials 1989, August, 468 475
45 Froes, F. H. and Suryanaryana, C. JOM 1989, June, 12 17
46 Bianchi, L. JOM 1991, May, 4547
47 Houston, J. and Feibelman, P. Advanced Materials & Processes 1991, March, 31-34
48 Siegel, R. W. MRS Bulletin 1990, October, 60-67
49 Lammermann, H. and Feurstein, A. Proc. International Gas Turbine Congress, 1991, Yokohama, Japan, pp. 269-281
50 Froes, F. H., Kim, Y. and Hehman, F. JOM 1987, August, 14-20
51 Suryananyana, C. and Froes, F. H. Intern. Mater. Rev. 1991, 36, 85-123
52 Fleischer, F. L. JOM 1985, December, 1/20
53 Wiedemeier, H. and Singh, M. J. Materials Science 1991, 26, 2421-2430
54 Froes, F. H. Material," & Design 1989, 10, 110-120
55 Stephens, J. R. and Nathal, M. V. In Superalloys 1988, Eds Reichman, S. et al., The Metallurgical Society, 1988, pp. 183-192
56 Bickerdike, R. L. et al. Inter. J. Rapid Solid. 1986, 2, 1-19
57 Evans, A. G. In Ceramic Mierostructures '86 Role of lnter/iwes, Eds J. A. Pask and J. A. Evans, Plenum Press, New York, 1986, pp. 775 794
58 Gupta, V., Argon, A. S. and Cornie, J. A. J Materials and Science 1989, 24, 2031 2040
59 Evans, A. G. Mater Sci & Engng 1988, 105, 65-75
60 Hayhurst, D. R., Lechie, F. A. and Evans, A. G. Proc. Roy, Soc. 1991, 434, 369-381
61 Gisaffe, S. J. Adv. Mater. & Proc. 1990, 1, 43-94
62 Deve. H. E. and Maloney, M. J. Acta Metall. 1991, 39, 2275 2284
63 Lu, T. C., Evans, A. G. and Hecht, R. J. Acta Metall. 1991, 39, 1853- 1862
64 Xiao, L. et al. Mater. Sci. & Engng 1991, A144, 277-285
65 D6ve, H. E. et al. Acta Metall. Mater. 1990, 38, 8, 1491-1502
66 Bose, A. and Lankford, J. Advanced Materials & Proeesses 1991, 7, 18-22
67 Roe, K. T. V. Odette, G. R. and Ritchie, R. O. Acta Metall. Mater. 1992, 40, 353-361
68 Prewo, K. M. In Proc. 21st Universi o, Con/erence on Ceramic Sciences, Eds Tressler, R. E. et al., Pennsylvania State University, Plenum Press, 1985, pp. 529-547
69 Nardonc, V. C., Strife, J. R. and Prcwo, K. M. Mater. Sci. & Engng, 1991, A144, 267-265
70 Qadri, S. B. et al. Sur[ace and Coatings Technology 1991, 49, 67-70
71 Yamada, T. et al. High Temp. Tech. 1987, 5, 193-200
72 Beregovsky, V. V. et al. Sur[ace & Coating* Technology 1991, 48, 13-18
73 Bickerdike, R. L. et al. Int. J. RapidSolidoqcation 1984, 1, 305-325
74 Kelly, A. Phil. Trans. R. Soc. Lond. 1987, A322, 409-423
75 Movchan, B. A. et al. Thin Solid Films 1982, 97, 215-219
76 Sans, C. et al. Thin Solid Films 1983, 107, 345-351
77 Clegg, W. J. et al. Nature 1990, 347, 455-457
78 Janssen, R. and Heussner, K. Powder Met. Intern. 1991, 23, 241 236
79 Matsuura, M. et al. Mater. Sci. & Engng 1991, AI33, 551-554
80 Makowiecki, D. M., Jankowski, A. F. and Kernan, M. A. Magnetron Sputtered Boron Films and Ti/ B multilayer Structures, Report UCRL-JC 103223, Lawrence Livermore National Laboratory, 1990.
81 Terry, B. and Jones, G. Metal Matrix Composites, Elsevier, Oxford, 1990
82 Feest, A. Metal*" and Materials 1988, May, 273-278
83 Williams, J. In Proc. 27th Joint Propulsion Con[krenee, 1991, AIEE, Sacramento
84 Bashford, D. P. Metals & Mater. 1992, 8, 79-84
85 Hehmann, F., Sommer, F. and Predel, B. Mater. Sci. & Engng 1990, AI25, 249-265
86 Smith, H. R., Kennedy, K. and Boericke, F. S. J. Vacuum Science and Technology 1970, 7, 48-51
87 Hughes, J. L. Metals Engineering Quarterly 1974, February, 1-5

Posts · by **Yohannes** » Sun Apr 22, 2018 4:39 pm

Out of character sentence: going to post this (and the next article) because hopefully they will be of help to the person who asked (by telegram) about the subject (if they are not helpful, please do tell by telegram and I'll see what I can do. Thank you!)

CONSIDERATIONS REGARDING THE NEXT GENERATION OF BALLISTIC PROTECTIVE EQUIPMENT SUCH AS “LIQUID BODY ARMOR”

BULETIN ŞTIINŢIFIC
Nr. 2 (38) 2014
Nicolae Balcescu Land Forces Academy; Sibiu Vol. 19, Iss. 2, (2014): 97-102.

AUTHORS:

LuminiţaCristina ALIL
aegyssusatm@yahoo.com

MILITARY TECHNICAL ACADEMY, BUCHAREST, ROMANIA

SimonaMaria BADEA
smsbadea@yahoo.com

SCIENTIFIC RESEARCH CENTER FOR CBRN DEFENSE AND ECOLOGY, BUCHAREST, ROMANIA

Florin ILIE
ilieflorinv@yahoo.com

“NICOLAEBĂLCESCU” LAND FORCES ACADEMY, SIBIU, ROMANIA

ABSTRACT

The development of ballistic protective equipment is marked by constant improvement of the performance and the level of protection while reducing the weight of the ballistic protection structures and individual armor. Generally, the protective systems are designed and implemented by the optimum combination of different types of materials, such as ceramics, metals, polymers, fibers and composite materials, in order to meet the specific requirements of the various types of threats. Thus, the ballistic protection is achieved by material layers with specific functions. Considering these aspects, research nowadays focuses on implementing simpler structures in terms of construction, enabling high mobility while keeping the same performance level or even a higher level of protection than previous generations. This article briefly presents some of the basic materials for one of the future solutions, the “liquid body armor.”

1. Introduction

Over time, people have used various materials to protect themselves from wounds in battle or other type of dangerous situations. The first clothes and shields were made of animal skins. With the development of civilization, people began using shields of wood and metal. Metal armor is still associated with heavy medieval knight image. By the 1500s, the advent of firearms made metal armor to become ineffective. The only real protection was offered by natural stone walls or obstacles (rocks, trees or ditches). Among the first records on the use of light armor is included the use of protective clothing made of natural silk in medieval Japan. This solution proved to be inefficient from a study conducted in the late nineteenth century by Americans.

A second generation of body armor appeared during the Second World War and was made of ballistic nylon. This type of armor was only partially effective because although it provided good protection from shrapnel, it was ineffective against bullets. Once the polyaramid fiber (Kevlar) was
discovered in 1970 within the company of DuPont, the manufacturing industry of ballistic protective equipments knew new directions.

Significant advances achieved by research in chemistry in recent decades have resulted, inter alia, in obtaining polymers whose strength exceeds up to ten times that of steel, and ceramics whose hardness is approaching that of diamond [1]. Besides, the development of ballistic protective equipment is marked by the continuous improvement of the performance and the level of protection while reducing the weight of the ballistic protection structures and individual armor.

Generally, the protective systems are designed and implemented by the optimum combination of different types of materials, such as ceramics, metals, polymers, fibers and composite materials, in order to meet the specific requirements of the various types of threats [2]. The choice of materials and geometry, as well as how they are assembled, represent key factors in the design of armor. Each material component serves a specific purpose not only in stopping the projectile kinetic energy or in mitigating the effects of blast shock waves, but also in maintaining the structural integrity of the ballistic protection structure.

Diagram in Figure no. 1 describes the theoretical basic configuration of an armor, which includes both high-density ceramics and porous materials, but also fibers, coatings, polymeric binders and adhesive joints. Complex architecture shown in Figure no. 2 uses several different materials and methods of assembling, so that the ballistic protection function is to be achieved by material layers with specific functions.

Figure no. 1 Schematic representation of the cross-section of a typical armor tile specifically used for the ballistic protection of a vehicle, which highlights the complexity of the armor architecture, generally speaking. The assembly comprises different types of materials, such as porous and dense ceramics, composite fibers, thermoplastic polymers, and adhesives. DEA – diethanolamine (this has the role of surfactant/corrosion inhibitor) [3].

Figure no. 2 Ballistic protective vest – layer structure [4].

Furthermore, the structural configuration of a ballistic protective vest can be schematized as shown in Figure no. 2. A ballistic protective vest has in its structure the following types of materials:

The behavior of a protection assembly against a specific threat is not, however, the mere sum of its parts responses. Thus, an integrated approach considering experimental and calculation aspects, allowing the variation of the microstructures of materials and, respectively, the dynamic characterization of materials behavior at high speeds, by themselves and as part of a protective structure, may underlie the development of better protection materials with lower density. Following, we present some potential materials for the realization of future generations of high-performance ballistic protection structures.

2. Generalities

One of the specific threats in theaters is explosion. The effect of an explosion which also represents the true threat for a military is the shock wave that propagates at very high speed. The human body is not homogeneous, it has many components with completely different structures. Some of the tissues are soft, others are more dense (depending on the content of water / liquid). Each organ, due to its physical characteristics, has a certain inertia, which makes the body a group of organs that are in a relative motion to each other, when applying a force in a specific area of the body. In the absence of a protective rigid structure, in case of explosion, the effect of the shock wave may cause pulling of the limbs, due to relative motion between them and trunk. (Figure no. 3).

Figure no. 3 Relative movement of the limbs and head to torso during an explosion illustration of limb pulling mechanism

The simplest model of protective equipment to comply with the above requirements would be represented by a totally rigid suit, but such equipment would eliminate almost all degrees of freedom, while the very definition of the fighter or intervention personnel presumes increased
mobility of the wearer. Similarly, we can mention the fact that, currently, most of the body armor for ballistic protection currently in use consists of heavy structures that limit the mobility and slow down deploying.

The situation described above may represent an application for some non-Newtonian fluids, which have the ability to instantly change their apparent viscosity in case high speed deformation load is applied, returning to their normal state after the dynamic load ceases. The following paragraphs describe two such fluids with future applications in “Liquid body” type armor systems [5].

3. Magnetorheological fluids

Magnetorheological fluids (MR) belong to the class of so-called fluids with controllable behavior (or controled fluids). A MR fluid is composed of dense micronic (range 0.1-10 μm) magnetic particles, held in suspension by a liquid medium (dispersion medium) of lower density, typically an oil (Figure no. 4).

Figure no. 4 MR fluid in idle state

When subjected to a magnetic field, the apparent viscosity of the fluid increases so much that it reaches a point where it behaves like a viscoelastic solid (Figure no. 5). Important to this behavior is that the yield stress of the fluid in the active state can be precisely controlled by varying the intensity of the applied magnetic field. It follows that it is possible to control the ability of the fluid of transmitting force by using an electromagnet.

Figure no. 5 MR fluid in active state

The magnetorheological fluid applications include: shock absorption systems to earthquakes (for buildings), suspension for vehicles, human prosthesis and, of course, liquid body armor.

4. Rheopectic fluids

There are bodies which, at constant shear rate, show a change in time of the shear stress, and, hence, their apparent viscosity. Moreover, the rheological behavior of some objects also depends on their “shear history”, that is the size and duration of previous loads that were applied to the body. Such bodies have a rheological behavior dependent upon time.

By loading a fluid at constant shear rate, the shear stress may remain constant or may change over time.

Figure no. 6 The variation in time of shear stress: 1 – rheopectic fluids; 2 – independent of time fluids; 3 – thixotropic fluids

The raise of stress in time indicates a rheopectic behavior and vice versa, the stress lowering in time indicates a thixotropic behavior.

The thixotropic behavior is manifested by a decrease in isothermal viscosity at progressively increasing shear rate as a result of fluid destructuring. The progressive decrease in shear rate determines the restructuring of the fluid. The opposite behavior to the phenomena described is represented by the rheopectic behaviour, also known as the anti-thixotropic behavior [6].

On increasing the shear rate the fluid structures itself and on lowering the shear rate it destructures. The rheopectic behavior was highlighted on: aqueous suspensions of clay, gypsum, bentonite earths, etc.

Currently, such a technology is tested and optimized in a “Liquid body armor” type structure by a group of researchers from Great Britain (BAE Systems).

AKNOWLEDGEMENT: This paper has been financially supported within the project entitled “Horizon 2020 - Doctoral and Postdoctoral Studies: Promoting the National Interest through Excellence, Competitiveness and Responsibility in the Field of Romanian Fundamental and Applied Scientific Research”, contract number POSDRU/159/1.5/S/140106. This project is co-financed by European Social Fund through Sectoral Operational Programme for Human Resources Development 2007-2013. Investing in people!

References

1 Simona Badea, Cercetări privind comportarea materialelor de protecţie la acţiunea undelor de şoc provocate de explozii în atmosferă, PhD thesis, (Bucharest: Military Technical Academy, 2011).

2 Florin Ilie, “Research on Protection Materials Behavior at Their Projectile and Armor Impact”, The XVIIIth International Conference The Knowledge-Based Organization, Land Forces Academy, 14-16 June, Sibiu (2012), 248-252.

3 National Research Council, Opportunities in Protection Materials Science and Technology for Future Army Applications, , (Washington D.C.: The National Academies Press, 2011).

4 Maria Nistor, “Vesta antiglonţ – Particularităţi privind tehnologia de confecţionare”, E-revista CCD Mureş, 8(2013).

5 Nasim Uddim, Blast protection of civil infrastructures and vehicules using composites, (Woodhead Publishing Limited, 2010).

6 Vékás, Ladislau, Nanofluide magnetice şi fluide magnetoreologice, Rezultate şi perspective în ştiinţa şi ingineria fluidelor, (Bucureşti: 2008)

Posts · by **Yohannes** » Sun Apr 22, 2018 5:26 pm

Impact Response of Shear Thickening Fluid (STF) Treated High Strength Polymer Composites – Effect of STF Intercalation Method

11th International Symposium on Plasticity and Impact Mechanics, Implast 2016
View online: Procedia Engineering 173 ( 2017 ) 655 – 662
View Table of Contents: http://www.elsevier.com/locate/procedia
Peer-review under responsibility of the organizing committee of Implast 2016

AUTHORS: Neelanchali Asija^a*, Hemant Chouhan^a, Shishay Amare Gebremeskel^a, and Naresh
Bhatnagar^a

a. Mechanical Engineering Department, Indian Institute of Technology Delhi, Hauz Khas, New Delhi-110016, India
* Corresponding author. Tel.: +91-011-26591715; fax: +91-011-26582053. E-mail address: neelanchali@gmail.com

Article history:
1877-7058 @ 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
Peer-review under responsibility of the organizing committee of Implast 2016
doi: 10.1016/j.proeng.2016.12.133

1. Introduction

Body armours from ancient times have been used to provide protection to humans against any threat or possible source of injury. These have been evolved from readily available materials such as animal skins or natural fibers made from cotton, thatch, silk woven in textile forms, to metallic shields of copper, steel or iron. Modern military operations and highly sophisticated weapons and ammunitions have necessitated the development of advanced ballistic protection body armour systems that are flexible, light-weight, comfortable-fit and possess high energy absorbing/ dissipating capacity [1]. Over the last three decades, the development in the technology of manufacturing high performance fibers and production processes have enabled the production of light weight armour with improved protection levels. One of the widely adopted ways to enhance the ballistic performance of high strength materials is by impregnating it with STF (Shear Thickening Fluid) [2–5]. STFs are a special class of field responsive non-Newtonian fluids that possess the ability to undergo transition from low to high viscosity under the imposed stress [6]. The last decade has witnessed the synergistic effect of STF and high performance fabrics, so as to produce a new flexible and light weight material possessing superior knife, stab and ballistic resistant properties than the existing material systems. In all the STF-treated ballistic materials an important parameter which dictates the ballistic performance of the hybrid structure is the method of STF intercalation. The various parameters which strongly influence the efficacy of STF intercalation are;

Affinity of ballistic fiber or fabric towards STF, so that it is efficiently wetted and impregnated.
The strength of the material which is used to hold the STF in a ballistic composite structure.
The optimal quantity of STF to be added to ballistic composite and the duration of STF impregnation.
The laying up sequence of ballistic composite to have appropriate number of STF treated ballistic fabrics sandwiched between untreated layers.
Efficient method of STF intercalation. This involves either placement of STF-filled high strength material pouches between ballistic fabrics or keeping STF treated ballistic fabrics between untreated layers.

This study involves determination of effective method of STF intercalation so as to enhance the ballistic resistance of the resulting composite. The STF treated composites were subjected to both low velocity impact testing on drop tower machine as well as high strain rate testing on in-house designed and fabricated SHPB (Split Hopkinson Pressure Bar) apparatus, to establish the efficacy of STF treatment.

2. Materials and Methods

2.1 Materials

(a) For Synthesis of STF – The materials used for STF synthesis include; dry Silica nano powder, Ethyl Alcohol and PPG (Poly Propylene Glycol). Rigid spherical silica particles of size 100nm were procured from Nippon Shokubai, Japan. PPG having an average molecular weight of 400g/mole, was chosen as the solvent for its high thermal stability, non-toxicity and easy availability. Ethyl alcohol was used as the solvent in ultrasonication process. PPG was obtained from Qualikems Fine Chemicals, whereas, ethanol was procured from Sigma-Aldrich. All the chemicals were used as received, without any further purification

(b) For Synthesis of Extrusion Foamed Polymer Sheet – CO15EG Impact Copolymer procured from Reliance Polymers was used for extrusion of porous sheet. It is a high impact Polypropylene (extrusion grade) with MFI (Melt Flow Index) of 1.8. CO15EG copolymer was compounded with Maleic Anhydride modified homopolymer Polypropylene OPTIM®P-425 which was procured from PLUSS Polymers, in order to enhance the wettability of CO15EG polymer with STF. The MFI of OPTIM®P-425 is 110g/10 min as per ASTM D1238.

(c) For Manufacturing of UHMWPE Composite Panels – For manufacturing of composite panels, two variants of UHMWPE (Ultra High Molecular Weight Poly Ethylene)i.e. Gold Shield and Spectra Shield were procured from Honeywell, USA, in the form of 0/90 prepregs. Both the materials are patented products of Honeywell. Spectra Shield composite panels were subjected to low velocity testing on drop tower machine whereas Gold Shield panels were subjected to high strain rate testing on SHPB. This was done since Spectra Shield is a laser-transparent material. Consequently, SHPB test specimens could not be cut from Spectra Shield composite panel, whereas laser cutting of Gold Shield panels was easily accomplished owing to its laser absorption characteristic.

2.2. Methods of Sample Preparation

(a) Shear Thickening Fluid (STF) - STF was synthesized by ultrasonic homogenization method. The known weight percentages of PPG and silica nano powder were mixed in excess amount of ethanol, and subsequently dispersed with high intensity ultrasonic horn (Ti-horn) at 20 kHz, 1200 W/cm² at amplitude of 42% for 3 hours. The weight percentage of silica was kept fixed at 67.5% wt. After the ultrasonication process, the fluid sample was placed in a preheated oven at 80^OC for 2 hours to evaporate the excess ethanol.

(b) Extrusion Foaming of Polymeric Films – CO15EG was compounded with OPTIM®P-425 in a twin screw extruder. The weight percentage of Maleic Anhydride Grafted PP (MAgPP) was kept fixed at 15 wt.% in CO15EG. All the compounded pellets were dried at 60^OC vernight prior to extrusion.

Fig. 1: Pictures of Extrusion Foaming Process

(c) Manufacturing of Spectra Shield^® and Gold Shield Composite Panels^® - Both UHMWPE Spectra Shield^® and Gold Shield composite panels were made in Santec 200 Ton compression moulding machine. The compression moulding parameters used for manufacturing both the composite panels are given in Table 1.

Table 1: List of compression moulding parameters for Gold Shield^® and Spectra Shield^® Panels

3. Experimental Results and Discussion

3.1. Low Velocity Drop Tower Testing

(a) Synthesis of STF-treated composite panels

The composite specimens for low velocity testing were synthesized in the following manner; 10 sheets of 150x150mm extrusion foamed MAgPP polymer films were taken and 10mL of STF was added at the center of it. Then it was kept in an oven at 80^OC for 4 hours to evaporate the excess of ethanol. Thereafter, these films were placed between two compression moulded Spectra Shield panels and packed by extruded film of LLDPE on all sides. Then this sandwich construction was compression moulded at 108^OC and 8 Bar pressure, with curing time of 5 minutes. Finally the composite panels were ready for drop tower testing.

(b) Drop Tower Testing

The low velocity drop tower testing was conducted on Instron Dynatup 9250 HV machine. Four sets of tests were done by varying the impact velocity in steps of 2, 3, 3.5 and 4 m/s. The images of the drop tower testing are shown in Fig. 2.

Fig. 2: Pictures of drop tower testing of composite panels

The tabulated results obtained from drop tower testing are given in Table 2. In this Table, ‘T’ signifies STF-treated and ‘U’ signifies untreated specimens.

Table 2: Tabulated results from the low velocity impact testing of composite panels

From the Table it is evident that energy at peak load is higher for STF-treated composite panels as compared to untreated panels. It is also observed that deflection at peak load is higher and the energy absorbed is lower for STF-treated panels as compared to untreated panels. This shows that STF treatment does not enhance the penetration resistance of composite panels. The higher energy at peak load for STF treated composite panels implies that the rise in energy with respect to time is rapid after STF addition than in the absence of STF. Therefore, it can be inferred that STF addition to panel induces quicker response in terms of attainment of peak load against impact loading. However, the peak load attained is lower for STF treated panels. This can be explained by the fact that by varying the impact velocity from 2-4 m/s, the strain rate is varied in the range of 4.9-9.7 s^-1. The experimental strain rate is calculated by dividing the impact velocity with the drop height of the hemispherical impactor.

Fig. 3: Flow curve of shear thickening fluid

From Fig. 3 it is evident that the critical shear rate of the STF under study is 15.3 s^-1, whereas in drop tower testing the strain rate encountered is less than 10 s-1. As the experimental strain rate is less than the critical shear rate required for the onset of shear thickening phenomenon, hence, STF addition did not had any synergistic effect on the improvement of impact resistance or enhancing the energy absorption capability of the composite panel.

3.2. High Strain Rate Testing on SHPB

The SHPB technique comprises of three bars; Incident Bar (IB), Transmission Bar (TB) and the Striker Bar (SB). During testing, the striker bar is propelled by compressed nitrogen gas, through a 2m long barrel, before striking the incident bar. A pulse shaper was placed at the impacting surface of the IB, prior to each experiment The advantage of using a pulse shaper is that it deforms slower than the bar material, thereby resulting in a near trapezoidal loading pulse (εi). For each experiment, a fresh pulse shaper was used. When the striker bar hits the IB at the specimen-IB interface, a fraction of the loading pulse is reflected (εr) back through the IB and the remaining is transmitted (εt) through the TB. The fraction of the loading pulse reflected and transmitted depends upon the acoustic impedance ‘z’ (z = ρc) mismatch between the specimen and the bar materials. The magnitude of these pulses (εi, εr and εt) are measured with the help of strain gages that are mounted in Quarter Bridge Type-II configuration, in which one active strain gage was attached at the centre of each bar. The analytical relations used for determination of specimen strain, strain rate and stress as a function of time are as follows:

Table 3: SHPB Experimental Parameters

3.2.1. Preparation of SHPB specimens

(a) Laser Cutting of Gold Shield^® panels for making SHPB specimens

Fiber laser of 400 W continuous power (Model: RS400, SPI Lasers, UK) was used to cut Gold Shield panels. The dimensions of the laser-cut Gold Shield^® specimens were; thickness 1.35 mm and diameter 11.35 mm. From the weight and volume of the Gold Shield^® specimens the density was calculated to be 1.23 g/cc.

(b) Treatment with STF

Three methods were adopted for treating Gold Shield^® specimens with STF. These are;

Method I – Using extrusion foamed MAgPP polymer sheets – In this method, 10 square pieces of 15 wt.% MAgPP extrusion foamed polymer sheets having each side 8.8 mm were cut. The foamed square pieces were then dipped in STF for 2 hours. Then they were stacked over one another and kept between two laser-cut Gold Shield^® specimens. Slight force was applied on the resulting composite structure with a tissue paper, to squeeze out excess of STF, and thereafter the pieces were kept in a preheated oven at 80^OC for 2 hours to evaporate the excess of ethanol. Finally, the specimens were tested on SHPB apparatus.

Method II – Keeping liquid STF specimen between two Gold Shield^® specimens - In this method, the liquid STF was directly applied to one of the faces of Gold Shield^® specimen with the help of a spatula while the other end was secured to the face of the incident bar of SHPB by applying high speed grease. The liquid STF was delicately balanced between two Gold Shield^® specimens. The other face of the second Gold Shield^® specimen was kept in contact with the transmitter bar face of SHPB.

Method III – By dipping the Gold Shield^® specimens in liquid STF - In this method, the laser-cut Gold Shield^® specimens were dipped in liquid STF for 4 hours. Prior to dipping, STF dispersion was diluted with ethyl alcohol. Thereafter the specimens were kept in preheated vacuum oven for 2 hours at 80^OC to evaporate the excess of ethanol. Finally, the STF impregnated Gold Shield^® specimens were subjected to high strain rate testing on SHPB apparatus.

3.2.2. Experimental Results and Discussion

I. Testing of foamed polymer-Gold Shield^® STF composites

In this, the specimens prepared by Method I were subjected to SHPB testing. For comparison, neat pieces of two Gold Shield^® laser cut specimens were also tested on SHPB at the same triggering pressure of 0.5 bar, without keeping the STF treated foamed polymer sheets between them. Fig. 4 shows the plot of Incident, Reflected and Transmitted voltage pulses (Vi, Vr and Vt respectively) with respect to time for both the SHPB tests.

Fig. 4: Plot of Vi, Vr and Vt wrt time for (a) STF treated foamed polymer-Gold Shield^® composite (b) two neat Gold Shield^® specimens kept together between IB and TB

From Fig. 4 it is observed that the transmitted pulse is very weak in case of composite specimen whereas strong transmission signal is obtained in neat Gold Shield^® specimens. Also, the compressional strain rate attained in the composite specimen is only 5.05 s^-1 whereas in neat Gold Shield^® specimens strain rate of 7040 s^-1 is attained. The experimental observations emphasize on the fact that the foamed specimen is ruptured even before the stress pulse is transmitted through the specimen to the transmission bar, thereby leaving no time for STF to give response against impact loading. This results in almost negligible transmission voltage pulse, thereby resulting in very low specimen stress. The strain rate in the specimen is also very low (less than 10 s^-1), owing to weak and fragile foamed specimen as compared to Gold Shield^® specimen. The reflected pulse observed in Fig. 5 is solely due to reflection from the first Gold Shield® specimen at the incident bar-specimen interface.

Fig. 5: Plot of Vi, Vr and Vt wrt time for (a) STF between Gold Shield^® specimens (b) two neat Gold Shield^® specimens kept together between IB and TB without STF

II. Testing of Gold Shield^® - Fluid STF specimens on SHPB

In this test the weak foamed polymeric sheet was completely discarded and liquid STF was applied directly on the in-contact faces of the two Gold Shield^® specimens with the help of spatula. After the STF application, both the faces of the Gold Shield specimens carrying STF were gently brought in contact, without inducing any compressive force on the applied STF layer, and immediately striker bar was fired. Fig. 5 shows the plot of Incident, Reflected and Transmitted voltage pulses (Vi, Vr and Vt respectively) with respect to time for both the SHPB tests.

From Fig. 5 it is evident that the level of incident voltage is almost similar for both the tests since both the tests were performed at same N₂ gas triggering pressure of 0.5 bar. However, a strong transmission signal is observed in the second case when STF is absent between two neat Gold Shield^® specimens, in comparison to the presence of STF in the first case. This is attributed to the fact that when the stress wave enters from one medium to another, the stress wave propagation is hindered by the acoustic impedance of the second material. If the stress wave encounters same material, a strong transmission is obtained. In the first case, there are four types of interfaces i.e. Titanium IB (incident bar) and first Gold Shield^® specimen, Gold Shield^® specimen and STF, STF-second Gold Shield^® specimen and finally second Gold Shield^® specimen and Titanium TB (transmission bar), and there is stress wave attenuation at each interface, resulting in weak transmitted signal. In the second case, stress wave encounters only two different interfaces i.e. entrance of stress wave at the Titanium IB and Gold Shield^® specimen interface and exit of the stress wave at the second Gold Shield^® specimen-Titanium TB interface. Consequently, the stress wave attenuation is smaller in this case, resulting in strong transmission signal. Moreover, neat Gold Shield^® specimen is stronger than neat STF fluid specimen in terms of producing higher specimen stress during the impact loading event.

III. Testing of STF- Impregnated Gold Shield^® specimens on SHPB

In this test, the specimens prepared from Method III were subjected to SHPB testing. The SHPB testing of both the impregnated as well as unimpregnated neat specimens was carried out to determine the effect of STF impregnation on the high strain rate behaviour of Gold Shield^® specimens. Fig. 6 shows the V_I(t), V_R(t) and V_T(t) plot for firing at 1.5 bar of N₂ gas pressures.

Fig. 6: Plot of Vi, Vr and Vt wrt time for (a) Untreated (b) STF treated Gold Shield^® specimen

From Fig. 6 it is evident that the STF-treated Gold Shield^® specimen exhibit higher reflection voltage, resulting in higher compressional strain rates as compared to untreated specimens. To further study the response of STF impregnated Gold Shield specimens under impact loading, the N₂ gas pressure was varied form 0.5 bar to 3.5 bar in steps of 1 bar, to vary strain rates in the range of 3000 to 16000 s^-1. The peak stress and peak strain values attained in each test are given in Table 4.

Table 4: Peak stress, strain and strain rate values obtained from SHPB testing of Gold Shield^® specimens

From Table 4 it is observed that the peak specimen stress increases progressively with the loading rate of the specimen. Loading rate implies to the maximum specimen strain rate attained in the loading part of the stress cycle. The impact toughness was computed for both the STF treated and untreated specimens, by calculating the area under the stress-strain curve for the loading part of the stress cycle, i.e. till the strain continues to increase with the specimen stress. Fig. 7 shows the variation of impact toughness with peak specimen stress and peak strain values.

Fig. 7: Variation of the impact toughness with specimen loading rate for STF-treated and untreated Gold Shield^® specimens

From the Fig. 7 it is evident that the impact toughness continues to increase progressively with the peak stress and peak strain values as well as the loading rate. It is also observed that the escalation in impact toughness is more sharp and steep for STF treated Gold Shield^® specimens as compared to untreated samples at higher loading rates. Therefore, all the experimental results manifest that the STF treated Gold Shield^® specimens exhibit higher impact resistance as compared to untreated specimens under dynamic impact loading.

4. Conclusions

From the experimental results it can be inferred that in order to enhance the impact resistance of a material by
treatment with STF, the following conditions must be met;

^®

Acknowledgements

The authors are indebted to IRD-IITD for a grand challenge grant (MI00810) for doing this research project. The authors also wish to express their sincere gratitude towards the staff of Production Engineering Laboratory, IIT Delhi, for extending their full support in the fabrication of SHPB.

References

[1] N. V. David, X.-L. Gao, J.Q. Zheng, Ballistic Resistant Body Armor: Contemporary and Prospective Materials and Related Protection Mechanisms, Applied Mechanics Reviews. 62 (2009) 050802.

[2] V.B.C. Tan, T.E. Tay, W.K. Teo, Strengthening fabric armour with silica colloidal suspensions, 42 (2005) 1561–1576.

[3] A. Srivastava, A. Majumdar, B.S. Butola, Improving the impact resistance performance of Kevlar fabrics using silica based shear thickening fluid, Materials Science & Engineering A. 529 (2011) 224–229.

[4] M. Hasanzadeh, V. Mottaghitalab, The Role of Shear-Thickening Fluids (STFs) in Ballistic and Stab-Resistance Improvement of Flexible Armor, Journal of Materials Engineering and Performance. 23 (2014) 1–15.

[5] M.J. Decker, C.J. Halbach, C.H. Nam, N.J. Wagner, E.D. Wetzel, Stab resistance of shear thickening fluid (STF)-treated fabrics, Composites Science and Technology. 67 (2007) 565–578.

[6] P.K. and N.K. Hamid Reza Baharvandi, Effect of silica weight fraction on rheological and quasi-static puncture characteristics of shear thickening fluid-treated, J. Industrial Textiles. (2015) 1–22.

Posts · by **Yohannes** » Mon Apr 23, 2018 8:02 pm

[ Out of character: I have access to three or so accurate research about "An end-to-end model of an electrothermal chemical gun". I know that the co-opening poster of this thread Lamoni (and I myself) have promoted ETC stuff in the past, and so I will not be posting those three resources here (in the hope of assuring peopel that I want this thread to be a neutral resources thread). However, if anyone would like to see the resources feel free to telegram me and I'll see what I can do. Next article will be about sound pressure in tank gun muzzle silencer instead. Thank you!

Posts · by **Yohannes** » Mon Apr 23, 2018 8:53 pm

CFD analysis of sound pressure in tank gun muzzle silencer

J. Cent. South Univ. Technol. (2011) 18: 2015−2020
DOI: 10.1007/s11771−011−0936−7

AUTHORS: Hafizur Rehman¹, Hanshik Chung¹, Taewhee Joung², A. Suwono³, Hyomin Jeong¹

1. Department of Mechanical and Precision Engineering, Gyeongsang National University, The Institute of Marine Industry, 445 Inpyeong Dong, Tongyeong 650-160, Gyeongsang Nam do, South Korea;
2. Eco-Friendly Heat & Cold Energy Mechanical Research Team, Gyeongsang National University, 445 Inpyeong Dong, Tongyeong 650-160, Gyeongsang Nam do, South Korea;
3. Faculty of Mechanical and Aerospace Engineering, Bandung Institute of Technology, Indonesia

@ Central South University Press and Springer-Verlag Berlin Heidelberg 2011

Article history:
Foundation item: Project(NRF-2010-013-D00007) supported by the National Research Foundation of Korea; Project supported by 2010 Year Research Professor Fund of Gyeongsang National University, Korea and NIIED Korea
Received date: 2010−02−09; Accepted date: 2011−04−15
Corresponding author: Hyomin Jeong; Tel: +82−55−772−9114, Fax: +82−55−772−9119; E-mail: hmjoeong@gnu.ac.kr

Abstract: The high pressure waves generated due to muzzle blast flow of tank gun during firing is a critical issue to examine. The impulsive noise from the gun has various negative effects such as damage of human bodies, damage of structures, creating an environmental, social problem and also military problems such as exposure of location of troops. This high pressure impulsive sound, generated during the blast flow, was studied and attenuated. An axisymmetric computational domain was constructed by employing Spalart Allmaras turbulence model. Approximately 90% of pressure and 20 dB of sound level are reduced due to the use of the three baffle silencer at the muzzle end of the gun barrel, in comparison with the tank gun without silencer. Also, the sound pressure level at different points in the ambient region shows the same attenuation in results. This study will be helpful to understand the blast wave characteristics and also to get a good idea to design silencer for large caliber weapon system.

1 Introduction

Due to firing of tank guns, a high intensity sound pressure is created in form of muzzle blast wave. In fact, this muzzle blast is produced due to the explosion of the propellant inside the gun barrel. The deflagration of the propellant in the chamber produces an abrupt expansion of gases. This rapid increase in volume causes pressure waves which accelerate the projectile into flight from the muzzle end of the barrel and as result of this high intensity muzzle blast, impulsive sound is heard. Compared with other sound, the impulsive sound has several special features and different properties, such as low frequency, strong directivity and long range propagation [1−2]. That’s why it can easily reach
surrounding areas and communities. The impulsive noise from the gun has various negative effects such as damage to human bodies, damage of structures, creating an environmental, social problem and also military problems such as exposure of location of troops.

The parameters which cause this high intensity noise are muzzle blast, sabot discard, projectile flight and explosion of the projectile at the target etc. There are two main sources of impulsive noise from the firing, i.e. gun blast noise and projectile bow shock noise [3]. The gun blast is highly directional, therefore sound effect at the locations directly in front of the gun is about 15 dB higher than that for equidistance locations directly to the rear of the gun. The projectile bow shock noise only occurs forward the gun, in a region determined by the supersonic velocity of the projectile. This noise is localized nearer to the gun if the slug is unstable in flight and thus decelerates quickly to subsonic speeds [3−4].

The noise from the tank guns and cannons is directed predominantly to the front and is attenuated much less over water than over vegetated land. According to some experimental investigation, the noise levels due to high pressure blast flow could be heard about 10 miles away from the firing point at a level of 90 dB [1, 3].

Thus, in the view of all above facts, the study of blast wave and impulsive sound attenuation is of great importance.

Silencers or mufflers are used to reduce this muzzle blast flow noise. Silencers have to be designed especially, so that they allow gun gases to expand into chamber volumes properly to get the maximum pressure reduction. The attenuation of silencer generally increases with its internal volume. Attenuation increases with the number of baffles but only up to a certain value and then decreases thereafter. The attenuation also depends on the length of the inlet chamber, the placement of the silencer, and the whole size of projectile. The suppression of the muzzle blast is important in both large caliber weapon system and small caliber weapon system designs. In case of large caliber weapon system, the design of silencer has relied heavily on experimental work and the development of empirical databases [1, 5].

The study on impulsive noise is divided into two categories: noise attenuation and blast wave analysis. In the present study, the impulsive sound attenuation, by using a three baffle silencer during high pressure blast flow, has been analyzed. For this study, large caliber 120 mm K1A1 tank gun has been selected especially.

For evaluation, the designing work, simulation and results, Gambit, CFD and Fluent software has been used. A three baffle silencer has been fitted at the muzzle end of the barrel. The simulated results of pressure and sound pressure level at different points inside the silencer and also at different points in the ambient region have been compiled and compared with the results at the same points without using silencer.

2 Governing equation

For a three-dimensional, turbulent, unsteady and compressible flow, the governing equation is expressed as

where Q, E, F and G denote the dependent variables and flux vectors.

The Spalart Allmaras equation for aerodynamic flow is expressed as

where

And c_t1 = 1 and c_t2 = 2

The far field boundary condition is

And the turbulent eddy viscosity is computed from

where

Additional definitions are given by the following equations:

Also, shear stress can be expressed by laminar viscous coefficient and eddy viscous coefficient as

As per ideal gas equation, the pressure p is given by

3 Numerical analysis and simulations

In order to do the simulation for this case by using appropriate numerical solver, a validation case study with sufficient quantitative and qualitative information about the very complex flow-field created by muzzle blast is necessary to properly validate the computational fluid dynamics (CFD) techniques. For this, a CFD analysis of 7.62 mm NATO rifle G3 with a DM-41 round was selected about the flow-field in the form of shadowgraph images and analysis. This problem was tested in Refs.[6−8].

Figure 1 shows the reference shadow graphs and the validated fluent shadow graphs. Also Fig.2 shows the pressure graph at initial condition of the reference for 7.62 mm NATO rifle G-3.

Fig.1 Reference shadow graphs (a) and fluent validated shadow graphs (b) at different time intervals: (a₁) and (b₁) t_exp≈3.5×10⁻³ ms; (a₂) and (b₂) t_exp≈3.7×10⁻³ ms

Fig.2 Reference pressure graph at initial condition for 7.62 mm NATO rifle G3

After getting validation of the above shadow graphs, computational fluid dynamics analysis has been applied to analyzing the supersonic blast flow and for design work Gambit software has been used. The basic domain has been made from the specifications and data of 120 mm caliber K1A1 tank gun barrel, as listed in Table 1. The designing work has been made by using Gambit and analyzed by using fluent.

Table 1 Specifications of 120 mm K1A1 tank gun

A density based axisymmetric, unsteady state condition with ideal gas as fluid has been used. First order implicit scheme is used for time integration and also the Spalart-Allmaras S.A (1-eqn) turbulence model is used. Additionally, the multi-block grid technique has been applied to analyzing the complicated geometry of the gun muzzle.

3.1 CFD modeling and simulation for high pressure blast flow field without silencer

To investigate the high pressure supersonic blast flow field with a silencer, first the same case without installing silencer at the muzzle end of the gun is analyzed. After that, the achieved data are compared with the results achieved with silencer case.

Figures 3 (a) and (b) show the schematic diagram of the computational domain, initial condition, and boundary condition of 120 mm K1A1 gun without silencer respectively and Fig.4 shows the maximum pressure graph at inlet condition.

Fig.3 Schematic diagram (a) and boundary condition and mesh diagram (b) without silencer for 120 mm tank gun

Fig.4 Pressure−time plot at inlet condition for 120 mm K1A1 tank gun

In order to get the impulsive sound pressure level in the open field area, different points have been taken at a radial distance of R2 and R4 from the muzzle end. These points have been taken at an angle of 0^O, 15^O, 30^O, 45^O, 60^O and 90^O, as shown in Fig.3(a).

3.2 CFD modeling and simulation for high pressure blast flow field with silencer

The schematic diagrams of computational domain, boundary condition, mesh and CFD diagram for the 120 mm K1A1 tank gun after installing three baffle silencer are shown in Figs.5(a), (b) and (c), respectively. To get the comparison result of sound pressure level in the ambient region after installing these silencers, all the points have been taken at same distances and angles as above case. Detail is shown in Fig.5(a).

Fig.5 Schematic and CFD animation diagram (a), boundary condition and mesh diagram (b) and CFD diagram (c) with three baffle silencers for 120 mm tank gun

4 Results and discussion

The pressure graph at five different points inside the silencer is shown in Fig.6(a) and CFD diagram of blast wave formation inside the barrel at the acquisition points is shown in Fig.6(b). Furthermore, Table 2 describes the comparison results at these points.

Fig.6 Pressure graph at five different points inside silencer (a) and blast wave formation diagram (b) inside gun barrel points

Table 2 Pressure and sound pressure level at five points inside silencer

From Fig.6(a) and Table 2, it is clear that due to influences of the propellant shock wave, pressure variation and sound pressure level at point a are recorded as 300 MPa and 263 dB, respectively. After first baffle at point b, the pressure gets to 200 MPa and the sound level is 260 dB. After second baffle at point c, the pressure gets to 170 MPa and the sound level is 258 dB. After third baffle at point d, the pressure gets to 105 MPa and the sound level is 254 dB. And at the exit point e of the silencer, the pressure and sound level are recorded as 50 MPa and 244 dB, respectively.

In view of the above results, it is concluded that, approximately 90% of pressure and 20 dB of sound pressure level have been attenuated due to use of the three baffle silencers.

Figure 7 shows the pressure graphs at different points with and without silencer, which have taken at different angles at a radial distance of R2 in the ambient region. Table 3 illustrates the comparison results of pressure and sound level at these points. Detail of the points has been already shown in Fig.5(a).

Fig.7 Pressures at different points taken at radial distance of R2 in ambient region: (a) 0^O; (b) 15^O; (c) 45^O; (d) 60^O; (e) 90^O

Table 3 Pressure and sound pressure level at different points at R2 in ambient region

Similarly, Table 4 gives the comparison result of pressure and sound pressure level at the different points with and without silencer, at same angles but at a radial distance of R4 in the ambient region.

Table 4 Pressure and sound pressure level at different points at R4 in ambient region

From Table 3, it is clear that the reduction in pressure and sound level at R2 and angle 0^O is 82% and 14 dB, whereas at 90^O the reduction is 92% and 22 dB. Similarly, from Table 4, the reduction in pressure and sound level, recorded at point R4 and angle 0^O is 85% and 16 dB, whereas at 90^O, the reduction is 90% and 20 dB, respectively.

5 Conclusions

The CFD analysis of impulsive sound pressure and also attenuation of sound pressure, generated by large caliber gun during blast flow is conducted. Due to use of three baffle silencers at the muzzle end of gun barrel, a considerable amount of propellant energy has dissipated inside the chamber volume of the silencer and also the sound pressure level is diminished. Approximately 90% of pressure and 20 dB of sound level are reduced, in comparison with the gun without silencer. The results of this study will be helpful to understanding the blast wave characteristics as well as in design of silencers for large caliber weapon system.

References

[1] KANG K J, KO S H, LEE D S. A study on impulsive sound attenuation for a high pressure blast flowfield [J]. Journal of Mechanical Science and Technology, 2007, 22(1): 190−200.

[2] COOKE C H, FANSLER K S. Numerical simulation and modeling of a muffler [R]. Memorandum Report BLR-MR-3735, 1989.

[3] FANSLER K S, von WAHLDE R. A muffler design for tank cannon acceptance testing [R]. Pentagon Report A360142, 1991.

[4] PATER L L, GRUBB T G, DELANEY D K. Recommendation for improved assessment of noise impacts on wildlife [J]. The Journal of Wildlife Management, 2009, 75(5): 788−795.

[5] FANSLER K S, COOK C H, THOMPSON W G, LYON D H. Numerical simulation of a multi compartmented gun muffler and comparison with experiment [R]. Technical Report BRL-TR-3145, 1990.

[6] CLER D L, CHEVAUGEN N, SHEPHERTY M S, FLAHERTY J E, REMACLE J. Computational fluid dynamics application to gun muzzle blast−A validation case study [R]. Technical Report ARCCBTR-03011, 2003.

[7] CLER D L. Techniques for analysis and validation of unsteady blast wave propagation. 2003

[8] FANSLER K S, LYON D H. Attenuation of muzzle blast using configurable mufflers [R]. Memorandum Report BLR-MR-3931, 1989.

(Edited by YANG Bing)

Posts · by **Yohannes** » Tue Apr 24, 2018 12:42 am

Nitramine-Based High Energy Propellant Compositions for Tank Guns

Defence Science Journal, Vol 50, No January 2000, pp.75-81
@2000, DESIDOC
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

AUTHORS: R.S. Damse^a and Haridwar Singh^b

a. Mr RS Damse obtained his MSc (Organic Chemistry) from University of Poona, in 1983. He joined DRDO at the High Energy Materials Research Laboratory (HEMRL) as Scientist B. His areas of work include: development of high explosives for filling of various warheads, design and development of combustible cartridge cases for tank guns of different calibre and ignition system for tank gun ammunition.

b. Dr Hardiwar Singh obtained his PhD in solid rocket propellant combustion from University of Poona. He joined DRDO at the HEMRL, Pune, in 1971 and became Director in 1990. He is a recognised postgraduate guide of University of Poona and Shivaji University and has supervised 20 post-graduate and doctoral theses. He was a Visiting Scientist at the Max-Planck Institute, Germany; High Pressure Combustion Laboratory, Pennsylvania State University, Army Research Laboratory; Maryland Science Application Centre, Santa Clara and Sandia National Laboratory, USA. He has been awarded the Astronautica[ Society of India Award for 1994. He was conferred with the DRDO Scientist of the Year Award in 1983 for his contributions in applied sciences and advanced solid propellants, high energy propellants, respectively. He is the recipient of DRDO Cash Award for the development of power plants for missile target and Best Paper Award of Defence Science Journa[ in 1984. He is the Chairman of the High Energy Materials Society of India, a member of AIM, International Advisory Board-Flame Structure & Combustion, and an honorary member of the Russian Academy of Astronautics. His areas of research include: rocketry, cast double-base (CDB) propellants, HTPB-based composite propellants, nitramine-based propellants, high energy composite modified double base (CMDB) for integrated rocket ramjet, AP-based p(opeIIants, and zirconium powder based fuel-rich propellants.

Article history:
High Energy Materials Research Laboratory, Pune -4 021
Revised 17 June 1999

NTRODUCTION

Conventional gun propellants have reached saturation level in terms of energy. To meet the requirements of tank gun ammunition, propellants of higher force constant at a relatively lower flame temperature (T_f) are required to minimise gun barrel erosion. Linear rate of burning coefficient

and pressure exponent

are important parameters in determining the combustion behaviour and suitability of a propellant for tank gun ammunition. A higher value of

necessitates increase in web size of the propellant grain. Increase in web size, in turn, poses problems in manufacture, loadability and brittle fracture of the grain, particularly at lower temperature. Similarly, a higher magnitude of a leads to exponential rise in burning rate and pressure, which affect the safety of the gun.

Extensive research is being carried out allover the world to improve upon the force constant by increasing the number of moles of combustion gases per unit mass rather than increasing

. The main constituents of the propellant combustion gases are carbon monoxide, carbon dioxide, hydrogen, water and nitrogen. To formulate a higher energy gun propellant, the ingredients should have higher percentages of hydrogen, carbon monoxide and nitrogen rather than water and carbon dioxide in their combustion product gases. Literature survey reveals that a number of new series of cool burning, high impetus and low molecular weight gun propellants have.been studied^1-4. Most propellants contain either cyclic or linear nitramines, such as cyclotrimethylenetrinitramine (RDX), cyclotetramethylene-tetranitramine (HMX), triaminoguanidine nitrate (TAGN), nitroaminoguanidine (NAGU) and triaminoguanidineethylene-dinitramine (TAGED) as energetic ingredients. The results of a systematic study on RDX-based propellant compositions aimed at obtaining higher energy for tank gun applications are presented in this paper.

2 EXPERIMENTAL WORK

Six different compositions based on nitrocellulose (NC) of 13.1 per cent nitrogen content, di-octylphthalate (DOP), carbamite and RDX of average particle size ( 5 μm) were formulated. Theoretical performance of the composition was computed using THERM program and the results are presented⁵ in Table 1. Propellant compositions were made on a laboratory scale (1 kg batch) by solven process⁶. First, fine RDX was dehydrated with ethyl alcohol and coated with required quantity of DOP on dry weight basis. Exact percentage of DOP and uniformity of its coating to RDX was confirmed gravimetrically using n-pentane as the solvent for the extraction of DOP. All the samples were subjected to impact and friction sensitivity tests to obtain safety-related information. Propellant compositions were prepared using a 30 per cent solution of acetone-alcohol (70:30) mixture. The ingredients were kneaded in an incorporator for 6 hr to obtain a homogeneous propellant dough. Five per cent extra solvent was required to have a good dough during the preparation of compositions containing 75-80 per cent RDX. The dough was subsequently extruded in cord form at around 50 bar using a hydraulic press. Extruded cord strands were cut to 12 cm and dried in an oven at 45-50^OC till the volatile matter got reduced to 1 per cent. Out of the six compositions, composition III, containing 65 per cent RDX and 28 per cent NC, was made in a multitubular configuration also to study its burning rate characteristics in that configuration. Dried propellant samples were tested for physical characteristics like web size and density, and finally fired in a 700 cc closed vessel (CV) at 0.20 g/cc loading density for the determination of ballastic performance. Results of CV tests are shown in Table 2.

Table 1: Chemical formulations and theoretical performance of propellant compositions

Table 2: Data on closed vessel firing of the propellant compositions

3. EVALUATION OF BALLISTIC PERFORMANCE

3.1 Measurement of Sensitivity

Impact sensitivity was measured by fall hammer method using 2 kg drop weight and 20 mg sample. The height mentioned in Table 3 refers to 50 percent probability of explosion of the compositions. Friction sensitivity was measured using Julius Peter apparatus and 10 mg sample. The results for impact and friction sensitivity are given in Table 3.

Table 3: Data on sensitivity of the propellant compositions

3.2 Thermal Characteristics

Deflagration temperature was obtained on 5 mg sample by gradually raising the temperature at the rate of 5^OC/min in Julius Peters furnac.e. The temperature at which the sample got ignited was recorded. Decomposition temperature was recorded using differential thermal analyser (DTA). DTA curves were recorded in an inert atmosphere using 10 mg samples in alumina crucibles at a heating rate of 10^OC/min. Calorimetric values of the compositions were determined using Julius Petets adiabatic bomb calorimeter of 300 cc at 1 atm. 1 g sample was ignited and the total heat output was measured (Table 4).

Table 4: Data on thermal characteristic tests of the propellant compositions

3.3 Thermal Stability

Thermal Stabilities of the compositions were determined by Bergmann and Junk test. 5g samples were heated at 120^OC for 5 hr and the total gaseous volume of nitrogen oxides was measured titrimetrically.

3.4 Mechanical Properties

For determining tensile strength and percentage elongation, ‘mini-samples’ (Fig. 1) were punched out of the propellant strips and dried up to ~1 percent V.M. level. Tensile strength and percentage elongation were determined using the Instron universal materials testing machine (model-1185). Flexural properties were determined using propellant strips of particular dimensions (100 mm x 10 mm x 3 mm) and Instron machine. For determining percentage compression of the composition containing 65 per cent RDX and 28 per cent NC, multitubular grains having L/D =1 (Fig. 1) were made. The Instron machine was used for this purpose.

Figure 1. Specimen samples for testing the mechanical properties.

4. RESULTS & DISCUSSION

The results of ballistic evaluation of the propellant compositions using a cv are given in Table 2. Compositions I and II exhibited comparatively lower values of force constant. However, composition III was found to be most suitable because of its higher force constant at a relatively lower

.

value was found to be within acceptable limit, but

value exceeded unity. Therefore, the same propellant composition was further studied in a multitubular configuration to assess its suitability with

and

. Results of cv tests on the subject composition indicate 14 per cent and 23 per cent decrease in the value of

and

, respectively when the cord configuration was changed to multitubular format. This difference appears to be due to approximation in the form function for multitubular shape and also ignition and burning characteristics of holes of the multitubular propellant⁷. Approximation in the form function is achieved due to anomalous combustion mechanism ofRDX propellant, as indicated by the slope break phenomenon. Slope breaks are related to the change in the mechanism of decomposition, probably due to changes in depth of melt-layer of the deflagrating propellant surface. The form function is thus suitably approximated in the multitubular rather than the cord configuration, which results in a low pressure exponent. Therefore, suitability of the subject composition for tank gun applications was better assessed when it was manufactured in a multitubular configuration. Secondly, p_vs dp/dt profile of the multitubular propellant (Fig. 2) was found to be of non-peaky nature as compared to the cord configuration. Forcompositions containing 70, 75 and 80 per cent RDX, even though the force constant increased successively, values of

,

, and

also increased drastically for beyond the desired levels for tank gun applications. Hence, these compositions have not been selected for tank gun applications. The same is confirmed by cv firing, wherein a sudden rise in p_vs dp/dt profile was noticed in the case of composition containing 80 per cent RDX and 16 per cent NC (Fig.3).

Figure 2. P vs dp/dt profile of composition III (multitubular configuration).

Figure 3. P vs dp/dt profiles of the propellant compositions (cord configuration)

The measured sensitivity values given in Table 3 indicate increasing trend of sensitivity from the compositions containing 55 per cent RDX and 36 per cent NC - 80 per cent RDX and 16 per cent NC. This may be attributed to a successively increasing order of oxygen balance^8,9.

Data on deflagration temperature (Table 4) indicate that gradual decomposition of NC extended from 180^OC to 200^OC, as yellow vapourisation was observed around 200^OC. DTA curves of the first four compositions containing 55 per cent RDX and 36 per cent NC - 70 per cent RDX and 24 per cent NC show uniformity and single mass decomposition, as only one exotherm was recorded. The regular increase in the decomposition temperature is attributable to successively increased solid loading of RDX within the NC matrix. For the last two compositions containing 75 per cent RDX and 20 per cent NC and 80 per cent RDX and 16 per cent NC, respectively, two exotherms have been recorded. The first exotherm at 203^OC for the former and at 207^OC for the latter indicate decomposition of propellant mass formed with the maximum loadable RDX within the NC matrix, whereas the second exotherm at 222^OC for both the compositions indicates decomposition of surplus RDX, which could not be loadable within 20 per cent and 16 per cent of NC matrix, respectively (Fig. 4). This observation indicates nonsuitability of these two compositions for tank gun applications.

Figure 4. DTA curves

Calorimetric value was found to increase regularly, indicating a successively higher energy output due to the more exothermic reaction between NC and increased content of RDX in comparison to the compositions containing 55 per cent RDX and 36 per cent NC - 80 per cent RDX and 16 per cent NC (Table 4).

Data on thermal stability obtained from Bergmann and Junk test (Table 5) indicate that all the compositions were thermally stable, and stability of the subject composition for tank gun applications was better assessed when it was manufactured in a multitubular configuration. Secondly, P _vs dp/dt profile of the multitubular propellant (Fig. 2) was found to be of non-peaky nature as compared to the cord configuration. For compositions containing 70, 75 and 80 per cent RDX, even though the
force constant increased successively, values of

,

, and

also increased drastically for beyond the desired levels for tank gun applications. Hence, these compositions have not been selected for tank gun applications. The same is confirmed by cv firing, wherein a sudden rise in P _vs dp/dt profile was noticed in the case of composition containing 80 per cent RDX and 16 per cent NC (Fig.3).

Table 5: Data on thermal stability (Bergmann and Jank test at 120^OC) of the propellant compositions

The measured sensitivity values given in Table 3 indicate increasing trend of sensitivity from the compositions containing 55. per cent RDX and 36 per cent NC - 80 per cent RDX and 16 per cent NC. This may be attributed to a successively increasing order of oxygen balance^8,9.

Data on deflagration temperature (Table 4) indicate that gradual decomposition ofNC extended from 180^OC to 200^OC, as yellow vapourisation was observed around 200^OC. DTA curves of the first four compositions containing 55 per cent RDX and 36 per cent NC - 70 per cent RDX and 24 per cent NC show uniformity and single mass decomposition, as only one exotherm was recorded. The regular increase in the decomposition temperature is attributable to successively increased solid loading of RDX within the NC matrix. For the last two compositions containing 75 per cent RDX and 20 per cent NC and 80 per cent RDX and 16 per cent NC, respectively, two exotherms have been recorded. The first exotherm at 203^OC for the former and at 207^OC for the latter ind;cate decomposition of propellant mass formed with the maximum loadable RDX within the NC matrix, whereas the second exotherm at 222^OC for both the compositions indicates decomposition of surplus RDX, which could not be loadable within 20 per cent and 16 per cent of NC matrix, respectively (Fig. 4). This observation indicates nonsuitability of these two compositions for tank gun applications.

Calorimetric value was found to increase regularly, indicating a successively higher energy output due to the more exothermic reaction between NC and increased content of RDX in comparison to the compositions containing 55 per cent RDX and 36 per cent NC - 80 per cent RDX and 16 per cent NC (Table 4).

Data on thermal stability obtained from Bergmann and Junk test (Table 5) indicate that all the compositions were thermally stable, and stability was higher in comparison to compositions containing 55 per cent RDX and 36 per cent NC - 80 per cent RDX and 16 per cent NC. This may be attributed to decreasing gaseous volume of nitrogen oxides, as the NC content decreased successively.

Data on mechanical properties given in Table 6 indicate the tensile strength and percentage elongation to be in the decreasing order from the composition containing 55 per cent RDX and 36 per cent NC - 70 per cent RDX and 24 per cent NC. However, mini-samples of the compositions containing 75 per cent RDX and 20 per cent NC- 80 per cent RDX and 16 per cent NC got broken at the grips, while fixing the jaws of the I1}stron machin~ during the course of experiment. This is indicative of poor mechanical properties. Additionally, the flexural test indicates decreasing order of flexibility in terms of displacement at yield from the composition containing 55 per cent RDX and 36 per cent NC - 80 per cent RDX and 16 per cent NC. This may be attributed to successively lower contents of NC and DOP, which are responsible for mechanical properties. The availability of long carbon-carbon chains in the molecular structures of NC and DOP causes absorption of strains and stresses, contributing towards improved mechanical properties. Higher value of
percentage compression for the selected composition containing 65 per cent RDX and 28 per cent in NC in comparison to the conventional NQ propellant indicates potentiality of its use in tank guns, particularly at high solid loading.

Table 6: Data on mechanical properties of the propellant compositions

5. CONCLUSION

The propellant composition III containing NC (13.1 per cent nitrogen) DOP/carbamite/RDX (5 μm) has been found to provide higher force constant (1200 j/g) with relatively lower T_f (3210 K), reasonably good burning rate characteristics and mechanical properties.

REFERENCES

1. Rao, K.P.; Umrani, P.K.; Nair, R.G.K. & Venkatesan, K. Studies on some aspects of propellants for improving the performance of tank guns. Del Sci. J., 1978,37(1),51-57.

2. Flanagan; Joseph, E.; Haury & Vernon, E. Cool burning gun propellants containing triaminoguanidine nitrate and cyclotetramethylene tetranitramine with ethyl cellulose binder. US Patent 3,909,323 dated 30 September 1975. 3p.

3. Flanagan, et al. Gun propellants containing ilitraminoguanidine. US Patent 4, 373, 976 dated 15 February 1983.

4. Flanagan; Joseph, E.; Haury & Vernon, E. Gun propellants. Brit. Patent UK 1,432,327, dated 14 April 1976. 3p.

5. Rao, K.P. Calculation ofthermochemical constants of propellants. Del Sci. J., 1979,29(1),21-26.

6. Singh, Surjit. Handbook on solid propellants, Vol. 2; Double base propellants, Ch. 3. India, 1976. CI (ME) Report 1/76, 124p.

7. Pillai, A.G.S.; Dayanandan, C.R.; Joshi, M.M.; Patgaonkar, S.S.& Karir, J.S. Studies on the effects of RDX particle size on the burning rate of gun propellants. Del Sci. J., 1996, 46(2), 83-86.

8. John, Mullay. Relationship between sensitivity and molecular electronic structure. Prop. Explos. Pyrotech., 1987, 12, 121-24.

9. Yolk, F.; Bohn, M.A. & Wunsch, G. Oetennination of chemical and mechanical properties of doublebase propellants during ageing. Prop. Explos. Pyrotech., 1987, 12, 81-97.

Posts · by **Yohannes** » Tue Apr 24, 2018 1:44 am

The following are Lamoni's resources (they can be downloaded from the internet for free (I believe, someone please correct me if I am wrong here), so I will just list them here (thank you to our dedicated Senior N&I RP Mentor Lamoni!):

http://www.introni.it/pdf/Navy%20Electr ... %20Hbk.pdf
https://pdfs.semanticscholar.org/a751/a ... 99f173.pdf
http://mirror.thelifeofkenneth.com/lib/ ... arfare.pdf
http://www.diva-portal.org/smash/get/di ... TEXT01.pdf
http://www.vtvt.ece.vt.edu/research/ref ... REPORT.pdf
http://nnt.es/Information%20warfare%20a ... curity.pdf
http://www.oldcrows.org.au/files/2008%2 ... Rogers.pdf
http://indianstrategicknowledgeonline.c ... sharma.pdf
https://www.researchgate.net/profile/Mi ... b0e225.pdf
https://dspace.mit.edu/bitstream/handle ... sequence=2
http://www.dtic.mil/get-tr-doc/pdf?AD=ADA478587
http://www.academia.edu/download/349357 ... rines_.pdf
https://pdfs.semanticscholar.org/1c58/4 ... d302e2.pdf
http://www.hydrogen-peroxide.us/uses-ox ... n-2001.pdf
https://courses.engr.illinois.edu/npre4 ... ystems.pdf
http://www.dtic.mil/get-tr-doc/pdf?AD=ADA538633
https://dr.ntu.edu.sg/bitstream/handle/ ... sequence=1
https://calhoun.nps.edu/bitstream/handl ... sequence=1
https://calhoun.nps.edu/bitstream/handl ... sequence=1
https://www.researchgate.net/profile/J_ ... 0-2000.pdf
http://www.dtic.mil/get-tr-doc/pdf?AD=ADA477822
https://hal.archives-ouvertes.fr/hal-01073612/document
https://pdfs.semanticscholar.org/192f/7 ... b7ac3d.pdf
https://dr.ntu.edu.sg/bitstream/handle/ ... sequence=1
http://citeseerx.ist.psu.edu/viewdoc/do ... 1&type=pdf
http://www.dtic.mil/get-tr-doc/pdf?AD=ada428713
http://jmss.org/jmss/index.php/jmss/art ... ad/266/280
https://dr.ntu.edu.sg/bitstream/handle/ ... sequence=1
http://www.dtic.mil/get-tr-doc/pdf?AD=ADA519221
http://www.dtic.mil/get-tr-doc/pdf?AD=ADA471258
https://fortunascorner.com/wp-content/u ... denial.pdf
http://www.dtic.mil/get-tr-doc/pdf?AD=ADA275348
https://calhoun.nps.edu/bitstream/handl ... sequence=1
http://www.dept.aoe.vt.edu/~brown/VTShi ... 2Paper.pdf
https://www.researchgate.net/profile/Da ... 5e08af.pdf
http://www.dtic.mil/get-tr-doc/pdf?AD=ADA461955
http://www.dtic.mil/get-tr-doc/pdf?AD=ADA461887
http://www.abcm.org.br/anais/conem/2010 ... 0-0856.pdf

Posts · by **Yohannes** » Tue Apr 24, 2018 2:23 am

[ Out of character: Idea for posting this comes from Palmyrion. Not sure if this will help. Thank you!

Ballistic performance of composite metal foams

Composite Structures 125 (2015) 202–211
Journal page: https://www.journals.elsevier.com/composite-structures
http://dx.doi.org/10.1016/j.compstruct.2015.01.031
0263-8223/ 2015 Elsevier Ltd. All rights reserved.

AUTHORS: Matias Garcia-Avila^a, Marc Portanova^b, and Afsaneh Rabiei^a*

^aAdvanced Materials Research Lab, Department of Mechanical and Aerospace Engineering, North Carolina State University, Raleigh, USA
^bAviation Applied Technology Directorate (AATD), U.S. Army Research, Development & Engineering Center, Fort Eustis, USA

*Corresponding author at: Department of Mechanical and Aerospace Engineering,
North Carolina State University, 911 Oval Drive, Campus Box 7910, Raleigh, NC
27695, USA. Tel.: +1 919 513 2674; fax: +1 919 515 7968.
E-mail address: arabiei@ncsu.edu (A. Rabiei).

Article history:
Available online 29 January 2015

1. Introduction

High-performance hard armor systems for ballistic protection of aircraft, ground and amphibious vehicles, and personnel have always been the subject of study for researchers. Hard armor systems typically consists of multiple layers, with a ceramic or ceramic composite plate at the strike face, backed with a ductile material such as ballistic steel or aluminum, or a high performance fiber reinforced composite. This hybrid arrangement of layers allows the armor system to defeat the projectile upon impact, with the ceramic layer blunting and eroding the projectile due to its high hardness, and the more ductile/high tensile backing plate absorbing the residual kinetic energy of the fractured or deformed projectile through plastic deformation [1]. A variety of armor options are already available, however each one has its own limitations restricting their widespread use in many applications. The development of light-weight combat technology, such as aircraft and amphibious vehicles, and the need to improve higher mobility for ground troops requires the continuous reduction of armor weight while increasing their ballistic performance.

Composite armors made with ceramic strike face and high-strength fiber reinforced composites have been widely studied as light-weight armors in the past. Several types of ceramic materials, such as aluminum oxide (Al₂O₃), boron carbide (B₄C), silicon carbide (SiC), silicon nitride (Si₃N₄), and combinations of those are typically used as the strike face plate in armor systems [2–6]. These ceramics are combined with high-tensile strength back plates made of aramid fiber composites such as Kevlar^TM and Twaron^TM, or polyethylene composites such as Spectra^TM or Dyneema^TM to absorb the kinetic energy of the projectile. Although some of these combinations perform to some extent, the high cost of the constituents along with their heavy weight leaves room for improvement.

Composite metal foam (CMF) is low-weight high-strength metal foam manufactured using hollow metallic spheres embedded in a solid metal matrix. This material has shown superior mechanical properties compared to any other metal foam [7–15]. These outstanding qualities of CMFs are further improved under high-speed impact type of loading (similar to that in ballistic impact) compared to quasi-static loading [16]. These properties have made composite metal foams strong candidates for applications in composite armor systems. In this paper, CMF manufactured with 2 mm steel hollow spheres, embedded in a stainless steel matrix, and processed using powder metallurgy technique, was used to fabricate a new light-weight high performance composite armor system. The CMF was bonded to a ceramic plate on the strike face. Some samples were tested without any backing plates and some used a thin layer of aluminum or Kevlar^TM backplate behind the CMF. Ballistic testing was performed using U.S. National Institute of Justice (NIJ) standard 0101.06 [17] for 7.62 x 51 mm M80 (Type III) and 7.62 x 63 mm M2 Armor Piercing (AP) (Type IV) threats.

A finite element approach was used to simulate ballistic impact and predict the energy absorbed by the composite metal foam (CMF) layer within the composite armor system. A full 3D model of the composite armor was studied using a Lagrangian formulation in Abaqus/Explicit 16.3 commercial solver.

2. Material processing

Steel–steel composite metal foam (S–S CMF) panels were manufactured using hollow spheres embedded in a stainless steel powder matrix and processed using powder metallurgy technique previously developed [8,10,12,16,18]. Hollow steel spheres with 2 mm outer diameter and 200 μm sphere wall thickness were manufactured by Hollomet GmbH in Dresden, Germany using lost core technique [19,20]. 316L stainless steel powder with 44 μm particle size from North American Hoganas high Alloy LLC was used as matrix material. Fig. 1A shows a 30 x 30 cm CMF panel after processing. Boron Carbide (B4C) ceramic tiles were used as the strike plate, and Kevlar^TM or aluminum 7075 panels were used as backing plates in the armor system. All plates were 30 x 30 cm with different thicknesses of ceramic or CMF to maintain a total thickness around 25 mm in all samples with or without backing plates.

Fig. 1. (A) CMF panel processed using powder metallurgy technique for the application in armor system and (B) schematic of the side cross-section of the complete armor system showing CMF panel between a B4C ceramic strike plate and a Kevlar^TM or Al-7075 backplate (thicknesses not to scale).

The multi-layered composite armor system was assembled by bonding the CMF panel to a ceramic tile on one side, and either no backplate, or either an Al-7075 (Ceramic-CMF-AL) or a simple weave Kevlar^TM plate with fiber ultimate strength of 2.9 GPa (Ceramic-CMF-KV) on the other side. The assembled sandwich panel was wrapped with a single layer of 6oz plain-weave fiber glass infused in epoxy and bonded using vacuum bagging techniques and room temperature curing, in an attempt to keep failed ceramic fragments from ejecting during impact. Fig. 1B shows the assembly of the composite armor system, with a backplate and a total thickness to about 25 mm. Table 1 shows some properties for each layer used in the composite armor plates, along with threat type and impact velocities.

Table 1
Some properties of composite armor plates, along with the threat types and impact velocities.

3. Ballistic experiments

Ballistic testing of the composite armor system was performed using the guidelines included in the National Institute of Justice (NIJ) standard 0101.06 [17] for Type III (7.62 x 51 mm M80) and Type IV (7.62 x 63 mm M2 AP) threats. Fig. 2 shows a top view sketch of the setup for the ballistic experiments. The composite armor system was placed against a heated Roma Plastilina No. 1 (clay), following the standard guidelines, in order to monitor the total out of plane deformation of the back of the armor, which is an indication of the potential body trauma caused by the impact. To prevent serious injury, NIJ 0101.06 specifies a maximum of 44 mm for the depth of penetration (DOP) into the clay and no limit on the diameter of the footprint, or back face signature (BFS). A ‘‘Mann’’ gun mounted on a two axis rig was used for the ballistic tests. Accurate measurements of the projectile speeds were possible using two velocity chronographs located between the gun and the target. Two high speed cameras were aimed at the impact face and the rear of the target to monitor the impacts. A 5 m distance was maintained between the gun and the target with a zero angle of obliquity the gun.

Fig. 2. Top view sketch of the ballistic test setup showing gun barrel, bullet velocity chronograph, target location, and high-speed cameras.

4. Finite element analysis

Studying the behavior of the armor system under ballistic impact using finite element analysis (FEA) provides an understanding of the failure ] mechanisms and a powerful and inexpensive tool for optimization of the ballistic system.

Hydrocodes are computer programs which handle propagation of shock waves, stress, strain, velocities, etc. within a continuum material as a function of time and position [21]. The relationship between these changes in the material state can be calculated using classical continuum mechanics such as conservation of mass, momentum, and energy. There are two major types of hydrocodes descriptions to create a system of differential equations, Lagrangian and Eulerian. To solve these equations, material properties are used to relate stress and strain and define failure mechanisms within the material, and equations of state relate internal energy and density changes with internal pressure [21]. Typically, Lagrangian solutions are simpler and require fewer equations to be solved than that of Eulerian definitions, thus requiring less computing power. For this reason, Lagrangian descriptions are preferred to solve the majority of finite element models.

4.1. Material models

Gordon Johnson and William Cook developed a constitutive model for ductile materials subject to high strain rates [22]. Their material model gives an expression of stress as a function of strain, strain rate, and temperature and has become the standard when modeling metals at high strain rates. Eq. (1) shows the expression of the Johnson–Cook material model, with

being the stress,

and

the effective plastic strain and reference strain rate respectively,

the homologous temperature, and five material constants
A, B, C, n, and m.

Constant A represents the yield strength, with B and n being strain hardening constants of the material which can be obtained through quasi-static loading tests. Constant C is the strain rate sensitivity of the material and it is found from high strain rate testing.

gives a material softening effect with increasing temperature and can also be found by varying the temperature of the sample while testing. Due to the accurate prediction of the material strength by this model, several materials models for metals have already been developed by Johnson and Cook [22].

The behavior of ceramic face plate and the bullet has already been studied by other researchers [23,24]. The purpose of this FEA analysis is to study the behavior and energy absorption of composite metal foams at high-speed impacts. As a result, the focus of this study is on the behavior of CMF with the assumption that the ceramic failure has already taken place and the bullet has already been blunted. Although this model does not include the complete behavior of the armor system, it can serve as a parametric tool to understand the behavior of the composite foam and the aluminum backing plate as a coupled system.

Composite metal foam (CMF) has unique material properties that are not easy to fit into any preexisting constitutive material model. Typical stress–strain curve under quasi-static compression for S–S CMF manufactured using 2 mm spheres and powder metallurgy technique is shown in Fig. 3. Similar to all metallic foams, steel–steel composite metal foam is characterized by an elastic region, followed by a yield and a plateau region. During the ‘‘plateau’’ region, the porosities continue collapsing under compression, until all porosities are collapsed and the material starts behaving like a solid material. In the case of composite metal foams the presence of a matrix between spheres causes a strain hardening effect during the period in which spheres are collapsing, which is seen as a tilted plateau in the stress–strain curve shown in Fig. 3. Further details about the typical stress–strain curves of CMFs under compression can be found elsewhere [7,9]. When loading S–S CMF under high strain conditions, the material exhibits an increase in yield strength due to the inertial effects and cushioning effect caused by the compression of the air trapped in the porosities [16,25,26]. This effect is observed in Fig. 3 by the dotted curve corresponding to the dynamic behavior of steel composite foams at a strain rate of 3277 1/s tested in a using Hopkinson Bar system. Further details related to that experiment and resulted data are presented elsewhere [25]. The strain rate sensitivity of composite foams and the improvement in their energy absorption capabilities (

) at high strain rates can be easily observed in this figure at strain levels up to 25–30% strain. At higher strain levels (above 25–30%) the strength of the material matches to that under quasi-static loading. In this case, the energy absorption of the material was estimated to be between 2–3 times higher than that of quasi-static loading conditions [25,26]. Although the strain rate in ballistic testing is much higher, defining the material property based on energy absorption is the only quantitative way to simulate the behavior of composite metal foams. This material definition provides a way to estimate the material energy absorption as a function of the actual compressive strain observed on the foam upon their inspection after ballistic impact. For these reasons, the stress–strain curve of CMFs under ballistic loading is predicted using the total value of the energy absorbed per unit volume of the compressed foam upon the inspection of the material after ballistic impact and considering the strengthening effect due to the strain rate sensitivity of CMFs.

Fig. 3. Typical stress–strain curves for composite metal foam for quasi-static and dynamic loading curve [23].

4.2. Model setup

In first step single layer CMF and aluminum 7075-T6 panels with 300 x 300 mm dimensions were modeled separately and meshed in Abaqus/Explicit. Modeling the erosion of the projectile and the ceramic layers is considered beyond the scope of this study mainly because it is well established in the literature. As the result, our focus will be on the behavior of CMF layer with a backing plate. Since all of the experimental studies indicated that the ceramic layer spread the load onto the CMF layer leaving a perforation area of about 12 mm upon the impact of bullet, a solid and non-deformable cylinder with 12 mm diameter was used to simulate the effect of bullet- ceramic layer group and perforate the armor
system similar to the experimental ballistic tests. Fig. 4 shows an illustration of the finite element model setup for a coarse mesh definition. Quadratic tetrahedral elements were used for all bodies and the model was constrained using a fixed support at the outside edges of the plates, as suggested in the literature [23,24,27–29]. The model setup shown in Fig. 4 corresponds to the coarse mesh definition, with large elements at the outside of the panel and a progressive finer mesh at the center of the panel. In order to obtain accurate results under bending, up to 3 elements were considered through the thickness of the aluminum layer in the coarser mesh definition. Smaller elements were considered for the finer mesh definition. Larger elements at the outside, where bending was not observed, should not affect the results at large.

Fig. 4. Finite element model setup in Abaqus/Explicit 16.3 showing the mesh for the projectile (red), CMF layer (yellow) and Aluminum 7075-T6 backplate (green), thicknesses and dimensions are not in scale. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

The ballistic clay is considered to have a yield strength of 1 MPa with an elastic-perfectly plastic behavior. A pressure of 1 MPa was considered as a support boundary condition on the back of the aluminum plate to simulate the resistance provided by the clay during the ballistic tests. Frictionless contact definition was defined for all surfaces [27] and general contact definitions were considered between all elements, in agreement with the literature [23,27,28]. Lagrangian formulation was used to solve the conservation equations. The energy absorbed by each panel was obtained from the simulation and compared to the experimental tests.

4.3. Material model definitions

Due to the complexity of the composite armor system, several material constitutive models were considered for each layer material:

4.3.1. Aluminum 7075-T6 backing plate

A Johnson–Cook constitutive material model was considered for the aluminum backing plates used in the composite armor with the parameters [30] shown in Table 2.

Table 2
Johnson–Cook and elastic material properties for Al7075-T6 plate.

4.3.2. Steel–steel composite metal foam

In order to maintain the shape of the stress–strain curve shown in Fig. 3 as a qualitative criterion under high-strain rate loading, a multi-linear stress–strain definition is used in the model to define the material response and shown in Fig. 5. Our previous studies indicated that the CMFs exhibit a strengthening effect due to the strain rate sensitivity of CMFs (Fig. 3) [25]. In this study, the energy absorbed per unit volume of CMF under ballistic impacts in experimental studies [26] is used to estimate the values of the yield and plateau strength. The density of the material is considered 2.8 g/cm³, with an elastic modulus of 13.2 GPa and Poisson ratio of 0.1. To minimize mesh size dependence of the FEA results, three different mesh sizes of coarse, medium, and fine were used to simulate penetration impact on an aluminum backing plate. Since the load would be applied at the center of the plate, a center area of the plate of 100 mm in diameter was meshed using smaller elements. For the coarse, medium, and fine meshes, the center area was meshed using 2, 0.9, and 0.7 mm maximum size elements and the outer area used 15, 10, and 9 mm maximum size elements respectively.

Fig. 5. Multi-linear stress–strain definition used for S–S CMF at ballistic relative strain rates.

5. Results and discussion

5.1. Experimental results

All various types of samples without backplates and with Kevla^TM or Al backplates were able to stop the projectiles with DOPs less than 44 mm, which is considered the maximum allowable penetration according to the NIJ 0101.06 standard.

Digital images of the front strike face and rear face of a CeramicCMF (no backing) armor panel after NIJ-Type III impact test are shown in Fig. 6A-B respectively. The combination of ceramic and S–S CMF showed superior ballistic performance under Type III threats at impact speeds at or above NIJ standard requirements. The ceramic strike plate successfully blunted and eroded the projectiles upon impact. As seen in Fig. 6A a very small amount of NIJ-Type III bullet jacket material is left embedded in the armor. Radial cracks on the ceramic plate are seen spreading out of the impact area, forming an outward crater seen in Fig. 6A. The ceramic layer successfully spread the load onto the CMF layer, which is going to absorb most of the kinetic energy of the bullet through plastic deformation and subsequent densification. As seen in Fig. 6B, this residual tensile stresses on the back of the CMF layer formed radial cracks extending from the impact area. Some fragments of the CMF were ejected from this area. This observation encouraged the idea of adding a thin layer of backplate behind the CMF to absorb those residual tensile stresses and catch any low velocity fragments.

Fig. 6. Digital images of a Type III impact area on an armor system without backing plate: (A) front strike face showing complete arrest of the bullet and (B) rear face showing bulging of CMF and small amount of cracking due to tensile stresses.

Fig. 7A-B shows front and rear faces on Ceramic-CMF-AL and Ceramic-CMF-KV composite armor respectively after an NIJ-Type III impact. Both types of composite armor systems successfully stopped the projectiles for single and multi-hit scenarios, for bullet speeds at or higher than those specified by NIJ 0101.06 standard. It is observed that the ceramic plate caused failure of the projectile upon impact. A post impact inspection of the CMF layer showed that it had successfully absorbed the kinetic energy of the projectile through compression and the aluminum and/or Kevla^TM backing plates supported the CMF layer as it was compressed, absorbing any residual tensile stresses and catching ejecta. Little or no fragments of projectile could be found after the NIJ-Type III tests on either one of the composite armor systems, which suggests a complete disintegration of the 7.62 x 51 M80 bullet for all samples tested.

Fig. 7. Front-strike and rear face digital images of impact area of NIJ-Type III tests showing complete arrest of the bullet and rear bulging of backing plate for: (A) Ceramic-CMF-AL, (B) Ceramic-CMF-KV.

Fig. 8 shows a comparison of the areal density for CMF (ECMF) for Ceramic-CMF (no backing), Ceramic-CMF-AL, and Ceramic-CMFKV samples for different impact velocities using NIJ-Type III projectiles. Using a stand-alone Ceramic-CMF composite armor system yielded successful results against Type III threats up to speeds at or above the NIJ standard requirements. A previous study by the authors suggested an increase on the yield strength of CMF material at high strain rates by over a factor of 2 at impact speeds up to 26 m/s [16]. In addition, a preliminary study of the ballistic properties of CMF suggested an energy absorption increase between 2–3 times higher at NIJ-Type III and IV impact speeds [26]. This increase in performance of CMF at high loading rates suggested a possible reduction in thickness of the ceramic and CMF layers, resulting in lighter and thinner armor plates. Also, adding Aluminum or Kevlar^TM backing plates to the back of the CFM allowed a reduction of the
thickness of both ceramic and CMF, which resulted in a weight reduction of 17% compared to the no backing samples.

Fig. 8. Areal density for Ceramic-CMF, Ceramic-CMF-AL, and Ceramic-CMF-KV composite armor tested under Type III conditions at different impact speeds.

Similarly, Fig. 9A-B shows front and rear faces of Ceramic-CMFAL and Ceramic-CMF-KV composite armor respectively after impact of NIJ-Type IV projectiles. As can be seen, similar behavior of the armor system was obtained for Type IV projectiles. In this case, the partially disintegrated the hardened steel core and part of the bullet jacket were left embedded in the armor, as shown in Fig. 10A. The AP projectiles tested on the Ceramic-CMF-AL and Ceramic-CMF-KV showed 40–65% mass loss at impact velocities between 860–890 m/s (Fig. 10B), depending on the thickness of the ceramic.

Fig. 9. Front-strike and rear face digital images of impact area of NIJ-Type IV tests showing complete arrest of the bullet and rear bulging of backing plate for: (A) Ceramic-CMF-AL, (B) Ceramic-CMF-KV.

Fig. 10. (A) NIJ-Type IV AP projectile embedded in a Ceramic-CMF-KV sample after ballistic test at 865 m/s projectile speed and B) recovered AP projectiles from Ceramic-CMF-KV and Ceramic-CMF-AL showing 40–65% bullet mass loss.

Areal density for CMF (E_CMF) versus projectile speed for NIJType IV tests are shown in Fig. 11 for Ceramic-CMF (no backing), Ceramic-CMF-AL, and Ceramic-CMF-KV composite armors. For the Ceramic-CMF-AL, and Ceramic-CMF-KV composite armor systems designed for NIJ-Type IV threats, the addition of the backing plate and the reduction in thickness of ceramic and CMF layers resulted in a 13% and 20% weight reduction respectively compared to the Ceramic-CMF armor, with the Kevlar backed samples being 5% lighter than the AL backed samples due to a thinner ceramic layer used.

Fig. 11. Areal density for Ceramic-CMF, Ceramic-CMF-AL, and Ceramic-CMF-KV composite armor tested under Type IV conditions at different impact speeds.

Fig. 12 illustrates the deformation mechanism of a Ceramic CMF-backing plate composite armor system. As discussed before, upon impact, the hard ceramic plate blunts the projectile due to large compressive stresses developed at the projectile tip. When the compressive stresses travel through the ceramic layer and reach the interface between ceramic and CMF layer, tensile stresses are created due to the sudden change in mechanical impedance between the two layers. These tensile stressed are then reflected back towards the impact face. The intersection between the compressive and tensile stress waves traveling through the ceramic layer creates a high stress concentration area at angles between 25^O–75^O normal to the outer surface of the ceramic, which results in the failure of the ceramic material forming a Hertzian cone zone [31]. This cone detaches from the ceramic and serves to distribute the compressive load at the ceramic-CMF interface over a larger area. The residual tensile waves in the ceramic form circumferential and radial cracks and due to this localized fracture and comminution
in the vicinity of the impact area, results in an outward crater at the impact face. As penetration progresses, compressive waves build up on the CMF layer until its yield point and further, deforming plastically at high compressive loads and absorbing the kinetic energy of the projectile. The light weight backing plate below the CMF layer absorbs any residual tensile stresses of the armor system, maintaining the integrity of the impact area and keeping debris contained inside the perforation. Using a combination of ceramic, CMF, and backing plate, provides a layer-based functional design solution where each constituent contributes in a collaborative fashion to the ballistic energy absorption process.

Fig. 12. Representation of the failure mechanism of Ceramic-CMF-Backing plate composite armor subjected to ballistic loading (thicknesses are not to scale).

In ballistic impacts, most of the kinetic energy of the projectile is transformed into brittle fracture of the ceramic under compression and tension, plastic deformation of the projectile and backing plate, and heat. For this study, and since the local temperature at the point of impact could not be measured, the heat generation is considered negligible for energy calculations. Using an energy approach previously discussed [26], the energy absorbed by each component in the composite armor system can be approximated. Upon impact, the kinetic energy of the projectile (E_KE) is transferred to the armor system as the energy used for plastic deformation of the bullet (E_bullet), energy absorbed by the ceramic (E_ceramic), energy absorbed by CMF layer (ECMF), energy absorbed by the backing plate (E_backing), and residual energy from clay deformation or debris ejected from the target in the event of complete penetration (E_res,), as shown in Eq. (2):

Similar studies on energy absorption of armor systems have been reported in the literature [32]. The energy per unit volume of material for the projectile, ceramic, backing plate, and clay can be calculated from their respective stress–strain curves by calculating the area under the curve using a strain energy (w_p) method according to Eq. (3)

where w_p is essentially the area under the stress–strain curve in J/m³. Using material properties of each layer and multiplying the value of this strain energy by the total amount of material under deformation per layer (bullet, ceramic, backing plate, clay), the total kinetic energy dispersed by each component of the composite armor system is calculated. Fig. 13A-C shows a representation of each method of calculating strain energy for each layer of the composite armor against both NIJ-Type III and Type IV projectiles, where

and

are the yield strength of the material and the corresponding strain (for ductile materials),

and

are the ultimate strength and corresponding ultimate strain, respectively. Properties of each component are obtained from the literature [22,30,33] and shown in Fig. 13.

Fig. 13. Stress–strain curve used for analytical method to calculate plastic strain energy for each layer with (A) corresponding to hardened steel bullet core of NIJ-Type IV projectile and NIJ-Type III projectile, (B) Kevlar^TM backing plate, and (C) aluminum 7075-T6 backing plate.

Residual energy (E_res) was calculated from BFS and DOP measurements on clay, a

of 1 MPa, along with the method shown in Fig. 13B, and residual velocities of particles obtained from high speed video.

Solving for E_CMF in Eq. (2) and substituting all energy values calculated gives the estimated energy absorption by the CMF layer. Values for all energy absorbed per layer in percentage of total kinetic energy are shown in Table 3.

Table 3
Energy absorbed by each layer in the composite armor system as the % of bullet kinetic energy for both NIJ-Type III and Type IV impacts.

Fig. 14 shows the kinetic energy absorbed by the CMF layer in the composite armor system for both NIJ-Type III and Type IV tests. As can be seen, with the appropriate arrangement of layers, CMF was capable of absorbing 60–70% of the kinetic energy of the bullet, proving the superior energy absorption capabilities of composite metal foams at high impact speeds. It is also observed that by adding a soft backing plate behind CMF, the areal density was decreased with no adverse effect on the energy absorption capabilities of the total composite armor system.

Fig. 14. Energy absorbed by CMF layer for both NIJ-Type III and Type IV tests in all composite armor systems tested.

5.2. Finite element analysis results

5.2.1. Mesh sensitivity study

The penetrator’s net nodal force reactions were obtained and plotted against projectile displacement for a depth of penetration of 20 mm (such DOP is selected based on our ballistic studies) (Fig. 15). No issues were encountered with over skewed elements for the 2 finer mesh definitions. It can be seen that the solution for both medium and fine meshes is comparable and a finer mesh does not provide a more accurate solution. For this reason, and to save on computing power, a medium mesh definition was used in the model.

Fig. 15. Force–displacement results obtained for a 20 mm DOP simulation for the mesh sensitivity study for coarse, medium, and fine mesh definitions.

Von-Mises stress plots for a 15.17 mm depth of penetration simulation are shown in Fig. 16 at 10, 30, 60, and 90 ls of penetration time. Compression of the CMF layer at 90 μs shows full densification up to 80% strain, with similar deformation pattern obtained in experimental tests. The aluminum backplate supports the rear face of the CMF, and deforms in tension absorbing the residual kinetic energy of the penetrator, leaving a bulging profile on the armor system similar to that shown in Fig. 7A and Fig. 9A.

Fig. 16. Cross-sectional Von-Misses stress contour plots on CMF and Al layer obtained for a depth of penetration of 15.17 mm at 1, 10, 30, 60, and 90 μs.

Fig. 17 shows the simulation results for energy absorbed by CMF and aluminum layers for both Type III and Type IV bullet speeds compared to the experimental results. As can be seen, a close prediction of the energy dissipated by the aluminum plate is obtained for all tests. For the CMF layer, an over-prediction of the energy absorbed by the FEA model is seen in Fig. 17. Fig. 18 shows the high x-y shear stresses developed in the CMF layer under puncture by the penetrator. These shear stresses could cause failure in the material and could hinder plastic flow under ballistic loading, artificially raising the energy absorption of the material in the FEA model. However, behavior of CMF under shear loading has not been studied extensively and as the result it was not taken into account in this material model. Although close prediction of the behavior of CMF has been obtained by this model, further characterization of CMF under shear loading is needed to consider complete material failure definitions and develop a more accurate model.

Fig. 17. FEA and analytical results for the energy absorbed by CMF and Al 7075-T6 layers for both NIJ-Type III and Type IV tests in all composite armor systems tested.

Fig. 18. Cross-sectional x–y shear plots for a 15.17 depth of penetration simulation, showing high shear areas on the CMF and aluminum layers.

6. Conclusions

Composite metal foam panels manufactured using 2 mm steel hollow spheres embedded in a stainless steel matrix and processed through powder metallurgy technique were used together with boron carbide ceramic and aluminum 7075 or Kevlar^TM back panels to fabricate a new composite armor system. This composite armor was tested against NIJ-Type III and Type IV threats using NIJ 0101.06 ballistic test standard. The highly functional layer-based design allowed the composite metal foam to absorb the ballistic kinetic energy effectively, where the CMF layer accounted for 60–70% of the total energy absorbed by the armor system, and allowed the composite armor system to show superior ballistic performance for both Type III and IV threats.

Finite element analysis results for ballistic loading of the armor system closely predicted the behavior and energy absorption of the CMF and aluminum layers. The Kevlar system was not considered in the simulation since the results for the energy absorbed by CMF in the system with the aluminum layer were successful. However, the failure mechanisms of CMF under ballistic loading are complex and further characterization of the material under shear loading is necessary prior to establish a comprehensive model of its behavior under ballistic loading.

Acknowledgments

The authors would like to acknowledge North Carolina State University’s Chancellor Innovation Fund (CIF) for its financial support that made this project possible. Special thanks to Dr. Robert Bryant and his team at the Advanced Materials and Processing Branch at NASA Langley Research Center, for granting access to their material processing facilities.

References

[1] Hetherington J, Smith P. Blast and Ballistic Loading of Structures. Oxford; Boston: CRC Press; 1994.
[2] Medvedovski E. Ballistic performance of armour ceramics: Influence of design and structure. Part 1. Ceram Int 2010;36:2103–15.
[3] Medvedovski E. Ballistic performance of armour ceramics: Influence of design and structure. Part 2. Ceram Int 2010;36:2117–27.
[4] David NV, Zheng JQ, Gao X-L. Ballistic resistant body armor: contemporary and prospective materials and related protection mechanisms. Appl Mech Rev 2009;62. 050802–050802.
[5] Tasdemirci A, Tunusoglu G, Güden M. The effect of the interlayer on the ballistic performance of ceramic/composite armors: experimental and numerical study. Int J Impact Eng 2012;44:1–9.
[6] Medvedovski E. Lightweight ceramic composite armour system. Adv Appl Ceram 2006;105:241–5.
[7] Rabiei A, O’Neill AT. A study on processing of a composite metal foam viacasting. Mater Sci Eng A 2005;404:159–64. http://dx.doi.org/10.1016/j.msea.2005.05.089.
[8] Neville BP, Rabiei A. Composite metal foams processed through powder metallurgy. Mater Des 2008;29:388–96.
[9] Vendra LJ, Rabiei A. A study on aluminum–steel composite metal foam processed by casting. Mater Sci Eng A 2007;465:59–67.
[10] Rabiei A, Vendra LJ. A comparison of composite metal foam’s properties and other comparable metal foams. Mater Lett 2009;63:533–6.
[11] Vendra L, Rabiei A. Evaluation of modulus of elasticity of composite metal foams by experimental and numerical techniques. Mater Sci Eng A 2010;527:1784–90.
[12] Rabiei A, Neville B, Reese N, Vendra L. New composite metal foams under compressive cyclic loadings. Mater Sci Forum 2007;539–543:1868–73.
[13] Vendra L, Neville B, Rabiei A. Fatigue in aluminum–steel and steel–steel composite foams. Mater Sci Eng A 2009;517:146–53.
[14] Vendra LJ, Brown JA, Rabiei A. Effect of processing parameters on the microstructure and mechanical properties of Al–steel composite foam. J Mater Sci 2011;46:4574–81.
[15] Brown JA, Vendra LJ, Rabiei A. Bending properties of al-steel and steel–steel composite metal foams. Metall Mater Trans A 2010;41:2784–93.
[16] Rabiei A, Garcia-Avila M. Effect of various parameters on properties of composite steel foams under variety of loading rates. Mater Sci Eng A 2013;564:539–47.
[17] NIJ 0101.06-U.S. Department of Justice. Ballistic Resistance of Body Armor NIJ Standard 0101.06 2008.
[18] Rabiei A, Vendra L, Reese N, Young N, Neville BP. Processing and characterization of a new composite metal foam. Mater Trans 2006;47:2148–53.
[19] Andersen O, Waag U, Schneider L, Stephani G, Kieback B. Novel Metallic Hollow Sphere Structures. Adv Eng Mater 2000;2:192–5.
[20] Stephani G, Kupp D, Claar TD, Waag U. Fabrication of Ti-based components with controlled porosity. Int Conf Powder Metall Part Mater 2001:50–8.
[21] Anderson Jr CE. An overview of the theory of hydrocodes. Int J Impact Eng 1987;5:33–59.
[22] Johnson GR, Cook WH. A constitutive model and data for metals subjected to large strains, high strain rates and high temperatures. In: Proc 7th Int Symp Ballist 1983; 21: p. 541–7.
[23] Bürger D. Rocha de Faria A, de Almeida SFM, de Melo FCL, Donadon MV. Ballistic impact simulation of an armour-piercing projectile on hybrid ceramic/fiber reinforced composite armours. Int J Impact Eng 2012;43:63–77.
[24] Feli S, Asgari MR. Finite element simulation of ceramic/composite armor under ballistic impact. Compos Part B Eng 2011;42:771–80.
[25] Rabiei A. Material with Improved Absorption of Collision Forces for Railroad Cars. Safety IDEA Project 20 final report, National Academy of Science 2014.
[26] Garcia-Avila M, Portanova M, Rabiei A. Ballistic performance of a composite metal foam-ceramic armor system. In: Proc metfoam 2013.
[27] Iqbal MA, Chakrabarti A, Beniwal S, Gupta NK. 3D numerical simulations of sharp nosed projectile impact on ductile targets. Int J Impact Eng 2010;37:185–95.
[28] Vanichayangkuranont T, Maneeratan K, Chollacoop N. Numerical simulation of level 3A ballistic impact on ceramic/steel armor. In: The 20th conference of mechanical engineering network of thailand 2006.
[29] Teng X, Wierzbicki T, Huang M. Ballistic resistance of double-layered armor plates. Twenty-fifth anniv celebr issue honouring profr Norman Jones his 70th birthd 2008; 35: p. 870–84.
[30] Brar NS, Joshi VS, Harris BW. Constitutive model constants for Al7075-T651 and Al7075-T6. Am Inst Phys Conf Ser 2009;1195:945–8.
[31] Fountzoulas CG, LaSalvia JC. Improved modeling and simulation of the ballistic impact of tungsten-based penetrators on confined hot-pressed boron carbide targets. In: Swab JJ, Halbig M, Sanjaythur, editors. Adv. Ceram Armor VII. John Wiley & Sons, Inc.; 2012. p. 209–17.
[32] Naik NK, Kumar S, Ratnaveer D, Joshi M, Akella K. An energy-based model for ballistic impact analysis of ceramic-composite armors. Int J Damage Mech 2012. 1056789511435346.
[33] Peroni L, Scapin M. Mechanical properties at high strain-rate of lead core and brass jacket of a NATO 7.62 mm ball bullet 2012;26.

Posts · by **Yohannes** » Wed Apr 25, 2018 7:19 pm

Military History: Acquiring Armour
Some Aspects Of The Australian Army’s Leopard Tank Purchase

The Author: Richard Pelvin

Richard Pelvin, who graduated from Monash University, also holds a Master of Defence Studies degree from the University of New South Wales. He worked in various areas of the Defence Department including the Navy and Army History Sections before becoming Curator of Offi cial Records at the Australian War Memorial. Resigning from the Public Service in 2002, he is now an independent military history researcher and Archival Consultant to the Army History Unit. He has published illustrated histories of Australia in World Wars I and II. A third volume on Vietnam is in preparation.Copyright of Full Text rests with the original copyright owner and, except as permitted under the Copyright Act 1968, copying this copyright material is prohibited without the permission of the owner or its exclusive licensee or agent or by way of a license from Copyright Agency Limited. For information about such licences contact Copyright Agency Limited on (02) 93947600 (ph) or (02) 93947601 (fax)

Article history:
Australian Army Journal, Vol. 3, No. 1, Summer 2005-2006: 193-212.

Using material from declassified National Archives of Australia files, the author of this article examines the purchase of German Leopard tanks in the first half of the 1970s by the Australian Army.¹ The files cover the years between 1971 and 1975, and contain correspondence from the Army’s Materiel and Operations branches. Given the topicality of the purchase of the new M1A1 Abrams tanks by the Australian Government, an analysis of the Leopard tank purchase highlights some useful aspects of the acquisition process as it occurred thirty years ago. The author does not attempt to recount the complete story of the Leopard acquisition but rather seeks to throw some light on the process as it was viewed in the then Department of Army (later Army Office).

The article seeks to demonstrate the vagaries of decision-making in the acquisition process. In the early 1970s, the Army chose the cheaper and less combat effective American M60 tank, aft er initially opting for the Leopard, and then reverted to the choice of purchasing the Leopard tank. The article also examines the criteria governing the numbers of tanks purchased, noting their resonance with the recent purchase of the Abrams. Finally, there is an assessment of the call, made in all seriousness, to examine the possibility of manufacturing the Centurion tank in Australia. The term ‘tank’ includes gun tanks and support vehicles: medium tank dozer (MTD), armoured vehicle launched bridge (AVLB), armoured recovery
vehicle (ARV) and medium tank mine clearing (MTMC).

Selecting a New Tank

In 1970, the Australian Army began to consider a replacement for its obsolescent Centurion tank fleet. Th e Centurion tank design was old and inferior to designs appearing in the region. Th e Centurion vehicle had also become unreliable, costly and difficult to maintain, causing serious disruption to the training process and placing heavy demands on maintenance and transport. In addition, the Centurion was being phased out worldwide, making spare parts expensive and their availability uncertain. The Centurion replacement project was designated Major Equipment Submission No. 9 (MES 9) and was initially submitted to the Defence Force Development Committee (DFDC) in August 1970. The DFDC advised the Minister for Defence, Malcolm Fraser, on the development of the Defence Force as a whole, based on current strategic assessments and resource levels. The committee’s membership comprised the Secretaries for Defence and Supply, the Chief of the Defence Force Staff , the Service Chiefs and invited members as appropriate. In respect of capital equipment proposals, the DFDC was advised by the Force Structure Committee.

At the behest of the DFDC the Army undertook an assessment of its future armour requirements. This assessment resulted in the publication in May 1971 of a study entitled Armour in the Australian Army 1975–1990. The study provided the general basis for the type and numbers of armoured vehicles requested ‘in that it describes the tasks to be carried out by the Medium Tank and the structure of armoured forces for the time-frame 1975-1990’. Outline Military Characteristics were derived and developed from the 1971 study and a number of tanks were selected for initial evaluation. For instance, an Army evaluation team assessed a number of designs in the United States, Britain, France, Germany and Japan during March and April 1971. Th e drafting of Operational Materiel Requirement (OMR) No. 1—The Medium Tank also commenced. This study outlined the operational characteristics required of the new vehicles. Such characteristics were to be the criteria against which Defence (or the Government?) would make a final selection of tank.

First Revise to MES 9

In October 1971 the Army submitted evaluation outlines to the DFDC in what became known as the First Revise to MES 9. The committee was advised that Armour in the Australian Army 1975–1990 had recommended, inter alia, that Australia required a medium-tank family of gun tanks and the full range of support vehicles. Sufficient tanks were required to equip not only three tank squadrons and their light aid detachments (LAD), but also the Armoured Centre, the Royal Australian Electrical and Mechanical Engineers (RAEME) Training Centre and the recovery platoons of four field workshops, maintenance support and five months of war stocks. Th e total number of vehicles of all types required was 139.

… the German Leopard
and the American M60A1
emerged as the two favoured
contenders for acquisition
by the Australian Army.

After the evaluation team had eliminated other tank types, the German Leopard and the American M60A1 emerged as the two favoured contenders for acquisition by the Australian Army. Initial data available on the two tanks indicated that both would satisfy the OMR to a large extent.² Th ere was little technical risk involved since both tank types had been in service in substantial numbers. Moreover, both types had been subjected to comprehensive product improvement programs to maintain their battle-worthiness in modern conditions, especially in fire control techniques. The German Leopard was considered to have more ‘stretch’ capacity than the American M60. Spare parts would be available over a sufficient period, with the opportunity for a ‘one-time last buy’ before the end of production. In addition, there did not appear to be any significant manpower implications involved in a purchase.

The Environment of Future Australian Military Operations

In its views on future armoured capabilities, the Force Structure Committee suggested that many of the assessments made in Armour in the Australian Army 1975–1990 on such issues as strategic policy, areas of operations and the future requirement for, and the role of, tanks had been made prior to the guidance available in another 1971 study paper entitled Environment of Future Australian Military Operations (EFAMO). As a result, the number and role of tanks and derivative vehicles needed to be reviewed in the light of EFAMO. Referring to armoured forces, EFAMO stated:

³

On 5 April 1972, the DFDC agreed with the FSC position and advised that the procurement of future tanks should be divided into two phases. Phase one was to satisfy the provisions laid out in the EFAMO paper and involved the provision of ‘sufficient tanks to ensure the Army retains a high standard of training in all respects of armoured warfare; and in particular, infantry/tank cooperation’. Acquisition
should be designed to enable the Army to keep abreast of developments in tactics and technology in armoured and mechanised warfare in the timeframe 1975–1990.

Phase two of procurement called for the provision of ‘sufficient additional tanks to permit the Army to deploy and sustain one tank squadron in operations’. Defence would take no procurement action in phase two unless changed strategic and military requirements made such action desirable. The numbers to be acquired would allow the deployment of one squadron on operations, as had been the case during the recent war in Vietnam. A phased acquisition program presented the Army with two problems. Th e first problem was the lead time required to put the extra tanks into service. Th e second problem was the probability of not being able to acquire tanks of the same type as were obtained under phase one—with all the associated operational, training and logistic complications that would be entailed.

MES 9 Second Revise

In September 1972, the Army submitted MES 9 (Revise 2) to the DFDC. In this second revision of its major equipment submission, the Army identified a need for a phase one purchase of 78 tanks. Of this number, 60 guns were to be tanks guns. The remaining armoured vehicles were to be divided between MTD and AVLB, ARV and MTMC types. Th e aim was to equip two tank squadrons (comprising 44 tanks), instead of three squadrons operating with Centurions. Tanks were also to be allotted to a field workshop (1), to the Armoured Centre (19) and to the RAEME Training Centre (2) while twelve vehicles were provided for maintenance support. Phase two would involve the procurement of a further 60 tanks to reinforce the two squadrons by eight tanks each and to equip a third tank squadron of 26 vehicles. A further 26 vehicles were earmarked for increased maintenance support and to provide a war reserve.

In the second revision, evaluation trials of the Leopard and M60 tanks would be completed by mid-1973 and fi rm orders placed early in fi scal year 1974–75. The two tank squadrons were also to be re-equipped by the end of 1976. In its consideration of the revised submission, the DFDC agreed that phase one should be deferred in order that planned authorisation of the first phase was not implemented in 1974–75.

The Armoured Fighting Vehicle Study

In September 1972, in Minute No 111/1972, the DFDC attempted to correct the deficiencies it perceived in the study, Armour in the Australian Army 1975–1990. The committee requested that the Army submit an armoured fighting vehicle study statement. Th e aim was to produce further useful information with regard to numbers and types of armoured vehicles in order to guide the further examination and development of the project.

Th e new study was ‘[t]o examine the need, role and required characteristics of armoured vehicles against the strategic environment of EFAMO and the tactical background of endorsed scenarios with the aim of deriving a rational vehicle range’. The operational roles of armoured vehicles were to be examined in relation to the strategic situations outlined in EFAMO. Th e study was also to be delimited by adherence to a number of scenarios endorsed by the DFDC in order to derive an appropriate range for the various vehicles. The study was to consider the resources required for two sets of circumstances: first, acquisition and retention of essential military skills; and second, to determine what was required as a basis for expansion.

Both of these examinations were to be ‘conducted with regard to the timing, likelihood and importance of situations given in EFAMO and with due regard to an optimisation of the mix of vehicles with cost’. Th e resulting paper was entitled The Armoured Fighting Vehicle Study and was submitted to the DFDC in August 1973.

Leopard or M60?

Meanwhile, a military team carried out evaluations of the two contending tank types, Leopard and M60, in the period between June 1972 and April 1973. The Military Board considered the evaluation team’s report at three separate meetings during August 1973 and the results of its deliberations were set forth in Military Board Minute No. 392/1973 of 30 August 1973. The evaluation team considered the main armament systems of the two vehicles to be excellent, although better fire control systems were required on both tanks in order to use their guns to full potential. Serious deficiencies were noted in the fighting arrangements for the crew commander on the M60. These deficiencies were attributed to the complex machine-gun cupola and a number of poor target-acquisition arrangements.

The evaluation team
considered that the
Leopard tank was superior
to the M60 in terms of
strategic movement.

The evaluation team considered that the Leopard tank was superior to the M60 in terms of strategic movement. Because the most eff ective method of movement in Australia was by rail, the size of the M60 precluded its carriage on the eastern coast railway network. As a result, if the M60 was chosen, tank-training activities would be eff ectively confi ned to the Puckapunyal area in Victoria. In contrast, the Leopard tank was portable over most Australian mainline railways and was thus more strategically versatile.

In addition, although the Leopard tank carried less armour, it was lighter, faster and more agile. Th e German tank could probably survive better on the modern battlefield since it presented a ‘smaller target for a shorter time’. The evaluation team also believed that the M60 did ‘not represent a significant improvement over the [mobility] characteristics of the Centurion [tank]’. A further consideration was that, although the M60 weighed 8 tons more than the Leopard vehicle, the former’s extra armour was spread over a greater surface area. Thus, the evaluators concluded that the ‘the degree of armoured protection provided by the M60 is only marginally greater than that provided by the Leopard’.

… the Leopard tank
was portable over most
Australian mainline
railways and was thus more
strategically versatile.

In respect to issues of mechanical reliability and ease of technical maintenance, both tanks were considered to be a great improvement on the Centurion. However, the Army’s Director of Electrical and Mechanical Engineers (DEME) expressed a strong preference for the Leopard. He noted that the latter vehicle, ‘being a more recently designed vehicle … has advanced maintenance engineering features when compared with the M60A1’. A factor that told heavily against the M60 was the fact that there was no armoured recovery vehicle based on the former’s chassis design. Th e US Army used the M88 ARV, which operated on an entirely different chassis and had no commonality with M60.

… the evaluators concluded
that the ‘the degree of
armoured protection
provided by the M60 is only
marginally greater than that
provided by the Leopard’

The introduction of even a few vehicles of this type into the Australian Army would place a considerable burden on the logistic system. The Army did not consider that the M88 had the full range of capabilities that the Armoured Regiment required. Indeed, the evaluation team suggested that the Army would have to give serious thought to three options: retain the Centurion ARV in service; develop a local ARV based on the M60; or purchase the Leopard ARV for use with the American tank. In contrast to these problems with the M60 ARV, the Leopard ARV (an example of which had taken part in the trials), was not only based on the gun tank’s chassis but was ‘highly regarded, both as a recovery vehicle and for its usefulness in tank maintenance’.

The Military Board concluded
that … ‘the Leopard is the
medium tank which best suits
the operational and training
needs of the Australian Army’

Although no firm cost estimates could be given, available estimates, based on the cost of the purchase of 100 gun tanks, suggested a unit cost of $A336,000 for the M60 and $A593,000 for the Leopard. Th e Military Board concluded that, while the M60 represented a great improvement on the Centurion, on balance, ‘the Leopard is the medium tank which best suits the
operational and training needs of the Australian Army’. In terms of a phased purchase, the Military Board strongly recommended that the Leopard tanks be bought in a single purchase, thereby averting the risk that the tank might be out of production when the second phase became necessary. The Army’s final evaluation report was completed in June 1973 and copies were passed to the Department of Defence in September of that year.

The Last Stand of the Centurion Tank

While trials on the German Leopard and American M60 tanks were being undertaken, the Army was also carrying out studies on the implications of extending the life of the British Centurion tank. The FSC had prompted these studies in October 1972, noting that the Israeli Army was in the process of re-engineering their Centurions for use in modern battle conditions. The FSC suggested that the frequency of Base Overhauls—reduced aft er July 1972—could be reviewed. Moreover, there was a possibility of acquiring spare parts by purchasing either cheap Centurions from disposed fleets or more of the tanks themselves from the United Kingdom.

The Army replied that the lead-time in acquiring spares from Britain was up to two years and would involve a quadrupling of the price of Centurion spare parts. The Army noted that purchasing tanks for purposes of cannibalisation was unlikely to provide adequate spares for Centurion. Australian armoured specialists also observed that the Israeli conversion of its Centurion fleet was untested in Australian conditions. Further, even if they were re-engineered, the Australian Centurions would require updating in areas such as armament, fire control and in suspension and cupola.

The necessary expertise for the conversion of the Centurions would have to be developed at the Army Design Establishment. The whole process would be expensive, would take a number of years to complete and, if unsuccessful, ran the risk of leaving the Army with a fleet of obsolete and unreliable tanks. In addition, the conversion process would impinge adversely on the capacity and direction of Australian Armoured Corps training and development.

Noting these concerns, the FSC requested in its Agendum 6/1973 of 20 February 1973 that the land force further explore the conversion option for incorporation in the next revision of MES 9. Th e committee even requested that the Army and the Department of Supply ‘should advise whether consideration has been given to the possibility of acquiring the jigs and tools from the UK to continue the manufacture of the Centurion, with a new engine, in Australia’.⁴ In response to this bizarre suggestion to reproduce an obsolete tank, the Army noted that it was unlikely to be feasible since the UK production line for the Centurion had been closed for some years. In addition, far more than the purchase of jigs was involved. There were, for example, unresolved problems of subcontract items alongside issues of component assembly and inspection.

At its meeting of 13 March 1973, the Force Structure Committee finally accepted the Army’s arguments that the use of the Centurion was incapable of being extended. The FSC then recommended to the DFDC that the option of extending the life of the Centurion be excluded—a recommendation that was accepted on 30 March 1973. The Army no doubt welcomed this decision since there was no rational reason why Army staff resources should have been wasted on examining ways and means of maintaining in service a tank design that was 30 years old, much less manufacturing it in Australia.

MES 9 Third Revise

The Army regarded the tank as
being fundamental to ground
combat due to the flexibility
conferred by its protection,
mobility and firepower.

The Third Revise of MES 9 was submitted to the Department of Defence on 31 August 1973. The document requested approval in 1974–75 for the purchase of 87 Leopard tanks and associated equipment at a net project cost of approximately $A58 million. Th e third submission drew on the conclusions of the Armoured Fighting Vehicle Study. In the wake of Vietnam, the primary purpose of the defence force had become defending mainland Australia, a task for which armoured forces would be required. Th e Army regarded the tank as being fundamental to ground combat due to the flexibility conferred by its protection, mobility and firepower. The beginning of a Defence of Australia policy in the 1970s had the effect of enhancing the requirement for acquiring a new medium-tank family for the period between 1975 and 1990.

The principle of the phased purchases remained in the third revision. In the first phase of acquisition—and in order to maintain the state of the armoured art—a purchase of 87 tanks would allow the re-equipment of the three tank squadrons, each less one troop (42 gun tanks and 12 support vehicles). Twenty-two tanks and support vehicles would be allocated to the Armoured Centre and the RAEME Training Centre while eleven gun tanks and four support vehicles would provide vital maintenance support. Th e second phase of acquisition called for the purchase of a further 40 vehicles required to bring the tank squadrons up to full strength, equip a field workshop, increase the maintenance support component and provide a war reserve.

The reduction in the number of tanks since the submission of the Second Revise in 1972 was due to the restrictions placed on Army numbers. The Armoured Fighting Vehicle Study had concluded that 99 tanks were required to maintain the state of the armoured art and to enable armour-infantry training. However, the Director of Army Development (DAD) advised the Army Development Committee (ADC) when submitting the draft of the Third Revise for approval that:

Following more recent ORBAT [Order of Battle] determinations related to [a] 34–36,000 man Army, structures in which a tank element of three squadrons each less one troop emerges, the number of tanks required for squadron service has been reduced to 87.No cost presentation was included with the third revised submission as updated costs were not readily available while the binding date for costs previously quoted had passed. The Department of Defence had asked the Leopard and M60 tank manufacturers for new estimates, which were to be forwarded when available.

The Resurgence of the M60 Tank

Once the new costings for both of the projected tanks and the future production proposals of each of the manufacturers became available, the Army’s initial support for the Leopard changed significantly. As a result, the Third Revise was withdrawn on 30 October 1973 for yet further revision. In a letter of 3 December 1973, the Special Deputy, Army Offi ce set out Army’s case.⁵ The Special Deputy considered that, although the Leopard was the superior tank, per se, there were several compensating advantages for the M60 that had led the Military Board to recommend its purchase over the German vehicle. These compensating advantages were that the M60 was much cheaper, was readily available from current production and had a favourable delivery rate. Th e M60 also off ered ready and prolonged availability of logistical support, using proven systems and procedures. Finally, there was a
greater expectation that M60s would be available should the Australian Army require further tanks.

Th e Army noted that production of the German Leopard tank was scheduled to conclude in 1974. Meanwhile, the M60A1 was in production but was scheduled for replacement by the greatly improved M60A3 tank, probably by 1976. The spare-parts situation would not allow the Centurion to remain in service much after mid-1976. In order to allow Australian tank training to continue unbroken, the land force estimated that the M60 tank would be in service by that time. The Army therefore proposed that M60A1 models be obtained immediately and then be upgraded to A3 standard retrospectively at a cost of $13,500 per gun tank after 1977.

The Army also recommended that Defence purchase the full number of 127 new tanks rather than use the phased arrangements that the DFDC had prescribed. To maintain the state of the armoured art required 99 tanks. Yet only 87 were considered sufficient as an interim establishment in view of reduced manning levels enforced by government policy. Neither strength, argued the Army, provided sufficient capability to meet training requirements or to sustain one squadron on operations.

The Army also pointed out that all American M60 production, including the upgraded A3, was scheduled to cease in January 1979. Th e final date for placement orders was September 1976. Therefore, should extra tanks be required aft er 1976, it might be possible to acquire rebuilt M60s from the United States—although they would be in the second half of their useful life. Their availability was not, in any case, guaranteed since the strategic circumstances that forced Australia to adopt a phase-two purchase might also apply to the United States. In such circumstances, Australia would conceivably be forced to purchase a different vehicle type. The land force warned:

A mixture of types of medium tanks within a single unit is entirely unacceptable from the point of view of the conduct of operations, although such a mixture might be acceptable across a range of armoured units in a large army.Consequently, re-equipping the Armoured Regiment with one type of tank could only be ensured if Defence ordered a total of 127 tanks in the period 1974–77.

In proposing a single purchase, the Army had moved out of step with the prevailing strategic assessments and corresponding policy guidance. The new Labor Government, which came to power in the election of 2 December 1972, had adopted a defence policy that had been foreshadowed by EFAMO. The immediate strategic environment was benign and, while not guaranteed, was likely to remain so for the next ten to fifteen years. Coincidentally, the life of the new tank was estimated to cover the period between 1975 and 1990. The DFDC had pronounced that a phase two purchase would only proceed if the strategic situation worsened. Hence, given the benign security outlook, it was most unlikely that a phase two purchase would be required before the anticipated end of the new tank’s service. While a single purchase by the Army of 127 tanks would allow the land force to deploy a squadron on operations, government policy was based on maintaining a ‘core force’ as the basis of development in order to maintain the state of the armoured art.⁶

MES 9 Third Revise Revised

In December 1973, another MES 9 revision contained the Army’s arguments as outlined above. Th e estimated cost of $A54.082 million in the document was based on an initial purchase of 87 M60A1 tanks. Th is acquisition was to be followed by obtaining a further 40 M60A3s upgraded tanks in the second purchase, along with product improvement kits to bring A1 gun tanks up to A3 standards. However, the paper suggested that the actual breakdown of purchase numbers would be contingent on how many A1s were needed urgently at the time of order. In the Army’s view, the fewer A1s that were bought the better, since if more A3s were acquired then there would be less upgrade costs.

In proposing a single purchase,
the Army had moved out of
step with the prevailing strategic
assessments and corresponding
policy guidance.

Concurrently with the drafting of the Third Revise, the Army began to address any drawbacks associated with the acquisition of the M60. Australian military staff in the Embassy in
Washington DC were asked to forward details of a product-improved M88 ARV, which might address initial Australian reservations over vehicle recovery. Defence specialists commenced further study on the feasibility of transporting the M60 in Australia. Yet, because the Army was so certain that the Government would accept its M60 proposal, senior Army offi cers
advised the military working group established to oversee the introduction of the new tank that any consideration of a Leopard acquisition should cease. Nonetheless, despite these views, the official Army and Defence position was that both the M60 and the Leopard suppliers should be treated equally until a fi nal decision on acquisition was made by Australia. This was a wise decision.

The Special Study Group on Armoured Fighting Vehicles—Interim Report

The Army’s position on tank acquisition was by no means accepted elsewhere in the Department of Defence. The FSC had established its own Special Study Group on Armoured Fighting Vehicles consisting of seven members from Defence Central. All except one, a Royal Australian Navy commander, were civilians. The study group provided an interim report to the FSC on
7 December 1973.

The Army’s position on tank
acquisition was by no means
accepted elsewhere in the
Department of Defence.

In that report the study group expressed immediate concern regarding the number of tanks to be purchased. Noting the current Army requirement for 127 tanks and the previous requirement for 99—subsequently reduced to 87 as proposed by the AFV Study—the study group announced that it was exploring the possibility of acquiring only 80 tanks. Th is number of vehicles would be used to equip three complete squadrons but would exclude 19 tanks allocated to the Armoured Centre. Armoured training would be carried out by one of the three squadrons. Th e study group also advised the Defence Department that it was considering the ramifications of reducing the number of tank squadrons from three to two.

The study group went on to question the Army’s proposal for a single large purchase immediately. Th e group noted that the closing date for M60 orders was January 1977 and expressed concern that an early purchase would result in a large number of M60A1s having to be retrofi tted to new A3 standards. While the Special Study Group acknowledged that the Centurion would leave service soon, it did not address a replacement tank type in the context of a later purchase date, nor did it acknowledge the Army’s intention to keep the M60A1 purchase as low as possible to avoid tank retrofitting. Finally, the Special Study Group stated that further work was required on the definitive costs and strategic mobility of the M60, and called for the addition of an Army and a Joint Intelligence Organisation representative to its membership.

The Numbers Game

In preparation for attendance at the FSC, the DAD briefed the Chief of Operations (C Ops) on the Army’s requirement for a single purchase of 127 tanks. As laid out in the Third Revise, a single purchase was preferable in order enable an armoured squadron to be deployed operationally. While it was not possible to estimate accurately the time period over which operations could be sustained, the War Reserve was calculated on the basis of having a squadron in operation for five months. Th e brief restated the Army’s intention to keep the number of M60A1 tanks as low as possible as during the phasing out of the Centurion.

… the Army warned that with
only 53 tanks, developing
logistic and tactical concepts
would be confi ned to academic
studies at the Armoured Centre.

The DAD brief warned that the 80-tank proposal by the Special Study Group was regarded as being ‘quite untenable’. The process of acquiring a lower number of vehicles overlooked the fact that the Armoured Centre required vehicles continually for various forms of training. If the tank squadrons were engaged in manoeuvres, they would not be available for Armoured Centre training purposes without risk of degrading the maintenance and development of tank capability in the field.

In the event, the study group revisited the 127, 99 and 87 tank number options as well as additional proposals for 53 and 76 tanks. Little progress was made except that group members
noted that less than 70 or 75 tanks (both numbers were used in the same paper) would mean a clear lack of capability for immediate operational deployment. In addition, training would be extremely constrained while any expansion of capability would be protracted. Th e group noted that strategic guidance prescribed the following:

The study group requested that the Army demonstrate the lead time required to bring a force of less than 99 tanks to the level of capability in strategic guidance.

On 18 December 1973, the Army produced a detailed summary of the capabilities of each quantity of tanks under consideration. It is worth examining this summary in some detail in the light of the 66 M1A1 gun tanks and M88 ARVs currently being procured at the beginning of a new century. Th e Army outlined the options open to the Defence Department as follows:⁷

An Acquisition of 53 Tanks

A $20 million purchase of 53 tanks would allow only one squadron of tanks to be fielded. Such a capability would allow the Armoured Centre to fulfil its charter while low-level crew and RAEME training could also be provided. However, such a number was clearly insufficient to allow combined arms training with infantry. Moreover, the development of practical mobile operations would not be possible since no tanks could be deployed on field exercises. As a result, the corps officer structure would be adversely affected. Finally, the Army warned that with only 53 tanks, developing logistic and tactical concepts would be confined to academic studies at the Armoured Centre.

An Acquistion of 75 Tanks

A $30 million purchase of this number of tanks would permit the fielding of two squadrons. Seventy-five tanks would also allow some tank-infantry training to occur, but barely enough to maintain the state of the art. Tanks might also be deployed in combined arms support exercises, but this activity would be at the expense of infantry-tank training and would be limited to a troop of four tanks. There would be no spare capacity for senior commanders and their staffs to practise the use of armour. The subsequent dilution of core skills expected would take five years to remedy.

An Acquisition of 87 Tanks

Estimated by the Army to cost $38 million, the acquisition of 87 tanks would allow three squadrons, each less a troop, to be deployed in the field. These restricted squadrons would be capable of undertaking some infantry-tank training, but not as effectively as that of a full squadron. Similarly, there would be restricted opportunities for higher-level training, the development of mobile-operations techniques in the defence of Australia and for RAEME activity in an operational setting. An 87-tank option provided only a ‘limited state of the art’ in armoured warfare that would be constrained by manpower restrictions. Overall, this number of tanks meant a serious reduction in effectiveness since the withdrawal of the tactical reserve of a squadron on operations would be a ‘fundamental tactical requirement’. Operational capability would be confined to a ‘one time’ effort that would ‘waste out’ the operational squadrons.

An Acquisition of 99 Tanks

Costed at approximately $42 million, this number of armoured platforms would, the Army believed, allow for the provision of three fully equipped squadrons. This number of tanks would allow the same training opportunities as the 87-tank restricted option, but without the severe limits imposed by the loss of one troop per squadron. Possessing nearly 100 tanks would confer a low-level capability to deploy a troop of tanks in a low-risk operational environment—such as a peacekeeping operation—but not for the traditional military operations needed to maintain the state of the armoured art.

An Acquisition of 127 Tanks

The Army estimated the cost of adopting this option at approximately $49.66m (later revised to $54.082m). Th is number of tanks provided all the capabilities of the 99-tank option outlined above but also allowed for the employment of one squadron in mid-intensity operations for some five months. If base repair facilities existed, such a squadron could be deployed indefinitely.

… the FSC noted that any
purchase by Australia of less
than 99 tanks would result in
the inability to field more than
a token force operationally

While the Army was engaged in drawing up its numbers options, the FSC had considered the interim report of the Special Study Group on 11 December 1973. The FSC endorsed the work of the Study Group, noting that, in seeking the full 127 tanks, the ‘Army is giving priority to the development of combat support covering armour and artillery, rather than infantry’. The committee also recorded the advice from the Army’s Chief of Operations that a squadron of the armoured regiment could not satisfactorily fulfil the requirements of the Armoured Centre.

Report of the Special Study Group on Armoured Fighting Vehicles

Th e Special Study Group believed that the argument that the Defence of Australia enhanced the requirement for medium tanks was ‘plausible’. However, in its report the group could make no ‘definitive judgment on this in the absence of a clear definition of the operational situations that could be encountered’. In its conclusions, the report restated the Army’s analysis of the capabilities and restrictions involved in purchases of different numbers of tanks. Finally, the report restated the arguments for and against the American M60 and the German Leopard tank types.

Adoption of the 87-Tank Option in 1974

On 5 March 1974, while considering the Study Group’s report, the FSC noted that any purchase by Australia of less than 99 tanks would result in the inability to field more than a token force operationally. If there were battle losses sustained, such force could not be supported for long without further degradation. However, the FSC again emphasised that strategic guidance indicated that no military threat to Australia was foreseen. Given these circumstances, the committee stated the following:

AFVs are required in order to allow the Army to maintain its capabilities to use armour and to keep up with developments in tactical doctrine covering the use of armour. The information supplied by Army indicates that this kind of capability could be maintained with 99 tanks, and in a more restricted way with 87, but not with fewerThe FSC noted the Army’s bid for 127 tanks but considered that, given prevailing strategic conditions, ‘an option larger than 87 tanks should not be contemplated at this stage’. Th e committee would make its final judgment on armoured vehicle numbers in the context of priorities of other defence items and program considerations. The FSC concluded that Defence could consider the possibility of a later purchase phase of 40 tanks beyond the projected program period.

On 6 March 1974, the DFDC considered the FSC’s report. By now the Army, including the Chief of the General Staff, Lieutenant General Sir Francis Hassett, accepted the inevitable result ‘that it was probably only realistic in view of the financial climate to limit procurement to 87 tanks in this program’. The DFDC decided that an initial purchase of 53 tanks followed by an additional 34 tanks would assist in the spread of expenditure and would probably reduce the number of vehicles requiring upgrades. Such a number was not, however, the total required by the Army, and in the longer term it would be important to develop skills in armoured warfare. In order for these skills to materialise, a fully equipped armoured regiment was required, consisting of 101 tanks (the 99-tank option plus two to give mobility to the regimental headquarters). The Defence Department had, by then, all but abandoned the concept of the deployable squadron and envisaged no increased use of armour in the Defence of Australia role. Rather, the Army’s Armoured Corps would seek to merely maintain state of the armoured warfare art on the basis of using restricted squadrons. Such an approach would maintain tank-infantry cooperation and retain a degree of capability for armoured warfare.

The committee agreed that the M60 was preferable to the Leopard in terms of cost and logistics but decided to pursue further detailed examination of contractual costs and the possibility of local industrial participation. Th is approach was reinforced by a Secretarial Note of 27 March 1974 that distributed a paper by the CGS dealing with the question of retaining Centurion support vehicles in order to provide support for the new gun tanks. The Secretary of Defence stated:

The paper prepared by the Chief of the General Staff assumes that the M60 is the preferred tank … the final decision on choice of tank is subject of further data to be provided on both M60 and Leopard on cost and the development of contractual position.The April 1974 Ministerial Announcement

In April 1974 the Minister for Defence, Lance Barnard, submitted a proposal to the Cabinet authorising a purchase of 53 medium tanks at a cost of $28m. The acquisition of a new tank was justified on the grounds of armour’s continued role in ground combat—a role validated by the 1973 Yom Kippur War in the Middle East. The Minister pointed out that the purchase of new tanks in limited numbers was intended only to maintain the state of the art in armoured warfare. Buying 53 tanks would allow the Army to re-equip one tank squadron, provide for maintenance support and re-equip the Armoured Centre. Such numbers would also permit ‘squadron crew’ training and some preparation in combined arms and mobile operations. The Ministerial proposal envisaged that Defence would seek a further decision from the Government in order to acquire additional tanks in 1976–77. However, the Army’s ability to deploy armour on operations would remain limited, and could only be undertaken at the expense of training efficiency.

The acquisition of a new
tank was justified on
the grounds of armour’s
continued role in ground
combat—a role validated by
the 1973 Yom Kippur War
in the Middle East.

The Minister’s submission was accepted by Cabinet and announced in Parliament on 9 April 1974. One can speculate that there must have been some concern in Army Office since the Minister’s statement conflated the capabilities of a larger tank force, which might be ordered in the future, with that of the 53 vehicles that would actually be purchased. However, in a later meeting with the Minister, the CGS was assured that the purchase of 53 tanks was only an initial acquisition and that ‘it was the Government’s intention to entertain the procurement of additional numbers of tanks in 1976–77.’

Tank Procurement Strategy

The Military Board considered that the M60 was the most suitable type for the Australian Army given its cheapness, sureness of supply and favourable logistic arrangements. The Army had good grounds to believe that the department would accept its expert advice in this matter. However, on 17 April, DFDC Secretarial Note No 43/1974 stated that, while the Government was aware that the M60 and Leopard had both been tested in Australia, the Cabinet had not made a final selection. A choice would depend on ‘negotiations, yet to be undertaken, with the two contending suppliers. Final orders defi ning prime equipment, maintenance support and training conditions, and Australian industry participation have still
to be negotiated.’

This statement was followed on 3 May 1974 by a Defence Department minute forwarding a suggested procurement strategy to the Army for consideration and giving the land force only a week to reply. Subsequently, a meeting took place between the Army and Defence Logistics Division on 10 May at which Army representatives argued that further negotiation on the Leopard tank was a waste of time and effort and was also probably unethical. With regard to a procurement strategy, the Army pointed out that it already had an established
project office to manage acquisition of the new tank.

The Military Board
considered that the M60
was the most suitable type
for the Australian Army
given its cheapness, sureness
of supply and favourable
logistic arrangements.

However, the Department of Defence was unmoved by the Army’s position. Indeed, in the Deputy Chief of Materiel’s opinion, Departmental officials were ‘obsessed’ with a negotiated procurement. Defence officials were following the Minister’s public statement that there would be negotiations, with off set production for Australian industry also considered. Moreover, officials pointed out that Italy was now producing Leopard tanks and might become an alternative source for these tanks if German production halted, as expected, at the end of 1974. In short, Defence officials believed that the Army would have to find the resources to comply with ministerial direction.

As a result, it was agreed between Defence officials and the Army that an approach should be made to the German manufacturers establishing whether the latter would provide a quote on a requirement for 53 tanks and whether a further 34 could be ordered by 1977. The Army undertook to provide Defence officials with details of its existing tank project office in order to establish a ‘determination of its [the project office’s] acceptability in terms of the requirements of the Procurement Strategy’. Defence Logistics would then revise the procurement strategy in the light of information that the Army had provided.

On 23 May 1974, a further meeting took place between the Army and representatives from Defence Logistics. Officials agreed that, while the M60 was the favoured choice of tank, following the Minister’s statement, it was necessary to obtain up-to date costs from both of the American and German competitors.

Conclusion

In May 1974, the files on the tank acquisition cease. However, the denouement was, of course, not the procurement of the M60 but of its rival, the Leopard. The Army’s defence of an M60 purchase appears strange when the accepted wisdom is that a Service that is modernising will always seek the most expensive item available. Yet, in the restricted budget environment of the immediate post-Vietnam years, the Army was also aware of other needs, particularly the requirement to replace its medium artillery and its anti-aircraft guns. Eventually, the Army purchased 101 Leopard tanks. Today, in the first decade of the 21st century, the Army is replacing its Leopards with a new armoured force consisting of 59 Abrams gun tanks and 7 ARVs.⁸

In 1974, in a benign strategic environment, many experts considered that, in order to maintain the state of the art in armoured warfare, the Army required a minimum of 87 tanks. Indeed, as we have seen, the Army considered that proposed purchases of 53 and 75 tanks respectively were not militarily viable—a view that the Special Study Group supported. Today, the purchase of a smaller number of Abrams MI tanks reflects a transition in warfare to an age of sensors and precision strike, in which materiel mass is considered to be less important than operational effect. To a non-specialist on armoured warfare or interested onlooker, it appears that the Abrams purchase reflects the coming of a new electronic battlespace that Army officers could only glimpse in 1974. It is a battlespace in which the capabilities of platforms, rather than their mere numbers, may determine operational
success or failure.

Endnotes

1. The files were acquired by the Army History Unit and comprise six files involved from Series A3688 and are numbered 367/R4/Parts A to H (Part B is missing). They are not in chronological order but they are contemporaneous.

2. The OMR was not promulgated until April 1972, but appeared to be sufficiently developed for its purpose at this time.

3. AWM121, Item 11/G/1.

4. A separated Department with functions similar to those of the Defence Materiel Organisation today.

5. Special Deputy was an interim title given to former permanent heads of the service departments after the abolition of those departments on 30 November 1973. The position was eventually abolished.

6. Ministerial Statement on Australian Defence, Parliamentary Debates, 22 August 1973.

7. All options include 19 tanks for the Armoured Centre.

8. Media Release by Senator the Hon. Robert Hill, No. 47/2004, M1 Chosen as Australian Army’s Replacement Tank, <http://www.minister.defence.gov.au/Hilltpl.cfm?CurrentId=3643>.

Posts · by **Yohannes** » Fri Apr 27, 2018 1:51 am

Electrothermal-Chemical Ignition Research on 120-mm Gun in Korea

[ Out of character: if anyone can spot the reference to K2 Black Panther MBT they get cookies for free ]

Journal page: IEEE Transactions on Magnetics; IEEE Transactions on Magnetics, Vol. 45, No. 1, January 2009

AUTHORS: Seong Ho Kim, Kyung Seung Yang, Young Hyun Lee, Jin Sung Kim, and Byung Ha Lee

Agency for Defense Development, Daejeon 305-150, Korea

Article history:
Manuscript received September 26, 2008. Current version published January 30, 2009.

Corresponding author: S. H. Kim (e-mail: shkim19@add.re.kr). Color versions of one or more of the figures in this paper are available online at http://ieeexplore.ieee.org.
Digital Object Identifier 10.1109/TMAG.2008.2008415

I. INTRODUCTION

ELECTROTHERMAL-CHEMICAL (ETC) propulsion was considered to be an attractive technology to control the pressure generated by chemical propellant in a combustion chamber. However, the required electric energies are so large as yet, and moreover, the control capability is not as efficient as that expected.

Completing basic researches of ETC technology using a 30-mm gun [1], the Agency for Defense Development (ADD), Korea, has been studying an electrothermal ignition using a 120-mm gun. The major difference between the previous and the current research is the role of electrical input energy which is now used only for the purpose of early ignition of propellant. The merit of ETC technology in the ignition is a control capability of the ignition power. The increase rate and the peak value of the chamber pressure vary with different ignition powers. Through an adjustment of the ignition power, a compensation for the variation in charge temperature is possible [2].

In the research of electrothermal ignition, properties of the capillary plasma injector, the burning of propellant in the closed vessel, and the gun-chamber simulator have been studied. The capillary plasma injector was designed to transfer electrical energies around 100 kJ. The purpose of the closed-vessel experiments was to investigate the burning effect of the capillary plasma on various propellant compositions. The gun-chamber simulator with a transparent acrylic tube was used to see the early phase of ignition in detail before real firing experiments using the 120-mm gun.

In this paper, the progress of studies on the capillary plasma injector, the closed vessel, the gun-chamber simulator, and firing tests using the 120-mm gun is described.

II. CAPILLARY PLASMA INJECTOR

The capillary plasma injector used in the research of 30-mm ETC gun was 15 mm long and 5 mm in diameter. In spite of delivering large amount of electrical energies, there was no severe structural problem except for the erosion of electrodes. In the research of 120-mm ETC gun, the plasma injector was designed to fit the threads of conventional primer, and the capillary was lengthened about two times larger than that used previously. With the increased capillary length, occasional failures have occurred in the firing experiments. These were mainly due to an increased pressure inside the capillary and the combustion chamber. Thus, both of the end electrodes of the capillary were designed more robustly.

The pulse forming network used in this paper consists of two 300-kJ capacitor bank modules connected in parallel to make a pulse longer than 1 ms by a delayed triggering [3]. Each module has a capacitance of 1236

, an inductance from 20 to 160

, and circuit resistances below 10

. A triggered vacuum switch and crowbar diodes were installed in each module. By adjusting charging voltages and triggering times of the individual modules, a desired current waveform is obtained. It delivers a current below 100 kA in 1 or 2 ms.

In order to verify the stability of discharge operations in the region of energy transfer above 100 kJ, various plastic liner materials have been tested. The capillaries were 31 mm in length and 5 mm in inner diameter. The inductances of the first and the second module were 80 and 40

, respectively. The triggering time of the second module was 650

. The discharge experiments were conducted in the open air. Table I shows the test results using polyethylene (PE) and polyoxymethylene (POM) capillaries without failures. Under the same discharge condition, POM capillary showed larger resistance and energy transfer than those of PE capillary with very smooth ablation surface.

Table 1: Capillary discharge tests

The failures have arisen mainly from the leakage of highpressure plasma in the closed-end side of the capillary or in the plastic–metal interface region near the nozzle exit. In the
case of failures, PE capillaries were extruded or bulged out of the nozzle. Sometimes, POM capillaries were cracked into fine fragments. Thus, further studies are required to choose POM material for a reliable liner.

Fig. 1 shows a picture taken at 484 after the discharge of POM capillary delivering an electrical energy of 175 kJ. The length of the luminescent plasma jet was 50 cm long. Currently, experiments using PE capillary in the region of energy transfer around 100 kJ have been done reliably without further difficulties.

Fig. 1. Discharge of POM capillary in the open air. Exposure: 2

; time after discharge: 484

; and total transferred energy: 175 kJ.

When a capillary plasma injector is used for ETC ignition, it needs to identify the electrical input information and to control the delivery of electrical energy to the combustion chamber. Not only the parameters of the pulse forming network but also the plasma resistance values are required to estimate the electrical input.

For a simple and fast estimation, a resistance model modified from that derived in [4] was useful in the estimation of electrical properties. In the case of controlling only the electrical parameters under the same capillary geometry, the resistance values measured from experiments were proportional to the values obtained from raising the current values to an empirically fitted power factor. Although the fitted values of proportional constant and power factor are useful to estimate the electrical behaviors in advance in the experiments, it is difficult to extract their physical properties.

In order to calculate the properties of capillary plasma without any fitting formula, 0-D fluid equations were solved referring to [5]. Considering the nonideal property of electrothermal plasma, a modified plasma resistivity was used [6]. The thermodynamic quantities and plasma properties needed in the calculation were obtained from Saha equations assuming a local thermodynamic equilibrium condition. The ablation model described in [4] was adopted, in which the black body radiation is a primary source of wall ablation and the stagnation enthalpy of plasma is chosen as the ablation energy.

To compare with calculations, capillary discharge experiments in the open air have been done. With different charging voltages in each bank module and a delayed triggering of 650

, double pulses of about 2 ms were applied. The measured and calculated signals are shown in Fig. 2, where the values in the first peaks agree well. However, in the second pulse, the calculated signals increasingly deviate from the measured values. Although the later part of the measured current could be lowered due to the limitation of current transformer used in this experiment for a large current, it is clear that the calculations do not estimate the measured values well in the viewpoint of the voltage signals. The erratic behaviors of the measured voltages in the later parts also implicate nonuniform obstacles such as debris.

Fig. 2. Measured and calculated electrical signals of the capillary discharges using different charging voltages of (square) 8.5/5.5 kV, (circle) 12/6 kV, and (triangle) 16/6.8 kV with a triggering delay of 650

. The capillaries used are 24.3 mm long and 5 mm in diameter. (a) Current (LC). (b) Voltage (LV).

In the comparison between PE and POM capillaries, the calculations showed almost the same estimations for both of the materials in the same discharge condition. It is because the thermodynamic properties per particle in the plasma consisting of the atomic and ionic elements of C, H, O, and N show no much difference in the temperature region of a few electronvolts, and the ablation rate in the model of [4] is controlled and limited by the plasma enthalpy irrespective of the prepared materials in the vapor layer.

However, the fact that the experimental result using POM capillary showed large resistance values than those expected implies a lower temperature of POM plasma compared to that of PE plasma. Although Loeb and Kaplan’s ablation model estimates the electrical behavior well for PE plasma without an introduction of uncertain factors such as radiation fraction or sublimation energy, the experiment using POM capillary reveals the necessity of detailed studies on the ablation model reflecting the material properties.

III. CLOSED-VESSEL EXPERIMENTS

There have been diverse explanations on the interaction between plasma and propellant in the ETC ignition [7], [8]. In this paper, a closed vessel shown in Fig. 3 was designed to see the burning properties of ETC ignition using several kinds of propellant compositions.

Fig. 3. Closed vessel for ETC ignition.

The vessel is similar to a conventional one except for including a connection line to supply a high-power electric pulse to a capillary plasma injector inside it. It has an effective volume of 236 cm³ and can be operated near 140 MPa. The capillary used in the vessel is 40 mm long and 5 mm in diameter. The ignition starts with a vaporization of a copper wire of 0.1 mm in diameter inside the capillary. The electric energy can be supplied up to 30 kJ within 1 ms.

The tested propellant compositions were JA2 and LOVA. Opaque JA2 propellant containing graphite and transparent JA2 propellant without graphite were used. JA2 propellant of cylindrical shape and without any perforations was used. LOVA propellant has a hexahedral shape with 19 perforations. The quantities used in the tests were about 23 g with a loading density of 0.1 g/cm³. Electric pulses of 1.4–17 kJ were supplied. Conventional ignitions were also done in the same vessel for comparison with ETC ignitions. The ignition starts from supplying a heating current into the nichrome wire surrounding priming powders.

The burning rates were analyzed referring to the methods described in [9]. In the case of ETC ignitions, electric pulses of 0.5 ms in width were applied with different charging voltages.

Fig. 4 shows pressure curves of opaque JA2 propellant in time, from which burning rates are calculated. The ignition delay time is defined as the time interval between the start of
electric current and the time to reach 10% of peak pressure. Conventional ignitions in these experiments showed delay times of about 83 ms for all kinds of propellants. All ETC ignitions showed short delay times as the transferred energies increased. Fig. 5 shows the ignition delay times versus transferred energy. Both opaque and transparent JA2 showed similar delay times. LOVA required more energy than JA2 for an earlier ignition.

Fig. 4. Pressure versus time of opaque JA2 propellant; and Fig. 5. Ignition delay time of propellants.

Short delay times can be explained by an increase of the initial burning surface which is provided by the interaction of propellants with the injected plasma. The fact that opaque and transparent JA2 propellants showed no difference in the delay time means that the small mixed graphite does not affect on the surface at early ignition times.

Fig. 6 shows the calculated burning rates for the experiments shown in Fig. 4. Below the transferred energy of 7 kJ, the burning rates were similar to that of the conventional ignition
up to the pressure of 70 MPa. As the electric energy increased further, the rates around the pressure of 30 MPa increased. Above the pressure of 70 MPa, the rates in ETC ignition were all lower than those of the conventional one.

Fig. 6. Burning rate of opaque JA2 propellant.

Fig. 7 shows the result using transparent JA2 propellant. Up to the ignition energy of 7 kJ, the burning rates did not deviate much from those of the conventional ignition. In the case of 13 kJ, the rate increased severely in the all-pressure region.

Fig. 7. Burning rate of transparent JA2 propellant.

Fig. 8 shows the burning rates of LOVA propellant, which were lower than those of JA2. The rates were almost the same as those of the conventional ignition in the energy up to 16.7 kJ. ETC ignition did not show an appreciable effect on the burning rates of LOVA.

Fig. 8. Burning rate of LOVA propellant.

Since the electrical ignition energies are transferred early within 0.5 ms and have no direct effects on the propellant after 0.5 ms, the burning properties reflect that the inside region of the propellant is modified by the interaction with plasma during the existence of the electric pulse. In the case of ignition energy of 13 kJ, the high burning rate of the transparent JA2 in the all-pressure region shows that the propellant was affected at full depth below its surface. This tendency also appears a little near the low-pressure region in the case of 7 kJ. In the opaque JA2 propellant showing a similar tendency near the low-pressure region, the interaction region is just below its initial surface, which can be enlarged as the energy increases. From these results, it can be concluded that the penetrated plasma radiation during the existence of the electric pulse plays a role in the increase of the burning rates. Since LOVA is opaque inherently, the plasma radiation has no effect on the modification inside the propellant in spite of the large ignition energy. Since these experiments have not been done repeatedly for the same energy value, further experiments are required to verify the reproducibility.

IV. EXPERIMENTS USING GUN-CHAMBER SIMULATOR

A gun-chamber simulator using an acrylic combustion chamber was fabricated to study the early ignition stage as shown in Fig. 9. It has the same structure as the real 120-mm
gun chamber except for an acrylic chamber wall withstanding up to 30 MPa. The capillary plasma injector is positioned along the central axis of the transparent acrylic chamber. The chamber with a volume of 10 L can be filled with JA2 propellant up to 8.4 kg. An electrical pulse is supplied from two capacitor bank modules through a coaxial cable.

Fig. 9. View of gun-chamber simulator.

For the purpose of fast propagation of plasma flame, a hollow tube was attached to the capillary plasma injector. Two different kinds of tubes were tested. The geometry of the metal tube is the same as the conventional ignition primer as shown in Fig. 9. The idea of introducing a combustible tube was obtained from the low efficiency of metal tube. It was made of the same material as the combustible cartridge case of 120-mm gun charge. It is in the shape of cylindrical tube with many holes on its surface, and the internal volume is maintained to be the same as that of the metal tube. Fig. 10 shows the chamber pressures measured when the plasma injectors were discharged in the vacant chamber without
propellant. At a transferred energy of about 45 kJ, the pressure of the chamber using combustible tube increased about four times faster than that of the metal tube.

Fig. 10. Comparison of pressure increase for metal and combustible tubes without propellant.

Fig. 11 shows the combustion of JA2 propellant inside the gun-chamber simulator just when the acrylic tube begins to be broken.

Fig. 11. ETC ignition of JA2 propellant.

Fig. 12 shows the chamber pressures when the plasma injectors were discharged in the chamber filled with JA2 propellant of 8.4 kg. At a transferred energy of about 45 kJ, the time to reach peak pressure in the experiment using the combustible tube was about 1.52 ms as short as the half of that in the experiment using the metal tube. As a result of these experiments, it seems that the combustible tube is more effective in enhancing the increase rate of pressure than that of the metal tube.

Fig. 12. Comparison of pressure increase for metal and combustible tubes with JA2 propellant.

Through tests using the gun-chamber simulator, the pressure profile and the delay time between the rising pressures of the breech and the projectile base have been studied. From this, an appropriate electrical energy and propellant compositions have been selected for real firing experiments.

V. 120-MM GUN FIRINGS

A 120-mm L55 barrel with a chamber volume of 10 L was assembled with a vertically sliding breech mechanism. The gun before a firing experiment is shown in Fig. 13, where the breech block is connected to coaxial cables to supply electric pulse energy.

Fig. 13. 120-mm gun for firing experiments.

Up to now, several tens of rounds of firings have been performed with JA2 propellant. In Fig. 14, the breech pressures and the transferred electric powers at three different charging voltages are shown. For each case, a double pulse of about 2 ms in duration was applied with a time interval of 0.65 ms. Projectiles of 8.3 kg were fired with JA2 propellant of 8.4 kg using the capillary plasma injector attached with a combustible tube. At a transferred energy of 46 kJ, the maximum breech pressure was about 655 MPa, and the muzzle velocity of the projectile reached 1755 m/s. At a transferred energy of 82 kJ, the maximum breech pressure was about 724 MPa, and the muzzle velocity reached 1792 m/s. The maximum breech pressure and the muzzle velocity increased as the transferred energy did. In this experiment, the duration of the electric pulse was fixed so that the electrical power also increased as the charging voltages were raised. It showed a possibility of control on the peak pressure by changing ETC ignition conditions.

Fig. 14. Result of full-scale firing experiments.

VI. SUMMARY AND CONCLUSION

The ETC research of ADD, Korea, has been concentrating on the plasma ignition applicable to a 120-mm gun. The ignition of propellant has been done by a capillary plasma injector which can be operated near an energy transfer of 100 kJ. Electrical properties were estimated with the help of plasma theories and compared with experimental results. The relatively large resistance values in the later part of the discharge and the large resistivity of POM capillary require more studies in the ablation process.

From closed-vessel experiments, it was concluded that plasma radiation plays an important role in the interaction with JA2 propellant. In the case of LOVA propellant, more increased energy should be supplied for a fast ignition.

Before real firings, a gun-chamber simulator of an acrylic tube was utilized to verify the early ignition stage using the designed plasma injectors. The combustible tube attached to the injector showed a superior ignition properties compared to the conventional metal tube.

A variety of firing experiments have been performed using a 120-mm gun. A possibility to change the muzzle energy by a control of ETC ignition has been verified, and it needs to improve and optimize the ignition processes. ADD, Korea, continues the research to enhance the efficiency of the plasma injector, to compact the pulse forming network, and to investigate the electrothermal ignition properties of LOVA propellant.

REFERENCES

[1] J. W. Jung, S. H. Kim, and K. S. Yang, “Overview of ETC research in Korea,” IEEE Trans. Magn., vol. 39, no. 1, pp. 22–23, Jan. 2003.
[2] T. H. G. G. Weise, J. Kruse, P. Schaffers, and H.-K. Haak, “Status and results of the German R&D program on ETC technologies,” IEEE Trans. Magn., vol. 37, no. 1, pp. 46–51, Jan. 2001.
[3] Y. S. Jin, H. S. Lee, J. S. Kim, G. H. Rim, J. S. Kim, Y. H. Lee, K. S. Yang, J. W. Jung, and H. J. Moon, “Performance of 2.4-MJ pulsed power system for electrothermal-chemical gun application,” IEEE Trans. Magn., vol. 39, no. 1, pp. 235–238, Jan. 2003.
[4] Loeb and Z. Kaplan, “A theoretical model for the physical processes in the confined highpressure discharges of electrothermal launchers,” IEEE Trans. Magn., vol. 25, no. 1, pp. 342–346, Jan. 1989.
[5] R. B. Mohanti and J. G. Gilligan, “Time dependent simulation of the plasma discharge in an electrothermal launcher,” IEEE Trans. Magn., vol. 29, no. 1, pp. 585–590, Jan. 1993.
[6] R. J. Zollweg and R. W. Liebermann, “Electrical conductivity of nonideal plasmas,” J. Appl. Phys., vol. 62, no. 9, pp. 3621–3627, Nov. 1987.
[7] M. J. Taylor, “Energy transfer mechanisms involved in plasma ignition of solid propellants,” in Proc. 20th Int. Symp. Ballistics, Orlando, FL, September 23–27, 2002.
[8] A. Koleczko, W. Ehrhardt, H. Schmid, S. Kelzenberg, and N. Eisenreich, “Plasma ignition and combustion,” in Proc. 19th Int. Symp. Ballistics, Interlaken, Switzerland, May 7–11, 2001.
[9] W. F. Oberle and D. E. Kooker, BRLCB: A closed-chamber data analysis program. Part 1. Theory and user’s manual Army Res. Lab., Aberdeen Proving Ground, MD, ARL-TR-36, 1993.

Posts · by **Yohannes** » Fri Apr 27, 2018 3:40 pm

Posts · by **Yohannes** » Fri Apr 27, 2018 4:33 pm

Future Armour Materials and Technologies for Combat Platforms

Journal page: Defence Science Journal, Vol. 67, No. 4, July 2017, pp. 412-419, DOI : 10.14429/dsj.67.11468
@ 2017, DESIDOC

AUTHORS: B. Bhav Singh^*, G. Sukumar, P. Ponguru Senthil, P.K. Jena, P.R.S. Reddy, K. Siva Kumar, V. Madhu, and G.M. Reddy

*E-mail: bhav_singh@dmrl.drdo.in
Defence Metallurgical Research Laboratory, Hyderabad – 500 058, India

Article history:
Received : 22 February 2017, Revised : 10 May 2017
Accepted : 18 May 2017, Online published : 03 July 2017

Reproduced with permission of copyright owner. Further reproduction prohibited without permission.

Abstract

The ultimate goal of armour research is to create better armour for battle worthy combat plat forms such as main battle tanks, infantry combat vehicles and light combat vehicles. In each of these applications, the main aim boils down to one of the two things; either reduce the weight without sacrificing protection or enhance the performance at same or even reduced weight. In practice, these ambitions can be fulfilled only if we have with us, appropriate improved armour materials, advanced and innovative technologies and also improved designs, which enable us to use them for creating next generation armour modules. Armour systems have progressed through improvements in metallic, ceramic and lightweight (low areal density) composite materials. Similarly, the advances in development of explosive reactive armour (ERA) and non-explosive reactive armour (NERA) have generated efficient armour system against contemporary high explosive antitank ammunition and missile threats for the armoured vehicles. Yet, to achieve armour performance exceeding that of the current light combat vehicles and main battle tanks, further advancements in armour materials, systems, and survivability technologies are required for new vehicular systems that weigh significantly less than the present combat platforms. Various approaches and advancements in the metallic and composite armour materials, ERA and NERA systems to improve the survivability of armoured vehicles in the futuristic multi-spectral battlefield scenarios are described.

1. Introduction

During the past several years there has been many fold increase in the threat level and the armour materials have, indeed undergone great changes to meet the challenge. There is a need to continue to develop materials and modules which can withstand all futuristic threats. This calls for novel concepts in design and testing methods for optimisation of armour even ahead of the ammunition, to create systems that are protected well. The weight of armour in combat vehicles has always been constrained by the overall weight of the vehicle and the power-to-weight ratio. Changes in the type of threats in recent years have led to a shift in focus on the need for protection against multi-spectral threats. Enormous efforts are being put world over on the development of armour materials and systems to provide greater ballistic protection with minimum weight penalty. For providing such a protection, it is essential to create high performance passive, reactive, dynamic and active armour technologies with creative armour design concepts. Today, no single material is capable of effectively defeating wide range of threats, and hence, a wide variety of armours have to be developed. The most important element of survivability is armour protection. In the beginning, battle tanks were made solely of steels. In recent years the situation has changed with the emergence of excellent armour materials.

The various types of candidate materials and systems for armour applications namely advance materials like steel armour, polymer matrix composite armour, laminated composite armour, explosive reactive armour, non explosive reactive armour for different types of threats are discussed.

2. Rolled Homogeneous Armour Steel (Spade steel)

Rolled homogeneous armour (RHA) steel (Table 1) has remained the standard armour world over on most of the tanks. Its low cost, reliability, availability of production infrastructure, concurrent utility as a structural material and its ease of fabrication have enabled this steel to hold on to its prime position. This steel armour continues to be used in the tempered martensitic microstructure after heat treatment which involves hardening to increase its resistance to penetration by projectiles and then tempering to make it tougher and therefore enhance the energy absorbing capability against impacting projectiles. Also, intense research in ferrous metallurgy has led to greater improvements in the ballistic performance of the steel. Ability to increase its hardness while maintaining adequate toughness has been the key to this success for achieving its high performance. These wonderful advancements have been achieved through micro-alloying, inclusion shape control, and thermo-mechanical processing and grain refinement.

Table 1. Chemical composition of RHA steel

3. Medium hardness Armour Steel

The development of medium hardness steel essentially focused on selection of suitable heat treatment cycles on rolled homogeneous armour (RHA) steel (Table 1) in order to obtain medium hardness with improved ballistic properties and without any cracking tendencies. This approach has advantage of using existing steel and existing production infrastructure making scale up to industrial level practicable. The activities involved are the optimisation of heat treatment on RHA steel in order to get medium hardness, subsequent mechanical property and microstructural evaluation, and ballistic evaluation against small arms and large caliber ammunition. Presently RHA steel is used for the manufacturing of structural parts of battle tanks in India and it has a hardness value of around 300 VHN. This steel was made by Steel Authority of India Limited and supplied in the form of rolled plates. In this work, objective was to enhance the hardness of RHA steel to about 400-450 VHN by employing suitable heat treatment procedures. The targeted hardness is nearly 50 per cent higher than the existing RHA steel being used. While increase in hardness and strength of steel results in improved ballistic properties, it is generally accompanied by a loss in impact properties and weldability^1-2. Weldability is an important issue since fabrication of structures, armour modules, etc. employ welding extensively for joining. Thus, it becomes essential to optimise processing parameters specially heat treatment to achieve higher hardness and strength without significant loss in weldability.

3.1 Mechanical Properties

By varying the tempering temperature of RHA steel, a wide range of mechanical properties are achieved. The yield strength and tensile strengths varied in the ranges of 1146 MPa - 1463 MPa and 1247 MPa - 1900 MPa, respectively. The hardness of the steel varies between 381 VHN - 586 VHN. The charpy impact energy varies in the range of 19J - 85 J depending on tempering temperature. Based on the hardness and CVN energy results, two tempering temperatures 450^OC and 500^OC are selected which produced a hardness of around 450 VHN coupled with good impact toughness. The mechanical properties of the steel in the as-received condition are compared with the mechanical properties of the steel in the modified tempering condition in Table 2. It can be observed that there is a considerable increase in the yield strength and ultimate tensile strength of the steel at the cost of little ductility in comparison to the presently used RHA steel by adopting modified tempering conditions. The hardness also shows a substantial rise at both the tempering conditions than the presently used steel hardness.

Table 2. Mechanical properties of RHA Steel at different tempering temperatures

3.2 Weldability

Weldability of the modified heat treated RHA steel was evaluated by Tekken tests and results have been found to be satisfactory up to 25 mm thickness in 450^OC tempered condition and up to 80 mm thickness in steel plates tempered at 500^OC. Figure 1 displays the 80 mm thick Tekken test specimen after the welding process. The Tekken test specimens were cut into slices across the thickness for further observation. Figure 2 illustrates the microstructures of 500^OC tempered welded samples of 80 mm thickness plate. No visual cracks were observed. Local critical stress was found to be about 560MPa for 500^OC tempered welded specimens as shown in Fig. 3.

Figure 1. 80 mm Tekken specimen after welding.

Figure 2. Microstructure of 500^OC tempered 80 mm thick Tekken specimens (a) Base metal, (b) Base metal – weldments interface, and (c) HAZ.

Figure 3. Stress-time curve for 500^OC tempering condition.

3.3 Ballistic Performance

The medium hardness steel plates tempered at 500^OC were ballistically tested against 125 mm FSAPDS ammunition (Fig. 4). The large caliber ballistic test results are as shown in Fig. 5 and it can be observed that ballistic performance of the 500^OC tempered plates is approximately 10 per cent - 15 per cent better than the presently used RHA steel plates. Against armour piercing projectiles such as 7.62 AP and 12.7 AP, medium hardness plates showed about 20 per cent - 25 per cent improvement in ballistic performance measured in terms of depth of penetration.

Figure 4. 500^OC tempered steel plates tested against 125 mm FSAPDS ammunition.

Figure 5. Large caliber ballistic evaluation results.

4. High hardness Armour Steel: DMR1700 Steel

For a given impact velocity, the DOP can be reduced by increasing the hardness and strength of the armour steel. By using high hardness steels, the protection can be increased considerably for a given weight of armour or the weight of the armour can be reduced for a given threat. High hardness steels can be used as components of armour modules fitted in the tanks or can be used as add-on armour for battle tanks. The high hardness steels are therefore going to play an important role in the design of fighting vehicles for improved performance against the kinetic energy threats.

DMR-1700 steel (Table 3) is a medium carbon low alloy high hardness steel developed by DMRL. This steel has shown promising results for armour application due to its high strength and hardness values. Figure 6 shows the microstructures of the steel plates after heat treatment. Tempered martensitic microstructure is observed in all the plates up to 50 mm thickness. Table 4 displays the comparison of mechanical properties of DMR-1700 steel plates and RHA steel. The strength and hardness of DMR-1700 steel are significantly higher in comparison to RHA.

Figure 6. Optical microstructure of 50 mm DMR 1700 steel plate.

Table 3. Chemical composition of DMR-1700 steel (wt%)

Table 4. Comparison of mechanical properties of DMR-1700 plates with RHA steel

4.1 Ballistic Performance

The ballistic performance of high hardness steel (DMR1700 steel) has been evaluated against various small arms ammunition as well as against large caliber ammunitions. DMR 1700 steel shows much improved ballistic performance in comparison to RHA. This can be attributed to the high strength and hardness of DMR 1700 steel in comparison to RHA. From the ballistic test it is seen that DMR-1700 steel exhibits improved ballistic performance of about 25 per cent against 7.62 AP ammunition and 20 per cent against long rod kinetic energy projectiles (125FSAPDS) (Fig. 7) as compared to RHA steel.

Fig. 7 Ballistic performance comparison of DMR 1700 steel and RHA steel.

5. High Nitrogen Steel

There has always been a demand to reduce the weight of armoured structures used in various ballistic applications. Weight reduction can be achieved by using newer materials having better ballistic performance or by changing the design of armour systems. Traditionally, high strength low alloy steel with tempered martensitic microstructure has been widely used in various armour applications. Apart from high strength low alloy steels, studies have also been carried out on other class of steels such as secondary hardening steels, auto tempered steels, maraging steels, bainitic steels and nickel free high nitrogen steels (HNS) for potential ballistic applications.

Defence Metallurgical Research Laboratory (DMRL), Hyderabad has recently developed high nitrogen steel (Table 5) for armour applications. High nitrogen steel has been produced industrially at M/s. Jindal stainless (Hisar) limited, Hisar, Haryana, India. The steel plates have been evaluated against various small arms ammunition as well as against large caliber ammunitions. From the ballistic test it is seen that HNS exhibits improved ballistic performance of about 35 per cent against soft projectiles, 10 per cent against armour piercing projectiles and similar performance against long rod kinetic energy projectiles (125FSAPDS) as compared to RHA steel³.

Table 5. Chemical composition of high nitrogen steel (wt%)

6. Glass Composites Laminates

Fighting vehicles comprising light weight armour made up of ceramic and polymer composite possess better mobility, fighting ability and fire power as compared to the vehicle with traditional steel armour. From the past two decades glass, aramid and ultra high molecular weight polyethylene (UHMWPE) fibre reinforced composites have gained considerable importance for structural and add-on armour applications due to their high specific strength and high energy absorption under dynamic loads^4-6. Effect of type of fibre reinforcements such as glass, aramid, carbon and ultra high molecular weight polyethylene (UHMWPE) on ballistic performance was reported. Energy absorption mechanisms
of different fiber reinforced composites have been addressed by various researchers through analytical, numerical and experimental studies^7-10. Though impact performance studies on E-glass/epoxy composites are extensively carried out by various researchers, there is a scanty information available on comparative performance of E-glass/epoxy and E-glass/ phenolic composites under high velocity impact especially against 125 mm FSAPDS. Therefore there is need to carry out comparative study on effect of matrix on ballistic performance
of glass composite laminates against long rod penetrators.

6.1 Ballistic Evaluation

Two different composite laminates namely E-glass/epoxy and E-glass / phenolic laminates were subjected to ballistic evaluation against 125 mm FSAPDS ammunition. Performance of the laminates is compared by measuring the depth of penetration in backing steel plates. Ballistic evaluation was carried out at PXE, Balasore range using 125 mm FSAPDS ammunition with tungston heavy alloy (WHA) penetrator. The penetrator was fired from smooth bore gun of T-72 tank at a impact velocity of 1650±30 m/s. Residual depth of penetration in backing RHA plates was measured. Schematic diagram for measurement of residual depth of penetration in RHA plates is as shown in Fig. 8. Thickness and mass efficiency of the composites were calculated and compared with RHA. Due to the limitations on availability of material only limited no of tests were performed.

Figure 8. DOP test configuration.

Mass and thickness efficiencies for the both laminates have been calculated and as shown in Table 6. From the data, it is observed that both the laminates show approximately similar performance against FSAPDS ammunition with limited ballistic data, phenolic shows slightly better performance but this need to be verified with more number of experiments. Composites laminates shows better mass efficiency compared to RHA but thickness efficiency is lesser than the RHA. The data suggests that it is required to use appropriate thickness of composites to get the optimum efficiency with respect to mass and thickness of armour modules.

Table 6. Mass efficiency and thickness efficiency of different glass composites against 125 mm FSAPDS

7. Kanchan Composite Armour

There has been a significant development in the penetration capabilities of kinetic energy (KE) projectiles starting from 1950 to 2000. This development of KE rounds led to drastic increase in the penetration of armour (RHA)¹¹. Subsequent design of newer weapons such as shape charge warheads also led to huge increase in penetration in RHA steel. These advances led to the development of improved armour materials as also the designs, since more RHA is needed to provide the increased requirements of protection.

It is now well-known that no single material can provide protection against all types of ammunition which work on quite different principles. Therefore, different materials are combined optimally in the form of composite armour to provide effective immunity. The first composite armour used by British was Chobham armour which consisted of a layer of ceramic between two plates of steel armour¹². Today, all modern tanks invariably use composite armour in many locations. The performance of composite armour materials can be further improved by choosing the right design. The stacking sequence of the layers is also an important factor in improving its ballistic performance. Composite armour gives protection against most of the ammunitions, i.e. K.E and shaped charge ammunitions.

The kinetic energy projectiles deliver momentum on the target and thereby destroy the armour, whereas, the chemical energy weapons produces high velocity jets and cause damage to the armour. Long rod kinetic energy projectiles are used to defeat thick armour plates used in main battle tanks¹³. Hence design of armour should be made such that it can reduce the penetration of the projectile by defeating the same. This
reduction in penetration can be achieved by appropriate designing of armour materials. Hence, selection of materials is very essential for design aspect. The properties of the materials should be such that it can blunt the projectile, should dissipate the shock waves generated during impact loading and to absorb the energy of the projectiles by undergoing severe plastic deformation. For blunting of projectile, a hard material such as high hardness steel or ceramic can be used at the front. However ceramic or high hardness materials cannot be used at the front due to their brittle behaviour and thereby losing the integrity of the laminate structure. Rolled homogeneous armour is considered to be a suitable candidate to face the initial impact. DMRL had designed and developed new Kanchan armour modules for improved ballistic performance against large caliber ammunitions by optimising the armour structures for improved protection which can absorb lot of impact energy.

8. Materials For Explosive Reactive Armour

Explosive reactive armour (ERA) was developed to defeat shaped charge warheads and found to effectively reduce the penetration. Explosive sheets were sandwiched between two metallic plates which when exposed will be initiated by the impact of the shaped charge. The moving metal plates interact with the jet and cause damage to the jet¹⁴.

The explosive reactive armour contains an explosive layer in between two metal plates. The functioning of explosive reactive armour is governed by two important mechanisms, Jet perturbation and metal cutting effect of the jet¹⁵.

When a shaped charge jet strike the ERA its explosive is initiated within microseconds. Since the detonation products are confined by two metallic plates, they attain velocity and density comparable to that of the jet. These detonation products collide with the incoming jet and the transverse impact of detonation products makes the jet lose its linearity and coherency and there by its penetration power comes down. This mechanism is effective only during a fraction of the total life of the jet. The second mechanism, metal cutting effect is caused by the interaction of the moving plate with the jet. The plate moves laterally with respect to the jet direction exposing new surface to the incoming jet, which causes the consumption of the jet¹⁶.

Defeat of long rod penetrators by explosive reactive armour occurs by breaking of the projectile by the moving plates. Breaking of the projectiles needs high strength flyer plates. The lateral movement of the plate with respect to the projectile consumes the projectile and reduces the penetration power. Defeat of long rod projectile needs not only high strength but also an optimum toughness so that the plate will not disintegrate and stay continuous during the flight which is critical for the performance.

Recently explosive reactive armour has been developed with hard armour steel plates to defeat shaped charge and to reduce the penetration of long rod projectiles.

High strength armour steels have been developed for use in explosive reactive armour. These steels are tempered martensitic steel. The steel plates were tempered at different temperatures such that the resulting mechanical properties will suit the requirements. They were used as part of ERA sandwich structure. The ERA developed have resulted in 80 per cent reduction in the shaped charge penetration (Fig. 9) and around 30 per cent reduction in the long rod penetration in RHA.

Figure 9. Experimental set up for ballistic evaluation of ERA against shaped charge warhead.

9. Materials For Non-Explosive Reactive Armour

Non explosive reactive armour (NERA) sandwiches are attractive in add-on-armour applications. NERA is also called bulging armour. While explosive reactive armour sandwiches are known to be extremely effective against shaped charge jets, they have inherent disadvantages such as those caused by undesired interaction of the flying plates with the main armour and the environment. To overcome the significant safety drawback posed by the explosive content of reactive armour, inert cassettes containing metallic sandwiches with inert filling materials were proposed by Held¹⁶ and Lundgren¹⁷, et al. Jet-metal interaction involving momentum exchanges in a direction transverse to the jet motion are considered to be the basic cause for jet disruption in a reactive armour¹⁸. Figures 10 shows the basic working principle of non explosive reactive armour. Difference in the energy content is the main feature that distinguishes inert cassette from the reactive sandwich.

Figure 10. Working principles of non explosive reactive armour.

Due to significant safety advantage, development of non explosive reactive armour (bulging armour) has been pursued at DMRL. Recent DMRL results of NERA against HEAT missile have shown that NERA can reduce more than half of the penetration of HEAT missile when compared to its penetration in monolithic RHA steel plate¹⁹. Figure 11 shows photographs of the test set up and bulging armour panels before and after penetration.

Figure 11. . (a) Photographs of the test set up for firing of NERA against shaped charge warhead and bulging armour panels (b) before and (c) after penetration.

10. Conclusions

The strength and hardness of the currently used RHA steel are increased substantially, to 1380 MPa and 450 VHN respectively by adopting suitable heat treatment procedures. Weldability of the medium hardness steel has been found to be satisfactory up to 80 mm thickness in steel plates tempered at 500^OC. The medium hardness steel displayed around 10 per cent - 15 per cent improvement in ballistic performance over currently used RHA steel. DMR-1700 steel plate’s shows good combination of strength and impact toughness. There is a significant improvement in ballistic performance of the high hardness steel compared to RHA steel. Due to better ballistic performance and lower cost, high nitrogen steel has potential to replace RHA (SPADE) steels in armour modules and add on armour structures. Significant weight saving can be achieved in armour solutions which are primarily used against deformable projectiles by using HNS in the place of SPADE steel. In case of armour solutions against armour piercing and long rod projectiles, application of HNS in place of SPADE steel will result in significant cost reduction.

Ballistic performance of laminated composite against 125mm FSAPDS residual depth of penetration shows similar for epoxy and phenolic based composite laminates. The study shows type of resin is not affecting on ballistic performance of E-glass composite laminates when it is subjected to 125mm FSAPDS ammunition. Laminate ballistic performance has been compared with RHA and found that mass efficiency is better than RHA whereas thickness efficiency is inferior to RHA plates. Kanchan composite armour can provide protection against kinetic energy projectiles as well as chemical energy based shaped charged projectiles.

High strength armour steels have been developed for use in explosive reactive armour. The ERA developed has resulted in 80 per cent reduction in the shaped charge penetration and around 30 per cent reduction in the long rod penetration in RHA. NERA can reduce more than half of the penetration of HEAT missile when compared to its penetration in monolithic RHA steel plate.

References

1. Jena, P.K.; Bidyapati, Mishra; Babu, M. Ramesh; Babu A.K.; Singh, Arvindha, K; Kumar Siva & Bhat, T.B. Effect of heat treatment on mechanical and ballistic properties of a high strength armour steel. Int. J. Impact Eng., 2010, 37, 242-249.

2. Metals handbook. Ed. 9. Welding, Brazing and Soldering, American Society for Metals, Metals Park, OH, ASM, Vol. No. 6, 1983.

3. Singh, B Bhav; Sukumar, G.; Kumar, K. Siva & Gogia, A.K. Ballistic studies on nickel free high nitrogen steel. Technical Report No. DRDO-DMRL-ADDG-1-120-2016, 2016.

4. Sutherland, L.S. & Soares, C.G. Impact characterization of low fibre volume glass reinforced polyester circular laminated plates. Int. J. Impact. Eng., 2005, 31, 1-23.doi: 10.1016/j.ijimpeng.2003.11.006

5. Aslan, Z.; Karakuzu, R. & Okutan, B. The response of laminated composite plates under low velocity impact loading. Composite Structure, 2003, 59(1), 19-27.doi: 10.1016/S0263-8223(02)00185-X

6. Sutherland, L.S. & Soares, C.G. Impact on low fibrevolume, glass polyester rectangular plates. Composite Structure, 2005, 68, 13-22.

7. Larsson, F. & Svensson, L. Carbon, polyethylene and PBO hybrid fiber composites for structural light weight armour. Compos. Pt. A Appl. Sci. Manufact. 2002, 33, 221-231.doi: 10.1016/S1359-835X(01)00095-1

8. Morye, S.S.; Hine, P.J.; Duckett, R.A.; Carr, D.J. & Ward, I.M. Modelling of the energy absorption by polymer composites upon ballistic impact. Compos. Sci. Technol. 2000, 60, 2631-2642.

9 Masta, M.R.O.; Crayton, D.H.; Deshpande, V.S & Wadley, H.N.G. Mechanisms of penetration in polyethylene reinforced cross-ply laminates. Int. J. Impact. Eng., 2015, 86, 249-264.

10. Karahan, M.; Jabbar, A. & Karahan, N. Ballistic impact behavior of the aramid and ultra-high molecular weight polyethylene composites. J. Reinf. Plast. Compos., 2015, 34(1) 37-48.

11. Lanz, W.; Odermatt, W. & Weihrauch, G. Kinetic energy projectiles: Development history, state of the art, trends. In 19th International Symposium on Ballistics, Interlaken, Switzerland, 2001.

12. Retrieved from http://en.wikipedia.org/ Wiki/Composite_armour (Accessed on 03 Februray 2016).

13. Senthil, P. Ponguru; Kumar, K. Siva & Gogia, A.K. Terminal ballistic eroding long rod impact DMRL technical report. Technical Report No. DRDO-DMRLADDG-019-2012.

14. Elshenawy, Tamer; Ismail M.M & Reyad, A. Optimization of performance of explosive reactive armours. In 21st International symposium on ballistics, Adelaide, Australia, 2004, 1, P. 227.

15. Yadav, H.S.; Bohra, B.M.; Joshi, G.D.; Sundaram, S.G. Kamat, P.V. Study on basic mechanism of reactive armour. Def. Sci. J., 1995, 45, 207-212.

16. Held, M. Schutzanordnung gegen Geschosse, insbesondere Hohlladungsgeschosse, 1973 German Patent No. 2358227.

17. Lundgren, R., Medin, G., Olsson, E. & Sjod, L. Reactive armor arrangements. 1987. US Patent No. 4881448.

18. Mayseless, M., Erlich, Y., Falcovitz, J., Rosenberg, G. & Wheis, D. (1984) Interaction of shaped charge jets with reactive armour. In 8th International Symposium on Ballistics, Orlando, USA.

19. Rao, S.S.; singh, B bhav; Papukutty, K.K.; Madhu, V.; Siva kumar, K. & Bhat, T Balakrishna. Bulging composite armour against high explosive anti tank ammunitions. DMRL, TR No: 2002 313, 2002

Acknowledgments

The authors would like to thank Director, DMRL for giving permission to publish this paper. The valuable suggestions of Dr T. Balakrishna Bhat and Dr A.K. Gogia have been very useful. The authors would like to thank PXE, Balasore for conducting large caliber ammunition trials and HEMRL, Pune for conducting HEAT ammunition trials on bulging armour panels and ERA panels. Support of various groups and small arms range of DMRL in different stages of this work is highly appreciated.

Contributors

Mr B. Bhav Singh obtained his MTech (Metallurgical & Materials engineering) from the Indian Institute of Technology Madras, in 2004. Currently he is working as Scientist E at Armour Design and Development Division, Defence Metallurgical Research Laboratory, Hyderabad. His research interests are in the areas of development of steel, titanium armour, bulging armour. In the present work, he has carried out the experimental studies on high hardness steel, medium hardness steel, high nitrogen steels, bulging and preparation the manuscript.

Mr G. Sukumar obtained his MTech (Metallurgical and Materials engineering) from the Indian Institute of Technology Madras, in 2009. Currently, he is working as Scientist-C at the Armour Design and Development Division, Defence Metallurgical Research Laboratory, Hyderabad. His research interests are in the areas of development of steel, titanium armour and bulging armour. In the present work, he was involved in the experimental studies on high nitrogen steels, bulging armour and writing manuscript.

Mr P. Ponguru Senthil obtained his MTech (Metallurgical and Materials engineering) from the Indian Institute of Technology Madras, in 2016. Presently he is working as Scientist-D at the Armour Design and Development Division, Defence Metallurgical Research Laboratory, Hyderabad. He is involved in the development of armour steel plates for advanced explosive reactive armour and medium hardness armour steel development. In the present work, he was involved in the experimental studies on explosive reactive armour.

Mr P.K. Jena obtained his BE (Metallurgical engineering) from Utkal Univesity, in 2001. Presently, he is working as Scientist ‘D’ in Defence Metallurgical Research Laboratory, Hyderabad. He is working in the areas of development of armour materials & systems for various types of protective applications. His research interests are in the areas of development of metallic armour materials like high hardness and medium hardness steel and aluminium armour. In the present work, he was involved in the development of high hardness armour steels, medium hardness steels and their weldability studies.

Mr P. Rama Subba Reddy obtained his MSc (Polymer Sci.) from S.K. University, Ananthapur, in 1998. Presently, he is working as Scientist E at Armour Design and Development Division, Defence Metallurgical Research Laboratory, Hyderabad. He is working in the areas of development of armour materials and systems for various types of protective applications. He has rich experience in various areas like armour composites and their evaluation for ballistic applications. In the present work, he has carried out the experimental studies on composites armour.

Dr K. Siva Kumar obtained his PhD (Metallurgical and Materials Engineering) from IIT Bombay, Mumbai, in 1995. Presently, he is working as Scientist G at Armour Design and Development Division, Defence Metallurgical Research Laboratory, Hyderabad. He is working in the areas of development of armour materials and systems for various types of protective applications. His research interests are in the areas of composite armour development, metallic armour materials and bulging armour. In the present work, he was involved in planning of work and arranging various materials and systems required for experiments.

Dr Vemuri Madhu, obtained his PhD (Applied Mechanics) from IIT Delhi, in 1993. Currently, he heads the Armour Design and Development Division, Defence Metallurgical Research Laboratory, Hyderabad. He is working in the areas of development of armour materials and systems for various types of protective applications. His research interests are: Ceramic and composite armour development, modelling and simulation of ballistic phenomena, high strain rate, shock and blast studies. In the present work, he was involved in design of experiments and analysis of test results.

Dr G. Madhusudhan Reddy obtained PhD (Metallurgical Engineering) from Indian Institute of Technology, Madras, in 1999. Presently he is working as Scientist ‘H’ and heading the Metal Joining Group of Defence Metallurgical Research Laboratory, Hyderabad. He has more than 300 scientific publications to his credit. In the present work, he has carried out the weldability studies of different armour materials such as high hardness steels and medium hardness steels.

Posts · by **Yohannes** » Fri Apr 27, 2018 8:38 pm

User-Friendly Explosive Reactive Armour - A Long Term Reality

Journal page: Dcfence Science Journal, Vo147, No 2, April 1997, pp. 265-273
@ 1997, DESIDOC

AUTHORS: S.N. Dikshit

*E-mail: bhav_singh@dmrl.drdo.in
Defence Metallurgical Research Laboratory, Hyderabad – 500 058, India

Article history:
Received 06 February 1995
Revised 14 August 1996

Reproduced with permission of copyright owner. Further reproduction prohibited without permission.

Col (Dr) SN Dikshit obtained his BE (Metallurgy) from REC, Warangal, in 1970, ME (Mechanical) in 1982 from Indian Institute of Science, Bangalore, in 1970 and PhD from Banaras Hindu University in 1994. His research work includes armour materials. He is working at Defence Metallurgical Research Laboratory, Hyderabad, where he is actively engaged in the design and development of light weight 'Kanchan' composite armour and 'Explosive Reactive Armour' for the protection of MBT Arjun, Vijayanta and Ajeya tanks. He has published several papers international and national journals. He is a life member of Indian Institute of Metals.

ABSTRACT

There is a strong need to develop explosive reactive armour (ERA) for protecting battle tanks against an emerging threat of kinetic and chemical energy missiles. In this context, global trends, principle and limitations of ERA and threat perception-based types of ERA have been dwelt upon. User-friendly ERA is a long-term reality. User-friendly ERA system is thus defined to be an efficient and protective system that not only provide full protection to the tank crew, but is also harmless to the supporting infantry.The indigenously-developed ERA system is close to be termed as a user-friendly ERA.

1. INTRODUCTION

For over 75 years, in any conventional ground action, a battle tank has been the key weapon due to its inherent characteristics of high mobility, firepower and crew protection. Accordingly, ever increasing quest for higher fire power, higher mobility and better protection, has led to the design of present day main battle tanks (MBTs), which fall in the category of heavy tanks. Threat by lethal kinetic as well as chemical energy projectiles further poses a formidable risk of maintaining such a high power-to-weight ratio with increased protection. Seemingly, the development of gas turbine engine might appear to provide some relief to salvage mobility problem for the tank designers. However, unaffordable overall cost associated with the demand of high order logistic support due to heavy fuel consumption and frequent high standard maintenance support due to poor reliability in the dusty terrain puts the clock back. A natural question that arises in the minds of the designers is where to go from here ? Hit avoidance, wherein the incoming projectile is destroyed far away from the surface of the tank, would be a logical and straightforward answer to this problem. However, this concept of active armour remains in conceptual state only and it would be a matter of decades before it comes into operation. In the absence of any realistic solution, an intelligent application of the age-old explosive power may thus offer the desired results in the years to come.

This paper highlights the importance of ERA in view of the development of high penetration kinetic energy projectiles and shaped charge missiles. The principle of design, limitations, and global trends in the development of user-friendly explosive reactive arm our (ERA) system for the protection of battle tanks have also been dealt with.

2. NEED FOR ERA

With the introduction of explosively formed projectiles (EFP) having striking velocity of more than 2000 m/s, tandem missiles, advanced antitank guided missiles capable of penetrating 1000-1200 mm of rolled homogeneous armour (RHA) steel and depleted uranium (DU) kinetic energy projectiles (KEP) have threatened the very existence of even MBTs of the world. Though depleted uranium penetrators may not find place in the tank due to obvious reasons, the tungsten (W) penetrator technology advancement also appears to be at its peak level. Present day tungsten (W) penetrators have very high penetration capability¹ as mentioned in Table I. It should be appreciated that a maximum of 23-25 per cent of the total weight of the tank can only be reserved for the purpose of protection of the tank². Whereas for protecting armoured fighting vehicles (AFVs) against present and futuristic ammunition, this figure can be as high as 30-40 per cent, thereby causing an imbalance in the design of AFVs.

Table 1. Latest ammunition development parameters

Against these impressive odds, passive armour may no longer be able to protect the crew of the tank in its present form. The ERA development thus assumes importance in providing protection to MBTs and old generation tanks held by various countries of the world. Some of the locations on the tank turrets may have slightly poor protection levels, especially in the case of old generation tanks. In these locations, bulging caused by ERA may not be of severe nature as part of the projectile energy is dissipated in bending and stretching of the plate material³.

3. PRINCIPLE & WORKING OF ERA

Many investigators have expressed their views about the working of ERA^4-12. However, there is no coherency in these expressions. Some explain that the projection of the metallic plates in the path of the jet adds to the thickness of the base armour. There are some who explain that it is the disruption of the jet caused by the detonation products thereby reducing the penetration capabilities of the jet of the shaped charge. Most or the investigators agree that the efficiency of ERA is drastically reduced at zero obliquity. They also confirm that basically the ERA system comprises explosive sheet sandwiched between two metallic plates.

While understanding the principle of working of ERA, it is to be noted that the mechanism of ERA functioning for a shaped charge differs from that of KEP. In the case of a high speed jet formed by a shaped charge, reduction in penetration can be achieved by:

The finding of this study is that both the flying-off of the metallic plates in the path of jet and the disruption of jet by the detonation product, play a role in reducing the penetration of the jet. However, a large number of experiments conducted in this area provide definite information that the major cause of reduction in penetration due to the disruption of the jet and plate-flying plays a secondary role in the functioning of ERA. It should be, however, appreciated that in a microsecond phenomenon of detonation of the explosive, plates are very much required for providing confinement to the detonation of products. It can thus be understood that apart from the projection of these plates in the path of the jet, indirect application of these plates is to assist the disruption of the jet by providing confinement for few microseconds. The very fact that the thickness of these top and bottom plates of ERA developed by various countries in the range of 2-12 mm is a point to ponder about their I role in the penetration reduction. Plate thickness of the order of 4-12 jet diameter is predicted for initiation of explosive by a shaped charge jet having a velocity of 7 km/s and jet diameter as 1.5-3 mm^10,13,14. Additionally. if plate cutting role is assumed to be a dominant factor, there is no reason as to why these plates should not be made out of a high density material like tungsten, which will offer drastic reduction in the penetration of the jet. The use of aluminium alloy, mild steel, armour grade material and even dense alumina (ceramic) as a plate material clearly points towards the fact that the disruption of jet is a major issue in the working of ERA.

With regard to the efficacy of ERA against long rod penetrator, information in the open literature is quite patchy. In a limited number of experiments conducted by us, it is evident that ERA can function against long rod penetrators, provided the sensitivity of the explosive is optimised to ensure its detonation at a very low striking velocity of these KEP (V = 1300-1500 m/s). Reduction in penetration of KEP is achieved by a combination of the deflection and fragmentation of KEP in the presence of the detonation product. Reduction in penetration by a KEP with low ratio of length to diameter of such projectiles, has been observed to be quite appreciable. For clear understanding of the working of ERA against long rod penetrators, a large number of experiments have to be conducted.

4. ADD-ON EFFECTS

ERA panels when mounted on the tank on its frontal arc, sides, nose plate and turret top will enhance its protection level against the missile threat. At the same time, these panels will have some overall adverse effects on the tactical functioning of the tank. Some of the important
points are:

¹⁵

Keeping in view, the gains achieved due to the employment of ERA on the battle tank, minor changes as indicated above have to be adopted. In any kind of add-on armour system adopted for enhancing the protection level of the tank, basically some kind of compromise in its tactical functioning has to be accepted by the tank crew. Add-on effects cannot be totally avoided while designing such armour.

5. LIMITATIONS OF ERA

ERA is a novel technique to. protect battle tanks against the threat of high-calibre, shaped charge projectiles, without affecting the mobility of
the tank. However, its use also results in a large number of functional restrictions imposed on the crew of the tank. Some of the most relevant limitations of ERA system are:

¹⁶

^O

⁰

6. DESIGN PRINCIPLE

The ERA system has three basic components, namely, two metallic plates, thin sheet of explosive and a container with an appropriate mode of
mounting the same. Before understanding the design and development of an ERA system, it would be essential to know the following:

Having gained information on the above parameters, it is, desirable to ascertain the performance of the ERA system used for enhancing the protection levels of the tank. Since explosive forms the heart of the ERA system, the constituents of explosive are to be optimised first. The optimisation of many more technical parameters is totally dependent on the explosive quality. The change of explosive leads to changes in many other design parameters. It is thus important to understand that ERA design revolves around the type of explosive being utilised in the ERA system. The speed of the shaped charge jet, its diameter and the velocity of detonation (VOD) of the explosive in relation to the mass (thickness) of the flying plates are the critical design parameters for the success of the ERA system. Mass of the flying plates and the mass of explosive used will play a deciding role in the design of such armour system. Like a fire triangle, ERA design can be summarised in the form of a speed-based triangle as shown in Fig. 1. Matching of these three speeds is the prime concern of an ERA designer. Since the optimisation of explosive depends on the level of technology developed by any country, it is seen that design philosophy of different countries is different. Such a difference in explosive technology will lead to different types of ERA products being available in the market. It is for this reason that the ERA panels offered by different countries are available in different shapes and sizes. The thickness of top and bottom plates and their material also differ in these ERA systems. In addition, the tank protection philosophies pursued by various countries will also be reflected in these varying ERA products.

Figure 1. Speed-based ERA design triangle.

While designing an ERA system, development work will include optimisation of the following parameters:

The above design parameters can be classified into vital, essential and desirable categories. Perfection in optimising the vital parameters cannot be neglected at any cost, and maximum development efforts are to be expended in it. However, optimisation of desirable parameters at times poses serious challenges, leading to overall changes even in the design of the vital parameters. In our development work, many years got wasted in just finalising the mounting mode of the ERA system, which otherwise appeared to be a simple task.

7. GLOBAL TRENDS

Information available suggests that the idea of protecting tanks with the application of ERA is quite old. Dr Held⁴ invented ERA and his basic patent was accepted in 1970. Some of the advanced countries have an experience of more than four decades in the design and development of ERA. In fact, most of the NATO countries have gone for ERA development and they appear to be engaged in joint collaborative research work leading to the development of ERA and its countermeasures. Merits, demerits and progress made by various countries in this area are available to a limited extent in open literature. Countries like Russia^17,18, Poland¹⁹ and Israel²⁰ have provided details of ERA development work in open literature. However, NATO countries are maintaining silence over ERA development and its application on battle tanks. The USA appears to have not released any information about using ERA on battle tanks, though it has been put on MICV Bradley along with/honeycomb structure¹⁴. Truly speaking, silence of some of the countries by no chance should be construed as lack of interest either on the part of the army or the designers. On the other hand, it would be quite reasonable to assume that the so called mark I models of ERA have been developed by most of the countries for emergency situations. These mark I models may be light in weight, indicating that they are primarily meant to counter the antitank guided missiles of eighties. All the countries might have conducted elaborate field testing of Mark I models and scientists may now be engaged in minimising the serious limitations that may have been noticed during these trials. It is therefore wise to appreciate that the basic aim of total silence by these countries may be striving towards excellence.

It appears to be an accepted and well-known fact that ERA is less effective at normal inclination, i.e. when path of the jet coincides with the normal of the ERA panel⁴. The higher the obliquity the better is the performance (Fig. 2). Some of the countries appear to have achieved appreciable degree of competence in this direction.¹⁹ Figure 3 provides probability-based hit locations on ERA panel. The selection of overall design parameters based on the worst situation is observed to be satisfactory.

Figure 2. Angle or attack and ERA performance

8. TYPES OF ERA

The need for and long-term gains of ERA have been understood by the world. The most confusing aspect of present day ERA development lies in knowing as to what type of ERA one has to design. There is no straightforward answer to this question. The choice of the type of ERA, i.e. light weight, heavy weight or medium weight will be governed by the tactical considerations, as dictated by threat perception. This clearly brings out the fact that ERA panels developed by one country may not be optimally suitable for another country. Design of the type of ERA is thus linked with the threat being visualised on the ground and on the existing protection level of the tanks. Thus, it is obvious that ERA panels designed for main battle tanks of the world may not provide satisfactory protection levels to the old generation tanks. The evolution of the type of ERA gets restricted to the total weight penalty as given in Table 2. Weight penalties expressed as percentage of total weight of the tank have indirect relation with the mobility of the tank turret. The option of heavy weight ERA would appear to be attractive; however, it may cause some degradation in the speed of rotation of the turret, especially beyond a slope of 25^O. This aspect will have to be examined in depth, before designing heavy weight ERA panels.

Table 2. Expected weight penalties and protection gains

9. USER FRIENDLY ERA

In a broader sense, user-friendly ERA system is imagined to be a kind of efficient protection system which would not only provide full protection
to the tank crew without any psychological barrier but also be harmless to the supporting infantry. In other words, any ERA system with the least number of limitations, qualifies to be termed as user-friendly ERA. Such a user-friendly ERA developed by any country is bound to stay for a long period, unless scientists lay their hands on a wonder material which can do the job of ERA while being totally insensitive. Thus the scientific community of the world is faced today with the task of converting 'explosive power' into 'friendly power' .Some of the logically desirable features of user-friendly ERA can be summarised as:

^21-25

While ERA offers unimaginable weight and space advantages in protecting world tanks against serious threat caused by missiles and KEP having much superior penetration capabilities, designers have to make ERA a user-friendly armour, at least to a reasonable extent. As per the open literature, so far, no country appears to have developed a user-friendly ERA fulfilling all the above requirements. The limitations of ERA system are quite noticeable; perhaps that is the reason why some of the armies of the world have not yet accepted introduction of ERA in the Services, though such armour is developed by them. Development work on indigenous ERA system suggests that we are very close to offer a user-friendly
ERA system.

10. CONCLUSION

Development of User-friendly ERA system is a long-term reality for the protection of battle tanks against the threat of high penetration KEP and shaped charge jets. Indigenous development work suggests that we are close to offer such a system in due course.

ACKNOWLEDGEMENT

The author wishes to express his gratitude to Shri S.L.N. Acharyulu, Director, DMRL, Hyderabad for granting permission to publish this paper.

REFERENCES

1. Int. Def. Rev., 1994, 27(4), 43.

2. Simpkin, Richard (Ed.). Tank warfare, design constraints and tialance. Brassey Publishers Ltd., London, 1979. p.J27.

3. Dikshit, S.N. and Sundararajan, G. The penetration of thick steel plates by ogive shaped projectiles-experimental and analysis. Int. J. Impact Engg., 1992,12(3), 373-408.

4. Military Technology, 1991,8.

5. Armour/ Antiarmour, SAL Documentation, London.

6. Janes Def. Wkly., November 1989.

7. Janes Def. Wkly., January 1990.

8. Armour, January-February 1988.

9. Janes Def. Wkly., May 1987.

10. Ogorkiewicz, R.M. Combat vehicles armour progress. Int. Def. Rev., Vol 6, 1995.

11. Dikshit, S.N. Emerging trehds in the development of explosive reactive armour, DMRL, Hyderabad, 1990. DMRLTR 90107.

12. Yadav, H.S.; Bohra, B.M.; Joshi, G.D.; Sundaram, S.G. & ~arrlat, P.V. Study on basic mechanism of reactive armour. Def. Sei. J., 1995, 45(3), 207-12.

13. Frey, R.; Lawrence, W. & Chick, M. Shock evaluation after shaped charge jet impact and its relevance to explosive initiation; Int. J. Impact Engg., 1995, 16(4), 563-70.

14. Military Technology, 1989, 3.

15. Dikshit, S.N. Complexities of add-on armour and remedial measures. Del 5ci. J., 1996, 46(2).

16. Military Technology, 1986, 4; 1987, 1

17. Int. Def Rev., 1990, 7, 728.

18. Int. Def Rev., 1988, 9, 1065.

19. Janes Sov. Intelligencel Rev., July 1989.

20. Armour, January-February 1983.

21. William, P. WaIters; Flis, W.J. & Ch~u, P.C. A survey of shaped-charge jet penetration models. Int. J. Impact Engg., 1988, 7(3), 307-25.

22. Pack, D.C. On the perturbation and break-up of a high speed, elongating metal jet. J. Appl. Phys., 1988,63(6), 1864-71.

23. Chou, Pei Chi,; Carleone Joseph & Karpp, Rabert R. Criteria for jet formation from impinging shells and plates. J. Appl. Phys., 1976, 47(7), 2975-81.

24. Singh, M.; Madan, A.K. & Bola, M.S. Particulation of metallic jets under high strain rate. Int. J. Impact Engg., 1994,15(5), 699-710.

25. Chou Pei Chi & Carleone Joseph. The stability of shaped-charge jets. J. Appl. Phys., 1977, 48(10).

Posts · by **Yohannes** » Mon Apr 30, 2018 6:02 pm

Experimental studies on protection systems of military vehicles against RPG type missiles

Solid State Phenomena
ISSN: 1662-9779, Vol. 240, pp 244-249
doi:10.4028/www.scientific.net/SSP.240.244
@2016 Trans Tech Publications, Switzerland

AUTHORS: Robert Panowicz^1,a*, Tadeusz Niezgoda^1,b

¹Military University of Technology, Faculty of Mechanical Engineering, Department of Mechanics & Applied Computer Science, Kaliskiego 2, 00-908 Warsaw, Poland
^arpanowicz@wat.edu.pl, ^btniezgoda@wat.edu.pl

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Article history:
Submitted: 2014-11-29
Accepted: 2015-02-18
Online: 2015-08-28

Introduction

RPG missile is an unguided, anti-tank rocket missile equipped with a shaped charge warhead and is produced in many countries all over the world. A weapon of this kind is currently utilized by approximately 40 countries as well as by a similar number of illegal organizations. It is estimated that annual production of these missiles is equal to approximately 4 million items. It became popular owing to low production costs, small dimensions, small mass, high efficacy (up to 100 m its efficacy is estimated at 96%) and penetration as well as service easiness. It moves on the trajectory at the velocity ranged from 230 to 280 m/s covering the distance ranged from 300 to 500 m, and the shape charge jet is able to penetrate 300 m of RHA steel.

The warhead is composed of a case, explosive material, a liner and an electric fuse (Fig. 1). The explosive material placed in the charge is appropriately formed, what allows orientation and concentration of energy of detonating explosive material. Charge formation consists in making a cavity with an angle of approximately 60^O, in its front part and placing a metal liner (e.g., OFHC Cu, ARMCO iron) in it. When the explosive material is detonated on the symmetry axis, the axially symmetrical detonation wave, starts to move in it. The wave reaches the liner at the certain moment and starts to move along it from inside to outside. The pressure of the detonation products on the liner causes its deformation as well as movement along and towards the symmetry axis. Rapid accumulation of the material of the liner takes place on the symmetry axis, what also results in a rapid growth of pressure especially on the axis of the system. On the direction where the pressure is lower (symmetry axis), the material is propagated and the jet is formed from it. The jet constitutes approximately 20% of the liner mass and moves along the symmetry axis of the charge with velocity of a few kilometres per second. The remainder of the liner creates a slug which moves with relatively low velocity. The jet is in the form of thin metal thread of velocity gradient between the jet tip, the temperature of which is about 800 K, and the tail remaining in contact with the slug of temperature of about 400 K. The gradient of velocity is responsible for fragmentation the jet integrity and reducing its efficiency. Its thickness in the central part is a few millimeters. The jet is a factor responsible for destruction. It creates a small hole of the depth up to 9 calibres of the liner.

Generating the jet requires making part of a shaped charge and its assembling with a very high accuracy maintaining the symmetry of the whole. It results from the principle of operation of the shaped charge that the greater the symmetry is the more the perforation increases. The symmetry of the charges increases along with the development of technology, what has been presented [1]. Modern charges satisfying such conditions are capable of perforating up to 9 calibres (a charge of a 100 mm-diameter perforates up to 900 mm of RHA steel).

Fig. 1. Shape charge

Protection against this type of threat is provided by different kinds of countermeasures such as reactive, bar and slat armours and an active protection system.

The purpose of slat and mesh armours fitted on a protected object is such destruction of a missile with the shaped charge warhead (the symmetry disorder) in which the shaped jet is not produced.

The same purpose is related to an active protection system, however, in this case, destruction of the missile takes place at a certain distance from the armour and without any contact with it.

It results from the presented principle of operation of the charge that it is possible in the following cases:

Research on passive protection systems

Research on passive protection systems (bar armours) of vehicles began with both static and dynamic tests on materials. To determine the strength properties of materials in the static range, a universal strength machine INSTRON was used, however, to determine the strength properties of materials in the dynamic scope/range, a stand with a split Hopkinson pressure bar was utilised. Material constants for a constitutive Johnson-Cook model can be defined based on a classic algorithm presented in works [2, 3], or an algorithm using genetic optimization methods [3].

Applying the capabilities of reverse engineering, CAD model of RPG missile together with a fuse was built, and subsequently it was simplified and divided into finite elements (Fig. 2). The simplification were introduced to the model in such a manner that they do not affect the behaviour of the warhead. It enabled reduction of the number of finite elements of the model and hasten numerical analyses of the considered system. The detailed data concerning this model are presented in works [4, 5]. Representative numerical analysis of the course of the entire phenomenon with taking into consideration geometric and material non-linearity, an influence of large deformations, strain rate and a process of missile destruction is presented in Fig. 3.

Fig. 2. CAD model of RPG missile, 1- head, 2 - rocket engine, 3 - stabilizer

Fig. 3. RPG interaction with rod armour (a), liner deformation and explosive crushing (b) at two different time

For numerical analysis there was applied Ls-Dyna software including a finite element method with explicit integration scheme using for analyses of different fast changing phenomena [6].

Owing to the content of dangerous materials in RPG missile, there was developed a model of the system expressing basic inertial characteristics of the system.

The conducted laboratory experimental studies and numerical analyses result in a conclusion that basic characteristics of the armour deciding about effectiveness of armour work are stiffness and inertia of the armour. If these properties are too small, the contact of the missile with the armour elements results in their fast extension. Therefore, there occur only local damages of the missile at the point of its contact with the armour.

The selected types of bar armours, with the geometrical and strength parameters previously determined based on numerical analyses, were examined during experimental fire-field tests. The course of the interaction process was recorded with the use of a fast camera Phantom V12, and efficiency of the armour work was estimated through the marks left by a shaped charge on the classic armour. The representative frames from the research results are presented in Figure 4.

Fig. 4. Experimental investigation, (a) film frames, (b) armour after investigation

There is observed a flash which is an effect of rapid contact of metal elements and destruction of the shaped charge warhead, from which crashed explosive material is spilt out. Only one element of the armour was deformed. With respect to the abovementioned, such an armour is effective against multiple firing.

The presented scheme of experimental tests was modified through fitting, behind the bar armour, a mirror allowing observation of the phenomenon course from the side as well as from the direction of the oncoming missile (Fig. 5) The representative results of the experimental research are shown in Figure 6. Figure 6a presents destruction of the warhead, whereas Figure 6b presents relatively small deflection of the missile case, which led to the short circuit in the fuse system. Therefore, the missile did not detonate after hitting the mirror.

Fig. 5. Experimental stand with mirror

Fig. 6. Experimental investigation, (a) head destruction, (b) fuse short circuit

Such an armour does not work in the case of the fuse contact with an element of the bar armour, which is the main drawback of the armour of this kind.

Research on active protection system

A similar methodology of tests was used in the case of research on active protection system.

Using computer methods in mechanics, there was developed a fragmentation warhead, consisted of a case, explosive material, a fragmentation layer and a fuse, which drives the fragments to the velocity allowing significantly damage the warhead of RPG missile [7, 8].

The average velocity of fragmentation elements determined based on numerical analyses was determined at the level of 600 m/s, while, from the analysis of experimental tests results, this velocity is in the range of 500 to 800 m/s. Taking into consideration the fact that the velocity of the missile is higher than 200 m/s, the collision of the fragment and the missile occurs when the resultant velocity is approximately 750 m/s.

The warhead was tested in the static conditions estimating the density of fragments arrangement and their energy(Fig. 7). Based on the conducted static analyses, it was estimated that each object of RPG missile dimensions which is in the angle of 20^O from the warhead and at the distance up to 6 m is hit by at least 2 fragments of energy allowing perforation of the armour case, the shaped liner as well as crushing of the explosive material.

Fig. 7. Static tests, (a) spatial distribution of fragments, (b) static tests of interaction of fragments and RPG missile warhead RPG

At the final stage of the research, the entire system was tested in simulated real conditions; Fig. 8. To detect RPG missile, an optoelectronic detective warhead was used [9].

Fig. 8. Fire filed tests of the system; 1) RPG missile, 2) detonation of fragmentation cassette, 3) detonation of missile with shaped charge warhead resulted from interaction with the fragments

The effectiveness of the missile operations was determined based on the films recorded with a fast camera as well as on the base of the interaction of the shaped jet and a classic armour. The representative results of the tests are presented in Figures 9.

Fig. 9. Classic
armour, (a) partial
destruction of shaped
charge warhead, (b)
full destruction of
warhead - no marks

Summary and conclusions

It can be concluded from the conducted experimental research on the passive and active protection systems that the course of both of the phenomena presents a stochastic character. In the case of bar armours, it is related to weather conditions which change the RPG trajectory, however, in the case of the active protection system, it is also related to participation in the experimental tests of other elements influencing the moment of activation of the fragmentation warhead.

The effectiveness of the bar armours exceeds 70%, however, it is strongly dependent on the angle of firing. This value decreases the fastest in the case of the armours made of flat bars, however it decreases slower in the case of the armours made of the element of the square and circular cross section. The drawback of the two latter armours is their bigger mass.

In the case of active protection, it is difficult to determine the efficiency of operation of the fragmentation warhead as the moment of its activation is influenced both by the moment of detection and the moment of fuse activation. Due to its significant importance, a special significance is attached to the random variables of the moment of fuse activation.

The operation of the entire system is greater than 80%, however, only in a few cases a complete destruction of the warhead occurs. In most cases, the destruction is so huge that the generating destructive element (now it is not a shaped jet any longer) perforates approximately 1 cm of steel, which means 30-fold diminishing of its effectiveness.

References

[1] R. M. Ogorkiewicz, Advances in armour materials, International Defense Review, 4 (1991).

[2] G. J. Johnson, W. H. Cook, A constitutive model and data for metals subjected to large strains, high strain rates and high temperatures, in: Proceedings of the Seventh International Symposium on Ballistics, The Hague, 541–547 (1983).

[3] L. Kruszka, M. Magier, M. Zielenkiewicz, Experimental analysis of visco-plastic properties of the aluminium and tungsten alloys by means of Hopkinson bars technique, Applied Mechanics and Materials, 566 (2014).

[4] K. Sybilski, R. Panowicz, D. Kołodziejczyk, T. Niezgoda, Validation study of the simplfied model of the missile with cumulative head, Journal of KONES powertrain and transport, 19, 3 (2012).

[5] T. Niezgoda, R. Panowicz, K. Sybilski, W. Barnat, Numerical analysis of missile impact being shot by rocket propelland grenades with rod armour, WIT Transactions on Modelling and Simulation, 51 (2011), WIT Press, doi:10.2495/CMEM110551.

[6] J.O. Hallquist, LS-Dyna manual, 2006.

[7] J. Nowak, R. Panowicz, M. Konarzewski, Influence of destructor case type on behaviour of fragments in military vehicles active protection system, Journal of KONES Powertrain and Transport, 21, 1 (2014).

[8] D. Kołodziejczyk, P. Kupidura, Z. Leciejewski, R. Panowicz, Z. Surma and M. Zahor, Counterprojectile for active protection system, 27th International Symposium on Ballistics, Freiburg, Germany, April 22–26, 2013.

[9] Z. Mierczyk, A. Kawalec, T. Niezgoda, B. Stec, R. Trębiński, Z. Leciejewski, A. Gawlikowski, P. Knysak, P. Kupidura, A. Młodzianko, R. Ostrowski, R. Panowicz, A. K. Rutkowski, Z. Surma, W. Susek, J. Wojtanowski, M. Zahor, M. Zygmunt, P. Kędzierski, Active protection system, Polish patent, P.408131 z 07. 05. 2014r.

26th Symposium on Experimental Mechanics of Solids
10.4028/www.scientific.net/SSP.240

Experimental Studies on Protection Systems of Military Vehicles against RPG Type Missiles
10.4028/www.scientific.net/SSP.240.244

DOI References

[3] L. Kruszka, M. Magier, M. Zielenkiewicz, Experimental analysis of visco-plastic properties of the aluminium and tungsten alloys by means of Hopkinson bars technique, Applied Mechanics and Materials, 566 (2014). 10.4028/www.scientific.net/amm.566.110

[4] K. Sybilski, R. Panowicz, D. Kołodziejczyk, T. Niezgoda, Validation study of the simplfied model of the missile with cumulative head, Journal of KONES powertrain and transport, 19, 3 (2012). 10.5604/12314005.1138156

[5] T. Niezgoda, R. Panowicz, K. Sybilski, W. Barnat, Numerical analysis of missile impact being shot by rocket propelland grenades with rod armour, WIT Transactions on Modelling and Simulation, 51 (2011), WIT Press, doi: 10. 2495/CMEM110551. 10.2495/cmem110551

[7] J. Nowak, R. Panowicz, M. Konarzewski, Influence of destructor case type on behaviour of fragments in military vehicles active protection system, Journal of KONES Powertrain and Transport, 21, 1 (2014). 10.5604/12314005.1134093

Military Related Engineering and Industrial Resources Thread

Military Related Engineering and Industrial Resources Thread

Re: Military Related Engineering and Industrial Resources Th

Re: Military Related Engineering and Industrial Resources Th

Re: Military Related Engineering and Industrial Resources Th

Re: Military Related Engineering and Industrial Resources Th

Re: Military Related Engineering and Industrial Resources Th

Re: Military Related Engineering and Industrial Resources Th

Re: Military Related Engineering and Industrial Resources Th

Re: Military Related Engineering and Industrial Resources Th

Re: Military Related Engineering and Industrial Resources Th

Re: Military Related Engineering and Industrial Resources Th

Re: Military Related Engineering and Industrial Resources Th

Re: Military Related Engineering and Industrial Resources Th

Re: Military Related Engineering and Industrial Resources Th

Re: Military Related Engineering and Industrial Resources Th

Re: Military Related Engineering and Industrial Resources Th

Re: Military Related Engineering and Industrial Resources Th

Re: Military Related Engineering and Industrial Resources Th

Re: Military Engineering and Industrial Resources Thread

Electrothermal-Chemical Ignition Research on 120-mm Gun in K

Lamoni Resources Batch 3

Future Armour Materials and Technologies for Combat Platform

User-Friendly ERA - A Long Term Reality

Re: Military Related Engineering and Industrial Resources Th

Who is online