Armor Basics
Armor Basics
Armor Basics
by Paul Lakowski
1.1 Basics
Modern AFVs. are rated in three important areas; firepower, armor and mo-
bility. Mobility is often the most important capability viewed from an opera-
tional context but armor and fire power determine success and failure on the
modern tactical battle field. Historically the battle between projectile and
plate has determined the out come of most tank battles. It’s probably true
that fire power is the more important of the two, but often it’s the level armor
that becomes the ‘rate determining step’.
In order to keep pace with gun penetration, designers were forced to focus
more armor to the front at the expense of flank protection. To combat this,
gun designers resorted to high tech ammunition, the and the battle went on.
After WW-II, Soviets and Americans both experimented with ERA equipped
tanks, while the Americans experimented with the silica ceramic armored T-
95. These technologies offered potential but were too costly and the main
solution adopted was to up the weight. The main battle tank went from 20—
30 tons in WW-II to 35—48 tons in the fifties. In other words, this year’s
heavy tank turned into next year’s medium tank just by changing the
name.
In the 60 & 70s the dramatic rise in the potential of the ATGM forced another
evolutionary step. The British resorted to the ‘heavy tank’ with 16 inches of
armor called the Chieftain, while the French opted for a medium tank, and the
Americans developed the M-60 which was a cross between the M-48 & M-
103 designs. The German solution was the Leopard 1, a 40 ton hybrid
tank with the turret armored like a heavy tank while the hull was a me-
dium tank – a smart solution. The Soviets developed the T-64, their own
version of the Leopard. In some respects this was still the heavy tank of the
40s & 50s as the armor of the T-64 was on the same level of the Chieftains in
most places but in other places it was medium tank armor. The Soviet armor
solution was similar to the German solution, just arranged differently.
The 80s saw the introduction of western Chobham armor to counter ATGMs
[Anti Tank Guided Missiles], while the Soviet solution was to add Explosive
Reactive Armor („ERA“) to the T-64—80 tanks, but the gun designers were
able to keep pace. By the 90s even these armors were obsolete and required
upgrading to compete against the latest warheads. The current solutions are
dU [Depleted Uranium] armor for the M-1s and Challengers, "Wedge armor"
for the Leopard 2A5, and K-5 for the Russians. The one thing in common
here is that the new heavy armor only covers about ½ the front profile.
So the first solution in the ever increasing upward need for more armor is to
transfer armor mass to the most vulnerable sections of the tank at the
expense of the less exposed vehicle areas. In addition special materials have
been relied on increasingly to help boost armor levels at some cost to the
design.
1.2 Geometry
Firstly, all projectiles will ricochet. The real question is at what angle
and velocity do they ricochet. Ricochet occurs when a attacking projectile
glances off the sloped armor of an AFV without digging in far enough to
penetrate the plate. If it has no time to dig in before it ricochets, it can’t pene-
trate even modest amounts of armor. A complex model has been developed to
predict the angle at which a projectile is expected to ricochet, this is called the
‘critical ricochet angle’.1
The longer the rod, the higher the ricochet angle and the faster the rod, the
higher the critical ricochet angle. In addition, heavy metal rods of WHA or
dU2 ricochet at higher angles that steel. The critical ricochet angle is meas-
ured from the vertical plane [i.e. 90° is horizontal]. A rod of 10:1 L/d [Length
to rod Diameter ratio] @ 1.7km/s should ricochet at ~78° when made of steel,
while its WHA /dU counterpart will ricochet @ 81°. Stretching the penetrator
to 15:1 L/d increases the ricochet angle to 82—83°, and it’s likely that 30:1
rods will ricochet at >84—85°. Tate’s ricochet formula predicts a ±5° varia-
tion around these values, so 50% of the 10:1 steel rods should ricochet @
~78°, while ricochet will occur as high as 83°and as low as 73°. The above
cases apply to thin plate targets, but if the plate is over 4:1 T/d [plate Thick-
ness / rod diameter ratio] the ricochet angles should go down a few degrees.
Since the time it takes a projectile to ‘turn’ is around 40—60 micro seconds,
and since the entire penetration event takes 300—400 microseconds [large
warhead], even shaped charge warheads [HEAT] will ricochet when the right
combination of striking velocity and angle are reached. Modern HEAT
rounds will ricochet as well, the only question is whether this is before or
after jet penetration. Modern shaped charges with standoff probes and base
initiation will start the jet penetration process before the main round impacts
the slope armor. Since this is a 400 micro second event [½ a millisecond], it is
quite likely that the main body of the round will not even have reached the
plate by then.
The second aspect of slope is the asymmetrical force acting on the penetrator.
When a projectile strikes a sloped plate, the side of the penetrator closest to
the plate will suffer more force, erosion, and damage than the opposing side.
This puts an unbalanced force on the rod, turning it in towards the plate – and
then into the opposite direction. The penetrator takes a longer overall route
through the armor, resulting in less penetration of sloped armor.3
1
See: J. Phys. D. Appl. Phys. Vol 12-1979 pp. 1825—1829.
2
Wolfram Heavy Alloy –Tungsten & depleted Uranium
3
See: Rheinmetall Handbook on Weaponry [figure 1128] (1982)
1.2.2 Projectile shape
This asymmetrical force on the penetrator varies from projectile to projectile,
but it is tied to the nose shape of the projectile. Anderson Jr et al has shown
that the effects of nose shape disappear after the projectile has pene-
trated to a depth of two projectile diameters. Now since AP shot only
reach two projectile diameters penetration, this nose effect is quite dramatic,
but for 20:1 and 30:1 L/d long rod penetrators at higher velocity the effect is
marginal at best. What it means is that by the time you stretch to these rod
lengths, any effect of slope is only a few percent at best and by the time you
reach shaped charge jet L/d ratios [100:1], the effect is no more than 1%.4
The change of effect from slope [45—60°]. All values are for a pointed rod and show how
much the LOS is increased by the change in plate thickness and increasing rod length.5
T/L AP APDS APFSDS
L/d 3:1 4.5:1 1st Gen.: 10:1 2nd: 20:1 3rd Gen.: 30:1
thin 1.03 1.02 1.00 1.0 1.0
spaced 1.33 1.19 1.09 1.04 1.025
moderate 1.4 1.22 1.11 1.05 1.03
Semi Inf. 1.46 1.26 1.13 1.06 1.04
The effect of increasing the armor resistance by slope can also be achieved by
curving the armor. The slope then is a combination of both the ‘tangent’ of
the horizontal and the vertical planes. To determine the net ‘compounded
armor slope’ the following formula is used.
1
2
1
2
+ − 1
cos(V ° ) cos( H ° )
COS= Cosine
V°= Vertical angle
H °= Horizontal angle
With the increasing use of special armors their impact on sloped armor must
also be assessed. When ceramics are struck the effect is to create a huge ‘shat-
ter zone’ radiating outwards in an elliptical pattern that’s larger than the same
damage into a steel target. When the ceramic plate is slanted, the effect is
to dramatically reduce the efficiency of the sloped armor. Tests on sloped
ceramic steel targets struck by AP shot show the effective resistance is only
1.6 times the Line Of Sight [LOS] thickness @ 60°. The same impact on a all
steel target should result in the effective LOS increasing from 2.1 to 2.5.6
Test of APFSDS on slanted ceramic steel targets report no difference in the
penetration compared to the LOS thickness, suggesting this problem doesn’t
apply to the all important APFSDS.7
4
See: Int. J. Impact Engng. Vol. 22, pp. 189—192 (1999) plus Int. J. Impact
Engng. Vol. 17, pp. 263—274 (1995).
5
See Int. J. Impact Engng. Vol. 22, pp. 189—192 (1999), Int. J. Impact
Engng. Vol. 17, pp. 263—274 (1995). Rheinmetall Handbook on Weaponry
[figure 1128] (1982)
6
See: Int. J. Impact Engng. Vol. 19, pp. 811—819 & Shock under Impact IV
pp. 91—101.
7
See: Int. J. Impact Engng. Vol. 23, pp. 771—782.
refers to the ratio of the thickness of the armored plate to that of the attacking
projectile, while lateral confinement refers to the ratio of the diameter of the
attacking projectile to the width of the armored plate. Tests done on armor
material will always yield different results if either the T/d or the lateral
confinement ratios are too low. For modern APFSDS & HEAT, the width
of the plate must be more than 30 times the diameter of the attacking rod / jet
for all results to be stable and transferable to another case for comparison.
Along the main turret walls of a real tank target, this effect is marginal, but
near the mantlet the effect reduces the armored resistance to 0.85—0.9. Fur-
ther, test on ceramic steel targets show the effect is much more dramatic.
Mantlet 1 o’clock 2 o’clock ( Front turret Hit location)
All Steel ~0.88 ~0.96 0.99 ( % reduction in resistance)
Ceramic/Steel ~0.78 ~0.85 ~0.95 (% reduction in resistance)
In all cases, the T/d must be 1.6 times the rod / jet expected penetration.
When this is achieved, the target is said to be ‘a confined semi infinite target’.
Confinement is important because as the shock wave of impact moves
through a target plate it reflects from the ‘free edge’, crosses back over
new waves emanating from the impact point, creating a ‘weakened zone’
through interference. In the case of ceramics and composites, this area is
much larger than steel and is visible in the form of ceramic tile shattering and
composite ‘delamination’.8
The T/d effect starts to diminish rapidly so that after 3:1 its not that much
different than the semi infinite case [3—5% below]. This has its greatest im-
pact on spaced armor. Against such plates the resistance of the plate is re-
duced to 95—60%.9
Lateral confinement has it’s greatest impact in the turret armor on modern
tanks. The gun embrasure area presents a ‘free edge’ which goes a long way
to explaining why most tank turret armor thickens as you approach the
mantlet area. In the past this effect was also responsible for reducing the
strength and resistance of glacis plates around the hatch and MG-port areas.
1.3.1 Steel
Any review of modern armor materials must start with steel. According to
the [American Steel Manufacturers] ASM-96 guide there are literally hun-
dreds of steels in use throughout the world, but only a few qualify as good
armor material.
Firstly the type of steel must be relatively cheap as its still the most common
material used in tank armor accounting for about ½ the weight. In order to
survive the pressure and strain of impact, this steel must be both strong and
ductile. A class of steels -that currently fit the bill- have been developed
called ‘high strength low alloy steel’ [HSLA], and the most common of these
in research papers is ‘Type 4340 steel’. This steel features low carbon [0.3—
0.5%], with moderate manganese content [1—3%] and good ductility [on the
order of 8—10%] and strength [~1.0—1.1 GPa – ultimate tensile strength].
The hardness range from ~250—300 Brinell Hardness Number [BHN, a
rating system for metal hardness] to BHN 350—390. Other steels are avail-
able that are stronger like ‘Maraging Steel’ and harder like ‘Tool Steel’,
but tests reveal these offer only 90% of the resistance of RHA.
8
See: Int. J. Impact Engng. Vol. 19, pp. 49—62.
9
See Int. J. Impact Engng. Vol. 23, pp. 639—649.
Usually Rolled Homogenous Armor [RHA] appears in three forms; armored
steel [RHA], semi hardened steel [SHS] & high hardness steel [HHS]. Ar-
mored steel is about 270—300 BHN. It is most often found in thick armor
and can appear as cast or rolled; all modern tanks feature rolled plate. It ap-
pears that modern cast RHA offer’s only ~90—92% of the resistance of
rolled plate, while WW-II cast can offer anywhere from 90—50% resis-
tance. All Soviet tanks and British tanks feature cast turrets. While the British
tanks and Yugoslav Versions of Russian tanks feature 270BHN cast armor,
Russian sources speculate that their cast is high nickel and therefore harder.
Thinner plates of RHA [several cm] can be machined still at 350—390 BHN
offering 12—18% more resistance than RHA vs. Armor Piercing Fin Stabi-
lized Discarded Sabot [APFSDS ]. The M-1 is reported to feature High Yield
–120 plate that is about 350BHN plate. Semi hardened steel is usually 400—
450 BHN and appears in moderate thickness of several cm and offers a
Thickness Effectiveness [TE]of 1.2 to 1.25, that’s 20—25% more resistance
than RHA. All western Chobham armored tanks feature semi hardened steel
as a part of their layered structure. This steel is harder to weld into the struc-
ture, which limits its use.
High hardness steel is about 500—600BHN and offers about 30—34 % more
resistance than armored steel, but its costly [twice the price of RHA], difficult
to weld, and can only be manufactured in thin rolled plates. Often this armor
has to be bolted on to the main armor wall The Leclerc tank and German
Leopard 1A3 feature this armor layered with RHA and SHS, it’s as-
sumed Leopard 2s also featured triple hardness steel. Layered steel with
250—430 and 515 BHN – as in the Leopard 1A3 – should offer an average
hardness of 18% higher than RHA, but the exact layering should increase this
by 25% to a TE of ~1.5 times the resistance of RHA for dual hardness and
~1.6 for triple hardness armor.
1.3.2.1 Aluminum
The post WW-II period saw a number of special armors developed to enhance
armor resistance to shaped charge warheads including, ERA, aluminum and
ceramic armor to name a few, but all these were to expensive except for alu-
minum. At 1/3 the density of steel, Aluminum was an attractive alternative to
steel especially in the construction of light AFVs. and support vehicles. Un-
fortunately along with the lighter construction comes a corresponding less
resistance, AL5083 [M113; M2/3 and LTVP-7 AFVs] offers only 60% of the
resistance of RHA [vs. API shot]. This type of aluminum is only 2.66 g/cm³
[compared to 7.83 g/cm³ for RHA], and resists corrosion well.
10
See: Int. Defence Review 4/91, pp. 349—352
and is included in the side hull skirting and rear armor of a number of tanks.
The front armor of the Leopard 2 series may feature aluminum.
1.3.2.2 Titanium
An interesting alternative to Aluminum is Titanium, which has a density of
only 4.5 g/cm³ and offers resistance of 80-90% of RHA [APFSDS]. However,
Titanium is many times the price of aluminum which itself is twice as expen-
sive as RHA. Titanium is known to be used in select items of the M-1’s armor
to reduce weight and maybe used in the modern version of BDD armor in
Russian tanks.11
The fuel cells mounted around the driver of the M-1 tank are reported to fea-
ture honeycomb structure to increase resistance in the front hull.12
In addition, Diesel fuel has been shown to be a reasonable armor and by inte-
grating it into the armor, it opens doors to increased protection. To model
fuel cells when estimating armor values, a value between water and
Methanol was used, Methanol has a TE of 0.63 against shaped charges.
Water cells offer a TE resistance of 0.15 vs. APFSDS. While a target of
600mm of water offers the same resistance to shaped charges as 300mm Alu-
minum (which is equivalent to 150mm RHA) suggesting a TE value of 0.15
KE and 0.45 HEAT.13
1.3.4 Composites
Many lightweight materials have also been tested like Fiberglas in an effort to
replace part of the dependence on heavy steel in AFV design. Usually these
composites involve fiber material that is suspended in a medium for rein-
forcement and stiffening. The mediums can be Epoxy, Thermoplastics, Vi-
nylester, Polyester or some Phenolic type material. These also boost the den-
sity of the material and allow it to change from ‘cloth or fabric’ to ‘panels’.14
Steltexolite is a example of a lightweight Russian Fiberglas that uses glass
cloth. It’s known to be used extensively in Russian tank armor. Steltexolites
material compares well with aluminum’ in terms of resistance vs. KE
projectiles and is slightly better vs. shaped charges, this despite the fact
that it is just 2/3 the density of aluminum.15
11
See: Int. J. Impact Engng. Vol. 20, pp. 121—129
12
See: Int. J. Impact Engng Vol. 19 pp. 361—379.
13
See: Int. J. Impact Engng. Vol. 23, pp. 585—595
14
See M. Szymczak in: DREV paper Sept’95.
15
See Int. J. Impact Engng. Vol. 17; pp. 751—762
compared to Fiberglas but comparable figures for APFSDS and HEAT. Not
as good as Steltexolites but lighter at just ¾ of the density, it’s a good solution
as a spall liner. The effect of spall is like a ‘small grenade’ going off inside
the AFV, with the addition of spall liners this is reduced to a ‘shot gun
blast’ [50% reduction in particles and blast cone]. Newer materials like
‘Spectra Shield’ and ‘Dyneema’ achieve the same effect but at 2/3 the weight
of Kevlar. Dyneema is of note as being the liner in German AFVs, and has
comparable resistance to Fiberglas at 1/3 the density.16
1.3.4.2 Ceramics
By far the most common ‘special armor’ studied to increase AFV protection
are ceramics. It’s assumed to be the main component in Chobham armor.
Ceramics are light but very hard materials, over 4 times as hard as the hardest
steel at only half the weight. This combination of light weight and high
hardness offers resistance to KE warheads comparable to RHA and,
more importantly, resistance to shaped charge warheads up to twice the
amount RHA offers. While this makes them good armor material, there are
several drawbacks to the use of ceramics in tank design. Firstly ceramics lack
mechanical strength and can’t be used as support structures. Furthermore, to
be most effective they must be encased in metal, therefore diluting some of
the weight and performance benefit. While the most basic ceramic, Alumina
[AL2O3] is about as expensive as Aluminum or hard steel [twice the price of
RHA], the really mass efficient ceramics can be up to 10 times the cost of
Alumina.
16
See: M. Szymczak in: DREV paper, Sept’95
Vs. HEAT @ 2:1 standoff 1.26 1.38 1.44
Vs. HEAT @ 6:1 standoff 1.32 1.79 1.65
Same target with………
plus rubber target @ 6:1 standoff 1.3 1.8 1.62
plus airgap target @ 6:1 standoff 1.22 1.65 1.72
Resistance of AD-97
Tungsten liner @ 2:1 standoff 1.05 1.1 1.05
dU or Tungsten lined shaped charges seem to offer almost the same penetration into ceramic
steel targets as all steel targets, suggesting they are unaffected by that special armor.
In addition, the spaced plates themselves also help to defeat the shaped charge
by erosion. Test on thin spaced plate’s show that the collapse of the plate
flows into the path of the jet, leading to a large disrupted zone. Since the
jet has little strength, it too is disrupted and the plate will offer a resistance
2—3 times the LOS thickness.
If the spaced plate arrangement is layered, the disrupted zone and shaped
charge loss of penetration is larger. A steel–aluminum–steel arrangement
offers a resistance 7 times the LOS thickness of the plates. The ‘Wedge
armor’ added to the Leopard 2A5 seems to be of this construction with sev-
eral plates of steel, probably of different hardness [triple hardness steel?]18
Sufficiently large enough spaced plates can also offer increase resistance to
kinetic energy attack [APFSDS], increasing plate resistance ~10% as well as
10% for slanted impact.19
If the layer includes an elastic material the plates will bulge at considerable
speed [200—500 m./s], increasing the effectiveness of the plate in much the
same way ERA works (see below). These kind of arrangements could offer
17
See: Int. J. Impact Engng., Vol. 18, pp. 1—22
18
See: Int. J. Engng. Sci. Vol. 20, pp. 947—961
19
See: Int. J. Impact Engng. Vol. 5, pp. 323—331
~10 times the LOS thickness against shaped charges. The Israeli EKKA
armor added to M113 and AAVP–7 are examples of this armor.20
ERA generally works in the following way: A flat layer of explosive is sand-
wiched between two steel plates, mounted some distance from the main armor
wall. When this array is struck by a sufficiently large enough force [HEAT jet
or KE penetrator], the explosive is detonated and the ‘Flyer plates’ are driven
apart. If this impact occurs at angle, the expanding movement of the
plates will cut across the path of the jet or rod, thereby eroding it. When
the rod interacts with the flyer plate, it will suffer enhanced erosion and mag-
nify its yaw resulting in 10—20% loss of penetration per plate. However it
must be noted that since this is tied to the ‘t/d effect’, any change in the rod or
plate thickness will effect the resistance the array can offer.
The effect on HEAT jets is similar to rod shaped projectiles, but since the jet
is already weak the disruption can be massive. In addition, tests on ‘asymmet-
rical sandwiches’ show that even at normal impact the HEAT jet is seriously
disrupt. Thin plates offered 7—10 times the resistance at normal impact.22
20
See: Int. J. Impact Engng. Vol. 21, pp. 294—305
21
See: Int. J. Impact Engng., Vol. 14, pp. 373—383
22
Int. J. Impact Engng. Vol. 23, pp. 795—802
23
„Principle Battle Tank“ pp. 59, Arsenal books & Kontakt 5 Patent
next ‘segment’, in other words: It might be reusable! Since only a maximum
of 2 of the 4-7 plates are ‘flyer plates’, the variation should be only about
±10% [instead of ±30% in K-1] .24
1.3.7 LAYERING
Test of AP shots on various aluminum–steel combinations has revealed that if
the less dense layer is on top, the array offers as much as 15% more resis-
tance than the other way around. Tests on APFSDS seem to show this same
effect. Test on ceramic with backing plates show resistance changes with
the backing material. The Ceramic/Aluminum, offering much less resis-
tance than the same Ceramic mounted on RHA. In addition, the same
ceramic mounted on tungsten plate offers more resistance still. In the case
of aluminum, this is less dense than the ceramic and thus it fits into the above
model. The case of the Tungsten backing is of note due to the possibility that
this might be a key to dU armor effectiveness.
Tungsten offers a TE of 1.44 compared to RHA. But when the ceramic was
mounted on Tungsten, the resistance of the ceramic increased by 33% over
the resistance offered by the Tungsten plate. Looking at it numerically the 1
part ceramic + 2 parts RHA offered 88% of RHA, making the ceramic only
0.75. The 1 part Ceramic +1 part Tungsten target was 1.16 times RHA. But it
should have offered 97% resistance making the combination 11% better. This
implies that the backing material increases the resistance of all the com-
ponents of the array.25
24
See: Zaloga, Steven: „Artillery & Design Practices 1945—present“, pp.
122, 124/125, 147, 436
25
See: Int. J. Impact Engng. Vol. 23, pp. 771—782
armor arrangement. It’s likely that the secret to the generations of dU
armor may be in the impact of high density and high hardness backing
materials.26
26
See: Int. J. Impact Engng. Vol. 17, pp. 409—418 & Int. J. Impact Engng.
Vol. 19, pp. 703—713 (1997)
1.5 Estimates of various tanks’ armor strength
Armor diagram Example
K-5 coverage seems to be about 60%, the T-90 without K-5 looks a lot like the T-72BVwith K-1
1.5.4.1 T-80U
A = 210mm KE & HEAT B = 520mm KE & 570mm HEAT
C = 70–120mm KE & 210 – 260mm HEAT H = 70–120mm KE &~ 210 – 260mm HEAT
D = 400mm KE & 510mm HEAT J =470mm KE & 730mm HEAT
E = 280–290mm KE & 370–410mm HEAT K = 490mm KE & 520mm HEAT
F =110–140mm KE & 180–270mm HEAT L = 480mm KE & 640mm HEAT
G = 60mm KE & 300–400mm HEAT M = 260mm KE & 340mm HEAT
K-5 coverage seems to be about 60%, the T-80 without K-5 looks a lot like the T-72BVwith K-1
K-5 coverage seems to be nearly 100%. Note these estimates assume quantitative not qualitative
increases, the figures should be higher.
1.5.7 General armor description: Challenger-1
This tank was designed with cast turret plus Chobham armor and hard steel
outer cover plates. The front turret thickness of the Challenger seems to range
from 920mm along side the gun, narrowing to 880mm and ~800mm at the
turret corner. Assuming the same volume as the Chieftain, then the weight
increase to Challenger suggests a 13% armor increase overall [should be
54/92; 48/88 & 44/80] suggesting an average density of 4.3-4.6 g/cm³. This
sounds like 1.5 part steel, 2 part Alumina [97%?] and 2 part GRP[average
4.4g/cm³], similar to an armor tested in a RARDE paper27. That’s 0.82 KE
and 1.22 HEAT. A 1991 IDR article reports the Challengers armor was
~1000mm HEAT and a 1985 engineering estimate put the front turret @ >
620mm KE armor.28