Mechmi~ ~ C o m m u n i c a l i o m , Vol. 27,No. 2, pp.

185-189, 2000
Cog~i~ 0 2~0 Ehe,~r ~ i m ~ Ltd
PflmM in the USA. All fil~ t~se~ved
0093-~I 3,~}0~-xe fiem matter
~n: see~4~D(oe)emm-x

ANALYTICAL SOLUTION FOR PENETRATION BY RIGID CONICAL 1MPACTORS

USING CAVITY EXPANSION MODELS

G. Ben-Dor, A. Dubinsky and T. Elperin

Depamnent of Mechanical Engineering, The Peaflstone Center for Aeronautical Engineering,
Ben-Gurion University of the Negev, P. O. Box 653, Beer-Sheva 84105, Israel

(Received 10 June 1998; acceptedfor print 31 January 2000)

~tnxhic~

The cylindrical cavity expansion approximation allows to determine the explicit analytical
expressions for the forces acting on a penetrating impactor at each location of a target-imlmctor
contact surface [1-3]. However in order to determine penetration characteristics (the balliatic limit
velocity, the residual velocity, the depth of the penetration) it is necessary to solve an equation of
motion of an impactor with appropriate initial! conditions. At this stage of the solution additional
simplifying assumptions (e.g., simplified deso!ption of the incomplete imanersion of the impactor
into a target and/or assumption about a constant velocity in calculating the target twtistance) are
introduced in order to obtain analytical formulae. In this study the analytical formulae for
penetration characteristics are derived without Using these.simplifying assumptions.

13eJu.-rimionof the model

Consider a normal penetration of a rigi d conical-nosed impactor of mass M into a target. All,the
notations are shown in Fig. 1. The coordinate h is defined as the distance of the nose of the
in-q~'tor from the front surface of the target. The coordinate ~ is associated with the target. Only
the nose of the impactor with the length L interacts with the target, and the impactor may have also
the cyfinthical part with the length l, r0 is the impactor's base radii and 0 is the half angle of the
apex of the cone. We take into account that for incomplete immersion of the impactor into a target
only a part of the impactor interacts with the target. This part of the lateral surface of the nose of the
impactor is located between the cross-sectionS x = x I and x = x 2 (see Fig. 1) where (for details
see [4])

0if0~h<b I hif0<h<_l
(1)
Ya(h)=[ h - b ifb < h < b + l ' x2(h) = [1 if 1_<h < b + l
185
186 G. BEN-DOR, A. DUBINSKY and T. ELPERIN

for a target with a finite thickness, and

~l(h) = 0 , ~2(h)= { h i f i0f<hh><l l l (2)

for a semi-infinite target, gl = Xl/L, x2 = x2/L, K = h/L, b = b/L.

l ~ ro
-J~1
0

! x2 ;- - ' R - - T
~b

--i /

.... ,L . . . . . . . . __]"
0 0 0
Fig. 1. Coordinated and notations.

The impactor-target interaction is described by the following solution of the problem of hole's
expansion from the time t---0 [1-3]:

p = a21~2 + alRR + a 0 (3)

where R is the radius of the hole, p is the pressure applied in the normal direction at the part of the
impactor's surface, coefficients aj (j = 0,1,2) depend on the properties of the material of the target,
and formulas for these coefficients can be found in the above cited studies. Since (see Fig. 1)
R = Ix, 1~ = l~h, R = ~h, x = h - ~, 13= tan O, Eq. (3) can be written as follows:
 PENETRATION OF RIGID IMPACTORS 187

p = a2~21~2 + al~2Xh + a 0 (4)

The force dF acting at the surface element (IS of the impactor at a given location at the surface of
the impactor which is in contact with the target earl be deacfibed as dF = (pfi + kl~0)(iS, where fi0
and ~0are the inner normal and tangent unit vectors at a given location at the surface of the impactor,
respectively, k is the friction coefficient. The total force F is determined by integrating the local
force over the impactor-target contact surface S. The drag force D is directed in the opposite
direction to the projection of the total force on the direction of the motion of the impactor h:

~(h)
D---OIId:2'IX I pxdx (5)
S x,(h)

Using the variable ~12 = v 2 = w(h) the equation of motion of the impactor Md2h/dt 2 = - D can be
written as follows:

[A0 +AiJ2(h)]-~ +[A2w+l]Jl(~)=0 (6)

where v is the velocity of the impactor, ~ = x/L,

(7)
x, (B) B+ 1

M al~2 >0, A 2 -- ~a2~2 (8)

A 0 = 4~a0~( ~ + k)L3 > 0, A! = 2a0 >a0
0,

Tar2et with a finite thickness

If an initial (impact) velocity Vin = v(0) is considered as an initial condition, the solution of Eq. (6)
can be written as follows:

~(Vin,V)= (h) (9)

wh~'~
 188 G. BEN-DOR, A. DUBINSKY and T. ELPERIN

V2

~F(vm,v) =
dw J (z)dz
(10)
fvJzA2w---+1' v%hJ : ~ A0 + A1J2(z )

If v = Vres for h = b + 1, we arrive at the relation between the initial velocity Vin and the residual
velocity vies, ~F(vin, Vies) = (b + 1). The ballistic limit velocity v,, i.e., the initial velocity which
corresponds to zero residual velocity, is determined from the ~uation ~(v,,0) = (b + 1).

Omitting the mathematical details, we present formulae which can be derived from the above
equations:

2 2 __. B / - ~ ,
v-~-~- B v-~2 = 1, v, (11)
v~ v~ ~/ A2

wh~-~,'e

B = exp[A2(b + 1)], (b + 1)= 3Gg2/(2AI) (12)

1~(g)- ~[(1-b)g] if b <1

G= I z(g)+(~_l)gif~> l , g=[(3A0/AI)+I] -I/3 (13)

i dq 1 . (2z+l'~ l~n (z-l)2 rt

g(z)=~l_g, - - ~ - a r c t a n ~ , - - ~ j - g z2--~z+ 1 6"~ (14)

- ,mi:iaflaim l

Equation (9) allows to find the depth of penetration of an impactor into a semi-infinite target h, for
a given initial (impact) velocity Van. Substituting v = 0 and h = h, in Eq. (9) yields the following
equation:

'l'(vm,0) =

where tIS and are determined by Eq. (10) taking into account Eq. (2). Since the function ~(h,)
increases from 0 to ~, when h, increases from 0 to ~, Eq. 05) has a unique solution for every
Vin > 0. Omitting the algebraic manipulations we present only the final formulas:
 PENETRATION OF RIGID IMPACTORS 189

h'=f /e[(A2v2n + 1)2Al/A2 -- 1~1/3 if Vin < V!

1+ 2Al(-e+l)I"~-I
l n ( A 2CA2
v i 2 n 3, +l)-~llln~--~)lif Vin -->V1 (16)

wh~

_ fff e + I~A2/(2AI) _ I~/A2 3A0

, e = ~ (17)
e / A1

The above formulae take into account two stages of penetration. In the first, initial stage, the
impactor's nose only partially penetrates into the target, and the penetration resistance depends on
the depth of penetration. If the impact velocity is relatively small, the penetration terminates at this
stage. If the impact velocity is sufficiently large for penetration to a depth more than L (the second
stage), the penetration resistance becomes constant when h > L.

.C,~cAidtng.ttm~

Some known phenomenological localized interaction models [5-7] can be described by Eq. (3) with
a I = O. The obtained above formulas can be adopted for this ease by passing to a limit A 1 --->O.

References

[I]. Kh. A. Rakhmatulin, A. Ya. Sagomonian and N. A. Alekseev Soils Dynamics [Voprosy
Dinamiki Gruntov]. Moscow Univ. Publ. House (1964). - In Russian.
[2]. A. Ya. Sagomonian, Vestnik Moskovskogo Universiteta, Seriya Matematika, Mekhanika,
No. 6, p. 104 (1975). - In Russian
[3]. M.J. Forrestal, F. P. Norwood and D. B, Longope, Ira. J. Solids Structures, 17,
p. 515 (1981).
[4]. G. Ben-Dot, A. Dubinsky andT. Elperin, Int. J. oflmpactEng., 21 (I-2), p. 51 (1998).
[5]. R.F. Recht, High Velocity Impact Dynamics: Analytical Modeling of Plate Penetration
Dynamics. In: Zukas, J. A., Ed. High Velocity Impact Dynamics. New York: Wiley
Interscience Publ., Chap. 7 (1990).
[6]. G. Ben-Dot, A. Dubinsky and T. Elperin, Theoret. andAppl. Fract. Mech., 26,
p. 193 (1997).
[7]. A.I. Bunimovich and A. V. Dubinsky, Mathemoaical Models and Methods of Localized
Interaction Theory, World Sci. Publ (t995).