A Study of Projectile Response to Ballistics Environment

by Stephen E. Ray, Michael J. Nusca, and James F. Newill

ARL-TR-3672 November 2005

Approved for public release; distribution is unlimited.

 NOTICES

Disclaimers

The findings in this report are not to be construed as an official Department of the Army position unless
so designated by other authorized documents.

Citation of manufacturer’s or trade names does not constitute an official endorsement or approval of the
use thereof.

Destroy this report when it is no longer needed. Do not return it to the originator.
Army Research Laboratory
Aberdeen Proving Ground, MD 21005-5066

ARL-TR-3672 November 2005

A Study of Projectile Response to Ballistics Environment

Stephen E. Ray
U.S. Army High Performance Computing Research Center

Michael J. Nusca and James F. Newill

Weapons and Materials Research Directorate, ARL

Approved for public release; distribution is unlimited.

 Form Approved
REPORT DOCUMENTATION PAGE OMB No. 0704-0188
Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering
and maintaining the data needed, and completing and reviewing the collection information. Send comments regarding this burden estimate or any other aspect of this collection of information,
including suggestions for reducing the burden, to Department of Defense, Washington Headquarters Services, Directorate for Information Operations and Reports (0704-0188), 1215 Jefferson
Davis Highway, Suite 1204, Arlington, VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to any penalty for failing to
comply with a collection of information if it does not display a currently valid OMB control number.
PLEASE DO NOT RETURN YOUR FORM TO THE ABOVE ADDRESS.
1. REPORT DATE (DD-MM-YYYY) 2. REPORT TYPE 3. DATES COVERED (From - To)
November 2005 Final September 2004–May 2005
4. TITLE AND SUBTITLE 5a. CONTRACT NUMBER
A Study of Projectile Response to Ballistics Environment DAAD19-03-D-001
5b. GRANT NUMBER

5c. PROGRAM ELEMENT NUMBER

6. AUTHOR(S) 5d. PROJECT NUMBER

Stephen E. Ray,* Michael J. Nusca, and James F. Newill 622618H8000
5e. TASK NUMBER

5f. WORK UNIT NUMBER

7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING ORGANIZATION

REPORT NUMBER
U.S. Army Research Laboratory
ATTN: AMSRD-ARL-WM-BD ARL-TR-3672
Aberdeen Proving Ground, MD 21005-5066
9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITOR'S ACRONYM(S)

11. SPONSOR/MONITOR'S REPORT

NUMBER(S)

12. DISTRIBUTION/AVAILABILITY STATEMENT

Approved for public release; distribution is unlimited.

13. SUPPLEMENTARY NOTES

*
U.S. Army High Performance Computing Research Center, Minneapolis, MN
14. ABSTRACT
The U.S. Army Research Laboratory’s interior ballistics code, ARL-NGEN3, is a state-of-the-art modeling tool providing interior
ballistics predictions and flowfield simulations of the ignition, flamespreading, and combustion of solid-propellant charges used to
launch projectiles from guns. Modern solid mechanics codes such as DYNA3D and EPIC can complement the ARL-NGEN3 code by
providing a capability to predict the mechanical response of the projectile to the unsteady, high-pressure environment within the gun
chamber. This report presents results from an ongoing effort to couple the ARL-NGEN3 code to these solid mechanics codes. The
test problem involves a notional gun ammunition concept in which the projectile afterbody “telescopes” or intrudes significantly into
the gun chamber that houses the solid propellant charge. Results from the ARL-NGEN3 simulation are shown, and they indicate that
large pressure waves move radially and, more significantly, axially through the combustion chamber during propellant ignition.
Results from EPIC and DYNA simulations, which utilize the ARL-NGEN3 pressure data and predict material response, are
compared; good agreement between the codes is demonstrated. Results are also presented from studies using the EPIC code for
several different internal projectile configurations.
15. SUBJECT TERMS
gun charges, solid propellant, multiphase flow, material stress
17. LIMITATION 18. NUMBER 19a. NAME OF RESPONSIBLE PERSON
16. SECURITY CLASSIFICATION OF: OF ABSTRACT OF PAGES
Michael J. Nusca
a. REPORT b. ABSTRACT c. THIS PAGE 19b. TELEPHONE NUMBER (Include area code)
UL 24
UNCLASSIFIED UNCLASSIFIED UNCLASSIFIED 410-278-6108
Standard Form 298 (Rev. 8/98) Prescribed by ANSI Std. Z39.18

ii
Contents

List of Figures iv

1. Introduction 1

2. Description of Test Case 2

3. ARL-NGEN3 Results 3

4. EPIC/DYNA3D Comparison 4

5. Further EPIC Results – 1.25-in Wall Projectile 6

6. Initial Rotation of Projectile 7

7. Propellant With Minimal Interface Drag 8

8. The 0.5-in Wall Projectile 8

9. Projectile With Payload 9

10. Summary and Conclusions 11

11. References 12

Distribution List 14

iii
List of Figures

Figure 1. Axisymmetric representation of notional CTA configuration – chamber, projectile

afterbody, and four propellant regions shown. Note that region IV fills the annular
volume (4)..................................................................................................................................2
Figure 2. Geometry of the notional CTA projectile and the surface of the mesh used in the
EPIC simulation (left) and the DYNA3D simulation (right).....................................................3
Figure 3. Pressure-time curves predicted using the ARL-NGEN3 code for geometry shown
in figure 1 – results for the locations labeled “R,” “M,” and “F” along the chamber wall:
granular propellant (left) and concentric wrap propellant in region IV having minimal
axial interphase drag (right). Note that the scales in the right-hand picture are lower than
those on the left (4, 9). ...............................................................................................................3
Figure 4. Total stress computed by EPIC (solid line) and DYNA3D (dashed line) near the
back of the projectile forebody (left) and near the midpoint of the projectile tail (right)..........5
Figure 5. Total stress at 2.4 ms into the gun firing computed by DYNA3D (left) and EPIC
(right), near the time of the highest total stress. The red region is where the total stress
surpassed 90 ksi (621 MPa); the material likely would have failed and the projectile
damaged in that region...............................................................................................................5
Figure 6. Total stress computed by EPIC using the full data set (red line) and the sparse data
set (blue line) near the back of the projectile forebody (left) and near the midpoint of the
projectile tail (right). ..................................................................................................................6
Figure 7. Total stress computed by EPIC for the 1.25-in wall projectile, near the projectile
forebody (left) and in the middle of the projectile tail (right). The blue lines show the
results from the axisymmetric geometry, and the red lines show the results with the
rotated projectile. The stress values did not notably diverge until after 4.5 ms. ......................7
Figure 8. Total stress computed by EPIC for the 1.25-in wall projectile, near the projectile
forebody (left) and in the middle of the projectile tail (right) using the pressure in the left
half of figure 3 (blue) and from a minimal-interphase-drag ARL-NGEN3 simulation
(red). The maximum stress for the latter case is around 150 ksi. .............................................8
Figure 9. Total stress computed by EPIC for the 0.5-in wall projectile, near the projectile
forebody (left) and in the middle of the projectile tail (right). Note the much larger
oscillations and higher peak stress compared to figure 6. .........................................................9
Figure 10. Total stress in the projectile at 2.4 and 4.4 ms into the gun firing. These times
correspond to the two peaks in the plots in figure 9. The red regions are where the stress
has surpassed 200 ksi (1.38 GPa), so it is very unlikely that any conventional metal could
have survived this gun launch....................................................................................................9
Figure 11. Geometry of the 0.5-in wall projectile carrying four disks; the disk colors
correspond to the lines in figure 12. Each disk is held loosely in place by a ridge of steel
on either side. ...........................................................................................................................10
Figure 12. Mass-weighted average of the acceleration of the disks shown in figure 11. The
disks are numbered in order with disk 1 on the left (rearmost disk) and disk 4 on the right
(foremost disk). The line colors correspond to the disk colors in figure 11. ..........................11

iv
1. Introduction

The interior ballistics environment of the case-telescoped ammunition (CTA) is quite

challenging since the gun chamber, normally occupied by the solid propellant charge and a
projectile afterbody of small diameter and significant taper, is redesigned so that fully one-
quarter of the available chamber volume is occupied by a projectile afterbody of large constant
diameter. Concurrently, an acceptable level of projectile muzzle velocity must be maintained so
that the propellant loading density of the gun chamber is quite high. Ignition of this
tightly-packed charge results in axial pressure waves of such significant amplitude and frequency
that the survivability of the projectile must now be called into question. What is needed is an
approach that couples the numerical codes used for charge design and the numerical codes that
model the structural response of the projectile so that the ammunition designer can arrive at a
solution that guarantees the successful launch of a useful payload at a lethal velocity.
Modern interior ballistics (IB) simulation codes are generally capable of modeling the ignition,
flamespreading, and combustion of densely-packed advanced charges, even those that include a
variety of propellant shapes, such as grains, sticks, disks, and concentric wraps. Most IB codes
cannot specifically simulate the CTA configuration since the presence of the projectile afterbody
represents both an axially discontinuous and a numerically significant area change in the gun
chamber. The U.S. Army’s ARL-NGEN3 code is a multidimensional, multiphase computational
fluid dynamics (CFD) code (1–3). The IB flowfield simulation using this code consists of a
discrete component (the propellant) and a continuous component (the gases initially in the
chamber and generated by propellant combustion). To capture both types of components, the
ARL-NGEN3 code uses a uniquely coupled Eulerian-Lagrangian approach (1–3). Previous work
has reported the utility of the ARL-NGEN3 code in simulating the CTA configuration (4)
including the simulation of large amplitude pressure waves around the projectile afterbody and
an unsteady pressure loading of the projectile surface.
While IB codes accurately compute the motion of the projectile, they do so while treating it as a
rigid body. Such codes cannot model the full response of the projectile to the pressure load
imposed on its surface. Vibration and possible damage to or deformation of the projectile are
excluded from IB models. These mechanical effects can be simulated using modern
computational solid mechanics (CSM) codes, such as DYNA3D (5) and EPIC (6). DYNA3D
and EPIC are both explicit finite-element codes for structural mechanics problems. When the
deformation of the projectile is limited to the elastic regime and contact between the projectile
and the gun tube occur at relatively low speeds, the accuracy of both the EPIC and DYNA codes
on such problems has been demonstrated; therefore, one expects both EPIC and DYNA to give
very accurate results in the IB simulation environment.

1
2. Description of Test Case

There are several goals in the first part of this work; among them is a demonstration of the
accuracy, efficiency, and capabilities of the proposed CFD/CSM coupled modeling approach and
a determination that the results from DYNA3D and EPIC codes are comparable. To meet these
goals, a simplified version of the projectile and charge for a notional large-caliber CTA
configuration was defined for this work (4, 7–10). Figure 1 shows the geometry of the projectile
afterbody, or tail, and the gun chamber. The chamber is axisymmetric in shape, with no tapering
or chambrage, and its diameter and initial length are 14 cm (5.5 in) and 84 cm (33.1 in),
respectively. The intruding projectile tail is also axisymmetric in shape, with an outer diameter
of 8 cm (3.1 in) and length of 70 cm (27.6 in); the presence of a projectile boattail (i.e., tapered
afterbody) and/or aerodynamic fins were excluded for simplicity.

Figure 1. Axisymmetric representation of notional CTA configuration –

chamber, projectile afterbody, and four propellant regions shown.
Note that region IV fills the annular volume (4).

The chamber shown in figure 1 was divided into four regions, each containing a different
propellant and/or different propellant geometry. Region I contained 25 g of M1 propellant,
which is initially ignited to start the ARL-NGEN3 simulation, and region II contained 25 g of
black powder. Regions III and IV were filled with seven-perforated JA2 granular propellant to a
loading density of 0.9 g/cm3.
Figure 2 shows the geometry of the projectiles used in the EPIC and DYNA3D simulations. The
codes used slightly different geometries for reasons dealing with the mesh generators used for
each of the codes and that will be further discussed later. The sole geometric difference between
the two projectiles was that the projectile nosetip was truncated for the projectile used in the
DYNA3D simulations. All of the walls of the projectiles were assumed to be 1.25 in (3.175 cm)
thick in the first set of simulations. With the EPIC code, a second set of simulations was run in
which the walls of the projectile were 0.5 in (1.27 cm) thick. The outer geometry of that
projectile was identical to that in the left side of figure 2. The projectiles were assumed to be
made of steel; however, the density of the steel was artificially lowered in order to match the
40-lb projectile weight used in the ARL-NGEN3 simulations.

2
Figure 2. Geometry of the notional CTA projectile and the surface of the mesh used in the EPIC simulation (left)
and the DYNA3D simulation (right).

3. ARL-NGEN3 Results

Detailed results from an axisymmetric ARL-NGEN3 simulation of this problem were previously
published (4, 7–10). The graph on the left of figure 3 shows the pressure from the ARL-NGEN3
simulation at the points labeled “R,” “M,” and “F” in figure 1. Notice that there is a pressure
spike that occurs at different times for the three axial locations. This is an indication of the
strong pressure wave that moves axially through the chamber. The mass of the projectile used in
the ARL-NGEN3 simulation was 18.1 kg (weight of 40 lb). Given that weight limit, the pressure
load shown in the left half of figure 3 was likely strong enough to damage or destroy a projectile
made of any conventional material.

Figure 3. Pressure-time curves predicted using the ARL-NGEN3 code for geometry shown in figure 1 – results for
the locations labeled “R,” “M,” and “F” along the chamber wall: granular propellant (left) and concentric
wrap propellant in region IV having minimal axial interphase drag (right). Note that the scales in the
right-hand picture are lower than those on the left (4, 9).

The pressure waves seen in the left half of figure 3 arose from several factors, including the high
loading density, the geometry of the projectile afterbody, and the resistance to the propellant in
region IV (see figure 1) to the axial flow of the high-pressure gas generated during propellant
ignition. Another simulation from ARL-NGEN3 explored the effects of the latter factor. In this
simulation, the propellant in region IV was assumed to be a concentric wrap propellant with the
same thermodynamic characteristics of seven-perforated JA2 granules but with minimal
resistance to gas flow in the axial direction (9, 10). The line plot on the right in figure 3 shows

3
the pressure at three locations in the combustion chamber. The chamber is seen to pressurize
more evenly as the pressure waves have been damped significantly. The peak pressure is about
one-third less than in the previous case and as a result a projectile made of a conventional
material might be able to survive the launch. The drop in muzzle exit velocity, however, is
noticeable: 1470 m/s in the former case and 1400 m/s in the latter. Detailed ARL-NGEN3 code
results for these simulations, including contour plots of the gas pressure and propellant porosity
fields are available in reference (10).

4. EPIC/DYNA3D Comparison

In a real gun system any amount of plastic deformation of the projectile is unacceptable. So in
the present solid mechanics simulations the projectile material was assumed to respond
elastically throughout the gun-firing simulation. The simulations determined if, when, and
where the projectile deformed plastically, but they did not predict the effects of such
deformation. Limiting the response to the elastic regime simplified the problem, and the results
from DYNA3D and EPIC were expected to agree with each other quite well.
The pressure on the surface of the projectile afterbody computed by ARL-NGEN3 was used as
input by DYNA3D and EPIC to simulate the mechanical response of the projectile. The
geometry of the projectile and the mesh on its surface used in the EPIC and DYNA3D
simulations was shown in figure 2. The fine details of the mesh were not expected to affect the
computed response in any appreciable way. Mesh generators that came packaged with
DYNA3D and with EPIC were used to generate the meshes employed during the respective
simulations. Therefore, while the size and number of elements in the two meshes were close, the
meshes were not identical. Based on the algorithms used in the EPIC mesh generator, removing
the point of the projectile’s nose would hurt the time-step size, while in the other mesh generator,
including the nose would hurt the time-step size. Provided the mass of the projectiles is equal,
the presence or absence of the nose was expected to have very little effect.
A set of simulations was run in order to compare the results from EPIC and DYNA3D; as
previously mentioned, the results should agree very well. The 1.25-in wall projectiles were used
in these cases. In these simulations, a smaller pressure file from ARL-NGEN3 was used. In
order to get this file, the pressure was taken from ARL-NGEN3 every 200 µs, even though
pressure data was available roughly every 0.3 µs; the resulting reduced pressure set was referred
to as the “sparse” pressure set. Using the sparse pressure set removed a potential source of noise
that would make comparisons more difficult. Figure 4 shows the total stress computed by EPIC
and by DYNA3D at two points on the projectile tail, one near the back of the projectile forebody,
and one roughly halfway along the tail. The DYNA3D results were previously presented (4, 8–10).
The meshes used in the simulations were not identical, and so the exact locations where the
stress was measured were slightly different in the EPIC simulations versus the DYNA3D

4
simulations. Nonetheless, the results of the two codes in figure 4 are quite close, agreeing well
on the amplitude, frequency, and phase of the oscillations. Figure 5 shows the total stress at
2.4 ms into the gun firing as computed by EPIC and DYNA3D. Again, the agreement between
the two solid mechanics codes is very good, even with the slight variation in geometry and mesh.
Based on the good agreement and previous validation of the codes, both codes were expected to
generate accurate results for this class of problems. The EPIC code was used to generate all of
the solid mechanics results presented in the remainder of this report.

160 100
DYNA DYNA
140 EPIC 90 EPIC
80
120
70
100 60
80 50

60 40
30
40
20
20
10
0 0
0 2 4 6 8 10 0 2 4 6 8 10
Time (ms) Time (ms)

Figure 4. Total stress computed by EPIC (solid line) and DYNA3D (dashed line) near the back of the projectile
forebody (left) and near the midpoint of the projectile tail (right).

Figure 5. Total stress at 2.4 ms into the gun firing computed by DYNA3D (left) and EPIC (right), near the time of
the highest total stress. The red region is where the total stress surpassed 90 ksi (621 MPa); the material
likely would have failed and the projectile damaged in that region.

The stresses shown in figure 4 show a strong frequency at about 4 kHz and then after 8 ms
another mode starts to be excited with a frequency at about 8 kHz. These are both natural
frequencies of the projectile. Recall that the solid mechanics and fluid dynamics solvers are not
fully coupled. Therefore, there is neither damping of the solid vibrational modes nor shifting of
the projectile’s natural frequencies by the gas.

5
5. Further EPIC Results – 1.25-in Wall Projectile

A simulation was run with EPIC using the full pressure set from ARL-NGEN3. This simulation
was run on a SGI Origin 3800 at the U.S. Army Research Laboratory (ARL) Major Shared
Resource Center. The full data set was too large for EPIC to read into memory, so the simulation
was broken up into several pieces. Figure 6 shows the stress computed at the same locations as
in figure 4 using the two data sets. The peak stress is much higher with the full pressure set, and
the overall response is more oscillatory.
140 250
Full data set Full data set
Sparse data set Sparse data set
120
200
100
150
80

60 100

40
50
20

0 0
0 2 4 6 8 10 0 2 4 6 8 10
Time (ms) Time (ms)

Figure 6. Total stress computed by EPIC using the full data set (red line) and the sparse data set (blue line) near
the back of the projectile forebody (left) and near the midpoint of the projectile tail (right).

As mentioned earlier, the full pressure set was too large for the machine being used. So, the
pressure was taken from every fourth time level in ARL-NGEN3’s data set, and this reduced set
was called the “quarter” data set. This data set did fit within the memory of the SGI Origin
3800. Another EPIC simulation used the quarter pressure set, and the results were compared
with those that came from the EPIC simulation using the full data set, which are shown in
figure 6. The two sets of lines almost perfectly overlaid each other, having no visible
differences, so the results are not shown here. All of the rest of the simulations were performed
using the quarter set.
A fast Fourier transform of the pressure from ARL-NGEN3 was performed in order to identify
the dominant frequencies of the input data. Figure 3 showed line plots demonstrating the trends
in the pressure in the gun chamber. The ARL-NGEN3 code uses explicit time-stepping, so the
time-step size varied during the simulation, and the code wrote out the pressure at every time
step. The pressure was taken every 9.3 µs in order to perform the FFT. The analysis revealed
that the pressure data had a strong mode near 100 Hz and progressively weaker oscillations near
500, 1100, and 1700 Hz. The 100-Hz mode was simply the average pressure rise and fall

6
during the gun firing. The 500-Hz mode was the axially-moving pressure wave that appeared as
the double peak in the left half of figure 3.
The frequencies of the computed results from the simulated launch of the 1.25-in-thick wall
projectile using the quarter pressure set were determined. As expected, there were very strong
signals at 100 and 500 Hz, with slightly weaker signals at 1100 and 1700 Hz. These were the
forced modes coming from the pressure computed by ARL-NGEN3. In addition, there was a
very strong signal at 3.7 kHz, with other peaks at 3.2, 7.4, 8.5, and 22.8 kHz, as well as other
smaller peaks. These frequencies are probably all-natural frequencies of the projectile.

6. Initial Rotation of Projectile

In another simulation, the projectile was initially rotated along the long axis by 1°, i.e., raising
the nose while lowering the tail. In order to accommodate this change, the inner and outer
diameters of the gun tube were slightly increased. Such changes would be expected to affect the
ignition, flamespreading, and combustion during the firing cycle, but that effect was neglected,
and the pressure computed by ARL-NGEN3 using the axisymmetric geometry was used in this
simulation. Figure 7 shows the total stress at the same two locations in the projectile tail as in
figure 6. The projectile began to move about 1.5 ms into the firing cycle, and contact with the
chamber wall was first made shortly thereafter. It is interesting to note that, comparing this case
with the previous (the axisymmetric projectile placement), the computed stress agreed very well
until much later in the firing cycle, about 4.5 ms. After that point, the stress is overall higher in
the case with the initially-rotate projectile.

250 140

120
200
100

150
80

60
100

40
50
20

0 0
0 2 4 6 8 10 0 2 4 6 8 10
Time (ms) Time (ms)

Figure 7. Total stress computed by EPIC for the 1.25-in wall projectile, near the projectile forebody (left) and in the
middle of the projectile tail (right). The blue lines show the results from the axisymmetric geometry, and
the red lines show the results with the rotated projectile. The stress values did not notably diverge until
after 4.5 ms.

7
7. Propellant With Minimal Interface Drag

A desirable feature of the propellant loaded into region IV (see figure 1) is low interphase drag
(resistance to flow) in the axial direction. An ARL-NGEN3 simulation studied the limiting case,
in which the granular propellant in region IV had no interphase drag (9). The computed pressure
looked very similar to that shown in the right half of figure 3, in which, because of the concentric
wraps in region IV, there is nearly no axial interphase drag. The launch of the projectile in the
initial axisymmetric placement was studied using the pressure results from the no-interphase-
drag simulation. Figure 8 shows the total stress from the EPIC simulation along with a repeat of
the computed stress from the case first shown in figure 6. The projectile stress for the no-drag
propellant does not peak until much later and the maximum is lower than for the real propellant
charge. Given the weight limit on the projectile, it would still be a challenge to design a
projectile to survive this environment, but it might be possible.
250 140

120
200
100

150
80

100 60

40
50
20

0 0
0 2 4 6 8 10 0 2 4 6 8 10
Time (ms) Time (ms)

Figure 8. Total stress computed by EPIC for the 1.25-in wall projectile, near the projectile forebody (left) and
in the middle of the projectile tail (right) using the pressure in the left half of figure 3 (blue) and from a
minimal-interphase-drag ARL-NGEN3 simulation (red). The maximum stress for the latter case is
around 150 ksi.

8. The 0.5-in Wall Projectile

A second projectile geometry was used in simulations; in this one, the walls of the projectile are
0.5 in thick. Again, the density of the metal was adjusted so that the mass of the projectile was
18.1 kg. EPIC simulated the launch of this projectile using the quarter pressure set coming from
the ARL-NGEN3 simulation involving granular propellant in region IV (shown in the left picture
of figure 3). Figure 9 shows the total stress at two points along the projectile tail. The maximum

8
total stress in this case was much higher than previously, surpassing 350 ksi (2.4 GPa), beyond
the yield stress of most metals. The stress was also noisier than in the previous cases. This
should all be expected and was caused by the decrease in the wall thickness. Figure 10 shows
the total stress in the projectile at 2.4 and 4.4 ms into the simulation. As can be seen in the
pictures, it is very unlikely that a projectile made of any conventional metal could have survived
the high stresses.
400 250

350

200
300

250
150

200

100
150

100
50

50

0 0
0 2 4 6 8 10 0 2 4 6 8 10
Time (ms) Time (ms)

Figure 9. Total stress computed by EPIC for the 0.5-in wall projectile, near the projectile forebody (left) and in
the middle of the projectile tail (right). Note the much larger oscillations and higher peak stress
compared to figure 6.

Figure 10. Total stress in the projectile at 2.4 and 4.4 ms into the gun firing. These times correspond to the two
peaks in the plots in figure 9. The red regions are where the stress has surpassed 200 ksi (1.38 GPa),
so it is very unlikely that any conventional metal could have survived this gun launch.

9. Projectile With Payload

As previously stated, one of the goals of this work is a demonstration of the capabilities of a
coupled CFD/CSM approach to studying gun launches. One of the capabilities lays in the study
of a payload launched aboard the projectile. The projectiles are hollow and could carry some
type of payload, perhaps an explosive or some type of electronics package; the 0.5-in wall
projectile especially has a lot of available volume. We know from the work that a 0.5-in wall
projectile made of a conventional material probably would not survive the launch. But if the

9
notional projectile were made of some yet-to-be-determined material that met the weight and
strength requirements posed by the pressure load, then the issue becomes what type of loading
the payload would experience. To study this, four disks were placed at different locations inside
of the projectile. Figure 11 shows the inside of the projectile with the disks. All of the disks
were flat circular plates of polycarbonate to represent perhaps a type of electrical component
aboard the projectile, and the deformations of the disk were limited to the elastic regime. The
foremost disk had a diameter of 10.7 cm (4.2 in) and a thickness of 3.9 cm (1.5 in), and the
diameter and thickness of the other disks were 5.2 cm (2.0 in) and 1.3 cm (0.52 in), respectively.

Figure 11. Geometry of the 0.5-in wall projectile carrying four disks; the disk colors
correspond to the lines in figure 12. Each disk is held loosely in place by
a ridge of steel on either side.

In order to study the survivability of the disks shown in figure 11, it could be important to know
the vibrational load and frequencies to which they were subjected. Figure 12 shows the mass-
weighted average of the acceleration of the four disks as a function of time. The foremost disk
(the green line in figure 12) showed the least oscillations, likely caused by the fact that it was by
far the largest of the disks and that the projectile forebody probably vibrated less (in terms of
amplitude) than did the projectile tail. The disk midway along the projectile tail (the red line in
figure 12) showed the most oscillations. Why that happened has not yet been determined, but it
likely was caused by the vibrational modes of the projectile tail that were excited during the gun
launch.
In addition to the magnitude of the acceleration, the frequencies at which the disks vibrate could
also be important, especially if they were electronic components. All of the signals showed
100-, 500-, and 1100-Hz frequencies, all seen in the pressure input. Above that, the disks
vibrated at 2.5, 5.5, 9 (a rather strong signal), and 13 kHz, and at other higher frequencies. Not
all of the disks vibrated with every one of these frequencies. There are methods for reducing the
vibrational load on the disks, but none were studied in this work. For instance, the disks were
allowed to move a little between the steel ridges holding them in place. Other means of securing
the disks in place could also have the effect of reducing the vibrational load transmitted to the
disks.

10
 1.E+06
Disk 2
Disk 1
8.E+05 Disk 3
Disk 4
6.E+05

4.E+05

2.E+05

0.E+00

-2.E+05
0 2 4 6 8 10
Time (ms)

Figure 12. Mass-weighted average of the acceleration of the disks

shown in figure 11. The disks are numbered in order
with disk 1 on the left (rearmost disk) and disk 4 on the
right (foremost disk). The line colors correspond to the
disk colors in figure 11.

10. Summary and Conclusions

The interior ballistics environment of the CTA is quite challenging since the gun chamber,
normally occupied by the solid propellant charge and a projectile afterbody of small diameter
and significant taper, are altered so that fully one-quarter of the available chamber volume is
occupied by a projectile afterbody of large constant diameter. Concurrently, an acceptable level
of projectile muzzle velocity must be maintained so that the loading density of the gun chamber
is quite high. Ignition of this tightly-packed charge results in axial pressure waves of such
significant amplitude and frequency that the survivability of the projectile must now be called
into question. What is needed is an approach that couples the numerical codes used for charge
design (ARL-NGEN3) and the numerical codes that model the structural response of the
projectile (DYNA3D and EPIC) so that the ammunition designer can arrive at a solution that
guarantees the successful launch of a useful payload at a lethal velocity. The current paper has
demonstrated that propellant ignition, flamespreading, and combustion results generated using
the ARL-NGEN3 code, can be as successfully linked to the EPIC code as was previously
demonstrated for the DYNA3D code. Results from EPIC and DYNA3D, as expected, compared
quite well with each other. This modeling effort will continue, with results from ARL-NGEN3
and EPIC being compared with experimental data coming from the work described in (11).

11
11. References

1. Gough, P. S. Formulation of a Next-Generation Interior Ballistic Code; ARL-CR-68; U.S.

Army Research Laboratory: Aberdeen Proving Ground, MD, September 1993.
2. Nusca, M. J.; Gough, P. S. Numerical Model of Multiphase Flows Applied to Solid
Propellant Combustion in Gun Systems; AIAA Paper No. 1998-3695, July 1998.
3. Nusca, M. J.; Conroy, P. J. Multiphase CFD Simulations of Solid Propellant Combustion in
Gun Systems; AIAA Paper No. 2002-1091, January 2002.
4. Nusca, M. J.; Horst, A. W. Progress in Multidimensional, Two-Phase Simulations of
Notional Telescoped-Ammunition Propelling Charge. Proceedings of the 39th JANNAF
Combustion Subcommittee Meeting, Colorado Springs, CO, December 2003; CPIA
Publication JSC CD-25.
5. Whirley, R. G.; Engelmann, B. E. DYNA3D – A Nonlinear, Explicit, Three-Dimensional
Finite Element Code for Solid and Structural Mechanics; UCRL-MA-107254 (Rev. 1);
Lawrence Livermore National Laboratory: Oak Ridge, TN, November 1993.
6. Johnson, G. R.; Stryk, R. A.; Holmquist, T. J.; Beissel, S. R. Numerical Algorithms in a
Lagrangian Hydrocode; WL-TR-1997-7039; Wright Laboratory: Eglin Air Force Base, FL,
June 1997.
7. Horst, A. W. Weapon System Constraints on Advances in Gun Propulsion. Proceedings of
the 5th Joint Classified Bombs/Warheads and Ballistics Symposium, Colorado Springs, CO,
1–2 June 2002.
8. Newill, J. F.; Nusca, M. J.; Horst, A. W. Advances in Coupled Projectile-Dynamics/Interior-
Ballistics Simulations: Coupling the DYNA3D Code and the ARL-NGEN3 Code.
Proceedings of the 21st International Symposium on Ballistics, Adelaide, South Australia,
April 2004.
9. Nusca, M. J.; Horst, A. W.; Newill, J. F. Multidimensional, Two-Phase Simulations of
Notional Telescoped-Ammunition Propelling Charge. Proceedings of the 52nd JANNAF
Propulsion Meeting, Las Vegas, NV, May 2004; CPIA Publication JPM CD-04.
10. Nusca, M. J.; Horst, A. W.; Newill, J. F. Multidimensional, Two-Phase Simulations of
Notional Telescoped-Ammunition Propelling Charge; ARL-TR-3306; U.S. Army Research
Laboratory: Aberdeen Proving Ground, MD, September 2004.

12
11. Williams, A. W.; Colburn, J. W.; Howard, S. L.; Brant, A. L.; Nusca, M. J.; Newill, J. F.
Full-Scale Simulator Studies of a Notional Telescoped-Ammunition Charge. Proceedings of
the 40th JANNAF Combustion Subcommittee Meeting, Charleston, SC, June 2005.

13
NO. OF
COPIES ORGANIZATION

1 DEFENSE TECHNICAL
(PDF INFORMATION CTR
ONLY) DTIC OCA
8725 JOHN J KINGMAN RD
STE 0944
FORT BELVOIR VA 22060-6218

1 US ARMY RSRCH DEV &

ENGRG CMD
SYSTEMS OF SYSTEMS
INTEGRATION
AMSRD SS T
6000 6TH ST STE 100
FORT BELVOIR VA 22060-5608

1 INST FOR ADVNCD TCHNLGY

THE UNIV OF TEXAS
AT AUSTIN
3925 W BRAKER LN
AUSTIN TX 78759-5316

1 DIRECTOR
US ARMY RESEARCH LAB
IMNE ALC IMS
2800 POWDER MILL RD
ADELPHI MD 20783-1197

3 DIRECTOR
US ARMY RESEARCH LAB
AMSRD ARL CI OK TL
2800 POWDER MILL RD
ADELPHI MD 20783-1197

3 DIRECTOR
US ARMY RESEARCH LAB
AMSRD ARL CS IS T
2800 POWDER MILL RD
ADELPHI MD 20783-1197

ABERDEEN PROVING GROUND

1 DIR USARL
AMSRD ARL CI OK TP (BLDG 4600)

14
NO. OF NO. OF
COPIES ORGANIZATION COPIES ORGANIZATION

1 DIR USARL 18 CDR US ARMY ARDEC

AMSRD ARL D D CARLUCCI
J MILLER R CARR
2800 POWDER MILL RD R CIRINCIONE
ADELPHI MD 20783-1197 S EINSTEIN
T GORA
3 DIR USARL J HEDDERICH
AMSRD ARL RO P P HUI
D MANN J LANNON
R SHAW E LOGSDEN
TECH LIB P LU
PO BOX 12211 B MACHAK
RESEARCH TRIANGLE PARK NC S NICHOLICH
27709-2211 P OREILLY
J OREILLY
8 US ARMY AVIATN & MSLE CMD J RUTKOWSKI
W CHEW A SABASTO
C DOLBEER J SHIN
JS LILLY R SURAPANENI
M LYON PICATINNY ARSENAL NJ 07806-5000
JM FISHER
BP MARSH 1 COMMANDER
RS MICHAELS US ARMY NGIC
D THOMPSON AMXST MC 3
REDSTONE ARSENAL AL 35898-5249 220 SEVENTH ST NE
CHARLOTTESVILLE VA 22901-5396
2 PM MAS
SFAE AMO MAS 1 COMMANDANT
LTC M BUTLER USAFC&S
PICATINNY ARSENAL NJ 07806-5000 ATSF CN
P GROSS
2 PM CAS FT SILL OK 73503-5600
SFAE AMO CAS
PICATINNY ARSENAL NJ 07806-5000 2 CDR NAVAL RSRCH LAB
TECH LIB
7 DIR BENET WEAPONS LAB J BORIS
M AUDINO WASHINGTON DC 20375-5000
R DILLON
R FISCELLA 1 OFFICE OF NAVAL RSRCH
R HASENBEIN J GOLDWASSER
E KATHE 875 N RANDOLPH ST RM 653
K MINER ARLINGTON VA 22203-1927
S SOPOK
WATERVLIET NY 12189-4050 8 CDR
NAVAL SURFACE WARFARE CENTER
1 COMMANDER J CONSAGA
RADFORD ARMY AMMO PLANT R DOHERTY
SMCAR QA HI LIB C GOTZMER
RADFORD VA 24141-0298 S MITCHELL
S PETERS
T C SMITH
C WALSH
TECH LIB
INDIAN HEAD MD 20640-5000

15
NO. OF NO. OF
COPIES ORGANIZATION COPIES ORGANIZATION

4 CDR 2 BATTELLE COLUMBUS LAB

NAVAL SURFACE WARFARE CTR TWSTIAC
J FRAYSEE V LEVIN
R FRANCIS 505 KING AVE
T TSCHIRN COLUMBUS OH 43201-2693
TECH LIB
DAHLGREN VA 22448-5000 1 BATTELLE PNL
M GARNICH
3 CDR PO BOX 999
NAVAL AIR WARFARE CTR RICHLAND WA 99352
A ATWOOD
S BLASHILL 1 BATTELLE EAST SCI & TECH
T PARR A ELLIS
CH1NA LAKE CA 93555-6001 1204 TECHNOLOGY DR
ABERDEEN MD 21001-1228
1 AIR FORCE RESH LAB
MNME EN MAT BR 2 JHU CHEM PROP INFO AGNCY
B WILSON W HUFFERD
2306 PERIMETER RD R FRY
EGLIN AFB FL 32542-5910 10630 LITTLE PATUXENT PKWY
STE 202
1 AIR FORCE OFC OF SCI RSRCH COLUMBIA MD 21044-3200
M BERMAN
875 N RANDOLPH ST 1 OUSD (AT&L)/STRAT & TACT
SUITE 235 RM 3112 SYS MUNITIONS
ARLINGTON VA 22203-1768 T MELITA
3090 DEFENSE PENTAGON
1 NASA LANGLEY RSRCH CTR RM 3B1060
D BUSHNELL WASHINGTON DC 20301-3090
MS 110
HAMPTON VA 23681-2199 1 BRIGHAM YOUNG UNIV
M BECKSTEAD
1 DIR SANDIA NATL LABS DEPT OF CHEMICAL ENGRG
M BAER DEPT 1512 PROVO UT 84601
PO BOX 5800
ALBUQUERQUE NM 87185 1 CALIF INSTITUTE OF TECHLGY
F E C CULICK
1 DIR SANDIA NATL LABS 204 KARMAN LAB
R CARUNG MS 301 46
COMBUSTION RSRCH FACILITY 1201 E CALIFORNIA ST
LIVERMORE CA 94551-0469 PASADENA CA 91109

2 DIR LAWRENCE LIVERMORE NL 2 UNIV OF ILLINOIS

L FRIED DEPT OF MECH INDUSTRY
M MURPHY ENGINEERING
PO BOX 808 H KRIER
LIVERMORE CA 94550-0622 R BEDDINI
144 MEB 1206 N GREEN ST
1 CENTRAL INTELLIGENCE AGNCY URBANA IL 61801-2978
J BACKOFEN
RM 4PO7 NHB
WASHINGTON DC 20505

16
NO. OF NO. OF
COPIES ORGANIZATION COPIES ORGANIZATION

5 PENNSYLVANIA STATE UNIV 1 BAE ARMAMENT SYS DIV

DEPT OF MECHANICAL ENGRG J DYVIK
K KUO 4800 E RIVER RD
T LITZINGER MINNEAPOLIS, MN 55421-1498
G SETTLES
S THYNELL 2 GEN DYNAMICS ORD/TACT SYS
V YANG N HYLTON
UNIVERSITY PARK PA 16802-7501 J BUZZETT
10101 DR M L KING ST N
1 ARROW TECHLGY ASSOC INC ST PETERSBURG FL 33716
1233 SHELBURNE RD D 8
SOUTH BURLINGTON VT 05403 3 GENERAL DYNAMICS ST MARKS
J DRUMMOND
2 ATK TACTICAL SYSTEMS H RAINES
W B WALKUP D W WORTHINGTON
T F FARABAUGH PO BOX 222
210 STATE RTE 956 SAINT MARKS FL 32355-0222
ROCKET CENTER WV 26726-3548
1 GENERAL DYNAMICS ARM SYS
1 ALLEGHENY BALLISTICS LAB J TALLEY
PO BOX 210 128 LAKESIDE AVE
ROCKET CENTER WV 26726 BURLINGTON VT 05401

1 ALLIANT TECH SYSTEMS INC 1 HICKS AND ASSOCIATES SAIC

C CANDLAND MN07-LW54 I MAY
5050 LINCOLN DR 7990 SCIENCE APPLIC CT
EDINA MN 55436 VIENNA VA 22182

3 ATK AMMO & ENERGETICS 1 PAUL GOUGH ASSOC INC

D A WORRELL P S GOUGH
W J WORRELL 1048 SOUTH ST
S RITCHIE PORTSMOUTH NH 03801-5423
RADFORD ARMY AMMO PLANT
RT 114 PO BOX 1 2 VERITAY TECHGY INC
RADFORD VA 24141-0299 R SALIZONI
J BARNES
2 ATK THIOKOL 4845 MILLERSPORT HWY
P BRAITHWAITE EAST AMHERST NY 14501-0305
R WARDLE
PO BOX 707 1 SRI INTERNATIONAL
BRIGHAM CITY UT 84302-0707 PROPULSION SCIENCES DIV
TECH LIB
1 ATK ELKTON 333 RAVENWOOD AVE
J HARTWELL MENLO PARK CA 94025-3493
PO BOX 241
ELKTON MD 21921-0241 1 SAIC
W WAESCHE
1410 SPRING HILL RD
SUITE 400
MCLEAN VA 22102

17
NO. OF NO. OF
COPIES ORGANIZATION COPIES ORGANIZATION

ABERDEEN PROVING GROUND AMSRD ARL WM TB

P BAKER
1 CDR
USAATC
CSTE DTC AT SL
APG MD 21005

42 DIR USARL
AMSRD ARL WM
W CIEPIELA
AMSRD ARL WM B
J B MORRIS
AMSRD ARL WM BA
D LYON
T KOGLER
AMSRD ARL WM BD
W R ANDERSON
R A BEYER
A L BRANT
S W BUNTE
T P COFFEE
J COLBURN
P J CONROY
B E FORCH
B E HOMAN
AW HORST
S L HOWARD
P J KASTE
A J KOTLAR
C LEVERITT
R LIEB
K L MCNESBY
M MCQUAID
A W MIZIOLEK
J A NEWBERRY
M J NUSCA (6 CPS)
R A PESCE RODRIGUEZ
G P REEVES
B M RICE
R C SAUSA
J SCHMIDT
A W WILLIAMS
AMSRD ARL WM BC
P PLOSTINS
AMSRD ARL WM EG
E SCHMIDT
AMSRD ARL WM M
S MCKNIGHT
AMSRD ARL WM SG
T ROSENBERGER
AMSRD ARL WM T
B BURNS
M ZOLTOSKI

18