Propellants, Explosives, Pyrotechnics 26, 201–208 (2001) 201

The Evaluation of Computational Fluid Dynamics Methods

for Design of Muzzle Blast Suppressors for Firearms
M. Keith Hudson* and Chris Luchini

Graduate Institute of Technology and Department of Applied Science, University of Arkansas at Little Rock, Little Rock,
AR 72204 (USA)

J. Keith Clutter+ and Wei Shyy

Department of Aerospace Engineering, Mechanics & Engineering Science, University of Florida, Gainesville, FL (USA)

Summary pressure levels from small arms firing to lessen the risk of
hearing loss. Interestingly, while suppression for hearing loss
Suppression of muzzle blast is important in both large and small reduction has received some study, there has been little
caliber gun designs. Key goals in the case of small caliber systems are
reported in the open literature over the many years that
the reduction in the incidence of hearing loss due to the acoustic signal
and signature reduction for military applications. Various devices have these devices have seen use. This is most likely due to
been used to reduce the muzzle blast, and the design of these devices strict US regulation of these devices in civilian applications.
has relied heavily on experimental investigation. The current study As in the case of the large caliber suppressors, the design
evaluates the utility of computational models in the design of sup-
pressors for small caliber guns. Experimental measurements are made
process for the suppressors has depended heavily on experi-
for a representative suppressor design and simulations are performed to ments and a cut-and-try procedure. Unlike the large caliber
determine the level of model sophistication needed to correctly predict work, no significant computational effort has been under-
the effects of the device. The current simulations correctly capture both taken. Therefore, the goal of the current study is to determine
the levels and characteristics of the acoustic signal generated by the
bare muzzle and suppressor configurations. These findings support the the applicability of computational tools developed for the
use of computational models in the suppressor design process. large caliber suppressors to the small caliber suppressors. Of
primary concern is the scaling of the blast phenomena and the
identification of the driving physics, which dictates the peak
overpressures and pressure signals. These two factors are key
1. Introduction to the acoustic signature of the suppressor and need to
be captured by any computational code to be used for
Devices for the suppression of overpressures from firearms suppressor design.
have been known and utilized for some time dating back to This report summarizes the initial experimental and com-
the work of Maxim around the turn of the century(1). putational investigation into suppressors for 0.22 caliber=
Currently, suppressors are used on both large and small 5.56 mm and 0.38 caliber=9 mm guns. The experimental
caliber guns for somewhat different purposes. In the case effort tested a commercial suppressor as well as a cylindrical
of large caliber guns, the primary goal of overpressure baffle design used to evaluate the computational code. The
suppression is to reduce the effects of blast on structures remainder of this document first discusses the experimental
and supporting vehicles. The design process of the suppres- details and highlights some of the predominate physical
sion devices has relied heavily on experimental work and the occurrences identified. Next, the computational model is
development of empirical databases(2,3). Some computa- reviewed and the simulations for the cylindrical baffle
tional efforts have been undertaken(4–6) but have been suppressor are presented and discussed. Conclusions are
limited primarily to large caliber gun systems. In the case then drawn as to the utility of computational codes in small
of small caliber guns, suppressors have been widely used as caliber suppressor design and the driving physics behind the
clandestine devices in sniper and other roles in warfare to acoustic signal.
avoid detection of the shooter. While this role has been
widely accepted for many years other applications of sup-
pression are being sought, particularly to reduce the acoustic 2. Experimental Investigation

2.1 Experimental Setup and Description

*Corresponding author; e-mail: [email protected]
+
Currently J. K. Clutter is with Analytical & Computational Firearms suppressor data collection requires that the
Engineering, PO Box 809, Helotes, TX 78023, USA researcher have a sound insulated laboratory with adequate

# WILEY-VCH Verlag GmbH, D-69469 Weinheim, 2001 1040-0397/01/0410–0201 $17.50þ:50=0

202 M. K. Hudson, C. Luchini, J. K. Clutter, and Wei Shyy Propellants, Explosives, Pyrotechnics 26, 201–208 (2001)

backdrop for projectile containment, or have the ability to set Acoustic data are collected using a Bruel and Kjaer 4135
up on an outdoor range that has adequate facilities to support condenser microphone powered by a 2801 power supply.
the planned experiment. A suitable range has been located Calibration data indicated that this unit is accurate to 100 kHz
that offers a sheltered area with utilities, but provides an and provides an output of 3.39 mV=Pa. The microphone is
adequate acoustic environment to make sound measure- positioned upright (pointed up) on a tripod and positioned
ments. All testing has been performed with the instrumenta- between 76 and 508 mm from the muzzle. The firearm is then
tion sheltered from direct sunlight, but with the firearms positioned to a point parallel to the microphone, and then
muzzle and microphone located just outside of the shelter to pulled back up to 254 mm from the microphone to establish a
avoid direct sound reflection effects on the collected data. grid of measurements (Table 1). The microphone is read by a
Figure 1 shows the general layout of the equipment and LeCroy Model 9400A, 175 MHz 8-bit digital storage oscillo-
tested firearm for all experimental trials. The equipment scope. Computer readouts of the sound tracings during firings
used includes a Competitor Corp. 0.38 special caliber are not available so peak data are recorded by hand. If there
(9629 R) action for all nominally 0.38 caliber testing appeared to be two major sound peaks, each peak is recorded.
and a AMT Lightning rifle for all 0.22 caliber LR testing. Measurements from three firings are made at each gage
Both actions have been modified to allow fitting of a position. Firings are made with the bare muzzle in all
commercial suppressor shell, utilizing a GEMTECH positions, followed by a similar set of firings with the
Model Vortec 9 for 0.38 caliber and a Vortec 2 for all suppressor attached. For all experimental firings, the sup-
0.22 caliber LR testing. Barrel length on the Competitor pressors consisted of a right circular cylinder body with one
action is 254 mm while the 0.22 rifle has a length of copper baffle held in place one third of the distance down the
508 mm. The cylindrical baffle suppressor dimensions are suppressor body by aluminum spacers (Figure 3). Limited
given in Figure 2. Handloaded ammunition has been used firing has been carried out using the commercial suppressor
in the 0.38 caliber unit consisting of a 160 grain Speer on the 0.22 caliber rifle in order to show that the cylindrical
jacketed bullet in a 0.38 caliber special casing, over 8.6 suppressor, used in the computational code evaluation study,
grains of Alliance Blue Dot Powder. The 0.22 caliber LR produces similar pressure reductions.
has used commercially available CCI Blazer brand ammu-
nition. During firing, the 0.38 caliber unit is held on a
sandbag, while the 0.22 caliber rifle is shoulder fired in the
normal manner. Care is taken to ensure the same relative
alignment of the pressure gages for each firing. Table 1. Placement of Pressure Gages for the Experiments and
Simulations
Caliber (mm) Gage X (mm) Y (mm)
0.22 5.56 1 0 191
0.22 5.56 2 89 191
0.22 5.56 3 178 191
0.22 5.56 4 0 254
0.22 5.56 5 127 254
0.22 5.56 6 254 254
0.38 9 1 0 254
0.38 9 2 127 254
0.38 9 3 254 254

Figure 1. Experimental layout.

Figure 2. Cylindrical baffle suppressor’s cross-section and specific Figure 3. Picture of the cylindrical baffle suppressor and 0.38 caliber
distances. action used in the tests.
Propellants, Explosives, Pyrotechnics 26, 201–208 (2001) Evaluation of Computational Fluid Dynamics 203

3. Experimental Results and Discussion noise. Also, audibly, the sound is suppressed to a level
where it is not objectionable to the un-protected ear. The
All the experimental measurements are presented in control firings made using the full commercial set of
Table 2 where ‘‘Sup’’ denotes the cylindrical baffled baffles are noted to be very quiet, although still sounding
suppressor and ‘‘Com’’ the commercial suppressor. Oscillo- like a firearm in general. Another distinct acoustic signal
scope traces from the unsuppressed firearms show a single noted during testing is the sonic crack generated by the
high-intensity peak with only minimal ringing type peaks supersonic bullet. This is especially true in the 0.22
over the rest of the measurement period. This of course caliber LR trials.
correlates with the sharp, high-intensity crack heard when a
firearm is discharged. For the positions further from the
muzzle the sound is seen to diminish with distance from the 4. Computational Model
microphone, as would be expected, and the tracing pattern
remains essentially the same except for the overall intensity
changes. 4.1 Governing Equations
Oscilloscope traces for the firearm firings using the
cylindrical baffle suppressor show a characteristic inten- The computational model used for the current study is a
sity spreading. The large single peak seen with the bare finite volume based computational fluid dynamics (CFD)
muzzle is gone, replaced typically by a set of peaks of code developed to aid in the design of gun muzzle devices.
similar intensity, often by two peaks of almost the same The governing equations for the gun blast problem are the
amplitude especially in the data for the 0.38 caliber data. full Navier–Stokes equations for a multi-species chemi-
The values of the two peaks are given in Table 2 and are cally reacting flow. The current study focuses on the
denoted with the 1 and 2 following the suppressor inviscid and real gas aspects of the problem to determine
designation. Also for the suppressor configuration, the their relative role in the generation of the acoustic
smaller peaks, which appear as ringing type peaks in signature. Therefore, the equations to be solved are the
the bare muzzle tests, are relatively larger when compared Euler equations for a multi-species flow with variable
to the peak signals. This is in agreement with the specific heats. When discretized, the equations take the
suppressor acting to ‘‘spread’’ the discharge sound out form
over a larger time scale, minimizing the peak value, but @JQ @JF @JG
giving a longer duration to the overall sound. Audibly, the þ þ þ JH ¼ 0 ð1Þ
@t @x @Z
authors heard this as a change in the characteristics of the
sounds to less of a crack and more of a loud hissing where the dependent variable and flux vectors are

Table 2. Experimental Measured Peak Pressures in Atmospheres for All Configurations

Caliber (mm) Y (mm) X (mm) Bare Sup (1) Sup (2) Com (1) Com (2)
0.22 5.56 191 0 1.0505 1.0074
0.22 5.56 191 89 1.0604 1.0098 1.0125
0.22 5.56 191 89 1.0658 1.0119 1.0113
0.22 5.56 191 89 1.0621 1.0106 1.0135
0.22 5.56 191 178 1.0757 1.0123 1.0169 1.0082 1.0061
0.22 5.56 191 178 1.0749 1.0124 1.0130 1.0071 1.0053
0.22 5.56 191 178 1.0782 1.0124 1.0139
0.22 5.56 254 0 1.0387 1.0048 1.0056
0.22 5.56 254 0 1.0391 1.0055 1.0053
0.22 5.56 254 0 1.0366 1.0051 1.0043
0.22 5.56 254 127 1.0511 1.0092 1.0097
0.22 5.56 254 127 1.0532 1.0087 1.0098
0.22 5.56 254 127 1.0536 1.0075 1.0094
0.22 5.56 254 254 1.0501
0.22 5.56 254 254 1.0583 1.0128
0.22 5.56 254 254 1.0600 1.0109
0.22 5.56 254 254 1.0565 1.0128
0.38 9 254 0 1.1500 1.0300 1.0416
0.38 9 254 0 1.1636 1.0266 1.0281
0.38 9 254 0 1.1670 1.0203 1.0237
0.38 9 254 127 1.1936 1.0479 1.0542
0.38 9 254 127 1.2105 1.0392 1.0532
0.38 9 254 127 1.1922 1.0489 1.0523
0.38 9 254 254 1.1704 1.0610 1.0987
0.38 9 254 254 1.1554 1.0629 1.0799
0.38 9 254 254 1.1626 1.0658 1.0842
204 M. K. Hudson, C. Luchini, J. K. Clutter, and Wei Shyy Propellants, Explosives, Pyrotechnics 26, 201–208 (2001)
2 3 2 3
r rU X
NS
P
6 7 6 7 e¼ ai hi ÿ
6 ru 7 6 ruU þ xx P 7 r
6 7 6 7 i¼1
6 7 6 7 ðT ð5Þ
6 rv 7 6 rvU þ xy P 7
6 7 6 7
6 7 6 7 hi ¼ h0fi þ Cpi dT
6 rE 7 6 UðrE þ PÞ 7
Q¼6 7; F ¼ 6 7;
6 7 6 7 TR
6 ra1 7 6 ra1 U 7
6 7 6 7 where TR is the reference temperature for the gas properties.
6 7 6 7
6 .. 7 6 .. 7 The specific heat, Cpi, of each species is a known function of
6 . 7 6 . 7
4 5 4 5 temperature. The representation of specific heats can vary
raNSÿ1 raNSÿ1 U from assuming they remain constant to a quadratic depend-
2 3 2 3 ence on T. If a high order function is used for Cpi then an
rV rv
6 7 6 7 iterative procedure must be used to extract the temperature in
6 ruV þ Zx P 7 6 ruv 7 each cell at each time level. Here, a compromise between
6 7 6 7
6 7 6 7 efficiency and sophistication is made by representing Cp as
6 rvV þ Zy P 7 6 rv 2 7
6 7 6 7 a linear function of T over the temperature range to be
6 7 1 6 vðrE þ PÞ 7
6
6 V ðrE þ PÞ 7 7 encountered during the simulations. By using the linear
G¼6 7; H ¼ 6 7 ð2Þ
6 7 y6 7 relationship, the temperatures at each point in the field can
6 ra1 V 7 6 ra1 v 7
6 7 6 7 be extracted by solving a simple equation while introducing
6 7 6 7
6 .. 7 6 .. 7 the effects of varying specific heats.
6 . 7 6 . 7
4 5 4 5
raNSÿ1 V raNSÿ1 v
4.3 Fluid Dynamics Operator
The dependent variable ai is the mass fraction of ith
species with the fluid being defined by NS total species. The fluid dynamics aspects of the problem are modeled
Note that the mass fraction of the NSth species is not using an explicit scheme. To maintain second-order accu-
explicitly modeled since the total density is included and racy, the fluid dynamics operator must be second order and
P
NS
the relationship r ¼ rai holds. here a prediction-correction scheme is used of the form(8)
i¼1
The suppressor design to be simulated is axisymmetric and Dt  ð1Þ 
Q ¼ Qn ÿ @x F þ @Z Gð1Þ þ Hi;j
therefore the axisymmetric form of the equations is used, and 2 ð6Þ
 
the effects of the third dimension are included by incorporat- Qnþ1 ¼ Qn ÿ Dt @x Fð2Þ þ @Z Gð2Þ þ Hi;j
ing the source term H.
with
The grid Jacobian J and the contravariant velocities are
defined as @x F ¼ Fiþ12;j ÿ Fiÿ12;j
ð7Þ
J ¼ x x y Z ÿ xZ y x @Z G ¼ Gi;jþ12 ÿ Gi;jÿ12
U ¼ xx u þ xy v ð3Þ and where the superscripts * and n denote the time level at
V ¼ Zx u þ Zy v which the fluxes are computed and the superscripts (1) and
(2) denote the spatial order of the numerical fluxes. Note the
The effects of the projectile are included in the simulation fluxes in x and Z are computed at the cell faces and the
by making a constant velocity assumption and determining at axisymmetric source term is computed based on the cell
each time interval the appropriate location of the projectile. average.
The cells that contain the projectile are identified and an The scheme used to define the inviscid numerical fluxes is
additional source term is added to denote the appropriate the Steger–Warming flux vector splitting algorithm which
volumetric change and impermeable surface boundary con- has been extended to model multi-species flows(9). The flux
ditions are applied to model the projectile’s surface. vector splitting algorithm decomposes the inviscid fluxes
into non-negative (Kþ) and non-positive (Kÿ) components
based on the eigenvalues of the Jacobian A ¼ @F=@Q and
4.2 Gas Properties likewise for G. The split fluxes take the form

The equation of state is derived by assuming the ideal gas K  ¼ l  

1 K1 þ l2 K2 þ l3 K3 ð8Þ
equation is valid for each species and has the form(7) where the eigenvalues are
X
NS
ai 1ÿ  

P ¼ rRu T ð4Þ l
k ¼ lk   lk 
M i
2
i¼1
l1 ¼ bk ð9Þ
During the calculations the temperature must be extracted
l2 ¼ bk þ cjHkj
from the conserved quantity of internal energy using the
relationship l3 ¼ bk ÿ cjHkj
Propellants, Explosives, Pyrotechnics 26, 201–208 (2001) Evaluation of Computational Fluid Dynamics 205

with Note Ii, j, the cell-length, is used to provide weighting for

non-uniform grid spacing. The same extrapolation procedure
yk ¼ k~ x u þ k~ y v is carried out for the fluxes in Z and can be performed on
k ky either the dependent or primitive variables. Previous inves-
k~ x ¼ x ; k~ y ¼ ð10Þ
jHkj jHkj tigations have shown that using primitive variables gives
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi better performance for flows with strong shocks and this is
jHkj ¼ k2x þ k2y the method used here(10).
The split flux components are
2 3 2 3
r r 4.4 Boundary Conditions
6 ru 7 6 rðu  k~ x cÞ 7
6 7 6 7
6  rv  7 6 rðv  k~ y cÞ 7 The present predictive code has been designed to model
6
g ÿ 16r h ÿ c 7 7 1 6 rðh  y cÞ 7
6
2
7 the launch phase of the ballistics problem and therefore, it is
K1 ¼ 6 t gÿ1 7 K2;3 ¼ 6 k 7
g 6 ra1 7 2g 6 ra1 7 assumed that boundary conditions near the muzzle exit are
6 7 6 7
6 .. 7 6 . 7 known. This alleviates the need to recompute the interior
4 . 5 4 .
. 5
ballistics phase for each computation, which reduces the
raNSÿ1 raNSÿ1 computational time. Typical boundary condition information
ð11Þ needed includes temperature, pressure, and velocity time
histories near the muzzle exit as well as the gun propellant
The above formulation gives K ¼ F when k ¼ x and
used. This information can be obtained either from an interior
K ¼ G when k ¼ Z. For the multi-species chemically react-
ballistic code or from experimental measurements. For the
ing flow, c is the frozen speed of sound where c2 ¼ gðP=rÞ
current study, the simulations were carried out in parallel to
and g is the effective specific heat ratio.
the experiments so some assumptions had to be made as to
As indicated in Eq. (7), the fluxes are evaluated at the cell
the boundary conditions. The exact boundary conditions
faces and are either 1st or 2nd order representations. The flux
achieved during the experiments may vary somewhat from
at the face is a function of the states in the neighboring cells
those assumed, however, the relative effects of the muzzle
and can be symbolically represented by
    device should be evident in the simulations.
Fiþ12;j ¼ Fþ QLiþ1;j þ Fÿ QR 1 ð12Þ The particular boundary conditions used for the simula-
iþ ;j 2 2
tions of the 0.38 caliber are a peak pressure of 870 kPa, peak
If a 1st order spatial representation is used, then velocity of 305 mps, and a peak temperature of 1315  C. It is
assumed that all quantities decayed to atmospheric condi-
QLiþ1;j ¼ Qi;j ; QR
iþ1;j ¼ Qiþ1;j tions over a time period of approximately 4 ms. For the 0.22
2 2
caliber, the peak pressure is lowered to 290 kPa but the
To achieve 2nd order accuracy, a MUSCL approach is
remaining variables were kept the same.
used in which cell-center values are extrapolated to the
The simulations presented here model the flow field as a
interfaces(10). Also, to guard against the interpolation intro-
combination of three species, these being the O2 and N2
ducing any nonphysical extremes into the field in the region
found in the ambient air and the gun propellant gas. The
of large gradients, a limiter must be used. The formula for the
properties for oxygen and nitrogen are available in various
neighboring states takes the form
sources(11). The gun propellant is known to be composed
QLiþ1;j ¼ Qi;j þ Fÿ
iþ1;j
primarily of the active agents CO and H2O as well as N2,
2 2
þ
ð13Þ which was assumed to be inert for the purposes of this model,
QR
iþ1;j ¼ Qiþ1;j ÿ Fiþ1;j and to a smaller extent the combustion products H2O and
2 2
CO2 resulting from the interior ballistics process. Therefore,
where the limiting function is
the properties used for the gun propellant (F) are formulated
liþ1;j h i to represent a mixture of CO and H2 and the boundary
Fþiþ12;j
¼ mmod DQþ ÿ
iþ1;j ; DQiþ1;j
2 conditions imposed near the muzzle exit specify the mass
h i ð14Þ
l i;j fraction to be aF ¼ 0.64 and aN2 ¼ 0:36: These assumptions,
Fÿ 1
iþ2;j ¼ mmod DQ ÿ
i;j ; DQ þ
i;j which simplify the gun gas composition, are done to reduce
2
the number of governing equations. Similar processes have
with
ÿ
been used previously with good results even when further
2 Qiþ1;j ÿ Qi;j combustion is included in the modeling(6).
DQþ
i;j ¼
liþ1;j þ li;j
ÿ
ð15Þ
2 Qi;j ÿ Qiÿ1;j
ÿ
DQi;j ¼ 4.5 Results and Discussions
li;j þ liÿ1;j
The only experimental data available for code evaluation
Here, the popular minmod limiter is used where
are the peak pressures measured in the experiments. There-
mmod½X;Y Š ¼ signðXÞ max½0; minðjXj; Y signðXÞފ ð16Þ fore, the only judgment as to the utility of the computational
206 M. K. Hudson, C. Luchini, J. K. Clutter, and Wei Shyy Propellants, Explosives, Pyrotechnics 26, 201–208 (2001)

code that can be made is whether the code correctly simulates

the general effect of the suppressors in reducing the pressure
levels and in turn the acoustic signal. This data can also be
used to determine if the inviscid and real gas effects being
modeled are dominant players in the determination of the
peak pressures and the acoustic signals. The data from the
experiments and simulations are presented with respect to
the gage location. The locations of the gages are given in
Table 1. The distances are measured from the exit of the
muzzle in the cases with no suppressor and from the exit of
the suppressor when it is used.
A comparison between the simulated and measured pres-
sures for the bare muzzle 0.38 caliber is presented in Figure 4
as well as data for the 0.38 caliber with the suppressor
present. The curves denoting the experimental measurements
Figure 5. Simulated pressure time history at gage location 3 for the
are fit to the average of the three firings made for each 0.38 caliber with and without suppressor.
configuration and gage location. During the firing with the
suppressor, two distinct peaks were measured by the gages
and these are denoted wave 1 and wave 2 with wave 1 being
the peak that arrived first. Likewise, the simulations showed
the initial peak to be accompanied by a second peak or
plateau (Figure 5). However, in the simulations the larger of
the two peaks always arrived first where as in the experiments
the larger of the two peaks arrived second. This discrepancy
may result from the assumptions about the projectile flight
velocity since the interaction of the projectile with the
pressure field as it is evolving in the suppressor can effect
the resulting pressures. Previous studies(6) have shown that
neglecting the projectile can affect the predicted overpres-
sures and the same results would be expected if there is error
in the projectile velocity. However, the simulated pressure
levels agree quite well with the experiments and do indeed
convey the effect of the suppressor in reducing the pressure
and in turn the level of sound generated. Figure 6. Comparison of peak pressures from the experiments and the
The peak pressures from the simulation for the 0.22 caliber simulations for the 0.22 caliber with and without suppressor.
case are presented in Figure 6 with the nomenclature the
same as earlier. The first observation is that even though the
simulation captures the trend in overpressures for the bare
muzzle case, the values are lower than those measured at all
gage locations. This indicates either the pressure assumed for
the boundary conditions in the simulation was somewhat
lower than those achieved during the experiments or the
inviscid non-reacting flow model is not capturing some of the
driving physics. It has been shown that including the
chemical reaction processes results in higher overpressures
are seen in simulations for gun blast(6). However, before
adding reaction for the cases in the current study, a closer
assessment of the true boundary conditions should be made.
The simulations of the 0.22 caliber with suppressor do
capture the general trend of the baffle design producing lower
pressures and in turn lower sound levels. However, the
simulated peak pressure values are somewhat larger than
the measurements for gages 3 and 6. Again it is believed that
some of the discrepancy is due to the assumed projectile
velocity but further investigation is needed. Even given these
regions of overpressure, the simulation does capture the
Figure 4. Comparison of peak pressures from the experiments and the effects of the baffle design in reducing the pressure levels
simulations for the 0.38 caliber with and without suppressor. and in turn the sounds generated.
Propellants, Explosives, Pyrotechnics 26, 201–208 (2001) Evaluation of Computational Fluid Dynamics 207

As in the case of the 0.38 caliber with suppressor, the suppressor. The contour levels have been set to highlight the
measurements for the 0.22 caliber also showed a coupling of pressure spectrum around 2.13 kPa (1 atmosphere). Evident
high-pressure peaks. But here the variations in the magni- in the figure is a sequence of pressure pulses emitted from the
tudes were much less and the larger of the two was not always suppressor. Points A, B, and C denote the peaks of the pulses
the peak that arrived first. In the simulations, the predominant where A is a pulse just being emitted while B and C are pulses
peak was followed by a lower peak or plateau in pressure which have traveled outward into the field. If the time
(Figure 7) much as the case for the 0.38 caliber simulations. evolution of the suppressor’s internal flow field is viewed,
As mentioned earlier, when fired the baffle design generates a it is evident that shocks are continuously reflecting off the
sequence of pressure waves emitted from the suppressor with face of the suppressor walls normal to the line of fire. This is
magnitudes larger than the ringing noted in the bare muzzle most likely the driving force behind the pulsating pressure
case. This phenomenon is also seen in the simulations and signature.
can be seen in Figure 8, which shows a pressure contour at
one instance in time during the firing of the 0.22 caliber with
5. Conclusions

Many types of muzzle devices are used to reduce the

overpressures generated when a gun is fired. An example of
these devices is the baffle configuration tested here. Both the
experiments and simulations show such a design reduces the
level of overpressures. The fact that the current simulations
capture these phenomena infers that the reduction in sound
by muzzle devices such as the baffle is due in a large part to
the inviscid aspects of the flow. This as well as the good
comparisons with the measured peak pressures is encoura-
ging to the engineer tasked to design muzzle devices since all
these simulations have been carried out modeling only the
inviscid and real gas aspects of the problem. Further accuracy
can be achieved by including the chemical reactions and
turbulence and this would be required to model muzzle flash.
Also, phenomenon such as suppressor erosion would require
some accounting for the particle loading and heat transfer to
Figure 7. Simulated pressure time history at gage location 2 for the the walls. Any investigation into these phenomena should be
0.22 caliber with and without suppressor. accompanied with a more detailed model of the internal flow
field to include the modeling of the turbulence. Further
investigation is needed to determine to what level these
aspects need to be modeled for engineering applications.
To correct some of the discrepancies identified here in the
prediction of pressure and sound generation, more attention
should be paid to the projectile flight parameters with one
option being to use the simulated pressures on the projectile
to dictate its flight velocity. However, any increase to the
model complexity should be weighed against its robustness
and efficiency for the task at hand. The current study does
show the utility of computational modeling in the design
process of suppressors, which is needed to reduce the
reliability on empirical databases and the expensive cut-
and-try procedure.

6. References

(1) E. C. Ezell, ‘‘Small Arms of the World’’, 12th ed., Barnes and
Noble, New York 1993.
(2) L. Stiefel, ‘‘Gun Propulsion Technology’’, Progress in Astro-
nautics and Aeronautics, Vol. 109, AIAA, Washington D.C.,
Figure 8. Simulated pressure contour at one instant in time for the 1988, pp. 183–259.
0.22 caliber=5.56 mm with suppressor. In (b) the scale has been set to (3) G. Klingenberg and J. M. Heimerl, ‘‘Gun Muzzle Blast and
highlight the pressure spectrum around 0.1 MPa. Pressure peaks are Flash’’, Progress in Astronautics and Aeronautics, Vol. 139,
denoted with A, B, and C. AIAA, Washington D.C., 1992, pp. 197–338.
208 M. K. Hudson, C. Luchini, J. K. Clutter, and Wei Shyy Propellants, Explosives, Pyrotechnics 26, 201–208 (2001)

(4) G. C. Carofano, ‘‘Blast Field Contouring Using Upstream (10) J. S. Shuen, ‘‘Upwind Differencing and LU Factorization for
Venting’’, ARCCB-TR-93009 (1993), US Army Armament Chemical Non-equilibrium Navier-Stokes Equations’’, J. Com-
Research, Development and Engineering Center. putational Physics 99, 233–250 (1992).
(5) G. C. Carofano, ‘‘A Note On The Blast Signature of a Cannon’’, (11) D. R. Stull and H. Prophet, ‘‘JANAF Thermochemical Tables’’,
ARCCB-TR-92014 (1992), US Army Armament Research, NSRDS-NBS 37, (1971).
Development and Engineering Center.
(6) J. K. Clutter, G. Abate, W. Shyy, and C. Segal, ‘‘Study of Fast
Acknowledgements
Transient Flow Phenomenon for Munition Application’’, AIAA
The authors wish to thank Armond Tomany for work in modifying
Paper 96-0829.
the firearms to accept the suppressor units and to Philip H. Dater, MD
(7) J. Anderson, ‘‘Hypersonic and High Temperature Gas Dynam-
of GemTech Division of Gemini Technologies and Antares Technol-
ics’’, McGraw-Hill, New York 1989.
ogies for supplying the suppressors used in this study.
(8) R. J. LeVeque and H. C. Yee, ‘‘A Study of Numerical Methods
for Hyperbolic Conservation Laws with Stiff Source Terms’’,
J. Computational Physics 86, 187–210 (1990).
(9) M. S. Liou, B. Van Leer, and J. S. Shuen, ‘‘Splitting of Inviscid
Flues for Real Gases’’, J. Computational Physics 87, 1–24
(1990). (Received May 21, 2001; Ms 2001=038)