UNCLASSIFIED

AD NUMBER
AD909325

NEW LIMITATION CHANGE

TO
Approved for public release, distribution
unlimited

FROM
Distribution authorized to U.S. Gov't.
agencies only; Test and evaluation; 26 Apr
1973. Other requests shall be referred to
Director, USA Ballistic Research
Laboratories, Attn: AMXBR-XM-SE, Aberdeen
Proving Ground, MD 21005.

AUTHORITY

USAARDC ltr, 8 Mar 1978

THIS PAGE IS UNCLASSIFIED

THIS REPORT HAS BEEN DELIMITED

AND CLEARED FCR PlUBLIC REI-EASF.

UNDER DOD DIRECTIVE 5200,20 AND
MO RESTRICTIONS ARE IPPOS.Er UPON'
ITS USE AND) DISCLOSURE;

DISTRIBUT(ON STATEMENT A

APPROVED FOP PUBLIC RELEASE;

DISTRIBUTION UNLIMITED,
 MEMORANDUM REPORT NO. 2276

MUZZLE DEVICES, A STATE-OF-THE-ART SURVEY

VOLUME I: HARDWARE STUDY

by
SD'.D c
EdadM. Schmidt F)~~r
'APR? 26 1973

February 1973 C

Distribution limited to US Government agencies only

Other requests for this document must be referred toUl*
DiSeztor, USA Ballistic Research Laboratories, ATTN:
AMXBR-XM-SE, Aberdeen Proving Ground, Maryland 21005.

USA BALLISTIC RESEARCH LABORATORIES

ABERDEEN PROVING GROUND, MARYLAND
Destroy this report when it is no longer needed.
Do not return it to the originator.

Secondary distribution of this report by originating or

sponsoring activity is prohibited.
Additional copies of this report may be purchased from
the U.S. Department of Commerce, National Technical
Information Service, Springfield, Virginia 22151

The findings in this report are not to be construed as

an official Department of the Army position, unless
so designatrd by othur audiorized documents.
BALLISTIC RESEARCH LABORATORIES

MEMORANDUM REPORT NO. 2276

FEBRUARY 1973

MUZZLE DEVICES, A STATE-OF-THE-ART SURVEY.

VOLUME I: HARDWARE STUDY

Edward M. Schmidt

Exterior Ballistics Laboratory

Distribution limited to US Government agencies only.

Other requests for this document must be referred to4
Director, USA Ballistic Research Laboratories, ATTN:"I%
AMXBR-XM-SE, Aberdeen Proving Ground, Maryland 21005.

RDT&E Project No. 1J562604A607

AE6ERDEEN PROVING GROUND, MARYLAND

 BALL I STI C RESEARCH LABOR ATORI ES

MEMORANDUM REPORT NO, 2276

EMSchmi dt/so
Aberdeen Proving Ground, Md.
February 1973

MUZZLE DEVICES, A STATE-OF-THE-ART SURVEY.

VOLUME I: HARDWJARE STUDY

ABSTRACT

A review of the salient literature addressing the engineering

design of muzzle devices is presented. Both theoretical and experi-
mental techniques applicable to specific hardware items are discussed.
The types of devices considered include: muzzle brakes, compensators,
blast deflectors, blast suppressors, and flash suppressors. The second
volume of this report addresses specific gas dynamic theories which are
applicable, but, in general, have not been utilized in the analysis of
muzzle gas flow fields.

3
 TABLE OF CONTENTS
Page

ABSTRACT .............. .......................... 3

LIST OF ILLUSTRATIONS .......... ................... 7
LIST OF SYMBOLS ........... ...................... 11
I. INTRODUCTION ............ ........................ 13
I. MUZZLE GAS FLOW ........... ...................... 13
III MUZZLE DEVICES: THEORY .......... .................. 19
A. Muzzle Brakes, Compensators and Blast Deflectors . . . . 20
B. Flash Suppressors ...... ................... .. 32
C. Blast Suppressors ....... ................... .. 37
IV. MUZZLE DEVICES: EXPERIMENTAL TECHNIQUES .............. 45
A. Muzzle Brakes, Compensators, and Blast Suppressors . . 46
B. Flash Suppressors ....... ................... .. 49
C. Blast Suppressors ....... ................... .. 52

V. CONCLUSIONS .......... ........................ .. 54

REFERENCES .......... ......................... 101
DISTRIBUTION LIST .......... ..................... 109

5
 LIST OF ILLUSTRATIONS
Figure Page
1. Formation of Precursor Blast and Free Jet .... ........ 55

2. Photograph of Precursor Flow Field ..... ........... 56

3. (a,b,c, and d) Schematic of Propellant Gas
Ejection Flow ........ ..................... 57
4. (a,bc, and d) Photograph of Propellant Gas
Ejection Flow ........ ..................... 61
5. Schematic of Wave Structure for an Underexpanded,
Steady Jet ........ ...................... ... 65

6. Plot of Deflection Angle Versus Pressure Ratio .... ..... 66

I
7. Plot of Deflection Angle Versus Pressure Ratio for
Various Ye ........ ...................... ... 67
8. Effect of Pressure Ratio Upon Free Jet Structure . ... 68
9. Categories of Muzzle Devices ...... .............. 69
10. Rateau Muzzle Brake ....... ................... 70
11. Rateau's Calculation of Force on 75mm Gun
Muzzle Brake ........ ..................... 71
12. German Double Baffle Muzzle Brake ..... ............ 72

13. Oswatitsch Brake Nomenclature ................. 73

14. Optimization Plot for w ..... ................. 74

15. Plot of a vs ae for Zero Projectile Hole Loss ...... ... 74

16. Comparison of Theory and Experiment for Oswatitsch

Brake ......... ......................... 75

17. Smith Muzzle Gas Flow Nomenclature ............. .... 76

18. Smith Brake Theory Results ..... ................ 76

19. PressurE Distribution on Brake Surface in a Steady

Jet .......... .......................... 77

20. Flash Phenomena ........ ..................... 78

7
Figure Page

21. Flash Preventive Devices ...... ................. 79

22. Watling Flash Suppressor Tehs .... .............. .. 80

Z3 .Supersonic Diffuser Silencer ...... ............... 81

24. Blast Suppressor Classification ...... ............. 82

25. Furrer Explosive Blast Measurements ... ........... .. 83

26. The Effect of Muzzle Brakes on Blast Pressures ........ 83

27. Westine's Universal Gun Blast Plot ................. 84

28. Correlation of Overpressure Factor for Various

Weapons ...... ... ........................ .. 85

29. Universal Impulse Plot and Correlation Plot .... ....... 86

30. Time of Blast Arrival ...... .................. .. 87

31. Bixler Blast Suppressor Model .... .............. .. 88

32. Sequence of Measured Blasts ...... ............... 89

33. Oswatitsch 2-D Measured Blasts .... .............. 90

34. Efficiency Factor, Theory and Experiment .... ......... 91

35. Efficiency Factor of Extended Brake ......... ....... 91

36. Axisyrnmetric Jet With Baffle ...... ............... 92

37. Cover Types and Effect of Efficiency .... ........... 92

38. Slade Recoil Measurements ..... ................ .. 93

39. Slade Water Table ....... .................... .. 94

40. Slade Optimal Turning Vane ....... ................ 94

41. Smith Muzzle Brake Rig ....... .................. 95

42. Aerodynamic index Versus Pressure Ratio ......... 96

43. Aerodynamic Index for Multibaffle Brakes .... ......... 96

44. Smith's Overpressure Measurements ..... ............ 97

8
Figure Page

45. Concentration Schematic ..... ... ................. 98

46, MRI Mach Numnber Measurements and Comparison With
Theory ......... ......................... .98
47. Relative Time and intensity of Flash .... ........... 99
48. Blast Suppressor Test Rig ......... ................ 100

49. Attention of Constant Length Suppressor .... ......... 100

9
 LIST OF SYMBOLS

A area

c gun caliber

D diameter

E propellant energy

F force

G mass flux parameter

I impulse

J momentum flux narameter

gun barrel length

m mass flux

M Mach number

p static pressure

Ps stagnation pressure

Q charge weight

r radial distance

R gas ccnstant

t time

T temperature

v velocity

We mean brake exit velocity

W Oswatitsch momentum parameter

co-volume
ratio of specific heats

ni Smith efficiency factor

I1
 LIST OF SYMBOLS (continued)
0 free ,let turning angle

p density
Oswatitsch efficiency factor

Superscripts
mean property
S)* sonic quantity

Subscripts

)e exit quantity

)m muzzle quantity

S)o side-on quantity

condition at infinity (ambient)

12
 I. INTRODUCTION

Muzzle devices have been used to control the trajectory of pro-

jectiles, the motion of the recoiling gun, and the observable effects
of the sudden release of the propellant energy. The shotgun choke is
an example of a device which controls projectile ballistics through
muzzle contour. in the mid-nineteenth century, the French attempted to
control recoil by redirecting propellant gases rearward through a series
of holes drilled in the gun barrel near the muzzle. Silencers and blast
suppressors began appearing around the turn of the century. The Maxim
multi-baffle silencer developed during this time frame is still one of
the most effective designs. flash hiders received considerable atten-
tion during the First World War; however, the most effective devices
namely, bar-type flash suppressors. did not make an appearance until
1949. At this time, the Franklin Institute accidently d~scovered the
flash suppression capabilities of bars or prongs placed in the muzzle
flow regime.

While the construction and use of muzzle devices has proceeded at

a brisk pace, the development of an insight into tht physical phenomena
occuring within the devices has lagged behind. There are no strong
Stheoretical analyses available which perivit the design of devices 'or a
large range of weaponry. What techniques are available are generally
empirically based and highly restrictive in nature. It is the purpose
of this report to present a state-of-the-art survey of the design tech-
niques applied to these devices. Consideration will be given to the
nuzzle gas flow phenomena associated with both the bare muzzle and the
muzzle device-equipped flow fields. Both experimental and theoretical
engineering design analyses will be investigated. To define the floN
environment in which the muzzle device is immersed, this report will
commence with a discussion of bare muzzle phenomena.

II. MUZZLE GAS FLOW

The development of the flow field associated with the discharge of

a gun may be broken into two phdses. The first phase involves the
formation of the precursor free jet and blast. This flow is induced b;
projectile in-bore travel forcing the tube air to be ejected. Additional-
ly, imperfect obturation pe,-nits the leakage of high press,,re propellant
gases around the projectile. The second phase consists of -he flow
field created as the high piessure propellant gases leave the y:-. tube
foilowing projectile uncorking. This phasc is of longer duraLion than
the precursor phase. It commences with the formation of a strong blast
and a coupled, highly underexpanded jet, continuing through the emptyinQ
process wherein the pressure in the gun tube is finally brought to equi-
librium with the ambient pressure.
 The formation of the precursor effects is depicted schematically
in Figure 1. Following ignition of the propellant, the projectile is
accelerated down the gun tube, forming compression waves in the quies-
cent tube air. This is the classical accelerating piston problem from
one-dimensional gas dynamics. The first wave propagates at the local
speed of sound. As it moves, the air it passes is accelerated in the
direction of wave motion, compressed and heated. Since the speed of
sound in air is proportional to the square root of the air temperature,
waves will propagate at a higher velocity through the compressed mediun.
This effect, coupled with tihe velocity imparted to the tube gas, causes
the subsequent compression waves to catch the lead wave and eventually
coalesce to form a shock. This process is shown in the x-t diagram.
As the projectile continues to accelerate and as high pressure propel-
lant gases leak around it, more compression waves propagate down the
tube strengthening the shock.
When the shock reaches the muzzle, the lateral constraints are no
longer present and the gases can expand in three dimensions. The
intrusion of this high pressure, moving tube gas stream into the quies-
cent atmosphere surrounding the gun causes the still air to be rapidly
displaced forming the precursor air blast. The tube gas stream expands
forming an axisymmetric free jet. If the flow velocity behind the
normal shock is subsonic at the time it reaches the muzzle, a one-
dimensional expansion fan will propagate back up the gun tube acceler-
ating the gases to a sonic velocity at the muzzle. If the flow behind
the normal shock is supersonic at the time it reaches the muzzle, no
waves can propagate back upstream. A spark shadowgraph of the pre-
cursor blast and free jet formed about the muzzle of a 5.56mm Mann
barrel firing M-16 ammunition is shown in Figure 2. This shadowgraph
was taken at the BRLl*, but the technique is an historically proven ex-
perimental tool utilized by early investigators 2 - 5 of muzzle gas flows.
The precursor blast is nearly spherical with its center displaced
forward of the muzzle. In this case the blast geometric center is
located approximately 3.2 calibers forward of the muzzle. The tube gas
jet is seen to possess a shock structure typical of underexpanded jets.
A point to be emphasized is that the blast and jet formation processes
are mutually dependent and occurring in a highly unsteady manner. The
jet develops in an environment which has been previously compressed
and set in motion by the passage of the precursor blast. In turn, the
growth of the jet effects the blast, as can be seen from the waves
which are propagating from the jet to the blast in the shadowgraph. An
additional complication is introduced if the unsteady nature of the
muzzle conditions is considered. The precursor muzzle flow forms the
ambient into which the high pressure propellant gases expand once the
projectile separates from the muzzle.
The growth and decay of the propellant gas jet and associated blast
have been studied by a variety of researchers. Cranz 3 , 4 utilized spark
Schlieren coupled with selective probing of the flow field. He noted
*Refcrenceo are lioted on page 101.

14
the qualitative agreement between his experimental results and the
linearized jet theory of Prandtl. Quayles used similar techniques to
investigate the muzzle gas flow but did not attempt a discussion of the
gasdynamics involved. World War II stimulated interest in muzzle phenom-
ena. Working on blast suppressor design, Slade 6 investigated the chrono-
logical history of muzzle phenomena and presented a qualitative descrip-
tion of these based on spark shadowgraphs of a caliber .30 rifle. Using
work performed on muzzle brakes during the war7a- 7 c, Oswatitsch6 analyzed
the development of the muzzle gas flow field and attempted to quantify
the phenomena occurring. From these works, the following description of
the muzzle gas flow field about a small caliber, high velocity weapon is
constructed. In particular, the flow from a 5.56mm Mann barrel firing
an M-193 projectile at 3200 feet per second is addressed.
Schematics of the salient flow features are presented in Figures
3a - 3d. The first figure in the series shows the propellant gases ex-
panding around the projectile soon after uncorking. A strong blast is
formed in the area exterior to the boundary of the tube gas free jet.
However, there is no evidence of shock formation at the interface
between the flow field of the tube gas jet and the developing propellant
gas jet. This may be due to the effect of the rapid expansion around
the projectile undergone by the propellant gases combined with the
presence of an established velocity field (the tube gas jet); thus, the
condition of continuity of pressure and velocity across a contact surface
could be satisfied. However, the opaqueness of the propellant gas cloud,
which also contains a large amount of particulate matter, prevents pene-
tration and makes observation of the exact interface location impossible.

Figure 3b shows the flow field at a somewhat later time. The pro-
pellant gases have moved over the projectile and the interface between
the two gas jets has moved forward of the projectile nose. Beginnings
of shocks are forming ahead of the interface as the propellant gases
move into what was the low velocity region of the tube gas jet (i.e.,
behind the tube gas jet Mach disc). The momentum 'of the propellant gases
in the axial direction plus the flow paths orovided by the precursor
flow field have caused the propellant gas jet and blast to become signi-
ficantly elongated in the axial direction. A shock structure internal
to the jet has begun to form. Oswatitsch 3 has shown that the muzzle
velocity of the propellant gases is at least sonic. The flow between-the
muzz;e and the projectile base expands rapidly to supersonic velocities.
That these vehiocities are higher than the projectile velocity is
evidenced by the shock standing at its base. The pressure at the base
of the projectile in this region is considerably higher than that on
the nose, and the projectile continues to accelerate. At the jet
boundary, shocks are beginning to form due to a coalescing of waves
reflected from the contact surface between the propellant gas jet and
the surrounding air. The jet boundary is observed to be hiqhly turbu-
lent indicating that mixing between the propellant gases and air is
occuring there. This mixing is an important factor in the examination
of flash phenomena.
15
 As the propellant gases expand supersonically, their pressure drops
rapidly. It is seen from Oswatitsch that the pressure drops from on the
order of 5000 pounds per square inch at the muzzle to 5 pounds per square
inch at a distance of 10-15 calibers. Since these sub-atmospheric pres-
sures must eventually recover to near-atmospheric pressure (the exact
recovery pressure is determined by compatibility of pressures and veloci-
ties at the interface between the jet and blast fields), shocks occur in
the flow to recompress the gases. In a highly underexpanded jet structure
such as the propellant gas jet, axial recompression occurs through the
formation of normal shocks (Mach discs in axially symmetric flow or
Riemann waves in planar flow). Early in the muzzle jet development, the
projectile provides sufficient obstruction to the flow to bring about
recompression; however, as the flow field grows larger relative to the
projectile, the influence of the projectile presence diminishes and
eventually can no longer recompress the flow to provide recovery. At
this point, the Math disc forms in the jet and becomes the means of
pressure recovery.

Figure 3c shows the flow field at this time. Across the Mach disc,
the flow is decelerated to subsonic velocity. Since the projectile
still moves at roughly the muzzle velocity, it no longer is moving
slower than the fluid, and it begins to experience drag ra~her than
thrust. As it moves through the propellant gases, a viscous wake is
shed. The blast wave is decelerating due to the effects of radial ex-
pansion and the projectile moves through it. As the projectile pene-
trates the blast, a bow shock forms and the drag increases still further.
Simultaneous to and at times preceding projectile penetration of the
blast, solid particles are seen to be penetrat.ing the blast and moving
at supersonic velocities into the undisturbed air. These are most likely
powder particles which may be still burning.

The jet structure at this time is nearly that of the typical under-
expanded jet. However, the shock structure possesses an irregular,
pentagonal configuration. This may be an effect of the precursor flow
influencing the subsequent development of the propellant gas flow field.
The precursor tube gas jet has been inundated by the rapid growth of the
propellant gas jet. One of the last vestiges of the precursor flow is
the precursor blast which will soon be outstripped by the stronger main
blast.

Subsequent to the projectile penetration of the main blast, the jet

attains its maximum growth, Figure 3d. The jet is now of a structure
identical to the highly underexpanded steady jet. The flow within the
bounding shocks is supersonic increasing in Mach number along the axis
and reaching a maximum. just at the normal shock. The thermodynamic
properties of the gas, namely, density, pressure, and temperature, drop
off rapidly from the muzzle values as the expansion region is traversed.
Across the shocks, these properties increase again, reaching a second
maximum behind the Mach disc. The flow traverses the peripheral oblique
shocks maintaining supersonic velocity; however, maintenance of super-

16
sonic velocities across a normal shock is not possible. Thus, a contact
discontinuity is formed between the flow which traverses the oblique
shocks and that which traverses the Mach disc. Across the discontinuity,
pressure and flow inclination are maintained while all other properties,
such as velocity and density, may be discontinuous. The boundary of the
jet is the only area where viscous effects are significant. The turbulent
mixing layer and recirculating flow region (smoke ring) bring about rapid
mixing of the propellant gas and air. The importance of the nature of
the jet boundary on the properties of flow interior to the bounding
shocks has been shown in steady jet studies 9 - 1 3 to be negligible. The
development of the prcpellant gas free jet and blast is vividly illustra-
ted in the series of spark shadowgraphs shown in Figures 4a -4d.
The state-of-the-art of blast and jet theory will be discussed in
the second volume of this report; however, sinc.e the flow in which a
muzzle device is immersed is jet-like over almost all of its duration,
it will be informative to consider the results of recent work relating
tc the parameters influencing jet structure. In his text devoted to
jets, Pail" resents the status of jet studies through the mid-fifties.
Love, et al, present an updated summary of work in the field in ad-
dition to presenting an in-depth study of axisymmetric free jet. The
upsurge of the15space program in the sixties brought about increased
investigation s'5 of rocket plumes, which are effectively underexpanded,
axisymmetric free jets. The work of most interest in the investigation
of the free jet established at the muzzle of a gun is that of Love .
9
I
In this paper, supersonic jet flows are examined both theoretically,
using the method of characteristics, and experimentally, using Schlieren
photographs. Parameters of i,,cerest are varied and the effect on jet
structure is observed. These parameters are: jet exit Mach number,
nozzle contour, ratio of specific heats and pressure ratio. For the time
being, it is convenient to asswue 8 that the muzzle velocity becomes sonic
upon projectile exit and remains so over the major portion of the jet
lifetime. Further, a bare muzzle with zero inclination to the bore axis
will be taken as the most straight-forward example of this type flow.
Figure 5 schematically depicts a wave pattern for such an expansion.
At each corner, the flow is expanded through a wave fan which may be
locally assumed to behave like a two-dimensional, Prandtl-Meyer expansion.
Since in inviscid flow theory the condition for a contact surface to be
maintained is that pressure and flow inclination be continuous across it,
the flow expands through an angle, o, sufficient to reduce the static
pressure from the exit pressure, Pe, to the surroundings pressure p..
However, when the waves from the opposite corner reach the jet boundary,
they would overexpand the flow to a pressure below that of the surround-
ings. In order to mai, tain pressure equivalence at the boundaries these
incoming expansion waves must reflect as compression waves. This d-s
in fact occur and the resulting reflected compression waves eventual y
coalesce to form the shock structure of the jet.

17
 Treating the corner expansion as two-dimensional, Love computes the
effert of variation in the exit-to-surroundings pressure ratio, pe/p_,
on the initial jet deflection angle, -,,Figure 6. The effect of in-
creased pressure ratio is quite obvious. For values less than 200, the
deflection angle grows rapidly with increasing pressure ratio. However,
further increase causes less rapid changes in deflection angle which
approaches the limiting expansion angle (for y = 1.4, (Inax - 1300) asymp-
totically. The effect of pressure ratio on the muzzle jet can be noted
if the initial inclination of the precursor tube gas jet, Figure 2, and
that of the propellant gas jet, Figure 4d, are compared. Love investi-
gated another parameter affecting the initial deflection angle, namely,
the ratio of specific heats,Ye, Figure 7. The strong dependence of the
initial deflection angle and therefore the overall jet structure upon
the ratio of specific heats, Ye, indicates the importance propellant
cheiistry can have upon the jet structure. Love continues his calcu-
lation past the initial expansion, calculating the jet boundary shape
and constructing the characteristic net. However, the effect of com-
pression wave coalescence into shocks is not ,reated. This coalescence
is treated by other jet analyses 1 5 , 2 6 and resultant property profiles
are presented or discussed.

A shock formation which is difficult to predict is the Mach disc.

Pail' discusses the evolution of the Mach disc in an underexpanded jet
as the pressure ratio is increased, Figure 8. These schematics, taken
from the original Schlieren photographs of Hartmann and Lazarus3o, show
that as the pressure ratio, pe/Pý, increases the flow undergoes a
stronger expansion initially and must, therefore, undergo stronger recolli-
pressions to satisfy the boundary pressure compatibility condition. In
the first instance, the exit and surrounding pressures are identical;
thus no pressure waves are generated in the inviscid flow field. As the
exit pressure is increased, these pressure waves do evolve. The second
schematic shows that for a relatively low overpressure the weakness of
the waves and confinement of the flow geometry inhibit coalescence until
the compression waves reach the outer boundary. As the pressure in-
creases further, the strength of the waves and geometric scale of the
flow increase and the shocks move in toward the axis. Further increase
causes a shock reflection at the axis. Eventually a pressure ratio is
reached where the oblique shocks become so strong that the flow can not
traverse both and remain supersonic. At this point, a Mach reflection
occurs and the normal shock or Mach disc is formed. Further increases
of pressure cause the compression waves to coalesce sooner, forminy the
shock bottle,and move the location of the Mach disc further downstream.
The position of the Mach disc as a function of exit Mach number, pressure
ratio, and ratio of specific heats has been investigated by a variety of
researchers9e1Ob13' 2 6 -3C For a sonic nozzle, the location of the Mach
disc in the axial direction is g1,,enZ 7 92- as

( "e 1 /2
x/D -- 0 .69
e p
where x = axial distance to shock
D = orifice diameter
The applicability of this type of relation to the muzzle jet can
be seen In comparing the photographs of the precursor and propellant gas
jets. Additionally, in another facet of the jet life cycle, its decay,
the analysis of free jet structure indicates what to expect. As the
pressure ratio decreases,the shock bottle should shrink in size with the
Mach disc moving in toward the muzzle. Thus, a reverse sequence to that
shown in Figure 8 would be anticipated. SladeG in his discussion of the
muzzle phenomena about a caliber .30 rifle indicates that such a process
does occur. In light of the direct relationship between the muzzle
phenomena after the bullet leaves the muzzle flow field and free jet
flows, it is not surprising that muzzle device research has made exten-
sive use of jet theory.

III. MUZZLE DEVICES: THEORY

A simplistic approach to the categorization of muzzle devices would
be to establish two groups; the first would contain devices which seek to
control or utilize the momentum of the propellant gas flow, while the
second would consist of those which control the rate or distribution of
propellant energy release, Figure 9.
Muzzle brakes and compensators redirect the propellant gas from
purely axial efflux utilizing the resultant forces generated to modify
the recoil characteristics of the weapon. Blast deflectors are designed
to ameliorate the effects of the propellant gas ejection upon the sur-
roundings. They may be directed to reduce the dust raised by the muzzle
gas jet or to eliminate blast wave impingement on the gunner or nearby
structures. Deflectors, like brakes and compensators, are essentially
turning vanes and/or channels which redirect the flow momentum to a
sector where ejection will give least harmful effects. The second
category of devices is concerned with the manner in which the residual
propellant energy is released into the atmosphere. Blast suppressors
attempt to minimize both the amount of energy released from the weapon
and the rate at which it is released. A wide variety of blast suppres-
sors has been designed, including multi-baffle, expansion chamber,
energy absorption and energy dissipation devices. Flash suppressors
function on the principle of controlling the distribution of propellant
gases and their energy such that the elevated energy states required to
initiate combustion do not occur simultaneously to or in the vicinity
of ignitable propellant gas-air mixtures,

The utilization of any particular muzzle device requires a consider-

able effort to interface the device to the weapon and to the tactical
role of the weapon. A muzzle device is useless if its design is im-
practical either in terms of cost, weight, bulk or interference with

19
proper weapon functioning and use. The maintenance associated with
certain muzzle device designs makes them impractical for general usage.
Consideration must be given in designing a device that can utilize cur-
rently available materials and manufacturing techniques to provide
sufficient structural strength for a long, failure-free life. With such
obvious practical considerations recognized, this report will be directed
to the consideration of the gas dynamic and chemical kinetic phenomena
associated with projectile ejection from a weapon equipped with various
forms of muzzle devices. It must be noted that while practical consider-
ations are important, they are at times a hindrance to theoretical and
experimental investigations of the detailed flow phenomena. Thus,
investigators are forced to make simplifying assumptions in their analyses
Which are at odds with practical realities. These assumptions and the
necessity of making them will, hopefully, be made clear in the dis-
cussions to follow.

The devices will be considered by functional groups. Brakes,

compensators and blast deflectors operate on identical principles and
will be discussed together. Blast and flash suppressors will be con-
sidered separately. Important contributions to the analysis of flow
through these devices will be discussed. The fourth section will present
related experimental techniques.

A. Muzzle Brakes, Compensators and Blast Deflectors

The most definitive source of information concerning the theory and

design of muzzle brakes is the survey performed by Hammer 31 just after
the close of World War 1I. If consideration is given to the fact that
the status of muzzle brake design has not advanced significantly from
the time of this report, a phrase from page one of Hammer's work is
quite telling: "One of the reosons for initiating this project was to
discover design formulas for muzzle brakes. Unfortunately, such formu-
las were not discovered." While this statement is true, the report
does present much valuable information, starting with a state-of-the-art
summary and concluding with a collection of translations of significant
7
foreign papers , 32- 38 . To provide an understanding of the development
of muzzle brake technology, select papers from this group will be
considered.

A basic problem in the computation of muzzle gas flow fields is the

estimation of muzzle flow properties as the gun tube empties. Early
researchers utilized the work of Hugoniot 3 2 to obtain this estimate.
Hugoniot considered the process of emptying a high pressure reservoir.
His analysis assumes quasi-steady flow in that no wave motion is allowed
in the reservoir. This implies instantaneous adjustment of the thermo-
dynamic state of the reservoir gases to the effects of outflow. The
calculations are carried out for either an isothermal or adiabatic
process. The rate of outflow may be computed by assuming a sonic orifice

20
or, if the pressure ratio across the orifice Is too low, by using the
steady, quasi-one-dimensional flow equations. Knowing the volume and
the initial state of the reservoir gases, it is possible to use the
computed rate of outflow to compute the subsequent state of the gases
for each type process. Rateau 3 3 modified the analysis of Hugoniot by
considering the effect of gas co-volume and approximating the initial
propeity distribution in the gun tube at the time of shot ejection by
one with zero velocity but an elevated pressure and temperature which
upon adiabatic expansion to the muzzle pressure produces the muzzle
velocity. CornerN compares the analyses of Hugonlot and Rateau, and
presents his own calculations for the emptying problem. His work is
based on one-dimensional, unsteady gasdynamics and provides a more
satisfactory comparison with theory than either Hugoniot or Rateau.
Hcwever, he indicates that the analysis of Rateau gives reasonably good
results if the assumption of an initial property distribution is dropped.
Oswatitsch8 utilizes the method of characteristics to calculate the
variation of muzzle properties. His main aim in this analysis is to
establish the validity of assuming a sonic muzzle and quasi-steady flow
conditions rather than providing detailed initial conditions to his
muzzle gas flow field calculations. The theory of interior ballistics
has developed considerably (e.g., Ref. 39-45) since these works; however,
with few exceptions, these advances have not been incorporated into
analysis of muzzle gas flow fields. For this reason, they will not be
discussed, but it is realized that future efforts which attempt to in-
terface the interior and transitional flow fields must consider these
improved techniques.

In addition to modifying Hugoniot's theory, Rateau 3 3 performed an

analysis of the flow through a muzzle brake and provided a technique
to calculate the force on the brake. 'he methods of Rateau are of
particular interest since they form the basis for the section on muzzle
brakes in the current Army Design Handbook1 6 . Rateau is realistic ir.
assessing some of the uncertainties facing analysis of muzzle gas flows;
among the difficulties he notes are:

1. Physical properties of the high temperature propellant gases

are inadequately known.

2. Chemical state is unknown; relative proportion of hydrogen and

water vapor greatly modifies coefficients.
3. The flow is highly unsteady while analyses utilize quasi-steady

assunptions.

4. Pressure and temperature of gases at muzzle are not well known.

Rateau defines the gun emptying problem as being analogous to the

emptying of a reservoir of volume equal to the volrixe of the gun con-
taining gas at a pressure and temperdture both higher than the gun
gas values. The gases then expand adiabatically through a converging

21
nozzle such that at the throat of the nozzle the flow velocity is equal
to the muzzle velocity of the gun and the pressure is equal to the
muzzle pressure. Using an equation of state of the form:
p(v - c) RT
where
p = pressure
v = specific volrume
a = co-volume of gas
R = gas constant
T = temperature
he calculates the property variation at the throat through the emptying
process. After the gases pass the throat, they are expanded through a
brake of the geometry shown in Figure 10. The conical expansion nozzle
has a Jeflection angle s which is set4 6 at some value less than 300.
This requirement is made to prevent flow separation at the expansion
corner. However, in light of recent work 9 - 3 0 on underexpanded rree jets,
this required angle is much too confining. This can be seen by consider-
ing that underexpanded jets have initial deflection angles of 900 or
more at high pressure ratios, Figures 6 and 7. Since propellant gas
jets have a low ratio of specific heats (y 1.25) and high pressure
'ý.

ratios (Pe/P,ý -v0(500)), large initial deflection angles would be antici-

pated. Figures 4a - 4d indicate this angle to be roughly 900. Therefore,
assuming a smooth transition section from the muzzle to the conical
section, turning angles greater than 300 could easily be obtained. Since
the important braking forces occur early in the emptying process while
the pressure ratio ,s on the order of 100, there should be no difficulty
in maintaining a 60 - 700 turn. There appears to be some confusion as
to the role of the conical expansion nozzle; rather than inducing ex-
pansion of the flow, the nozzle channels the expansion to insure maximum
impingement upon brake surfaces.
Rateau assumes the nozzle to be designed such that adiabatic ex-
pansion occurs through it, reducing the reservoir pressure to atnospheric
pressure at the entrance to the brake. Using the isentropic flow
relations, it is possible to calculate the resulting velocity at the
brake. From his reservoir emptying calculations and under the assanptlon
of quasi-steady flow, this velocity may be computed as a function of time.
Knowing the mass flow through the muzzle and constraining the channel
geometry to be such that a quasi-one-dimensional flow may be assumed
the mass flow into the brake and projectile hole may be divided in direct
proportion to their relative areas, Figure 10:

ml SI + S2

ý2SlS ii. s-
-

22
where hi* = mass flux through muzzle
m1 2 = mass flux through brake, projectile hole
sl,2 = area of brake, projectile hole entrances.
Rateau equates the force on the muzzle brake to the change in mo-
mentum through it, assuming flow properties constant through the turn:
F1 = m1 u + 6 1 u sin a, = m* S'+S
S u (1 + sin a,)
2

where: u = velocity of flow expanded to p.

For subsequent baffles, the mass flux is again taken as area de-
pendent and the flow properties remain constant. Thus:

m: m2 s (s1•
n•*s3s S2_
) S3
=2S
53+54 S2
-'4

F3 sS+2 S3(5)) u (1 + sin a2).

Using-this analysis, Rateau calculates the force on a muzzle brake from

a 75mm gun, Figure 11. The shaded portion of the curve represents the
effect of projectile residence in the brake. While there is not a
significant increase in impulse (area under the curve) due to projectile
residernce, there is an increase in peak force which must be accounted
for in the structural strength of the device. Rateau claims, but does
not show, good success in comparison between tests and analysis.

Rateau's analysis has several shortcomings. He does not account

for non-isentropic effects such as friction and, more importantly, shock
formations. His assumption of expansion to atmospheric pressure within
the conical nozzle is not substantiated. The assumption of turning
through the brake without property change is obviously incorrect. These
simplifying assumptions make it difficult to use Rateau's analysis to
perform optimization studies. The only parameters to be varied are the
relative areas of the brake channels, the number of brakes and the
turning angle. This leads to the obvious optimization of turning the
maximum amount of flow through the greatest angle. The addition of more
brakes can be shown to follow the law of diminishing returns as succes-
sive brakes work with a mass flux diminished by the flow-through through
its predecessors.
Even with its shortcomings, Rateau's analysis formed a basis for a
number of subsequent reportsN-37. The British improved upon the anal-
ysis by including the effect of area variation and radial expansion 4 7 .
This form of Rateau's analysis is essentially that presented in the
current Army Design Handbook 46 .

23
 With the approach of World War II, the Germans began research into
the design of muzzle brakes. This program was largely experimental and
produced an unbelievable number of different brake designs 31 . The design
which came into most use both by the Germans and the Allies was the
double baffle brake, Figure 12. According to the classification of
brakes set forth in Reference 46, this is an open brake since the flow
expands freely to the brake; whereas the device of Rateau, Figure 10, is
a closed brake since the flow is channeled to the brake surfaces.
During the war, the British tested a captured German double baffle brake
against a variety of high performance designs4 8 . The conclusion was
that although better braking action could be obtained through the use of
closed type brakes with greater rearward flow deflection, the penalty of
increased blast overpressures in the crew area could not be tolerated.

The German research effort also included basic investigations of the

gas dynamics of muzzle brakes°/a-?c. Oswatitsch assumed that the greatest
braking forces were imparted during the high pressure, supersonic portion
of the tube emptying cycle and, further, that this flow could be assumed
Lo be quasi-steady. Examining a typical brake, Figure 13, Oswatitsch
sought to optimize the design details of the device such as spacing,
turning angle, brake contuur, etc. To estimate the forces transmitted
to the brake surface, Oswatitsch performed a momentum flux calculation:

Fx = Ap-p+cu2) dAj - ;. (p-p.+pu 2 )dA 3 + cos ae I (P-P+C+u2 ) dA 2

At the muzzle and brake projectile opening,p ý,p,; therefore neglect p.

with respect to p:

FX= 2 ) dA, - fiA3 (p+pu 2

'Alp+ou ) dA 3 _ cos c1e fA2 (p-pi+puf) dA 2

Examine the last integral and introduce mean property values over the
discharge cross section:

fA2 (p-p. + pu2 ) dA2 = 'A 2 Pu

2
(1 + r)dA2

-(1 + Pe-P )fA d-

=W(
-2•e )e A
2 udA2

Define: Ai = f (p + pu) dAi

GA = fA pu dA

The force equation becomes:

24
 FX = JAI " JA 3 + cos aee (1 + eeP) GA

From conservation of mass:

GA2 = 'A1 - 'A
3

and defining an efficiency factor, a, to be:

ay FX/JA,
or:
JAI +e•e
co 4P 'A1 '
qpl+cos " JwAJ

Oswatitsch now assumes that the conditions at the muzzle are sonic and
uniform across the opening. Under this assumption (asterisks indicate
sonic throat properties):

GA l , pudAI p* u* A,
JAl fAt0 +puzdA (P* + p* u* 2 )A1

= p* u*/p* =M*- I .*
+p* u*Z 1 + yM*°Z
P*

Since M* = 1:

GAI ='

For a quasi-one-dimensional flow the maximum expansion that can occur

is to a condition of zero static temperature implying a complete con-
version of internal gas energy into kinetic flow energy. The energy
equation can be used to calculate the maximum velocity that occurs in
such an expansion:

Wmax y-1

Thus:

25
 A Wm ax
and: JA, We f~ e~ / 1 'A
0 -- ~ +-- +. .1 ~ 6
G COSC~
A1 W1 ax -Al

To evaluate the efficiency of a particular brake design, it is neces-

sary to obtain estimates of the flow behavior within the brake. The
propellant gases leave the muzzle at the velocity of sound and are
expanded to sL'nersonic velocities. The deflection of thp supersonic
stream, at the brdke surfaces causes shocks to form. Thus,if the deflection
is great enough, the flow undergoes a non-isentropic process resulting
in a decreased stagnation pressure. This represents a decrease in the
effective muzzle pressure. Since the deflection of the flow is not
constant at all points in the brake, the effective muzzle pressure varies
fromn point to point. However, as a first approximation, Oswatitsch
assumes the flow to compress uniformly through a normca shock re-
sulting in a decrease in effective muzzle pressure from ps to P§M
From this condition, it accelerates along the brake surfac'T and is
ejected at a supersonic velocity. Examining the efficiency equation, it
is apparent that maximization of the third term is essential. For the
time being neglect projectile hole losses, i-e., assume JA3 E A3 E0.
Then:
+Pewe
+ 4 - P. /77 COS c•
Wmax t PeWe j 2-1Y
Define:

Wmax Pe
W We 1+ e- P J

w~ ~ e-

Wmax I e

WeeI

Oswatitsch examines the maximization of this term. Since shocks occur

in the brake, let:
26
 P. P. PSM,

Pe Psm Pe

Assuming an isentropic expansion from pm to •., then R can be calcu-

lated from the isentropic relation:

PS~m ~
and:

e Tee
Wmax y RT*

Me yR Tse
e

-I Ts--*

across a shock, T T
Since stagnation temperature is constant
thus:

e w Me vy R/(I+L AM
"mWrax vYR7--

Using these relations, Oswatitsch calculates W as a fut'ction of Psm/Pe

for various psm/p., Figure 14. From the limit of sonic exit flow on
the left, the W factor increases to a maximum which occurs when the exit
pressure is equal to the ambient pressure, Pe ' P-' and then decreases
as the flow is overexpanded, Pe - p,. The brake efficiency is directly
related to W:

0=1 + W C cLa

curves of Figure
To use the m/5,"and c/. 14, it is necessary to be capable of cal-
the detaileA
cula~ing P/ an psIp•. This requires knowledge o"
of characteri-;tics calculao-•,n
brake flow.' Oswatitsci utilizes a method of the
number along the centerline
to obtain the variation of Mach at the'
flow passes through a normal shock
muzzle jet. Assuming the
27
location of the brake, Psm/P may be calculated from the Rankine-Hugoniot
relations and the known muzzle and ambient pressures. A computation of
P' /Pe is more difficult as it requires a technique to determine how the
flow expands along the brake surface. Oswatitsch does not present such
a technique and instead assumes Pe = p.. This effectively reduces the
analysis to a method of accounting for shock losses in a thruster nozzle.
Since if the flow passes a strong shock, P~m/Po is decreased thereby
decreasing a. The analysis is valuable in that it represents the first
attempt to account for internal shocks within a framework of multi-di-
mensional flow.

Assuming a typical value of psm/p= = 1000, Oswatitsch picks

W = 0.87 and calculates the resultant brake efficiency as a function of
the deflection angle, Figure 15. The effect of increased deflection is
to increase brake efficiency. It should be noted that the calculations
presented by Oswatitsch do not allow a detailed investigation of the
brake surface design since they in no way account for the method of
turning the flow through the angles depicted.

Oswatitsch recognized that this analysis did not account for pro-
Jectile hole losses. Thus, examining the momentum flux term under the
assumption of uniform properties over the cross section of the muzzle
and projectilc hole:

S= +23.
JA; P1 + Plu 2
A,

P.•-. +1
= P3u _ pUuR 2 A-
p + 1 PlU1 2 A1

1 + G
L+M2" GA3 U3

yM 1 2 A,

Now if the flow is assumed to pass through a normal shock at the brake,
it is decelerated to subsonic velocities. In passing through the pro-
jectile hole,it expands to sonic velocity at the throat. Since the flow
is steady and since the stagnation temperature is constant across the
shock:
T____ - 1
1*-
TS3 (1+ 2 M3 *2)

28
and 3 2
2 Ts*

but
TS 3 Ts* Ts
1 *

TT* 2 T *_ 2 T = T3*

and
u3* = a3* = rRýT*= U*

or, at each subsequent throat in the brake, the flow is sonic with a
velocity equal to the muzzle exit velocity. Thus:
A GA

but: _I

GA., Pq*ul*A3 Pq*A14

A3

Pt 3*
""•Ptl* A,
Pt1

Since M3* = MI* 1 p, and:

Pt3* ptl*

GAp' ptl, AI3

LAI s A-
GAi Pt * A, PS Al
29

S. ... . . ..................... ~~~~~~~~

| .... .ii
l| || || i |. _. .i•. i.F
Substituting into the expression for efficiency:
o= --I P +wIy•ll -• cs•
%s +mW 2. Pism A cos

PI-- _ ~ o
m
PS Al W 2-PI1

This exordssion indicates that shock losses to the flow passing through
the projectile hole have a favorable effect on efficiency. Since Psm/Psm
decreases with increasing Mach number and since the Mach number increases
along the muzzle jet centerline, this indicates that projectile hole
losses can be minimized if the distance from the muzzle to the brake is
increased(i.e.,with increased brake standoff, the mass flux through the
projectile hole decreases). However, this must be balanced by the effect
of decreasing W with Psm/Pp Ideally, a brake should be designed to in-
duce a normal shock to stand at the projectile hole while inducing mini-
mum shock development on the turning vanes. Obviously, such a device
would be impossible to construct. However, utilizing this concept,
Oswatitsch designed a "back-effect-free" baffle which he tested and com-
pared with his theory, Figure 16. The agreement is quite good, especial-
ly in light of the development of oblique rather than normal shocks in
the projectile hole.
7C
Additionally, Oswatitsch develops a method of characteristics
construction of the flow through a muzzle brake. He develops the shock
structure and investigates the effect the first baffle has upon the
efficiency of the second. His conclusion is that the efficiency of a
second brake suffers from two main causes:
1. Reduced mass flow due to presence of the first brake.
2. Reduced jet pressure ratio due to shock structure.
The reduced pressure ratio results in less expansion in the second jet.
This tends to develop a higher concentration of mass flux near the axis
thereby allowing for increased projectile hole losses. Thus, to obtain
more efficiency from a given diameter second brake, it must be placed
further from the projectile hole of the preceding brake than the first
brake is from the gun muzzle, note Figure 12. Oswatitsch develops an
experimental test facility which he used to corroborate his theory and
study the internal brake shock structure. However, a discussion of this
facility will be deferred until Section IV.
During the war, research in the United States concentrated on the
design of blast deflectors. The motivation of these studies was to

30
decrease the obscuration generated in firing high velocity guns. The
work6 , 4 9 , 5 0 concentrated on experimental investigations which will be
discussed in Section IV. Millikan 4 9utilized the work of Corner 3 8 to
obtain a relation for the momentum flux from a gun. He then used this
to calculate the downthrust on a gun under the assumption of complete
gas efflux at sonic velocity through holes of a specified geometry. The
analysis was overly simplistic failing to allow identification of signi-
ficant geometric and gas dynamic influences.
Following World War I1, interest in muzzle device research waned.
Some reportss- 5 s 6 were produced which considered flow through muzzle
brakes. Of these, only the work of Smith 5 3 "56 contains an effort to
integrate advances in gas dynamic theory with muzzle gas flows. Smith's
approach is similar to that of Oswatitsch. He assumes the flow to be
steady and utilizes a method of characteristics construction 5 7 of a
free jet to obtain the flow properties.

Smith notes that Owen and Thornhill postulate the flow within the
first shock bottle to be universal, i.e., independent of the pressure
ratio. This statement is substantiated by Ladenberg'sS 8 experiments.
Smith then assumes the force on a baffle placed within the shock bottle
may be estimated by assuming "the flow which would normally pass through
the disc is simply removed from the flow picture with a corresponding
removal of its thrust component. The same result would be obtained if
we considered that this flow were deflected normal to the jet axis,
without disturbing the remainder of the flow," Figure 17. Smith notes
that an optimal value of the axial brake location exists. For small
axial separation from the muzzle, there are large losses through the
central core flow. As the brake is moved further from the muzzle,
these losses dimninish due to greater flow impingement upon the brake
surface. However, continued displacement tends to increase the flow
outside or around the brake. These two effects must be balanced.

To ex-iiine the optimization of brake, Smith calculates the aero-

dynamic index, n, of a selected brake versus the axial location, x,
within a jet having Pe/P' > 1.0. The aerodynamic index, which is
identical to the brake efficiency, u, of Oswatitsch, is dcfined to be:

n1 T

where: T - thrust on barrel with brake

To =thrust on barrel without brake.

The calculated results are shown in Figure l8. Smith compared his
theory which was calculated for pe/p_, with results obtained from a
steady jet 5 3 which had a pressure ratio p /p, = 53. Theory and ex-
periment agree well. To assess the effeci of variable Y, a computation
for -Y 1.67 was made and showr in Figure 18. It is interesting to note
31,
 that the experimental data agree better with this computation. However,
a consideration of the earlier discussion on jet flows shows this is not
surprisinq. Since increasing I or decreasing Pe/Pý produce similar
changes in the initial jet deflection angle, the jet calculation for
Pe/P. - -, Y = 1.67 should result in a flow more geometrically similar
to the finite pressure ratio jet (Pe/P- = 53, y = 1.4) of the experiTnents.

In later work 5s, Smith uses the analysis to predict the pressure
distribution on the brake and compares this with experiment, Figure 19.
The ratio of the pressure difference between the front and back surfaces,
to the reservoir pressure is plotted as a function of radial location.
Apparently Smith calculates pressure on the front surface by assuming it
equal to the stagnation pressure behind a normal shock standing at the
location of the surface in a flow field computed from the method of
characteristics. The technique to calculate thie pressure on the rear
surface is not indicated. Possibly, the rear surface pressure may be
assumed to be equal to the ambient pressure.
To validate the usc of a steady jet analysis for the coo,putation of
brake effectiveness of an actual weapon, Smith compares the value of the
aerodynamic -ndex obtained in h'.s steady flow experiments with that ob-
tained in firing a 7.6am rifle. The comparison is good. Since the un-
steady flow experiments cor,.pare well with the steady jet theory, Smith
concludes that the analysis is valid.

The similarities between the work of Smith and Oswatitsch are quite
apparent. Both assme steady flow models, ut;lize the method of char-
acteristics to construct a flow field, and then evaluate brake effective-
ness with identical indices. !n Section IV, the similarities of their
experimental approaches wfil be exai};ed. The work of these two men
essentially represent the current state of muzzle brake theory.

B. Flash Suppressors

Of all the phenomena occuring at the muzzle of a gun, flash is the

most difficult to analyze. Not only must the gas dynamics be investi- -_

gated, but the chemical kinetics associated with propellant gas/air

mixtures must be coupled with the flow field. The Franklin Institute
has performed a great deal of flash related research 59-6 5 and, in fact,
produced the current Amy Handbooks on the subject• 6 ,d6 Reviews of the
research performed both at Franklin Institute and elsewhere are avail-
able 6 7 , 63 . These show that flash research is concentrated largely on
experimental investigations while theory is quite basic and largely
qualitative in nature.

A schematic of the flash phenomena showing its salient features"l

is shown in Figure 20. As the projectile is forced down the tube, a
certain amount of high pressure propellant gases leak around it. The
leakage mixes with the tube gas and is ejected from the gun tube foniiing
the precursor flow field, Figure 1. If the tube gas was largely air,

t3
the ejected propellant gas/air mixture could be burning upon exit, or, _1
if the tube gas was mainly the propellant gases remaining from a pre-
viously fired round, then mixture with the atmosphere could cause
ignition. In any caselthe flash which occurs prior to the round break-
ing the muzzle is known as preftaah.

With the uncorking of the projectile, the high pressure propellant

gases are released. At the muzzle, the propellant gases are at high
temperature and are likely incandescent. This bright orange incan-
descence forms what is known as pr.nmary flash. As the gases move away
from the muzzle, they expand very rapidly experiencing a large drop in
temperature. This temperature drop quenches the incandescent radiation.
However, a dull reddish glow has been observed within the shock bottle
and may be due to energy release from internal energy levels which can
not adjUst to the rapid changes in flow properties. This continued
radiation in the expansion region is known as muzzLe glow. When the
gases pass the normal shock, they are slowed to subsonic velocities,
compressed and brought to a high static temperature. This conversion of
the kinetic flow energy to internal energy of the gas causes the propel-
lant gas to become incandescent once more. This region is called
intermediate flaah.

By far the most severe flash phenomena is 8econa•zry fZaah. This

flash results from the combustion of the propellant gas/air mixture at
the boundary of the jet. The propellant gases mix with the air in the
turbulent shear layer and in the vortex ring. The propellant gases
consist largely of carbon monoxide and hydrogen with lesser amounts of
carbon dioxide, nitrogen, water vapor, traces of hydrocarbons and nitro-
gen oxides. Since the propellant gases are oxygen poor, mixing with the
oxygen rich air forms a combustible mixture requiring only a suitable
source of ignition. Among the possible sources of ignition are:

1. ignition by preflash,
2. spontaneous ignition,
3. ignition by intermediate flash,
4. burning powder particles,
5. hot muzzle,
6. tracer rounds.

An experimental investigation conducted by Stephens 6 0 indicated the

possibility of ignition by the preflash. In a set of firings of a cali-
ber .50 gun, axial holes of various diameters were drilled in the pro-
jectile. As the size of the hole increases, the amount of propellant
gas leakage and, thereby, the preflash effects grow stronger. Stephens
also observed that the secondary flash increased with increasing hole
size indicating a connection between the two phenomena.

33
 Spontaneous ignition was examined by StephensýC who developed a
technique to calculate the temperature of the air gas mixture upon
ejection from an ideal flash suppressor, i.e., one that ejects the gas
at atmospheric pressure. He points out that even under optimal con-
ditions it may not be possible to prevent spontaneous ignition once the
propellant gas and air reach a certain mixture ratio. Fay 6" indicates
that spontaneous ignition is the most likely source of ignition.

1Initicn by the intermediate flash is also noted as possible by

Fay". The effectiveness of bar type suppressors in eliminating inter-
mediate flashS",CEG 8 , and,through itsecondary flash demonstrates the
ignition capability of these phenomena. A set of photographs by Hodil 70
seem to indicate the growth of the secondary flash from the intermediate
flash. The remaininý sources of ignition have been noted by various
researchers 6 8,Ge717...

The literature on muzzle flash contains no analyses which attempt

to model the chemical kinetics of the propellant gas at various points
in the flow field. The reason for this is quite readily ascertained.
There is no available model which accurately describes the non-reacting
flow. Although the method of characteristics has been widely used in
muzzle brake analysis, the assumption of an inviscid flow is not applic-
able if flash is to be modeled. Additionally, the chemistry is dependent
upon an accurate appraisal of pressure, temperature and residence times
the mixture undergoes. The current state-of-the-art does not appear
capable of providing an analysis which could handle all of the phenomena
associated with a propellant gas jet.

The consideration of a technique to suppress flash is thus ldrgely

a matter requiring empirical'y based principles. The current suppres-
sion techniques may be categorized as:
1. Addition of flash inhibitors to the propellant.

2. Utilization of propellant energy through interior ballistic

design to reduce residual, post-ejection propellant gas energy.

3. Utilization of muzzle devices.

The first two techniques will not be considered here. There are two
types of muzzle devices used on weapons"L,, the flash hider and the flash
suppressor, Figure 21. The flash hider merely covers the light-emitting
portion of the mnzzle gas flow. The conical flash hider is shown cover-
ing the primary flash and the muzzle glow. If made significantly larger,
this type of device could also shield against intermediate flash. How-
ever, it does not eliriinate secondary flash, nor does it prevent obser-
vation from the front.

Flash suppressors are active de.ices. They are utilized to control

the efflux of propellant gases. 'ht- onical flash suopp-ssor was the

3 41

Iii
device most surveyed during World War 1168,73-75. Supposedly, the
device expands the propellant gases until they reach atmospheric pres-
sure at the exit. This prevents the formation of an underexpanded jet
and its associated shock structure, eliminating intermediate flash and,
hopefully, secondary flash by removing its ignition source. Since the
jet is being exhausted at supersonic velocities, and at an initial non-
zero deflection angle, shocks will form within it emanating from the
viscoug mixing layer. However, the shocks will not be as strong as the
boundary shocks and Mach disc formed within the bare muzzle jet. A
point worthy of note is that a device of this design is essentially a
supersonic thruster nozzle, and as such, will increase the recoil of
the weapon.

The bar type flash suppressor is similar to the conical suppressor

in that it is designed to alter the flow shock structure. The device
consists of an odd number of bars placed around the bore-line of the
weapon. The slots formed are essentially slit nozzles which allow the
flow to expand radially in a quasi-planar manner. The utilization of
an odd number of bars prevents symmetric wave reflection, Figure 5,
from the centerline which eventually builds up into the boundary and
normal shock structure. Both this and the conical flash suppressor
control the tendency of the flow to overexpand and then recompress
through a normal shock. The elimination of the normal shock prevents
intermediate flash and retards secondary flash.

The capabilities of a variety of muzzle devices to suppress the

flash from a multiple shot weapon were experimentally examined by
Watling 7 6 . Mounting the devices on a caliber .30 machine gun, he fired
a multiple bur-st sequence which consisted of firing for four seconds,
resting for eleven seconds, and recycling. The resultant rate of fire
was 150 rounds per minute. He defines an efficiency index:
number of rounds fired
number of rounds that flash
The optimal value of n is infinity which implies zero flashing, and
the worst value is one,implying all rounds flash. A sample of the
types of devices he tested and the resulting indices are shown in
Figure 22.

The results are somewhat surprising. The standard bar type sup-
pressor does not perform as well as some of the more exotic designs.
However, the method of firing must be considered since multiple bursts
will heat up the barrel and muzzle device,thereby decreasing the flash
suppression capability of the system. Thus a device which is extremely
effective for single shot or short burst firing may not work well in a
prolonged firing role. Additionally, there is data available which
does not completely agree with Watling's results. Bar type suppressors
have been shown sufficient for miniguns 71 ,72 while the addition of a
blast suppressor (silencer slots) to other weapons 7 0 did not notice-
ably effect the flash. These inconsistencies indicate the need to

35
correlate the available information with respect to the type of device
used, the weapon fired, the ammunition, the mode of firing, etc. The
lack of this correlation is partially responsible for the lack of firm
design principles for flash suppressors.

The last device shown in Figure 22 is interesting. The device has

an expansion chamber followed by a converging-diverging nozzle. Ap-
parently the nozzle is to function like the diffuser section of a super-
sonic wind tunnel. At its optimal condition, the device should operate
as shown in Figure 23. The flow expands from a sonic muzzle, then is
turned parallel to the chamber walls through a shock. The shock is
reduced to a Mach wave through interaction with the reflected expansions
(note: waves reflect from solid boundaries as the same type waves).
This type of "bounded jet" was investigated by Barakauskas 7 7 who claimed
the resultant flow was similar to that from a Laval nozzle. If this
were the case, an expansion into a chamber with a radius larger than
that of the muzzle would produce supersonic flow. This supersonic flow
then passes through the diffuser. The diffuser throat must be suffi-
ciently large to pass the mass flux of the propellant gas flo,,w. If not,
choking will occur and the chamber will be shocked down to subsonic
velocity with sonic conditions at the diffuser throat. The flow would
then expand to supersonic velocities in the diverging section of the
diffuser which is the situation this design seeks to avoid. Assuming
choking does not occur, the ideal case would result in the development
of a normal shock downstream of or at the throat which will bring the
flow to subsonic velocities behind it. The flow then passes the diver-
gent section subsonically which, according to quasi-one dimensional gas
dynamics, decreases the flow velocity and increases the static pressure.
Through proper design,such a diffuser would eject the flow at low
subsonic velocity and a static pressure which was nearly ambient.

These arguments are somewhat misleading in that they consider a

steady flow. On this basis, calculation shows for a chamber-to-
muzzle radius ratio of 6.0 that a chamber Mach number of 5.50 results.
At this Mach number and with atmospheric ejection, the muzzle stag-
nation pressure could be only about 350 psi to maintain the ideal
state. For higher pressures, the normal shock would be pushed out of
the diffuser resulting in supersonic exit. However, the effect of the
unsteady nature has not been considered. The expansion of the gases
into the nozzle chamber does not occur with steady properties at the
muzzle. Rather, the pressure, temperature and density are continually
dropping; thereby making it conceivable that while the nozzle-diffuser
would be initially supersonic throughout, at later times an ideal flow
situation could evolve. However, Watling makes no attempt to analyze
the flow through these devices.

The detailed investigation of flash suppressor physics requires

more information than is currently available. The basic principles
of the devices are only vaguely known. A unified data base would be
desirable in order to allow for an identification of significant

36
parameters effecting flash in various weapons. A technique to mathe-
matically model the flow field within these devices would be extremely
useful.

C. Blast Suppressors
Gun blast adversely effects the gunner, his concealment, surround-
ing personnel and structures. The blast forms when the excess propel-
lant gas energy is released into the atmosphere. The propellant gas
energy is transferred to the surrounding air by viscous shear, radiation
and heat conduction, but the most rapid transfer of energy occurs as the
propellant gas expands into the atmosphere doing work by displacing the
surrounding air. This displacement at the muzzle propagates as a series
of compression waves to all parts of the fluid. The compression waves
coalesce rapidly into a blast wave. The intensity of the blast is
greatest near the muzzle and is selective in the sense that the highest
intensity occurs where the most air displacement is effected, i.e., to
.the front of the weapon due to the directed kinetic energy of the pro-
pellant gas. As the blast expands radially away from the muzzle, the
propellant gas energy is deposited over an ever increasing surface area
displacing greater volumes of air. This, coupled with the decreasing•
rate of energy deposition at the muzzle as the gun tube empties, causes
the intensity of the blast to drop rapidly as it travels away from the
weapon. At sufficiently large distances, the blast wave decays to a
sound wave.
The control of weapon blast centers on the technique of releasing
the excess propellant gas energy into the atmosphere. The history of
device innovation may be traced through the list of patents 7 8 on designs
to accomplish this energy control. Basically, the devices attempt to
reduce blast through energy absorption, energy dissipation, and energy
containment and controlled release, Figure 24. Energy absorbing devices
use heat transfer from the hot propellant gases to cold metal fibers or
device channel walls. This heat transfer lowers thee gas temperature
thereby decreasing the amount of energy available to perform work.
Dissipative devices attempt to force the propellant gas to perform work
on the muzzle device prior to release into the atmosphere. The work
can be in the form of viscous shear on channel walls or through fibrous
packing, or it may perform work on a movable device such as a rotor.
Energy containment devices consist of chambers into which the propellant
gas expands,decreasing the volumetric energy concentration and allowing
release at reduced pressure, temperature and velocity. These categories
are not at all exclusive and many devices utilize combinations of some
or all of them.
There is great difficulty in translating basic principles such as
these into field-worthy. effective hardware. The state of blast sup-
pressor technology testifies to this fact. Existing theory is largely
empirical in nature,being an off-shoot of research into large scale
37
explosions. Westine 7 9 , 80 develops scaling laws for the blast field
about guns based upon a technique presented by Hopkinson 8 l. The
Hopkinson scaling law was for the blast field about conventional ex-
plosions. Schlenker 8 2 . 8 " devized a technique to compute the blast
field around artillery pieces with and without muzzle brakes. His
analysis used the theoretical results of Brode 8 5 , 8 6 which was produced
for point and spherical source explosions. A

Furrer 8 7 produced an interesting report which commences with an

experimentally derived scaling law for conventional explosions and then
attempts to obtain a similar law for weapon blast. Using quartz micro-
phones, Furrer obtained traces of the blast pressure level at a given
location as a function of time, Figure 25. The first plot represents
a pressure-time trace for the blast of a spherical charge. Furrer ob-
served that both Po and to, the maximum side-on overpressure and
positive phase duration, respectively, scale with the charge weight, Q.
Additionally, he noted that po drops as the inverse of the distance i
from the charge. He gives the following scaling law:

Po 0.34 r
-- r

to 2.85 Q-.!"

where: PO - Kg/cm 2
to - milliseconds
r - meters
Q - Kg TNT

Performiny the same experiments with a gun, he noted the pressure-time

traces were not as simple. Spikes appear on the trace not only due to
the main blast arrival, but also due to flash blast and ground reflec-
tion. However, he noted that at large distances from the weapon, the
blast waves coalesce and resemble the blast from a spherical charge.
In firing 2 an through 15 on cannon,he noted the following scaling law
to be applicable:

Po =0.43.7--

to : 1.80 Q•.*•

where measirements were taken along a line perpendicular to the gun

bore and even with the muzzle. Furrer also measured the blast field
around a gun with and without a muzzle brake noting the resultant dis-
tortion of the pressure contours, Figure 26.

3BI
 Research conducted in the United States during World War II indi-
cated that the scaling law for gun blast was not as simplistic as Furrer
indicates. Westine 79 ,80 reviews this research8 8 , 8 9 and extends it to
develop more comprehensive scaling laws. Both Reynolds 88 and the Navy8 9
noted from experimental data that the peak overpressures about guns would
be equal at identical geometric locations if:

1. All distances were measured in calibers.

2. Projectile (muzzle) velocities were equal.

3 fi=*= (fe = 2
where: M = projectile mass
E = propellant energy
c = gun caliber
k = gun barrel length
4. Neglect heat conduction, viscosity and gravitational effects.
The Navy noted that the scaling could be extended to the impulse, the
area indicated by cross-hatching in Figure 25, if the impulse was
divided by the gun caliber. Westine 7 9 points out that these scaling
laws are similar to the Hopkinson 8 1 spherical charge scaling law:

PO : f(•3
d-:f(r)

where: r = radial distance from charge

d = charge diameter
10= positive phase impulse.
The principal drawback of the Navy scaling laws is the lack of flexi-
bility. The requirements of equal muzzle velocity and scaled projectile
mass, propellant charge and barrel length mean that the blast. field
about a weapon could not be scaled from existing data unless these exact
conditions happened to be available.

Westine notes that Barton, et al 9 0 ,intrcduce an extension to this

scaling law which makes it considerably more flexible. Barton approxi-
mates the blast field about a gun by the blast field around a spherical
charge of reduced energy, w, located a distance ro from the muzzle.
The reduced energy is calculated from:
w = n (E - 1/2 Mv2 )

39
 where: w= reduced energy
n = empirical correlation factor
E= propellant energy
M= projectile mass
v = muzzle velocity.
The distance rn is empirically determined as the approximate location
of the center of a gun blast shock envelope. The overpressures were
then said to be equal at equivalent geometric locations where the dis-
tances were scaled by the cube root of the reduced energy:

This scaling eliminated the restrictions of equal r.,: zle velocity and
scaled projectile mass, propellant energy and barrel length.
Westine 7 9 notes that all of the aforementioned scaling laws do not
consider the ratio of bore length to diameter as being significant. For
this reason, Westine claims the techniques can not scale all types of
weapons equally well, i.e., rifles would not scale with mortars or
pistols. To alleviate this difficulty, Westine proposes an improved
scaling law: PQ
bcc
IOC2_= f (Xy
1 9--
W c' C

where: w = E-1/2 Mv2

x = axial dimension
y = transverse dimension

The functions are then obtained from experiment and presented as a

universal plot of non-dimensional overpressure versus distance in cali-
bers. Westine observed from the experimental results that the over'pres-
sure parameter could be approximated as:

This is seen to be equivalent to multiplying the original parameter by

Z/c. The resulting relation is seen to be quite general and Westine
claims it to be valid to an extremely large variety of guns. His plot
of overpressure is shown in Figure 27. As an indication of the range of

40
applicability possessed by the "universal" pressure field, Westine
cross plots his data. Figure 28 presetits the results obtained by
plotting the overpressure parameter against y/c for x/c = O, Also
presented is data from experiments on a wide variety of weapons. The
correlation is very good, and the range of weapons considered is ob-
viously extensive. Westine indicates that some of the scatter could be
due to the difficulty in calculating the available energy, w. The tests
examined were T1ot condicted to specifically examine the Westine scaling
law, and, therefore, did not contain all of the data required for the
computation, e.g., muzzle velocity and propellant specific energy.

Westine also presernts a universal curve for the side-on, positive

phase impulse which was experimentally determined to scale according to:

Westine notes that this is not a non-dimensional parameter, but should

also contain the speed of sound in air. However, he assumes this to be
a constant and does not choose to include it giving the above parameter
the dimersions of inverse velocity. His results are shown in Figure 29.
Again the data is seen to correlate well when cross plotted and com-
pared with the results of a variety of weapons tests.
The third important blast parameter Westine considers is the time
of arrival of the blast front. Westine writes the scaling law for time
of arrival as:

c C'c' C' c

Westine assumes this function can be approximated by representing the

muzzle blast front as being equivalent to the blast front propagating
from a point charge located at some distance, L/c, along the boreline
away from the gun muzzle. Thus the blast is assumed spherical and the
propagation distance, r, is measured from the center of explosion,
y/c 0, x.'c = L/c. The location of the point charge is empirically
0
determined to be a function of c/w'/3 . The geometry, point charge
location and time-distance results of Westine are shown in Figure 30.
The scaling for time of blast arrival is more complex than overpressure
and impulse scaling. However, this is in part due to the lack of time
of arrival data. Westine's results indicate that once the center of
the blaAt is located as a function of c/wi/ 3 then:

ILf
T, -_ f Y_
c ' ' c' C'
In this sense the scaling is similar to that previously obtained for po
and I.. It is interesting to note that the blast data pnssess a slope
41
alwost equal to the sonic slope. This indicates that after an initial
period of propagttion aL sneeas in excess of the local speed of sound
in air, the blast strength drops quickly and the wave speed, .-r/,,.To,
rapidly approaches sonic velocities.

The scaling laws discussed thus far have been for bare muzzles.
The addition of a muzzle device increases the complexity of the problem
considerably. Schlenker"-"' perfonrs an analysis to approximate the
blast field around guns equipped with muzzle brakes. His approach is
similar to that (f Barton, ot ala.. A reduced gas energy at the time
of shot ejection is computed. This energy is allucated to two puirt
sources located or either side of the m'uzzle brake. The resultant pres-
sure field is then covputed utilizing the point source analysis of
6rode.. Levin9 1 performis an analysis si:nilar to that of Scl.lenker with
the exceptior that Levin allocates the reduced energy not only through
the brake ports but also through the projectile orifice. Levin utilizes
data obtained from recoilless rifle fitings as his hasic source of blast
infotmation. This data en3blcd h,,,,. to include the effects of port flow
Jdirectionaity. .dit,-nally, Levin's analysis allows for the consider-
atic,, of projectil. residency within the brake.

The analysis of the effects of blast suppressor design upon the

blast has not been extensive.
I.wuzzle No significant attemots at ,-.odel-
ing energy absorption or dissipation devices were uncovcred. However,
;;ttemntst have been made iinthe investigation of devices utilizing energy
containment and controlled reliease. bixler, et ali ',studied iulti -baf
fle devices both theoretically and experimentally. Three theoretical
aoproaches are offered: acoustic theory, blast theory and quasi-one-
dimensional flow theory. The acoustic theory,which forns the basis for
the design of conventional (e.a., autoriobile, iruffiers,assumes linear
or sonic wave notion. This is obviously inapplicable to the strong,
non-linear i:-,uzzle blast, and 1Li.ler indicates that the resultant appli.
cation is not valid. The hiast theory put forth in this report con-
siders a situation sm,:ilar to Figure 31.

The blast front expands spnerically at a velocity relative to its

geoietric centero.a to vs. At the saýe ti,,e, the geoometric center
translates at a constant velocity vc. This blast expands irto a rmulti-
baffle blast suppressor with a geomletry as shown. Bixler tOen applies
the quasi-one-dirensional shock propagation theory of 'Whitha•r-- to the
multi-chambered channel of the blast suppressor. The application of
the Whiitham theory to this configuration is som'ewhat tenuous. In it,
area variations are small in order to permit the ut:--ation cf one-
dimensional characteristic theory. For the geometr, _onsidt.reo by
Bixler, the drea changes occur suddenly, cnompletely destroying the une-
dimiensiionality of the flow. However, bixlei na.'Ks use ot Whitha•i,'
relatior, fnr a strong shock expanding into a varyi,;n area channel:

P
 where: p = static pressure
A = area
Ký = function of shock Mach number
= 0.394 for M. and y : 1.4 -

Bixler assumes the blast to propagate as shown in the sequence in

Figure 31. The projectile effect is neglected allowing the blast to
propagate in accordance with the geometry specified earlier. In a
certain time interval, At, the blast front reaches the first baffle
surface, sequence schematic 3. Since the blast front propagates at a
velocity equal to vc + vs, and since the distance from the muzzle to the
first baffle is s, the time interval is calculated to be:
At: s
VC + VS
To use Whitham's pressure formula, Bixler must calculate an equivalent
channel area. This he approximates as being equal to the surface area A
of a spherical blast front of radius vs At,less the bore area: 4
2 T.
A -4,(vs At) 2 .2

Then utiiizing Whitham's strong shock equation, Bixler obtains

L2-v11~~ L 0.394

where: Pm =muzzle pressure

p, = pressure at the exit of the first chamber.

Substitution of the above relation for ýt allows the equation to be A

rewritten:
== fl ) r
f I1 22

In order to obtain a relation for the pressure drop in expanding into

subsequent chambers, Bixler assumes the process occurs identically in

all chambers. Thus assuming identical geometry and constant shock
propagation velocities, he obtains:
Pn n

pm pm Pi Pn-1

F S12 ~VS~ 7 0.394 n

16 v + V

43
The assumptions used to obtain this expression are not justifiable.
His method of sequential expansion of the blast implies that the blast
boundaries remain frozen in a chanber once the front of the blast moves
into the next chamber. This is not a physical reality. The blast con-
tinues to expand, reflecting from chamber boundarics and raising the
chamber pressure drastically. Thus the pressure within the chamber is
not constant ncr is it as low as would be predicted by Bixler's method.
This fact is noted by Bixler in his comparison of the theory with
experiment. He finds that the predicted attenuation is considerably
higher than was experimentally observed.

Bixler's third theory, the shock tube theory, attempts to relate

the blast propagation to a diaphragm rupturing at the muzzle causing the
shock to propagate at a Mach number, M1 o, into the first chamber. He
notes that the gas is not at rest when the diaphragm bursts (bullet
uncorks), but is travelling at the projectile velocity, Vp. He then
obtains a shock Mach number at the muzzle by adding the two:
M.ý
MoFl~ ++ bp.
VP

This type of addition is deceptive. The fact that the gas has a finite
velocity when it breaks the muzzle necessitates that the kinetic energy
be accounted for in the calculation of M,,, i.e., the effect is not
linear. Bixler then applies Whithan's theory to the expansion into the
chamber, calculating th2 resultant shock Mach nu!,,ber:

M) K /2

where: M, = Mach nu~iber at exit to first chamber

Al = cross-sectional area of first chamber

Am = muzzle bore area.

The pressure is then apparently calculated from known conditions at the

1;1uzzle and the conputed Mach number. This prozedure is then continued
through subsequent baffles. Again, the assumiption of quasi-one-
dimensional f.low, no reflections of blast and a linear diaphragm velo-
city addition are not justifiable. The resulting coulparisons with cx-
periments reflects the lack of physical reality in this theory.

An additional problem with this analysis is its inability to

account for projectile presence in the device. Skochko and Greveris"F
conducted extensive tests on a wide variety of suppressor designs.
Their results show that projectile presence effects blast in the sense
that it presents a significant blockage to propellant (;as eApansion.
This blockage is not comnplete and part of the high Pressure gases leak
around the projectile forning a distinct "blow-by" blast m..easured prior

40
to tne main propellant gas blast. The measured blast structure as con-
structed from these experiments is shown in Figure 32. The four distinct
blast pulses were measured near the muzzle. Further away there is a
tendency for them to coalesce into a single disturbance.

Another effect of the bullet presence would be the facilitation

of the development of a quasi-steady flow structure within the device.
With the bullet standing in the exit from a chamber, the gases would be
forced to expand completely into the chamber prior to emergence through
the hole. If the gas flow attains supersonic velocity, shock formation
would cause a loss in stagnation pressure which, coupled with the volu-
metric expansion, would attenuate the resultant propellant gas blast
strength. Such discussion is largely conjecture as no theoretical model
is currently available to handle such a complex gas dynamic problem.
The current basis of suppressor design remains to be empirical laws.

IV. EXPERIMENTAL TECHNIQUES

The complex nature of the shot ejection phenomena is reflected not
only in the difficulties entailed in obtaining a theoretical model, but
also in conducting valid experimental investigations. The fact that
ejection phenomena such as turbulent mixing, chemically reacting flows,
detonation physics, free jet flows, and blast phenomena are the unre-
solved subjects of extensive research efforts in fields quite rermote
from ballistics indicates that the muzzle gas flow field is not in
imminent danger of being completely defined. However, judiciously con-
ducted experiments can and have shed light on isolated phenomena which
are important to the design of muzzle devices.

An important tool in many experimental studies is the muzzle Flow

simulator. These simulators may be divided into two main categories,
steady and unsteady flow devices. The steady flow devices examine the
quasi-steady free jet structure and the interaction of this structure
with muzzle devi-es. The gas tested is generally one of known thermo-
dynamic properties, e.g., air, although trace impurities may be delib-
erately introduced to examine mixing phenomena. lhe elimination of
time dependence and chemical reaction greatly simplifies the investi-
gation of the flow field. However, care must be taken to insure that
the effects of free jet parameters such as pressure ratio, ratio of
specific heats, exit Mach nuimber and flow inclination are understood
and taken into account. The unsteady flow devices include time
dependence but simplify examination of the ejection phenomena through
elimination of chemical kinetics, reduction in physical scale or number
of dimensions, or control of interior ballistics. Such devices as air
guns, light gas guns, and scaled down hardware simulators fall into
this category.

45
 The techniques utilized to survey the muzzle gas flow and associated
phenomena are not unique to this field being taken directly from the
current technology base. The remainder of this Section will be devoted
to consideration of research efforts conducted on the various type muzzle
devices.

A. Muzzle Brakes, Compensators, and Blast Suppressors

To study the muzzle gas flow through a brake, Oswatitsch" constructed

a simulator capable of taking gross force measurements and Schlieren
photographs, Figure 33. The device was two-dimensional so that the in-
ternal flow could be observed. A problem with this type of two-dimet.-
sional device is the influence of wall boundary layer growth. Ladenburg 38
noted this problem in his free jet studies. Additionally, Oswatitsch
assumed that the forces exerted on the brake could be approximated by a
steady flow device. Due to limitations on pumping capacity, the maximumn
I
pressure ratio that could be obtained was Pe/pý = 10.0. This is two
orders of magnitude lower than the actual pressure ratio- however,
Oswatitsch presents a comparison between his previously discussed theory
with com'putations based on this pressure ratio and the resulting force
measurements obtained with this apparatus, Figure 34. Oswatitsch notes
that for small a, i.e., near the muzzle, the measured efficiency factor
is greater than the corresponding theoretical calculation. This is due
to the brake deflecting the flow through a greater angle than that of
the brake surface. At larger values of a, the measured braKe efficiency
is lower than theory. This effect is attributed to losnes around the
brake. To examnine these effects in detail, Oswatitsch designed an
extended brake, Figure 35. The effect at values of a greater than one
is as expected. The increased baffle surTace results in increosed flow
deflection and increased efficiency. However, at lower values of a,
the brake efficiency is lower than the previous experimental values.
This effect is not adecuately explained, but it may be due to an over-
expansion along the baffle surface.

The results of Oswatitsch's research is a brake design similar to

"That shown in Figure 34, with a radius increased to 2.4 calibers. This
device was fabricated and tested on an actual weapon achieving an
efficiency compatible with the steady state theory and experiment,
Figure 16.

A practical design feature considered by Oswatitsch was the de-

tails of the ruzzle brake cover. A cover is necessary in order to
connect the brake to the weapon and to selectively direct the exhausting
gases preventing dust obscuration or blast damage. Oswatitsch utilized
an axisymmetric free jet impinging upon a flat plate baffle, Figure 36.
By varying the cover design and separation from the axis, Oswatitsch
could measure the variation in brake efficiency frow the no cover
efficiency, Figure 37. The effect of the covering on the brake effi-
ciency is complicated; however, two main effects are noted. First,the
covering decreases the brake impingemnent area which tends to decrease

I
I

i iI I I I II I I I I I I I I : ! i•
efficiency. Second, the cover can direct more flow onto the available
brake surface causing greater average rearward deflection. This latter
effect is positive in nature, but it must be noted that increased flow
also passes the projectile hole under this circumistance. Oswatitsch's
result shows cover type IV to be the most effective, and it is this
design which is recommended.

Simultaneous to the work of Oswatitsch, researchers6,4 9 ,s° in the

United States were investigating the design of blast deflectors. Al-
though the design of the two types of devices are quite often identical,
a detailed study of braking action was not attempted. Rather, various
device designs were fired over dust tables and the resulting obscuration
estimated. Slade 6 utilized a caliber 0.30 rifle to simulate the dust
obscuration effects of larger caliber weapons. Successful deflectors
were constructed full size and tested. Slade noted that such scaling
did not always -'ork well. The ohscuration caused by dust clouds is a
subjective measu-ement increasing non-linearly with the gun caliber.
Thus deflectors which worked well on small caliber weapons were not
effective on larger caliber weapons. Slade did note that the effective-
ness of a device to deflect blast and thereby reduce obscuration was
intimately related to the effectiveness of the device as a muzzle brake.
Pursuing this concept, a caliber 0.30 rifle was mounted in a ballistic
pendulum and the resultant recoil energy measured. A flat plate baffle
was placed on the weapon, the plate diameter and distance from the
muzzle was then systematically varied, Figure 38. His measurenents
show similar behavior to those of Smith 5 3 , 5 , shown in Figure 18.
Initially, the curves overlap since each baffle deflects the gases
identically and the projectile hole losses are equal. However, at
greater distances the inability of the smaller diameter baffles to
deflect the gases is shown.

To observe the internal shock structure of the blast deflectors,

Slade made use of the free-surface liquid analogy. This analogy is
based on the fact that there are similarities between the flow of a
compressible fluid and the motion of the surface of a liquid. Cranz'
references the analogy in noting the similarities between the wave
pattern around a moving ship and the spark shadowgraphs of a supersonic
projectile. Slade constructed a water table to allow the observation
of the flow patterns associated with the blast deflector, Figure 39.
The analogy was quite complete, including the bore/chamber profile,
baffle profile, and even a projectile. From water table studies and
the braking criterion, Slade arrived at an optimal turning vane shape,
Figure 40. The design is similar to the optimum brake design of
Oswatitsch, Figure 34, with the addition of a guide nozzle. Slade also
noted that the only method to completely eliminate obscuration was to
duct the gases rearward and eject them vertically at the trunnions.

Pursuing the same goal as Slade, Millikan' 9 and Robinsun•- also

made use of a muzzle flow simulator. However, the device was an
air-powered gun. Later in their programs, a steam-powered gun, w'-,

47
utilized to obtain higher pressure ratios. The air gun consisted of a
caliber 0.50 bore separated from the chamber by a diaphragm. The
chamber was hydraulically pumped to a maximum pressure of 3000 psi at
which time the diaphragm ruptured, and the projectile was accelerated
down the bore, Using wooden projectiles, muzzle velocities up to 1200
feet per second were obtained. The main advantage of this device was
the elimination of contaminants in the gas thereby allowing Schlieren
or shadowgraph techniques to be applied with a resultant optical pene-
tration of the muzzle jet. Robinson 50 utilized the device over a dust
table photographing the resultant dust cloud grcwth with time. His
conclusion was that baffles could be designed which effectively elimi-
nated forward muzzle jet flow; however, obscuration still evolved from
venting the gases to the side. Millikan4 9 conducted a two-dimensional
Schlieren survey of the flow interior to a blast deflector. He used
these results coupled with caliber 0.30 firings to propose certain
radical anti-obscuration concepts. These id(as were later tested by
Munch, et aW! and found to be generally imprictical.

More recently, Smith3-35 has performed e,,tensive research on

muzzle phenomena and, specifically, on muzzle brakes. Like Oswatitsch,
Smith assumes that a good approximation of the thrust on a muzzle brake
can be obtained with a steady flow simulator. However, where Oswatitsch
utilized a two-dimensional simulator with a maximum pressure ratio of
10, Smith constructed an axially synmnetric apparatus with a maximum
pressure ratio of 264. Both tested mainly in air; although Smith did
attempt testing in nitrogen, he noted no significant effect of the
ratio of specific heats. Smith's apparatus is shown schematically in
Figure 41. The technique of measuring thrust differs from that of
Oswatitsch in that rather than measuring the thrust on the brake, Smith
measures the overall thrust on the weapon.

Using this apparatus, Smith conducted extensive tests on a variety

of muzzle brake designs, including discs, cusps, reverse cones, cowled
brakes, open ar;d closed brakes. He also varied brake dimensions, dis-
tance from the muzzle, numbers of brakes and spacing of the brakes.
His results for a single brake are presented as plots of his previously
defined aerodynamic index versus 3xidl location, Figure 18, and versus
pressure ratio, Figure 42. The first plot has been discussed. The
second plot is interesting as iL shows the effect of jet growth on the
brake thrust. At low pressure ratios, the jet has a small diameter,
Figure 8, and most of the mass passes through the muzzle brake projec-
tile hole. As the pressure ratio increases, the jet grows resulting in
greater flow impingement upon the brake surface and increasing aerody-
n airic index. However, a traximum is reached as increasing pressure ratio
results in spillage around the outer edge of the baffle. With further
increase of the pressure ratio this spillage or inco:•plete deflection
effect predominates and the aerodynamic index drops. Additionally,
Smiiths:. took pressure measurements on the front and rear surfaces of
orakes mounted in this simulator. The resultinU ,'easurements coipared
favorably with his steady flow theory, Figure 19.
 Smith's investigation of multi-baffle brakes produced interesting
results, Figure 43. This plot illustrates what economists call the law
of diminishing returns. Due to the deflection of mass by previous
baffles, the additional gain in aerodynamic index by each subsequent
baffle diminishes; e.g., in this case An, = 0.48, ir,2 = 0.28, n3 = 0.10.
A second effect is the decrease in the effect jet pressure ratio at each
baffle due to shock losses. This requires increased spacing to achieve
maximum efficiency, e.g., in this case tl = 1.10, L•x2 = 1.30, Ax 3 = 2.10.
These effects were also noted by Oswatitsch; however, Smith's results
illustrate them quite clearly.

To check the effect of pressure ratio and the steady flow ass~inp-
tion vn the validity of his results, Smith mounted identical muzzle
brakes on a 7.62nm rifle. The resulting aerodynamic indices obtained
from firings with the rifle mounted in a ballistic pendulum compared
favorably with his steady state measurements (generally within 10%).
Smith notes that to examine dimensional scaling, large caliber firings
would be required.

Recognizing that changes in blast effects due to the utilization of

a muzzle brake were important, Smith 5 3 applied a multi-flash shadowgraph
technique to obtain estimates of peak overpressure levels. The pressure
jump across a shock can be obtained from the Rankine-Hugoniot relations
if one is given the shock propagation velocity into a medium of known
thermodynamic state. Firing a 7.62mn rifle, Smith took multiple shadow-
graphs of the shot ejection phenomena on a single photographic plate.
Since the distance travelled by the shock and projectile between flashes
can be measured directly and assuming the projectile velocity to be
constant (2680 - 20 FPS), the projectile travel divided by the projec-
tile velocity will give the time between flashes, thus allowing the
shock velocity to be calculated. The conditions of the am5ient into
which the shot is fired are known; therefore, the overpressure contours
may be computed, Figure 44. The effect of the brake is seen clearly;
the overpressure levels on the lower portion of the figure, i.e., with
the brake, are higher further to the rear than are those without the
brake. Likewise, in the forward direction,overpressure at a fixed lo-
cation is decreased with the application of a muzzle brake. This over-
pressure data suggests an interesting possibility, namely, the combi-
nation of the scaling laws of Westine&9 , e c with data such as this to
provide a "universal" blast field plot for various muzzle brakes. The
inclusion of the brake efficiency into such scaling laws is a further
attractive possibility which could yield a single scaled overpressure
field of wide applicability.

B. Flash Suppressors

Experimental work with flash and flash suppressors has generally

been qualitative in nature, addressing flash largely as a yes-no
phenomena. The bdsic experimental tools •re photography and hu.nan
49
observation73-76. With the basic physical principles of the flash
phenomena in hand, a muzzle device is designed which will supposedly
effect some portion of the flash cycle. The technique to test the
device is, first, place it in a dark room and photograph the weapon
firing. If no flash shows up, place an observer in the room and fire
again. If the observer sees no flash or "reduced" flash then the design
is considered verified.

However, research conducted at Franklin Institutes5-•6 was directed

at obtaining quantitative infornation relating to the nature of gun muz-
zle flash. Their research is summarized in the Antry Handbook on flashi
and in the Midwest Research Institute state-of-the-art surveys• 7 6b
The approach taken by these investigators was two-fold, consisting of
the utilization of steady flow simulators and actual weapons firings.

The steady flow simulator 6 ý discharges air at a stagnation pressure

of 500 psi through a 0.025 inch nozzle into a chamber evacuated to 1.0
psi. This gives a static pressure ratio, Pe/P•, of 264. The simulator
was utilized to study both jet mixing and jet structure. Mixing was
studied through the use of tracer gases in the evacuated chamber. Both
NH-, and CO, were tried, but CO, was used more extensively due to its
less offensive nature. The CO-. was introduced into the chamber until
a 5<.concentration was established in the chamber air. Air from the
nozzle is dried in a KOh tower and posesses negligible CO2 , thus any CO,;
detected in the resulting free jet can be ascribed to entrainment frii
the chamber atmlosphere. The resulting concentration of CO? is then
indicative of the percent of air which would be entrained by a propel-
lant gas jet.

The CO2 concentration is determined by drawing gas from selected

locations in the jet with a hypodermic needle. The gas is passed
through a liquid nitrogen trap in which the CO, is frozen. The reain-
ing air is drawn off and the rate of air flow measured. Knowing the
sampling time, the mass of air taken can then be computed. The mass of
CO: can be determined by allowing it to expand into a known volume at
a known temperature. The resulting pressure is nieasured and utilized
in the equation of state to calculate the CO density. Since the
volume is known, the CO. mass is also calculated. With bcth the mass of
air and CO: sampled, the CO-. concentration is calculated. A schematic
depicting the results of this study is presented in Figure 45. The in-
clusion of the gas-air mixture within the shock bottle is somewhat sur-
prising and may be due to probe interference effects.

A similar flow simulator was constructed by the Midwest Institute-

However, this device utilized a 0.30 inch nozzle and was unsteady in
nature utilizing a burst diaphrag'n to initiate flow. The flow field
produced by this device was surveyed utilizin-: conical tipped probes
to determine local Mach nuwber. This technique, used earlier by Cranz
and Glatzel , consists of placing a conical probe in a su[:ers•ric
flow, taking a Schlieren or shadowgraph picture o4 the probe in the
flow, and measuring the resultant shock angle produced on the tip. Frcrn
supersonic flow theory, this angle can be related to the local Mach
number for a given tip angle. The resulting Mach number distribution is
shown in Figure 46. The results are seen to be highly erratic and com-
parison with theory is almost non-existent. The reasons for this are
first the effect of probe interference is very strong in the vicinity of
the Mach disc and possibly throughout the flow field,which was geometri-
cally small. Secondly, the jet was unheated and the resultant expansion
undoubtedly caused liquefaction of nitrogen in the test air. These
results are presented to indicate the difficulties connected with sur-
veying this flow field and to point out the care that must be given to
analysis of experimental results.

In addition to attempting a survey of the local flow properties and

mixing of the muzzle gas with air, Franklin Institute 6 6 conducted
spectroscopic surveys of the muzzle flash. While the data did not yield
information on radiation from discrete poritons of the flow field, it
did provide valuable output concerning the gross emission properties of
the various flash phenomena. The radiant energy was concentrated mainly
in the infrared region of the spectrum with less than one percent of the
energy being in the visible range. Additionally, impurities in the pro-
pellant gases such as potassium, sodium and compounds of calcium and
copper were primary emitters of distinct spectral bands. Since the
visible spectrum is of concern in flash suppressor design, certain
results obtained by this research will be considered.

Tests were conducted utilizing a caliber 0.50 barrel connected to

a 20mm chamber. To survey intermediate flash, an 85 inch barrel was
used in order to provide internal energy absorption sufficient to elimi-
nate secondary flash. The spectral analysis of this flash phenomena
indicated that continuum emission was the predominant source of radiation,
this shows that the major source of intermediate flash is incandescent
solids. The barrel length was then successively decreased and firings
conducted at each new length to examine the effect of barrel length upon
flash. As expected, secondary flash began occuring more frequently as
the barrel was shortened. It should be noted that the occurence of
secondary flash was intermittent until a barrel length of 61 inches was
reached at which length all rounds flashed.

The barrel was apparently shortened to 45 inches and a spectral

analysis of secondary flash undertaken. The resuits showed a majority
of the radiant energy was emitted in selected bands which are indicated
below:

Band % Visible Energy

CuO, CuOH 50
CaOH 25
Na 7
Continuum 18

51
The utilization of steel bullets rather than the copper jacketed bullets
used in constructing the above table produced an altered distribution
curve. The curve reflected the elimination of copper impurities and a
drop of radiation in those bands (CuO, CuOH).

A survey of the sequence of flash phenomena was undertaken. Uti-

lizing a piezoelectric transducer mounted in the chamber of the gun, the
breech pressure as a function of time was recorded on an oscilloscope.
Superimposed on this trace were the flash intensity measured by a photo-
cell and the shot ejection recorded by a microphone close to the muzzle.
Two cases were observed: with and without secondary flash, Figure 47.
These traces are purely qualitative in nature, but they do indicate the
relative intensity and duration of the flash phenomena. The secondary
flash is seen to occur relatively late in the tube emptying cycle pos-
sessing a great amount of intensity which continues for a long period of
time. Measurement of the temperatures of the two flash phenomena using
the line reversal method shows that intermediate flash possesses a
potassium line temperature of 1250 0 K while the corresponding secondary
flash temperature is 2200 0 K. These observations serve to illustrate
what is already well known; namely, secondary flash is the most severe
flash problem.

No research was uncovered which attempted to apply intensive

f'ow field measurement techniques to the muzzle effluence from weapons
equipped with flash suppressors. As stated earlier, flash suppressor
effectiveness is largely measured as a yes-no parameter. The investi-
gation of flow through these devices and the effect of design changes
on the flow and flash phenomena is definitely a prime requirement for
advancing the state of understanding of flash suppression by physical
means.

C. Blast Suppressors

The understanding of the weapon blast phenomena requires a com-

prehensive catalog of information ranging from the detailed interior
ballistics of the weapon through blast formation due to gas ejection.
Flow through the blast suppressor must be examined, and the resulting
modification of gas efflux related to changes in the blast field. This
latter requirement is not available in the present technology base.
Interior ballistics of small arms are reasonably well in hand, and the
measurement of blast fields can be readily accomplished with currently
available techniques 7 8 ,87,94- 9 8 . However, there have been few signi-
ficant efforts uncovered which attempt to examine the gas dynamics of
blast suppressors.

This lack of a good data base relating suppressor design to blast

prompted Skochko and Greveris 7 8 to test a group of silencers. Utilizing
condensor microphones, the noise levels five meters to the side of these
silenced weapons were measured. Fifteen different silencers were tested;
however, the weapons upon which the devices were mounted varied frxii

52
test to test. This makes it difficult to obtain definite conclusions
as to silencer effectiveness. Additionally, the weapons were firing
reduced charge rounds leading to subsonic exit velocities. This con-
dition is not applicable to the typical field blast suppressor. Their
work does provide insight into the sources of gas dynamically generated
sound, Figure 32, and its inclusion of a multitude of device designs and
patents forms a useful compendium.

Bixler, et a1 9 2 conducted a coordinated experimental investigation

attempting to relate internal design changes to sensed changes in the
blast field. Utilizing a 7.62ni rifle and a cliber 0.45 pistol, a set
of test devices was constructed which allowed systematic variation of
geometric parameters. The tests concentrated on multi-baffle devices
and of necessity required certain restrictions on the number of para-
meters considered. To test the effcct of varying tOe number and spacing
of the baffles upon the weapon muffling characteristics, Bixler con-
structed a fixed length test device which allowed baffler rearrangement,
Figure 48. Sound levels were measured with a condenser microphone, and
the results are shown in Figure 49. The full variations examined by A
Bixler are not included in this figure in order to simplify its inter-
pretation. The blast attenuation is seen to increase rapidly with the
number of baffles. A maximum attenuation is attained and remains con-
stant for a considerable increase in baffles until a gradual decline is
experienced as the baffles begin to fill the expansion chamber thereby
acting more as a channel rather than flow impediments. Bixler notes
that the weight of the device increases with the number of baffles added.
From an initial weight of one pound without any baffles, a weight of two
pounds is incurred
if fifteen baffles
are utilized.
Bixler also considered a device which allowed the overall length to
vary. As expected, the attenuation increased with increasing length.
The diameter of the expansion chamber is also an important parameter.
Therefore Bixler tested a chamber which was 12" long by 20" in diameter.
The resulting attenuation is shown as the closed data point in Figure 49.
The attenuation is quite high. Such a device is obviously impractical;
however, this data point can be considered an indicator of the iaximum
attenuation level which could be achieved for this particular weapon
with a well-designed device.
Bixler also presents spark shadowgraphs showing the internal flow
through a multi-baffle device. The photographs clearly illustrate the
multiple shock reflections from the baffle surfaces and the effect of
"blow-by" in the formation of the multi-blast sequence of Skochko and
Greveris. Bixler's experimental work is a gooa example of the type of
work which is required in the examination of blast suppression. Control
of the test weapon or simulator must be exercised. The blast suppressor
designs should then be parametrically controlled in order to establish
a set of scaling laws which have g(neral applicability.

53
 V. CONCLUSIONS

Various approaches to the theoretical and experimental analysis of

muzzle devices have been presented. However, the complexity of the
unsteady, chemically reacting, muzzle gas flow coupled with the numerous
types of muzzle device for each functional group have prevented the con-
struction of definitive design criteria. Existing data and techniques
suffer from a lack of compilation into a unified, coherent data base.
In general, the technology applied to these devices is not apace with
the current state-of-the-art.

A majority of the research into muzzle devices is directed to the

design of muzzle brakes. The work of Ostwatisch 7 a-7c and Smith5?-5 6
represent the most advanced approaches to the problem. Although both
are quasi-steady analyses, correlation with experimentally obtained data
is excellent. It is noted that these techniques have not been incorpo-
rated in current Army design litcrature"6 . The research into the design
of flash and blast suppressors nas been largely empirical in nature,
producing information applicable to a particular design of muzzle device
and class of weapon, In all cases, the most pressing requirement is for
a compilation of available data. This data base would permit the
development of programs to fill obvious gaps in the design techniques.
Another obvious requirement is the accurate definition of the muzzle gas
and its flow field. A combined experimental and theoretical approach to
provide this essential infornation utilizing currently available tech-
nology is being implemented by the BRL under a SASA-supported muzzle
device prograd.

I5
 to 4

t2

Free Jet

Precursor
tot -- Blast

t3
t4 .•- Shock Path

t5 "-- Projectile Path

FIGURE I FORMATION OF PRECURSOR BLAST

AND FREE JET

5
55

-AX
F. -

I
I
I
=1

F ]
Si
0

LL

a
U-

a
'A

I
S.-

:5

(NJ

S.-

1
 I-J

WW
cc oCJ 4 0
cr. ) LL.

w 0 La0

co ~a- a
LaJ

z
4w
CLa
a- -
0 C0

LLL

00

(0
(0 0o

57
 wI
(n L. U
4 0
LU LU
z ir LL.
0 c-z z
2u 0
z < I-
x 0u
Q)U

LJ

U)U) 0j

L)) L
LLLU

U00
U)
 a: W U:
5;W Ww

030
ww

00

-j

0-
0~
a.
Ln U-

00
X4

ww
00

COO
 0

_ Z
pz p

00

u-i

Cr)

(1) 0)

LLJ 0

LLI tx

(L) wL ;
Mr -i w Li

LLJ co X. 3:
or Ca-t-Q
ALL

U.'-

61-.7 . -
 a
3

U-
vi

1
/

a
C.)

0-

cI

I I
I
I

i
1
62
 I

0i

U-

';'•It'; ' I

LiL

bob3.

63
T4~
 czzz
0./
:CL zz 4

o'I
a,

00
cr-

I
0

c n Cl)=
0 0.Cw
zz
-U<

0 w
Vm

10
Li..

65
 P.0

Me 1.0
1.4
:e4

90-
80"
70-
e
(Deg.) 60
50--

40
.3o"
20"-
10--

0, 14i - - -i , , - I i i
0 200 400 600 800 1000
Pe / Poo

FIGURE 6 PLOT OF DEFLECTION ANGLE VERSUS

PRESSURE RATIO

•.'--'
 40-

30--l

(Deg) A
20-

I0-

Pe / Poo

FIGURE 7 PLOT OF DEFLECTION ANGLE VERSUS

PRESSURE RATIO FOR VARIOUS ye'
Me% [.0 ee 0

t6
 Pe Pooj

Pe- 1-8 POO

Pe 2.3 m

K72.5P
zz_,•

aPe ,, 10.0 POO

UPON FREE
FIGURE 8 EFFECT OF PRESSURE RATIO
JET STRUCTURE
68
 !. MOMENTUM =

a. UTILIZATION :
MUZZLE BRAKES
COMPENSATORS
b. CONTROL:
BLAST DEFLECTORS

2. ENERGY

a. RATE CONTROL:
BLAST SUPPRESSORS (SILENCERS)
b. DISTRIBUTION:
FLASH SUPPRESSORS

FIGURE 9 CATEGORIES OF MUZZLE DEVICES

 w
14

w
6 N-
 IAI
z

E
IC)

0
w

0 0

i-- QLLJ

CCl

0 LfL 'L*N

D
U,-
 TOP

I~~S DE2Z
==J

FRONT

FIGURE 12 GERMAN DOUBLE BAFFLE MUZZLE

BRAKE
 21
I

I.'.

C~e J

FIGURE 13 OSWATITSCH BRAKE NOMENCLATURE

73
 047-
/e.

""
1000

I
0.6-

0.4-' - -
6 10 20 30 In Psm/T
0.--
FIGURE 14 OPTIMIZATION PLOT FOR W

2.5

G ?_.0--

1.50

0°30" 60" 90P ae

FIGURE 15 PLOT OF a- vs ae FOR ZERO PROJECTILE

HOLE LOSS AND :9/7

74
 - C-i
4.8 CAL. _

2.0-48 CAL.0

- -7733
I.ST
a" 550

2.5-

0 0.5 1.0 1.5 2.0 2.5

a ( Calibers)

FIGURE 16 COMPARISON OF THEORY AND

EXPERIMENT FOR OSWATITSCH BRAKE
75
 STREAMLINES
SCORE FLOW REMOVED BY
BRAKE

FIGURE 17 SMITH MUZZLE GAS FLOW NOMENCL.ATURE

.5
.4
-THEORY, y 1.4
77 .3 --- THEORY, 1.67
0 EXPERIMENT 71.4
.2- Pe/Poo: 53

02 4 -6 8 10
x (Calibers)
FIGURE 18 SMITH BRAKE THEORY RESULTS
 Ps 10
T 125
1.125

0.3 ''THEORY
" ''i EXPERIMENT

0.2-

Ps 0.1-

0 0.25 0.50 0.75 1.00 .25 7

I 0.3
X z1.25

Ps O.I-

°0 0.25 0.50 0.75 1.00 1.25 7

FIGURE 19 PRESSURE DISTRIBUTION ON BRAKE

SURFACE IN A STEADY JET
 PREFLASH

SL SECONDARY

MUZZLE
FGR 0GLAOA

FIGURE ?-0 FLASH PHENOMENA

78
 PRIMARY

MUZZLE

CONICAL FLASH HIDER

CONICAL FLASH SUPPRESSOR

6 FRONT

BAR FLASH SUPPRESSOR

21 FLASH PREVENTIVE DEVICES

FIGURE
 CON E 1.77 1
BAR amp===1_ 5.4 -

MIXIG
• dZI SOTS
4"T7 -.-- 5. -

MIXING SLOTS , e 4
+ CONVERGING •,= • 77 1.5
NOZZLE

SILENCER SLOTS ..
_,_--7-- x78

SINGLE SLOT _ ..
,__
__-_ 7) s

SLOTS + CONE SINGLE SLOT "

C7785 \CONE 77- 100 1

CONVERGING •
CONE 77 8,1

CONVERGING -
DIVERGING 7) 300
CONE

FIGURE 22 WATLING FLASH SUPPRESSOR TESTS

 CC.
- w

LL.
U-

A 0

w
(L

0 w
I U)
 Meali Ficbe r

ENERGY ABSORPTION

Vortex Generating Baffles

•ii
ENERGY DISSIPATOR (MAXIM)

ENERGY CONTAINMENT AND

CONTROLLED RELEASE

FIGURE 24 BLAST SUPPRESSOR CLASSIFICATION

P In P,0 r coflst

In totF

000

Broke
WWithout
iWith Broke

180r

FIGURE 26 THE EFFEC1 OF MUZZLE BRAKES ON

BLAST PRESSURES

3-
 1.14

80A
xA
WETINE GN BASTPLA
FIGUE 2 UNVERAL

100~
 0

00
10 = . l0a w.

0 0.22 Rifle
o• stow
0 M-16
1-4 VM-60
1o M-14
I•M-I
o M-? Carbine
* 0.45 Pistol
* 0.45 Grease Gun
A0.50 Machine Gun x o
O 40MM Grenade Launcher
O XM-129 i3r. Launcher(High V)
S20 MM Aircraft Gun
0 195MM Howitzer
5-

5 10 50 100 500

FIGURE a_ CORRELATION OF OVERPRESSURE FACTOR FOR

VARIOUS WEAPONS (x/c=O)
,,V.:
 90
7 5 x10 8
80 0.452 Z T 1.25 xT

y/c 60

50 S~2,71

40 K /

20
10o 5.42 " .5 2•
I000 I I I ! t I I--1
0 . I !
10 20 30 40 50 60 70 80 90
-350-20 -10 0

x/c

, , ,
lOO ci * *"
8000

600
400 0X/Cz 0
400
0 Crossplot
IC El 200 w0

wx~olo aSPIW
I00*0M-160 -•~

80 1
výM-60
60 P M-14
40 X 0.45 Pistol KI

A 0.50 . .
o13/50 Navy Gun~rsp
8"/.50 Navy Gun
x
20
n 14"/50 Navy Gun
o 146/50 Nao? Gun , ,

4 6 10 20 40 60 100 200 400

y/c

UNIVERSAL IMPULSE PLOT AND CORRELATION PLOT

FIGURE 29

I-?
 10. 0 0.30 RIFLE
0 o.30(REO. CG.()
L I o.45 PISTOL
L r C, 8I 3"/54 NAVY GUN

T 6,C 5"/50 NAVY GUN

0
0.005 0,010 0.015
3
C/w1 (IN 3 /LB1 )/

200

160[ 0F
la.o 4Z
r
c - SON IC
WAVE
80

40

0 4 8 12 16
To/c (ms/ IN)

FIGURE 30 TIME OF BLAST ARRIVAL

 ~~ v-qsatI
vs
BLAST FRONT

VcVt
"- --- Vc'•t •Vs _

GEOMETRY OF THE BLAST FIELD

A

III I

SEQUENCE I / S
LBAS
XION
P ROPGA1
:3:
,II

GEOMETRY OF THE BLAST SUPPRESSOR

BLAST

BLASTJ

SEQUENCE OF BLAST PROPAGATION

FIGURE 31 BIXLER BLASZT eSUPPRESSOR MODEL

L r&N
 PRECURSOR TUBE GAS ILAST )

BLOWY DEVELOPMENT

BLOWSY BLAST ®

PROPELLANT GAS BLAST ®

"FLASH BLAST ®

INTERNAL QUASI- FREE JET SHOCK

STEADY SHOCKS

FIGURE 32 SEQUENCE OF MEASURED BLASTS

 ate:

15.5 mm

TOP VIEW

LASS P L ATES
I-G SEX
dv\" ,."i-BRAKE

2 mm

TEST CONDITIONSSIDE VIEW

TSE a 5400R
YE = 1.4
PE m 45 PSIA
Poo c 4.5 PSIA

NO PROJECTILE

FIGURE 33 OSWATITSCH 2-D MUZZLE FLOW

SIMULATOR

I I I I I I I II I I I I I I II I j
 \.0

0.8-
o" o.6-PE/t to1 I
2.0 CAL. 04 oaEO s, -

AND ~~EXPERMN
34 EFFICIENCY FACTOR, THEORY
FIGURE
1.0 1.

0 "J

i • \ 0.8

S•. 0.6-
aEP/P"60
0.4

0 AND EXEIMN
1.0 2.0
1.

a (CALIBERS)

FIGURE 35 EFFICIENCY FACTOR OF EXTENDED BRAKE

 2-0 COVERING

Q 1.25WITH COVERING
WITHOUT COVERING

FIGURE 36 AXISYMMETRIC JET WITH BAFFLE

........
0.2.

n IVI

0.6 0.7 0.8 0 . 1. 1. 13

N0 -- Z (CAL,)

IV

I93-~-0.2

FIGURE 37 COVER TYPES AND EFFECT ON EFFICIENCY

 48.

--- NO BAFFLE- --

4.0- T0
RECOIL x
ENERGY
(FT-LB)
3.2,

2.4- 5. 00"

1 .61

0 2

X(IN.)

FIGURE 38 SLADE RECOIL MEASUREMENTS

,) j A
 BORE BAFFLE
E tMBER

,NtTIAL CHAMBER WATER LEVEL

fI
-. L~.-PROJECTILE

II
FIGURE 39 SLADE WATER TABLE

FIGURE 40 SLADE OPTIMAL TURNING VANE

 SUPPLY
AIR •--.

A I R

HINGED
BELLOWS

SPRING RI0.30o, 0.40" BORE

BALANCE -

SWINGING PLATFORM"-•

MUZZLE DEVICE]

TEST CONDITIONS:
Tse : 540OR

Xe z 1.40

PeMAX" 61 PSIA
POOMIN 1 0.122 PSIA

NO PROJECTILE

FIGURE 41 SMITH MUZZLE BRAKE RIG

 0.6 1.02 2.0 CAL.

0.4 -:

0-
2 5 10 20 50 100
pE/o/
PE PO

FIGURE 42 AERODYNAMIC INDEX VERSUS

PRESSURE RATIO (7 1.0)

:1lJ I I

1.0 - 2 5 CAL.
S~0.8

S~~0.8-S--
f7

0.4 P 264

0.2

05 12 3 45
0

X (CALIBERS)

FIGURE 43 AERODYNAMIC INDEX FOR

MULTIBAFFLE BRAKES
 2.5WITHOU
BRAKE

10 10-200 CAIER0

OVER PRES SURES IN PSI

FIGURE 44 SMITHIS OVERPRESSUREI MEASUREMENIS -

 ALL AIR IXT UTRE

X MUZZLE GAS/AIR

ALL MUZZLE GAS

SHOCK BOUNDARY

FIGURE 45 CONCENTRATION SCHEMATIC

M3.8
M____ 1.8 4.5 2.0 1.5 1.2 1.6 1.4 j< .0
S0 o- 0 0 0 0 o 0

10 THEORY: OWEN & FhORNHILL5"7

9 J p£,, poo : o

7 0 EAPERIMIENT
6 ~PE / Pcoz 359
4 0 TSE 540 0 R
3

20
0 I 2 3 4 5 6 7 8 9 10 I1
X/D

FIGURE 46 MRI MAKCH NUMBER MEASUREME'G'TS AND

COMPARISON WITH THEORY
 PR ESSU RE SBREECH
U

FLLASH INTENSITY

t MILLI SEC)
M

SHOT EJECTION

WITHOUT SECONDARY FLASH

FLASH
INTENSITY SECONDARY FLASH

BREECH
PRESSURE
P ,f- SECONDARY FLASH
(REDUCED SCALE)

WITH SECONDARY FLASH

FIGURE 47 RELATIVE TIME AND INTENSITY OF FLASH

99
 COVERING BAFFLE- INSERTS ýTHREADED ROD

zi J44j ~2.75"A

FIGURE 48 BLAST SUPPRESSOR TEST RIG

30'
0
25 0 00 0
0
ZO~o0
ATTENUATION 0 D 2.75"
(db) 15- 00 D2"
S~I0-

5
o 4 : IA iI i 1,
0 2 4 6 8 10 12 14 16 18 20
NUMBER OF BAFFLES

FIGURE 49 ATTENUATION OF CONSTANT LENGTH

SUPPRESSOR
100
 REFERENCES
I. E. Schmidt and D. Shear, "Experimental Transitional Ballistics at
BRL", Ballistic Research Laboratories Memo Report (to be published).
2. W. A. Hyde, U. S. Navy Ordnance Pamphlet No. 422, 1913.
3. C. Cranz, and B. Glatzel, "Die Ausstromung von Gasen Bei Hohen
Anfangsdrucken," Ann. der Physik, Vol. 43, 1914.
4. C. Cranz, Lehrbuch der Ballistik, J. Springer Verlag, Berlin, 1926.
5. P. Quayle, "Spark Photography and its Application to Some Problems
in Ballistics," Scientific Papers, Bureau of Standards, Vol. 20,
No. 508, 1925.
6. J. J. Slade, "Muzzle Blast: Its Characteristics, Effects, and
Control", National Defense Research Council, Report A-391, March
1966.
7. K. Oswatitsch, "Flow Research to Improve the Efficiency of Muzzle
Brakes':
a. "Part I: Tests on Baffle Surfaces With One-Dimensional Flow",
German Air Research Report 6601, July 1943.

b. "Part II: Efficiency Factor, Momentum Relation, Two-Dimensional

Flow, and Flow With Covering", Research and Development, Army
Ordnance, Report 1001, October 1944.
c. "Part III: The Axial Symmetric Flow Problem, a Comparison of
Tests of Muzzle Brakes Free of Reaction", Gottingen, March 1945.

8. K. Oswatitsch, "Intermediate Ballistics", Deutsche Luft und

Raumfahrt FQ 64-37, DVL Beright 358, December 1964.
9. E. Love, C. Grigsby, L. Lee, and M. J. Woodling, "Experimental and
Theoretical Studies of Axisymmetric Free Jets", NASA TR R-6, 1959.
10. T. Adamson, and J. Nicholls, "On The Structure of Jets From Highly
Underexpanded Nozzles into Still Air", J. of the Aern/SDace Sciences.
Vol. 26, 1959.
11. W. Sheeran and D. Dosanjh, "Observations on Jet Flows From a Two-
Dimensional, Underexpanded Sonic Nozzle", AIAA J., Vol. 6, No. 3,
1968.
12. S. Crist, P. Sherman, and 0. Glass, "Study of the Highly Underex-
panded Sonic Jet", AIAA J., Vol. 4, No. 1, 1966.

101
13. H. Ashkenas and F. Sheman, "Structure and Utilization of Super-
sonic Free Jets in Low Density Wind Tunnels", Rarefied Gas Dynamics
Fourth Symposium, Vol." II, Academic Press, New York, 1066.
14. S. Pai, Fluid Dynamics of Jets, D. Van Nostrand, New York, 1954.

15. M. Moe and B. Troesch, "Jet Flows with Shocks", ARS J., 1960.

16. K. Bier and B. Schmidt, "7ur Form. der \!erdichtungsstose un Frei

Expandierenden Gasstrahen", Z. Angen. Physik, 1961.
17. M. Sibulkin, W. Gallaher, "Far--Field Approximation for a Nozzle
Exhausting Into a Vacuum", AIAA J., Vol. 1, No. 6, 1963.

18. A. Charwat, "Boundary of Underexpanded Axisymmetric Jets Issuing

Into Still Air", AIAA J., Vol. 2, No. 1, 1964.

'i9. F. Albini, "Approximate Calculation of Underexpanded Jet Structure",

AIAA J., Vol. 3, No. 8, 1965.
20. J. Hill and J. Draper, "Analytical Approximation for the Flow From
a Nozzle Into a Vacuum", J. Spacecraft, Vol.3,No. 10, 1966.
21. E. Wedepeyer, "Asymptotic Solutions for the Expansion of Gas Into
Vacuum", Ballistic Research Laboratories Report 1278, April 1965.
(AD 465 240)
22. F. Boynton, "Highly Underexpanded Jet Structure: Exact and Approxi-
mate Calculations", AIAA J., Vol. 5, No. 9, 1967.

23. K. Korkan and L. Knowles, "Boundary of Underexpanded Axisymmetric

Jets in Still Air", AIAA J., Vol. 7, No. 2, 1969.
24. E. Muntz, B. Hamel, and B. Maguire, "Some Characteristics of Exhaust
Pluwe Rarefaction", AIAM J., Vol. 8, No. 9, 1970.

25. G. Simons, "Rarefaction Effects in High-Altitude Rocket Plumes",

AIAA J., Vol. 10, No. 3, 1972.

26. D. Easbnan and L. Radtke, "Location of the Normal Shock Wave in

the Exhaust Plume of a Jet", AIAA J., Vol. 2, No. 4, 1964.
27. C. Lewis and D. Carlson, "Normal Shock Location in Underexpanded
Gas and Gas-Particle Jets", AIAA J., Vol. 2, No. 4, 1964.
28. M. Werle, D. Shaffner, and R. Driftmeyer, "Freejet Terminal Shocks",
AIAA J., Vol. 8, No. 12, 1970.

29. M. Abbett, "Mach Disk in Underexpanded Exhaust Plumes", AIAA J.,

Vol. 9, No. 3, 1971.

102
30. J. Hartman and F. Lazarus, "The Air Jet With a Velocity Exceeding
That of Sound", Phil. Mag. and Journ. Sci., Ser. 7, Vol. 31, No.
204, 1941.
31. E. Hammer, "Muzzle Brakes, Volume I: History and Volume II: Theory",
Franklin Institute, June 1949. (AD 111481) XE

32. M. Hugoniot, "On the Diversified Movement of a Gas Compressed in a

Reservoir Which Empties Freely Into the Atmosphere", Comptes Rendue, A
Vol. 103, 1886.

33. A. Rateau, "The Theory of Muzzle Brakes", Memorial de L'Artillerie

Francaise, Vol. XI, 1932; Comptes Rendue, Vol. 168, 1919.

34. L. Kazincky, "Theoretical Considerations in Regard to the Con-

struction of MuzzlE f3rakes", Memorial de L'Artillerie Francaise,
Vol. IX, 1930.

35. E. Ravelli, "A Study of the Theory of Muzzle Bra:es", Rivista di

Artiglieria E Genio, January 1929.

36. M. Gabeaud, "The Theory of Recoil Brakes", Memorial de L'Artillerie

Francaise, Vol. XI, 1932.

37. M. Gentil, "A Summary of the Studies of Muzzle Brakes", Memorial de

L'Artillerie Francaise, Vol. XV, 1936.
38. J. Corner, "The Internal Ballistics of a Gun After Shot Ejection",
A.C. 4502/1.B 201/Gn. 227, OSRD Liaison Office Reference No.
WA-859-1.

39. Engineering Design Handbook, Ballistic Series, Interior Ballistics

of Guns, AMCP 706-150, U.S. Army Materiel Command, February 1965.

40. P. Baer and J. Frankle, "The Simulation of Interior Ballistic

Performance of Guns by Digital Computer Program", Ballistic Research
Laboratories Report 1183, December 1962. (AD 299 980)

41. P. Vottis, "Digital Computer Simulation of the Interior Ballistic

Process in Guns", Watervliet Arsenal, WVT-6615, October 1965.
(AD 805 248)

42. 0. Heiney, "Simplified Interior Ballistics of Closed Breech Guns",

Air Force Armament Laboratory, AFATL-rR-67-42, April 1967.
(AD 822 447)

43. J. Corner, Theory of interior Ballistic of__Lu, John Wiley and

Sons, New York, 1950.

44. F. Hunt, Internal Ballistics, Philosophical Library, New York, 1951.

103
 A

45. C. Thornhill, "A New Special Solution to the Complete Problem of

Interior Ballistics", Royal Anrament Research and Development
Establishment, Report'2/67, April 1967. (AD 818 009)

46. Engineering Design Handbook, Gun Series, Muzzle Devices, AMCP

706-251, U.S. Amy Materiel Command, May 1968.

47. "Design Data Sheet, DH5", Armament Design Department, Ministry of

Supply, Great Britain, 1943.
48. V. Dear and F. Smithies, "Tests of Muzzle Brakes for the Q.F. 6 PR.,
7 cm. Gun", Gun Design Committee, A. C. 4162/GN. 260, June 1943.

49. C. Millikan, E. Secher, and P. Buhler, "A Study of Blast Deflectors,

Final Report", National Defense Research Council, Report A-351,
October 1945.

40. E. Robinson and E. Wilson, "Reduction of Smoke and Bl3st Obscuration

Effect", National Defense Research Council, Report A-325, May 1945.

51. N. Munch, M. MacKay, and R. Greengrove, "80rmm Gun Blast Deflector

Development", Power Generators, Inc., Report 1398, September 1955.
(AD 836 751)

52. P. Townsend, "Development of a Gas Gun to investigate Obscuration

Effects", WECOM, Report 66-3281, November 1966. (AD 804 815)
:3. F. Smith,,"Model Experiments on Muzzle Brakes", Royal Armament
Research and Development Establishment, Report 2/66, June 1966.
(AD 487 121)
54. F. Smith, "Model Experiments on Muzzle Brakes, Part III: Measure-
ment of Pressure Distribution", Royal Armament Research arid
Development Establishment, Report 3/68, February 1968. (AD 845 519)

55. F. Smith,"A Study of Gun Blast in Relation to That From a Moving

Source", Royal Armament Research and Development Establishment,
Memo 28/70, 1970. (AD 878 659)

56. F. Smith, "Loads on Surfaces Due to Gun Blast", Royal Amament

Research and Development Establishment, Memo 29/70, 1970.

57. P. Owen and C. Thornhill, "The Flow in an Axially Syrnnetric Super-

sonic Jet From a Nearly Sonic Orifice Into a Vacuum", Royal
Armament Research and Development Establishment, Report 30/48,
1948. (AD 57 261)

58. R. Ladenburg, C. Van Voorhis, and J.

Winckler, "Interferometric
Studies of Faster Than Sound Phenomena, Part Ii: Arildysis of
Supersonic Air Jets", Physics Review, Vol. 76, No. 5, September
1949.
104
59. A. Horton, J. Aunew, et al, "Basic and Technical Work on Military
Propellants", Frankli, Institute, Final Report F-2092-15,
December 1949.

60. E. Stephens, G. Wachtell, and S. Carfagno, "Physical Suppression of

Gun Muzzle Flash", Franklin Institute, Interim Report 1-23C4,
September 1953. (AD 21 736)

61. J. Agnew, "Activities in Connection with Basic and Technical Work

on Infrared Radiation From the Flash of Military Propellants",
Franklin Institute, Progress Report 1897-10, February 1948.

62. L. Herczeg, "Investigation of the Spectral Emittance of Military

Propellants", Franklin Institute, Final Report F-2258, August 1954.
(AD 66 009)

63. S. Carfagno and G. Wachtell, "Emergence of Muzzle Gases Between

Suppressor Bars", Franklin Institute, Interim Report 1-A1828-5,
January 1958 (AD 300 766)

64. S. Carfagno and 0. Rudyi, "Relationship Between Propellant Compo-

sition and Flash and Smoke Produced by Combustion Products (U)",
Franklin Institute, Final Report F-A2132, December 1960.
(AD 321 118)(Confidential)

65. S. Carfagno, "Handbook on Gun Flash (U)", Franklin Institute,

November 1961 (AD 327 051) (Confidential)

66. Engineering Design Handbook, "Spectral Characteristics of Muzzle

Flash", AMCP 706-255, U.S. Army Materiel Command, June 1967.

67. and Flash in Small Arms

W. Gillam, S. Fisher, and H. Young, "Smoke1948.
Ammunition", Midwest Research Institute,

68. H. Young, "Smoke and Flash in Small Arms Ammunition, 1948-1955",

Midwest Research Institute, 1954. (AD 88 537)

69. J. Fay, "The Physical Mechanism of Gun Flash", Midwest Research

Institute, Progress Report 1, June 1954.

70. E. Hodil and D. Merrill, "Weapoo Development, Multiple Flechette

Weapon Systen Development Contract", Winchester-Western, October
1970.

71. J. Wyatt, "Flash Hider for GAU - 2B/A (7.62 Miniguns)", Eglin AFB,
June 1969. (AD 864 225L)

72. C. Burket, "Integrated Engineering and Service Test of the Flash

Suppressor Assembly for M134, 7.62mw Miniguns", U.S. Anmy Aviation
Test Board, August 1970. (AD 875 334L)

105
73. R. Ladenburg, "Report on Muzzle Flash", Ballisti,: Research
Laboratories Report No. 426, 1943.

74. R. Ladenburg, "Suppression of Muzzle Flash From Caliber, .30 Machine

Gun', Ballistic Research Laboratories MR 404, 1945. (AD 492 761)

75. R. Ladenburg, "Studies of the Muzzle Flash and its Suppression",

Ballistic Research Laboratories Report 618, 1947, (AD 224 762)

76. W. Watling, "Acoustical and Visual Attenuation Through Dynamic

Regulation of Muzzle Gas Flow", Springfield Armory, June 1966.
(AD 634 649)

77. A. arakauskas, "Sudden Expansion of a Bounded Jet at High Pressure

RaL"," AIAA J., Vol. 2, No. 9, September 1964.

78. L. Skochko and H. Greveris, "Silencers", Frankford Arsenal, Report

R-1896, August 1968.

79. P. Westine, "Structural Response of Helicopters to Muzzle and

Breech Blast. Volume I of Final Technical Report: Blast Field
About Weapons". Southwest Research Institute, Report 02-2029,
November 1968.

80. P. Westine, "Modeling the Blast Field Around Guns and Conceptual
Design of a Model Gun Blast Facility", Southwest Research Institute,
Report 02-2643-01, September 1970. (AD 875 984).

81. B. Hopkinson, British Ordnance Board Minutes 13565, 1915.

82. G. Schlenker, "Contribution t-. the Analysis of Muzzle Brake Design",

Rock Island Arsenal, Report 62-1794, May 1962.

83. G. Schlenker, "Theoretical Study of the Blast Field ot Artillery

With Muzzle Brakes", Rock Island Arsenal, Report 62-4257, December
1962.
84. G. Schlenker and S. Olson, "A revised Method of Computing the
Overpressure Isobars in the Blast Field of Artillery", Rock Island
Arsenal, Tech. Note 65-2, July 1965. (AD 817 024L)

85. H. Brode, "Point Source Explosion in Air", Rand Corporation

RM-1824-AEC, December 1956. (AD 133 030)

86. H. Brode, "A Calculation of the Blast Wave From a Spherical Charge
of TNT", Rand Corporation, RM-1965, August 1957. (AD 144 302)

87. W. Furrer, "The Acoustics of Detonatl,i", Schweizer Archiv, July

1946.

106
88. G. Reynolds, "Muzzle Blast Pressure Measurements", Princetun
University, Report PMR-21, April 1944.

89. U.S. Navy Gun Blast Committee, "Survey of Research on Blast (U)",
First Interim Report, 1946 (Confidential).

90. H. Bartor,, R. Heyman, and T. Schiffman, "Correlation of Muzzle

Blast Pressures Over Flat Surfaces", Amour Research Foundation,
1956.

91. S. Levin, "Concept and Feasibility Studies of Muzzle Brake Blast

Suppression Devices for 105mm and 155amm Howitzers", Ordnance
Engineering Associates, Report P.N. 2070, February 1964
(AD 601 728)

92. 0. Bixler, H. Kahlke, R. Kaplan, and J. Van Houten, "Analytical

and Experimental Studies of Weapon Muffling", Ling Tempco Vought,
TR 0-71200/7TR-123, August 1967. (AD 821 879L)

93. G. Whitham, "On the Propagation of Shock Waves Through Regions of

Non-uniform Area or Flow", J. Fluid Mech., Vol. 4, 1958.

94. G. Garinther and J. Moreland, "Transducer Techniques for Measuring

the Effect of Small Arms Noise on Hearing", U.S. Am-ny Human
Engineerinq Laboratories, TM 11-65, July 1965.

95. N. Ethridge, "A Procedure for Reading and Smoothing Presure-Tinie

Dat3 Fý-cm H.E. ;.u Nuclear Ex sions", B'Iistic Research
Laboratories Memorandum Report 1691, September 1965. (AD 479 775)
96. R. Reisler, R. Raley, and D. Lefevre, "Air Blast Measurements
Recorded by Standard and Developmental Instrumjentation", P.O.R.
4030, Operation Sailor Hat, NDA, July 1967. (AD 819 459)

97. Materiel Test Procedure, "Airblast Pressure Measurement-Electric",

Test and Evaluation Command, July 1970. (AD 875 696)

98. C. Kingery, "Parametric Analysis of Sub-Kilotun Nuclear and High

Explosive Air Blast", Ballistic Research Laboratories, February
1968. (AD 833 698)

10C
 DISTRIBUTION LIST

No. of No. of
Copies Organization Coies Organization

2 Commander 1Commander
Defense Documentation Center U.S. Arm), Materiel Command
ATTN: TIPCR ATTN: AACRD-W
Cameron Station SOGI Eisenhower Avenue
Alexandria, Virgini '22314 Alexandria, Virginia 22304

1 Director 1 Commander
Defense Nuclear Agenc," U.S. Army Aviation Systems
Washington, DC 2030:" Command
ATTN : AMSAV:-E
I Commander 12th & Spruce Streets
U.S. Army Materiel Command St. Louis, Mtissouri 63166
ATTN : AMCDL
5001 Eisenhower Avenue I Director
Alexandria, Virginia 22304 U.S. Army Air 'Mobility Research
and Development Laboratory
Ames Research Center
Commander
U.S. Army Materiel Command .- cffett Field, California 9403;
ATTN: AMCRD, MG S. C. Meyer
5001 Eisenhower Ave-nue 2 Commander
U.S. Army Electronics Command
Alc::andria, Virginia 22304
ATT.N AMSE.- CE-
A
AMSEL-RD
1 Commander
U.S. Army Materiel Command Fort Moonmouth, New Jersey 07703
A.TTN: AMCRD, Dr. J.\'.R.Kaufman
5001 Lisenhower Avenue C1Co0oZnj.er
U.S. Army Missile Command
Alexandria, Virginia 223f-4
A.T'fN : N.kSM I -R
I -RBI,
.'M5l
Commander sS5F09
Redstone Arsenal, Nlabama
U.S. Army Materiel Command
AITN: NMCRD-TP
5 Commander
$001 Eisenhower Avenue
U.S. Armoy Missile Command
Alexandria, Virginia 22304 .MsI -RDIK
ATTN
Mr. R. Bec't (4 cys)
Commander Mr. R. Deep
U.S. Army Materiel Command
Redstone Arsenal, Alaibama .ýM509
ATTN: AMCRD-TE
5001 Eisenhower Avenue
1 Commander
Alexandria, Virginia 22301
U .S. .Arwr'x lank ,At @mot i \e
Comm and
Commander ,\I*N: AMS:V,- RIIF L
U..3. Army Materiel Command
ATTN :,MCRD-MfI" arr:,• i, ,i.m i
5001 Eisenhower Avenue
Alexandria, Virginia 223041
I C9
 DISTRIBUTION LIST

No, of No. of
Copies Organization Copies Orani zat ion

2 Commander 4 Commander
U.S, Army Mobility Equipment U.S. Arm), Weapons Command
Research & Development Center ATTN: AMSW`E-RE1
ATTh,: Tech Docu Ctr, Bldg. 315 AMSWE-RES
AIISME-RZT AMSWE-RDF
Fort Belvoir, Virginia 22060 AMCPM-RS
Rock Island, Illinois 01202
2 Commander
U.S. Army Munitions Command I Commander
ATTFN : AMSMU-RE US Army) Watervliet Arsenal
AMSMU-RE-M Watervliet, New York 12189
Dover, New Jersey 07,01
I Commander
5 Commander U.S. Army Safeguard Systems
U.S. Army Frankford Arsenal Command
ATTN:
SMr.
SNIUFA-U2lOO
J. Mitchell
SMUFA-U31 )0 1
Huntsville, Alabama 35804

Di rector
~1iMr. S. Fulton U.S. Army Advanced Materiel
SMIJFA-U3300 Concepts Agency
Mi. S. Hirshman 2461 Eisenhower Avenue
SMUFA-U3300 Alexandria, Virgin.a 22314
Mr. A. Cianciosi
L4100-1S0-2 1 Commander
Mr. C. Sleischer, Jr. U.S. Army Hlarry Diamond
Philadelphia, Pennsylvania Laboratories
19137 ATTN: ANLXDO-TP/O02
W¢ashington, DC( 20438
7 Commander
U.S. Army Picatinny Arsenal 1 Commander
ATTN: SMUPA-DR-D, U.S. Army Mnteriels and
Mr. S. Wasserman Mechanics "esearch Center
SMUPA-DR-,<, Mr. A. Loeb ATTN: AJMXMR-..vlL
Mr. D. Mertz Watertown, MUssach,'ott s 02172
SMUPA-D, Mr. Lindner
SMUPA-V, Mr. E. Walbrecht 1 Con~mander
Mr. S. Verner U.S. Armv Natick Lahor:itories
SMUPA-VE, Dr. Kaufman A'rTN.: ,.IXR-, Dr. ). SielinP
Dover, New Jersey 07801 Natick, Massachusetts 017o2

PLASTEC I DeLuty ..Assistant Secretary

U.S. Army Picatinny Arsenal of thie Army (RWP)
ATTN: SMUPA-FR-M-D ATTN: Mr. C. L. Poor
Dover, New Jersey 07801 Washington, DC 20310

110
 Di3s'rRIrBUrION LIST

No. of
No. of
0rai z at i~on Sa Ogiizat i on
Commander
I ,S, Army Research Office 4 Director
U.S. Naval Research Laboratory

(Durham) ATTN: Tech Info Div

CRD-AA-EII Code 7700, Dr. A. Kolb
ATTM:
Box (31, Duke Station Code 7720, Dr. E. McClean
2770U Mr. I. vitkovitsky
Durham, North Carolina
Washington, DC 20390
SDirector
U.S. Army Advanced BalliStics 3 Commander
missile Defense Agency U.S. Naval Weapons Laboratory
P. 0. Box 1500 AT'N: Code CX, Dr. W. Kemper
35809 Code V\215, Mr. J. Edwards
Huntsville, Alabama
Code MIAS, Mr. M. Jones
Commanider Dahlgren, Virginia 22448
3
U.S. Naval Air Systems Conmmand
ATI'N : AIR-604 2 Commander
Washington, DC 203b0 U.S. Naval Ordnance Station
Indian Head, Maryland 20640
3 Conander
U.S. Naval Ordnance Systems 1 AEDC" (AL1")
Command Arnold AS
Tennessee 37389
ATIN: ORD-9132
Washington, DC 203oL)
2 AD'TC (ADBPS-12)

Commander and Director vglin :fl-

2
U.S. Naval Ship Research and Florida 32342
Development Center
ATTN: Tech Lib I APATIL (PLR)
Aerodynamic Lab -glin AFB
Florida 324
Washingtun, DC 20007

Commander I AF:ATL (DLRO)

3
U.S. Naval Weapons Center E'glin AFB
753, Tech Lib Florida 3?542
ATTN: Code
Code S0704,
Dr. W. Haseltine 1 AFVAIL ([DLRV)
Code 3007, Mr. A. Rice Eglin AI-A
California 935S5 1lorida 3254,
China Lake,
Commander A\FVATL (DIY.'.A, Mr. V. Burgess;
4
'lech Oib)
U.S. Naval Ordnance Laboratory
ATTN: Code 031, Dr. g. Lobb [d1lin AFIB
Code 312, Mr. R. Regan Vioida 325 4
Ir.- S
S ilactiog..
Code 730, Tech 2 ib
Silver Spring, Maryland 2091'0

Iii
 DISTRIBUTION LIST

No. of No. of
Copies Organization Copies Organization

AFWL (DEV) 2 Director

Technical Library National Aeronautics and
Kirtland AFB Space Administration
New Mexico 87117 George C. Marshall Space
Flight Center
1 ARD (ARIL) ATTN: MS-I, Lib
Wright Patterson AFB R-AERO-AE, Mr. A. Felix
Ohio 454.33 Huntsville, Alabama 35812

SARL 1 Director
Wright-Patterson AFB National Aeronautics and
Ohio 45433 Space Administration
Langley Research Center
ASt) (ASBEE) ATTN: NIS 185, Tech Lib
Wright-Patterson AFB Langley Station
Ohio 45433 Hampton, Virginia 2S365
RTD (FDFE, Mr. Sedderke) 1 Director
Wright-Patterson AFB National Aeronautics and
Ohio 45433 Space Administration
Lewis Research Center
Director ATTN: MIS 60-3, Tech Lib
National Bureau of Standards 21000 Brookpark Road
ATiN: Tech Lib Cleveland, Ohio 44135
U.S. Department of Commerce
Washington, tLC 20234 1 Advanced Technology Laboratories
ATTN: Dr. J. Erdos
Ileaduarters a100 Jericho Turnpike
National Aeronautics & Space Jericho, New York 11753
Administration
A[TN: Code EP, M. Adams ARO, Inc.
Washington, DC 20546 ATTN: Tech Lib
Arnold AFS
Director Tennessee 37389
NASA Scientific 6 Technical
Information Facility 1 F- :hnical Director
ATTN: SAK/IDL :It Firearms Corporation
P. 0. Box 33 150 Hluvshore Avenue
College Park, Maryland 20740 Hartford, Connecticut 1410(l
IDirector 1 Gener. I Electric Corporation
•.t PL'ulsit"on Laboratory Ar.ae.t Div:.ii"
AVI',N • Terh Lib ATTN : Mr. R. Wht e
4800 Oak (;rove Drive Lakeside Avenue
PIasdena, California 91i03 Bur.lington, Vermont 0540 1

112
 DISTRIBUTION LIST

No. of No. of
Organization C Orsanization
Cop-es
I Massachusetts Institute of
I Winchester-Western Division
Olin Corporation Technology
,tr'VN; Mr. D. Mettill Department of Aeronautics
New Haven, Connecticut 06S04 and Astronautics
ATTN: Tech Lib
Sandia Corporation 77 Massachusetts Avenue
I
ATTN: Aerodynamics Dept Cambridge, Massachusetts 02139
Org. 9320, R. Maydew
P. 0. Box 5800 1 ,irector
Guggenheim Aerospace Labs
Albuquerqt'e, New Mexico 87115
New York University
New York Heights
I California Institute of
'ew York, New York 10053
Technology
Aeronautics Department
Prof. H. Liepmann I Ohio State University
ATTN:
Department of Aeronautics
Pasadena, California 91102
and Astronautical Engineering
.\TTN: Tech Lib
Guggenheim Aeronautical
Columbus, Ohio 43210
LaboratoryInstitute of
California
I Polytechnic Institute of
Technology
Brooklyn
ATTN" Tech Lib
91104 Graduate Center
Pasadena, California Dr. G. Morretti
ATIN:
Farmingdale, New York 11735
3 Calspan Corporation
ATTN: Mr. .1. Martin
"Dr. W. Wurster 1 Director
Dr. G. Skinner Forrestal Research Center
[
Princeton University
P. 0. Box 235
Princeton, New lersey 08S40
Buffalo, New York 14221
2 Franklin Institute IPrinceton University

I AnlN: Dr. Carfagno

Dr. Wachtell
Race & 20th Streets
Vorrestal ImAboratories
ATTN: Dr. S. I. Uheng
Princeton, New Jersey OS-, 10

Philadelphia, Pennsylvania
University of Michi 'n
19103
Department of Aeronautical
t'nginerring
1 Director Dr. A. Kuethe
NTq'TN:
Applied Physics Laboratory
Dr. M, .Sichle1
The Johns liopkins University
East Engineering Building
1
8621 Georgia Avenue
20910 Ann Arbor, Miich i lan 4RC104
Silver Spring, Maryland

1 1%
 DISTRIBUTION LIST

No. of
CoSpi Or anization

Aberdeen Proving Ground

Ch, Tech Lib

Marine Corps Ln Ofc
CDC Ln Ofc
CO, USALWL
CO, USASASA (S cys)
ATlTN Mr. B. Jezek
CDR, USATECOM
ATTN: AMSTE-BE
Mr. Morrow
NMSTE-TA-R
Mr. Wise
Dir, USWASAA
ATT•: Dr. J. Sperrazza
 Security Classificationl
DOCUMENT CONTROL DATA.- R &D
(SOCuFIly Classificatlion Of fill&, body of abstract and Indexing seinuotallofl must be entered wh~en the. o~.rall report L. classified)
I.ORIGINA TINQ AC TIVITY (Corporate e.lhot) &a..REPORT SECURI TY CLASSIFICATION

U. S. Army Ballistic Research Laboratories Unclassif ied

Aberdeen Proving Ground, Maryland 21005 11b. 000OUP

3 REPORT TITLE 4
MUZZLE DEVICES, A STATE-OF-THE-ART SURVEY. VOLUME I: HARDWARE STUDY

4. OESCRIPTIVX N4OT9S (7~p* off

,epu and Inchleiv. dala#)

5AU THOR(S (Firt. name*. middle Mintal. is&# name.)

LDWARD M. SCHMIDT

4. REPORT OATE 70. TOTAL NO- OF PAGES Tb. No. OF REFS

FEBRUARY 1973 114

Ga. CON TRAC T OR GRANT NO. PA. 01siGINATOWS REPORT NUMBER(S)

bI. PROJECT . RDT&E 1J562604Ab07, BRI. MEMORAJNDUMI REPORT NO. 227t

c. 9b. OTHgR HEPORT NO($) (Any other nueMbetO that troy be eAssined

d.
10 DISTRIOUTION STATWEUER~T
Di St rib Ution limited to UIS Government agencies only. Other requests for tli
docunieni must be referred to Director, UiSA Ballistic Research Laboratories, ATIN:
AMKBR-XM-SE, Aberdeen Proving Ground, Maryland 2100S.
II. SUPPLEMENTARY NOTES ISPNORING MItl TARY ACTIVITY

U. S. Army Materiel Command

I Washington, D. C. 20315
13. AMSTRAC T

A review of the salient literature addressing the engineering design Of MuIZz1e

devices is presented. Both theoretical and experimental techniques applicable
to specific hardware items are discussed. The types of devices considered incluide:
muzzle brakes, compenlsators, blast deflectors, blast suppressors, and flash
5iuppressors. The second volume of this report addresses specific gas dynamic
theories which are applicable, bat, in general, have not buten iitil-ý:cdl ;1 thL,
analysis of muzzle gas flow fields.

DID ~..,,1473 _______ _11_1_I'll)

Security Classification
 UNCLASS lIFI )
Security Classiftcation

LINK A LINK 8 .INK C

KEY WONOI
moLE *1 ROLE WT POLK WT

Muzzle Devices
Muzzle Brakes
Silencers
Flash Suppressors
Blast Suppressors

INC(LASS I1F 11]1)

Lsecurity Clssifii caiton