AD-A234 090

AD

TECHNICAL REPORT ARCCB-TR-90030

D YNA MIC S TRA IN WA VES-

A DEVELOPMENT PERSPECTIVE

R. HA SENBEIN B. ARTUS

A. GABRIELE G. CUNNINGHAM

D. FINLA YSON R. GAST

0%ELECTE ft
OCTOBER 1990 19MAPR2 21991,

IUS ARMY ARMAMENT RESEARCH,

_ DEVELOPMENT AND ENGINEERING CENTER
4 1 CLOSE COMBAT ARMAMENTS CENTER
BENEiT LABORATORIES
WAERVLIET, N.Y. 12189-4050'

APPROVED FOR PUBLIC RELEASE; DISTRIBUTION UNLIMITED

94 1029
 DISCLAIMER

The findings in this report are not to be construed as an official

Department of the Army position unless so designated by other authori:ed

documents.

The use of trade name(s) and/or manufacturer(s) does not constitute

an official indorsement or aporoval.

DESTRUCT ION NOTI C

For classified documents, follow the procedures in DoD 5200.22-M1,

industrial Security Manual, Section 11-19 or DoD 5200.1-R, Information

Security Program Regulation, Chapter IX.

For unclassified, limited documents, destroy by any method that will

prevent disclosure of contents or reconstruction of the document.

For unclassified, unlimited documents, destroy when the report is

no longer needed. Do not return it to the originator.

SECURITY CLASSIFICATION OF THIS PAGE (When Date Entered)
REPORT DOCUMENTATION PAGE READ INSTRUCTIONS
REPORT__DOCUMENTATIONPAGE_ BEFORE COMPLETING FORM
1. REPORT NUMBER 12. GOVT ACCESSION NO. 3. RECIPIENT'S CATALOG NUMBER

ARCCB-TR-90030 i
4. TITLE (and Subfttle) 5. TYPE OF REPORT & PERIOD COVERED

DYNAMIC STRAIN WAVES - A DEVELOPMENT PERSPECTIVE Final

6. PERFORMING ORG. REPORT NUMBER

7. AUTHOR(&) 8. CONTRACT OR GRANT NUMBER(&)

R. Hasenbein, A. Gabriele, D. Finlayson,

B. Artus, G. Cunningham, and R. Gast
9. PERFORMING ORGANIZATION NAME AND ADDRESS 10. PROGRAM ELEMENT. PROJECT. TASK
AREA & WORK UNIT NUMBERS
U.S. Army ARDEC AMCMS No. 6436.39.6430.012
Benet Laboratories, SMCAR-CCB-TL PRON No. 4A7HF7YFlA/F
Watervliet, NY 12189-4050
11 COkITROLIMG 0O:C= NAME AND ADDRESS 12. REPORT DATE

U.S. Army ARDEC October 1990

Close Combat Armaments Center 13. NUMBER OF PAGES
Picatinny Arsenal, NJ 07806-5000 16
14. MONITORING J rent from Controlling Office)
AGENCY NAME & ADDRESS(diffe 15. SECURITY CLASS. (of this report)

UNCLASSIFIED
IS. DECL ASSI FI CATION/DOWN GRADI NG
SCHEDULE

:6. DISTRIBUTION STATEMENT (of this Report)

Approved for public release; distribution unlimited.

17. DISTRIBUTION 3TATEMENT (of the abstract entered In Block 20, It different from Report)

IS. SUPPLEMENTARY NOTES

Presented at the Sixth U.S. Army Symposium on Gun Dynamics, Tamiment, PA,
15-17 May 1990.
Published in Proceedings of the Symposium.
19. KEY WORDS (Continue an reverse side Of noceoeav mnd Identlfy b, block number)

Projectile Passage Circumferential Strain

Dynamic Strain Time
Peak Static Strain Flexural Waves

20. AR rAC? (Comsa dorevre

e *If nesy0 6Rd Identify by block nubmber)
A discussion of dynamic strain waves in large caliber cannon tubes and an
extensive analytical treatment of this phenomenon was previously presented
at the Fifth U.S. Army Symposium on Gun Dynamics.

ni report considers the implications these dynamic strains had on the

development of a particular cannon tube, including measurement techniques
which evolved during tests at the proving ground, predictive design methods
(CONT'D ON REVERSE)
DO Fo0010147
I'mO O,,ov6 s L
I JAN 73 EDYTnOf OF I11V S IS O LETE UNCLASSIFIED
SECUmTvY CLASSIFICATION OF THIS PAGE (WIren Date Entered)
SECURITY CLASSIFICATION OF THIS PAGE(Whan Data Znterfd)

20. ABSTRACT (CONT'D)

which have since become standard analytical tools, and potential

problems which have been identified for future study.

UNCLASSIFIED
SECURITY CLASSIFICATION OF THIS PAGE(oWhIn Data Entered)
 TABLE OF CONTENTS
Page

TNTRODUCTION ............................................................... 1

BACKGROUND .................................................................. 1

PROVING GROUND TEST METHODOLOGY ............................................ 3

FINITE ELEMENT ANALYSIS .................................................... 7

DESIGN IMPLICATIONS OF DYNAMIC STRAINS ..................................... 8

CONCLUSIONS ................................................................ 9

REFERENCES ................................................................. 10

LIST OF ILLUSTRATIONS

1. 120-mm XM25 tube ....................................................... 11

2. Strain gage application to tube ........................................ 12

3. Initial strain versus time trace ....................................... 12

4. Correlation between strain traces (analytical

versus experimental) ................................................... 13

5. Dynamic strain trace with event markers ................................ 13

6. Typical graph of "peak dynamic strain versus

tube axial position". .................................................. 14

7. Typical graph of "dynamic strain amplification

versus tube axial position".. .......................................... 14

8. Typical "circumferential strain versus time"

curve from dynamic finite element analysis ............................. 15

9. Typical "peak dynamic strain versus tube axial

position" curve from dynamic finite element analysis ................... 15

Accession For

, • . T

i T , " !
INTRODUCTION

A discussion of dynamic strain waves in large caliber cannon tubes and an

extensive analytical treatment of this phenomenon was previously presented at

the Fifth U.S. Army Symposium on Gun Dynamics (ref 1).

This report considers the implications these dynamic strains had on the

development of a particular cannon tube, including measurement techniques which

evolved during tests at the proving ground, predictive design methods which have

since become standard analytical tools, and potential problems which have been

identified for future study.

BACKGROUND

The 120-mm M256 cannon is the main weapon of the U.S. MIAI Abrams Tank.

This cannon was originally designed and developed by Rheinmetall in the Federal

Republic of Germany, and it ranks among the most powerful tank weapons in the

world. However, since its adoption by the U.S. Army into the Abrams Tank

System, the armor on threat tanks has become increasingly more formidable;

therefore, considerable interest existed in the mid-1980s in "upgunning" the

M256 cannon simply by increasing the length of its tube (i.e., increasing its

length of projectile travel in order to achieve higher muzzle velocity). The

experimental tube that was envisioned to accomplish this was designated the

120-mm XM25 tube. System planners indicated a desire to make an absolute mini-

mum of other changes to the Abrams Tank to achieve this increased firepower

capability.

When the M256 cannon was integrated into the MiA1 Tank, it was a relatively

easy task to balance this gun about its trunnions. This was deemed desirable

since it simplified the weapon stabilization problem. However, when the tube

1
was extended to XM25 length using wall thicknesses similar to those in the M256

tube, designers found that a considerable imbalance resulted. For a new system

design, this problem could be addressed by several methods such as the use of

equilibrators, counterweights, or enhancements to the elevation/stabilization

system. However, since the system guidance indicated a desire to make minimum

changes to the existing MIA1 Abrams Tank, tube designers were left with the pri-

mary responsibility for minimizing the imbalance. It was quickly realized that,

in the design of this conventional all-steel tube, the only method of

accomplishing this was to reduce wall thicknesses towards the muzzle end to

values less than the previous design practice might have 'deemed judicious.

System planners, however, indicated that the higher risk of doing so would be

acceptable for this experimental tube and urged that this approach be taken.

As expected, when the wall thicknesses towards the muzzle end of the XM25

tube were decreased, calculated stresses and strains increased since there was

less material to contain the same amount of pressure. While it appeared that

these values would be acceptable from a single-shot strength viewpoint, concern

arose that the critical fatigue zone in the tube might shift from the chamber

area to the muzzle. As a result, fracture mechanics and fatigue experts in

Benet's Research Division were consulted, and they made appropriate recommen-

dations for laboratory testing in the forward tube sections. At the same time,

however, they warned that significantly reducing tube wall thicknesses near the

muzzle could result in unknown end effects and loading patterns which might

increase strains beyond those that might otherwise be predicted. Their con-

sidered advice that the muzzle end of the tube be studied intensively during

engineering tests began the process which led to the later identification of the

dynamic strain phenomenon.

2
PROVING GROUND TEST METHODOLOGY

The 120-mm XM25 tube (Figure 1) was subseqt-ntly designed with wall

thicknesses towards the muzzle end as low as 12.7 mm; for reference, the minimum

wall thickness of the M256 tube is 17.0 mm (one-third greater). After the tube

was fabricated by the Watervliet Arsenal (NY), it was shipped to Aberdeen

Proving Ground (APG) (MD) for, engineering tests under the lirection of the

Combat Systems Test Activity (CSTA). Test plans requested that multiple strain

gages (oriented both circumferentially and longitudinally) be placed at several

axial locations on the tube from the bore evacuator to the muzzle and that

available DM13 APFSDS-T cartridges be utilized when firing. During the initial

stages of the test, considerable difficulty was encountered simply keeping the

gages attached to the tube, particularly those located towards the muzzle.

After a period of experimentation by CSTA, however, a satisfactory application

procedure was determined (shown schematically in Figure 2). Subsequent results

obtained when firing the DM13 cartridge produced "strain versus time" traces (an

example is shown in Figure 3) that contained what appeared to be anomalies. For

example, severe peaks were present in the strain signals well beyond those which

would be predicted using equations of statics. Further, significant compressive

circumferential strains were observed as the projectile approached the strain

gage locations. And finally, longitudinal strain gage results generally

oscillated about "zero strain," but at amplitudes which approached those of the

circumferential strain gages.

Two significant observations were made at this early juncture of the test.

First, it was noted that when firing the high speed DM13 round, the lack of

agreement between observed peak strains and calculated strains decreased signif-

icantly in the rearward gage locations where projectile velocity is lower.

3
Second, when a lower velocity M831 HEAT-TP cartridge was fired (grateful

acknowledgement should be given to Mr. Clyde Musick, CSTA Test Director, for

this suggestion), there was much better agreement between predicted and exoeri-

mental results at all gage locations. These two observations led designers to

plot "strain amplification (defined as 'peak strains measured by the gages

divided by calculated static strain') versus projectile velocity" (at that gage

location) for the two different cartridges. The resulting curve seemed somewhat

well-behaved, the ratio being approximately 1.1 at lower velocities and

increasing monotonically to approximately 4 as projectile velocity increased.

This provided the first clue that the phenomenon might be somehow related to

projectile velocity.

Dynamic'sts in Benet's Research Division were asked to consider whether the

observed data were the result of an actual physical phenomenon or simply an

instrumentation problem. Their subsequent closed-form analytical efforts (which

were presented at the Fifth U.S. Army Symposium on Gun Dynamics (ref 1))

revealed that the phenomenon being observed was indeed real, and they provided

significant insights into its nature. Disturbingly, however, the proving ground

strain traces (Figure 3) did not bear a resemblance to those predicted by the

Benet researchers. After looking closely at the predicted analytical strain

waves and considering the strain measuring methodology at the proving grounds,

Benet researchers were able to propose two significant modifications to the lat-

ter:

0 First, it was suggested that the filters being used during recording o,-

the strain signals (10 kHz lowpass) be increased to a higher value (30 kHz

lowpass), since the anticipated frequencies of the strain wave should be on the

order of 15 kHz;

4
 Second, it was suggested that the time "window" for presentation of

individual strain traces be decreased from the previous 100 milliseconds to

around 5 to 10 milliseconds in order to better observe details of the wavelike

nature of the strains. The time "window" should be the same for all strain

gages on the entire tube for a given round:

" beginning slightly before the projectile arrives at the rearmost

strain gage, and

" ending slightly after the projectile exits the tube.

The above suggestions were incorporated into the CSTAiAPG firing tests, and

the benefits were immediately apparent. Good correlation was noted between -he

analytically-predicted dynamic strain traces and those obtained from firing

tests (Figure 4), including frequencies and amplitude. Moreover, specific

events such as projectile passage, projectile exit from the tube, and predicted

static strain cc!ld be superposed on the traces to assist with the data reduc-

tion (Figure 5).

Since that time, several additional "rules of thumb" have evolved for the

conduct of dynamic strain tests at proving grounds. Best results are obtained

when the test includes a wide variety of cartridge types, ranging from the

slowest of those which will be fired from the tube in actual service (or

training) to the fastest. In some cases it may be desirable to pre-condition

the cartridge to elevated temperatures to attain the highest possible muzzle

velocities. Generally, five (minimum) to ten (preferred) rounds of each type

should be fired. Strain gages should be applied at several axial locations on

the tube, taking care to select positions which include lower projectile veloci-

ties (e.g., slightly forward of mid-tube) as well as higher velocities towards

the muzzle. Five to six axial positions are currently selected, depending on

5
the number of available channels for recording data at the proving ground. At

each of these locations, four circuTferential strain gages should be attached as

showi in Figure 2, taking care to locate them at precisely the same axial posi-

tion. Longitudinal gages may also be used, but these are often of less general

interest and are more difficult to interpret.

In reducing dynamic strain data, the following procedure is now genlrally

used:

, Determine the peak strain value at the time associated with projectile

passage for each strain gage. In order to select the correct peak, projectile

in-bore location must be kncn or estimated. This is done wi.' the greatest

accuracy if projectile muzzle velocity is concurrently measured during the

dynamic strain test and later used to refine interior ballistic model predic-

tions.

* In reporting the peak dynamic strain at a given axial location for any

individual round fired, it is statistically best to use the average of the peak

values indicated by all four strair gages. This tends to cancel out the addi-

tional strains which might be added/subtracted by bore eccentricity and axial

tube flexure.

* Dynamic strain values are most simply portrayed by plotting "peak strain

versus tube axial location" for each cartridge type and pre-conditioning tem-

perature (see example in Figure 6). It is often instructive to also show the

calculated static strain on the same graph. Note in Figure 6 that the results

of each round are shown, producing a (real) array of possible results. Mean

values (for later comparison with analytical predictions) and standard

deviations (which increase significantly towards the muzzle) are also often

calculated for the entire group of five to ten rounds and displayed on a similar

graph.

6
 * Aljo of interest are plots of "dynamic strain amplification versus tube

axial position" for each cartridge type and pre-conditioning temperature.

Again, strain amplification is defined as the peak dynamic strain divided by the

calculated static strain at that location and pressure. Figure 7 shows an

example of this type of plot, and it is based on the mean dynamic strain value

at each of the axial locations.

* A significant additional output of a dynamic strain test is a plot of

"dynamic strain amplifications versus projectile veloc'ty." One benefit of this

type of curve is that the results from all cartridge types and pre-conditioning

temperatures may be combined into one figure. The disadvantage, of course, is

that projectile velocities are often classified, thereby limiting the oppor-

tunities for presentation of these types of results.

FINITE ELEMENT ANALYSIS

Apart from urderstanding the phenomenon of dynamic strains and being able

to test for them, it is important for tube designers tc have at their disposal

techniques for analysis of dynamic strains during the design phases prio- to

manufacture. The algorithms must be specifically geared towards the actual tube

geometry and the array of ammnjnition (either existing or envisioned) to be fired

through it. A satisfactory methodology for doing this has evolved which has two

separat. steps:

• First, the loading conditions on the tube must be determined,

specifically, Lne applied pressure and the projectile velocity. To accomplish

this step, appropriate data are supplied as inputs to an interior ballistics

computer code, and output files containing the following information are created

(both being functions of time):

7
 " pressure in the bore at the base of the projectile;

" axial location of the projectile.

* Second, the data files generated in the ballistic analysis above are

used as inputs into a dynamic finite element analysis (FEA). An axisymmetric

gridwork which duplicates the interior and exterior diameters of the cannon tube

is created for use with a non-linear finite element code capable of performing

dynamic analyses (such as ABAQUS). Although this type of FEA is potentially

large in scale, it can be performed in a timely manner using a supercomputer.

Results produced using this technique have compared quite favorably with strain

gage data from proving ground tests. A typical output graph of "circumferential

strain versus time" is shown in Figure 8 and one of "peak dynamic strain versus

tube axial position" is shown in Figure 9.

This analytical method can provide additional insights regarding a given

tube's dynamic response in areas where actual measurements are either difficult

or impossible (for example, at the extreme muzzle of the tube, at the bore, or

within the walls of the tube). Outputs can also be used to create video anima-

tions which further clarify the physical nature of the dynamic strain waves.

DESIGN IMPLICATIONS OF DYNAMIC STRAINS

There are various implications of dynamic strains on developmental cannon

tubes, some of which are reasonably well understood and some of which are

excellent candidates for ongoing research. A partial list would include the

following items:

" effect on tube strength (failure criteria);

* effect on tube fatigue life (high strain rate loading, multiple strain

cycles per round fired);

8
 * effect on projectile behavior due to local clearances or constrictions

at tube-projectile interfaces (e.g., initiation of balloting, sabot tip-

off, etc.);

* effect on adhesion/cohesion of bore coatings such as chromium plating;

* creation of local accelerations in the tube walls.

It is interesting that the last item on this list, which looks most innoc-

uous, may be of the most immediate concern due to the severe environment which

these accelerations create for attached components such as Muzzle Reference

System Collimators. Resulting local accelerations near the muzzle can be on the

order of ± 100,000 g's and may result in breakage of delicate optical com-

ponents.

CONCLUSIONS

Dynamic strains are a phenomenon which have only recently been identified

and understood. The effort to do so has been the result of an intensive collab-

oration between research-oriented people, who are fundamentally interested in

understanding and explaining physical phenomena, and development-oriented

people, who are required to deliver functional hardware in a timely manner.

From the development perspective, this problem has been (and continues to be) a

prime example of the mutually beneficial relationship that can exist between the

two, for truly without the researchers, the dynamic strain problem would never

have been observed and understood at all.

(SPECULATIVE POSTSCRIPT: It is likely that researchers likewise feel a perverse

reciprocal appreciation for developers who create exciting new problems like

these in the first place by attempting to expand the limits of hardware

performance.)

9
 REFERENCES

1.T. E. Simkins, "Resonance of Flexural Waves in Gun Tubes," in: Proceedings

of the Fifth U.S. Army Symposium on Gun Dynamics, ARCCB-SP-87023, Benet

Laboratories, Watervliet NY, 23-25 September 1987, pp. 65-78; also ARCCB-

TR-87008, July 1987.

10
40

2 .
 1. U-BOND 610
2. MICROMEASUREMENT STRAIN
TOP VIEWGAGE (EA-08-250BF-350)
TOP IEW3. RUBBER PAD
4. FOAM RUBBER SPONGE
5. GUN TAPE
S. SINGLE STRAND OF WIRE
7. TABSTRIP
6 8. U-COAT
9. WIRE FROM CABLE
10. CANNON TUBK

SIDE VIEW

Figure 2. Strain gage application to tube.

Z3

. a00 a 00 .0 '4 '0 "

Tit.SE

Fiue3 ntilsri esu ietae

z1
 -- measu~red

-predcted i

TI ME

Figure 4. Correlation between strain traces

(analytical versus experimental).

-PROJECTILE PASSAGE

- PEAK DYNAMIC STRAIN

W j:- PEAK STATIC STRAIN

TIME

Figure 5. Dynamic strain trace with event markers.

13
 30 5 ROUNDS ,
..-

,.I- :530

k+000
U4 4
M4 4
500 0
* AVG. OF 4 CIRCUMFERENTIAL STRAIN GAGES FOR EACH ROUND
0 PREDICTED STATIC LAME STRAIN

AXIAL LOCATION

Figure 6. Typical graph of "peak dynamic strain

versus tube axial position."

AVERAGE OF 5 ROUNDS

AXIAL LOCATION

Figure 7. Typical graph of "dynamic strain amplification

versus tube axial position."

14
 4000

I 3000

C 2000

1000
L.J

0-
Li
LI

1 4800 mm from rear face of tube

-1000

1.4 1.5 1.6 1.7 1.8 1.9 2 2.1 2.2

TIME (x1OE-03 5EC)

Figure 8. Typical "circumferential strain versus time"
curve from dynamic finite element analysis.

5000-

S4000- predicted dynamic strain

__ - - -L=4 static strain
Ul

- 3000

LJ 2000 ' "

,.J
2000-

1000
DISTANCE FROM REAR FACE OF TUBE (MM)
F4gure 9. Typical "peak dynamic strain versus tube axial position"
curve from dynamic finite element analysis.
 TECHNICAL REPORT INTERNAL DISTRIBUTION LIST

NO. OF
COPIES

CHIEF, DEVELOPMENT ENGINEERING-DIVISION

ATTN: SMCAR-CCB-O I
-DA 1
-DC I
-DI 1
-DP 1
-DR 1
-DS (SYSTEMS) 1

CHIEF, ENGINEERING SUPPORT DIVISION

ATTN: SMCAR-CCB-S 1
-SE 1

CHIEF, RESEARCH DIVISION

ATTN: SMCAR-CCB-R 2
-RA 1
-RE 1
-RM 1
-RP 1
-RT 1

TECHNICAL LIBRARY 5
ATTN: SMCAR-CCB-TL

TECHNICAL PUBLICATIONS & EDITING SECTION 3

ATTN: SMCAR-CCB-TL

DIRECTOR, OPERATIONS DIRECTORATE 1

ATTN: SMCWV-OD

DIRECTOR, PROCUREMENT DIRECTORATE 1

ATTN: SMCWV-PP

DIRECTOR, PRODUCT ASSURANCE DIRECTORATE 1

ATTN: SMCWV-QA

NOTE: PLEASE NOTIFY DIRECTOR, BENET LABORATORIES, ATTN: SMCAR-CCB-TL, OF

ANY ADDRESS CHANGES.
 TECHNICAL REPORT EXTERNAL DISTRIBUTION LIST

NO. OF NO. OF
COPIES COPIES

ASST SEC OF THE ARMY COMMANDER

RESEARCH AND DEVELOPMENT ROCK ISLAND ARSENAL
ATTN: DEPT FOR SCI AND TECH 1 ATTN: SMCRI-ENM
THE PENTAGON ROCK ISLAND, IL 61299-5000
WASHINGTON, D.C. 20310-0103
DIRECTOR
ADMINISTRATOR US ARMY INDUSTRIAL BASE ENGR ACTV
DEFENSE TECHNICAL INFO CENTER ATTN: AMXIB-P
ATTN: DTIC-FDAC 12 ROCK ISLAND, IL 61299-7260
CAMERON STATION
ALEXANDRIA, VA 22304-6145 COMMANDER
US ARMY TANK-AUTMV R&D COMMAND
COMMANDER ATTN: AMSTA-DOL (TECH LIB)
US ARMY ARDEC WARREN, MI 48397-5000
ATTN: SMCAR-AEE 1
SMCAR-AES, BLDG. 321 1 COMMANDER
SMCAR-AET-O, BLDG. 351N 1 US MILITARY ACADEMY
SMCAR-CC 1 ATTN: DEPARTMENT OF MECHANICS
SMCAR-CCP-A 1 WEST POINT, NY 10996-1792
SMCAR-FSA 1
SMCAR-FSM-E 1 US ARMY MISSILE COMMAND
SMCAR-FSS-O, BLDG. 94 1 REDSTONE SCIENTIFIC INFO CTR 2
SMCAR-IMI-I (STINFO) BLDG. 59 2 ATTN: DOCUMENTS SECT, BLDG. 4484
PICATINNY ARSENAL, NJ 07806-5000 REDSTONE ARSENAL, AL 35898-5241

DIRECTOR COMMANDER
US ARMY BALLISTIC RESEARCH LABORATORY US ARMY FGN SCIENCE AND TECH CTR
ATTN: SLCBR-DD-T, BLDG. 305 1 ATTN: DRXST-SD
ABERDEEN PROVING GROUND, MD 21005-5066 220 7TH STREET, N.E.
CHARLOTTESVILLE, VA 22901
DIRECTOR
US ARMY MATERIEL SYSTEMS ANALYSIS ACTV COMMANDER
ATTN: AMXSY-MP 1 US ARMY LABCOM
ABERDEEN PROVING GROUND, MD 21005-5071 MATERIALS TECHNOLOGY LAB
ATTN: SLCMT-IML (TECH LIB) 2
COMMANDER WATERTOWN, MA 02172-0001
HQ, AMCCOM
ATTN: AMSMC-IMP-L 1
ROCK ISLAND, IL 61299-6000

NOTE: PLEASE NOTIFY COMMANDER, ARMAMENT RESEARCH, DEVELOPMENT, AND ENGINEERING

CENTER, US ARMY AMCCOM, ATTN: BENET LABORATORIES, SMCAR-CCB-TL,
WATERVLIET, NY 12189-4050, OF ANY ADDRESS CHANGES.
 TECHNICAL REPORT EXTERNAL DISTRIBUTION LIST (CONT'D)

NO. OF NO. OF
COPIES COPIES

COMMANDER COMMANDER
US ARMY LABCOM, ISA AIR FORCE ARMAMENT LABORATORY
ATTN: SLCIS-IM-TL 1 ATTN: AFATL/MN
2800 POWDER MILL ROAD EGLIN AFB, FL 32542-5434
ADELPHI, MD 20783-1145
COMMANDER
COMMANDER AIR FORCE ARMAMENT LABORATORY
US ARMY RESEARCH OFFICE ATTN: AFATL/MNF
ATTN: CHIEF, IPO I EGLIN AFB, FL 32542-5434
P.O. BOX 12211
RESEARCH TRIANGLE PARK, NC 27709-2211 METALS AND CERAMICS INFO CTR
BATTELLE COLUMBUS DIVISION
DIRECTOR 505 KING AVENUE
US NAVAL RESEARCH LAB COLUMBUS, OH 43201-2693
ATTN: MATERIALS SCI & TECH DIVISION 1
CODE 26-27 (DOC LIB) 1
WASHINGTON, D.C. 20375

DIRECTOR
US ARMY BALLISTIC RESEARCH LABORATORY
ATTN: SLCBR-IB-M (DR. BRUCE BURNS) 1
ABERDEEN PROVING GROUND, MO 21005-5066

NOTE: PLEASE NOTIFY COMMANDER, ARMAMENT RESEARCH, DEVELOPMENT, AND ENGINEERING

CENTER, US ARMY AMCCOM, ATTN: BENET LABORATORIES, SMCAR-CCB-TL,
WATERVLIET, NY 12189-4050, OF ANY ADDRESS CHANGES.