k• - apps.dtic.mil · numerical results aru available. the sequence of evento in projectile...

$: K• - apps.dtic.mil · numerical results aru available. The sequence of evento in projectile impact shock initiation is most clearly illustrated in the series of mass fraction contour$
<CVRtAD-

TECHNICAL REPORT ARBRL-TR-02576

NUMERICAL MODELING OF PROJECTILE IMPACTSHOCK INITIATION OF BARE AND COVERED

COMPOS ITION-B

John Starkenbery . ..... -

Yun Huang .Alvin Arbuckle 'K• AUS 1 1984

Auri't 1984 A

US ARMY APMAMENT ).P31 AND DEVEOPM[NT CENTER"BALLISTIC REr '•CH LABORATORY

ABERDEEN 7' ViNG GROUND, MARYLAND

* Approved for public release;, distribution unlimited.

C-,

. 4 08 .30 006, : : . . . . .. :. : : . . . . -,. . . , ., . . , , . , . . , , , . . . . . . . . . , : . : .

Destroy this report when it is no longer needed.Do not return it to the originator.

Additional copies of this report may be obtainedfrom the National Technical Information Service,U. S. Department of Commerce, Springfield, Virginia22161.

0

The findings in this report are not to be construed as an officialDepartment of the Army position, unless so designated by otherauthorized documents.

The use of trade names-or manufacturers! names in this re~portdoes not constitute indorsement of any commercial product.

.. . . . . . . . . ..-

E~

SECURITY C...AISIFICATION OF THIS PAGE (3?in Date Entoted)

REPOT DCUMNTATON AGEREAD INSTRUCTIONS______ REPORT_____________________PAGE_ BEFOR1PCOMPLETINGFORM

I. REPORT NUMBER 2GVT ACCESSION NO. 3. RECIPIENT.rS CATALOC NUMBER

Technical Report ARBRL-TR-O02576 I'J)-9Vio _ _________

4. TITLE (and Su~dtgtI) S. TYPE OFREPORT & PERIOD COVERED

NUMERICAL MODELING OF PROJECTILE IMPACT SHOCKINITIATION OF BARE AND COVERED COMPOSITION-B, Final

6. PERFORMING ORG. REPORT NUMBER -

7. AUTNOft(s) 8. CONTRACT OR GRANT NUMBER(r)

J. StarkenbergY. Huang,A. Arniuckle _____________

0. PERFORMING ORGANIZATION NAME AND AODRESS 10. PROGRAM ELEMENT. PROJECT, TASK

Us Army Ballistic Research Laboratory AREA AWORK UNIT NUMBERS

ATTN: DRXBR-TBDAberdeen Proving Ground, MD 21005- 5066 1L162618Af180

11. CONTROLLING OFFICE NAME AND ADDRESS 12. REPORT DATE

US Army Ballistic Research Laboratory August 1984AMT: DRXBR-OD-ST -F3. NUMBER Oli PAGESAberdeen Proving r-round, NT) 21005-5066 _____________

1.MONITORING AGENC NAME A OODRESS(if different feam Cm.UtlItirg Office) 1S. SECURITY CLASS. (of thisreport)

UNCLASSIFIED15s.. DECLASSIFICATION/DOWNGRADING

SCHEDULE

IS. DISTRIBUTION STATEMEN4T (of UrnS. Paport)

Approved for public release, distribution unlimited.

I?. DISTRIBUTION STATEMENT (of I%* albstact entered In Sleek 2.it different how irepeft)

IS. SUPVPLEMENTARY MOT9S

It 9 QQ (Continue orn reveree old& It necessary and identify by blec& number)

fragment impactmunitions vulnerabilityprojectile impactshock-to-detonation transition

26- A0$rRACT (C40UNW ass' ro 00WUS Wa.1* ay =W i IdeaitymI by' block numba)ý WdIC)This report concerns our numerical modeling of the projectile impact

shock Initiation of composition-B (comp-B). We have Considered both bare'andcovered charges Impacted by cylindrical steel projectiles Using the Los Alamos2DE code. We have examined the flcw fields In s0oe detail and compared pre-dicted critical velocities with published experimental values. For barecharges, we observed two different mechanisms by which the critical velocity Is

DO ~ 103 IO9WW@)MOVB'SOS03T~UNCLASSIFIED

iacumny, cLASSIICATIlW Oir ThIX PAQ9 (IlloData Entered)

, j.*; * *.s . . . . 0 0 .6i ,9 ?.#.. . ..

"-i.-UIN|ITY CLASSOIriCATiON OF THIS PAOGR(Wht, D00 Zntme")

"20. ABSTRACT (continued)

determined. For imrects by projectiles of sufficiently large diameter initia-tion occurs as the impact induced shock wave builds to detonation by reinforce-ment due to burning behind the shock. For smaller diameter, high velocityprojectiles, we saw that detonation or near detonation breaks out immediatelyon impact, but may be quenched by the ensuing rarefactions. We found that 2DE

predicted the critical velocity accurately, We also checked fpidt values

along the initiation threshold and found them to be relatively constant. Wecompared the shock to detonation transition paths to the Pop-plot for comp-Band found them to agree in the ease of a planar shock buildup but not in thecase of projectile impact, for which multiple paths to detonation were ob-served. We also simulated the special projectile geometries con.-,;ered byMoulard and found that 2DE provided a qualitative explanation of his obser-vations.

In the case of covered projectiles we found flow fields similar to thebare charge case. The thickest ,cover plates allowed the rarefaction to over-take the shock before they entered the explosive and significantly raised thecritical velocity. The predicted Initiation thresholds agree with Howe'sresults but not with Slade and Dewey's.

-S.

* ~UNCLASS IFIlED

SECU11,1TY CLASMINICATIONrOiP THIS PAGIEU(t.. bef. EnfmaG

4-°

TABLE OF CONTENTS

Page

LIST OF ILLUSTRATIONS ......... ................ 5

I. INTRODUCTION. . . . . . . . . ........... 7

II. BRIEF DESCRIPTION OF 2DE ..................... 8

III. PROJECTILE IMPACT SHOCK INITIATION OF BARE COMPOSITION-B. . . . . 8

Geometry and Computational Considerations . . . . . . . . . . . . 8

Results 10

IV. PROJECTILE IMPACT SHOCK INITIATION OF COVERED COMPOSITION-B . . 25

Geometry and Computational Considerations . . . . . ... . . . . . 25

ResUltA . . . . . . .. . . . . . . 25

V. -SUMMARY ' 36

REFERENCES........ . .................... 39

DISTRIBUTION LIST .......... . . . . . . . . . . . . . 41

3

LIST OF ILLUSTRATIONS

Figure Page

1 Axisymmetric Geometry Used in Projectile ImpactComputations . . . . . . . . . . . . "*9

2 Sequence of Mass Fraction Contour Plots for theSuperaritical Impact of a 10 mm Diameter Steel.Projectile at 1.1 ka/s against Bare Composition-B . . . 12-13

3 Critical Impact Velocity as a Function of ProjectileDiameter - Comparison of 2DE Predictions withPublished Experimental Data for Bare Composition-B... 15

4 Sequence of Mass Fraction Contour Plots for the'Supercritical Impact of a 4 mn Diameter SteelProjectile at 1.7 km/s against Bare Composi.tion-B 16-17

5 Sequence of Mass Fraction Contour Plots for the Sub-critical Impact of a 4 mm Diameter Steel Projectile"at 1.6 km/s against Bare Composition-B ......... 18-19 t

6 Sequence of Pressure and Mass Fraction Profiles forthe Subcritical Impact of a 4 mm Diameter SteelProjectile at 1.6 km/s agaitst Bare Composition-B . . 20-21

7 Comparison of Computed Paths to Detonation withPop-plot - Flyer Plate Impact against BareComposition-B . . . . . . . 22

8 Comparison of Computed Paths to Detonation withPop-plot - Projectile Impact against BareComposition-B .............. . . .. .w • 23 1_.

9 Sequence of Mass Fraction Contour Plots for theSuperoritical Impact of an Annular Cross-SectionSteie Projectile at 0.9 km/s against BareComposition-B . . . . . . 26-27

I

10 Sequence of Mass Fraction Contour Plots for the-Superoritical Impact of a "Rectangular" Cross-Section Steel Projectile at 1.25 Msi. . . . . . . . . 28-29 .-

11 Sequence of Mass Fraction Contour Plots for theSubcritioal Impact of a "Rectangular' Cross-Section Steel Projectile at 1.1 km/s against BareComposition-3 .. . . . .• 30-31

12 Sequence of Mass Fraotion Contour Plots for the -.Supercritical Impact of an 8 - Diameter SteelProjectile at 1.75 km/s against a Composition-BTarget Protected by a 4u m Thick 3teel Cover Plate. . . 32-33

S °

•, .• •2 '2".

I

13 Sequence of Mass Fraction Contour Plots for theSubcritical Impact of a 6 mm Dmeter SteelProjectile at 1.4 km/s against a Comrcsition-B

* Target Protected by -a 1.5 mm Thick Steel Cover Plate. 34-35. dl/2

"14 Correlation of Ved with h/d - Comparison of 2DEPredictions with Experimental Data. 37

'I

.U

°J*

ml

"Lo

'I

I

I-

.°6

I. INTRODUCTION

Projectile impact shock initiation of high explosives has long been a

subject of interest in the energetic materials community. Considerable1-4I

experimental data has been generated over the years. Numerical modeling ofprojectile impact shock initiation for comparison with experiments has been

reported in at least one case. However, only the predicted critical velocitiesand no detailed analysis of the flow fields revealed by the computations were -: -.

5presented. In another numerical study, Ilader and Pimbley modeled the pinitiation of explosives due to the impact of shaped charge jets using the same

computer code used in the present report.

This report concerns, our numerical modeltng of the projectile impact shock

initiation of COmposition-B (comp-B). le have considered both bare and covered

charge3 impacted by cylindrical steel projectiles using the Los Akamos 2DE I

code. We have examined the flow fields in some detail and compared predicted

critical velocities with published experimental values. Our earlier work on

this subject was found to have made use of insufficient artificial viscesity . 6

This has been corrected in the present report and the work extended to a numberof different areas.

ID. c. Slade and J. Dew~ey, "High-o~rder Itnitiation of 2W~ M(.litaz'y Ek~plosivesby Projectile Impact," Balliatic Research Laboratory Report No. 12022, JU~y1957'. AD 145868

2s. M.,Brown and S'. G. Whitbread, "The Initiation of Detonation by shockWvsof Known Dur'ation and Intensity," Lee ondes de Detonation., C.AT..R.S.

No. 109, pp. 69-80, Paris, 1962.3L. A. Roa Lund J. W. Watt, and N. L. Coleburn, "INitiation Of WarheadEx'ploeives by the Impact of Controlled Fragments, r Noirma Impact, 1 NavaZOrdnance Laboratory Technical Report NOLTR-?3-124,.August 1974.

4`X. I.. Bahl, *H, C. Vdntine, and .R. L. Weingarts, "fThe Shock rnitiation of, Bareand Covered Ezptos~ives by Projectile Imnpact," Seventh Sjrpoaim~ (International)onr Detonation, erw, 1981, pp. 325-335.

5C. L. Mader and G. H. Pimb",Y, VJet Initiation and Penetration of EkpZo-ives,-j.ournaZ of Ime tgetic Materials, Vol. 2, No. 1,' 1983, pp 1-44.

OT. K. Huang, d. .arkenh2rg, ahd A. L. Arbucke, "A itV erio Study.of.So.kInitiation of composition-B by High Speed Impact of , ," . Steem Projectiles,"

BRL Report to be pubZlihed.

7

II. BRIEF DESCRIPTION OF 2DE

The 2DE code 7 is a two-dimensional, reactive hydrodynamic computer codewhich makes use of the equations of motion in Eulerian form. It incorporatesthe HOM equation of state and, most important for our application, the ForestFire model for shock initiation of high explosives. In our computations we usedan elastic-plastic constitutive model to account for the behavior of steel.

III. PROJECTILE IMPACT SHOCK INITIATION OF BARE COMPOSITION-B'

Geometry and Computational Considerations

The aktsymetric geometry used in the bare charge projectitl impactcomputatooat is Illustrated in Figure 1. We have considered cylindrical steelWhitnead

,2

projectiles of unit aspect ratio (t/d) since Brown and havedemonstrated that different aspect ratios do not produce different criticalvelocites for shook initiation, except in the case of thin discs (L/d(1/4).Computations were made for projectile diameters, d, of 4, 5, 8, 10, 12, 15 and18 m-. Sufficient target material is provided when the length and diameter ofthe explosive charge are each three times the corresponding projectile

* dimensions.

We set up these Impact problems for 2DE calculation with axisymmetric gridsas summarized in Table 1. Here ar is the radial, cell dimension, I the numberof cells along the radial axis, at the axial cell dimension, J the number of.ells along the axis of symetry, At the time step for each computational cycle,

and N the total number of cycles to be completed.

TABLE 1

Input Data for 2DE Computational Grids - Bare Charges

d Ar,Az I - J At N(m) (m) (&s)

4 0.200 40' 90 0,005 4005 0.250 __0 70 0.006 4008 0.20Y 60 105 0.006 50010 0.340 60 105 0.010 " 50012 0.400 60 105 0.008 50015 0.500 60 105 0.010 50U18 0.600' 60 105 0.010 400

7 J.D. &erahner and c. L. fAbder, "2DE, -A 21oo-DimeneionZ Contivtoue SuZarianHydrodyjnpiw . Code for Computing W4tiomponent Reactwv( HYidrodymnicn Prob-Zero," Loa AZizos S*ientific Laboratory Report L.-4846, March 2972.

8

rd/2- 3 d/2

PRO6JECTILE C014POSITION-B TARGET

Figurt 1. Axisyimuotric Geomuetry Used in Projectile Impact Computations.

Results

Flow Field Observations. A number of graphical representations of ournumerical results aru available. The sequence of evento in projectile impactshock initiation is most clearly illustrated in the series of mass fractioncontour plots of Figure 2. The mass fraction varies from one to .ro aschemical reaction in the explosive runs to completion. These plot3 show resultsfor the impact of a 10 mm diameter projectile at 1.1 km/s, just above thecritical velocity. "te position of the impact shock is also shown. At 1.0 jusafter impact, burning is observed throughout the region between the shock andthe projectile, but detonation has not yet begun. Detonation, which may berecognized by the close spacing of the contour lines, is first observed to breakout after the shock has propagated some distance from' the impact point. Thedetonation then spreads outward along the shock and is well established by1.5 As. it also propagates back toward the projectile through the partiallyreacted material.

Determination of Critical Impact Velocity and Comparison with Experimental Data.By varying impact velocity in computations of this type we were able todetermine the critical velocity as a function of projectile diameter. In Figure3we have plotted the 2DE results together with the experimental result& of

Dewey and Sladez as well as the Jacobs-Roslund empirical formula 3 . Theagreement is excellent, with the 2DE go and no-go predictions all bracketing theempirical-curve. We have also Included the data for ISL comp-B (65/35) reported

by Moulard at the last detonation symposium. 8 This-explosive is reported to begenerally less sensitive than U.S. comp-B and particularly less sensitive whenimpscted by small projectiles!

We observed a different mode of critical shock initiation at the smallerdiameters. When the projectile diameter is small and the impact velocity ishighdetonation appears almost immediately on impact as shown in Figure 4 at0.3 #s after the impact of a 4 aI diameter pro.ectile at 1.7 km/s. In this casedetonation continues to propagate and a considerable amount of explosive hasbeen consumed by 1.2 *s. On the other hand, when the impact velocity 1s

reduced to 1.6 km/s, the detonation that breaks out immediately does notcontinue to propagate but is quenched by the action of following rarefactions asshown in Figure5. Thus, the mass fraction contour lines begin to spread out by

.0.7 9s. Little or no progress is made between 1.0 js and 1.5 #s as thedetonation dies out leaving a bubble of detonation products in its wake, Itshould be noted that these detonations are not overdriven as were those computed

5by Nader and Pimbley for shaped charge jet Impact. An overdriven detonation incomp-B produced by the impact of a steel projectile would require an impactvelocity exceeding approximately 2.8 km/s. Actually, full CJ pressure is neverachieved in these detonations. Figure 6 shows a series of pressure and mass

85* MouZard, "CriticaZ Conditione for Shock Initiation of Deto-ation by ZProjectiZe Impact," Seventh Sympoeiwn (?nter'nationaZ) on Detonation, Jrune'2981, pp. 316-324.

10

fraction profiles along the axis of symametry at various times for the 4 meprojectile impact at 1.6 ku/s. While the mass .'aection drops rapidly to zerothe pressure never rises above about 23 GPa.

In order to issess the relationship between the 2DE predictions of critical

velocity and the critical p 2t concept, we made a series of compu.attonh in whichthe Forest Fire model was deactivated and the explosive treated as an inertmaterial. By observing the pressure history of the target explosive adjacent to

the Impact point we were able to caltulatefp dt. Computations corresponding to

our highest suberitica1 impact velocities for projeeftile dimseters rmnging from5 = to 18 - were made. The results, sumarized In Table 2, show that, while

peak shook pressure decreases with decreasing impact velocity, fpdt ruains

fairly constant along the initiation threshold. Thus, the critical p t concept

appears consistent with Forest Fire.

Table 2, Response of "Inert* Composition-B to Critical Projectile Impact

Projectile Impact Peak Shook '. 2Diameter Velocity Pressure p dt

(m) (km/s) (GPa) COPs 2- )

5 L.4 9.6 328 1.1 7.0 32

10 1.0 6.0 3112 0.9 5.3 2815 0.8. 4.5 2818 0.7 3.8 25

Shock to Detonation Transition Paths. The Forest Fl"e Model is based in part onthe Single curve buildup hypothesis. Thus, the Pop-plot is interpreted asdescribing the process of bulidup to detonation Inathe shock pressure - distanceto detonation plane. This is true at least for the planar geometries In whichsingle curve buildup has been observed. In an earlier numerical study using2DE, we had occasion to consider the planar problam arlsing whea a flyer plate

of sufficient lateral extent strikes a coup-B target.9 Ones the distance alongthe axis of symetry' at which steady state detonation first appears haa beendetermined, the prolgress of shock buildup toward detonation as a function ofdlistance- of run to detonation can be compared to the Pop-plot. This has beendone ,for the problems of 10 am~thick flyer plates striking Ooup-B targets at 0.6and 0.7 ku/s in Figure. 7. The results indicate that 2D0 reproduces the singlecurve buildup. phenomenon in the planar geometry. The projectile impact data,however, does not appear to produce a single curve buildup along the axis ofsymmetry as illustrated in Figure 8.

9J. St alenbezrg, Y. .X. Hzang, and A. L. AbuokZle, 'A !t,.DiweneionaZ Numericalstudy o~t Detonation Pr'opagation Between Mmiiti one by Yewu of Slzock initiation, "BRL Report ARBRL.TR.O2522* Se ptnber 1983. AD ALWM

I __________._.

18.16.-

14.12.-10.-

6.

4.2'.0.-

0. 5. 10. 15. 20. 25. 30. 35. 40. 45.MF •AT I INTERVAL- 1.OOUOE-01

TIME- 1.00 CYCL!- 120

14.

4.12.0.-

0. S., 10. 15. 20. 25. 30. 35. 40. 45.HF HAT I INTERVAL- -.OOOOE-01

TIME- 1.50 CYCLE- 180

Figure 2. Sequence of Mass Fraction ConLour Plots for the SupercriticalImpact of a 10 mm Diameter Steel Projectile at 1.1 kn/s againstBare Composition-B

12

"i " " . .�*f%" " *" " " " *= % '% ,, = '~ : " .- a - 4 ". '*. " .e- 9 * - " .4 .-" ' Q .~ ". *

. ~18.-

14. -12.10.

6.

2.

0. 5. 10. IS. 20. 25. 30. 35. 40. 45.MF MAT 1 INTERVAL= l.OOOOE-01

TIME1 2.00 CYCLE- 240

s8.16.14.

12.10.'

•. S.'• 6. -

-0. S. 10. IS. 20. 25. 30. 35. 40. 45.IF MAT I INTERVAL- 1.OOOOE-01

TIME" 2.50 CYCLE- 300

Figure 2. (continued)

13

-'-'~ l .. : %.,- .~

"*\%% .%*|...*. :~ ~ **~**

This page intentionally left blank.

14

2.0 2D. GO*2{NOGO, 0

JACOBS ROSLUN D FORMULA-SLADE AND DEiEY (3

2 1.5MOULARD (ISL 65/35 COMP-B) 0

01.0- 3.26 mm3 /2/t&s

91. V*2 b'

> d,

hiI-

~0.5

I/vTJ VA Vv 72 AA0 1/4O 1/./r 11-4

I..: D, NO GO ______

0.2 0.3 0.4 0.5ii/0 (m1m12)

Figure 3. Critical' Impact Velocity as a Punction of Projectile Diameter -Coparison

2DE Predictions with Published-Experimental Data for Bare Composition-B..

i5

.................... *...'..****....

................ .. . .,.,,! .. . .A* .. * *Iu* * q

9.

8.

7.

6.

4.

3.

2.

0. C7

0. 2. 4. 6. 8. !0. 12. 14. 16. 18. 20.MF MAT 1 INTERVAL- 1.OOOOE-01

TIME- .3 CYCLES 60

9.8.

7.

6.5.

4.

3.

2.

1.

0. 2. 4. 6, '8. 10. 12. 14. 16. 18. 20.MF MAT I INTERVAL- 1.0OOOE-01

TIMES .6 CYCLE- 120

Figure 4. Sequence of Mass Fraction Contour Plots for the SupercriticalImpact of a 4 mm Diameter Steel Projectile at 1.7 m,/s againstBare Composition-B

16

• •.•.• •' .. * o. , '. , ''•.. '•.~ • *. * . ", ... • *# ti . %.. .'' .'. d 4 .'; "o• ".. 4, • " */ ' " %* **

* 9.

8.

7.

5.

4.

3.

2.

0.0. 2. 4. 6. 8. 10. 12. 14. 16. 18. 20.

MF MAT 1 INTERVAL- 1.0000E-01TIME- 1.1 CYCLE- 210

9.•'. 8.,-

7.

6.5.

4.

3.

2.

"0- " I0. 2i. 4. 6. 8. 10. 12. 14. 16. 18. 20.,

"MF MAT I INTERVAL- I.OOOOE-01"% TIME 1.2 CYCLE- 240


17kd

2 8.

7.

6.

5.

4.

3.

2.f0

"0. 2. 4. 6. 8. 10. 12. 14. 16. 18. 20."MF MAT ,I INTERVAL- 1.OOOOE-01

TIME= .5 CYCLE- 100

7.

"6.

5..I 4.-

3.2.

0.-

0. 2., 4. 6. 8, 10. 12. 14. 16. 18. 20.MF MAT. 1 INTERVALS 1.OOOOE-0I-

TIME- .7 CY LE- 150

'Figure 5. Sequence of Mass Fraction Contour Plots for the SubcriticalImpact of a 4 mm Diameter Steel Projectile at 1.6 km/s againstBare Composition-B

U -- 18

0' S

i

i8.

7.6.

4.

1 1.-L

0.

.O0 2. 4. 6. 8. 10. 12. 14. 16. 18. 20.MF MAT 1 INTERVAL- I•OOOOE-O0


L

I

S27.•

14.

1 19

0.. 2. 4. 6. 8. 10. 12. 14. 16. 18. 20.SMF MAT 1 INTERVAL- 1.OOOOE-01

S~TIME- 1.5 CYCLE- 300

'• Figure 5. (continued)

o.: 19

L

25 - - -I ------- 4AO

t19

o..20 -ý9 .18.17 .

TIM 25 .3 "ICROS-W.D -

21-5 9 .16 - <_

~~~~~~~1 -5 -M •.E- .'omesc.•,-

10 HASS FRACTION

LlI __ PRESSUPE (CPA) P cj),

0 2 4 6 8 10 12 14 16 18

DISTANCE (MM)

CIo- 20 "II • ' --

W 15 I I TME- .40 MICROSECOND .6 <

c/) I I _.oW I 1---MASS FRACTION(1) _ elsu.c, "3 •.o~ ~ ....... PRESSRE (GPAI

.0 V II I - - 1 1 1 I i -0 2 4 6 8 10 12 14 16 18

DISTANCE (MM)

Figure 6. Sequence of Pressure and Mass Fraction Profiles for theSubcritical Impact of a 4 mm Diameter Steel Projectileat 1'.6 im/s against Bare Composition-B

20

I..

25 - -

- - -- - --- C-0-. 2C- 20.63 0

-. •.18 -

X TIME- .*OMICROSECOND .16.I15 U_

-./,10 - I '1-) . .. 14SKASS FRACTION'"5 PRESSURE (PA)

T I .--"-0 2 4 6 8 i0 12 .14 16 18

"DISTANCE (MM)

25 - -- - - - -<I 19_a 20-:: •i i.17 - •iX .,6 < ,.,,o In I 7,• e ,co•oo i

TIME . 0 * MICROSECOND I Ct.

L~5 1*. .IMASSFRACTIONc 10 - 1I 'l1'U.""* I I I I I~ I "I4'- I I

"0 2 4 -6. 8 10' 12 14 16 18

DISTANCE (MM)

- Figure 6. (continued)

21

C~4J

C4)

C*4.

000 ii.

%0 N U.

0 (A

OOC LUixO 0

CLl%0 0

o4.1

V 10

=4).

o<I144Cu4 0 0

04)

u 5

N o004.U

(ww) OUVN.00 OL LU

22'

C~4

.0

"00

0

_ X,

0i ixp~

LI) 0VLL4J

-h4J

04.

4.40

LI 4

I0 04

w$ 4j,

040 ce. I.M

*ww Noiq.N13 00N

23

Special Geometries. At the Seventh Symposium on Detonation, Moulard presentedsome Interesting experimental results in which ISL coup-B was impacted by

projectiles of circular, annular and rectangular cross-section. Although theprojectiles were designed to yield the same overall shock duration, he observedconsiderable differences in the critical velocities produced by each.Specifically he found that the cylindrical projectile required the highestcritical velocity (about 2 km/s), the aumular projectile required the lowestcritical velccity (less than 1 km/s), and the rectangular ceross-section producedan intermediate critical velocity. He sought to explain this by introducing acritical crea concept. We were interested in determining whether the classie1shock initiation concepts incorporated in 2DE could explain these observationswlthout recourse to additional concepts. The circular and annularcross-sections could be represented exactly. Indeed, the circular cross-sectioncomputation had already been completed. The impact of a rectangularcrosz-section projectile is, strictly, a three dimensional problem, but werepresented it by the Impact of a slab of infinite breadth having the thicknessof the snallest dimension (5 mm) of the projectile used in the experiments. Theexperimentally and numerically determined critical velocities are listed inTable 3. The results show that the classical concepts on which the ForestFire Model is built are sufficient for a qualitative explanation of the Moulardobservations. The principal reason for the absence of quantitative agreesent is'the different formulation of ISL comp-B rnd its reported lower senstivity tosmall diameter projectile Impact.

Table 3. Comparison of Moulard Experiments with 2DE Simulation.

Critical Velocity (km/a)Projectile

Cross-section Moulard Experiments 2DE Predictions

(65/35 comp-B) (60/40 comp-B)

MMSn , 1.95- 2.02 1.40 -1.50

S M i d.S1S D o.d. 1.06 -1.11 0.80 -0.90

S MxM 11 M 1.33 1.42 i.10- .15

Some of the reasons for thevariations in critical velocity 4h projectilecross-section become apparent when we observe the flow fields pro aed. Figure9 shows a series Of mass fraction contour plots for the impact .of the anniularprojectile. The relatively low impact velocity produces no imbdlate reactionadjacmnt to the impact point. However, at a later time, shock reflection(probably Mach stem formation) at the axis of symmetry produces higher pressuresthan the circular cross-section projectile impact at the same velocity. Thus,the detonation is observed to break out along the axis of symmetry at lowerimpact velocities. Of perhaps greater interest, then, 1s the difference between

?24

the circular and rectangular cross-section results. The only major diiferencehere is the rate at which the rarefaction quenches the incipient detonation..Remember that in the small diameter case we observed immediate detonation whichwas then auenched by the action of the followin& rarefaction. The results forthe rectangular cross-section are shown in Figures 10 and 11. Figure 10 showsmass fraction contours for a supercritical impact at 1.25 km/e. Detonationarises as a result of shock to detonation transition. In the subcritical impactat 1.1 km/s in Figure 11, no detonation occurs. It remaJns to be determinedwhether this strong effect on critical velocity is manifested with projectilesof larger dimensions for which simple shock to detonation transition occurs forthe cylindrical projectiles also.

'IV. PROJECTILE IMPACT SIOCK INITIATION OF COVERED COMPOSITION-n

Geometry and C'mputational Considexations

We have also addressed the related problem of projectile impact of coveredcomp-B by introducing a steel plate of thickness, h, htaween the projectile andthe explosive. Projectile diameters of 6, 8, and 10 mm and cover plates of 1/8,1/4, 1/2, and 3/4 as thick as thediameter In each case were cor3idered. Gthergeometrical considerations are as described for the bare charge problem. Theaxisyumetrio computational grids are described in Table 4.

Table 4. Data for 2DE Computational Grids - Covered Charges

d h/d ar,Az I J at N(mm) (mm) (As)

6 1/8 0.15 45 100 0.00o4 41006 1/4 0.15 45 100 0.004 6006 1/2 0.15 45 100 0.004 6506 3/4 0.15 45 100 0.004 7008 1/8 0.20 45 100 0.005 4508 1/4 0,20 45 100 0.005 GOO8 1/2, 0.20 45 110 0.005 80n8 3/4 0.20 45 120 0.005 800

101 1/6 0.25 45 100 0.005. 50010 1/4 0.34 55 125 0.008 70010 1/2 0.20 55 140 0.005 75010 3/4 0.20 55 1.40 0,005 800

Results

Flow Field Observations. Typical results are shown in the mass fractLon zontourplot sequence of Figure 12. This is for the 1.75 km/s impact of an 8 r.qdiameter projectile against a comp-B target preteoted by a 4 mm thick coverplate. In general, the flow fields were quite sxiilar to those observed in thebare charge Cases, Only the case of the 6 m- diameter projemtile with the 1.5mm thick cover plate showed the small diameter effect observed in the b.irecharge problems. In this case we observed a quick shook to detonationtransition followed by quenching. This is shown in Figure 13. However, we didnot consider quite as small projectiles in the covered charge problem. With the

25

25._ _ _ _ _ __

20.

15.

0. 5. 10. 15. 20. 25. 30. 35. 40. 45. 50.*MF MAT 1 INTERVAL- I-OOOOE-01


30.

25.

* 20'.

15.

10.

5.

0. 5. -10. I5.' 20. 25. 30. 3S.- 40. 45. 50.MF MAT 1 INTERVAL- I.0000E--0

TLrlEw 2.5 CYCLE* 200

-Figure 9. Sequence of Mass Frac 'tion Contour Plots for the SupercriticalImupact of an-Annular Cross-Section Steel Projectile at 0.9 km/sagainst Bare Co'uposition-B

26,

'--" 30. -

25.

20.

10.

.0.

.- 0. -

0. S. i0. I5. 20. 25.. 30. 35. 40. 45.' SO.MF MAT I INTERVAL= I1.OOOE-OI* TIME= J.0 CvCLE=,:240

30. .

* ~25.1

20.'

15.-

0.-r I0. 5. 10. ts. 20. 25. 30. 35. 40. 45. SO."MF MAT I INTERVAL' i.O000E-O;

T "IME- 4.0 CYCLE= 320


27

12.

10.-

8.

6.

4.

2.

0.-0. 5. 10. 15. 20. 25. 30.

MF MAT 1 INTERVAL= 1.OOOOE-01


12.

10.

8.

6.

.- 4.

2.i"0 o. J,-

1. S. 10. 15. 20. 25. 30.,.MF MAT I INTERVAL- .OOOOE-01

TIME- 1.20 CYCLE& 240

"Figure 10. Sequence of Mass Fraction Contour Plots for the Supercritica!Impact of a'"Rectangular' Cross-Section Steel Projectile at 1.25 km/s

28

L °b • * * .

12.-

10.-

8.

"6.

4..

2.

.0. 5. 10. 15. 20. 25. 30.MF MAT 1 INTERVAL- 1-OOOOE-01


12.

10. -

8. -

6.

4.

2.

0. 5. 10. 15. 23. 25. 30.MF MAT 1 INTERVALS 1.OOOOE-01


Figure 10. (continued).

29

12.

•" 10.-

*-6.

2.

-"0. 5". I0. Is. 20. 25. 0"MF MAT 1 INTERVAL- 1.OOOOE-01


12.

10.

4e

2.

0' . 10. Is. 20. 25. 30'.MFMAT 1 INTERVAL- I-OOOOE-01

I. TIME-- 1.. CYL--0

p.

Figure 11. Sequence of Mass Fraction Contour Plots for the SubcriticalImpact of a "Rectangular" Cross-Section Steel Projectile at1.1 km/s against Bare Composition-9

30

% %

¢.5.0 . .15 . 20~ . 25*. . 30.%%t&1%.t. .. * .HF.%lA . % NEVL ].OO- 0

B 12.

10.8."

4.

2.-0 . I~ IrInIs

0 0. so 10. 15. 20. 25. 30.MF MAT 1 INTERVAL- I.OOOOE-01


12.

10.8..

i

4.-

2.

0. 5. 10. 1I5i 20. 25r. 30,"M" MAT 1 INTERVAL= 1.OOOOE-01



31

• . ...* . . .*... *..... ".. . " C-' N" " a. ' l V ."

I

9.'

*" 7.

I 6.5.-

4.3.2.1 1.m

0. 5. 10. 15. 20. 25.MF MAT 1 INTERVAL- I.OOOOE-01


I 10

9.

78.I 6.S4.

o3.

0. 51 15. '20. 215.MF MAT 1 INTERVAL- 1.OOOOE-01

I 95.-M~ 1.5CCE 5

Figure 12. Sequence of Mass Fraction Contour Plots for the Supercritical*Impact of an 8 mm Diamaeter Steel Projectile at 1.75 km/s against

a Composition-B Target Protected by a 4 amm Thick Steel Cover Plate

32

I --

. . . . . . .. . . . . . . . .

4.

°4.

3.2.

~10.-

0...0. 5. -0. 15. 20. 25

MF MAT I INTERVAL= I.OOOE-OeiTIMEu 1.5 CYCLE= Ž90

| •10.-

9.-1.

"6. -

4.3.-2.

0.-"0O. 5. 10. is. 20. 25.

MF MAT 1 INTERVAL- -.OOOOE-O1STIME- 1.65 CYCLE= 330

SFigure 12. (continued)

33

.i

*7.6. l iin . . llb

5.

4.

3.

2.

0. I0. 2. 4. 6. 8. 10. 12. 14. 16.

MF MAT 1 INTERVAL- 1.O000E-01TIME- .70 CYCLE- 175

7.

6.

4.

3.

2.

0. -0. 2. 4. 6. 8. 10. 12. 14. 16.

MF MAT 1 INTERVAL- l.OOOOE-01TTIlE- .90 CYCLE- 225

Figure 13. Sequence of Mass Fraction Contour Plots for the SubcriticalImpact of a 6 mm Diameter Steel Projectile at 1.4 km/s againsta Composition-B Target Protected by a 1.S mmThick Steel CoverPlate

34

7.

6.

5.

4.

3.

2 . 2. 4. 6. 8. 10. 12. 14. !6.

MF MAT INTERVAL= •.O000E-O,TIME- 1.50 CYCLE= 375

7,.

6.

5.,

4.

3.

2.

1. I

0. 2. 4. 6. 8. 10. 12. 14. 16.MF MAT 1 INTERVAL- 1-OOOOE-01TIME= 1.7 CYCLEF 425


35..

thickest cover plate (h/d:3/4) the rarefaction was observed to overtake theshock completely within the cover plate before propagating into the explosive.In these cases, detonation, when produced, develops at a decaying shock wave andthe critical velocity is higher than night be anticipated.

Determination of Critical Impact Velocity and Comparison with Experimental Data.A limited quantity of covered comp-B experimental data for comparison isavailable. This includes the early results of Slade and Dewey as well as some

more recent results obtained by Howe10 for projectile attack against 105 mammunition. The Jacobs-Roslund formula for covered explosive suggests that theproduct of critical velocity and square root of projectile diameter depends onlyon the h/d ratio. Thus, in Figure 14 we have plotted our 2DE results together

with the aforementioned experimental data in the V'd 1 / 2 -h/d plane. The 2DEpredictions appear independent of projectile diameter but do not produce astraight line In this plane. They agree quite closely with the Howe results atthe two smaller h/d ratios and not as well with the Slade and Dewey experiments.This is a curious result since the Slade and Dewey experiments, with relativelythin cover plates, almost certainly produce shock initiation and Howe hasinterpreted his observed initiations with thicker shell casings as due to ashear mechanism. The matter is further complicated by the fact that the 2DEcomputations do agree with the Slade and Dewey bare charge results. A straightline has been fitted through the 2DE and Howe results for 0.2<h/d<O.6. Both theexperimental and theoreticalresults lie above the straight line for h/d>O.6 anddo not agree closely with one another.

If the 2DE results for covered charges are correct, then the Slade andDewey results become suspect and How '3 experimental initiations Must be due toshock. The difference observed at tte larger h/d values would then indicatelower computational accuracy, possibly due to the inadequacy of the Forest Firemodel when the rarefaction imediately follows the ini--tiing shock. It appearsmore likely, however, that the Slade and Dewey data are correct. Then, the 2DEresults must be in error and Howe's initiations may properly be attributed to amechanism other than shock.

V. SUMMARY

Our computational study of projectile impact shock initiation ofcomposition-B revealed details of the flow fields produced and providedpredictions of critical impact velocities for both bare and, covered explosivetargets.

For bare charges, we observed two different mechanisms by which thecritical velocity is determined. For impacts by projectiles of sufficientlylarge diameter initiation occurs as the impacttinduced shock wave builds todetonation by reinforcement due to burning behind the shock. For smallerdIameteri high velocity projectiles, we saw that detonation or near detonationbreaks out immediately on Impact, but may be quenched by the enusingrarefactions. We found that 2DE predicted the critical velocity accurately, We

also chtckedfP2dt values along the initiation threshold and found them to be

I op. M. Howe, personaZ oomnnication.

36

8 2DE SLADE & DEWEY *

d,6m O I0OWE E8rm ED10mam ¢ .i8

6--

1/2 MM12 m 31

V~d V3.29c 2 51+1.6h/d) (0 mop

C4 -O

.• 20E BARE2 CHARGE RESULTS

0• I I I

0 0.2 0.4 0.6 0.8 1.0hid

Figure 14. Correlation of V*d with h/d - Comparison of 20EPredictions with Experimental Data.

3

. 3,, -

relatively constant. We compared the shook to detonation transition paths tothe Pop-plot for comp-B and found them to agree in the case of a planar shockbuildup but not in the Case of projectile impact, for which multiple paths todetonation were observed. We also simulated the special projectile geometriesconsidered by Moulard, and Iound that 2DE provided a qualitative explanation ofhis observations.

In the Case of covered projectiles we found flow fields Limilar to the barecharge case. The thickest cover plates allowed the rarefaction to overtake theshock before they entered the explosive and s3gnifleantly raised the criticalvelocity. The predicted initiation thresholds agree with Howe's results but notwith Slade and Dewey's.

38

REFERENCES

1. D. C. Slade and J. Dewey, "High-Order Initiation of Two Military Explosivesby Projectile Impact," Ballistic Research Laboratory Report No. 1021, July1957. AD 14S868

2. S. M. Brown and E. G.. Whitbread, "The Initiation of Detonation by ShookWaves of Known Duration and Intensity," Les Ondes de Detonation, C.N.R.S.No. 109, pp. 69-80, Paris, 1962.

3. L. A. Roslund, J. W. Watt, and N. L. Coleburn, "Initiation of WarheadExplosives by the Impact of Controlled Fragments. I Normal Impact," NavalOrdnance Laboratory Technical Report NOLTR-73-124, August 1974.

4. K. L. Bahl, H. C. Vantine, and R. L. Weingarts, "The Shock Initiation ofBare and Covered Explosives by Projectile Impact," Seventh Symposium(International) on Detonation, June 1981, pp. 325-335.

5. C. L. Nader and G. H. Pimbley, "Jet Initiation and Penetration ofExplosives," Journal of Energetic Materials, Vol. 1, No. 1, 1983, pp 1-44.

6. ., K. Huang, J. Starkenberg, and A. L. Arbuckle, "A Numerical Study of ShockIr•tiation of Composition-B by High Speed Impact of Small Stee)Projectiles," BRL Report to be publiched.

7. J. D. Kershner and C. L. Nader, "2DE, A Two-Dimensional Continuous EulerianHydrodynamic Code for Computing Multicomponent Reactive HydrodynamicProblems," Los AlaMos Scientific Laboratory Report LA-4846, March 1972.

8. H. Noulard, "Critical Conditions for Shock Initiation of Detonation by SmallProjectile Impact," Seventh Symposium (International) on Detonation, June1981, pp. 316-324.

9. J. Starkenberg, Y. K. Huang, and A. L. Arbuckle, "A Two-DimensionalNumerical Study of Detonation Propagation Between Munitions by Means ofShook I!itation," BRL Report ARBRL-TR-02522, September 1983.ADA 133680

10. P. N. Howe, personal oommuinication.

39

DISTRIBUTION LIST

No. of No. ofCopies Organization Copie Organization

12 Administrator 1 Directcr

"Defense Technical Info Center US Army Air Mobility Research"ATTN: DTIC-DDA and Development LaboratoryCameron Station Ames Research CenterAlexandria, VA 22314 Moffett Field, CA 94035

1 Chairman ,1 CommanderDOD Explosives Safety Board US Army Communications Rsch

- ATTN: Dr. T. Zaker and Development CommandRoom 856-C ATTN: DRSEL-ATDDHoffman Bldg 1 Fort Monmouth, NJ 077032461 Eisenhower Avenue"Alexandria, VA 22331 1 Commander

US Army Electronics Research1 Commander and Development Command

' US Army Materiel Development Technical Support Activityand Readiness Command ATTN: DELSD-L

ATTN:, DRCDRA-ST Fort Monmouth, NJ 077035001 Eisenhower Avenue

' Alexandria, VA 22333 1 CommanderUS Atuy MissileCoumand

5 Commander ATTN: !)RSHI-RArmament R&D Center Redstone Arsenal, AL 35898US Army AMCCOMATTN: DRDAR-TDC(D) 1 Commander

DRDAR-TSS(D) US Army Missile CommandDRDAR-LCE(D)Dr. R. F. Walker ATTN: DRSMI-YDLDRDAR-LCE(D)Dr. N. Slagg Redstone Arsenal, AL 35898DRDAR-LCN(D)Dr. P. Harris

Dover, NJ 07801 1 CommanderUS Army Missile Command

1 Commander ATTN:' JRS4E-RK, Dr. R.G. RhoadesUS Army Armamefnt. Munitions Redstone Arsenal, AL 35898

and Chemical CommandATTN: DRSHC-LEP-L 1 CommanderRock Island, IL 61299 US Army Tank Automotive"" , Command

"I Director AM:aDR*•. Benet Weapons Laboratory ATTN: .DRSTA-TSL

Armament R&D Center Warren, MI 48090US Army AMCCOMATTN: DRSMC-LCB-TL(D) I DirectorWatervliet, NKY 12189 US Army TRADOC Systems

"Analysis Activity1 Commander ATTN: ATAA-SL

US Army Aviation Research White Sands Missile Rangeand Development Command NM 88002

*"ATTN: DRDAV-E4300 Goodfel'low Boulevard HQDASt. Louis, 1O 63120 DAA-ART-2

";'- 41 Washington, DC 20)10

q .

-4"

%0

DISTRIBUTION LIST (continued)

No. of No. of" Copies Organization Copies Organization

1 Commandant 9 Commander* US Army Infantry School Naval Surface Weapons Center

ATTN: ATSH-CD-CSO-OR ATTN: Mr. L. Roslund, R122Fort Benning, GA 31905 Mr. H. Stosz, R121

"Code X211, Lib1 Commander E. Zimet, R13

US Army Research Office R.R. Bernecker, R13- ATTN: Chemistry Division J.W. Forbes, R13

P.O. Box 12211 S.J. Jacos, RIOResearch Triangle Park, NC 27709 K. Kim, Ri3

Dr. C. Dickinson1 Commander Silver Spring, M 20910

Office of Naval ResearchATTN. Dr. J. Enig, Code 200B 4 Commander800 N. Quincy Street Naval-Weapons CenterArlington, VA' 22217 ATTN: Dr. L. Smith, Code' 3205

Dr. A. Amster, Code 3851 Commander Dr. R. Reed, ,Jr., Code 388

Naval Sea Systems Command Dr. K.' J., Graham, Code 3835ATfN: Mr. R. Beauregard, China Lake, CA 9'3555

"SEA 64EWashington, DC 20362 1 Commander

Naval Weapons Station1 Commander NEDED

Naval Explosive Ordnance ATTN: Dr. Louis Rothstein, Code 50'.. :Disposal Facility Yorktown, VA 23691

SATTN: Technical LibraryCode 604 1 Commander

Indian Head, MD 20640 Fleet Marine Force, AtlanticATTN: G-4 (NSAP)

- 1 Commander Norfolk, VA 23511Naval Research Lab

"'ATTN: Code 6100Washington, DC 20375 1 AFRPL

ATTN: Mr. R. Geisler,Cozmander Code AFRPL MKPANaval Surface Weapons Center Edwards AFB, CA .93523ATfN: Code G13Dahlgren, VA 2244,8

SCommander 3 Air Force Armament Lab (AFSC)

US Army Development and ATTN: Dr. Martin Zimmer. (DLJ)Eglin Air Force Base, FL 32542,.Employment Agency

... ~ATTN: MODE-TED-SAB ' 'A :'. 1 AFWL/SULFort Lewis, WA 98433 Kirtland AFB, N4 87117

42

** * . . . '.-...* .. • ,.• .- . .'*-'• • - - -• • - . . .•. . •- ,.* .. *.. .. - ,

,6

DISTRIBUTION LIST (continued)

SNo. of No. of•pies Organization Copies Organization1 Commander

Ballistic Missile Defense Aberdeen ProvingGroundAdvanced Technology Center Dir, USAMSAA

"ATTN: Dr. David C. Sayles ATTN: DRXSY-DP.O. Box 1500 DRXSY-MP, H. Cohen3 * Huntsville, AL 35807 Cdr, USATECOM

ATTN: DRSTE-TO-FI Director

". Lawrence Liveore Laboratory ".- Cdr, CRDC. AMCCOMc L r L ATTN: DRSMC-CLB-PAUniversity of California DRSMC-CLNATTN: Dr. M. Finger DRSMC-CLJ-LLivermore, CA 94550

I DirectorLawrence Livermore LaboratoryUniversity of CaliforniaATTN: Dr. R. McGuireP.O. Box 808"Livermore, CA 94550

"1 Director"Los Alamos Scientific Laboratcry'ATTN: John Ramsey

* P.O. Box 1663Los Alamos, NM 87544

"1 DirectorLos Alamos Scientific LaboratoryATTN: Edward CourtP.O. Box 1663Los Alamos, NN 87544

"* 1 Director"Los Alamos Scientific LaboratoryATTN: Dr. C. Mader

* P.O. Box 1663* LOs Alamos, NN 87544

1 Director'. Los Alamos Scientific Laboratory-

ATTN: Dr. A. BowmanI P.O. Box 1663* Los Alamos, NI4 87544'

1. DirectorSandia National LabATTN: Dr. J. KennedyAlbuquerque, N4 87115

43

e° ,. *. *j •/* * : * ; . % : d. >-; : .

USER EVALUATION SHEET/CHANGE OF ADDRESS

This Laboratory undertakes a continuing effort to improve the quality of thereports it publishes. Your comments/answers to the items/questions below willaid us in our efforts.

1. BRL Report Number Date of Report

2. Date Report Received

3. Does this report satisfy a need? (Comment on purrose, related project, orother area of interest for which the report will be used.)

4. How specifically, is the report being used? (Information source, designdata, procedure, source of ideas, etc.)

S. Has the information in this report led to any quantitative savings as faras man-hours or dollars saved, operating costs avoided or efficiencies achieved,etc? If so, please elaborate.

6. General Comments.' What do you think should be changed to improve futurereports? (Indicate changes to organization, technical content, format, etc.)

Name

CURRENT OrganizationADDRESS Address.

City, State, Zip

7. If indicating a Change of Address or Address-Correction, please provide theNew or Correct Address in Block 6 above and the Old or Incorrect address below.

Name

OLD OrganizationADDRESS

Address

City, State, Zip

(Remove this sheet along the perforation, fold as indicated, staple or tape.closed, and mail.)

•. .-"- . .. "- . " .•," -. be'• .'. .-. . • ". . •.'e•.-•.•.'• . ,,.'. . '.p • '...:•'•°•.•:"•"40"%

k• - apps.dtic.mil · numerical results aru available. the sequence of evento in projectile...

Documents