unclassified - DTIC

UNCLASSIFIED

AD NUMBER

AD-116 878

CLASSIFICATION CHANGES

TO UNCLASSIFIED

FROM CONFIDENTIAL

AUTHORITY

OCA; Mar 16, 1968 IAW Document Marking

THIS PAGE IS UNCLASSIFIED

UNCLASSIFIED

AD NUMBER

AD-116 878

NEW LIMITATION CHANGETO

DISTRIBUTION STATEMENT: A

Approved for public release; Distribution is unlimited.

LIMITATION CODE: 1

FROM No Prior DoD Distr Scty Cntrl St'mt Assgn'd

AUTHORITY

USNSWC Ltr; Dec 4, 1974

THIS PAGE IS UNCLASSIFIED

UNCLASSIFIED

AD .,1• 74...

,A

DEFENSE DOCUMENTATION CENTERFOR

4 SCIENTIFIC AND TECHNICAL INFORMATION

CAMERO4 STATION ALEXANDRIA. VIRGiNIA

DOWN•GRADED AT 3 YEAR INTERVALS:DECLASSIFIED ApTER 12 YEARS

DOD DIR 52C0O,1 i

UNCLASSIFIED

, ....L..- ,,,i .... ... .... . . .......... ....... .... . .-

I A-

-'- F *- 641 -S,iq it Ural* La 1- 'i--1

Reprducj}b

tolte fllowng ddes AmdiServices TehicaInfrmation igencys

GOVIMENOPOCURIEMET SERAIOP ?V!CE U GOENENITTEREYICRNO REEPONRBIIT OR AN OBULDI[GATO DrAYTOUIE 2,D OHEIOT HTH

1?YPrICE: WHN GVMN OR OTHREA I NYMNER DAICNGSN SPECIIIN HOLDR COER DANYCTA

PERESPONOR CRPORTYO, NORCOVYN ANY OBIAIOiJaTSOER AN TEhSO TOMNFACTUTATTE,

UISE OR SELL ANY PATENTED INVENTION TEAT MAY I1 ANlY WAY BE RELATED THERETO.-

-r- . rx~

Of4~ ~J1A-,ff ~ffr-tile

' V 2~ ~2 IM

DISCLAIMER NOTICE

THIS DOCUMENT IS BEST QUALITYPRACTICABLE. THE COPY FURNISHEDTO DTIC CONTAINED A SIGNIFICANTNUMBER OF PAGES WHICH DO NOTREPRODUCE LEGIBLY.

j

i "!

.t)

U''

NOTICE: THIS DOCUMENT CONTAINS INFORMATION AFFECTING THE

NATIONAL DEFIENSE OF THE UNITED STATES V.T'IT!N THE MEiANING

OF THE ESPIONAGE LAWS, TITLE 18, U.S.C., SECTIONS 793 and 794.

THE TRANSMISSION OR THE REVELATION OF ITS CONTENTS IN . -

ANY M••NER TO AN UNAUTHORIZED PERSON IS PROM-BITED BY LAW.

01

""'

_ - ,~ ~ ~ I * "~ - -- i. .• • .j•...., •

- ° .. - -. , .-... 4 *. dln..•m m• - .l . ab d. ... ',. -

NAVORD REPORT 4236

THE DEVELOPMENT OF IMPACT SENSITIVITY TESTSAT THE EXPLOSIVES RESEARCH LABORATORY

__- BrRUCETON, PENNSYLVANIA DURING THE YEARSN t 1341-1945

16 MARCH 19'56

4 OF

JT

•AA• ::• . >.m•

* Us So NAVAL OflD!AUCE LIEILAT&I16Y

WRITE OCA1, MAR•YLAND

EksslfI.,, . . .. . ........

S• IINCLASSIFIED ...

"This publication is CONFIDENTIAL and shalIlbe safe-

guarded in accordance with the security provisions of

the U. S. Navy Regulations andthe provisions ofappli-

cabL.e Federal Statutes and Executive Orders for Safe-guarding Classified Information. As stated in Article

0906.2 of OPNAVINST 5510,iA, dated Z October 1954,it is forbidden to make extracts from or to copy thisclassified document without authorization."

I

" • UNCLASSIFIED• .i L - -- . -- j i

/-

NAVORD Report 4Z36

THE DEVELOPMENT OF IMPACT SENSITIVITY TESTSAT THE EXPLOSIVES RESEARCH LABORATORY

BRUCETON, PENNSYLVANIA DURING THE YEARS1941-1945

Edited by:H. Dean Mallory

Approved by: E. C. NoonanChief, Fuels and Propellants Di% itriaw

* ABSTRACT: This NAVORD Report consists of reproductions of rep•ortswhich are no longer generally available. They report work ca-ried out in1942-45 at the Explosives Research Laboratory, Bruceton, P4. -heBruceton Impact Machine (now used at tre Naval Ordnance Labor-atory)is described, and the development work with it is fully reported-. I is as

* a result of this investigation of 14 different tool types (harmmner azd anvilcombinations) and of other variables affecting the test value that thepresent NOL standardized impact sensitivity test for high explosi.es wasselected.

Most of the ERL work was carried out by Rogers F. Davis* whose progress reports are the major portion of this NAVORD Report.

Summary reports of his work have appeared in OSRD reports 8C (19442)and 5744 (1945). The first of these is also reproduced in this re.oort.

t

Explosives Research Departrnm.:.U.S. NAVAL ORDNANCE LABORATORY

White Oak, Silver Spring. Maryland

' -. ------

ii,

if I

NAVORD Report 4236 16 March 0956

1 The interest shown in impact sensitivity tes•tig at the Naval Ordnance

Laboratory conference on Explosive Sensitivity, Za-Z9 June 1955,'i pointed to the need for information on this test method. The presentreport represents the whole of the _RL data available to Naval Ordnance

i j Laboratory workers. The complete reports have n%At previously been• javailable to others.

This information is issued under Task NO 301-664143006/12040. Thereport is for information only and is not intended as a basis for

* official action.

.. W. W. WILBOURNE

PAUL M. FYEBy direction

S I I

r I

It

I III III _________I________I II I•

4 NAVORD Report 4Z36

I! t

MAST'_'R TABLE OF CONTENTS

ABSTRACT ----------------------------------------- if " PROM JLGATICN------------------------------------ U

t Introduction ----------------------------------------.. 1TABLE OF CONTENTS - REPORT I ----------------------

I. Direct Imnpact Tests ---------------------------------- I-lb-3Hardening Procedure 1for K-*.-s Anvils and ,tr--_rs ------ I-lb-3

Design No. 1 ----------------------------------------- -6-8Design No. 3 ------------------------------ -7-9

SDesign N. Z----------------------------------------- -1-11-13Design No. 4 ----------------------------------------- 1-11-13Design No. 5 ---------------------------------------- 1-12-14

* Design No. 6 ---------------------------------------- I-IZ--14i ' TABLE OF CONAENTS - REPORT 11 ---- - 1-1-1!! , A Report on the Behavior of Explosives to Mec.Lanical

Shock ------------------------------------------ 11-3-18Striker ?.nd Anvil Diagrams ------------------------------ 6-21Lrnprin.4 of Striker and Awil Surfaces ------------------- 11-7-22

I The Bruceton No. 3 Design ----------------------------- -10-25S j TThe Bru. -*.r& No. I Design --------------------------- II-33-45a

The No. 5 Design ------------------------------------- 11-37-48The c.,. 7 Deeign------------------------------------- 1-50-59The... J3 IDesign --------------------------------- -- 50-59

Tb, Nz,. 9 Design ------------------------------------- 11.53-62Th. .- 10 Design ------------------------------------- 11-5665Th.- I. I Design--------------------------------- ..- 59-68T.ý.... IZ Design -------------------------------------- 82-90

is.*.c ts 12a, I L and 13 --- ------------------------- -85-93CW.r-luding Remarks --------------------------------- 11-105-113

R. %;renccs -------------------------------- 1-107-1ISYABL. Oc CCNTENTS - REPORT I .------------------- 111-1-117

Suppiemernt to Reports of March 13, 1944 and July 4, 1944' Concerning Bruceton Design No. IZ for St.adying the

Behavior of Explosives to Impact ------------------ . I-Z-118TAB.LE OF CONTENTS - REPORT IV ------------------ IV-1-168

* .Con-e,,-,ing Discussion of the Problem of the Behavior ofJ IWplosives to Impact ----------------------------- IV- •-169

-I

| , _

NA Report 4MZ36

Nii3

page

"Summary --------------------------------------------- IV- Z- 169Introductory Statements ------------------------------- IV- 2- 169

I History of the Probltm -------------------------------- IV-Z-169"- 'Statement of zhe Problem ------------------------------ IV-3- 170

. 'Experime',:al Procedure ------------------------ IV-4-1ITExper•n-ental 7'esults --------------------------.-------- IV. 10-177Conclt.,-ionc ------------------------------------------ IV- 15-182

A. Gen-rai Theory ------------------------------- IV-15-18ZB. Recom-nendations for Future Development IV-20-187

References ------------------------------------------- IV-23-190Other Rer.-ts ,which L.clude Reccnt Impact Sensitivity

Work at the Navs.; Ordnance Laboratory, White Oak"Silver Spring, Miaryland --------------------------- IV-24-191

Misce!--; eius X..pc.. s -------------------------------- IV-29-196Com-nent .*n ..ernmiitivity and the Use of the Impact

,i M achines ---- --- ---- --- ---- --- ---- --- --- 197

Statement of Pre, bAern and Background ------------------ 197'mpa-t M. cb'ne Test --------------------------------- 198

• I:• • ILLUSTRATIONS

* REPORT I

TABLE I A Typical Set of Impact Trials ---------- 1-5-7TABLE 11 Height Vs. % Explosions of Tetryl on

No. I Design ------------------------ 1-8-10TABLE IT! Comparative Sensitivities with Some of

the Designs Tried -------------------- 1-8-10TABLE IV Effect of a Binder on 50% Explosion

Heights on No. 3 -------------------- 1-8-10-- ,TABLE V Reliability of the Bruceton Method 1. 1-9-11

TABLE VI A Comparison of the Bruceton Methodwith the Conventional Method ---------- I-10-1ZI '

* • •.REPORT U

TABLE I D;ta by Conventional Method for DesignNo. 3 -------------------------------. I-1?-32

iv i -a

* .. r

NAVOR.D Report 4Z36

5 - Page

REPORT II (cont'd)

TABLE II S'..nm.-nary of 50% Explosion H-eights for1_PEIN and RDX Under Varied Conditionswith Bruceton No 3 Dsigr. ------------ II-24-39

TABLE III Th-e Effect of the Amount of ExplosiveTested in Design ."o. 3 ---------------- II-6-40a

TABLE IV S...=.ary of 50% Explosion Heights of RDX-PE',N Mixtures as Te--t7• by Brucetc-iDesign No. 3 ------------------------ I-30-43a

TABLE V S-.:-nnary of Design No. ! by ConventionalProcedure and Large i.pact Machitie - U-33-45a

TABLE VI Ev:;sion Probabilities of Nitrocellulosecf Varied Nitrogen Content asL-vestigated by Design No. I --------- II-36-47a

TABLE VII S'...-nary of Sensitivity Data for Explosivesin Molten State --------------------- II-40-50a

TABLE VIII Furtlher Data at Elevated TemptratbJresvwith Design No. 5 (Valuzs :ire 50%,aExplosicn Heights in Crn) ------------ 11-42-52

* TABLE IX Cc:nparative Data by Design No. 5 for TNTS* at 85-900 C and 80/20 Nitroglycerin-

Dimethyl Phthalate at lla.n.&TLmperature ------------------------ 11-43-53

TABLE X T-ne Effect of the Particle Z%'a.ure on thzSensitivity of Coraites ar-1 Ballifitit-,As Studied by Desitn '.'. I ---------- U-45-54a

* TABLE 2.1 'he Effect of Elevated Temperature- o,.&*he Behavior of British Cordite toLr Inpact ------------------------------ 11-46-55

t TABLE XII Conpaxativ., Sensitivities (50% Explosion* * Drop-lHeights) by Designs Nr,. 3 and

No. 5 ------------------------------- 11-47-56TABLE XIII Comperati.ve Da'a i•, Design No. 7 ------ 11-51-60

. TABLE XIV Coniparative I.ata Obtained by DesignNo. 8 wi~n a 5 Kilogram Weight ------- 11-5--61

TABLE XV Co:npar.,tive Data Obtained by Designi No. 8 with a.Z. 5 Kilogram Weight ---- 11-52-61

ATABLE XVI SAm.ary of Ste., Disc MIeasurements in1 Connection with Design 9 ------------ 11-54-63

PIT

" " 7- -. , - -" a,- '

I oN _rt 35

SPageft REPORT 11 (cont'd)

TABLE XVU S rnmary oi 50 Exploscn Heights by1k ' Design No. 9 (Z. S Kilogram Hammer)-- II-55-o4

"TABLE XVIII Comparative 5!.W" Explosioa Heights

Obtair.ed by Design 10 and Z. 5 KI',,grarm

100" Hammer --------------------------- U--58-67TABLE XIX Data from PETN tu Show Particle Size

Eiff-L. in No. I I Sensitivity Design!I ,With a Z. 5 Kilo-ramn riammer (Large

L jpic. Machine) -------------------- I1-6 -70I' TABLE AX Comuarative Data with Screened. PSTN as

, Tested by Design i1 with DifferentFlint Papers and 2. 5 Kg Hammer ----- 11-66-74

TABLE XXI Conizarative PETN Data by Design No. 11

*'n* z.-.iall L-pnact Machine -The %E is

* the Result of 100 Trials (Unless,Indicated) -------------------------- 11-66-71

TABI.- XXI! Sr.all Machine Data v-ith Screened RDX byDesign 11 and 2.0 Kilogram Drop

I. H-rnmer --------------------------- 11-72-80

TABLE XXIII Large Impact Machine Data with Screened

RDX by Design I I and 2.5 KilogramDrop Hammer ---------------------- II-7Z-80

TABLE XXIV S,-mmary of Design No. ii Sensitivity4 Lata ------------------------------- - -74-8Z

TABLE XIV Comparative Data for PETN by Design IZ- 11-83-91TAbLE Ak -! Comparison of Regular and "Fensitive"

TNT by Design I -------.------------- 11-84-92

TABLE XXVII Comparative TNT Data by Design No. 12 - U-84-92TA3LO. XXVIUI Ammonium Nitrate Behavior witL Varied

* Papers for No. 12 Design ------------ 11-86-94TABLE XXIX t..,.mparative Data with Sensitive Substances

ofr Designs lZb d 13.----------------- IE-87-95

TABLLZ XXX D'ta for Liquids with Di.sign 13 .---------. -90-98

TABLE XXXI Data for Solid Explosives from DesignINo. 13.------------------------------- 1- •*-99

TABLE XXXII Summary of Significant Graphical Data for| CulCrtain Liquid Explosives av Tested by

Desikz. No. 13 ----------------------- II-95a- 103&,

! iiin- I

R•D Report 4236

I Page

REPORT II (cont'd)

TABLE XXXIII Simmary of Significant Graphical Datafor Certain Solid Explosives as Testedby Design No. 13 -------------------- 11-96-104

TABLE XXXIV Data Obtained by Design 1Zb ------------ 11-99-107TABLE XXXV Summary of Design 13 Evaluation Data

(Solids) ---------------------------- 11-10Z-110

PLATE I The Bruceton Design No. 3 ------------- 11-11-26PLATE 11 Strikers and Anvils for Various Designs - II-12-627

1 PLATE I[ Anvil-Striker Holders ------------------ -13-23-PLATE IV The No. 5 Design as Used for Explosives

in Molten Form --------------------- 11-38-49PLATE V The No. 5 Design with Striker in Position

for Impact ------------------------- 11-39-50SPLATE VI General View of Design No. 10 ---------- I-57-66'6 * PLATE VII 2/0 Flint Paper 1Z. 5 Magnification ------ 11-65-73

PLATE VIII 510 Flint Paper 12. 5 Magnification ------ 11-65-73I PLATE IX Pulverized 210 Silica from 12 Cm Drop

of 2. 5 Kg Hammer ------------------ 11-65-73

PLATE X Pulverized 210 Silica from 150 Cm Dropof Z. 5 Kg Hammer --------- --------- 11-65-73

PLATE XI Dtes;.gn No. 13 -------------------------- -88-96PLATE XII Damage to Large Strikers by Powerful

Explosions of Nitroglycerin, WhenTested by Design 13 ----------------- U-98-106

FIGURE 1 T'ra,:tical Sensitivities by Design No. 3 --- 11-18-33FIGUR.E - Theortial Sensitivities by Design No. 3- U1-19-34FIGURE 3 .:omparative Sensitivities by Design No. 3-11-ZO-35FIGUReit 4 Thenrtical Sensitivities by Design No. 3- U1-21-36FIGURE . Sensitivities by Design No. 3 ------------- 1 -22-37

FIGURE. 6 itneitivities by Design N,.. 3 ------------ U-23-38FIGURE 7 Comparative Sensitivities of PETN Charges

of Varied Weight --------------------- -27-41

SFIGURE 8 Comparative Sensitivities of RDX Chargesj of Varied Weight -------------------- IT-28-42

FIGURE 9 Comparative Sensitivities of Tetryl•! s . Chax,.;s o." Varied Weight 11- 29-43

FIGURE 10 Szo as Ute Fu-ction of Weight of Charge'I 1 Toastud fur ZETN', RDX and Tet ryl byDesign No. 3 ----------------------- 11-31-44

NIP "g i

~ ~,'. , / '.

ffl**ppor4236

1:

Page! •

REPORT I! %ont'd)

FIGURE I I 50% Explosion lieights of RDX-PETN.kMixtures as Tested by Design No. 3 ---- U-31.-45

1 FIGURE 12 Comparative Sensitivities of PETN, ROKand EDNA by Design No. I ------------ 11-35-47

FIGURE 13 Sensitivity of TNT Ls a Function ofTemperpture as Tested by Design No. 5- 11-41-51

FIGURE 14 !---,nparatirt benbitivities oi Molten TNTand RC'Zc NG-Dimet;.,-" Phathlate as:

ý.' aed 1- De tA n N~' ------------------------ 11-44-S4* LI1GURE 15 0!i.•vity c:PE i"N by De-";vn .l". ii .....---- 6Z-71

FIGURE 16 Behavioi ,f :.L'IN to Different Flint Papersand Different Drop-Ham~ners as Tested

ii by Design !No. 'I --------------------- 67-75I FIGURE '7 'ompar,." -° ..•ET SesisiLivities as

.:.vesu., .. ed Ly Design 11 and 5(0 FlintPaper -,- --- --------------- - 11-68-76

a IGU)J4 !8 Conap•p." e FETN Sensitivities asIn-eet'S.'ted by Design 11 and Z/0 FlintPaper -------------------------------- 1-69-77

FIGURE 19 Comparative Sens"tivities of PETN asStudied by Design No. 11 -------------- 11-70-?7

FIGURE z0 Practical Comparative Sensitivities byDesign No. 11 ---------------------- 11-76-84

FICURE 2; Theoretical Senqitiwities 'y Design No. 11-- 11-77-85FIGURE ZZ Cc-m'anra•0'- Bensi•ivi' 4 es by Design No. II- 11-78-86"FIGUPRE 23 Compar_'.vi &r.i---ities by Design No. 11- 11-79-87FIGURZ Z4 -'-,eors.-c-i* Sensilivit - by Design No. 11- 11-80-88FIGURE Z5 Conc.-;irative Sensit; ,ii-.. by Design No. 11- 1-81-89:-1GURE Z6 C•rmpar".J.e Practical S&usitivities of

Liquids by Design No. 13-------------- 11-92-100FIGUJAE 27 --c,.np.rative .racticl Sensitivities by

1 ODesiemn Vu. 13 ----------------------- 11-93-101* F!r..,Ur C mpaz.'t.ive Practix.l Sensitivities by

Deaign-No. 13 ----------------------- 11-94-102"FIGURE 29 Ccm'paratiwv Practical Sensitivity by

f: Design No. 13 ------------------------- 1-95-103

I ..- viii

I- .. . " "

-- o

CONFIDENTIAL4NAVORD Report 4Z36

Page

REPORT 11 (cont'd)

FIGURE 30 The Effect of Desensitizer on t.he 500%

Explosion Heights of Nitrc-lyct-.l. ;.ndDiethylene Glycol Dinitrate as Test-.By Design No. 13--- --.... 1-97-135

FIGURE 31 Comparative Sensitivities by Design No. 12b- I1- V.0- 108FIGURE 32 Sensitivities by Design No. 13 with

Evaluation Appiied --------------------- 1- 103- 111FIGURE 33 Sensitivities by Design No. 13 with

Evaluation Applied -------------------- 11- 104-112.

REPORT II

T.EBLE I S'tmmary of Significant Data by ImpactDesign No. 12.------------------------ 111-9-125

TABLE U Summary of Decign Z Piat;%. ior Class IExplosives Drop--leight of 2. 5 KilogramHammer in Cma ------------------------ 111-10-126

TABLE M! Summary of Design 12 Data for Clas. IIExplosio'es Drop-Height of 2. 5 .Y-'ngramHammer in Cm ----------------..------ M-11-127

TABLE IV Summary of Design 1Z Data for .La- : 1T

Explosives Drop-Heig*-! of .7. S Kil-.,;mHammer inCa----------------------- _ 1-12-1284 TABLE V Summary of Design 12 Data for Clab• IVExplosivee Drop-Height of 2. 5 KiL~kramHammer in Cm ----------------------- 111-13- 129

TABLE VI Summary of Recent Data witb Ee.-r. Z2Material Screened throu•ii 1| er. 50 meshDrop Height of Z. 5 Kilogram F!-mrner in

A VCt .---------------------------------- - 1-14-130STABLE VII Summary of Graphical (Probabiliiy Graphs)

50% Explosion Drop-Heigbh.s for Desig3No. 12 .-------------------------------. 1-15-131

TABLE VIII Direct Comparison of Evaluated 51)%Explos! .n Drop-Heights --------------- - - I-6-13Z

CDNFIDENTIAL

7...j......SY

* - ~.ONID1.4rlALNAVORD Report 4236

Pad"e

i' :REPORT 111 (cont'd)

TABLE IX Surmary of Graphical 50% ExplosionHeig':ts for Coarse M/ateriala byDesign No. 1Z--------------------- 111- 17-133

1' TABLE X Surmmary of RDX Data Showing the Eifect'oof Added Grit ------------------.----. I17-133

TABLE XI Summary of Data for the Effect of ChargeWeight on the Average % Exploions -- 111-18-134

PLATE I Method oi Loadin8, -)r Dcsign No. .1 ---- ni-7-123

_ PLATE II the Design No. 1Z with Charge Read- to1! Receive Impact ---------------------- fn-8-1Z4

FIGURE . Comparative Sensitivities of Class IExp!03iVes by Design No. 12 -- TT-- I-!~19-135

FIGURE Z Comparative Sensitivities of Class IExp•osi es by Design No. 1Z --------- M--0-136

FIGURE 3 Ccrmparative Sensitivities of Class l:S- - • ..... s by Design No. 1Z ..... m-ZI- 137

FIGUaE 4 Comparative Sensitivities of Cla:& 7T

Explosives by Design No. 1Z --------- II-ZZ-138FIGURE 5 Comnparative Sensitivities of Class 11FIUR"6 ExpFosives by Design No. 1Z ----------- -23-139Si FIGURE 6 Compar3tive Sensitivities of Class III

Explosives b-? Design No. 1Z --------- I1I-4-140FIGURE 7 Comparative Sensitivities of Class IV

Explosives by Design No. 12 --------- --M 5-Z141FICTIRE 8 Reference Curve fo- Sensitivity of Trinitro-

toluene as ':'-dhxd by Design No. IZ ---- III-26-14Z

FIGURE 9 Comparative Sensitivities of Coarse (Scr.ienedthrough 16 mesh on 50 m-.shi Mi- ,ialsas Studied by Design No. !Z ----------- 111-Z7-143

. FIGURE 10 Comnp-rntive Sensitivities of Coarse (Screenedthrough 16 mesh on 50 meshi Materiala

FIGURE 11 ,zss Studied by Design No. 12----------- Ii-28-144SFIGURE I I Comparative Senshnhvities of Class I Explosives

by Design No. '-I , Doubtful Explosions

Being Counted as Failures ------------ 111-29-145

'i I

CONFIDENTIAL

.... . _. ..

CONFIDENTIALNAVORD Report 4236 fI

Page

REPORT III (cont'd)

FIGURE 12 Comparative Sensitivities of Class 11(One Class 1) Explosives by Design No. .IZ, Doubtful Explosions being Countedas Failures 3--------------------------- 111-30-146

FIGURE 13 Comparative Sensitivities of Class 11Explosives by Design No. 12, DoubtfulExplosion- 1;ing Counted as Failures --- III-31-147

FIGURE 14 Comparative Sensitivities of Class oIIExplosives by Design No. 12, DoubtfulExplosions bein-g Counted as Failures --- MI- 3 8i

FIGURE 15 Comparative Sensitivities of Class IIIExplosives by Design No. 12, DoubtfulExplosions being Counted as Failures --- 111-33-149

FI4CURE 16 Comparative Sensitivities of Class IVExplosives by Design No. 12, DoubtfulExplosions being Counted as Failures --- M11-34-150

FIGURE 17 Reference Curve for Sensitivity of Trinitro-toluene as Studied by Design No. 12,

Doubtful Explosions Being Counted asFailures a------------------------------. -35-151

FIGURE 18 Comparative Sensitivities of Coarse (Screenedthrough 16 mesh on 50 mesh) Materials asStudied by Design No. 12, Doubtful Explosionsbeing Co•,ntcd as Failures -------------- 11-36-152

"FIGURE 19 Comparative Sensitivities of Coarse (through16 on 50 mesh) Materials as Studied byDesign No. 12, Doubtful Expl:.sions beingCounted as Failures ------------------- -M-37-153

' iIGUR, 20 The Effect of Added Grit on the Sensitivityof RDX, as Studied by Design No. IZ --- III-38-154

FIGURE 21 The Effect of Added Grit on the Sensitivityof RDX, as Studied by Design No. 12 --- IU1-39-155

i ,"jURE kZ The Effect of Added Grit on the Sensitivity"of TNT, as Studied by Design No. 12 --- 111-40-156

Z'3'JM." E 23 The Effect oi Added Grit on the Sensitivityof TNT, as Studied by Design No. 12 --- 111-41-157

xiCONFIDENTIAOL

, . _ _

.. 4 i - a- a. . . . a . -- - .m-

I.

CONFID ENTL%ý1NAVORD Report 4Z36

Page

REPORT IMI (cont'd)

Y1:: iKIE 24 The Effect of Added Grit or. the 20%Explosion Height of TNT and PDX, as

STested by D#esign No. lZ ---------. - U-4Z-15av FIGURE 25 The Effect of Added Grit on the Sensitivity

of RDX, as Studied by Design No. 12 -- MI-43-159

FIGURE 26 The Effect of Added Grit on t.e Sensitivity"of RDX, as Studied by Design No. 12 -- 111-44-160

FIGURE 27 The Effect of Added Grit on the Sensitivityof TNT, as Studied by Desiga No. 1Z -- II..45-161

FIGURE Z8 The Effect of Added Grit on tue Sensitivity"of TNT, as Studied by Design No. 12 -- Mi-46-162

FIGURE Z9 The Effect of Added Grit on thu 50% ExplosionHeight of TNT and RDX, as Tested byDesign No. 1Z ---------------------- IIT-47-163

FIGURE 30 Comparative Fensi-ti-ities of TNT ý.rgerof Var,= '." .i.ht ;.-a Studied by DesignNo. IZ .... ..---------------------- 11-48-164

"FIGURE 31 The 50% Explosion -r.'p-Heiht ,'f TNT asa Fuuc tica of the Weight of Maternal '

Tested by Plceign No. 12 ----------- 11-49-165

1FiGURE 32 Comparative Sensirivliies of RDX Chargesof Var~ed Weight as Studied byDesign No. IZ ----------------.---- M1-50-166

FIGURE 33 The 50%. L.'1p--sn £rop-Height of RDXas a Function of the #eig-t uf 1/-ptorialA I Tested by Design No. 1Z ----- - -----.- 51-167I I

"REPORT IVTABLE I Summary of lmportaL Types of Piston-

I. ' Anvil Combinations Developed fir theLapact Test at Bru,;eton ........ IV-5-17z

TABLE IU Comparison, 'f Ob.eerver and Me.h-anica,Light !or Mezan Values of :5ý -Trial "Upa.d Down" Runs from Type 12 ImpactTests -------------------------- IV-11-178

xiiCONFIDENTIAI.

ii u||

CONFIDENTIALZAVOKD Report 4Z36

Page

REPORT IV (cont'd)

TABLE III Data Showing Reproducibility of ResultsObta•ined by Personal Judgement fromthe Type IZ Impact Tests Involving50 Trial "Up and Down" Runs -------- IV_2-l179

TABLE lia Data Showing Reproducibility of ResultsObtained by Trigger Circuit from theType IZ Impact Test InvoLving 50 Trial"Up and Down" Runs ----------------- IV- 13-180

TABLE IV A Rough Comparison of Some CommonExplosives as to Hardness and ImpactSensitivity ---------- --------------- IV-17-184

FIGURE I Design No. I ------------------------- IV-6-.173FIGUR.E Z Design No. (14) ----------------------- IV-6-173

* FIGURE 3 Design No. 4 ------------------------- IV-7-174FIGURE 4 Design No. 5 ------------------------- IV-7-174FIGURE 5 Design No. 3 ------------------------- IV-8-175FIGURE 6 Design No. It -------- ft.----------- IV-9-176FIGURE 7 Design No. 12 ------------------.----- IV-9-176

xii

I *

!*

./

it- CONFIDENTIAL* .NAVORD Report 4Z36

TIHE DEVELOPMENT OF IMIPACT SENSITIVITY TESTSAT THE EXPLOSIVES RESEARCH LABORATORY

BRU':ETON, PENINSYLVANIA DURING THE YEARS1941-1945

I "Introduction

4, Darin•" World War 1f. interest in explosives research wasstimulated and much was accomplished. The aspect of the subject relatedto the ease of initiation and ?ropagation was dealt with by a number of

Zr• groups but principally those associated with Bowden and Ubbelohde inthe United Kingdom. Their results have been published both in theclassified and open literature. Professor Bowden and his co-workers

* have published extensively in the Proceedings of the Royal Society, andin addition the monogra;.h by Bowden and Yoffe, Initiation and Growthof Explosion has been widely read. Professor Ubbelohde and his

I co-workers ha% e published their collected. works in Phil. Trans.,A241, i97-Z96 (1948). Impact testing has been an important phaseof the British work.

I .A large part of the United St•ates' Impact Sensitivity work,especially the development of methods, was done at the ExplosivesResearch iaboratory, Bruceton, Pennsylvania during the period1941-1945. Results of the work were published as OSRD Report 804(1942) and OSRD Report 5744 (1945). 1he first of these is in the formof a brief progress report (which is included in the present collectedreport) while Lhe second is a summary of the findings which includesa short description of each tool type and sorme oa the results obtained.

Dam from most impact machines will, in a general way, be

found to match the sensitivity orders as deternined by one of the ERLS. •:-.,• tool types. It is hoped that information such ae t'.-'."s will be an aid in theI understanding of impact data. By following thu, atep by•step i.rogress

i / made during the evolution of the various tools, some groups may decide

their tests can be improved upon by adaptin& a different tool design totheir available machine.

j jWith the exception of the first few pages which are from• ]t OSRD 804, the present report is made up of detailed progress reports

by Rogers F. Davis to Dr. E.H. Eyster at the ERL, Bruceton, Pa.The original editions ware limited to about five copies of each. It is

CONVIDENTM6L

CONIFIDENTLAL

I, NAVORD Report 4236

I thereiore obvious why most persons engaged in impact testing are notacquainted with them although they have probably read the two OSRD

SRt..'its. 804 and 5744.

IAlthough it might have been possible to reissue the two OSRD* Reports instead of the present detailed material, it was felt importaut

that all the information be made available especially now that furtherattempts are being made to understand the mechanisms of initation.

,It seems reasonable to ouppose that the hot spot idea, an suggested byd iUbbelohde and developed by Bowden, can now be extended by considering

t"he effects of heating times. It is rnot too difficult to detect the timeelement, in a rough fashion, in Davis' data as teola are used whichba-he various degrees of confinement; the hot spot temperatures reportedby Beu'den may in this way be somewhat dependent on tool type.

I -This collected report has been made up from the following:

1. *,.. 804 (the impact part covering work up to July 1942)

2. The Behavior of Explosives to Mechanical Shock (coveringthe period August 1942 through June 1944)

. 3. The Behavior of osives to Mechanical Shock asStudied by Bruceton Impact r ai No. 12 (covering the period August1943 through July 1944)

-/ 4. Concluding r- .ioi. , the Problem of the Behavior of- Explosives to Impact (dat. )ctober 1945)

"The reports have been kept in the above sequence to make it easier tofollow the chronological development ol the work.

| A uniform page numbering has been adopted throughout this

I? report. Every page carries a Roman numeral referring to the numberof the ortZinal Bruceton report, followed by a number designating the

, page within that pacricular report (I -1-13; U-l-108;III-l-51; IV-I-29)..Ir,,mediately below is the page number of the collected report (1-200).

CONFIDENTIAL

. . - - - - -. . .

CONFIDENTLA.LNAVORD Repost 4Z36

* - TABLE OF CONTENTS

REPORTI

*I •

S I Direct Impact Tests

S•A-.DENING PROCEDURE FOR KETOS ANVILSAND STRIKERS

Ii IT TABLE I - A TYPICAL SET OF IMPACT TRIALS

* Design No. I

Design No. 3

TABLE II - HEIGHT VS. % EYPLOSIONS OF TETRYL

STABLE II - COMPARATIVE SENSITIVITIES WITH SOMEOF THE DESIGNS TRIED

' ITABLE IV - EFFECT OF A BINDER ON 50% EXPLOSION

HEIGHTS ON NO0. 3

• CONF13DENTIALi w ...... • ~ ~- . .... 1"... .

I/"" O T - -

ICONFIDENTIAL

NAVORD Report 4236

II

TABLE OF CONTENTS (Cont'dj

REPORT I

TABLE V - RELIABILITY OF THE BRUCETON

METHOD

TABLE VI - A COMPARISON OF THE BRUCETON

METHOD WITH 711Z CONVENTIONALM"ETHOD

Design No. ZtDesign No. 4

I '1 Design No. 5

it Design No. 6

tj j* 4 . CONFIDENTIAL

U . _ ___ ___ ___ ____ ___ ___ __.

CONFIDENTI.ANAVORD Report 4Z36

Report I (from OSR.D 804)

* |Direct Impact Test*

The work de sc-bed in this section has been perftrmned byMr. H. Knud~er. Dr. C. H. Sage, Dr. Z. H. Eyster, and, . Mr. Rogers Davis.Mr. ROf the three impact machines of the Bureau of Mines only the

Sso-called small one has been found suitable for general testing workon military explosives. It is described'in the Bureau bulletin No. 346,but certain rnodiiications have been introduced for the purposes of thiswork. Thus the heavy striker of the original machine has been replaced

4: by a bmail steel rod, Z inches long and i1Z inch in diameter made of ahigh -grade a/ioy steel (Ketos steel) properly hardened by heat

II treatment.

S* HARDENING FROCEDURE FOR KETOS ANVILS t-ND STRIKERS•iPut piece t,- be hardened in furnance when temperature reaches

680 0 C. Raise te.mnperature quickly and hold constant for fifteen mianuLesas follows:

l/2 inch piece 1 1/4 inch pieceKetos Steel 800 0 C 815 0 C

Qiench in water-free oil until cool enough to hold in hand.Then transfer immediately to a tempering oil bath for two to three hours.Temperature held at 230 0 C.

Ketos steel is obtained from the Crucible Steel Co.mpany.

The anvil has been made in the form of a rod of the same steel,1 1/4 inch in diameter and from I to 1 l/Z inch long depending on the typeof holder used. Both are mounted in a steel frame (See Figure 1),* which

jis reai-ily removed from the machine for interchange of damaged parts.The striker slides in a vertical sleeve opposite the center of andperpendicular to the top surface of the anvil; the latter is firmly heldin the frame and rests directly on the heavy base plate of the machine.The 2 kg. weight of the original machine has been replaced by a 5 kg.

I * weight for the majority of tests.

* *Figure not availakie in original I- breport.

3jj ICONFDZNT•I6L

1. _ - -~ *- ..- -

/ aI i I I i

All

CONFIDENTIALI NAVORD Report 4236

I •A second zmachine has been built, using the same striker-anvilunit but emLodyi-. several changes in the mechanical details ofconstruction. The most insortant one of these is a mechaadcal releaseof the weight, rat.3er than a magnetic one. Actual experience has shown,however, tbat the magnetic device is really preferable, as giving a morereproducible tra~ectory of fall of the weight. Thid second machine has

*"been provided w.th a device to preent rebound. The observation ofd detonation or. i.-.ct is auditory on both machines. With some explosives,

particularly the ;ess sensitive ones, only partial detonations are usually,, observed, -which are sometimes difficult to classify; this, however, isi! only one of the lesser d.ificulties of these tests.

In accord with other reports on the subject, the work with these

machines coon s6.owed that the precise form and nature Ot the strikingsur'aces, as we'l as the manner of distribution of the explosive and itsform have prof.-.=d effects on the results of the tests.

Experiments have been made with a considerable number ofditferently fash'-i'=ed strikers and anvils, but most of these have beenfound impracticable because of a variety of reasons. The chief of themwere th3t: a) the metal parts did not stand up well under the punishment,giving irrepr-'oduci;e results; b) the- relative sensitivities of the knownexplosives did iflt fall into a series commonly accepted, the duplicationof which seemed to be desirable, and c) the designs were too "insenaitive"

i. e., only very sensitive compounds could be fired with the maximumavailable energy, 5 kg. weight from 100 cm. height.

Extended series of trials have been made with the following) .designs of the striker-anvil combination:

I) A Cat-ended striker of 1/2 inch diameter on a flat anvil.

* 2) Same design as No. 3, but the flat anvil is replaced by a

truncated cone, the flat top area facing the center of the cap being1/4 inch in diameter.

3) A Cat-ended striker turned down at its lower end to0. 306 inch diameter, over which slips a brass cap (standard percussioncaps, obtai.eaed from the Western Cartridge Company, East Alton,illinois) of 0. 310 inch internal diameter and about 1/8 inch high. The

41' _______

I. ______

S. . . I I *I

CONFIDENTILLNAVORD Report 4Z36

anvil is flat and the explosive is placed in the cap.

4) Thste striker has a flat bottom 318 inch in diameter with abevel 3/64 Inch wide ground around the edge at such an angle that goodcontact is made with the surface ,oj a spherical• depression ground out inthe anvil on a 1IZ inch radi6s.

5) The striker is flat ended and fits snugKly into a flat-bottomeddepression in the face of the anvil, 1/16 inch deep ar.d 3/8 inch indiawmeter. The explosive is place$ in this ciepression and covered by ?-layer of tin foil. The striker is then pressed into the depression so ib'atthe foil acts as a gas seal around the edges.

6) Same design an No. 2, but a spherical depression 3/16 inchin diameter is ground out on the top face of the anvil on a 1/4 inchradius.

With all designs it is essential to use a. standardized quantity ofthe explocive, the striker must be presse4 by hand or-to the sanmple beforedropping the weight and the parts must be washed with a suitable solventafter each trial. Even though all efforts are made to standardize theparts and their heat treatment, it as not always possible to duplicate thedata of one particular set of parts with the next one, sa-emingly identical.Less satisfactory designs among those enumerated above, give heightsfor 50% explosions which vary by as much as a factor of two from oneset of parts to another. Furthermore, the .eaze set of parte Ehowschanges in results due to proSressive wear. although such chanles arefrequently not pronounced. Nonetheless it has been found essential tomake tests on standard substances at very frequent intervals. This.naturally, delays the progress of the tests and, hence a relatively smallnumber of trials is now made on each material. "Wh.,t is lost thereby instatistical accuracy of the result, is gained through bhtter preservationof parts for the next series of trials with a standard substance. Unlessthe parts show obvious signs of wear and of irreproducible behavior,the presently accepted test includes but twenty trials on the new material,followed by an equal number on a standard.

The materials are measured out for the trials not by weight b*tby means of a small spoon holding about 20 mm 3 of the explosive. Thisprocedure is faster and yet does not irnpai&, the results significantly.

1-3

* CONFIDENTIAl.

-"- - .-.- - ----- -L

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

CONFIDENTIAL

NAVORD Report 4236

* All materials are carefully dried before triala but otherwise mostnew materials go into the tests as received. Afterwards they are

remeasetred after grinding and sieving, only the frz.,tion passing No. 100and retained on No. 200 being used. All standards are used after similarsieving.

With mixtures, particularly including substances of greatlydiffering eensitivities, sieving must be avoided to p.revent segregation.

d# Milled mixtures, particularly those including waxy substances, are notsubjected to further grinding to avoid changes in the state ofphlegmatization. Cast mixtures also often give quite differentsensitivities from those of a simple mechanical mixture of the sameingredients. The procedure has therefore been developed of casting verythin wafers of such mixturea between glass plates, stripping them, cuttingto size, inserting into th; brass caps of Noe. 2, 3, or 6 design, andIl cautiously fusing again or the bottom of the cap by heating.

, The operational procedure is as follows: a trial is made at somearbitrary height of fall and for the next trial either the next lower or thenext higher height is chosen, depending upon whether explosion did ordid not take place. This is continued until the test is completed. Trialsare made at intervals of I cm. or more, depending on the total height.The trials are continued until sufficient data have been accumulated and

* then a standard Is run in the same manner.J7

Table I shows a typical set of trials, capital E's indicatingexplosions, capital N's absence of same. All partial audible explosionsare counted as E's but a mer.- visual charring of parts of the charge isconsidered to be an N. in cal.:',lating the height of fall which gives

"500 explosions, it is assumed that: a) if, in a trial from a given height,explosion occurs, explosior would have occurred in this trial from anfgreater height; b) if, in a trial from a given height, no explosion occursno explosion would have occurred in this trial from any lower height.To express these assumptions, lower case n's are written in Table Ibelow each capital N and lower case eos are written above each E. Tocalculate the percentage explosions for each height, one takes the sumof all N, n, E, e for this height and divides by this the sum of E and o,thus:

ZiE + re x 100Spercentage explosions - X + e + ZN +E

£This prucedure is a convenient and rapid one for determining the

1-4

64 CONFIDENTIAL*1'I_ _ _ "" 'o

CONFIDENTUALaNAVORD Report 4236

height for 50% explosions, if it is combined with the above describedprocedure for carrying out the trials. It should not be used with the moreconventional procedure of making equal numbers of trials at preselectedheights, as erroneous results will be obtained. It is also not correct fordetermining the height for any other percentage of explosions, except the50%0 height. Finally, it must be noted, that, all. data are discarded inthe beginning of each series of trials unti a break is reached, i.e., ifat first explosions were obtaines, the data are taken from the first trialwithout explosion and if the trials started with non-explosions, then thefirst valid result is an explosion obtained bysucceL"sive raising of theheight of fall.

TABLE I

A. TYPICAi. SET OF IMPACT TRIALS

Heightof Wgt. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

60 cm e 0 ee e •E a 0I I55cm E I E e e E N E E I e

50 cm E N E I N N" N E E

45 cm N n E N n n n n N

40 cm n n N n n n n n n

60 cm 11 = 100 ".E 50%pt. 50 cm.

T11

S50 cm 5 =50% E

4S 4cm I - IE

40 cm 0 = 0%E

., 7ii 45CONFIDFNTAL

u27

CONFIDENTIALNAVORD Report 4Q36

We shall now consider the different designs of the strikingsurfaces one by one.

Design No. I

This design gives reproducible results on Cyclonite (23 cm.) andmore sensitive materials, like PETN (15 cm.) etc., if the area of theexplosive tinder the striker is controlled. TNT gives partial and onlyoccasional detonations at heights of fall even as great as 300 cm. Even

4# ,,for Tetryl this design is unsuitable. The following Table II shows the

percentages of explosions (calculated by the conv-ntional procedure ofmaking equal numbers of trials at preselected heights) as function of

xl 'height, obtained for Tetryl. It will be observed that the percentage ofexplosions first increases with height of fall, but then decreases and doesnot reach 50% at any height tried. The explanation of this behaviour isfound in the easy explusion of the material from between the smooth steelsurfaces of the striker and the anvil. Befoe enough energy has beensupplied to particles of Tetrfl to ignite them, they are scattered about.Only a small fraction of the initial sample is left under the striker and thisis found in the form of an extremely thin, wax-like, non-crystalline layer,

Iwhich is probably -ust as difficult to bring to detonation as is gelatinizednitrocellulose (compared with the fibrous material). The observeddecrease of percentage explosions at greater heights is probably due toboth causes; more complete '-jec'jon of the material and more complete

'ge"leltzdzation" of the remaining fraction.Very striking results are obtained by placing tin foil (0. 0005 inch

thick) on the anvil, then the sample, then foil ag-'in. The heights for moresensitive explosives are not greatly changed thereby b;t now Tetryl isfound to bemore sensitive than Cyclonite. (See Table Ill). This is a" finding which, as will be seen later, is obtained with all designs providinglarge resistance to lateral motions of the explosiv.e particles and placingthe sample in a condition of "high confinement" as regards the freedom ofJ escape of the products of explosion.

When a small depression is ground out in the anvil of No. I, the height offall for 50% detona .on with Cyclonite is depressed to 7 cm. althoughneither Tetryl nor TNT can be located on the scale reaching up to 100 cm.Thus with such design we find conditions opposite to those obtained with tinfoil.

1-6

CONFMIENTIAL L

r7 4.

Ae

CONFIDENTIALNAVORD Report 4236

D._ign No. 3

This has recently been adopted as standard for explosiv-es of equalor greater sensitivity thar. Tetryl. The advantzges are a rapid executionof tests, good reproducibility ar•d a "reasonable" sequence ofseitsitivities of the several common explosives tried. Wit-h %,.ight! of f-l1lof more than about 70 cm. the striker deteriorates rapid7y and yet TNTwith this design is stitl above 100 cm. The excellent repz.-.ducibiLity isshown in Table "-/ which gives results of tests on a series of materialsconsisting of a rather insensitive crystalline filler with a resinous binderwhich incr'ases the sensitivity.

The type 3 anvil and striker combinatioa has now been in use longenough so that a critical study can be made. both c-f this pearticular device,and also of the operationas procedure adopted for determin-i.g the 50%fexplosion points. During the recent intensive study of the sensitivities ofcyclonites prepared by various means, a sample of cyclo=-_te received"fro-n England has been used as a standard. We now have data on thismaterial which include well over a thousand falls of the weight, and hencea reliable statistical treatment should be possibl.e. Mor=a.Uy one run oftwenty shots on the standard is made ea; c y, and this -- nmary includesresults fro.n fifty-four such runs, extending over a two and a half-mo.zperiod, and no results have been discarded. During this ti.ne a largenumber of strikers and a smaller number of different anrv-s have been used.

Figure Z is a plot of the per cent explosions obtai=ed against theheight of fall of the 5 kg. weight. Because of the procedmre adopted, inwhich the height of fall is lowered or raised after each trail, dependingupon whether an explosion is obtained or not, most of the trials areconfined to a rather narrow region in the neighborhood cf the fifty per cent"point. For example, there were 353 trials at 50 cm., 2S1 t7ials at55 cm., Z52 trials at 45 cm., but only eighteen times was it necessary togo to a height of b5 cm., -"r one of 35 cm. From the gra,-3h it is seen thatthe 50% point is at 50 cm. (It should be pointed out that. - resultsinclude only actual ;alls of the weight, and no "assumed" results are used,as is done in our standard method of treating the results.)

In order to give an idea of the reliability of our stamdard method ofobtLi.ing the sensitivity from a run of twenty rhots (for want of a bettername we shall refer to this as the "Bruceton riethod" to distinguish it fromthe conventional method) we give in Table V the values obtained for thefifty percent explosion points in each of the fifty-four runs of twenty shotseach."•Figure Z not available in original report.

CONFIDENTALL

tIMMUNE

S• CONFIDENTIAL".. ~NAVORD R~eport 4236

TABLE I.

HEIGHT VS. % EXPLOSIONS OF TETRYL ON NO. I I0LSTGN.

Height, cm. 0 30 50 75

. Explosion 0 25 z0 10

ji TABLE IU

"COMPARATIVE SENSITIVITIES WITH SOME OF THE71 : DESIGNS TRIED

1; Figures given give height in cmn. for 50% explosions with a' i 5 kg. weight

Design COMPOUNDS

PETN Cyclonite Tetryl P. A. TNT Trinitroanisole

INo. I is Z3 >100 >1003 No. 1; tin foil 6 17 8 >100

* No. 3 z0 50 47 >100No. z 45 44 >100

* No. 2; tin foil 21 18 62f .-" No. 4 10 19 20 about 100 >100No. 5; tin foil 8 20 10 45No. 6 33 56 64

TABLE IV

EFFECT OF A BINDER ON 50% EXPLOSION HEIGHTS ON

1 NO.3

% Binder 0 1 2 3 5 7 10 is; Height, cm. 88 27 Z3 18 17 14 13 10

1-8

10CONVIDENTIAL

7-

CONFIDENTIA,NAVORD Report 4236

TABLE V

RELABILITY OF THE BRUCETON-!I METHOD

Fifty perecnt Explosion Height No. of Tiz-.,:. Obtained

-i 41 cm.42 cm. t43 cm.44 cm. 145 cm.46 cm. z45 cm. I48 cm. 949 cm. 550 cm. 756 cm. 6iI52 cm. S

53 cm. 8i ~54 cm.

55 cm.Ii ~56 cm.

S57 cm. 0

58 cm. 059 cm. 0I.14~ cm. I

4 Total 54

j "It is seen ;rom this table that in fifty-four trials, the 50% pointfound was never more than 10 cm. from the true value, 50 cm., and in40 cases, or 74%, the value was witain I cm. of the correct one.

A few experiments have also been made to compare more direct-ly the results with the "Bruceton" method and with the conventional one.On five different days, a run of twenty shots was made on our standardcyclonite and the 50% point calculated by the Bruceton method.Immediately thereafter, forty shotts were made alternately from aheight fiecentimeters above and five centimeters below this fiftypercent height. Frorn these forty shots, a second fifty percent heightwas obtained. In order to give an idea of the variations experienced, the

/ e

.................................................. • • : .. . - - . .'.--.•. -- '--,,•. • %k•k:

V CONFIDENTIAL"NAVORD Report 4Z36

the results are given for the first twenty and second twenty shots of eachbatch of forty shots. These are shown in Table VI, where E meansexplosion and N means no explosion.

r TABLE V1

A COMPARISON OF THE BRUCETON METHOD WITH THECONVENTIONAL METHOD

Expt. Bruceton 1st Z0 Znd 20 40 ConventionalNo. 50% Height Shots Shots Shots 50% Height

1 50 cm. 55cm. 8E ZN 7T 3N 15E SN 45cmS !. 45 cm. 5E SN 5E 5N 1OE ION

2 " 50 cm. 55cm, 6E 4N 6E 4N IZE 8N1445cm. ZE SN 6E 4N 8E IZN 50 cm.

3 49 cn.. 54rcm. 7E 3N E .0N 13E 7N44cm. 4E 6N .Z 9N 5E 1SN 50 cm.

! 4 18cm. 55cm. 9E IN 8E ZN 17E 3N45 cm. 2E SN IE 9N 3E 17N 50 cm.

* 1 5 53cri. 58 cm. 6E 4N 6E 4N iZE SN• " 48cm. 3E 7N SE 5N 8E IZN 53 cm.

The agreement between the two sets of results is very good, and4 "both remain close to the true value of 50 cm.

• The important feature of the Bruceton method is that it makes itextremely probable that the fifty percent height will lie within thereasonably narrow range in which shots have been made. The particularmethod adopted tor finding the fifty percent height from the measurements

; iis simply an objective way of smoothing the results and obtaining ananswer. Actually, if one takee instead merely the midpoint of the rang*over which shots have been nmode, the results are very little affected.S• Theae statements ar* not meant to imply that the conventional method ofmaking equal number o( shots at several preselected heights will not

give good results, but rather that the Bruceton method, we believe, will

t .... 1-10

jsji .r1

II II I . ....... - - -


give a reliable answer with a smaller number of trials.

Design No. 2

This design lowers considerably the heights of fall for TNT,bringing it almost onto the scale of the 100 cm. machine, but alters theheights for the more sensitive comspounds but sUghtly.

Tin foil, however, has very profound effects on the results, aswas the case with design No. 1. Table MI gives heights of fall for501/ explosions for designs Nos. 3 and 2, the latter with and without tinfoil. Equally striking results are observed in Figure 31which shows thesensitivities of mixtures of Cyclonite and TNT as function of composition

* and of the machine design. It is evident that almost any dependence ofsensitiv-ity on composition can be obtained at will and therefore we do notattribute any fundamental significance to the composition-sensitivity curvesdescribed by Urbanski*.

It is interesting to note that the effect. here descriLed for tin foilare apparently not connected with the chemical nature of this material.Very similar, although not necessarily identical results, h;ave been also

* obtained on using cellophane or thin rubber membranes. Painting ofthe striker with rubber cement or depositing on it a thin layer of wax alsoresult in extensive lowering of. the 50% points and in "inversions" of thesensitivity order of some explosives. And yet, wax mixed with the sameexplosives acts as a phlegmatizer, i.e.. raises the 50% points.

Design No.4

4i This was used extensively during the summer of 1941 but is nownot much used because of poor reproducibility of the results, in particularbecause of difficulties in preparing similarly acting metal cmponents.

S~It covers about the same range as No. 2. TXTL does not fit onto the 100 cm.scale but all. more sensitive compounds can be studied by this machine

design. Table II gives comparative heights of fall for 50% explosions, butthese have little absolute significance because of the wide scattering of

* data.

SeZeit. F. das ge;."Schiess.-und Sprengsw. 33, 41 (1938)1 .'igure 3 not av~lable in aoriginal r& ot.

13CONFIDENTIAL

Ii _______.. ... . ,_._

S ...... • • .,. .... 7! ".: . . .• .• ..

CONFIDENTIALNAVORD Report 4Z36

Design No. 5

Design No. 5 is the extreme cace of iniversions of the "conventional"order of sensitivities of common explosives, which is undoubtedlyassociated with the extreme confinement to which the materials aresubjected here on impact. It is a vary "sensitive" design, i. e., even

* " insensitive explosives give low 50% points with it. Table llI shows someof the results obtained, but absolute figures have little significance sincethe results are no. very reproducible. This design is now in use only forvery insensitive pure materials and for liquids, for which it Is moresuitable than the others. The chief Gbjections are the difficulty of makingreproducible indentations in the anvil, poor reproducibility of the .esultsand the "unnatural" order of sensitivities of some of the common explosives,particularly of phlegmatized mixtures.

Design No. 6

Design No. 6 is still in a very exp -e ' -• g :'-_--sut appears to* be suitable for work on insensitive rompounds, provided it can be made

sufficiently reproducible. It is more "'.nsitive" than the similar designNo. 2, a finding that could have lbeen expected because in other cases alsothe introduction of a central indentation in the anvil lowered the 50% points.In the present case, the lowering is particularly marked for insensitivematerials, as is shown by Table II. This design is being considered forwork with materials of the TNT type.

I The preceding dezcription seems to us to be ample evidence thatit is futile to speak of the impact sensitivity of an explosive, even relative4 " to that of a standard, unless the design of the machine used has beenrigidly specified. It has been frequently stated in the literature thatimpact sensitivities of explosives form a very definite series but thatfrictional sensitivites form a different series, and that, depending on theamount of friction in a blow, different results may be obtained. This isundoubtedly true and, in fact, explains much of the data presented above.However, from th.- mechanical point of view, all machine designsdescribed here deliver direct impacts, not frictional ones. We mean by adirect impact one in which the moving surface strikes a perpendicular blowon another one, while by a frictional impact is meant one in which aglancing blow is delivered. From this point of view all designs here givenare substantially identical. They differ, however, very much in the extentto which the explosive itself is moved by the impact. In design No. 1,

the explosive is freely scattered from under the striker; in designs

S14COFE7A

-7 *

- .

CONFIDENTULNAVORD Report 4236

Nos. Z and 3 it is driven outwa-.ds againt the resistance of the brass cap,which is alway. found bulged by a ring of the explosive substance after anegative trial; in designs Nos. 4 and 6 the material is partially driventowards the center of the anvil, where it is subjecte4 to high compression.Finally, in designs using tin foil or other soft materials, particularlyin design No. 5, the motion of the explosive is greatly hindered by the

r " increased friction against the soft, yielding material. These statementsare all confirnred by actual C'iservations on the explosive* after negativetrials and they must account for the wide variety of the results obtained.

In service use an explosive may be subiected to a great variety.ofsudden stresses ar.,4 no single laboratory test can be expected to reproducethem all at once. Worse than that, it is very difficult to decide whichlaboratory design correctly reprodulces a given service hazard. A large

, body of empirical knowledge has been accumulated, however, which VOuldsuggest that designs numbered before 4is Nos. 1, 2, 3, =nd 6( without tin

* foil) measure reasonably faithfully the relative hazards encountered inhandling several explosives considered. These designs, therefore, we now

* prefer to use, although we believe that in order to explore more thoroughlythe dangers of a given new material, it should be tested on more than onedesign in the laboratory and then subjected to very extensive large scaletrials.

11-13

CONFZDENTIAL

.... I I "" I - • I II - IZ" 7- - --- - •--- • -- I . . .... . I I•L-.


A

TABLE OF CONTENTS

SI REPORT II

SINTRODUCTION

General Di3cussion of the Impact ProblemTreatment of Data

.0, EXPERIMENTAL RESULTS

.' The Bruceton No. 3 Impact Design.. Equipment and Procedure

Treatment of Data* General Results with GraphsS .Effect of Poor Apparatus

Effect of Weight of Material TestedInvestigation of a Series of Mixtures

The of Sens:tive and Intermediate Materials

The Bruceton No. I Impact DesignV Procedure and Resultsj. Effect of % Nitrogen on Sensitivity

of NitroceUuloseThe Bruceton No. 5 Impact Design

Equipment and Procedure

Results with Liquid and Molten ExplosivesImpact S,-neitiveness of Cordite and BallistiteSome General Results by Designs No. 3and No. 5

jI,,Impact Designs No. 7-10 with General Results andDiscussion

The Bruceton No. 11 DesignProcedureGeneral DiscussionEffect of Varied Flint Papers and"Particle Sise

General Resalts (with Graphs) for DesignNo. I1

•'"" 16

CONFIDENTIAL

, .* * . ..

{,I: CONFIDENTIALS~NAVORD Report 4L36

ii ~TABLE O F CONTENTS ( Cont'd)

REPORT 1M

40 The Bruceton !'. 1Z Impact DesignStudy of "Sensitive" TNTZffect of Paper Base of Flint Paper

imnpact Designs No. IZa, IZb and 13Equipment and ProceduresGeneral DiscussionData for Design No. 13 (with Graphs)Data for Design No. 1Zb (with Graphs)A Different Interpretation of Results

CONCLUSIONS

Evaluation of Certain Designs with ConcludingRemarks

REFERENCES

'I7

IoI

II-'j i!_ ICONFIDENTIA


- REPORT II

i .Bruceton, PennsylvaniaJuly 4, 1944

Re--Port to: Dr. E. H. Eyster

From: Rogers F. Davis

I S•bject: A Report on the Behavior of Explosives toMechanical Shock

From some of the literature (1, 2 ...... ZZ) and past research"there has developed the idea that to investigate explosives as to theirresponse to mechanical shock will arrange these materials in a generalord.er of behavior or so-called stnsitivity. This general order is thoughtto be of value in regard to practical handling of various explosives in thati~.±i-.duals may learn of dan;erous, shock-sen,,itive materials and maythereby minimize accidents by observing careflness. Too, this generalord.er is thought to reveal commercial and military possibilities of a given

S - exp-osive as far as general handling is concerned.

Investigating the behavior of explosives to impact or shock usuallySi..-ivolves placing a small quantity of material on a firm base or anvil and

SI inserting through a guide ring a piston or striker which is brought to restS ato-?. the small charge of explosive. A weight ov hammer of known mass is

the= 3ermitted to fall under gravitational influence so as to - -use impact ontdhe striker-explosive combination. The necessary height All or drop-hei•ht to cause explosion becomes the characteristic evaiuation of the

- sezsitivity of the explosive. Most investigators use as the criterion themi=ýmurn drop-height needed to produce an explosion in at least 10 trials.Ma.y evaluations have been reported in past literature (1, Z ..... 20) withoutthe proper emphasis on a description of the method of testing the explosive

* ' and on the degree or intensity of exp!osions resulting from the listed drop-heih-.ts or impact energies. The method of testing is most important, as

Ss•..-,t variation in the construction of strikers and anvils will change theSeorders of sensitivity of explosives. Strong evidence of these phenomena

is seen from various designs described in this report.

U-3

I -- =IIT"--2-.A.


Other methods of ev,&uating sensitivity involve measuring theamount of gas produced during a given explosion. The Rotter machine isperhaps the most familiar in this realm. Such procedure is definitelymore scientific than previous evaluating methods, but has the disadvantageof being time consuming.

Sore investigators (Dr. W. S. Koski, Hercules Powder Company)

have attem.p:ed to measure the intensity of sound of explosion as acriterion for the amount of material exploding; however, interferingnoises make this approach difficult.

At Bruceton explosions are detected by auditory means, and it hasbecome the practice to identify and compare explosives by the drop-heightneeded to produce explosions in 50% of the trials. Also kept in mind arethe minimum drop-heigut to produce an explosion and the drop-height atwhich explosions occur in every trial.

The 50% explosion drop-height was chosen, as an abbreviated20-trial determination or "run" was develaped (ZL) in 1941 for which the501' explosion height was accurate. The procedure involved is discussedlater in this report under the section describing the No. 3 Bruceton design.The conventional procedure is to carry out a series (20 or more) c1 trialsat varioaxs drop-heights to obtain 0- 100% explosibility; which involves atleast 100-ZIO trials per explosive or sample. Because of the largenumber of samples at Bruceton and the limited supply of impact machines,it was essential that an abbreviated procedure be developed for routinedeterminations. The conventional procedure has likewise been employedat Bruceton whenever extensive studies were pursued.

O.S. R. D. Report No. 804 (ZI) discussed sensitivity studies withsome six Bruceton designs used at that time. By a design is meant agiven set of conditions in the form of the type of striker and anvil used.Since August, 1942, there have been developed methods designated as

* Designs No. 7 - 13. All of these latter designs are used with a largeimpact machine which his a maximum drop-height of 337 cm. or about11 feet with either a 2.5 or 5.0 kilogram weight. Designs No. 1-6 wereall used with smaller machines of 100 cm. maximum drop-height. Thepresent writing is to discuss the more important results obtained withdesigns 7 - 13 and additional (since 1942) dama fruzn designs 1, 3 and 5.

The inte rpreta:ion of sensitivity data obtained by the conventionalmethod of testing has produced some interesting and confusing aspects.

I zU-4

CONFIDENTIAL

7... .. .. Y "

V. J-,.

• .•-=

-441 ,CONFIDENTIAL

NAVORD Report 4236

We obviously know that the percent explosibility or the probability toexplode is a function of the drop-height, but doubt has existed as to theexact nature of this function. When probability to explode (synonyrnouswith % explosions or % explosibility) is plotted as a function of thelogari th'm or the drop-height, an elongated S-shaped curve results. Thisbecomes more prominent when a large tat least 1001 number of trials arecarried out for a given drop-height. Taylox and. Weal. (9116) attempted.] to treat the curve as a statistical distribution obeying the Maxwell-Boltzman distribution law. This was Hikewise thought by this author inthat -he. curves are similar to the integrated form of the Guassiannorf-'.1 error function and that probability is involved in sensitivitystudies. However, it seems that the curves are similar only in shape.

The relationship may be a linear one, with a general type equationof the for-m Sk = EA, where S is the drop-height, E the % explosibility orprobability to explode, K is the slope and A the theoretical E-intercept(when S-'O) on a log-log plot. A has a large negative value and posseshes

* no physical significance. It is needed only to give a specific equation foreach explosive; thus with knowledge of A and the slope, K, the theoreticalsensitivity curve can be drawn.

Salthough the theoretical relationship between E and S =ay be linear,the practical relationship becomes the complex S-shaped curve. An"equation to fit these curves is meaningless, as many trials are needed toobtain the exact shape of the tails of the curve. The tails are most likelycaused by the fact that the drop-hanmrer does not fall in the same mannerfor each trial. Identical hits or impacts on the striker are likewise notobtained !or each trail; and as a result, certain stress concentrationsfrom irregular impacts produce certain or doubtful explosions (depending

"I . upon the eitlosive) at drop-heights which theoretically should produce noexplosions, or likewise produce a failure to explode at drop-heights which

theoretaically should produce explosions in every trial. Even the slightestf t clearance, . 001 - .002", between the guide ring and the striker leaves a

region for mobility of the striker and produces irrequla: impacts.Unfortu-nately, such clearance is needed with a striker-guide ring design,to permit removal and insertion of the striker. Even a slight irregularityduring an impact process of this kind can cause enormous energy

S dissipation or concentration. Diagram* 1-4 give 2 few of the possibilities

during impact of this type which would cause deviations. Another factor,discussed later, is that the elastic properties of a given explosive vary,

which in turn causes deviations in the pressures produced during suchimpact processes.

21-SI 20CONFIDENTIAL

/

CONFIDENTIALNAVORD REPORT 4236

F

IR!NG

IUNGER

I" I; I I 1, ~ ~~It t

EXPLOSIVEIf.J "•ANVIL,,

i Azi( I) EXPLOSION AT UNEXPECTED

"IDEAL CONDITIONS DROP- HEIGHT

4 Ii

P H/ AI P/SI4 " / II//I / Ii

I I

mo3m 1 141.

DOUBTFUL EXPLOSION AT FAILURE AT UNEXPECTEDUNEXPECTED DROP- HEIGHT DROP- HEIGHT

A FEW OF THE POSSIBLE DEVIATIONS DURING AN IMPACTS"t "PROCESS SUCH AS IS PRESENT IN SENSITIVITY STUDIES.

] 21SCONFIOENT"IAL

A~• t . . . .' / " - •

CONFIDENTIAL

NAVORD REPORT 4236

tt

.V

(K

(; 4) (5) (6)

I J

I 7 181319

SHOWN ABOVE ARE 9 SUCCESSIVE CARSON PAPER IMPRINTS OF THE IMPACT

OF A SO CM. FALL OF A 2.5 KILOGRAM HAMMER ON A 11/4" OIAMErER STRIKER.THESE IMPRINTS SHOW THE BEHAVIOR OF THE HAMMER DURING IMPACT. ALSO

SHOWN ARE SUCCESSiVE IMPRINTS OF THE STRIKER-ANVIL. SURFACES FROM

THE IMPACT OF A 2.5 KILOGRAM HAMMER FALLINO 5OCM. THESE SERVE TO

ILLUSTRATE SLIGHT VARIATIONS AS THC COLOR INTENSITY IS SEEN TO VARY

S[ AMONG IMPRINTS

DROP-HAMMER ON STRIKER-. I r1-7

22CONFIOENTIAL

*4.

\ .4

'I &.-

~, I* *~ *

CONFIDENTIAL"NAVORD REPORT 4236

.y-Ky

1C

'00A (4) (5) (6)

i8 .9

(7) (8) (9)

STRIKER-ANVIL SURFACES

25

CONFIOENTIAL

low

i in li I i

CONFIDENTLUL.1 • NAVORD Report 4U36

Another approach to the interpretation of sensitivity results is to* plot the % explosions as a function of the logarithm of the drop-height on

probability graph paper. The tails ")f the S curve are here avoided and a,I linear curve results. The use of probability paper is based upon the

assumption that theoretically the S curves are asymptotic at the lower andupper ends, i. e., there is always the probability of an explosion as zero

Ir drop-height is approached and likewise there is always the probabiUtyof a non-explosion as infinite drop-height is approached.

e;

In certain portions of this writing sensitivity curves will be shown.These will be identified as the practical or actual S-shaped curves obtainedloy plotting the Is explosions or probability to explode (P.) as a functionof Lhe drop-height on semi -logarithmic graph paper; as a general theoreticallinear curve on a log-log plot of the same, allhough the opinion of late isthat the log-log plot is only an approximation. In reality this plot removesonly the lower tail of the S-curve and produces strange ordcre ofsensitivity at the X-axis intercept; however, these graphs are presented toillustrate this point of discussion.

Practical curves will be identified by the graphical value of the50% explosion drop-height and the slope of the curve at that particularpoint. "Theoretical" curves will in most cases be identifie'i by a generalequation of the form: log E = K log S-log A. where E is the % explosions,

j . S the drop-height, K the slope and A the negative E intercept when S=O.Log A and K will be calculated from known rounded values of E and S.

i In addition to these two types of plots, there will appear plots onprobability graph paper. These curves will be identified by the 50%explosion drop-height and the slope of the, curve. A large value of theslope indicates that the explosive in question requires a wide range ofdrop-height to produce <1 to >99% explosibility; while a small slope

* likewise indicates that the explosive is influenced greatly by a reasonablys mall range of height.

It will be observed that with these latter probability plots,

different orders of sensitivity may occur at the <116 end of the curve,i.e., the y-intercept. It must be remembered that such values are

* extrapolations and undoubtedly are beyond the accuracy of the impactmachine.

Slopes of the elongated S-curves will be determined by drawing atangent at the 50% explosion poilnt of the cut ee and measuring with a

"" II-9

• Z4CONVIDENTIAL4

, !,__ _ __ _ __ _ __ _.I S..


protractor the angle formed with the X-axis. The tangent (from tables)of this angle becomes the slope of the curve at the 50% explosiou point.

For the probability curves, th! slopes will be determined by the

familiar formula YZ - Yl' where X2 is .93

XL2- XI

(90%) and XI is . 10 (10%) in most cases. W'here a curve does not reach• • 90% explosions, appropriate %'s are chosen.

The Bruceton No. 3 Design

Since 0. S. R. D Report No. 804 (Z!) the No. 3 or brass cup designhas been adopted as one of the standard sensitivity tests at Bruceton. Thedesign is illustrated in Plate I. The anvil or base (A) consists of ahardened ketos steel cylinder of 1 1/4" diaeter and Z" height. Theplunger or striker tS) is L/2" diameter ketos drill rod which is machinedto a taper at one end to fit a brass cup of 0.308" l.d. The tapered end isground to 0. 336" diameter for the standard test. Plate 11 shows arotherview of the anvil (A-3), striker (S-3) and brass cups (C-3). Plate IIIpresents a side view of the striker-anvil holder device (H-Il). The same

* design is applicable to the large impact machine and the respective partsare seen in Plates II and UII. Plate II shows this holder with a designNo. I striker inserted.

The No. 3 design is ideal for the initiating and booster type ofexplosive, but is limited for materials of the TNT class. The maximumdrop-height is restricted to 100 cm., as t:e striker tips tend to bulge soas not to fit the cup. These tips likewise will develop cracked edges, whichlead to erratic results due to localized pinching and confinement.

Booster-type of explosives (RDX, Tetryl) are most affected by crackedstrikers, while initiators such as PETN, lead azide and lead styphnate

are unaffected.

For routine sensitivity determinations the so-called "BrucetonMethod" is used in reporting results. The principle was developed in] 1941 by Drs. G. H. Messerly and D. P. MacD~ugalI (21). A series of20 trials is carried out as follows: Assu=e that a trial registered E,or explosion, at 50 cm.; the procedure is to lower the drop-height in5 cm. increments until a failure to explode, or N, is obtained, then theheight is increased by 5 cm. until an explosion results, and so on.

II-0

*1 11-10• Zs

CONFIDENTIAL

I 77 ____

IIIP4

CONFIDENTIALNWVORD REPORT 4236

:12

. "

IIn

W - 5.0 KILOGRAM OROP-HAMMEN

I • R -- GUIDE ROD$ TO CONTROL PATH OF FALLING HAMME[R

S[C - CRANK TO RAISE AWO.LOWEA HAMMEIR-ANVIL- STRIKER HOLDER

M--IrLECTROMAGNETIC DEVICE TO HOLD "AMMER

" r'• f S -- STRIKER INSERTED IN BRqASS CUP""A -ANVIL OR BASE

SPLATE I

I -

Sl CONFIDENTIAL

-ii

"'" I I L CU I -I CR KTORAS AND --JWE HAMlE. ..

CONFIDE NTIALN4yORO REPORT 4236

IiI 11 Iz# 13

A-4 A-S A-26/, p / 26 ' C

ILC-3,6 Ile

II

5-1.4 s STRIKER FOR OESIONS NO. I, ,

5-3,2,6 0 " 3,2,s-S 31 • DESIGN NO. 3, LARGE IMPACT MAC;IINE

S-4

3-7

A-I , 3,7, Os. I't, it, 13 0 ANVIL FOR OESINS NwO., 3,?, 9. It, 12,13

-I A-tS s ANVIL FOR DESIGNS NO. log

A-4 a • 3 4

C-536 a tRASS CUPS FOR DESIGN NO.1

C-7 a COPPER CUPS FOR DESIGN NO.?

PLATE 11

STRIKERS AND ANVILS FOR VARIOUS DESIGNS

CONFIDENTIAL

- -. - b•S .-..

* CONFIDENTIALNAVORD REPORT 4236

E

H- HLEFODEINNO1, ,1ADUEDWT AG

IMAC MCINE EINN.I 6SON-n HODE FO SML IMATIC[ EINSN.10

H-!

2-13

Ct I - (J1

ISi ~H-lI • HOLDEIR FOR DESIGNS NO. I, 7, 6 AND USED WITH LARtGE

S~IMPACT MACHINE. DESIGN NO. I IS SHOWN.

' ~H-U, HOLDERt FOR SMALL IMPACT MACNINE DESISNS NO. I-6.

* SIDE VIE[W OF NO. S IS ILL.USTRtATEO.

• I PLATE TNI! ~ANVIL- STRIKER HOLDERS•

' b

2oSi €~ONFIDENTIAL l

4 *

L .... . .

CONFIDENTIAL

NAVORD Report 4236

Thus, Trial Hegh Result.

I 50 E

2 45

3 40 N

14 45S

5 40 N

6 45 N

7 50 E etc.

The 50% mark., or height at which explosions occu- in 50% of the

Strials, is used to identily explosives in routine determninatis. its. This' ~valuet is calculated as fcL~ows: Taking an actual deterzninati.,in or "rum•" for

' RDX:

* Trial Height Result Trial 1,eight Result

1 50 E 11 50 E

Z 45 N Iz 45 N

3 50 E 13 50 N

4 45 E 14 55 E

5 40 N 15 50 N

6 45 E 16 55 N

7 40 N 17 60 E

8 45 E is 55 N

! 9 40 N 19 60 E

10 45 N 20 55 E

Condensing actual trials:

Height ;-E

60 2 0 100

55 a 2 50

50 3 z 60

45 3 3 50

40 0 3 0

A , -14

CONFIDENTIAL

-1.,4 1....-

LL" , .•. ;.j2 .et

CONFID •N TIAL~, NA,'ORD Report 4Z36

If a trial registered E at 50 cm. it is assumed that under thesame conditions for that particular trial there would also have beenexplosions at any drop-height above 50 cm. Likewise if a trial at 50 cm.regittered failure, it is assumed that failures would have occurred at anyA height below 50 cm. Applying these principles to actual trials we obtain,

60 10 0 10055 8 z s0so 6 4 6045 3 7 3040 0 10 0

By inspection we observe that the 50%1 explosion height is between45 and 50 cm. To calculate the value in cm., to be subtracted from 50 cm.

* j ; or added to 45 cm., proportion is applied.

the height in cm. to be added to 45 cm. = 50%=E at 45 cm.

the increment of height between 45 and 50 cm. %'E at 50 cm. -j0E at 45 cm.

I •numerically equivalent to:

I X = 2..0 X a 3.3 and 5011- explosion height is 45+3.3 or 48.3 cm.

also,

the height in cm. to be subtracted from 50 cm. - %E at 50 cm. -50"the increment of height between 45 and 50 cm. %E at 50 cm. - %E at 45cm.

I numerically equivalent to:IIL.X a 0 , X--l. 7 and 50% explosion height is 50-1. 7 or 48.3 cm.

5 ro"

I With sensitive materials of 50% explosion heights in the order of6- 15 cm., the increment of height is A cm. instead of 5 cm. The

I increment amounts in a general way to 10% of the height.

• 30

CONFIDENTIAL

I Kr m a -7

CON FIDENTIALNAVORD Report 4Z36

A standard is a common explosive such as RDX, PETN, orTetryl which is tested daily to indicate the condition of the design forthat particular date. RDX is the most common standard substanceemployed for Design No. 3.

The 50,% explosion height represents tze most ar-curate portionof the curve ob:ained when probability to explode is plotted as a iunctionof drop-height in the case of a large number of trials at given drop-heightsfrom 0 to 1. 0 GOA% E) probability of E. The Bruceton Method isaccurate within t, 5- 10% for the 0. 5 probability height and serves as arapid means of evaluating explosives in a comparative sense. It mustbe remembered, however, that this time saving method is accurate forthe 5014 explosion drop-height only.

The No. 3 design was likewise employed to investigate a numberof common and newer explosives in the conventional procedure. Thesedata are summarized in Table I and treated graphically in Figures 1-6.

4 Significant data are listed on the graphs.

From experience it his been found that certain factors affectresults obtained with Design No. 3. Such variables as the diameter of

the striker tips, cracks deeloing along striker tip edges, the nature ofthe anvil surface, the nature of the container and the amount of explosiveplaced in the cups all affect rerults.

Conditions of the striker and anvil were varied in aninvestigation involving PETN and RDX. It was observed that strikers withcracked edges which also possessed a diameter of <0. 306" affected RDXmore thin PETN. A relatively sensitive substance such as PETN seems"so classed regardless of conditions, it would seem. Changing the diametercf the striker tip from normal 0. 306" to 0. 304-0. 300" deviated the RDX50% explosion height from normal 48 an. to 35 cm., while strikers withcracked tip edges and undersized diameters gave values in the order of30-35 cm. PETN was lowered, if any, about 2-4 cm. below normal 30 cm.Table U shows a summary of theme resmlts.

The weight of explosive tested was varied with the explosives RDX,PETN and Tetryl. Individually weighed charges were used for 20 trialsat each drop-height to obtain data presented in Table M. Abbreviated"runs" by Bruceton Method (ibid.) with PETN gave for 5, 10 and 20 mg.charges 50% explosion drop-heights of 19. Z, 17.5 and ZI.0 cm.respectively. o.mal values here with 30-35 mg. charges average about30 ,,-m.

1~ 3'CONFIDENTIAL

P 7& .*

I1.., i i l

1 CONFIDENrALNAIVOR REPORT 4236

7, DATA av CMVKiTIOkOAL 04ETUD0roo DESIGN No.)i

40 L004

44 4 0 iS 4 0 16 Is

so 4 is 44 45 0 Los0

is 9 4 Is I" O 0*

I' 0 ooo I it s0 0so 40 so 4 t 9t o

40 .20 *000 2- Is

'. 3 00 9 140 40 00. S 10 s 40 4.5 40 is so tTo 13__ _ at__ _ 65__ *070 *a0 In S4t __0 _ to _ __

* I00 4 0 i o )r0 to 4Sz 41 Go 40 476as Q t

Is *00j000 Is'* to *040 67's as $:1 ,L~ 0 so 10is - s :L o so Is *I S

vh1 __ -1__ 7--t.__ T~ TI T. *0la Z*0 *0 t~... E%C I-ý %C 00 %0 %c *0 1* i -

_s 6

*A 101

AS . 10 4t as

00 ___________so

-. 5 a I.? _

if -as as Ias ~~~ ~~ ~ ~ 4 Is1 sIss o oa

-1 s. D00pw~ a e 4.6 0.. 6- UM. N&S44

32

COWCM A


0 0

z - 4

to

aa

CL I.-a w w

.7 7 x i m j L Z g L C c - --- -!¶ z

~1L z

-LJ

0- 00-

I. xIm4

D 1 (30 x 0+

a~I *q Itt i0 0 0 0 8

(8d) 100-4X3 OL AflniVOU4

33

CONFIDENTIAL

CONcIDEHkMAL

NAVORUi REPORT 4236

coc

1.- A ,

.J ~ ~ . 'p~~ II c4*5 1 ZO 'o

-AinO U.A I q jW ZW..hh0 W W5

2*X +- ow

34

CONFIDENTIAL

.1 CONFIDENTIALNAVORD REPORT 4236

:7 '.77 Z: z

000I1 zt: 2, w

x0 c 0 v 4I 0:. 0

354a 4a WY ly V) CI

CONFIDENTIAL

- NAVORD REPORT 4236

LEEN

0LEAD STYPHNATE 19.6 ....

0 PETX 74 AfA-

40 GE'%=-AL EQUATIONS

-MERCURY FULMINATE,LOG E 3.4 LO S -2.92 -

1! _ ~~LEAD STYPMNATE F-LOGE* 54 LOGS-524

2U PETN- ...

::7:77:Z ~ -7!. 7:

__ Ft o

I 4 6 S Ic 20 40 60 so0100pDROP HEIGHT (S) OF A 2.0 XG. HAMMER

FIG. 4 THEORETICAL SENSITIVITIES BY DESIGN NO. 3

fl-21

CONFIOENTIAL

*t .*_ -S±.A


-~ -AS

. .... ' ...

7I :7:., 0--

w cm

.tt

cc 0

.0 in

U zf -7 : '.,-

w o ... o u) (1

g&Ld

2 =.. .... .... ... to-- ---- 1A*

-.. A..

z IL -7'7

-0, 0 S 0 0

a 77

Ngoo - -db'M N~ 9 .40 (113~) ±N1h314 dOme O01

- ~ ~ . - C O N F ID E N IhA L *x.

Li __

* CONFIDENTIAL

NAVORD REPORT 4236

.. . . ... -.. hi

ki k.4 oI -

Z: 7: ------ - -

.. ..i ...4.. .

... ... ... 0

0~ In

a~. ... ... 2::I.....i ..L [... .. .4..InI

DIDi~ -0 e4-x - j

ev 0 r- V 0 0*1 wO U ad 0 -0 lo - -0

.4 144 A........ ... .. .... .

____ ____ ___ ____ ___ ____ I - * '.d

CONFIDENTLA.LNAVORD Report 4Z36iti

Figures 7-9 show graphic treatment of data from Table 13.As the amount of material tested decreases for PETN, the materialrequires less energy for detonation; but as the charge weight exceeded30 mg.. a relatively constant drop-height was needed to give an explosionprobability of 0.5. With RDX, the 50% marks increased with the chargeweight; however, as the drop-height exceeded 70 cm., the probability ofexplosion appeared constant and the curves merge at 75 cm. The same1 behavior held with Tetryl, with merging at 90 cm. drop-height.

i]TABLE I.

SUMMARY OF 50% EXPLOSION HEIGHTS FOR PETN AND RDX UNDERJ 'IVARIED CONDITIONS WITH BRUCETON NO. 3 DESIGN

4 Substance: PETN Striker Diameter- Condition 0.306" 0.304" 0. 302" 0.300'

Striker Edges normal (slightly sanded) 37.7 31.Z 31.0 35.5normal procedure . 7. 5 27.1 34.2 ....25.0 30.0 34.Z 38.9

J

Striker edges sharp, no sanding 27.9 22. 5 28.3 31.327.5 32.0 37.7 33.2

27.S 23.7 28.0 29.0

q Striker edges normal, anvil rough 28.7 30.0 30.0 31.9

Striker edges sharp, anvil rough 32.9 29.0 29.0 30.0

Striker edges cracked, normal anvil 21.9 33.5 27.5 28.1

Striker edges cracked, rough anvil 21.9 27.5 27.1 37.5

Striker edges normal, 2 scoops PETN6(60 mg) 39.1 27.5 36.0 40.0

31.1 30.6 31.8 37.4

11-2439

S~C ONFIDE.N TILq4 :;- .m I ....

- A I II II II


TABLE It (cont'd)

SUMMARY OF 50% EXPLOSION HEIGHTS FOR PETN AND RDX UNDERVARIED CONDITIONS WITH BRUCeTON NO. 3 DESIGN

Substance: PETN Striker Diameter

Condition - 0.306" 0. 304" 0. 30Z" 0. 300"

Striker edges sharp, 2 scoops PETN 3Z. 1 30.6 45.0 32.336.2 29.2 36.2 38.8

Striker edges normal, Z scoops, roughanvil 34. Z 30.6 .... ....

Striker edges sharp and cracked.2 scoops, and rough anvil .... . 5 31.4 34.2 35.0

34. Z 28. 1 3G. 6 34.0

Substance: RDX

t.Normal Proc, iure 47.4 33.3 40.8 34.0

Striker edges sharp 38.1 36.8 45.0 39.1

, Striker edges sharp and cracked 37.5 30.6 32.3 34.0

f Normal striker, Z scoops RDX

(70 mg) 49.2 40.0 47.5 42.0

I'-I

40CONFIDENTIAL

. . ... . /.

CONF1DENTUL

HAVoRD Report 4236

1'~

IZ

0

Z(- 03 ,a,

0

LX 0

a 410-MM H .6

CONFIDENTIALNAVORqD REPORT 4236

400

40

_ILJ

>

C',000004-rw

- _ ~ ____ I +0I ___14_quo

0!0

I(d) 3001dX3 01 AJS1SUV*VOOd

nl-2741

CONFIDENTIAL


Iw

1*>

'I0

J V _ _ _ _

V -a:

I _________________________________________________________________J-4_

/ _ 4

r~~~T TrII'I

in

CCJNF'EPTIALNAVORD REPvjRT 4236

0 C

w..... .. 0 -

WIL

------- > -44_U)

-0r

t -1- 1-i 3

N U..

3 0 a(ad) 3001dX3 O1 AJ"IUWvoOld

11-2943

CONFIDENTIAL

L/2~~m pLLi~ WE d-. -..


Figure 10 shows the 50% explosion height as a function of thearnont of material tested for PETIN, RDX and Tetryl on a log-log plot.PETN data with a Z kilogram hamrrer am also shown and each point for thiscurve is the average of two Bruceton Method "runs: using the indicatedcharge weights. it is of interest to note that as the weight of chargeapproaches zero the 50% explosion height becomes independent of the massof the hanrnmer. This is in agreement with Koski and Lawrence (22).

Various mixtures of RDX and PETN were investigated by the

SNo. 3 Design and the 5056 explosion heights obta-ined. These data aresummarized in Table IV and graphed in Figure 11. with the 50% explosionheight as a function of the PETN content. The relationship is linear,as expected. The PET'N used at the time of this study posseazed very finecrystals and was more sensitive than more recent material. Thisphenomenon is likewise in agreement with Lawrence (ibid.) in that thefinely divided PETN consisting of agglomerates was found to be moresensitive than material of a coarser nature.

TABLE IV

SUMMARY OF 50% EXPLOSION HEIGHTS FOR RDX-PETN MIXTURES ASTESTED BY BRUCETON DESIGN N.O. 3

Ave. 5C. Explosion Ht.% PETN %RDX 'Run" I "Run" U1 Actual Rounded*

0 100 51.4 54.0 52.7 5310 90 43.8 50.0 46. 5 4720 80 40.0 40.8 40.4 4030 70 43.3 48.3 45.8 46"40 60 36.6 36.1 36.3 36so so 5 36.9 38.4• 37.6.8

60 40 21.9 40.9 31.4 3170 30 24.4 20.9 ZZ. 6 Z380 20 21.8 21.5 21.6 2290 10 21.9 25.6 23.7 24

100 0 17.9 16.5 17. Z 17

*Used in Plot

U1-3043a

CONFIDENTIAL

I p

CONFIDENTIAL

NAVORD IIEPORT 4236

L&..

Z o

o o o o0

LL

- % 40

SItT'

-44f's0

w'

qU)

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

CONFIDENTIAL

I7

i--

CONFIDENTIALNAVORO REPORT 4236

l;0::

.1

b Log ,z .......... 26

S50 ..." ... . ..... . ..

7 77"040 - Lo "~ ........ PET +172

,_ _... . _ • 30 _ _ _ _

* 'C _ _._ "=-..• .. _•._:. = • • • -30

I _' _-- __ . . _-___ _

0. tSI 4 - ---. --

30 10 2 30 40 50 60 7 80 so 90 00

J °% PET. IN MIXTURE

FIG. II 50% EXPLOSION HEIGHTS OF RDX- PETNMIXTURES AS TESTED BY DESIGN NO. 3

I

4 uCON i I OENTI l l Il I II 11

* -.•~ . . . . ...


The Bruceton No. I Design

Perhaps the most common method of testing the beha-vior ofexplosives to impact is this particular design. It consists essentially ofa 1/2" diameter leetos steel striker of 3" length with the anvil being the

same as the No. 3. Plates II (S-I, A-I) and III (H-I) illustrate tus design.The charge of explosive is merely centered on the anvil and the strikerplace l atop the charge tc be irn position for receiving impact. The designis fairly satisfactory for brisant explosives, but is not suitable formaterials uf soft, waxy consistency such as Tetryl, TNT, Composition B:Fivonite, Emmet and TNT mixtures. The main difficulty is the eacapeof the material during the impact process.

Table V and Figure 12 prese..- data by conventional procedure forthe behavior of RDX, PETN and EDNA as tested by Design No. 1. Theerratic behavior of even these fundamnenzai explosives is easily observedhere.

Recently the sensitivity of ritrocellulose samples of variednitrogen content was studied using design No. 1. These data are seen inTable VI in which the prcbabilitiee of e::plorion in 20 trials for differentdrop-heights are listed. From these data it is evident that sensitivityis not a function of nitrogen content when studied by Design No. 1.Several of these nitrocelluloses were bulky in nature and these erraticelastic properties undoubtedly contributed to results obtained.

TABLE V

SUMMARY OF DESIGN NO. I DATA BY CONVENTIONAL PROCEDUREAND LýARGE IMPACT MACHINE

Drop-Height EDNA EDNA* PETN RDX(Cm.) Trials _E Trials Tria P Trials.P

15 20 020 20 .175 Z0 .65 20 .20z5 20 03040 20 .55 20 .4550 40 .Z75

11-33"45a

4 i* CONFIDENTIAL

4 5.J- - -K!. .' .. . . .•"- .. -.

• • . , .,, •. •.• : .• . -,, .•:

/ *


TABLE V (cont'd)

SUMMARY OF DEXICGN NO. I DATA BY CONVENTIONAL, PROCEDURL.AND LARGE IMPACT MACHE

Drop-Height EDNA EDNA* PETN RDX(Cmn..I Trials Trials P. rials P T'rials Ps

60 20 .30 20 .6S zo .6570 40 .5580 ZO .90 20 .4590 60 .43

100 20 .60 20 .45120 20 .30 20 .35150 20 .20 ZO .35 20 .90 20 .85180 20 .30 20 .40200 20 1.00 20 1.00210 20 .25 20 .35240 20 .50270 20 .40300 20 .40330 20 .45

*Data with 2. S Kg. Hammer

*. CONF IDENTIAL

Il I_--Li j '"-q• .6

CONFIDENTIAL

NAVORD REPORT 4236

ilz~x

h- -: N

0 U04-u

wIL-

af

V. :7. 0.

z 4... ..

ZOFDETA

(L M *

* NAvORD Report 4Z36

:1 0

00

0o

I, IdE-4

0 000

z 11 il e 14 0ill;'44

040

0D-

04Ux 6IENM

- ... ..-- -


T"he No. 5 Design

During 1941-42 and part of 1943 a method designated as No. 5

was used to study explosives which failed to show a 50% explosion drop-height when tested by designs No. 1 and No. 3. The anvil (Plate II,

X. •labeled A-5) consisted of a 1 1/4" diameter ketos steel cylinder ofZ I Z" height with a central cavity or depression of 3/8" diameter and'a 1116" depth. The cavity served to contain a 30 rag. charge of explosive,which was in turn covered with one ply tin foil of 0. 0005" thickness. Atap-,red striker (Plate U, S-5) was then forced atop the explosive. TheII result was a snug fit, as the fail acted as a seal and gave a highconfinement. Beca-use of the localized confinement, this design provedcapable of detonating most explosives whether in solid or liquid forms.Plates IV and V illustrate the apparatus for testing explosives in moltenform. It was by t.s test that molten TNT showed appreciable

sensitiveness. O.her solid explosives were likewise found to be moresensitive as the melting point was approached and exceeded. These

interesting data are seen in Table VII. Figure 13 serves to show the50%. explosion drop-height of TNT as a function of temperature. From the-eraph it is seen that the 50% explosion height changes from 60 at room,!mperature to 16-17 cm. at the melting point of TNT.

Composition B, Pentolite (50/50) and TNT were recentlytested at varied temr-peratures with different sets of No. 5 strikers and

4 anvils. The striker-anvil depression clearances appeared identical,but results of Table VIII indicate differences. Notice the deviationbetween two runs of Pentolite on different auvils; and that TNT behavederratically at room temperature, as the usual 50% explosion drop-heightis 45-55 cm. Molten explosive values of Table VIII likewise are notin agreement with those of Table VII.

* t

11-37

444-

CONFIDENTIAL

_____ -.. . .. .. //--


I (

A CHARGE OF UNMELTED EXPLOSIVE IS REAOY TO 9S COVERED WITH ONE PLYOF TIN FOIL. C INDICATES CHARGE CENTERED IN ANVIL. IEPRESSION.

PLATE SZ

THE NO.5 DESIGN AS USED FOR EXPLOSIVES ON MOLTEN FORM

1-36If 49

CONFIDENTIAL,

S.... .. .• -••'•, ..


[0. . . - . -

.4 ~iiL ~ . -. , .£r

I1I11

Ivi.

t~ -T A

S a STRIKER

"W" 5.0 KILOGRAM DROP-HAMMER ATTACHED TO M OR ELECTROMAGNETt T s THERMOMETER

SSPsSMALL SPOON OR SCOOP USED TO MEASURE EACH CHARGE OF EXPLOSIVE

PLATE Nr

THE NO.5 DESIGN WITH STRIKER IN POSITION FOR IMPACT

50 "

CONFIDENTIAL -•

7T?7.. ev--two.

iv CONF1IDENTML

4' NAVO"W Report 4236

>1-

(A * I I >1 ;0

t~0-4

in00f% 0 U

44 :

0

Ad 0 0 ~ 0

01 05

U-40

50aC.ONFWEN~TL%

CONFIDENTIAL4, NAVORD REPORT 4236

-Z O

H-1

~Z 0

N~~~~~~~ NIZlII NI~d3O

5'ZCONFIENTIA

Li W


TABLE VIII

FURTHER DATA AT ELEVATED TEMPERATURES WITH DESIGN NO. 5(Values are 50% Explosion Heights in Cm)

Explosive ZO-25 0 C 95. 1000 C 108aC 121 0 C

Composition B 26.0 Z3. 4 25. 8

Composition B 28. 6

50/50 Pentolita 31.7 5Z. 6

SO/bO Pentolite 32.1 10.0

TNT >90 20.8 17.0

Liquid TNT was studied more extensively by the conventionalprocedure of numerous trials at each drop-height using design S.Comparative results with 80/20 nitroglycerin-dimethyl phthalate areshown in Table IX and Figure 14. The nitroglycerin-phtha~late mixtureproduced complete detonations in most cases, while the TNT explosionswere of a partial nature.

Attempts to explode liqu~d TNT by design No. 3 indicated thematerial was not sensitive. Ten trials at 10-90 cm. (10 cm. increments)

were for each height all failures. Much of the liquid was squeezed outaround the striker during the impact process, which was most likelya factor in that no explosions resulted.

11-425Z

CONFIDENTIAL6

'Z WT"- . ....

CONFIDFEN TIALNAVORD Report 4236

TABLE IX

COMPARATIVE DATA BY DESIGN NO. 5 FOR TNT AT 85-90°CAND 80 / 20 NITROGLYCERII- DWM ETHYL PHTHALATE

AT ROOM TEMPERATURE

i Drop-Height of Molten TNT 80/20 NG-Phthalate'4 5 Kg. Hammer Trials Ave. Pe Trials Ave. P

4 20 0

6 20 .OZ5

8 40 .237 20 0

to 40 .325 40 .187

15 40 .537 40 .625

20 60 .40 40 1.0

30 20 .475

J 35 20 .30

40 20 .70

.1

11-43

S3CONFDENTLAL/


- No

w'Zz

-w

d* -

000

_ _ _ P.)

- ~ d o d d d Al(d) 30O~~~~dX3 01. AII9S.d3VMii ~1 t >

CONIDETIA

* w

*1~~~~~_____


A study of the behavior of Cordite and Ballistite to impact byDesigns No. 1 and No. 5 was carried out in September, 1942. It wasfound that both propellants became sensitive when the particle sizeapproached thin shavings in form. When tested at elevated temnperaturesL. the No. 5 design, Cordite was seen to explode furiously at very lowdrop-beights. The explosions were actually near-detonations in intensity.Table X shows 5016 explosion heights by Design No. I as a function ofparticl.! size, while Table XI indicates drop-heights for explosion

S S probaoility of 0. 5 for Cordite as tesLed by design No. 5 a.t elevatedtemnperatures. The heated Cordite was in the form of single pieces

1/4 x 114 x 1/16" in dimensions. A 5. 0 kilogram drop-hammer was usedthroughout these studies.

TABLE X

THE EFFECT OF THE PARTICLE NATURE ON THE SENSITIVITYOF CORDITES AND BALLISTITE, AS STUDIED BY DESIGN NO. I

Nature of British Type British Rocket U. S. NavyMaterial Cordite Cordite Ballistite

Sing!e piece,1/4 x I14 x 1/16" 53.0 52.6 29.2

8-10 pieces,3xZx I mm. 40.00* 26.1 17.4

Shavings (30-40)

I x I x I nim. 37. 5** 18.1

*600C

**75 0C

11-4554&

CONFIDENTIAL4

*. .~

CONFIDENTIALNAVORD Report 4;.36

TABLE XI

THE EFFECT 0r ELEVATED TEMPERATURES ON THE BEHAVIOR OFBRITISH CORDITE TO IMPACT

Temp. Design Design D--sign Re marks"°C. No. 5 No.5* *o. _

. 15b.8 Z0.7 53.0 *No tin foil60 10.8 12.6 40.0 covering75 !0.0 13.6 37.5140 10.0 10.8 **Ballistite175 3.3 3.5175 Z. Z**

By experience, it was found that the clearance between thestriker and the anvil depress~on was most important for the No. 5 design.Results were irreproducible unless 0.001" clearance could be maintainedduring testing. Maintenance of such clearances proved to be very diffictutand the design was practically discontinued in L943.

During the zouzse of two years, many explosives were examinedas to impact behavior. Table XII lists in sunmraary form the 50% explosiondrop-heights for common and newer substances as investigated bydesigns No. 3 and No. 5. Unless indicated, the values are the result of

* one "run" by the Bruceton Method.

During the period October, 194Z until Ma-,P, 1943, time wasdevoted in search of a method of impact testing which would be suitablefor the study of the TNT class of explosives. This period saw thedevelopment and discontinuation of designs 7, 8, 9 and 10. These andsubsequent designs (1 1, 12, 1 3) involved the %!se of the large impactmachine at Bruceton.

! 11-46

* 1 55

* ICONFIDENTIAL

2' "'.'

*NAVORD Reprt 423

4 TABLE XII

COMPARATIVE SENSITIVITIES (50% EXPLOSION DROP-HEIGHTS) BY_____ ____DES!GNSNO. 3 and NO. 5 ______

Chemical Nan.e Comm onSof Explosive Desigr~ation No. 3 No. 5

1 Lactitol Decanitrate 1.5 -

Mannito! Hexanitrate NitromannlteDulcitol Hexanitrate Nitrodulcite It -

'ILead Trinitroresorcinate Lead Styphnate 11 -

StarcE- litrate Nitrostarch 12 -

1 -Guarxyl -4- nitro samino-guanyltetrazene Tetracene 12 -

Blasting Gelatin 16 -

Lead Azjide~ tr. Lead Axide 19

Tntrtrarte* 22N z

113(N-nitroguanyl)ethyl

Nitrate_/ 24 27Diethanolnitraniine DINA 27-I Pentaerythritol Tetra-

nitrate PE TN 29 -

TetrarnethylolcyclohexaziolPentanitrate Fi'vo~lie -9

Ethyl enedinitrarnine EDNA 30-40 35

* . Cyclotriznethylene Tri-nitramine Cyclonite, Canadian

Erithritol Tetranitrats ETN 38 -

P(2, 4, 6-trinitrophenylo-* nitrami~ne)Ethyl

Nitrate Pentryl 4Z -

* Dinitroxyethylnitroxainide NENO 42 -

Cyclotrimethylene Tri-

f nitramine Cyclonite, U.S. -48 Z3British

1U-4756

CONFIDENTIALL

CONFIDENTIA

NAVORD Report 4Z36

TABLE X1I (cont'd)

COMPARATIVE SENSITIVITIES 150% EXPLOSION DROP-HEIGHTS) BYDESIGNS ZNO. 3 and NO. 5

Chemical Name Common.of oExplosive Desig~nation No. 3 No. 5

-Uminonium Perchlorate Ammnirorum Perchlorate 55 46Z.,4, 6-Trinitrophenyl-

methylnitrarnine Tetryl 56 20II 2,6-Dinitro-2, 6 bishydroxy-

methyl-11, 7 heptanediolfl tet ranitrate 82--

Baranal 84 -

Torpex-11 86-Torpex-I 88s

Ammronium Pic rate Explosive D 80 19*2, 4, 6- Trinitrophenol Picric Acid >90 22

N -nitro -n- Methyl'hydroxy-acetarnide Nitrate Hyman >90 Z4-Methyl 3, 3, 3 Trinitro-butane -- >90 Z4

Tetramethylolcyclo-hexanone Tetranitrate Sixonite >90 29

Z, 4, 6 Trinitrobenzene ThE >90 Z9r 2-Methyl-Z-'litro- 1, 3-

Propanediol Dinitrate Nitroisobutylglycol >90 37Dinit rate

* I Tetrarnethylolcyclo-pentat.one Tetranitrate Fivonite >90 38

N. N' -dimne thyl-N, N'dinitro-oxyanaide MNO >90 39

3-Methyl 2, 2, 3 Trinitro-pentane - >90 46

,4, 6 Trinitrotol- t. TINT >90 48

&ablinied TNT >90 50Ethylenediamrraae O)initrate -- >90 522, 4, (6 Trinitra',?- sole TMA >90 552,4 Dinitrobet - ne DNB >90 58

2, 4 Dinitrotoic, it DNT >90 58

CONFIDENTLAL.

I' ~~~ ~ ........... _________


I TABLE XII (cont'd)

COMPARATIVE SENSITIVITIES !5O% EXPLOSION DROP-HEIGHTS) BY

I - ~~~~DESIGNS NO. 3 and NO. 5 _______

Chemical Name Commonof Exolosive ____De signation No. 3 No. 5

11, ~Ethyit rint ethyl olme thaneTrinit rate Emnmet >90 58

Z, 3. 3 Trinitro- 2, Z Methyl.I'butane - >90 70

2, 4 Dinitroaniline DNA >90 85Urea Nitrate Urea Nitrate >90 85

~,4 Dinitrophenol DNP >90 97Nitrourea Nitrourea >90 >1OOHexamethylene tetrAmine

Dinitrate Hexamine Dinitrate >90 >100* 3-Xthyl-Z, Z, 3 Trinitro-

pentane -- >90 - -

Nitromethane Nitrornethane -- sot Tetranitromrethane Tetranitromethans - 83Diethyleneglycol Dinitrate* DEGN -- Z-3Glyceryl Trinitrate* Nitroglycerin -- 3.0

Blasting Gelatin -- 3. ZSorbitol Hexanitrate* Nitrosorbitan -- 7Marinitol I-Hexanitrate* Nitromannite -- 9Glyceryl, Lactate Tri-

nitrate GLTN -- 9Butine-Z diol- 1, 4 Di-

nitrate -- 9Methylt rirnethyloirrethane

Trinitrate Memmet -- 17

7iirpeo CONFIDENT0IAL0

NN-iirppeaie>0 >0------------KiormHme

Aglmeae PETX

CONFID ENTIALNAVORD Report 4236

The N•. 7 Vesi!

In principle, this design was a modification of the No. 3. Sincestrikers with the No. 3 showed a tendency to bulge and dev-elop crackededges from d-op-heights exceeding 90 cm. with a 5.0 kilogram drop-hammer, it was reasoned that a larger striker with larger cups wouldelirn-.-.ate this difficulty. A LIZ" diameter ketos drill rod tapered to

0 0. 445 + 0. 001" to fit a copper cup was the result. Plate II illustrates thestriker (S-7), copper cups (C-7) and anvil (A-7) used in this design.Experience proved that copper was the improper material for the cups,as it tended to flow from the impact shock. Explosives possessing softcrystals were able to escape most of the impact energy by creeping intothe space between the cup wall and striker. Results in general proved tobe er:atic and were definitely a function of the amount of explosive testedin addction to the diameter of the striker. The design was discontinuedfor these reasons in Feburary, 1943. Table XIII shows a t-_ummary of50% explosion drop-heights of several Bruceton runs on common explosives.A few runs were made with one ply tin foil covering the sample, and theseare Likewise shown in Table XIUI.

It is possible that a heavy walled steel cup fitting a 112" diameterstriker may eliminate the main disadvantage of the No. 7 design. Anotherpossibility is a small metal ,disc covering the explosive which in turnrests atop a similar disc of abrasive paper. Investigations along thisline have not as yet been pursued.

The N., 8 Design

JJ Design 8 was a combination of designs No. I and No. 3. Anex,-3losive-containing brass cup was centered under a L/Z" diameterstriker and the drop-hammer released to smash the cup. This procedurewas seen to detonate, with a loud r, Viort, even the TNT class of explosives.Reautis appeared reproducible car' n the study, but deviations did occuras more data were obtained. Var ýis in the hardness of the brass cupsand i--n the technique of centering a 1ad d1 cup under the striker werefactors causing deviations. Tables XIV and XV summarize the 50% explosiondrop-heights of common explosives examined. Note that the order ofsensitiveness is altered when a 2. 5 kilogram drop-hammer is used. Thedesign saw usage during October - December, 1942, and was thendiscc.tinued.

11-5059

CONFIDENTIAL

Za

!A

C ON F ID £-NTiAL

NAVOR.D p.eport 4236

a Ca

-Ii %0 emf . 0.

no b. 40 - -W a '

AA

0*

Hz

>0C

6' 0

z go-

ini o

14 -4 -4040I45 a0

tie Zc 000

-l~~~ v. -0D 0 ' v1vi 44'

.0

- -beN

s 60

140 144,14

N

COMMIDNTIALNAYOPLD IRzport 4136,

it3t

0 lid

.'i# a,

0~

z '4

I- K11 t a 4 Z .

al 0

.0 0 '

0 1o p oll.

400

* a~

0.1

rii ITU

CONFIDENTIAL* NAVORD Report 4Z36

,I

The No. 9 Design

I' This method was studied during February-May, 194Z.Conditions were essentially those of a modified No. 8 design. Thedifference was that a steel cup, similar in shape and dimensions to thebrass cups, was flattened into the form oi a disc prior to using. TheI disc was prepared by resting a LIZ" striker atop the steel cup, andbmashing the cup by a 105 cm. drop of the 5.0 kilogram hammer. Theresulting disc possessed a crater into which a small charge of explosive

) could be placed. It was thought that if explosive could be entrapped(during impact) under the overlapping edge of the disc, a good detonationshould result from exceptional confinement. It was also reasoned thatthe dissipati-on of energy, as in design 8, should be reduced materiallyas most of the smashing was completed in preparation of the discs.

Unfortunately, observations indicated that the prepared discswere not uniform in thickness or depression depth. Twenty discs were

measured with a micrometer for thickness and crater depth both before* and after a run with TNT as the explosive. These measurements are listed

in summary form in Table XVI, where deviations are easily seen,particularly in column 4. Table XVII summarizes 50% explosion drop-heights on explosives studied.

Lead azide and lead styphnate displayed very peculiar behaviorwhen tested by the No. 9 design. Both substances are normally quitesensitive, but appeared decidedly insensitive by method 9. Thesematerials tended to be compressed into a disc of the exact shape as theIi crater w.ithin the steel disc, and were able to escape most of the impactenergy. Mobility of the explosive to the confining space under the edgesof the disc did not occur with these salts on trials registering failure toexplode. The likely reason for this behavior is strange elastic propertiesof these particular substances. The design was discontinued in May, L94Zbecause of uncontrollable deviations of the steel discs.

11-5362

k CONFIDENTL4,L

4


TABLE XVI

SUMMARY OF STEEL DISC MEASUREMENTS IN CONNECTION WITHDESIGN 9

(Dimension values are X10'• in.)

U4.3

'V. 6. . 3. ..-. 51 0 4

4A .,16

a. .

S 0 6 140. .. 0

0

25.9 16.8 9.1 5.4 1.0 -4.4 205 E

24.9 16.4 8.5 3.1 1.6 -1.5 190 El

25.1 19.3 5.8 3.9 4.5 +0.6 17S E

25.0 19.7 5.3 2.8 4.9 +2.1 160 N

25.7 20. 6 5. 1 4.5 5.4 +0.9 17S24.3 17.9 6.4 3.1 3.7 +0.6 190 N

Z5.7 19.9 5.8 4.1 4.7T +0.6 205 N

-75.6 16.5 9.1 2.8 1.7 -1. 1 220 E25.3 18.9 6.4 3.3 4.1 +0.8 205 E

25.3 17.5 7.8 3.1 3.5 +0.4 190 El

25.9 21.2 4.7 5.9 5.8 -0.1 175S25.7 18.4 7.3 4i.7 3.Z -1.5 190 NZ4. 0 16.6 7.4 3.2 1.7T -1.5 205 E25.2 17.3 8.1 3.0 2.5 -0.5 190 E26.0 19.3 6.7 4.2 5.9 +1.7 175 _25.0 17.6 7.4 Z.0 2.8 +0.8 160 N

26.9 18.4 8.5 6.5 2.0 -4.5 175 E1Z6.5 ZO.6 5.9 4.3 4.1 -0.z 160 N

26.4 16.9 9.5 3.2 1.1 -2.1 175 N27.1 18.3 8.8 5.3 3.1 -2. 2 190 E

['5463

CONFIDENTIAL

I. I ___ _ _ _ _ _

- j ._

~~ I CONFZDE-NVALF ~NAVORD Reaport 4136

.4fIV 10 09 .0 40

z CF

0@

N 'ILI 0.9

.4 .4.

j~. (1 09 0 01 .

) -~ .. 01

00

- --- - -.' - -- 9 -1. -

z .4.

1-. o z xIS ::1 5 a3,~

a N z 1 . A .u . .uI

, :*4 -

I: CONFIDENTIALNAVORD Report 4116

pI The No. 10 Design

This method was used briefly during May, 194Z. Conditions were

Sa modification of Method No. 1. Instead of a I/2" diameter striker, the2. 5 kilogram hammer itself became the striker. As seen in Plate VI,

jthe striking area of the hammer was in reality a press fitted ketos steel

anvil of 1 1/4" diameter surface. The 1 14" diameter gave the needed

I,.sur.ace to prevent the escape of the explosive, as was the disadvantage

0 SwitN design No. I. Another reason for the use of the hammer as striker

was that the velocity of the hammer at impact is greater. than that of an

inserted striker, unless the weight of the striker is negligible in

'cj ci-marison with the drop-harwmer. Witha L/Z" striker of weight of

65 g;rns, the .elocity difference is negligible, altrough present; however,wia. larger strikers weighing 500 grams such as the type for design l1b,

this effect becomes more prominent. To insure the absence of the velocity

df:'erence was the other reason in mind while using the hammer as the

striker.

Plate VI shows a general view of design 10. The charge of

explosive was merely centered atop the elongated anvil and the hammerreleased from desired drop-heights. In some instances, liquid explosives

I in particular, the charge was covered with a single ply of 0.0005"(thicl'ess) tin foil to prevent to a considerable degree the escape of the.uil during impact. A single drop from a common medicine dropper

bec•-ne the sample for liquid explosives. This amounted to 40-50 mg. of

material, in the case of nitroglycerin. Explosions produced with liquids

were most violent in nature, and the operator was compellcd to use cotton

ear plugs to endure the sound intensity.

"Solid explosives were observed to behave erratically in the No. 10

test. The flushness of the impacting surfaces was the most importantvariable. A carbon paper imprint of the impact served to check the

condition of the striking surfaces. Unsatisfactory surfaces were correctedby refacing in a lathe with grinding wheel attachment; however, this

proved to be awkward and time consuming and it was decided to discontinueusing the hammer as a striker.

Several solids were tested in the presence of sand blasted surfaces,

but resul^.ing explosions soon removed the grit effect and smooth, slippery

surfaces again were the conditions. Table XVIII presents data from

method 10.

U-5665

CONFIENTU.L

Iw

"CONFIDENTIAL

"NAVORD REPORT 4236

FE[ ..--.

J,

SW, s 2. KILOGRAM DROP- MAMMelt

FE - ELECTROMAGNETIC HOLDER FOR W#EIGHT

As ELCNGATErD ANVIL WITH PIN TO PREVENT ESCAPE OF

ANVIL. DURING IMPACT

PLATE Z

GENERAL VIEW OF DESIGN NO. iO

; ! ,•CONFIDENTIAL

i. . "

CONFIDENTIALNAVORD Report '(.L36

uU

z go

z '-'-

40 >n S

'4 inin

00

0 ooc 0 v

> 04 ~ to'dCU)*

0A

Go C

-4 0 (,C

0 >

F44

o0

0-*

0 to00 .0 04

1001 0 0U 0 a

0 o

tc% 4 OJ4 5.

67CONFIDENTILk

//

CONFIDENTIAULNAVORD Report 4236

IThe No. I1 Design

The period May - Augnst, 1943 resulted in the development of theelev-.:th in a series of Bruceton impact methods. The procedurepresented a new approach to the sensitivity problem in that the explosivewas rested upon a I/f2" square of abrasive surface before being subjectedto impact. The abrasive surface was in the form of a square of00 specification Armour flint paper, which has the following screenanalyses as supplied by the Armour Sandpaper Company, Chicago,illinois:

Mes zScr en e__ngs Size of OpeningsM Lesh -)f Screen %X 10" in. (64

* on 125 mesh 6 42 107

on 157 mesh 76 33 84

Through 157 mesh 18 33 84

Through 180 mesh 5 25 63

The idea of the No. I I design was suggested byDr. D. P. MacDougaLl in hopes of obtaining reliable data for comparinga series of EDNA samples. EDNA was erratic in behavior withidesigns No. I and No. 3, and such a test as No. 1! was definitely needed."Design 11 was the same as Ao. I except that the sample was placed on a11/2" square of abrasive paper before the striker was placed atop thecharge. The hammer for most No. II data had a mass of 2. 5 kilograms.

The design proved to be satisfactory for its original purpose ofcomparing EDNA samples. Results were reproducible in general, as anew square of flint paper was used in each trial to present a nearly constantsurface. Strikers and anvils deteriorated in time and needed occasioualSresurfacing or refacing. The anvils showed more deterioration as a

1/2" diameter circular area possessing scratches and pits becameapparent After 500 trials. Strikers with scratched suriaces dil notaffect results, however.

Physical difficulties were encountered with the No. I I techniquein the case of drop-heights >100 cm. for explosives of soft. waxy

CONTIDINTALa ?

i ;.

\ V-

///"

CONFIDENTIAL.NAVORD Report 4Z36

consisteicy. Such materials as TNT, Composition A, Composition B, and

50/50 Ednatol behave normally until the drop-heigtt becomes >100 cm. Ata, these heights, e: plosions are difficult to identify and !00%- explosiun points

are not obtained during the conventional method of testing.

There are several possible reasons for this behavior. It was

thought, at first, that the heat produced from the impact of stcel on flint

crystals was inelting TNT (the explosive which first displayed the above

effect); which in turn was forced to flow past the flint crystals and becomeabsorbed in týie piper base and form an insensitive TNT-paper mixture.

This hypothesis was supported by the following: for all trials observed asfailure to explode, the middle porticn (under the explosive) of the flintpaper was driven firmly against the anvil and had to be scraped off beforeproceeding to the next trial. The appearance was as if the charge had

melted and refroze in a short time interval. To reproduce the appearanceoi these trials, the usual square of flint paper was made wet with water

1' and then subjected to impact. This indicated tiat a liquid had formed in thecase of TNT.

* 4* While investigating British Composition A, the escaping tendency

Swas again observed. The heat of impact was no doubt great enough to melt

a the wax present, b-at likely not >200°C to melt the RDX. Hvl the melted* wax penetrated the r-at-r base, pure RDX would have remained and

violent eetonaticns should have occurez~d. The heat effect :s real, asduring an examination of arnmoniurn picrate, a failure to explkde at150 cm. was observe-d anc. the unmelted, compressed charge was hot totouch upon rapid removal of the flint paper square from the f.ring chamber.

Arnother possible explanation in the case of both TNT andComposition A is that the mrnpact pressure of >100 cm. drops pulverizes

* the flint crystals to such a fine stat- of subdivision that mixed with initially

insensitive explosives, this mixture is rnore inert The pulverized silica:nay form gaps between explosive particles and in turn inhibit the chain

* reaction.

Particie size affects results in certain cases with design 11. The

effect was first noticed while investigating PETN. Three portions of a

sample were exanined, namely, the "as received" material, the screened

fractions on 100 mesh and through 100 theseh on 200 mesh. It was observei

that the "thrs 100 on 200" fraction was about 1. 7 times (50% explosion

point) as inse'nsitive as the other two portions. This effect was real, as

100 trials rer t.rop-height were carried out. These data are shown inTable XIX and Figure 15. The graphs show nicely the practical shape

of sensitivity curves.

i ' 69CONFID •N.IAL

-

CONMEPNTIALNAVORD Report 4236

'I e 40 C P I

>4

Nr~ 0

P0 41 .4

z .<

w W c ZI F0

-r-~ 3000

04 I

07>O ZENA

~z

CON FIDEN TIALNAVORD REPORT 4236

Li-L .

2w~~ 0

I--

0* 0

00 71

MA Z


It was thought that crystals of the less sensitive fraction wereforced or packed around the flint crystals to escape much of the impactenergy. Impact pressures from 8-16 cm. drops were not great enoughto pulverize, to much degree, the flnt crystals and friction elfects aremuch less than with drops of 25-30 crn. Particles screcmd through 100mesh are of the order 75-145ýa (Ll0, average) in diameters, while theflint particles themselves average 7S5L. A compariaon of particle sizehere would indicate that the explosive was encountering friction and thatour hypothesis concerning crystal packing is incorrect; however, sincePETN must be screened while water wet and be brushed or washedthrough the screen meshes, there is a strong tendency for sharp,irregular edges to be removed and decrease the sensitivity to frictionaleffects; whereas crystals of the more sensitive portions are apt to beirregular in shape and surface and in turn be more sensitive to thesefriction effects. Too, the I 0•i diameter particles are able to be packedinto the spaces between the flint crystals, as these spaces are of theorder . 08 - .20 mm. (80-Z00j) in width. The packed crystals act as acushion for crystals above and thus minimize friction effects. Crystalsfrom the other portions are unable to pack into spaces as their diametersare >150g and consequently receive more frictional effects and explodewith greater ease.

At a later date the less sensitive fracIxuJ was retested usingflint papers of 2/0, 3/0 and 5/0 specificatione. Flint particles of3/0 specifications are of the order 58-8ZtL diameters while 5/0 are from30-60A in diameter. The intercrystal spaces approximate 40-801L for5/0 paper. The choice of a finer grained flint paper was an attempt toeliminate the packing effect. The effect of the finer papers was comparedwith 20 trials at 12 cm. for each type and using the "on 200 mesh"portion of the PETN. The probability to explode was for 2/0, 3/0 and5/0 respectively . 55, . 55 and .90. This was concerning as there wereno explosions in 100 trials at 12 cr. in the case of the previous work with2/0 paper. The different behavior was attributed to Larger crystals beingtested. The sample which exploded with a probability of . 55 was theremainder of that used for the I00 trial/height study, and it was reasonedthat larger crystals, i.e., the upper limit of the 75-145 particles, werepresent in the sample bottle. Since. only 1-Z grams of the original10-15 grams was remaining, it seemns logical that larger, heaviercrystals of 145 is diameters would settie to the bottom of the samplebottle. The use of a finer paper eliminated to & greater degree theparticle size effect, as seen in Table XX. More extensive comparisonof 2/0 and 5/0 paper with PETN, as received, is shown in Table XXI and

U--63T7a

CONFIDENTIAL

|7

CONFIDENTIAL

NAVORD Report 4236

Figure 16. These latter data were obtained on the small impact.machine. It is of interest to note that 50% explosion heights are similarin magnitude for 2/0 and 5/0 flint papers.

Recent data with colloid milled and screened PETN indicate theseto be more sensitive than as received material. Figures 17-19 show theprobability graph paper plots of these data from Table XXI. This mostrecent work tends to reverse the earlier findings and the screened PETWwas seen to be the mnost sensitive. This leaves dow..bt z3 to the explanation

of earlier .henromena. There is one other explanation which has not beenmentio-ed. It is possible that the 1/2" s:riker was pitted and that the

screened PETN was not receiving the same impact as the as-receivedsample. This is only a hypothesis, as the striker for that original workhas since been refaced; however. RDX and TNT data agree with recentfindings with PETN.

It is of interest to note (recent data) that as the PETN crystalsbecome finer, the material becomes independent of the size of flint paperusedj and practically identical 50% explosion heights result. The greatestdifferences between 2/0 and 510 papers are seen to occur at 10 and <10 cm.drop-heights (see Table XXI).

Plate VII shows a photomicrograph of 2/0 flint paper under amagnification of 12. 5. The circular area represents about 4 mm. ofsurface. Notice the inter-crystal spaces into which small crystals couldpack. The spaces appear as black areas at the base of the particles and

* vary in diameter from 1-3 mm. which correcting for magnification givesS .08 - . 20 mm. diameters.

Plate VIII shows a similar photomicrograph of the 5/0 flint paper.Notice that inter-crystal spaces are much reduced in magnitude.

Plate IX shows the surface of 2/0 flint paper following the impactof a Z. 5 kilogram hammer falling from 12 cm. Plate X shows another2/0 surface following an impact frcmi a 150 cm. drop. The pulverizingof the flint is negligible in Plate IX, while in Plate X the pulverizedsilica could well mix with insensitive explosive during impact tq inturn, form a still less sensitive mixture.

1.SHU-64

CONFIDENTIAL

Innnni - - l-II -I - lll II-ll Ilpn

S.... ...-.--.-,--I


r h.;,].... .!I

PLATE 3Z PLATE2/0 FLINT PAPER 5/0 FLINT PAPER

12.5 MAGNIFICATION 12.5 MAGNIFICATION

• . .' .. • .,. '

o: .Y.,,

PLATE IX PLATE X

PULVERIZED 2/0 SILICA PULVERIZED 2/0 SILICA

FROM I. CM. DROP FROM 150 CM. DROP

OF 2.5 KG. HAMMER OF 2.5 KG. HAMMER

i--65'73

CONFIDENTIAL

"1: r

CON FIDENTIAL

NAVORD Report 4Z36

.-. ur- U

4 'a 0%a-4

"f 0d

U nC.0

-4

00

141 0

04 .4~5 1

>4 >

E-44 oW~~t

04 0.0

(* 4 H

N [a

~* II4"CONonDEt-

1-0 w

'1 - >in S 0 F

9-----0


0w

- 2-

U-

a.

I-J(

f::. U. W

~- 67

w ----- I. I.' __ __ __.

CIL~~ IL a.

CONE iOENTfALNAVORD REPORT 4236

1 ~ ~ .. .... ....

i... .... . . ....-. t. . ... .... . ..

+1iZ J i.;.-1 ---:-.,.---i!: rii r i -! i Ir _:'.- _ <,,. - • .- . , -- _

- .... .> ,

:- -== = = = == = = = = :: : :: : ::: :: :: : :: : : :: : : :: :

'-:

_ _ ..> r-7: .i

oh-

... o. z

0 . 9- .z::A 0-

o .... ... . . .. ., , ; . :. :i3..

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

~~~~~. ..... .... -- ,....... ..... :............* C•.,_S~ ~~~............o.,..............................•. ...

•, . v , . .. . . .: i. . . . , ... . L'LI: i:: ;::l: :

0-.-

........ . . . . w

* 4g(

ow.-,,r - a 4 4

• • cONFIroEPrIAL,

7- - ... ....

II.l Ia -

CONFIDENTIALNAVORO REPORT 42 36

Ia

w_ _ _

4. i ... .. -. ... .

. . ... . . . • .....

.. . .. ... .. i

_. _____=_====__= .==o== == = == ==== === = "4-,W

-) :.1 v:: I~; i:>:::• ..... _

- : :::: ::2"

z

,:,1 ", .!1. ]! ;:I:I- \lr.:iii

:t:0 td

M~f3~vH1)Y 9g 9 10O IP53; J.LHOlH-daato 901

=-6_9E-0

77

CONFIDENTIAL

0 CLA

-9 at.. .

CONF' ErTIALNAVORD REPORT 4236

t a

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

-I- 7.+ ...' • • ; .~~~~~...L I.. - .... ! .... .... .... !.. .. ,-!, - .. , . . .t

S. •_ ..... V • , --.t-\\ }~ ~ ~~... •. . ...... .... .... I - • . .. . -

-- T--_ . " I:" . ':-: , 7 -" ..... , ........ -

T. *. . ... .. . .. ...... ... U)

_7 7 7... --. _I --- , -' , -

0~ ~ T F-. ~ ~ ~ ..

-0w .. .... ---- --- .-

-I-Allr "0W1 Cl.0 ') ZHO:H~~lI 0

\I : -:_________0__

C0 CL__ 7-TI

t 1 .... ......

0~~~~~~n LA'wi- .. _ _ _

. . . . I L

0iVVY 0 0 ao US~ (H1)-dM OOIm.. ... ... o

78 ::7

CONFIENTIA.. ..... . .I . .. .. . ..

CONFID EN TIALNAVORD Report 4036

Screened portions of RDX were investigated by the No. II design

as to the possiaility of a packin6 effect similar to the earlier PETN

data; however, the reverse was found to be true, as the finer particles

appeared mc'e scitsitive on Z/0 paper than thc, as received portion.

Table XXII jo-v these data fox 50 trials pe: drop-height. Further

comparative data with a Z. 5 kilogram h•.rnmerare shown in Table XXUII.

The effect of particle size was further seen with two TNTsamples. Again, the effect was seen with Z/O flint paper. A TNT,

coded R-1321, appearea more sensitive than another, c-'ded R-1628.Microscoujic evaamination showed a difference in particle sizes.Particles of R-13ZI were in the or-ler of 10CGi lis di;&meter, while thoseof R-1ald appeared> Z00i&. The behavior seems in agre~ement with RDX

and rerent data with PETN.

Direct comparison of R-13ZI and R-1628 on the same day showedthe following average probability to explosion for Z0 triali.

Drop Height 7 10t0 1z5

R-13ZI Pe = .60 Pe .95 Pe =725

R-1628 PC .425 Pe .60 Pe =.60

Overall comparison of the same gave the following data,indicating again a difference.

Drop-Height (cm.) R- 13ZI R- 1628

of 2. 5 N~g. Hammer Trials Ave._P4 Trials Ave. P.

50 60 .133 '00 .16

55 40 .175

60 40 .555 100 .255

65 40 .6570 40 .75 100 .37

75 80 .788 20 .425

s0 40 .90 100 .46

90 100 5Z

100 80 .975 120 .683

125 20 .725 120 575

150 100 595

11-7179

CON FIDENTIAL

S o"s

-,~~~~~~~~~~~~~ ------= .. . . 9.'7--": -=Z ,•'• •

"CONFIDENTIALNAVORD Repart 4236

,i .<0i•,4 .,., o~ 1 o .. ,

co 0

•<i >° > 01I ! ! , o ,.,o . _./

Iwo , '•E CID 0

0..0

I , -a-//o z

-A1 -. / O li ... I " O O0,

5, 0 0 01o-ooo

*1l I..

14 so

Iii "" /0/ .

0~I %0&r-0

Z- -•.O c - M-

.44 v A

go q4. .. raIAi0

94, - -1 -

A4, %nc, Q

CONFIDLN TTALNAVORD Report 4236

in the case of TNT, the sensitivity to frictional offects is lessas the crystals are easily compressed and have saft, easily fractureledges. It seems possible that this material may be compressed pastflint crystals and escape friction. A confinement effect may also bepresent. Smaller crystals of R-1321 are able to a greater degree to fitinto d.z•ressio;•s or spaces between the flint crystals and are confinedto a greater degree than larger particles (>Z00O) of R- I6ZS. Theseentrapped crystals represent more confinement during impact, and inturn give more explosions and of more violent intensity than largercrystals which lack the initial confinement from entrapment. In thecase of PETN, the mateial is friction sensitive and less dependentupon confinement. Too, the drop-heights are much different for PETN

and TNT to cause a tremendous difference in confinement during impact.

Surface of flint paper (cross section)

(arrows indicate spaces forentrapment of crystals)Il I

Plunger

Anvil

X entrapped crystal

Confinement During Impact

These are under more cotzALkement than an explosive crystal restingatop the flint crystal, especially if drop-height is >20 cm. With PETN,very little, if any, crushing of the flint crystals occurs and the

phenomena become tribochemical in nature at drop-heights of 8- 16 cm.

for larger crystals.

Several explosives were tested in t; e conventional procedure bydesign II, and these results are shown in Table XXIV akd Figure 20-25,with significant data likewise present.

11-73* 1 81

CONFM.ENTUL1

IM

V)

CONFIDENTIAL

NAVORD REPORT 4236

Drop-Height Nitromannite PETN RDX(cm) of 2.5 Ave. Ave. Ave.kg Hammer E D N LE E D N E D N LE E

2 0 0 20 0

14 2 0 38 5 . . . .

6 18 0 22 45 0 0100 0

8 25 0 35 41.7 1 0 99 1 - - - 0

10 32 o 8 8o 7 0 93 T 7 1 32 18.8 2.

12 38 0 2 95 35 0 65 35 - - - -

14 19 0 1 95 84 o 16 84 - - - -

15 - - - - - 20 0 1OO 16.7 6

16 20 0 0100 6 0 o 4 96 - - - -

18 - - - - - - - - 25 0 35 141.7 -

20 98 0 2 98 1514 1 85 45.4114

25 - - - - 83 0 57 59.2 17

30 115 0 25 82.2 19

35 78 0 2 97.5 19

40 /,4o 0 0100 -

145 * '20 0 0 100

50

60

70

80

n- 7482

CONFIDENTIAL

,,Ig.-c- .

4 1I 9

NENO Tetryl Pentolite EDNAAve. Ave. Ave. Ave . A

- 0 0 20 0 . . . .

18.8 2 4 13 20 0 1 19 2.5

16.7 6 5 9 42.5 0 0 20 0 0 4 16 10 1 0 39 2.5

41.7 - - - - - - - -- - - - -

45.4 14 2 4 75 0 5 15 12.5 7 1 32 18.8

59.2 17 1 2 87.5 4 7 9 37.5 15 4 41 28.3

82.2 19 0 1 95 55 9 76 42.9 11 4 5 65 68 290 g 43.1

97.5 19 0 1 95 38 0 22 63.3 14 5 1 82.5 69 3 48 58.7

100 - - - 52 0 8 86.7 19 1 0 97.5 63 2 35 64

100 40 0 o0 - - - - 40 0 20 66.7

20 0 0 100 20 0 0 1CO 84 0 16 84

- - - 54 0 6 90

37 0 3 92.5

TABLE XXIV

SUMMARY OF DESIGN NO, 11 SENSITIVITY DATA

-.-- -

ý) 0

A. A e. ?4 96.'I "

19 _ _ _ o_ __ __ _ _ _

Av? ? 2 ].. , , 3 2 ]. Av.9 Av.

)0 0 5 0 3 4

298 1. 1 !85 45.4 14 70 0 5 15

e3 - 0 57 59.2 17 1 2 87.5 2.5 7 9

115l]. 0 25 82.,2 19 0 1. 95 55 976 42.9 11] 4. 5 6

!78 0 2 97.5 19 0 1 9.5 .38 0 22 63.3 1.4 5 1 8.5 0 o - - - - 58 86.T 19 1 o 97. -

20 0 0 - 4 0 -oo0 l o - - - -

- 0. 20 0 0 100 20 0 0 100

• •, TABLE %%C"V

7 0 9 5 91SUMMARY OF DE6N NO. 11 SITIVITY DATA

160 0 100 -- - - 52088 .7191 97.I

AmmoniumPentolite EDNA Perchlorate Fivonite

Ave,. AVe!. A AED N E £ D N E D N E D N

o 1 19 2.5

o 14 16 10 1 0 39 2.5

0 5 15 12.5 T 1 32 18.8 0 1 19 2.5 7 2 11 J4)

4 7 9 37.5 15 4 41 28.3- - - - 11 2 7 60

11 4 5 65 68 2 90 43.1 7 3 1o 42.5 17 3 0 92.5

14 5 1 82.5 69 3 18 58.7 - a a - a a -.

19 1 0 97.5 63 2 35 64 11 2 7 60 18 2 0 95- - - - 40o 20 o 66T - - - a a -

200 0 100 8 01684 1 2 2 6 65 19 10 97.5

- - a 54 o0690 14 1 5 72.5 -. --

37 0 3 92.5 - - a -

a - a a 17 0 3 85

CTA

iV.


D -me- sixo:t.e Pi.cric Acid b-cstton U Ccopositlon

(cm) o .5 a Aye. ,ve. Ave. Ave, Ave.

1 EE_ E D -W -i4 _______ E________ __m_____Ea__%_E

2

6

18

to 2 1 19 2.5 0 01 20 912 -

85 0 2 IS 716

20 2 31097.5 72 5 19 7 a Ii 2.525 10 1 9 5.

30o0" 2 90 - - - - 2 17 15 17.5 0 5 15 12.5

35 18 1 1 0.1 7 1 2 .5 . .- . .5 --

70 19 1 0 o 8 12 10 6 1 0,1 -2.5 2 a to 30

45 - - - -5 5 12 3 - 70 - - - -

50 19 1 0o 7. 11 7 2 72.5 5 1 14i 2.5 2 3 15 17.56

60 17 ,3 0 92.5 5 0 15 25 :,2 Ik 14 4T.5 1

TO0 - - - ,3 0 IT IS 6 23 32.5 - k4

75 20 0 0 100 . . . .- 26 1 33 44.2 5

8O 8 0 1.2 4*0 12 4• 4, TO ." -

90 12 0 d 0o 12 4 4 70 . . .

100 18 0 2 '.0 4 2 b7.5 6 6 8 5 72

110 20 0 0 too 1 6 1 60 -o

120 1 85 . .

125 - 3 - 13 25 56

1o0 :6 , 1 87.5 - - -

140 17 3 0 92.5 - - -

150 I 2 1 75 6 4 10o 0 52

IT175 _ _ _ _ _ _ _ - - - - - - - -

200 j 1613 20

225 . . . -275 I

U-75 TAK. 2111 CeSd.

63 1. Op =non No. 11 3M0TVM DAfTACON FAUT

ON

AAXmonlum Pot.assiua JA,.e. Aveve. e. ,ve. Ave. (cm) of 2.5

E 31 )L E E _____ N PE kg E D N %

a

6

8

10

12

16" ~18

- . - - -- --- 20

5 1 1.9 2.5 0 0 20 Q 25

- .-- - - .30

17.5 0 5 15 12.5 - - -

- -40

32.5 2 10 )0 --- -_ _

- - - 2 0o 8 5 50

21.5 2 5 As 17.5 6 24 70 l8e - - . 60

is J c23 32.5 - - -26 50 37 1, 0 29 27.5 0 20 0 75

- 26 1 33 4.2 5 7 8 42.5 .. 8o

40 t2 4 4 70 . . . 3620 M 146 . . ... 90

0o 12 % 4 70 . . . 42 20 38 52 20 5 17 58.8 (6 0 14 30 100

,,o 9 2 b7.5 6 6 8 45 7z 22 26 69.1 - 110

too ;3 6 1 60 . . - ... 120

- 15 4 1 85 - - 3- 0 9 77.5 4 0 16 20 125

-- - - 3 4 13 25 56 22 42 56 - --- - * 130

:6 , 1 87.5 . . .. . 140.

17 3 92.5 - - - - - 315 2 ,3 90 6 0 14 30 150

th 2 4 75 6 41 to 4 52 15 33 59.51 " 0 197.5 -.-- - ITS

19 L 0 97.. 8 o 012 4 200

161320 - - 3 L 16 17.5 25

14 0 6 TO 20

33 0 7 82.$ 275

15 0 5 75 3o6

TAKA ]Zzly, Coated,

IG Mm OF MCGA O. •11 SKTIzIVm •ATA

.A k

COIN FIENTIAL

kAVORD R~r'ORT 423$

00

+- - - -

- ______U

->

0

I- U)

IL

C.,

L.)

00dN

*d 0owx ci di~wa d

a.counloCN"

4. -


100 0i

60

0R~ MITOMGHTT OP2 G AME

PCENTOLOEET29

CONF IDE NTIALNAVORD REPORT 42 36

a= -

L za

... ... _

00

be

Un

Isus

L-4-

x U)

* O)I~4 ~-1 ... C4a

060CONFIENT0A

CONFIOENTIALNAVORO REPORT 4236

4 c0zz

+ 0

4%J

00

mo I- J

Lc) xO~K .~ .~IIS~ .A

CONFIDENTIAL

~A P

SCc'ý NF N 7; A LNAVORO REPORT 4236

iA

if 7L7 "!1VFF

t60- X (F)tt

L Q (IS)

I-- A .... -. " -oli "+.' + "+ -r . -; -j ji • '

0 (CA) _ :T t :-

* IJ,

4C -- - 5oEXLOIO

S• _ -T 'rI- --- T--T W I-• : .. , • # ! I -. ' "

I EXPLOSI4 0EPOII

Fiv t IK I F) 24

01/ 0P-Ei .,S J CM , I T. -"E/ IS 5

FG,24 THEORETIC I SENST".VITIEB DSIG 5O.11

COMPOSIETI ONAAL) 6

..... ...- "EIGH (CM I OF- 2. KG .HAMMER (

FI, 24. THEREICA SESIIITE B--Y DEIN-

!--•.• T /r -T' i- " FVOIZ-80 4asl

I~~i tl CONI DE"NTIALl~~PI;

-- • -i--l .....•.,OSTO. .-..- 7M

CO?.IE IDENTIALI NAVURD REPORT 4Z36

OI

0.0 a%-Wi)

1 00

0 -j *L .' ..

0 0 ILI--0.

Cd

1 O Z~ I.L 3 O c . .aa

I.-S0 -j 0089

CC2ONTA

,.,.~ o' S,

CON FIDENTIAL

NAVORD Report 4Z36

In conclusion it may be said th-at, for explosives in the higherand intermediate range of &ensitivity, design 11 is satio5factory. Thecushioning effect of the paper itself enters the picture with suchsubstances as mercury fuim-.niate, lead styphnate and lead azide. Onemay expect explosions in the region of 1-8 cm.; but until the cushioningfactor is reduced (>10 cm. drops) explosions will be few. Unfortunatelythis design is not satisfactory for certain explosives of low sensitivityas at drops >100 cm. the identification of e'cplosion becomes difficultand the 100% explosion marl -s not reached. For materials of soft,waxy consistency, not much. better than 50-75% explosions are obtainedas the drop-height is increased. To overcome this difficulty, a strikerof 1 1/4" diameter was subs'.tuted for the 1/2" striker of design No. 11and the development of des; gn 12 was u.ndertakena.

The N•o. 12 De sign

As rmentioned above. "this method was a revised No. 11. Thelarge striker was found to .e satisfactory and 100% explosiondrop-heights were obtained. This pointed in general direction thatonce again, the escape of L-e sample was occuring in the No. 1L case.The No. 12 design also involved the use of 5/0 flint paper in hopes ofelui-inating any particle size effect, as was the case with a flint paperof 00 specification.

A previous report (March, 1944) summarized most of theimportant comparative data for Design 12; however, some additionaldata axe presented here. *

Table XXV shows Lk-,a: the particle size difficulty has beenminimized for PETN, wh.en tested by Design 12.

*The report referred to is not available although it is believed thatthe material is included in report 3 which is the next section.

U_- a90

CONFIDENTIAL

• 4

CONFID E!jN TI A LNAVORD Report 4236

TABLE XXV

COMPARATIVE DATA FOR PITN BY DESIGN 12

Screened Material ThroughDrop-Height As Received Mat'l 100 mesh on ZOO meshZ. 5 Kg. Hammer Trials Trials Pt

10 z0 .35 40 .575

1z z0 .60 40 .60

14 20 .70 40 .95

In September, 1943, several polymorphic forms uf TNT fromCornell University were investigated by Design 12. Several of thesesamples appeared more sensitive than regular TNT at intermediate(40-75 cm.) drop-heights. The more sensitive material was very lightaiid fluffy in texture and the amount used per trial amounted to aboat1/3 that of regular TNT. It was learned that the same lower orderof sensitivity could be obtained wiLh regular TNT if a 5-8 mg. chargebe used instead of the usual 20-25 mng. charge. Also, if low density(0. Z5) TNT were tested, the same lowerorder of sensitivity was present.It was interesting to note that the difference in sensitivity disappearsat drop-heights >75 cm. and also if the samne weight of sample beused for comparative testing. Table XXVI presents in summary form thedata discussed above.

Designs I I and 12 involved the use of a new idea in qualitativeinterpretation. To approach quantitatively the fact that TNT may charor feebly "explode" at a drop-height at which PETN detonates furiously,it was decided to evaluate such weak decompositions as a D or doubtfulexplosion. In conventional procedure, it amounted to I/2 explosion.For example, in a series oa 20 trials if 10 showed E or explohion,7 N or failures and 3 D nr doubtful explosions, the %E + D with D = Ewould be 65 while JE with D x N would be 50. The average % E would be65 + 50 divided by 2 2 57. 5. The above procedure gives a quantitativeapproach to the question of degree or intensity of explosion, which ismost important in sensitivity studies. Explosions are qualitativelyseparated into various types according to the degree of explosion with

11-835 91

CONFIDENTIAL

..Li

CONFMEN TIAL

NAVOR[• Report 4Z36

> %n

e~ fl NI 0f.) sn

~d > LnI oP/ r, N •• ! ,, o '

,n V 7"i -E /u

.I . 1 - .-

0•.l • l l -

z0

0

r4~ N0A--d

. k4. 'da'

S'•' -i'" " •' .... • •Iell .- "*,e '

' I" "j o "

uAI<,, II- o " " " 0 . . .

zH 0

H o-! *I %M a x "

- 11 <t •, 0 e 0 _ 0 0

0 oNO N *i-U M

0-4m

,o. S.

N

f4.

I.•

L CONFIDENTIAL%n

..A

t- .... . E

a f- Cpl-z ii00 L

4____a ___

_____ __Co__a


sound as the main criterion. With certain aluminized explosives,doubtful explosions are also classified as to those accompanied with orwithout visible sparks or flame. Such factors as visible flame (f),loudness (I), completeness of detonation (c), nearly complete (Z, andpartial or low order explosibility (p) are kept in mind in symbolizingE, Ep, El, Ef, Epf, Elf, Ec, E-, Ecf, E'f, D, D., and Df.

While the qualitative classification does not affect the valueof ••E or synonymous Pe, it does give information that should beconsidered. For instance, during a rather extensive investigation ofvarious Torpex sam-ples following an explosion at Yorktown, Va., itwas found that a certain RDX-aluminum mixture exploded with thesame intensity of explosion at the same drop-heights as mercuryfulminate. AJthough the frequency of explosion was less than thefulminate, it was important to know that once an explosion did occur,it was most violent.

One of the disadvantages of the No. 1Z design was foundto be the tendency of certain oxidizing agents to react with the paperbase of the flint paper. This was first noticed while investigatingammonium nitrate. A further study with this particular substanceshowed that appreciable reaction is experienced with certain types ofnon-abrasive paper. Table XXVIII. The procedure of loading theexplosive was identical with Design IZ, except that these various typesof paper were substituted for the 5/0 flint paper.

Designs 12a, l2b and 13

Following the extensive program of sensitivity study fromAugust through October, 1943, attention was directed to the developmentof a design suitable for testing liquid explosives. Development in late1943 and early 1944 saw the use of designs IZa, 1Zb and 13.

Method lZb was simply No. IZ without the 5/0 flint paper.A drop of liquid from a common medicine dropper was placed in thecenter of the usual ketos anvil, the 1 1/4" diameter striker placed atopthe charge and the weight dropped from desired heights. Method IZaemployed a I/Z" square of No. 615 filter paper to absorl the liquidbefore impact. The striker and anvil for lZa were identical to thostof 1Z and 1Zb.

43CONFIDENTIAL

CONFIDENTMLNAVORD Report 4236

.9

E40411.. 0

"In~

6,0o

0 ~

0 ~0

0 Id

"4 z.I

In~

CONTXDEN TIAL


Design 13 was developed with the idea of hitting the liquid

explosive semi-directly and eliminating to a greeter degree thesqueezing of the material from between the striker and anvil surfaces.

A small wooden applicator (.075t1 diameter and 3 1/4" long) or"toothpick" was inserted through a 1/4" diameter hole at the top of the

striker and placed across thu outer guide ring to elevate the striker

about 7 mr-n. above the charge of explosive. Plate XI illustratesdesign 13, while Plate VI shows 1Z, IZa or l1b in position to receive

impact.

The wooden toothpick was elastic and naturally required some

energy to be stretched so that the striker made contact with the

explosive. With sensitive solid explosives the drop-heights are greaterfor design 13 than for 1Zb. Liquids, on the other hand, show greater

probability to explode at lower drop-heights with design 13. Thesqueezing tendency present in IZa or 12b is avoided with No. 13 and

liquids are able to receive more direct impact at a localized or

concentrated point of application. Thin layers of poor propagatingmaterial are present in the procedure of 12b which undoubtedly require

greater impact energies. Table XXIX shows comparative data fordesigns 12b and 13 for a sensitive liquid and solid explosive.

TABLE XXIX

COMPARATIVE DATA WITH SENSITIVE SUBSTANCES FOR DESIGNS

12b AND 13

DropHeight Lead Styphnate Nitroglyce rin

2. 5 Kg. Design Ib Design 13 Design lZb Design 13Hammer Triais Pec Trials ±ec Tral.s tec Trials -et

6 20 .05 10 0 10 0 ZO .158 20 .55 ZO 0 ZO 0 20 .80

10 ZO .75 20 .20 20 .05 20 .8512 10 .50 20 .75is 20 1.00 20 .85 10 .80ZO 20 .95 20 1.00

11-87

CON FIDENTIAL

. ,- .. .. .. -_-


;" : , .I. , .:. .I,;4.--4'.: •. !

""Fe .. i.

.•;:... * .. *.... ..A ,- . :, 4 -. .

A a t.4 IL

G INNER GUIDE RING

S •ELEVATED STRIKER

P a "TOOTHPICK" TO ELEVATE THE STRIKER

W W0 2.$ KILOGRAM DROP- HAMMER

Me• ELECTROMAGNET TO HOLD AND RELEASE THE HAMMER

Rno SLIDE-PLATFORM TO RECEIVE HAMMER ON REBOUNDS

PLATE 31

DESIGN NO. !3

96

CONFIDENTIAL

* I

•. .. ~€- ,

L../, 2> ..

AI*I I V L


S~It is seen from Table XXIX that as the drop-height becomes

>d0 cm., the probabnltities of explosion become the same for bothdesigns. For explosives of intermediate id high insensitivity, the

' energy lost in stretch-ing the wooden pin becomes negligible.

S~Design 13 was employed during the study of numerous liquid

:explosive mixtures. These data are summarized in Table XXX and

XXXII. Graphical treatment of these data is seen in Figure Z6. Solidexplosives tested by design L3 are summarized in Tables XXXI andXXXIIX and Figures Z7-ZX.

Plate XU Mustrates the general damage to 1 114" strikers by

powerful explosions of 1--drop (40-50 rmg.)of nitroglycerin. These arethe most violent explosions encountered in sensitivity studies. Thepitted anvil (A) of Plate XII is due to the action of mercury fulminateexplcding. Strikers and anvils are usable for about 5-10 trials with

this substance, and then require refacing.

Figure 33 shows the linear relationship between the logarithm ofthe 501, explosion height and the percentage of desensitizer in a liquidexplosive mixture of sensitive and insensitive substances. Note the amountof desensitizer required for the nitroglycerin to make it safe, yet explosive.

This latter value in % desensitizer is in the order 35-50%.

Table XXXIV and Figure 31 show data for now discontinueddesign IZb. This design hab been discontinued with solid explosivesbecause of a disturbing reason. It was found that the 50% explosion

height for PETN was about 70 cm., provided the striker was merelyrested atop the usual heaped charge. However, if the charge wereflattened by pressing firerly on the striker and at the same time spinningthe striker to give a thin layer of PETN of I 1/4" diameter, it wasfound that explosibility from a drop-height of 337 cm. was only about20%. The reason for this was likely the uneven distribution of energy

to the thin charge. Energy dissipation likewise was encountered here.Since such results would place PETN in a class with TNT, it was

decided to discontinue design 12-b.

As long as PETN charges were not flattened (as described) beforeimpact, the material showed 100% explosibility at 100- 125 cm. But thisdanger is apt to happen during any determination involving design 12-b.

/ ' 1-899.

97CONFIDEN TLAL

CON FIDENTIALNAVORD Report 4234

0 0 Z .<~Pd P

z

(0m a 000Sj n44. -; P4 04 1

iin TIo'0 d

:.zI I

* WI 0'~0 0 Pd

-1 I. ax oa, Cdddd D

n W

z .zwn1

0000

z 0 a 0 .0 0 '44

A.

t. 0 .

0is.

: '1.1 '4 p4.4.4.

-. *7-

-mom ~ 0--*.

CO N F I DEN TIA L

NAVORZ) Report 4136

TAB~LE XXXI

DATA FOR SOLID PLOSIVES FROM DESIGN NO. 13

Drop Lead

H~eig~ht tpnt PETX( Pertiolito RDX Tatryl op4

0 zo 20is) 20 as

to to 10 20 o 0

30 t0 to0 t0 0 0to3

40 to t30 .S 2 0 t 0 t0 I*

so ~ ~ ~ ~ ~ ~ ~ 2 5oA 2 .5 2 s t 0 2.

60 8o z A 0 20 z5

80 t0 60 oI

90 t0 84) AD7T.5

l00 t0 95 Zo 100 20 40 20 100 40 85

131 to Ss 10 99

ISO A0 70

175 30 70

zoo A0 67.5

2501z0 90 __________

10 0 30 100

*%C (or Pentolit* and soacgquout explosiwes is AVG. SE

Drop

He.ight boron Torpex miti@I-Z Cop 8 T1T Pleric Acid Explosiv* D,

A ~(Cmn.? ~ T ~ T T %t T ~ T

30 10 0

40 20 2.5so to £1. 5 20 6 0 t .~ 3P0 a

40 t0 4a. s 20 0

7075 to l7.5 30 m.st 30.s 40 10 10 a

90

100 30 100 30 47. 5 20 45 40 23.1320 42.5 20 0

Its AD 7z.s

ISO to 97.S 0 to a 40 48.6 20 Is

175 A0 as

lo0 10 1" Z 391.5 40 "AD to 0

ISO 40 48.6

300 AD 95 40 41.5 30 1.5

3730 68.3 20 it.s

U-91

c 9' 1 TIA1

I CONFIDENTIALI NAVORO REPORT 4238

ZI:. .. , as Q Q

1w ........ Z

0 o_ 0 00 3

U.

LA1+-9

a.a.

I~NTIAL

NAVO9?D REPORT 4236

0

o Wo

-- - -- - --

Ii z

I.coN

_ _ _ _ _ _ _

c-

-~-.--.

*g~- 77

CON4F IDENTIgALNAVORO REPORT 4236

CL 0

zU)

~ra.

CL

0 c0

SUISoIldX3 '$SAV

102

CONFIOENTIAL


_ 0

2w

-3-2 ~~~ -Z7--7--:Z

wU)

2

a.I w

8 ~~ x

SNMOCbOWX3 %'9AV

103

CONFIDENTIAL.

LI

CONFID•ENTIALNAVORD Report 4236

TABLE XXXII

SUMMARY OF SIGNIFICANT GRAPHICAL DATA FOR CERTALN LIQUIDEXPLOSIVES AS TESTED BY DESIGN NO. 13

•a

u 0.- - " o a.

00

Na eo4xlsie.• 0 .

Xo u 00..X 1 m 4 w>

X XU . a Ixý

Nirsicri . 10 4, .W• 1.0 .0

00/ N 000 . 0. A4

o. 0 10 0. 4 ..

D.M.0 0hhlt 4a. 4. w148 . .

0.0

Namneof Explosive 0 W o 1

Nitroglycerin 8.8 1.0 4.4 1.0 16* 1.0 1.0

80/20 NG-DGTN/D. M. Phthalate 36.5 3.9 10 2.3 40 2.5 2. 9

75/25 NG-DGTN/0 M. Phthalate 42.5 4.8 21 4.8 60 3.8 4.5

Diethylene glycol -dinitrate (DEGN) 48 5.5 15.8 3.6 98' 6. 1 5.1

9S/5 DEGNT/DNT 62 7.0 23 5.2 14Z* s.9 7.0

90/10 DEGNIDNT 16 8.6 21 4.8 M8e' 11.8 8.4

70/30 NG-DGTN/D. M. Phthalate 36 9.8 30 6.8 159' 11.8 9.5

85/1S5DEGN/DNT 109 12.4 35 8.0 2400 15.0 11.8*Extrapolated

11-95a103 a

CONFWDENTiAL

- .

S S m w u m 1 I l

CONFIDENTIALNAVORJD Report 4Z36

TABLE XXXIII

SUMMARY OF SIGNIFICANT GRAPHICAL DATA FOR CERTAIN SOLIDEXPLOSIVES AS TESTED BY DESIGN NO. 13

.o u 0 00o

S2

I4.- o€

. o 0r. 'a. Z0.~ Z 0. g -.

o .*0

U 0 Q 6 a >0

0. % 0%. - .-4- O d a 0 d an a >c

0*6 0 is t: 0U C a

Explosive > >* • >0 0 . o , o

to 140.W 1310 4.40

Lead Styphnate 6.7 10.7 5 3.6 14 19 3 9PETN 3.3 64 29 21 35 105* 17.5 Z7Boron Torpex 8. 6 66 30 33 55 92 15 3350/50 Pentolite 7. 1 66 30 30 50 98 16 32

Torpex-Z 3.4 74 34 31 52 150* 25 37Tetryl 7.6 83 38 40 67 100 17 41Compositio., B 4.0 102 46 56 93 350 53 66Minol-2 4.3 104 47 48 80 192 32 53Picric Acid 5.1 109 49 70 117 242* 40 69R.DX 1.5 110 50 22 37 337 56 48TNT 1.8 220 100 60 100 6000 100 100Ammonium Picrate -- >337 >100 Z70 450 >337 >100 >100

*Extrapolated

uA

V A , CONYIDWNTIA.L

Li _____ "

CONFI0DENTIALNAVORD REPORT 4236

1000 z z. -r- ____ _

800

400 _ _ _ _

- r-100 I4

2 40

00 -i---- - ________TEDINTRTOUE

o 60 jESENSTIZE+

. .........

$ ~~10 V7Wi LOGY .1XE 093

0.7 -7

a0 0 20 soE 4E0GLCOO 50 AT(ON0ROO70N% DEENITZE INSE:NITIZER

EXPOSIN HIGHS O NITROGLYCERINE ANDET4Y DITALTHYEE

CONFINSETTIER


CL..1

SA1.

I4 EI

SA a PITTED ANVIL PROM EXPLOSIONS OP MERCURY FULMINATE "I

PLATE "Xlr

DAMAGE TO LARGE STRIKERS BY POWERFUL EXPLOSIONS 4;OF NITROGLYCERIN, WHEN TESTED BY DESIGN 13.

, €ONFIDENTIAL

/

V..

_ _ _-_ _,


U w

Sv -0

I.S

X "a

JA

to ,n dOQ •,

000000 0

Wit4 N- na n4 a %

at -adaW 0 00 0

1We .010 Z 1f 1%1

C O N FI D E N T AL

,a,,. -. ooll ' °lII'°°

-an .4o.4 '

100 0 0 0

SOI SI~iT

.45 ,51I4141

-' "

SIll m y I l i ._ L0 0 0 0 3 0 04

....... .. -....... ... • • 7 . ...

.•- , .IA

CONFIDENTIALJ ~ NAVORD REPORT 4236

3.LWNHdA.LS OV31 ONV 31Tia-mif4 Aufl3J3VI HIOA NO)~ lH9t3I-dOlO 00-1

. .. . . . ...... .... ............. 0 0.a. .

of 1-~~I Z W: z

la - a 1£ ... .~ ftO~v--ee 0

JW 0

. . . . . . . . . .. - .I

I~ I u........... w

Ia.

z -0o

Z z 2 ~'......


In addition to determining the average % explosions at various 4

drop-heights for design 13, a quantitative approach was attempted. Thevarious qualitative types of explosions obtained were given arbitraryevaluations in an attempt to interpret the behavior of PETN and Pentoliteas studied by design 13. From Figure 27 it is seen that the 50% explosionheights are 64 for PETN and 66 for Pentolite. Practically, these arevastly different because of the difference in the degree of intensity ofexplosions of these two explosives.

The procedure was as follows: arbitrary evaluations for varioustypes of e-tplosions became: on the basis of 50 for a complete detonation(EC), N 20, D , D. 10, Ep = 0.50. E . Z. 50, El = 3.75, Ea 5= 4. ?5.These evaluations also represent the amount of material exploding, asexperience has proven that a D represents about 2% of the charge as explod-ing, an Ep as about 10%, an E as about 50%. El as 75%, EZ-as 95% andEc as 100% exploding.

* The evaluation principle was then applied to conventional data of20 or more trials per drop-height. Each type of explosion wasmultiplied by its evaluation to obtain a total evaluatioc or :E.. Thistotal actually represented the average '%o of material exploding pertrial at a given drop-height. By this scheme, the most sensitive ofseveral substances tested at a given drop-height, would have a X Ev of 100.

£

The above approaches a gas measurement, but of -:ourse isentirely personal and represents only an approximation to an actualevaluation from a measure of the gas evolved during an t .cplosion.

'I, Table XXXV summarizes results from design 13 ,vith theevaluation principle applied, while Figures 3U-33 show £Ev as afunction of the logarithm of the drop-height as plotted on probabilitygraph paper. From these data, it is seen that PETN ard Pentoliteare in better order. RDX behaves out of line by design 13, and recentlyIt has been decided to use this design for liqtid explosives only.

1K-i(,;

___ ,U_-101

109, ~CONFID EN TIAL

CONFrDENT1ALNAVORD REPORT 4236

°rop-Hee ight _Tetr!(ca.) or !-&aJ Styph"ate PETN T__r____2.5 Kg. rHan&r £p Z El E~r E, D N SE, E£ E El E,• Ec D M X E. Ep E El SLW

6 0 0 0 0 0 0 10 08 0 0 0 0 0 0 20 0

10 0 0 0 a a 0 16 20

15 0 0 0 1 O 3 85 0 TI

S20 0 0 0 0 20 0 0 100 0 0 0 0 0 0 20 0

0 - - - - - 0 1 1 0 0 0 1, 6.3

40 0 0 1 0 2 a i7 13.8 0 0 0 0

500 0 0 1 5 0 14 29.8 0 1 a 0

60 0 0 0 2 3 0 14 2'4.5 - - -

TO 0 0 0 2 10 0 8 59.5 - -

75 • .. 0 2 4 0

80 0 0o 0 o o 1 0 2 89 0 2 8 0

______ - 0 0 0 4 12 0 479 -0 a 1) 0

100 0 0 0 3 16 0 1 94.3 0 2 16 0

125

Urop-Height(co.) of PicrIc Aci, Minol-2 FOX2.5 Kg.Hammer Ep £ El Ef' to D X ZE Eq £ El E E£ 1) N T Ev £p E E Z?

20 0 0 0 0

30 0 0 2 0

.0 0 0 2 0

50 0 0 0 0 0 0 20 0 0 03 0

60 -

70 . .. .. . . .. . . 0 0 3 0

75 0 0 0 0 0 0 20 0 0 0 0 0 0 9 11 .9 - -

100 0 6 1 0 0 3 l'i 19.1 2 4 0 0 0 f 7 11.7 0 0 6 1&6

125 0 5 9 0 0 1 5 46. - . . .. . . 0 0 0 11

150 0 5 "0 0 0 A6 1 50.6 3 8 8 0 0 1 0 51.6 0 1 a 11

175 1 it 0 0 2 2 52 - 0 0 1 37

200 2 3 5 0 0 0 052".6 0 1 322

250- -

300 0 0 0 18

337 0 0015s

~ '5 W.OEV110 LUo.- o o. -

I II El Er lr D N & Sp E E1 Er Et D m ZE, Ep Z El Er E, 0 V %E,

I i

II

20 0 ... . . ...

it 13.8 0 0 0 0 0 0 10 0 0 0 0 0 0 3 it .3 0 0 0 0 0 5 25 .5

14 219.8 0 1 0 0 0 19 2.5 o 3 0 0 0 4 13 T.9 I 10 0 0 19 19 3.2

14 24.5 . .. . 0 3 0 0 0 6 i1 8.1 2 C 0 0 0 6 12 1.68 59.5 ....-- - - - ......

- 0 2 4 0 0 2 12 20.2 0 13 2 0 0 2 3 41.9 3 6 1 0 0 6 4 20.9

2 89 0 2 8 0 0 L 9 35.1 - . . - - . .

4 .9 0 2 13 0 0 1 4 53.9 .- - -

1 94.3 0 2 16 0 0 0 0 12.5 1 10 9 0 0 0 0 9.3 6 22 3 0 0 6 3 36.1

o 0 9 o 0 1 0 67.8

._____,____ E Cgcroaltlon 5 ¶WT

m rEE EP K1 El E £ E 0 E, -E D 3 9 El El, E, 0 N 9 4 EP K 91 Er K, 0 V 9El

0 0 0 0 0 0 20 0

0 0 2 0 0 0 18 7.5

0 0 2 0 0 0 18 7.5

200 0 0o 3 a 0 0 1T 11.3.0 0o 0 0 0 1 19 .1 0 0o0 0 0 0o20 0

*- --- - - - - - - - 0 0 0 0 0 0 20 0

- 0 0 3 0 0 0 17 11.3 - - - - - --

11 .9 - - * 0 0 0 0 0 3 17 .3 1 0 0 0 6 33 .6

7 11.7 0 0 6 :4 4 0 36 36.30o 9 0 0 0 8 3 U- 0 6 0 0 0 72 7r .9

...0 0.0 L. 0 0 9 52.3-o o..

c 51.6 0 1 a it 0 0 14 62.3 0 a 8 0 0 2 2 50.3 0 14 0 0 0 11 15 18.1

. . o o 01 37 0 1866.7- . - . - -...... .... ....-

o054.6 0 13 22 1 0 2762.3 0 6 13 0 0 1 0 63.t9 010o0 0 01it18 13.1

S ... o 0 o- o . . 1 12 0 0 916 19.50 0 0 18a 0 0 285.50 o a 0 0 I o 6a.S6 0 0 0 & 12 27.9

0 o 00 1 5 0 0 96.2 1 150 0 o 9 5 25.9

TAB.&& X111S•OS4AR Of DES1I1 1) NAL•IUON DaTA I"• OUN)

... .- .. L.. _

,I

. Tetr , Pentolite, T

El ke X D N ZE, E El EL, EC D v ZZ, rE E 91 EWr EC D N j. KEy E I

0 0 0 0 20 0 .ii1 0 0 0 13 6.3 . . . . . . 0 0 0 0 0 0 20 0 0 0 01 0 2 0 17 13.8 0 0 0 0 0 c 10 0 0 0 0 0 0 3 17T .3 0 0 0

0 1 5 0 14 29.8 0 1 0 0 0 0 19 2.5 0 3 0 0 0 4 15 7.9 1 1 0

3 2 3 0 14 24.5 0.. 3 - 0 0 0 6 11 8.1 2 0 0

0 2 10 0 8 6,9.5 - -. - - -

- - -1 0 2 8 0 0 1 12 20.2 0 13 2 0 0 2 5 47.9 3 6 1

6 S10 028• 9 0 2 8 0 0 1 935.1 .. .... . .

O 4 12 0 4 T. 0 2 13 0 0 1 1 53.9 - .. .. . .

0 3 16 0 1 9%.3 0 2 18 0 0 C 0 72.5 1 10 9 0 0 0 0 59.3 6 22 3

0 0 9

* I4ino),-2 mmX C•;RX omposttol n a ____

1 E~ 2, ED 9 I Ey E 91 ET ye~ 3 N Zl~ ItK El D~ N~ DUE Ep L

1 0 0 0 0 0 20 0

• 0 0 0 0 18 7.5

a 0 2 0 0 0 18 T.5

0 0 0 0 20 0 0 0 3 0 0 0 17 11.3 0 0 0 0 0 1 19 .1 0 0 4

- - - . -- -- - - - - - - - - - - - - - - - - - - - - 0 0 '

0.. . . . 0 03 0 0 a 17 11.)3- - -

0 0 0 9 '11 .9 . . . . . . .- 0 0 0 0 0 3 17 .3 1 0

0 0 0 7 7 11.7 0 0 6 1 4 0 36 36.3 0 9 0 0 0 85 23.3 0 6

.. . - 0 0 011 0 0 9 52.3-- ----..

0 0 5 .051.6 0 1 2 11 0 0 I1 62.) 80 8 0. 0 a a.2 0. 01 i.0... . 0 0 1 37 6 0 8 6.? - --

5 0 0 0 o 54.6 0 1 1 1 0 27 61).0 6 13 0 0 1 0 63.9 0 10

. • * I. to1 0 001 0 0 a 85.5 a 27 0 0 1 0 62'.5 019

a 0 001s 5 0 096.1 15I

?A]" I=

tot

.or• -2 0 Boron 0rorvex (.5.) 0.11 7 1 1 0 0 0 1 9 . 1 0 0 0 3 •6 .50r

6

8

to

1,' .3 0 0 0 0 0 5 25 .5 0 0 0 0 o I 19 .1 ..

SUi 8.1 i 0 0 0 0 6 12 1.6 2 3 0 0 0 T 8 9.2 60

- 7T.9 3 6 1 0 0 6 4 20.9 1 13 5 0 0 1 0 51.9 75

0 0 59.3 6 22 3 0 0 6 3 36.1 0 8 12 0 0 0 0 65 1oo

0 0 9 0 0 1 0 67.8 - - - 125

iN' A',ontua Picratc (Co.) of2. 5 Cc.

0 N I3 £E Rp z EE D X l £ E Z El 4 I[€ D V Z5 1 Hair

20

30

1 19 .1 0 0 0 0 0 0 20 0 50

0 00 0 0 0 20 0 60

3 17 .3 1 00 0 0 6 33 .6 -- . *Sr

83 23.3 0 6 0 0 0 7 27 7.9 0 0 0 0 0 0 20 0 10

a Z !0.2 0 %• 0 0 0 11 1i 18.1 10 - 15

1 0 63.9 0 10 0 0 0 l218 13.1 0 0 0 0 0 0 6 0 2cm

- a 1 12 2 0 0 916 A 19.• . . . . .. -----.

1 0 62.6 0 19 2 0 0 8 11 27.9 a 0 0 0 1 tt 2.6 )w

is 0 0 0 e 9 2 0 0 0 0 1 1 1

II L

II_ _II__ _ _ _I -I I I I

-. . . . . . . . . . . . ... .. -- - - -. . .- - - . . . . . . . . .- . ~ - . . - - - -

CONFIDENTIAL 4

NAVORD REPORT 4236

- -i- -• •, i -

S:.I:.I::..::::)

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

"I "I1..:.• :z "50"2 0. .- 0

I Tmh w:I1::e- I"... . ... N a .z

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

-a.0

. ".. .. .. ..

-1 '! - , - , . . .o . . .

40

..... .. ...

A -.. j -. .-1 • -w; . . ..

7 : :i..... . .: i : p ".. .... ........ -

-.- b -- 0 .- i .- -

iii " '- :- -:.*. ~~j7::::

. . , ...................

.,,. . . .. .. . o . . . . . . . . . IJL

..i. .._.. .... . .... .... ...... . ..

Ij!I

,I- ...... . . .-. .

C3%, tD-:N~lA

NAVCORD REPORT4236

- L

~~~~- I4- I u c a

I... LJ 0 .

MOIi-TOW

04 00a 4ý t- w:77 M

7r 004i 9I' ___

ID I

CONFIDENTUL

NAVORLD Report 4Z36II

CIc.1n1!uding Remarks:

* By way of evaluating various designs we may conclude that

N ." 3, 12 and !. are the ma, satisfactory at the present writing.

i! The No. 3 is satisfactory for cr.-nparing materials in the low and

intermediate levels of sensitivity, the No. 1Z is satisfactory forS..mparing insensitive materials, while the No. 13 seems fairly

{ satisfactory for inveptigatini c •rnnparative behavior of liquid explosives.Other designs all contain disadvantages which caused a discontinuationol each.

As for future work, a newer approach to sensitivity work may

be attempted. To obtain better reproducibility in impacts, it is hopedthat an inverted pendulum type of machine will suffice. The ideaarose after an unicrz',nate conclusion that the falling weight machine"possesses, by nature, uncontrc2•able deviations in the form ofirreproducibility oi impacts as the hammer hits the striker. These

.4 slight differences are shown earlier in the form of I0 iuccessiveimprints. The new machine, if equipped with excellent bearings for the

1* arm of the drop-hammer and aiso proper facilities to obtain good align-ment of hammer and anvil surfaces, should produce more reproducibleimpacts. This type of machine also gives a practical test in that the

explosive is struck directly by the gu-ded weight corresponding to theblow of a claw harame: or a sledge hammer, dependink upon the massof the proposed weight. Such a machine could be used for testing anytypes of explosive, i. e., solid, liquid, molten or frozen material.

Another idea for future consideration is the substitution ofj 3/4" diameter abrasive-coated metal discs for the flint paper of design

12. It is hoped that the reactio", of certain oxidizing agents with the

paper base of the flint paper will be thereby removed.

It is likewise hoped that the measurement of the gas evolvedduring explosions from impact may be carried out in the near future..

It is this writer's opinion that solid explosives will show

deviations regardless of the method of testing. The elastic propertiesof a given solid explosive seem to be the important factor in impact.Hard, brittle, inelastic substances are usually sensitive to impact;while soft, waxy, elastic substances appear in the intermediate and

finsensitive categories. The pressure rise during impact is rapid snd

reaches an enormous maximum value. The maximum pressure isi 1 U-lOS

*11I CO.FMENTL4

CONFYDMENTIALNAVORD Report 4236

dependent upon the modulus of elasticity of the subLtiaace receivingimpact. With stcel, these pressures are reproducible, as the modulusof elasticity is contant. However, with explosives. variations inpressure occur while investigating the same substance, as the elasticproperties vary throughout the mass. Each scoop of a given explosivemay vary as to its modulus of elasticity; and as a result, differentpressures of impact are produced even though the harnmer be dropped.from a given height for successive trials. The pressure variationstogether with the noa-u-.iforrm impacts of the hammer against thestriker, will account for variations in the form of doubtful, partial,common, laud and complete explosions all occurring at a constantdrop-height and also the irreproducibility of the same number of eachtype if a given series be repeated.

The pressure variations due to divergences of the raodulus ofelasti,:ity are most likely minimized in the case of liquid explosives, aselastic properties are more nearly uniform here. In general, liquidswill explodte with greater intensity than solidi. This seems logical,as in liquids the molecules are more compact and once the explosivechain reaction be initiated, it can continue with greater easet whereaswith solids, the gaps between individual molecules tend to hinderpropagation and less intensive explosions are produced. Pastexperience has shown that solid explosives are usually slightly morecensitive when in a fine state of subdivision. This likewise seemslogical in that molecules are slowly approaching compactness and aliquid in behavior as the subdivisions become finer and finer.

The unfortunate conclusion concerning solids is that at bestwe will have to be content with comparative data representing aaverage value of the limit of the true value.

Rogers F. Davis

i.

11-106I " 114

CON IDEN TAL

--wim• • • m I l I I I l I 1 P "

CONFIDENTUALNAVOB.D Report 4Z36

References:

(1) H. Kast (Trans. by W. J. Williams), J. Frank. Inst. 169,(1943).

(Z) F. Lenze, Z. Ges. Schiess-Sprengstoffw.1, Z87 (1906)1 Section

III B, Intern. Cong. Applied Chem., Rome, 2, 522 (1907).

.4 (3) H. Met-egang, Section III B, VI Intern. Cong. Applied Chem.,

Rome, Z , 53's (1907).

(4) H. Kast, 7th Intern. Cong. Applied Chem. J. Chem. Soc.

Ind. Z8, 747 (abst.); Z. Ges. Schiess - Spregnstoffw.4 Z6T(1909).

(5) J. Eggert, Z. Elektrochem. Z7, 547 (1921).

(6) R. Robertson, J. Chem. Soc. 119, 1 (1921).

(7) W. Taylor arnd C. Rankenbach, J. Frank. Inst. 204, 369 (1927).

(8) G. Tammann and C. Kroger, Z. Anorg. Ailgem. Chem. 169.,1 (1928).

(9) W. Taylor and A. Weale, Proc. Roy. Soc. A 138, 92 (1932).

(10) L. Wohler and 0. Wenzelberg, Angew. Chem. 46, 173 (1933).

(11) H. Muraour, Mem. Artill. francais 12, 559 (1933).

(12) M. N.rni, J. Soc. Ordn. and Expl. (Japan) 30, No. Z,106 (1936) Japan. J. Eng. 16, 6Z.

(13) J. D. Hopper, J. Frank. Inst. 225, 219 (1938).

(14) T. Urbanski, Z. Gee. Schiess-Sprengstoffw. 33, 41 (1938).

(15) T. Urbanski, Ibid. 33, 62 (1938).

(16) W. Taylor and A. Weale, Trans. Far. Soc. 34, 995 (1938).

1 1-107

115CONFIDENTIAL

-1

CONFIDENTIALNAVORD Report 4U36

t i.

I References (cont'd)

(17) T. Urbanald, Ibid. 34, 206 (1939).

(18) T. Urbaneki. Wladomosci Tech. Usbrojeni43. "#1?(1939);; I1 Mem. Artill. franc. 18, 1013 (1939).

(19) A. Berthmann, Chem. App. L7, Z43 (1940).

S"')) A. Stettbacher, Protar 8 , 81 (194Z).

"(Z1) D. P. MacDougail, OSRD Report .o. 804, August. 194Z.

4 (ZZ) R. W. Lawrence, OSRD Report No. 1288, March, 1943.

j I

*11.-10

iii

116CON FIDEWIZAI.

JrLi _ _ _ _-

a __ _ _ _ _ _ __ _ _ _ _ _

CONFID E."i'TAL

NAVORD Report 4236

-. TABLE OF CONTENTS

.". Report III

INTTR.ODUCTION

General Discussioa

~ APPARATUS

DiscussionIllustrations

"RESULTS

Summary of Data from Rcport of March 13, 1944Summary of Cornpiete Data for 42 Explosives

Graphical Treatrnent of DataSummary of Data ior Explosives of a Constant

1 Grist SizeGraphical Treatment of Data

Summary of Graphical 50% ExplosionDrop-Heights of 4Z Explosives and Seven ofCommon Grist Size

2 The Effect of Grit on the Sensitivity of RDX and TNTas Studied by Design No. 1Z

DiscussionActual DataGraphical Treatment

The Effect of Weight of Sample on Results"DiscussionActual DataGraphical Treatment

CONCLUSION

1174 CONFIDENTIAL

RE

CONFIDENTIALNAYORD Report 4236

REPORT 11

Bruceton, Pa.August 7, 1944

R eport to: Dr. E. H. Eyster

]rorm: Rage rs F. Davis

00' Subject: Supplement to Reports of March 13, 1944 andJuly 4, 1944 Concerning Bruceton Design No. 12for Studying the Behavior of Explosives to Impact

This present report is to discuss recently acquired data withDesign No. 1Z and also to present the probability plots of otheravailable data by the same design.

The procedure involved with Design No. 12 was discussed inthe report of Marc.h 13, 1944 (Copies: Dr. D.P. MacDougall andDr. J. C. Holtz); however, a brief review cannot be harmful. Themethod involves the use of a 1 1/4" diameter Ketos steel striker of3 1/Z - 3 11/16" in length and the usual Ketos anvils of 1 114" diameterand 2-2 I/2"in height. Illustrations of the design are shown inPlates I and II.

The sample of explosive, measured volumetrically by means ofa small stoup or spoon of 17- 18 mm 3 volume, is placed atop a IZ/"square of 5/0 Armour flint paper, which in turn is centered atop the1 1/4" diameter anvil. The large striker is pressed atop the flintpaper-explosive combination and tamped gently by a 1/2 cm. drop ofthe 2. 5 kilogram weight or drop-hammer. The tamping procedure is topresent a nearly constant surface of crystals to receive the impact.

Explosives are investigated by the so-called conventionalprocedure, i. a., at least 20 trials are carried out for various drop-heights to obtain 0-100% explosibility.

The practical elongated S-shaped curves of sone 42 explosiveswere shown in the March 13 report. These curves are obtained byplotting the average 1% explosions as a function of the logarithm of thedrop-height on semi-logarithmic coordinate paper. The newerinterpretation involving probability graph paper is shown in thepresent writing.f." 111-2

' 118ititsiI CONFIDENTIL&t I. -

rt . . .. 2 __ _ _" ..

CCO F-DENTIALNAVORD Report 4Z36

Table I show. in summary iorm. the data used in the March 13report. It is a phoLograph of the large table appearing in this previouswriting.

Tables 11 - V show actual data for 42 explosives investigatedduring August-October, 1943. These explosives are divided into fourclasses according to their seasiti,'eness. The classification is basednot entirely upon the 50% explosion drop-heights, but on the overallcurve for each explosive.

Table VI presents actual data for seven explosives of a commonparticle size. These seven are not to be compared with any of the42 other substances, as the particle sizes are much different. Thelatter seven explosives represent a coarse grained material and areto be compared with each other only.

Figures 1-8 show the probability plots of data from TablesU - V. The average 0/6 cxploaic"s ic. plotted as a function of thelogarithm of the drop-height. These values are obtained by treatingD or doubtful explosions as I/Z explosions (see page 11-83-91).

Figures II -17 show probability plots of data from Tables aI-Vobtained by treating doubtful explosions as failures, i. e., plotting% explosions only. These plots are made to reveal the 50% explosiondrop-heights only.

Figures 9, 10 and 18, 19 treat data from Table VI in anidentical manner.

• Table VUI summarizes the graphical 50916 explosion drop-heights

of 42 explosives as determined from probability plots of actual datafrom Tables II-V. Tw 50%, explosion heights are listed, namely,

tfor average I* explosions and for %Y explosions only (doubtfulexplosions being treated as failures to explode). These 50% explosionheights are also converted to an evaluation scale with TNT set as thestandard at 100 units. Table VILU shows a direct comparison of thetwo evaluated 50% explosion heights. It is seen that the same ordersand values (TNT values), with few exceptions, exist.

These comparisons were made because it has often been theopinion to note doubtful explosions, but to actually sum themasfailurss. To sum doubtfuls as explosions would change the practical

IW- 3'119

W CON•TMDENTIAM

#1


sensitivity order; and it is not fair or sensible from a practicalviewpoint, because doubtfuls are v low order "explosions".

Table IX shows the graphical data for the seven coarse grainedexplosives, with TNT evaluations applied and compared.

An early argument against Design No. ILZ was that itU could notd !be used to determine the effect of added grit upon the sensitivity of a

given explosive. It %as argued that the presence of grit in the5/0 flint paper wouild nullify any sensitizing effects of grit within theexplosive. A study of the effect of added grit upon the sensitivity of

•I RDX and TNT was recently completed with the employment of the* 4No. 12 design. It was encouraging to find that the presence of

>>1% grit was easily detected. Table X sunimarizes these data, whileFigures 20-23, 25-28 treat these data graphically. Figures 24 and29 show "he 50% average explosion drop-height of TNT and RDX plottedab a fAtaction of the % of grit in the explosive. The curves obtained areas expected, indicating a maximum sensitiveness is reached whenabout 20% grit is present. Two particle sizes of grit were used duringthe study, but both gave approximately the same sensitizing effect inmagnitude and general behavior. As the % of grit to cause greatestsensitiveness was exceeded, the curve again rises to an extrapolatedinfinite drop-height for 100% grit.

abov The 50% explosion height for TNT without grit is somewhatabove normal, but these data were obtained by a different observer thanfor the reference curve of Figure 8. Too, recent data indicate thatTNT Is apt to vary in its 50% explosion height from 75-95 cm.,depending upon the observer. The difficulty here is in the interpretationof certain trials as doubtful explosions or as partial explosions.

* Results from Design No. 12 are in general reproducible.Tabulated below are 50% explosion drop-heights for IZ Canadian samplesrecently tested compared with 50% explosion drop-heights taken fromTable VII. These explosives were examined by the same observer,although the samples were different.

" 4/' III-4

120, •CONFIDENTIALIIlk,

[.-- -

L.A _____um u I


Explosive Recent Data Table VU Data

PETN 8 9.5Blasting Gelatin 13 ---

RDX 15 I5NENO 13 23

" DINA Z5 25Pentolite 26 32SEDN.. 34 31.5

Tetryl 37 32Picric Acid 46 36MNO so 66.5Composition B 56 55a TNT 79 75

j "Another variable recently studied with Design No. 12 was the

effect of the weight of material tested upon the drop-heights necessaryto cause explosions. Data for thisatudy are . een foriWX Vf z=1nTable XI. Graphical treatment is seen in Figures 30, 32. Figures 31,33 show the 50% explosion drop-heights as a function of the weight of

material tested.

For TNT it is seen that lower 50% explosion heights are obtained

with smail samples; and as the layer of explosive becomes thicker, thedrop-height must be increased. The same effect for RDX was notdisplayed until >50 mg. charges were tested.

From data of Table XI it is seen that the greatest effect of theweight of sample appears at the <50% explosibility side of the curve,

while the effect disappears at drop-heights to cause >50-75% explosibility.

It was shown in the report of July 4, 1944 (see page 11-83-91) t)'at

reaction between ox-gen-rich explosives and the paper base of theA 5/0 flint paper does occur. Such a phenomenon should be considered

when studying this type of explosive compound. The use of metal discs

coated with nbrasive would undoubtedly eliminate these reactions. Suchprocedure is for future consideration.

Now that sufficient data are available, it may be said that DesignNo. 12 appears to be suitable for practically all solid explosives. The

III-S

, ~CONFIDENT'IAL,


order of sensitivity obtained by this design appears reasonable from

the practical viewpoint.

111-6

* i t

i4

11-

1 /Z

CONFIDENTIAJ..

7

CQNFIDENTIALNAVORD REPORT 4Z36

G~- S

' ! t c1 ii*.

AT.i

I I .

R

.4 L.

i ..

A a KETOS STEEL ANVIL OF I-I/4" DIAMETE&

C2 CHARGE OF EXPLOSIVE ATOP 1/2* SQUARE O? 5/0 FLINT PAPER

GO GUIDE RING FOR 1-I/4*0IAM[TER STRIKER

MR DEVICE TO CATCH DROP-HAWMER ON RESOUND

+ PLATE I

METHOD OF" LOADING FOR DESIGN NO.12

123CONFIDEONTIAL.

SP

-... . ,,, + _ __- _ ___ ___ . ....__


I Lr

-SS 4t-•

"1 N, A i.m

L $(TD STE kV 'O -14 l: gg

"i " - .... dI' " +"

,4 I Ai:+ ,.;

4SPIKE OR PIN TO PREVENT 3TRIEf[t'S ESC:APE[

I I

" ; FOLLOWING IMPACT FROM A RofOto Op3 IIS CIL.Ofrj " ;THE DROP- HAMME[R.

! • * I 1-1/4 ' DIAMI[TZPR. 4*1/2' LONG XITOSI STEE[L •lrRIKE111,Oft PLUNGER4

i W • 2.5 KILOGRAM OROP-NqAMMMEQ Cot WEiIGHT".

; N •Us ELECTROMAGNE[TIC DEVICE[ yo NOLC %No RE[LEA$SE, THt WE14"T OR OROP-14AMMEIL

It II OEVICZ TO CATC14 DROP-NIAMMEM ON# R[IOUlgo

j + l PL ATE 31i,[ THE DESIGN NO.i2 WITH CHARGE READY TO RECEIVE IMPACT

124C' D* IT"; €cONlFlOUNTtAL

I _ _ _ _ S*,

1I 0l l ii i "i' '

CODNFIDENTIALNAVORO REPORT 4256

o 0 Z 5

W# 4~ 1. 150 1 i .

'!4 1.6j 4

1-4 t - 44

S. .. toh 6o 1.6 10. to .

Pz * -3u : -32r

. 10. 3< !

Tt ace is . 1 4 s 1.6i 4.1 6 24 IS0'

t..8 Ar 9Vl ="~ *1 40 a 7i ZT I S .9.. U Z ~ 4 3.

39.* m 3~ I. 40'~ D SO 30 9 £4 .10 9 :0 '

La. 14 !. -KU .0%. R - 96 as 8 3. 4 0 a 38 j* 9 0 I 9 9 9 3 .0 3 0 £ .

4. 5. w0 3 P D IC ý 1O ]I -1 0 I9 I s .14 9:.4)9 .9 3 9 0 3 3 . 1 $ .a L 91 i s. 0

i. 6 3 "44 a 6 . 1.9 1 97 39 94. S3. 3 3 .9 39.0S s0 9 0 .3 0

I 3M it- .. 9 9.19 9. 9.0 S8 15 3 5.9 IQ.0 31 9 1.

It-79 I8.s 30 .1. 4 St * 4 3 . 3 9 £ 9 0 3 3 3

00 iS XD*A.0l"t A.&T4 '40.4 30 4. Is Is 9 .9 Is7.: * ~ 8 8: 3

33. Lw.'A. t713,3 14. II .9 9 1 so 1a. 33 30 9 9. . 1 4.

34. r79iM9 TRi~ 9 .3)Nl S4. 30 1.9 S ~ ~ 99 .0 37Is .9 3. 1 1 4.

£9~~~. Im.* P '~ I.0 0 01. 39 2. 4s i00 £0 7 90 £0 30 3 3 £90 £ 0 4

1.933 10.9 30 3.9 9 S 6 S 5.9 'Q.O so 79 II .8 30.0 01 10 7.0

"'s. 14 .1PW W f 1-99 S9- 33 9 S 7 Is 39 . 4 30 . 0 3 3 934. .Is 81 34 .0

34. 3n .9tr 4 w.93 £7.010 39 ).9 7 Is0 £0o.9 8. 03 300 40 3.9.1 00 I31 313 413.0

8a l9t 30R 11 169 14. 3.9 1

.69a. IO U to~S 1.0 . 8 9 * .9 l, -39 40 3i. 6.9 bS 310 t.9 .1

so S 0.a F e * O.~ 31 61 8. 10 3 .8 9 80 40) 31. 1 S .0 1? I00 40 3 .9. s . 9 331 33

19.e4 %0~ 4,.S N NO lo 91 ~ H : ' 4 94 3 00 39190 300 3 90 14

00.S-b p 99 . lol S.94 18.

to .

43. *aI~WU~ .919 331 34 3. 3 9?1.0 30.* 39 3044 l% 33. 300 1% 301 o

~ ~~3. ~~ ~f 0 s 3.00 £41 31 3I93s34 30 . ?S .98 3 133.- 0 3 3 ~

I T , U N OMis .9 a O9 I t s0 % 10 0cAo Da A 3 1 39

CON944I ft 77. 1

AýWmwdft ft"Wrate 06 'I "'R - 164 24.11I

IF 5 0 -4Z

a 4 11 4 ~ -- k M

I 4 ST. 4t 1 4 6 4 S 4 I 6

"1 0 1 4. I 0. "0 9 6 5 1's . **100~ a0 4 o 1 1. a '1 i0

1 .041 b Is 010

is .T cs., 0 .4 4 2 S toI 7 ý 6 os e if 312

S.+ ~ ~ ~ - T- Is. IbT ll OI-II I I l l~i I&C4 6 .0 Oj I 13 i£ 10k Il~ 0 =+l| *..0 0.-i I 0*.14~. 1. 0 * 70 IS I. is5124

.0 I T S.., 4.4 -n -i- - i A~ ýr '. a .1 6 Is as Al

I ' S . 0 6 1

i - 14.7 4.0 & 0 Is . .- A I ? 10. Ili M

:.. I 1 ". , 07 4 .1 I I&.4 Is so 1,4 1 is I is 4•0

I5 00 4 l.l 0 54 07 l 04 4.0. 1 09.4 40 l'Ih.i 0.430 II 3)l • 4I4

S i. .O 4 1 ISz1.0 ao 3 as It 40 . . 4 .9 GA J•

7 sIa ; I I Z9 iiis 4.9 I0.0 SO 1 0 9.9 I0 II 1,1 9.I 8.441 101 31 7 4

"*0i s 0 4 41 v 4.1 t7 13 0 01 ST .04 10 7.i .&- m to1 140IS 9 4S.• 10 so IS a t.0 I .0 7$ to. I 2 4 0.1•4 1 .1, i 11 141

00 V9 11.0 11 so 4.0 I t.0 4 4I a S IS -1 . 41 .0 11.1 S 22.0 7 . 07 I 3 i 0.4 . 00 40 1. b4 Z. * 00• 40 at 17

* 1 4 0.0 04 4 0'1 0 0 It. 00. 0 1 07.4 0 I.I I.9 0A ;NO

0 t I. 14 0 0 4210 04. 4 00 4.' Ill .49.1 4.00 14 42.114 a S7 a % 1)2.q S4S " 1, 59 4-.0 I S 04 w I0 7 0 1 e4 9 1 4 s Is -33 44 %l l

7.9 M 1 I S1 I It bT~ 4!. 0 16. i It ILL its I 0 :'.1 4 .• 14 6z- S 64,b o10 7.9 1 S0 Is 47 46.9 1 90.. 9 I 40 10i %4 4.8 4 I 1 4 711 64 4.4

)a 11.0 I4.0 1& L0O to. 0 10.4 29 ? 1. I1 ". 4.10 05 411 46C 96. AS atIuA to .M 113.1 0.4.)

I4 S.9 $ 0 00 0 44.3 0 1..1 Zi 111.11 4. 4i.s0 0. 3. 14 .1 14S

)3 0.8 11i - 1 4.0 0 1 S It s |W 4. a 4. o0 14 9 ' .T 1I0ll .1 008 10.0 *l 0 i 000 108T.0 04 1.7 30 914? 014 ,I 4' .741 0I8 87-.9 0 440

1* 9, 4 S.0 II . I 1 .1 04 41.0 , 01 " R,4 1. 1 31 In 0? 000N4 11.8 1 4 141 000 61.0 Z41, *4 1 119 0.3 4.10 a4 48.1 14 a3a04 0I 8 000 0 0 S0 401 it 0 14 1 0 (411 1. 11 4.0 IS 004 95% C b 4O•1 IS 1 . Tom"

04 10.1 I0S.0 U 100 70. 1 *7.6 is 07.4 1 IT%-i+ Uto 11. ?A Z;

0 .1 Ill i, 10.0 AV IN ". 4 .4 s6 T

O I0 1 1. 1 11 1" I0.0 . 1 " 0.0 T0oW4I1 0.1 0 190 Is 0 .11 Z0.0 ,4 I1 09 IS 7 ! .5 j 4.7l 1460 IS. i 4 3 " 41

A a IS -%. v 4. u it to >) " t 6 1 14 1,. . T 'rsee, 1 40 04 434 103 &A •16 am

40 001 As. * 09 9 4).0 46 1. 1 1 80 " ,000 I. oaI m I. 10 a" 0.ts z L+I4'm40 It.? I S.0 9 M i ts 44.4 Is :1 l l0 .4L1. I . as 040 " 0 usi

a 0.14 9 0 0 90 34 A111 =l1 as4.0 0 >is* 44 0 00 17. 6% a"sit *M.S II? ;ll, I I *O~ II I. >l'',~,l 1 m1t )l 1414 3.0,1 Is jo 90 10J*6l A-* ammh

:is.4 10. 11 IS.$ >1, 1 1I "" . 1 A 411000 3 .4g a 341 .

IMAS O" mIciaICA•7T DATA I1 IAMACT 011IGN NO. II

*... .. .... . .

*• ,. i. . , .... _ -J,' -• •.

NAVORD 4E

2 4 681Cocmen Name or Ave. Ave. Ave. Ave.

E;xnlcslve D' _ N % E N E DIi

Erythrttol Tetra- 0 0 20 0 o 5 1 i z 27.5 17 0 3 85 20 0 0 100 20altrate

Tetracene 0 020 0 4 0 16 20 15 0 5 (5 19

Xitro•-•.-te 0 0 40 0 :5 0 25 37.5 27 0 13 67.5 39 0 1 ,-7.51 39

PETN . - I"P,., 0 40 0 3 0 .57 7.5 9 0 31 e2.5 30

*C -Wa-RDX 0 0 20 0 2 3 15 17.5 3 6 11 30 6 7 7 1:1.5 10

Anhydroenneahep- 0 0 20 0 a 1 19 2.5 0 3 17 7.5tttol Pentant-trate L

.6 t8 10 12 TComm-on aria of Ave. IAve. Ave. Ave.

Ex~~v D H LE E D N LF, E L N IE E

50/50 P-E'T- 0 0 20 0 0 19 5 0 2 1S 5 4 5 11. 2.5 8Ft won iteI

0 u 40 0 3 36 6.J 7 1 32 18.8 5 2 3 1 1>

RDX 0 0 0 0 2 0 58 5 5 0 55 S.3 10 3 21 b 20

sE+;O 0 0 40 0 0 2 38 .5 0 6 34 7.5 4

DINA 0 0 4o 0 0 1 39 1.3 1 2 37 5 _

---- to

. •

• +..+ ,•l.eT1l•.,l+

i I

1 t

_ 2 14 I 16 jA v:-! Ave. Ave. e

%F_____ D N L E D N E N %E

"0 100

. 0 ' 5 20 0 0 1 o0

""7.5 , 6 0 2 95 20 0 0 100 _

- 0 ,0 50 34 0 6 05 38 o 2 95 39 0 1 97.5 N

S& 65 :. 72.-ý L 2 7 ',2.5 15 4 1 85 17,

5 13 6 5 9 4~2.5jo j

15 20 2.5 30 35 -, Ave, Ave. Ave. Ave. IAve.

; . 8 6 6 5£ 5 65 15 3 2 62.5 1,) 1 0 97.5 20 0 0 100

( 5.8 20 1) 31 40.8 50 5 3 65.8 54 3 3 92.5 55 5 0 95.8 z-

•4 13 23 26.3 8 23 313 12 21 7 56.3 21 14 5 70 26 14I 0 82.5

1 5 4 8.8 9 8 23 32.5 17 8 15 52.5 17 11 12 56.3 19 11 io 61.3 27)

TABL- I I

SUVWARY Of DESIGN 12 DATA FOR CLASS I EXPW[VZESDAQP-HEIGUT O 3.5 XII4)GRAW H•MR IN Cu.

II#

( % ""milli

I I8 10 1" r: •Ave. Ave. Ave. Ave. Ave.

0 100 20 0 0 100

0 5 75 19 0 1 95 20 0 0 100

:0 1 9(.5 39 0 1 97.5 ,S 0 2 95 20 0 0 100

S0 31 2-.5 30 0 30 50 34 0 6 85 38 0 2 95 39 0 1

7 7 47.5 10 6 4 65 Wi T 2 72.5 11 2 7 72.5 15 4 1

3 17 7.5 2 5 13 22.5 659 42.5

12 15 20 1 25 30 35Ave. Ave. Ave. Ave. Ave.

D E D N E D N LE E D N LE E D N LE E D N

15 1 32.5 8 6 6 55 11 4 5 65 15 3 2 82.5 19 1 0 97.5 20 0 0

2 33 15 17 3 204o6.3 •2 9 2 83.8 34 4 2 90 40o 0 0 i00

3 27 2a8.8 20 9 31 40.8 50 3 3 6 5.8 54 3 3 92.5 55 5 0 95.8

_ 13_ 23 26.3 1 9 23 31.3 12 21 7 56.3 21 14 5 70 26 14 0

1 5 34 8.8 9 8 23 32.5 17 8 15 52.5 17 11 12 56.3 19 ii i0

TA3LZ I I

I SUXfXY O DESIGN .2 DATA FOR CLASS I EXPLOSIVES

DROP-EIGHT OF r2.3 KILOGRAM HAOLA R IN Cu.

'II

)

v" ". 2 A'. ve. Ave. 30 Ave. Ave.E D N Av. E D N'e LE E )N D

39 0 1 97.5 20 0 0 100

15 4, 1 15 17 3 0 92.5

12.5 0 7 3 67.5 14 6 0 85 IT 3 0 92.5 20 0 0 100

7eye. D Ave. Ave. 50 Ave.LE E D N :!: K D N! 5 E D N E D m LE ________

7..5 20 0 0 100

100 iC 0 0 100

95.8 20 0 0 100

70 26 14 0 o2.5 20 0 0 100 20 0 0 100

56.3 19 11 10 61.3 27 7 6 76.3 33 6 1 90 2 0 0 100

.1_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ A

-- A

C¢NPONENTIALHAVORD REPORT 423S

8- 10z• 20 25Comoc kuee or8 Ave. Ave. Ae ve.Av.

_____ 0 NL, I N ISE

5o- PC'0X- 0 0 20 0 0 1 19 2.5 2 2 16 "5 6 4 )0 40 8 4 8 50

50/50="-- 0 0 40 0 0 1 39 1. 3 0 5 35 6.3 1 10 29 15?1voo4.¢

Baraa., 0 0 20 0 1 1 is 7.5 3 5 12 2Z1.5 8 210 o 45

EDNA oo4 0 0 14 2 37 5 1 3 36 6.3 8 9 23 31.

io 1 il 25 20 'Coum.0. 3 i6 of Av*. Ave. Ave

, lvutte. 0 0 40 0 6 9 25 26.3 14 11 15 48.8 239• 8 68.-.-,..- °

To.pezI 0 0 20 0 0 4 16 10 2 41* 20 9 5 6 57.

Tetryl 00 o 40 0 0 5 35 6.30 8 32,10 3 7 30 16.3 9 16 15 42.

Beot 0 0 40 0 20 3 8 5 1 1 35 3.8 4 5 31 16.3 7 7 26 -,6.

N%14 C1c0 0 0 20 0 1 2 17 10 6 5 7 52.

50/5•O entoltte 0 0 40 0 0 8 32 10 3 12 45 1 5 14 21 30 18 11 31 39.

_ _ _o_10 20 !0o _ _ ....Comisoa Name 07 Ave. Ave. Ave. Ave. Ave.Eyrlcmtve 2 0 N LE £0 N X fE Z 0 N__MLE

,ouiatte fe ro- 0 0 20 0 0 1 is 2.5 5 12 32.5 8 7 5 57.

sizet.ts 0 0 1 0 0 0 1 39 1.3 F 6 31 15 4 13 23 26.) 13 11 16 46.

I CON-IA

I GNF IOrNTIAI.

"4.'I

_Ave. Ae. Ave. V a. .Ave. ,E E 4_1 E • D _N E • D N i . N 9 .• iz D N_ f! z LE f D N

8 8 fl0 13 6 4 75 13 3•4 72.51 5 1 82.5 IT 3 0 92.5 19 2 0 91

12 15 1) 8.a 16 159 58.8 la1 la ý67.5 50 10 0 9I e 2 .0 145 16 22 65 la81 1 92.5116 1 3 82.526 o1 65 9 1 0V

8 923 31.3 16 1o 1, 52.51 6 15 58.8 21 12 7 67.5 24 11 5 T3.8 130 7 3 83.8 39 1 0

,,e, Ave. tt r _ i A N ýE E D -4 %E Itg D m 1

23 9 8 6..8 22 11 7 68.8 28 120 65 33 4 86.3

9 5 6 57.5 1436362 80 11 T 2 72.5 181 1 92.5

5 9 6 15 42.5•22 9 9 66.3 23 11 6 T7.3 5 13 2 78.8 31 9 0 88.8 33 T 0 91.3 36 2 0

3 7 T 26 26-3 16 8 16 50 2, 6 10 67.5 31 • 5 83.8 3 6 1•3

6 5 7 52.5 t 1 .5 65 15 2 3 80 16 3 1 87.5 18 2 0 5

18 11 31 39.2 18 10 12 57.5 28 21 11 64.2 23 14 2 77.5 460 19 1 82.5 55 5 0 95.1 20 0 0 1

Ave ve. Ave. A.ve. Ave.

5 8 7 5 5T.5 9 ' A6 57.5 It 3 6 62.5 15 46 1 85 18 1 1

3 1), 116 4663 1 •5•131 2 53,81 S 14 8 62.5 22 12 6 70 21 13 6 68.8 9 8 9 3

TANJ I IlI

SWEMARY OF VISIOG 12 DATA FOR CLASS It IMPL40SIVTIDROP-UIGE? Or 2.5 KILOGRAM NAMNNS IN Cm.

,! I II I

Ave. •v. •.e.Ae.' Ave. Ae

D _ _ .. x N 9-- E . E D W SE D N LE E DD

4 1@ 4O 4 8 50 13 4 3) 75 13 3• 4 T2.5 1,4 5 1 82.59 17 3b

10 29 15 12 15 13 - 5.1

5 12 27.5 8 2 10 45 16 2 2 Z i- S 1 92.5 16 1 8 62.5 260

S 36 6.3 3 23 31.3 16 10 14 52.5 16 15 9 52.e 21 12 7 67.•5 Z 11 5 73.8 30 7

SAVO. Awe. Av. 'Ave .' Ave. Ave.

23 9 8 68.8 22 11 7 65.6 261•2 o 65 33 3

S9 5 6 57.5 14 a 2 S0 11 7 2 72.5 18 1

7 30 16.3 9 16 15 £2.5 22 9 9 6•-.3 23 11 6 71.3 5 13 2 7T.6 31 9 0 68.6 33 7

5 31 16.3 7 7 26 26.3 i.E a 16 s0 2£ 6 10 67.5 31 4

2 17 10 8 5 7 52.5 it 4 5 65 15 2 3 o80 16,

14 21 10 11 1 9. 1 1 1 5;.5 2321± 11 6.2Z 4 7.5%0 19 1 82.5 55 5

IO £o __ __ _ 6o0 I .

- 40 AV 1;. Ave. AVe. 70 re. 7 ve.D D I D E E D 9 lZ I D 4L ON D! E 0

12 32.5 8 5 57.15 9 7 4 57.5 11 3 6 62.5 15 4 1 85

f 1323 26.3 11 16 £6.3 15 13 12 5,.•8 14 8 62., 22 12 6 70 21 13

TAME III

XMART Of DS1.GI 12 DTA FOR CLA£S It W iTIUU£ OP..IGrT OF 2.5 C ILOGRAi 2.1tKI 1A Co.

A -

AVe. Ave. Ave. e. .We.E~ ~ D H %E D N L_ E £ % 0 9 U Q N%

I.T ) 0 92.5 L' Z 09S5 20 0 0 100

18 8 4 67.5 -0 091.7 40 3 0 t•

26 ,1465 1i 1 09.5 20 0 0 0 Ic

4 11 5 73.6 30 7 3 83.8 3 1 0 98.8 20 C 0 100

I o 60 -0o 5 ...... •so 100Ve Ave. Ve. tw. .* A.'. o;. Av,*

12 D ,; : ý : E: N z Va AE £ 3 N %E E D N %E26 12 0 65 33 3 86.3 20 a 0 10C 10 0 0 100

"1 7 2 72.5 18 1 1 92.5 1S 2 0 95 20 0 0 100

.6 5 9 0 88.6 33 7 0 91.3 8 2 0 97.5 20 0 0 100

24 6 10 61.5 31 4 5 83.8 36 1 3 90.8 2 b 0 95 20 0 0 100

15 2 3 80 16 3 1 87.5 18 2 0 95 19 1 0 97.5 20 0 0 100

.5 40 19 1 82.5 55 5 o 95.1 J20 0 0 100 20 0 0 100

I e o j C _ o_______,.._SVAve AV*. 100 we. 125 Av,. Ave.

9 D N %& z eE 2 !3 L-- _ E DM___

Js 4 1 85 16 1 . 92.5 19 1 0 57.5 20 0 0 100

21 13 6 68.6 29• 6 3 82.5 27 10 , 60 17 5 0 2.5l

'*'. J. IT"-

a- -- ,..

COMOEND•?TIALNA•VORD REPORT 4236

10 1 20 1Co mona .e of O e AVe. ATK. Ave. Ave.

atnol2 0 020 a 1 0a 9 5 1 0 19 5 1 t15 22.5 11 1 8 57.5

?Per Acid 00 40 0 2 3 35 8.8 11 11 38 27. 5 13 17 30 )5.8 17 21 22 45.8

50/50 OSJlAtl 00 40 0 0 2 38 2.5 0 1 39 1.3 6 12 22•30 10 10203 7.5

_tntro____t__- 0 0 20 0 0 1 19 2.5 3 2 15 20

75/25 Tetrytol a 0 040 0 14 36 5 a 6 TI 2 ;8 4 18i5o 21 4 15 57.5

Compositon 0 0 0 0 0 456 3.3j A6 4o 20 4 20 36 23.3

Common Kaa of Ave. ATe. 2 A0 5 Ave.I ~ ~ q, D N 9K D N N 1)~~I 0 K K 0 N~

Eetrtnltronplt•a-O 0 20 0 0 1 19 2.50 317 7.5 1 3 16 12.5 8 4 8 50lene

Trtn•tmrbenzeon 0 0 20 0 0 1 19 2.5 6 3 11 3T.5 2 13 30 10 0o10 50

S50/50 AamtOl 0 0 20 0 03 17 7.5 3 6 11 3 5 12 27.5 6 3 11 37.5

British Coqgo- 0 0 40 0 0 1 39 1.3 0 8 32 10 83 9 23 1.3 11 19 30 34.2* ~s iti on

12 ____PO_____30 40

Common Name of )Ave. Ave2.2.5

fat _ _ _ _-W. _ _ _ t IRXMIDoo 0 0 o_0_ C 39 1.3 0 1 39 5.3 35 1 16.3 6 10 2 27e.5

Co Mamo ;F._o Ave- AV*. Avre. AV*. Ave.

0 0 40 0 0 , 1.3 0 I36 3 a 7 51 9.2 , 12 65 11.,

. .- - n+• ; .+ . . .,-• 41I- -" fl Jlll I

itt...

b. ,,so , 60 70 ,s• o.Ae. Aev. Ave Ave. Ave.e

D. .I N IF D 1l D ! N %9 a DR 2 E !~

1. 8 57.5 It 1 8 51.s 13 4 3 75

21 22L 5 .8 24 21 15 57.5 14 2 65 1 T 2 90.8 19 10 9r.5

10 2 37.5 a 16 16 40 11 12 17 4;i.5 17 11 12 56.3 20 12 8 65 33 5 2 8..

S13 22.5 58 7 .5 11 67.5

A15 57.5 23 7 106. 2Ji 7 77 5 29 8 3 82.S 3_____ 6$ 6092.51IS2 0 95

23 36 23.3ý 1 20 42 35 j 2') 17 23 47.51 2719 14 60.8 _____ 18 -857T6.7150 91 g

Tw Ave.6 T AT*. 75 "; T" AV00SD W %3N zve D eDma

I 4 850 s 12 32.5 13 1 6 6T.5 15 4 1 85

0 010 50 5 3 9 47.-5 11 09 55 11 18 57.5 15 4 1 85 15 3 2 82.~

6311it37.-5 6 86 5o 9 7 4 62.1

1 19 30 34.2 20 20 20 50 41 i

4" so To _o 90 100i'.•e. [A've. ' Ave;,e. Av - &' ".Ave. Ave.

________L_ 2 i LiA6 1 o 2 27.5 11. 8 21 37.5 12 9 19 4 j-l 1 9 17 86.3 5 6 9 4o 9 6 5 6o, 11 4 5 65 1

a so Avg. 7 AVe, Ave. v•, Ave. Ave.

9D 0l 1! *1 Z U D N Z 3D 13 DM.

312 65 11.3 15 24 81 22.5 ~35 .3, 88 56 19 25 65.541 12 7 16-4 41 9 6 $2.3,59 1 0 99.~

mJ•|•rt 4I .8 SICFI8•I

-I - .- - - i -- i l TAU S IIT

"MIA OW ,WG 12,,,wC~S I ~p~

I

A--ye. K. - - r.io Ave. AA£ D 4 U DN E DH

1t 1 8 57.5 11 1 8 57.5 13 I4 3 75

17 21 22 45.8 24 21 15 57.5 44 14 2 85 51 7 2 9o.8 I1 1 0

10 10 20 37.5 8 16 16 40 11 12 17 142.5 17 It 12 56.3 20 12 8 1

3 2 15 20 2 5 13 22.5 5 8 7 45 115t4 67.5

21 4 15 57.5 23 7 1 i 66.3 29147 77.5 29 8 3 e2.5 31 6 0

S20 o !5 '23.3 R O• 1 •412 i 2 17 • .23 1 7. 2T 19 14 60.8 _7 i8 i

AVe. Ave.9 D D KN %L g~N 2 gL2 %E

"8 4 8 50 5 3 12 32A 13 1 6 67.5

S10 0 10 50 8 3 9 4"4.5 11 0 9 55 11 1 8 57.5 15 4 1

6 3 11 37.5 6 8 6 50

11 19 30 34.2 20 20 20 50

Ave. Ave. AV.. Ave. Ave. A

"6 1024 27.5 11 8 21 37.5 1i 1'x 9 17 6.3 5 6 9 140 9 65

14 50 79, • s ' 10 , j IS"' OA•__. Av_. Ave. Ave. AV;- A

__ __ _ D___I____ Iz g_ __ _ I v_ _ _ a __ _ _ -e t

________ £ D W~ E ?AM& ITN D ~ 1 D

SW Or INJISIG 1 DATIA 70a ' U I .II 11 778l4'l9

Oin~OPIGNT W 3 .101143 .IAM RAlNCkl 1I CO

in-nwi0 .5u "am31 NC

' .

A.A© ve. Ave. Ae

so_______ 10 120 8.15 2 1 90 20 0 0 100

1; 1 0 W-.5 is 1 1 92.5 20 o 0 loo

7.5 13 & 3* 75 20 0 0 100

3, 6 0 92.5 .8 2 o 95 1 0 97.5 20 0 0 100

.Y7 18 S 76.71 50 9 1 90.8 56 4 0 96.7 20 0 0 100

• •. vAv. Ave. Ave. AVe.

15 4 1 85 14 , 2 80 20 0 0 0oo

)7.5 15 1 d 5 15 3 2 e2.5 13 2 5 TO 9 1 0 95

9 7 4 62.5 15 : 2 62.5 19 1 0 9 5 20 0 0 100

34 16 8 7-.6 33 4 3 87.5 3T 3 0 96.3 20 0 a 100

i~,1 ii 125 17~We. Age. Av.. Ave. AveIwge Ave.

0 9 6 1 6o it 5 65 1 4 2 80 18 2 0 95 19 1 0 97.T5 ! 00 100

•',9. 175 "_

78., 45 9 6 62.5 59 0 0 10, a( o_ _ o _100

II- ~ ,.

.e u u uuu *|| | ~ i| Bl

CONFIDENTIALNAVORD REPORT 4Z36

Common Name o. AveI Ave. Aw.Ex0lost8e E D X %17 9 D I L9 K D N %Z E V S

Diamnonium 0 0 20 0 5 2 13 30 5 10 3T7.5 8 5 T 5Ednate

Common biame of" Ae. 3 Ave. Ave. •

Erplosive E ED I! ýE E D N E D N %E E 0D N

Amonium Pcrate 0 0 40 0 0 1 39 1.3 1 5 34 8.8 S T 45 1

30 4o 5o0, 75C,'ommon Naas of AVa. Ave. Ave. B

E~fotev ED - ED N L E D N ED N

Potaseium Chlo- 0 0 20 0 1 1 18 7.5 1 4 15 17.5 8 6 6rate

Amonium Nitrate 0 0 20 0 0 20 0 0 0 20 0 0 0 20

PotasstuaPer- - -"- - - 0 0 20 0 - -

chlorate

Uttrepantdine

Ouantdine Nitrate

CONFIDENTIAL

7..

.Mumm(

- 60 T0 80 .........2 12Ave."a Ave. Ave.Av. Ave.

I EID I I

M ýE E D N 4- E D if 9 N j D LE E

0 37.5 8 5 7 52.519 4 T55 115 4 67.512 3 5 6T.5j13 4i 3 75

so__ __ _ 75 ,100lo 125 - VAve. AVG. Ave. ;,Va.

E ~ D N E! D W ýE ED N LE Z D ~~E

A8.6j8 7 45 1.3.2 2613 21 511.2 30 14i 16 6 1 .j 3 8 16 6 76.T E42 15 5 80.8

75 100 -15i5

Ave Ave Ave. Ave. Ae

x c EDN47 E L E0Mý E D H 19 E D )

15 17.5 8 6 655 659 42.5 114 5 65 12 1. T52.5FU k 5 65

20 0 0 a 20O 0 1 1 i8 7.5 - - - - T 0 13 35 -- -

20 0 ---- 0 1 19 2.5 - - - - 2 018s10 - --

0 0 10 0 - - - - - - - - - -

00 10 0 - - -- -- ---

TAXZLKT

SWQ'sRY 0F DES10N 12 DATA FOR CLAD.S IV EXPLOSIVES

DROP-MKZLiIT Of 2.5 KXWOOR AMM1OE IN Coo

Ave. Ave. -0 Ave.800

4c o 75 100 -2

Ave__ Ave Ave.L*N E E D N %E: { D N 5 11 E 6751 X 5 6E. 13 ý

Ae 5 Ave. 10 Ae. 15Ave. Aý ve.LDN E E D X %LE E D N LE E

153 8.85 8 67655. 2642 5 9 42.511 415 65127316 6762.7 11

070 10 0 - -5 -

E• D

0 20 0 0 0 2~0 0 1 0 108 . - - - 7 0 13 5 -

SLWqAY OFP DESIGN 12DATA FOR CLASS 1V EXPW31VESI DROP-HEIGHT Of 2.15 KILOGRA14 HAMMER IN Co.

Mm

- 1

joo125 15o 200 o

.Ave. Ave 1 Ave. Ave. Ave, 30 Ave.-~ ~ liL E D N f.rE 0 L F EV-NM

567.5 13 4 3 1- 12 3 5 67.5 10 6 65 6 3 1 75 9 1 0 95

125 . 15o 175 . 200 225 300Ave. A%,. Ave. Ave. Ave.

N g D N - ESD H ý!.-g-9D N %I A 2 g1 LS_ _

676.7 42 15 5 80.8 35 4 1 92.1 18 1 1 92.51 _ 17 2 1 90 19 1 0 97.5

150 175 200 250 300 3Ave. Ave. Ave. Ave. Avg.

E D LE fK -V N E D N tq 3 D N E D N

T7 62.5 11 4 5 65 13 1 6 67.5 17 1 2 87.5 IT 1 2 87.5 10 0 0 100

-13 35 - - - 7 T 9 45 - - - - 11 36 62.5 14 3 3 7 7.5

1 - - - - 6 1 13 32.5 3 5 12 27.5 7 1 12 37.5

.. . . 0 1 9 5 1 0 9 10

. . . . .- 0 0 10 0 .. . . . . .- 0 0 5 0

PLO31VES

w CI..

NAVORJ RU-ORT 4236

Ct J! - 0

I;

o H4'*

M-1

100CON;1 IA

CONFIDENTIALNA VORD Report 4A36

z

O >

w xa w

~0

jzZ :ý n- f nC

0INIA

C ON F IDE£NTI AL

NAVORO Report 4ZJ6

oz:

z A A

Ita lo

0

0 0

o -

0 z%

.0

14

-0 ... .0.

z 0.

0 z

z~ E 7

3 4iiZ .w jjZU

fU

0 a

o4. ZimU

I-j

Ol uj o Z : E~ C , z ;

IIr

CONFIDtNTIAL

CO NFLiDENT IA LI ~ NAVORD Report 41.36

0U

- - - --- -- ----

o Z.

'-I f

3%-

A 0

a 0

=XC 0 & 0E~w a.w

w z .- 6 a 0. z %m .wi. m z 3

.9Is * I *corI0.IA

CON FIDEN TIAL.

NAVORD R~port 4236

I6oz2 64

IZoo U f~

Jr "~ ~!Ap WIFA

0f -

a. Cbo.C.4ýv - 0

0~

4 _________0

F6

o A

00

-u - a S4 131c

.4 .! DC TA

0 *:

CONY )I DNTIA LNAVORD Rvport Z3

0 v

-

t -. 0

z

U..

4 0 .

CAA

3.3n uh

0 9axI-0

'Ilac 00.0MT A

CUN F!DEN TI AL

4 NAVORD Report 4-I6

Ir

IX

4%

x-to

-e

.2

.4r

a% 0

itO

. .2 .-.

M0 u

o •

t AA

)c

Sa i

I ° "

cj ! o r11) C N T IAto

CONFIOENTIAL.1 ~NAVORD REPORT 42226

I t a s a 4 4

I&&--Ito I

OAf a"elIl:cVC "AWC

2~~~~ -. P Lae.,lIA- 9.j~~t W. 44-... .... .. l A- .. 4D.4. .L... C -8,. .C- %.Ca '%c .. 4 se *4 %,.....1 .I ..... do

.96 be of & .1 _ _ _ _ s _ _ _ A 4 .606~*

441 16 90 1 1 9 I fIt 4ii 0*3 .4 n A4 on4. *

~ * * * - 91% 44 Me

14 1,ofaW, A *A.' .4 61.A.A el4 l 4

141 It III A lip .. 04, 111, AA* s

______~~6 1, .0011 &.a" 4..4 0 na ;

3 .4-., ~______ A ~ low

_0 _ __to

Si~17 0 '-F.EIN TLA-LNAVO:SD Report 4Z36

> u

z >

"I .

0-~ X 4

IN

L. Z>

U oUbu•C

0 . 0

,, E x0 i

00 00 > coQ a

z 13

N t i 0 1 0a

~tti

fE X0a >1 0 non% 0 .i0

*0 - 0 nar 0

P4 e4

0

CA - .2p - 0 t- fý a rl -l 00134{ ~CONFIENTIAL

CONFIDENTIALNAVOFR) REPORT 4236

o ~ 0

.. .... ... .

.I. ... jZ Z. .. .... . .. .... .

-r a. L .

IA...........

S. ..... ....0 0 ~ . 77 - -;

FA

-... ... .0 . .UIz

-7- a :t*.. .. , 0... <

I A.

d £ý

x o4 #4 o

b3nAYH ONI %' JO LM013H -dO0MG 001

135

CONFIOENTIAL

CONFIDER'TIALNAVORO REPORT 4236

..... . .. .......:: ..... .....

.. ........ w

::'A l 4.. .. .... ..I I x

7-: :7 t 0

-i ... ...

cc

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

0.1 7..: ... ..

a in 4f I

*~~..

... ..V~ .. .: .4.i

o J . I.. . .... ::::..A..

0z 0

CL

IL ........

*~ ~ ~~~ ... ..... . . . . . . . . . . . . . .

~ ...................................................... .... ... c

.... ....0 .... ......

r z 4& 4L Fb C

136

7 ------- - ..

CONIFIDOENTIALI NAVORO REPORT 4236

I 7F-- T--Azj:iP. . * i:Ii' _

... ............ I 0

.. . . . . . . . . . . . ....i-~Cl 4

- - ---- ---

I ~zo

Ha W

~r0!0

* 2I

IXZ .. . .. ...4.'W

*XCO... .... ....7~ ...

4 ed d3sq~ Q~ - TF 0HIM-OQ

V3"hw siiG'I4-03H-dWQ901

M-21137

*~Cot4FiDENTIAL.

IZ

CoNC IDE NTIALNAVORD REPORr46

Nl

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

U 1 L

-70

0 (

'9 - C.)

-~ - 20

~t *~ - -4! :wL4

1.7.

-~~.. .. al-dM 0

138JA

2~ 0

v* 0,-,

ý',ONFSNTIANAVORO rZEPORT 4236

'11711. ...0 0

1 -717.

Qr L-I

U,,

:7 .. LL.

Z i 0

o 0 Aiýi2 pt O

U). ý- 0 .

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

- - . - - . . - - _ _

on- -t,

--- - -

CONFIDENTI1AL

NAVORD REPO~RT 4236

* ~01

T0:: 7- TF

.. ........ t.-* r~0j-4..--j7~ -A, Ht~L

-0j

0O -- 0o n4

Z4in ftm.,~ e,

a 44#x za10 *I r o

ILO~OI I. 0

Itwu

-. 0

j z ~I3flV 9~g~ O H0IH-8~iU00

CONFIENTI>

~ -.--- -- _ ____ ___ ___ ___ ____ ___ ___ ____ ___ ___ __dw J Z

CONFIDENTfALNAVORD REPORT 4236

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

. .... * -

.... ... ...

a w: Wa- In f

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

4 :z::.:X:~-.------.--- - -. a1 :

Z ,- 0.

I- --- ------. - .log*. -

ta

uE~u N~ u ls~ u s3N~ ao2 CC~Ii.. -25l

.. - .. .- L

p. NAVORD REPORT 4236

.......... .. w1

¾~. ... ...- - 4......... .... -..

-71 47

S9-.

w ::: :TT.......... .

L-z

(0 ._. ...

- - ;.-:::il. 2.... .... L

7 I ... .... C/) L

.. .... .....

o W

j_ CCm 2

0 - -*

CONFIDENTIALNAVORD REPORr 4236

z

LAj'7 1T Tj.~ .1. 77

.. j. ~-. U'

r...t.... ... -7.1DI

1-(1

a..a~L

VI x i

- X 40 _____w

A% :::::..J

tit 1::: ____ I.:................................. .... Z

f-i.~~ie4 7 7-7 L....t

-72

143

CONFIDENTIAL

Wf I ~. A LNAVORD REPORr 4236

71

". .. .. " .. I .T .. ..

~~-1

ww

tW w

-N i ýOw ............. .. ...

.. ... ...

C.~~~~~U Uqj .H11 .0... K .. H .... ... .. . ..

; er

... ... > ~ t

e4 .... ..

CON FtDE!4TALNAVORO REPORT 4256

0

jc 5

-7 ----- .: : 7:: 2 ,. -

W

0 0

:-A244,7L-

ota* LL. a.

CO,,FIDO- -_,

NAVCPRD RE=Pz?"T 1236

S. . ;... ... . ... . . -- .. . . . . .--... . . .- :, ' : : - -. ,T• ,T T

...' . ......... U)

-. 4-

" - ... .....,:==. : := : E--6r-==-= = q :-.. = == == : 0, - ... , . .. . .

- - - . . . .* , .. . . .. C.. . . - - - " ... .. . ."".-

"".......................... ........ .

.J zz.~ 0~~i

i ~~t ... .... =4•~ ::l:l: Q ..L. ... .. ....... l--i =..o ... .. . .. 0

Sz :., .. ..... :.0•,•..•,...... x ... . ...... .• .. . " "0 •.. .,:: . : t-- .• ! . .. i- .... .... .... ..-, t. wZ 1 0.

-4cn-

, ! ., - -. •, .... _, 0

.... .... .... t :%lox

u .- IX

;:;7 W. .. ... .. .... I .. - . . ,,,J

, - -.. ... .. . .

0 >.

t1 T1 -: - - a-....... ...... .................... . .

a x&... .. .. 1 .. 1... -: :. .... .... . .S• 1 •,ll .... . . t C.. . L • ,*t* f

[IF a0N

,3y'IWvI cxGz Jo LHIIH-dovo Sol111-30

CONFIOENTIAL

i4

/ • •n I- -


... .. .. .... :T T;L---:- --- -t- 7!'-44 ..... .. ;

'.7 0;7f7F7

0 -n

2 ~ - l..... .... . .. > (

.. . . . . .

* I LaJeI~ k - -- -

crI:2~

b-O. ~' 7-2U ~

14=7

IL z

CNFl,1:E 1.1T IA LNAVORD REPORT 4236

5 .!a. -

tt

40

77 w

00

...-.--. ~ z

C4 -

.1 -

L cccs~IW

U .6, - ---------

4z -

"o '4

0-' O n o 4,b ... -- " I-

CON'Fi DENT INAVORD REPORT 4-36

d

a:

V .....Z

0 ..... _ _ _ _' ' . 2-.... • . ..

..... i~.-- ---------- -- I.:.-.-•-~-. _0-•.. . .. I - --

- - - .. . . .... . . -. . . ., " ' 4 -:- ..... _J . ___ L" iu

-,-1/ •:::-- _-_--_:-- _. -.-. - •--* -- .... -._• . Cr l

/- -9-- 0-'

-- - --- ---

UJ.4 • ...9- :2 • : .. .... ..__ _ _ __ ._ _

=••o ..:_,..• •-::f:::*:: ::*-:-.-::::: =

01 -1 7 (00<

o E . .. -• - -.. -.I. -_.---. .... ; ,- -,0 Z

x to J

0- .. . .. .. w) .C .

: : .. .. .. .... .. . .

- - ----------- -G6

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

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

r •t•: : "' i•' I l' : I I i " •

w a *A

z _ _ . . .... ... ..,- 0 --; • ::| -°. ; . -- - -... :77 .: ... ' .. _. ,.....e -- .

z- 7777 o . . / : :! : I . ... , l . ,, L.-

t3•IV 'at t' 01 -XIXd• 00

M-33 .149I

CONFIDENTIAL

NA'¢OR• 2:L, ,:'?Y 4236 Zo_0•0

; ,, , , ,, gt -- J"I ! I • i--•--- ;:I ! I I ..... , i,•ilz ! l• ,•,•" •-: : xS..... , .... t ........ -....... t.:--T--,-•--:-,- : :: ----: ,- --:• w=========== =========== ::::::::::::::::::::: • •,•.-::- i-; • i :i'.4 L

• -- " t--T'l

-r-"'t-7--r-" ! " i : " •::: ;• ; o • • - . I'-::::::::::::::::::::::::::::::: : : :• :::Ii:: -£: : :::•::-: ====================== • • [

" .::!:,: ::::'--:::" -:' ::::::::::::::::::::: •-•IIS............. ' ................ ' ....... -• •" ::':-] O J"

• ;-, x ,• m •-. 0 •..t."7•••.•, .....-. _• ....... . ........ ; ........ , .... •.• .- N • N --.. I , ""l'•:--'i .... ,;,""I" , ,..•o . -, 0'•'•,"I .... x-• .... • .... i•.-:L:---•-7•V-• i-

- ::i:::: \ ...... • "'• ::.'t::-•.:: <• ; o _jf:::: •-• ••".•:...-••

o [_:: .:::: ® x 0 e'O'"

S" : "•::i.'::°'•',". :'-.• .;'..•-:• .:.--:L•;:: :::---:::: :,':"

:I" $ .. . . -- • .....

- :::'::::;. ":i:; 0, • $

S:::';::::[::::i ::;:.'::: ..... S........... • • •j....: :',--:I::: , .......... :: :::::1:::.:•.x • ....... •.So

• . mS....... o ......... ... •

I•! •.,...,*--..... ''' , .... . .... , .... •, .... •, .... f ........ ., ::::i!!: !:!: ::::t::::t ....... . ...

::::i::::l .... I-- .. .............S• ': .

: ...... ::::::: :::::::' ,11 :1

S- .. . • N o - S " •q-' 8S5- •1111•1•1•1 1)1 II'1 • JLNgilN-dOIdQ •)0•1

S• CONFIOt[NTIAL

• " • .... :•T4•r"'•X"a• " .... F • . ... " % ................ -•

CONFIOENTIAL-

NAVORO f�EPORT 4226

02:

zi Z

C,

�I2IT: 1IL�___ t. ...J. V. 4 0

- I.,..

__ .- J

171: -J

S-. .--- �-- (I.)

o �

>�

4I I ___ -� j ___ � �5z

Z I A; * 1 i - J0o t I__

Sf \.- ____ t1.-S..--- I I--� ---- � Lr�

.%- 1 0 >2:w

$ -- �.-..-- I! I __ 'a

'ilL -. LLJW�

ILiti LL�I�.1.- L.....I-i.. 0

s-I - LtVL. .� .

I 4.�ffi' -- C)

a 0

___ * L�.I

N N - Li-i

I*3APWM �5*� G� iO LI4�a3M-dO�0 Sol

I�- 55

CO)NIDENTPAL U)NAVORD 'EPORT 4236

ppu

o - T-- -- t - • -T-Tr -'T- -. • •• r7T"TT' •T(.j.

In'

"_____.__.-____ 2 -.- ,-.

S. .... .. ..- .-...... toLU U

, I r+

77 Zl, zli•I 0- -e+

z . .

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

r:: i in:

_, -z

t__ __ _ 00..*

0.2, -NF ,-NIA

0 J"

.... . W C.'4- -7: i'.4:

If- _

... ... ... .... 7at :: . *: ý * . . .I- -- - - -

~. 152

.7..

* U)

__ __ __ __ _ _L.-__ _ ... ......... .'it Tw

CONFIDEtJT1AL

14AVORD REPORT 4236 C

w

71.7J -7-

.... ... ..U(a

_ -

. . . . . . . . . . . . . .. . . . . . . . . . . . . .UD

.~-:--7--.~.

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

t ul xa wf

; :~ .

j ~ 3- .. ... ). .. ..c 1.... ... ... ... ... . . ... ...

.. ... .... ... : :

SF

az

K-)?

'53 IA.CONFIMC~AL

CONFIDENTIALNAVORD REPO1RT&4236

I IS

qm 00

0 4 U

-- - -- L c

U* Z

z S

1i3VIAVH x) xZ jO ±HIHO 0

to31 ' Ac0 ---- --- CONFIENTIA

CL W

I -~. - . __' '


0a

...~ ~ ~ r.:.. ... ....-.. .. %6e .

00C 0 4A

SW uo-0~. ..... .. 'T' 0 0

cc

_____ Cs)

.. ... ... .. .

___0 z

I0 U

o an ------- 0

(LVVV TZ (Ld L~3Md~X -39

. ....... . M

DEE L.

CONFiCENTIALNAVORD REPORT 4236

-4 ------- ---

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

. ~~.... ......... ......... ........ _ 2

.. .. . . . . ..

-1 _ _7 7.. -. .. .. . . .. . . .

__J _ . Z*7 ..... .... .. .. I O C

4~. ..0.. .....1*.....................---:2IW . .. . .. ..

w,- Ca

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

-a 1+4, -11

404c:, .. 56

1..

CONS :DENTIALNAVORO REPORT 4236

........................... .... .... .... .... ...... ... .. .. L... ... .... ~F FT ~ j 2 .. ..I........

r ., ..... .. ..

~~~T A1(.. ...... .... Z

(A. -so 0 V Pi

40 0

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

It 'A.........-...

ub -

UJI 0

001

U. -. -

CON FMENTIAL*1 NAVORO REPORT 4235

.,2.

-4 U4

--

IT - HIHtdIOd3 o 0m-42-

*~~~1 CCFDETA

'p

cc' w

NAVORD REPORT 4236

S. . .. ..... ... ....- . .. I -- -.- .. .. -_..

L N

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

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

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

z... ... n77 r7L: -

4 0 I O

t49. j---_..

u,

.* 0

z200

0

QL .. , ::7:: :::: : ::::: :::::::: :::: : C3.... .........................................................,,,' ._.. I ,..: .X1 us :. '_~ - -

._. . . . . . . . . .44 CANF0NTal

2 I

0 0Q0

: 1 .. . . leg

'.... ... .. . .L.

., - \. h

CON F:DE NT "LNAVORO REPORT 4256

- - - - -- - -

1--~------ --- .

W I-

10

0 0>I 011____C_

L WC

... ....... 0xjN~. H-:_:::-\ ----------

I ~LzI i

* r~vi. .. . . ...-

*@ '.4

I.A~~i~~vgR..... ... .* wQj4 1 U d i 0

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

.... .... .... . . . 7 ,0.... .. C. ... O.... .

C*

CON"FIoENTIAl-NAVORD REPORT 4236

.x:1"H .-

"" I . . ." ... .. . • i -... . 1..t --

o 1 , . I 10 ; ; 'I 0 . .t .!

-- " '~~......... . ... ". .. .. . . . . • ,- : .

1 0 fa z:1. . ,. I . . . . . .: " :

- .. - .--- ...-.- -- .. ... ._.w=taJ :-:I z

ih I ;: '.' " t " I•'" : ' VJ .: l. . ! .... 4 ." --. . . -. -_ .a 0

p.-! : 1:.: .... Q... ... .. :::- • : . ... _ . . -.Lai z: 0

0, 0 ......

o •.: . ... . I - I ~ . .. ... .. ... ...- .. .... ... . . . _ o ,,Lii ... -4 0 0

= .-, :..- -..- .- .. .. . . .... ... .... .. . .... ... ... . . . .. ..SIL-- I-I.. . -! . • * "ow

]0-

- - - -

d - -

k13WVflH VX G' AD I.H013I4-dO~U 001

X-45161

CONFIDENTIAL

CONFIDENTIALNAVORO REPORT 4256

ty 0 Z

1*ILa I L~flu i

* ~ ~~~~ cc.. .:!2~ *T.

WJ uu

*d 0a5 z

10-.

l0eL

000

CONFiDE NTIALNAVORD RE.PC!7 4236

.............................- ~ --- 0

S..-~ ..-- .3

--- 7 --- ---

-4 _j -0

0~ Z,U.. Q

&-. -- LL

I-'

___ _a_ -W-.t:77 x-- .--- -~. _ _ _

00

w. 0 U____

$. L. . 4 w!. .!....

~ .4.Z:2p-77--7

m- 4. U

*~~_ ____ _ _ _

-, 41

NAVORO REPCR >

-7-

t

9z--

uh 000w

7i

4 -

w

iv; 7=7 ii ---- --30

~ 48 Li..

K - - . ......... . ......

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

CQUNFrnENrTAL%~AVORD REPORT 4236

_ _ L/ -~-I t z~u~. __ _ __ ___

-. 1 __________-~ I:1 __ ___ -

- _ _ _ _ ...

:rz::.:::~:if~ 4ILw_ _ _ _ _ _ j r7 ---

-7 ~ ~ ~ ~ - 0. z z f ~ ...

A-t - -

T. _ _7 _ _

165 U_CONFIDENTIAL

CONF'OIDI'~.L

NAVORO REPORT 4236

3i W

Ps C4~ a t

.. .... ..... * p 4 4

. .. .. . ... ..

i

- - .--- I- ~ ~ -fi

2 LI

t ~0 uNA 0 a .

s 4 a

=r .: ----- ld L

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

N7 7

---- ---- 7'

II Rle IIL l" .

ERE R

j~j dIanmo 97 Jo193_ 0

CON!FI DENTIA ULNAVORD REPORT 42360

47 4

-- -Do

T- j-

~.-~.-.4 _r ---

--.- -

FEZ_

q~~ - - 1-a_

x w

. *9 .!i.>

NAVORD '.Zport 4236

REPORT IV

Table of Contents

INTRODUCTION

Summa r-Introductory StatementsHistory of the ProblemStatement of the Problem at Bruceton

EXPERLMENTAL PROCEDURE

A. Developing a Sensitivity Test that Would Permit

a Definite Comparative Evaluation of allSolid Explosives

B. Developing Methods of Interpretating the Results

of the Impact Sensitivity Test

EXPERIMENTAL RESULTS

C. Developing a Sensitivity Test that Would Give

* i. Reproducible ResultsD. Developing an Impact T!st that Gave Results

in Agreement with Practice

CONCLUSIONS

A. General Th-oryB. Recommendations

* BIBLIOGRAPHY

I

I

SIV_-I

CONFIDENTIAL.

ON

.- i ' J *- _l i •- . . .

CONFIDENTIAL14AVORD Report 4Z36

REPORT IVBruceton, Pa.September Z8, 1945

Report to: Dr. Eugene H. Eyster

From: Rogers F. Davis

Subject: Concluding Discussion of the Problem ofthe Behavior of Explosives to Impact

I ~S,-u'ama ry

This. report serves to discuss a brief history of the impact3ensitivity problem at Brucetcn and conclusions reached after threeyears of developmental procedures. There are also discussedrecommendations which may ease the difficulties to be encountered infuture research on the problem.

* Introductory Statements

A survey of developmental work on the impact problem atBruceton was presented ;a two reports to Dr. E. H. Eyster (1). It isassumed that the reader is familiar with these reports in addition toOSRD Report No. 804 (2), a-.d numerous OSRD Interim Reports inwhich impact sensitivity is discussed by Drs. MacDougall and Eyster.

History o, the Problem

The general problem of the impact sensitivity of explosives isbuilt around the general query of: from a practical viewpoint, howsensitive to impact are explosives of commercial and militarypossibilities? This may be expanded to include the safety aspects asfar as manufacture, general handling around the manufacturing plantand in shipping, and the use that a particular explosive will be subjectedto if it is to be adopted by the Armed Forces.

To answer these problems, the so-called impact sensitivity testhas been developed in the course of the last 40 years (1). Unfortunaiely,there has also developed the general conce;3tion that a relatively simpleimpact test conducted on an explosiv-, will reveal its military andcommercial aspects from a safety point of view. This particularapproach is filse in that the overall safety evaluation of an explosive

IV-2169

CONFIDENTIAL

J ;

[, ii . . . . . • • ...•w.7,.

CONFIDE.' 71ALNAYORD Reýrt 4236

cannot be obtained through the medium of a small scale test whichcertainly does not reproduce all of the cuaditione of practice.

With the exception of bullet sensitivity tests, it in only duringthe past five years (3, 4, 5, 6) that the important idea of developinga particular test to reproduce field condi.tions has materialized.

In the final aaalysis, an impact m'achine can give an ordering in ad comparative sense of the behavior of exploaives to the particular impact

test conducted. We have found that by varying the conditions(confinement mainly) that vastly different comparative sensitivities are

bobtained.

I We quickly arrive at the general question as to why we have thesensitivity test. The real value of such a test is to obtain inform...tionconcerning the comparative behavior of explosives to a test that hasbeen conducted many times on many explosives. Such a test usually isthe result of a period of development, and the comparative order ofexplosives by this test is known. Likewise, the impact test is developedalong lines to give orders of sensitivity which are in reasonably goodagreement with practice. For instance, from & practical viewpointwe would consider an impact test as misguiding .f we found that Mercury

Fulminate were more insensitive than solid TNT or that Tetryl were moreinsensitive than Explosive D. We know the safety history of these

* substances from practice. Thus, if we can develop a small scale testthat will give us a practical ordering of the sensitivity of explosives, wehave a preliminary criterion to evaluate the safety of the material. Thisis indeed valuable in that long range developmental procedure on a givenexplosive will not be attempted if initial tests indicate that it would beunwise from a safety viewpoint.

Statement of the Problem

At Bruceton we have been confronted with four general aspects

S of the impact sensitivity problem:

.' A. To develop a sensitivity test that would permit a definite

S comparative evaluation of al) solid explosives.

B. To develop methods of interpreting the results obtained.

"C. To develop a test that would give reproducible evaluations.

I';. IV-1t •170

I CONFIDENTILk


D. To develop a sensitivity test that would agree with practicein its evaluations.

Experimental Procedure

A. Developing a sensitivity test that would permit a defi-'itecomparative evaluation of all solid explosives:

Impact on an explosive may be direct or indirect. Fcor the problemat Bruceton, we have employed indirect impact, i.e., the small amount(35 mg.) of explosive is placed between two parallel metallic surfacesand struck indirectly by means of a falling body of known mass which hitsthe upper of the two parallel surfaces. The procedure involving indirectimpact is advantageous in that the metallic surfaces are subjected todeformation from impacts and explosions, and frequently must be replaced;thus a simple piston-anvil arrangement permits a more careful control ofsurfaces. Too, hits are more reproducible by using indirect impact in thata falling body does not always follow the same downward path; but thedeviation is minimized by having the weight striker a rigidly ' eld piston(striker or plunger) - anvil combination.

Table I and Figures 1-7 summarize the important types of piston-anvil combinations which were developed over a perio,' of three yearsat Bruceton.

The general conclusion regarding surfaces is that unless impactenergy be concentrated over a reasonably small area of non-flowingexplosive, the chances are that no or wveak explosions will occur. Fromthe surfaces discissed in Table I, we can see that the atove principleis nearly achieved in only one design, namely that of resting the explosiveupon abrasive materials.

"A B. Developing methods of interpreting the results of the impactsensitivity test.

A discussion of the so-called "Bruceton Up and Down"abbreviated and the conventional interpretation of resulte is to be found inOSRD Report No. 804 and in report to Dr. E.H. Eyster from this writer.

A newer statistical interpretation is discussed in the AppliedMathematics Panel Report of July, 1944 (7). This report treats statisticalmethods of calculating the height of fall to produce explosions with &probability of 0.01, 0. 10, 0.50, and 0.90 from a mnininunt of 100 trialsper explosive.

IV-4171

CONFIDENTIAL

/i

CAOORPON~T 4236

SUMMARY Of OLPO&IAZT T1P9S or vwmis-Amm.UQW0112'r-rtar's vZXLzC-vr rcm mx !IMPACT ?g1 AT BAUGEW6

NRe... se 1.4.~. a I ý S - 1T. . s.I. U

i fl~~w.ommd Le ... p.. d..av

-4fl.6n M, prCa. -.. At.M

pA6. 1. . A-. W .p ta.* Mk.Sig

sm. taA. at -SQ *..a-i 4at ae.aat .flfl

t-e b p A4 dA *ta ".A.. Lb- 60a.pp..y thea

eSf.t t's ' .e* C.u-.4':4. cad .b . ~. mf.u absth

b~tb pam wdmue l W :.c....,m -,"$

.44 ase &... pes .. 1 X.I.st all- a%*.

bs,6bmlf- -4 00 1.. A6A'LVh1 dA .

A-U C-.. tisa6.1=1 Wt. -- v

a. u~. ~ a. C ~ W.ee.'.ad Adf esa,-m ejbý. b...e 6'.t **b.eke

-d -,- ." eo6 . ".b..eaer

me...-I. .aA. AbS. mA.

I. .6 .6 Ait&y a

I~.-, bids -cs

a .UP. b.;& t .P .taa-,. ed

%We"6'.. -4beý 4. . 15 %A. ~.~ ei. ~ Tb... .An6e -1.aa .- , -b -A.__ ____ ____ -ad -1 -iqc . M... b .4

OI 6 _ _ _._ _ is, I I- Abd,† † † †., & - - Nw

I CONFiOENTIALI NAVORD REPORT 4?36

I

!,z ,

' __ __.____

Ii

,PISTON

r1

iiiN

.".

II-

D 173

CONFIDENTIAL


,a

503

a + 16( 116

z

8. w

I 44

I- KI

i"

IPISTIN

Il

PITO

L 174

CONFIDENTIAL

Ip o w

::'•;;•' •'•':- " --- - • 1 I I n . .. .. .. .... IIII.I ' -l " • " -" , r i


\

I I_ '"--

II

9 z

wu0

-----,-"---- -,-""

IT

PISTON <

F£ CONFIDENTIAL

175w'I,

dj . _______"_____" __

CC.FIDENTI4L

NAVORD REPORT 4Z36

3 oo l

2 T4

I I (0'

. -4v 2 PSN

g :."

J Jm

IE

fT

43 a. g.

ih

I 9

IW

[Il4z

* U)

* 4w

V. 1~X-9176

CONFI DENTIAL

I4

CONFDWENTIAL

NAVORD Report 4Z36

Experimental Results

C. Developi..g a sensitivity test that would give reproducible

results:

Obtaining reproducible results has been the most difficult phase

of the impact problem at Bruceton.

The general conclusion is that some mechanical means of

determining the extent of explosion fro-n impact must be developed, instead

of relying on - -rsonal judgement. This has been attempted by measuring

the sound r plosion with a microphone-amplifier-oscilloscope set-up.

Unfortuna' not enough data were obtained toanw'er the question of

reproduc. ity, as some eleven explosives were tested only once by the

conventio.,al scheme.

Recently an electrical trigger circuit has been connected in series

with a microphone, amplifier, and oscilloscope. In the trigger circuit,

a small neon light is actuated by thesoud blast from an explosion.

The trigger was calibrated to fail to ightwhe•ne -. 5 Kg., drop-hammer

fell 11 feet (maximum drop-height of prtsent large impact machine) to

strike a "blank explosive". In our case this was a "charge" of sodium

chloride. Any louder sound would have actuated the light. The light isnot cor cct ir all cases, however, and the circuit needs the attention

of an electrical engineer at the present time. Although the circuit is

not perfect, the method of qualitatively interpreting results seems to bein the right direction. (Subsequently, work was done at the Naval

Ordnance Laboratory on the neon light system. Final results are

described in Appendix I of NAVORD Rerort 359Z, "Factors Affecting the

Behavior of Explosives to Mechanical Shock", G. Svadeba,18 December 1953.)

A comparison of mean values of common explosive for the Type 12impact machine may be seen in Table U. Tables Mll and Ula show the kind

of reproducibility in results obtained by personal judgement and also by

the trigger circuit.

IV-10177

CONFUIMNTIAL

fI

CON1 z" r1. NTIALNAVORD Report 4Z36

i

TABLE H

COMPARISON OF OBSERVER AND MECHANICAL LIGHT FOR MEANVALUES OF 50-TRIAL UP AND DOWN" RUNS FROM TYPE 12

IMPACT TEST. (Numbers in Parentheses Indiicate Number ofRuas to Obtain Listed Mean Value.)

. Common Name Observer (J. M.) Light

"of Explosive Mean(cm.) TNT=100 Mean(cr.) TNT=100

PETN 12 (4) 7 10 (1) 7RDX 19 (7) 12 19 (1) 13DINA 22 (1) 13 25 (2) 17*Tetryl 41(5) Z5 41(1) 27*Pentolite 4Z (6) 26 28 (1) 19

EDNA 42 (2) Z6 28 (1) 19*Torpex, Unwaxed 50 (3) 30 88 (1) 59*Minol-Z 60 (3) 37 30 (1) z0

`- -etrytol 67 (1) 41 103 (l) 69*Composition A 7,! (5) 44 75 (3) 50*55/45 Ednatol 7'i I2) 45 87(1) 58*Fivonite 74 (1) 45 90 (1) 60*Compositiont B 8Z (10) 50 70 (3) 47*Torpex-2, Waxed 83 (17) 51 86 (1) 57

*HBX, Paraffin 116 (3) 71 140(1) 93*Torpex D-1, Paraffin 136 (16) 83 Z09 (1) 90x*UWE 141 (9) 86 Z07 (1) 90x

*TNT 164 (i3) 100 150(13), 231 (1)** 100Explosive D 255 (1) 155 Z72 (1) lie*50/50 Amatol 33 (1) 22

SXTNT = 231 cm . Others, TNT 150

*] *Material screened 50% through 16 on 30 mesh and 50% through 30 on50 mesh.

**Value obtained on day Torpex D-1, UWE, and Explosive D were

tested.

178

SCONrFIfENTIAL

jig ___'_

CONF;W :N TIAL,

NAVOf'.D FReport 423L

TAbLE III

DATA SHOWING RE)RODUCIBILITY OF RESULTS OBTAINED BYPERSONAL JUDGEMENT FROM THE TYPE 12 IMPACT T-ST

INVOLVING 50 TRIAL "UP AND DOWN" RUNS

S Explosive Mean Values (Rounded)

PETN 10.5, 13, 1 , 12PRDX 18, 15, 15, 18.5, Z3, 21, 24Tetryl 39, 33, 4Z, 53. 40Pentolite 44, 39, 44, 44, 38, 41EDNA 4G, 44Tnrpex, Unwaxed 46, 57, 4660/40 Cyclotol 55, 66DBX 52, 75"Composition A 69, 75, 69, 75, 7155/45 Ednatol 71, 77Composition B 86. 65, 76, 7Z, 74, 68, 8Z, 64,

116, I14, 81Torpex, Waxed 81, 86, 81, 85, 82, 107, 90. 83,

60, 6Z. 5, 7Z, 85, 69, 9?, 96, 85,91

HBX, Paraffin 107, IZ, 2Z0HBX, S•anolind 141, 120, 132, 113, 1Z7, lZ3,

IZ5, 107Torpex D-1, Paraffin 15Z, 130, 132, 136, 134, 134,

IZ6.5, 100, 107, 212, 1o0, 145,112, 134, 118. 99

UWE IC', 105, 118, lZ, 13Z, 169,

;.03, 167, 149Minol 65, 68, 47

TNT 130, 145, 171, 146, 161, 123,138, 141, 165, 156, -13, 176,207, Z53, 189

CON;LDENTIAL. I


TABLE Wa

DATA SHOWING REPRODUCIBILITY OF RESULTS OBTAINED BY

TRIGGER CIRCUIT FROM THE TYPE 1Z IMPACT TESTINVOLVING 5Q TRIAL "UP AND DOWN" RUX'S

V

4 Oplosive Mean Values MRounded)

Composition B 70, 64, 77

Composition B-2 71, 70, 71.5

Composition A 68, 73, 83

i DINA 24, Z5

TNT 130, 134, IZ5, 134, 132, 143, 140,1.3, 130, 155. 182, 197, 231(arranged chronologically)

A

The above values were obtained by using the neon light as thecriterion for an "up" or "down" in height during the actual runs. On theday the value of 231 cm. for TNT was obtained, the Llg~ht was checkedagainst the oscilloscope by photographing the traces; and it was foundthat the light was incorrect. Deflections of maximwu rise on the tracewere as great as 5. 6 mm. for TNT and yet registe-redl as N by the light.Past experience has indicated that deflections greater lbah 3. 0 mam. arenearly always an indication of an explosion.

Below are shown data from the actual run which i•flistrates theerror of the neon light interpretation. The explosive is TNT and themean for the run was calculated to be Z30. 8 cm. The observer is seento be correct most of the time.

1 IV--13180

CONFIDENTIALL

V./,

gone"

CONFIDENTVALNAVORD Report 4Z36

Height Light Observer Deflection H L. 0 D

75 N E 4.9 ZLZ E Ei 2.7

106 N Ep 5.8 150 N N 5.0

150 N N 1.2 a ZI N N 2.1

.Z1 E El 7.9 300 N D 5.0

150 N N 1.7 (424) (E) (E) ---

212 N D 3.1 300 E E 7.0

300 E E 6.3 21Z N D 3.4

Z1a N D 4.8 300 E Ep 5.8

300 N E 4.7 21 N N Z. 0

7' (424) (E) (E) - 300 E E-1 8.3

" 300 £ .l 2. Z Z12 N Ep 5.1

212 E E 7.3 300 E EP 6.0

O150 N N 1.6 Z21 N D 4.5

212 N N 1. 6 300 N Ep 5.0

300 E D 4.0 (424) (E) (E) ---

212 E E 6.6 300 E E 6.0

150 L E 6.4 Zi E E 8.0

106 N N 1.2 1SO N N 0.8

150 N E 4.1 21Z N N 2.3

"ZI1 E 3.9 300 E E 5.8

150 N N 1.1 212 E El 7.1

Z21 N N 1. z 150 N N 1. 4

300 E El 7.0 zlz N D 4.1

212 N N 1.0 300 E E 5.9

212 N N 1.1

300 E E 6.0 300 N D 3.0

one may undoubtedly expect better reproducibility. At this writing the

light has not been corrected because of the absence from this

laboratory of the designers of the circuit, Mr. Axlerod and Mr. Kollar.

It may likewise be an aid to obtaining better reproducibility among

results to use small explosive pellets of initially constant dime nsions

instead of testing random shaped heaps of explosive for the impact

"charge". Also, by measuring the gas evolved during explosion in a

closed impact system, or by measuring the blast pressure evolved

during explosion from impact may render results which are in better

agreement. IV-14

* '181CONFIDENTIAL

, CONFIDENTL4ALNAVORD Report 4Z36

I D. Develuping an impact test which gave resultj in agreementSI. with practice.

This has not been realized in an overall sense with any designr yet developed at Bruceton. The Type IZ impact machine cornea netrestto giving an ordering oi sensitivities which is in agreement with practice.The deviations for this par:icular design appear with oxygen-richsubstances, in which the sensitivity evaluations appear tog. L 'h.,'vsensitive than expected).

Ni We have verification (1) that oxygen-rich substances react with thepaper base of the flint paper, and are thereby sensitized. This isespecially true of KCI0 3 . NH 4 CI0 4 , NH4 NO 3 and mixtures containing

NH 4 NO 3 such as Amatol, Minol, and DBX.4 3

An approach to eliminating this diffic-ity was suggested sometimeago by Dr. MacDougall, in that abrasive coated metal disks should besubstituted for the flint paper. This was attempted, but the disks

appeared to have too fine an abrasive coating and the explosive flowedduring impact as with smooth surfaces. Disks coated with coarserabrasive should elimina:e this difficulty.

Conclusionsj 1'A. General Theory

It is a general statement that an explosion occurs when the heatsupplied is greater than the heat removed in a given reaction. This

likewise infers that an explosion occurs when the rate of heat application

exceeds the rate of heat removal (8).

In the impact test:

Heat may be applied by: Heat may be removed by:

. a. A resistance to rapid a. Conduction.compression.b. Frictional phenomena. b. Radiation,

Sc. Adiabatic compression of c. Convection.% air entrapped around crystals.

S1 •CONFIDENTIALL

man

A-6 Y_ _


Whether or not a substance undergoes rapid compression willdepend. upon its hardness. Soft cyrstals will offer less resistance torapid compression, as their plastic. flow pressure will be low. Thevesubstaices will permit the impact energy to be concentrated over agreater area than for a hard substance. This property will in turnminimize frictional effects, the adiabatic compression of entrapped air(it should flow outward with the flowing of the solid), and as a resultexplosions from impact should be low in freque.ncy and of a low order inintensity.

A hard crystal, on the other hand, will tead to resist rapidcompression as its abiiity to flow is practically non-existent. Hardsubstances are always brittle and possess an exceelingly great or anindeterminate plastic flow pressure, as they usually fracture under greatpressures. This property will permit the impact energy to beconcentrated over a small area, and as a result explosions should behigh in frequency and of greater violence than the case of soft, semi-plastic explosives.

* Can we then say that, in general, hard explosive crystals aresensitive to impact while soft crystals are insensitive? The answer isaffirmative in most cases, but exceptions do occur. This is best seenin Table IV which roughly compares some common explosives as tohardness and sensitivity.

The exceptions seen in Table IV, notably Nitromannite andLead Azide, indicate that a factor other than hardness is involved inestimating impact sensitivities. This iactor is most likely the thermalactivation energy of the explosive. Thermal activation energies forseveral explosives were determined by Henkin (9) for gradual applicationof heat. In the impact process thg heat is most likely applied over aperiod of time of the order of 10- to 10"4 sec. The values of andcomparative orders of thermal activation energies for heat applied inthis interval of time may not necessarily be comparable with thosemeasured b', Henkini but his data are worth considering.

Hentin immersed a copper vial containing about 25 mg. ozexplosive into a molten alloy both of known temperatures, and measuredthe time for explosion to occur. Naturally, it required a finiite time toapply heat to the explosive. The lowest explosion time Henkin measuredwas about 0.07 sec. in our case, the impact process has undergonecom;.'etion in 0. 07 sec.

IV- 16183

CONFIDENTIAL

!S

CONFIDENTIALN'AYOT D Report 4Z36

TABLE IV

A ROUGH COMPARISON OF SOME COMMON EXPLOSIVES ASTro iARDNESS AND IMPACT SENSITIVITY

ComparativeComparative Impact Sensitivity

Hardness* Design 3 Deign 12Explo sive H. I. L. H. I. L. H. L L

Lead Styphnate ? XNitromannite X X XLead Azide X X XPETN X X XRDX X X XDINA X X XNEWO X X XEDNA X X XTetryl X X XPentolite X X XFivonite X X XComposition B X X XPicric Acid X X XEmmet X X XComposition A X X XTNT X X X

Explosive D X X X

Ha rdne ss Sen sitivitr

H = High Hard Sensitive"I = Intermediate a -

L a Low a Soft Insensitive

;By personal Judgement and estimation from experience in handlingthese substances.

"* IV-17184

CONFIDENTIAL

~t

CONFIDENTkALNAVORD Report 4Z36

Henkin plotted logl 0 t (time or induction period) as a iunction ofthe reciprocal of the Absolute temperature, and obtained a linearrelationship over most temperature ranges studied. The slope of thestraight line determnines the therinal activation energy from thewell-known equation:

E Ilog~ Z. 303 y + constant

where t is the length of the induction periodR is the gas constant, expressed in caloriesT is the absolute temperatureE is the thermal activation energy,

expressed in calories

Shown below is a comparison of Henkin values of E (9) with roughevaluations of hardness and impact sensitivity.

Comparative Comparative ImpactHardness Sensitivity

, •_xplosive E(Calories) H. I. L. H. I. L.

Ls ead Styphnate 58, 800 ? XPETN 22,000 X XDINA 12,000 X X (X)NENO 17,ZOO X XEDNA 10,000 X X (X)Tetryl 14,400 X XFivonite 13, 500 X X (X)

Emmet 15,000 X X (X)Picric Acid 27, 400 X X (X)

(X) Type 3 Impact Machine

Soft materials with low activatin energies are: Fivonite, Tetryl,DINA, EDNA, NENO, and Emmet. These appear generally in theintermediate class of sensitivity. Their low activation energies maytend to shift them from the insensitive class of explosives, as the

* IV- 18185

CONFDENTAL.,

*

I +•" - r -~ r' ' -

- i .A. .I• .o +

/I

: - CONFIDENTALNAVORD Report 4Z36

latter would be expected on the basis of their softness. However, under

high confinement where their flow cannot counteract the low activationS/ energy, these substances appear as relativcly sensitive substances. This

is seen in previousil reported data (1) from the Type 5 impact machine.

These substancem likewise appear insensitive under impactconditions where they may flow relatively freely. This is seen in datafrom the Type I and Type 3 impact tests (1).

* The notable exception of high activation energy and high"sensitivity seems to be Lead Styphnate, but its comparative hardness isquestionable. The high value of E may explain why this substanceexplodes with greater violence than most other explosives. Once itbecomes activated, the decomposition is most violent and complete innature. The presence of the heavy lead atom may explain the exceptionin that appreciable internal strain is undoubtedly present within themolecule. This may also be true in the cases of sensitive MercuryFulminate and Lead Azide, where again heavy metal atoms are heldby comparatively weak linkage.

Another chemical factor which may determine sensitivity to acertain degree is the heat of explosion. Explosives with the higher heatof explosion may tend to be more sensitive, in that a group of moleculesactivated to explosion will activate neighboring molecules by the heat ofexplosion and propagation occurs. This property may account for thedifferent propagation properties among explosives.I' calculated by Brinkley and Wilson (10) and are correlated fairly well

"1 {- • with impact sensitivity, as seen below.

ComparativeHeat of Explosion Impact Sensi- Propaga-

Material K cal. MKg. (Z5 0 C) tivity tion

PETN 1410 High CIOodRDX 1240 "

* NENO IZ1I IntermediateEDNA 981Tetryl 890

MNO 736 Low FairTNT 650 "

j186'I •" ? •CONFIDENTIAL

' .•2 '_ --.---- 2.•1

CONFLDENTI&LNAVORD Report 4236

These three properties of an explosive, namely, the comparativehardness, the thermal activation energy, and the heat of explosion seemto give a reasonably good preliminary estima•le of the sensitivity classof the explosive. Heavy metal salts seem to be the only outstandingexception here.

It seems logical to assert that future reaearch directed to themeasurement or calculation of theme properties may reveal importantinforrmation.

General correlation of sensitivity with available heats of explosio 1 0 }),thermal activation energies (9), and comparative hardness may be seenbelow. Comparative hardness could possibly be determined by measuringthe velocity of sound in the explosive, as this represents the speed of thecompression wave through the material. Materials in which the velocityof sound is low are soft, while hard substances permit sound to travelthrough them with greater velocities (LI). Sonet (12) has recentlydeveloped a method for measuring the velocity of sound in explosives, buthis data are restricted to mixtures and only one (TNT) of the above puresubstances.

6£

Heat of The rmalComparative Comprrativa Explosion ActivationSensitivity Hardness K cal. /Kg. Energy

Explosive H. I. L. H. I. L. Calories

PETN X X 1410 22,000RDX X X 1240 ------

NENO X X 1211 17,200EDNA X X 981 10,000Tetryl X X 890 14,400MNO X X 736 ---

TNT X X 650

B. Recommendations for Future Development

(a) A Type 12 impact machine should be developed in which theabrasive paper is replaced with an abrasive coated steel disk of aboutI" diameter and 1/3Z" thickness. (Glass cloth has been used at the NOL

IV-L0

CON1FIDENTUA•

CONFIDENTIAMNAVORD Report 4Z36

but it is hard on the operator; see NAVORD Report 359Z.)

(b) A closed system employing a desioa as di3cussed under (I)above should be developed to measure the blai. pressure during anexplosion from ixnpact.

(c) The explosive charge should be either a small catting orpellet, cylindrical in -hape and of constant initial dimensions. Pelletsor castings of 0. 065-0. 075" in height and 0. Z25 diameter would beadequate. The random shape. and contact areas of an initial scoop orheap of explosive would be greatly minimized, should pellets be adopted.

(d) The striking surface of the falling weight should be curvedinstead of flat. Ui mechanically possible, a hardened steel sphere whichcould be rotated after each impact would be ideal in that repeated impactswould not be occurring over the same area to soon cause a deformationof the curved surface.

(e) 1. The base of the piston-anvil holder should be circularinstead of square and should be made rigid by at least four instead ofthe usual two cap screws. Lock washers should also be used with thesecap ecrews.

2. The piston-anvil holder should be one piece of equipmentinstead of a two, piece holder, as are present holders.

I 3. A press fitted hardened steel insert should rest directlyunder the anvil to eliminate d,.'ormation of this area, which occurs withcold rolled steel.

4. The only interchangeable part of the piston-anvil holdershould be the means of holding the anvil rigid and the piston guide ring(press fitted). The present arrangement seems adequate here.

(f) A revised Type 3 impact test could be developed in which the0. 306" diameter striker or piston tips are changed to 0. 500" diameter,

and the brass cups are replaced by hardened steel cups of about the samedepth as current brass cups but of about 0. 502" i. d. The cup materialis important in that if it flows during impact, it permits creepage of theexplosive and in turn enables the explosive to escape much of the impactenergy. Hardened steel cups should not flow appreciably and explosionsof greater frequency and violence may be obtained with insensitive

• IV-Zl- E188

'" CONFIDENTIAL41 -4.

* o.

CONFIDE TIAL

NAVCRD Report 4Z36

materials. Tco, the 1/2" diameter piston would be durable for impactenergies greater than 500 Kg. -cm.

(g) If the intensity of sound is continued to be used as acriterion of explosion, the tests should be conducted. i a constanthumidity and temperature room.

t (h) Dr. K. W. Tuckey of the AMP at Princeton University,*# Princeton, New Jersey recently visited us and has sug£gested that the

interior walls of th.• fi.ring chamber of the impact machine be coveredw-ith a sound absorbing material to eliminate possible echo effects,

deflections of the sound waves, and perhaps guide the compression wavedirectly to the microphone. (this procedure is in use at the NavalOrdnance Laboratory.)

(i) Fundamental studies with impact are naed,-d:

I. Investigation of the theories of Hertz ('3) and_ St. Venant (14) should be undertaken to verify or discre4it them as the

case may be.

2. Contact times between bodies repreuen~ed by the surfaces,velocities, and masses of the present impact test shoult be measured.

3. The impact pressures involved in the imp Lct test shouldbe known. These could be measured with a large, stron:,y builtcondenser gauge.

(j) An impact machine should be developed in which velocities25 ft. /sec. could be obtained. A compressed air-driven drop-hammercould easily approach velocities encountered in bomb dropping tests andmost likely those attained in rifle bullet tests.

(k) More work should be directed towards the development ofspecific tests to answer a specific impact problem.

*1 * IV-22

CONFIAINT!AL


) References

(1) Bruceton Reports to Dr. E. H. Eyster from R. F. Davis,July 31 and August 26, 1944 (reports unavailable).

(Z) D.P. MacDougall, O.S..4D. Report No. 804, August, 194Z.

* (3) Eastern Laboratory, E. I. du Pont de Nemourb and Company, Inc.,0. S. R. D. Report No. 4315, March 31, 1945.

A (4) H. J. Fisher and C. 0. Davis, Division 8, N.D.R.C. ReportNo. L-1, December 16, 1944.

I (5) Ordnance Board Proceedings, British, WA-4182-21, March 21, 1945.

, (6) Ordnance Board Proceedings, British, WA-2610-1Z, July 21, 1944.I(7) "Statistical Analysis for a New Procedure in Sensitivity

Experiments", Applied Mathematics Panel Report No. 101. IR,J SRG-P No. 40, July, 1544.

i (8) Getman and Daniels, An Outline of Theoretical Chemistry, ?. 350.SI John Wiley and Sons, New York, N. Y. 1937.

J (9) H. Henkin, 0.S. R. D. Report No. 1986, November 4, 1943.

(10) S. R. Brinkley, Jr. and E. B. Wilson, Jr., O.S.R.D. ReportNo. 1510, June 14, 1943.

(II) Handbook of Chemistry and Physics, 23rd Edition, pp. 1511-12,Chemical Rubber Publishing Co., Cleveland, Ohio, 1939.

(12) J. L. Jones, Naval Ordnance Laboratory Memorandum No. 5640,June 27, 1944. Also N.O.L.M. No. 6405, January 22, 1945.

(13) J. Hertz. J. reine, Angew. Math. 92, 156 (1881), atso "CollectedWorks".

(14) St. Venant. J. de Math. Liouville, Paris, Series 2, Vol. jZ (1867).

IV-23190

CONFIDENTIAL.


Other Reports Which Include Recent Impact Sensitivity Work at theNaval Ordnance Laboratory, White Oak,

Silver Spring, Maryland

The following list of reports contains sensitivity data obtained atthe Naval Ordnance Laboratory subsequent to termination of the war-timework at the Explosives Research Laboratory, Bruceton, Pennyslvaniain 1945. Termination of the Bruceton work resulted in moving theimpact sensitivity facility to the Naval Ordnance Laboratory. The listis intended to be complete although some omissions may have occurred.Some NOL reports other than impact reports have been included becausetheir subject matter is considered to be relevant to the general field.

A few reports, other than those originating at NOL, are alsolisted for the purpose of calling attention to the ideas and data whichthey contain.

* (1) OSRD 66Z9 J. M. Downard and R. W. LawrenceSensitivenees of High Explosives, 30 March

1946, Confidential (Final report on contract() ROEMsr-719 with the Hercules Powder Co.).

(2) NAVORD Report ___________________

66-46 George F. StrolloContainer Dent Sensitivity of Fxplosives,1 April 1946, Confidential.

(3) NAVORD Report85-46 R. J. Seeger

Final Report on Comparison Test of ImpactSensitivity of Military Explosives, Part I,Summary of Data, 14 Auguat 1946, Confidential.

(4) NAVORD Repo t

94-46 R. J. Seeger; Final Report on Comparison Test of Impact

Sensitivity of Military Explosives, Part U,• i IV-24

S~191

CONFIDENTIAL

,

7

CONFIDENTIAL

NAVORD Report 4Z36

Discussion of Results, 14 August 1946,

Confidential.

(5) NOLM 10, 003 E. H. Evster aud L. C. Smith (NavalOrdnance Laboratory Memorandum, WhiteOak, Maryland)Studiez of the ERL Type 1Z Drop-WeightImpact Machine ,L NOL, 25 January 1949,

Confidential.

(6) NOLM 10, OZZ E. H. Eyster and L. C. SmithGasometric Studies on the 140L Drop-WeightImpact Machine, 16 February 1949,Confidential.

(7) NAVORD Report1589 N. D. Mason

nImpat.t Sensitivity Determinations ofExplosive Compounds Tested During the

Period from I January 1950 to .1950, 1 November 1950, Confidential.

(8) NAVORD Report

2111 G. Svadeba

Impact Sensitivity of Primary Explosives,I June 1951, Confidential.

(9) NAVORD Report"Z140 G. Svadeba

Impact Sensitivity of Perchlorate Explosives,28 June 1951, Confidential.

(10) 14AVORD Report2181 G. Svadeba

Impact Sensitivity Ltermninations of ExplosiveCompounds Tested During the PeriodI November 1950 to I August 1951,1 August 1951, Confidential.

9zCONFIDENTIAL

N. .


(11) NAVORD Report2197 0. H. Johnson

Preliminary Studies of the Desensitizationof Ex.plosive Compositions of the TypeAluminum /.•mnmonium Perchlorate / f.DX,17 September 1951, Confidential.

(12) NAVORD Report2433 0. Svadeba

Sensitivity of Explosives to Impact; Periodof 1 August 1951 to 1 May 195-, 5 May 1952,Conlidential.

(13) NAVORD ReportZ579 Russell McGill

The Sensitivity of Explosives, 7 August i952,(Translation of a Japanese Impact SensitivityLavesti-ation by Mlasayoshi Niimi, Journalof The Military Explosives Society, 30No. 2 106-111 (1936).

(14) NAVORD ReportZ647 G. Svadeba

Impact Sensitivity of Primary Explosives,I November 1952, Confidcntial.

(15) NAVORD Report.i, 832 G. Svadeba

8bDesensitization of Ammonium Perchlorate

Explosives, 9 April 1953, Confidential.

(16) NAVOPD Report2940 G. Svadeba

Sensitivity of Explosives to Impact, Period1 May 1952 to 1 July 1953, 1 July 1953,Confidential.

(17) NAVORD Report3592 G. Svadeba

Factors Affecting the Behavior of i•xplosives' to Mechanical Shock, 18 Decembez L953,

Confidential.

i 193CONFIDENTIAL

rI

77

I

CONFIDENTIALIW * NAVORD Report 4Z36

(1 18) NA•VORD Report3853 D. 7. Addonizio and G. Svadeba

Microscopic Antlysis of DesensitizedZxplosives, Z9 October 1954, Confidential.

(19) ARF Project (Final report for the Naval OrdnanceC-0Z3 Laboratory)

S.Kiyo Ha(.,i and W. C. McCraneDesensitization Studies, 10 June 1955,ConfideAtial (covering the period of contract15 April 1951 to 15 May 1955).

.1(201 NAVORD Report3955 George Svadeba

I Sensitivity of Explosives to Impact, Periodf I July 1953 to I November 1954, Z3 May 1955,

I Confidential.

(ZI) NAVORD Report4058 L.E. Starr, S. Duck, G.W. Reynolds

Sensitivity of Explosives to Impact, Period1 November 1954 to I May 1955, Confidential.

Gap Sensitivity Reports(1) NOLM 10, 336 (Naval Ordnance Laboratory Memorandum,

I | White Oak, Maryland){ | E. H. Eyster, L. C. Smith and S. R. Walton

The Sensitivity of Explosives to Pure Shocks,14 July 1949. Confidential.

(2) NOLM 10, 577 R. H. Stresau and L. E. Starri" Some Studies of the Propagation of Detonation

Between Small Confined Explosive Charges,15 Juiy 1950, Confidential.

(3) NAVORD Report2370 C. C. Lovenberg

S]Booster Sensitivity Investigatio;s During thej . Period from July 1949 to March 1952,

IV-27

119 - -CON TL

NAVORD Report 4Z36

Z0 March 1952, Confidential.

(4) NAVORD Report-494 W. E. Dimmock, Jr.

A Small Scale Gap Sensitivity Test, Z July 1952,Confidential.

(5) NAVORD Report2589 C. C. Lovenberg

Booster Sensitivity of Ammonium Perchlorate,19 August 1952, Confidential.

(6) NAVORD Report2884 G. D. Edwards and T. K. Rice

Liquid Monopropellants: Progress ReportNo. Z: Detonation Sensitivity, 25 OctoberI 1953, Confidential.

(7) NAVORD ReportZ997 J. Savitt

A Sensitivity Test for Castable Liquid Explo-sives Inciuding Results for Some NewMaterials, Z2 October 1953, Confidential.

(8) NAVORD Report2.938 J. Savitt

Effect of Acceptor Explosive ConfinementUpon Acceptor Sensitivity, 11 November 1953,"Confidential.

(9) NAVORD Report3608 3. Savitt

Some Observations Which Suggest a PossibleNew Procedure for Certain Sensitivity Tests,21 January 1954, Confidenti..

(10) NAVORD Report

3753 J. SavittSome Observations on the Growth of Detonation,

Z5 August 1954, Confidential.

IV- 28195

;4e

* I • , | m i| /

AVORD Report 4Z36

Miscellaneous Revorts

() OSRD 5617 E. MA. Boggs et alj ~ Initiation Stu.dies on Solid Explosives,

23 October 1945, Confidential (ERL,

1~ Bruceton, Pa.).

t(Z) OSRD 5609 H. A. StreckerInitiation, Propagation and Luminosity

*1~ Studies of Liquid Explosives, 3 December 1945,Confidential (ERL, Bruceton, Pa.).

(3) NOLIA 10288 3. M. RosenTaliani Apparatus for the Measurement of

I' Thermal Stability of Explosives, 10 January1950, Confidential.

(4) NOLM 108Z4 3. M. RosenI The Thermal Stability of Miscellaneous High

Explosive Samples and Related Materials,30 January 1950, Confidential.

V (5) NAVORD ReportI3703 0.3J. Bryan and 3. W. Enig

Ignition of Propellants by Hot Gases. Part 11Ignitability Comparisons of Twelve Propellants

* ~and Sam-e Effecis of Shape on Ignitability,

~ 10 December 1954, Confidential.

(6) NAVORD Report3748 J .Ei

Ignition Erne rgie s of Solid Propellants,1 29 April 1954, Confidential.

IV2

Ul =o

NAV ýRpct 736

COMMENT ON SEINSITIVITYAND THE USE OF IMPACT MACiM=

D. Price

Statement of Problem and Background

The sensitivity of a material cannot be exactly defined but it i£sSthe tendency of the material to show some exothermic reaction u="-- a

wide variety of conditions. The reaction may be as mild as a rapi•"Tquenched burning or as violent as a high order detonation, If itlead uo an accident while the material is being handled or used, itmust be considered evidence of undesirable sensitivity of the ma-ter-'--

There is apparently no compilation of field accidents in theand handling of high explosives from which some statistical scaleof over-all sensitivity might be compiled. Consequently variousinvestigators have been using a variety of small scale tests to asSeS

"sensitivity" and have been unable to designate any of the numerous -a.ingsobtained as best reflecting the field accidents. Naturally a highLyr

confused and uniatisfactory situation has resulted.

The small sca e te'sts used for sensitivity measurements g'-_resingle point value in circumstances where a family of curves sho=d bobtained if the mate.-ial is to be adequately evaluated. To make t!-_ssituation clearer, an analogy in the study of metals may be conside-•d-_No one now expects to rate metals by measuring the stress necesW--:-7to produce ene particular value of strain. Instead, the entire strasei,-strain curve is obtained for each n.ietal1 and if two curves happen t=cross, a reversal in the rating of the metals at different stress Le-e.lsis expected. Finally, it is recognized that different rates of stre-sapplication result it. different stress-strain curves.

The situation in sensitivity evaluation is very comparable- M: isbelieved that the total thermal energy received by the material and "•"rate at which it is received are the two factors completely deter? -its tendency to react. Of course, it is more difficult in this case z.A-in the case of metals to deterinine how much of the energy provdedactually goes into the material; that depends on the heat conductiCtiand capacity of the material itself.

197

"1I - _ . . . .•-- "... - . .. _,_ _ _ J.'" - -

Recently G.B. Cook has integrated the heat conductivity-selfnheating equation* for boundary conditions to be expected in isothermz.Ibaths for five explosives for which the necessary physical valueswere available. The results demonstrate that for a straightforwardl"application of thermal energy, reversals in rating will occur forratings carried out At •,zf...... týmrpn tures. In cth-ar wordp,

I reversals under different test conditions are to be expected and ax.J confusing only becaube the cntire curve has not been determined.

In the case of propellants, most accidents occur during use oU.-thematerial as a fuel, i.e., as a result of its exposure to thermal ener'y,In the case of high explosives, with the exception of cook-off phenorti-na,most ac.cidents are associated with some type of mechanical impact:(dropping a shell, penetration by a bullet, sympathetic detonation).-In t;S.s latter case, the same factors of total thermal energy and ratee -atwhich it is received are still fundamental, but an additional one of thMeffectiveness of converting the impact energy into thermal energybecomes equal'y important. Thus the general statement that heatsensitivity and impact sensitivity are unrelated seems incorrect. Ittseems most probable that the mechanism of initiation is thermal andckiis the same in both cases, but that physical factorz, ý,ur~h as hardneam of

Sthe material affect the conversion of the impact energy into thermallenergy so that the rate at which themergy reaches two explosives difffereven when the impacts are identical. Thernmal tests (vacuum stabilUtyat 170 0 C, cook-off temperature) show tetryl to be more reactive tha=RDX. Mechanical tests (rifle bullet, booster, impact) show the ro'eevse.This might be explained by the fact that RDX has a Moh hardness ofi2. 5 while that of tetryl is 0.7.

"Impact Machine Test

Si The mechanical test in widest use (i.e.. the easiest and quickeest)j is the impact machine test. Comparison of test results from one labbwith those of another show a chaotic condition, but this test will altmnutcertainly continue to be used for some time as the chief criterion offjudging safety in handling. It is, therefore, of some importance ta Lfunda way of achieving 3ome agreement in the test results.

;'• *This includes both factors: total thermal energy and rate at whictiiit Isreceived. See G.B. Cook, The Theory of Thermal Explosions,A.R.D.E. Reports 19/55 and 27/55, Oct. 1955.

-- . -

This test is a single point evaluation, and, therefore, neverrep.-seaLtative of sensitivities to be expected over a range of conditions.In or:er to be a satisfactory impact test. the point of evaluation mustbe s4 chosen that the test values satisfy two criteria:

1) They must be reliably reproducible.

S) They must order explosives in a rating equivalent to thatw!:.: would be found by a statistical evaluation of field accide :s dueto -- eýhanical impact.

12zUnfortunately, no statistical evaluation of field accidents is

avai-a1:-e. However, some well established qualitative informationcan 'e used. For instance, TNT has been handled safely for manyyears, and Explosive D (ammonium pcrate) is even less sensitive thanTNT as evidenced by the use of picr:atol (Explosive D/TNT, 52148)in a:•.or piercing and semi-armor piercing shells. On the other hand,

IODX is so sensitive to muchanical impact that it cannot be used alone.It is used only in combination with less sensitive materials such as TNTor m-•.. Oth.er qualitative inforrmation of this type is available, but thepresent examples serve to show that a useful impact test must rate*these materials in the order of RDX, TNT, Explosive D for decreasingsensitivity. Unless an impact test rating is in agreement -with theavai•lable handling information, its test values should never be consideredas se-sitivity ratings of the explosive.

The difference in sensitivity ratings from laboratory to laboratoryarises, of course, from using test values obtained at different evaluationocints as sensitivity ratings. Different tool types are used in differentmachines and tool types similar to 1o. 5 (whicU does not satisfy

criterion Z) are quite prevalent. While there is general agreement thatthe 'ýZ% explosion level is easiest to measure, there is some opinionfa%-aring the use of the 10% level in a safety evaluation.

a The purpose of the work reproduced in this NAVORD Report wasto develop a set of tools by use of which impact test values satisfying

the two criteria of a useful impact test could be obtained. It isrepro.duced here so that other investigators may profit by it, and not

need to retrace the initial developmnt work. By the end of hisinvestigation, Davis had found that, of the tools tested, type IZ best

the criteria for most solid high explosives, and type 3 was'rimary explosives and very sensitive high explosives.

f.6- *8i''I

UCIASSIFIED

MM "Pp36r03

H!e rv <x•=ane-de- type 1'3 toolsAior liquid high explosives and the&sca•-r-ng of the ,other tr•'pes oýf :ools studied for rearons noted in therepoc$s. Why E..fster a..n. Dav-ris subsequently recommended (OSRDI) 5744)tes=.F. -'aith ,-ol w-'pes 3", 5, azd i±. not cl.,r. r. samne outa wereconasi- red in both ca se-Y. and ta-ae, data (those reported here) indic.%te=art cv--s 5 t.,.ols esTrult2.-d in, va.Lues which satisfied neither of the test

-he prese=e. data-.-.ndicatce that ratings at the 50% level weregene--.-ly the san.e as rratings :at the 10% level if the type 1Z toolb,2. 5 ls4... weight ad 35 rg .ita-==iard sample size and preparation wereused. The sa,.z rating- -at two: r4iferent levels wvuld not nccessarilybe e=".zted for any genatral teist. It must also be kept in mind that theirpc-, %ast &tve s a statt 'sticaL,-evaluation and is, therefore, subject tothe -- itatioc-s oE any stzatistic.al study. In particular, one cannot expectto o si uig=i.!-iant vai".es forr the lower levels with a statistical samplesiz- c4 only PZ5-L-D0 shcu;s.

"e•_o,. . 22 C.-.ols (oxr- a bettter design) should be establihed, there is-j. s:cie c-,-,t..1 worsu. that sk-iould be carried =-t on this test. It

' -•sa of ve•gatu.,; the eiffect on test values of

1it. Crga.=ic material.((-sandpaper and its adhesive)

21-. Oher weigh.its thaz. the 2. 5 Xg one now used

3. T•--.eraturre variaation

S-&.. Pirical staate of-_sample

(a.) Use acf preussed and cast pellets rather than powder.

Sb) Presseed petllets preferable for comparison of high.

ewp•osive teest valmes with those ot propellants.

um~ --UNCLASSIFIEDWO"-

+,1 A+,. . . . . -. . +, , -. . ., + + - •

./

NA Re poor 4236

DISTRIBUTION LIST

No. ofCopies

Olin-Mathieson Chemical Corporation, East Alton, Ill.,Via: InsMatAttn: T. G. Blake .......... * .................... 3-Attn: Mr. F. R. Seavey ...... . ..... .............. I

Eastern Laboratory, E. I. DuPont de Nemours and Co.,Wilmington, Delaware, Via: Ins~at,Attn: Dr. Clyde 0. Davis ........... ............... • 3Aitn: Dr. D. L. Coursen ........................ .... •

Armour Research Fo.ndation, Chicago, Ill., Via: InsMat,Attn: Dr. Walter C. McCrone ... .................. ..... 3

National Northern Division, National Fireworks OrdnanceCorporation, West Hanover, Mass., Via: InsMat,Attn: Dr. A. D. Mirarchi ........................... 3

Department of Physics, Pennsylvania State University,State College, Pennsylvania, Via: ONE,Attn: Prof. Raymond Pepinsky ........................ 1

Atlas Powder Co., Reynolds Experiment Station,Tamaqua, Pennsylvania, Via: InsMat .................. 2

Redstone Arsenal, Huntsville, Alabama,Attn: Rob.i and Haas .......... .. ........ • 1Attn: Thiokol Corporation ...........................

* Attn: Rocket Development Laboratory ................. 1Holston Defence Corporation, Kingsport, Tenn., Via: InsMat. 1Wabash River Ordnance Works, Newport, Indiana, Via: InsMat. 1Commanding Officer, U. S. Naval Ammunition Depot,

Crane, Indiana ...... .......................... ... 1University of Wisconsin, Naval Research Laboratory,

Department of Chemistry, Madison 6, Wisconsin, Via:ONRAttn: Prof. J. 0. Hirshfelder ... o ........ 1

Jet Propulsion Laboratory, Pasadena, Calif, Via: InsMat,Attnt Mr. G. Robillard . ...... 2

Pe-ojet-Gen-iral Corporation, Azusa, California, Via: InsMat,Attn: "r. M. H. Gold 2..... .... ... 2Atlantic F ._rch Corporation, 812 North Fairfax Street,k!exa -?as, Va., Via: InsMat . ................

University %.f Utah, Salt Lake City, Utah., Via: ONR,Attrn: Prof. Ztelvin Cook *****see *a* 2

Southwest R.searnh Institute, San Antonio, Texas, Via: ONR,Attn: Dr. James H. W~igand .......................... 2

Chief, Armed Forces Special Weapons Project, P. 0. Box2610, Uashington, D. C. 0...................... • ... 3

'WChief, az*J!-au of Aeronautics, Department of the Navy,Wanton 25, D. C... .. .

t;t

19CLftSSIFIE4 _" " '

DISTRIBUTION LIST, Cont'd.

No. ofC01e

CGhie£, Ltrpau of Otd ance, Washington, D. C.,

At( ui Re06 ........... .. *.. . ........... . ... ... 1j. A ttf•i' ReU3 ....................................... 1Atil Ad6 1j A Ad6 ...... e.e~.s.. ............ •..... •.... ... .. ..

Chief, )1'teau of Ships, ,ashington, D. C.,Atitts Code 423 ..... . 5. .. .. .. ... ...... . . . . . 2

Chief, Uirtice of Naval Research, Washirgton, D. C.,Attlit Dr. Ralph Roberts ............................ 3SAttWit; Dr. L. Butz ..... . .. . . .. . . . . . . . .

I C)ef of GCrdnance, Department of the Atmy, Research andDeVtdopment Division, Washington, L. C.,A trIt ODTM .*. . .... .. .. .. .. .. .. 5

Chief ol' Enginieers, Department of the Army,Wao•[ington, D. C.,Att III Research and Development Division ............ 3

Commandor, Air Material Command, Wright-Patterson AirForce Base, Dayton, Ohio ........................... 2

Commandog', Naval Ordnance Test Station, China LaLe, Calif.,Attlki Dr. W. A. Gey .... 1...... ......... .. ...... 1

I ; AttlI: K. S. Skarr .................... . . . . . . . 1

ALMtT. 0. K. Penningtan ............................. 1Attlit Dr. W. S. McEwan .. 2..........*........... 2Attn: Library .. ,.. .......... * .................. 2

CommandvP, Air Force Armament Center, Eglin Air ForceBafiO Eglin, Fla ............ ... ....... ... 1

Command Inf. General, Frankford Arsenal, Philadelphia, Pa.. 3Commanding Officer, Engineer Research and Development

Labnratories, Fort Belvoir, Va . ..................... 3Commanding" Officer, Naval Mine Depot, Yorktown, Va ...... 1

S'Attn J. T. Mianley. Research and Development" ,or tor .1...........

Attri: G. Fvac.eft, Research and DevelopmentLga)orat ;ry ......... ......... .................. 1

- Commanding Officer, Naval Powder Factory, Indian Head,14tipyland ......... I

SAttnt Dr. Sol Skolnik 2At~ntA. OwyangI

Atu 8 J. A. O'Callagham 1Attn: Re H &-ay ...Commander, Air Materiel Test Center, Eglin Air ForceB10Eln Fla. 0 00 0000*6000se0 **o1

L. Commandor, Norfolk Naval Shipyard, Portsmouth, Va.,Attn$ Code 290 .......................... ........ 1

ASTIAj, Payton Ohio

* .UINCLASIFIED ______•. ,,- • •'" , • , ... -'" • •'•' • '"• " ,t. t •" I "• • L• "'"

UNCU SSIFIED

DISTRIBUTiON LIST, Critl,No. orCopieo

Conmi•anding Officer, Picatinny Arsenal, 7rover, 1. J . ....... 1Attn: Mr. Karl Ottosen ........... ...... ...... .... * 3Attn: Mr. A. Bulfinch ..... . ...... 1Commanding Offtier, Watertown Arsenal, "iterto,.n., Kass .... 2Director, New Kexico Institute of Kiint,; and Te:hnologyRe3earch and Development Dtvis~on, Socorr.New Mexico, Via: ONR ..............d I'irector, Naval Research Laboratory, Wahiington, D. C ..... 2Attn: Harry W. Foy. •........... I ........ .. 1Attn: Oscar Levine ................ 1Director, National Bureau of Standards Washingt n, D. C.,Attn: 0.-dnance DIvision ........... 3Direator, Los Almirs Scientific Labora.ry, P. 0. Zox 1663,Los Alamos, New Mexico, Via: Ins:Ait .................. 3Attn: Dr. L. C. Smith ............. ......... 1

Director, Applied Physics Laboratory, -z3hns Hopki 3University, 8621 Georgia Ave., Silver Spring. md.,Via: ONR,Attn: Dr. G. B. Swartz .... .......Director, U. S. Bureau of Mine -.b. :;eet,Pittsburgh, Pa. 2Attn: Dr. R. W. 3==================@ ===== 2

Director, Ballistics Research . .borat.-yy, Aberdeen ProvingGround, Aberdeen, Md. .. ...... ........ .......... •..* 3I ~~~~~Attn: D. C. Vest .• ....... .. ...... o•. . . . . .Attn: Dr. Jane Dewey ............................ 1Hercules Powder Co., Experiment Stattic, Wt~mingtoi,Delaware, Via: InsMat 1....... ......... ... ... 1Attn: Dr. R. W, Lawrence ..... 1..................... 1Attn: Dr. A. F. Giacco ...... ...................... 1Attn: Dr. Julius Roth ............. •...............• 11 Carnegie Ia.titute or TechnolcQ , Sc•..iey Park,Pittsburgh 13, Pa., Via: C-.R,Attn: Prof. E, M, Pugh 2

California Institute of Technolc.-Y, Fasadena 14, Calif.,Via: ONR,Attn: Prof. Norman Davidso, .... ,..................... 1

Arthur D. Little, Inc., Cambridgý, N-so., Via: InsMyit,Attn: Dr. E. K. Kreidl ............. *.......•.... 2Attn: Mr, Lee Ashton *1..*.. 0..**00 ...Attn: Mr. A. R. Almeida ..... .....

Commanding General, Headquarters, i0.1ld Command, AFSWP,Sandia Base, Albuquerque, Nt i •s4xico Fort SI.N' 4 1P2Corporation, Attn: Mr. Fratz-c •sa . . .

Attn: Mr. C. F.. -;ld .... ... ... s..... 1.I~ s ttn: Dr. H. Dt'r. Kallory . 1

~ AACLA$SSFIED

'j 1

unclassified - DTIC

Documents