UNCLASSIFIED

AD NUMBER

AD 861893

NEW LIMITATION CHANGE

TOApproved for public release, distributionunlimited

FROMDistribution authorized to DoD only;

Specific authority; Sep 1969. Otherrequests shall be referred to DoD Armed

Services Explosives Safety Board,Washington, DC.

AUTHORITY

DoD Explosives Safety Board, ltr 7 Jul2000, 2461 Eisenhower Ave, Alexandria, VA22331-0600.

THIS PAGE IS UNCLASSIFIED

"FOR OFFICIAL USE ONLY

MINUTESOF THE ELEVENTH

cz EXPLOSIVES SAFETY SEMINAR00

SHERATON-PEABODY HOTELMEMPHIS, TENNESSEE

9 -10 SEPTEMBER 1969 0DII~c ID EC-

VOLUME II

I 11aoh transmittal ot this docuwnt outslde the Dopwt

"getoase oust have prior approva ot "

ARMED SERVICES EXPLOSIVES SAFETY BOARDWashigton, IC.. 20315

FOR oKFFiSE SLY

L L

FOR OFRCIA USE ONLY

I HERAT0I4-JEABODY Pc7ELO

JEMPHIS. ~NESSEEW

9-10 §eptember 1969.

UOLME~

Conducted by ' l

ARMED SERVICES EXPLOSIVES SAFETY BOARD

Washington, D. C. 20315 (

FOR OFICA MSE ONLY

k~s ý-A: 6W &99k=

TABLE OF COI4TENTSO

HWSYSTEM JAFETY CAN REDUCE a XPLOSIVE BAMIADS' OUO~ ~1rUwulauf, Boeilig CompanyIea~tle, Wash- 1 )~)

~JEAETY jsiacs (fr tH NERT-JI LUENT JROCESSCPALED OIOELLATS ?RO&fSS1NG% (FOUD),-trry D. HeAderson, U*.S. Naval 0jnce Station,Indian Head, Md. 31

ii5

HOW* zv.

FOR OFFICIAL USE ONLY

"HOW SYSTEM SAFETY CAN REDUCE

THE EXPLOSIVE HAZARDS"

BYI

J. L. UMLAUFSYSTEM SAFETY MANAGERSRAM PROGRAMAEROSPACE GROUP, THE BOEING COMPANY

SEATTLE, WASHINGTON 96124

FOR OFFICIAL USE ONLY

1a

kýI

p.

bcawe of the short time allowed to present this topic, I cannot penetrate to the

depth required to cover such a comprehensive subject. I hope I can stimulate

enough Interest for further discussln In our questIon/amwer period.

The design, development, test and deployment of military weapon systems are

usually associated with the we of hazardous materials. Chemical or nuclear energy

sources to produce heat and overpressure effects are the most used. The paramount

objective of the US Air Force Nuclear Safety Program Is the prevention of accldentlal

or premature occurrence of a nuclear detonation. Conrfidence that the events necessary

to produce a nuclear yield, arm or fire a warhead, or release or launch a vehicle

will not occur except when authorized. These same objectives generally apply to

arny weapon or hazard source. Since a nuclear weapon susposes cataclysmic

consequences In the event of a catastrophy, a systems approach with total Involve-

ment of people, hardware, software, polIcy, concepts, etc., Is necessary to assure

a low level of risk.

The purpose of this presentation Is to relate the concepts and methodology of System

Safety to the reduction of explosives hazards. We can best fulfill this purpose by an

example and examine some of the results of an Integrated system safety program for

a modern sophisticated weapon system called SRAM.

SRAM Is the acronym for the Short Range Attack Missile capable of air launch from

the B-52 and F -I I A aircraft. We also call It AGM-69A (Air to Ground Missile)

or Weapon System 140A (Chart I). When we refer to the Weapon System we Include

the aircraft andspecIal "black box" equipment, the missile, people, Aerospace

Ground Equipment (AGE), Technical Orders, Training Equipment. All the hardware,

software, people and the systems necessary for acquisition.

The system features (Chart 2) Indicate the Interrelated aspects of missile safety,

nuclear safety, range safety, ground safety, exploslves safety, flight safety and

2

X P

I ~-_7 rt -< 4

Indust rial safety are encompassed In ilils weapon system. The missile (Chart 3)

is propelled by a split pulse type of booster/crulseClass II solid propellant motor.

An Ines. -I/solid state computerized guidance system steers the missile, which

Is an aerodynamic shape, with moveable fins. Various ordnance Initiated subsystems

are used for ejection from the aircraft, battery activation, controls power boost

activation and fins unlock devices. A linear shaped charge (LSC) Is used on the missile

motor for test launches to render the motor non-propulsive and meet range safety

requirernents.

Chart 4 Is a paraphrased extract from the contract statement of work which requires.

systematic analyses and a mathematical model for probability prediction and assessment

of numerical safety requirements. Chart S Is a summary of the applicable specilfications,

manuals and regulations that specify qualitative design and operating safety requirements.

One of particular Interest Is the application of MIL-STD-833, to minimize the hazards

of Electra Magnetic Interference (EMI). All the SRAM squibs and Ignition devices are

designed to meet this standard to minimize premature or Inadvertent Ignition. I

assume most of you are familiar enough, generally, with these requirements without

dwelling on them. Charts 6 and 7 show the requirements to meet federal, state and

local codes and Insure protection of employees and property during manufacture

and transport of explosive components.

The application of System Safety Engineering (Chart 8) consists of: analysis for

Identification of hazards by cross, subsystem and system analyses; categorization of the

hazard as to degree of severityl; necessary correction of the system to eliminate or

reduce the hazard. Preliminary and Critical Design Reviews (PDR, CDR) are key

correction points. Once hardware Is built, the Engineering Change Proposal (ECP)

Implements changes. The system analyses culminate Into Inputs for the system math model,

our major tool for assessing the degree of achievement of our numerical safety goals.

(Chart 9). The system safety numerical requirements (Chart 10) are allocated by phases

which are further allocated Into the mhinon cycle (Chart i1).

3

--L - _ _ _ __._,_ _- ' 4"+•++%-'+-:+-+•.:.- .+ + +"•.::++ :..?+' - .++ .. + + , : ++++ '_• +-+ + . .. + .. +_ . . *+ + . . .. .. . .+ . ... ....

'C • ...-- -+ ++ +, / +:

S.

The logic model, for the numerical allocation and assessment program, uses tho Fault

Tree technique. Chart 12 shows the top system tree for the FB-I I 1 and 8-52 Emergency

War Order (EWO) or Operatlonal Test Launch (OTL) critical or catastrophic events.

The hazardous events Identified In the system (Chart 13) wre Identified by hazards

analyses of the system. Probabilities for these events are allocated by phase

and an mssessmnt/pmrdlction made by a computer simulation of over a million trials

of the system model through Its mission cycle. Data sources (Chart 14) draw from

reliability, biotechnology and maintainability estimates for the system. The

failure rates are Inductively summed by Bolean to give the probability for the top

events. The Monte Carlo technique Is ed to simulate the system model and derlvq the

probabilities by phaes.

An esample of how the system fault tree/computer simulation technique can be used to predict

explosive hazarc, and evaluate the need for change to re~uca the hazor&i and the effect,

I would like to discuss some of the anulyses leading to the first live launch of a SAM

from the 5-52. Chart 15 shows the molar hazardous events that could occur during the

launch. The hazard of mast concern was a missile motor explre "n prior to or after

launch. Premature activation of the command destruct linear shaped charges could

cause Ignition Of the propellant with cam rupture and dispersal of motor cate fragmets.

The events that could cause this are shown In Chart !6.

The other major event was a missile motor explosion caused by the events shown In

Chart 17, which are In addition to failures In the motor mechanical components. Chart 18

shows the prediction for the first launch wing a 1.5 second Ignition time after release

for the boost pulse. A probability of 1.44 x 10-4 was predicted for missile/carrier collision

cawed by faments from a motor case rupture due to overpreusurIzatlon striking the aircraft

and cawing critical damage. The relatively low probabilities for command destruct and

premature Ignition by the motor Ignition circultry Indicated they were sufficiently safe.

The most significant contribution to the event wr physical failures In the motor. A

secondary parametric study was made of the probability of motor cae overpressurization

(explosion) and the probability of ragments striking the aircraft and coming sorlous demage.

4

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

Charts 19 and 20 show the logic of events and probabilities for the motor explosion,1. 1 X 10-3 which Indicated a need to Increase th. separation time between the

missile and aircraft since at 1.5 secondi Ignition time "h missile was only about 25

to 40 feet away from the aircraft. Using the model of Chart 21, which assume& the motor

fragments In a subtended cone strike th aircraft, with the penetration criteria of

Chart 22, and empirical date fhom motor firings which overpressurized and produced

firagments, the graph on Chart 23 was derived From maxImurn penetration heights

for critical wing panels on the aircraft. The Chart 24 was developed for the probability

of fragments striking the aircraft at varlous separation distances. Using a 2000 foot

separation distance (maximum fragment distance from motor firing dole) a corservative

11. 5 second delay was used for Ignition time of the motor boast pulse. The predicted

probability of 1 x 10 for one of 50 particles striking the aircraft gave an Increase

of somne 20 orders of magnitude. A motor Improvement program has been started

to eliminate the cause of failure predicted by the analysis. As more empirical motor

firing data are generated to Indicate higher confidence with lower risk, the firing

time for tho motor will be reduced to Ownormall1.3seconds. Chart 2 sha summary

of aie launched missile Incident&, Including operational and developmen t launche..

It Is Interesting to note that thvese emirical data correlates with the curren t predicticiwfor SRAM.

This briefly describes the degree of Involvement required for a system. safety analysisand how the hazards associate with teuse of explosives can be effectively reduced.

VJ ¶

:42

- '1CD

LOW

rd-r

qII

V9L

LLJ kA

iaS.

Lhi

ce. U -A

*~ ~~ a U

= %-#

a ~ LiJ L enhiO 4L @

CeNAnLin.0 cc z A I.

b ^LL

0 C.D O

Ed) gz =I

0

0 0U

UA-z

U.

z uJ

OU

4c

CLU

t3

LA.I

U) %

0 -R) L&JU

LA. LAW U u

C)(. LL

'L = LU U) U

LUIA & -= LLn jj

tA BE;1~ * * 0 0 to

U) U)w

0 0

' I'T

.~., .7 'I a- &

IA z. ILZ4A

oz I/b Io

MA, V;0 d

Z_ 4cZ (10 U5

Uj z2 zo-WW -e.

z Z z >.

of uJZZ 0 WO~MJL /b

E:-- I.- tA

Li. :tL.!WYI 4

ZA tAw

LOWs z w i - 2

0-- U. -z me .

10

ROM'

-JJ

-JJU'A

LL

0I-

LU 0JL -j -j

< Z3 CA~

:5 a. 0 0

IL 4A

5- UU

,~~LU ~ U' >. ()L

V140-. ----

47.(U

F _ _ _ p

z

oom

g0 L&Ju

C3 4A LA.

00 0

0 L -0F-Z

s4< is.-. i C

CD .1A1^U C L. cC ) L

ccJ 0A 0C L

C3 0 cIC

0 I '~c12

SYSTEMA SAFETY ENIGIUEEflING (rdItL.S.3130)

GROSS

ANALYSIS Mna=SUBSYSTEM

SYSTEM

HAZARDS

IDENTIFICATION_______-I

IV

PDR

*CORRECTION mmeCDR

ECP

Irk 13

4I4A

01 3XI am0.

LAIKcl:

sum~

I LUJ

VP0

14

LAI

4A*CA

9 4A

VL4

LJJA

LU~C CIO~--

LA X tA)i~

C! C%4

am - C

z LLA 4A ...-0 0 U.' 2A)

x ts ICL* *4A2 - - CA

>0 x C5

1 co

____UZ

Iz

_C 0

H

IOo

,d I.- i

ItA

16__ _ _ _ _

4t4

F p

I�

IIIII aBa

S -I

I

'I'llUrnSUman

NII I ill,�Id '�

2*

'8

ill � *1�

�gtq� 1

II I

4:

101:�Ii.�I

17

-- �.-4� _

4-4- - � � -� -�

l+F4 N 1:B E4Ili

aoe' show

• "":"a-",.

411

I b

18

1r___ L

IIMi)I

I-,V

LU L&JJ

C IL16Jws %A .i w

EI~-a Z m4J~~

3- * U -J ~cc

LU 0 A

'4ZO a~m4 ; la

c LU:CLO!

1 21

LAS

203

illdo

.E3

OE

21u

? I.- :E 5

F *

-� � I7�

Ak

liiIA'U

.1U0at

S6�

S

Iii II

Ak

11'

3 �.

r'--- �'

�

A 'S �AAAAAAAJ �

-½ �� 4�

y 0 z 0

r gI

L)zIo Qz

nil

usn

23

IO

ILt

LOi

fig

aal

ILIR

2 24

F 7

si t". .

viq

6~IIF

250

O'II

CACL.9,

4A4

LUL

CAI

26

p

0 z46,,4

( /.

0 V0

27 1"!

%.4

UJI

a.a

. ...... I I. 0 o e .

la____

_ o_

4 .o.

* 28

-L. ' I7lr, J.

1~~tin

I 31'"~*Um~out

go

1l

29a

m mI

So

ioi

- -

I I Ia

st l l

1F

30-

. I...I

m m m j, I ,,

I ii!! 11

Ii -1_ , 11 s Ii.dHI e

an

-l • I -I -" I

' I I

30

•-.• :•:• • -- ,. . ••)• .; • . . • ,i •..•..= • : .•- ,.-•,, •-,.: •;• .:.: ., . ": .k ..•: / .: ,-•, i .• ..• ',.• ."• .• •' .•:•,' ,k• .:. :.,.::,-' -..ii • •• ,• ...

I

FOR OFFICIAL USE ONLYTHE SAFETY ASPECTS OP THE INERT-DILUENT PROCESS

AS APPLIED TO PROPELLANTS PROCESSING

Larry D. HendersonU.S. Naval Ordnance Station, Indian Head, Md.

The Naval Ordnance Station is actively investigatingcontinuous processing techniques usable for rocket propellants.One of these techniques, the Inert-Diluent process has been.under study for the past nine years. This process is basedon the Quickmix process patented by Rocketdyne Division ofNorth American Rockwell. This process has the advantagethat all of the propellant ingredients investigated aredesensitized when held in emulsion or suspension in theheptane carrier, in the line sizes Oncountered in the process,and at concentration greater than/those used in the process.The results of early safety tests have shown that, byseparation of the various processing areas, no chance existedfor transmittal of an explosion from one processing area toanother by means of the process. Since small quantities ofmaterial are in process at any instant, the chance of havingan incident of large magnitudo (such as have occurred inlarge batches of explosives) is greatly reduced.

The Naval Ordnance Station currently has two Inert Diluentfacilities. A pilot plant, which has the capacity forone thousand pounds per hour or less and where quantities ofup to five thousand pounds may be processed, and a largerfacility where a maximum rate of twenty-five hundred poundsper hour may be produced in quantities of thirty thousandpounds. Both facilities use similar equipment, but thelarger facility has considerably more sophisticated equipmentand buildings.

Of particular interest within the Inert Diluent Processare remotely controlled solids handling systems and castingsystems.

The solid feeder, developed for this process, does notuse the conventional belts or screw conveyors, normallyassociated with continuous solid feeders. Instead thefeeder uses a weighing mechanism which greatly reduces thefriction imposed upon the solids. A small weighing hopper,which holds less than one pound of typical explosive solidIngredients, has been fitted with a bottom discharge valve.

FOR OFFICIAL USE ONLY

31

FOR OFFICIAL USE ONLY

This valve is similar in form to the commercially availableRed Jacket Valve but is constructed of fifty thousandths(0.050") thick rubber tubing enclosed In a light gage metalhousing. When the annular space between the metal housingand the rubber tubing is pressurized (with between three andfour pounds per square inch gage air), the rubber tubegently reduces in diameter, folds, and closes off thedischarge orifice of the hopper. This sphincter-like actiongently shuts off flow from the weigh hopper.

In both IDP facilities the solids from manually loadedhoppers flow through the special solid feeders to a pumpagitated solids/carrier dispersion vessel. The solids arethen transported as slurries in heptane to the mixing pointwhere they are mixed with the carrier emulsified explosiveplasticizer. The carrier is decanted in a stilling vesselwhile the explosive mix is either degassed into a castingpot or directly into a rocket motor or other casting configu-ration at 10 mm Hg absolute pressure.

In the pilot plant the explosive flows through hosesinto. two alternately filled, evacuated, casting pots. Asthe explosive enters the casting pot it is degassed by passingthrough a slit plate. When a casting pot is full, &@ observedby closed circuit television on a dial type weight indicator,by addition of controlled nitrogen the remaining free vol1meis slowly brought to a pressure sufficient to transfer theexplosive through a casting manifold into rocket motors whichhave previously been placed on hand carts. After the .DP"process has been shut down, these carts are manually moved toseparate curing facilities.

Control of the casting is done via remotely controlled,Saunders valves and Red Jacket valves, with the aid ofclosed circuit television.

In the larger IDP facility the explosive material flowsthrough the separator, transfer hose, and slit degassingdevice, directly into a rocket motor positioned in an evacuatedcasting vessel. This casting veusel is a itainless steel cladpit fifteen feet in diameter by thirty feet deep with vacuum

FOR OFFICIAL USE ONLY32

iý4

ot Li

FOR OFFICIAL USE ONLY

"capability. A series of several castings may be positionedon a turntable system which is utilized to position themotors directly beneath the alit degassing device. Operationof the system is so interlocked as to avoid underfill oroverfill. Rotation of the turntable is interlocked to thecasting system so that explosive may only flow into an emptyor filling rocket motor, never into a full unit. Explosivedrips are caught on either Teflon-lined trays or on electri-cally conductive plastic sheets. The casting operation iseasily observed over a closed-circuit television system.Operator override is possible on the system, but only to shutoff the flow of explosive or stop rotation of the turntable.Capacitance or float type level switches are used to shut offexplosive flow into the rocket motor when it is full. Aftercasting, the vacuum on the casting vessel is slowly releasedwith nitrogen and curing remotely progrmmed for the cast-in-place rocket motors. After curing, the lid to the casting pitis removed and the motors are transferred for final inspection,assembly, and shipment. The scrap explosive is easily peeledfrom the hardware after it is cured. At the end of a run allexplosive transfer hoses are washed with heptane into a scrapvessel located in the casting pit.

In both facilities no personnel are exposed to explosivesduring mechanical handling of the material. In all casesall of the operators remain in concrete blockhouses severalhundred feet from the casting process so that exposure touncured material is minimized or eliminated.

&1

FOR OFFICIAL USE ONLY

Z..,

=W WOW

Comments:

Discussion and comments on IDP centered around technicalaspects of the process rather than the safety aspects. Theseincluded:

a. A discussion on the moisture removal qualities ofcarrier liquid during processing and the capability ofremoving the carrier from the matrix to below the detectablerange prior to cure of the cast material. Formulationsprocessed in the 50,000 to 100,000 cp viscosity range thatwere vacuum cast and deaerated, gave essentially voidlesscastings. The process is particularly well suited to double-base formulations in this viscosity range.

b. A discussion on the suitability of IDP for otherformulations revealed that composite and cross-linked formu-lations were not well suited. A Rocketdyne representativedeclared that they were looking at the suitability of thesystem to process Tritonol and H-6 using water as a carrier.Comments from others questioned the removal of water (below2% level) from TNT but agreed that it may be a safer operationin that it would reduce the number of people normally exposedto the melt and load operation.

c. Other comments were on the wisdom of applying continuousand remote processes and technology of this type to explosivefill operations. The consensus was that it would be a goodmove.

34

44

S.... .• - -- .-,- , • .- , - .i•.I] .,

... • . . . . .". .:'"" •-".,'•,,.i,•:-. ••',, '••"-:' '-* •i..•;' •! .•":; :•, • ,,- •• •-•.-'•'-' ••; .... . ..•• , , .....A,• •-

DEPARTMENT OF DEFENSE EXPLOSIVES SAFETY BOARD -

2461 EISENHOWER AVENUEALEXANDRIA, VIRGINIA 22331-0600

DDESB-KMC 0 7 JUL 2000

MEMORANDUM FOR DDESB RECORDS

SUBJECT: Declassification of Explosives Safety Seminar Minutes

References: (a) Department of Defense 5200.1-R Information Security Program, 14 Jan 1997

(b) Executive Order 12958, 14 October 1995 Classified National SecurityInformation

In accordance with reference (a) and (b) downgrading of information to a lower level ofclassification is appropriate when the information no longer requires protection at the originallylevel, therefore the following DoD Explosives Safety Seminar minutes are declassified:

a. AD#335188 Minutes from Seminar held 10-11 June 1959.b. AD#332709 Minutes from Seminar held 12-14 July 1960.c. AD#332711 Minutes from Seminar held 8-10 August 1961.d. AD#332710 Minutes from Seminar held 7-9 August 1962.e. AD#346196 Minutes from Seminar held 20-22 August 1963.f. AD#456999 Minutes from Seminar held 18-20 August 1964.g. AD#368108 Minutes from Seminar held 24-26 August 1965.h. AD#801103 Minutes from Seminar held 9-11 August 1966.i. AD#824044 Minutes from Seminar held 15-17 August 1967.j. AD#846612 and AD#394775 Minutes from Seminar held 13-15 August 1968.k. AD#862868 and AD#861893 Minutes from Seminar held 9-10 September

1969.

The DoD Explosives Safety Seminar minute cbe publicrelease, distribution unlimited.

DANIEL T. TOMPKINSColonel, USAFChairman

Attachments:1. -Cover pages of minutes

cc:DTIC

MINUTES

OF THE ELEVENTHEXPLOSIVES -SAFETY SEMINAR

SHERATON-PEABODY HOTELMEMPHIS, TENNESSEE -"

9-10 SEPTEMBER 1969(

VOLUME II

Conducted byS_ - ARMED SERVICES EXPLOSIVES SAFETY BOARD

Washington, D. C. 20315

C _____________