TECHNICAL. REPORT ARFRL-TR-O2410

(Supersedes IMR •4o. 721)

E"TINGUISHING AMMUNITION HRES

Anthony E. FinnertyS~DTIC

July 1982 JUL131982

US ARMY ARMAMENT RESEARCH AND DFVELOPMENT COMMANDFQ USBALLISTIC RESEARCH LABORATORY

ABERDEEN PROVING GROUND, WARYLAND

Approved for public releara ; Oistributlon unlimited.CC)

LAJ • {.•'0 01 O~ ~j1 (jr

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

Secondary distribution of this report is proliibited.

Additional copies of this report may be obtainedfrom the National Technical Infor.'ation Service,U. S. Department of Commerce, Spr,1gfield, Virgin!ia22161.

I

1I

The findings in tiis report are not to be construed asan official Department of the Army pnsition, unlessso designated by other authorized doLuments.

?he me* of tt'ftAj 'nww or *onfaacu~wr*' name in this rupor~ti**a not coa"ctsit.d i~rO- of a'iY Owuromial produot.

* I,

'h -E

UnclassifiedSECURITY CLASSIFICATION OF THIS PAGE (W~hen Dael Flr'a - 4TUCIN

REPORT DOCUMENTATION P~AGE - 19FORE COMPLETING FORMV.RPR NUMBER C~T. RECIPIENT'S CATALOG NUMBER

Technical Report ARBRL-TR-02410__________

fl'F.TITLE (antdSubtuite) 5. TYPE OF R9PORT A PERIOD COVERED

Extinguishing Ammunition Fires Final

6. PERFORIJING ORG. REPORT NUMBER

7. AUTHOR(s) 11. CONTRACT OR GRANT NUMFE-R4.)

Anthony E. Finnerty

9. PERFORMING ORGANIZATION NAME AND ADDRESS 110. PROGRAM ELEMENT. PROJECT, TASKUS Amy Ballistic Research Laboratory AREA & WORK UNIT NUMBERS

ATTN: DRDAR-EBLT . L162601AH91Aberdeen Proving Ground, MD 21005

II. CONTROLLING OFFICE NAME AND ADDRESS 12. REPORT DATE

US Army Armament Research 4 Development Command July 19~32US, Army Ballistic Rese-T.ch Laboratory (DRDAP.-BL) 1T. MUN6EP! OF PAGES

Aberdeen PT2 ing Ground, Mn rip1 314, MONITORING AGENCY NAME &ADORE*SS(f diffEfraitfi, Controlling Office) IS. SECURittY CLASS. (of title report)

Unclassified

Oii;:- DUS PIC ATION/ DOWN GRADI NO

16. DISTRIBUTION STATIEMENT (of thi fiteportO

Approved for public release; distribution unlimited.

17, DISTRIBUTION STATEMEV 'r (of it. absttact entered In Block 20, it different frow ieport)

1S. SUPPLEMENTARY NOTES

This report supersedes BRL IMR No. 721

IS. KEY WORDS (Continue oat revere* old@ it n~eesawy and Identtify by block numbw)shaped charges Halon 1301

IIfire extinguishers wae

M30 propellant

20. ABSTRACT (Continue an rerwev side It noo&eaeaelnad tdontity by block umimbs)

It is generally conceded that ammunition fires are the main reason for thetotal loss of armored vehicles in combat. We are investigating methods ofextinguishing these fires. We have thus far demo-astrated that it is possibleto extinguish propellant fires by the use of conventional extinguishing agents.Since propellant is the principal combustible in aimmunition campartments andother storage places, this opens up the possibility of controlling ammunitionfires. (Continued on reverse side)

AlDO~ 1413 KDITýON OF I NOV 651IS OBSOLETE Unclassified

..- ~.. ..... T:'Y CAFATI Of THIS PAGE (When Dae. BtereM

SEUIAA lýOO THIS PA6I(WIhen Date #nte@.vi-

'

20. Abstract (con't)

We have achieved the following iesults:

1. In the event of an ignition of a propellant bed, we can detect the£fire and activate extinguishers in about 30 milliseconds. It is necessaryto cool the burning solid in order to fully quench the fire. Water based foa shave been successfully used for this purpose. The water cools the burningmaterial while the expansion of volume due to the foaming additive all-wsgreater coverage from a given amount of extinguishant. Use of an additionalextinguisher containing Halon 130Cl allows even quicker extinguishment ofthe fire.

2. In the event of a shaped charge jet attack on a cartridge casecontaining M30 propellant, we can save 2 of the original 4 kg of M30, usingboth water based foam and Halon 1301. In addition, the propellant thatburns does so inefficiently with greatly reduced intensity. A lot of theheat liberated is absorbed by the water. The propellant burn wouldprobably not be catastrophic to either vehicle or crew members.

3. Shaped charge jets have been used to attack fully loaded 105 mmkinetic energy rounds. Most of the M30 propellant burned, but with poorefficiency. The flames associated with the burning process were signifi-cantly reduced. It is quite possible that crew members could have survivedthe burning of the propellant in the presence of the extinguishing agents.

t

4! a

SAD

IiCflYCLASSIPICATIOM OP- "HIS-PAGE;(Wren Dlat. RnteMd)

..... ----- ----1~

TABLE OF CONTENTSPage

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

LIST OF TABLES ........................... 7

I. INTRODUCTION ...................... .............. 9

II. EXPERIMENTAL .... .................. .. ....... 10

A. Materials ....................... 10

B. Procedures ......... ..................... .... 11

III. RESULTS AND DISCUSSION ..... ................. is

IV. CONCLUSIONS ....... ............ 24

DISTRIBUTION LIST ............... ................. 29

A.oossion yor

unannolc ad/O-

tDistribtion/.--.-C---'-'SAV ajjabilitY Code

Dis special

are

LIST OF ILLUSTRATIONS

Figures Page

1. Gas Chamber. .. .. .. ...... ....... ....... ....... .........12

2. Mo~dified Oxygen Tudex Flammability Tester .. .. ..... ... ...... 13

3. 10S mw~ Cart-ridge Case and Aluminum Sleeve. .. .. ... .........14

4. Field Test Setup. .. .. .... ..... .................16

*I5. Field Test Setup for Foam Spray. .. .. .. ...... ....... . .......

6. Field Test Setup with Extinguisher Secured at Box . - 18

7. Fates of a Grain of Propellant Under 3 Different Conditions 21.

s161

LIST OF TABLES

Tables Page

1. Chamber Tests Results as Flame Entered Chamber .... ....... 19

2. Results Using Condensed Phase Extinguishants ....... ... 19

3. Summary of Results Using Electric Match Igniter ........... 23

S4. Summary of Results Using 81 mm Shaped Charges and CartridgeCases . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

S. Summary of Results Using 81 mm Shaped Charges and M392A2 (105mm)Rounds . . . . . . . . . . . . . . . . . . . . . . . . . . 27

7/1

- -A

I. INTRODUCTION

The total loss of an armored vehicle in combat is generaily caused byan ammunition and/or fuel fire. In an effort to roduce such lo3ses, the U.S.Army is equipping its latest tank, the Ml, with a system to automaticallydetect and extinguish fuel fires. This ,Aystem incorporateýs the latest stateof the art electronic and mechanical components to achieve the millisecondrange response time needed to protect both the crew and the vehicle. Testshnve demonstrated that it is possible to detect a diesel fuel fire andextinguish it in 100 milliseconds.

In the Ml tank, ammunition is stored in a vented compartment at therear of the turret. This compartment is ncrmally sealed off from the crewvolume, so that the cEew and vehicle may survive a direct hit on the mainammunition supply. However, no technology has been developed to extinguishammunition fires. Even In an arrangement such as the Mlls ammunitioncompartment, a large portion of the ammunition may be destroyed even if onlyone round is struck initially.

r EIn all other US armored vehicles, including the M60 taak, ammaunitionr is stored in the crew volume using the vehicle's armor as protection.A hit which perforates the armor and strikes the ammunition will probablycause complete destruction of the vehicle and its crew. There is an obviousneed for an extinguishing system which is capable of quenching an ammunitionfire, thus limiting the destructive effects to only the round initially struck.K Therefore, a decision was made to determine if it would be possible, usingtechnology similar to that developed for the Ml tank, to detect and extinguishan ammunition fire in an armored vehicle in the short period of time betweena hit and a catastrophic event.

Previous tests conducted at the Ballistic Research Laboratory havedemonstrated convinc:Lngly that a direct hit by a shaped charge jet on theexplosive contgined in a 105 mm HEAT warhead causes immediate detonation ofthe explosive. However, other rounds positioned alongside the warheadtake from many milliseconds to several seconds before ignition or cook offoccurs. This is long enough for a rapid system to extinguish the fire.Therefore, in this case of a shaped charge jet striking a high explosivewarhead, there may be sufficient time to act and prevent additional roundsfrom exploding.

1. Dicker, R.J.L., "Countering the Crew Compartment Explosion,"International Defense Review 5, pp 1.20-122, 1.979.

2.Melani, G., Frey, R.B., and Carter, S.M., "Determination of the ResidualJet Parameters Required to Initiate Cased Explosive Charges," BRL MemoReport No. 2753, May 1977. (AD B0199?1L)

9 .

Looking at a different situation, in which attack by a conditionedshaped charge jet occurs on the propellant section of the -ound, normallyonly a small portion of the propellant reacts immediately. Most of thepropellant grains are scattered about to ignite and burn a short time later.A fast response extinguisher may prevent the burning of most of thispropellant from the round initially struck. This shculd prevent ignitionof adjacent rounds which have been cracked open by fragments from thefirst round.

II. EXPERIMENTAL JA. Materials

Three brands of fire extinguishers were employed. Two of these arecurrently in use in the Ml tank. These use extremely fast discharge ratevalves. They were manufactured by Crown Systems and Marotta ScientificControls, Inc. These extinguishers are solenoid valve types requiring a 24volt DC (10 amp) source for actuation. The third system, "Prt,-Calc"manufactured by Chemetron Fire Systems, is slower in initial response anddischarge time. However, it utilized a 110 volt AC solenoid operated pilotvalve. Since this was more couvenient to use and the extinguisher flow wasadequate for our needs, the, "Pre-Calc" extinguisher was used for most of thiswork.

The M30 propellant and the empty 105 mm cartridge cases were standardtypes. They and the M392A2 rounds were obtained from the APG AmmunitionSupply. These rounds are 105 mm discarding sabot kinetic energy type. Theyare one of the rounds used in the M68 gun, standard on the Ml and M60 tanks.These rounds were chcsen bpcause they are directly applicable to the problemwe are attacking and they do not present 'he problem in handling that shapedcharge rounds cause after they have been exposed to a fire.

The 81 mm diameter precision shaped charges were obtained fi.m the IWarhead Mechanics Branch of the Terminal Ballistics Division of the BRL.

The ethylene glycol used was standard automotive type inhibitedpermanent antifreeze.

Foam concentrate was purchased from National Foam Inc. Two types offoams were used, a standard one known as AOW+3 and a low temperature typecalled AOW3 + Cold Foam. Since ethylene glycol was used in formulating thefoam solutions, concentrations up to 20% foam agent were used, inste. ofthe normally recommended 3%. The high concentration was needed to obtain

i ~3,IlMaere2s, J.N., !erei,'.s, A.B., "Observations of Shaped-Charge Jet/M30 IProcil :nt Reactons," 3RL Tehnical Report No. 02108, Sept. Z978. (AD A062299)

10

SI•

good foam expansion at -140 C. At OC, 6% AOW3 + Cold Foma gave good resultseven with one-third antifreeze rresent in the formulation.

Photographic coverage was obtained by using two Teledyne Milliken 16 mmcameras. Normally one was operated at 100 fps, the other at either 500 or300 fps.

B. Procedures

1. An experiment was conducted to determine if it were possible to extin-guish propellant fires using conventional extinguishing agents. A train of over-lupping M30 propellant grains (the propellant used in the U.S. Army's 105 mmtank gun round) was constructed on a section of angle iron and placed into agas chamber. The train extended out of the chamber so that the propellant couldbe easily ignited using a propane torch. The chamber was approximately 75 cmlong by 35 cm diameter. A 12 mm diameter pipe was attached to one end of thechamber and a 60 mm hole cut into the other end. The angle iron with two over-lapping rows each containing 25 grains of M30, could be easily slid into or out ofthe chamber. A schematic of the test setup is given in Figure 1.

In use the chamber was fl, shed for 10 minutes with a test gas. The propellantwas placed onto the angle iron with some grains extending out of the chamber.The propellant was ignited and observations on the combustion behavior were madeas the burning progressed into the chamber. Depending on thle gas inside the box,combusticn either continued with visible flame or the flame was extinguished andonly fizz burning occurred inside the box.

2. In other tests, the angle iron, with M30, was placed on the g'oundand ignited. A stream of liquid extinguisher was directed at the burningpropellant. Observations were made as to how easily combustion was stopped.

3. The combustion of single grains of M30 was studied using a modifiedOxygen Index Flammability Tester. A grain of M30, wrapped in a nichrome wire,was suspended in a glass chimney. Mixtures of two gases, in this case extinguish-ant plus air, could be passed through the apparatus. The single grain could beignited electrically in the controlled atmosphere. The most efficient extinguish-ing agent would be the one which prevented flame with the highest percent of airin the mixture (lowest percent of extinguishant). A schematic of this apparatusis presented in Figure 2. The grain of M30 was suspended in the glass chimney andthe system flushed with test gas for 10 minutes. The nichrome wire was thenheated and observations were made on the combustion of the M30.

4. For field testing, a box was constructed from 25 mm thick steel. Theinterior dimensions were 41 cra wide, 74 cm high and 104 em deep. The front ofthe box was left open. This pe:.mitted high speed (500 fps) camera coverage evenwhen the front was sealed with a plexiglas sheet. A vent 30 cm by 41 cm was cutinto the top of the box. A pile of 4 kg of loose M30 was placed on the bottomof the box to simiilate a burning 105 mm round. An acceptor round containing4 kg of M30 was exposed to the flamhes from the burning material. The acceptorround consisted of a slotted sleeve and a slotted 105 mm cartridge case. Theslots were aligned to allow the Dropellant in the case to be exposed to theenvironment. Figure 3 shows the slots in the sleeve and case.

11

CHAMBER GAS INLET

2 ~ \ RAY CONTAINING 50 GRAINS M30 PROPELLANTPROPANE TORCH

FIGURE 1. GAS CHAMBER

12

PROPELLANT- NICHROMEF' GRAIN WIREU

GLASS BEADS-

TEST GAS -

FIGURE 2. MODIFIED OXYGEN INDEX FLAMMABILITY TESTER

13

155 MM X 25MM 1 U00 MM X 6 MM SLOTSLOT I, AL SLEEVE IN CARTRIDGE CASE

FIGURE 3. 105 MM CARTRIDGE CASE AND ALUMINUM SLEEVE

1

14 •'

An electric match and black powder were used to ignite the 4 kg ofloose M30. A solar cell detector signaler, a sequential timer when the firestarted. An exdinguisher was activated electrically and Halon 1301 or otherextinguishants n-uped into the test box. A plexiglas front cover was used in anattempt to keep the halon inside the test box. A representation of the testsetup is given in Figure 4.

5. Subsequent experiments were performed using water based foamingagents as fire extinguishants. For this work, the plexiglas shield wasremoved and the extinguisher situated 3 meters from the front of the box.A foam generating nozzle was fabricated and screwed into the discharge port ofthe extinguisher. Upon detection of a fire, the foam was sprayed in thegeneral direction of the box. Foam ,,as used to increase coverage of the spray.Only materials generally considered non-toxic were considered for theformulAtions. Therefore the materials in the formulation have been limitedto water, calcium chloride, ethylene glycol and convenrtional fire fightingfoaming agents. A schematic of the setup is given in iYigure 5.

6. In an attempt to model the real combat situation more closely, testswere conducted in which a cartridge case containing 4 kg of M30 propellantwas placed on the bottom of the test box. One hundred seventy-eight mm ofarmor were placed over the hole in the top of the box. An 81 mm shaped chargewas set up to fire its jet through the armor and into the box. The cartridgecase was positioned in such a way that the jet passed through the case and propel-lant or the case was placed out of the path of the jet. In the latter situationthe case was hit only by spall from the armor plate. Initially the extinguisherswere set up the same way as in procedure 5. However, the blast from theexplosive ii the shaped charge was sufficient to knock over the extinguishers.Therefore, holes were cut into the sides and back of the box. The extinguisherswere secured at the box so that the oxtinguishant would flow right into thetest setup. A schematic of the setup is given in Figure 6.

7. In this procedure, M392A2 rounids (105 mm kinetic energy type) wereset in the bottom of the box. Armor was placed over the opening in the top

of the box and an 81 mm shaped charge fired through the armor into the box.Conditions could be arranged so that the round would be hit by the jet orjust by spall. The extinguishers were secured at the test box.

III. RESULTS AND DISCO'SSIN

The chambcr experiments (procedure 1) showvd that it is indeed possibleto prevent gas phase burning of M30 propellant if it is subjected to theproper atmosphere. The results of these tests are given in Table 1.

i is11

7, 1,*

SLOTTE SLEEVE WITHSLOTED ASECONTAINING4 GM30 PROPELLANT

FIRE EXTINGUISHER

SOLAR CELL

DETECTOR

4 KG 'i3O PROPELLANT BED

FIGURE 4. FIELD TEST SETUP

-~ .- .. .- ~ .- .: .-

1|

SLOTTED SLEEVE WITHSLOTTED CASE CONTAINING

SOLAR CELL 4 KG M30 PROPELLANTDETEC T ORQ _ _

iI

/,'-4

1 4 KG M30

V- "PROPELLANT BED

FIRE EXTINGUISHER

FIGURE 5. FIELD TEST SETUP FOR FOAM SPRAY

17

I I -V .-

1 EXTINGUISHER

CASE WITH 4 KG 30 PROPELLANT

FIGURE 6. FIELD TEST SETUP WITH EXTINGUISHERSECURED AT BOX

18

Table 1. Chambor Tests Results as Flame Entered Chamber

Gas

Halon 1301 Flame extinguished.solld continued to fizz burnNitrogen Flame extinguished solid continued to fizz burnCarbon Dioxide Flame e.ttinguished solid continued to fizz burnArgon Flame extinguished solid continied to fizz burnHalon 1211 Intense fire inside chamber

In no case was it possible to stop fizz burning. However, it was feltthat fizz burning could be accepted as long as there was no flame to transferburning to nearby exposed propellant. The surprise was that the M30 burnedvery strongly in a Halon 1211 atmosphere. The Halon 1211 is considered anexcellent extinguishing agent for hydrocarbon fires.

In procedure 2, using liquid and 6olid agents directed at burning grainsof MSO, it was found that flame could be extinguished if a high vapor pressureextinguishant was used to displace the air in the vicinity of the burninggrains. Fizz burning continued. The evaporating agents were able to quenchthe gas phase burning but not the solid fizz burn of grains untouched byliquid. When all of the burning grains were covered with liquid or solidagent all combustion ceased. Here a cooling mechanism was sufficient tocompletely quench the M30. Results of these tests, using several condensedextinguishants, are given in Table 2.

Table 2. Results Using Condensed Phase Extinguishants

Agent Rasults

Water M30 grains continued to burn with flame even whenlower sections of grains were immersed in liquid. Onlywhen liquid was directed at burnirng grain was flame

', extinguished.

Liquid Nitrogen Evaporating gas extinguished flame. Solid continuedto fizz burn. All burning ceased when grains werecovered with liquid agent.

Carbon Dioxide Solid carbon dioxide formed on M30. Flamo wasextinguished. Solid M30 continued to fizz burn untila large quantity of agent was directed at the burninggrains.

These experiments demonstrated that it is possible to completely quenchthe burning propellant if cooling material is directed onto the gains.

Even though none of the gaseous extinguishants used in the chamber testsstopped fizz buining, four of the agents immediately quenched flames inside, the

19

~i~I

chamber. There was no way to distinguish among these four as to ei-iciency.This situation would probably be true in respect to any future agents to betested. Therefore, the single grain extinguishing experiment w~s tried. Inthis setup of procedure 3, the atmaosphere surrounding the propelloat graincan be varied. Initially neat gaseous agent was used. In subsequentexperiments controlled ratios of agent to air were to be tested. Unexpectedly,the single grain heaced electrically in neat Halon 1301 burst into flame.The evolved gases simply pushed the Halon 1301 gas away from the M30.Dilution of the evolved gases did not occur since the flow of the halon was veryslow and the volume of the system was small. While air may aid in flameformation, flame can be established if the evolved gases are in a high enoughconcentration. In the chamber experiments of procedure 1 the gas flow wasfaster and the volume of the chamber larger. T1he test gases were able todilute the evolved gases so flames were extinguished. These two experimentsshowed that a dilution miechanism is important in preventing flames. If thegases coming off a fizzing grain can be diluted with an inert gas, flames willnot be establishA. A representation of the possible behavior of a grain ofM30 under different conditions at atmospheric pressure is given in Figure 7.

In procedure 4, experiments using the steel box and Halon 1301 showedthat if the halon was piped into the box within several hundred millisecondsafter the solar %ell detected the ignition of the black oowder and M30, it waspossible to extinguish flames. The M30 continued to fizz burn. In a secondor two after the end of the halona flow, sufficient gases were evolveu tosurround the burning solid with combustible vapors. Flame reappeared andfull scale burning occurred. The acceptor shell was ignited in about fourseconds. This behavior, extinguishment of flame, followed by its reappearance,was the best experimental result obtained with Helon 1301. It became obviousthat it would be necessary to quench all burning of the M30 to prevent ignitionof the acceptor round. Liquids, with their inherently greater cooling effect,would be needed.

Tests were then conducted using water - ethylene glycoi mixtures. Afterdetection of fire by the solar cell, the liquid extinguishant was piped intothe steel box through 3 meters of tubing as in Figure 4. Several hundredmilliseconds elapsed between ignition of the black powder and delivery of theliquid extinguishant to the b rning propellant. It was found that a minimumof seven liters of extinguishant was required to quench the pile of 4 kgof burning M30. This large amount of extinguishant was requiredbecause the fire had become well established in the time span between ignitionand when the liquid actually got to the fire. However, the acceptor rounddid not ignite even though it was exposed to flame during this time.

In procedure 5, foaming agents were added to a water - ethylene glycolmixture to obtain a large volume of coverage. This means that the extinguishantdoes not have to be aimed directiy at the fire site. As long as the materialflows in the correct general direction, extinguishant will get to the fire.The premixed solutions were put into conventional "Pre-Calc" fire extinguishers.The extinguishers were merely aimed at the front opening of the steel box(no plexiglas covering the front). Using this method of getting material to

20

AIR I NCREASEDj COMBUSTION

FLOW OF INERT FIZZ BURN

GAS SOLID PHASE

BURNINGPROPELLANT

COOLIN•G INERT AGENT EXTINGUISHMENT

FIGURE 7. FATES OF A GRAIN OF PROPELLANT

UNDER 3 DIFFERENT CONDITIONS

21

the fire, flow of liquid extinguishant could be detected on the filmsseveral milliseconds after ignition. It was possible to extinguish theiM30 fire using only four liters of foam extinguishant. This is rationalizedon the basis that the quicker the extinguishant is delivered to the firesite, the less intense the fire and the easier it is to extinguish.Therefore less liquid is required to quench the fire and save the aczeptorcartridge case.

Experiments were then done using two extinguishers, one containing 7 kgof Halon 1301 and the other four liters of foam solution. Sincevehicles are already being equipped with halon to suppress hydrocarbon fires,it was reasoned that halon extinguishers would probably be activated bypropellant fires as well as by hydr.;carbon fires. It was found that the f13Meswere extinguished more quickly when the two extinguishers were used. TheHalon 1301, possibly due to lower viscosity, begins discharge about tenmilliseconds before flow of the water based foam begins. The halon is capableof stopping gas phase burning, as we had already discovered. The foam whichfollowed easily quenched the fizz burning propellant. Therefore,the twooxtinguishants together performed better than would be expected on a simpleadditive basis, since the halon cannot control the M30 fire by itself.Results of tests using electric match igniters are given in Table 3.

The thought of combining Halon 1301 and foaming extinguishant in onecontainer is very attractive. However, It is well known that Halon 1301and similar materials must be kept completely free of water since hydrolysisoccurs giving acid products. This is unacceptable in normal extinguishers.There is also the problem of expelling the halon before the foaming agent.The beneficial interaction of the two extinguishants depends on getting thehalon to the fire with the foan following closely behind. If a singleexti•aguisher were engineered to expell the halon first, there might beinsufficient pressure remaining to discharge the foam efficiently.

In procedure 6 a shaped charge jet was conditioned by passing through178 mm of armor and directed into the experimental box. For the firstshaped charge shots, the device was aimed so that the jet would miss thecartridge case of M30. Spall from the armor plate hit the case. Often the)spall merely peppered the case with no punctures or made only a few small holesin the case with no fire. The fine spall particles apparently lost theability to penetrate the steel cartridge case after passing through 56 cm ofair. In one instance the case was struck by a large spall particle (3pall ring)and fire was observed. There were also cases in which the small spallparticles did indeed cause ignition. Spall initiated fire is particularlydifficult to extinguish since the fire is inside the cartridge case. There isno efficient method of getting the extinguishant to the site of the fire. Inmost cases, the shaped charge was aimed so that the conditioned jet wentthrough the cartridge case and propellant. It was found that only about 50%of the M30 in the case could be saved, However, the presence of theextinguishant greatly reduced the intensity of the fire.

22

44 0.0.

oc~ I A4)t d) 4);.4

(A ~ 4 0 4 j P4 04 I4)4) -4 4J 00

lwOC. k 4.t. 4: t.4

*H- V-4

~ 4J

4)

0

.9-4 u

$4 4) 0~* 4.4 4)

0 40

0 0 4)

000

4.) 4V)V

.0 A0 4.)

41~

440 0 8~+ M

U)) CD+14 r00 0 H0 .ýo , n

23cdtn4

The M30 that burned did so at reduced efficiency as evidenced by greatlydiminished visible flame. Much of the heat that was generated was absorbedby the extinguishant. A large amount of the water was vaporized. Thisaction limited the temperature rise. Therefore, it may well be possible thateven though 2 or 3 kg of M30 were consumed, the temperature increase in thevicinity may not have been too great. Crewmen inside a vehicle might havebeen able to survive such an event. Typical results are given in Table 4.

As described in procedure 7, conditioned jets were shot into thepropellant section of M392A2 rounds. When only one round was used, about20% of its propellant could be saved using one foam plus one halon extinguisher.However, the intensity of the fire was greatly diminished due to the coolingeffect of the water-based extinguishant. Tests were done using two M392A2rounds in the box. These rounds were placed side by side. Thc jet hit oneround. Here also, about 20% of the propellant from the struck round wassaved. The second round was only dented. It neither burned nor broke open.The intensity of the fire was low, just as in the case of a single round.Results are given in Table S.

IV. CONCLUSIONS

The following conclusions may be drawn from this work:

1. Using readily available components, it is possible to detect andquench a fire initiated by an electric match buried in a bed of propellantfast enough so that it should not become a catastrophic event.

2. Halon 1301, commonly used as an extinguishant for hydrocarbon fires,is not sufficient to extinguish a fire in a bed of M30 propellant.

3. Water based foams are sufficient to extinguish fires in propellantbeds.

4. A combination of Halon 1301 and water based foam, in separate Ipressurized containers, extinguishes a propellant bed fire even more quicklythan foam alone•.

5. When the propellant in a M392A2 round is initiated by an extremelyenergetic source, such as a shaped charge jet, rapid application of bothHalon 1301 and water based foam saves at least 20% of the propellant andreduces the intensity of the fire associated with the propellant that does Iburn.

6. Use of appropriate extinguishing agknts can reduce the destructiveeffects of propellant fires. This should lead to an increase in thesurvivability of armored vehicles and lessen the danger to crewmembers.

24

.1i

IA

k) 0

+j 4r-4 V- V

0 A

4 1 V1.

4 1 . 4

0 O93 4 5 2 4

L) 4) U'i

0v-4 %0*U 0 0 zU)0 -4 44 .41 40

0-44 p4.J -H'bb 4.M4 4D()0 D -

1 4 -i V- co

94-444 1414-U)000 0

I. 'o,-

44-

C4 *ro

4)N1 4.) m) )

"+ 4J 4).

!ý 1.J1 N1O

'-)'4 0J v-I to n

4 w14u 0N0 0n 0 4)(n rq'0.4

'A P WO +G -i)4

4-1 0. PA p. + 4 14 ~ +:' .40 th + p )4 4

0"W 4J (d0c

,-) 4-4" co wo0) -44

r4 .r4 00 0 0u t i

'd t4.4" 4..)

U.. 44) 4

-44 0 o 4 "44j) "d 41 4) A.r. 14 ow 0) ~co1. . v! Q 04 A 4

Or-) P8 U)24. 4) 1944 lln 0 U00t4.)~~C H.4." )04 4.000

4144 -4 i In 04 0 r4J0 414 04.H~C/ A %)Q 0113'

p.. P"U)n 1 I 4 1011 J4 .4

C) I) 4)4)4 4)

0~ bo0 u0 0 b

25

U) coI >r

4.1

44.1

'4J

0 z

4) 4)%0C a

104

0

r. 44

4.1 ý1 44) 4) 4J

0~ 4.1

W1 00. b1

*i.Hf W W rf 0 wba

0 14

26 4

4J 4)4- 1 4)U

"H too 4-)~

0 444) .1 4J) .0P4 0 0 j . 4 0

o 1:14 r4 r

0 .U)..A

0 0~4 4J4- 100 r

4-1 0U) *H)1 J 00.

41

Ud 4

4- 4JC 4)

z z2 4~+ U CUd

bO "~4'ý

ognU) G)~ ~ 4~ ~ )U4C

W) tn~ -ct ý -H $4 P

W~ ~ 4-4WC +a\0C 4.J+

'4-4~ ~~ ~~ +~~. ~ U (~'*- C

41 -P 0n kO OI0 -4'0P4 C

4j. d) U)'

4) -p r4-J Q 4c

44 .* cd N %D H t.C N 'i -"

o2

- ~(D

ACKNOWLEDGMENT

The author wishes to thank Mr. Lowell Bryant who was responsible fora large portion of the experimental program, including all ignition anddetonation aspects. The assistance of Mr. Charles Stumpfel is greatlyappreciated for providing technical assistance tn acquiring and interpretingrelevant data. Mr. Stanly Mahan provided timely and skillful recharging (andsometimes repair3 of the fire extinguishers. Mr. Bernard IzdebFki's aidin providing the electronic work for the solar cell detectors and thesequotial timer is appreciated.

Encouragement and support of this project was provided by Mr. KarlBrobeil of the Tank *ind Automotive Command, Warren, Michigan.

2i8S~28,

1 i

DISTRIBUTION LIST

No. of No. ofCopies Organization Copies OEanization

12 Administrator 1 ComitanderDefense Documentation Canter 'US Army ARPADCOMATTN: DTIC-DDA Bexiet Weapons LaboratoryCameron Station ATTN: DRDAR-LCB-TL

Alexandria,VA 22314 Watervliet) NY 12189

SCommander 1 CommanderUS Army Materiel, Development US Army Communications Rsch&. Readiness Command & Developmient (Comuiand.

ATTN: DRCDMD-ST ATM'N: fDRDCO-LPPA-SA5001 Eisenhower Avenm.a Port Monmouth, NJ 07703Alex~andria, VA 22333

F C~ommaitzder

3 Commander US Army Electronics Research .1-US Army Armament Research • Development Command

i Development Command Technical Support ActivityATTN: DRDAR-TSS (2 cys) ATTN: DELSD-L

DRDAR-TDC Fort M~nmouth, NJ 07703Dover, NJ 07801

3 Commander 2 rommanderUS Army Armament Research US Army Missile Command& Development Command ATTN: DRSMI-RATTM J. Frasier DRSMI-YDL i

J. Pearson Redstone Arsenal, AL 35898

G. Randers-PehrsonDover, NJ 07801 Commandsr

US Army Tank Automotive1 Commander Research & Development Cmd

SUS ,-my Armament Materiel ATTN: DRDTA-ULReadiness Command Warren, MI 48090

ATI,,: DRSAR-LEP-L,Tech 'Lib.Rock Island IL 61299 Director

US Army TRADOC Systems1tConander Analysis Activity

US Army Aviation Research ATTN: ATAA-SL:,Tcb Lib& Development Command Khite Sands Missile Range"ATTN: DRDAV-E NM 880024300 Goodfellow BlvdSt.LouisMO 63120 Commander

US Army Ma,'erials & Mechanics.1 Director ReýearLh Center

US Army Air Mobility Reserach ATIN: Techrnical Library& Development Laboratory DRXMR-ERAmes Research Center Joe PriftiMoffett Field, CA 94035 Eugene De Luca

Watertown, MA 02172

29

I,.

DISTRIBUTION LIST

No. of No. ofCopies -. ranization Copies Organization

3 Commander 1 Michigan Technical UniversityUS Army Ndtick Resftrch ATTN: W.Predebon

SDevelopment Cad Houghton, MI 49931ATTN: DRXRE/Dr.D.Sieling

Williaw Crensh~w 2 Southwest Research InstituteArthur Murphy ATTO: Dr.C.Anderson

Natick, MA 01762 "Mr.T.Jeter8D00 Culebra Rd.

Commandier San Antonio,TX 78228Army Research Off:2ce 2 CoumanderATTN: E.Saibel 2 Comman AorP.O.Box 12211 US Army Tank Automtive R&D (;m

Research Triangle Park ATTN: DRDTA-RCAF

NC 2'77%9 Mr.Karl Brobeil, r.Charles BeaudetteSCommander Warren, MI 48090

ATTN: M.e.Bacpman, Code 383S 1 Marotta Scientific Controls,Inc.

China Lake, (A 93555 ATTNSMr.Thomas RogersBoonton Ave.

National Aeronautics 4 Space Boonton, NJ 07005

Admini strationLyndon R.Johnson Space Center Program ManagerATTN: B.G.Cour-Palais 11 Abramns Tank System

nTX 77058 ATTN: Mr.Terry DeanWarren, MI 48090

N&2ional Aeronautics 6 SpaceAdministration Aberdeen Proving Ground

Langley Research CenterATTN: D.H.Humes Dir, USAMSAAHampton, VA 23365 ATTN: DRXSY-D

DRXSY-MP, H. Cohen

3 Kaman AviDyne Cdr, USATECOMATTN: Mr.R.Milligan ATTN: DRSTE-TO-F

Mr.G.Zartarian Dir, USACSL, Bldg. E3516, EAMr.R.Yeghiayan ATTN: DRDAR-CLB-PA

83 Second Avenue Dir, USAMTDNcrthwest Industrial Park ATTN: S. MahanBurlington, MA 01830 T. Julian

30

7

USER EVALUATION OF REPORT

Please take a few minutes to answer the questions below; tear outthis sheet, fold as indicated, staple or tape closed, and placein the mail. Youir commeiLts will provide us with information forimproving future reports.

1. BRL Report Number

2. Does this report satisfy a need? (Comment on purpose, relatedproject, or other area of interest for which report will be used.)

3. How, specifically, is the report being used? (Information

source, design data or procedure, management procedure, source ofi ideas, etc.)

4. Has the information in this report led to any quantitativesavings as far as man-hours/contract dollars saved, operating costsavoided, efficiencies achieved, etc.? If so, please elaborate.

5. General Comments (Indicate what you think should be changed tomake this report and future reports of this type more responsiveto your needs, more usable, improve readability, etc.) _.

6. If you would like to be contacted by the personnel who preparedthis report to raise specific questions or discuss the topic,please fill in the following intormation.

Name:

Telephone Number:

Organization Address:

.I ._._._ ._._._._._._._._.