WIXLIB

2.The'ISummary andDiscussion of Resultstests discussed in thispaper showed thatthe bulb-type devices when used as hand extinguishers were less effective than the 1-qt pumpgun carbon-tetrachloride extinguisher. In thefirst place, the capacity of the devices was, ingeneral, too small to be practically effective and,in the second place, the manner in which theextinguishing liquid was applied made the devicesless effective than the pump-gun extinguisher forthe same volume of liquid used.The first test showed that none of the devicesput out a relatively simple test fire that the 1-qtpump guncan usually extinguisheasily.Theshowed that to put out this test fire,more (in most cases much more) extinguishingliquid was required.The third and fourth testsshowed that even with test fires reduced in extentand severity to a point where they could be extinguished easily by not more than one-half of thesecondtestcharge of a 1-qt pump-gun extinguisher, thedevices did not put out the test fires except inthe case of one of the units in one of three trials.With the bulb-type extinguishers, the liquid isnot as thoroughly broken up and as readily vaporized as it is with the pump-gun extinguisher, fromwhich the liquid is forcibly projected in a relatively compact stream that changes to a fine spraywhen it strikes objects in the fire. Furthermore,when the liquid is applied by sprinkling from, acontainer, it cannot be as effectively directed as itcan be with the pump gun and, apparently, a considerable portion of the liquid serves to dilute theburning gasoline rather than to cause the formation of a smothering gas blanket.In the trials in which the test fire was notextinguished, the flames were somewhat diminished during the application of the liquid, but thefuel continued to burn with the production ofdense, acrid fumes in great volume.The testswere made in a room with a total volume ofa test space of 2,000 ft. 3 This enclosure containeda single window 3 ft 4 in. wide by 4 ft 10 in. high,with the bottom 3 ft from the floor. It was provided with a fireproof shutter, pivoted horizontallyat the center, which was so arranged that it couldbe set at any given inclination to give the desiredventilation.A tightly fitting door was constructed in the partition to give access to thetest space.This door was kept closed when thetests were in progress.The space in the buildingoutside of the test space contained three wndows,each 3 ft 4 in. by 4 ft 10 in. and one door, all ofwhich were kept open during the tests.During each test, the temperature of the airsurrounding the extinguishers, as well as the temperature at various other points in the test space,was measured with thermocouples. The progressof the test fires was observed by temperaturemeasurements taken at various points in the pathsof the flames.fullapproximately 90,000 ft 3When 1 qt of extinguishing liquid was applied in the manner described above, it was found that shortly afterwardthe fumes became so irritating that the operatorwas driven from the room, and they became soopaque that it was impossible to distinguish objectsat distances greater than 3 or 4 ft.IV. Tests of Devices asAutomaticExtinguishersThe tests of the devices when used as automaticextinguishers were conducted in one of the Bureau's fire-test buildings that contained a singleroom 29 ft long by 15 ft 3 in. wide by 10 ft 1 in.high. The walls were of brick and the floor andceiling, concrete.fireproof partition was erectedacross the building 13 ft from one- end to provideA1.ExtinguishersAtoFAccording to the claims made forAfireextin-F whenpresented for the tests, thetheory of operation was that should a fire occurin the protected space the temperature would riseand the release mechanism would cause the dischargeof the carbon tetrachloride, whereupon the liquidwould vaporize and be carried by the convectioncurrents over the burning area to smother the firewith an oxygen-excluding blanket. According tothis theory, these devices were intended to placea smothering gas blanket locally over a small areaof burning fuel and not to provide vapor sufficientto dilute the atmosphere of the protected space toa point where combustion could no longer be posThissible because of the low oxygen content.theory excluded the use of the devices for fires inordinary combustible materials for which carbontetrachloride when locally applied is not consideredsuitable, and the fires used in the tests were, therefore, confined to fires of flammable liquids (class Bguisherstofires).Because of the danger involved in the handlingof gasoline in a small room, oil rather than gasolinewas used for the fuel. The oil was a mixture oftwo parts of light engine oil with one part of kerosine.The fuel was contained in a pan located onTheFires of two sizes were used.the floor.larger of the two consisted of 2 gal of burning fuelFor the secondin a pan 3 ft sq having 6-in. sides.fire, }i gal of fuel was used in a pan 2 ft sq having1-in. sides.The number of extinguisher units recommendedby the makers was one for each 1,000 ft3 of spaceAccordingly, trials were first madeprotected.with two units in the 2,000-ft test space. Another series of trials was then made, in which thenumber of units was doubled33

Results of Tests(a)ExFirst test with two extinguisher units.was chosen for this test. As showntinguisherin table 1, this extinguisher had a capacity intermediate between that of the smallest and theto F.Inlargest extinguishers in the gioupeach trial, the two units were mounted 5}i ft abovethe floor on the vertical center lines of the twoend walls of the test space.In every trial, the test fire pan was on the floor.Trials were first made with the pan at the centerof the test space (location a), then with the pannext to one of the 13-ft walls directly under oneof the two extinguishers (location b), and finallywith the pan in a corner of the test space (locationThe three locations of the pan are shown inc).Location b is also indicated in figure 3.figure 2.(1)CAWINDOWtest were the following: Extinguisher A, one ofthe two extinguishers having the smallest capacity(0.44 qt) of those tested; extinguisher F, havingthe largest capacity (1.00 qt) and in one trial,extinguisher E, having an intermediate capacityL 1;i,(0.84 qt).TEST- FIREPANINLOCATION CEXTINGUISHERJ!IEXTINGUISHERINTEST- FIREPANLOCATION aTEST- FIREINPANLOCATION bCLOSED DOOR,-2-11TwoAto Ffirst series of trials of devicesas automatic extinguishers.units used; 3-foot-square test-fire pan in trials 1 to 3; 2-foot-square panFlGTJRE2.Plan forin trials 4 to 8.During these trials, the shutter was tilted sothat the ventilation through the window, bothabove and below the shutter axis, was just enoughto permit the fuel in the test space to burn freely.The results of the test are given in table 3.extinguisher units.(2) Second test with jourThe second test, in which four rather than twoextinguisher units were used in each trial, was madebecause of the negative results shown in the firstThe extinguishers chosen for the secondtest.4FigureThe3.testAtoTrial 5, first series of tests of devicesas automatic extinguishers.pan was located directly beneath one of the extinguisherunits, asshown.F

ITableTestResults of tests of devices3.AFtoas automatic tityAreaLoca-oftiontestofpanfiregal222.do.do - .do.doVlHH.doityunitsunitswhichFuelof .extin-guisher?cabDurationof ionfireput tionNumber 0100100Less14141512min141414110.711.911than 1YesLessthan X3aAadddAEFFSee figure 2 for test 1, and figure 4 for test 2.2 The ambient temperature at 1 extinguisher was 205 to 210 F for \\ i minand at the other extinguisher, 200 to 250 F for 2}i min.3 The ambient temperature at the 4 extinguishers for 10 min during the test1In each trial of the four samples under test, twowere mounted on the 15-ft 3-in. wall and two onthe partition opposite.The extinguishers weremounted at a height of 5}i ft from the floor andwere spaced 3 ft 10 in. from the 13-ft walls, asshownin figure 420.43044 0131514131313131313was 370 to 415 F, 255 to 310 F, 355 to 410 F. and 245 to 285 F, respectively.4The ambient temperature at the 4 extinguishers for 10 min during the testwas 230 to 300 F, 275 to 310 F, 315 to 355 F, and 2o0 to 300 F, respectively.Onaccount of the nature of the results obtainedin the first test, it was not considered necessaryin the second test to include all the test conditionscovered in the first. Therefore, only one of thetwo test fires was used, the fire of 2 gal of oil andkerosine in the 3-ft-sq pan. The trials were madeat location a with the pan on the floor at thecenter of the test space, and at location d, withthe pan on the floor adjacent to one of the 13-ftwalls.During thetrials,the lower half of thewindowin the 15-ft 3-in. wall was sealed off, and the shutter was tilted to leave an opening 40 by 20 in. inAdditional ventilation was prothe upper half.vided for these tests by an opening approximately40 by 20 in. in the fireproof partition opposite.Both openings were above the midheight of thetest space, which was a condition recommendedby some of the makers of the devices tested. Thetotal area of the ventilating openings amounted to4 percent of the wall area above the midplane ofthe enclosed space. Trials with no extinguisherspresent showed that these openings provided justenough ventilation to permit the test fires to burnfreely until the fuel was consumed.The resultsof these trials are given in table 3.Figure4.Plan for secondseries of trials of devicesAtoFas automatic extinguishers.Four units used;3-foot-square test-fire pan.5

Summary(b)of ResultsThe tests of the devices in group A to F whenused as automatic extinguishers were conductedin a room of 2,000-ft 3 capacity in which the ventilation was sufficient to permit the fires to burnThe testsfreely until all the fuel was consumed.showed that when two extinguisher units(ex-tinguisher C) were used the devices were noteffective in extinguishing or checking relativelysimple test fires located on the floor, except whena small fire was located directly under one of theextinguishers so that all its liquid was promptlydischarged into the pan of burning oil. When theunits were doubled in number and located so thatno extinguisher was directly above the test fire,the devices (extinguishers A, E, and F) failed toextinguish or check the fires either when theywere in the center of the room or against the wall.The thermal release mechanisms of some of theunits failed to operate as intended.2.Extinguisher2 in. apart in each layer,Those of each layerwere laid transversely to those of the preceding'layer.A cube was thus formed approximately24 in. on the side, with air channels to affordready combustion. The cube was built onsupports Iji ft from the floor and was ignitedby K gal of gasoline in a pan placed 1 ft below11the cube.Theresults of the test are given intable 4.GExtinguisher G was in the nature of an installedsystem that operated on the principle of dilutingthe atmosphere of the entire protected space to apoint where combustion would no longer beThepossible because of the low oxygen content.claims for this device did not exclude fires inordinary freely burning materials, as was the caseHence,for the extinguishers in group A to F.in the tests of this extinguisher, trials were madeboth on fires of freely burning material (classFigure5.fires offlammable liquid(classBautomatic extinguishertrials ofwoodGon'Gonffire.Two extinguisherAandfires)Plan forunits used.fires).The ventilating openings in the test space werethe two 40- by 20-in. windows used in the secondto F.series of trials of extinguishersThe number of extinguisher units specified bythe maker for extinguishing fires in ordinary freelyburning materials was one for every 1,000 ft 3of space protected and for putting out fires ingasoline, oil, and like materials, one for everyAccordingly, in the trials on the wood750 ft 3fires, two extinguishers were used and in the trialson the oil fires, three extinguishers. The unitswere suspended from the ceiling on the longitudinalcenter line of the test room and were equallyspaced, as shown in figures 5 and 6.A.(a)Results of Testswith two extinguisher units onFor the class A fires, one of the woodfires frequently employed as "standard" for the2%-gal soda-and-acid hand extinguisher was used.(1)classFirsttestA fires.The fuel for this fire consisted of 78 pine sticks2 by 2 by 24 in. long.The sticks were arrangedin 13 layers.They were laid parallel and aboutFigure6.Plan fortrials ofoil fireonautomatic extinguishernfloor.Three extinguisher units used, 3-foot-squaretest-firepan.t

Table4.TestResults oftests ofGautomatic extinguisherAVasAreaof testpanLocationof firegalWood 3.adodOil3X33X33X323-doand wood3used3.-of unitsextin-whichguishers?Fuel con-sumedIduration2of o10040100See figure 5 for test 1 and figures 6 and 7 for test 2.fires, this was the time when the stack collapsed.See description in text.One unit failed to operate, although the ambient temperature at the extinguisher was about 400 F for 8 min.One unit discharged cnly one-half its contents, although the ambient temperature at the extinguisher was 390 to 480 1of fire,operated22TESTOil.of unitsby theft121Numberof oilTEST231NumberDurationfireput city13«423»132222For the wood23466Durationof oil fire.WoodFforover 2 min.continued to burn. 8 with three extinguisher units onFor the class B fires, 2 or 3 gal of\class B fires.the mixture of two parts light engine oil and onepart kerosine were used in the 3-ft-sq pan havingl(5-in. sides.The results of these trials are givenin table 4.The third trial in the second test was madebecause it appeared probable that, since the firein the previous trial was located directly underone of the extinguishers, the extinguishment ofthe fire had been accomplished by a local gasblanket rather than by dilution of the wholeatmosphere of the enclosed space. Hence, in thethird trial, the same test fire was used, but thepan was placed beneath a wooden bench and about18 in. off the floor, as shown in figure 7.(2)Secondtest(b)Summary EXTINGUISHER ',4 FTSQUARE BENCH TOPTTEST-FIREPANIN LOCATION fELEVATIONof ResultsG were conducted in2,000-ft3 room used for similar tests ofsmaller extinguishers. As in those trials,sufficient ventilation was provided to permit thefires to burn freely so that they would not extinguish themselves.As indicated in table 4, in both of the trials withfire) thea fire of freely burning material (classextinguisher, when one unit was used for each1,000 ft3 of space, failed either to extinguish orThethethetests of extinguishersameAcheck the fire.In the tests with the oil fires (class B fires), inwhich one unit was used for each 667 ft3 theextinguisher failed to put out or check the test,firesintwoof the trialsand extinguished thefireIn that trial, the fire was placeddirectly beneath one of the units.It appeared probable that, under the conditionsof the third trial, the fire was not extinguished bydilution of the atmosphere of the enclosed spaceinonebutbytrial.a local blanketing effect.PLANFigure7.Plan andtinguisher Gwood bench.elevation for trial of automatic exon oil fire in S-foot-square pan shielded byThree extinguisher units used.7

in the number recommended by the manufacturer,and six of them were also tried in double thatV. Conclusionsnumber. 2The tests indicated clearly that when handoperated the bulb-type carbon-tetrachloride extinguisher, as represented by the samples tried,was distinctly less effective than the 1-qt pumpgun extinguisher, the smallest generally consideredacceptable, principally by reason of the mannerof hand extinguishers recommended for use on fires in flamfires) are described in the Manual of Fire-Loss Preven(classHandbook of Fire Protection,tion of the Federal Fire Council and the1BNFPATenth Edition. The Manual may be purchased from the Superintendentof Documents, Government Printing Office, Washington 25, D. C, at 30cents a copy. The Handbook is published by the National Fire ProtectionAssociation, 60 Batterymareh Street, Boston 10, Mass.2 Water-sprinkler systems are commonly employed for the automatic protection of enclosed spaces in which the combustibles are of the usual freelyburning solids. For flammable-liquid fires in enclosed spaces, automaticallyoperated systems employing carbon dioxide or foam are sometimes installedwhich the liquid was applied.devices when employed as automatic extinguishers were not effective in protectingThe seven devices were testedthe enclosed space.1inOther typesmable liquidsThe bulb-typeWashington, Julyo24, 1950.

fire Extinguisher Desig-nation I'upae-ity Methodofoperation Number ofunits used Was fire extin-guished? Test1 ft qt 1 4X4 A 0.44 Throwing No 2 4X4 B i.44 Sprinkling. No 3 4X4 C.59 do No 4 4X4D.75 do No 5 4X4 E.84 do No 6 4X4F 1.00 Throwing. No Test2 1 4X4 A 0.44 Throwing. 5 No 2 4X4 0.59 Sprinkling. 3 Yes 3 4X4 C.59 do 4 Yes