£r advanced flame arrestor materials ^ … tr-72-12 & £r advanced flame arrestor materials ^...
TRANSCRIPT
AFAPL TR-72-12
&
£r ADVANCED FLAME ARRESTOR MATERIALS ^ AND t- TECHNIQUES FOR FUEL TANK PROTECTION
§ Quentin C. Malmberg
Edwin W. Wiggins
MCDONNELL AIRCRAFT COMPANY
TECHNICAL REPORT AFAPL-TR-72-12
March 1972
Approved for Public Release; Distribution Unlimited
R«prodiK»d by
NATIONAL TECHNICAL INFORMATION SERVICE
SpnnglKld Va 2JI5I
Air Force Aero Propulsion Laboratory Air Force Systems Command
Wright-Patterson Air Force Base, Ohio
[Coo
THIS DOCUMENT IS BEST QUALITY AVAILABLE. THE COPY
FURNISHED TO DTIC CONTAINED
A SIGNIFICANT NUMBER OF
PAGES WHICH DO NOT
REPRODUCE LEGIBLYo
NOTICES
When Government drawings, specifications, or other data are used for any purpose other than in connection with a definitely related Government procurement operation, the United States Government would thereby encour- age no responsibility nor any obligation whatsoever; and the fact that +he Government may havo formulated, furnished, or in any way supplied the said drawings, specifications, or other data, is not to be regarded by any implication otherwise as in any manner licensing the holder for any other person or corporation, or conveying any rights or permission to manufacture, use, or sell any patented Inventlo.-.s that may in any way be related thereto.
Cop lea of this report should not be returned unless return is required by security considerations, contractual obligations, or notice on a specific document.
m f,'»fwn«(... a,
i-
saifled S«-c»nty Classifit »nnn
DOCUMENT CONTROL DATA R&D iSerunty clmtaificmtion ol titl«, body uf nbstrmct find indetiinf unriuimtion mw-l bf mntmrvd whT» Ihr uvermll report t» clmaallitd)
1 ONiCtNATlNC ACTiviTv (Corpor»t9 muthor)
McDonnell Douglas Corporation P. 0. Box 516 St. Louis, Missouri 63166
tm. REPORT iECuRtTV CLASSIFICATION
Unclassified 2b. CBOOP
J R«:PO"T riTLE
Advanced Flame Arrestor Materials and rochnltiues for Fuel Tank Protection
« DESCHPTivC NOTES (Typ» ol rmport and Imtutiv. dmtrm,
Final Technical Report - 28 December 1970 - 26 November 1971 9 AUTHONiii (Ftrtt nmnt», middt» iniumt. imat n*m»}
Quentln C. Malmberg Edvin W. Wiggins
ft NEPOR T DATE 7«. TOTAL NO OF PAGES
iue Tfc. NO OF RE FS
•«. CONTRACT OR GRANT NO
b. PROJECT NO 30^8
r Task 30U807
d. Work Unit 30U807038
9«. ORIGINATOR'» REPORT NUMhtR(»»
9b OTHER RE PUNT NOiti (Any other number» thmt mmy b» amal^imd thl» npotl)
AFAPL-TO-72-I2 10. DISTRIBUTION ST A TCMCXT
Approved for Public Release, Distribution Unlimited
11 SUPPLEMENT ART NOTES I* SPONSORING MILITARY ACTIVITY
Air Force Aero Propulsion Laboratory Wright Patterson Air Force Base, Ohio ^33
13 ABSTRACT
Tbe purpose of Phase I was to develop and test concepts for minimizing the weight and volume displacement penalties of polyuretbane foam explosion arrestor suppression systems. Both structural and integral isolation concepts for arrestor voiding techniques were investigated. For fuselage tanks the integral concept of large hollow cylinders offered th^ greater percentage void (5öt) for unpressurized tanks while *he voided foam lined wall configuration was the better approach for pressurized fuselage tanks. The small six-celled wing tank egg crate pattern provided for 95 and 87* void at 0 aid 2 psig initial system pressure respectively while for the large wing tank kt$ void at 0, 2 and 5 psig initial system pressure was possible.
Phase II was the materials investigation portion of the program and evaluated flame arrestor effectiveness, fuel flow resistance and thermophyslcal properties of representative candidate materials and configuration. The most efficient of the sixteen material candidates from an arrestor effectiveness standpoint was 3M polyester Scotch Brite felt. The material thermophyslcal properties of thermal conductivity, specific heat, density and surface area have only a small effect on material explosion suppression performance.
Collation of data from Phase I and II was accomplished In Phase ill of the program.Qnpirical relationships for the test data were developed through coii5)uterize( regression analysis and the relative Importence of the applicahl* variables was determined.
DD ,^.,1473 ,PAGE " I NOV
S/N 0101.807-630 1
Unclassified Securit, ClassificMtion
Uncla«sifled Security t'Ussihcation
K F v «OMD«
Flame Arrestor Cor.figurations Void Percent Polyurethane Foam Fuel Tank Protection Explosion Arrestor System
DD;r..1473 '**«'
NO L C I * T
1
Unclassified Scruritv CI«»»ific«lion
^^^^^^^^^^^■^^a^Ma j
ADVANCED BLA)^ ARBZSTOP MATSRIALr. ATO
TlffT-TW Ri FUEL TAt.TC PHOTSCTIOÜ
iuentln C. I'Almberg Edwin H. Ulgfilns
Details of Illu?frnti11■? ?n " this dorumcnl niay bl : • tter
s, studied on niicrc'.ciie
Approved for public release; distribution unlimited
. --
I
FOREWORD
Thle report waa prepared by ). &• Mairaberg and E. W. Wiggins, Survivabllity/Vulnerabillty Design Section, McDonnell Aircraft Company, McDonnell Douglas Corporation, St. Louis, Missouri. The work reported herein was carried out under Contract No. F33615-71-C-1191, Project No. 30^8, "Advanced Flame Arrestor Materials and Techniques for Fuel Tank Protection," and was administered by the Fire Protec- tion Branch, Air Force Aero Propulsion Laboratory, Air Force Systems Command, Wright-Patterson Air Force Base, Ohio with A. J, Ferrenberg as Project Engineer. The period covered by this report is December P8, 1970 to ".'ovember 26, 1971.
This report was submitted by the authors December 27, 1971.
This technical report has been reviewed and is approved.
itf^&jZ- B. P. Botteri, Chief Fire Protection Branch, Fuels Lubrication and Hazards Division, Air Force Aero Propulsion Laboratory, Weight-Patterson Air Force Base, Ohio
11
ABSTRA:T
Tlie "Flame Arrestor Miterials and Techniques for Fuel Tank Protection" work program was subdivided into three phaces. rhases I and II involved tho testing of reticulated polyurethanc foam configurations and installation tech- niques and other candidate materials for flame arrestor effectivencsn. Phase III consisted of the reduction and analysis of all the recorded iatu.
The purpose of Phase I was to develop and test concepts for minimizing the weight and volurce displacement penalties of polyurethane foam explosion suppression systems. Both structural and integral isolation concepts for arrestor voiding techniques were investigated. These two appro«jhes adjust weight and volume displacement penalties with respect to allowable fuel tank structural limits and system over-pressures. 'The structural isolation con- cept utilized natu—'l aircraft structural compartnentization whereas the integral isolation concept used closed compartments within the foam itself in the form of hollow bodies. For unpressurized fuselage tanks the integral concept of large hollow cylinders offered the greater percentage void (58.5"') while the voided foam lined wall configuration was the better approach for pressurized fuselage tunks. The mont effective foam configurations for small and large simulated wing tank systems were the egg crate and lined wall concepts. The small six-celled wing tank egg crate pattern provided 92 and 87*5 void at 0 and 2 psig initial system pressure respectively, while for the large three cell wing tank U?' void at 0, 2 and 5 psig initial system pressure was possible. Limited tests on the lined wall configuration showed 80 und l»7]t voiding to be possible for the small six cell and largo three cell wing tank systems respectively.
Irstallation and fabrication techni-iues were addressed throughout thlc portion of the program and the hollow body and egg crate designs proved to be the best approach to simple Installation and fabrication.
Phase II w .6 the materials investigation portion of the program and evaluated flame arrestor 'ffectiveness, fuel flow resistance and thermophysical properties of representative candidate materials and configurations. The most efficient of the sixteen material candidates evaluated from an arrestor effec- tiveness standpoint was 3M Scotch Britc. Following this came the Scott Paper Co. fire extinguishing foam and the 2) pore per inch (ppl) low density reti- culated polyurethane foam materials. Fluid flow tests reversed the sequence of these materials from a pressure drop performance standpoint, '.vetting agents and coatings improved arrestor effectiveness to only a smell degree but showt-u hhnt with the ^rorer material configuration they could contribute significantly to reduce flame penetration. The material thermophyr-ical pro- perties of thermal conductivity, specific heat, density and surface area exhibited only a small effect on material explosion suppression performance.
Collation of data from Phases I and II was accomplished in Phase III of the program. Empirical relationships for the test data were developed throu^ computerized regression analysis and the relative Importance of the applicable variables was determined.
Hi
TABLE OF GONT'HTG
SECTION
I
II
PAGE
INTRODUCTION AMD nUMMAHY 1
PHAJE 1 TEST PROGRAM 5
1.0 TEST GET ITT 5
1.1 Teiit specimen 5
1.2 Instnmontatlon 5
1.3 Explosive Mixture 5
1.1+ Ignition Gyoteir. 10
2.0 TEGT CONFIGURATIOA' 10
2.1 General 10
2.2 Fuselace Tank Ccnfigurations 10
2.2.1 Lined Wall Conficur:ition 10
2.2.2 Voided Lined .-.'all Configuration .... 10
2.2.j Lari:o Dlaiwter 'lollow CyLiad*rt (Flat inda) 15
2.2.1+ Gnall Dlaineter Hollow Cylinderc. .... 15
2.j '''ine Tunk Configuratioris 15
3.0 TEGT PMCBOWS 15
k.O PSGULTG tSD DZSCU88X01I OF RSSÜLTS 25
k.1 Fusela^e Tank Re suite 27
'+.1.1 Lined Vail ;onf I cjuratior. 2?
k.1.2 •.'oiled Lined ..-alls 2?
^.1.3 Large Jianeter Hollow Cylinlcr.. .... 28
I*.I.1- .;mall Diomoter Hollow Cylinlcr^ .... 26
U.1.5 Fuselafie Tank SuOMOy 29
U.2 3ix Cell Ving Tank Refills ?9
Preceding page blank V
TAbLi: OF coi.TrriTC (co:'r':))
ACTION
III
rv
l'.?.l 31X Cell L: : I
h.2.? six Cell 3g| 'r. ; rieuratlo
U.S.j Six C«ll Larno !(ollow Cyllnden
U.C.-1. L; x Cell SMll Hollow Cylindere
'••2,5 Si A. Cell Configuration SuOBBry
'•.:, Thruu "ell Wing Tanks Resulte ....
U.j.l ntreo Cell Lined .-.'all ....
»...2 Ihroo :oli agg Creta ConfJ '•jrat
''.i.j Throe Call onflguretlon Sunmar,
coMCLurio::;" Atn RSCOK-EW^ATIONS
6.0 RSCOfl'-JTOATlO!© . . ,
PHA5Z XI I2OT r^rTA:: . ,
1.0 K3T CiT-UP
l.i Coabuotlon Cct-Up . .
1.2 Instruacntatlon . . ,
1.3 Sxploalv Mixturu . .
I.- Ifinltion ryotea . . .
1.5 p'tttl Flu« Pr-.-.-.'.r: Drop "Jt-Cr
1. ;.l "UJI Flow Tcut Instrunentntlon
1.6 ntemophysi sal Pr^iorjtlea Test Set-U
1.6.1 Theraal Coniuntlrlty ....
1.6.8 Sp elf la »«at
1.6.3 alffsrontlal Theymü taalyzcr
1.6. • ;;u•f^^l,■■ Area
LL-. 2;
29
29
30
30
30
30
30
a 33
33
33
79
79
79
79
79
79
79
60
80
00
bo
60
80
vi
SECTion
VI
TABLK OF CONTCMTT. (CODT'',))
PAGi'
2.0 MATERIAL TEST CONFIGURATION; 80
2.1 rioneral 80
2.2 Material Conficurations 80
j.O ÜBST PROCSnüRE 81
3.1 Combustion Touts 81
3.2 Fuel Mow Tsttl 61
3.3 Thermophy., KV'J. Properties Tests 82
ii.o RoCXTr. AM,, niGcussio:; OF RESULTS 83
U.l CatJbustiOtl Tests 83
l>.2 Flow fest Pesidts 8^
'-.} Ihennphyalcftl Proptrti«« Tests 85
COMCLTffiKWS AHD RECOMKENDATIOUS 87
5.0 coacLuaicaB 87
-J.O RECa-BESD/WlOIIS 87
PRASE III 137
1.0 ANALYSIS OF PPOCPAl'. RESULTS 137
1.1 General 137
1.2 Jxploslon Suppression ^ysto-n "'.odel Ij7
1.2.1 Pnase I Dat. and Model Ar.alysic, .... 1J9
1.2.2 Phase I Computer RtgreUlon Analysis . . 1^0
1.2.3 Phase II Data Analysi:-. : M
1.3 Explosion Suppression 3ysterr.s//!rcr,ift Paranetcr Cons idontions 1^2
1.3.1 Fuel Tank ConfiGurntlons 1^2
1.1.1.1 Fuselage Tank lU2
1.5.1.2 'Jiag Tanks 1*3
vil
T/TL: QV :o:iT.:::zr. (corrr1:')
üj;cTiori
1.3.1.3 Advanced Material!
l.j. . .VM-T. Jyctor. Punaltiea
i.j.r:.! Croai Voided Ponra Fange r-nalti ■;
I.J.P.? Grots Voided FV»ar. Cros- 7aico»Cff . . ght Panaltles
l.>.?.3 AdditlMial Val^Jt rar.ilticü . .
l.j.3 Syoten Sffact«
1.3.3.1 PUsl "low
L.3.3.2 ?Mol Level Sff»CtO
1.3. ■ Installation Ctawiieratlona
l.-.-.l ::■.•*■ Aircraft Cyatcaa
I.;.-,? Ratroflt ."yct^:...
■'rrjr. i;:.; : - |?csulta if the Äroraatoßraphlc Analysis cf Propona TaiBplee . . .
PAGE
1^3
1-j
1*3
Ik}
Ikk
Ikk
1UU
1UU
145
U5
U5
vlll
FIGURES
Fltjure 1
Figure 2
Fl.-rur-' :
Figure ■
Fl.'urc 5
Plgura o
Fifauro 7
Fieur-; 3
Figur« 9
Figure 10
Figure 11
Figure 1?
Fijure 13
Figure I1'
Figure 15
Figore 16
Fifure 17
Figure 13
Figure 19
Figure 20
Fl.;ure ?1
Figure 22
Figure ^3
Li:'T OF .--i.:'
_j
Fuselago Fuel Jell Tut .ft Up ^
SSanlatad Six C«ll Win« Tank 7
■ . • '• :: . 8
Test Iressure Trace 9
i ir ■ fml T tnk Foari Conflguratlona 11
iiicd „'all Conflffuratlon 12
Voii »all ;:eg:nentü I1*
15 Inch Oianeter Cylinder In^tallnt u r. • SlBulr.t« :'..•■ Lage Twk 16
7.5 Inch Dlonetcr Cylinder l.-utall-ition in Cinulated F'.selr., ; TanX 17
Typical 7.5 Inch "iar.etcr Cylinder k I ::orr.ispher;o:'.l Elude 10
Three Cell 41ng Ta::x Fcan Ccmflguretlona 19
Six Cell Wing TaaK Foon Configuration 20
.".cherr.atic of ■ ,jg Craft rrengement 22
Foxr. Sgg Crate 23
Tnrcc Cell Wing Tank ^gg Crate Configuration .... 2k
Fuselage Tank Test Set Up 26
01 Poem Wall Veld Configuration kk
10" Foam '.-.'all "oil Configuration ^5
l\i' Foar. tfall Voi-i Configuration k6
25.1 Foan ./all .'oid Configuration V7
15 Inch Diameter Cylinder Configuration kB
7.5 Inch niameter Cylinder Flat 2r.i Configuration . ^9
7.5 Inch Plameter Cylinders (Roil End) Configuration 50
ix
LIST OF FIGURES (COTT'lp
FIGUR5G rAJE
Ficure 2!» Pressure Ratio vr,. Relief to Combuütion V'llame Ratio - Lined Wall 51
Ficurc 25 Prescure Ratio vr,. Relief to Coir.bustion Vclume Ratio - 10 percent Voided Lined Wall .... 52
Figure 2D Pressure Ratio vr,. Relief to Combustion Volume Ratio - 15 percent Voided Lined Hull .... 53
Figure 27 Pressure Ratio vs. Relief to Combustion Volume Ratio - 25 Percent Voided Linel •-.'all .... ^
Figure 28 Pressure Ratio vs. Relief to Combustion Volume Ratio - 15 Inch Diameter Voided Cylinder . . 55
Figure 29 ProsRure Ratio vs. Relief to Combustion Volume Ratio - 7»5 Inch diameter Voided Cylinder Flat ?MM 56
Figure 30 PTMSur« Ratio vs. Relief to Combustion Voianc Ratio - 7.5 incn DlMWter Voided Cyllnd'.-i.' Honi "ndi 57
FlL;ure 31 Frejsur-- Ratio tf. Relief to Combustion Volume Ratio - 10 percent Voided Lined Wall (15 ppl Foam) 50
Figure 32 Fuüelagc Tank Results 59
Figure 33 Six Cell ..'ing Tank ügg Crate Toütc 6k
Figure 3U ZU Cell Wlnr Tank Cyllnd-r Tests 65
Figure 35 Six Cell WlOfl TaiJt Cylinder Teste 66
Figure 36 Pressure Ratio vs. Relief to Combustion VolUM Matio - Egg Crate Gix Cell ■;ing Tank 6?
Figure 37 Pressure Ratio vs. Relief to Combustion Volarie Ratio - 15 Inch Diameter Cylinder Clx Cell Wing Tank 68
Figure 38 Pressure Ratio vs. Relief to Combustion Volume Ratio - 7.5 Inch Diameter Cylinders Six Cell Wing Tank 69
Figure 39 Six Cell Wing Tanks Results 70
' LIST OF FIGURES (COMT'D)
FIGURZS PAGE
Figure Uo Three Cell Wing Tank Tests (Lined Walls) 73
Figure Ul Three Cell Wing Tank Egg Crate Tests 74
Figure ^2 Pressure Ratio vs. Relief to Combustion Volume Ratio - Lined Wall Three Cell Wing Tank 75
Figure U3 Pressure Ratio vs. Relief to Combustion Volume Ratio - Egg Crate Three Cell Wing Tank 76
Figure 1*U Three Cell Wing Tank Results 77
Figuro U5 DTA Trace - 25 PPI Foam 88
Figure HO DTA Trace - Fire Extinguishing Foam 89
Figure 1*7 DTA Trace - 3M Scotch Brite 90
Figure U8 DTA Trace - 25 PPI Foam Coated with Polysuiride . . 91
F!pure l»9 DTA Trace - 25 PPI Foam Coated with Glas. Hesin . . 92
Figure 50 DTA Trace - 25 PPI Foam Coated with Redar Viton . . 93
Figure 51 3TA Traco - 25 PPI Foam Coated with Kel-F 91*
Figure 52 WA Trace - Polyester Screen 95
Fit'ura 53 DTA Trace - I.'omex Honeycomb 96
Figure 51* Flow vs. Pressure Drop of 25 PPI Polyurethane Foam Uncoated and Coated with Polysulfide Ill
Figure 55 Flow vs. Pressure Drop - 25 PPI Polyurethane Foam Plated with Copper and Aluminum Tube Core 112
Figure 56 Flow vs. Pressure Drop - Fire Extinguishing Foam and Polyester Pelt 113
Figure 57 Flow vs. Pressure Drop - Aluminum Honeycomb .... ll1*
Figure 58 Flow vs. Pressure Drop - Aluminum Honeycomb .... 115
Figure 59 Flow vs. Pressure Drop - flomex Honeycomb 116
Figure 60 Thermal Conductivity of Flame Arrestor Materials . . 118
Figure 6l Specific Heat of Flame Arrestor Materials 119
jci
^JH
LIST OF ngtgEg (CONT'D)
FIGURES
Figure 62
Figure 63
Figure 64
Figure 65
Figure 66
Plgora 67
Figure 68
Figure 69
PACE
Precs'iro Ratio va. Relief to Conibustion Volume Ratio Ploxlglasr. Tube Tests 133
Pressure Ratio vr.. Relief to Conbuation Volume Ratio Plexiglass Tube Tost.; l^h
Pressure Ratio vs. Relief to '"ombust 1 on Voltune Ratio Plexiglas;-, Tube Test: . , 135
Pressure Ratio vs. Relief to Combustion Volume Plexifdass Tube Tests 136
Single Coll todel ]37
Multi Cell rodol 137
Multi Cell Nod«! 138
Foan V'eight an3 Volume Penalties - ^ of Fuel vs. Fuselage System Voiding l'+7
xll
mi
LIGT OF TA3LZG
L.::-
r-bic i
raala tl
Tahlo in
T.'.bl,; iv
Tabln V
:-.t ■. VI
:VnU VTI
Tttbla ■.'III
Table K
Tatto ::
■:-.ibi- :<i
?oblf XII
" :ble .'ITT'
?■ bi • ■•:T:
r-.bic ::v
'nbl xvi
roUc CTII
raU.« A'.'ITT
i1ablf :<ix
ri.bi- AX
.abl'- XXI
Vo'dtd '/oiiia Lined Wall Ccmri(;uratio!i
'.CB Crate Conflivirntlon Dlnonsiono
Fuselage Fuel Tank Jystem Tost i'ata ./all Configuration
Puaelogi FU.JI Tank iystem Tt>st 'lata Foan '-yall Conf If.urat lori
ruMXage Fui;l T ink Syatwn T^st "at-i ■;-.' cyltwtera (i; :nch i)ir..) . . .
PuMlaoo Fttal Tank Systm fcit Data 3ai CyllMoru (7.5 fodi :na.) . . .
"\iEclat;,; Fuel Tan/. Syatm T^Jt .iat:i Seottidierlcal Bnd Cylinders (7.5 In
Purclage Pual Tan!; Systan Test )ata Vbldad 'oar. ..' 11 CcT-.flf^iratlnn (15
-1.. . -1
r-x Coll
Six C«ll
Thr
;!. rank Teot üat« .
lr.»5 Tan;: Test Data .
Ing rar.'rv Tjst )ata .
Ins ^-•rr- *e*t 5ata .
^•11 •:.■ - Toni? Lina ! ..all
coll rina Tank T-st ^te
I'jbXeriaX -onf'.gjratlona ....
Müt^rlal Properttaa
MtarlaX Te»X :-i(^t
Mt «rial .'--J. .- ■• : r.ssuru .Yjp
l'.nt-Tl!.! Cüornophyi.l.-al rroporti..'
ttoteriaX Codbustlon Tests . . .
pfc&s« 1 ^ysi.T- Void Nrcontap .
)at!
Line 1
■o 11--'.
•Int.
■'lat
h 3laJ
''AC"
13
21
31*
35
37
uo
42
60
61
62
63
71
72
97
96
101
109
117
120
IM
xlll
SECTIOM I
ir.TBOSUCTIOH AUl) GlTJ-g^Y
Fuel tank fires and explosions are a major cause of aircraft LOSBM in oombat. Considerable research and dovelopmcnt has been duvotdi to .xplorin;: fuel tank explosion protection concepts. Nltroccn dilution, chenieal iuench- ine and polyurethane foaii void filler material omerce as ttW prlr.ury candidate systems. Of these the passive, logistics free polyurethanc fOM syötons appears Ideal. Light-weight (lov danslt;') fcan and /iross voidln;-; techni.xuec are used to reduce the weight and vol>ino penalties encountered in contemporary foam system installations. Previous work by KCAIR, in cooperation with Tcott Paper Co., successfully demonstrated a low-density reticulated polyurethanc foM explosion suppression system with 80 to 1^0 percent voiding. Xhia degree of voiding had only been demonstrated in tankr. that were Jublivided into a number of intercommunicatint; cells such as aircraft wing tank:; which are inherently segmented by ribs and spars, V/here compa-tmentization is not inhi.ront as in the case of aircraft fuselage tanks, a lesser percent voiding is possible for equal allowable combustion over-pressures as evidenced by the work performed by Bureau of Mines where up to 1*0^ voiding was achieved for .'ingle cell configura- tions.
Basically two types of ^ross voiding concepts pre ;.T.tly exitit. The first is a structural isolation design as in aircraft vint; fuel tanks where the structure offers natural compartmentizatlon with intercommunicating open- ings between cells. Foam is used to isolate firer- to the combustion cell by acting as a flame arrestor, stopping the flume propogntion to the adjacent cells. Pressure generated by the combustion process in the ignited cell is relieved through the foam and intercommunicating hol^s. The second variation of this concept uses hollow foam bodies to provide flame isolated compartments with walls of sufficient thickness to locally isolate fires. Combustion pressure is relieved through the foam and into isolated volumes. 3oth of these concepts have been investigated in Phase T of this progrnm with an arbitrary system success jriteria of 10 psi allowable combustion over-pressure.
The Phase I effor». of this program was designed to improve and optimize installation concepts and techniques for foam fire and explosion suppression in simulated aircraft fuselage and wing fuel tanks. Configurations designed to accomplish these progrp.n objectives were established to provide data that would optimize the system operation from a foam void standpoint. Foam volumes were predetermined for all systems in order that test void volumes could be increased or decreased in increments of 5^ of the total tank volume. Combustiori tests were conducted at each void increment and the pressure an: temperature in each cell of the specimen was recorded. Kach configuration was tested at 0, 2 and p psig initial pressure with successively larger or smaller void volumes until an over-pressure from combustion of 10 psi was reached. In some cases whore it was obvious that data could be successfully extrapolated, the tests were completed when sufficient data were obtained to establish a curve.
Phase II investigated material flame arrestor effectiveness with respect to combustion over-pressure and fuel flow resistance. ThermophysicO.. project' . determinations of candidate arrestor materials were conducted Including thern:-.! conductivity, specific heat, melting temperature, heat of fusion, bulk density.
specific fuel retention and ourfacj area. Combustion testing was divided into three tusks: (l) arrestor material confifuration and screening tests, (2) basio material evaluation tests and (3) coating tests. The success c-itericn of the arrcstors was based on a combustion over-pressure limit of ?5 PCI.
All material combustion and fuel flow pressure drop tests were con- ducted in an eight-inch diameter plejcifjlass tube lu.ing a material thickness of two inches. Combustion tests wore conducted at 0, 2 and 'j pslg initial system pressures and flow tests were run at 50, 100, and 150 ^prn flow rates using JP-U as the fluid media. Thermophysical properties tests were conducted on those materials that performed satisfactorily through at least one set of combustion parameters.
Data from phases I anil II of the program were reduced and collated dur- ing Phase III. Application of the data from the Phase I effort was directed toward weight and volume displacement panaltles as well M system advantages and disadvantages for design of explosion suppression systems for both retro- fit and new aircraft. Arrestor effectiveness data was analyzed to determine the relationships of chemical interactions and system parameters and the relative significance of those parameters on the performance of the material as a flame arrestor. Jmpirical relationships were uovelopcd by computer regression analysis with all applicable data included. These relationships give direction to the future development of flame arrestor materials by pointing to the parameter's relative effectiveness toward flame suppression.
Jata from the program have load to the following conclusions;
o The naxi.iium void percentage obtained in these tests for the simulated 100 gallon aircraft fuselage fuel tank is 58.5'/o «t 0 psig initial system pressure using a 10 pslg over-pressure success criterlcn and was accomplished using large (15 inch diameter) hollow foam cylinders.
o The ten percent voided foam lined wall fuselage tanks configuration offers the lightest weight and subsequently the largest void percent- age (52 and ^7,9%) system of those tested for 2 and 5 pslg initial pressure.
o Egg crate type patterns offer thi greatest degree of design freedom as well as the most efficient flame barrier system with the greatest amount of void volume (ggft at 0 oslg, 8?' at 2 pslg, and JY, at 5 psir initial test pressures) of the ccr^ipurations tested for the six-cell simulated aircraft wing t:mk.
o The egg crate voiding configuration performed to the 50'' void volume level for the three-cell 300 gallon simulated aircraft wing tan!;.
o Systems with a number of small voids may not be as uffectlve as those with fewer larger voids.
o All void vapor volume In the hollow body configurations tested burned when the system combustion over-pressure reached 5 psig from the initial ignition.
o Successive ignitions using the same foam arc possible for fire and explosion suppression systems.
13 ppi foam Is not as efficient as 23 ppl foam In voided fire and explosion suppression systems at percentages greater than 20i.
The performance of hollow body configurations Is hindered by the fact that for a projectile simulated line source Ignition, combustion occurs simultaneously Inside and outside the foam wall, thus burning a greater portion of combustibles within the foam.
Combustion volume to foam thickness is the primary design parameter for gross voided foam explosion suppression systems.
Ihe use of other matttrials in a gross voided configuration for explosion suppression is possible.
3M polyester felt, "Scotch Brite," showed the best possibility as a substitute material of those tested.
Where fuel flow pressure drop is an Important system operation parameter the 2? ppl Scott reticulated polyurethane foam appeared to be the best material tested.
Wetting agents are more effective in eliminating burn through than reducing combustion over-pressure when the relief to combustion volume (Vr/Vc) ratios are small.
The thermophysical properties investigatt1 have a negligible effect on the explosion suppression ca, abilities of the material.
Arrestor material geometry appears to be the mo^t Important parameter in eliminating flame propagation with foams and felts being the oest configurations tested.
:
SECTION II
PHASE I TEST: PRCXSAM
1.0 TEST SET-UP
1.1 Test Specimen
■niree specimens were used throughout Phase I testing. The specimens were (1) a 100 gallon fuselage tank, (2) a 300 gallon 6 cell wing tank, and (3) a 300 gallon 3 cell wing tank. The three specimens were assembled from six 30 x 2U x 15 inch elemental boxes. Each elemental box was basically a steel angle iron frame which had match drilled sides so that the elements could be assembled in any combination. The boxes wert de- signed with both steel and plexiglass side and cover plates capable of withstanding Uo psig over-pressure. Plexiglass covers provided the capability to photographically record the explosion events.
The fuselage tank was constructed by assembling two sets of francs and panels to produce a 30 x 30 x 2U inch tank (Fig. 1). This siae was selected to meet the military standard 100 gallon fuselage test tank so that the data generated could be ccrapared to other similar work.
The three and six celled wing tanks were assembled from the six elemental frames to produce 90 x 1*8 x 15 and 90 x 30 x 2U inch tanks respectively. Figures 2 and 3« The three celled tank had, 55^ cell to cell intercommunication while the six celled tank was provided with only 5^ cell to cell intercommunicating open area.
1.2 Instrumentation
The Instrumentation consisted of strain gage type transducers, and uQ gauge cromel-alumel thermocouples in each cell of the respective specimen. The pressure, temperature and ignition time data was recorded on oscillo- graph traces at 15 Inches per RtOMl as shown in the sample trace of Figure U. The thermocouple outputs were mei only to monitor ignition and flame propa- gation In that combustion tempe. azures exceeded their useful range.
1.3 Explosive Mixture
Premlxed propane/air mixtures near stoichioraetric conditions were used as the combustible media in each test and the foam was wetted with JP-5 fuel. This simulated explosive aircraft tank condition was used for test simplicity in lieu of JP-1* flushing. If JP-U had been used, st:;ichlometric vapor/air conditions could have been achieved only by cooling the entire test article oelow Uo0F. The use of propane/air JP-5 permitted testing at all ambient and specified initial pressure conditions. Previous tests using propane/air proved that it produces results equivalent to JP-V^ir combustion.
Preceding page blank
FIGURE 1 FUSELAGE FUEL CELL TEST SET UP
GP 71 739 4
Bomb Sampler
Ignition
Oill
FIGURE 2 SIMULATED SIX CELL WING TANK TEST SET UP
GF71 0937 25
FIGURE 3 SIMULATED THREE CELL WING TANK TEST SET UP
GP71 093/ 26
8
Bomb Sample Test For Propane'Air Mixture Combustion Effectiveness Value
96 psl Simulated Fuselage Fuel Tank Comjustion Pressure Rise With 40% Void Lined Walled Configuration at 0 psig Initial Pressure
psi '—7.7pji *—7 7 psi
Ignition Ignition P1P2P3(10ps' 'nchl P4 (25Psl/inchl
Nott Paper Speed 15 inches per second
FIGURE 4 TEST PRESSURE TRACE
GP71 0;t37 39
I.1» Ignition Bystga
The ignition aystem useil wa'-. a stanluri B\ureau of MilWI laiilnous tub^ tnnsfonmr ret-up operatoi at (ü to 73 volts primary. The i(;nitor portion of the ^et-up was varied Tlightiy to simulate a straight line projectile path Lgnltion aourc"?. Thiu was aceompuiihed by placing the ignltor In a perforated tube which extended fron wall to wall, simultaneously i^nttlnt all individual voids along its path.
2.0 TT'ST COr.'FTGUPATIOKS
2.1 General
Tha test eonflgimtioni for all three upcclnenr, were desired around the structural and integral voiding concepts. The percent volllng ranc«-' was based upon the "CAIR raodi.'l of Arrestor Suppressed ^xplosionr. Tills ^odel relater. the t anK over-pressure to the ratio of relief ^lumc to com- bustior. volume, and the initial pressure. Using t;.i.- model a;,' the prof;ra.T. leflnod success criteria of 10 psig over-pr'.-ssure the initial test vola was calcul-ited an I the testing proceeded fron; that point in + j''-. incremental void Jhaii^os. Initially all test configurations were teste 1 using 2^ ppi, I.36 Ib/ft-^ reticulated polyurethane foan. After the best configuration for the '"uselage tank specimen was established, 15 ppi, loo lb/ft' r^tlcy'ated polyurethane foan was tested In that configuration.
2.2 Fuselage Tank :onf'r^uratic:
Four fuselage foam configurations were investigated. These configura- tiDns were (a) lined walls, [h] voided lined walls, (a) largt flat end hollow cylinders and (1] small flat and hemispherical ended cylinders shown in Figure $• These configurations wire varied in total void percentages as described in the following paragraphs.
2.2.1 Lined '.■.'all 'Jonfiguration
I-'odel analysis of tlM lined wall configuration for Initial pressures of 0, 2 and p pslg, predicted that the maximur. allowable voii percentages would be 1*6, ^S» and 39t void respectively. This variation in void per- centage was adopted and predetermined thicknesses of foam to obta'n 5^ incremental voids from '„Z to ö percent were stacked on the specimen walls. Tests wore run by removing the inner most layer of foam after each ignition.
2.2.2 Voile 1 Line! .;all "onfiguratlons
This configuration was basically the sane as the lined wall config'iratlon with the exception that sixteen voids within the foam were included as shown in Figure 6. The total void perc:ntTge was the sam of thusc internal voids and the center void. The total void " for the configuration was varied from 30 to 60"' using 10, 15 and PS" internal voids nnd 15 to -5"' center voiding. The void sises and wall thickness used ^re given in Table 1 and a photej-raph of an actual set of voided wall segments is presented In Figure 7.
10
Mn
Foam Foam
Lined Wall Configuration A
Voided Lined Wall Confiquration B
Foam Foam
Large Diameter Hollow Cylinders
Configuration C
Small Diameter Hollow Cylinder?
Configurations D&E
FIGURE 5 FUSELAGE FUEL TANK FOAM CONFIGURATIONS
GP71 0937 27
11
tmmm ftfWw ■■ *> nt r
16 Total Foam Voids per Systenn
iiv ■mi .1 ■ ^Mllliiti'i li'li
Center Void
lynition Void Volume —
■wT^frpr^ri*r'^^??'T^^TTT^r^pT'^
-^ , ■« ■>■■»<
• •'- -'
«MiiliäUiilMMiilMilMdät
• Ignition Location (Typ 3 Places)
FIGURES VOIDED WALL CONFIGURATION
GP71 093 7 33
12
—
V:)
iDID
r"'
fJ!
LlU.
..ICD
W..u
..L C
lm.~
l'GU
H.it
l.'I
OH
FD.ft
11 V
OID
DD
~I.:
>IOl
l3
,.....~--.
.,~~-~~~F~
o!SI~:. ~V:
o:i-d':::!
..---.-~--
~-~[_· C
~ter y
oid
'· l,
To.al.
I Fo
.., ~oJ
.dT-~o
.., ~o
id -~I F
o~~ ~~
~~~· fT
~f!l
I ~~
lll
o..
(i
n)
(~)
L,-
;-rr
;;)T
(%)-
1 lo
luu
e (r
n3
) O
ute1
· Jm
. (:u
>.
Inn
er D
im.
(in
) C
en
ter
Voi
d [
(%) ~
11
1.6
X 7
.JJ
X ll
10
.0
:_30
X 1
).2
X
9.2
2(
).5
20,5
00
.J0 X
28
X
22
J0 :
:t 2
4.8
X
1<1.
8 4
.8 i
n.
)0.5
l 12
l IJ
4 5
16
7
I 8
.....
\,o
J 9 1
0
11
12
13
14
lS
16
17
1.6
X
7.)
3 X
ll
10
.0
J0 X
1
6.4
X
10
.4 l
25
.0
1.6
X 7
.Jj
X 11
1
0.0
30
X 1
7.6
X
11
.6
29.9
1.6
X 7
.JJ
X
]1
10
.0
JO X
J.
S.S
X
12
.8
J5.J
1.6
X
7 .J
J X
11
1
0.0
30
X
19
.8 X
1
).8
14
0.0
1.6
X
7 .J
J X
11
1
0.0
.J0
X
21
.7 X
1
5.7
5(
).0
1.6
T..
7.J
J X
11
1
0.0
.J0
X 1
9.8
X
18
.8
4Q.0
1.6
X
7 .JJ
X
11
10
.0
.J0 X
2
0.8
X l4
.S
45
.0
1.6
X
7 •. J
J X
11
1
0.0
.J0
X 1
3.8
X
12
.8
)5.)
1.6
X 7
.JJ
:A 1
1
10
.0
J0 X
19
.S X
1
).8
4
0.0
2.1.
X 7
.JJ
X 11
15
.. 1
.J0 X
1
8.8
X
12
.8
J5.J
2.4
X 7
.JJ
X
11
15
.1
30 X
1
7.6
X
11
.6
29
.9
2.4
X
7.J
J X
11
1
5.1
J0
:.:
17
.6 X
1
1.6
2
9.9
1.6
X
7 .J
J X
11
1
0.0
.J0
X 2
1.7
X
15
.7
5().
0
4.0
X 7
.JJ
X ll
25
.0
J0 X
1).
6 X
7.6
1
5.0
4.0
X 7
.))
X
11
25
.0
JO :
.:: 1
5.2
X
9.2
2
0.5
4.0
.X 7 .)
) X
11
2
5.0
)0
:;
16.!,
. :<.
10.1~
25
.0
J0 X
2S
X
22
)0
X
24
.8 X
1
8,8
.J0 X
28
X
22 i JO
X
24
.8 X
1
8.8
)0 X
28
X
2.~
J0 X
2
4.8
X
18
.8
.J0 X
28
X
22
.J0
X 2
4.8
X.
18
.8
.J0 X
28
X
22
.J0
X 2
4.8
X
18
.8
JO X
28
X
22
)0
X
24
.8 X
1
8.8
.JO X
28
X
22
.J0
X 2
4.8
X
18
.8
.J0 X
28
X
22
J0
X
24
.8 X
1
8.8
.J0 X
28
X
22
.J0
X
24
.8 X
1
8.8
J0 X
2
8 X
. 22
JO
X 2
).2
X
17
.2
.J0 X
23
X
22
.J0
X
2).
2 X
1
7.2
JO X
. 28
X
22
)0
X
2).
2 X
1
7.2
J0 X
~ X
2
2
.J0 X
2
4.8
X
18
.6
JOx
.28
x2
2
)Ox
20
xl4
J0:;
;2ax
22
J0
x2
0x
l4
4-2
in
.
).6
in
.
].0
in
.
2.5
in
.
1.5
5 i
n.
2.5
0 i
n.
2.0
0 i
n •
J.OO
in
.
2.50
in
.
2.2
0 i
n.
2.8
0 i
n.
2.8
0 i
n.
1.5
5 i
n.
).2
0 i
n.
2.4
0 i
n.
1.0
0 in
.
35.0
'39
-9
45
.J
. 5
().0
60.0
,50.
0
55.0
l
45-3
50.0
50
.4
145.0
45
.0
6o.O
4() ..
0
u.s
so.o
FIGURE 7 VOID WALL SEGMENTS
GP7i 0937 34
Ik
2.2.3 Laree Diameter Hollow Cyliniers (Flat Ends)
Void percentages for the 15 inch diameter 1? inch Ion?: -jylinder:. were varied from ^0 to 60,y). Two sets of cylinder;: with varied wall and end thicknesses were used for all testing. The firct Get of cyliniers were configured with 2.75 inch walls and 1.0 inch ends and wori' used for tests 1 through 6 Given in Table 5. The second set was fabricated vith 1.95 inch wall and ends and used for test number 1. Alteration of this sot to 1.85 inch wall and end thickn^sü was made for tests 3 through Ij. Tests 11 and 12 used the first sot of cylinders with 2,J« inch thl^k walls and ends. Figure 8 is a photograph of this configuration installed in ths, fuselage tank.
2.2.'» Small Diameter Hdlow Cylinders
Two configurations of 7.5 inch diameter hollow cylindcrri were t:-3t i, flat end 2h inch long cylinders and heni spherical end 12 inch Ion;; jylir icr... The void percentage range for each configuration was obtained by varyin,: the inside diameter of the cylinders. The flat end configuration was tent- ! fro- 30 to 50^ total void while the hemispherical end cylinders were tested I'rorn 1*0 to 60t total void. Figure -,< is a photograph of thrt installed flat Dfi cylinders and Figure 10 is a photograph of a hemispherical end hollow ojl:..i.-,.
J.3 Wing Tank Configurations
Two tank specimens were used, a three and six cell sinulaxti 300 callon wing tank previously described. Lined wall and egg crate foam eonflRuratloni' us described for the fuselage tanks were evaluated in these specimen". 1. addition to these configurations 7.5 and 15 inch diameter cylinlerr. wen: tested In the six cell specimen. Figures 11 and 12 are schematics of the.-v wing tank configurations.
3xtenslve work vas conducted on the egg crate design since both the void volume and foam wall thickness were investigated. {Table 2) The nunb.r of void volumes wore varied frorr b to 2h per cell d'-pending on the totrj void desired. A schematic oi a typical Installation is shown in Figure 1J. while photographs of the six and three cell wing tank installation ar presented as Figures \,k and 15.
Total voiding for these specimens was considerably higher than th.-.t, of the fuselage tanks in that the relief to combustion volume for subdivide 1 t-vv Is naturally much higher. The total void percemages tested for these con- figurations ranged from **0 to 90y: for the six cell. 1 specimen and 30 to 801 for the three cell wing tank specimen.
3.0 TE3T PROCEDURE
The test procedure for all three tank specimens and all foam confLTuratic was basically the same. After the desired foam configuration was fabricate) and installed in the test specimen the foam was wetted with jr-5 fuel. All excess JP-5 fuel was drained off before ignition. The specimen was then sealed and evacuated to 5 psia with a water educator in preparation for the
15
''
FIGURE R IB INCH DIAMETER CYLINDER INSTALLATION IN 8IMUI ATED FUSFI AOF TANK
it- < t n«n< t■
i*.
^^
FIGURE 9 7.5 INCH DIAMETER CYLINDER INSTALLATION IN SIMULATED FUSELAGE TANK
GP71 0937 38
17
—
* 1 7 5"DIü.CVI 1*1 ill ft 1 ■ Spherical Ends ■ Vl \Ä\ ■
I I45" Una I ^IMmi
FIGURE 10 TYPICAL 7.5 INCH DIAMETER
CYLINDER WITH HEMISPHERICAL ENDS
GP71 0937 49
18
xf ^
v.>
^
^y
Configuration A Lined Walls
Foam
Foam
/ ^^l^t^y* /^
1 H LH L^ K-J W-J U'_-i
r—» r—» r—» n r—i r—> r—"i
H!3 MM IM u r—1 »•—T r —T
kkj bbj HbJ r-i r—» r —T M* ** ^p
V v-j K3 bkJ / y' Configuration B
Egg Crate
FIGURE 11 THREE CELL WING TANK FOAM CONFIGURATIONS
C.P71 0937 53
19
s- y ^—
oy oy
S 1 ^-
Oy
EEiZ]
ZIC21Z] [2100
[ZDIZ1IZ] 1Z3IZ1IZ] IZ]IZ](Z3
EKZId
[Z1IZ1LZ] Z1IZJIZ]
Configuration A Lined Walls
Configuration B Egg Crate
Configurations C & D Cylinders ILoro & Small)
FIGURE 12 SIX CELL WING TANK FOAM CONFIGURATIONS
20
GP7I 0937 50
■
g
IJI
113
P. o
« e « r- O C«"* T -H J3 q
■Ir
fl-S 5g
I I:
IA >A
IA H W> tf>
«A «A IA • • • I I
I
<A H
(V t
O
H ■0
A H iA
lA
•o M IA
M IA CM
X5
H o
M
O
H
H
Ä 8
o
M o
K o i
■« c^ 2( 21 "*
1 * % 9 I J^ I I i a »
21
J
Outside Walls
Wing Tank Spec
FIGURE 13 SCHEMATIC OF EGG CRATE ARRANGEMENT GP71 OS3 7 51
22
mm
•^ t ♦
25 PPI Reticulated Foam
w Ik FIGURE 14 FOAM EGG CRATE INSTALLATION
GP71 0937 15
23
'
FIGURE 15 THREE CELL WING TANK EGG CRATE CONFIGURATION
QP71 0937 55
2U
introduction Of the premixed pressurlzc-l propuno/'Ur mixture. The propane/ air mixture was node up In a ceparato mixing tank by Introiucinp n pre- determined partial preojure of propane into th«; t.ink and pressurising the tank with shop air to the ualculutcd tot;il pre^üuro ro-iulrod. The aixlng tank and the specimen tank were manifolded together and bro i^ht to pressure equUlbrlliH >i^ the intorcomo^tln,; plumb In : shown ichonatleally in Figure 1 . After e^uillbrlun wae established the mixing tank w:is Isolate 1 arvl the test specimen was bled down through the bomb sampler to the desired inltltil pressure 0, ? or 5 pain. The bomb sampl r was then Isolated from the specimen by valvin-. The sample mixture was ignited and the peak pressure recorded prior to each ter.t. This proceduri vcrlfle.! the explosive mixture oonditions and established th.- aiiabatic expansion factor used in th'- model and data analysis.
The ißnitlon of the test specimen's UtplMive mixture Immediately followed that of the bomb BOB^I«. After the completion of this tott sequence the complete system war. purged ar.i mai. ready for UlQ next test. Foam void configurations were varied r.s were the initial pressures until the resultant specimen tank over-pressure of 10 pslt: "as exceeded ani Buf^'cltnt iata was obtained to lefine the configuration performar..;.-. Then a na", con- figuration of foam was introduced into the pronram.
>».0 B3SULTS AND DI8CUSSI0H CF RESULTS
All of the data for the three tank Bpeciaona and their respective foam configurations are presented in four ways. The raw data is presented as tables, whereas the reduced data is presented ßTSphicelly (l) as physl.-aiiy installed and tested (?) according to the theoretical model ur.d (j) in summary bar form. The grafha of the "as tested" are In the form of total installed void percent "ersus peak combustion to initial absolute pressure ratio. The total installed void percent is also the percent reduction in wcipht and volume penalties for the foam system and th'is of direct value to the designer. Absolute peak combustion to initial pressure ratios arc used Instead of Ar's in that theoretical coasldentions predict that this approach will normalize the data for iifforcr.t initial pressures thereby consolidating the graphical data. Althout^1. this normallzlnc was not realized it is Still felt that the pressure ratio presentation is useful and convenient.
The second form of the data correlatior. presents the actual relief to combustion volume ratio to the varlius observed pe.-ik combustion to initial absolute pressure ratios. Flottint: the data La the relief/combustion volume theoretical form provided a meai.s of noi.itorinc the lata for nross icviations and interpretation of the results. In many cases particularly when: Internal void eonfIgurations were bein^ tested several pressure peaks were observei on the oscillograph traces. The first observed over-pressure correlated with the initial ignited volume whereas the intermediate peal; over-pressure was shown to correlate well with the total voil volume, indicatlr.f; burn through or ignition of the secondary inner voids. Glncc straight line, simulated pro- jectile Ignition was used in the testing the relief volume to combustion volume varied from one configuration to another even the-.gh the total void percent was the same.
25
Vent
Ignitor
150 PSIG Transducer
40 PSIG Relief Valve
Ignition Bomb
Test Tank Vsnt
^
FIGURE 16 FUSELAGE TANK TEST SETUP QP71 0937 50
26
^^^^^-^ — . ■
the bar graphs are presented to provide a simple visual comparison of the foam voiding configuration capabilities as compared to each other. From these graphs the superior configuration for any of the program test parameters can be easily determined as indicated by the solid bars.
H.I Fuselage Tank Results
Tables 3 through 8, present the test conditions, configuration and combustion over-pressure data for the simulated 100 gallon fuselage tank. Only the maximum or peak combustion over-pressures are given in these tables. The average peak over-pressures were converted to absolute pressure ratios and plotted against total void volume. These plots are presented in Figures 37 through 32. Tbe theoretical relief/combustion volume curves are presented as Figures 2U through 31.
•..1.1 Lined Wall Configuration
The lined wall foam configuration data are plotted in Figure 17 and can be seen to correlate quite well with the theoretical curve up to ■♦0"'- voiding even though the Initial pressures were varied from 0 to 5 pslg. These same data also correlate quite well with the theoretical model parameters as shown la Figure ^1«. The divergence of data from that pre- dicted theoret'cally is caused by penetration of the flame front into the foam. Flame peneti;.tion Into arrestor materials were observed in the movies and Is known to be a function of both pressure and foam pore diameter. Tt can be clearly seen that both parameters prevail by inspecting the three curves; theoretical, 25 and 15 pores per inch foam in Figure 18. The 25 ppi foam configuration was adopted as the base line to compare with other configurations and therefore appears in all the fuselage tank configuration graphs.
1.1.2 Voided Lined Walls
This configuration was essentially the same as the previous discussed lined wall configuration with the exception that closed "hidden" voids con- stituting 10, 15 and 251) of the total tank volume were Incorporated in the foam liner. In all cases this additional voiding was divided into 16 individual cells of which two were included in the initial Ignition volume. Consequently one-eighth of the additional voiding was negated thereby displacing the curve above the base line as seen in Figures 18 through ÜO. In the case of the 10t voided wall configuration the theoretical shift of 7/8 of the additional voiding was obtained out to the '»5* baseline point. However in the case of the 15 and 2y- voided lined wall configuration lull theoretical improvement was not obtained as several of the wall voids ignited during the test. The data are good however, as can be seen in Figures 26 and 27. In these plots the peak over-pr ^sures are plotted against the theoretical curve parameters and are seen to correlate. In addition the initial or Intermediate pressures are plotted against i.he initial combustion relief volumes where applicable and maximum pressure against the assumed total void combustion volume to relief volume. The latter plot correlates well with the theoretical curve and therefore it can be
27
concluded that all Inner voids ore burning, but on a layed baslo. "ftie Bize of the center void volume. In addition to wall thicknocr,, affoct.. the performance of this approach. Thiü is to bu expected in that -as pre-'iounly diHcusscl the larger the center void volum-? ^he greater the reüultLnt heat flux, prucsnra and depth of penetration. Graphically this can be observed in Figures 19 and 20 for the 15/) and P5* voided wall configuratiun. The center void volume in the 15* voided lined wall configuration was croater than that of the 25y) voided wall configuraticr. for the r.ame total void /., therefore flame penetration ani burning of the void.; within the foam occurs sooner for the l^i configuration. This did not occur with the 10% voided foam liner since the foam th'cknor'ö to combustion volume was greater. These results would lead to the oonrluclon that an optimum void volume to foam thickness exists and that if all voids w^rc uniform the optimum total voiding could be reached. This optimum configuration would be an egg crato configu. .-luin with more and more voids as the tank volume got larj-cr. This relationship is further supported by results from other configurations and larger cell tests to be diseutMd«
't.l.j Large IMametcr Hollow Jylinderr:
Tills configuration was ^uite succeosi'ul. The small number of voldr and the resulting greater foaa thickness to 'ombustion volume can pro^bly b'- credited with the succesü. There wen: four internal voids of vhicn two In addition to the void external to the cylinders were Initially ignited. The expected improvement In voiding performance was realized at zero psif initial pressure as can be seen from Figure 21, At higher initial pressures however the depth of flame penetration caused all of the internal voids to burn as evidenced in Figures 21 ani 23. './h^n all uu vsld« burned the eanflgurfttlon reverted to the baseline performance.
U. 1.'• .'j.all Uareter Hollow Cylinders
Two configurations of hollow 7.5 inch diameter cylinder;; were tajtei. Flat end twenty four inch long and hemiyi h^rical end twelve inch long cylinders of four wall thicknesses wer.3 run. Figures 23 and 23 show the performance of these t'JO configurations. It Is quite apparent that the shorter hemispherical end cylinders were superior. Of interest here is the divergence from predicted performance. The different I :iti:il pressures ;-ave different curves which plot below the baseline. Gome explanation of the pressure phenomena can be gained by looking at Figures 29 and iO. In these it is -iuite obvious that nearly all the flat end cylinders »ore penetrated as were all but three test conditions for the smaller hemisph_rlcal end cylinders. The flat end cylinders had a smaller Initial combustion volume and yet performed more poorly. The principal difference war the flat end cylinders had small extorral combustion volume channels which tend to slow flame speed. This slow flame freely penetrates the foam. One explanation for the data below the baseline is the fact that in the positive initial pressure tests where this is evident, higher resulting combustion prersure occurs thu: the flame penetrates into the foam from both sides and subsequently burns more vapor. This effect was discussed when the baseline was establish^. (Paragraph '-.l.l)
28
^.1,3 Fuselape TanK .Sumnar'
Figure 38 summnrlzed th- results of all thu fuselage conflcuratlons tested. UM three bar grapns for 0, 2 and 5 psig show that the large hollow cylinders are best for 0 psig tankage while the 10VJ voided lined wall configuration is better for the 2 and 3 psig initial tank pressure system. These rusults might well be dj fferent for larger or smaller tanks in that the eombustion volume has a direct proportional effect in system resultant pressure rise. Further, the multicell two or three tier egg crate conficurntion not investigated raltjht well out perform either of the two best candidates thus fur evolved as the (-eonctry of this configuration allows for greater void percentages while keeping the combustion volumes small.
4.2 Six Cell V/lng Tank I'.e.sultü
The tank speclrac-n usefl in the following series of tejt:j contained cix cell;- 3r.,:h of ij0 gallon capacity. There veto four fonm configurations tested. The recults and test parometerr, are presented in Tabl^r '-J through 12. Th2 test configurations were lined '..'alls, eg^ crates, and r;r.all and li'.rge hollow cylind:rs. Figures 33 through 3(' percent the data in graphical form.
I».2.1 r.ix Cell Lin?d Wall
UMI lined wall system was fjiven only a cursory investigation since previou:. MC AIR tests reverded that this lesign used in an 80"' void cor.f ipuration in a relatively small scale test fixture produced excellent results. Testin/' for this configuration was limited to an 80^ void arrangement at 0, 2 and 5 psig inltin.]. pressare to determine the applicability to larger size wing tanks. The limit of cell size to voiding for this confirursttion ll still now known and should b3 further investigated.
4.2.2 .:ix Cell Xgg Crate onf'.guration
Sxtensiv»* work wus con iucte 1 on thi' tgg, T.-.te design since both the void volume and foam wall thickness had to be optimized. Table 10 contains the dnta for this configuration while Pigurce 33 and 3^ present the result;; in graphical form. It cr.n be seen that this configuration works best vith one inch foam separation walls. Thi;; thickness is charaot:ristic of the 25 ppi foam. Smaller pores would require Leas thic'r.nesr. ufaila larger pores would require a greater thickness. The percent void allowable for the 10 psig over-pressure crit-jrir; vac found to IM ""', 87, and 7C for 0, 2 and 5 pa.lg initial system pressures.
•»,2.3 3ix Cell Large Hollow "ylinaers
Figures 3'* nn* 37 present the results for this configuration whil. Table 11 contains th_- data. 7he performance here matched that of th.; fuselage tank tests, '..'hile only two points were established for the wing tank they correlated well with the fuselage tank result.-, u shown in F^gur' 3U,
29
-
'•.2.U Six Cell rsmall HoLlov Cyllr.-lers
Figures 35 ar.d 38 present the results for this configuration while the data are contained In Table 12. From Figure 35 It can be seen that three curves, one for each Initial pressure test condition, were obtained. The smaller 7.5 Inch diameter cyll.-.ders in this size tankage performed better than the 15 Inch diameter cylinders. Equivalent results however were obtained In the fust.'-.age tank at greater void percentages.
^.2.5 Six Cell ConflgurAtlon nummary
In general for the six c«ll 00 gallon wing tank It can be concluded that the egg en te followed by th« lined wall configuration were the better designs (See Figure 39). The hoMow bodies are more sensitive to Initial pressure and when flame penetration Into the Inner voids occurs greater »han the theoretical minimum pressure results. This latter fact Is probably due to a pumping action where the outside ignition pushes some combustibles Into the foam followed by Internal ignition pushing it back out to be further reacted. This action does not occur with the lined will or egg crate design since there Is no external/Internal voiding. The lln'?d wall and egg crate configurations act In one direction resulting In data that falls very near to the theoretical predictions.
U.3 Three Cell Wing Tank Pesults
Tests for this segment of the program were carried out In a three cell 300 gallon specimen. Two foam configurations were tested the results of which are presented In Tables 1-, and lu. The configurations tested were lined wall and egg crate. The data are graphically presented in Figures Uo through -•L'.
-.3.1 Three Cell Lined Mall
Only ambient prorsure system tests were conducted with the lined wall configuration as the performance r^sulttd in only a Wf* void system. The degradation of performance of this configuration in the thr^e celled specimen Is due to the larger combustion voloraes with respect to foajn and relief volumes. It Is Interesting to note that this void percent Is the predicted value for the equivalent size fuselage tank.
^.3.2 Three Cell Sgg Crate ConfIfuratlon
The egg crate configuration performed to a ^2^ void at the 0, 2 and 5 pslg Initial system pressure. Once again the larger voids In the three celled specimen for the egg crate configuration dictated the limit of performance. In only one case did the flame fall to penetrate the walls of the egg crate voids. From Figure U3 it Is obvious that for the remaining tests all avail- able void volume was Ignited as the calculated data for combustion of all voids plot within testing tolerances of the theoretical curve.
iO
U.3.3 Three Cell Configuration Sumroary
Tests In the three cell rpeclmen reveal that lined wall and egR crate foam configuration perform equally well. (3ee Figure W*). The larger Initial Ignition volumes substantially reduce the void possi- bilities because of the reduction of Vr/Vc when considering vc equal to ignition volume only. Reducing the size of the voids does not in- crease the total possible system void if a one inch minimun foam wall thickness is maintained.
SECTION III
CONCLUSIONS />• RKCOM^ENDATIONS
5. CONCLUSIONS
The Phase I tests of this proßram have shown that aircraft fuel tani fire and explosion suppression nay be readily accomplished by the use of voldea reticulated polyurethane foam systems. Fuel taik size and shape as well ns foam cor'lguratlon have proven to dictate to a degree the nnount cf voiding possible when using the pro^'ara success criterion of 10 pal combustion over-pressures.
o The largest percentaga void (59.5'') for 100 t'allon fuselage tanks at 0 psig initial pressure, wan obtained using fifteen inch diameter hollow foTW cylinders.
o The greatest amount of voiding in fuselage tanks was obtained with the 10"t voided lined wall configuration with 52 and ''T^ percent total void for the ? and 5 pslß initial pressure.
o The SO"*, void, lined wall foara conficuration was successfully tostci at 0, Z and 5 psij initial pressures in the jOO gallon, six cell simulated wing tank.
o The six coll wing tank egg crate pattern performed within 10 psi maximum combustion over-pressure to ftt void at 0 psig Initial system pressures.
o The lined wall foam configuration performed satisfactorily up to ^5' for the three ce.1l 300 gallon simulated wing tank.
o The egg crate confiRuratlon for the three cell simulated wing tank specimen provided UO'j, 30^ and 501 maximum vcid with 0, S and 5 psig initial system pressure respectively.
o i.'o special installation techniques were required for any of the foajn designs tested. The hollow body configurati'-.iü were assembled with adhesive prior to installation while the lined wall and egg crate patterns were held in place within the respective tank by cutting the pieces oversize and compressing them in place.
6.0 RgCOWggDATiag
It la recommended that further testing of the lined wall and egg crate configurations be conducted using actual aircraft tanks and a gunfire ignition source. Further it is recommended that tests be conducted to determine the maximum and minimum opening between cells to determine their effect on the void concept. Hollow spheres should be tested in future work as this configuration appears to be promising and ^ulte adaptable to system retrofit.
Preceding page blank
33
mm
r---r
I T
est
, C
cnft
g.
Desc
rip
tio
n
I i
r -1--t-
-A
--" -:
(j;~
-~-i
ll~-
--
1·, 2
Cen
ter
Cm
ty
3 4
A
Lin
ed '
li ills
Cen
ter
Cav
ity
Lin
ed
Wa
lls
Cen
ter
Cav
ity
IJ.n
ed '
llia
lls
Ce
nte
r C
av
ity
FllS
El.A
GE
FU
EL
T:t.N
K SY
ST.i:
M T
ES
T D
X1'A
t:r
nEO
WAL
L CO
NFI
.GU
RJ.T
ION
,-
---.-
---.-
1 rO~:
l Foii~;
-:u ! ~'
orun ~~
To
·::a
l A
mb.
'
Init
ial
1 ~~r
e ~1s
it;r
I Vc'
c!. ! '
.l'~~~
I P~e
ssu~e:
:1
yst
c
'1:, :2l.t·
i5z~ ;
__
__
__
'_) -l-
:_··_1
: __ ?7
.. -.-1·----
-i--~-~·_·
_;-+-: -~ {_
,-~_'~-~~_
)_''-+_Ri_
~_;J_{_;_·
._~g __ )J A
~~~-
2_~~
-p} J
I
1. 3
5 0
-14
.13
14.1
3 95
-2.5
I "
.25
I <
.w ! !, '
!J) I
t 25
1.
35
o 35
.0
14.3
1 14
.31
95-.3
,.
6.c
o
' 6.
;:J I t>
.lO
1 11
I ; 7.
70
';.(
)) f
39
.8
14-3
7 0
14.3
7
l3or
:tb
S ar
:rp 1·
~ "!
"".-.
.....-
·::;.
25
95.0
25
1.35
1.
4.37
14
-37
96.1
16
.10
13.7
0 l!
,.:J
() I
5 L
iDed
W'a
.lls
25
1.3
5
0 35
.0
14-3
3 16
.33
7.5
7
7.62
I
7.7
4
Cen
ter
Cav
ity
w
~
I 6
Lin
ed W
alls
25
1
.35
0
39
.8
14.1
9 16
.19
102.
5
t"! rn
({)
-1
~
It~ > r= !1 r,; n 0 ,
7 A
a A
9 10
A
11
Cen
ter
Cav
it;y
Lin
ed
Wal
ls
25
Cen
ter
Cav
it;y
Lin
ed
Wa
lls
Cen
ter
Cav
it;y
Lin
ed
Wa
lls
Ce
nte
r C
avit
y
Lin
ed
Wa
lls
Cen
ter
Cav
it7
25
25
25
25
1.35
1.35
1.3
5
1-35
1.3
5
0 JO
.O
14-1
3
0 3
5.0
14
-48
0 40
.0
14.4
8
0 JO
.O
14-5
2
0 3
5.0
14
.52
19.1
3 12
5.5
19.4
8 12
J.8
19.4
8 I
136.
9
16.5
2 10
8.2
. 16.
52
U6
.8
I 7
.04
7.
00
7.09
8.7
8
1).
90
6.0
8
s.a6
a.ea
lJ. 7
5 114
.05
6.22
6.
09
< ~--~----~------------L----L--------~--~----~-------~------~----------~----._ ___
_ ._ __ ~
H
g
§
i
i £ R a 11 a ^
äg H *» HO <
• M IA
1 1 si
• 8 9
ill (V ri X ^ 3 IA n 4 3
lated
Tank
ssur
e
< rA -i •
UN vfi ■ • • 3
d iH « B 4) (.
>1> 4 iA 4 • tf s IA
CM IA rH •
< «^ -* IA sO «X5 9 O 9
O) -^N rH DO A C }> -H
s.S3a >o ir\ (^ en CO o rH (M tO .p «s^ • • • • • •
lljl ^ g i g rH
B o rH l'*i
■ 3||3 S 3 3 3 9 « O
(A iiji i 4 •
4 •
4 4 rl rH vO rH
• 3 «8 o o 3 9 o vO rA Q ja n -H •3 r^ r> rS ||J 4 * 4 4 i 4 4 4
g-o irv o • • en o O O • ;j
o otx i IT» IA R 3 R IA f- >^
o B««t o o o o Q q o O U. >v-5 H r-l r-l H H r-t rH rH
^t _■♦ -* <J -.-f -* -? ^.-* o n^v ■ u, a A .H iH rH i-l H iH rH rH
S rH
D <U O -H B !. N ►>. ITi l/N A IA IA IA IA LA P O -rj PS N CM (V cv iM CV CM fV C-, «- "5 V^
h >> >, s >. ^ S >. g
•H |
•p *> v fl ü V m T3 «J TJ +1 V •! •3 *>
»-4 +3 o J ♦JO
D -rl
►4 *> o
CJ iH
J +1 O
P ri I«! rJ *> O
C -rH
■ m •H a •H B •H •H TH va ® '' •) i;! i5i « ' ' u 'S (1) l5l| 1-i
n 8 o ■* V h\ ■0 i-t ♦> ii] Isj •0 rH *J
131 T) H +> •rl H C o a) o o ci v
•G rH .p
131 £y.c > r- ü > ?ü ^ r^ o > "^ CJ > r- o > 'T ü > :• o
a a ■ ■ (C n a n
■ ■ i-t cv I^V -* IA vO r- (C H
'
35
■M»
&g rN « 3 fl It c? 9 n a •
IS8- <
<> i 3 >o a i -* t^ s £ M N^ cv 4 9 R r^ K ft }S }( s
Lat
ed
Tan
k is
ure
< • •
rf vO a ri -i t^ 3
l-t —. n 5^ • * 5 8 cT S <
&• H rt vO o B ~j f-^ q H H s (V M
u -->. H M a o 5 oj ■ «IMA to o m in m «N o
i 1
fljl i 1 H O
lA i a rH a B M ä -H a a
4) jiiQ Q (N -* fV (V ^ vO «0 (v (*\ ir\ in >o >o p^ n in
A 4 vO 4 ^ 4 3
. 91« O fV -* f^ OJ 3 vO n fv til r^ »A u> vO «O R r^ m • • • 4 4 ^ 4 4 4 4 ^ 4
tfaa m o ^f q q o o m o
R R -* -* 8 5 in si
82^ o o^a O o in in in o m in in b. > w r^ M rH H r-l rH t\ N fV
■ -H «M -* -* <f <f -* -4 -* -t -* O B*^ i ■
&. a jo H H l-i r-( rH rH rH rH H S M
_ - — ^^
■ • ■ «ri ■ b H a. IA ITl in m in m in in in 0 O -H M. fV tv rj f>J CJ fV <V W tv
s lh >, k S s», s kk S § 1 i I«! M *J CJ ►3 +1 o J +> o
1 i i-I ♦> U
1 ? 1 g rJ +5 o
73 -P 4) -rl
►4 «J O 1. i t3 i-I <J o •H 3 •rt •H a •H •iH •H •H •rH 1 l5i ■ 1 I L.
4) <o " u ^g^fc |* 0) 1S I |2 I s ■ « ■
O «j D O « V
T3 r- t-t t- O «
■P is i o « o
T3 H -P T3 rH
9>
73 rH +J T3 rH ■p •0 rH p 1 1 iil 13 ^sj 1 r?g f iai > - Ü > R o > ' Ü > ^ o > 9 O p fl Ü > r- o > r- u
tf •H
I en a PQ m P3 m CO m o
*> I o rH 9 r> 4 m vO t^ 0) o H r-t rH rH rH rH H
i6
ii a ■ Ü
B
U C o 5 ^ 1 R 3 i M 0 « •H'-N H ♦> M
-i -* tn d -* 4
CM ft ^ 1 R H
ulated
1 Tank
essure
< -i -* r> d ^ 4
m* H ^ H R R R < -* -* (»"i o -^
u ^, rH BO vfl P«
O vO o IA CM es • • £ c D aj § i p £ 8 •- « .H a ■
- * 46 |Q i vO -^ w is R ^ .r4 *> 0) -rl • • • 4-> n n n 4 4 ^ 4 4 rH
0)
.1? ,3 I -H i * ^ R R H t- 0] CO ■ • t • •
4 4 ^ 4 4 ri
^S ♦J ^'-N m i/\ ■A I« in IA p o ■• -, <t 3 -* 3 -^ -f t-. >N_^
h
CD
■ • • • 8 • 8 • 2
-* -t p-» CN IA -■t
*" ^_^
>c^
Foan
nsit
b/ft
• • • • • ^d rH rH rH rH r-t rH
^_ let'ä IA i/\ }Q ir> ir> L-v O O »H »-« N fV t\ fV N C S "i^
s •rl
: -p 3= 1 O T3
ft V cn -P n -P
E 9i ^ ? „ ^ ■ f0 4) vtl ^
o n
H in c ^ -O NH o a e
Q rH b U ro Ü <M P OV-.H ? 1)
tr •H i O Ü o O o o
*J I 1 rH N r> ^ u-\ vO
37
IJ-a III III m n d M «
en a.
a.
a.
«V
«
a
a •i
a « IT» rH
8 8
S 4
i
is
3
a 00
s
0) -"-s
ilia w 0) «) f< n ■c3 il 8
to I
8 a 8 Q
O
8
■4J jo en n a a
4 -* i/\ 3
4 3 3
0) -H n m a 4
* a 4
3 4 4
O OV!. s a a
i 1 1
a 5£
Ad
£ -»
%
a
o o -w U
§ •H
•rl E o n 5
IA UN IA IA IA
O -J O V lAH rH R lA H 1
AI I AI I 41 I if I 0 4) oi) oc or
n5 3 ?dll ArH , H <
rfiP4J I
4) IA H 4)
E
IA IA r-l •« CC
&$ P4H MI
•H «lAIA H »lAIA r-l*C3e0 HOttJ-O CJ T* rH / ■
55
Ü-Ör-trH OTJOJCJ O^OMrV
M
0) m 98
■MM
i. I I
ll
9
C
äN
0.
<]
fall I» s
illl
ill l9a Sä»
I US
a <M
illl
1 I I
s
I
4
n
«
N
I I
8
8
R
^1 i fid« 1
.-I * Ü0 » r^ • «^«^•d
3 .i| Ö.|
-y
39
-■ ■
1 Q \f\ ^ CA >• s 1 1 1 1 1 i » K O a 4 HI
N 9 p 5 £ IA -4 '3 2 % i 9 s
ill Mt* « r- n
< • • •
r- CO H rH rH A 3
ip ft. • 1 • 5 • 8 • M
IA lA
00
R ^
2 9
• ^ s
^ ^ 4
<u ^> M H a c D -H
s.2!3 a o o D O O o /> O O 0 • • • •
fllll i i K IA fA
3 3 s LA 1 PA ä I**»
9 sO « «3 3 « % K 3 aj 3 ^ J ^ 3 • 4 3 3 «
4 4 4 a > 0 H 3
< g, F- *—s H ^^
1
B d a
• 3 (d
111 • 4
• 4 3
3 • 4
5 4
9 4 4 4 4
3 4
lA
c
5 a« O O^ R s s LA A LA IA R R S
n
C-. > v^
it
i F-"
1
1
H 3 Si
CO P ^ R «0 a 30
f. 5 • -* • • • •
-4 -* -* -t -t •
:o a 2 1 fi
^~s
3 K « « « % rA
sO ^0 rA
0 PA (A
0 fA
0
• • • • rH H rH -, -<! rH H
13 ^^
C 4) 4) -rt « L. N U 0 o -w u R f9 «8 r* ^J rA «3 C £ 01 ^^
^-N ^-N ^^ ^*\ • in (0 (0 tf) rH rH H H r^ ■\J p^
c ^| 1 ir\ o
5« I iv 1 C 4>
IA O 5 O | O 0 0' _o
o
t e-5 . . ^31 1 -H H *>
t .SI . rS ^ 4J
n •P n « « n H-* tn ■
2^^ «.3 ^ 4) 4) F-
4) t-
4) t-
V 3 ■ ■ •§^55 •§^55 \n&i ^5.5
■ H • IA LA
O ^ H H
a | 1 ^ * ä a I •H • ./N IA ■H • IA t/\ 5 . IA VA ■ 4) 41 4) H 4) 4} r-l • (V tV
^^ A A r-l « ^5 O 1 1 I 1 [1 1 I
to
o o o Q Q a a n a a a a
tl I il r-1 rv (^1 -^t IA NO •*«. CO 0 s d H
1 S
Fuel
Tank
Igni
tion
Pressure (peig)
<i I i i 1 1
f R R » R • « « • «0 n (*\ rj
<3 9 • 5
8 8 t
8 t it a • • • • a^ -^ <n rj
H P.
<
9 • 8 • R 9
9 i
i • 3 R R » « § . •
eo f» «^ H
V
ll lit 8 M J.
•H I (U
> n
"a
o
B o i o
3 o i i
O tV N 1A • # • « «n t- » <v •3 ^ * s
^8? ■H -P «
8 4 <n <•>
i i • 4
^ ^ i i 4 4 4a
• 3'«? III 5 <n
4 K 4 4
i
4 5 ^ ^ ^ 4 4 4 4
4« O OVl 3 ^ K R R ^ ^ Ä Ä
US 1 00 to CO
• <o
-* g; ^ g; ^ • • • •
3^5
Is NO
4 • PS H
• H
g 0) 0) -H B U H M< 0 0 -rl U R n
(V R ^ «5 ?? S?
1 1 1
en
i ■ 8
1 liuta ■H • ITS IA
i I s 1
a d z
1 s
1 inii
S S S S
i ä ä ä i I f i p p p p 9 9 8 a
* * * *
Q Q Q Q o Q Q Q Q
1 a a 4 H
41- & a 5 R
s
5 i
M
I
»A
i
111 IM III 3 £'c
s a JO ft 9 8
£ 1 1 1 1 • • • 4
• •
o* • 8 9 • « 8 ft 8 • • 9 • R ft • • a. <1
c^ i ^ c- • 3 a n ft
tf & 3 g 1 3 5 a SJ JQ • « • • < 9 t- in
H CM o • SI (A
H Ü ft 0) ^
00 O O IA O >A O CV f^ • « • • • • •
Boc
.b
Ign
i P
res
Ris
e 3 3 a* H rH
£ 0 fc a a
lal
too
sure
la
)
NO ft ft 8 a S 3 S 9 |*|J •
4 ^ R •
4 ^3 Ä •
4 ft Ä
Amb.
P
ress
ure
(p
cia)
ft ft ft 9 1 ft 3 3 3 i • •
4 4 •
4 • •
4 4 i 4 4
iis R ft ft 3 3 9 8 J 8 H >W
1 US IB «0 o ü £ £ <0 ft R •
-* • • -*
• -*
• •
id ^ sO vO vO •« >o VÜ r^v f^l c^ f^ (»\ r-\ n « iH i-I H H H rH
• rH H
y-N
IIJI M m iri ir> IA »A IA IA IA N rvl CM OM CV tM (M CV CM
b. 0- V)-*
• H H . ~t -* , t- IN
1 1
l/N O £ i iA 5 ^N i i IA B
• •
1 1 H Ei
9 S
^3i 11 s a
rH rH H^
1 1 s a
<s 1 1 11 1 . iA tA 3 • N tv 11
•
1 M w w H w w M M w
I H fM r> ■* ki IA vO t^ «V 0^
3 s
fl i 0. in H
I B
I
£0 I u
A,
0«
ft.
o»
»If?
I ills |^8 a
tij
i
I M
il§
»i «., N O "H
b ft. CO
i V m
&
I
as
Ö?
3 4
3 4
K
^
N
ff
CD
0>
3>
i
(V
3 4
3 4
9 4
3
IA
o
9
«0 B
&
?r3
Si. ►.
+> ^ü fc
It |l K
+>
in in CQ to m
C\ ^3
I
< > \ \ g» ■
7 ■1
\ a. Q- a CM 1 in
\ \ \ \ \ \ \ 1 i
\ i \ \ [\ \ \ \ \
\ \ \ \ \ \ \ \ \ \ . \ \ \ V \ \ \ \\\
\ A\\ W \ V x\ \ Y N\\\ VVVA A-W
%\\ « v\\ 1 V \\ Q. VW
"O 4rt ~ x\\.
Vo
i 0,
2.
5 pp
^ = c (v; v^\s. -11 s
i f s 1 Iff J
s L >>\ oo
. L... -
z o < ■ g K z 8
I 9 ; h < w
■ < O K
t
D a
S S n leioj |)IOA K
GP71 09.(7 28
kit
z o <
o u. z o o Q 5 >
i < o
o
GP71 09?7 30
^5
,
\ v 1
1 5 \
Wal
l V
oid
. 0
psig
em
Pre
ssur
e (2
5 p
(Vai
l V
oid
Co
nfi
gu
(2
5 p
pi)
\
D
15%
Fo
im
Init
ial
Sys
l
O 0
% F
oam
\ (B
ase
Lin
e 1 \\
\
z o
0 . Q
a- _i
| < O
i tr Z) o
I R leioj. piQA ..„
GP71 0937 31
U6
J
|B10i piOA %
('.P;i 09,17 ,1?
»♦7
"
|R»01 piOA %
GP71 0937 .15
40
.-■
CM Z o
< cc D
? O u 1 LU
< ■ -1 fN u.
oc a LU
a. D 7
J J >■
00 u - cc
|BJOi piOA %
GP71 0937 37
U9
z o
< oc 3
u.
O u Ü z ai
cc LU Q z > o cc LU
< 5 i o z in
w IT
a
inox I'IOA %
CPn 0P37 30
50
Li
u
9
8
7
6
5
4
3
2
1
0
i i i O Combustion Data tor Ignition
of all Available Void
1 \ 1
I /- Theoretical Curve Pc
P1
K
I
VTK i
V„/V-, \
VTK-VC(K-1)
Adiabatic Expansion VR/VC
\
Vy Total System Volume Vc Combustion Volume V- - V_ V„ Rehpf Vohimp J
\
1 i; M
\
Cä^ b
, ̂ _
2 3
Pc/P1 PSI
; a
FIGURE 24 PRESSURE RATIO vs RELIEF TO COMBUSTION VOLUME RATIO-LINED WALL
FUSELAGE TANK
GP71 0'):i' JO
51
VR/VC 5
FIGURE 25 PRESSURE RATIO vs RELIEF TO COMBUSTION VOLUME RATIO 10 PERCENT VOIDED LINED WALL
FUSELAGE TANK
GP 7 1 093 7 4 I
52
VVC 5
O Combustion Data for Ignition of Center Cavity Plus Two Wall Voids
A Intermediate Pressure
□ Data Circulated for Combustion of all Voids
VTK
>, VT K-VC(K-1)
< ■ Adiabatic Expansion J-r - Total System Volume //> ■ Combustion Volume /T - Vc = VR = Relief Volume
Pc/P1 PSI
FIGURE 26 PRESSURE RATIO v$ RELIEF TO COMBUSTION VOLUME RATIO 15 PERCENT VOIDED LINED WALL
FUSELAGE TANK GP71 0937 43
53
10
VRVC 5
r ' ' t ■■■'
0 Comhustion Ddta t Centei Cavity Plus
i —r i ii Ignition of i wo Wall Cavities
A Intermediate Pressure
D Data Calculated foi ComlKistinn of
\0
all Voids
\ — Theotelical Curv > 1
_ 1 V : p c
I VTK
HI
\ K
VT K Vc IK tl
ArlMh.itic Expdnsn
\ < ? VC
V/
Total System Volume Combustion Volume
\
vT -c • R MC"r' ^
Vi ! (
PP, PS.
FIGURE 27 PRESSURE RATIO vs RELIEF TO COMBUSTION VOLUME RATIO 25 PERCENT VOIDED LINED WALL
FUSELAGE TANK
r,P7i o-t i; J i
5^*
.. -.
Vvc 5
pi;pi PS'
FIGURE 28 PRESSURE RATIO vs RELIEF TO COMBUSTION VOLUME RATIO 15 INCH DIAr "TER VOIDED CYLINDER
FUSELAGE TANK
'i
55
V'VC &
4
3
2
O Max Pressure
A Intermediate Pressu re
D Data Calculated lor Combustion of all Voids
1 1
V - Theoretical Curve
\ Pc , D
VTK '
\/ 1/ V/ /I/ 11
«■ Arti^hatir Fxpammn
VT Total System Volume V- - Combustion Volume VT VC VR Relief Voljme 0
' 1 ; 1 9? 7
. ^Ä; A ii k
8 9 10
PcP1 pti
FIGURE 29 PRESSURE RATIO vs RELIEF TO COMBUSTION VOLUME RATIO 7.5 INCH DIAMETER CYLINDER FLAT ENDS
FUSfcLAGE TANK
50
VVC &
0 Maximum Pressure Only
D '^'aii Calculated for Combustion
1 \ of all Voids
\ r- Theoretical Curve
1 y Pc VTK
l \ P, VTK-VC(K-1)
K ■ Adiabatic Expansion
i \ V-j- • Total System Volume Vo ■ Combustion Volume
! \ NT-VC V ^ neue'
t yJb0
1 il i u ^HEr^-£-J
1 1
Pc/Pl ^
FIGURE 30 PRESSURE RATIO ws RELIEF TO COMBUSTION VOLUME RATIO 7.5 INCH CYLINDER HEMI ENDS
FUSELAGE TANK
57
VVC 5
4
3
2
O Com hi Center
A Intern
Q DataC of all \
i ~r i
islion Data tot Ignition of Cavity Plus Two Wall Cavities
ediate Pressure alculatert for Combustion
/oids
i i j
1 /— Tiieuieiii-dt Cu ve 1 1 1
p(: VTK
P
V V
, VT K VCIK 11
A 9 'j Total System Volume P Comhuation Volume
\ k 9
o V T VC VR Relief Volume
1 1 5 •
8
" r ^
PcPl l,s'
FIGURE 31 PRESSURE RAT'O vs RELIEF TO COMBUSTION VOLUME RATIO 10 PERCENT VOIDED LINED WALL
FUSELAGE TANK 15PPI FOAM
l,PM IVI MM
58
XHM
1 >
E
100
90
80
70
60
50
OPSIG
I 40 ■ ■ 5 30
20
10
2PSIÜ
>
E 3 E
>
Walls Voideil Walls
FIGURE 32 FUSELAGE TANK RESULTS
5V
7 5 In Did Cyhmtei HemiEnd
ÜP7) 093)4H
a
I-8
*
ll
•as 05
■ > n -H
« o l|J
S5S
MS o 2
O O -r-l U
§
f &
60
_fc!_
to I
CM cn CO I
I CM
o o tM -*
CO -* ~* ^ 5 » CO
Ann
444
8 8 8
co co es -i -* -*
NO sO >o ('S (*% fA
A IA tfl IN CM CV
< < •<
N r> 60
smmt
s 9 s s 9 8 8 R Ä £ a Si a a t «^ (»> & 4 H N CM CM 3
s 9 R » 9 Ä R Ä Ä a 5 a a 8 8 a
Ill 11
f <^ tn & Ä H <M cv C^ i 9 9 R R 9 8 Ä Ä Ä a 3 a a s 8 8
tf m «^ a a H N CM CM OJ c^ j 9 9 1 3 9 3 ^ Ä Ä UN
«0 3 a vO S 8 a f »-> «^ 4 a H <v CM CM to c^ i
9 9 S fl 9 3 a 1) Ä & 5 UN CM J $ q ft
f r\ CN 4 H <M CM o, S t^ !l ^H
9 9 a a 9 8 8 IJ ^ a 3 k •lo ^ 8 8 r\ «^ X a H C* CM «v ^ C>- 8
ill - 1 Oijj & n •rT i
H • H H
<M
1 CM
I; ■
a o
1 o •
9
«0 • «0
fl 1 B «0
a CN O
B B r»
^
o r-j r 1
Initial
System
Pressure
(pei
a)
UN
4 i i • 4
a • 4
a a
a a
a • 4
a a a
5 a
>o •
4 i a • 4
a si
Amb.
Pressure
(peU)
• 4
• 4
UN en • 4
a • 4
a • 4
a • 4
a • 4
a 4
a • 4
a 4
5 • 4
>o
4 4 4 4 5 4
Total
Void
S 8 8 a a a a a a a a 8 S S a a
>H
1? «0 «0 «0 «0 «3 «0 «0 «0 «0 «0 «5 «0 «o oo M M
-* -* ■^ -* -* -* -* -» -» -» -* -* -* -* -* -*
.- '
Iff « « « A a a a a a a a a H H iH f-i ri iH H H H iH H H r) rH H H
^-S
io£^ CM i (V UN a UN UN
CM UN UN CV CM
UN N
UN N
:ription
0
1 | «
o
• 1 o
»
1 O
i «
6
•
6
•
1 i • 4)
o Ü f
4) i 3 1 1 9 1 1 1 1 1 1 s S ä § 1 1
i 1 CD (0 00 n m CD OS OQ o CO ca m CD CQ OQ CQ
i H <M «*> -* u\ <o t«- ^61^ JL & ?l » 4 UN
rH S
I
Is
H S
1«
tf • 9 t a • • •
a it q 3 •
1 • ^ a » ^ t d
tf • 8 R S • • • s ^ ^
• 5
^ 4 £ 8 Ä ä a »
• 5
^
H o • 3 S 8 3 3 I
8 • 5
BU • ? 8 Ä ä 4 4 i
l 4) O h —
ll||ü LA • 1 i i a
Init
ial
Pressure
(peU)
4 1 3 1 1 *
3
- 4
Amb.
Pressure
(peia)
4 i t 1 ||S % 3 3 3 s
1 \h * 4 5 5 ! -* ^ -* 5
iff • H
iS^ ^ Ä H -; H
•
3 o -H 5. h. ft, !0^
■I N ^J « « N
1 0} *^N ^-^ ^^S
•1|I1 i«|fi c^fss iijiii läjii^ |5.g^ ^d-ÜNN ^SNN »3a4i4
,^1 q -S so co
f o o o o Ü
1 H ^ 62 ■*
/>
1 ■J
jl f 3 4
S g 8 S ^ ^ 2J KF
1A
1 • • Ä Ä ^ 8 • • • •
?? " 5 »
tf
C
g » » 8 S « ^ S s » ^ & i? - d a
£ 2
i oi 3 9 9^ 5 6(; 3 «
t 1 1
3 g 8 R 3 ^ 3 J<
ililü O «A
3 i ^ CM O
3 3 5 1 ili« 3 3 ^ ^ ^ ^
2J 4 ^ 5
l|J 1 1 Ä A 5 i 4 4 4 4
i 55w 3 S 3 2 3 2
* It * » ft * ^ ^ ^ <r
H H
rS c^ CN ^\ H -i r^ H
^Q ^Q ?} -Q ^ ^3
1 | |5.55 l-S .55
Ml i|il iril i|il li'ijl fi'ii li"5i h^ii 5'A5/\/\ 5vr\5iAA 5U^5AIA 5A5IAA
►»t«-<vi • • St«-N • • SC^fM • • St^CVJ . . O^HtMN 0«^HHH O^^HHH O—'HrH-t
4 Q Q Q O Q Q
H N <»> -* .A .o 63 '
50
40
30
-0&2PSIG Initial Pressure I" Wall Egg Crate
O 0 PSIG Initial System Pressure
D2 P3IG Initial System Pressuie
A5 PSIG Initial System Pressure
20 10 12 14 20 22 24 1.6 1.8
Pc/P) psi
FIGURE 33 SIX CELL WING TANK EGG CRATE TESTS am 0397 te
6U
80
70
60
% Total Void 50
40
30
20
•
OPSIG Initial P •essure —.
/
y^ 2 psic ' Initial 'ressure —.
1
/ \
5PSI( Initial Pressure—'
\
1.0 1.2 14 1 6
Pc/P, psi
1.8 20 22
FIGURE 34 SIX CELL WING TANK CYLINDER TESTS 15 INCH DIAMETER
65
GP71 0937 17
% Total Void 50
FIGURE 35 SIX CELL WING TANK CYLINDER TESTS 75 INCH DIAMETER HEMI ENDS
Gi'7 1 rro» iH
OO
100 on
g&-o-£ ̂ «P nn I 1 I
> 1
701 ■ 1 '
.J 1 [
50 t I
40
30
CD 0
O Maximum Pressure A Inleimudidte Prdssuie Tt D Oata Calculated from
Combustion of all Voids 1 20 1 1 ^
Pc .- VTK
P, VTKVCIK-1I
I K Adiabatlc Expansion ;
10 I 'j Total System Volume.
VQ Combustion Volume 9 8
\
-V- 7 V— VT - Vr Relief Volume "fj I \ , - . ;;
4
3
V -
Ii£Dj v ^The jretical Curve
V 2
1 0
\ m
\
v
\ Q
^ V x^
7 ^ "v^^
6
5 •s^ 4
3
B 2
1 0 1.2 14 16 18 20 22 24
Pc/PI
FIGURE 36 PRESSURE RATIO vs RELIEF TO COMBUSTION VOLUME RATIO EGG CRATE SIX CELL WING TANK
GP71 0937 19
67
mm mm^m J
V0 V
90 80
70
60
50
40
30
20
O Maximum Pressure U Uatrt
Com LalculateiJ Irom instion of all Voids
1
P,. V TK
10
P1
K
VTK VC(K i;
AJiabatic Expansion
9 8
\ \ VT Total System Volume
\ V Combustion Volume —
6
C 5
4
\ VC
\ VT Vc VR Relief Volume
\ .A1 Fheoreticai Curve i
3
2
y '
< 5 O
r\
1 0
i \ i
V>
»V Q
7 i fc. 6 qD\^
"^v^
4
? 1 0 12 1 4 1 6 1 8 20 22 24
P P. c 1
FIGURE 37 PRESSURE HATIO vs RELIEF TO COMBUSTION VOLUME RATIO 15 INCH DIAMETER CYLINDER SIX CELL WING TANK GP71 0'> 1 ' . 1
68
100 90 80
70
60
50
40
30
20
VR/VC 5
10 9 8 .7
6
■
O Maximum Pressure D Data Calculated from
Combustion of all
\ P c
VTK
p. VTK VC(K 1) —I \ r1
\ K Adiabalic Expansion
\ VT Totai System Volume
\ Vr-
\ C
V^ V„ Reiiel Volume J
V ^Thec retical Cu rve
N ' 0 N 1 c
Si )
1 9
s_ "V. ^«J
"Jt. °\ ̂j
1 1 i 0 1.2 14 1 (> 18
P„/Pi
20 2.4
r2 rc'r1
FIGURE 38 PRESSURE RATIO vs RELIEF TO COMBUSTION VOLUME RATIO 7.5 INCH DIA. CYLINDERS GPJI o» 7 -o
SIX CELL WING TANK
A
g p >
E I E
90
80
70
60
50
40
JO
20
1C
0
'00
90
80
/O
60
40
30
20
10
0
"^
.i.i«d W.iiis
OPSIG
5PSIG
Fc|ii Cr»ti 15 in ill.i Cylinriei 12 in Long
^i
i
1 [_ 7 5 in ill., r, Imili'i
12 in Long
FIGURE 39 SIX CELL WING TANK RESULTS OP/I O'l.l.' •,.■
,•-
H 5
8 I
Ü a
II CM i
4) ^-^
"8
• a m
Ml bow.
I ll 2 —
55
^ d xi
r> o «> o
«* 5 4 ^ CM <0 O O
8 J^ 3 ^
>0 [- N "N
s a ^ ^
o o H H ON ^ t
WS tr\ Q CM
4 4 4 4
4 4 4 4
vO IA
CO 00 «0 CO
-t -* -i -*
M IM >A B f^ r^ O «»N
g | ■H E o n
I
IT» l/N l/S U> «N <V <\ CM
II H
•<-<"<•<
H CM f
•71
i > ?5 I
i
g l»> 8 (V » s R 8 •H--^ a. • • • •
< R H a o^ » i
II I1
.^ B H
• 9 «5
8 »
it 8 • • • R R • • < rH o^ t^ d o
Bomb
Sample
Ignition
Preaaure
Riae
(pel«)
to (N o Q O • R
« i H «0
lifi 3 s eo «0 «0
4> -P •> K ill« 4 4 •
5 4 si
Ut^^ * B * 3 3 K «Q <0
S » 5? i * •
4 •
4 4 4 £
a« iis 8 3 & Et 3 R
^^ 14 &>S «0 «0 so «0 • «0 CO
o uZ-> •* -* -t -t -i -* t*
^^
■A vt< o >o >o NO VO <»> C^i n (^ (^ f^
Fo
Dena
(lb/
r-J rH r-i H rH i-l
•-*
O O -H B l/\ U\ »Tv irt l/\ « «x f\l OJ (M o* tv
b, ft, W^' 1
g V ■ V V 0) 1) 1 * I i 1 ti ti I B E E E t. kl u o ü o u u
1 1 I 1 1 1 1 1 tc v
1 n CD ■ ca (0 »
1 r-H OJ CN -* in vO H
72
M
HO
10
60
% Total 50 Void
40
30
70
10
^ OPSIG nitial Pressu e
So
y <r
\y ^
15 16 1.7 18 2 1 19 2.0
Pc/P1 PSI
FIGURE 40 THREE CELL WING TANKS TES" LINED WALLS
2.2 23 24
GP71 0)37 23
73
■J
C
\
\
\
\
\
\ N 1 \ 1 I
tial
Pre
ssur
e tia
l Pr
essu
re -
ti
al P
ress
ure
\ L
c c c o* at 01
a a 0. 0 "N m
0D< 1
c ) N
a. u a
to
I- < u Ü o
z
u LU UJ oc X
cr
o
0 0 O O 0 9 0 O 01 00 r^ (0 in «» (N
I GP71 0937 f.4
7U
100 90 80
70
60
5C
40
30
20
VR/VC
I I 1 1 1 _ 1 II 0 Mii»ir"um Pressure —
•A Inlerniediate Pressure —
. D Data Calculated from - Combustion of all Voids
Pc VTK
L P, VTK - V^(K-I) ^ 0 3 T V K Adiabatic Expansion
I > L VT Total System Volume
\ \ "c Combustion Volume
\ \ vT- vc V R "* iet voi jme
\ ^
k > k N 1
c ] H D
Theore fical CL rve—^
\ N w i ^ ^N
tX L i \
\^ s s N
I 5
■
10 nq 0.8 0 7
06
0.5
04
03
02 10 12 14 16 1.8 2.0 2 2 2.4 26 2 8 3 0
Pc'P1
FIGURE 42 PRESSURE RATIOS vs RELIEF TO COMBUSTION VOLUME RATIO ■ LINED WALL WTIOM»«
THREtCELL WING TANK
.
75
Vvc
100 90 80 70
60
50
40
30
20
1.0 09 OH 0.7
0.6
0.5
04
0.3
02
r I I (J MaximuiT) fiessure D Cala Calculpted from
Uomoustion ol all voias I l I 1
Pc VTK
- Theo retical C L
.urve P1 VTK - VC(K-1)
C K !w<v-
K Acli./h.itK Expansion
Total System Volume
r iS s^ ̂
VC VT
Com
-vc justion VR R
Volume elief V( Jlume
V
^ H ■^c ?
' I \ \ \ V
\.
\ ! l I
\
\ 3 ; N " v \ sv S
V N. K ^X
S 5
10 12 14 16 18 2 0 ?.a ?4 2.6 2 8 3 0
Pc/Pl
FIGURE 43 PRESSURE RATIOS vs RELIEF TO COMBUSTION VOLUME RATIO EGG CRATE THREE CELL WING TANK üP7i Ü937 b?
76
■o ö >
E
I
■ o >
a9
I
I
5 >
100
90
80
70
60
50
40
30
20
10
0
100
90
80
70
60
50
40
30
20
10
0
100
90
80
7n
60
50
40
30
20
10
0
OPSIG
2PSIG
5PSIG
Lined Walls Eqq Crate
FIGURE 44 THREE CELL WING TANK RESULTS
cp/l en; 58
77
.
SECTION IV
PHASE II TEST PROGRAM
1.0 TEST SET-UP
1.1 Combustion ."Set-Up
The Phase II test cct-up for evaluating arrester effectiveness characteris- tics consisted of two jO-lnch lone 8-inch diameter 3/°" wall thickness plexiglass tubes, Joined with a divider plute containiiig 50^ open area. The tube set-up was designed to withstand the full combustion over-precFures generated by either a stolchioraetric JP-U/air or propane/air mixture. Combustion volume to relief volume ratios of 1 to 1, 1 to 5 and 1 to 10 were provided by using predetermined lengths of plexiglass tubing for the combustion side of the system.
l.P Instrument.-'.t i on
Instrumentation consisted of strain gauge type pressure transducers located in each end of the plexiglass tube arrangement and a l»0-gauge cromel-alumel thermocouple placed In the combustion side of the system. Pressure, temperature and ignition time data were recorded at 1000 cycles/second system response on oscillograph traces at lb inches per second. The temperature date was used as a fire verification only. The combustion side of the tube was cnilbratod with spaced tape strips along its length and high speed movies (1000 frames/second) were taken to obtain flame speeds.
1.3 Explosive fixture
The explosive r.ixturo used for the combustion tests was the same as described for the Phase I effort, basically stoichiometric propane/air mixture. A chemical analysis of the commercial propane used In the program Is included in Appendix I.
1. •• Ignit-'on 'ystem
The power supply for the ignition system for the Phase II combustion ter.tc is described in the Phase I test program. Ignition was accomplished by a point source ignitor with a l/U-lnch spark gap. fjpark ignition was initiated at the mid-point of the combustion side of the tube set-up.
1.5 Fuel Flow Pressure Drop Get-Up
Fuel flow pressure drop tests were conducted using the 8-inch plexiglass tube -ct-up described in Paragraph 1.1. The two-inch thick test specimens were bonded to the divider plate containing a hole which was 50'" of the cross sectional area. A flow straightoner consiiting of a ?-inch thick piece of aluminum honeycomb was Installed in the flow tuoe inlet stction. Fluid pumping rates wore varied from 50 to 150 gallons per minute using JT-1* at ambient temperatures a; the i'low media.
Preceding page blank 79
1.5.1 Fuel Flow Teat Instrumentation
Data instrumentation consisted of a turbine; type flowmeter connected to a frequency counter to measure fuel flow, a U-cube manometer for mcanurinc test specimen differential pressure and a pressure gauge to measure fuel inlet pres- sure. Flow measurements were taken from a dicital read-out type Secknan counter. System accuracy for the flow instrumentation was + 2 percent.
1.6 Thermünhysieal Propertios Test Get-Up
1.6.1 Thermal Conductivity
A guarded hotplate apparatus with ±2* accuracy within a moan temperature range of 150 to 3000F was used to generate the thermal conductivity data. Measurements were taken, with the material in a vacuum, in increments of IOC ? o-."er a mean temperature range of 150 to l*00oF.
1.6.2 Specific Heat
Data fur the specific heats of selected materials was taken with a Thcrno- pbyslea Model AC-lOO adiabatic calorimeter. Temperature increments of 100 ? over a range of 0 to u00oF were used for specific heat data points with an accuracy of ±2'.
1.6.3 differential Thermal Analyzer
The Robert L. Stone Model KA-2HD differential thermal analyzer was used to determine material melting points, latent heats and heats of reaction. XblB technique measures the differential temperature between an inert reference and the sample material due to chemical reactions as a function of temperature. Data is recorded on thermograms which are shown in Figures U5 through 55.
1.6.^ Surface Area
Surface area measurcnentr. were made by the American Instrument Co. using a :.'unlnico Model A?A k. This device measures surface area by the lev; temperature gas absorption technique where the quantity of gas necessary to form a mono- layer of gas molecules on the surface of the Siaterial is lueacurcd and recorded as surface area per unit material weight.
2.0 MATERIAL TiilST COiVFICURATIONS
2.1 General
All material specimens for this phase of the program were tested in 2-inch thicknesses, cylindricai:.y-cut to fit inside the plexiglass test set-up.
2.2 Material Configurations
Sixteen basic materials and configurations were tested to demonstrate their arrestor effectiveness. In addition, eleven coatings were applied to these various materials and configurations varying their chemical and thermophysicol properties to determine any relationships applicable to their arrestor effectiveness. The material configurations and coatings tested are presented in Table 15. Table 16 describes these materials and their respective properties.
80
3.0 TSiT rROC^DUR:-
3.1 Combustion Tests
In order to evaluate the many material confiruratlonc with a minimum number of tests, a test procedure was set up so aj to test naterials at th- least MVtr« tost conditions first where failure would eliainate further tostinn; of that material. This was accomplished by arraruUnr; the test set-up with a com- bustion volume to relief volume ratio of 1 to 10, initial system pressure of 0 psip and the natirlal wetted with JP-^. Specimens that failed to show flaaie arrestor capabilities within the defined success criteria at these condition;, vere olialnated from further t;st':nj>
Icnitlon tost procedures were the same for each specimen and test para- meter- The sample was bonded to the plexiglass tube divider plate and weighed. The specimen was then wetted when applicable, allowed to drain for five minutes and acaln weired. ■":: • llvlder plate and attached specimen were installed in the set-up and the pressure in the plexiglass tube reduced to 0.5 PSIA by a water eiuctor system. A pr«--mixed stoichiomctric propane/air mixture was introduced into both ends of the tort fixture to a positive pressure, which was then vented through the bomb-scnplor. ./hen the desired test pressure was established, the vent was closed and the mixing tank and bomb-sanpler isolated by closing the necessary valves. The bomb sample mixture was ignited ani the pressure monitored to verify the stoichiometrio mixture of the propane/air media. The mixture in the test article was then lt-nitcd and pressure and temperature measurements recorded on an oscillograph trace.
After the system was purgsd« the test plate was again weighed (Table 17). If the tested specimen was effective in limiting the system over-pressure to ■.ithin the defined success criteria, the test was repeated at 2 and subsequently 5 psig initial syster. pressure. This procedure was repeated with combustion to relief volume ratios of 1 to 5 and 1 to 1.
Coatings as listed in Table 15 wire applied to materials that showed a measure of success Including all the honeycomb configurations in spite of their poor performance. Tests on these coated materials followed the sa^ae sequence as described in the previous paragraph.
Both wet and dry sample tests were conducted. Wetting agents included JP-5 and water.
3.2 Fuel Flow Tests
Fuel flow pressure-drop tests were conducted on candidate mat trials using JT-1* as the fluid media. The specimen vas bonded to the plexiglass tube divider plate and installed in the test set-up. Pressure-drop readings were taken while increasing and decreasing flow rates from 50 to 150 and back to 50 gpm. I'atu for these are shown in Table 13 and Figures 5^ through 59«
81
3.3 Thermophysleal Properties Tests
Thermophysical properties testa were co.-lucted or. car:dldate materials as listed in Table 14. Thermal coriductlvlty moasurjmei.tr, within + if were made in increments of 100oF over a mea,, temperature ra.-.t.e of 150 to UOO0?'. Each of the ti st samples was sandwiched between the heater plate and a water-cooled heat 'ink.. The heat throuf;h each te^t sample (one-half of the total electrical pow^r supplied to the ce.-.tral heater) was raa..ually adjusted by rheostates. Rheostates were also used to manually match the temperature of the guarded heater to the temperature of the co; tral heater, 1.1.us ensurini a unidirectional heat flux through the test samples. By measuring the electrical power supplied to the central heater, the temperature drop across each test sample, and the thick- ness of each test sample, the thermal conductivity of the test material was calculated at steady-state by means of the one-dlmenslonal form of Fourier's law of heat conduction. Data for these tests are shown in Figure 60.
Specific heatn of the materials were measurer», In Increments of 10G F over the range of 0 to '♦00oF. The sample was Installed In the small flat box calorimeter which waa lowered i:ito the adiabatic guard chamber. A radiation shield was folded around the tiuard chamber and B vacuum bell-jar placed over the entire assembly. After evacuation f the enclosure, liiuid nitrogen flowed throu,^h the cooling coil which wis an ir.tegral part of the adiabatic r.uard chamber, thus lowering the chamber and calorimeter to their Initial starting temperature. Upon reaching the desired low tempcr:iture, the heater on the calorimeter was turned on and a regulated preset D.Ci power, was supplied to the calorimeter and contained sample. The snorgy suppll-i to the calorimeter serves only to raise the tefperature of the calorimeter plus the enclosed sample. The heat supplied per legree rise in tompernture represonts the heat capacity of the two. -'Ubtractlng the predetermined heat capacity of the calorimeter from the total yields the heat cnpaclty of the jample. The L-peclfic heat of the sample was obtained by dividing the sample heat capacity Dy Ui«i ..as.^ of the sample. Data for those tests are shown in "iguro 6l.
Melting point, latent heat a:.i heats of reaction were determined by differ- e:tial thermal an:dysir. The 3pecimei. wao place I In cj;itact with or.w Junction of a differential thermocouple while the other Junction wat placed in contact with an amount of high purity alumina having the came thermal r.nec as that of the specimen. Temperature was then increased at a pro^remmed rate in a con- trolled atmosphr-e. Whan the specimen undergoes an exothermic or endcthermic reaction, the Junctions of the differential thermocouple become unbalanced and an emf is generated and recorded, free Figure ^5 through pj,
Specimen displacement volumes were obtained by submerging a weighed sample in a graduated cylinder partially filled with a known fluid. The displacement volume, 7+ in percent of total bulk volume is given by:
V. . 0\ - Pb/Pt (100)
pt = true density-
Po = bulk density
82
AiM
Fuel retention of the three succesoful ccrabustlcn candidate materials was determined by submerging a weighed test sample in a room temperature fuel bath. The sample was then drained and re-weighed.
Surface area wnc measured by a low temperature t^as absorption technique which measures the iunr.tity of gas necossary to form a nonolaycr of gas molecules on the surface of a weighed sample.
1*.0 RESULTS AMD DISCUGGIOH OF RESULTS
U.l Combustion Tests
Data for the combustion tests of the Phfse II effort are presented in tabular and graphical form. Table 20 and Figures 6:-' through 65. The tables represent the raw data and reflect pres ■ re increases from combustion for both the Ignition and receiver sides of the test article. Testing of each specimen was initiated at the least severe system paraneters of combustion to relief volume ratio of 1 to 10 and a JP-5 wetted speciiieii. ^here material flame arrestor effectiveness at this condition did not meet the defined success cri- teria as previously discussed, no further combustiDn testing was performed with that specimen. It can be seen from Table 20, that considerable effort was naved using this method in that a great number of material configurations did not warrant further testing.
Graphs of the Phase 11 data are presented in the form of relief volume to combustion volume ratio versus the absolute final combustion pressure to initial system precsure ratio. Only 9 materials could be tested at a sufficient number of system parameters to successfullj 'e graphed. There are shown in Figures 62 through 65 and give excellent correlation with the theoretical curve as pre- viously establishec".. Other material data points where obvious failure occurred correlate with- this clicoretlcal curve when the combustion volume is made e^ual to the total tube system volume minus the material volume. These were not plotted as the pressure rises and rise rates were too great to be useful as an aircraft fuel tank suppressant system. It can be seen that data correlation using this method Is accomplished with the Phase I effort of the program. Of the nine configurations tested only three base materials produced results which warrant their consideration as effective flame arrestors. Tnese were 25 ppl reticulated polyurethane foam, fire extinguishing foam and i.V Scotch Brite felt. Aluminum tube core and polyester scretn produced erratic results and therefore their degree of success is questionable with respect to this testing.
The 25 ppl foam successfully suppressed the explosions for the following test conditions:
(1) Vr/Vc ratio of lO/l at 0, 2 and 5 psig initial system pressores with wetted, water-wetted and dry materials.
(2) Vr/Vc -atio of 5/1 and l/l at 0 and 2 psig initial pressure with JP-5 and water-wetted material.
83
The fire extinguishing foam was effective for the following test condlt' ms:
(1) Vr/Vc ratio of 10/1 and 5/1 at 0, 2 and 5 psig initial system pre'aure with JP-5 wetted, water-wetted and dry matericC .
(2) Vr/Vc ratio of l/l at 0 and 2 psig initial pressure with both wet and dry material.
Scotch Brite material was successful at all ratios of Vr/Vc tested for 0, 2 and 5 psig and wetted or dry material.
Coatings were applied to various materials as listed in Table If and In one case showed marked Improvement in the results. Aluminum tube core cjated with flourel In a Vr/Vc set-up of 10/1 at ambient initial pressure reduc 1 the combustion over-presrrure from 17.5 psig to 1.5 psig. This only performed with such over- whelming results in one case of several similarly coated materials, but indicates that with the right combination of materials, configurations and coatings con provide Improved flame arrestors. Other coatings tested showed Irtprovements also, reducing combustion over-pressures by from 12 tO 2h%. Polysulflde-coated l/8-inch perforated aluminum honeyconb showing a 12''. reduction in over-p.essures and glasc; resin-coated fiberglass providing the 2U^ change. Results of <ther material and coating combinations ranged between this minimum and maximum ycrcent improvement as can be scon in Table 20, Coated foams showed little if ary improvement over the base material results and in the case of the K3r and KI rjatlngs, higher com- bustion over-pressurob were obtained. Copper-coated foam h's a greater flame arresting effectiveness than nickel-coated foam or tne MSU polyurethane foam materials from an over-pressure standpoint.
V.'ator and JP-5 wetted arrestor material performed better than dry naterids with respect to limiting pressure rise and flame propagation. The water wetted samples performed only slightly better than the JP-5 wonted and dry specimen at large Vr/Vc ratios in spite of the jverwhelminc thsnul sink. This would indicate that the action was more physical jr chemical than thermal. At low Vr/Vc ratios; i.e., 1:1 only the '..'etted material was successful in 3limlnating flame propation. Merc again the water wetted samples only slightly out performed the JP-5 wetted materials indicating strong cr.emical effects.
Flame speeds for the combustion tests conducted at 0 psig initial pressure with a stolchlometric propane air mixture were measured by high speed motion pictures (1000 frames/sec). The number of frames that were spent showing the propagation of the ignition kernal in the calibrated tube were counted and the flame speed calculated. The average flame speed for these tests was 19 ft/sec.
U.2 Flow Test Results
Results of these tests are given in Table 18 and shown in Figures 51* through 59. Of the three most successful arrestor test materials, the 25 ppi foam resulted in the lowest pressure drop. As can be seen from the tabulated data, pressure drop reading for the foam and felt materials do not repeat as the flow is cycled from 50 to 150 to 50 gpm rates. This is due to the collapsing of the material as the flow is increased and the failure of the material to regain
8k
mm
Itö original shape as the flow is decreased. After a period of time, the material does return to its original shape and the tests results wore rjper.table. Materials felt to have promising flame arresting capablliti'.»: in adlition to the three discussed above, were also flow tested for latu correlation purposes and syatoin analysis.
Ix ', Thünnophysical Properties Test
Thormophysical properties measurements, included thermal conductivity, specific heat, surface area, bulk density, specific fuel retention, melting temperature and heat of reaction of selected materials. Con- figurations that showed little or no potential flame arresting capabilities were not carried through the complete thermophysical properties testing in that no benefit could be realized toward the comparison of materials without successful combustion test data. Table 19 shows the data applicable to the various materials. Figures 60 and 6l represent the thermal conductivity and specific heats at various test temperatures. Figures ^5 through 53 show the traces from the DM neasurements. Thermal conductivity and specific heat data have been included as a variable in the regression analysis of Phase III of the program and are given in Section VI of this report.
85
-■- - ■ _
SECTION V
CONCLUSIONS AND RECOMMENDATIONS
5.0 Conclusions
Phase II of tbe program has shown that other materials and configurations can be effectively employed as fire and explosion attenuators In aircraft fiel tanks. Although the number of material configurations that performed vlthln the test article vas not representative of an actual aircraft fuel system.
o Hie most efficient flame arrestor mattrials tested for all combustion parameters considered were the 3M polyester felt, (Scotch Brite) followed by fire extinguishing and 25 ppl Scott reticulated polyurethane foam.
o Hie proper coobination of material configurations and coatings can be effective flame arrestors. Felts and foams appear to be the most promising configurations.
o Large pore diameter materials are not effective flame arrestor configurations.
o Wetting the material affected system performance from an over-pressure standpoint and showed a greater flame arresting effectiveness where the Vr/Vc ratios were small.
o More material damage resulted in tests with dry specimens.
o Ttxe combination of fiberglass honeycomb coated with glass resin produced the greatest Improvement in arrestor effectiveness due to coating addition.
o JP-U fluid pressure drop is lower for the 25 ppl reticulated polyurethane than other equally effective foam and felt materials.
o The material thermophysical properties of thermal conductivity and specific heat have a small effect on the explosion attenuating effectiveness.
6.0 RECOMMENDATIONS
It is recommended that further testing of arrestor materials be conducted where specimen thickness to combustion volume can be varied and the L/D ratios for effective flame arresting can be determined. It is also recommended that these testj concentrate on foam and felt materials. Ignition energy also needs to be investigated as a variable for this type system.
Preceding page blank a?
J
Hnat of ~on ~
1 34 enl/gr
Hout of R<HH:tion ··' 30BO c:11l/gr cp • 0.6'1 Bwl# -°F JJ Temp of Fusion " 200°C ~ ?:' Temp of Reaction .• 2ll8°C ./ i :;
I
~ II":: : I'' : ji 500°C (932°F) -
1 t.< 1·>_) ••..•. .. ·;
~.' I ;~ ····• . ,, f; :· ;; .·•·
.. ;, '"*:: .. ~: \} < ·--
~ t_ ii'···' .. ; :.}_t' :,/i.· <·· ::::';.,. .. ,:, .... ,·,:,. ·::H ......._ (•':
:tl;& r.;; 400°C (752°F)- (' ... ···•·. ·.'
' .: .. ;: : . •... /> ....
•· ,· ';
I 'J ••........ •.·'.;·.
. ;:.; .;·· ..... '
.•.... ; •. '
~
\... ~~i :·;·f!:(·· I
I -----1--i; :
' •••••
I ,. I __ L 300°C (572°F) ~ I
' r--
~t\ ~~
..
Melting ,.-Point
2oo0c 1392°Fl
Exothermic Endother I!-me
FIGURE 45 OTA TRACE
25 PPI FOAM 150 IJV 2.239 :ng Sempl•
OP1,.,60Ml
88
Heat of Fusion ■ 17 6cal/gr
Heat of Reaction » 2530 cal/gr
cp-0.48Btu/#-0F
Temp of Fusion ■ 210oC
Temp of Reaction = 227
-Exothermic -•- Endothermic —
FIGURE 46 DTA TRACE
FIRE EXTINGUISHER FOAM 150 ^V 2.451 mg Sample
GP71 1606 11
89
. .
... i .
~--. ---~ I \ t
,__ ____ ,:·---·-...__u52°F) 400°C ;:::J::=;:t-----t-"1 _ __._ _ __.___.,
I I I ..
r i : : t : 1~-. -~!--, __...._: - i I I t ! : j 11 I 1' -+--t:t-t-+---1-------i ..___ ___ .._ ___ . __.__,_ --~
:',~r i I' V 1-- Mtllting Poln
1t
FIGURE 47 DTA TRACE 3M SCOTCH BRITE 150 ~iV 2.647 mg Sample
01'71 tl\0!\ ':>
()
1
Heat of Fusion ■
I Heat of Reaction ■ 763
3520
cal/gr
cal'gr
1
/
\i 00C(
"1
fcp-0.47 Btu/«UF
Temp of Fusion 2250C
Temp of Reaction ■ 2750C 3320f
s / l=bU 1
1
^ /
L4 S <*"- ̂
1
i
N L. "i
4( X)0C( 7520F)—~- ^/
1 \
1 i -N \^
N \
300"C. 157 ?0FI- L.-\ 1
i
i L - -j |
Reverse \.
Polarity—^ \
1 '
—1—
2 OCCI 392° c»—^
1 1
■1
! ; i t- E xotnei mic —— » 1 — Endothe rmicJ
FIGURE 48 DTA TRACE 25 PPI FOAM COATED WITH POLYSULFIDE 150 fiV 3 581 mg Sample
C.P7) 160'j 14
91
r----~·------------·~--~--~~~-r--,-~~~ Ht!iJI of Fw;ton •· 70.0 Clll/or
Hu;n c•f Heactinn '" 2700 cal/gr
cr) " D. 5!1 Rtu/# -°F ·- h.t-+--+-+-+---1--"-P--t r~'rnp of != u~lon " 230°C - .~ .. ( Temp of Roilction .. 202°C/ r- 5oo0c WJ2° r: 1 +r----flf-~-o·~, ~--o...w--+--t--+---1--+---i
;·
/l',f
. .r:-Melting .. .r I Point . _
. I 1----'1---t--+--"t--t-Exothermic ~f---t+--Endothermic-. I I I : .
200°C(3920~~:::t:: :
i l l I
I
FIGURE 49 DTA TRACE
25 PPI FOAM COATED WITH GL/\SS RESIN 150 11V 3.300 mg Sample
C\P11 11\05 10
92
------:r-Ht•nt of F •;c./on · t1tJ c;d/gr
Hent of R~ti!Cl ion c 2/tl\1 c;;f/gr
cp ~ 0.33 Bru;:l;:_o,~
Temp of F'"'on ' 244°C I ~ Tilrnp of ne .. ctioll 28fl°C I i '
-·l·-·-r- :mooc (932°F 1 =-1 .
I I ~~i .• : ···._··•···•····· I I !l 4{ .... ,, ~-J_. ~L~ r .. I it,ll'; I i I I . ,; ~~· '.~:. ::{;·;. I l i I . '.····
I I I I ~' ·\l'.;.,/ .. t;;;·2:t -~--~w .. · .... eli{
I v , . ·.· ,· j·... !. " ~' i i CJ~--- i >u }:'> ;.:;).~:; < ,,
I I 1 l ··:~ . / . i·\ , ! ::./;. I IO<l"·' <752?1 1'\ • ·.~. • .: I f-H v I '[\,)·~ .< '-+-H / ... ''· I>F, I I I \ .<~' ··•·•·· . i~:ti\;} I H-t ! .I \ •. iT •. M, I••·;)•~ I : I "\ •. I< . I> !;11 ' ' t l. i 1 i K . .'l' •': l:.i:j\} i I I ! ... f" !)· ·.···'1<?:·.
1
I I
~········'''' I I ! f I 300°C (572°F} ~a;
i I i I
,....,
~ i l -~
·-i ;
I ~ i i
! I I I
~~~ ' i ' I I I I I
---l- f-- Meltin11 Point I
I I I I I I I I I
. I
I I 200oc (3920F)-1-
cp700 I I Exothermic Endothermic-
FIGURE 50 DTA TRACE 25 PPI FOAM COATED WITH REDAR VITON 150 IJV 4.137 mg S1mplo
OP 7 1 -1110l\ 1
93
I
:' I
I I
l
H;' I 1', .··. ! t,.
I ~~ .. I ~~ -~--~~,~~ ... ++~-~~· -+~~
I ! i I i\. I i
~
;
-
I
I
I I I I I I ' I
r-------+l __ :_.l_l3oo0 c 1sn°F~-\ I J
I : I l-f+· · !. j_·4-+~r-~~~'-·~ i,
i ; i
[
I' I I I I '· I I I i i I
! I i . '1, i II I : I i i ;
: 1 I : j i ' 1 · ).- Meltinq
I ' I I I i /iii Pninl I ; : ~ I j I : J
H·-t--1 ll I ! L I I ! I ! I I I l I I 200°C (392fT l I
~~-__ L: i'. I I I I -++-+--+--+---4
LJ_____ - I J El<olt~Prmlc--l·--Emlnthermk FIGURE 51 DTA TRACE
25 PPI FOAM COATED WITH KEL·F 150 IJV 3.057 mg Sample
01' 1l I 001'111
1
i i i u...i ,,t * '--
1 1 97 7 <-»l/nr
1 Heat of Reaction ^ 1830cal/gr
I CP = 0.18Btu/«-0F
mp of Fusion ■ 206oC J00oC
1 1 Te
h V f Temp ot Reac ion « 250UC
t J00o(
__ ;(752 0F|-
,
___
J
4-
1
,
1
1
1 '
3C 0oC( J720F 1
< I
s S \ 1
'
|
■
— 1
—,
•*■ —
. ^
.
i ^_ Itinq f 'omt
1 1 l 7nn0r (iq70Fi—
s
RxoTh etmic —Fn lother mic-
POLYESTER SCREEN
FIGURE 52 DIA TRACE
150/JV 6.285 mg Sample GP7 1 1605 13
95
1 la^FIMCC L r -^-5 + - Heat of Fusion 0
^ k ^__ 4 -4- ' Heat of Reaction 3160 cal P \ cp 0 41 Biu a 0F >
V, i ; Temp of Fusion ?630C \ 1 Temi )of R eartion 2590C
^ A 1
♦ -—^-V—
■ —
400oC(752oFl A
LJ—__—A. T \ |
—i—i
^ r \
i 1
300oC(57?oFl—t- \ I : 1 -J
—'
J K -.
\ .
—
(
\ —
\
41
-Exo thprtr K -^ 1 — En (tothe mic
FIGURE 53 DTA TRACE NOMEX HONEYCOMB 150 ^iV 3 581 mg Simple
LiP/l 1r-0b 1!
96
mmm
>...: ;..;
f---4-4---it---t--t --+--t--+---+---+------------.-----~ .. ·-----·-·1 Ji'oem 15 ppi
-1---4--+--t-----------··--·-· ---··---·-.--· 1-"ire Ex:tinguiehing Foam
,3M S<:otch Brite
Aluminum. Honoycoob
Aluminum 1/8" Hex.agnol Honeycomb
-----4-·--~-4--~---+---4-4----i,__-4--------------·· .. -;..;
:1: *: . r--l--1----l
Perforated Aluninum Honaycomb
Aluminum 'l'ube Core (Dimpled)
Fiberglass Round Honeycomb {Tubea)
________ ._, .... ,_ -··-··--
XI~ 1--+-+·-f--1---+---!-:--f--+-+-+---------.. ---------
l~ ·---
Fiberglass Sine \-lave Honeycomb (Cot•re~tion) 1 .. -
Nomex RoWld Honeyc\-mb r---t--+--4----t--+---+----+---+--+---+--------·-··----·--·------
Nomex. Sine Wave Honeycanb (Corregation) ><I
1--1:---+--+-+--1--t--+--i--t---+------·----·-·-·--~-··-· Stainleos Steol Tubes .3/16 Diameter 0.0~9 Wall
Viton Tuboe
Polyester Scre~.n (Rollod)
1 1---+--1---+---+--1---·-·-··-+--...._....-J ____ i--------------·-.,·-l
St&inleas S!;,eel ScnH.:n 11
(Chovron) I
t---t--+--+-+---+--t---i----+--~-4-------·-~----·--------l
Copper Plated ~~--~~=--r·~~~----·1 t-k-+-"'--l--+--+---i--+--t--+--r--+-N-ic_k_o_l_P_l_at_o_· a_· <5 _!Pi ~-~
'&, __
____
., ____
---···-·------~---......---· ------------,-
---------
.:cr
.. :i~~.:=:tti
on
;,~
~~si
:,--
,nat
ion
l.
Ec:r
:~::
.:c..:
.b
Al-
l/S
-505
2-.a
:m:-t
... 5
2.
11~eycor~b
Al-
-?/3
0-50
52-
.C(n
:W-4
.3
J.
l'on
eycc
:co
AL
-1/'3
-505
2-.(X
)_)-
12.1
4-
Hon
e:rc
cc:b
A
L-1
/S-5
052-
.001
]!1
<:
Hon
e-;c
O"-
:b
.,.
H!m
-10
-1/S
-3 .o
I 6.
Hon
eyco
rl:
H:r
u-lO
-F/5
0-3.
5
7.
Hon
.eyc
a-:.b
H
?..f-
J/1
6-4
.0
g_
Hon
eyco
nb
H?.
F-r
/50-
3.5
9.
=~tal r~;b~s
::ate
rial
.~u:
-:--
L'1c
:::
All
oy
58
52 I
l39
Al'..
l":'..i
nu.':l
A
llo
y
5052
HJ9
.Alu
r:d.
r::z
: A
llo
y
5052
H39
Alur.in~ A
llo
y
5052
HJ:'
Mo
nt
~kr:.ex.
?Iyl
on F
arer
[)..l
f'ont
Ha>
ex
z:yl
on f
ap
cr
Gla
ss F
a"br
ic
Zle
in.fo
rced
Il
ssti
c
Gla
ss F
abri
c P.
ei.:1
.!' o
i'c ed
. H
asti
c
Al t:
.--.in
t:::
ID
-.m
;"
Cell
.3
ze
{In
ch
1/o
1/S
1/8
l/3
J/16
.137
5
~ ·1
· ~ ·
,
I -----~T
r0
1
1:1
2C
r:J1
e1
3
. i
(In
ch
) l
Cell
i:a
lls
~Cell
Sh
afe
.cx.
11
' rJ
on
-rerf
cra
ted
Jl
exas
on
al
t t
i
lllc
n-f
.orC
ora
t'd
I ?l
"~--Ccre
! I
I
4.5
j I
~.3 '
.CO
J P
erfo
rate
d
Hex
aso
nal
I
12.1
.001
3 T
ube
Co
re
!Ion
-Per
fora
ted
Hex
:a.g
cnal
).
0
Uon
-1 e
rfor
ated
. F
lex
-Co
re
).5
:~on-Per!."orated
Hex
.ago
nal
Uon
-Fer
i'ora
t.ed
Fle
x: -c
ore
.049
0
,12
. I l I
lJ.
1.4
.. ' ;
1.5.
16
.
1.7.
1.8 .
. I 1.
9.
20
..
21.
Scr
;:en
:::
loth
(6
9'\
op
en
a.re
F..)
Scre
e!l
Clo
th
{ 5.
-~
';J.
J ~ o
pen
ei-
:!a
)
Ret
ic:u
l.at
ed
?oar
:.
Ret
icu
late
d
Foam
.
Ret
icu
late
d
Foam
Felt
(S
cotc
h B
rite
)
PC 1
.981
, 3M
Com
pany
R/M
L-3
20
3-6
, R
aybe
stos
/Man
hatt
an
Typ
e 65
0 G
lass
R
esin
, O
wen
s-U
lin
ois
Red
ar-R
F-9
02-o
l.,
R.Z
. D
arl.
ing
CCIIf
i.Sa.Y
I I Ste~l
?oly:.u-~t..i:m ..
'le
Po
lyu
reth
ane
Po
lyu
reth
ane
vit
h
fire
reta
rdan
t ch
em
ica
l.s
Po
lyes
ter
Kel
-F
Flo
ure1
Po
1y
sult
ide
Gla
ss R
esin
Red
ar V
ito
n
:,r-
!nsi
ty
.7/Y
t.3
500
.
I ! lG
G
""""
'li"
·I
1 · .. ;
ire
jia..
=
' !/~ ~
.. !.b
! ~
' t li
near
beh
! .0
2)
I 1
I i
I l
I I
25 ?
Pi
l 1
.}6
'
I I
I .1
5 p
pi
I I 1.
36
' '
I I
35 p
pi
I I I I
J I
I
.....
0 0
I i I L..
...
:~
I .::c
:. ! ! 2
).
24
.
25. '
::::
o:1f
igur
at i•
)!<
~ c
Des
ign
atio
:.
l
Vib
rath
a.1
1e
A-7
61,
U:1
iroy
al C
hem
ical
Reti
cu
late
d F
oam
Ret
:\.c
ulat
ed F
oam
Tef
ion
-S 9
58-2
07
Gre
en,
Dup
ont
. i I
-------~-
Mat
eria
l
Po
lyu
reth
ane
:a P
lati
ng
S
olu
tio
n
Cu
Pla
tin
g
So
luti
on
Tef
lon
-·~~
\~G~
·:
.< 'J r
( c·~
\: :·
r' ~.)
~·~;;:~E::i
IAL
!-'?
CJP-
·:RI
.,I:
~:,
Cell
Siz
e
-. .;all
I'h
i:k
ness
(I
nch
) (I
nch
)
I
j -~
! ~ .. '":
'T"\
~1._
... !
" i
Cel
l ·J
al.l
s C
ell
.3h
ape ~ ~;?t:Y I
' i
I I
' .
I I I I
I
' I
'
Bef
ore
?:i.
r:t:n
.g
Aft
er F
irin
g
I Sa
mpl
e .,.
T-
---P
late
:1
ate
rial
ilett
ing
Sa
mpl
e D
ry
' I
Tes
t W
ai@
lt w
eigh
t Ag
en:~
\·
Ieig
ht
S1!m
ple
(A~~
)) I
;;~~
~ N
o.
Mat
eria
l (O
z)
(Oz)
(O
z)
~~)
Wei
ght
~)
(Oz}
I 1
Foam
25
pp
1
49
.5
52.0
50
.5
49.0
1.
5 JP
-5
2 F
aa
25
pp
i 49
1.
0 52
.0
51.5
50
.0
I .;;
JP
-5
I I 3
3M S
eo
tch
-Brl
te
49
2.0
52
.0
52.0
-
0 I
JP-.5
I
I 4
Fo
a 15
pp
i 48
1
.5
52.5
50
.0
I 2.
5 JP
-5
I -
5 F
o•
25 p
p1.
48
2.0
53.0
52
.0
-1.
0 JP
-5
6 3
ll S
co
teh
-Bri
te
48
1.5
51.0
50
.5
.5
JP-5
7 JH
Sco
tch
-Bri
te
49
1.5
52.0
52
.0
l 0
JP-5
I
8 JH
Seo
tch
-Brl
te
51.5
51
.0
50.0
.5
H;
P
9 JH
Sco
tch
-Brl
te
52.5
52
.0
51.0
.5
H
2()
10
JH
Sco
tch
-Bri
te
52.5
52
.0
51.0
.5
H
2()
ll
Fir
e H
ltti
ng
u1
shin
g
52.0
51
.5
51.0
.5
JP
-5
Fosa
.
12 Fire~
53.0
52
.5
52.0
.5
JP
-5
Fo
a I
I
13
Fir
e !
ltti
Dg
uia
hin
g
53.0
52
.5
51.5
.5
JP
-5
Fo
a
14
Yo
. 2S
pp1
54
.0
51.5
50
.5
1.0
~
l
I '
'
!est
No.
15
16
l7
t i t 1
8
I i 19
l
...-I
CJ
I ~
:'\)
! j 21
t
22
2J
24
25
2S
2"'
2~
~-~
Fo
e 2
5 p
pi
FO!IO
25
pp
i
Fi~ :x
tin
gu
:is.
.lrl
ng
F
oa
Fir
e E
rt.l
ng
uis
hin
g
Fo
•
Fo.
a 25
pp
i
Fo
e 2
5 n
i
FOS!
l 25
Pi=
-\
Fo
a 25
PFi
FO
lD 2
5 p
t'i
Fo
• 25
~i
Fir
e ~ir~ing
!"':;
)8
Vir
e !X
ting
-JJ.
.shi
ng
?oaa
~~eh-:ite
,:-,..:
-· .. :~-;~: ~.
•.:
_...._ __
_ " ___
_ ..._ --
--.... -~
4B
43
4B
43
48
~
~
43 .. ~
55.0
54
.0
56.0
56.0
54
.0
57.0
5
7.0
1.5
51.0
2.5
52.5
52
.0
2.5
52.0
50
.5
1.5
54
.5
5J.
5
2.0
55.0
52
.0
2.0
54.5
5
).5
).0
54
.5
52.0
4. 5
58
.0
55.0
J.C.
55.0
5
).)
,;a.
,.i
5!.
-.. -
-~--
... -· _
_ .., __
_ -~~--·-
-~-------·
52.0
52.0
5J.5
55.0
50.5
51
.0
52.0
51.~
·---
---·
·
Age
nt
1.0
2.0
2.0
0
JP-5
.5
JP-5
1.5
JP
-5
1.0
H
;,P
).0
H
;fl
1.0
I
H;!
)
2.5
1
JP-s
).0
I J
P-S
l.O
I v
'P-S
I
_<-~ J
~~--_' _
___.
,-,-
I I
i !
Tea
t I
No.
M
ater
ial
I
JO
I S
cctc
ll-&
:tte
l
Jl
Fir
e E
lcti
ng
uis
.llln
.g
Foa
:a
)2
Fo
m 2
5 p
pi
JJ
Yo
el 2
5 pp
1
J4
Fo
sa 2
5 p
p1
35
Fo
. 25
p¢
. .- 0
J6
P'o
a 2
5 pp
1.
w
J7
Seo
t.ch
-Eri
t.e
JS
Fir
e 1
Xti
.ng
uis
h1
ng
P
'oaa
39
Fa
. 25
pp1
.
1/)
F
oa
2S
pp1
41
Fa
. 25
p¢
42
fire
Erl
.i.•1
gn1
aMn
g
lo.
43
Fir
e P
.d.i
ngu1
shin
g Y
o.
n.re
Ea:Ua&~dsb~ na
48
F
o.a
).0
1.5
2.0
1.5
1.5
2.0
4.0
2.0
2.0
1.0
2.0
55.0
52.0
56.0
59.0
5l.O
49.0
50.0
4B.5
48.5
49.5
50.0
52.0
52.0
50.5
54.0
54.0
so.o
50.0
so.o
so.s
51.0
Dr-
y
Dry
t ~;
{~
Dry
L>ry
Dr7
Dry
Dr7
Dry
J.O
H;P
1.5
JP-5
DrT
2.0
JP-5
s.o
H40
..-----~
I I
Ka
teri
a.l
I
Tes
t I
I N
o.
, -
I ~
I -
l 4 5
I
F'ir
e E
:rli
n,&
'J.b
hing
'
1 T~
!I_
Sco
tch
Bri
ta
4 7
Sco
tch
Bri
te
48
49
50
83
8S
S6
88
Seo
teh
Bri
te
Sco
tch
Brl
te
Foam
25
pp
i P
ink
Fa
.
.U-1
/8"
Hex
ap
1
Boo
eyca
:ab
Kel
-P
Co
atin
g
Al-
Tab
e C
ore
Kel
-F
Coa
t!D
g
Al-
1/B
" H
ex.e
gnol
H
c:lD
e;Jc
c.b
Xel
-F
Co
aU
ng
Plb
erg
l&sa
1/8
-~
Roa
nd J
lon
eJca
lb
lel-
Y C
oet
1n
g
JJ.-
1/8"
Pe
rto
rat«
<
~Xel-P
Coa
t.!D
g
lc:a
Dt
!ten
S1a
e W
aft
Jfc
De.1c
CDb
Kel
-P
Co
atin
g
Pla
te
Wei
ght
(Oz)
48
L.8 48
48.0
413
48
J.5
2.0
).0
).0
2.0
4.5
7-5
).9
2.9
9.3
55.0
5J.O
59-5
59.0
52.0
52.5
51.0
51.0
49.5
53.8
57.9
55.0
51.5
58.4
52.0
1.0
.)
1.6
4.0
.J
.s
----
-l I
-,;;
!~~
I
~a
JF-5
JP-5
JP-5
JP-5
Ji'-5
1
I B
efo
re F
irin
g
Alt
er F
irin
g
--I
I I
Sam
ple
T
Pla
te
Mat
eria
l W
etti
ng
Sooi
pl.IT
; Dr
.r T
eat
W.e
ight
. W
eig
ht.
A
gent
. W
eig
ht
Scp
le
(A}-
(B}
We
ttin
g
Bo.
}{
at eri
al
(Oz)
(O
z}
(Os)
(O
z) ~
(Oz)
A
gent
(A
) (B
) I
(Oz~
89
ll<:
iim~
J/J.
.btf
Rou
nd
48
J.4
5
2.2
5
1.2
I
--I
1.0
JP
-5
HG
"ne7
Ccm
b K
el-F
I
I C
oa
tin
g
90
J.l'
iber
gl.u
a 1/
8" S
ine
48
J.O
52.0
51
.5
-.s
JP-5
' f I W
lrfe
Honey-~ Kel~F
Co
atin
g
92
Al-
1/8'
• P
erl'o
rate
d 48
9.
0 57
.7
57.0
-
.7
JP-5
~Glasa
l R
esin
93
Al.-1~ l
f«u
gn
ol
4S
4.1
5).
6
5J.
l.
~
.5
JP-S
~Gl.Ms
Res
in
94
.U-l.
/8'•
Hex.a
gno1
48
4.
5 54
.2
53-9
-
.J
JP-5
H
ale7
ccab
Gl.
ue
R
esin
95
Fib
erg
1&
ss 1
/811
R
ound
48
J.
4
5).9
5
).0
-
.9
JP-S
Hc
nt~J
ccab
Gla
sa
. '
l?.e
aira
l
96
Jl-T
ub
e C
ore
Gla
ss
48
7.8
58.4
58
... 2
-.2
.na
-s P
.esi
n
.
rn
.U-1
/8"
Per
fora
ted
48
9.
1 58
.7
58.7
-
0 JP
-5
lfcae
7cca
b R
edar
V
itoD
Co
atin
g
98
.U-1
/8"
Jfe:x:
a&n0
1 48
4.
S 54
.4
54.1
.)
.na
-s flo
ne:rc
aatb
Vit
on
C
oat
ing
;
L .
--·-
'----· -
--
lDl
102
lOS
106
Fib
erg
lass
1/8
" S
ine
\rla
ve H
on:q
c:C
Db
Vit
on
C
oat
ing
Al-
Tub
e C
ore
Vit
on
C
oat
ing
Fib
er&
].as
s 1
/811
R
ound
H
on
eyc
cab
Po
l;:r
sul!
ide
Co
atin
g
llc:
:as
3/16
" R
ound
H
on
eyc
cab
Po
l.y
sul!
'id
e C
oat.
iDg
Fib
erg
lass 1
/8'*
Sin
e W
ave
Jfone
Jcam
b P
ol,
.wti
de C
oati
l'lg
Al-
1/t
r' H
exag
n.ol
lfo
ne7c
c:ab
Po
l:y
sult
ide
Co
atin
g
9.0
6.2
6 .. 7
5.9
8.)
48
s.a
-~-
---<
~ -w
-.----
·
JP-5
52.8
52
.5
.J
JP-5
53.4
52
.8
.6
JP-5
58.9
.8
-
JP-5
5J.3
56
.0
JP-5
56.0
55
.4
.6
JP-5
.)
JP-S
57.1
57
.0
.1
.lP-S
58.7
58
.4
.)
.lP-S
---
·--·--
--~ -·«~
~ ---
--__ ,...
....,.
, . -
,---~---
"~--------
----
----
----
----
----
----
----
-·--
--.-
----
----
----
, B
efo
re n
r....r
:.g
Aft
er F
irin
g_
--
--.-
---
r---i
Sam
ple
,..
I P
l.i"l.
te
r~'ltedd
',;:-
;;tt
ing
Smnp
..1u
nz,y
! ,_
.,:
l Te~
\l:l
igh
t ',•
:eig
ht
i~~t:.nt
;-led
g."l
t
I S:r.
mp1
c
I N
o.
rr.a
terh
l ((
}.;)
(O
s)
(<4)
(0
!1)
Wei
ght
(A)
(B)
{03
)
I lD
S
A1-
l/&1
He.
u.gn
o1
I 49
7
.4
57.0
56
.a
I •
Hon
e;rc
rn.tb
Pol
:nru
J.,r
ide
I C
oa
tin
g
I I
I I
t.-J.4
62
.9
I 10
9 A
l.-1
/811
P
erf
ora
ted
4
0
1).
0
t
r l
(A)-
(s)
wet
ting
1j
( 0:1
) !g
ent
.2
JP-5
.5
JP-5
H
oney
eCD
b PoJ.r.ru.lfi~e
Co
atin
g
11
0 Po~ R
oll
48
lS
.O
68.5
JP
-5
lll
n-Tu
be C
ore
48
14.0
6
).5
6
).0
-5
JP
-5
. il1
2
25 p
pi-
Fo
a
48
1.9
6).
4
52.'6
49
.9
Gla
ss R
esin
10
.5
JP-5
11
) 25
pp
i-F
oa
48
J.5
6)
.9
52.0
ll
.9
JP-5
lf
i C
oati
ng
114
25 p
p1
-FO
B
48
2.6
5).6
48
.0
50.6
5.
6 JP
-5
Ke1
-F C
oat
ing
115
25 p
pi-
Fo
e
48
2.6
I 54
.7
5J.
J 5Q
.6
Glu
e
Res
i.'l
'116
25
pp
1-
Fa
. 48
2.
6 55
.7
52-~
50.6
'
Vit
on
Coa
t:in&
I
I
1.4
JP
-S
).2
JP
-5
I 56
.2
117
25 p
p!.
-F
o.
48
4.2
54
.0
2.2
JP
-5
Pol7
5Ulf1
de C
oati
ng
us
Al-
Tnb
e C
ore
48
8
.8
59.0
58
.7
.)
R-5
F
lou
rel
Co
atin
g
' 11
9 P
ol7e
ster
Rol
l 48
I
3.4
I 20
.0
121
25 p
p1 -
Fo
m
49
_:~~:r
. Ccat
1.-:_~
-----·
. L __
____
... 2.
1 ..
I 5~-2
: 54
.1.
I ---
------
:_ ___
__
,_
_,,,,,,, ---
---.._
_ __
___
-----
----
--·-
-I
JP-.5
" "'T
-5
. ....
J!
,.--
-·
--·---
----
----
--
f Aft
e:P
Fir
ing
~;fo:t~ F
irin
g
----
·-·-
----
-..--
I --,-
--, r·--
-----r-
---l
Sei
nple
I
I -
1 T~,,
~ l
P1:~
t6
! M
ate
dE
ll
H"e
tti.
ng
Sm
::tpl
e D
ry
" '
( ..
-::;
, .. ,.
, .. ~
; !
! ...
i; ·.
·: ·~~
ii &
...:,.
,:· .~.
·;~ .. t:
J. :;
\ r"~
}..-··
.._
• !. ~:
! ..
-l '
~ -· 4 !
.. .;
:. .·
i (~
y·
i i
·:....
.to
. =~terhl..l
<oz)
I (~~)
(Oz)
I
We
igh
t (0
:1)
Ag~ .. ·t
, (A
) (B
) (a
S)
' -
I I
56.<
::
I 122
25 p
pi -
Foar
.:t
48
;.o
;4.2
53
.0
2.0
JP
-5
Po
ljr'
sul!
ide
Co
atin
g
123
.25 p
pi
-Fo
6&U
48
2.6
5
J.l
51.8
.5
().6
1.
3 JP
-5
p •
I G
l.ass
.. e
s:L"
'l
124.
25
pp1
-Fo
am
48
.l.o
53
.1
52.5
50
.6
.6
JP-~
J K
el-F
Coa
tin
g
' '
; :
b (D
;
I I
l
·. ·,
TABLE XVIII
H.ATERI.ALS FUEL P'WW PRESSURE J:roP DATA I
"'~·-~------~-----
Fuel P'low R4lte Pre1aure Drop K~?.tCrt:"ial (GPM) (In. Fuel)
,-..,.... .. .,.,_ ...
ppi Pmtrn 52.0 1.6 103.0 6.2 148.0 l2.b 10),0 b.8
--..-~ 49.0 l.b -52.0 1.7
103.0 6.4 150.0 17.b 102.0 7.5
5'!.0 2.0 59.0 1.8 ·•. ""
103 .. 0 5.3 150.3. 12.0 105.0 7.0
Fire Extinguishing Poem 49.0 12.1 10b.O 15.4 140.0 28.8 102.0 16.3
51.0 .L5.2 49.0 13.9
102.0 16.1 140.0 ,30.,3 102.0 16.7
53.0 15.2 ....... -------~-~-
311. Scotch Brite 55-4 6.1 100.6 17.0 156.0 39.4 104.0 19.1
!>b.S b.? 102.0 18.2 15,3.0 41.0 102.0 19.5
57.7 s.J 104.0 15 • .3 152.0 .32.5 60.0 s.B
---"-Al-1/8" Hre:reanb 2" Thick 6,3.0 0.4 h. 5 lb/ft 105.0 1.0
·150.0 2.0 106.0 ... .,._ 1.1
54.0 0.2 6,3.5 0.4
105.0 1.0 150.0 2.0 106.0 1.1 ..
54.4 0.2
-·-··
l09
M!\torial
AlHon
1/8" Per!or1.tod 2" 'J'htr.i< eyeanb 12.1 lb/ft2
1/8" Haxegnol ? 11
1eyccnb 4.3 1b/ft2 'rh J.ck
----Al Tube Cora
I r--;:~ er~la.ss l/A11 Round 2"
eycomb 3.0 1b/ft2 Thick
1 Hon
~--Fib
,,_ . .._.,
erp;lrus:l 1/S" 3L"1e
I e Honev-ccmb
11 lb /rt2 • I
\~av .., ...;•
I __ _ "----
I
i
25 ?"' • J.
rri Fomr. Copper a ted
L_
?" Thick
·-I 25
Co ppi Fe~ Polysulfide
I a ted
·-" __ . _ _.._, ____ ... __
~--
"""""-
Fuel Flow Rate Pres8W"e Dro~ (OPH) Cin. Fuel)
s6.s 0.6 10,3.0 1.4 155.0 "- ).0 104.0 1.5
55.5 o.s -
52.0 0,4 104.C 1..3 -150.0 2.3 102.0 1..3 51.0 0,4
59.0 o.e 103.0 1.7 151.0 ,3.5
"-57.5 0.3
104.0 1.0 150.0 2.0 103.0 0.9 65.5 o.s 54.4 0.2
102.0 o.s 146.0 1.7 100.0 o.s ;6.5 0.2 1
l I 54.0 4 • .3 .
105.0 s.s I 150.0 11.1 102.0 5.6 5).0 5.1
53.0 2.0 104.0 7.9 141.0 20.8 10).0 10.1
50.0 1.9 51.0 '·- . ...._ 1.8
103.0 8.2 14.3.0 22.4 104.0 10.8 53.0 2.1
, .... , .. ._. ~ ."
uo
ö u. ui Z < I
11 Ei R| u. > O Q a UJ o s i « Q8
M I S2 > o
si
lanj ssipui ■ dojQ «jnstsjj
1U
to Ku
LU lU QC CD
S5
S 1 in *
^^
oi UJ a cc a.
co O to (J
OL •-
S S s g) o g
IT) in
D
\ar\j sanniii dojQ sjnsssi,)
UX'
jo 2
- §
-i
R O V) CC LU
o >
ii | F B <
a QC _. a. S
o > O £ "-
o ■ §i X U.
si s is
-«
-8
Irid-j saipuj - dojo sinssa/j
113
i |3nj ssipui doiQ ainssajj
LU
i.inj ssgauj doiQ ajnssajj
115
\anj sat) mi doiQ ajmsajj
116
---------------------------~~~~~~~~~---~~~---~~-~-------
~1'Ar.L.t.
;, .i.
A
l·~rrRiftL
'I'HER
?<C
FR"i
SIC
,:.L
F-~~ OP:::n?F~S
-----------·--:::;:::::.,-----.---·-.,..------r----r·---l-------r-7"0~=,.. -..
.,--------.
r -_
_::
::::
:::j
1
spr.:
cn-:_
,; ,
?R
OP
M:T
ES
---------
~,K B
TU-r
-,~;
T
H
EA
T O
F P
nLK
__
Ft
rl!L
SUR
FAC
E .
__..-
-·•
Cp
;.aT
R
EAC
TIO
N
DS
NS
11Y
D
ISP.
uAC
EHEN
'J R
ETEN
TIO
N
ARE
A
~-----
----
~.A'IERI
AL R
R-n
-2
F Y:I
DJ/ #
-F
F'
-+-B
TU_f_
if_' -+-
#/_n_3
-+---~
-o ---
t--#
_'f_
n_3 -t
--~-~_
'1 __
1
r FOA.
.. 2
5 p
pi
i 0
.47
__
+--
_.6
3 _
__
}_?2
556o
!.
66
2
2.5
6 (5
~) 48
eo
,
j ~~
-mr~
~--
-o_._l
3 __ 4
-_
__
._53 _
__ .~_
41_9~r
--4~--
--+---
2_.~--
-~----
2,_o_9
__ -+~
-·-~-{
_7_·3_
~_)-r-
---4~-
·----~
'
I !
FO
AM
-25
pp
i KEL~F
0.3
2
.58
4)(
59
90
--
-63
50
FOA
M-P
OLY
SULF
IDE
0.33
43
7 63
50
--
-34
20
0.41
.5
9 41
.6
4980
-
--
685o
0.21
40
40
--
-44
oo
t l
...-FO
AM-K
Br
0.3
0
.5
--
--
-10
750
~~-------------------+-------r-------1----~----~-----~-------~-----~r----------;
roA
H-K
I 0.
32
.52
--
--
-)I
J20
FIR
E E
XTI
MG
UIS
HD
IG F
OAM
0.
24
' -5
5 41
0 3-
77
4.4
0.3
9
--
--
--
24llo
~----------------------+--------r-------4----1~----1-------r-------~r-------r----------,
1.52
-
: -
--
--
1965
.18
401
3300
.41
5700
O 25 ppi Foam
• Scotch Brite I3MI
A 25 ppi Foam Kel F
A 25 PP< Foam I Polysulfidc O 25 ppi Foam t Gla's Resin
B 25 ppi Foam t Redai Vitun
0 25 ppi Foam t KBt
' 8P f 25ppiFoa.-, I Kl Q Fire Extinguishing Foam
; |_ ^ 25 ppi, Foam Ni Plated
■^•25 ppi, FoarnCu Plated
200 3a Mean Temjierature F
FIGURE 60 THERMAL CONDUCTIVI fY OF FLAME ARRESTOR MATERIALS
500
GP/1 1606 6
118
1.01 0.9 t--------+------+----- -1-----+-----+------f
0.8t-----------+----------+----------+----------+----------+--------~
0.2
0 25 ppi Foam
0 Scotch Brite i3M)
D. 25 ppi Foam Ke! F
0 25 ppi Foam+ Polvsulfide
<> 25 ppi Foam .f Class Rt!sin \l 25 ppi Foam + K Br
0 25 ppi Foam+ Kl 0 Fire Extingui~hing Foam
0~.-----·-----~----------._ ______ --_. __________ ._ ________ ___ 50 100 150 200
Temperature· °F
250 300
FIGURE 61 SPECIFIC HEAT OF FLAME ARRESTOR MATERIALS
119
350
OP 11 Ulotl 5
I
I A
."'lb
ient
I
::ater:iCl~
l re
ssu
re
I (r
sia)
I i ,- I Fo
e;::
25
r:ri
J..
4.50
r I F
oar:t
25
rri
11;..4
.4
FOS
."':l
25 r
:ri
11;..
27
Foan
25
:rri
11
;..48
Fo
an
25
Fri
14
.43
Foa;
:: 25
rri
14
.31
Fo
sn 2
5 rr
i 14
.5
Foa
:: 2
5 rr
i 14
.5
Foc
r: 2
5 :r
ri
14.1
..1
Fo
s::
25 rr
i 11
... 2
3
Fo
an
2'>
J=ri
14 I
''
•-"0
Fos;:
1 25
rri
.
;
14.3
1
Foan
25
rri
14
.4S
:o~r
t 25
n
i
1.4.
!.,8
Fo
e.
25 r
r1
14.3
1
7o
r. 2
5 rr
i lJ
.,.JS
Fo
r. 2
5 rr
i 14
.41
:··oa:-
. 2 >
lJ i
14
.)7
. .
TAB
LE
XX
:':.A
Till
iAL
CO
l ffi
lJJT
IU;
T-::·~·)"TS
1Ube
V
olur
:e
J.at
io
Init
ial
tres
sure
H
~>r
.. ,, ~ ,,
(pia
) :.
I_;
10/1
]4
. 50
10
/l
14.4
1 ..
10/1
12
..27
5/l
14
.43
5/1
14-4~
-'I
)J
, 14
.31
I
1/1
14
-.)
1/1
14
-.)
1/1
11..4
1
10/1
16
.23
10/1
16
.4S
10/1
16
.31
5)1
16.4
J
5/1
16.4
n
5/1
16.3
1
5/1
16.3~
1/1
16
.41
1'"
I~
16.J
7
DT
61
I
• >
. \.o
: :bc..:~t io
n
Jeceiv
er
Jid
e ~·,..~
·~,i
etti
:..g
.:
:J~u
e
Cis
ig)
(}:s
ig)
.A.~e.J.t.
' I -
2.0
2.1
JI 5
i
1.7
1
.3
H20
1.e
1.8
iJ
ry
4.1
3.
9 JF
5
].5
2.
5 H 20
3.8
3-7
Dry
17.0
14
.5
Jt:- .. ,
17.0
14
-5
Jf5
62
.9
62.6
D
ry
2.4
2
.4
Jr5
2.0
2.
1 H 20
T!l
.:
Xce
ss O
f 35
fP
. 'i
}ri
'
' 4.
3 4.
1 Ji
5
!._.2
4.
2 n 2o
1.9
1.~
0:7
4.5
4.
4 D
ry
1;.
o
14.3
Jl
5
J..4.<
) 1.
4.4
H 20
,-~----------------------------,
-r --r--·
T
l k
l.b
ien
t.
'/o
lu::
e T
ube
llt
Lu-
••
t .
· ·8
-~r
1al
F
ress
ure
I E
at.i
o II
'.it.
ial
C•)
<:-.b
ust.i
cn
;~eceive:t:
·.-.;ett:L~
1 (r
sia)
'1 .. /V
... 1-fess~e
Sid
e S
ide
Ag
en
t I
~'-
\.,
~~~~
(~ig}
(~j
r i
........
I t
.. • ;)
~{:1
.rr~
14.~6
10/1
19
.46
-I
-JF
5 ' .
?car
:. 25
rr
:t
u~.:s
10
/l
"':-
,...
., . ·'
I }
.0
2.5
I
H2o
..... "1
" 4.
-~)
fOI!
t:'l
25
rr
i 1
4.)
1
10/1
1
9.)
1
79.5
-
I Dr
:r F
oan
25 rr
i 14
.!,·3
5
/l
1').4
<:1
In :
..:X:
cess
o
f 35
rs
i JI
-5
Fos
:: 2
5 n
·i
14.4
3 5
/1
19
.4d
In
:.:X
cess
of
35 r
si
H2o
;
I
I
• f
l'ub
e
' L
\r
! !
.. t:b
i~~~
nt
Vol
u.-.
e :n
itis
l C
Dr:'
,b:..
ts t i
on
1
leceiv
er
:;ett
lng
r~ate:--1al
Fre
ssu
re
:lati
o
1-r
essu
re
Sid
s S
ide
Ag
en
t
I (r
sia.
) V
:/V
, (J
si:
;)
(I si,~)
(rsi
;s)
.. \.
,
'?::
I ri
Foa:~
11;. ·
'·
10
/1
14-4
2.
5 2.
2 I
JF5
I ... .
~
' : ·
:..:l:
rs :ll
:""'
i d '"'"'
! • I
,-rr
i ?o
e::
I 14
.7
)/1
1.
47
! ...
:..
4.1
JF
5 ..,
:J
• 1
Po l:
rsCll~-il e Coatin~.
25 r
ri ?
oa
-:
u~. 5
J.
/1
14.5
-
-Jf
5
Fo
l-rs
t.tE
ide
Go
at in.~
25 r
ri ;;
"oa-
: 1.
4.4.
1
0/1
16
.4,
2.5
2.J
Jl
-5
Red
.ar-
V it
on
' .,-
rP.
?ca
rt
1.4.
7 5'
1 14
-7
4-7
4.4
Jf
5 ...
;~
# ~
Vi t
on
t:oatin~
I
25 n
:i F
oa
o
14.5
1
/1
14.5
5S
.a
58.6
Jf
5
Vit
o:t
Co
ati
n1
I ::
ate
dal
I I
-j
o•tin•
l k.
:~~i
t"!r
. t
'/o
l:::
:c
Go"'"
,:_, I
.~1
"'*'-
.. <..
. J
I r~
Jsst
.:.;
:"~_
: ~~t.io
.:d
tial
Cr:J
.-.~
~5 t
ion
I
:~ec
civ.
c;r
A-_~cnt
(rsia
) v..
, ·v-.
es
surc
.3
ide
3id
e :/
'v
(rsia
) (t
:sig
) (:
;:si
d
~· ·
~· 11
...00
10
/1
11,..
60
1.7
1
.7
JI-5
m
-,.n
L~~
- I ""'"
~int~~shL~~
I .,
1
t .-<
\.
10
/l
'1
, t
~~
i 1
.5
1.4
}i
zO
::!.
re
I .J.
..i ..
.. "-
i--'•-
) .l.
...L+
•""''
-"
Foa
n I
I
Fir
e =
xt.i
n~s.
l-tL
'\rt
: 11
•• 5
0 10
/1
la.o
o 2
('
.u
2.0
Jf
5
FO&
:I I
Fir
e :!din~ shilL~
JJ ...
;.e
10/1
l6
.l~J
l.
9
1.7
H
2o
FO
fll
Fir
e !
:Xt~
··ui
shi.
"'t.
~ ~-42
1;1
1
6.4
2
15.6
1
5.2
D
ry
' F
os:t
'
l '
Fir
e ?
!cti
n";t
lisi
li.r
•. :~
11-+
.Go
10/1
1'
}.60
2.
2 2.
2 JF
5 P
'osa
Fir
e :
:!d
in ~S.lria1
14.4
S
10/1
1
9.4
3
2.1
2.1
H 20
'?o
a
Fir
e :
xti
n "':
Uis
!l i."
'l. "'!
14.1
.3
10/1
19
.:~)
2.
4 2
.)
De
y F
oa
l
FJr
e ~1n":Uis.'lin ~
u.:
... J 5/
1 19
.1.3
·:'!
5.
0 4.
7 Jl
-5
Po
•·
.. -
<
f. F
ire
~inmdshin'!
14-4
3 5/
1 l'
].4
J ~~
-5
I 4.
4 ~
H 0
2
Poap
.
Pi:
e ::
xt in
"';U
ishi
n1
14
.)1
5/
1 1
9.)
1
4.5
4.5
;r.y
F
or.
Fir
e :X
t.in
":Ui
s.~i
.nF;
1
4.4
1
l./1
1
?.4
1
17.2
16
.7
Jf5
?oa:
:
Fi:-
o .X
tinr;
.us!
rln
:"!
14.!~1
l/1
19
• .+1
16
.!>
16.0
H
O
fOS
"t
2
' Fir~ ~in~in~
1' ••
:..2
1/l
11 1
.42
;
93.5
-
;)ey
Fo
e:
~-------------T--
1 I t,;~~~
~, ....
... ...
..., .•
a~i:!rl.a.:..
l 'i::-;1~:;
:c ;t
.,'lt
io
..,/~ 1/'
! 1., .•
v
10
/:
1/1
T'J
be
I1ri
tial
! :'
"~SS
Ur0
2,}
Cc:
.h:1
st.i
on
1.6
11~.
9
\ .
......
.. l
~
. ,;
.ece
.lve
:-
l.b
Age
nt
,J15
~------------------~--------------4------------+----------~-------------r---------------r--------~
):: .:;cot<±-Bri~e
]:~
3co
t6-I
3ri
te
)I:
Sco
tch
-3ri
te
1.4.
:,.4
1.4.
)1
.)!~
Sco
tch
-Dri
te
14.4
0
):f
Sco
!.ch
-B::
-ite
10
/1
1
10
/1
10
/1
t; 1
-,
i ..L
5/1
1/1
1/1
1')
.!.0
19.6
8
19.4
4
19.1
.4
19.1
.4
1')
.)1
19.4
0
1•).
40
19
.40
2.l
2.)
2.5
5.2
17
.'/
17.6
20
.)
2.3
2.)
17
.)
17.4
....
l A.-.
b.=.er
~t I
\'ol
ur·~
e L
..-u
tial
C
o:·l
n:st
io:n
I
,,.•+
iJL
-. I
I !
:~ecciver
I ::
ate
r-ia
l P
.r"E
.3S
C:.:
:"'<!
Ia
tio
1-
ress
:..L.
-e
Sid
e S
ide
~· t;
-J'\
.o
' ::
, '
I :~
~
(Isia
) 'r
"'
(}
.:si
~)
(J.-
si:;
) f:
t:si
:;)
' • :
i/ • c
i
rU-"
?a:.:
e Ccr~
ll. .
. 7
10
/l
!.4.
7 59
.0
57
.2
Ji-5
1
:
.Al-
'l'..
:be
Cc,
..,.
14.4
1
0/1
l4
I.
5!...
.3
';.~2.7
J15
, ...
Al-'l'ub~
Cor
i! 14
.5
lOfl
lA
.5
I 17
·'·
I 17
.5
Jl-5
..U
-1'u
h'}
Cor
e , '
-l0
/1
16.5
71
.1
57 • .
2 Jf
5 -'
-'•·
)
I A
l-'?
:lbe
Cor
e 14
-43
10/1
14
-43
27.4
2
4.3
n
v
2
A • -
T-i!
:i.::
Cor
e F1
aur1
11
14.4
10
/1
14
.4
1.5
1.4
Jl
-5
Coati.~
f i
Al-
Tub
e C
ore
Flo
Lre
1 14
.4
10/:
. 16
.4
66.5
64
.7
Jt-
; I
Co
atin
g
Al-
'l'ub
e C
ore
Flo
ure
l•
14.7
5/
1 14
-7
51.9
44
-J
Jf5
C
oat
in~
Al-
Tu
bc
Cor
e 1.
4.7
10
/l
14-7
1
5.0
1
5.0
Jf
-5
Gla
ss C
oati
n-:
Al-
'I'u
l;e
Cor
e k
l-:i
-'
14.7
1 10
,1
14.7
1 6
1.0
52
.. 9
Jf-5
Coatir~
' A
l-1'
.Jb9
Cor
e tf
ito
n
14.7
10
/1
14.7
5
1.)
I
50
.0
Jf)
C
oat
in~
..U
-Tu
be
Cor
e 7
e!"
lon
14
.5
10/1
1.
4.5
51.~
55.2
JI
5 C
oa
t in
'!
Al-1"'~
Cor
e 14
.5
lOil
14
.5
51.4
55
.1
.ns
J ol;
r1re
than
e r.
cati
.n't
J
v -x
> v tc ^ .rt T3 -H ^i 0) -H n
^< P »A.
^ <5 ^ ^ ^ ^ ^
<
■ o
■H 4> ^^ (0 0) (J S TI •rl
n 'T w b ^-' O
a ri t;
j ♦> n in
■ ^ o u 3 -H ?. o « .-^ > rT >
c 3 « «1 01 -H
•H n n
j
o n n I o '1
r^i
".H O (0 n o
•I" HI
I in n 4) c
9
t O
n m u l I in n 4) F
3 4 4
O H O
c> ITv C^ Q m O O o r\ <^ m ~t r\ r\ -* 444444444
I o
I •H
■u •« x; O ID «)
n n H o o o Ü rj o
B B 6 ■ i i
m 0
o o Ü u
•o
« o o
o 'J
Ü
•H
U t- U U U, U »-.
i o o
B i
iA in iA m
UMM
120
c o o
5
1 hi
M U^ IA 'A 1 l/> IA IA IA IA IA 1 'A "5 1 ■1
-a -5 -s Al -a •^ "»
o C-- (V m ■0 rH to (M C-- O 00 Al
•H T3 -H H
rv m i r IA si «0 IA
M «0 IA IA IA
'- 0 •H
n l- t■ O O -*
n o *•-
rH
O 'A
(A
r-i IA
O rA
i CA
IA
UN ■
SS
fA • IA IA
D
o <d d t.-
t~ jx. r- ir> IA 'A -* -* P ?! c- t--
*3I~ 4 4 4 4 \i 4 4 4 4 4 4 4
L: o o d -H > rH rH rH rH H rH rH rH H iH rH ■^
o « n 3 IA 'S» in rH O
rH <r 3 Si o rH 3 s & §
ent
sure
ia
)
r- t- f^. u> IA IfN -* Al r- Al
r- r^ r-
IfZ ^ 4 4 4 4 4 4 4 4 4 4 4 *~
-a •H L o I
rH rH
1 rH
8 rH H
rH o
8. 1 V t:
rl
|
5
1 o
B o C (H
■
rj rH a) ci ll •» rH IJ 5 "
C 01 ll S X ,^
I) 1- 4) 0)
«1 -H »1 •• I o ■ XI — * X X
■ 11; ri ■P •ü •tJ d 4) ") 4 rH V
1 X
: 0
rH « +> 1 C a
t 1 I
rH E n
! n 1
r—1 1
ll 00 u
11 11| 5 SU lb l c n
M O O O ■p o •H > 3^: rl O O
< X Ü =355 35«! 3&Si
127
X
8
«I
<1
L 0) ^> > 0) no
•H T) -H (U -rt CO Ü '0 M,
C o
•H
3 uT- ^i J -O -H
»A a. "9
(U ^ 3 M
n -H
m-
u
+J t.'^ C 3 C8
•H (o n
IP S 4
I P. 1
-Li -3
A. lA
'S
IA to IA
o «
*
4
4 4 IA
4
0) •o
g'r [3 f
III d 11
X o
III I C «
H O O < s o
il ■ £
(1) J if ■
|Ä
11
128
!*
5§
>
I«
!55
Ij 12
9 11 II
III il rll
c « o o X Ü
c o
r-i
||5
3£5
-3
o r I
c^- IA -* »/> f\i 0 TO N
IA IA i IA 8 IA IA IA IA
4
4
•P c E B
l&
^j E |o|
__ , __
__
J Tub~
AP
J
AP
A
:nbi
ent
Vol
ume
Init
ial
Com
bust
ion,
R
eceiv
er
Pre
ssu
re
Rati
o
fl'C
Ssu
re
Sid
e j
Sid
e W
etti
ng
J
l~'tterial
(pai
g)
\'R/V
c (p
sig
) (p
sig
) (p
sig
) A
gent
Al-
l/8"
Per
fora
ted
14.4
10
/1.
14.4
68
.8
55.5
JP
5 H
oney
com
b
Al-
1/8"
Perf
cra
ted
I
14.7
10
/1.
14.7
59
-7
58.0
JP
5 H
on
eyc
om
b
Al-
1/8"
Perf
ora
ted
14
.7
10/1
. 14
.7
63.8
62
.0
JP5
Hor
.eyc
omb
Ke
l-F
co
atin
g
Al-
1/8
" P
er:t
ol'
l'lt
ed
14.7
10
/1.
14.7
55
·3
53.5
JP
5 H
ooey
eam
b P~ultide
Co
ati
ng
Al-
1/8
" P
er:
tora
ted
1.
4.7
10/1
1.
4.7
59-7
57
-5
JP5
Hon
eyco
mb
Re
da
r V
ito
n C
oat
ing
-
I
I I
o o
H a <
■ iA IA IA lA lA lA 1. il % % I 1 % 1
S§ 0) < ?t Is ■»
h 41 ^-N 00 OJ CO o tA IA r»" > O Ü0 • • • • • •
CU ^H T* "H
S t) CO a
ON 1A 9 5 ^R lA
IA 9 D >_ tc
c o
■H ^^ +J 4) Ü0 IA OJ lA m CO IA o a. m -o -H • • • • • •
^ 3 -H in ^ .Q w a t- «s 0! s M £
S ^- o u
■ ^ a) 3 'S
OJ -H ü) «H ir\ lA t^ N t- r- t- ^ ,J U5 Ü) • • • • 9 «4 M A
EH C C^ j- -» ^t -t ■t * .* r* r^ ■H rH rH A H
N a,
i o o rH r-l ^1 iH rH rH rH 3 -^t > ^ "•^«^ -^ >^ ■^ "^ r-l +J -^ o o o o o o o" o m ,r. H r-( r^ rH H rH rH > K >
V
c 3 « lA LA N r^ r- 1^ t>- »1 Ul -H ■ ■ • • • • •H CO U) -» -* ■r -s -* -* * x> v o. H r-l rH i-t rH rH rH
MES
(U 1) 0) i) ■ D T3 ■Ö | > > > •H ■H 1 a) I 0) V^ G r« DC •» »3 l ll 3 to (U V V b. v in 3 fc t c c C 1 B >> 0 i o >. •H f-f •rt r^ S r-t K rH K H
•H CO CO (0 4) tn o 4) 0 h M C i X r Pu 1 - ■ - ■ ^O \o
+> CO P 00 ^ CO j> CO £> rH Ä rH n
1 ^ B 51 a >» a ro 0 M
•^ a ■ r-l O rH 0 H" 0 00 rH O 60 no 0 M rH
t) ■ ü C O C ill O X >> X >> X >>-H X >>-H X >»-H ■ li 11 0) 1) •»->
SCO) 0 o o
Jl 4) +1 a c d o o ö o o 3 11? K
2 Ä 5 as ss as o s ac o a ae o z as o r-
130
'-·-
----
---·
···-
·---
--·-
-r--
·-l'
-..m
_b_i
.-e-
-n·t
---·
-··-
---·
-·--
~~-~
~~--
--~-
-~~~
al
Press~
,.;..r
e R
ati
o
I P
res5
ul"
e
~---2-
5-p-~-
1-t;-~
~-i-al
-----i
:---:-
:-:-~-
)---rl
~:~:c
I ::~
::) t·.
Po
tass
ium
I~U
de
I
. C
oat
ing
II.
25 p
pi F
oam
P
ota
ssiu
m I
od
ide
I C
oati
ng
25 p
pi F
oam
P
ota
ssiu
m B
rom
ide
Co
att.
ng
·
25 p
pi
Foa
m
Po~sium B
rOid
de
Co
ati
ng
c) p
pi
Foa
m
Gla
ss R
esi
n
25 :
pp
i F
oam
G
lass R
esin
25 p
pi
Foam
G
lass
Resi
n
14
.4
25 p
pi
Fo
ul
Cla
ss R
esi
n
i 1
4.5
25 ;
pi
Foam
K
e1
-F C
oa
tin
g
25 p
p1 F
oam
K
el-F
Co
atin
g
I
14.5
10
/1
16
.58
10
/1
14
.)8
t
10
/1
16
.58
10
/1
14
.7
10
/1
16
.4
5/1
1/1
5/1
1/1
.O.P
C
om
bu
stio
n
Sid~
(psi
g)
13.5
71
.0
12
.4
62.0
2.5
4.5
4.4
62.0
AP
R
ecei
.. er
Sid
e
(psi
g)
1*.6
13
.7
2.2
4.8
4.2
l
\letti
ng I
Age
nt
i
JP~ I
~5 I
JP5
JP5
JP5
.JP5
JP5
JP5
JP5
JP5
.... w
i\)
r w
ett
ing
A
gen
t Y~t
e~.~--
---T-~
;;~pe
--~-i~
~rr---
t-,~~:
-r,l··
c~iE~:
;:~-r?
:~b~r
'----
----
----
----
----
4---
----
----
---+
----
----
. ----~~------------~--------~
r !--
I I
! 25
;::::
: :!.
-F'
oar:J
l:
... 7
lC
/l
h. 7
i H
o :;
at a
4 B
urn
'rn
ru
I ;a
co
ati
ng
l
j
25
pp
i -
Fo?..
!!l
·j 1
4.4
4
10
/1
16
.44
P
ress
ure
in
I:::X
:ces
s o
f 3
5 p
si
!11
Co
ate
d.
25
pp
i -
Foam
U
i C
oat
ed
i I
14
.44
1
0/1
1
9.4
4
Pre
ssu
re i
n ~ces5
of
35 p
si
I I I
I ~-
----
----
·~'~
----
--~-
_._
L_
. __
j ___
___ _
JF5
JP5
JP5
10
VVC 5
i 1 111 0 Foam 25 ppi HjO
1 1 A Foam 25 ppi JP 5 1 1 1 I
1 \ \
\ A"Th 'oreticdl Curve
\ /
V A ^^^ A
2 psig Initial Pressure
,
1 6 7 8 9 10
Pc,Pl
FIGURE 62 PRESSURE RATIO ws RELIEF TO COMBUSTION VOLUME RATIO
PLEXIGLAS TUBE TESTS
GP71 1605 4
133
VR/VC 5
^H-l
Q Fire Extinguisher Foam (Dry)
H.^M'.M
1 (7A
\ rJt*° etical Curve
v. *
. -
1 R 9 10
Pc'P1
FIGURE 63 PRESSURE RATIO vs RELIEF TO COMBUSTION VOLUME RATIO
PLEXIGLAS TUBE TESTS GP71 IliOb 1
13^
■—■ii — i .
10
9
8
7
6
VVC 5
4
3
2
1
0
|-TÄ>
I O Scotch Bnte (H20)
A Scotch Brite (JP 5) l ill
T
\ r11" ■MM Curve
\ /
V ^J ^<*.
8 9 10
Pc/Pl
FIGURE 64 PRESSURE RATIO vs RELIEF TO COMBUSTION VOLUME RATIO
PLEXIGLAS TUBE TESTS GP71 1605 I
135
-
rf*>
Vvc
I — O 25 ppi Foam Viton Co.ifmq (JP 5)
O 25 ppi Foam GIJS'. Coating (JP 5)
D 25 ppi Foam Cu CnalM) (Dry) V 25 ppi Foam Cu Coaled (JP 5)
_ A 25 ppi Foam Kel F Coalpri (JP 5)
Q 25 ppi Foam Polysultide Coaled (JP 51 ^— < -L-
FIGURE 65 PRESSURE RATIO vs RELIEF TO COMBUSTION
7 8 9 10
VOLUME RATIO
PLEXIGLAS TUBE TESTS
136
UP 71 I i.Of, }
■MM
SECTIOU VI
I'HAGE III
1 .0 J\iJALYSI~~ OF PROGnAl·l RZSULTS
1.1 Gcnernl
The ob,~ect of Phnr.e III of the progrrun wnc clcfin~d o.a rl'lt!J nnaly:;lr; of :•ttns<.!B I end II results. The analysis \.IO.S to lead to nn engine•:ring cctil!'.at! of the future potential of the concepts und tcchn iquen dev·~lopari and an •}vnlunt.ion of th·~ t:I:J.terio..la invest1e;ntcd with respect to th•.:ir :J.nr.tc nrrnat~.n · d'fe~t:i'.rcnc~.;s. In order to accomplish thts tnsk, an e.nnlytico.l mod'!l of th(! ex::~losion snppr·::G~ion system we.G established nnd ll compu\...Jr rc!e:rr.!;;s.!.on <:n:U.;::;::: 0f 'th·~ :l::ttn was conducted.
The sJmrlc ... t model of a relieved explosion which gimu.l:1ted thl! t0cta pe!"formcd is a s~ngle-cell configuration us shown in Figur~~ 66-,\. ~r: th!:> !:lodcl Vc i::; the combuntion volume nnd Vf the arrestor volume. 711~ r:::!.•.!f volt<me in this case is supplied by the arrestor materinl only. If, h.:-·..z.;v::r, th0 •l2pth of the O.lTesto:r m:lteria.l is r;reo.ter th:m that n<:!e;ied to ·~li~lnnL fl<'f:;'.! p·orO.f"/lt:ton, then voiding behind the arrestor matcr'!.al i:~ p(');;~!bl:: {'.:> sho·,m :!.n :·'ir.~ur·~ i)() .... n. The relief volu!ne no'W i'1 '.'r plun ·:r '"'ith b·t< i~;tlly n.:. r.n~mrJ;t~ ::1 th·.: model parrun•Jters.
l ( ,',)
FIG. 66
( il)
v r
Furth•1r expn.n:;ion of the cornplexity of the rnorl~l to cimulr.tt~ a ;..ult !c•f'll ·.:!tw t.'J.nk, :.Jould b':) to place n wall behind the nrrcstot• matnri:ll 'Wh~ch ·J.~.-.·.-··~ :'~·~· ·:,urc COTI'JnUnJen.tion, bUt wtth l.l VO.riable dcr,rce Of reGtr.i.ct~On ar. :;h:y.;n ::1 I·'igur': 67-!, .~ n.
(A) FIG. 6'7
MULTI CBIJ:, tiOtl~,
137
Vr
(P)
1'11 1:1 mo(tel configurnt.ion Jo !JOt:cntinlly u two-ctJll modal which enn net aa thtJ <>inf!,.\1:' cell C·.1g. a;/',~~- B) whC~1 t,hf! X'f!BtrictLon is zero or infinity. 'I'he molf(tl c~~·, b1~ further ·~xpm:de'l to n multict.'ll conf!curntion by ntld1nc rnorc unite, cimulntl!ll'; ~t wing tn.nk with :1 munbcr of cells formed by the porforntcd ribn nnd 5por:s.
,\ll fc<ttr modt!l:• cnn be nnnly1.cd by uo1ng the mo<lel .i.n P'lr,ur(: 67~'5 within the ;;tnt11d limJtn :: :tn.;u t'.w other modvl conf1gur11t 1onn nr11 !'Jlmply npcc1ul CflO!..'r! -Jf Figure 67 -B.
·:'lJ'! nnn.lysi.t> ,J!' the mod .. !l twa:::: tho :::impl!: r, '1 1 T, relntionchipo before nnd r.ft.:r the combuatlon rroce~HJ while n.llowin~ frc•! •!XpMnion to occur. Although .~t.:ti;i•:: e.J.Uil:ibrinm condl.tionr. nrc nssumed, the :.mulya!:::: is valid for the non!··.:.:tr~~~t·.:J r~onnr;m·t~ttonc, nn(t by lntro(iucinc; nn experlrr.·~ntnlly determined ~:·t·':·nc:•: rc~;tr1etion fader, tho mo.ximum dyno.mic r.re:Hl\::~IJ in the combustion '!Ol 1U:':·-' ~.~w be correlnt'.:!d,
··~;,c:r· .:n th•: c.Ynbur:tlon c!' hydrocnrbons ·.-~ith r.J.lr, little or no chanc:! occur~ Ln t;~'-' :n··:rar~-· "ol·~'~\LL'1r •,w t[3ht or totnl mole:.: of ~fW pr~.:sent, the following r·::l~:tlon~hip is a~~:ti.L"neJ to be trut~.
p1 '/1/'i], ~-
~-'tl:rth·~r, :;tnce tht! ~n..x.lrr.um rntio of (T2/'1'1) i::; .::ir,ht fer most hydrocarbon/ ;'li,r ::toi(!hlor:.etrtc: :~:xtur-·.::r> of inter·~~Jt and lG ~n·~·.!pcndcnt of all other mo:iel pnrnmeter:~, it ts c:on~;~·to::!r·~·d a con:1tnnt, K, the n.dinbo.tic cx:pnnsion fnctor in the n.nnl;n:.~•. Th:1s tl:·.: ::ombnstlon ·.·occnu, can be wr!.tten ns
:"'fl . . •• • t p 'p "' ;.: · '· l '1 "' 1 ;) ':::' or · 21 1 ··
':'1:1·~ nbov .. ~ ~~-iuntiu:: i:: G:.ttiGfac_tory for unrelieved ex:plo.Q..ions; hO\olt.•V~~r,
•,.;i1'?r1 free CXJ::Ct..'1Sion i:::; (t].lO',.IC\l nnd no.me !'I'Opagntion i:::; lirni ted to the UVn.ilnbl:.! ·r;::.bu.;tion volwnl!, :;~verr...l nttunua.ting effcctr; taY.!.! pla.c~. FirGt ao:r;c of the ~,v:t a:,blc unr~rtctGrl and pn.rtinlly rco.ctert (qucnchf!d) combustible ca:;cr. nrc tr~t:1srorted throur;h tho :1rrestor mnte:rinl into the relief volume. Con:ocquently, not ,il of th~ orlginnlly a.vnilnble co:nbuotlon voltL~t~ is reacted, Concurrently, h::cnus·::: of rn~G:> trun::;fer into the relief volume 1 the corrc::;pond · nc relief prc:l- ·· · :;urc increases lirnitine; subse-1wmt mas:J tran!:fcr. The restriction plntc behind the ':U'r~J:::;tor ( fora:r) rein forces this lflttor c f!'cct resul tine in hi{l)ler effective relief pressures nnd thereby higher combustion cham'!Jer over-pressures.
Returning to th·~ mod•.;!l nnd introducinG Vx ( ;.'lc;, 68) o.s thn.t porticn of t.~!U combu~.,;tlon vol.tlf.ic twtuo.lly rco.ctcct and which when fully c:<pnnrtE"d np;ninot r.:.~: relief bn~k-prc:;::nrr'1 cxpondn just to the nrrentor fnce, we obtnin thC! f'olJ.o•..,.ln;; equ':1.tlon: -
hl Vc
FIG. 68 HULTICE!L t·~ODSL
Vr J
(3)
'~here P1 is the initial pressure, and n is the reversible polytropic exponent
>;hert~ rvn = constclnt. Using t;ne relationohip PV"•C and summing the initi&l. PV" ,'lroducts to final equUibrll.lm, r"sulta in tbe following I'elationehips:
P1KVx0
+ Plvcn - P1vxn + P1vr" + P1vrn • P2vt", and (4)
P ,, n L n ,, n p V n p V n c " c ·, ' c " 1' + r r 1111 2 t { 5)
Ecl'_w.tin~r Eqa. (4) and (5) and substituting Eq. (3) to eliminate Vx yields:
Pc K(An + B0 + 1)~
pl {KA0 + KBn + ~) (6)
\!here A "" Vr/Vc and B a Vr/Vc. The )\ term io the average restriction value dcfinc::J as the rrv io of Pc/Pr· It can be seen from Eq. (6) that as the relief volume goes to ze:·o, A, goes to zero and the eq_uation reduces to that of a slnp;Le cell 'Ji th ;·c orifice restriction fnctor, and (Pc • P2):
K(Bn + l) "" (KBn + l) (7)
LU:e11ise, 8f3 B e.pproaches zero the equation reduces to
Pc P,, "' t:.
::t K, PJ. n
~ ' ·'·
the mruc!.mum exploe;ton over-pressure given in Eq. (2).
Since the p:rc·cess modeled is no't constant pr~t~sure, volume or temperature "'<.he Vli!.ue of n m.riea. The value used in correlating the data herein was n • l ( looth,:!"mal).
1. 2.1 fhas~ I D~~:~-~ nnd Model Analysis
Trw Ph,.toe ! i•tt.n gcnero.llJ' followed the model analytih. Divergence from ti'l~ a.rwlyticllt mcul•:!l rHd occur 'With respect to initial pressure yielding a ·l!rttnc~t. curve for esch initial p:reosure. This effect was more pronounced ror t.h·~ hollow body C".,nfJgurationa. For thio reason, the initial. preuure waa 1 n·.:lurled, Jn the ':·•:nputer r.nnlynio o.~ one of the variables. It vas al.[lo ·1otteed th1:Ti.t div"":·;Jcnce from tho model anal.ynis occurred at Vr/Vo ratioa 'below L ). T~H! cxplttn·,·~ !.on of thin divergence is felt to be the tact that h18bor prenHLI!''!!£ rer:ntlthfl, f'ro~n t.he combustion cause deeper penetration ot tber name :\·on"'. J :rt, the I'1Hlll theroby increasing 'the eom'buation volume.
139
I
'..:ho:rtl rrog:reno:!vo burning occurred which wno only in the mult1coU or multi-votd confie;urntloM atrone; deviation from the model o.nol.yota \#O.s intlico.wd. ~ t 1.1ns f<.-,und by ot.ur1y of the ooclllogrnph trace:: a however that tho initial. rrr•:sr.ure pen}{ cor~·ol~~ted well with the model Md thn.t it the preosure a.t \#hieh ~:lw follov:l.rl(~ pi:Ht},o :'ltnrt.ed wns uaed uo ru1 initin.l prccnure 1 thiD data, too, :~orreltJ.tt::<l 'Well w J.th "'.:!t•.! nnnlyticlll model. Dome small diocrcpe.nclos with thio r;rocedurt~ were not·:!d ~!1 cnncs where the c~'ecrnAJ. volume arouurl the hollow bodios ·::1c; f1::~nll. !n thin clwe, the G·i_Uece;ec efn:ct of cliffcrentin.l burnina and thore-
p:r<:~osurc pumping Ol" circulation of the combustibie mixture from within the hrJllo·.,r bf.:'trt:L!G wns conjectured. Gtudy of high-opc,;,d motion pictures ot' these ronfi[',urntion::~ would indico.te thin posr;ibllity. Hhon the external voi<l ia lnrc': nnd hn•; g,)O'i fl.cu~~~ pnths, then the cxternn.l it:)I'lition is rnpid and wtit'orm, ;.;:i.ving b•~tt~~r resuJ. ts. Thin ca.n be seen by compnring the ocvcn end one-half irlch d iwneter cyl !ndcr date.. The ohorte:r hemiephericnl head cylinders with t:w int~rconnr~cted extel"nnl voids out perf'orrood the long :nat end cylinders ' ':i1ich hnd a number of small independent external voids. 'J.lhc external ignition n thf: l:).tter case lo.cted much longer allowing ~or greater ci!"cula.M.on of'
unbm·ncc~ 3n.sen a11d. therefore more combustible vo.po:rc were burned.
l • 2.:? P!!nsc I Conreute r Rc gre o s ion Anlllls is
rJ\m methods of e':l.untinG the Phase I 9ystcm parrunetera \tere used in nn ::':t.tempt to get further insight into the explosion supprcsnlon syntem. Tile firGt n.nal.ysls e,1unted prensure-rioe with the various nyatem parameters in nn nttempt tn conclo.te the t'!nt.:l with the .model analysio par0111etcrs- !H.nety-nine t•.;5t po:Lnto including o.ll initioJ. preonures, o.nd fuaelage as well as wins tDxl.k test do.tn were run. T'ne equation derived from this o.nalysis was as follows:
where:
tr = .00357 (P1)3•095
Vr. )"2973 (Vr y72'J7 <vc Vc
llP = Over pr~ssure (psid) ?1= Initi&J. prcasu:re (poio.)
Vr= 'rotnl volume - conioustion volwnc ( in3) 'If"" Ar:re:;tor volwnc (in3) Vc= Combu:Jtion volume (in3)
It cn.n be sc1~n frr-;m thin cmpcrico..l rel().tic,nship thnt initial .prcusurc f,,llo'.1•!d by the ;·ntioc of forun volwnc to combt~stion volume and rel1of volume to conbu~ti.on voJ.1.tme rn•c? the order of rclcvcnce of tho test pnrrunctcro incltrl•lrt. 'Ji1'! ~:'=pnn~nt of Lhc inltittl preoauro ohown why nomnJ.:tzntion o:r tho finnl. to .Jr.lti'l.L presn1tre in the model nnn.lyoio didn't fit the dntn yielding on 'n(lcp(m(!!mt ;·urv~': for each initial prerHilU'Q.
'ihc ~;cctmd computer nnu..1.ysio of tho dntn equo.ted forun volume to tho other •··!ot pl<.\·nmetr•r.n. 1~1 thio nnn~yaio oighty .. eight points were rwt again including
140
~
l
I
:~i
-1 'l
1
! '~ :j ,, ~ :l :-Jl
,·: :I ~ r~
' ;;; ,.
nll l:litiul prr~nnux•cs and tank confjguratio;,c, The resulting empirical Cto~Uiltio;; ~.~~;t~J no followr~:
,!t88 (Vc) .150 (Vr) ·574 (Pl) .277
~) .0~5
(Vr) = Totol volum':! - combustion volume (1n3);104
(Vr) "' Foam volume (in3);104
L\1' = Ove'l:'Pressure (psid) F1 == Idtia.l pr~ssure (psia.)
'I'he reur:o:: for the relief volume i:1crca.si:1g '-'ith the foam •tolume i--: thl~ equation is that a portio:-. of the l"~lief volume in each oyst:er.t co::figuratio:l is fo::un volume. ':!:he expor:ents of the variables o.re a~: indica.tior or their L. importar:ce '-'ith :respect to fonm volume. The exp0!1ents '-'ill not chru:ge if the 10 fo.ctors n.re removed. ns this procedure will only change the constant. The reaso:. for lntorduc:i!lg this factor was to accommodate the computer '-'hieh o:tl.y fits the ::;ignificnnt numbers.
1.2. 3 Pha:::Je ii ::~to. Annl.Y;sis
The data generated in this portion of the program was an attempt to evnluate the flrune a.rrent.tnr, effectiveness of 1'1 variety of materials and. configuratior.s with respect to their thr~rmAl 1 physical and chetnical properties. Their ~ffectiveness cf.l.n only be measured with respect to system pressure rise nnd re-1uired thi.ckness of mnte:rinl. to eliminate flnme propagation. For those cases '-'here flan:e ;Jropagation through the arrestor mo.teria.l didn't occur the data correlo.teri well 'r!ith the model nns.lysia. The divergence from the model could be taken as nn, i~dicntion of the arrestor effectiveness. This is a weak. effect, however, whereen the rc~uired arrestor thickness '-'ith respect to combustion volume gives the st:ronger delineation. Unfortu.."lntely insufficient data ""ns cenernted '-lith respect to th l s lnnt parm:K!tc:r o.nd. there fore th iG evaluation cannot be made.
Computer n.no..lysi"l of the pressure rise dntn '-'ith respect to the material properties was mariE:. 'I'he best f:lt log-log regresc ion annlysis of the data yi•!lrkd th·~ follmd.ng e-s,untion 1 from th~ 52 test conditions entered •
• 054 (Vc) .9h83 (Pl)l.076 (K) .1762 (f) .2536 (A) .2147
(C )•2164 p
141
I
AP ::. Ovcrprcr;nu:ru (prJtd) '\ "' lultinl prcoDure (poic.)
'..1 c "' Co:r.buntlo:. volume (ir)) /100
rn .. ::.~:c~ cnpn.dty of mnt~:r!·tl (n~·n/ f/-~)
!'~ '~ 'i'h<:!md conductivity ( n~·u- IN/ { °F .. ft~
~ "" :)e!:r.ity - (#jrt3) ,\ .. Gurfnce nrco. (ft2/"1) ; 103
t • Thick1msa (2 inches conatnnt)
The datn bput for thin e~p!.ricnl- u~1nly:;t~ 1~1clu:ieci r. tc;~._:to O'!C Vf'r:J~til')r . . ~'"·r<bU!lt :!:.~1•: 'lolw~ 11h ~J.·: )101-:l int~ th•• r~:l ir:?f volu~~ ~onoto.nt. 1:. c.ct·l it ion ··~
th~:.:, tl:r: other :;;r~·t;~r, vr.rlnblt~ of lnitirJ. prt':i3ure \mr. varied by 30 · .• :cv>Jrn.l ~ont.tn~~ n.r.d. thr:-!c bunlc r:~~t•;!rialr; n:;ri. their thcrmc.:phyr.jcrtl prO:"~P.rttes 'Were :·l;.o inclurl.c:d. '2hc n:1nlys:L::: reo;ultcd 1~1 n curv1! f1t •:lthin 10 ~ of thn ~xperi, ... ·: tr,l vnlue s.
It ir; uppo.re:·t fro~ thi:> empirical rclotion:;h.!p thRt by increnr-1~~ thl! he~t (:':"?-~~city of th~· :u-r'~stor ·..:hil; red.ucine; itz thermnl. conductivity n:1d d.~nsity, ~m}~rOV(!·i. nrr~utor cffcct:!.V'..!nnsr. occurr:. It r;houl.:i h~: note:\ l\dre 1 however, that '.ncr,~:-~:;LlC th•~S'! pr,rr.m·.!t :rc.;, tht;:!.r off.:ct on i:JJ~ r\ecrl1ncna du~ to th~~ fro.ctio:~al
~x:'O'.~rts. ':'hin fits 'W:~ll with the foct 'thnt water wP.t n.nd ,JP-5 "o~et Rrr.,r.tcrs ;):::rf0rm sorcl•J\-Ih:-Jt b·.:ttnr thun the dry specimen::. ;Usc '"''.!tt:.:.~,~ thP. spccim~~;
· .. ·cttld hcV·..! " t'.!!;:!.c:~cy tn rt~,iucw the microourfnct! or01' . .Lnmrovinr. th~ perfor:J.'' '.cc. !i:.J'.-ICVcr, by comp·1.rb1~ thn rl11tc. of tho.1r:: ::pec1m·~·,r: which fnilerl a~ •. i. :1·.er2b:,c •,;er·.~ ::ot ir~,~lu"i•..:,1 i:: tho n:1o.lys in, th" nnhlyn1 G of th!! ~urfrtc·~ ::rl~,.. .... · ;--r~~"::t ~.::. :Jorn,;\,·!:rtt qu(!rtlo;~~blt:.
. ) 1 ...... ~ ......
::::~: Jata r~·)r.:;r:-:t'!-i l:1 thin proGram inrUr.nt·~n th..,t tho;! t.?Xi":\.~')n!:'l" :-:uri)rc:~:;:!_o., ·:_:.;t•::-:-: c:o~~r:r'\J.r.".t!.f:r-. :t1; .tlcv.v11. 'b:· tho :-i.~" ·; .. ·l t:r:v: ::;f fncl tnnt: b~l:~r ~on-.' ~. ·~·~· '. ':'h·~ t•,Jn ty:·::: of r.o·1fiL.;urnt:ton r~r·· nlr,rl•.! c·~ll nr'i. nulticell <:'r ::n"..tl~·rol<t ~:r~:;t,_ms. L"nis ~~sult ls 11U:l.tn cv:!·i'.!nt b:: lnr.pectinc; the .tntP. :r·:~u:-~+~:"1 in ·~:::1.bl0 'll.
~ .. ~ .1.1 Rl'oroduced from b~o~t available copy,
;~11:nln.g€? tn-1'-':r\ (~<1n·:r;1lli fall i:1to the :; !nrl• .. !~~~11 clr:u.iA 1 h~··:•·!w'r, 'l~ th~
·::·.:11 b•:L'0!""'.~0 l~tT(I;.:!l' 1 th•J !'·!-!lli!'t'l f'OO.l'l'l VCl\l:r.•. t~ !~j'!lbu:"'.tion V!)lU.'n•: i!'~rl~tl\•1C to ··, rol:.t .,...h,)r•~ 1111'1 t.i ~v~1 i cor.fir;urntionn p...:rfcrm b~ttcr than n!nclc <":·-11. con:'icur~t·
t l :):·,::;. In thi::; pr,;l3rO..':\ ,,·herl: the :;imulo.tutl fu:::elt.lt~L.l t!lnk "o~nA of lCJ ~:nllo•HJ
~:; :-•(:!t:r, U1•: 10', vol.ki lined w~Lll confl('(nrnt~cn pqrfo:rm~..d th·.~ b•1r:t ':IVC!" th•' r•1't[~/l oro, 2 nnl. ) pclv, initinl tu.nk prcrH:.urc.~. ?or thin oonflr,uration, the t,:~.01r. voLl voluml] perr.~ntc. for 10 plliC ov~rprc:wurl.) ,.,.crt:, 52, 46 on;\ 3fl percent r•·:m::cl.ivcly. 'i'htl l.5-lr.c:h dinmctar cylindt:rn fu."\~ttoncd orfuet.lvoly up to 58· vo.L\ vnhtr.:c, but .,.,~l"G llmitc:,.i to 0 puia initi!\l pr(Jusuro.
l42
The lined wn.U and egg crnte configurations o.ppeo.r to be tho most otf'ic1ent explosion suppre~sion eyatems configurationo for multieell wing tonke. Asain ns t.hc combuntion volume gets larger the multivo:td (egg crate) configuration cut-pcrf'onnn the s j.n(l,lc void (lined. woJ.l.) confip;uration. Voiding percentages ro~ the (300-gnllon) siX·cell simulated wing tank oxoeoded 90% ror both the egl)-crnte nnd l:l.ned wall configurations. When the cell stze was doubled, (the three-c(.)J.l, 300 ga.llor1 wing tank) the preferred :foom configuration becasm thE' egg crate ntyle providing up .to 58;t voiding for the lO _ _paig overpreosure critf:rion. 'rhe lined. wall cont'igurn.t:l.on wns only good to 43,; voiding as experienced in the single cell funelage tank. Hnd the system been tested with voided lined wol.ls, the results might have bMn q,uite different, with up to 55.~ or larger voiding realized as indicated in the fuselage tanks with a simUnr foam configuration.
1.3.1.3 Advanced l,io.teriol.s
Substitution of the 3f,t felt or Scott's fire extinguishing foam evaluated in the Pha:;e II portion of this program might well change the allowable void percentage for n~J. of the configurations tested. Both of these materials have superior flrune arresting properties at elevated initial pressures as indicated by the Phase II results. Their increased voiding, however, ~ould have to make up for t~eir grenter oJ.splncP.ment and absorption penalties; as discussed below, in o1·der to compete \4e ig.i.twise with the 25 ppi foam systems.
1.3.2 Fon.m S;rstem re~~
~he r~suJ.ts of the gross voided faa~ explosion suppression system configurati.on test!> CP-rt now be converted into aircraft penalties. The penalties of concern ore ra.t1ge pcma.lties and gross tnke-off weight penalties.
L 3.2.1 ~___2oided Foam - nange Penaltiea
R~rtge pennltie~ are of greater concern in the case of a retrofit system :>hlce ::!.n ne"' nirc:ro.ft design, the direct fuel loss can be offset in the desibn by increasing the tank size. '!hE> retrofit range loss for the first approxima· tion is BL"liply the volume percent penalty, This is the sum of the percent fuel c:J.~plnccmcnt rmd percent fuel o.bcorption. Figure 69 shows the relationship of 10) 15 and 25 ppi forun nnd allowable voiding for this penalty. The values of fuel d:tr,placcm:::nt and ful!l retention req,uir'2d to plot volume and weight penalti,~s for the ?5 ppi foum were obtained in Phase II and are given in Table 19. Do:tn. for the 10 and i5 ppi foam material curves shown in Figure 69 were obtained. from previously suppl i.cd Air Force test reports. It can·. be seen that the increonc in voidtng permitted by decrensed llOre oize over ohadows the veisht anrl volu.m·~ pcnnJ.t:!.co incurred by mnteriol absorption and diaplacemont.
1. 3.2 .2 ~roG!J Voided Foom - Gros_s Take-off Weight Penalties
The gross tnkc-off weight penalty for n. new design aircraft is not shown, 'but Jr; nJ.mply the -weif".ht pennl.ty M given in Figure 69 times the g:rovth to.ctor for i:h"' typic~:tl fle;h+.er or eargo n.ircrnrt. '·1eight penalty 11 tho re-r~ent by WC1(Y)1t fomn minun the percent by weight ruel displaced, Since both or these
factor~; for the smn.llcr pore n lze J.ow-dcncity fonr!::J rLra c~w,J. anrt J..!o::~ thnn ttl'J 10 ppi cto.ndnrd. hifjl-dcnnity fou:n the n.dvnntace of ::'5 ppi fCJn.r:t n._.~,t not b'.! 'mhonccd hy the l)ro::JG voiding tcchni~ues (Figure 69). Ho.,..'evor
1 the voitiine
clo0s improve the ctnturc of th·~ syotel'l mr.tkinB it \Wi('.ht compct1:tivo •,11th oth~r LnertinR systems.
~-io far the mntcrin.ls phynicn.l weig.ht pen~ltic:l nro o..ll th:1.t hn.vc b~·:n r.cc·.!~>G·Jci. In thu ~~ur.c of c;roo~J voidin~ 1 o.ttn.chncnt of th(! fco.r.1 or th.:: O.BS1!r.bl~ni~ of the ~·o;:un requi:rcr. in some cnncc mecho.nicol fnr.tt!ncrn or ndhcc1ve-bon11ne •r:hich menifost ttwmcelvcs in nn ndrHtiona.l weight :rennl:ty. i!!xpcrience has sho'"!: thnt bondinr, of thir. r:mteria.l nearly doubles the forun weight penalty. ConcreIJllcntly, adhesive bon1ing should be nvo:i.ded if pot:nlble. In th·~ ens': of the hollO'..: bo<Ues 1 thi~; em be nccompllshcrl by l:'inl,.tnr: th·J:x .t'omr. !.n tvc nr r.lO!:"':
t·.;L~~;~~opinc po.rts. Compression p.1cl.o:.inr; of th':·~·) bcdinn ln tb-.J cubJ . .::et tnnk :1 iJ!llnnt2::: any further fo.::;tcninr re :1uir(mcr.t. In the cn.~0 cf cr,;j cro.t lng l)r .Llrv~(1 \-Jttlls, lnterloc1,~1.ng fo~un noner.1blit~~ h;. .. v':! pr'C"."::n r~nt1n-r:-cto-r:,r, ·.T!1·:r~ ::-,c~ch:mic::tl fnst(•n,~r::; rt1"C" rt:!-:i_Uirert, 1'1cinc: h~·'3 b: .... •n fc11:1rl ]1\·:t,.. ':'nt1:-f::~!tnry ·"·ith only n. 10 to 30:' nd.ditionnJ. ~,·clr;ht pennJ.ty,
;\n.~t.~·~r n.Gp(~ct cf the gro~s •roll·.;:l f<J:lr.l cynt.':!:n which mu::t bP- fl:ilr•..;;;r,c•i it:; c ffc ct upon fuel system operation::;. ·.ath et l:!.ttln cnr:lnct~rin~, M~~t
()rob:!.r~;-;s cn.n be hrcnclJ.ed. l'lumhlnr;, r,a.ugin1~ c:.nd fll,1l tr.:m'1fr~r 'lr•.: the pr!!",r.r:' ::ucl c:.y.~~·2m fu11ction:'; of c:onct~rn. ·~·l~:! hcllow bcvly r;ros!:: ._._.,,d~nr: cC"ne •:·t r>·!:,rt: 'l uit'-: ~fell :in thnt rrope r slz. i n(j r-".:i ·~vntro~.lcd p~c;~lng r•rav1der. vo; d:; b0.L:• ··~1 +.h~ ·r~ul:ies thernr;·2lv•::s .'md thf! tnnl: ·.·1:-tll:; t:.•l•.quntc fer gnu;~:.i.ng prober., pmr.~ .;_nlct::: or v-ent v~1.lvin[~. 7his Js :~:.~o tr,.Jc fc1· th~~~ 1 ~nr-(1 wn.ll C" .. nd :·~<t~ "!T'~-.t·"' <.:rJnJ'·Jp;urntlon ln th~~ vr!rtic~Ll. v.!.-:-~1. '.,'here.: 1.-.'nll plumbJn~ cnmes thru h. hol·.~ ;n the fonn, proper sc,.J.lnr.; 'will ~lo•,.,r t.hL~ r.:,•ct·~~n to fu.nction
1u'itc well.
l. 3 . 3 .1 F'uc 1 Plow
i;er:Lnl refuel1np: repreGentc th(~ hlc;h rnte of fu,:l flow 1v·~n by ~~.l:r~rn t't fuel oyntemn. !"iinct:! the n.bsolutc pr(~G::~urc drop j:: lt1a: for thd r:rc'"' ,,.,i'i~o rorun eonfit;urntion than the fully p::~.cketl :Jyr.tt~mc \olhich nrc ~·.,:·:~·ntflbh:
1 JJ."l ~:r:"l
blemo o.:r.e llnticipatcrl.. Two pr(~ViOilSly performed tr;)!JtS Of r,ro;.:.; VOld·~·'· r~')~r, :LnstalJ.nt1.ono in rr.odern f'ighr~r o.ircrnft !'ucl nyot.emr. b'Htr out. this conclur.ton • .Ln these test, maximum rcftteling flow rrtteo \ole~ impooe,\ wlth ~ln(l wlthnut the r:roc:a voided forun syst(.!m \~ithout dif'f'icultics or .tnm·onr.,.!d rcfuellnr t:troo. ;:., l tJ.ow prcr.oura drop throue;h th:J r.m!Ulcr port.~ dinmc•tt:r foum 1 wldl·~ gr""t<'r, !s 101 t .a problt!m r.J.ncc the prounu.rc l'irop in n. function or th-~ thlnltnr'\r,,., t.hrnurh which the fur:]. mw;t prts!'l. ,:;rosn voidinr, draoticl\lly reduc·!n th~ th:tclmes•• t":tf tht~ 25 ppi foum result. inc; in n lc::;r.cr pr<!souro .\rop th1\11 thnt of the M\l~l~ 1 h :n":·~:lnrG~r pore diameter i'nrun.
Forun rl:l nt".t'lU•lt!cn nntl plru:~.:!mnnt r.hculd b~ ,icao ir,M.-i to !M\ll'\) thnt tl~·~ p·. · " ;~elJcf volume! to cor.tbuotion volume rt'1tiJ ia mrtintolnc,! nt. ~1l'!. fuol lev·:L.>.
1.44
I
ThJ::; 1;; not u Pl'obJ.mn in n.ny of the ~Jngle-ceU d.eoisn configurationo, however in the cnr.~ of the mLtt tJ ~cell lined wall coufiguration, conSiderations or t.hie :tcquirement ncedn nttcnt:con. The multi-cell system utilizeo the interconrnunieatin~ OpC'n:l.ngs between cella for preasure relief of the ignited ceU. Fuel levela li:nlt nnd ln oom.:.: cnncs toto.l.ly eliminate this commtUiication, l'ed.Ucing the syatem effecttvely to n o:tne;lt.:: cell which requires o.dd1tiono.l. relief volume
1 thuo more
fo~t!n. Ily proportioning the forun EJUCh that a greo.ter percentage is located in tho top (normal usnge volur.~e) of the tank, the problem disappears. Ouch u oy~tem ht't!:i been den:i.gned nnd ;,tualified on o.n advanced fighter aircraft.
l. 3. 4 InctoJ.l~!.;i .. ~~ Considcrntion::~
'J.".-~o typen of in::;tnllntion are of concern; new nircrnft and retrofit ~ro~:: volded !'orun systcmn.
l.j.h.l ~w A~.f.t Systcns
1~1 the case of new aircro.ft, erentur dt.'lsir,n latitude and free acr:ena to the tr~aks r~duce the insto..llntion rrcblcru:;. Of the erose: vo!dcd t'onm confi~t1r3-i~i.on:: under ccnnidcrntion 1 the lined wo.l.l and ce;c; cl"o.te multicell (wine ta.n:t~ s::·~:t.::!r:: ::.::: th~ r1ost difficult to inotnll. Isolation of each cell re'luires te:llou:::. d..::nign to insure the :ninimum thickness of forun in aJ.l flame pc.ths and. to obtnin scaln around :UJ. pass th.rouo:h components including bracketry a.nd 1:;rtnc; nut :.:hanncl::;. Fuel tro.nofer holec and tank venting holes between compnrtrr:c:-\t::; :>.r': pr>..rt::.culnrly cliffictllt in tha.t the wing skins form part of the pcrif:r; o:' tht.: holes I:L'"'\,:'1. fnstcnerG o.rc nearly DJ.wa.ya present. These problems are not il~~:~urmou.ntnble' but 1o present nn cnainccrlng dcoien problem. Tho enginoerine e:f'Cort ia w.:!ll .,.,.orth it however, in that 8o to 90,~ voiding is possible as compo:~:! tc th~ l10 to voiding avo.1lo.blc ".~hen uning hollow free bodies.
1.3.4.2 Retrofit Systems
'.:'he retrr;fit cn.::;e 15 ,tuitc different. J\cccoc to to.nks particul.o.rly winc; to.n}~s t;cmcro.lly iJ ::liffi~ul t and in some co.oes rc:;.uirea mnjor o.irero.ft otructur<: re-work. Here for the ::ml':.e of simplicity 1 the hollow body eonfigurR.tion is :pnrticula.rly ·.~·311 suited. The hollow bodies cc.n be compressed to po.sa throucjl small openings anll possibly cv"!n be strung to fncilitRte removol.. Random hcllow body ortentc.tion "hich occurs throue;h thi;:; method of pnckins could. tr.terf'crc •.d th fuel nystcm component a 1 by~ thio should not be n mo.jor problem sin":"e all mccho.nic:o.l cg,uir>•nent in mo.de o.ccerlG:!.ble due to maintenance· requirements. R~movt.tl of the component to plo.ce n nylon or some other mnterinl co.ge around the critic~ 5 ttcr. '\Jill. alleviate the problem. Uniform distribution of the hollow bod leo in n.lJ. cells is not nec:essnry, nevertheless 1 · n good degNe of untformi ty ohould be atri'l(:d for.
I
TAB1~-KXI l.,aee I 8Y't'Jt(J':'I Void Percentages
•> ·M-'"·""'-"'- ---~·. -·-•><'~"' .. "''-~¥"'_"_ .. __ ... _
~ -··-·-------·-F-'u_e_e_lag..,_e_T_aru_as_(_lOO __ «_all_,o_na_) -------r-------i I I
'
, r ,1 nod ;·Jal)
1 :;;~ Void eel IJ . .-1ed ','i'all
2).~ Voided L:tned <!s.ll
c) .li a. (>rll ntlrr'~ . Al :'~JdS
7.5'' m.a. Cyl.lnrl~r:J HI'.J:\i. J·nd~.
F, f, Voi.k'·l Lined .!n.U ': •. rr\ ,
I I
I I I
I I
0 PSIG
41.0%
53.0%
47.'5%
44.5%
58.5%
'i0.5%
50.5%
29.0%
• ··--······- ••• -·· •• J
2 Psm
40.0%
. 52.0.%
44.5%
-u.O% 35.~
45.0%
-. ,.,. ,.
···-~--- .. --f, CeLl. ::tnp; Tank (SO r,a1lons each)
s Psm
37 .C>:'
47.'),
---
23.0~
32.5!£
-
... . ....... r·- ---· ......... -----.. --.. --------,---------f
7. S" ,Jia. Cy linrlP.rs i ~>rJ . :Jids.
80.0%-r
92.0%
'>9.0%
70.0%
! Rnproduced from I best :wail~ble copy.
80.0%-t
B?.O~
40.()%
so.~
79.()%
I ............ -·-. ----------'"-------..,.--1---------"'--------f 3 CeU ~.Jinv: Tank (100 p:allon each)
I. .......... --·-· ---·---·-·---~--------,-·---------...,.-------.....f
I JS-45Z *
J --~~-~---- )0.0~ 1·:,-.rr Crate so.~ . -· "" --·-· ____ ........ ______ __. : rnrrreesive FaDur'" 1 r<f'!SR Ouild Up Df!pcnds on number and aize or Cella.
·--§, 'v ~ --; ::> u. iii .... 0 I-
0 1f1
?: iii c QJ a. ... J:: .5:' OJ
~ ::::. 2
.:,(.
"' I-t! 0
l5
3,(!
'
'\ --·-
\ 3.7 4.0
1\ __ .,.
~ \ !Allowable Foam Voiding (FuselaQe-l", 1\ ~- I
Configurations)
~'--. ~---+-- e-.--
~~ ~~~\. I l--r \. t--4--· ~ ~-\. ~l\;--+---- ·-·-···- --·- 1-----
i
! 1- ! ~~ \1 .
2.8
2.4
2.0
3.0
'1/1. . ?: I ~ \ .
I --\~\ 1.13 I
I
2.0 J e
"-," ,j:- ~f\! -N \ .. ·- I
~ '
~+.·-~ ~-- -~~ -
K ---···~
""15& . ~ . !\. ""l
, .2
0.8 1.0
""!~ 1' ~
~ ~ ----~ I ~ ~":: ~~~ 10 PI?!_ 15 ppl ~ ........ r-::::
0.4
~ 0 0 0 10 20 30 40 50 60 70 80 90 100
Foam Voiding ·%
FIGURE 69 FOAM WEIGHT AND VOLUME PENALTIES·% OF FUEL vs FUSELAGE SYSTEM VOIDING
.2 0 >
RESULTS OF TH.i CHROKATOJRATHIC ANALYSIS OF PROPANE GAMPLSS
Contaminants
Acotylona
ethane
rropanu
Butane
0.138
.0001
.025
91.107
Ö.730
jctuctlon l,i:nitD (')
.00 so
.0001
.0001
.0001
.0001
148
--