hpt I. FAA-RO-78-122 /1

TESTS OF CRASH-RESISTANT FUEL SYSTEM

FOR GENERAL AVIATION AIRCRAFT VICID

William M. Perrella, Jr.

C.2 DECEMBER 1918

FINAL REPORT

DDCDocument is available to the U.S. public through

the National Technical Information Service, r'J 7.]Springfield, Virginia 22161. JAN 29 1979

Prepsed for A

U.S. DEPARTMENT OF TRANSPORTATIONFEDERAL AVIATION ADMIISTRATION

Systems Rese-ch & Development Service

D.C - _

NOTICE

The United States Government does not endorse products

or manufacturers. Trade or manufacturer's names appearherein solely because they are considered essential tothe object of this report.

.... + + , " " ' " + + + ] II

II

I -• I

Technical Report Documentation Page

1. Report No 2. Government Accession No. 3. Recipient's Catalog No,

'~FAA 78-122e an ............~ 5. Re art Date

LTESTS OF RASH-JESISTANT FUEL SYSTEM FOR GENERAL / DecinP78/

AVIATION AIRCRAFT 7 V____________ 6. Performing Organization Code

8. Performing Organization Report No.

i a e lJr FAA-NA-78-48 -9 . Performing Organzton Name and Address I0] Work O inT oT TRA I)

,Federal Aviation AdministrationNational Aviation Facilities Experimental Center 11. Contract or Grant No.

Atlantic City, New Jersey 08405 184-5,l-10013. Type,6" Re" ort and Period Covered

12. Sponsoring Agency Name and AddressU.S. Department of Transportation Final AFederal Aviation Administration Feb p 76 - Juli78,Systems Research and Development Service . ]Washington, D.C. 20590

15. Supplementary Notes

/r16. AbstractA significant percentage of general aviation aircraft accidents result in post-crash fires due to the ignition of fuel spillage, often contributing injury ordeath to the aircraft occupants.

Testing was performed to demonstrate the performance of light-weight, flexible,crash-resistant fuel cells combined with the use of frangible fuel line couplings.Included in these tests were four full-scale crash tests of a typical light twin .

aircraft. In three tests, the crash-resistant fuel system performed satisfactory.The fourth and final test, where the lightest weight tanks were used, resulted ly

in tank failures and demonstrated a possible lower strength limit to the tankmaterial.

I\

17. Key Words 18. Distribution Statement

Crash-Resistant Fuel Cells Document is available to the U.S. publicAircraft Fuel Tanks through the National Technical Infor-Crashworthy Fuel Cells mation Service, Springfield, Virginia

22161

19. Security Classif. (of this report) 20. Security Classif. (of this page) 21. No. of Pages 22. Price

Unclassified Unclassified 31

Form. ')T F 1700.7 (8-72) Reproduction of completed page authorized - "

7\ L

N 0

~00~ 0 . a6

o1- -- 0g 1 0C; N

cc 0

-0I

3..

CAP V) -EE 0oz 1 016a 0ill

I'S0 1, 10 , 1 ,3 ,K * =-es

oU

A' INI -!iqac

'.3. I ~ o~ si I L SZ I t 91 ~ I I '0 6 L S CCU I I ii i i I

E- 0

TABLE OF CONTENTS

Page

INTRODUCTION 1

DISCUSSION 1

Tank Construction 1

Installation of Tanks in Aircraft 4Full-Scale Crash Tests 5

RESULTS 5

CONCLUSION 7

REFERENCES 7

APPENDIX

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

iii

0[IR 0 4A IAlf .~ 2

Ii

LIST OF ILLUSTRATIONS

Figure Page

1 Left Main Fuel Cell 8

2 Two-Ply Fuel Tank 8

3 Installation of Two-Ply Fuel Tank 9

4 Frangible Coupling Mounted to Fuel Cell Mounting Flange 9

5 Typical Frangible Coupling Installation (Wing Root) 10

6 Nose Gear Towing Attachment 10

7 Crash Test Site 11

8 Aircraft in Position for Test 11

9 Typical Acceleration Pulse 12

10 Damage to Left Wing Two-Ply Fuel Tank, Test 1 12

11 Failure of Main Spar, Left Wing, Test 1 13

12 Main Acceleration Pulse, Test 1 13

13 Aircraft Impact, Test 2, (Note Water Spray from Right Wing) 14

14 Damage to Right Wing, Original Aircraft Bladder Cell, 14Test 2

15 Damage to Right Wing, Original Aircraft Bladder Cell, 15Test 2

16 Damage to Left Wing, Two-Ply Fuel Tank, Test 2 15

17 Regular (Non-Crash-Resistant) Bladder Cell after Impact 16

18 Damage to Regular Aircraft Cell due to Pole Impact 16(Inboard)

19 Damage to Regular Aircraft Cell due to Pole Impact 17(Outboard)

20 Main Acceleration Pulse, Test 2 17

I iv

oiI-

LIST OF ILLUSTRATIONS (Continued)

Figure Page

21 Damage to Left Wing, Test 3 18

22 Damage to Right Wing, Test 3 18

23 Effect of Rock Impact, Test 3 19

24 Main Acceleration Pulse, Test 3 19

25 Damage to Outboard Area of 5.5 oz Tank 20

26 Damage to Inboard Area of 5.5 oz Tank 21

27 Damage to Bottom Inboard Area of 5.5 oz Tank 22

28 Tears in 8.0 oz Tank From Outboard Pole Impact 23

29 Tears in 8.0 oz Tank From Inboard Pole Impact 24

30 Tear in Gage Fitting Flange of 8.0 oz Tank 25

31 Damage to Inboard Left Wing 26 132 Damage to Outboard Left Wing 27

33 Damage to Outboard Right Wing 28

34 Damage to Inboard Right Wing 29

V

LIST OF TABLES

Table Page

1 Fuel Tank Characteristics 2

2 Physical Properties of Fuel Tank Construction (MIL-T- 327422B)

3 Crash Test Data 6

vi

- '1

INTRODUCTION A contract was awarded to theUniroyal Corporation to design andfabricate crashworthy tanks for a

From 1973 through 1976 the National typical light twin-engine aircraftTransportation Safety Board's (NTSB) (reference 2). These tanks were toyearly summary of general aviation be equipped with Aeroquip typeaccidents showed over 16 percent of DE5175-1-8A frangible couplingsall accidents result in fatalities, on the filler and vent fittings. Thewith 30 percent of these involving contract specification is shown inpostimpact fire. It is apparent that appendix A and includes contractonce ignition occurs iil the presence modifications (i.e., lightest weightof large quantities of spilled fuel, tanks and values).the survival chances of the aircraftoccupants are greatly reduced, evenwhen fire-fighting equipment is DISCUSSIONimmediately on the scene. The onlyfeasible way to decrease the inci-dence of postcrash fires is by the TANK CONSTRUCTIONreduction of fuel spillage andignition sources. Therefore, the Construction materials devised byFederal Aviation Administration (FAA) industry to meet MIL-T-27422B pro-initiated a program to evaluate a way vided a starting point for theof preventing massive spillage of program. The initial contractfuel during a crash, i.e., crash called for the construction of threeresistant flexible bladder cells used left-hand tanks of two-ply construc-with self-sealing frangible couplings tion, and three right-hand tanks ofat critical points in the fuel lines, three-ply construction. To assist in

reducing construction weight,The United States (U.S.) Army has the drop test requirement of theunquestionably established that fuel above specification was reduced fromcan be contained by flexible fuel 65 feet to 39 feet. This reductiontanks, thereby eliminating the resulted in a change in impactpotential of postcrash fire (refer- velocity from 65 ft/s to 50 ft/s,ence I). Though the Army does not which was considered more represent-now collate accident records, they do ative of a general aviation airplanekeep injury/fatality records. To accident environment. All tankdate, these data show only five fire fittings were initially specified torelated injuries and one fatality meet NIL-T-27422B requirements.that showed evidence of inhalationof fire by-products in a nonsurviv- At the Uniroyal facility, a left-handable accident. These tanks, while two-ply tank was filled with 59.2very effective, impose weight and gallons of water and dropped from acost penalities which could be height of 39 feet. The tank success-significantly reduced for general fully withstood the impact on itsaviation aircraft. Consequently, the leading edge with no visible damage.major thrust of this program was to Based on that fact and the results ofdemonstrate by full-scale test, a full-scale aircraft crash testeffective low-cost, lightweight, described later, it was decided notcrash-resistant fuel cells. to fabricate two of the three-ply

-r -

cells; two single-ply types were 9.6 pounds each. Refer to figure 1.specified. The tanks were to be The contractor was required to per-fitted with Uniroyal-designed fit- form the following material tests

tings similar to MS33581, with the per MIL-T-27422B:addition of a third ring. Thesefittings, are lighter in weight and (1) Constant rate tear (4.6.5.1)lower in cost than the Uniroyal Wall (2) Impact penetration (4.6.5.2).Expansionand Fibre-Lok®fittings used (3) Impact tear (4.6.5.3)on the other tanks, and have been (4) Panel strength calibrationdemonstrated as satisfactory in (4.6.5.4).65-foot free-fall impacts whenmounted in a 2 foot by 2 1/2 foot by Table 2 shows the results of these2 1/2 foot test tank, per paragraph tests. The Army helicopter accident4.6.6.2 of MIL-T-27422B. Table 1 record with crash resistant fuelsummarizes the characteristics of the systems installed is impressive. Ittanks delivered for testing at NAFEC. is believed that this technology can

be transferred to small generalThe main bladder cells with which theaircraft is normally equipped weigh

TABLE 1. FUEL TANK CHARACTERISTICS

Uniroyal Fabric Total Weight TankConstruction Fabric Weight of Fittings Weight

Tank Type Qty. Code No. Plies (oz/sq yd) (Ib) (Ib)

L. H. 3 *US758 2 12.75 6.50 27.0

R. H. 1 US759 3 12.75 7.76 38.0

R. H. 1 **US756 1 25.50 2.75 24.2

R. H. 1 US764 1 12.75 2.75 18.0

R. H. 1 US762 1 8.00 2.75 17.5

L. H. 1 US768 1 5.50 2.75 15.0

* One of these tanks successfully passed the 39-foot drop test.

** This material has FAA Technical Standard Order (TSO-C80) approval.

2

-A

N ~ ~ ~ 0 go N n V)t.

0

Pz40n0 0 0c

P-4

C1 14 .I-

P4) V.

E-'4.C/)IIA4 )

- --- * - - 1~

M.-- -W- -- - -- . -9"P

-A

aviation airplanes with appropriate sharp, broken components (ribs,consideration being given to differ- stringers, etc.) will not penetrateent accident environments, structural the tank. In addition, for theconfigurations, and tank shapes. lightest possible weight, the protec-Tank crash loads in a small fixed- tive effect of the structure must bewing aircraft are generally different considered in the design process andthan in a rotor wing aircraft. A in qualification testing. Forhelicopter often has a significant example, a lighter tank could be usedvertical velocity component during a if it were located behind, instead ofcrash, whereas, the general aviation in front of the main spar. To modifyairplane has a more significant an existing aircraft in that waylongitudinal velocity component. would not be feasible; however, forSince helicopter fuel tanks are new aircraft design, crashworthinessgenerally located under the floor, features could be very easily incor-

concentrated mass loads overhead in porated at minimal cost.the structure, suzh as engines andtransmissions, can impact a tank from INSTALLATION OF TANKS IN AIRCRAFT. Asabove while lower structure impacts crash-resistent tanks are stifferfrom below, due to deformation caused than standard bladder cells, theyby ground contact. Additionally, could not be installed in the openinghelicopter tanks are often of a shape in the wing normally used for thatclose to cubical, so that very little purpose. Figure I shows main celldeformation is required to develop components while figure 2 illustrateshigh hydraulic loads. In most a two-ply tank.fixed-wing aircraft, there are noheavy masses to sandwich a tank The crashworthy tanks were installedagainst the groundt by removing the bulkhead rib at the

wing root after the wing was removedAlso, the shape of a typical wing from the aircraft. The tank was slidtank is favorable in regard to into the wing, after which the ribhydraulic loads. Impacted against was riveted back in place, and thethe leading edge, the tank volume wing was reinstalled. Referring to

will tend to increase to a larger figure 3, it is seen that the tankspercent of its former volume, so the occupy the wing leading edge. Nohydraulic pressure will remain low. modifications were done on the wingThe primary design criteria for such where the filler flange fittings,tanks are tearing and puncture access, and gauge fittings wereresistance. There are some aircraft located. The frangible couplingswhich use belly tanks or nacelle (figure 4) were installed at the fueltanks of compound shapes or shapes outlet and vent fittings. Asnot conducive to volume increase upon bulkhead ribs interfered with theseimpact. The testing performed in this fittings, 2-inch diameter clearanceprogram is not applicable to such holes were cut in the ribs. To

tank types. provide a more severed operatingcondition for the couplings, aluminum

In any design of a crash-resistant tubing was used in lieu of thefuel system, attention must be given flexible tubing normally used for

to the tank/structure interaction in fuel lines. Acuating arms wereorder to obtain optimum results. installed to impact these linesFirst, the tank must be placed in a when the arms contacted the groundcavity which will not fail so that during the crash (figure 5).

4

It was found that there was a de- from the wings. Spoilers werecrease in the volumetric capacity of installed along the upper wingthe crash-resistant tanks relative to surface to keep the airplane fromthe existing aircraft cells, each of flying. At a distance of 10 feetwhich holds 59 gallons: the three-ply from the I-beam, poles were sunk intotank held 53 gallons, the two-ply, 55 the hill to a depth of 18 inches.gallons, and the single-ply US764 These poles were spaced symmetricallyheld 57.6 gallons. The decrease in off the centerline of the hill, at 42capacity occurred because these inches and 108 inches each. Thepreproduction tanks did not closely poles were hollow mild steel tubing,conform to the inner contours of the 4.375-inches outside diameter,wing. Ln a production tank, the fit 0.188-inch wall thickness and were 10would be more precise, probably feet in length. Small rock pilesres,'ting in a reduction in volume of were located on the hill to furtherless than a gallon. increase the severity of the crash

conditon (figure 7 and 8). There areA crashworthy fuel tank must not fail no standards in general use for awhen the aircraft experiences "sur- crash site as is used in this typevivable" crash accelerations. An of test; hence, the selection of theaircraft crash is considered surviv- type of poles, rocks, and hill wereable if the acceleration levels and selected to produce a destructivedurations are within certain limits environment to the fuel tank loca-which do not result in fatal injuries tion. The crash site was intended toto the occupants. These tests are be at least as severe as a crash atnot designed to bring the cabin an airfield involving airport struc-

environment up to the limits of tures such as approach lights.survivability, but they are designedto expose the fuel tank location to a In all tests, the aircraft main tanksdestructive environment, were filled with water. Accelero-

meters, CEC type 4-203-0001, wereFULL-SCALE CRASH TESTS. installed on the floor of the air-

craft at the longitudinal center ofThe crash tests were performed at the gravity location (station 126).National Aviation Facilities Experi- Accelerations in the vertical andmental Center (NAFEC) catapult longitudinal direction were recordedfacility. A compressed-air catapult on an oscillograph. The data werewas used to accelerate the test filtered at 90 hertz (Hz).aircraft along a 90-foot track. Atthe end of the catapult stroke, theaircraft, which was pulled by its RESULTSnose gear, was released to impact anearthern hill of 40 slope. At thebase of the hill, a 12-inch by The acceleration pulses had the12-inch I-beam was installed to break general form illustrated in fig-off the aircraft's landing gear. The ure 9. Time zero started as the nosenose gear was strengthened to with- gear struck the I-beam, resulting instand the catapult pulling force the initial spike shown. This spike,(figure 6), while the main landing typical in all tests, was about 100gear mounting bolts were sawed in g's longitudinal, for a duration of 5half to effect an easier separation milliseconds. Following this event,

for a period of 0.27 to 0.33 seconds

5

or so, depending on the test, there the highest speed possible with thewas no one acceleration peak distin- type of catapult used. Therefore,

guishable from the accelerations the empennage and engines were notcaused by the vibration of the installed on the airframe. From anstructure. The aircraft was deceler- analysis of the film, it was con-ating at an average level of approxi- cluded that the probable effect ofmately 2 g's. During this period of the engine mass on the local fueltime, the main landing gear was cell impact loads would have beenbroken off. The main acceleration negligible. However, the dynamicpulse was typically about 0.1-second behavior of the wing and aircraftduration, during which the aircraft after the impacts would have beenwas experiencing retarding forces significantly different had thefrom ground contact, as well as engines and empennage been installedrock impacts. Only in test 3 did the. (reference 3).pole impacts on the wing happen tocoincide with the main acceleration Major results of the four crash testspulse. After an analysis of the are summarized in table 3. The firsthigh-speed films, it was found that test evaluated three-ply and two-plythe pole impacts, which occurred tanks. The left wing received theapproximately 0.21 seconds into the most severe impacts. It containedcrash event, did not produce any the two-ply tank, which, upon latersignificant acceleration peaks within visual inspection, was found to bethe cabin of the aircraft. While the undamaged. The self-sealing Aeroquipeffect of these impacts on the wing couplings all actuated with nowas severe, resulting in much damage, leakage. The left wing was nearlythe inherent chord-wise flexibility torn from the fuselage, with only aof the wing prevented the ,force small part of the spar web holding ittransferral to the fuselage. (figures 10 and 11). Acceleration

levels are shown in figure 12. TheIt was necessary to keep the aircraft impact speed of the aircraft wasweight light in order to obtain 93 feet per second (ft/s).

TABLE 3. CRASH TEST DATA

AircraftWeight Maximum

lb Impact Acceleration, gTest Fuel Tank Fuel Tank Speed,No. Date L. II. R. II. - Empty Tanks Full Ft/s Fwd Damage

2-Ply* 3-Ply None to either

1 2/18/76 US758 US759 1,700 2,600 93 15 5 tank

2-Ply Original None to L. II. tank2 8/8/76 US758 Aircraft 1,710 2,660 93 29 7.5 R. If. tank ruptured

Bladder Cell

2-Ply I-Ply None to either tank5/18/77 US758 US764 1,660 2,598 95 27 55

6/30/78 I-Ply -Ply 1,680 2,590 95 - - Both tanks failed;

US768 US762 R. If. tank receivedminor daniage

*This tank was previously drop tested from 39 ft.

6

Test 2 compared the existing bladder occurred in three areas (figures 28

cell (right wing) with a two-ply to 30). Due to instrumentationcrashworthy cell. The two-ply tank problems, acceleration data were not

survived the 93-ft/s impact with no obtained. Refer to figures .31 throughdamage, but the original bladder cell 34 for structural damage.failed catastrophically, spraying out

its contents almost instantaneously. The single-ply 25.5 oz tank was notThe cell failed predominantly by tested in view of the results with

tearing (figures 13 through 19). the single-ply 12.75 oz tank. TheDuring impact, the aircraft rotated results obtained show that effectivecounterclockwise about its center of crash-resistant fuel systems can begravity as viewed from above, as the constructed which have small weight

forces delivered to the left wing and volume penalties. The use ofwere higher due to the stiffening these systems would undoubtedlyeffect of the two-ply tank. It came result in the saving of lives which

to rest at about a 300 angle. otherwise would be lost in postcrashAcceleration levels are shown fires.in figure 20.

Slight leakage was observed from both CONCLUSION

frangible couplings on the two-plytank after this test. Inspectionrevealed this was caused by corrosion It has been demonstrated that light-

in the flapper valve assemblies. The weight, flexible, crash-resistanttanks had been filled with water for fuel cells used with self-sealing,several weeks prior to the test. As frangible, fuel-line couplings canthe couplings are designed to operate effectively reduce postcrash fuel

with aviation fuels, this leakage was fires in general aviation aircraft

not considered a problem which would equipped with wing tanks. A single-occur in service, ply tank constructed with 12.75 oz

nylon reinforcement was the lightestTest 3 was performed to evaluate the tank which sustained no damage. Aperformance of a 12.75 oz single-ply single-ply 8.0 oz tank received only

tank installed in the right wing. A minor damage, and a 5.5 oz tanktwo-ply tank was run concurrently. failed catastrophically.Both of these tanks survived the95-ft/s impact with no damage dis-

cernable to a visual inspection REFERENCES(figures 21 through 23). Acceler-ation levels are shown in figure 24.

1. System Safety Newsletter, U.S.Test 4 was run with a single-ply 8.0 Army Agency for Aviation Safety,oz tank in the right wing and a Vol. 4 No. 4, 1975.,

single-ply 5.5 oz tank in the leftwing. At the impact speed of 95 2. Piper Navajo Fuel Tanks, FAAft/s, both tanks failed. The 5.5 oz Crash Resistant Modifications,

tank failed by tearing, which propa- Tanks and Testing, Uniroyal Reportgated extensively (figures 25 to FC-1641-77, March 1977.27). The 8.0 oz tank was onlyslightly damaged, the tears were 3. Dynamic Resp2onse of Struc-limited to less than 2 inches and tures, Pergamon Press, N.Y., 1971.

....

GFITTING, FUEL DRAIN VALVE

@ ACCESS FITTING

GAFILE FITTING

a FIL R A INGVLEFTIG~~ A WAN ER E N T RO I D

FIGUREWER 1.T LEFRMANIDELEEL

FIGURE~I 2CTOELYFULSAN

p b

E - -- - ~ - , - - , - .

FIGURE 3. INSTALLATION OF TWO0-PLY FUEL TANK

OEOUIV. TO MS33656-8

(D EQUIV. TO MS33514-12 MMO

OEQUIV, TO MS33649-12

ZMOUtJTING FLANGE

L5HXNT137 HEX NUT 77-48-4

FIGURE 4. FRANGIBLE COUPLING MOUNTED TO FUEL CELL MOUNTING FLANGE

RIB (REF)WING (REF)

FUEL TANK

MAIN SPAR (REF)

LEADING EDGE (REF)

2.011 DA. CLEARANCE HOLE BULKHEAD FITTING(CAPPED FAR SIDE)

FRANGIBLECOUPLING FUEL LINE (0. 5" OD, .032"

WALL 2024 - T3)

IMPACT ACTUATING ARM 77-48-5DIRECTION

FIGURE 5. TYPICAL FRANGIBLE COUPLING INSTALLATION (WING ROOT)

FIGURE 6. NOSE GEAR TOWING ATTACHMENT

10

FIGURE 7. CRASH TEST SITE

, 7

FIGURE 8. AIRCRAFT IN POSITION FOR TEST

ii

_- -- , -

ZA

0

TIME

MAIN PULSE 77-48-15

FIGURE 9. TYPICAL ACCELERATION PULSE

-1 16

FIGURE 10. DAMAGE TO LEFT WING TWO-PLY FUEL TANK, TEST 1

12

FIGURE 11. FAILURE OF MAIN SPAR, LEFT WING, TEST 1

25

-LONGITUDINAL

---- VERTICAL

20

.14

010

0.33 0.35 0.37 0.39 0.41 0.43 0.45 0.47 0.49TIME, SECONDS 77-48-18

FIGURE 12. MAIN ACCELERATION PULSE, TEST 1

SI13Il

FIGURE 13. AIRCRAFT IMPACT, TEST 2, (NOTE WATER SPRAY FROM RIGHT WING)

,I .

FIGURE 14. DAMAGE TO RIGHT WING, ORIGINAL AIRCRAFT BLADDER CELL, TEST 2

14

77-0-21

FIGURE 15. DAMAGE TO RIGHT WING, ORIGINAL AIRCRAFT BLADDER CELL, TEST 2

A?

FIGURE 16. DAMAGE TO LEFT WING, TWO-PLY FUEL TANK, TEST 2

15

.7'

77-48-2

FIGURE ~ ~ 18. DAAET EUA-ICATCL U OPL MAT(NORJ

~ -16

-V. -, --- -- -

-, - ,

,44

FIGURE 19. DAMAGE TO REGULAR AIRCRAFT CELL DUE TO POLE IMPACT (OUTBOARD)

- LONGITUDINALVERTICAL

Z5

20

J

0 0.4 0 . 44 O,.46

0. 34 0.3 3 0 O. 3b 7.70O.4

vmz LI, SECONDS 77-8-26

FIGURE 20. MA IN ACCELERATION PULSE, TEST 2

17

FIGURE 21. DAMAGE TO LEFT WING, TEST 3

FIGURE 22. DAMAGE TO RIGHT WING, TEST 3

18

'.4j

FIGURE 23. EFFECT OF ROCK IMPACT, TEST 31

6oA ---- LONGITUDINAL

-VERTICAL

so -

40

Io I

20.1 0 . 02 .2 0 4 02 .Z

FIGURE 24.8 M.IN ACEE. TO PUSE TEST 0.6 3z

I1

U7-4

00

200

00

00

-4

4 44

'-4r4

21

.00

0,

0

.4'4

22

'Aq

E-i

0

04

N

00o

cc

23

Fn I

0

P4

0

E-4'

r14

24

FE-

544

CYo

25-

co'

1-4

cn.

26I

1 44

e 1-4

27

.9 h t -.9 9 . 9***** '9 ~*'

A

'r

I * I4.

'9 3%

1

'a

A>5 .9 4'V

C.D

z

'-4

5-4

CD-' H

* 0* H

9 '' 9 5-4

Cl

Cl

CDH

~

28I

ii

I

I-I

II II111;P

FIGURE 34. DAMAGE TO INBOARD RIGHT WING

b

II

29

APPENDIX A

ORIGINAL CONTRACT SPECIFICATIONS

STATEMENT OF WORK

A. Introduction B. Detailed Requirements

Postcrash fire accidents continue to The contractor shall provide thecause a significant number of fatali- necessary qualified personnel,ties in general aviation operation. facilities, materials, equipment, andThe most promising method of con- services to perform and conduct thetrolling postcrash fires, thus following in the fabrication,reducing these fatalities in small testing, and installation of candi-aircraft, is fuel containment, date crashworthy fuel cells for aSuitable fuel containment can be typical general aviation aircraft.provided by using flexible bladder-type fuel cells which through special 1. Fabricate three crash-construction are resistant to worthy fuel cells for a typical lightbursting and tearing when subjected twin aircraft. These three cellsto impact forces associated with (referred to herein as B.1 cells) aresurvivable-type accidents. to conform in size and shape to the

original main left fuel cell.Special areas for consideration are: All three cells shall contain two

plies of 12-ounce-weight nylon fuelI. Tank seam failure or tank cell fabric.

rupthire

One of these three cells will be used2. Tank impact penetration for the crash impact test of MIL-T-

27422B, Part 4.6.7.9. as modified in3. Fitting pullout paragraph B4 below. These cells

shall include the following arrange-ment of materials in the construc-

United States Army Aviation Material tion.Laboratories programs have developedcrashworthy fuel cell materials for a) An innerliner coatinghelicopter applications and have plus barriers and cementsproduced prototype ceils for onecurrently-manufactured small fixed- b) Twelve-ounce nylonwing aircraft. These cells, while fabric, applied at 450most effective, impose weight andcost per'itties which could be reduced c) Twelve-ounce nylonfor civ; Iplications. fabric, applied straight

The purpose of this effort is to d) An outercoat of fuelfabricate and test relatively light- cell materialweight, low-cost crash-resistant fuelcells which will prevent massivefuel spillage in small aircraftsurvivable accidents.

A-1

2. Fabricate three crash- 4. Four (4) samples each ofworthy fuel cells for a typical light the constructions of the two tanks,twin aircraft conforming in size and B.1 and B.2, shall be subjected toshape to the original main right each of the five (5) compositefuel cell. All three cells (referred construction tests of Part 4.6.5 ofto herein as B.2 cells) shall contain Military Specification MIL-T-27422B,three plies of 12-ounce-weight nylon and (1) each of tbaks B.1 and B.2fuel cell fabric. If the left shall be subjected to the crashfuel cell fails when subjected to the impact test of MIL-T-27422B, Partcrash impact test of MIL-T-27422B, 4.6.7.9, with the exception that thePart 4.6.7.9, as modified in para- test tank shall be dropped from agraph B.4 below, then one of these height of 39 feet onto the forward orthree cells will be used for the leading edge of the tank. If con-cited crash impact test. These cells struction B.1 passes the 39-footshall include the following arrange- drop, construction B.2 need not bement of materials in the construc- drop tested. All tests shall betion. conducted at Contractor's facilities.

a) An innerlayer 5. All materials, includingcoating plus barriers and cements the fittings, will conform to the

requirements of MIL-T-27422B.b) Twelve-ounce nylon

fabric, applied at 450 6. All workmanship will bein conformance with the high quality

c) Twelve-ounce nylon requirements of MIL-T-27422B, andfabric, applied straight those of the aircraft industry.

d) Twelve-ounce nylonfabric, applied at 450

e) An outercoat offuel cell material

3. All openings in thetanks shall incorporate fittingsadaptable to the breakway valve or tothe frangible tank port to wingsurface structure, whichever isrequired. Valve fittings shall besized to that breakaway tank to fuelline valve which is an off-the-shelfitem and closest in size to the fuellines used in a light twin aircraftsystem. Other port fittings shallconform to the sizes existing inthe operational cells presentlymanufactured for a typical lighttwin aircraft. If required, addi-tional openings shall be provided tofacilitate installation of valves.

A-2