Implementation o f “MSG-3” for Crack Growth Analysis of...

JAST, Vol. 10© Iranian Aero

Im

Grow

AD ATA ED EFT FAA FCG FD FDA FEM FS GLS LC LDD LDP MRB MSG-3 NDI NDT OEM OEW SI SIF SLD SSI

1. PhD Candid2. Professor

0, No. 1, pp. 1- 1ospace Society, W

mplem

wth A

1, 2. De

The man SSI caircraft mfailure apMoreoverdesign apcorrespongrowth afractograp

Keyword:

Nomen

Accidental Air TranspoEnvironmenExtended FFederal AvFatigue CraFatigue DamFatigue DamFinite ElemFatigue StriGeneric LoLoad CrackLogic DeciLogical DeMaintenancMaintenancNon DestruNon DestruOperation EOperation EStructural IStress IntenStructural LSignificant

date (Correspond

15 Winter-Spring 201

mentat

nalysi

A.

epartment ofRese

Po

main goal of thicomponent of bmaintenance anpproachmeaninr, it cansignificpproach and ndence with theanalysis is inveaphy for the purp

SSI Component

nclature12

Damage ort Associationntal Deteriorat

Fatigue Testingiation Adminisack Growth mage mage Analysis

ment Method iation

oad Spectrum k sion Diagram cision Processce Review Boace Steering Grouctive Inspectiouctive Testing Empty Weight Empty Weight Integrity nsity Factor Logic DiagramStructural Item

ding Author)

13

tion o

is of A

R. Nikouee

f Mechanicalearch Branch

ostal Code: 14

*arnc

is article is to ib747 aircraft. d its safety leveg that MSG-3 r

cantly help savithe developme

e SLD of MSG-3estigated. The rpose of explori

s, Aircraft, Crack

n tion g stration

s

es ard oup 3 on

m m

of “MS

Aircra

e1* and S. A

l and Aerosph, Islamic Az

477893855,Te

[email protected]

implement the MThis process i

el. MSG-3 logicreduces the maing thetime andentof maintena3,an SSI compoapproach of t

ing the loading.

k, Analysis

TDI

The incluaircraairwoexpernew can bof Atask-3[1]. struclogicthe pConsdamainto tThis scheddesigconseundeits suscedifficthis can b

SG-3”

aft SS

Adibnazari

ace Engineerzad Universit

ehran, IRAN

om

MSG-3 processis expected toic took a top-dowaintenance cost d providing finaance programionent is selectedthis study is to

I Tea

definition of udes those taskaft systems, orthy conditirience in devaircraft has s

be developed tATAtask force

oriented mainOne of the m

tural treatmenc evolves into possibility ofsiderations oage, age explthe logic diag

section coduled maintegned to relate equences of tected. Each ssignificance

eptibility to anculty involvedis establishedbe developed

” for C

SI Com2

ring, Sciencety

s in structure fieincrease the ewn or consequeand upgrades

ancial benefits fing. In this std and itsfatigueo use the FEM

ar Down Inspe

Introduct

f maintenanceks required tocomponents,

ion [1].Airliveloping scheshown that mthrough the use led to the dntenance procmain parts ofnt. In this tra form which

f structural dof fatigue, loration, and

gram and arerontains guide

enance tasks scheduled m

f structural structural item

to continy form of damd in detectind, scheduled

which can be

Crack

mpone

e and

eld for ase of

ence of safety. for the tudy,in e crack M and

ection

tion

e in industry o restore or m

and structurne and man

eduled maintemore efficient se of LDP [2]development cess defined f the MSG-3 reatment, the h more directldeterioration p

corrosion, others areinc

outinely consielines for d

for aircraft maintenance ta

damage tham is assessed ininuing airwmage, and the

ng such damastructural mae effective in

k

ents

generally maintain an

res in an nufacturer

enance for programs

.The work of a new, as MSG-process is structural

y assesses processes. accidental corporated idered [2]. developing

structure asks to the at remain n terms of

worthiness, e degree of age. Once aintenance

n detecting

2 / Journal of Aerospace Science and Technology Vol. 10/ No. 1/ Winter-Spring 2013

Adibnazari S. and Nikouee R.

and preventing structural degradation due to fatigue, environmental deterioration, or accidental damage throughout the operational life of the aircraft. The structural maintenance task(s) are developed as part of the scheduled structural maintenance [2]. In this study,the aircraft SSI component of a Boeing 747 wasselected for assessment of crack growth using SLD, implementing crack growth analysis through the experimental study of fatigue crack . Smaller cracks

were investigated with fractography and lead cracks were investigated using“FEM “. Upon part selection, the general spectrum of loading was defined, and after designating loading spectrum, with the aid of modeling, fatigue crack growth and the element design were analyzed based on the finite element method.

Background

MSG-3 Development

In 1968 the MSG was created with a mandate to formulate an LDP for development of the initial scheduled maintenance requirements for new aircrafts. That same year, representatives of the steering group developed “MSG-1” which, for the first time, used an LDD to develop the scheduled maintenance program for the new Boeing 747 aircraft. In 1970, MSG-1 was updated to MSG-2 to make it applicable for later generations of aircraft.MSG-2 decision logic was subsequently used to develop scheduled maintenance programs for the aircraft of the 1970s.

In 1979, the ATA task force sought to improve on MSG-2 in order to address a new generation of advanced technology aircraft (B757 and B767)[1]. The work of the ATA task force led to the development of a new task-oriented maintenance process defined as MSG-3[1]. Today, MSG-3 is the only method used by commercial airplane manufacturers. Policy states that the latest MSG analysis procedures must be used for the development of routine scheduled maintenance tasks for all new or derivative aircraft. In MSG-3, the structural inspection program is designed to provide timely detection and repair of structural damage occurring during commercial operations. Detection of damages such as corrosionand fatigue cracking by visual and/or NDT procedures are considered [1].The primary objective of the scheduled structural maintenance is to maintain the inherent airworthiness throughout the operational life of the aircraft in an economical manner. To achieve this, the inspections must meet the detection requirements of each of the AD, ED and FD assessments. Inspections related to the detection of AD/ED are applicable to all aircraft when they first enter service [2]. Also inspections related to FD detection in metals are applicable after a threshold [2]. Additionally, accidental Damage (AD) is characterized by the occurrence of a random discrete event [2]. Besides, Environmental Deterioration (ED) is

characterized as structural deterioration resulting from a chemical interaction with its climate or environment [2]. Finally, Fatigue Damage (FD) is described as the initiation of a crack or cracks due to cyclic loading and subsequent propagation [2]. Moreover, in order to increase the level of safety and economical gain for operators as well as manufactures and to ease the oversight of authorities in the structural division, at first all aircraft structural elements shouldbe classified. Secondly, the group classification of the types of inspections and maintenance intervals has tobe examined.The above-mentioned elements consist of [2]:

1. SSI which in fact are primary structural parts of the aircraft.A Structural Significant Item (SSI) is any detail, element or assembly, which contributes significantly to carryingflight, ground, pressure or control loads, and thefailure of which could affect the structural integrity necessary for thesafety of the aircraft [2].

2.Other elements, known as “Other Structure”,is judged not to be a structural Significant Item. It is defined both externally andinternally within zonal boundaries [2].

The structural logical diagram“SLD” in the MSG-3. process is shown in Fig. 1.

Extended Fatigue Testing Between 2002 to 2005 three articles were published by Bakuckas and Carter (2002 -2003) and Mosinyi, Bakuckas, Ramakrishnan, and Lau-Tan-Awerbuch (2005) on extended fatigue testing onsome parts of a

scrapped Boeing 727 (i.e. extended fatigue testing to evaluate the structural integrity of high age aircraft) [3,4,5]. In fact these articles were the result of a common project accomplished by a team of representatives from the Federal Aviation Administration, Delta Airlines, and Drexel University whichlasted four years and involved tear-down

inspection and extended fatigue testing on a scrapped Boeing 727 structure with the total cycle equal to 60000 [3,4,5]. These activities were accomplished on suspected widespread crack growth points, and for this purpose, eleven aircraft fuselage panels were dismantled from the aircraft;seven panels with unique damage were investigated by Non-destructive testing (NDT) and the four remaining uniquely damaged panels were examined in an FAA-approved laboratory under complete aircraft fuselage testing[3,4,5].In fact, possible existing cracks and crack growth during this test were actively inspected. There were no signs of crack after 43500 simulated flight cycles (FC). In an aircraft which was equal to 60,000 cycles before being completely scrapped, after 104,000 flight cycles on the test panel, there was no sign of crack [3, 4, 5].

/ 3 Journal of Aerospace Science and Technology Vol. 10 / No. 1/ Winter - Spring 2013

Implementation of “MSG-3” for Crack Growth Analysis of Aircraft …

Figure 1. SLD in MSG-3 process

Theoretical Investigation

Introduction of the Part to be investigated

As mentioned in the previous section, the main goal of this study wasto implement the MSG-3 process on selected SSI elements of a Boeing 747 aircraft.The study focused on a part of the structure and according to the SLD and MSG-3 process, the intended part was designated through a structure logic diagram. The selected SSI parts of a Boeing 747; with accumulated 16756 FC and 64555 FH were removed from section 41 of this aircraft andlocated on Frame Station 300[6].The main task in this study was to investigate crack growth rate in crack locations, in a defined direction and under a specified default loading spectrum. Figs. 2 through 4are illustrations of the part under investigation. A hole is shown in one of these areas. The general form of these cracks for the selected part is shown in detail in Fig. 4.

Figure 2. Theimage of frame station 300, 280 and selected component on aircraft fuselage

4 / Journal ofVol. 10/ N

Figure 3. Im

Figure 4. The

Surface Cra

Fractographyfailure (fatphenomena).Bused to investreported in thspecifically fo

Althougnumber of cyfatigue crackcould show using an opticrack sampcorrespondenstriation founcyclic load.Tfor crack groeach loadingconfigurationrole in determit also provida part can bethe defined sof fracture swas designaSchijve, NixMatsunaga; wconfirm fatialuminum aConsequently

f Aerospace ScienNo. 1/ Winter-Sp

mage of selected

e exact location

acks Fractogr

y is a methodtigue, corrosBecause of sutigate and show

he 1951 articleocusingon “fatgh “Orowan”yclic operationk at a critical p

fatigue crackical microscopple. They nce betweennd in each loThey also preowth rate and g cycle as a n. Not only tmining fatigudes a situatione defined by cspace and distriation, fatig

ated by Lairdx and Flowewho declared igue crack alloy unless y, if there was

nce and Technologring 2013

d component at

n of the crack un

raphy

d of analyzinsion, and urface fractionw damage and

es issued by Zatigue striation”” announced ns are requireposition, Forsk in each loape and one aluestablished

n the numbeoading cycle esented reasofatigue crackresult of fr

this result haue crack growtn in which loacounting fract

stance [7].In rue crack grow

d,Smith, McMer, and also that there wagrowth in ein a cycle-

s no sign of fr

gy

frame STA. 30

nder test analys

ng damage another simil

n, this method d failure signs, affe and Word” [7].

that a greed to start initissyth and Rydading cycle buminum fatigua one-to-on

er of fractuand number

onable evidenk propagation racture striatioas a significath property, bading history oture striation relation to forwth mechanis

Millan, PellouMcEvily an

asno evidence either steel -by-cycle basracture striatio

0

sis

nd lar is as

den

eat ial

der by ue ne

ure of

nce in on ant but on in

rm sm ux, nd to or

se. on,

therefor necesof ocand sand s

strucare showeinspetoolsmicrothe eto invan elfractocrackthe lmodesize columedge of stcrackcracksurfaresulsurfafrictifatigureasotrigganalyresul

Tab

(8

89

e would be sodamage and ssary to desigccurrence andsurface fractusmooth materiMechanism tural elementshort-term loadever, it is poection of whic in fractogoscope, and slectronic micrvestigate surfalectronic micrography. Figsk on the part (eft side of the is given in Tfrom the firstmn is the exaof fracture su

triation beingk growth rate.k surface frace fracture shlted from fatace fracture onon from rubbiue panels aons for craered this siysis was accolting from fati

ble1. The result

) Dista

fracsurfac

8 2.8 .3 1

Fig 5. Electron

ome other caufailure ana

gnate their md define the pure specificatioials [7]. for the int

s failure whicding, fatigue, ssible to see

ch is done by graphy are similar equipmroscope used w

face fractures oroscope (SEMs. 6, 7, and 8(Fig 4the smalhe hole). CracTable 1. The ft edge of fracct location of

urface,the thirdg considered,. Figs. 9, 10, ractography. howed that thtigue striation the fractogring was one o

and crack grck initiationituation. Themplished by gue.

t of fractograph

ance on cture ce ( )

Nof st

6.6 15 0.7

nic microscope fractograph

S and Nikouee R.

uses, such asalysis. Henceechanism and

part operation on, especially

troduction ofch could be in

corrosion, or surface frac

fractography. the x-ray, ment[7]. In twas LEO 440on selected pa

M) (Fig 5) wa8 are related tllest crack is lck growth ratfirst column iscture surface,thf striation fromd column is th and forth cand 11 showFractography

he lead crackon; but thereraphy.Therefoof the reasons.rowths were

n, accidental erefore, fatigassuming crac

hy around small

Number f fatigue triation

18 8 5

used for the prhy

Adibnazari S.

s accident, e, it was d situation

condition y forbrittle

f aircraft vestigated

r creeping; ctures, the The main electronic

this study, 0i. In order art cracks; s used for to a small located on te for this s the crack he second m the first he number column is

w the large y of this k initiation e was no ore surface Although

possible damage

gue crack ck growth

ler cracks /( )

1.49 1.87 2.14

roject

Implementat

Figure 6.

Figure 7.

Figure 8.

Figure 9

tion of “MSG-3”

. Fatigue striatiwith 500

Fatigue striatiowith 1000

Fatigue striatiowith 1130

. Fatigue striatiwith 1134

for Crack Growt

on at 8 mm of ctimes magnific

on at 88mm of 0 times magnific

on at 9.3 mm of 0 times magnific

ion at 2mm of c4 times magnific

th Analysis of Air

crack initiationcation

crack initiationcation

f crack initiationcation

crack initiation cation

rcraft …

point

n point

n point

point

L

Inilanpinsposathgdemancrspdhlo6vcewspselifl

JoVo

Figure 10. Fapoin

Figure 11. Fatiw

Loading

n order to perllustrated in nalyzed usingurpose, the ndicators. As pecification of flight, a reample for larghe general loaeneral model efine loadingmpty operationd aircraft opruising level, peed, fuel cifferential caistory and aioading spectru000 flight cyariable coeffienter of gravi

wasrelated to fpectrum waservice life of ife of such alights. The

ournal of Aerospacol. 10 / No. 1/ Wint

atigue striation ant with 1560 tim

igue striations awith 2180 times

rform fatigueFigs. 5 and g the finite first stage it is a comp

f applied loadference wasuge commerciaad spectrum, w

of loading hg included:inteon weight, mperation data cruising spe

onsumption abin pressure ircraft reactioum consisted oycles, which icient of loadty and differe

flight altitude.one-tenth of a wide-bodie

aircraft is typregular fligh

ce Science and Tenter - Spring 2013

at 6.3mm of crames magnificati

at 25.3mm of crs magnification

crack growth6, the bigge

element methwasto desigplicated task

ding during diused to defineal aircraft andwhich wasthe

history. Paramended aircraft

maximum paylsuch as rang

ed, takeoff spas well as variations, a

on in wind gof the continuwasa combin

ding (∆g) in ential cabin pr The number

f the maximud jet aircraft,apically define

ht profile of

/ 5chnology

ack initiation ion

rack initiation

h analysis, aser cracks arehod. For this

gnate loadingto designate

ifferent stagese the loadingd to designateen utilizedas ameters used toft dimensions,load capacity,ges of flight,peed, landingthe trend ofaircraft flightusts [8].Each

ued loading ofnation of thethe aircraft’s

ressure, whichof flights in aum designedas the serviceed as 60,000f commercial

s e s g e s g e a o , , , g f t h f e s h a d e 0 l


aircraft is dshown in Fig

Table 2. Loa

Stage no.

1 2

3-8 9

10-15 16 17 18

Figure 12.

In diffenamed in tabcould affectgroups [8]:

- Aircraft

climb, an

- Aircraft

preflight

taxiing, t

- Wind gu

- Internal

The firsweredifferenwereacceleragravity. Samasthe numbethe period wflight cyclesthe aircraft reported in Taircraft’s fustructure, n


divided into g. 12and name

ad Environment

Stage

Preflight TaTakeoff Ru

Climb Cruise

Descent Landing

Landing RoPostflight Ta

. Ordinal diagra

erent flight mble 2, the cir

t fuselage str

flight maneu

nd landing sta

ground op

t movement, c

the landing ph

ust loading dur

cabin pressure

st three circunt from iated changesmples of theier of cycles iwasconverted

s [8].The finaexperience i

Table C21 refuselage as numbers for


different staged in table 2 [8

t at Each Stage

Load e

axi un

oll axi

TaxG

Gust anGust anGust an

LandRolTax

am for stages of

modes shown rcumferential ructurewasdiv

uvering in d

ages

erations incl

consistent surf

hase, and land

ring weather d

e

umferential lointernal cabs in the aircir variations in a specified

d to equal val accelerationin the centerference [8]. Ca cylindrica

r the shear

gy

ges, which a8].

of Flight Profil

environment

xi Ground Ground nd Maneuvernd Maneuvernd Maneuver

ding Impact ll Ground xi Ground

f flight profile

in Fig. 12 anloading whic

vided into fo

different cruis

luding aircra

face movemen

ding impact

deterioration.

oadings (whicbin pressurcraft center werecalculated period. The

values of 600n spectrum thr of gravity Considering thal monocoqur stresses

are

le

nd ch

our

se,

aft

nt,

ch re) of ed en 00 hat

is he ue in

longiwerecylindiffealtitu[8]. asthecentediffedirecEqualoadiobtai

σ =Wheσ: direc

σ : Spress/2:aircr∆g ∶table

: on m

σwoul

pane

aircrselecnose loadilocatshow

F

itudinal aneomitted andndrical pressuerential cabinude and is recThe stress s

e sum of tower of graverential”in thction of the ation 1.This ing spectruinedfrom Tabf(p) + f(g)re:

stress in ction at point

Stress in dam

sure equals 2 for circumf

Stress in daft inertia and

Load coeffice

Stress definmaximum des

.It wasassum

ld occur at th

l whileσ w

aft and zero cted compone

section.Finaings and thetion would b

wn in Fig.14.

igure 13. Viefind

nd circumd calculatedurized vessel.n pressure icorded in Tabspectrum of w term, the “vity” and the longitudinfuselage cou

equation cm to a

ble C21 in ref= σp + (1longitudinal

of damage

age location / for lonferential crac

damaged locd aerodynam

cient accordin

ed in longitusign stress, s

med that ma

he upper cro

would be zero

distance [8]ent is shown ially the come stress spece based on th

ew of the locading spectrum

S and Nikouee R.

mferential d as for a Internal andis related to

ble C20in the the damagedstress spectruthe “cabin nal or circumuldbe calculaconverts thestress spec

ference [8]. ∆g)σ1g

l or circum

because inter

ngitudinal crks

cation as a ic loading

ng to loading

udinal directishown in the

aximum stres

own plate of

o at each en

]. The locatioin Fig.13 and

mbination of trum in the he following

ation of damam loading

Adibnazari S.

directions a simple d external o aircraft reference

d location um in the

pressure mferential ated using e general ctrum as

(1)

mferential

rval cabin

racks and

result of

spectrum

ion based e form of

ss (σ )

the wing

nd of the

on of the d 4m from

different damaged diagram,

age for

Implementat

Figure

Finite El

One mamechaniccrack tipat that pThe subj

Figure 1

Critical a

The initiain fracturfactor eqthen cracgrowth wwith the s

Whecrack grothe unknintensity structureseverywhelength in

tion of “MSG-3”

e 14. Sample osho

lementMode

ain point of cs is to find t

p. In order to point, the finject wasassu

15. (a) Two dimw

and Initial Cr

al crack lengthre mechanics,quals the fracck length willwould beunstaspeed of sounen the loadingowth in the fatnown paramet

factor equals s, loading aere. So, as a n the integrat

for Crack Growt

of loading specown in Fig. 13

eling and Ana

fatigue anathe stress intefind the stre

ite element mumed to be

mensional modewith second ord

rack Lengths

h was assume, if the crack cture toughnesl be critical. able and the cd.

g for crack grotigue mechanter is the lengfracture tough

and stress aresult, to findted structure,

th Analysis of Air

trum for locati

alysis

alysis in fraensity factor ess intensity fmethod was two–dimensi

els produced froder elements ac

s

d 0.05 inch [8tip stress inte

ss of the maIn this case,

crack is propa

owth is knowismis well-degth, and the hness. In integ

are not the d the critical , the crack l

rcraft …

ion

acture at the factor used. ional.

Asathmpwcaseinpfa

crth

restv(

om initial 1 milcompanying bo

8] and ensity

aterial, crack

agated

wn and efined,

stress grated same crack

length

shseenwar

JoVo

Additionally, ame crack, thhe loading wa

mentioned analanetary stre

werefrom thealculation neelected directn X and Yropagated in actor in thef werack conditiohe (equ

was calcuesults is showtress intensitarious crack ) = (

llimeter crack aoundary layer an

hould be studee the criticalnd points of

wings. For crirea, the follow

- Calculate th

- Specify theaxis of Fig.

- Select the f

- Read the ccurve from

- This numbepoint


it was plannehe same bounas applied accalyses werebess, and thee second oretwork is illtion for crackY direction compose mofirst and secrecalculated on and appliuvalent of ulated from wn in Fig 16ty factor verssizes. ) + ( . )

round designatend loading cond

died at critical points in the

f this area, mitical length

wing procedur

he stress field

e calculated st. 19

first curve und

correspondingguide diagram

er is critical c


ed with similndary layer cocording to Fi

based on twoe elements der with 8 lustrated in Fk propagation

causedthe de and the strcond mode and for the ecation of Pa ) wequation (2). This diagrasus longitudi

)

ed hole (b) Mesdition

l points. Refere first zone fusmeaning the at arbitrary p

re should be ad

at the arbitrar

tress field in t

der the K-IC o

g number of m

crack length at

/ 7chnology

lar holes, theonditions andg. 15(a). The-dimensionalto be usednodes. The

Fig15(b).Then and loading

crack toberess intensity( ).evaluation of

aris equation,wasused. The[11] and the

am shows theinal stress in

(2)

sh generation

r to ref [8] toselage are thethreshold of

points of thisdopted:

ry point

the horizontal

of the material

f the selected

t the arbitrary

e d e l d e e g e y . f , e e e n

o e f s

l

l

d

y


For examthe critical seend points/th

Location of At this pointand critical investigationfor crack probecause poin(deck) and longitudinal stress level ofor selectedforcrack pconsidered:

- being far

- being nea

the faced


mple, doing thection of the fhreshold of t

Figur

Initial Crackt, the initial c

crack lengn of this studyopagation at pnt “O” is ver

reduces thbending Stre

of points A, Bd points.Thespropagation,

from neutral a

ar to end poi

zone fuselage


his procedurefirst zone fusethe wings an

re 16. Critical

k crack length, cgth were dey showed that point “O” in ry close to thefluctuation ess. Therefore, C, D are mo

se are the hhence tw

axis (deck)

ints/threshold

e

gy

e for 3 points oelage means thnd, as a resu

crack length

crack directioetermined. Th

the stress levFig. 17 is lowhe neutral axrange of th

e, obviouslythore than point highestpotenti

wo featuresa

of the wing

on hat ult,

criticmm,

on the critical

on, he

vel w, xis he he O

ial are

in

pointmarkThis

Fig

cal crack lengand 10mm,re

l section of fir

For this purpts are shown ked by red, bcoding waske

gure 17. Illustra

gth for A, B, spectively in

rst zone fusela

pose, 4 pointin Fig.17,na

black, green, ept in all curve

ation of selectedgrowth analy

S and Nikouee R.

C points areFig16.

age

ts wereselectamed A, B, Cand blue, reses and diagram

d point for fatigysis

Adibnazari S.

6 mm, 19

ed. These C, D and spectively. ms.

gue crack



Fatigue Crack Growth Analysis

In order to calculate the component fatigue life limit from a fracture mechanics point of view, it wasneeded to consider the subject of fatigue crack growth. In the finite element analysis for crack growth modeling, we need the specific criteria (i.e. finite elements output data) which, in this field of study, are obtained from variable functions in structural analysis. In fact these variables are displacement. If a quantity such as stress, strain, or energy, which is related to these displacements, is considered in the final evaluation, the stress intensity factor which introduces fracture mechanics elasticity wouldhelp find the property thatis in a linear situation with the field of stress at the final stage of the finite element analysis. With the aid of a model based on this property, we canachieve a certain indicator for fatigue crack growth. The item in accord with fatigue analysis from a fracture standpoint in the finite element method is, in fact, crack growth. In this case, using the model and growth criteria hand in the same result, and the fatigue crack rate (i.e. growth length in each loading cycle) wouldbe drawn in the form of extremities of stress intensity factor variations at the point of the crack. Fatigue crack growth rate diagrams (i.e. crack length in each loading cycle) are presented for 3 types of aluminum based on the tensional stress variation of the coefficient extremity at the crack tip [9]. This logarithm diagram is divided into 3 zones which, for the intermediate zone (linear part) power equation, are known as Paris’ law (Equation3,4) [10, 11]. = ( ) (3) = (∆ ) (4)

The material used for the aforementioned modeling wasaluminum T6-7075 and the two constant values“c” and “m” were calculated whilethese two constant values for the proposed aluminum wereselected as 3.6e-11 and 4.1 [9]. Consequently, the part service life time was calculated with the integration in Equation 5being per se derived from Equation (4) [10, 11]. = (∆ ) (5)

By assuming the final crack a critical crack, its length will be located at the upper limit of the equation (5).In this equation, N wouldbe the final service life of the component. The location and direction of the initial cracks in the structure will determine the amounts of the stress intensity factor during growth and consequently, by assuming a model f(∆k) function, it wouldbe possible to calculate the integral (4). The size and direction of initial crack is the same as the bigger crack, and the situation around the crack is the same as the component used for proposed modeling (bigger crack on right hand side of hole in Fig. 4).With respect

to the number of loading cycles beingknown, the calculationof crack growth was carried out. The calculation of the crack growth rate (∆ ) in this number of cycles was considered and calculated using equation (6), and in this case integration of N wouldbe made [10, 11]. = (∆ ) (6)

In other words, ten blocks of loading equaling 6000 cycles were applied to the model at each block. The sum of the block increment for each of them was calculated. Therefore loadings, except cabin deferential pressure, werecalculated as a variation of acceleration at aircraft c.g for 6000 flight block, as reported in Table 21in the reference [8]. Thusfinal acceleration spectrum which the aircraft wouldexperience at c.g was calculated.This acceleration spectrum became fuselage longitudinal stress and after combining with longitudinal and lateral stress caused from internal cabin pressure, the stress spectrum for 1 period of 6000 flights was achieved. Therefore we would be able to do the same calculation for 10 blocks of 6000 flights.

If da/dN is a constant value, it is possible to calculate crack growth rate for 1 loading cycle (1 flight) and then multiply any number of cycles, which result in final length. In fact the practical method is to divide the number of spectrum cycles by 6000 flights, and the number of cycles for 8 different stages of each flight couldbe calculated. This helps the specified code to be capable ofsolving the finite element model for 6000 flights and causes the exact number of flights required for cracking to reach the required specified length.

The worst and most unrealistic method is to apply 60000 flights simultaneously in order to make the crack reach critical distance, because it is not possible to distinguish at what number of flights critical value is recorded. The nominal method for solving this problem is to divide 60000 flights into 10 different parts andthe final crack length in each stage of the initial crack wouldbe the default value for the next stage. Each of the ten parts consists of 8 loading blocks. In each block there exist 6000 flight cycles. For example, at the first block there are 6000 aircraft surface movements before commencing the flight, and in the second block there are 6000 takeoffs, and the situation is the same for climb, cruise, landing, and surface movement after landing and flight. Consider, if the crack for the fifth period of loading reaches critical value, that event wouldoccur between 24000 and 30000 flights. In order to find the precise number of failure situations, it is possible to increase the number of loading periods, for example to increase 10 periods of 6000 block 60 of 1000 block will help calculate the results for less than 1000 flights.



In Fig.18, the stress intensity factor at maximum and minimum loading for60000 flightsis as illustrated for the 4 points of A, B, C, and D at the first zone of the fuselage. Also inthe landing phase an excessive load wouldbe applied to the fuselage, but as interior cabin pressure approaches zero, there is no sign of increase in stress values. For as we see in Fig.19, from maximum and minimum values it is possible to find the range of variation in values.

In Fig.20 we can see ∆ for 60000 flight cyclesof 4 specified points at the first zone of fuselage. The limit for cruise flight and landing of aircraft can also be seen in the diagram. One period of calculation consisting of 6000 flights of 54000 to60000 flight cycles for 4 points at the first zone of the fuselage is shown in Fig.21. The main point which should be noticed in comparing Figs.19 and 21 is the significant situation for ∆ at the landing phase and the during cruising conditions. As we can see, the fuselage experiences the most deviation of stress intensity factor during landing phase of the flight; but that does not mean that most crack growth occur during this phase of the aircraft. By referring to reference [8] table 21, we can see that, for 6000 flights the aircraft will experience 6.8 million cycles, but for the aircraft landing phase, only 6000 cycles is applied during 6000 flights. That is to mean that in each flight more than 1 million cycles of variation in the ∆ domain is applied to the fuselage while it cruises, but ∆ for the landing phase occurs only once. For this reason, in order to have a share of each stage of flight represented in crack growth it is better to apply the number of cycles in the calculations.

Fig.22 shows the crack growth rate with an initial length of 1 millimeter for assigned points. As we see, crack growth during the aircraft landing phase (third point from right) is approaching zero. In Fig.23 (a, b), crack growth variation based on each phase of flight is shown. It is obvious that a crack with an initial length of 1 millimeter at point A and B does not have considerable variation (the red and black curves). In point C the range of variation is small and rises incrementally with increases in crack length. Even at point D crack length shows no considerable changes, and it is possible to see ascending and descending trends for ∆ . If the calculation of the flights is continued for more than 60000, once again the trend of changes increases. This significant change for ∆ is a result of stress effect respecting the location of the crack. The stress situation at the tip of the crack changes continually, however. In Fig.24, ∆ variation of values can be seen. Also in Fig. 25 these parameters are illustrated in integrated form, including the 1millimeter initial length of the crack. As we can see, cracks with an initial length of 1 millimeter will increase to 3 millimeters as a result of 60000 flight cycles. By limiting flights to 6000, the crack growth value for 6000 cruise flights can be seen. In order to distinguish the trends of change for crack length as shown in Fig.24, the points for the end of each of the 6000 flights is presented.

In Fig. 25, the fatigue crack growth curves are compared with linear conditions, and it is obvious that as a result of less crack growth, the trend of these changes approaches linear conditions. In case of a continued increase in cyclic loading, the growth trend wouldleave the linear situation, and its shape woulddepend on the length of the crack. In addition, we see that at no point wouldthe crack reach the critical length.

Figure 18. equivalent stress intensity coefficient at maximum and minimum loading of60000 flights for 4 points at

the first zone of fuselage

Implementat

Figure 1

Figure 2

tion of “MSG-3”

9. Equivalent st

20. Equivalent

for Crack Growt

tress intensity cfrom

domain of stres

th Analysis of Air

coefficient at m54000 to 60000

ss intensity coe

rcraft …

minimum and ma0 on 4 points a

fficient change

JoVo

aximum loadint the first zone

s for 60000 flig


g in 1 period ofof fuselage

ghts on 4 points


f calculation,on

s at the first zon

/ 11chnology

n 6000 flights

ne of fuselage

1



Figure 21. The equivalent domain for stress intensity coefficient changes during 1 period of calculation, on 6000 flights from 54000 to 60000 flights on 4 points at the first zone of fuselage

Figure 22. Crack growth rate with initial length of 1 millimeter during 1 period of calculation, on 6000 flights from 54000 to 60000 flights on 4 points at the first zone of fuselage



Figure 23. (a, b). Crack growth rate for initial crack of 1 millimeter ofsample during 60000 flights on 4 points at first zone of fuselage



Figure 24. Crack length value of first 1 millimeters of sample as a result of 60000 flights on 4 points at the first zone of the fuselage

Figure 25. Crack length value of first 1 millimeters of sample as a result of 60000 flights on 4 points at the first

zone of the fuselage

Conclusion

The best ways to promote an aircraft structural design and maintenance program could be using the MSG-3

process (on structure). This process is based on LDP and covers the accidental, environmental and fatigue damages for SSI and other structural items. In this article we investigated the MSG-3 process for SSI



components. For this purpose fractography and the crack growth analyses by FEM wereused.In the selected component and for smaller crack the result of fractography for the surface fracture indicated that the lead crack initiation growth rate and also for bigger crack under life limit cyclic loadingin none of the selected points the crack would reach the critical length;hence, specified periodic inspection would be done. As mentioned before, the MSG-3 process covers all damage types. In future plan the first step for promoting the maintenance program in structure field is that all cracks in SSI components are going to be placed under investigation in selected aircraft tocover the fatigue damage and then the plan on the other components and damages type will be developed.

REFERENCES

1. Shannon, P. Ackert, “Basics of Aircraft Maintenance Programs for Financiers,” Date of Issue: October 1, 2010 www.aircraftmonitor.com

2. ATA (Air Transport Association) “MSG-3 Operator/Manu facturer Scheduled Maintenance Development” revised 2003.

3. Mosinyi, Bakuckas, Awerbuch, Lau,Tan “Crack Growith Assessment of High-Usage in-Service Aircraft Fuselage Structure,” Drexel University /

FAA / Delta Airline -Aging Aircraft Coference FAA/DOD 2005.

4. John G. Bakuckas, Jr - Aubrey Carter “Destuctive Evaluation and Extended Fatigue Testing of Retired Aircraft Fuselage Structure,” Proceedings of the 6th Joint FAA/DoD/NASA Conference on Aging Aircraft, September 2002, San Francisco, CA

5. John G. Bakuckas, Jr - Aubrey Carter “Destuctive Evaluation and Extended Fatigue Testing of Retired,” Aircraft Fuselage Structure: Projet Update Proceedings of the 7th Joint DoD/ FAA/NASA Conference on Aging Aircraft, New Orleans, Louisiana, September 2003,

6. BOEING COMPANY “SRM (Structure Repair Manual)- D6-13592 B747” 2010

7. McEvily A.J. and Matsunaga, H., “On Fatigue Striations,” Scientia Iranica Vol. 17, No. 1. Sharif University of Technology, February 2010.

8. FAA and USAF “RAPID (Repair Assessment Procedure and Integrated Design)” Version 2.1. May 1998.

9. ALCOA Co “Alloy 7075 Technical Sheet, www. millproducts- alcoa.com”

10. A. G, Grandt, Fundamental of Structural Integrity, Wiley and Sons 2003.

11. Broek, David, “Fracture Mechanics,” The Practical Use, 1988.

Implementation o f “MSG-3” for Crack Growth Analysis of...

Documents

Transcript of Implementation o f “MSG-3” for Crack Growth Analysis of...