A Study On The Use Of Neuber's Rule In Fatigue Crack

Initiation Predictions

by

SANJEEV K. VISVANATIlA, B.Eng.

A thesis submitted to

the Faculty of Graduate Studies and Research

in partial fulfillment of

the requirements for the degree of

Master of Engineering in Aerospace Engineering

Department of Mechanical and Aerospace Engineering

The Ottawa-Carleton Institute for Mechanicd and Aerospace Engineering

Carleton Universis.

Ottawa, Ontario, Canada

September 1998

O copyright

1998, Sanjeev K. Visvanatha

1*1 National Library
of Canada
Bibliothkque nationale
du Canada
Acquisitions and Acquisitions et
Bibliographie Services services bibliographiques
395 Wellington Street 395,nie Wellington
Ottawa ON K1A O N 4 Ottawa ON K IA ON4
Canada Canada
YOMfile Votre rëferenœ

Our f i Notre reférence

The author has granted a non- L'auteur a accordé une licence non
exclusive licence allowing the exclusive permettant à la
National Library of Canada to Bibliothèque nationale du Canada de
reproduce, loan, distribute or sell reproduire, prêter, distribuer ou
copies of this thesis in microfom., vendre des copies de cette thèse sous
paper or electronic formats. la forne de microfiche/^ de
reproduction sur papier ou sur format
électronique.

The author retains ownership of the L'auteur conserve la propriété du

copyright in this thesis. Neither the droit d'auteur qui protège cette thèse.
thesis nor substantial extracts fiom it Ni la thèse ni des extraits substantiels
may be printed or otherwise de celle-ci ne doivent être imprimés
reproduced without the author's ou autrement reproduits sans son
permission. autorisation.
Abstract
The local strain approach to fatigue life prediction contains a number of assumptions

which can lead to considerable error in the prediction of crack initiation Me. One

assumption is the use of an approximate relationship known as Neuber's rule to estirnate

the stress and strain at the notch root of a component. The applicability of Neuber's rule

in the local strain method was examined for two coupon geometrïes through a finite

element analysis. Ln addition, the ability of the local strain method to predict the lives of

the coupons subjected to s p e c t m loading was assessed by comparing local strain

predictions for the two coupons with results fiom a coupon test program. The findings of

this study verified the applicability of Neuber's rule in plane stress situations. A method

of estimating multiaxial elastic-plastic notch stresses and strains was verified to be an

effective means of accounting for notch root multiaxiality. A method of estimating total

life, composed of crack initiation and crack propagation, was proposed which accounts

for the notch size effect displayed in sharply notched coupons.

Acknowledgements
A mentor once told me that choosing a gnduate school is sirnilar to choosing real estate,

where the three most important criteria are "Location, Location, Location7'. He went on

to Say that it was the supervisor done who woufd either "make or break'' the two-year

experience, leading to the parallel mle for graduate school: "Supervisor, Supervisor,

Supervisof'. My two supervisors, Professor Paul V. Straznicky and Dr. Roy L. Hewitt,

confirmed this d e and were instrumental in making my graduate degree an experience to

remember! Their sage advice inspired me to End? and in many cases, construct the roads

which led to here.

Several others were extremely helpful throughout the duration of this thesis. Thanks go

to Mr. Jan Weiss for always taking the time out of his busy schedule to discuss the fmer

points of C-CI89, Ms. Pascale A E e for her assistance in the expenmental program, and

Mr. Luc Lafieur for coaching me on the use of the MTS rig.

Sincere gratitude is expressed to The Natural Sciences and Engineering Research Council

(NSERC) for the award of a Post Graduate Scholarship, The Canadian Space Agency for

the award of a supplement to the NSERC scholarship, Carleton University for the award
of a teaching and research assistantship, and to the hstitute for Aerospace Research at the

National Research Council of Canada for the generous use of their facilities.

Thanks go to my fiiends for the stirnulating discussions and meaningless jokes which

helped to keep uisanity at bay through the years. Cheers!

Finaily, gratitude is expressed to my parents, my siblings and their families. You have

been supportive throughout my academic career. Thank you.

Table of Contents

Absfract .......................................................................................... III

...

Acknowledgements .................... ....m..mms~m~m..mmmmmm..mmmm..m.mm..sm.. mm.mmmm.miv

Table of Contents ................................................. .w~mm8mm.mms~.mm.

List of Tables ................................................................................... xi

List of Figures ..............................................mm~.m..mmmm~m.mmmm.m

3.

List of Appendices ........................................................................ .xv

Nomenclature................... ..........m.smm.wms.mmm.sm.mmLm.s~.smw.mmm..m.mmw. .xvi

Units ................................................................................................ xx

.......mm~~mm.~mmI.m
Chapter 1 - Introduction

Chapter 2 - Review of Fatigue Crack Initiation Prediction ............5

2-1 Introduction ......................,

........................................................................ 5
..
2.2 Aspects of Fatigue Crack Initiation......................................................................... 5

2.3 Nominal Stress (NS) Approach............................................................................... 6

2.4 Local Strain (LS) Approach ...................................................................................... 8

2.4.1 Principle of LS Method ...................................................................................... 8

2.4.2 LS Method ................. .

................................................................................... 9
 2.5 Damage Accumulation ........................................................................................... 16

2.6 Principle of Equivalence ..................................................................................... 18

2.6. 1 Concerns ............................................................ ,

... ................................. 18
2.6.2 Notch Stress-Strain Estimation ..............................
,..................................... 19
2.6.3.1 Neuber' s Rule and irs Variations .............................................................. 19

2.6.2.2 Equivalent Strain Energy Density (ESED) .............................................. 24

2.6.2.3 Finite Element (FE) h a l y s i s ...................,... ........................................... 25

2-63 Notch Severity..................................................................................................27

2.7 Fracture Mechanics .....................................,.......................................................... 29

2.8 Surnrnary ................................................................................................................. 30

-
Chapter 3 Project Definition.................... ................'.32
. . . g . mm . . . . . .

Chapter 4 - Coupon Test Program .................................m.~~. ...34

4.1 Introduction ....................................................................................-..-.....................

34

4.2 Coupons and Test Sequences ..........................

...
............................................ 34
4.3 Test Procedure ......................................................................................................... 36

4.4 Detection of Crack Initiation ......................,.,.......................................................... 36

4.5 Test Results ....................

,.,..., ........................................................................ 37
4.6 Accuracy of Applied Loads ...........................
..................................................... 38
4.7 Crack Initiation Sites ......................................................................................... 41

4.8 Summary ....................... .

...................................................................................... 42
Chapter 5 - Local Strain Software .....................................m.....m...
#.43
5.1 Introduction .....................................
.,............. 43
.

vii
 5.2 Description of C-CI89 ............................................................................................ 43
5.3 McCracken Fatigue Life Prediction Program ......................................................... 44

5-3-1 Material Properties Dialog Box ............................. .

..................................... 45
5.3.2 Spectrum Dialog Box .................
.............................--.........................-......45
5.3.3 Prediction Methods Dialog Box ...........................
........*...............................46
5.3.4 Executing the Prediction .................................................................................. 47

5.3.5 Documenting Results ................... .

................................................................ 47
5.4 Validation of McCracken ......................
............................................................ 47

5.4.1 Overview of Validation ............

...................................................................... 47
. , . . .5.4.2
. . . . Material
. . . . . . .Data
. . . ...............
. . . . . . .. .,
.. .48
5.4.3 Cornparison of C-CI89 and McCracken Predictions ....................................... 48

5.5 Cornparison of Local Strain Prediction Programs................................................... 49

5.5.1 C-CI89 vs . LOOPIN8 ...................................................................................... 49

5.5.2 CC189 vs . McCracken .................................................................................... 50

5.6 Summary ................................................................................................................. 52

Chapter 6 - Finite Element Analysiç ......m.mmm.m.........................

m.mmmmmm53

6.1 Introduction ............................................................................................................. 53

6.2 Constitutive Models .............................................................................................. 53

6.3 Loading.................................................................................................................... 54

6.4 FEA of Low Kt Coupon .................................................................................... 55

6.4.1 Geometry of Low Kt FE Mode1........................................................................ 55

6.4.2 Elastic FEA of Low Kt Coupon - Verification........

...
............................ 56
 6-43 Elastic-Plastic FEA of Low Kt Coupon ........................................................ 57

6.5 FEA of High Kt Coupon....................................................................................... 58

6.5.1 Geometry of High K, FE Mode1................................................................. 58

6.5.2 Elastic FEA of High Kt Coupon - Verification ............................................... 59

6.5.3 Elastic-Plastic FEA of High Kt Coupon........................................................... 60

6.6 Results of Elastic-f Iastic FEA ................................................................................ 60

6.6.1 Low Kt Coupon ................................................................................................ 61

6.6.2 High Kt Coupon .................................... ....................................................... 62

6.7 Discussion o f FE Results ........................................................................................ 62

6.8 Summary ................................................................................................................. 64

Chapter 7 - Sensitivity Study .....................................m.mmmmm~.m.Immœ

7.1 Objective ................................................................................................................ 65

7.2 Description of the Stress and Strah Estimation Methods ....................................... 66

7.2.1 Solution Technique for Neuber's Rule .......................................................... 66

7 - 2 2 Solution Technique for Glinka's ESED Method ............................................ 67

7.2.3 Solution Technique for H o & m and Seeger's Generalized Meîhod ............ 67

7.3 Sensitivity of Notch Root Stress and Strain ..................................... .

... 69

7.3.1 Low K, Coupon ............................................................................................. 69

7.3 -2 High Kt Coupon ............................................................................................... 70

7.4 Sensitivity o f Crack Initiation Predictions .....................

...
............................. 71
7.4.1 Low Kt Coupon ................................................................................................ 71

7.4.2 High K, Coupon ............................................................................................... 72

 7.5 Summary ..........................*..................................................................................... 73

Chapter 8 - Discussion of Results ................................................75

8.1 Introduction ............................................................................................................. 75

8.2 Applicability of Neuber's Rule ............................................................................. 75

8.3 Agreement between LS Predictions and Test Results............................................. 77

77
8.3.1 Introduction ......................................................................................................

8.3.2 Crack Initiation vs . Total Life ................... .-....-........

,................................78
8.3.3 Fatigue Concentration Factor ...........................................................................81

8.3.4 Estimating Total Life ....................................................................................... 83

8.4 Material Properties .......................

.................................................................... 85
8.5 Equivalent Strain Equations ................................................................................. 86

-
Chapter 9 Conclusions ................................................................ 88

88
9.1 Conclusions .............................................................................................................

9.2 Recomrnendations for Future Research .................................................................. 90

9.3 Summary of Contributions .................................................................................... 91

References.............................m...... ..............................................3 2
List of Tables
Table 4- 1:Test Results for Low Kt Coupons.............................. ................................ 37

Table 4-2: Test Results for High Kt Coupons ................................................................ 38

Table 5-1: Cornparison of C-CI89 and McCracken Predictions for tef-man35 ............... 48

Table 6- 1:Force Convergence for Low Kt Coupon ........................................................ 58

Table 6-2: Force Convergence for High Kt Coupon ....................................................... 60

Table 7-1: McCracken Inputs for Crack initiation Sensitivity Study................................71

Table 8-1: Indication of Crack Propagation Phase for Low Kt Coupons .......................... 79

Table 8-2: Crack Length at First Detection for Hi& Kt Coupons .................................... 80

Table 8-3: Fatigue Concentration Factors for Low and Hi& Kt Coupons ....................... 81
List of Figures
Figure 2-1: Equivalence between Smooth and Notched Specirnens................................. 99

Figure 2-2: Local Stmin Method - Load Spectnim and Cyclic Stress-Strain Curve ......... 99

Figure 2-3: Local Strain Method - Hysteresis Loop Tracking ........................................ 100

Figure 2-4: Volume of Critically Stressed Material at Blunt and Sharp Notches ........... 101

Figure 4-1: Low R Coupon Geometry........................................................................... 102

Figure 4-2: High Kt Coupon Geometry........................................................................... 103

Figure 4-3 : Possible Crack Initiation Sites..................................................................... 104

Figure 5-1: McCracken Prediction Environment ............................................................ 105

Figure 5-2: Materid Properties Dialog Box .................................................................... 105

Figure 5-3 : Spectrum Dialog Box ................................................................................... 106

Figure 5-4: Prediction Methods Dialog Box ................................................................... 106

Figure 5-5: CycIic Stress vs .StressWrain Curve - AI 7050-T74................................... 107

Figure 5-6: Strain-Life Curve - Al-7050-T74 .............................................................. 107

Figure 5-7: Cornparison of SWT and LOOPINS Equivalent Strain Equaiions .............. 108

Figure 5-8: C-CI89 Representation of Stress vs. Stress*Strain Curve ........................... 108

Figure 6- 1: Aluminum 7050-T74 Stress-Strain Curve..................

............................. 109
Figure 6-2: Low Kt Coupon Mesh Convergence Study ................................................. 109

Figure 6-3: Finite Element Geornetry of Low Kt Coupon ............................................... 110

Figure 6-4: Cornparison between FE results and approximate relationship for stress vs.

distance...........,
..........,,........................................................................................ 110

Figure 6-5: High Kt Coupon Mesh Convergence Study.................................................. 111

Figure 6-6: Finite Element Geometry of High K, Coupon .............................................. 111

Figure 6-7: SP3 vs . Distance fiom Notch Root - Mid-Thickness of Low Kt Coupon .... 112

Figure 6-8: SP3 vs .Distance fiom Notch Root - Surface of Low K, Coupon ................ 112

Figure 6-9: EP3 vs . Distance f?om Notch Root - Mid-Thickness of Low Kt Coupon .... 113

Figure 6-10: EP3 vs . Distance £?omNotch Root - Surface of Low K, Coupon .............. 113

Figure 6-1 1: von Mises Stress vs. Distance f70m Notch Root - Mid-Thickness of Low Kt

Coupon ................................................................................................................. 114

Figure 6-12: von Mises Stress vs . Distance fiom Notch Root - Surface of Low Kt

Coupon ...............:................................
................................................................ 114
Fiame 6-13: SP3 vs .Distance from Notch Root - Mid-Thickness of High Kt Coupon . 115

Figure 6-14: SP3 vs . Distance fiom Notch Root - Surface of High Kt Coupon.............. 115

Figure 6-1 5: EP3 vs . Distance frorn Notch Root - Mid-Thichess of High K, Coupon . 116

Figure 6-16: EP3 vs . Distance from Notch Root - Surface of High Kt Coupon ............. 116

Figure 6-1 7: von Mises Stress vs . Distance fiom Notch Root - Mid-Thickness of High Kt

Coupon .................................................................................................................... 117

Figure 6-18: von Mises Stress vs. Distance fkom Notch Root . Surface of High Kt

Coupon .................................................................................................................... 117

Figure 6-19: Notch Root Stress vs .Net Section Nominal Stress - Low Kt Coupon ....... 118

Figure 6-20: Notch Root Strain vs .Net Section Nominal Stress - Low Kt Coupon ....... 118

...
Xlll
Figure 6-2 1:Notch Root Stress vs.Net Section Nominal Stress .High Kt Coupon .....-1 19

Figure 6-22: Notch Root Strain vs.Net Section Nominal Stress .High Kt Coupon ...... 119

Figure 7-1 : Low Kt Coupon .Stress Estimation ............................................................. 120

Figure 7-2: Low Kt Coupon .Strain Estimation ............................................................. 120

Figure 7-3: Hi& Kt Coupon .Stress Estimation ................*........................................... 121

Figure 7-4: High Kt Coupon - Strain Estimation ............................................................ 121

Figure 7-5: Low Kt Coupon .Crack Initiation Prediction Sensitivity Study .................. 122

Figure 7-6: High Kt Coupon - Crack Initiation Prediction Sensitivity Study ................. 122

Figure 8-1: Stress Intensiq Factor vs .Crack Length ................................................ 123

Figure 8-2: Low Kt Coupon - Prestrain and Non-prestrain LS Predictions .................... 123

Figure 8-3 : Equivalent Strain Equation Study - High Kt Coupon................................... 124

xiv
List of Appendices

Appendix A Coupon Test Program........................................... 125

O

Appendix B = Format of McCracken hput and Results Files ....138

Nomenclature
Crack Initiation Life

Damage surn

Increment of damage

Design Limit Stress

Elastic moddus, Secant modulus

Principal strains, ABAQUS naming system

Dimensioniess function of geometry (Equation 8- 1)

Cyclic hardening coefficient (Equation 2-4)

Fatigue concentration factor

Theoretical stress concentration factor

Equivaient stress concentration factor (Equation 7-3)

Local strain concentration

Local stress concentration

Cycles to failure of smooth specimen

Nurnber of cycles to failure at load level i (Equation 2-1)

Number of load excursions at load level i (Equation 2-1)

Test result for sequence A (Equation 2-2)

Improved prediction for sequence B (Equation 2-2)

Predicted lives for sequences A and B (Equation 2-2)

Strain ratio

Stress ratio

Net-section nominal stress

Principal stresses, ABAQUS naming system

Fatigue strength

Damage parameter (Equation 2-1 3)

Petersons's material constant (Equation 2-23)

Elastic stress ratios (Equations 7-1 and 7-2)

Fatigue strengtli exponent (Equation 2-9)

Fatigue ductility exponent (Equation 2-9)

Notch depth (Equation 8-1)

Net section nominal strain

Size of crack at initiation (Equation 8-1)

Number of load levek (Equation 2-1)

Cyclic hardening exponent (Equation 2-4)

Notch radius

Coordinate directions

Principal m i n ratio

Principal plastic strain increment, i = 1,2,3 (Equation 2-1 9)

Equivalent plastic strain increment (Equation 2- 19)

Notch root strain

 Fatigue ductility coefficient (Equation 2-9)

Principal strains

Principal plastic strains, i =1,2,3 (Equation 2-20)

Equivalent plastic strains, i =1,2,3 (Equation 3-20)

Poisson's ratio and modified Poisson's ratio (Equation 7-7)

Notch root stress

Principal elastic messes

Deviatoric stress, i = 1,2,3 (Equation 2-19)

Fatigue strength coefficient (Equation 2-9)

Principal stresses

Hydrostatic stress tensor

Stress tensor

Deviatoric stress tensor

Prefixes:

A range

Subscripts:

U ultimate

Y yield

eff effective

eq, q equivalent

xviii
ma maximum or peak

O mean

PL plastic

Superscripts:

(notched) notched specimen

(smooth) smooth specimen

xix
Units
In keeping with the practices of the North Amencan aerospace industry and the hstitute

for Aerospace Research (IAR) at the National Research Council of Canada (NRC), the

British Lmperiai system of units is employed in this thesis. Système International

d'unités (SI) equivaients are provided in brackets within the text where practicai. The

following conversion factors are useful:

1 inch = 25.4 mm

1 lbf = 4.4482 N

1 ksi = 6.8948 MPa

1 kip = 1000 lbf

 -
Chapter 1 Introduction
Meta1 fatigue is a process which causes the failure of an engineering component subjected

to repeated loading. Typical engineering structures are cornplex, and are subjected to

irregular load histories. This? added with the cornpiex nature of the fatigue process,

makes it diEcuIt to accurately predict the life of a structure. Nevertheless, fatigue

analysis methods have been developed over the years to aid the design engineer. Today,

fatigue life prediction is a fundamental undertaking in the design of many cornponents

and structures used in the automotive, aerospace and offshore industries.

Fatigue is a primary mode of failure for aifiames. In general. fatigue cracks initiate

within the aifiame at points of stress concentration which c m occw due to a ïnaterial

flaw, or a geometric feature such as a cutout or a rivet hole. Unless detected by an

inspection prognm, these cracks may progress through the stnicture until failure occurs.

Thus. for convenience, the fatigue process is often divided into two phases: crack

initiation and crack propagation.

Fatigue life prediction methods in use today are based on the Nominal Stress (NS), Local

Strain (LS). and Fracture Mechanics approaches. The NS approach uses constant

amplitude stress-life c w e s to calculate the fatigue damage based on the nominal stress in
the component. A total life (initiation + propagation) prediction results fiom the use of

the NS approach. The LS approach differs fiom the NS approach in that the stress and

strain state at the notch is considered. The use of the LS approach results in a prediction

of life to crack initiation. Finally, the fracture mechanics approach predicts the growth of

a small crack to one which will cause failure of the component. An advantage to using

the fracture mechanics approach is that damage is quantified in terms of a visible

parameter> the crack length. This is in contrast to the NS and LS approaches where

damage is quantified in terms of a numerically calculated damage sum.

The local strain approach is typically used in situations where life is defmed as the onset

of detectable flaws. One example is in the design of the CF4 8 aircrafi. The local straui

approach is also being used in analysis work for the International Follow-On Structural

Test Program (Simpson, 1997). IFOSTP, as the project is known, is a full scale fatigue

test of the C F 4 8 airframe being conducted by the Canadian Forces (CF) and the Royal

Australian Air Force (RAAF). The a f t fuselage and empennage tests are the

responsibility of the Australians. while the wïng and centre fuselage are Canada's

responsibility. The centre fuselage test is currently underway at Bombardier hc.,

Canadair Defense Systems Division (BKDSD). Preparations are being made for the

wing test at the Structures, Materials and Propulsion Laboratory of the lnstitute for

Aerospace Research (SMPL-IAR) at The National Research Council of Canada (NRC).

The full scale test is performed by agplying a representative load history to the test article

through a system of hydraulic acniators. IFOSTP has adopted a load spec-, derived

fiom flight test data which represents 279 flights of combined CF and RAAF usage.

To reduce the testing tirne, a process known as tnincation is adopted whereby small load

cycles which do not contribute to fatigue darnage are removed fiom the load spectrum

applied to the test article. The SMPL-IAR is currently performing spectnim -cation

sensitivity studies to determine the level of truncation to apply to the wing load spectnim.

A local strain based cornputer program, C-CI89 (Klohr, 1990), has been adopted by

IFOSTP for use in the spectrum tnincation sensitivity tests being conducted at SMPL-

IAR. Variants of the -189 progam were used in the design of the CF-18, and are

currently used by the CF for fleet management purposes.

The locai strain method contains a nurnber of assumptions which can cause considerable

error in predictions. One assurnption is the use of Neuber's mle to estimate the stress and

strain at the notch root of a component. Neuber's rule was derived for a specific

geometry and loading, but is generally used unconditionally in the LS method. The

objective of this thesis is to anaiyze the applicability of Neuber's rule in the local strain

approach.

The layout of the thesis is as follows. Chapter 2 contains a review of research regarding

the prediction of fatigue crack initiation, includinp a review of Neuber's rule and its

limitations. Having established the background, Chapter 3 presents the project definition.
Chapters 4 through 7 describe the analyses performed in support cf the project definition.

A discussion of the results of the study is given in Chapter 8. Conclusions and

recommendations for future research are given in Chapter 9.

 -
Chapter 2 Review of Fatigue Crack Initiation

Prediction

2.1 introduction

This chapter gives an overview of the prediction of fatigue crack initiation. A review of

the Nominal Stress method will be given fmt to make clearer the discussions which

foliow. The Local Strain method will then be reviewed and areas of conceni will be

discussed.

2.2 Aspects of Fatigue Crack Initiation

Since the "initiation" of a fatigue crack is not a single physical phenomenon, it must be

arbitrarily defined by the user. The definition of fatigue crack initiation therefore varies

in the literature. For instance, it is defined as the number of cycles to grow a crack 2-3

mm long in (SAE, 1988). However, most aerospace related literature quote the crack

length at initiation equal to 0.01" (0.254 mm), e.g. (Baotong and Xiulin, 1993). The

definition of a crack length at initiation is limited by the ability of non desmctive

inspection (NDI) techniques to reliably fmd cracks in a structure.

A useful criterion for the assessrnent of fatigue life prediction concepts is the ratio of "test

results/prediction2 obtained Corn a large number of predictions. The perfect prediction

method would give a ratio of 1.0 every time. This is not achievable due to the complex

nature of the fatigue process and the large number of simpli@ing assumptions present in

fatigue 1ife prediction methods. According to Buch (1980): a prediction concept works

sufficiently well if the ratio for al1 predictions lies within the rage of 0.5 to 2.0.

2.3 Nominal Stress (NS) Approach

The NS approach was the first fatigue life prediction method and is still used even though

more complex methods have been developed. Although the NS approach yields a total

life (initiation + propagation) estimate, a review of this method will make cleârer the

discussion of the LS approach which follows.

The b a i s for the method is the stress-life. or S-N curve. The S-N curve is usually

generated by rotating bending tests which are performed for a number of stress ratios to

account for mean stress efTects. The tests are m until the specirnen ruptures. Stress

concentrations are taken into account by using S-N curves which are obtained for

different values of the theoretical stress concentration factor, Kt.

The Palmgren-Miner Rule (Miner's rule) is used to accouot for the variability of loading

with time. Miner's mle assumes that failure of the component occurs when the darnage

sum equals unity. The damage sum, D, is defined as the fraction of life used up by a

senes of damaging load excursions. The mle is expressed as:

where n is the number of load levels in the spectrum? i is the current load level, Ni is the

number of load excursions at level i, and Nti is the nurnber of cycles to cause specimen

failure at Ioad level i.

If the results of variable amplitude loading tests are available, an irnproved prediction can

be made using the "Relative-Miner7' approach (Heuler and Schütz, 1986). The Relative

Miner approach suggests that it is not necessary for the darnage surn at failure to be unit%

but only that the darnage surn at failure be the sarne for spectra which are similar.

Consider two spectra labelled "A"

and "B". The mle is expressed as:

wliere NA and NprcdAare the test result and prediction for spectrum A, NpredSB
is the

prediction for spectnun BI and Ne is the improved prediction for spectnun B. The

definition of the similarity of the spectra is open to interpretation, but can include sirnilar

peak values and global stress ratios.

The NS approach has several weaknesses which led to the development of the local strain

and fiacture mechanics approaches. The weaknesses outlined by Bannantine et al. (1990)

are:

1. The NS method is completely ernpirical in nature and lacks the physical

insights into the mechanisms of fatigue demonstrated by other methods.

2. The actual stress-stmk response of materials is ignored, thus mean residual

stresses resulting from sequential loading effects cannot be modelled. This

implies that the NS approach may have problems dealing with spectra which

are no t "close" to constant amplitude.

Even though the NS approach has senous shortcomings, it is still used since there is a

large amount of fatigue data available.

2.4 Local Strain (LS) Approach

2.4.7 Principle of LS Method

The LS approach was developed to overcome sorne of the problems inherent in the'^^
approach. The principle behind the LS approach, depicted in Figure 2-1, is that smooth

specimens tested under strain-control can sirnulate the fatigue darnage at the notch root of

an engineering component. Equivalent fatigue darnage is assumed to occur at the notch

root and in the smooth specimen when both are subjected to identical stress-strain

histories. This is known as the pBnciple of equivalence.

Since the smooth specimens are tested under strain control, the LS approach uses the

strain-life or E-N c w e . The LS approach is considered to be an estimation of life to

crack initiation since it is assumed that once the equally stressed volume of material in the

srnooth specimen fails (Figure 2 4 , the equally stressed volume in the notched specimen

will fail. ïherefore? cycles to failure (specimen rupture) of the smooth specimen is

considered to be equal to cycles to crack initiation of the notched specimen.

2.4.2 LS Mefhod

The LS approach estimates the fatigue crack initiation life for a notch located in a

component subjected to variable amplitude loading. The LS method is composed of four

steps:

1. Notch Stress and S train Calculation

2. Cycle Counting

3. Mean Stress Correction

4. Darnage Calculation

The method tracks the notch root stress-strain response to identiQ damaging events by

use of a cycle counting procedure. For each damaging event, the effect of the mean stress

is accounted for by an equivalent strain equation. The damage for each event is then

calculated fiom the matenai strain-life curve. A description of the four steps follours.
Step I - Nor& Stress-Sfrain Calculation:

The first step in the LS approach is to establish a relationship between the net section

nominal stress range and the local stress-strain ranges at the notch root of a component.

This couid be accomplished numencally using finite element methods (FEM), or

experimentally using strain gauge readings. Both of these approaches are usually dropped

in favour of approximate relationships. such as those reviewed by Seeger et al. (1977).

Due to its simplicity, the most widely used of these relationships is the one proposed by

Neuber (1961). Known as Neuber's nile, it states that the geometric mean of the stress

and strain concentration factors is equal to the theoretical stress concentration factor.

Generally, this is expressed in tems of stress and strain ranges for the case when the

stress range remote to the notch is linear elastic. Neuber's rule has the following form:

where do and A s are the notch root stress and strain ranges respectively, Kt is the

theoretical stress concentration factor, AS is the net-section nominal stress range, and E is

the elastic modulus of the material. Equation 2-3 is solved using the material stress-strain

curve to calculate the notch root strain fiom the applied stress. In fatigue loading, the

cyclic stress-strain curve obtained fiom cornpanion samples or the incremental step test is

used to calculate stress and strain amplitudes, whereas the hysteresis c u v e is used to

calculate stress and strain ranges. Massing (1926) proposed that the hysteresis curve is

twice the cyclic curve if the tensile and compressive responses of the material are
identical. When the cyclic stress-strain c u v e fiom tests is not available, the following

Rarnberg-Osgood approximation may be used:

where AG and A& are the notch root stress and strain ranges respectively, K? is the cyclic

hardening coefficient, n' is the cyclic harde- exponent, and E is the elastic rnodulus.

Step 2 - Cycle Counting:

A darnaging event is identified by use of a cycle counting procedure. Standard practices

for cycle counting in fatigue analysis are detailed in ASTM Standard E 1049-85 (ASTM,

1995). Counting procedures such as the Rainflow method and its derivatives are

considered to be superior since they are able to identiQ the overall largest cycle in the

spectrum.

The notch root stress-strain calculation and the cycle counting steps are generally

performed simultaneously. The combined procedure for cycle counting and notch stress-

strain estimation is best explained using an example problem. The procedure uses

Neuber's rule to calculate the notch root stress and strain fiom the applied load spectrum

and the material stress-strain curve. The notch root response is tracked to identi& closed

hysteresis loops. A derivative of the Rainflow method called Closed Hysteresis Loop

Counting will be used.

Cornbined rnethod for notch root stress-main calculation and cvcle cozrntina:

Figure 2-2 presents a typical spectnim and cyclic stress-strain curve for a component.

The application of load A causes the notch root stress-strain response to reach point A in

Figure 2-3[A]. The notch root stress and strain are calculated using the cyclic cuwe and

Neuber's mle. The application of loads B and C follow the hysteresis cuve until points

B and C are reached in Figure 2-3 [^] . The stress and strain ranges, AB and BC, are

calculated usïng the hysteresis curve and Neuberos rule. Up to this point, a closed

hysteresis loop bas not been identified.

The application of load D causes an unloading from C (Figure 2-3@3]). The stress-strain

path follows the hysteresis curve fiom C until B is reached. Point B corresponds to the

previous largest valley load. The stress-strain path fiom B to D is calculated as an

extension of the path from A to B. This phenornenon is known as "material merno@',

and restncts the stress-strain paths fiom crossing each other. Loop BC is closed, and a

cycle of magnitude BC is counted.

n i e application of load E is s h o w in Fi=we 2-3 [Cl. The stress-strain path foIlows the

hysteresis curve from D until A is reached. Point A corresponds to the previous largest

peak load. The stress-strain path fiom A to E is calculated as an extension of the path

fiom the origin to A. The cyclic curve is used to characterize tliis portion of the notch

stress-strain response. Again, material rnemory restncts the stress-strain paths fiom

crossing each other. Loop DA is closed, and a cycle of magnitude DA is counted.

Application of load F is computed in the same marner as for load B, and is shown in

Fi,pre 2-3 [Dl.

Step 3 - M e m Sîress Correction:

The combined procedure described above is used to identm closed hysteresis loops. For

each closed hysteresis loop the effect of mean stress is accounted for by adjusting the

strain amplitude of the loop so that it represents some "equivalent" strain amplitude at a

stress ratio of -1. This is accomplished using equivaient strah equahons, such as

those listed by Forness et al. (1989):

SWT:

Goodman: 2
=
(, -00)

Gerber:

-
Soderberg: 2
where Ad2 is the strain amplitude, o,, is the maximum notch root stress, o, is the mean

stress' GY is the material yield stress, ou is the matend ultimate stress? and the subscript

"eq" denotes the equivalent strain amplitude.

The Smith-Watson-Topper (SWT) equation is comrnonly used in the LS approach. The

formulation of the SWT equation requires that the strain-life curve be modified to

represent a c u v e compatible with the SWT equation (Smith et al., 1970). Instead of

plotting strain amplitude versus life, the parameter ,/:$, is plotted versus [Xe. The

onginal strain-life curve and the SWT compatible curve are identical in the region where

the strain amplitude is in the elastic range of the material.

Step 4 - Dûmage Calczrlation:

Final15 the number of cycles to failure of the smooth specimen is calculated for each

closed hysteresis loop from an experimentally obtained strain-life curve (at %=-1). If the

SWT equation is used to correct for mean stress, the SWT compatible strain-life curve is

used. %%en the strain-life c u v e is not available fiom experiment, the following Manson-

Cofin relation may be used:

where Ad2 is the strain amplitude, cp' is the fatigue strength coefficient, Nf is the number

of cycles to failure of the smooth specimen, EC is the fatigue ductility coefficient, b is the

fatigue strength exponent. and c is the fatigue ductility exponent.

The darnage Iiom each loop (Di) is calculated as the inverse of the cycles to failure of the

smooth specimen (NKi):

The total damage @) for the component is then computed using Miner's linear damage

de:

Once the entire spectrurn is analyzed, the crack initiation life (CIL) of the notched

cornponent is calculated as the inverse of the total darnage of the component:

CIL = -
D
Heuler and Schütz (1986) identined two areas of concern for the LS method: the damage

accumulation and the principle of equivalence. These will be discussed in Sections 2.5

and 2.6, respectively-

2.5 Damage Accumulation

The first area of concem. darnage accumulation, &ses fiom predicting the fatigue

behaviour in spectnim loading by h e a r damage accumulation calculated fiom smooth

specimen constant amplitude strain-life data (at &=1). Some attempts to overcome the

damage accumulation problem have involved introducing a modified baseline damage

parameter curve. Ln this approach, the parameter, strain range, is replaced by quantities

which are assurned to descnbe the fatigue damage properties of the material under

constant as well as under variable amplitude loading- One such darnage pararneter, Zd, is

denved from shoa crack behaviour descnbed on the basis of the cyclic J-integral

approach (Heitmann et ai., 1983). The darnage parameter has the following form:

where Acerr is the effective stress range calculated as the difference between the applied

stress and the stress required to open the crack, E is the elastic modulus, n' is the strain

hardening exponent, Ac is the notch root stress range, and AspI is the notch root plastic

strain range. This damage parameter includes the effect of rnean stress which is modelled

on the basis of a crack closure argument.

Another approach to overcoming the darnage accumulation problem in spectnim loading

has involved the use of prestrained s&-life data. Prestraining refers to initial overioads

applied to the smooth specimens used to generate the strain-life curve. The rationale

behind the use of prestrained data in spectrum loading is to take into account the

influence of large cycles on the following smdler ones (Bergmann et al., 1979). Conle

and Topper (1980) demonstrated that the use of non-prestrained data led to non-

conservative life predictions in a variable amplitude loading study.

In variable amplitude loadhg, the peak load in the s p e c t m is applied once per block, but

strain-life data is generated through constant amplitude tests: or in some cases, with initial

prestrain. It has been suggested that a periodic overload must be applied to the smooth

specimen when generating the strain-life data to reduce the non-conservatism of

predictions made using non-prestrain data. Code and Topper (1980) report that

predictions made using periodically overstrained constant amplitude data closely

approximate the test results. but are still non-conservative. More recently, DuQuesnay et

al. (1995) report that the fatigue limit of srnooth specimens made fiom aluminum 2024-

T351 is significantly reduced when an overload of yield magnitude is applied

periodically.

Finally, nonlinear darnage accumulation mies have been proposed for use in place of

Miner's d e . Some of the nonlinear darnage accumulation d e s require new material

constants wliich must be determined fiom tests. This is a disadvantage when cornparhg
these models to Miner's nile. Schütz (1979) expressecl the opinion that nonlinear damage

niles had not reliably shown convincing improvements in prediction accuracy when

compared to Miner's d e . However, Bleuzen et al. (1994) have recently identified the

ONERA LS model with nonlinear darnage accumulation to be appropiate for predicthg

fatigue life under complex loading sequences.

2.6 Principle of Equivalence

2.6.7 Concerns

The second area of concern in the LS approach is the assumed eqiiivalence between the

smooth specirnen and the notched member. Concerns with the principle of equivdence

include:

Determination of stress and strain at the notch of a component: The stress and

strain at the notch must be known in order to calculate the fatigue damage of a

notched component from smooth specimen test data This presents a problem

since only the elastic solution is known exactly, and approximate relationships

such as Neuber's rule must be used in the plastic regirne.

Effect of notch severiw on fatigue life: Expenmental observations have shown

that stress concentrations in notched members have less eflect in fatigue than is

predicted by the stress concentration factor K,. This effect is dependent on

materid and the geometry of the notch.

Differences between test resuits and predictions can b e as large as an order of magnitude

due to limitations in the principle of equivalence. The two concems wiîh the principle of

equivalence will be discussed in Sections 2.6.2 and 2.6.3.

2.6.2 Notch Stress-Sfrain Estimation

The cornparison of the fatigue damage at the notch of a component and in the smooth

specimen used to generate the test data requires an estimation of the stress and strain at

the notch. Lf the loading is hliy elastic, then the exact solution can be found using

Hooke's law and the stress concentration factor Kt. When there is yielding at the notch,

the exact soiution is not known. and approximate relationships such as Neuber's d e , and

more recently, Glinka's Equivalent Strain Energy Density (ESED) method are used.

Approximate relationships contain assumptions which may not be valid in certain

situations. For instance, Neuber's rule does not mode1 the multiaxial stress state present

at the notch root of many engineering components. For this reason, many authors feel

that a detailed elastic-plastic finite element (FE) analysis is the best way to obtain an

accurate estimation of the notch stress and strain. The next sections will describe

Neuber-s rule. Glinka3 ESED method, and FE analyses in more detail as they apply to

local strain fatigue predictions.

2.6.2.1 Neuber's Rule and ifs Vatfations

Neuber showed that for a shear-strained prismatic body with an arbitrary non-linear

stress-strain law, the geometric mean of the stress and strain concentration factors (K,
and K 3 is equal to the theoreticai stress concentration factor? Kt. This is expressed as

follows:

The stress concentration factor, GT

is the ratio of the notch root stress and the net-section

nominai stress:

The strain concentration factor, K,, is the ratio of the notch root strain and the net-section

nominal strain, e:

Combining equations 2- 14 through 2- 16, Neuber's mle is re-written as:

Generally, this is re-written in terms of stress and strain ranges for the case when the

stress range remote to the notch is linear elastic:

n i e form presented in Equation 2-18 is widely used in fatigue life calculations using the

LS approach. It shows that the product of the notch stress and strain ranges can be

estimated by knowing the theoreticai stress concentration factor. the appIied stress range.

and the elastic modulus of the material-

Although Neuber's rule was denved for a monotonic loading case, it was applied to

fatigue Loading by Manson and Hirschberg (1966). They suggested using the cyclic

stress-strain curve instead of the monotonic curve for fatigue loading. Topper er al.

(1969) were the first to refer to Neuber's work as 'Neuber's rule". They showed that

smooth specimen fatigue data could be used to adequately predict fatigue iives of notched

rnembers made from 2024 and 7075 duminum alloys under fully reversed loading.

Many authors have questioned the unconditional use of Neuber's rule in fatigue life

predictions, for example, Tipton (1991) and Glinka et al. (1988). The grounds for this

challenge are the following:

Neuber's rule was derived for a specific notch geometry and loading condition,

but is used without question in many cases where the loading and notch

geometry deviate substantially fiom Neuber's original work.

 It has been found in mmy cases to predict larger notch root strains when

compared to finite element studies or direct measurernent. resulting in overly

conservative predictions.

Neuber derived his original work for a hyperbolic notch profile Ioaded under two-

dimensional monotonie shear. The shear stress and strain at the notch root were the only

non-zero stress and strain components, far fiom being a general solution to the notch

stress-strain problem. Neuber recognized this limitation in his original work, and gave

suggestions as to how the formulation may be generalized for arbitrary stress states by

means of one of the well known theories of failure.

Tipton (1991) suggested that the conservative nature of Neuber's mle was due to the

mdtiaxial stress state at the notch root which is not accounted for in the approximation.

When a notched engineering component is Ioaded in tension. the material at the notch

root tries to contract perpendicular to the axis of loading. The Iesser stressed bulk

material on either side of the notch prevents this fiom occumng, and a transverse stress at

the notch root is created.

Hoffmann and Seeger (1985) developed a generalized method for estimating the

multiaxial elastic-plastic notch stresses and strains using a modularïzed approach. Their

method closely follows the recommendations set by Neuber in his original work and

consists of two steps:

 1. Relating the applied stress (S) to the equivalent notch stress and strain (o,and

~q)-

2. Relating the equivalent stress and strain to the principal stress and strain (oi

and si).

The fxst step is accomplished by using Neuber's rule? but replacing the uniaxial

quantities with an equivalent quantity calculated under the von Mises or Tresca yield

criterion.

The theory of plasticity offers two possibilities to correlate the equivalent stress and strain

to the principal stress and strain. The stress tensor [a]is often divided into a hydrostatic

stress tensor [pIj and a deviatoric stress tensor [d].

The flow rule of Prandtl-Reuss

relates the principal plastic strain increments dsiP to the deviatoric stresses ci',the

equivalent plastic strain increment dsqP,and the equivaient mess o,. The Prandtl-Reuss

flow mle has the following form:

These equations represent the exact solution for isotropie hardening under the assurnption

of von Mises yield critenon. If the ratio between the deviatoric stress components
remains constant during loading, the Prandtl-Reuss law reduces to Hencky's rule which

assumes the principal plastic strains ciPto be a function of the deviatoric stresses:

Hoffmann and Seeger (1985) assume that the ratio between deviatoric stress components

does not change substantially in the vicinity of the notch. Thus, Hencky's d e will

deliver results with sufficient accuracy. Hoffmann and Seeger have dernonstrated the

success of their method for thick bars and thick pressure vessels.

2.6.2.2 Equivalent Strain Energy Density (ESED)

Altemate approxirnate relationships have been proposed for use in place of Neuber's rule.

One such relationship is the Equivalent Strain Encra Density (ESED) method proposed

by Molski and Glinka (1981). In this approach, it is assumed that the strain energy

density at the notch root does not change significantly if the localized plasticity is

surrounded by predominantly elastic material. In other words, the computation of the

strain energy density at the notch root will yield identical results for either the elastic or

the elastic-plastic material Iaw. The ESED relation has the following form:
where Ki is the theoretical stress concentration factor, S is the net-section nominal stress,

E is the elastic moduius. q,,is the notch root stress as a function of E, the notch root

strain.

According to Glinka el al. (1988): the ESED mode1 is supenor to Neuber's rule when

predicting elastic-plastic notch stresses and strains. Sharpe el al. (1992) make the

assertion that Neuber's mle is the single best model for cases which are plane stress. It

was demonstrated in some cases that the measured or FE results for local strain lie

between the predictions fiom the Glinka ESED mode1 and Neuber's d e , with Neuber's

ruLe giving the larger estimate for a given value of K& e.g. (Molski and Glinka, 1981),

(Sharpe et aL. 1992). In these instances, it was difficult to determine which model best fit

the experimental data. It has been suggested that estimations made fkom Glinka's ESED

model and Neuber's rule will give lower and upper bounds on the local strain, which can

be used to tag an uncertainty with the life prediction (Sharpe et al., 1992).

2.6.2.3 Finite Element (FE) Analysis

Many authors feel that a detailed elastic-plastic finite element analysis is the best way to

obtain an accurate stress-strain estimation at the notch root of an engineering component.

However, elastic-plastic FE analyses are far fiom being ready to use in everyday design

applications, especially in cases of complex geometry, irregular loading and realistic

transient rnaterial plasticity behaviour (Tipton, 1991). Even though this is the case, FE

analyses are attractive since the Ievel of detail used in the model c m be tailored to a

particular research program. For instance, some authors have opted to model a multiaxial
stress state at the notch root (Tregonuig? 1992), while others have included the effects o f

creep (Umeda et al., 1987).

The Ioading used in an FE analysis is generally monotonic (using cyclic material data),

which is analogous to Neuber's d e which was formulated for monotonic loading, but

extended to fatigue loading. FE analyses which use a representative load history are not

common since they are time consurning?and are not possible in many FE packages.

Some researchers have used FE packages which have an integrated fatigue life program,

such as the PC-based NISAENDURE (Kahonen, 1991) and the MSCEATEGUE

(Doerfler, 1997) programs. With these integrated packages, aifiame cornponents can be

subjected to representative load histones to identify fatigue critical areas.

Doedler's study showed that MSCFATIGUE could benefit the preliminary design of

airfiame structures since it has the ability to locate areas in a structure which may be

susceptible to crack initiation. This global ability does corne at a high price when one

considers the number of load cases that are studied in a typical aircrafi development

program.

Combined with the results from coupon and component test prograrns, packages such as

NISA/ENDURE and MSCFATIGUE could become valuable tools in an engineering

design office.
2.6.3 Notch Severity

As mentioned previouslyoexperimental observations suggest that stress concentrations in

notclied members have less effect in fatigue than is predicted by the stress concentration

factor Kt. A so-called fatigue concentration factor, Kr is ofien used in place of Kr to

represent this phenornenon. The fatigue concentration factor is defined as the ratio of the

fatigue strengths of smooth and notched specimens, Se(mooth) and s,(notckd) , respectively.

The fatigue concentration factor is expressed as:

In order to calculate Kf fiom Equation 2-22, it is necessary to have the fatigue strengths of

both smooth and notched specirnens available fiom experirnent. Usually, Kf is calculated

at N = ~ o 'cycles for a stress ratio of &=-

1 (Nie et al., 1994). Since experimental results

are tirne consuming and expensive to obtain, empirical formulae such as those listed by

Weixing et al. (1995) are used to calculate Kr. The formula proposed by Peterson (1959)

is popular and has the following form:

where "a" is a material constant, and r is the notch radius. The accepted rationale behind

using Kf instead of Kt is to account for size effects at notches. The size effect is depe~dent
on material and stress gradient at the notch. A sharp notch has a steeper stress gradient

and a smailer volume of highly stressed material when cornpared to a blunt notch (Figure

2-4). When a crack has grown outside the highly stressed material in the sharp notch

case, the crack growth slows d o m since the effect of the stress concentration diminishes-

In addition, as the volume of highly stressed materid increases, the probability of fatigue

failure increases since the likelihood of a flaw existing w i t b the Iarger volume is greater

than for a smaller volume @owling, 1979).

It is apparent that Kr calculated from Equation 2-23 is constant. However, Nie e f al.

(1994) propose a "Variable Kf " of the following fom:

where E and E, are the elastic and secant moduli. Their study showed that predictions

based on the variable KG the Manson-CoEn reiationship and Miner's rule were in

agreement with test results for a critical location of a wing spar made fiom AISI-4340

steel and subjected to spectrum loading.

Walker (1977) noticed that the application of Neuber's rule with the fatigue concentration

factor disregarded the multiavial stress state at the notch root. He demonstrated that the

success of this approach was not accidental, and that there was an equivalence between

the following two methods:

 1. Use of Neuber's rule with the fatigue concentration factor KG

2. Use of Neuber's rule with Kt and consideration of the rnultistxial stress state.

Hence, Walker's approach suggests that the use of Kf may be an empirical method of

accounting for notch root multiaxiality. However, this approach disregards the notion put

forth by many that Kr accounts for size effects.

Bannantine et al. (1990) noted that there appears to be a limiting value of Kr based on

observed behaviour. The limiting value is dependent o n material, but has been s h o w to

be approximately 5 or 6, even though K, could be as large as 15. Bannantine et al.

propose to use Rf to account for the smaller initiation life in sharp notches, where a

si_pificant crack propagation stage may exist. Thus, for sharp notches, local strain

predictions made using Kr would be in better agreement with total life test results when

compared to local strain predictions made using KI. This suggests that the use of Kr may

be an empincal method to account for crack propagation in sharp notches.

2.7 Fracture Mechanics

A new area of fatigue research involves the use of fracture rnechanics approaches to

characterize crack initiation. Linear Elastic Fracture Mechanics (LEFM) can only be used

when the stress state at the crack tip can be described by an elastic stress field. This

lirnits the applicability of LEFM to low stress levels and reiatively long initial defects on

the order of 1 mm (Miller, 1987). LEFM is unable to characterize the behaviour of short
cracks since the stress field at the tips of small defects in highly stressed matenals cannot

be described by an elastic stress field. Elastic-Plastic Fracture Mechanics (EPFM)

approaches are beuig researched to simdate the growth of short cracks &om the level of

microstructure up to sizes compatible with LEFM.

Shoa crack behaviour is important when studying notches since a crack emanating fiom a

notch will be growing in a stress field which is usually above the LEFM lirnit. In

addition. as the crack grows out of the highly stressed material at the notch root. crack

grow-th rate slows d o m , possibly to the point where the crack becomes non-propagating.

n i e local strain approach was developed to overcome some of the problems identified

with the nominal stress approach, narnely its empirical nature, and the disregard for the

actual stress-strain response of the matenal.

The basis for the LS approach is that srnooth specimens tested under strain-control c m

simulate the fatigue damage at the notch root of an engineering component. Equivalent

fatigue damage is assumed to occur at the notch root and in the smooth specimen when

both are subjected to identical stress-strain histories. The local strain approach is

considered to be superïor to the NS approach since it considers the stress and strain state

at the notch root of a component. However, the local strain method contains assumptionç

that are not necessady valid. Two particular areas of concem are the darnage

accumulation and the principle of equivalence.

Attempts to overcome the damage accumulation problem have resulted in nonlinear

damage theories, use of prestrained and periodicaily overloaded material data, and the use

of rnodified baseline damage parameters.

Limitations in the principle of equivdence c m lead to order of magnitude differences

between measured and predicted life. Neuber's rule was derïved for a state of plane stress

but is generally used unconditionally in situations which deviate from plane stress. An

altemate approach. Glinka's ESED method. has been proposed which considers the strain

energy density at the notch root. Approximate relationships such as Neuber's nile and

Glinka's ESED rnethod are usehl to make first estimates of the notch root stress and

strain, but a more accurate prediction would require M e r computational andor

expenmental study of the case at hand.

The observed behaviour of the fatigue strengths of notched and smooth specimens has led

to the use of a fatigue concentration factor (Kr) in LS predictions. The use of Kf may be

an empirical rnethod of considering the crack propagation stage in sharply notched

meinbers,

Current research focus in the area of fatigue crack initiation lies in using fracture

mechanics to describe the growth of a microstructuraI flaw to a crack of suficient size,

compatible with LEFM.

 -
Chapter 3 Project Definition
In Chapter 2, it was established that there are areas of concem in the LS approach. One

concern is the use of Neuber's rule (Neuber, 1961) to describe the notch root behaviour of

an engineering component. As was seen. Neuber's rule provides an easy method for

estimating the notch root stress and strain of the component. Neuber's rule was denved

for a specific geometry and loading, but is generally used without question in many cases

which deviate substantially fkom the specific case for which it was derived. It has been

shown to be accurate in situations which are plane stress' but has been found to

underestimate notch root strains in situations where a multiaxial stress state exists. In

addition, Neuber's rule only represents the stress concentration of the notch through the

use of the stress concentration factor, and does not take into account factors such as strain

gradient and size effects. With these points in mind, it becomes apparent that Neuber's

rule is limited in its application, and may be inappropriate to use in certain situations.

The objective of this thesis is to analyze the applicability of Neuber's d e in the local

strain method. The predictive capability of Neuber's rule will be exarnined for two

coupons, one having a low stress concentration, the other having a Iiigh stress

concentration. The ability of the local strain method to predict the lives of the two

coupons will be assessed.

To meet the objective, three tasks were carried out:

Completion of a coupon test program to obtain total life test data for the two

coupons subjected to spectrum loading (Chapter 4).

Development of a local strain crack initiation prediction program simiiar to

CC189 so that LS predictions could be made using different implementations

of the notch root stress-strain estimation (Chapter 5).

Developrnent of elastic-plastic finite element rnodels of the two coupons to

observe the notch root stress and strain as a fùnction of the net section nominal

stress (Chapter 6).

The applicability of Neuber's rule in the LS approach was tested by completing a notch

root stress-strain sensitivity study. This involved cornparhg the stress-strain estimation

from Neuber's rule to the results of the FE analysis and other approximate relationships.

The ability of the local strain method to predict the lives of the two coupons subjected to

spectnun loading was tested by completing a crack initiation prediction sensitivity study-

This invoived comparïng LS predictions with the coupon test results fiom Chapter 4.

Local strain predictions were made for each of the methods used in the notch root stress-

strain sensitivity study. The results of the two sensitivity studies are presented in Chapter

7.
Chapter 4 = Coupon Test Program

4.1 Introduction

A coupon test program was undertaken at SMPL-IAR to test two coupon geometries

subjected to a CF-18 trading edge flap hinge moment sequence. Two sequences were

tested on each coupon geometry, one containing rnanoeuvring loads? and the other

containing combined (manoeuvring and buffet) loads. n i e tests were performed as a part

of a spectmm truncation sensitivity study at SMFL-LAR.

4.2 Coupons and Test Sequences

The coupons were manufactured fiom aluminum 7050-T745 1. The fkst geometry, "Low

Kt". contained a double edge semi-circular notch (Figure 4-1). The second geometry.

"High K,'?. contained a centrally Located slot (Figure 4-2). The coupons wil1 hereon be

referred to as Low Kt and High K,

The coupons have the general designations of "P1L 548 YYY 030": and "P2L 540 YYY

OjO", respectively. The reference label, 'YYY', refers to a coordinate within the block of

aluminum fiom which the coupons were machined. The reference label is unique for

each coupon. Only the reference labels will hereon be quoted when referring to a

particular coupon.
The sequences tested were representative of a trailing edge flap hinge moment sequence

on the CF-18 aircraft. The sequences were developed by Bombardier Inc., Canadair

Defense Systems Division (BVCDSD). Each CF- 18 is equipped with a Maintenance

Signai Data Recording System (MSDRS) which records flight pararneters. engine data

stores data, weapons data, and seveo channels of strah data autornatically at specified

fiequemies and on some specified events (Hewitt et al.. 1996). Since the tmiling edge

flap hinge moment is not directly measured by the MSDRS, it is predicted fiom a

regression of measured flight loads against measured flight pararneters through a

Parametric Load Formulation (PLF). Buffet loads are also not measured by the MSDRS.

The buffet loads were measured during a series of CF-18 flights and were added to the

manoeuvre spectrum to obtain the combined (manoeuvre + buffet) spectnim. The test

sequences represent 279 flights of combined Canadian Forces and Royal Australian Air

Force usage.

The first sequence? designated tef-rnan05. contained manoeuvring loads only, and was

bnincated at 5% peak load range, resulting in 17534 turning points per block. The number

of tuniing points corresponds to the number of load reversals in the spectnim. The

second sequence, containing the combined (manoeuvrïng and buffet) loads and

designated tef-sum05, \vas also truncated at 5% peak load range, resulting in 409380

turning points per block. In each instance, one block represents 279 unique flights with a

total duration of 326 spectnim fli@ h o u s (sfh). Five coupons of each geometIy were

tested for each sequence, resulting in a total of 20 fatigue tests.

4.3 Test Procedure

A test procedure was developed using the MTS Teststar and TestWare SX software. The

details of the test procedure can be found in Appendix A. Two load frames were used,

each dedicated to a particular coupon geornetry- A 55 kip (244 Idu) load frame was used

to test the Low K, coupons. and a 22 kip (97.9 kN) load fiame was used to test the High

Kt coupons. The apparatus set-up and calibration information can also be found in

Appendix A.

Initially. two coupons of each geometry were tested under the tef-man05 sequence so that

the stress level for the test could be verified. The sequences and coupons were selected at

random for the subsequent tests.

4.4 Detection of Crack Initiation

Pnor work at SMPL-IAR on tnuication sensitivity for the CF-18 centre fuselage

spectrum. used acoustic emissions monitoring to estimate the crack initiation lives on

some specimens. This study showed good correlation between initiation lives and total

lives. Therefore, since crack initiation measurements are time consurning, and there was

an urgency to compIete the testing, SMPL-IAR decided not to measure initiation lives on

the specimens tested with the trailing edge flap sequences.

In addition, SMPL-IAR was p h & l y interested in predicting the relative effects of

spectrurn truncation rather than absolute lives and thus the distinction between initiation

and total Iife was initially not a concern for them.

4.5 Test Results

Test results for the Low Kt coupons are presented in Table 4-1. The teman05 sequence

was tested using the peak load levels of 9.04/-4.85 kips (40.2/-21.6 kN). The load level

was selected by SMPL-IAR so that the tested life was approximately 2.5 times the

expected a i b e lifetime. In the case of the tef sum05 sequence, the peak load levels

were 10.05/-4.85 kips (44.7/-21.6 kN). The load level for the tef-su05 sequence was

selected by SMPL-IAR so that the peak manoeuvring stress was the same as in the

tef-man05 sequence.

1 YWI 1 Turning ( Missedl

Spectrurn

tef-man05
1
083 1
Start

14/01198, 1150
End

15/01198
Points
796735
1 ErrorTol
1 0190 '/5-
Blocks Hours Crack

Log Mean
tef-sumo5 1 On 1 16/01/98, 09:32 1 18/01/98 1 4571310 1 0190
409380 311 20/01/98, 1O:45 22101198 4869689 0190
pts. / block 257 24/01/98. l3:28 26/01198 3937343 O190

Loa Mean

Table 4-1: Test Results for Low Kt Coupons

Test results for the Hi& Kt coupons are presented in Table 4-2. The tef-man05 sequence

was tested using peak load levels of 7.27/-3.90 kips (32.Wl7.3 kN). In the case of the

tef-sumo5 sequence, the peak load levels were 8.08/-3.90 kips (35.9/-17.3 W. These
load levels were selected to obtain equal lives with respect to the Low Kt coupons. A

multiplier of 80.44% was used with respect to the Low Kt Load Ievel.

Table 4-2: Test Results for High Kt Coupons

It should be noted that the first three High KI coupons, 460, 454, 377, were tested before

the LVDT and load cell were re-calibrated on 22 and 23 January 1998, respectively. The

change in calibration was minimal: but required the MTS system to be re-tuned.

4.6 Accuracy of Applied Loads

The performance of the MTS system was tracked during the course of the testing to

determine the accuracy with which the loads were applied to the coupons. Quantities of

interest were missed end leveIs and the relative error between cornmanded and attained

load. Missed end levels is a MTS specific term which refers to the number of turning

points in the spectrum which were not applied to the coupon within a specified threshold.

In most cases, the threshold value was 90 lbf (400 N).

For the Low Kt coupon. the loads were applied with a very hi& degree of accuracy. For

the tef-man05 sequence. no end levels were rnissed, indicating that all of the loads in the

spectrum were applied to the coupons within the threshold of 90 lbf (400 N). For the

tef-sumo5 sequence, only specimens 131 and 239 had missed end levels, corresponding

to 1 end level and 12 end levels respectively. The missed end levels are insignificant in

these cases since the spectrum size 1s 409380 turning points per block, and the resulthg

Iives for the coupons are 9.18 and 9.76 blocks.

n i e accuracy of the application of the peak load was very high when compared to the

commanded load for the Low K, coupon. As an example, specirnen 077. subjected to

tef-sumo5 had a peak load of 10050 lbf (44704 N) in tension and -4850 Ibf (-2 1574 N) in

compression. Six random measurements of the applied peak load resulted in an average

of 10079 lbf (44834 N) in tension and -4855 Ibf (2LS96 N) in compression. The relative

error when compared to the comrnand was 0.29% and 0.10% respectively.

For the High Kt coupon. the situation is different. For the tef-sumo5 sequence, the

Largest number of missed end levels was 17136 for the first specimen tested, number 407.

In this case. only 19 end levels were missed during application of the first block, while

the MTS systern was in "leamhg mode". The majority of the 12136 missed end levels

occurred between blocks 14 and failure (16.35 blocks)? after the specimen had cracked

approximately one-third of the way across the net-section.

The tef-sumo5 sequence is significantly different than the tef~na.1105 sequence in that it

contains a large number of smail amplitude cycles which are applied very quickly since

the loading rate is constant. The large number of rnissed end levels for specimen 407 is

likely due to the MTS system undershooting the small amplitude cycles by more than the

tolerance of 90 lbf (400 N). This is due in part to the loading rate being constant as

explained above, but also due to the fact that the coupon was cracked when the majority

of end levels were missed. The MTS system "learns" how to apply the spectnun to the

coupon in the fust block when the coupon is not cracked. In this leamhg mode, the MTS

system generates a computer file which is used d e r application of the first block. The

computer file allows the MTS system to anticipate the next load demand. When cracking

occurs, the computer file is no longer valid since the stiffness of the coupon is not the

sarne as when it was not cracked. Thus? for specimen 407. the missed end levels are due

to the MTS system undershooting the small amplitude cycles due to the loading rate being

constant and the stifiess being reduced fiom the un-cracked case.

For the tef-man05 sequence, the largest nurnber of missed end levels was 1193 for

specimen 40 1. This coupon was the fiat to be tested afier re-calibration of the LVDT and

load cell. In addition. the error tolerance was lowered fiom 100 lbf (445 N) (used for

specimens 160,454, and 377) to 90 lbf (400 N). The re-calibration and the change in the

error tolerance reqiiired the MTS system to be re-tuned. The majority of missed end

levels occurred during the re-tuning effort, while the MTS system was in "leaming

mode'?.
The accuracy of the application of the peak load was very high when compared to the

commanded load for the High Kt coupon. As an exarnple, specimen 395' subjected to

tef-sumo5 had a peak load of 8084 lbf (35959 N) in tension and -3901 lbf (17353 N) in

compression. Seven random rneasurements of the applied peak Ioad resulted in an

average of 8122 Lbf (36128 N) in tension and -3903 lbf (17361 N) in compression. The

relative error when compared to the comrnand was 0.47% and 0.05% respectively.

As a whole. the loads were accurately applied to both the Low and High Kt coupons.

4.7 Crack Initiation Sites

Tables 4-1 and 4 - list the crack initiation sites, by number? for each of the coupons

tested. The numbers refer to the locations defined in Fimure4-3. The site fkom which the

crack initiated was determined by a visual inspection of the fracture surface.

For the Low Kt coupon. the dominant failure mode was a single corner crack. in the case

of specimens 083 and 239. there is evidence of cracks ïnitiating fiom two corners of the

net-section. For the High Kt coupon, the dominant failure mode was a double edge crack

with initiation at the mid-thickness of the coupon.

A coupon test program was undertaken as part of a spectrum truncation sensitivity study

at SMPL-IAR. The following resuits were obtained:

Total life results are available for the Low and High Kt coupons subjected to the

tef~nan05and tef-sum05 sequences. A sample size of five specimens per

sequence was used for each coupon geornetry.

For the Low Kt coupon, the application of the tef-man05 sequence results in a life

of 14232 sfh (iogarithmic mean), while application of the tef-sumo5 sequence

resuits in a life of 3339 s£h (logarithmic mean).

For the High Kt coupon, the application of the t e m a n 0 5 sequence results in a life

of 25180 sfh (logarithmic mean), while application of the tef-sum05 sequence

results in a life of 5308 s a (logarithmic mean).

 -
Chapter 5 Local Strain Software

5.1 Introduction

As mentioned in Chapter 1, the local strain cornputer program. C-CI89. is used in

IFOSTP. This chapter presents a description of the local strain method used in C-CI89.

In addition, a description of the local strain prograrn designed for this study will be

presented. FinaIly. differences between -189 and another local strain program,

LOOPINI, will be highlighted.

5.2 Description of C-Cl89

The LS method used in -189 is composed of four steps (Fomess et aL, 1989):

1. Cycle Counting - Closed Hysteresis Loops

2. Notch Stress and Strain Calculation - Neuber's Rule

3. Mean Stress Correction - EquivaIent Strain Equations

4. Darnage Calculation - Miner's Rule

The LS approach used in CC189 is similar to the method outlined in Chapter 2. Data

used in the C-CI89 LS approach are the following:

 1. Applied stress sequence manged in peak-valley pairs.

2. Material cyclic stress vs. stress*strain curve

3. Material strain amplitude vs. life curve at k-

1

4. Material elastic rnodulus and proportional limit

5.3 McCracken Fatigue Life Prediction Program

To meet the project objective defmed in Chapter 3' it is necessary to study different

irnpiementations of the notch root stress and strain estimation used in the local strain

approach. The computer prograrn C-CI89 is proprietary and is neither clearly wrïtten nor

weli documented. Making changes to C-CI89 to explore different implementations of the

notch root stress and main was therefore considered undesirable. A computer program

similar to C-CI89 was developed for this research prograrn with the following purposes:

1. To aid in the understanding of the local strain approach, and

2. To serve as a platform upon which capabilities not found in CC189 could be

added.

The McCracken Fatigue Life Prediction Program (hereafter referred to as McCracken)

was conceived of in response to these two points.

McCracken was irnplemented in Microsoft Visual C++ version 4.0, and is a Windows

95NT 4.0 application. McCracken is executed by running McCracken.exe fiom

Windows Explorer. The McCracken prediction environment is depicted in Figure 5-1.

Running a prediction using McCracken requires the completion of four steps:

1. Configure the material properties

2. Confiigure the Load spectnun

3 Configure the prediction

4. Go

A menu option and associated button on the toolbar are provided to allow the user to

complete these four steps.

5.3.1 Material Properties Dialog Box

The matenal properties dialog box is displayed in Figure 5-2. It allows the user to

confi,pe the material used in the prediction. The inputs in this dialog box are the cyclic

stress vs. stress*strain curve filename, the strain-life curve filename, the modulus of

elasticity. the proportional limit. and the ultimate tensile strength. The required format of

the material data fites is presented in Appendix B.

5.3.2 Spectrum Diaiog Box

The spectnim dialog box is displayed in Figure 5-3. It allows the user to configure the

load spectrum used in the prediction. The inputs in this dialog box are the load spectrum

filenarne, the reference value, the number of tuming points, the coupon design limit stress

(DLS), the stress concentration factor, and the number of flight hours per block. The

trailing edge flap spectra developed at SMPL-IAR are expressed in terms of a hinge
moment. The reference value provides a means of converting the load sequence to an

equivalent stress sequence. The DLS refers to the peak stress in the load spectrum.

An option in McCracken is the ability to run the predictions for a number of different

design limit stresses. This is acconplished by entering values for the maximum Kt*DLS,

the Kt*DLS hcrement, and the nurnber of Kt*DLS breakpoints to calculate. This

information c m be used to plot Kt*DLS vs. Life for the component The required format

of the load spectrum file is presented in Appendix B.

5.3.3 Prediction Methods Dialog Box

The prediction methods dialog box is presented in Figure 5 4 . It ailows the user to

configure the way in which the prediction is to be made. The user has the option to

configure five aspects of the local strain approach:

1. Nominal stress vs. notch root stress-strain relationship

2. Cycle ordering

3. Cycle counting

4. Equivalent strain relationship

5. Material properties

For aspects 1 and 4, multiple options are listed to enable the user to configure the manner

in which the prediction is made.

The standard adopted for IFOSTP is to use the foliowing options when making a

prediction using -189 (Foster, 1993):

Use of prestrained material data

Move peak load to the front of the spectrum

Use of "Closed Hysteresis Loop Counting"

Smith-Watson-Topper (S WT) equivalent strain equation

This standard was used in this thesis, except where noted.

5.3.4 Executhg the Prediction

the program is
Afier the material, Load spectrum and prediction methods are c~nfi~gured,

ready to be executed. The toolbar button labeled "Go" executes the local strain

algorithrns. The results are displayed to the screen once the caiculations are complete.

5.3.5 Documenfing Results

The results c m be saved to disk after the prediction is made using the menu option

"Results-Save". The user c m specifi a filename for the results file. Al1 results files are

given the ".res'? extension. The format of the results file is given in Appendix B.

5.4 Validation of McCracken

5.4.1 Ovetview of Validation

The McCracken software was validated by comparing predictions to those made by

CC189 for the trailing edge flap hinge moment sequences described in Chapter 4.
5.4.2 Material Data

The materials database in C-CI89 contains data for alurninum 7050-T74. This data was

used in both the C-CI89 and McCracken predictions. The cyclic stress vs. stress*strain

and strain-life curves are shown in Figures 5-5 and 5-6. respectively. The prestrain and

non-prestrain data are ploned on Figure 5-6, as well as the SWT compatible strain-life

c u v e which is generated by the program.

5.4.3 Cornparison of C-Cl89 and McCracken Predictions

Crack initiation predictions for the tef-man05 sequence are presented in Table 5-1.

[ksil

Table 5-1:Cornparison of C-CI89 and McCracken Predictions for tef-man05

The McCracken predictions are in excellent agreement with the C-CI89 predictions. The

maximum error (relative to C-CI89) over the Kt*DLS range in question is 0.75%.

The CC189 predictions were made at SMPL-IAR and used the maximum manoeuvre

range as the reference value. This was done since SMPL-IAR was interested in the
effects of truncation which is typically quoted as a function of the largest cycle in the

spectrum. The predictions made for this study use the maximum value in the specûum as

the reference. Therefore, a conversion had to be made to express the -189 predictions

in a format suitable for cornparison with McCracken predictions. The error between

CC189 and McCracken predictions is due to interpolation error when converting the

C-CI89 prediction to one in t e m s of the maximum spectrum value.

5.5 Cornparison of Local Strain Prediction Programs

5.5.1 C-C189 vs. LOOPIN8

The Northrop local strain program, LOOPIN8 (Porter, 19831, was initially reviewed for

use in IFOSTP. LOOPIN8 is similar to C-CI89 in its local strain algorithm. However. it

uses a custom equivalent strain equation which is not included in C-CI89, In addition,

LOOPIN8 does not allow the use of other equivalent strain equations.

It was noted by Foster (1993) that predictions made using LOOPIN8 were si-gnïficantly

lower than those from C-CI89 even though the material data used showed good

agreement with the data used in C-CI89. A study was performed to ven& the

discrepancy between the use of the SWT and LOOPIN8 equivalent strain equations. The

LOOPIN8 equivalent strain equation has the foliowing form:

where REand & are the strain and stress ratios. and q-,
is the mean stress. From Figure 5-

7- it can be seen that the LOOPIN8 equivalent strain equation predicts a lower Life for the

t e f ~ 1 a n 0 5sequence than the SWT equation. The difference is significant and is

consistent with the results of Foster (1993).

5.5.2 C-Cl89 vs. McCracken

Some of the numerical techniques used in the C-CI89 program were improved upon in

McCracken. An audit of the C-CI89 source code revealed that the solution of Neuber's

rule was not strictly correct. -189 calculates the product of the notch root stress and

strain using Neuber's rule? and then interpolates through a stress vs. stress*s& data file

to solve for the notch root stress. The method used in McCracken is consistent with the

method presented by Bergengren et al. (1993). In this method. the stress vs. strain curve

is used instead of the stress vs. stress*strain curve. The product of the notch root stress

and strain is calculated using Neuber's rule. Next. iteration is used to find a point on the

stress vs. smin cun7eso that the product of stress and strain for that point is equal to the

value calculated using Neuberosrule.

The two solution techniques only display a difference in the interpolated regions of the

stress-strain curve. The diKerence is most pronounced in the elastic region of the stress-

strain c w e . In this region the C-CI89 technique leads to erroneous results. -189 uses

the stress vs. stress*strain c u v e which is created by multiplying the abscissa of the stress

vs. s a i n curve by the ordinate, and re-plotting this value as the new abscissa. In dohg
so. any regions of the original stress vs. strain curve which were Iinear (Le. elastic region)

are now non-linear in the new stress vs. stress*strain curve. Even though this is the case,

-189 plots the elastic region of the stress vs. stress*strain c w e as linear. Figure 5-8

plots the CC189 representation of the stress vs. stress*strain curve next to what one

would expect. Use of C-CI89 data and the -189 solution to Neuber's mle in the

elastic region results in erroneous predictions of notch root stress and strain. For this

reason, CC189 uses Hooke's Law to calculate the notch root stress and strain when in the

elastic region of the material.

Outside the elastic region. the difference between the C-CI89 and McCracken solutions

to Neuber's mle is small. The smdl difference in stress-strain estimation rnay not justify

the extra computational effort required by the McCracken solution. However, the

McCraclien solution of Neuber's rule is valid in al1 regions of the stress-strain curve,

including the elastic region. Ln addition, the McCracken solution uses the more familiar

stress vs. strain curve instead of the unconventional stress vs. stress*strain curve.

When calculating the cycles to failure for a closed hysteresis loop, CC189 does not allow

the life to go beyond 10~' cycles. This can be a problem in a spectrurn with a large

nurnber of small amplitude cycles, where the long life region of the strain-life curve is

extrapolated to get the cycles to failure. McCracken does not force the life to a maximum

of 10" cycles in cases where an extrapolation has to be made.

A computer prograrn to predict crack initiation life using the local strain approach was

developed and validated. The relative error behveen the McCracken and CC189

predictions is small (maximumerror on the order of 0.75%) for the matenai and spectnim

tested. The difference is due to the conversion of the CC189 predictions to express them

in terms of the maximum spectrum value. The McCracken program presents a suitable

plaform for implementation of new notch root stress-strain estimation methods since it

was specifically designed to allow such modification.

 -
Chapter 6 Finite Elernent Analysis

6.1 Introduction

An elastic-plastic finite element analysis (FEA) was performed for both the Low and

High Kt coupons. The objective of the FE analysis was to observe the notch root stress

and strain as a function of the net section nominal stress for each coupon.

The FEA for each coupon was divided into two cornponents:

1. Initial elastic FEA. and

2. Elastic-plastic FEA

The initial elastic FEA was performed to verie the stress concentration factor.

The FE analysis was perforrned using the AJ3AQUS finite element program (HKS Inc.

1997), and run on a Silicon Graphics 'Indy' cornputer.

6.2 Constitutive Models

The material considered in the analysis was aluminurn 7050-T74. The material data

presented in the C-CI89 materials database (Klohr, 1990) formed the b a i s for the
constitutive modets. The cyclic stress vs. strain curve for aluminurn 7050474 is
presented in Figure 6-2 -

The ABAQUS FE program requires that the material be descnbed in the elastic region by

Young's modulus, and in the plastic region by stress vs. plastic strain data. Figure 6-1

shows the stress vs. plastic strain data plotted next to the stress vs. strain curve.

The "Classical Metal Plasticity" model in ABAQUS was used with the von Mises failure

criterion. The plasticity model uses an associated plastic flow rule to deterrnine the

inelastic deformation rate. The flow rule specifies the inelastic deformation rate to be in

the direction normal to the yield surface as the material yields. According to HKS Inc.

(19971, this assumption is acceptable for most calculations with metals.

6.3 Loading

When the coupons are tested in fatigue, they are placed in the hydraulic gips of the MTS

load frarne such that the longitudinal axis of the coupon is aligned with the hydraulic

actuator. The picture of the Low Kt apparatus set-up in Appendix A demonstrates this.

The coupons are subjected to axial loading which c m be approximated by a uniformly

distributed load applied to the top and bottom surfaces.

6.4 FEA of Low Kt Coupon

6.4.7 Geometry of Low KtFE Mode1

The Low K, coupon is depicted in Figure 4-1. It consists of a rectangular strïp with two

semi-circular edge cutouts. The design stress concentration factor is 1.4. Three planes of

synmetry at x = O": y = 0": and z = O" allowed one-eighth of the coupon to be considered.

S y m e t r i c boundary conditions were used on the one-eighth FE rnodel. Nodal

displacements were not allowed for nodes lying on the following planes:

1 Plane 1 Definition 1

A three dimensional FE mode1 was constructed using 20 node 3-D elements. A mesh

convergence study \vas performed using models whîch contained 432, 1008, 1800, and

3600 elements. Each of the models were loaded with a 5 ksi (34.47 MPa) distributed load

applied to the top surface of the coupon. The von Mises stress dong the edge defined by

y = 0". z = O", was plotted for the 432 and 3600 element meshes to determine

convergence. Figure 6-2 shows that the results from the 432 element mesh are not as

refined as the results fiom the 3600 element mesh. The results from the 1008 and 1800

element meshes were found to agree with the results from the 3600 element mesh.
The mesh selected for the analysis. shown in Figure 6-3: has 1800 elements and 8913

nodes. Bergengren et al. (1993) used a sirnilar mesh in an elastic-plastic FE analysis of

double edge notch coupon made fiorn steel DP400.

-
6.4.2 Elastic FEA of Low Kt Coupon Verifcation

The Low Kt FE rnodel was loaded elastically to venfy the design stress concentration

factor. Kt. and to ver@ the stress distribution at the net section. The stress concentration

factor found in the analysis was 1.41, whereas the design was for 1.4. Other sources were

used to veri@ the design Kt with the following results:

Source Kt
-
FE mode1 1.41
SMPL-IAR Design Kt (Weiss, 1997) 1-40
(Young, 1989) 1.40
(Pilkey, 1994) 1.42

The stress concentration factor obtained in the analysis agrees with what is presented in

literature for this geometry. It should be noted that life predictions which are made for

this coupon at SMPL-[AR use a stress concentration factor of 1.39, which was obtained

from a 2-D plane stress FE analysis.

The following approximate relationship, presented by Glinka and Newport (1987), was

used to ven& the elastic stress vs. distance response at the net section:
where x is the distance fiom the notch root, r is the notch radius, S is the net-section

nominal stress, and Kt is the stress concentration factor.

Figure 6-4 presents a cornparison of the FE results nith those predicted by Equation 6-1.

Glinka and Newport's relationship was not in agreement with the FE results at the net

section of the Low K, coupon. Glinka and Newport (1987) assert that the relation is valid

for a spmetrical semi-circula edge notch in the region x < 3r. However, Equation 6-1

was verified by Glinka and Newport (1987) for a coupon which was very wide when

compared to the notch root radius. For the Low Kt coupon, the notch root radius is large

when compared to the overall width of the coupon. With this in mind, the approximate

relationship may not be suirable to predict the elastic response of this geometry.

6.4.3 Elastic-Plastic FEA of Low KtCoupon

n i e Low Kt FE model was loaded incrementally as described by HKS Inc. (1997) up to

the maximum applied stress. The criterion for force convergence was based upon

minimizing the residual force. The residual force is deGned as the difference between

external Loads and intemal forces. The force convergence of the model is summarized in

Table 6- 1.
 Parameter value 1 Units
Maximum Stress @oss section) 28 ksi]
Number of Load Incrernents 14 -
Largest Residual Force 0.0004586 ["/O]

Run Tirne 15.5 plours]

Table 6-1: Force Convergence for Low Kt Coupon

The critenon for the residual force was 0.5%. As c m be seen, the largest residual force

for the mode1 was negligible, on the order of 0.0004%. This indicates that force

convergence was well widin the required tolerance, and that the number of load

increments is suficient to model the coupon up to the maximum applied stress.

6.5 FEA of High Kt Coupon

6.5.1 Geometry of High Kt FE Model

The High Kt coupon is depicted in Figure 4-2. It consists of a rectangular strip with a

centrally located dot. Three planes of syrnmetry at s = 0"' y = 0'': and z = O" dlowed

one-eighth of the coupon to be considered. Symmetric boundary conditions were used on

the one-eighth FE model. Nodal displacements were not dlowed for nodes lying on the

following planes:

1 PIane 1 Definition 1

2 - "'net section" Y = 0:'

3 z = O?'
A three dimensional FE model was constnicted using 20 node 3-D elements. A mesh

convergence study \vas performed using models which contained 600, 13 14, 2280, and

4500 elements- Eacli of the models were loaded with a 2 ksi (13.79 MPa) distributed load

applied to the top surface of the coupon. The von Mises stress along the edge defined by

Y = (-yT.z = O", was plotted for the 600 and 4500 element meshes to determine

convergence. Figure 6-5 shows that the 600 element mesh provides a crude

approximation of the stress gradient near the notch. The results from the 1314 and 2280

element meshes were found to agree with the results fiom the 4500 element mesh.

The rnesh selected for the analysis, shown in Figure 6-6. has 2280 elements and 11153

nodes.

-
6.5.2 Elastic FEA of High Kt Coupon Verifkation

The High K, FE model was loaded elastically to ver@ the stress concentration factor, K,

The stress concentration factor found in the analysis was 3.18. The value used at SMPL-

[AR for life predictions is 2.88, which was obtained from a 2-D plane stress FE analysis.

Other sources were used to veriQ the design Kt with the following results:

Source Kt
FE mode1 3.18
(Young, 1989) 3.22
(Pilkey, 1994) 3.13
(ESDU, 1983) 1 3.23
The stress concentration factor obtained in the analysis agrees with what is presented in

literature for this geornetry.

6.5.3 Elastic-Plastic FEA of High Kt Coupon

n i e High Kt FE model was loaded incrementally as described by HKS Inc. (1997) up to

the maximum applied stress. The criterion for force convergence was based upon

minimizing the residual force. The force convergence of the model is summarized Table

6-3.

Parameter Value UniCs

Maximum Stress (@ gross section) 45 bil 1

1 Number of Load hcrements 1 16 1 [-1

- - 1
1 1
I

Largest Residual Force 1 0.0003158 1 f%7

1 Run Time 1 22.3 1 [hours] 1
Table 6-2: Force Convergence for High Kt Coupon

The criterion for the residual force was 0.5%. As c m be seen' the largest residual force

for the model was negligible' on the order of 0.0003%. This indicates that force

convergence was well within the required tolerance? and that the number of load

increments is sufficient to model the coupon up to the maximum applied stress.

6.6 Results of Elastic-Plastic FEA

The following plots were generated for each of the FE analyses:

Maximum principal stress (SP3) vs. distance from notch root

 Maximum principal strain (EP3) vs. distance from notch root

von Mises stress vs. distance fi-orn notch root

The ABAQUS naming convention for principal stresses and strains is used, where

SPPSP2>SP 1, and E P P E P D E P 1.

The plots were generated for two locations at the net-section, the mid-thickness (Y = O",

= O?'), and the surface of the coupon (y = O", z = 0.125"). The results are presented for

various levels of net section nominal stress.

6.6.1 Low Kt Coupon

Figures 6-7 and 6-8 present SP3 vs. distance from the notch root for the rnid-thickness

and surface of the Low Kt coupon, respectively. Although the mid-thickness results are

larger than the surface results for a given value of net-section nominal stress, the

difference between the two is small.

Figures 6-9 and 6-10 present EP3 vs. distance from the notch root for the mid-thickness

and surface of the coupon. The largest strain is at the notch root, with the mid-thickness

strains larger than the surface strains for a given value of net section nominal stress.

Figures 6-1 1 and 6-12 present the von Mises stress vs. distance from the notch root for

the mid-thickness and surface of the coupon. Although the von Mises stress was net of
prhnary interest in this study. it was ~bservedsince it was used as the failure criterion in

the FE program.

6.6.2 High Kt Coupon

Figures 6-13 and 6-14 present SP3 vs. distance fiom the notch root for the mid-thickness

and surface of the High Kt coupon, respectively. As expected? the stresses at the mid-

thickness are larger than the stresses at the surface for a given value of net section

nominal stress. The differences between the mid-thickness and surface results are more

pronounced than in the Low Kt case.

Figures 6-15 and 6-1 6 present EP3 vs. distance fiom the notch root for the mid-thickness

and surface of the coupon. The Iargest strain is at the notch root. with the mid-thickness

strains larger than the surface strains for a given value of net section nominai stress.

Figures 6-17 and 6-18 present the von Mises stress vs. distance from the notch root for

the mid-tluckness and surface of the coupon.

6.7 Discussion of FE Results

The results of the elastic-plastic analysis (Figures 6-7 to 6-12 and 6-13 to 6-1 8) illustrate

the progression of the stress state at the net section. As the value of the net-section

nomind stress increases. the maximum value of SP3 moves inwards from the notch root.

This is due to the influence of transverse stresses and is consistent with yielding at a

notch as described by Broek (1989). The movement of the maximum value of SP3 is

more pronounced in the High K, coupon due to the strong triaxial stress state.
For the Low K,coupon, the notch root stress and strain are plotted against the net section

nominal stress in Figures 6-19 and 6-20 respectively. The results for the mid-thickness

and the surface of the coupon are nearly identical, indicating that there is not a large

degree of constraint associated with this geornetry. Sharpe et al. (1997) suggest that the

ratio of principal strains, a. can be used as a measure of the constraint at the notch root.

When a is very close to the negative of Poisson's ratio for the material? the c o n s ~ a i nis
t

low, and the stress state approaches that of plane stress. For the Low Kt coupon, a was

found to be -0.3. which is close to the negarive of Poisson's ratio for aluminum 7050-

T745 1.

The situation for the High Kt coupon is different. Figures 6-21 and 6-22 plot the notch

root stress and strain against the net section nominal stress. A large difference exists

between the surface and mid-thickness results, indicating a large degree of constmint

associated with this geometry. In this case, a was found to be -0.22. Sharpe er al. (1992)

suggest that as cr approaches zero? the notch root constraint increases and the stress state

approaches that of plane strain.

Although the stress concentration factors found in the analyses agree with the "handbook"

values, the value of Kt for the High Kt coupon was found to be in serious disagreement

with the value used at S M P L - M . The value found in this analysis was 3.18 which

agrees with the values given in (ESDU, 1983), (Young. 1989), and (Pilkey, 1994). The

value used at SMPL-IAR is 2.88 which was obtained f/rorn a 2-D plane stress FE analysis.
The Hi& Kt coupon is constrained at the notch root as explained previously. Thus, the

assumption of 2-D plane stress is not valid.

The following surnmarizes the FE analyses performed on the Low and Hi@ K, coupons:

The elastic FE analyses verified the stress concentration factor with

"handbook" values.

The progression of the stress state at the net section bas shown that the

maximum vahe of the largest principal stress moves inwards fiom the notch

root as yielding progresses. This is consistent with known theory.

The stress States at the notch root of the Low and Hi& Kt coupons are

different. The Low K,coupon is close to plane stress. whereas the High Kt

coupon experiences a tnaxial stress state at the notch root with a significant

amount of constraint.

Notch root stress and strain vs. net-section nominal stress data have been

derived in a form suitable for cornparison to Neuber's rule and other

approximate relationships.
 -
Chapter 7 Sensitivity Stuciy

7.1 Objective

As mentioned in Chapter 2, the local strain approach to fatigue life prediction requires a

relationship between the net section nominal stress and the notch root stress and strain.

This chapter presents a sensitivity study performed on the stress and strain estimation

used in the local strain approach. The sensitivity study was performed for the Low and

High Kt coupons by varying the method used to calculate the notch root stress and stmin.

The study was divided into two components:

1. Sensitivity of notch root stress and strain estimation.

2. Sensitivity of crack initiation predictions for the trailing edge flap hinge

moment sequences described in Chapter 4.

The objective of the study was to identi@ the best stress-strain estimation technique when

compared to the FE analysis, and to assess the ability of the local strain method to predict

the lives of the coupons when compared to the test results in Chapter 4. The McCracken

Fatigue Life Prediction Prograrn described in Chapter 5 was used to perform the analysis.
 66

7.2 Description of the Stress and Strain Estimation Methods

The methods analyzed in this study are:

e Neuber's rule (Neuber, 1961)

Glinka's Equivalent Strain Energy Density Method (Molski and Glinka 198 1)

Hoffmann and Seeger GeneraIized Method (Hoffmann and Seeger, 1985)

Elastic-Plastic FE analysis (Chapter 6 )

A brief description of the solution technique for Neuber's rule. Glinka's ESED method.

and Hoffmann and Seeger's method will be presented, followed by the results of the

stress-strain estimation study and the crack initiation prediction study.

7.2.1 Solution Technique for Neuber's Rule

The derivation of Neuber's rule was presented in Chapter 2. The two forms analyzed in

this study are:

Equation 2-1 7 will be referred to as Neuber (Nonlinear), while Equation 2-18 will be

referred to as Neuber rule. Equations 2-17 and 3-18 are solved using the following

method:
 1. Calculate the ri& sides of the equations. In the case of Equation 2-1 7, the net-

section nominal strain is calculated as the strain value on the cyclic stress-strain

curve whose stress value is equal to the net-section nominal stress.

2. Use iteration to find a point on the cyclic stress-strain curve whose product of

stress and strain is equal to the value calculated in Step 1.

7.2.2 Solution Technique for Glinka's ESED Method

Glinka's equivalent strain energy density (ESED) method (Molski and Glinka, 1981) is

espressed as:

Equation 2-21 is solved using the following method:

1. Calculate the left side of the equation.

2. Use iteration to find a point on the cyclic stress strain cuve such that the

integral calculated in Equation 2-2 1 is equai to the value calculated in Step 1.

7.2.3 Soiution Technique for Hoffmann and Seeger's Generalized Method

The method to estimate rnultiaxial elastic-plastic notch stresses and strains proposed by

Hoffmann and Seeger (1985) was described in Chapter 2. The rnethod is comprised of

two steps:
1. Relating the applied stress to the equivalent notch stress and strain. The

GC,od?obtained h m an elastic
principal elastic stresses at the notch root oelz

FE analysis are used to calculate the equivalent stress concentration factor Ktq.

The equivalent notch stress and strain (o,and E,) are calculated from Neuber's

rule.

3. ReIating the equivalent notch stresses and strains to the principal notch stresses

and strains. This is accomplished by using the fmite Law of Hencky (Equation

2-20). In order to solve the set of equations, it is necessary to make an

assumption regarding one of the principal strains. Hoffmann and Seeger

(1985) make the assumption that the ratio of principal strains (a)is the same in

the elastic and elastic-plastic cases. The solution of Hencky's equations,

expressed in teims of the principal strains (q,s2)- the principal stresses (ci,02,

Poisson's
03)? ratio (v), and the equivalent quantities From step 1 (o,,E,) is

then:
 (CO nstant)

(traction@ee srirface)

3= 411-- au'a
ta-
~
Eq -

Equations 7-9 and 7-10 are the notch root stress and strain respectively. In Chapter 6, it

was found that the High Kt coupon has a large degree of constraint at the notch root. The

Low Kt coupon was found to be in a state of nearly plane stress. As such, the HoEnann

and Seeger method will only be used for the High K, coupon.

7.3 Sensitivity of Notch Root Stress and Strain

The independent variable in each of the estimation methods presented above. including

the FE analysis, is the net section nominal stress. Therefore, the notch root stress and

strain are presented as functions of the net section nominal stress for each method.

7.3.1 Low Kt Coupon

n i e notch root stress and strain results are presented in Figures 7-1 and 7-2, respectively

for the Low K, coupon. Figure 7-1 shows that Neuber's mle provides an adequate
estimation of the notch root stress when compared to the FE results. From the strain

results of Figure 7-3: it c m be seen that a similar statement c m aiso be made.

When compared to Neuber's rule. Glinka's ESED method predicts a smaller notch root

stress and strain for a given value of the net section nominal stress. It would appear that

Glinka's ESED method provides a lower bomd on the notch root strain. This is

consistent with the results of Sharpe et al. (1992).

7.3.2 High Kt Coupon

The notcli root stress and strain results are presented in Figures 7-3 and 7-4, respectively

for the High K, coupon.

The results of the sensitivity study are quite interesting. In Chapter 2, it m-as established

that Neuber's rule overestimates the notch root strain in a situation where a multiaxial

stress state exists. This is certainly evident in this analysis. In addition. Figure 7-3

illustrates that Neuber's rule provides a lower estimate of the notch root stress when

compared to the FE results. The Hoffmann and Seeger modification to Neuber's rule

(Equation 2-1 8) provides a reasonable approximation of the rnultiaxial effect. As was the

case with the Low Kt coupono Glinka's ESED method provides a lower estimate of the

notch root stress and strain state when compared to Neuber's rule and the FE results.
7.4 Sensitivity of Crack initiation Predictions

Local strain crack initiation predictions were made for the trailing edge flap hinge

moment sequences. n i e predictions were made using the IFOSTP standard options.

outlined in Chapter 5, and wiîh the McCracken inputs Listed in Table 7-1.

Coupon Spectrurn Turning ~ e f e r e n c eValue Kt DLS

Points [in*kîps] [ksi]

tef sumo5 1 409380 1 223.45

Table 7-1: McCracken Inputs for Crack Initiation Sensitivity Study

7.4.7 Low Kt Coupon

Cornparisons between the crack initiation predictions and the test results (total life) are

presented In Figure 7-5. The results show that predictions made using Neuber's rule to

estimate the stress and strain at the notch root are in good agreement with the test results

(total life) for both the tef-man05 and t e h m 0 5 sequences. In addition, the predictions

made using the FE results and Glinka's ESED method are nearly identicai to those made

using Neuber's rule.

7.4.2 High Kt Coupon

Cornparisons between the crack initiation predictions and the test results (total life) are

presented in Figure 7-6. Unlike the results for the Low Kt coupon, the results for the

High Kt coupon show that the predictions made for the t e m a n 0 5 and tef-sumo5

sequences are very conservative with respect to the test results (total life). In addition. the

predictions made using the FE results are "more conservative" than those made using

Neuber's rule. This is a surprising result since it was established in Section 7 . 3 2 that

Neuber's rule overestimates the notch root strain for the High K, coupon when compared

to the FE results. Since the number of cycles to failure for each closed hysteresis loop is

calculated fiom the strain-life curve (Figure 5-6). it was expected that predictions made

using Neuber's mle would be consemative with respect to predictions made using the FE

results.

This result is explained by noting the multiaxid stress state at the notch root of the High

Kt coupon, and the use of the Smith-Watson-Topper equivalent strain equation. In

Section 7.3.2, it was established that the FE results demonstrate a smaller notch root

strain and a larger notch root stress when compared to Neuber's rule. Moreover, the

SWT equation modifies the strain amplitude according to Equation 2-5 which is Listed

here for reference:

The net effect of the lower strain and the higher stress for the rnultiaxial case teads to a

Iarger value of equivalent strain when compared to using Neuber's nile. Hence,

predictions made using the FE results for the High K, coupon will be conservative with

respect to predictions made using Neuber's rule.

7.5 Summary

The following sumrnarizes the results of the sensitivity study:

Neuber's mle provides a reasonable estimation of the notch root stress and

strain for the Low Kt coupon.

The Hoffmann and Seeger modification to Neuber's rule provides adequate

estirnates of the multiavial effect for the High K, coupon.

For both the Low and High Kt coupons. Glinka3 ESED method predicts a

lower notch root stress and strain than Neuber's rule and the FE results.

Crack initiation predictions made using Neuber's rule are in agreement with

the test results (total life) for the Low Kt coupon.

Predictions made for the High Kt coupon are conservative when compared to

the test results. Due to notch root muitiaxiality and the use of the SWT

equation, crack initiation predictions made using the FE results are

conservative when compared to those made using Neuber's nile.

The local strain method displays a considerable difference in predictive

capability between the Low and High K,coupons. A discussion of this point
will be made in Chapter 8.
 -
Chapter 8 Discussion of Results

8.1 Introduction

As stated in the project definition (Chapter 3, the objective of this thesis is to examine

the applicability of Neuber's rule in the local strain approach and to assess the ability of

the local straîn method to predict the lives of the Low and High Kr coupons. With this in

mind. it becomes clear that this thesis rnust answer two questions:

1. 1s Neuber' rule applicable for the two coupons presented ?

2. Does the local strain method adequately predict the Iives of the two coupons

subjected to s p e c t m loading ?

The discussion of these two questions will be made in the following sections' using

results Eom the studies presented earlier. In addition, a discussion will be made for two

other sources of error in the LS approach: materid data and equivalent strain equations.

8.2 Applicability of Neuber's Rule

Neuber's nile was denved for a two dimensiond state of stress, and has been shown to

provide accurate estimations of notch root stress and strain in plane stress situations.

Sharpe el al. (1992) make the assertion that Neuber's rule is the single best mode1 for
plane stress situations? which was supported by the results for the Low Kt coupon. In the

case of the High KI coupon, a multiaxial stress state exists. hence Neuber's nile is

inadequate to describe the notch root response. This result suggests that Neuber's rule

should not be used unconditionally, but should be vetified by an elastic-plastic finite

element analysis when possible. The Hoffiann and Seeger modification to Neuber's rule

appears to be a promising way of accounting for the multiaxial stress state at the notch

root.

The form of Neuber's rule which does not assume Hooke's Law. Neuber (Nonlinex). is

not commonly used in the literature even though it accounts for non-linearity in the net

section nominal stress-nominai strain response. ui this study, it was found that

predictions made with the traditional form of Neuber's rule were closer to the FE results

than those made using Neuber (Nonlinear). It is not known if other researchers have

observed similar performance of the Neuber (Nonlinear) method.

In this analysis, Glinka's ESED method provided a lower bound on both the notch root

stress and notcli root strain for the Low and High Kt coupons. The hasis for the ESED

method is the assumed equivalence between the strain energy density calculated fiorn the

elastic and elastic-plastic material laws. This assumption is valid when the local plastic

zone is small in comparison with the elastic portion of the material surrounding the notch.

Figures 6-7 and 6-13 show that yielding occurs across a significant portion of the net-

section as the nominal stress increases. This indicates that the Glinka ESED method may
not be appropriate to use once the net-section nominal stress is geater than the yield

stress.

The results of this study suggest that Neuber's rule is adequate for the Low Kt coupon,

and Neuber's rule with the Hoffmann and Seeger modification is adequate to estimate the

notch root stress and strain of the High K, coupon. In this study, the method used to

estimate the notch root stress and strain has a relatively small effect on the predictions of

crack initiation life as demoostrated in Figures 7-5 and 7-6. However, the effect of the

notch root stress-strain estimation rnethod in the LS approach is dependent on material.

spectra and notch geometry. Variations in these three parameters will yield different

results than those presented in this thesis. Therefore. in situations different from those

considered in this thesis, it cannot be concluded with certainty that the notch root stress-

strain estimation will ais0 have a relatively small effect on the predictions of life to crack

initiation. Further study involving different coupon geometries- spectra and material

would provide insight ïnto the effect of notch root stress-strain estimation on the

prediction of crack initiation life.

8.3 Agreement between LS Predictions and Test Results

8.3.1 Introduction

The results of the sensitivity study presented in Chapter 7 display a large difference in the

capability of the local strain method to predict the lives of the Low K, and Hi& K,

coupons subjected to the trailing edge flap hinge moment sequences. In short, crack

initiation predictions made for the Low K, coupon are in apreement with the test results

(total life). while the predictions made for the High Kt coupon are extremely conservative
with respect to the test results. This raises the question of why the local strain method is

able to predict the lives of the Low Kt coupons and not able to predict as well for the

High Kt coupons. This point will be discussed in terms of the observed behaviour of the

coupons with respect to crack initiation life versus total life. and in terms of the fatigue

concentration factor. In addition. a method to caiculate total life will be presented which

includes a defuiition of the size of a crack at initiation.

8.3.2 Crack Initiation vs. Total Life

The coupon tests described in Chapter 4 were part of a program at SMPL-IAR to study

the relative effects of truncation on life and were not directly aimed at correlating test

lives with predictions. For the purpose o f this thesis. the assumption was made initially

that the crack initiation phase cornprised a very large portion of the total life. During the

course of the testing. attempts were made to "catch" the crack initiation of the coupons by

visual obsewation.

For the Low Kt coupon. the attempt to "catch" the crack initiation was in part

unsuccessful~as a crack was never observed during testing. However. notes were made at

intervals during testing of when these inspections were made which allows the tirne to

initiation to be estimated. The last recording of this information for the coupons listed in

Table 8-1 indicates that the crack propagation phase is smail, and could be smaller than

9% of the total Life in the case of specimen 077, and smaller than 26% in the case of

specimen 3 11. At the tirne this information was recorded, no visible cracks were present

in the coupons. indicating that the crack "initiated" after these recordings were taken.
 Specimen Last record of Turning points Life remaining

turning points applied at failure after Iast record

077 4153094 4571310 9%

Table 8-1: Indication of Crack Propagation Phase for Low Kt Coupons

In the case of the High Kt coupon? visible cracks were detected after they had grown

across the net section of the coupon. Table 8 - lists the approximate crack length.

number of turning points applied to the coupons at first detection of the crack- and

percent life remaining (based on the number of turning points at failure). The Hi& Kt

case is ver); diEerent from the Low Kt situation in that cracks were visibly seen and

propagated for a significant portion of the total lives of the coupons. The total amount of

time spent propagating the cracks is not known since the number of cycles to initiation

was not isotated.

 Specimen Number of turning Crack length [in] Life Remaining

I 1 points applied (from notch root) Wd

Tabie 8-2: Crack Length a t Fint Detection for High Kt Coupons

Experimental observations suggest that the crack propagation phase of the Low K,

coupon accounts for a very small portion of the total Me. In the case of the Hi& K,

coupon. observations suggest that the crack propagation phase accounts for a sign5cant

portion of the total life. These observations are supported by a limited fracture mechanics

study of the coupons usinj the FRANCZDL crack propagation simulator (James and

Swenson. 1997). The FRANC?D/L program is a two-dimensional finite element program

which uses a regenerative rnesliing algorithm to propagate a crack using linear elastic

Fracture mechanics (LEFM). A crack with length 0.01" (0.254 mm) was placed at the

notch root of both coupons. Using the simulator, the stress intensity factor at the crack tip

was plotted versus crack length as measured frorn the notch root for both the Low and

High K, coupons. The results of this study, presented in Figure 8-1, show that the Low Kt

coupon has a much larger stress intensity at the crack tip than the High K, coupon for a
given crack length. As a result, the Low Kt coupon would experience a shorter crack

propagation stage than the High Kt coupon. consistent with observed behaviour.

8.3.3 Fatigue Concentration Factor

In Chapter 2, the fatigue concentration factor was discussed as a method of improving

local strain predictions by accounting for size effects in sharp notches. The most direct

and reliable way to determine Kr is by expenment (Weixing et al.. 1995). However. an

attempt of such magnitude was not feasible during the course of this research program

due to time and economic constraints. The problern with using empirical formuiae to

calculate Kf is that they require material constants which c m only be calculated from data

for smooth and notched specimens available from previous tests. In addition, data may

not be available for the specific material in question. An estimate of the material constant

for the Peterson formula was given as 0.025" (0.635 mm) for alurninum alloys (Peterson,

1959).

The values of Kr calculated h m Equation 2-23 for the Low and High K, coupons are

listed in TabIe 5-3.

Coupon

Low Kt

High Kt
--

Table 8-3: Fatigue Concentration Factors for Low and High Kt Coupons
The Kr value for the Low Kr coupon is very close to the Kt value, resulting in little

difference in predicted life. The use of Kr instead of Kt in the local strain predictions for

the Hi& K, coupons result in larger predictions of life when plotted on the K*DLS curve

of Figure 7-6. The predictions are more in line with the test results for total life. This

result is consistent with the observations of Topper et al. (1969): who showed that

measured notched fatigue lives (initiation + propagation) and lives predicted From smooth

specimens agree within a factor of 2 when Kf is used in place of Kt in Neuber's nile.

The apparent success of this method is questionable since there is no way of knowing

how applicable the material constant "a" is for the material used in this analysis. The

value of 0.025" (0.635 mm) is supposedly constant for al1 aluminurn alloys according to

Peterson (1959). Topper ei al. (1969) use a value of 0.028" (0.71 1 mm) for both 2024

and 7075 aiuminurn alloys. Efforts to obtain other values for aluminurn were

unsuccessful. The rnajority of data available for the calculation of Kr are for steels and

are given in references such as die S M Fatigue Design Handbook (1988).

Another concern with the use of Kf is that the stress and strain estimation at the notch root

is altered. Using Kf instead of K, in both Neuber's rule and Glinka's ESED method

resuits in srnaller predictions of the notch root stress and strain. The FE analyses

presented in Chapter 6 cannot be modified to represent the stress concentration by Kf

instead of K, leading to an important consequence of this method: the stress and strain
experienced by the notch root of the coupons in testing will be larger in magnitude than

the stress and strain predicted by Neuber's nile with Kr for local strain predictions.

Furthemore. the notion put forth by Bannantine et al. (1990) that the use of Kr is an

empirical method to account for crack propagation in sharp notches cannot be dispelled.

The basis for the use of Kr in place of Kt is the observed difference in the fatigue

strengths of smooth and notched specimens. The fatigue strength of the notched

specirnen is taken as the stress level to cause failure at 10' cycles. In the case of the

notched specirnen. the crack propagation stage may comprise a significant portion of the

total life. Hence, Kr inherently includes a first order consideration of the crack

propagation in the specimens used to generate the "notched specimen" S-N data.

The use of Kr in the LS approach is questionable when considering the above discussion.

In the case of the High Kr coupon. crack propagation was observed to be a significant

portion of the total life. Therefore. crack propagation must be taken into account to

obtain an accurate estimate of total life for the High Kt coupon. A method to predict total

life. consisting of initiation and propagation is discussed in the next section along with a

definition of the size of the crack at "initiation".

8.3.4 Estimating Total Life

Several authors have supported the notion that crack propagation analyses must be

performed to obtain an accurate estimate of total life as the severity of a notch increases.

Nelson and Socie (1 982) indicate that in some cases crack growth analyses are required to
properly assess the component fatigue life, while in other cases such as blunt notches,

crack initiation analyses atone are suficient.

Dowling (1979) noted that the diRering strain gradients between srnooth and notched

specimens will cause error in local strain predictions if a significant portion of life

considered as initiation is acîually spent in crack growth. Dowling proposed that a

definition of initiation as a crack of ~ ~ c i e n t small

l y size overcomes this problem.

Crack initiation analyses are valid until a crack reaches the initiation size, and

propagation analyses are performed £kom the initiation size to failure. The proposed

initiation size is derived fiom fracture mechanics and has the following form:

where F is a dimensionless function of geometry and c is the notch depth. The value of F

c m be found fiom handbooks listing stress intensity factors such as the one by Tada el al.

(1978). For moderate to sharp notches, the value of 1 ' falls in the range of r/4 to r/20. For

the Low Kt coupon. the value of 1' is approximately 0.05" (1.27 mm), whereas for the

High Kt coupon, the value is approximately 0.0 1" (0.254 mm).

Dowling demonstrated that combined initiatiodpropagation predictions were in excellent

agreement with test results for the initiation and total life of blunt and sharply notched

coupons made fiom AISI 4340 steel.

Dowling's approach shows promise since the notch size effect is considered in a more

rigorous manner ihan in the method which considers Kfïnstead of Kt, namely, the crack

propagation out of the highly stressed volume of material at the notch root. When used in

conjunction with crack initiation and growth measurernents, this method would prove

usefùl in predicting total life in fatigue test progams such as the spectrum truncation

sensitivity studies at SMPL-IAR.

8.4 Material Properties

One source of error cornmon to al1 fatigue predictions methods is the applicability of the

material data used. In this study, it is not known how well the CC189 materials database
models the behaviour of the materials and f o m s used at SMPL-IAR. ASTM Standard E

606-92 (ASTM? 1995) describes the standard practice for strain-controlled fatigue testing,

and Iists recommended requirements when reporting data. The requirements include:

rnaterial and specimen description; description of equipment, testing environment. testing

conditions and procedures; and finally, the test results for the cyclic stress-strain and

strain-life properties. A simila.reporting scheme was used by Endo and Morrow (1969)

in their study on cyclic stress-strain and fatigue behaviour of representative aircrafi

metals. By contrast, the CC189 materials database only includes the cyclic stress-strain

and strain-life cmres for the materials,

Even though this is the case? no logical reason exists to use data f?om other sources

(Straznicky, 1996). nie values of the material fatigue properties are only one factor
contributing to the inaccuracy of the local strain predictions. Therefore it cannot be

decided wïthout support £tom a detailed study that the use of one particular data set will

provide the best agreement with experimental results.

The -189 materials database contains strain-life data for duminum 7050-T74 in

prestrain and non-prestrain conditions. The difference between the prestrain and non-

prestrain data for alurninum 7050-T74 is shown in Fiemire 5-6. The effect of the

prestraining reduces the cycles to failure for a given strain amplitude. Figure 8 - shows

predictions made for the Low Kt coupon subjected to the tef-man05 sequence using the

prestrain and non-prestrain material data. The prestrain predictions are closer to the test

results, while the non-prestrain predictions are non-conservative with respect to the test

results. This limited study suggests that the prestrain data should be used to account for

the influence of large cycles on the following smaller ones in variable amplitude loading.

The use of strain-life data generated with periodic overloads seems to be a promising

approach when considering the damage accumulation problem described in Chapter 2.

However strain-life data generated with periodic overloads is not available in the

literature for the matenal considered in this thesis. In general, such data is only available

for specific research studies.

8.5 Equivalent Strain Equations

In Chapter 7, it was demonstrated that the LS predictions made using the FE results for

the Hi& Kt coupon were conservative with respect to the predictions made using

Neuber's rule. This was explained in terms of the multiaxial stress state at the notch root
of the Hi& Kt coupon and the use of the SWT equivatent strain equation. Even diough

the S W T equation is commoniy used in the literature, for example by Bergmann et al-

(1979), the inclusion of the SWT equation in the LS predictions for this study must be

justified. Figure 8-3 presents LS predictions for the High Kt coupon subjected to the

tef-man05 sequence. The predictions were made using the FE results and for the

following equivalent strain equations: SWT, LOOPINI, Modified Goodman, Gerber, and

Soderberg. The prediction made using Neuber's rule with the SWT equation is plotted

for reference on Figure 8-3. The results show a large difference between the use of the

SWT equation and the LOOPIN8 and Gerber equations. According to Bannantine et al.

(1990). the Soderberg equation is seldom used since it is considered to be conservative.

while test results generally fall between the Goodman and Gerber curves. Fomess et al.

(1989) recornmend the use of the SWT equation due to its independence fiom empincally

derived constants as in the case of the LOOPINI equation. In addition, the SWT equation

becomes undefined when omax<O,

indicating that the SWT equation predicts zero darnage

for a closed hysteresis loop that is fùlly compressive. For the other equations presented,

this condition m u t be irnposed, while for the SWT equation, the condition is "built-in".

For this study. the SWT equation has shown good agreement with test results for the Low

and High Kt coupons, and shodd continue to be used in EFOSTP.

Chapter 9 Conclusions-
9.1 Conclusions

This thesis analyzed the applicability of Neuberos rule in local strain crack initiation

predictions. In addition. the ability of the local strain method to predict the lives of the

two coupons used in this study was assessed. Based on the research performed for this

thesis, the following conclusions are drawn:

1. Local strain crack initiation predictions contain a number of assumptions which can

lead to considerable error. Two particular areas of concem identified in this study are

the damage accumulation problem and the principle of equivalence. The damage

accumulation problem arises fiom predicting the fatigue behaviour in spectnim

loading by linear damage accumulation calculated from smooth specimen constant

amplitude strain-Iife data (at &=-1). Concems with the principle of equivalence

include determination of stress and strain at the notch of an engineering component.

and the effect of notch severity on fatigue life.

2. The notch root of the Low K, coupon is very close to a state of plane stress. The notch

root of the High K, coupon is constrained, leading to a rnultiaxial stress state.

3. Neuber's rule is adequate to predict the notch root stress and strain of the Low Kt

coupon. In the case of the High Kt coupon, the Ho&ann and Seeger modification to

Neuber's rule provides adequate estimates of the notch root stress and strain. If

possible, elastic-plastic FE analyses should be performed to verify the applicability of

Neuber's rule in fatigue test programs.

4. Experimental observations and a limited fracture mechanics study suggest that the

crack propagation phase in the Low K, coupon is small? whereas in the High Kt

coupon, it could account for a significant portion of the total life.

5. Local strain predictions for the trailing edge flap hinge moment sequences dernonstrate

a large difference in predictive capability behveen the Low and High Kt coupons. This

is due to the differing crack propagation phases in these coupons.

6. Althougli the fatigue concentration factor provides local strain predictions which are

closer to the test results (total life) for the High Kt coupon, its use is questioned for

three reasons: the availability and applicability of material data; the modification of

the stress-strain estimation which results f?om the use of Kt; and the empirical manner

in which this method accounts for the crack propagation phase in sharply notched

coupons.
7. A method to predict total life (initiation and propagation) has been idenhfied which

accounts for the notch size effect in a more rigorous marner than in the method which

uses the local strain approach with the fatigue concenmtion factor.

9.2 Recommendations for Future Research

The following is proposed for future research in the area of the prediction of fatigue crack

initiation:

1. Isolate crack initiation and monitor crack growth in future coupon test programs at

SMPL-IAR with the intent of estimating total life using the combined

initiatiodpropagation approach identified in this study.

2. Develop an in-house rnatenals database for the materials appropriate to the specific

CF-18 locations under study. Although this is a large undertaking, any uncertainty in

using CC189 material data at SMPL-IAR would be removed. This effort should

ideally include the development of non-prestrain, prestrain and periodically overloaded

strain-life data in addition to the development of cyclic stress-strain properties.

3. Study the problem of damage accumulation in the local strain approach. This would

provide insight into the problern of predicting the fatigue behaviour in spectnun

loading by linear damage accumulation calculated fkom constant amplitude strain-life

data. The study should include darnage parameters and non-linear damage rules, both

of which may better represent the fatigue damage under spectrum loading.
4. Study the growth of microsmictural flaws to characterïze crack initiation at notches.

The behaviour of such flaws requires the use of Elastic-Plastic Fracture Mechanics.

9.3 Surnmary of Contributions

This thesis has made the folIowing contributions to the general knowledge in the field of

the prediction of fatigue crack initiation:

1. The applicability of Neuber's rule to plane stress situations was verified by an elastic-

plastic finite element analysis.

3. A method of estimating multiaxial elastic-plastic notch stresses and strains was

verified to be an effective means of accounting for notch root multiaxiality.

References
Amencan Society for Testing and Materials (ASTM). (1995). Standard E 606-92.
Standard Practice for Strain-Controlled Fatigue Testing.

Atnerican Society- for Testing and Materials (ASTM). (1995). Standard E 1049-85.
Standard Practices for Cycle Counting in Fatigue Analvsis

Bannantine, J.A., Corner, I.J., Handrock, J.L. (1990). Fundarnentds of metal fatigue
analvsis. Englewcod Cliffs, NJ, U.S.A.: Prentice-Hall Inc.

Baotong, L., Xiulin, Z. (1993). An approach for predicting fatigue crack initiation life of
a low alloy steel below room temperature. Engineering Fracture Mechanics. 46 (2): 339-
346.

Bergengren, Y.. Gustavsson~A.' Larsson, M., Melander, A., Bork' C.P., Golmano, M..
Hünecke, J.. McDowell. D.L. (1993). Fatigue properties under constant and variable
amplitude ioading of notched fati.gue specimens of a high strength sheet steel (Report No.
M-3 121). Stockholm, Sweden: Swedish Institute for Metds Research.

Bergmann. J., Seeger, T., Weisgerber, D., Sippel, K.O. (1979). The sensitivity of the
local approach to the quality of input data in the fatigue service life prediction of aircraft
components. Proceedines of the International Svrnposium on the Low-Cvcle Fatigue
Strength and Elasto-Plastic Behaviour of Metals. Stuttgart. 1979.30 1-3 17.
Bleuen, C.?Chaudonneret. M., Farcy, L.? Flavenot J.F.. Ranganathan. N. (1994). Fatigue
testing and life prediction for notched specimens of 2024 and 7010 ailoys subjected to
aeronauticd spectra. In C. Amzallag (Ed.), Automation in Fatiwe and Fracture: Testing
md Analysis. ASTM STP 1231 (pp. 508-530). Philadelphia, PA, U.S.A.

Broek, D. (1989). The practical use of fracture rnechanics. Dordrecht, The Netherlands:
Uuwer Acadernic Publishers.

Buch? A. (1 980). Verification of fatirne crack initiation life prediction results ( T A . No.
400). Haifa. IsraeI: Israel hstitute of Technology.

Conle' A., Topper, T.H. (1980). Overstrain effects during variable amplitude service
history testing. International Journal of Fatirne, 2 (3,
130-136.

Doerfler. M.T. (1997). An evaluation of service life analysis of metallic airfrarne

structure with MSC/FATIGUE. MacNeal-Scwendler Corporation Website. Paper 3397.
Website: http://vmw.rnacscli.com/aerospace/

Dowling N E . (1979). Notched member fatigue life predictions comblliing crack

initiation and propagation. Fatime of Engineering Materials and Structures. 2: 129-138-

DuQuesnay. D.L.. MacDougall. C., Dabayeh, A.. Topper, T.H. (1995). Notch fatigue
behaviour as influenced by periodic overloads. International Journal of Fatigue, 17 (3,
9 1-99.

Endo, T., Morrow, J. (1969). Cyclic stress-strain and fatigue behaviour of representative
aircraft metals. Journal of Materïals, 4 (1 ), 159- 175.

Engineering; Sciences Data Unit (ESDU). (1983). Elastic stress concentration factors
peometric discontinuities in flat bars or strips of isotropic materials. Item No. 69020.
Forness, S.D.. McFarland. J.L.. Meyer, ES., O'Neill, LW. (1989). Final Report:
Structural methods standardkation team number 8. Task: Crack initiation- McAir.

Foster. W.A. (1993). An investigation into the fatigue crack initiation p r o m m C CI89
(Technical Note 45). Melbourne, Australia: Department of Defense, Defense Science and
Technology Organisation. Aeronauticd Research Laboratory.

Glinka G.. Newport. A. (1987). Universal features of elastic notch tip stress fields.
International Journal of Fatigue. 9 (3)- 143-150.

Glinka G., Ott W.? Wowack H. (1988). Elasto-plastic plane strain anaiysis of stresses and
strains at the notch mot. Journal of Enszineering Materials and Technology, 110 (3): 195-
204.

Heitmann. K., Neumann. P.: Vehoff. H. (1983). Random load fatigue of steels: service
life prediction based on the behaviour of micro-cracks. Proceedings of the International
Conference on the Application of Fracture Mechanics to Matends and Structures.
Freiburg. FRG, 1983.

Heuler. P.. Schütz. W. (1986). Assessrnent of concepts for fatigue crack initiation and
propagation life prediction. Journal of Materials Technolow. 17 (1 1): 397-405; 17 (12),
449-456.

Hewitt, R.L.. Hiscocks, R.J., Bernard, G. (1996). Load spectnim determination for an
aircraft with a digital flight control systern - CF-18 centre fuselage exarnple. Canadian
Aeronautics and Space Journal, 42 (1)- 5-9.

Hibbit, Karlsson & Sorensen (HKS) Inc. (1997). ABAQUS/Standard User's Manual.
Vol. 1-III. Version 5.7.
Hoffmann M., Seeger T. (1985). A generalized method for estirnating rnultiaxid elastic-
plastic notch stresses and suains, Parts 1 and 2. Journal of Engineering Materials and
TechnoIom. Transactions of the ASME. 107,250-260.

James' M.? Swenson, D. (1997). FRANC2D/L: A crack propagation simulator for plane
lav-ered structures. Short user's guide version 1.4.

Kahonen A. (1991). FEM and fatigue life calculation usine the PC-based NISA
LIENDLIRE software package (Report No. 735). Espooo Finland: Technical Research
Centre of Finland.

Klohr, J.D. (1990). Computer program documentation for C CI89. McDonnell Aircraft
Company

Manson S.. Hirschberg M. (1966). Crack initiation and propagation in notched fatigue
specimens. Proceedings of Sie First International Conference on Fracture. Vol. 1. 1966.

Massing, G. (1926). Proceedings of the 2nd International Congess on Ap~Iied

Mechanics. Zurich. 1926.

Miller. K.J. ( 1987). The behaviour of short fatigue cracks and their initiation, Part Il - A
general surnmary- Fatigue and Fracture of Engineering Materials and Structures, 10 (3,):
93-1 13.

Molski K., Glinka G. (198 1). A method of elastic-plastic stress and strain calculation at
a notch root. Materials Science and Engineering, 50 (1): 93-100.
Nelson. DY.,Socie, D.F. (1982). Crack initiation and propagation approaches to fatigue
analysis. In P.R. A b e k s & C.M. Hudson (Eds.), Design of Fatigue and Fracture
Resistant Structures. ASTM STP 761 (pp. 110-132). Philadelphia, PA, U.S.A.

Neuber H. (1961). Theory of stress concentration for shear-strained prismatical bodies

with arbitrary nonlinear stress-strain law. Journal of Applied Mechanics. Transactions of
the ASME, 28 (4), 544-550.

Nie K.. Wu F.M., Liu J.F. (1994). A variable Kr - Neuber's rule for predicting fatigue
crack initiation life- Fatigue & Fracture of Engineering Matends & Structures, 17 (9).
1015-1023.

Peterson R.E. (1959). Notch-sensitivity. In Sines & Waisman (Eds.), Metal Fatieue @p.
293-306). New York, U.S.A.: McGraw-Hill.

Pilkey, W. (1994). Formulas for stress. strain. and structural matrices. New York.
U.S.A.: John Wiley & Sons Inc.

Porter, P.G. (1983). A rapid method to predict fatigue crack initiation Volume Ii -
computer program user's instructions. Northrop Corporation - Aircraft Division.

Schütz. W. (1979). The prediction of fatigue life in the crack initiation and propagation
stages - a state of the art survey. Engineering Fracture Mechanics, 11 (2): 405-421.

Seeger T., Beste A., Arnstutz H. (1977). Elastic-plastic stress-strain behaviour of

monotonie and cyclic loaded notched plates. Proceedings of the 4L" International
Conference on Fracture. Waterloo. ON.. Canada, June 1977,943 -95 1.
Sharpe W.N.' Yang CH., Tregooing R.L. (1992). An evaluation of the Neuber and
Glinka relations for monotonie loading. Journal of Applied Mechanics. Transactions of
the ASME, 59 (21, S50-S56.

Simpson, D.L. (1997). Review of Canadian aeronautical fatigue work 1995-1997

(Report No. LTR-ST-2094). Ottawa ON.. Canada: National Research Council Canada
hstitute for Aerospace Research.

Smith. K.N., Watson, P., Topper. T.H. (1970). A stress-strain function for the fatigue of
rnetals. Journal of Materials. 5 (4), 767-778.

Society of Automotive Engineers (SAE). Inc. (1988). SAE Fatiwe design handbook
AE- 1O (znded.).

Straznicky, P. (1996). Effect of low amplitude cycles in spectrum loading (Report No.
IAR-CR-3 1). Ottawa ON.? Canada: National Research Council Canada Institute for
Aerospace Research.

Tada H.. Paris. C.. Invin, G. (1978). The stress analysis of cracks handbook. St. Louis,
U.S.A.: Del Research Corporation.

Tipton S. (1991). A review of the development and use of Neuber's rule for fatigue
analysis. SAE Transactions. 100 ( 5 ) , 143-148.

Topper T., Wetzel R., Morrow J. (1969). Neuber's rule applied to fatigue of notched
specimens. Journal of Materials. 4 (l), 200-209.

Tregoning R.L. (1992). Elasto-plastic notch root strains: Measurements versus finite-
elernent predictions. Experimental Techniques, 16 (6), 4 1-45.
Umeda H.. Sakane M., Ohnami M. (1987). Cornparison of local s d n at the notch root
beisveen FEM analysis and experimental strain measurement under creep-fatigue
conditions, JSME International Journal. 30 (268), 1543-1550.

Walker E.K. (1977). Multiaxial stress-strain approximations for notch fatigue behaviors.
Journal of Testin.2 and Evaluation, 5 (2). 106-113.

Weiss. J.. Note to Tuegel, E.' 22 Apnl? 1997.

Weixing. Y., Kaiwuan. X., Yi- G. (1995). On the fatigue notch factor, Kg International
Journal of Fatieue, 17 (4)- 2 6 - 2 5 1.

Young, W.C. (1989). Roark's formulas for stress & strain (6Ih ed.). New York. U.S.A.:
McGraw-Hill.
 Smooth Specimen Notched Specirnen

Figure 2-1: Equivalence betiveen Smooth and Notched Specimens

Figure 2-2: Local Strain Method - Load Spectnim and Cyclic Stress-Strain Cunre
 Strain

Strain

Figure 2-3: Local Strain Method - Hysteresis Loop Tracking

 Blunt Notch Sharp Notch

Figure 2-4: Volume of Critically Stressed Material at Blunt and Sharp Notches
 Section A-A
Low Kt Coupon (EnIarged)

Section A-A
High Kt Coupon
(En larged)

Figure 4-3: Possible Crack Initiation Sites

 Load Spcdnim File =
R t f t r t n c c S h a h = 1.000000
Tuming Points = 0
flight Hours per Block = 0.000000
Kt = 1.000000

Cydic Suess-Shin Data =

Strain-Cife CU- =
Modulus of Elasticity = 1.000000
Cycfic Proporüonal Umit = 1.0001100

-.- --
-1 NUM ,

Figure 5-1: McCracken Prediction Environment

- Cydic Siress m Sberr5ham Data ---- ---.-

UTS
17

Figure 5-2: Material Properties Dialog Box

. Spectrum Information --______- -_----_---- -

-
Kt Stress Conwnirdion
1

M aK tmS KWLS lncrement Nu* d Points

O O 10

Figure 5-3: Spectrum Dialog Box

- R a * s f o r N d Sbessvs. Ndch RodSbarsfftrain--- -

13 Options:
r ~euber'~~uie-~- F S& N e w s Ruie aCe C-Ci89

r R w ü s frornFEA
-------
-cyde OrdtrLig
l
-Countn?gMethod - - -- -
' [3 Peak to Frm 13 tlosed Hyrteresis b p s ,
--

Figure 5-4: Prediction Methods Dialog Box

 Figure 5-5: Cyclic Stress vs. Stress*Strain Curve - Al 70504'74

I.OOE+Ol 1.00E+02 1.OOE+O3 1.OOE+O4 1.00E+05 1.OOE+O6 1.00E+07

Life [cycles]

Figure 5-6: Strain-Life Curve - AI-7050-T74

 1O000 100000
Life Efh]

Figure 5-7: Cornpanson of SWT and LOOPIN8 Equivaient Strain Equations

Figure 5-8: C-Ci89 Representation of Stress vs. Stress*Strain Cuwe

 -- l
-Stress vs. Plastic Strain
b

/
6

1
6
- -- ,
a - - - - -

__- - r--- - - r
- - - -

I l

O 0.01 0.02 0-03 0.04 0.05 0.06 0.07

Strain [inlin]

Figure 6-1 :Aluminum 7050-T74 Stress-Strain Curve

-432 elements
,- 3600 elements j
f
i _ _ _- ___

0.75 0.85
x [in]

Figure 6-2: Low Kt Coupon Mesh Convergence Study

 Figure 6-3: Finite EIement Geometry of Low Kt Coupon

Yield Stress = 40 [ksi] --

i - - - - - -Glinka+Newport
- - .1
4

I - -
Notch Root Radius = 0.587 [in]
i - approximate
. :-. -
relation
- - r -
j I
I
0
I

l 1 I !

Figure 6-1:Cornparison between FE results and approximate relationship for stress

vs. distance
 0-6 0-7
x [in]

Figure 6-5: High Kt Coupon Mesh Convergence Study

IL,
Figure 6-6: Finite EIement Geometry of High Kt Coupon
 . ---..-----A - --- ----
- .. -

i l ~ < Ï & a lstress/'fietd Stress

-
I i
Yield Stress = 40 [ksi]
.i I-
L - -

0.71 \ -
:--- j-031
.- - - -
! I
Notch Root Radius = 0-587 [in] -- - -i - - I
,--
i ! 1
t
t
-

I
a - - - -
- -1.28 j-
1.54 1

Figure 6-7: SP3 vs. Distance from Notch Root - Mid-Thickness of LOW Kt Coupon

Figure 6-8:SP3 vs. Distance from Notch Root - Surface of LOWKt Coupon
Figure 6-9: EP3 vs. Distance from Notch Root - Mid-Thickness of Low Kt Coupon

Yield Stress = 40 [ksi] ~ & n i n a~tresshield

l Stress
E = 10000 [ksi] ,
, .. _ _ _I
. . -

-* - !
-
0.71
Notch Root Radius = 0,587[in] ; l___
: 1 1.O3

Figure 6-10: EP3 vs. Distance from Notch Root - Surface of Low Kt Coupon
 O 0.1 0.2 0.3 0-4 0.5 ; - - ' -- 1,671
- .-- 0-7
xlr

Figure 6-11: von Mises Stress vs. Distance from Notch Root - Mid-Thickness of Low

Kt Coupon

-- . - - -
-. - StressMeid
Yield Stress = 40 [ksi] -- - - -- - -
t A
Nominal Stress
- - - - - -.
f l

Notch Root Radius = 0.587 [in] ,!

-

1
I V
0.71 i

Figure 6-12: von Mises Stress vs. Distance €rom Notch Root - Surface of Low Kt
Coupon
 -
Figure 6-13: SP3 vs. Distance from Notch Root Mid-Thickness of High Kt Coupon

- - -
l
Nominal ~ t t e s s ~ i eStress
ld
Yield Stress = 40 [ksi] i--
1

Notch Root Radius = 0.125 [in] I,- - - - 0.32i

0.68 1
' * * - - - -1.01 !

Figure 6-14: SP3 vs. Distance from Notch Root - Surface of High Kt Coupon
 I l
Yield Stress = 40 [ksu Nominal StressMeld Stress
I
E = 10000 [ksi] O.%!! ---
\ Notch Root Radius = 0-125 [in] i,,,0.68 !

O 1 4 5
YIr

Figure 6-15: EP3 vs. Distance from Notch Root - Mid-Thickness of High Coupon

Yield Stress = 40 [ksq c - Stress

Nominal StressNield
-

E = 10000 [ksi] - -- P. 0.32 i -- - -. -- .

Notch Root Radius = 0.125 [in] : - - - 0.68 '

I
- - - - - -1.01 ,
. -. - - - -A - -- - -- - - - .
---- 1.35
'

r----

Figure 6-16: EP3 vs. Distance from Notch Root - Surface of High Kt Coupon
 -
Figure 6-17: von Mises Stress vs. Distance from Notch Root Mid-Thickness of High

Kt Coupon

Figure 6-18: von Mises Stress vs. Distance from Notch Root - Surface of High Kt

Coupon
 O 10 20 30 40 50 60
Nominal Net Section Stress [ksi]

Figure 6-19: Notch Root Stress vs. Net Section Nominal Stress - Low Kt Coupon

O 10 20 30 40 50 60
Nominal Net Section Stress [ksi]

Figure 6-20: Notch Root Strain vs. Net Section Nominal Stress - Low Kt Coupon
 10 20 30 40 50
Nominal N e t Section Stress [ksi]

Figure 6-21: Notch Root Stress vs. Net Section Nominal Stress - High Kt Coupon

10 20 30 40 50
Nominal Net Section Stress [ksi]

Figure 6-22: Notch Root Strain vs. Net Section Nominal Stress - High Kt Coupon
 O 10 20 30 40 50 60
Net Section Nominal Stress [ksi]

Figure 7-1: Low KrCoupon - Stress Estimation

- - - -.- - . -L . - -. 1. 1 1
1 -
Neuber i I

1 ... ! - -- 6- - --
' O O 1
I Neuber (Nonlinear) I I

-FE Results - Mid-Thickness i.- -_ . . . - - . . . .- ..-

I
... -. I
-r --
. +Glinka ESED

-
-
u U

O IO 20 30 40 50 60
Net Section Nominal Stress [ksi]

Figure 7-2: Low Kt Coupon - Strain Estimation

 , _ - - _ --FE Results - Mid-Thickness
/ +Neuber-Modified by Hoffmann and Seeger
1 - - il - -Neuber (Nonlinear)-Modified by Hoffmann and Seeger :

O 10 20 30 40 50 60
Net Section Nominal Stress [ksi]

-
Figure 7-3: High Kt Coupon Stress Estimation

O 1O 20 30 40 50 60
Net Section Nominal Stress [ksi]

Figure 7-4: High Kt Coupon - Strain Estimation

 1 O000
Life [sfh]

Figure 7-5: Low Kt Coupon - Crack Initiation Prediction Sensitivity Study

I O tef-man05 - test results @ K fDLS

.1 A
. . ,
tef-sumo5 . -test results @ .K- f-.
7-r-i DLS
; 1 :-i- - -

1 O000
Life [sfh]

Figure 7-6: High Kt Coupon - Crack Initiation Prediction Sensitivity Study

 . -.
1-
Law Kt Coupon j --
i
-1
!
k-
- -- -: -L - -i/-L-

0.00 0-05 0.10 0-15 O -20 0.25 0.30 0.35

a [in]

Figure 8-1: Stress Intensity Factor vs. Crack Length

10000
Life [sfh]

-
Figure 8-2: Low Kt Coupon Prestrain and Non-prestrain LS Predictions
 -
Appendix A Coupon Test Program
1. Loiv Kt Coupa

Mode1 No. Serial No. r -

Capacity 1 Calibration

I LVDT 602

1392
N/A

250 kN / 55 kips
12 March 1997

12 March 1997

Hydraulic Grips 449 350 kN / 55 kips N/A

1

446 250 kN / 55 kips N/A

Schematic and Calibration Information for 55 kip Load Frame

 COUPON TEST SET-UP - LOW Kt
Load Frame #2

Test Specimens

Specimens for the trailing edge flap study were used. The coupon is called "Low KZ', and
is made from alurninum 7OSO-T7451.

Test Procedure

Nore : Aper 5 specimens, remove the shimsfrom hydrauZic gr@ and grit blast. Then
re-instaZl in their prei~ioioirsZocations (murked).

1. Start TestStar and open appropriate configuration file (*.TCC).

- for Lotv Kt: E F-A1 .TCC

2. Start hydraulic system.

- Reset the POD (if errors still there, turn APC control switch "on"). vene that CLC
mode is selected.
- Put HPS on Yo+?. then "high".
- Put HSM on "low" and extend actuator s
- Put HSM on "high".

3. Check tuning values in "CLC" and "Force Sg" modes.

The following values should be used :

- CLC? for both titanium and alurninurn : P = 30 (master)?P = 6 (limit)

- Force Sg, for aluminum : P = 7.50,I = 0.9: D = 0.02, F = 0.04

Note :rhese values shozdd be iunedfor a speczjic coupon andor specrrzrrn.

1. Check input signais range and limits

- A) Load transducer
for man05: range = 10000 Ibf
upper limit = 10000 lbf
lower lirnit = -5500 Ibf
for sum05: range = 25000 Ibf
upper limit = 11000 Ibf
lower limit = -5500 Ibf
- B) Displacement
range = 5 inches
 Zero out load transducer and lock zero.
on POD, switch fiom present mode to the "Displ Pod" mode
Teststar window : > Adjust
3 Input Signals
> Force
> Check and record shunt calibration d u e s
> Unlock zero
> Autozero
> Lock zero
on POD. switch from "Displ Pod" to "CLC"

lnstall specimen in hydraulic grips

make sure that the "CLC" mode is selected, make sure APC set to OFF.
place crosshead 47-48 inches above the base, corresponding to a post length of 47-48
inches,
place specimen (selected in random order) in bottom grip and center it between the
two screws that are needed to maintain the inseas. The specimen should be placed
with the reference hole at the top right position.
tighten specimen slightly in bqttom grip.
extend acniator slowly with the shim on top of the specimen so that the shim just
enters the top gnp inserts.
extend hydraulic cylinder for specimen to be the closer to its final position. Use the
distance fiom bottom grip to notch as a rough estimate of how far to raise actuator.
turn APC to OFF.
use square and insert built for this purpose to align specimen. Tighten bottom screws
first to align the square to the actuator. Apply hydraulic pressure to bottom grip.
remove the square from the actuator once alignrnent is complete.
extend actuator until shim has bottomed out in top grip ( Do not exceed -5 to -8 Ibf of
compressive force when doing so). Tum APC to OFF, then apply hydraulic pressure
to top _ep.

Check on any Ioad change (with hydraulic pressure applied to grips, ioad
should be between 30 and 40 Ibf)

8. Zero out displacernent transducer and lock zero.

- on POD, switch fiom "CLC" to "Force POD"
- TestStarwindow-: > Adjust
> Input Signals
> Displacernent
> Switch range fiom 5.0 in to 0.5 in
(configuration should be saved with 5.0 in range to
dlow installation of specirnen)
> Unlock zero
> Autozero
 > Lock zero
- on POD. switch fiom "Force POD" to "CLC" and make sure APC set to OFF

9. Start TestWare-SX and open appropriate template (WJJVGTST.000)

Select desired procedure :

- UTNGTST Default Procedure

- TEF-ACT Mean05 haversine AI
- TEF-ACT S m 0 5 haversine Al

10. Change data file name for each specimen

11. For the first specimen in a series, create a new SAC file

NOTES:

1. Run man05 with 300% rate multiplier, and sumo5 with 200% rate multiplier.
2. To remove coupon, set displacernent range to 5".
TestWare-SX Procedure File for t e m a n 0 5 Sequence - Low Kt Coupon

Procedure Narne = TEF-ACT mean05 haversine AL

File Specification = C:\TS2\TWSX\WINGTST~O10
Software Version = 3.OB
Printout Date = 03-1 0-1998 LZ:O4:5 1 PM

Data File Options

File Format = Excel Text File
Log Events = Yes
Include Procedure Description = No

Recovery Options
Autosave disabled.

Ramp to mean : Step

Step Done Trigger 1= ramp up

ramp up : Monotonic C o r n a n d
Start Trigger = Step Start
End Trigger = <none>
Segment Shape = Ramp
Time = 3 (Sec)
AxrAL
Control Mode = FORCE SG
End Ievel = 50 ( Lbf)

Spectrum Cycling : Step

Step Done Trigger 1 = Play Spectrurn

Record PeaksNalleys : Data Acquisition

Start Trigger = Step Start
End Trigger = Play Spectnim
Mode = Peak / Valley
Buffer Type = Circular
Master Channel = FORCE
Data Header = PeaksNalIeys
Sensitivity = 10 ( lbf )
Buffer Size = 16000

Play Specû-um :File Playback Command

Start Trigger = Step Start
End Trigger = <none>
File Narne... = C:\TSITWSX\TE-MANOSSFP
Passes =O
Rate Multiplier = 300 ( % )
Compensation = SAC
AXIAL
Control Mode = FORCE SG
Level Reference = O ( lbf )
Level Multiplier = 100 ( % )
Table Options
Create new SAC table. = No
Load SAC table fiom file. = Yes
Use existing SAC tabIe. = No
Additional Options
Save SAC table to tile. = No
Default Table Limits = No
Error Tolerance = 90 ( lbf)
Upper Table Limit = 10000 ( lbf )
Lower Table Limit = -5500 ( lbf )
SAC Load File.., = C:\TS3\TWSX\TEFMOSSAC
SAC Save File ... =
2. High Kt Coupon

I Component 1 Mode1 No. 1 SenalNo. 1 -

Capacity Calibration
I
LVDT N/A 101 22 January 1998
1
I
Load Cell

Hydraulic Grips
661.2lA-03

647.10A
1
4684

02841 11
I 100 kN / 22 kips

100 kN / 22 kips
23 January 1998
I
100 kN / 22 kips

Schematic and Calibration Information for 22 kip Load Frame

 COUPON TEST SET-UP - HIGH Kt
Load Frame #8

Test Specimens

Specimens for the trailing edge flap study were used. The coupon is called "High Kt?''
and is made fiom aluminum 7050-T7451.

Q Test Procedure

1. Start TestStar and open appropriate configuration file (*.TCC).

- for High Kt: TEF-AP.TCC

2. Start hydraulic system.

- Reset the POD (if errors still there, turn M C control switch "on")? ver@ that CLC
mode is selected.
- Put W S on "Io+', then "hi*'
- Put HSM on "low" and extend actuator slightly to ensure it is free
- Put HSM on "hi&"

3. Check tuning values in "CLC" and "Force Sg" modes.

The following values should be used :
- CLC, for both titanium and duminum : P = 30 (master), P = 6 (Limit)
- Force S g for aluminum :P = 7.50,I = 0.9, D = 0.02. F = 0.04

Abte :these values should be tlrnedfor a specijic coupon an&r specfrum.

4. Check input signals range and limits

- A) Load transducer
for man05: range = 10000 Lbf
upper lirnit = 10000 lbf
lower Iimit = -5500 Ibf
for sum05: range = 25000 lbf
upper Iimit = 1 1000 lbf
lower Iirnit = -5500 lbf
- B) Displacement
range = 5 inches
 Zero out load transducer and lock zero.
on POD, switch fiom present mode to the "Displ Pod" mode
TestStarwindow: > Adjust
> Input Signals
> Force
> Check and record shunt calibration values
> Udock zero
> Autozero
> Lockzero
on POD. switch fiom "Displ Pod" to "CLC"

Install specimen in hydraulic grips

make sure that the "CLC" mode is selected. make sure APC set to OFF.
place crosshead 38 inches above the base, corresponding to a post iength of 38 inches.
place specimen (selected in random order) in bottom g i p against iower stop. The
specimen should be installed with the reference hole at the top nght position.
tighten specimen slightly in bottom @p.
extend actuator slowly so that the specirnenjust enters the top grip wedges.
extend hydraulic cylinder for specimen to be the closer to its final position. Use the
distance fiom bottom grip to slot as a rough estimate of how far to raise actuator.
tm APC to OFF.
use upper stop to align specirnen. Ven& alignment with digital protractor. Apply
hydraulic pressure to bottom grip.
extend actuator until specimen bas bottomed OUI in top g i p ( Do not exceed -5 to -8
Ibf of compressive force when doing so). Tum APC to OFF, then apply hydraulic
pressure to top grip.

Check on any load change (with hydraulic pressure applied to grips, load
should be between 20 and 20 lbf)

8. Zero out displacernent transducer and lock zero.

- on POD, switch fiom "CLC" to "Force POD"
- TestStar window : > Adjust
> Input Signds
> Displacement
> Switch range fiom 5.0 in to 0.5 in
( ~ o ~ g u r a t i oshould
n be saved with 5.0 in range to
allow installation of specimen)
> Unlock zero
> Autozero
> Lock zero
- on POD. switch fiom "Force POD"to "CLC" and make sure APC set to OFF.
9. Start TestWare-SX and open appropriate template (WWGTST.000).
Select desired procedure :

- WINGTST Default Procedure

- TEF-ACT Meanûj haversine Al
- TEFACT Sumo5 haversine AL

10. Change data file name for each specimen.

11. For the first specimen in a series, create a new SAC file.

NOTES:

1. Run man05 with 300% rate multiplier, and mm05 with 200% rate multiplier.
2. To remove coupon, set displacement range to 5".
TestWare-SX Procedure File for tef-man05 Sequence - High Kt Coupon

Procedure Name = TEF-ACT mean05 haversine AL

File Specification = C:\TS2\TWSX\WINGTST.O IO
Software Version = 3.1 B
P ~ t o uDate
t = 03-10-1998 12: 10:20 PM

Data File Options

File Format = Excel Text File
Log Events = Yes
hclude Procedure Description = No

Recovery Options
Autosave disabked.

Ramp to mean : Step

Step Done Tngger 1 = ramp up

ramp up :Monotonic Command

Start Trigger = Step Start
End Trigger = <none>
Segment Shape = Ramp
Time = 3 ( Sec )
Control Channel I
Control Mode = Force Segment
End level = 50 lbf

Spectrum Cycling :Step

Step Done Trigger 1 = Play Specinim

Record PeaksNalleys :Data Acquisition

Start Trigger = Step Start
End Trigger = Play Spectrum
Mode = Peak / Valley
Buffer T*pe = Circular
Master Channel = Force
Data Header = PeaksNaIleys
Sensitivity = 10 lbf
Buffer Size = 16000

Play Spectrum : File Playback Command

Start Trigger = Step Start
End Trigger = <none>
FileName... =C:\TS2\TWSX\TE-MANO1.SFP
Passes =O
Rate Multiplier = 200 ( % )
Compensation = SAC
Control Channel 1
Control Mode = Force Segnent
Level Reference = O ( kip )
Level Multiplier = 80.44 ( % )
Table Options
Create new SAC table. = No
Load SAC table fiom file, = Yes
Use existing SAC table- = No
Additional Options
Save SAC table to file. = N o
DefauIt Table Lirnits = No
Error Tolerance = 90 lbf
Upper Table Limit = 10000 lbf
Lotver Table Limit = -5500 lbf
SAC Load File.., = C:\TS2\TWSX\TEFMOS.SAC
SAC Save File... =
 -
Appendix B Format of McCracken Input and

Results Files

The McCracken Fatigue Life Prediction Program (McCracken) is available through the

Department of Mechanical and Aerospace Engineering at Carleton University.

Individuals interested in obtaining the McCracken application are encouraged to submit a

wntten request to the following address:

Department of Mechanicd and Aerospace Engineering

Cnrleron University

1125 Colonel By Drive

Ottawa, Ontario. Canada

KIS 5B6

McCrackenVersion 21

Copyright @ 1997,1998
Sanieev K Visvanalha
1. Cyclic Stress vs. Stress*Strain Curve

Filename exiension: sed

Format of Data File:

3 &Y2
etc. etc.

n Number of data pairs (integer)

xn Stress*Strain at index "n'?
y, Stress at index "n?'

Data pairs ananged in order of increasing values of Stress*Str;tin. Le. increasing

Xn*
Exumple of Data File:
2. Strain-Life Curve

Filmame extension: -end

Format of Data File:

Ietc. - I etc. I

n Number of data pairs (intepr)

XII Strain Amplitude at index "n"
Y, Life at index "n"

Data pairs arranged in order of increasing values of Strain Amplitude. i.e.

increasing X,.

Exornple of Dala File:

3. Spectrum File

Filename extension: -trunc

Format of Spectrzm File:

1 AI BI CI DI EI FI
2 A2 BZ Ci D2 Ez F2
3 A3 B3 C3 D3 E3 F3
etc.

Notation:

1 A Spectrum flight reference nurnber 1

B B l d
C Gauge strain (FE)
D Nominal weight off wheels occurrence time
E Angle of attack
F Dvnamic pressure

Colmm Alignment:

The columns start at the following locations:

Example of Spectrzrm File:

etc.
4. Results File

Filename extension: .res

Format of Results File:

Load Spectrum File =

Reference S t r a i n =
Turning Points =
F l i g h t Hoürs pet B l o c k =
Kt =

Cyclic S t r e s s - S t r a i n Data =
S t r a i n - L i f e Data =
Modulus of E l a s t i c i t y =
Cyclic P r o p o r t i o n a l Limit =

Elapsed Time

Elapsed T i m e ( t o t a l seconds) =

Life
K t f IjLS
Life ( L 1 KttDLS (1)
I

etc. etc.
I
I
etc.
L i f e (n)
I KC+DLS (n)

Notation:

I n 1 Total number of Kt*DLS values for which life 1

predictions were made
Life(n) Life prediction for index "n"
1 K,*DLS(n) Kt*DLS for index Y? 1

Kt*DLS vs. Life data is arranged in order of decreasing K,*DLS. The first data pair will
be for the coupon DLS.
4. Results File - continued

Example of Reszdts File:

L o a d Spectrum
-------------
L o a d Spectrum F i l e = C : \ M c C r a c k e n \ L o a d Spectrums\wflu~~mean00000500,trunc
Refererice S t r a i n = 2083,000000
T u r n i n g Points = 33082
F l i g h t H o u r s per B l o c k = 326.000000
K t = 1-300000

Cyclic Stress-Strain Data = ~:\~ilcCracken\Warrerial\Aiuminum

from CI89 - EIP.sed
Strain-Life Data = C:\McCracken\Material\Aluminum from CI89 - P-end
X o d u l u s of Z l b s t i c i c y = 10000.000000
C y c l i c P r o p o r t i o n a l L i m i t = 10.000000

E l a p s e d Time
------------
E l a p s e d Time ( t o t a l s e c o n d s ) = 65,000000

Life
Kt' DLS
l4499.3OO363
3088.743874
4263.657052
5550 -064118
7850,674271
12622.446882
24444.55140â
56113.L46176
 IMAGE EVALUATION
TEST TARGET (QA-3)

-
APPLIED 4 IMAGE.inc
---
--
= 1653 East Main Street
-
---
--
--
--
-a Rochester, NY 14609 USA
Phone: 7161482-0300
Fm-7161288-5989

O 1993. Applied Image. Inc. Ail Rights Reserved