You are on page 1of 38

Analysis Methods Document

RAPIDC 2.0

Repair Assessment Procedure


and Integrated Design for Commuters

August 16, 2004


Prepared by:
Chun Chen

Reviewed by:
Jeffrey Tom
Program Manager

ii
Table of Contents
1.0 Introduction ...........................................................................................................................1

2.0 Static Module ........................................................................................................................1

2.1 The Mounting Plate/Repair Doubler Allowable and Margin of Safety...........................1


2.2 The Fastener Joint Allowable and Margin of Safety.....................................................2
2.2.1 Fastener Joint Allowable in Straight Shank Holes ............................................3
2.2.2 Fastener Joint Allowable in Countersunk Holes...............................................3
2.2.3 Fastener Joint Margin of Safety......................................................................4
2.3 The Shear Margin of Safety........................................................................................5
2.4 Margin of Safety as a Criterion...................................................................................5
2.5 The Stiffness Check of the Repair...............................................................................6
2.6 The Fastener Bending Check of the Repair .................................................................6
2.7 The Inter-Rivet Buckling Guidelines ............................................................................6

3.0 Damage Tolerance Module ....................................................................................................6

3.1 Critical Locations.......................................................................................................7


3.2 Initial Cracks and Subsequent Growth Assumptions................................................. 10
3.3 Fastener Load Transfers and Stress Gradients .........................................................16
3.4 Stress Intensity Factors ............................................................................................16
3.5 Residual Strength and Limit Stresses.........................................................................23
3.5.1 Residual Strength: Fracture Toughness Criterion............................................23
3.5.2 Limit Stresses...............................................................................................23
3.6 Load/Stress Spectrum and Equivalent Stress.............................................................24
3.7 Crack Growth Rate Data .........................................................................................25
3.8 Crack Growth Analysis Methods..............................................................................27
3.9 Inspection Threshold and Intervals............................................................................30

4.0. Analysis Procedure Flow Charts...........................................................................................30

5.0. General Notes and Program Limitations................................................................................32

6.0. References...........................................................................................................................32

Appendices

A - Static Strength Analysis of Fastener Joints


B - RAPID-FEM Analysis
C - Stress Intensity Factor Solutions
D - Load/Stress Spectrum Development
E - One-Cycle Equivalent Stress Calculation

iii
F - General Notes and Program Limitations

iv
1. Introduction

This document describes technical details of the analysis methodology implemented in the static strength
and damage tolerance analysis modules of the Repair Assessment Procedure and Integrated Design for
Commuters (RAPIDC) program, Version 1.4. In this document, the static strength analysis method for
repaired fuselage skin and skin modifications to antenna installation is first described. A description of
the damage tolerance analysis method is next provided.

RAPIDC is a software tool that assesses the static strength and performs damage tolerance analysis of
repaired skin and modified skin to antenna installation. The static strength analysis determines if
doublers and fasteners are statically adequate in the design, and the damage tolerance analysis provides
residual strength and crack growth to assist users in determination of inspection intervals for
maintenance schedule of the repaired/modified skin.

With the 2-D stress analysis capability, RAPIDC calculates load transfers at critical fasteners and stress
gradients along assumed crack growth paths in repairs and antenna installation configurations.

Static strength and damage tolerance analysis methods used in RAPIDC are described in subsequent
sections.

2. Static Strength Analysis

Typical repairs on the commuter aircraft fuselage skin generally involve removing the damaged area and
replacing it with repair doublers. The repair doublers are mechanically fastened to the skin around the
skin cutout.
Typical antenna installations on the fuselage of commuter airplanes generally involve cutting a hole in the
skin for the antenna installation. A doubler with a hole the same size of the skin cutout is then
mechanically fastened to the skin around the cutout.

To assess the static strength of repaired/modified skin, three independent criteria are used to evaluate
the margins of safety. The margins of safety are calculated using the doubler allowable, the joint
allowable, and the shear allowable.

2.1 The Doubler Allowable and Margin of Safety

(A) Repairs

To assess the strength restored by the doubler in a repair, the margin of safety of repair doublers is
calculated as

Pd u
Margin of Safety = −1
Pd

1
where Pdu , the doubler force allowable per inch, is calculated using the equation

N
Pd u = ∑ Ft u t D
1

in which Ftu and tD are the ultimate tensile strength of the doubler material and the thickness of each
repair doubler, respectively, and N is the total number repair doublers.

The skin internal force per inch, Pd , is calculated using the equation

L
Pd = ∑ σ u t s
1

where σu and ts are the design ultimate tensile stress of the skin material and the thickness of each skin
layer, respectively, and L is the total number skin layers.

(B) Antenna Installations

To assess the restoring strength capability of the doubler due to strength loss of skin cutout, the margin
of safety of the doubler is calculated as

Ppu
Margin of Safety = −1
Pp

where Ppu , the doubler load allowable, is calculated using the equation

Ppu = Ftu (wp – D) tp

in which Ftu, wp, D, and tp are the ultimate tensile strength, width, cutout diameter, and thickness of the
doubler, respectively.

The skin internal force, Pp , is calculated using the equation

Pp = σu D ts

where σu and ts are the design ultimate tensile stress and thickness of the skin, respectively.

2
2.2 The Fastener Joint Allowable and Margin of Safety

In the vicinity of a repair or an antenna installation on the skin, skin stresses are transferred to the
doublers through fasteners via fastener shear and hole bearing. The fastener together with the skin and
doublers represents a fastener joint. A joint can only transfer a certain amount of load until it fails. The
point at which this fastener joint fails is the joint allowable. The calculation of the joint allowable is
based upon two failure cases, fastener shear and hole bearing. Two hole types also need to be
considered when determining the joint allowable, the straight shank and countersunk.

2.2.1 Fastener Joint Allowable in Straight Shank Holes

Straight shank holes are used for protruding head fasteners and for doubler that have a flush head
fastener installed but are not countersunk. The joint allowable is the lower of the fastener shear
allowable or the hole bearing allowable. An allowable is calculated for the skin and doubler that the
fastener goes through.

The single shear allowable for straight shank holes is calculated using the following equation

Psu = Fsu A f S CF1

The double shear allowable for straight shank holes is calculated using the following equation

Psu = 2 Fsu A f S CF2

In both equations, Fsu is the ultimate shear strength of the fastener material, Af is the cross-sectional
area of each fastener, SCF1 is the single shear correction factor and SCF2 is the double shear correction
factor. SCF1 and SCF2 are used only for solid rivets and can be found in MIL-Handbook 5F Table
8.1.2.1(b).

The hole bearing allowable for the straight shank hole is calculated using the following equation

Pb r u = Fbru d t

in which Fbru is the ultimate bearing stress of the plate material, d is the fastener hole diameter, and t is
the thickness. In RAPIDC, the Fbru for the case e/d (edge distance to hole diameter ratio) equal to 2.0
is used.

The joint allowable for a given joint is the lower of Psu or Pbru denoted by Pjoint.

3
2.2.2 Fastener Joint Allowable in Countersunk Holes

The fastener joint allowables for countersunk holes are different from the straight shank holes. They are
determined by tests and can be found in MIL-Handbook 5F Section 8.

2.2.3 Fastener Joint Margin of Safety

The fastener joint margin of safety is determined by summing each fastener joint allowable in the skin
and doubler layers. It is calculated for each side of the doubler using the equation

(Pjoint )n
K
Ptotal = ∑
n =1

in which Ptotal is the total fastener joint load, (P joint )n is the fastener joint load for the nth fastener, and K
is the number of fasteners. If multiple skin layers are present, the fastener joint allowable for each skin
layer is added together for the Ptotal for the skin layers. If multiple-repair doubler layers are present, the
fastener joint allowable for each doubler layer is added together for the P total for the repair doublers.

The Ptotal for the skin and doubler are then compared. The smaller of the two is the fastener joint
allowable for that side of the doubler and is used in determining the margin of safety for the fastener
joints.

An applied load is needed to determine a margin of safety. That load is the design ultimate load of the
structure.

(A) Repairs

For repairs, the applied load Papplied is calculated by

L
Papplied = ∑ (σ u l t s )n
n =1

where σu is the design ultimate tensile stress, ts is the thickness of each skin layer, l is the length of
damage treatment (skin cutout) normal to the fastener load direction, and L is the number of skin layers.

The margin of safety is calculated by

Ptotal
Margin of Safety = −1
Papplied

(B) Antenna Installation

4
For antenna installations, this applied load Papplied is calculated by

Papplied = σu D ts

where σu is the design ultimate tensile stress, ts is the skin thickness, D is the diameter of skin cutout
hole.

A detailed description of fastener joint margin of safety calculation together with illustrative examples is
provided in Appendix A.

2.3 The Shear Margin of Safety

(A) Repairs

The shear margin of safety of a repair is calculated by

number of layers
∑ (Fsu t ) doublers
Margin of Safety = i=1 −1
number of layers
∑ (Fsu t) skin
k=1

where FSU is the ultimate shear strength of the skin or doubler material. The equation is used for each
side of the repair.

(B) Antenna Installations

The shear margin of safety of the antenna installation is calculated by

( FSU ( w plate − D ) t ) plate


Margin of Safety = −1
( FSU D t ) skin

where FSU is the ultimate shear strength of the skin or doubler material. The equation is used for each
side of the doubler.

2.4 Margin of Safety as a Criterion

The margins of safety (MOS) based on the doubler allowable and the fastener joint allowable are
calculated to determine the adequacy of the repair or antenna installation on fuselage skin.

5
• MOS < 0 Design is statically inadequate
• MOS = 0 Design is marginally adequate
• MOS > 0 Design is statically adequate

When the design is not statically adequate, it must be redesigned to ensure the adequacy of the design.
2.5 The Stiffness Check

The stiffness ratio between the doubler and the skin is calculated using the following equation

( E t ) doubler
Stiffness Ratio =
( E t ) skin

The design is considered adequate if the ratio is between 1.0 and 1.5. It is too stiff when the value is
greater than 1.5 and not stiff enough when less than 1.0.

2.6 The Fastener Bending Check

The fastener bending is checked using the following equation

ts+t p
Q=
d

where d is the fastener diameter, ts and tp are the thickness of skin and doublers, respectively. The
parameter Q is the fastener-bending indicator. For aluminum fasteners, the bending is important. A Q
value above 2 may indicate that the aluminum rivet will not fill the hole but instead may buckle in the
hole. In such a case, RAPIDC recommends steel or titanium fasteners be used. For steel and titanium
fasteners, there is no constraint for typical fuselage repairs and antenna installations.

2.7 The Inter-Rivet Buckling Guideline

To avoid inter-rivet buckling in the repaired skin, the fastener spacing should be four to six times the
diameter of the fastener shank diameter.

3. 0 Damage Tolerance Analysis

To perform the damage tolerance analysis of a repaired/modified skin, critical fasteners and their load
transfers as well as skin stress gradients at an open hole must first be determined. Assumptions of initial

6
flaws at fracture critical locations and the continuing damage need to be made. The stress spectrum
must also be prescribed. In addition, the following data are needed:

§ Crack growth rate data of the skin material


§ Fracture toughness of the skin material
§ Stress intensity factors of relevant crack configurations

Damage tolerance analysis procedure is described in the flowchart shown in Figure 1. Each element in
the analysis procedure is described in subsequent sections.

Critical Fastener Stress Crack Growth


Location & Load Spectrum Rate Data

Initial Crack
Geometry

Stress Intensity
Factors

Crack Crack
Advance Growth

Yes Residual
σRes > σLimit?
Strength

No

Inspection
Threshold & Interval

Figure 1. Flow chart of damage tolerance analysis procedure

It is noted that the stiffeners are not considered in the damage tolerance analysis. By ignoring the
stiffener effects on stress intensity factors of a crack growing toward a stiffener, the crack tends to grow
faster which renders a conservative crack growth life.

3.1 Critical Locations

Damage tolerance analysis of a repaired or modified skin begins by postulating initial flaws at the critical
locations. The criteria used to determine the critical locations are

(a) Center and corner fasteners in a fastener row,


(b) Highest fastener load transfer, and
(c) Stress concentration at an open hole

7
Critical locations determined based on the criteria are described as follows.

(A) Rectangular repairs

In rectangular repairs, central fasteners in the outermost row normally take a constant load transfer. The
highest load transfer occurs at the corner fastener. These critical locations are illustrated in Figure 2.

(i) One doubler (ii) Two stack-up doublers (iii) One internal and
one external doublers

Figure 2. Critical fasteners in rectangular repairs

(B) Circular Repairs

In circular repairs, the highest load transfer occurs at the outermost fastener. Figure 3 shows the critical
fasteners.

Figure 3. Critical fasteners in circular repairs

(C) Splice Joint Repairs

In splice joint repairs, the critical fastener location normally occurs in the outermost fastener row.
For butt joint repairs with the skin cutout and doubler flushed on one skin, the center and corner
fasteners in the outermost fastener row parallel to the splice doubler are considered critical. The
fastener next to the free edge of the repaired skin in the outermost fastener row normal to the splice
doubler is also critical.

For butt joint repairs with a skin cutout on one skin and a doubler extending over the other, the center
fastener in ht e outermost fastener row parallel to the splice doubler in the damaged skin takes the

8
highest load transfer. The corner fastener on the undamaged skin normally yields higher load transfer as
compared with that on the damaged skin. In addition, the fastener next to the free edge of the damaged
skin in the outermost fastener row normal to the splice doubler is also critical.

When skin cutouts and the doubler are on both skins, critical fasteners are located at the center and
corner in the fastener row parallel to the splice doubler in the damaged skin with a smaller skin cutout.
The fastener next to the free edge of the larger damaged skin in the outermost fastener row normal to
the splice doubler is also critical.
For lap splice joint repairs, critical locations are located at the center and corner fasteners in the
outermost row parallel to the splice doubler on the upper damaged skin. The fastener next to the free
edge of the same skin in the outermost fastener row normal to the splice doubler is also critical.

Examples of critical locations in splice joint repairs are shown in Figure 4.

(1) Butt joints

(i) Cutout & doubler (ii) Cutout on one skin & (iii) Cutout & doubler
flushed on one skin doubler on 2 skins on 2 skins

(2) Lap joints

(i) Cutout on upper (ii) Cutout on upper (iii) Cutout on upper &
skin with a flushed skin with a Rect lower skins with 2
doubler nd a T-doublers Rect and a T-doublers

Figure 4. Examples of critical locations in splice joint repairs

(D) Antenna Installations

9
In antenna installations, an irregular fastener placement is allowed in the design. The critical fastener
location is determined at the fastener that takes the highest load transfer.

Due to the cutout in both skin and doubler in an antenna installation, high skin stress takes places at the
edge of the skin cutout. The skin cutout hole is thus also considered as the critical location.

Figure 5 shows possible critical locations in antenna installations.

(i) Circular (ii) Elliptical (iii) Rectangular

(iv) Sausage-shaped (v) Teardrop-shaped

Figure 5. Examples of critical locations in antenna installations

3.2 Initial Crack and Subsequent Growth Assumptions

With critical locations identified in a repair or an antenna installation, initial flaw configurations and
subsequent crack growth scenarios are assumed as follows.

(A) Rectangular repairs

Scenario 1: Center fastener hole in the outermost fastener row

10
Initial Cracks: Two diametric through cracks of lengths 0.05” and 0.005”, respectively,
emanating from the center fastener hole along with a 0.005” crack at one side
of other holes

0.005” 0.005” 0.05” 0.005” 0.005”

Subsequent Growth: All cracks grow concurrently but independently and interaction between
cracks is ignored when the primary crack grows toward the adjacent hole. The
amount of growth δa1 for the 0.005” crack is added to its original length when
the 0.05” crack grows into the adjacent hole. The tip-to-tip primary crack
continues to grow, with crack tip interaction, toward adjacent holes. All
0.005”+ δa 1 0.005”+ δa 1

0.005”+ δa 1 + δa 2 0.005”+ δa 1 + δa 2

secondary cracks grow concurrently but independently. The same process


continues in successive growth until the residual strength is no longer greater
than the limit stress or the crack growth life reaches 300,000 flights.

Scenario 2: Corner fastener hole in the outermost fastener row

Initial Crack: Two diametric through cracks of lengths 0.05”, pointing toward the adjacent hole,
and 0.005”, respectively, emanating from the corner fastener hole along with a
0.005” crack at one side of other holes

0.005” 0.005” 0.05” 0.005”

Subsequent Growth: All cracks grow concurrently but independently and no interaction
between cracks is considered when the primary crack grows toward the
adjacent hole. The amount of growth δa for the 0.005” crack is added to its
original length when the 0.05” crack grows into the adjacent hole. The tip-to-
tip primary crack continues to grow with crack tip interaction. All secondary
cracks grow concurrently but independently. The same process continues in
successive growth until the residual strength is no longer greater than the limit
stress or the crack growth life reaches 300,000 flights.
0.005”+ δa 1L 0.005” + δa 1R

11
0.005”+ δa1L + δa2L 0.005”+ δa 1R + δa 2R
It is noted that successive growths of the 0.005” at corner fastener hole δa1R,
δa2R, etc. are different from the growths at other holes δa1L, δa2L, etc. because
the former does not grow toward an adjacent hole but others do.

(B) Circular repairs


For circular repairs, it is postulated that two initial diametric of 0.05” and 0.005” at the critical fastener
hole exist in the outermost fastener ring. The cracks grow concurrently with crack interaction.

Scenario 3: Critical single fastener hole

Initial Cracks: Two diametric through cracks of lengths 0.05” and 0.005”, respectively,
emanating from the critical fastener hole

0.005” 0.05”

Subsequent Growth: Cracks grow concurrently with crack interaction and continue to grow until
the residual strength is no longer greater than the limit stress or the crack growth
life reaches 300,000 flights.

Critical length

(C) Splice joint repairs

In splice joint repairs, fasteners are assumed to place in a straight pattern parallel or normal to the splice
doubler direction. Crack growth analysis is performed on the critical fastener hole in the outermost row
similar to that in rectangular repairs. Since the fastener next to the free skin edge normally takes the
highest load in the row, the following additional scenario is considered.

Scenario 4: Fastener hole next to the free edge

12
Initial Crack: Two diametric through cracks of lengths 0.05”, pointing toward the free edge, and
0.005”, respectively, emanating from the edge fastener hole along with a
0.005” crack at other holes

0.005” 0.005” 0.005” 0.005” 0.05”


Free edge

Subsequent Growth: All cracks grow concurrently but independently and interaction between
cracks is ignored when the primary crack growing toward the free edge. The
amount of growth δa for the 0.005” crack is added to its original length when
the 0.05" crack breaks the ligament to become an edge crack.
The primary edge crack continues to grow toward the adjacent hole. All
secondary cracks grow concurrently but independently. The growths continue
until the residual strength is no longer greater than the limit stress or the crack
growth life reaches 300,000 flights.

0.005” + δa 0.005” + δa 0.005” +δa 0.005” +δa

Free edge

(D) Antenna installations

13
In antenna installations, fasteners can be placed in an irregular pattern. Crack growth analysis is
performed on the critical fastener hole. Because of irregular placement of fasteners, the number of
fasteners on both sides of the critical fastener hole needs to be determined. An assumption is made that
the crack will run into the next fastener hole if it is located within a strip of three times the fastener hole
diameter. The procedure to determine the number of fasteners on each side of the critical fastener is
illustrated in Figure 6.

(1) Draw a line (solid line), normal to the stress direction, passing through the critical hole

Critical

(2) Draw 2 lines (dotted lines) a distance 1.5 times the fastener hole diameter, one above and the
other below the line drew in Step 1

1.5 fastener
diameter

Critical

(3) Ignore fasteners located outside the dotted lines band, using the center of the fastener hole, and
reposition the fasteners on the line passing through the critical hole

14
Critical

(4) Calculate averaged fastener pitch and fastener diameter if necessary, and rearrange the fasteners

Critical

Figure 6. Determination of number of fasteners


on each side of the critical fastener

Once the number of fasteners on each side of the critical fastener has been determined, the scenario is
described as follows.

Scenario 5: Critical fastener in a fastener row

Initial Cracks: Two diametric through cracks of lengths 0.05” and 0.005”, respectively,
emanating from the critical fastener hole along with a 0.005” crack at other
holes

0.005” 0.005” 0.05” 0.005” 0.005”

Subsequent Growth: All cracks grow concurrently but independently and interaction between
cracks is ignored when the primary crack grows toward the adjacent hole. The
amount of growth δa1 for the 0.005” crack is added to its original length when
the 0.05” crack grows into the adjacent hole. The tip-to-tip primary crack
continues to grow with crack tip interaction. All secondary cracks grow
concurrently but independently. The same process continues in successive
growth until the residual strength is no longer greater than the limit stress or the
crack growth life reaches 300,000 flights.

0.005”+ δa 1 0.005”+ δa 1

0.005”+ δa 1 + δa 2 0.005”+ δa1 + δa 2

15
When it is determined that there are no fasteners in a row on both sides of the critical fastener, scenario
3 is then used.

Scenario 6: Antenna connector cutout hole

Initial Crack: Two diametric through cracks of lengths 0.05” and 0.005”, respectively,
emanating from the antenna connector cutout hole along with a 0.005” crack at
other fastener holes

0.005” 0.005” 0.05” 0.005” 0.005”

Subsequent Damage: All cracks grow concurrently but independently and interaction between
cracks is ignored when the primary crack grows toward the adjacent hole. The
amount of growth δa for the 0.005” crack is added to its original length when
the 0.05” crack grows into the adjacent hole. The tip-to-tip primary crack
continues to grow, with crack tip interaction, toward adjacent holes. All
secondary cracks grow concurrently but independently. The growths continue
until the residual strength is no longer greater than the limit stress or the crack
growth life reaches 300,000 flights.

0.005”+δa 1,S 0.005”+ δa 1L 0.005”+ δa 1,S

0.005”+δa 1,S+δa 2,S 0.005”+ δa 1L + δa 2L 0.005”+ δa 1S + δa 2S

It is noted that successive growths of the 0.005” at skin cutout hole δa1L, δa2L, etc. are
different from the growths of 0.005” crack at fastener holes δa1S, δa2S, etc. because the
hole sizes and fastener pitches are different.

16
3.3 Fastener Load Transfers and Skin Stress Gradients

To calculate the stress intensity factor of a crack emanating from a fastener hole in a repaired or
modified skin, the fastener load needs to be known. For an open hole such as the antenna connector
hole in skin, the stress gradient along the assumed crack growth path needs to be described.

Fastener load transfers in skin of repairs and antenna installations are obtained using 2-D RAPID-FEM
analysis. A description of the analysis is provided in Appendix B. Stress gradients in skin for antenna
installations are also obtained from RAPID-FEM analysis.

3.4 Stress-Intensity Factors

The stress-intensity factor of a crack, denoted by K, characterizes the stress field near the crack tip. It
has been used in crack growth prediction models under cyclic load as well as in fracture toughness
measurements. The value of K depends on the crack configuration in the structural geometry and the
loads applied in the structure.

Stress-intensity factor solutions for the crack geometries in the crack sequences described in section 3.2
are required for damage tolerance analysis. Three methods that RAPIDC uses are described as
follows.

Method 1: Superposition method

This method is used for the case in which the crack is located at the edge of a pin-loaded hole in a plate
subjected to gross, bearing, and bypass stresses as illustrated in Figure 7.

σ Gross σ Gross + σ Bypass

σ Bearing
a
= 1/2 a + a
σ Bearing
σ Bearing

σ Bypass σ Gross + σ Bypass

Figure 7. Superposition of stress intensity factors

Referring to Figure 7, the stress intensity factor can be obtained as

K = K Far Field + K Point Loads

17
where the term 'Far Field' refers to σGross + σBypass, and ‘Point Loads’ represents σBearing.

Let BRF and BPF denote the bearing factor (σBearing /σGross ) and the bypass factor (σBypass/σGross ),
respectively. The stress intensity factor K can be written as

K = K Far Field + K Point Loads

= 1
2 (1+ BPF ) σ Gross π a β FarField + 1
2 BRF σ Gross πa β Po int Loads

where βFar Field is the geometry factor obtained by normalizing the stress intensity factors for the problem
shown in Figure 8 with respect to the Griffith crack. βPoint Loads is the geometry factor obtained by
normalizing the stress intensity factors for the problem shown in Figure 9 with respect to the Griffith
crack.

Thus, the above equation can be rewritten for the overall beta factor, β, as

K = β σ Gross πa
where

β = ½ [ (1 + BPF) βFar Field + BRF βPoint Loads ]

in which the geometry factors βFar Field and βPoint Loads are given by the curves as shown in Figures 8 and
9 respectively.

σ∞

R a

σ∞

18
3.5

Normalized Stress Intensity Factor: K/Ko


3

2.5

1.5

0.5

0
0 2 4 6 8 10
Normalized Crack Length: a/R

Figure 8. A crack emanating from a hole under


uniform far field stress: K o = σ ∞ πa

R a

19
1.6

Normalized Stress Intensity Factors: K/Ko


1.4

1.2

0.8

0.6

0.4

0.2

0
0 2 4 6 8 10
Normalized Crack Length: a/R

Figure 9. A crack emanating from a hole under


a pair of point loads: K o = F π a
2 Rt

Method 2: Green’s function method

If the stresses vary along the crack paths as shown in Figure 10(a), the stress intensity factor can be
obtained by summing up the stress intensity factors for each pair of splitting point loads along the crack
surfaces as shown in Figure 10(b).

X X
σ(x) = σο φ(x) p

R a 20 R a
(a) Stress Gradients (b) A Pair of Point Loads

Figure 10. Illustration of the Green’s function method

Consider a crack emanating from a hole and is subjected to a pair of point loads as shown in Figure
10(b). The stress intensify for the crack tip can be written as

p
∆K = G( λ ,ω )
πa

where the non-dimensional parameters λ and ω are defined as λ = a/R and ω = x/a, respectively.

Let the stress gradient be σ(x) = σ0 φ(x) as shown in Figure 10(a), the load per unit thickness at a
distance x from the edge of the hole can be expressed by

p = σ 0 φ ( x ) dx

Therefore, the stress intensity factor of the crack subject to the stress gradient shown in Figure 10(a)
can be calculated as

a σ 0 φ (x )
K=∫ G( λ, ω ) dx
o
πa

and can be expressed using the equation

a 1
K=
π ∫o σ o φ (aω ) G(λ,ω ) dω

The geometry factor β is can be obtained as

β = K / Ko

21
where K 0 =σ 0 πa .

Method 3: Compounding method

To account for the effect of a geometry modification ahead of a crack tip, the geometry corrector factor
is compounded with the stress intensity without the geometry modification to take into consideration of
the effect. The method is illustrated by the example below.

Consider a crack emanating from a hole and growing toward an adjacent hole as shown in Figure 11.

P
σ(x) = σ0 φ(x)

D a

Figure 11. A crack at a hole approaching an adjacent hole

The stress intensity factor can be estimated as

K = σ 0 π a β Compound

where βCompound is the compounded geometry factor computed by

βCompounded = βAt a Hole x βApproaching a Hole

For the case of a crack emanating from a hole in a wide plate subjected to stress gradients on crack
faces as shown in Figure 10(a), the geometry factor βAt a Hole can be obtained using the Green’s function
method described in Method 2. For the case of a crack growing toward a hole (Figure 12), the
geometry factors βApproaching a Hole are obtained from an open literature and are presented in Figure 13.

σ∞

22
c

2a
b R

σ∞

Figure 12. A crack approaching a hole

K o = σ ∞ πa

2.5
Normalized Stress Intensity Factor: K/Ko

2.25 R/c = 0.8 0.7 0.6


0.5
2

0.4
1.75

0.3
1.5

0.2
1.25
0.1

1
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Normalized Crack Length: a/b

Figure 13. Geometry factors for a crack approaching a hole

With βAt a Hole and βApproaching a Hole determined, the value of βCompounded can then be calculated.

Using the method, the stress intensity factor for a crack passing through multiple holes can be estimated.
The solutions are provided in Appendix C.

23
3.5 Residual Strength and Limit Stresses

The critical crack length is defined as the crack length in skin at limit stress. The methods used to
determine the residual strength evaluation and limit stress calculation are described in the following
sections.

3.5.1 Residual Strength: Fracture Toughness Criterion

The residual strength using the fracture toughness criterion is given by

Kc
σ Re sidual =
β π a

where K C is the toughness of the skin material, β is the geometry factor, and a denotes the crack length.
Fedderson revised the curve by drawing a tangent line from residual strength curve that intersects the
material yield stress, σy, as shown below in Figure 14.

Kc
Residual Strength

σy β πa

Crack Length

Figure 14. Residual Strength Curve

The residual strength diagram is therefore the straight tangent line followed by the curved portion of the
residual strength curve obtained from the fracture toughness method.

3.5.2 Limit Stresses

The limit stress in the circumferential direction, for the longitudinal crack, is calculated using the equation

pR R
σ Limit,Circumfere ntial = 1 .1 + 0 .6
t t

24
where p is the pressure differential, R and t are the radius and thickness of the fuselage at the location
where the damage tolerance analysis is performed. The stress due to the additional pressure 0.6 psi in
the equation is to account for the external aerodynamic suction pressure (0.5 psi) plus a 0.1 psi
regulator tolerance per FAR 23.574.

The limit stress in the longitudinal direction, for the circumferential crack, is determined using the
equation

pR R n1 σ 1G
σ Limit, Longitudinal = + 0 .6 +
2t 2t γ

where p is the pressure differential, n1 is the maneuvering limit load factor, γ is the payload reduction
factor assuming a value of 0.7, R and t are the radius and thickness of the fuselage, and σ1G is the one-G
stress, under normal operating condition, at the location where the damage tolerance analysis is
performed. The additional pressure 0.6 psi is added per FAR 23.574.

3.6 Load/Stress Spectrum and Equivalent Stress

To predict the crack growth of a repaired or a modified skin, the stress spectrum at the repair or
antenna location needs to be described. RAPIDC generates the stress spectrum from the load
spectrum that RAPIDC also generates. In addition, RAPIDC accepts user-provided stress spectrum in
the crack growth analysis.

The stress spectrum, either generated by RAPIDC or provided by the user, can be used in the cycle-
by-cycle crack growth analysis. A one cycle equivalent stress representing a flight cycle can be
calculated from the stress spectrum and used in the simplified crack growth analysis. The user can
provide RAPIDC with the equivalent stress as well. The analysis process is described in Figure 15.

25
User Input User
Aircraft Data Input

Load
Spectrum

Stress Stress
Spectrum Spectrum

Cycle-by-Cycle Equivalent
Crack Growth Analysis Stress

Simplified
Crack Growth Analysis

Figure 15. RAPIDC Analysis Process

Load and stress spectra generation methods are described in Appendix D. For the calculation of one
cycle equivalent stress used in simplified crack growth analysis, the method is described in appendix E.

3.7 Crack Growth Rate Data

The crack growth rate data obtained experimentally from constant amplitude coupon tests in the form of
da/dN versus ∆K for the material are generally documented in a tabular form. The tabulated data can
be used directly in the crack growth calculation using the cycle-by-cycle method. In RAPIDC, the
following da/dN versus ∆K tabular data of thirteen materials are in the material database.

(1) .................................................... 2024-T3 Clad Sheet, -T42 Bare Sheet, L-T RT LA DW


(2) .................................................... 2024-T3 Clad Sheet, -T42 Bare Sheet, T-L RT LA DW
(3) ........................................................ 2024-T351 Plate, -T3511 Extrusion, L-T RT LA DW
(4) ................................................................................7050-T7452 Forging, L-T T-L LA RT
(5) ........................................................... 7050-T74511, -T76511 Extrusion, L-T RT LA DW
(6) ...............................................................................7050-T74511 Extrusion, L-T RT STW
(7) .............................................................. 7050-T7651, -T7451 Plate, L-T T-L RT LA DW
(8) ...............................................................................7050-T76511 Extrusion, L-T RT STW
(9) ...........................................................................................7475-T7351 Plate, L-T LA RT
(10)................................................................................. 7475-T7651 Plate, L-T LA RT DW
(11)..........................................................................7475-T761 Clad Sheet, L-T RT LA DW

26
(12)..................................................................................... 7075-T6 Clad Sheet, L-T RT LA
(13)...................................................................... 2014-T6 Sheet, T=0.05 L-T RT LA 10 HZ

where

L-T:..........................................................................longitudinal-transverse material orientation


T-L:..........................................................................transverse-longitudinal material orientation
RT:..........................................................................................room temperature test condition
LA:...........................................................................laboratory air environmental test condition
DW: ........................................................................ distilled water environmental test condition
STW:...................................................................sump tank water environmental test condition

RAPIDC also provides user with direct input of da/dN tabular data or Walker’s coefficients. The input
format of the da/dN tabular data is described as follows.

Line 1 1
3 30 (2024-T3 CLAD SHEET,-T42 BARE SHEET, L-T RT LA DW)
0.05 0.40 0.80
Line 4 0.100E-31 0.100E-07 0.300E-07 0.500E-07 0.700E-07 0.800E-07 0.100E-06 0.200E-06
0.300E-06 0.500E-06 0.800E-06 0.100E-05 0.200E-05 0.500E-05 0.800E-05 0.100E-04
0.200E-04 0.400E-04 0.800E-04 0.100E-03 0.200E-03 0.500E-03 0.100E-02 0.300E-02
0.800E-02 0.200E-01 0.500E-01 0.100E+00 0.200E+00 0.100E+01
Line 8
2.8990 2.9000 2.9010 3.0000 3.2000 3.3000 3.6000 4.7500
5.5000 6.2000 6.6000 6.8000 8.0000 10.1000 12.2000 13.2500
16.9000 20.0000 24.2000 25.6000 29.6000 36.0000 41.0000 50.5000
61.0000 70.0000 78.0000 87.0000 92.0000 93.0000
Line 12
2.3480 2.3490 2.3500 2.3600 2.5000 2.6000 2.8800 3.9000
4.5000 5.1500 5.7000 6.0000 7.0000 8.7500 9.9000 10.5000
13.2500 16.0000 19.5000 21.0000 24.5000 29.0000 33.5000 41.5000
Line 16 48.0000 51.0000 54.0000 56.5000 57.0000 58.0000
1.7490 1.7500 1.7600 1.7700 1.8200 1.8750 2.0500 2.8500
3.4000 4.0000 4.5000 4.8000 5.6000 7.0000 8.0000 8.4000
9.9000 11.5000 13.0000 13.5000 15.3000 18.0000 19.8000 20.0000
20.0010 20.0020 20.0030 20.0040 20.0050 20.0060

Line Acronym Type Definition


1 NMAT Integer The material number that corresponds with the material
number containing the Walker’s C, p and q values
2 MRATIO Integer The first integer number is the number of different ratio
values in line 3; the maximum value is 10
2 MVALUE Integer The second integer value is the number of da/dN values in
lines 4-7, and ∆K values in lines 8-11, 12-15, and 16-19;
the maximum value is 40
2 - - Description of the material; optional
3 RATIOV Real The different ratio values in ascending order; there are

27
MRATIO values
4-7 DADN Real The da/dN values; there are MVALUE values for this y-
axis
8-11, DELTAK Real The ∆K values; there are MVALUE values, and MRATIO
12-15, sets for this x-axis
16-19

3.8 Crack Growth Analysis Methods

Two methods have been implemented in RAPIDC for the crack growth analysis, the simplified method
[1] and the cycle-by-cycle method. In the cycle-by-cycle method, the retardation effect due to
occasional stress overloads is accounted using the Generalized Willenborg Model [2,3,4]. In RAPIDC,
the crack growth analysis is performed up to 300,000 flight cycles.

(A) The Simplified Method

In the simplified crack growth analysis method, the number of flights for a crack growing from the size ai
to the size aj (aj > ai ) can be calculated by the equation

1 -p
Nij = C ( S G ij )

where C and p are the coefficients in the Walker’s equation, S is the equivalent stress, and Gij is the
geometry term. The quantity Gij represents the cumulative geometry effect and is calculated by the
equation

−p

∫ [ ]
aj
G −ij p = β( a) π a da
ai

where β(a) is the geometry factor. The integration is carried out numerically using the Gaussian
quadrature method [5]. In this method, the integral can be approximated by the following equation

aj −ai K
G −ij p =
2
∑w
k =1
k g (a k )

where
g (a k ) = [ β (a k ) πa k ]−p
and
(a j − ai ) t k + (a j + ai )
ak =
2

28
In the above equations, the variables wk and tk are the k-th weighting coefficient and root of the K-th
order Legendre polynomial PK (t) = 0. Values of w and t for the case of k = 4 are given below.

k Wk tk
1 0.347855 -0.861136
2 0.652145 -0.339981
3 0.652145 0.339981
4 0.347855 0.861136

It is noted that the simplified method calculates the number of flights Nij for a crack growing from a size
a to a next size aj (aj > ai). To generate crack growth, the procedure is carried out successively
i
between two consecutive crack lengths.

(B) The Cycle-by-Cycle Method

The cycle-by-cycle crack growth analysis method is illustrated by the example shown in Figure 16. The
analysis procedure is described in the flow chart shown in Figure 17.

σ
σmax,i
σmax,2
σmax,1
σmin,i

R a σmin,2
σmin,1
Time Sequence
σ
(a) Crack at hole in a wide plate (b) Stress Spectrum
Crack Growth Rate Data: da/dN

R1 R2 R3
Normalized SIF: K/Ko

Ko = σ πa

Normalized Crack Length: a/R Stress Intensity Factor: ∆K

(c) Normalized Stress-Intensity Factor (d) Fatigue Crack Growth Rate Data

Figure 16. Description of Crack Growth Analysis

29
i=1

Crack Length
a i-1 i = i+ 1

Stress Excursion
σmin,i → σmax,i

Stress-Intensity Factors
K min, i = σ min, i β πai−1
K max, i = σ max, i β π ai −1

Plastic Zone Size


2
1  Kmax,i 
rp ,i =  
2π  σ 
 ys 

No Yes
i=1
Set Set
γ i = ai −1 + r p, i a OL = a i-1
rOL = rp,i
Yes No γOL = a OL + rp,i
γ i ≤ γ OL
KOL = Kmax,i

Reduced SIF SIF Differential


ai −1 − aOL ∆Ki = Kmax,i - Kmin,i
K R, i = K OL 1 − − K max,i
rOL
Stress Ratio

Effective SIF
Ri = K min,i /K max,i
Eff
Kmax,i = Kmax,i - φ KR,i
Eff Incremental Crack Length
Kmin,i = Kmin,i - φ KR,i
∆a i = f (Ri, ∆Ki ) = (da / dN )i
Effective SIF Differential
Eff Eff Eff Updated Crack Length
∆Ki = Kmax,i - Kmin,i
ai = a i-1 + ∆ai
Effective Stress Ratio
No
/KEff
Eff
Eff K min,i a i ≥ a cr
Ri =
max,i
Yes
Incremental Crack Length
Stop
Eff Eff
∆a i = f ( Ri , ∆Ki ) = (da / dN )i

Figure 17. Flow Chart of the Cycle-by-Cycle Crack Growth Analysis

30
Let a0 be the initial crack length. The crack growth analysis starts with an initial crack of length a0 at i =
1. For the first cycle stress excursion, the stress-intensity factors and the plastic zone size are
calculated. With the stress ratio R and the change of the stress-intensity factor ∆K, the incremental
crack length is obtained from the da/dN data. The crack is incremented to the length a1.

At each sequential i- th cycle, the crack growth starts with a length equal to ai-1. The stress-intensity
factors and the plastic zone size are calculated using the i-th stress excursion. A test is then made to
determine if the growth of the progressing crack is retarded. A positive answer leads the analysis to use
the Generalized Willenborg retardation model in the calculation of incremental crack length. Otherwise,
the retardation effect is bypassed in the calculation. The crack then propagates the length ai.

The process continues until the crack grows to the critical crack length determined from the residual
strength of the structure at the limit stress.

In RAPIDC, the compression-tension stress cycle is treated as a zero-tension stress cycle and the
compression-compression stress cycle is ignored in the crack growth calculation. Furthermore, the
parameter φ introduced in the generalized Willenborg retardation model is calculated by the equation

∆ K Threshold
1−
∆K i
φ=
S −1
OL

where ∆K Threshold is the threshold stress-intensity factor level associated with zero fatigue crack growth
rates, and SOL is the overload (shut-off) ratio required to cause crack arrest for the given material. A
value of 2.3 is used in RAPIDC.

3.9 Inspection Threshold and Intervals

With the crack growth of the repaired or the modified skin provided by RAPIDC, the user determines
the inspection threshold and the inspection interval at the detectable crack length for the desired
inspection method.

4.0 Analysis Procedure Flow Charts

The RAPIDC analysis program consists of a static module and a damage tolerance analysis module.
The static module performs the static strength analysis of the repaired or the modified fuselage skin.
Margins of safety are calculated based on three independent criteria. RAPIDC calculates the load
carrying capacity lost, due to the damage treatment or a skin cutout, and the margin of safety based on
the joint allowable which includes the fastener shear and hole bearing allowables. It calculates the
margin of safety based on the ultimate tensile strengths of the doubler and the design ultimate tensile

31
strength of the skin. The margin of safety for shear is also calculated based on the shear allowables of
the skin and the doubler. In addition, the stiffness ratio between the doubler and the skin layers, and the
bending of the fastener going through the stack-up layers are also calculated. The program then checks
adequacy of the repair. RAPIDC either recommends redesign due to inadequate static design or deems
the repair to be statically adequate. The analysis procedure is described in the flow chart in Figure 18.

Static
Module

Carrying Load Lost Skin and Doubler Skin and Doubler


Due to Skin Cutout Tensile Load Shear Load
Stiffness
and
Fastener Bending
Fastener Shear Hole Bearing Skin and Doubler Skin and Doubler
Load Allowable Strength Allowable Tensile Allowables Shear

Margins
of Safety Acceptable

Redesign No All Margins Yes Yes


Damage
Advice of Safety > 0 Tolerance
No

Figure 18. Flow Chart of the Static Module

The damage tolerance module performs the crack growth and the residual strength analysis of the
repaired or the modified skin. The analysis is performed for cracks initiating at the critical fastener
holes. Initial crack configurations are assumed for repairs and antenna installations. Fastener load
transfers and skin stress gradients are obtained from RAPID-FEM analysis. Stress-intensity factors of
relevant crack configuration are then calculated. The residual strength is calculated and the crack
growth analysis is performed using either the simplified or the cycle-by-cycle method. The residual
strength is then checked against the limit stress to determine whether the crack advances. Based on the
damage tolerance analysis results, the user determines the inspection threshold and inspection intervals
for the selected inspection method. The analysis procedure is described in Figure 19.

32
Damage Tolerance
Module

Initial Crack Locations Stress Spectrum


and Lengths or
Equivalent Stress

Gross, Bearing, and


Bypass Factors

Stress Intensity
Factors

Crack Residual Crack


Advance Strength Growth

Yes
σRes > σLimit

No

Inspection
Threshold & Intervals

Figure 19. Flow Chart of the Damage Tolerance Module

5. 0 General Notes and Program Limitations

General notes and program limitations of the RAPIDC 2.0 program is provided in Appendix F.

6. 0 References

1. T. Swift, “Repairs to Damage Tolerant Aircraft,” in Structural Integrity of Aging Airplanes,


Editors: S. N. Atluri, S. G. Sampath, P. Tong, and Springer-Verlag, 1991.

2. J. D. Willenborg, R. M. Engle, and H. A. Wood, “A Crack Growth Retardation Model Using


an Effective Stress Concept,” AFFDL-TM-71-1-FBR, 1971.

3. J. P. Gallagher and T. F. Hughes, “Influence of the Yield Strength on Overload Affected


Fatigue-Crack-Growth Behavior of 4340 Steel,” AFFDL-TR-74-27, 1974.

4. R. M. Engle and J. L. Rudd, “Analysis of Crack Propagation Under Variable-Amplitude


Loading Using the Willenborg Retardation Model,” AIAA Paper No. 74-369, 1974.

33
5. Handbook of Mathematical Functions with Formulas, Graphs, and Mathematical Tables, Edited
by Milton Abramowitz and Lrene A. Stegun, Dover Publications, Inc., New York, 1972.

34