DESIGN AND ANALYSIS OF AN EQUIPMENT RACK STRUCTURE OF A

MEDIUM TRANSPORT AIRCRAFT

A THESIS SUBMITTED TO
THE GRADUATE SCHOOL OF NATURAL AND APPLIED SCIENCES
OF
MIDDLE EAST TECHNICAL UNIVERSITY

BY

MEHMET EFRUZ YALÇIN

IN PARTIAL FULFILLMENT OF THE REQUIREMENTS

FOR
THE DEGREE OF MASTER OF SCIENCE
IN
AEROSPACE ENGINEERING

SEPTEMBER 2009
 Approval of the thesis:

DESIGN AND ANALYSIS OF AN EQUIPMENT RACK STRUCTURE OF A

MEDIUM TRANSPORT AIRCRAFT

submitted by MEHMET EFRUZ YALÇIN in partial fulfillment of the

requirements for the degree of Master of Science in Aerospace Engineering
Department, Middle East Technical University by,

Prof. Dr. Canan Özgen _____________________

Dean, Graduate School of Natural and Applied Sciences

Prof. Dr. İsmail Hakkı Tuncer _____________________

Head of Department, Aerospace Engineering

Prof. Dr. Yavuz Yaman _____________________

Supervisor, Aerospace Engineering Dept., METU

Examining Committee Members:

Assist. Prof. Dr. Melin Şahin _____________________

Aerospace Engineering Dept., METU

Prof. Dr. Yavuz Yaman _____________________

Aerospace Engineering Dept., METU

Assist. Prof. Dr. Güçlü Seber _____________________

Aerospace Engineering Dept., METU

Assoc. Prof. Dr. Metin U. Salamcı _____________________

Mechanical Engineering Dept., Gazi University

Dr. Muvaffak Hasan _____________________

Chief of Structural Analysis, TAI

Date:
I hereby declare that all information in this document has been obtained and
presented in accordance with academic rules and ethical conduct. I also declare
that, as required by these rules and conduct, I have fully cited and referenced
all material and results that are not original to this work.

Name, Last Name :

Signature :

iii
 ABSTRACT

DESIGN AND ANALYSIS OF AN EQUIPMENT RACK STRUCTURE OF A

MEDIUM TRANSPORT AIRCRAFT

Yalçın, Mehmet Efruz

M.Sc., Department of Aerospace Engineering
Supervisor : Prof. Dr. Yavuz Yaman

September 2009, 130 pages

In this study, equipment rack structure for a medium transport aircraft was
designed and finite element analysis of this design was performed. The equipment
rack structure, which was designed for a modernization project, was positioned and
dimensions were determined by regarding the geometry of primary structures of the
aircraft. The structure was designed such that it satisfies the pre-defined margin of
safety values. Design of the structure was prepared in Unigraphics, and the finite
element modeling and analysis phases were carried out using MSC.Patran and
MSC.Nastran programs. For the fastener analysis, which is usually carried out by
hand calculations, two analysis tools were prepared by using FORTRAN and
Microsoft Office Excel programs. These tools were found to greatly facilitate the
analysis and save time. As these tools can be used in other finite element analyses, in
which MSC.Patran and MSC.Nastran programs are used, user manuals were
prepared.

Keywords: Aircraft Structural Design, Fasteners, Finite Element Analysis

iv
 ÖZ

ORTA BÜYÜKLÜKTE BİR NAKLİYE UÇAĞI İÇİN EKİPMAN RAFI

TASARIMI VE ANALİZİ

Yalçın, Mehmet Efruz

Yüksek Lisans, Havacılık ve Uzay Mühendisliği Bölümü
Tez Yöneticisi : Prof. Dr. Yavuz Yaman

Eylül 2009, 130 sayfa

Bu çalışmada, orta büyüklükte bir nakliye uçağı için ekipman rafı tasarımı ve
bu tasarım için sonlu eleman analizi yapılmıştır. Bir modernizasyon projesi için
oluşturulan ekipman rafı uçaktaki ana yapılara göre konumlandırılmış ve
boyutlandırılmış, önceden belirlenmiş olan güvenlik faktörlerini sağlayacak şekilde
oluşturulmuştur. Tasarım çalışmaları Unigraphics programında yapılmış, sonlu
elaman modelinin hazırlanmasında ve sonlu eleman analizlerinde MSC.Patran ve
MSC.Nastran programları kullanılmıştır. El hesapları ile yapılmakta olan bağlayıcı
analizleri için FORTRAN ve Microsoft Office Excel programları kullanılarak bu
analizleri daha kolay ve hızlı yapacak olan iki adet program oluşturulmuştur. Bu
programların MSC.Patran ve MSC.Nastran programları kullanılarak yapılan diğer
sonlu eleman analizlerinde de kullanılabilmesi için kullanım prosedürleri şekillerle
açıklanmıştır.

Anahtar Kelimeler: Hava Araçlarının Yapısal Tasarımı, Bağlantı Elemanları, Sonlu

Eleman Analizi

v
 to my family and friends who
never withhold their love, care and patience from me

vi
 ACKNOWLEDGMENTS

I would like to express my deepest gratitude to my supervisor

Prof. Dr. Yavuz Yaman for his guidance, assistance, encouragements,
insight and patience throughout this study.
I must thank my superiors in TAI, Dr. Muvaffak Hasan, İlknur
Yılmazer and Okay Karakaş for providing me the necessary data and
their guidance.
I would like to express my appreciation to my colleague Kadir
Akkuş, for his help in the Unigraphics.
I want to express my gratefulness to my best friends, Levent
Ünlüsoy, Erdoğan Tolga İnsuyu, Fatih Yaşar and Anıl Öztürk, for their
assistance, support and patience.
Finally, I would like to thank my family and friends for their
insight and encouragements.

vii
 TABLE OF CONTENTS

ABSTRACT ................................................................................................................ iv
ÖZ ................................................................................................................................ v
ACKNOWLEDGMENTS ......................................................................................... vii
TABLE OF CONTENTS .......................................................................................... viii
LIST OF TABLES ....................................................................................................... x
LIST OF FIGURES .................................................................................................... xi
LIST OF SYMBOLS ................................................................................................ xvi
LIST OF ABBREVIATIONS .................................................................................. xvii
CHAPTER
1. INTRODUCTION.................................................................................................... 1
1.1 Scope and Objectives .................................................................................... 1
1.2 Thesis Lay-Out .............................................................................................. 2
2. LITERATURE SURVEY ........................................................................................ 3
2.1 Finite Element Method .................................................................................. 3
2.1.1 Finite Element Method Basics ............................................................... 4
2.1.2 Finite Element Modeling ....................................................................... 6
3. DESIGN AND ANALYSIS OF AN EQUIPMENT RACK STRUCTURE OF A
MEDIUM TRANSPORT AIRCRAFT .................................................................. 10
3.1 Overview ..................................................................................................... 10
3.2 Preliminary Design Phase............................................................................ 14
3.3 Finite Element Modeling and Preliminary Analysis Phase ......................... 21
3.4 Detailed Design Phase ................................................................................. 35
3.5 Buckling Analysis Phase ............................................................................. 85
3.6 Fastener Selection Phase ............................................................................. 88
3.7 Dynamic Analysis ....................................................................................... 94
3.8 Fitting Design Changes ............................................................................. 102
4. CONCLUSIONS .................................................................................................. 105
viii
REFERENCES......................................................................................................... 107
APPENDICES
A. FASTENER CHECKS ........................................................................................ 109
A.1 The Tension Failure ...................................................................................... 109
A.2 The Shear Failure .......................................................................................... 110
A.3 The Combined Shear and Tension Failure .................................................... 110
A.4 The Bearing Stress Failure ............................................................................ 111
A.5 The Pull-through Stress Failure .................................................................... 111
B. USAGE OF DEVELOPED FASTENER TOOLS .............................................. 113
B.1 The FORTRAN Based Executable Tool ....................................................... 113
B.2 The Microsoft Office Excel Tool .................................................................. 126

ix
 LIST OF TABLES

TABLES
Table 3.4.1 Material Properties .................................................................................. 42
Table 3.4.2 Maximum Force Components at Understructure Connections at Each
Fitting (Forces are given in absolute values) ............................................................. 44
Table 3.4.3 Major Differences between the von Mises Stress Distributions Obtained
for Preliminary Design and Detailed Design ............................................................. 76
Table 3.4.4 Major Differences between the Deflections Obtained for Preliminary
Design and Detailed Design ....................................................................................... 84
Table 3.6.1 Margin of Safeties (MS values) for the Fasteners at Understructure
Connections ................................................................................................................ 92
Table 3.6.2 Used Fasteners and Fastener Properties .................................................. 93
Table 3.7.1 The First 10 Natural Frequencies for the Dynamic Analysis of the Rack
Structure ..................................................................................................................... 96
Table 3.8.1 Resultant Changes in the Fitting Designs ............................................. 103

x
 LIST OF FIGURES

FIGURES
Figure 3.1.1 Design and Analysis Process Flowchart ................................................ 12
Figure 3.1.2 The Outer Dimensions of the Equipment Rack to be Designed ............ 13
Figure 3.2.1 Horizontal and Vertical Profiles Formed Following the Aim of
Minimum Possible Number of Discontinuities .......................................................... 15
Figure 3.2.2 Separation of Two Modules of the Rack Structure ............................... 16
Figure 3.2.3 Equipment Placement Pattern ................................................................ 17
Figure 3.2.4 Forming Smaller Sections ..................................................................... 18
Figure 3.2.5 Rack Structure after Placement of Equipment Supports ....................... 19
Figure 3.2.6 Locations and Labels of the Front Fittings ............................................ 20
Figure 3.2.7 Locations and Labels of the Back Fittings ............................................ 21
Figure 3.3.1 Midplane Modeling Example ................................................................ 23
Figure 3.3.2 Finite Element Model of the Rack Structure ......................................... 24
Figure 3.3.3 Different Mesh Size Example................................................................ 25
Figure 3.3.4 Finite Element Model of Equipment Rack, Front View ........................ 26
Figure 3.3.5 Fastener Modeling Example .................................................................. 27
Figure 3.3.6 High von Mises Stress Regions for Rack Structure under Ultimate
Forward Acceleration (7g in -X) Case ....................................................................... 29
Figure 3.3.7 High von Mises Stress Regions for Rack Structure under Ultimate
Sideward Acceleration (7g in +Y) Case .................................................................... 30
Figure 3.3.8 High von Mises Stress Regions for Rack Structure under Ultimate
Downward Acceleration (7g in -Z) Case ................................................................... 31
Figure 3.3.9 Displacement of the Rack Structure under Ultimate Forward
Acceleration (4.67g in -X) Case ................................................................................ 32
Figure 3.3.10 Displacement of the Rack Structure under Ultimate Sideward
Acceleration (4.67g in +Y) Case................................................................................ 33

xi
Figure 3.3.11 Displacement of the Rack Structure under Ultimate Downward
Acceleration (4.67g in -Z) Case ................................................................................. 34
Figure 3.4.1 Location and an Example of Upper Fittings .......................................... 36
Figure 3.4.2 Formed Diagonal Elements ................................................................... 37
Figure 3.4.3 Finite Element Model after the Addition of Upper Fittings and Diagonal
Profiles ....................................................................................................................... 38
Figure 3.4.4 Finite Element Model after the Addition of Cover and Cabling Masses
.................................................................................................................................... 40
Figure 3.4.5 Highlighted Cover and Cabling Mass Elements.................................... 40
Figure 3.4.6 Fitting Labels ......................................................................................... 47
Figure 3.4.7 Location and Connection Pattern of Fitting B1 ..................................... 48
Figure 3.4.8 Finalized Geometry of Fitting B1 .......................................................... 49
Figure 3.4.9 Location and Connection Pattern of Fitting B2 ..................................... 49
Figure 3.4.10 Finalized Geometry of Fitting B2 ........................................................ 50
Figure 3.4.11 Location and Connection Pattern of Fitting B3 ................................... 51
Figure 3.4.12 Connection of Fitting B3 to Both Modules ......................................... 52
Figure 3.4.13 Finalized Geometry of Fitting B3 ........................................................ 53
Figure 3.4.14 Location and Connection Pattern of Fitting B4 ................................... 53
Figure 3.4.15 Finalized Geometry of Fitting B4 ........................................................ 54
Figure 3.4.16 Location and Connection Pattern of Fitting B5 ................................... 55
Figure 3.4.17 Finalized Geometry of Fitting B5 ........................................................ 56
Figure 3.4.18 von Mises Stress Distribution on the Back Fittings under Ultimate
Sideward Acceleration (7g in +Y) Case .................................................................... 57
Figure 3.4.19 von Mises Stress Distribution on the Fitting B4 under Ultimate
Sideward Acceleration (7g in +Y) Case .................................................................... 57
Figure 3.4.20 Location of Two Front Fittings at the Separation of Rack’s Modules 59
Figure 3.4.21 Examples of Finalized Geometry of the Front Fittings ....................... 59
Figure 3.4.22 von Mises Stress Distribution on the Front Fittings under Ultimate
Forward Acceleration (7g in -X) Case ....................................................................... 60
Figure 3.4.23 von Mises Stress Distribution on the Fitting F12 under Ultimate
Forward Acceleration (7g in -X) Case ....................................................................... 61
Figure 3.4.24 Location and Connection Pattern of Fitting U1................................... 62
Figure 3.4.25 Finalized Geometry of Fitting U1........................................................ 63
xii
Figure 3.4.26 Location and Connection Pattern of Fitting U2................................... 64
Figure 3.4.27 Finalized Geometry of Fitting U2........................................................ 65
Figure 3.4.28 Location and Connection Pattern of Fitting U3................................... 65
Figure 3.4.29 Finalized Geometry of Fitting U3........................................................ 66
Figure 3.4.30 Location and Connection Pattern of Fitting U4................................... 67
Figure 3.4.31 Finalized Geometry of Fitting U4........................................................ 67
Figure 3.4.32 Location and Connection Pattern of Fitting U5................................... 68
Figure 3.4.33 Finalized Geometry of Fitting U5........................................................ 69
Figure 3.4.34 von Mises Stress Distribution on the Upper Fittings under Ultimate
Sideward Acceleration (7g in +Y) Case .................................................................... 70
Figure 3.4.35 von Mises Stress Distribution on the Fitting U4 under Ultimate
Sideward Acceleration (7g in +Y) Case .................................................................... 70
Figure 3.4.36 Actual Geometry of the Maximum Stress Region on Fitting U4 ........ 71
Figure 3.4.37 Zoomed View of Highest von Mises Stress Region on the Fitting U4
under Ultimate Sideward Acceleration (7g in +Y) Case ........................................... 72
Figure 3.4.38 von Mises Stress Distribution on the Inner Parts of the Rack Structure
under Ultimate Forward Acceleration (7g in -X) Case .............................................. 73
Figure 3.4.39 von Mises Stress Distribution on the Inner Parts of the Rack Structure
under Ultimate Sideward Acceleration (7g in +Y) Case ........................................... 74
Figure 3.4.40 von Mises Stress Distribution on the Inner Parts of the Rack Structure
under Ultimate Downward Acceleration (7g in -Z) Case .......................................... 75
Figure 3.4.41 Front View of the Rack Structure Showing the Materials in Different
Colors (Blue: Al 2024 T3511, Yellow: Al 7075 T6 Clad, Green: Al 7075 T7351,
Red: 17-4PH stainless steel, Pink: 17-7PH stainless steel) ........................................ 77
Figure 3.4.42 Back View of the Rack Structure Showing the Materials in Different
Colors (Blue: Al 2024 T3511, Yellow: Al 7075 T6 Clad, Green: Al 7075 T7351,
Red: 17-4PH stainless steel, Pink: 17-7PH stainless steel) ........................................ 78
Figure 3.4.43 Thicknesses of the Parts of the Rack Structure, Front View ............... 79
Figure 3.4.44 Thicknesses of the Parts of the Rack Structure, Back View................ 80
Figure 3.4.45 Deflection Plot of Rack Structure Profiles under Limit Forward
Acceleration (4.67g in -X) Case ................................................................................ 81
Figure 3.4.46 Deflection Plot of Rack Structure Profiles under Limit Sideward
Acceleration (4.67g in +Y) Case................................................................................ 82
xiii
Figure 3.4.47 Deflection Plot of Rack Structure Profiles under Limit Downward
Acceleration (4.67g in -Z) Case ................................................................................. 83
Figure 3.5.1 Buckling Analysis Results after the Detailed Design Phase ................. 86
Figure 3.5.2 Buckling Critical Profiles after Detailed Design Phase......................... 87
Figure 3.5.3 Buckling Analysis Results Obtained after Increasing Buckling Critical
Profiles’ Thicknesses ................................................................................................. 88
Figure 3.6.1 Front Fitting Screw and Hi-Lok Locations............................................ 90
Figure 3.6.2 Back Fitting Screw and Hi-Lok Locations ............................................ 90
Figure 3.6.3 Upper Fitting Screw and Hi-Lok Locations .......................................... 91
Figure 3.7.1 Boundary Conditions Used for the Dynamic Analysis of the Rack
Structure ..................................................................................................................... 95
Figure 3.7.2 1st Mode Shape of the Rack Structure Excluding Masses ..................... 97
Figure 3.7.3 2nd Mode Shape of the Rack Structure Excluding Masses .................... 97
Figure 3.7.4 3rd Mode Shape of the Rack Structure Excluding Masses..................... 98
Figure 3.7.5 4th Mode Shape of the Rack Structure Excluding Masses ..................... 98
Figure 3.7.6 5th Mode Shape of the Rack Structure Excluding Masses ..................... 99
Figure 3.7.7 1st Mode Shape of the Rack Structure Including Masses .................... 100
Figure 3.7.8 2nd Mode Shape of the Rack Structure Including Masses ................... 100
Figure 3.7.9 3rd Mode Shape of the Rack Structure Including Masses .................... 101
Figure 3.7.10 4th Mode Shape of the Rack Structure Including Masses .................. 101
Figure 3.7.11 5th Mode Shape of the Rack Structure Including Masses .................. 102
Figure B.1.1 Addition of Multi-Point Constraint Forces and Grid Point Force
Balance Option into Analysis .................................................................................. 114
Figure B.1.2 Reporting the Multi-Point Constraint Forces ...................................... 115
Figure B.1.3 Input of MSC.Nastran® Input File Name in Execution Window........ 115
Figure B.1.4 Input of Multi Point Constraint Force Report Name in Execution
Window .................................................................................................................... 116
Figure B.1.5 Choice of Contents of Output File ...................................................... 117
Figure B.1.6 Number of Load Cases to be Analyzed .............................................. 118
Figure B.1.7 Fitting Factor Application for Whole System ..................................... 119
Figure B.1.8 Input of Fitting Factor Values for Each Load Case ............................ 120
Figure B.1.9 Output File Format Showing Only the Maximum Forces .................. 121

xiv
Figure B.1.10 Output File Format Prepared for the Input of Microsoft Office Excel
Tool .......................................................................................................................... 122
Figure B.1.11 FORTRAN Based Program and Output ........................................... 123
Figure B.1.12 Maximum Tension Forces on Fasteners 1324 and 1325 .................. 124
Figure B.1.13 Maximum Shear Forces on Fasteners 1324 and 1325 ...................... 125
Figure B.2.1 Data Inputted in Formatted Input Sheet .............................................. 126
Figure B.2.2 Fasteners Corresponding to Multi-Point Constraint Elements ........... 127
Figure B.2.3 Fastener Properties Inputted ............................................................... 128
Figure B.2.4 Material Properties Inputted ............................................................... 129
Figure B.2.5 Fastener Analysis Results ................................................................... 130

xv
 LIST OF SYMBOLS

Am Maximum Minor Area of Fastener

D Diameter
E Modulus of Elasticity
F Force
Fbru Ultimate Bearing Strength
Fsu Ultimate Shear Strength
Ftu Ultimate Tensile Strength
Fty Yield Tensile Strength
Fx Force in X Direction
Fy Force in Y Direction
Fz Force in Z Direction
g Gravitational Acceleration
Kp Bolt Tension Preload Coefficient
R Ratio of Applied Load to Allowable Load
t Thickness
T Torque
β Wrench Torque Ratio
ν Poisson’s Ratio
ρ Density
σ Stress

xvi
 LIST OF ABBREVIATIONS

all Allowable
app Applied
CAD Computer Aided Design
ff Fitting Factor
kg Kilogram
LC Load Case
m Meter
mm Millimeter
MPa MegaPascal
MPC Multi-Point Constraint
MS Margin of Safety
N Newton

xvii
 CHAPTER 1

INTRODUCTION

1.1 Scope and Objectives

Aerospace industry is a continuously developing branch. With the developing

technology and experience gained in the branch, newly designed aircrafts become
much more functional than the old ones. But, since design of a completely new
aircraft is a costly process for the manufacturers, as like replacing the aircrafts in use
with the new ones for the customers, modernization of the existing aircrafts with the
new technology is a widely applied process. The equipment rack structure, which is
to be designed in this thesis study, is a part of one of these modernization projects
carried on a medium transport aircraft. This rack will be designed to carry newly
developed equipments, which will increase the functionality of the aircraft.
Regarding the needs of the customers, it was aimed to add 29 new equipments to
mentioned medium transport aircraft and the space needed for these equipments is
the reason why they are to be placed in a rack structure. The scope of this study is
designing the equipment rack structure such that, it will resist to the ultimate
acceleration factors of 7g applied separately in each of the three translatory flight
axes. This aim should be achieved by a light-weight, easily producable and cost
effective structure.

1
1.2 Thesis Lay-Out

This thesis study is to be presented in four main chapters. First chapter is the
introduction chapter and gives the scope and objectives of the design study. In the
second chapter, literature study about the finite element method and modeling, which
will be used for the analysis of the designed structure, will be presented. Third
chapter is for the design and analysis phases. In this chapter preliminary design,
finite element modeling and preliminary analysis, detailed design, buckling analysis,
fastener selection and dynamic analysis phases are to be introduced. Finally, fourth
chapter is for the conclusions. In the appendices, fastener failure checks and the user
manuals of two tools, prepared for the fastener analysis, will be presented.

2
 CHAPTER 2

LITERATURE SURVEY

2.1 Finite Element Method

Finite Element Method, which is also referred as Finite Element Analysis, is a

computational technique which aims obtaining approximate solutions for the
dependent variables in the boundary value problems. These problems are the
mathematical problems in which one or more dependent variables must satisfy a
differential equation and boundary conditions in the domain of interest, which is
called “field”. The variables in the field are the dependent variables in the boundary
value problem and usually used in the representation of a physical structure. The
specified values that the dependent variables take on the boundaries of the field are
called “boundary conditions” [1].
Finite Element Method is the idealization of a continuous system by a
discrete system consisting of smaller elements in which a field quantity is allowed to
have only a simple spatial variation. This idealization enables the conversion of a
structural system with infinite degrees of freedom into one with finite degrees of
freedom to make the problem solvable. Since this process reduces the complexity of
spatial variations, Finite Element Method provides an approximate solution. This
approximate solution distinguishes from other numerical procedures with two
characteristics [2]:

• For generating a system of algebraic equations an integral formulation is

utilized
• For the approximation of the unknown quantity or quantities continuous
piecewise smooth functions are used
3
 In addition to these characteristics differences, Finite Element Method has the
advantages listed below over other numerical solutions [2]:

• Finite Element Method is applicable to any field problem.

• There is no geometric restriction.
• Boundary conditions and loading are not restricted.
• Material properties are not restricted. It may even change within an element.
• Components that have different behaviors and different mathematical
descriptions can be combined.
• A finite element structure closely resembles the actual body or region to be
analyzed.
• The approximation is easily be improved by grading the mesh so that more
elements appear where field gradients are high and more resolution is needed.

2.1.1 Finite Element Method Basics

Finite Element Method is a numerical approach for reduction of a problem

which has infinitely many degrees of freedom, to a problem which has finitely
degrees of freedom and can be solved by using a computer [3]. When the problem to
be solved is the analysis of a structure, for the reduction of this problem structure is
cut into several elements whose behavior can be represented in a simple way. This
process allows forming equilibrium equations at the connection points of the
elements which are usually to be solved by implementation of computer tools in the
stress analysis.
Finite element structure is the assembly formed by the finite elements.
Element connection points are called as nodes. The particular arrangement of
elements is called as mesh which is represented by a system of equations to be solved
for unknowns at nodes. Solution of nodal quantities, when combined with the
assumed field in any given element, completely determines the spatial variation of
the field in that element. Thus the field quantity over the entire structure is
approximated element by element, by using the fact that all displacement
4
components at a node are to be same for all elements that share same node, and in
piecewise function [2].
A major application of the Finite Element Method is the determination of
fastener or internal forces and stresses at any location of any component in a design
process, noting that the stress term stands for the von Mises stress [4] throughout this
thesis except the cases it is described specifically. As the aim of the design process is
to be obtaining safe design, if the fastener or internal forces, or stresses exceed
allowable values, the component can be redesigned and then reanalyzed until the safe
design is obtained.
The stress analysis procedure in the Finite Element Method can be
summarized in the following steps [5, 6]:

• Discretization: Dividing the object to be analyzed in to finite elements whose

size and shape vary for reflecting the boundaries of the elastic body to be
analyzed and expected form of solution
• Selection of the Interpolation Function: Selection of the element type or
interpolation function which is to be used in the approximation of the
displacements and strains
• Derivation of the Stiffness Matrices of Elements: Determination of the
element stiffness matrix of each element which relates the forces and
displacements in respecting element
• Formation of the Global Stiffness Matrix: Assembling the element stiffness
matrices such that formed stiffness matrix will relate the forces and
displacements in the whole elastic body
• Rearrangement of the Global Stiffness Matrix: Setting up simultaneous
equations such that they can be solved for the unknown values of forces and
displacements after the application of mechanical and geometrical boundary
conditions
• Derivation of the Unknown Values for Forces and Displacements: Solving
the equations formed for the derivation of unknown forces and displacements
whose solutions will give the reaction forces and deformations of the elastic
body

5
 • Computation of Strains and Stresses: Obtaining the strains and stresses on the
elastic body by using the strain-displacement and stress-strain relationships

2.1.2 Finite Element Modeling

In the Finite Element Modeling the essence is having a simplified

representation of a physical structure which serves a particular and perhaps limited
purpose [5]. The simplified representation is achieved by the simulation of the
physical structure as a combination of finite elements. This combination should be
prepared such that, it should reflect the physical properties and boundary conditions
of the actual structure adequately. In the preparation of the model badly shaped
elements and elements which are too large to represent important variations of the
field quantity should be avoided. But on the other hand, using many more number of
elements than the smallest number, needed for the adequate representation of the
field and its gradients, should be avoided also. The reason behind this is the wish of
not wasting the analyst time and computer resources [7].
There are many computer package programs providing analysis capabilities
by using the Finite Element Method. ANSYS® [8], ABAQUS® [9],
MSC.Patran®/Nastran® [10] are a few to mention. In this thesis
MSC.Patran®/Nastran® are used. The reasons stemmed from the fact that these
programs are widely used in aerospace industry throughout the world and its
capabilities are well suited for aerospace applications.
There is a lack of publication about the procedures of finite element
modeling. This is mostly because of the fact that used techniques differ for each
person as; although basic information can be learned from the textbooks, rest is
acquired with the experience. In this section some common modeling techniques will
be presented:

Fastener Modeling:

The distribution of load between the components of a structural assembly is

affected by the stiffness of fasteners between them, in addition to the dimensions and

6
material properties of the components. Thus, representation of the fasteners affects
the accuracy of the results of finite element analysis.
Although it is possible to model each fastener separately with the manually
calculated stiffness values, the commonly used procedure is simulation of the
fasteners in three common groups:

• Single-Point Constraint: A Single-Point Constraint is a constraint that may be

applied at a grid point to restrain its displacement [10]. This type of constraint
is used for modeling the fasteners connecting modeled structure to ground, or
to understructure and simulates the boundary conditions. The Single-Point
Constraint is a rigid connection and assumes no elasticity for the
understructure. The rigidity created by the Single-Point Constraint causes
stress values obtained at the connection region to be higher than the real case,
as no load is to be absorbed by the elasticity of the understructure. Since in
the real case understructure deforms with the structure under consideration
some of the load would be absorbed during this deformation. But, as having
higher stress values in the finite element analysis results makes these results
more conservative, using Single-Point Constraints at the understructure
connections is a commonly applied procedure as carries no risk of unexpected
failure. The general usage of the Single-Point Constraints is the application of
zero displacement into translation directions.
• Multi-Point Constraint Element of Type 2 (RBE2: Rigid Body Element, Type
2): RBE2 is a tool which rigidly connects the selected components of several
grid points [10]. This type of element is commonly used for the simulation of
the fasteners connecting the two profiles of the structure under consideration.
As this element has infinite rigidity the selected components, which can be
transition and rotation components, of the connected nodes will be same. This
means that the connected nodes will translate and rotate in same amount,
which may not be the case in real life. For simulating the real case better, this
type of elements is used while connecting only the translations. The idea
behind this application is that the rotations are to be restrained by the force
couples in the fasteners instead of moment on a single fastener. As fasteners
have stiffness properties, which are taken as infinite in RBE2 connections,
7
 amount of load absorption by the fastener is neglected. As like the Single-
Point Constraints, RBE2 type connection element also yield higher stress
values in the connection regions and since the results are on conservative side
when compared to real case. So, using RBE2 type of elements for modeling
the fasteners between two profiles is a commonly applied procedure, as it
carries no risk of unexpected failure and avoids the requirement of manual
calculation of fastener stiffness values.
• Multi-Point Constraint Element of Type 3 (RBE3: Rigid Body Element, Type
3): RBE3 is a tool which is used for the distribution of applied loads and mass
in a finite element model [10]. RBE3 type of element distributes the applied
loads and mass to connected points based on its geometry and local weight
factors. This element type is used in the connections of concentrated masses,
usually equipment masses modeled in their center of gravity, to connection
points of the non-modeled parts, usually equipments. The RBE3 type of
elements does not add additional stiffness to the structure under
consideration. So, the deformation results obtained at the load or mass
application points will be higher than the ones in real case. This type of
element is one of the best elements for the load and mass distribution and
because of easy and time saving application process they are the most
commonly used ones.

Midplane Modeling:

When a solid structure is to be modeled with two dimensional shell elements

one dimension of the actual structure is lost in the finite element model. For the
better reflection of the actual structure this dimension is to be thickness of the three
dimensional structure, which is to be inputted in the properties of the elements for the
finite element analysis. MSC.Patran® and MSC.Nastran® identify this thickness such
that the two dimensional finite element assumed to be at the middle of the actual
three dimensional element. Half of the thickness is assumed to be added to both faces
of the two dimensional element and so, this element forms the symmetry plane of the
three dimensional element. Thus, three dimensional profiles are modeled at the

8
middle of the thickness directions, midplane, as a common application for the better
simulation of the actual structure.

Rules to be Followed for Satisfactory Mesh:

Number and locations of the nodes and element types of the finite element
model to be formed should be selected such that associated analyses would be
sufficiently accurate. Although several methods include automatic-mesh generation
with capability of producing and iteratively refining the mesh according to user-
selected error tolerance, obtaining satisfactory mesh may require manual
optimization of the mesh. There are several practical rules which can be followed for
obtaining satisfactory mesh [11]:

• Nodes should be located at the concentrated load application points.

• Location of the nodes should be selected such that, there will be node at the
boundary condition points.
• Nodes should be located in the concentrated spring and mass locations.
• There should be nodes placed along lines and surface patches, over which
pressures, compliant foundations, distributed heat fluxes and surface
convection are applied.
• Nodes should be located at the applied traction or heat flux discontinuities.
• Nodes should be located along the lines of symmetry.
• There should be nodes located along the interfaces between different
materials or components.
• Ratio of largest to smallest element dimensions should be smaller than five.
• For the symmetric configurations, symmetric meshes should be formed.
• For the domains with higher gradients, density of elements should be greater.
• Interior angles in the elements should be between 45° and 135°.
• Mesh size variations should be gradual rather than abrupt.
• In the sub-domains, which have low gradients, meshes should be uniform.

9
 CHAPTER 3

DESIGN AND ANALYSIS OF AN EQUIPMENT RACK

STRUCTURE OF A MEDIUM TRANSPORT AIRCRAFT

3.1 Overview

The aim of this thesis is to design a light-weight, easily producible and cost
effective structure to carry 29 equipments having a total mass of 343 kg, for a
medium transport aircraft. The designed structure should resist to the ultimate
acceleration factors of 7g separately in each of the three translatory flight axes with a
minimum margin of safety of 0.2. The margin of safety is a factor used in aerospace
industry as an index for reliability. It is calculated by division of allowable load by
applied load and by subtracting one (Eqn 3.1.1) [12]. The requirement of minimum
0.2 margin of safety comes from the Federal Aviation Regulations since the designed
structure will not be exposed to any test procedure. This factor is used for the
uncertainties which could arise when the strength of parts of the design are not
proved with any load test. In aerospace industry, a value called as fitting factor is
generally used to overcome the defects that could arise during the
removal/installation procedures of the parts, as the designed structure should be
partially or completely removable. Hence a fitting factor of 1.15 will also be used in
the calculation of margin of safety. The calculation of margin of safety with the
fitting factor is given in Eqn 3.1.2 [12].
Margin of Safety Calculation:

Allowable Load
MS = −1 (Eqn 3.1.1)
Applied Load

10
Margin of Safety Calculation with Fitting Factor:

Allowable Load
MS = −1 (Eqn 3.1.2)
Applied Load × Fitting Factor

In order to design the required system, the following steps will be undertaken:

1. A preliminary design will be conducted by using the profiles with

generally used section properties and standard materials. This
preliminary design phase will be carried out by using the Computer
Aided Design (CAD) program of Unigraphics.
2. Analysis of the preliminary design will be carried out by performing
finite element modeling and using MSC.Patran® and MSC.Nastran®
package programs. Also the buckling analysis will be performed by
using analytical methods. The idea behind these analyses is to find out
the parts that are insufficient for the loads they are designed to carry. In
addition to this, parts with margin of safety values much higher than the
required values can be found. By examining the results of finite element
analysis it will be possible to find more effective results for insufficient
and oversafe parts.
3. Redesign the parts which malfunction due to insufficient features. And
also it is possible to redesign the oversafe parts to make them
lighter/cheaper/easy to be manufactured if there is appreciable gain.
4. Reanalyze the whole system with the newly incorporated parts. This
step may require performing redesign process again.
5. Selection of the fasteners by using the analytical methods to perform
necessary checks (i.e. tension, shear, combined tension and shear,
bearing, pull through failure checks). This step may require changes in
the designed parts.

Design and analysis steps are shown in a flowchart in Figure 3.1.1.

11
 Preliminary
Design

Unsafe Analysis Safe

Design Design

Selection of
Redesign Fasteners
Impossible

Possible

Final
Configuration

Figure 3.1.1 Design and Analysis Process Flowchart

During the design of a secondary structure for a modernization project of an

aircraft, the existing (both primary and secondary) structures of the aircraft play a
very important role. The outer geometry and the understructure connection points of
the newly designed part are to be determined according to these existing structures.
Since the inner geometry of the new design should reflect a coherent setup with the
outer geometry of same structure; it can be concluded that the whole structure
depends on the existing structures. This is one of the aspects taken into the
consideration during the preliminary design of the structure, in this thesis. Figure
3.1.2 shows the outer dimensions that the designed structure will fit in.

12
 Figure 3.1.2 The Outer Dimensions of the Equipment Rack to be Designed

As it can be seen in Figure 3.1.2, the rack structure to be designed have outer
dimensions of 4.07 m in length (corresponding to aircraft’s longitudinal direction)
1.32 m in height (corresponding to aircraft’s vertical direction) and a variable width
of 0.52 m to 0.69 m (corresponding to aircraft’s horizontal direction). This structure
corresponds to 9 frames of the aircraft which will be used for the connection of
designed structure to the aircraft. The curvature at the back side of the rack structure
is due to the curvature of these frames. The variable width of the rack structure will
be 0.58 m at the bottom and increase up to 0.69 m when the aircraft fuselage has the
maximum width and again decrease to 0.52 m at the upper side. While determining
these dimensions, it is important that there should be no undesired contact between
the structure and the understructure. For avoiding any contact between the new
structure and understructure outer dimensions mentioned here was determined such
that there will be at least 10 mm gap to the understructure in every location and every
direction. These dimensions will form the base of the preliminary design.

13
3.2 Preliminary Design Phase

The preliminary design of the equipment rack will be formed in a CAD

(Computer Aided Design) program of Unigraphics. The aerospace companies’
design policies have long been proved that in order to have an effective design,
which is in our case to maximize the ratio of the weight to be carried to the structural
weight, it is essential to form the load paths as smooth as possible. The structure
should be smooth and continuous from the load application points, which are the
equipment connection points for the rack structure, to the connections to
understructure. The idea behind this application is to avoid the possible stress
concentrations which could arise at the discontinuities. By this way the necessity of
using profiles with higher thicknesses and larger sizes and consequently with higher
load carrying capability can be avoided. The load paths can be assumed as if they
were the currents in a circuit and hence it can be easily imagined that the
combination of two or more load paths leads to higher loads at the intersections and
following profiles. The load paths, in the structure to be designed, are to be used for
transmitting the current to the understructure. Hence, like the current in a circuit,
having higher number of load paths - leading to the understructure connection points
- results in lower loads in these paths. The connection points mentioned here is the 9
frames that the rack structure corresponds. For transmitting the load to the
understructure, 9 vertical profiles in the front side and 5 vertical curved profiles in
the rear side were formed as can be seen in Figure 3.2.1. These profiles must be
connected to frames with connection elements called as fittings, whose locations are
labeled in the aforementioned figure. Although these vertical profiles were aimed to
be continuous throughout the structure, two of the front profiles had to be divided
because of the geometric features of the equipments to be placed. In addition to
vertical fittings, rack structure to be designed was divided into three shelves by
horizontal profiles which will transmit the load from equipment locations to vertical
profiles. As it can be seen in Figure 3.2.1, horizontal profiles at the back of the
structure are continuous between two ends. In the middle section, horizontal profiles
in the front of the structure were divided for the placement of biggest equipment to
be installed.

14
 Figure 3.2.1 Horizontal and Vertical Profiles Formed Following the Aim of
Minimum Possible Number of Discontinuities

The future maintenance consideration is another important aspect to be taken

into account during the design of a structure. The rack structure is to be designed in
such a way that it can be removed completely or partially from the aircraft. These
possible removal operations could be required for the maintenance of any part of the
aircraft (or the rack structure itself), any defect in the parts of the rack or any change
in the configuration or structure. In order to satisfy the proper maintainability, the
rack structure to be designed will be divided into two parts from the middle section
(from the fifth front vertical profile and from the third rear curved profile) to allow
partial removal. In this section, two similar profiles facing opposite directions on
both front and rear side of the rack structure are to be formed. These similar profiles
will form the outer geometry of two separate modules of the rack structure and these
modules are to be available for the separately removal/installation. For the
connection of these modules small connector plates, which will be easily removable,
will be formed. The profiles forming the outer structure of two modules and the
connector plates used for the connection of these modules can be seen in Figure
3.2.2.

15
 Figure 3.2.2 Separation of Two Modules of the Rack Structure

The placement pattern of equipments inside the rack structure also play an
important role in the design. This placement has to be determined by taking the
magnetic, geometrical, electrical (cabling) and heat based (cooling system) aspects
into account. In addition to these; forming smooth load paths and obtaining the
minimum interface loads transferred to understructure were aimed in the preliminary
design phase. The former was obtained by dividing of the structure into continuous
rows as mentioned and the latter was achieved by the placement of the heavier
16
equipments at the bottom of the rack as so formation of the small moment arms with
respect to bottom fittings. The idea behind this application is that the dominating
factors on interface loads are the moment arm between the center of gravity and the
understructure connections, mass properties and applied accelerations. The moment
arm was aimed to be kept at the smallest possible value during the equipment
placement process. As a result; the equipment placement pattern, formed to have
continuous rows and minimum moments with respect to bottom fittings, is shown in
Figure 3.2.3. Equipments on the formed three shelves are shown without details and
with outer dimensions.

Figure 3.2.3 Equipment Placement Pattern

After the determination of equipment placement pattern, it is important to

divide the structure to smaller sections for the placement of equipment supports. For
the achievement of smaller sections, horizontal profiles in the direction of aircraft’s
horizontal direction, which is the front-back direction for rack structure, were
formed. These profiles were placed between the connection points of front vertical
and horizontal profiles and back profiles of the rack, where the geometrical features
17
of the equipments to be placed generate no restraint. These profiles, which can be
seen in blue in Figure 3.2.4, would distribute the load coming from equipment
supports to front and back profiles.

Figure 3.2.4 Forming Smaller Sections

In the procedure of forming equipment supports, the geometrical features and

the connection pattern of equipments play an important role. Since equipments are
standard parts, dimensions and fastener connection patterns cannot be changed. In
addition to these, another aspect in the formation of equipment supports is the
cooling requirements of some of the equipments. These requirements could be
minimum distance to any other structure or openings around the equipments (i.e. it
could be required that support should not cover the bottom of the equipment
completely). The rack structure after the placement of equipment supports, which
were formed according the mentioned requirements and with the aim of load
distribution to the surrounding structures, can be seen in Figure 3.2.5.

18
 Figure 3.2.5 Rack Structure after Placement of Equipment Supports

As mentioned before, it was aimed to have minimum possible interface loads

transferred to understructure of the aircraft and equipments were placed to achieve
this. But, the rack structure is thought to carry an approximately 343 kg of
equipment. In addition to this, there will be extra mass of 21 kg from the connectors,
30 kg from the cables and 46 kg from the covers. When the designed structure mass
is added to others, the total mass would be above 500 kg. All those are going to be
subjected to a 7g ultimate acceleration on each of the translatory axes during the
flight. To distribute the high forces and moments, caused by moment arms between
the load application points - which are the equipments’ and structure’s center of
gravities - and understructure connection points 21 fittings were used in the bottom
of rack structure. These fittings were formed with basic geometrical features and
understructure connection pattern as these will be used in preliminary analysis. They
were designed to have a total of 73 fasteners connecting rack structure to the
understructure by regarding the geometrical restraints and requirement of load
distribution. There are 16 fittings formed for the connection of front vertical profiles
to the understructure (one at each end profile and two at each middle profile). These
fittings will be labeled as front fittings from now on and will be numbered from F1 to

19
F16 in the aircraft’s longitudinal direction. Location of the front fittings can be seen
in Figure 3.2.6.

Figure 3.2.6 Locations and Labels of the Front Fittings

In addition to front fittings, rack structure has 5 fittings at the back side
connecting the curved vertical profiles to the frames under the structure. These
fittings will be labeled as back fittings and they will be numbered from B1 to B5
from aircraft’s nose to tail. The back fittings have 2 fasteners connected to
understructure except the Fitting B1, as this fitting is fastened to understructure with
only one fastener because of the interference of the understructure. Location of these
fittings can be seen in Figure 3.2.7.

20
 Figure 3.2.7 Locations and Labels of the Back Fittings

Placement of the fittings completes the load path from equipments to the
understructure, which means the achievement of minimum requirement for the finite
element analysis of the designed structure. Since the general dimensions and layout
(i.e. shelf and module structure) and equipment placement pattern of the structure to
be designed is completed, now it is possible to detail the design procedure with the
help of finite element analysis.

3.3 Finite Element Modeling and Preliminary Analysis Phase

For the finite element analysis, which is to be done by MSC.Nastran®, a finite

element model will be prepared by using MSC.Patran®. One of the most important
aspects in the finite element modeling is the preparation of a model which has
minimum degree of freedoms while reflecting the actual characteristics of the real
model as much as possible. The idea behind the small number of degree of freedoms
is that the smaller sized models require less storage and processor requirements in
addition to the shorter computational times. Most common application for obtaining
a model with smaller number of the degree of freedoms is modeling the structure as a
21
combination of one (rod, beam) and two (shell elements: quad, tria) dimensional
elements. But, since fasteners of the structure to be designed will be analyzed, finite
element model of this structure will be formed with two dimensional elements. The
idea behind this application is that the one dimensional elements do not yield
accurate results for each fastener as the correct location cannot be simulated.
When a three dimensional structure is reduced to two dimensional elements,
it is obvious that one dimension of this structure is omitted. The missing dimensions
in these elements are modeled by the input of sectional properties as the element
properties. The important point here is that when the sectional properties are inputted
to the two dimensional element, section is formed such that the finite element
corresponds to the middle section of the assumed element. The sectional properties to
be inputted to the two dimensional element includes the thickness value which is the
omitted dimension of three dimensional structure. The element is assumed to have
half of the thickness at each side corresponding to omitted dimension’s direction.
Thus, the two dimensional element stands in the midplane of the section, which
requires that this element should be placed to midplane of the actual structure for
better reflection of the characteristics of the actual structure. Figure 3.3.1 shows the
midplane modeling of a part used in rack structure. In Figure 3.3.1-a, the part to be
modeled with finite element modeling is shown. In Figure 3.3.1-b, the finite elements
used in the modeling of this part are shown. Finally, in Figure 3.3.1-c, the side view
of the modeled part is shown and the surface, where the finite elements were located,
can be seen between the actual faces of three dimensional part.

22
 Figure 3.3.1 Midplane Modeling Example

As mentioned before the finite elements, which can be seen in Figure 3.3.1-b,
were located in the midplane of the three dimensional structure shown in Figure
3.3.1-a. Thus, when the thickness of the structure is added as the properties of the
two dimensional elements, the reflected structure seems like the Figure 3.3.1-c,
where upper and lower lines represent the two faces of the three dimensional
element. The line between upper and lower ones represents the actual location of the
two dimensional element.
Another way of decreasing the total degree of freedoms is using the minimum
possible number of elements. This can generally be done by using larger sized
elements, but the larger elements results in less accuracy in the final results. The
finite element model of the rack structure of this thesis was prepared in such a way
that the smaller sized elements were used in the critical sections like rack to

23
understructure connections and equipment connections, whereas the larger sized
elements were used in the non critical regions like the sections of vertical and
horizontal profiles of the rack structure where no other profile is connected. As a
common procedure the required size of the elements were determined such that all
the elements have minimum dimensions larger than the thickness of the modeled
structure and there would be more than one element in each dimension of the
structure except the thickness. The application of this method led to accurate results
in the critical sections and less number of degree of freedoms in the non-critical
sections. Figure 3.3.2 shows the finite element model prepared for the preliminary
analysis of the rack structure. The boxed region of Figure 3.3.2 was further
highlighted in Figure 3.3.3. Figure 3.3.3 shows an example of different element
sizes. As the fitting connecting rack to the understructure is a critical part, the
elements used in this part are smaller than those in the other relatively less critical
parts of the figure.

Figure 3.3.2 Finite Element Model of the Rack Structure

24
 Figure 3.3.3 Different Mesh Size Example

Equipments placed in the rack are modeled as point elements, which have the
corresponding equipment’s mass, at the center of gravity of each equipment. These
elements are connected to other structure at the fastener points by RBE3 type of
multi-point constraint elements as a common finite element modeling procedure. The
function of these elements is to distribute the load to the connection points by taking
the elasticity of the equipments’ structures into account. The utilization of this
procedure makes it possible to reduce the degree of freedoms remarkably. By this
application, each equipment is modeled with 6 degree of freedoms instead of
thousands. In addition to this, equipments are to be supplied by supplier firms and
these firms should have done all the structural analysis. By not modeling the
equipment structures, results for the equipment structures, which are out of scope of
the design procedure, would be hindered. Figure 3.3.4 shows a view of finite element
model of rack structure in which the each equipment and its connection elements can
be seen.

25
 Figure 3.3.4 Finite Element Model of Equipment Rack, Front View

The common procedure of modeling fasteners in MSC.Patran® is removing

the fastener holes from the connected structures and then connecting the center of
these holes with multi-point constraint type of elements having infinite rigidity in the
user specified directions, which are the three translational axes usually. Figure 3.3.5
shows an example of fastener modeling. Figure 3.3.5-a shows the modeled parts and
Figure 3.3.5-b shows the finite element model of the same section. Black circles in
both figures represent the fastener locations and multi-point constraint elements used
in place of fasteners are shown in red circles in Figure 3.3.5-b. These types of
elements ensure that the two connected nodes always keep the same distance and
allow the load transfer in both directions like the real life application of the fasteners.
In the rack structure model, the fasteners at the connections between the rack
structure and the understructure were modeled as single point constraints, which are
boundary conditions of zero displacement in all translatory axes. Idea under this
application is that as the connected structure is assumed to be infinitely rigid,
prevents the motion in all directions, the finite element analysis yield more
26
conservative results. In addition to this, by using the single point constraint elements
modeling of the surrounding structures are avoided, which means smaller sized
models and less time spend in modeling.

Figure 3.3.5 Fastener Modeling Example

27
 The developed finite element model of the rack structure was further analyzed
for the ultimate acceleration factors of 7g at each translational axis separately. The
analysis was done by using the linear static module of MSC.Nastran® package
program. These acceleration factors were obtained from the company’s resources and
for obtaining the limit acceleration factors they should be divided by 1.5. Since
ultimate acceleration factors were used in the analysis, the ultimate material physical
properties were also assigned. As the used analysis type is linear static it is possible
to obtain the limit load results by dividing the ultimate loads by 1.5.
By using the results of the finite element analysis of the preliminary design, it
is possible to strengthen the insufficient parts, which cannot resist the loads they are
supposed to carry, and/or change the design concept of any part of the structure
completely. Since these results yield the malfunctioning parts of the structure and
parts undergoing high loads or large displacements, the design phase, from now on,
can be more focused and detailed. Figures 3.3.6, 3.3.7 and 3.3.8 show the von Mises
stress distribution on the designed structure for forward, sideward and downward
acceleration respectively. It should be noted that the legends showing the range of
the results were limited by 300 MPa for determination of critical parts. This 300 MPa
limit was determined randomly since most of the standard aluminum parts have
higher ultimate strengths, and so, it is an acceptable value for these aluminums. In
Figures 3.3.6, 3.3.7 and 3.3.8 regions which are shown with red color represent the
regions undergoing von Mises stress of at least 300 MPa, which are the critical
sections of the structure designed.

28
Figure 3.3.6 High von Mises Stress Regions for Rack Structure under Ultimate
Forward Acceleration (7g in -X) Case

29
Figure 3.3.7 High von Mises Stress Regions for Rack Structure under Ultimate
Sideward Acceleration (7g in +Y) Case

30
 Figure 3.3.8 High von Mises Stress Regions for Rack Structure under Ultimate
Downward Acceleration (7g in -Z) Case

Figure 3.3.6 shows the stresses on the aluminum parts of the structure, under
the 7g forward acceleration, according to the legend. As it can be seen in the figure,
stress is concentrated at the connections of the front vertical profiles and the
horizontal profiles of the shelves, and sections undergoing high stresses enlarge from
top to bottom. The reason of the enlargement in these sections is that the heaviest
equipments were placed on the lower shelves and the load on the upper shelves flows
through the vertical profiles to the lower front fittings. These sections are the stress
concentration sections which are the results of combining load paths. Figure 3.3.7
and Figure 3.3.8 show that the stresses in the parts forming the structure are not high
except in some sections, which can be lowered with small changes.
Another important result of the finite element analysis is the deflections of the
structure. It is important to have no undesired contact, between the designed structure
and the understructure or between the parts of the designed structure, under the limit
loads representing the flight conditions. For the designed rack structure, the critical
31
regions for the deflection profile are the curved and horizontal profiles at the rear
part of the rack structure. The minimum distance between these parts and the frames
is about 10 mm which is the case for top and bottom sections of curved profiles. This
distance is bigger than 10 mm in the other sections of curved profiles and this
indicates that the maximum deflection, in the back profiles of the structure, had to be
less than 10 mm under flight conditions. As mentioned before, the limit loads can be
obtained by dividing ultimate loads with a factor of 1.5 (i.e. 7g/1.5=4.67g). The
deflection plots for three critical limit loads, which are in the forward, sideward and
downward directions are given in Figures 3.3.9 to 3.3.11 respectively. The
displacement values shown in these figures are in mm.

Figure 3.3.9 Displacement of the Rack Structure under Ultimate Forward

Acceleration (4.67g in -X) Case

32
Figure 3.3.10 Displacement of the Rack Structure under Ultimate Sideward
Acceleration (4.67g in +Y) Case

33
 Figure 3.3.11 Displacement of the Rack Structure under Ultimate Downward
Acceleration (4.67g in -Z) Case

As it can be seen in Figures 3.3.9 to 3.3.11 rack structure, formed in the

preliminary design phase, undergoes relatively large displacements under the limit
load conditions. Since the displacements are aimed to be below 10 mm, especially in
the outer structure of the designed structure, stiffness of the structure in X and Y
directions (i.e. longitudinal and horizontal directions of the aircraft) should be
increased. Although the maximum displacements in the forward and sideward
acceleration conditions are on the outer structure of the new design, maximum
deflection in the downward acceleration is in the inner structure in which the gaps
are larger because of the equipments’ geometrical and electrical restraints.
By using the stress distributions and displacement pattern of the rack
structure, developed in preliminary design phase, it is possible to perform the
detailed design.

34
3.4 Detailed Design Phase

In the detailed design phase of the structure all the geometrical and
mechanical (i.e. material etc.) features of the parts forming the rack structure are to
be determined. This phase requires the finite element analysis of the structure
designed in the preliminary design phase. The finite element modeling/analysis will
be carried on simultaneously with the design procedure. By obtaining the finite
element analysis results for each change made in the rack structure, the design
procedure will be more precise since it will be possible to see if the change is
appropriate or not.
As mentioned in the chapter “Finite Element Modeling and Preliminary
Analysis Phase” it is important that the stiffness of the rack structure in the X and Y
directions (i.e. longitudinal and horizontal directions of the aircraft) should be
increased, displacements should be limited to 10 mm in the outer structure in the
limit loads and the high stress sections in the forward acceleration condition should
be eliminated. It seems that number of understructure connections are not enough
since combination of the load paths yields high stress regions and the displacement in
the upper section of the rack is much higher than the aimed value. Since there are
frames of the aircraft at the rear side of the rack structure, it is possible to add fittings
at the upper rear side of the structure connecting this section to understructure. These
fittings will prevent the high displacements at the upper side of the rack and will
form new load paths leading to understructure, which will be helpful in the
decrement of the stress concentrations. Since there are 5 curved profiles which are in
the vertical direction, 5 fittings will be formed on the upper side of these profiles.
These fittings will be called as “upper fittings” and labeled from U1 to U5 in the
aircraft’s longitudinal direction and from nose to tail. Figure 3.4.1 shows the location
of these fittings.

35
 Figure 3.4.1 Location and an Example of Upper Fittings

In addition to upper fittings, diagonal elements will be used to increase the

stiffness of the structure in the planes which are critical in terms of displacement or
stress distribution pattern. The idea behind the usage of these diagonal elements is
that by forming two triangular elements instead of one rectangular element more load
paths are formed and since the load is distributed between more parts, load carrying
capacity of the structure in the plane, which the diagonal elements are formed,
increases. The diagonal elements formed in the rack structure can be seen in Figure
3.4.2 in yellow.

36
 Figure 3.4.2 Formed Diagonal Elements

As it can be seen in Figure 3.4.2, 16 diagonal elements were formed in the

longitudinal direction of the aircraft. These diagonal elements were placed such as 8
of them lies in the bottom section of the rack structure and the remaining ones were
placed in the top section. These diagonals will be useful in the forward acceleration
condition and placed to reduce the stress concentrations at the connections of
horizontal and front vertical profiles. In addition to these profiles, diagonal elements
between the rear curved and front vertical profiles were formed for increasing the
load carrying capability in the sideward direction. These profiles were placed to all
sections at which curved vertical profiles exist at the rear side of the structure. This
placement was determined by using the fact that the rear profiles transmit the load to
the back and upper fittings and so the load on these profiles will be relatively higher.
With the help of the diagonal elements formed in aircraft’s vertical direction, load on
the curved profiles will be lowered. This will contribute to the load carrying
capability of the structure, mostly in vertical and horizontal directions.
Since the design applications and the finite element analysis will be
conducted simultaneously in this phase, it is important to reflect the design changes

37
to finite element model. The finite element model formed after the addition of upper
fittings and diagonal profiles can be seen in Figure 3.4.3.

Figure 3.4.3 Finite Element Model after the Addition of Upper Fittings and Diagonal
Profiles

In the detailed design phase cover and cabling designs should be completed
also. The covers will be placed in the sections where the rack structure have access
(i.e. faces except the back face of the rack which is closed by understructure). These
covers will be placed to every section where the horizontal and vertical profiles
forms rectangular areas and connected to these profiles. The connection points of
covers to structure will be designed such that no load will be transmitted from the
structure and only loads, which could arise because of the mass of the covers, will be
transmitted to rack structure from these points. The idea behind this application is
that the covers can be lighter as they will carry only their own mass. The
achievement of this one way load flow is to be satisfied by using special fasteners at
the connections of covers. The fastener holes at these locations will be designed to be
larger than the fasteners that will be used in these connections. Since the difference
38
between the fasteners and hole diameters will allow the fasteners to move freely in
the holes, no load will be transferred to covers through the fasteners. There will be
140 points of cover connections which will transmit the load created by the 46 kg of
cover mass to the structure. In addition to covers, harness routing will be done in this
phase. Since power and data transfer between the equipments are done through
cables, these cables should be routed and supported in the rack structure. The
procedure used in the harness routing is supporting the cables with the rack structure
by the help of simple supports called as harness brackets. These brackets will be
connected to rack structure and used for holding the cables. Since loads that would
arise because of the mass of the cables will be transferred to structure from these
brackets, higher number of brackets is preferred. As harness connection points are
usually located in the rear side of the equipments and it is not desired to have cables
at the front side for not to prevent access to equipments, these brackets will be
located at the rear side of the rack structure. Another reason of having harness
brackets at the rear side of the rack structure is that the cabling between the
equipments in the rack and other devices in the aircraft is to be routed on frames and
stringers of the aircraft, which are located at the rear side of the structure. When all
the cabling is done, there will be 60 brackets connected to structure and carrying a
total of 30 kg of harness.
It is essential to add the masses of the covers and harnesses to the finite
element model as total mass of the covers and the harnesses are high enough to affect
the final configuration of the rack structure. The total mass of the covers covering the
rack structure were modeled with point elements at the cover connections. As
mentioned before the fasteners connecting covers to the rack structure were designed
such that load flow from rack to covers is prevented. That is the reason why point
elements were used instead of two dimensional elements while modeling the covers
as two dimensional elements could lead to incorrect load paths. The incorrect load
paths, if occurred, causes misleading finite element analysis results, which could end
up with unexpected failures. As the change in the load path could result in obtaining
smaller stress results from the finite element analysis, when compared to actual case,
failures could be missed. Like the covers, the cabling mass on the rack structure was
distributed to the bracket points at which the cables are connected to rack structure
by point elements. The connectors connecting the cables to the equipments were not
39
included as separate finite element models, but their mass values were added to
connected equipments’ masses. Figure 3.4.4 shows the finite element model after the
addition of cover and cabling masses. When the section boxed in this figure is
zoomed, point elements representing the cover connection points and cabling
supports can be seen as in Figure 3.4.5.

Figure 3.4.4 Finite Element Model after the Addition of Cover and Cabling Masses

Figure 3.4.5 Highlighted Cover and Cabling Mass Elements

40
 Figure 3.4.4 shows the finite element model after point elements representing
the cover and cabling masses added. Additionally, Figure 3.4.5 shows the section,
which is boxed in Figure 3.4.4, while showing the point elements representing the
cover connections and cabling supports in this section circled. The connection points
of the cover, placed between the two horizontal and two vertical profiles, are circled
with red color. As it can be seen in the figure, the cover placed in this section has 4
connection points distributed to front vertical profiles. The mass of the cover was
distributed to these points with point elements whose locations are shown with red
circles. In addition to cover masses, the point elements, used for the simulation of the
mass of the cabling in this section, are shown in black circles. These two circles,
which are at the rear face of the rack structure, represent the locations of the two
harness support locations. These two brackets support the cable between the
equipment placed in the zoomed section and the largest equipment of the rack, which
is at the next section. As it can be noted in the figure, the distance between two
support locations were kept relatively small to have smaller mass values divided on
each element. And also this is the case for all of the harness routing supports in the
rack structure as 60 brackets were used for the total harness mass of 30 kg.
As the finite element model represents the final configuration of the rack
structure formed up to this point, it is possible to apply the finite element analysis to
obtain new stress and displacement distribution pattern on the structure. By using the
results, which will be obtained from the finite element analyses that will be done
after each design change, all the detailed features (i.e. dimensions, utilized materials
etc.) of the rack parts can be determined precisely. In the determination of these
features, surrounding parts and material selection play important roles. As all parts of
the structure are to be connected to other parts, general dimensions of the parts affect
each other. So, changing these dimensions is not an easy task except changing the
thickness values. As a result, material selection becomes more important in the
strength of the designed parts.
The materials, which will be used in the structure, are to be chosen from the
aluminums (2024 and 7075 series) and steels (PH series) which are frequently used
in the aerospace industry. For the rack structure, Aluminum 2024 T3511, Aluminum
7075 T6 Clad, Aluminum 7075 T7351, 17-4PH stainless steel and 17-7PH stainless
steels are to be used. The material and thickness configuration will be determined by
41
carefully analyzing the density, cost, processability and strength characteristics.
Among the listed materials, Aluminum 2024 series have the lowest strength, cost and
highest processability whereas steels have the highest strength and cost but the
lowest processability. Although Aluminum 7075 series stand between the Aluminum
2024 series and steels in the strength, cost and processability characteristics, and
density of the 7075 series are nearly same with 2024 series which is lower than the
density of steels. To fulfill the aim of light-weight, easily producible and cost
effective structure, selection of the materials of the parts of the structure is to be done
regarding the physical properties of the materials, which are listed in Table 3.4.1.
The required strength value for any part, which will be determined from the finite
element analysis results, is to be achieved with the material selection and thickness
change by using the physical properties of the materials following the listed aims.

Table 3.4.1 Material Properties

Modulus of
Density Poisson’s
Allowable (MPa) Elasticity
Material (kg/m3) Ratio
(MPa)
Ftu Fty Fsu Fbru E ρ ν
Aluminum 2024
393 289 200 744 74463 2768 0.33
T3511
Aluminum 7075
469 393 255 889 71705 2796 0.33
T7351
Aluminum 7075 T6
489 420 289 979 71016 2796 0.33
Clad
17-4PH Stainless
1068 999 772 1723 196500 7833 0.27
Steel
17-7PH Stainless
1220 1034 772 2420 199948 7640 0.28
Steel

As mentioned before, it is important that all of the parts of the rack structure
should be safe under both ultimate and limit load conditions, with minimum margin
of safety values of 0.2 as required. It should be noted that for the listed materials

42
ultimate strength to yield strength ratios are smaller than 1.5, which is equal to the
ultimate load to limit load ratio. This means that, since linear static module of the
MSC.Nastran® is used in the finite element analysis, margin of safety values
evaluated in the limit load cases will be bigger than the ones in the ultimate load
cases. So, material selection can be carried out by using the results for ultimate loads,
regarding the ultimate properties of the materials. For each part of the rack structure
alteration of utilized material, thickness and geometry should be done by considering
the aims of this design study, which are forming light-weight, easily producible and
cost effective structure having smallest possible interface loads transmitted to
understructure. Using the physical properties listed in Table 3.4.1 with the results
obtained from finite element analysis makes it easier to fulfill these aims.
It is advantageous to finalize the fitting designs firstly, as these parts are very
important in this design study, since displacement and interface load distribution of
the structure depends upon the elasticity of these fittings. As the entire load, created
in the rack structure because of the acceleration factors applied on the masses, will
flow to understructure through this parts, they should be strong enough to resist
relatively high loads. So, it is important to choose a utilized material and required
thickness couple for each fitting which will yield the lightest structure without
omitting the requirements of producability and cost effectiveness.
Since the sections that the fittings will be placed determines the basic
geometry of these fittings, the results of the finite element analysis obtained by using
these basic geometries gives a general idea for the detailed design. As the loads on
the fittings and the general stress distribution were obtained from the finite elements
analysis, materials and the thickness of the sections in touch with the understructure
were determined by regarding these results. Material that will be utilized to all the
fittings of the rack structure were chosen to be 17-4PH stainless steel because of the
high stress values on these parts and the necessity of having small displacements in
the rack structure. As this material is much stiffer than the aluminums, displacement
distribution over the rack structure decreases relatively when this material was
utilized to all of the fittings. In addition to this, thickness of the fittings, in the
sections where the understructure connection fasteners exist, were found to affect the
load distribution on the understructure significantly. Since this distribution is aimed
to be homogenous as much as possible, different thickness values were tried in many
43
analyses. The maximum values of components of the forces, which are transmitted to
understructure, at each fitting location are presented in Table 3.4.2. These force
components were obtained for the case, which yielded smaller differences when
compared to other cases. This was the case when the understructure connection
fastener sections had 5 mm thickness for back fittings, 3 mm thickness for front and
upper fittings.

Table 3.4.2 Maximum Force Components at Understructure Connections at Each

Fitting (Forces are given in absolute values)

Acceleration
Fitting Fx (N) Fy (N) Fz (N)
Case
Forward 1552,24 697,84 3572,46
B1 Sideward 1596,11 879,27 3752,15
Downward 153,40 29,45 431,20

Forward 1586,11 640,59 2914,78

B2 Sideward 2610,55 1716,13 9023,57
Downward 238,86 117,79 905,33

Forward 528,76 353,90 1625,97

B3 Sideward 2659,92 1591,51 8200,91
Downward 574,05 351,41 3613,16

Forward 274,58 129,33 668,15

B4 Sideward 2152,42 1357,27 4753,02
Downward 561,56 395,92 1530,21

Forward 457,74 980,31 1448,97

B5 Sideward 2125,68 2007,54 4069,49
Downward 160,21 439,02 1725,13

Forward 227,83 225,07 1341,20

F1 Sideward 242,92 256,79 978,52
Downward 28,87 22,23 610,69

Forward 902,49 184,17 4743,34

F2 Sideward 421,38 209,18 1163,11
Downward 71,55 84,50 775,32

Forward 798,41 177,11 4985,81

F3 Sideward 475,42 346,57 1632,99
Downward 307,20 120,58 770,86

44
Table 3.4.2 (continued) Maximum Force Components at Understructure
Connections at Each Fitting (Forces are given in absolute values)

Acceleration
Fitting Fx (N) Fy (N) Fz (N)
Case
Forward 797,84 175,73 5075,00
F4 Sideward 555,77 317,71 1721,22
Downward 129,09 54,94 546,87

Forward 474.06 32.85 830.22

F5 Sideward 277.11 313.57 1225.31
Downward 114.24 43.11 790.30

Forward 784.12 170.10 5062.43

F6 Sideward 374.57 177.99 1045.97
Downward 97.18 36.51 578.96

Forward 752.38 194.64 4846.85

F7 Sideward 392.06 261.23 1196.04
Downward 153.84 77.80 394.01

Forward 806.33 131.88 4991.10

F8 Sideward 360.59 220.33 1249.84
Downward 97.51 16.49 188.46

Forward 795.45 140.03 5101.63

F9 Sideward 352.04 194.73 1445.13
Downward 110.66 81.27 994.42

Forward 822.83 149.84 4793.55

F10 Sideward 489.18 354.67 1571.76
Downward 264.97 123.89 701.15

Forward 825.86 183.73 4732.05

F11 Sideward 581.95 323.09 1660.50
Downward 236.53 61.84 875.18

Forward 714.09 145.94 186.49

F12 Sideward 258.02 300.97 1222.87
Downward 84.58 48.83 719.82

Forward 815.57 209.88 4976.04

F13 Sideward 341.05 196.32 1183.30
Downward 139.32 41.63 949.25

45
 Table 3.4.2 (continued) Maximum Force Components at Understructure
Connections at Each Fitting (Forces are given in absolute values)

Acceleration
Fitting Fx (N) Fy (N) Fz (N)
Case
Forward 769.95 172.95 5037.87
F14 Sideward 354.76 249.39 1321.26
Downward 300.36 88.23 1007.78

Forward 847.13 201.41 4880.47

F15 Sideward 388.47 222.71 1191.72
Downward 173.89 32.09 672.88

Forward 303.04 125.23 920.99

F16 Sideward 162.01 179.84 742.63
Downward 50.56 42.45 876.86

Forward 1418.71 2034.01 1809.95

U1 Sideward 526.35 1890.19 1529.56
Downward 240.49 622.44 555.88

Forward 1674.02 2798.75 2721.58

U2 Sideward 574.45 2558.63 2276.94
Downward 96.49 907.52 960.15

Forward 1168.77 2134.54 2433.51

U3 Sideward 147.95 2884.12 1201.74
Downward 303.35 1343.90 980.88

Forward 1832.56 2711.61 2552.28

U4 Sideward 613.95 2580.54 2372.04
Downward 419.38 1178.32 751.51

Forward 1724.86 2101.86 2141.33

U5 Sideward 368.36 1202.37 1283.61
Downward 51.03 549.32 415.74

Since the material and the thickness of the fastener connection regions
changes the von Mises stress distribution on the fittings, detailed design of these
parts are to be done with new finite element analysis results. In the light of these new
results each fitting will be studied separately and will be shown according to labels
shown in Figure 3.4.6.

46
 Figure 3.4.6 Fitting Labels

Back Fittings:

Back fittings are the lower fittings located at the rear side of the rack structure
and labeled from B1 to B5. These fittings are used for connecting the rear curved
profiles and lower horizontal profiles, connected to these curved profiles, to the
frames under the rack structure. Since the connected parts’ geometry and
understructure connections are different in each section, back fittings will have
different geometries to match these sections.
Fitting B1 is located at the bottom of the curved profile which is the one
nearest to the aircraft’s nose. This fitting has connections to the curved profile and
the horizontal profile, between the front vertical and rear curved profiles, at the upper
side and a fastener connection to frame at the lower side. The location of Fitting B1
can be seen in Figure 3.4.7.

47
 Figure 3.4.7 Location and Connection Pattern of Fitting B1

As it can be seen in Figure 3.4.7, Fitting B1 was designed such that it has 3
fastener connections, shown with holes, to curved and horizontal profiles. In addition
to this, change in the geometry of the fittings at the sections intersecting with other
profiles can be seen. As mentioned before, surrounding structures’ geometry play an
important role in the detailed design of fittings.
For the detailed design of this fitting, von Mises stress distribution obtained
from the finite element analysis, which was conducted with the preliminary fitting
designs, was analyzed priorly. It would be possible to have an idea about the required
section thicknesses by regarding von Mises stress levels over the fitting. But, the
fundamental effect on the design of the fittings was the geometrical restraints created
by the connected structures and the required thickness for the understructure
connections. So, the design procedure used in the design of the fittings came out to
be such that, a fitting design, which was consistent with the geometrical restraints,
was prepared and by using the finite element analysis it was figured out if the
required strength obtained or not. If the designed fitting did not satisfy the
requirements, applied procedure was strengthening the malfunctioning sections or
designing a new geometry. Figure 3.4.8 shows the final design for Fitting B1 which

48
was consistent with the geometrical restraints and which satisfied the required
margin of safety results in each section.

Figure 3.4.8 Finalized Geometry of Fitting B1

Fitting B2 is located at the bottom of second curved profile and connected to

this profile and the horizontal profile lying in the horizontal direction of the aircraft
as shown in Figure 3.4.9.

Figure 3.4.9 Location and Connection Pattern of Fitting B2

49
 The 3 connection points of Fitting B2 to each of the curved and horizontal
profiles can be seen in Figure 3.4.9, which are shown as holes. The finalized
geometry of this fitting, which is consistent with the surrounding geometry and has
the shown connections, was obtained as shown in Figure 3.4.10. This fitting has also
two fasteners connected to understructure whose locations can be seen in the lower
section of the fitting.

Figure 3.4.10 Finalized Geometry of Fitting B2

Third back fitting which is labeled as “Fitting B3” is located at the separation
of the rack’s modules. This fitting is connected to the curved and the bottom
horizontal profiles of the module, which is in the front in the aircraft’s longitudinal
direction. The location of the fitting and the 6 fastener locations, which are shown
with holes, connecting this fitting to the mentioned profiles can be seen in Figure
3.4.11.

50
 Figure 3.4.11 Location and Connection Pattern of Fitting B3

As it can be seen in Figure 3.4.11 Fitting B3 has 3 fasteners connected to

curved profile, shown with green color, and 3 fasteners connected to horizontal
profile, which is shown with dark blue color. Since the understructure, under this
section of the rack structure, only allows one row of understructure connection
fasteners, it was not possible to add another fitting, which would be connected to
other module. For obtaining connection of this module to understructure an extra part
was added to Fitting B3 to connect it to the bottom horizontal profile of the module,
which is at the rear side of the aircraft. This additional part and the connection of the
fitting to the profiles of both modules can be seen in Figure 3.4.12.

51
 Figure 3.4.12 Connection of Fitting B3 to Both Modules

After the addition of the extra part, which was formed for the connection of
the modules of the equipment rack structure, the final geometry of the Fitting B3 is
shown in Figure 3.4.13. This geometry was obtained as the one satisfying required
margin of safeties in each section, after some finite element analyses conducted with
different geometries.

52
 Figure 3.4.13 Finalized Geometry of Fitting B3

Fitting B4 is the fitting located at the bottom of the fourth curved profile of
the rack structure. It is connected to the curved and bottom horizontal profiles like
the other back fittings. The location and the connection pattern of the Fitting B4 are
shown in Figure 3.4.14.

Figure 3.4.14 Location and Connection Pattern of Fitting B4

53
 As it can be seen in Figure 3.4.14 Fitting B4 has 3 connection points to both
of the curved and the vertical profiles, which are shown with green and dark blue
colors respectively. In addition to this, upper part of the fitting was designed such
that it would be consistent with the connected profiles. The final state of the
geometry of the Fitting B4, which has 2 fastener connections to understructure, is
shown in Figure 3.4.15.

Figure 3.4.15 Finalized Geometry of Fitting B4

Last of the back fittings is the Fitting B5, which is located at the rack’s
nearest section to the aircraft’s tail. Connection of this fitting to the rear vertical
profile and the horizontal profile was done with 3 fasteners each. The fastener
connection and the placement of this fitting are shown in Figure 3.4.16.

54
 Figure 3.4.16 Location and Connection Pattern of Fitting B5

The profiles which restrain the Fitting B5 on the upper side can be seen in
Figure 3.4.16 with green and dark blue colors. On the lower side the 2 fastener
connections made to the understructure played an important role in the formed
geometry of this fitting. The geometry of the Fitting B5, formed to have mentioned
fastener pattern and required margin of safety values, is presented in Figure 3.4.17.

55
 Figure 3.4.17 Finalized Geometry of Fitting B5

Since by the completion Fitting B5 the detailed design of the back fittings
was completed it was important to perform one more analysis to obtain final results.
As change in the load carrying capacity of one fitting affects the loads in the other
ones it was important to use the finalized designs of all of the fittings. The highest
von Mises stress distribution on the back fittings was obtained under ultimate
sideward acceleration. This stress distribution is shown in Figure 3.4.18 for all of the
back fittings. The circled fitting in this figure, which is Fitting B4, is the one with the
highest stress values and the zoomed view of this fitting is shown in Figure 3.4.19.

56
Figure 3.4.18 von Mises Stress Distribution on the Back Fittings under Ultimate
Sideward Acceleration (7g in +Y) Case

Figure 3.4.19 von Mises Stress Distribution on the Fitting B4 under Ultimate
Sideward Acceleration (7g in +Y) Case

57
 Finite element analysis was pointed out that; ultimate sideward acceleration
case is the case which yields the maximum stress distribution on the back fittings. As
it was mentioned before, Figure 3.4.18 shows the stress distribution on the back
fittings for the mentioned acceleration case and the fitting with the highest stress
values can be seen in Figure 3.4.19. As it can be seen in Figure 3.4.19 the maximum
stress value on the Fitting B4, so on the back fittings, is 504 MPa. This value is an
acceptable value for the utilized material in the back fittings, which is the 17-4PH
stainless steel. Since the highest stress values on the other back fittings are smaller, it
is obvious that no failure is expected on the back fittings under the given loading
conditions. Although the obtained maximum stress value was relatively small for a
steel part, back fittings were kept in shown geometries. The reason why these parts
were not made thinner was the requirement of high load carrying capability. Since
any size reduction in these would cause higher loads on the other fittings, this was
not preferred in the design.

Front Fittings:

Front fittings are the parts used for the connection of the front vertical profiles
to the understructure. These fittings are labeled from F1 to F16 and formed in two
different geometries to fit to the surrounding structures. Figure 3.4.20 shows the
location of the two fittings, which are located at the separation of rack’s modules and
connected to vertical profiles at this section.

58
 Figure 3.4.20 Location of Two Front Fittings at the Separation of Rack’s Modules

One example of each front fitting, which was designed to be consistent with
the vertical profiles and the understructure connections, is shown in Figure 3.4.21.

Figure 3.4.21 Examples of Finalized Geometry of the Front Fittings

59
 As it can be seen in Figure 3.4.21 each front fitting has 8 connections to
vertical profiles and 4 connections to understructure. The finite element analysis
conducted with the final geometry of these fittings yielded that, ultimate forward
acceleration case creates the highest stress distribution on the front fittings, which is
shown in Figure 3.4.22. The fitting, which has the maximum stress value on it, is
circled in this figure and the stress distribution on this fitting is shown in Figure
3.4.23.

Figure 3.4.22 von Mises Stress Distribution on the Front Fittings under Ultimate
Forward Acceleration (7g in -X) Case

60
 Figure 3.4.23 von Mises Stress Distribution on the Fitting F12 under Ultimate
Forward Acceleration (7g in -X) Case

As it can be seen in Figure 3.4.23 maximum stress value on the Fitting F12 is
at the region, where the vertical profile, having y direction as normal direction, and
the horizontal profile, having z direction as normal direction, are connected.
Although there exist a fillet in that region in the real case and it was not modeled in
finite element model, this stress value of 620 MPa is in acceptable range for utilized
17-4 PH stainless steel material. With the effect of the fillet, the stress in the real case
would be lower. As the highest stress values on the other front fittings are smaller no
failure is expected on these fittings under ultimate forward, sideward and downward
accelerations. Although the margin between the ultimate tensile strength of the
utilized material and the obtained maximum stress makes it possible to use smaller
thickness values for the front fittings, that was not preferred in order not to reduce
the load carrying capacity of these fittings.

61
Upper Fittings:

Upper fittings are located at the upper rear side of the rack structure and
connect the rack structure to the frames of the aircraft. These fittings are labeled from
U1 to U5. Because of the connected parts’ geometry and understructure connection
pattern, geometries of these fittings are different from each other. As like all of the
other fittings, the upper fittings’ finalized geometries were obtained by forming the
fittings that fit the surrounding structure and analyzing these to see if it satisfies the
required margin of safety values for each loading condition. The procedure used, if
the designed part was not satisfactory, was designing the fitting in a new geometry,
but still under the effect of geometrical restraints, or changing the thickness of the
insufficient sections, and conducting more finite element analyses.
Fitting U1 is located at the upper side of the curved profile which is the one
nearest to the aircraft’s nose. This fitting has connections to the curved profile and
the horizontal profiles, which lie in the aircraft’s horizontal and longitudinal
directions. At the upper side, this fitting is connected to frame of the aircraft which
corresponds to this section. The location of Fitting U1 can be seen in Figure 3.4.24.

Figure 3.4.24 Location and Connection Pattern of Fitting U1

62
 As it can be seen in Figure 3.4.24 Fitting U1 was designed such that it has a
total of 7 fastener connections, shown with holes, to the rack structure profiles. There
are 2 fastener connections to each of the green vertical profile and dark blue
horizontal profile. But, all the 7 fasteners are connected to orange horizontal profile
as this profile was extended throughout the connection region of the fitting. The
finalized geometry of the Fitting U1 can be seen in Figure 3.4.25.

Figure 3.4.25 Finalized Geometry of Fitting U1

As it can be seen in Figure 3.4.25 there are 8 holes in the upper part of the
fitting. This means that there are 8 fastener connections between the fitting and the
understructure.
Fitting U2, which is located at the top of second curved profile, is connected
to two orange horizontal, one dark blue horizontal, one yellow diagonal and one
green vertical profile which can be seen in Figure 3.4.26.

63
 Figure 3.4.26 Location and Connection Pattern of Fitting U2

As it can be seen in Figure 3.4.26 a total of 8 fastener connections were

formed between the Fitting U2 and the connected rack profiles. These fasteners were
distributed such that there are 2 fasteners connected to each of the green vertical,
dark blue horizontal and yellow diagonal profiles. In addition to these, there exist 4
fastener connections to each of two orange horizontal profiles, lying in the aircraft’s
horizontal direction, bearing in mind the fact that there are common fasteners
connecting three different parts. The finalized geometry of Fitting U2 is shown in
Figure 3.4.27.

64
 Figure 3.4.27 Finalized Geometry of Fitting U2

Fitting U3 is located at the separation of the modules of the rack structure.

This fitting is connected to two green vertical, two orange horizontal, two dark blue
horizontal and one yellow diagonal profile of the rack structure, as can be seen in
Figure 3.4.28. The gap which can be seen between the green profiles represents the
separation of two modules. This fitting was designed such that it provides connection
between the two modules of the rack structure in addition to rack to understructure
connection.

Figure 3.4.28 Location and Connection Pattern of Fitting U3

65
 As it can be seen in Figure 3.4.28, Fitting U3 has 8 fasteners at each of the
rack and understructure sides. This fitting has 2 fasteners connected to each of green,
dark blue and yellow profiles in addition to 4 fasteners connected to each of the
orange horizontal profiles. Also, as it can be seen in Figure 3.4.28 the understructure
connections of Fitting U3 lay on a flat surface, unlike the other upper fittings. The
reason of this is the geometrical aspects of the understructure. Frame corresponding
to this fitting has a wider flange than the other fittings and all of the fastener
connections, between the fitting and the frame, correspond to this flange. The
finalized geometry of Fitting U3 is given Figure 3.4.29.

Figure 3.4.29 Finalized Geometry of Fitting U3

The fittings labeled as “Fitting U4” is located at the upper side of the fourth
curved profile of the rack structure. It is connected to rear curved profile, diagonal
profile, and horizontal profiles lying in aircraft’s horizontal and longitudinal
directions. The location and the connection pattern of the Fitting U4 are shown in
Figure 3.4.30.

66
 Figure 3.4.30 Location and Connection Pattern of Fitting U4

As it can be seen in Figure 3.4.30 Fitting U4 has a total of 8 connection points

to the rack profiles. There are 2 connections between each of the green curved,
yellow diagonal and dark blue horizontal profiles and the fitting. In addition to these,
there are 4 fasteners, on this fitting, connected to each of the orange horizontal
profiles as some of them are common with other profiles. The finalized geometry of
the Fitting U4 is shown in Figure 3.4.31.

Figure 3.4.31 Finalized Geometry of Fitting U4

67
 The 8 holes which can be seen in Figure 3.4.31 in the upper part of Fitting U4
represent the understructure connections of this fitting. The reason why 4 of these
fasteners are lying on different direction from the others is the geometrical restraints
created by the understructure.
Fitting U5, which is the last of the upper fittings, is located at the upper side
of the last curved vertical profile, in the direction going form aircraft’s nose to tail.
This fitting has connections to curved vertical profile, shown with green color,
horizontal profiles lying in aircraft’s horizontal and longitudinal profiles, shown with
dark blue and orange colors respectively, and diagonal profile, shown with yellow
color, as can be seen in Figure 3.4.32.

Figure 3.4.32 Location and Connection Pattern of Fitting U5

As it can be seen in Figure 3.4.32, Fitting U5 was designed to have 8 fastener

connections to understructure, in the upper section of the fitting, in addition to rack
profile connections. The connection patterns of these fasteners were determined by
regarding the understructure geometry in this section. The finalized geometry, which
fits the surrounding structures’ geometry and satisfies the required margin of safety
results, of the Fitting U5 is given in Figure 3.4.33.

68
 Figure 3.4.33 Finalized Geometry of Fitting U5

After the completion of the detailed design of the upper fittings, finite
element analysis, which would yield the final results, was performed. The reason
why this final analysis was needed is the effect of the change in the load carrying
capacity of one fitting on the loads on other fittings. The highest von Mises stress
distribution on the upper fittings, which was obtained under ultimate sideward
acceleration, is given in Figure 3.4.34. The fitting, which has the highest stress
distribution when compared to others, is circled in this figure and zoomed view of
this fitting is shown in Figure 3.4.35.

69
Figure 3.4.34 von Mises Stress Distribution on the Upper Fittings under Ultimate
Sideward Acceleration (7g in +Y) Case

Figure 3.4.35 von Mises Stress Distribution on the Fitting U4 under Ultimate
Sideward Acceleration (7g in +Y) Case
70
 As it can be seen in Figure 3.4.35 maximum stress developed on the Fitting
U4 is the highest stress on the fittings. This condition makes Fitting U4 one of the
most critical parts in the rack structure. Although having a stress value of 959 MPa
on the fittings was not acceptable, analyzing the zoomed view, of the region with
maximum stress values, yielded different comments. For more detailed analysis,
actual geometry of this region is shown in Figure 3.4.36 and the stress distribution is
shown in Figure 3.4.37.

Figure 3.4.36 Actual Geometry of the Maximum Stress Region on Fitting U4

71
 Figure 3.4.37 Zoomed View of Highest von Mises Stress Region on the Fitting U4
under Ultimate Sideward Acceleration (7g in +Y) Case

When compared to geometry shown in Figure 3.4.36, the black line in Figure
3.4.37, which is drawn at 4 mm away from the connection line of the surfaces,
represents the thickness of the three dimensional profile. In addition to this, second
line, colored in blue, is drawn in the place of end of the fillet. In the finite element
model, load transfer between the shown profiles is done through the connection
nodes of two sections, which are on a single line. But, in real case, same load is
transferred through the area between the ends of the fillets in both directions. As this
maximum stress value was created by the load concentration on the connection
nodes, which is not the case in real life, stress distribution outside the fillet region
was taken into consideration as a common application. Since the stress values outside
the fillet were in the applicable range for the utilized material of 17-4PH stainless
steel, this part was accepted as a safe design. As the von Mises distributions on the
other upper fittings were smaller, it was concluded that finalized geometries of these
fittings satisfies the required margin of safety values.

72
 After the completion of the detailed design of fittings connecting rack
structure to understructure, stress distribution on the other parts of the rack structure
was changed. By using the finite element analysis results obtained using the finalized
geometry of the fittings, material and the thickness distribution, yielding the required
safety margins on each part of the rack structure, can be formed. Von Mises stress
distribution, on the inner parts of the rack structure, for the three critical loading
conditions, which are the forward, sideward and the downward acceleration cases,
can be seen in Figures 3.4.38 to 3.4.40.

Figure 3.4.38 von Mises Stress Distribution on the Inner Parts of the Rack Structure
under Ultimate Forward Acceleration (7g in -X) Case

73
Figure 3.4.39 von Mises Stress Distribution on the Inner Parts of the Rack Structure
under Ultimate Sideward Acceleration (7g in +Y) Case

74
Figure 3.4.40 von Mises Stress Distribution on the Inner Parts of the Rack Structure
under Ultimate Downward Acceleration (7g in -Z) Case

The von Mises stress distribution plots shown in Figures 3.4.38 to 3.4.40
show that there were stress concentrations in the vertical profiles of the rack
structure. In addition to vertical profiles, stress distributions on the horizontal profiles
of the rack were relatively higher than the other profiles. Major differences between
the stress distributions obtained for the preliminary design, which were given in
Figures 3.3.6 to 3.3.8, and for the detailed design, can be summarized as shown in
Table 3.4.3.

75
Table 3.4.3 Major Differences between the von Mises Stress Distributions Obtained
for Preliminary Design and Detailed Design

Location Acceleration Case Preliminary Design Detailed Design

Lower Horizontal - Ultimate Forward Stress is between
Stress is higher than
Vertical Profile Acceleration (7g in 135 MPa and 169
300 MPa
Intersections –X) MPa
Lower Horizontal Ultimate Sideward Stress is between
Stress is higher than
Profile which has Acceleration (7g in 35.6 MPa and 75.2
300 MPa
Lowest X Location +Y) MPa
The Most The Most
Ultimate Forward
Widespread Stress Widespread Stress
Whole Structure Acceleration (7g in
Range is 80 MPa to Range is 33.9 MPa
–X)
100 MPa to 67.7 MPa
The Most The Most
Ultimate Sideward
Widespread Stress Widespread Stress
Whole Structure Acceleration (7g in
Range is 60 MPa to Range is 0.01 MPa
+Y)
80 MPa to 35.6 MPa
The Most The Most
Ultimate Downward
Widespread Stress Widespread Stress
Whole Structure Acceleration (7g in
Range is 40 MPa to Range is 0.03 MPa
–Z)
60 MPa to 35.6 MPa

By using stress values on each profile which can be read by using the stress
distribution plots in Figures 3.4.38 to 3.4.40 and the legends shown in the right hand
side of the figures, required material and thickness couple for each part, yielding the
required margin of safety values, can be found. In the determination of this couple,
mechanical properties of the materials, which were listed before, were used
following the aim of obtaining light-weight, easily producible and cost effective
structure. The materials utilized in the profiles of the rack structure are presented in
Figures 3.4.41 and 3.4.42.

76
Figure 3.4.41 Front View of the Rack Structure Showing the Materials in Different
Colors (Blue: Al 2024 T3511, Yellow: Al 7075 T6 Clad, Green: Al 7075 T7351,
Red: 17-4PH stainless steel, Pink: 17-7PH stainless steel)

77
 Figure 3.4.42 Back View of the Rack Structure Showing the Materials in Different
Colors (Blue: Al 2024 T3511, Yellow: Al 7075 T6 Clad, Green: Al 7075 T7351,
Red: 17-4PH stainless steel, Pink: 17-7PH stainless steel)

As mentioned before materials utilized in the fittings, connecting rack to the

understructure, were chosen to be 17-4PH stainless steel and these parts were shown
with red colors in Figures 3.4.41 and 3.4.42. Other steel parts in the rack structure are
the four pink colored horizontal profiles carrying the heaviest equipment. Because of
the geometry and connection concept of this equipment, vertical profiles carrying the
equipment were connected to rack with these four profiles. Since these profiles have
to carry high forces and moments, using stainless steel of 17-7PH in these profiles
removed the necessity of using very thick profiles, if any aluminum was used. As
mentioned before, von Mises stress values on the vertical profiles of the rack
structure were high. When the stress plots of these profiles and the mechanical
properties of the materials were compared, it was concluded that using Aluminum
7075 series in these profiles would yield these parts to be lighter. As it was possible
to use Aluminum 2024 series in the horizontal profiles of the rack structure without
78
using very high thicknesses, this was preferred in design as Aluminum 2024 series
has the highest processibility and lowest cost among the materials used in this study.
The parts, which were made from Aluminum 2024 series, were represented in dark
blue color in Figures 3.4.41 and 3.4.42. The materials utilized in the supports of the
equipments were chosen by using the stress distribution values on these profiles.
The thicknesses of the equipment rack parts, when the material configuration
is as shown in Figures 3.4.41 and 3.4.42, are given in Figures 3.4.43 and 3.4.44
according to the legend. For some of the fittings, thickness distributions were given
in zoomed views as the highest thickness values were used in the fittings.

Figure 3.4.43 Thicknesses of the Parts of the Rack Structure, Front View
79
 Figure 3.4.44 Thicknesses of the Parts of the Rack Structure, Back View

As it can be seen in Figure 3.4.43 and Figure 3.4.44, maximum thickness

values in the rack structure were used in the fittings. This was the case because of the
fact that; since the loads created at the center of gravity of each part flows to
understructure through these fittings, they were to endure high loads. As these parts
are connected to several parts of the rack, load flowing from these connected parts

80
creates load concentration on the fittings. Also, curved profiles at the back of the
structure and the front vertical profiles were designed to be thicker than the
horizontal parts of each shelf, as these profiles transmit the forces of shelves to the
fittings. Finally, horizontal profiles are thicker than most of the equipment supports,
as these profiles carry more equipments than the equipment supports. The cross
profiles, throughout the rack structure, have one of the smallest thicknesses in the
rack structure.
In addition to margin of safety requirements for the stresses on the profiles
designed, another requirement was having small displacement values, especially in
the profiles which are close to understructure. This small deflection requirement
comes from the idea that; no undesired contact between the rack profiles and the
understructure should occur under the flight conditions, which are the limit load
conditions. The deflection plots for the three critical limit loading conditions of
forward, sideward and downward acceleration cases are given in Figures 3.4.45 to
3.4.47.

Figure 3.4.45 Deflection Plot of Rack Structure Profiles under Limit Forward
Acceleration (4.67g in -X) Case

81
Figure 3.4.46 Deflection Plot of Rack Structure Profiles under Limit Sideward
Acceleration (4.67g in +Y) Case

82
 Figure 3.4.47 Deflection Plot of Rack Structure Profiles under Limit Downward
Acceleration (4.67g in -Z) Case

It can be seen in Figures 3.4.45 to 3.4.47 that the maximum deflection near
the frames is about 6 mm and is smaller than the minimum gap, which was designed
to be 10 mm. Therefore, no failure is expected due to the loads which would arise
because of undesired contacts. Further study of the deflection plots yielded out that
the maximum displacement on the inner profiles of the rack structure was about 18.1
mm, on the supports of heaviest equipment, which was way below the distance to the
surrounding parts. As the deflections on the other profiles were not higher than the
gaps to the neighbor profiles, no problem was expected because of the displacement
values throughout the rack. Major differences, between the deflection plots obtained
for the preliminary design, which were given in Figures 3.3.9 to 3.3.11, and for the
detailed design, can be summarized as shown in Table 3.4.4.

83
Table 3.4.4 Major Differences between the Deflections Obtained for Preliminary
Design and Detailed Design

Location Acceleration Case Preliminary Design Detailed Design

Limit Forward The Maximum The Maximum
Whole Structure Acceleration (4.67g Deflection is 40.70 Deflection is 18.10
in –X) mm mm
Limit Sideward The Maximum The Maximum
Whole Structure Acceleration (4.67g Deflection is 69.40 Deflection is 12.00
in +Y) mm mm
Limit Downward The Maximum The Maximum
Whole Structure Acceleration (4.67g Deflection is 14.10 Deflection is 10.70
in –Z) mm mm
Limit Forward Deflections Change Deflections Change
Upper Shelf Acceleration (4.67g from 24.40 mm to from 0 mm to 6.04
in –X) 40.70 mm mm
Limit Sideward Deflections Change Deflections Change
Upper Shelf Acceleration (4.67g from 37.00 mm to from 0 mm to 4.01
in +Y) 69.40 mm mm
Limit Downward Deflections Change Deflections Change
Upper Shelf Acceleration (4.67g from 4.71 mm to from 0 mm to 7.87
in –Z) 13.20 mm mm
Limit Forward Deflections Change Deflections Change
Middle Shelf Acceleration (4.67g from 10.90 mm to from 2.42 mm to
in –X) 35.30 mm 18.10 mm
Limit Sideward Deflections Change Deflections Change
Middle Shelf Acceleration (4.67g from 18.50 mm to from 0.80 mm to
in +Y) 60.20 mm 12.00 mm
Limit Downward Deflections Change Deflections Change
Middle Shelf Acceleration (4.67g from 3.77 mm to from 0 mm to 8.58
in –Z) 14.10 mm mm

84
 Table 3.4.4 (continued) Major Differences between the Deflections Obtained for
Preliminary Design and Detailed Design

Location Acceleration Case Preliminary Design Detailed Design

Limit Forward Deflections Change Deflections Change
Lower Shelf Acceleration (4.67g from 2.71 mm to from 1.21 mm to
in –X) 19.00 mm 7.25 mm
Limit Sideward Deflections Change Deflections Change
Lower Shelf Acceleration (4.67g from 4.68 mm to from 0 mm to 4.01
in +Y) 32.4 mm mm
Limit Downward Deflections Change Deflections Change
Lower Shelf Acceleration (4.67g from 0 mm to 9.43 from 0 mm to 10.70
in –Z) mm mm

After the completion of the stress and displacement checks of the profiles of
the equipment rack structure buckling analysis was to be performed.

3.5 Buckling Analysis Phase

Buckling is the case in which the profile under discussion suddenly becomes
sharply curved under compression load, instead of remaining straight. As the buckled
structure can carry some amount of load, although not in the way that it was designed
for, buckling is not considered as a failure case. That is the reason why the buckling
of the rack structure is being presented in a separate section.
The buckling analysis was performed under the limit loading conditions as a
common procedure and the buckling module of MSC.Nastran® package program was
used for the profiles in the rack structure. The condition of having no buckling in any
profile is satisfied when the eigenvalues are higher than 1 for all of the loading
conditions. The eigenvalue mentioned here is a term used in MSC.Patran®, which
stands for the ratio of allowable load to applied load for each part [10]. So, buckling
analysis was performed to obtain eigenvalues for each of the limit forward, sideward

85
and downward accelerations. Figure 3.5.1 shows the results of buckling analysis
which was performed using the rack structure formed after the detailed design phase.

Figure 3.5.1 Buckling Analysis Results after the Detailed Design Phase

The boxed area with red color, in Figure 3.5.1, shows the first five
eigenvalues for each of the limit acceleration factors in forward, sideward and
downward directions. The function of these eigenvalues is indicating buckling in one
of the profiles, at the load which can be obtained when the applied load is multiplied
with the extracted eigenvalue. Which means that, when the limit forward acceleration
factor is multiplied with the first eigenvector of 0.80576, the profile, colored in the
left side of Figure 3.5.1, which is one of the diagonal profiles under the bottom shelf,
buckles. So, the profiles which would buckle, at the loads corresponding to the
eigenvalues in boxed in blue color, were to be strengthened. The three profiles,
yielding eigenvalues smaller than 1, are shown in Figure 3.5.2 in red color.

86
 Figure 3.5.2 Buckling Critical Profiles after Detailed Design Phase

As it can be seen in Figure 3.5.2, the three profiles which were buckling
critical are the diagonal profiles at the top and bottom section of the rack structure.
The thicknesses of these profiles were used as 1.27 mm after the detailed design
regarding the stress and deflections on them. As these profiles were obtained to be
buckling critical, using the buckling analysis procedure, they were to be strengthened
for not to buckle under limit loads. For having no buckling, the thicknesses of these
diagonals were increased to 1.6 mm. Buckling analysis results with these new
thicknesses is given in Figure 3.5.3.

87
Figure 3.5.3 Buckling Analysis Results Obtained after Increasing Buckling Critical
Profiles’ Thicknesses

The eigenvalues in the boxed area in Figure 3.5.3 showed that the buckling
loads of all of the profiles in equipment rack structure were above the limit loads as
required. So, it is obvious that no buckling in any profile is expected under the given
limit load conditions.
As thicknesses of three profiles were increased, which would yield higher
margin of safety values for these profiles, and no buckling is expected under limit
loads, the design study can be carried to next phase, which is the fastener selection
phase.

3.6 Fastener Selection Phase

The fasteners are usually modeled as multi-point constraint type rigid

elements in the finite element models. The so called fastener check process includes
the obtaining of tension and shear forces from MSC.Patran®. Then, by using the
mechanical properties of the fasteners and connected materials, the margin of safety
results for tension, shear, combined tension and shear, bearing and pull-through
88
failures can be found (see Appendix A). When the structure to be analyzed contains a
large number of fasteners, this procedure becomes unwieldy and almost impossible
to be applied. Although, only the maximum forces for the similar fastener types are
checked as a common procedure, the required process is time consuming and prone
to make mistakes. Hence to facilitate the analysis and avoid the likelihood of
mistakes, a FORTRAN Based Executable Tool and a Microsoft Office Excel Tool,
whose usage will be explained in Appendix B, is prepared for performing the
fastener check process for all fasteners.
Fasteners, connecting fittings of rack structure to understructure, were
modeled with single-point constraint elements, unlike the other fasteners which were
represented with multi-point constraint elements. The idea behind this approach is to
try to simulate the effects of understructure. As this application does not consider the
elastic characteristics of the understructure, the results obtained from the finite
element analysis become higher than the real case. This is a generally applied
procedure and because of the conservativeness of the results, carries no risk of
having unexpected failures.
The fittings were connected to understructure by screw type fasteners at the
lower side and hi-lok type fasteners at the upper side because of the geometrical
constraints. As upper fittings were connected to rack structure with screw type
fasteners, the understructure connections do not prevent the removability of the
structure. In addition to these, lower fitting connections to rack profiles were made
with hi-lok type of fasteners. The fasteners on the fittings are shown in Figures 3.6.1
to 3.6.3 in which the screws were modeled in their places and the hi-lok locations
were represented as holes.

89
Figure 3.6.1 Front Fitting Screw and Hi-Lok Locations

Figure 3.6.2 Back Fitting Screw and Hi-Lok Locations

90
 Figure 3.6.3 Upper Fitting Screw and Hi-Lok Locations

The developed tools for the fastener check processes were prepared for the
fasteners modeled with multi-point constraint elements hence the fasteners at the
understructure connections should be investigated separately. Maximum tension and
shear forces, obtained using the single point constraint forces obtained from finite
element analysis, and the margin of safety values obtained by using these forces are
given in Table 3.6.1. The formulas used for the calculation of these margin of safety
results are presented in Appendix A.

91
 Table 3.6.1 Margin of Safeties (MS values) for the Fasteners at Understructure
Connections

MS
Maximum Maximum
Region / MS MS Combined MS MS Pull
Tension Shear
Fastener Tension Shear Tension Bearing Through
Force (N) Force (N)
and Shear
Front
Fittings / 5101.63 904.56
0.592 7.552 0.361 19.721 3.666
NAS 1801- (Forward) (Forward)
08-8

Back
Fittings / 9023.57 3124.11
0.363 2.343 0.208 10.558 5.647
NAS1143- (Sideward) (Sideward)
10

Upper
3767 2670
Fittings / 0.550 1.904 0.269 7.114 4.567
(Forward) (Forward)
HL20PB5-8

As it can be seen from the Table 3.6.1 minimum margin of safety for the
fasteners between the fittings and the understructure is 0.208, which is greater than
the minimum required value of 0.2. Since the preload coefficients at these fittings
were taken to be 0.5, which is close to the maximum, it is highly likely that no failure
is expected on these fasteners.
Fasteners inside the rack structure were selected such that the equipment
supports and the connection profiles between the rack’s modules can easily be
removed. In the connection points, except the screws used in mentioned locations,
rivets and hi-loks were used. For all of the fasteners inside the rack structure the
developed fastener analysis tools were used and these fasteners were selected in
order to satisfy the minimum margin of safety requirement of 0.2.

92
 For satisfying minimum required margin of safety values for the failure cases
of tension, shear, combined shear and tension, bearing stress and pull-through stress
at every connection point, 10 different fasteners were used throughout the rack
structure. Table 3.6.2 shows the used fasteners and the required properties of these,
in the margin of safety calculations.

Table 3.6.2 Used Fasteners and Fastener Properties

Allowable Allowable Nominal Head

Fastener Pre-Tension
Fastener Tension Shear Diameter Diameter
Type Force (N)
Load (N) Load (N) (mm) (mm)

MS20470E4 Rivet 2664 2362 3.18 6.35 -

MS20470E5 Rivet 3442 3540 4.00 7.94 -

MS20470E6 Rivet 3838 4746 4.80 9.00 -

NAS1801-08 Screw 9341 8896 4.17 6.27 1054

NAS1101-08 Screw 9964 8940 4.17 6.74 1208

NAS1143 Screw 14145 12010 4.82 9.48 2662

HL19PB5 Hi-Lok 5738 8918 4.00 6.58 2224

HL19PB6 Hi-Lok 8896 11965 4.82 7.60 3113

HL20PB5 Hi-Lok 9697 8918 4.00 7.98 3558

HL20PB6 Hi-Lok 14145 11965 4.82 9.33 4893

93
 The fastener selection was the final phase of the equipment rack design study
conducted in this design study. With the completion of this phase, the rack structure
design - having minimum margin of safety value of 0.2 in every failure case under 7g
ultimate acceleration factors in each translatory axis of the aircraft - was finalized.
For the ultimate acceleration conditions in forward, sideward and downward
directions, no failure is expected in any part of the rack structure.

3.7 Dynamic Analysis

Dynamic behavior of the equipment rack was investigated for two different
cases. In the first case, all of the equipment, connector, cover and cable masses were
removed and in the second case, all of those were retained. The natural frequencies
and the mode shapes of the structure were determined for the both cases. The
dynamic analysis was performed by using the normal modes module of
MSC.Nastran®. This analysis was performed for the boundary conditions shown in
Figure 3.7.1.

94
Figure 3.7.1 Boundary Conditions Used for the Dynamic Analysis of the Rack
Structure

95
 As it can be seen in Figure 3.7.1, the rack structure was fixed in all degrees of
freedom at the understructure connecting fastener points. This means that the three
translational and the three rotational degrees of freedom of these points were
restrained. The first ten natural frequencies of both cases are given in Table 3.7.1.

Table 3.7.1 The First 10 Natural Frequencies for the Dynamic Analysis of the Rack
Structure

Only Structure Whole Structure

Mode
Profiles Including Masses
1 37.08 Hz 8.87 Hz
2 39.70 Hz 10.27 Hz
3 43.00 Hz 10.41 Hz
4 43.15 Hz 10.63 Hz
5 50.67 Hz 12.00 Hz
6 53.17 Hz 13.26 Hz
7 55.83 Hz 13.86 Hz
8 56.15 Hz 14.12 Hz
9 56.25 Hz 14.25 Hz
10 56.69 Hz 14.43 Hz

As it can be seen from the Table 3.7.1 the inclusion of the masses, as
expected, reduces the natural frequencies. Figures 3.7.2 to 3.7.6 give the first five
mode shapes of the rack structures.

96
Figure 3.7.2 1st Mode Shape of the Rack Structure Excluding Masses

Figure 3.7.3 2nd Mode Shape of the Rack Structure Excluding Masses

97
Figure 3.7.4 3rd Mode Shape of the Rack Structure Excluding Masses

Figure 3.7.5 4th Mode Shape of the Rack Structure Excluding Masses
98
 Figure 3.7.6 5th Mode Shape of the Rack Structure Excluding Masses

The modes for the rack profiles were usually the local modes. Since there are
many profiles in the rack structure, it is obvious that there are many local modes for
the rack structure.
For the second case, which includes equipment, connector, cover and cable
masses, the first five mode shapes are shown in Figures 3.7.7 to 3.7.11.

99
Figure 3.7.7 1st Mode Shape of the Rack Structure Including Masses

Figure 3.7.8 2nd Mode Shape of the Rack Structure Including Masses

100
Figure 3.7.9 3rd Mode Shape of the Rack Structure Including Masses

Figure 3.7.10 4th Mode Shape of the Rack Structure Including Masses
101
 Figure 3.7.11 5th Mode Shape of the Rack Structure Including Masses

Like the case without the masses, the modes obtained for the second case
were also mostly local modes.

3.8 Fitting Design Changes

Changes, which were made in the fitting designs throughout the design
process, are summarized in Table 3.8.1.

102
 Table 3.8.1 Resultant Changes in the Fitting Designs

Fitting First Design Final Design

5 mm lower section thickness
4 mm lower section thickness
17-4PH stainless steel
B1 Aluminum 7075 T7351
Additional flange in the middle
section
5 mm lower section thickness
4 mm lower section thickness
3 mm middle section thickness
4 mm middle section thickness
B2 17-4PH stainless steel
Aluminum 7075 T7351
Additional flange in the middle
section
5 mm lower section thickness
4 mm lower section thickness
6 mm curved profile connection
4 mm curved profile connection
section thickness
B3 section thickness
17-4PH stainless steel
Aluminum 7075 T7351
Additional flange in the middle
section
5 mm lower section thickness
4 mm lower section thickness
5 mm middle section thickness
4 mm middle section thickness
B4 17-4PH stainless steel
Aluminum 7075 T7351
Additional flange in the middle
section
5 mm lower section thickness
4 mm lower section thickness
6 mm curved profile connection
4 mm curved profile connection
section thickness
B5 section thickness
17-4PH stainless steel
Aluminum 7075 T7351
Additional flange in the middle
section
F1 to 4 mm thickness 3 mm thickness
F16 Aluminum 7075 T7351 17-4PH stainless steel

103
 Table 3.8.1 (continued) Resultant Changes in the Fitting Designs

Fitting First Design Final Design

6 mm middle section thickness 8 mm middle section thickness
U1
Middle section geometry change
6 mm middle section thickness 8 mm middle section thickness
U2
Fillet design change
6 mm middle section thickness 8 mm middle section thickness
U3
Middle section geometry change
6 mm middle section thickness 8 mm middle section thickness
U4
Fillet design change
U5 6 mm middle section thickness 8 mm middle section thickness

104
 CHAPTER 4

CONCLUSIONS

In this thesis study, an equipment rack structure carrying 343 kg of equipment

mass, 46 kg of cover mass, 30 kg of harness mass and 21 kg of connector mass was
designed. This structure was designed by using the Computer Aided Design (CAD)
program of Unigraphics and finite element analysis of the designed structure was
performed by using MSC.Patran® and MSC.Nastran® package programs. By
changing the design of inadequate parts until margin of safety values of 0.2 were
obtained, for each part and for every failure case, finalized design was formed. After
the formation of the finalized geometries of the profiles, fasteners to be used in the
connection points were selected. In the selection process of these fasteners, two
computer programs were developed for the automation of the fastener failure checks,
which are usually performed manually.
The fatigue analysis was not performed for this design since this is a structure
which is not safety critical. This structure is a secondary structure inside the aircraft
and any fatigue failure of any part of this structure will not be a catastrophic failure.
Since, failure of any part will not yield malfunctioning of a primary structure this
structure is not regarded as a safety critical part and fatigue analysis was not
performed by following the general design policy.
In the design study carried out in this thesis, it was concluded that for
obtaining a light-weight, easily producible and cost effective structure finite element
analysis and design process should be carried out simultaneously. In addition to this,
the distribution of the interface loads is affected by the number of smooth load paths,
between the load application points and the understructure.
Finally, load carrying capability of a part mostly depends on its dimensions
and utilized material’s strength values. In the designed rack structure, equipment
105
supports generally have relatively smaller thicknesses and weaker materials. On the
contrary, fittings, which are the understructure connections, have relatively higher
thicknesses and stronger materials. Using these facts and required thickness and
material information of the middle parts, it was concluded that; the load carrying
capability of the parts should increase from load application points to the
understructure, which means through the load paths.

106
 REFERENCES

[1] Hutton, D. V., “Fundamentals of Finite Element Analysis”, McGraw-Hill,

2004
[2] Cook, R. D., Malkus, D. S., Plesha, M. E., Witt, R. J., “Concepts and
Applications of Finite Element Analysis”, John Wiley & Sons Inc., 2001
[3] Chen, Z., “Finite Element Methods and Their Applications”, Springer, 2005
[4] Curtis, D. H., “Fundamentals of Aircraft Structural Analysis”, McGraw-Hill,
1997
[5] Woolfson, M. M. and Pert, G. J., “An Introduction to Computer Simulation”,
Oxford University Press, 1999
[6] Stolarski, T., Nakasone, Y., Yoshimoto, S. “Engineering Analysis with
ANSYS Software”, Elsevier, 2006
[7] Cook, R. D., “Finite Element Modeling for Stress Analysis”, John Wiley &
Sons Inc., 1994
[8] ANSYS®, User’s Manual for Revision 5.1, ANSYS® Inc., 1994
[9] ABAQUS®, User’s Manual Version 6.4, ABAQUS® Inc., 2003
[10] MSC.NASTRAN®, User’s Guide, MSC.Software©, 2001
[11] Peery, J. D., Azar, J. J., “Aircraft Structures”, McGraw-Hill, 1982
[12] Joint Service Specification Guide, “Aircraft Structures”, Department of
Defense, 1998
[13] Vaughan, E. R., Daniel, M. F., “The Correct Use of Finite Element for Stress
Analysis of Aircraft”, American Helicopter Society 60th Annual Forum, 2004
[14] Niu, M. C. Y, “Airframe Stress Analysis and Sizing”, Hong Kong Conmilit
Press Ltd., 1999
[15] Bruhn, E. F., “Analysis and Design of Flight Vehicle Structures”, Tri-State
Offset Company, 1973

107
[16] Nicholson, David W., “Finite Element Analysis: Thermomechanics of
Solids”, CRC Press, 2003
[17] Beer, F. P., Johnston, E.R., DeWolf, J. T., “Mechanics of Materials”,
McGraw-Hill, 2002
[18] Dowling, N. E., “Mechanical Behavior of Materials”, Prentice Hall, 1999

108
 APPENDIX A

FASTENER CHECKS

The fastener checks include the calculation of margin of safety results for
tension, shear, combined shear and tension failures of the fasteners and pull-through
and bearing failures of the profiles connected. The procedures used for the
calculation of the margin of safeties are given below:

A.1 The Tension Failure

The failure due to tension can be observed when the tension load on the
fastener exceeds the tension strength of the respective fastener. As a general rule in
aerospace industry the screws are used in high tension zones and the rivets are used
for the shear loads. The important point to be considered for the screws and similar
fasteners like hi-loks is that; because of the torque applied during the installation of
fastener there exist a pre-tension force which should be taken into account. But,
when the considered fastener is a rivet or the applied tension force is too high that
causes seperation between the connected surfaces, the pre-tension force is not used in
the calculations and only the applied force is used as the total tension force. The
margin of safety for tension load is determined from [14];

Fall ,tension
MSTension = −1 (A.1)
Fapp ,tension × ff

(when fastener type is rivet or applied tension force causes separation between the
connected surfaces although the fastener type is screw or hi-lok)

109
 Fall ,tension Fall ,tension
MSTension = −1 =
(Ftension ) × ff (Fpre−tension + K p × Fapp,tension )× ff − 1 (A.2)

(when fastener type is screw or hi-lok and no separation occurs)

where pre-tension force can be found by;

T × Am
Fpre−tension = (A.3)
β

The bolt tension preload coefficient, Kp value, which is used in the

calculation of the tension force with pre-tension, may yield the total force which is
actually bigger or smaller than the applied force. As applied tension force should be
directly used when the separation occurs, for the conservative results, the larger one
of the force calculated with the pre-tension and applied force is used in the analysis
process.

A.2 The Shear Failure

The shear failure can be observed when the shear load on a fastener exceeds
the shear strength of that fastener. Usually the rivets are used at the high shear
regions although it is also possible to use the screws. The calculation of the margin
of safety for the shear load is performed by [14];

Fall ,shear
MS Shear = −1 (A.4)
Fapp ,shear × ff

A.3 The Combined Shear and Tension Failure

Since having pure shear or pure tension load on a fastener is a rare case, the
effects of having combined shear and tension load has to be checked also. For
performing this check the following formula is used [14];

110
MS Combined = 1 − ((R
Tension )2 + (RShear )2 ) × ff (A.5)

where [14];

Fapp ,tension Fapp ,shear

RTension = and RShear = (A.6)
Fall ,tension Fall ,tshear

A.4 The Bearing Stress Failure

Shear force on the fasteners may cause bearing on the connected profiles.
Bearing is related with the pressure created on the touching area between the fastener
and profile because of the shear force on the fastener. The area described is assumed
to be a rectangle having edges as the diameter of fastener, D, and the thickness of the
profile, t. Therefore margin of safety for bearing stress can be found by [14];

σ all ,bearing σ all ,bearing

MS Bearing = −1 = −1 (A.7)
σ app ,bearing × ff Fapp ,shear
× ff
D×t

A.5 The Pull-through Stress Failure

When the tension force on the fastener is relatively high, it is possible that
one of the connected profiles torn around the fastener hole and fastener pulls
through, hence causing a separation between the mating surfaces. As mentioned
before the pre-tension force is only used in the calculations when the used fastener
type is screw or hi-lok and the separation between the connected surfaces do not
occur. The margin of the safety for the prescribed pull-through stress can be found as
[14];

111
 σ all ,shear σ all , shear
MS Pull =through = −1 = −1 (A.8)
σ app , pull =through × ff Fapp ,tension
× ff
0.6 × π × Dwasher × t
(when fastener type is rivet or applied tension force causes separation between the
connected surfaces although the fastener type is screw or hi-lok)

σ all , shear σ all , shear

MS Pull =through = −1 = − 1 (A.9)
σ app , pull =through × ff (F pre − tension + K p × Fapp ,tension )
× ff
0.6 × π × Dwasher × t
(when fastener type is screw or hi-lok and no separation occurs)

where the pre-tension force can be found by;

T × Am
Fpre−tension = (A.3)
β

In the calculation of margin of safety of pull-through stress it is possible to

use the head diameter of the fastener instead of the washer diameter as usually head
diameter is smaller. Using the head diameter of the fastener yields more conservative
results.

112
 APPENDIX B

USAGE OF DEVELOPED FASTENER TOOLS

For the application of fastener failure checks to all of the fasteners in the
finite element model, two computer tools were developed as being a FORTRAN
based executable file and a Microsoft Office Excel file. The executable file finds and
lists the maximum tension and shear forces and the connected material informations
for all of the fasteners, modeled with rigid multi-point constraint elements, between
two profiles. Since this is a time consuming process and especially when fasteners’
tension directions do not coincide with any of the global axes the manual application
becomes very tedious, the developed tool provides an opportunity in time saving
which increases by the increasing number of fasteners. The Microsoft Office Excel
file on the other hand, performs the fastener failure analyses, which are listed in
Appendix A, by using the output file of executable file. This tool automitizes the
margin of safety calculations and is found to be more beneficial with increasing
fastener number. These tools are detailed in the following sections.

B.1 The FORTRAN Based Executable Tool

This developed tool helps user in finding the required data for the failure
checks. For the executation of this file two input files are required. First input file is
the MSC.Nastran® input file, which is the output file of MSC.Patran® and has an
extension of ”bdf”, of the prepared finite element model. The second file required is
the multi-point constraint force report, obtained from MSC.Patran® after the
completion of MSC.Nastran® analysis. The user should be aware of the fact that the
addition of Multi-Point Constraint Forces or Grid Point Force Balance options is
needed for obtaining the required report file. For the addition of these options,

113
“Subcases” window should be opened by using the button called “Subcases” in the
“Analysis” tab of MSC.Patran®. In the newly opened window, load case which will
be used in the analysis should be selected and “Output Request” window should be
opened. For the addition of the required outputs clicking on the desired output’s
name is sufficient. This process is presented in Figure B.1.1.

Figure B.1.1 Addition of Multi-Point Constraint Forces and Grid Point Force
Balance Option into Analysis

After the completition of the finite element analysis, the user should print the
report file containing multi-point constraint forces for the preferred load cases. This
report can be written by using the “Report” option in “Results” tab of MSC.Patran®.
When preferred load cases and results to be reported are selected in the first pane,
report will be written in a file whose name can be changed in the third pane, as
shown in Figure B.1.2.

114
 Figure B.1.2 Reporting the Multi-Point Constraint Forces

When these input files are provided the executable file, fastener.exe, can be run
and following steps can be followed:

• Firstly the name of the MSC.Nastran® input file should be inputted to

execution window with the extension corresponding to file type (ex:
example.bdf). This step is shown in Figure B.1.3 for an example case in
which MSC.Nastran® input file name is “rack.bdf”.

Figure B.1.3 Input of MSC.Nastran® Input File Name in Execution Window

115
• Secondly name of the multi-point constraint force report file should be
inputted with the extension corresponding to file type (ex: example.rpt). Input
of the report file “fast_force.rpt” is shown in Figure B.1.4 as an example.

Figure B.1.4 Input of Multi Point Constraint Force Report Name in Execution
Window

• At this step the program presents a choice to the user. The user can prefer
obtaining an output file which only contains maximum tension and shear
forces on the fasteners and corresponding load cases or another output file
which is prepared as the input for the Microsoft Office Excel file which is the
developed tool for the failure analyses. As it is explained in the execution
window integer of 1 should be inputted, as shown in Figure B.1.5, to obtain
input file for Microsoft Office Excel Tool

116
 Figure B.1.5 Choice of Contents of Output File

• At the fourth step, program requests the number of load cases reported in the
multi-point constraint force report file. Figure B.1.6 shows the example in
which the number of load cases reported is 3.

117
 Figure B.1.6 Number of Load Cases to be Analyzed

• The fifth step includes the question of fitting factor. As the fitting factor
application may change throughout the finite element model, the program
asks if it is same for the whole structure or not. Figure B.1.7 shows this step
for the case in which fitting factor is same for all of the parts in the structure.

118
 Figure B.1.7 Fitting Factor Application for Whole System

• The sixth step is followed if the fitting factor application is same for the
whole structure. In this step, the tool requires the fitting factor values for each
reported subcase. It is obvious that, if the fitting factor application changes
for each fastener, calculation of margin of safety values should be performed
manually. Alternatively, when it is the case program prepares the output file
without fitting factor information and without asking any more input, the
fitting factors can be inputted by using the Microsoft Office Excel Tool.
Figure B.1.8 shows an example case in which the fitting factor values are
same for 3 load cases.

119
 Figure B.1.8 Input of Fitting Factor Values for Each Load Case

After the execution the results can be obtained from the file “output.txt”. If
only the fastener forces are requested, the columns in the file correspond to multi-
point constraint element ID, maximum tension force on the fastener, corresponding
load case, maximum shear force on the fastener and corresponding load case
respectively. An example of this output format is given in Figure B.1.9.

120
 Figure B.1.9 Output File Format Showing Only the Maximum Forces

Figure B.1.9 shows an example of the case in which only the maximum
tension and shear forces for each fastener were required. The columns represent the
multi-point constraint element ID, maximum tension force on the fastener,
corresponding load case, maximum shear force on the fastener and corresponding
load case respectively.
If Microsoft Office Excel Tool input file is requested, these columns change
to multi-point constraint element ID, first profile’s material and thickness, second
profile’s material and thickness, maximum tension force on the fastener and
corresponding load case, maximum shear force on the fastener and corresponding
load case and fitting factors corresponding to maximum tension and maximum shear
load cases respectively. An example of this output file is shown in Figure B.1.10.
This arrangement can be seen better when the file is imported in to Microsoft Office
Excel Tool.

121
 Figure B.1.10 Output File Format Prepared for the Input of Microsoft Office Excel
Tool

The output data format shown in Figure B.1.10 is the format prepared for the
input of Microsoft Office Excel Tool. The columns represent the multi-point
constraint element ID, first profile’s material and thickness, second profile’s material
and thickness, maximum tension force on the fastener and corresponding load case,
maximum shear force on the fastener and corresponding load case and fitting factors
corresponding to maximum tension and maximum shear load cases respectively.
In order to verify the results of the developed tools, two fasteners whose
tension direction do not coincide with any global axis were chosen and results
obtained from the developed tool and MSC.Patran® were compared. Figure B.1.11
shows the input steps and output of tool where lines corresponding to chosen
fasteners (multi-point constraint elements 1324 and 1325) were highlighted. In the
output section LC stands for the load case and first load case corresponds to forward
acceleration case.

122
 Figure B.1.11 FORTRAN Based Program and Output

As it can be seen in Figure B.1.11, for the element number 1324 the
maximum tension and shear forces were obtained as 213.07 N and 2263.05 N
respectively, in forward acceleration case. For the element number 1325 these values
were 108.97 N and 298.23 N respectively. To obtain tension and shear forces from
MSC.Patran® a new coordinate system was formed. In this coordinate system, the
tension direction of the fasteners was defined as an axis. The force components along
this tension direction was presented in Figure B.1.12. These force components
correspond to tension on the fasteners. For the shear forces, the resultant of the force
components on the remaining axes were plotted and the obtained figure is given in
Figure B.1.13.

123
 Figure B.1.12 Maximum Tension Forces on Fasteners 1324 and 1325

Figure B.1.12 shows the tension forces on multi-point constraint elements

1324 and 1325, which were labeled with black. The tension forces on these fasteners
were shown with vectors and as it can be seen in the figure, tension forces obtained
from MSC.Patran® were nearly same with the results obtained from developed tool’s
output.

124
 Figure B.1.13 Maximum Shear Forces on Fasteners 1324 and 1325

Figure B.1.13 shows the shear forces on multi-point constraint elements 1324
and 1325 which were labeled with black. The shear forces on these elements were
obtained by combining the force component on the two remaining axes when tension
direction excluded. As it can be seen in the figure, shear forces obtained from
MSC.Patran® were nearly same with the results obtained from developed tool’s
output.
As both the shear and tension forces obtained from MSC.Patran® and
FORTRAN Based Executable Tool were nearly same, it was concluded that the
developed tool yields the correct results. So, for the calculation of margin of safety
values for each fastener, output obtained from FORTRAN Based Executable Tool
can be used as input of Microsoft Office Excel Tool.

125
B.2 The Microsoft Office Excel Tool

This developed tool automizes the margin of safety result calculations of tension,
shear, combined shear and tension, bearing and pull-through failure cases for all of
the fasteners modeled with rigid multi-point constraint elements. The usage of this
tool requires the previous run of the FORTRAN Based Executable Tool which
prepares the input file format for Microsoft Office Excel Tool.

• For the input of the “output.txt” file which is to be prepared by FORTRAN

Based Executable Tool, this file should be in same folder with Microsoft
Office Excel Tool and the button on the sheet called “Input Data” should be
used. This button runs an application which automates the input procedure. If
the fitting factors are not same for the whole structure, and so, are not listed
in the input file, user should manually input these values for each fastner. The
inputted data can be seen in the sheet called “Formatted Input” in the required
format, as shown in Figure B.2.1. This figure shows an example case in
which fitting factors are same for whole structure.

Figure B.2.1 Data Inputted in Formatted Input Sheet

126
• Second step is the filling of “MPC-Fast Match” sheet. This sheet requires the
input of fasteners corresponding to multi-point constraint elements. Although
it is enough to input the fastener name for the rivet type elements; for the
screw type elements, the grip lenghts should also be inputted. The idea
behind the requirement of a grip length is that the Kp values, which are the
bolt tension preload coefficients, are determined by the grip lengths.
Although these values depend on the distance between bolt head and nut and
the distance between the mid-points of connected profiles, for the simplicity,
they are assumed to be a function of grip length. It is also possible for the
user to input manually the obtained Kp values by assigning a number to each
value and using these numbers in the grip length column. It is highly
recommended that for the structures, whose fastener pattern is determined in
the preliminary design phase, this sheet to be filled during the preparation of
finite element model. An example of data to be filled in “MPC-Fast Match”
sheet is shown in Figure B.2.2.

Figure B.2.2 Fasteners Corresponding to Multi-Point Constraint Elements

127
• Third step is the input of required fastener properties in the “Fasteners” sheet.
Fastener name, fastener type (screw or rivet type), tensile and shear strength
limits, the nominal and head diameters and pre-tension forces for the screw
type fasteners are to be input for all of the used fasteners in the designed
structure. This sheet also includes the bolt tension preload coefficient section,
which is to be filled according to the user’s preference as mentioned in
predecessor step. Figure B.2.3 shows an example of data to be inputted in
“Fasteners” sheet.

Figure B.2.3 Fastener Properties Inputted

• “Materials” sheet requires the input of the required mechanical properties of

the utilized materials. For the correct run of this tool it is essential to have
material names used in MSC.Patran® same with the material names in this
sheet. The importance of this comes from the fact that Microsoft Office Excel
Tool matches the materials listed in these sheet, with the data obtained from
FORTRAN Based Executable Tool, in the analysis process. Also, since the
FORTRAN Based Executable Tool uses the data from MSC.Patran® output
file, material names should be same. Figure B.2.4 shows material properties
inputted for 5 different materials.

128
 Figure B.2.4 Material Properties Inputted

• The next step in the analysis is the operation of the analysis macro. The
button which is located in the “Analysis” sheet runs the developed macro for
the margin of safety calculation process, which is to be performed using the
formulas presented in Appendix A. When the this process is completed, the
inputted data and margin of safety results for tension, shear, combined shear
and tension, bearing and pull-through failure cases for each fastener are to be
listed in the “Results” sheet. Figure B.2.5 shows the results of the analysis
process carried out by using the data presented in the previous steps.

129
Figure B.2.5 Fastener Analysis Results

130