Modelling and Optimization of Laser Cut

Minia University
Faculty of Engineering
Production Eng. & Design Department
“Modelling and Optimization of Laser Cutting

Operations”
Thesis
Submitted in Partial Fulfillment for M.Sc. Degree in

Mechanical Engineering
By
Hany Mohamed Abdu Mohamed Elsayed
B.Sc. Production Engineering and Design
Minia University
Supervisors:
Prof.Dr.Eng. Mohamed Hassan Prof.Dr.Eng. El-Giushi Mokhtar

Gadallah Ali
Professor of Industrial Engineering Professor of Production Engineering and
Mechanical Design & Production Mechanical Design
Department, Faculty of Engineering, Faculty of Engineering
Cairo University Minia University
Prof. Dr.Eng. Yehia Mahmoud Ismail

Professor of Production Engineering and
Mechanical Design
Minia University
2015
Minia University
Production Eng. & Design Department
“Modelling and Optimization of Laser Cutting

Operations”
Thesis
Submitted in Partial Fulfillment for M.Sc. Degree in
Mechanical Engineering
By
Hany Mohamed Abdu Mohamed Elsayed
B.Sc. Production Engineering and Design
Minia University
Supervisors:
Prof.Dr.Eng. Mohamed Hassan Prof.Dr.Eng. El-Giushi Mokhtar

Gadallah Ali
Professor of Industrial Engineering Professor of Production Engineering and
Mechanical Design & Production Mechanical Design
Department, Faculty of Engineering, Faculty of Engineering
Cairo University Minia University
Prof. Dr.Eng. Yehia Mahmoud Ismail

Professor of Production Engineering and
Mechanical Design
Minia University
2015
ACKNOLEDGEMENT
First and foremost, grateful to Allah SWT for making it possible for me to
complete this thesis on time.
I wish to express my deepest gratitude to our supervisor Prof. Dr. Mohamed
Hassan Gadallah , Prof.Dr.El-Giushi Mokhtar Ali and Prof.Dr.Yehia
Mahmoud Ismail for their abundantly helpful and offered invaluable
assistance, support, and guidance.
My sincere gratitude and extreme thanks to all who have given hand in the
Faculty of Engineering, Minia University, especially all my Lab mate and
members of the staff of Mechanical Engineering Department, who helped me
in many ways.
My deepest gratitude goes toAss. Prof. Dr. Hammed Mahmoud Abo El Enin
for giving me their valuable time to guide me throughout my higher education.
I would also thank the Research Center in Helwan Tebeen – Cairo for their
valuable feedbacks helped me to using laser machine.
Finally, I would like to express special thanks and appreciation to my family
for their patience, continuous encouragement, and great help through the
years of this study.
Eng. Hany Mohamed Abdu Mohamed
2015
v
Contents
CONTENTS
Page
ACKNOWLEDGMENTS------------------------------------------------------ i
ABSTRACT----------------------------------------------------------------------- ii
CONTENTS----------------------------------------------------------------------- iii
LIST OF TABLES-------------------------------------------------------------- iv
LIST OF FIGURES------------------------------------------------------------- x
NOMENCLATURE------------------------------------------------------------- xv
CHAPTER 1: INTRODUCTION------------------------------------------- 1
1.1. Introduction ------------------------------------------------------------------- 1
1.2.Taguchi Approach------------------------------------------------------------- 1
1.3. Response Surface Methodology (RSM)----------------------------------- 5
1.4. Brief Introduction of Laser Beam machining (LBM)------------------- 6
1.4.1. Features of Laser Beam--------------------------------------------- 6
1.4.2 .Laser Beam Cutting-------------------------------------------------- 7
1.4.3. Types of laser Cutting------------------------------------------------ 8
1.4.4 .Type of Lasers--------------------------------------------------------- 9
1.4.5. Advantages and Limitations----------------------------------------- 11
1.5. Stainless Steel----------------------------------------------------------------- 13
1.5.1. Applications for Stainless Steels------------------------------------ 16
1.5.2. 316 L stainless steel Applications---------------------------------- 17
1.6. Problem Statement----------------------------------------------------------- 17
1.7. Thesis organization----------------------------------------------------------- 18
CHAPTER 2 : LITERATURE REVIEW---------------------------------- 19
2.1. Introduction------------------------------------------------------------------- 19
2.2.Taguchi Approach------------------------------------------------------------- 19
2.2.1. Description of Taguchi Method and Quality Engineering------ 20
2.2.2 .Orthogonal Arrays ------------------------------------------ -------- 21
2.3. Response Surface Methodology (RSM) ---------------------------------- 23
2.4. Laser Beam Machining (LBM)--------------------------------------------- 26
2.5. Modelling and Optimization of LBM Process Parameters------------ 27
2.6. Laser Cutting Parameters---------------------------------------------------- 31
2.7. Stainless Steel----------------------------------------------------------------- 32
CHAPTER 3: EXPERIMENTAL WORK -------------------------------- 35
3.1. Introduction------------------------------------------------------------------- 35
3.2. Experimental Investigations------------------------------------------------ 35
3.3. Selection of Measured Responses and Control Variables-------------- 36
3.4. Plan of Experimentation----------------------------------------------------- 36
3.5. Material and Specimen Shape---------------------------------------------- 40
3.6. LBM Equipment-------------------------------------------------------------- 41
i
Contents
3.7. Laser Cut Quality Characteristics------------------------------------------ 44

3.7.1 .Cut kerf width--------------------------------------------------------- 45
3.7.2. Cut edge surface roughness----------------------------------------- 45
3.7.3. Heat affected zone--------------------------------------------------- 46
3.8. Measurement of the Process parameters---------------------------------- 46
3.8.1. Measurement of the kerf width (Ta)-------------------------------- 47
3.8.2 .Measurement of the surface roughness (Ra)---------------------- 48
3.8.3 Measurement of the heat affected zone (HAZ)-------------------- 50
CHAPTER 4: MODELLING AND OPTIMIZATION OF LBM
PROCESS ------------------------------------------------------------------------- 51
4.1. Introduction------------------------------------------------------------------- 51
4.2.Analysis of Experimental Results------------------------------------------ 51
4.2.1.Experimental results analysis----------------------------------------- 51
4.2.2.Verification of Experimental Results-------------------------------- 59
4.3. Mathematical Models using Response Surface Methodology (RSM) 61
4.3.1. Motivation------------------------------------------------------------- 62
4.3.2. Mathematical Models for the kerf taper (Ta)--------------------- 63
4.3.3. Mathematical models for Average Surface Roughness (Ra)---- 78
4.3.4. Mathematical models for the heat affected zone (HAZ) -------- 93
CHAPTER 5: CONFIRMATION EXPERIMENTS---------------------- 107
5.1 Introduction-------------------------------------------------------------------- 107
5.2 Experiments-------------------------------------------------------------------- 107
5.3 Conclusion --------------------------------------------------------------------- 108
CHAPTER 6: DISCUSSION AND ANALYSIS OF RESULTS ------- 113
6.1. Analysis of variance---------------------------------------------------------- 113
6.1.1. Analysis of variance (ANOVA) based on S/N ratio
indicates several results--------------------------------------------- 113
6.1.2. Analysis of mean (ANOM) based on mean indicates several
results ----------------------------------------------------------------- 114
6.1.3. Analysis of variance (ANOVA) based on standard deviation
indicates several results---------------------------------------------- 115
6.2.Selection of optimum level-------------------------------------------------- 116
6.2.1 The effect of different operating parameters on S/N (Ta
and Ra) ----------------------------------------------------------------- 116
6.2.2 The effect of different operating parameters on Mean (Ta
and Ra) ----------------------------------------------------------------- 118
6.3 Discussion of Validation----------------------------------------------------- 119
CHAPTER 7: CONCLUSIONS AND RECOMMENDATIONS
FOR FUTURE RESEARCH-------------------------------------------------- 113
7.1.Conclusions-------------------------------------------------------------------- 121
7.2. Future Work------------------------------------------------------------------- 122
REFERENCES------------------------------------------------------------------- 123
ii
Contents
APPENDIX A: TWO-LEVEL AND THREE LEVEL

ORTHOGONAL ARRAYS---------------------------------------------------- 127
APPENDIX B : ANOM RESULTS------------------------------------------ 131
APPENDIX C: ANOVA RESULTS---------------------------------------- 133
APPENDIX D: PLOTS OF CONTROL FACTORS EFFECTS ------ 137
APPENDIX E: SUMMARY OF CONTROL FACTORS EFFECTS- 142
ARABIC SUMMARY
iii
List of figures
LIST OF FIGURES
Figure No. Description Page

Figure 1.1 Process diagram with input-output responses------------------------ 2
Figure 1.2 Taguchi robust design matrix------------------------------------------- 3
Figure 1.3 Taguchi Loss Function -------------------------------------------------- 3
Figure 1.4 Taguchi design procedure----------------------------------------------- 4
Figure 1.5 Response Surface plot--------------------------------------------------- 6
Figure 1.6 Laser beam cutting process--------------------------------------------- 7
Figure 1.7 Terms related to the cutting process of the work piece------------- 8
Figure 1.8 Schematic of Nd:YAG laser beam machine-------------------------- 10
Figure 1.9 Principle of a CO2 laser------------------------------------------------- 11
Figure 1.10 Available stainless steel alloy systems-------------------------------- 14
Figure 2.1 A schematic illustration of the laser cutting head and work piece- 31
Figure 2.2 Laser cutting parameters------------------------------------------------ 32
Figure 3.1 Linear Graph L27OA----------------------------------------------------- 37
Figure 3.2 Dimensions of Specimen------------------------------------------------ 40
Figure 3.3 Specimen Shape---------------------------------------------------------- 40
Figure 3.4 The general view of the laser cutting machine----------------------- 42
Figure 3.5 Block diagram of laser cutting system-------------------------------- 43
Figure 3.6 Layout of the LBM Equipment----------------------------------------- 43
Figure 3.7 Rofin Sinar laser cutting machine used in the present work------- 44
Figure 3.8 shows the laser supply assist gas-------------------------------------- 44
Figure 3.9 The schematic representation of laser cut kerf---------------------- 45
Figure 3.10 the cut edge surface roughness----------------------------------------- 46
Figure 3.11 General view of the tool-maker microscope------------------------- 47
Figure 3.12 Plot of average surface roughness (Ra)------------------------------- 49
Figure 3.13 General view of the TAYLOR–HOBSON instrument------------- 49
Figure 3.14 Method of measurement of HAZ-------------------------------------- 50
Figure 4.1 Measured Vs. Predicted S/N ratio response (Ta)-------------------- 63
Figure 4.2 Effect of power and pressure on Ta at constant frequancy and
cutting speed based on S/N ratio-------------------------------------- 65
Figure 4.3 Effect of power and frequancy on Ta at constant pressure and
Figure 4.4 Effect of power and cutting speed on Ta at constant pressure and
frequency based on S/N ratio------------------------------------- 66
Figure 4.5 Effect of pressure and frequency on Ta at constant power and
Figure 4.6 Effect of pressure and cutting speed on Ta at constant power
and frequency based on S/N ratio------------------------------------- 67
Figure 4.7 Effect of frequency and cutting speed on Ta at constant power
and pressure based on S/N ratio--------------------------------------- 67
Figure 4.8 Measured Vs. Predicted standard deviation response (Ta)-------- 68
X
List of figure captions
Figure 4.9 Effect of power and pressure on Ta at constant frequency and

cutting speed based on standard deviation--------------------------- 70
Figure 4.10 Effect of power and frequency on Ta at constant pressure and
Figure 4.11 Effect of power and cutting speed on Ta at constant pressure and
frequency based on standard deviation-------------------------- 71
Figure 4.12 Effect of pressure and frequency on Ta at constant power and
Figure 4.13 Effect of pressure and cutting speed on Ta at constant power and
frequency based on standard deviation-------------------------- 72
Figure 4.14 Effect of frequency and cutting speed on Ta at constant power
and pressure based on standard deviation---------------------------- 72
Figure 4.15 Measured Vs. Predicted Mean response (Ta)------------------------ 73
Figure 4.16 Effect power and pressure on Ta at constant cutting speed based
on mean------------------------------------------------------------------- 75
Figure 4.17 Effect power and frequency on Ta at constant pressure and
cutting speed based on mean------------------------------------------- 75
Figure 4.18 Effect power and cutting speed on Ta at constant pressure and
frequency based on mean---------------------------------------------- 76
Figure 4.19 Effect pressure and frequency on Ta at constant power and
cutting speed based on mean------------------------------------------ 76
Figure 4.20 Effect pressure and cutting speed on Ta at constant power and
frequency based on mean----------------------------------------------- 77
Figure 4.21 Effect frequency and cutting speed on Ta at constant power and
pressure based on mean-------------------------------------------- 77
Figure 4.22 Measured Vs. Predicted S/N ratio response (Ra)------------------- 78
Figure 4.23 Effect of power and pressure on theaverage surface roughness at
constant frequency and cutting speed of cut based on S/N ratio-- 80
Figure 4.24 Effect of power and frequency on the average surface roughness
at constant pressure and cutting speed of cut based on S/N ratio- 80
Figure 4.25 Effect of power and cutting speed on the average surface
roughness at constant pressure and frequency of cut based on
S/N ratio------------------------------------------------------------------ 81
Figure 4.26 Effect of pressure and frequency on the average surface
roughness at constant power and cutting speed of cut based on
S/N ratio------------------------------------------------------------------ 81
Figure 4.27 Effect of pressure and cutting speed on the average surface
roughness at constant power and frequency of cut based on S/N
ratio------------------------------------------------------------------------ 82
Figure 4.28 Effect of frequency and cutting speed on the average surface
roughness at constant power and pressure of cut based on S/N
ratio------------------------------------------------------------------------ 82
Figure 4.29 Measured Vs. Predicted standard deviation response (Ra)-------- 83
XI
Figure 4.30 Effect of power and pressure on the average surface roughness at
constant frequency and cutting speed of cut based on standard
deviation------------------------------------------------------------------ 85
at constant presure and cutting speed of cut based on standard
deviation------------------------------------------------------------------ 85
roughness at constant presure and frequency of cut based on
standard deviation------------------------------------------------------- 86
roughness at constant power and cutting speed of cut based on
standard deviation. ------------------------------------------------------ 86
roughness at constant power and frequency of cut based on
roughness at constant power and pressure of cut based on
Figure 4.36 Measured and predicted Mean response (Ra)----------------------- 88
Figure 4.37 Effect of power and pressure on the average surface roughness at
constant frequency and cutting speed based on mean----------- 90
at constant pressure and cutting speed based on mean------------- 90
roughness at constant pressure and frequency based on mean---- 91
roughness at constant power and cutting speed based on mean--- 91
roughness at constant power and frequency based on mean------- 92
roughness at constant power and pressure based on mean--------- 92
Figure 4.43 Measured Vs. Predicted S/N ratio response (HAZ)----------------- 93
Figure 4.44 Effect of power and pressure on the heat affected zone (HAZ) at
constant frequency and cutting speed of cut based on S/N ratio-- 95
Figure 4.45 Effect of power and frequency on the heat affected zone (HAZ)
at constant pressure and cutting speed of cut based on S/N ratio- 95
Figure 4.46 Effect of power and cutting speed on the heat affected zone
(HAZ) at constant pressure and frequency of cut based on S/N
ratio------------------------------------------------------------------------- 96
Figure 4.47 Effect of pressure and frequency on the heat affected zone
(HAZ) at constant power and cutting speed of cut based on S/N
ratio------------------------------------------------------------------------- 96
XII
Figure 4.48 Effect of pressure and cutting speed on the heat affected zone
(HAZ) at constant power and frequency of cut based on S/N
ratio------------------------------------------------------------------------- 97
Figure 4.49 Effect of frequency and cutting speed on the heat affected zone
(HAZ) at constant power and pressure of cut based on S/N ratio- 97
Figure 4.50 Measured Vs. Predicted standard deviation response (HAZ)------ 98
Figure 4.51 Effect of power and pressure on the heat affected zone at
constant frequency and cutting speed of cut based on standard
deviation------------------------------------------------------------------- 100
Figure 4.52 Effect of power and frequency on the heat affected zone at
constant presure and cutting speed of cut based on standard
deviation------------------------------------------------------------------- 100
Figure 4.53 Effect of pressure and frequency on the heat affected zone at
constant power and cutting speed of cut based on standard
deviation------------------------------------------------------------------- 101
Figure 4.54 Measured Vs. Predicted Mean response (HAZ)--------------------- 102
Figure 4.55 Effect of power and pressure on the heat affected zone at
constant frequency and cutting speed based on mean--------------- 104
Figure 4.56 Effect of power and frequency on the heat affected zone at
constant pressure and cutting speed based on mean----------------- 104
Figure 4.57 Effect of power and cutting speed on the heat affected zone at
constant pressure and frequency based on mean--------------------- 105
Figure 4.58 Effect of pressure and frequency on the heat affected zone at
constant power and cutting speed based on mean------------------- 105
Figure 4.59 Effect of pressure and cutting speed on the heat affected zone at
constant power and frequency based on mean----------------------- 106
Figure 4.60 Effect of frequency and cutting speed on the heat affected zone
at constant power and pressure based on mean---------------------- 106
Figure 5.1 Plot for comparison of Experimental and Predicted results for
kerf taper------------------------------------------------------------------ 111
surface roughness-------------------------------------------------------- 112
heat affected zone-------------------------------------------------------- 112
Figure D.1 Plot of control factors effects (S/N ratios) for Ta-------------------- 137
Figure D.2 Plot of Control factors effects (Standard deviation values) for Ta 137
Figure D.3 Plot of control factors effects (Mean values) for Ta----------------- 138
Figure D.4 Plot of control factors effects (S/N ratios) for Ra-------------------- 138
Figure D.5 Plot of Control factors effects (Standard deviation values) for Ra 139
Figure D.6 Plot of control factors effects (Mean values) for Ra----------------- 139
Figure D.7 Plot of control factors effects (S/N ratios) for HAZ----------------- 140
Figure D.8 Plot of Control factors effects (Standard deviation values) for
HAZ------------------------------------------------------------------------ 140
XIII
Figure D.9 Plot of control factors effects (Mean values) for HAZ-------------- 141
XIV
List of tables
LIST OF TABLES
Table No. Description Page

Table 1.1 Application for stainless steel products------------------------------- 17
Table 2.1 3
Orthogonal array OA4 (2 ) --------------------------------------------- 21
Table 2.2 7
Orthogonal array OA8 (2 ) --------------------------------------------- 22
Table 2.3 1 6
Orthogonal arrays: OA18 (6 x3 ) and OA18--------------------------- 22
Table 2.4 Percentage chemical composition (wt.%) of stainless steel 316L- 33
Table 2.5 Mechanical properties of stainless steel 316 L----------------------- 34
Table 3.1 Assignment of levels to control factors------------------------------- 37
Table 3.2 L27 OA Orthogonal array------------------------------------------------ 38
Table.3.3 L27OA used to plan experimentation---------------------------------- 39
Table.3.4 Top& bottom kerf widths and Ta-------------------------------------- 48
Table.4.1 Experimental results of kerf taper (Ta) and Average Surface
Roughness (Ra) --------------------------------------------------------- 52
Table.4.2 Summary of S/N ratios using L27OA--------------------------------- 53
Table.4.3 Summary of mean values using L27OA------------------------------- 54
Table.4.4 Summary of Standard deviation values using L27OA-------------- 55
Table.4.5 ANOVA results for the kerf taper (Ta) based on S/N ratios------- 57
Table.4.6 ANOM for The kerf taper results-------------------------------------- 58
Table.4.7 ANOVA results for the kerf taper (Ta) based on Standard
deviation------------------------------------------------------------------- 58
Table.4.8 Results of the confirmation experiment for S/N ratios values ---- 59
Table.4.9 Results of the confirmation experiment for mean values----------- 60
Table.4.10 Measured versus predicted S/N ratio response for the kerf taper
(Ta) (Model Validation)------------------------------------------------ 64
Table.4.11 Measured versus predicted Standard Deviation response for the
kerf taper (Ta) (Model Validation)------------------------------------ 69
Table.4.12 Measured versus predicted Mean response for the kerf taper (Ta)
(Model Validation)------------------------------------------------------ 74
Table.4.13 Measured versus predicted S/N ratio response for the average
surface roughness (Ra) (Model Validation)-------------------------- 79
Table.4.14 Measured versus predicted standard deviation response for the
average surface roughness (Ra) (Model Validation)---------------- 84
Table.4.15 Measured versus predicted mean response for the average
surface roughness (Ra) (Model Validation)-------------------------- 89
Table.4.16 Measured versus predicted S/N ratio response for the average
surface roughness (HAZ) (Model Validation)----------------------- 94
Table.4.17 Measured versus predicted standard deviation response for the
average surface roughness (HAZ) (Model Validation)------------- 99
Table.4.18 Measured versus predicted mean response for the average
surface roughness (HAZ) (Model Validation)----------------------- 102
Table 5.1 Confirmation Experiment ---------------------------------------------- 109
iv
Contents
Table 5.2 The confirmation and predict results of the Average Surface
Roughness (Ra)---------------------------------------------------------- 110
Table 5.3 The confirmation and predict results of the kerf taper (Ta)-------- 111
Table 5.4 The confirmation and predict results of the heat affected zone
(HAZ)---------------------------------------------------------------------- 112
Table.6.1 Effect of factor on S/N (Ta)-------------------------------------------- 116
Table.6.2 Effect of factor on S/N (Ra)-------------------------------------------- 117
Table.6.3 Effect of factor on S/N (HAZ)----------------------------------------- 117
Table.6.4 Effect of factor on Mean (Ra)------------------------------------------ 118
Table.6.5 Effect of factor on Mean (Ta)------------------------------------------ 118
Table.6.6 Effect of factor on Mean (HAZ)--------------------------------------- 119
Table.6.7 Comparison of experimental and predicted (Ta& Ra &HAZ)
based on S/N ratio------------------------------------------------------- 120
Table.6.8 Comparison of experimental and predicted (Ta& Ra &HAZ)
based on Mean------------------------------------------------------------ 120
Table.A.1 L4 Array (23) ------------------------------------------------------------- 127
Table.A.2 L8 Array (27)-------------------------------------------------------------- 127
Table.A.3 L12 Array (211) ----------------------------------------------------------- 128
Table.A.4 L16 Array (215) ----------------------------------------------------------- 128
Table.A.5 L9 Array (34) ------------------------------------------------------------- 129
Table.A.6 L27 Array (313) ------------------------------------------------------------ 130
Table.B.1 ANOM for Average surface roughness (Ra) results---------------- 131
Table.B.2 ANOM for The heat affected zone (HAZ) results------------------- 132
Table.C.1 ANOVA results for the Average surface roughness (Ra) based
on S/N ratios-------------------------------------------------------------- 133
Table.C.2 ANOVA results the heat affected zone (HAZ) based on S/N
ratios----------------------------------------------------------------------- 134
Table.C.3 ANOVA results for the Average surface roughness (Ra) based
on Standard deviation--------------------------------------------------- 135
Table.C.4 ANOVA results for the heat affected zone (HAZ) based on
Standard deviation------------------------------------------------------- 136
Table.E.1 Summary of control Factors effects (S/N ratio and Mean values) 142
vi
Nomenclature
NOMENCLATURE
LBM Laser Beam Machining

RSM Response Surface Methodology
DOE Design of Experiments
Ta Kerf Taper
Ra Average Surface Roughness
HAZ Heat affected zone
S/N Signal to Noise Ratio
OA Orthogonal Array
ANN Artificial Neural Network
ANOVA Analysis of Variance
ANOM Analysis of Mean
SS Statistical Sum
Nd Neodymium
Nd: YAG Neodymium: yttrium-aluminium-garnet
CO2 Carbon dioxide
L27OA 3Levels, 27 experiments OA
X1 Power
X2 Pressure
X3 Frequency
X4 Cutting Speed
xv
ABSTRACT
Taguchi's parameter design is a simple systematic approach to optimize

performance, quality and cost. Response surface methods (RSM) are
analytical techniques employed by process modelers to determine a
cause-effect relationship between different responses and control input
variables. On other side that Laser beam cutting machining (LBM) is a
non-traditional machining process, LBM are widely used for cutting,
drilling, marking, welding, sintering, and heat treatment. Stainless steel
(316 L) have become a popular field of research because Stainless steel is
used extensively in a number of everyday applications in the home,
industry, hospitals, food processing, farming, aerospace, construction,
chemical, electronics, heat exchangers and energy such as electricity
generation plants and is in boilers and containers of fluid; the austenitic
grade of stainless steel is the most used by far.
The objective of this study is to apply Taguchi optimization methodology

to optimize Laser beam cutting machining parameters of Stainless steel
(316L) to achieve optimal Average kerf taper (Ta) and Surface
Roughness (Ra). A series of experiments are conducted using (LBM) to
relate machining parameters to several quality responses. Analysis of
variance (ANOVA), Analysis of mean (ANOM), Orthogonal array
(L27OA) and signal to noise ratio are employed to analyze the influence
of process parameters. The machining parameters are machining on
power (Watt), oxygen pressure (MPa), pulse frequency (Hz) and cutting
speed (Cm/min). Another objective is to build mathematical models for
Average kerf taper and average surface roughness as function of
significant process parameters using Response Surface Methodology.
Experimental results for both S/N ratio and mean response values show
that power, oxygen pressure, and cutting speed are the significant
parameters that influence Kerf taper at confidence levels 99%, 95%, and
90% respectively. On the other hand, power, and oxygen pressure are the
significant parameters that influence average surface roughness at
confidence levels 99%95%, and 90% respectively, consequently both the
power and pressure of oxygen are the criteria that affect the impact of the
heat affected zone at confidence levels 99%95%, and 90% respectively.
RSM models are developed for mean responses, S/N ratio, and standard
deviation of responses. Optimization models are formulated as single
objective problem subject to process constraints. Models are formulated
based on Analysis of Variance (ANOVA) using via optimization toolbox
MATLAB. Optimum solutions are compared with Taguchi Methodology
results.
xix
Chapter One. Introduction
Chapter 1
INTRODUCTION
1.1. Introduction
Laser cutting is nowadays the most significant application of lasers

in materials processing in terms of market share. Some of the metals that
are commonly laser cut in industries such as car production and ship
manufacturing include :low alloy steel, stainless steel and aluminum.
Economical criteria affecting the choice of a suitable laser system for a
particular laser cutting application is now gaining much importance as
manufacturers using laser cutting in their production procedures are
particularly interested in high cutting speeds for maximization of productivity,
attainment of high cut quality so that rework of cut pieces can be
eliminated, and cutting reproducibility. Increased process efficiency, quality,
and flexibility help to reduce costs.
In this chapter, The Taguchi approach is introduced. Classification of

variables into Controllable and uncontrollable variables are also shown.
Response Surface Methodology (RSM), Laser beam cutting machining (LBM)
and workpiece used Stainless steel are discussed.
1.2. Taguchi Approach

Taguchi approach offers a simple and systematic approach to optimize
performance, quality and cost. Signal to noise ratios and orthogonal arrays are
two major tools used in robust design. Signal to noise ratio measures quality
with emphasis on variation, orthogonal arrays, and accommodates many design
factors simultaneously [1, 2]. The motive of robust design is to improve the
quality of a product or process by achieving Performance targets and
minimizing performance variation. Variables are usually classified into
1
 Control Variables X: variables that can be controlled during the

manufacturing process.
 Noise variables Z: variables either not controllable, or too difficult

(expensive) to control in the manufacturing process. Noise variables can
cause variation of responses Y and lead to quality loss.
 Response variables Y: They are performance characteristics, the system

outputs, and are functions of control and noise variables Figure 1.1
shows a process with input variables and output responses.
Figure 1.1: Process diagram with input-output responses [1].
Taguchi's robust design evaluates the mean performance and its variation by
crossing two Arrays: an inner array, designed in the control variables, and an
outer array, designed in the noise variables. Figure 1.2 shows a two level
factorial design for both the inner and outer arrays. For each row of the inner
array, response values are generated for each noise variables combination. For
example, inner array row 1 with outer column 1 leads to the response value
y11, inner row 1 with outer column 2 leads to response value y12, and so on.
This design leads to multiple response values for each combination of control
variables, the response mean )μ(, and variance )σ2( can be obtained [1].
2
Figure 1.2: Taguchi robust design matrix [1].
Given the mean and variance for each inner array row, the experiments can be
compared to determine which set of control settings best achieves mean on
target and minimum variation performance goals. Taguchi uses the signal-to-
noise ratio and quality loss (measured using a loss function) to combine the
effects of mean performance and performance variation. The S/N ratio
calculation depends on the particular response being investigated. The second
performance characteristic used by Taguchi Robust Design Techniques, the
Loss function, is generally used to measure the loss of quality associated with
deviation from a target performance value, as shown in Figure 1.3 [1].
Figure 1.3: Taguchi Loss Function [1].
3
Quality loss is measured by the deviation from the target. This means loss of
quality occurs gradually when the quality characteristic moves in either
direction from the target value, rather than as a sharp cutoff with the
conventional approach.
The standard form of the loss function L(y) is given as follows:
L(y( = k(y − T( 2 (1.1)
Also the S/N ratio characteristics can be divided into three categories given by
as follows:
Nominal is the better characteristic: S/N = -Log (y 2/s2y) (1.2)
Smaller is the better characteristic: S/N = -Log (1/n) (∑y2) (1.3)
Larger is the better characteristic: S/N = -Log (1/n) (∑1/y2) (1.4)
Where y is the average of observed data, s2y is the variance of y, n the number
of observations, and y the observed data. For each transformation, the higher
the S/N ratio the better. Steps of Taguchi parameter design are shown as
following Figure 1.4:
Figure 1.4: Taguchi design procedure [2].
4
1.3. Response Surface Methodology (RSM)

Response surface methodology (RSM) is a collection of mathematical
and statistical techniques for empirical model building. By careful design of
experiments, the objective is to optimize a response (output variable) which
is influenced by several independent variables (input variables). An
experiment is a series of tests, called runs, in which changes are made in
the input variables in order to identify the reasons for changes in the output
response.
The response of interest is influenced by several variables and the objective is
to optimize this response [5].
RSM is an important branch of experimental design. The objectives of quality
improvement, including reduction of variability and improved process and
product performance, can often be accomplished directly using RSM. An
example for response surface plot is shown in Figure 1.5. It is well known that
variation in key performance characteristics can result in poor process and
product quality. During the 1980s, considerable attention was given to process
quality, and a methodology is developed using experimental design. Attention
is given to:
 Designing and developing products and processes robust to component
variations.
 Minimizing variability in the output response of product or process target
value.
 Designing products and processes that are robust to environment
conditions. Robust means that the product or process performs
consistently on target and is relatively insensitive to factors that are
difficult to control.
5
Figure 1.5: Response Surface plot.
 Find factor settings (operating conditions( that produce the “best”

Response surface methods may be employed to:
response.
 Find factor settings that satisfies operating or process specifications.
 Identify new operating conditions that produce demonstrated
improvement in product quality over the quality achieved by current
conditions.
 Model a relationship between the quantitative factors and the response.
1.4. Brief Introduction of Laser Beam machining (LBM)
1.4.1. Features of Laser Beam
Laser, Light Amplification by Stimulated Emission of Radiation, is a

high-energy beam of electromagnetic radiation .Laser Beam Machining or more
broadly laser material processing deals with machining and material processing
like heat treatment, alloying, cladding, sheet metal bending etc. Such
processing is carried out utilizing the energy of coherent photons or laser beam,
6
which is mostly converted into thermal energy upon interaction with most of
the materials. Nowadays, laser is also finding application in regenerative
machining or rapid prototyping as in processes like stereo-lithography, selective
laser sintering etc.
Laser beam can very easily be focused using optical lenses as their wavelength
ranges from half micron to around 70 microns. Focused laser beam as indicated
earlier can have power density in excess of 12 MW/mm. As laser interacts with
the material, the energy of the photon is absorbed by the work material leading
to rapid substantial rise in local temperature. This in turn results in melting and
vaporization of the work material and finally material removal [6].
1.4.2. Laser Beam Cutting
In this process, the focused laser beam hits the work piece where it locally
fuses the material and also partly or completely vaporizes it. By the impulse of
a gas jet emerging from a nozzle the material is removed and leaves the kerf
due to the relative movement of beam and work piece Figure.1.6. The gas jet is,
at the same time, meant to protect the focusing optical system against vapor and
weld spatter.
Figure 1.6: Laser beam cutting process [6].
7
Laser cutting is a thermal cutting process in which a cut kerf (slot) is formed by
the heating action of a focused traversing laser beam of power density on the
order of 104 W mm-2 in combination with the melt shearing action of a stream
of inert or active assist gas.[8] The focused laser beam melts the material
throughout the material thickness and a pressurized gas jet, acting coaxially
with the laser beam, blows away the molten material from the cut kerf. The
terms related to the cutting process are illustrated in Figure 1.7.
Figure 1.7: Terms related to the cutting process of the work piece [7].
1.4.3. Types of laser Cutting:
In laser cutting, three versions of the process are to be distinguished:
 Oxygen laser cutting

 Fusion laser cutting
 Evaporative laser cutting
Oxygen Laser Cutting: The laser beam heats the material to ignition
temperature. The oxygen injected into the kerf burns the material and expels the
slag formed. The combustion process generates additional energy. With the
quality of the cut being continuously high, a distinct connection between the
purity of the oxygen and the maximum possible cutting speed can be proven.
Fusion Laser Cutting: In this version of the procedure, the material gets fused
in the crossover point by laser radiation. The melt is expelled from the kerf by
an inert gas. High-pressure fusion laser cutting is proving to be increasingly
8
successful in oxide-free cutting of stainless steels. It is also successfully used in

cutting mild steels and aluminum. As a rule, nitrogen is used as the cutting gas.
The cutting gas pressure at the cutting nozzle can be 20 bar and above.
Evaporative Laser Cutting: In evaporative laser cutting the material to be cut

is evaporated at the crossover point of the laser beam. An inert gas, e.g.,
nitrogen or argon, expels the by products from the kerf. This cutting process is
used with materials that have no liquid phase or melt, as is the case with paper,
wood, several synthetic materials or plastics, textiles, and ceramics.
At present, CO2 lasers with performances of up to 5 kW and Nd: YAG lasers

with performances of up to 2 kW are in use for laser cutting. Special CO2
cutting lasers with performances of up to 5000 W allow process-reliable
machining of mild steel plates with a thickness of up to 25 mm. With high-
speed thin-sheet cutting, cutting speeds of up to 40 mm/min are achieved. New
drive mechanics allows positioning speed of up to 300m/min [8].
So far, CO2lasers have proven suitable tools for fast 2-D laser cutting of thin
sheets due to their good focus ability and high laser beam performances. By
increasing the beam quality of solid-state lasers through the use of diode-
pumped Nd:YAG lasers, with new resonator programs, launching into ever
smaller fibers becomes possible with which, in the meantime, suitable focusing
or high-speed cutting has become practicable.
1.4.4. Type of Lasers
Lasers are basically of two types, i.e. solid laser and gas laser. These can
be pulsed or of continuous type:
Solid Lasers
Solid state lasers (viz., ruby and Nd : glass), operate below 1 or 2 Hz, frequency
and are used only for low pulse applications like spot welding, drilling, etc.
9
However, Nd : YAG laser is also used for operations like cutting. Many laser
materials are now available, viz., calcium fluoride crystals doped with
neodymium (Ca + F2Nd). The round crystal rod with reflective ends is used.
Crystalline ruby (Al2O3+ Cr2 (0.05 %)) is another material used for laser
action. Flash lamp surrounding the ruby rod produces light, and both are
enclosed in a cylinder. This cylinder has highly reflective internal surfaces
which direct the light from the flash lamp into the rod. This light excites the
chromium ion of ruby crystal to high energy levels. While on return journey to
the normal state, these excited ions at high energy levels release the photons (or
energy in the form of short duration pulses). The schematic of LBC is shown in
Figure 1.8.
Figure 1.8: Schematic of Nd:YAG laser beam machine.
Gas Lasers
CO2 lasers typically emit light with a wavelength of 10.6 µm and possess
overall efficiencies of approx. Figure 1.9. ''10 to 13 %. CO2 lasers always
10
utilize a gas mixture to generate the laser beam. This laser gas mixture
invariably consists of helium, nitrogen, CO2 and possibly other additives. The
laser active medium, CO2 gas, is stimulated by electrical gas discharge. During
this process, the nitrogen molecules transfer energy from electron impact to the
CO2 molecules. The transition from energetically excited CO2 molecules (upper
vibrational level) to a lower energy level (lower vibrational level) is
accompanied by photon release, ultimately leading to emission of a laser beam.
By colliding with the helium atoms, which comprise the major share of the gas
mixture, the CO2 molecules return to the ground state and are now available for
another cycle. Basically, low pressure (typically 100 – 250 MPa) is needed to
allow these events to proceed in the laser gas. The excess energy that cannot be
used to generate the laser beam must be removed from the system as heat. The
pertinent technologies depend on the design of the laser instrument and are to
be discussed in greater detail. [6].
Figure 1.9: Principle of a CO2 laser [10].
1.4.5. Advantages and Limitations

Material cutting is one of the major steps involved in a manufacturing process.
This step must be performed accurately and efficiently so that the subsequent
11
steps of the manufacturing process can be carried out properly. Current

manufacturing units use lasers to perform this task. Laser cutting machines are
integrated with a computer-controlled programing system which determines
where and how the cut has to be made on the material. Even though laser beams
have great utility, they also have some limitations .
Advantages
 Holding the work piece in right position is easier in case of laser cutting
as compared to mechanical cutting
 Cuts obtained are extremely precise and do not require a lot of time.
Rather, the entire process of cutting the material is quite easy and
accomplished in less time than that required by conventional cutting
machines.
 As the cut is made by the help of a laser beam, there is no direct contact
of the workpiece with any cutting instrument, thereby eliminating the
risk of material contamination.
 Laser cutting machines use less energy for cutting metal sheets as
compared to plasma cutting technology.
 Laser cutting is controlled by the help of computer programs, thereby
saving substantial amount of manpower.
 As the machine does not require human involvement except for repairs
and test runs, the incidence of accidents and injuries is also reduced.
 Laser cutting is extremely versatile and can be used to cut or engrave
simple to complex designs on a workpiece.
 Manufacturing units with space constraints benefit a lot by installing
laser cutting machines because one or two laser cutters are capable of
performing the job of several other machines used for cutting.
12
Limitations
 Rate of production is not consistent when laser cutting is used. It highly
depends on thickness of the workpiece, type of material, and type of
laser used.
 Carelessness in adjusting laser distance and temperature may lead to
burning of some materials. Certain metals tend to discolor if the
intensity of the laser beam is not as per requirement.
 Cutting plastic with the help of these machines may cost a lot of money
because plastic emits fumes when subjected to heat. Due to this, the
entire setup has to be made in a well-ventilated room which can be quite
expensive. Also, fumes released during the process can be toxic and may
prove to be fatal.
 Human involvement is only needed in case of test runs and repairs.
During these tasks, if by mistake, a worker comes in contact with the
laser beam, he may suffer from serious burns.[11]
1.5. STAINLESS STEEL
STAINLESS STEELS are iron-base alloys that contain a minimum of

about 12% Cr, the amount needed to prevent the formation of rust in unpolluted
atmospheres (hence the designation stainless). Few stainless steels contain more
than 30% Cr or less than 50% iron. They achieve their stainless characteristics
through the formation of an invisible and adherent chromium-rich oxide film.
This oxide forms and heals itself in the presence of oxygen. Other elements
added to improve particular characteristics include nickel, manganese,
molybdenum, copper, titanium, silicon, niobium, aluminum, sulfur, and
selenium. Carbon is normally present in amounts ranging from less than 0.03%
13
to over 1.0% in certain grades [12]. Figure1.10 provides a useful summary of

some of the compositional and property linkages in the stainless steel family,
Figure 1.10: Available stainless steel alloy systems. [12]
By the metallurgical phases present in their microscopic structures:
 Ferritic stainless steel

 Martensitic (including precipitation hardening steels)
 Austenitic stainless steel
 Duplex steels, consisting of mixture of ferrite and austenite
Ferritic stainless steels: consist of chromium (typically 12.5% to 17%) and

iron. Ferritic stainless steels are essentially nickel-free. These materials contain
very little carbon and are non-heat treatable, but exhibit superior corrosion
resistance to martensitic stainless steels and possess good resistance to
oxidation. They are ferromagnetic and, although subject to an impact transition
(i.e. become brittle) at low temperatures, possess adequate formability. Their
14
thermal expansion and other thermal properties are similar to conventional

steels. Ferritic stainless steels are readily welded in thin sections, but suffer
grain growth with consequential loss of properties when welded in thicker
sections.
Martensitic stainless steels: consist of carbon up to 1.2%, chromium (11-18%)

and small amounts of manganese and nickel. These materials may be heat
treated, in a similar manner to conventional steels, to provide a range of
mechanical properties, but offer higher hardenability and have different heat
treatment temperatures. Their corrosion resistance may be described as
moderate (i.e. their corrosion performance is poorer than other stainless steels
of the same chromium and alloy content). They are ferromagnetic, subject to an
impact transition at low temperatures and possess poor formability. Their
thermal expansion and other thermal properties are similar to conventional
steels. They may be welded with caution, but cracking can be a feature when
matching filler metals are used.
Austenitic stainless steels: consist of chromium (16-26%), nickel (10-22%)

and iron. Other alloying elements (e.g. molybdenum) may be added according
to the desired properties to produce derivative grades. The austenitic group
contains more grades that are used in greater quantities, than any other category
of stainless steel. Austenitic stainless steels exhibit superior corrosion resistance
to both ferritic and martensitic stainless steels. Corrosion performance may be
varied to suit a wide range of service environments by careful alloy adjustment
e.g. by varying the carbon or molybdenum content. These materials cannot be
hardened by heat treatment and are strengthened by work-hardening. Unlike
ferritic and martensitic stainless steels, austenitic grades do not exhibit a yield
point. They offer excellent formability and their response to deformation can be
controlled by chemical composition. They are not subject to an impact
transition at low temperatures and possess high toughness to cryogenic
15
temperatures. They exhibit greater thermal expansion and heat capacity, with
lower thermal conductivity than other stainless or conventional steels. They are
generally readily welded, but care is required in the selection of consumables
and practices for more highly alloyed grades. Austenitic stainless steels are
often described as non-magnetic, but may become slightly magnetic when
machined or worked.
Duplex stainless steels: consist of chromium (22-25%) nickel (5-7%),

molybdenum up to 4%, copper and iron. These stainless steels have a
microstructure consisting of austenite and ferrite, which provides a combination
of the corrosion resistance of austenitic stainless steels with greater strength.
Duplex stainless steels are weldable, but care must be exercised to maintain the
correct balance of austenite and ferrite. They are ferromagnetic and subject to
an impact transition at low temperatures. Their thermal expansion lies between
that of austenitic and ferritic stainless steels, while other thermal properties are
similar to plain carbon steels. Formability is reasonable, but higher forces than
those used for austenitic stainless steels are required [12].
The Difference between 316 and 316L Stainless Steel is that 316L has a 0.03
max carbon and is good for welding whereas 316 have a mid-range level of
carbon. 316 and 316L are austenitic alloys, meaning that these stainless steel
products gain corrosion resistance from use of a nonmagnetic solid solution of
ferric carbide or carbon in iron in the manufacturing process.
1.5.1. Applications for Stainless Steels
Stainless steels are used in a wide variety of applications. Most of the

structural applications occur in the chemical and power engineering
industries, which account for more than a third of the market for stainless steel
products (table 1.1). These applications include an extremely diversified range
of uses, including nuclear reactor vessels, heat exchangers, oil industry tubular,
16
components for chemical processing and pulp and paper industries, furnace
parts, and boilers used in fossil fuel electric power plants. The relative
importance of the major fields of application for stainless steel products are as
follows:
Table 1.1 Application for stainless steel products [12].
Application Percentage
Industrial equipment
Chemical and power engineering 34
Food and beverage industry 18
Transportation 9
Architecture 5
Consumer goods
Domestic appliances, household utensils 28
Small electrical and electronic appliance 6
1.5.2. 316 L stainless steel Applications
316L Austenitic Stainless Steel is mainly used for anti-corrosion material

in commercial installation, such as structure parts, tower, trough, vessels, pipes
and exhaust manifolds, furnace parts, heat exchangers, pharmaceutical and
photographic equipment, valve and pump trim, chemical equipment, containers
of fluid, evaporators, pulp, paper and textile processing equipment, parts
exposed to marine atmospheres and tubing.[13].
1.6. Problem Statement
This study models the laser Cutting Machining process using Response
Surface Methodology and design of experiments over a realistic domain of
process variables 316L stainless Steel are used for experimentation. Three
objectives are modeled formulated and optimized, these are; a. The Kerf
Taper (Ta); b. Average surface roughness (Ra); c. Heat Affected zone (HAZ).
Process parameters include machining on: Power (Watt); oxygen pressure
17
(MPa); pulse frequency (Hz) and cutting speed (Cm/min). Optimization of

single and multi-objective formulation are given and discussed.
1.7. Thesis organization
This thesis consists of six chapters. The first chapter presents an

introduction to the study. Literature review is given in Chapter 2. Chapter 3
presents the experimental work Chapter 4 presents modeling of LBM process
using RSM, Chapter 6 gives Confirmation Experiments, Chapter 6 gives
discussion and Analysis of Results. Finally, Conclusions and recommendations
for future research are given in Chapter 7.
18
Chapter two Literature Review
Chapter 2
LITERATURE REVIEW
2.1. Introduction
In this Chapter, a review is given under the following headings.
 Taguchi Approach (Robust Design).

 Response Surface Methodology (RSM).
 Laser Beam Machining (LBM).
 The Austenitic Stainless Steel (316 L).
 Modeling and optimization of LBM process.
2.2. Taguchi Approach
Taguchi’s two most important contributions to quality engineering are the

use of quadratic loss function to quantify quality and the development of robust
designs (System, Parameter and Tolerance design(. Taguchi’s robust designs
have widespread applications upstream in manufacturing to fine tune a process
so that the output is insensitive to noise factors [17].
These authors confirm that Dr. Taguchi made important contributions to quality
engineering; however, it may not be easy to apply his techniques to real life
problems without some statistical knowledge. Specifically, the use of signal-to-
noise ratios in identifying the nearly best factor levels in order to minimize
quality losses. Three important discussions on Taguchi methods are published.
Some other Performance measures are given and discussed as alternative to
signal-to-noise ratios. Taguchi’s parameter design is discussed extensively by a
group of scientists in a discussion panel chaired [14].
The major point is that Taguchi methods do not have a statistical basis and
signal-to- noise ratios pose some computational problems. Taguchi’s parameter
design is studied and claimed that putting controllable and uncontrollable
19
factors in two separate arrays, inner and outer, will result in more experimental
runs [14].
It is expensive to arrive at a process having on target mean and minimum

variance with Taguchi methods. They suggest an alternative model based on an
asymmetric quality loss to obtain the most economical process mean. Previous
arguments and alternative approaches to Taguchi methods are summarized.
Alternative performance measures are discussed and compared with signal-to
noise ratios [14].
2.2.1. Description of Taguchi Method and Quality Engineering
Taguchi proposed a three-step Approach:

 System design.
 Parameter design.
 Tolerance design.
In system design, the engineer applies scientific and engineering knowledge to
produce a basic functional prototype design. The objective of parameter design
is to optimize the settings of the process parameter to improve quality
characteristics. In addition, it is expected that the optimal process parameter
values obtained from parameter design are insensitive to variation and other
noise factors.
Tolerance design is used to determine and analyze tolerances around the

optimal settings recommend by the parameter design. Tolerance design is
required if the reduced variation obtained by the parameter design does not
meet the required performance, and involves tightening tolerances on the
product or process parameters for which variations result in a large negative
influence on the required product performance.
Accordingly, parameter design is the key step in Taguchi method to achieve

high quality without increasing cost [4].
2.2.2. Orthogonal Arrays
An orthogonal array (OAS) is an n x m matrix whose columns have the property

that in every pair of columns each of the possible ordered pairs of element
appears the same number of times. Tables 2.1 and 2.2 display OA4 (23) and
OA8 (27) respectively. Note that in every pair of columns, each of the 4 ordered
pairs (0, 0), (0, 1), (1, 0), and (1, 1) appears exactly once. Similarly, every pair
of columns in table 2.2 contains each of the four pairs (0, 0), (0, 1), (1, 0), and
(1, 1) exactly twice. The letter L in this notation stands for Latin square, and it
indicates that orthogonal arrays are generalized Latin squares.
Orthogonal arrays can be viewed as plans of multifactor experiments where the

columns correspond to the factors, the entries in the columns correspond to the
test levels of the factors and the rows correspond to the test runs. Fractional
factorial plans based on orthogonal arrays irrespective of the degree of
fractionation are necessarily orthogonal plans. This is the primary reason for the
popularity of fractional factorials based on orthogonal arrays. Table 2.3
displays two orthogonal arrays: OA18 (61x36) and OA18 (21x37) [3], [36].
Table 2.1: Orthogonal array OA4 (23) [3].
Trial no. Colum no.
0 1 3
1 0 0 0
2 0 1 1
3 1 0 1
4 1 1 0
1
Table 2.2: Orthogonal array OA8 (27) [3].
Trial no. Colum no.
1 2 3 4 5 6 7
1 0 0 0 0 0 0 0
2 0 0 0 1 2 1 1
3 0 1 1 0 0 1 1
4 0 1 1 1 2 0 0
5 1 0 1 0 2 0 1
6 1 0 1 1 0 1 0
7 1 1 0 0 2 1 0
8 1 1 0 1 0 0 1
Table 2.3: Orthogonal arrays: OA18 (61x36) and OA18 (21x37) [3].
Trial no. Colum no.
1` 1 2 3 4 5 6 7 8
1 0 0 0 0 0 0 0 0 0
2 0 0 0 1 1 1 1 1 1
3 0 0 0 2 2 2 2 2 2
4 1 0 1 0 0 1 1 2 2
5 1 0 1 1 1 2 1 0 0
6 1 0 1 2 2 0 0 1 1
7 2 0 2 0 1 0 2 1 2
8 2 0 2 1 2 1 0 2 0
9 2 0 2 2 0 2 1 0 1
10 3 1 0 0 2 2 1 1 0
11 3 1 0 1 0 0 0 2 1
12 3 1 0 2 1 1 0 0 2
13 4 1 1 0 1 2 0 2 1
14 4 1 1 1 2 0 1 0 2
15 4 1 1 2 0 1 2 1 0
16 5 1 2 0 2 1 2 0 1
17 5 1 2 1 0 2 0 1 2
18 5 1 2 2 1 0 1 2 0
* Columns 1`, 3, 4, 5, 6, 7 and 8 from OA18 (61 * 36).

Columns 1, 2, 3, 4, 5, 6, 7 and 8 from OA18 (21 * 37).
2.3. Response Surface Methodology (RSM)
Methods to achieve robust design are described for single response and
achieve yield maximization. The Cp and Cpk capability indices are applied to a
single response problem and two multiple response problems. Results showed
that the proposed method is capable of producing good manufacturing yield and
robust design simultaneously.
The simulated Annealing with fractional factorial design is combined to deal

with global optimization problems. The standalone technique is inefficient and
requires numerous iterations before a good solution is found. Several problems
are tested: these are multimodal functions, noise corrupt data fitting, non-linear
dynamic control and large parameter optimization problems. The hybrid
algorithm outperforms the standalone method.
The concept of dual response surface approach is developed to quantify

variability of vehicle launch design. Design of experiments and design analysis
codes are employed to construct the mean and standard deviation models.
These models are included in an optimization procedure to minimize variability
subject to mean weight constraints.
The dual response approach based on quadratic models is studied. The study
concluded that high order polynomial models are more effective than regular
low order models. The three used signal-to-noise based models are proposed for
optimization. The study stressed that the accurate models are obtained before
the optimization stage [5].
The methods of manufacturing process optimization are described using

Taguchi experimental design with historical data collected during normal
production. The approach is applied to manufacture of wide chord fan blades
for aircraft engines using the super-plastic forming process.
Several alternatives exist are mentioned; these are linear regression methods,
simple and inverse interpolation functions, and Artificial Neural Network
(ANN). They Concluded that:
 Direct Monte Carlo simulation technique has low performance in

complex systems.
 The first order reliability method, RSM and Neural Network (NN) are
suitable alternatives despite loss of accuracy
 Finally, both RSM and ANN techniques have comparable levels [5].
Two approaches for experimental optimization are developed of Gas Metal Arc
Welding (GMAW) process. These approaches are the RSM and Genetic
Algorithms (GA). Control Parameters include reference voltage, wire FR and
welding speed based on four quality Measures. These measures include
deposition efficiency, bead width, and depth of penetration and
reinforcement.
The use of surrogate models is evaluated in optimization of certain assembly

operation. Shape optimization is carried in conjunction with the RANS analysis
to evaluate the performance of different surrogate models. RSM, NN, and the
multiple surrogate model press-based averaging (PBA) are employed.
Anderson et al. (2007) presented an alternative hybrid approach combining
RSM and Principal component analysis (PCA) to optimize multiple correlated
responses in turning process with correlated responses. The problem is then
transformed into an optimization form using a single objective function and a
main constraint function. The objective function is aggregated from six
objective measures.
The non-linear diffusion process is investigated using RSM and transformation

of variables. D-optimal experimental design is performed and diffusion rate is

calculated. Residuals are both normally and randomly distributed. Variable
transformation techniques are used prior to RSM model building and
optimization.
A study on powder mixed electrical discharge machining (PMEDM) is

presented using RSM. Several factors are analyzed: these are pulse on time,
duty cycle, peak current and concentration of silicon powder. Their effects on
material removal rate and surface roughness are determined.
Statistical Taguchi approach and Back Propagation (BP) is utilized to evaluate

the effect of various parameters and identify the optimal parameter setup values
in gas-assisted injection molding process. Signal-to-noise ratios are used as
System responses. The effect of learning rate and number of hidden nodes on
the efficiency of NN are studied.
RSM is utilized to model the drilling operations. Three-level full factorial

design is employed to carry experimentation. The developed models are
verified by regression analysis and resulting models predict optimal process
parameters for typical drilling processes. The optimization of the developed
RSM is carried via a new technique known as Tribes.
A similar procedure is followed to determine the optimal heat treatment
conditions of different Ni-Co-Mo surfaced layers.
RSM and ANN are developed to predict the surface roughness on mould
surfaces. Model variables include radial force, speed, axial-radial depth of cut
(DOC) and machining tolerances. Both modelling techniques are compared via
computational cost, cutting forces, tool life and dimensional accuracy. ANN
model requires large number of iterative calculations compared with a single
step for RSM. The problem is even worse for highly non-linear and large size
problems [5].
2.4. Laser Beam Machining (LBM)
LBM is accomplished by precisely manipulating a beam of coherent light

to vaporize unwanted material. LBM is particularly suited to making accurately
placed holes. It can be used to perform precision micromachining on all micro-
electronic substrates such as ceramic, silicon, diamond, and graphite. Examples
of microelectronic micromachining include cutting, scribing and drilling all
substrates, trimming any hybrid resistors, patterning displays of glass or plastic
and trace cutting on semiconductor wafers and chips.
Generation of the laser beam involves stimulating a lasing material by electrical

discharges or lamps within a closed container. As the lasing material is
stimulated, the beam is reflected internally by means of a partial mirror, until it
achieves sufficient energy to escape as a stream of monochromatic coherent
light. Mirrors or fiber optics are typically used to direct the coherent light to a
lens, which focuses the light at the work zone. The narrowest part of the
focused beam is generally less than 0.3 mm in diameter. Depending upon
material thickness, kerf widths as small as 0.1 mm are possible. In order to be
able to start cutting from somewhere else than the edge, a pierce is done
before every cut. Piercing usually involves a high power pulsed laser beam
which slowly makes a hole in the material.
There are three main types of lasers used in laser cutting. The CO 2 laser is
suited for cutting, boring, and engraving. The neodymium (Nd) and
neodymium yttrium–aluminum–garnet (Nd:YAG) lasers are identical in style
and differ only in application. Nd is used for boring and where high energy but
low repetitions are required. The Nd–YAG laser is used where very high power
is needed and for boring and engraving. Both CO2 and Nd–YAG lasers can be
used for welding. Common variants of CO2 lasers include fast axial flow, slow
axial flow, transverse flow, and slab [6].
A survey of the literature indicates that most of the published papers are based
on studying of laser cutting of metals ceramics and composites. Some authors
have performed preliminary investigations on the laser cutting of difficult to
laser cut materials i.e. highly reflective and thermally conductive materials,
such as aluminum and its alloys and stainless steel.
The applications of these materials in technologically advanced industries

demand narrow and intricate cuts. The geometrical accuracy of laser cut
specimens mainly depends upon the kerf width, kerf taper and kerf deviation
along the length of a cut. Findings of different papers show that most of the
researchers have concentrated their study on optimizing the kerf taper and kerf
width without considering the kerf deviation along the length of a cut. The kerf
deviation along the length of cut becomes more important when the material is
difficult to laser cut. As for the best of author’s knowledge, no work has been
found related to the Taguchi based fuzzy logic multi objective optimization of
the laser cutting process. Considering all these facts, the objective of the present
research work is to analyze the laser cutting performance of Duralumin sheet
with the aim to optimize the multiple quality characteristics such as kerf width
and kerf deviations simultaneously by using hybrid approach of Taguchi robust
parameter design method.
2.5. Modelling and Optimization of LBM Process Parameters
The literature related to modelling and optimization of LBM is mainly

using statistical DOE such as Taguchi method and response surface method.
Several analytical methods based on different solution methodologies, such as
exact solution and numerical solution, have also been examined related to
LBM.
Tosun and Ozler [15] applied Taguchi methodology for optimization of surface
roughness and tool life simultaneously during hot turning of high manganese
steel workpiece using the sintered carbide tool on a lathe. They also studied the
effect of hot turning parameters (cutting speed, depth of cut, feed rate, and
workpiece temperature) on multiple performance characteristics.
Almeida et al. [16] have used factorial design for experimentation to study the
effects of the energy per pulse, overlapping rate and type of assist gas on the
surface roughness and dross formation during Nd: YAG laser cutting of pure
titanium (grade 2) and titanium alloy Ti-6Al-V (grade 5) sheets.
Tam et al. [17] reported the use of the Taguchi technique of experimental
design in optimizing the process parameters for drilling deep-holes in nickel
based super alloy, Inconel 718. The thickness of the material was 25.0 mm.
Oxygen was the assist gas and the focal length of the focusing lens was 300
mm. The effects of five process parameters-pulse energy, pulse duration, pulse
shape, focal position, and assist gas pressure were explored. The various
parameters were assigned to an L18 orthogonal array. The primary response
under study was the drilling time. It was predicted that a minimum drilling time
of 31.51 s was needed to drill a hole with a pulse energy of 30.0 J, a pulse
duration of 1.8 ms, a ‘‘treble’’ pulse shape, and an oxygen pressure of 0.35
MPa.Chen et al. [18] discussed the use of the Taguchi method of experimental
design in optimizing process parameters for micro engraving of iron oxide-
coated glass using a Q-switched Nd:YAG laser. The effects of five key process
parameters- beam expansion ratio, focal length, average laser power, pulse
repetition rate and engraving speed-were explored. The primary response under
study was the engraving line width. An L16 orthogonal array was used to
accommodate the experiments. The study indicated that a minimum line width
of 18 mm could be obtained with beam expansion ratio of 59, focal length of 50
mm, laser average power of 0.4 W, pulse repetition rate of 5 kHz, and
engraving speed of 5mm/min.
Dubey and Yadava [19] applied a hybrid approach of Taguchi method and
principal component analysis (PCA) for multi-objective optimization of pulsed
Nd:YAG laser beam cutting of nickel-based superalloy (SUPERNI 718) sheet
to achieve better cut qualities within existing resources. The three-quality
characteristics kerf width, kerf deviation (along the length of cut), and kerf
taper were considered for simultaneous optimization. The input parameters
considered were assist gas pressure, pulse width, pulse frequency, and cutting
speed. Initially, single-objective optimization was performed using Taguchi
method and then the signal-to-noise (S/N) ratios obtained were further used in
PCA for multi-objective optimization. The results included the prediction of
optimum input parameter levels and their relative significance on multiple
quality characteristics.
Li, Tsai, and Yang [20] applied Taguchi’s robust design methodology to study
the depth of cut, width of cut and HAZ during laser cutting of Quad Flat No-
lead (QFN) packages using a diode pumped solid state laser (DPSSL) system.
They found that three control factors such as laser frequency, cutting speed, and
laser driving current contributed greatly to improve the laser cutting quality.
RSM is a statistical technique employed to design experiments with a reduced

number of experimental runs to achieve optimum responses. It is used to
establish mathematical models which correlate the responses and the
independent control parameters. Sharma and Yadava [21] performed
experiments in a 200W pulsed Nd: YAG laser beam machine system with CNC
worktable. Surface quality i.e. kerf taper and surface roughness are the required
measures of response. Process parameters which are considered that affect
these responses are assist gas pressure (x1), pulse width (x2), pulse frequency
(x3), and cutting speed (x4).The relationship of process parameters and output
responses is represented mathematically. Li et al. [22] have applied the LBM
and EDM sequentially for micro-drilling of fuel injection nozzles .They
9
initially applied the laser drilling to produce the micro-holes and then EDM was
used for rimming the drilled micro-holes. They claimed that this hybrid
approach has eliminated the recast layer and heat affected zones (HAZs)
typically associated with laser drilling. They also claimed that the hybrid
process enabled 70% reduction in drilling time as compared with EDM drilling.
Electro-chemical or chemical etching processes are combined with laser beam
for localized etching to enable selective material removal. The use of LAE has
improved the etched quality and etching rate of super-elastic micro-gripper
prepared by cutting of nickel–titanium alloy [23].Prusa et al. [24] have
developed a numerical model for the calculation of heat conduction losses,
cutting speed and temperature distribution in HAZ in laser cutting of thick
materials. Tam et al.[25]applied Taguchi method to study the laser cutting
process for 4.5 mm thick mild steel sheet using Rofin Sinar RS500 laser. The
S/N ratio of overall figure-of-merit (FOM) is considered as quality function.
FOM function integrates weighted effects of quality characteristics (kerf width,
surface roughness, micro-hardness, slope of cut edge and HAZ) and cost
components (cutting speed, oxygen pressure and beam power).
Laser beam cutting (LBC) is a process in which a high intensity focused laser
beam is used to melt; and possibly vaporize the work piece along the traverse
contour. A pressurized assist gas jet is used to eject the molten or
vaporized layer. The cut separating the work piece (known as the kerf) is
created by the relative motion between the incident laser beam and the work
piece. The principle components of the laser cutting system include: the laser
which generates the used laser beam, the beam guidance train (i.e. fiber
optics or mirrors), the laser cutting head which consists of the focusing optics
and assist gas nozzle assembly, and the work piece handling equipment. A
schematic illustration of the laser cutting head and work piece is shown in
Figure 2.1.
Figure 2.1 A schematic illustration of the laser cutting head and work piece.
2.6. Laser Cutting Parameters
The quality of the laser cutting process and consequently the resulting cut
edge quality is governed by a number of parameters related to the laser
system, material, and the process Figure 2.2 [26]. The laser system
parameters include the wavelength of the laser radiation, maximum output laser
power, and laser beam quality, the material parameters ''including the
material type and thickness'', and the processing parameters '' including the
used laser power'', cutting speed, focal length of the focusing lens, focal point
position relative to work piece top surface'', type and pressure of assist gas,
nozzle diameter, and nozzle standoff distance. For the cutting of a specific
material (work piece thickness); using a particular laser system, the
processing parameters can be altered by the operator so as to optimize the
cutting process and obtain high cut quality at a high cutting speed for high
productivity. It is noted , that laser system parameters - which are
characteristics of the used laser system - cannot be modified by the operator.
1
Figure 2.2 Laser cutting parameters.
2.7. Stainless Steel
Austenitic stainless steels are iron-chromium-nickel alloys which are hard

enable only by cold working. Nickel is the main element varied within the
alloys of this class while carbon is kept to low levels. The nickel content may
be varied from about 4% to 22% - higher values of nickel are added to increase
to ductility of the metal. When chromium is increased to raise the corrosion
resistance of the metal, nickel must also be increased to maintain the austenitic
structure.
These alloys are slightly magnetic in the cold-worked condition, but are
essentially non-magnetic in the annealed condition in which they are most often
used. The austenitic types feature adaptability to cold forming, ease of welding,
high-temperature resistant, and, in general, the highest corrosion resistant.[12]
The following are brief descriptions of some of the most commonly ordered
stainless steels:
Type 302 stainless steel has a great corrosion resistance but less work
hardening than Type 301. This basic alloy of the austenitic group often referred
to as 18:8. Machinability: 40%. Drawing or stamping: good and welding: very
good, consist of 0.15 carbon.
Type 304 stainless steel has lower carbon to minimize carbide precipitation. It
is less heat sensitive than 18:8 steels. Used in high-temperature applications.
Machinability: 45%, drawing or stamping; very good and welding: very good,
tough welds, consist of 0.08 carbon.
Type 304L stainless steel has an extra low carbon content to avoid harmful
carbide precipitation in welding applications. Its corrosion resistance is
comparable to type 304. Machinability: 44%. Drawing or stamping: very good,
welding: very good; recommended for welding, consist of 0.03 carbon.
Type 316 stainless steel contains molybdenum for better corrosion resistance -
particularly to pitting. Machinability: 45%. Drawing or stamping – good,.
welding: very good, tough welds, consist of 0.08 carbon.
Type 316L stainless steel has a carbon content lower than 316 to avoid carbide
precipitation in welding applications. Machinability - 45%. Drawing or
stamping ; good. Welding : very good, recommended for welding, consist of
0.03 carbon. [12].
Major properties of stainless steel (316 L) are listed on table 2.4 and 2.5,
including the chemical composition and Mechanical properties.
Table 2.4: Percentage chemical composition (wt.%) of stainless steel 316L [12]
Ni Cr C Mn Si S Mo P
10-14 16-18 0.03 2 0.75 0.03 2-3 0.045
Table 2.5: Mechanical properties of stainless steel 316 L [12]
Tensile Yield Elongation Reduction Brinell

Strength Strength in 4d of Hardness
(MPa) (0.2% Area,% HB
proof(MPa))
485 170 45 65 217
Chapter three Experimental Work
Chapter 3
EXPERIMENTAL WORK
3.1. INTRODUCTION
Metal cutting is one of the most significant manufacturing processes

in the area of material removal application. It is defined as the removal of
metal from a workpiece in order to obtain a finished product with desired
attributes of size, shape, and surface roughness.. In high speed cutting
operations such as laser cutting, dimensional accuracy, kerf width and quality
of surface finish are three factors that manufacturers must be able to
control among various process variables. Surface finish is an important factor
for determining the quality of a workpiece. Surface roughness is harder to attain
and track than physical dimensions; because relatively many factors affect
surface finish. Some of these factors can be controlled and some cannot. Key
controllable process parameters in laser metal cutting are cutting speed, assist
gas pressure and laser power. The parameter used to evaluate surface finish in
this study is the Roughness Average, Ra value.
The other parameter of interest is the kerf width. In laser machining the quality
of the cut is assessed by the narrowness of the kerf. This will result in minimal
loss of the material being processed.
3.2 . Experimental Investigations
In order to achieve the objectives of this study, experimental

investigations were performed using the Nd : YAG laser. The cutting tests
were specifically designed to investigate three research problems concerning
the laser cutting of thick-section metal. These research problems are outlined
below:
 The laser requirement for cutting stainless steel (316l) , the effects of
processing parameters and the resulting of cut edge quality in laser
oxygen.
 The processing parameters that influence the kerf width (Ta) , surface
roughness (Ra) and heat affected zone(HAZ) during laser cutting of
stainless steel (316l).
 The requirements for modelling and optimization of each Kerf Taper and
Surface roughness.
3.3. Selection of Measured Responses and Control Variables

Three different responses are selected: Kerf Taper (Ta), Average Surface
Roughness (Ra) and heat affected zone (HAZ). LBM process parameters are
Laser Power, Cutting Speed, Focal Length, Focal Position, Type of Assist Gas,
Duty cycle percent, Nozzle diameter, Nozzle stand -off distance and Frequency.
In this research, four control variables are used; the Machining parameters are
laser power (Watt), cutting speed (Cm/min), frequency (Hz) and gas pressure
(Mpa).
3.4. Plan of Experimentation

Four factors are used at three levels. Table 3.1 which indicate the factors to
be studied and the assignment of the corresponding levels. The array chosen is
the L27 (313) that has 27 rows corresponding to the number of tests (26 degrees
of freedom) with 13 columns at three levels, as shown in Table 3.2. The plan of
experiments is made of 27 tests in which the first column is laser power (Watt),
the second column to assist pressure (Mpa), the third column frequency (Hz),
the fourth column is cutting speed (Cm/min) and the remaining are assigned to
the interactions as shown in Figure 3.1. The tests are replicated, resulting in a
total of 81tests, to allow the analysis of variance and signal to noise ratio
calculation. The full factorial array in this experiment is L81 (34), but fractional
factorial array L27OA is used in this study. Table 3.3 gives L27OA used for
experimentation of LBM process.
Table 3.1: Assignment of levels to control factors.
Level 1 Level 2 Level 3
Power 150 Watt 450 Watt 750 Watt

(Watt)
Control Pressure 0.5 Mpa 1 Mpa 1.5 Mpa
Variables ( Mpa)
Frequency 25 Hz 75 Hz 125 Hz
(Hz)
Cutting speed 10 Cm/min 20 Cm/min 40 Cm/min
(Cm/min)
Figure 3.1: Linear Graph L27OA [27].

Table 3.2: L27 OA Orthogonal array [20].
Trial Colum no.

No. 1 2 3 4 5 6 7 8 9 10 11 12 13
1 1 1 1 1 1 1 1 1 1 1 1 1 1
2 1 1 1 1 2 2 2 2 2 2 2 2 2
3 1 1 1 1 3 3 3 3 3 3 3 3 3
4 1 2 2 2 1 1 1 2 2 2 3 3 3
5 1 2 2 2 2 2 2 3 3 3 1 1 1
6 1 2 2 2 3 3 3 1 1 1 2 2 2
7 1 3 3 3 1 1 1 3 3 3 2 2 2
8 1 3 3 3 2 2 2 1 1 1 3 3 3
9 1 3 3 3 3 3 3 2 2 2 1 1 1
10 2 1 2 3 1 2 3 1 2 3 1 2 3
11 2 1 2 3 2 3 1 2 3 1 2 3 1
12 2 1 2 3 3 1 2 3 1 2 3 1 2
13 2 2 3 1 1 2 3 2 3 1 3 1 2
14 2 2 3 1 2 3 1 3 1 2 1 2 3
15 2 2 3 1 3 1 2 1 2 3 2 3 1
16 2 3 1 2 1 2 3 3 1 2 2 3 1
17 2 3 1 2 2 3 1 1 2 3 3 1 2
18 2 3 1 2 3 1 2 2 3 1 1 2 3
19 3 1 3 2 1 3 2 1 3 2 1 3 2
20 3 1 3 2 2 1 3 2 1 3 2 1 3
21 3 1 3 2 3 2 1 3 2 1 3 2 1
22 3 2 1 3 1 3 2 2 1 3 3 2 1
23 3 2 1 3 2 1 3 3 2 1 1 3 2
24 3 2 1 3 3 2 1 1 3 2 2 1 3
25 3 3 2 1 1 3 2 3 2 1 2 1 3
26 3 3 2 1 2 1 3 1 3 2 3 2 1
27 3 3 2 1 3 2 1 2 1 3 1 3 2
Table 3.3: L27OA used to plan experimentation.
Control Variables in Experiments (L27OA)
Power Pressure Frequency Cutting speed

Trial (Watt) ( MPa) (Hz) (Cm/min)
X1 X2 X3 X4
#
1 1 1 1 1
2 1 1 1 1
3 1 1 1 1
4 1 2 2 2
5 1 2 2 2
6 1 2 2 2
7 1 3 3 3
8 1 3 3 3
9 1 3 3 3
10 1 1 2 3
11 2 1 2 3
12 2 1 2 3
13 2 2 3 1
14 2 2 3 1
15 2 2 3 1
16 2 3 1 2
17 2 3 1 2
18 2 3 1 2
19 2 1 3 2
20 3 1 3 2
21 3 1 3 2
22 3 2 1 3
23 3 2 1 3
24 3 2 1 3
25 3 3 2 1
26 3 3 2 1
27 3 3 2 1
9
3.5. Material and Specimen Shape
The experiments are carried out on the austenitic stainless steel (316l), the
specimen shape is in the form of cuboids with length 50 mm, width 50 mm and
thickness 3 mm as shown in Figures 3.2 and 3.3.
Figure 3.2: Dimensions of Specimen.
Figure 3.3: Measurement of multi-performance responses.
The laser machine used in this investigation is a ND: YAG laser operating in
continuous and pulse mode. The output power capacity is 2.2KW, the laser
beam is focused by "Zn, Se" lens with focal length 200 mm. The minimum
diameter of focused beam is about 0.47 mm. The relative movement between
laser beam and work piece is realized CNC (computer numerical control) X-Y-
Z nozzle. For alignment procedure a Pulsed laser beam was transmitted along
the optical axes. Laser power of 150 W, 450 W, and750 W, at a wavelength of
10.6, were used in this study. The cutting speeds were used 10, 20, 40 Cm/min.
frequency at 25, 75 and 125 and oxygen flowing at 0.5,1 and 1.5 MPa were
used as the shielding gas with a length of cut 15 mm.
3.6. LBM Equipment
The LBM machine ROFIN DY x55-022which contains four units:
 Chiller (chamber).
 Laser source.
 Controller system.
 Cutting gas system.
The Chiller unit consists of three systems:
 Cutting robot.
 Cutting head.
 Work piece table with Positioning System; which consists of three-axis
(x, y, z) table to adjust the position of specimen manually as shown in
Figure 3.6
Laser source unit consists of two systems:
 Laser control unit

 Making control
Controller consists of two systems:
 Oscillator control pc.

 Robot controller.
Cutting gas system consists of two systems:
1
 The oxygen assist gas cylinder

 Control valves and the compressed air measurement
The general view of the laser beam cutting machine and the auxiliary supply
devices is given in Figures 3.4, 3.5 shows the block diagram of the laser
cutting machine assembly. Apart from the a ND: YAG laser tube, it consists
of a dc power unit, assist gas unit, a mother board for the cutting machine, a
computer, a control panel display and the x- y motion assembly. The dc power
supply provides the power for all the sub-units. The assist gas unit provides
pure assist gas which is used to remove the debris generated during the cutting
process. The focusing of the beam on the work is done with the help of helium-
neon laser. The entire unit is housed in a chamber with complete protection to
prevent beam leakage. Extra care is taken for a ND: YAG laser as it
delivers extremely high power output in the infra-red region. Figure 3.6
shows the layout of the equipment used.
Figure 3.4: The general view of the laser cutting machine.
As the beam is invisible, sufficient precaution is necessary during the

cutting operation. All polished objects which can reflect IR radiation must
be kept away from the machine units.
Figure 3.5: Block diagram of laser cutting system
Figure 3.6: Layout of the LBM Equipment.
Figure 3.7 shows the laser cutting head. It consists of optical wave guide
focusing lens and a coaxial nozzle to supply assist gas during the cutting
operation. The important units of laser cutting assembly are beam delivery
system, lenses, mirrors used in the waveguide and the polarizer , Figure 3.8
shows the laser supply assist gas.
Figure 3.7: Rofin Sinar laser cutting machine used in the present work
Figure 3.8 shows the laser supply assist gas
3.7. Laser Cut Quality Characteristics
The characteristics of the laser cut edge that can be used to define
the laser cut quality include: the cut kerf width(Ta), surface roughness(Ra),
and heat affected zone (HAZ) Figure 3.9.
3.7.1. Cut kerf width

The kerf is the cut slot that is formed during through thickness cutting
[28] . The cut kerf width is the distance separating the two cut surfaces of
the cut slot Figure 3.9, which represents also the amount of material removed
during the laser cutting process. The cut kerf width depends on the focused
spot size, laser power, and to some extent cutting speed.
The cut kerf width is also affected by the oxygen pressure during laser cutting
of stainless steel using oxygen assist gas. The dynamic nature of the erratic
exothermic oxidation reaction produces an irregular cut kerf width and
deep grooves (striations) on the cut edge. A relatively uniform cut kerf
width is formed during laser cutting of steel using an inert assist gas jet and the
striations formed on the resulting cut edges can be associated with the melt
flow mechanism and the moving melting front respectively.
Figure 3.9: The schematic representation of laser cut kerf.
3.7.2. Cut edge surface roughness
The cut edge surface roughness is the unevenness or irregularity of

the cut surface profile; which is observed as striations on the cut
edge like those shown in Figure 3.10. The dynamical behavior of
the laser cutting process affects the shape of the cutting front and the
melt flow mechanism resulting in the formation of striations on the
cut edge.
Figure 3.10 the cut edge surface roughness.
3.7.3 Heat affected zone

In metal cutting operations, in general oxygen is used while argon or helium is
used for wood or plastic cutting [29]. When using oxygen as assist gas, it will
not only drag the melt away but will also provide exothermic reaction in the
cutting section enhancing the energy available for increasing the cutting speed
[30].To maintain a high production rate and an acceptable level of quality for
the cut parts, it is important to select the optimum combination of process
parameters, as these parameters impact on the special microscopic and
macroscopic characteristics of the finished parts, as signified by the kerf width,
the width of the heat affected zone (HAZ) and the surface roughness after
processing[31]
3.8. Measurement of the Process parameters

It was necessary to keep all parameters under strict control. The laser
machine was equipped with various measuring devices to permit detailed
examination of the interrelationship between the main process parameters. The
process parameters that must be taken in account before starting the
experimental tests are as the follows.
3.8.1 Measurement of the kerf width (Ta)

Laser cutting of stainless steel sheet was carried out along the curved path. The
curved profile was considered as quarter part of a circle (arc) Figure 3.9 .The
cut quality was evaluated in terms of the computed output performance
parameter Ta: three replications for each experimental run have been performed
to obtain the 15 mm cut length [18], table 3.4 gives the top and bottom kerf
widths; those were measured using the tool-maker microscope figure 3.11.
Figure 3.11: General view of the tool-maker microscope.
The kerf taper has been computed using the following formula [32]:
k w − k w × 8
Ta (deg.) = (3.1)
π× k
Table 3.4: Top& bottom kerf widths and Ta.
st rd
Ta
Bottom 1 Bottom 3
nd
Run Top Top Bottom 2 Top (Mean)
Trial Trial Trial
# deg
1 0.300 0.271 0.28 0.215 0.181 0.33 0.182 0.145 0.35 0.320
2 0.258 0.234 0.22 0.216 0.185 0.30 0.176 0.156 0.19 0.236
3 0.242 0.216 0.25 0.243 0.219 0.22 0.183 0.148 0.33 0.266
4 0.201 0.182 0.18 0.275 0.252 0.22 0.250 0.221 0.27 0.223
5 0.364 0.335 0.27 0.321 0.285 0.34 0.256 0.213 0.41 0.340
6 0.233 0.204 0.27 0.234 0.201 0.31 0.231 0.197 0.32 0.300
7 0.214 0.189 0.23 0.364 0.344 0.19 0.211 0.187 0.22 0.213
8 0.373 0.330 0.41 0.370 0.336 0.32 0.388 0.334 0.51 0.413
9 0.220 0.189 0.29 0.201 0.168 0.31 0.367 0.329 0.36 0.320
10 0.341 0.281 0.57 0.233 0.189 0.42 0.377 0.337 0.38 0.456
11 0.288 0.248 0.38 0.292 0.244 0.45 0.431 0.390 0.39 0.406
12 0.280 0.218 0.59 0.219 0.176 0.41 0.333 0.285 0.45 0.483
13 0.321 0.263 0.55 0.362 0.314 0.45 0.347 0.276 0.67 0.556
14 0.371 0.322 0.46 0.353 0.284 0.65 0.450 0.393 0.54 0.550
15 0.231 0.188 0.41 0.250 0.190 0.57 0.361 0.291 0.66 0.546
16 0.254 0.161 0.88 0.363 0.284 0.75 0.411 0.312 0.94 0.856
17 0.387 0.306 0.77 0.408 0.338 0.66 0.420 0.329 0.86 0.763
18 0.252 0.182 0.67 0.321 0.239 0.78 0.398 0.305 0.88 0.776
19 0.412 0.320 0.87 0.436 0.374 0.59 0.436 0.367 0.65 0.700
20 0.365 0.300 0.62 0.309 0.216 0.88 0.379 0.302 0.73 0.743
21 0.461 0.391 0.66 0.453 0.378 0.71 0.458 0.366 0.87 0.746
22 0.473 0.380 0.88 0.413 0.319 0.89 0.415 0.315 0.95 0.906
23 0.367 0.286 0.77 0.392 0.300 0.87 0.399 0.305 0.89 0.843
24 0.492 0.389 0.98 0.499 0.427 0.68 0.492 0.414 0.74 0.800
25 0.470 0.311 1.51 0.403 0.219 1.75 0.389 0.260 1.23 1.496
26 0.462 0.325 1.30 0.431 0.268 1.55 0.455 0.329 1.20 1.350
27 0.491 0.323 1.60 0.488 0.336 1.45 0.494 0.354 1.33 1.460
3.8.2 Measurement of the surface roughness (Ra)

The average surface roughness parameter (Ra) is used to quantify the surface
roughness of the machined surface. Ra is the arithmetic average height of
surface component (profile) irregularities from the mean line within the
measuring length used to describe the vertical dimensions of roughness Figure

3.12. Ra is measured using a contact type stylus based on surface roughness
tester. The device employed shown in Figure 3.13 is TAYLOR–HOBSON
instrument (SURTRONIC 3, 112/1500 – 1150483, DENMARK). Surface
roughness is measured on the LBM machined area three times and the average
value is used, the unit of measure is µm. Figure 3.13 show an optical
microscope used for measuring of kw, All measurements were acquired using
4.00 mm evaluation length.
Figure 3.12: Plot of average surface roughness (Ra).
Figure 3.13: General view of the TAYLOR–HOBSON instrument.
9
3.8.3 Measurement of the heat affected zone (HAZ)

The dimension of the heat affected zone (HAZ) was studied under a
microscope. The heat affected zone is the area of the work material, which was
not melted during cutting but whose microstructure and mechanical properties
were affected and or altered by intense heat during laser cutting. The heat from
the laser cutting process and subsequent re-cooling causes this change in the
area surrounding the cut. The extent and magnitude of property change depends
primarily on the base material, and the amount and concentration of heat input
by the laser cutting process. Figure. 3.14 [33] shows a scheme with a typical
section of laser cut specimen and define the methodology to quantify dimension
of the heat affected zone (HAZ).
The tool-maker microscope was used to measure the width of HAZ along the
15mm segment of the cut edge. The measurements were repeated three times to
obtain averaged values.
Figure 3.14: Method of measurement of HAZ [33]

Chapter four: Modelling and Optimization of LBM Process
Chapter 4
MODELLING AND OPTIMIZZATION OF LBM PROCESS
4.1. Introduction
In this chapter, experimental results of LBM Process are given. Process

variables include Power, X1; Pressure, X2; Frequency, X3; and Cutting speed,
X4 respectively. Process responses include: the kerf width (Ta), surface
roughness (Ra) and the heat affected zone (HAZ) respectively. Statistical
analysis and design of experiments are performed on the mean response, the
standard deviation response and signal-to-noise ratios respectively. This results
in significant variables (at 90%, 95%, and 99% confidence levels) that affect
mean response, standard deviation (or variance response) and signal-to-noise
ratios respectively. Later, significant variables are used to develop
mathematical models for process modeling analysis and optimization.
4.2. Analysis of Experimental Results
4.2.1. Experimental results analysis

For analysis of the results obtained from experiments the S/N ratios, mean, and
standard deviation values are calculated. In this case, two response values for
each experimental condition as shown in Table 4.1. The S/N ratio for the Kerf
Taper for first trial is given in Eq. (4.1)
Ta1 = -10 log [(1/3) (0.282+0.332+0.352)] = 9.897dB (4.1)
The S/N ratio for the Surface Roughness (Ra) for first trial as shown in Eq.
(4.2)
Ra1 = -10 log [(1/3) (3.502+2.332+4.002)] = -10.5024 dB (4.2)
The S/N ratio for the heat affected zone (HAZ) for first trial as shown in Eq.
(4.3)
HAZ1 = -10 log [(1/3) (0.1102+0.0972+0.1002)] = 18.44421 dB (4.3)
1
The S/N ratio, mean, and standard deviation are determined similarly. The
L27OA array is employed as a 3-level 4 variables problem with interactions.
This represents 27 / 81 fraction of the required experiments. For example, 2-
level array such as L16OA or L32OA can be used. Similarly, an L81OA or
higher size can be used to model 3-level experiments.
Table .4.1 Experimental results of kerf taper (Ta), Average Surface Roughness
(Ra) and the heat affected zone (HAZ).
Kerf Taper (deg) Average Surface HAZ (mm)

Roughness (µm)
(Lower The Better) (Lower The Better)
(Lower The Better)
Run
# 1st 2nd 3rd 1st 2nd 3rd 1st 2nd 3rd
Trial Trial Trial Trial Trial Trial Trial Trial Trial
1 0.35 0.33 0.28 4.00 2.33 3.50 0.100 0.097 0.110
2 0.19 0.30 0.22 3.40 4.50 3.80 0.118 0.104 0.113
3 0.33 0.22 0.25 3.00 3.60 3.40 0.131 0.144 0.132
4 0.27 0.22 0.18 4.90 3.50 3.50 0.127 0.141 0.131
5 0.41 0.34 0.27 3.33 4.60 4.50 0.123 0.126 0.141
6 0.32 0.31 0.27 3.75 4.60 4.66 0.157 0.139 0.141
7 0.22 0.19 0.23 4.63 4.17 4.75 0.165 0.140 0.144
8 0.51 0.32 0.41 4.50 4.99 5.20 0.145 0.164 0.168
9 0.36 0.31 0.29 5.75 5.00 5.50 0.140 0.158 0.164
10 0.38 0.42 0.57 5.03 5.92 5.87 0.169 0.152 0.142
11 0.39 0.45 0.38 5.65 5.86 6.33 0.172 0.155 0.144
12 0.45 0.41 0.59 5.50 6.88 5.57 0.162 0.133 0.148
13 0.67 0.45 0.55 4.30 6.50 5.33 0.221 0.242 0.253
14 0.54 0.65 0.46 5.94 6.52 6.37 0.221 0.253 0.249
15 0.66 0.57 0.41 5.37 6.53 6.55 0.232 0.279 0.26
16 0.94 0.75 0.88 6.40 6.83 6.00 0.240 0.265 0.281
17 0.86 0.66 0.77 6.31 6.68 6.30 0.294 0.254 0.260
18 0.88 0.78 0.67 6.60 6.50 6.98 0.248 0.264 0.313
19 0.65 0.59 0.87 6.87 6.89 7.50 0.322 0.301 0.371
20 0.73 0.88 0.62 7.22 6.94 7.22 0.334 0.373 0.420
21 0.87 0.71 0.66 7.44 6.89 7.16 0.366 0.408 0.460
22 0.95 0.89 0.88 7.01 7.81 7.30 0.422 0.488 0.372
23 0.89 0.87 0.77 7.75 8.20 9.83 0.491 0.353 0.622
24 0.74 0.68 0.98 8.87 9.20 9.58 0.357 0.503 0.574
25 1.23 1.75 1.51 8.96 8.85 9.40 0.702 0.536 0.567

26 1.20 1.55 1.30 9.10 9.40 9.19 0.593 0.435 0.521
27 1.33 1.45 1.60 9.85 9.87 9.40 0.512 0.552 0.650
Table 4.2: Summary of S/N ratios using L27OA.
Kerf Surface Heat

Experimental Control Factors Taper Roughness Affected
Run L27OA (Ta) (Ra) Zone
# X1 X2 X3 X4 X1. X1. X1. S/N (dB) S/N (dB) (HAZ)
X2 X3 X4 S/N (dB)
1 1 1 1 1 1 1 1 9.860398 -10.5024 19.78687
2 1 1 1 1 2 2 2 12.35326 -11.8799 19.02986
3 1 1 1 1 3 3 3 11.35094 -10.4818 17.34231
4 1 2 2 2 1 1 1 12.92146 -12.0871 17.51446
5 1 2 2 2 2 2 2 9.249399 -12.4303 17.70523
6 1 2 2 2 3 3 3 10.43511 -12.7828 16.71954
7 1 3 3 3 1 1 1 13.39135 -13.1096 16.47425
8 1 3 3 3 2 2 2 7.52354 -13.8136 15.95480
9 1 3 3 3 3 3 3 9.860398 -14.689 16.23058
10 2 1 2 3 1 2 3 6.670905 -14.9971 16.20821
11 2 1 2 3 2 3 1 7.790207 -15.4954 16.05869
12 2 1 2 3 3 1 2 6.205736 -15.5875 16.58651
13 2 2 3 1 1 2 3 4.976184 -14.7299 12.43075
14 2 2 3 1 2 3 1 5.10651 -15.9612 12.34453
15 2 2 3 1 3 1 2 5.092926 -15.8123 11.77690
16 2 3 1 2 1 2 3 1.306705 -16.1493 11.61600
17 2 3 1 2 2 3 1 2.296144 -16.1675 11.37566
18 2 3 1 2 3 1 2 2.142669 -16.517 11.16965
19 3 1 3 2 1 3 2 2.929411 -17.0162 9.56080
20 3 1 3 2 2 1 3 2.487979 -17.0592 8.46606
21 3 1 3 2 3 2 1 2.475418 -17.1066 7.67834
22 3 2 1 3 1 3 2 0.846002 -17.362 7.33132
23 3 2 1 3 2 1 3 1.463221 -18.7299 6.00572
24 3 2 1 3 3 2 1 1.825669 -19.2958 6.25903
25 3 3 2 1 1 3 2 -3.58918 -19.1551 4.35101
26 3 3 2 1 2 1 3 -2.65800 -19.3048 5.67396
27 3 3 2 1 3 2 1 -3.32856 -19.7436 4.81772
Table 4.2 gives L27OA and corresponding responses using lower the kerf taper,
lower the better surface roughness and heat affected zone respectively.
Variables include: X1- power (Watt), X2 - pressure (MPa), X3 –frequency(Hz),
X4 – cutting speed (Cm/min).Three interactions are studied; these are (X1. X2),
(X 1. X3) and (X 1. X4) respectively (19).
Table 4.3: Summary of mean values using L27OA.
Experimental Control Factors

Run L27OA (Ta) (Ra) (HAZ)
X1 X2 X3 X4 X1. X1. X1. (Mean) (Mean) (Mean)
# X2 X3 X4
1 1 1 1 1 1 1 1 0.320 3.27 0.102
2 1 1 1 1 2 2 2 0.236 3.90 0.112
3 1 1 1 1 3 3 3 0.266 3.33 0.136
4 1 2 2 2 1 1 1 0.223 3.96 0.133
5 1 2 2 2 2 2 2 0.340 4.14 0.130
6 1 2 2 2 3 3 3 0.300 4.33 0.146
7 1 3 3 3 1 1 1 0.213 4.51 0.150
8 1 3 3 3 2 2 2 0.413 4.89 0.159
9 1 3 3 3 3 3 3 0.320 5.41 0.154
10 2 1 2 3 1 2 3 0.456 5.60 0.154
11 2 1 2 3 2 3 1 0.406 5.94 0.157
12 2 1 2 3 3 1 2 0.483 5.98 0.148
13 2 2 3 1 1 2 3 0.556 5.37 0.239
14 2 2 3 1 2 3 1 0.550 6.27 0.241
15 2 2 3 1 3 1 2 0.546 6.15 0.257
16 2 3 1 2 1 2 3 0.856 6.41 0.262
17 2 3 1 2 2 3 1 0.763 6.43 0.269
18 2 3 1 2 3 1 2 0.776 6.69 0.275
19 3 1 3 2 1 3 2 0.700 7.08 0.331
20 3 1 3 2 2 1 3 0.743 7.12 0.376
21 3 1 3 2 3 2 1 0.746 7.16 0.411
22 3 2 1 3 1 3 2 0.906 7.37 0.427
23 3 2 1 3 2 1 3 0.843 8.59 0.489
24 3 2 1 3 3 2 1 0.800 9.21 0.478
25 3 3 2 1 1 3 2 1.496 9.07 0.602
26 3 3 2 1 2 1 3 1.350 9.23 0.516
27 3 3 2 1 3 2 1 1.460 9.70 0.571
Table 4.4: Summary of Standard deviation values using L27OA.
Experimental Control Factors

L27OA Ta Ra HAZ
Run X1. X1. X1. (Standard (Standard (Standard
# X1 X2 X3 X4 X2 X3 X4 deviation) deviation) deviation)
1 1 1 1 1 1 1 1 0.036056 0.85711 0.006807

2 1 1 1 1 2 2 2 0.056862 0.55678 0.007095
3 1 1 1 1 3 3 3 0.056862 0.30551 0.007234
4 1 2 2 2 1 1 1 0.045092 0.80829 0.007211
5 1 2 2 2 2 2 2 0.070000 0.70613 0.009644
6 1 2 2 2 3 3 3 0.026458 0.50895 0.009866
7 1 3 3 3 1 1 1 0.020817 0.30615 0.013429
8 1 3 3 3 2 2 2 0.095044 0.35921 0.012288
9 1 3 3 3 3 3 3 0.036056 0.38188 0.012490
10 2 1 2 3 1 2 3 0.100167 0.50003 0.013650
11 2 1 2 3 2 3 1 0.037859 0.34818 0.014107
12 2 1 2 3 3 1 2 0.094516 0.77730 0.014503
13 2 2 3 1 1 2 3 0.110151 1.10074 0.016258
14 2 2 3 1 2 3 1 0.095394 0.30105 0.017436
15 2 2 3 1 3 1 2 0.126623 0.67550 0.023643
16 2 3 1 2 1 2 3 0.097125 0.41509 0.020664
17 2 3 1 2 2 3 1 0.100167 0.21656 0.021572
18 2 3 1 2 3 1 2 0.105040 0.25325 0.033867
19 3 1 3 2 1 3 2 0.147422 0.35809 0.035921
20 3 1 3 2 2 1 3 0.130512 0.16165 0.043062
21 3 1 3 2 3 2 1 0.109697 0.27501 0.047089
22 3 2 1 3 1 3 2 0.037859 0.40501 0.058184
23 3 2 1 3 2 1 3 0.064291 1.09436 0.134515
24 3 2 1 3 3 2 1 0.158745 0.35529 0.110639
25 3 3 2 1 1 3 2 0.260256 0.29103 0.088263
26 3 3 2 1 2 1 3 0.180278 0.15394 0.079103
27 3 3 2 1 3 2 1 0.135277 0.26576 0.071002
Analysis of variance (ANOVA): The objective of ANOVA is to investigate

the process parameters which affect the output parameters significantly.
ANOVA is performed as follows:
Calculation of S/N ratio for quality characteristics: For example, control

factor X1 level 1(or X11) of the kerf taper show as table (4.5) and appendix
B&C:
X11= [9.8604+12.3533+11.3509+12.9215+9.2494+10.4351+13.3913+
7.5235+ 9.8604] = 96.9459dB
Similarly, other mean and standard deviation values can be calculated.
Degree of freedom (DOF) : Factor X1 has three levels, it has two

degrees of freedom. In general, DOF associated with a factor is one less than
the number of levels. DOF total is one less than all number of trials, and DOF
associated with error is difference between DOF) total and DOF) factor.
Calculation of the total sum of squares: SS =[∑yi2] –T2/N (4.4)
Where yi: Response value at each run i, T: Sum of S/N values, and N:
total number of trials.
Calculation of sum of squares due to various factors (SS): Sum of

squares due to factor X (1) are calculated as:
SS) X1= [[X11]2/9+[X12]2/9+[X13]2/9-T2/N] (4.5)
Calculation of sum of squares due to error (SS) Error): The sum squares
due to error = Total sum of squares (SST) – Total of sums of square due to
various factors.
 Calculation of mean square (V): Mean square = sum of square /

Degree of freedom, as following example :
V) X1 = SS) X1 / DOF) X1 (4.6)

 Calculation of F value [F) calculated] is calculated as:
F) Calculated = Mean Square of each factor / SS) Error (4.7)

 Calculation the percentage of contribution (P%) is calculated as:
P) X 1 = [SS) X 1 /SS) Total] *100 (4.8)
 Calculation of F) Tabulated: according to specified confidence levels

or risk levels: 90%, 95% or 99% respectively.
 Pooled error variance: The error variance calculated by pooling the

smaller factorial effects.
Table 4.5: ANOVA results for the kerf taper (Ta) based on S/N ratios.
S/N by factor level (dB) Degree Sum of Mean Value Contribution

of Square Square of Percentage
1 2 3 Freedom ( SS) (V) F) P%
(DOF) calculated
X1: Power
96.945 41.588 2.452 2 500.924 250.462 74.883 79.86 %
X2: Pressure
62.124 51.916 26.944 2 72.784 36.392 10.880 11.61 %
X3: Frequency [Pooled Factor]
43.445 43.697 53.842 2 7.797 - - -
X4:Cutting speed [Pooled Factor]
39.164 46.244 55.576 2 15.048 - - -
X1.X2: Interaction Factor[Pooled Factor]
49.313 45.612 46.059 4 0.897 - - -
50.907 43.052 47.024 4 4.121 - - -
X1.X4: Interaction Factor [Pooled Factor]
52.338 42.753 45.892 4 5.286 - - -
Error 16 53.516 3.3447 8.53 %
Total 26 627.224 100%
Values of F) Tabulated at DOF) Factor = 2 and DOF) Error = 16 are [33]:
[DOF) Factor = 2 , DOF) Error =16]

Confidence Level 90% 95% 99%
%
F) Tabulated 2.67 3.63 6.23
Table 4.6: ANOM for The kerf taper results
Mean by factor Degree Sum of Mean Value Contribution

level of Square Square of Percentage
(DOF) calculated
X1: Power
2.631 5.392 9.044 2 2.2997 1.1498 72.314 67.45 %
X2: Pressure
4.356 5.064 7.647 2 0.6670 0.3335 20.975 19.56 %
5.766 6.514 4.787 2 0.1669 - - -
X4:Cutting speed
6.780 5.447 4.840 2 0.2191 0.1095 6.886 6.43 %
5.726 5.644 5.697 4 0.0006 - - -
5.497 5.863 5.707 4 0.0077 - - -
5.481 5.896 5.690 4 0.0098 - - -
Error 14 0.2238 0.0159 6.56 %
Total 26 3.4096 100%

%
F) Tabulated 2.73 3.74 6.51
Table 4.7: ANOVA results for the kerf taper (Ta) based on Standard deviation.
Standard deviation by Degree Sum of Mean Value Contribution

factor level of Square Square of Percentage
(dB) Freedom ( SS) (V) F) P%
1 2 3 (DOF) calculated
X1: Power
0.4432 0.8670 1.2244 2 0.03402 0.01701 8.257 44.47 %
X2: Pressure [Pooled Factor]
0.7699 0.7346 1.0301 2 0.00582 - - -
0.7130 0.9499 0.8717 2 0.00327 - - -
X4:Cutting speed
1.0577 0.8315 0.6454 2 0.00951 0.00475 2.306 12.43 %
0.8549 0.8304 0.8493 4 0.00006 - - -
0.8032 0.9331 0.7983 4 0.00133 - - -
0.7391 0.9936 0.8019 4 0.00394 - - -
Error 16 0.03297 0.00206 43.10 %
Total 26 0.0765 100%

%
F) Tabulated 2.67 3.63 6.23
4.2.2. Verification of Experimental Results
Once the optimal level of design parameters has been selected, the final step is
to verify the improvement of quality characteristics using the optimal level of
design parameters. The estimated optimal level of parameters can be calculated
as [1]:
9
Ypredicted =Ymean + ∑[Yi- Ymean] (4.9)
Yi is the mean (S/N ratio, mean, and standard deviation) response at optimal
level for the main design parameters that affect the quality characteristics, and
Y Mean is the total mean (S/N ratio, mean, and standard deviation) response.
Tables 4.8 and 4.9 show the comparison of the predicted and actual the kerf
taper and average surface roughness. Good agreement between the predicted
and actual (S/N ratio and mean) responses is observed. In other words, the
experiments results confirm prior design and analysis for optimizing the
process parameters.
Table 4.8: Results of the confirmation experiment for S/N ratios values.
The kerf taper

Experiment Prediction
Optimal Level X13,X23 X13,X23
The kerf taper -48.893 -47.944
S/N ratio (dB)
Surface roughness
Optimal Level X11 X11
Surface roughness -126.732 -133.565
S/N ratio (dB)
The heat affected zone
The heat affected zone -1387.299 -1386.945
S/N ratio (dB)
Table 4.9: Results of the confirmation experiment for mean values.
The kerf taper

Optimal Level X11,X21,X43 X11,X21,X43
The kerf taper 1.706113 1.673976
Mean values
Surface roughness
Optimal Level X11, X21 X11, X21

Surface roughness 14.546 15.068
Mean values
The heat affected zone
The heat affected zone 0.2831 0.2818
Mean values
4.3. Mathematical Models using Response Surface Methodology (RSM).
RSM attempts to analyze the influence of the independent variables

on a specific response. The purpose of mathematical models relates the process
responses to facilitate the optimization of the process. The mathematical model
commonly used for the process responses is represented as:
Y = F (X1, X2, X3………….( + ε (4.10)
Where X1, X2, X3… Xn are process parameters and ε is the error which is
normally distributed about the observed response Y.
Coefficients of process parameters using RSM is represented as [34]:
[B] = Inverse ([Z]T *[Z])*[Z]T*[F] (4.11)
Where [B]: array of coefficients of process parameters, [Z]: orthogonal array

with values of selected process parameters, [F]: array with values of measured
response, and [Z ]T :transpose array of [Z].
In order to judge the accuracy of the predicted model, Percentage of deviation

Ф i and average percentage of deviation Ф ^ are defined as:
Ф I = [(Absolute[R measured – R predicted]) /(R measured)]* 100 (4.12)
Where Ф i: percentage deviation of single sample data.

R measured: measured response.
R predicted : predicted response.
1
Ф ^ =(∑Ф i)/n (4.13)

where Ф ^: average percentage deviation of all sample data.
n: the size of sample data.
4.3.1. Motivation
This question deals with the issue of how to construct a metric, a statistic that
may be used to ascertain the quality of the fitted model. The statistic should be
such that for one range of values, the implication is that the model is good,
whereas for another range of values, the implication is that the model gives a
poor fit [35].
Sum of absolute residuals
Since a model's adequacy is inversely related to the size of its residuals, one
obvious statistic is the sum of the absolute residuals.
�� = ∑��=�|��| (4.14)
Average absolute residual

A better metric that does not change (much) with increasing sample size is the
average absolute residual:
∑�
�=�|��|
�� = (4.15)
�
With n denoting the number of response values. Again, small values for this
statistic imply better-fitting models.
4.3.2. Mathematical Models for the kerf taper (Ta)
A mathematical model is developed for the kerf taper based on S/N ratio as in
equation 4.12. Figure 4.1 gives the measured vs. predicted the kerf taper based
on S/N ratio.
Ta1 =12.3339 – 0.002X1 – 2.2722X2 + 0.0499X3 – 0.0304X4 – 0.0073X1.X2 –

0.0001X1.X3 (4.16)
15
Experimental result
Predicted result
10
Kerf Taper (dB)
-5
0 5 10 15 20 25 30
Experiment Number
Figure 4.1: Measured Vs. Predicted S/N ratio response (Ta)
Figures 4.2, 4.3, 4.4, 4.5, 4.6, and 4.7 show the estimated three-dimensional
surface and contour plots for the kerf taper response values (dB) as function of
X 1, X 2, X 3,X4, X 1. X 2 and X1.X3. In all these figures, two of the four
variables are hold constant at center points.
Table 4.10 gives the corresponding measured vs Predicted S/N ratios. The
model deviation varies from 0.21% to 99.93%.
Table 4.10: Measured versus predicted S/N ratio response for the kerf taper
(Ta) (Model Validation).
Run Measured response S/N Predicted response S/N deviation %

# ratio (dB) ratio (dB)
1 9.8603 10.9188 10.73
2 12.3532 10.9188 11.61
3 11.3509 10.9188 3.80
4 12.9214 10.6762 17.37
5 9.2493 10.6762 15.42
6 10.4351 10.6762 2.31
7 13.3913 10.1296 24.35
8 7.5235 10.1296 34.63
9 9.8603 10.1296 2.73
10 6.6709 7.8068 17.02
11 7.7902 7.8068 0.21
12 6.2057 7.8068 25.80
13 4.9761 6.1852 24.29
14 5.1065 6.1852 21.12
15 5.0929 6.1852 21.44
16 1.3067 2.6126 99.93
17 2.2961 2.6126 13.78
18 2.1426 2.6126 21.93
19 2.9294 3.2148 9.74
20 2.4879 3.2148 29.21
21 2.4754 3.2148 29.86
22 0.8460 1.2432 46.95
23 1.4632 1.2432 15.03
24 1.8256 1.2432 31.90
25 -3.5891 -2.9734 17.15
26 -2.6580 -2.9734 11.86
27 -3.3285 -2.9734 10.66
Average Residual 570.83
Average Absolute Residual 21.14
Average percentage of Model accuracy % 78.86 %
Figure 4.2: Effect of power and pressure on Ta at constant frequency and

cutting speed based on S/N ratio
Figure 4.3: Effect of power and frequancy on Ta at constant pressure and

cutting speed based on S/N ratio
Figure 4.4: Effect of power and cutting speed on Ta at constant pressure and
frequency based on S/N ratio
Figure 4.5: Effect of pressure and frequency on Ta at constant power and

cutting speed based on S/N ratio.
Figure 4.6: Effect of pressure and cutting speed on Ta at constant power and
frequency based on S/N ratio.
Figure 4.7: Effect of frequency and cutting speed on Ta at constant power and
pressure based on S/N ratio.
A mathematical model is developed for the kerf taper based on standard

deviation as in equation 4.13. Figure 4.9 gives the measured vs. predicted
Average surface roughness based on standard deviation over the 27
experiments.
Ta2= 0.0132+0.0001X1+ 0.0280X2+0.0002X3–0.0015X4+0.0062 (X2)2
(4.17)
0.25
Standard deviation response
0.20
0.15
0.10
0.05
Experimental
Predicted
0.00
0 5 10 15 20 25 30
Experiment Number
Figure 4.8: Measured Vs. Predicted standard deviation response (Ta).
Table 4.11 gives the corresponding measured vs. predicted mean values. The
model deviation varies from 0.0009 % to 69.58 %. Figures 4.10, 4.11,
4.12,4.13,4.14and 4.15 show the estimated three-dimensional surface plots
for the kerf taper standard deviation response values as function of X 1, X2,
X3 and X 4 In all these figures, two of the three variables are hold constant at
center points.
Table 4.11: Measured versus predicted Standard Deviation response for the kerf
taper (Ta) (Model Validation).
Run Measured response Predicted response

# (Standard Deviation (Standard Deviation deviation %
values) values)
1 0.036056 0.03375 6.39
2 0.056862 0.03840 32.46
3 0.056862 0.04615 18.83
4 0.045092 0.0474 5.11
5 0.070000 0.05515 21.21
6 0.026458 0.04275 61.57
7 0.020817 0.04915 136.1
8 0.095044 0.03675 61.33
9 0.036056 0.04140 14.82
10 0.100167 0.03340 66.65
11 0.037859 0.04115 8.69
12 0.094516 0.02875 69.58
13 0.110151 0.11015 0.0009
14 0.095394 0.09775 2.46
15 0.126623 0.10240 19.13
16 0.097125 0.07675 20.97
17 0.100167 0.08915 10.99
18 0.105040 0.08915 15.12
19 0.147422 0.11115 24.6
20 0.130512 0.09875 24.33
21 0.109697 0.10340 5.74
22 0.037859 0.06275 65.74
23 0.064291 0.06740 4.83
24 0.158745 0.07515 52.65
25 0.260256 0.13640 47.59
26 0.180278 0.14415 20.04
27 0.135277 0.13175 2.60
Average Residual (AR) 819.53
Average Absolute Residual (ARR) 30.35
9
Figure 4.9: Effect of power and pressure on Ta at constant frequency and

cutting speed based on standard deviation.
Figure 4.10: Effect of power and frequency on Ta at constant pressure and

Figure 4.11: Effect of power and cutting speed on Ta at constant pressure and
frequency based on standard deviation
Figure 4.12: Effect of pressure and frequency on Ta at constant power and

1
Figure 4.13: Effect of pressure and cutting speed on Ta at constant power and
frequency based on standard deviation.
Figure 4.14: Effect of frequency and cutting speed on Ta at constant power and
pressure based on standard deviation.
A mathematical model is developed for the kerf taper based on Mean values as
in equation 4.14. Figure 4.17 gives the measured vs. predicted the kerf taper
based on Mean values.
Ta3= - 0.057+0.0012X1+0.3657X2–0.0011X3– 0.00664X4+0.0091 (X2)2

(4.18)
1.6
1.4
1.2
Mean response
1.0
0.8
0.6
0.4
Experimental
Predicted
0.2
0 5 10 15 20 25 30
Experiment Number
Figure 4.15: Measured Vs. Predicted Mean response (Ta).
Table 4.12 gives the corresponding measured vs. predicted mean values. The
model deviation varies from 2.04% to 33.74%. Figures 4.18,
4.19,4.20,4.21,4.22and 4.23 show the estimated three-dimensional surface
plots for the kerf taper mean response values as function of X 1, X 2 ,X3and
X4In all these figures, two of the three variables are hold constant at center
points.
Table 4.12: Measured versus predicted Mean response for the kerf taper (Ta)
(Model Validation).
Run Measured response Predicted response deviation %

# (Mean values) (Mean values)
1 0.320 0.214225 33.05
2 0.236 0.221050 6.33
3 0.266 0.232425 12.62
4 0.223 0.275675 23.62
5 0.340 0.282500 16.91
6 0.300 0.293875 2.04
7 0.213 0.270725 27.1
8 0.413 0.277550 32.79
9 0.320 0.288925 9.71
10 0.456 0.326850 28.32
11 0.406 0.338225 16.69
12 0.483 0.320025 33.74
13 0.556 0.653900 17.6
14 0.550 0.665275 20.95
15 0.546 0.647075 18.51
16 0.856 0.880350 2.84
17 0.763 0.891725 16.87
18 0.776 0.873525 12.56
19 0.700 0.776025 10.86
20 0.743 0.757825 1.99
21 0.746 0.764650 2.49
22 0.906 0.936075 3.31
23 0.843 0.917875 8.88
24 0.800 0.924700 15.58
25 1.496 1.263125 15.56
26 1.350 1.244925 7.78
27 1.460 1.251750 14.26
Figure 4.16: Effect power and pressure on Ta at constant cutting speed based on
mean.
Figure 4.17: Effect power and frequency on Ta at constant pressure and cutting
speed based on mean.
Figure 4.18: Effect power and cutting speed on Ta at constant pressure and
frequency based on mean.
Figure 4.19: Effect pressure and frequency on Ta at constant power and cutting
speed based on mean.
Figure 4.20: Effect pressure and cutting speed on Ta at constant power and
frequency based on mean.
Figure 4.21: Effect frequency and cutting speed on Ta at constant power and
pressure based on mean.
4.3.3. Mathematical models for Average Surface Roughness (Ra).
A mathematical model is developed for the average surface roughness based on

S/N ratio as in equation 4.15. Figure 4.19 shows the measured vs. predicted
average surface roughness based on mean over the 27 experiments.
Ra1 = – 8.98366 – 0.0073X1 – 0.54769X2 –0.00593X3 – 0.04662X4 –

0.00314X1X2 + 0.000025X1X4 (4.19)
-10
Experimental
Predicted
-12
-14
Ra (dB)
-16
-18
-20
0 5 10 15 20 25 30
Experiment Number
Figure 4.22: Measured Vs. Predicted S/N ratio response (Ra).
Table 4.13 gives the corresponding measured vs. predicted S/N ratio response.
The model deviation varies from 0.30 % to 7.22%. Figures 4.20,
plots for the kerf taper mean response values as function of X1, X 2,X3 ,X4and
X1.X2 In all these figures, two of the four variables are hold constant at center
points.
Table 4.13: Measured versus predicted S/N ratio response for the average
surface roughness (Ra) (Model Validation).

1 -10.5024 -11.1579 6.24
2 -11.8799 -11.1579 6.07
3 -10.4818 -11.1579 6.45
4 -12.0871 -12.3929 2.52
5 -12.4303 -12.3929 0.30
6 -12.7828 -12.3929 3.05
7 -13.1096 -14.0570 7.22
8 -13.8136 -14.0570 1.76
9 -14.6890 -14.0570 4.30
10 -14.9971 -15.0915 0.62
11 -15.4954 -15.0915 2.60
12 -15.5875 -15.0915 3.18
13 -14.7299 -15.3032 3.89
14 -15.9612 -15.3032 4.12
15 -15.8123 -15.3032 3.21
16 -16.1493 -16.0455 0.64
17 -16.1675 -16.0455 0.75
18 -16.5170 -16.0455 2.85
19 -17.0162 -17.1757 0.93
20 -17.0592 -17.1757 0.68
21 -17.1066 -17.1757 0.40
22 -17.3620 -18.5959 7.10
23 -18.7299 -18.5959 0.71
24 -19.2958 -18.5959 3.62
25 -19.1551 -19.5009 1.80
26 -19.3048 -19.5009 1.01
27 -19.7436 -19.5009 1.22
Average percentage of Model accuracy % 97.14%
9
Figure 4.23: Effect of power and pressure on the average surface roughness at
constant frequency and cutting speed of cut based on S/N ratio.
Figure 4.24: Effect of power and frequency on the average surface roughness at
constant pressure and cutting speed of cut based on S/N ratio.
Figure 4.25: Effect of power and cutting speed on the average surface
roughness at constant pressure and frequency of cut based on S/N ratio.
Figure 4.26: Effect of pressure and frequency on the average surface roughness
at constant power and cutting speed of cut based on S/N ratio.
1
Figure 4.27: Effect of pressure and cutting speed on the average surface
roughness at constant power and frequency of cut based on S/N ratio.
Figure 4.28: Effect of frequency and cutting speed on the average surface
roughness at constant power and pressure of cut based on S/N ratio.
A mathematical model is developed for the Average surface roughness based

on standard deviation as in equation 4.16. Figure 4.27 gives the measured vs.
predicted Average surface roughness based on standard deviation over the 27
experiments.
Ra2= 0.908422 -0.00026X1- 0.16631X2-0.0006X3+0.007159X4–0.1112

(X2)2+0.0000081 (X3)2 (4.20)
1.2
Experimental
Predicted
1.0
0.8
0.6
0.4
0.2
0 5 10 15 20 25 30
Experiment Number
Figure 4.29: Measured Vs. Predicted standard deviation response (Ra).
The model deviation varies from 9.29% to 206.87%. Figures 4.28,
plots for the kerf taper mean response values as function of X 1, X 2,X3 and
X4.In all these figures, two of the four variables are hold constant at center
points.
Table 4.14: Measured versus predicted standard deviation response for the
average surface roughness (Ra) (Model Validation).

values) values)
1 0.85711 0.745385 13.03
2 0.55678 0.620885 11.51
3 0.30551 0.399885 30.89
4 0.80829 0.63941 20.89
5 0.70613 0.51491 27.08
6 0.50895 0.29391 42.25
7 0.30615 0.540615 76.58
8 0.35921 0.416115 15.84
9 0.38188 0.195115 48.90
10 0.50003 0.574925 14.97
11 0.34818 0.394425 13.28
12 0.77730 0.537425 30.86
13 1.10074 0.44023 60.00
14 0.30105 0.25973 13.72
15 0.67550 0.40273 40.38
16 0.41509 0.424255 2.20
17 0.21656 0.243755 12.55
18 0.25325 0.386755 52.71
19 0.35809 0.312565 12.71
20 0.16165 0.496065 206.87
21 0.27501 0.331065 20.38
22 0.40501 0.303770 24.99
23 1.09436 0.487270 55.47
24 0.35529 0.322270 9.29
25 0.29103 0.169075 41.90
26 0.15394 0.352575 129
27 0.26576 0.187575 29.41
constant frequency and cutting speed of cut based on standard deviation.
constant presure and cutting speed of cut based on standard deviation.
roughness at constant presure and frequency of cut based on standard deviation.
at constant power and cutting speed of cut based on standard deviation.
roughness at constant power and frequency of cut based on standard deviation.
roughness at constant power and pressure of cut based on standard deviation.
A mathematical model for the average surface roughness based on mean is

given in equation 4.17 Figure 4.35 shows the measured vs. average surface
roughness based on mean over the 27 experiments.
Ra3 = 1.459249 +0.006813X1+ 1.44X2–0.00138X3+0.007159X4–0.0239

(X2)2+0.0000361 (X3)2 +0.0000386 (X4)2 (4.21)
10
8
Mean response
4 Experimental
Predicted
3
0 5 10 15 20 25 30
Expe rime nt Numbe r
Figure 4.36: Measured and predicted Mean response (Ra).
The model deviation varies from 0.05% to 12.76%. Figures 4.36,
plots for the kerf taper mean response values as function of X 1, X2,X3 and
X4.In all these figures, two of the four variables are hold constant at center
points.
Table 4.15: Measured versus predicted mean response for the average surface
roughness (Ra) (Model Validation).
Run Measured response Predicted response deviation %

# (Mean values) (Mean values)
1 3.27 3.251017 0.58
2 3.90 3.402012 12.76
3 3.33 3.686817 10.71
4 3.96 3.962027 0.05
5 4.14 4.078282 1.49
6 4.33 4.467307 3.17
7 4.51 4.709887 4.43
8 4.89 4.930362 0.82
9 5.41 5.249907 2.95
10 5.60 5.422762 3.16
11 5.94 5.561807 6.36
12 5.98 5.924287 0.93
13 5.37 5.847412 8.89
14 6.27 5.951717 5.07
15 6.15 6.418417 4.36
16 6.41 6.730682 5.00
17 6.43 6.939207 7.91
18 6.69 7.336427 9.66
19 7.08 7.224607 2.04
20 7.12 7.441327 4.51
21 7.16 7.738082 8.07
22 7.37 8.214207 11.45
23 8.59 8.396187 2.25
24 9.21 8.797162 4.48
25 9.07 8.604117 5.13
26 9.23 8.890317 3.68
27 9.70 9.221812 4.92
9
constant frequency and cutting speed based on mean.
constant pressure and cutting speed based on mean.
9
roughness at constant pressure and frequency based on mean.
at constant power and cutting speed based on mean.
91
roughness at constant power and frequency based on mean.
roughness at constant power and pressure based on mean.
9
4.3.4. Mathematical models for the heat affected zone (HAZ).
A mathematical model is developed for the heat affected zone based on S/N
ratio as in equation 4.22. Figure 4.49 shows the measured vs. predicted average
surface roughness based on mean over the 27 experiments.
HAZ1= 24.0619–0.01789X1–3.6727X2+0.001112X3+0.034543X4-0.00000041
(X1)2–0.11025 (X2)2–0.000049 (X3)2-0.00025 (X4)2 (4.22)
20.0
Experimental
17.5 Predicted
15.0
HAZ (dB)
12.5
10.0
7.5
5.0
0 5 10 15 20 25 30
Experiment Number
Figure 4.43: Measured Vs. Predicted S/N ratio response (HAZ)
The model deviation varies from 0.18 % to 24.00 %. Figures 4.50,
plots for the heat affected zone S/N response values as function of X1, X 2,X3
,X4and interactions In all these figures, two of the four variables are hold
constant at center points.
9
Table 4.16: Measured versus predicted S/N ratio response the heat affected
zone (HAZ) (Model Validation).

1 19.78687 19.82286 0.18
2 19.02986 19.34638 1.66
3 17.34231 18.27096 5.35
4 17.51446 18.31254 4.55
5 17.70523 17.61106 0.53
6 16.71954 17.21064 2.93
7 16.47425 16.92265 2.72
8 15.95480 16.89617 5.90
9 16.23058 16.04575 1.13
10 16.20821 15.24367 5.95
11 16.05869 15.00725 6.54
12 16.58651 14.36395 13.39
13 12.43075 12.35163 0.63
14 12.34453 11.89021 3.68
15 11.77690 11.92191 1.23
16 11.61600 10.44951 10.04
17 11.37566 10.66309 6.26
18 11.16965 10.24479 8.28
19 9.56080 9.251197 3.23
20 8.46606 9.004097 6.35
21 7.67834 8.35281 8.78
22 7.33132 7.919507 8.02
23 6.00572 7.447407 24.00
24 6.25903 7.47112 19.36
25 4.35101 4.802467 10.37
26 5.67396 5.005367 11.78
27 4.81772 4.579080 4.95
9
Figure 4.44: Effect of power and pressure on the heat affected zone (HAZ) at
constant frequency and cutting speed of cut based on S/N ratio.
Figure 4.45: Effect of power and frequency on the heat affected zone (HAZ) at
constant pressure and cutting speed of cut based on S/N ratio.
9
Figure 4.46: Effect of power and cutting speed on the heat affected zone (HAZ)
at constant pressure and frequency of cut based on S/N ratio.
Figure 4.47: Effect of pressure and frequency on the heat affected zone (HAZ)
at constant power and cutting speed of cut based on S/N ratio.
9
Figure 4.48: Effect of pressure and cutting speed on the heat affected zone
(HAZ) at constant power and frequency of cut based on S/N ratio.
Figure 4.49: Effect of frequency and cutting speed on the heat affected zone
(HAZ) at constant power and pressure of cut based on S/N ratio.
9
A mathematical model is developed for the heat affected zone based on

standard deviation as in equation 4.23. Figure 4.56 gives the measured vs.
predicted the heat affected zone based on standard deviation over the 27
experiments.
HAZ2= 0.027581- 0.0000913*X1- 0.03379* X2 +0.00014*X3 +0.0000723*X1X2

-0.0000013*X1X3 +0.000000248(X1)2 +0.000000234(X3)2 (4.23)
0.14
0.12
0.10
0.08
0.06
0.04
Experimental
0.02
Predicted
0.00
0 5 10 15 20 25 30
Experiment Number
Figure 4.50: Measured Vs. Predicted standard deviation response (HAZ).
The model deviation varies from 0.86 % to 43.66 %. Figures 4.50,
plots for the heat affected zone standard deviation response values as function
of X1, X 2,X3 ,X4and interactions In all these figures, two of the four variables
are hold constant at center points.
9
Table 4.17: Measured versus predicted standard deviation response the heat
affected zone (HAZ) (Model Validation).

values) values)
1 0.006807 0.008081 18.71
2 0.007095 0.008081 13.89
3 0.007234 0.008081 11.7
4 0.007211 0.005558 22.91
5 0.009644 0.005558 42.36
6 0.009866 0.005558 43.66
7 0.013429 0.014736 9.73
8 0.012288 0.014736 19.92
9 0.012490 0.014736 17.98
10 0.013650 0.015876 16.3
11 0.014107 0.015876 12.53
12 0.014503 0.015876 9.46
13 0.016258 0.016398 0.86
14 0.017436 0.016398 5.95
15 0.023643 0.016398 30.64
16 0.020664 0.025171 21.81
17 0.021572 0.025171 16.68
18 0.033867 0.025171 25.67
19 0.035921 0.041011 14.16
20 0.043062 0.041011 4.76
21 0.047089 0.041011 12.9
22 0.058184 0.099628 71.23
23 0.134515 0.099628 25.93
24 0.110639 0.099628 9.95
25 0.088263 0.079796 9.59
26 0.079103 0.079796 0.87
27 0.071002 0.079796 12.38
99
Figure 4.51: Effect of power and pressure on the heat affected zone at constant
frequency and cutting speed of cut based on standard deviation.
Figure 4.52: Effect of power and frequency on the heat affected zone at
constant presure and cutting speed of cut based on standard deviation.
1
Figure 4.53: Effect of pressure and frequency on the heat affected zone at
constant power and cutting speed of cut based on standard deviation.
1 1
A mathematical model is developed for the heat affected zone based on Mean
as in equation 4.24. Figure 4.56 gives the measured vs. predicted the heat
affected zone based on standard deviation over the 27 experiments.
HAZ3= 0.078062+0.000000622*(X1)2+0.055224*(X2)2–0.00000185*(X3)2-
0.0000248*(X4)2 (4.24)
0.6
0.5
Mean response
0.4
0.3
0.2
Experimental
Predicted
0.1
0 5 10 15 20 25 30
Experiment Number
Figure 4.54: Measured Vs. Predicted Mean response (HAZ).
Table 4.18 gives the corresponding measured vs. predicted Mean response. The
model deviation varies from 0.22 % to 24.83 %. Figures 4.50,
plots for the heat affected zone Mean response values as function of X1,X 2,X3
,X4 and interactions In all these figures, two of the four variables are hold
constant at center points.
1
Table 4.18: Measured versus predicted Mean response the heat affected zone
(HAZ) (Model Validation).

# (Mean values) (Mean values) deviation %
1 0.102 0.102226 0.22
2 0.112 0.102226 8.72
3 0.136 0.102226 24.83
4 0.133 0.126954 4.54
5 0.13 0.126954 2.34
6 0.146 0.126954 13.04
7 0.150 0.147724 1.51
8 0.159 0.147724 7.09
9 0.154 0.147724 4.07
10 0.154 0.167736 8.91
11 0.157 0.167736 6.83
12 0.148 0.167736 13.33
13 0.239 0.227854 4.66
14 0.241 0.227854 5.45
15 0.257 0.227854 11.34
16 0.262 0.317194 21.06
17 0.269 0.317194 17.9
18 0.275 0.317194 15.34
19 0.331 0.402916 21.72
20 0.376 0.402916 7.15
21 0.411 0.402916 1.96
22 0.427 0.442324 3.58
23 0.489 0.442324 9.54
24 0.478 0.442324 7.46
25 0.602 0.539304 10.41
26 0.516 0.539304 4.51
27 0.571 0.539304 5.55
1
Figure 4.55: Effect of power and pressure on the heat affected zone at constant
frequency and cutting speed based on mean.
Figure 4.56: Effect of power and frequency on the heat affected zone at
constant pressure and cutting speed based on mean.
1
Figure 4.57: Effect of power and cutting speed on the heat affected zone at
constant pressure and frequency based on mean.
Figure 4.58: Effect of pressure and frequency on the heat affected zone at
constant power and cutting speed based on mean.
1
Figure 4.59: Effect of pressure and cutting speed on the heat affected zone at
constant power and frequency based on mean.
Figure 4.60: Effect of frequency and cutting speed on the heat affected zone at
constant power and pressure based on mean.
1
Chapter Five: Confirmation Experiments
Chapter 5
CONFIRMATION EXPERIMENTS
5.1. Introduction
In This chapter, Confirmation experiments for the models developed earlier are
carried. Three models are developed for the kerf width, surface roughness and
heat affected zone. The study is carried using to 3 mm stainless steel specimen.
5.2. Experiments
Table 5.1 gives the settings of the confirmation experiments for the laser cutting
process. The five settings are taken at the lower and maximum limits of the
Power, Oxygen Pressure, Frequency and Cutting speed. Three replications are
taken for the Kerf Taper (Degree), Average surface roughness (µ m) and Heat
Affected Zone (mm). The mean, standard deviation and signal-to-noise ratios are
calculated and compared later to prediction models.
Table 5.2 gives a comparison between the surface roughness measurements (µ m)

using Taguchi and RSM approaches .This comparison is gives in terms of the
mean, standard deviation and signal-to-noise ratios. Using the mean as a
measure, the models developed earlier deviate from actual measurements from
– 4.99 % to + 9.32 %.
Using the standard deviation as a measure, the models developed deviate from
actual measurements from –146 % to +769.8 %. Using the S/N ratio as a measure,
the models developed deviate from actual measurements from 1.12 % to 14.776
%. Accordingly, it is recommended to use the developed models to predict the
average and signal to noise ratio of surface roughness.
1
Table 5.3 gives confirmation and prediction results for the kerf taper in degree.
Using the mean as a measure, the developed earlier deviate from the actual
measurements from – 6.450 % to + 2.43 %.
Using the signal to noise ratios as a measure, the models deviate from the actual
measurements from – 105 % to +149 %. Using the standard deviations as a
measure, the models deviate from the actual measurements from – 649 % to
+12.79 %. According, it is recommended to use developed models to predict
mean kerf taper in degree.
Table 5.4 gives confirmation and prediction results for the Heat affected zone.
Using the mean as a measure, the different between the developed and predicted
models vary from – 4.35 % to + 8.24 %. Using the standard deviation as a
measure, the different between the developed and predicted models vary from
– 778 % to + 462 %. Using the S/N ratio as a measure, the different between the
developed and predicted models vary from – 53.4 % to 66.4 %. Accordingly, it
is recommended to use the developed models to predict the average HAZ.
5.3. Conclusion
Ten confirmation experiments are carried to verity models developed previously.

The models developed show good prediction capabilities for the Kerf width,
Surface roughness and Heat affected zone.
1
Table .5.1 Confirmation Experiments.
Average Surface
Experiment
Confirmati
Frequency
Pressure
HAZ (mm)
Cutting
Oxygen
Kerf Taper (deg) Roughness (µm)

Power
Mean
Speed
Mean
Mean
on
1st 2nd 3rd 1st 2nd 3rd 1st 2nd 3rd

Trial Trial Trial Trial Trial Trial Trial Trial Trial
1 150 1.5 75 25 0.431 0.387 0.430 0.416 5.341 4.466 4.476 4.761 0.174 0.190 0.185 0.183
2 250 2.5 125 35 0.813 0.761 0.805 0.793 7.421 6.741 7.219 7.127 0.366 0.388 0.404 0.386
3 350 2.75 175 15 0.965 1.131 1.279 1.125 7.976 8.540 9.374 8.630 0.513 0.449 0.591 0.518
4 450 0.50 65 12 0.486 0.498 0.504 0.496 5.215 4.932 5.816 5.321 0.157 0.214 0.241 0.204
5 550 0.25 190 18 0.402 0.362 0.307 0.357 6.543 7.126 5.609 6.426 0.195 0.174 0.198 0.189
6 200 1 200 40 0.082 0.069 0.071 0.074 6.211 5.310 5.942 5.821 0.052 0.041 0.045 0.046
7 600 3 50 10 1.813 1.671 1.664 1.716 9.784 9.831 9.980 9.865 0.589 0.564 0.605 0.586
8 400 2 125 25 0.913 0.815 0.954 0.894 8.441 7.950 8.632 8.341 0.362 0.410 0.296 0.356
9 200 3 200 10 0.986 1.102 0.984 1.024 9.210 8.124 7.122 8.152 0.511 0.530 0.501 0.514
10 600 1 50 40 0.785 0.689 0.659 0.711 7.320 6.865 7.445 7.210 0.345 0.336 0.342 0.341
1 9
Table .5.2 The confirmation and predict results of the Surface Roughness (Ra)
Experiment
Confirmati
Frequency
Pressure
Average Surface Roughness
Cutting
Oxygen
Average Surface Roughness
Power
Speed
(µm)
on
(µm)
Taguchi RSM
Mean S/N ST.DV Mean S/N ST.DV
1 150 1.5 75 25 4.761 -13.586 0.50232 4.8900 -13.1231 0.549294
2 250 2.5 125 35 7.127 -17.065 0.34921 7.3025 -16.2945 0.034774
3 350 2.75 175 15 8.63 -18.739 0.70333 8.6031 -17.6728 -0.23048
4 450 0.50 65 12 5.321 -14.541 0.45143 5.3934 -14.0588 0.761597
5 550 0.25 190 18 6.426 -16.199 0.76524 6.7472 -15.2856 1.024166
6 200 1 200 40 5.821 -15.318 0.46253 5.7540 -14.4701 1.069272
7 600 3 50 10 9.865 -19.882 0.10233 9.7486 -21.2714 -0.68546
8 400 2 125 25 8.341 -18.429 0.35183 7.5635 -17.1677 0.257539
9 200 3 200 10 8.152 -18.273 1.04428 8.1701 -15.5729 -0.36771
10 600 1 50 40 7.210 -17.164 0.30524 7.3325 -17.3566 0.751522
11
Table 5.3. The confirmation and predict results of the kerf taper (Ta).
Experim
Pressure
Confirm
Frequen
Oxygen
Cutting
Power
Speed
Kerf Taper (deg)
ation
ent Kerf Taper (deg)
cy
Taguchi RSM
1 150 1.5 75 25 0.416 7.608 0.02512 0.4435 8.840 0.0616
2 250 2.5 125 35 0.793 2.011 0.02800 0.8442 3.639 0.1194
3 350 2.75 175 15 1.125 -1.023 0.15708 1.1453 0.510 0.1845
4 450 0.50 65 12 0.496 6.089 0.00917 0.5169 8.609 0.0687
5 550 0.25 190 18 0.357 8.895 0.04769 0.3664 8.145 0.0865
6 200 1 200 40 0.074 22.589 0.00700 0.0722 12.96 0.0474
7 600 3 50 10 1.716 -4.697 0.09276 1.7206 -9.63 0.2080
8 400 2 125 25 0.894 0.955 0.07142 0.8873 1.627 0.1215
9 200 3 200 10 1.024 -0.218 0.06756 1.0756 6.413 0.1980
10 600 1 50 40 0.711 2.938 0.06582 0.7172 2.760 0.0574
111
Table 5.4. The confirmation and predict results of the heat affected zone (HAZ).
Experim
Pressure
Confirm
Frequen
Oxygen
Cutting
Power
Speed
HAZ (mm)
ation
HAZ (mm)
ent
cy
Taguchi RSM
1 150 1.5 75 25 0.183 14.745 0.008185 0.19040 16.12701 -0.01776
2 250 2.5 125 35 0.386 8.261 0.019078 0.40280 9.968842 -0.03850
3 350 2.75 175 15 0.518 5.665 0.071116 0.50967 5.971949 -0.04528
4 450 0.50 65 12 0.204 13.681 0.042884 0.20643 14.30810 0.008152
5 550 0.25 190 18 0.189 14.457 0.013076 0.19484 12.15608 -0.04692
6 200 1 200 40 0.046 26.703 0.005568 0.04448 15.92827 -0.01472
7 600 3 50 10 0.586 4.638 0.020664 0.59036 1.55622 -0.02091
8 400 2 125 25 0.356 8.897 0.057236 0.35407 9.13435 -0.02284
9 200 3 200 10 0.514 4.589 0.014731 0.52347 7.03958 -0.05338
10 600 1 50 40 0.341 9.344 0.004583 0.31290 10.31207 0.040256
11
Chapter Six: Discussion and Analysis of Results
Chapter 6
DISCUSSION AND ANALYSIS OF RESULTS
In This chapter, are related about influences of The kerf taper and surface
roughness and finding the result which factors control parameters like
power ( X1) , pressure (X2) , pulse frequency (X3) cutting speed (X4) is
most important with help of Taguchi method.so we will explain Discussion
and Analysis of Results for this thesis.
6.1. Analysis of variance
6.1.1. Analysis of variance (ANOVA) based on S/N ratio indicates several

results:
 Using the kerf taper and S/N ratio transformation, power and pressure are
significant factors in affecting S/N ratio at 99% confidence level. All
interactions effects: X 1. X 2 and X 1. X3 and are not significantly at any
confidence level.
 Using average surface roughness and S/N ratio transformation, Power is

significant and pressure factors in affecting S/N ratio at 99 % confidence
level. All interactions effects: X1. X 2 and X1. X 4 and are not significantly at
any confidence level.
 Using the heat affected zone and S/N ratio transformation, power and
pressure are significant factors in affecting S/N ratio at 99% confidence
level. All interactions effects: X 1. X 2 and X 1. X3 and are not significantly
at any confidence level.
 It should be noted that mathematical models should include the individual

effects as well as interaction effects (complete models). Some researchers
would include the insignificant effects as well as interaction effects (Meta
11
models), accordingly, it is wise to include all terms (individual and

interaction effect) in the model development phase.
 It is observed from Table 4.8 that X13, X 23 are the optimal levels of the
design parameters for the kerf taper which implies of power at high level
and pressure at high level and this combination gives the best the kerf taper
with in the specified range. On the other hand, it is observed from Table 4.8
that X 11,X21 are the optimal levels of the design parameter for surface
roughness ratio which implies percent of power at low level pressure at low
level, this level give the best surface roughness within the specified range
from the same table that X13, X23 are the optimal levels of the design
parameter for the heat affected zone which implies of power at high level
and pressure at high level and this combination gives the best the heat
affected zone with in the specified range.
6.1.2. Analysis of mean (ANOM) based on mean indicates several results:
 Using the kerf taper as response indicates that. Power, pressure and cutting
speed are factors in affecting mean at 99% confidence level. All interaction
effects are statistically in significant at any confidence level. It is observed
that several parameters are insignificant, although they contribute a good
percentage to the total statistical sum of squares. For instance, cutting speed
(X4) has a contribution of 6.43%. This means that all parameters should be
include in the resulting model.
 Using average surface roughness, Power and pressure are significant factors
in affecting S/N ratio at 99% confidence level. All interactions effects:
X 1. X2 , X 1. X3 and X 1. X 4 and are not significantly at any confidence
level.
 Using the heat affected zone, power and pressure are significant factors in
affecting mean response at 99% confidence level. All interactions effects:
11
X 1. X 2 and X 1. X3 and are not significantly at any confidence level.

 It is observed from Table 4.9 that X11, X 21and X43are the optimal levels of
the design parameters for the kerf taper which implies of power at low
level, pressure at low level and cutting speed at high level and this
combination gives the best the kerf taper with in the specified range. On the
other hand, it is observed from Table 4.9 that X 11,X21 are the optimal levels
of the design parameter for surface roughness ratio which implies of power
at low level and pressure at low level this level give the best surface
roughness within the specified range.
X11,X21 the optimal levels of the design parameter for the heat affected zone
which implies of power at low level and pressure at low level and this gives
the best the heat affected zone with in the specified range.
6.1.3. Analysis of variance (ANOVA) based on standard deviation indicates

several results:
 Using the kerf taper, power is statistically significant at 99 % confidence

level and cutting speed is statistically significant at 90 % confidence level.
Yet statistically insignificant at any confidence level. All interaction terms
are insignificant at any confidence level.
 Using average surface roughness, only pressure is statistically significant at

90 % confidence level. The interaction effect X1. X3 has a contribution of
8.60 %, yet statistically insignificant at any confidence level. Both of power,
frequency and cutting speed and all interaction terms are insignificant at any
confidence level.
 Using the heat affected zone, power is significant response at 99%

confidence level, pressure is significant response at 95% confidence level
and frequency is significant response at 90% confidence level. All
interactions effects are not significantly at any confidence level.
11
 It is observed from Table 4.7 that X11and X43are the optimal levels of the
design parameters for the kerf taper which implies of power at low level
and cutting speed at high level and this combination gives the best the kerf
taper with in the specified range. On the other hand, it is observed from
Table 4.10 that X23 is the optimal level of the design parameter for surface
roughness ratio which implies of pressure at high level this level give the
best surface roughness within the specified range
6.2. Selection of optimum level
6.2.1. The effect of different operating parameters on S/N (Ta, Ra and

HAZ)
The effect of different operating parameters on S/N ratio comprising the kerf
taper is shown in Table 5.1 and Figure D.1 It is clear that, optimum levels of
different control factors for obtaining minimum kerf taper is: cutting speed at
level 1(150 watt), pressure at level 1(0.5 MPa), pulse frequency at level 3
(125 Hz) and cutting speed at level 3 (40 cm/min). Relative contribution of the
controlling parameters on kerf quality is shown in Table 6.1.
Table 6.1: Effect of factor on S/N (Ta)
S/N ratios (dB)

Symbol. Factors Level 1 Level 2 Level 3
X1 Power 2.452a 41.588 96.945
X2 Pressure 26.944a 51.916 62.124
X3 Frequency 53.842 43.697 43.445a
X4 Cutting Speed 55.576 46.244 39.164a
a
optimum level
On other hand The effect of different operating parameters on S/N ratio

comprising the surface roughness is shown in Table 6.2 and Figure D.4It is
clear that, optimum levels of different control factors for obtaining minimum
11
kerf taper is: cutting speed at level 1(150 watt), pressure at level 1(0.5 MPa),
pulse frequency at level 1 (25 Hz) and cutting speed at level 2 (20 cm/min).
Relative contribution of the controlling parameters on kerf quality is shown in
Table 6.2.
Table 6.2: Effect of factor on S/N (Ra)
S/N ratios (dB)

X1 Power 111.777a 141.417 164.773
X2 Pressure 130.126a 139.191 148.650
X3 Frequency 137.086a 141.584 139.297
X4 Cutting Speed 137.571 137.316a 143.080
a
optimum level
The effect of different operating parameters on S/N ratio comprising the heat
affected zone is shown in Table 6.3 and Figure D.7 It is clear that, optimum
levels of different control factors for obtaining minimum kerf taper is: cutting
speed at high level 3(750 watt), pressure at level 3 (1.5 MPa), pulse
frequency at level 1 (25 Hz) and cutting speed at low level (10 cm/min).
Relative contribution of the controlling parameters on kerf quality is shown in
Table 6.3.
Table 6.3: Effect of factor on S/N (HAZ)
S/N ratios (dB)

X1 Power 156.758 119.567 60.144a
X2 Pressure 130.718 108.087 97.664a
X3 Frequency 109.916a 115.635 110.918
X4 Cutting Speed 107.554a 111.806 117.109
a
optimum level
11
6.2.2. The effect of different operating parameters on Mean (Ta, Ra and

HAZ)
The effect of different operating parameters on Mean comprising the kerf taper
is shown in Table 6.4 and Figure D.3It is clear that, optimum levels of different
control factors for obtaining minimum kerf taper is: cutting speed at level 1(150
watt), pressure at low level (0.5 MPa), pulse frequency at high level (125 Hz)
and cutting speed at high level (40 cm/min). Relative contribution of the
controlling parameters on kerf quality is shown in Table 6.4.
Table 6.4: Effect of factor on Mean (Ta)
S/N ratios (dB)

X1 Power 2.631a 5.392 9.044
X2 Pressure 4.356a 5.064 7.647
X3 Frequency 5.766 6.514 4.787a
X4 Cutting Speed 6.780 5.447 4.840a
a
optimum level
The effect of different operating parameters on mean comprising the surface

roughness is shown in Table 6.5 and Figure D.6 It is clear that, optimum levels
of different control factors for obtaining minimum kerf taper is: cutting speed at
low level (150 watt), pressure at low level (0.5 MPa), pulse frequency at high
level (125 Hz) and cutting speed at level 2 (20 cm/min). Relative contribution
of the controlling parameters on kerf quality is shown in Table 6.5.
Table 6.5: Effect of factor on Mean (Ra)
S/N ratios (dB)

X1 Power 37.74a 54.84 74.53
X2 Pressure 49.38a 55.39 62.34
X3 Frequency 55.20 57.95 53.96a
11
X4 Cutting Speed 56.29 53.32a 57.50

a
optimum level
The effect of different operating parameters on mean comprising the heat

affected zone is shown in Table 6.6 and Figure D.9 It is clear that, optimum
levels of different control factors for obtaining minimum kerf taper is: cutting
speed at low level (150 watt), pressure at low level (0.5 MPa), pulse frequency
at high level (125 Hz) and cutting speed at level 3 (40 cm/min). Relative
contribution of the controlling parameters on kerf quality is shown in Table 6.6.
Table 6.6: Effect of factor on Mean (HAZ)
S/N ratios (dB)

X1 Power 1.222a 2.002 4.201
X2 Pressure 1.927a 2.540 2.958
X3 Frequency 2.550 2.557 2.318a
X4 Cutting Speed 2.776 2.333 2.316a
a
optimum level
6.3. Discussion of Validation

Validation of the predicted results with the experimental results is done in
order to conform the predicted results to be acceptable for practical use.
Percentage of prediction error shows the amount of variation with the actual
experimental results.
Validation of RSM models indicates average percentage deviations of the kerf

taper, surface roughness and the heat affected zone based on S/N ratio values
are 21.14, 2.86 and 6.58 respectively. Table 6.7 indicates average percentage
accuracy in the kerf taper, surface roughness and the heat affected zone based
on S/N ratio values are 78.86%, 97.14% and 93.42 %respectively.
119
Table 6.7: Comparison of experimental and predicted (Ta& Ra& HAZ) based
on S/N ratio.
Deviation %
The kerf taper Surface The Heat affected
Roughness zone
Average 570.83 77.24 177.79
Residual
Average
Absolute 21.14 2.86 6.58
Residual
Average
percentage of 78.86% 97.14 % 93.42 %
Model accuracy
%
Validation of RSM models indicates average percentage deviations of the kerf

taper, surface roughness and the heat affected zone based on Mean values are
15.29, 4.99 and 9.00 respectively. Table 6.8 indicates average percentage
accuracy in the kerf taper, surface roughness and the heat affected zone based
on S/N ratio values are 84.71 %, 95.01 % and 91.00 %respectively.
Table 6.8: Comparison of experimental and predicted (Ta& Ra &HAZ) based

on Mean.
Deviation %
Surface The Heat affected

The kerf taper Roughness zone
Average 412.96 134.83 243.06
Residual
Average
Absolute 15.29 4.99 9.00
Residual
Average
percentage of 84.71 % 95.01 % 91.00 %
Model accuracy
%
1
Chapter Seven: Conclusions and Recommendations for Future Research
Chapter 7
CONCLUSIONS AND RECOMMENDATIONS FOR FUTURE
RESEARCH
7.1. Conclusions
In this thesis, a L27OA (3-levels, 4 variables with interactions) is employed to

plan experimentation of the LBM process. This is equivalent to 34 = 81 plus the
interactions. The use of L27OA is equivalent to 27/81 = 1/9 of experiments
required based on Taguchi design was performed for Matlab software was
used for analysis the result and theses responses were partially validated
experimentally .Any higher orthogonal array such as L81OA and L243OA
could have been used stainless steel used for experimentation. Any material
can be used instead; however the importance of the chosen stainless steel is
justified by the increasing applications. The models for the kerf taper and
average surface roughness are specific for the chosen material and range for the
control variables. The results are concluded as following:
 Experimental results for both S/N ratio and mean (real) response values
show that power and pressure are the significant parameters among all
controllable factors that influence the kerf taper at confidence level 99%.
On the other hand, power, and pressure are the significant parameters
among all controllable factors that influence the heat affected zone
ratio at confidence level 99%. Finally, experimental results show that
power and pressure are the main parameters among all controllable factors
that influence average surface roughness at confidence level 99%.
 Optimum levels recommended obtaining better the kerf taper for the
specific range are power (150 watt), low level of pressure (0.5 MPa) while
optimum levels for the heat affected zone ratio are power (750watt), and
high level of pressure (1.5 MPa). Optimum levels for average surface
roughness are power at low level (150watt), low level of pressure (0.5
MPa) and high level of depth of cut (1mm).
 Validation of RSM models indicates average percentage deviation in the

kerf taper, surface roughness and the heat affected zone ratio based on
S/N ratio values are 21.14 %, 2.86 %, and 6.58 % respectively. Based on
mean values, the average percentage of deviation are 15.29 %, 4.99%, and
121
Chapter Seven: Conclusions and Recommendations for Future Research
9.00% respectively. Based on standard deviations, the average percentage

of deviation in the kerf taper, surface roughness and the heat affected zone
ratio are 30.35%, 39.17 %and 18.61 % respectively.
 The accuracy of these models can be enhanced by increasing the number

of trials over large domains of process variables. More interactions can be
included in model development.
 Optimization of RSM models is presented. Optimum process settings are

known to minimize the kerf taper (Ta), minimize average surface
roughness (Ra) and minimize the heat affected zone (HAZ) respectively
or as a combined multi-objective optimization problem with equal and
different weights.
 The models have been found to be accurately representing both to the kerf
taper, surface roughness and heat affected zone values with respect to
experimental results.
7.2. Future Research
 Other LBM process parameter such as: Focal length, Focal point position,
Duty cycle % and different type's gases. In addition, several quality
responses can be measured such as Material Removal Rate (MRR).
 Other 2- level, 3-level and mixed type orthogonal arrays can be used to
study LBM process. Cost of carrying experimentation can be used as an
objective besides already studied objectives.
 The machine condition needs further elaboration in relation to process

variables range. Noise related to the machine employed requires further
quantification.
122
References
References
[1] Park, S.H, Robust Design and Analysis for Quality Engineering, 2nd
edition, Chapman Hall, London (1996).
[2] Zhang, J.Z., Chenb, J.C., Kirby, E.D. Surface roughness optimization
in an end-milling operation using the taguchi design method. Journal of
Materials Processing Technology,vol. 184, p. 233-239,(2007).
[3] Raghu N. Kacker, Eric S. Lagergen, and James J. Filliben, “ Taguchi’s

Orthogonal Arrays Are Classical Designs of Experiments”, Journal of Research
of the National Institute of Standards and Technology, Vol. 96, 1991, pp. 577-
591,(1991).
[4] Kai Yang, and Basem.S.EL-Haik, Design for Six Sigma, 2nd edition,
McGraw-Hill, (2003).
[5] Gadallah, M.H, “Modeling and Synthesis Using Response Surface

Methodology: a Comparative Study", Int. J. Experimental Design and Process
Optimization, Vol. 1, No.2/3, 2009, pp.202 – 239,(2009).
[6] Rofin Introduction to industrial laser materials processing. http://

www.obrusn.torun.pl/htm0/prod_images/rofin/Laserbook.pdf. Accessed 23
August 2012, (2004).
[7] European Committee for Standardization, “Thermal cutting – Classification

of thermal cuts, Geometrical product specification and quality tolerances”, EN
ISO 9013, (2002).
[8] N. Beier, D.Ditzinger: Laser strahlschneiden oder Stanzen – Kriterien für

den Entscheidungsprozess, Int. Conf. Cutting Technology pp. 25–30, in
German,(2002).
[9]J.K.S. Sundar, S.V. Joshi, Laser cutting of materials, Centre for Laser
Processing of Materials, International Advance Research Centre for Powder
Metallurgy and New Materials, Hyderabad,(2005).
[10] http://www.photonics.com/EDU/Handbook.aspx?AID=25155
[11] http://www.buzzle.com/articles/advantages-and-disadvantages-of-laser-
cutting.html
1
References
[12] J. Beddoes and J. G. Parr, Introduction to Stainless Steels, ASM

International, Materials Park, OH,(1999).
[13] http://www.aisi-stainless.com/Selling-list/316L-Stainless-Steel-supplier-
316L-stainless-steel-Applications.html
[14] Rajesh Kumar, Ritual Chandrakar, Anil Kumar, Haldhar ram chandrakar,
“Taguchi Loss Function Optimized Model for Supplier Selection and
Evaluation”, International Journal of Advanced Engineering Technology,
Vol.III, pp. 268-270,(2012).
[15] N. Tosun and L. Ozler, ‘Optimization for hot turning operations with
multiple performance characteristics’, Int. J. Adv. Manuf. Technol. Vol. 23, pp.
777-782,(2004).
[16] I.A. Almeida, W.de. Rossi, M.S.F. Lima, J.R. Berretta, G.E.C Nogueira,
N.U. Wetter, and N.D.Vieira Jr, ‘Optimization of titanium cutting by factorial
analysis of the pulsed Nd: YAG laser parameters’, J. Mater. Process. Technol.
Vol. 179, pp. 105-110, (2006).
[17] Tam SC, Yeo CY, Jana S, Lau WS, Lim EN, Yang LJ Optimization of
laser deep hole drilling of Inconel 718 using the Taguchi method. J Mater
Process Technol, (1993).
[18] Chen YH, Tam SC, Chen WL, Zheng HY Application of Taguchi method
in the optimization of laser micro-engraving of photomasks. Int J Mater Prod
Technol (1996).
[19] Dubey AK, Yadava V, Multi-objective optimization of Nd:YAG laser

cutting of nickel-based superalloy sheet using orthogonal array with principal
component analysis. Opt Lasers Eng, (2008).
[20] C.H. Li, M.J. Tsai, and C.D.Yang, ‘Study of optimal laser parameters for
cutting QFN packages by Taguchi’s matrix method Opt. Laser Technol. Vol.
39, pp. 786-795, (2007).
[21] Sharma, A. and Yadava,V, “Modelling and optimization of cut quality

during pulsed Nd:YAG laser cutting of thin Al alloy sheet for straight profile”,
Optics & Laser Technology,(2012).
1
References
[22] L. Li, C. Diver, J. Atkinson, R.G. Wagner, H.J. Helml, Sequential laser
and EDM micro-drilling for next generation fuel injection nozzle manufacture,
Annals of CIRP 55 (1) 179–182, (2006).
[23] A. Stephen, G. Sepold, S. Metev, F. Vollertsen, Laser-induced liquid-

phase jet-chemical etching of metals, Journal of Material Processing
Technology 149 (1–3) 536–540, (2004).
[24] J.M. Prusa, G. Venkitachalam, P.A. Molian, Estimation of heat conduction

losses in laser cutting, International Journal of Machine Tools and Manufacture
39 431–458, (1999).
[25] S.C. Tam, L.E.N. Lim, K.Y. Quek, Application of Taguchi method in the
optimization of the laser-cutting process, Journal of Materials Processing
Technology 29 –63–74, (1992).
[26] Hügel H. New solid-state lasers and their application potentials, Optics
and Lasers in Engineering, 34, pp. 213-229, (2000).
[27] P. Ross, Taguchi Techniques for Quality Engineering-Loss Function,

Orthogonal Experiments, Parameter and Tolerance Design, McGraw-Hill, New
York, (1988).
[28] Ion, J.C. Laser Processing of Engineering Materials: Principles, Procedure

and Industrial Application. Elsevier Butterwort-Heinemann, (2005).
[29] Karatas -C, Keles-O ,Usta Y. Laser cutting of steel sheets: influence of
workpiece thickness and beam waist position on the kerf width and stria
formation. J Mater Process Technol; 172:22–9, (2006).
[30] Salem HG, Mansour MS, Badr Y, Abbas WA. CW Nd:YAG laser cutting
of ultra-low carbon steel thin sheets using O2 assist gas.J Mater Process
Technol;196(1–3):64–72,( 2008).
[31] Yu LM. Three-dimensional finite element modelling of laser cutting J

Mater Process Technol; 63:637–9, ( 1997).
[32] Avanish Kumar Dubey, Vinod Yadava .Optimization of kerf quality

during pulsed laser cutting of aluminium alloy sheet. Journal of materials
processing technology 204412–418, (2008).
1
References
[33] Bilkent University web site, May 2013,

http://www.bilkent.edu.tr/~berument/f.pdf
[34] Phadke, M.S, Quality Engineering Using Design of Experiment, Quality

Control, Robust Design and The Taguchi Method, 2nd edition, Wadsworth &
Books, California, (1988).
[35] http://www.itl.nist.gov/div898/handbook/pri/section5/pri5992.htm
[36] Genichi Taguchi, Yu-in Wu, “Off-line Quality Control”. Central Japan
Quality Control Association, Nagaya, (1979).
1
Appendix A
Appendix A: Two-Level and Three-Level Orthogonal Arrays

Table.A.1: L4 Array (23) [36].
Colum no.
Trial no. 1 2 3
1 1 1 1
2 1 2 2
3 2 2 2
4 2 2 1
Colum no.
Trial no. 1 2 3 4 5 6 7
1 1 1 1 1 1 1 1
2 1 1 1 2 2 2 2
3 1 2 2 1 1 2 2
4 1 2 2 2 2 1 1
5 2 1 2 2 2 1 2
6 2 1 2 1 1 2 1
7 2 2 1 2 2 2 1
8 2 2 1 1 1 1 2
1
Appendix A
Table.A.3: L12 Array (211) [36].
Trial Colum no.

no
1 2 3 4 5 6 7 8 9 10 11
1 1 1 1 1 1 1 1 1 1 1 1
2 1 1 1 1 1 2 2 2 2 2 2
3 1 1 2 2 2 1 1 1 2 2 2
4 1 2 1 2 2 1 2 2 1 1 2
5 1 2 2 1 2 2 1 2 1 2 1
6 1 2 2 2 1 2 2 1 2 1 1
7 2 1 2 2 1 1 2 2 1 2 1
8 2 1 2 1 2 2 2 1 1 1 2
9 2 1 1 2 2 2 1 2 2 1 1
10 2 2 2 1 1 1 1 2 2 1 2
11 2 2 1 2 1 2 1 1 1 2 2
12 2 2 1 1 2 1 2 1 1 2 1
Table.A.4: L16 Array (215) [36].
Trial Colum no.

no
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
2 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2
3 1 1 1 2 2 2 2 1 1 1 1 2 2 2 2
4 1 1 1 2 2 2 2 2 2 2 2 1 1 1 1
5 1 2 2 1 1 2 2 1 1 2 2 1 1 2 2
1
Appendix A
6 1 2 2 1 1 2 2 2 2 1 1 2 2 1 1
7 1 2 2 2 2 1 1 1 1 2 2 2 2 1 1
8 1 2 2 2 2 1 1 2 2 1 1 1 1 2 2
9 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2
10 2 1 2 1 2 1 2 2 1 2 1 2 1 2 1
11 2 1 2 2 1 2 1 1 2 1 2 2 1 2 1
12 2 1 2 2 1 2 1 2 1 2 1 1 2 1 2
13 2 2 1 1 2 2 1 1 2 2 1 1 2 2 1
14 2 2 1 1 2 2 1 2 1 1 2 2 1 1 2
15 2 2 1 2 1 1 2 1 2 2 1 2 1 1 2
16 2 2 1 2 1 1 2 2 1 1 2 1 2 2 1
Colum no.
Trial no 1 2 3 4
1 1 1 1 1
2 1 2 2 2
3 1 3 3 3
4 2 1 2 3
5 2 2 3 1
6 2 3 1 2
7 3 1 3 2
8 3 2 1 3
9 3 3 2 1
1 9
Appendix A
Table.A.6: L27 Array (313) [36].
Trial Colum no.

No. 1 2 3 4 5 6 7 8 9 10 11 12 13
1 1 1 1 1 1 1 1 1 1 1 1 1 1
2 1 1 1 1 2 2 2 2 2 2 2 2 2
3 1 1 1 1 3 3 3 3 3 3 3 3 3
4 1 2 2 2 1 1 1 2 2 2 3 3 3
5 1 2 2 2 2 2 2 3 3 3 1 1 1
6 1 2 2 2 3 3 3 1 1 1 2 2 2
7 1 3 3 3 1 1 1 3 3 3 2 2 2
8 1 3 3 3 2 2 2 1 1 1 3 3 3
9 1 3 3 3 3 3 3 2 2 2 1 1 1
10 2 1 2 3 1 2 3 1 2 3 1 2 3
11 2 1 2 3 2 3 1 2 3 1 2 3 1
12 2 1 2 3 3 1 2 3 1 2 3 1 2
13 2 2 3 1 1 2 3 2 3 1 3 1 2
14 2 2 3 1 2 3 1 3 1 2 1 2 3
15 2 2 3 1 3 1 2 1 2 3 2 3 1
16 2 3 1 2 1 2 3 3 1 2 2 3 1
17 2 3 1 2 2 3 1 1 2 3 3 1 2
18 2 3 1 2 3 1 2 2 3 1 1 2 3
19 3 1 3 2 1 3 2 1 3 2 1 3 2
20 3 1 3 2 2 1 3 2 1 3 2 1 3
21 3 1 3 2 3 2 1 3 2 1 3 2 1
22 3 2 1 3 1 3 2 2 1 3 3 2 1
23 3 2 1 3 2 1 3 3 2 1 1 3 2
24 3 2 1 3 3 2 1 1 3 2 2 1 3
25 3 3 2 1 1 3 2 3 2 1 2 1 3
26 3 3 2 1 2 1 3 1 3 2 3 2 1
27 3 3 2 1 3 2 1 2 1 3 1 3 2
1
Appendix B
Appendix B: ANOM Results
Table B.1: ANOM for Average surface roughness (Ra) results.

Freedom ( SS) (V) F) P%
1 2 3 (DOF) calculated
X1: Power
37.74 54.84 74.53 2 75.319 37.659 114.69 83.76 %
X2: Pressure
49.38 55.39 62.34 2 9.348 4.674 14.23 10.39 %
55.20 57.95 53.96 2 0.927 - - -
56.29 53.32 57.50 2 1.028 - - -
52.64 56.51 57.96 4 1.681 - - -
56.51 55.50 55.10 4 0.117 - - -
56.45 55.27 55.39 4 0.094 - - -
Error 16 5.256 0.3285 5.85 %
Total 26 89.923 100%

%
F) Tabulated 2.67 3.63 6.23
1 1
Appendix B
Table B.2: ANOM for The heat affected zone (HAZ) results.

(DOF) calculated
X1: Power
1.222 2.002 4.201 2 0.5303 0.2652 139.58 85.59 %
X2: Pressure
1.927 2.540 2.958 2 0.0597 0.0299 15.74 9.64 %
2.550 2.557 2.318 2 0.0041 - - -
X4:Cutting speed[Pooled Factor]
2.776 2.333 2.316 2 0.0151 - - -
2.400 2.449 2.576 4 0.0018 - - -
2.446 2.516 2.463 4 0.0003 - - -
2.512 2.441 2.472 4 0.00026 - - -
Error 16 0.0296 0.0019 4.77 %
Total 26 0.6196 100%
Values of F) Tabulated at DOF) Factor = 2 and DOF) Error = 16 are

%
F) Tabulated 2.67 3.63 6.23
1
Appendix C
Appendix C: ANOVA Results
Table C.1: ANOVA results for the Average surface roughness (Ra) based on
S/N ratios.

of Square Square of
Freedom ( SS) (V) F) Percentage
1 2 3 (DOF) P%
calculated
X1: Power
111.777 141.417 164.773 2 156.764 78.382 130.35 84.53 %
X2: Pressure
130.126 139.191 148.650 2 19.067 9.534 15.86 10.28 %
137.086 141.584 139.297 2 1.125 - - -
X4:Cutting speed[Pooled Factor]
137.571 137.316 143.080 2 2.357 - - -
135.109 140.842 142.016 4 3.036 - - -
138.710 140.146 139.111 4 0.123 - - -
139.469 139.574 138.924 4 0.028 - - -
Error 16 9.621 0.6013 5.19 %
Total 26 185.452 100%

%
F) Tabulated 2.67 3.63 6.23
1
Appendix C
Table C.2: ANOVA results the heat affected zone (HAZ) based on S/N ratios.

of Square Square of Percentage
(DOF) calculated
X1: Power
156.758 119.567 60.144 2 527.670 263.835 268.39 87.03 %
X2: Pressure
130.718 108.087 97.664 2 62.938 31.469 32.01 10.38 %
109.916 115.635 110.918 2 2.073 - - -
107.554 111.806 117.109 2 5.093 - - -
115.274 112.615 108.580 4 2.525 - - -
113.454 111.700 111.315 4 0.289 - - -
112.310 113.466 110.693 4 0.432 - - -
Error 16 15.728 0.983 2.59 %
Total 26 606.336 100%

%
F) Tabulated 2.67 3.63 6.23
1
Appendix C
Table C.3: ANOVA results for the Average surface roughness (Ra) based on
Standard deviation.
Standard deviation by Degree Sum of Mean Value Contributio

factor level of Square Square of n
(dB) Freedom ( SS) (V) F) Percentage
1 2 3 (DOF) calculated P%
X1: Power [Pooled Factor]
4.1556 4.4758 2.9509 2 0.1437 - - -
X2: Pressure
3.5720 5.5978 2.4125 2 0.5776 0.28880 2.723 25.59 %
3.8093 4.3597 3.4133 2 0.0502 - - -
3.9915 3.5394 4.0514 2 0.0174 - - -
4.8111 3.3819 3.3890 4 0.1505 - - -
X1.X3: Interaction Factor
4.8572 3.7208 3.0043 4 0.1940 0.0485 0.457 8.60 %
2.9817 3.9817 4.6222 4 0.1518
Error 14 1.4849 0.10606 65.81 %
Total 26 2.2565 100%

%
F) Tabulated 2.73 3.74 6.51
1
Appendix C
Table C.4: ANOVA results for the heat affected zone (HAZ) based on
Standard deviation.
Standard deviation by Degree Sum of Mean Value of Contribution

factor level of Square Square F)
(dB) Freedom ( SS) calculated Percentage
1 2 3 (DOF) (V) P%
X1: Power
0.0861 0.1757 0.6677 2 0.0218 0.01090 32.06 70.78 %
X2: Pressure
0.1895 0.3874 0.3526 2 0.0025 0.00125 3.68 8.12 %
X3: Frequency
0.4006 0.3073 0.2216 2 0.0018 0.0009 2.65 5.84 %
0.3168 0.2289 0.3838 2 0.0013 - - -
0.2604 0.3388 0.3303 4 0.0004 - - -
0.3561 0.3083 0.2651 4 0.0005 - - -
0.3093 0.2834 0.3368 4 0.0002 - - -
Error 14 0.0047 0.00034 15.26 %
Total 26 0.0308 100%

%
F) Tabulated 2.73 3.74 6.51
1
Appendix D
Appendix D: Plots of control factors effects

The control factor effects are displayed graphically in Figure D.1, Figure D.2,
Figure D.3, Figure D.4, Figure D.5, Figure F.6, Figure F.7, Figure D.8, and
Figure D.9 which makes it easy to visualize the relative effects of the
various factors on all three characteristics (i.e., The kerf taper, Surface
roughness and The heat affected zone)
Figure D.1: Plot of control factors effects (S/N ratios) for Ta.
Figure D.2: Plot of Control factors effects (Standard deviation values) for Ta.
1
Appendix D
Figure D.3: Plot of control factors effects (Mean values) for Ta.
Figure D.4: Plot of control factors effects (S/N ratios) for Ra.
1
Appendix D
Figure D.5: Plot of Control factors effects (Standard deviation values) for Ra.
Figure D.6: Plot of control factors effects (Mean values) for Ra.
1 9
Appendix D
180
160
140
HAZ [S/N Ratio]
120
100
80
60
40
20
x1.x2l1
x1.x2l2
x1.x2l3
x1.x3l1
x1.x3l2
x1.x3l3
x1.x4l1
x1.x4l2
x1.x4l3
x11
x12
x13
x21
x22
x23
x31
x32
x33
x41
x42
x43
Control Factor Level

Figure D.7: Plot of control factors effects (S/N ratios) for HAZ.
Figure D.8: Plot of Control factors effects (Standard deviation values) for HAZ.
1
Appendix D
4.5
3.5
3
HAZ [Mean]
2.5
1.5
0.5
Control Factor Level
Figure D.9: Plot of control factors effects (Mean values) for HAZ.
1 1
Appendix E
Appendix E: Summary of control factors effects

Table E.1: Summary of control Factors effects (S/N ratio and Mean values).
1-Summary of control Factors effects (S/N Values(.

The Kerf Taper Surface Roughness Heat affected zone
)Ta) (Ra) (HAZ)
Control
S/N by S/N by S/N by
Factor F F F
factor factor factor
level Value Value Value
level level level
)dB( )dB( )dB(
X1
1 96.945 -111.777 156.756
2 41.588 74.883 -141.417 130.35 119.567 268.39
3 2.452 -164.773 60.144
X2
1 62.124 -130.126 130.718
2 51.916 10.880 -139.191 15.86 108.087 32.01
3 26.944 -148.650 97.664
X3
1 43.445 -137.086 109.916
2 43.697 - -141.584 - 115.635 -
3 53.842 -139.297 110.918
X4
1 39.164 -137.571 107.554
2 46.244 - -137.316 - 111.806 -
3 55.576 -143.080 117.109
X1.X2
1 49.313 -135.109 115.274
2 45.612 - -140.842 - 112.615 -
3 46.059 -142.016 108.580
X1.X3
1 50.907 -138.710 113.454
2 43.052 - -140.146 - 111.700 -
3 47.024 -139.111 111.315
X1.X4
1 52.338 -139.469 112.310
2 42.753 - -139.574 - 113.466 -
3 45.892 -138.924 110.693
Over all
1
Appendix E
Mean 5.222 -15.480 12.462
2-Summary of control Factors effects (Mean Values(.

The Kerf Taper Surface Roughness Heat affected zone
)Ta) (Ra) (HAZ)
Control
Mean by Mean by Mean by
Factor F F F
factor factor factor
level Value Value Value
level level level
X1
1 2.631 37.74 1.222
2 5.392 72.314 54.84 114.69 2.002 139.58
3 9.044 74.53 4.201
X2
1 4.356 49.38 1.927
2 5.064 20.975 55.39 14.23 2.540 15.74
3 7.647 62.34 2.958
X3
1 5.766 55.20 2.550
2 6.514 - 57.95 - 2.557 -
3 4.787 53.96 2.318
X4
1 6.780 56.29 2.776
2 5.447 6.886 53.32 - 2.333 -
3 4.840 57.50 2.316
X1.X2
1 5.726 52.64 2.400
2 5.644 - 56.51 - 2.449 -
3 5.697 57.96 2.576
X1.X3
1 5.497 56.51 2.446
2 5.863 - 55.50 - 2.516 -
3 5.707 55.10 2.463
X1.X4
1 5.481 56.45 2.512
2 5.896 - 55.27 - 2.441 -
3 5.690 55.39 2.472
Over all
Mean 0.632111 6.18926 0.275
1
‫مل ص الرسالة‬
‫يعت ر م و "التاجوشى" لتصميم المتغيرا أ اة هامة لتصميم قو يقد ن ج متناسق بسيط‬

‫لتحسين اأ اء الجو ة التكل ة‪ .‬تعت ر من جية أس ح اإستجابة من التقنيا التحليلية ال عالة‬
‫لتوضيح عاقة الس ب التأثير بين مدخا العملية أستجابت ا ‪ .‬على الجانب ااخر يعت ر عملية‬
‫اسع للق ع‬ ‫ق ع باللي ر عملية من عمليا التشغيل غير التقليديه ‪ ,‬تست د على ن ا‬
‫‪,‬الح ر‪,‬عمل العاما ‪ ,‬اللحا ‪ ,‬التل يد اخيراً فى المعالجة الحرارية‪.‬اص ح لل وا المقا للصدأ‬
‫شع ية ك ير فى ال حو ‪ ,‬أنه يتم است دا ال وا المقا للصدأ على ن ا اسع فى عد من‬
‫الت يقا اليومية لك فى الصناعا المن لية ‪,‬المستش يا ‪ ,‬تج ي اأغ ية ‪ ,‬ال راعة ‪,‬ال يرا‬
‫‪,‬ال ناء ‪ ,‬الكيما يا ‪ ,‬االكتر نيا ‪,‬الم ا ا الحرارية ال اقة م ل مح ا توليد الك رباء‬
‫يتم ل فى الغايا حا يا الموائع ‪,‬يعت ر النوع اأ ستينى من ال وا المقا للصدأ هو ااك ر‬
‫ست داما ً حتى اآ ‪.‬‬
‫ال دف من ه الدراسة هو التصميم اأم ل لمتغيرا عمليا الق ع باللي ر لل وا المقا للصدأ‬
‫‪ ,‬لك ب دف الوصو لى القيمة الم لى لمعد كا من ت تق الشق متوسط خشونة الس ح‪ ,‬لقد‬
‫أجريت سلسله من التجار بأست دا ماكينة الق ع باللي ر لك لربط متغيرا العملية بعناصر‬
‫اإستجابة السابقة ‪.‬متغيرا التشغيل العينة للعملية هم ال اقة( ا ) ‪,‬ضغط اا كسجين (ميجا‬
‫باسكا )‪,‬تر الن ض(هيرت ) سرعة الق ع (سم‪ /‬قيقه)‪ .‬ت دف الدراسة لى تكوين نما ج رياضية‬
‫لمعد ت تق الشق ‪ ,‬متوسط خشونة الس ح المن ة المتأثرة بالحرارة كدالة فى عناصر العملية‬
‫الم ثرة‪.‬‬
‫قد أ ر النتائج التجربية على أساس كا من نس ة اإشار لى الضوضاء المتوسط الحسابى‬

‫أ ال اقة ضغط اأ كسجين سرعة الق ع هم العوامل الم ثرة على معد ت تق الشق لك عند‬
‫مستويا ثقة ‪ ٪90 ,٪95 ,٪99‬على الترتيب‪ .‬كما أ ال اقة‪ ,‬ضغط اأكسجين هي المعايير‬
‫ال امة التي ت ثر على متوسط خشونة الس ح عند مستو ثقة ‪ ٪90 ,٪95 ٪99‬على الترتيب‬
‫بالتالى فإ كاً من ال اقة ضغط اأ كسجين هى المعايير التى ت ثر على تأثير من قة الق ع‬
‫عند مستو ثقة ‪ ٪90 ,٪95 ٪99‬على الترتيب‪.‬‬
‫النما ج الرياضية المستنتجة من من جية أس ح اإستجابه تم ت ويرها لكل القيم على أساس نس ة‬
‫اإشارة لى الضوضاء أ المتوسط الحسابى أ اإنحراف المعيار ‪,‬أما المو يل الرياضى فقد تم‬
‫نم جته فقا ً لتحليل الت اين بإست دا أ ا التصميم اأم ل ل رنامج ‪.MATLAB‬‬
‫ل‬
‫ل‬
‫كعملفليسلامرءليودلعالـ ـال‪...‬لوليسلأخولعملم لهولجاهـل ل‬
‫ل‬
‫ول إانلكبيـرلال وملللعملعـ ـد ل‪...‬لصغ ل إا الالتوܒلعليهلاجحافل ل‬
‫ل‬
‫ول إانلصغيـرلال ومل إانلانلعامـال‪...‬لكبـ ل إا ال دتلاإليهلاحـافـل ل‬
‫ل‬
‫لولل لكناللالعمل إاللبس تܑل‪.....‬لسمنبيكلع لكوصيلهالببيان ل‬
‫ل‬
‫اءلوحرصلواجهادلولبلغـܑل‪.....‬لولحبܑلأس تا لوطولل مـان‬
‫الإهـــــــــــداء ل‬
‫ل‬
‫اإىل…………لم لحܒلقدهالتم لاجنܑ‪،‬لاإىلأيلا نون‪ .‬ل‬
‫اإى…………لم لجعللمشوا يلالعل يلم ا‪،‬لاإىلأيلالرحم‪ .‬ل‬
‫اإى…………لم لملس ندىلىلا يا لأخواىلالعزاء‪ .‬ل‬
‫اإىل…………لم لسانديلوأ يليلد ي‪،‬لاإىل وجيلالصابر ‪ .‬ل‬
‫اإىل…………لم للجلهملرتليلاد ‪،‬لاإىلأبنايلمرل لوحبيبهلالعزاء‪ .‬ل‬
‫اإىل…………لم لانوالجوا ىلدام ًال‪،‬أصدقاىلالوفياء‪ .‬ل‬
‫اإى ل…………لم لمللامعروفليدوملواميـللدا لفوـوالللكو ـرلىليـومل‬
‫أنىلأنكلوقوܒلج ىلعىلطولل‪،‬اإىلخاىلالغاى‪ .‬ل‬
‫ل‬
‫اإلهملميعالأهديلهديلامتواضـ لهـ ال اجيـالللالإطـا لبـمعر مللـ ول‬
‫مر لهدم‪ .‬ل‬
‫ل‬
‫ل‬
‫للللللللللللللللللللللللللللللللللللللللللللللللللللهند ل‬
‫للللللللللللللللللللللللللللللللللللللللللللللهاىلمدلعبد لمدلالس يد ل‬
‫أ‬
‫جــامعـــــــ المنيـــــــــــــــــــا‬
‫كليــــ ال ســـــــــــــــــــــــ‬
‫قسم ه س اإنتا التصميم‬
‫ال مـــــ جه اأم ليـــــــــــــــه لعمليــــــــــــــــــــــا‬

‫القطـــــــــــــــــــــع باللــــــي‬
‫س المي اني ي‬ ‫ج الماجستير فى ال‬ ‫سال مق م للحصو على‬
‫إعداد‬
‫‪ /‬هـــانى محمـ عبـ محمـ الســي‬ ‫المــ‬
‫بكالوريوس هندسة اإنتاج التصميم ‪ -‬كلية ال ندسة – جامعة المنيا‬
‫تح إشراف‬
‫اأستا ال كتو‬ ‫اأستا ال كتو‬

‫الجيوشى م تا على‬ ‫محم حسن جا ه‬
‫أستاذ بقس هندسة اإنتاج التصمي‬ ‫أستاذ ال ندسة الصناعية‬
‫ك ية ال ندسة – جامعة المنيا‬ ‫قسم التصميم الميكانيكى اإنتاج‬
‫كلية ال ندسة‪ -‬جامعة القاهرة‬
‫اأستا ال كتو‬
‫يحيى محمو إسماعيل‬
‫أستاذ بقس هندسة اإنتاج التصمي‬
‫ك ية ال ندسة – جامعة المنيا‬
‫‪2015‬‬

Modelling and Optimization of Laser Cut

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Modelling and Optimization of Laser Cut

Uploaded by

Copyright:

Available Formats

Minia University

“Modelling and Optimization of Laser Cutting

Submitted in Partial Fulfillment for M.Sc. Degree in

Prof.Dr.Eng. Mohamed Hassan Prof.Dr.Eng. El-Giushi Mokhtar

Prof. Dr.Eng. Yehia Mahmoud Ismail

“Modelling and Optimization of Laser Cutting

Prof.Dr.Eng. Mohamed Hassan Prof.Dr.Eng. El-Giushi Mokhtar

Prof. Dr.Eng. Yehia Mahmoud Ismail

Eng. Hany Mohamed Abdu Mohamed

3.7. Laser Cut Quality Characteristics------------------------------------------ 44

APPENDIX A: TWO-LEVEL AND THREE LEVEL

Figure No. Description Page

Figure 4.9 Effect of power and pressure on Ta at constant frequency and

Table No. Description Page

LBM Laser Beam Machining

L27OA 3Levels, 27 experiments OA

Taguchi's parameter design is a simple systematic approach to optimize

The objective of this study is to apply Taguchi optimization methodology

Laser cutting is nowadays the most significant application of lasers

In this chapter, The Taguchi approach is introduced. Classification of

1.2. Taguchi Approach

 Control Variables X: variables that can be controlled during the

 Noise variables Z: variables either not controllable, or too difficult

 Response variables Y: They are performance characteristics, the system

Figure 1.1: Process diagram with input-output responses [1].

Figure 1.2: Taguchi robust design matrix [1].

Figure 1.3: Taguchi Loss Function [1].

The standard form of the loss function L(y) is given as follows:

L(y( = k(y − T( 2 (1.1)

Nominal is the better characteristic: S/N = -Log (y 2/s2y) (1.2)

Smaller is the better characteristic: S/N = -Log (1/n) (∑y2) (1.3)

Larger is the better characteristic: S/N = -Log (1/n) (∑1/y2) (1.4)

Figure 1.4: Taguchi design procedure [2].

1.3. Response Surface Methodology (RSM)

Figure 1.5: Response Surface plot.

 Find factor settings (operating conditions( that produce the “best”

1.4. Brief Introduction of Laser Beam machining (LBM)

1.4.1. Features of Laser Beam

Laser, Light Amplification by Stimulated Emission of Radiation, is a

1.4.2. Laser Beam Cutting

Figure 1.6: Laser beam cutting process [6].

1.4.3. Types of laser Cutting:

In laser cutting, three versions of the process are to be distinguished:

 Oxygen laser cutting

successful in oxide-free cutting of stainless steels. It is also successfully used in

Evaporative Laser Cutting: In evaporative laser cutting the material to be cut

At present, CO2 lasers with performances of up to 5 kW and Nd: YAG lasers

1.4.4. Type of Lasers

Figure 1.8: Schematic of Nd:YAG laser beam machine.

Figure 1.9: Principle of a CO2 laser [10].

1.4.5. Advantages and Limitations

steps of the manufacturing process can be carried out properly. Current

1.5. STAINLESS STEEL

STAINLESS STEELS are iron-base alloys that contain a minimum of

to over 1.0% in certain grades [12]. Figure1.10 provides a useful summary of

Figure 1.10: Available stainless steel alloy systems. [12]

By the metallurgical phases present in their microscopic structures:

 Ferritic stainless steel

Ferritic stainless steels: consist of chromium (typically 12.5% to 17%) and

thermal expansion and other thermal properties are similar to conventional

Martensitic stainless steels: consist of carbon up to 1.2%, chromium (11-18%)

Austenitic stainless steels: consist of chromium (16-26%), nickel (10-22%)

Duplex stainless steels: consist of chromium (22-25%) nickel (5-7%),

1.5.1. Applications for Stainless Steels