ANALYSIS OF MACHINING QUALITY IN EDGE TRIMMING OF CARBON FIBER REINFORCED COMPOSITE. A Thesis by. Neebu Alex Urban

Transcription

1 ANALYSIS OF MACHINING QUALITY IN EDGE TRIMMING OF CARBON FIBER REINFORCED COMPOSITE A Thesis by Neebu Alex Urban B.Tech, Kerala University, Kerala, India, 2001 Submitted to College of Engineering and the faculty of the Graduate School of Wichita State University in partial fulfillment of the requirements for the degree of Master of Science December 2005

2 ANALYSIS OF MACHINING QUALITY IN EDGE TRIMMING OF CARBON FIBER REINFORCED COMPOSITE I have examined the final copy of this thesis for form and content and recommend that it be accepted in partial fulfillment of the requirements for the degree of Master of Science, with a major in Industrial Engineering. Dr. S. Hossein Cheraghi, Committee Chairman We have read this thesis and recommend its acceptance: Dr. Jamal Sheikh-Ahmad, Co-Advisor Dr. Behnam Bahr, Committee Member ii

3 DEDICATION Dedicated To My Parents iii

4 ACKNOWLEDGEMENTS I would like to thank my advisors Dr. S. Hossein Cheraghi and Dr. Jamal Sheikh- Ahmad for their guidance and support during my stay at Wichita State University. I acknowledge their valuable suggestions and committed efforts towards the successful completion of this work. My acknowledgements go to Dr. Behnam Bahr for his suggestions and acceptance to serve in my thesis committee. I also take this opportunity to thank Dr. Krishna K Krishnan for his guidance and support during my stay at Wichita State University. I am grateful to my family for giving me the emotional support. memorable. I thank my entire colleague s and friends who made my time in W.S.U This project was funded in part by the Wichita State University Manufacturing Innovation and Development in Aviation Initiative (MIND). I take this opportunity to thank the MIND team for their valuable support and input throughout this research project. iv

5 ABSTRACT Carbon fiber reinforced composites (CFRP) have found extensive application in today s industries such as aerospace, automotive and shipping industries for their high strength and light weight. But due to its inhomogeneous nature they encounter numerous machining problems. Service life of a component is highly dependent on the quality of machining. An experimental investigation is conducted to determine the effect of process parameters spindle speed and feed rate and tool condition on the surface quality of a machined CFRP composite material. Machining operation used is edge trimming. The aim of the experiment is to set optimum parameters for obtaining quality machined surfaces. Surface Quality was quantified based on delamination depth and surface roughness. It was found that delamination depth and surface roughness increase with an increase in feed rate and an increase in cutting distance and decrease with an increase in spindle speed. So the cutting conditions for best surface quality are high spindle speed and low feed rate and the cutting conditions for worst surface quality is low spindle speed and high feed rate. Results from this work were interpreted in the form of line graphs, 3D graphs and microscopic pictures for process optimization. Statistical analysis was done to validate the experimental results. v

6 TABLE OF CONTENTS 1 INTRODUCTION Research Focus and Objectives Report Organization LITERATURE REVIEW Fiber Reinforced Plastics (FRP) Machining of FRP Surface Quality of Machined FRP. 7 3 METHODOLOGY Introduction Experimental Set-up Workpiece material Cutting Tool Machine Set-up Workpiece Clamping Experimental Procedure Surface Roughness Measurement Delamination Measurement ANALYSIS OF RESULTS Introduction Type of Cutting Configuration Surface Finish of Machined CFRP Delamination Frequency Delamination of Machined CFRP Process Optimization graph Effect of Chip Thickness on Surface Roughness and Delamination Depth Effect of Chip Thickness on Surface Roughness in Longitudinal Direction Effect of Chip Thickness on Surface Roughness in Transverse Direction Effect of Chip Thickness on Delamination Depth STATISTICAL ANALYSIS OF EXPERIMENTAL DATA Statistical Design Matrix Statistical Output Analysis ANOVA Output for Delamination Depth Residual Analysis for Delamination Depth Factor effects of Delamination Depth Response Surface of Delamination Depth. 83 vi

7 5.2.5 ANOVA Output for Surface Roughness in Longitudinal Direction Residual Analysis for Surface Roughness in Longitudinal Direction Factor effects of Surface Roughness in Longitudinal Direction Response Surface of Surface Roughness in Longitudinal Direction ANOVA Output for Surface Roughness in Transverse Direction Residual Analysis for Surface Roughness in Transverse Direction Factor effects of Surface Roughness in Transverse Direction Response Surface of Surface Roughness in Transverse Direction CONCLUSION AND FUTURE WORK Conclusions Future Work REFERENCE.103 APPENDIXES vii

8 LIST OF TABLES Table 3.1 Fiber orientation of CFRP material. 14 Table 3.2 CNC Router Specification 15 Table 3.3 Experimental Matrix for Machining. 19 Table 5.1 Statistical Design Matrix.. 76 Table 5.2 ANOVA Table for Average Delamination Depth 77 Table 5.3 ANOVA Table for Surface Roughness in Longitudinal Direction 86 Table 5.4 ANOVA Table for Surface Roughness in Transverse Direction. 94 viii

9 LIST OF FIGURES Figure 2.1 Different Chip Formation. 6 Figure 2.2 Types of Delamination. 10 Figure 3.1 Diamond Interlocked Knurled Tool. 15 Figure 3.2 Computer Numeric Controlled Router. 16 Figure 3.3 Workpiece Clamping 17 Figure 3.4 Cutting Operation. 17 Figure 3.5 Schematic Diagram of Up-milling Operation Figure 3.6 Surftest SJ Figure 3.7 Clamping for Measuring Roughness 22 Figure 3.8 Measurement in Longitudinal Direction.. 23 Figure 3.9 Measurement in Transverse Direction.. 23 Figure 3.10 Optical Microscope Set up for Delamination Figure 3.11 Measurement of Delamination.. 26 Figure 4.1 Average Delamination Depth for Up milling and Down milling. 28 Figure 4.2 Average Surface Roughness in Longitudinal Direction for Up milling Down milling 28 Figure 4.3 2X-Magnification Scale 29 Figure 4.4 Optical micrograph of machined surface (S5000 rpm-f100 ipm). 30 Figure 4.5 Optical micrograph of machined edge (S5000 rpm-f100 ipm) 30 Figure 4.6 Optical micrograph of machined surface (S5000 rpm-f200 ipm) 31 Figure 4.7 Optical micrograph of machined edge (S5000 rpm-f200 ipm) 31 ix

10 Figure 4.8 Figure 4.9 Figure 4.10 Figure 4.11 Figure 4.12 Figure 4.13 Figure 4.14 Figure 4.15 Figure 4.16 Figure 4.17 Figure 4.18 Figure 4.19 Figure 4.20 Surface Roughness Longitudinal Direction (a) Speed 5000 rpm (b) Speed rpm (c) Speed rpm for feed rates of 100,200 and 400 ipm 33 Surface Roughness Transverse Direction (a) Speed 5000 rpm (b) Speed rpm (c) Speed rpm for feed rates of 100,200 and 400 ipm. 34 Surface Roughness Longitudinal Direction (a) Sharp Tool (130.5 inches) (b) Worn Tool (1020 inches). 36 Surface Roughness Transverse Direction (a) Sharp Tool (130.5 inches) (b) Worn Tool (1020 inches). 38 Shows Microscopic pictures of the machined surface for speed 5000 rpm at feeds (a) 100 ipm (b) 200 ipm and (c) 400 ipm. 39 Shows Microscopic pictures of machined surface for speed rpm at feeds (a) 100 ipm (b) 200 ipm and (c) 400 ipm 40 Shows Microscopic pictures of machined surface for speed rpm at feeds (a) 100 ipm (b) 200 ipm and (c) 400 ipm. 41 Delamination Frequency for speed 5000 rpm-feed 400ipm sharp tool (a) Type I Delamination (b) Type II Delamination (c) Type I/II Delamination (d) Type I, Type II, Type I/II and Type III Delamination. 44 Delamination Frequency for speed 5000 rpm-feed 400 ipm worn tool (a) Type I Delamination (b) Type I/II Delamination (c) Type III Delamination (d) Type I, Type II, Type I/II and Type III Delamination 46 Delamination Frequency for speed rpm-feed 100 ipm sharp tool (a) Type I Delamination (b) Type II Delamination (c) Type I/II Delamination (e) Type I, Type II, Type I/II and Type III Delamination. 48 Delamination Frequency for speed rpm-feed 100 ipm worn tool (a) Type I Delamination (b) Type III Delamination (c) Type I, Type II, Type I/II, Type III Delamination.. 50 Average Delamination Depth (a) Speed 5000 rpm (b) Speed rpm (c) Speed rpm for feed rates of 100,200 and 400 ipm.52 Average Delamination Depth>1.0mm (a) Speed 5000 rpm (b) Speed10000 rpm (c) Speed rpm fro feed rates of 100,200 and 400 ipm..53 x

11 Figure 4.21 Figure 4.22 Figure 4.23 Figure 4.24 Figure 4.25 Figure 4.26 Figure 4.27 Figure 4.28 Average Delamination Depth> 1.5mm (a) Speed 5000 rpm (b) Speed rpm (c) Speed rpm for feed rates of 100,200 and 400 ipm. 54 Average Delamination Depth (a) Sharp Tool (130.5 inches) (b) Worn Tool (1020 inches). 56 Average Delamination Depth >1.0mm (a) Sharp Tool (130.5 inches) (b) Worn Tool (1020 inches).. 57 Average Delamination Depth >1.5mm (a) Sharp Tool (130.5 inches) (b) Worn Tool (1020 inches) Microscopic pictures of machined edge for speed 5000 rpm at feed 100 ipm (a) Bottom surface (b) Top surface.. 60 Microscopic pictures of machined edge for speed 5000 rpm at feed 200 ipm (a) Bottom surface (b) Top surface.. 61 Microscopic pictures of machined edge for speed 5000 rpm at feed 400 ipm (a) Bottom surface (b) Top surface.. 62 Microscopic pictures of machined edge for speed rpm at feed 100 ipm (a) Bottom surface (b) Top surface.. 63 Figure 4.29 Figure 4.30 Figure 4.31 Figure 4.32 Figure 4.33 Figure 4.34 Microscopic pictures of machined edge for speed rpm at feed 200 ipm (a) Bottom surface (b) Top surface.. 64 Microscopic pictures of machined edge for speed rpm at feed 400 ipm (a) Bottom surface (b) Top surface 65 Microscopic pictures of machined edge for speed rpm at feed 100 ipm (a) Bottom surface (b) Top surface.. 66 Microscopic pictures of machined edge for speed rpm at feed 200 ipm (a) Bottom surface (b) Top surface. 67 Microscopic pictures of machined edge for speed rpm at feed 400 ipm (a) Bottom surface (b) Top surface. 68 Effect of Chip Thickness on Surface Roughness in Longitudinal Direction (a) Sharp Tool (130.5 inches) and Worn Tool (1020 inches)..71 xi

12 Figure 4.35 Figure 4.36 Figure 4.37 Figure 4.38 Effect of Chip Thickness on Surface Roughness in Transverse Direction (a) Sharp Tool (130.5 inches) (b) Worn Tool (1020 inches)...71 Effect of Chip Thickness on Average Delamination Depth (a) Sharp Tool (130.5 inches) (b) Worn Tool (1020 inches).. 73 Effect of Chip Thickness on Average Delamination Depth >1.00mm (a) Sharp Tool (130.5 inches) (b) Worn Tool (1020 inches). 73 Effect of Chip Thickness on Average Delamination Depth > 1.5mm (a) Sharp Tool (130.5 inches) (b) Worn Tool (1020 inches). 74 Figure 5.1 Normal Probability Plot: Average Delamination Depth 79 Figure 5.2 Outlier Plot: Average Delamination Depth 79 Figure 5.3 Residual vs Speed plot: Average Delamination Depth.. 80 Figure 5.4 Residual vs feed plot: Average Delamination Depth. 80 Figure 5.5 Residual vs Cutting Distance plot: Average Delamination Depth.81 Figure 5.6 Residual vs Predicted plot: Average Delamination Depth 81 Figure 5.7 Effect of Speed on Average Delamination Depth. 82 Figure 5.8 Effect of Feed on Average Delamination Depth 82 Figure 5.9 Effect of Cutting Distance on Average Delamination Depth 83 Figure 5.10 Contour Plot: Average Delamination Depth. 84 Figure D Surface plot: Average Delamination Depth 84 Figure 5.12 Cube Plot: Average Delamination Depth. 85 Figure 5.13 Normal Probability Plot: Surface Roughness (Longitudinal Direction) Figure 5.14 Outlier Plot: Surface Roughness (Longitudinal Direction) Figure 5.15 Residual vs Speed plot: Surface Roughness (Longitudinal Direction) 88 xii

13 Figure 5.16 Figure 5.17 Figure 5.18 Residual vs Feed plot: Surface Roughness (Longitudinal Direction). 88 Residual vs Cutting Distance plot: Surface Roughness (Longitudinal Direction) 89 Residual vs Predicted plot: Surface Roughness (Longitudinal Direction) 89 Figure 5.19 Effect of Speed on Surface Roughness (Longitudinal Direction). 90 Figure 5.20 Effect of feed on Surface Roughness (Longitudinal Direction) 91 Figure 5.21 Effect of Cutting Distance on Surface Roughness (Longitudinal Direction) 91 Figure D Surface Plot: Surface Roughness (Longitudinal Direction) 92 Figure 5.23 Contour Plot: Surface Roughness (Longitudinal Direction). 93 Figure 5.24 Cube Plot: Surface Roughness (Longitudinal Direction).. 93 Figure 5.25 Normal Probability Plot: Surface Roughness (Transverse Direction) Figure 5.26 Outlier Plot: Surface Roughness (Transverse Direction). 95 Figure 5.27 Residual vs Speed plot: Surface Roughness (Transverse Direction). 96 Figure 5.28 Figure 5.29 Figure 5.30 Figure 5.31 Residual vs Feed plot: Surface Roughness (Transverse Direction) 96 Residual vs Cutting Distance plot: Surface Roughness (Transverse Direction).. 97 Residual vs predicted plot: Surface Roughness (Transverse Direction) Interaction effect of Speed and Feed on Surface Roughness (Transverse Direction) Figure D Surface Plot: Surface Roughness (Transverse Direction) Figure 5.33 Contour Plot: Surface Roughness (Transverse Direction). 99 Figure 5.34 Cube Plot: Surface Roughness (Transverse Direction). 100 xiii

14 1 INTRODUCTION Carbon fiber reinforced composites (CFRP) are extensively used in today s industries such as aerospace, automotive and shipping because of their high specific strength and high specific stiffness. In addition they possess good thermal resistance, corrosion resistance, damping resistance and dimensional stability. In aerospace industries where high strength and light weight is a major criteria, composites become the preferred choice of material. But due to the inhomogeneous nature of CFRP, machining of composite materials becomes a major cause of concern in the industry. Reliability of a machined component for its service application is mainly dependent on the quality of the machined surface. Though composites are produced nearnet shape, they still have to undergo finishing operations to meet the dimensional requirements for assembly with other components. CFRP mainly undergoes finishing operations like trimming, drilling and turning. But due to the inhomogeneous nature of fiber reinforced composites, the machined surface will be rough and undergoes many forms of damage like delamination, spalling and splintering. Machining also causes changes in the mechanical and chemical properties of the individual constituents used in the composite. So the performance of a composite is critically dependent on the surface condition produced by machining. Therefore quality of a machined surface for its service life is of considerable importance. Knowledge of the machining characteristics of these materials is very limited and there has been little work done to study the influence of process parameters on the quality of machined CFRP composites. This work concentrates on studying the effect of process parameters like spindle speed and feed rate in edge trimming of a CFRP composite material on machined surface quality. 1

15 1.1Research focus and Objectives This study focuses on quantifying the amount of machining damage on a CFRP composite material undergoing trimming operation based on delamination depth and surface roughness. The purpose is to identify the effect of process parameters such as spindle speed and feed rate on the surface quality of machined CFRP parts. The objective is to develop a machining database to obtain quality CFRP parts by optimizing spindle speed and feed rate. This objective is achieved by conducting trimming experiments on CFRP parts at various spindle speeds and feed rates and analyzing the resulting data on delamination and surface roughness using statistical methods. 1.2 Report Organization This report is comprised of six chapters. Chapter one is introduction. Chapter two gives background information and literature review on CFRP. Chapter three discusses the experimental set up and measurement procedures used. Chapter four deals with analysis of results of the experiment. Chapter five discusses the statistical analysis of experimental data. Chapter six discusses conclusions and future work. 2

16 2. LITERATURE REVIEW 2.1 Fiber Reinforced Plastics (FRP) Composites are inhomogeneous materials consisting of two phases that are deliberately combined to form structures with desired properties. The two phases are the matrix phase and the reinforcement phase. The phases are macroscopically distinguishable with distinct boundaries between them known as interface. The role of the matrix phase is to hold the reinforcement phase, distribute the applied load and protect the composite from hostile environments such as heat, cold, moisture and corrosion. The reinforcement phase supplies strength to the composite, conducts or resists corrosion, resists or conducts electricity. The materials used for the matrix and reinforcement phase can be a metal, polymer or ceramic. Reinforcement materials in composite construction are used in the form of flakes, particles, sheets, whiskers and fibers. Composites are classified according to the type of matrix materials and the form of reinforcement material used. Since reinforcement improves the properties of the composite, selection of reinforcement material should be in such a way that they possess high strength and stiffness. Reinforcement materials retain their strength and stiffness in fibrous form than in any other form. Thus fibers are the most commonly used reinforcement materials in composite construction and hence called as fiber reinforced polymers (FRP). In general fiber reinforced composites are available in laminate form, which is obtained by stacking a number of thin layers of fibers and matrix and consolidating them into desired thickness. By controlling the stacking sequence and orientation of fiber in each layer a wide range of physical and mechanical properties can be obtained with the 3

17 composite specific to the application purpose. This gives composites the edge over metals and alloys in industrial applications. The properties of fiber reinforced composites are highly dependent on the lay and the individual components of the composites that is the fiber-matrix interface. Fiber reinforced composites offer high strength-weight ratio, high modulusweight ratio, high fatigue strength-weight ratio, high fatigue damage tolerance, low coefficient of thermal expansion and high internal damping [1]. These properties make fiber reinforced composites emerge as the major structural material in the aerospace, vehicle and shipping industry where weight reductions as well as exceptional physical and mechanical properties are of major concern. 2.2 Machining of FRP The need to understand the machining of FRP in greater depth is necessary to optimize the machining process parameters. Usually composites are produced to their final shape in the curing process but they still have to undergo finishing operations to meet the dimensional requirements for fitting in with other parts. The common finishing operation used in industry is edge trimming. Edge trimming can be done with conventional router machine and with non-conventional machining process such as electrical discharge machining, ultrasonic machining, laser machining and abrasive water jet machining [2]. Since non-conventional machining processes have disadvantages like heat-affected zone and low material removal rate, the preferred choices of trimming are with conventional router and abrasive water jet machining. Machining of FRP is entirely different from machining of metals and alloys. Since FRP s are heterogeneous materials, material behaviour is inhomogeneous and 4

18 depends on the properties of fiber and matrix and fiber orientation. So due to this inhomogeneity, the machined surface of FRP will be rougher and less regular than machined metal surface. It has been shown that the machined surface of a fibrous composite is highly dependent on the chip formation process and the fiber lay direction with respect to cutting direction [3]. Chip formation process in machining of FRP is entirely different from machining of metals. Machining of FRP relies on three mechanisms for generating a new surface and they are abrasion, plowing and cutting [4]. In abrasion the tool cuts the softer matrix and separates the chip. In plowing the tool pushes the material and deforms plastically and in cutting it is simply the severing of fibers. The combination of these three mechanisms, fiber orientation and cutting direction will determine if the chip formation is by deformation, rupturing or shearing of the composite material. When cutting a composite laminate with fiber orientation along the cutting direction the failure mechanism is rupture of fibers and debonding between the matrix and fiber. So when the tool moves parallel to the fiber orientation, the fibers fracture into pieces. While when machining a composite laminate with fiber oriented at an angle to the cutting direction the failure mechanism will be deformation or shearing [4]. So when the tool moves to fiber oriented at an angle the fiber gets crushed and fractures sharply, as a result the fractured fibers protrude from the machine surface, plies get separated, surface becomes rougher and large machine damage such as delamination and cracks are observed. Figure 2.1 shows the rupture, deformation and shearing of a composite laminate subjected to different fiber orientation. 5

19 Fig 2.1 Different Chip Formation [4] So due to the inhomogeneity and abrasive nature of fibers, tool selection in machining FRP is also a major concern. High hardness of the fiber limits the tool 6

20 selection to extremely hard and expensive materials. Cutting tools used for trimming of FRP composites are polycrystalline diamond or carbide tools since they have high abrasion resistance, crumbling resistance and prolonged tool life [5]. In composite machining the top and bottom layers have the biggest effect on machining quality. Therefore selection of tool geometry should also be made according to their properties. For trim routing of glass or carbon fiber reinforced composites, tools with multiple cutting edges made of polycrystalline diamond or carbide cutting materials are preferred since they withstand intermittent cutting of the hard and soft material, have long tool life and at the same time maintain surface quality. Another serious problem associated with machining of FRP is the generation of air borne dust, which could cause serious health problems as well as damage to the electric components of the machine. So due to the abrasive nature of the fiber tool wear increases rapidly which results in poor surface quality and also generation of chips in the form of dust, which are harmful, leads us to optimize the machining process. 2.3 Surface Quality of Machined FRP Machined surfaces are evaluated using different methods each of which has its own unique characteristics corresponding to the quality requirements. The choice of inspection technique depends on the available equipment, the investigator and the terms in which the quality control guidelines are established. Mechanical performance of homogeneous material is dependent on the residual stress induced in the material and surface topography [6]. Since FRP do not develop residual stress in machining, the quality of machined FRP is evaluated based on surface profilometry and visual techniques. 7

21 The notion quality in machining fiber reinforced composite is based on two main aspects; they are surface topography and machine damage. Surface topography is characterized by surface roughness and machine damage is characterized by delamination. Surface of a machined FRP mainly consists of small holes, fiber cracks, fiber chipping, and blurs of the matrix material. Quality of a machined surface is dependent on fiber orientation and type of fiber. Glass and carbon fiber undergo brittle fracture when subjected to tensile, bending and shearing stress while aramid fibers fracture when subjected to simultaneous tensile and shearing stress. From studies conducted, it is seen that at fiber orientation of 30 and 45 the surface is very poor, since the fiber is cut by combined compressive and bending stress, while better surface is obtained at fiber orientation of 90 [9]. Other factors that influence surface roughness are tool wear, feed rate, and temperature. Studies show that surface roughness increases with increase in tool wear, feed rate and temperature. The parameters used to characterize a machined surface fall into two categories; they are roughness parameters (Ra, Rq, Rz, Rt) and statistical parameters such as skewness, kurtosis and frequency height distributions. The various roughness parameters are arithmetic average height (Ra), root-mean square height (Rq), peak to valley height (Rt), valley to mean height (Rv) and ten-point average height (Rz). However studies have shown that roughness parameters Ra and Rq show limited variation in their values with respect to fiber orientation [6]. So the preferred roughness parameters for representing the surface features of composites are peak to valley height (Rt) and ten-point average height (Rz), which is the average of five peak points and five valley points [3]. The role of these roughness parameters is to evaluate the surface produced by a machining process and to 8

22 quantify the amount of machining damage for different process parameters such as cutting speed, feed rate and depth of cut. It has been shown that lower the value of surface roughness, the better is the quality of machined surface. Roughness values also indicate changes in the mechanical properties of machined FRP. Studies have shown that with increasing roughness the fatigue strength and impact strength decreases [7]. Roughness is measured on a machined surface by using stylus profilometer, which is the common instrument used in the industry. Roughness value is given by the vertical displacement of a diamond stylus tip, which moves along the machined surface. Result of roughness measurement greatly depends on the stylus path, since fiber direction changes from layer to layer. Better method of taking roughness measurements is to keep the stylus in one layer and take readings at different locations of this layer or take readings at different location for different layers and take the average. Another important thing to be considered in taking roughness measurement of a composite surface is that matrix smearing, fiber protruding and fiber clinging to the stylus tip will obliterate the reading and will not give an exact description of the surface. So a visual inspection in combination with the profilometer reading is necessary to quantify surface topography. Fiber reinforced composites undergo many types of machining damages, important among them are delamination which is the separation of the interply, debonding which is the failure of fiber-matrix adhesion and fiber pullout which is the removal of fiber from the matrix and fibers protrude from the surface. Machining damage experienced by fiber-reinforced composites is highly dependent on the fiber orientation relative to cutting direction. Multidirectional laminates undergo different forms of 9

23 damage during machining and therefore maintenance of surface quality becomes difficult. From studies conducted on edge trimming of FRP, it is seen that there are four types of delamination present irrespective of cutting mode and condition as shown in figure 2.2 and they are characterized as Type I, Type II, Type I/II and Type III delamination [8]. Type III Fig 2.2 Types of delamination [8]. Type III Delamination 10

24 Type I delamination is characterized as areas where plies have broken inwards and are seen as areas where plies have been missing from the trimmed edge. Type II delamination consists of uncut fibers protruding from the trimmed edge. Type I/II delamination consists of plies which have broken inwards and extend outwards from the trimmed surface. Type III delamination consists of fibers detached from the trimmed surface along the cutting direction. These types of delamination occur in edge trimming operations since the surface plies are not stabilized by the adjacent plies. Also machining forces act in one direction and tend to lift the layers resulting in serious delamination. Studies conducted showed that cutting parameters influence machine damage as well as strength of the composite. Machine damage was found to decrease with increasing cutting speed and lower feed rate. In aerospace application a threshold value of 2.5mm is set for delamination damage depth for the top and bottom plies [8]. Based on this threshold value process parameters like cutting speed, feed rate, cutting mode (up milling or down milling) and depth of cut are selected which produces parts with delamination below 2.5mm. Analysis of delamination for particular process parameters in one study was conducted by cutting a number of workpieces at a particular cutting speed and feed rate. Then the parts were analyzed for delamination depth above 1mm (in that study 1mm was set as the threshold value for delamination depth). Delamination depth was represented in terms of percent delaminated, which is by taking the number of delaminated parts above 1mm and dividing it by the total number of parts machined at that particular cutting speed and feed rate [8]. The idea was to see the effect of machining parameters on surface quality. In that study cutting speed was maintained constant and feed rate was varied. The 11

25 workpiece used was a graphite/epoxy composite material composed of eight plies. Two types of cutting tools of different diameters were used. One was a carbide tool of 10 helix angle and the other was polycrystalline diamond tool of 30 helix angle. It was seen that the tool condition also influenced delamination. As tool wear increased machine damage also increased. In that study it was found that polycrystalline diamond tool generated less delamination since its wear was very slow. The study also found that slow feed rate has significant effect in reducing delamination. This inhomogeneity makes FRP unique and at the same time creates problems in machining which leads to optimization of the machining process. So based on the present and future application of fiber reinforced composites, it is necessary to understand the effect of process parameters on the quality of CFRP materials. There has been little work done to know the influence of process parameters on the quality of CFRP composite materials. This study is mainly conducted to know the effect of process parameters and tool wear on the machined surface quality of a CFRP composite material. The experiment is conducted by varying spindle speed and feed rate on a CNC router using edge trimming operation and monitoring delamination depth and surface roughness. The idea is to develop a machining database to obtain optimum cutting speed and feed rate in routing composite materials. This machining database will be useful for the operator to set the optimum cutting parameters for obtaining acceptable machined part consistent with the quality criteria. 12

26 3. METHODOLOGY 3.1 Introduction The inhomogeneous nature of CFRP makes it s machining very difficult, especially in obtaining quality machined surfaces. The only option in which acceptable machined surfaces can be obtained is by controlling the process parameters. This work presents a study on determining optimum cutting speed, feed rate and tool condition for obtaining quality machined surface based on delamination depth and surface roughness. The machining operation used for this study is edge trimming, which is a type of milling operation and is one of the major finishing operations done for CFRP materials in the aerospace industry. Machine used for the experiment is a 3-axis CNC router manufactured by Timesavers. The machine has a maximum spindle speed of 15,000 rpm and maximum feed rate of 400 ipm. An experimental matrix for machining was prepared based on three spindle speeds of 5000, and rpm and three feed rates of 100, 200 and 400 ipm. Cutting tool used for the experiment is a diamond interlocked knurled tool of 0.25 inches diameter manufactured by Ultra Tools which, is the common cutting tool used in the industry for edge trimming process. A constant radial depth of cut of inches is maintained throughout the experiment. Type of cutting configuration used for the entire experiment is up milling. Dimensions of the CFRP board used for the experiment are inches. Surface roughness was measured using SJ-400 surface profilometer manufactured by Mitutoyo Corporation and delamination was measured using an optical microscope. 13

27 3.2 Experimental Set-up Workpiece Material The work piece material used in the experiment is a multidirectional continuous carbon fiber reinforced composite with 0.1 inch thickness. The workpiece was cut into blanks of inches for the experiment. The CFRP material used has a 10-ply lay up and the orientation of the fiber used in the composite is given in Table 3.1. Ply or Part Number Fiber Code Orientation P2 WGA 45 0 /135 0 P3 WG 0 0 /90 0 P4 WG 45 0 /135 0 P5 WG 0 0 /90 0 P6 WG 45 0 /135 0 P7 WG 45 0 /135 0 P8 WG 0 0 /90 0 P9 WG 45 0 /135 0 P10 WG 0 0 /90 0 P11 WG 45 0 /135 0 Table 3.1 Fiber orientation of the CFRP material Cutting Tool The tool used in the experiment is diamond interlocked knurled tool manufactured by Ultra Tools. The tool is made of sub micron grade tungsten carbide material. The word diamond is used since the tool has a number of single cutting edges in the form of 14

28 diamond shape. Diameter of the tool is 0.25 inches, overall length is 2.5 inches, and cutter length is 0.75 inches with a drill point angle of 135. The diamond interlocked tool used for the experiment is shown in the Figure 3.1. Fig 3.1 Diamond Interlocked Knurled Tool Machine Set Up The machine used for this experiment is a 200 series 3-axis CNC router of model number 235-I manufactured by Timesavers. Specifications of the machine are given in Table 3.2. Figure 3.2 shows the CNC router machine. Parameters Spindle Type Spindle Speed Horse Power of the Spindle Depth of Cut (Maximum) CNC controller Maximum X-axis travel distance Maximum Y-axis travel distance Maximum Z-axis travel distance Work Surface/Table Specifications Air cooled, quick change RPM 7.5 hp 400 IPM General Numeric 810 CNC control (Siemens) 62 inches 38 inches 6 inches 36 inches Χ 60 inches Table 3.2 CNC Router Specifications 15

29 Fig3.2 Computer Numeric Controlled Router Workpiece Clamping The workpiece used for the experiment is mounted on a Kistler force dynamometer as shown in Figure 3.3. The workpiece was clamped securely and carefully on the dynamometer using a fixture so that minimum vibrations occur while cutting. 16

30 Fixture Tool CFRP Board Dynamometer Fig 3.3 Workpiece Clamping Experimental Procedure The experiment consists of short-cut experiment carried out on the CNC router as shown in Figure 3.4. Cutting Tool Clamp fixed to dynamometer Workpiece Tool Feed Direction (Up-mill) Fig 3.4 Cutting Operation. 17

31 Linear cuts were made on the 4.0 inch side. Cutting mode is up milling. Figure 3.5 shows a schematic diagram of up-milling operation. Radial depth of cut is inches (25% of tool diameter). Two passes were made on the 4.0 inch side and after that the workpiece is removed from the dynamometer and analyzed for delamination using optical microscope and surface roughness using stylus profilometer. The experiment is conducted following the matrix as shown in Table 3.3. a eff a e V c V f Fig 3.5 Schematic diagram of Up-milling operation Where a e - radial depth of cut in inches V f Feed rate in ipm V c Cutting speed in ipm = πdn, D- diameter of cutter, N-spindle rpm a eff Effective chip thickness in inches = a e V f / V c 18

32 Spindle Speed, N (rpm) Cutting Speed, V c (ipm) Feed Rate, V f (ipm) Effective Chip Thickness, a eff (inches) Table 3.3 Experimental Matrix for Machining As there are three spindle speeds and three feed rates there will be nine matrix points. The idea is to see how the quality of the machined surface varies with respect to each combination of spindle speed and feed rate. Surface quality of the machined CFRP is analyzed for the tool when it is sharp and worn out. Wear of the tool is achieved by cutting on a long CFRP panel of 24.5 inch x 25 inch dimension. So each combination of spindle speed and feed rate consists of five runs. The first two runs consist of seven cuts, of which five cuts are made on the long panel and two cuts on the short panel. The next three runs consist of twelve cuts each, of which ten cuts are made on the long CFRP panel and the last two cuts on the short CFRP panel. At the end of each run the short workpiece is analyzed for delamination and surface roughness. So the nine matrix points consist of 45 runs. The entire matrix is repeated once again for validating the results. So a total of 90 workpieces of 4-inch 1.5- inch were prepared for the experiment. For each matrix point a new tool is used. So a total of 18 cutting tools were used for the experiment. A CNC program was written for conducting long cuts and short cuts. 19

33 Before starting this experiment a trial experiment was conducted to determine the type of milling operation (up-milling or down-milling) to be used for the above experiments. For this a spindle speed of 5000 rpm and feed rates of 100 ipm and 200 ipm were selected. A threshold value for delamination depth is selected based on the worst operating condition when the first experiment set is conducted. Results of these experiments are discussed in detail in chapter four. 3.3 Surface Roughness Measurement At the end of each run the workpiece is removed from the dynamometer and is analyzed for the surface variation created by the particular matrix point of spindle speed and feed rate. Surface finish of the workpeice is measured with a surface roughness tester Surftest SJ-400 manufactured by Mitutoyo. It is the common type of roughness measuring instrument used in the industry for machined surface inspection. SJ-400 is capable of measuring surfaces with various parameters; in our study we use Rz for quantifying surface finish since it is the better roughness indicator for composites. Roughness value is determined by the vertical displacement of the stylus produced by the surface irregularities when the tip moves along the surface. Roughness values are obtained on a display unit. Figure 3.6 shows the SJ-400 roughness tester and the display unit. 20

34 Display unit SJ-400 Fig 3.6 Surftest SJ-400 Surface roughness measurement for this study is done by first cleaning the machined surface with air to remove the powdered chips and broken fibers so that these don t hinder the stylus measurement. Then the workpiece is placed in a pocket on a block with the machined edge on top. Figure 3.7 shows workpiece clamping for surface roughness measurement. 21

35 Fig 3.7 Clamping for measuring roughness Surface roughness is measured both in a longitudinal direction that is along the cutting direction and a transverse direction that is perpendicular to the cutting direction. In the longitudinal direction a cut-off length of 0.8mm is taken which is the industrial standard and a traverse length of 4.0mm is taken for longitudinal measurement. For the transverse measurement a cut-off length of 0.25mm and traverse length of 1.25mm were taken. This constraint in the transverse measurement is due to the small thickness of the workpiece which is 0.1 inches, and the maximum traverse length we can obtain is 1.25 mm. Roughness measurement for both longitudinal and transverse directions are taken at six different locations of the workpiece and the average was taken. Figures 3.8 and 3.9 show how measurements in the longitudinal and transverse directions were taken. 22

36 Fig 3.8 Measurement in Longitudinal Direction Fig 3.9 Measurement in Transverse Direction 23

37 3.4 Delamination Measurement At the end of each run the short workpiece is analyzed for delamination type and depth using an optical microscope. The optical microscope setup consisted of a 10X zoom lens system (NAVITAR), a CCD video camera (PULNIX-TM-200), a DC regulated halogen light source for illumination, XCAP image capturing software and an adjustable table with a micrometer dials attached to it. The lens attached to the camera gets the picture of the workpiece of desired magnification and the software displays the image for analysis. Figure 3.10 shows the setup of the optical microscope. Fig 3.10 Optical Microscope Set up for Delamination 24

38 For delamination measurement the workpiece is placed flat on the moving table so as to view the machine edge of the workpiece. To capture the image, a 2.0X lens and a zoom scale of 1.0X were used. Illumination is adjusted so as to get a clear image of the machined edge. Once the software displays the image the edge is scanned for delamination. Then the delaminations are classified as Type I, Type II, Type I/II and Type III as discussed in the literature review. For each type of delamination the depth of delamination is measured as shown in Figure The depth is measured with the micrometer dial attached to the moving table. Procedure for delamination measurement is as follows: Capture the image Identify the type of delamination Table is moved so that image on the screen changes accordingly so as to get to the point where the damage depth is extended. Micrometer dial is set to zero and the table is moved till the damage is extended Readings on the micrometer is noted and recorded This procedure is repeated for the entire 4.0inch length of the workpiece on both the top and bottom surfaces. Delamination depth for all types of delamination are noted on both the top and bottom surface of the entire workpiece and combined average of all types of delamination was taken. Figure 3.11 shows how delamination depth is measured for each type of delamination. 25

39 Type I/II Type II Type I Type III Fig 3.11 Measurement of Delamination Depth 26

40 4 ANALYSIS OF RESULTS 4.1 Introduction Fiber reinforced composites application for service life is highly dependent on the quality of the machined surface. Due to the abrasive nature of CFRP, machining of these materials becomes a major problem. Even though CFRP composites are produced near net shape they still have to undergo finishing operations such as trimming and sawing to meet the dimensional requirements for the service application. These finishing operations create numerous machining damages to CFRP composites. Therefore the quality of a machined surface is of critical importance. This chapter discusses the results obtained from experiments conducted for quantifying the surface quality of a machined CFRP composite using edge trimming operation on a CNC router. Results are explained in a sequence in which the experiment is conducted. 4.2 Type of Cutting Configuration Before the start of the actual experiment a trial experiment was conducted to determine the type of cutting configuration, namely up milling or down milling, to be adopted for the actual experiment. The criterion is based on which cutting mode gives minimum delamination depth and lower surface roughness values. For this trial experiment a spindle speed of 5000 rpm and feed rates of 100ipm and 200ipm were selected for determining the cutting configuration. Short cut experiments using a force dynamometer were used for this study. All cuts were linear cuts and for each cutting mode the experiment was repeated twice and the average delamination depth and surface roughness in longitudinal direction were determined. Figures 4.1 and 4.2 show results of the experiment. 27

41 Average Delamination Depth-S5000 Avg Depth (inches) Up milling Down milling Feed Rate (ipm) Fig 4.1 Average Delamination Depth for Up Milling and Down Milling Average Surface Roughness-Longitudinal Direction-S Rz(um) Up milling Down milling Feed Rate (ipm) Fig 4.2 Average Surface Roughness in Longitudinal Direction for Up Milling and Down Milling. 28

42 Figures 4.1 and 4.2 show that both average delamination depth and surface roughness are lower for up milling than for down milling. This is because in down milling the matrix and fibers pile up inside the flute of the tool and obstruct the cutting which in turns produces rough surfaces and deep damages. The optical microscope pictures in Figures also show that up milling is better compared to down milling. The microscopic pictures are taken at 2X zoom scale and Figure 4.3 shows a 1mm calibration scale taken at the same magnification for reference. One division on this scale is 50µm and 1mm is equal to 20 divisions. All the microscopic pictures are taken at 2X-scale. Fig 4.3 2X-Magnification scale (1mm) 29

43 (a) Up Milling (b) Down Milling Figure 4.4 Optical micrograph of machined surface. Machining along horizontal direction from left to right for up milling and from right to left for down milling (S5000 rpm-f100 ipm). (a) Up Milling (b) Down Milling Fig 4.5 Optical micrograph of machined edge. Machining along horizontal direction from left to right for up milling and from right to left for down milling (S5000 rpm-f100 ipm). 30

44 (a) Up Milling (b) Down Milling Figure 4.6 Optical micrograph of machined surface. Machining along horizontal direction from left to right for up milling and from right to left for down milling (S5000 rpm-f200 ipm). (a) Up Milling (b) Down Milling Fig 4.7 Optical micrograph of machined edge. Machining along horizontal direction from left to right for up milling and from right to left for down milling (S5000 rpm-f200 ipm). 31

45 Also if we look into the pictures for down milling we see that there is fiber breakage, fiber pull out, and matrix removal from between fibers and cracks. While for up milling we see more cracks with less amount of fiber breakage and fiber pullout. This results in poor surface quality for down milling compared to up milling. Based on these facts it is decided to utilize up milling for the remaining experimental matrix. 4.3 Surface Finish of Machined CFRP Surface roughness plays a vital role in specifying smooth finish of a machined surface in the production environment. Manufacturing processes can drastically change the surface layer, which in turn results in change in the mechanical properties of the composites. There are many factors that affect the surface layer such as cutting speed, feed rate, depth of cut, tool geometry and cutting configuration. So quantification of surface layer for generating a smooth surface is imperative. The roughness parameter helps in obtaining a smooth surface finish by optimizing the process parameters. In this study at the end of each run the workpiece is analyzed for its surface variations produced by the process parameters. Surface roughness was measured in two directions on the workpiece, one along the feed direction (longitudinal direction) and the other across the feed direction (transverse direction). For longitudinal direction the sampling length (cutoff length) is 0.8mm and traverse length is 4.0mm. For transverse direction the sampling length is 0.25mm and traverse length is 1.25mm. This constraint is due to small thickness of the workpiece, which is 0.1 inches. The experimental matrix was repeated twice and an average was taken for surface roughness values for both longitudinal direction and transverse direction. Results of the average surface roughness along the longitudinal direction and transverse direction are 32

46 shown in Figures 4.8 and 4.9. The actual surface roughness values from both experiments for longitudinal and transverse direction are given in the appendix. Average Surface Roughness-Longitudinal Direction-S5000 Rz (um) Cutting Distance (inches) aeff f100 aeff f200 aeff f400 (a) Average Surface Roughness-Longitudinal Direction-S10000 Rz (um) Cutting Distance (inches) aeff f100 aeff f200 aeff f400 (b) Average Surface Roughness-Longitudinal Direction-S15000 Rz (um) Cutting Distance (inches) aeff f100 aeff f200 aeff f400 (c) Fig 4.8 Surface Roughness Longitudinal Direction (a) Spindle speed 5000 rpm (b) Spindle speed rpm (c) Spindle speed rpm for feed rates of 100,200 and 400 ipm. 33

47 Average Surface Roughness-Transverse Direction-S5000 Rz (um) Cutting Distance (inches) aeff f100 aeff f200 aeff f400 (a) Average Surface Roughness-Transverse Direction-S10000 Rz (um) Cutting Distance (inches) aeff f100 aeff f200 aeff f400 (b) Average Surface Roughness-Transverse Direction-S15000 Rz (um) Cutting Distance (inches) aeff f100 aeff f200 aeff f400 (c) Fig 4.9 Surface Roughness Transverse Direction (a) Spindle speed 5000 rpm (b) Spindle speed rpm (C) Spindle speed rpm for feed rates of 100,200 and 400 ipm. 34

48 In the above figures, for longitudinal direction, it is seen that surface roughness increases with an increase in cutting distance and increases with an increase in feed rate for all the three spindle speeds. The increase in surface roughness with cutting distance is due to the wear of the tool. As tool wear increases with increasing cutting distance surface roughness also increases resulting in higher surface roughness values. Surface roughness increases proportionally with increase in feed rate. As feed rate increases, thickness of the chip increases, which results in an increase of surface roughness. For transverse direction we don t see any trend with respect to cutting distance and feed rate. It is noted that the surface roughness in the transverse direction is several magnitudes higher than the surface roughness in the longitudinal direction. This is because of more waviness in the transverse direction due to ply stack up. One reason we can give for seeing no trend in the transverse direction roughness is the restriction we face in setting the sampling length to 0.8 mm, which is the industrial standard, required for estimating the machined surface for a particular process. Another reason we can give is the plies are separated in an irregular pattern, which results in giving inconsistent roughness values. For the purpose of identifying which process parameters that is, which spindle speed and feed rate, give better surface finish 3D-graphs are plotted. 3D-graphs are plotted for the first and last run of each matrix point. First run is assumed as the state in which the tool is sharp and the last run is assumed as the state in which the tool is worn out. This is to see how wear of the tool and the process parameter affect surface finish. Figure 4.10 shows the 3D-graphs for surface roughness in longitudinal direction for the tool when it is sharp and when it is worn-out. 35

49 Average Surface Roughness-Longitudinal Direction-Sharp Tool 30 Rz (um) F100 F200 Feed Rate (ipm) F400 S15000 S10000 S5000 Speed (rpm) S5000 S10000 S15000 (a) Surface Roughness Longitudinal Direction Sharp Tool Average Surface Roughness-Longitudinal Direction-Worn Tool Rz (um) F100 F200 Feed Rate (ipm) F400 S15000 S10000 S5000 Speed (rpm) S5000 S10000 S15000 (b) Surface Roughness Longitudinal Direction Worn Tool Fig 4.10 Surface Roughness Longitudinal Direction (a) Sharp Tool (130.5inches) (b) Worn Tool (1020 inches). 36

50 From the 3D-graphs we see that surface roughness increases with increase in feed rate and decreases with increase in spindle speed for both conditions of the tool. Roughness values are higher for worn tool than for sharp tool, this is due to wear of the tool since wear causes increase in roughness. The best cutting condition for surface finish is high spindle speed and low feed rate and the worst cutting condition is low spindle speed and high feed rate. So, based on these two extreme conditions we can set the optimum spindle speed and feed rate for surface finish depending on quality requirements for the trimming operation. Figure 4.11 shows the 3D-graphs for surface roughness in transverse direction for sharp and worn tool. In these graphs we see that surface roughness increases with increase in spindle speed except at spindle speed15000 rpm and feed rate 400 ipm for both conditions of the tool. Also roughness values are higher than in longitudinal direction. This is because in transverse direction waviness is more and the plies are separated in an irregular manner, resulting in inconsistent surface roughness values. Transverse direction graphs do not give any information for optimizing spindle speed and feed rate. Optical microscope pictures are also taken for each matrix point for the state of the tool when it is sharp and worn which is the first run (R1) and last run (R5) of the matrix combinations. Figures show the microscopic pictures of the machined surfaces. 37

51 Average Surface Roughness-Transverse Direction-Sharp Tool 30 Rz (um) F100 F200 Feed Rate (ipm) F400 S15000 S10000 S5000 Speed (rpm) S5000 S10000 S15000 (a) Surface Roughness Transverse Direction Sharp Tool Average Surface Roughness-Transverse Direction-Worn Tool Rz (um) F100 F200 Feed Rate (ipm) F400 S15000 S10000 S5000 Speed (rpm) S5000 S10000 S15000 (b) Surface Roughness Transverse Direction Worn Tool Fig 4.11 Surface Roughness Transverse Direction (a) Sharp Tool (130.5 inches) (b) Worn Tool (1020 inches). 38

52 (a) S5000-F inches-sharp S5000-F inches-worn (b) S5000-F inches-sharp S5000-F inches-worn (c) S5000-F inches-sharp S5000-F inches-worn Fig 4.12 Shows Microscopic pictures of the machined surface for spindle speed 5000 rpm at feed rates (a) 100 ipm (b) 200 ipm and (c) 400 ipm. Machining along horizontal direction from left to right. 39

53 (a) S10000-F inches-sharp S10000-F inches-worn (b) S10000-F inches-sharp S10000-F inches-worn (c) S10000-F inches-sharp S10000-F inches-worn Fig 4.13 Shows Microscopic pictures of machined surface for spindle speed rpm at feed rates (a) 100 ipm (b) 200 ipm and (c) 400 ipm. Machining along horizontal direction from left to right. 40

54 (a) S15000-F inches-sharp S15000-F inches-worn (b) S15000-F inches-sharp S15000-F inches-worn (c) S15000-F inches-sharp S15000-F inches-worn Fig 4.14 Shows Microscopic pictures of machined surface for spindle speed rpm at feed rates (a) 100 ipm (b) 200 ipm and (c) 400 ipm. Machining along horizontal direction from left to right. 41

55 From the microscopic pictures we see that surface gets deteriorated when cutting distance reaches 1020 inches from inches. Surface gets rougher with increase in feed rate and surface finish gets better with increase in spindle speed. This observation is in agreement with the 3D-graph results for longitudinal direction in Figure Also we see that ply separation is more obvious with increase in spindle speed resulting in peaks and valleys. This is why we see an increase in roughness with an increase in spindle speed in the 3D-graphs for transverse direction in Figure Also we don t see any trend with respect to feed rate in the transverse direction, this can be due to fiber reorientation and matrix smearing caused by machining which obstructs the stylus in giving a representative value of the machined surface. So we can only consider the longitudinal direction surface roughness values for optimizing the process parameters. 4.4 Delamination Frequency Delamination frequency is used to analyze the occurrence frequency of the different types of delamination and their depth. Delamination frequency shows which types of delamination are mostly occurring, their frequency of occurrence and how their depth is distributed. From the distribution we can determine which statistical parameter to select such as average, standard deviation or range to draw general conclusions about surface quality. For this we selected the best cutting condition of spindle speed of rpm and feed rate of 100 ipm and the worst cutting condition of spindle speed of 5000 rpm and feed rate of 400 ipm for the tool when it is sharp and worn and found out the distribution of all the types of delamination in a combined way and also separately. 42

56 (a) Frequency of Occurrence S5000-F400-TI-D-Sharp Tool Delamination Depth (inches) (b) Frequency of Occurrence S5000-F400-TII-D-Sharp Tool Delamination Depth (inches) Frequency of Occurrence S5000-F400-TI/II-D-Sharp Tool Delamination Depth (inches) (c) 43

57 Frequency of Occurrence S5000-F400-TI,TII,TI/II,TIII-D-Sharp Tool Delamination Depth (inches) (d) Fig 4.15 Delamination Frequency for spindle speed 5000 rpm-feed rate 400 ipm sharp tool (a) Type I Delamination (b) Type II Delamination (c) Type I/II Delamination (d) Type I, Type II, Type I/II and Type III Delamination. In Figure 4.15 for the worst cutting condition of spindle speed of 5000 rpm and feed rate of 400 ipm, and when the tool is sharp we see that Types I and I/II delamination occurring frequently and Type II delamination occurring at lower frequency. Type III delamination did not have a significant occurrence frequency and hence it is not shown here. When the delamination types are combined we see almost a normal distribution for the delamination depth. 44

58 Frequency of Occurrence S5000-F400-TI-D-Worn Tool Delamination Depth (inches) (a) (b) Frequency of Occurrence S5000-F400-TI/II-D-Worn Tool Delamination Depth (inches) Frequency of Occurrence S5000-F400-TIII-D-Worn Tool Delamination Depth (inches) (c) 45

59 Frequency of Occurrence S5000-F400-TI,TII,TI/II,TIII-D-Worn Tool Delamination Depth (inches) (d) Fig 4.16 Delamination Frequency for spindle speed 5000 rpm-feed rate 400 ipm worn tool (a) Type I Delamination (b) Type I/II Delamination (c) Type III Delamination (d) Type I, Type II, Type I/II and Type III Delamination. In Figure 4.16 for the worst cutting condition of spindle speed of 5000 rpm, feed rate of 400 ipm and when the tool is worn we see Types I and I/II delamination occurring at high frequency. When the types of delamination are combined we see a normal distribution pattern for the delamination depth. Also in the Figure for combined types of delamination we see the delamination depth is more for worn tool than for the sharp tool. For both conditions of the tool we see Type I and Type I/II delamination occurring most. 46

60 (a) Frequency of Occurrence S15000-F100-TI-D-Sharp Tool Delamination Depth (inches) Frequency of Occurrence S15000-F100-TII-D-Sharp Tool Delamination Depth (inches) (b) (c) Frequency of Occurrence S15000-F100-TI/II-D-Sharp Tool Delamination Depth (inches) 47

61 Frequency of Occurrence S15000-F100-TIII-D-Sharp Tool Delamination Depth (inches) (d) Frequency of Occurrence S15000-F100-TI,TII,TI/II,TIII-D-Sharp Tool Delamination Depth (inches) (e) Fig 4.17 Delamination Frequency for spindle speed rpm-feed rate 100 ipm sharp tool (a) Type I Delamination (b) Type II Delamination (c) Type I/II Delamination (d) Type III Delamination (e) Type I, Type II, Type I/II and Type III Delamination. In Figure 4.17 for the best cutting condition of spindle speed rpm, feed rate 100 rpm and when the tool is sharp we see Type I, Type II, Type I/II, Type III delamination with Type II and Type I/II delamination more prominent. Here we see the occurrence of the types of delamination and the depth of delamination is minimum since 48

62 this is the point at which the chip load is minimum. Here the combined frequency graph of delamination types follows a normal distribution pattern. (a) Frequency of Occurrence S15000-F100-TI/II-D-Worn Tool Delamination Depth (inches) Frequency of Occurrence S15000-F100-TIII-D-Worn Tool Delamination Depth (inches) (b) 49

63 (c) Frequency of Occurrence S15000-F100-TI,TII,TI/II,TIII-D-Worn Tool Delamination Depth (inches) Fig 4.18 Delamination Frequency for spindle speed rpm-feed rate 100 ipm worn tool (a) Type I/II Delamination (b) Type III Delamination (c) Type I, Type II, Type I/II, Type III Delamination. In Figure 4.18 we see Type I/II and Type III delamination with Type I/II more prominent. Here the delamination depth is more as compared to that in Figure 4.17, this is due to the tool wear since as wear increases delamination depth increases. The combined frequency graph follows a normal distribution pattern. In general and based on these two extreme cutting conditions we see Type I and Type I/II delamination are the most common types of delamination damage occurring. This type of damage phenomena can be assumed for the other combinations of spindle speed and feed rate. Also the frequency distribution graph for the combined delamination types follows a normal distribution pattern for the two extreme cutting conditions and for both conditions of the tool. Since the data for the delamination frequency are normally distributed, the average and standard deviation can be used to draw general conclusions about delamination depth for the selected process parameter. The average gives the extent of delamination depth for each combination of spindle speed and feed rate. Standard 50

64 deviation gives the variation of the delamination data for each combination of spindle speed and feed rate. The average and standard deviation values for delamination depth frequency distributions obtained in these experiments are given in appendix. Furthermore the average delamination depth of the combined frequency distribution will quantify the extent of delamination of any given cutting condition from this point forward. 4.5 Delamination of Machined CFRP CFRP undergoing trimming operations are subjected to serious delamination on the top and bottom surface plies because these plies are not constrained on both sides by adjacent plies. Acceptance/rejection of these machined pieces depends on the size of delamination. As per the industrial standard, threshold delamination depth was taken as 2.5 mm [8]. After each run of the experiment the workpiece is analyzed for delamination depth for each matrix combination. The experimental matrix is repeated twice and average delamination depth of the two experiments is calculated and plotted. The actual delamination values for the two experiments are given in appendix. Figures show the line graphs plotted for average delamination depth, average delamination depth for threshold values of 1.0 mm and 1.5 mm. Average delamination depth means the average of all types of delamination seen. Average delamination depth for threshold value of 1.00 mm means the average of all types of delamination greater than 1.00 mm. Similarly average delamination for threshold value of 1.5mm means the average taken for all types of delamination greater than 1.5mm. These graphs are plotted to see the extent of delamination depth for each combination of spindle speed and feed rate and set the optimum process parameters. 51

65 Average Delamination Depth-S5000 Avg Depth (inches) Cutting Distance (inches) aeff f100 aeff f200 aeff f400 (a) Average Delamination Depth-S10000 Avg Depth (inches) Cutting Distance (inches) aeff f100 aeff f200 aeff f400 (b) Average Delamination Depth-S15000 Avg Depth (inches) Cutting Distance (inches) aeff f100 aeff f200 aeff f400 (c) Fig 4.19 Average Delamination Depth (a) Spindle speed 5000 rpm (b) Spindle speed rpm (c) S15000 rpm for feed rates of 100,200 and 400 ipm. 52

66 Average Delamination Depth>1.0mm-S5000 Avg Depth (inches) Cutting Distance (inches) aeff f100 aeff f200 aeff f400 (a) Average Delamination Depth>1.0mm-S10000 Avg Depth (inches) Cutting Distance (inches) aeff f100 aeff f200 aeff f400 (b) Average Delamination Depth>1.0mm-S15000 Avg Depth (inches) Cutting Distance (inches) aeff f100 aeff f200 aeff f400 (c) Fig 4.20 Average Delamination Depth >1.0mm (a) Spindle speed 5000 rpm (b) Spindle speed rpm (c) Spindle speed rpm for feed rates of 100,200 and 400 ipm. 53

67 Average Delamination Depth>1.5mm-S5000 Avg Depth (inches) Cutting Distance (inches) aeff f100 aeff f200 aeff f400 (a) Average Delamination Depth>1.5mm-S10000 Avg Depth (inches) Cutting Distance (inches) aeff f100 aeff f200 aeff f400 (b) Average Delamination Depth>1.5mm-S15000 Avg Depth (inches) Cutting Distance (inches) aeff f100 aeff f200 aeff f400 (c) Fig 4.21 Average Delamination Depth > 1.5 mm (a) Spindle speed 5000 rpm (b) Spindle speed rpm (c) Spindle speed rpm for feed rates of 100,200 and 400 ipm. 54

68 In Figures it is seen that average delamination depth increases with an increase in cutting distance, increases with an increase in feed rate and decreases with increase in spindle speed. The increase with cutting distance is due to the wear of the tool since wear of the tool increases with an increase in cutting distance. Wear of the tool causes cutting forces to increase and cutting forces cause delamination. Delamination depth increase is more prominent with respect to feed rate since as feed rate increases the chip load increases, and thus increases cutting forces resulting in serious delamination. Delamination decreases with increase in spindle speed since an increase in spindle speed results in a decrease in chip load. Figures show the average delamination depth for threshold values at 1.0 mm and 1.5 mm respectively. Here also we see delamination depth increase above the threshold is more prominent with respect to feed rate. Delamination depth increase with respect to cutting distance is there but not obvious. For graphs plotted for threshold value of 1.5mm we see no delamination depth higher than 1.5mm (0.06inch) at spindle speed of rpm and feed rate of 100 ipm. These line graphs show that delamination depth is minimum at high spindle speeds and low feed rates, which corresponds to low chip load. Then 3D-graphs are plotted for the state of the tool when it is sharp and worn for process optimization. Figure 4.22 shows the 3D-graphs for average delamination depth for the state of the tool when it is sharp and worn. Graphs show that average delamination depth increases with respect to feed rate and decreases with increase in spindle speed. Also we see that with sharp tool there is less delamination than with the worn tool. This is because for sharp tool the contact area between the tool and workpiece will be less 55

69 resulting in clean cut with less damage while for worn tool the contact area between tool and workpiece will be more resulting in rough cut with heavy damage. Average Delamination Depth-Sharp Tool 0.08 Avg Depth (inches) S5000 S10000 S F100 Feed Rate (ipm) F200 F400 S15000 S10000 S5000 Speed (rpm) (a) Average Delamination Depth Sharp Tool Average Delamination Depth-Worn Tool 0.08 Avg Depth (inches) S5000 S10000 S F100 Feed Rate (ipm) F200 F400 S15000 S10000 S5000 Speed (rpm) (b) Average Delamination Depth Worn Tool Fig 4.22 Average Delamination Depth (a) Sharp Tool (130.5 inches) (b) Worn Tool (1020 inches). 56

70 Average Delamination Depth>1.0mm-Sharp Tool 0.08 Avg Depth (inches) S5000 S10000 S F100 Feed Rate (ipm) F200 F400 S15000 S10000 S5000 Speed (rpm) (a) Average Delamination Depth > 1.0mm Sharp Tool Average Delamination Depth >1.0mm-Worn Tool 0.08 Avg Depth (inches) S5000 S10000 S F100 Feed Rate (ipm) F200 F400 S15000 S10000 S5000 Speed (rpm) (b) Average Delamination Depth > 1.0mm Worn Tool Fig 4.23 Average Delamination Depth >1.0mm (a) Sharp Tool (130.5 inches) (b) Worn Tool (1020 inches). 57

71 Average Delamination Depth >1.5-Sharp Tool Avg Depth (inches) S5000 S10000 S F100 Feed Rate (ipm) F200 F400 S15000 S10000 S5000 Speed (rpm) (a) Average Delamination Depth > 1.5mm Sharp Tool Average Delamination Depth >1.5mm-Worn Tool 0.08 Avg Depth (inches) S5000 S10000 S F100 Feed Rate (ipm) F200 F400 S15000 S10000 S5000 Speed (rpm) (b) Average Delamination Depth > 1.5mm Worn Tool Fig 4.24 Average Delamination Depth >1.5mm (a) Sharp Tool (130.5 inches) (b) Worn Tool (1020 inches). 58

72 In Figure 4.24, it is seen that irrespective of the state of the tool at spindle speed of rpm and feed rate of 100 ipm the average delamination depth is below 1.5mm. This is well below the standard industrial threshold value of 2.5mm. So at this combination of spindle speed and feed rate there is minimum delamination and better surface finish. Surface quality at this point indicates that the wear of the tool is minimum. At this point the tool life will be more which results in reducing production cost. All the above graphs indicate that high spindle speed and low feed rate is better for obtaining machined surfaces with minimum damage. Optical microscopic pictures were also taken to validate the graphical results. Figures below show the microscopic pictures of the machined edges when it is subjected to sharp tool (R1) and to worn tool (R5) conditions for the three spindle speeds and three feed rate combinations. In all the pictures we see that delamination increases with an increase in feed rate, increases with an increase in cutting distance and decreases with an increase in spindle speed. These results are same as those shown by the line and 3D graphs. We also see in these pictures that bottom plies are subjected to more damage than the top plies. This is due to the tool geometry in which the axial tool force acts downward, resulting in more damage on the bottom plies of the laminate. We see that for all combination of spindle speeds 5000, and rpm and feed rate of 100 ipm and when the tool condition is sharp the machined edge is almost free of delamination. From these observations we see that the worst cutting condition for surface quality is low spindle speed, high feed rate and best cutting condition for surface quality is high spindle speed and low feed rate. 59

73 (a) S5000-F inches-sharp S5000-F inches-worn (b) S5000-F inches-sharp S5000-F inches-worn Fig 4.25 Microscopic pictures of machined edge for spindle speed 5000 rpm at feed rate 100 ipm (a) Bottom surface (b) Top surface. Machining along horizontal direction from left to right. 60

74 (a) S5000-F inches-sharp S5000-F inches-worn (b) S5000-F inches-sharp S5000-F inches-worn Fig 4.26 Microscopic pictures of machined edge for spindle speed 5000 rpm at feed rate 200 ipm (a) Bottom surface (b) Top surface. Machining along horizontal direction from left to right. 61

75 (a) S5000-F inches-sharp S5000-F inches-worn (b) S5000-F inches-sharp S5000-F inches-worn Fig 4.27 Microscopic pictures of machined edge for spindle speed 5000 rpm at feed rate 400 ipm (a) Bottom surface (b) Top surface. Machining along horizontal direction from left to right. 62

76 (a) S10000-F inches-sharp S10000-F inches-worn (b) S10000-F inches-sharp S10000-F inches-worn Fig 4.28 Microscopic pictures of machined edge for spindle speed rpm at feed rate 100 ipm (a) Bottom surface (b) Top surface. Machining along horizontal direction from left to right. 63

77 (a) S10000-F inches-sharp S10000-F inches-worn (b) S10000-F inches-sharp S10000-F inches-worn Fig 4.29 Microscopic pictures of machined edge for spindle speed rpm at feed rate 200 ipm (a) Bottom surface (b) Top surface. Machining along horizontal direction from left to right. 64

78 (a) S10000-F inches-sharp S10000-F inches-worn (b) S10000-F inches-sharp S10000-F inches-worn Fig 4.30 Microscopic pictures of machined edge for spindle speed rpm at feed rate 400 ipm (a) Bottom surface (b) Top surface. Machining along horizontal direction from left to right. 65

79 (a) S15000-F inches-sharp S15000-F inches-worn (b) S15000-F inches-sharp S15000-F inches-worn Fig 4.31 Microscopic pictures of machined edge for spindle speed rpm at feed rate 100 ipm (a) Bottom surface (b) Top surface. Machining along horizontal direction from left to right. 66

80 (a) S15000-F inches-sharp S15000-F inches-worn (b) S15000-F inches-sharp S15000-F inches-worn Fig 4.32 Microscopic pictures of machined edge for spindle speed rpm at feed rate 200 ipm (a) Bottom surface (b) Top surface. Machining along horizontal direction from left to right. 67

81 (a) S15000-F inches-sharp S15000-F inches-worn (b) S15000-F inches-sharp S15000-F inches-worn Fig 4.33 Microscopic pictures of machined edge for spindle speed rpm at feed rate 400 ipm (a) Bottom surface (b) Top surface. Machining along horizontal direction from left to right. 68