Study on failure of surface and interface of the laminated composite structures of a small ship

Doctor of Philosophy Study on the failure of surface and interface of the laminated composite structures of a small ship The Graduate School of the University of Ulsan Department of Me

Pham Thanh Nhut

Doctor of Philosophy

Study on the failure of surface and interface of the laminated composite structures of a small ship

The Graduate School

of the University of Ulsan Department of Mechanical Engineering

Pham Thanh Nhut

Study on the failure of surface and interface of the laminated composite structures of a small ship

이 논문을 공학박사 학위 논문으로 제출함

2013년 12월

울산대학교 대학원 기계공학과

Pham Thanh Nhut

Study on the failure of surface and interface of the laminated composite structures of a small ship

Supervisor: Prof Young Jin Yum

A Dissertation

Submitted to the Graduate School of the University of Ulsan

In partial Fulfillment of the Requirements

for the Degree of

Doctor of Philosophy

by

Pham Thanh Nhut

Department of Mechanical Engineering University of Ulsan, Republic of Korea

Dec., 2013

Pham Thanh Nhut 의 공학박사 학위 논문을 인준함

Study on the failure of surface and interface of the laminated composite structures of a small ship

This certifies that the dissertation

of Pham Thanh Nhut is approved

Committee Chairman Prof Seok Jae Chu

Committee Member Prof Il Sup Chung

Committee Member Prof Doo Man Chun

Committee Member Prof Seung Tae Choi

Committee Member Prof Young Jin Yum

Department of Mechanical Engineering University of Ulsan, Republic of Korea

Dec., 2013 ACKNOWLEDGEMENTS

I would like to express my heartfelt gratitude to my advisor, Prof Young Jin Yum, who not only gives me priceless guidance in scientific work but also supports me the financial help for three years studying in the University of Ulsan

I would like to thank the professors in the committee for their suggestions and comments throughout the study Sincerely thank to University of Ulsan that grants me the scholarship from the 2011 Research Fund I also express my gratitude to professors and staff members of the School of Mechanical Engineering for their kind help Special thank

to Korea for living and studying opportunity and Korean people for their civilization and friendship

I would like to thank my family members, who encourage and suffer a separation during the period I live away from home Specially thank to all members of the Structure and Composite Materials laboratory who support me a lot in my research works

I would like to thank the Union of Vietnamese Students at University of Ulsan for spiritual help and consulting abroad living and studying experience, and close friends in Vietnam who always look forward to my returning our motherland

Ulsan, Republic of Korea Dec., 2013

PHAM THANH NHUT

CONTENTS

Abstract iv

List of figures vi

List of tables ix

Nomenclature x

Abbreviations xii

Chapter 1 : Introduction ……… 1

1.1 Overview ……… 1

1.1.1 Thermoset composite materials ……… 1

1.1.2 Properties and application of composite materials in shipbuilding ………… 3

1.2 Surface characteristics of lamina composite ……… 4

1.3 Problem statement, necessity, objectives and thesis layout ……… 6

Chapter 2: Analysis of typical surface structures and design of test specimens …… 9

2.1 Joining structures ……… 9

2.1.1 Analysis of structures ……… 9

2.1.2 Design of test specimens ……… 11

2.2 Propeller structures ……… 14

2.2.1 Analysis of structures ……… 14

2.2.2 Design of test specimens ……… 16

2.3 Coating structures ……… 19

2.3.1 Analysis of structures ……… 19

2.3.2 Design of test specimens ……… 19

Chapter 3: Failure of lap-joints with different surface properties ……… 21

3.1 Introduction ……… 21

3.2 Theory and experiment ……… 23

3.2.1 Method ……… 23

3.2.2 Theory ……… 23

3.2.3 Materials and making of specimens ……… 25

3.2.4 Mechanical tests ……… 25

3.3 Results and discussions ……… 28

3.3.1 Result of tensile and in-plane test ……… 28

3.3.2 Shear strength in y-z plane ……… 32

3.3.3 Shear strength in x-z plane ……… 35

3.3.4 Shear strength in x-y plane ……… 37

3.3.5 Combination of shear stress for maximum and average values ……… 39

3.4 Conclusions ……… 41

Chapter 4: Failure of propeller blades surface by cavitation erosion phenomenon 42

4.1 Introduction ……… 42

4.2 Theory and experiment ……… 44

4.2.1 Method ……… 44

4.2.2 Materials and samples preparation ……… 44

4.2.2.1 Manufacturing process of mold ……… 45

4.2.2.2 Evaluation of mold ……… 46

4.2.2.3 Manufacturing process of propeller ……… 56

4.2.2.4 Preparation of circulating water channel (CWC) ……… 57

4.2.3 Test conditions ……… 58

4.3 Results and discussions ……… 59

4.3.1 Weight loss of blades ……… 59

4.3.2 Position and damage on blade surface ……… 62

4.3.3 Area and propagation of cavitation ……… 65

4.4 Conclusions ……… 68

Chapter 5: Failure of gelcoat surface by print-through phenomenon ……… 70

5.1 Introduction ……… 70

5.2 Theory and experiment ……… 71

5.2.1 Method ……… 71

5.2.2 Theory ……… 72

5.2.3 Materials and specimens preparation ……… 73

5.2.4 Experimental organization ……… 75

5.3 Results and discussions ……… 76

5.3.1 The values of surface roughness ……… 76

5.3.2 The microphotograph of surface ……… 80

5.3.3 Prediction of surface roughness ……… 86

5.4 Conclusions ……… 91

Chapter 6: Conclusions and future research ……… 93

6.1 Conclusions ……… 93

6.2 Future researches 94

6.2.1 Single lap-joints problem 94

6.2.2 Composite propeller problem 97

6.2.3 Gelcoat surface problem 97

References ……… 99

Bibliography ……… 103

List of publications ……… 106

Appendix ……… 108

Lamina composite is used in most structures of composite ship like hull, bulkhead,

frame or propeller Surface and interface properties of this material should be considered

thoroughly because they can affect the strength, hardness or aestheticism of products The

failures of surface and interface were studied and evaluated by experimental investigation

through three typical structures of a composite boat

For failure of single lap-joints, unsaturated polyester/glass fiber (woven and mat) was

used for composite panel samples manufactured by hand lay-up and vacuum method at

room temperature The polyester resin was used as an adhesive Joining process of the

samples was carried out on six types of surface with two different methods: terminal-joint and secondary-joint Geometry properties of all bondings were the

same Mechanical properties of adhesive, composite adherends, terminal-joint and

secondary-joint specimens were determined by experiments Combinations between the

experiment results and bonding theory were used in this study The maximum and average

shear stresses were calculated by maximum tensile and in-plane force and geometry

parameters of specimens The results of maximum and average shear stresses were

compared and evaluated in six joints together The results showed that grinding,

grinding/acetone terminal-joint and mat-mat, mat-woven secondary-joint had the high

strength Conversely, no treatment and woven-woven bonding had very low strength In

addition, the failure occurred strongly at two ends and then moved toward the middle area

of overlap length

For failure of composite propeller blades surface, the cavitation erosion phenomenon

was tested directly A small three blades propeller was made from fiberglass combined

with epoxy, polyester and gelcoat resin This propeller was tested in the circulating water

channel whose revolution and depth were determined by conditions of cavitation number

Cavitation erosion was observed and evaluated by images, weight loss and cavitation area

From the results, it was observed that the cavitation erosion occurred around 0.4R and

0.7R (on both pressure and suction surface and both leading and trailing edge) Most

damage regions propagated along the circumferential direction of propeller Ge-GF blade had the best property, Po-GF blade had the worst property and the quality of Ep-GF blade was medium

For failure of gelcoat surface, print-through phenomenon (PTP) was investigated on the specimens The samples were fabricated with gelcoat resin material only Three input factors (variables) were studied including thickness of layer, hardener fraction and temperature of making The specimens were tested to determine the roughness and microphotograph of PTP Then, the multiple regression analysis was used to obtain the experimental regression equations Results indicated that the effect of three input factors

on PTP was different: hardener fraction was the strongest and thickness was the weakest The PTP increases quickly with upper 25oC and 2% hardener The amount, area and depth

of points and cracks also increased when the input factors value increased Besides, the values of roughness or input factors can be predicted by experimental regression equations Finally, future works were given and performance method was suggested

Key words: surface, terminal-joint, secondary-joint, composite propeller, cavitation

erosion, gelcoat, PTP

LIST OF FIGURES

Fig 1-1 Influence of reinforcement type and quantity on composite properties………… 2

Fig 1-2 Manufacturing of E-glass mat and woven roving……… 4

Fig 1-3 Surfaces of lamina composite structure……… 5

Fig 2-1 Co-curing bonding……… 9

Fig 2-2 Making of secondary layer……… 10

Fig 2-3 Making of terminal-joint……… 11

Fig 2-4 Three surface pairs of secondary-joint……… 11

Fig 2-5 Surface of terminal-joint: (a) Three pairs of surface, (b) Morphology of surface 12 Fig 2-6 Geometry of tensile test specimen……… 12

Fig 2-7 Geometry of in-plane test specimen……… 13

Fig 2-8 Position of propeller of small boat……… 14

Fig 2-9 Types of small propeller……… 14

Fig 2-10 Structure of small propellers……… 15

Fig 2-11 Linking of blade and hub……… 15

Fig 2-12 Aluminium alloy mold of blade……… 17

Fig 2-13 Circulating water channel standard……… 17

Fig 2-14 Vibratory apparatus for cavitation erosion test……… 18

Fig 2-15 Composite circulating water channel……… 18

Fig 2-16 Application of gelcoat resin……… 19

Fig 2-17 Print-through phenomenon……… 20

Fig 2-18 Structure of mold and manufacturing process……… 20

Fig 3-1 Results of tensile loading test: (a) No treatment terminal-joint, (b) Grinding terminal-joint, (c) Grinding/acetone terminal-joint, (d) Mat/mat secondary-joint, (e) Mat/woven secondary-joint, (f) Woven/woven secondary-joint ………… 27

Fig 3-2 Results of tensile loading test: (a) No treatment terminal-joint, (b) Grinding terminal-joint, (c) Grinding/acetone terminal-joint, (d) Mat/mat secondary-joint, (e) Mat/woven secondary-joint, (f) Woven/woven secondary-joint ……… 30

Fig 3-3 Results of in-plane loading test……… 32

Fig 3-4 Shear stress in y-z plane along y……… 33

Fig 3-5 Ratio of (yz)/(yz)ave vs.y/c in tensile test case……… 34

Fig 3-6 Shear stress in x-z plane along y……… 34

Fig 3-7 Ratio of (xz)/(xz)ave vs y/c in in-plane test……… 37

Fig 3-8 Shear stress in x-y plane along y……… 38

Fig 3-9 Ratio of (xy)/(xy)avevs.y/c in in-plane test……… 39

Fig 3-10 Average shear stress for all cases……… 40

Fig 3-11 Maximum shear stress for all cases……… 40

Fig 4-1 Aluminium propeller model……… 45

Fig 4-2 Parts of composite mold……… 45

Fig 4-3 Making of mold from propeller model……… 46

Fig 4-4 Hardness specimens and testing……… 46

Fig 4-5 Bending and compression testing……… 48

Fig 4-6 Load-displacement of mold only……… 48

Fig 4-7 Load-displacement of mold & mixture inside……… ……… 49

Fig 4-8 Results of deflection along radius: a) displacement at 0.7R , b) Load-displacement at 0.8R , c) Load-Load-displacement at 0.9R ……… 51

Fig 4-9 Comparison of deflection……… 51

Fig 4-10 Position and method of measurement……… 52

Fig 4-11 Error of Al propeller/GFPG mold pair……… 54

Fig 4-12 Error of Al propeller/GRP propeller pair……… 55

Fig 4-13 Manufacturing process……… 56

Fig 4-14 Composite propeller……… 56

Fig 4-15 Composite CWC……… 57

Fig 4-16 Experiment of composite propeller……… 59

Fig 4-17 Weight loss of three blades……… 61

Fig 4-18 Weight loss rate of three blades……… 61

Fig 4-19 Weight loss ratio of three blades……… 61

Fig 4-20 Position and area of cavitation for each time……… 64

Fig 4-21 Microphotograph of collapse of surfaces……… 65

Fig 4-22 Propagation of cavitation……… 66

Fig 4-23 Development of cavitation area……… 67

Fig 4-24 Rate of cavitation area every hour……… 67

Fig 5-1 Gelcoat and hardener 73

Fig 5-2 Manufacturing process of specimens: a) Nine specimens at 15oC, b) Nine specimens at 25oC, a) Nine specimens at 35oC……… 75

Fig 5-3 Measurement of roughness 76

Fig 5-4 Effect of thickness on roughness: a) Roughness Ra, b) Roughness Rz………… 77

Fig 5-5 Effect of hardener fraction on roughness: a) Roughness Ra, b) Roughness Rz…78 Fig 5-6 Effect of temperature on roughness: a) Roughness Ra, b) Roughness Rz……… 79

Fig 5-7 Forms of PTP on gelcoat surface: a) At 0.5mm thickness, b) At 1.0mm thickness, c) At 1.5mm thickness 83

Fig 5-8 Change of surface state……… 85

Fig 6-1 Change of overlap length 95

Fig 6-2 Reinforcement at two ends 95

Fig 6-3 In-plane test specimens: b) T shape b) Y shape 96

Fig 6-4 PIRANHA propellers are 17% stronger than aluminum propeller 97

Fig 6-5 Change of underlying lamina structure……… 98

LIST OF TABLES

Table 3-1 Geometric characteristics of joint samples……… 26

Table 3-2 Material properties of polyester resin and composite.……… 27

Table 3-3 Results of tensile loading test and the calculated shear stresses from theory… 33 Table 3-4 Results of in-plane loading test and calculated shear stress xz……… 36

Table 3-5 Results of in-plane loading test and calculated shear stress xy……… 38

Table 4-1 Characteristics of propeller model……… 44

Table 4-2 Hardness of materials……… 47

Table 4-3 Roughness of materials……… 47

Table 4-4 Coordinate of pressure surface……… 53

Table 4-5 Thickness of blade……… 53

Table 4-6 Coordinate error of mold and thickness error of blade compared to Al propeller……… 54

Table 4-6 Weight of blade after 500 hours of experiment……… 60

Table 4-7 Rate of weight loss……… 60

Table 4-8 Ratio of weight loss and weight of blade……… 60

Table 4-9 Area of cavitation erosion……… 66

Table 5-1 Code of specimens……… 73

Table 5-2 Properties of gelcoat resin……… 74

Table 5-3 Input data for SPSS software……… 86

Table 5-4 Simple regression analysis: a) For dependent variable Ra, b) For dependent variable Rz ……… 87

Table 5-5 Multiple regression analysis for group 1: a) For dependent variable Ra, b) For dependent variable Rz ……… 88

Table 5-6 Multiple regression analysis for group 2: a) For dependent variable Ra, b) For dependent variable Rz ……… 88

Table 5-7 Result of roughness error: a) Error of Ra, b) Error of Rz 89

Table 5-8 Ra & Rz of hull surface from practical conditions……… 91

Table 5-9 Hardener fraction at the same Ra (minimum)……….……… 91

Table 5-10 Hardener fraction for different surfaces at 20oC, 1mm……… 91

Ga : Shear modulus of adhesive (MPa)

h : Depth of immersion of propeller axis (m)

pv : Saturation vapour pressure (KN/m2)

R : Radius of propeller (m), coefficient of correlation

R2 : Coefficient of multiple determination

Ra : Arithmetic mean value of roughness (m)

Rz : Mean height of profile irregularities of roughness(m)

ta : Thicknesses of adhesive, mm

ti : Thicknesses of inner adherend, mm

to : Thicknesses of outer adherend, mm

y : Length of joint (mm), dependent variable

z : Number of blade propeller

Greek symbols

i : Coefficientdetermining the contribution of the independent variable

 : Density of sea water (T/m3)

 : Cavitation number, standard deviation

 : Shear stress of outer adherend in y-z plane (MPa)

τult : Ultimate shear stress

COP : Cavitation occurrence probability

CWC : Circulating water channel

PVC : Polyvinyl chloride

rpm : Revolution per minute

RTM : Resin transfer molding

C h a p t e r 1

Introduction

1.1 Overview

1.1.1 Thermoset composite materials

A composite material can be defined as a combination of two or more materials that results in better properties than those of the individual components used alone In contrast

to metallic alloys, each material retains its separate chemical, physical, and mechanical properties The two constituents are a reinforcement and a matrix The main advantages of composite materials are their high strength and stiffness, combined with low density, when compared with bulk materials, allowing for a weight reduction in the finished part The reinforcing phase provides the strength and stiffness In most cases, the reinforcement is harder, stronger, and stiffer than the matrix The reinforcement is usually a fiber or a particulate [1] Popular fibers include glass, carbon and aramid, which may be continuous

or discontinuous Continuous-fiber composites normally have a preferred orientation, while discontinuous fibers generally have a random orientation Examples of continuous reinforcements include unidirectional, woven cloth, and helical winding, while examples

of discontinuous reinforcements are chopped fibers and random mat Continuous-fiber composites are often made into laminates by stacking single sheets of continuous fibers in different orientations to obtain the desired strength and stiffness properties with fiber volumes as high as 60 to 70 percent The type and quantity of the reinforcement determine the final properties (Fig 1-1)

The continuous phase is the matrix, which is a polymer, metal, or ceramic Polymers have low strength and stiffness, metals have intermediate strength and stiffness but high ductility, and ceramics have high strength and stiffness but are brittle The matrix

(continuous phase) performs several critical functions, including maintaining the fibers in the proper orientation and spacing and protecting them from abrasion and the environment

In polymer and metal matrix composites that form a strong bond between the fiber and the matrix, the matrix transmits loads from the matrix to the fibers through shear loading at the interface

Fig 1-1 Influence of reinforcement type and quantity on composite properties

The advantages of composites are many, including lighter weight, the ability to tailor the layup for optimum strength and stiffness, improved fatigue life, corrosion resistance and with good design practice, reduced assembly costs due to fewer detail parts and fasteners The specific strength (strength/density) and specific modulus (modulus/density)

of high strength fibers are higher than those of other comparable aerospace metallic alloys Applications include aerospace, transportation, construction, marine goods, sporting goods, and more recently infrastructure, with construction and transportation being the largest In general, high-performance but more costly continuous-carbon-fiber composites are used where high strength and stiffness along with light weight are required, and much

Fiber volume fractions (%)

lower-cost fiberglass composites are used in less demanding applications where weight is not as critical For the marine industry, corrosion is a major source of headache and expense Composites help minimize these problems, primarily because they do not corrode like metals or wood

1.1.2 Properties and application of composite materials in shipbuilding

During the second world war, just after polyester resins were first developed, glass fibers became available following the accidental discovery of a production process using blown air on a stream of molten glass Soon, fiberglass reinforced plastic (FRP) came into use and FRP boats started to become available in the early 1950’s Since the 1950’s, resins (polyester, epoxy and vinylester) have improved steadily and GRP has become without doubt the most prevalent composite used in boatbuilding Today, the reinforcement materials are widely used as E-glass, S-glass and carbon fiber Here, the E-glass fiber is most commonly used with two types: fiberglass mat and fiberglass woven roving (Fig 1-2) Mat is a non-woven felt consisting of chopped glass fibers 1" to 2" in length, crisscrossed and randomly interlocked and held together with a binder Since it is not a woven cloth it can be stretched to fit into difficult areas Because it is highly absorbent it soaks up a great deal of resin, and the resultant structure is stiffer than a layer of woven cloth, although not

as strong Matting is often used between layers of woven cloth or roving, or as a waterproof layer when there is no gel-coat Woven rovings are composed of direct rovings woven into a fabric The input rovings are designed to give rapid wet-out and excellent laminate properties The construction gives bi-directional (0°/90°) reinforcement and the strength of continuous filaments Woven roving is a high strength, coarsely woven fabric widely used in all phases of fiberglass molding It has replaced all other lightweight grades where speed of wet-out, durability and flatness of weave are important Higher physical properties are possible in laminates of this grade because of the reduced number of interstices in the fabric

Because of low cost, unsaturated polyester resin, fiberglass mat (density of 225, 300, 450g/m2, etc.) and fiberglass woven roving (density of 150, 360, 570g/m2, etc.) are used for most structures of ship (hull, deck, bulkhead, etc.) These products can be manufactured

by hand lay-up or machine at room temperature Epoxy, vinylester resin and carbon fiber are used only for high strength or resistant chemical structures (propeller, engine-bed or fuel tank) For the products making from glass or carbon fiber and polyester, epoxy or vinylester resin, the mold and surface layer are very important The mold can be made from wood or metal and most surface layer is gelcoat resin Gelcoats are designed to be durable, providing resistance to ultraviolet degradation and hydrolysis Specialized gelcoats can be used to manufacture the molds which in turn are used to manufacture components Suitable resin chemistries for the manufacture of gelcoats vary, but the most commonly encountered are unsaturated polyesters or epoxies

Fig 1-2 Manufacturing of E-glass mat and woven roving

(fs-yingzhe.en.made-in-china.com; rongbay.com)

1.2 Surface characteristics of lamina composite

The lamina composite products are formed from several layers So the surface is first considered including interface and outer surface (Fig 1-3) Interface occurs in whole of structures of composite boat and called joining (layer – layer, layer – plate, plate – plate,

etc.) There are three kinds of joints for glass fiber layers: co-curing, secondary-joint and terminal-joint Co-curing is the joining of layers together continuously during curing process (lay-up process) When pre-curing upper and lower adherends (unperfected surface) were bonded together, it is secondary-joint If upper and lower adherends are perfect surface, that bond is called terminal-joint Co-curing is used usually for manufacture since

it is easier to carry out and has a better ability to link between layers, while surfaces of secondary and terminal-joint must be well prepared to obtain necessary mechanical properties [2] Clearly, the quality of joints depends greatly on the preparation of surface of the adherends

Interface

Topcoat

Fiber/resin layerCoating resin

Molding

Out-surface

Fig 1-3 Surfaces of laminated composite structure

Besides, the outer surface (surfaces contact to mold) is quite important When fabrication of composite structures, a resin layer is applied first and after finish and removing the mold, it becomes the coating It relates to the strength, hardness and aesthetics (smooth, color, etc.) of product The outer surfaces can be gelcoat resin, polyester or epoxy resin For marine boat, the corrosion of surface occurs easily The gelcoat resin is used for this purpose with resistant corrosion property However, for some typical structures of ship such as propeller, the damage can cause by corrosion or Outer surface

cavitation erosion phenomenon And the surface of propeller is made from different resins,

so the cavitation is also different

In terms of aesthetics, the roughness of outer surface is an important parameter in manufacturing process The roughness can depend on the structure of inside layers (properties of resin, type of fiber, or order of layers, etc.) [3], [4] Or it is influenced by manufacturing conditions (thickness, temperature, or hardener fraction, etc.)

1.3 Problem statement, necessity, objectives and thesis layout

The amount and role of surface in composite boat is quite great The interface decides the strength of structures and outer surface impacts on the working efficiency and aesthetics So study on surface properties is necessary to improve the quality of boat In this study, we mention to some structures of composite ship which have typical surfaces such as:

 Thick plate, connected plate: for interface with lap-joint problem

 Composite propeller: for outer surface with cavitation erosion problem

 Gelcoat layer: for outer surface with print-through problem

A lot of studies have been done by researchers related to above three problems For joint problem, the most popular problems are single lap-joint and double lap-joint of composite, metal, wood, etc For cavitation erosion problem, most researches evaluated cavitation of ship propeller by theory or experiment with vibratory apparatus The print-through problem is also studied for gelcoat layer with influence of making conditions and gives the prevention methods However, these three problems are studied on other materials (for lap-joint) or other methods (for cavitation and print-through phenomenon) From above analysis, three problems of thesis are performed as follows:

 Problem 1 (joint): Tensile and in-plane tests will be performed for the single joint samples with different surface conditions (mat to mat, mat to woven and woven

lap-to woven fiberglass surfaces) and different treatment methods (no treatment, grinding and grinding/cleaning the surfaces by acetone) and the bonding theory [5] will be used to evaluate the joint strength

 Problem 2 (cavitation of propeller): Composite propeller is fabricated based on an aluminium alloy propeller (plug) The circulating water channel is also made from composite materials with enclosed equipments (electric motors, speed controller and shaft propeller) Rotational speed and depth of propeller are determined by conditions of cavitation number [6] The propeller is tested for 500 hours and cavitation erosion of blade surface is checked at each 100 hours The results are evaluated according to three parameters: weight loss, area of the cavitation and propagation of cavitation

 Problem 3 (print-through of gelcoat layer): The specimens are made of gelcoat resin material with change of three input factors: thickness of layer, hardener fraction and temperature The roughness and microphotograph of gelcoat surface are tested by roughometer Simultaneous, multiple regression model [7] is also used to give a experimental equation The PTP of gelcoat surface is evaluated on the basis of the experimental results and regression equation

Furthermore, the dissertation also used some programs to measure, calculate and process the data like Autocad 2008, Microsoft Excel Office 2007 and Statistical Package for the Social Sciences (for Multiple Regression equation)

The objectives of this research are as follows:

 Design and fabrication of tensile test specimens following ASTM D3165–07 international standard [8] and in-plane test specimens with new idea to evaluate the quality of lap-joints by different interfaces

 Design and fabrication of composite propeller and the equipments for cavitation test

An aluminium propeller model is a plug for manufacture of composite propeller mold Three blades of composite propeller are made from this mold with three different materials: fiberglass/polyester, fiberglass/epoxy and fiberglass/gelcoat resin The main equipment for experiment is circulating water channel It is also made from composite materials with enclosed equipments (electric motors, speed controller and shaft propeller)

 Design and fabrication of the roughness test specimens with gelcoat resin layer only These specimens are made based on changes in hardener fraction, temperature and thickness

 Evaluation of fracture load and shear stresses of lap-joint specimens from the results

of tensile and in-plane test and from bonding theory

 Evaluation and comparison of cavitation erosion of propeller blades by weight loss, position, area and propagation standard

 Evaluation of quality of gelcoat surface by roughness values, microphotographs of surface The roughness or manufacturing conditions are also estimated by the experimental regression equations

Judging from the objectives, the dissertation can be organized as follows:

 Chapter 2 analyzes the interfaces and outer surfaces of structures in composite ship The typical structures are selected to determine the test specimens Amount and geometric characteristics of specimens are designed to meet the theory and international standard

 Chapter 3 presents the manufacturing process of lap-joint specimens and results of tensile and in-plane test Bonding theory is used to calculate the shear stress values in y-z, x-y and x-z plane The strength of joints is evaluated to compare the quality of different surfaces

 Chapter 4 shows the method of fabrication and experiment of composite propeller and evaluation of cavitation phenomenon on propeller blade surface Composite propeller is made from available propeller model Experimental equipments are designed in accordance with standards Experimental conditions are calculated according to cavitation theory

 Chapter 5 introduces the role of gelcoat layer in composite structure and test method

to determine the roughness of gelcoat surface The results are evaluated by surface roughness and experimental regression equations

 Chapter 6 gives results from this study, gives conclusions and proposes future works

Application of first layer Application of second layer

Fig 2-1 Co-curing bonding

With the structural thickness and area significantly, we have made them many times (because of the continuous manufacturing can reduce product quality) It means that we have to use secondary-joint (SJ) In this case, generally old surface is intact without any treatment methods So, we can predict that the adhesive between old and new surface is not good For example, for a boat, special structures of this problem include hull, deck and bulkhead With 15 meters length of ship, the hull has around 12mm thickness (12 layers of mat and woven glass fiber) and 75m2 of area Commonly, we can divide fabrication process into four stages Each stage consists of 3 layers and is performed in a day It takes over 48 hours for curing of material before fabricating the next stage So we have four secondary-joints (Fig 2-2)

Fig 2-2 Making of secondary layer (aeroworks.nl; tecnologiadelosplasticos.blogspot.com)

When making the assembly or reinforcement of hull (frame, bulkhead, etc.), we have to paste different details together Besides, when repairing or reusing the semi-products, we have to connect the existing products (perfected surfaces) together, called terminal-joint (TJ) The quality of these joints depends greatly on the properties of adherend surface It means that the bonding will become better if we prepare good surface by different treatment methods (Fig 2-3) In this thesis, we consider three surfaces of bonding:

 The surfaces are remained (no treatment)

 The surfaces are processed by grinding method

 The surfaces are processed by grinding method and then are cleaned by Acetone chemical solution

Fig 2-3 Making of terminal-joint (www.bowdidgemarinedesigns.com)

2.1.2 Design of test specimens

Today, the two most common materials used are E-glass mat and E-glass woven And for secondary-joint, all bonds used unperfected surfaces So, we study on joint strength of three pairs of surface: mat-mat (SJ1), mat-woven (SJ2) and woven-woven (SJ3) (Fig 2-4) Also above analysis, all terminal-joints used perfect surfaces with different treatment methods or without treatment the surface Therefore, the number of surface pair of terminal-joint is also three: no treatment (TJ1), grinding (TJ2) and grinding/acetone (TJ3) (Fig 2-5)

Fig 2-4 Three surface pairs of secondary-joint

a) Three pairs of surface

b) Morphology of surface Fig 2-5 Surfaces of terminal-joint

This surface is quite smooth by

releasing agent film from mold

On the other hand, bonding theory [5] showed that the strength of single lap-joint is studied on three parameters: shear stress of adhesive in x-z and y-z plane (xz a & a yz) and shear stress of adherend in x-y plane (xy o ) For a yz, stress value is determined from result

of tensile test For a

xz

 andxy o , stress values are calculated from results of in-plane test Hence, we have to design two types of specimen for two tests Tensile test specimen is made by ASTM D3165-07 standard [8] (Fig 2-6) but in-plane test specimen is carried out

by different methods In this work, it is also designed for tensile test but has geometric shape especially (Fig 2-7) In these figures, y axis is along the center line of specimens (or joint area)

Fig 2-6 Geometry of tensile test specimen

Fig 2-7 Geometry of in-plane test specimen

2.2 Propeller structures

2.2.1 Analysis of structures

A propeller is a rotating fan which is used to propel the ship by using the power generated and transmitted by the main engine of the ship The transmitted power is converted from rotational motion to generate a thrust which imparts momentum to the water and the ship is moved forward (Fig 2-8) Propellers can be classified by number of blades (two, three, four, etc.), pitch of the blade (fixed pitch and controllable pitch) or materials (Fig 2-9)

Marine propellers are fabricated from corrosion resistant materials against sea water environment The materials used for making marine propeller are aluminium and stainless steel alloy Other popular materials used are alloys of nickel, aluminium and bronze which have low weight and high strength The most recent material used to make propellers is high-tech composites With advanced resins (epoxy, gelcoat, etc.) and fibers (nylon, glass and carbon), we also manufacture the composite propellers for small boats which have good properties but low cost Another important advantage is that the deformation of the composite propeller can be controlled to improve its performance [9]

Structure of big propellers has only two parts: blades and hub However, propeller of small boats can have three main parts: blades, hub and core (Fig 2-10) Blades and hub can be made together (seamless blade) or separated For separate blade, the blade is assembled into hub by different methods (Fig 2-11)

2.2.2 Design of test specimens

Because of complex shape of propellers and properties of composite materials, manufacturing

of a composite propeller is difficult Besides, even if any one blade is damaged we have to use a new propeller for type of seamless blade This is both inconvenient and costly Therefore, in this study, we made a composite propeller with separate blades On the other hand, three kinds of composite will be studied to evaluate the cavitation Hence, the composite propeller is fabricated with three different composite materials for three blades

 First blade: mat and woven glass fiber and epoxy resin (Ep-GF)

 Second blade: mat and woven glass fiber and gelcoat resin (Ge-GF)

 Third blade: mat and woven glass fiber and polyester resin (Po-GF)

In order to obtain the demanding geometry, the metal propellers are made by casting and milling methods, but only casting method for composite propellers Almost casting molds for composite propellers are fabricated from aluminium alloy (Fig 2-12) Some studies has been done by researchers related to composite material propellers and manufacturing methods Ching-Chieh Lin et al [9] fabricated a model of propeller from the prepreg of Toho HTA1200 carbon fiber/ACD8801 epoxy with a metal mold Jukka Tervamaki [10] presented a simple method of making a two bladed composite propeller with an open mold using an existing propeller as a plug McDermott J [11] mentioned the manufacturing of marine composite propeller by resin transfer molding method and gave some advantages for this method In this study, a model of small aluminium alloy propeller was used as a plug to manufacture the composite mold Then, composite propeller was made from that mold Hardness and roughness of surfaces mold, strength and deformation and compression force of mold were measured by experiment The advantages and disadvantages of composite mold were evaluated from testing results and observation

Fig 2-12 Aluminium alloy mold of blade [9]

Properties of propeller are mainly determined by experiment with circulating water channel standard (Fig 2-13) However, most cavitation erosion phenomenon of propellers

is tested by vibratory apparatus through plate specimens (Fig 2-14) In this work, we design and make a circulating water channel by composite material Depth and speed of propeller are controlled to accord with experimental conditions (Fig 2-15)

Fig 2-13 Circulating water channel standard (www.felco.ne.jp)

Fig 2-14 Vibratory apparatus for cavitation erosion test

Fig 2-15 Composite circulating water channel

2.3 Coating structures

2.3.1 Analysis of structures

Gelcoat is the decorative surface found on fiberglass parts such as boats, bathtubs and seats, typically 0.5 mm to 1.0 mm thickness This outer layer is needed for aesthetics and protection of the underlying laminate structure Chemically, it is unsaturated polyester and vinylester resin that is unreinforced but heavily filled with a complex variety of additives These additives are used to determine its color, ultraviolet stability and chemical resistance

In addition, the manufacture of fiberglass parts typically requires a gelcoat layer to aid in the release of the parts from the mold So, the outer surfaces of hull and most structures of composite boat use gelcoat layer (Fig 2-16)

Fig 2-16 Application of gelcoat (boatpaintguide.com; multitechproducts.wordpress.com)

2.3.2 Design of test specimens

Properties of gelcoat layer depend on a lot of factors such underlying laminate structure, room temperature, hardener fraction, thickness of itself or surface of mold In this work,

we study the print-through phenomenon of gelcoat layer only (not applied to other layers) and consider the effect of three factors: temperature, hardener fraction and thickness Print-through phenomenon occurs on surface with different levels (Fig 2-17) For high level, it can be observed by naked eye and called a wrinkle or a crack For low level, it is also called roughness and measured on microscope equipment So, test specimens will be fabricated from gelcoat resin only whose maximum thickness and area must be satisfied the size of microscope The mold used is glass plate because of high smoothness From

practical conditions and requirements of manufacturer for gelcoat resin, every input factor

is divided into three levels as follows:

Fig 2-17 Print-through phenomenon (www.fao.org; www.foxnews.com)

Fig 2-18 Structure of mold and manufacturing process