EXPERIMENTAL STUDIES ON MECHANICAL PROPERTIES OF ALUMINIUM / RED MUD METAL MATRIX COMPOSITE FABRICATED BY STIR CASTING METHOD A PROJECT REPORT
Submitted by
PON SWARNA RAJA PANDIAN.S (9908009128) RAGHUNATH.K (9908009138) RAJKUMAR.D (9908009149) In partial fulfillment for the t he award of the degree Of
BACHELOR OF TECHNOLOGY IN
MECHANICAL ENGINEERING
DEPARTMENT OF MECHANICAL ENGINEERING
KALASALINGAM UNIVERSITY (Kalasalingam Academy of Research and Education) KRISHNANKOIL - 626 126 Academic Academic Year (2011-2012) 1
DEPARTMENT OF MECHANICAL ENGINEERING
KALASALINGAM UNIVERSITY (Kalasalingam Academy of Research and Education) Krishnankoil- 626 126.
DECLARATION
We hereby declare that the project work done in ―EXPERIMENTAL STUDIES ON MECHANICAL PROPERTIES OF ALUMINIUM / RED MUD METAL MATRIX COMPOSITE FABRICATED BY STIR CASTING METHOD ‖ in partial fulfilment of
requirements for the award of Bachelor of Technology in Mechanical Engineering, under the supervision of Mr.RAJESH.S, Assistant Professor-I, Department Department of Mechanical Engineering, Kalasalingam University, Krishnankoil. Krishnankoil.
PON SWARNA RAJA PANDIAN.S 9908009128
RAGHUNATH.K
RAJKUMAR.D
9908009138
9908009149
B.Tech (Mech. Engg)
B.Tech (Mech. Engg)
B.Tech (Mech. Engg)
2
DEPARTMENT OF MECHANICAL ENGINEERING KALASALINGAM UNIVERSITY (Kalasalingam Academy of Research and Education) Anand Nagar, Srivilliputhur KRISHNANKOIL - 626 126
BONAFIDE CERTIFICATE
Certified that this project report ―EXPERIMENTAL STUDIES ON MECHANICAL PROPERTIES OF
ALUMINIUM / RED RED MUD METAL METAL MATRIX COMPOSITE COMPOSITE
FABRICATED BY STIR CASTING METHOD ” is the bonafide work of ― PON SWARNA
RAJA
PANDIAN.S
(9908009128),
RAGHUNATH.K
(9908009138),
RAJKUMAR.D (9908009149) ” who carried out the project work under my supervision .
Dr.S.Rajakarunakaran
Mr.Rajesh.S
HEAD OF THE DEPARTMENT
SUPERVISOR, Assistant Assistant Professor Professor I Department of Mechanical Mechanical Engineering
Department of Mechanical Engineering
Project Viva-voce held on _______________
Internal Examiner
External Examiner
3
ACKNOWLEDGEMENT
A project of this magnitude and nature requires kind cooperation and support for successful completion. At the outset, we wish to express our sincere gratitude to our ViceRadhakrishnan for giving us kind permission to do this project. chancellor Dr. S. Radhakrishnan
We express our deep sense of gratitude and indebtedness to Mr. S. Rajesh, Assistant Professor – Professor – I, I, Department of Mechanical Engineering, Kalasalingam University, for his kind guidance and encouragement encouragement throughout t hroughout this project work. We also express our hearty gratitude to Dr. S. Rajakarunakaran, Senior Professor and Head of the department, Department of Mechanical Engineering, Kalasalingam University, for his valuable suggestion. We express our deep sense of gratitude and indebtedness to Dr. J.T.Winowlin Jappes, Professor & Head of CENTER FOR COMPOSITE MATERIALS, Department of
Mechanical Engineering, Kalasalingam University and Prof. N. Sureshkumar, Associate Professor, Department of Civil Engineering, Kalasalingam University, for their valuable suggestions and helping us in carrying out our experiments in the Tribometer and Brinell hardness testing Machine respectively respectively during the t he course of our project. We also express our hearty gratitude to all the Faculty members of CENTER FOR COMPOSITE MATERIALS , Department of Mechanical Engineering, and Kalasalingam
University, for their valuable advises in our progress. We also extend our sincere thanks to Mr. A. Marimuthu, Lab Technician (Machine Shop) Mr. R. Muthumani, Lab Technician (Machine Shop), Mr.B. Palmurugan, Lab Technician (Fitting Shop), Mr. C. Pandimuneeshwaran, Lab Technician (Center For Composite Composite Materials) Mr. M. Suryakumar , Lab Technician (Strength of Materials Lab) for providing us help throughout the project.
4
Abstract Red mud emerges as the major waste material during production of alumina from bauxite by the Bayer‘s process. It comprises of oxides of iron, titanium, aluminium and silica along with some other minor constituents. Based on economics as well as environmental related issues, enormous efforts have been directed worldwide towards red mud management issues i.e. of utilization, storage and disposal. Different avenues of red mud utilization are more or less known but none of them have so far proved to be economically viable or commercially feasible. It is generally agreed that resistance to wear of MMCs is created by reinforcement and also the wear properties are improved remarkably by introducing hard inter metallic compound into the aluminium matrix. The reinforcing materials are generally SiC, Al2O3, TiB2 etc and are costly. The present research work has been undertaken with an objective to explore the use of red mud as a reinforcing material as a low cost option. This is due to the fact that red mud alone contains all these reinforcement elements and is plentifully available. Experiments have been conducted under laboratory condition to assess the wear characteristics of the aluminium red mud composite under different working conditions in pure sliding mode on a pin-on-disc machine. This has been possible by fabricating the samples through usual stir casting technique. The samples are fabricated with LM 25 matrix material with different % of red mud material. Dispersion of red mud particles in aluminium matrix improves the hardness of the matrix material and also the wear behaviour of the composite. While increasing the % of reinforcements of the material there is decrease in wear characteristic. Finally, Artificial Neural Network (ANN) was employed to develop mathematical model for wear prediction using Decision Prediction Tool 5.7.1. The accuracy of the developed model is 83%.
i 5
TABLE OF CONTENTS CHAPTER
TITLE
NO
1
2
3
4
5
6
PAGE NO
ABSTRACT
i
LIST OF TABLE
ii
LIST OF FIGURE
iii
LIST OF SYMBOL
iv
INTRODUCTION
1
1.1 Need of metal matrix composite composit e
1
1.2 Application of metal matrix composite
1
1.3 Materials and structure
2
LITERATURE REVIEW
4
2.1. Objective of the Project
6
COMPOSITE MATERIALS
7
3.1 Classification of composite materials
8
METAL MATRIX COMPOSITE PROCESSING METHODS
13
4.1 Liquid state fabrication techniques
13
4.2 Solid state fabrication techniques
15
4.3 Vapour deposition
18
PROPERTIES OF COMPOSITE MATERIALS
19
5.1 Various mechanical properties
19
5.2 Physical properties
21
5.3 Effect of various parameter
23
5.4 Wear characteristics
25
ARTIFICIAL NEURAL NETWORK
31
6.1 Background
31
6.2 Mathematical model
32
6.3 Activation functions
33
6.4 Network functions
35
6.5 Learning
36
6.6. Supervised learning
36
6.7 Unsupervised learning
37
6.8 Applications
37
6.9 Types of neural networks
38
6
LIST OF EQUIPMENTS USED
39
7.1 Stir casting setup
39
7.2 Hardness testing machine
40
7.3 Center lathe
41
7.4 Pin on disc Tribometer
42
7.5 scanning electron microscope
43
EXPERIMENTAL METHODOLOGY
44
8.1 Material selection
45
8.2 Dispersion phase – red mud
45
RAW MATERIAL PRE PROCESSING
48
9.1 Processing Processing of of raw material material
48
9.2 Combination of raw materials
48
FABRICATION OF METAL MATRIX COMPOSITE
50
10.1 Preparation of mould
50
10.2 Melting and casting of specimen
51
10.3 Dispersion processing
52
10.4 Hardness test
53
10.5 Density of cast specimen
54
10.6 Wear test
54
10.7 composite fabricated by stir casting method on micro structure
55
11
RESULT AND DISCUSSION
56
12
REFERENCE
63
13
ANNEXURE
65
14
CONCLUSION
66
7
8
9
10
7
LIST OF TABLES Table
Page
No
No
5.1
Symptoms Sympto ms of wear
30
8.1
Chemical composition composit ion of Al LM25
45
8.2
Mechanical Mechanical properties of Aluminium LM 25
45
8.3
Chemical composition composit ion of Red mud
47
9.1
Combination of raw materials
49
11.1
Hardness values of casted specimens
59
11.2
Density values of casted specimens
59
11.3
Wear test on casted specimens specimens
59
Ii
8
LIST OF FIGURES Figure
Page
No
No
1.1
Types of composite composit e materials
2
1.2
Structures of composites composites
3
3.1
Classification of composite materials with metal matrix
11
3.2
Types of Particles
11
4.1
Squeeze casting infiltrat infiltration ion
14
4.2
Stir casting setup
15
5.1
Archimedes principle
22
6.1
Mathematical process of artificial neural network
33
6.2
Style of Neural computation
35
7.1
Stirrer setup
39
7.2
Brinell hardness testing machine
40
7.3
Impact testing machine machine
41
7.4
Centre lathe
42
7.5
Pin on disc Tribometer
43
8.1
Flow chart of experimental procedure
44
8.2
XRD Image of Red Mud
47
9.1
Stir casting
49
10.1
Green sand mould
51
10.3
Sample specimen for hardness hardness test
53
10.7
SEM image of Specimen
56
11.1
Graph showing change in hardness
58
11.2
Graph showing change in density
58
iii
9
LIST OF SYMBOLS Symbol
Explanation
Σ(sigma)
Summation
Φ(phi)
Activation function
ν(nu) ν(nu)
Threshold Threshold function
€(euro) €(euro)
Computation
iv 10
CHAPTER 1 1. INTRODUCTION Material is anything made of matter, constituted of one or more substances. Wood, cement, hydrogen, air and water are all examples of materials. Sometimes the term "material" is used more narrowly to refer to substances or components with certain physical properties that are used as inputs to production or manufacturing. In this sense, materials are the parts required – matrix to make something else, from buildings and art to stars and computers.Metal computers.Metal – matrix composites (MMCs) are known as the most useful and high-tech composite materials in our world. MMCs are very important because of their high ratio of strength and weight, high Young modulus and high abrasive properties.
1.1 NEED OF METAL MATRIX COMPOSITE The need for composite materials has become a necessity for modern technology, due to the improved physical and mechanical properties. Metal matrix composites (MMC) have been developed in recent years. Metal Matrix Composites have emerged as a class of material capable of advanced structural, aerospace, automotive, electronic, thermal management and wear applications. A composite material is a material consisting of two or more physically and/or chemically distinct phases. The composite generally has superior characteristics than those of each of the t he individual components. components. Usually, the reinforcing component is distributed in the continuous or matrix component. When the matrix is a metal, the composite is termed a metal-matrix composite (MMC).
1.2APPLICATION 1.2 APPLICATION OF METAL MATRIX COMPOSITE
Metal composite materials have found application in many areas of daily life for quite some time Metal Matrix Composites (MMC‘s) are increasingly found in the automotive industry These materials are produced in situ from the conventional production and processing of metals. Metal matrix composites (MMCs) are a class of materials with the ability to blend the properties of ceramics (high strength and high modulus) with those of metals or alloys (ductility and toughness) to produce significant improvements in the mechanical properties of the composite over those of the monolithic metal or alloys. Among the various matrix materials available, aluminium and its alloys are widely used in the fabrication of MMCs. 11
Aluminium MMCs exhibit an excellent combination of high specific strength and stiffness, high wear resistance, good seizure resistance, improved fatigue and creep strength as compared to the base alloy. The MMC finds various applications in different fields as follows
MMCs represent the Next Generation of solutions for today‘s electronic requirements
Prototyping for the Space Shuttle,
Commercial airliners
Bicycles & automobiles
Electronic Electro nic substrates substr ates and Aerospace
Comparable construction unit characteristics are attainable only with the application of powder metallurgical aluminium alloys or when using heavy iron pistons. The reason for the application of composite materials is, as already described the improved high temperature properties. Potential applications are in the area of undercarriages, e.g. transverse control arms and particle-strengthened brake disks, which can be also applied in the area of rail mounted vehicles, vehicles, e.g. for undergrounds and railway (ICE).
1.3 MATERIALS AND STRUCTURE
Fig 1.1 Types of Co mposite mposite Materials
12
Fig 1.2 Structures Str uctures of composites
The difference between a material and a structure is not clearly defined. Many draw the lines between what you understand as a homogeneous material when you see it with your bare eyes, and the inhomogeneous material structure that you clearly see is made up of a fixed geometry or mixing of materials. For instance an alloy is by this definition a material even though it consists of two or more components, components, but a honeycomb honeycomb core built up of two different components is a structure. Materials are often classified into the six broad classes that are shown in figure 1.1; metals, ceramics, glasses, elastomers, polymers and composites. But, when we also include material structures, the number is bigger, and the classification of the term ‖materials and structures‖, ev even en though it is not a conventional way of making a classification, may look like the one purposed in figure 1.2.
13
CHAPTER 2 2.LITERATURE REVIEW
In recent scenario, there is an increasing trend towards using composite materials in order to achieve better performance in engineering materials. Thus, production and application of metal matrix composites (MMC) have increased in recent years (Kok M, 2005). MMCs are very important because of their high strength to weight ratio, high Young modulus and wear resistance properties, when compared to most conventional materials (Degischer HP, 1997). It also possesses high thermal conductivity and corrosion resistance properties (Hossein Abdizadeh et al 2011). Therefore, MMC are being used for wide variety of application such as, connecting rod, automobile drive shafts, cylinder liner, cylinder block, rotors (Rasit Koker et al 2005), heat sink (Dobrzanski LA et al 2006, Torralba et al, 2003) , crane bearing and motor blocks (Liu HN et al 1999, Moustafa SF et al 1997).
To fabricate MMC, metallic matrix materials should be embedded with reinforced materials, which is in the form of continuous fibers, short fibers, chopped fibers, whiskers or particulates (Yusuf Sahin, 2010). In general, aluminium, copper, nickel, silver, zinc, magnesium will be used as matrix material and Silicon carbide (Yusuf Sahin, 2010), Alumina oxide (Sun Zhiqiang et al 2005, Arslan G et al 2001, Turan S et al 2001), boron nitride, graphite, beryllium oxides, graphite (F. Akhlaghi et al 2009) silicon nitride (Sun Zhiqiang et al 2005, Fujii Hidetoshi et al 1993, Jenfin Lin et al 2001) will be used as reinforcement material. Among the different matrix material aluminium and its alloys are promising materials in automobile and aerospace application owing to their excellent mechanical properties (F. Akhlaghi et al 2009), and other major advantage is better corrosion resistance properties, moreover aluminium cheaper than other light matrix materials. However, low wear resistance of pure aluminium is a serious drawback in using many applications. Addition of ceramic reinforcement materials in the aluminium matrix material would improve the strength, hardness, wear resistance and corrosion resistance of the matrix materials. mater ials. (Mehdi Rehimian et al 2011, Toress B et al 2002, Sahin Y 1996). In order to embedded the reinforcement material with matrix material, there are different manufacturing methods are available.
14
The manufacturing methods are classified as: liquid state processing, solid state processing and deposition techniques. The liquid state processing includes, stir casting, semi solid processing (Bekir Sadik Unlu 2008), spray casting, infiltration and in situ processing. Solid state processing includes powder metallurgy (Bekir Sadik Unlu 2008,Liu B et al 1994), diffusion bonding, pultrusion, and attriter milling. Deposition technique includes chemical vapour deposition, and physical vapour deposition. In case of liquid state processing methods, the ceramic material are added in to t o the melt and stirred, st irred, because of stirring ceramic material with matrix material has some advantage ie better matrix and ceramic material bonding and easier control of micro structure. (Bekir Sadik Unlu 2008, Bermudez MD et al 2001,Mazeen AA et al 1992,Sahin Y et al 2003, Gahr KH et al 2000). However liquid state processing methods has two major drawbacks, firstly the ceramic particles are not wetted with matrix material and secondly, the particles tend to sink or float depending on their density relative to the liquid metal and so that dispersion of the particles are not uniform, where as solid state process, especially powder metallurgy process offers uniform distribution of reinforcement with matrix mater ial.( ial.( Bekir Sadik Unlu 2008, Bai M et al 1995, Nesarikar AR et al 1991, Soliman FA et al 1997).
15
2.1 OBJECTIVE OF THE PROJECT From the literature review it was found there is wide opportunity to reduce the research gap in the field of metal matrix composites. Therefore, in this work attempt has been to fulfil the research gaps. The objective of the proposed work is 1.
To fabricate aluminium based red mud metal matrix composite with different composition of red mud using stir casting process.
2. To investigate various mechanical properties of specimen material by conducting various tests like hardness, density, and wear resistance. 3. To study the effect of red mud reinforcement on mechanical properties. 4. To compare the mechanical properties of various compositions of red mud material. 5. Based on the outcome of mechanical studies, mathematical modeling for wear will be formulated and suitable decision making model using intelligent techniques techniques like Artificial Neural Network will be developed.
16
CHAPTER 3 3. COMPOSITE MATERIALS Composite materials, often shortened to composites or called composition materials, are engineered or naturally occurring materials made from two or more constituent constituent materials with significantly different physical or chemical properties which remain separate and distinct at the macroscopic or microscopic scale within the finished structure. Composite material is a material composed of two or more distinct phases (matrix phase and dispersed phase) and having bulk properties significantly different from those of any of the constituents. Composite materials are materials that combine two or more materials (a selected filler or reinforcing elements and compatible matrix binder) that have quite different properties that when combined offer properties which are more desirable than the properties of the individual materials. The different materials work together to give the composite unique properties, but within the composite you can easily see the different materials; they do not dissolve or blend into each other.
The key characteristics of composites is the • Specific strength (the strength to weight ratio) • Specific stiffness or specific modulus (the (the stiffness-to-weight ratio) • Tailored material ( since composites are composed of 2 or more‖phases‖,
They can be formulated to meet the needs of a specific application with considerable ease) Composites are not a single material but a family of materials whose stiffness, strength, density, and thermal and electrical properties can be tailored. The matrix, the reinforcement material, the volume and shape of the reinforcement, the location of the reinforcement, and the fabrication method etc. can all be varied to achieve required properties
17
3.1 CLASSIFICATION OF COMPOSITE MATERIALS
o
MATRIX PHASE
o
DISPERSED PHASE
MATRIX PHASE
The matrix is the monolithic material into which the reinforcement is embedded, and is completely continuous. This means that there is a path through the matrix to any point in the material, unlike two materials sandwiched together. In structural applications, the matrix is usually a lighter metal such as aluminium, magnesium, or titanium, and provides a compliant support for the reinforcement. In high temperature applications, cobalt and cobalt-nickel alloy matrices are common. .
DISPERSED (REINFORCING) PHASE
The second phase (or phases) is embedded in the matrix in a discontinuous form. This secondary phase is called dispersed phase. Dispersed phase is usually stronger than the matrix, therefore it is sometimes called reinforcing phase. Many of common materials (metal alloys, doped Ceramics and Polymers mixed with additives) also have a small amount of dispersed phases in their structures, however they are not considered as composite materials since their properties are similar to those of their base constituents (physical properties of steel of steel are similar to those t hose of pure iron).
3.1.1CLASSIFICATION BASED ON MATRIX PHASE There are two classification systems of composite materials. One of them is based on the matrix material (metal, ceramic, and polymer) and the second is based on the material structure:
POLYMER MATRIX COMPOSITE
Polymer Matrix Composite a polymer (resin) matrix combined
(PMC) is the with
a fibrous
material reinforcing
consisting dispersed
of phase.
Polymer Matrix Composites are very popular due to their low cost and simple fabrication methods. Use of non-reinforced polymers as structure materials is limited by low level of their mechanical properties: tensile strength of one of the strongest polymers - epoxy resin is 20000 psi (140 MPa). In addition to relatively low strength, polymer materials possess low impact resistance.
18
CERAMIC MATRIX COMPOSITE
Ceramic matrix composites (CMCs) are a subgroup of composite materials as well as a subgroup of technical ceramics. They consist of ceramic fibers embedded in a ceramic matrix, thus forming a ceramic fiber reinforced ceramic (CFRC) material. The matrix and fibers can consist of any ceramic material, whereby carbon and carbon fibers can also be considered a ceramic material. Ceramic Matrix Composite (CMC) is a material consisting of a ceramic matrix combined with a ceramic (oxides, carbides) dispersed phase. Ceramic Matrix Composites are designed to improve toughness of conventional ceramics, the main disadvantage of which is brittleness. Ceramic Matrix Composites are reinforced by either continuous (long) fibers or discontinuous (short) fibers. Generally, CMC names include a combination of type of fiber/type of matrix. For example, C/C stands for carbonfiber-reinforced carbon (carbon/carbon) ( carbon/carbon),, or C/SiC for carbon-fiber-reinforced silicon carbide. Sometimes the manufacturing process is included, and a C/S iC composite manufactured with the liquid polymer infiltration (LPI) process (see below) is abbreviated as LPI-C/SiC.
METAL MATRIX COMPOSITE
Metal Matrix Composites (MMC‘s) are increasingly found in the automotive industry. They consist of a metal such as aluminium as the matrix, and a reinforcement that could be continuous fibres such as silicon carbide, graphite or alumina, wires such as tungsten, beryllium, titanium and molybdenum, and discontinuous materials. Metals containing ceramic particles, whiskers or (short or long) fibers are also gaining importance. MMC are said to be materials for the demands of the future. When the demands for high thermal conductivity, conductivity, reduced weight, heat dissipation, dissipation, and high strength are factors for design.
19
3.1.2 BASED ON REINFORCEMENT
SILICON CARBIDE
Silicon carbide matrix composi co mposites tes are fabricated by chemical vapor infiltration or liquid phase Infiltration methods of a matrix material into a preform prepared from silicon carbide fibers. Silicon carbide matrix composites are used for manufacturing combustion liners of gas turbine engines, hot gas re-circulating fans, heat exchangers, rocket propulsion components, filters for hot liquids, gas-fired burner parts, furnace furnace pipe hangers, immersion burner tubes.
ALUMINA AND ALUMINA-SILICA
Alumina
and
alumina-silica
(mullite)
matrix
composites are
produced
by sol-gel
method, direct metal oxidation or chemical bonding. Alumina and alumina-silica (mullite) matrix composites are used for manufacturing heat exchangers, filters for hot liquids, thermophotovoltaic burners, burner stabilizers, combustion liners of gas turbine engines.
CARBON-CARBON
Carbon-Carbon Composites are fabricated by chemical vapour infiltration or infiltration methods of a matrix material into a preform prepared from carbon fibers. Carbon-Carbon Composites are used for manufacturing high performance braking systems, refractory components, components, hot-pressed dies, heating elements, turbojet engine compon co mponents. ents.
3.1.2.1 BASED OF REINFORCEMENT GEOMETRY
PARTICULATE REINFORCED REINFORCED MMC MM C (PRM):
Metal matrix composite with a particulate reinforcement reinforcement occupying occupying a volume vo lume fraction greater than 5% in the t he material (otherwise, the particulates are generally considered to be inclusions). inclusions).
DISPERSOID REINFORCED MMC:
Metal matrix composite with a dispersed reinforcement occupying a volume fraction greater than 5% in the material (otherwise, the material is considered to be a dispersion strengthened metal - which incidentally may form the matrix of any type of MMC, i.e., a MMC with dispersion-strengthened dispersion-strengthened matrix). 20
CERMETS:
A metal matrix with a three-dimensionally percolating ceramic reinforcement, typically with far more ceramic than metal (strictly speaking they contain less than 20% metal per volume and are are thus not not considered as MMC).
According to the percolating percolating structure of both
constituents cermets could thus be considered as both a ceramic and a metal matrix composite.
Fig 3.1 Classification of Composite materials with metal matrix
Fig 3.2 Types of particles
21
Particulates are the most common and cheapest reinforcement materials. These produce the isotropic property of MMCs, which shows a promising application in structural and in automobile fields. Reinforcement materials are metallic, non- metallic and ceramic materials such as;
Red mud,
Silicon carbide (SiC),
Aluminium oxide (Al2O3),
Boron nitride,
Tungsten carbide,
Titanium diboride,
Titanium carbide,
B4C,
Silicon nitrate.
22
CHAPTER 4 4. METAL MATRIX COMPOSITE PROCESSING METHODS The Metal matrix composites are fabricated by various techniques which are generally classified in to two major methods such as * Liquid State Fabrication Fabrication * Solid state fabrication fabrication Here we have focused on the Liquid state fabrication techniques
4.1 LIQUID STATE FABRICATION TECHNIQUES The methods of liquid state fabrication of Metal matrix composite
Infiltration
Gas Pressure Infiltration
Squeeze Casting Infiltration
Pressure Die Infiltration
Stir Casting
4.1.1 INFILTRATION Infiltration is a liquid state method of composite materials fabrication, in which a preformed dispersed phase (ceramic particles, fibers, woven) is soaked in a molten matrix metal, which fills the space between the dispersed phase inclusions. The motive force of an infiltration process may be either capillary force of the dispersed phase (spontaneous infiltration) or an external pressure (gaseous, mechanical, electromagnetic, centrifugal or ultrasonic) applied to the liquid matrix phase (forced infiltration).
4.1.2 GAS PRESSURE INFILTRATION Gas Pressure Infiltration is a forced infiltration method of liquid phase fabrication of Metal matrix composites, using a pressurized gas for applying pressure on the molten metal and forcing it to penetrate into a preformed dispersed phase. Gas Pressure Infiltration method is used for manufacturing large composite parts. The method allows using non-coated fibers due to short contact time of the fibers with the hot metal. In contrast to the methods using mechanical force, Gas Pressure Infiltration results in low damage of the fibers.
23
4.1.3 SQUEEZE CASTING INFILTRATION Squeeze Casting Infiltrat Infi ltration ion is a forced infiltration infiltr ation method of liquid phase fabricat ion of Metal matrix composites, using a movable mould part (ram) for applying pressure on the molten metal and forcing it to penetrate into a performed dispersed phase, placed into the lower fixed mold part. Squeeze Casting Infiltration method is similar to the Squeeze casting technique used for metal alloys casting.
Fig 4.1 Squeeze Casting Infiltration
PRESSURE DIE INFILTRATION
Pressure Die Infiltration is a forced infiltration method of liquid phase fabrication Metal matrix, using a Die casting technology, when a preformed dispersed phase (particles, fibers) is placed into a die (mould) which is then filled with a molten metal entering the die through a sprue and penetrating into the preform under the pressure of a movable piston (plunger).
4.1.4 STIR CASTING Stir Casting is a liquid state method of composite materials fabrication, in which a dispersed phase (ceramic particles, short fibers) is mixed with a molten metal matrix by means of mechanical stirring. Stir Casting is the simplest and the most cost effective effective method of liquid state stat e fabrication. fabrication.
24
Stir Casting is the simplest and the most cost effective method of liquid state fabrication. The liquid composite material is then cast by conventional casting methods and may also be processed by conventional metal forming technologies t echnologies.. Stir Casting is characterized by the following features:
Content of dispersed phase is limited (usually not more than 30 vol. %).
Distribution of dispersed phase throughout the matrix matr ix is not perfectly homogeneous:
1. There are local clouds (clusters) of the dispersed particles (fibers); 2. There may be gravity segregation of the dispersed phase due to a difference in the densities of the dispersed and matrix phase.
The technology is relatively simple and low cost.
Fig 4.2 Stir casting setup
4.2 SOLID STATE FABRICATION TECHNIQUES 4.2.1 POWDER METALLURGY: The majority of the structural components produced by fixed die pressing are iron based. The powders are elemental, pre-alloyed, or partially alloyed. Elemental powders, such as iron and copper, are easy to compress to relatively high densities, produce pressed compacts with adequate strength for handling during sintering, but do not produce very high strength sintered parts. Pre-alloyed powders are harder, less compressible and hence require higher pressing loads to produce high density compacts. However, they are capable of producing high strength sintered materials. Pre-alloying is also used when the production of a homogeneous material from elemental powders requires very high temperatures and long sintering times. The best examples are the stainless steels, whose chromium and nickel contents have to be pre-alloyed to allow economic production by powder metallurgy. As a general rule both mechanical and physical properties improve with increasing density. 25
Therefore the method selected for the fabrication of a component by powder metallurgy will depend on the level of performance required from the part. Many components are adequate when produced at 85-90% of theoretical full density (T.D.) whilst others require full density for satisfactory performance. Sintering is t he process whereby powder compacts compacts are heated so that adjacent particles fuse together, thus resulting in a solid article with improved mechanical strength compared compared to the powder p owder compact. This ―fusing‖ of particles results in an increase in the density of the part and hence the process is sometimes called densification. There are some processes such as hot isostatic pressing which combine the compaction and sintering processes into a single step. The density of the component can also change during sintering, depending on the materials and the sintering temperature. These dimensional changes can be controlled by an understanding and control of the pressing and sintering parameters, and components can be produced with dimensions that need little or no rectification to meet the dimensional tolerances. Note that in many cases all of the powder used is present in the finished product, scrap losses will only occur when secondary machining operations are necessary.
4.2.1 HOT ISOSTATIC PRESSING: Powders are usually encapsulated in a metallic container but sometimes in glass. The container is evacuated; the powder out-gassed to avoid contamination of the materials by any residual gas during the consolidation stage and sealed-off is shown in fig. 4.3. It is then heated and subjected to isostatic pressure sufficient to plastically deform both the container and the powder. The rate of densification of the powder depends upon the yield strength of the powder at the temperatures and pressures chosen. At moderate temperature the yield strength of the powder can still be high and require high pressure to produce densification in an economic time. Typical values might be 1120°C and 100 MPa for ferrous alloys. By pressing at very much higher temperatures lower pressures are required as the yield strength of the material is lower. Using a glass enclosure atmospheric pressure (15 psi) is used to consolidate bars and larger billets. The technique requires considerable financial investment investment as the pressure vessel has to to withstand the internal gas pressure and allow t he powder to be heated to high temperatures. temperatures.
26
As with cold isostatic pressing only semi finished products are produced, either for subsequent subsequent working to smaller smaller sizes, or o r for machining to finished finished dimensions.
4.2.3 HOT FORGING (POWDER FORGING): Cold pressed and sintered components components have the great advantage advantage of being b eing close to final shape (near-nett shape), but are not fully dense. Where densification is essential to provide adequate mechanical properties, the technique of hot forging, or powder forging, can be used. In powder forging an as-pressed component is usually heated to a forging temperature significantly below the usual sintering temperature of the material and then forged in a closed die. This produces a fully dense component with the shape of the forging die and appropriate mechanical properties. Powder forged parts generally are not as close to final size or shape as cold pressed and sintered parts. This result from the allowances made for thermal expansion effects and the need for draft angles on the forging tools. Further, minimal, machining is required but when all things are considered this route is often very cost effective.
4.2.4 METAL INJECTION MOULDING (MIM): Injection moulding is very widely used to produce precisely shaped plastic components in complex dies. As injection pressures are low it is possible to manufacture complex components, even some with internal screw threads, by the use of side cores and split tools. By mixing fine, typically less than 20 mm diameter, spherical metal powders with thermoplastic binders, metal filled plastic components can be produced with many of the features available in injection moulded plastics. After injection moulding, the plastic binder material is removed to leave a metal skeleton which is then sintered at high temperature. Dimensional control can be exercised on the as-sintered component as the injected density is sensibly uniform so shrinkage on sintering is also uniform. Shrinkage can be large, due to both the fine particle size of the powders and the substantial proportion of polymer binder used.
27
4.2.5 DIFFUSION BONDING Diffusion Bonding is a solid state fabrication method, in which a matrix in form of foils and a dispersed phase in form of long fibres are stacked in a particular order and then pressed at elevated temperature The finished laminate composite material has a multilayer structure. Diffusion Bonding is used for fabrication of simple shape parts (plates, tubes). In the Foil diffusion bonding Layers of metal foil are sandwiched with long fibres, and then pressed through to form a matrix.
4.3 VAPOUR DEPOSITION:
There are two methods of solid state fabrication of Metal matrix composite; composite;
4.3.1 PHYSICAL VAPOUR DEPOSITION: The fibre is passed through a thick cloud of vaporized metal, coating it. In situ fabrication technique. technique. Controlled unidirectional solidification of a eutectic alloy can result in a two -phase microstructure with one of the phases, present in lamellar or fiber form, distributed in the matrix.
4.3.1 ELECTROPLATING / ELECTROFORMING: ELECTROFORMING: A solution containing metal ions loaded with reinforcing particles is co-deposited forming a composite material.
28
CHAPTER 5 5. PROPERTIES OF MATERIALS Material property may be a constant or may be a function of one or more independent variables, such as temperature. Material's properties often vary to some degree according to the direction in the material in which they are measured; a condition referred to as anisotropy. Materials properties that relate two different physical phenomena often behave linearly (or approximately so) in a given operating range, and may then be modeled as a constant for that range. This linearization can significantly simplify the differential constitutive equations that the property describes.
5.1 VARIOUS MECHANICAL PROPERTIES
Compressive strength Density Ductility Fatigue limit Flexural modulus Flexural strength Fracture toughn to ughness ess Hardness Plasticity (physics) Poisson's ratio
Shear modulus Shear strain Shear strength Softness Specific modulus Specific weight Tensile strength Yield strength Young's modulus
TENSILE STRESSES:
Tension (or tensile) stresses develop when a material is subject to a pulling load; for example, when using a wire rope to lift a load or when using it as a guy to anchor an antenna. "Tensile strength" is defined as resistance to longitudinal stress or pulls and can be measured in pounds per square inch of cross section. Shearing stresses occur within a material when external forces are applied along parallel lines in opposite directions. Shearing forces can separate material by sliding part of it in one direction and the rest in the opposite direction. A material that is stressed repeatedly usually fails at a point considerably below its maximum strength in tension, compression, or shear. For example, a thin steel rod can be broken by hand by bending it back and forth several times in the same place; however, if the same force is applied in a steady motion (not bent back and forth), the rod cannot be broken.
29
The tendency of a material to fail after repeated bending at the same point is known as fatigue. HARDNESS
Hardness is the measure of how resistant solid matter is to various kinds of permanent shape change when a force is applied. Macroscopic hardness is generally characterized by strong intermolecular bonds, bonds, but the behaviour of solid materials under force is complex; therefore there are different measurements of hardness: scratch hardness, indentation hardness, and rebound hardness. Hardness is dependent on ductility, elasticity, plasticity, strain, strength, toughness, viscoelasticity, viscoelasticity, and viscosity.
IMPACT STRENGTH
The Charpy impact test, also known as the Charpy v-notch test, is a standardized high strainrate test which determines the amount of energy of energy absorbed by a material during fracture. This absorbed energy is a measure of a given material's toughness and acts as a tool to study temperature-dependent brittle-ductile transition. It is widely applied in industry, since it is easy to prepare and conduct and results can be obtained quickly and cheaply. But a major disadvantage is that all results are only comparative. TOUGHNESS:
Toughness is the property that enables a material to withstand shock and to be deformed without rupturing. Toughness may be considered as a combination of strength and plasticity. ELASTICITY:
When a material has a load applied to it, the load causes the material to deform. Elasticity is the ability of a material to return to its original shape after the load is removed. Theoretically, the elastic limit of a material is the limit to which a material can be loaded and still recover its original shape after the load is removed.
30
PLASTICITY:
Plasticity is the ability of a material to deform permanently without breaking or rupturing. This property is the opposite of strength. By careful alloying of metals, the combination of plasticity and strength is used to manufacture large structural members. For example, should a member of a bridge structure become overloaded, plasticity allows the overloaded member to flow allowing the distribution of the load to other parts of the bridge structure. BRITTLENESS:
Brittleness is the opposite of the property of plasticity. A brittle metal is one that breaks or shatters before it deforms. White cast iron and glass are good examples of brittle material. Generally, brittle metals are high in compressive strength but low in tensile strength. As an example, you would not choose cast iron for fabricating support beams in a bridge. DUCTILITY AND MALLEABILITY:
Ductility is the property that enables a material to stretch, bend, or twist without cracking or breaking. This property makes it possible for a material to be drawn out into a thin wire. In comparison, malleability is the property that enables a material to deform by compressive forces without developing defects. A malleable material is one that can be stamped, hammered, forged, pressed, or rolled into t hin sheets.
5.2 PHYSICAL PROPERTIY Metals in general have high electrical conductivity which depends on their valency of ions, thermal conductivity, lustre and density, and the ability to be deformed under stress without cleaving. ELECTRICAL CONDUCTIVITY:
Electrical conductivity or specific conductance is the reciprocal quantity, and measures a material's ability to conduct conduct an electric current.
31
THERMAL CONDUCTIVITY:
Thermal conductivity, k, is the property of a material's ability to conduct heat. It appears primarily in Fourier's Law for heat conduction. Heat transfer across materials of high thermal conductivity occurs at a faster rate than across materials of low thermal conductivity. Correspondingly materials of high thermal conductivity are widely used in heat sink applications and materials of low thermal conductivity are used as thermal insulation. Thermal conductivity of materials is temperature dependent. In general, materials become more conductive to heat as the average temperature increases. The reciprocal of thermal conductivity is thermal t hermal resistivity. DENSITY
The density of a material is given by the ratio of mass per unit volume. The density of MMC varies due to infiltration of Red mud in to the Aluminium alloy. The density of a material can be found using Archimedes principle which states that the weight of the displaced fluid is directly proportional to the volume of the displaced fluid (if the surrounding fluid is of uniform density). In simple terms, the principle states that the buoyant force on an object is going to be equal to the weight of t he fluid displaced by the object, or the density of the fluid multiplied by the submerged volume times the t he gravitational constant, g. Thus, among completely submerged objects with equal masses, objects with greater volume have greater buoyancy.
32
Fig 5.1 Archimedes principle
5.3 EFFECT OF VARIOUS PARAMETERS
EFFECT OF REINFORCEMENT VOLUME FRACTION It was predicted that there exists a critical reinforcement volume fraction above which the composite strength can be improved relative to that of the unreinforced material and below which the composite strength decreases, owing to the ineffective load transfer from matrix to reinforcement in MMCs. For low volume fraction of reinforcement, the composite strength was observed to be governed by the residual matrix strength, which decreases with increasing reinforcing volume fraction.
EFFECT OF PARTICLE SIZE The deformation and fracture behaviour of the composite revealed the importance of particle size. A reduction in particle size is observed to increase the proportional limit, yield stress and the ultimate tensile stress. It is well established that large particles are detrimental to fracture toughness due to their tendency towards fracture. It would be highly desirable to have a composite system where the reinforcing particles are relatively fine so as to get the stiffness benefits of a composi co mposite te without w ithout significantly lowering fracture toughness.
EFFECT OF REINFORCEMENT DISTRIBUTION Apart from the reinforcement level, the reinforcement distribution also influences the ductility and fracture toughness of the MMC and hence indirectly the strength. A uniform reinforcement reinforcement distribution distr ibution is essential for effective utilization of the t he load carrying capacity capacity of
33
the reinforcement. Non-uniform distributions of reinforcement in the early stages of processing was observed to persist to the final product in the forms of streaks or clusters of uninfiltrated reinforcement with their attendant porosity, all of which lowered ductility, strength and toughness of the material.
FRACTURE
The fracture behaviour of MMCs has been identified not only for extending their applications but also for improving mechanical properties, especially strength and ductility. Better understandings of the underlying mechanisms affecting composite properties are essential if the properties of the composite material are to be improved. Tensile fracture of conventional alloys is considered in terms of the micro void coalescence model (MVC). Void nucleation in unreinforced alloys occurs at constituent particles, either through particle failure, through interface decohesion. Decohesion is most common, but particle cracking occurs with elongated particles. In composites, there are three possible mechanisms for void nucleation particle part icle cracking, interface decohesion, and matr matrix ix void void nucleation is the same mechanism as occurs in the unreinforced alloys
MICROSTRUCTURE
The most important aspects of the microstructure is the distribution of the reinforcing particles, and this depends on the processing and fabrication routes involved. The oxides of reinforcing particles used in the composites have a varying density. Density of the particles is one of the important factors determining the distribution of the particles in molten metal. Particles having higher density than molten metal can settle at the bottom of the bath slowly and particles of lower density can segregate at the top. During subsequent pouring of the composite melt, the particle content may vary from one casting to another or even it can vary in the same casting from one region to another. Therefore uniform distribution of the particles in the melt is a necessary condition for uniform distribution of particles in the castings. The properties of composites are finally dependent on the distribution of the particles. Hence the study of the distribution of the particles in the composite is of great significance. Several investigators have examined examined the t he fracture samples of different metal matrix composites; it was observed that the fracture occurred mainly through the matrix in a ductile manner.
34
5.4 WEAR CHARECTERISTICS Wear can also be defined as a process where interaction between two surfaces or bounding faces of solids within the working environment results in dimensional loss of one solid, with or without any actual decoupling and loss of material. Aspects of the working environment which affect wear include loads and features such as unidirectional sliding, reciprocating, rolling, and impact loads, speed, temperature..Wear temperature..Wear is related to interactions between surfaces and more specifically the removal and deformation of material on a surface as a result of mechanical action of the opposite surface. The need for relative motion between two surfaces and initial mechanical contact between asperities is an important distinction between mechanical wear compared to other processes with similar outcomes.
5.4.1 TYPES OF WEAR The study of the processes of wear is part of the discipline of tribology. The complex nature of wear has delayed its investigations and resulted in isolated studies towards specific wear mechanisms or processes processes..[6] Some commonly referred to wear mechanisms (or processes) include: 1. Adhesive Wear
4. Fretting Wear
2. Abrasive Wear
5. Erosive Wear
3. Surface Fatigue
A number of different wear phenomena are also commonly encountered and presented in the literature. Impact-, cavitations-, diffusive- and corrosive- wear are all such examples. These wear mechanisms, however, do not necessarily act independently and wear mechanisms are not mutually exclusive. "Industrial Wear" are comm co mmonly only described as incidence of multiple wear mechanisms occurring in unison. Another way to describe "Industrial Wear" is to define clear distinctions in how different frict ion mechanisms mechanisms operate, for example distinguish between mechanisms with high or low energy density. Wear mechanisms and/or sub-mechanisms frequently overlap and occur in a synergistic manner, producing a greater rate of wear than the sum of the individual wear mechanisms.
35
ADHESIVE WEAR Adhesive wear can be found between surfaces during frictional contact and generally refers to unwanted displacement and attachment of wear debris and material compounds from one surface to another. Two separate mechanisms operate between the surfaces. 1. Adhesive wear are caused by relative motion, "direct contact" and plastic deformation which create wear debris and material transfer from one surface to another. 2. Cohesive adhesive forces, holds two surfaces together even though they are separated by a measurable distance, with or without any actual transfer of material. The above description and distinction between "adhesive wear" and its counterpart "cohesive adhesive forces" are quite common. Usually cohesive surface forces and adhesive energy potentials between surfaces are examined as a special field in physic departments. The adhesive wear and material transfer due to direct contact and plastic deformation are examined in engineering science and in industrial research. Two aligned surfaces may always cause material transfer and due to overlaps and symbiotic relations between relative motional ―wear‖ and ―chemical‖ cohesive attraction. Generally, adhesive wear occurs when two bodies slide over or are pressed into each other, which promote material transfer. This can be described as plastic deformation of very small fragments within the surface layers. The asperities or microscopic high points or surface roughness found on each surface, define the severity on how fragments of oxides are pulled off and adds to the other surface, partly due to strong adhesive forces between atoms but also due to accumulation of energy in the plastic zone between the asperities during relative motion. The outcome can be a growing roughening and creation of protrusions (i.e., lumps) above the original surface, in industrial manufacturing referred to as galling , which eventually breaches the oxidized surface layer and connects to the underlying bulk material which enhance the possibility for a stronger adhesion and plastic flow around the lump. The geometry and the nominal sliding velocity of the lump defines how the flowing material will be transported and accelerated around the lump which is critical to define contact pressure and developed temperature during sliding. The mathematical function for acceleration of flowing material is thereby defined by the lumps surface contour.
36
It‘s clear, given these prerequisites, that contact pressure and developed temperature is highly dependent on the lumps geometry. Flow of material exhibits an increase in energy density, because initial phase transformation and displacement of material demand acceleration of material and high pressure. Low pressure is not compatible with plastic flow; only after deceleration may the flowing material be exposed to low pressure and quickly cooled. In other other words, you can‘t deform a solid material using direct contact without applying a high pressure and somewhere along the process must acceleration and deceleration take place, i.e., high pressure must be applied on all sides of the deformed material. Flowing material will immediately exhibit energy loss and reduced ability to flow due to phase transformation, if ejected from high pressure into low pressure. This ability withholds the high pressure and energy density in the contact zone and decreases the amount of energy or friction force needed for further advancement when the sliding continues and partly explain the difference between the static and sliding coefficient of friction (μ) if the main fracture mechanisms are equal to the previous. Adhesive wear is a common fault factor in industrial applications such as sheet metal forming (SMF) and commonly encountered in conjunction with lubricant failures and are often referred to as welding wear or galling due to the exhibited surface characteristics, phase transition and plastic flow followed by cooling. The type of mechanism and the amplitude of surface attraction vary between different materials but are amplified by an increase in the density of ―surface energy‖. Most solids will adhere on contact to some extent. However, oxidation films, lubricants and contaminants naturally occurring generally suppress adhesion and spontaneous spo ntaneous exothermic exother mic chemical reactions react ions between betw een surfaces sur faces generally produce a substance with low energy status in the absorbed species.
ABRASIVE WEAR Abrasive wear occurs when a hard rough surface slides across a softer surface. ASTM (American Society for Testing and Materials) defines it as the loss of material due to hard particles or hard protuberances that are forced against and move along a solid surface. Abrasive wear is commonly classified according to the type of contact and the contact environment. The type of contact determines the mode of abrasive wear. The two modes of abrasive wear are known as two-body and three-body abrasive wear. Two-body wear occurs when the grits or hard particles remove material from the opposite surface. The common analogy is that of material being removed or displaced by a cutting or plowing operation.
37
Three-body wear occurs when the particles are not constrained, and are free to roll and slide down a surface. The contact environment determines whether the wear is classified as open or closed. An open contact environment environment occurs when the surfaces are sufficiently displaced to be independent of one another There are a number of factors which influence abrasive wear and hence the manner of material removal. Several different mechanisms have been proposed to describe the manner in which the material is removed. Three commonly identified mechanisms of abrasive wear are: 1. Plowing 2. Cutting 3. Fragmentation Plowing occurs when material is displaced to the side, away from t he wear particles, resulting in the formation of grooves that do not involve direct material removal. The displaced material forms ridges adjacent to grooves, which may be removed by subsequent passage of abrasive particles. Cutting occurs when material is separated from the surface in the form of primary debris, or microchips, with little or no material displaced to the sides of the grooves. This mechanism closely resembles conventional machining. Fragmentation occurs when material is separated from a surface by a cutting process and the indenting abrasive causes localized fracture of the wear material. These cracks then freely propagate locally around the wear groove, resulting in additional material removal by spalling.
SURFACE FATIGUE Surface fatigue is a process by which the surface of a material is weakened by cyclic loading, which is one type of general material fatigue. Fatigue wear is produced when the wear particles are detached by cyclic crack growth of micro cracks on the surface. These micro cracks are either superficial cracks or subsurface cracks.
FRETTING WEAR Fretting wear is the repeated cyclical rubbing between two surfaces, which is known as fretting, over a period of time which will remove material from one or both surfaces in contact. It occurs typically in bearings, although most bearings have their surfaces hardened to resist the problem. Another problem occurs when cracks in either surface are created, known as fretting fatigue. It is the more serious of the two phenomena because it can lead to 38
catastrophic failure of the bearing. An associated problem occurs when the small particles removed by wear are oxidised in air. The oxides are usually harder than t han the underlying metal, so wear accelerates as the harder particles abrade the metal surfaces further. Fretting corrosion acts in the same way, especially when water is present. Unprotected bearings on large structures like bridges can suffer serious degradation in behaviour, especially when salt is used during winter to deice the highways carried by the bridges.
EROSIVE WEAR Erosive wear can be described as an extremely short sliding motion and is executed within a short time interval. Erosive wear is caused by the impact of particles of solid or liquid against the surface of an object. The impacting particles gradually remove material from the surface through repeated deformations and cutting actions. It is a widely encountered mechanism in industry. A common example is the erosive wear associated with the movement of slurries through piping and pumping equipment. The rate of erosive wear is dependent upon a number of factors. The material characteristics of the particles, such as their shape, hardness, and impact velocity and impingement angle are primary factors along with the properties of the surface being eroded. The impingement angle is one of the most important factors and is widely recognized in literature. For ductile materials the maximum wear rate is found when the impingement angle is approximately 30°, whilst for non ductile materials the maximum wear rate occurs when the impingement angle is normal to the surface.
39
5.4.2 SYMPTOMS OF WEAR A summary of the appearance and symptoms of different wear mechanism is indicated in Table 5.1 and the same is a systematic approach to diagnose the wear mechanisms.
Table 5.1 Symptoms of Wear
40
CHAPTER 6 6. ARTIFICIAL NEURAL NETWORK An artificial neural network (ANN), usually called neural network (NN), is a mathematical model or computational model that is inspired by the structure and/or functional aspects of biological neural networks. A neural network consists consists of an interconnected interconnected group of artificial neurons, and it processes information using a connectionist approach to computation. In most cases an ANN is an adaptive system that changes its structure based on external or internal information that flows through the network during the learning phase. Modern neural networks are non-linear statistical data modeling tools. They are usually used to model complex relationships between inputs and outputs or to find patterns in data.
6.1. BACKGROUND The original inspiration for the term Artificial Neural Network came from examination of central nervous systems and their neurons, axons, dendrites, and synapses, which constitute the processing elements of biological neural networks investigated by neuroscience. In an artificial neural network, simple artificial nodes, variously called "neurons", "neurodes", "processing elements" (PEs) or "units", are connected together to form a network of nodes mimicking the biological neural networks — hence the term "artificial neural network. Because neuroscience is still full of unanswered questions, and since there are many levels of abstraction and therefore many ways to take inspiration from the brain, there is no single formal definition of what an artificial neural network is. Generally, it involves a network of simple processing elements that exhibit complex global behaviour determined by connections between processing processing elements and element e lement parameters. While an artificial neural network does not have to be adaptive per se, its practical use comes with algorithms designed to alter the strength (weights) of the connections in the network to produce a desired signal flow. These networks are also similar to the biological neural networks in the sense that functions are performed collectively and in parallel by the units, rather than there being a clear delineation of subtasks to which various units are assigned. Currently, the term Art ificial Neural Network (ANN) tends to refer mostly to neural network models employed in statistics, cognitive psychology and artificial intelligence.
41
Neural network models designed with emulation of the central nervous system (CNS) in mind are a subject of theoretical neuroscience and computational neuroscience. In modern software implementations of artificial neural networks, the approach inspired by biology has been largely abandoned for a more practical approach based on statistics and signal processing. In some of these systems, neural networks or parts of neural networks (such as artificial neurons) are used as components in larger systems that combine both adaptive and non-adaptive elements. While the more general approach of such adaptive systems is more suitable for real-world problem solving, it has far less to do with the traditional artificial intelligence connectionist models. What they do have in common, however, is the principle of non-linear, distributed, parallel and local processing and adaptation.
6.2 MATHEMATICAL MODEL Neural network models in artificial intelligence are usually referred to as artificial neural networks (ANNs); these are essentially simple mathematical models defining a function f:X→Y X→Y or a distribution over X or both X and Y, but sometimes models are also intimately
associated with a particular learning algorithm or learning rule. A common use of the phrase ANN model really means the definition of a class of such functions. When creating a functional model of the biological neuron, there are three basic components of importance. First, the synapses of the neuron are modelled as weights. The strength of the connection between an input and a neuron is noted by the value of the weight. Negative weight values reflect inhibitory connections, while positive values designate excitatory connections. The next two components model the actual activity within the neuron cell. An adder sums up all the inputs modified by their respective weights. This activity is referred to as linear combination. Finally, an activation function controls the amplitude of the output of the neuron. An acceptable range of output is usually between 0 and 1, or -1 and 1. Mathematically, Mathematically, this process pro cess is described in the figure 6.1
42
Fig 6.1 Mathematical process of artificial neural network
From this model the interval activity of the neuron can be shown to be:
The output of the neuron, y k , would therefore be the outcome of some activation function on the value of v k .
43
6.3 ACTIVATION FUNCTIONS As mentioned previously, the activation function acts as a squashing function, such that the output of a neuron in a neural network is between certain values (usually 0 and 1, or -1 and 1). In general, there are three types of activation functions, denoted by Φ (.) . First, there is the Threshold Function which takes on a value of 0 if the summed input is less than a certain threshold value (v), and the value 1 if the summed input is greater than or equal to the threshold value.
Secondly, there is the Piecewise-Linear function. This function again can take on the values of 0 or 1, but can also take on values between that depending on the amplification factor in a certain region of linear operation. o peration.
Thirdly, there is the sigmoid function. This function can range between 0 and 1, but it is also sometimes useful to use the -1 to 1 range. An example of the sigmoid function is the hyperbolic tangent function.
The artificial neural networks which we describe are all variations on the parallel distributed processing (PDP) idea. The architecture of each neural network is based on very similar building blocks which perform the processing and is shown in figure 6.2
44
Fig 6.2 Style of Neural Computation Co mputation
6.4 NETWORK FUNCTION The word network in the term 'artificial neural network' refers to the inter – inter – connections connections between the neurons in the different layers of each system. The most basic system has three layers. The first layer has input neurons, which send data via synapses to the second layer of neurons, and then via more synapses to the third layer of output neurons. More complex systems will have more layers of neurons with some having increased layers of input neurons and output neurons. The synapses store parameters called "weights" that manipulate the data in the calculations. The layers network through the mathematics of the system algorithms. The network function f(x) is defined as a composition of other functions gi(x), which can further be defined as a composition of other functions. This can be conveniently represented as a network structure, with arrows depicting the dependencies between variables. A widely used type of composition is the nonlinear weighted sum, f (x) (x)=K (∑iwigi(x)), where (commonly referred to as the activation function is some predefined function), such as the hyperbolic tangent. It will be convenient for the following to refer to a collection collection of functions as simply a vector. g=(g1 ,g2 ,....,gn)
45
6.5 LEARNING What has attracted the most interest in neural networks is the possibility of learning. Given a specific task to solve, and a class of functions, learning means using a set of observations to find f*€f find f*€f which which solves the task in some optimal sense. This entails defining a cost function C: € F , (i.e., no solution has a cost f*) ≤C ( f ) ¥ f € F→R such that, for the optimal solution f*, C ( f* less than the cost of the optimal solution).The cost function is an important concept in learning, as it is a measure of how far away a particular solution is from an optimal solution to the problem to be solved. Learning algorithms search through the solution space to find a function that has the t he smallest smallest possible possible cost. There are three major learning paradigms, paradigms, each corresponding to a particular abstract learning task. These are supervised learning, unsupervised learning and reinforcement learning.
6.6 SUPERVISED LEARNING x€ X ,y€ Y and the aim is to In supervised learning, we are given a set of example pairs( x,y), x€ find a function f: x→ y in the allowed class of functions that matches the examples. In other words, we wish to infer the mapping implied by the data; the cost function is related to the mismatch between our mapping and the data and it implicitly contains prior knowledge about the problem domain. A commonly used cost is the mean-squared error, which tries to minimize the average squared error between the network's output, f(x), and the target value y over all the example pairs. When one tries to minimize this cost using gradient descent for the class of neural networks called multilayer perceptrons, one obtains the common and wellknown back propagation algorithm for training neural networks. Tasks that fall within the paradigm of supervised learning are pattern recognition (also known as classification) and regression (also known as function approximation). The supervised learning paradigm is also applicable to sequential data (e.g., for speech and gesture recognition). This can be thought o f as learning with a "teacher," in the form of a function that provides continuous feedback on the quality of solutions obtained thus far. Basically supervised learning is classified in two types. These are error connection gradient descent and stochastic. Error are also classified into least mean square and back propagation.
46
6.7 UNSUPERVISED LEARNING In unsupervised learning, some data x is given and the cost function to be minimized, that can be any function of the data x and the network's output, f .The cost function is dependent on the task (what we are trying to model) and our a priori assumptions (the implicit properties of our model, its parameters and the observed variables).As a trivial example, consider the 2 f(x)) ]. Minimizing this cost will give model(x)=a, where a is a constant and the costC=E[( x- f(x))
us a value value of a that is equal to the mean mean of the data. The cost cost function can be much more more complicated. Its form depends on the application: for example, in compression it could be related to the mutual information between x and y, whereas in statistical modeling, it could be related to the posterior probability of the model given the data. (Note that in both of those examples those quantities would be maximized rather than minimized).Tasks that fall within the paradigm of unsupervised learning are in general estimation problems; the applications include clustering, the estimation of statistical distributions, compression and filtering.
6.8 APPLICATIONS The utility of artificial neural network models lies in the fact that they can be used to infer a function from observations. This is particularly useful in applications where the complexity of the data or task makes the design of such a function by hand impractical. The tasks artificial neural networks are applied to tend to fall within the following broad categories: Function, or analysis, including time series prediction, fitness approximation and modeling. Classification, including pattern and sequence recognition, novelty detection and sequential decision making. Application areas include system identification and control (vehicle control, process control), quantum chemistry, game-playing and decision making (backgammon, chess, racing), pattern recognition (radar systems, face identification, object recognition and more), sequence recognition (gesture, speech, handwritten text recognition), medical diagnosis, financial applications (automated trading systems), data mining (or knowledge discovery in databases, "KDD"), visualization and e-mail spam fi ltering.
47
6.9 TYPES OF ARTIFICIAL NEURAL NETWORKS
An artificial neural network is a computational simulation of a biological neural network. These models mimic the real life behaviour of neurons and the electrical messages they produce between input, processing by the brain and the final output from the brain. The systems can be hardware and software based specifically built systems or purely software based and run in computer models.
o
Feed forward neural network
o
Radial basis function (RBF) network
o
Kohonen self-organizing self-organizing network
o
Learning Vector Quantization
o
Recurrent neural network
48
CHAPTER 7 7. LIST OF EQUIPMENTS USED
7.1 STIR CASTING SETUP A Self fabricated mechanical stirrer was used to perform stir casting. The equipment was fabricated by keeping the major factors as variable speed, temperature resistance, load resistance. resistance. Portabili Po rtability ty and cost.
Fig 7.1 Stirrer Setup
MAJOR COMPONENTS
DC Motor
Speed controller (VARIAC)
Stirrer blade
49
SPECIFICATION DC Motor Make
:
Crompton Greaves Commercial Manufacturer
Model
:
JLG52586
Speed
:
1400 rpm
Speed Controller – VARIAC Type
:
4AMP- 1PH
Maximum Load
:
4 AMP
Maximum kVA
:
1.08
7.2 HARDNESS TESTING MACHINE The hardness of the casted specimen was tested using Brinell Hardness Tester. The Brinell hardness is found by identifying the amount of indentation of the indenter ball in the material when a constant amount of load is applied. The result is given in HB or BHN unit.
Fig 7.2 Brinell Hardness testing machine 50
SPECIFICATIONS Make
:
SE Udyog Private Ltd.,
Model
:
B/3000/8
Maximum Load
:
3000 kgf
7.3 CENTER LATHE The sample specimens for wear test were machined to the required standard dimensi d imensions ons and surface finish was machined using a manual center lathe.
Fig 7.4 Center lathe
SPECIFICATIONS Make
:
Sona Industries
Model
:
PL4 – PL4 – Lathe Lathe
Bed length
:
4
51
1 2
‖
7.4 PIN ON DISC TRIBOMETER The
High
Temperature
Pin
on
Disc
Tribometer
designed
and
developed
by magnum engineers is primarily intended for determining the Tribological characteristics of wide range of materials under conditions of various normal loads & temperatures. A stationary pin mounted on a pin holder is brought into contact against a rotating disc at a specified speed as the pin is sliding, resulting frictional force acting between the pin and disc are measured by arresting the deflecting pin holder against a load cell. Both normal load and speed can be set as desired.
Fig 7.5 Pin on Disc Tribometer
SPECIFIACTION Make
:
Magnum engineers
Normal load range
:
up to 200N
Frictional force range
:
up to 200N with a resolution of 1N with tare facility
Wear measurement measurement range
:
0-4mm with tare facility facility
Pin length length
:
25-30mm
Wear Disc material
:
EN32 Steel
52
7.5 SCANNING ELECTRON MICROSCOPE
SPECIFICATION
Resolution
3.0nm
Magnification
X5 to 300,000
Display
20 inch, high resolution FPD
Filament
Pre-centered W hairpin filament
Accelerating voltage
0.5 to 30 kv
53
CHAPTER 8 EXPERIMENTAL METHODOLOGY
The processing of Aluminium Red mud metal matrix composite involves various step by step processes. Each and every step was planned in the aspect of safety, accuracy and cost. The steps to be carried out are shown in the flow chart below.
RAW MATERIAL PRE PROCESSING
MATRIX PHASE
•
FABRICATION COKE FEEDING
•
SPEED SETTING
MELTING OF AL
•
WEIGHING OF MATERIALS
•
REINFOCRC EMENT PHASE
•
OPTIMIZING UNSING ANN
•
DIE PREPARATI ON
•
SURFACE CLEANING
•
ADDING DEGASSIFIER
•
SLAG REMOVAL
•
STIR CASTING
•
MATERIAL SELECTION
FURNACE AND STIRRER INITIATION
MOULDING
•
Fig 8.1 Flow chart of Experimental Procedure
EXPERIMENTAL PROCEDURE
COKE FEEDING
MELTING OF ALUMINIUM
PRE HEATING OF RED MUD
SOLIDIFICATION IN AIR PROCESS
DIMENSIONING THE SPECIMEN
54
SIZING FOR HARDNE SS,IMPA CT, TEN SILE
•
SIZING OF SPECIMENS
• Wear charecteristi cs of Al RM Composite Composite is optimized using Artficial Neural Network tool in MATLab
8.1. MATERIAL SELECTION LM 25 finds its application in many areas like automobile, aerospace, food and beverage machine making industries, because of its light weight, and good mechanical properties. Though it has many advantages it is also having a major drawback that it has low wear resistance property. To overcome this problem, material like Alumina, SiC, and Tio2 can be added to increase the resistance property. The following are the chemical, mechanical and physical properties of Al LM 25
Table 8.1 CHEMICAL COMPOSITION OF Al LM25 Al
Si
Fe
Mg
Mn
Cu
Ti
Ni
Zn
Pb
Sn
98.45
0.7
0.5
0.40
0.3
0.2
0.2
0.1
0.1
0.1
0.05
Table 8.2 MECHANICAL PROPERTIES OF ALUMINIUM LM25 Density
2.7 g/cc
0.2% Proof Stress (N/mm2)
80-100
Tensile stress (N/mm2)
130-150
Elongation (0%)
2
Brinell Hardness Number
55-65
Endurance Limit Limit (5 ( 5 X 108 Cycles +
70-100
N/mm2) Modulus of elasticity ( X 103 N/mm2)
71
8.2 DISPERSION PHASE – Red Mud (RM) Reinforcement Reinforcement increases the strength, stiffness and t he temperature resistance capacity and lowers the density of MMC. In order to achieve these properties the selection depends on the type of reinforcement, its method of production and chemical compatibility with the matrix. MMC reinforcements can be metallic (such as tungsten and cobalt), non-metallic (most often carbon, graphite, or boron), or ceramic (for example, silicon carbide (SiC),aluminium oxide (Al2O3), boron nitride, tungsten carbide, titanium diboride, titanium
55
carbide). Here selected reinforcement material is Red Mud which posses very heavy/ dense having low porosity. It is beyond doubt that activity of primary industries often yields substantial amounts of by-products. The disposal in the original industrial site is favoured by economic reasons though traditional storage in nearby dumps can be impractical owing to the considerable masses involved and environmental environmental restrictions. rest rictions. The local exploitation of these by-products is therefore a growing technological aspect of basic industries and one tenable option is their r euse as starting materials for other productions.
This huge amount of industrial by-products / wastes which is becoming a client for increasing environmental pollution & generation of a huge amount of unutilized resources. An emblematic case is the ‗red mud‘ discharged by industry producing alumina from bauxite: alkaline digestion of 2.5 t of bauxite bauxite affords alumina and ≈1.5 t of red mud. The chosen Red Mud was in the size of 150 microns, and was bought from Madras Aluminium Company Ltd., (MALCO) The red mud finds good properties to be used as the reinforcement material. Some of the advantages are
Mechanical Mechanical compatibility (a thermal expansion expansion coefficient which is low but adapted to the matrix),
Chemical compa co mpatibility, tibility, Thermal stability,
High Young‘s modulus,
High compression and tensile strength,
Good process ability, economic efficiency.
56
CHEMICAL COMPOSITION OF RED MUD
The X-Ray diffraction image of the selected Red Mud is studied and the composition of the RM were listed out
Fig 8.2 XRD Image of Red mud Table 8.3: Chemical Composition of Red Mud Constituents
% Weight Weight
Constituents
% Weight Weight
Al2O3
15.0
Fe2O3
54.8
TiO2
3.7
SiO2
8.44
Na2O
4.8
CaO
2.5
P2O5
0.67
V2O5
0.38
Ga2O3
0.096
Mn
1.1
Zn
0.018
Mg
0.056
Organic C
0.088
L.O.I
Balance
57
CHAPTER 9 8. RAW MATERIAL PRE PROCESSING
The fabrication of Al RM metal matrix composite requires certain pre processing in order to make the process more effective and legible. The processes involved are explained briefly.
9.1 PROCESSING OF MATRIX AND REINFORCEMENT MATERIAL
The aluminum pieces were sized to the requirement i.e. to full fill the weight combination criteria. After sizing and weighing the raw material it is need to be cleaned for impurities. Aluminium pieces were washed in the bath of Sodium Hydroxide solution (NaOH) for 10 minutes. The quantity of the bath was 20ml in a glass beaker. The Red mud was crushed in Ball mill for about two hours in order to get the even sizing of the material. Then the milled red mud was sieved using three different sieves such as 150 microns, 90 microns and 75 microns. The sieved red mud were taken and the RM with 150 microns were taken for further processing
9.2 COMBINATION OF RAW MATERIALS The fabrication of Al Red Mud composite was planned to done in different compositions such as 0%, 5%, 10%, 15%, 20%, 25%, 30% of Red mud and remaining parts of Aluminium respectively. Initially two combinations were fabricated for pre studies. The combinations were tabulated in Table
58
Table 9.1 Combination of Raw materials S
Combination In Terms of Percentage
Combination In terms of Weight (gm)
No.
Al
RED MUD
Al
RED MUD
1
100
0
1000
0
2
95
5
950
50
3
90
10
900
100
4
85
15
850
150
5
80
20
800
200
9.3 INITIATION OF FURNACE AND STIRRER The furnace needs to initiate before the beginning of the melting process. To do so, the furnace was fed with coke in different layers and then it is fired using char coal. The furnace needs to attain certain temperature of range 800 to 1000 degree Celsius in order to melt the aluminum. For stir casting process, the stirrer should be set to a speed of 600 rpm. Then the to and fro motion of the stirrer is checked and oiled for free movement.
Fig 9.1 Stir casting setup 59
CHAPTER 10 10. FABRICATION OF METAL MATRIX COMPOSITE The method chosen for fabricating Al / RM Metal Matrix Composite (MMC) is Stir casting. This is because the processing expenses are low and also a better method to achieve dispersion in a low time and cost. Since the DC motor is used to stir there is no problem of clustering and uneven dispersion. The process of fabrication also includes die preparation. Here for different studies, three different moulds were prepared on Green sand mould using specified patterns. The different patterns used were a cylindrical rod of diameter 30mm and length of 300mm, a square of size 10X10 depth of 50mm, and a rectangle of size 10X50 and depth of 10mm
10.1 PREPARATION OF MOULD The part to be made and its pattern must be designed to accommodate each stage of the process, as it must be possible to remove the pattern without disturbing the molding sand and to have proper locations to receive and position the cores. A slight taper, known as draft, must be used on surfaces perpendicular to the parting line, in order to be able to remove the pattern from the mold. This requirement also applies to cores, as they must be removed from the core box in which they are formed. The sprue and risers must be arranged to allow a proper flow of metal and gasses within the mold in order to avoid an incomplete casting. Should a piece of core or mold become dislodged it may be embedded in the final casting, forming a sand pit, which may render the casting unusable. Gas pockets can cause internal voids. These may be immediately visible or may only be revealed after extensive machining has been performed. For critical applications, or where the cost of wasted effort is a factor, non-destructive testing methods may be applied before further work is performed. The green sand mould was selected as die because it is the simple and cheapest way to get castings in different dimensions. The green sand was made wet to the required consistency by adding water to it in a slower manner. Then the pre shaped wooden patterns of three different dimensions were made. They were a cylindrical rod of diameter 35mm and another rod of diameter 15mm and a rectangular pattern of size 100x100x50. The green sand is rammed well and vent holes were made at required places
60
Fig 10.1 Green sand mould
10.2 MELTING AND CASTING OF TEST SPECIMEN The weighted quantities of aluminium were melted to desired superheating temperature of 800 degree Celsius in graphite crucible. The coke fed furnace was used for melting. After melting was over, the required quantity of red mud particulates, preheated to around 400 degree Celsius were then added to the molten metal and stirred continuously by using mechanical stirrer. The stirring time was maintained between 60-80 seconds at an impeller speed of 550 rpm. After the aluminium is melted, Hexa chloro ethane was added as degasifier. The slag formed was removed periodically to avoid impurities in the casting. During stirring to enhance the wet ability small quantities of Magnesium was added to the melt.. The melt with t he reinforced particulates were then poured to a prepared moulds. After pouring is over the melt was allowed to cool and solidify in the mould. The casted specimens are displayed in the Fig 7.1
Fig 10.2 Fabricated Specimens 61
10.3 DISPERSSION PROCESSING In dispersion processes, schematically represented in Figure 8.1, the reinforcement is incorporated in loose form into the metal matrix. Because most metal reinforcement systems exhibit poor wetting, mechanical force is required to combine the phases, generally through stirring. This method is currently the most inexpensive manner in which to produce MMCs, and lends itself to production of large quantities of material, which can be further processed via casting or extrusion. The simplest dispersion process in current use is t he Vortex method, which consists of vigorous stirring of the liquid metal and the addition of particles in the vortex. Stirring with a specially designed impeller has the advantage of limiting the incorporation of impurities, oxides, or gases reduced vortexing. Ingots of such composites are now commercially produced in large quantities. Other methods being investigated include the bottom-mixing process, where a rotating blade is progressively lowered into an evacuated bed of particles covered with molten aluminium, and the injection of particles below the surface of the melt using a carrier gas. Compared with the unreinforced metals and alloys, the particle reinforced metal matrix composites (MMC) exhibit markedly higher stiffness and strength. So the composites are attracting more attention of automotive, aircraft and aerospace constructions. The production techniques have been well advanced in recent years, such as powder metallurgy, extrusion process and liquid infiltration. However in practice, it is often difficult to obtain a homogeneous distribution of reinforced particles. Further, it has been found that the mechanical properties of MMC are greatly influenced by the spatial distribution of the particles. Existing experimental and theoretical evidences suggest that the homogeneity of particle spatial arrangement plays a key role in controlling the yield strength, ductility, fatigue and fracture behaviour of MMC]. Although these behaviours are still quite poorly understood, there is general agreement that the microstructures with particle clustering tend to result in poorer mechanical properties. Therefore further understanding of the relationship between particle distribution and deformation mechanisms in MMC is of major importance for their engineering applications. At present the numerical analysis has been employed by a number of researchers to predict the effects of particles on the MMC. By and large, these analyses approached the problem by considering the unit cell model, where one particle was embedded in matrix. In addition, the shape of the particle was assumed to be cylindrical, spherical, rectangular or cubical]. The simplification aims at computations but fails capture the morphology such as particle size, shape and distribution. As a result, it fails to predict the 62
overall mechanical properties of MMC. Also, when more than one particle is considered, the particles are generally assumed to be the uniform random spatial arrangement and identical shape. In reality, the particle part icle microstructure is quite complex.
TESTING OF SPECIMEN 10.4 HARDNESS TEST Hardness is the property of a material to resist permanent indentation. Because there are several methods of measuring hardness, the hardness of a material is always specified in terms of the particular test that was used to measure this property. Rockwell, Vickers, or Brinell are some of the methods of testing. Brinell hardness testing is the most common method for hardness testing. In Brinell tests, a hard, spherical indenter is forced into the surface of the metal to be tested. The diameter of the hardened steel (or tungsten carbide) indenter is ranges from 5-10mm. Standard loads range between 500 and 3000 Kg; during a test, the load is maintained constant for a specified time (between 10 and 30 seconds). Here we used 5mm indenter ball diameter of the hardened steel, the 750 Kg load is maintained constant for a 10 seconds. Then indentation on the specimen is measured by hand microscope Fig. 10.3
Fig 10.3 Sample specimen for Hardness test
63
10.5 DENSITY OF CAST SPECIMEN A material's density is defined as its mass per unit volume. It is, essentially, a measurement of how tightly matter is crammed together. The principle of density was discovered by the Greek scientist Archimedes. The SI unit of density is kilogram per cubic meter (kg/m3). It is 3 also frequently represented in the cgs unit of grams per cubic centimetre (g/cm ).Archimedes‘ ).Archimedes‘
principle: An object weighs less in water than it does in the air. This loss of weight is due to the up thrust of the water acting upon it and is equal to the weight of the liquid displaced. The weight of the performs were taken in a digital balance that is weight of the perform in water. The readings are noted and tabulated The density was calculated using the formula = Wa/ (Wa-Wb) ( b
b)
where Wa is the weight in air, W b the weight in water and
is the density of water.
64
10.6 WEAR TEST
Experiments have been conducted in the Pin-on-disc type Friction and Wear monitor Pin On Disc Friction & Wear Testing Machine (19238622)with data acquisition system, which was used to evaluate the wear behaviour behaviour of o f the composite, against hardened ground steel disc (En 32) having hardness 65 HRC and surface roughness (Ra) 0.5 μm. It is versatile equipment designed to study wear under sliding condition only. Sliding generally occurs between a stationary Pin and a rotating disc. The disc rotates with the help of a D.C. motor; having speed range 0-2000 rev/min with wear wear track diameter 10-140 mm which could yield sliding sliding speed 0.26 to 10m/s. 10m/s. Load is to be applied on pin (specimen) by dead weight through pulley string arrangement. The system has a maximum loading capacity of 200N. The tests have been carried out under the following conditions: • The specimens under tests were fixed to the collect. The collect along with the specimen (Pin) is positioned at a particular track diameter 90mm. This track diameter is to be changed after each tests i.e. a fresh track is to be selected for each specimen. During experiment the specimens remains fixed and disc rot ates.
• Load is applied through a dead weight loading system to press the pin against the disc. • Frictional force arises at the contact can be read out from the controller. • The speed of the disc or motor rpm can be varied through the controller. • For a particular type of composite 27 sets of test pieces were tested. • Each set of test was carried out for a period of 6 hrs run. After each one hour run the test pieces were removed from the machine and weighted accurately to determine the loss in weight.
65
CHAPTER 11 11. RESULTS AND DISCUSSION A detailed study was undertaken to pool-up the existing literature on Aluminium based MMCs and efforts were put to understand understand the basic needs of the growing Composite industry. This includes various aspects such as Characterization, fabrication, testing, analysis and correlation between microstructure and the t he properties obtained.
The conclusions drawn from this study are • Pure aluminium matrix is preferred to various alloy matrices due to the high temperature stability of the aluminium as compared with aluminium alloys. Lower working temperature‘s in case of alloy matrices is attributed to lo lower wer stability of the alloy matrix and coarsening of the grains. In addition, the load transfer in case of pure aluminium matrix is more effective due to the clean interface. • There exists a wide range of database in the literature for different types of reinforcements reinforcements in Aluminium Metal Matrix Composi Co mposites. tes. • In particle reinforced composites, the fracture mode was observed to depend on reinforcement purity, reinforcement particle size, and nature of interface, volume fraction of reinforcement, fabrication route adopted, and extent of hot working, presence of any intermetallic precipitates and extent of coherency of second phase with the matrix. • There are varieties of techniques available for production of metal matr ix composite. composite. Each having its own merits and demerits.. In particular, some are far more expensive than others. The manufacturer generally prefers the lowest cost route. Therefore, stir-casting technique represents represents a substantial substantial proportion of t he MMCs in commercial sectors today.
Thus the priority of this work will be to prepare MMC using red mud (an industrial waste from Bayer‘s process) as reinforcement material and to study its wear characteristics. The effect of different dependant factors primarily sliding velocity, normal load, effect of heat treatment temperature and cooling media are also to be studied.
66
Table 11.1 Hardness values of Casted specimens Load: 750 kgf
Indenter Ball diameter : 5 mm S.No
Indentation Time : 10 Sec
SPECIMEN
HARDNESS (BHN)
1
0
116.854
2
5
136.657
3
10
139.0031
4
15
162.854
5
20
169.261
6
25
172.072
7
30
175.300
Table 11.2 Density values of Casted Specimens S.No
SPECIMEN
DENSITY (g/cc)
1
0
2.68
2
5
2.96
3
10
2.98
4
15
2.99
5
20
2.96
6
25
2.95
7
30
2.97
Table 11.3 Wear test on casted Specimens
S.No
SPECIMEN
LOAD
VELOCITY
WEAR
WEIGHT
Frictional
(N)
(m/s)
(microns)
LOSS (g)
force(N)
1
Pure
30
2
163
0.0133
2.1
2
5
50
3
180
0.0268
6
3
10
70
4
434
0.0596
6.2
4
15
30
2
69
0.0054
5.3
5
20
50
3
257
0.0671
17.8
6
25
70
4
414
0.0627
27.4
7
30
30
2
98
0.0110
3.7
67
BHN 200 180 160 N H B n i s s s e n d r a H
140 120 100 80
BHN
60 40 20 0 0
5
10
15
20
25
30
% of Red Mud
Fig 11.1 Graph showing change in hardness
3.05 3 2.95 2.9 c c / g n i y t i s n e D
2.85 2.8 2.75 Density
2.7 2.65 2.6 2.55 2.5 0
5
10
15
20
25
% of Red Mud
Fig 11.2 Graph showing change in density
68
30
Artificial Neural Network (ANN) was employed to develop mathematical mode modell for wear prediction using Decision Prediction Tool 5.7.1. The accuracy of the developed model is 83%. The image of output obtained during the operation of Neural Network tool were given below
69
70
CHAPTER-12 12 .REFERENCE 1. Kok M. Production and mechanical properties of Al 2O3 particle-reinforced 2024 – aluminium alloy composites. Journal Material Processing Technology (2005); 161:381 161:381 – 7. 2. Degischer HP. Innovative light metals: metal matrix composites and foamed aluminium. Material Design (1997);18 :221 – 6. 6. 3. Rasit Koker , Necat Altinkok , Adem Demir, Neural network based prediction of mechanical properties of particulate reinforced metal matrix composites using various training algorithms, Materials and Design 28 (2007) 616 – 627. 627. 4. Dobrzanski LA, Wlodarczyk A, Adamiak M. The structure and properties of PM composite materials based on EN AW-2124 aluminium alloy reinforced with the BN or Al2O3 ceramic particles. J Mater Process Technology, (2006); 175: 186 – 186 – 91. 91. 5. Torralba JM, daCost CE, Velasco F. P/M aluminium matrix composites: an overview. J Material Processing Technology 2003-133:203 – 2003-133:203 – 6. 6. – Cu 6. Liu HN, Ogi K. Dry sliding wear on Al2O3 continuous fibre reinforced Al Al – Cu alloy against steel counterface. J Mater Sci 1999-34:5593 – 1999-34:5593 – 9. 9. 7. Moustafa SF, Soliman FA. Wear resistance of d-type alumina fibre reinforced Al – Al – 4% 4% Cu matrix composite. Tribol Lett 1997-3-311 – 5. 5. 8. Hossein Abdizadeh , Maziar Ashuri , Pooyan Tavakoli Moghadam , Arshia Nouribahadory , Hamid Reza Baharvandi, Improvement in physical and mechanical properties of aluminium/zircon composites fabricated by powder metallurgy method, Materials and Design, xxx (2011) xxx – xxx. xxx. 9. Yusuf S-ahin, Abrasive wear behavior of SiC/2014 aluminium composite, Tribology International 43 (2010) 939 – 939 – 943. 943.
71
10. Mehdi Rahimian, Nader Parvin, Naser Ehsani, The effect of production parameters on – Al2O3 microstructure and wear resistance of powder metallurgy Al Al – Al2O3 composite, Materials and Design 32 (2011) 1031 – 1031 – 1038. 1038.
11. Mehdi Rahimian, Nader Parvin, Naser Ehsani, The effect of particle size, sintering temperature and sintering time on the properties of Al – – Al2O3 Al2O3 composites, made by powder metallurgy, Journal of Materials Processing Technology 209 (2009) 5387 – 5387 – 5393. 5393. 12. Mazahery A, Abdizadeh H, Baharvandi HR. Development of high-performance – 4. A356/nano-Al2O3 composites. Mater Sci Eng A 2009;518:61 2009;518:61 – 4. 13. Ansary Yar A, Montazerian M, Abdizadeh H, Baharvandi HR. Microstructure and mechanical properties of aluminium alloy matrix composite reinforced with nanoparticle MgO. J Alloy Compd 2009;484:400 – 4. 4.
14. Ansary Yar A, Montazerian M, Abdizadeh H, Baharvandi HR. Microstructure and mechanical properties of aluminium alloy matrix composite reinforced with nanoparticle MgO. J Alloy Compd 2009;484:400 – 4. 4. 15. Sahin Y, Murphy S. The effect of fibre orientation of the dry sliding wear of borsicreinforced 2014 aluminium alloy. J Mater Sci 1996;34:5399 – 1996;34:5399 – 407. 407. 16. Metals handbook, powder metallurgy. 9th ed. vol. 7. American Society for Metals; 1993. 17. Sahin Y. Preparation and some properties of SiC particle reinforced aluminium alloy composites. composites. Mater Des 2003; 24:671 – 9. 9.
72
CHAPTER-13 ANNEXURE List of Publications:
This project work was presented in th thee National level student‘s conference on ―DESIGN, MATERIAL & CONSTRUCTION‖ on 25 th & 26th August 2011 conducted by VEL TECH DR.RR & DR.SR TECHNICAL UNIVERSITY, Avadi, Chennai. We were awarded with the First best paper award, under the stream MECHANICAL.
73
CHAPTER 14 CONCLUSION
Form the experimentation; it was found that while using low particle size reinforcement reinforcement material the possibility possibility of float or tend tend to sink will be more more depends upon their density liquid metal and so that dispersion dispersion of the particles are not uniform. Therefore it is decided to use different stirrer model with varying speed and timing. Finally it is decided to use four blades with 90 degree angle, stirring speed 550 rpm and time is 60 sec to get proper distribution of the reinforcement material. Red mud, the waste generated from alumina plant can be successfully used as a reinforcing material to produce Metal-Matrix Composite (MMC) component in aluminum matrix to be used in wear environment. It can be successfully used in place of conventional aluminum intensive material. Increase in % of reinforcement of the material increases the hardness of the work piece, since reinforcement material harder than the matrix material. The density of the composite will obviously increases since increase in reinforcement increases the density of the material as per the rule of mixture. The wear rate of the composite decrease with increases in % of reinforcement materials. Finally, the wear property of the composite depends on many factors, such as sliding velocity, sliding distance and load. Computation through neural networks is one of the recently growing areas of artificial intelligence. Neural networks are promising due to their ability to learn highly non-liner relationship. It can also be gainfully employed to simulate property-parameters correlation ship in a space larger than the experimental domain. It is evident from the present study that the artificial neural technique has the potential to predict and analyze the wear behavior of metal matrix composites if it is properly trained.
74
