ARME Vol-1 No-2 pdf

An International Peer-Reviewed Journal on Mechanical Engineering

w w w . t r p . o r g . i n

Asian Review of Mechanical Engineering

Journals Division

THE RESEARCHP U B L I C A T I O Nw w w . t r p . o r g . in

ISSN 2249 - 6289Vol. 1 No. 2

July - December 2012Asian Review of Mechanical Engineering

ARME Special Issue

AFTMME ‘12

Special Issue

Published from the Proceeding of International Conference

on

Advancements and Futuristic Trends in Mechanical and Materials Engineering

AFTMME’12

October 5 - 7, 2012

Editors

Dr. Buta Singh Sidhu

Dr. H.S. Bains

Dr. Hazoor Singh Sidhu

Er. Pardeep Kumar Jindal

Er. Sukhpal Singh Chatha

Er. Rakesh Bhatia

Er. Harish Garg

Organized by

Punjab Technical University

Jalandhar -Kapurthala Highway

Kapurthala,

Punjab – 144 601, India

In collaboration with

Department of Science and Technology (DST)

Defence Research and Development Organisation (DRDO)

Council of Scientific and Industrial Research (CSIR)

Preface

It is our immense pleasure to publish the papers of International conference on “Advancements and

Futuristic Trends in Mechanical and Materials Engineering” (October 5-7, 2012) organized by Punjab

Technical University (PTU), Jalandhar.

The world of engineering is rapidly changing and the growth of technological know-how has been

catalytic in the recent achievements and the ones to come in the near future. The success of engineers and our

profession depends on how well we can adapt to these changes. Futuristic trends in mechanical engineering

are the need of the hour. The parameters that determine the standard of achievement are precision, quality,

efficiency, economy, reliability and acceptability. This conference provides a common platform to the

mechanical engineers in the emerging interdisciplinary areas and recent technologies to meet the future

challenges. The contributed papers have been selected through a peer review process by the distinguished

experts. A word of thanks to our team, who has worked hard day in and out in compiling and publishing the

proceedings of this conference. We are greatly indebted DST, DRDO and CSIR without whose assistance; it

would not have been possible to organize such a resourceful event.

A special thanks to all the individuals and institutions who contributed to the success of the conference,

the authors for submitting papers as well as the sponsors for their generous financial, and logistical support.

Editors

PATRON

Dr. Rajneesh Arora (Vice-Chancellor)

CONVENER

Dr. Buta Singh Sidhu (Dean Academics)

CO-CONVENER

Dr. H.S. Bains (Registrar)

ORGANIZING SECRETARY

Dr. Hazoor Singh (Associate Prof., YCoE, Talwandi Sabo)

Er. Pardeep Jindal (Assistant Prof., YCoE, Talwandi Sabo)

CO-ORGANIZING SECRETARY

Er. Sukhpal Singh Chatha (AP, YCoE, Talwandi Sabo)

Er. Rakesh Bhatia (AP, YCoE, Talwandi Sabo)

Er. Harish Kumar Garg (AP, GZSCET, Bathinda)

EVENT CO - ORDINATOR

Ms. Madhu Midha (Asstt. Librarian)

LOCAL ADVISORY COMMITTEE

Dr. Nachhattar Singh Advisor to VC

Dr. H. S. Bains Registrar

Dr. A. P. Singh Dean, Student Affairs

Dr. R. P. Bhardwaj Director, Recruitment

Mr. S. K. Mishra Director, Finance

Dr. Balkar Singh Director, Secrecy

CONFERENCE COMMITTEE

Dr. Satya PrakashEmeritus Professor,Department of Material and Metallurgy Engineering,IIT, Roorkee, Uttrakhand,India

Prof. Raman Singh, DirectorDepartment of Mechanical & Aerospace Engineering, Monash University - Clayton Campus, (Melbourne), Vic 3800, Australia

Prof. Mousa S. MohsenEditor-in-Chief, Jordan Journal of Mechanical andIndustrial Engineering, Hashemite University,Zarqa, Jordan

Dr. Md. Aminul IslamProfessor of Materials & Metallurgical Engg.,Bangladesh University of Engg. & Tech., Dhaka -1000, Bangladesh

Dr. Manoj GuptaDept. of Mechanical Engineering, National University of Singapore, Singapore - 119260

Dr. R. ArulmaniESSO-KTT Project, Singapore

Prof. B. Ben-NissanFaculty of Science, University of Technology, Sydney, Broadway 2007, NSW, Australia

Prof. M Gwyn HockingProfessor in Material Chemistry,Imperial College London, United Kingdom

Dr. A.S.W. KurnyBangladesh University of Engineering & Technology, Dhaka-1000,Bangladesh

Dr. Shantanu BhowmikSenior Scientist, Singapore Institute ofManufacturing Technology, Singapore - 638075

International Advisory Board

Dr. Pardeep KumarProfessor, Department of Mech. Engg.,IIT Roorkee, Uttrakhand, India

Dr. Subhash ChanderDepartment of Mech. Engg., NIT, Jalandhar, Punjab, India

Dr. S.P. SinghProfessor, Department of Mech. Engg, IIT Delhi, India

Dr. RajnishTyagiBanaras Hindu University, Varanasi, Uttar Pradesh, IIT Kanpur, India

Dr. Harpreet SinghDepartment of Mech. Engg, IIT Ropar, Punjab, India

Dr. D.S. HiraVice-Chancellor, GKU, Talwandi Sabo, Punjab, India

Dr. Sunil PandayDirector, SLIET, Longowal, Punjab, India

Dr. Kulwant SinghDept. of Mech.Engg., SLIET, Longowal,Punjab, India

Dr. A.S. KhannaProfessor, Department.of Metallurgical Engg. & Material Sci., IIT Bombay, India

Dr. S.K. RoyProfessor, Department of Metallurgical Engineering & Material Science,IIT Kharagpur, India

Asian Review of Mechanical EngineeringVolume 1 Number 2 July - December 2012

CONTENTS

Sl. No. Title Page No.

1. A FEM and Image Processing Based Method for Simulation of 01

Manufacturing Imperfections

Arshad Javed, A. K. Sengar and B.K. Rout

2. Structural and Optical Investigation of Aluminium-Lithium-Borate Glasses 06

Gurinder Pal Singh, Parvinder Kaur, Simranpreet Kaur, Deepawali Arora and D.P. Singh

3. Performance Comparison of Single and Double Layer Microchannel Using 09

Liquid Metal Coolants: A Numerical Study Deewakar Sharma, Harry Garg and P.P. Bajpai

4. The Thermodynamic Study of Turbocharger Pressure Ratio and Ambient Temperature 18

Variation on Exergy Destruction Estimation of Homogeneous Charge Compression

Ignition Engine Cogeneration System Shailesh Kumar Trivedi and Abid Haleem

5. Experimental Investigations of Traveling Wire Electro-Chemical Spark Machining 24

(TW-ECSM) of Borosilicate Glass

Basanta Kumar Bhuyan and Vinod Yadava

6. Performance Characteristics of Diesel Engine Fueled by Biodiesel of 30

Jatropha Oil and Soybean Oil Ashish Malik and Parlad Kumar

7. Machining Study of TI-6AL-4V Using PVD Coated TiAlN Inserts 34

Narasimhulu Andriya, Venkateswara Rao P and Sudarsan Ghosh

8. Investigation of the Structure and Mechanical Properties of 41

Friction Stir Welded Aluminum Alloy

A. Chandrashekar, B. S. Ajay Kumar, V. Anandkumar and P. Raghothama Rao

9. Multi-Objective Optimization of the Electro-Discharge 45

Diamond Surface Grinding Process

Shyam Sunder and Vinod Yadava

10. Enhancing Wear Resistance of Low Alloy Steel Applicable on 51

Excavator Bucket Teeth Via Hardfacing

Shivali Singla, Amardeep Singh Kang and Jasmaninder Singh Grewal

11. Creep Modeling in An Orthotropic FGM Cylinder 55

Ashish Singla, Manish Garg, Dharmpal Deepak and V. K. Gupta

A FEM and Image Processing Based Method for Simulation of Manufacturing Imperfections

Arshad Javed*, A. K. Sengar and B.K. Rout Department of Mechanical Engineering

Birla Institute of Technology and Science, Pilani – 333 031, India

*Corresponding Author E-mail: [email protected]

Abstract - Use of appropriate methods to capture

manufacturing imperfection at the conceptual stage is a major

challenge for the designer and researchers in industry.

Imperfections are observed in almost all type of in macro,

micro and nano-machining domain of manufacturing process.

These imperfections lead to undesirable performance in

application phase. In the present work, a simulation based

approach to handle manufacturing imperfection is

implemented using image processing operators. This method

simulates the image of the component due to manufacturing

imperfections. The usage of these image processing operators

facilitates a realistic simulation of manufacturing errors, in

macro, micro, and nano domain manufacturing. The

simulated image is further processed for its structural

properties i.e. maximum deflection, reactions, Von Mises

stress, and change in amount of material, corresponding to its

intended application. In order to generate these results based

on modified image of beam, the concept of "Solid Isotropic

Material with Penalization"(SIMP) is utilized along with 2-D

finite element routine. An example of a simple cantilever beam

is selected to illustrate the proposed methodology, and the

results are analyzed. The present work discusses a simple and

easy method to predict the behavior of designed component

prior to its manufacturing.

Keywords: Manufacturing imperfection, Nano-fabrication,

Micro-fabrication, FEM, Image processing, SIMP

I. INTRODUCTION

Manufacturing errors and imperfection are a common

part of any product or component. Imperfections emerged

out by various means. One of the approach is to use a very

accurate process, where cost and effort increases

exponentially. Other approach is to reduce the

consequences of manufacturing imperfection, by

performance simulation prior to application. By simulating

the performance, the parameters of boundary can be

adjusted to get the intended results. The imperfections are

observed at macro level and also spread across the domain

of micro and nano scales. At this level, the manufacturing

errors lead to imperfections of the components, which

change the property of the component drastically. Thus, the

required performance cannot be achieved [1-4]. In such

scenario, it is very difficult to produce a precise component.

Hence, consideration of imperfection becomes a vital issue.

Present work focuses on the simulation issue of geometric

imperfection of a structural component. The manufacturing

processes considered are, milling for macro, etching for

micro and electron beam lithography and laser

micromachining for nano domain. To carry out simulation

process, image processing operators i.e. dilate and erode are

used [5, 6]. Presently, researchers are using these operators

for the simulation of geometric imperfections [7], however,

the application is limited. In order to use the image

processing operators efficiently, it is necessary to know its

corresponding effect on the property of the structural

component. To fulfill this requirement, the concept of solid

isotropic material with penalization (SIMP) is applied.

SIMP is a methodology to optimize the topology of a

structural component as well as synthesizing a compliant

mechanism even at nano level [8, 9]. SIMP applies a

particular approach to get the details of structural property

of any arbitrary shape using the image of structural

component. Based on image processing operator and SIMP,

the simulations are carried out and effects on the structural

component are recorded. The objective of this work is to

explore the behavior of the property of a structural

component, over the variation of image processing

operators. A methodology is proposed to fulfill the

objective. It can be applied to any type of shape of structural

component. It will help the designer and practitioners to set

a particular value of image processing operator to simulate a

real time geometric imperfection.

The manuscript is organized in following manner. In

section two, the details of the image processing operators is

presented. The approach of SIMP is given in section three.

The overall methodology to apply the simulation process is

discussed in section four. The details of selected problem

and its simulation results are discussed in section five and

six respectively. Finally conclusion is drawn.

1 ARME Vol.1 No.2 July - December 2012

2

II. IMAGE PROCESSING OPERATORS

Image processing operators are used for single

processing of the original image. There are several image

processing operators available to modify or process an

image into desired form. From vast ranges of operators, two

specific operators are selected i.e. Dilate and Erode, for

binary image. This selection is based on their capability to

simulate the actual manufacturing imperfections [7].

Dilation (Dilate) allows the image boundary of empty

(white) region to expand in filled (black) portion. This

makes the total area of empty region to increase. Erosion

(Erode) allows the filled image boundary to expand in the

empty region. Thus, the total area cover by filled region of

the image increases. The effect of Dilate and Erode can be

observed in Fig. 1. These operators simulate the

imperfections corresponding to milling, etching, electron

beam lithography, and laser micromachining. The

operators can be tailored by the proper selection of the

structuring element (SE) that decides exactly how the

object will be dilated or eroded. Based on the suitability to

simulate the manufacturing imperfection few are chosen in

the present work. These SEs are line, disk, square, and ball

(Table I).

III. SOLID ISOTROPIC MATERIAL WITH

PENALIZATION (SIMP)

In the present work, the structural property of the

component is computed using SIMP approach, associated

with finite element method (FEM). In this section, a brief

detail of SIMP is presented. Initially, Bendsoe and Kikuchi

[10] developed a homogenization approach, which is the

basis of the present state of art in SIMP. The problem is

worked out in a descritized approach, where the whole

design domain is considered element wise. Each element is

assigned a density parameter, which expresses the

existence of “material” or “no material”. Strictly speaking,

this design parameter is the material presence expressed in

fractional values for the each element and it is constant

within each element. The value of density parameter is

relaxed in a continuous interval from zero to one, by

inclusion of gray elements. Physically, the gray elements

Fig.1 (a) Original image (b) Dilated image(c) Eroded image

are equivalent to an intermediate state between solid or

void. The assumed intermediate states are purely

mathematical and it cannot be implemented at

manufacturing level. To make it feasible, density

parameters are penalized using a power-law approach that

produces an approximate discrete solution. A very high

penalization turns the density parameter near to one or zero

that represents empty or solid condition. Usually

penalization power is taken as three. This technique is

termed as SIMP [11].

In the present work, the SIMP is used in a reversed way.

Here the process starts with a given density value, which is

generated from the simulation of image. The stepwise

methodology of this process is given in next section.

IV. SIMP METHODOLOGY

The proposed methodology is implemented by creating

a MATLAB code. The steps of this process are summarized

below.

Step 1. Selection of a structural component (beam).

Step 2. A black and white image of the beam is created,

where black represents the presence of material.

Step 3. Image is modified using Dilate or Erode

operators. It represents the manufacturing

imperfection.

Step 4. The modified image is read and a matrix of its

corresponding pixel values is generated.

Step 5. The values of the pixel matrix are normalized

between zero to one. Gray elements are

represented by intermediate values.

Step 6. The elements of the pixel matrix, which are

having values equal to zero, are assigned a min

value (0.001). It makes valid operations of

matrixes while applying FEM. The resultant

matrix is called as "density matrix".

Step 7. Different modulus of elasticity values are

generated for each element of density matrix. It is

carried our using power law approach of SIMP.

Step 8. The values of modified elasticity values for the

elements are preceded for FEM based routine.

Step 9. Using FEM, the values of reaction on beam-

supports, nodal deflection, and Von-Mises stress

are computed.

V. EXAMPLE

In the present work, a cantilever beam shown in Fig. 2 is

selected for the application of proposed methodology. The

ARME Vol.1 No.2 July - December 2012


3

length of the beam is 20 mm, width is 4 mm and the

thickness if 1 mm. The beam is having internal curves. The

modulus of elasticity and the Poisson's ratio are selected as

200 GPa and 0.3 respectively. At the lower right end of the

beam, a load of 10 N is applied as shown in Fig. 3. The

simulations are performed on this beam, and the results are

provided in the next section.

VI. RESULTS AND DISCUSSION

As per the methodology given in section four, the

simulations are carried out. The Dilate and Erode operators

are set for few SEs which are given in Table I.

TABLE I STRUCTURING ELEMENT TYPES SELECTION FOR SIMULATION

The actual image of beam (Fig.2) is dilated and eroded,

for each SE values. Simulated images are shown in Fig.4. In

these images, the effect of dilation (Fig. 4(a)), and erosion

(Fig. 4(b)) can be seen with respect to its SE number. In

dilation, the amount of material of beam is reduced, while in

erosion, the amount of material of beam is increased. The

variation in the material of beam i.e. mass fraction, is

controlled by the SE parameters. The SE parameters are

tuned to produce a particular amount of variation, which are

referred from the actual case of imperfections [12]. The

variation of mass fraction from the actual one, can be

Fig. 2 Dimensions of selected beam, in mm

Fig. 3 The corresponding black and white image of the cantilever beam

SE No. Type1

Line 12 Line 23

Disk 1

4

Disk 25 Square 16 Square 27 Ball 18 Ball 2

observed in 5(a). It shows that for SE-4, the mass fraction is

having highest and lowest values. For SE-7, the mass

fraction is very near to the actual values. The deflection in

upward and downward direction can be observed from

Fig.5 (b) & (c) respectively. When the load is applied, the

curvature of the curved part of the beam is changed and the

nodes in that region will have the upward deflection. It can

be seen that, deflection in both cases are high for dilated

beam. It is due to the removal of material from the beam.

This relationship is not direct, as for dilated-SE-5 the

upward deflection is increased though the mass fraction is

higher for SE-4. For SE-7, the deflections are near to the

actual values in both cases. The downward deflection is

highest for dilated-SE-4, and lowest for eroded-SE-4.

Maximum reaction values at the fixed end of the beam are

also simulated for the selected beam, as shown in Fig. 5(d).

The reaction is highest for dilated-SE-1 and lowest for

eroded-SE-4. The reactions are almost equal to the actual

value at SE-7, in both cases. For the failure check, Von-

Mises stress is computed for each case of the beam, as

shown in Fig. 5(e). For dilated-SE-1, 3, 6, 7 & 8, it is

approaching to the actual value. It is highest for dilated-SE-

4, where the mass fraction is lowest. Apart from eroded-SE-

2 & 7, the Von-Mises stress values for all SEs are lower than

the actual values. The lowest value is observed at eroded-

SE-4, where the mass fraction is highest.

In these result, the effect of different SEs can be

observed. The highest and lowest effect is made by disc2

and ball1 SE, respectively. Hence, for a fine variation of

properties, ball1 SE should be used in simulation and for

coarse variation, disc2 SE should be used. It can be seen that

the values are in direct relationship with mass fraction,

leaving few cases. The reason for this may be the way in

which material is dilated or eroded from the beam. In

addition, the mirror image property of dilate and erode

operators can also be observed. The maximum variation of

the mass fraction is also same for the dilation and erosion,

i.e. around 9% of the actual value. The obtained values

depend on the topology of the beam. However, the

observation will be similar among the different properties.

The obtained results indicate that the method is beneficial to

simulate the manufacturing imperfections efficiently.



4

Fig. 4(a) Dilated images for SE numbers

Fig. 5(b) Variation of upward deflection

Fig. . 5(c) Variation of downward deflection

Fig. 5(d) Variation of maximum reaction

Fig. 4(b) Eroded images for SE numbers

Fig. 5(a) Variation of Mass fraction



5

Fig. 5(e) Variation of maximum Von-Mises stress

VII. CONCLUSION

In any manufacturing process, imperfections are

obvious. However, it is imperative to simulate the

imperfection before the actual process. In the present work,

the imperfection are simulated which are valid for

machining, etching, electron beam lithography, and laser

micromachining. In this simulation process, image-

processing operators are used along with SIMP method.

The variations among the different SEs are observed for a

selected cantilever beam problem. Their effects on beam

deflection, reaction, and Von-Mises stress are simulated

and analyzed. Present work will be helpful to the

practitioner to select a specific SE to simulate the

imperfections. This work can be extended for fine

examination of each SE in different topologies of structural

components.

REFERENCES

[1] Nicolae L. and Garcia E. (2005), Mechanics of Microelectro

mechanical Systems, New York, Kluwer Academic, pp. 343-381.

[2] Daniel R. K., Dominik V. S. and Robert H. B. (2004), “Drastic

Enhancement of Nanoelectromechanical-System Fabrication Yield

Using Electron-Beam Deposition”, Applied Physics Letters, Vol. 85,

pp. 157-159.

[3] Steve R. and Robert P. (2001), "A Review of Focused Ion Beam

Applicat ions in Microsystem Technology", Journal of

Micromechanics and Microengineering, Vol. 11, pp. 287–300.

[4] Hyunseok K., Chulki K., Minrui Y., Hyun-Seok K. and H. B. Robert

(2010), “Local Etch Control for Fabricating Nanomechanical

Devices,” Journal of Applied Physics, Vol. 108, pp. 1-3.

[5] William K. P. (2007), Digital Image Processing,

[6] Solomon, C. and Breckon, T. (2011), Fundamentals of digital image

processing: a practical approach with examples in Matlab,

Chichester, West Sussex: Wiley-Blackwell.

[7] Sigmund O. (2007), "Morphology-Based Black and White Filters for

Topology Optimization", Structural and Multidisciplinary

Optimization, Vol. 33, pp. 401-424.

[8] Javed A., Rout B. K. and Mittal R. K. (2007), "A Review on Design

and Synthesis of Compliant Mechanism for Microactuation",

Proceedings of 2nd ISSS National Conference on MEMS, micro

sensors, smart materials, structures and systems, Pilani, India, pp. 1-

10.

[9] Elesin Y., Wang F., Andkjær J., Jensen J.S. and Sigmund O. (2012),

Topology optimization of nano-photonic systems, Integrated

Photonics Research, Silicon and Nanophotonics (IPRSN), Theory,

Modeling & Simulations I: Numerical Methods (IM2B), Colorado

Springs, Colorado.

[10] Bendsøe M. P. and Kikuchi N. (1988), "Generating Optimal

Topologies in Structural Design using a Homogenization Method”,

Computer Methods in Applied Mechanics and Engineering, Vol. 71,

pp. 197-224.

[11] Bendsøe M. P. and Sigmund O. (2003), Topology Optimization:

Theory, Methods and Applications, Berlin, Springer, 2003, pp. 2-68.

[12] Sigmund O. (2009), "Manufacturing Tolerant Topology

Optimization”, Vol. 25, pp. 227-239.

John Wiley & Sons.

Acta Mechanica Sinica,



Structural and Optical Investigation of Aluminium-Lithium-Borate Glasses

Gurinder Pal Singh, Parvinder Kaur, Simranpreet Kaur, Deepawali Arora and D.P. SinghDepartment of Physics, Guru Nanak Dev University, Amritsar – 143 005, India

E-mail: [email protected]

Abstract - Glass samples of compositions xAl O -(30- ) 2 3 x

Li CO –70B O with x varying from 0 to 8% mole fraction are 2 3 2 3

prepared by melt quench technique. Decrease in the band gap

from 3.12 to 2.91 eV for lithium borate glasses with an increase

in the Al O content has been observed and discussed. The 2 3

FTIR spectral studies have pointed out the conversion of

structural units of BO to BO . Due to the formation of BO and 3 4 4

AlO units, changes in the atomic structure with Al O 6 2 3

composition are observed and discussed.

Keywords: X-ray diffraction, Optical properties, FTIR

I. INTRODUCTION

The study of oxide glasses has received considerable

attention due to their structural peculiarities [1]. These

glasses have wide applications in the fields of electronics,

nuclear and solar energy technologies and acoustic-optic

devices [2]. In addition, they are often used as dielectric and

insulating materials and it is known that borate glass

constitutes a good shield against IR radiation . It is well

known that the main structural units of the borate network

which are [BO ] triangles and [BO ] tetrahedral, may form 3

different super -structural units; boroxol and meta-borate

rings, meta-borate chains, penta-borate,tri-borate, diborate

and pyro-borate [3].

The addition of alkali oxides can improve many

properties of borate glasses as well as modify, even improve

their preparation conditions. Lithium is an important alkali

cation and Al O is an important modifier. Glasses based on 2 3

the lithium aluminum system have attracted considerable

interest in recent years due to their significant applications

in science and industry. Borate glasses containing Lithium

have been extensively studied due to their technological

applications as solid electrolyte in electro chemical devices

such as batteries [4]. Alkali borate glasses are highly useful

mater ia ls for vacuum ul t ra v io le t opt ics and

semiconductors lithography owing to the presence of stable

glass forming range and transparency from the near UV to

the middle infrared region [5].

4

Another oxide, Al O cannot form a glass by itself. It can 2 3

form glass once it is added with another suitable oxides and

it will take part in the formation of the glass structural unit.

Alkali free alumina lead borate glasses are very stable

against devitrification possess high mechanical strength,

toughness, moisture resistant and excellent electrical

properties. Due to this these glasses have application in

battery sealing and microelectronic packing [6-8]. The

addition of an Al O is anticipated to enhance the chemical 2 3

durability of the glasses while simultaneously increasing

the glass transition temperature and reducing the thermal

expansion coefficient [9-11].

The present work investigates the dominant role of

Al O on structural and optical properties in Li CO -B O 2 3 2 3 2 3

glass system. The structural properties are studied by using

XRD (x-ray diffraction), Fourier transform infrared

spectroscopy (FTIR) techniques. The optical properties of

glasses are determined by using UV-visible spectroscopy

measurements.

II. EXPERIMENTAL DETAIL

A. Sample Preparation

Glass samples xAl O -(30-x)Li CO –70B O with x 2 3 2 3 2 3

varying from 2 to 10 mol % are prepared by the

conventional melt quench technique. The raw materials of

Lithium Carbonate (Li CO ), Aluminium Oxide (Al O ) and 2 3 2 3

Boric oxide (B O ) of appropriate amounts are mixed 2 3

together and melted in silica crucible at temperature range

of 1100° C for 60 minutes until a bubble free liquid was

formed. The melt is then poured in to preheated steel mould

and annealed at temperature of 380° C for 1 hour to avoid

breaking of the samples by residual internal strains. The

obtained samples are grinded with different grade of silica

carbide and polished with cerium oxide in order to obtain

maximum ? atness. The nominal composition of the

prepared glasses is given in the Table I.

The amorphous/crystalline nature of the samples is

confirmed by X-ray diffraction (XRD) study using

(Shimadzu, Japan) X-ray diffractometer at the scanning

rate of 2 degree/min and 2è varied from 10–70°.

ARME Vol.1 No.2 July - December 2012 6

The Optical Absorption spectra of polished samples are

recorded at room temperature by using UV-Visible

Spectrophotometer (Perkin Elmer) in the range from 200-

800 nm.

The infrared transmission spectra of the glasses are

measured at room temperature in the wave number range -1400–4000 cm by a Fourier Transform computerized infra-

red spectrometer type (Thermo Nicolet 380 spectrometer).

The prepared glasses are mixed in the form of ? ne powder

with KBr in the ratio 1:100 mg glass powder: KBr,

respectively. The weighed mixtures are then subjected to a 2pressure of 150 kg/ cm to produce homogeneous pellets.

The infrared transmission measurements are measured

immediately after preparing the pellets.

III. RESULTS AND DISCUSSIONS

A. X-Ray Diffraction

The x-ray diffraction pattern (Fig. 1) does not reveal any

crystalline phase in Al O -Li CO –B O glass samples 2 3 2 3 2 3

which indicate the amorphous nature of the samples.

TABLE I NOMINAL COMPOSITION (MOLE %), AND BAND GAP OF

GLASSES

B. FTIR

The infrared transmittance spectra of glasses in the -1400–4000 cm region shown (Fig.2) has large, medium,

weak and broad peaks.

According to literature survey, the borate spectra are

divided into following three regions [12-14].The regions

are;

-1(a) 600-800 cm for the B-O-B vibrations

-1(b) 800-1200 cm for BO groups4

-1(c) 1200-1600 cm for BO groups3

-1There is another band from 2300-4000 cm , which is

due to hydrogen bonding in OH group [15].-11. The band centered at 699 cm has been assigned to B-O-

B bending vibration of BO and [BO ] groups [16]. Its 3 4

Intensity increases with the increase in contents of

aluminum contents, which is due to presence of [AlO ] 6

group of aluminium in glass network [16].-12. In sample A1, the band observed at 1024 cm is due to

B-O bond stretching of [BO ] groups [15]. 4

3. This band is shifting towards the lower wave number -1(from 1024 to 981 cm ) side in sample A5 with the

increase in the percentage of Al O . Also, its intensity 2 3

increases with the increase in contents of Al O , which is 2 3

due to increase in tetrahedral [BO ] groups in the borate 4

network [15].

Glass Code

Al2O3 Li2CO3

B2O3

Band Gap

A1 0 30 70

2.91

A2 2 28 70 2.95A3 4 26 70 3.00A4 6 24 70 3.08A5 8 22 70 3.12

10 20 30 40 50 60

Sample-A5

Sample-A3

Sample-A1

Inte

ns

ity

[a.u

.]

2?[degree]

Fig. 1 XRD of glass samplesFig. 2 FTIR spectra of Al O -Li CO –B O2 3 2 3 2 3

glass with varying concentration of Al O2 3

500 1000 1500 2000 2500 3000 3500

Tra

nsm

ittance

[a.u

.]

Wave Number (cm-1)

981

997

997

1024

1024

679

679

679

692

699

1376

1349

1382

1349

1349

A5

A4

A3

A2

A1


Structural and Optical Investigation of Aluminium-Lithium-Borate Glasses

4. As the concentration of Al O increases shifting of band 2 3

-1arises (from 1024 to 981 cm ) which is due to presence

of [AlO ] units of aluminium. This is attributed to 6

combined presence of aluminium [AlO ] group and 6

tetrahedral [BO ] groups of borate [16]. 4

-15. The band in the region 1200-1500 cm , centered at 1382 -1cm is due to B-O stretching of [BO ] groups in ortho and 3

meta-borate units [17].

C. Optical Band Gap½1. The plots between (áhí) and energy (hí) of glasses are

used to determine the optical band gap as shown in fig.3.

2. The optical band gap energy value E , decreases with opt

an increase of tungsten oxide and lithium oxide

contents.

3. It indicates that a compact structure is formed.

4. With the addition of aluminium at the expense of lithium

content, a large number of oxygen ions become

available in the glass network and changes it from

trigonal [BO ] to tetrahedral [BO ] which results in 3 4

compact the network [17].

5. The gradual increase in the concentration of aluminium

ions cause to increase in tetrahedral group [AlO ] units 6

[17].

6. Formation of [AlO ] has shifted the absorption edge to 6

the lower energy that leads to a significant shrinkage in

the band gap. This change in band gap shows that the

aluminum enters the glass structure as network

modifier [18].

2.0 2.5 3.0 3.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

A5

A4 A2

A3

A1

(?h

?)1

/2

Energy(eV)

Fig. 3 Optical band gap of Al O -Li CO –B O2 3 2 3 2 3

glasses with varying concentration of Al O2 3

IV. CONCLUSION

In conclusion, with the increasing contents of Al O 2 3

against the decreasing Li CO and fixing the B O leads to 2 3 2 3

the compaction of glass network due to the formation of

tetrahedral [BO ] units of borate and [AlO ] unit of 4 6

aluminium. In this way it has been observed that it decreases

the optical band gap energy. The FTIR study shows the

incorporation of [BO ], [BO ] and [AlO ] units as network 3 4 6

modifiers with B-O-B vibration in glasses network. It has

also been observed that Al O content helps in converting 2 3

[BO ] group to [BO ] units. This reveals that aluminium 3 4

ions also enter the glass structure as a network modifier.

REFERENCES

[1] R K Brow, D r Tallant and G L Turner, J. Amer Cerm. Soc., 80 (1997),

pp. 239.

[2] G. Pal Singh, Simranpreet Kaur, Parvinder Kaur, Sunil Kumar and

D.P. Singh, Physica B, 406 (2011), pp.1890.

[3] E.I.Kamitsos, A.P. Patsis and G.D.Chryssikos, J. Non-Cryst. Solids,

152 (1993), pp. 246.

[4] J F MacDowell , J. Amer Cerm. Soc., 73 (1990), pp. 2287.

[5] V V Golubkov and E A Porai –Koshits, Sov J. Glass Phys Chem., 17

(1991), pp. 458.

[6] Monika Arora, S. Baccaro, G. Sharma, D. Singh, K. S. Thind, D. P.

Singh, “Nuclear Instruments and Methods in Physics Research

Section B” Beam Interactions with Materials and Atoms 267 (2009),

pp. 817-820.

[7] A.B. Corradi, V. Cannillo, M. Montorsi, C. Siligardi, J. Mater. Sci. 41

(2006), pp. 1573.

[8] M.S. Reddy, G.N. Raju, G. Nagarjuna, N. Veeraiah, J. Alloys Compd,

pp. 438 (2007) 41.

[9] S.R. Elliott, Section 3, Physics of Amorphous Materials, Longman

Science and Technology, Essex (1984), pp. 53–133.

[10] K. Arai, H. Namikawa, K. Kumata, T. Honda, Y. Ishii, T. Handa, J.

Appl. Phys. 59 (1986), pp. 3430–3436.

[11] S. Tanabe, J. Non-Cryst. Solids, 259 (1999) pp.1–9.

[12] E.I. Kamitsos, M. A. Karakassides, G. D. Chryssikos, J. Phys. Chem.,

90 (19) (1986), pp. 4528.

[13] E.I.Kamitsos, A.P. Patsis, G.D.Chryssikos, J. Non-Cryst. Solids, 152

(1993), pp. 246-57.

[14] J. Krogh-Moe, J.Non-Cryst. Solids, 1 (1969) pp. 269.

[15] R.D Husung, R H Doremus, J. Mater Res. 5(10) (1990), pp. 2209-

2217.

[16] L.Stoch, M.Sroda, J. Mol. Struct.511-512 (1999), pp. 77.

[17] M.S. Gaafar, N.S Abd El-Aal, O.W. Gerges, G.El-Amir, J. Alloy

Compd. 475(1-2) (2009), pp. 535-542.

[18] G. Pal Singh, D.P. Singh, Physica B 406(3) (2011), pp. 640-644.

[19] G. Pal Singh and D.P. Singh, Physica B, 406 (2011), pp. 3402.


Gurinder Pal Singh, Parvinder Kaur,Simranpreet Kaur, Deepawali Arora and D.P. Singh

Performance Comparison of Single and Double Layer Microchannel Using Liquid Metal Coolants: A Numerical Study

1 * 2 2Deewakar Sharma , Harry Garg and P.P. Bajpai1,2 Central Scientific Instruments Organisation, Chandigarh -160 030, India

* Corresponding Author E-mail: [email protected]

Abstract - With increase in demand for new cooling solutions,

double layer configuration of microchannels has been

extensively studied. Recently liquid metals have also been

proposed to further improve cooling owing to their high

thermal conductivity. However, their advantages with double

layer system are yet to be explored. A comparative study is

made between single and double layer microchannel using

liquid metals (liquid gallium) as the cooling medium. The type

of configuration (counter or parallel) best suited is analysed

and the results are compared with single layer for four

different lengths. The cross-sectional area of single layer is

such that it has same flow area as that of double layer

microchannel. The performance of both is compared under

the conditions of same flow rate and pump power. It is judged

on the basis of maximum temperature attained and minimal

temperature variations at the heated surface. It is observed

that with liquid metal (gallium) as coolants, the double layer

arrangement doesn't prove much advantageous and better

results can be obtained using single layer. Results also favour

liquid metals for small lengths of microchannels showing their

favourability for miniaturized cooling systems.

Keywords: Conjugate heat Transfer, Double layer, Liquid

metals, Single layer

I. INTRODUCTION

With recent developments in field of microelectronics

and ever increasing demand for higher computational speed

and superior performance, power density levels have

increased several manifolds. The peak power consumption

in high performance desktop applications is expected to

touch the 198 W mark by 2015[1] and expected dissipation

of heat flux in next generation microprocessors and 2 microelectronic components is over 1000W/cm [2].

Consequently, dissipation of large amount of heat within a

small space has increased temperature levels. To ensure

consistent operating parameters and reliability of the

circuits, there is a need to maintain operating temperatures

within certain limits. The conventional cooling systems

such as air cooling, heat pipes, thermoelectric cooling etc.

are either incompatible with new microelectronic

components or seem to have reached their practical limit.

This has motivated researchers to come with several

solutions which have been summarized in , according to

their heat removal capacity. Among these, of the coolant

temperature at entrance and exit. As a result, large

temperature gradients prevail in substrate which over a

period degrade the performance and reduce the components

reliability. Several solutions have been proposed to

overcome this problem. The concept of flexible

microchannel heat sink, which utilized flexible soft seals to

enhance heat transfer, was proposed by Khaled and Vafai

and has been discussed in detail in their work [13, 14].

A significant amount of work has also been done

regarding flow pattern to reduce such gradients. The liquid

cooling using microchannel have caught most of the

attention owing to their several advantages such as their

direct integration on the substrate (electronic chip) which

can reduce thermal contact (internal) resistance almost to

zero. Moreover, reduced hydraulic diameters allow for

significantly high values of heat transfer coefficients, of the 3 2order 10 W/m .

Tuckerman and Pease were the pioneers who introduced

the concept of microchannel cooling. They demonstrated

that with microchannels 50 µm wide and 300 µm deep, very -6 2small thermal resistance (9x10 K/ (W/m )) is possible with

2power density of 790 W/cm . Following this, significant

contributions have come up in this field with investigations

related to hydraulic and thermal performance of

microchannels which can be classified as experimental and

numerical [5-8]. Several review articles have also been

published pertaining to geometrical, experimental and

numerical reviews[9-12]. Despite several advantages, heat

dissipation within small region results in significant

variation early work of Missaggia and Walpole presented

single layer counter flow heat sink in which flow of water

took in opposite directions in adjacent channels which

reduced on chip temperature variations. A novel concept of

double layer microchannel was proposed by Vafai and Zhu

[16] which mainly addressed the issue of reducing

temperature gradients and variations at the heated surface.

It consisted of two layers of heat sink in which flow of


coolant (water) in opposite directions aided in reducing

temperature variations at the heated surface. Following

this, substantial research has been done in this field to

further investigate and explore the advantages of such kind

of system. The concept of double layer microchannel has

been further extended to stacked microchannels[17,18] .

Majority of the work discussed so far was carried out

using water as cooling medium or similar high Prandtl

number fluids. Recently, use of liquid metals with low

melting point for cooling of high power density devices has

been proposed by Liu and Zhou [19]. All such possible

liquids have been reviewed in detail by Kunquan and Liu

[20] amongst which liquid gallium and its alloys were

considered most suitable owing to their overall superior

thermo-physical properties. These can be summarized as

high thermal conductivity, low melting point, non-toxic

nature, boiling point higher than 2200°C eliminating power

density as a limiting factor etc. The experimental study of

Miner and Ghoshal [21] showed that heat transfer 2coefficients of the order 10W/cm /K are achievable using

68 20 12gallium alloy,Ga In Sn . Further, experiments performed

by Li et al [22]. under different flow and heat dissipation

rates using liquid gallium showed that temperature drop in o ocase of liquid gallium was 46.7 C against 51.9 C with water.

The need for smaller radiator size was also observed when

using liquid gallium as cooling medium.

Whereas high thermal conductivity of liquid metals is

advantage on one hand, their low specific heat is a

disadvantage which may cause higher temperature

difference at fluid inlet and exit. The difference in thermo-

physical properties of liquid metals as compared to high

Prandtl number fluids has motivated the current study to

compare the performances of double layer and single layer

microchannels. Further, investigating the performance

dependence of double layer microchannel on type of fluid

was also a propelling factor for this work.

In this study, the performance of single and double layer

microchannel is compared using conjugate heat transfer

analysis using liquid gallium as the cooling medium. Since

the temperature gradients are expected to be in all the

directions, a three dimensional model is adopted in the

present work. To see if there is any effect of length on the

performance, four different lengths have been considered.

For sake of brevity SL, DLCF, DLPF is used throughout

representing single layer, double layer counter flow and

double layer parallel flow configurations respectively.

II. ANALYSIS

In this section, various aspects related to analysis are

described briefly. These include experimental units

(computa t iona l domain) , boundary condi t ions ,

assumptions etc.

A. Computational Model

Single and double layer microchannel heat sink (parallel

and counter configurations depicted using arrows) along

with the coordinate system is shown in Fig.1 (a-b)

respectively. These also represent computational domains

under study.

The cross section of the geometry used in this analysis is

same as used by Vafai and Zhu[16] . For comparing both the

systems, total flow area (inlet cross sectional area) of SL is

same as that of DL system. Only one half of both the type of

heat sinks has been included in the computational domain

owing to symmetry conditions. 'H ' and 'W ' represent ch,DL ch

the height and width of each channel of DL respectively.

'H ' and 'W' denote the total height and width of the total,DL

computational domain, respectively. 'W ' is the thickness of s

solid region while 'L , L , L and L ' represent four different 1 2 3 4

lengths considered for analysis. For SL system, 'H ' and ch,SL

'H ' denote channel height and computational height total,SL

respectively. Rest of the dimensions are same as that of DL

heat sink. Table 1 summarizes all the dimensions. These are

also represented pictorially in Fig. 2.


Deewakar Sharma, Harry Garg and P.P. Bajpai

Fig. 1 Three dimensional computational domains (a) Double Layer Microchannel (b) Single Layer Microchannel

The analysis is based on the following assumptions:

i. Steady state flow.

ii. Incompressible fluid.

iii. Laminar flow.

iv. Constant properties of both fluids and solid.

v. Effects of viscous dissipation are negligible.

TABLE I VARIOUS DIMENSIONS USED IN THE ANALYSIS

Based on the above assumptions the governing

equations of mass, momentum and energy as applied to the

fluid region were:

Continuity:

Momentum:

Energy:

Fig. 2 Various Dimensions Used in analysis

Variable Dimension (in µm)

Wch 30Ws 30W 60

Hch,DL 100Hch,SL 200

Htotal,DL

260Htotal,SL

230L1

2000L2

4000L3

6000

L4 8000

DUC

75

DLC

75DSL 92.3

where the variables prepresent fluid

velocity, viscosity, density and thermal diffusivity

respectively. 'P' and 'T' denote pressure and temperature

while the subscript 'f' denotes fluid. The following energy

equation was applied to solid region. Energy (for heat

transfer):

'T ' represents the temperature of solid region with s

subscript's' representing solid region.

Further, hydraulic diameters (required to calculate

Reynolds Number) are calculated as:

i. Double Layer system

ii. Single layer system

Here 'a' and 'S', are the area and perimeter of the channel

respectively while subscripts UC,LC and SL denote upper

and lower channel of DL systems and SL respectively. The

hydraulic diameters are also tabulated in Table I.

The Reynolds number is defined as

where 'w ' is mean flow velocity of the fluid while 'D' avg,f

denotes the hydraulic diameter of channel and other

subscripts have same meaning as described before.

B. Boundary Conditions

In any computational analysis, boundary conditions play

a significant role describing the type of analysis and nature

of equations that are being solved. The boundary conditions

as applied to computational domain in present study are 6 2shown in Fig. 3. Uniform heat flux, q'' (=10 W/m ) is

applied at the base, at y = 0 µm. The adiabatic conditions

were applied at the following faces:

i. Top surface as the heat sink cover is usually made of

poorly conducting material.

ii. The entrance and exit walls of the solid region

considering heat transfer due to fluid as dominant

factor.



iii. Outer wall of solid region owing to symmetry condition.

Uniform velocity and temperature conditions were

imposed at the inlet of both the systems. For DLCF, Z=0

represents inlet for lower channel and outflow for upper

channel while for DLPF it denotes former for both the

channels. For SL system, inlet conditions were imposed at

Z=0. Uniform pressure condition at the outlet was applied

in all cases. Continuity of temperature and heat flux as well

no slip condition was assumed at solid-liquid interface

while symmetry conditions were imposed on the plane X=0

in all the cases.

The solid region was assumed to be made of Silicon.

Table II lists all the material properties used in this present ostudy. Since melting point of gallium is 29.8 C (≈303 K)

[20], the inlet temperature was assumed to be 305 K and

thermo-physical properties of liquid gallium at 313 K were

used.

TABLE II PROPERTIES OF MATERIALS

C. Solution Method and Grid Independence

The continuity, momentum, and energy equations were

solved using general purpose finite volume based

commercial code, FLUENT. The standard scheme for

pressure discretization, SIMPLE algorithm for pressure

Fig. 3 Boundary Conditions

Double Layer Single Layer

MaterialProperty

Silicon Liquid Gallium

Density (kg-m-3)

2328 6088a

Specific Heat (J-kg-1K-1)

705 400b

Thermal Conductivity (W-m-1K-1)

150 29

Viscosity ( Ns-m-2)

- .000187c

a,c: from Ref. b: from Ref.

velocity coupling and the second order upwind scheme for

momentum and energy equations were used. For grid

independence, three grid sizes were tested separately for

each length using counter flow arrangement in case DL

system. In lieu of computational resources and time, further

refinement of grid was stopped when variation in results

upon further decrease in grid size was below 1%. Similar

procedure was adopted for SL system. For example, grid

size of '24x66x50' and '24x76x50' was used (in x x y x z

directions) for DL and SL system (L length).1

III. RESULTS AND DISCUSSION

The performance of liquid gallium in counter and

parallel arrangement for DL and SL is compared for each

length. The range of flow rate considered for comparison -8 3 -8 3varies from 0.72x10 m /s to 3.6x10 m /s, which

corresponds to Reynolds number 147 to 732 for DL and 180

to 902 for SL. It is to be noted that flow rate mentioned here

is the total flow rate in DL microchannel (i.e. sum of flow

rate in upper and lower channel). The following

abbreviations have also been used extensively for the sake

of simplicity, 'PF' for parallel flow, 'CF' for counter flow,

'UC' for upper channel 'LC' for lower channel and L , L , L 1 2 3,

and L representing lengths of the microchannel as 4

described in Table I. The term flow rate has been used

instead of total flow. In addition, where results follow same

trend for all lengths, results pertaining to only two of the

lengths have been shown.

A. Comparison Based on Maximum Temperature Attained

at the Heated Surface

In this section the performance of both the systems is

evaluated on the basis of maximum temperature attained at

the heated surface. This forms one of the basic criteria in any

comparison as it is always desired to keep maximum

temperature as low as possible.

1. Maximum Temperature Versus Flow Rate

Fig. 4 (a-b) shows maximum temperature reached at

different flow rates for DLCF, DLPF and SL system for

lengths L and L considered in the study. 1 3

It is observed that for both the lengths and at all flow

rates, maximum temperature reached in DLCF is much

higher as compared to DLPF and SL system. This behaviour

of liquid gallium may be attributed to its low specific heat

which results in significant rise of temperature of liquid

gallium in DLCF. This is attributed to heat flow at inlet of LC

not only from base but also from UC and vice-versa in case



Here, 'T ' is maximum temperature attained at base and max

this relation is defined for same flow rate. Similar relations

hold for parallel configuration.

The results shown are normalized (w.r.t absolute value)

for each category. It is to be noted that since, similar pattern

was observed in maximum temperature versus flow rate for

all four lengths, only one length is considered here for sake

of brevity. However the results for remaining lengths have

been tabulated in Table III. It is interesting to note that

superior performance between DLPF and SL system is flow

rate dependent. This is evident from negative values at low

flow rates which shows lower temperature attained in case

of DLPF. The same trend is followed for all four lengths as

can be seen from Table III. It can be deduced from the Fig. 5

and Table III that at low flow rates, performance of DLPF is

superior whereas with increase in flow rate, SL heat sink

system is more suitable.

Fig. 5 Difference between Maximum temperature attained in DLCF and DLPF w.r.t SL system

(Normalized) versus Flow Rate (L length)2

of UC which doesn't happen in case of DLPF. This prevents

significant rise in temperature of coolant (liquid gallium). It

can also be seen that the difference between maximum

temperature attained at low and high flow rate is quite

significant for counter arrangement in comparison to DLPF

and SL configurations, especially for longer length. In other

words, rate of decrease of maximum temperature attained at

the base with flow rate is significantly higher for DLCF

whereas it tends to remain nearly same for other two

configurations under study. This is true for all the lengths

considered. This means the performance of DLCF is

significantly affected by flow rate. This may be explained as

follows. The temperature attained by liquid gallium at exit

decreases with increasing flow rates. This means flow of

heat from UC to LC and vice versa is lower thereby allowing

for more efficient heat transfer and distribution.

For further analysis of the results, difference between

maximum temperature attained for DLPF and DLCF w.r.t

SL system is depicted in Fig. 5 for L length. It is defined as:2

Fig. 4 Maximum Temperature reached at the base for all three configurations i.e.

SL,DLCF,DLPF for the following lengths (a) L (b) L1 3

TABLE III NORMALIZED TEMPERATURE DIFFERENCE OF

DL SYSTEM CONFIGURATIONS W.R.T SL SYSTEM FOR L L AND L1, 3 4

Flow Rate x

103

[ml/h]

L1 L3 L4

ÄT

DLCF-SL

[K]

ÄT

DLPF-

SL

[K]

ÄT

DLCF-

SL

[K]

ÄT

DLPF-

SL

[K]

ÄT

DLCF-

SL

[K]

ÄT

DLPF-SL

[K]

0.6 1 -1

1

-1

1 -1

1.2 0.70

0.73

0.62

0.66

0.63 0.671.8 0.41 0.81 0.37 0.70 0.38 0.662.4 0.25 0.78 0.23 0.68 0.23 0.643.0 0.16 0.76 0.15 0.62 0.15 0.59



Another important observation to note is that there is

first increase in normalized temperature difference

followed by its decrease at all the lengths. This shows that

after a certain flow rate, even though the performance of

single layer is superior, but its rate decreases beyond a

certain flow rate. However, this should not be of much

concern because at higher flow rates, the flow regime may

become turbulent requiring different analysis.

B. Thermal Resistance Versus Pump Power

In practical approach, the performance of heat sink is

limited by flow rate which depends on available pumping

power. Further, thermal resistance addresses the

performance more appropriately as it accounts for both heat

flux and maximum temperature rise. Hence, for further

understanding the phenomena of cooling using liquid

gallium (metals) in DL and SL systems and making the

analysis more relevant to practical systems, thermal

resistance versus pump power is shown in Fig. 6 (a-b) for

lengths L and L .2 4

We define, thermal resistance as:

Similarly pump power can be defined as:

Here ' P' represents pressure drop while 'Q ' represents F

the total flow rate. For sake of transience, only results

pertaining to DLPF are evaluated owing to its superior

performance as shown in previous section. Further, results

are shown only for L and L length. As seen from Fig.6 2 4

(a-b), SL offers lower thermal resistance at same pump

power for both the lengths.

This may be attributed to additional conduction

resistance due to base of upper channel. The base behaves

like a fin which has certain effectiveness which may reduce

overall effective area thereby increasing convective

resistance. Since the overall thermal resistance, in addition

to bulk fluid resistance, is affected by combination

convective and conduction resistances, dominance of latter

may explain the observed phenomena.

C. Comparison on the Basis of Minimum Temperature

Variations

The main idea of DL system as proposed by Vafai and

Zhu[16] was to minimize temperature variations along the

flow direction. This prevents thermal stresses to

accumulate over a period which may otherwise lead to

failure of the component. This aspect is also important for

higher reliability of the system. Hence it is important to

analyse whether such a system holds any advantage while

using liquid gallium (metals).

1. Minimum Temperature Variations Versus Flow Rate

Fig. 7 shows flow rate versus temperature variations for

L length for DLCF, DLPF and SL. For remaining lengths, 2

these are listed in Table IV.The temperature variation is

defined as:

It is observed that performance of DLPF is better at low

flow rate in terms of minimal temperature variations at the

base of heated surface. However, with increase in flow rate,

Fig. 6 Thermal Resistance versus Pump Power for lengths: (a) L (b) L2 4



SL shows superior results as compared to DLPF at same

flow rate conditions as can be deduced from Table IV.

Moreover, there is increase in minimum flow rate beyond

which performance of SL is superior with length as can be

observed from Table IV. For example, at length L1, DLPF -8 3performs better up to flow rate 0.72x10 m /s whereas this

-8 3increases to 1.44x10 m /s for the remaining lengths. It is to

be noted that the limits described here are due to values of

flow rates considered in this study. For exact limits, more

values of flow rates need to be considered.

In continuation of results from previous sections, it can

be seen that the performance of SL is superior both in terms

of maximum temperature attained at the base as well as

minimum temperature variations at the base at higher flow

rates.The difference being very minimal, which may also be

attributed to computational limit or error, creates the need to

investigate the performance of DLPF and SL further i.e. on

the basis of pump power.

2. Pump Power Versus Minimum Temperature Variations

As explained above, pumping power is one of the major

constraints in the area of microchannel applications. The

use of DL system was proposed to achieve minimum

temperature variations as compared to SL systems at the

cost of low pump power. This was possible in the studies in

Vafai and Zhu[16] and others due to high specific heat of

water which showed positive results for such a design,

especially in counter flow arrangement. However, for

liquid gallium (metals) the case may not be true as can be

deduced from above sections. Further insight in this regard

is found by considering Fig. 8 which depicts temperature

variations against pump power for DLPF and SL. For sake

of brevity only results corresponding to L and L length 2 4

have been shown while those of DLCF have been omitted

owing to its inferior performance.

Fig. 7 Tempertaure variations at heated surface (base) for Maximum Temperature reached at the base for L length.2

TABLE IV TEMPERATURE VARIATION (ÄT [K]) AT HEATED SURFACE FOR L , L AND L1 3 4

Length

Flow

Rate

x 103

[ml-h-1]

L1 (ÄT [K])

DLCF DLPF SL

25.92 15.54 12.14 12.2751.84 10.10 6.56 6.5577.76 6.82 4.60 4.58

103.68 4.98 3.62 3.59129.6 3.89 3.03 3.00

Length

FlowRate

x 103

[ml-h-1]

L3 (ÄT [K])

DLCF DLPF SL

25.92

131.63 39.51 39.65

51.84

80.84 20.24 20.24

77.76 50.16 13.72 13.71

103.68

33.47 10.46 10.44

129.6 23.92 8.51 8.48

Length

Flow

Rate

x 103

[ml-h-1]

L4 (ÄT [K])

DLCF DLPF SL

25.92

226.29 53.20 53.3451.84 141.57 27.08 27.0877.76 87.09 18.29 18.27

103.68 57.48 13.88 13.86129.6 40.60 11.24 11.22

Fig. 8 Pump Power versus Temperature variations at base for lengths: (a) L (b) L2 4



It is observed that SL channel is suitable in terms of

temperature uniformity at the base. Even though at low flow

rate conditions, performance of DLPF was better under same

evaluation criteria, lower hydraulic diameter in case of DL

system deteriorated its overall performance. Hence it can be

deduced that use of DL system may not prove advantageous

in case of liquid metals and creates a need to for an alternate

design. This also suggests while using liquid gallium (or

metals) as coolants, the complications and reliability aspects

practically linked with DL system will not be encountered.

This further supports use of liquid gallium (metals) as

cooling medium for compact systems as suggested in the

studies of Li [22].

IV. CONCLUSION

Three dimensional conjugate heat transfer analysis is

carried out to compare the performance of single and

double layer microchannel with liquid gallium as cooling

medium. Owing to different thermo-physical properties of

liquid metals especially high conductivity and low specific

heat, the performance of single layer microchannel system

is found to be superior as compared to double layer system.

Under the condition of same flow rate, single layer

configuration was favoured under both the evaluation

criteria i.e. maximum temperature at the base as well as

minimum temperature variations beyond certain flow rate

for all the lengths considered in the study. However,

comparison on the basis of same pump power showed that

the overall performance of single layer is superior in all

aspects. Since pump power is more realistic approach, it can

be concluded that under the conditions of same flow rate in

both the channels of DL system and equivalent total flow

rate in SL system, single layer heat sink is suitable while

using liquid gallium (metals) as cooling medium. This also

suggests that liquid gallium (or metals) is more suited for

compact and reliable systems.

ACKNOWLEDGEMENT

The authors are thankful to Dr. Pawan Kapur, Director

CSIR-CSIO, for supporting their work. The authors also

express their gratitude to Mr. Vinod Karar (CSIR-CSIO)

and Mr. Sumer (IIT-Kanpur) for their valuable inputs and

suggestions.

Nomenclature

a Cross sectional area of microchannel

D Hydraulic diameter (µm)

DLCF Double layer counter flow

DLPF Double layer parallel flow

H Height of each channel in double layer ch,DL

microchannel (µm)

H Height of channel in single layer microchannel ch,SL

(µm)

H Total height of computational domain in case of total,DL

double layer microchannel (µm)

H Total height of computational domain in case of total,SL

single layer microchannel (µm)

L Length of microchannel (µm)

LC Lower channel

P Pressure (Pa)

PP Pump power (mW)

q'' Heat flux (W/m2)

Q Flow RateF

R Thermal Resistance (K/(W/m2))th

Re Reynolds Number

S Cross sectional perimeter (µm)

SL Single layer (microchannel)

T Temperature (K)

UC Upper channel

w Mean flow velocity in z directionavg,f

W Total width of computational domain in double and single layer microchannel systems (µm)

W Width of solid region (substrate) (µm)s

W Width of microchannel (µm)ch

x,y,z coordinate system (µm)

Subscripts

base base of microchannel (y= 0 µm)

ch channel

ch,DL channel of double layer system

ch,SL channel of single layer system

f fluid

LC lower channel

max maximum

min minimum

s solid region

UC upper channel

1,2,3,4 lengths of microchannel (2000, 4000, 6000 and

8000 µm respectively)

Greek Letters

á thermal diffusivity

ñ density

µ dynamic viscosity

Ä delta (change in Pressure)



REFERENCES

[1] "2007 International Technology Roadmap for Semiconductors,

Executive Summary," pp. 82-83.

[2] L. Chai, G. Xia, M. Zhou, and J. Li, "Numerical simulation of fluid

flow and heat transfer in a microchannel heat sink with offset fan-

shaped reentrant cavi t ies in s idewal l ," Internat ional

Communications in Heat and Mass Transfer, vol. 38, pp. 577-584.

[3] X. C. Tong (2011) Advanced Materials for Thermal Management of

Electronic Packaging, 1st ed. New York: Springer Science +

Buisness Media

[4] D. B. Tuckerman and R. F. W. Pease (1981), "High-performance

heat sinking for VLSI," Electron Device Letters, IEEE, vol. 2, pp.

126-129.

[5] B. X. Wang and X. F. Peng (1994), "Experimental investigation on

liquid forced-convection heat transfer through microchannels,"

International Journal of Heat and Mass Transfer, vol. 37,

Supplement 1, pp. 73-82.

[6] J.-Y. Jung, H.-S. Oh, and H.-Y. Kwak (2009), "Forced convective

heat transfer of nanofluids in microchannels," International Journal

of Heat and Mass Transfer, vol. 52, pp. 466-472.

[7] P.-S. Lee and S. V. Garimella (2006), "Thermally developing flow

and heat transfer in rectangular microchannels of different aspect

ratios," International Journal of Heat and Mass Transfer, vol. 49,

pp. 3060-3067.

[8] T.-Y. Lin and S. G. Kandlikar (2012), "A Theoretical Model for

Axial Heat Conduction Effects During Single-Phase Flow in

Microchannels," Journal of Heat Transfer, vol. 134, pp. 020902(1-

6).

[9] J. P. McHale and S. V. Garimella, "Heat transfer in trapezoidal

microchannels of various aspect ratios," International Journal of

Heat and Mass Transfer, vol. 53, pp. 365-375.

[10] P.-S. Lee and S. V. Garimella (2006), "Thermally developing flow

and heat transfer in rectangular microchannels of different aspect

ratios," International Journal of Heat and Mass Transfer, vol. 49,

pp. 3060-3067.

[11] G. Hetsroni, A. Mosyak, E. Pogrebnyak, and L. P. Yarin (2005),

"Fluid flow in micro-channels," International Journal of Heat and

Mass Transfer, vol. 48, pp. 1982-1998.

[12] G. Hetsroni, A. Mosyak, E. Pogrebnyak, and L. P. Yarin (2005),

"Heat transfer in micro-channels: Comparison of experiments with

theory and numerical results," International Journal of Heat and


[13] A.-R.A. Khaled and K. Vafai (2004), "Control of exit flow and

thermal conditions using two-layered thin films supported by

flexible complex seals," Int. J. Heat Mass Transfer, vol. 47, pp. 1599-

1611.

[14] A.-R. A. Khaled and K. Vafai (2002), "Flow and heat transfer inside

thin films supported by soft seals in the presence of internal and

external pressure pulsations," International Journal of Heat and


[15] L.J. Missaggia and J. N. Walpole (1991), "A microchannel heat sink

with alternating directions of water flow in adjacent channels,"

Integrated Optoelectronics for Communication and Processing, pp.

106–111.

[16] K. Vafai and L. Zhu (1999), "Analysis of two-layered micro-channel

heat sink concept in electronic cooling," International Journal of

Heat and Mass Transfer, vol. 42, pp. 2287-2297.

[17] X. Wei, Y. Joshi, and M. K. Patterson (2007), "Experimental and

Numerical Study of a Stacked Microchannel Heat Sink for Liquid

Cooling of Microelectronic Devices," Journal of Heat Transfer, vol.

129, pp. 1432-1444.

[18] X. Wei and Y. Joshi (2004), "Stacked Microchannel Heat Sinks for

Liquid Cooling of Microelectronic Components," Journal of

Electronic Packaging, vol. 126, pp. 60-66.

[19] J Liu and Y. X. Zhou (2002) China Patent No. 02131419.5.

[20] M. A. Kunquan and J. Liu (2007), "Liquid metal cooling in thermal

management of computer chips," Frontiers of Energy and Power

Engineering in China, vol. 1, pp. 384-402.

[21] A. Miner and U. Ghoshal (2004), "Cooling of high-power-density

microdevices using liquid metal coolants," Applied Physics Letters,

vol. 85 pp. 506-508.

[22] T. Li, Y.-G. Lv, J. Liu, and Y.-X. Zhou (2005), "A powerful way of

cooling computer chip using liquid metal with low melting point as

the cooling fluid," Forschung im Ingenieurwesen, vol. 70, pp. 243-

251.

[23] K. E. Spells (1936), "The determination of the viscosity of liquid

gallium over an extended range of temperature," Proceedings of the

Physical Society, vol. 48, pp. 299-311.

[24] G. B. Adams, H. L. Johnston, and E. C. Kerr (1952), "The Heat

Capacity of Gallium from 15 to 320°K. The Heat of Fusion at the

Melting Point," Journal of the American Chemical Society, vol. 74,

pp. 4784-4787.



The Thermodynamic Study of Turbocharger Pressure Ratio and Ambient Temperature Variation on Exergy Destruction

Estimation of Homogeneous Charge Compression Ignition Engine Cogeneration System

1* 2Shailesh Kumar Trivedi and Abid Haleem1* Research Scholar, Department of Mechanical Engineering

Faculty of Engineering and Technology, Jamia Millia Islamia, New Delhi- 110 0252Professor and Head, Department of Mechanical Engineering

Faculty of Engineering and Technology, Jamia Millia Islamia, New Delhi - 110 025E-mail of corresponding author: [email protected]

Abstract - Homogeneous Charge Compression Ignition

(HCCI) technology is different from conventional combustion

technologies. It has the combination of lean and premixed fuel

air mixture and charge is compression ignited so it has multiple

ignition points throughout the combustion chamber thus

eliminating the high peak temperature inside the combustion

chamber. This new engine technology is helpful in production

of ultra low NO and particulate emissions. The use of lean and X

unthrottled operation yields higher efficiency and better fuel

economy also. In this paper, a new HCCI engine combined

cycle cogeneration system is proposed and studied. The system

is equipped with turbocharger, fuel vaporizer, engine,

catalytic converter, different components of Organic Rankine

cycle (ORC) and further heat recovery steam generator

(HRSG) for waste heat utilization from the exhaust gases to

obtain process heat. An exergy analysis is applied to the

different components of HCCI engine cogeneration system to

examine the thermodynamic losses in terms of exergy

destruction and the effect of ambient temperature and

turbocharger pressure ratio is obtained. The result shows a

ranking among the components of the system on the basis of

thermodynamic losses or the exergy destruction. This paper

shows how an exergy method can yield effect of ambient

conditions and design parameters values to reduce losses in

various components of HCCI engine cogeneration system.

Keywords: Cogeneration, Exergy analysis, Exergy destruction,

HCCI engine, Wet ethanol

I. INTRODUCTION

IC Engines have played a key role, both socially and

economically, in shaping the modern world. However in

recent decades, serious concerns have been raised with

regard to the environmental impact of exhaust and

particulate emission arising from operation of these

engines. In addition, concerns about world's finite oil

reserves have lead to rising fuel prices. These two factors

lead to massive pressure on vehicle manufacturer for

Research and development (R&D) to produce ever cleaner

and more efficient vehicles. R&D efforts always focus on

improving engine efficiency while meeting future national

and state emissions regulations through a combination of

combustion technologies that minimize in-cylinder

formation of emissions. In an effort to combine the benefits

of both SI and CI engines, homogeneous charge

compression ignition (HCCI) engines are being developed.

The start of ignition in HCCI engines is not directly initiated

by an external event such as the firing of a spark plug in SI

engines or the beginning of fuel injection in standard diesel

engines; instead HCCI relies solely on the fuel auto ignition

process to control the combustion [1].

II. RECENT AUTOMOTIVE TECHNOLOGIES

The ultimate intention of emission legislation to drive

technologies to the position where realistic reasonably

priced in close proximity to zero emission with satisfactory

performance becomes a reality. Recent progresses in

conventional SI and CI engine technology have allowed

huge improvement in emission and fuel consumption. The

adoption of three way catalytic converter in SI gasoline

engine has considerably reduced the emission of carbon

monoxide (CO), unburned hydrocarbon (HC), and oxides of

nitrogen (NO ). High speed direct injection (HSDI) diesel X

engines and stratified charge gasoline direct injection

(GDI) engines permit lean combustion by managing fuel

flow rate. Therefore these approaches achieve significant

reduction in fuel consumption. Alternative Technologies

such as fuel cells and electric vehicles that have been

introduced in the market come with associated problems.

These include high cost, changes required to the fuelling

infrastructure and lack of development to support these

technologies.


An alternative combustion technology commonly

known as homogeneous charge compression ignition has

emerged and it has the potential to achieve high efficiency

and very negligible NO and virtually no smoke emissions. X

It has the abilities to meet current and future emission

legislation, without the need for expensive exhaust gas

treatment systems. In fact HCCI combustion is a new

combustion process in reciprocating internal combustion

engines. In 1979, the most recognized original work in lean

combustion process for IC engine i.e. HCCI was reported

by Onishi et al [2]. They discussed HCCI combustion and

called it active thermo atmospheric combustion (ATAC).

They applied it on 2 stroke gasoline engine with lean

mixture at part load operation and consequently achieved

improved fuel economy and reduced exhaust emissions

along with lower noise and vibration. Noguchi et al (1979)

[3] Studied a gasoline engine combustion by observation of

intermediate reactive products during combustion. They

observed that the air fuel mixture burns in the reaction zone

with flame front as it propagates across the combustion

chamber and it has a clear separation between burned

charge and unburned charge where as in the case of

homogeneous charge compression ignition (HCCI) engine,

all the charge is consumed simultaneously as the charge

auto ignites and it has multiple ignition points. However

this combustion process is at a lower rate. These works

were motivated by their desire to control the irregular

combustion caused by auto ignition of cylinder charge to

obtain stable lean burn combustion of 2- stroke gasoline

engines. Thring R.H. (1989) [1] introduced the

terminology homogeneous charge compression ignition

(HCCI) for this type of combustion process and it was

further adopted by many researcher of present time to

describe such combustion process in both gasoline and

diesel engine. He suggested that the passenger car engine

can run on HCCI mode at idle and light load operation to

obtain fuel economy and smooth operation while switching

to conventional gasoline engine operation at full power for

good specific power output. Olsson and Johnsson (2001)

[4] used a modified 12 liter six cylinder, turbo diesel engine

mainly used in truck application for study of HCCI engine.

They achieved HCCI combustion over a large speed and

load range by employing combination of isooctane and

heptane through a close loop control, as well as turbo

charging, high compression ratio and intake air heating.

The technology of HCCI is attractive as there is no need for

huge modifications to the existing hardware of IC engines

and its fuelling system and it further considerably reduces

NOx emissions.

A. Ethanol in HCCI Engine

Recently a lot of research is being carried out for the use

of alternate fuels in HCCI engine. Mach et al (2009) [5]

investigated 4-cylinder 1.9 liter engine running in HCCI

mode fuelled with wet ethanol. They investigated the effect

of ethanol water fraction on the engine's operating limits,

exhaust emission, intake temperature and heat release rates.

Saisirirat et al (2011) [6] investigated the auto ignition and

combustion characteristics in HCCI using ethanol/n-

heptane mixture with varying alcohol percentage up to 57%

by volume. Diesel engine fuelled with alcohol/n-heptane

blend was used at constant equivalence ratio of 0.3, with 0intake temperature at 80 C operating at 1500 rpm. Wu et al

(2011) [7] investigated the reduction in smoke and NO of a X

partial HCCI engine using premixed gasoline and ethanol as

a fuel. The experiments were conducted under different

engine speed of 1200, 1500 and 1800 rpm and at different

loads. The result shows the successful operation of HCCI

engine with ethanol resulting good efficiency and

effectively reduced emissions.

Exergy analysis of wet ethanol fuelled HCCI engine for

cogeneration application is very much missing in the

literature. Therefore in order to meet out the simultaneous

demand of power and thermal energy from a sustainable

fuel in efficient and environment friendly manner, an

exergy analysis of wet ethanol fuelled engine in HCCI mode

for cogeneration of power and heat has been carried out in

this research paper. Magnitude of exergy destruction in

various components of the cogeneration cycle has been

evaluated and discussed.

III. SYSTEM DESCRIPTION

A schematic diagram of the wet ethanol operated HCCI

engine cogeneration system is shown in Figure 1. This

schematic is adopted from Frias et al [8] and modified for

cogeneration application. Ambient air enters the

compressor which delivers air at high pressure and

temperature followed by the regenerator, this raises the air

temperature. Next liquid ethanol in water is injected into the

vaporizer, where it evaporates and mixes with air. The

evaporation process in the vaporizer produces a

homogeneous mixture of ethanol, water vapor and air,

which then enters the HCCI engine. The ethanol water air

mixture inducted into the cylinder heats up as it mixes with

residual gases within the cylinder. After combustion,

exhaust gases enter the catalytic converter at a higher

temperature and exit the converter at further higher

temperature due to heat release from conversion of


The Thermodynamic Study of Turbocharger Pressure Ratio and Ambient Temperature Variation on Exergy Destruction Estimation of Homogeneous Charge Compression Ignition Engine Cogeneration System

unburned fuel, hydrocarbon (HC) and carbon monoxide

(CO) which were not reacted in the engine combustion

chamber. Gases from catalytic converter at higher

temperature flow into the turbine, generating power that

drives the turbocharger compressor. After circulating

through turbine, the exhaust gases exchange heat with the

intake air in the regenerator and then leave the regenerator

system at ambient pressure and higher temperature. These

exhaust gases at higher temperature are assumed to be

routed through the evaporator where heat transfer occurs

between the exhaust stream and the organic working fluid

(R113). In this study a counter flow heat exchanger

(evaporator) configuration is considered to maximize heat

transfer between the engine exhaust gases and the organic

fluid. Thermodynamically this is preferred configuration

because the temperature difference between the hot fluid

and the cold fluid is minimized, thereby reducing the exergy

destruction.

The heated organic vapor is then expanded in the

turbine, heat is rejected to the ambient in the condenser, and

the cooled working fluid is pumped with pump, back in to

the evaporator. Heat of hot exhaust gas is utilized in the heat

recovery steam generator to generate steam and hence to

produce the process heat.

12

Steam to process

15

16

Compressor

6

Air 1

Air

Air

Fuel vaporizer

Wet

ethanol

HCCI

engine

Turbine

Catalytic converter

2

3

5, Air, steam, fuel

4

7

8

Exhaust

Regenerator

HW

Pump

11

13

14

Organic

Rankine Cycle

Turbine

Evaporator

9

CW

Condenser

10

17

Process

R113 Vapor

Condensate

return

Heat

recovery

steam

generator

Fig. 1 The schematic diagram of wet ethanol operated HCCI engine cogeneration system with organic Rankine cycle and process steam. (Adopted from Frias et al [8] and modified for cogeneration application)


Shailesh Kumar Trivedi and Abid Haleem

A. Exergy Destruction Model

Development of exergy destruction model shows the

analyst how the performance of a system departs from the

ideal limit and to what extent each component contributes to

this departure, and what can be done to design better (less

irreversible) systems. The general exergetic balance

applied to a fixed control volume is given by Moran and

Shapiro [9]

[1]

where is the heat transfer rate to the system, W the

mechanical power produced by the system, the

irreversibility rate or exergy destruction, and is the flow

exergy associated with the stream of matter.

[2]

where h & s represent the specific enthalpy and 0 0

specific entropy at dead state respectively.

01...

0.

=−−+−

− ∑∑∑ D

outoutout

ininin

j

j EememWT

TQ

jQ.

DEe

))]()(([ 000 ssThhme iii

i −−−=∑

TABLE I EFFECT OF VARIATION OF AMBIENT TEMPERATURE ON EXERGY DESTRUCTION

IN DIFFERENT COMPONENTS OF THE WET ETHANOL OPERATED HCCI ENGINE COGENERATION SYSTEM FOR PR=3, ÇC=80% Å =79%

TABLE II EFFECT OF VARIATION OF TURBOCHARGER PRESSURE RATIO ON EXERGY DESTRUCTION IN DIFFERENT COMPONENTS OF THE WET ETHANOL OPERATED HCCI ENGINE COGENERATION SYSTEM FOR T0=300 K, ÇC=80%, Å =79%

IV. RESULTS AND DISCUSSION

Cogeneration is applied to the wet ethanol operated

HCCI engine system to enhance the system overall

efficiency and to reduce the emissions. The exergy

destruction or thermodynamic losses in each component,

and the exergy efficiency of the cogeneration cycle have

also been investigated under the exergy balance approach.

The properties of ethanol, organic fluid (R113) and related

details are taken from Heywood [10] and Perry's chemical

engineers hand book [11]. Processes in the engine are

typical polytropic compression and expansion and near

constant volume combustion as Otto cycle [Osborne et al

(2003)] [12]. Equations used for evaluating performance

parameters have been referred from Trivedi et al (2010) [13]

and Khaliq et al (2011) [14]. Fuel used is 35% ethanol in

water mixture which improves the energy balance of

ethanol production and it can efficiently run HCCI engine.

[Frias et al (2007)] [8].

Amb. Temp.

T0

(K)

ED,

Turbo.

comp.

(kJ/kg)

ED,

Regenator

(kJ/kg)

ED,

Fuel Vap.

(kJ/kg)

ED,

HCCI Engine

(kJ/kg)

ED,Cat.

Conv.

(kJ/kg)

ED,

Turbo

Turbine

(kJ/kg)

ED, ORC

HRSG

(kJ/kg)

ED, ORC

Turbine

(kJ/kg)

ED,

Condenser

(kJ/kg)

ED, ORC

Pump

(kJ/kg)

ED,

Cogen.

HRSG

(kJ/kg)

ED,

Exergy

lost to

Env.

(kJ/kg) 290 23.540 75.732 67.018 2815.001 126.849 19.719 20.122 3.422 15.130 0.421 17.408 30.788 295 24.022 74.037 66.006 2816.315 127.341 19.984 21.739 3.747 14.189 0.461 17.708 28.858 300 24.509 72.409 65.003 2817.591 127.673 20.242 23.460 4.088 12.886 0.503 18.008 27.028 305 25.003 70.528 64.029 2818.818 127.796 20.490 25.425 4.472 11.517 0.551 18.308 25.298 310 25.503 68.957 62.984 2820.056 128.041 20.743 27.328 4.836 9.491 0.596 18.609 23.666

Pr. Ratio

Pr

ED,

Turbo.

comp.

(kJ/kg)

ED,

Regenator

(kJ/kg)

ED,

Fuel Vap.

(kJ/kg)

ED,

HCCI Engine

(kJ/kg)

ED,Cat.

Conv.

(kJ/kg)

ED,

Turbo

Turbine

(kJ/kg)

ED, ORC

HRSG

(kJ/kg)

ED, ORC

Turbine

(kJ/kg)

ED,

Condenser

(kJ/kg)

ED, ORC

Pump

(kJ/kg)

ED,

Cogen.

HRSG

(kJ/kg)

ED,

Exergy

lost to

Env.

(kJ/kg) 2.50 20.235 89.504 76.701 2814.00 128.144 14.652 22.663 3.926 12.373 0.483 18.008 27.028 2.75 22.454 80.246 70.546 2815.884 127.805 17.526 23.038 4.002 12.615 0.493 18.008 27.028 3.00 24.509 72.409 65.003 2817.591 127.673 20.242 23.460 4.088 12.886 0.503 18.008 27.028 3.25 26.428 65.580 60.095 2819.070 127.273 22.797 24.088 4.215 13.285 0.519 18.008 27.028 3.50 28.230 59.105 55.595 2820.433 127.026 25.229 24.932 4.383 13.816 0.540 18.008 27.028



Table I & II reveals the exergy destruction in each

component of wet ethanol operated HCCI engine

cogeneration system. This exergy study shows that the

maximum exergy is destroyed in the HCCI engine which is

2817.591kJ/kg at mean operating conditions. Exergy

destruction in Heat transfer processes e.g. regenerator and

fuel vaporizer accounts for about 72.409kJ/kg and

65.003kJ/kg respectively at mean conditions. The exergy

destruction in catalytic converter is 127.673kJ/kg. The

exergy destruction in HCCI engine and catalytic converter

is high because the effect of chemical exergy in these

components predominates over the effect of physical

Figure 2 clearly indicates that cogeneration system has a

good thermal performance with first and second law

efficiencies of 46.47% and 38.5% respectively for the mean

operating conditions of T =300K, Pr=3, ç =80%. Thus the 0 C

recovery of waste heat is considerably increasing the

system efficiency. That is why various engineering

applications throughout the world are considering

cogeneration system for improving efficiency.

V. CONCLUSION

In this article, the thermodynamic analysis of the wet

ethanol operated HCCI engine cogeneration system is

performed. The exergy analysis is aimed to evaluate the

exergy destruction in each component as well as the

exergetic efficiencies. The fuel used is 35% ethanol in water

mixture and this blend is directly formed in the process of

ethanol production from biomass. This study further

explores the use of wet ethanol as a fuel for HCCI engines

exergy. Exergy destruction in ORC evaporator, ORC

turbine, ORC condenser, pump and cogeneration heat

recovery steam generator (HRSG) is very less compared to

main HCCI engine components. It indicates that the exergy

analysis is providing ranking among the components of the

system. The component with higher exergy destruction is

very much responsible to deteriorate the performance of the

system as compared to the components with lower exergy

destruction. It further indicates that which component

needs to be repaired or serviced first for maintenance

purpose.

Figure 2. First and second law efficiency of wet ethanol based HCCI engine

cogeneration system at mean operating conditions

0

5

10

15

20

25

30

35

40

45

50

Cogeneration system First law efficiency Cogeneration system second law efficiency

Firs

t and

sec

ond

law

eff

icie

ncie

s (%

)

while using exhaust heat recovery to provide the high input

energy required for igniting wet ethanol. The heat

exchanger (regenerator) was used to preheat the intake air

allowing HCCI combustion without electrical air heating.

The thermal efficiency of the overall plant is found to be

46.47% and the exergetic efficiency is 38.5%. The results of

this study show that HCCI engines can use ethanol fuels

with 35% ethanol in water mixture while maintaining

favorable operating conditions. This can remove the need

for the most energy-intensive portion of the water removal

processes i.e. distillation and dehydration of wet ethanol

during ethanol production.

The main conclusions from the current study can be

summarized as follows:

1. HCCI combustion process is highly different from

combustion process of SI and CI engines as it lacks

flame propagation therefore it has superior potential for


Shailesh Kumar Trivedi and Abid Haleem

achieving high thermal efficiency compared to SI or CI

engine. It is concluded that the cogeneration cycle has a

good thermal performance with first and second law

efficiencies of 46.47% and 38.5% respectively for the

mean operating conditions of T =300K, Pr=3, ç =80%.0 C

2. Maximum exergy was destroyed in the HCCI engine

which is 2817.59kJ/kg at mean operating conditions.

Exergy destruction in Heat transfer processes e.g.

regenerator and fuel vaporizer accounts for about

72.409kJ/kg and 65.003kJ/kg respectively at mean

conditions. The exergy destruction in catalytic

converter is 127.673kJ/kg.

3. The exergy destruction in HCCI engine and catalytic

converter is high because the effect of chemical exergy

in these components predominates over the effect of

physical exergy.

4. Exergy destruction in ORC evaporator, ORC turbine,

ORC condenser, pump and cogeneration heat recovery

steam generator (HRSG) is very less compared to main

HCCI engine components.

5. It indicates that the exergy analysis is providing ranking

among the components of the system. The component

with higher exergy destruction is very much responsible

to deteriorate the performance of the system as

compared to the components with lower exergy

destruction. It further indicates that which component

needs to be repaired or serviced first.

Heat recovery from automotive engines has been

predominantly for turbo-charging or for cabin heating.

Studies relative to application of the recovered heat to run

organic Rankine cycle (ORC) is scarce. In this paper an

ORC is attached with hot exhaust gases of HCCI engine.

Mathematical model results suggest that the concept is

thermodynamically feasible and could significantly

enhance system performance of the engine. It will largely

benefit considering the cost advantage particularly fuel cost

in long run and emission control. This would definitely

provide a right platform for rapid and qualitative

development of internal combustion engines and will bring

economic development.

REFERENCES

[1] Thring R.H.,(1989) “Homogeneous charge compression ignition

(HCCI) engines”, SAE International, SAE Paper no. SAE 892068

[2] S. Onishi, S. H. Jo, K. Shoda, P.D. Jo, S. Kato, (1979), “Active

thermo-atmosphere combustion (ATAC)-A new combustion process

for internal combustion engines”, SAE International, SAE Paper no.

790501.

[3] Noguchi. M., Tanaka, Y., Tanaka, Y., and Takeuchi. Y., (1979), “A

study on gasoline engine combustion by observation of intermediate

reactive products during combustion,” SAE International, SAE

Paper no. 790840.

[4] Olsson, J.O., Tunestal, P., and Johnsson, B., (2001), “Closed loop

control of an HCCI engine” SAE International, SAE Paper no. 2001-

01-1031.

[5] Mach, J.H., Aceves, S.M., Dibble, R.W., (2009) “Demonstrating

Direct Use of Wet Ethanol in a Homogeneous Charge Compression

Ignition (HCCI) Engine,” Energy, Vol. 34, pp. 782-787

[6] Saisirirat, P., Togbe, C., Chanchaona, S., Foucher, C., Rousselle, M.,

and Dagaut, P., (2011) “Auto- Ignition and Combustion

Characteristics in HCCI and JSR Using 1-butanol/n-Heptane

Blends,” Proceedings of Combustion Institute, Vol. 33, Issue 2, pp.

3007-3014.

[7] Wu, H.W., Wang, R.H., Ou, D. J., Chen, Y.C., Chen, T.Y., (2011)

“Reduction of Smoke and Nitrogen Oxides of a Partial HCCI Engine

Using Premixed Gasoline and Ethanol with Air,” Applied Energy,

Vol. 88, Issue 11, pp. 3882-3890.

[8] Frias, J.M., Aceves, S.M., Flower, D.L.,(2007), “Improving Ethanol

Life Cycle Energy Efficiency by Direct Utilization of Wet Ethanol in

HCCI Engines,” ASME Trans., Journal of Energy Resources

Technology, Vol. 129, pp. 332-337.

[9] Moran, M.J., and Shapiro, H.N., (2008), Fundamentals of

Engineering Thermodynamics, 6th ed., Willey New York.

[10] Heywood, J., (1988), Internal Combustion Engine Fundamentals,

McGraw Hill, New York.

[11] Green, D.W., Perry, R.H., (2008) Perry's Chemical Engineer's Hand

Book, Eighth ed., McGraw Hill, New York.

[12] Osborne R.J., Li, G., Sapsford, S.M., Stokes, J., Lake, T.H.,(2003)

“Evaluation of HCCI for Future Gasoline Power Trains,” SAE Paper

No. 2003-01-0750.

[13] Trivedi, S.K., Khaliq, A., Sharma, P.B., (2010) “An Examination of

Exergy Destruction in Wet Ethanol Operated HCCI Engine Based on

First and Second Law Analysis,” Proceedings of First International

Conference on New Frontiers in Biofuel, Sharma et al., eds., DTU,

New Delhi, India, pp295-305.

[14] A., Khaliq, S.K., Trivedi, I., Dincer, (2011) “Investigation of wet

ethanol operated HCCI engine based on first and second law

analysis,” Applied Thermal Engineering, vol.31, Issue 10, pp.1621-

1629.



Experimental Investigations of Traveling Wire Electro-Chemical Spark Machining (TW-ECSM) of Borosilicate Glass

Basanta Kumar Bhuyan and Vinod YadavaDepartment of Mechanical Engineering,

Motilal Nehru National Institute of Technology,

Allahabad – 211 004, Uttar Pradesh, India

E-mail: [email protected], [email protected]

Abstract - Traveling Wire Electro-Chemical Spark Machining

(TW-ECSM) is an innovative hybrid machining process,

combining the features of the Electro Chemical Machining

(ECM) and Wire-Electro Discharge Machining (WEDM). It is

more suitable for machining of electrically non-conductive

engineering materials. Conventional machining methods and

some unconventional machining methods cannot be

effectively applied for machining of borosilicate glass due to

the resulting problems of air borne dust, tool wear and

thermal damage. In this paper an inhouse TW-ECSM setup

has been designed and fabricated successfully and employed

for experimentation. The results about the feasibility of the

process and its performance during machining of borosilicate

glass have been illustrated. Borosilicate glass, which is

frequently used as a material for fabrication of micro

structures, was used as a workpiece. Experiments were

carried out to investigate the effects of specimen thickness

along with different type of wires on material removal rate

(MRR). For same set of input parameters, material removal

rate is found to decrease initially but after achieving certain

value, it begins to increase with increase in specimen

thickness.

Keywords: Traveling Wire Electro-Chemical Spark

Machining (TW-ECSM), Borosilicate Glass, Hybrid

Machining Process, MRR

1. INTRODUCTION

Borosilicate is a more stable form of glass, and

undergoes less expansion and contraction with changes in

atmospheric temperature and pressure. It transmits and

reflects a wide range of the light spectrum. For that reason,

it is useful in telescopes and optical instruments used in

space. It can be ground into a great variety of useful shapes,

ground to a high gloss, and coated with useful industrial and

scientific materials. It has also been formed into a variety of

utilitarian household objects, such as towel rods,

coffeepots and light fixtures, as well as jewelry and

accessories. Borosilicate glasses are also found in

semiconductor application in the micro electro-mechanical

systems development.

New ways of cutting hard and brittle non-conductive

engineering materials which are difficult-to-machine by

conventional methods continue to attract attention. While

electrical methods such as electro chemical machining

(ECM) and electro discharge machining (EDM) have

proved useful, drawbacks such as the expense of tooling for

forming large cavities and low rate of material removal have

hindered their wider acceptance. Ultrasonic machining

(USM), abrasive jet machining (AJM), laser beam

machining (LBM), and electron beam machining (EBM)

are some of the advanced machining processes that can be

used for literature by different researchers, such as Electro

chemical arc machining, Electro chemical discharge

machining and Spark assisted chemical engraving [3]. The

diversity of name illustrates the complexity of the process.

Machining these materials, but dimensional accuracy

and surface quality of the machined surfaces are the major

concern. Recently, a new trend has been introduced to

combine the features of different machining processes.

Such machining processes are called as hybrid machining

processes (HMPs). HMPs are developed to exploit the

advantages of each of the constituent machining processes

and diminish the advantages of each constituent process. It

has been observed that sometimes, hybrid machining

process enhances the material removal rate (MRR),

increases the capabilities of the constituent processes, and

widen the area of application of the constituent processes.

HMPs also reduce some adverse effects of the constituent

processes when they are applied individually. Electro-

chemical spark machining (ECSM) is one of the HMPs,

which combines the features of electro chemical machining

and electro discharge machining, has stemmed from its

ability to remove metal at high rates, as much as five and

fifty times faster than ECM and EDM, respectively under

the same parameter setting [1]. The ECSM process uses

Electro-Chemical Discharge (ECD) phenomenon for

generating heat for the purpose of removing work material

by melting and vaporization. This was presented for the first


time in 1968 by Kurafuji as “Electro-chemical Discharge

Drilling” for microholes in glass [2]. Several other names of

ECSM are used in ECSM with ECD have been tried in

many configurations: Die Sinking-ECSM, Hole Sinking-

ECSM, Die Drilling-ECSM, Hole Drilling-ECSM, Wire

Cutting-ECSM, Disc Cutting-ECSM, Cylindrical

Grinding-ECSM, Surface Grinding-ECSM and Pocket

Milling-ECSM. Die Sinking–ECSM operation usually

involves machining of cavity using non-rotating tool

electrode, where as in hole drilling–ECSM a rotating tool

electrode is used with the main focus on the surface quality

of side wall of the hole. Wire cutting–ECSM is capable of

slicing large volumes and machining complex shapes of

non-conducting materials without the need of full form tool

electrode. In contour milling–ECSM, a simple shape tool

electrode is used to produce a three-dimensional (3-D)

cavity by adapting a movement strategy similar to

conventional milling. Success in the application of sinking

and drilling ECSM has stimulated interest in studying the

prospects of TW-ECSM. The development of the Traveling

Wire Electro-Chemical Spark Machining (TW–ECSM) is

the outcome of machining requirements. In 1985, Tsuchiya

et al. [4] proposed TW-ECSM first time for cutting non-

conducting materials such as glasses and ceramics. Various

researchers have put forth explanations of ECD

phenomenon based on their experimental studies.

Bhattacharya et al. [5] conducted experiments on

alumina and concluded that the most effective parametric

combination for moderately higher machining rate and

dimensional accuracy are 80V and 25% NaOH

concentration. Tool tip geometry was also found to play an

important role in a controlled spark generation in ECDM.

Kulkarni et al. [6] proposed the discharge phenomenon

similar to arc discharge in gases. They proposed that

hydrogen gas bubbles get accumulated at the tool-electrode

tip leading to combining of bubbles into a single large

bubble which isolates the tip completely from the

electrolyte. This causes the local electric field gradient

between the tool and electrolyte interface to go beyond the

breakdown limit of 25V/µm leading to an arc discharge.

Basak and Ghosh [7] treated the discharge phenomenon as a

switching off process due to bubble bridges. Hofy and

McGeough [1] carried out experimental studies on the

effects of mode of electrolyte flushing, wire erosion,

machining speed on metal removal rate during TW-ECAM.

Their recommendation was to use coaxial mode of flushing

for maintaining the machining action and its accuracy and

also reported the values of bubble diameters, 1µm.

Peng and Liao [8] verified that TW-ECDM can be

applied for slicing meso-size non-conductive brittle

materials of several millimeters thick. They have shown

that pulsed DC power shows better spark stability and more

spark energy than constant DC power. Nesarikar et al. [9]

carried out experimental study for the feasibility of TW-

ECSM process for precision slicing of thick Kevlar-epoxy

composite. They did comparison between the experimental

and calculated values of MRR and average diametral

overcut with the variations in electrolyte conductivity,

applied voltage and specimen thickness. Jain et al. [10]

carried out experiments on their self developed setup of

TW-ECSM for cutting Glass epoxy and Kevlar epoxy

composites using NaOH electrolyte. They found that the

wire wear rate and the overcut follow a similar behavior as

the machining rate but the wire wear rate is about two

magnitudes smaller than the MRR. They also tried to study

the effect of artificially introducing some bubbles into the

process during machining and found that the MRR as well

as the overcut decreases slightly. Yang et al. [11] carried out

experimental study to improve the overcut quality by

adding SiC abrasive to the electrolyte. They discussed the

effects of abrasive on expansion, roughness and MRR on

the various machining parameters in Wire Electro-

Chemical Discharge Machining (WECDM). Singh et al.

[12] attempted to explore the feasibility of using TW-

ECSM process for machining of electrically partially

conductive materials like lead zirconate titanate and carbon

fiber epoxy composites. They found that MRR increases

with increase in supply voltage. MRR also increases with

increase in concentration of the electrolyte up to around 20

wt. %. Beyond this concentration, it starts decreasing. They

also observed that machined surface shows evidence of

melting. Large cracks are sometimes observed when the

machining is done at higher voltage. However, such

cracking is not seen at lower voltage.

Based on the above literature survey, studied in depth, it

has been found that the ECSM process in general and the

TW-ECSM process in part icular has not been

commercialized and literature available for this process is

still scarce. The focus of the present work is on developing

an inhouse TW-ECSM setup and experimental analysis of

the non-conducting engineering materials. The effects of

specimen thickness and type of wires on material removal

rate have also been observed.

II. FUNDAMENTAL OF ECSM

The ECSM process involves a complex combination of

the electro chemical action and electro discharge action.



The electrochemical action helps in the generation of the

positively charged ionic gas bubbles, i.e. hydrogen gas (H ). 2

The electro discharge action takes place between tool and

workpiece due to breakdown of the insulating layer of the

gas bubbles. In ECSM process, an electrochemical cell

consists of two electrodes dipped in electrolyte, one is

larger electrode (anode) called as dummy electrode or

auxiliary electrode and another one is smaller electrode

(cathode). The distance between two electrodes is 30-50

mm and smaller size electrode is dipped 2-3 mm below the

electrolyte as shown in Fig. 1.

When an external potential is applied between the

electrodes, electrical current flows through a cell resulting

in electrochemical reactions such as anodic dissolution,

cathodic deposition, electrolysis of electrolyte etc. Surface

area of the cathode dipped in the electrolyte is very small

compared to anode hence high current density at the

cathode results in rapid generation of hydrogen gas bubbles

Fig. 1 Electro-Chemical cell with two electrodes of grossly different size

and oxygen gas bubbles at the anode due to electrochemical

reactions. Boiling of electrolyte near small electrode would

occur due to Ohmic heating of the electrolyte. Beyond a

certain value of the applied potential, electric sparks appear

at the smaller electrode and the cell current drops. This is

known as Electro-Chemical Discharge (ECD) phenomenon

[2]. There are various theories [3] proposed to explain the

mechanism of spark generation at the cathode. However,

none of them has been verified experimentally.

In TW-ECSM process, D.C power is supplied between

the wire and the auxiliary electrode and the sparking takes

place between the wire and electrolyte and hydrogen gas

bubbles are accumulated and insulating layer is formed near

the wire surface. With the further increase of applied

voltage, sparking from wire takes place. If the workpiece is

kept in the vicinity of the spark zone, material is removed by

melting and vaporization. Thus the material removal

process in traveling wire electrochemical spark machining

is very complex in nature which is governed by various

process parameters. The experimental setup of TW-ECSM

has to be developed so that the process parameters are to be

properly controlled to achieve the good machining

performance.

III. DEVELOPED EXPERIMENTAL TW-ECSM SETUP

Traveling Wire Electro-Chemical Spark Machining

(TW-ECSM) setup has been designed and fabricated

keeping in view the fundamental mechanism of the process

and basic functional requirements of different parts. The

setup was performed after assembling various indigenously

developed basic components such as, Machining chamber,

Wire driving system, Electrolyte supply system and Power

supply system. A schematic diagram of the Traveling Wire

ECSM setup is shown in Fig. 2 (a) and a photograph of the

setup is also shown in Fig. 2 (b).

Fig. 2 (a) Schematic diagram of TW-ECSM setup and (b) Photographic view of the developed tabletop TW-ECSM setup



The machining chamber of size 400mmx250mm

x110mm is made of Plexiglass holds the electrolyte, as it is

an electrically insulating, transparent and corrosion

resistant material. It is kept on the lower platform of a

wooden table. On the middle wall of the machining

chamber electrode positioning and job-feeding unit is

fixed. At the bottom of machining chamber a hole is

provided to drain out electrolyte from chamber. Within

machining chamber the tool electrode is just touching the

non-conducting workpiece such as borosilicate glass. The

auxiliary electrode is a vertical graphite rod and a

horizontal scale is attached at the centre of the top edge of

the vertical rod. The horizontal scale is provided in order to

measure the horizontal displacement of the auxiliary

electrode which in turn helps to measure and control the

inter electrode gap. In the base and side wall of the

machining chamber, pulleys are attached, which helps in

movement of wire throughout machining chamber. The

electrolyte reservoir is attached with the side wall of the

machining chamber in order to supply electrolyte. The

workpiece holder is made of iron plate. A vertical up and

down movement up to 30mm can be made to change the

depth of the workpiece in the electrolyte by using depth

control mechanism.

The wire driving system consists of a feed spool, a take-

up spool, a set of pulleys and a stepper motor. The step angle

of the stepper motor is 1.8º. The rpm of the stepper motor

can be varied from 1 to 80. The programmable Logic

Controller (PLC) is used to rotate the stepper motor

smoothly. The input voltage to the stepper motor is 24V and

the input current to the stepper motor is 2.8A. The torque of

the stepper motor is 20kgcm. The wire electrode is fed

towards the workpiece at a constant rate from a feed spool

through a set of pulleys to the take-up spool. The pulley that

is submerged in the electrolyte is made of Teflon and other

pulleys are made of copper. A stepper motor drives the take-

up spool to pull the wire gently at a constant speed. An

anode made of graphite is attached to the pulley mount and

its distance from the cathode (wire) can be adjusted. The

distance between two electrodes is 30-50mm from each

other.

The electrolyte supply system consists of a pump and a

flow control valve. The electrolyte is supplied to the cutting

site on the work specimen can be immersed thoroughly in

the electrolyte. There are two different modes of electrolyte

flushing, such as (a) electrolyte flushing perpendicular to

wire and (b) electrolyte flushing coaxial with wire. The

electrolyte is added to the machining chamber from the

reservoir in the form of drops instead of flow from pipe. If

electrolyte is fed with high velocity, there will be no

formation of insulating layer or gas bubbles. Hence for this

thermal consideration the electrolyte should be added drop

by drop.

The power supply system used in TW-ECSM is mostly

DC power supply voltage able to maintain about 40V across

the cathodic tool-electrode and anodic auxiliary electrode.

Pulsating current can be applied to increase the

performance of TW-ECSM. Pulsating current has three

parameters such as pulse on-time, pulse off time and peak

current density. In the pulsed TW-ECSM process, a pulse

generator is used to supply the voltage pulses across the

electrodes. Pulsing is applied to this D.C by means of a timer

control. The main 230 volts, 3 phases, AC power supply are

converted to low voltage D.C power supply by a step down

transformer and silicon controlled rectifier unit. The

positive terminal of the power supply unit is connected with

auxiliary electrode and one end of the coil heating the

electrolyte. The negative terminal of the power supply unit

is connected with wire and another end of heating coil. Thus

temperature of the electrolyte is controlled electrically o ofrom 20 C to 60 C. The voltage and current can be recorded

with a voltmeter and ammeter.

IV. EXPERIMENTATION

Experiments of TW-ECSM have been conducted by

varying specimen thickness, keeping other parameters

constant. Initial experiments were performed in

borosilicate glass with graphite rod (diameter 8mm, length

55mm) as anode and copper wire of diameter 0.70mm, brass

wire of diameter 0.25mm and stainless steel of diameter

0.50mm as cathode. A rectangular borosilicate glass of size

40mm×35mm×2mm was adopted as a workpiece.

Workpiece was held at constant distance of about 35mm

from the anode. Cathode (wire) was always kept in physical

contact with the workpiece which was mounted on the

supporting platform. Copper wire and stainless steel wire

were broken frequently even at 64 volts but brass wire was

broken at 50 volts because of its low current carrying

capacity. Very low wire speed would lead to the situation

similar to the stationary tool resulting in overheating and

finally breaking of the wire. Too high wire speed was also

not desirable because it would be uneconomical. Hence,

wire was driven by stepper motor at a constant speed of 1.2

m/min. An aqueous solution of NaOH with a 200g/l solution

at 20ºC, was used.



NaOH has higher specific conductance, reactions take

place at higher rates, so a larger amount of gases were

evolved. Hence, higher MRR was achieved. Therefore, all

the experiments reported in this paper were carried out

using NaOH solution as electrolyte. Each experiment was

tested for about 5 to 12 min, during which voltage and

current were recorded on a voltmeter and ammeter,

respectively. The minimum linear feed rate to the

workpiece which could be achieved using the present setup

was 0.008mm/s. This feed rate was higher than the cutting

rate observed during the experiments. The nature of graph

observed that MRR of borosilicate glass using copper wire

and stainless steel are greater than the MRR using brass

wire of same material, keeping other parameter constant.

This is because of copper wire is a more electrically

conductive than brass wire and as per the diameter

concerned, stainless steel wire diameter was greater than

brass wire diameter. Copper possesses excellent

conductivity rating, low tensile strength, high melting

point and low vapour pressure rating which severely

limited potential. As new materials and demands came,

developers subsequently experimented with the use of

brass in order to meet the new demands. Brass is an alloy of

copper and zinc. Generally, the higher the zinc percentage,

the better the wire is for EDM.

V. RESULTS AND DISCUSSION

A. Machining of Borosilicate Glass

Micrograph of the specimen after machining has been

studied with the help of Optical Measuring Microscope

(OMM) as shown in Fig. 3.

Fig. 3 (a) shows after slicing of borosilicate glass with

keeping other parameter constant and (b) represents the

micrographs of machining portion of the specimen by using

optical measuring microscope 10X. Initial experiments

were performed in borosilicate glass with graphite rod as

Fig. 3 Machining of borosilicate glass (a) groove cut and (b) micrograph of groove cut using

optical measuring microscope 10X

anode and brass wire of diameter 0.25mm as cathode.

Likewise, other experiments were performed. Each

experiment was tested about 5 to 12 minutes, during which

time the machining of current and voltage were recorded as

ammeter and voltmeter respectively. Brass is an excellent

thermal and electrical conductive material as well as less

corrosive. So that current rises at the machining of the

materials within less time, results more hydrogen bubbles

accumulate at the tool tip and more material removes.

B. Material Removal Rate

After machining, the workpiece was washed, dried to

evaporate any water remaining on the surface and

reweighed using a weighing digital micro balance

(accuracy 10µg, CAS India Private Limited). The

difference between the original weight and the final weight

gave the amount of material removed. In this work, material

removal rate in millimeter per minute was evaluated as

shown in Fig. 4. The efficiency of the machining process is

also usually evaluated in terms of material removal rate in

milligrams per second.

1. Effect of Specimen Thickness

The effect of specimen thickness on MRR for

borosilicate glass with considering copper wire, stainless

steel and brass wire as shown in Fig. 4. MRR is decreasing

with increase in specimen thickness in the first part of the

curves. This is because of the obstruction in the current path

resulting in less bubble formation on the wire passing

through the groove and the energy generated on the wire

away from the workpiece is wasted in heating the

electrolyte. Similarly, second part of the curve, MRR

increases with increase in specimen thickness. This is due to

the discharge zone in the vicinity of the wire gets shifted to

the top of the workpiece and more bubble concentration on

the wire passing through the groove.

Fig. 4 Effect of specimen thickness on MRR



The Fig. 4 also shows that machining of borosilicate

glass with brass wire is better than copper and stainless steel

wire but low MRR than other two, due to low voltage. We

also observed that MRR decreases initially up to 3mm

thickness of the specimen and then increases with an

increase in the thickness of the specimen.

VI. CONCLUSIONS

Based on the experimentally obtained results during

machining of electrically non-conductive borosilicate

glass on designed and fabricated TW-ECSM setup and

thereafter discussion on the investigated results, the

following conclusions are drawn as listed below.

1. It has been found that machining of borosilicate glass

with brass wire is better than the copper wire and

stainless steel.

2. MRR has been observed to decrease initially with an

increase in specimen thickness upto certain value and

then increases continuously.

ACKNOWLEDGEMENT

Financial support for this work has been granted by the

Council of Scientific and Industrial Research (CSIR),

Government of India through project no. 22/

(0486)/09/EMR-II entitled “Experimental and Numerical

Study of Traveling Wire Electrochemical Spark Machining

of Advanced Engineering Materials”.

REFERENCES

[1] Hofy H. El. and McGeoug J. A. (1988), “Evaluation of an Apparatus

for Electrochemical Arc Wire-Machining,” Journal of Engineering

for Industry, Vol. 110, No. 2, pp. 119-123.

[2] Kurafuji H. and Suda K. (1968), “Electrical discharge drilling of

glass,” Annals of the CIRP, Vol. 16, pp. 415-419.

[3] Wuthrich R. and Fascio V. (2005), “Machining of non-conducting

materials using electrochemical discharge phenomenon- an

overview,” International Journal of Machine Tools & manufacture,

Vol. 45, No. 9, pp. 1095-1108.

[4] Tsuchiya H., Inoue T. and Miyazaiki M. (1985), “Wire

electrochemical discharge machining of glasses and ceramics,”

Bulletin Japanese Society of Precision Engineering, Vol. 19, No.1,

pp.73-74.

[5] Bhattacharyya B., Doloi B. N. and Sorkhel S.K. (1999),

“Experimental investigations into electrochemical discharge

machining (ECDM) of non-conductive ceramic materials,” Journal

of Material Processing Technology, Vol.95, No. 1-3, pp. 145-154.

[6] Kulkarni A., Sharan R. and Lal G. K. ( 2002), “An experimental

study of discharge mechanism in electrochemical discharge

machining,” International Journal Machine Tools & Manufacture,

Vol. 42, No. 10, pp. 1121-1127.

[7] Basak I. and Ghosh A. (1996), “Mechanism of spark generation

during electrochemical discharge machining: a theoretical model and

experimental investigation,” Journal of Materials Processing

Technology, Vol. 62, No. 1, pp.46-53.

[8] Peng W. Y. and Liao Y. S. (2004), “Study of electrochemical discharge

machining technology for slicing non-conductive brittle materials,”

Journal of Materials Processing Technology, Vol. 149, No. 1-3, pp.

363-369.

[9] Nesarikar V. V., Jain V. K. and Choudhury S. K. (1994), “Traveling

wire electrochemical spark machining of thick sheets of Kevlar-

Epoxy composites,” Proceedings of Sixteenth All India

Manufacturing Technology Design and Research Conference,

Bangalore, India, pp. 672-677.

[10] Jain V. K., Rao P. S., Choudhury S. K. and Rajurkar K. P. (1991),

“Experimental investigations into traveling wire electrochemical

spark machining (TW-ECSM) of composites,” ASME Transaction,

Journal of Engineering for Industry, Vol. 113, No. 1, pp. 75-84.

[11] Yang C. T., Song S. L., Yan B. H. and Huang F.Y. (2006), “Improving

machining performance of wire electrochemical discharge

machining by adding SiC abrasive to electrolyte,” International

Journal of Machine Tools & Manufacture, Vol. 46, No. 15, pp. 2044-

2050.

[12] Singh Y. P., Jain V. K., Prashant Kumar and Agrawal D. C. (1996),

“Machining piezoelectric (PZT) ceramics using an electrochemical

spark machining (ECSM) process,” Journal of Materials Processing

Technology, Vol. 58, No. 1, pp. 24-31.



Performance Characteristics of Diesel Engine Fueled byBiodiesel of Jatropha Oil and Soybean Oil

1 2*Ashish Malik and Parlad Kumar

1Department of Mechanical Engineering, ABES Engineering College,

Ghaziabad, Utter Pradesh, India2University College of Engineering, Punjabi University, Patiala, Punjab, India

* Corresponding author E-mail: [email protected]

Abstract - Biodiesel is considered as an important renewable

and alternative fuel of future. This paper focuses on the

performance of biodiesel made from jatropha oil and soybean

oil. The biodiesel was produced through transesterification

process and it was blended with the fossil diesel. The blended

mixtures were tested in an IC diesel engine attached with a

dynamometer. On the basis of performance tests it was found

that brake thermal efficiency of mixed jatropha and soybean

biodiesel blends is nearer to pure diesel oil at different rpm.

The brake specific fuel consumption obtained with biodiesel

blend of mixture of jatropha and soybean oil is comparable

with fossil diesel.

Keywords: Biodiesel, Alternative fuel, Jatropha oil, Soybean

oil, Brake thermal efficiency

1. INTRODUCTION

Diesel engines are used in a number of applications like

heavy automobile engines, cars, small irrigation water

pumping systems and small electricity generators etc.

Therefore, diesel fuel is used much more than any other

gasoline fuels. Diesel is produced from non-renewable

fossil fuels which are limited in quantity and one day they

would be exhausted. Biodiesel is an alternative and

renewable fuel that can be produced from various vegetable

oils, animal fats and waste cooking oil. Biodiesel can be

blended with fossil diesel to create a biodiesel blend. This

blend can be used in compression-ignition engines or oil-

fired boilers and furnaces with little or no modifications.

Since biodiesel is produced from plants, it is considered

as carbon neutral because after combustion it produces the

carbon which has been absorbed during photosynthesis

process. The use of biodiesel can reduce the use of

petroleum based fuels and possibly lower the overall

greenhouse gas emissions. Biodiesel, due to its

biodegradable nature, and absence of sulfur and aromatic

contents, produces less toxic emissions. Therefore it is

considered as an environment friendly fuel. Many

researchers have produced biodiesel by using different

edible and non-edible vegetable oils and these biodiesels

have been tested in diesel engines for their performance.

Hammerlein et al. conducted experiments using filtered

rapeseed oil in diesel engine and found that the brake power

and torque using rapeseed oil were 2% lower than that of

diesel oil. But, NO and particulate emissions were lower as x

compared to diesel fuel combustion. However, noise and

emission of CO and HC were higher [1]. Chio blended the

diesel and biodiesel in the ratio of 80:20 and 60:40 by

volume. The prepared blend was tested on a single cylinder

caterpillar engine, using both single and multiple injection

system. During high loads and single injection, the

particulate matter and CO emissions were decreased.

However, a small increase in NO was noticed when the bio-x

diesel concentration is increased. But in the case of multiple

injection a decrease in particulate emission was observed

with little or no effect on NO [2]. Agarwal used ethanol x

blended with diesel to study its performance and emissions

in an Engine. The tests demonstrated that almost all the

important properties of biodiesel are in very close to the

fossil diesel [3]. Rao et al. found that the brake thermal

efficiency for biodiesel and its blends was slightly higher

than that of diesel fuel at tested load conditions. It was found

that for jatropha biodiesel and its blended fuels, the exhaust

gas temperature increases with increase in power and

amount of biodiesel [4]. Godiganur et al. found that the

blend of Karanja oil methyl ester and diesel can be used

successfully as an alternative fuel without any affect on

engine power and performance. Moreover, the amounts of

exhaust emissions are lower than those of diesel fuel [5].

Mohanty et al. used Polanga oil to run an IC engine to

evaluate the combustion performance and emission

characteristics of diesel engine. It is found that Polanga oil

can be used as an alternative fuel in diesel engine without

affecting the performance of the engine [6].

Almost all of the vegetable oils can be used as raw

material to produce biodiesel. But it is suggested that only

non-edible oils should be used for this purpose due to

obvious reasons. In the present study different blends of

biodiesel from jatropha oil and soybean oil has been used in


Blend Ratio

(By volume)Given Name

Blend-1

Biodiesel from Jatropha 20% +

Diesel 80%Jatropha blend

Blend-2

Biodiesel from Soybean 20% +

Diesel 80%Soybean blend

Blend-3

Biodiesel from Soybean 10%

+Biodiesel from Jatropha 10% +

Diesel 80%

Soya+Jatropha blend

a diesel engine to check its performance and emission

temperature. The jatropha oil is a non-edible oil and it can

be produced from the carcass of jatropha plant. This plant

can be grown on non-irrigated land without any special

care.

II. PREPARATION OF BIODIESEL

Vegetable Oils (Triglycerides) are ester of glycerol and

fatty acids. A chemical reaction of these triglycerides with

the alcohol in the presence of a catalyst produces methyl

esters (biodiesel) and glycerol. This reaction process is

known as transesterification. Currently, most commercial

biodiesel is produced with the transesterification processes

using a homogeneous alkaline catalyst, generally NaOH or

KOH. The alkaline catalyst provides a faster reaction as

compared to an acidic catalyst. The alcohol which is

generally used is methanol due to its relatively low cost and

a high cetane number of the produced biodiesel [7].

Methanol has an ability to react with triglycerides quickly

and the alkali catalyst is easily dissolved. The rate of

transesterification process depends upon the amount of

methanol, catalyst, reaction temperature and reaction time.

The chemical reaction is given below:

CH COOR2

| Catalyst

CHCOOR + 3CH OH →CH OH) CHOH+3CH COOR3 2 2 3

|CH COOR2

Catalyst

(Triglyceride + Methanol→ Glycerol+Methyl Esters)

Theoretically, three moles of alcohol are required for

each mole of triglyceride to produce three moles of fatty

acid alkyl ester and one mole of glycerol. But since

transesterification is a reversible reaction, excess amounts

of alcohol are used to ensure that the oils or fats will be

completely converted to esters, and a higher alcohol

triglyceride ratio can result in a greater ester conversion in a

shorter time. Therefore maintaining a high alcohol to oil

ratio is essential. The commonly employed molar ratio for

two-step acid transesterification is 6:1 and 9:1 for alkali

catalyzed transesterification [8]. If molar ratio is further

increased it makes it difficult to separate the glycerol from

the oil [9]. The process temperature is constrained to be

below the boiling point of the alcohol used. Reported

reaction times for typical biodiesel production ranges from

30 minutes to over 2 hours, with catalyst concentrations that

vary between 0.1 and 2% [10, 11]. In this study jatropha oil

and soybean oil have been used as raw materials for

producing two different types of biodiesels. KOH was used

as catalyst. Table-1 shows properties of diesel and the oils

which have been used for the production of biodiesel.

For producing the biodiesel the vegetable oil was mixed owith methanol and KOH. The mixture was heated at 60 C

for 3 hours while stirring with magnetic stirrer. Mixing is

very important in the transesterification reaction as

triglycerides are immiscible in the alcohol solution. oTransesterification was carried out at 60 C, which is just

below the boiling temperature of methanol. The reaction

mixture was allowed to settle to get biodiesel at the top and

glycerin at the bottom. After separation the biodiesel was

filtered to remove any solid particles then it was heated at o100 C for 10 minutes in stainless steel tank to remove any

water contents.

After the preparation of two different types of biodiesel

by using two different raw materials, three different types of

blends of the biodiesel and conventional fossil diesel were

prepared as shown in the Table II.

In the first blend 20% of jatropha biodiesel is mixed with

80% of conventional diesel. In the second blend 20% of

Soybean biodiesel is mixed with 80% of conventional

diesel. In the third blend 10% of jatropha biodiesel and 10%

of soybean biodiesel is mixed with 80% of conventional

diesel. After the preparation of blends, tests were conducted

to determine the performance characteristics of the engine.

TABLE I PROPERTIES OF JATROPHA OIL, SOYBEAN OIL AND DIESEL

Fuel

Dynamic Viscosityat 20oC

(cP)

Density(g/cc) at

30oC

Calorific Value

(KJ/Kg)

Flash point

oC

Cetane Number

Diesel

4.83

0.845

45,870 35 49.6

Jatropha Oil

50.73

0.93292

45,456 240 51

Soybean Oil

60

0.9239

38,000 254 34.8

TABLE II DIFFERENT BENDS OF BIODIESEL AND DIESEL


Performance Characteristics of Diesel Engine Fueled by Biodiesel of Jatropha Oil and Soybean Oil

III. EXPERIMENTAL SET UP AND PROCEDURE

Experimental investigations were carried out on a

double cylinder DI diesel engine to examine the suitability

of jatropha and soybean biodiesel blends as alternate fuels.

Two cylinder four stroke Kirloskar engine test rig with

attached hydraulic dynamometer was used to conduct

performance tests on engine. Specifications of test rig are

given in Table III.

After warming up and stabilizing the engine, tests were

conducted. For all the tests the speed of the engine varied

from 1000 to 1500 rpm at a constant load. The jatropha and

soybean oil blends were supplied as fuel to the enginr. Each

experiment was repeated three to five times to calculate the

mean value of the experimental data. The performance

characteristics of the engine were evaluated in terms of

brake thermal efficiency (BTE), brake specific fuel

consumption (BSFC) and exhaust gas temperature (EGT).

These characteristics were compared with the results of

pure diesel.

IV. RESULTS AND DISCUSSION

A. Engine Performance

Different blends of diesel and biodiesel are tested and

compared with pure diesel for calculating BTE at different

speeds as shown in Fig. 1. The graph reveals that, maximum

value of BTE of different blends as well as pure diesel is

exhibited at 1000 rpm of the test engine. It is also observed

that at 1100 rpm the BTE of the jatropha blend approaches

to that of pure diesel, however on increasing the speed it

goes on decreasing. Soybean blend shows lower values of

BTE at all selected rpm. It may be due to its lower calorific

value. The soy+jatropha blend provide more BTE as

compared to blends of only soybean or only jatropha. But

the BTE of this blend is also less than the pure diesel.

TABLE III DIESEL ENGINE TEST RIG SPECIFICATIONS

ManufactureKirloskar Oil Engines Ltd.

Engine Type4 stroke, double cylinder

Power 16 HPR.P.M 1500Cylinder bore

87 mmCylinder stroke

110 mmInjection opening Pressure

200 bar

Cooling medium

Water cooledDynamometer type

Hydraulic

Comparison Of Brake Thermal Efficiency Of Different Oil

Blends

6

8

10

12

14

900 1000 1100 1200 1300 1400 1500 1600

RPM

Bra

ke

Th

erm

al

Eff

icie

nc

y

Diesel Jatropha Soybean Soya+Jatropha

Fig. 1 Comparison Curves For BrakeThermal Efficiencies of Different Oil blends

B. Brake Specific Fuel Consumption

Brake specific fuel consumption (BSFC) has been

evaluated for different blends and at different rpm as shown

in Fig. 2.

It is found that pure diesel has the lowest BSFC value.

This is primarily due to the fact that diesel has the highest

calorific value among three fuels, and needs the lowest fuel

consumption rate for achieving the same engine brake horse

power output as by the other three blends. The values of

BSFC for soy+jatropha blend are slightly higher than pure

diesel at all values of rpm. It has been found that for all types

of tested fuels, as the rpm of engine increases BSFC also

increases under constant load conditions. BSFC is higher in

case of soybean blend with its lower BTE at all engine

speeds.

C. Exhaust Gas Temperature

Exhaust gas temperature (EGT) increased with increase in

engine speed for all the cases as shown in Fig. 3.

Comparion Of Brake Specific Fuel Consumption for

Different Oil Blends

0.5

0.6

0.7

0.8

0.9

1

1.1

1.2

1.3

900 1000 1100 1200 1300 1400 1500 1600

RPM

Bra

ke S

pecif

ic F

uel

Co

nsu

mp

tio

n (

kg

/kw

.hr)

Diesel Jatropha Soybean Soya+Jatropha

Fig. 2 Comparison Curves For Brake SpecificFuel Consumption of Different Oils


Ashish Malik and Parlad Kumar

Comparison Of Exhaust Temperatures Of Different Oils

100

120

140

160

180

200

220

900 1000 1100 1200 1300 1400 1500 1600

RPM

Ex

ha

us

t Te

mp

era

ture

Diesel Jatropha soybean soya+jatropha

Fig. 3 Comparison Between Exhaust Temperatures of Different Oils

EGT values for jatropha blend are less than the pure

diesel at 1100 rpm and it remains nearer to the EGT values of

pure diesel upto 1300 rpm. After 1300 rpm EGT of jatropha oblend increases and reaches to maximum value of 190 C at

1500 rpm. The EGT of soya+jatropha blend is near to pure

diesel at 1400 rpm and slightly less than the pure diesel at

1500 rpm. EGT of soybean is higher as compared to tested

blends at various speeds of the engine. EGTs of biodiesel

blends are different from pure diesel oil, it may be due to

their different calorific values and different trends in

specific fuel consumption during the tests.

V. CONCLUSIONS

On the basis of performance tests it is found that BTE of

mixed jatropha and soybean biodiesel blends is nearer to

pure diesel oil at different rpm. The brake specific fuel

consumption obtained with biodiesel blend of mixture of

jatropha and soybean oil is comparable with pure diesel.

Exhaust gas temperature of soybean oil is maximum at all

recorded engine speeds as compared to other oil blends and

pure diesel oil. The performance test carried out showed

that blend of diesel and biodiesel (soya+jatropha blend) can

be successfully used as an alternative fuel for diesel engines

without any modifications.

REFERENCES

[1] Hemmerlein N., Korte V., Richter H., Schröder G. (1991),

Performance, Exhaust Emission and Durability of Modern Diesel

Engines Running on Rapeseed Oil, SAE Paper Series 910848,

International Congress and Exposition, Detroit, Michigan.

[2] Choi C.Y. (1997), Effect of Bio-diesel Blended Fuels and Multiple

Injections on D.I. Diesel Engines, SAE 970218.

[3] Agarwal A.K. (2007), “Biofuels (alcohols and biodiesel)

Applications as Fuels for Internal Combustion Engines”, Progress in

Energy and Combustion Science, Vol. 33, No. 3, pp. 233-271.

[4] Rao Y.V.H., Voleti R.S., Raju A.V.S. and Reddy P.N. (2009),

“Experimental Investigations on Jatropha Biodiesel and Additive In

Diesel Engine”, Indian Journal of Science and Technology, Vol.2,

No. 4, pp. 25-31.

[5] Godiganur S.K., Murthy C.S. and Reddy R.P. (2010), “The Effect Of

Karanja Oil Methyl Ester On Kirloskar HA394DI Diesel Engine

Performance And Exhaust Emissions”, Thermal Science, Vol. 14,

No. 4, pp. 957-964.

[6] Mohanty C., Jaiswal A., Meda V.S., Behera P. and Murugan S. (2011),

“An Experimental Investigation on the Combustion Performance and

Emissions of a Diesel Engine Using Vegetable Oil-Diesel Fuel

Blends”, The Essentional Automotive Technology Event- April 12-

14, SAE International, USA.

[7] Noiroj K., Intarapong P., Luengnaruemitchai A. and Jai-In S.(2009),

“A Comparative Study of KOH/Al2O3 and KOH/Na Catalysts for

Biodiesel Production via Transesterification from Palm Oil”,

Renewable Energy, Vol. 34, No. 4, pp. 1145-1150.

[8] Ala'a Alsoudy, Mette Hedeggard Thomsen and Isam Janajreh (2012),

“Influence on Process Parameters in Transesterification of Vegetable

and Waste Oil -A Review”, International Journal of Research and

Reviews in Applied Sciences, Vol. 10, No. 1, pp. 64-77.

[9] Srivastava A. and Parsad R. (2000), “Triglycerides Based Diesel

Fuels”, Renewable and Sustainable Energy Reviews, pp 111-113.

[10] Alamu O.J., Waheed M.A. and Jekayinfa S.O. (2007), "Biodiesel

Production from Nigerian Palm Kernel Oil: Effect of KOH

Concentration on Yield", Energy for Sustainable Development, Vol.

11, No.3, pp 77-82.

[11] Alamu O.J., Waheed M.A. and Jekayinfa S.O. (2008), “Effect of

ethanol-Palm Kernel Oil Ratio on Alkali-Catalyzed Biodiesel

Yield”, Fuel, Vol. 87, No. 8-9, pp 1529-1533.


Performance Characteristics of Diesel Engine Fueled by Biodiesel of Jatropha Oil and Soybean Oil

Machining Study of TI-6AL-4V Using PVD Coated TiAlN InsertsNarasimhulu Andriya, Venkateswara Rao P and Sudarsan Ghosh

Department of Mechanical Engineering, Indian Institute of Technology Delhi, New Delhi -110 016, India

Abstract - This paper deals with machining Ti6Al4V material. The experimental analysis was carried out using Response Surface Methodology (RSM). The detailed experiments under wet and dry conditions using the PVD coated TiAlN tools. In the present work the relationship of Ti6Al4V’s surface roughness and cutting forces with critical machining parameters and conditions, based on experimental input and output data, has been derived during the turning operation. It has been found through design of experiments technique that linear model is best fitted for predicting feed force and surface roughness under both dry and wet cutting environment. Linear model is also fitted for thrust force prediction during dry cutting. However under wet cutting condition a quadratic model is more suited for prediction of the thrust force. 2FI (2 Factor Interaction) model is found to be fitted for cutting force prediction under both the cutting environment.

Keywords: Ti6Al4V-alloy, PVD Coating, TiAlN tool, RSM

I. IntroductIon

Titanium and its alloys are considered as extremely difficult to machine materials. Titanium and its alloys have several promising inherent properties (like low strength-weight ratio, high corrosion resistance etc.) but their machinability is generally considered to be poor. Titanium and its alloys have high chemical reactivity with most of the available cutting tool materials. Also due to the low thermal conductivity of these alloys the heat generated during machining remains accumulated near the machining zone. Consequently the cutting tools are more prone to thermal related wear mechanism like diffusion, adhesion wear. Hence, on machining, the cutting tools wear out very rapidly due to high cutting temperature and strong adhesion between tool and workpiece material. Additionally, the low modulus of elasticity of titanium alloys and its high strength at elevated temperature makes the machining further difficult [1-3].

To a large extent, machining of titanium and its alloys follows criteria that are also applied to common metallic materials. Compared to high strength steels, however, some restrictions have to be recognized, which are due to the unique physical and chemical properties of titanium and its alloys. The lower thermal conductivity of titanium alloy hinders quick dissipation of the heat caused by machining. This leads to increased wear of the cutting tools. The lower modulus of elasticity of titanium leads to significant spring back after deformation under the cutting load. This causes titanium parts to move away from the cutting tool during machining which leads to high dimensional deviation in the workpieces. The lower hardness of titanium and its higher chemical reactivity leads to a tendency for galling of titanium with the cutting tool and thereby changing the important

tool angles like the rake angles Titanium alloy machining performance can be increased by selecting improved cutting tool materials and coated tools [4-5]. Now a days, most of the carbide cutting tools come with hard coatings deposited on them either by the CVD or PVD technique. PVD coated tools have been found to be better performing compared to their CVD counterparts. Also in PVD thinner coatings can be deposited and sharp edges and complex shapes can be easily coated at lower temperatures [6]. PVD–TiAlN-coated carbide tools are used frequently in metal cutting process due to their high hardness, wear resistance and chemical stability. Also, they offer higher benefits in terms of tool life and machining performance compared to other coated cutting tool variants.

Currently in machining industries hard turning process is being used to obtain high material removal rates. For successful implementation of hard turning, selection of suitable cutting parameters for a given cutting tool - workpiece material and machine tool are important steps. Study of cutting forces is critically important in turning operations [7] because cutting forces co-relate strongly with cutting performance such as surface accuracy, tool wear, tool breakage, cutting temperature, self-excited and forced vibrations, etc. The resultant cutting force is generally resolved into three components, namely feed force (Fx), thrust force (Fy) and cutting force (Fz).

Machining of titanium and its alloys differs from conventional turning of engineering materials like steel, in several key ways, mainly because the thermal conductivity of the material is very low when compared to the steel (KTi is 7.3W/mK and KSteel is 50.7W/mK) [8]. This low thermal conductivity results in high heat accumulation at the machining zone (shear zone) and heat dissipation is very less when compared to conventional turning of steels.

II. LIterature revIew

CNC Turning is widely used for machining of symmetrical components in a variety of industries such as automotives, aerospace, chemical, biomedical, textile and other manufacturing industries. In the machining process, errors may occur due to the problems in the machine tool, machining methods and the machining process itself. Of these, the errors that arise due to high cutting forces are the major problems for machining process. In turning, cutting forces and surface finish are important parameters by which the performance can be assessed. Hence it is important to minimize the cutting forces and maximize the surface finish.

Sun et al [9] studied the characterization of cutting


forces in dry machining of titanium alloys considering input parameters like cutting speed (60-260 m/min) , feed ( 0.12 to 0.3 mm/rev)and depth of cut (0.5 to 2 mm) using uncoated inserts and they have reported that cutting forces increases with increase in feed and increase in depth of cut. Venugopal et al [10], Hong et al [11], have studied the cutting forces under dry and wet cutting environment for machining of Ti-6Al-4V using uncoated inserts and they compared the results with cryogenic machining. Jawaid et al [12] have studied the machining of titanium alloys using PVD TiN coated and CVD coated (TiCN+Al2O3) in wet cutting environment and they assessed the wear mechanism of coated inserts. Nalabant et al [13] have investigated extensively the effects of uncoated, PVD and CVD coated cutting inserts and the various cutting process parameters on surface roughness and they have found that the best average surface roughness values were obtained at cutting speed of 200 m/min with a feed of 0.25 mm/rev using a 2.3 µm thickness PVD coated TiAlN-coated cutting tool.

Recently Yuan et al [14] studied the machining of titanium alloys using uncoated cemented carbide inserts under three different cutting environments such as dry, wet, MQL with room temperature and MQL with varying temperature of cooling air. Fang et al [15] did a comparative study of the cutting force in high speed machining of Ti-6Al-4V and Inconel 718 and they have explained the similarities and differences both quantitatively and qualitatively in terms of force related quantities.

Most of the experimental investigations on titanium machining have been conducted using two-level factorial design (2k) for studying the influence of cutting parameters on cutting forces and surface roughness[11, 15-16]. In two-level factorial design, one can identify and model linear relationships only. For studying the nonlinearity present in the output characteristics at least three levels of each factor are required (i.e. three-level factorial design, 3k). A central composite design which requires fewer experiments than alternative 3k design is usually better. Again, sequential experimental approach in central composite design can be used to reduce the number of experiments required. Keeping the foregoing in mind, the present work is focused on investigations of cutting forces and surface roughness as a function of cutting parameters in titanium machining using sequential approach in central composite design technique. The study was conducted on Ti-6Al-4V alloy using coated tools under dry and wet environment to analyze and compare the measured output parameters. Regression equations correlating input parameters viz., Cutting speed, feed, depth of cut and effective rake angle with output like forces and surface roughness were established based on experimental data.

The review of literature suggests that for the machining titanium alloys most researchers have used the input machining parameters like cutting speed, feed and depth of

cut. But there are hardly any paper where researchers have used different rake angles as also an input parameter. In the current paper the effective rake angle is considered as another input parameter. The major objective of the present work is to experimentally find the magnitude of the cutting forces and the surface roughness of the turned components and compare them under dry and wet cutting environment.

III. experImentaL detaILs

The details of experimental conditions, instrumentations and measurements and the procedure adopted for the study are described in this section.

A.Workpiece Material

Titanium alloys have found wide applications owing to its unique characteristics like low density [2]or high strength to weight ratio (density of titanium is about 60% of that of steel or nickel-based super alloys) and excellent corrosion resistance (for biomedical, chemical and other corrosion-resistant environments). Titanium is an expensive metal to extract, melt, fabricate and machine. Titanium alloys are considered to be difficult-to-machine materials. This is due to certain inherent metallurgical characteristics of these alloys that make them more difficult and expensive to machine than steels of equivalent hardness. Titanium alloys have low thermal conductivity due to which the heat generated in the cutting zone cannot be rapidly conducted away into the fast-flowing chip.

In the present study Ti-6Al-4V alloy bars of 60 mm diameter and length 200 mm were used. They were annealed and their chemical compositions are given in the Table I.

Table I ChemICal ComposITIon (%) of TI–6al–4V

B.Cutting Tool

In the present experiments, 5 levels of rake angle were used. The -6 degree default rake angle tool holder for CNMG tool inserts was used and for VNMG inserts the tool holder default rake angle -10 degrees was used. So, the rake angles


Machining Study of TI-6AL-4V Using PVD Coated TiAlN Inserts

obtained by such combination of inserts and tool holders are -10, -6, 0, 7 and 14 degrees.

C.Machine Tool

A rigid, high precision T-6 (Leadwell, Taiwan) lathe equipped with specially designed experimental setup was used for carrying out the experiments. For increasing rigidity of machining system, workpiece material was held between chuck (three jaw) and tailstock (revolving center).

D.Cutting Conditions

The experiments have been conducted using tool holders with -6 and -10 degree default rake angle. In this study the input parameters and their levels are shown in Table III.

E.Cutting Force Measurement

The cutting forces were measured using Kistler® piezoelectric dynamometer (model 9257B) mounted on specially designed fixture. Kistler® tool holder (model: 9129AA) was used for holding the 20×20 shank size cutting tool. The charge generated at the dynamometer was amplified using three-charge amplifier (Kistler®, Model: 5070A). The input sensitivities of the three-charge amplifiers were set corresponding to the output sensitivity of the force dynamometer in the x, y and z directions. The amplified signal was acquired and sampled using USB data acquisition system and stored in computer using Dynaware software for further analysis. The sampling frequency of data was kept at 300 samples/s per channel and the average value of steady-state force was used in the analysis.

Table III The leVels and InpuT parameTers

F.Surface Roughness Measurements

The measurements of average surface roughness (Ra) were made on the Taylor Hobson Surface roughness measuring machine with Ultra Surface Finish Software V5 version. Three measurements of surface roughness were taken at different locations and the average value was used in the analysis.

G.Response Surface Methodology

Response surface methodology (RSM) is a collection of mathematical and statistical techniques that are useful for the modeling and analysis of problems in which a response of interest is influenced by several variables and the objective is to optimize this response [18].

H.Experimental Plan Procedure

Planning of experiments is an important stage. Number of experimental runs was decided by using the response surface methodology. In this study, cutting experiments are planned using five-levels of each of the input parameters. Cutting experiments are conducted considering four input parameters or factors: Cutting Speed, feed, depth of cut and rake angle. A total of 30 experiments were performed on a CNC turning center (T-6 Lead well). The cutting experiments involved in the machining of Ti–6Al–4V with TiAlN-PVD coated carbide tools, five levels of cutting speeds, feeds, and depth of cut and effective rake angles. Two sets of environments have been used to compare the experimental output.

Iv. resuLts and dIscussIon

The results are analyzed in Design Expert V8.0.6 software. An ANOVA summary table is commonly used to summarize the test of the regression model, test of the significance factors and their interaction and lack-of-fit test. If the value of ‘Prob > F’ in ANOVA table is less than 0.05 then the model, the factors, interaction of factors and curvature are said to be significant. Finally, % contribution column is added in ANOVA summary table and it often serves as a rough but an effective indicator of the relative importance of each model term [18].

A. Force Components: The Cutting, Thrust Force And Feed Force Against Input Parameters

Anova analysis shows that the model is significant and feed (B) and depth of cut (C) are only the significant factors (terms) in the model. All other terms are insignificant. In default the central composite design the curvature is insignificant which says that the model is linear. The lack of fit also confirms the insignificance as depicted from Anova analysis thereby indicating that the model fits well with the experimental data.

The various R2 statics ( i.e R2, adjusted R2and Predicted R2) of the cutting force are exported for Anova table for dry and wet cutting environment. The value R2 = 0.9748 for Dry and the value for R2 = 0.9749 for wet cutting environment of Fz force indicates that 97.48% for dry and 97.49% for wet of the total variations are explained by the model. The adjusted R2 is a static that is adjusted for the size of the model. The value of the adjusted R2 = 0.9719 for Dry and the value of adjusted R2 = 0.97206 for Wet cutting environment indicates that 97.19 % for Dry and 97.2% for wet of the total variability



is explained by the model after considering the significant factors. Predicted R2 = 0.967 for dry and Predicted R2 = 0.9674 for wet cutting environment is in good agreement with adjusted R2 and shows that the model would be expected to explain 96.7% for Dry and 96.74% for Wet of the variability in new data [18]. ‘C.V.’ stands for the coefficient of variation of the model and it is the error expressed as a percentage of the mean ((S.D./Mean)×100). Lower value of the coefficient of variation (C.V. = 8.20%) indicates improved precision and reliability of the conducted experiments.

The same procedure was applied on thrust force (Fy) and resulting ANOVA with R2 statistics for models (considering only the significant terms) generated. For the thrust force, the cutting velocity and effective rake angle is insignificant and feed and depth of cut are significant.

The response surface eqauations as obtained from the Anova analysis and are follows

Fx =96.49+387.437*feed -- (1)

Fx= 66.493+450.1*feed -- (2)

Fy = 15.397 + 160.7861 * depth of cut -- (3)

Fy=7.43+0.0019*V+3.955*doc0.2621*gama+0.00142*v* gama+18.9177*f*doc+0.6797*f*gama+18.9177*f*gama+

208.44*f^2+0.42709*gama^2 -- (4)

Fz= 15.89+61.833*f+62.58*doc+1548*f*doc -- (5)

Fz = -8.451+164.541*f+61.68*doc+1426.45*f*doc -- (6)

From equations 1to 6 are alternet Dry and wet cutting environments respectively. The normal probability plot of the residuals (i.e. error = predicted value from model−actual value) cutting force is shown in Fig 1.1- Fig 1.2 for dry and wet cutting environment and reveal that the residuals lie reasonably close to a straight line, giving support that terms mentioned in the model are the only significant[18].

Design-Expert® SoftwareFz

Color points by value ofFz:

478

134.8

Internally Studentized Residuals

Norm

al %

Pro

babi

lity

Normal Plot of Residuals

-3.00 -2.00 -1.00 0.00 1.00 2.00 3.00

1

5

10

2030

50

7080

90

95

99

Design-Expert® SoftwareFz

Color points by value ofFz:

448.8

118.4


Norm

al %

Pro

babil

ity


-3.00 -2.00 -1.00 0.00 1.00 2.00

1

5

10

2030

50

7080

90

95

99

Fig.1 Normal Probality & Residuals

Fig. 2 explains the comparision of the significant factors with the input parameters. Fig 2.1 and Fig 2.2 explains that the most significant factors for the inrease in the cutting force are feed and depth of cut. Fig 2.3 shows that the significant factor for feed force is feed and as feed increases the feed force also increases. As shown in Fig 2.4 feed is also the most significant factor for increase in the surface roughness.

Fig.3 shows the scanning electron microscope (SEM) images under the different input parameters. SEM images are obtained to study the rake face and cutting edge behaviour for the extreme cutting conditions. Fig.3.1 shows the 14 degrees rake angle with a fresh cutting edge.

The same insert is shown in Fig.3.2 & Fig.3.3 after machining. Fig.3.2 shows the extreme (high levels) coniditions of all the input parameters (cutting speed (140m/min), feed (0.2 mm/rev), depth of cut (1.7 mm) and rake angle (14 degrees)), it can be observed that from Fig.3.2 the formation of built up edge is more and also it can be observed that peeling off of the coating from the rake face has occured resulting in the tool failure. It is also observed from the Fig.3.4 to Fig.3.6 that wear of the nose radius has taken place and also sizeable crater wear is seen on the rake face (Fig.3.5 and Fig.3.6).



Fig. 2.1 Comparision of f & Fz

Fig. 2.2 Comparision of doc & Fz

Fig. 2.3 Comparision of f & Fx

Fig. 2.4 Comparision of f & Ra

Fig. 2 Comparing the significant factors for forces and surface roughness.

Fig. 3.1 SEM micrographs of a fresh cutting edge of 14 degess rake angle cutting tool inserts

Fig. 3.2 SEM micrograph of cutting tool insert under the following cutting conditions: V=140 m/min; f = 0. 2 mm/rev and doc =1.7 mm and 14 degess

rake angle

Fig.. 3.3 SEM micrograph of cutting tool insert under the following cutting conditions: V=100 m/min; f = 0.12 mm/rev and doc =1.1 mm and 14

degress rake angle



Fig. 3.4 SEM micrograph of cutting tool insert under the following cutting conditions: V=100 m/min; f = 0.12 mm/rev and doc =1.7 mm and 0 degess

rake angle

Fig. 3.5 SEM micrograph of cutting tool insert under the following cutting conditions: V=100 m/min; f = 0.2 mm/rev and doc =1.1 mm and 0 degress

rake angle

Fig. 3.6 SEM micrograph of cutting tool insert under the following cutting conditions: V=140 m/min; f = 0.12 mm/rev and doc =1.1 mm and 0 degress

rake angle

B. Surface Roughness and Input Parameters

The normal probability plot of the residuals for surface roughness in dry condition (Ra-D) and the normal probability plot of the residuals for surface roughness in wet condition (Ra-W) is shown in Fig.4. The Figures prove that the residuals lie reasonably close to a straight line, giving support that terms mentioned in the model are the only significant [18]. The final response surface equation for linear model of surface roughness is shown below in coded values.Ra=1.5102-0.01536*V-0.275*feed +0.21471*doc+.0983*V*feed -- (7)

Design-Expert® SoftwareRa

Color points by value ofRa:

0.99724

0.3007


Norm

al %

Pro

babi

lity


-2.00 -1.00 0.00 1.00 2.00 3.00

1

5

10

2030

50

7080

90

95

99

Design-Expert® SoftwareRa

Color points by value ofRa:

0.8835

0.1385


Nor

mal

% P

roba

bilit

y


-2.00 -1.00 0.00 1.00 2.00 3.00

1

5

10

2030

50

7080

90

95

99

Fig.4 Normal Probality & Residuals

Iv. concLusIon

The following main conclusions are drawn from the comparative study of the effect of cutting speed, feed, depth of cut and effective rake angle on the feed force (Fx), thrust force (Fy), cutting force (Fz) and surface roughness (Ra) in



the machining of Ti-6Al-4V using PVD TiAlN coated inserts.• The central composite design is beneficial as it saves

number of experimentations required when compared with the full factorial design for the same factors and for the same levels.

• Linear model is fitted for feed force and surface roughness for dry and wet cutting environment, where as Linear model is fitted for thrust force in dry cutting and quadratic model is fitted in for thrust force in wet cutting environment and 2FI (2 Factor Interaction) model is fitted for cutting force in both the cutting environment.

• For the feed force model: feed is most significant factor in both the cutting environment with 41.04% and 50.47% contribution in the total variability of model whereas depth of cut has a secondary contribution of 5.11% in the model.

• For the thrust force model: the feed and depth of cut are significant factors with 2.12% and 67.39% contribution in the total variability of model, for wet cutting environment where as in dry cutting environment the feed and the depth of cut are significant factor with 1.5% and 66.77% contribution in the total variability of model, respectively.

• For the cutting force model: the feed and depth of cut are the most significant factors affecting cutting force and account for 46.88% and 47.59% contribution in the total variability of model, respectively for wet cutting environment, where as in for dry cutting environment the feed and depth of cut are the most significant factors affecting cutting force and account for 46.88% and 47.59% contribution in the total variability of model, respectively. The interaction between these two provides a secondary contribution of 1.28%.

• For the Surface roughness model: the cutting velocity and the feed provides primary contribution and influences most significantly on the surface roughness.

From conclusions drawn from the analysis of the results for Ti-6Al-4V machining using PVD coated TiAlN inserts the best suited environment for the selected process parameters is wet condition. Such detailed experimental work enable researchers to choose the optimized process parametric conditions including cutting tool geometry (rake angle mainly) to machine Ti alloy material effectively and efficiently without sacrificing on the material removal rate.

reFerences[1] Ramesh, S., L. Karunamoorthy, and K. Palanikumar, “Fuzzy

Modeling and Analysis of Machining Parameters in Machining Titanium Alloy,” Materials and Manufacturing Processes, Vol.23, No.4, pp. 439-447, 2008.

[2] Lutjering G, W.J., Titanium, Springer, Berlin, 2003.[3] Ramesh, S., L. Karunamoorthy, and K. Palanikumar, “Surface

Roughness Analysis in Machining of Titanium Alloy,” Materials and Manufacturing Processes, Vol. 23, No.2, pp. 174-181, 2008.

[4] Bouzakis, K.D., et al., “Application in milling of coated tools with rounded cutting edges after the film deposition,” CIRP Annals - Manufacturing Technology, Vol. 58, No.1, pp. 61-64, 2009.

[5] Corduan, N., et al., “Wear Mechanisms of New Tool Materials for Ti-6AI-4V High Performance Machining,” CIRP Annals - Manufacturing Technology, Vol.52, No.1, pp. 73-76, 2003.

[6] Özel, T., et al., “Investigations on the effects of multi-layered coated inserts in machining Ti–6Al–4V alloy with experiments and finite element simulations,” CIRP Annals - Manufacturing Technology, Vol. 59,No.1, pp. 77-82, 2010.

[7] Shaw, M.C., Metal Cutting Principles, Oxford University Press, Oxford, NY, 1984.

[8] Mathew, J.D., Jr., Titanium: A Technical Guide. ASM International, 1988. for Steel-http://www.engineeringtoolbox.com/thermal-conductivity-d_429.html

[9] Sun, S., M. Brandt, and M.S. Dargusch, “Characteristics of cutting forces and chip formation in machining of titanium alloys,” International Journal of Machine Tools and Manufacture, Vol. 49, No.7-8, pp. 561-568, 2009.

[10] Venugopal, K.A., S. Paul, and A.B. Chattopadhyay, “Growth of tool wear in turning of Ti-6Al-4V alloy under cryogenic cooling”, Wear, Vol. 262,No.9-10, pp. 1071-1078, 2007.

[11] Hong, S.Y., Y. Ding, and W.-c. Jeong, “Friction and cutting forces in cryogenic machining of Ti–6Al–4V,” International Journal of Machine Tools and Manufacture, Vol. 41, No. 15, pp.2271-2285, 2001.

[12] Jawaid, A., S. Sharif, and S. Koksal, “Evaluation of wear mechanisms of coated carbide tools when face milling titanium alloy,” Journal of Materials Processing Technology, Vol.99, No.1-3, pp. 266-274, 2000.

[13] Nalbant, M., et al., “The experimental investigation of the effects of uncoated, PVD- and CVD-coated cemented carbide inserts and cutting parameters on surface roughness in CNC turning and its prediction using artificial neural networks,” Robotics and Computer-Integrated Manufacturing, Vol. 25, No.1, pp. 211-223, 2009.

[14] Yuan, S.M., et al., “Effects of cooling air temperature on cryogenic machining of Ti–6Al–4V alloy,” Journal of Materials Processing Technology, Vol. 211, No.3, pp. 356-362, 2011.

[15] Fang, N. and Q. Wu, “A comparative study of the cutting forces in high speed machining of Ti-6Al-4V and Inconel 718 with a round cutting edge tool,” Journal of Materials Processing Technology, Vol. 209, No.9, pp. 4385-4389, 2009.

[16] Bermingham, M.J., et al., “New observations on tool life, cutting forces and chip morphology in cryogenic machining Ti-6Al-4V,” International Journal of Machine Tools and Manufacture, Vol. 51, No.6, pp. 500-511, 2011.

[17] Chemical Composition ASTM B348 Grade 5. http://www.smithshp.com/downloads/Ti-6Al-4V%20_Grade%205_SHP%20.pdf.

[18] Montogomery, D.C., Design and Analysis of Experiments, 5th ed, John Wiley & Sons Inc, 2001.



Investigation of the Structure and Mechanical Properties of Friction Stir Welded Aluminum Alloy

A. Chandrashekar1*, B. S. Ajay Kumar2, V. Anandkumar3 and P. Raghothama Rao4

1,2Dept. of Mechanical Engg, Bangalore Institute of Technology, Bangalore, India3Dept. of Industrial Engg. & Management, MVJCE, Bangalore, India

4Dept. of Mechanical Engg, SJC Institute of Technology, Chickballapur, India*Correspondence author E-mail: [email protected]

Abstract - Friction stir welding is now extensively used in aluminum industries for Joining and Material processing applications. The Technology has gained increasing interest and importance. In the present paper, the Mechanical and Microstructural properties of AA 5083(AlMg4.5Mn0.7) butt joints produced by Friction Stir Welding have been investigated. Different welding trials with two rotating speed of the tool have been done. The Mechanical properties of the welded joints have been evaluated though micro hardness measurements (HV) and Mechanical Tests. Metallurgical characterization has been done by means of optical microscopy to investigate size, morphology and distribution evolution of the metallic matrix and precipitates present in this type of aluminum alloy. The change in microstructure across the welded joints was characterized by the presence of severely deformed grains in the region of the weld nugget.

Keywords : AA5083, Friction Stir welding, Structural properties, Mechanical testing, Tool Rotating speed

I. IntroductIon

FSW is one of the method in which the heat formed due to friction and plastic deformation is used. The process has been invented and experimentally proven at “The Welding Institute” UK in December 1991[1]. FSW is a technique, which allows aluminum, lead, magnesium, titanium, steel & copper to be welded continuously with a non-consumable tool. A non-consumable rotating tool that stirs the material of welded parts shown in Figure 1, at temperatures well below their melting point produces the joint. Main advantage of FSW is the low welding temperature eliminating many problems of conventional welding processes. Due to the low temperature, materials such as Al-Cu-Mg alloys difficult to weld by fusion processes are easily welded by FSW [2].

A number of different process variables affect the quality of a joint produced by friction stir welding; tool design, tool rotation and travel speed, tool heel plunger depth and tilt angle, welding gap, thickness miss match and plate thickness variation [3]. Higher tool rotation rates generates higher temperature because of higher friction heating and result in more intense stirring and mixing of material.

Fig.1 Friction stir welding process

The design of the tool is a critical factor as a good tool can improve both the quality of weld and the maximum possible welding speed [4]. It is desirable that the tool material is sufficiently strong, tough and hard at the welding temperature. Effect of welding speed on microstructure and mechanical properties of friction stir welded aluminum alloy was investigated by Sakthivel et.al. [5]. The influence of FSW parameters on the grain size of the stir zone and the formability of friction stir welded 5083 aluminum alloys was examined by Tomotake Hirata et.al.,[6]. The Aluminum plates were friction stir welded at various rotational speeds (850-1860rpm) and travel rates of 30 to160 mm/min with welding forces ranging from 2.5 to 10 MPa, using different diameters welding heads was investigated by Wang and Liv [8]. From these experiments it has found that dimensions of the welding head are critical to produce sound weld.

II. experImentaL procedure

A. Fabrication of FSW Tool

FEW tool made of High Speed steel having a pin profile of straight circular was used to weld the joints. Tool has a shoulder of diameter 14mm, a pin diameter of 4mm and a pin length of 4mm.

B.Friction Stir Welding of Aluminum Alloy

The aluminum alloy AA 5083 was selected for friction stir welding process. The chemical composition of Aluminum Alloy AA 5083 is as shown in Table I. Test plates of size 125mm*60mm*6mm were prepared from AA5083 alloy.


Table I ChemICal ComposITIon of The 5083 alumInum alloy (In WT %)

The experimental set up consists of a Friction stir welding machine as shown in Figure 2. The plates are positioned in the fixture, by using mechanical clamps so that the plate will not be separated during welding. The machine can be operated over a wide range of tool rotational speeds, welding speeds & tool axial forces.

Fig.2 Friction stir welding machine

C. To Assess The Mechanical And Structural Properties of FSW Joints

Eight jobs were produced at two different speeds by using FSW tool. The welding parameters are presented in Table II. A sample of a friction stir welded plate is shown in Figure 3.

Fig.3 Friction stir welded joint

Table II WeldIng parameTers of fs WeldIng

1. Tensile Test

The Tensile test specimens were prepared according to the ASTM E8 standard & the transverse tensile properties of the FS welded joints were evaluated using a computerized UTM (Universal Tensile Machine). For each speed welded joint, three tensile specimens were prepared & tested.

2.Structural Analysis

For Microstructural Studies, the samples were polished & etched with chemical solution(Keller’s reagent) that contained 90ml distilled water, 5ml nitric acid, 3ml hydrochloric acid and 2ml hydrofluoric acid for about 80 seconds before being observed under the microscope. The micro structural details of the welded and HAZ were studied with the help of Optical Metallurgical microscope (model: NIKON Epiphot 200).

3.Micro Hardness

Micro hardness values along the cross sections (transverse to weld direction) of samples was measured by using Vickers micro hardness testing machine. Hardness measurements were taken at different points for an applied load 100gms using Vickers micro hardness testing method IS: 1501-2002.

III. resuLts and dIscussIon

A. Tensile Test

A typical three specimens from a job has been drawn to evaluate transverse tensile properties of the welded joint. The specimens prepared for tensile test are shown in Figure 4. The results obtained from those specimens are tabulated and presented in Table III.



Fig.4.Specimen made for tensile test

Table III TensIle TesT resulTs of aa 5083

B. Microstructure for AA5083

Fig.9 The macrostructure of the welded specimen

Figures (5), (6), (7) and (8) show the microstructure of the specimen with the positions of micro-structural zones. Figure (9) show the macrostructure of the specimen. Characteristic structural zones of FSW can be clearly identified from recorded images of microstructure and different zones. Those zones are: unaffected material or parent metal, heat affected zone - HAZ, thermo-mechanically affected zone - TMAZ, and so called “weld nugget” zone – NZ, which provide a clearer view of the observed structure of welded joints, as well as the grain size. Macrostructure consists of fine precipitates of alloying elements along the grain boundary in the matrix of aluminum solid solution, in the base metal and fine columnar grains at the weld zone.


Investigation of the Structure and Mechanical Properties of Friction Stir Welded Aluminum Alloy

C. Micro Hardness Tests

It is observed that the Hardness is minimum at the weld centre i.e. at the centre of the weld nugget. Markings are made at a distance of 10 mm from centre of the weld on either side. Hardness values are shown in table 4. The results of the Vickers micro hardness are presented in Figure 10.

Table IV mICro hardness measuremenT for fsW Welded speCImen

Fig.10 Micro hardness profile of AA 5083

Iv. concLusIon

On the basis of the experimental characterization conducted on FS welded joints of AA 5083, the following Conclusions can be drawn:

• It was found that the weld imperfections significantly reduce the tensile strength and hardness of welded joints.

• Hardness was found to be lower in the weld region compared to TMAZ, HAZ & BM regions.

• Micro-Structure of the parent metal shows dendrites of aluminum solid solution and the grains are fined columnar at weld zone.

acknowLedgements

The authors are grateful to the Management and Department of Mechanical Engineering, Bangalore Institute of Technology, Bangalore, India, for extending the facilities to carry out this Investigation.

reFerences

[1] W M Thomas, I M Norris, D G Stainer and E R Watts, “Friction stir welding development and variant technique”, The SME summit 2005, USA.

[2] Vladvoj Oceneasek, Margarita Slamova, Jorge F Dos Santos and Pedro Vilaca , “Microstructure and properties of FSW aluminium alloys”, Hradec nadmoravici, Metal, pp. 24-26, 2005.

[3] A J Leonard & I A Lockyer , “Flaws in Friction Stir Welds”, 4th International Symposium on FSW, Park city, USA,2003.

[4] Prado R A, Murr L E, Shindo D S, Soto H F,” Tool wear in the friction stir welding of aluminium alloy 6061+20% Al203: A preliminary study”, Scripta Materialia 45, pp. 75-80, 2001.

[5] Sakethivel T, Sengar G S, Mukhopadhyaya J , “Effect of welding speed on micro-structural and mechanical properties on friction stir welded aluminium”, International Journal of Advanced Manufacturing Technology, Vol 43, pp. 468-473, 2009.

[6] Tomotake Hirata, Taizo Oguri, Hideki Hagino, Tsutomu Tanaka, Subg Wook Chung,Yorinobu Takigawa and Kenji Higashi, “Influence of friction stir welding parameters on grain size and formability in 5083 aluminium alloy”, Material Science and Engineering, A 456, pp. 344-349, 2007.

[7] Wang D, Liv S, “Study of FSW of aluminium”, Journal of materials science, Vol.39, pp. 1689-1593, 2004.



Multi-Objective Optimization of the Electro-Discharge Diamond Surface Grinding Process

Shyam Sunder* and Vinod YadavaDepartment of Mechanical Engineering, Motilal Nehru National Institute of Technology,

Allahabad (U.P.) – 211 004, India* Corresponding author E-mail: [email protected]

Abstract - Grinding of Metal matrix composites (MMCs) which are making inroads in various engineering applications have proved to be extremely difficult to machine due to presence of hard ceramic reinforcement. Electro-discharge machining (EDM) of MMCs containing electrically non conducting phases possess few problems in terms of hampering the process stability and impeding the material removal process. Use of combination of grinding and EDM has potential to overcome these problems. This article presents the optimization design of an electro-discharge diamond surface grinding (EDDSG) process performed on aluminum-metal matrix composite (Al-MMC). The major performance characteristics selected to evaluate the process are material removal rate (MRR) and average surface roughness (Ra). The input machining parameters used in the present study were current, pulse on-time, wheel speed, and duty factor. Experiments were carried out on newly self developed surface grinding setup for electro-discharge diamond grinding (EDDG) process for Al-10wt.%SiCp composites. The experimentations are planned as per L9 orthogonal array. Grey relational analysis (GRA) is used for optimizing the machining parameters. Principal component analysis (PCA) is coupled with GRA to evaluate the weighting values corresponding to various performance characteristics that their relative importance can be properly described. The most significant factor has been found as pulse on-time effecting the robustness of electro-discharge diamond surface grinding (EDDSG) process.

Keywords : Electro-discharge diamond surface grinding, Aluminium-metal matrix composites, Grey relational analysis, Principal component analysis

I. IntroductIon

In recent years the critical need for less expensive structural materials that can provide an optimum level of performance has generated considerable research interest in the development and application of metal matrix composites (MMCs). Clyne and Withers [1] discussed that use of MMCs provide significant benefits including performance such as component’s life, and improved productivity. Kannnan and Kishawy [2] mentioned that MMCs provide economic advantage through energy savings or lower maintenance cost and environmental benefits of lower noise levels. Compared to monolithic metals, MMCs have high strength-to-weight ratio, better fatigue resistance, better elevated temperature properties, lower coefficient of thermal expansion, improves thermal conductivity, and excellent wear resistance. However, the utilization of MMCs in different industries is not as generalized as expected due to difficulties in machining of it. Cost effective machining has not been, yet, proven. MMCs

have been successfully applied in aerospace industries since 1970s and in the middle of 1980s these materials reached the automobile industry and nowadays its use is gaining importance [3]. Al2O3 is widely used in mechanical, optical, and microelectronic applications because of its excellent chemical resistance, good mechanical strength, high hardness, transparency, high abrasive and corrosion resistance [4–5].

In traditional machining processes, grinding is one of the viable method of machining because of high dimensional accuracy and surface quality. But grinding of composite materials using conventional surface grinding process shows poor surface finish and accuracy [6]. The decreasing cutting ability of the wheel during the grinding of MMCs may be caused by the following phenomena: (1) break out and fragmentation of grains due to abrasion of reinforcement; (2) attrition wear of the active grains; (3) clogging of the wheel caused by the adherence of the chips. The last two forms of damage determine the formation of wear flats on the wheel surface [7-8]. EDM is an inefficient machining process. Thermal modeling of the process [9] has indicated that the fraction of molten material which is physically not removed but re-deposited on the parent material could be as high as 80%. EDM of composite materials containing electrically non conducting phases possess a few problems. The non-conducting material particles hamper the process stability and impede the material removal process [10]. These problems can be taken care of in EDDG which is a hybrid machining process comprising of diamond grinding (DG) and electro-discharge grinding (EDG). MRR is enhanced as the abrasive grains eradicate the non-conducting material particles, with spark discharges having thermally softened the surrounding binding material.

This hybrid machining process has been developed by combining EDM with metal bonded diamond grinding. In this process, synergetic interaction effect of electro-discharge action and abrasion action are employed to increase the machining performance of constituent processes. The electrical discharges of EDDG cause considerable decrease in grinding forces, and grinding wheel wear; and also effectively re-sharp the grinding wheel. The abrasive action in this process helps to increase material removal rate (MRR) and surface quality.

EDDG can be operated in three different configurations (1) electro-discharge diamond cut-off grinding (EDDCG) (2) electro-discharge diamond face grinding (EDDFG) (3)


electro-discharge diamond surface grinding (EDDSG).

EDDSG is used to machine flat surfaces by using periphery of the metal bonded diamond grinding wheel. Since rectangular workpiece is held in a horizontal orientation, peripheral grinding is performed by rotating the grinding wheel about a horizontal axis perpendicular to the downward motion of servo system. The relative motion of the workpart is achieved by reciprocating the workpiece. While machining the rotating wheel is fed downwards under the control of servo system. The metal bonded grinding wheel and work surface are physically separated by a gap, the magnitude of which depends on local breakdown strength of the dielectric for a particular gap voltage setting. The workpiece is thus simultaneously subjected to heating due to electrical sparks occurring between the periphery of metal bonded grinding wheel and the workpiece, and abrasion action by abrasives of diamond wheel having protrusion height more than the inter-electrode gap (IEG).

Through grey relational analysis (GRA), a grey relational grade is defined as an indicator of the multiple-performance characteristics for evaluation. Lu et al. [11] used GRA coupled with principal component analysis (PCA) to optimize process parameters of high-speed end milling of SKD61 tool steel. Yang et al. [12] employed GRA method to determine optimal machining parameter setting for the end milling of high-purity graphite under dry machining conditions. Most of the researchers used their subjective judgment to establish the weighting values of various performance characteristics to calculate the values of grey relational grade.

Pearson [13] proposed PCA which was subsequently developed as a statistical tool by Hoteling [14]. This approach preserves as much original information as possible by significantly simplifying a large number of correlated variables into fewer uncorrelated and independent principal components. In recent times, PCA has gradually become an analytical tool for the optimization of a system with multiple-performance characteristics [15].

The context is organized in the following manner. Section 2 describes about newly developed experimental setup on EDM machine and Taguchi methodology based experimentation. Section 3 presents optimization using GRA coupled with PCA and finally paper concludes with confirmation tests and summary of this study.

II. detaILs about experImentaL set up and taguchI methodoLogy based experImentatIon

An attachment was developed and mounted on a Smart ZNC die-sinking EDM machine. The EDM machine was supplied by Electronica Machine Tools Ltd. Pune, India.

The metal bonded diamond grinding wheel mounted on the ram of the machine with an attachment. Fig.1a,b [16] respectively shows schematic diagram of electro-discharge diamond surface grinding set-up and dimension details of

fabricated attachment attached to Z axis replacing original tool holder of ZNC EDM machine. The grinding wheel was driven with the help of variable-speed D.C. motor through a belt pulley arrangement. The speed of the motor was varied by changing supply voltage with the help of a variac. The set up consists of a metal bonded diamond grinding wheel, D.C. motor, shaft, pulley,V-belt, bearing etc.

(a)

(b)

Fig.1a,b (a) Schematic diagram of electro-discharge diamond surface grinding set-up and (b) Dimension details of fabricated attachment attached

to Z axis replacing original tool holder of EDM machine

Since the experiment was to be performed in surface grinding mode, so an automatic table feed arrangement was made. The lead screw of the machine table was driven by reversible synchronous motor. Since for automatic to and fro motion of the table motor should automatically rotate both in clockwise and anticlockwise direction as and when it is required, therefore a reversible synchronous motor control circuit was designed using relay switch, two limit switches and regulated power supply. Working of this automatic control is very simple. Suppose the motor is rotating in clockwise direction and as a result table is moving in forward direction. When a lever attached to machine table presses the limit switches, polarity of the motor will automatically changed and motor will start rotating in anticlockwise direction and therefore table will move in reverse direction.



The input process parameters taken are wheel speed, current, pulse on-time, and duty factor. The output parameters analyzed are MRR and Ra. Experiments were performed on aluminum-silicon carbide (Al-SiC) MMC.

Each workpiece was machined for 90 minutes before measuring output parameters. Three repetitions have been done in each set of experiments. Amount of material removal after 90 minutes was obtained by finding weight difference before and after machining using precision electronic digital weight balance with 0.01mg resolution. The MRR is calculated by using the following formula:

where is initial weight of workpiece in gram (before machining); is final weight of workpiece in gram (after machining); t is machining time in minutes. A Talysurf surtronic 25 at 0.8 mm cutoff value was applied to measure the Ra of each machined workpiece. The specification of grinding wheel is shown in Table I.

Table I speCIfICaTIon of grIndIng Wheel

Abrasive DiamondGrain size 80/100Grade M Concentration 75%Bonding material BronzeDepth of abrasive 5 mmWheel diameter 100 mmThickness of wheel 10 mm

Three levels of each process parameters have been selected without considering the interaction effect. The numerical value of process parameters at different levels for machining of Al-SiC composite is shown in Table II.

Pilot experiments were performed to decide the range of parameters. The initial level of process parameters for machining Al-SiC composites is: wheel speed- 1000 RPM, current- 8 A, pulse on-time- 100 µs, and duty factor- 0.578. The experiments were performed as per standard L9 orthogonal array (OA) (Table III). Actual photograph of the setup is shown in Fig.2a,b.

(a) (b)

Fig.2a,b EDSSG set up (a) Fabricated attachment attached to Z axis replacing original tool holder of EDM machine and (b) Fabricated

attachment attached to X axis for an automatic table feed arrangement

Table II maChInIng parameTers and TheIr leVels used In The experImenT for al-10WT%sIC ComposITe

Table III experImenTal obserVaTIons for al-10WT%sIC ComposITe usIng l9 oa

III. optImIzatIon usIng gra coupLed wIth pca

In GRA, when the range of sequences is large or the standard value is large, the function of factors is neglected. However, if the factors measured unit, goals and directions are different, the GRA might produce incorrect results. Therefore, original experimental data must be pre-processed to avoid such effects. Data pre-processing is the process of transforming the original sequence to a comparable sequence. For this purpose, the experimental results are normalized in the range of zero and one, the process is called grey relational generating. Three different types of data normalization according to whether we require the smaller-the-better (SB), the larger-the-better (LB), and nominal-the-better (NB). The normalization is taken by the following equations.

Smaller-the-better (SB)

Larger- the-better (LB)

Nominal-the-better (NB)



Where is normalized value of the kth element in the ith sequence, is desired value of the kth quality characteristic, max is the largest value and min is the smallest value of , , is the number of experiments and is the quality characteristics.

A grey relational coefficient is calculated to display the relationship between the optimal and actual normalized experimental results. The grey relational coefficient can be expressed as

where is the relative difference of kth element between comparative sequence and the reference

sequence , is the absolute value of difference between ,

is a identification coefficient and its value lie between zero and one. In general it is set to 0.5

The average grey relational coefficient is the grey relational grade but, the importance of each quality characteristic may be different The grey relational grade is a weighting-sum of the grey relational coefficients. It is defined as follows:

where represents the weighting value of the kth performance characteristic, and the corresponding weighting values are obtained from the principal component analysis.

A. Principal Component Analysis

Pearson and Hotelling initially developed PCA to explain the structure of variance-covariance by way of the linear combinations of each quality characteristic.

1. The Original Multiple Quality Characteristic Array

X =

where is the number of experiment and is the number

of quality characteristic. X is the grey relational coefficient of each quality characteristic.

2.Correlation Coefficient Array

The Correlation coefficient array is evaluated as follows:

where : the covariance of sequences

and ; : the standard deviation of sequence

; : the standard deviation of sequence .

3. Determining the Eigenvalues and Eigenvectors

The eigenvalues and eigenvectors are determined from the correlation coefficient array,

where eigenvalues,

eigenvectors corresponding to the eigenvalues .

4. Principal Components

The uncorrelated principal component is formulated as:

Where is called the first principal component, is called the second principal component and so on.

The principal components are aligned in descending order with respect to variance, and therefore the first principal component accounts for most variance in the data. In order to objectively reflect the relative importance for each performance characteristic in grey relational analysis, PCA is specially introduced here to determine the corresponding weighting values for each performance characteristic.



Table IV The CalCulaTed grey relaTIonal Co-effICIenT, grey relaTIonal grade and rank

Table V The eIgen Values and explaIned VarIaTIon for prInCIpal ComponenTs

Table IV represent the grey relational coefficient of each performance characteristic. These data are used to evaluate the correlation coefficient matrix, and determine the corresponding eigen values shown in Table V. The eigenvector corresponding to each eigenvalue is listed in Table VI & and its square can represent the contribution of the corresponding performance characteristic to the principal component.

Table VI The eIgenVeCTors for prInCIpal ComponenTs

Table VII shows that the contributions of MRR and Ra are indicated as 0.4999 and 0.4999. Moreover, the variance contribution for the first principal component characterizing the whole original variables, i.e. the two performance characteristic, is as high as 62.76%.Hence for this study, the squares of its corresponding eigenvectors are selected as the weighing values of the related performance characteristic, and the coefficients , in are thereby set as 0.4999 and 0.4999 respectively.

The main effects of each control factor on grey relational grade are given in Table VIII. The use of the grey relational grade to perform the ANOVA analysis is shown in Table IX. Wheel speed, duty factor and pulse on time are the most significant process parameters for affecting the multiple process responses. From the response table for grey relational grade the best combination of input parameters is the set with S2C3T2DF3 i.e., wheel speed at 1200 RPM, current at 24 A,

pulse on-time 150 µs and duty factor 0.697. The percentage contribution of each control factor to the total variance is pulse on-time 31.14%, duty factor 29.08%, wheel speed 28.15%, current 11.63%.

Table VII ConTrIbuTIon of QualITy CharaCTerIsTIC

Iv. conFIrmatIon experIment

Once the optimal level of the machining parameters is identified, the next step is to verify the improvement of the performance characteristics using this optimal combination. Table X compares the results of the confirmation experiment using the optimal machining parameters (S2C3T2DF3) with those of the initial machining parameters (S1C1T1DF1). Three confirmation experiments were conducted at the optimum setting of the machining parameters. The average value of MRR, and Ra at optimum level were found to be 0.137 g/min, and 6.12 μm.The result of confirmation test shows that quality characteristics MRR has been improved considerably, while Ra deteriorates slightly.

Table VIII response Table

Table Ix resulT of anoVa

* Pooled factors



Table x resulTs of ConfIrmaTIon experImenT aT opTImum parameTer leVel (usIng gra and pCa)

v. concLusIons

Based on experiments, results the conclusions are summarized as follows:

1. The principal component analysis, used to determine the corresponding weighting values of each performance characteristics while applying grey relational analysis to a problem with multiple performance characteristics, is proven to be capable of objectively reflecting the relative importance for each performance characteristic.

2. The factors setting found as best combination of process variables is wheel speed- level 2 (1200 RPM), current- level 3 (24A), pulse on-time- level 2 (150 μs) and duty factor - level 3 (0.698). The percentage contribution of each control factor to the total variance is pulse on-time 31.14%, duty factor 29.08%, wheel speed 28.15%, current 11.63%.for simultaneous optimization of MRR. and Ra. Hence, the most significant factor affecting the EDDSG robustness has been identified as pulse on-time.

3. Improvement in MRR by 61.17%, but deterioration in Ra by 2.68% have been found during EDDSG at the optimum parameter setting against initial parameter setting while performing simultaneous optimization of multiple quality characteristics.

reFerences

[1] Clyne T, W., and Withers P. J. (1992), An Introduction to Metal Matrix Composites, Cambridge University Press London.

[2] Kannan S., and Kishawy H. A. (2006), “Surface Characteristic of Machined Aluminum Metal Matrix Composites, International Jourmal of Machime Tools and Manufacture, Vol. 46/15, pp. 2017-2025.

[3] Mohan B., Rajadurai A. and Satyanarayana K. G. (2002), “Effect of SiC and Rotation of Electrode on Electric Discharge Machining of Al-SiC Composite,” Journal of Material Processing Technology, Vol. 124, pp. 297-304.

[4] Handke B., Simonsen J. B. Bech M. Li. Z. and Møller P. J. (2006), “Iron Oxide Thin Film Growth on Al2O3/NiAl,” Surface science, Vol. 600, pp. 5123 - 5130.

[5] Murakami T., Ouyang J. H. Sasaki S. Umeda K. and Yoneyam, Y. (2007), “High Temperature Tribological Properties of Spark-Plasma-Sintered Al2O3 Composites Containing Barite-Type Structure Sulfates,” Tribology International , Vol 40(2), pp. 246-253

[6] Aguair P. R., Dotto F. R. L. and Bianch E. C. (2005), “Study of Thresholds to Burning in Surface Grinding Process,” Journal of the Brazillian Society of Mechanical Sciences and Engineering, Vol. 27(2), pp. 150-156.

[7] Zhu, Y., and Kishawy H. A. (2004), “Influence of Alumina Particles on the Mechanics of Machining Metal Matrix Composites,” International Jourmal of Machime Tools and Manufacture, Vol. 45, pp. 389–398.

[8] Di Ilio A., Paolett, A. Tagliaferrz V. and Venial F. (1996), “An Experimental Study on Grinding of Silicon Carbide Reinforced Aluminium Alloys,” International Jourmal of Machime Tools and Manufacture, Vol. 36/6, pp. 673–685.

[9] Erden A., and Kaftanoglu B. (1981), “Heat Transfer Modeling of Electric Discharge Machining, in: Proc. 21st Int. Mach. Tool Des. Research Conf., London.

[10] Konig W., Dauw D. F. Levy G. and Panten U. (1998), “EDM—Future Steps Towards the Machining of Ceramic,” Ann. CIRP., Vol. 37(2), pp. 623–631.

[11] Lu H. S., Chan C. K. Hwang N.C. and Chung C. T. (2009), “Grey Relational Analysis Coupled With Principal Component Analysis for Optimization Design of the Cutting Parameters in High-Speed End Milling,” Journal of Material Processing Technology, Vol. 209, pp. 3808–3817.

[12] Yang Y.Y., Shi R. and Huang C.H. (2006), “Optimization of Dry Machining Parameters for High Purity Graphite in End-Milling Process,” Mateials and Manufacturing Processes, Vol. 2, pp. 832–837.

[13] Pearson K., (1901), “On Lines and Planes of Closest Fit to Systems of Points in Space,” Philosiphical Magazine series, Vol 62, pp. 559–572.

[14] Hotelling H. (1993), “Analysis of a Complex of Statistical Variables into Principal Components,” Journal of Education Psychology, Vol. 24, pp. 417–441.

[15] Fung H.C., and Kang P.C. (2005), “Multi-Response Optimization in Friction Properties of PBT Composites Using Taguchi Method and Principal Component Analysis,” Journal of Material Processing Technology, pp 602–610.

[16] Yadav S K S., and Yadava Vinod. (2008), “Experimental study and parameter design of electro-discharge diamond grinding,”International Journal of Advanced Manufacturing Technology, pp. 34-42.



Enhancing Wear Resistance of Low Alloy Steel Applicable on Excavator Bucket Teeth Via Hardfacing

Shivali Singla1, Amardeep Singh Kang2* and Jasmaninder Singh Grewal3

1Department of Mechanical Engineering, Baba Hira Singh Bhattal Institute of Engineering and Technology, Lehragaga, Distt. Sangrur, Punjab, India

2Department of Mechanical Engineering, Punjab Technical University, Kapurthala, Punjab, India3 Department of Production Engineering, Guru Nanak Dev Engineering College, Ludhiana, Punjab, India

* Corresponding author E-mail: [email protected]

Abstract - New developments in the field of continuously operating earth moving equipment demand a new way of improving wear resistances of these equipment parts which directly involved with different types of sand and rocks during their operation in harsh field environment. Wear caused by the impact and abrasion action of hard particles is a major problem in the area of earth moving machinery. The objective of this study was to enhance the useful life of the excavator bucket teeth in order to decrease the idle time required to reinstate the teeth periodically during working. The objective was carried out by means of hardfacings, where the effect of the hardfacings on the extent of wear and the wear characteristics of the excavator bucket teeth were examined. Four types of iron-based hardfacing electrodes with a wide range of C (0.75-5% by weight) and Cr (2-33% by weight) were selected to deposit by manual metal arc welding process on the low alloy steel. It was observed that the wear rates of the hardfaced low alloy steel were significantly lower than those of the un-hardfaced steel, indicating a great improvement in the wear protection provided by hardfacings.

Keywords : Excavator bucket teeth, Wear, Hardfacing, Pin-on-disk

I. IntroductIon

Wear is a surface phenomenon and occurs mostly at outer surfaces. Every part that is moving in service will be subject to wear at the contact point with other parts. The consequence of this wear is that the parts need to be replaced, which costs more and causes downtime on the equipment. The surface characteristics of engineering materials have a significant effect on the serviceability and life of a component thus cannot be neglected in design. Abrasive wear produces the premature failure of many components of the extraction machinery with considerable economic costs [1-3]. The ongoing challenge of engineers in these fields is to find or design materials that are the most wear resistant, in order to extend the life of the parts and reduce the frequency of part replacement. Surface engineering is an economic method for the production of materials, tools and machine parts with required surface properties, such as wear resistance [4-5]. Wear related failure of machinery components counts as one of the major reasons for inefficient working of machines in a variety of engineering applications. Many such applications involve handling of abrasive materials or contact with the

material in service. Abrasion is one of the important and commonly observed wear modes in these cases.

Wear resistance of materials can be improved through bulk treatment and surface modification. While bulk treatment has been practiced for a long time, surface treatment is fairly recent and gaining importance. Because wear is a surface phenomenon, it is possible to use a relatively inferior bulk material for a specific (wear related) application by modifying the surface characteristics of the material economically. A variety of techniques/materials exist for modifying the surface properties of substrates. However, their success depends on an appropriate selection of the techniques/materials depending on the application of the modified components. This emphasizes the need to characterize the modified surfaces accordingly [6-7]. Among many proven techniques of surface modification, hardfacing has been especially effective in cases not requiring close dimensional tolerances. Any equipment needs to be maintained properly to work effectively so that it is to regularly inspect for signs of wear, corrosion, fatigue and cracks. The equipment is subjected to various types of wear especially the abrasion wear. The wear causes hundreds of tons of material to be lost and productivity that can never be recovered. This represents significant expenses to companies in the recovery or replacement of these wear prone elements. Surfacing is a process of depositing a material layer over substrate either to improve surface characteristics like corrosion resistance, wear resistance, and hardness, etc. or to get the required size or dimension [8-10]. A variety of bulk materials (ferrous and non-ferrous metals and alloys) can be modified by alloying, mixing, compositing, and coating to achieve adequate resistance to wear corrosion and friction. Hardfacing technique is discussed in the current research. In order to properly understand the hardfacing technique it is necessary to understand the wear that has occurred and its causes.

The purpose of this research was to study the wear characteristics and wear resistance of low alloy steel that were processed with four different commercial hardfacing alloys by shielded metal arc welding (SMAW) and comparing these with the un-hardfaced low alloy steel under laboratory conditions. The wear rates obtained from laboratory can then be used to predict service lives of the excavator teeth.


II. experImentaL procedure

A. Materials and Weld Metal Overlays

Excavator teeth used in the experiment were made from low alloy steel with a composition given in Table I. It can be hardened and tempered to offers good combination of ductility and hardness combined with excellent resistance to shock. The substrate material is the material from which the excavator bucket teeth are made up of. Spectroscopy is done to determine the actual composition of the original bucket teeth. Spectroscopy was done at National Institute of Secondary Steel Technology (NISST), Mandigobindgarh, Punjab (INDIA). From Spectroscopy analysis it was found that the excavator bucket teeth are made up of low alloy steel - 27Mn2 (EN14B).

Table I ChemICal ComposITIon (WT %) of loW alloy sTeel

Selection of the hardfacing electrodes was done on the basis of the chemical composition of the excavator teeth. The actual chemical composition of the bucket teeth had been determined with the help of Optical Emission Spectrometer. The result obtained from the spectroscopic analysis of original bucket teeth helped in the selection of various hardfacing electrodes. Also, from the literature survey it was found that number of alloying elements like Cr and C etc. can be added into the substrate in the form of weld consumables to improve wear resistance. The percentage composition of the four different hardfacing electrodes used in the current research work is as given in Table II.

Table II ChemICal ComposITIon (WT%) of hardfaCIng alloys used In presenT researCh Work

For laboratory tests, samples were prepared in the form of cylindrical pins having final diameter of 8 mm and length of 30 mm with the help of lathe machine, surface and cylindrical grinding machines.

B. Deposition of Hardfacing Layers

Shielded metal arc welding (SMAW) technique was used to deposit the hardfaced layers. Constant current type power source was used, the reason being that with this type of characteristics, the welding current remains constant, irrespective of small variation in arc length and consequent slight change in arc voltage, which are unavoidable even in the case of a skilled worker.

Table III WeldIng parameTers for eaCh hardfaCIng eleCTrode

As the welding current was fairly steady, the weld quality is consistent. The various welding parameters are given in Table III. DC was used for welding because DC has the advantage of two polarities, which means that the electrode can be made negative or positive. Straight polarity (i.e. electrode negative) can be used for SMAW of all steel, but not for most non-ferrous metals. With straight polarity, more of the arc heat is concentrated on the electrode and consequently melting and a deposition rate higher, welding is more rapid and the distortion of work piece is less. On completion of the weld deposits, each test piece was cooled down in air.

III. Laboratory test

The specimens were then subjected to wear tests on a pin-on-disc test apparatus, which is shown schematically in Fig 1.

Fig. 1 Schematic illustration of the pin-on-disc wear test



The pin-on-disk test is generally used as a comparative test in which controlled wear is performed on the samples to study. The mass lost allows calculating the wear rate of the material. Since the action performed on all samples is identical, the wear rate can be used as a quantitative comparative value for wear resistance. The device used was “Wear and Friction Monitor Tester TR-201 of Make-M/S Ducom Instruments Pvt. Ltd., Bangalore-INDIA, conforming to ASTM G99 standard as shown in Fig 2. This testing apparatus is designed to study the wear and friction characteristics in sliding contacts. It is also used for evaluating the rate of wear and ranking of materials. It is fully guarded for operator safety. It is operated with a pin perpendicular to the flat circular disc. The sliding path is a circle on the disc surface.

Fig. 2 Wear Testing Apparatus as per ASTM G99 standards

Three samples of each hardfacing i.e. H 1, H 2, H 3 and H 4 on EN-14B were subjected to wear on Pin-On-Disc wear test rig at normal loads of 30N, 40N and 50N respectively. Three samples of EN-14B substrate were also subjected to wear on Pin-On-Disc wear test rig at the same loads. The comparisons of wear with respect to sliding distance at different values of load acting have been made by plotting the graphs of different hardfaced and unhardfaced samples.

The Fig 3 shows the comparison of weight loss between hardfaced and un-hardfaced samples at load of 30N.

Fig. 3 Variation of cumulative weight loss with the sliding distance at 30 N subjected to dry sliding wear

It shows that H 1 hardfacing alloy is at the lowest position on the graph and un-hardfaced sample at the top most position on graph and it can be seen that the wear w.r.t. sliding distance lowest slope for H 1 alloy. The wear goes on increasing w.r.t. sliding distance for all the samples. The curves of H 2 and H 3 are approximately overlapping each other. Similarly the comparison for wear between hardfaced and un-hardfaced samples has also been made from the graph of wear with respect to sliding distance at 40 N and 50 N load acting (Fig 4 & Fig 5).

Fig. 4 Variation of cumulative weight loss with the sliding distance at 40 N subject to dry sliding wear

Fig. 5 Variation of cumulative weight loss with the sliding distance at 50 N subject to dry sliding wear

Iv. concLusIons

Based upon experimental results obtained in the present work, the following conclusions have been drawn:

• H 1, H 2, H 3 and H 4 hardfacing alloys have successfully been deposited on EN-14B substarte using SMAW process.

• The specimen’s hardfaced with H 1, H 2, H 3 and H 4 alloy on low alloy steel showed significantly lower


Enhancing Wear Resistance of Low Alloy Steel Applicable on Excavator Bucket Teeth Via Hardfacing

cumulative weight loss as compared to uncoated EN-14B substrate.

• Cumulative weight loss for hardfaced and un-hardfaced (EN-14B) specimens increases with increase in load.

• The cumulative weight loss for H 1 hardfacing alloy was observed to be minimum in the present study.

• The wear resistances for EN-14B, hardfaced with H 1, H 2, H 3 and H 4 alloys followed a general trend as given below: H 1 > H 2 > H 3 > H 4 > Low alloy steel.

reFerences

[1] D. Dowson, History of Tribology, 2ndEdition, 1998.[2] J. Neale, Component Failures, Maintenance and Repair: A Tribology

Handbook, 1995.[3] W. Scotts, Proceedings of the International Conference on Tribology

in Mineral Extraction, Nottingham, 1984.

[4] D.L. Olson, C.E. Cross, Friction and Wear in the Mining and Mineral Industries, ASTM Handbook, Vol. 18, Center for Welding and Joining Research, Colorado School of Mines, USA, pp.649–655.

[5] E. N. Gregory, “Surfacing by welding,” Weld. Inst. Res. Bull. 21 (1) (1980) 9–13.

[6] G. Schmidt and S. Steinhauser, “Characterization of Wear Protective Coatings”, Tribo.Int” 1996, 29, pp. 207-14.

[7] M. F. Buchely, J. C. Gutierrez, L. M. Leon, A. Toro, “The effect of microstructure on abrasive wear of hardfacing alloys”, Wear, 2005, 259, pp 52-61.

[8] Kirchgaßner, M., Badisch, E., Franek, F, “Behaviour of iron-based hardfacing alloys under abrasion & impact”, Wear, 2008, Vol 265(5-6), pp. 772-777.

[9] Dasgupta, R.; Prasad, B. K.; Jha, A. K.; Modi, O. P.; Das, S.; Yegneswaran, A. H,“Low stress abrasive wear behavior of a hardfaced steel”, Journal of Materials Engineering and Performance, Vol. 7, No. 2, pp.221-226.

[10] W.Wo, L.-T.Wu, “The wear behavior between hardfacing materials”, Metall. Mater. Trans. A27A (1996) 3639–3648



Creep Modeling in An Orthotropic FGM CylinderAshish Singla1, Manish Garg2, Dharmpal Deepak1 and V. K. Gupta1*

1Department of Mechanical Engineering, University College of Engineering, Punjabi University Patiala, Punjab, India2Department of Physics, Punjabi University Patiala, Punjab, India

*Corresponding author E-mail: [email protected]

Abstract - A mathematical model has been developed to estimate steady state creep in an orthotropic cylinder made of functionally graded composite. The FG cylinder is assumed to undergo creep according to Power law. The model developed has been used to investigate the steady state creep response of the FGM cylinder for varying orthotropicity of the material. The results obtained are compared with those estimated for a similar FGM cylinder but having isotropic properties. The result reveals that the presence of orthotropicity in the FGM cylinder may significantly modify its creep response.

Keywords : Modeling, Steady state creep, Cylinder, Functionally Graded Material, Orthotropic

I.IntroductIon

Functionally graded materials (FGMs) are microscopically inhomogeneous composite materials in which the volume fraction of two or more materials is varied smoothly and continuously as a function of position in certain direction(s) of the structure from one point to another [6, 13]. These materials are mainly constructed to operate in high temperature environments and are made from a mixture of metal and ceramic or a combination of different metals. FGMs have been developed as ultra high temperature resistant materials for potential applications in aircrafts, space vehicles and other structural components exposed to elevated temperature [5].

Bhatnagar et al. [2] have presented the analysis for an orthotropic thick-walled cylinder undergoing creep due to the combined action of internal and external pressures, and rotary inertia. It is observed that the cylinder with more strength in the radial direction has lower effective stress and performs better. Gupta et al. [4] analyzed creep stresses and strain rates in a rotating non-homogeneous thick-walled cylinder by using Seth’s transition theory. The study indicates that for a rotating non-homogeneous cylinder, with compressibility increasing radially, the circumferential stress is maximum at the external surface at lesser angular speed but at higher angular speed it becomes maximum at the internal surface. The compressive value of axial stress, observed at the external surface, increases with the increase in angular speed. Chen et al. [3] analyzed creep behavior of an FGM cylinder subjected to both internal and external pressures. It was assumed that the properties of graded material are axi-symmetric and depend on the radial coordinate. An asymptotic solution was derived on the basis of Taylor expansion series. The approximate solutions calculated by taking different higher-order terms in the Taylor series were compared with the results of finite element (FE) analysis performed in ABAQUS software. It

is observed that although the use of higher-order terms may help to obtain a more accurate result for the time-dependent behavior of the cylinder, a fifth-order form is sufficiently accurate to calculate the distribution of creep stress with satisfactory approximation. Singh and Gupta [9] investigated the steady state creep in a transversely isotropic functionally graded cylinder operating under internal and external pressures. They described creep behavior of the cylinder by a threshold stress based creep law. The effect of anisotropy was investigated on the creep stresses and creep rates in the FGM cylinder. The study reveals that in the presence of anisotropy, the radial and tangential stresses are marginally affected whereas the axial and effective stresses vary significantly. The strain rates as well as strain rate inhomogeneity decrease significantly when the extent of anisotropy (α) reduces from 1.3 to 0.7. Sadeghi et al [7] carried out strain gradient elasticity formulation to analyze FG micro-cylinders. The material properties were assumed to obey a power law distribution in the radial direction. A power series solution for stresses and displacements in FG micro-cylinders subjected to internal and external pressures was obtained. Numerical examples were presented to study the effect of characteristic length parameter and FG power index on the displacement field and stress distribution in the FG cylinders. It is observed that the characteristic length parameter has a considerable effect on the stress distribution of FG micro-cylinders. The increase of material length parameter leads to decrease the maximum radial and tangential stresses in the cylinder. The study also reveals that the FG power index has a significant effect on the maximum radial and tangential stresses.

The literature consulted so for reveals that a number of studies have been undertaken to investigate the creep behavior of composite cylinders. The studies pertaining to creep behavior of FGM cylinder are rather limited. Further, the studies on FGM cylinders, however limited, assume the material to be isotropic. In actual, the FGMs are anisotropic in nature. The anisotropy may be induced during processing such as forging or due to initial creep deformation. Therefore, it is imperative to consider the effect of anisotropy on the creep behavior of the FGM cylinder subjected to thermo-mechanical loading.

II. dIstrIbutIon oF reInForcement

The cylinder is made of Al-SiCp composite with SiCp content decreasing linearly from the inner to outer radius. Therefore, the density and the value of creep parameters B and n will vary with the radial distance. The content (vol. %)


of SiCp, V(r), at any radius r of the cylinder is given by Singh and Gupta [9],

( ) ( )( ) [ ]minmaxmax VV

abarVrV −

−−−=

(1)

Where maxV and minV are respectively the maxim

um (at the inner radius) and t he minimum (at the outer radius) SiCp content in the cylinder.

The average SiCp content in the cylinder can be expressed as,

(2)

Where l is the length of cylinder.

Substituting V(r) from Eq. (1) into Eq. (2) and integrating, we get,

( )( ) ( )3

32max

2

min32

231113

λλ

λλλλ

+−

+−−−−=

VVV avg

(3)

Where )( ba=λ is the ratio of inner to outer radius of the cylinder.

III. creep Law and parameters

The creep behavior of the FGM cylinder is described by Norton’s power law.

nee Bσε = (4)

Where eε is the effective strain rate, eσ is the effective stress, B and n are material parameters describing the creep performance in the cylinder.

It is evident from the study of Singh and Ray [8] that the values of creep parameters B and n appearing in Norton’s law depend on the content of reinforcement, which vary with the radial distance. It is also revealed that the effect of varying SiCp content on the creep parameters B and n is opposite to each other. The value of B decreases with the decrease in SiCp content but the value of n increases with the decrease in SiCp content. In the light of this, the values of Power law multiplier (B) and stress exponent (n) appearing in the creep law (Eq. 4), at any radius r of the FGM cylinder are estimated by following equations.

[ ]φavgo VrVBrB )()( = (5)

and

[ ] φ−= avgo VrVnrn )()( (6)Where Bo and no are respectively the values of creep

parameters B and n respectively and ϕ is the grading index. The values of Bo, n and ϕ are the taken from the study of Chen et al. [3].

Table I Values of Creep parameTers [3] and dImensIon of The model

Iv. mathematIcaL FormuLatIon

Let us consider a thick-walled hollow cylinder made of functionally graded Al-SiCp composite having inner and outer radii as a and b respectively. The cylinder is subjected to internal and external pressures denoted respectively by p and q.

For the purpose of analysis the following assumptions are made:

(i)The material of the cylinder is orthotropic and

incompressible i.e. 0=++ zr εεε θ

where r, θ and z are taken respectively along the radial, tangential and axial directions of the cylinder.

(ii) The cylinder is subjected to internal pressure that is applied gradually and held constant during the loading history.

(iii) Elastic deformations in the cylinder are neglected as compared to creep deformations.

The cylinder is sufficiently long and hence is assumed under plain strain condition (i.e. axial strain rate, 0=zε )

The radial ( rε ) and tangential ( θε ) strain rates in the cylinder are given by:

(7) and rur

=θε (8)

Where is the radial displacement rate and u is the radial displacement.

Eqs (7) and (8) may be solved to get the following compatibility equation,

(9)

The cylinder is subjeconditions,



pr −=σ at ar = (10)

qr −=σ at br = (11)

Where the negative sign of rσ implies the compressive nature of radial stress.

By considering the equilibrium of forces acting on an element of the cylinder in the radial direction, we get,

(12)

Since the material of the cylinder is incompressible, therefore,

0=++ zr εεε θ (13)

The constitutive equations under multi axial creep in an orthotropic cylinder, when the principal axes are the axes of reference, Bhatnagar and Gupta [1] are given by,

( ) ( ) ( )[ ]θσσσσσ

εε −+−+

= rzre

er HG

HG

(14)

( ) ( ) ( )[ ]rze

e HFHG

σσσσσ

εε θθθ −+−+

=

(15)

( ) ( ) ( )[ ]rzze

ez GF

HGσσσσ

σεε θ −+−

+=

(16)

Where F, G and H are the anisotropic constants, eε and

eσ are respectively the effective strain rate and effective stress in the FGM cylinder.

The Hill’s yield criterion, when the Principal axes of anisotropy are the axes of reference, Dieter [11], is given by,

( ) { }21

222 )()()(1

−+−+−

+= θθ σσσσσσσ rrzze HGF

HG (17)Under plain strain condition ( 0=zε ), one may get from

Eqs. (7), (8) and (13),

rCur =

(18)

Where C is a constant of integration. Using Eq. (18) in Eqs. (7) and (8), we get,

2rC

r −=ε (19) and 2r

C=θε (20)

Under plane strain condition, Eq. (16) becomes,

)()(

GFFG r

z ++

= θσσσ

(21)

Substituting zσ from Eq. (21) in to Eq. (17), we get,

(22)

Substituting rε and zσ respectively from Eqs. (19) and (21) into Eq. (14), we obtain,

(23)

Using Eqs. (4) and (22) in Eq. (23) and simplifying, one gets,

nr

r

I/21=−σσθ (24)

Where, (25)

Substituting Eq. (24) into Eq. (12) and integrating, we get,

pXr −= 1σ (26)

Where, (27)

Substituting Eq. (26) into Eq. (24), we obtain,

pr

IX

n−+=

/21

1θσ (28)

To estimate the value of constant C, needed for estimating , the boundary conditions given in Eqs. (10) and (11) are used in Eq. (26) with X1(Eq. 27) integrated between limits a to b. to get,

qpdr

r

Ib

a nn

−=−∫ +21

(29)

Substituting the value of 1I from Eq. (25) in to Eq. (29) and simplifying, we obtain,

n

XqpC

−=

2 (30)

Where, and (31)

Using Eqs. (21) and (22) into Eqs. (14) and (15), one obtains,


Creep Modeling in An Orthotropic FGM Cylinder

(32)

The analysis presented above yields the results for isotropic FGM cylinder. When the anisotropic constants are set equal i.e. F=G=H.

v. estImatIon oF anIsotropIc constants

The Hill’s yield criterion for orthotropic material, as given by Eq. (17), involves constants F, G and H, the values of which are required for estimating involve creep response of the FGM cylinder. If X, Y and Z are the tensile stresses in the principal directions of anisotropy, then according to Hill, Dieter[11].

+=+=+= GF

ZFH

YHG

X 2221;1;1

(33)

The above set of equations may be solved to estimate the values of anisotropic constants as given below,

−+=

−+=

−+=

222

222222

1112

;1112;1112

ZYXH

YXZG

XZYF

(34)

For isotropic material the ratio of anisotropic constants is unity i.e. F/G = G/H = H/F = 1.

If the material of cylinder is subjected to uniaxial loading in r and θ directions, the corresponding stress invariant may be expressed in terms of observed tensile strength and Hill’s anisotropic constants as given below,

yreHG σσ

2+=

(35) and y

HFe θσσ

2+=

(36)

where YYr θσσ , are respectively the yield strength of

composite in r and θ directions and σe is the isotropic yield stress. If the material of cylinder is tested under uniaxial loading in z direction, the stress invariant may similarly be written as,

yzeGF σσ

2+= (37)

where yzσ is the yield strength of composite in z

direction.

It is assumed that during processing of FGM cylinder the whiskers get aligned in the tangential (θ) direction, leading to anisotropic behavior. Therefore, in FGM cylinder the direction θ becomes longitudinal direction and the remaining directions (i.e. r and z) may be taken as transverse directions. For axisymmetric problems like cylinder, the directions r,

θ and z may be taken as the principal directions. Thus, the anisotropic constants given by Eqn. (34), may be expressed as,

2222

111e

rz yyy

F σσσσθ

−+=

(38)

2222

111e

rz yyy

G σσσσ θ

−+=

(39)

2222

111e

zr yyy

H σσσσ θ

−+=

(40)

When assume the yyr θσσα = and

yyz θσσβ =,The

value of G/F > 1 and H/F < 1 implies that the yield strength of FGM cylinder in the tangential direction is the highest and lowest in the axial direction. On the contrary, G/F < 1 and H/F > 1 imply that the yield strength of the cylinder is the highest in tangential direction but lowest in the tangential direction.

To study the effect of anisotropy on the stress and strain rates, following numerical values of anisotropic constants taken from Kulkarni et al. [10] has been used.

Table II Values of anIsoTropIC ConsTanTs Taken from [10]

vI. numerIcaL scheme oF computatIon

To begin the computation process, the vlaue of X2 is estimated from Eq.(31) after substituting the value of anisotropic constants F,G and H and the values of creep parameters B and n, as estimated from Eqs. (5) and (6) respectively.Thereafter,the value of constant C is estimated from Eq. (30) and using this in Eq.(25) the value of I1 is obtained. The value of I1, thus estimate is used in eq. (27) to

get X1.Knowing X1, the stresses rσ and θσ are obtained respectively from Eqs. (26) and (28).The values of and are substituted in Eq. (21) to estimate the distribution of axial

stress ( zσ ) in the cylinder. Knowing rσ , θσ and zσ , the values of and are obtained respectively from Eqs. (4)

and (22). Finally, the strain rates rε and θε are estimated respectively from Eqs. (14) and (15).



vII. resuLts and dIscussIon

A. Validation

Before discussing the results obtained in this study, it is considered necessary to validate the analytical procedure used in this study. The values of B and n are assumed to be constant for the cylinder as 2.77×10-16 MPa-n/h and 3.75, similar to the work of Chen et al.[3].The distribution of tangential stress estimated in the cylinder is compared with those reported by Chen et al.[3], in Fig. 1. An excellent agreement is observed between these results, validating the present study.

Fig. 1 Validation of Chen et al. [3] vs. Analytical

B. Variation of Creep Parameters

Figure 2 shows the variation of creep parameters B and n with radial distance in FGM cylinders. The value of parameter B and n in the FGM cylinder is supposed to decrease and increase respectively with increase in radial distance, as is evident from the Eqs. (5) and (6). The variations observed in parameters B and n are attributed to decreasing SiCp content, V(r), in the FGM cylinders with increasing radius (r), as evident from Eq. (1). Owing to similar distribution of reinforcement (SiCp) in the different FGM cylinders C1-C3, they have similar variations of parameters B and n.

Fig. 2: Variation of creep parameters B and n in cylinder

C. Effect of Anisotropy on Stresses and Strain Rates

Figure 3 shows the effect of anisotropy on radial stress in the FGM cylinders. The radial stress remains compressive over the entire cylinder with a maximum (compressive) and zero values reported at the inner and outer radii respectively, under the imposed boundary conditions given in Eqs. (10) and (11). the results obtained through analytical technique are not affected by varying degree of anisotropy in the FGM cylinders. This is attributed to the fact that the term X1 used in Eq. (26) for calculating radial stress is not affected by varying the extent of anisotropy in the FGM cylinders.

Fig.3 Effect of anisotropy on radial Stress

Figure 4 shows the effect of anisotropy on tangential



stress in the FGM cylinder. The tangential stress remains tensile throughout the FGM cylinders and is observed to decrease with the increase in radius. The results obtained by analytical procedure reveals that the tangential stress in the FGM cylinder is not affected by varying the extent of anisotropy in the FGM cylinders. This is attributed to the fact that the terms X1 and I1/r2/n, used in Eq. (28) for calculating tangential stress, are not dependent on the extent of anisotropy.

Fig. 4 Effect of anisotropy on Tangential Stress

Figure 5 shows the effect of anisotropy on effective stress in the FGM cylinder. The effective stress decreases with increasing radial distance. The results of analytical procedure reveal that the effective stress is observed to be minimum for FGM cylinder C1 and maximum for FGM cylinder C3 when compared with the isotropic FGM cylinder C2.

Fig. 5 Effect of anisotropy on Effective Stress

Figure 6 shows the effect of anisotropy on radial and tangential strain rates in FGM cylinders. The radial and tangential strain rates in the cylinder are equal in magnitude but opposite in nature under the assumptions of incompressibility (Eq. 13) and plain strain condition ( 0=zε ). The effect of anisotropy on radial and tangential strain rates in the FGM cylinder decreases with increasing radius. The radial strain rate is the lowest in FGM cylinder C1 and the highest in FGM cylinder C3 when compared to isotropic FGM cylinder C2. The effect of anisotropy on effective strain rate (Fig.7) is observed to be similar as for tangential strain rate in Fig.6.

Fig. 6 Effect of anisotropy on radial and tangential strain rate

The effect of anisotropy on effective strain rate in the cylinder similar those described for radial and tangential strain rates (refer Fig. 6).

Fig. 7 Effect of anisotropy on effective strain rate



vIII. concLusIons

The present study has led to the following conclusions:

1. The effect of anisotropy on radial and tangential stresses increases near the inner radius but exhibit are decrease towards the outer radius.

2. The effective stresses in the FGM cylinder with yyy zr σσσθ >> is lower everywhere as compared to any

other FGM cylinder. The effect of anisotropy on the effective stress decreases with increasing radius.

3. The strain rates (radial, tangential and effective) in the FGM cylinder is the lowest for the FGM cylinder with

yyy zr σσσθ >> and the highest for the FGM cylinder with yyy rz θσσσ >> .

4. The effect of anisotropy on strain rates decreases with increasing radius.

reFerences

[1] Bhatnagar, N.S. and Gupta, S.K., (1969), “Analysis of thick-walled orthotropic cylinder in the theory of creep”, Journal of the Physical Society of Japan, Vol. 27, No. 6, pp. 1655- 1662.

[2] Bhatnagar, N.S., Kulkarni, P.S. and Arya, V.K. (1986), “Creep analysis of orthotropic rotating cylinder considering finite strains, International Journal of Non-Linear Mechanics, Vol. 21, No. 1, pp. 61–71.

[3] Chen, J.J., Tu, S.T., Xuan, F.Z. and Wang, Z.D. (2007), “Creep analysis for a functionally graded cylinder subjected to internal and external pressure”, Journal of Strain Analysis of Engineering Design, Vol. 42, No. 2, pp. 69-77.

[4] Gupta, S.K., Sharma, S. and Pathak, S. (2000), “Creep transition in non-homogeneous thick walled rotating cylinders”, Indian Journal of Pure and Applied Mathematics, Vol. 31, No. 12, pp. 1579–1594.

[5] Noda, N., Nakai, S., Tsuji, T. (1998), “Thermal stresses in functionally graded materials of particle-reinforced composite”. JSME International Journal Series, Vol. 41, No.2, pp. 178-184.

[6] Reddy, J.N. (2000), “Analysis of functionally graded plates”, International Journal for Numerical Methods in Engineering, Vol. 47, No.1-3, pp. 663–684.

[7] Sadeghi, H., Baghanib,M. and R. Naghdabadi (2012), “Strain gradient elasticity solution for functionally graded micro-cylinders”, International Journal of Engineering Science, Vol. 50, No.1, pp. 22–30.

[8] Singh, S.B. and Ray, S. (2001), “Steady state creep behavior in an isotropic functionally graded material rotating disc of Al-SiC composite”, Metallurgical and Materials Transactions, Vol. 32A, No. 7, pp.1679–1685.

[9] Singh, T. and Gupta, V.K. (2011), “Effect of anisotropy on steady state creep in functionally graded cylinder”, Composite Structures, Vol. 93, No. 2, pp. 747-758.

[10] Kulkarni, P. S., Bhatnagar, N. S. and Arya, V. K.(1985), “Creep analysis of thin-walled anisotropic cylinders subjected to internal pressure, bending and twisting”, Proceedings of the workshop on solid mechanics, pp. 13-16.

[11] Dieter, G.E. (1988), Mechanical Metallurgy, 3rd ed. London, McGraw-Hill Publications.

[12] Shen Y.L. (2010), Constrain deformation of materials: devices, heterogeneous structures and Thermo-mechanical modeling, Springer Publication, Ist edition, p.no.179.

[13] Suresh, S., Mortensen, A. (1998), Fundamentals of functionally graded materials: processing and thermomechanical behavior of graded metals and metal-ceramic composites, IOM Communications Limited, London.



Documents

ARME Vol-1 No-2 pdf