+ All Categories
Home > Documents > Pipe Line Socket Thesis

Pipe Line Socket Thesis

Date post: 14-Apr-2018
Category:
Upload: tania-joshi-sharda
View: 224 times
Download: 0 times
Share this document with a friend
81
-1- MODIFICATION OF CORE -SHAPE FOR IMPROVEMENT OF PRODUCTIVITY OF MALLEABLE IRON PIPE FITTINGS (SOCKET 1", 1 " ) THESIS Submitted in partial fulfilment for the award of the deg ree of MASTER OF TECHNOLOGY In PRODUCTION ENGINEERING Submitted By Krishan Dev 100628181031  PUNJAB TECHNICAL UNIVERSITY JALANDHAR  
Transcript

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 1/81

-1-

MODIFICATION OF CORE -SHAPE FOR 

IMPROVEMENT OF PRODUCTIVITY OF

MALLEABLE IRON PIPE FITTINGS

(SOCKET 1", 1

"

)

THESIS

Submitted in partial fulfilment for the award of the degree of 

MASTER OF TECHNOLOGY

In

PRODUCTION ENGINEERING

Submitted By

Krishan Dev 

100628181031 

PUNJAB TECHNICAL UNIVERSITY 

JALANDHAR  

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 2/81

-II-

MODIFICATION OF CORE -SHAPE FOR 

IMPROVEMENT OF PRODUCTIVITY OF

MALLEABLE IRON PIPE FITTINGS

(SOCKET 1", 1

"

)

THESIS

Submitted in partial fulfilment for the award of the degree of 

MASTER OF TECHNOLOGY

In

PRODUCTION ENGINEERING

Submitted By: Under Guidance of:

Krishan Dev Er. Gautam Kocher

Roll No. 100628181031 Asst. Prof.

Production Engg. Deptt.

R.I.E.T Phagwara, Punjab (INDIA) 

Department of Mechanical Engineering

R AMGARHIA INSTITUTE OF ENGG. & TECHNOLOGY 

PHAGWARA.

AFFILIATED TO PUNJAB TECHNICAL UNIVERSITY, JALANDHAR 

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 3/81

-III-

DECLARATION

I hereby certify that the thesis titled: Modification of Core-Shape for Improvement of 

Productivity of Malleable Iron Pipe Fittings (Socket 1ʺ, 1 ʺ) in partial fulfilment for 

the Award of Degree of  Master of Technology in Production Engineering at

Ramgarhia Institute of Engineering & Technology, Phagwara is an authentic record of 

my work carried out under the supervision of Er.Gautam Kocher, Assistant Professor,

Mechanical Engineering Department Ramgarhia Institute of Engineering & Technology,

Phagwara. I have not submitted the matter presented in the thesis anywhere for the award

of degree.

(Krishan Dev)Roll No. 100628181031

CERTIFICATE

This is to certify that the above statement made by the candidate is correct to the best of 

my knowledge and belief.

Er.Gautam Kocher

Assistant Professor  

The Viva-Voce examination of Krishan Dev (Roll No.100628181031) has been

conducted successfully on __________________ 

Supervisors HOD External

Examiner 

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 4/81

-IV-

ABSTRACT

The term pipe fittings include a number of fittings e.g. elbow, tee, socket, union etc.

These fittings should have chamfering at outlet. The chamfering is provided to assist

assembly and prevent damage to the start of thread. The chamfering should have an

included angle of 900

± 50. Presently there is practice of chamfering malleable iron pipe

fittings after these have been galvanized.

In this work, new shaped cores for malleable iron sockets 1ʺ, 1 ʺ have been

developed. With the use of these new cores, we get sockets after casting which don’t

need chamfering anymore. The occurrence of chamfering during casting process savesthe valuable time required for chamfering operation. The sockets casted by using these

new cores are lighter in weight than the existing sockets. Other dimensions of new

sockets are kept unchanged. Thus the number of sockets that can be made from the same

amount of molten metal has increased. Thus the scrap production due to chamfering

operation has reduced. The labour cost of chamfering has eliminated due to use of these

new cores. Thus the productivity has increased due to reduction in time of production,

scrap production and labour cost.

Keywords: Chamfering, Malleable, Core, Pipe Fittings.

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 5/81

-V-

ACKNOWLEDGEMENT

I am heartily grateful to the supreme power almighty with whose blessings I have been

able to complete this study.

I owe my profound gratitude to my guide Er.Gautam Kocher Assistant Professor

Mechanical Engineering Department of Ramgarhia Institute of Engineering. &

Technology Phagwara for his inspiring and pains taking supervision, valuable guidance,

suggestions at every stage of my thesis. Without his critical review and valuable

discussions this work could not have been produced to its present form.

I would like to acknowledge the cooperation and help rendered by Er. R.K.Dhawan,

Principal and Er. Harvinder Lal Assistant Professor, Ramgarhia Institute of Engg. &

Technology.

I am also thankful to the proprietors of B.N. Industry, Jalandhar city for providing me

infrastructural facilities.

KRISHAN DEV

Roll No. 100628181031

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 6/81

-VI-

TABLE OF CONTENTS

Topic Page

Declaration & Certificate III

Abstract IV

Acknowledgement V

Table of Contents VI-VII

List of Figures VIII-IX

List of Tables X

List of Publications XI

Chapter 1: Introduction 1-28

1.1 Process of manufacturing of G.I. Malleable Pipe fittings 2

1.2 Terminology 16

1.3 Types of Fittings 17

1.4 Core 19

1.5 Core Materials 20

1.6 Core Sands Which Require Heat Treatment 21

1.7 Core Sand Which Do Not Require Heat Treatment 23

1.8 Washes, Pastes, Powders and Other Dressings 24

1.9 Core Frames 27

1.10 Machine Core making 28

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 7/81

-VII-

Chapter 2: Literature Review 29-35

Chapter 3: Problem Formulation 36

Chapter 4: Experimental Work 37-53

4.1  Experimental Introduction 37

4.2 Work Place 37

4.3 Procedure for making core box, Pattern of socket 37

4.4 Manual Core making 40

4.5 Procedure for making moulding sand, mould and casting of socket 43

4.6 Tools & Equipments Required 47

Chapter 5: Results 54-67

5.1 Analysis 61

5.2 Discussion 63

Chapter 6: Conclusion 68

6.1 Scope of Improvement 68

References 69-70

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 8/81

-VIII-

LIST OF FIGURES

Figure No. Page

Figure 1.1 Pipe Fittings 1

Figure 1.2 Induction Furnace 3

Figure 1.3 Annealing Furnace 4

Figure 1.4 Equipment for Compression 5

Figure 1.5 Grinding Machine 6

Figure 1.6 Shot Blasting Machine 7

Figure 1.7 Dry Galvanizing Process 12

Figure 4.1 Core Box 37

Figure 4.2 Core Box 1ʺ (sectional view) 38

Figure 4.3 Core Box 1 ʺ (sectional view) 38

Figure 4.4 Pattern of Socket 39

Figure 4.5 New Cores 40

Figure 4.6 New Cores of Socket 1ʺ 41

Figure 4.7 Existing Core of Socket 1ʺ 41

Figure 4.8 Existing Core of Socket 1 ʺ 42

Figure 4.9 New Core of Socket 1 ʺ 42

Figure 4.10 Making of Moulds 44

Figure 4.11 New Pieces of Socket 45

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 9/81

-IX-

Figure 4.12 Shovel 47

Figure 4.13 Trowel 48

Figure 4.14 Trowel 48

Figure 4.15 Lifter 48

Figure 4.16 Lifter 48

Figure 4.17 Strike Off Bar 48

Figure 4.18 Vent Wire 49

Figure 4.19 Draw Spike 49

Figure 4.20 Slick 49

Figure 4.21 Sprue Cutter 50

Figure 4.22 Bellow 50

Figure 4.23 Rectangular Flasks 51

Figure 4.24 Handle Ladle 51

Figure 4.25 Handle Ladle 52

Figure 4.26 Muller 52

Figure 4.27 Rotary Furnace 53 

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 10/81

-X-

LIST OF TABLES

Table No. Page

Table 1.3.1 Types of fittings 17

Table 1.3.2 Details of wall Thickness 19

Table 4.3.1 Dimensions of Pattern of socket1ʺ 39

Table 4.3.2 Dimensions of Pattern of socket1 ʺ 39

Table 4.4.1 Materials Required for core 40

Table 4.4.2 Dimensions of new core of socket 1ʺ 41

Table 4.4.3 Dimensions of existing core of socket 1ʺ 42

Table 4.4.4 Dimensions of existing core of socket1 ʺ 42

Table 4.4.5 Dimensions of new core of socket1 ʺ 43

Table 4.5.1 Materials Required for Mould 44

Table 4.5.2 Materials Required for Making socket 45

Table 4.5.3 Dimensions of new piece of socket 1ʺ 45

Table 4.5.4 Dimensions of existing piece of socket 1ʺ 46

Table 4.5.5 Dimensions of new piece of socket 1 ʺ 46

Table 4.5.4 Dimensions of existing piece of socket 1 ʺ 46

Table 4.6.1 Tools and equipments required 47

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 11/81

-XI-

LIST OF PUBLICATIONS

1.  Optimization of Core Shape for Malleable Iron Pipe Fittings (Socket 1 ʺ),

Published in “International Journal on Emerging Technologies” 3(I):151-

153(2012)

2.  Productivity Improvement technique for Malleable Iron Pipe Fittings

(Socket 32 mm),  Published in International conference on “advancements and

futuristic trends in mechanical and materials engineering”(PTU,AFTMME -2012)

 pp. 65-66

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 12/81

-1-

CHAPTER: 1

INTRODUCTION

The term “Pipe fittings” include a number of fittings e.g. elbow, tee, socket, union etc.

These fittings are used to connect pipes. Male fittings are those which have only male

threads. Female fittings are those which have only female threads. Male-Female

fittings are those which have male and female threads at outlet. We have done work 

on Blackheart Malleable iron pipe fittings. Malleable iron is a cast iron-carbon alloy,

which solidifies in the as-cast condition in a graphite free structure, that is, Total

carbon content is present in its combined form as cementite (Fe 3C). The properties of 

material are obtained by heat treatment. Malleable iron castings may be either white

heart, blackheart or pearlitic, according to the chemical composition, Temperature and

Time cycle of Annealing process and properties resulting there from. Blackheart

Malleable iron castings obtained after annealing in an inert atmosphere have a black 

fracture. The microstructure developed in the castings is mainly ferritic with temper 

carbon.

Figure: 1.1 Pipe Fittings

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 13/81

-2-

1.1 PROCESS OF MANUFACTURING G.I. MALLEABLE PIPE

FITTINGS

Various steps in manufacturing of pipe fittings are -:

  Casting

  Annealing

  Grinding

  Shot Blasting

  Galvanization

  Machining including Chamfering, Tapping.

1.1.1 Casting: In Casting process pipe fittings are made from molten metal. The

molten metal consists of 60% pig iron and 40% steel scrap. The Sand Casting (Green

Sand) moulding process utilizes a cope (top half) and drag (bottom half) flask set-up.

The mould consists of sand (usually silica) and molasses. When molasses and sand

are mixed the bonding characteristics of the clay are developed which binds the sand

grains together. When applying pressure to the mould material it can be compacted

around a pattern, which is either made of metal or wood, to produce a mould having

sufficient rigidity to enable metal to be poured into it to produce a casting. The

 process also uses coring to create cavities inside the casting. After the molten metal is

 poured and has cooled, the core is removed. The material costs for the process are low

and the sand casting process is exceptionally flexible. The mould material is

reclaimable, with between 90 and 95% of the sand being recycled, although new sand

and additions are required to make up for the discarded loss.

The sand must exhibit the following characteristics:

Flow ability: The ability to pack tightly around the pattern.

Plastic Deformation: Have the ability to deform slightly without cracking so that the

 pattern can be withdrawn.

Green Strength: Have the ability to support its own weight when stripped from the

 pattern, and also withstand pressure of molten metal when the mould is cast.

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 14/81

-3-

Permeability: This allows the gases and steam to escape from the mould during

casting.

The melting process is either carried out in rotary or induction furnace.

Figure: 1.2 Induction Furnace

An induction furnace is an electrical furnace in which the heat is applied by

induction heating of metal. The advantage of the induction furnace is a clean, energy-

efficient and well-controllable melting process compared to most other means of 

metal melting. Most modern foundries use this type of furnace and now also more

iron foundries are replacing cupolas with induction furnaces to melt cast iron, as the

former emit lots of dust and other pollutants. Induction furnace capacities range from

less than one kilogram to one hundred tones capacity and are used to melt iron and

steel, copper, aluminium and precious metals. Since no arc or combustion is used, the

temperature of the material is no higher than required to melt it; this can prevent loss

of valuable alloying elements. The one major drawback to induction furnace usage in

a foundry is the lack of refining capacity; charge materials must be clean of oxidation

 products and of a known composition and some alloying elements may be lost due to

oxidation (and must be re-added to the melt). Operating frequencies range from utility

frequency (50 or 60 Hz) to 400 kHz or higher, usually depending on the material being melted, the capacity (volume) of the furnace and the melting speed required.

Generally, the smaller the volume of the melts, the higher the frequency of the furnace

used; this is due to the skin depth which is a measure of the distance an alternating

current can penetrate beneath the surface of a conductor. For the same conductivity,

the higher frequencies have a shallow skin depth - that is less penetration into the

melt. Lower frequencies can generate stirring or turbulence in the metal. A preheated,

1-tonne furnace melting iron can melt cold charge to tapping readiness within an hour.

Power supplies range from 10 kW to 15 MW, with melt sizes of 20 kg to 30 tones of 

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 15/81

-4-

metal respectively. An operating induction furnace usually emits a hum or whine (due

to fluctuating magnetic forces and magnetostriction), the pitch of which can be used

 by operators to identify whether the furnace is operating correctly or at what power 

level.

1.1.2 Annealing: The pipe fitting material after casting needs to be annealed. It is

very important process in the manufacturing of pipe fittings. The purposes of 

annealing are:-

1 To improve ductility

2 To improve malleability

3 To relieve internal stresses

Figure: 1.3 Annealing Furnace

Process: Material is first packed in annealing furnace. The annealing furnace is

generally made from heat resistant material like fire bricks. The material is generally

 packed inside rings of steel. The layers of material are separated from each other by

sand which acts as a conductor of heat as well as stops products from sticking to each

other when heated. The material is heated to a temperature of 960 0C. The material is

then held at this temp for about 4-6 hours. The material is then allowed to cool slowly

in the furnace itself. The cooling process takes generally 2-3days.After cooling

 process is over and furnace cools to suitable temp, the material is then taken out. The

annealed material is tested for its malleability by the compression test as explained

further.

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 16/81

-5-

Compression Test: This test is conducted on the material that has been annealed. It is

conducted to judge the malleability of the pipe fittings and shall be carried out as:

A ring of 10mm width for socket 1ʺ is cut from one end of the unfinished socket after 

the annealing process to form a test piece. The outside of the test piece is measured

over the points 450 off mould joints. The test piece shall be placed on the equipment

as shown in FIG 1.4 [A hand vice can be used in place] and shall be compressed

gradually at the rate of 17 to 20mm/min until the amount of compression reaches 5

 percent of one original outside diameter. The test should not show any crack on any

 part of the work piece.

Figure: 1.4 Equipment for Compression

1.1.3 Grinding: A machine tool operation which is mostly used of finish within close

tolerances flat, cylindrical or other surfaces by the abrasive action of a high speed

grinding wheel is known as Grinding. A machine tool which is usually designed to

finish close tolerances flat, cylindrical or other Surfaces by the abrasive action of a

high-speed grinding wheel is called grinding machine or grinder. 

Grinding Action: Grinding wheels are composed of an abrasive material of very high

harness, approaching the hardness of diamond held together by a set adhesive

substance called, a bond. Upon rapid rotation of the grinding wheel, its grainscontacting the work piece remove thin chips from it. The cutting ability of separate

grains varies since the form of grains varies and their sharp cutting edges are arranged

differently. For this reason, the material being ground is cut by certain grains in the

same manner as by the cutting teeth of a milling cutter. Some other grains scrape and

scratch the work. Some grains only rub against the surface of the work piece.

Consequently, in grinding a metal, not only chips of various forms are produced, but a

metallic dust as well is produced .The particles of this dust and pieces of the chips are

cemented together at the high temperature generated in grinding. This temperature is

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 17/81

-6-

due to friction of grinding wheel on the ground surface. In grinding without coolant,

the temperature is in the zone where chips cut may reach 2000oC. The harder the

material to be ground, the more rapidly the grains of the grinding wheel will be

dulled. For this reason, wheels, in which the bond presents less resistance to breaking

out of the dulled grains, are used for grinding the harder materials. This facilitates self 

sharpening of the wheel. 

Figure: 1.5 Grinding Machine

1.1.4 Shot Blasting: Shot Blasting is a process in which an abrasive material is forced

through a jet nozzle using compressed air pressure. This creates a fast and effective

way of  cleaning or preparing surfaces for recoating using steel shots. Steel shots are

sharp, hard abrasive which is used  to prepare surfaces on non‐ferrous metals before

recoating. 

Shot blasting machines incorporate 5 basic elements:

1. One or more wheel units

2. A cabinet that contains abrasive material as it performs its cleaning function.

3. A means of presenting the work to be cleaned the abrasive action.

4. A system to re‐circulate and clean the abrasive, removing sand, fines and

contaminants from the abrasive mix before returning effective abrasive to the blast

wheels.

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 18/81

-7-

5. Dust collector to remove all dust and abrasive fines from blast machine to provide

an environmentally safe operating atmosphere.

The wheel: The key component in airless blast cleaning machine is the

abrasive‐throwing wheel. The intensity of the radial and tangential forces it develops,

the abrasive flow volume and velocity it generates, the accuracy and stability of its

 blast pattern in the target zone‐all are vital to the effectiveness and economy of the

 blast cleaning operation. 

Figure: 1.6 Shot Blasting Machine 

Blasting: Abrasive from an overhead storage hopper is fed to the centre of the wheel

unit which is rotating at high speed. A cast‐alloy impeller rotates with the wheel and

carries the abrasive to an opening in the stationary control cage form where it is

discharged onto the bladed wheels. At this point, the abrasive is picked up by the

inner ends of the throwing blades and is rapidly accelerated as it moves to the

 periphery of the wheel. When the blast wheel is properly adjusted and its elements are in good working

condition, the full effect of the blast stream will be attained for maximum efficiency.

Some of the more common causes and cures of wheel malfunctions are highlighted on

these pages.

Control Cage: One of the most critical components in the wheel is the control cage.

This governs the directions of the blast out of the wheel and onto the work to be

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 19/81

-8-

cleaned. As little as 10% misadjustments of the “hot spot” can reduce cleaning

efficiency by 25% or more.

Impeller: Of equal importance in maintaining the desired blast pattern is the effect of 

abrasive wear on the impeller. If the leading edge of any segment of the impeller 

 becomes worn so that it becomes parallel with the blade, the abrasive will cut through

the bottom of the blade and could create wear on the steel spacer and blade slots. Not

only will the blast pattern be affected but also wheel imbalance will result creating

additional serious problems. 

Wear Affects Efficiency: Complaints of longer than normal cleaning cycles,

inadequate cleaning and high maintenance costs can usually be traced directly to loss

of directional control over the blast pattern. Of course, the blast pattern must be set

 properly in the first place but it can change for a variety of reasons. Wear on the

wheel parts which control the “Hot‐Spot”‐ control cage, impeller and blades  –  is the

chief cause for changes in this pattern. Inspect them regularly and replace them as

soon as excessive wear is detected. 

Ammeter  –  A Tool to Control Cleaning Efficiency: The ammeter on each Shot

Blasting machine wheel motor is an important tool to help you control cleaning

efficiency. It is the only way of determining at a glance how much abrasive is being

thrown by the wheel. For example, on a typical Shot Blasting machine, with a 19 ½ ʺ 

diameter by 2 ½ʺ wide wheels using a 15 HP motor on 440 volts approximately 8

amperes would be used without any abrasive flowing into the wheel. Under full load,

20 amps would be used. The abrasive thrown under full amperage would weigh about

375 pounds per minute or about 31 pounds for each abrasive load ampere. If the

wheel were operating at 17 amps rather than 20, there would be over 25% reduction in

wheel efficiency. When the wheel is operating at less than full amperage (as stamped

on the plate above the ammeter) this usually means there is an insufficient amount of 

abrasive in the machine but it may also indicate poor adjustment of the wheel parts, it

is important that the cause of this low amperage be corrected immediately since

longer periods of blasting are required under these conditions to produce the desired

cleaning results. 

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 20/81

-9-

Wheel Housing: The blast wheel is enclosed within a housing whose primary

function is to serve as a safety guard and abrasive seal around the rapidly rotating

wheel. To minimize wear on the housing, a series of protective liners are installed

inside this housing. The latest protective liner kit, which is recommended for all “M”,

“R” and “RLM” wheels, consist of only nine places. Labyrinth seals provide an

abrasive tight closure between top, side and end liners. The curved top liner 

minimizes ricochet of abrasive back into the wheel. 

To Reduce Blast Cleaning Wheel Noise: One of the chief sources of noise at the

 blast machine is the opening where abrasive is fed to the blast wheel. The Sound

Abator totally enclosed abrasive control valve, which can be installed on most Shot

Blasting machine machines, is specifically designed to combat noise from this source

 by sealing the opening to the wheel. It can reduce the noise level of a typical blast

wheel, measured three inches from the abrasive feed inlet, by 25 decibels (A scale).

The Sound Abator also serves another important function by modulating the volume

of abrasive flowing to the blast wheel.

The Abrasive Handling System: Every Shot Blasting machine blast cleaning system

contains the following elements: 

1. The abrasive elevator.

2. A device to move abrasive from the elevator and provide preliminary screening of 

the abrasive before it enters the separator  – this may be by gravity or a screened rotary

screw conveyor.

3. An air wash abrasive separator to remove all dust, fines and contaminants from the

abrasive.

4. A hopper to collect refuse removed for the abrasive.

5. An abrasive control device (Sound Abator) to control and meter flow of abrasive to

the blast wheel.

6. A means of moving spent abrasive, sand and other contaminants to the elevator.

This could be a helicoids screw, shaker conveyor or gravity.

Ventilation and Environmental Protection: Since the blasting action removes sand,

scale, rust, etc. from work and reduces the material removed to varying degrees of 

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 21/81

-10-

fineness, an adequate and properly operating dust collector system is necessary for 

efficient operation of the blast equipment. The predetermined flow of air from each of 

the vent points on the Shot Blasting machine must exist for proper operation of the

machine. Although the failure to maintain this flow will soon become obvious

through a reduction in cleaning efficiency and dusting at the machine, a periodic

check on air volumes will preclude the possibility of an unobserved gradual

degradation of operation. The dust collector is also the air source for the abrasive

separator. The condition and efficiency of the dust collector have an important

influence on separator efficiency. When proper ventilation is being experienced at

each venting point, static pressure readings (with a manometer) should be taken in

each of the vent pipes and these readings recorded as standards for future comparison.

Should any future readings show material change in static pressures, it indicates an

upset in the condition of air flow, the cause of which should be investigated

immediately. 

Importance of Abrasive: The final element of blast cleaning is the abrasive itself.

Three important factors should be considered to evaluate the performance of the

abrasive: 

1. The amount of cleaning the abrasive will do in a given length of time.

2. The quality of the cleaning.

3. The cost of performing a given amount of work.

This performance is determined by abrasive breakdown characteristics, the abrasive

size distribution in the blast machine and the abrasive hardness. Abrasive breakdown

rate affects the shape of the abrasive in the operating mix, and therefore the

maintenance on the blast equipment. Abrasive size distribution is also influenced by

the breakdown rate. The smallest size abrasive possible should be selected for each

 job. The size of the abrasive selected, however, is not the factor influencing

consumption. Rather, it is the size at which the abrasive is removed for the machine.

Abrasive hardness is the third major consideration in arriving at proper product

selection. The harder, tougher and more resistant the abrasive, the more useful energy

it will impart to the cleaning task.

When possible, use a low breakdown and high hardness product, characteristics found

in Shot Blasting machine Steel Abrasive, for lowest maintenance costs and maximum

cleaning efficiency.

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 22/81

-11-

1.1.5 Hot Dip Galvanization: The galvanizing process consists of three basic steps:

surface preparation, galvanizing and inspection.

Surface Preparation: Surface preparation is the most important step in the

application of any coating. In most instances incorrect or inadequate surface

 preparation is generally the cause of a coating failing before its expected service

lifetime. The surface preparation step in the galvanizing process has its own built-in

means of quality control in that zinc simply will not metallurgically react with a steel

surface that is not perfectly clean. Any failures or inadequacies in surface preparation

will immediately be apparent when the steel is withdrawn from the molten zinc

 because the unclean areas will remain uncoated and immediate corrective action must

 be taken. Surface preparation for galvanizing typically consists of three steps: caustic

cleaning, acid pickling and fluxing.

Caustic Cleaning: A hot alkali solution often is used to remove organic contaminants

such as dirt, paint markings, grease and oil from the metal surface. Epoxies, vinyls,asphalt or welding slag must be removed before galvanizing by grit-blasting, sand-

 blasting or other mechanical means.

Pickling: Scale and rust normally are removed from the steel surface by pickling in a

dilute solution of hot sulphuric acid or ambient temperature hydrochloric acid.

Surface preparation also can be accomplished using abrasive cleaning as an

alternative to or in conjunction with chemical cleaning. Abrasive cleaning is a process

whereby sand, metallic shot or grit is propelled against the steel material by air blasts

or rapidly rotating wheels.

Fluxing: Fluxing is the final surface preparation step in the galvanizing process.

Fluxing removes oxides and prevents further oxides from forming on the surface of 

the metal prior to galvanizing. The method for applying the flux depends upon

whether the galvanizer uses the wet or dry galvanizing process.

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 23/81

-12-

In the dry galvanizing process (see Figure 1.7), the steel or iron is dipped or pre-

fluxed in an aqueous solution of zinc ammonium chloride. The material is then dried

 prior to immersion in molten zinc. In the wet galvanizing process, a blanket of liquid

zinc ammonium chloride is floated on top of the molten zinc. The iron or steel being

galvanized passes through the flux on its way into the molten zinc.

Figure: 1.7 Dry Galvanizing Process

Galvanizing: In this step, the material is completely immersed in a bath consisting of 

a minimum of 98% pure molten zinc. The bath temperature is maintained at about 840

0F (449 0C). Pipe Fittings items are immersed in the bath until they reach bath

temperature. The zinc metal then reacts with the iron on the steel surface to form a

zinc/iron inter-metallic alloy. The articles are withdrawn slowly from the galvanizing

 bath and the excess zinc is removed by draining, vibrating and/or centrifuging. The

metallurgical reactions that result in the formation and structure of the zinc/iron alloy

layers continue after the articles are withdrawn from the bath, as long as these articles

are near the bath temperature. The articles are cooled in either water or ambient air 

immediately after withdrawal from the bath. Because the galvanizing process involves

total material immersion, it is a complete process; all surfaces are coated. Galvanizing

 provides both outside and inside protection for hollow structures. The galvanizer’s

ability to work in any type of weather allows a higher degree of assurance of on-time

delivery. Working under these circumstances, galvanizing can be completed quickly

and with short lead times. Two- or three-day turnaround times for galvanizing are

common. 

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 24/81

-13-

Inspection: The two properties of the hot-dip galvanized coating that are closely

scrutinized after galvanizing are coating thickness and coating appearance. A variety

of simple physical and laboratory tests may be performed to determine thickness,

uniformity, adherence and appearance. Products are galvanized according to IS 6745-

1972. The mass of zinc coated pipe fittings is determined by stripping method as

explained: 

Stripping Solution: Dissolve 20g of antimony trioxide (Sb2O3) or 32g of antimony

trichloride (SbCl3) in 1000ml of concentrated hydrochloric acid (sp gravity 1.16).  

Immediately before test, prepare the stripping solution by adding 5ml of the solution

 prepared under to 100ml of concentrated hydrochloric acid (sp gravity 1.16). Mix

well. 

Procedure: Weigh the cleaned test piece whose mass is less than 200 g nearest to

0.01 g, for test piece whose mass is between 300 to 1000 g, weigh to nearest 0.1 g and

for masses over 1000 g, the accuracy of weighing shall be nearest to 0.5 g. After 

weighing immerse each test piece singly in test solution prepared and allow remaining

there until the violent evolution of hydrogen ceases, and only a few bubbles are being

evolved. This requires about 15 to 30 seconds except in the case of sherardizedcoatings which require somewhat longer time. 

Calculation

Where

M=mass of zinc coating in g/m2of surface,

M1=original mass in g of the test piece,

M2=mass in g of the stripped test piece, and

t= thickness of the stripped test piece in mm.

Performance of Galvanized Coatings: Galvanized coatings have a proven

 performance  under numerous environmental conditions. The corrosion  resistance of 

zinc coatings is determined primarily  by the thickness of the coating but varies with

the severity of environmental conditions. Environments in which galvanized steel and

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 25/81

-14-

iron  are commonly used include indoor and outdoor atmospheres, the storage of 

hundreds of different chemicals, in fresh water, sea water, soils, and/or concrete.

Because of the many years galvanizing has been used for corrosion protection, a

wealth of real-world, long-term exposure data on zinc coating performance in a wide

variety of environments is available.  

Atmospheric Exposure: Zinc oxide is the initial corrosion product of zinc in

relatively dry air. This is formed by a reaction between the zinc and atmospheric

oxygen. In the presence of moisture, this can be converted to zinc hydroxide. The zinc

hydroxide and zinc oxide further react with carbon dioxide in the air to form zinc

carbonate. The zinc carbonate film is tightly adherent and relatively insoluble. It is

 primarily responsible for the excellent and long-lasting corrosion protection provided

 by the galvanized coating in most atmospheric environments. Exposure atmospheres

may be divided into five types. They are: 

Moderately Industrial: These environments generally are the most aggressive in

terms of corrosion. Air   emissions may contain some sulphides and phosphates that

cause the most rapid zinc coating consumption. Automobile, truck and plant exhaust

are examples of these emissions. Most city or urban area atmospheres are classified as

moderately industrial.

Suburban: These atmospheres are generally less corrosive than moderately industrial

areas. As the term suggests, they are found in the largely residential perimeter 

communities of urban or city areas.

Temperate Marine: The service life of galvanized coatings in marine environments

is influenced by proximity to the coastline and prevailing wind direction and intensity.

In marine air, zinc corrosion follows a different mechanism; chlorides from sea spray

can react with the normally protective zinc corrosion products to form soluble zinc

chlorides. When these chlorides are washed away, fresh zinc is exposed to corrosion.

The addition of calcium or magnesium salts to the surface of the zinc can extend the

service life of the coating. 

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 26/81

-15-

Tropical Marine: These environments are similar to temperate marine atmospheres

except they are found in warmer climates. Possibly because many tropical areas are

often relatively far removed from heavy industrial or even moderately industrial areas,

tropical marine climates tend to be somewhat less corrosive than temperate marine

climates.

Rural: These are usually the least aggressive of the five atmospheric types. This is

 primarily due to the relatively low level of sulphur and other emissions found in such

environments. 

Corrosion Performance in Fresh Water: Galvanizing is successfully used to protect

steel in fresh water exposure. “Fresh water” refers to all forms of water except sea

water. Fresh water may be classified according to its origin or application. Included

are hot and cold domestic, industrial, river, lake and canal waters. Corrosion of zinc in

fresh water is a complex process controlled largely by impurities in the water. Even

rain water contains oxygen, nitrogen, carbon dioxide and other dissolved gases, in

addition to dust and smoke particles. Ground water carries microorganisms, eroded

soil, decaying vegetation, dissolved salts of calcium, magnesium, iron, and

manganese, and suspended colloidal matter. All of these substances and other factors

such as pH, temperature and motion affect the structure and composition of the

corrosion products formed on the exposed zinc surface. Relatively small differences

in fresh water content or conditions can produce relatively substantial changes in

corrosion products and rate. Thus, there is no simple rule governing the corrosion rate

of zinc in fresh water. Hard water is much less corrosive than soft water. Under 

conditions of moderate or high water hardness, a natural scale of insoluble salts tends

to form on the galvanized surface. These combine with zinc to form a protective

 barrier of calcium carbonate and basic zinc carbonate.

Corrosion Performance in Soils: More than 200 different types of soils have been

identified and are categorized according to texture, colour and natural drainage.

Coarse and textured soils, such as gravel and sand, permit free circulation of air, and

the process of corrosion may closely resemble atmospheric corrosion. Clay and silt

soils have a fine texture and hold water, resulting in poor aeration and drainage. The

corrosion process in such soils may resemble the corrosion process in water. 

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 27/81

-16-

1.1.6 Machining: After galvanization, the pipe fittings are chamfered. Then sockets

are threaded with the help of taps. In this way entire manufacturing process is over. 

1.2 TERMINOLOGY:

Various terms used in pipe fitting are as given below:-

  Fittings- The connecting pieces connecting one or more parts.

  Equal Fittings- Where all outlets are of the same size.

  Unequal Fittings- When two or more outlets are of different size irrespective

of the number of outlets.

  Male Fittings- Fittings having only male threads.

  Female Fittings- Fittings having female threads on the outlet.

  Male-Female Fittings- Fittings having male and female threads at the outlets.

 Reinforcement- An additional material at the outside diameter of an internallythreaded fitting in the form of band or bead.

  Rib- Locally of axially aligned additional material on the outside or inside of a

fitting for assistance in the assembly or manufacturing.

  Run- Two principal axially aligned outlets of a tee or cross.

  Branch- Side outlet(s) of a tee or cross.

  Chamfer- Removal of a conical portion at the entrance of a thread to assist

assembly and prevent damage to the start of the thread.

  Face-to-Face Dimension- Distance between two parallel faces of axially

aligned outlet of a fitt ing.

  Face-to-Centre Dimension- Distance from the face of an outlet to the central

axis of angularly disposed outlet.

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 28/81

-17-

  Centre-to-Centre Dimension- Distance between the two parallel central axis

of the outlet of a fitting.

1.3 TYPES OF FITTINGS:

It is denoted as elbow, bend, tee, cross, etc. the diagrammatic representation of the

various types of fittings is given in table-1.3.1

Table 1.3.1: Type of Fittings [26]

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 29/81

-18-

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 30/81

-19-

Detail of wall thickness of various fitt ings is as shown in table-1.3.2

Table 1.3.2: Details of Wall Thickness

Sr. No. Size Designation Wall Thickness

Basic Size Tolerances

(1) (2) (3) (4)

I.  ½ 2.5 -0.5

II.  ¾ 3.0 -0.7

III.  1 3.0 -0.7

IV.  1 ¼ 3.5 -0.7

V.  1 ½ 3.5 -0.7

VI.  2 4.0 -0.7

VII.  2 ½ 4.5 -1.0

VIII.  3 5.0 -1.0

IX.  4 6.0 -1.0

X.  5 6.5 -1.0

XI.  6 7.5 -1.0

 No limit for plus tolerance

1.4 CORE:

A core is that portion of the mould which forms the hollow interior of the casting or 

hole through the casting. The core is a mass of dry sand which is prepared separately:

 baked in an oven and then placed in the mould. The core gives hollow portion in the

casting which cannot be readily obtained by the mould. Core is used in moulding

where big size hole is to be obtained in the casting. Small size hole cannot be obtained

in the casting by core. In pit moulding, the entire mould is made of core. Sometimes,

the cores are also used to reduce metal erosion in gates and runners to retard foreign

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 31/81

-20-

matter in the melt and to provide a cap on the top of the mould.

1.4.1 Characteristics of Cores: The core should be sufficiently strong and hard so

that it may be able to support its own weight and withstand force of molten metal.

  It should be permeable to escape core gases.

  It should be able to withstand high temperatures of the molten metal.

  It should be capable of collapsing shortly after the molten metal has solidified

around it.

  It should produce minimum amount of gas when in contact with molten metal.

1.4.2 Core Making: The cores are made separately in a core box. The core boxes are

made of wood or metal and designed in several types to aid in-core removal. The

various steps in core making are ramming of core sand in the box, venting,

reinforcing, removing of core from the box, baking, pasting and sizing. The cores are

made either by hand or by machines designed for this purpose. The cores of 

symmetrica1 cross-section can be made by extruding core sand mixture through a

suitable die opening. Usually the cores are made by core blowing machine for 

 production work.

1.4.3 Core Baking or Core Drying: After making the cores, they are dried to drive

off the moisture and to harden the binder. The cores are dried in ovens equipped with

drawers, shelves or other holding devices. They are dried in batches or continuously

over moving shelves. The heat in oven is produced by burning oil or by electric

resistance. The core drying time depends upon the quantity of moisture and binder 

used in the sand, size of the core and temperature of the oven.

1.5 CORE MATERIALS:

Dry sand cores are usually made of clear river sand which is mixed with a binder and

then baked to give the desired strength.

1.5.1 Core Sand: Core sand must have the proper type of strength, porosity or 

 permeability, smooth surface and sufficient refractoriness. Strength depends on the

type of sand and binding material used. Sharper grains of sand will bond together and

form a stronger core. Porosity or permeability depends on the size of the sand grain

and its freedom from fines in between grains. Smooth surface of a core results in a

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 32/81

-21-

smooth surface of the opening in the casting. In the attempt to produce a smooth

surface, care must be used not to go too far since permeability is lost as the fineness of 

the sand increases. Refractory property is possessed by sand to a great extent. A

 binder must be selected that will withstand the temperature required of the core. A

thin coating of graphite or similar material adds considerably to its ability to

withstand the intense heat momentarily.

It is not desirable to have the core remain hard after the metal has cooled. The binding

material used should disintegrate or be burn out by the prolonged contact with the hot

metal so that the core may be removed easily from the finished casting. This so

 prevents shrinkage cracks during cooling.

1.5.2 Core Binders. It has already been stated that silica sand is used for preparing

cores. This sand has no natural bond. Hence some other materials are added to it

which acts as binders. The binder cement the sand grains and give sufficient strength

to cores to prevent breakage, distortion, erosion during core making, moulding and

casting.

Various commercial binders are available in the market like core oil, resins, sulphite,

liquor, molasses and proteins. The core oil mainly contains vegetable oil like linseed

oil or corn oil. Core oils are very economical and produce better cores. Rosin and

 pitch are thermoplastic binders. The powdered binder is mixed with the core-sand. On

heating, the binder liquefies and coats the grain sands. On cooling the dispersed liquid

 binds the sand grains together fulfilling a united mass. Rosin is a form of resin.

Petroleum and coal for resins are also used as binders. Pitch compounded with dextrin

and steam coal is used for large cores. Phenol and urea-formaldehyde, thermosetting

 plastic core binders are more suitable. Molasses give hardness to the core but lacks in

strength.

Protein binders like gelation, glue, casein, etc. are used where easily collapsible cores

are required. Sulphite liquor is used where high strength, hardness, quick drying and

high temperature resistance is required.

1.6 CORE SANDS WHICH REQUIRE HEAT TREATMENT:

For the strengthening of cores include oil-bonded, clay-bonded, and resin-bonded

sand mixture (the bonding resins in the last type of sand are fast-curing synthetic

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 33/81

-22-

resins). Oil-bonded and clay-bonded  core sands show satisfactory properties are

comparatively cheap and suitable for use in hand and machine production of cores in

sand blowing, jolt and squeeze machines. The cores made from these required thermal

heating to impart strength, which lengthens the production process, lowers the

operating efficiency of the foundry and generates a need for installing driers.

That is why such core sands find application in piecework and short run work.  Resin

bounded core sands are made with synthetic resin binders of Class  B-1

(carbamidebase), B-3 (lignin) and Class A-I (powdered Bakelite).

These binders are capable of hardening at 230-250°C for a short time (2 or 3 min to

30-50s depending on the composition and size of cores). Catalysts (both organic and

inorganic acids) may be added to speed up the process of  curing. The core sand

composition also includes such additives as ferric oxide and crystalline graphite

which improve the heat conduction and increase the specific heat of the sand all

thereby enable the core to heat through and harden more speedily. Other additives

diminish stickiness and improve flow ability. The core sand hardens directly in a

metal core box heated by a gas or by electrical heaters. These are the so-called hot

 boxes. The sand hardens as a result of policondensation of a binder (resins of B-1 and

B-3 classes) or its polymerization (powdered Bakelite) the core gains high strength,

up to 100 kgf cm-2 and B-3 class resins have low green strength, they flow easily and

thus readily fill the cavities of complex core boxes. The cores are taken off the boxes

already hard. So the castings show improved dimensional accuracy. The core sands

deform well, shake out with ease from the castings but have insufficient thermal

stability. Resin-bonded sands go into the production of cores of all classes for casting

thin-walled small pieces 150 to 200 kg in mass from iron, steel and non-ferrous

alloys. These sands are prepared from washed sands of the first and second classes,

which are more expensive than common quartz sands; the cost of binders is high too,

400 to 800 rubles per ton. The cores are moulded in complex, costly metal boxes, so

that resin-bonded sands are envisaged for use in high volume and high run production.

In this case it pays to automatize the production process with a view to increasing the

efficiency of manufacture, cutting down the costs and improving the quality of 

castings. Along with the sands mentioned above, powdered Bakelite-bonded sands are

used for the manufacture of hollow shell-type cores.

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 34/81

-23-

1.7 CORE SAND WHICH DO NOT REQUIRE FOR HEAT

TREATMENT

Core sands which do not call for heat treatment are most promising since they allow

the foundry to dispense with the heating of boxes and to simplify substantially the

 production process and moulding equipment. Cores can be made in wooden, plastic

and metal core boxes. These sands are highly suitable for use in various types of 

 production.

Synthetic resin-bonded cold-curing core sands contain such binders as carbamide,

carbamide-furan, phenol-furan; phenol formaldehyde resins (binders of B-1 class).

Catalysts are added to speed up the hardening of binders. These are commonly

organic and inorganic acids such as benzenesulphonic, orthophosphoric and nitricacids. The core sands feature high flow ability and strength from 14.7 to 15.20 kgf 

cm-2 and also good gas permeability, deformability and collapsibility.

An important characteristic of a sand mixture is its life that is the time during which

the sand still remains mouldable. The life of sands can be controlled by varying the

amount of catalyst added to the sand. As the quantity of catalyst increases, the sand

life shortens. So, knowing the time it takes to fill the box and ram the sand, we can

add such an amount of catalyst as is necessary to provide the desired span of life for 

the sand. As the sand hardens, its strength gradually grows. The rate of strength

growth is directly proportional to the added amount of catalyst. The maximum value

of strength for the given sand decreases with the increased quantity of catalyst.

As the thermal stability of sand decreases, burning-on becomes more probable.

Phenolic and phenol-furan resins feature the highest thermal stability and make

suitable binders of sands for steel castings. Carbamide-furan resins have lower 

thermal stability. They serve as binders of iron core sands. Carbamide resins have the

lowest thermal stability. These are the binders of core sands for casting non-ferrous

alloys.

Cold curing sands has a lower strength than sands curable in hot boxes and therefore

they largely go into the production of cores of the third and fifth classes. The setting

time these sands take until they acquire a maximum strength comes to few hours. The

merit of cold-curing sands is they allow for the production of cores in wooden, plastic,

and metal boxes. That is why they have found the widest applications in the batch

 production of moderate sized and large sized castings from iron and steel. The use of 

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 35/81

-24-

these sands makes it possible to exclude drying, mechanize the core production

 process, improve the quality of castings and increase the output.

Cold-curing silica-bonded sands include core sand mixtures with a liquid silica glass

as a binder. The cores are dried by blowing carbon dioxide through the rammed sand.

They can also harden under heat. The core sands have high strength, good gas

 permeability, but show lowed formability and poor collapsibility. Sawdust

(aboutl.5%) and asbestos powder (up to 5 %) make the cores more deformable and

collapsible. These core sands are applicable in the piece and batch production of steel

and iron castings.

 Liquid self-set core sands which compare in properties to the previously described

sands have come into extensive use in the batch production of large castings. The use

of these sands enables the foundries to raise the output per man-hour, mechanize the

 process of production of cores for casting parts both piecemeal and in small lots and

improve the quality of castings

1.8 WASHES, PASTES, POWDERS AND OTHER DRESSINGS:

Core and mould washes and pastes are called upon to prevent metal penetration or 

 burning -on, increase the surface strength, decreases the crumbliness of mould and

core walls and provide clean surfaces and smooth casting appearance. Anti-

 penetration washes consist of refractory materials which form the base and binding-

agents. The washes applied to the surface of mould and cores form a strong refractory

coating which keeps the molten metal and its oxides from penetrating into pores

 between sand grains and thus eliminates the burnt- on-effect.

1.8.1 Moulding Washes: These paints must conform to the following requirements:

(l) Have a high melting temperature to sand up to the fusion effect of contactingmetal.

(2) Produce no fusible compositions when in contact with the metal.

(3) Remain invariable in composition during preparation, storage and when in use.

(4) Have good covering capacity.

(5) Form a strong skin on mould and core walls, free of cracks after drying.

(6) Firmly adhere to the mould. . .

(7) contain as little foreign matter and difficulty available materials as possible.

The choice of washes depends on the kind of metal cast, mass of the casting and

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 36/81

-25-

moulding method. Washes for large iron castings contain such anti-penetration

materials as black lead with the additions of bentonite and binders: In washes for 

small and medium size iron castings, silica flour mixed with coal and ground coke

substitutes for graphite. Washes for steel castings usually consist of silica flour or 

zirconium silicate which serves as a refractory base, mixed with the same-binders as

those used for iron castings.

In casting iron parts, it is advisable to add 5% coal and charcoal dust to the wash in

order to create-a reducing atmosphere in the mould; silica flour, graphite and ground

anthracite account for 95% of this reducing wash.

The compositions of washes for moulds and cores of iron castings are described here.

The steel moulding washes of the 2nd, 3rd and 4th types are put to use for cores in

castings steel parts with walls 20 to 40 mm in thickness. In painting cores or moulds,

it is advisable to stir the wash regularly to make it stay in suspension. When applying

the wash on cores by dipping, one should shake off the excess of wash to avoid

influxes and insure against sealing of the vents, it is good practice to check the coat

tightness and its surface hardness by applying the wash to sample moulds and cores or 

to standard specimens. To enable a better sticking of the wash to the surface of 

moulds and cores, foundries make use of priming paints consisting of 25% lignin,

75% water and 25% pectic gel. The paints are applied by the common methods. Air -

drying washes speed up the process of drying of coated moulds and cores. The

composition of a wash of this type includes 10% crystalline graphite, 12% black lead,

3.5% polyvinyl butyral and 74.5% solvent 646 (or a solution of ethyl acetate and

alcohol in the proportion 1. to 1). This wash is applied to moulds and cores for iron

castings. In the wash used for steel castings, graphite is changed by zircon.

1.8.2 Pastes: If washes do not give a sufficiently smooth casting surface nor ensure

the desired dimensional accuracy of casting, it can be useful to apply pastes on to the

surface of cores to exclude surface blemishes. Pastes find rare uses, however, because

they involve manual labour. Coating pastes are made anhydrous. They usually consist

of four parts crystalline graphite and one part vegetable oil (by volume). Sometime

lignin serves as a binder Instead of the expensive oil and talc together with graphite

makes the base. After applying the paste, the cores are dried at 220-240°C.It is

advisable to use oil-free pastes of the following composition (% by mass): 50% talc,

15% chamotte, 25% crystalline graphite, and 15% clay. The dry powder is then

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 37/81

-26-

dissolved in 0.6 or 0.51 of water per kg dry mass. If the paste is applied to a hot core,

the coat does not require additional drying.

1.8.3 Putties: These find use for repairing purposes and for sealing seams which may

form while cementing the cores. It is only the cores which have small cracks and

dents in unimportant places that are subject to repair. The cores with open fissures and

large fractured parts are considered non-repairable. The putty of the following

composition is most popular: 65% grade 2KOO63 sand, 25% crystalline graphite, and

10% moulding clay screened through No. 016 sieve. The ingredients-are properly

mixed, and the mass is then blended with water (O.31kg water for 1 kg powder);

 powdered soap is added in an amount of 0.5% by mass to give plasticity. Puttyapplied to the cores for steel castings consists of 40% refractory clay, 30% silica flour 

and 30% quartz sand; the powder is then mixed with 12% lignin and 13% water.

1.8.4 Core Cements (Pastes): These serve to .bond together core boxes. The

composition commonly includes water-soluble binders, clay and bentonite. In wide

use are the core pastes of the following compositions:

(1) 0.50% lignin, 50% moulding day, 20% water (the rupture strength of dry paste is

not less than 685 kPa, or kgf cm-2);

(2) 40% dextrin and 60% clay mixed with water (65 parts water to 100 parts by mass

of powder).

1.8.5 Parting Powders and Dusts: Patterns and core boxes are dusted with facings to

 prevent the moulding sand from sticking to their surfaces. Powders for in a water-

impermeable coat and thus exclude sand adhesion. Foundries use a lycopodium

 powder and its substitutes for the purpose.

The lycopodium powder  is a white to yellow substance which is light in mass, fluid

and fine-grained (it fully passes through No. 0063; No. 055 catches 5% powder). This

 powder is costly and not readily available.

1.8.6 Artificial Dusts: (substitutes for lycopodium powder) are produced from fine

 powders of tripoli, dolomite and other similar materials. The powders are treated with

 paraffin, fat, and wax to provide a thin film on powder grains. Other materials which

 prevent sand sticking are kerosene with crystalline graphite or the mixture of 10%

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 38/81

-27-

oleic acid and 90% kerosene. Heating a pattern plate to 40°C also helps eliminate

sticking. The film of kerosene on the pattern surface precludes its moistening with

water and thus excludes sand adhesion. For economy, the mixture may consist of 50%

kerosene and 50% black oil. Heating the tern dries its Surface and thus impedes

moisture condensation if the sand is still hot.

1.9 CORE FRAMES:

These are the reinforcement means moulded into cores to increase strength. The

frames are made from wire or shaped cast iron plates. The core--reinforcement must

fulfil the following requirements: give sufficient strength and rigidity to the core, not

spring or come off the core sand (soft, annealed wire will do for the purpose), deform

readily to allow for contraction of the casting, not stand in the way of vent holes being

made and permit easy shakeout of the core from the casting.Thin cores are reinforced

with I or 12 mm wire inserted into the core boxes during core moulding. Small and

moderately sized cores are made with 6 to 10 mm wire frames whose separate parts

are fastened with a thinner wire. The reinforcement means for large sand-clay cores

are iron and steel cast frames with 6 to 10 mm cast in wire inserts.

The framework for medium-sized and large cores includes lifting arrangements by

which the cores are suspended on the crane for delivering them to the assembly site.

Wire and cast frames are made in various shapes. Wire frames are laid along the

length in the core. They should terminate at least 2 or 3 mm short of the core ends.

The frame should pass into the prints to add to the core strength. If the  core has two

 prints located opposite each other the frame should extend into both.

The wire that makes up the frame proper is the basic wire and which runs around the

frame periphery and strengthens individual parts of the core is the binding wire. It is

impermissible to place the frame wire too close to or directly on the surface or the

core otherwise the frame may weld to the casting and cause the formation of 

 blowholes and hot tears. The distance from the wire frame to the core surface must be

5 to 10 mm. For cast frames, this distance in cores measuring 500 x 500 mm, (500 -

1000) x (500 - 1000) mm and over 1000 x 1000mm ranges from 20 to 30 mm, 25 to

30 mm and from 30 to 35 mm respectively.

If a core should have only one frame, this is placed in the centre of its cross section.

Several frames should be positioned uniformly over the core cross section. If a core

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 39/81

-28-

has a curved axis, it is better to set up a few thin frames instead of a single thick frame

in order to facilitate the core shakeout from the casting. It is undesirable to use core

reinforcement in the mass production of castings since this complicates the core

making process and the core shake out procedure. The required strength of cores is

often possible to achieve by using high strength core sand.

1.9.1 Making a Large Core in Turnout Core Box with Loose Side: In the

 preliminary operations, the core maker first prepares a cast frame (grid) for core

moulding. For this, he bends wire about the contour of the core working cavity and

checks it for the right position by inserting the frame into the box. After this done, he

cleans the box of the adhered sand, wipes its working surface with a kerosene-moisten

drag, fills the box with core sand to a depth of 50 - 70 mm and rams it.

The core maker now inserts the clay coated frame into the box and sets up steel pins

or hooks to reinforce the protruding and narrow portions of the core. Then he lines the

working surface of the box with core sand and rams it into narrow pockets and

recesses. Next he inserts into the box, a wooden piece to form a cavity for coke ash

and fills the box around the wooden piece with sand and compacts it. The core maker 

then removes the piece, makes vent holes, fills the cavity with coke ash and sand and

rams the core. Using the necessary handling equipment, he places a drying plate on

the top and inverts the box.

Further, the core maker raps the box to facilitate the removal of the upper part. After 

this is done; he gently takes aside loose pieces of the box. Now he finishes up the

core, checks it for compactness, repair the portions damaged in the core removal

operation, rounds off and fillets sharp angles and inserts metal pins into thin parts and

at corners. The finished core is conveyed for drying and then for coating.

1.10 MACHINE CORE MAKING:In the mass and large-lot production of castings, cores are made on core making

machines which at present are progressively introduced into foundries producing

castings in small lots. Core making machines increase the productivity of labour,

make easier the work of operators and produce cores of high accuracy, largely from

sands of lowered green strength which flow readily into deep pockets of the box.

There is a variety of core making machines available to the foundry such as core

 blowing, slinging, jarring, squeezing, core shooter, screw-feed (extrusion) types and

others.

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 40/81

-29-

CHAPTER: 2

LITERATURE REVIEW

A comprehensive review of the work reported by various researchers in the field of improving productivity of castings is described as

C. Li and B. G. Thomas [2000] investigated the theoretical limits of the shell

thickness, casting speed, and productivity of the steel continuous casting process as a

function of steel grade, section size, and mould length, assuming ideal liquid flux

lubrication. The predictions are based on the maximum casting speed that is just able

to produce a thin shell with the critical thickness needed to withstand the ferrostatic

 pressure below the mould and avoid a longitudinal rupture from excessive creep stain.

The calculations are performed with a finite-element thermal-stress model that has

 been validated with numerical solutions and plant data. The critical shell thickness is

 predicted to be on the order of 3 mm. It is surprisingly insensitive to steel grade and

superheat, but decreases with decreasing section size and increasing working mould

length. The theoretical maximum casting speed and potential productivity both

increase with decreasing critical shell thickness. The theoretical limits to casting

speed are predicted to be extremely high, exceeding 21 m/min for a conventional 700-

mm working mould length, 200-mm square bloom mould, which corresponds to 3.5

million tonnes / year. The infeasibility of these limits in practice is likely due to other 

 problems such as achieving shell thickness uniformity and liquid flux lubrication.

Attention should return to focussing on these other problems which limit productivity.

This work suggests that if shortening mould length can solve lubrication, taper, and

other problems, then it should be explored as a potential means to increase

 productivity. A uniform shell would be strong enough to withstand the ferrostatic

 pressure even with a shorter mould length and a higher casting speed. To overcome

the other problems which limit casting speed and productivity, design changes

regarding fluid flow, mould powder, mould taper, and machine length are also

required. This should be the concern of the designers of future continuous casting

 processes.

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 41/81

-30-

Thomas Schroeder [2004] reviewed the formulations of starch and dextrin based

adhesives used in making cores. Various advantages of these adhesives are

availability and low costs, stability of quality, insolubility in oils and fats, non-toxic

and biodegradable nature, heat resistance.

Nicolas Perry, Magali Mauchand, Alain Bernard [2004] described that the costs

controls is a major decision tool in the competitiveness of the companies. After 

defining the problems related to this control difficulties, they presented an approach

using a concept of cost entity related to the design and realization activities of the

 product. They tried to apply this approach to the fields of the sand casting foundry.

This work high lightened the enterprise modelling difficulties (limits of a global cost

modelling) and some specifics limitations of the tool used for this development. A

Cost Entity is a grouping of costs associated with the resources consumed by an

activity.

L.A. Dobrazański, M. Krupiński, J.H. Sokolowski, P. Zarychta and Wlodarczyk -

Fligier [2006]  presented methodology of the automatic quality based on analysis of 

images obtained with the X-ray defect detection, employing the artificial intelligence

tools. The methodologies developed made identification and classification of defects

 possible and the appropriate process control made it possible to reduce them and to

eliminate them at least in part. The reduction of defects in casting resulted in increase

of productivity.

J. Dańko, M. Holtzer, R.  Dańko [2007]  presented a paper which dealt with such

 problems of scientific and development research concerning the reclamation of used

foundry sands as: management of used sands generated in foundry production,

recommendation of selection of effective reclamation techniques and assessment

methods of the reclaimed material quality, identification methods and an

environmental impact assessment of spent sands from foundry technologies, moulding

and core sands of an increased reclamability and a decreased harmfulness for 

environment. The reclamation of used sand helps in increasing productivity.

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 42/81

-31-

A. Fedoryszyn [2007] observed that an increase of productivity requires a wide-scale

mechanisation of the equipment used for casting production on modern foundry

moulding lines. Modernisation of foundries is expected to help in creation of optimum

conditions for casting production, satisfying all the requirements regarding quantity

and quality of castings produced. Modern designs of moulding lines were described,

including moulding machines and the related equipment.

F. Peters, R. Voigt, S. Z. Ou and C. Beckermann [2007] described that steel

castings produced in sand moulds, the expansion of the sand has a significant impact

on the final size and shape of the casting. Experiments were conducted using a

cylindrical  casting to study this effect for different sands (silica and zircon) and

different sand binder systems  (phenolic urethane and sodium silicate). The type of 

sand has a significant effect on the final casting  dimensions, in particular because the

expansion of silica sand can be irreversible. The sand  expansion effect is enhanced by

the presence of sodium silicate binder. In addition, the size of the core strongly affects

the internal and external dimensions of the resulting casting.

Liu Weihua, L i Yingmin, Qu Xueliang, Liu Xiuling [2008] proposed productivity

improvement by preparing a new aqueous alkaline resol phenol-formaldehyde resin

from phenol and formaldehyde using NaOH as catalyst; With addition of some cross-

linking agents, after passing carbon dioxide gas through the resin bonded sand, high

as-gassed strength and 24 h strength are achieved.

E.O. Olakanmi and A.O. Arome [2009] contributed to productivity improvement by

characterising core-binding properties of beniseed and melon oils with a view to

finding alternatives to the imported foundry core oils that deplete Nigeria’s foreign

exchange. Clay and sand samples collected from Niger and Plateau states respectively

were blended with varying proportions of core-oils in order to assess their functional

 properties. Baked strength of 3,223.41 kNm-2 obtained for the oils suggest that the oil

samples can be used as substitute for imported core oil. On the basis of ranking

according to functional properties, beniseed oil was found to possess more desirable

functional properties in terms of bulk density, strength and collapsibility. Practical

applications of these core oils reveal that they are suitable for castings of large,

medium and small sizes. On the basis of results obtained in this study both beniseed

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 43/81

-32-

oil and melon oil can be used as substitutes for imported core oils like linseed oil,

corn oil and fish oil. Beniseed core oil binder was found to be the most desirable for 

core making because it was able to impart higher bulk density, green and baked

strength to the core mixes. Moreover, its collapsibility was faster than that of the

melon core oil binder.

Slavomír Peľák, Rudolf Mišičko, Dagmar Fedáková, Jana Bidulská [2009]

evaluated the relationships between the chemical composition, the dendrite structure

 parameters, the casting technology parameters and the occurrence of defects in

continuously cast slabs. For calculation of the selected indices, the sulphur and

 phosphorus content, the overheating temperature, the casting speed, the dendrite arm

spacing, and the central zone share were chosen. These indices determined the

susceptibility of steel to the defect formation. The defects were formed in steels with a

high overheating degree, a low share of central zone, high dendrite arm spacing and

an exceeded recommended sulphur and phosphorus content.

A.K.M.B. Rashid [2010]  presented methods of controlling hot tears in casting in

order to increase productivity. He proposed the use of local chilling of hot spots,

reduction in casting temperature to reduce hot tearing. He also suggested that fins are

a major source casting constraint. So casting should be checked straight out of mould,

not after machining.

Lakshamanan Singaramu [2010] The Taguchi method is a powerful problem

solving technique for improving process performance, yield and productivity. Green

sand process involves many process parameters which affect the quality of the

castings produced. An analysis of significant process parameters of green sand casting

 process has been made in this paper. The parameters considered were Green strength,

moisture content, permeability and mould hardness. The outcome of the paper was the

optimised process parameters of the green sand casting process which lead to

improved process performance and thus minimum casting defects. 

R. Venkataraman [2010]  presented innovative ideas for improving foundry

 productivity and casting quality. He discussed case studies in areas like design of 

 product, methoding practice, pattern manufacturing, moulding, core making, meltingand pouring and post processing after casting evaluation. He suggested time study of 

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 44/81

-33-

 processes like drag moulding time, cope moulding time, core setting time, box closing

time and pouring time. Implementation of new ideas is helpful for increasing

 productivity by reducing cycle time. Metal spurting can be prevented by a mobile

cover moving along with ladle covering top area of mould. This ensures safety of 

 people , plant and facilities.

Nishikant M Ejrekar, Vicky Kapur [2011] contributed to productivity improvement

 by presenting a practical approach to coating technology in production of high quality

castings. Refractory coatings play an important role in preventing many casting

defects like metal penetration and fusion. The main objective while using a refractory

coating is to apply a uniform coating layer free from runs, drips and cracks.

Anil Barik [2011] discussed  effects and controlling measures of directional

solidification in castings towards quality improvements. Directional solidification can

 be achieved through use of riser, chills, chaplets, insulating pads and sleeves for 

risers, mouldable exothermic sleeves. 

O.S.I. Fayomi, O.O. Ajayi and A.P.I. Popoola [2011] investigated the use of local

oils, namely groundnut oil, cotton seed oil and palm oil with Nigeria local clay and

silica sand for the production of foundry cores on varying composition. Addition of 

cassava starch, local clay, oil and moisture to sand are used to produce strong and

efficient core. These oils were tested and it was found that the three could be used to

 produce foundry cores. The best Composition was found to be core comprising 2.5%

starch, 2.5% clay, 8% oil, 8% moisture and 68% sand and baked at 150oC for 1 h

30min. The tensile strength of the core was as high as 600 KN/m2. The study have

 been used to consider the effectiveness and suitability of Nigeria Ochadamu sand and

clay with other bounding binder proportion of groundnut oil, cotton oil, palm oil, and

starch variation. The results showed after tensile analysis that the sand and oils

employed were good binders if baked between temperature of 150 and 200°C and

with 2.5% starch content.

J.S. Colten [2011] suggested solution for various casting defects to increase

 productivity. Shrinkage can amount to 5-10% by volume. So part and mould need to

 be designed to take this amount into consideration. Porosity is caused due toentrapped gases and shrinkage. Remedies suggested for gas bubbles are controlled

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 45/81

-34-

atmosphere, proper venting, and proper design of runners, gates and addition metallic

elements to react with gases. Solutions for porosity/shrinkage are use of runners and

use of chills. 

R. Vinayagasundaram, V.R. Nedunchezhian [2012] studied the application of 

modern technology in foundries and its impact on productivity and profitability. In

order to produce the cast components in foundries with high quality and at lowest cost

the foundry owners have to introduce modern technology to improve the productivity. 

This study brought out how technically qualified entrepreneurs of selected foundries

have carried out technological innovations, mainly due to their self-motivation and

self-efforts. Introducing the modern technology in the process and changing product

designs, as desired or directed by the customers resulted in cost reduction, quality and

 productivity improvement. These incremental innovations have enabled the selected

foundries to enhance competitiveness, grow in the domestic market and penetrate the

international market and grow in size over time. And have achieved technological

innovations successfully based on their technological capability and customer needs,

enabling them to sail through the competitive environment.

Bharathi Rajkumar k. and Gukan Rajaram [2012] implemented lean philosophy

in foundry with a spotlight on increasing productivity. The purpose was to develop

kaizens to eliminate waste in the foundry. In this case study, the industry could not

meet the customer requirements due to the increasing cycle time of the individual

 process and rework due to excess rejection. Processing time was more in core making,

mould making and finishing section of the foundry. Time study and motion study was

conducted and the respective operation idle time and busy time were taken. This paper 

described some of the quality improvement tasks that reduced the rejection rate which

also affect the productivity. The improvement of work process was executed by

eliminating and combining of work process, which reduces production time, number 

of process and space utilisation. The results of process were analysed by conducting

time study after implementation with cost benefit analysis to show the financial

 benefits.

Noberto T. Rizzo Downes, Ramon D. Duque  proposed ten steps for increasing

yield in ductile iron castings. These included use of ceramic foam, shorter gating

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 46/81

-35-

systems, non-turbulent gating system gating design, thinner runners and ingates, and

 proper and efficient use of risers. Many of these techniques were reported in the

literature. It was suggested that cost-savings achieved by application of these steps

could be significantly higher than those achieved by conventional scrap reduction

 procedures.

These researchers have suggested various means to increase productivity in foundry.

These researches emphasize use of easily available core binders, proper composition

of contents of cores to reduce wastage and production cost of cores. Use of refractory

coating on mould and core, use of proper sand binders of mould are other means of 

having defect free castings. Vinayagasundaram and Nedunchezhian emphasized the

use of modern technology in foundries to increase productivity. These modern

technologies include – Just in Time, Lean Manufacturing, and Enterprise Resource

Planning. Technological challenges for foundry according to him are optimal design

of pattern and core to minimize scrap.

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 47/81

-36-

CHAPTER: 3

PROBLEM FORMULATION

A lot of researches have been done by many researchers in the field of casting. There

is large scope of improvements in this field. Researches in the field of foundry include

core making techniques, use of different type of binders for core making, mould

making techniques, use of different techniques to improve quality of sand casting etc.

Major foundries are producing auto parts, sanitary fittings, pipe fittings and machine

 parts. Pipe fitting industry is a small scale industry. Pipe fittings are mainly made of 

malleable iron. These products are first casted in foundry. Productivity in these

industries is low because of non-optimal shape of core boxes in which cores are made.

It has been seen that pattern and core makers engaged in these industries are not

technically aware of productivity improvement techniques. So a need is felt to make a

core of optimum shape in order to improve productivity.

OBJECTIVES

Various Objectives of our work are:-

1)  To reduce time of chamfering and labour cost.

2)  To reduce the cutting tool cost for chamfering.

3)  To increase productivity by reducing scrap production due to chamfering.

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 48/81

-37-

CHAPTER: 4

EXPERIMENTAL WORK 

4.1 EXPERIMENT INTRODUCTION:

In this present work a new shaped core for socket has to be developed and cast a

socket using this core. The effect of using this core on labour cost of chamfering,

scrap production during chamfering, time of chamfering and cutting tool cost will be

studied.

4.2 WORK PLACE:

The experimental work has been done in B.N. Industry, Jalandhar city.

4.3 PROCEDURE FOR MAKING CORE BOX, PATTERN OF

SOCKET:

The core box and pattern were first made from wood on lathe machine in the machine

shop. The core box and pattern were then got casted of aluminium metal. These were

then machined to the dimensions and shape. The core box is a split type two piece

core box.

Figure: 4.1 Core Box

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 49/81

-38-

Figure: 4.2 Core Box 1ʺ (sectional view)*

Figure: 4.3 Core Box 1 ʺ (sectional view)*

*All dimension are in mm.

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 50/81

-39-

Figure: 4.4 Pattern of Socket

Table: 4.3.1 Dimensions of Pattern of Socket 1ʺ 

Sr. No. Dimension Pattern of Socket

1. Length of Pattern excluding core print 43 mm

2. Length of Pattern including core print 61 mm

3. Diameter of Pattern 35 mm

Table: 4.3.2 Dimensions of Pattern of Socket 1 ʺ 

Sr. No. Dimension Pattern of Socket

1. Length of Pattern excluding core print 50.5 mm

2. Length of Pattern including core print 70.5 mm

3. Diameter of Pattern 43.5 mm

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 51/81

-40-

4.4 MANUAL CORE MAKING:

In the single piece and small-lot production of castings, it is common practice to make

cores manually in core boxes and with the aid of templates. The production of cores in

core boxes is the most widespread method which is rather simple and effective. Firstthe box halves are fastened with clamps and the box is placed vertically on the bench.

The sand is then gradually rammed, layer after layer, into the box, it is trimmed off 

level along the top face, a reinforcement wire is inserted, the core is vented and then

the box is gently knocked with a mallet to facilitate the core removal. It should be

remembered that too sharp and hard blows can deform the core. Next the clamps are

loosened; one box half is lifted, the core is withdrawn from the box, and placed it on

 plate for transportation to an oven. Here it is dried properly. The core is now ready for 

use.

Table: 4.4.1 Materials Required for Core

Sr. No. Material Quantity

1 Sand 60%

2 Dextrin 40%

3 Water 

65 parts water to 100 parts by mass of 

 powder.

Figure: 4.5 New Cores

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 52/81

-41-

Figure: 4.6 New Core of socket 1ʺ* 

Table: 4.4.2 Dimensions of New Core of socket 1ʺ 

Sr. No. Dimension New Core

1.  Length of core including core print 61mm

2.  Length of core excluding print core 43mm

3.  Diameter of Core 29mm

4. 

Diameter of core at ends 31mm5.  Length of core print 9mm

6.  Diameter of core print 29mm

Figure: 4.7 Existing Core of Socket 1ʺ**All dimensions are in mm.

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 53/81

-42-

Table: 4.4.3 Dimensions of Existing Core of socket 1ʺ 

Sr. No. Dimension Existing Core

1. Length of core excluding core print 43mm

2. Length of core including print core 61mm

3. Length of core print 9mm

4. Diameter of core 29mm

Figure: 4.8 Existing Core of Socket 1 ʺ*

Table: 4.4.4 Dimensions of Existing Core of socket 1 ʺ 

Sr. No. Dimension Existing Core

1. Length of core excluding core print 50.5mm

2. Length of core including print core 70.5mm

3. Length of core print 10mm

4. Diameter of core 36.5mm

Figure: 4.9 New Core for socket 1 ʺ*

Table: 4.4.5 Dimensions of New Core of socket 1 ʺ 

*All dimensions are in mm.

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 54/81

-43-

Sr. No. Dimension New Core

1 Length of core including core print 70.5mm

2 Length of core excluding print core 50.5mm

3 Diameter of Core 36.5mm4 Diameter of core at ends 38.5mm

5 Length of core print 10mm

6. Diameter of core print 29mm

4.5 PROCEDURE FOR MAKING MOULDING SAND, MOULD

AND CASTING OF SOCKET: 

 None of the natural sand possesses the required qualities to the required extent. They

may lack in one or more of these properties which we have to make up by artificial

means to make the sand suitable for use. Sand mixing is the process through which we

add those materials to the sand which are rich in such characteristics which the sand

lacks. Sand to be used in moulding should be properly conditioned before use in order 

to obtain good castings, since most of the defects which occur in castings are due to

improper conditioning of the sand. It holds good equally for the new as well as old or 

used sand. Proper conditioning means the uniform distribution of the clay bond over the said grains, even distribution and proper control of the moisture content in the

sand and sorting out the foreign materials like nails,  gaggers and other metal pieces

from the sand by ridding and a thorough mixing of the entire sand mass. Even today

above operation is carried out by hand in most of the small foundries. Since no testing

equipment is normally available in such foundries, the sand condition is judged by

moulders themselves by virtue of their practical experience only and the quality of the

castings produced in such foundries entirely depends upon this factor. A common

 physical test, which is generally followed by most of, moulders, for judging the sand

condition is to grip a handful of the prepared foundry sand and then relieve the

 pressure of the fingers. The sand mass thus produced is broken into two pieces by

hand and the edges formed at the broken section are carefully observed. If there is no

deformation in the edges the sand is supposed to properly conditioned. If the upper 

surface of the broken pieces appears to be setting, down gradually, as if it is being

compressed, it indicates high moisture content. Gradual separation of sand grains, as

if they are being sprinkled from the parted surfaces, indicate a weak-bond and low

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 55/81

-44-

moisture content. Mixing of sand by hand is performed by first collecting the sand,

together with the other constituents to be mixed in it, in the form of a help and then

 pouring adequate amount of water on to it. after keeping it as such for some time it is

turned upside down by means of a shovel and the operation repeated several times to

ensure thorough mixing of different constituents. It is then riddled to remove the

foreign material from it and thus it is ready to use. The sand for mould is properly

mixed with molasses in Muller. This sand is used for making the mould. The core is

 placed in the drag. The cope and drag are then fitted together and clamped properly.

Molten metal is then poured into it through sprue. After sometime, the casting gets

cooled. In this way we get the required specimen.

Figure: 4.10 Making of Moulds

Table: 4.5.1 Materials Required for Mould

Table: 4.5.2 Materials Required for Making Socket

Sr. No. Material

1 Sand

2 Molasses

3 Coal dust

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 56/81

-45-

Figure: 4.11 New pieces of Socket

Table: 4.5.3 Dimensions of new piece of Socket 1ʺ 

Sr. No. Dimension New piece of socket

1.  Length 43mm

2.  Thickness 3mm

3. Outer Dia. 35mm

4. Inner Dia. 29mm

5. Weight 172gms

Table: 4.5.4 Dimensions of Existing piece of Socket 1ʺ 

Sr. No.  Material Quantity

1 Pig Iron 60%

2 Mild steel scrap 40%

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 57/81

-46-

Sr. No. Dimension Existing piece of socket

1. Length 43mm

2. Thickness 3mm

3. Outer Dia. 35mm

4. Inner Dia. 29mm

5. Weight 180gms

Table: 4.5.5 Dimensions of New piece of Socket 1 ʺ 

Sr. No. Dimension New piece of socket

1. Length 50.5mm

2. Thickness 3.5mm

3. Outer Dia. 43.5mm

4. Inner Dia. 36.5mm

5. Weight 274gms

Table: 4.5.6 Dimensions of Existing piece of Socket 1 ʺ 

Sr. No. Dimension Existing piece of socket

1. Length 50.5mm

2. Thickness 3.5mm

3. Outer Dia. 43.5mm

4. Inner Dia. 36.5mm

5. Weight 286gms

4.6 TOOLS & EQUIPMENTS REQUIRED:

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 58/81

-47-

Table: 4.6.1 Tools & Equipments Required

The function of various tools &equipments is explained as:

4.6.1 Shovel: It consists of iron pan with a wooden handle. It can be used for mixingand conditioning the sand and then transferring the mixture in some container.

Figure: 4.12 Shovel

Sr. No. Name of Tool/Equipment

1 Shovel2 Trowels

3 Lifters

4 Strike off bar 

5 Vent wire

6 Draw spike

7 Licks

8 Sprue pin9 Sprue cutter 

10 Bellows

11 Files

12 Moulding boxes

13 Ladle

14 Riddle

15 Muller 16 Furnace

17 Rotary blower 

18 Weighing scale

19 Weights

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 59/81

-48-

4.6.2 Trowels: These are used for finishing flat surfaces and corners inside a mould.

Common shapes of trowels are shown as under. 

They are made of iron with a wooden handle.

Figure: 4.13 Trowel Figure: 4.14 Trowel

4.6.3 Lifter: A lifter is a finishing tool used for repairing the mould and finishing the

mould sand. Lifter is also used for removing loose sand from mould.

4.6.4 Strike off bar: It is a flat bar, made of wood or iron to strike off the excess sand

from the top of a box after ramming. Its one edge made bevelled and the surface

 perfectly smooth and plane.

Figure: 4.17 Strike Off Bar

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 60/81

-49-

4.6.5 Vent wire: It is a thin steel rod or wire carrying pointed edge at one end and a

wooden handle or a bent loop at the other. After ramming and striking off the excess

sand it is used to make small holes, called vents, in the sand mould to allow the exit of 

gases and steam during casting.

Figure: 4.18 Vent wire

4.6.6 Draw Spike: It is a tapered steel rod having a loop or ring at its one end and a

sharp point at the other. It is used to tap and draw patterns from the mould.

Figure: 4.19 Draw Spike

4.6.7 Slicks: They are used for repairing and finishing the mould surfaces and edges

after the pattern has been withdrawn. The commonly used slicks are heart and leaf,

square and heart, spoon and bead and heart and spoon.

Figure: 4.20 Slick 

4.6.8 Sprue Pin: it is a tapered rod of wood or iron which is embedded in the sand

and later withdrawn to produce a hole, called runner, through which the molten metal

is poured into the mould.

4.6.9 Sprue Cutter: It is also used for the same purpose as a sprue pin but there is a

marked difference between their uses in that the cutter is used to produce the hole

after ramming the mould. It is in the form of a tapered hollow tube which is inserted

in the sand to produce the hole.

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 61/81

-50-

Figure: 4.21 Sprue Cutter

4.6.10 Bellow: It is used to blow but the loose or unwanted sand from the Surface and

cavity of the mould.

Figure: 4.22 Bellow

4.6.11 Moulding Boxes or Flasks: The moulding boxes or flasks used in sand

moulding are of two types:

(a) Closed moulding boxes.

(b) Open type of snap flasks.

These boxes used in sand moulding may be made of wood, cast iron or steel. They

consist of two or more parts. The lower part is called drag, the upper part cope and all

the intermediate parts, if used, cheeks. All the parts are individually equipped with

suitable means for clamping during pouring. Wooden flasks are generally used in

green sand moulding. Dry sand moulds always require metallic _boxes because they

are heated for drying. Large and heavy boxes are made from cast iron or steel and

carry handles and grips as they are manipulated by cranes or hoists etc. The closed

metallic flasks may have a rectangular or round shape.

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 62/81

-51-

Figure: 4.23 Rectangular flasks

A snap flask is hinged in one corner and it is rectangular in shape and it is made of 

wood. It is used for bench moulding and number .of moulds can be made from same

set of flask.

4.6.12 Ladles: They are used to receive molten metal from the melting furnace and

 pour the same into the mould. Their size is designated, by their metal holding

capacity. Small hand shank ladles, used by a single moulder, are provided with only

one handle and are made in different capacities up to a maximum of 20 kg. Medium

and large size ladles are provided with handles on both sides to be handled by two

moulders. They are also made in different sizes with their capacity varying from 30

kg to 150 kg. A typical hand shank ladle is shown.

Figure: 4.24 Handle Ladle

Extremely large sizes, with capacities ranging from 250 kg to 1000 kg are found in

crane ladles. Geared crane ladles can hold even more than 1000 kg of molten metal.

They facilitate a better pouring control than the ungeared ladles and ensure more

safety for workers.

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 63/81

-52-

All the ladles, irrespective of their size, consist of an outer casing made of steel or 

 plate bent in proper shape and then welded. Inside this casing is provided a refractory

lining. At its top the casing is .shaped to have a controlled and well-directed flow of 

molten metal.

Figure: 4.25 Handle Ladle

4.6.13 Muller

Figure: 4.26 Muller

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 64/81

-53-

It is possible to mix foundry sand by hand, and quite successful moulds have been

made using this method for many generations, but the amateur foundry man will not

have the space or time to laboriously mix small quantities of sand to the quality that a

Muller can.

4.6.14 Rotary Furnace

Figure: 4.27 Rotary Furnace

The rotary melting furnace is the most flexible and universal design of equipment to

recycle aluminium scrap. Due to the nature of its operation all scrap forms can be

recycled with good results. The rotary furnace is rotated either by a friction drive

wheel system or a positive rack/pinion or chain drive depending upon the size and

 production requirements. A single door is utilized with either vertical, horizontal

rotation or a pendulum type swing arrangement depending upon the plant layout. A

high efficiency fume extraction system is provided either fixed directly on the furnace

which tilts with the furnace and exhausts through a rotary joint or simply by a high

level fume collection hood / housing. Rotary furnaces have been traditionally static

 but over recent years tilting designs have been implemented due to the many

advantages with regard to reduced cycle times, yields and consumptions. Mechatherm

engineer rotary furnaces to suit all applications and client requirements. We can

supply oxy-fuel burner system, cold air burner operating on either fuel oil or gas.

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 65/81

-54-

CHAPTER: 5

RESULTS 

1. 

Reduction in Scrap Production for socket 1ʺ: - For calculating the reduction inscrap production due to chamfering, first the weights of sockets casted by using

existing and new core are noted. These two types of weights of sockets are

compared for calculating the reduction in scrap production. For this purpose, 25

sockets of each type are weighed and their weights are as given below:

Socket No. Weight of socket 1ʺ in gms.

with existing core

1 180

2 180

3 181

4 180

5 179

6 182

7 180

8 180

9 180

10 179

11 180

12 181

13 180

14 181

15 182

16 182

17 180

18 179

19 179

20 17921 179

22 179

23 179

24 179

25 180

Total weight of 25 sockets 4500

Weight per socket 180

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 66/81

-55-

Difference in weight of one socket (with existing and new core) = 8gms

Reduction in scrap production = 8gms

% age of scrap reduction = 8 x 100/180

= 4.4%

Socket No. Weight of socket 1ʺ in gms.

with new core

1 1722 172

3 172

4 172

5 173

6 173

7 172

8 172

9 172

10 172

11 172

12 171

13 172

14 172

15 172

16 173

17 173

18 173

19 171

20 171

21 171

22 17223 171

24 172

25 172

Total weight of 25 sockets. 3300

Weight per socket 172

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 67/81

-56-

2.  Reduction in Scrap Production for socket 1 ˝ : - Reduction in scrap production

is calculated in the same manner which has been used previously in case of  socket

1ʺ and is explained as below: 

Socket No. Weight of socket 1 ʺ in gms.

With existing core

1 286

2 286

3 286

4 2865 286

6 288

7 286

8 286

9 287

10 286

11 286

12 286

13 286

14 28615 286

16 287

17 287

18 286

19 286

20 285

21 285

22 284

23 286

24 285

25 286

Total weight of 25 sockets 7150

Weight per socket 286

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 68/81

-57-

Socket No Weight of socket 1 ʺ in gms

with new core

1 274

2 274

3 274

4 274

5 274

6 274

7 274

8 274

9 275

10 275

11 274

12 27413 274

14 274

15 274

16 274

17 274

18 274

19 275

20 274

21 274

22 27223 274

24 273

25 274

Total weight of 25 sockets 6850

Weight per socket 274

Difference in weight of one socket (reduction in scrap) = 12gms

Reduction in scrap production = 12gms

% age of scrap reduction = 12 x 100/286

= 4.2%

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 69/81

-58-

3.  Elimination of Chamfering Time: - The sockets casted by using new cores do

not require chamfering. The time consumed for chamfering sockets casted by

using existing cores is saved. The time required for chamfering a lot of 10 sockets

each casted by using existing cores is observed and is given as below:

Observation No. Time of chamfering Socket 1ʺ 

with existing core

1 576 seconds

2 576 seconds

3 579 seconds

4 575 seconds

5 574 seconds

Total 2880 seconds

Observation No.Time of chamfering Socket 1 ʺ 

with existing core

1 578 seconds

2 576 seconds

3 576 seconds

4 575 seconds

5 574 seconds

Total 2880 seconds

Chamfering time for both the sizes of sockets is equal. Therefore reduction of time is

the same in both cases.

Time of chamfering for 10 sockets (with existing core) = 576 seconds

Time of chamfering for 10 sockets (with new core) = Nil

Therefore reduction in time of chamfering for 10 sockets = 576 seconds

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 70/81

-59-

4.  Reduction in Labour Cost of Chamfering: - The sockets casted by using new

cores do not require any labour cost because chamfering is not needed in this case.

The labour cost for chamfering sockets casted by using existing cores is saved.

The labour cost required for chamfering sockets casted by using existing cores is

saved. The labour cost required for chamfering sockets casted by using existing

cores is calculated by taking observations on hourly basis on a machine which is

used for chamfering as given below:

Observation No. No. of Sockets 1ʺ Chamfered

1 66

2 64

3 604 60

Total 250

Observation No.No. of Sockets 1 ʺ Chamfered

1 60

2 623 66

4 62

Total 250

Total number of sockets chamfered in 4 hours is the same in both cases. Therefore

reduction in labour cost also the same.

 No. of pieces chamfered in 4 hours = 250

 No. of pieces chamfered in 8 hours = 500

Labour cost of 8 hours = Rs. 225

Labour cost per piece (with existing core) = 45 paisa/piece

Labour cost per piece (with New core) = Nil

Reduction in labour cost per piece = 45paisa/piece

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 71/81

-60-

5.  Reduction in Cutting Tool Cost for Chamfering: - The sockets casted by using

new cores do not require chamfering. Therefore there is no cutting cost for 

chamfering in this case. The cutting tool cost for chamfering sockets casted by

using existing cores is saved. The cutting tool cost required for chamfering

sockets casted by using existing cores is calculated as given below:

For Socket 1ʺ 

 No. of cutting bits used = 3

Cost of 3 cutting bits = Rs. 360

 No. of sockets chamfered (with existing core) = 5000

Cost of chamfering of sockets (with existing core) = 360 X 100/5000

=  7.2 paisa/socket

Cost of chamfering (with new core) = Nil

Reduction in expenses on cutting tool = 7.2 paisa/socket

For Socket 1 ʺ 

 No. of cutting bits used = 3

Cost of 3 cutting bits = Rs. 360

 No. of sockets chamfered (with existing core) = 5000

Cost of chamfering of sockets (with existing core) = 360 X 100/5000

= 7.2 paisa/socket

 No. of sockets chamfered (with existing core) = 5000

Cost of chamfering (with new core) = Nil

Reduction in expenses on cutting tool = 7.2 paisa/socket

Cutting tool cost of chamfering for both the sizes of socket is equal. Therefore

reduction in cutting tool cost is the same in both cases.

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 72/81

-61-

5.1 ANALYSIS

The overall cost estimation method is used for analysis of results. Cost estimation

includes only scrap cost, labour cost of chamfering, cutting tool cost for chamfering.

Only these elements are included because these are the element which get affected

due to use of new core. Other cost elements remain the same for both cases. In case of 

casting with existing core one Ton casting contains 5555 sockets of 1ʺ.

Labour cost of chamfering 1 Ton sockets = 5555 x 45

= 249975 paisa

2500Rs.

Cutting tool cost of chamfering 1 Ton sockets = 5555 x 7.2

= 39996 paisa

400Rs.

Scrap produced during chamfering 1 Ton sockets = 44 Kg.

Scrap cost = 44 x 35

= 1435 Rs.

COST ANALYSIS – EXISTING CORE vs. NEW CORE

Summary of cost (in Rupees) Per Ton for socket 1ʺ 

Cost ElementUsing existing core

Rs. Per Ton

Using New core

Rs. Per TonScrap in chamfering 1435 0

Labour cost of chamfering 2500 0

Cutting tool cost of chamfering 400 0

Total cost 4335 0

Profit = 4335 Rs.

By comparing the new core with the existing core use, the overall profit per ton

increased by Rs. 4335 in case of adopting new core in socket 1ʺ.

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 73/81

-62-

In case of casting with existing core one Ton casting contains 3496 sockets 1 ʺ 

Labour cost of chamfering 1 Ton sockets = 3496 x 45

= 157320 paisa

1573Rs.

Cutting tool cost of chamfering 1 Ton sockets = 3496 x 7.2

= 25171.2 paisa

251Rs.

Scrap produced during chamfering 1 Ton sockets = 42 Kg.

Scrap cost = 42 x 35

= 1470 Rs.

Summary of cost (in Rupees) Per Ton for socket 1 ʺ 

Cost ElementUsing existing core

Rs. Per Ton

Using New core

Rs. Per Ton

Scrap in chamfering 1470 0

Labour cost of chamfering 1573 0

Cutting tool cost of 

chamfering251 0

Total cost 3294 0

Profit = 3294 Rs.

By comparing the new core with the existing core use, the overall profit per ton

increased by Rs. 3294 in case of adopting new core in socket 1 ʺ.

So it shows the advantages of using new cores.  

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 74/81

-63-

5.2 DISCUSSION

Scrap Production :- Experiments carried out with new shaped core show that scrap

due to chamfering is not produced using these cores because chamfering is not needed

in this case whereas scrap produced due to chamfering increases with an increase in

number of pieces in case of using existing core. The variation in scrap production

with existing and new core for socket 1ʺ is shown graphically as: 

0

100

200

300400

500

600

700

800

900

10 20 30 40 50 60 70 80 90 100

Pieces with existing

core

Pieces with new shapedcore

 

Number of Pieces

Graph 1- Plot of scrap production during chamfering for pieces with existing and

new core for socket 1ʺ 

0

1000

2000

3000

4000

5000

6000

7000

8000

10 20 30 40

Pieces with existing core

Pieces with new shaped

core

 

Graph 2- Bar graph showing weight of socket 1ʺ with existing and new shaped core 

   S  c  r  a    W  e   i   h   t   i  n   G  m  s .

No. Of Pieces

   W  e   i   h   t   i  n

  m  s .

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 75/81

-64-

0

200

400

600

800

1000

1200

1400

10 20 30 40 50 60 70 80 90 100

Pieces with existing core

pieces with new shaped

core

 

Graph 3- Plot of scrap production during chamfering for pieces with existing and

new core for socket 1 ʺ 

0

2000

4000

6000

8000

10000

12000

10 20 30 40

Pieces with existing core

Pieces with new shaped core

 

Graph 4- Bar graph showing weight of socket 1 ʺ.with existing and new shaped core. 

   W  e   i  g   h   t   i  n  g  m  s .

 

No. Of Pieces

   S  c  r  a    W  e   i   h   t   i  n   G  m  s .

 

No. Of Pieces

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 76/81

-65-

Labour Cost of Chamfering :- Our experiments have shown that there is no labour 

cost of chamfering using new core. With existing core labour cost is 45  paisa/piece.

The variation in labour cost of chamfering for existing and new core products is

shown graphically as: 

0

100

200

300

400

500

100 200 300 400 500 600 700 800 900 1000

Pieces with existing

core

Pieces with new shaped

core

 

Number of Pieces

Graph 5- Plot showing labour cost of chamfering for pieces with existing and new

core for sockets 1ʺ and 1 ʺ.

0

50

100

150

200

250

300

350

400

200 400 600 800

Pieces with existing core

Pieces with new shaped

core

 

Graph 6 –  Bar graph showing labour cost of chamfering for pieces with existing and

new core for sockets 1ʺ and 1 ʺ.

   L  a   b  o  u  r  c  o  s   t   i  n   R  s .

 

   L  a   b  o  u  r  c  o  s   t   i  n   R  s .

 

No. Of Pieces

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 77/81

-66-

Cutting Tool Cost of Chamfering :- Our experiments have shown that there is no

cutting tool cost of chamfering using new core. With existing core cutting tool cost is

7.2 paisa/piece. The variation in cutting tool cost of chamfering for existing and new

core products is shown graphically as: 

0

50

100

150

200

250

300

350

400

1000 2000 3000 4000 5000

Pieces with existing core

Pieces with new shaped

core

 

Number of Pieces

Graph 7 –  Plot showing cutting tool cost during chamfering for pieces with existing

and new core for sockets 1ʺ and 1 ʺ 

0

50100

150

200

250

300

350

400

1250 2500 3750 5000

pieces with existing

core

pieces with new core

 

Number of Pieces

Graph 8 –  Bar graph showing cutting tool cost during chamfering for pieces with

existing and new core for sockets 1ʺ and 1 ʺ.

   C  u   t   t   i  n    t  o  o   l  c  o  s   t   i  n   R  s .

 

   C  u   t   t   i  n    t  o  o   l  c  o  s   t   i  n   R  s .

 

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 78/81

-67-

Time of Chamfering :- Our experiments have shown that no time is needed for 

chamfering using new core. With existing core time consumed is 576  seconds for 10

 pieces. The variation in time of chamfering for existing and new core products is

shown graphically as: 

0

1000

2000

3000

4000

5000

6000

7000

10 20 30 40 50 60 70 80 90 100

Pieces with existing

core

Pieces with new

shaped core

 

Number of Pieces

Graph 9 –  Plot showing time of chamfering for pieces with existing and new core for 

sockets 1ʺ and 1 ʺ.

0

50

100

150

200

250

300

350

400

1250 2500 3750 5000

Pieces with existing core

Pieces with new shaped

core

 

Number of pieces

Graph 10 –  Bar graph showing time of chamfering for pieces with existing and new

core for sockets 1ʺ and 1 ʺ.

   T   i  m  e  o   f  c   h  a  m   f  e  r   i  n  g   i  n  s  e  c  o  n   d  s

 

   T   i  m  e  o   f

  c   h  a  m   f  e  r   i  n  g   i  n  s  e  c  o  n   d  s

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 79/81

-68-

CHAPTER: 6

CONCLUSION

1) Experiments have shown that no scrap is produced due to chamfering in case of using new core. It is because chamfering is not needed in this case. Scrap produced

during chamfering in case of existing core with socket 1˝ is 8 gm. per piece. Hence in

case of metal casting of 2 tonnes there is saving of 88 kilogram of scrap. The market

cost of scrap is about Rs. 22 per kilogram, whereas the cost of malleable casting is Rs.

62 per kilogram. Hence there can be saving Rs. 3520 in that case. In case of socket

1 ʺ, there is saving of 12 gm per piece.

2) No time is needed for chamfering in case of using new core whereas time of 

chamfering n case of using existing core is 576 seconds for 10 pieces.

3) There is no cutting tool cost in case of using new core. In case of using existing

core, the cost of cutting tools for chamfering is Rs. 360 for 5000 pieces.

4) There is no labour cost for chamfering in case of using new core. In case of using

existing core the labour cost is 45 paisa per piece.

The productivity has improved with the use of new shaped core due to

reduction in scrap production, time needed for chamfering, cutting tool cost and

labour cost for chamfering.

6.1 SCOPE OF IMPROVEMENT

The chamfering work is needed in all other pipe fittings including Tee, Elbow and

Union of different sizes. There is scope of eliminating the need for chamfering these

items also. New shaped cores for these items need to be developed. Productivity of 

 pipe fitting industry will improve to great extent by using such cores. 

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 80/81

-69-

REFERENCES

1.  C. Li and B.G. Thomas 2000, “Analysis of the potential productivity of continues

cast moulds.” pp. 595-611

2.  Thomas Schroeder 2004, “Starch and dextrin based adhesives.” 

3.   Nicolas Perry, Magali Mauchand, Alain Bernard 2004, “Costs models in design

and manufacturing of sand castings products.” IDMME. 

4.  L.A. Dobrazański, M. Krupiński, J.H. Sokolowski, P. Zarychta and Wlodarczyk -

Fligier 2006, “Methodology of analysis of casting defects” Journal of 

Achievements in materials and manufacturing engineering, Vol. 18 Issue 1-2

5. 

J. Dańko, M. Holtzer, R. Dańko 2007, “Factors influencing selection of effectivereclamation techniques and assessment methods of the reclaimed material

quality.” Archives Of Foundry Engineering, ISSN (1897-3310) Vol. 7, pp. 29-32

6.  A. Fedoryszyn 2007, "Assessment of systems for mechanisation of casting

 production.”, Archives of Foundry Engineering, Vol. 7 pp. 83-86

7.  F. Peters, R. Voigt, S.Z. Ou and C. Beckermann 2007,“Effect of mould expansion

on pattern allowances nib sand casting of steel”, International Journal of Cast

Metals Research , Vol.20 No. 5

8.  Liu Weihua, Li Yingamin, Qu Xueliang, Liu Xiuling 2008,“Study on binder 

system of CO2-cured phenol-formaldehyde resin used in foundry”, China

Foundry, Vol. 5 No. 2

9.  E.O. Olakanmi and A.O. Arome 2009, “Characterisation of the core-binding

 properties of fatty based oils”, International Journal of Physical Sciences, Vol.

4(11), pp. 623-628

10. Slavomír Peľák, Rudolf Mišičko, Dagmar Fedáková, Jana Bidulská 2009,

“Between the dendrite structure quality, the casting technology and the defects in

continuously cast slabs.” Materials Engineering, Vol. 16 No. 4 

11. A.K.M.B. Rashid 2010, “Casting defects.” Advanced Topics in Foundry

Engineering

12. Lakashamanan Singaram 2010, “Improving quality of sand casting using Taguchi

method and Ann analysis.” International Journal on Design and Manufacturing

Technologies, Vol. 4 No. 1

7/29/2019 Pipe Line Socket Thesis

http://slidepdf.com/reader/full/pipe-line-socket-thesis 81/81

13. R. Venkataraman 2010, “Innovative ideas for improving foundry productivity and

casting quality.” Transactions of 58th IFC

14.  Nishikant M Ejrekar, Vicky Kapur 2011, “Refractory coating –  A Practical

A pproach.” Foundry, Vol. 23 No. 1 

15. Anil Barik, “Effects and controlling measures of directional solidification in

castings towards quality improvement.”, foundry Journal, Vol. 23 No. 5

16. Q.S.I. Fayomi, O.O. Ajayi and A.P.I. Popoola 2011,“Suitability of local  binder 

compositional variation on silica sand for foundry core-making”, International

Journal of Physical Sciences, Vol. 6(8), pp. 1940-1946

17. J.S. Colton 2011, “Casting defects & design issues.” Manufacturing Processes and

Systems

18. R. Vinayagasundaram, V.R.  Nedunchezhian 2012, “Application OF Modern

Technology in Foundry and its Impact on Productivity and Profitability”,

European Journal of Social Sciences, Vol. 29, pp. 320-330

19. Bharathi Rajkumar K. and Gukan Rajaram 2012, “Productivity enhancement in

foundry industry using lean tools.” Indian Foundry Journal, Vol. 58, No. 9 

20.  Norberto T. Rizzo Downes, Ramon D. Duque, “Ten steps to improving casting

yield in ductile iron foundries.”

21. IS: 4759-1979, Indian Standard Specification for Hot-Dip Zinc Coatings on

Structural Steel and Other Allied Products.

22.  IS: 2108-1977, Indian Standard Specification for Black Heart Malleable Iron

Castings.

23.  IS: 2107-1977, Indian Standard Specification for White Heart Malleable Iron

Castings.

24. IS: 2633-1972, Method for Testing Indian Standard Uniformity of Coating on

Zinc Coated Articles.

25. IS: 6745-1972, Indian Standard Method for Testing Mass of Coating on Zinc


Recommended