simulation and experimentation in feeder design_secure

SIMULATION AND EXPERIMENTATION IN
CASTING FEEDER DESIGN
A PROJECT REPORT

Submitted by

PATEL ARTH G.
POONAWALA TAHA Y.
SANGHANI DARSHAK V.
SUKHADIA DHAVAL V.

in partial fulfillment for the award of the degree
of

BACHELOR OF TECHNOLOGY
in

MECHANICALENGINEERING

DEPARTMENT OF MECHANICAL ENGINEERING
FACULTY OF TECHNOLOGY
DHARMSINH DESAI UNIVERSITY, NADIAD

DEC 2014

i

ACKNOWLEDGEMENT
This report would not have been possible without the kind support and help of many
individuals. We would like to extend our sincere thanks to all of them. We thank Prof. R. V.
SOLANKI, Prof. N. A. VORA and Mr. H. T. PATEL for their kind co-operation with
enough encouragement towards the completion of this report.
We are highly indebted to Dr. MAYUR SUTARIA (CSPIT, Changa) for his guidance as
well as for providing necessary information regarding this work and his support in completing
it. We thank Prof. RAKESH BAROT (BVM, V V Nagar) for guiding us at the right moment.
We would like to express our gratitude and thanks to industry person Mr. RAJESHKUMAR
TOSHNIWAL (M/s. UTSAV METALS, NADIAD) for using his foundry to carry out the
experiment. We specially thank Mr. BALKRISHNAN for providing a helping hand at various
stages.
We are also grateful to the lab technicians of our departmental workshop for giving a helping
hand at all times without any discontent.
We would like to express our gratitude towards our parents for giving us moral support to work
hard. Our heartfelt thanks and appreciations to our classmates who supported us in taking up
this work and people who have willingly helped us out with their abilities.

ii

ABSTRACT

In metal casting, defect free castings which require least finishing operations has been the
primary goal since the inception of technology. There is always a compromise between the cost
involved in the production of cast component and the quality required. Besides, it is always
desired that the yield of casting is maximized against the volume of feeder/riser accommodated
to meet the solidification shrinkage requirement.
The shrinkage porosity defect is one of the most common solidification defects of sand casting
process. It occurs in the thickest sections of casting which is possessing lasting freezing point.
The practical approach of design of feeder has high factor of safety and due to that oversized
feeders have normally been designed and tested on shop floor. This consumes lot of time and
resources. Thus, there is a need for computer aided optimal feeder design coupled with
solidification simulation so as to reduce the no. of the shop-floor trials and obtain enhanced
yield and high quality, in minimal possible time.
The initial design is the aluminium casting part (without feeder) which is simulated online in
Efoundry to detect the location of hotspot. Then a feeder is designed on the following steps:
determination of the feeder-neck connection point on the casting surface, initial feeder design
and feeder shape optimization using Efoundry till the hotspot is obtained in the feeder itself.
The same part is then experimentally poured and verified with cut-section. It is observed from
actual pouring that shrinkage cavity had shifted towards the feeder whereas it remained at the
center of the junction in the non-feeder part. It is concluded at the end that the selection of
proper feeder affects the quality of casting during solidification.

TABLE OFCONTENTS
NO. TITLE PAGE
Acknowledgement i
Abstract ii
Table of Contents iii
List of Figures v
List of Tables vi
Nomenclature vii
1. Introduction to Metal Casting 1
1.1. Founding or Casting 1
1.2. History of Casting 1
1.3. Casting Process Steps 2
1.4. Applications 2
2. Quality of Castings 3
2.1. Need of Quality 3
2.2. Defects in casting 4
2.3. Significance of Defects 5
2.4. Methods of Improvement 6
2.5. Use of Computers 6
3. Literature Survey 7
4. Research Problem Definition 9
4.1. Motivation 9
4.2. Goal 10
4.3. Research Objectives 10
4.4. Research Approach 10
4.5. Scope 10
5. Solidification and Shrinkage 11
5.1. Mechanism of Solidification in Pure Metals 11
5.2. Shrinkage 11
5.2.1. Liquid Contraction 11
5.2.2. Solid Contraction 12
5.2.3. Solidification Contraction 12
5.3. Solidification Simulation 12

6. Pouring and Feeding Castings 13
6.1. Gating System 13
6.1.1. Elements of Gating System 13
6.1.2. Gating System Design 14
6.2. Feeders 16
6.2.1. Principles of Feeding 16
6.2.2. Types of Feeders 17
6.2.3. Feeder Design 18
6.3. Casting Yield 20
7. Feeder Design Simulation and Experimentation 21
7.1. Junction Definition 21
7.2. Pattern, Feeder and Mould Design 22
7.2.1. Pattern Allowances and Design 22
7.2.2. Feeder Design using Caine’s Method 22
7.2.3. Sprue, Runner and Gate Design 24
7.2.4. Moulding Sand & Mould Box 26
7.3. Simulation in Efoundry 26
7.4. Experimentation in Foundry 30
8. Experimental Results and Discussion 32
9. Conclusion 36
Future Work 37
References 38

v

LIST OFFIGURES
Fig.No. Description Page No.
1.1 The casting process 1
2.1 Defects in casting 4
4.1 Foundry Defect Spectrum 8
5.1 Development of columnar crystals 11
5.2 Solidification contraction regimes in liquid, freezing and solid range 11
6.1 Elements of Gating System 13
6.2 Solidification of a cube casting 16
6.3 Classification of Casting Feeders 17
6.4 Top & Side Feeder Shapes 17
6.5 Progressive Directional Solidification 19
7.1(a) Orthographic View - Selected ‘X’ Junction geometry 21
7.1(b) Isometric View - Selected ‘X’ Junction geometry 21
7.2 Geometry with top feeder 24
7.3 Mould Box Section 25
7.4 Drag pattern 25
7.5 Cope pattern with riser and sprue 26
7.6(a) Efoundry Simulation Step - 1 27
7.6(b) Efoundry Simulation Step - 2 27
7.6(c) Efoundry Simulation Step - 3 28
7.6(d) Efoundry Simulation Step - 4 28
7.7(a) ‘X’ junction solidification simulation 29
7.7(b) ‘X’ junction solidification simulation with feeder of dia D = 62 mm 29
7.7(c) ‘X’ junction solidification simulation with feeder of dia D = 64 mm 29
7.8(a) Sand Preparation 30
7.8(b) Facing Sand application for Drag 30
7.8(c) Rammed Drag 30
7.8(d) Mould cavity in Drag 30
7.8(e) Cope preparation by placing feeder & sprue 31
7.8(f) Venting in the cope 31
7.8(g) Final Drag & Cope Assembly 31
7.8(h) Metal Pouring 31
7.8(i) Final Cast Product 31
8.1 Cast job without feeder 32
8.2 Cut Plane Section in the job without feeder 32
8.3 Cast job with feeder 33
8.4 Cut Plane Section in the job with open top feeder 34
8.5 Top Feeder and its Cut Plane Section 34

vi

LIST OFTABLES
Table
No.
Description
Page
No.
5.1 Solidification shrinkage for major cast metals 12
7.1 Physical Properties of Geometry 21
7.2 Pattern Allowances 22
7.3 Moulding Sand Properties and Composition 26
7.4 Pouring Condition Parameters and Alloy Composition 31

vii

NOMENCLATURE
M Modulus
C Efficiency factor for gating system
H Height
SA Surface area
H Potential head
P Pressure
v Liquid velocity
V Volume
W Specific weight
g Gravitational acceleration
Q Rate of flow
A Area of cross section
W Weight of casting
T Pouring time
Mass density of the molten metal
Solidification time
∈ Feeding efficiency
Shrinkage factor
D Diameter of feeder
X Freezing ratio
Y Yield
Length of neck
Diameter neck
Sprue base diameter
Sprue top diameter
Sprue top area
Gate area
Choke area
k Mould constant
Temperature of solid

Subscripts
c casting
f feeder
N neck

SIMULATION AND EXPERIMENTATION IN CASTING FEEDER DESIGN
Page | 1
Chapter 1
INTRODUCTION TO METAL CASTING
1.1 Founding or Casting
The process in which the metal is first liquefied by properly heating it in a suitable furnace and
pouring the molten metal in a previously prepared mould cavity where it is allowed to solidify
is termed as metal casting. Subsequently, the product is taken out of the mould cavity, trimmed
and cleaned to the required shape. Casting is one of the oldest manufacturing process, and even
today it is the first step in manufacturing of most products.
1.2 History of Casting
Casting is one of the oldest manufacturing methods which dates back to 4000 B.C. In early
years, the axe heads of copper were cast in open stone moulds. During that period gold, silver,
copper, bronze, etc. were widely cast. Some decorated bronze castings could be seen in the
European churches. Also, earlier castings of cast iron were those of cannon shots and grave
slabs. Later, the first foundry center came into existence in the days of Shang dynasty (1766-
1122 BC) in China. The Greeks and Romans revealed the use of decorated ornaments and metal
bells. Moreover, the process of casting was known to certain families only and was considered
as an art and craft.
In 1540, Biringuccio wrote on Metal Founding which was further adopted and reworked by
Reaumur who studied the various factors influencing the production of white, malleable and
gray irons. A no. of foundries came into existence after the British Industrial Revolution.
During the mid-20th
century newer techniques came into existence, the phenomenon of casting
[Fig. 1.1] – The casting process

Page | 2
could be understood better and more no. of young men took interest to develop the subject.
Today, no. of methods have been discovered and patented to obtain the best casting for different
materials e.g. lost wax casting process, centrifugal casting, die casting, etc. Besides, computer
simulation has also been put to effort to analyze and optimize the various factors influencing
the process.
1.3 Casting Process Steps
 Make the Pattern out of wood, metal or plastic
 In case of Sand Casting, select, test and prepare the necessary sand mixtures for mould and
core making.
 With the help of patterns prepare the Mould (a container having a cavity of the shape to be
cast) and necessary Cores (body of sand which is employed to produce cavity in casting).
 Melt the metal/alloy to be cast.
 Pour the molten metal/alloy into the mould and remove the casting from the mould after
the metal solidifies.
 Clean & Finish the casting.
 Test & Inspect the casting.
 Remove the Defects, if any.
 Relieve the casting stresses by Heat Treatment.
 Again inspect the casting.
 The casting is ready for shipping.
1.4 Applications
The growing demand of high precision castings and of intricate designs at lower costs has
helped considerably in the development of Foundry industry. Hardly there is any product
which does not have one or more cast components. Few such applications are:
1. Automobile parts
2. Machine tool structures
3. Turbine vanes
4. Power generators
5. Railway crossings
6. Pump filter and valves
7. Aircraft jet engine blades
8. Communication, Construction and Atomic Energy applications

Page | 3
Chapter 2
QUALITY OF CASTINGS
2.1 Need of Quality
Casting quality is one of the keys to survival in foundry industry. High quality castings depend
on the ability of a casting producer to test and inspect all the raw materials. The ability to
discriminate between imperfections greater than the allowable severity due to various causes
plays an important role in reducing the rejection rate of castings. It is essential that enough
focus is laid down to implement a proper cast product in single pour so as to minimize the
amount of metal loss which is lost as vapourized metal oxide during the melting process. Major
points of concern during the design of a cast product are:
1. Appropriate allowances on the pattern and mould cavity
 As molten metal cools it shrinks depending on its properties i.e. thermal expansion
coefficient and hence proper shrinkage allowance should be added
 The cast product obtained does not have good surface finish generally in sand casting
method and hence proper machining allowance should be added
 Both these factors add to the oversizing of the cavity than the actual job which adds to
the cost of casting
2. Fettling process
 The pathway through which molten metal enters the mould cavity comprises of
elements like runners and gates
 Riser or feeders are placed to accommodate the metal requirement during solidification
 These elements have to be removed of the cast product after solidification and is sent
to the furnace for melting
 This process induces large losses in terms of loss of metal and the defects caused in
the cast product due to improper design of runner, gates and risers
3. Properties of sand and other raw materials
 When the molten metal cools in the mould cavity heat transfer takes place through
the surfaces of mould cavity
 If these surfaces do not provide adequate amount of heat transfer chances are there
that the cavity might blow off or hot spots remain in the core of the component or any
other type of defect is obtained
 Due to chemical reactions or variation in the solubility of molten metal at different
temperature and phase, gases are evolved and need to be vented out properly

Page | 4
Considering the above points and various other affecting factors, the casting need to be
optimized for compromise against a defect free part and the metal loss.
2.2 Defects in Casting
Defects arise in castings due to faulty casting design, faulty method of casting and faulty
workmanship. The casting design defects include faulty a) pattern equipment, b) flask
equipment and rigging, c) gating and risering and d) sand and cores compositions. A defect
may arise from a single clearly defined cause or more generally may be due to a combination
of causes interacting with each other. Common defects in castings as observed are shown in
the fig and they are defined as follows:
1. Blow – a fairly large, well-rounded cavity produced by the gases which displace the molten
metal at the cope surface of the casting due to inadequate venting
2. Scar – a shallow blow, usually found on a flat casting surface
3. Blister – a scar covered by thin layers of metal
4. Gas Holes – entrapped gas bubbles having nearly spherical shape, and occur when an
excessive amount of gases is dissolved in the liquid metal
[Fig.2.1] – Defects in casting

Page | 5
5. Pin Holes – tiny blow holes which occur either at or just below the casting surface and are
found generally in large castings with uniform distribution over the entire casting surface
6. Porosity – very small holes uniformly dispersed throughout a casting due to decrease in gas
solubility during solidification
7. Drop – an irregularly shaped projection on the cope surface of the casting caused by
dropping of sand from cope or other overhanging projections into the mould
8. Inclusion – non-metallic particles in the metal matrix
9. Dross – lighter impurities appearing on the top surface of the casting
10. Dirt – small angular holes obtained when sand particles dropping out of the cope get
embedded on the top surface of the casting and are removed
11. Wash – a low projection on the drag surface of a casting commencing near the gate, caused
by erosion of sand due to the high velocity of jet of liquid metal in bottom gating
12. Buckle – a long, fairly shallow, broad, V-shaped depression occurring in the surface of a
flat casting of a high temperature metal because of the bulging of mould face
13. Scab – a rough, thin layer of metal, protruding above the casting surface, on top of a thin
layer of sand resulting due to separation of sand at that place and the flow of metal through
that path between sand and mould
14. Rat Tail – a long, shallow, angular depression found in thin casting
15. Penetration – in a soft and porous mould surface, the molten metal may flow between the
sand particles up to a distance, into the mould causing rough, porous projections
16. Swell – found on vertical surfaces of casting if the moulding sand is deformed by the
hydrostatic pressure caused by high moisture content in sand
17. Misrun – liquid metal due to insufficient superheat freezes before reaching the farthest
point of the mould cavity
18. Cold shut – insufficient mixing of metal at the intersection of two streams
19. Hot tear – a crack developed due to high residual stresses
20. Shrinkage cavity – improper riser design leads to quick solidification of the metal inside
the mould causing internal or external voids as the shrinkage is not compensated
21. Shift – misalignment between two halves of a mould or of core
2.3 Significance of Defects
Under working conditions, some defects in the casting may be inherent and their significance
can only be established in relation to the function of the casting. Behaviour under service
stresses and environment in most cases is the over-riding consideration and its appearance as
well. Besides, the defect affects its mechanical properties and surface condition.

Page | 6
2.4 Methods of Improvement
A no. of variables like raw material composition variations, furnace operation variation,
variations in pouring of liquid metal, variations in mould, cooling conditions, etc. are to be
manipulated simultaneously to obtain a sound casting with good surface finish. The foundry
which was once considered as an ‘Art’ is now the ‘Science’ of foundry.
Improved strength in a casting can be obtained by manipulating the internal structures of the
casting which in turn involves controlling the rate of heat transfer through the mould and
cooling of casting and modifying additions to influence solidification of casting. Good surface
finish can be obtained by minimizing mould-metal interface reactions and by controlling
evolution of gases during solidification stage. Dimensional accuracy of castings are influenced
by fluidity of melts, shrinkage due to liquid-solid phase transformation, mould materials and
methods of mould making and the casting method employed.
2.5 Use of Computers
The computer represents the most significant and universally applicable development in
business and commercial activity. In particular, there is an overlap between the cost estimating
activity and the determination of the method of production, encompassing both pattern layout
and gating and feeding system designs. Earlier computer applications in foundry industry saw
application in the field of optimizing the weight and cost estimation.
It is possible at present to select the correct choice of mould making, core making and the
casting process if the input data based on design considerations are available by Value
Engineering and Value Analysis. Value Engineering co-relates function with cost and Value
Analysis concerns the selection of correct manufacturing process. Value Analysis finds out the
manner in which the casting part will be stressed in service whereas Value Engineering
suggests the casting procedure. For improving strength and quality of castings Multiple Linear
Regression analysis data, Optimisation and Simulation methods are generally applied.
Regression analysis plots experimental data to find co-relations between the various variables.
Gating and Feeding system design for the given casting are simulated in various softwares like
Novocast (Dewtec Computer Systems Ltd.,UK), ProCAST, Flow-3D, Star-Cast, MAGMA and
SOLIDCast with FLOWCast and OPTICast. These programs perform coupled simulation of
mould filling and solidification for a given casting design with complete feeders, runners and
gating. The widespread availability of powerful, yet low cost computers has opened the
possibility of creating, analyzing and optimizing virtual castings so that quality components
can be produced in minimum no. of shop-floor trials.

Page | 7
Chapter 3
LITERATURE SURVEY
1. Heine et al. in 1968 gave the principle of directional solidification which meets the feed
path requirement. It states that, “If the feeder can be placed on the highest modulus section
of the casting, with progressively thinner (lower modulus) sections extending away, then
the condition of progressive solidification towards the feeder can be met.”
2. Jong et al. in 1991, described that the casting can be separated into different feeding
sections by dividing into simpler shapes at different sections called feeding unit. A feeding
unit is a group of casting sections in which modulus decreases progressively from the
highest modulus section to the lowest modulus section. Each feeding unit requires a
separate feeder.
3. Wu et al. in 1992, gave the critical modulus gradient which is needed to be maintained in
all feeding units to ensure proper feed direction.
4. Chvorinov in 1996, gave one of the earliest geometric based optimization efforts which
was proposed on the modulus method and related it directly to solidification time of a
casting.
5. Wlodawer used Chvorinov’s rule to design the feeders in such a way that the modulus
(M) of the feeder is greater than that of the casting and must increase by 10% from
the casting across the ingate to the feeder for ensuring adequate feeding. He proposed a
relationship between casting, neck and feeder modulus as
: : 1 ∶ 1.1 ∶ 1.2
6. Ravi & Srinivasan in 1996, proposed the Vector Element Method (VEM) which
determines the feed path and location of hot spot inside the casting, using the direction
of the largest thermal gradient at any point inside a casting to move along a path which
leads to a hot spot (a local maxima of temperature with gradients tending to zero).
7. Campbell in 2004 laid out the feeding rules which gives an idea for feeder size
calculation and its location.
8. Jacob, Roschen et. al. in 2004 presented a novel approach to the problem of feeder design
by augmenting genetic algorithms with CAD to optimize the feeder dimensions. Genetic
algorithms based on empirical rules were used as an optimization tool. A 3-dimensional
CAD model of casting is modeled using CAD software. The casting is further divided into
feeding sections and their volume, surface area were calculated. For each feeding section

Page | 8
feeder is designed by using parallel search in the domain of all possible solution and thus a
population of probable solutions are prepared. A fitness function is defined on the basis of
population of solutions, and is used in maximization of casting yield.
They also showed that the required modulus of the feeder is given by the relation:
∗
The value of multiplication factor mainly depends on the cast metal (for steel is 1.3, ductile
iron is 1.15, grey iron is 1.0) and it was proposed by Ravi et al., in 1997.
9. B. Ravi et al. presented various works related to casting simulation in the previous years
some of which are:
a. Location of feeder and its shape selection
b. Framework of feeder design and optimization in order to maximize the yield and
productivity against high rejection rate
c. Feeder neck proportions and a taper section so that modulus of the neck increases
as it moves away from the casting
10. D Joshi, B Ravi in 2009, presented the classification and simulation based design of 3D
junctions in casting wherein VEM was employed to predict the extent of shrinkage porosity
defect and it was validated experiments. A benchmark part with 3D junction was also
presented to show how simulation can be used to predict and prevent the defect by
modifying the junction design.
11. Elizabeth Jacob, Dundesh S. Chiniwar, Savithri S, Manoj M and Roschen Sasikumar
in 2013, carried out simulation based feeder design for metal castings wherein the casting
part alone is simulated and the solidification profile is used to identify the hotspot and
design the feeder. The feeder was further improved and verified with simulation in “Virtual
Feed” software which was then simulated in AUTOCAST software (3D Foundry Tech) to
check for hotspots.
12. E‐Foundry, developed at IIT-B is a part of the NKN (National Knowledge Network)
mission to connect knowledge providers and seekers through a high bandwidth network.
Users can freely access the teaching content developed in IIT‐B, to update their knowledge
in casting design and simulation. It also offers online simulation lab, which accepts a 3D
CAD model and generates solidification images. It can be accessed through the following
link : http://efoundry.iitb.ac.in/

Page | 9
Chapter 4
RESEARCH PROBLEM DEFINITION
4.1 Motivation
Manufacturing of defect-free components at low cost and high productivity is important for the
casting industry today. The major challenges that the industry faces are large number of shop
floor trials, high rate of rejection and low casting yield. These can be overcome by adopting
solidification simulation technology. A typical foundry defect spectrum is shown below which
indicates that shrinkage holds a majority in defect related issues.
Solidification of the molten metal after being poured is an important phase in the casting
process which greatly affects the casting quality (produces shrinkage defects) and its yield. To
compensate for the shrinkage during the phase change, the required liquid is obtained from the
adjacent liquid regions. The last freezing regions are the most probable locations of shrinkage
cavities, which need feeders appended at suitable location on the casting. The total volume of
the feeder should be minimized to improve casting yield and productivity. The design and
optimization of the feeder requires intensive human interaction and numerous trial and error
iterations. The assistance of simulation tools for determining the optimal shapes, sizes and
locations of the feeders while compromising against the quality and cost constraints are
difficult to achieve.
Several optimization methods have been integrated into commercially available software like
AUTOCAST, CAST PRO, etc. which yield an easy design of the entire cast component
including feeders, gates, etc. The ready availability of computer technology, to automate the
casting design and optimization process makes both economic and engineering sense.
[Fig.4.1] – Foundry Defect Spectrum

Page | 10
4.2 Goal
Goal of the research is “design of feeder for a selected cross junction widely found in various
casting products which has the root cause of creating a shrinkage cavity and to validate this
design by experimentation.”
4.3 Research Objectives
• Study the solidification simulation of the junction
• Design of the feeder
• Re-simulate the junction along with its feeder
• Experimentation of the design by pouring molten metal
4.4 Research Approach
Research approach is divided into following steps to achieve the objective:
• Literature study on Solidification, Shrinkage and Feeder Design using Efoundry’s video
lectures by Dr.B.Ravi
• Selection of metal (Aluminium), moulding method (sand moulding) and junction
parameters and simulating it in Efoundry for locating the hotspot
• Feeder design using Caine’s method, locating it on the selected junction and simulating it
in Efoundry to check for ‘No hotspots at the junction’
• Design of pattern and making the appropriate mould cavity for metal pouring in a junction
with and without feeder
• Verifying the results by observing the cut-plane section at the hotspot region in both the
jobs
4.5 Scope
Scope for this project is being limited to design of feeder and validating the design by actual
pouring for Aluminium (ADC 12 grade) in Green Sand Casting. This project is focused on the
junction with single hot spot in the geometry and which has been addressed by a single feeder.

Page | 11
Chapter 5
SOLIDIFICATION AND SHRINKAGE
5.1 Mechanism of Solidification in Pure Metals
Liquids need to be cooled below their freezing points before the solidification begins. This is
because energy is required to create surfaces for new crystals. The degree of super cooling
necessary is reduced by the presence of other surfaces which serve as the initial nuclei for the
crystal growth. When a liquid metal is
poured into a mould, initially the temperature
everywhere is . The mould face itself acts
as the nucleus for crystal growth, and if the
conductivity of the mould is high, randomly-
oriented small crystals grow near the mould
face. Subsequently, a temperature gradient
results within the casting, as indicated in fig.
for and . As the solidification progresses
gradually inwards, long columnar crystals,
with their axes perpendicular to the mould
face, grow. This orientation of crystal growth
is desirable from the point of view of strength
of casting.
5.2 Shrinkage
The molten metal in the mold cavity occupies
considerably more volume than the solidified
castings that are eventually produced. This is for
the compensation of volumetric contraction which
metal exhibits. There are three quite different
contractions to be dealt with when cooling from
the liquid state to room temperature, as shown in
the fig. 4. They are:
5.2.1. Liquid Contraction: This contraction
occurs while metal is cooling in liquid state, since
liquid grows in density as it cools. This type of
contraction in the liquid state does not pose a significant problem because most of the
[Fig.5.1] – Development of columnar crystals
[Fig.5.2] – Solidification contraction regimes
in liquid, freezing and solid range
(adapted from Campbell, 2003)

Page | 12
superheat of a melt is usually lost during or quickly after pouring. It is compensated by
pouring more molten metal.
5.2.2. Solid Contraction: The solid contraction occurs after the casting has solidified and
as it cools from the solidification temperature to room temperature. To ensure that the
dimensions of the castings are correct, the pattern used to produce the given casting is
usually made slightly larger than the casting dimension.
5.2.3. Solidification Contraction: Contraction during solidification occurs at the freezing
point, since density of the solid is greater than density of liquid. This type of contraction is
the root cause of solidification related defects which in turn causes shrinkage porosity or
cavity. To compensate solidification contraction extra metal needs to be fed to the solidifying
casting. This extra metal is provided by separate reservoir of metal called as feeder, since
its action is to feed the metal to casting.
5.3 Solidification Simulation
The solidification process involves the transformation of the hot liquid metal to solid and
then subsequent cooling of the solid to the room temperature. Solidification of molten
metal after being poured into a mold cavity is an important phase in the casting process
which greatly affects the product quality and yield. During the past two decades, computer
modeling of solidification simulation has been widely used in foundry with an aim to:
 Predict the pattern of solidification, including shrinkage cavities and associated defect
predictions for various ferrous metals like steel, grey iron, ductile iron and nonferrous
metals like aluminium, copper, etc. as well as in precious metals like gold, silver,etc.
 Simulate solidification in various orientation of casting, with various metal-process
combinations, so that optimal position can be selected.
Such simulation can be obtained in Efoundry, developed by a team of faculty members under
Dr.B.Ravi’s supervision at IIT-Bombay. It gives an approximate visualization of the
solidification in a given casting. This can be used to know how efficient a designed feeder will
work so as to reduce the number of experimental trials.
Metal
Melting Point
(C)
Liquid density
(kg/m3
)
Solid density
(kg/m3
)
Shrinkage
(%)
Aluminium 660 2385 2700 7.1
Copper 1084 8000 8960 5.3
Cast Iron 1370 6900 7100 3.0
Cast Steel 1640 7015 7870 5.0
[Table 5.1] – Solidification shrinkage for major cast metals

Page | 13
Chapter 6
POURING AND FEEDING CASTINGS
6.1 Gating System
Gating system refers to all those elements, which are connected with the flow of molten metal
from the ladle to the mould cavity. The aim of the gating system is to provide a defect-free
casting. The various elements that are connected with a gating system are discussed as follows.
6.1.1 Elements of Gating system
1. Pouring Basin
The molten metal is not directly poured into the mould cavity because it may cause mould
erosion. The molten metal is poured into a pouring basin, which acts as a reservoir from which
it moves smoothly into the sprue. The main function of a pouring basin is to reduce the
momentum of the liquid flowing into the mould by settling first into it.
2. Sprue
Sprue is the channel through which the molten metal is brought into the parting plane where it
enters the runners and gates to ultimately reach the mould cavity. Sprue is tapered to gradually
reduce the cross section as it moves away from the top of the cope so that velocity of flow at
the bottom is increased.
3. Sprue Base Well
This is a reservoir for metal at the bottom of the sprue to reduce the momentum of the molten
metal. The molten metal as it moves down the sprue gains in velocity, some of which is lost in
the sprue base well by which the mould erosion is reduced. This molten metal then changes
direction and flows into the runners in a more uniform way.
[Fig.6.1] – Elements of Gating System
7. Riser

Page | 14
4. Runner
It is generally located in the horizontal plane (parting plane), which connects the sprue to its
in-gates, thus allowing the metal enter the mould cavity.
5. Runner Extension
The runner is extended a little further after it encounters the in-gate so as to trap the slag in the
molten metal.
6. Gate or In-gate
These are the openings through which the molten metal enters the mould cavity. The shape and
the cross section of the in-gate should be such that it can readily be broken off after casting
solidification and also that it allows the metal to enter quietly into the mould cavity. Depending
on the application, various types of gates used in the casting design are: Top Gate, Bottom
Gate, Parting Gate and Step Gate.
7. Riser
Most of the foundry alloys shrink during solidification. As a result of this volumetric shrinkage
during solidification, voids are likely to form in the casting unless additional molten metal is
fed into these places which are termed hot spots since they remain hot till end. Hence, a
reservoir of molten metal is to be maintained from which the metal can flow into the casting
when the need arises. These reservoirs are called risers.
6.1.2 Gating System Design
The liquid metal that runs through the various channels in the mould obeys the Bernoulli’s
theorem which states that the total energy head remains constant at any section. The same stated
in the equation form ignoring frictional losses is
2
where, = potential head, m; = pressure, Pa; = liquid velocity, m/s; = specific weight
of liquid, N/m3
; = gravitational acceleration
Though quantitatively Bernoulli’s theorem may not be applied, it helps to understand
qualitatively, the metal flow in the sand mould.
Another law of fluid mechanics, which is useful in understanding the gating behavior is the
law of continuity which says that the volume of metal flowing at any section in the mould is
constant. The same in the equation form can be

Page | 15
where, = rate of flow, m3
/s; = area of cross section, m2
; = velocity of metal flow, m/s
It is preferred that sprues are tapered in order to reduce the aspiration of air due to the increased
velocity as the metal flows down the sprue. This conclusion was drawn by applying the above
equation of continuity along with the Bernoulli’s equation.
The three parameters which are used in the gating design are:
1. Pouring Time
The time for complete filling of a mould termed as pouring time, is a very important criterion
for design. Too long a pouring time requires a higher pouring temperature and too less a
pouring time means turbulent flow in the mould which makes the casting defect prone. There
is thus an optimum pouring time for any given casting.
The pouring time depends on the casting materials, complexity of the casting, section thickness
and casting size. For nonferrous materials, a longer pouring time would be beneficial since they
lose heat slowly and also tend to form dross if metal is poured too quickly. Generally, a thumb
rule used for calculation is given below, though various empirical relations are available.
√
where, pouring time,s; W = weight of casting
2. Choke Area
After calculating the optimum pouring time, it is required to establish the main control area
which meters the metal flow into the mould cavity so that the mould is completely filled within
the calculated pouring time. This controlling area is called choke area.
The choke area can be calculated using Bernoulli’s equation as

2
where = choke area, mm3
; W = casting mass, kg; = pouring time,s; = mass density of
the molten metal, kg/mm3
; = effective metal head (sprue height), mm; = effeciency
factor which is a function of the gating system used
The effective sprue height , of a mould depends on the casting dimensions and type of gating
used.
3. Gating Ratios
The gating ratio refers to the proportion of the cross sectional area between the sprue, runner
and in-gates and is generally denoted as sprue area : runner area : in-gate area.

Page | 16
A typical gating ratio recommended or used in practice is 1:2:1. Besides, it is a general practice
to cut the runner in the cope and the in-gate in the drag to help in the trapping of slag. Moreover,
the in-gates are made wider compared to the depth, up to a ratio of 4. This facilitates in the
severing of gating from the casting after solidification. Small casting may be designed with a
single in-gate, however, large or complex casting require multiple in-gates to completely fill
all the sections of the castings effectively.
6.2 Feeders
In most cases, the terms risers and feeders are used interchangeably. Riser is something which
is open to the atmosphere and the metal can been seen rising in the mould cavity. Feeder on
the other hand has the job of feeding liquid metal to the hotspot. It can be blind or open.
Therefore, all risers are feeder but all feeders are not risers.
6.2.1 Principles of Feeding
The function of a feeder/riser is to feed the casting during solidification so that no shrinkage
cavities are formed. The requirement of the feeder depends to a great extent upon the type of
metal poured and the complexity of the casting. Various materials have different volumetric
shrinkages of which grey CI sometimes has a negative shrinkage and some metals such as
aluminium and steel have high volumetric contraction and hence, risering is required.
Shrinkage cavity development can be understood from the following example. Fig 6.2(a)
shows a cube which is completely filled with liquid metal. As time progresses, the metal starts
losing heat through all sides and as a result starts freezing from all sides, equally trapping the
liquid metal inside as shown in fig 6.2(b). Further solidification of the metal causes a
subsequent volumetric shrinkage which leads to metal concentration and thus, causes void
formation. The solidification when complete, finally results in the shrinkage cavity as shown
in fig 6.2(d).
[Fig.6.2] – Solidification of cube casting
(a) (b) (c) (d)

Page | 17
The reason for the formation of the void in the above cube casting is that the liquid metal in
the center which solidifies in the end is not fed during the solidification, hence the liquid
shrinkage occurred ends up as a void. Such isolated spots, which remain hot till the end are
called ‘hot spots’. A casting designer has to reduce all this hot spots so that no shrinkage
cavities occur.
In this connection, the term directional solidification is normally used in the casting
terminology. It means that the solidification of the metal should start at the remotest point of
the casting from the feeder. Since the cooling is achieved by the removal of heat from all
surfaces which are exposed to the atmosphere or sand, cooling normally starts from the point
which is thinnest or is exposed over a larger surface area.
6.2.2 Types of Feeder
A general classification of feeders is shown below:
An open feeder is exposed to the atmoshpere whereas a blind feeder is closed at its top. The
top feeders are placed above the hot spot, whereas the side feeders are placed at the side of
the hot spot, usually at the parting line. The various shapes used for such feeders are shown in
fig. 6.4.
[Fig.6.3] – Classification of Casting Feeders
[Fig.6.4] – Top & Side Feeder Shapes

Page | 18
6.2.3 Feeder Design
Solidification of the casting occurs by losing heat from the surface and the amount of the heat
is given by the volume of the casting. Hence, the cooling characteristics of a casting can be
represented by the surface-area-to-volume ratio. Since the riser is similar to the casting in its
solidification behavior, the riser characteristics can also be specified by the ratio of its surface
area to volume.
If this ratio of the casting is higher, then it is expected to cool faster. Chvorinov has shown that
the solidification time of a casting is proportional to the square of the volume-to-surface area
of the casting. The constant of proportionality called the Mould Constant depends on the
pouring temperature, casting and the mould thermal characteristics.
2
where = solidification time, s; = volume of casting; = surface area; = mould
constant; = modulus of casting
Some general rules have been laid out for feeder design on the basis of the above
characteristics. They are discussed as below:
1. The modulus criterion or heat transfer criterion is that the feeder must solidify at the
same time as, or later than the casting. This is satisfied by ensuring that the feeder has a modulus
(volume to surface area ratio) that is sufficiently larger than the casting by a multiplication
factor. The required modulus of the feeder is given by
∗
Multiplication factor for steel is 1.3, ductile iron is 1.15, grey iron is 1.0.
2. The casting can be separated into different feeding sections by dividing into simpler
shapes at different sections called feeding unit. A feeding unit is a group of casting sections in
which modulus decreases progressively from the highest modulus section to the lowest
modulus section. Each feeding unit is isolated from other feeding units by the low modulus
regions in between them.
3. The volume criterion states that the feeder must contain sufficient molten metal to meet
the volume contraction requirements of the casting. This is satisfied by ensuring that the feeder
has sufficient volume to feed all the shrinkage. The feeder volume should be at least equal to
the minimum volume given by
∗

Page | 19
where, feeding efficiency is volume fraction of the feeder that is actually available for
feeding and = shrinkage factor
4. The feed path criterion states that there should be positive feed paths to flow from the
liquid to all parts of the casting it is supposed to feed. In order to meet the feed path requirement,
the principle of directional solidification is followed. If the feeder can be placed on the highest
modulus section of the casting, with progressively thinner (lower modulus) sections extending
away, then the condition of progressive solidification towards the feeder can be met. The
number and position of feeders should be designed based on this criterion.
5. On the basis of the feeder location and type of connectivity, various parameters related
to feeder shape are considered. Taller feeders are used for steel castings (e.g. for cylindrical
feeders H/D = 2, where H and D are height and diameter of cylinder, respectively), which
exhibit shrinkage pipe, whereas in iron and aluminum castings, H/D value can be about 1.5.
6. Efficiency of feeder is characterized by modulus i.e. volume/heat transfer area. By
selecting different feeder shape we can have different efficiency. Casting yield is depending on
volume of feeder so it is necessary to reduce the volume of feeder. For small castings,
cylindrical feeders are widely used. For larger castings, cylindrical feeders with spherical
bottom (side location) or spherical top (top position, blind type) are widely used.
7. After determining the feeder dimension, shape and connection point, feeder neck is
assigned. Feeder neck is an important parameter, designed in a way to ensure decreasing
modulus towards the casting. This is done to ensure that the neck should solidify after the
casting hotspot and to maintain the flow of liquid metal from the feeder to casting hotspot.
∗
The multiplication factor is 1.2 to 1.5 depending on the cast metal.
8. The shape of the feeder-neck depends on the feeder shape, feeder position and the
connected portion of the casting. The most widely used neck shapes are cylindrical (for top
cylindrical feeders) and rectangular (mainly for side feeders). The neck may also be tapered
down towards the casting, thereby gradually reducing the modulus towards the casting.
9. Above all, casting yield should be maximized by using optimization techniques.
[Fig.6.5] – Progressive Directional Solidification

Page | 20
6.3 Casting yield
All the metals that is used while pouring is not finally ending up as a casting. On completion
of the casting process, the gating system used is removed from the solidified casting and
remitted to be used again as raw material. Hence, the casting yield is the actual volume of
casting required to the volume of metal poured into the mould cavity.
∗ 100
The higher the casting yield, the higher is the economics of the foundry practice. It is therefore
desirable to give consideration to the maximizing the casting yield, at the design stage itself.
Generally, those materials which shrink heavily have lower casting yield. Also, massive, and
simple shapes have higher casting yields compared to small and complex parts.

Page | 21
Chapter 7
FEEDER DESIGN SIMULATION AND
EXPERIMENTATION
7.1 Junction Definition
‘X’ junction is widely encountered in parts where 4 streams of metal get together. They are the
main regions where hotspot formation is always possible and hence leads to shrinkage cavity.
One such section of size 100 x 100 mm is selected. The part dimensions, isometric view,
orthographic view and physical properties of the job are shown below:
Physical Properties
Material ADC 12 (Al-Si : 83.4% - 11.2%)
Vcasting 233376 mm3
S.A.casting 29756 mm2
ρcasting 2823 kg/m3
mcasting 0.658 kg
Mcasting 7.842 mm
[Fig.7.1] – Selected ‘X’ Junction geometry
7.1(a) Orthographic view
7.1(b) Isometric view [Table 7.1] – Physical Properties of geometry

Page | 22
7.2 Pattern, Feeder and Mould Design
7.2.1 Pattern Allowances and Design
A wooden pattern for the selected junction is prepared. The various allowances so provided
on the pattern for Aluminium casting are as follows:
Allowance type Theoretical Allowance Actual Allowance Provided
Shrinkage 0.015 mm per mm 0.01 mm per mm
Draft ½-2 ⁰ ½-2 ⁰
7.2.2 Feeder Design Using Caine’s Method
Keeping in mind the various laid out rules for feeder design in the previous section and using
Caine’s method, the design of feeder is as follows:
The ‘freezing ratio’, X, of a mould is defined as the ratio of cooling characteristics of the casting
to the riser.

In order to be able to feed the casting, the riser should solidify last and hence its freezing ratio
should be greater than unity. It may be argued that the sphere has the lowest surface-area-to-
volume ratio and hence that it should be used as a riser. But in a sphere, the hottest metal being
at the centre, it is difficult to use it for feeding the casting. The next best is the cylindrical type
which is most commonly used for its ease in moulding.
Based on the Chvorinov’s rule, Caine developed a relationship empirically for the freezing
ratio as follows:

where

;
, and are constants whose values for Aluminium are 0.10, 0.06 and 1.08 respectively.
Design steps:
Volume of casting = 233376 mm3
Surface area of casting = 29756 mm2
[Table 7.2] – Pattern Allowances

Page | 23
Considering a cylindrical riser with H/D = 1,
Riser Volume =
Surface area of riser = 1.25
Freezing ratio,
/
1.25 2
/0. 3 0.0255

0.25
3
233376
1.0712 ∗ 10
3
Substituting the values of X & Y in Caine’s relation, we get
0.0255
0.1
1.0712 ∗ 10
3
0.06
1.08
By rearranging the terms and solving the above equation we get,
62.19166 62
Feeder neck dimensions due to geometry restrictions gives neck diameter, 25
The empirical relations for top riser neck dimensions are:
Length of neck, max /2 and 0.2 ∗
∴ 12.6
Modified surface area of casting due to neck placement is 29756 0.785 ∗ 25 ∗ 25
29265.375 mm2
.
New casting modulus is
.
7.97
Modulus of feeder,
∗ .
12.4
here, . Hence, design is safe.
Also, 0.785 ∗ 0.785 ∗ ∗ 193171.23 mm3
∴ Casting Yield, ∗ 100 54.70 %
Simulating the above results in Efoundry and optimizing the hotspots we have the final feeder
dimensions as:
64 ; 25 12.2
, 52.427%; 12.8 6.25
Although modulus of feeder neck is less than modulus of casting, its effective modulus is
always higher than the casting due to high heat transfer zone in the surrounding region of the
neck. Hence, to compensate for the required diameter according, 1.1 ∗ , a taper
is provided on the neck with the diameter at feeder connection equal to 30 mm.

Page | 24
7.2.3 Sprue, Runner and Gate Design
The route path which molten metal follows to enter into the mould cavity is pouring basin to
sprue to base well to runner to in-gates to mould cavity. The design of such path is as follows:
Weight of casting, = 0.658 *2 + 0.692 = 2.008 kg
Pouring time, √ = 1.417 s
taking, = 10 s (approximately)
Mass density of the molten metal, = 2439.8 kg/m3
Assuming top gating system with the entire sprue to be located in the cope itself, effective
metal head (sprue height), = 101.7 mm
Efficiency factor, = 0.73 (gating system with two runners)
∴ Choke area,
.
. ∗ ∗ . ∗ √ ∗ . ∗ .
= 79.8542 mm3
∴ Choke diameter, = 10.085 mm. let, 15 mm
Hence, choke area, = 176.625 mm2
.
Assuming the gating ratio of 1:2:1,
choke area = gate area
∴ gate area, ∗ (20 * thickness) which gives thickness of gate
10 mm. The gate is here located at the parting line because of 50 mm depth of the mould
cavity. so that metal enters the cavity at the mid plane thereby reducing erosion.
[Fig.7.2] – Geometry with top feeder
Top Feeder Φ64 x 64 mm
Feeder Neck
Φ25 x 12.2 mm

Page | 25
The area of runner obtained using the gating ratio accounts for a tapered runner section which
has been neglected here due to small job geometry and considering the pattern manufacturing
considerations.
Using Bernoulli’s equation and considering 10 mm height for the pouring basin from the top
of the cope,
Sprue area at top, ∗ 176.625 ∗
.
.
186.0064
∴ sprue top diameter, = 15.39 mm. let, 30 mm
Also, sprue base well diameter is 2.5 ∗ = 50 mm
The mould cavity so designed is shown in the figure below. The actual cope and drag patterns
are prepared from wood and those used for preparing the mould cavity are also shown below.
[Fig.7.3] – Mould Box Section
[Fig.7.4] – Drag Pattern
Locating pins
Junction
pattern
Runner

Page | 26
7.2.4 Moulding Sand
The moulding sand used for foundry practice is obtained commercially. The properties and
the composition of the sand used are:
Sr.No. Description Range
1. Grain Size AFS 70 to 90
2. Green Compressive Strength 700 to 1500 gm/cm
3. Permeability 90 to 120
4. Compactibility 35 to 48
5. Moisture 3.5-5%
6. VM 4.5-5.5%
7. Active Clay 8-10%
7.3 Simulation in Efoundry
Efoundry has inbuilt video classes which provides a good insight into casting design and
simulation technique. The video classes have been conducted by Dr. B Ravi (IIT-B). It also
includes an online library which contains technical papers and dissertations which have been
carried out in the same field. Certain ebooks are also available which are published by Dr.B
Ravi. It also holds an alloy database which gives the composition, properties, applications and
standards for ferrous and non-ferrous metals. Besides, it also provides online tutorial for
designing a casting component. A number of 3D models are also available in the library which
can be directly used or edited online in the CollabCAD software for dimensions. Online
simulation for solidification of casting for a given geometry can be done in the Sim Lab option.
A number of case studies are also uploaded to provide a better insight into the actual casting
industry problems.
[Table 7.3] – Moulding Sand Properties and Composition
[Fig.7.5] – Cope Pattern with riser and sprue
SprueTop
Feeder

Page | 27
The steps to carry out simulation in efoundry are as follows:
1. Login to efoundry website
2. Prepare the geometry file in .stl (stereolithography) file format
3. Select the metal and sand mesh parameters
4. Upload the model and interpret the simulation results.
7.6(a) – step 1
7.6(b) – step 2

Page | 28
The temperature contours obtained from efoundry for the selected ‘X’ junction are shown in
fig.7.6. These contours locate the hotspot region in the given geometry where white region is
7.6(c) – step 3
7.6(d) – step 4
[Fig 7.6] – Simulation steps in Efoundry

Page | 29
the hottest of all and indicates probable location of shrinkage cavity. The simulation is done
with the following parameters:
 Metal : Aluminium
 Sand Mesh : Coarse
Initially, the geometry was prepared in Creo 2.0 and was imported to ‘.stl’ (stereolithography)
format. The ‘.stl’ file was then imported to ‘Sim Lab’ in efoundry and then simulated.
Fig 7.6(a) – indicates the solidification zone in the junction without feeder. It is seen that at the
centre of the junction a local hotspot formation leads to unavailability of feed metal during
solidification causes and hence, causes shrinkage cavity of large volume.
[Fig.7.7] – Solidification simulation in Efoundry
7.7(a)
7.7(b); D = 62 7.7(c); D = 64
AMBIENT 582 ⁰C

Page | 30
Fig 7.6(b) – indicates the solidification zone in the junction with the feeder of diameter D = 62
mm. It is seen that the hotspot is not completely removed from the junction and has some
hotspot zone left inside the casting. Indicated by dark yellow regions.
Fig 7.6(c) – indicates the solidification zone in the junction with the feeder of diameter D = 64
mm. It is seen that the hotspot is completely removed from the junction and the hotspot has
completely shifted inside the feeder.
7.4 Experimentation in Foundry
The entire mould cavity was prepared in the foundry using the available green sand. Metal was
poured into the cavity and the casting was then analysed internally by cutting it diagonally to
observe internal defect (shrinkage cavity). Below figures show the various steps followed for
experimentation.
[Fig.7.8(a)] – Sand Preparation [Fig.7.8(b)] – Facing Sand application for Drag
[Fig.7.8(c)] – Rammed Drag [Fig.7.8(d)] – Mould cavity in Drag

Page | 31
The following pouring process was carried out at M/s. Utsav Metals, Nadiad. The parameters
noted during pouring condition are as follows:
Actual Pouring time 26 seconds
Actual Weight of Casting 2.150 kg
Al Alloy Composition (%)
Al Si Cu Zn Pb
83.41 11.24 2.88 0.99 0.18
[Fig.7.8(e)] – Cope preparation by placing feeder & sprue [Fig.7.8(f)] – Venting in the cope
[Fig.7.8(g)] – Final Drag & Cope Assembly [Fig.7.8(h)] – Metal Pouring
[Fig.7.8(i)] – Final Cast Product
[Table 7.4] – Pouring Condition Parameters and Alloy Composition

Page | 32
Chapter 8
EXPERIMENTAL RESULTS AND DISCUSSIONS
In order to validate the designed feeder for the job, two mould cavities are prepared pertaining
to the same jobs where one job is fed with open type top feeder whereas the other has no feeder.
The cavities are prepared inside a single mould box so that it ensures same pouring condition
for both the jobs and hence, provides a base for comparison. This is an effective means of
observing how feeder affects the quality of casting and helps in minimizing internal defects.
The runner and gate is separated from the two jobs and each of the job is cut diagonally across
the junction to observe the shrinkage cavity located inside it. The discussions related to each
of the job is as follows:
 Job without feeder
[Fig.8.1] – Cast job without feeder
[Fig.8.2] – Cut Plane Section in the job without feeder
Cutting Plane

Page | 33
A clean surface at the top of the junction indicates no riser/feeder location at its top. Due to the
breakage of mould cavity while preparing the cope, the cavity lost its corners on the top two
arms. This caused accumulation of the molten metal at those points. The cavity broke due to
the sticking of large amount of sand at the corners. One of the reasons for this is improper draft
available for pattern removal. High moisture content in the sand or excessive stickiness in the
moulding sand can also be its cause.
It is seen from the above cut-plane sections of the job that porous holes have been developed
at the center of the junction. Distributed shrinkage porosity is observed as in a case of long
freezing range alloy. This was predicted by the efoundry simulation as well which shows that
the location of hotspot lies at the center of the junction. Although, a bigger shrinkage cavity
was expected at the hotspot zone which was in reference to a short freezing range alloy i.e.
pure metal, the cavity obtained here is in the form of porous holes. The large amount of molten
metal being available at the edges of the junction allowed the junction to solidify progressively
thereby, reducing the amount of shrinkage. Large amount of shrinkage in terms of surface
defects is visible in the spread out metal area at the arms of the geometry.
Due to misalignment of the cope and drag after assembly, parting line shift is also observed.
The amount of parting line shift is about 3 to 5 mm. Parting line shift is never desired in the
casting.
This indicates clearly that if appropriate amount of metal is available for the casting during
solidification then shrinkage cavity can be eliminated.
 Job with feeder
[Fig.8.3] – Cast job with feeder

Page | 34
A circular spot on the top of the job indicates the feeder location on the job. Here also, the
breakage of cavity at the edges of the junction caused the metal to spread out of the job
randomly. This resulted in increase of the job size and hence the solidification as expected per
simulation is not obtained. A cut section of the top feeder is shown below. It is seen that there
is no cavity formed inside it.
Due to increase in the size of the job, the feeder would not have been able to feed the casting
and hence, it worked oppositely by feeding itself from the job. This may have been possible as
the shrinkage cavity pores have shifted from the parting line towards the top of the casting i.e.
towards the neck. If appropriate radiograph of the casting can be carried out, the actual
shrinkage zones could be observed accurately.
The shrinkage pores obtained here are least dense and smaller in size as compared to those in
the previous case by certain extent. There is a possibility that the neck may have solidified
earlier than the hotspot and hence, feeding path is blocked. Moreover, a sink is observed at the
top surface of the feeder which shows that some metal has been fed to accommodate the
[Fig.8.4] – Cut Plane Section in the job with open top feeder
[Fig.8.5] – Top Feeder and its Cut Plane Section

Page | 35
shrinkage. This dip is easily visible in the cut plane section of the feeder indicated by a
curvature at the top of the surface.
In general, it is expected that no shrinkage be formed inside this part, but due to various issues
related to sand preparation, mould making and pattern making makes it difficult to achieve for
inexperienced researchers. If the given job is casted out completely under the guidance of a
proper experienced foundryman, then a more in-depth analysis can be performed.
Moreover, theoretical casting yield,

* 100
∴
0.658 ∗ 2
2.008
∗ 100 65.5%
and actual casting yield from pouring ,

∴ ′
0.658 ∗ 2
2.150
∗ 100 61.2 %
It is also clear that actual casting yield is less than that of the theoretical casting yield since the
actual amount of metal poured is always more than that of theoretical value due to various
factors like mould cavity errors, oxidation loss through the sand mould, volumetric contraction
of the molten metal absorption of metal in the sand. Feeding of metal during solidification also
accounts to this factor.

Page | 36
Chapter 9
CONCLUSION
The selected ‘X’ junction is simulated in Efoundry to locate the hotspot formation. An
appropriate feeder is designed using Caine’s method and keeping in mind the rules laid out by
Campbell for feeder design. The feeder diameter so obtained is 62 mm. The feeder is then
added to the geometry and is simulated till the hotspot gets shifted to the feeder itself. The final
feeder diameter is 64 mm where the entire hotspot formation is in the feeder itself. The feeder
selected is an open type top feeder.
A mould cavity is prepared with two jobs of the ‘X’ junction where one job is assigned a feeder
and the other has no feeder. A single mould box ensures that the pouring condition remains
same for both the jobs and hence, allows them to be compared experimentally for shrinkage.
The shrinkage cavity in the junction without a feeder is obtained in the form of small amount
of porosity holes distributed across the parting line or the core of the junction. This confirms
to the hotspot location obtained in the Efoundry simulation. An actual type of cavity as expected
is not obtained due to the actual alloy composition whereas efoundry simulated for pure Al.
One of the other reasons is that while preparing the cope portion of the mould, the cavity as
desired was not obtained and it broke along the corners and edges of the arms of the job.
Moreover, parting line shift was also encountered. This caused the metal to spread out along
its edges. The extra metal so available aided the solidification feed paths of the job and hence,
reduced the shrinkage cavity size. This indicates that if appropriate amount of metal is available
during solidification, then feed metal paths exist and it tends to minimize the shrinkage cavity.
The shrinkage cavity in the junction with feeder is shifted above the parting line and it moved
towards the feeder neck. This indicates that feed paths were available from the top feeder.
Moreover, a spherical dip was observed in the top surface of the feeder which also indicates
that feeding had taken place. The only setback here was the parting line shift and the mould
cavity breakage in the cope portion along the edges and corners of the junction. This increased
the size of the job to be cast which may have increased its modulus than the feeder. This may
have caused a reversed feed path removing the cavity from the feeder. Again, a drop in the
modulus of the feeder neck due to such reasons caused the neck to solidify earlier which
blocked all feed path from the feeder to the job and vice versa.
Taking proper care while mould preparation can produce appropriate results as obtained from
the simulation.

Page | 37

FUTURE WORK

 Current work is limited to a junction with single hotspot which can be further expanded for
a junction with multiple hotspot.
 Optimization of the feeder can be done by employing the available numerical optimization
techniques and other casting simulation software. This ensure maximum yield.
 The desired junction can be tested for other materials and observed for the same defect. It
can be specially checked for steels or CI where graphitization causes negative shrinkage.
 Use of other methods like Vector Element Method, Modulus method, Naval Research
Laboratory method, etc. and others should be employed. The feeder dimensions so obtained
can be compared for best feeding efficiency and yield.
 A benchmark product can be taken as a case study where there are more than one hotspot
and the feeder dimensions obtained by different method can be tested and compared. The
main motto here should be to consider the economic factor associated with the cast product.
 Improved feeder design by incorporating insulated or exothermic feeder which gives high
feeding efficiency can also be validated.

Page | 38

REFERENCES
1. Amitabha Ghosh, Asok Kumar Mallik, Manufacturing Sciences, Affiliated East-West
Press Pvt. Ltd., 1981
2. Dr. P. C. Mukherjee, Methods of Improving Strength and Quality of Castings
3. O. P. Khanna, Foundry Technology, Dhanpat Rai Publications, 2011
4. P. N. Rao, Manufacturing Technology – Vol. 1 (Foundry, Forming and Welding),
McGraw Hill Education (India) Private Limited
5. Richard W. Heine, Carl R. Loper, Philip C. Rosenthal, Principles of Metal Casting,
Tata McGraw Hill Education Pvt. Ltd.
6. Elizabeth Jacob, Dundesh S. Chiniwar, Savithri S, Manoj M., and Roschen Sasikumar,
Simulation-Based Feeder Design for Metal Castings, Indian Foundry Journal, Vol.59,
No.12, December 2013, p.39-44
7. M. Jagdishwar, Casting Feeder Design Optimization Based on Feed Path and
Temperature Analysis, M.Tech Dissertation, IIT Bombay, 2012
8. D. Joshi, B. Ravi, Classification and Simulation Based Design of 3D Junctions in
Castings, AFS Transactions 2009
9. E-Foundry Academy, Casting Design and Simulation Video Lecture,
h // f d ii b i

simulation and experimentation in feeder design_secure

More Related Content

What's hot (20)

Viewers also liked (20)

Similar to simulation and experimentation in feeder design_secure (20)

simulation and experimentation in feeder design_secure