See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/273261132

Estimation of shrinkage porosity of a cast aluminium alloy

Article · February 2012

DOI: 10.26634/jme.2.2.1567

CITATIONS READS
0 1,715

4 authors, including:

Samavedam Santhi Srinivasan Sundarrajan

Mahatma Gandhi Institute of Technology Sri Vishnu Educational Society
18 PUBLICATIONS   14 CITATIONS    54 PUBLICATIONS   349 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Flow and Fill characteristics of cast Al-Si alloys View project

Estimation of porosity and shrinkage of cast Al-Si alloys through casting process simulation View project

All content following this page was uploaded by Samavedam Santhi on 09 March 2015.

The user has requested enhancement of the downloaded file.

 RESEARCH PAPERS

ESTIMATION OF SHRINKAGE POROSITY OF A

CAST ALUMINIUM ALLOY
By
S.SANTHI * S. B. SAKRI ** D. HANUMANTHA RAO *** S.SUNDARRAJAN ****
* Assistant Professor, MGIT, Hyderabad.
** Technical Officer 'C', DRDL, Hyderabad.
*** Professor, M.V.S.R. Engineering College, Hyderabad.
**** Director, National Institute of Technology, Trichy.
ABSTRACT
Shrinkage characteristic of US A356 cast aluminium alloy has been discussed in the present study. The reduction in
specific volume leads to shrinkage in castings and it can be visualized as a casting defect. The shrinkage characteristic
has been studied using finite difference based casting process simulation software and it is quantified using
mathematical formulae. CAD has been used for construction of three dimensional model of the shrinkage defect.
Shrinkage characteristic has also been quantified through experimental validation studies and compared well with
casting process simulation. Shrinkage characteristic study and control is essential for producing defect free castings.
Influence of casting shape on the shrinkage characteristic has been studied in this paper.
Keywords: Shrinkage Porosity, Aluminium Alloy, Casting Shape, Process Simulation, Solid Model.

INTRODUCTION castings and use material data. In order to predict

The aluminum alloys (Al-alloys), are prone to defects, such
as shrinkage, one of the chronic problems, which impact
the quality of the castings. Due to the increasing use of
aluminum castings, shrinkage characteristics are useful for
improving mechanical properties of the castings [1]. A. Reis
etal stated [2] that the physics of formation of shrinkage
porosity has been involving interactions among many
physical phenomena such as heat transfer, fluid flow in the
liquid stage including natural convection, flow in the mushy
zone, solidification shrinkage, deformation of the solid skin
due to the formation of under pressure and many more.
Prediction of the behaviour of the liquid metal during
solidification has been important to consider the different
modes of shrinkage and trace the evolution of the liquid
metal free surface [3]. The quantity of shrinkage and
porosity has been depending on the thermal cooling
conditions, the alloy content and the casting temperature.
The casting process simulation is time and temperature
dependent. Solidification or freezing of castings is a non-
linear transient phenomenon and it involves modification
of phase with liberation of latent heat from a moving liquid-
solid boundary. The casting simulation programmes have
been predicting temperature distribution or hot spots in the
shrinkage and porosity the thermal properties are defined
as a function of the temperature [4, 5]. Shrinkage porosity
has been predicted by detecting open, partially closed
and closed liquid regions by utilising appropriate boundary
conditions. Casting process simulations have been carried
out using Virtual Casting software. It is a program for the
simulation of the solidification process of industrial castings
using FDM. This software has been developed by Regional
Research Laboratory, Trivandrum. This casting simulation
software has been installed at Institute of Indian
Foundrymen (R&D) Center, Hyderabad.
1. Experimental Plan
It has been observed from the literature that some of the
influencing parameters for shrinkage characteristics are
alloy composition, shape of casting, mould coat, bottom
chill, pouring time, mould sand and pouring temperature
[6].
For the present study casting shape, bottom chill, pouring
temperature and alloy composition were considered as
the process parameters.
Alloy composition affects the shrinkage characteristics,
mechanical properties and structure of the cast product.
The chemical composition of the alloy considered for the

18 i-manager’s Journal on Mechanical Engineering, Vol. 2 l

No. 2 l
February - April 2012
 RESEARCH PAPERS

present study has been given at Table 1. Exp run Alloy Chill Pouring Casting shape
order temperature (0C)
Three basic industrial casting shapes have been
1 USA356 Chill T+50 Cylinder
considered for the present study, namely rectangular, 2 USA356 Chill T+50 Cube
cube and cylinder. The rate of heat exchange depends on 3 USA356 Chill T+50 Rectangular
4 USA356 Chill T Cylinder
thermo-physical properties of liquid metal, wall thickness of 5 USA356 Chill T Cube
6 USA356 Chill T Rectangular
casting and its shape [7]. The dimensional details and 7 USA356 N T+50 Cylinder
pouring technique of the castings have been provided at 8 USA356 N T+50 Cube
9 USA356 N T+50 Rectangular
Table 2. 10 USA356 N T Cylinder
11 USA356 N T Cube
Pouring temperature [8] influences fluidity, porosity, strength 12 USA356 N T Rectangular
and structure of the casting. Hence pouring temperature Table 3. Shows the Details of the Number of Experiments
with 500C of super heat was considered as one of the
process parameter for the present study.
Another processing parameter is bottom chill which shows
significant difference in the casting characteristics,
promotes directional solidification, help in eliminating
shrinkage porosity defects. Mild steel bottom chill is
considered for the present study. The details of the number
of experiments conducted have been given at Table 3.
2. Casting Process Simulation
2.1 Solid Model
Simulation studies on shrinkage characteristics of
aluminium alloy have been conducted using the Virtual Figure 1. Solid Model/Test Piece for Cube Shape Casting
casting commercial software [9]. The solid models for
for casting solidification process [9], predicting and
shrinkage simulations were created using the solidworks.
analysing the occurrence of shrinkage defects. The input
The solid model or test piece for cube shape casting is
data for the casting process simulation is a solid model of
given in Figure 1.
the casting with methoding and material properties or
2.2 Simulation Studies thermal properties. Virtual Casting consists of three major
Virtual Casting is based on finite difference method. Virtual processes; Pre-processing, Solving the Governing
casting software has been creating a virtual environment Equations and Post-processing or Visualization of results.

Alloy Si Fe Cu Mn Mg Ni Zn Pb Sn Ti Each Total Al

USA356 6.5-7.5 0.55 0.2 0.35 0.2-0.65 0.15 0.15 0.15 0.05 0.05-0.25 0.05 0.15 Rem

Table 1. Chemical Composition (%wt)

S.No Shape Dimension, mm Pouring Technique Volume, cc

1 Rectangular 115x100x48 552

2 Cube 82x82x82 551368

3 Cylinder Ø90x90 572.265

Table 2. Dimensional Details of the Casting Shapes for Shrinkage Characteristic

i-manager’s Journal on Mechanical Engineering, Vol. 2 l

No. 2 l
February - April 2012 19
 RESEARCH PAPERS

The interfacial heat transfer coefficient is the rate of heat macro shrinkage and micro shrinkage has been
loss through the metal/mould interfaces which influence envisaged. The contour plots showing the location and
the mould filling and shrinkage or volume deficit magnitude of porosity formed during the solidification for
characteristics. However, interfacial heat transfer simulation run order number 10 is been shown below in
coefficient is not a simple material property and is Figure 2.
dependent upon chemical, physical interfacial 2.3 Construction of 3D Shrinkage Porosity
conditions, mould and casting material properties and
The present study shows the volume of the shrinkage
casting geometry. The selection of interfacial heat transfer
porosity of the casting at various locations, which is difficult
coefficient values as well as boundary conditions at the
to measure in 2D. Hence 3D solid works software has been
metal/mould interface affect the accuracy of the
used for generating 3D model of the images. The large
simulations. In the present investigations, interfacial heat
tolerance level typical of (0.1mm) foundry process has
transfer coefficient values for conformity of computer
been considered for the present study [14].
simulation and the experimental measurements are in
The contour plot data for simulation run order 10 at four
consistency. The thermo-physical properties of the US A 356
locations starting from 80mm from the flat bottom to
alloy, bottom chill and silica sand [10] for the casting
92mm have been converted to solid model part (sldpart).
simulation have been given at Table 4 [11,12,13].
Figure 3 shows constructed porosity of the simulation run
The solid model of shrinkage characteristic which is
order 18 using Solid Works software is given below.
imported in the stl file format as the solution domain,
Shrinkage porosity distributed for simulation run order 10
divides into small finite cells of casting and mould with a
from 80-92mm from the flat bottom is given at Table 5.
material id. Boundary Conditions are assigned at all
material interfaces like metal, mould, ambient and mould Porosity distribution in these four locations is 0.499 which is
coat. For shrinkage simulation the exit fraction of solid is 1. obtained from the contour plot for this particular simulation
The Exit solid fraction is the total volume percent of solid at
which the solver exits. For example the value of 1 for exit
solid fraction means that the simulation will be done till the
casting is 100% solidified.
Shrinkage occurs in metallic materials during freezing and
cooling due to reduction in specific volume, a physical
characteristic. The defects can be reduced by reducing
the shrinkage, controlling the distribution of shrinkage and
balance it using feeders and chills. The information
regarding the shrinkage and distribution is thus essential in
minimizing casting defects. Using virtual casting software
Figure 2. Typical Contour Plots Showing Porosity Distribution
S.No Parameter US A 356 Sand MS Chill
for Simulation Run Order 10
- - 84mm from bottom 88mm from bottom
1 Melting point, (°C) 660
2 Thermal conductivity, W/mm.K 0.159249 90.27x10-5 4.5x10-2
3 Density of solid (g/cc) 2.68496 1.5219 7.84
4 Liquidus temperature (°C) 613 - -
5 Freezing range (°C) 127.8 - -
6 Latent heat of fusion, J/Kg 388442 - -
7 Specific heat, J/Kg.K 962.944 1076.0076 460.548
8 Heat Transfer Coefficient, HTC, W/m2 K -
- -
Metal-mould 0.0035
Metal- coating mould 0.0015 - -
Volume is 15.49288 cc Volume is 10.8578 cc
Table 4. Thermophysical Data for Casting Process Simulation Figure 3. Shows Constructed Porosity of the Simulation Run Order 10

20 i-manager’s Journal on Mechanical Engineering, Vol. 2 l

No. 2 l
February - April 2012
 RESEARCH PAPERS

S.No Distance from the flat bottom, mm Volume cc taking the liquid metal from the interdendritic region. Once
1 80 15.49288 the dendrite mesh thickens, the interdendritic channels

2 84 10.63793
become narrow, and gradually more resistant to the flow of

3 88 10.8578
the residual liquid. This eventually leads to shrinkage pores
in the casting.
4 92 15.55

Total Volume 52.53861 Influence of Casting Shape on Shrinkage Porosity

Table 5. Shrinkage Porosity for Simulation Run Order10
run order 10. The amount of shrinkage porosity distribution in
these regions has been calculated and given at Table 6.
The porosity distribution for all the 12 simulation runs has
been calculated in the similar way. The results are
tabulated and given at Table 7.
2.4 Results and Discussion
US A356, long freezing range alloy liquid metal flow has
been ceased once the amount of fraction of solid reaches
a critical value. When the solute concentration is
increased, the solidification mode changes from growth of
columnar grains with more or less planar front to the
formation of equiaxed dendrites. Though a thin skin has
been initially formed on the mould walls, solidification has
not proceeded progressively inwards. Solidification
commenced by the advance from the mould walls
towards the interior with the liquidus isotherm. When there is
an insufficient amount of liquid metal, the dendrite mesh is
Total volume of shrinkage porosity 52.53861
Amount of shrinkage porosity Total volume x maximum porosity
distribution in these regions distribution in these regions
52.53861 x 0.499
26.21677
Amount of shrinkage porosity 26.21677/ volume of the casting
in the casting from table
26.21672/572.265
0.045812
The decrease in volume occurs from the pouring
temperature of liquid to ambient temperature is volume
deficit or shrinkage. Macro cavities decrease with casting
shape changes from thick to flat castings. This can be
explained based on the solidification behaviour of the
casting shape. The solidification morphology of the casting
in the localized areas has been influenced by the shape of
the casting because the edges play a considerable role in
the heat extraction. The edges of the test casting
combined with the relative composition of the alloy decide
the variation of formation of macrocavities. Relative
solidification time affects the macrocavities, with increase
in solidification time, the macro cavities increases.
Cylindrical shape castings are promoting maximum
cavities and it has been shown in the Figure 4.
3. Experimental Validation Studies
The experimental validation studies have been conducted
to validate the casting process simulation results. On the
basis of experimental validation studies it is possible to
compare and validate the simulation processes results with
reality and optimize. To evaluate the influence of process
parameters 3 (simulation run order 10, 11 and 3)
experiments have been conducted and schematic
diagram of the testing arrangement for cube shape
5
Cylinder
4.5 Cube
Rectangular

% Amount of shrinkage porosity in 4.5812 4

the casting for simulation run order 13
3.5
Table 6. Shrinkage Porosity Distribution in the Regions
3
% Shrinkage

for Simulation Run Order 10

2.5
Exp run order %Shrinkage porosity Exp run order %Shrinkage porosity
2
1 3.473 7 3.8
1.5
2 2.845 8 3.24
3 2.628 9 2.895 1

4 3.85 10 4.582 0.5

5 3.25 11 3.3 0
6 2.8 12 3.1 1 2 3 4
Figure 4. Influence of Casting Shape on Shrinkage
Table 7. Shrinkage Porosity for 12 Casting Process Simulations Porosity of USA356

i-manager’s Journal on Mechanical Engineering, Vol. 2 l

No. 2 l
February - April 2012 21
 RESEARCH PAPERS
casting as shown in Figure 5. These experiments have been
designated as exp run order 1 to 3. Prepared mould,
overflow core and pouring basin for rectangular, cube and
cylinder shape castings have been shown in Figure 6. The
overflow core is placed over the mould in order to ensure
that a fixed quantity of metal only is poured each time into
the mould.
3.1 Moulding, Melting and Pouring
Moulds are prepared using green sand process. The sand
composition consists of Bentonite 5-6% of sand weight and
water is 5-8% of sand weight. Moulds are prepared with
slight ramming. The patterns have been stripped after 3
hours. The alloy is melted in an electric resistance furnace
of capacity 20Kg provided with mild steel crucible.
Figure 5. Schematic Diagram for the Volume Deficit
Experiment for Cube Shape Casting
Figure 6. The Assembled Mould for Volume Deficit Experiment
22 i-manager’s Journal on Mechanical Engineering, Vol. 2 l
Temperature is measured with the help of a thermocouple.
The furnace is put off and the crucible is lifted and put in a
tilting device. The metal is tapped into a smaller crucible for
pouring into the mould. Figure 7 shows the rectangular,
cylinder and cube shaped solidified castings of the
experimental validation studies.
3.2 Shrinkage Porosity Calculations for the Experimental
Studies
Shrinkage porosity calculations for the experimental studies
have been given below
·The test casting is taken out of the mould and the cone
portion is cut off. The volume of the pipe has been
measured by keeping the casting under a burette and
distilled water with wetting agent is dropped into the
cavity till it is completely filled. The titration volume Vtitr is
read from the burette. The macrocavity, Vm is given by
Vm = Vcone + Vtitr
·The weight of casting in air and while immersed in water
are determined using a sensitive balance of accuracy
0.001gm
V = (weight in air – weight in water)
·The theoretical volume, V theoretical is obtained as
follows Vtheor = Weight in air / (Theoretical maximm
density where the theoretical maximum density is
obtained from chill specimen)
The internal porosity Vint is computed as follows
Vint = V – Vtheor
Shrinkage porosity is given by = (Vm+Vint)/Vmould
Shrinkage porosity for experimental run order1 is given
below at Table 8 using the above calculations.
The shrinkage porosity values for the remaining 2
experiments have been calculated in similar way using the
above calculations and are given at Table 9.
Rectangular Cylinder Cube
Figure 7. Solidified Castings of the Experimental Validation Studies
No. 2 l
February - April 2012
 RESEARCH PAPERS

S.No Parameter Exp run order 1 The shrinkage porosity values from the simulation results are
1. Gair(Wt),gm 1508 matching with the experimental studies data. Porosity is
2. Vtheor 562.6865672 major concern for foundry industry which is to be
3. V=(Gair-Gwater)cc 563 addressed to reduce the in-process rejections in castings.
4. Vtitr, cc 10 Present study mainly focuses on predicting the porosity in
5. Vcone,gm 17.5 aluminium alloy casting through process simulations which
6. Vcone,cc 6.529850746
will help the foundry industry in predicting and controlling
7. Vint (V- Vtheor)cc 0.313432836
the porosity. A novel technique for quantifying the porosity
8. Vtitr+Vcon+ Vint, cc 16.84328358
from casting process simulations is implemented with the
9. Vmould,cc 591.553
help of CAD application programme.
10. Shrinkage porosity 16.84328358/591.553= 0.028472992
0.028472992 Conclusion
11. % Shrinkage porosity 2.847
Shrinkage characteristic of the casting has been quantified
Table 8. Shrinkage Porosity for Experimental Run Order1
using finite difference based casting process simulation
Exp run order %Shrinkage porosity
software. Macro cavities decrease with casting shape
1 2.65
changes from thick to flat castings. Cylindrical shape
2 2.628
castings are promoting maximum shrinkage because the
Table 9. Shrinkage Porosity values for the Remaining 2 Experiments
edges play a considerable role in the heat extraction.
3.2 Critical Analysis of Simulation and Experimental Quantification of shrinkage porosity from casting process
Studies simulations is implemented with the help of CAD
The results from the simulation runs and experiments have application programme. The experimental validation
been compared. The simulation results are in agreement studies results are in agreement with casting process
with experimental studies data and it is shown in the Table simulation results.
10 & Figure 8. Acknowledgements
The authors reward their thanks to Directorate of
Cylinder USA356Exp run USA356Sim run
Engineering and Director DRDL for providing support and
Cylinder 2.84 3.5
permission for carrying out this R&D work
Cube 2.65 3.162
ectangular 2.628 2.895 References

Table 10. Critical Analysis of Simulation and Experimental Studies [1]. D.M. Stetanescu, (1992). ASM Metals Handbook
Volume 15, Casting, ASM International, The Materials
3.6
USA356Exprun Information Company.
USA356Simrun
3.4 [2]. A. Reis., Zhan xu., Rob Van Tol., A.D. Santos., A.
Barbedo Mayalhoos., (2005). ASM Metals Handbook
% shrinking porosity

3.2
Volume 15, Casting, ASM International, The Materials
3 Information Company Modeling of the Underpressure
Occurring During the Shrink Porosity Formation, Viii
2.8
International Conference On Computational Plasticity,
Complas Viii, E. Oñate And D. R. J. Owen (Eds), Cimne,
2.6
Barcelona.
2.4
[3]. Simulating Porosity in Ductile Iron Castings, Calcom SA,
Cylinder Cube Rectangular
Parc Scientifique EPFL, CH-1015 Lausanne, Switzerland,
Figure 8. Shrinkage Porosity Values for Simulation and phone: +41 21 693 2918 / fax: +41 21 693 4740,
Experimental Studies

i-manager’s Journal on Mechanical Engineering, Vol. 2 l

No. 2 l
February - April 2012 23
 RESEARCH PAPERS
mail@calcom.ch / www.calcom.ch
[4]. P. Scarber., Jr. H., Littleton., (2008). Simulation of Macro-
porosity in Aluminium lost foam castings, Paper 08-145, AFS
Transactions, pages 1061-1068.
[5]. Arno Louvo & Matti Sirrio (1994). Use of Simulated
Porosity for Avoidance of Casting Defects, VTT
Manufacturing Technology, Finland Presented at World
Foundry Conference in Düsseldorf, Germany.
[6]. S. Sundarrajan., H.Md. Roshan., E. G. Ramachandran.,
(1984). Studies on shrinkage characteristics of binary Mg-Al
alloys, Transactions of the Indian Institute of Metals, Vol.37,
No.4, August.
[7]. John R. Brown, (2008). Foseco Non-Ferrous Foundry
man's Handbook, Eleventh edition, Revised and
Butterworth Heinemann publisher.
[8]. A. Kielbus, (2007). The influence of casting temperature
on castability and structure of AJ62 alloy, Archives of
Materials Science and Engineering, Volume 28 Issue 6,
June , Pages 345-348.
[9]. Virtual casting user manual
[10]. A. Pani Kishore & T. S Prasanna Kumar, (2004). Fixing
Boundary Conditions For Solidification Simulation of
24 i-manager’s Journal on Mechanical Engineering, Vol. 2 l
View publication stats
Aluminium Alloy Plate Casting, International Symposium of
Research Students on Material Science and Engineering,
December 20-22, Chennai, India, Department of
Metallurgical and Materials Engineering, Indian Institute of
Technology Madras.
[11]. A Boundary Condition Coupling Strategy for the
Modeling of Metal Casting Processes, Michael Trovant,
National Library of Canada, Acquisitions and Bibliographic
Services, 395 Wellington Street, Ottawa ON K1AON4,
Canada, 1998.
[12]. M. Venkataramana., V. Vasudeva Rao., R. Ramgopal
Varma., & S. Sundarrajan, (2008). Instrumentation to
Measure Heat Transfer Coefficient at the Metal Mold
Interface, J. Instrum. Soc. India 37(3) 157-163.
[13]. G.S. Cellini., L. Tomesani., (2008). Metal head –
Dependent HTC in sand casting simulation of aluminium
alloys, Journal of Achievements in Materials and
Manufacturing Engineering, Volume 29 Issue 1 July.
[14]. Fiorello Losano., Gabriella Marinsek., Alberto Maria
Marlo., Miriam Ricci., (1999). Computer Tomography in the
Automotive Field Development of new engine Head case
study, DGZfP Proceedings BB 67-CD, Paper 10.
No. 2 l
February - April 2012