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

net/publication/352761045

The effect of heat transfer coefficient and cooling schedule on porosity sizes
of low-pressure die casting aluminium alloy wheels

Conference Paper · June 2021

CITATIONS READS

0 95

4 authors, including:

Pawat Chunhachatrachai Worawut Janhom

National Taiwan University of Science and Technology Maxion Wheels
1 PUBLICATION   0 CITATIONS    1 PUBLICATION   0 CITATIONS   

SEE PROFILE SEE PROFILE

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

TSME ICOME 2019 View project

All content following this page was uploaded by Pawat Chunhachatrachai on 27 June 2021.

The user has requested enhancement of the downloaded file.

 The 10th TSME International Conference on Mechanical Engineering
10th – 13rd December 2019
Pattaya, Thailand

ICOME-2019-0050

The effect of heat transfer coefficient and cooling schedule on

porosity sizes of low-pressure die casting aluminium alloy
wheels

Pawat Chunhachatrachai 1,*, Worawut Janhom 2, Ninja Rakkitsiri 2, Kamthon Septham 1

1
Department of Mechanical Engineering, Faculty of Engineering, King Mongkut’s
University of Technology Thonburi, Bangkok, 10140, Thailand.
2
Maxion Technical Centre Asia Pacific (MTAP), Maxion Wheels (Thailand) Co., Ltd.,
Saraburi, 18140, Thailand.

* [email protected]

Abstract. Wheel is one of the most important parts of road vehicles sustaining all static
and dynamic loads. Thus, wheel manufacturing must conform to rigorous quality
requirements and safety standards. Shrinkage porosity plays a critical role in casting
quality of aluminum alloy wheels since it dramatically reduces the mechanical
properties of the wheels. The porosity formation results from the effect of solidification
shrinkage. It is certain that the heat transfer coefficient (HTC) and the cooling schedule
are two major parameters affecting the solidification of aluminum alloys. This paper
presents the effect of HTC and cooling schedule on porosity sizes of A356 aluminium
alloy wheels using the low-pressure die casting (LPDC) process. Numerical simulation
is employed using ProCAST, a finite element solver. The shrinkage porosity field is
employed to predict the porosity size and distribution. The HTC is increased at each
1,500 W/m2K interval from 12,500 to 21,500 W/m2K. The start time of the cooling
system is initially operated at 140 seconds after the mould is completely filled with
molten alloys, then it is gradually delayed up to 200 seconds. It shows that porosity sizes
become smaller when the start time of the cooling system is retarded. The variation of
HTC has more influence on porosity sizes for an earlier start time. Last, the start time
of the cooling system has more influence on porosity sizes of A356 aluminum alloy
wheels than the HTC.

Keywords: Shrinkage porosity/ Computational fluid dynamics/ Casting simulation/

Heat transfer coefficient/ Cooling schedule

1. Introduction
Wheel is considered one of the critical components for ground vehicles. Wheel failure could cause a
severe accident. Thus, extremely high quality and safety standards are needed. There are several factors
that affect the quality of wheels such as materials, design, and manufacturing. In the manufacturing
process, shrinkage pores dramatically weaken the mechanical properties of wheels since they can
significantly increase the local stress concentration [1]. The formation of porosity occurs during material
solidification. It is suggested that significant parameters that influence shrinkage porosity are wheel
geometry, pouring rate and temperature of molten alloy, ambient temperature, heat transfer coefficient
(HTC) between casting and mould, moulding material and finishes, and cooling strategies (e.g., cooling
schedule and cooling regions) [2].
According to the aforementioned parameters affecting shrinkage porosity, the heat transfer
coefficient (HTC) between casting and mould and the cooling schedule are focused since they can be
varied without mould modifications, which significantly increase the cost of production. Proper values
of HTC and cooling schedule not only minimise porosity sizes, but also might improve productivity as
the cycle time decreases. Moreover, both HTC and cooling schedule are closely connected with the well-
known Niyama criterion [3,4,5], which is widely used to predict locations and sizes of shrinkage
porosity.
The purpose of this study is to investigate the effect of the heat transfer coefficient between casting
and mould and the cooling schedule on porosity sizes of low-pressure die casting (LPDC) aluminium
alloy wheels. The Maxion 4883 wheel model is selected in this study since it is highly susceptible to
shrinkage porosity. Numerical simulation is employed using ProCAST, a finite element casting
simulation solver. The size and distribution of the porosity shrinkage are predicted based on the
shrinkage porosity field [6]. It is expected that this study will provide guidelines to reduce shrinkage
porosity defects in the commercial production of future wheel models.

2. Methodology

2.1 Low-pressure die casting and the details of the cooling system
In low-pressure die casting, wheel cast is enclosed by three main components: top die, side die, and
bottom die as shown in figure 1. The cooling system positions can be located on all three components
depending on mould designers. In general, the layout of cooling system and cooling schedule are
carefully planned and controlled in order that the solidification process is unidirectional. figure 2 shows
the details of the cooling system of the Maxion 4883 wheel model. The positions of the cooling system
are located along the direction of solidification, which initially begins at the rim (side die) and finally
ends at the wheel hub. In this wheel model, there are three main cooling

Figure 1. Main components of LPDC process [2] Figure 2. The positions of the cooling system

positions at A, B, and C, respectively. Positions A and B are cooled by air at both bottom die and top
die while position C is cooled by water at the bottom die. Water cooling is used at position C, the thickest
region (wheel hub), because water has greater value of thermal conductivity than air. In this study, only
the variation of HTC and cooling schedule of the cooling system at position C is focused since the
shrinkage porosity mostly appears around the hub region due to insufficient molten metal feeding during
the solidification process [7].
2.2 Numerical simulation

2.2.1 Governing equations

In ProCAST, a finite element casting simulation, it is assumed that molten metal is an incompressible
Newtonian fluid [6]. Thus, the governing equations are as follows:

(1)

(2)

(3)

, where 𝑉⃗ is the velocity vector, ρ is the density, Φ is the velocity component, t is the time, μ is the
dynamic viscosity, P is the pressure, g is the gravitational acceleration, T is the temperature, Cp is the
specific heat of molten metal, λ is the thermal conductivity, L is the latent heat, and fs is the solid phase
fraction at the solidification stage.
2.2.2 Meshing
In numerical simulation of low-pressure die casting, top die, side die, bottom die, cast wheel, and all the
cooling channels are included in the simulation domain. Mesh size is one of the significant parameters
related to accuracy of the simulation, especially for the prediction of locations and sizes of shrinkage
porosity. Table 1 presents the mesh size used for each component in this study.

Table 1. Mesh sizes of the simulation domain

Components Mesh size (mm.)

Cast wheel 1.5-2.3
Cooling channels 1.5-2.3
Dies (top, bottom, and side) 1.8-2.5

2.2.3 Shrinkage porosity model

According to ProCAST User Manual [6], shrinkage porosity field, a default shrinkage porosity model
in ProCAST, is employed to visualise the isosurfaces called “Shrinkage porosity”. Liquid pockets,
bounded by a mushy zone (two-phase region) and a solid cover, are formed during the solidification
process. Regions of low solid fraction appear. As solidification proceeds, a region can be split into many
regions. When a region is cooled down, solidification shrinkage occurs due to the increasing density
with reducing temperature. Porosities are then formed due to the effect of solidification shrinkage [8].

2.3 Design of experiment

The effect of HTC and cooling schedule on the size and distribution of the shrinkage porosity are
investigated following the procedure presented in figure 3. It is noted that, for boundary and initial
conditions defined in table 2, a number of production cycles are required for the temperature of dies
(top, bottom, and side) to reach steady-state conditions. The start time of the cooling system is initially
operated at 140 seconds after the mould is completely filled with molten alloys, then it is gradually
delayed up to 200 seconds. The HTC is increased at each 1,500 W/m2K interval from 12,500 to 21,500
W/m2K.
 Table 2. Boundary conditions and Initial conditions

Molten material temperature 700oC

Cooling schedule at position A 90 s
Cooling schedule at position B 110 s
Cooling schedule at side die 60 s
HTC at position A (Air) 800 W/m2K
HTC at position B (Air) 1400 W/m2K
HTC at side die (Water) 12000 W/m2K
HTC mould interface (Contacted) 0.5-1.5 kW/m2K
HTC mould interface (Bolt) 2-5 kW/m2K
Top die temperature@ 1st cycle 500oC
Bottom die temperature@ 1st cycle 550oC
Side die temperature@ 1st cycle 400oC

Figure 3. Experimental procedure

3. Results & Discussion

3.1 Solidification
There are two phases involved in the solidification process - solid and liquid. The solidification process
at t = 250 seconds after the mould is completely filled with molten metal is considered and compared.
Figures 4-6 present the solidification process for the start time of the cooling system at t = 140 seconds,
170 seconds, and 200 seconds, while the HTC is varied from 12,500 to 21,500 W/m2K. Every case is
not unidirectional solidification since the hub region turn into a solid phase earlier than the spoke region.
It is obvious that the fastest solidification process occurs at the case presented in figure 4(c) since the
cooling time is the longest and the value of HTC is the largest. Last, the variation of HTC has a slight
effect on the solidification time, while the start time of the cooling system is much more important.
 Figure 4. Fraction solid at 250 seconds for the start time at 140 seconds
(a) HTC = 12,500 (b) HTC = 17,000 (c) HTC = 21,500 W/m2K

Figure 5. Fraction solid at 250 seconds for the start time at 170 seconds
(a) HTC = 12,500 (b) HTC = 17,000 (c) HTC = 21,500 W/m2K

Figure 6. Fraction solid at 250 seconds for the start time at 200 seconds
(a) HTC = 12,500 (b) HTC = 17,000 (c) HTC = 21,500 W/m2K

3.2 Porosity size and distribution

The purple isosurfaces represent the porosity shrinkage captured using the shrinkage porosity field in
ProCAST. Figures 7-9 show that the porosity shrinkage is mainly found at the hub region for every case.
Moreover, porosities appear at approximately the same location even the HTC and cooling schedule are
varied. For the case of the start time at 140 seconds, the porosity size appears to be the largest. The
porosity size becomes smaller when the start time of the cooling system is delayed.

Figure 7. Porosity size and distribution at 250 seconds for the start time at 140 seconds
(a) HTC=12,500 (b) HTC=17,000 (c) HTC=21,500 W/m2K
 Figure 8. Porosity size and distribution at 250 seconds for the start time at 170 seconds
(a) HTC=12,500 (b) HTC=17,000 (c) HTC=21,500 W/m2K

Figure 9. Porosity size and distribution at 250 seconds for the start time at 200 seconds
(a) HTC=12,500 (b) HTC=17,000 (c) HTC=21,500 W/m2K

Further investigation is performed as shown in figure 10. Clearly, the total pore volume, the
summation of porosities, decreases when the start time of the cooling system is retarded. For the start
time at 140 and 150 seconds, the variation of HTC significantly affects the total pore volume. The lower
the HTC, the smaller the total pore volume. For the start time at 160 and 170 seconds, the variation of
HTC has less influence on the total pore volume. The total pore volume dramatically decreases when
the start time is delayed. For the start time at 180, 190, and 200 seconds, the total pore volume
monotonically decreases with the start time, and the variation of HTC has no influence on the total pore
volume. It is likely that the start time of the cooling system significantly affect the total pore volume.
Earlier solidification may prevent filling of molten material to substitute the shrinkage volume in the
hub region.

Cooling Schedule (s)

Figure 10. The effect of the HTC and cooling schedule on the total pore volume
 4. Conclusion
To sum up, the effect of the heat transfer coefficient between casting and mould and the cooling schedule
on porosity sizes of low-pressure die casting (LPDC) aluminium alloy wheels is investigated. Only the
variation of HTC and cooling schedule of the cooling system at position C is studied with the aid of
casting simulation software, ProCAST. The obtained results are summarized as follows:
(1) The total pore volume decreases when the start time of the cooling system is delayed. For
an earlier start time, the variation of HTC significantly affects the total pore volume. The lower
the HTC, the smaller the total pore volume. However, for a later start time, the total pore
volume monotonically decreases, and the variation of HTC has no influence on the total pore
volume. Thus, an earlier start time is not a proper choice due to larger total pore volume.
(2) The later the start time, the smaller the total pore volume. In order to obtain acceptable
total pore volume and minimise the cycle time, it is suggested that the start time between 170
and 180 seconds is the optimum operating point for this wheel model.

Acknowledgement
This research was financially supported by Maxion Wheels (Thailand) Co., Ltd. and King Mongkut’s
University of Technology Thonburi. ProCAST software from the ESI group is gratefully acknowledged.

References
[1] Teng, X., Mae, H., Bai, Y., and Wierzbicki, T. 2009, Pore size and fracture ductility of aluminum low
pressure die casting, Engineering Fracture Mechanics 76, 983–996
[2] Zhang, B., Cockcroft, S.L., Maijer, D.M., Zhu, J.D., and Phillion, A.B. 2005, Casting Defects in Low-
Pressure Die-Cast Aluminum Alloy Wheels, JOM 57 (11), 36-43.
[3] Niyama, E., Uchida, T., Morikawa, M., and Saito, S. 1982, AFS Cast Met. Res. J. 7, 52–63.
[4] Gao, W., Pan, L., and Zhen, L. 2011, Applicability of shrinkage porosity prediction for casting with Niyama
criterion, J. Beijing Univ. Aeronaut. Astronaut. 37, 1534–1540.
[5] Carlson, K.D., and Beckermann, C. 2009 Prediction of Shrinkage Pore Volume Fraction Using a
Dimensionless Niyama Criterion, Metallurgical and Materials Transaction A 40 (1), 163-175.
[6] ProCAST User’s Manual 2009, ESI Group, The virtual try-out space company.
[7] Sharma, P.K., Jain, A., and Bisht, P. 2016, Minimization of Defect in Aluminium Alloy Wheel Casting Using
7 QC Tools, International Journal of Scientific & Engineering Research 7 (3), 744-751.
[8] Wang, J., Fu, P., Liu, H., Li, D., and Li, Y. 2012, Shrinkage porosity criteria and optimized design of a 100-
ton 30Cr2Ni4MoV forging ingot, Materials & Design 35, 446–456.

View publication stats