Research Article

Advances in Mechanical Engineering

Numerical simulation and optimization 12(5) 1–12

! The Author(s) 2020
DOI: 10.1177/1687814020923450
of casting process of copper alloy journals.sagepub.com/home/ade

water-meter shell

Kaikui Zheng1 , Youxi Lin2, Weiping Chen3 and Lei Liu2

Abstract
The water-meter shell has a complex-structured thin-walled cavity, and it can cause casting defects such as shrinkage and
misrun. On the basis of structural analysis of a water-meter shell, a three-dimensional model and a finite element model
of the water-meter shell were constructed using the SOLIDWORKS and ProCAST software as a modeling tool and a
casting numerical simulation tool, respectively. Three processes associated with the bottom gating system without a
riser, a step gating system with a preliminary riser, and a step gating system with an optimum riser were successively
numerically simulated. The mold-filling sequence, temperature distribution, liquid-phase distribution during solidification,
and shrinkage distribution of these three processes are discussed here. The numerical simulation results indicated that
optimization of the casting process and the rational assembling of the riser led to the shrinkage volumes at the inlet
position, regulating sleeve, and sealing ring of the water-meter shell decreasing from 0.68 to 0 cm3, 1.39 to 0.22 cm3, and
1.32 to 0.23 cm3, respectively. A comparison between model predictions and experimental measurements indicated that
the castings produced by the optimized process had good surface quality and beautiful appearance, without casting
defects, demonstrating that numerical simulation can be used as an effective tool for improving casting quality.

Keywords
Copper alloy, water-meter shell, casting process optimization, shrinkage defect, finite element analysis

Received 14 January 2020; accepted 9 April 2020

Handling Editor: James Baldwin

Introduction qualification rate of the water-meter shell has signifi-

cant economic benefits.4
With the growing shortage of global energy and water The trial-and-error method is mostly used in the tra-
resources, as well as the enforcement of legal metrolo- ditional casting process design, and the design is
gy, ensuring the measurement precision of water meters improved on the basis of the experience of engineers
has become imperative.1–3 The water-meter shell con- and workers. In particular, in the process of developing
sists of a cavity containing water and a gear for mea-
suring water flow, and high importance has been
1
attached to its quality and performance. However, in School of Advanced Manufacturing, Fuzhou University, Jinjiang, China
2
School of Mechanical Engineering and Automation, Fuzhou University,
actual production, due to its complex structure, the
Fuzhou, China
castings are prone to defects such as shrinkage and 3
Mechanical and Electrical Engineering Practice Center, Fuzhou
misrun. Furthermore, the water-meter shell should University, Fuzhou, China
withstand pressures of 1.5 and 2.0 MPa for intervals
Corresponding author:
exceeding 15 and 1 min, respectively. The current qual- Youxi Lin, School of Mechanical Engineering and Automation, Fuzhou
ified rate of water-meter components is not sufficiently University, Fuzhou 350116, China.
high, and in particular, the improvement of the Email: [email protected]
Creative Commons CC BY: This article is distributed under the terms of the Creative Commons Attribution 4.0 License (https://
creativecommons.org/licenses/by/4.0/) which permits any use, reproduction and distribution of the work without further permission
provided the original work is attributed as specified on the SAGE and Open Access pages (https://us.sagepub.com/en-us/nam/open-access-at-sage).
2 Advances in Mechanical Engineering

and designing large castings and new products, it is effectively eliminating the porosity and the shrinkage
necessary to repeatedly design the products and defects of the leaf spring bracket casting.
improve the design many times to meet the design In 1965, the finite difference method was used by
requirements. Apart from resulting in a long design general electric engineers to perform numerical
cycle and considerable material wastage, this process simulation studies on large thick-walled parts, and a
makes it difficult to guarantee the quality. Therefore, predictable solidification model was developed.16 The
copper alloy castings designed and manufactured by solid-phase method was first used by Henzel and
the traditional method are expensive. Keverian17 to predict the shrinkage defects of castings,
Since the 20th century, with the development of and it was subsequently used for the numerical simula-
computer technology, numerical calculation methods, tion of large steel castings; the steel castings produced
and heat transfer and related disciplines, developments on the basis of the simulation passed verification tests.
in casting numerical simulation technology have been In 1970s, the numerical simulation of the casting solid-
rapid, and the technology has been widely used. This
ification process was studied in depth. On the basis of
technology can not only shorten the casting process
such a numerical simulation, a new prediction criterionpffiffiffiffi
design cycle and product trial cycle in the design and
for shrinkage defects, involving the expression G= R
development of casting products but can also signifi-
(where G and R denote the temperature gradient and
cantly reduce energy consumption and material loss.
The mold-filling process involving metal liquids was cooling rate, respectively), was proposed by Niyama
first studied by Hwang and Stoehr5,6 using fluid et al.,18,19 and it has been widely used in practical engi-
simulation software. Chen et al.7 improved the two- neering. In the 1980s, research on the numerical simu-
dimensional (2D) message authentication codes algo- lation of the casting solidification process reached a
rithm to make it more versatile for studying liquid climax, and both theoretical and practical applications
metal flows with various symmetries and the processing of numerical simulations yielded improved results. The
of the intersection surface of the flows. Thomas and meshing of complex castings was studied by Ohnaka
Beckermann8 improved a method used for geometric and Li20 to achieve the meshing of castings of arbitrary
information analysis and the pressure iterative method, shapes, and the researchers numerically simulated cast-
thereby reducing the computation time of numerical ings with complex shapes in the study. According to the
simulations. Lost foam casting was used by Hirt and formation mechanism of shrinkage defects, various
Nichols9 for numerical simulation studies. These methods involving numerical simulation results of the
authors incorporated a shrinkage defect prediction casting solidification process have been proposed for
model in the commercial software FLOW-3D and suc- predicting shrinkage defects.21,22 Commonly used pre-
cessfully applied it to automotive castings. The gating diction methods include the direct simulation method,
system of a camshaft was studied by Jolly et al.10 using solid-phase line method, critical solid-phase rate
numerical simulation techniques, and they showed that method, temperature gradient method, and Johor
the porosity and the inclusion defects in the casting Bahru criterion method.23–29
were caused by the excessive turbulence of the molten With the increasing demand for improving the qual-
metal in the casting system. The pressure casting of ity and reducing the cost of products, the prediction of
thin-walled magnesium alloy parts was studied by Hu shrinkage defects in metal castings is of critical concern
et al.11 using casting numerical simulation technology, and a potential research area. In this study, in view of
and the casting system was optimized and designed.
the complexity of the copper alloy water-meter shell
Bansode et al.12 used Taguchi’s method to investigate
structure, the filling and solidification processes in the
the effect of critical process parameters of investment
fabrication of copper alloy water-meter shell castings
casting on dimensional variations of thin-walled, com-
were simulated using the numerical simulation software
plex geometrical stainless steel component. Prabhakar
et al.13 used the 2D computational fluid dynamics ProCAST for reducing the trial cost and design cycle
method to simulate the melt flow and solidification time in the traditional trial and error method.
stage in the lead sand casting process. A numerical Furthermore, the Niyama criterion was used for ana-
simulation approach based on the finite element casting lyzing the formation of shrinkage defects in the solidi-
software ProCAST has been developed by Sun et al.14 fication process. The use of casting numerical
Compared with insulation riser, shrinkage porosity simulation technology helped not only to overcome
proportion decreased from 23.10% to 17.01%, and the problem of not being able to observe changes in
the shrinkage cavity proportion decreased from the molten metal in the mold cavity during the tradi-
1.002% to 0.530%. Liu et al.15 used the finite element tional casting process but also to optimize the casting
model and ProCAST software to simulate and optimize process scheme, resulting in the fabrication of a higher-
the filling and solidification process of the casting, quality water-meter shell.
Zheng et al. 3

Development of casting model alloy water-meter shell and a sand core are shown in
Figures 2 and 3, respectively. The casting was simulat-
Numerical simulations of castings are generally per-
ed using a virtual mold in ProCAST. The character-
formed by assuming that the casting forming system
istics of the copper alloy water-meter shell casting were
shows geometrically limited dispersion. Changes in
the flow field, temperature field, and stress field as follows: the grid size of the water-meter shell was set
during the casting process were analyzed using numer- to 2 mm, the grid sizes of the casting system and sand
ical calculations, and the casting defect prediction core were set to 4 mm, and finer meshes were used in
method was combined with numerical simulations to some sensitive areas. These values facilitated an accu-
predict the casting defect distribution. The copper alloy rate numerical simulation of the copper alloy water-
water-meter shell studied in this work is a water-meter meter shell and simplified the calculations. The casting
component that protects the parts of the precision mea- is shown in Figure 4.
suring instrument inside it. A 2D sectional view of the
copper alloy water-meter shell is shown in Figure 1. Structural optimization and simulation
The water-meter shell is a thin-walled cavity with a
complex structure. The minimum and maximum wall
analysis
thicknesses are 2.3 and 8.75 mm, respectively. The cast- Bottom-injection-type no-riser design
ing material used in this study was ZCuZn40Pb2, and its
chemical composition is presented in Table 1. The The fabrication of a casting with a bottom-injection
liquidus and solidus temperatures of the casting mate- horn casting method without any riser was numerically
rial were 898 C and 887 C, respectively, and its specific simulated to determine the region where shrinkage
heat capacity was 0.377 J/(kgK); the quality was about
1.1 kg. Water-meter shell castings are required to have
clear outlines and keep the porosity, cold insulation,
misrun, and other defects at the lowest possible level.
The structure is required to be compact and to have a
certain strength and hardness so that it can withstand
pressures of 1.5 and 2.0 MPa over intervals of 15 and
1 min, respectively.
A complete water-meter model consists of a copper
alloy water-meter shell casting, a gating system, and a
sand core. For use in research, models of these water-
meter components are established by SOLIDWORKS
software. Three-dimensional (3D) models of a copper

Figure 2. Three-dimensional model of a copper alloy water-

meter shell.

Figure 1. Cross section of copper alloy water-meter shell: 1,

inlet; 2, outlet; 3, regulating sleeve; and 4, sealing ring. Figure 3. Three-dimensional model of a sand core.

Table 1. Chemical composition of casting materials.

Chemical composition Cu Fe Pb Sn Zn Impurity

Mass fraction (wt%) Balance 0.1 0.5–2 0.75 37 <1

4 Advances in Mechanical Engineering

defects were likely to be concentrated and the filling The mesh size of the casting was 2 mm, the total
capability of the casting system, and a preliminary number of grids was 520,296, the casting temperature
design of the gating system of the casting was devel- was 1000 C, the casting time was 3 s, and the initial
oped. Further, the mold-filling capability was improved mold temperature was 150 C. The temperature field of
and casting defects were reduced. The 3D model of the the filling process of the copper alloy water-meter shell,
bottom-injection-type no-riser design is shown in shown in Figure 6, was obtained from an analysis of
Figure 5. numerical simulation results. The temperature field of
the solidification process of the copper alloy water-
meter shell is shown in Figure 7, and the distribution
of shrinkage defects after solidification is shown in
Figure 8.
In Figure 6, in the bottom-injection-type no-riser
design, the molten metal rises smoothly during the fill-
ing process without splashing, and the gas in the cavity
can be smoothly discharged, so that the occurrence of
defects such as pores can be effectively avoided.
However, because of the chilling effect of the metal
mold, the thin-walled region of the casting easily sol-
idifies prematurely. This leads to the supplementary
molten metal being blocked, resulting in the formation
of misrun defects. Therefore, it is necessary to include a
Figure 4. Meshing of the casting in ProCAST.
supplementary runner in the upper part of the water
meter to compensate for the insufficient filling capabil-
ity of the bottom-injection-type no-riser design.
In Figure 7, the core cavity wall of the water-meter
shell begins to solidify first, for this wall is the thinnest.
As the solidification progresses, four isolated liquid
regions are gradually formed. These are the locations
where shrinkage defects of the water-meter shell are
concentrated. From an overall perspective of the
entire solidification process, the heat distribution of
the bottom-injection type is not uniform, which is det-
rimental to the sequential solidification of the casting.
An analysis of the casting temperature field revealed
the presence of a large number of hot spots at the
inlet and outlet of the casting, at the sealing ring, and
at the adjusting sleeve. It is evident from Figure 8 that
the largest shrinkage defect areas of the casting were at
the adjusting sleeve and the sealing ring, consistent
with the location of the isolated liquid region in the
Figure 5. Three-dimensional model of the bottom-injection-
liquid-phase distribution diagram. These results show
type no-riser design.
that these shrinkage defects were generated because of

Figure 6. Temperature field of the filling process for the bottom-injection-type no-riser design: (a) t ¼ 0.68 s, (b) t ¼ 1.55 s,
(c) t ¼ 2.73 s, and (d) t ¼ 3.00 s.
Zheng et al. 5

Figure 7. Temperature field of the solidification process for the bottom-injection-type no-riser design: (a) t ¼ 3.00 s, (b) t ¼ 3.77 s,
(c) t ¼ 4.88 s, (d) t ¼ 6.10 s, (e) t ¼ 7.59 s, and (f) t ¼ 11.84 s.

Figure 9. Three-dimensional model of the step-type trial-riser

design.

Step-type trial-riser design

On the basis of an analysis of the numerical simulation
results for the bottom-injection-type no-riser design,
Figure 8. Distribution of shrinkage defects in the bottom- the casting was supplemented with a riser, and the
injection-type no-riser design. step casting design was adopted to enhance the filling
capability. With these steps, misrun defects were effec-
tively avoided. The 3D model of the step-type trial-riser
the large wall thickness, which resulted in a long solid- design is shown in Figure 9.
ification time. By contrast, the wall in the adjacent part The mesh size of the casting was 2 mm, the total
was thinner, leading to premature solidification, and number of grids was 551,932, the casting temperature
the supplementary runner was broken to form an iso- was 1000 C, the casting time was 3 s, and the initial
lated liquid region. mold temperature was 150 C. The temperature field of
6 Advances in Mechanical Engineering

Figure 10. Temperature field distribution of the filling process for the step-type trial-riser design: (a) t ¼ 0.85 s, (b) t ¼ 2.14 s,
(c) t ¼ 2.52 s, and (d) t ¼ 3.00 s.

Figure 11. Temperature field of the solidification process for the step-type trial-riser design: (a) t ¼3.00 s, (b) t ¼ 4.11 s, (c) t ¼ 6.84
s, (d) t ¼ 7.38 s, (e) t ¼ 9.20 s, and (f) t ¼ 13.16 s.

the filling process of the copper alloy water-meter shell, upward and downward, which is favorable for the
shown in Figure 10, was obtained from an analysis of sequential solidification of the casting.
numerical simulation results; the temperature field of A comparison and an analysis of the temperature
the solidification process of the copper alloy water- fields of the solidification process, shown in Figures 7
meter shell is shown in Figure 11, and the distribution and 11, reveal that in the step casting design, the posi-
of shrinkage defects after solidification is shown in tion of the isolated liquid region of the casting
Figure 12. remained unchanged despite the addition of the supple-
It is clear from Figure 10 that the improvement of mentary runner; the addition of the riser significantly
the gating system and the addition of a supplementary enhanced the feeding effect of the isolated liquid
runner solved the problem of insufficient filling capa- region. Figure 12 shows that through the addition of
bility of the bottom-injection-type no-riser design. risers, shrinkage defects in the castings can be trans-
When the filling capability is affected by a temperature ferred, indicating that the choice of the riser position
decrease, the molten metal entering from the supple- strongly influences the feeding of the casting; the
mentary runner opened in the upper part can merge shrinkage defect appears to be significantly smaller,
with the molten metal entering from the bottom but is not completely transferred, which indicates that
runner. Consequently, a temperature gradient is the structure and size of the riser are inappropriate and
formed such that the temperature in the middle of the insufficient, respectively. Thus, the full feeding of the
casting is the lowest, and the temperature increases casting and the complete transfer of the shrinkage
Zheng et al. 7

defects were not achieved with the step-type trial-riser wedge-shaped wall was formed in the casting structure
design. to achieve a temperature gradient that gradually
increased toward the riser. The temperature gradient
Step-type improved-riser design considerably increased the effective feeding distance
of the riser.
On the basis of an analysis of the numerical simulation A thickness compensation design for the inlet and
results for the filling and solidification processes for the outlet and a regulating adjustment were used in this
step-type trial-riser design and due to the satisfactory study, and they are shown in Figure 13. The improved
effect observed for the step casting design, the casting riser, shown in Figure 13(a), had the following features:
method was not changed and the design of the riser was the compensation thickness at the inlet and the outlet
optimized. Since the inlet and outlet of the water-meter of the riser was 3 mm, the size of the annular riser was
shell and the adjusting sleeve had cylindrical structures, 33 mm, the diameter of the inner circle was 14 mm, and
the elimination of shrinkage defects could not be per- the height was 20 mm. Furthermore, the improved reg-
ceived only by increasing the riser diameter and height. ulating adjustment part, shown in Figure 13(b), had the
For the complete elimination of shrinkage defects, the following features: the compensation thickness at the
wall thickness was gradually increased toward the regulating adjustment of the riser was 1.5 mm, the riser
riser on the part of the casting near the riser, and a structure was changed to a trapezoidal step, the diam-
eter of the large end was 11 mm, the diameter of the
small end was 9 mm, the diameter of the inner circle
was 4.5 mm, and the height was 15 mm.
The mesh size of the numerical model was 2 mm, the
total number of meshes was 571,551, the filling time
was 3 s, and the solidification time was 22.76 s. The
temperature field of the solidification process of the
copper alloy water-meter shell with a step-type
improved-riser design was obtained from an analysis
of numerical simulation results, and it is shown in
Figure 14; the distribution of shrinkage defects after
solidification is shown in Figure 15.
It is evident from Figure 14 that the problem of the
isolated liquid phase at the sealing ring, inlet, outlet, and
regulating adjustment can be considerably improved
using an improved riser. Figure 14(a) shows the temper-
ature field of the casting at the instant filling was com-
pleted. The temperature of most parts of the casting is
near the liquidus, and the temperature of the thinner
cavity wall is below the liquidus. As shown in Figure
14(b), at 3.70 s, the cavity began solidifying, and the
solidification sequence started from the bottom and pro-
gressed upward. At this time, the temperature of all parts
except the sealing ring, inlet, outlet, and regulating
adjustment was below the liquidus. In Figure 14(c), at
Figure 12. Distribution of shrinkage defects in the step-type 6.09 s, the thinner cavity wall had solidified, and the
trial-riser design. unsolidified part of the casting was separated into three

Figure 13. Optimization of the riser: (a) inlet and outlet of the riser and (b) regulating adjustment part.
8 Advances in Mechanical Engineering

Figure 14. Temperature field of the solidification process for the step-type improved-riser design: (a) t ¼ 3.00 s, (b) t ¼ 3.70 s,
(c) t ¼ 6.09 s, (d) t ¼ 7.27 s, (e) t ¼ 9.63 s, and (f) t ¼ 11.63 s.

adjustment was divided into two parts (Figure 14(e)).

However, the size of the liquid-phase region in the regu-
lating adjustment was smaller than that in the case of the
trial riser. For the step-type improved-riser design, the
liquid phases at the inlet and outlet were disconnected
from the gating system, forming isolated liquid phases.
During the solidification process, the liquid phases at the
two locations slowly moved toward the risers and finally
disappeared at the risers.
It is clear from Figure 15 that with the improved
riser design, shrinkage defects were completely elimi-
nated at the inlet and the outlet of the casting and
were present in small numbers at the regulating adjust-
ment and sealing ring. The distribution of shrinkage
defects in different processes and different parts is
shown in Figure 16. The volume pairs of shrinkage
defects in different processes and different parts are
presented in Table 2.

Experiment
Casting process
Sand core production. Since the water-meter shell is thin-
Figure 15. Distribution of shrinkage defects in the step-type walled and encloses a cavity, it is necessary to fabricate a
improved-riser design. sand core. The sand core is used to form the inner cavity,
holes, and some specially shaped pits on the casting.
During casting, most of the sand core is covered by the
portions, namely, the sealing ring, adjusting sleeve, and molten metal and is subjected to heat transfer and influ-
inlet and outlet of the gating system. The solidification at enced by the flow of the high-temperature metal liquid.
the sealing ring was completed at 7.27 s (Figure 14(d)), Therefore, the quality of the sand core directly determines
and at 9.63 s, the isolated liquid region at the regulating the quality of the casting. The sand-core-making
Zheng et al. 9

Figure 16. Comparison of the distribution of shrinkage defects between different processes and different parts.

Table 2. Volume comparison of shrinkage defects between dif-

ferent processes and different parts.

Adjustment
Inlet (cm3) sleeve (cm3) Seal ring (cm3)

No riser 0.68 1.39 1.32

Trial riser 0.40 0.23 0.56
Improved riser 0 0.22 0.23

Figure 18. Sand core prepared in this study.

Figure 17. DL-400-B automatic double-head core shooting

machine.

equipment used in this experiment was the DL-400-B

automatic double-head core shooting machine, and it is
shown in Figure 17. The materials used for preparing the
sand core in this study and their amounts were as follows:
Figure 19. Smelting furnace DL-GYT-III.
silicon (50 kg), a curing agent (180 mL), a resin binder
(800 mL), and an additive (30 mL); the sand core pre-
pared is shown in Figure 18. between the metallic elements and its interaction with
gases, lining materials, and the like. Since copper alloy
Metal smelting. Metal smelting is a complex physico- is easily oxidized, it is the decisive factor to obtain
chemical process that involves the chemical interaction qualified metal liquid in strict accordance with the
10 Advances in Mechanical Engineering

process. The main steps in the smelting process are mold can be effectively extended. Moreover, graphite
copper ! charcoal ! melting ! lead þ zinc ! particles are very small, and therefore, the fluidity of
slagging-off ! heating and spraying fire ! addition the metal liquid can be increased to some extent, which
of P-Cu þ Al ! tapping and casting. The smelting fur- is convenient for filling the mold.
nace DL-GYT-III was used in this study, and it is
shown in Figure 19. Casting. The quality of the copper alloy water-meter
shell mostly depends on the quality of the casting. In
Coating on the mold. The metal mold is made of forged this study, the pouring temperature was 1000 C, the
red copper. The mold used for obtaining a casting of casting time was 3 s, and the initial mold temperature
the copper alloy water-meter housing requires a coating was 150 C; the mold was tilted by 60 during the cast-
on its surface for each casting. The coating method ing process and rotated at a constant speed so that the
used in this study was soaking the mold in graphite– gas in the casting could be effectively released and the
water mixture. Graphite is a heat-resistant substance flow rate of the molten metal could be lowered for fur-
that is suitable for separating the high-temperature ther improving the quality of the casting. The casting
metal liquid from the mold, so that the life of the equipment used in the experiment was the gravity cast-
ing machine SD-450-B, shown in Figure 20.

Post-treatment of casting. The casting of the copper alloy

water-meter shell was small. Therefore, after the cast-
ing was cooled, the casting system was removed by
manual tapping, and the riser was cut off by a sawing
machine to obtain the final casting for subsequent
machining.

Comparative analysis of casting quality

On the basis of the results of the optimization of struc-
tural and casting process parameters, the production of
the copper alloy water-meter shell before and after
optimization was studied in depth. A comparison of
the casting surface quality is shown in Figure 21, and
a cross-sectional comparison of the castings is shown in
Figure 20. Gravity casting machine.
Figure 22.

Figure 21. Comparison of surface quality of castings: (a) before optimization and (b) after optimization.
Zheng et al. 11

Figure 22. Comparison of surface quality of castings: (a) before optimization and (b) after optimization.

Conclusion
The copper alloy water-meter shell was selected as the
research object in this study. The casting process
parameters were numerically simulated and optimized
for eliminating defects such as shrinkage and misrun.
From numerical simulations, experimental compari-
son, and an analysis of the casting process, the follow-
ing conclusions were drawn:

1. On the basis of an analysis of the structure of the

water-meter shell, a preliminary design of the gating
system was developed. The step gating system was
adopted, and the ingates were set at the inlet and
outlet of the casting. A preliminary design of the
Figure 23. Longitudinal section of the casting. riser was made at the positions of the inlet, outlet,
sealing ring, and regulating sleeve of the water-meter
shell. Subsequently, the 3D model and finite element
model of the casting were developed.
In Figure 21(a), before optimization, there were 2. On the basis of a numerical simulation analysis of
many gas porosities and shrinkage defects at the the casting structure and the optimization of the
water outlet of the casting, and the surface quality at riser, the shrinkage volume at the position of the
the water outlet was poor. As Figure 21(b) shows, after inlet, regulating sleeve, and sealing ring of
optimization, the surface quality of the casting was the water-meter shell was reduced from 0.68 to
good, and there were no defects such as misrun, cold 0 cm3, 1.39 to 0.22 cm3, and 1.32 to 0.23 cm3, respec-
separation, or pores. tively, indicating significant reductions.
In Figure 22(a), the cross section of the water-meter 3. Experimental comparison between and an analysis
shell before optimization shows shrinkage defects and of the water-meter shell casting produced using the
gas porosities of different sizes at the inlet and the optimized process parameters and that produced
outlet of the casting and the adjusting sleeve, which using nonoptimized process parameters indicated
would seriously affect the quality and performance of that the optimized copper alloy water-meter shell
the casting. The cross section of the casting after the exhibited a beautiful appearance and had good sur-
optimization of the casting structure and casting pro- face quality, without macrodefects such as cracks,
cess parameters is shown in Figure 22(b). The section cold shuts, and misrun. The microstructure of the
quality of the casting is good, without shrinkage, cross section of the casting was compact and without
porosity, or any other defect. Furthermore, the opti- microdefects such as shrinkage or porosity. This
mized casting was cut longitudinally to examine a lon- result shows that the optimized process parameters
gitudinal section for defects; the longitudinal section is obtained in this study can effectively improve the
shown in Figure 23. There is no shrinkage cavity, quality of castings and increase productivity by min-
porosity, or any other defect in the longitudinal sec- imizing the number of rejections, thereby reducing
tion, and the quality is good. production cost.
12 Advances in Mechanical Engineering

Declaration of conflicting interests 12. Bansode SN, Phalle VM and Mantha SS. Optimization
of process parameters to improve dimensional accuracy
The author(s) declared no potential conflicts of interest with
of investment casting using Taguchi approach. Adv Mech
respect to the research, authorship, and/or publication of this Eng 2019; 11: 1–12.
article. 13. Prabhakar A, Papanikolaou M, Salonitis K, et al. Sand
casting of sheet lead: numerical simulation of metal flow
Funding and solidification. Int J Adv Manuf Technol 2019; 106:
The author(s) disclosed receipt of the following financial sup- 1–13.
14. Sun D, Lv R, Xie J, et al. Casting defect and
port for the research, authorship, and/or publication of this
process optimization of steel crossing. ISIJ Int 2020;
article: This project was supported by the National Natural
612: 1–8.
Science Foundation of China (Grant Nos 51875106 and 15. Liu J, Yang L, Fang X, et al. Numerical simulation and
51975123) and Fuzhou Science and Technology Plan optimization of shell mould casting process for leaf
Project (Grant No. 2019-G-42). spring bracket. Chin Foundry 2020; 17: 35–41.
16. Cao HJ, Song YP and Wang WY. Current application
ORCID iD and future development of computerized casting simula-
Kaikui Zheng https://orcid.org/0000-0002-8756-0267 tion. J Henan Univ Sci Technol 2006; 27: 5–8.
17. Henzel JG and Keverian J. The theory and application of
a digital computer in predicting solidification patterns.
References J Miner Met Mater Soc 1965; 17: 561–568.
1. Liu A and Mukheibir P. Digital metering feedback and 18. Niyama E. Computer-aided design for metal casting.
changes in water consumption—a review. Resour Conserv Annu Rev Mater Res 1990; 20: 101–115.
Recycl 2018; 134: 136–148. 19. Niyama E and Anzai K. Solidification velocity and tem-
2. Luciani C, Casellato F, Alvisi S, et al. Green smart tech- perature gradient in infinitely thick alloy castings. Mater
nology for water (GST4Water): water loss identification Trans 1995; 36: 61–64.
at user level by using smart metering systems. Water 20. Ohnaka I and Li XP. Research and development status
2019; 11: 405. of Japanese solidification simulation technology. Foundry
3. Vesal M, Rahmati MH and Hosseinabadi NT. The exter- 1991; 5: 43–45.
nality from communal metering of residential water: the 21. Carlson KD, Hardin RA, Ou S, et al. Development
case of Tehran. Water Resour Econ 2018; 23: 53–58. of new feeding-distance rules using casting simulation:
4. Beardsley JC. The use of water meters. J Am Water part I. Methodology. Metal Mater Trans B 2002; 33:
Works Assoc 2017; 109: 54–57. 731–740.
5. Hwang WS and Stoehr RA. Fluid flow modeling for 22. Stefanescu DM. Computer simulation of shrinkage relat-
computer-aided design of castings. JOM 1983; 35: 22–29. ed defects in metal castings–a review. Int J Cast Metal
6. Hwang WS and Stoehr R. Computer simulation for fill- Res 2005; 18: 129–143.
ing of castings. Trans Am Foundrymen Soc 1987; 95: 23. Li WZ. A study on numerical simulation of microstructure
425–430. and shrinkage cavity formation during solidification pro-
7. Chen S, Johnson DB and Raad PE. Velocity boundary cess of casting. Beijing, China: Tsinghua University, 1995.
conditions for the simulation of free surface fluid flow. J 24. Niyama E. A method of shrinkage prediction and its
Comput Phys 1995; 116: 262–276. application to steel casting practice. Imono 1982; 54:
8. Thomas BG and Beckermann C. Modeling of casting, 507–517.
welding, and advanced solidification processes VIII. In: 25. Li WZ, Liu BC and Wang CL. Research on predicting
Proceedings of the 8th international conference on model- shrinkage cavity an porosity of steel casting. Foundry
ing of casting and welding processes, San Diego, CA, 7–12 Technol 1995; 16: 37–46.
June 1998. Pittsburgh, PA: The Minerals, Metals & 26. Reis A, Xu ZA, Tol RV, et al. Model for prediction of
Materials Society. shrinkage defects in long and short freezing range mate-
9. Hirt CW and Nichols BD. Volume of fluid (VOF) rials. Int J Cast Metal Res 2007; 20: 171–175.
method for the dynamics of free boundaries. J Comput 27. Tavakoli R. On the prediction of shrinkage defects by
Phys 1981; 39: 201–225. thermal criterion functions. Int J Adv Manuf Technol
10. Jolly M, Cox M Gebelin JC, et al. Modeling the invest- 2014; 74: 569–579.
ment casting process: some preliminary results from the 28. Hajkowski J, Roquet P, Khamashta M, et al. Validation
UK research program “fundamentals of investment tests of prediction modules of shrinkage defects in cast
casting” (FOCAST). Transactions of the American iron sample. Arch Foundry Eng 2017; 17: 57–66.
Foundrymens Society 2001; 109: 919–935. 29. Jabbari M, Baran I and Mohanty S. Multiphysics model-
11. Hu BH, Tong KK, Niu XP, et al. Design and optimisa- ling of manufacturing processes: a review. Adv Mech Eng
tion of runner and gating systems for the die casting of 2018; 10: 1–31.
thin-walled magnesium telecommunication parts through
numerical simulation. J Mater Process Technol 2000; 105:
128–133.