International Communications in Heat and Mass Transfer 61 (2015) 102–110

Contents lists available at ScienceDirect

International Communications in Heat and Mass Transfer

journal homepage: www.elsevier.com/locate/ichmt

Warpage control of thin-walled injection molding using local

mold temperatures☆
Shih-Chih Nian a, Chih-Yang Wu b, Ming-Shyan Huang b,⁎
a
Department of Power Mechanical Engineering, National Taitung Junior College, 889 Jhengci N. Rd., Taitung City 95045, Taiwan, ROC
b
Department of Mechanical and Automation Engineering, National Kaohsiung First University of Science and Technology, 2 Jhuoyue Road, Nanzih, Kaohsiung City 811, Taiwan, ROC

a r t i c l e i n f o a b s t r a c t

Available online 29 December 2014 Currently, 3C products are required to be lightweight, portable, and convenient. Injection molding is among the
most used techniques for mass production in plastic processing industries; however, producing thinner parts
Keywords: that do not warp is challenging. Although plastic components warp for numerous complicated reasons, warpage
Injection molding primarily is caused by variations in shrinkage during the injection process of plastic part manufacturing. Material
Neutral axis theory properties, part design, mold design, and processing conditions are factors influencing variations in the part shrink-
Thin-walled molding
age. For example, inconsistent thickness in component geometry, poor sprue–runner–gate or cooling design in the
Warpage
injection mold, and improper molding condition settings may cause plastic parts to warp excessively. Warpage
causes unpredictable component shapes, which may cause poor assembly quality. Although mold cooling achieved
by adjusting mold temperatures improves warpage, the conventional single mold temperature setting for each
male or female mold plate limits the cooling capability. Therefore, this paper describes local mold temperature set-
tings for a cooling system that can prevent severe warpage in an asymmetric plastic cover for handheld communi-
cation devices. The neutral axis theory is introduced to analyze the temperature distribution in the cross section of a
part, and then predict the warping trend. Through simulation and experiments conducted in this study, the feasi-
bility of using an effective local mold temperature setting in a cooling system to reduce part warpage was verified.
© 2014 Elsevier Ltd. All rights reserved.

1. Introduction The main cause of warpage in injection-molded parts is the uneven

Plastic parts that are thin (b1.5 mm) and feature a high ratio of flow
length to thickness (N 100) form quickly solid layers when molten poly-
mer enters mold cavities, thereby facilitating a short shot caused by the
sharp decrease in the flowing channel. Thus, a high injection speed is
required to complete filling and packing during the injection molding
process [1]. This high-speed injection molding requires using high injec-
tion pressure to force molten polymer into mold cavities and overcome
flowing resistance, and thus considerable changes inner pressures, par-
ticularly those of polymeric materials located near and far from the
gates. This phenomenon may induce uneven shrinkage of plastic parts
that warp easily after mold release [2–4]. Thin-walled plastic parts are
particularly prone to severe warping because of their weak mechanical
structures; nevertheless, the effects of improper molding condition
settings and uneven cooling result in sectional shrinkage variations.
Therefore, warpage control is crucial in manufacturing industries to pre-
vent quality problems during the successive assembly process, and
warpage must be minimized within dimensional tolerance.
☆ Communicated by W.J. Minkowycz.
⁎ Corresponding author.
E-mail address: mshuang@nkfust.edu.tw (M.-S. Huang).
volumetric shrinkage from high to low temperatures. The volumetric
shrinkage level of an injection-molded part can be described using the
pressure–volume–temperature (PVT) diagram depicted in Fig. 1. The
pattern from Point 1 to 2 represents the filling stage, in which molten
polymer enters mold cavities, and cavity pressure is gradually increased
corresponding to the degree of enforcing injection pressure. Point 2 is
the end of the volumetric mold filling, which is followed by packing;
the molten polymer pressure in cavities achieves maximum at Point 3.
The pattern from Point 3 to 4 is the transition of packing to static holding
that is frequently set at a relatively lower value than the injection pres-
sure; the cavity pressure is slightly reduced by the back flow of molten
polymer during the transition. The static holding stage (Point 4 to 5) is
performed at constant pressure, compensating for the specific volume
reduced by cooling. Notably, compensation achieved by the holding
pressure is effective only when molten polymer at the gates is not fro-
zen. The pattern from Point 5 to 7 represents the cooling stage, during
which the pressure decreases continually with the degree of volumetric
shrinkage during mold cooling; Point 5 to 6 and Point 6 to 7 indicate the
constant specific volume and constant pressure conditions, respectively.
At Point 7, the molded parts are ejected from mold cavities and then
cooled to room temperature (Point 8) under atmosphere pressure.
The travel distance between Point 6 and 8 determines the degree of vol-
umetric shrinkage in an injection-molded part.

http://dx.doi.org/10.1016/j.icheatmasstransfer.2014.12.008
0735-1933/© 2014 Elsevier Ltd. All rights reserved.
 S.-C. Nian et al. / International Communications in Heat and Mass Transfer 61 (2015) 102–110
Fig. 1. PVT diagram.
Uneven volumetric shrinkage can be induced by thermal effects,
pressure, part geometry, and flowing orientation effects. The details
are presented as follows.
1.1. Thermal effect
Temperature primarily causes the specific volume change in proc-
essed polymers. As depicted in the PVT diagram (Fig. 1), the specific
volume of injection-molded parts cooled from high temperatures and
surrounded by constant pressure change, causing a high shrinkage
rate. In thin-walled molding, molten polymer that enters cavities is
quickly cooled with energy dissipated from cavity surfaces. Conse-
quently, polymer temperatures close to the mold surface are lower
than those in the polymer's center, generating a relatively lower shrink-
age rate. Moreover, the thick portions shrink more than do the thin
Fig. 2. Geometry of the portable cover.
103
portions. Thus, injection-molded parts with uneven thicknesses under-
go nonuniform shrinkage and subsequent warpage.
In addition, differential cooling results in variations in sectional
shrinkage. The temperature difference between the upper and lower
surfaces causes differential shrinkage between the cavity and core, pro-
ducing a bending moment after the part is ejected from mold. This
bending moment creates warpage or residual stress, depending on the
mechanical stiffness of the part [5,6]. The cooling channel layout and
the core and cavity material properties also affect the cooling rate uni-
formity. Basically, deviation in the cooling rate causes uneven shrink-
age, particularly in thick parts that shrink substantially. In addition, a
hot mold surface shrinks more than does a cold surface. The thermal
effect contributes to generating internal stress in injection-molded
parts during the cooling stage; the outer layer along the thickness direc-
tion solidifies first and limits the shrinkage direction of the inner part
that cools later. Cooling channels are favorable to be placed close to
the part whenever possible. However, varying the distance between
the cooling channels and the part can facilitate controlling the differen-
tial cooling effects.
1.2. Pressure effect
Pressure is another crucial factor affecting the specific volume of
polymer. As shown in the PVT diagram, injection-molded parts cooled
under high pressure shrink less. The pressure level is associated with
the location of molten polymer in the cavities. For example, the polymer
near the gate is surrounded with high pressure, and therefore shrinks
less. By contrast, the polymer far from the gate and treated with low
pressure shrinks more. In thin-walled molding, molten polymer is
rapidly cooled, generating a substantial pressure gradient along the
thickness direction that produces nonuniform residual stress, and then
creates uneven shrinkage after cooling, thereby generating part warp-
age. The molten polymer can efficiently release internal stress at a
high temperature and shrinks less when sufficiently cooled.
Packing profiles can be used to establish an approximately uniform
distribution of volumetric shrinkage throughout a molded product. In
general, lower pressures cause volumetric shrinkage to increase, where-
as higher pressures reduce volumetric shrinkage. A constant packing
pressure results in volumetric shrinkage and is maximal at the end of
104 S.-C. Nian et al. / International Communications in Heat and Mass Transfer 61 (2015) 102–110

flow and minimal near the gate region. A decayed pressure profile gen-
erates an approximately uniform volumetric shrinkage by causing the
parts of plastic cooling near the gate to freeze as effectively as if they
were under the same pressure in regions further from the gate.

1.3. Geometry effect

Part geometry and mechanical properties of materials also play a

crucial role in warpage; the final warpage of a part depends on mechan-
ical stiffness, which is a function of the geometrical configuration and
the material's mechanical properties. If a part has greater mechanical
stiffness, it will warp less because of high variations in the shrinkage;
a part with less mechanical stiffness will warp more [6]. In addition,
the geometrical structure of parts may cause two flowing behaviors,
the hysteresis effect and the race-tracking effect, that substantially
affect the temperature and pressure distribution inside injection-
molded parts. These two effects are exerted when parts exhibit appar-
ent thickness deviation. During mold filling, molten polymer consistent-
ly flows to the low resistance area first and then to the high resistance Fig. 4. Melt-front patterns used during mold filling: (a) 85% filling time; (b) 99% filling
area. time.

1.4. Flow orientation effect

Shrinkage is also affected by the direction of material orientation.
During shear flow, polymer molecules align themselves with the flow
direction. The extent of orientation depends on the shear rate to
which the material is subjected. When the material ceases flowing,
the induced molecular orientation begins to relax at a rate that depends
on the material's relaxation time. If the material freezes before relaxa-
tion is complete, the molecular orientation is frozen. The shrinkage
amount varies in directions parallel and perpendicular to the direction
of material orientation.
Molding conditions are perhaps the easiest factors with which to
alter the orientation effects exerted on a part. By increasing melt tem-
perature, the viscosity of the material is reduced, as is the frozen layer
thickness, indicating that less molecular orientation is frozen in this
layer. The material remains hotter and longer, thereby enabling relaxa-
tion to occur; the molecular alignment is randomized, reducing the ori-
entation effect. Increasing the injection speed increases the shear
heating in the mold, and thus, the material is highly oriented. The addi-
tional shear heating reduces the viscosity of the material, which remains
hotter longer, allowing relaxation to occur and reducing the orientation.
In addition, reducing the thickness generally enhances orientation, but
reduces area shrinkage. Analysis has frequently indicated that the max-
imal orientation area is near the gate. Thus, the number and position of
gates should be determined to achieve both uniform and effective
shrinkage values at the end of flow paths.
direction. In particular, the flowing ability of molten polymer is sensitive
to part thickness reduction, and high resistance occurs when melting
plastics with decreasing part thickness. For example, the completeness
of thin-walled molding is affected by injection speed, injection pressure,
melt temperature, and mold temperature [6]. Among these, injection
speed and pressure are the major factors, followed by melt and mold
temperatures. A high injection speed and injection pressure facilitates
effective thin-walled molding without generating a detrimental short
shot; however, severe inner pressure deviation occurs along the flowing
direction, causing further warpage. Conventionally, increasing the gate
number is a quick solution used to reduce the injection pressure
required to complete mold filling, but this method has limitation.
Other approaches, such as high melt and mold temperature settings,
are effective, but prolong the cycle time [7–10]. Nevertheless, a low
mold temperature can easily cause hysteresis and further unbalanced
flowing behaviors, whereas a high mold temperature stabilizes the fill-
ing quality [7].
Nonuniform temperature and pressure distributions during in-
jection molding are known to cause local shrinkage, and associated
internal stresses can cause various local strengths, which produce
warpage [11]. Holding pressure can prevent shrinkage and warpage.
A high holding pressure and a high mold temperature has been
verified to eliminate shrinkage and warpage [12–13]. The Taguchi
method and the ANOVA analysis are frequently applied to optimize

2. Literature review

Compared with conventional injection molding, thin-walled injec-

tion molding forms a higher ratio of solid layers along the depth

Fig. 3. Sprue–runner–gate system in the injection mold. Fig. 5. Injection-molded portable cover: (a) front; (b) back.
 S.-C. Nian et al. / International Communications in Heat and Mass Transfer 61 (2015) 102–110
Fig. 6. Initially designed cooling channels with 8-mm diameter.
process parameters and improve shrinkage and warpage [14–17]. Indi-
vidually adjusting mold temperatures on the male and female mold
plates facilitates efficiently improving warpage and reducing the
cooling time [18–20]; they involve the dynamical control of mold and
cooling temperatures.
This study focused on adjusting local mold temperature settings to
eliminate the warpage in thin-walled molding. First, a commercial sim-
ulation software was utilized to predict the filling, cooling, shrinkage,
and warpage corresponding to initial design cooling channels. A neutral
axis theory was considered to analyze the warping direction based on
the information of temperature distribution along the part thickness,
enabling the local mold temperature settings to be identified. A new
design cooling system that facilitates local mold temperature setting is
then tested to verify its ability to reduce warpage.
3. Product geometry and warpage simulation using an initial cooling
channel design
Fig. 2 depicts the geometrical dimensions of the thin-walled injec-
tion molded part investigated in this study. The frame was 164-mm
long, 98-mm wide, and had a rectangular hole in the center (131-mm
long, 80-mm wide). The top and bottom edges were symmetrical; the
right edge was wider than the left, and both edges had different cross-
sectional structures. The thickness ranged from 1 mm to 3.5 mm
and the overall average thickness was 2 mm. Fig. 3 shows the sprue–
runner–gate system in the injection mold, in which the four fan-gates
Fig. 7. Warpage analysis of the initially designed cooling channels.
105
Fig. 8. Neutral axis on a bending structure.
with identical dimensions are positioned symmetrically to enable the
molten polymer to flow in a balanced manner during mold filling.
Consequently, both uniform and acceptable amounts of shrinkage at
the end of the flow paths can be achieved. Fig. 4 depicts the simulated
melt front used during mold filling, satisfying the traditional criteria of
filling all flow paths of the same distances in a given time; thus, four
pairs of melt fronts convene nearly simultaneously (i.e., 1% filling time
difference). The processed polymer was a PC and ABS blend polymer
(PC/ABS 385) made by the Chi-Mei Company (Taiwan). A high-speed
entirely electric injection molding machine with 100-tonnage maximal
clamping force (Robotshot i100β) made by the Fanuc Company (Japan)
was employed in the experiment. Fig. 5 shows the injection-molded
plastic part.
The initial cooling channel design for the thin-walled frame used in
portable 3C devices is shown in Fig. 6. The system includes three 8-
mm diameter cooling channels that surround the cavity and core at dis-
tances of 26 mm and 16 mm, respectively. The cooling layouts designed
for female and male mold plates differed; a pair of wrapped channels
was placed on the four edges of the female mold plate, whereas one
wrapped channel was used in the male mold plate. To predict the defor-
mation of the thin-walled part produced by injection molding, the
Moldex3D commercial simulation software was employed to analyze
the warpage. When the temperature setting for the entire mold was
70 °C, the simulation results (Fig. 7) indicated that the left and right
edges severely warped upward, whereas the top and bottom edges
warped downward, a phenomenon that is consistent with actual mold-
ing conditions.
4. Neutral axis theory
To control the warpage defects frequently occurring in thin-walled
molding, the neutral axis theory was considered to determine possible
warping direction. According to literature, the neutral axis is defined
106 S.-C. Nian et al. / International Communications in Heat and Mass Transfer 61 (2015) 102–110

Fig. 9. Cross-sectional temperature profile of initially designed cooling channels: (a) partial C-C section; (b) A-A section.

as an axis in the cross section of a beam or shaft along which there are
no longitudinal stresses or strains (Fig. 8). If the section is symmetric,
isotropic, and not curved before a bend occurs, then the neutral
axis is at the geometric centroid. If all polymers on one side of the
neutral axis are in a state of tension, those on the opposite side are
in compression. Thereby, there is a compressive strain at the top of
the beam, and a tensile strain at the bottom of the beam. From statics,
a moment consists of equal and opposite forces; therefore the total
amount of force across the cross section must be zero [21].
Basically, the approach used in this study analyzed the cross-
sectional temperature distribution along the part thickness immediate-
ly after the filling process (i.e., the molten polymer within cavities was
under zero pressure). Without pressure affecting the temperature
distribution, the effects temperature exerted on warpage were able to
be observed. Molten polymer at a high temperature shrinks substan-
tially after cooling to room temperature. The cross-sectional tempera-
ture distribution and the location of its neutral axis can indicate the
Fig. 10. Newly designed cooling channels with 8-mm diameter.
warping trend. For example, Fig. 9 illustrates the cross-sectional tem-
perature profiles associated with long and short edges. According to

Fig. 11. Warpage of the long edge w.r.t. with various mold temperature settings.
 S.-C. Nian et al. / International Communications in Heat and Mass Transfer 61 (2015) 102–110 107

the temperature profile in the C-C section of the short edge (Fig. 2), the
maximal temperature was located above the neutral axis and caused
relatively substantial shrinkage during cooling, suggesting that the
short edge will warp, forming a concave shape after cooling. By contrast,
the maximal temperature is located below the neutral axis in the A-A
section of the long edge (Fig. 2), which will warp into a convex shape.
Therefore, the neutral axis theory can be considered to explain the
warpage behavior depicted in the simulation (Fig. 7). Moreover, if the
temperatures of female and male mold plates are adjusted to enable
the maximal temperature to be located exactly on the neutral axis, the
warpage caused by uneven shrinkage can be eliminated.

5. Warpage simulation and experiment with local mold temperature

setting

Compared with the initial design cooling system (Fig. 6) that is lim-
ited when improving the warpage of thin-walled parts, a new design
cooling system that enables establishing local mold temperatures for
the four edges is proposed in this paper (Fig. 10). Fig. 11 shows the
comparison of z-directional warpage in the long edge with respect
to a uniform mold temperature setting (70 °C mold temperature)
and a local mold temperature setting (80 °C and 30 °C mold tempera-
tures for female and male mold plates, respectively). The female plate
was deformed in the range of −0.88–1.0 mm, whereas the male plate
was less deformed in the range of − 0.63–0.7 mm. Fig. 12 depicts the
cross-sectional temperature profiles of the long edges according to
various mold temperature setting. The simulation determined that a
local mold temperature setting at the long edge enabled the maximal
temperature to be near the neutral axis. Consequently, the simulation Fig. 13. Sensing positions for the z-directional position of the portable cover (back side).
verified the feasibility of using local mold temperature settings to elim-
inate the warping of thin-walled molding based on the neutral axis
theory. balanced. The height of each sample was measured and then compared
To further verify the effectiveness of using local mold temperature to with its standard value. Fig. 13 depicts the sensing positions for the
reduce the warpage of long edges, relevant experiments were conduct- z-directional height on the back side of the portable cover. Fig. 14
ed in this study. shows curvature profiles of the four cover edges that correspond to var-
ious mold temperatures, and Table 1 lists the warpage amount according
5.1. Single mold temperature setting to various mold temperature settings, including range, straightness,
and flatness analyses. Because the measurement was conducted by
In the single mold temperature experiment, the mold temperature probing the back face of the part, the long convex edges on the outer
was set at 50, 60, and 70 °C, and some samples were acquired after 40 surface became concave in shape on the inner surface. Similarly, the
cycles of injection molding ensured that data collected would be short edges that were concave on the outer surface were convex on

Fig. 12. Cross-sectional temperature profiles of the long edge w.r.t. with various mold temperature setting.
108 S.-C. Nian et al. / International Communications in Heat and Mass Transfer 61 (2015) 102–110

4.60
a respectively, and the warpage was minimal among these four levels of
4.50
single mold temperature settings.
4.40

4.30 5.2. Local mold temperature setting

4.20
Z (mm)

4.10 According to simulation analysis, the long edges warped upward

4.00
and the short edges warped downward. According to the neutral axis
theory, the local mold temperature of the female mold plate at the
3.90
long and short edges was higher and lower, respectively, whereas the
3.80
local temperature of the male mold plate at the long and short edges
3.70 was lower and higher, respectively. During the local mold temperature
3.60 experiment, the mold temperature was set to approximately 40 °C to
3.50 52 °C (Table 2), samples were acquired after 40 cycles of injection
P2 P4 P6 P8 P10 P12 P14 P16 P18 P20 P22 P24 P26 P28 molding, and each sample was measured.
Position Fig. 15 depicts the cooling channel layout that enabled setting the
4.60
local mold temperature during injection molding of a portable device
4.50
b cover. To reduce the warpage in the two long edges, the A1 and A3
cooling channels in the female mold plate were given high tempera-
4.40
tures, and the corresponding B1 and B3 channels in the male mold
4.30 plate were given low temperatures. By contrast, to reduce the warping
4.20 of the two short edges, the A2 and A4 channels were given low temper-
Z (mm)

4.10 atures in the female mold plate, and the corresponding B2 and B4 chan-
4.00
nels were given high temperatures in the male mold plate. The local
mold temperature setting was constrained by the capacity of available
3.90
mold heating devices in this experiment. Table 2 shows the actual
3.80
local mold temperatures; A1 and A3 (49 °C–52 °C) were hotter than
3.70 B1 and B3, and B2 and B4 (40 °C–42 °C) were hotter than A2 and A4.
3.60 To ensure the mold temperature control in the eight mold regions was
3.50 stable, eight temperature sensors were installed in each region
P2 P4 P6 P8 P10 P12 P14 P16 P18 P20 P22 P24 P26 P28 (Fig. 16). Fig. 17 shows the record of cycle-by-cycled mold tempera-
Position tures, which were controlled steadily. Fig. 18 shows the deformation
profile for the thin-walled injection molding that fabricated the portable
4.60

4.50
c device cover using local mold temperature setting. The greatest defor-
mation was located at P12 of the short edge (3.85 mm). Table 3 lists
4.40 the straightness of the four edges: 0.11 mm and 0.06 mm on the right
4.30 and left edges, respectively, and 0.19 mm and 0.23 mm on the top
4.20 and bottom edges, respectively. The flatness and range were merely
Z (mm)

4.10 0.31 mm. Depicting a comparison of the optimal warping control

4.00
using local mold and conventional single mold temperature settings,
Fig. 19 indicates that the deformation of the cover fabricated using
3.90
local mold temperature settings was superior to that produced by a sin-
3.80
gle mold temperature setting of 50 °C, in which the flatness are 0.31 mm
3.70 and 0.65 mm, respectively. Experimental and simulation results both
3.60 verified the feasibility of applying the neutral axis theory and local
3.50 mold temperature settings to effectively control the warping of a thin-
P2 P4 P6 P8 P10 P12 P14 P16 P18 P20 P22 P24 P26 P28 walled part.
Position
6. Conclusions
Fig. 14. Z-directional deformation profile of the molded part fabricated with conventional
single mold temperature settings: (a) 50 °C; (b) 60 °C; (c) 70 °C.
This study included a neutral axis theory to estimate the warping
direction of thin-walled parts, which severely warp because of high
the inner surface. Fig. 14 depicts the part profiles and Table 1 indicates injection velocity and pressure used during fabrication, improper
that at a 50 °C mold temperature, the straightness of the right and left design of sprue–runner–gate and cooling channels, and weak me-
edges were 0.39 mm and 0.30 mm, respectively, and those of the top chanical structure. A cooling system that enables establishing indi-
and bottom edges were 0.31 mm and 0.13 mm, respectively. The vidual mold temperatures was designed. The simulation and
range and flatness of the four edges were 0.71 mm and 0.65 mm, experimental results demonstrated that the straightness of the
warpage area in parts fabricated using local mold temperature
Table 1
Warpage level w.r.t. using various single mold temperature settings.
Table 2
Mold Straightness (mm) Range Flatness Setting of local mold temperatures.
temp. (mm) (mm)
Right Left Top Bottom Area Temperature (°C) Area Temperature (°C) Difference (°C)
edge edge edge edge
A1 87 B1 38 49
50 °C 0.39 0.30 0.31 0.13 0.71 0.65 A2 38 B2 78 40
60 °C 0.55 0.33 0.35 0.17 0.78 0.72 A3 92 B3 40 52
70 °C 0.53 0.42 0.33 0.28 0.91 0.87 A4 36 B4 78 42
 S.-C. Nian et al. / International Communications in Heat and Mass Transfer 61 (2015) 102–110 109

Fig. 15. Layout of local mold cooling channels: (a) female mold plate; (b) male mold plate.

Fig. 16. Mold temperature sensing positions: (a) female mold plate; (b) male mold plate.

settings can be improved substantially compared with when using
conventional single mold temperature settings. The following con-
clusions were made:
(1) The study employed Moldex3D to predict the thermal effect
cooling channels exerted on the part warpage. The model
yielded a temperature profile along the part thickness
100
direction and a warpage profile according to various mold
temperature settings.
(2) A computer simulation and experiment was performed to
predict the warpage of the injection-molded portable cover,
fabricated using mold temperature settings of 50, 60, and 70 °C.
The results indicated that the left and right thin edges warped se-
verely upward, whereas the top and bottom thick edges warped
downward, results that were consistent with the actual molding.
A mold temperature of 50 °C achieved the minimal warpage with
a flatness of 0.65 mm.

90 4.60

4.50

80 4.40
Temperature (oC)

4.30
70 4.20
Z (mm)

4.10
60
4.00

3.90
50
3.80

40 3.70

3.60

30 3.50
P2 P4 P6 P8 P10 P12 P14 P16 P18 P20 P22 P24 P26 P28
0 50 100 150 200 250 300 350 400 450 500 550
Position
Time (sec)
Fig. 18. Z-directional deformation profile of the molded part fabricated with local mold
Fig. 17. Cycle-by-cycle local mold temperature profiles. temperature settings.
110 S.-C. Nian et al. / International Communications in Heat and Mass Transfer 61 (2015) 102–110

Table 3 References
Warpage of local mold temperature settings.
[1] J. Fasset, Thin wall molding different in processing over standard injection molding,
Straightness (mm) Range (mm) Flatness (mm) SPE-ANTEC11995. 430–433.
[2] P. Staeblin, Thin-wall technology increases component utility, Kunststoffe Plast. Eur.
Right edge Left edge Top edge Bottom edge
87 (1997) 21–22.
0.11 0.06 0.19 0.23 0.31 0.31 [3] M.S. Despa, K.W. Kelly, J.R. Collier, Injection molding of polymeric LIGA HARNs,
Microsyst. Technol. 6 (1999) 60–66.
[4] J.W. Bozzelli, J. Cardinal, B. Fierens, Pressure loss in thin wall moldings, SPE-
ANTEC11997. 425–428.
[5] M.C. Huang, C.C. Tai, The effective factors in the warpage problem of an injection-
molded part with a thin shell features, J. Mater. Process. Technol. 110 (2001) 1–9.
[6] N.R. Subramanian, L. Tingyu, Y.A. Seng, Optimizing warpage analysis for an optical
housing, Mechatronics 15 (2005) 111–127.
[7] M.C. Song, Z. Liu, M.J. Wang, T.M. Yu, D.Y. Zhao, Research on effects injection process
parameters on the molding process for ultra-thin wall plastic parts, J. Mater. Process.
Technol. 187 (2007) 668–671.
[8] Y.J. Weng, S.Y. Yang, S.C. Nian, S.T. Huang, Y.J. Weng, Optimal process conditions of
shrinkage and warpage of thin-wall parts, Polym. Adv. Technol. 22 (2011) 891–902.
[9] S.C. Chen, Y. Chang, T.H. Chang, R.D. Chien, Influence of using pulsed cooling for
mold temperature control on microgroove duplication accuracy and warpage of
the Blu-ray Disc, Int. Commun. Heat Mass 35 (2008) 130–138.
[10] M.S. Huang, Y.L. Huang, Effect of multi-layered induction coils on efficiency and uni-
formity of surface heating, Int. J. Heat Mass Transf. 53 (2010) 2414–2423.
[11] R. Sánchez, J. Aisa, A. Martinez, D. Mercado, On the relationship between cooling
setup and warpage in injection molding, Measurement 45 (2012) 1051–1056.
[12] B. Ozcelik, I. Sonat, Warpage and structural analysis of thin shell plastic in the plastic
injection molding, Mater. Des. 30 (2009) 367–375.
[13] C.P. Chen, M.T. Chuang, Y.H. Hsiao, Y.K. Yang, C.H. Tsai, Simulation and experimental
Fig. 19. Comparison of z-directional deformation profiles for the cover fabricated with a study in determining injection molding process parameters for thin-shell plastic
single mold temperature of 50 °C (dashed line) and local mold temperatures (solid line). parts via design of experiments analysis, Expert Syst. Appl. 36 (2009) 10752–10759.
[14] H. Oktem, T. Erzurumlu, I. Uzman, Application of Taguchi optimization technique in
determining plastic injection molding process parameters for a thin-shell part,
Mater. Des. 28 (2007) 1271–1278.
[15] Y.M. Deng, Y. Zhang, Y.C. Lam, A hybrid of mode-pursuing sampling method and
genetic algorithm for minimization of injection molding warpage, Mater. Des. 31
(2010) 2118–2123.
(3) Applying the neutral axis theory can effectively indicate the [16] M.D. Azaman, S.M. Sapuan, S. Sulaiman, E.S. Zainudin, A. Khalina, Shrinkages and
warpage of each edge and suggested local mold temperature set- warpage in the processability of wood-filled polypropylene composite thin-walled
tings for the female and male mold plates at long and short parts formed by injection molding, Mater. Des. 52 (2013) 1018–1026.
[17] S.H. Tang, Y.J. Tan, S.M. Sapuan, S. Sulaiman, N. Ismail, R. Samin, The use of Taguchi
edges, respectively. In the local mold temperature experiment, method in the design of plastic injection mould for reducing warpage, J. Mater.
the mold temperature was set at approximately 40 °C–52 °C, as Process. Technol. 182 (2007) 418–426.
is listed in Table 2. The flatness (0.31 mm) of the cover fabricated [18] P. Larpsuriyak, H.G. Fritz, Warpage and Countermeasure for injection-molded in-
mold labeling parts, Polym. Eng. Sci. 51 (2011) 411–418.
using local mold temperature setting was superior to that fabri- [19] S.C. Chen, S.H. Tarng, C.Y. Tseng, Using pulsed cooling to reduce cycle time and im-
cated using a single mold temperature of 50 °C. Experimental prove part warpage, SPE-ANTEC22010. 867–871.
and simulation results both verified the feasibility of applying [20] H.L. Chen, S.C. Chen, W.H. Liao, R.D. Chien, Y.T. Lin, Effects of insert film on asymmet-
ric mold temperature and associated part warpage during in-mold decoration injec-
the neutral axis theory and local mold temperature settings to tion molding of PP parts, Int. Commun. Heat Mass 41 (2013) 34–40.
effectively control the warping of a thin-walled part. [21] http://en.wikipedia.org/wiki/Neutral_axis.