Numerical Simulation of Cavity Roughness Effects On Melt Filling in Microinjection Molding
Numerical Simulation of Cavity Roughness Effects On Melt Filling in Microinjection Molding
Numerical Simulation of Cavity Roughness Effects On Melt Filling in Microinjection Molding
Introduction
njection molding is one of the most common processes for cost-effective mass production of plas-
Therefore, the conclusions obtained from these research may not be applicable to the polymer melt
ow, which is high viscous, non-Newtonian, and in
low Reynolds number.
The roughness effects in microinjection molding
have been investigated by some researchers. Smialek
and Simpson8 pointed out that the increase of mold
surface roughness can prevent slippage during cavity lling and lead to a more appealing surface for
the molded parts. Theilade et al.9 found out that a
rough cavity surface yields a lower linear shrinkage.
Grifths et al.10 observed that the cavity roughness
has inuence in the level of turbulence of the melt
ow. However, its effect on the slip-stick phenomena was not identied. In our previous work,1,11,12
experiment was performed to investigate roughness
effects on ow area (or volume) through the lling of
a thin circular cavity with polyoxymethylene (POM).
Figure 1 shows the schematic diagram of the cavity
lling investigated in the experiment. The wall of
the cavity insert is smooth with comparatively negligible surface roughness. The core wall of the cavity
has different surface roughness, but the same roughness mean lines on its two semicircular halves. The
separating line between the two semicircular halves
of the core insert was adjusted vertically in the mold
such that gravity has the same effect on the two cavity halves during the cavity lling. As a result, the
two semicircular cavity halves are lled under the
same processing conditions, and the difference in
ow areas between the two halves is predominantly
caused by the difference in surface roughness between the two halves of the core insert. In the experiment, roughness values, cavity thicknesses, mold,
and melt temperatures were varied.
In this work, an inlet-velocity iterative numerical
procedure was developed to simulate the case encountered in our previous experiment. The rough-
90
During cavity lling, polymer melt ow is governed by the mass, momentum, and energy conservation equations. The melt and air are assumed to be
incompressible. The gravity effect is neglected, considering the fact that gravity force is very small as
compared with viscous force for the melt.13 The surface tension and wall slip effects are also neglected as
the dimensions of the molded parts discussed in this
work are much larger than 10 m.14,15 Furthermore,
the properties of the uids, such as density, thermal
conductivity, specic heat, and so on are assumed to
be constant. As a result, the governing equations for
the polymer melt ow are written as
Continuity equation
v = 0
(1)
Momentum equation
v
+ v v
t
= p + ( ),
= v + ( v )T
(2)
Energy equation
C p
T
+ v T
t
= (kT) + 2 ,
=
1
:
2
(3)
(4)
DOI 10.1002/adv
a i (y)
A
i=1
(5)
(T, , P) =
(6)
where
T (P) = D2 + D3 P
A2 = A2 + D3 P
n, , D1 , D2 , D3 , A1 , A 2 are constants.
DOI 10.1002/adv
+u
= 0,
t
|| = 1
(7)
(9)
+
1
h=
+
sin
2
1
<
||
(10)
>
91
V1 U
V1 + U
P>0
P<0
(13)
V2 = 2Uin V1
where U is the change in the inlet velocities for
each iteration, which can be determined by using
the following equation:
U = Uin
(14)
(11)
(12)
92
(15)
DOI 10.1002/adv
SOLUTION PROCEDURE
TABLE I
22
DOI 10.1002/adv
* (Pa)
0 28.5 51.6
TABLE II
Properties of the POM Melt, Air, and Mold Material
Material
POM melt
Air
Mold (steel)
0.14
0.037
29
2101
1.0
460
2 H + d 2U t
dini
eff
1 in
4Heff
(16)
93
94
INVESTIGATIONS OF ROUGHNESS
EFFECTS
In the simulation, and PL are set as 1% and 1%,
respectively. Nonuniform grids were used to discretize the computing domain, with ner grids near
the roughness layer. Figure 8 shows the development of the inlet velocities on the two semicircular
DOI 10.1002/adv
halves. The cavity thickness was 300 m. The injection velocity (i.e., Uin ) was 23.5 m/s1 , and the mold
and melt temperatures were 323 K and 453 K, respectively. It can be observed that the melt ows faster on
the smoother half than it does on the rougher half.
Furthermore, the ow speed increases with time on
the smoother half and decreases correspondingly on
the rougher half. Figure 9 shows the pressure development on the two cavity halves. It can be observed
that the pressures on the two halves obtained by
using the iterative procedure are quite close.
Using the inlet velocity values obtained, the ow
area on each cavity half at the nth time step can be
calculated by using the following equation:
Area =
n
Ain t
Vi
2Heff
(17)
i=1
DOI 10.1002/adv
(18)
95
96
DOI 10.1002/adv
References
Conclusion
The roughness effects on ow area were investigated numerically with the aid of a developed
inlet-velocity iterative procedure. The procedure
can simulate the development of melt ow front
quite accurately. The numerical results revealed
that cavity roughness does resist polymer melt
ow during lling the hot melt into the cavity at
a lower temperature. However, the signicance of
roughness effects is dependent on mold and melt
temperatures, cavity thickness, and injection rate.
The trends of the roughness effects for different
mold and melt temperatures obtained from the
simulation are in good agreement with those
obtained from the experiment, that is, when the
melt temperature is kept constant, an increase in
the mold temperature will reduce the roughness
effects. Furthermore, the mold temperature is more
signicant for the roughness effects when the melt
temperature is kept constant at a low level. Similarly,
when the mold temperature is kept constant, an
increase in the melt temperature will also reduce
the roughness effects. The melt temperature is more
signicant for the roughness effects when the mold
temperature is kept constant at a low level. The relation between k value and the cavity thickness can be
approximated by a logarithm function. There exists
1. Zhang, H. L.; Ong, N. S.; Lam, Y. C. Polym Eng Sci 2007, 47,
20122019.
2. Croce, G.; Agaro, P. D. Superlattices Microstruct 2004, 35,
601616.
3. Kandlikar, S. G.; Joshi, S.; Tian. S. R. Heat Transfer Eng 2003,
24, 416.
DOI 10.1002/adv
7. Qu, W. L.; Mala, G. H.; Li, D. Q. Int J Heat Mass Trans 2000,
43, 39253936.
8. Smialek, C. D.; Simpson, C. L. In ANTEC 1998, Vol. 3,
pp. 33733377.
9. Theilade, U. R.; Kjaer, E. M.; Hansen, H. N. In ANTEC 2003,
Vol. 1, pp. 463467.
10. Grifths, C. A.; Dimov, S. S.; Brousseau, E. B.; Hoyle, R. T. J
Mater Process Tech 2007, 189, 418427.
11. Zhang, H. L.; Ong, N. S.; Lam, Y. C. Int J Adv Manuf Tech
2008, 37, 11051112.
12. Zhang, H. L.; Ong, N. S.; Lam, Y. C. Polym Eng Sci 2008, 48,
490495.
13. Haagh, G.; Vosse, F. Int J Numer Methods Fluids 1998, 28,
13551369.
14. Yao, D. G.; Kim, B. J Micromech Microeng 2002, 12, 604610.
15. Young, W. B. Appl Math Model 2007, 31, 17981806.
16. Osher, S.; Sethian, A. J Comput phys 1988, 79, 249.
17. Yap, Y. F.; Chai, J. C.; Wong, T. N.; Toh, K. C.; Zheng, H. Y.
Numer Heat Tr B-Fund 2005, 50, 455472.
18. Chang, Y. C.; Hou, T. Y.; Merriman, B.; Osher, S. J Comput
phys 1996, 124, 449464.
19. Sussman, M.; Puckett, E. G. J Comput Phys 2000, 162, 301
337.
20. Sussman, M.; Smereka, P.; Osher, S. J Comput phys 1994, 114,
146159.
21. Chang, R. Y.; Yang, R. Y. Int J Numer Methods Fluids 2001,
37, 125148.
22. Patankar, S. V. Numerical Heat Transfer and Fluid Flow;
Hemisphere: Washington, DC, 1980.
97