- Email: [email protected]
Enhancement of induction heating efficiency on injection mold surface using a novel magnetic shielding method
International Communications in Heat and Mass Transfer 50 (2014) 52–60
Contents lists available at ScienceDirect
International Communications in Heat and Mass Transfer journal homepage: www.elsevier.com/locate/ichmt
Enhancement of induction heating efficiency on injection mold surface using a novel magnetic shielding method☆ Shih-Chih Nian a,⁎, Ming-Shyan Huang b, Tzung-Hung Tsai b a b
Department of Power Mechanical Engineering, National Taitung College, 889 Jhengci N. Rd., Taitung City 95045, Taiwan, ROC 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
Available online 5 December 2013 Keywords: Ferrite material Induction heating Magnetic shielding Proximity effect
a b s t r a c t Mold temperature is a major factor in the quality of injection molding process. A high mold temperature setting is feasible to enhance the molding quality but prolongs the cooling time. Induction heating is the method currently used to heat the mold surface without increasing the molding cycle. However, one unresolved problem of induction heating is the proximity effect resulting from two adjacent coils with different current directions. The proximity effect substantially decreases heating efficiency, which then causes non-uniform heating. This effect is difficult to avoid in a single-layer coil. The most common solution, which is to use magnetic concentrators to reduce the proximity effect, does not obtain satisfactory results. In the novel magnetic shielding induction heating method developed in this study, heating efficiency and temperature uniformity are enhanced by using ferrite materials to separate the conflicting magnetic fields caused by the repulsive proximity effect. Three typical single-layer coils are investigated in this study, including a reciprocated single-layer coil, a single-layer spiral coil, and a rectangular frame coil. Appropriate placement of ferrite materials on these induction coils successfully eliminated the proximity effect, increased the heating rate, and improved temperature uniformity. © 2013 Elsevier Ltd. All rights reserved.
1. Introduction In the injection molding process, mold temperature is an important factor in the quality of injection molding process. A high mold temperature setting is feasible to enhance the molding quality but increases the cooling time and parts' demolding temperature. To enable cavity filling under a high mold temperature and parts' ejection under low temperature, a variotherm process was developed for high-precision injection molding, thin-walled injection molding and micro-injection molding [1–3]. The variotherm process dynamically controls the mold temperature, preheats the mold to a higher temperature before the melt enters the mold cavities, and then decreases the mold temperature to the ejection temperature after the cavities are filled. Common mold heating approaches for a variotherm process include infrared heating [4,5], gas-assisted heating [6], thin-film resistance heating [7], resistance cartridge heating [8,9], steam heating [10], and induction heating [11–14]. The advantages of induction heating, in which only the mold surface is heated, include rapid heating and cooling, good controllability, energy saving, and allowing local heating. Compared with conventional oil heating, induction heating has better potential for increasing mold temperature and reducing cycle time. The induction heating method has been widely used in the various industrial manufacturing processes. Since the skin effect of ☆ Communicated by W.J. Minkowycz. ⁎ Corresponding author at: Department of Power Mechanical Engineering, National Taitung College, 889 Jhengci N. Rd., Taitung City 95045, Taiwan, ROC. 0735-1933/$ – see front matter © 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.icheatmasstransfer.2013.11.017
induction heating concentrates the resulting Joule heating at the surface of processed workpiece, the surface can be heated by using a high-frequency current. Therefore, induction heating is an efficient variotherm process for injection molding. Studies show that induction heating rapidly increases mold surface temperature, substantially shortens cycle time [11], increases the molding ability of thin-walled injection molding and microinjection molding [12], improves the replication of micro-features [13], and reduces the welding line defect of injection molding parts [14]. Induction coil designs can be classified into four types: unmovable external hung-up type, movable external hung-up type, wrapped type, and inserted type. The first type is easily implemented because the induction coil is independent of the processed workpiece. Chen [15] used induction heating to increase the surface temperature of injection molds and found that both heating rate and uniformity are satisfactory when the area of the processed workpiece is 100 mm × 100 mm. The movable external type can be either a fixed induction coil with a movable processed workpiece or a fixed processed workpiece with a movable induction coil. The former is more common, particularly in metal welding or surface treatment [16–18]. In wrapped induction heating, induction coils are used to embed the whole workpiece. Chen [19] reported that wrapped induction heating is effective for injection molding. In the inserted type, the induction coils are placed behind cores and cavities so that the penetration effect and heat conduction provide indirect heating of the mold surface. This technique is superior in terms of reduced cycle time because the induction coils do
S.-C. Nian et al. / International Communications in Heat and Mass Transfer 50 (2014) 52–60
not have to be moved. However, the placement of induction coils inside the molds requires consideration of the layout of the cooling channel, the layout of the ejection system, and the rigidity of the mold structure. Induction heating is apparently an effective solution for heating the mold surface without increasing the molding cycle period. However, the main unresolved problem of using induction heating in injection molding is the non-uniform temperature distribution on the heating surface. The various causes of differences in heating temperature include the coil turn space, heating distance, cavity geometry, magnetic interference, and the proximity effect. For instance, different directions of electric current along induction coils can cause a non-uniform temperature distribution by introducing a repulsive proximity effect. For a practical coil design, however, opposite directions of coil current are difficult to avoid. In a study of the effect of repulsive proximity on non-uniform temperature distribution, Sung et al. [20] compared temperature uniformity in the heating surfaces of different induction heating coils. The experimental results showed poor heating efficiency in the singlelayer coil with opposite current directions. However, the double-layer reciprocating coil and the coil with magnetic flux concentrator efficiently increased heating speed and provided uniform temperature distributions. Huang [21] designed a multi-layer coil for improving heating efficiency and temperature uniformity in a conventional single-layer coil. The experimental results showed that the multi-layer coil has a more uniform temperature distribution and better heating efficiency compared to the conventional single-layer coil. Although, the double-layer reciprocating coil can avoid the heating layer (i.e., the layer near the heating surface) with opposite current direction, the double-layer coil must have reduplicate coil length of single-layer reciprocating coil. However, the increased coil length increases coil manufacturing cost and energy consumption. The local multi-layer coil has a more uniform temperature distribution and better heating efficiency; applications of the local multi-layer coil are limited
53
Fig. 2. Magnetic flux fields of two adjacent opposite current coils: (a) with magnetic concentrator; (b) with magnetic shielding material.
to spiral-like forms. Applying magnetic flux concentrator to control magnetic flux field and eliminate proximity effect is the most popular method in the induction heating process, but fails to completely isolate the conflicting magnetic fluxes when opposite current coils were induced.
Fig. 1. Effect of current direction on magnetic flux line: (a) coils with identical current direction; (b) coils with opposite current direction.
Fig. 3. Dimensions of the heated plate with cooling channels.
54
S.-C. Nian et al. / International Communications in Heat and Mass Transfer 50 (2014) 52–60
Fig. 4. Dimensions of experimental coils and the heating area: (a) single-layer reciprocated coil; (b) single-layer spiral coil; (c) rectangular frame coil.
2. Induction heating principles Induction heating of metal is performed by electromagnetic induction. According to the Faraday law and the Lenz law, passing an electrical alternating current through heating coils produces an alternating magnetic field that can be used to heat an object. If processed magnetic or non-magnetic conductive workpieces are placed in the alternating magnetic field established by the heating coils, cutting-of-flux produces Table 1 Magnetic properties of ferrite materials. Properties
Initial permeability, μi
Relative permeability, μr
Resistivity
Ni–Zn ferrite Mn–Zn ferrite
1500 2300
1.19 × 108 1.83 × 108
107 Ω-m 8 Ω-m
current at different depths. The workpieces' resistance and the flow of eddy currents therein generate heating power [22]. Induction heating exploits two effects, the Joule effect, which is based on hysteresis loss and eddy current loss, and the electromagnetic effect, which mainly comprises the skin effect and the proximity effect. These effects are described further below. 2.1. Skin effect Passing an alternating current through a conductive wire generates a magnetic field both inside and outside the conductor. Varying the current then causes a variation in the magnetic field in the conductor, which generates an induction electromotive force and, thus, a current in the conductor. The current in the conducting wire section is distributed as follows: the current near the center of the conducting wire is very
S.-C. Nian et al. / International Communications in Heat and Mass Transfer 50 (2014) 52–60
55
Fig. 5. Ferrite materials' location and heating temperature distribution of reciprocated single-layer coil: (a) without ferrite slices; (b) without ferrite slices after 10 s heating; (c) without ferrite slices after 20 s heating; (d) with ferrite slices; (e) with ferrite slices after 10 s heating; (f) with ferrite slices after 20 s heating.
small whereas that at the surface is relatively large. The cause of the difference is known as the skin effect. In induction heating, a highfrequency current passing through the coil produces the largest induction eddy current at the workpiece surface. The eddy current declines exponentially with distance from the surface.
2.2. Proximity effect The proximity effect can be observed when conductive wires are placed in the vicinity of each other. The magnetic fields of the two conductive wires affect each other, and the associated magnetic field on the mold surface changes [23]. Fig. 1 shows the effect of current direction on magnetic flux line. When two coils have an identical current direction, the magnetic flux lines are combined throughout the mold surface, which enables uniform heating of the mold surface. However, in coils with opposite current directions, the magnetic flux lines are separated and driven away from the center of the coils. Therefore, excluding the magnetic field substantially decreases induction heating efficiency and temperature uniformity.
3. Methodology This study proposes a novel magnetic shielding induction heating method to solve the repulsive proximity problem. By using ferrite materials to separate the conflict magnetic fields to eliminate the influences of repulsive proximity effect, the heating efficiency and temperature uniformity were thus enhanced. Fig. 2 shows the magnetic flux fields of two adjacent opposite current coils combined with a magnetic concentrator and magnetic shielding material. Whereas the magnetic concentrator can separate the magnetic fluxes between the magnetic concentrators and concentrate the magnetic flux below the coils, the magnetic concentrator cannot avoid the proximity effect that occurs under the coils. Thus, the center of the workpiece has a lower heating efficiency and a lower temperature. In contrast with the magnetic concentrator method, the proposed magnetic shielding method completely separates the magnetic flux fields from different coils and
2.3. Magnetic concentrator A magnetic concentrator has material properties resembling those of a transformer coil, i.e., high conductivity and low resistance. It is used in the induction heating process for local enhancement of the magnetic field, which thus improves the heating effect. When the processed workpiece is close to the induction coil, the proximity effect changes the current distribution. The current is distributed along the coil surface, which causes a loose distribution of eddy current on the processed workpiece. Thus, the surface heating effect is insufficient. Use of a magnetic concentrator can enforce the coil current and eddy current distribution, which then enhances its heating rate [23].
Fig. 6. Dimensions of 9-point measurement.
56
S.-C. Nian et al. / International Communications in Heat and Mass Transfer 50 (2014) 52–60
4.2. Experimental coil design This experiment used three typical single-layer coils, the reciprocated single-layer coil, the single-layer spiral coil, and the rectangular frame coil, to perform a series of induction heating experiments. Fig. 4 shows the coil designs and heating area (shaded area). The coils were composed of 5 mm diameter copper tube, and the heating distance from the heated-plate surface to the coil was 15 mm in all three coils. Fig. 4(a) shows that the reciprocated single-layer coil has a 25 mm pitch and a heating area of 75 mm × 75 mm. Fig. 4(b) shows that the single-layer spiral coil has a 15 mm pitch and a heating area of 75 mm × 75 mm. Fig. 4(c) shows the rectangular frame coil around a 121 mm long × 76 mm wide rectangular area. The heating area was imitated with the front cover of a 5-inch smartphone. 4.3. Magnetic shielding material In the magnetic shielding induction heating method, ferrite materials are used to separate the conflicting magnetic fields. The normal ferrite materials mainly have two types, the Mn–Zn and the Ni–Zn ferrites. Both ferrite materials are widely used in transformer coils, and are easily magnetized and demagnetized. Table 1 shows the magnetic properties of the Mn–Zn and Ni–Zn ferrites. The Ni–Zn ferrite was selected for this experiment. The Ni–Zn ferrite has much higher resistivity (108 Ω-m) compared to the Mn–Zn ferrite (8 Ω-m). The higher resistivity is advantageous to avoid the ferrite to be heated during the induction heating process, and therefore has less extra energy consumption. 4.4. Setup for magnetic shielding induction heating
Fig. 7. Temperature profile of single-layer reciprocated coil: (a) without ferrite slices; (b) with ferrite slices.
drives the magnetic flux uniformly throughout the workpiece surface. This study performed a series of induction heating experiments in three typical single-layer coils: the reciprocated single-layer coil, the single-layer spiral coil, and the rectangular frame coil. Both heating efficiency and temperature uniformity were compared.
4. Experimental setup 4.1. Experimental tools A series of induction heating experiments were performed to evaluate the use of ferrite materials in the proposed magnetic shielding induction heating method. A normal mold steel AISI-P20 with dimensions of 170 mm long × 170 mm wide × 20 mm thick was heated by a high frequency induced heating machine (HP-25KW, Honor, Taiwan). Fig. 3 shows the geometry and cooling channel (diameter: 8 mm) dimensions of the heated plate. The mold temperature controller was used to pass 60 °C heat-transfer oil through the cooling channels during all heating/cooling process. A thermometer and infrared ray thermal imaging system (ThermoVision A20, FLIR, USA) were also used to record the temperature variation of the heated-plate surface.
Figs. 5(d), 8(d), and 10(d) show the locations of the ferrite materials relative to the three experimental coils. The ferrites were placed either at the corners of the coil or in the middle between the copper-tube conductors with opposite current directions. Fig. 5(d) illustrates the locations of the ferrite materials in the reciprocated single-layer coil. The coil current directions of the vertical conductors were opposite to the adjacent others, and a repulsive proximity effect was induced between the vertical conductors during the induction heating process. Five ferrite slices with 4 mm thickness were used in the reciprocated single-layer coil. Three ferrite slices were inserted into the middle between each adjacent vertical conductor to eliminate the proximity effect. Additionally, two ferrite slices were placed at the outside of coil right/left boundaries to limit the effect area of magnetic flux. Fig. 8(d) shows the locations of ferrite materials in the single-layer spiral coil. The circling coil current direction in the spiral coil induced a repulsive proximity effect at the spiral core. Therefore, the center of the heated surface had a lower temperature during the induction heating process [22]. A 4 mm diameter ferrite cylinder was inserted into the center of the spiral core to eliminate the proximity effect. Additionally, a 100 mm × 100 mm square ferrite slice (thickness: 4 mm) was placed over the spiral coil with 5 mm distance from coil surface to ferrite surface, to increase heating speed. Fig. 10(d) shows the locations of the five ferrite slices used in the rectangular frame coil. Four ferrite slices were placed at the corners of the coil to increase the heating efficiency of the corner area, and a ferrite slice was placed at the middle of the coil inlet and outlet, where the copper-tube conductor had an opposite current direction. 5. Results and discussion 5.1. Experimental results for reciprocated single-layer coil The reciprocated single-layer coil shows a large repulsive proximity effect between the adjacent copper-tube conductors with opposite current direction. In this experiment, five ferrite slices were used to improve the heating efficiency of the reciprocated single-layer coil.
S.-C. Nian et al. / International Communications in Heat and Mass Transfer 50 (2014) 52–60
57
Fig. 8. Ferrite materials' location and heating temperature distribution of single-layer spiral coil: (a) without ferrite materials; (b) without ferrite materials, after 10 s heating; (c) without ferrite materials, after 20 s heating (d) with a ferrite cylinder; (e) with a ferrite cylinder, after 10 s heating; (f) with a ferrite cylinder, after 20 s heating; (g) with a ferrite cylinder and covered with a ferrite slice; (h) with a ferrite cylinder and covered with a ferrite slice, after 10 s heating; (i) with a ferrite cylinder and covered with a ferrite slice, after 20 s heating.
Fig. 5 shows the thermal images of mold surface captured by the thermometer and infrared ray thermal imaging system after the mold surface was heated for 10 s and 20 s by reciprocated single-layer coil with/without ferrite slices. Fig. 5(b) and (c) shows that repulsive proximity effect substantially limits the heating effect of the reciprocated single-layer coil on the mold surface. The maximum temperature inside the heating area (inside black lines) increased only 5 and 10 °C after 10 and 20 s heating, respectively. Specifically, the center area shows a very small surface temperature increase. Fig. 5(e) and (f) shows that the ferrite slices effectively solve the heating problem in the reciprocated single-layer coil. The maximum temperature inside the heating area increased from 60 to 125 and 158 °C after 10 and 20 s heating. The temperature distribution shows that the temperature of the mold surface under the copper tube was
higher than that of the mold surface in contact with the ferrite slices, which resulted from the area of the mold surface in contact with the ferrite slices which also shows a proximity effect (Fig. 2(b)). The temperature profiles of the heated area during the conduction heating process were obtained by a nine-point measurement in experiments performed in a reciprocated single-layer coil and a single-layer spiral coil. Figs. 6 and 7 show temperature profiles obtained by the 9point measurement at points P4, P5, P7, and P8 during conduction heating. The experimental results show that the coil without ferrite had a small temperature variation (Fig. 7(a)) that the coil with ferrite slices had a faster heating speed at beginning 5 s, and that the speed of increase in surface temperature slowed after 20 s (Fig. 7(b)). The experiment also confirms that the high resistivity (108 Ω-m) of Ni–Zn ferrite can avoid the ferrite slices to be heated by the induction heating coils. Therefore, the ferrite slices can maintain at a low temperature.
58
S.-C. Nian et al. / International Communications in Heat and Mass Transfer 50 (2014) 52–60
proximity effect and substantially increase the heating speed of the reciprocated single-layer coils. 5.2. Experimental results for the single-layer spiral coil The core of the single-layer spiral coil shows a severe repulsive proximity effect. This study compared three ferrite arrangements in terms of their effects on spiral coil heating, including a conventional spiral coil (without ferrite material) (Fig. 8(a)), a spiral coil center inserted into a ferrite cylinder (diameter: 4 mm) (Fig. 8(d)), and a spiral coil inserted into a ferrite cylinder and covering a ferrite slice (thickness: 4 mm) (Fig. 8(g)). Fig. 8 shows thermal images of the mold surface after 10 and 20 s of heating by the spiral coils. The maximum and minimum temperatures of the heated area are shown at the top-right and center of the thermal images. The thermal images of the mold surfaces heated for 10 s show that: (1) inserting a ferrite cylinder into the core of the spiral coil substantially increases core surface temperature from 73 °C (Fig. 8(b)) to 82 °C (Fig. 8(e)) and decreased the temperature deviation from 20 °C (max. 93 °C, min. 73 °C) to 15 °C (max. 97 °C, min. 82 °C). (2) Inserting a ferrite cylinder slightly increased heating speed and increased the maximum temperature from 93 °C (Fig. 8(b)) to 97 °C (Fig. 8(e)); (3) covering a slice of ferrite above the spiral coil substantially increased heating speed, increased the maximum temperature from 97 to 117 °C, and increased the surface core temperature from 82 to 97 °C (see Fig. 8(e) and (h)). Furthermore, when the induction heating time was increased to 20 s, all surface temperatures increased, but the variation in temperature deviation showed a different trend. When the heating time was increased from 10 to 20 s, the temperature deviation in the conventional spiral increased from 20 to 23 °C (Fig. 8(b) and (c)), but the temperature deviation in the spiral coil with ferrite cylinder decreased from 15 to 12 °C (Fig. 8(e) and (f)). In an actual injection molding system with external induction heating, 2 to 5 s are needed to remove the coil from the mold, clamp the mold, and inject the melt into the cavity. During this interval, the heated surface cools. Fig. 9 shows the temperature variation measured at points P4, P5, P7, and P8 (see Fig. 6) relative to heating/cooling time in an experimental plate induction heated for 20 s then allowed to stand for 5 s. The experimental results show that, at the induction heating stage, the heated surface of the spiral coil with ferrite cylinder had a smaller temperature deviation and that the spiral coil with ferrite cylinder and ferrite slice had a faster heating speed and a larger temperature deviation. During the cooling stage, due to the heat transfer effect, the core temperature (at P5) had slower cool-down speed, which enhanced temperature uniformity. The temperature deviation of the spiral coil with cylinder ferrite was decreased by 5 °C after standing for 5 s (Fig. 9(b)). The spiral coil experiment confirmed that inserting a ferrite cylinder into the core of the spiral coil reduced the proximity effect and increased the surface core temperature, whereas covering a ferrite slice above the coil substantially increased the heating speed of the spiral coil. 5.3. Experimental results for the rectangular frame coil
Fig. 9. Temperature profile of single-layer spiral coil: (a) without ferrite materials; (b) with a ferrite cylinder; (c) with a ferrite cylinder and covered with a ferrite slice.
Problems caused by overheating decreased, and energy consumption was reduced. The experimental results for the reciprocated singlelayer coil confirmed that the ferrite slices efficiently eliminate the
The rectangular frame coil showed a large repulsive proximity effect between the coil inlet and outlet area and a low heating efficiency at the four corner areas. In this experiment, inserting five ferrite slices (Fig. 10(d)) effectively solved the above problem of non-uniform heating in the rectangular frame coil. Fig. 10 shows thermal images of the mold surface after heating by the rectangular frame coils for 10 and 20 s. The maximum and minimum temperatures of the heated area are displayed at the top-right of the thermal images. The images show that the ferrite slices efficiently improve temperature uniformity during conventional rectangular frame coil heating. The thermal images of the mold surface heated for 10 s revealed three findings: (1) the large repulsive proximity effect reduced the heating effect of the conventional coil in the area between
S.-C. Nian et al. / International Communications in Heat and Mass Transfer 50 (2014) 52–60
59
Fig. 10. Ferrite materials' location and heating temperature distribution of rectangular frame coil: (a) without ferrite slices; (b) without ferrite slice, after 10 s heating; (c) without ferrite slice, after 20 s heating; (d) with ferrite slices; (e) with ferrite slices, after 10 s heating; (f) with ferrite slices, after 20 s heating.
the inlet and outlet copper-tube conductors. The smallest increase was 2 °C (from 60 to 62 °C) (see Fig. 10(b)). (2) Inserting a ferrite slice between the inlet and outlet copper-tube conductors efficiently eliminated the repulsive proximity effect. The temperature increased from 2 to 9 °C (from 60 to 69 °C) (see Fig. 10(e)), and the temperature deviation decreased from 11 °C (max. 73 °C, min. 62 °C) to 7 °C (max. 76 °C, min. 69 °C). (3) The thermal image of the conventional rectangular frame coil shows a corner retract phenomena at the rectangular frame corners; that is, compared to the coil corner, the heated region offsets to the inside direction of the corner and significantly decreases the heating temperature in the corner area (see Fig. 10(b)). (4) Inserting ferrite slices on the outside of the rectangular frame coil corner efficiently eliminated the corner retract phenomena in the heated surface and withdrew the offset heated areas to the correct positions below the coil corner (see Fig. 10(e)). Further, when the induction heating time was increased from 10 to 20 s, all surface temperatures increased, but temperature deviations only slightly increased; the temperature deviation of the conventional rectangular frame coil increased from 11 °C (max. 73 °C, min. 62 °C) to 13 °C (max. 78 °C, min. 65 °C) (see Fig. 10(b) and (c)), and the temperature deviation of the rectangular frame coil with ferrite slices increased from 7 °C (max. 73 °C, min. 62 °C) to 8 °C (max. 78 °C, min. 65 °C) (see Fig. 10(e) and (f)). 6. Conclusion This study developed a magnetic shielding method that effectively eliminates the repulsive proximity effect during the induction heating process. A high-resistivity Ni–Zn ferrite was used as magnetic shielding material to separate the conflicting magnetic fields. The three typical single-layer coils used in this study included the reciprocated singlelayer coil, the single-layer spiral coil, and the rectangular frame coil. After recording temperature variation at the mold surface with a thermometer and infrared ray thermal imaging system, the recorded thermal images were used to analyze heating efficiency and temperature uniformity. The experimental results confirm that the magnetic shielding
method is a simple and feasible scheme for eliminating the proximity effect, increasing induction heating speed, and increasing the uniformity of temperature distribution. The following conclusions are drawn: (1) The Ni–Zn ferrite is a suitable shielding material for separating opposing magnetic fields. According to the high resistivity property of Ni–Zn ferrite, the Ni–Zn ferrite cannot be heated by induction coil, and thus has less heat influence to the mold surface and extra energy consumption. (2) The conventional reciprocated single-layer coil has a severe repulsive proximity effect between the copper-tube conductors, which reduces heating efficiency. The maximum temperature inside the heating area was increased only 5 °C after 10 s heating. Heating efficiency was improved by inserting ferrite slices between the copper-tube conductors, which increased the maximum temperature of heating area from 60 to 125 °C after 10 s heating. (3) In the conventional spiral coil, the lower temperature at the core area of the mold surface causes a non-uniform temperature distribution. However, inserting a ferrite cylinder into the core substantially improves heating speed and temperature uniformity. Additionally, covering a ferrite slice above the spiral coil increases the heating efficiency of spiral coils. (4) The conventional rectangular frame coil shows a severe repulsive proximity effect between the copper-tube conductors' inlet and outlet area, which reduces heating efficiency. Inserting a ferrite slice between the inlet and outlet conductors efficiently eliminates the repulsive proximity effect and solves the heating difficult problem. (5) A problem of the conventional rectangular frame coil is that the corners of the rectangular frame have retracting phenomena. That is, the heated region compared to the coil corner produces an offset to the inside direction of corner. Inserting ferrite slices on the outside of the rectangular frame coil corner efficiently eliminates the corner retract phenomena and withdraws the offset of heated areas to the correct position below the coil corners.
60
S.-C. Nian et al. / International Communications in Heat and Mass Transfer 50 (2014) 52–60
In summary, all experimental results in this study agree that the magnetic shielding method is feasible and efficient for reducing the repulsive proximity effect, for increasing heating speed, and for increasing temperature uniformity during the induction heating process. Acknowledgment The authors thank the Precision Mold and Die Research and Development Center for providing assistance with the experimental equipments. Ted Knoy is appreciated for his editorial assistance. Reference [1] G. Wang, G. Zhao, H. Li, Y. Guan, Analysis of thermal cycling efficiency and optimal design of heating/cooling systems for rapid heat cycle injection molding process, Mater. Des. 31 (2010) 3426–3441. [2] S.Y. Yang, S.C. Nian, S.T. Huang, Y.-J. Weng, A study on the micro-injection molding of multi-cavity ultra-thin parts, Polym. Adv. Technol. 22 (6) (2011) 773–1082. [3] S.C. Nian, S.Y. Yang, Molding of thin sheets using impact micro-injection molding, Int. Polym. Process. 20 (4) (2005) 441–448. [4] P.C. Chang, S.J. Hwang, Simulation of infrared rapid surface heating for injection molding, Int. J. Heat Mass Transf. 49 (2006) 3846–3854. [5] P.C. Chang, S.J. Hwang, Experimental investigation of infrared rapid surface heating for injection molding, J. Appl. Polym. Sci. 102 (2006) 3704–3713. [6] S.C. Chen, R.D. Chien, S.H. Lind, M.C. Lin, J.A. Chang, Feasibility evaluation of gas-assisted heating for mold surface temperature control during injection molding process, Int. Commun. Heat Mass Transfer 36 (8) (2009) 806–812. [7] D. Yao, B. Kim, Development of rapid heating and cooling systems for injection molding applications, Polym. Eng. Sci. 42 (12) (2002) 2471–2481. [8] G. Wang, G. Zhao, Y. Guan, Research on optimum heating system design for rapid thermal response mold with electric heating based on response surface methodology and particle swarm optimization, J. Appl. Polym. Sci. 119 (2) (2011) 902–921. [9] G. Wang, G. Zhao, Y. Guan, Thermal response of an electric heating rapid heat cycle molding mold and its effect on surface appearance and tensile strength of the molded part, J. Appl. Polym. Sci. 128 (3) (2013) 1339–1352.
[10] M.C. Jeng, S.C. Chen, P.S. Minh, J.A. Chang, C.S. Chung, Rapid mold temperature control in injection molding by using steam heating, Int. Commun. Heat Mass Transfer 37 (9) (2010) 1295–1304. [11] S.C. Chen, Y.W. Lin, R.D. Chien, H.M. Li, Variable mold temperature to improve surface quality of microcellular injection molded parts using induction heating technology, Adv. Polym. Technol. 27 (4) (2008) 224–232. [12] M.-S. Huang, N.-S. Tai, Experimental rapid surface heating by induction for micro-injection molding of light-guided plates, J. Appl. Polym. Sci. 113 (2009) 1345–1354. [13] S.C. Chen, W.R. Jong, Y.J. Chang, J.A. Chang, J.C. Cin, Rapid mold temperature variation for assisting the micro injection of high aspect ratio micro-feature parts using induction heating technology, J. Micromech. Microeng. 16 (2006) 1783–1791. [14] S.C. Chen, W.R. Jong, J.A. Chang, Dynamic mold surface temperature control using induction heating and its effects on the surface appearance of weld line, J. Appl. Polym. Sci. 101 (2006) 1174–1180. [15] S.C. Chen, H.S. Peng, J.A. Chang, W.R. Jong, Simulation and verification of induction heating on a mold plate, Int. Commun. Heat Mass Transfer 31 (7) (2004) 971–980. [16] R. Rudolf, P. Mitschang, M. Neitzel, Induction heating of continuous carbonfibre-reinforced thermoplastics, Compos. A: Appl. Sci. Manuf. 31 (11) (2000) 1191–1202. [17] K.Y. Bae, Y.S. Yang, C.M. Hyun, S.H. Cho, Derivation of simplified formulas to predict deformations of plate in steel forming process with induction heating, Int. J. Mach. Tools Manuf. 48 (15) (2008) 1646–1652. [18] I. Magnabosco, P. Ferro, A. Tiziani, F. Bonollo, Induction heat treatment of a ISO C45 steel bar: experimental and numerical analysis, Comput. Mater. Sci. 35 (2) (2006) 98–106. [19] S.C. Chen, Y.T. Lin, J.A. Chang, Investigation of Mold Heating by Induction Heating With Embedded Coils Within Mold Interior, SPE-ANTEC, Orlando, Florida USA, 2010. 1431–1434. [20] Y.T. Sung, Y.N. Lin, S.J. Hwang, H.H. Lee, D.Y. Huang, Design of Induction Heating Module for Uniform Cavity Surface Heating, SPE-ANTEC, Orlando, Florida USA, 2012. 1427–1436. [21] M.-S. Huang, Y.-L. Huang, Effect of multi-layer induction coils on efficiency and uniformity of surface heating, Int. J. Heat Mass Transf. 53 (2010) 2414– 2423. [22] V. Rudnev, R. Loveless, R. Cook, M. Black, Handbook of Induction Heating, Marcel Dekker, New York, 2003. 99–120. [23] T.J. Ahmed, D. Stavrov, H.E.N. Bersee, A. Beukers, Induction welding of thermoplastic composites—an overview, Compos. Part A 37 (2006) 1638–1651.
Contents lists available at ScienceDirect
International Communications in Heat and Mass Transfer journal homepage: www.elsevier.com/locate/ichmt
Enhancement of induction heating efficiency on injection mold surface using a novel magnetic shielding method☆ Shih-Chih Nian a,⁎, Ming-Shyan Huang b, Tzung-Hung Tsai b a b
Department of Power Mechanical Engineering, National Taitung College, 889 Jhengci N. Rd., Taitung City 95045, Taiwan, ROC 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
Available online 5 December 2013 Keywords: Ferrite material Induction heating Magnetic shielding Proximity effect
a b s t r a c t Mold temperature is a major factor in the quality of injection molding process. A high mold temperature setting is feasible to enhance the molding quality but prolongs the cooling time. Induction heating is the method currently used to heat the mold surface without increasing the molding cycle. However, one unresolved problem of induction heating is the proximity effect resulting from two adjacent coils with different current directions. The proximity effect substantially decreases heating efficiency, which then causes non-uniform heating. This effect is difficult to avoid in a single-layer coil. The most common solution, which is to use magnetic concentrators to reduce the proximity effect, does not obtain satisfactory results. In the novel magnetic shielding induction heating method developed in this study, heating efficiency and temperature uniformity are enhanced by using ferrite materials to separate the conflicting magnetic fields caused by the repulsive proximity effect. Three typical single-layer coils are investigated in this study, including a reciprocated single-layer coil, a single-layer spiral coil, and a rectangular frame coil. Appropriate placement of ferrite materials on these induction coils successfully eliminated the proximity effect, increased the heating rate, and improved temperature uniformity. © 2013 Elsevier Ltd. All rights reserved.
1. Introduction In the injection molding process, mold temperature is an important factor in the quality of injection molding process. A high mold temperature setting is feasible to enhance the molding quality but increases the cooling time and parts' demolding temperature. To enable cavity filling under a high mold temperature and parts' ejection under low temperature, a variotherm process was developed for high-precision injection molding, thin-walled injection molding and micro-injection molding [1–3]. The variotherm process dynamically controls the mold temperature, preheats the mold to a higher temperature before the melt enters the mold cavities, and then decreases the mold temperature to the ejection temperature after the cavities are filled. Common mold heating approaches for a variotherm process include infrared heating [4,5], gas-assisted heating [6], thin-film resistance heating [7], resistance cartridge heating [8,9], steam heating [10], and induction heating [11–14]. The advantages of induction heating, in which only the mold surface is heated, include rapid heating and cooling, good controllability, energy saving, and allowing local heating. Compared with conventional oil heating, induction heating has better potential for increasing mold temperature and reducing cycle time. The induction heating method has been widely used in the various industrial manufacturing processes. Since the skin effect of ☆ Communicated by W.J. Minkowycz. ⁎ Corresponding author at: Department of Power Mechanical Engineering, National Taitung College, 889 Jhengci N. Rd., Taitung City 95045, Taiwan, ROC. 0735-1933/$ – see front matter © 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.icheatmasstransfer.2013.11.017
induction heating concentrates the resulting Joule heating at the surface of processed workpiece, the surface can be heated by using a high-frequency current. Therefore, induction heating is an efficient variotherm process for injection molding. Studies show that induction heating rapidly increases mold surface temperature, substantially shortens cycle time [11], increases the molding ability of thin-walled injection molding and microinjection molding [12], improves the replication of micro-features [13], and reduces the welding line defect of injection molding parts [14]. Induction coil designs can be classified into four types: unmovable external hung-up type, movable external hung-up type, wrapped type, and inserted type. The first type is easily implemented because the induction coil is independent of the processed workpiece. Chen [15] used induction heating to increase the surface temperature of injection molds and found that both heating rate and uniformity are satisfactory when the area of the processed workpiece is 100 mm × 100 mm. The movable external type can be either a fixed induction coil with a movable processed workpiece or a fixed processed workpiece with a movable induction coil. The former is more common, particularly in metal welding or surface treatment [16–18]. In wrapped induction heating, induction coils are used to embed the whole workpiece. Chen [19] reported that wrapped induction heating is effective for injection molding. In the inserted type, the induction coils are placed behind cores and cavities so that the penetration effect and heat conduction provide indirect heating of the mold surface. This technique is superior in terms of reduced cycle time because the induction coils do
S.-C. Nian et al. / International Communications in Heat and Mass Transfer 50 (2014) 52–60
not have to be moved. However, the placement of induction coils inside the molds requires consideration of the layout of the cooling channel, the layout of the ejection system, and the rigidity of the mold structure. Induction heating is apparently an effective solution for heating the mold surface without increasing the molding cycle period. However, the main unresolved problem of using induction heating in injection molding is the non-uniform temperature distribution on the heating surface. The various causes of differences in heating temperature include the coil turn space, heating distance, cavity geometry, magnetic interference, and the proximity effect. For instance, different directions of electric current along induction coils can cause a non-uniform temperature distribution by introducing a repulsive proximity effect. For a practical coil design, however, opposite directions of coil current are difficult to avoid. In a study of the effect of repulsive proximity on non-uniform temperature distribution, Sung et al. [20] compared temperature uniformity in the heating surfaces of different induction heating coils. The experimental results showed poor heating efficiency in the singlelayer coil with opposite current directions. However, the double-layer reciprocating coil and the coil with magnetic flux concentrator efficiently increased heating speed and provided uniform temperature distributions. Huang [21] designed a multi-layer coil for improving heating efficiency and temperature uniformity in a conventional single-layer coil. The experimental results showed that the multi-layer coil has a more uniform temperature distribution and better heating efficiency compared to the conventional single-layer coil. Although, the double-layer reciprocating coil can avoid the heating layer (i.e., the layer near the heating surface) with opposite current direction, the double-layer coil must have reduplicate coil length of single-layer reciprocating coil. However, the increased coil length increases coil manufacturing cost and energy consumption. The local multi-layer coil has a more uniform temperature distribution and better heating efficiency; applications of the local multi-layer coil are limited
53
Fig. 2. Magnetic flux fields of two adjacent opposite current coils: (a) with magnetic concentrator; (b) with magnetic shielding material.
to spiral-like forms. Applying magnetic flux concentrator to control magnetic flux field and eliminate proximity effect is the most popular method in the induction heating process, but fails to completely isolate the conflicting magnetic fluxes when opposite current coils were induced.
Fig. 1. Effect of current direction on magnetic flux line: (a) coils with identical current direction; (b) coils with opposite current direction.
Fig. 3. Dimensions of the heated plate with cooling channels.
54
S.-C. Nian et al. / International Communications in Heat and Mass Transfer 50 (2014) 52–60
Fig. 4. Dimensions of experimental coils and the heating area: (a) single-layer reciprocated coil; (b) single-layer spiral coil; (c) rectangular frame coil.
2. Induction heating principles Induction heating of metal is performed by electromagnetic induction. According to the Faraday law and the Lenz law, passing an electrical alternating current through heating coils produces an alternating magnetic field that can be used to heat an object. If processed magnetic or non-magnetic conductive workpieces are placed in the alternating magnetic field established by the heating coils, cutting-of-flux produces Table 1 Magnetic properties of ferrite materials. Properties
Initial permeability, μi
Relative permeability, μr
Resistivity
Ni–Zn ferrite Mn–Zn ferrite
1500 2300
1.19 × 108 1.83 × 108
107 Ω-m 8 Ω-m
current at different depths. The workpieces' resistance and the flow of eddy currents therein generate heating power [22]. Induction heating exploits two effects, the Joule effect, which is based on hysteresis loss and eddy current loss, and the electromagnetic effect, which mainly comprises the skin effect and the proximity effect. These effects are described further below. 2.1. Skin effect Passing an alternating current through a conductive wire generates a magnetic field both inside and outside the conductor. Varying the current then causes a variation in the magnetic field in the conductor, which generates an induction electromotive force and, thus, a current in the conductor. The current in the conducting wire section is distributed as follows: the current near the center of the conducting wire is very
S.-C. Nian et al. / International Communications in Heat and Mass Transfer 50 (2014) 52–60
55
Fig. 5. Ferrite materials' location and heating temperature distribution of reciprocated single-layer coil: (a) without ferrite slices; (b) without ferrite slices after 10 s heating; (c) without ferrite slices after 20 s heating; (d) with ferrite slices; (e) with ferrite slices after 10 s heating; (f) with ferrite slices after 20 s heating.
small whereas that at the surface is relatively large. The cause of the difference is known as the skin effect. In induction heating, a highfrequency current passing through the coil produces the largest induction eddy current at the workpiece surface. The eddy current declines exponentially with distance from the surface.
2.2. Proximity effect The proximity effect can be observed when conductive wires are placed in the vicinity of each other. The magnetic fields of the two conductive wires affect each other, and the associated magnetic field on the mold surface changes [23]. Fig. 1 shows the effect of current direction on magnetic flux line. When two coils have an identical current direction, the magnetic flux lines are combined throughout the mold surface, which enables uniform heating of the mold surface. However, in coils with opposite current directions, the magnetic flux lines are separated and driven away from the center of the coils. Therefore, excluding the magnetic field substantially decreases induction heating efficiency and temperature uniformity.
3. Methodology This study proposes a novel magnetic shielding induction heating method to solve the repulsive proximity problem. By using ferrite materials to separate the conflict magnetic fields to eliminate the influences of repulsive proximity effect, the heating efficiency and temperature uniformity were thus enhanced. Fig. 2 shows the magnetic flux fields of two adjacent opposite current coils combined with a magnetic concentrator and magnetic shielding material. Whereas the magnetic concentrator can separate the magnetic fluxes between the magnetic concentrators and concentrate the magnetic flux below the coils, the magnetic concentrator cannot avoid the proximity effect that occurs under the coils. Thus, the center of the workpiece has a lower heating efficiency and a lower temperature. In contrast with the magnetic concentrator method, the proposed magnetic shielding method completely separates the magnetic flux fields from different coils and
2.3. Magnetic concentrator A magnetic concentrator has material properties resembling those of a transformer coil, i.e., high conductivity and low resistance. It is used in the induction heating process for local enhancement of the magnetic field, which thus improves the heating effect. When the processed workpiece is close to the induction coil, the proximity effect changes the current distribution. The current is distributed along the coil surface, which causes a loose distribution of eddy current on the processed workpiece. Thus, the surface heating effect is insufficient. Use of a magnetic concentrator can enforce the coil current and eddy current distribution, which then enhances its heating rate [23].
Fig. 6. Dimensions of 9-point measurement.
56
S.-C. Nian et al. / International Communications in Heat and Mass Transfer 50 (2014) 52–60
4.2. Experimental coil design This experiment used three typical single-layer coils, the reciprocated single-layer coil, the single-layer spiral coil, and the rectangular frame coil, to perform a series of induction heating experiments. Fig. 4 shows the coil designs and heating area (shaded area). The coils were composed of 5 mm diameter copper tube, and the heating distance from the heated-plate surface to the coil was 15 mm in all three coils. Fig. 4(a) shows that the reciprocated single-layer coil has a 25 mm pitch and a heating area of 75 mm × 75 mm. Fig. 4(b) shows that the single-layer spiral coil has a 15 mm pitch and a heating area of 75 mm × 75 mm. Fig. 4(c) shows the rectangular frame coil around a 121 mm long × 76 mm wide rectangular area. The heating area was imitated with the front cover of a 5-inch smartphone. 4.3. Magnetic shielding material In the magnetic shielding induction heating method, ferrite materials are used to separate the conflicting magnetic fields. The normal ferrite materials mainly have two types, the Mn–Zn and the Ni–Zn ferrites. Both ferrite materials are widely used in transformer coils, and are easily magnetized and demagnetized. Table 1 shows the magnetic properties of the Mn–Zn and Ni–Zn ferrites. The Ni–Zn ferrite was selected for this experiment. The Ni–Zn ferrite has much higher resistivity (108 Ω-m) compared to the Mn–Zn ferrite (8 Ω-m). The higher resistivity is advantageous to avoid the ferrite to be heated during the induction heating process, and therefore has less extra energy consumption. 4.4. Setup for magnetic shielding induction heating
Fig. 7. Temperature profile of single-layer reciprocated coil: (a) without ferrite slices; (b) with ferrite slices.
drives the magnetic flux uniformly throughout the workpiece surface. This study performed a series of induction heating experiments in three typical single-layer coils: the reciprocated single-layer coil, the single-layer spiral coil, and the rectangular frame coil. Both heating efficiency and temperature uniformity were compared.
4. Experimental setup 4.1. Experimental tools A series of induction heating experiments were performed to evaluate the use of ferrite materials in the proposed magnetic shielding induction heating method. A normal mold steel AISI-P20 with dimensions of 170 mm long × 170 mm wide × 20 mm thick was heated by a high frequency induced heating machine (HP-25KW, Honor, Taiwan). Fig. 3 shows the geometry and cooling channel (diameter: 8 mm) dimensions of the heated plate. The mold temperature controller was used to pass 60 °C heat-transfer oil through the cooling channels during all heating/cooling process. A thermometer and infrared ray thermal imaging system (ThermoVision A20, FLIR, USA) were also used to record the temperature variation of the heated-plate surface.
Figs. 5(d), 8(d), and 10(d) show the locations of the ferrite materials relative to the three experimental coils. The ferrites were placed either at the corners of the coil or in the middle between the copper-tube conductors with opposite current directions. Fig. 5(d) illustrates the locations of the ferrite materials in the reciprocated single-layer coil. The coil current directions of the vertical conductors were opposite to the adjacent others, and a repulsive proximity effect was induced between the vertical conductors during the induction heating process. Five ferrite slices with 4 mm thickness were used in the reciprocated single-layer coil. Three ferrite slices were inserted into the middle between each adjacent vertical conductor to eliminate the proximity effect. Additionally, two ferrite slices were placed at the outside of coil right/left boundaries to limit the effect area of magnetic flux. Fig. 8(d) shows the locations of ferrite materials in the single-layer spiral coil. The circling coil current direction in the spiral coil induced a repulsive proximity effect at the spiral core. Therefore, the center of the heated surface had a lower temperature during the induction heating process [22]. A 4 mm diameter ferrite cylinder was inserted into the center of the spiral core to eliminate the proximity effect. Additionally, a 100 mm × 100 mm square ferrite slice (thickness: 4 mm) was placed over the spiral coil with 5 mm distance from coil surface to ferrite surface, to increase heating speed. Fig. 10(d) shows the locations of the five ferrite slices used in the rectangular frame coil. Four ferrite slices were placed at the corners of the coil to increase the heating efficiency of the corner area, and a ferrite slice was placed at the middle of the coil inlet and outlet, where the copper-tube conductor had an opposite current direction. 5. Results and discussion 5.1. Experimental results for reciprocated single-layer coil The reciprocated single-layer coil shows a large repulsive proximity effect between the adjacent copper-tube conductors with opposite current direction. In this experiment, five ferrite slices were used to improve the heating efficiency of the reciprocated single-layer coil.
S.-C. Nian et al. / International Communications in Heat and Mass Transfer 50 (2014) 52–60
57
Fig. 8. Ferrite materials' location and heating temperature distribution of single-layer spiral coil: (a) without ferrite materials; (b) without ferrite materials, after 10 s heating; (c) without ferrite materials, after 20 s heating (d) with a ferrite cylinder; (e) with a ferrite cylinder, after 10 s heating; (f) with a ferrite cylinder, after 20 s heating; (g) with a ferrite cylinder and covered with a ferrite slice; (h) with a ferrite cylinder and covered with a ferrite slice, after 10 s heating; (i) with a ferrite cylinder and covered with a ferrite slice, after 20 s heating.
Fig. 5 shows the thermal images of mold surface captured by the thermometer and infrared ray thermal imaging system after the mold surface was heated for 10 s and 20 s by reciprocated single-layer coil with/without ferrite slices. Fig. 5(b) and (c) shows that repulsive proximity effect substantially limits the heating effect of the reciprocated single-layer coil on the mold surface. The maximum temperature inside the heating area (inside black lines) increased only 5 and 10 °C after 10 and 20 s heating, respectively. Specifically, the center area shows a very small surface temperature increase. Fig. 5(e) and (f) shows that the ferrite slices effectively solve the heating problem in the reciprocated single-layer coil. The maximum temperature inside the heating area increased from 60 to 125 and 158 °C after 10 and 20 s heating. The temperature distribution shows that the temperature of the mold surface under the copper tube was
higher than that of the mold surface in contact with the ferrite slices, which resulted from the area of the mold surface in contact with the ferrite slices which also shows a proximity effect (Fig. 2(b)). The temperature profiles of the heated area during the conduction heating process were obtained by a nine-point measurement in experiments performed in a reciprocated single-layer coil and a single-layer spiral coil. Figs. 6 and 7 show temperature profiles obtained by the 9point measurement at points P4, P5, P7, and P8 during conduction heating. The experimental results show that the coil without ferrite had a small temperature variation (Fig. 7(a)) that the coil with ferrite slices had a faster heating speed at beginning 5 s, and that the speed of increase in surface temperature slowed after 20 s (Fig. 7(b)). The experiment also confirms that the high resistivity (108 Ω-m) of Ni–Zn ferrite can avoid the ferrite slices to be heated by the induction heating coils. Therefore, the ferrite slices can maintain at a low temperature.
58
S.-C. Nian et al. / International Communications in Heat and Mass Transfer 50 (2014) 52–60
proximity effect and substantially increase the heating speed of the reciprocated single-layer coils. 5.2. Experimental results for the single-layer spiral coil The core of the single-layer spiral coil shows a severe repulsive proximity effect. This study compared three ferrite arrangements in terms of their effects on spiral coil heating, including a conventional spiral coil (without ferrite material) (Fig. 8(a)), a spiral coil center inserted into a ferrite cylinder (diameter: 4 mm) (Fig. 8(d)), and a spiral coil inserted into a ferrite cylinder and covering a ferrite slice (thickness: 4 mm) (Fig. 8(g)). Fig. 8 shows thermal images of the mold surface after 10 and 20 s of heating by the spiral coils. The maximum and minimum temperatures of the heated area are shown at the top-right and center of the thermal images. The thermal images of the mold surfaces heated for 10 s show that: (1) inserting a ferrite cylinder into the core of the spiral coil substantially increases core surface temperature from 73 °C (Fig. 8(b)) to 82 °C (Fig. 8(e)) and decreased the temperature deviation from 20 °C (max. 93 °C, min. 73 °C) to 15 °C (max. 97 °C, min. 82 °C). (2) Inserting a ferrite cylinder slightly increased heating speed and increased the maximum temperature from 93 °C (Fig. 8(b)) to 97 °C (Fig. 8(e)); (3) covering a slice of ferrite above the spiral coil substantially increased heating speed, increased the maximum temperature from 97 to 117 °C, and increased the surface core temperature from 82 to 97 °C (see Fig. 8(e) and (h)). Furthermore, when the induction heating time was increased to 20 s, all surface temperatures increased, but the variation in temperature deviation showed a different trend. When the heating time was increased from 10 to 20 s, the temperature deviation in the conventional spiral increased from 20 to 23 °C (Fig. 8(b) and (c)), but the temperature deviation in the spiral coil with ferrite cylinder decreased from 15 to 12 °C (Fig. 8(e) and (f)). In an actual injection molding system with external induction heating, 2 to 5 s are needed to remove the coil from the mold, clamp the mold, and inject the melt into the cavity. During this interval, the heated surface cools. Fig. 9 shows the temperature variation measured at points P4, P5, P7, and P8 (see Fig. 6) relative to heating/cooling time in an experimental plate induction heated for 20 s then allowed to stand for 5 s. The experimental results show that, at the induction heating stage, the heated surface of the spiral coil with ferrite cylinder had a smaller temperature deviation and that the spiral coil with ferrite cylinder and ferrite slice had a faster heating speed and a larger temperature deviation. During the cooling stage, due to the heat transfer effect, the core temperature (at P5) had slower cool-down speed, which enhanced temperature uniformity. The temperature deviation of the spiral coil with cylinder ferrite was decreased by 5 °C after standing for 5 s (Fig. 9(b)). The spiral coil experiment confirmed that inserting a ferrite cylinder into the core of the spiral coil reduced the proximity effect and increased the surface core temperature, whereas covering a ferrite slice above the coil substantially increased the heating speed of the spiral coil. 5.3. Experimental results for the rectangular frame coil
Fig. 9. Temperature profile of single-layer spiral coil: (a) without ferrite materials; (b) with a ferrite cylinder; (c) with a ferrite cylinder and covered with a ferrite slice.
Problems caused by overheating decreased, and energy consumption was reduced. The experimental results for the reciprocated singlelayer coil confirmed that the ferrite slices efficiently eliminate the
The rectangular frame coil showed a large repulsive proximity effect between the coil inlet and outlet area and a low heating efficiency at the four corner areas. In this experiment, inserting five ferrite slices (Fig. 10(d)) effectively solved the above problem of non-uniform heating in the rectangular frame coil. Fig. 10 shows thermal images of the mold surface after heating by the rectangular frame coils for 10 and 20 s. The maximum and minimum temperatures of the heated area are displayed at the top-right of the thermal images. The images show that the ferrite slices efficiently improve temperature uniformity during conventional rectangular frame coil heating. The thermal images of the mold surface heated for 10 s revealed three findings: (1) the large repulsive proximity effect reduced the heating effect of the conventional coil in the area between
S.-C. Nian et al. / International Communications in Heat and Mass Transfer 50 (2014) 52–60
59
Fig. 10. Ferrite materials' location and heating temperature distribution of rectangular frame coil: (a) without ferrite slices; (b) without ferrite slice, after 10 s heating; (c) without ferrite slice, after 20 s heating; (d) with ferrite slices; (e) with ferrite slices, after 10 s heating; (f) with ferrite slices, after 20 s heating.
the inlet and outlet copper-tube conductors. The smallest increase was 2 °C (from 60 to 62 °C) (see Fig. 10(b)). (2) Inserting a ferrite slice between the inlet and outlet copper-tube conductors efficiently eliminated the repulsive proximity effect. The temperature increased from 2 to 9 °C (from 60 to 69 °C) (see Fig. 10(e)), and the temperature deviation decreased from 11 °C (max. 73 °C, min. 62 °C) to 7 °C (max. 76 °C, min. 69 °C). (3) The thermal image of the conventional rectangular frame coil shows a corner retract phenomena at the rectangular frame corners; that is, compared to the coil corner, the heated region offsets to the inside direction of the corner and significantly decreases the heating temperature in the corner area (see Fig. 10(b)). (4) Inserting ferrite slices on the outside of the rectangular frame coil corner efficiently eliminated the corner retract phenomena in the heated surface and withdrew the offset heated areas to the correct positions below the coil corner (see Fig. 10(e)). Further, when the induction heating time was increased from 10 to 20 s, all surface temperatures increased, but temperature deviations only slightly increased; the temperature deviation of the conventional rectangular frame coil increased from 11 °C (max. 73 °C, min. 62 °C) to 13 °C (max. 78 °C, min. 65 °C) (see Fig. 10(b) and (c)), and the temperature deviation of the rectangular frame coil with ferrite slices increased from 7 °C (max. 73 °C, min. 62 °C) to 8 °C (max. 78 °C, min. 65 °C) (see Fig. 10(e) and (f)). 6. Conclusion This study developed a magnetic shielding method that effectively eliminates the repulsive proximity effect during the induction heating process. A high-resistivity Ni–Zn ferrite was used as magnetic shielding material to separate the conflicting magnetic fields. The three typical single-layer coils used in this study included the reciprocated singlelayer coil, the single-layer spiral coil, and the rectangular frame coil. After recording temperature variation at the mold surface with a thermometer and infrared ray thermal imaging system, the recorded thermal images were used to analyze heating efficiency and temperature uniformity. The experimental results confirm that the magnetic shielding
method is a simple and feasible scheme for eliminating the proximity effect, increasing induction heating speed, and increasing the uniformity of temperature distribution. The following conclusions are drawn: (1) The Ni–Zn ferrite is a suitable shielding material for separating opposing magnetic fields. According to the high resistivity property of Ni–Zn ferrite, the Ni–Zn ferrite cannot be heated by induction coil, and thus has less heat influence to the mold surface and extra energy consumption. (2) The conventional reciprocated single-layer coil has a severe repulsive proximity effect between the copper-tube conductors, which reduces heating efficiency. The maximum temperature inside the heating area was increased only 5 °C after 10 s heating. Heating efficiency was improved by inserting ferrite slices between the copper-tube conductors, which increased the maximum temperature of heating area from 60 to 125 °C after 10 s heating. (3) In the conventional spiral coil, the lower temperature at the core area of the mold surface causes a non-uniform temperature distribution. However, inserting a ferrite cylinder into the core substantially improves heating speed and temperature uniformity. Additionally, covering a ferrite slice above the spiral coil increases the heating efficiency of spiral coils. (4) The conventional rectangular frame coil shows a severe repulsive proximity effect between the copper-tube conductors' inlet and outlet area, which reduces heating efficiency. Inserting a ferrite slice between the inlet and outlet conductors efficiently eliminates the repulsive proximity effect and solves the heating difficult problem. (5) A problem of the conventional rectangular frame coil is that the corners of the rectangular frame have retracting phenomena. That is, the heated region compared to the coil corner produces an offset to the inside direction of corner. Inserting ferrite slices on the outside of the rectangular frame coil corner efficiently eliminates the corner retract phenomena and withdraws the offset of heated areas to the correct position below the coil corners.
60
S.-C. Nian et al. / International Communications in Heat and Mass Transfer 50 (2014) 52–60
In summary, all experimental results in this study agree that the magnetic shielding method is feasible and efficient for reducing the repulsive proximity effect, for increasing heating speed, and for increasing temperature uniformity during the induction heating process. Acknowledgment The authors thank the Precision Mold and Die Research and Development Center for providing assistance with the experimental equipments. Ted Knoy is appreciated for his editorial assistance. Reference [1] G. Wang, G. Zhao, H. Li, Y. Guan, Analysis of thermal cycling efficiency and optimal design of heating/cooling systems for rapid heat cycle injection molding process, Mater. Des. 31 (2010) 3426–3441. [2] S.Y. Yang, S.C. Nian, S.T. Huang, Y.-J. Weng, A study on the micro-injection molding of multi-cavity ultra-thin parts, Polym. Adv. Technol. 22 (6) (2011) 773–1082. [3] S.C. Nian, S.Y. Yang, Molding of thin sheets using impact micro-injection molding, Int. Polym. Process. 20 (4) (2005) 441–448. [4] P.C. Chang, S.J. Hwang, Simulation of infrared rapid surface heating for injection molding, Int. J. Heat Mass Transf. 49 (2006) 3846–3854. [5] P.C. Chang, S.J. Hwang, Experimental investigation of infrared rapid surface heating for injection molding, J. Appl. Polym. Sci. 102 (2006) 3704–3713. [6] S.C. Chen, R.D. Chien, S.H. Lind, M.C. Lin, J.A. Chang, Feasibility evaluation of gas-assisted heating for mold surface temperature control during injection molding process, Int. Commun. Heat Mass Transfer 36 (8) (2009) 806–812. [7] D. Yao, B. Kim, Development of rapid heating and cooling systems for injection molding applications, Polym. Eng. Sci. 42 (12) (2002) 2471–2481. [8] G. Wang, G. Zhao, Y. Guan, Research on optimum heating system design for rapid thermal response mold with electric heating based on response surface methodology and particle swarm optimization, J. Appl. Polym. Sci. 119 (2) (2011) 902–921. [9] G. Wang, G. Zhao, Y. Guan, Thermal response of an electric heating rapid heat cycle molding mold and its effect on surface appearance and tensile strength of the molded part, J. Appl. Polym. Sci. 128 (3) (2013) 1339–1352.
[10] M.C. Jeng, S.C. Chen, P.S. Minh, J.A. Chang, C.S. Chung, Rapid mold temperature control in injection molding by using steam heating, Int. Commun. Heat Mass Transfer 37 (9) (2010) 1295–1304. [11] S.C. Chen, Y.W. Lin, R.D. Chien, H.M. Li, Variable mold temperature to improve surface quality of microcellular injection molded parts using induction heating technology, Adv. Polym. Technol. 27 (4) (2008) 224–232. [12] M.-S. Huang, N.-S. Tai, Experimental rapid surface heating by induction for micro-injection molding of light-guided plates, J. Appl. Polym. Sci. 113 (2009) 1345–1354. [13] S.C. Chen, W.R. Jong, Y.J. Chang, J.A. Chang, J.C. Cin, Rapid mold temperature variation for assisting the micro injection of high aspect ratio micro-feature parts using induction heating technology, J. Micromech. Microeng. 16 (2006) 1783–1791. [14] S.C. Chen, W.R. Jong, J.A. Chang, Dynamic mold surface temperature control using induction heating and its effects on the surface appearance of weld line, J. Appl. Polym. Sci. 101 (2006) 1174–1180. [15] S.C. Chen, H.S. Peng, J.A. Chang, W.R. Jong, Simulation and verification of induction heating on a mold plate, Int. Commun. Heat Mass Transfer 31 (7) (2004) 971–980. [16] R. Rudolf, P. Mitschang, M. Neitzel, Induction heating of continuous carbonfibre-reinforced thermoplastics, Compos. A: Appl. Sci. Manuf. 31 (11) (2000) 1191–1202. [17] K.Y. Bae, Y.S. Yang, C.M. Hyun, S.H. Cho, Derivation of simplified formulas to predict deformations of plate in steel forming process with induction heating, Int. J. Mach. Tools Manuf. 48 (15) (2008) 1646–1652. [18] I. Magnabosco, P. Ferro, A. Tiziani, F. Bonollo, Induction heat treatment of a ISO C45 steel bar: experimental and numerical analysis, Comput. Mater. Sci. 35 (2) (2006) 98–106. [19] S.C. Chen, Y.T. Lin, J.A. Chang, Investigation of Mold Heating by Induction Heating With Embedded Coils Within Mold Interior, SPE-ANTEC, Orlando, Florida USA, 2010. 1431–1434. [20] Y.T. Sung, Y.N. Lin, S.J. Hwang, H.H. Lee, D.Y. Huang, Design of Induction Heating Module for Uniform Cavity Surface Heating, SPE-ANTEC, Orlando, Florida USA, 2012. 1427–1436. [21] M.-S. Huang, Y.-L. Huang, Effect of multi-layer induction coils on efficiency and uniformity of surface heating, Int. J. Heat Mass Transf. 53 (2010) 2414– 2423. [22] V. Rudnev, R. Loveless, R. Cook, M. Black, Handbook of Induction Heating, Marcel Dekker, New York, 2003. 99–120. [23] T.J. Ahmed, D. Stavrov, H.E.N. Bersee, A. Beukers, Induction welding of thermoplastic composites—an overview, Compos. Part A 37 (2006) 1638–1651.