2020 Volume 61 Issue 3 Pages 528-533
Effects of reduced pressure and casting design on changes in mold filling due to coat permeability in the expendable pattern casting process of aluminum alloy were investigated experimentally. The effect of coat permeability on the melt velocity of molten aluminum alloy in the expendable pattern casting process was investigated experimentally under the conditions of reduced pressure and top pouring. Aluminum alloy plates were cast under conditions of the reduced pressure and top pouring, using eight kinds of coats with different permeabilities. The melt velocity was measured and the results showed that the difference in the melt velocity was not large depending on the casting design. The application of the reduced pressure condition and use of high permeability coats led to higher melt velocities. However in the high coat permeability region, the melt velocity did not increase much, even when the coat permeability increased. The experimental values of the melt velocity were compared with the calculated values based on the mold filling model used in the previous study. When the coat permeability was low, the experimental values were in relatively good agreement with the calculated values. However in the high coat permeability region, the experimental values of the melt velocity were lower than the calculated values. By considering the heat transfer resistance of the liquid resin at the EPS surface to the mold filling model, even in the high coat permeability region, the experimental values of melt velocity showed more or less good correlation with the calculated values.
This Paper was Originally Published in Japanese in J. JFS 91 (2019) 737–742.
The expendable pattern casting (EPC) process is very attractive particularly for complex-shaped casting, because near net shape castings are obtained and it is possible to eliminate machining steps. Furthermore, because no molding binder is added to the mold in the EPC process, the environmental load can be reduced. Because of the extremely high recycling rate of sand, the cost of packing sand can also be reduced, under situations where silica sand is replaced with artificial sand.1) However in the EPC process, the molten metal is poured into the cavity produced by the thermal decomposition and liquefaction of the expendable polystyrene (EPS) pattern, and the thermal decomposition gas and liquid resin discharge into the dry sand through the coat layer. Therefore, the mold filling mechanism in the EPC process is more complicated than that in the cavity mold process.2) When the coat permeability changes, the thickness of the thermal decomposition gas layer changes. Therefore, it seems that the coat permeability and melt velocity do not have a simple relationship.3,4) In addition, when the pressure in the flask reduces or casting design changes, the mold filling also seems to change owing to a change in the thickness of the thermal decomposition gas layer. The liquid resin produced by molten metal heat increases the possibility of residual resin defects caused by entrapment of the resin in the molten metal.2) Moreover, the heat transfer resistance of the liquid resin reduces the thermal decomposition rate of the EPS pattern. As a result, the liquid resin seems to affect the mold filling. In particular, in the case of aluminum alloy, as the melt temperature is much lower than that of cast iron, the rate of liquefaction of the EPS pattern by the molten metal heat is higher than that of gasification.2) As described above, as the mold filling mechanism in the EPC process is complicated, there are few analysis systems for mold filling that can exactly simulate the EPC process.5) Therefore, the accumulation of experimental data for accurate mold filling analysis is required.
Recently, a relatively large number of investigations have been conducted on the mold filling for the EPC process, such as the study on molten cast iron by Maruyama et al.6,7) However, few studies have been performed on the mold filling mechanism in the EPC process, such as on the quantitative effect of the process parameters on the melt velocity.3,8) In particular, almost no studies on the melt velocities for a wide range of coat permeabilities have been reported.9)
In a previous study,9) an aluminum alloy plate was cast using eight kinds of coats with different permeabilities (Range: 0.13 to 40), and the effect of the coat permeability on the melt velocity was examined. It was found that the use of a high permeability coat led to a high melt velocity. In the high coat permeability region, even when the coat permeability increased, the melt velocity did not increase significantly. The experimental values of the melt velocity were compared with the values calculated based on the mold filling model.3) In the low coat permeability region, the experimental values were in relatively good agreement with the calculated values. However in the high coat permeability region, the experimental values of the melt velocity were lower than the calculated values. However, this discussion of the obtained results is not comprehensive. For a wide range of coat permeabilities, the effects of the casting design and reduced pressure in the flask on the melt velocity have not been examined.
In the present work, the effects of the reduced pressure and casting design on changes in mold filling due to the coat permeability in the EPC process of aluminum alloy were investigated. For eight kinds of coats with different permeabilities, an aluminum alloy plate was cast, and the melt velocity was measured. By considering the heat transfer resistance of the liquid resin at the EPS surface for the mold filling model, even in the high coat permeability region, the possibility of a correlation existing between the experimental values of melt velocity and calculated values was examined.
Figure 1 shows a schematic diagram of the casting apparatus used in the experiments, which was similar to that used in a previous study.9) The steel molding flask is a cylindrical vessel with an inside diameter of 200 mm and depth of 300 mm. The EPS pattern shown in Fig. 1 has a plate shape with a width, height and thickness of 70 mm, 200 mm, and 10 mm, respectively, and a depth direction length of 70 mm. The cluster of the bottom pouring system was assembled with the EPS pattern, as shown in Fig. 1(a). The cluster for the top pouring system using the EPS pattern with the same size was also assembled, as shown in Fig. 1(b).
Schematic diagram of casting apparatus for measurement of mold filling. (a) Bottom pouring, (b) Top pouring.
Two kinds of EPS patterns with expansion ratios of 30 times (density = 29 kg·m−3) and 60 times (density = 18 kg·m−3) were used. A ceramic tube (inside diameter, outside diameter, and length of 25 mm, 35 mm, and 300 mm, respectively) was used as a sprue from the pouring basin to the runner. The pouring basin is an insulating sand mold with an inner diameter of 80 mm, depth of 80 mm, and thickness of 15 mm. The runner is a plate EPS with a cross section of 35 × 10 mm and a length of 125 mm. The cross section of the ingate is 35 × 10 mm. A double sided adhesive tape (base: paper, adhesive: acrylic, thickness: 0.09 mm) was used to connect the runner and EPS pattern. It was attached only to the circumference in the cross section of the ingate.
Silica sand with a mode diameter of approximately 0.15 mm (AFS grain fineness number 62) was poured into the flask.
Eight kinds of coats shown in Table 1 consisting of one mica base coat (coat A) and seven kinds of silica base coats (coats B–H) were used in this study. The values of coat permeability, listed in Table 1, are in the 0.13–45 range, and are in accordance with the JIS mold permeability (JIS Z2603).10) These values were the measured values reported in a previous study,9) and were calculated using the following eq. (1).
| \begin{equation} K = \frac{\delta v}{s\Delta P} \end{equation} | (1) |
The aluminum alloy JIS AC2A (A319 equivalent) was used as a casting material. It was melted in a high frequency electric induction furnace and cast directly from the furnace. The pouring temperature was set as approximately 1073 K. During the pouring process, the distance from the bottom of the pouring basin to the melt surface was maintained at approximately 50 mm. The pouring operation was continued, until the height of the melt surface in the basin did not changed, due to either the complete mold filling or melt flow stop. As a result, the melt head at the melt surface during mold filling has almost the same value in each experiment. The melt head at the melt surface during mold filling decreases from approximately 350 mmAl to 150 mmAl with the bottom pouring, and increases from approximately 150 mmAl to 350 mmAl with the top pouring. An atmospheric pressure and reduced pressure of 13.3 kPa (differential pressure with atmospheric pressure) were applied as pressure conditions in the flask during the pouring process and mold filling. The pressure in the flask was reduced by aspiration from a suction port on the flask side. A plastic film was used to cover the top of the flask and maintain the reduced pressure in the flask.
In order to measure the arrival time of the molten metal in the flow direction distance, five touch sensors6,7) of the molten metal were inserted into the EPS pattern at 10, 55, 100, 145, and 190 mm from the ingate, as shown in Figs. 2(a) and 2(b) for the bottom pouring and top pouring, respectively. The touch sensors are tungsten wires with diameters of 0.5 mm. These wires passed through the center of the 70 mm wide EPS pattern. As shown in Fig. 2(c), the voltage across the resistance increases stepwise, every time the molten metal contacted a touch sensor. The arrival time was defined as the voltage rising time in Fig. 2(c).
Schematic diagram of touch sensor of molten metal. (a) Bottom pouring, (b) Top pouring, (c) Output voltage.
Figure 3 shows an example of the experimental results of the mold filling for the arrival time of the molten metal ta without reduced pressure for an EPS pattern expansion ratio of 60 times and a coat permeability K = 1.7 with the casting design as a parameter. The arrival times of the molten metal in Fig. 3 are shown as a function of the distance from the ingate at the pattern edge y, and were obtained by the output from the touch censors of molten metal. At y = 190 mm, there is almost no difference in the experimental value of the arrival time corresponding to the difference in the casting design. In the middle region of the mold filling distance, the arrival times for the top pouring are slightly longer than those for the bottom pouring. The reason for this may be as follows. For both the bottom pouring (a) and top pouring (b) shown in Fig. 4, as the melt head increases, the discharge rate of the decomposition gas between the melt and EPS pattern increases. On the other hand, as the melt head increases, the gas layer thickness decreases conversely. Thus, with an increase in the melt head, the heat flux between the melt surface and EPS pattern, the thermal decomposition rate, and the melt velocity all increase. As the molten metal fills, the melt head gradually decreases for the bottom pouring, and gradually increases for the top pouring. From these results, differences in the arrival times in the middle region of the mold filling due to differences in the casting design, seems to occur. Therefore, for the melt velocity in the following figures, the average velocity, which was calculated from the arrival time difference at 10 mm and 190 mm, was used.
Effect of casting design on arrival time without reduced pressure for coat permeability 1.7 and EPS expansion ratio of 60 times.
Mold filling model with thermal decomposition of EPS pattern. (a) Bottom pouring, (b) Top pouring.
Figure 5 shows the arrival time of the molten metal ta for a bottom pouring, which was similar to that of Fig. 3, with the degree of reduced pressure as a parameter, for an EPS pattern expansion ratio of 60 times and a coat permeability K = 1.7. As shown in Fig. 5, the arrival time with the condition of the reduced pressure is shorter than that of non-reduced pressure. With the reduced pressure condition, the thermal decomposition gas thickness in the schematic diagram of Fig. 4, decreases due to the increase in the discharge rate of the gas. As a result, it seems that the increase in the heat flux from the molten metal to the EPS pattern enhances the thermal decomposition of the EPS pattern.13)
Effect of reduced pressure on arrival time of bottom pouring for coat permeability 1.7 and EPS expansion ratio of 60 times.
The solid and broken lines shown in Figs. 3 and 4 represent the calculated values based on the mold filling model from a previous study.3) A summary of this mold filling model is given as follows. As the heat transfer from the melt surface to the EPS pattern through the decomposition gas layer is due to the radiation and heat conduction, the heat flux q can be expressed as follows:
| \begin{equation} q = \cfrac{5.67}{\cfrac{1}{\varepsilon_{a}} + \cfrac{1}{\varepsilon_{m}} - 1}\left[\left(\frac{T_{a}}{100}\right)^{4} {}- \left(\frac{T_{m}}{100}\right)^{4}\right] + \frac{\lambda}{\delta}(T_{a} - T_{m}) \end{equation} | (2) |
Figures 6 and 7 show the experimental values of the melt velocity u, as a function of the coat permeability K, with the casting design as a parameter, for non-reduced pressure condition and EPS pattern expansion ratios of 30 times and 60 times. The range of the coat permeability K is 0.13–45, which indicates that the largest K value is approximately 350 times the smallest value. Therefore, the horizontal axis is shown on a logarithmic scale. As shown in Figs. 6 and 7, the use of a high permeability coat or high EPS pattern expansion ratio led to a higher melt velocity. With increasing coat permeability or EPS pattern expansion ratio, the thickness of the thermal decomposition gas layer shown in the schematic diagram of Fig. 4 decreases. As a result, the thermal decomposition of the EPS pattern seems to be enhanced owing to the increase in the heat transfer from the molten metal to the EPS pattern. As observed for a bottom pouring in a previous study,9) when the coat permeability is less than approximately 2 (K < 2), even for the top pouring, the melt velocity increases relatively monotonously with increasing coat permeability. However, when the coat permeability is larger than approximately 2 (K > 2), the melt velocity does not increase significantly with increasing coat permeability. The difference in the melt velocity due to casting design is not so significant. In the schematic diagram shown in Fig. 4(b), for a top pouring, when the thickness of the gas layer that existed under the molten metal is sufficiently thin, the gas seems to hardly rises owing the buoyancy in the molten metal or between the molten metal and coat. However, with a coat permeability K = 0.13, there is the tendency for the melt velocity of the top pouring to be slightly larger than that of the bottom pouring. In the case of the coat low permeability, because of the thick gas layer, the gas rises easily owing the buoyancy in the melt or between the melt and coat.2) Therefore, it seems that a decrease in the gas layer thickness controls the decrease in the heat flux from the molten metal to the EPS pattern. This tendency is somewhat remarkable especially with an EPS pattern expansion ratio of 60 times.
Effects of coat permeability K and casting design on melt velocity u for EPS expansion ratio of 30 times.
Effects of coat permeability K and casting design on melt velocity u for EPS expansion ratio of 60 times.
The solid lines in Figs. 6 and 7 represent the values calculated based on the mold filling model from a previous study. When the coat permeability is less than approximately 2 (K < 2), the experimental and calculated values of the melt velocity are found to be in relatively good agreement, except with the small coat permeability of K = 0.13. However, when the coat permeability is larger than approximately 2 (K > 2), as the experimental melt velocity does not increase significantly with increasing coat permeability as described above, the experimental values of melt velocity are smaller than the calculated values.
Figures 8 and 9 show the experimental values of the melt velocity u as a function of the coat permeability K, with the degree of reduced pressure as a parameter, for the bottom pouring and EPS pattern expansion ratios of 30 times and 60 times. As shown in Figs. 8 and 9, the use of a reduced pressure condition led to higher melt velocities for both the EPS pattern expansion ratios of 30 times and 60 times. With the reduced pressure condition, as the thickness of the thermal decomposition gas layer decreases, it seems that the increase in the heat flux from the molten metal to the EPS pattern enhances the thermal decomposition of the EPS pattern. Similar to the results for the non-reduced pressure condition shown in Figs. 6 and 7, with increasing coat permeability, the melt velocity increases. When the coat permeability is less than approximately 2 (K < 2), the melt velocity increases relatively monotonously with increasing coat permeability. However, when the coat permeability is larger than approximately 2 (K > 2), the rate of increase in the melt velocity decreases. The experimental values were compared with the calculated values, which are represented by a solid line. Similar to the results for the non-reduced pressure condition shown in Figs. 6 and 7, when the coat permeability is less than approximately 2 (K < 2), the experimental values of the melt velocity are in relatively good agreement with the calculated values. However, when the coat permeability is larger than approximately 2 (K > 2), the experimental values of the melt velocity are smaller than the calculated values.
Effects of coat permeability K and reduced pressure on melt velocity u for EPS expansion ratio of 30 times.
Effects of coat permeability K and reduced pressure on melt velocity u for EPS expansion ratio of 60 times.
In the cases shown in Figs. 6, 7, 8, and 9, in the high coat permeability region, as the increase in the experimental melt velocity due to the increase in the coat permeability decreases, the experimental values of the melt velocity are smaller than the calculated values. Based on the mold filling model described by eq. (2) considered in this work, with increasing coat permeability, the thickness of the thermal decomposition gas layer decreases, and the heat flux due to the heat conduction described by the second term in eq. (2) increases. As a result, even in the high coat permeability region, the melt velocity increases monotonously with increasing coat permeability. As shown in Figs. 6, 7, 8, and 9, the calculated values of the melt velocity in the high permeability region are different from the experimental results. As the cause for this, the heat transfer resistance of the liquid resin at the surface of the EPS pattern as shown in Fig. 10 was considered. In the low coat permeability region, as the thickness of the thermal decomposition gas layer is thick, the heat transfer resistance at the gas layer is more controlled than that at the liquid resin. However, in the high coat permeability region, which has a thin gas layer, the heat transfer resistance of the liquid resin at the EPS pattern surface is more controlled. Thus, the heat flux seems to asymptotically approach a finite value. In the high coat permeability region of Figs. 6, 7, 8, and 9, there is the tendency for the experimental melt velocity to asymptotically approach a constant value.
Heat transfer model with heat transfer resistance of liquid resin.
Therefore, the heat flux due to the heat conduction described by the second term in eq. (2) was changed to that obtained by summing the heat conduction resistances at the gas layer and at the liquid resin. As a result, the heat flux q can be expressed using the heat transfer coefficient at the liquid resin hf, as follows:
| \begin{align} q &= \cfrac{5.67}{\cfrac{1}{\varepsilon_{a}} + \cfrac{1}{\varepsilon_{m}} - 1}\left[\left(\frac{T_{a}}{100}\right)^{4} {}- \left(\frac{T_{m}}{100}\right)^{4}\right] \\ &\quad + \cfrac{1}{\cfrac{\delta}{\lambda} + \cfrac{1}{h_{f}}}(T_{a} - T_{m}) \end{align} | (3) |
In order to examine the effect of the reduced pressure and casting design on the mold filling changes due to the coat permeability in the EPC process of aluminum alloy, aluminum alloy plates were cast using eight kinds of coats with different permeabilities. Furthermore, the melt velocity was measured. The following conclusions were obtained under the conditions of this work.
