**3. Results and Discussions**

#### **3. Results and Discussions**  *3.1. Effect of Part Thickness on the Mold Temperature Control*

*3.1. Effect of Part Thickness on the Mold Temperature Control*  In injection molding, the part geometry is an important element that not only impacts on the part formation but also the molding parameters. With a thin product, there are many methods for improving the melt filling by molding at high temperature, such as induction heating and heater heating. Of these methods, gas-assisted mold temper-In injection molding, the part geometry is an important element that not only impacts on the part formation but also the molding parameters. With a thin product, there are many methods for improving the melt filling by molding at high temperature, such as induction heating and heater heating. Of these methods, gas-assisted mold temperature control has shown many positive results [20–23]. With gas-assisted mold temperature control, the structure of the stamp insert is often used to increase the heating efficiency. Based on the results of these research works, the stamp thickness is one of the most important parameters of mold design and is impacted by the part thickness. Therefore, in this paper,

to estimate the heating ability of In-GMTC, an insert with a size of 77.4 mm × 77 mm was inserted into the cavity plate, and the heating process was achieved with a hot gas temperature of 300 ◦C; the gap between the gas gate and the heating surface was 3.5 mm (Figure 5c). By simulation with the model, as shown in Figure 6, the results of the variation in the mold temperature (at sensor S1—Figure 5) versus time for a heating time of 20 s are described as shown in Figure 8. For an initial mold temperature of 40 ◦C, it can be seen that the In-GMTC heated the plate to above 170.4 ◦C. This shows that the In-GMTC can support a heating rate of 6.5 ◦C/s. With the change in part thickness from 0.1 to 5.0 mm, the temperature at sensor S1 varied from 164.3 to 170.4 ◦C. This temperature range was higher than the glass transition temperature of almost all common plastic material. In former papers on gas-assisted mold temperature control [20–22], when the GMTC was applied for a heating area of 58 mm × 30 mm, a gas flow rate of 500 L/min and a gas temperature of 300 ◦C, the maximum heating rate was only about 2.2 ◦C/s [25]. This means that the In-GMTC, with the system design and heating process shown in Figure 2, has a great advantage in terms of the heating efficiency of the mold surface. size of 77.4 mm × 77 mm was inserted into the cavity plate, and the heating process was achieved with a hot gas temperature of 300 °C; the gap between the gas gate and the heating surface was 3.5 mm (Figure 5c). By simulation with the model, as shown in Figure 6, the results of the variation in the mold temperature (at sensor S1—Figure 5) versus time for a heating time of 20 s are described as shown in Figure 8. For an initial mold temperature of 40 °C, it can be seen that the In-GMTC heated the plate to above 170.4 °C. This shows that the In-GMTC can support a heating rate of 6.5 °C/s. With the change in part thickness from 0.1 to 5.0 mm, the temperature at sensor S1 varied from 164.3 to 170.4 °C. This temperature range was higher than the glass transition temperature of almost all common plastic material. In former papers on gas-assisted mold temperature control [20–22], when the GMTC was applied for a heating area of 58 mm × 30 mm, a gas flow rate of 500 L/min and a gas temperature of 300 °C, the maximum heating rate was only about 2.2 °C/s [25]. This means that the In-GMTC, with the system design and heating process shown in Figure 2, has a great advantage in terms of the heating efficiency of the mold surface.

ature control has shown many positive results [20–23]. With gas-assisted mold temperature control, the structure of the stamp insert is often used to increase the heating efficiency. Based on the results of these research works, the stamp thickness is one of the most important parameters of mold design and is impacted by the part thickness. Therefore, in this paper, to estimate the heating ability of In-GMTC, an insert with a

*Polymers* **2021**, *13*, 1004 10 of 19

**Figure 8.** Temperature comparison at the center of the heating area (Point S1) in simulations with different product thicknesses. **Figure 8.** Temperature comparison at the center of the heating area (Point S1) in simulations with different product thicknesses.

To observe the influence of part thickness on the heating step, five stamp thicknesses were used for the experiment. Figure 8 shows that with thicknesses of 0.1, 0.2, 0.3, 0.4, and 0.5 mm, the temperature at sensor S1 varied from 164.3 to 170.4 °C; this means that the heating rate increased from 6.2 to 6.5 °C/s. The increase in heating rate when the part thickness increased can be explained by the thermal energy needed to heat the stamp volume. Because the stamp and the mold plate were separated by an isolation layer, with the same heating source, the heating result mainly depended on the stamp volume. So, with a thinner stamp, more thermal energy is needed to increase the stamp temperature. Based on the design shown in Figure 5c, because the stamp To observe the influence of part thickness on the heating step, five stamp thicknesses were used for the experiment. Figure 8 shows that with thicknesses of 0.1, 0.2, 0.3, 0.4, and 0.5 mm, the temperature at sensor S1 varied from 164.3 to 170.4 ◦C; this means that the heating rate increased from 6.2 to 6.5 ◦C/s. The increase in heating rate when the part thickness increased can be explained by the thermal energy needed to heat the stamp volume. Because the stamp and the mold plate were separated by an isolation layer, with the same heating source, the heating result mainly depended on the stamp volume. So, with a thinner stamp, more thermal energy is needed to increase the stamp temperature. Based on the design shown in Figure 5c, because the stamp thickness and gap between the gas gate to the heating surface were 5 and 3.5 mm, respectively, with a thicker part, the less material there was at the cavity; therefore, a better heating rate was achieved with a thicker product. However, the difference in the heating rate when the part thickness changed from 0.1 to 0.5 mm was small. Therefore, in thin wall injection molding, this property is an advantage of this heating method, which could support mold temperature control with a thin wall part and a thickness that could be varied to lower than 0.5 mm.

The temperature curves in Figure 8 show that the heating rate was extremely high in the first 2 s, with the heating rate varying between 23.5 and 24.5 ◦C/s. This heating rate is higher than those of many heating methods reported in recent years [19–25]. After the first 2 s, the heating rate decreased. Although the temperature of the heating surface was not the same as the cavity surface (which was measured) as in Figure 5c, the temperature change was almost the same as those shown in other research on gas-assisted mold temperature control, which can be explained by the absorption of thermal energy and heat transfer. In this research, the heating surface absorbed the thermal energy, and the thickness at sensor S1 was thin, meaning that this thermal energy was held within a small material volume, resulting in a rapid temperature increase at sensor S1. This phenomenon was clearly apparent in the first 2 s. However, the higher temperature meant that the thermal energy at the thin cavity location transferred to another area with a lower temperature. So, this phenomenon lowered the heating rate at the cavity surface. In addition, with a heating time of 20 s, the temperature curves did not show the same limitation as in other studies [24,25]. Therefore, the temperature at the cavity surface still increased with a longer heating time or a high-powered heating source. However, as mentioned above, the temperature at 20 s was high enough to facilitate the melt flowing, and so based on this simulation, the heating time of 20 s was used for the following cases.

In our former study, when the internal GMTC was used for mold surface heating, there was a temperature difference between the inlet and outlet area [20–22]. Therefore, in this research, to evaluate the uniformity of the heating process under various stamp thicknesses, the temperature distribution at the cavity surface of the stamp, as shown in Figure 5a, was measured and compared using a simulation and experiments. Figure 9 shows the simulation result regarding the temperature distribution of the cavity. The temperatures of five sensors were measured at the end of the heating step and are compared in Figure 10. These results show that the highest temperature was located at the top of the stamp (sensor S1), which was closest to the hot gas gate, and the temperature was lower at the bottom of the stamp. This kind of distribution was better than that of previous research on the internal GMTC, which often found unbalanced temperatures between the two sides of the cavity area [20,22]. In addition, compared with the induction heating method [14,16,17], the In-GMTC method solved the problem regarding the low temperature at the center of the heating area and would therefore be better for application in real molding products. This result also shows that the temperature differences between the five sensor locations were 81.3, 81.1, 80.5, 79.1, and 78.2 ◦C with product thicknesses of 0.1, 0.2, 0.3, 0.4, and 0.5 mm, respectively. These results also prove that, with a heating time of 20 s, for all types of stamp thickness, the temperature of the cavity varied from over 83.8 to approximately 164.5 ◦C and the higher temperature was close to the melt entrance, which could lead to a greater reduction in the frozen layer. Thus, the pressure drop of the hot melt was limited and the melt could flow faster. This means that this temperature distribution was suitable for use in the injection molding process. In addition, these results also show that the lowest temperature could almost satisfy the mold temperature of common plastic materials and the highest temperature was not so high that degradation of plastic materials would occur.

**Figure 9.** Temperature distribution of the molding area with an initial temperature of 40 °C, a **Figure 9.** Temperature distribution of the molding area with an initial temperature of 40 ◦C, a gas temperature of 300 ◦C, an inlet gas pressure of 7 bars, and a heating time of 20 s. *Polymers* **2021**, *13*, 1004 13 of 19

gas temperature of 300 °C, an inlet gas pressure of 7 bars, and a heating time of 20 s.

**Figure 10.** Temperature along the center line in simulations with different product thicknesses. line of the stamp were 54.2, 79.1, and 85 °C, with inlet temperatures of 200, 300, and **Figure 10.** Temperature along the center line in simulations with different product thicknesses.

*3.2. Effect of the Inlet Temperature on the Heating Process* 

s.

Figure 5 and Figure 6 with gas temperatures of 200, 300, and 400 °C, a product thickness of 0.5 mm corresponding to the stamp thickness of 1 mm and a heating time of 20

Figures 11 and 12 show the temperature distribution under different air temperatures with a stamp thickness of 1 mm. This result shows that with the higher inlet temperature, the heating process became more effective, resulting in a higher temperature at the center of the plate, as well as the temperature difference on the plate increasing. In detail, based on the simulation result, the temperature differences along the center

In the GMTC method, the gas temperature is an important element, which can be

To verify the accuracy of the simulation result, the experiment achieved the same boundary condition as that in the simulation. The experiment was performed 10 times for each case; after that, the average value was represented for each case. Then, the temperature at the sensors was measured and compared with the simulation result, as shown in Figure 10. The comparison shows that the temperature difference between the simulation and experiment was lower than 10 ◦C. This difference is due to the measurement delay of the sensor, especially as, in this state, the thermal energy transferred quickly from the higher temperature area to the lower temperature area. However, in general, this result shows that the results of the simulation and the experiment have a good agreement.

#### *3.2. Effect of the Inlet Temperature on the Heating Process*

In the GMTC method, the gas temperature is an important element, which can be represented by the heating source. In the application of GMTC, a higher gas temperature causes the heating rate to increase. However, the amount of wasted energy will also increase. Therefore, in this study, this was investigated with the model shown in Figures 5 and 6 with gas temperatures of 200, 300, and 400 ◦C, a product thickness of 0.5 mm corresponding to the stamp thickness of 1 mm and a heating time of 20 s.

Figures 11 and 12 show the temperature distribution under different air temperatures with a stamp thickness of 1 mm. This result shows that with the higher inlet temperature, the heating process became more effective, resulting in a higher temperature at the center of the plate, as well as the temperature difference on the plate increasing. In detail, based on the simulation result, the temperature differences along the center line of the stamp were 54.2, 79.1, and 85 ◦C, with inlet temperatures of 200, 300, and 400 ◦C, respectively. In this paper, a comparison was performed between three different inlet temperatures, as shown in Figure 13a. To verify the accuracy of the simulation result, the experiment was performed with the same gas gap. The temperature at five sensors was collected and compared, as shown in Figure 13b–d, and the temperature distribution is shown in Figure 12. The comparation shows that the simulation and the experiment have a good agreement. In other research on the GMTC [20–25], the heating process was influenced by another wild source; however, with the In-GMTC, the heating was achieved in the private volume, meaning that the simulation and the experimental result exhibited good agreement. Therefore, this heating method could be easier to predict by simulation than the external GMTC. Figure 13b–d shows that the mean temperature in the experiment was slightly lower than the simulation result. This was due to the fact that the simulation results show the temperature at the end of heating step exactly; however, in the experiment, there was a delay time associated with the thermal camera obtaining the thermal picture, and in this delay time, the thermal energy at the higher temperature area was transferred to the lower temperature area, resulting in a lower temperature being obtained by the thermal camera. *Polymers* **2021**, *13*, 1004 14 of 19 400 °C, respectively. In this paper, a comparison was performed between three different inlet temperatures, as shown in Figure 13a. To verify the accuracy of the simulation result, the experiment was performed with the same gas gap. The temperature at five sensors was collected and compared, as shown in Figure 13b–d, and the temperature distribution is shown in Figure 12. The comparation shows that the simulation and the experiment have a good agreement. In other research on the GMTC [20–25], the heating process was influenced by another wild source; however, with the In-GMTC, the heating was achieved in the private volume, meaning that the simulation and the experimental result exhibited good agreement. Therefore, this heating method could be easier to predict by simulation than the external GMTC. Figure 13b–d shows that the mean temperature in the experiment was slightly lower than the simulation result. This was due to the fact that the simulation results show the temperature at the end of heating step exactly; however, in the experiment, there was a delay time associated with the thermal camera obtaining the thermal picture, and in this delay time, the thermal energy at the higher temperature area was transferred to the lower temperature area, resulting in a lower temperature being obtained by the thermal camera.

**Figure 11.** Temperature distribution of the stamp under different air inlet temperatures with a **Figure 11.** Temperature distribution of the stamp under different air inlet temperatures with a heating time of 20 s.

heating time of 20 s.

**Figure 12.** Temperature distribution of the cavity plate after 20 s of heating under different hot gas temperatures.

(**d**)**.**

heating time of 20 s.

400 °C, respectively. In this paper, a comparison was performed between three different inlet temperatures, as shown in Figure 13a. To verify the accuracy of the simulation result, the experiment was performed with the same gas gap. The temperature at five sensors was collected and compared, as shown in Figure 13b–d, and the temperature distribution is shown in Figure 12. The comparation shows that the simulation and the experiment have a good agreement. In other research on the GMTC [20–25], the heating process was influenced by another wild source; however, with the In-GMTC, the heating was achieved in the private volume, meaning that the simulation and the experimental result exhibited good agreement. Therefore, this heating method could be easier to predict by simulation than the external GMTC. Figure 13b–d shows that the mean temperature in the experiment was slightly lower than the simulation result. This was due to the fact that the simulation results show the temperature at the end of heating step exactly; however, in the experiment, there was a delay time associated with the thermal camera obtaining the thermal picture, and in this delay time, the thermal energy at the higher temperature area was transferred to the lower temperature area,

resulting in a lower temperature being obtained by the thermal camera.

**Figure 11.** Temperature distribution of the stamp under different air inlet temperatures with a

**Figure 12. Figure 12.** Temperature distribution of the cavity plate after Temperature distribution of the cavity plate after 20 s of heating under different hot gas temperatures. 20 s of heating under different hot gas temperatures. *Polymers* **2021**, *13*, 1004 15 of 19

*(\*) Si: Simulation ; Ex: Experiment*

**Figure 13.** The comparison of temperatures along the center line by simulation (**a**) with different inlet temperatures with a heating time of 20 s and stamp thickness of 0.5 mm under gas temperature of 200 °C (**b**); 300 °C (**c**) and 400 °C **Figure 13.** The comparison of temperatures along the center line by simulation (**a**) with different inlet temperatures with a heating time of 20 s and stamp thickness of 0.5 mm under gas temperature of 200 ◦C (**b**); 300 ◦C (**c**) and 400 ◦C (**d**)**.**

#### *3.3. Improve the Melt Flow Length of the Polyamide 6 Thermoplastic Composites by Internal 3.3. Improve the Melt Flow Length of the Polyamide 6 Thermoplastic Composites by Internal Gas Heating for the Gate Temperature Control*

*Gas Heating for the Gate Temperature Control*  To verify the efficiency of In-GMTC for use in the mold temperature control, the mold of melt flow length testing was used for experiment. The dimension of cavity is shown in Figure 5a. The melt flow thickness varied from 0.1 to 0.5 mm. The injection molding experiment was carried out with PA6/GF30% plastic and the molding parameters are shown in Table 3. For the common injection molding cycle, the mold temperature should be set in the range of 20–80 °C; however, with the thin wall product as in this case, to fill the cavity, the mold temperature must be set as high as the system can tolerate. This set up allows for easy flow due to the reduction in the freeze layer of the melt flow [3]. However, when the mold temperature is high, energy is inevitably wasted, and other problems occur such as warpage or flashing. To avoid these prob-To verify the efficiency of In-GMTC for use in the mold temperature control, the mold of melt flow length testing was used for experiment. The dimension of cavity is shown in Figure 5a. The melt flow thickness varied from 0.1 to 0.5 mm. The injection molding experiment was carried out with PA6/GF30% plastic and the molding parameters are shown in Table 3. For the common injection molding cycle, the mold temperature should be set in the range of 20–80 ◦C; however, with the thin wall product as in this case, to fill the cavity, the mold temperature must be set as high as the system can tolerate. This set up allows for easy flow due to the reduction in the freeze layer of the melt flow [3]. However, when the mold temperature is high, energy is inevitably wasted, and other problems occur such as warpage or flashing. To avoid these problems, local mold temperature control is presented in this paper. Instead of maintaining the entirety of the mold plate at the high

lems, local mold temperature control is presented in this paper. Instead of maintaining the entirety of the mold plate at the high temperature, local mold temperature control

the melt flow when it passed the area [10]. Figure 7 shows the cavity plate with an insert, which included the cavity area and the gate area. In the same manner as the above structure, the gate area was re-designed with a steel insert to improve the heating efficiency. This insert had a dimension of 77 mm × 77.4 mm × 5 mm. To observe the influence of gas temperature on the heating process, gas with temperatures of 200, 250, 300, 350, and 400 °C were used with a heating time of 20 s. To verify the heating efficiency as well as the ability to perform local heating, an infrared camera was used to obtain the temperature distribution at the end of the heating step. After that, the real

temperature, local mold temperature control was achieved for the cavity area by applying local air pre-heating at the beginning of the molding cycle. The high temperature at the gate area reduced the pressure drop of the melt flow when it passed the area [10]. Figure 7 shows the cavity plate with an insert, which included the cavity area and the gate area. In the same manner as the above structure, the gate area was re-designed with a steel insert to improve the heating efficiency. This insert had a dimension of 77 mm × 77.4 mm × 5 mm. To observe the influence of gas temperature on the heating process, gas with temperatures of 200, 250, 300, 350, and 400 ◦C were used with a heating time of 20 s. To verify the heating efficiency as well as the ability to perform local heating, an infrared camera was used to obtain the temperature distribution at the end of the heating step. After that, the real molding cycle was achieved with the parameters shown in Table 3. For each gas temperature, the molding cycle was operated for 20 cycles to stabilize all of the systems; then, the product of the next 10 cycles was collected to compare the melt flow lengths. Figure 14 shows the temperature distribution and the melt flow pattern with different gas temperatures. The gate temperature and melt flow length were measured. They are shown in Table 3. The temperature distribution shows that the high temperature was focused only at the gate area, which was heated by the hot gas for 20 s. Therefore, the mold plate was maintained at the low temperature in all molding cycles, which reduced the warpage and flashing, as well as the amount of energy wasted when compared with the common case. In the case without heating, the melt flow pattern shows that the melt length was 17 mm when the gate temperature was only about 50 ◦C. With the In-GMTC, when the gas temperature increased from 200 to 400 ◦C, the gate temperature varied from 50 to 216 ◦C with a heating time of 20 s. In other research, when the mold temperature was higher than the glass transition temperature, the melt flowed easily [16–22]. Thus, in this paper, Figure 14 shows that when the gas temperature was higher than 350 ◦C, the cavity was completely filled. In addition, these results for a 0.5 mm part thickness also prove that the In-GMTC leads to a large improvement in the melt flow length for the thin wall injection molding product—an increased melt flow length from 36.9% to 100% (fully filled) was observed when the mold plate was maintained at approximately a common temperature. The filling ability of PA6/GF30% for other part thicknesses was compared, as shown in Figure 15. In general, these results show that the In-GMTC method has a positive influence on the flow ability of PA6/30%GF, which was verified with flow with part thicknesses of 0.1–0.5 mm.



peratures.

peratures.

**Figure 14.** The melt flow length of a 0.5 mm part thickness with and without In-GMTC under different hot gas tem-**Figure 14.** The melt flow length of a 0.5 mm part thickness with and without In-GMTC under different hot gas temperatures. **Figure 14.** The melt flow length of a 0.5 mm part thickness with and without In-GMTC under different hot gas tem-

.

**Figure 15.** The melt flow length and gate temperature under different gas temperatures for In-**Figure 15.** The melt flow length and gate temperature under different gas temperatures for In-**Figure 15.** The melt flow length and gate temperature under different gas temperatures for In-GMTC.

#### GMTC. GMTC. **4. Conclusions**

**4. Conclusions**  In this study, an internal gas-assisted mold temperature control (In-GMTC) under different flow thicknesses (t) (0.5, 1.0, 1.5, and 2.0 mm) and a gas temperature that varied from 200 to 400 °C was applied to achieve rapid mold surface temperature control. Then, the In-GMTC was applied to verify the melt flow length of PA6/GF30% material in the thin wall injection molding cycle. Based on the results, the following conclusions **4. Conclusions**  In this study, an internal gas-assisted mold temperature control (In-GMTC) under different flow thicknesses (t) (0.5, 1.0, 1.5, and 2.0 mm) and a gas temperature that varied from 200 to 400 °C was applied to achieve rapid mold surface temperature control. Then, the In-GMTC was applied to verify the melt flow length of PA6/GF30% material in the thin wall injection molding cycle. Based on the results, the following conclusions In this study, an internal gas-assisted mold temperature control (In-GMTC) under different flow thicknesses (t) (0.5, 1.0, 1.5, and 2.0 mm) and a gas temperature that varied from 200 to 400 ◦C was applied to achieve rapid mold surface temperature control. Then, the In-GMTC was applied to verify the melt flow length of PA6/GF30% material in the thin wall injection molding cycle. Based on the results, the following conclusions were obtained:


difference at the cavity surface. The temperature differences along the center line


**Author Contributions:** Conceptualization, T.T.D. and P.S.M.; funding acquisition, T.T.D. and P.S.M.; investigation, P.S.M. and T.M.T.U.; project administration, T.T.D. and P.S.M.; supervision, T.T.D., and P.S.M.; visualization, T.T.D., P.S.M., and T.M.T.U.; writing—original draft, P.S.M. and T.M.T.U.; writing—review and editing, T.T.D. and P.S.M. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the Ministry of Education and Training, project Grant No. B2020-SPK-01, and hosted by Ho Chi Minh City University of Technology and Education, Vietnam.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The data used to support the findings of this study are available from the corresponding author upon request.

**Acknowledgments:** The authors thank Bui Duc Duan, Nguyen Hoai Nam, Hoang Thien Bao, Tran Van Duy, Vo Son Lam, Hoang Le Minh Quan, and Nguyen Huynh Nam Huy for their assistance with and discussions surrounding the experiment and simulation.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**

