Efforts to relocate production to low-wage countries have led to increasingly urgent cost pressures on injection molding industry, forcing processors to implement cost-reducing measures as quickly as possible, such as new production processes, higher integration and supply changes on. Production now also offers considerable room for savings.
One way is cycle time, which can exceed the shortest possible value by more than 40%. Reasons here lie in poor mold thermal design and excessive residual cooling time, which is usually pre-set subjectively based on estimates. Machine operators also often add a safety margin to compensate for process and tooling deviations, such as temperature and control fluctuations, and bring production to a steady state. But this is often at expense of suboptimal cycle times, and so is costly.

 

Cooling not only directly affects process costs, but also has a significant impact on quality, which in turn affects production efficiency. Residual cooling time is now entered as a fixed value and cannot compensate for occasional disturbances in production process, such as fluctuations in process, machine control and material. Result is variation in distortion, dimensional accuracy, surface quality and other part properties during mass production.
To solve such problems, Germany's South-West-phalia University of Technology, Kistler Instruments and well-known processing companies have jointly participated in a research project to develop a system in which demolding point does not depend on a fixed cycle time, but when plastic part reaches a certain thermal state. This step no longer keeps cooling time stable, but thermal properties of mold.
By measuring surface temperature of plastic part and cavity pressure, thermal state of plastic part is determined cycle by cycle in process. Residual cooling time is automatically calculated and sent directly to machine control unit.

Trial and setup times are reduced as it is no longer necessary to estimate and enter residual cooling time;
Significant economical improvement as cooling times and cycle times are kept as short as possible;
Because demolding temperature is stable, quality of plastic part is kept as stable as possible.

Ideal demolding temperature is determined not only by plastic used, but also by shape of part. Because of complexity of shrinking and twisting process, fact that mold wall and melt temperatures cannot be accurately determined in advance, it is impossible to accurately calculate demolding temperature of an injection molded part.
Estimates can only be made based on guide values from raw material manufacturers and experience of molder, because during demolding, with traditional wall thicknesses, temperature throughout section of plastic part will not be completely equal, there will always be a temperature gradient from middle of plastic part to outside.

 

Figure 2: Process conditions measured with a combined cavity pressure and temperature sensor
Therefore, in order to determine ideal demold point, use common cooling time formula. This enables a theoretical calculation of residual cooling time. In order to determine demolding point at a certain temperature, plastic properties, meaning wall temperature and melt temperature, must be known at beginning of cooling process.
However, as mentioned before, last two parameters mentioned are unknown. Of course, melt temperature during injection cycle can be obtained experimentally during injection cycle by means of thermocouples, which must be introduced into mold cavity each cycle. However, this method of temperature determination is not suitable for production and is mainly used for scientific research.
Therefore, in order to automatically calculate residual cooling time in mass production, it is necessary to develop a method to accurately determine mold wall temperature distribution and melt temperature at the beginning of cooling, i.e. after actual mold filling process. This requires application of a combined temperature/pressure sensor (Figure 1).
With this specially designed sensor, temperature measurement is carried out on sensor surface and temperature in contact with melt is recorded as soon as melt reaches sensor. Until this point, sensors accurately record mold temperature and provide necessary information about temperature distribution of mold wall. An integrated cavity pressure sensor probes location of volumetric fill at onset of cooling (Figure 2).
Therefore, all information can be used to automatically solve cooling time formula cycle by cycle through an in-depth algorithm, providing injection molding machine with actual cooling time to start demolding process.

In the interim, method described above was also tested on a large ABS housing part (plastic part weight about 1 kg). Its surface is mirror polished and must be absolutely defect-free. The total cycle time of process was 61 seconds, of which 42 seconds were taken up by cooling time.
Average mold temperature is 62℃. After thorough manual optimization, plastic parts can be made with sufficient quality under a stable cooling time. Reducing cooling time to a fixed value of 40 seconds rotated process to an indeterminate quality state (Table 1). Cracks appeared on the surface of plastic parts during indefinite interruptions (Fig. 3). This is caused by ejector pin because plastic part has not reached correct temperature for ejection point. A 2 second reduction in cooling time necessitates 100% visual inspection of plastic part.
Cooling times between 39.8 and 40.3 seconds can be achieved with automatic cooling time calculations at a stable demolding temperature of 72℃. In this case, cracks do not occur even during long production times. In this way, cooling time is reduced by 5% compared to safety margin.

 

Figure 3: Occasional cracks appear on plastic parts without automatic cooling time calculation
Figure 4: SmartAmp Load Amplifier with Combined Automatic Cooldown Calculation
To check system performance, average mold temperature was raised to 70℃. System now automatically increases cooling time and demolds at same temperature of 72℃. Plastic surface in this case is also flawless. In addition to surface properties, there are more test parameters for plastic parts. If demolding temperatures remain stable, they stay within desired tolerances.
This example shows that for a process that has been manually optimized to limit, automatic calculation of cooling time and possibly the earliest ideal deformation point can lead to further reductions in cycle time, while maintaining usual state of the best quality.

For industrial applications, it is important to integrate this method into an injection molding machine. Method of intervening in process with controlled movements must be integrated into machine in a stable manner. An approved solution was selected to implement this approach. Algorithm is integrated into an industrial load amplifier (Figure 4).