“Plastics are going digital” – Material characterisation using the example of pvT data

The motivation

Moulded plastic parts are generally designed specifically for their place of use and the corresponding application. The focus is often strongly on the installation situation in the assembly and the loads that prevail during use. However, too little attention is often paid to an elementary “intermediate step” between development and application – production in the injection moulding process! And this is precisely where material characterisation comes in for the creation of a software-specific material data set for process simulation. In order to optimise the plastic moulded part for the manufacturing process and avoid unnecessary correction grinding and adjustments to the subsequent injection moulding tool, a process simulation/injection moulding simulation must be carried out as accurately as possible, but this can only be promising if the material data has been determined conscientiously. The well-known guiding principle applies: The simulation results can only be as good as your input. By conscientiously and precisely analysing the entire production process at an early stage, even during the development phase, potential errors can be avoided, saving both time and money.

This aspect can be illustrated using the rule of ten for error costs (see Figure 1). It states that the later an error is discovered, the more cost-intensive it becomes as the project progresses – by a factor of 10!

Process control in the injection moulding process

The pvT behaviour (pressure-volume-temperature) of plastics as a description of the specific volume as a function of pressure and temperature effect plays an important role in processing (see Figure 2) in the injection moulding process:

A-B (filling phase): In this phase, the pressure is increased at a constant temperature, causing the specific volume to decrease.

B-C (holding pressure phase): The specific volume decreases at constant pressure and decreasing temperature. At the end of this phase, at point C, the sealing point is reached and the sprue gate solidifies.

C-D (holding and cooling section): The melt solidifies completely and the specific volume remains constant with decreasing pressure until the ambient pressure is reached.

D-E (component shrinkage): In this phase, the component shrinks. At point E, the demoulding temperature is reached and the moulded part can be ejected

pvT measurement method: Isobaric and isothermal

In terms of measurement technology, pvT data can be recorded in two different ways (see Figure 3):

Isobar: The pressure on the sample is kept constant while the melt is cooled at a constant rate. Once the lowest temperature value has been reached, the next pressure stage is approached.

Isothermal: The temperature is kept constant while the pressure on the sample is gradually increased. Once all pressure levels have been reached, the next temperature level is approached.

Looking back at Figure 2, it can be seen that only isobaric process control can map the reality of the injection moulding process and that isothermal determination is far away from the process (see Figure 4). In addition, as with any metrological acquisition of data, it is not only the measurement method that is decisive, but also the sampling rate and generation of data points. A glance at the majority of available material data sets quickly shows that the isobaric/isothermal classification is often missing and the number of data points is very low.

Comparison of isobaric and isothermal pvT data of a KEBATER PBT BF130

The following diagram shows isobaric pvT data in cooling at 10 K/min, isothermally measured and “commercially available reduced” isothermally measured pvT data for the KEBATER PBT BF130. The difference between the measurement methods can be seen in the specific volume. Faster cooling of the isobaric measurement leads to the formation of fewer crystals, which in turn results in a lower density and a higher specific volume. Slow cooling produces more crystals, resulting in a denser system and a smaller specific volume.

The measurement duration of the pvT data differs massively:
Isobaric [13h], isothermal [48h], isothermal market standard reduced [4h].

Influence of the pvT data in the simulation

The description of the material behaviour using pvT data is crucial for injection moulding simulation, as the pvT data shows how the volume of the material behaves at different pressure levels and temperatures in the cavity. The influence on the filling behaviour and also on shrinkage/distortion is therefore clear.

To illustrate the differences, four simulations were carried out, each with different pvT data sets (cf.
pvT data sets (see Figure 4) and one validated data set:

1. validated material data set
2. pvT isobaric
3. pvT isothermal
4. pvt market-reduced isothermal

Influence on the printing behaviour

The overestimation of the pressure in variant 4 (standard reduced isothermal) is clearly visible, whereas all three other simulation results show almost identical results. It can be seen that a reduction in the sampling rate in the pvT measurement can lead to a potential error in the pressure behaviour of the simulation result.

Influence on the holding pressure behaviour / shrinkage results

As pvT data provides information on how the volume of a plastic changes at different temperatures and pressures, this information is crucial for correctly mapping the shrinkage behaviour and the holding pressure phase in the injection moulding simulation. As shown in Figure 6, the increased pressure requirement (see Figure 5) leads to an underestimation of the warpage behaviour in the simulation result of the standard reduced data set. Here too, the other simulation results behave in the same way as the validated data set.

Conclusion

Based on the results presented, it is clear that precise recording and application of the pvT data in the injection moulding simulation is of crucial importance. Looking back at the rule of ten for error costs (see Figure 1), it becomes clear that an injection mould built under incorrect simulation assumptions could result in considerable modification costs. These costs and time delays are disproportionate to a conscientious material data characterisation, in particular the process-related isobaric determination of the pvT data.