Search

The processing conditions and the mould design are important parts of the powder injection moulding (PIM) process. In the present work, the mould filling phase of the process has been studied in situ with a special mould equipped with a sight glass using a high speed camera. Effects of features such as the mould temperature, the melt temperature, the surface roughness of the mould cavity, the gate dimensions, the mould design and the flow rate on the filling behaviour were demonstrated. In particular it was noted that a rougher mould surface to some extent facilitated the filling of the mould. The mould temperature, the nominal melt temperature and the flow rate affected the position of weld lines via the cooling of the feedstock and the corresponding change of the melt viscosity. The experimentally determined weld-line positions were compared with predictions from numerical simulations performed with the commercial software Moldex3D. The agreement between the experiment and the prediction was in most cases quite fair or good. The in situ observation of the mould filling also provided the possibility to study how the gate dimensions and the flow rate influenced the jetting behaviour.

Powder injection moulding (PIM) is an important and accepted industrial technique for net shaping of precision components which can have a rather complex geometry. In order to meet the imposed, often rather strict, requirements with regard to dimensional accuracy, it is important to have an adequate knowledge and control of the rheological behaviour and the related processing properties of the powder/polymer melt (feedstock). Such a knowledge is furthermore of crucial importance in numerical simulations of the PIM-process. In the present work, a model system, consisting of steel powder, poly(ethylene glycol) and wax, is used in order to illustrate how the viscometric properties as well as thermal properties, such as the conductivity and the specific heat, of the system can be related to the corresponding properties of the polymeric binder system. In a similar way, the pvT (pressure-volume-temperature)-behaviour of the model system is analysed and discussed. The pvT-behaviour, which has not been extensively reported on for PIM-feedstocks, is considered to be of significant relevance for controlling the outcome of the injection moulding process.

In our recent work, it was shown that the rate of filling during injection moulding did not significantly affect the tensile and impact properties of recovered and virgin polypropylene, whereas significant changes in the properties were observed after ageing at 100 °C. The purpose of the present work was to extend the former study to include the effects of processing and ageing on surface-appearance-related properties, in particular the gloss. In-mould rheology and gate-seal analyses were performed in order to select the filling and post-filling processing windows, respectively. The influences of filling rate, holding and cooling times on the gloss measured in the gate area and at the far end of the moulded specimens were evaluated. The mould used had an embossed surface giving test plaques with a matte appearance. The results showed that the gloss varied by about 30 to 40% within the filling window obtained by in-mould rheology. The slight gloss difference between the two PP-grades noted at the lowest injection speed vanished with increasing filling rate. Gloss differences were observed between the gate area and the far end of each processed moulding, again at the lowest injection speed. The materials processed at a high filling rate exhibited a fairly constant gloss as the holding time increased regardless of the position of the analysed region, whereas a more significant decrease was noted in the case of the mouldings processed at the low filling rate. The cooling time had a negligible effect on the gloss in comparison to the effect of filling rate. Ageing at 100 °C for 240 h increased the gloss level, especially at the far end of the moulding, but only for specimens manufactured at the lowest filling rate. Roughness measurements with a profilometer showed that the gloss development during processing and ageing was related to the topographical matching between the developing polymer surface and the mould topography.

The simulation of the advancing flow front during mould filling in the injection moulding cycle is important as a way of anticipating manufacturing defects, particularly different types of welds, air traps, cold spots and hot spots. The purpose of the present work was to evaluate a model (the distance model) simulating the advancing flow front and predicting potential issues related to the progression of the flow front such as welds and air traps. The distance model is based on a mathematical theory of Hele-Shaw flow for strongly shear-thinning fluids, i. e. fluids with a power-law index, n, equal to 0,3 or preferably less. Two grades of general-purpose polystyrene were selected according to their abilities to shear thin (weakly or moderately strong) at the selected processing temperature and in the typical shear rate range of the injection moulding process. The two grades were injection moulded into twelve mould configurations derived from two similar single-gated moulds. The flow length from the gate to the weld was measured in the mouldings obtained at three distinct rates of filling (conventional, slower and faster) and compared to the flow lengths obtained by the distance model. Good agreement was found between the predicted flow lengths and the experimental flow lengths at the conventional and faster filling conditions. Larger deviations between simulation and experiment were found at the slower filling rate, particularly for the weakly shear-thinning polystyrene grade. Some comparisons were also made with the predictions obtained using the commercial simulation code Moldflow. A comparison between the advancing front predictions of the distance model and experimental short shots of a commercial polypropylene grade in a lure-box type of mould geometry showed that the distance model, despite its simplicity, could probably be used to detect welds and air traps in more complex and practice-related mouldings.