Advertisement

Lightweight Design worldwide

, Volume 10, Issue 3, pp 40–45 | Cite as

New Mold Concept for Fiber-Plastic Composite Components

  • Felix Parschick
  • Michael Hübner
  • Marcus Feige
  • Markus Oettel
Production Reswer Project
  • 298 Downloads

In the ResWer project a new mold design with innovative tempering was developed and implemented by means of advanced molding and casting processes to obtain reproducible fiber plastic composite parts more quickly, less expensively and at higher quality. New casting alloys were processed, and additive manufacturing was used as well.

Limited natural resources and ever-increasing energy prices demand more efficiency in the life cycle of products manufactured using molds, as well as in the manufacturing of the mold itself. There is still room to increase resource efficiency in mold-making, from mold design and production up to actual use. In addition to constantly refining and optimizing the techniques used in manufacturing, an efficient utilization of energy and resources above all entails implementing new manufacturing methods.

Components made of fiber-reinforced plastics are generated in a sophisticated technique combining process engineering and shaping. Most of these components are still made manually. This procedure is not conducive to large-batch production. However, intelligent manufacturing and die (mold) concepts extend the suitability of these techniques for larger scale production. Optimal mold and die tempering is essential for economical industrial production of laminates with demanding properties.

The die temperature is an essential parameter in the manufacturing of fiber reinforced plastic parts. Tempering of the cavity is crucial when fiber composite components with a large-surface area have to be produced. Tempering in a narrow temperature range is required both in the feeding and curing stages. Tempering involves both heating and the intentional cooling down of the cavity as well. The mold is conventionally heated electrically, in an inductive way or through additional media. The cycle times can be reduced by an active cooling, using water and nitrogen. The channels required are conventionally drilled or milled in the mold. In these techniques, one disadvantage is the relatively large and variable distance of the heating-cooling channels to the die contour, which negatively affects both manufacturing process and product, above all at critical points in the die, the so-called hot spots.

Objective

The goal was to develop new dies (molds) for the manufacturing of large-area fiber composite parts with an innovative tempering system, combining the geometrically free design of casted molds with additive manufacturing techniques. The task was to engineer technologies and manufacturing process chains for a new type of lamination molds suitable for large-batch production, based on pre-cast, near-net shape preforms. The results regarding design and technology were to be implemented in practice in a mold demonstrator for the automotive industry. In the project, the composite mold demonstrator had a size of approximately 1400 mm x 200 mm and was developed to produce a benchmark part for electric mobility application. High speed and high performance cutting strategies and parameters were intentionally chosen for the mold materials and specific mold regions and issues, such as cavity, mold structure, or tempering.

Methodology

The target values for innovative mold tempering were defined according to the analyzed specifications, and the mold inserts were designs by 3D-CAD. Unlike the compact mold design typically used, the project utilized a shell design. Figure 1 elucidates the 3D-CAD model of the unmachined part after casting simulation, optimized, with ejectors and sensors. The walls were made as thin as possible for casting techniques. As a result, the mold structure was suitable for the expected stresses and included large cavities, which, in turn, allowed efficient tempering of the mold.
Figure 1

3D-CAD model of the demonstrator mold after casting without the runner system (© Direkt Form Projektgesellschaft mbH)

Based on the 3D-CAD files both the temperature characteristics and the flow inside the mold inserts were simulated. The cavity in the mold represents the structure which the tempering medium then streams through.

Multiple variants or cycles were developed in subsequent iterations, Figure 2, aimed at the most uniform flow and temperature distribution possible. A mold surface temperature of 80 °C was assumed. Coolant of 20 °C was introduced into the mold. Several variants evinced a very heterogeneous temperature distribution in the upper mold region; as a result, the heat removal in this region would be insufficient. When installing the mold overhead theoretically — whereby punch and die are inverted — the temperature distribution to be seen in simulation was clearly enhanced as a result of natural convection. All variants evidenced a low drop in pressure, at only about 0,3 bar.
Figure 2

Three variations of the cooling system in the demonstrator mold (1–3 from top), E = Inlet and A = Outlet (© Fraunhofer IWU)

To avoid the dead water zone as in variant 1, another configuration of connecting elements was developed, mainly with a centric inlet and two lateral outlets. Variant 3 shows a significantly more homogeneous temperature distribution along the mold’s longitudinal axis (without dead water zones), which results in a more uniform surface temperature.

Manufacturing of the Demonstrator Mold

Selecting the casting alloy plays a key role when transferring the results obtained through theoretical considerations (simulation) into practice (demonstrator mold). Stainless steels, such as 1.4008, are typical mold materials due to their high resistance to strength and corrosion. Aluminum alloys are also widely in use thanks to their sound thermal conductivity. Both materials have a relatively high thermal expansion that can cause problems when the component is ejected from the mold. Especially for large and complex workpieces, the result can be inaccurate parts.

To minimize this, tests with INVAR 36 (1.3912) were performed. INVAR 36 (1.3912), made of 64 % iron and 36 % nickel, is acceptable in terms of thermal conductivity; its thermal expansion is about seven times lower than that of steel. However, when using this alloy rather than conventional stainless steel, tool life is significantly lower.

The experimental INVAR 36 castings made over the course of the project provided very good results on a laboratory scale. When casting INVAR 36, some special requirements have to be considered, such as the molten alloy’s strong affinity towards gas absorption, which can cause pores during solidification. The alloy is also not hard and strong enough, giving rise to deformations in the manufacturing process — above all in large-surface area molds.

The material was subjected to many investigations as well as casting tests to guarantee the optimal casting quality of the functional surfaces by optimizing the melting and pouring technologies. After successful casting tests, in principle, there is no obstacle to using the INVAR 36 material for cast molds or other castings for which low thermal expansion is desired.

Tempering involves both heating and the intentional cooling down of the cavity as well.

Nevertheless, steel 1.4008 was chosen for the demonstrator mold. This decision was made because casting parameters based on experience were available for this material, allowing for successful pouring even in the case of a complex component like the demonstrator mold. Furthermore, the results achieved can also be directly compared in terms of tempering with other molds, allowing for direct conclusions regarding the efficiency of the new mold concept. As an outlook for further projects the use of INVAR is also planned, from which further positive effects can be expected in the mold. However, ongoing parameters must first be investigated regarding the material’s reliable usability in studies.

Solidification in the casting mold was simulated, based on this material and the desired casting geometry. The simulation was performed in several iterations. As a result, a casting system — consisting of gate, filter, run, feeder — was developed, reducing the risk of casting defects in the mold to a minimum.

Afterwards, the casting mold gets designed in 3D-CAD, based on the casting model including ingate system. The mold consists of three main elements summarized in Figure 3: drag box, core and cope box. After mold assembly, the casting material was poured. After cooling down, the casting was extracted from the mold and shot-blasted, after which the elements of the abovementioned casting systems were cut off. In the next step, the grinding, inspecting and welding operations were performed. The component was subsequently tempered to achieve the required material properties. The last step is finishing, shown in Figure 4. Very little material had to be machined thanks to the near-net-shape geometry that this casting technology makes feasible.
Figure 3

Machined casting model of the demonstrator mold (left: cope box; center: drag box; right: core) (© Direkt Form Projektgesellschaft mbH)

Figure 4

inish machining of the demonstrator mold (© ZVZ-Composite Werkzeug- und Vorrichtungsbau GmbH)

Implementation in a Near-series Test Mold

The process chain developed here was demonstrated for a specific industrial test mold by using the results obtained with the demonstrator mold. The ifc Composite GmbH company, which was the final customer as well as an experienced partner, was committed to providing a real component shape and the associated mold model. The large-surface fiber composite “housing shell“, Figure 5 (component is colored red) is a part of the stator housing in advanced wind generators. Two mold halves — lower and upper mold, in between which the intentional component is formed — are required for manufacturing.
Figure 5

3D-CAD model of the component “housing shell” (left) and the first design of the near-series test mold (right) with the component (red) (© Fraunhofer IWU)

The mold was designed and validated with a stress analysis, Figure 6 (top), considering the previously developed process chain. The goal was to combine minimal resources use with maximal performance. As had already been done for the demonstrator mold, the required flow volume of the tempering medium was calculated with a flow simulation, Figure 6 (bottom).
Figure 6

Finite element analysis (top) and computational fluid dynamics (bottom) insight the cope box (left) and the drag box (right) (© Fraunhofer IWU)

Thermal fluid analysis was performed by means of simulation; the flow conditions (course, direction) were analyzed in both mold halves and optimized for heating or cooling down the mold when processing the fiber composite components. In the first design iteration, with piping outside the mold, a shortfall was observed: flow through the mold halves was not as uniform as desired. A redesign was done, and the result obtained was analyzed with a thermal fluid flow simulation again. The goal of the redesign was to optimize the channel cross sections and minimize the occurrence of dead water zones. The final results for both mold halves were significantly better.

Four contour ejectors were integrated in the lower mold for easier component extraction because the component is relatively high and has an arched shape. The function of the ejectors has to be guaranteed over the entire temperature range during the manufacturing process of the fiber composite components. The four ejectors were experimentally made using different technologies for a comparison aimed at identifying the manufacturing technique with the best ratio between tempering performance and manufacturing costs. One cast, one milled and two made by additive manufacturing — laser beam melting — contour ejectors were produced.

For high functional integration, several aspects were considered when designing the cast body of the mold. Material recesses for the ejectors were also considered. The ribbings were designed according to the flow characteristics of the medium, using the simulation results, and the structural elements were designed for casting and the required stiffness.

The blowhole behavior illuminated by simulation, and the cooling simulation as a whole, showed that complete pouring of the part, free of volumetric defects is technically impossible, even if the mold maker has considered all mold structure requirements and has designed the mold in the best possible way for casting. Thus, the challenge for the casting technology was to minimize the unavoidable volumetric defects and to relocate them to non-critical mold sections which are far from finish-machining areas and those with high mechanical stresses.

Near-contour tempering channels were integrated into the design of the contour ejectors to be produced by additive manufacturing. The channels were integrated to test this very efficient tempering solution in a mold. The ejector inserts were made of maraging hot work steel 1.2709 by laser beam melting, with minimal allowance for finishing and fitting in the lower mold.

The experimental INVAR 36 castings made over the course of the project provided very good results on a laboratory scale.

The hybrid tooling approach — casted base body (1.4008) and additively applied functional structure (1.2709) — was investigated during the project at the Fraunhofer IWU. These studies have shown that the materials could be joined by fusion in the transition area, as a result of which 1.4008 and 1.2709 were firmly bonded, Figure 7 (graded transition of the materials). Testing the hybrid processing of casting material 1.4008 and the laser molten tool steel 1.2709 can be considered successful and will be investigated further.
Figure 7

Hybrid manufacturing casting-additiv— cross section of the transition between the casted stainless steel (1.4008) and the laser beam molten maraging hot work steel (1.2709) (© Fraunhofer IWU)

The goal of the redesign was to optimize the channel cross sections and minimize the occurrence of dead water zones.

The industrial test mold was manufactured in the same way (process chain) as developed for the demonstrator. After manufacturing, the mold halves were completed and assembled at the ZVZ tool making company. Machining did not disclose any discontinuities, such as pores or blowholes, which indicates a correctly designed casting mold and process.

Tryout of the Near-series Test Mold

The application test was performed under near-series conditions of use, after successful completion of the test mold. For the test trials, the mold was heated to operational temperature. Heating of the molds demonstrated significant time savings and was sped up by the new tempering system. The mold temperature distribution after heating was recorded by probes and thermal imaging, and documented. The medium had a constant 125 °C temperature, which also resulted in a uniform temperature distribution in the mold as a whole, thanks to the innovative tempering sytem.

Figure 8 shows the setup of the application test under real conditions: the mold is fixed in the forming press with a stack of fiberglass mats, impregnated by a mixture of resin and hardener. The stack is pressed in the mold afterwards, and the resin-hardener mixture cures exothermally.
Figure 8

Set up of the near-series test mold in the application trial with a stack of impregnated fiberglass mats (© IFC Composite GmbH)

Due to the homogeneous tempering system the fiberglass-reinforced plastics (GFRP) parts showed extremely high surface quality. The temperature was homogeneously distributed across the component despite its strongly varying cross sections. The innovative mold concept also made it possible to reduce the wave-like layer structure that was typical under previous manufacturing conditions, Figure 9. This way, the component structure becomes more homogeneous, and the properties of the GFRP component are affected positively. The automated pressing procedure by the mold ensures very high reproducibility of the component properties, which is a highly significant advantage in comparison with manufacture-like production.
Figure 9

Comparison of the manufactured component “housing shell” with the previous manufacturing technique (top) and the new casted mold (bottom) (© IFC Composite GmbH)

Notes

Thanks

This research and development project was funded by the German Federal Ministry of Education and Research (BMBF) within the Funding Action “SME — Innovative: Resources and Energy Efficiency” (funding number 02PK2430 — 02PK2433) and managed by the Project Management Agency Karlsruhe (PTKA). The author is responsible for the contents of this publication.

Copyright information

© Springer Fachmedien Wiesbaden 2017

Authors and Affiliations

  • Felix Parschick
    • 1
  • Michael Hübner
    • 2
  • Marcus Feige
    • 3
  • Markus Oettel
    • 4
  1. 1.Direkt Form Projektgesellschaft mbHDeutschland
  2. 2.Edelstahlwerke Schmees GmbHDeutschland
  3. 3.ZVZ-Composite GmbHDeutschland
  4. 4.Fraunhofer Institute for Machine Tools and Forming TechnologyDeutschland

Personalised recommendations