Molding filled resins is a different ball game

For all their engineering beauty and usefulness, injection molds have a number of drawbacks: They’re costly, difficult to alter once made, and you can’t see what’s going on inside while they fill.

Computer software makers get around this problem with systems such as Moldflow, and the result has generally been fewer expensive mistakes and faster mold turnaround by fewer design iterations. What computer software doesn’t do, however, is explain some of the subtleties of what happens inside the mold, especially for complex materials.

Take glass-filled versus neat thermoplastics, for example. A simple container mold in, say, polyethylene, may be designed with a draft angle between one-eighth and one degree to aid part ejection, but in cases where draft isn’t allowable on the finished part, it’s usually possible to engineer a stripper system for efficient ejection.

What happens if the material is switched to a heavily filled resin? One possibility is fracture of the container when the stripper momentarily distorts the part to “pop” it out of the mold.

Another issue commonly encountered around glass-filled polymers is fibre orientation. Near the mold surface, fibres tend to orient along the direction of flow, while in the bulk of the part, they’re aligned across the flow direction. Parts designed for high tensile strength in one direction may have radically different (and lower) capability in another. If a high degree of randomness in fibre orientation is desirable to give the part similar mechanical properties in all directions, higher melt temperatures are needed. That’s fine for the part designer, but heat in has to be taken out in the mold. Production would rather cut melt temperature to the minimum to shorten cycle times. The result is a classic quality/productivity compromise. And how about increased mold cooling to get fast cycle time with hotter melts? Unfortunately, visible weld lines, bubbles, or an undesirable matte finish in the molded parts can result from indiscriminate mold cooling.

The dynamics of melt flows contribute to another annoying consequence of injection molding: shrinkage. Shrinkage would be easier to control if it occurred evenly, but it doesn’t, and the resulting “anisotropy” is a major cause of mold redesign and rework. For typical commodity resins, shrinkage is greatest along the flow path. In the transverse direction, values in the range of 70 to 95 percent of the linear shrinkage are common. The effect varies with gate size and part thickness, with thinner parts being much less susceptible to the effect.

Now consider the same commodity polymer, only glass-filled. Now the shrinkage rule works the other way around, with shrinkage in the flow direction smaller than transverse shrinkage. And the difference between the two can be as high as fifty percent for the glass-filled case. With a difference that great, differential shrinkage must also be controlled by the use of fan or disk gating, along with reduced mold and melt temperatures, higher injection pressures and faster mold filling.

Shrinkage is a common enough problem, but fibre reinforcement can cause other troubles as well. The fibres are meant to add strength to the finished part, but to be effective, the fibres must be completely “wetted” by the resin matrix. Ideally, each individual fibre would be equally spaced, with an identical “stress field” around each one, allowing every fibre and its surrounding matrix to share applied loads evenly. Parts with sharp corners, however, not only introduce stress risers, but can also cause the fibres to gather at the corners, creating regions which are far too rich in reinforcing material. Since the secret of composite materials is in the interface between matrix and reinforcement, excessive fibre density in some regions of the part create local areas of lower strength. And in the “sharp corner” case, that lack of mechanical properties occurs right where stresses are most likely to concentrate. One possible “solution” could be lower aspect ratio, plate-like reinforcement, but this usually trades away overall mechanical properties, and definitely works against tensile strength in parts subjected to loads along a single, preferred direction. It’s hard to make a good drawbar out of talcum powder. A realistic solution is gentle radii in the part, and the elimination of sharp corners wherever possible. If they’re unavoidable, a design for added wall thickness or reinforcing ribs to share some of the applied loads is a good idea. I’m amazed that failures happen as little as they do; it’s really an indication of the remarkable strength of fibre-reinforced thermoplastics.

None of this is intended to suggest that filled thermoplastics can’t be successfully injection molded, but rather that adding a little chopped glass to a commodity resin isn’t like Hamburger Helper, but in fact results in an entirely new material. Don’t throw the rule book away, but expect to go back to school when molding a reinforced thermoplastic.