Three days before President Obama stated that “The blueprint [for a lasting economy] begins with American manufacturing,” in the 2012 State of the Union [1], the New York Times published a widely read article about why the Apple iPhone is not manufactured in the United States [2]. Two key reasons identified were that “factories in Asia can scale up and down faster and Asian supply chains have surpassed what’s in the U.S.” The article does not explicitly point out, however, that once a supply network develops in or moves to a geographic region, the cost of relocating that entire network is extreme, and the network serves as a magnet for attracting other manufacturers. From this perspective, at least a short-term blueprint for strengthening American manufacturing would include investing in and enhancing the flexibility of production segments in which current U.S. supply networks have competitive capacity. Polymer processing is one such segment with significant potential for continued future growth. The commercial importance of polymer processing has substantially increased in the past 40 years, with 5.6% annual growth in plastic consumption [3]. This is more than ten times that of steel and more than twice that of aluminum. Among production processes, extrusion accounts for 40% of all manufactured plastic parts primarily because it offers relatively low cost and high production rates [4]. Injection molding (25%), blow molding (25%), thermoforming (5%), and others (5%) comprise the remainder. The output of a common single screw extruder can exceed four tons of plastic product per hour [5]. In contrast, the maximum production rate of a single injection molding machine is only about 500 pounds per hour because cycle times are dictated by the time required to solidify the part, which depends on the material and size [6]. Besides cost, other advantages of extrusion include the abilities to knit different materials together such that portions of the part have different properties and/or to apply coatings to select regions of the exterior surface. Structural components, such as steel spines, can also be embedded within an extruded profile. Looking forward, demand for extruded plastics in the United States is projected to grow 2.4 percent yearly to nearly 40 billion pounds in 2013, valued at over $23 billion [7]. The current conventional extrusion process uses a stationary die land to manufacture continuous plastic products with a constant cross-section. While extremely popular in a wide range of applications including pipes, household siding, decorative molding, drinking straws, gutter systems, weatherstripping, and window framing components, the extrusion process does not have the flexibility to accommodate a wider range of more complex products due to three fundamental limitations of fixed die design.

A. Constant die land profiles

General, large-scale changes in the shape of the die land profile are not possible with current technology. Creating extruded parts with variable cross sections requires secondary processes such as forming, trimming or adhering additional pieces to the base profile, which significantly increase part cost and decrease production output, thereby negating the two primary advantages of extrusion. Thus, extrusion is presently limited to parts of constant cross section.

B. Limited die land adjustments

The shape of the die land generally does not exactly match the shape of the extruded part because the part swells during cooling [8-10]. Current technology provides no insitu ability to make small modifications to the die land shape to compensate for die swell. (The adjustable die lips on some sheet dies allow only small thickness changes, not shape modifications.) Therefore, an iterative trial-and-error process of evaluating the cooled product and re-machining the die is often implemented to determine the final die land shape [11]. This procedure slows effective production rates and raises part costs, particularly for cases in which too much die material is removed at any stage. Complex parts represent increased risks for these difficulties, further limiting extrusion of parts with precise dimensions to relatively simple cross-sectional shapes.

C. Limited process control adjustments

High quality parts require a uniform velocity distribution in the polymer at the die orifice because unbalanced flows induce internal residual stresses that are manifested as part defects such as shark skins and melt fractures [12]. The die channel is a transition region that funnels polymer from the circular exit of the extruder barrel to the profiled shape at the die land in order to achieve the uniform velocity distribution as the polymer exits the die orifice. Control parameters that affect the flow include screw speed and temperature in the barrel, line speed of the extrusion exiting the die, rate of change of the die’s extrusion profile, and alternative path (leak) flow. The complexities of molten polymer flow through non-trivial geometries usually result in an iterative approach to determining the appropriate combination of these parameters for a given die geometry [8]. Cost- and time-effective extrusion of parts with variable cross-sections requires prediction of the processing parameter values and real-time control to update them as the die land geometry is changed.

The DIMLab is working with the Locomotion and Biomechanics Laboratory at the University of Notre Dame to design and actuate a variable geometry extrusion die.