March 2012 Design Tip

Cut to the Bone

We've all seen cartoons of the caveman with his waist-high stone wheel. It literally weighs a ton and never seems to be attached to anything. We've seen images of chest-high, solid wood wheels on a peasant's lumbering cart; better than nothing but still far from race-ready. The real breakthrough in wheel design came when some genius realised that a wheel needs a hub and a rim but that almost everything in between is superfluous. It was only with the invention of spokes that the wheel-suddenly lighter, more economical, and with less angular momentum to spin up and brake down-came into its own.

The same principle applies to plastic parts. Plastic is inherently light, generally economical, and relatively easy to form. But that doesn't mean that it can't, with thoughtful design, be lighter still and even more economical, all without sacrificing performance or complicating manufacturing. Accomplishing this requires the elimination of unnecessary material, which raises two questions: why and how.

"Why" has several answers, and the first is cost. A recent item in TPE (The Plastic Exchange) pointed out that buyers were snapping up supplies of resin in late January 2012 to avoid price increases expected in February. The impending increases ranged from £0.01 to £0.04 (0,02? to 0,05?) per pound on some relatively inexpensive resins: polyethylene and polypropylene. If pennies-per-pound savings are that motivational, any significant reduction in the per-part volume of resin could be even more worthwhile, and such savings would be proportionally greater on more expensive resins.

The second reason to trim down is weight. From bicycles and cars to military gear and aerospace, weight reduction has become a kind of Holy Grail. Plastic is, of course, lighter than metal, but less plastic is lighter still. As the saying goes, "You can never be too thin or too rich," and in today's market being lighter can make you richer by reducing cost and increasing functionality.

Another argument for trimming is style. Perhaps because skeletonisation is associated with high-tech, sport, and the like, it has developed a certain cachet. Minimalism sells, which brings us to the next question, which is "how?"

One of the simplest ways to lose weight in a plastic part is to core out thick areas. This not only saves money and weight, but also prevents problems like sink, voids, and warps (see past Design Tip on coring). Another approach is skeletonisation. The concept has been used for years and applied to all sorts of materials. Wood and metal trusses are designed to maintain strength while minimizing material requirements. So is a honeycomb. In many cases, hollow cylinders can replace solid posts, and properly positioned arches can redirect forces and replace bulky solid supports. All of these techniques can be incorporated into plastic parts. The challenge is to remove material without impairing function. Figure 1 shows an example of the process.

Figure 1: This part evolved from a solid disk to one with cored-out sections to a fully-skeletonised spoke wheel with no loss of functionality.

Figure 1: This part evolved from a solid disk to one with cored-out sections to a fully-skeletonised spoke wheel with no loss of functionality.

Since resins vary widely in their characteristics, the amount of material that can be eliminated from a particular design will depend on the material, and choices of material and form will interact as the designer searches for an optimal design. Finite element analysis (FEA) software is great during the early virtual phase of development, but prototypes using actual materials are necessary for testing and refining the design as the process moves forward.

Thoughtful prototyping by rapid injection moulding can provide maximum information at minimum cost. You can test multiple resins in the same mould (as long as you remember that resins with different shrink rates will produce parts of slightly different dimensions). Depending on what you plan to test, e.g., individual part performance rather than a final assembly, this may not be a problem. Also, if you plan to test varying degrees of skeletonisation, remember that metal is easy to remove from a mould but hard to add, and less metal means more plastic. In other words, start with the most trimmed down version of your part. If it doesn't stand up to testing, you can add plastic for your next iteration by milling more metal from the mould instead of starting from scratch. As with skeletonising a part, it's a way of trimming fat from your development process, maintaining effectiveness while reducing cost and effort.