A move from metal to plastics makes sense from a cost basis. Changing the plastic material to another plastic? Often the performance trade-off just isn’t worth it.
Recently, I worked on a project that converted an aluminum diecast part to one injection molded from highly reinforced thermoplastic polyester. The cost of the part dropped from $2.02 to $0.76, a reduction of more than 60%. It could have likely been made for even less in a thermoset polyester had the end user had a qualified vendor for components molded from thermoset materials, but that’s a topic for another day.
Given the part weight and the fact that the anticipated volume for the part will likely preclude the building of more than one cavity, subsequent cost reductions will be relatively small and will likely involve poor decisions regarding lower-cost “equivalent” resins, a lower-cost molder, or both. And while the quality of the product potentially suffers, the probable additional cost reduction will likely be no greater than $0.15 as opposed to the initial savings of $1.26.
The lesson here is that when it comes to the relentless search for cost reductions, molders of plastic products and their customers benefit to a much greater degree when they replace metal than they do when they cannibalize applications that are already in plastic by using another, less expensive plastic. Metal replacement has long been a focus of the plastic materials industry. Many suppliers of high-performance raw materials have used this approach to the market ever since the golden age of polymer chemistry took us beyond the realm of commodity resins. But there is a general loss of focus on this strategic thrust in the industry and most of the activity today focuses on achieving cost reductions by substituting one plastic material for another in an effort to hold on to the low end of the market.
These strategies follow some predictable patterns. First, because tooling has already been built for the part, a sensible approach attempts to replace an amorphous material with a less expensive amorphous material or a semicrystalline material with a less costly semicrystalline product. A common progression in amorphous resins goes from polycarbonate (PC) to an ABS/PC alloy, or perhaps all the way to ABS. In the semicrystalline realm, a common conversion is nylon (polyamide) to polypropylene (PP).
While this approach can pay some initial benefits, it is a finite process. It’s a given that if the material specification for the part is repeatedly downgraded, a point will be reached where the part will fail to meet requirements. This tendency to revisit an existing application multiple times tends to replace new product development because it appears to be a quick way to cut costs without investing in upfront development work. If the trend continues, the organization goes into creative atrophy and the only avenue for cost reductions involves shaving a certain percentage from existing applications year after year.
Let’s see how it works
A typical scenario goes something like this: A polycarbonate application is targeted for cost reduction. The first step may be simply to find a PC supplier who is offering a more favorable price on the material. This satisfies the cost reduction requirements for the next fiscal year. But in the following year a new cost reduction goal is set and the application comes under review again. This time the procurement people find that an ABS/PC alloy is a less expensive material than a pure polycarbonate. This information may come through a material supplier, the molder of the part, or some independent research on resin pricing conducted by the purchasing people themselves. A qualification process goes forward that may or may not be thorough enough to uncover all the possible shortcomings of the new material. If the new material passes all the internal tests, the product will go to the field, where the market will decide on the wisdom of the new selection.
Fast-forward 18 months to the next cost reduction review. A new purchasing agent has been installed in plastic part procurement and in this what-have-you-done-for-me-lately environment, it’s time to go back to the well for another 3-5% reduction. This time the decision is made to try an ABS. Once again, parts are sampled, measured, and tested and a new generation of product is released to the marketplace. This cycle may even be repeated again with an attempt to replace the ABS with high-impact polystyrene (PS). On each rung of the material family ladder, there may be discussions with multiple suppliers of a given material in an effort to gain a better price. There will also likely be discussions with different molders, trying to find that holy grail of a well-molded part that costs next to nothing.
While this process may make everyone in the organization feel busy and productive, consider the gains even if we assume that everything goes perfectly at every step in the process. Let’s put some hard numbers on the activity using the part mentioned in the opening paragraph. The research tells us that approximately 50% of the cost in the selling price of the average molded part is raw material. We will assume that our part is typical and therefore $0.38 of each $0.76 part is the cost of the PET polyester. We will assume a selling price for the raw material of $2.15/lb. The cost reduction scenario starts when someone finds a PBT polyester with the same amount of glass fiber for $1.90/lb, and the price of the part drops to $0.715, or about 6%.
The PBT only has about 60% of the modulus of the PET, so the product tends to flex more under the applied loads associated with assembly and field use. There also appears to be an increase in the percentage of brittle parts falling out on the assembly line. And there is an indication based on customer complaints that some of the brittle product may be reaching the field as well. The increased brittleness has nothing to do with the inherent properties of the PBT. The two polyesters have comparable ductility. However, PBT is more susceptible to thermal degradation during processing and demands a narrower melt temperature window. If this is not adhered to, the resin loses its inherent toughness. (By the way, none of the costs associated with the assembly line losses or the customer service and warranty costs related to product replacement will be applied against the savings realized by the material change.)
In the meantime, someone finds a data sheet for a nylon 6/6 that is nearly as rigid as the original PET and appears to have better impact strength. It sells for $1.75/lb and has a 12% lower specific gravity, so the material content portion of the part cost drops another $0.06 and reduces the price of the part to $0.655. But after the parts get out to the field, the end user discovers that they increase slightly in critical dimensions and the modulus, which appeared to be nearly equivalent to that of the PET in the as-molded state, has declined to a point where it is now lower than the PBT it replaced. These phenomena are, of course, related to the effects of moisture absorption.
Finally, someone poring through a database comes across the jackpot: a 50% long-glass-fiber-reinforced polypropylene that restores the modulus of the original material, does not pick up moisture with the consequent change in properties and dimensions, sells for $1.48/lb, and reduces the weight of the part another 7%. Unfortunately, the part has a tendency to warp to a greater degree in this material, resulting in a longer cycle time that reduces the cost savings to only $0.02 or a final cost of $0.635.
In addition, the gates and runners in the mold were never designed for long glass, so unbeknownst to the molder and the end user, much of the improved performance derived from using long glass is not realized because the restrictive flow paths in the mold cause a significant level of glass fiber length reduction. This results in excessive creep.
Stop the Cannibalism
Reading this, you may think that it represents an unlikely series of events. But it happens frequently as projects are constantly revisited for some additional cost reduction. In addition, consider all the work that is involved in these iterations. Molds must be sampled, parts must be inspected and measured, and assemblies containing the new parts must be tested according to protocols whose complexity will depend upon the demands of the industry that they serve. In some cases, tooling changes will be required to bring parts back into compliance with part drawings or the drawings will need to be changed to reflect the new part dimensions.
Given all this effort, the relative benefit is very small compared to the initial cost savings realized from the metal-to-plastic conversion. The drive to turn everything into a commodity application has a secondary detriment to the molder. As the material selection is downgraded, the level of skill required to mold the part also becomes less of a selling point. It takes a certain amount of processing acumen to consistently produce good parts in PET and PBT polyester. A much larger population of processors is capable of processing polypropylene, so the competitive field automatically becomes larger.
Here is the big problem with continuing the advancement of plastics into the metal replacement market. Plastic materials are fundamentally different from metals both in their inherent short-term properties and in the way they react to the application influences of temperature, time, and load.
Metals also enjoy a level of standardization that eludes the plastics industry. When someone specifies 380 aluminum, there is a clear set of specifications that govern the nominal levels of the various constituents in the alloy and the allowable variation around those nominal values. But in the plastics industry, you might sample 10 different nylon 6/6 materials with the same stated nominal glass-fiber loading and get a large variation in performance due to differences in parameters such as the molecular weight and molecular weight distribution of the polymer, the length of the glass fibers, and the types of additives employed in the compounds.
In the plastics industry we talk in derogatory terms about a “metals mentality.” This usually describes someone who is comfortable with the performance of metals and shows no inclination to learn about and take advantage of the benefits of plastics. But imagine someone making material selection decisions between metals and plastics by performing the same type of data sheet comparisons that we just illustrated above. For example, aluminum has a modulus of 10 million psi. You can search through a lot of databases devoted to plastic materials and not find a plastic material with even half this stiffness. Obviously, if it were as simple as a direct comparison of properties, the plastics industry would not have enjoyed its past success. But to continue to convince those used to working with more traditional materials that the merits of plastic are worth consideration, we need to educate the marketplace.
We also need to educate ourselves about the fundamentals of plastic material behavior. People who work with metals are used to having reliable data with which to make engineering decisions. When they go to the information resources available on plastic materials, they are stunned at how little we know. That same lack of knowledge is what leads to the convoluted material selection process outlined in our theoretical case study above. Very often we cannot predict with any certainty the result of converting from polyester to nylon and then to polypropylene because we lack the hard data to do so.
In the second part of this article, we will look at a traditional approach to plastic material selection and then illustrate the knowledge gap that exists and how to fill it. The good news is that we know how to generate the information necessary to make informed judgments. We must do so if we are to continue to seek out the opportunities that can keep the industry profitable and prevent our activities from descending into a world where we eat our own in an effort to stave off extinction.
About
Michael Sepe Michael Sepe has worked in the plastics industry since 1975 in a variety of roles involving both manufacturing and research and development. He is an independent consultant based in Arizona with clients throughout North America. He assists clients with material selection, designing for manufacturability, process optimization, troubleshooting, and failure analysis. Learn
more about Michael Sepe.
