Sponsored By

November 28, 1999

9 Min Read
The Materials Analyst, Part 27: Contamination springs from cost pressures

This series of articles is designed to help molders understand how a few analytical tools can help diagnose a part failure problem. Michael Sepe is our analyst and author. He is the technical director at Dickten & Masch Mfg., a molder of thermoset and thermoplastic materials in Nashotah, WI. He has provided analytical services to material suppliers, molders, and end users for the last 12 years. He can be reached at (414) 369-5555, ext. 572.

It is almost impossible to carry on a conversation of any length in the world of plastic manufacturing these days without hearing or uttering the phrase “cost savings” or “cost reduction.” We seem to have become obsessed with the Lopezian idea that we can cut our way to prosperity and that the easiest way to accomplish this is on the backs of our vendors. It appears that we have abandoned the notion that business is about revenue generation.

This is unfortunate, because very often the mind-set of the cost reducer is antithetical to the mind-set of the innovative person who is looking for opportunities to improve the company’s portfolio of products and services. And at this particular point in our history, when faced with a choice between the two, most companies are going with cost reduction. Like it or not, it is the world that we live in, and it will probably get worse before it gets better.

The strategy of cost reduction, by itself, is not a bad one. Certainly most of us who were around to remember the 1970s acknowledge that elimination of entire layers of the organization and our newfound infatuation with quality were the keys to a resurgence in an American economic system that was taking its lumps after becoming fat and complacent following the second World War. But taken to its extremes, the drive for lower cost results in the use of substandard materials and shoddy workmanship.

This lesson appears repeatedly in our analysis of failed parts. The single biggest contributor to the poor performance of the molded part is an excessive loss of molecular weight.

Sometimes this is the result of poor processing, but increasingly it is caused by incoming material that has been so badly abused that the molder has almost no chance of producing a consistently good product. The second largest contributor is contamination. Here again, the molder may be at fault. But many of the instances that have appeared recently reveal that the problem starts before the molder ever sees the material. In some cases, the problem results in high scrap rates as the contaminated product has obvious flaws that prevent it from being usable. Surprisingly, these processors are the lucky ones.

The truly unfortunate participants are those that can actually make what appears to be an acceptable part, only to find out later that the part does not perform to customer specifications. This is when the panic really sets in. In these cases, lines are down, large end users are throwing out scary terms like “$2 million an hour,” and everyone is in a hurry to find the root cause. While the process can be a painful one, the good news is that contamination can be readily identified with simple lab tests.

The approach depends upon the exact nature of the problem. We will focus on true polymer contamination, the mixing of resin families to produce a material with a new set of properties, rather than a glass content that is off the mark because someone dumped the wrong bag of material into the hopper. We will also focus on contamination that occurred before the material reached the molder, since this is the trend that seems to be on the rise as processors seek to find better deals on raw material prices.

Unhealthy Blend
The first case comes from a molder of thin-walled parts being produced in nylon 6/12. This was an existing application that had run well for an extended period of time when the product began to exhibit an uneven surface finish. The surface felt as though it had lumps of unmelted material in it. Nylon 6/12 melts at 218C (424F) and the melt temperature of the material had been measured directly at 252C (485F).

With these thermal conditions, and with the relatively high shear rates associated with the part geometry, it did not seem likely that there would be unmelted pellets in the material stream unless the screw and barrel were badly worn. The molder was knowledgeable about the role of screw and barrel wear in melting efficiency, and the throughput on the machine was high for the size of the cylinder. But before going to the trouble of taking everything down for a day to clean and measure all of the components, the molder wanted to be sure the problem was not with the material. The problem had come on fairly suddenly and a wear condition would be expected to develop more gradually.

Because the problem was related to the ability of the material to melt, we decided on DSC testing, since one of its strengths is its ability to distinguish between materials of different melting points. We took two samples from a single molded part. One sample had a visible lump in it while the other one was normal from a cosmetic standpoint.

Figure 1 shows the result. The area that contained the irregular lump had a significant percentage of a second material with a much higher melting point. The melting point was a textbook match for nylon 6/6, and the melting point of 263C (505F) explained a lot about the problems with unmelted material. The surprise came from the area that appeared to be free of contamination. A much smaller amount of nylon 6/6 was found there as well, indicating that the contamination was fairly widespread. The punch line was that this processor did not mold a pound of nylon 6/6 anywhere in its facility.

The material, a low-cost version of nylon 6/12 that lists for nearly $4/lb, had become contaminated at the material supplier’s plant. It was costing the molder significantly in scrap and had almost prompted management to shut down the equipment for a day. While nylon 6/6 and 6/12 should mix well, they will only do so if the melt temperatures are high enough to soften both polymers. Otherwise, problems with unmelt, plugged gates, and even weak spots caused by stress concentration are the result.

Contamination Abounds
The second example comes from a large molder of automotive parts. Anyone who has dealt with the automotive industry knows that price pressures are relentless. There is also a newfound fascination among car companies with the notion of using recycled materials. In some cases this interest has reached the level of a mandate.

This molder had worked with a compounder, who in turn worked with a reclaimer of polyolefins, to produce an inexpensive grade of black 30 percent talc-filled polypropylene homopolymer. The resulting problem, unfortunately, got past the molder and to the end user’s assembly line.

During assembly, these parts were placed under a significant compressive load, a load that usually presented no problem for the part. Quite suddenly, the parts began to compress and deform during assembly to such a degree that the required torque could not be developed on the fasteners. When the sample got to us, the prevailing theory was that the material was low on talc, a condition that would certainly lower the modulus of the material and make it perform in this way. Because of the focus on the filler, we first ran a TGA test to determine the filler content. It was 29.2 percent, well within specification, and it was all talc.

Since this first approach was a blind alley, we decided to compare the composition of a good and a bad part using DSC. Figure 2 shows a comparison of the two parts as they melted. In reality, neither of these materials was a homopolymer; both contained polyethylene. The difference was that the good part was a copolymer. The manner in which the polyethylene melted was characteristic of a material in which the ethylene was reacted into the polymer backbone to make a true copolymer. The amount of ethylene in the compound was significant, and the modulus would definitely be lower in a material of this type than in a true homopolymer. But with the addition of the talc it was apparently stiff enough to make it through the assembly process and was considered to be representative of a good part.

The bad part was another story. The sharp and distinct melting point associated with the lower temperature transition also belongs to polyethylene. But in this case we were dealing with a mixture rather than a compound, and there was a high level of contamination. It is difficult to estimate the exact percentages of the two polymers using this method. However, it was clear that the amount of polyethylene was much greater in the bad part than in the good one and our estimate was that anywhere from 30 to 50 percent of the polymer in the bad part was polyethylene. This produced a much more flexible product that could not withstand the compressive forces of the assembly process. The good compatibility of the two materials, combined with the lower melting point of the contaminant, almost guaranteed that this particular problem would get past the processor undetected.

When the processor notified the compounder of the results of our tests, it in turn examined the feedstocks it was using to produce the filled material and found that the contamination was traceable back to the incoming reclaimed material.

Caution as a Defense
The flurry of activity that this type of event sets off as everyone scrambles to replace discrepant product and then deal with the task of assigning responsibility is something that most of us have been through at some point in our careers. It is regrettable that the costs associated with such a debacle are never fully captured and directly applied to the price of the raw material so that the true cost savings can be assessed. The only reliable defense against a recurrence is increased testing of the incoming product or the installation of separating equipment at the reclaim installation that can tell the difference between polyethylene and polypropylene.

In other words, someone has to spend money and invest in equipment in order to engineer the quality back into a system that has dropped below a critical level. Pricing has to reflect this investment or manufacturers will simply begin to walk away from the business.

Cost reductions that focus on material prices have a certain logic to them. All of the analysis shows that raw material represents 45 to 55 percent of the cost pie for molders, so it is a logical place to look for savings. But we need to remember that business is about revenue generation, and cost reduction strategies taken to an extreme inevitably result in poor quality unless they are coupled with innovative methods and good awareness of the fundamentals. The above examples are cautionary tales of what happens when we confuse price reductions with cost reductions.

Sign up for the PlasticsToday NewsFeed newsletter.

You May Also Like