Sponsored By

September 18, 1998

5 Min Read
The Materials Analyst, Part 7: Melt flow revisited

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 10 years. He can be reached at (414) 369-5555, Ext. 572.

In the first article of this series we illustrated the use of the melt viscosity or melt flow rate test to solve a performance problem with a molded part. In that case, we did not have raw material to use as a comparison, but we knew the historical range of values for the virgin material. The change in the melt flow rate from pellet to part was so large that there was no question about the role of improper processing in the failure of the part. But sometimes the comparison of a molded part to the generic specifications for a given material does not give a clear answer. In these cases, it is essential that the part be compared directly to pellets from the same lot used to mold the product.

This was the problem when a molder sent in samples of a box and cover molded from a PC/PET polyester blend. The material has good chemical resistance, good electrical properties, and adequate heat resistance, all of which were required for the application. But the most important performance criterion was impact resistance; the parts formed the exterior of an enclosure that had to pass a demanding drop test. Both the box and the cover were approximately 4 inches wide and 10 inches long and the nominal wall was only .060 inch. With a maximum flow length of 12 inches from the gate to the end of flow, the flow length to wall thickness ratio was 200:1. Originally, the part had been specified in a 12-melt-flow grade of the material. However, the pressures required to fill the cavity were beyond the capability of the equipment. Efforts to reduce viscosity to manageable levels by raising the melt temperature of the material had resulted in thermal degradation. The problem was "solved" by moving to the next available flow rate--a 30-melt-flow grade. However, the parts failed the drop test.

Unfortunately, this is a common occurrence today. In the quest for weight reduction, material savings, and faster cycles, wall thicknesses are pushed to lower and lower levels while, at the same time, part consolidation efforts increase the complexity of the design. When the part fails to fill, the easy solution is to go to a lower viscosity material. However, lower viscosity almost always means lower molecular weight. And when molecular weight is reduced, the first property to suffer is impact resistance.

When we talk about material degraded from excessive heat or improper drying, we are really talking about molecular weight reductions. While material suppliers produce high-flow grades of material under more careful conditions, the result is still a reduced molecular weight, and many of the same performance trade-offs that appear in a degraded polymer will occur in a high-flow virgin resin. The relationship between molecular weight, viscosity, and toughness is well established.

Were the impact failures due to the grade change or a problem with processing? PET polyester and polycarbonate are both very sensitive to moisture, so the possibility existed that the material had degraded during processing. It was also possible that the material was in its intended state, but that the 30-melt-flow rate was not compatible with the impact requirements. The melt-flow rate of the molded parts was 48. At first glance, it seemed obvious. A 30-melt-flow material had changed by 18 points; this was a 60 percent increase in melt flow. For an unfilled material, this fell well outside guidelines of a 30 percent maximum shift for good product and a 40 percent maximum shift for borderline product.

However, the problem was not so simple. A check of the specification range for this material showed that the acceptable range of melt flow values was 26 to 34. If the virgin pellets had actually started at 26, then the shift would have been 22 points for an increase of 85 percent. On the other hand, if the material had begun at 34, then the change was only 14 points for a shift of 41 percent.

In this case, we needed to compare the parts with raw material in order to determine the problem's source. We obtained a sample and ran the melt flow test. To our surprise the raw material checked out at 42! The raw material was out of specification, and the actual shift in melt flow attributable to the molder was only 14.3 percent. The molder had done an excellent job of processing; the problem was in the pellets. It was not surprising that the mold had filled so easily.

Once the problem was understood, a new lot of material was tested to certify that it was within specification at a 29.5 melt flow. The resulting molded parts checked out at 36, a 6.5 point shift or 22 percent. The impact tests were better, and with a slight increase in the nominal wall to .070 inch and the rounding of some sharp corners, the design was sound. This case brought into focus the interdependence of tool design, part design, material selection, and processing. A simple melt-flow-rate test helped to keep the development process on track and brought matters to a much quicker resolution than guesswork and finger pointing.

Sign up for the PlasticsToday NewsFeed newsletter.

You May Also Like