Crystalline resins finally yield to accurate shrink & warp simulation.

Semi-crystalline thermoplastics such as nylon and polyester have limited the accuracy of shrinkage and warpage simulation because crystallization throws material properties into flux. C-Mold Inc., Louisville, Ky., claims to have found a way around the problem.

Last month, the company adopted a new method for characterizing this troublesome class of materials for CAE purposes. Together with a related software enhancement, the new characterization method promises to cut the shrinkage and warpage prediction error for semicrystalline materials down to as little as 10% - a figure backed up by a validation study performed at the General Motors Research and Development Center in Warren, Mich., and funded by the National Institute for Standardization and Testing (see tables).

Given that the shrink/warp prediction error formerly associated with simulation of semi-crystalline resins commonly topped 40% and sometimes reached 100%, the accuracy gains "are a breakthrough," says C-Mold CEO Peter Medina.

[TABULAR DATA OMITTED]

'REAL-WORLD' CRYSTALLINITY

According to C-Mold project leader Manju Mahishi, the new method improves simulation accuracy by taking into account the effect of crystallinity on material properties. "Changes induced by crystallization in pvT and transition temperature significantly influence shrinkage and warpage simulation results," he says. The effect is most noticeable in nylon, PET, and other slow-crystallizing polymers.

In place of traditional pvT data, C-Mold's new characterization method produces what company officials term "fast-cool pvT data." According to Mahishi, "fast-cool pvT" reflects the real-world cooling conditions that govern crystallinity - especially the fact that plastic tends to cool much faster in production molds than in standard pvT tests. "The faster the plastic cools, the less opportunity it has to crystallize," Mahishi explains. "And the slower the plastic cools, the more crystalline it becomes. The more crystallization that occurs, the higher the shrinkage in that area."

Warpage can likewise be traced back to cooling. "Shrink is rarely uniform across a part, which is what causes warpage," notes Medina.

FAST-COOL pvT

C-Mold derives its "fast-cool" data from standard equilibrium pvT and transition-temperature information, along with measurements of crystallization kinetics - that is, the amount and size of crystal formation. The fast-cool pvT data becomes an input for the mold-filling simulation.

C-Mold has started to test resins according to its fast-cool procedure and to provide the results to its customers. But the company has its work cut out for it: Medina estimates that semi-crystalline resins and blends occupy about 40% of C-Mold's material-property database.

ANALYSIS BY LAYERS

The software is also more accurate because it now calculates shrinkage in up to 20 layers through the part, as opposed to applying a single shrinkage value to the entire part. "Using a single-point value is common but misleading," Medina argues.