Using simulation technology for plastic part design has become nearly a given in the past five years. A recent project from Lexmark Electronics (formerly Plastics Technology Center) underscores the reasons why more designers and molders are turning to moldfilling analysis. The former product designers and developers from PTC are now contract manufacturers who rely on experience and CAE to meet customer needs.
Early in the design phase of Lexmark's Optra S laser printer, 53 out of 360 parts were simulated for moldfilling. These represent about 80 percent of the plastic in the printer. At mold trial, for one set of molds, 19 out of 19 tools filled completely the first time. Lexmark estimates the upfront design analysis avoided roughly $10 million/year in manufacturing and opportunity costs. Want to know more?
According to Jim Rioux, a Lexmark development and simulation engineer involved in the project, three of the parts analyzed comprised the heart and soul of the machine. They constituted the internal frame for the printer: a center part (EP, for electro-static printer), left frame, and right frame members. Optimizing these parts had a significant effect on the overall cost-effectiveness of the printer. "We focused our simulation efforts here, where major opportunities could present themselves," he says. Norm Smith, development manager for Lexmark, adds, "Our motto is 'Good parts-first shot-no excuses.'"
One of the first benefits of simulation presented itself during early supplier involvement meetings. "The toolmaker was brought in early to help with part design," says Rioux, "and GE Plastics also attended those meetings to give advice and recommendations about the material, a glass and mineral-filled mPPO (Noryl HM4025). Simulation technology used all inputs to optimize the design."
|Moldfilling analysis helped Lexmark designers, toolmakers, and material suppliers agree early in the design phase on an optimum wall thickness. Melt front advancement predicted via CMold simulation for the center EP frame part (right) for Lexmark's Optra S laser printer also showed the design team that moving most critical tolerance parts to the center would not impede processability. An identical simulation for the right frame (above) helped ensure the thin-wall (under 2.5-mm) part would fill.
Simulation offered a second benefit for all 53 parts simulated, especially the three critical parts. Lexmark wanted to minimize wall thickness as much as possible, but with relatively steep flow lengths, Rioux and the team needed to know how those wall sections would fill. "We used simulation to see how much we could minimize wall thickness and still fill the cavity. We also needed to reach required stiffness, appearance, and cost goals along with other performance properties," he says. "Simulations allowed us to reduce wall sections to under 2.5 mm (from a previous program where wall thicknesses varied from 3 to 4 mm) and still meet our criteria." Using CMold software, the team flagged all major problems that could occur during filling: shear rate too high, pressures too high, and gross changes in the melt front advancement.
Finally, the design team at Lexmark used moldfilling analysis to support a creative design for the center EP part. "Part designers at Lexmark placed the majority of tight tolerances on this center frame to reduce critical molding in the various other parts of the printer," Rioux says. "By doing this, they were able to consolidate parts and keep costs to a minimum, especially in terms of tooling dollars, yet still maintain high quality. Simulation helped us determine if this strategy would work."
To make the best use of simulation, Rioux recommends using trends and magnitudes rather than hard numbers. "Exact warpage prediction can happen, but it's not 100 percent reliable. Perhaps 80 percent is a more reasonable figure for amorphous materials. For fiber-reinforced materials, the figure drops somewhat." Remember, a few iterations can help give tooling designers valuable direction. "Tooling engineers are talented. If a part only needs a better gate location, with their input, we run through the intent and can find it quickly."
Reliability is a core issue for simulation technology. Lexmark's moldfilling simulation team has an extensive background in analysis and checks results against actual parts to see how well they correlate. "When we ran the printer parts analysis, CMold didn't have the module for fiber orientation. Today, CMold integrates the information from its fiber orientation module into its shrinkage and warpage calculations," explains Rioux. "Now that we have it, we've benchmarked the original parts against the results, and there is a good correlation with warpage direction."
Surprisingly, Rioux advises not every part needs simulation, which he believes is a great tool for teaching designers and analysts about the injection molding process. It is not necessary for parts based on the same shape as a previously successful part, those with loose dimensional tolerances, or those with only one or two gating scenarios. Following good design practices for plastic parts and for the specific polymer (i.e., putting on a rib at the recommended wall thickness) is often enough. For these simpler parts, a simulation expert, experienced plastics designer, or tooling engineer can spot problems during design review without performing simulation.
However, when trying to minimize wall thickness and still fill the cavity, simulation is a must. Ditto for parts sensitive to warpage and those with tight dimensional tolerances. To minimize material usage on a nonappearance part by including holes, Rioux says, combine moldfilling with structural analysis.
|Key simulation elements explained|
|Lexmark's Steve DeFosse found that not understanding simulation technology jargon can seriously impede progress among designers performing moldfilling analyses. His tutorial, "Computer Simulation of Injection Molded Plastic Parts," is excerpted here to help clarify these terms.
Materials characterization: Data needed to conduct process simulation. For flow phase simulation, designers need shear rate vs. melt viscosity curves generated at three or four temperatures spanning the recommended processing range. This phase also requires heating and cooling rates, Tg, thermal conductivity, no-flow temperature, CTE, specific heat, and eject temperatures. Pack and hold phase analysis requires PVT (pressure-volume-temperature) relationships of the resin, as well as solid and melt stage densities. Cooling phase analysis uses properties measured for flow and pack. Warpage and shrinkage simulations need bulk modulus, compressive modulus, in-plane shear modulus, stress-strain curves, and Poisson's ratio in two directions.
Model building: Runners, gates, and cooling lines are modeled with 1-D elements because fluid flows in only one direction through them. Parts are modeled with 2-D, thin-shelled, triangular elements meshed over a centerline wireframe of the part design. Layers are defined by the software to represent each wall section. Mid-plane generators, available with most CAD packages, can simplify mesh generation.
Most models are imported from CAD files. Having to build a centerline model manually can double the time to complete an analysis.
Flow phase analysis: Offers melt front control to achieve balanced cavity fill, the ability to determine where vents should be located, and knit line location and size.
Pack phase analysis: Identifies temperature and pressure gradients across the primary walls of the part that can lead to anisotropic shrinkage and, thus, warps, sinks, voids, or other defects.
Cooling phase analysis: Takes mold steel, coolant, coolant flow rate, inlet temperature, and cooling channel configurations into account to predict how the mold surface temperature changes as a function of time during the molding cycle. Objectives include reducing cycle time by optimizing the process, mapping thermal efficiency of the cooling system, evaluating mold designs, improving part quality aesthetically or dimensionally, and looking for potential hot spots (10 deg C difference, as a rule of thumb) in an initial part design.
Warpage analysis: Recently developed tool that predicts direction and magnitude of warpage for the combination of material, design, and process. As with any new, complex software, features are still being refined. Lexmark has found highly accurate results in some cases, which reflect actual molded part behavior, with others showing accuracy in only direction of distortion rather than magnitude.
Phone: (606) 232-6599
Fax: (606) 232-2103