Structural analyis codes already do a good job predicting the performance of molded plastic parts. Yet, their results never tell the full story because the mechanical properties of those parts depend so heavily on the molding process. That’s why plastics-savvy engineers often supplement their structural analysis work with mold filling simulations that can reveal the influence of the molding conditions.
In the past, engineers had to do quite a bit of importing, exporting, and tweaking of meshes and models to synthesize the results of these two different types of simulations. And juggling all those results took time. “Some analyses that made sense in theory just weren’t practical to run,” notes Paul VanHuffel, a senior engineer and simulation expert at Cascade Engineering, a large injection molding firm that recently launched a CAE consulting service.
Now, Moldflow Corp. has redefined what is and what isn’t practical. Taking a multiphysics approach, the company has fully coupled its filling simulations with its own 3D structural analysis code. While it doesn’t replace dedicated structural analysis products, this multiphysics capabilility has helped Moldflow’s power users learn new things not only about their part performance but also about the economics of their molding processes. “Multiphysics has opened up new realms of analysis for us,” says Van Huffel.
The most important of these new realms involves performing a structural analysis of the the mold and molding machine components during the filling and packing phases of the injection molding cycle. Simulation codes--and maybe a few engineers--have long assumed that the mold and machine elements remain rigid during the high-pressure onslaught of the injected plastics. But that’s just not so in the real world. “Any engineer who has spent time around a molding shop knows that the metal moves,” says VanHuffel.
According to VanHuffel, understanding how the mold and machine deflect provides valuable insight into a part’s performance attributes, including weight and shape. It also can provide information useful in tool design. And it can help determine the best size molding machine for a job, a decision that will affect piece part costs.
Moldflow offers its structural analysis capabilities as part of the “Core Shift” module found in its Version 5.0 software. As its name suggests, this module began as a way to simulate the deflection of mold cores subjected to the pressure of the plastic melt flow. The software works by taking the pressure distribution results from the filling analysis and plugging them into a structural analysis to calculates core deflection. “The pressure distribution of the melt on the core becomes a boundary condition for the structural analysis,” explains Moldflow product manager Murali Annareddy.
The software then uses this deflection result to adjust the thickness of those mesh elements that represent the melt flow around the core--so that subsequent time steps in the filling analysis reflect any changes in core position. This iterative process repeats throughout the filling and packing cycles at user-defined time steps, providing a quasi-dynamic view of how deflecting cores can influence filling as their movement changes the shape of the cavity.
And the module doesn’t stop with deflection of the cores. The same multi-physics approach can also look at the deflection of any mold feature--including slides, ejector systems, and even the mold plates. It can also be used to analyze the movement of inserts inside the tool. Over the past year, Moldflow users have analyzed core shift in wide variety of applications. These include cases where the tooling has delicate features, such as the molds for electrical connectors. They also include applications whose molds have long, unsupported cores--from syringes up to industrial trash containers.
Beyond Core Shift
VanHuffel has taken Core Shift even further. He’s devised a way for to predict both the stretch of the tiebars on the molding machine’s clamp end as well as an injection mold’s tendency to flash.
Tiebar stretch is a normal part of a molding machine’s clamp operations, but it can becomes excessive when molders try to run molds on the smallest feasible molding machine in an effort to keep hourly production costs in check. In these “over-tonnage” situations, the tiebars can stretch enough for the mold to open slightly. In the worst cases, the mold will flash as plastic leaks through the open parting line, driving up rejects or triggering costly flash removal operations.
Even if no flash occurs, VanHuffel has found that parts gain weight since they slightly open tool holds more plastic. On big parts, that extra material adds up. “Smaller presses might seem to save money, but not if parts end up heavier than they should be,” he says. “On smaller parts, the extra material may not matter as much from a cost standpoint,” he continues. “But you run a bigger risk of tool damage since flash can erode the parting line.”
In the past, VanHuffel managed to simulate this over-tonnage behavior “manually” by combining the Moldflow simulation with calculations made in an Excel spreadsheet. Aside from taking too much time, this method had accuracy limitations because it assumed the tool was rigid. It thus neglected the deflection of the tool steel and the stiffness enhancements from the mold’s support components. Both of these factors tend to offset the influence of tiebar stretch, VanHuffel notes.
With Core Shift, VanHuffel now meshes the entire core-side of the mold along with the tiebars and most other parts of the molding machine’s clamp end. He also adds a series of vanishingly thin (1E-6) flow elements along the parting line of the tool. These “flash” elements approximate a zero-thickness element, something that the Moldflow code doesn’t currently support.
When the multiphysics analysis runs, VanHuffel “stretches” the tiebars with a displacement load corresponds to the clamp force required for a given molding job. If that load stretches the tiebars enough to open the tool, it shows up in the coupled filling analysis: The thin “flash” elements get thick enough to allow plastics to flow into them, indicating that the tool would flash. Conversely, if the mold closing forces can overcome the tiebar stretch, the flash elements will, in effect, remain at zero.
So far, the results of the analysis have been dead on. “We’ve been able to predict deflection within 0.001 inch of the deflection measured on actual parts,” Van Huffel reports.
And he say that Moldflow’s multiphysics capabilities do more than predict flash. To take one example, similar techniques can also help balance runner systems in large parts with many drops--since unbalanced conditions can exert unequal forces on the tiebars. He has also applied his flash elements method to predictions of whether mold vents will clog.
As the multiphysics capability evolves, VanHuffel believes he’ll be able to do even more with the software. His next project could be the most valuable one yet. Once Moldflow adds the ability to display not only deflection but also stress distributions--a feature that’s currently in the works--VanHuffel says he’ll actually be able to predict fatigue life of tooling components. “That will be a huge development,” he says.
Cascade Engineering has used Moldflows multiphysics capabilities to analyze tooling and machine deflection during the injection molding process. The analysis approach has been particularly helpful with large parts, such as this garbage cans.
In addition to its injection molding business, Cascade Engineering has launched a consulting service that will run mold-filling and structural analyses. For more information, visit http://www.cascadeng.com/center/consulting.htm.
For more information on Moldflow, go to www.moldflow.com. A white-paper on Over-tonnage Prediction using Moldflow can be found at http://www.moldflow.com/stp/pdf/eng/MPI_Core_Shift_White_Paper.pdf