Simulation can determine whether the mold has a low shear rate and shear stress area on the metal surface, where the polymer will eventually degrade, and can help processors design molds that are more suitable for their projects. #Optimal practice#dies

In Part 1 of this two-part series, we studied the actual process setup for the production of medical tubes. It was found (through customer trials and then through simulations) that the equipment available for the processor was too large for the relatively low output required by the project, so the polymer degraded on the screw surface. It was finally determined through simulation that smaller extrusion opportunities with screws optimized for low speeds would work better. 

It is also determined through simulation that the mold cut by the processor for this project is not optimized; it has many areas of low shear rate and shear stress on the metal surface, and the polymer will eventually degrade in these areas. This is a critical issue because it is often difficult to realize during physical testing that if it occurs on the inner surface of the mold, degraded material will form in the mold. In the second and final part, we report the analysis using VEL software (from Compuplast's virtual extrusion laboratory), which helps the processor design a mold that is more suitable for the project.  

The pressure and velocity contours predicted by the VEL using the side feed tube die are monitored. In fact, only half of the chips were simulated—where possible, symmetry was used in the simulation to reduce computational requirements. The velocity contour at the exit of the mold shows a small preferential flow towards the entrance.

The velocity contour at the end of the distributor indicates a poor design, resulting in a much larger flow at the inlet. The resistance in the flow path through the distributor seems to be correcting this change in order to be more uniform at the exit. However, this is not as effective as having a properly designed dispenser.

In the full size range, it is difficult to see the critical low shear rate region, which can lead to longer residence time and polymer degradation. Figure 1 shows the color profile of the shear rate in the flow field, but the scale is adjusted to emphasize the low shear rate area (blue area). Similarly, the color contour of the shear stress is drawn with a maximum scale setting of 20 kPa. The surface area of ​​the mold where the wall shear stress is less than 20 kPa is more likely to cause polymer stagnation and degradation.

Figure 1. Shear rate profile in a side feed tube die.

In addition, the geometry appears to have several "step changes", which lead to stagnant zones where the polymer eventually degrades. A more appropriately designed mold will avoid these types of problem conditions.

The pressure profile on the pipe mold with the spiral mandrel-type distributor was also monitored. The pressure drop appears to be too large, approximately twice the pressure drop through the side feed mold. The velocity contour is drawn on the cross section and drawn in relief at the exit of the mold. The latter seems very uniform.

Figure 2 shows the relief velocity profile at the end of the spiral spindle distributor. The velocity contour at the end of the dispenser looks very uniform, indicating that the spiral mandrel successfully dispenses the material.

Figure 2 Speed ​​contours of the spiral spindle distributor.

The color profile of the shear rate is drawn at the maximum scale set to 10/sec. It seems that the critical low shear rate (blue area) on the wall is mainly caused by a step change in the flow field. The color profile of the step-change shear stress is also very low, leading to stagnant areas where the polymer is most likely to degrade.

There is also a very low stress area on the outer surface at the beginning of the tapered transition. These areas of low shear stress should be removed by improving the mold geometry, as they will limit the duration of the mold's operation before it needs to be cleaned. In addition, more importantly, the extremely low stress areas on the inner surface can cause polymer degradation. This degradation may break and eventually enter the inside of the product, quality control may not see it, and it may be delivered to the customer.

Figure 3 shows some path lines in the spiral mandrel distributor. The path line is usually used to see how far the flow from each spiral is distributed on the circumference. The farther the material spreads on the circumference of the mold, the better the uniformity of the final product.

Figure 3 The path line on the pressure profile in the spiral mandrel distributor.

All molds designed to produce annular products inevitably produce welding lines in the product. Table 1 shows the direction of weld lines in three different types of ring molds.

In most cases, the weld line does not matter; however, if it is formed from a material that has a long residence time — and may degrade — it may affect the strength of the final product in the area. The orientation of the weld lines in the spiral mandrel mold results in an inherently stronger structure. The pipes produced with spiral mandrel-type distributors are also more resistant to "kinks."

The multi-lumen tube has two or more channels for conveying different materials during medical procedures. The overall diameter of the pipe is very small, usually less than 5 mm, and may contain three or four separate channels. Consider here a multi-lumen tube with four channels and a total diameter of 2.5 mm.

Due to client confidentiality, geometric dimensional details are edited. Processors report that pipelines require a long-term trial and error development process to produce acceptable products. In addition, every start-up company needs to "make major adjustments to bring the product back to standard." In addition, the process is reported to be very unstable, and product dimensions often exceed tolerances.  

Flow balance involves adjusting the geometry of the mold outlet to try to achieve the desired product shape. In addition, there are many sudden changes in geometry, which is undesirable for stable operation.  

Figure 4 shows the color profile of the velocity from the flow simulation of the original mold exit geometry, analyzed using semi-symmetrical. The calculated percentage flow (shown in light blue or dark blue) and the required percentage flow (black between brackets) for each area of ​​the cross section are based on the desired final product shape. The light blue number indicates that the flow rate is higher than the requirement, while the dark blue number indicates that the flow rate is below the requirement. 

Figure 4 Simulated MFR distribution (semi-symmetrical) at the exit of the die in a multi-cavity tube.

The color contour of the pressure field at the tip of the mold is drawn. The maximum pressure drop through this area is expected to be approximately 38 MPa (5500 psi), which is quite high for such a relatively short area.  

Monitor the color profile of the velocity in this area-especially the maximum velocity in the three areas of the flow field. It was found that the maximum speed into this area was 7.9 mm/s, then increased to 185 mm/s in the area with the smallest diameter, and then suddenly dropped to 78 mm/s before the die exit. These large and sudden changes in speed indicate that the polymer melt is undergoing large changes in stretching (or stretching) deformation.

Monitor the path lines in the flow field and plot the elongation rates along these path lines. The graph shows that when the material flows from the tapered area to the smallest diameter area, its elongation deformation suddenly increases to about 200 times/sec, and then drops to about -350 times/sec when it enters the area just before the die exit. Negative elongation is essentially a kind of compression.  

These sudden changes in tensile deformation make the polymer's extensional viscosity work, which is the resistance to acceleration or deceleration. This is another material characteristic of polymers. The extensional viscosity can be measured and incorporated into the mold design, but this characteristic cannot be well controlled during the resin production process and may vary from batch to batch. Table 2 summarizes the analysis results of the original mold. 

The evaluation summary shows that the main problems of the mold are large tensile deformation and poor polymer distribution. The pressure drop is higher than necessary, but this does not really cause a problem, it's just that the extruder has to work harder to deliver the material. The analysis also reviewed relatively small viscous heating and wall shear stresses, showing small shear stresses and potential degradation areas. Therefore, it is recommended to modify the die design to correct these problems.

There may be countless ways to balance the flow in the mold, but they all involve adjusting the relative resistance to achieve the desired flow distribution. Nevertheless, some methods are often more practical than others.    

The velocity profile in the original mold outlet shape is drawn together with the flow balance option: 

 • Option 1 is to modify the outer diameter of the geometry to change the resistance in different areas, but this may affect the roundness of the final product. • Option 2 is to modify the inserted needle to create multiple cavities, but this may affect the cavity shape. In addition, these pins are very small and can easily be deflected by flow. • Option 3 is to add smaller pins in high-speed areas to create greater resistance and reduce flow in these areas. However, the determination of such pins is also too small and impractical.  

The simulation determined that the most practical solution was to make a mold with basically three new parts. The new design will include a "balance plate" where the flow distribution will be very precisely balanced to the requirements of the final product. In addition, the length of this part will be designed to have the greatest resistance within the chip. This will ensure that this part has the greatest control over the flow distribution. 

Next, there will be a short, tapered transition plate that leads to the die exit plate. The exit plate will have basically the same shape as the final product, but scaled up about five times to account for the drop.

Observe the close-up of the balance plate and the three restrictors cut into the outer plate through simulation. The restrictor is positioned in the open area with excessive flow and optimized in size to obtain a flow distribution within 5% of the target distribution, as shown in Figure 5.

Figure 5 Flow balance of the cross section of the balance plate (1/3 symmetry).

The pressure drop of the new design is expected to be about 32 MPa, which is about 6 MPa smaller than the original design.

Draw the elongation along the path line in the new design. It is obvious from the scale of the chart that the tensile deformation in the new design is significantly reduced. Table 3 summarizes the standards in the new design.

The proposed new design was delivered and the board was manufactured accordingly. According to reports, after preliminary tests, some minor modifications are required to achieve the required product specifications. After that, it is reported that the start-up speed is much faster, the start-up waste is little, and then the production is very stable, with negligible sensitivity to material batch-to-batch changes. Cpk of critical dimensions> 2.0. 

Facts have proved that the extrusion simulation based on VEL software is very useful in predicting problems and running "what if" scenarios to adjust the size of the extruder and design related molds for plastic extrusion projects. Simulation should be used in the early stages of any project to avoid trial and error and rework tools.

About the author: Dr. Ben Chochaoui is a technical consultant for Windsor Industrial Development Laboratory Inc. in Windsor, Ontario, and has been with the company since 2009. During his career, he promoted his employers in FEA/CAE and CFD/Cam; established laboratories and simulation tools for demanding functional systems; numerical calculations related to product processing and durability tests; supervised engineers; Manage government-sponsored research; develop and maintain customer accounts; and train new engineers and R&D personnel in FEA and CFD. Contact: 519-991-8919; bencho@windsorlabs.ca.

This is a quick guide to solving four annoying problems in handling PET bottles.

The molder should be aware of the influence of process conditions on the final performance of the part.

Understand and solve the possible root causes of welding problems to keep production going on normally.

© 2021 Gardner Business Media, Inc. Privacy Policy [Login]