Process simulation cuts the guesswork in blow molding & thermoforming.

New software that simulates stretching and thinning of a hot parison or sheet helps you to optimize your process and tool design without trial and error. Two case studies demonstrate how it can pay off.

Even highly experienced molders must still rely on educated guesswork when designing blow molded or thermo-formed parts, tools, and the process conditions to go with them. Rules of thumb can go only so far in anticipating all of the factors that will affect the final results. Computerized simulation can show how the blow molded parison or thermoformed sheet looks at different time steps in the process and allows other data, such as wall thickness, stretch ratio, or temperature at any point on the part surface, to be viewed at each time step.

Computer simulation - which is still quite new for these processes - an provide a clearer direction for part, parison, and mold design prior to tool manufacture. It also provides better data on whether or not a part will achieve minimum wall thickness, optimal cycle time, and structural requirements. Less obvious, but equally important, simulated data on wall thickness, cooling, and area stretch ratio can be "mapped back" to the original parison or sheet, providing a useful guide for parison programming or heater placement to achieve optimum wall thickness and uniformity.

The following two case studies demonstrate the benefits that can be achieved when wall thickness, cooling time, and area stretch ratio are simulated early in the design process. The first is a blow molding capability study based on an existing product in which original parison thickness was reduced by 33%. The second is an actual thermoforming application in which the simulation provided data that were used as a basis for structural analysis.

REDUCING PARISON THICKNESS

The objective of the first example, which was based on an existing extrusion blow molded bottle for dishwashing liquid, was to reduce the original extruded parison thickness from 0.090 in. to 0.060 in. The challenge was to determine if the final thickness would still meet product specification and to predict how much cooling time would be reduced.

In Fig. 1, our C-Mold simulation software shows the predicted final wall thickness for both starting parison thicknesses. The thinnest areas are in the comers of the bottle - approx. 0.019 in. for the original parison and 0.015 in. for the reduced parison a wall-thickness reduction of about 21%. Because the thick area near the cap does not stretch significantly, it dictates the cooling time. With the 0.060-in. parison, faster cooling at the pinchoff area should allow the bottle to be ejected sooner. Because cooling time is proportional to the square of thickness, the 0.060-in. parison should give a better cycle time with a relatively small trade-off in the wall thickness at the corners.

Figure 2 shows that the 0.060-in. parison has a considerably lower cycle time. The estimated cooling time can be reduced from 17 sec for the original parison to 7 sec for the new design - a 59% reduction that has a significant impact on production rate.

The software can also display the distribution of area stretch ratio - i.e., how much different areas of the parison stretch relative to its original surface. Areas at the corners stretch significantly more than the other areas of the part. This stretch data can be mapped back to the original parison [ILLUSTRATION FOR FIGURE 3 OMITTED], indicating very precisely how to modify the profile of the 0.060-in. parison to achieve maximum uniform wall thickness. Furthermore, mapping back the area stretch ratio to the parison allows the parison to be programmed to be thicker in selected areas. For example, if the part specifications had required that the corners remain thicker, the parison could be programmed thicker in the area that blows into the corner without reverting to a thicker parison for the whole part. Although it was not included in the analysis in this case, modifying the thinned-down parison profile with the aid of mapped-back area-stretch data could have provided the optimal balance between material savings and part quality.

TUB THICKNESS ANALYSIS GIVES PEACE OF MIND

This thermoforming analysis is based on an actual ABS bathtub application analyzed by AC Technology and GE Plastics (Canada) for a customer. The C-Mold simulation was performed to provide more accurate thickness-distribution data as input for structural analysis.

As in the blow molding application, wall-thickness distribution data [ILLUSTRATION FOR FIGURE 4 OMITTED] was a key parameter in the analysis of the thermoformed tub. Wall-thickness data can be transferred directly from C-Mold to the Ansys finite-element structural-analysis package. As would be expected, the simulation shows the tub wall is thickest at the rim where there is minimal stretching and thinnest at the bottom corners and outside wall, where significant stretching of the sheet occurs.

Because of the tub's size, a relatively thick sheet was used to achieve the required final wall thickness. The thick sheet resulted in a long cycle time because of its cooling requirements. In Fig. 5, the areas in red take the longest time to cool.

As in the case of blow molding, thermoforming simulation can display the distribution of area stretch ratio in the finished part. The locations of the highest area stretch have the thinnest walls. Figure 6 shows the area stretch data mapped back to the original sheet. This helps the process engineer to adjust the profile of heater zones in order to optimize the wall-thickness distribution.

Stewart Barton is Engineering Services Manager at AC Technology North America in Louisville, Ky. The company supplies C-Mold software for workstations and PCs and performs simulations on a consulting basis.