Sponsored By

August 9, 1998

12 Min Read
Better closures through design and analysis -- Part II


Figure 1. In this FEA simulation, plug draft is 1.8°, while bottle draft is 1.0°. This combination spreads the contact over a larger area. The result--lower contact pressure--is indicated by the graph lines (showing both magnitude and direction of the pressure) emanating from the contact point. At the interface between the cap and bottle, dark blue lines show pressure is at its lowest.

When the product you are designing will be molded in production quantities measuring billions, it makes sense to get the design right the first time. And when those same products are closures that must get to market quickly, your goal will be to shorten the time it takes to get that optimal design. Fortunately, help is available. Predictive, nonlinear structural analysis methods such as finite element analysis are the keys to accelerated development, evaluation, and improvement of closure system performance.

Editor's note: Ready for more insights into designing better closures? Part I (see July 1998 IMM) gave concrete details on how to design for sealing ability, select compatible materials, and understand the basics of each common closure type-plug, claw, gasketed, and liner. In the second and final part of this article, the team of authors-Clinton Haynes, Margret DeYoung, and Jay Yuan-from Stress Engineering Services (Mason, OH) details the results of their finite element analysis projects and the design recommendations that grew out of this research.

There are hundreds of subtle, and not so subtle, geometric variations that can be introduced into the design of a closure system. Conventional trial and error methods of development don't offer any information regarding the local contact pressure between mating components. This is the critical information needed to evaluate and improve closure systems. Another reason to use computer aided analysis: project schedules and budgets are generally exhausted long before all the various good closure concepts have been investigated. Simulating designs via computer cuts down on the time required to consider all the viable concepts developed. It also allows you to evaluate their performance before selecting the most effective design.

Analysis reveals different optimum geometries for different seal types. Also, a comparison of each seal offers some considerations on where they should be applied. These results can be applied to specific closure systems by designers looking to get it right and make it snappy


Figure 3. Molding variablility can affect more than just sealing performance. These graphs illustrate how variability can change the overall performance of a closure system.

Plugging Away?
Plug seals are particularly versatile because they offer a great deal of axial (rotational) compliance. If properly designed, sealing can be achieved over a broad range of axial travel (or cap rotation) requiring relatively little strength on the user's part. This feature is particularly attractive for child-resistant closures where the timing between two independent actions is needed.

Specific design attributes of the plug depend on the bottle molding process. The bottle finish ID for injection blown containers can be carefully controlled, but containers produced on a shuttle or wheel machine generally have more variability, particularly when the containers are made of glass. It is especially critical to specify the draft on the finish ID for an IM preform or a container manufactured on a shuttle machine with a blow pin. The key issues are:

  • ensuring localized contact pressure,

  • controlling application torque, and

  • avoiding brittle material failure of the cap.

Blown polyethylene finishes, typical of a wheel machine, are less critical provided they are machined with a post reaming operation. Glass variability is generally similar to that found with a wheel-produced, blowmolded bottle.

Plug seal draft should be specified so the point of contact between the bottle and cap remains localized near the tip of the plug. Using nonlinear finite element methods, designers can analyze the effectiveness of a design to be sure this condition exists during assembly, before the closure is produced. Initial contact is made near the tip of the plug, and with the proper draft angle, contact remains more or less at this location throughout the assembly process. FEA methods can also be used to predict the application and removal torque for closure systems while still in design. Selecting draft angle combinations that do not offer this stress localization cause lower contact pressure and, thus, leakage (Figure 1).




Figure2. If improperly designed, plug seals can cause movement at the contact surface during assembly. The simulations above illustrate this condition, in which plug draft is 5° and bottle draft is 1°. FEA simulation can help designers avoid this problem.

To illustrate further, when the taper on the plug is 5° and the taper on the bottle is 1°, the peak contact pressure shifts from the central region of the plug to the intersection of the top of the bottle ID and the plug OD upon assembly (Figure 2). This drives the peak stress into the most significant geometric stress concentration on the cap, leading to closure cracking due to ductility exhaustion, accelerated environmental stress cracking (ESC), as well as significantly higher application torque levels. Torque increases because the stiffness of the plug is a cubic function of plug length. Therefore, the plug stiffens dramatically as the contact point moves closer to the top of the cap.

Plug and bottle contact geometry is only half of the sealing picture. Magnitudes of contact stress and pressure caused by interference are also important to a successful design. The definition of yield for plastics, for our purposes, is the maximum stress at which drawing begins. For conservative, robust design, it is recommended the closure be designed to achieve a minimum contact stress equal to the yield stress of the most compliant material (either the cap or the bottle material) at the point of contact for the full range of dimensional variability.

For extrusion blow molded bottle finishes, where dimensional variability can be as much as ±.010 inches on the diameter, selecting the appropriate dimensions can be a challenging task. One recommended process for selecting dimensions is to calculate the OD of the plug and ID of the bottle needed to achieve yield for the minimum interference condition. Remember, the scale of the yielding, gross local vs. microscale behavior depends on the specific application. After defining the minimum interference condition, determine nominal and maximum dimensions based on the expected variation and the desired quality level (Figure 3). For finishes with large variation (or standard deviation), the maximum interference can be significant. Analysis can be used to determine if, for the maximum interference condition, the material ductility is exceeded, creating the possibility of accelerated ESC problems or ductility exhaustion and cracking due to material creep.

Claw Seals
Functionally, the claw seal is an axial seal where the claw feature on the cap makes contact with the top of the bottle finish. Since there is no axial "stop," peak application torque on the closure occurs when applied torque and interfacial forces between the deforming claw and bottle top reach equilibrium.

Several problems can arise with this style closure:

  • To reach the recommended condition, where the claw seats more or less in the center of the finish cross section, designers need to plan ahead. If there is an unacceptable level of clearance between the T & E dimensions on the bottle and cap, the claw can slip off one side of the bottle finish. This can cause leakage because no method of radial alignment between the bottle and the cap is provided.

  • Claw-style closures are susceptible to leakage because they do not accommodate surface defects well. This is true for containers produced on a wheel machine. The relatively low structural stiffness of the claw and the typical quality of the sealing surface produced by a blown finish on a wheel machine can result in high leakage rates. Solution? Post-machine the finish to face the top of the bottle and remove the parting line defect. This is true of containers produced by a shuttle molding machine where the blow pin forms the ID of the finish. Claw performance on an injection blown finish is significantly improved relative to other blowmolding processes.

  • A high percentage of the material at the tip of the seal can yield. FEA results show that, at times, the entire cross section will yield. As a result, there is little follow-up force to maintain contact pressure between the bottle and the closure as the material relaxes with a corresponding drop in contact sealing pressure that doesn't recover with time.

Various claw design concepts incorporate displacement stops (Figure 4). These features tend to limit the volume of material that yields for high levels of application torque. However, problems associated with defect sensitivity and misalignment are not resolved. For this and other reasons, claw seals are best suited for containers that will hold viscous fluids.

The most difficult part of designing a claw seal is balancing the flexibility of the claw with the ability to generate high local contact pressure to create a seal. The solution is a function of both geometry and material selection. Unlike plug seal designs, there are no robust design rules. Predictive analysis methods are the best way to evaluate how well the balance has been achieved on a case by case basis.




Figure 4. Adding a displacement stop to claw seal deigns helps to limit material yielding at high levels of application force. These three concepts along with their respective predicted plastic strain contours shown in red are examples of this technique.

Nub Seals
Nub seals are the predecessor to claw seals and are intended to function in a generally similar fashion. The key difference between the claw and nub is the nub seal is based on the premise of very high local contact stresses to generate significant amounts of plastic strain.

The nub closure was originally intended to dig itself a seating trench in the top of a plastic bottle finish. For a glass bottle finish, the goal of the nub seal is to plastically deform itself so the deformations can fill in irregularities that are present on the sealing surface. As materials and packaging technologies evolved, nub seals are less prevalent as the primary sealing feature. They have given way to plugs, claws, and gasketed systems. However, they still see application, doubling as secondary seals, application torque limiting features, and stress localization features on closures using a liner.

Nub seals present problems with compliance. These seals offer little sealing capability until they have reached their fully loaded condition, which occurs over a minute range of angular cap rotation. This deficiency is further highlighted in production manufacturing operations, where the application torque variability can vary 30 percent or more from the nominal torque setting. This type of seal is also sensitive to nonplanarity of the sealing face of the bottle as well as its own planarity.

Similar to claw seals, the challenge of nub seal design is generally a balancing act between the cross section of the sealing feature and the yield characteristics of both the cap and bottle materials. The details of design concepts are best worked out via nonlinear finite element simulation. Also, nub seals are better suited to a supporting role rather than to the primary sealing mechanism. Another problem with the nub is its sensitivity to planarity. Because the seal is stiff with minimum axial compliance, it does not accommodate out-of-plane geometry, such as warpage or sinks.

Liners and Gaskets
Any third material added to the sealing system between the bottle and closure constitutes a gasket or liner. EVA, plastisol, Shell's Kraton, Concept's C-flex, AES's Santoprene, silicone, rubber, and paper-based materials are commonly used to create a gasket.


Figure 5. Thermoplastic elastomers make excellent gasket materials, especially for medical products where microbial sealing is necessary. The TPE gasket (light colored rectangle) deforms to accommodate a wide range of defects on sealing surfaces.

Liner performance can vary dramatically depending on the material selected and the environmental conditions it will experience. Thermoplastic elastomers generally offer excellent sealing capability over a wide range of torque variation and dimensional variability. This class of materials also does an excellent job of accommodating defects on bigger sealing surfaces that are much larger than anything possible for plastic/plastic seals. For this reason, TPEs are quite often used in medical product applications where microbial sealing is necessary (Figure 5).

Figure 6 shows a finite element analysis simulation of a closure design featuring a silicone gasket as the primary seal and a plug as a secondary seal on an injection molded finish. Because of the relatively low Young's modulus of the silicone gasket (1000 psi) and the high deformations of the gasket, the contact pressure distribution is much broader than the distribution on the plug.


Figure 6. Nonlinear FEA simulation (A, B, and C) shows how a closure system using a silicone gasket as primary seal and plug as secondary seal behaves during assembly. Contact pressures just after assembly (D) for the gasketed seal are better distributed compared to those for the plug seal (E). This occurs because silicone has a low Young's modulus (1000 psi) and because of its high deformation.

When designing sealing systems that utilize TPEs, primary technical hurdles include potentially high removal torque values, significant material relaxation at elevated temperature, and the difficulty of handling small, soft components for assembly. The handling problem can be addressed with advanced molding processes such as two-shot molding. But, as in the case of the other sealing systems, the details of the geometry and its response to assembly and thermal loading is best evaluated using nonlinear FEA methods. For elastomeric-type seals, this includes evaluating the level of constraint placed on the gasket by the molded cap. Structural features that either stiffen or soften the response of the gasket as the finish makes contact can also be evaluated.

Contact information
Stress Engineering Services
Mason, OH
Clinton Haynes
Phone: (513) 336-6701
Fax: (513) 336-6817
Website: www.stresseng.com

Sign up for the PlasticsToday NewsFeed newsletter.

You May Also Like