Originally Published MDDI March 2004

Cover Story

To maximize the benefits of injection molding for production of parts made from engineering plastics, designers must
balance many interrelated variables.

Hedden Miller and Ronald R. Lamonte

Most plastic parts in medical devices and diagnostic equipment are formed by injection molding. This process, which can involve temperatures above 600°F (315°C) and pressures that reach 30,000 PSI (2040 bar), balances many part-design, resin, and molding factors to create sophisticated components that withstand the rigors of medical use.

Despite its complexity, injection molding is simple in concept: plastic is melted, injected into a cavity, and removed when it has hardened. As shown in Figure 1, the machines used have an injection section that first plasticizes, then pushes the resin into the mold, as well as a clamping mechanism that opens and closes the mold. The process is economical, efficient, and precise, and yields little waste. In designing plastic components, medical manufacturers need to account for the intricacies of this processing method and the considerations involved in its use.

How It Works

Injection molding starts with small plastic pellets, usually about 1¼8 in. (3.2 mm) in length. These pellets are fed from a hopper on a molding machine into a reciprocating screw, which carries them through a heated barrel. Shear generated by the flights of the screw melts the plastic as it is conveyed toward the front of the barrel and mixes the melt so it is uniform. 

The screw retracts as molten plastic accumulates in front of it. When enough plastic accumulates to fill the mold, the screw is pushed forward hydraulically. This forces the melt, or shot, through the machine nozzle and into the closed mold. 

In the mold, the plastic flows through channels called runners and passes into part cavities through gates. Water or another fluid circulating through a cooling system in the mold extracts heat. The plastic is held at high pressure until it solidifies, or freezes off, at the gates. After parts have cooled and solidified enough to be handled, the mold is opened and the parts are removed. 

As parts cool in the mold, the next shot is plasticized in the barrel. After ejection, the mold is closed and the process is repeated. Because it is usually the longest step in the process, cooling time often determines cycle time and, thus, production rates. 

Medical designers can choose from among many variations on the basic injection molding process. Some of these include: 

•Double-shot molding, a two-step process that injects one color or material first. After the first shot hardens, a second color or material is injected into or around the initial shape. 
•Push-pull molding, which involves multiple layers having different orientations. Part properties are more uniform than if they were molded from just one direction. 
•Fusible cores, which create complex, hollow structures. Heating the part after molding melts the core, which flows out of the part. 
•Gas-assisted injection molding, in which an inert gas (often nitrogen) is forced into the melt as it enters the mold. The gas packs the plastic into the cavities to reduce cycle time, part weight, warpage, and stress in the cooled part, and minimize other problems.

Balancing Variables

In working with injection molding, designers and manufacturers must pay attention to many interrelated variables in order to create viable medical parts. Each aspect of the process, from part design and the resin used to the machine and the conditions chosen, involves numerous choices. 

•Moldability is often a key factor in choosing a resin. How well a plastic flows, transfers heat, and shrinks as it cools affects the molding process. Viscosity, for instance, influences cavity filling and the type, size, and placement of gates, as well as the temperature and pressure used. 
•Heat-transfer coefficient affects how the cooling system is configured to prevent part warpage from differential cooling and shrinkage. Shrinkage is also important in sizing mold cavities so part dimensions fall within the tolerances set.
•Shrinkage is influenced by injection temperature, how quickly a part cools, and orientation (how the molecules and/or any reinforcing fibers in a resin align as it flows into a mold). It is also affected by a part's wall thickness; e.g., thin areas shrink least. Shrinkage can be the deciding factor in material selection. 

Selecting a molding machine and determining the optimum processing variables to produce a specific part also involve a great many variables. These range from barrel temperature, injection pressure, screw speed, and shot size in the injection section, to cavity pressure, mold temperature, gate size, venting to allow the escape of gases, the clamping force that holds the halves of the mold together, and much more.

Tool Design 

Mold tools form the heart of the molding process. They should be designed to allow as broad a processing window as possible so such variables as hydraulic pressure, barrel temperature, and screw and barrel wear can shift over time without harming part quality. Tool design must account for how a resin enters a cavity, what happens once it is there and how the part can be released smoothly. Elements to be considered include gating, draft, runners, cavity number, slides, cooling channels, placement of parting lines where the mold opens, ejectors, and the metal used for the mold. 

Gates should allow the molten plastic to flow smoothly to the limits of the cavity. Designers adjust gate type, size, and position to account for factors such as cavity pressure, mold-filling time, and how resin molecules and fibers align to strengthen parts. Gates can be moved to hide flow lines that form on a part's surface as resin passes through a gate or to relocate weld lines (weak areas where two or more melt streams meet after flowing around a core) to noncritical areas. 

Many part elements are given a slight taper, or draft, in the direction the mold moves as it opens. This enhances removal of parts from the tool after the resin shrinks and cools. The draft is usually less than 3° per side. It can be affected by surface finish (the higher the polish, the less the taper), the length of the projection, and resin shrinkage. 

The runner system affects how much scrap is produced. Conventional cold runners create a lot of scrap, which is often reground and reused. Hot runners yield no scrap, but cost more and are harder to operate. They are especially useful when runner volume is large relative to part size. 

The number of cavities per tool varies with part size and production volume. In other words, higher volumes and smaller parts usually mean more cavities per tool. Precision parts are often made in tools having fewer cavities. 

The cost of mold tools is driven up by the need for special features. Parts having deep recesses, screw threads, or holes or depressions perpendicular to the direction the mold opens, for example, may call for unscrewing or collapsing cores, side-action slides, multiple plates, automatic unscrewing devices, intricate parting lines, and other release mechanisms. 

In one case, special tool features were essential in molding the lid of a disposable surgical device from Stryker Instruments (Kalamazoo, MI) that mixes and applies polymethyl methacrylate bone cement (see Figure 2). The device has three injection-molded elements made of a cyclic olefin copolymer (COC): the lid, a mixing chamber and a delivery cartridge. The 2-in. (51 mm) diam mixing chamber screws onto the lid that also has a vacuum port with a Luer fitting. The tool for the lid required three unscrewing cores for its three threaded portions. Also, since the COC used has almost no mold shrinkage, draft was carefully adjusted to prevent parts from sticking in the mold. 

Special features were also important for the Y-shaped manifold that ties key components together in an ultrasonic imaging catheter made by Boston Scientific (Natick, MA) and used to detect aneurysms in arteries (see Figure 3). Molding this part involved an intricate tool and two fragile core pins with diameters that match the inner diameter of the insert-molded lumen tubes. The pins are inserted into the tubes and extend up the arms of the Y so the lumens do not collapse as resin flows in the mold. The manifold is made in a single-cavity tool with one gate and a cold runner. 

Design Aids

The complex interplay of a great many variables in the injection molding process poses severe challenges for designers. Just having to meet tight tolerances, for instance, affects the resin and fillers, mold cavity layout and the gating, cooling system and release mechanisms chosen. The molding machine itself also affects tolerances in terms of how well it controls temperature, pressure and clamping force. 

Designers cope with the complex, often nonlinear task of configuring mold tools by using computerized approaches like finite element analysis (FEA) for structural and mold-filling analyses (MFA). Linear structural FEA can be used to determine displacement and various stresses and strains in a part. 

MFA evaluates filling pattern and pressure and temperature distributions to optimize resin flow, gate position, the location of weld lines, and other factors. Other numerical tools include mold cooling analysis, which gauges the effectiveness of the mold temperature distribution system, and shrinkage and warpage analyses, which evaluate dimensional control and predict molded-in stresses and warpage. 

Mold flow analysis was essential for the designers of an element in a syringe pump from Alaris Medical Systems (Basingstoke, UK) that meters and dispenses drugs during drip feeding (see Figure 4). Alaris used a polybutylene terephthalate (PBT) to replace five metal and plastic parts in the declutch tube of its P-series syringe pumps with a single injection-molded component. The 173-mm-long declutch tube on this electromechanical pump disengages 5-to 100-ml syringes from the unit's transmission.

The tube is injection molded around a core pin. Mold flow analysis was used to ensure that the cavity would fill without problem. This involved evaluations of the melt front, temperature and pressure across the mold, cooling time to define cycle time, and assessment of weld lines. In molding, the PBT enters the cavity from a gate at the base of the component and gives an even fill in just over 0.5 seconds. Mold cooling was adapted to ensure that the core pin remained straight as the part hardened. 

Design for moldability can involve close coordination between the designer, molder, and resin supplier. This was the case with the complex ergonomic handles for a minimally invasive surgery forceps made by Surgical Innovations Ltd. (Leeds, UK). The handles have eight molded elements, including a cone that transmits rotation and a complex central core with stainless-steel inserts (see Figure 5). The core supports the other elements. 

The elements in the handle are molded in glass fiber– reinforced polyphenylene sulfide, which meets tolerances as demanding as 0.002 to 0.003 in. (0.05 to 0.08 mm). Computer-aided flow and structural analyses for the core demonstrated that the thin-walled cone would fill out in molding. Gating for the core was adjusted so the resin was fed from the end to control weld lines and stress (most cylindrical parts are fed from the center). 

Wall thickness is important in designing for moldability, especially because of its effect on cycle time. Too thick a wall, even in just a small portion of the part, can lengthen cooling time and increase part cost. If stress or deflection under load is high, ribs or other reinforcing features can be added to build strength without thickening walls. To avoid a heavy mass that can extend cycle time, several thinner ribs are better than a single large rib. 

Mold design should account for how a part will be assembled. Many assemblies use snap fits to join parts economically and rapidly. These have molded elements that flex and return or nearly return to their unflexed positions. Snap fits generally need undercuts, thus requiring molds with side action features. If mated male and female threads are used, the internal threads will need an unscrewing or collapsing mechanism. External threads can often be molded by splitting them across the parting line.

Conclusion

As medical devices grow more sophisticated, designers are asking more from the molding process and materials that they use. The LabCD from Tecan (Männedorf, Switzerland) is a good example of a unit that pushed the limits of molding technology. 

This disposable, microfluidic device performs 48 drug-screening assays simultaneously. The disk is 5 mm thick and 124 mm in diameter and has upper and lower halves that are bonded together. One half has microchannels as small as 50 µm in width and depth; the other contains storage and measurement wells. Tecan used COC for the disk because, in part, it is an optically clear material that replicates the microchannels and reservoirs with a high degree of accuracy. 

In processing, the molder, Weidmann Plastics Technology (Rapperswil, Switzerland) used a proprietary injection molding process and a modified press to form the microstructures. Weidmann developed special mold inserts having 50-µm-deep channels by adapting advanced etching and micromachining methods. This process gave such fine control that a draft angle between 1° and 3° was added to the channels. The inserts are made of a nickel alloy with a Rockwell hardness of about 50 so that they hold all critical dimensions. 

Whether it involves a straightforward part or one that extends injection molding into new territory, the process finds extremely broad use in fabricating elements for medical devices and equipment because it is often the most efficient and lowest-cost way to make a component. In order to make the best possible use of this process and to gain the high performance needed in medical products, manufacturers must understand the process in detail and the interplay of the many part, resin, and molding factors it involves. 

Hedden Miller is marketing manager for engineering thermoplastics and Ronald R. Lamonte, PhD, is a development associate for Ticona (Summit, NJ).

Copyright ©2004 Medical Device & Diagnostic Industry