John L. Johnson
   R&D Manager
   AMTellect Inc.
   State College, Pa.
   www.amtellect.com

Randall M. German
   Brush Chair Professor in Materials
   Penn State University
   University Park, Pa.




This 15 X 5 X 4-mm part is a F-15 (Kovar) electronic package. PIM holds flatness and hole diameter tolerances to 50 and 10 µm, respectively.

 

This 17-4PH stainless-steel optic-fiber connector has an effective density of 0.5 gm/cm3 (7% of theoretical). It was initially die cast out of zinc, but did not have sufficient mechanical properties. PIM provides a 30% cost savings over machining the part from wrought stainless steel.

 

A disk-drive latch was designed specifically for PIM. To reduce part mass, the section around the hole was hollowed out.

 

A schematic decision tree helps make a first cut at whether PIM suits a given design by considering annual production quantity, geometric attributes, and materials.

 

This component weighs 27 gm and if machined would require an initial volume of material over 15 cm3 (volume defined by the imaginary box). Accordingly, the effective density is 1.8 gm/cm3.

 

A scatter plot on log-log scaling shows the general range of PIM products ranked by mass and largest dimension.

 

Frequency distribution for different mass ranges in a random large collection of metallic, ceramic, and cemented carbide components emphasizes the success of PIM with lower mass components.

Most industrial design guides for powder injection molding (PIM) try to show how to create shapes suited for the process. But designers often focus on what's needed for the target form, function, and features. Consequently, components put out for quotation are designed for the customer's application, not for the production process. Indeed, the bulk of the design community seems to ignore most rules, suggestions on draft angles, and materials and charts of properties. Instead of learning design rules, designers should evaluate part qualities that justify PIM so they can take better advantage of its processing capabilities.

The drivers behind new design concepts include improved performance, social and market trends, and reinvigoration of existing products. Engineers can only transform new concepts into successful designs by figuring out the best manufacturing strategy. This may involve a compromise between the optimal design and traditional fabrication technologies.

Even designers familiar with the shape-forming capabilities of PIM may get confused by design-guide rules for specific features. It might be better to identify generic features of PIM parts that give clues to whether the optimal design is suitable for PIM. If it is, PIM vendors can supply recommendations for modifying the design to improve manufacturability.

PIM is increasingly used to reduce the manufacturing costs of automotive, aerospace, electronic, industrial, and medical components. The technology has matured to the point where ISO-9002 and QS-9000 certification can assure quality and delivery. PIM components made from many high-performance materials have properties comparable to those of wrought and cast materials.

The process

PIM begins with mixing a metal or ceramic powder with a polymer that gives the mixture fluid behavior. This compounded feedstock is formed into the desired shape by molding it into a cavity using a plastic-type injection-molding machine. After molding, application of heat, solvents, or a combination of both extracts the binder. Then the debound part is thermally processed, or sintered, to produce a high-density, metallic or ceramic component.

All of the process steps are highly interrelated. The manufacturing details can be left to PIM vendors, who constantly tweak their processes and can ensure high process yields and tolerances. Further, PIM vendors know which design factors lower tooling costs and improve quality. For example, a long molding cycle time is expensive because roughly two-thirds of molding expenses depend on time. Obviously, if design changes can lower molding cycle times, piece-part costs drop. Other cost-saving factors that molders can shed light on include assembly options, part combinations, process yields, and material availability.

With respect to tool cost, the mold can be machined with a single cavity, two cavities, or more. Each cavity boosts the initial tool cost that must be amortized over the production run. But multiple cavities reduce the variable cost associated with molding. Early consultation with a PIM design advisor can determine how many are best for the application.

Identify candidates early

The PIM process produces complex, net-shape components from metals, ceramics, cemented carbides, and cermets. Cost in PIM is largely a reflection of the component size, tooling cost and complexity, mold cycle time, debinding and sintering speed, and other relatively straightforward factors. As with plastic injection molding, many shapes are possible, but certain features greatly impact processing ease, yields, and costs. So how do you know a good candidate for PIM? A simple schematic decision tree may answer that question.

First, determine the annual production quantity. PIM historically has been well suited for production quantities from 5,000/yr (specialty firearm sights) to over 100,000,000/yr (cell-telephone vibrator weights).

If the target is in that range, next consider the engineering specification. PIM works best for three-dimensional geometries where there are at least 10 discrete specifications (dimensions, locations, and surface finish) on the engineering spec, but no more than 100. PIM components are in production outside this window, but they are the rare 1% exceptions.

Then comes material. PIM materials must be small grain-sized powders that sinter density without extraordinary processing cycles. For ceramics, this usually requires sinter-enhancing additives. For metals, it's best to avoid strong oxide formers as well as reactive, volatile, or toxic metals, such as beryllium, mercury, lead, manganese, and magnesium. PIM works best for materials with melting points above 1,000°C (1,832°F). For lower melting-point materials, die casting is often a better option. Typical PIM materials include stainless and tool steels, alumina, zirconia, titanium alloys, and nickel and cobalt-based superalloys.

Designs meeting these geometry and material criteria are PIM candidates. Cost becomes the next consideration. The small grain-sized powders can be expensive, so good PIM candidates tend to have high component manufacturing or machining costs that out-weigh material costs. Materials account for about 15% of PIM manufacturing costs.

Hard materials prove difficult and expensive to grind or machine. So applications that benefit most from PIM are those with materials that don't machine easily (tool and stainless steels, titanium, and ceramics) or that have difficult geometries.

Other factors include tolerances and surface finish. For rough surfaces, setup dominates the machining cost. But smooth surfaces require lengthy machining times, which become the dominant cost. Accordingly, there is a region of smooth surfaces between 0.4 and 5 µm, where PIM is most effective.

Effective density is another metric that makes a component a PIM contender. It is the mass of a component divided by the outer envelope volume from which grinding or machining would start. For example, suppose a component machined from a 15 cm3 block of material has a mass of 27 gm. Its effective density would then be 1.8 gm/cm3 (mass divided by the volume). This is less than 23% of the original block of material, assuming the minimum original block volume is available.

Such wasteful use of time and material adds to manufacturing costs, reduces yields, and may degrade product quality. But these are all areas where PIM can excel. Most PIM components are shapes with low effective densities, typically in the range of 25% of the bulk density - 2 gm/cm3 out of nearly 8 gm/cm3 for stainless steel, as an example.

In simple terms, performance needs dictate a component's practical cost. Consumer products tend to migrate toward low-cost plastics, while PIM is a favorite for higher-performance metallic and ceramic products in medical or dental devices, firearms, defense and aerospace systems, sporting goods, and appliance and industrial components. Telecommunication devices, hand tools, instrumentation and sensors, cutting tools, automotive engines, electronic packaging, or marine equipment are also suitable for PIM.

Geometric attributes

Many complex geometries can be fabricated via PIM, but certain component properties prove to be more attractive. Early identification of matches ensures better technical and economic success.

A logarithmic plot of the mass versus maximum size dimension for more than 220 PIM parts is one way to illustrate the geometric envelope that is open to designers. There is only a loose correlation, but the general trend shows mass increasing with the largest size. In the sample set, the largest component was 193 mm (7.5 in.) in length with a 43-gm mass, and the largest length-to-thickness ratio was 120. The median PIM component has a maximum dimension of just over 1 in. (26 mm) and a mass of 8 gm.

The mass distribution of several hundred PIM-steel and stainless-steel components serves as another aid for distinguishing potential PIM candidates. The mass distribution has two peaks. The first is associated with small structures under 0.5 gm, such as low-mass orthodontic brackets, while the second peak is in the 8 to 16-gm range. The largest mass was 1,097 gm (2.4 lb) with a maximum dimension of 137 mm (5.4 in.). When combined with the earlier statistics, it becomes evident PIM excels at forming smaller components where machining would remove a lot of mass.

There are larger components in production by PIM, but these are specialized structures. From an economic view, many factors work against PIM for large structures. First, the tooling is more expensive. Further, the equipment is larger, boosting capital expense. Also with large parts, the ebb and flow of heat is slower during molding, debinding, and sintering, thus cycle times lengthen with size. High capital expense and lengthier processing further burden each component with higher costs. Accordingly, amortization of these costs drives PIM toward higher production quantities and smaller components. An example is evident in the turbocharger applications, where PIM cost is 40% lower for the smaller automotive units, but is only 10% lower for larger diesel truck units.

On the other end of the mass range, PIM parts as small as 0.003 gm are in production. Such "microsized" metal and ceramic parts are difficult to produce in high volume by other fabrication means. For parts in this size range and smaller, several technical challenges emerge related to PIM tool creation, mold filling, part ejection, dimensional control, cleanliness, and handling. Recent advances in micromolding of plastics, however, have provided some of the technologies needed to overcome these barriers.

Another gauge of PIM compatibility is the typical number of call-outs or dimensions on the engineering spec. A simple shape, such as a washer, has only a few features, while a microcomputer circuit has millions of features; both would be poor applications for PIM. Common PIM successes involve several dimensions - a wristwatch case is one example that matches well with the technology. The most challenging components currently in production have 130 dimensional specifications. If a component requires tight tolerances, often PIM is used to shape the body, but critical dimensions are obtained in postsintering machining steps.

The process can lead to outstanding successes when the designer anticipates PIM fabrication and invites expert participation in early design decisions. Advice and comment by an advisor knowledgeable in PIM invariably leads to lower costs, better quality, longer tool life, and the reconciliation between intended function and PIM processing attributes.


This Fe-50% Ni pneumatic valve for a pump weighs about 180 gm and was originally a four-part assembly consisting of two powder-metal parts and two machined parts.

 

This cell-phone hinge was designed specifically for PIM. Maximum dimensions along all three axes are about 1 cm.

Ten design factors that justify PIM

The criteria for identifying a PIM candidate component can be captured in a few key considerations. These 10 important design factors help identify good PIM candidates.

  • Low mass/volume ratio.
  • Quantity above 5,000/yr, preferably above 20,000/yr.
  • Material difficult to machine.
  • Moderate to high complexity.
  • High performance requirements.
  • Specialized surface features or finishes.
  • Difficult combination of tolerances.
  • Consolidation of assembled parts.
  • Noncritical locations for blemishes, such as gates and ejector pin marks.
  • Novel compositions or combinations of compositions.

Many materials are available via PIM, but a few dominate the field and provide solutions over a broad range of engineering tasks. Best applications are components that have a low effective density, reflecting much material loss and time consumption in grinding or machining. Geometric features need not be sacrificed for PIM, but there are certain geometric qualities the technology handles quite well. These tend to be in smaller components (below about 25 mm or 1 in.) with three-dimensional complex shapes that are designed for the molding step. Some design features create problems and might best be added after sintering, while others can be added in molding without any real cost.

Tooling costs make PIM a better option for higher production quantities. Because the process is expensive, it is most cost competitive for high-performance components that can't be fabricated using plastics. Combining several components into one PIM piece brings benefits such as reduced part count and elimination of assembly operations. Further, when a smoother surface is desirable, PIM fares well against other fabrication routes.