Copy to clipboard 
You’ve developed a product using machined parts and things seem to be going well, but the part price is bogging you down. You know you could improve your bottom line and grow your market share if you could get some cost out of those components. Switching to molded parts could make a difference, but when is the right time? Numerous factors must be considered when deciding to go molded.
Consider why you developed the product using machined parts in the first place. Machining parts from solid stock was a quick way to develop and test your design. Although the per-unit cost was high, the outlay was minimal because quantities were low. The commitment was low, too. If a design change was needed, a change in the machining program would do the trick. This was a fast and easy approach with a low total cost.
Now the design has settled down. Changes are fewer and further between; the product is ready for (or is gaining momentum in) the marketplace and it would be nice to start making a profit. Now is the time to switch to molding.
Making the decision to use molded parts can be intimidating. What are the real savings and how can you be sure that the timing is right? The answer is the standard economist’s answer: it depends.
All other things being equal, the simple financial decision is how quickly the part cost savings will pay for the mold. The three most important factors in this analysis are part complexity, material selection and the quantity of parts needed.
The cost of the mold is greatly influenced by part design complexity, which is also a key machining cost driver. Undercuts, side-actions (features not in the mold’s line of draw), threads and fine details increase mold cost. However, because these features also increase the cost of machining, the molded unit cost becomes more attractive as part complexity increases. The decision now depends upon material selection and part quantity.
Material waste plays a significant role in cost considerations, especially with more expensive grades like implantable PEEK. With some materials costing upwards of $4 per gram, your part cost could double due to the amount of material that is machined away. The molding process generates some waste, too, but is usually a much more efficient use of material.
Justifying the up-front mold cost with molded part price savings and material savings then becomes a question of quantity. Since the cost of each molded part is significantly less than a machined part, you must determine how many parts it will take to pay off the mold. You may be able to justify molded parts with as few as 100 parts per year if the machining costs are high or if machining generates significant material waste. Often, the expense of building a mold can be recovered by part cost savings within the first year of production—sometimes in six months or less.
Now to address the “all other things being equal” assumption. (Hint: All of the other things are not equal.)
• Stable design. The cost to revise a mold is generally greater than the cost to adjust a machining program. This is especially true if the added or revised feature is not in the mold’s line of draw. Prototype tooling is a great option when the part design is still in development. A “steel safe” approach leaves selected part features small in the mold so that revisions to thicken them require removing steel—a much less expensive proposition than adding steel.
• Validation costs. Since validation and submission costs are significant, they should be considered early in the process. A valid strategy may include machining parts for development and mechanical testing, and molding parts for design validation,
submittal and market launch. In some cases, the best solution is to mold a near-net shape and machine any part details that would increase the mold cost.
• Surface finish. While it is true that material in bar-stock form has typically lower residual stress than a molded part, machining operations remove the smooth resin-rich surface and add stress risers in the form of micro cracks—cracks that are not present in molded parts.
• Knit lines. Molded parts will have knit lines (sometimes called meld or weld lines) whenever the material flows around a hole and re-joins on the opposite side. With proper part design, mold design and scientific molding process, knit line strength will approach the strength of a non-knitted feature. Additives such as glass fiber may compromise knit line strength as the fibers do not efficiently cross the flow fronts, leaving a resin-rich (glass-starved) zone. In many cases, revisions to the part design or the addition of an overflow feature can minimize this effect. This issue must be thoughtfully considered if your strategy includes machining parts for testing and molding parts for production.
• Cleanliness. Machined parts are susceptible to burrs and particulate—requiring secondary deburring and washing operations.
Packaging a molded part right out of the press avoids these potential sources of contamination.
• Multi-material options. It is becoming commonplace to combine multiple materials into one part without the need for assembly.
This molding technology provides the flexibility of multiple material properties (such as strength plus a resilient seal) without
leaving the cleanroom.
• Lead time. Parts can be machined in a matter of days, but prototype molds can also be produced in a short lead time, especially if the part geometry is not too complex. Molded parts, however, win the leadtime challenge as quantities increase.
Machining parts might have been an effective way to work out the bugs in your product design. But when the pressure is on to increase part quality, cleanliness, lead time and profitability, it’s time to consider molding.
Jeff Randall, PE, is Vice President of Engineering at MRPC, a manufacturer of silicone, medical rubber and thermoplastic components and assemblies for the medical device industry. He received degrees in mechanical engineering and business administration and a Master’s certificate in project management from the University of Wisconsin-Madison.
www.mrpcorp.com
