This technical information has been contributed by
Proto Labs, Inc.

Rapid Injection Molder's Automated Process Creates Strategic Advantages for Clients

Injection Molding Plastic Prototypes

A software-driven contract manufacturing company has "written the book" on rapid injection molding - how to go from 3D CAD to functional plastic parts in as little as a day. Now, it's looking to apply the advantages of its technology to new areas of product development.

Since its founding in 1998 by Larry Lukis, Protomold has sought to "radically change the economics and lead times associated with prototype and low-volume production of injection molded parts." Thanks in large part to the time-saving advantages of its proprietary software, Protomold has swiftly achieved recognition as a fast-growing provider of rapid injection molding services, a market segment positioned between rapid prototyping and high-volume injection molding.

Earlier this year, Protomold and First Cut Prototype, a startup launched by Protomold to provide CNC-machined plastic prototypes, became divisions of Proto Labs, Inc., a new corporate entity that will continue to explore and develop new applications for the company's proprietary software technology. Proto Labs recently acquired a new 96,000-square-foot facility in Maple Plain, Minnesota to house the headquarters and rapidly-growing production operations of Protomold and First Cut Prototype. All told, Proto Labs now occupies approximately 160,000 square feet of combined manufacturing and office space in Maple Plain. The company also has a fully-operational manufacturing facility in England with 50 employees, and has plans to open another plant in Japan next year.

"The successful launch of the First Cut Prototype division has demonstrated that the company's software has applications well beyond our rapid injection molding origins," said Brad Cleveland, president and CEO of Proto Labs, Inc., in a statement announcing the new facility. "The new Proto Labs name reflects the company's broadened mission to help product developers get functional prototype parts quickly, easily, and cost effectively."

Both of Proto Labs' divisions—Protomold and First Cut Prototype—provide quick-turn prototypes and low volumes of plastic parts for applications across all industries. Both offer their customers the strategic advantage of extremely quick turnaround from 3D CAD files to real, plastic parts. "Our biggest strength is the speed with which we can ship real injection-molded or CNC machined parts—as fast as one business day in some cases," says Cleveland. "We've automated the process of going from a customer's 3D CAD model to a highly illustrative, interactive quotation, and then to the manufacture of real, plastic parts, faster than anyone in the world."

Protomold's motto is "Nobody's Faster in the Short Run!®". Design-2-Part Magazine's Rich Novicky recently spoke with Brad Cleveland about the company's efforts to make its vision a reality.

D2P: Protomold's website says, "Our single focus is to provide new product designers with the easiest, fastest, and least-expensive way to obtain low volumes of plastic injection-molded parts." Conventional wisdom usually indicates that easy, fast, and least-expensive are mutually exclusive when it comes to manufacturing, particularly when you're not able to spread tooling costs over large volumes. How is Protomold able to accomplish this?

BC: We have dramatically accelerated—streamlined—the process of manufacturing injection mold tooling. By nearly automating the process of designing and manufacturing aluminum injection molds, we have been able to significantly reduce its cost, and dramatically accelerate the speed with which it can be made. And it's metal tooling, so you get real injection-molded parts significantly faster than via conventional molding. We can also make these tools at very low costs and pass those savings on to our customers. We have some limitations on the size and complexity of the parts we can make, but if it's a part that we can make, nobody can turn the parts around faster. Or if the volume is low, like under a thousand parts, very seldom can anyone else make them less expensively, either.

So, the answer to your question is we are both fast and inexpensive due to the automation of the design and manufacturing of the mold tooling.

D2P: So, the trick is in the tooling?

BC: The trick is in the way we've automated the process of designing and manufacturing injection mold tooling, right.

D2P: Can you give us any background on how you're able to do that?

BC: Sure. We actually developed our own proprietary software that runs on what we call a "compute cluster." We actually have a number of compute clusters. So we have proprietary software that runs in a parallel processing environment, the software and the environment both Proto Labs-designed and developed. And the software almost entirely automates the process of converting a customer's 3D CAD model to a manufacturing plan for the injection molding tooling. So there is an up-front requirement that our customers must have a 3D CAD model that they prepare. Our software analyzes that and prepares what's called a ProtoQuote®, which is a very sophisticated, interactive web-based quotation that provides a moldability analysis and pricing options interactively. When our customer decides that it's something that he or she wants, they order over the internet and send us a PO. Then our software continues with the process and designs a mold, generating tool-paths for our 3-axis high-speed milling machines. So, nowhere in the process is there an engineer designing a mold, nowhere in the process is there a toolmaker involved. We have removed the engineering and toolmaker requirements for making molds.

D2P: So basically, you're taking the customer's specs and putting them directly into the CAD program?

BC: Exactly. Technically, we don't put them (the specs) into the CAD program; we put them into the tool paths for 3-axis high-speed mills, which is more accurately a machine command file. And the intermediate steps are the quotation and the development of the mold design, all of which takes place inside the compute cluster, and then the generation of the tool paths from that. But in a matter of minutes we can go from a 3D model to a tool path, where conventional toolmakers will spend days or weeks doing the same thing manually. That's the front end of our process, which is very, very streamlined. And then we have an exceptionally fast, lean manufacturing process for assembling the molds, doing the injection molding, and getting the parts out the door.

D2P: You mentioned that your process uses 3-axis milling. Is that a limitation, or can you handle more complex parts as well?

BC: Well, today Protomold only uses 3-axis CNC mills to make molds. It is not a fundamental limitation; we can extend that to 5-axis, and that's actually in the plan. But you can make some very complex shapes using only 3-axis mills. We can have very complex parting lines on our parts, and we can support up to four side actions on a mold per part.

D2P: Side actions?

BC: Imagine a part that has an undercut: If a part has undercuts, you need to have some kind of a moving action on the mold to pull the metal that's providing that feature out of the way before the mold opens for part removal. Those are generally known in the industry as side actions. The mold opens up in one direction, and another action in the mold would move in a different direction to produce the undercut. And we can do up to four side actions per mold; and we can do very complex shutoff surfaces and parting lines. So you can get some very complex parts despite these limitations.

But that is certainly the primary limitation of our process—the complexity of the parts. What our software does for us, upfront in the quoting stage, is analyze the part for the customer. And in the ProtoQuote®, it gives them a 3D color-coded, annotated illustration of their part that they can rotate around, zoom in on, et cetera, to see exactly what our software is telling them they need to change to fit within our process. In other words, if we identify an undercut, we'll show that to them. If it's an undercut that we can't support for some reason, let's say it's an internal undercut which we cannot support today, it'll show them that. It'll say you need to remove this feature if you want to order from us.

In addition to helping them understand the impact of our limitations on their design, we also give them a significant amount of moldability analysis. This is not to help them design a more functional part, that's not our role, but it is to help them design parts that are moldable, not just at Protomold, but any place. There are a lot of design "rules of thumb" in injection molding: You want a consistent wall thickness, you want a certain radius on fillets, you want to have different design techniques so that the part will fill properly, will eject from the mold smoothly, et cetera. Draft is a big concept in injection molding. So we provide a lot of that advice to help our customers design a more moldable part, which is really useful because a lot of our customers haven't done a lot of designing of injection molding parts before.

So they may come to us with a design that's very thick because they currently machine the part, and we go back and we say this design really isn't well suited for injection molding—it's too thick here, it's not consistent there, et cetera. So the bottom line is that our ProtoQuote® provides guidance with respect to the limitations of our process, and it provides generic injection molding guidance as well.

D2P: Because of the nature of your business, does Protomold find itself concentrating on specific industries or applications?

BC: I often say the only industries that use Protomold are industries that use little plastic parts [laughs]; in other words, pretty much every industry. You name any industry, and I can name a plastic part they use that Protomold can handle. Having said that though, we have thousands of customers, thousands of active companies just in the United States, plus we have a very fast-growing business in Europe as well. But just in the United States, we've got thousands of companies that buy prototype or low-volume production parts from us on a regular basis. I would say that if you actually did an analysis, you'd probably find that the largest identifiable chunk is probably associated with the medical industry. It may be that 15 to 20 per cent of our business is, one way or another, medical.

And then there is an enormous amount of business for mechanical equipment manufacturers, like cams, switches, and gears. We also do a lot of enclosures for hand-held devices, generally appliances or electronics. There's also a lot of work we do for consumer products. Say someone has a new product for cleaning out a toilet bowl. We would make the prototype handles for that before they go into high-volume production. Those are some of the industries we're in.

D2P: What types of plastics does Protomold work with?

BC: Well, we will do injection molding in almost any thermoplastic, which means hundreds and hundreds of different thermoplastics. The limitations that we set on it though, are we do not work in materials that have a very high melting temperature, or that are highly corrosive. A couple of examples of things that we will not do: PEEK™, we do not do Teflon®, and we do not do PVC. Those are some examples of things that we often turn away, because we manufacture aluminum injection molds, and although those materials are doable in aluminum injection molds, they tend to be very, very difficult to work with. In many instances, it doesn't work as well as we need it to, because we're so focused on speed. So we say "no" to those materials and others like them, for those reasons.

D2P: You mentioned that Protomold provides moldability information. Does that also include making suggestions to your customers on types of materials?

BC: First of all, we never provide design assistance to our customers to help them make a more functional part, because that's our customers' job. We do have some companies that we refer our customers to if they need help with their design. In terms of materials, we do have some information on our website, and we have customer service engineers that are very knowledgeable in the different materials.

So, occasionally, if a customer asks if we have any materials to recommend, we refer them to our website and we might share a little experience with them. But we tend to let our customers do what our customers do, which is the designing and selection of the best material for their specific application, and then we just try to give them advice on how they might want to tweak their design to fit within our process, to make sure it's moldable. So there's some of that advice, but mostly it's just the information that's on our website, which is actually pretty extensive. In general, we stay out of the design world and just focus on making parts.

D2P: The plastic injection molding and rapid prototyping industries can be very competitive. Why do your customers choose Protomold, and what keeps them coming back?

BC: Let's talk about rapid prototyping for a second. First of all, we don't specifically consider what we do to be rapid prototyping. Protomold makes prototypes really fast, but in general, rapid prototyping technologies are essentially systems for "printing" in 3D. That's not what we do—we do rapid injection molding. The difference is, the reason people come to us instead of buying rapid prototypes, is because what we make are real parts [not concept models]. We can actually make the parts using injection molding, which is the production manufacturing process they want when they're actually going to market, and we can make it in the same material they want their finished product to be.

Every single rapid prototyping process out there has dramatic limitations on the materials it can work with. And our "real" injection molded parts also have the surface finishes the customer wants—we can put textures on surfaces, we can have smooth surfaces, et cetera, whereas rapid prototyping is a really cool technology, but they're not "real" parts; they're prototypes.

We recently surveyed our customers, and about 80% use rapid prototyping, and then they use rapid injection molding. Engineers are smart. They know the value of getting a prototype done rapidly, and when they're ready to do testing or take it to a trade show, or low-volume production, then they use rapid injection molding for the next step. Then if they eventually need a million parts, they go to a large-volume molder.

So we're the next step beyond prototyping. We make real parts, but we make them quickly. In fact, we're shipping real injection molded parts within a day now, if the design is relatively small and simple. We can ship them the day after we get the order. So we're just as fast as a rapid prototype, but we ship real parts. The other thing you should know is with rapid prototyping, if you want one part, it might cost you fifty or a hundred bucks, so it can be significantly less than our minimum price of $1,795. Rapid prototyping has its advantages, but again, you can't really get real parts that way.

D2P: You mentioned your process has volume limitations. What is the upper limit of the number of parts Protomold typically handles?

BC: Well, that depends on the part, the customer, and the requirements. For the most part, if somebody knows for sure that they're going to need more than 10,000 parts, and they know they're not going to change their design, they'll tend to go off and have someone make a production steel tool for them. It will be worth the expense then, to spend maybe $20,000-$30,000 on a tool rather than $2,000-$3,000 with us. Because a part with us might cost a buck apiece, whereas a part from a multi-cavity steel tool might be ten cents. So in the high volumes, typically anything over 10,000 parts, they'll come out ahead with a production tool.

Having said that, pretty much once a week we get an order for 40,000 or 50,000 parts from customers. If a customer comes to us and says, "You know, I need this part, but I need it tweaked a little bit; please make me a new tool, and then go ahead and make me 40,000 parts," we can do that and then ship the parts a week or two later. Whereas if they were using injection molded tools from a conventional (production-volume) manufacturer, you're talking months. So sometimes speed is what the customer is really after, and they don't mind if it costs more for the flexibility that they never have to buy that $40,000 production tool. So sometimes we get large orders, but for the most part, our orders are typically under 10,000 parts.

Another angle of interest: Aluminum is a softer metal than steel, so depending on the resin that you're shooting, our molds may wear faster than steel molds used for large-volume production. We use filled resins, like glass-filled nylon, and these resins are very abrasive. If you're making a few hundred parts, nobody cares. But if the customer wants 5,000 parts and he wants to order them from Protomold, what we tell them is that if your mold starts to wear and you still want to keep buying parts from us, then we'll make a new mold (at our cost) and keep shipping parts to you. So when people ask us "How long will a mold last?" the answer is basically "forever" because we'll replace it if wear becomes an issue. The "technically accurate" answer is that the mold finish will last varying lengths of time depending on the resin used, but if it does wear, and you're still buying parts from us, we will remake the mold to support the customer's requirements.

D2P: So, is most of your business from customers who need parts quickly, rather than in volume, or are they using Protomold for the "startup" part manufacturing before they're ready to commit to volumes?

BC: The answer is yes, of course [laughs]. Customers use us for many different reasons. An example is what we did for Xerox; they're continually designing new printers, copiers, et cetera. They have a new product development group that wanted to bring a printer to market very quickly. What they do, is they first manufacture something less than 100 units, then test and demo them for customer feedback. And each of these units might have 40 or 50 different plastic parts in it—that's a lot of molds to make.

They came to us because of our speed and breadth. We have a pretty large-scale operation now, so we were able to take all those designs and parallel manufacture those molds, and get all those parts out the door quickly. They came to us with about 40 3D CAD models for these parts. We quoted them all, they bought them, and we delivered 100 of each part in 10-15 business days. So the way they used us was to very, very quickly get their prototype printers built. Once they had verified their design, then they went elsewhere to get production parts made in very high volumes. So Xerox used us to help get their new product to market faster than their competition. They told us they didn't know anyone else who could handle 40 molds at once. Their alternative was to go out and source them to a collection of different vendors, each of whom would have a longer lead time than Protomold. And from a purchasing standpoint, it would have been a nightmare, coordinating say 8-10 different vendors.

There are three main ways our customers use Protomold's speed to their advantage: First, people use us to get to market faster with a new product, as Xerox did. The second way that people use us is to reduce the risk when they go to market. Think of a medical device manufacturer. They have a very specific date when they need to go in for FDA testing, say, ten months down the road. If they can iterate on their product ten times during that time period rather than just twice, the improved chances these additional refinements add to passing the FDA test could mean the survival of their company. So many companies, such as medical device manufacturers, use us to reduce the risk associated with going to market. They don't necessarily get their product to the FDA faster; they iterate their product three times more and dramatically improve their chances of passing the test.

The third way that people use us is to go to market with multiple versions of their product, each with a slightly different twist, in the same amount of time that they otherwise could have only gone to market with just one version of their product. An example of this is a toy manufacturer. Because our tooling costs are so low, along with our speed, that meant that within a few months they were able go to market with six different versions of a new toy, rather than just one. Then when they saw which one looked like it would be the biggest seller, they chose that one for their high-volume manufacturing.

D2P: Brad, thank you for your time. Is there anything else that you would like to add?

BC: We talked earlier about how Protomold does rapid injection molding, not rapid prototyping. Well, now we have a division that does subtractive rapid prototyping. We take a block of plastic, put it in a milling machine, and cut a part out of it. It's a division of Proto Labs called First Cut Prototype, and they can deliver CNC machined plastic parts as fast as the next business day. This division of ours competes directly with rapid prototyping service bureaus, but the advantage they have is they can actually deliver the material you really want, and you get the end-product surface finishes you expect. And they're very synergistic with Protomold. If you need, say, 1-to-10 parts and you don't want to buy a tool yet, you just want to get a few real parts (actual material and surface finish), you can go to First Cut Prototype. And then, when you're ready for a few hundred parts, Protomold will quickly make the tool and injection mold the parts for quick delivery.

For more information on Protomold, First Cut Prototype, or Proto Labs, Inc., visit

PEEK is a trademark of Victrex plc

Teflon is a registered trademark of DuPont

Protomold is a registered trademark of Proto Labs, Inc.

This technical information has been contributed by
Proto Labs, Inc.

Home |  About Us |  Back To Technical Library |  Contact Us
Copyright © 1996-2010 All Rights Reserved.
General or Technical Questions? E-mail