In our latest Technology Tuesdays podcast, Design World’s Michelle Froese speaks with Jonathan Cottrell, lead program manager with PTI Engineered Plastics, about design for manufacturing and the injection-molding process.
We discuss the key factors to take into account when designing plastic parts, including the material, gate, wall, and draft considerations, as well as tips for improving the aesthetics of a part without compromising the integrity.
A lightly edited transcript of this conversation follows below.
Design World (DW): Hello! I’m Michelle Froese. Welcome to Design World’s technology, Tuesday’s podcast. Thanks for joining us. Today’s topic is design for manufacturing injection molding, and we’re going to hear from Jonathan Cottrell, lead program manager with PTI.
PTI is a custom injection molder and manufacturer of plastic components and assemblies that specializes in low-volume production. With nearly 25 years of experience, Jonathan has developed products in several industries such as automotive, aerospace, military, agriculture, medical devices, and others.
He’s had the opportunity in his career to follow products throughout their lifecycle — from concept to completion — which has led to valuable experience to draw from when producing quality parts using design for manufacturing practices. Of course, when designing any component and particularly a plastic part, there are several key factors to take into account including the uniform wall thickness, the gate location, and even applying the proper draft to your part model. So, this discussion is intended to shed light on such key areas of design with insight from our guest.
Jon, thank you so much for joining us today. It’s great to talk with you.
Jonathan Cottrell (JC): Thanks for having me.
DW: For sure! Before we begin, I was wondering if you could please share a little more about your background and career focus, especially considering design for manufacturing is such a broad topic.
JC: As you noted, I’ve been in the industry for close to 25 years now, starting in a small injection-molding house shortly after high school. My formal education included a bachelor’s in mechanical engineering, with a specialty in plastic product design. Afterward, I got my MBA. But also, as you pointed out, I’ve been managing products as a design engineer, a design release engineer, a program manager, a project engineer, and all sorts of different names throughout the years in many different sectors.
Typically, the parts that I’ve launched are for automotive interiors. They also include more of the middle-ground plastics and not the specialty, thin-wall molding, or high-volume type products. So, this includes the components that you might handle on a day-to-day basis. The mouse that you use with your computer, the devices that you see in a hospital room, monitors, smartphone cases, and things like that — more of the general, classic type of components.
DW: Got it! So you’ve got experience with quite a few different components and avenues within manufacturing and design?
JC: I’d like to think so!
DW: Why do you think it’s so important for a designer to consider the manufacturing process early on in a project, and at the beginning of the design phase of a product?
JC: Well, fully understanding how the parts can be made allows a designer to apply principles that are specific to that manufacturing process. Often we get designs that look great on paper, however, the design is not conducive to injection molding or what would be ideal for a machining operation. I can’t tell you how many times over the years I’ve received a part drawing or a part model, only to realize it isn’t made for injection molding. So, by not understanding the requirements needed for the molding world, a lot of time and costs are lost in an attempt to bring the product to fruition.
DW: Can you tell us more about the injection-molding process? For example, what are the main design aspects a designer should think about when developing a part?
JC: I could probably go on about this topic for hours, but we’ll kind of keep it focused on a few specifics. This includes the material selection, which is more commonly termed, the resin that’s being used. But for today, we’ll refer to it as the materials. The main design considerations include how thick the wall stock is, adding draft angles, parting lines, where the part’s going to be gated, how do we want to inject the part, the design of bosses of ribs. Also, if it’s not a standalone part, how will that part be joined or made into its sister part?
DW: Right. You mentioned the materials, are there any features to consider when selecting the material used for a project?
JC: Definitely. One of the first questions that I would ask is… what’s the environment that the material is going to reside in? Is it going to encounter any chemicals or any extreme temperatures? Will it be a cosmetic part or is it something that people are going to touch and feel? Another thing to consider is: do you need a higher, engineered-grade material for performance? I typically would lean on the material resin manufacturers to help guide us after we properly understand the application.
Companies, such as PTI… well, we have our preferred suppliers and we can help guide a customer as to the ideal material grade to use. However, fully understanding who’s going to use it, what it’s being used for, where it’s being used, and how are they’re going to use it goes a long way. It’s basically developing a story. It’s similar to what we learned in grade school— the who, what, where, when, why, and how. Once we answer those questions, we can hone in on the right type of material that we need to use.
DW: The five W’s and the H.
JC: Exactly.
DW: Based on your experience, Jon, what would be considered too thin or perhaps even too thick of a wall? And how much does the type of resin typically play a factor in this decision?
JC: As I pointed out, I’m taking this from the point of view of your standard plastic part. So, we recommend anywhere from two to three millimeters is a good wall thickness, and that helps ensure a more consistent and predictable post-mold shrinkage. Also, as I pointed out, we’re not talking about thin-wall molding. We see a lot of plastic packaging, such as water bottles and containers, which are really thin. But I’m referring to more general plastics or the general design of parts, which are specialized, thin walls, and high volume.
DW: Right.
JC: A good story here is that early on in my career — and it’s something I will never forget — I was working at a company that made these really cool levers for a large furniture supplier. They looked like wood and they were about three-quarters of an inch in diameter. I believe they had some type of wood filler because they always smelled like a bonfire when we ran those parts. Anyway, the intent of the part was to replace a wood handle, so they always used the exact same part design.
Unfortunately, such a large mass, at three-quarter-inch diameter, caused a lot of processing problems. It meant the parts actually had to be dumped into a chilled bath for about five minutes to cool. If they didn’t go in that bath, they would warp and sink and twist. They’d look like a coiled-up snake laying on the floor or something.
Eventually, that long, dirty process was redesigned to core out the part and provide it with better design principles. This eliminated the need for that water bath and reduced the cycle times, so it saved a lot of cost to the consumer and the manufacturer, simply by applying proper design principles. Now, every time I see a really thick part, my mind immediately goes to that handle and what it did and then what it became. So, this is definitely something to think about… every time I see an extremely thick part, it takes me back to that story.
DW: Good experience though.
JC: Yeah. And, to answer the second half of your question, you were asking about materials and wall stock. When we look at materials and how thick or thin the part is, it’s important to choose the material that has the right flow characteristics for that wall stock. Let’s say we have a large flat part, if we have a material that’s not very viscous, it’s not going to fill easily and this leads to a manufacturing issue. So, we need to be cognizant of the material choice versus how far the material needs to flow.
DW: Good point. Do these flow characteristics affect why the gate location is important? And are there different types of gates?
JC: Yes, to both of those questions. Gating the part — and, by definition, gating is where we inject the material into the cavity for the part geometry — is important because it dictates how far the part flows, where the flow fronts are, if there are any holes to go around, where the knit lines would be, etc.
The gating also affects what we see, so the visual end of things, and what we touch, the touched surfaces. For instance, a gate on the A surface of the part may leave a bit of a gate vestige. If, let’s say, it’s a medical device that a surgeon needs to touch and they’re wearing a rubber glove, that gate vestige could tear the glove, and now the surgeon is exposed to whatever they’re working on. This is clearly something we’d want to avoid.
This leads me to the next point during the tool-design process. We typically go through several questions and ensure we receive the gate location approvals and such from the customer. But, if the designer has these considerations in their mind early on, such as: “Where do we want to gate?” or “Is there an area in the part that’s a non-tactile surface or a non-visible surface?” And, if we can identify that on the part drawing or even a feature on the part at the design stage, then sometimes this eliminates that design cycle once we get into the tool design.
DW: This is such an important consideration…prior to the manufacturing. So, what are the different types of gates available?
JC: I guess, fortunately and unfortunately, there are a lot. But one of the main ones is an edge gate. As an example… remember when we were kids, we did models, or we might help our kids with models now? Or, maybe it’s a toy that comes in a frame. That frame typically has the toy parts in it and the runner. When you break those pieces off, there’s often a little standing vestige on the side of the part — this is conducive to an edge gate. Unless you have a talented operator that’s trimming it with a knife or some clippers, the edge pieces are going to be leftover.
There’s also what’s called a sub gate. Sub gates are sheared off during the part-ejection process. They’re cleaner, but also more complex to manufacture. Then, there’s a thing called a cashew gate. This is something that wraps around to the backside of the part. It’s also more complex, but it hides the gate location from the user and end-user.
There are even more complex direct gates and valve gates, which are part of a hot manifold system. At this point, we suggest that the designer get to know the gate styles and consider the parts to use, or consult our tooling experts. Even a quick internet search can help to find what options are out there for the part design
DW: This makes sense, considering they are so many options.
JC: Yes. And these are only the common ones. There are more options available. Plus, the tooling experts and process engineer experts will all have their own opinions as to the best option. One thing that we need to be wary of though when choosing different materials is that the material selection and gating styles must work with one another. So, if it’s a brittle and highly engineered material, for example, it may not be conducive to the cashew gate because those gates require material flexibility. You often can’t have the best of both worlds, so to speak. It has to be a combination that works together.
DW: And does cost heavily influence the gate choice?
JC: Cost definitely plays a part, for sure. The designer may not care about costs too much, but yeah… when thinking of the system as a whole, we certainly need to take costs into account.
DW: You mentioned part ejection a little bit earlier, Jon. What variables should be considered when ejecting apart from the mold?
JC: Ejecting a part out of the mold is ridiculously critical in terms of quality because I can give you a really good plastic part, but if we can’t get it out of the tool, you get a piece of steel with it.
We want to ensure that through the design process, the designer keeps this in the back of their mind. If we add a feature on the part, can we eject it? Or, if we identify — on the drawing or even in the CAD model areas — where to eject or not to eject? This is important. Another thing to consider is the draft of the part. If we don’t have any draft on a wall, then that wall may not release from the tool variable.
So… what is draft? This might be a question that you’re asking. Draft is if a part has a vertical wall, say perpendicular to the direction of the tool steel, that perpendicular wall needs to have a bit of an angle to it so that it releases from the tool. A lot of times, we see designs that don’t have any draft. Therefore, the part would likely stick to the tool. The part could also end up with some kind of defamation or it could crack or stress. But a bit of draft would help with the release and help prevent this.
I’m asked a lot about how much draft is needed. The typical rule of thumb, and what we would recommend, is one-and-a-half degrees. This is a starting point. And as a molder, I’m likely going to ask for more. As a product designer, I’m always going to ask for less. We certainly play back and forth a bit because different materials can mold just fine with less draft than others.
Through the years, I’ve seen parts with a little bit of draft and a lot of ribs. And it’s interesting when you try to mold that because sometimes we get to eject the pins that blow right through the part or we get parts that break and shatter as you try to get them out of the tool. Nobody likes to see that though because it’s a mess to clean up!
DW: No doubt.
JC: Yeah. One of the things that should be noted about surface draft and materials, especially for the A-side of a part or the visual side… a lot of people like to apply a texture to the part. In the automotive world, you’ll see a lot of animal prints or fake leather-type textures. Those textures have their own draft requirements and those draft requirements are typically published either by the texture house or by the OEM saying, all right… this is how much draft this type of texture of the surface finish needs. So, there’s documentation out there to help guide a designer on what that should look like.
DW: I was just going to ask you about the aesthetics or look of a part. How can a designer potentially improve the aesthetics of a component, but without compromising its integrity or reliability?
JC: First of all, for most designers, their parts look beautiful on the A side. Of course, every designer wants their parts to look great, but it’s a lot of the B-side stuff that we need to be wary of in terms of aesthetics. And we touched on a few things already, like about how we gate it and about adding drafts.
Ideally, that draft needs to be thought about when we discuss the materials or the components on the B-side. Those features on the backside are typically your structural components or your connection points, such as the ribs, bosses, and clips. And again, those backside features and how well they’re designed affects the A-side or the front side of the part. For instance, a rib should only be about 50% as thick as the walls it’s connected to. If it’s thicker than that, you’ll see what’s called read-through, and you can get a witness or maybe a little sink on the A-side. So, you end up with a bit of extra mass there.
In addition to that, let’s say there’s no draft on the rib or someone says it’s not possible to add any draft. Well, another thing you can get is called suck-back or pull. This means that as you push the part out of the tool, that area pulls back and leaves a bit of a witness where the rib is — because that rib is pulling on the A-side surface of the part. It’s like tugging on a string.
I saw a great example of this today. This morning, I’m sitting at my desk and I’m prepping for this discussion. I’ve got speakers by my computer and I noticed a witness all the way around them. I popped the cover off to see and that’s exactly where the rib is. Now, the witness doesn’t affect the function at all. The speakers work great but, visually, I can see the fault. As somebody in the plastics world, I notice it. A general consumer may or may not. But again, were the speakers designed properly? Probably not. But does it meet the intent of the function of the part? Definitely.
DW: I imagine you notice a lot of components now and how they’re designed, and how well the design holds.
JC: My wife, she picks on me for every time I get into a car I look at all the plastic and critique it.
DW: It’s a side effect of your job.
JC: Exactly.
DW: What about tolerances, Jon? What tolerances should a designer consider for an injection molded part? And does material play a factor here?
JC: So the tolerance question is probably hands down the number one question that I’m asked.
DW: Interesting.
JC: When I look at a drawing for the first time and start looking at the tolerances, if I say, “This looks a little tight for this feature,” the question that follows is always: “What can you hold?” Or, it’s: “How tight can we go?” The good thing is there’s a lot of documentation out there. There are white papers, books, and industry guides that have been published over the last 50 years — and probably ever since Bakelite was invented, which is one of the first plastics. So, I like to help guide the designer to use those tools for developing the tolerances.
Every grade or material family has basic guidelines available, but the one thing to know is that those guidelines are typically from a generic geometry. It’s a flat plate or it’s a round part, and consistent wall stock, but we break all of those rules because I know plastics. There’s no perfect design out there. I mean everything that we design or we injection mold, these are all complex shapes. Rarely do we get a nice flat plate or a nice flat dog bone from which most of the specifications or criteria were developed. So again, we go back to the baseline recommendations, and then we can work from there.
DW: It sounds like you get a little more freedom in design though with plastics, perhaps?
JC: You know what, there’s…sometimes. We always start with the standard recommendations, just like we were talking about, and then we can play from there.
DW: Earlier you mentioned connection points. Can you say more about these and about how plastics are typically joined together?
JC: I will start and say joining is not my specialty, but I do know enough to be dangerous. The first thing I and a lot of my colleagues will say, is that we recommend staying away from adhesives, such as glue or epoxy. Other people may have other thoughts on that. But, in my career, they seem to tend to be messy and inconsistent. And, if not used correctly, we could have a chemical impact on the plastic where the adhesive could degrade the plastic and it falters. This goes back to not properly vetting the application and the adhesive. However, over the last couple of years, in the world of adhesives, I’ve seen more and more improvements. At PTI, we’ve recently put in a six-axis robot that’s using two-part epoxy to join a few parts — but it was not without some reservations.
In addition to those adhesives, there’s a litany of snap-fit and press-fit design concepts that can be used. Our design team offers a large library of recommendations for snap and press fits. Again its first asking: What’s the application? Where’s the part being used? How is it being used? This leads to what kind of snap or press fit that we would recommend. We also see a lot of use for thread-forming screws, often called PT screws. They’re used if two components are joined together that don’t have to be serviced or taken apart and put back together for any reason.
We also see the use of thermal inserts and a capping screw, and those are typically are used for components that are serviced. So, if you have to take a door off to replace a battery or get into a circuit panel, or something like that. Again, another example of this is if you have kids with a new battery-run toy. Before, when we had to replace its batteries, it was usually a small door or panel with a snap-fit design. Think of even your TV remote control. It’s easy to get in and out of the battery compartment.
But now, some of the toys with batteries have a little screw that you have to unfasten. And, to keep them cheap, it’s often a self-tapping screw that, after one or two times of replacing the batteries, we strip out that screw hole or that screw boss. In my opinion, this application was used incorrectly. They use something that works one time but it doesn’t work many times. On higher-end products, you’ll instead see that the screw is going into a thermal insert or a molded insert that has a thread, so that it can be serviced without degradation. This goes back to how the product is going to be used, who’s going to use it, and how often? Asking the right questions.
One of the last joining techniques that we use, and by no means is it the only one, is ultrasonic welding. If you’re unfamiliar with the term, basically it involves using a piece of equipment that puts a lot of energy into the surface of a plastic part, exciting the molecules and re-melting them at a point of contact with the other plastic. It’s not just a mechanical bond, it’s a chemical bond to put those two pieces together. This provides a great seal and it can be hermetic. The strength is quite high and it’s typically used on parts that may or may not have great geometry or have a snap or a screw put into them. So again, the joining technique is a function of what the design needs are and what the design allows it to have.
DW: I edited quite a bit in the fastening industry, so I know one of the key points is to consider fastening earlier on in the design stage. You make some good points here! Out of curiosity, did the project with the epoxy work out OK?
JC: Definitely, it is. It’s still, we’ll say, in the pre-launch phase but from a functional standpoint, it’s working.
DW: That’s good. Are there any other more recent technical advances specific to injection molding that a designer should be aware of in relation to manufacturing?
JC: This is kind of a tough question for me because when we’re looking at the technical advances of how we injection mold or the equipment, there’s always the next thing, the newer and better, or the latest piece of equipment. However, the basics of designing a part stay the same. Where we see the advances like technology versus the design, what I would recommend is that we first go through the initial concept of the part and follow the good design principles. Then, we can start looking at the system design, which includes the new technology and how it might affect what the part design looks like.
Going back to when we were talking about gating earlier, there are new technologies and new ways of gating, and those play into how the part is designed. What comes first? The chicken or the egg? Do we design around the process? Do we design around its use? It needs to be clear where we need to get to and then consider where or how do we start. I’ve been involved with a lot of those new technology introductions over the years and we started with a part, made sure the part was of sound design, and then we brought in the questions like: How are we going to make it with this new technology? Or what new technology are we going to introduce? And then, what do we need to tweak on design for that new technology to work best?
DW: So, beginning with the fundamentals first.
JC: You got it. I guess my overall suggestion for a parts designer is to start with the general principles, gain a good understanding of what the manufacturing process is, and the needs from the design perspective, and then engage the manufacturer. There’s a lot of knowledge out there, either documented at the material supplier level, at the molding level, or within the tool build world. Use the resources out there. Not one person knows it all, of course, but there is a collective of knowledge out there that will help to get a good product to market faster.
DW: You mentioned quite a few here, but any other good practices that our listeners might want to consider when it comes to design or injection molding?
JC: I guess my biggest thing is this: doing a little homework goes a long way. Over the years, I’ve seen a lot of people design a product that they believe is good without doing research that would provide them with better information. So yeah, it goes back to, do a little homework and talk with the molders. You’ll gain a lot of knowledge quickly. And most molders will work with you.
For us here at PTI, we love getting in on the ground floor and at the design end of things. Our design team here is well-versed and designed for manufacturing practices. We have a nice format that we follow with specific questions, and a lot of them have to do with the items that we’ve talked about today.
DW: That’s great! Thanks so much for this information, Jon. We’re nearing the end of our time available. Is there anything else you’d like to touch on or any final points before we end our discussion today?
JC: You can reach out to us….we have a lot of design documentation available. You can find us at teampti.com. Also, even within our website, there’s documentation on our DFM process. Feel free to reach out to us through that portal and see what we have to offer.
DW: Excellent! Thanks again for joining us, Jon, and providing such great insight. And thank you listeners for your time and attention. To learn more about us, visit Design World’s site at designworldonline.com, and be sure to subscribe and share this wherever you listen to your podcasts. Thanks, Everyone! Have a wonderful day.
PTI
teampti.com
Filed Under: Adhesives • epoxies, Commentaries • insights • Technical thinking, PODCASTS
