Should Design Engineers understand Manufacturing Processes?
Yes, of course, design engineers should understand manufacturing processes. But the real question is how much of the process should they understand? If a little bit of knowledge is a dangerous thing, just think how dangerous say 90% of the knowledge you need would be. Just so you know up front, I’m not going to take a side in this argument. I’m just going to play both sides against the other. Most of the parts I work on are plastic, and I’ll say that I understand injection molding capabilities fairly well. I’m maybe no expert, but I know enough to understand several types of processes. My question is should a guy like me be the one making design decisions that determine what kind of tooling must be used?
I’ve been pretty lucky. I’ve got several friends who are actual mold designers, so when I run up against a design problem that requires some molder expertise input, I’ve had several people I can go and ask. The problem tends to be that if you ask ten different mold builders how to tool for a particular issue, you’ll get ten different and contradictory opinions.
As an example, I want to make my models accurate, with the parting line and draft. A lot of designers will just throw a part without draft at the molder, and expect the molder to either add it or not, putting it into the tool paths without a solid model. I don’t know what they expect is going to happen, but I don’t like to chance it. If you ask mold guys about that, you get your conflicting answers again. Without doing a Moldflow analysis, I wouldn’t be comfortable specifying gates on a large complex part, but the parting line and draft can greatly affect the design and functionality of the finished parts, so yes, I’m going to specify those to the best of my ability. If the molder wants something else, that’s cool, but I’m still going to model it and make sure it doesn’t affect the design and function.
On the other hand, if I knew absolutely nothing about injection molding, a molder might have to come back and essentially re-engineer all my parts. Break parts into halves with assembly features, put parting lines across cosmetic faces, put gates into ergonomic areas. select a material that is too stiff, or can’t be colored, or whatever. Yes, I understand how a CNC machine works, but should I really be the one making decisions about setup that affect the design? Should the CNC operator really be making design decisions because it’s more efficient to fixture the work this way or that way?
At one large plastics manufacturer where I did month-long contracts to complete an entire line of products, say mop buckets, carts, sheds, or whatever, the CAD engineer was the focal point of several departments. Each department would send someone in to look at the proposed design, and give their feedback. Right up front I’d be taking input from the people working on the loading dock to make sure they could maximize the room on a pallette or shipping container.
It sounds a little crazy, working with all of those people looking over your shoulder and criticizing, and needing to integrate what may be contradicting requirements from various departments perspectives. But that’s what we do as engineers, we solve problems, we integrate ideas, we make stuff work.
And then we have to take the scenario one step further, of course. It’s not just injection molded parts, but there are also sheet metal and microelectronics involved. Now I’ve got experience with all of that, but not like I’ve been working with those processes day in and day out for 20 years. I can’t redesign a circuit board because a plastic assembly boss has to move away from that wall or it will cause a thick section sink.
What it comes down to, and I’m sure most of you have reached this conclusion long ago, is that the design engineer needs a familiarity, but doesn’t necessarily need to be the manufacturing process guru. There’s just too much to know. There’s a reason why we have manufacturing engineers and design engineers, and tooling engineers, and yes, even marketing people. It takes input from all over the company to get the job done. You can’t expect your design engineers to assume all of those roles, even at a small company. After you’ve been at the job for a few years, yeah, maybe you know what you need to know to make decisions for the other departments, but get the departments in the room one by one and hear what they have to say. You might have to have an arm wrestling match to settle some disputes, but it’s best to know about problems before the product is on the shipping dock.
Just like the ” full stack developer ” from the software industry, I think a good design engineer need to have some sort of ” full stack ” capability in order to produce a better design.
I come from the opposite side – Manufacturing Engineer that doesn’t know Design. I’ve been fortunate enough to work in several different machining facilities over the years: Permanent-mold aluminum casting, Piping (think of chemical-processing plants), and Aerospace/Defense. Each one had different design rules for manufacturability.
For aluminum casting, one of the design rules was the shrink factor. Every CAD model had to be scaled up to account for shrinkage as the molten metal turned solid. Another design rule was minimum draft. Unlike plastic injection molding, our minimum was 1.5 degrees, but we preferred 3. We also had minimum thickness rules for ribs (too thin and the rib won’t fill completely). Gates and riser design were more art than science as we tried to avoid turbulent metal flow to decrease the amount of porosity.
In the Piping world, we dealt with really hard-to-machine materials, so design emphasis was on making as little machining work as possible. Designers consulted the different raw stock material guides to determine the closest proximity to the actual finished diameter, and using tube stock if available.
Aerospace and Defense had requirements like design the corner radius of pockets slightly larger than the expected cutting tool diameter (to avoid tool chatter in the corners). On pockets that had canted walls, we were allowed to leave extra material under the wall, matching the diameter of the smallest cutter we planned to use. Fixturing was another issue, as the primary purpose of our machining efforts was to take a billet of material and reduce it to the size of the designed part. That meant lots of stress (warpage) introduced into the machining operations, so we had several setups (Roughing/Semi-finishing/Finishing) to achieve the final design without warpage.
In many instances, I actually re-modeled the design file for manufacturability. It made me a good modeler, but it did not make me a designer. I have no idea how to design a part and ensure it is suitable for a given purpose.
I worked as a machinist (manual and CNC) prior to and while going to college for mechanical engineering, and I keep the machinability of my designs in mind at all times. An extra hour here and there spent considering those things might push back the release of a design by a day or two, but then those designs go through fabrication without those major issues that can cause delays of days or weeks.
I also did plastic part design once upon a time, but also had to do the mold design. As I got more familiar with injection molding and mold fabrication, my designs became more effective.
The more familiar you are with the relevant manufacturing processes, the better.