Cortex Design uses 3D printing to rapidly prototype hardware ideas in a physical form. Our in-house Objet polyjet 3D printer allows us to quickly print parts that could cost hundreds or thousands of dollars with subtractive CNC prototyping methods. It’s exciting technology, and it’s easy to understand why startups and Fortune 500 companies alike think about how 3D printing can augment their product design process, making it faster, cheaper, and leaner. In the future, 3D printing might completely replace other prototyping methods used in product design. It might even make advanced manufacturing in North America affordable again, and it might deliver a personal replicator to every home.
But we’re not there yet. Not even close.
Take a look at these pictures. Here’s a screenshot of a CAD design for a part:
Here is that same part made with subtractive CNC prototyping:
Here’s the same part made with a Polyjet 3D printer:
Now here’s the same part made with a cheaper FDM printer:
What is your reaction to each of these parts as a consumer? Or as a stakeholder? Or as an investor?
At Cortex Design, we work with companies at all levels of funding and maturity, and we see a lot of product ideas influenced by the limitations of 3D printing. We have invested a lot of time in learning and testing what today’s most common 3D printers can and cannot do. We help our customers understand how 3D printers can hold back the product design process and create just as many problems as they solve. This is especially important for customers launching a manufactured hardware product for the first time.
We sometimes encounter an expectation that 3D printing is a stand-in for subtractive CNC prototyping – where parts are carved from solid blocks of the target production material by large computer-controlled milling machines. Subtractive prototyping is much more time-consuming (taking days rather than hours to complete a product iteration), much more expensive typically, and requires skilled model makers to hand-finish a part in order for it to look like what you’ll see in production. All of this makes 3D printers much more tempting to use. But if we rely on 3D printers too much in the prototyping phase, you can end up designing for the 3D printer instead of for production. You can also encounter performance issues that you will not have with subtractive prototyping. This can open up a Pandora’s box of problems and totally derail the product design process.
Over-reliance on 3D printing can lead to lazy and lousy design.
3D printers don’t allow us to prototype with the same variety of materials we would use in the production of your product. The technology can make parts cheaply, which is great for figuring out an initial form for your product. But the sooner we can work with production-grade materials, the better. If we prototype a design using CNC machines, we can produce parts to a very high degree of dimensional accuracy that match the engineering performance and surface quality that an injection molding machine or a die casting machine will produce.
Great design is based on human experience. A big part of that is tactile response. If you have ever handled a 3D printed prototype, you’ll notice the difference in quality immediately. How does a product feel in your hands? How does it look? 3D printers are making strides in terms of quality, but most 3D printers create everything in layers, so you’re always going to have a “terraced” look. You can’t create compound continuous curved surfaces, so you can’t give your product the tactile, high-quality feel that will create a positive customer response. For that reason, we want to get high-quality prototypes in a stakeholders’s hands.
Cheap prototypes can be useful. Really cheap prototypes create issues by forcing you to fix a design that ain’t broke.
Not all 3D printers are created equal. FDM (fused deposition modeling) printers – which are the most inexpensive and common machines – have a minimum extruded bead width with which they can print, and tend to be weak between layers. If you design with this limitation in mind, you may need to print a prototype with thicker walls than you need to have in production, clunky ribbed features holding circuit boards in place, or excessively large screw bosses to stop them from snapping off in the prototype.
If you 3D print a mission-critical part – rather than producing it via CNC – and it fails, it’s not a useful failure. You may not know whether the part failed because of a design issue, or because the part isn’t strong enough. The solution might be to make the 3D printed part fatter, chunkier, or bigger than it needs to be in production, just so it is successful in prototype. You end up chasing a problem that may not even be a problem.
These design detours might lead to a prototype that has that tell-tale “playschool” aesthetic, a design that lacks the refinement of truly great hardware products. It’s easy to think that you will just make the appropriate changes and adjustments when you get to production design, but by that point it’s time consuming and expensive to bring back the elegance you were originally looking for in your product.
So why go off track in the first place?
If we work with CNC prototypes, we can design your product with production in mind, instead of making design decisions based on the limitations of the 3D printers. Take a look at this set of tips from Sculpteo for designing wall thickness for 3D printers. Now take a look at this this set of tips from Proto Labs for designing wall thickness for injection molding machines (quick-turn injection molding services come with their own set of design limitations, but alas that’s for another article). Which set of design considerations leads you closer to a production-ready product? Which is causing you to solve issues that have no relevance to the way your product will be manufactured?
Mass-volume production via additive manufacturing: doesn’t happen.
There are a few companies that are pioneering the use of additive manufacturing rather than more traditional methods in mass production, but the vast majority of hardware companies don’t have access to that kind of capital technology investment or process change. Until they do, why design for a manufacturing method that we aren’t going to use? Using CNC prototyping may cost a little more. It may take a little more time. But in the end, it’s aligning your process towards mass-volume production. CNC prototypes are still the closest thing to the injection molding and die casting processes that will manufacture your product at scale. Rather than designing for success in 3D printing, our approach is to design products for success in production.
Don’t get me wrong: 3D printers are useful.
We use our Stratasys polyjet 3D printer at Cortex Design all the time, and a bevy of great service bureaus for SLA, SLS, and FDM prototypes. These are good for making small jigs, holding fixtures, check fixtures, and other small non-critical parts. They are useful for rapid testing of ideas. They are great for evaluating our design decisions in between delivering a prototype to a customer. But I would never recommend putting a fully 3D printed prototype in the hands of a stakeholder – it doesn’t provide them with what they need to make good decisions.
It’s probably only a matter of time until additive manufacturing becomes a tool for mass production, but we’re not there yet. I’m confident that we’ll eventually close the technological gap when it comes to material quality and also cycle time. At that point it will make sense to rely on 3D printers for product design and send the CNC machines to the big factory in the sky.
Until then, let’s design for the production processes we have available to us today.
If you’re interested in learning more about what process works for each stage of product design, get in touch! I’m always looking to talk about design practices and where things are going. Leave a comment or reach out to me with a message!