Last week, we decided to finally solve a problem we have had here in the office for a while and thought it would be a good opportunity to give a mini-glimpse into the product development process. Normally, the development process has five main steps: Define, Research, Design, Engineering, and Production. These five steps happen both consecutively, and in some places, concurrently.
In the define step, we need to figure out what the problem we are trying to solve is. This involves identifying the problem and its scope, and defining the end goal of the project. In our case, the problem we had was that we didn’t have keys for the hand towel holders here in the office bathrooms. The end goal was to have a set of functioning keys so that we didn’t have to use paper towel rolls any more. Obviously this wasn’t a large problem, and didn’t require a huge amount of actual “product development”, but it gave us an opportunity to show off a bit of what we do here at PAI.
The research step would normally involve researching the potential users, and analyzing the market/competition. For this project, the research step was abbreviated since we were the only ones that would be using the finished product. A quick web search gave us an idea of what the key we would need might look like.
Even though we now had a general idea of what the key was supposed to look like, we still needed to do a bit more tinkering. By playing around with the lock mechanism, we determined exactly how the key needed to work, and by extension what it needed to look like. With a set of lockpicks, and some can-do spirit, we discovered that your paper towels are actually not as safe as Big Paper Towel would have you believe… these locks were (perhaps not surprisingly) very easy to open.
The next steps in the process are the design and engineering steps. Although these are two distinct steps, they tend to happen concurrently because each of the steps can inform changes to the other. For example, there may be engineering issues (production process, structural stability, size, etc.) to consider when working on the design of the product and design issues (ergonomics, storage, etc.), and changes made in one of the steps can have cascading consequences to the other.
Just as with the research step, everything in these two steps was very abbreviated when compared to the normal process. After taking measurements of the keyhole (width, length, depth), we had enough to draw up a key in CAD.
David doing that CAD... look at him go!
Now that we had a drawing of what our key needed to look like, it was time to turn theory into reality. Our next questions were: What will be our “manufacturing process”, and what material will we use? Although these sound like two distinct questions, in reality, the answer to one informs the answer to the other. The requirements we had for this project were that the key needed to be stiff/strong enough that it wouldn’t twist/bend/flex when we were trying to twist it in the lock, but also not so brittle that it would break, because the last thing you want is a broken key stuck in a lock.
Our first idea was that we could use a laser cutter to cut the key out of some type of plastic. This has the advantage that it is extremely easy, inexpensive, and quick to laser cut something of this size. We ruled this out because the plastic we had on hand (acrylic) was stiff enough to not flex when turned in the lock, but, we worried that the acrylic would end up being too brittle (especially at this size) and snap inside the lock. This left us with 3D printing a new key. 3D printing is a great solution since there are multiple different materials with which to print, each with their own distinct advantages and disadvantages. Nylon was our cheapest option, but it lacked the stiffness and rigidity that we desired. The next material we considered was one made by the same company that makes some of our 3D printers, Markforged. This material is called Onyx, and is a nylon that is infused with chopped of carbon fiber which adds strength to the material. While Onyx is stronger than nylon, we still wanted to make sure that we had the maximum strength possible for our key. To that end, we did decide to utilize Markforged’s Onyx, but with a continuous strand of carbon fiber running through it. This carbon fiber running through the interior of the key would give the finished product even more rigidity and resistance to bending/flexing/twisting. The Markforged machines specialize in these composite materials and it’s printed properties were a perfect fit for this application.
It fits, but just a tad too short...
Once we had finished the initial prototype, it was time to test! In the PAI process, prototyping is key (HEY-O) to test the functionality of the product and make any necessary refinements. While the dimensions of the key were spot on to fit into the keyhole, it was slightly too short in length to easily reach the locking mechanism. It required a bit too much downward force in order to unlock the dispenser. Product design is always an iterative process; no matter how large or small the project you have to understand that your first design is likely to need minor refinement before it works as desired. Even with a project this simple we had to make a couple iterations in CAD before the keys worked well.
Product design is iterative! (SAD) Revision 1 is on the left, and the final product is on the right.
It works! And now we have functioning hand towel dispensers! All in a day's work here at PAI...