In Part 1 of this series, we showed our design process, live at the Nashville Maker Faire. If you haven’t read that yet, go check it out. In this post we’re going to take the concept to the next level of finish.
We had already accomplished our initial prototyping goal — design an IoT device and make a working prototype in 2 days — but we wanted to push it farther. We could use our 2-day prototype for basic user testing, but not many people will be fooled into thinking it’s a retail-ready device.
This can be distracting and may skew user reactions so we wanted to refine it to the point that we could get genuine user reactions without having to explain to everyone, “This is a prototype. Act normal.” To do that, we re-built the housing to make it look and feel like a real production part.
We re-printed the final housing on the Form 2 at a 25 micron layer height (an 8x improvement over those earlier prototypes). It took about 14 hours to complete, but dang, does it look great!
That killer resolution was a great starting point, but we wet sanded it with increasingly fine sandpaper up to 1500 grit so it was buttery smooth for priming, painting and clear coat. If we had skipped this, the print layers would still be visible after painting. We used Tamiya primer and spray paint, and a matte clear Plastidip to finish. The Plastidip gives it a textured rubbery feel, which looks like molded plastic. It takes patience, finesse, and elbow grease to paint parts like this, but the finished part looks and feels like a real production piece off an injection mold. When you need that kind of fidelity, there’s just no way around it.
To interact with the electronics, we used Blynk, a drag-and-drop interface for programming IoT devices. Blynk made it easy to prototype the electronics and read the data using simple API calls to the device.
To demonstrate that it was working, we took a quick sample and held the sensor to see if it would pick up the warmth of a hand, and graphed that data to see the change.
This device, like many IoT devices, is defined by data. It’s whole purpose in life is to collect information, and make it easy to understand. That’s why prototyping the data may be just as important as prototyping the product itself. In the long run, this data would be displayed in pixel-perfect detail in a mobile app. For now, this graph will do.
An interface is just a way for a person to communicate with a computer— like the little one in our device — and for a computer to communicate back again. In our case, we communicate to the device with a button, and the device communicates back to us with a ring of lights. Touching the button is our way to ask, “How wet is the soil?” and the number of lights is the computer’s way of answering us. It’s pretty simple.
So far, we’ve used a lot of words to explain how we got from a basic idea to this functional prototype in a little over 2 days, but hopefully the results speak for themselves.
If we did all of this in just over 2 days, why do typical products take months or years to develop? In Part 3 we will get into the business side of developing products, using our plant sensor as an example.
Thanks for reading! Do you have any questions or want to talk shop? Leave it in the comments.