As we are gearing up towards our launch, we have entered a critical phase of the product development: testing. As you can imagine, a connected device such as SmartHalo requires a bevy of components to work together: several finely assembled components on a circuit board (the hardware) interact with a mobile device (the software), all within a plastic casing that must protect each of the components while being optimized to diffuse light and sound to the user. And this must work on thousands of units.
As if this wasn’t enough of a challenge by itself, this experience must meet the expectations of cyclist’s needs and perception. Every light pattern must be fine tuned to make sure our users understand at all times what the device is instructing them to do. Since the genesis of SmartHalo, we oriented our efforts towards creating a minimalist interface, but the art of simplicity is a puzzle of complexity; creating a simple interface is actually a lot of work.
To make sure each of these parts integrates into an awesome end product, test procedures are designed to ensure a level of quality you expect. These happen at three different levels of product development:
- Functional tests validate each of the products output on the assembly line
- Software tests make sure our code works properly
- User tests focus on how cyclists actually use SmartHalo
We’ll detail each of these in the following updates, with a focus on Functional tests today.
Making the first SmartHalo prototypes was hard. Going from those early prototypes and designing the product to be ready for mass production? Even harder. To make sure every component works well on every unit that is produced on the assembly line, our engineers have worked in partnership with Averna to develop test jigs. Averna is a world renowned test company that has worked with some of the biggest tech companies in the world. As a matter of fact, you probably use tech products that have been tested by them daily.
What’s a test jig? It’s a station equipped with various sensors that validates if each unit executes every features adequately in as little time as possible (since we’re producing thousands of units!). It checks every LED’s output, checks if the Bluetooth antennas work well, tests the speaker’s output and so on. Here’s a video explaining that process:
Once fully assembled, each unit from the assembly line is placed inside this jig while tests are run (this typically takes a minute per unit once the testing procedure is optimized). If a unit fails one or more tests, it is sent back to the assembly line to figure out what went wrong.
Next update we’ll detail how we test all those lines of code we’ve written.