What testing goes into the components we design and build?

The Velo Lumino TMAT stem switch and AT fender taillight were developed not because we wanted to sell stuff, but out of a desire for components for our own bikes that don’t already exist. No one makes a switch like the TMAT stem switch. Some independent fabricators have built outwardly similar switches that mount in the same location, but these are mostly DIY products that involve assembling multiple widgets and using off-the-shelf switching components, and they compromise the field serviceability of the bike because of their specialized installation requirements. So Tom and I re-thought the concept of a stem switch from the ground up. Tom designed the mechanical aspects of the switch, while I designed the circuitry contained inside. One of the most important criteria for us was reliability. We wanted the switch to last a long time, and we wanted it to perform reliably in any environment.

On our first prototype, Tom tested whether the mechanism would withstand 10,000 continuous cycles of rotation by rigging it up to his lathe. The unit got really warm from testing, but the mechanism not only didn’t disintegrate, it also felt the same after 10,000 cycles as it did before starting the test. If you browse through an electronics supply catalog and look at switch specifications, most consumer switches are rated to less than 10,000 duty cycles.

We also wanted to test whether the switch would work in extreme operating environments. So we subjected the switch to freezing– deep freezing. We froze the switch in a calibrated laboratory deep freeze freezer, all the way down to -80C (-112F). Yes, minus one hundred and twelve degrees Fahrenheit! IMG_6134Okay, admittedly the switch did freeze up at that temp, and wouldn’t turn (the Phil Wood grease had hardened at that low temp). But honestly, there isn’t anywhere on earth that cold, and if there were, there would not likely be cyclists! So we tested it at a more moderate -13F, closer to Minnesota frigid. The switch worked fine, and didn’t feel any different than at normal temperature. Then we tried baking the switch. We heated it in a calibrated incubator to +80C, or +176F. That would be like baking the switch on the dashboard of a closed car. In the sun. In Death Valley, California. Again, the switch felt the same, and functioned the same, as at normal temperature, but I had to wear oven mitts to test it!

We designed the switch around magnetic reed switch technology because reed switches are impervious to environmental conditions. They are immune to moisture and corrosive agents. So we wanted to put this to the test. We submerged the switch in a tank of water, and asked if it could reliably switch a test light on and off. Repeatedly. For hours. It was unaffected by being in water. (We also designed the switch to have built-in drainage, so that if water enters the switch body– which we designed knowing it will happen, it can effectively drain out the bottom via the steerer tube and not cause potential long-term problems).

15539184633_361aa78904_z
On-bike test rig for prototype #1. My winter commuter has a quill stem, so the switch was installed in a short piece of steerer tube and zip-tied to the stem!

Lastly, we asked whether the switch would withstand prolonged vibration, such as a bike would experience on many roads, especially when ridden on dirt and gravel roads– one of my favorite types of riding. So we subjected the switch to a shake test. We clamped it to a calibrated orbital shaker, and shook it at 900 RPM for three hours. In case you’re wondering, 900 RPM on this shaker is like riding really fast on a washboarded dirt road. The kind of shaking that rattles your bones and shakes parts on your bike loose. When was the last time you and your bike suffered like this for three hours straight? All the while, the switch was hooked up to a test light, and never wavered.

17464719896_e1fb6ceb30_z
An early TMAT stem switch prototype. The switch that went through the “torture suite” of tests now serves daily duty on my summer commuter.

On top of controlled lab testing, the switch was also tested in situ on my daily commuter for over a month, during which time I made a point of turning the switch on and off at least a dozen times each ride. And it’s perhaps the on-bike testing that was among the most valuable, as that allowed us to fine-tune aspects of the switch’s design that only come under scrutiny while riding a bicycle. Like the feel of the rotary knob: early on, we thought we had the feel of the detents that define the Off and On states of the knob perfectly dialed in. Then one cold January day when it was -4F outside and I was wearing bulky lobster gloves, I couldn’t tell whether I was turning the knob or not without looking to see. This led us to dial in more tension in the ball bearing detent mechanism that provides tactile feedback to the user about what the knob is doing. We not only designed the switch to be reliable in extreme situations, we also designed it to “feel” right.

Oh, and that early prototype that went through all that testing? It’s now in regular use on one of my bikes, where I don’t even think about it: I expect it to continue to remain as durable and reliable as the production units.

The AT fender taillight went through similar stress testing before going into production, which is why we’re confident enough to stand behind it, and all Velo Lumino components, with a 3-year warranty. Any components that we develop will go through similar rigorous testing and will be held to the same high standards.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s