Engineering Design Process Series: Testing a Prototype

By Marc Bucchieri

In creating a new whimsical education tool (as with any engineering project) one of the most important steps of the design process is testing the prototype. This step is the gateway that can turn a fun experiment into a product that other people actually use. One of our current developmental projects at WMSI is the Owl Radio Unit pair. These units are built to practice foxhunt scenarios, in which one unit finds the other using radio communication. After a week of happy tinkering and soldering, it was finally time to test our prototypes on the streets of Bethelehem, NH.

                                            Prototyping in full swing.

                                           Prototyping in full swing.

Of course, testing is only meaningful if there exist specific standards or metrics by which to judge the success of a prototype. In this case we focused on testing the range of our radio unit pairs. In the first round of tests, we set out to discover the maximum range of a couple different radio unit models. Since these units require a relatively clear line of sight, the transmitter was placed on top of a snowy planter near the high point of Main Street.

 Little Owl out on it’s own for the first time.

Little Owl out on it’s own for the first time.

In this first round of tests our receiver Owl was driven down Main St., stopping periodically every time the signal strength dropped another level. This kind of testing was not the most conclusive, since conditions could change so much between tests. However, it did give us some exciting “maximum range” values for a few different prototypes!

 GPS waypoints used to measure transmission range.

GPS waypoints used to measure transmission range.

Later we ran the Owl prototypes through a more controlled test to more thoroughly understand how signal strength drops off with distance. The Owl transmitter was left on a snow bank outside of WMSI headquarters on Park Ave, while the receiving unit was walked down the street. This time waypoints were taken every time the signal dropped off by another 10 dBm. From these results we were able to compare the units for two different types of antennas: a DIY wire antenna, and a commercial antenna purchased from Sparkfun.

 Yay for graphs! As expected, the signal strength drops off a little bit faster for the wire antenna than the commercial antenna.

Yay for graphs! As expected, the signal strength drops off a little bit faster for the wire antenna than the commercial antenna.

This graph illustrates another important concept in the engineering design process: testing against a control. By comparing the relative performance of two prototypes with only one difference between tests (the type of antenna), we’ll be better equipped to make decisions about what features to include in our final product. This data will be extremely useful as we iterate over our prototype design and get it ready for production. Check back soon for more posts involving our product development and the engineering design process!