Environmental energy harvesting has always been an attractive solution to the search for sustainable energy sources. Combined with the miniaturization and power consumption reduction of sensors and portable electronic devices, the possibility of powering using environmental energy, like vibration energy, instead of conventional batteries increases exponentially. Vibrational energy harvesting can be achieved using devices based on different physical phenomena. Electromagnetic based devices seem to be among the most effective when considering the power output levels within low frequency applications. However, these devices have always had inherent challenges including size and manufacturing cost reduction as well as limitations on the range of effective harvesting frequencies. The purpose of this thesis is to improve the design and manufacturing of an existing electromagnetic energy harvester, while maintaining comparable output power and range of effective harvesting frequency of the new harvester. The improved harvester is designed to maintain functionality based on nonlinear dynamics of an impact oscillator, which increases the effective range of harvesting frequency. This harvester is also designed to be manufactured using additive manufacturing techniques, which reduces the manufacturing cost and time significantly.

To achieve this purpose, the design of a fully metallic vibration energy harvester made by a partner group in the University of Waterloo was used as the starting point. The new harvester was redesigned for manufacturing using 3D printing technology as the rapid prototyping manufacturing process. This technology was implemented using the most cost effective material and equipment possible. Three consecutive iterations of the new harvester were produced including multiple versions of each iteration. These iterations presented the gradual improvement in design and manufacturing of the new harvester to overcome technical challenges while maintaining comparable performance levels. Iteration 1 was the baseline, process verification iteration, which included replacing the base of the original harvester with a redesigned and 3D printed base. In the second iteration the mechanism used to obtain the linear displacement of the seismic mass with respect to a fixed structure of the harvester was redesigned and 3D printed. The final iteration produced a fully 3D printed electromagnetic energy harvester for the first time. This new harvester was founded on the same displacement and generation nonlinear dynamic principles as the original metallic harvester with improvements in the design and material of the structure, as well as the selected components and mass of the harvester.

In spite of all the manufacturing challenges, iteration number 1 achieved a maximum output voltage of 138mV using a 76g metallic seismic mass, which was considered as the baseline for performance measurements. The price of the harvester base was reduced from $111.26 considering a metallic base to $28.16 with a 3D printed base. Iteration number 2 achieved a 78% output voltage compared to iteration number 1 with a cost reduction of 18.84%. Iteration number 3 achieved almost 50% maximum output voltage compared to iteration number 1 with a cost reduction of around 73% but remarkably with a seismic mass reduction of 65%.

The consecutive reductions of mass and cost of the springless vibration energy harvester with less proportional reduction in produced energy brings closer the possibility of further implementation of energy harvesters in daily applications including consumer products and education applications. The reduction in output voltage is linked to the type and quality of material used in 3D printing causing higher friction rates beside the reduction in seismic mass. Understanding these factors opens many possibilities for future work to improve the output of 3D printed magnetic-based energy harvesters even farther.