Vibration energy harvesting can become a durable source of energy for wireless
sensors or other low power applications like pacemakers. Huge savings in ecological footprint, production and maintenance costs can be achieved by replacing batteries for vibration energy harvesters. Most of the time, newly developed energy harvesters are tested in a lab environment on an electrodynamic shaker. The problem is that the standard lab experiments in the form of a sinusoidal or Gaussian noise signal excitation are not representative for the real world applications. In a classification of ambient vibrations it was observed that most vibrations found in the real world consist of a series of dominant frequencies, shocks and noise. It was also seen that among real world vibrations, there is a lot of variation in the power distribution among the classes. In the aim to bring the vibration energy harvester performance tests closer to the real world applications, an experimental benchmarking of energy harvester performance has been conducted. An energy harvester is designed and applied in the real world on the engine of two different cars. Successively, three different lab experiments are performed on an electrodynamic shaker, each experiment with its own type of vibration control. It is found that only taking the FFT data of a real world vibration is not sufficient. Using a sinusoidal excitation matching a single amplitude and frequency, or even a noise excitation matching the entire power spectrum, results in an under or overestimation of 50% compared to the real world performance. Therefore, to accurately predict the performance of an energy harvester in the real world, simulation or experimental testing need to be performed on the actual or a replication of the intended real world vibration.