The recent advent of machine learning and additive manufacturing allows us to design and manufacture complicated microstructures. Most current machine learning algorithms utilize simulated data due to limited developments in high throughput experiments. In this study, a fully automated split Hopkinson bar setup is designed and developed to generate large data sets of dynamic experiments. The experimental configuration comprises three main components: the gas propulsion system, the bar repositioning mechanism, and the automated sample placement system. The gas propulsion system is automated using a network of solenoids linked to an 8-channel relay under the control of an Arduino microcontroller. The bar repositioning mechanism is orchestrated through the collaboration of a sliding actuator, driven by a stepper motor, and a vacuum system. The sample placement system employs a sliding actuator integrated with a custom 3D printed cartridge, allowing for convenient automated reloading of experimental samples. This setup holds great potential for advancing material characterization and optimization through its sophisticated automation and data-driven approach.