Design of a high-throughput vibration-based fatigue testing framework for microscale metal beams

The discovery and development of new structural metals is often limited by the time required for characterization. In particular, fatigue characterization is both resource-intensive and time-consuming, and yet remains critical for the deployment of new metals for applications in aerospace, energy, and transportation where structural materials must withstand millions of cycles of loading. Conventional fatigue testing involves sequential experiments, evaluating one sample at a time. This significantly slows the exploration of new alloys, processing conditions, and additive manufacturing parameters.

This project aims to develop a high-throughput, vibration-based fatigue testing framework that enables simultaneous excitation of multiple microscale metal beam samples. We hypothesize that a single, coupled vibration system can be engineered to yield independent fatigue measurements across multiple samples. The core concept is a network of micro metal cantilever beams excited using a single shaker input. By leveraging resonance-driven loading, each beam experiences cyclic loading, enabling parallel fatigue testing within a compact experimental setup.

The primary design challenge is that the beams are dynamically coupled through the supporting structure: the measured vibration response is a superposition of interacting vibration modes, resulting in a non-trivial identification of the local response of individual samples.

This project addresses this challenge through two objectives:

1. The development of a vibration analysis framework that maps the global and local system response to beam-level dynamics
2. The optimization of network topology and sensor placement to maximize the resolution of individual beam responses

We will combine analytical modeling, finite element analysis, optimization, and experimental system identification. Using elastic beam theory as a foundation, and building on mobility and power-flow concepts, the framework will predict mode shapes, energy distribution, and dissipation across the network. The global response, measured using strain gauges, accelerometers, and laser vibrometry, will be decomposed into individual beam contributions, enabling the identification of local resonance shifts associated with fatigue damage. A demonstrator network will be manufactured using additive manufacturing and conventional machining to validate the models.

Beyond demonstrating feasibility, this project aims to establish a scalable framwork for high-throughput fatigue testing to accelerate material discovery. More broadly, this framework positions architected structures and materials as instrumented test platforms, enabling distributed sensing and characterization within a single vibrating structure.