Many steel bridge superstructures contain details prone to fatigue or distortion-induced cracking. Such cracks can lead to brittle fracture, greatly reduced load capacity and possible collapse of the bridge.

While transportation departments are aware of the problem, visual inspection is the primary way to determine if cracks are present. Cracks may form between inspections and remain undetected for some time.

This project will develop a system that can nondestructively locate and image dynamic stress concentrations on steel bridge superstructures. The system will use a low-cost thermal imager, dedicated field computer and special signal-processing algorithms. It would allow transportation agencies to quickly and nondestructively detect load-induced stress concentrations and intervene before cracks actually form. The device also could help evaluate the effectiveness of intervention in reducing or eliminating stress concentrations or treating existing fatigue cracks.