From Lab to Track: Prototyping the IR-TRAIN System for Rail Base Inspection
Tracks
NDT UNLOCKED
Knowledge Level - NDT Level I/NDT Level II
Knowledge Level - Student
Presentation Topic Level - Intermediate
Target Audience - General Interest
Target Audience - Research/Academics
Target Audience- NDT Engineers
Target Audience- Technicians/Inspectors
| Wednesday, October 8, 2025 |
| 11:30 AM - 12:00 PM |
| Monterey 2-3 |
Speaker
Tsuchin Chu
Professor
Southern Illinois University Carbondale
From Lab to Track: Prototyping the IR-TRAIN System for Rail Base Inspection
11:30 AM - 12:00 PMPresentation Description
Ensuring the structural integrity of railway infrastructure is critical for safe and efficient train operations. Although ultrasonic and other nondestructive testing (NDT) methods are effective for assessing the rail head and web, they offer limited capability for inspecting the rail base, a region susceptible to fatigue and stress-related defects. This work investigates the application of Line-Scanning Thermography (LST), an active infrared NDT technique, for detecting both surface-breaking and subsurface defects in the rail base area. Two experimental platforms were developed: 1) the Infrared Rail Inspection System (IRIS), a laboratory-scale setup using a motorized cart outfitted with three 2000 W linear quartz heaters and a mid-wave infrared (MWIR) camera; and 2) the InfraRed Thermographic RAil INspection (IR-TRAIN) system, a full-scale mobile prototype tested on a 30-foot rail panel at Southern Illinois University Carbondale (SIUC).
Thermal contrast was induced via transient heating and captured using the IR camera for post-processing. Defect detectability was analyzed based on temperature differentials between flaw and non-flaw regions, using metrics such as thermal contrast and signal-to-noise ratio (SNR). High-carbon steel and standard rail steel samples containing artificial defects, including bottom drilled holes (BDH), electrical discharge machined (EDM) notches, and surface cracks, were inspected at various speeds. The IRIS system demonstrated clear detection of BDHs with aspect ratios ≥ 2 at a speed of 1.34 m/s under 6000 W of heating power. The IR-TRAIN prototype extended this performance, successfully identifying surface-breaking defects like 0.3 mm wide hairline cracks at speeds up to 2.24 m/s. However, detection sensitivity decreased at higher speeds due to reduced thermal diffusion and shorter observation windows.
These results validate LST as a viable and scalable method for real-time rail base inspection. The findings support ongoing development toward a field-deployable, autonomous LST system capable of enhancing the safety and efficiency of railway maintenance operations.
Thermal contrast was induced via transient heating and captured using the IR camera for post-processing. Defect detectability was analyzed based on temperature differentials between flaw and non-flaw regions, using metrics such as thermal contrast and signal-to-noise ratio (SNR). High-carbon steel and standard rail steel samples containing artificial defects, including bottom drilled holes (BDH), electrical discharge machined (EDM) notches, and surface cracks, were inspected at various speeds. The IRIS system demonstrated clear detection of BDHs with aspect ratios ≥ 2 at a speed of 1.34 m/s under 6000 W of heating power. The IR-TRAIN prototype extended this performance, successfully identifying surface-breaking defects like 0.3 mm wide hairline cracks at speeds up to 2.24 m/s. However, detection sensitivity decreased at higher speeds due to reduced thermal diffusion and shorter observation windows.
These results validate LST as a viable and scalable method for real-time rail base inspection. The findings support ongoing development toward a field-deployable, autonomous LST system capable of enhancing the safety and efficiency of railway maintenance operations.
Short Course Description
Biography
Connor Seavers is a Mechanical Engineering PhD candidate in the School of Mechanical, Aerospace, and Materials Engineering at Southern Illinois University Carbondale (SIUC). Connor has been working as a graduate research assistant in the Intelligent Measurement and Evaluation Laboratory (IMEL) under the advisement of Dr. Tsuchin Chu since 2018. His research has largely focused on applications of nondestructive evaluation techniques for inspection and characterization of various materials. Connor also spent three semesters working as an intern at NASA Marshall Space Flight Center, where he gained valuable experience working on NDE of additively manufactured materials for space flight applications.
