Influence of Joint Gap on the Natural Frequency of Rail Joints in High-Speed Railways
Applied Sciences · 19.11.2026 · 16 (22) · 11496
DOI: 10.3390/app162211496 ↗Kurz gefasst
Fachartikel in der Zeitschrift Applied Sciences, online erschienen am 19.11.2026. Verfasst von Jung-Youl Choi. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Unlike continuous rails, rail joints are discontinuous structures composed of rail ends, joint bars, and joint bolts. Thus, changes in the rail end gap may affect the dynamic characteristics of the rail joints. This study investigated the influence of rail-joint gap variation on the first bending-mode natural frequency using impact-hammer modal testing and three-dimensional finite element analysis. Field measurements were conducted on an experimental concrete slab track comprising 60E1 rail segments shorter than 50 m connected by mechanical rail joints, rather than on an in-service continuously welded rail system. Field modal tests were conducted on a parent rail and on rail joints with gaps of 10, 13, 16, and 17 mm, all installed in the same type of track structure. The first natural frequency of the parent rail was approximately 168 Hz. For the 10 and 13 mm gap conditions, the average frequency was 166.5 Hz, comparable to that of the parent rail. In contrast, the average frequencies decreased to 127.5 and 118.5 Hz for the 16 and 17 mm gap conditions, respectively. A three-dimensional finite element model incorporating the 60E1 rail, joint bars, joint bolts, rail-support conditions, and the initial stress state was developed. A prestressed modal-analysis procedure was employed to account for the stress-stiffening effect associated with the longitudinal rail force, while the joint and fastening conditions were maintained consistently across all analysis cases, and response analysis was performed under the same gap conditions as those investigated in the field tests. The differences between the first natural frequency calculated through numerical analysis and the field measurement results ranged from 0.3% to 3.41%, and the trend of decreasing natural frequency with increasing gap was reproduced. The combined experimental–numerical approach provides quantitative evidence of the sensitivity of the first bending-mode frequency to rail-joint gap variation. The observed frequency changes may serve as supplementary indicators of variations in local joint stiffness and structural continuity, thereby providing a preliminary basis for vibration-based condition assessment and preventive maintenance. From a railway-safety perspective, identifying such changes may contribute to the early recognition of joint conditions requiring further inspection. However, the findings do not establish a direct relationship between natural frequency reduction and structural failure, and their applicability remains limited to the investigated experimental configuration. The findings of this study are limited to the investigated experimental concrete slab track comprising 60E1 rails, KR-type fastening systems, and mechanical rail joints. The experimental configuration does not fully represent the longitudinal restraint conditions of in-service continuously welded rails, and the actual rail temperature and longitudinal axial force were not directly measured. Accordingly, the results should be interpreted as a controlled assessment of the first bending-mode frequency sensitivity to joint-gap variation rather than as a general prediction of the dynamic behavior of all high-speed railway tracks.
Zitieren
APA
Choi, J. (2026). Influence of Joint Gap on the Natural Frequency of Rail Joints in High-Speed Railways. Applied Sciences, 16(22), 11496. https://doi.org/10.3390/app162211496
Vancouver
Choi J. Influence of Joint Gap on the Natural Frequency of Rail Joints in High-Speed Railways. Applied Sciences. 2026;16(22):11496. doi:10.3390/app162211496
BibTeX
@article{Choi2026,
author = {Choi, Jung-Youl},
title = {Influence of Joint Gap on the Natural Frequency of Rail Joints in High-Speed Railways},
journal = {Applied Sciences},
year = {2026},
volume = {16},
number = {22},
pages = {11496},
doi = {10.3390/app162211496}
}
EUVIMED