MxV RAIL R&D, RAILWAY AGE NOVEMBER 2025 ISSUE: In North American interchange, journal roller bearings removed from an axle, whether for bearing-, wheel- or axle-related causes, must go through a reconditioning process before being reused. This process includes visually inspecting rolling surfaces in search of either repairable or condemnable defects. MxV Rail has tested the effectiveness of a nondestructive evaluation (NDE) technology to enhance the inspection of bearing cups (or outer races) and find flaws that may otherwise go unnoticed and shorten the bearing’s useful life. (Wenger, M., A. Poudel, and B. Lindeman. (2025). “Reconditioned Bearing Performance – Phase 3.” Technology Digest TD25-005. MxV Rail/AAR. Pueblo, Colo.)
Under the Association of American Railroad’s Strategic Research Initiative (SRI) program, MxV Rail used a commercially available flexible eddy current array (ECA) technology to search for minute surface and near-surface raceway flaws on newly reconditioned bearing cups. These cups then underwent accelerated life testing in a laboratory bearing rig to determine if the suspected flaws would develop into spalls within 240,000 miles of the simulated service. The theory was that cups with ECA indications would spall during rig testing while those without ECA indications would not spall.
MxV Rail scanned 192 bearing cups using ECA technology and found seven with indications of surface or subsurface cracks that were not detected during a normal reconditioning inspection. These ECA indications ranged in size from 0.003 to 0.048 in2 (2 to 31 mm2). The seven suspect cups plus eight other non-indicating cups were reassembled with new roller assemblies, seals and grease for subsequent rig testing at the University of Texas Rio Grande Valley (UTRGV).
Rig tests were run for up to 240,000 miles with intermediate teardown inspections every 60,000 miles to assess cup raceway health. The rig was configured to operate at an equivalent train speed of 85 mph to accelerate testing. The bearings were subjected to a constant vertical load of 34,400 pounds to simulate the GRL (gross rail load) of a fully loaded 110-ton car. Any ECA indications, if present, were aligned directly under the vertical load for the duration of testing to evaluate the durability of the raceway material in those suspected flaw locations. UTRGV monitored the temperature and vibration of the bearings to sense if and when spalling occurred.
Of the seven cups with ECA indications, six spalled within 41,500 miles, while one survived the test, giving the ECA inspection an 86% true positive rate. Seven of the eight cups without ECA indications endured the 240,000-mile test without spalling, resulting in a true negative rate of 88%. The presence or absence of ECA indications predicted whether a cup would spall prematurely with an overall accuracy of 87%, or 13 out of 15 cups.
Figure 1 (above) showcases one cup that showed two separate ECA indications approximately 1.3 inches apart. The red circles illustrate where the ECA indications were located on the raceway before and after 17,120 miles of rig testing. Severe spalling enveloped the raceway where both indications were sensed.
The single false positive result occurred on the cup with the smallest ECA indication (0.003 in2). Ongoing research at MxV Rail is focused on the effectiveness of the ECA technology in accurately detecting small flaws and determining whether such flaws reduce bearing life.
The Technology Digest upon which this article is based can be found in our eLibrary along with more than 1,000 other publications describing the railway research, testing and analysis from the SRI program. The MxV Rail team has the software, laboratories, and test tracks to meet the railway industry’s needs for both publicly accessible research and contractual tests for individual customers. We invite you to learn more at www.mxvrail.com.
