Service Wheel Temperatures and Car Condition in Relation to Thermal Mechanical Shelling

The Wheel Defect Prevention Research Consortium (WDPRC) examined data from a wayside wheel temperature detector (WTD) located near the bottom of a grade in order to explore the root causes of hot wheels and thermal mechanical shelling. Not surprisingly, the data showed that most hot wheels, defined in this paper as a wayside WTD reading of 260°C (500°F) or greater, are found in trains descending the grade (descending trains), although they can be found in trains ascending the grade (ascending trains) as well. The majority of cars with hot wheels in ascending trains have the brakes applied at all wheel locations in the car, with unreleased or partially released hand brakes as a possible cause. While relatively few descending trains (15 out of 393) had many cars with hot wheels, these trains accounted for more than 20 percent of the descending cars with hot wheels, indicating that operational improvements could substantially reduce the quantity of hot wheels. Seventy-six percent of the descending cars with hot wheels had only a single wheel at or above 260°C (500°F). While the wheels in these cars are generally at higher temperatures than the wheels of other cars in the train, there were large temperature differences between individual wheel locations. Evidence of repeated hot wheel behavior was found in about 37 percent of the group of descending cars with hot wheels and about 20 percent of individual hot wheel locations. Two different car inspections were conducted based on the WTD data. First, a “near-real-time” inspection was conducted in which cars were quickly checked for obvious problems without removing them from the train. Next, an intensive inspection/test/teardown was conducted on bad actor cars, which showed repeated hot wheel behavior. Good actor cars, which repeatedly did not show hot wheels, were also present at the inspection/test/teardown for comparison. The cause of the hot wheels was not evident for the majority of cars at both inspections, however, bad actor cars were found to have twice the historical wheelset replacement rate of good actor cars.