Subscribe

NTSB Determines Probable Cause for October 2023 BNSF Derailment

Figure 1: The derailment site. (BNSF Image, Courtesy of NTSB)
Figure 1: The derailment site. (BNSF Image, Courtesy of NTSB)

The National Transportation Safety Board (NTSB) has released the final report for its investigation of the Oct. 15, 2023, accident involving a BNSF train that struck and partially collapsed a bridge that crosses over Interstate 25, killing a nearby driver in Pueblo, Colo.

What happened?

On Oct. 15, 2023, about 3:24 p.m. local time, southbound BNSF train C-ATMCRD0-31D derailed 31 hopper cars loaded with coal and subsequently killed a nearby driver in Pueblo, Colo. (see Figure 1, above), according to the report that was issued Feb. 12 (download below). The hopper cars, NTSB said, derailed near a track switch north of a railroad bridge that crosses over Interstate 25; the derailed cars struck and partially collapsed the bridge over the northbound lanes. Six derailed hopper cars fell to the interstate below, with at least one falling on a northbound truck-tractor in combination with a utility trailer (combination vehicle), killing the driver. The combination vehicle came to rest beneath the collapsed bridge span, derailed hopper cars, and lading. The NTSB said that neither of the two train crew members were injured. The train comprised 124 cars and three locomotives. Damages were estimated at $15.6 million.

Probable Cause

The probable cause of the derailment “was the train encountering a rail break that occurred when a mismatched thermite weld failed; the thermite weld cracked near the rail’s base because the welder, for unknown reasons, likely did not use a compromise kit during the welding as required by BNSF Railway procedures,” the NTSB determined.

NTSB Analysis

According to the NTSB, the derailment resulted from a broken rail north of the bridge. “Video from the lead and rear locomotives of train C-BTMSPS0-33D showed that the rail broke as this train passed through the area of the accident,” it said. “Laboratory examination of the rail sections showed that fatigue cracking had initiated at the base of the rail and propagated upward until the rail fractured in bending under the weight of the last train to traverse the area before the accident train. The rail break occurred at a thermite weld completed less than five months before the derailment, and the rail fractured from a small fatigue crack in the base at the thermite weld. This fatigue crack initiated along a layer of underlying weld flash, which had flowed into a gap beneath the base during welding. This gap was created by a mismatched weld configuration. Measurement of the rail sections showed a difference in rail profile that, under BNSF rules, would have required a compromise kit that would have prevented flash from pooling under the base of the rail, but the welder likely did not use one of these kits. The resulting accumulation of weld flash at the base served as an adequate stress concentration to initiate for fatigue cracking while the rail was under cyclic loading.”

The NTSB noted that railroads are required to regularly inspect sections of track likely to develop into broken rails, as well as individual rails for internal defects, such as a void/pocket inside the steel of the rail. “Ultrasonic testing is one common method, but defects near a rail’s base can be difficult for this method to detect because the geometry of the rail can reduce the strength of ultrasonic signal reflections from these flaws,” the agency pointed out. “BNSF conducted inspections of the track and individual rails near the derailment site but did not identify any defects. Track circuits are designed to detect track occupancy, not to detect broken rails or prevent trains from traversing them. In some cases, the system may indicate a restricting signal when a rail is broken, but that did not occur here.”

According to the NTSB, a truck driver was killed following the derailment because the train derailed near a bridge. Derailed hopper cars, it said, had enough energy to strike the bridge, partially collapse it, and fall to the interstate below, landing on the combination vehicle and killing the driver.

Lessons Learned

BNSF, in response to this accident, adopted a policy of analyzing every failed weld to determine why they failed, according to the NTSB. “If welding practices contributed to the failure, the responsible team is retrained,” it reported. “Additionally, briefing materials have been distributed to employees who weld rails to emphasize the importance of proper kit selection. BNSF also enhanced welding oversight by implementing mandatory audits conducted by welding supervisors. Each supervisor is auditing 10 random thermite welds per month to confirm that the correct kit was used.”

Further Reading: