Skip Ribbon Commands
Skip to main content
Safety Recommendation Details

Safety Recommendation R-17-034
Details
Synopsis: On May 6, 2015, at 7:21 a.m. central daylight time, six cars (81 through 86) of a BNSF Railway unit train derailed near Heimdal, North Dakota. The train—three locomotives, two buffer cars, and 107 tank cars—was carrying crude oil. At the time of the derailment, the train was traveling at 45 mph. After the derailment, the train traveled about 4.8 miles before stopping. The train separated from the cars after car 81 impacted a highway-rail grade crossing at milepost 149.01 about 1 mile east of Heimdal. Five of the derailed tank cars breached and released about 96,400 gallons of crude oil, which fueled a fire and a large, toxic smoke plume. (See figure 1.) About 30 people were evacuated from Heimdal and the surrounding area. The estimated damage was $5 million. At the time of the accident, the sky was overcast and the temperature was 57°F.
Recommendation: TO THE FEDERAL RAILROAD ADMINISTRATION AND THE ASSOCIATION OF AMERICAN RAILROADS: Collaborate in the evaluation of safe kip thresholds to determine the remedial actions for suspected defective wheels conditions in high-hazard flammable train service based upon equipment detector data, and revise the Federal Railroad Administration Safety Advisory 2015-01 and the Association of American Railroads interchange rules.
Original recommendation transmittal letter: PDF
Overall Status: Open - Acceptable Response
Mode: Railroad
Location: Heimdal, ND, United States
Is Reiterated: No
Is Hazmat: No
Is NPRM: No
Accident #: DCA15FR009
Accident Reports: Railroad Accident Brief: BNSF Railway Crude Oil Unit Train Derailment, Heimdal, North Dakota
Report #: RAB-17-12
Accident Date: 5/6/2015
Issue Date: 1/11/2018
Date Closed:
Addressee(s) and Addressee Status: Association of American Railroads (Open - Acceptable Response)
FRA (Open - Acceptable Response)
Keyword(s): Hazmat

Safety Recommendation History
From: NTSB
To: FRA
Date: 9/16/2019
Response: On June 28, 2018, you told us that you are collaborating with the railroad industry to determine the reliability of various wayside defect detection equipment to inform industry standards and allow railroads to select the equipment that best suits their operations. As you explained, wheel condition can potentially be accurately and reliably assessed when a variety of tools are used together. You also stated that you have no plans to mandate any remedial actions because the cost of revising the current regulation would outweigh the benefit. We note that you are also working with the AAR to identify and implement industry standards for remedial actions once the true mechanism causing defects, such as vertical split rim wheel failures, is understood. As the AAR explained in an April 16, 2018, letter, “the industry has shown that a high kip reading and wheel rim thickness of one inch or less together are a predictor of a potential broken wheel.” We are aware that, in January 2016, the AAR amended its rules to authorize the removal of wheels with a dynamic load of 50 kips or more in combination with a rim thickness of 1 inch or less. You indicated that, due to the AAR’s rules revision, you had no plan to revise Safety Advisory 2015-01. Pending completion of your research and development of remedial actions, Safety Recommendation R-17-32 is classified “Open—Acceptable Response.” We urge you to reconsider your position on developing the recommended rulemaking; pending that action, Safety Recommendation R-17-33 is classified “Open—Unacceptable Response.” We note your ongoing collaboration with the AAR and encourage you to update and distribute your safety advisory accordingly with the information in the AAR’s rules. Such action will provide another tool for the industry to use and would be responsive to Safety Recommendation R-17-34, which is classified OPEN--ACCEPTABLE RESPONSE.

From: FRA
To: NTSB
Date: 6/28/2018
Response: -From Ronald L. Batory, Administrator: Wheel Impact Load Detector (WILD) readings are helpful, but insufficient to determine the condition and compliance of a particular wheel or wheel set. Further, mandating remedial actions based on WILD readings would be ineffective. However, wheel condition can potentially be accurately and reliably assessed when a variety of tools are used together. FRA and the railroad industry are collaborating on ways to test and determine the reliability of various wayside defect detection equipment. The results of this research will inform industry standards, and allow railroads to select the equipment that best suits their operations. FRA will also work with AAR to identify and implement industry standards for remedial actions once the true mechanism causing defects, such as vertical split rim wheel failures, is understood. In its letter to the NTSB dated April 16, 2018, AAR explained that "research shows that kip readings are not an indication of imminent wheel failure"; nevertheless: ... the industry has shown that a high kip reading and wheel rim thickness of one inch or less together are a predictor of a potential broken wheel. Consequently, in January 2016, AAR rules were amended to authorize the removal of wheels with a dynamic load of 50 kips or more in combination with a rim thickness of one inch or less. Because AAR has changed its rules accordingly, FRA does not see the need to revise Safety Advisory 2015-01 . It must also be noted that requiring the use of wayside defect detection equipment, or calibration of the equipment, would require the current regulation to be revised. The cost of this revision would, based on the extremely low number of accidents that could be prevented by accurate and reliable wayside defect equipment, far outweigh the benefit. Therefore, FRA respectfully requests that the NTSB close Safety Recommendations R-17-3 2, R-17-33, andR-17-34. I appreciate your interest in these important safety issues.

From: FRA
To: NTSB
Date: 2/8/2018
Response: -From Karl Alexy, Director, Office of Safety Analysis, Federal Railroad Administration: Thank you for the January 11, 2018, report, BNSF Railway Crude Oil Unit Train Derailment, RAB-17112, Heimdal, North Dakota, sent to the Federal Railroad Administration (FRA) concerning, in part, the National Transportations Safety Board (NTSB) Safety Recommendations R-17-32 and R-17-33 to FRA, and R-17-34 to FRA and the Association of American Railroads. Improving safety is FRA's top priority, and FRA will continue to work to make rail shipments as safe as possible. FRA is committed to working with the NTSB to prevent future accidents and save lives. FRA welcomes and will consider all recommendations that will further this goal. If you have further questions or concerns, please contact Mr. Kenton Kilgore, Program Analyst.

From: NTSB
To: FRA
Date: 1/11/2018
Response: On December 29, 2017, the NTSB adopted its brief BNSF Railway Crude Oil Unit Train Derailment, RAB-17/12. The details of this accident investigation and the resulting safety recommendations may be found in the attached report, which can also be accessed at http://www.ntsb.gov. Among the Safety Recommendations are three issued to the Federal Railroad Administration, which can be found on page 11 of the report. The NTSB is vitally interested in these recommendations because they are designed to prevent accidents and save lives. We would appreciate a response within 90 days, detailing the actions you have taken or intend to take to implement these recommendations.

From: NTSB
To: Association of American Railroads
Date: 7/6/2018
Response: You wrote that the FRA’s Safety Advisory 2015-01 is based on a misunderstanding of why the AAR adopted its original interchange rule on kip readings, which was that high kip readings resulted in wheel removal because of the impact the high levels have on the rails, not because the readings indicated imminent wheel failure. We appreciate that your organization is currently involved in research that supports removing wheels that have accumulated mileage, high temperature, and fatigue cracks on vertical split rims, as we are aware of the safety concerns these issues present. Pending your recommended collaboration with the FRA to evaluate safe kip thresholds and determine remedial actions for suspected defective wheel conditions, Safety Recommendation R-17-34 is classified OPEN--ACCEPTABLE RESPONSE.

From: Association of American Railroads
To: NTSB
Date: 4/16/2018
Response: -From Edward R. Hamberger, President and Chief Executive Officer: As explained below, AAR has, and continues to, evaluate how kip readings can be used to enhance safety. Unfortunately, FRA's Safety Advisory was based on a misunderstanding of the reason AAR's original interchange rule on kip. readings was adopted. AAR's interchange standard pertaining to the removal of wheels because of high kip readings was based on the impact of high kip levels on rails and the objective of decreasing broken rails. Industry research shows that kip readings, are not an indication of imminent wheel failure. Data from about a million high kip wheels show that only 0.02 percent (219) of wheels with a kip reading of greater than 80 experienced a wheel failure. Furthermore, 74 percent (608 of 827) of broken wheels did not reach 80+ kips before they failed. However, the industry has explored the possibility of using kip readings in combination with other factors to identify wheels that might be prone to failure. In fact, the industry has shown that a high kip reading and wheel rim thickness of one inch or less together are a predictor of a potential broken: wheel. ·consequently, in January .201.6; AAR rules were amended to authorized the removal of wheels with a dynamic load1 of 50 kips or more in combination with a rim thickness of one inch or less. Note that the causal wheel in the Heimdal derailment had a rim thickness greater than one inch and therefore would not have met this combination threshold. A current research project is focusing on damage occurring at the outside of the tread, which is a known precursor to a vertical split rim (VSR). The Wheel Impact Load Cracked Rim Detector (WILDCARD) measures wheel impacts outside of the tapeline. This research may well yield a new way in which wheel impact measurements can be used to identify wheels prone to failure. As a separate issue, Page 6 of the Heimdal Accident Brief identifies a need for more information on VSR failures and AAR agrees. To further identify wheel failures that are the result of a VSR, rather than all types of a cracked or broken rim, a new defect code will be implemented on July 1, 2018, to provide specific VSR data. In conclusion, AAR has and will continue to explore ways of using kip data to identify wheels prone to failure. Accordingly, AAR respectfully requests that Recommendation R-17-034 be given a status of Closed -Acceptable Alternate Action.

From: NTSB
To: Association of American Railroads
Date: 1/11/2018
Response: The National Transportation Safety Board (NTSB) is an independent federal agency charged by Congress with investigating every civil aviation accident in the United States and significant accidents in other modes of transportation—railroad, highway, marine, and pipeline. We determine the probable cause of the accidents and issue safety recommendations aimed at preventing future accidents. In addition, we carry out special studies concerning transportation safety and coordinate the resources of the federal government and other organizations to provide assistance to victims and their family members affected by major transportation disasters. On December 29, 2017, the NTSB adopted its brief BNSF Railway Crude Oil Unit Train Derailment, RAB-17/12. The details of this accident investigation and the resulting safety recommendations may be found in the attached report, which can also be accessed at http://www.ntsb.gov. Among the Safety Recommendations is one issued to the Association of American Railroads, which can be found on page 11 of the report. The NTSB is vitally interested in this recommendation because it is designed to prevent accidents and save lives. We would appreciate a response within 90 days, detailing the actions you have taken or intend to take to implement this recommendation.