Investigation and Mitigation of Insulated Joint Electrical Failure: A Guide (2025)
Chapter: Appendix E: Industry Survey
Chapter E-1 Purpose and Background
Purpose
A major component of TRB TCRP Project D-20, “Investigation and Mitigation of Insulated Joint Electrical Failure” was a survey and interviews of transits with electrified railways related to arcing at insulated joints (IJs), with particular emphasis on IJ arcing issues related to the implementation of vehicles with alternating-current (AC) traction motors. This appendix summarizes how the survey was performed, the results of the survey, and information obtained in the interviews.
Background
IJs are special rail splices that mechanically connect rail ends and electrically isolate the connected rails. The electrical separation in rails provided by IJs is required for the train control (signaling) systems that govern train operations and prevent conflicts in the movements of trains. IJs are also used to isolate traction power negative return (TPNR) in adjacent track segments. Different devices and strategies used to allow TPNR to bypass the IJs should still provide electrical isolation between the rail ends required for the train control systems.
Electrified railways experiencing the IJ arcing covered in this study distribute the electric traction power as direct current (DC) using either contact (third) rail or overhead contact wire/catenary (OHC) systems. Some properties have trains that use alternating-current (AC) traction motors. On these trains, the DC power collected from the traction power system is converted to AC power onboard the trains to power the AC traction motors that drive the trains. Any power drawn by trains from a particular traction power substation (TPSS) must be returned to that same TPSS. Traction power is returned to TPSSs primarily using the running rails of the tracks, in addition to cables and bonding (cable connections). This return circuit is called TPNR.
TPNR is not the same as an electrical earth ground; it is a separate electrical circuit. Electrical potential difference exists between the TPNR and earth, and the difference in electrical energy can be significant at tens of volts and thousands of amps or more. TPNR systems are electrically isolated from earth ground to control stray currents that can damage the rail and adjacent electrically conductive infrastructure.
After the introduction of new vehicle fleets with AC traction motors, some DC electric railways noted IJ failures due to electrical arcing. Electrical arcing is not normal in electrified railways and is indicative of other underlying issues. Typically, IJ arcing and associated rail damage are caused by symptoms of other issues that create excessive TPNR electrical potential differences at the rails joined by the IJs, by a mechanical or insulation failure in the IJ assemblies. Where arcing occurs at IJs, the running rails, IJs, train control systems, and equipment on trains can be damaged, and the impacts are further compounded by delays to train operations caused by the arcing damage. While IJ arcing can occur on electrified railways with AC power distribution systems, the mitigation efforts are very different from those for DC systems and are outside the scope of this work.
Chapter E-2 Industry Survey
Survey Details
The industry survey was conducted over a four-month period. The initial plan was to send an email survey form to various electrified railways. TCRP panel members recommended using an electronic survey platform in lieu of a simple questionnaire. Various online survey platforms
were reviewed. Initially, two criteria were considered: ease of use for participants and security of user information and data security. The initial investigation of the various platforms provided a better understanding of available features, and two additional criteria were also included: capability to include images with the survey questions and dynamic presentation of questions.
Some key negative features that led to the rejection of some of the platforms for use in this survey included the following:
- License agreements that allow unrestricted use of the survey responses by others
- Collection of survey responder information
- Restrictions on web browser compatibility
- Potential operating system restrictions for users
The Microsoft Forms® survey platform was ultimately chosen, primarily because of data security but also ease of use. The included dynamic features significantly reduced the time load of the participants by not wasting their time on questions not relevant to their railway. Depending on the participant responses, the time to complete the questionnaire ranged from 5 to 10 minutes. Other survey platforms that met all the needs for performing this online survey were found, but cost was the final deciding factor. There was a learning curve in applying some of the more complex features, but self-help (FAQs, online videos, and support) was found to be readily available. The main disadvantage of the survey platform used was the limited capabilities for the size, formatting, and layout of the graphics used in conjunction with the questions.
Before sending out the survey to potential participants, a duplicate “sand box” test site was set up. Members of the research team took the survey using various web browsers, hardware platforms (desktops, laptops, smart phones, tablets), and operating systems. Doing so allowed for debugging to ensure (1) proper functioning of the layered questions, (2) sound survey results, and (3) ease of use for participants. None of the participants had any questions about the survey, and no technical issues were encountered during the survey. Additionally, the survey was configured so any person receiving the link could participate, allowing direct recipients to forward the invitation and survey link to others.
Survey Invitations
Various sources were used to identify potential participants. Additional contact suggestions were provided by TCRP panel members. The primary sources for participants included the following:
- American Railway Engineering and Maintenance-of-Way Association (AREMA) membership list
- APTA membership list
- The Pocket List of Railroad OfficialsTM (PocketList.com)
- Personal contact lists
Survey Transmission
Invitations to participate in the survey were sent to 46 electrified railways. The survey was transmitted via email using a standard transmission. Automatic email responses (e.g., bad email addresses, addressee no longer available) were followed up on to attempt contact with other possible participants from that organization. Follow-up phone calls were made wherever possible.
Dynamic Survey Questions
The number of survey questions participants were asked to respond to varied between 6 (nonelectrified railways with no IJ arcing) and 15 (electrified railway experiencing IJ arcing). The questions asked varied by the responses, so, for example, if the respondents stated they were not part of an electrified railway, the questions related to the type of electrification and distribution were skipped (hidden from the respondents). All survey participants had access to the same questions; however, the questions asked were varied so survey participants did not have to waste time responding to questions that were irrelevant to their system. The survey was structured so participants could change their responses during the survey, review their responses before saving/sending their responses, and print a copy of their responses. Details of the survey questions and the logic for the dynamic questions are provided in the last section of this appendix.
Survey Participant Confidentiality
Survey participants were ensured privacy so their answers could not be directly attributed to them personally or their employer. Additionally, during the interviews, MxV Rail informed the railroads that the data would be stripped of any information that could be directly attributed to their railroad, except for published or public reports. While required to receive information from many of the railroads, the confidentiality commitment was not detrimental to the work, since the information received can easily be presented in this guide to illustrate the information relevant to this work generally, conceptually, and without direct attribution. Additionally, the original graphics provided may be copyrighted, proprietary, or often required to meet the TCRP copyright requirements.
Chapter E-3 Survey Responses
Survey responses were obtained for 13 different systems. Eleven survey responses were submitted directly, and 2 additional survey responses were obtained directly through interviews. System voltage, obtained through research and interviews, was added to the survey data. Some of the responses were updated based on additional information obtained in the interviews. Table E-1 shows a summary of the technical responses.
Chapter E-4 Industry Interviews
Many electrified railways were reticent to discuss TPNR issues, citing concerns with possible outside perceptions concerning TPNR-related stray current issues. In the interviews, all were interested in the IJ arcing phenomenon. After the survey and interviews, some railways contacted the research team for more information and a description of low-level sparking issues at IJs, rather than severe destructive arcing (see Figure E-1).
The interviews with various transits extended over a much longer period and began before the survey was sent. Many of the initial interviews occurred during the literature review, as the team performed outreach to various electrified railways for internal reports concerning IJ arcing from various electrified railways. Also, as word that the TCRP research effort was underway spread through the industry, some electrified railways contacted the research team directly for information concerning IJ arcing. These contacts were used to extend the reach of the survey and interviews. At the same time as the D-20 work was underway, research team members were involved with other electrified railway projects unrelated to this project, and opportunities were taken to discuss IJ arcing and other TPNR-related arcing issues with those railways.
Long Description.
The table has 13 rows and 13 columns. The column heading from left to right are department, electric traction power distribution, supply voltage, DC traction power rectification, traction motor type, has your railroad experienced electrical arcing at insulated rail joints, when did electrical arcing at insulated rail joints issues begin, are there impedance bonds at the insulated rail joints (rain, snow, temperature, high or low humidity, water on track), has the electrical arcing at insulated rail joints been resolved, briefly describe the mitigations implemented or planned, how is the traction power and negative return distributed on cars or units, and how is traction power and return distributed on the cars or units. The rows are numbered 1 through 13.
Interview participants provided useful information that was used to focus the direction of the research efforts and analysis. The details of the various incidents directly related to this work that included a variety of situations where interviewees had experienced TPNR and IJ arcing were of great interest. Much of the information received was used for the case studies provided in this guide to illustrate the conditions and situations that may be conducive to arcing and mitigations that were applied. These case studies will be invaluable by providing examples that may help other electrified railways mitigate IJ arcing issues and aid others in the prevention of IJ arcing.
Chapter E-5 Key Takeaways
- IJ arcing is not normal for electrified railways.
- In most cases, IJ arcing is not caused by the failure of the IJ. Rather where large TPNR potential differences exist on adjacent running rails separated IJs, IJ arcing is caused by wheel-to-rail arcing or rail-to-rail arcing initiated by passing train wheels.
- Low-level IJ arcing (sparking) is more common than destructive arcing. Some of the respondents indicated that they are living with some low-level IJ sparking, since the issues are not causing operational or maintenance issues.
- IJ arcing is commonly found in locations where the TPNR for separate adjacent TPSSs separates/isolates from each other using IJs, by design.
- In most cases, IJ arcing appears to be related to the layout/design/capacity of the TPNR system.
- Destructive levels of IJ arcing are not common. Only two of the responding railways noted destructive IJ arcing issues at multiple locations.
- Many respondents had experience with destructive IJ arcing, and a commonly cited cause was loose, missing, or defective TPNR cable connections or cable connections to the running rails (bonds).
- At any IJ location, arcing may only occur on some but not all trains.
- When IJ arcing occurs, it is common for the condition to be initially misdiagnosed as a “bad IJ” and for arcing to recur after the installation of a new IJ. In some cases, the IJ was replaced multiple times. The underlying issue(s) were further investigated only after the arcing recurred.
- One electrified railway detailed its experience with IJ arcing issues (low-level sparking) related to rail end metal flow over IJ end posts.
- One electrified railway noted arcing issues at IJs leading to maintenance shop building where the IJs separated the TPNR from the earth-grounded rails in the shop.
- IJ arcing issues can result in detrimental impacts to other railway systems, including train control equipment, switch machines, conduits, concrete (rebar), generation of electrical touch hazards, etc.
- Typical successful mitigation results were obtained during a team effort between multiple maintenance and engineering disciplines (track, train control, electrical, vehicle maintenance).
- IJ arcing was reported on systems that distribute their traction power by both third rail and OHC.
Chapter E-6 Survey
Figure E-2 through Figure E-13 show the survey.
Long Description.
Below is a link to a Survey related to destructive electrical arcing at insulated rail joints. This Survey has at most 15 questions and takes approximately 5 to 10 minutes to complete. It is designed for both desktop or laptop computers and smartphones. Survey Link. Https colon forward slash forward slash forms dot office dot com forward slash r forward slash evTUrPCzYs question mark origin equals lprLink. A QR code. This survey is a part of a Transit Cooperative Research Program (TCRP) sponsored research project D-20 “Investigation and Mitigation of Insulated Joint Electrical Failure” https colon forward slash forward slash apps dot trb dot org forward slash cmsfeed forward slash TRB NetProjectDisplay dot asp question mark Project ID equals 5066. The phenomenon that is being investigated is not related to mechanical failure or deterioration of insulation on rail joints. It is related to electrical arcing across the insulation as train wheels pass over the insulating end posts, which has been experienced on some electric railroads. Even if you do not experience such electrical arcing, a response is still appreciated. This aids in quantifying how widespread this issue is. Feel free to forward his survey to others within your organization or to peers at other railroads. Multiple responses from the same railroad or organization are appreciated. The information you provide will be summarized and generalized, but not directly attributed to, or associated with you, or your company. Only the MxV Rail research team will have access to your response and communications. Please contact me if you have any issues accessing the survey, have questions, or would prefer to provide the information in another manner. Thank You. MxV Rail, A subsidiary of the Association of American Railroads (AAR).
Long Description.
The logic diagram has nine sections, each with one or more questions. The heading at the top section (Section 1) reads, Header, Electrical Arcing at Insulated Rail Joints. The first section has 4 questions, Section 2 has one question, Section 3 has 2 questions, Section 4 has one question, Section 5 has 4 questions, Section 6 to 8 have 1 question each, and Section 9 has a footer that reads, end of questionnaire. Thank you for your participation. An arrow from Section 1 leads to Section 4. An arrow from Section 4 leads to Section 7. An arrow from Section 1 leads to Section 3. An arrow from Section 4 leads to Section 5. An arrow from Section 7 leads to Section 9.
Long Description.
The page shows photos of rail joints at the top, followed by the text as follows. This is a survey of railroads related to electrical arcing at insulated rail joints. This information is for a Transit Cooperative Research Program (TCRP) sponsored research project D 20, Investigation and Mitigation of Insulated Joint Electrical Failure, https colon forward slash forward slash apps dot trb dot org forward slash cmsfeed forward slash TRBNetProjectDisplay dot asp question mark Project ID dash 5066. The information you provide will be summarized and generalized, but not directly attributed to, or associated with, you or your company. If you have any questions about this survey or would like more information about electrical arcing at insulated rail joints please contact, MxV Rail, a subsidiary of the Association of American Railroads (AAR) at aar dot com, www dot mxv rail dot com.
Long Description.
The page shows a question as follows. 5. If DC traction power, what type of rectification with radio buttons for simple, multiphase, and don’t know?
Long Description.
The page shows two questions as follows. 6. What type of traction motors are used on your motive power slash revenue vehicles with radio buttons for direct current (DC) traction motors, alternating current (AC) traction motors, both DC and AC traction motors, and don’t know. 7. How is traction power and return distributed on the cars or units with radio buttons for the illustration, for each truck or axle, on each car or unit, train lined across multiple cars or units, don’t know, and a blank space with text other.
Long Description.
The page shows a question as follows. 8. Has your railroad experienced electrical arcing at insulated Rail Joints with radio buttons for yes and no?
Long Description.
The page shows 4 questions as follows. 9. When did electrical arcing at Insulated Rail Joints issues begin with check boxes for since the beginning of operations, with the introduction of new vehicle fleet, after modification or upgrades to the traction power system, signal system, or track, and unknown. 10. Are there impedance bonds at the Insulated Rail Joint electrical arcing occurs with radio buttons for yes, no, and unknown. 11. Does weather influence electrical arcing at Insulated Rail Joints? (rain, snow, temperature, high or low humidity, water on track) with radio button for yes, no, and unknown. 12. Has the electrical arcing at Insulated Rail Joints been resolved with radio button for yes, no, and mitigations are planned or underway.
Long Description.
The page shows a question as follows. 13. Briefly describe the mitigations implemented or planned, followed by a blank space with text, and enter your answer.
Long Description.
The page shows a question as follows. 14. Are you or someone in your organization interested in participating in an interview to discuss electrical arcing at Insulated Rail Joints with radio buttons for yes and no?
Long Description.
The page shows a question as follows. 14. Contact information for interview, followed by a blank space with text to enter your answer.
Long Description.
The page reads as follows. End of questionnaire. Thank you for your participation. If you have any questions or would like more information about this research, please contact MxV Rail, A subsidiary of the Association of American Railroads (AAR) at aar dot com, www dot mxv rail dot com. Please click, submit, below to save your response. A blank space with text to enter your answer is at the bottom.
This page intentionally left blank.
