Railroad Protection
EMA recently published a paper at the AREMA Railway Exchange conference in Minneapolis, MN entitled: “SYSTEM LEVEL FULL SCALE LIGHTNING TESTING OF PTC WAYSIDE C&S SYSTEMS.” A summary of the paper follows:
Positive Train Control requires a high degree of reliability and availability. In order to meet these requirements, CSX needed to ensure the lightning protection of its wayside installations. In order to verify the CSX lightning protection design, an extensive full scale test and validation of all lightning protection devices and their applications was accomplished.
The following items are discussed:
- CSX Lightning Protection Approach
- Summary of CSX PTC Lightning Protection Features
- PTC Lightning Test Articles
- Application of Numerical Simulations of Lightning Strikes to Wayside Signal Systems
- Example Test Results
- Summary of Results
- Recommendations for Improvement
A PowerPoint presentation of this paper can be viewed at the bottom of this page.
INTRODUCTION
CSX’s implementation of Positive Train Control must ensure the reliability of wayside equipment.
A critical part of reliability is effective lightning protection. In order to verify the lightning protection approach, CSX performed extensive full scale testing of wayside PTC installations. This PTC system-level lightning test is the first of its kind in the railroad industry.
This test of two fully CSX configured PTC bungalows to threat level lightning is an innovative concept that provided a unique opportunity for verification of the CSX lightning protection approach that has been applied since 1998.
The test demonstrated the effectiveness of the protection approach, and also provided suggestions for future improvement.
CSX LIGHTNING PROTECTION TECHNICAL APPROACH
The Foundation of CSX Lightning Protection
In 1998, CSX decided that the foundation of its lightning protection approach must be consistent with the following objective:
Eliminate Train Control service interruptions and signal equipment losses due to lightning related incidents.
This objective is consistent with that of other highly reliable systems in other industries for which there is no option for system failure. These include military command and control facilities, commercial aircraft, and other ground based critical systems.
Basis of Approach
The CSX protection approach is similar to that of other industries, and is based on the considerations of stress vs. strength.
In this case, stress is the incident lightning environment, which is quantified in terms of peak current, available energy, available charge, and current rate of rise.
Strength is a measure of the ability of the system to withstand the stress.
Stress and strength are compared to create the effectiveness of the protection approach. This effectiveness is stated in terms of the margin which equals the ration of strength/stress. For effective protection, the margin should be much larger than 1. If the stress is greater than strength, more protection is required.
Stress, the strength of the lightning environment, can be determined from several sources. First, field experience can be used to estimate the lightning levels that would be responsible to cause the observed damage. Second, Computational Electromagnetics can be used to perform numerical simulations of lightning strikes to wayside systems in order to characterize lightning currents that are incident on the wayside electronics and protection devices. Finally, the AREMA Signal Manual provides guidance in the lightning environment suitable for protection design.
SUMMARY OF CSX PTC LIGHTNING PROTECTION FEATURES
Basic Lightning Protection Rule
The basic rule of lightning protection is: Don’t allow lightning inside the bungalow! The critical PTC electronics are inside the bungalow, so if lightning cannot penetrate the bungalow, then the electronics are safe.
In order to accomplish this, the following protection features are implemented:
- Metal bungalow
- Faraday cage
- GE Tranquell AC arrester
- RF coaxial in-line bulkhead arresters
- Hybrid Low Voltage Arrester (HLVA) arrester for circuits having operating voltages less than about 30 volts
- Grounding of spares and shields at both ends of cables
These are discussed in the following paragraphs.
Metal Bungalow
The metal skin provides an excellent shield upon which to apply protection for the lightning points of entry (POEs).
 Aluminum house providing the basic barrier for lightning shielding of critical electronics |
|
Faraday Cage
The cage is basically an extension of the bungalow surface inside the house, such that the house interior is a clean environment, and the interior of the Faraday cage contains the arresters and is the dirty external environment. Clean wires exit the cage with a very short lead length from the arrester, minimizing dirty environment coupling to the clean wires.
 The CSX Faraday cage |
|
GE Tranquell AC arrester
This is mounted in its own small metal box (another Faraday cage) and has nearly zero lead length for its arrester return paths. The surface area and small lead length of this arrester ensures a low impedance path that will minimize let-through voltages.
 GE Tranquell AC power protection |
|
RF Coaxial In-Line Bulkhead Arrester
Below is an interior view of a bulkhead mounted RF coaxial cable arrester.
 Interior view of bulkhead mounted RF coaxial cable arrester |
|
Hybrid Low Voltage Arrester (HLVA)
This is the HLVA that is used to protect low voltage circuits. It has the smallest let-through voltage of any arrester CSX has tested since 1998. It has never failed in a shorted mode in either extensive testing or in the field.
 CSX Hybrid Low Voltage Arrester |
|
Grounding of Spares and Shields
The image illustrates the grounding of shields and spares at both ends. The grounding is done within the dirty volume of the Faraday cage.
 Grounding of shields and spares |
|
PTC TEST OVERVIEW
There were 155 test shots with as much as 120 kA injected lightning current as follows:
- Track circuits (Electrologixs) and arresters: 73 shots
- Signal lamps (Electroblox and relay) : 31 shots
- Communication and I/O cables between houses: 25 shots
- AC power: 7 shots
- RF links: 19 shots
- Satcom: 10 shots
- iNet: 5 shots
- Cellular: 4 shots
PTC LIGHTNING TEST CONFIGURATION
The test articles consisted of two fully configured PTC bungalows as shown below. The internal equipment consisted of the following:
- CNA 2000 Communications Network Adaptor
- ElectroLogIXS VLC and EC5
- ElectroBlox Standalone Wayside Interface Unit
- MDS iNET 900 transceiver
- A shielded six pair cable connection was also included from the A house to the B house, of which one pair was used.
- A shielded 12 wire signal I/O cable also connects the A and B houses
- Satellite connection with the iDirect Evolution 5 box via a satellite dish
- Cellular connection via the Digi Transport WR44 box
- RuggedCom RS930L switch
- Battery chargers and batteries
- Internal Faraday Cage containing arresters
- House ventilation and lighting systems
CSX provides redundant IP communication paths into the PTC equipment enclosures as follows:
- Local IP connection via telecom cable if available
- Cellular connection
- Satellite connection
- A few remote locations may use a 220 MHZ radio to communicate to the locomotive directly instead of the IP connection back to the control center and to the 220 MHz base stations.
Two PTC bungalows for lightning testing
NUMERICAL SIMULATIONS OF LIGHTNING STRIKES TO WAYSIDE SYSTEMS
The PTC lightning test was supported with computational electromagnetic simulations of strikes to wayside installations. This provides a first principles estimate of the lightning current waveforms (i.e., the stress) incident on wayside systems to supplement AREMA waveforms.
Stress is a combination of the following attributes:
- Peak current with units of amperes
- Available energy: Action Integral = ∫i(t)2dt with units of Joules/ohm (also = A2-seconds)
- Total charge: ∫i(t)dt with units of Coulombs
Lightning strikes to a typical CP location were numerically simulated with the software EMA3D from Electro Magnetic Applications (www.ema3d.com), which has been used on aerospace and military projects worldwide.
The incident lightning current is the standard 1% waveform commonly used for protection design of aircraft and ground stations.
The computational model is shown below. Four 200 kA lightning strikes were analyzed as follows:
- Strike to Bungalow
- Strike to Track
- Strike to Ground
- Strike to Signal Mast
The computed results included currents on track wires and signal wires. An example is shown below. The largest lightning currents induced on track wires and signal lamp wires are as follows:
- Track wires
- Largest, for a strike to a rail
- Amplitude: 95 kA
- Action integral: 430 kJ/Ω
- Second largest, for a strike to the bungalow or signal mast
- Amplitude: 26 kA
- Action integral: 55 kJ/Ω
- Signal cable, for a strike to a signal mast
- Amplitude: 170kA
- Action integral: 1600 kJ/Ω
- Largest, for a strike to a rail
Wayside EMA3D computational model
Example of computed current flow
PTC TEST RESULTS SUMMARY
The results of the PTC test are summarized as follows:
- No electronic units in the houses were damaged.
- Momentary upsets were observed, all self recovered
- Electrologixs: 30 sec recovery
- iNet radio: momentary dropout, 30 sec recovery (closing the Faraday cage doors remedied this)
- VDSL: momentary dropout: 30 sec recovery
- AC Protection: GE Tranquell performed best of the arresters tested
- HLVA was the best performing low voltage arrester
- RF interfaces: all passed, but inconsistent behavior of Satcom arrester
- Bonding of cable shields and spare wires to bungalow skin on both ends is a highly effective protection method
- Faraday cage: highly effective
- Clean/dirty isolation
- Arrester ground lead short for minimum let-through
SUGGESTED DESIGN CHANGES
Several design changes were recommended as a result of this test: These included:
- The Tranquell installation could be improved by using MOVs on only L1 and L2, and by connecting the neutral directly to bungalow skin ground.
- Bulkhead arresters be added to the Satcom antenna design to eliminate lightning current penetration into the house.
- The low voltage arresters on the signal lamp circuits could be eliminated
