PART III: The devil is in the details.
While so-called “policy makers” generally practice the proverbial 50,0000-foot view, it is necessary for someone else to be (also in the proverbial sense) “in the weeds,” that is, checking the landing gear. This is a requirement to ensure that the “aircraft” will bring the “policy” folks to a safe landing. The saying of a few generations ago was that “the devil is in the details.” Large engineering programs succeed or fail on details rather than general concepts. Given that the concept of upgrading the NEC beyond NECIP functionality and criteria to those associated with VHSR (very-high-speed rail) is a good objective, it is necessary to look at some of the details in order to ascertain if this objective is desirable—that is, is the payback worth the cost and effort? As noted in Parts I and II, the New Jersey High Speed Rail Program (NJHSRIP) provides the primary source of data for such an analysis, particularly with respect to the cost data, including service impacts and production time as well as money. This program is also one of the primary data sources for the benefit side.
The NJHSRIP is being executed on a fast-track schedule, effectively as a design-build rather than a design-bid-build effort. The Program Requirements Document (PRD) along with the System Safety Plan established much of the design criteria for the individual rail systems (track, traction power, right-of-way, train control). Since it was necessary to commence fabrication, installation and construction prior to completion of all designs, a robust system safety and systems integration effort was employed from the program’s inception. The development and execution of the program is best reviewed by considering original design concepts; design changes necessitated during installation/construction and integration of the independent rail systems to optimize investments; innovations, including technical and business methods; and system safety requirements.
Original design concepts
NJHSRIP’s signalization design was a derivative of the HDIS (High Density Interlocking Signal) system deployed in the mid-1990s on the High Line, the two-track portion of the NEC between Newark and New York. This is a fixed-block, nine-aspect, cab/no-wayside ATC system that is not to be confused with ACSES (Advanced Civil Speed Enforcement System, Amtrak’s version of PTC, an overlay on the basic ATC system). Codes are delivered at two separate and distinct frequencies.
The second frequency gives the speed upgrade. Only selected trains are able to read it. For example, Amtrak trains are capable of reading the codes for MAS (maximum allowable speed) of 125 mph. Freight trains, which do not operate on the High Line, were assigned an MAS of 50 mph. This assignment was jointly worked out by Amtrak with Conrail, CSX and Norfolk Southern, and required FRA concurrence. This route and aspect will become the new NEC standard for cab signals/ATC.
Of note is the block layout. Because of the CONOPS (Concept of Operations) prescribing the primary diversion route of a high-speed train to an outside track, it was necessary to provide at least a 125-mph cab speed on tracks 1 and 4. It was also desirable to provide higher capacity (within the 90- to 100-mph speed regime) on these tracks.
This necessitated the use of extremely short blocks. Where 4,500-foot blocks were used on the inside (No. 2 and No. 3 HSR) tracks, 3,000-foot blocks were used on tracks 1 and 4. Figure 1 provides a typical arrangement of blocks, comparing tracks 1 and 4 with 2 and 3. Thus, as a direct result of the operational requirements as expressed in the CONOPS, signalization of the outside tracks was significantly more costly than on the two designated high-speed tracks. Presumably, this concept will also apply elsewhere on NEC, since it provides a workable solution to the capacity lost when signaling for the long braking distance required for VHSR. This allows the superimposition of VHSR on sections of the NEC that also support heavy commuter traffic, without degradation to commuter train capacity, assuming such capacity is based upon 90- to 100-mph braking.
Since the track in this territory has traditionally been well-constructed and maintained, most of the improvements were limited to interlockings. However, MIDWAY interlocking at MP 41.7 was the exception. A universal interlocking consisting of all No. 20 (45-mph) crossovers, MIDWAY’S retirement and replacement with a high-speed interlocking had been planned since NECIP. Such planning at the policy level gives budget makers an opportunity not to invest, and as a consequence this interlocking, located within the existing 135-mph territory, was in questionable condition.
For example, gas-fired switch heaters fed from wayside propane tanks were still in use. These heaters are routinely blown out by the passage of high-speed trains, thereby requiring full-time coverage by B&B employees during storms. Based upon the CONOPS, train movements presently scheduled at MIDWAY will be relocated to the two new high-speed (80-mph) interlockings located at DELCO (MP 32) and ADAMS (MP 34). MIDWAY will remain for use only as a “block breaker.”
After significant internal debate, the decision was reached to replace MIDWAY with new crossovers, but these would again be No. 20s, suited for 45 mph, not high-speed crossovers. This decision was driven by maintenance costs and continuing reliability as much as by capital costs: A four-track universal interlocking comprised entirely of high-speed crossovers would be nearly two miles between opposing home signals and would contain 60 switch machines, all equipped with snow melters. Performance of such relatively simple requirements as FRA-mandated monthly obstruction tests become a challenge.
One major improvement was the widening of track centers within the interlocking from 12.5 feet to 15 feet. This allowed improved crossover geometry. Former NJ Transit Vice President Rail Operations Kevin O’Connor (now with Metro-North) fully concurred with this decision, based on reliability considerations.
Minimal bridge work was required. At certain locations adjacent to roadways, however, the safety case mandated installation of automotive-type guardrails to serve as barriers against vehicular intrusion.
The most significant right-of-way improvements centered on improved drainage and placing the new signal houses at an elevation to accommodate a major flood. A speed-restricted reverse curve limited to 130 to 140 mph depending on equipment considerations exists between MP 39 and MP 40. The original concept called for realigning these curves to achieve 160 mph; this proved impractical from environmental (wetlands), property-taking and cost viewpoints, so the alignment will remain as is. This decision would be important to the cost containment required for catenary renewal. It was also desirable to widen track centers from the nominal 12.5 (and in some cases lower) feet. This was accomplished at interlockings, but for institutional reasons not throughout the designated 160-mph territory, resulting in other safety mitigations.
Catenary considerations
Improvements to the electric traction (catenary) system include a new frequency converter rated at 70 to 80 megawatts located at the existing 25-MW facility in Metuchen, and a new substation near Trenton. A new cable-in-trough signal power distribution system will replace the original Pennsylvania Railroad open-wire aerial distribution. This is expected to significantly reduce weather-based failures.
The original program scope also called for installation of independently registered constant-tension catenary over all four tracks for the full program length (MP 32 to MP 54). The conversion to constant-tension catenary proved to be one of the most problematic aspects of the program. The original PRR catenary is a heavy-duty, fixed-tension system with its own 25 Hz, 138 KV transmission lines carried on an overbuild.
The presence of a commercial utility overbuild (Public Service Electric & Gas) atop the railroad overbuild raised significant constructability and work scheduling issues. The existing catenary poles are spaced approximately 250 feet apart along the right-of-way. Placing new poles at 180 feet apart or less would require an outage of the utility lines, and ultimately their relocation to new structures.
The trolley wire and associated auxiliary and messenger wires for each track are generally supported by body spans—cabled assemblies that span all four tracks and supported by catenary poles on either side. Under this design, a pantograph-caused dewirement on one track is likely to tear down catenary on adjacent tracks.
A major improvement to reliability will be attained by replacing the body spans with a beamconverting each location to a portal structure. Thus, all new or converted structures will be portals, allowing independent registration of catenary over each track (below) directly to the new portal beam.
The original design concept was based on optimization of pantograph/wire dynamics for a 160-186 mph speed regime. This design called for spacing of portal structures at variable distances, but in no case could the new structure spacing exceed 180 feet with a constant-tension wire design. This created several major constructability, investment cost and maintainability challenges, including:
• Construction of new foundations along an active right-of-way, part of which is through designated wetlands.
• Erection of more than 400 portal beams over an electrified railroad that supports 24/7 operations.
• Retention, repair and maintenance of the old catenary structures until the Amtrak transmission lines can be relocated (currently this is not funded). The PSE&G overbuild would also require relocation.
• Coordination with PSE&G and the N.J. Bureau of Public Utilities to obtain the utility transmission outages required for erection of new poles. Since this is an electrical grid trunk line, these outages are seasonally limited, adding an undesirable scheduling constraint to catenary renewal.
Further engineering analysis indicated that speeds on the order of 140-145 mph could be achieved with new fixed-tension catenary using the existing pole spacing, with body spans converted to portal structures. This speed roughly matched that which would be achieved on the existing alignment for the reverse curve between MP 39 and MP 40. Analysis of train performance indicated that, given this restriction, consideration of high-speed braking and acceleration rates showed that a speed in excess of 140 mph could not be achieved between MP 39 and MP 32 (the eastern limit of catenary renewal). Thus, the 160-mph target between the locations was essentially a “paper” speed.
Also, the PSE&G overbuild extends east of MP 32. This, and a curve on a stone viaduct over the Raritan River (MP 31) eliminated any possibility of exceeding 140 mph east of MP 32. In fact, all the way through to Penn Station New York, there is no chance of exceeding even the current Acela Express MAS of 135 mph. East of MP 32, the timetable MAS is 125 mph for only about 10 of the 33 miles. Due to the curve-heavy alignment, there is very little opportunity to increase MAS for any class of train (including the anticipated new Tier III high-speed equipment) above the current timetable speed of 125 mph. Further analysis showed that, between MP 32 and MP 41, the running time difference between 140 mph and 160 mph is less than 30 seconds.
An integrated review of right-of-way, catenary, operational performance and capital and maintenance costs resulted in Amtrak’s decision to retain fixed-tension catenary between MIDWAY and MP 32. Such decisions are common in fast-track design-build programs as program needs are compared to a project’s schedule performance and “hard money” budgets. Such decisions are “made in the weeds,” based on a thorough knowledge of engineering systems, the railroad’s physical characteristics and operational requirements.
Upon review of data and the decision-making process, FRA funding-grant managers concurred with this change. The condition for FRA’s concurrence, however, was that the fixed-tension catenary would be comprised of all-new material and would be designed with portal structures and independent registration. The benefit of retaining fixed-tension catenary could be that the portal beams would be erected on the existing poles without the need for PS&G transmission line outages. As a result, two “beta” catenary configurations will emerge from NJHSRIP: one for fixed-tension and one for constant-tension. Based on their performance, each configuration will evolve to a standard applications design for use on the balance of the NEC.
Conclusion
Part III of this series has reviewed some of the engineering details and design adjustments that arose during the construction process. Many of these adjustments, such as the catenary changes, were driven by the hard money ($459 million)/hard schedule (5.5 years) requirements delineated in the grant. Part IV will discuss some of the innovative methods employed to achieve the grant requirements, including a special project labor agreement with BMWE.