Reducing Noise and Vibration on NYCT

by Bob Tuzik, January 1, 2005

Any problem affecting the New York City Transit is by default a big problem. The process of transporting more than four million passengers per day in 6,000 vehicles on more than 700 miles of elevated, underground at-grade mainline tracks, on a system that operates 24 hours a day, seven days a week, presents plenty of opportunity for problems to occur. One area that the NYCT and its riders have had to contend with is noise and vibration, particularly on elevated and underground lines. Vibration tends to create problems for the tenants of buildings near underground lines, while screech and wheel squeal are hard on residents’ and passengers’ ears.

Over the past several years, NYCT has implemented a number of track and vehicle treatments to reduce noise and vibration levels on the system. These treatments have reduced the average noise level on elevated and subway lines from 98 dBA in the 1970s and ’80s to an average 88 dBA since 1990, Antonio Cabrera, NYCT’s Director of Track Engineering, told delegates at Interface Journal and Advanced Rail Management’s Rail Transit ’04 Wheel/Rail Interaction Seminar.

On the vehicle side, quieter traction motors installed during scheduled maintenance programs reduced noise levels by 5 – 7 dBA. Ring-damped wheels, which were installed on new and overhauled cars, and the installation of composition brake shoes, which reduce screech during braking, reduced screech noise by as much as 15 – 20 dBA. A wheel truing program aimed at eliminating “flat” wheels has also reduced noise levels. “Trued wheels are 10 – 15 dBA quieter than flat wheels,” Cabrera said. New initiatives include the acquisition of 1,100 new cars with quieter traction motors, air compressors and composition brake shoes. NYCT’s motor shop is equipped with dynamometer to monitor vibration levels of remanufactured motors.

On the track side, NYCT has made modifications to several of its existing track types, which include:

• Ballasted track in a concrete invert (type I).
• Dual-block timber ties embedded in concrete (type IIM), which incorporates long ties every fourth tie to which the third rail is mounted. This design represents the standard for subway reconstruction. “If the track remains dry, it can last 100 years,” Cabrera said.
• Open-deck, steel elevated structure (type III).
• Ballasted track on fill, at grade (type VI).
• Direct fixation to structural invert (type VIII), which incorporates rubber rail seats resting on concrete or epoxy.

One of the first things NYCT did to reduce track-generated noise and vibration was to install cwr wherever possible. NYCT uses standard 115 RE rail section in open track, but is committed to a 110-pound ARA-B rail section in subway lines, most of which were built between 1904 and 1944, in order to maintain clearances. To date, NYCT has installed more than 320 miles of cwr—nearly half of the mainline track miles on the system. The use of cwr is not appropriate on old elevated structures, which require expansion joints every 200 feet, or in some tight radius curves on the system.

“Rails wear quickly in curves,” Cabrera said. “With our high traffic levels and 24/7 operation it would be very difficult to change out cwr.”

Overall, the use of cwr reduces noise by 5 to 7 dBA, compared to jointed rail on the system. The use of cwr also eliminates pounding at joints, which provides for a smoother ride and eliminates bolt hole cracks and broken rails. The use of resilient fasteners provides an additional 3-dBA reduction in noise.

A custom designed resilient, spring-loaded fastening system by Dynamic Engineering, which incorporates a resilient pad fixed between cast steel base and top plates that are fastened with springs under tension and clipped to the rail with a Pandrol Fastclip, has further reduced noise by 3 dBA to 5 dBA on elevated lines. This system, which is also used on subway lines, does not trap moisture (which can be a source of rail base corrosion), as rubber rail seats used in the Type VIII direct fixation to structural invert do.

The use of resilient plates on its subway track, which typically consists of timber tie blocks embedded in concrete, has been shown to reduce the track modulus from the standard 10,000 – 20,000 psi range to 3,000 – 4,000 psi, which approximates the modulus of typical ballasted track.

NYCT is also experimenting with an 800-foot section of Low Vibration Track, which it is considering for all future construction. LVT, designed by the Permanent Way Corp., features reinforced concrete blocks with resilient pads that are encased in a rubber boot around which concrete is poured. “This provides a resilient track with components that are replaceable, which is very important when operating a 24-hour-per-day, seven-days-per-week system,” Cabrera said.

Friction control

Lubrication and friction management have also been invaluable tools in reducing noise (and vibration, to some degree), particularly at stations in underground sections of the system. Grease is applied to the gauge faces of the running and restraining rails to control wear and reduce screech and squeal on sharp curves. For a heavy rail rapid transit system, NYCT has its share of sharp curves, many of which are 350-foot radius or less. The average noise on one 180-foot radius curve at the South Ferry underground station, for example, was 109 decibels—even with grease applied to the gauge faces of the running and guardrails. With trains operating at two-minute headways at times, the noise bordered on unbearable. A water spray treatment reduced the noise levels to 90 dBA, but degraded the ties and corroded the rails very quickly.

This convinced NYCT to adopt the use of Kelsan Technologies’ KELTRACK® top-of-rail friction modifier. This, in conjunction with gauge-face lubrication, reduced the noise levels to 85 dBA, offering better noise attenuation properties than the water spray, without its detrimental effect on the track structure, Cabrera said. Other tests of top-of-rail friction modifiers on a 300-foot radius curve on a 3.8% ascending grade showed a maximum average noise reduction of 7.3 dBA (an average 5.2 dBA in the 31.5 – 200 Hz frequency and 13.3 dBA in the 1,000 – 20,000 Hz frequency). Top-of-rail friction modifiers also appear to moderately reduce vibration (by 3 – 5 dB at 31.5 – 200 Hz), he said.

Only by looking at all aspects of the system—vehicles, track and operating conditions—NYCT has been able to effectively reduce noise and vibration on the system.

This article is based on “Maintenance and Noise Abatement Programs at New York City Transit,” a presentation by Antonio Cabrera, Director of Track Engineering, New York City Transit, at Interface Journal and Advanced Rail Management’s Rail Transit ’04 Wheel/Rail Interaction Seminar.