CBTC: will SEPTA be first? - Transit Update - radio-frequency communications-based train control - Brief Article

Railway Age, June, 2002 by Tom Sullivan

SEPTA's Subway Surface Trolley Line may go into full revenue service with radio-frequency (RF) communications-based train control before MTA New York City Transit or BART. Could a conservative design based upon COTS (commercial-off-the-shelf) products and open standards be the reason?

SEPTA Deputy Chief Engineer John LaForce often likes to describe the agency's CBTC needs this way: "We want an apple off the cart." In other words, by keeping things simple, SEPTA hopes to reduce project risk and schedule delay. It's looking like a good strategy.

This approach may work, because Flexiblok[TM], Bombardier's (former Adtranz) RF CBTC system being installed at SEPTA, is also being installed at Seattle Tacoma and San Francisco airports. San Francisco is in final testing and Sea-Tac expects to have substantial completion next year.

Why is RF CBTC so hard to implement? Inductive loop CBTC systems, in operation around the world, have proven to be safe and reliable for decades. But with RF, it is difficult to guarantee high availability and continuous communications coverage in subway tunnels. Complex and diverse structures and narrow clearances between trains and tunnel walls conspire to cause significant signal fading, multipath, and other nasty forms of interference. Advanced modulation techniques such as spread spectrum help, but it's nearly impossible with RF to obtain coverage equal to that of a simple near-field inductive loop.

For example, despite extensive testing and modeling, BART recently discovered it needs more trackside CBTC radios than it originally calculated. Recognizing most suppliers were operating on the bleeding edge of RF CBTC, Bombardier took a more conservative and traditional approach to RF coverage in tunnels.

To meet SEPTA's low-risk project approach, Bombardier's Joe Zwastetzky decided to retain the leaky coax antenna design pioneered by its radio partner, Andrew Corp., for use in subway tunnels. More commonly known by its trade name, Radiax[R], leaky coax-based systems have open slots along the length of the transmission line to permit the signal to propagate continuously along the right-of-way right next to carborne antennas.

Much like inductive loops for first-generation CBTC systems, leaky coax has provided reliable and continuous voice radio coverage in tunnels for decades. But second generation RF CBTC systems based upon leaky coax provide a potential side benefit: the ability to lease additional RF services such as mobile phone and other new digital wireless services using the same antenna.

Zwastetzky, the father of Flexiblok[TM], also selected a common communications protocol and used many COTS products in the design. For example, Flexiblok uses IEEE-1473-L, the open Control Network standard based upon LonWorks, to monitor and control trackside RF amplifiers that feed the five miles of leaky coax in SEPTA's subway. SEPTA's design also is based upon the same standard radio frequency identification tag attached to every U.S. freight car and that costs under $40 apiece.

In SEPTA's case, the tags are placed on each railcar and read by wayside readers. But there are also 139 similar tags along the subway tracks. Called "Norming Points," by Bombardier, these are used as milestone markers and read by carborne readers to determine absolute train position.

SEPTA's CBTC tunnel work is complete and half of its Subway Surface Trolley Line fleet has been retrofitted. Operational tests are planned for this summer. Many are watching NYCT's Canarsie Line and BART's Advanced Automatic Train Control RF CBTC projects. But it may take several years before either of these two programs will be able to show the measurable increase in performance and safety SEPTA hopes to demonstrate sometime next year.