Of all the stock excuses given to explain a stalled train or a delay in service on the New York City subway, “signal problems” can be the most infuriating. It’s a vague excuse, offering neither the human element implied by a “police investigation” or a “sick passenger,” nor the mystery that comes along with a lack of acknowledgement of the delay.
But those signals are a vital piece of the subway’s infrastructure, and they’re a crucial element of the larger issues that have contributed to the deterioration of the transit system. Understanding the signals and how they work goes a long way toward understanding the subway itself—and how it’s become so messed up in recent years.
It’s a complex topic, so read on for an overview of how signals work, and what the MTA is doing to get them to work better.
In a nutshell, the signal system regulates the amount of space physically allowed between trains. This is a safety protocol that impacts each train’s speeds and times. Trains operating at the established speed limit can, in theory, cruise along without issue; trains that travel above the limit will have their emergency brakes automatically triggered when passing a signal.
You’ve likely seen a subway signal on your commute: They look like traffic lights, and use the same color pattern—red: stop, green: go, yellow: proceed with caution—to let train operators know if it’s okay to move in and out of a station.
Some of the signals that are currently in use throughout the subway system are extremely outdated—this MTA video shows the signal system at the West 4th Street station, which dates back to the 1930s. “It works, but it’s an antiquated way to run a subway,” Wynton Habersham, the former MTA head of subways, says in the video.
Signal systems in use: Block vs. CBTC
The subway uses two kinds of signal systems: Automatic Block Signaling and Communications-Based Train Control (CBTC).
Block signaling—the kind seen in the MTA video above—is a manually operated method that has been in use since the subway’s inception. It has two schemes: A Division and B Division, which were put in place when the subway was privately owned and operated by three separate companies (before the MTA was even a thing).
As with streets, subways have blocks, each typically some 1,000 feet long. Fixed-block signals are visible from subway platforms, and the information they provide to train operators are based on the location of the most recent train to have passed—this is known as a moving block system. But this method is imprecise, and because of the age of the signals, subway personnel do not actually know the exact location of the subway cars using block signaling. Much of the current system was installed from the 1930s to the 1960s, and requires custom replacement parts to be made in-house because the machinery is so outdated.
While block signaling relies on manual operation, CBTC uses automatic, computer-based signaling. The equipment involved in CBTC is less visible and far more durable, making it less vulnerable. It’s also far more precise than its manually operated counterpart: It’s currently in use on the L line, which “consistently operates with an on-time performance higher than 90 percent, roughly 30 percentage points better than the system as a whole,” as Aaron Gordon recently reported. CBTC was recently rolled out along the 7 line, but the implementation didn’t go quite as smoothly as anticipated.
Key dates in subway signal history
1904: Block signaling is used on the city’s first subway line, the IRT Lexington Avenue line.
1962: An experimental automated train is destroyed in a fire at Grand Central, delaying the installation of a modern signal system in NYC, despite the technology being adopted for transit systems elsewhere.
1970: The practice of keying-by—“slowly moving past a red light if the track ahead was clear,” as the New York Times put it in 1973—is made illegal, making red signals mandatory with no exceptions and increasing the amount of delays caused by signal problems.
1991: In 1991, a drunk motorman speeds through a signal, causing a subway to derail near the Union Square subway station, killing five. This catastrophe finally leads NYCTA to consider automating trains.
1993: The safety of the L line’s signal system is questioned following a crash where one train rear-ended another, sending 45 people to the hospital.
1995: Following a crash on the Williamsburg Bridge in which a J train hurtled past an ancient red light and into the back of an M train, killing the driver and injuring 54 straphangers, maximum train speeds were reduced and “grade-time” signals were installed to further limit train’s speeds in the name of safety. These precautions, however, have significantly added to many of the chronic delays straphangers experience today.
2009: The L line becomes the first in the system to be fully outfitted and operational with CBTC.
2018: The rollout of CBTC on the 7 line is completed. In the latter half of the year, NYCT’s SPEED unit—which is tasked with getting subway trains to move faster—begins to study signal timers throughout the system to ID ones that are malfunctioning, and thus causing trains to go slower.
The future of the signal system
Back to that “signal problems” issue: As we’ve noted, many of the subway’s signals are antiquated, so if one breaks—which happens regularly, given the age of the system—it can take forever to fix. Some signals are misconfigured—trains initiate automatic braking even if they’re going at or below the proper speed. And subway operators are penalized for tripping signals, which incentivizes them to run trains extra slowly and avoid castigation, according the New York Times. (That piece also illustrates the impact on the greater system when a single subway car is delayed.)
NYCT president Andy Byford has vowed to overhaul the old signal system as part of his Fast Forward plan, which calls for replacing block signaling with CBTC on 10 subway lines within 10 years. It sounds good in theory, but there’s a cash flow problem: The full Fast Forward plan is expected to cost up to $40 billion to implement, and as of right now, a dedicated funding stream does not exist. (Congestion pricing and a weed-for-rails tax may help with that.)
There has been some progress as far as signal maintenance goes: Thanks to NYCT’s Save Safe Seconds campaign and its SPEED unit, a chunk of the more than 320 improperly calibrated signal timers have been fixed, allowing trains in those blocks to move more quickly.
And then there’s ultra-wideband radio technology, which is not yet in use in any transit system, but has nevertheless been touted by Gov. Andrew Cuomo as a way to help the ailing subway. It uses a wireless radio system to transmit data, and would, in theory, help communicate information more quickly between signals. It was among the winning proposals in the MTA’s Genius Challenge earlier this year, but it’s unclear how well it would actually work.
Signal types and terms
Automatic Train Supervision: This technology enables train dispatchers to know more or less where trains using block signaling are in the system from Operations Control Centers, as well as what enables countdown clocks to be generally accurate.
Automatic Block Signaling: This is the formal name for the fixed block signal system, in which wayside signals are fixed and unmoving, and automatic train stops are in place as a safety precaution to prevent trains from speeding, derailing, or otherwise malfunctioning.
Block: The precise unit used to measure where a train is between signals. Blocks are bookended by track circuits, which conduct electric current, and insulated joints (also known as block boundaries). When a train is in a block, it’s said to be “occupied” until the train leaves; then, it becomes “vacant.”
Chaining: A bit of a wonky term that essentially boils down to how fixed block subway trains are located within the system. It’s a form of measurement by which a train’s distance is measured feet from a fixed point on each line—what is referred to as “chaining zero.”
Interlocking: When multiple tracks intersect—imagine large junctions like DeKalb Avenue or Times Square—they are connected by railroad switches and “interlockings,” which are more complex than the ones at non-intersecting track points. These are controlled by NYCTA employees from signal towers. When they break (and they often do), multiple lines are impacted, as straphangers are intimately familiar with.
Moving block system: Unlike fixed block signaling, CBTC is a moving block system, where speed and capacity are increased by a train’s superior ability to communicate and lack of dependence on fixed infrastructure. Moving block systems thus require less equipment and are capable of more transmissions.
Relays: Part of fixed block signaling’s communication infrastructure, allowing off-site personnel and machines to understand what’s going on with any given train.
Transponders: Installed between track rails, transponders communicate with interrogator antennas in CBTC-equipped cars, which then communicate with transit personnel via radio.
Wheel detectors: These use car axle movement to sense train speed.