CBTC vs Conventional Signaling
Fixed-block signaling is not one technology but a century-long family: mechanical interlockings, relay logic, color-light aspects, cab signaling. CBTC is a tightly defined architecture in IEEE 1474.1. This sheet compares them where the comparison is fair — on the eight axes that change what a line can actually do — and attaches a real number to each.
Free to use and share with attribution. No email required.
What's inside
- All eight axes in one table, from separation logic to supply-chain concentration.
- Capacity figures from real conversions: Paris Line 1, Hong Kong Tsuen Wan, and the NYCT L Line.
- The capital-cost ranges per 10 km for a conventional resignaling versus a CBTC retrofit.
- Wayside hardware counts before and after — where the maintenance workload actually goes.
- Why every other difference on the list follows from the first one.
One difference, seven consequences
A fixed-block system divides the track into sections and lets one train occupy each. Minimum headway is therefore the time to traverse a block, plus the following train's braking distance, plus margins. A zone controller in a CBTC system knows each train's position to within one to two meters, knows its declared braking performance, and computes a Movement Authority that ends a safe distance behind the rear of the train ahead. The buffer is virtual and resized continuously.
Everything else on the sheet follows from that. Detection moves from the wayside to the train. The capacity ceiling rises 20 to 40 percent. Safety moves from operator compliance with a discrete aspect to continuous supervision, so the SPAD disappears as a category rather than being reduced. A hundred and fifty wayside signal heads collapse into a handful of zone controllers. And the supplier base narrows from a deep, second-sourceable market to roughly five companies.
The numbers worth quoting
Paris Métro Line 1 went from about 20 to 27 trains per hour after its Siemens conversion. Hong Kong's Tsuen Wan Line went from 18 to 24. The L Line, the US flagship, sustains 24 to 29 — up from roughly 20, with no civil works. On cost: a conventional resignaling on a typical US metro line runs $8–15 million per 10 km of track; a CBTC retrofit on the same line runs $15–25 million per 10 km before fleet equipment. Those two ranges, side by side, are the entire capital argument.
Sources & method
Adapted from Communications-Based Train Control, Volume 1, Chapter 1, by Chunjun (Francisco) Wang, and the companion article on the eight differences. Capacity figures are for peak trains per hour per direction on the named lines; cost ranges are US brownfield practice and exclude fleet equipment unless stated.