Slicing the Signal Block BNSF's Virtual Track Blocks Squeeze More Trains Onto the Same Rails

This application covers a method of railroad track control that takes an existing physical signal block, the stretch of rail between two insulated joints, and divides it electronically into several virtual track blocks, so that train positions and broken rails can be pinpointed to a fraction of a block. The applicant is BNSF Railway Company, one of the largest freight railroads in North America and a subsidiary of Berkshire Hathaway.

The Problem

Railways have kept trains apart for well over a century using block signalling: the line is partitioned into track blocks, and automatic signals, typically red, yellow and green lights, control movement between them. On a single direction track this lets trains follow each other with minimal risk of rear end collisions.

The specification identifies two stubborn disadvantages in the conventional approach. First, track capacity cannot be increased without building more infrastructure. If a railroad wants trains running closer together, it has to install additional signals and the control equipment that goes with them, block by block, along the line. Second, conventional block signalling cannot pinpoint a broken rail. A track circuit sees an electrical discontinuity somewhere in a block, but once a train is sitting in that block shunting the circuit, the system loses its ability to tell rail damage from normal occupancy. For a heavy haul freight railroad, an undetected rail break is one of the classic precursors to derailment.

What This Invention Does

The method in claim 1 is compact: partition a physical track block, defined by first and second insulated joints at each end of a length of track, into a plurality of virtual track blocks; detect an electrical circuit discontinuity in one of them; and generate a virtual track block position code indicating which virtual block the discontinuity is in. In the preferred embodiment each physical block is cut into four virtual segments, so the system knows not just that something is in the block, but which quarter of the block it is in.

The machinery behind that sits in the signal control houses already located at the insulated joints. Each house transmits two kinds of coded signal through the rails themselves. The first, called TC-A, is a coded message exchanged between neighbouring houses, similar to the Electrocode signalling commonly used by railroads: as long as house A can hear house B through the rail, the rail between them is intact and unoccupied. The second, TC-B, works more like an echo sounder. A transmitter and receiver at the same location send energy down the rail and read the reflection, and the signature of that reflected energy is proportional to the distance from the insulated joint to the nearest axle of the train. That distance measurement is what places a train, or a break, within a specific virtual block.

The occupancy picture is packaged as a Virtual track block Position message, a string of bits such as 11111000, one bit per virtual block, and transmitted to the computers onboard locomotives in the vicinity, preferably over a wireless link. Because train spacing is now managed against braking capability by the onboard computer rather than by lights beside the track, the specification notes that wayside signals are no longer needed at all. The document walks through worked scenarios with tables of TC-A, TC-B and position codes as a train moves through successive blocks, including operation with multiple trains inside the same physical block.

Key Features

• Physical blocks divided into virtual ones. A block bounded by two insulated joints is partitioned electronically, in the example into four segments, without laying new cable or adding wayside hardware between the joints.
• Two complementary rail signals. TC-A codes pass between adjacent signal houses to prove rail continuity, while TC-B reads reflected energy from the same location to measure how far away a train’s nearest axle is.
• Position codes, not just occupancy. The system generates a virtual track block position code identifying exactly which segment contains a train or a discontinuity, rather than a single occupied flag for the whole block.
• Broken rail detection inside occupied blocks. Because occupancy is resolved to virtual block level, a rail break can be detected within a physical block even while a train is in it, something conventional circuits cannot do.
• Braking managed onboard. Position messages are radioed to locomotive computers, which maintain braking distance directly, removing the need for lineside colour light signals.
• Capacity from existing infrastructure. Trains can safely follow each other at virtual block spacing rather than full physical block spacing, increasing line capacity with the track and joints already in place.

Who Is Behind It

BNSF Railway Company, headquartered in Fort Worth, Texas, operates one of the largest freight rail networks in North America, with roughly 32,500 route miles across 28 US states and parts of Canada, hauling coal, grain, intermodal containers and industrial goods. It has been wholly owned by Warren Buffett’s Berkshire Hathaway since 2010. It is unusual among applicants on the Australian register in being a railroad rather than a signalling supplier: this is an operator patenting its own control technology.

Four inventors are named: Jerry Wade Specht, Ralph E. Young, Kent Robert Shue and Mitchell Wayne Beard. The application is a divisional of Australian application 2024200532, itself a divisional of 2022246424, in turn a divisional of 2018261733, the Australian national phase of PCT/US2018/030325. The family claims priority from US provisional 62/502,224 filed on 5 May 2017.

Why It Matters

Railway capacity is usually bought with concrete and copper: more passing loops, more signals, shorter blocks. This specification pursues the same outcome in software and coded track signals, which is the same philosophy behind moving block signalling on modern metros and the onboard enforcement of positive train control mandated on US main lines. Fitting virtual subdivisions onto legacy track circuits and insulated joints is a pragmatic middle path for a network of BNSF’s size, where full resignalling would take decades.

The filing strategy tells its own story. A 2017 priority date carried through a PCT filing, an Australian national phase and three successive divisionals to a 2026 application indicates a family being kept deliberately alive, with claim scope adjusted at each generation. Australia is a natural jurisdiction for it: the country’s iron ore and coal heavy haul railways run some of the longest and heaviest freight trains in the world, and capacity and broken rail detection are exactly their operating constraints.

Related Concepts

• Automatic block signaling – the conventional fixed block system the virtual scheme is designed to improve on.
• Track circuit – the rail based electrical detection principle the TC-A and TC-B signals build upon.
• Moving block – the fully dynamic train separation concept that virtual blocks approximate using fixed infrastructure.
• Positive train control – the US onboard enforcement framework that the wireless position messages complement.
• Derailment – the failure mode that undetected broken rail can lead to.
• BNSF Railway – background on the applicant railroad and its network.

---

AU 2026202101 was published in the Australian Official Journal of Patents on 9 April 2026 and is open for public inspection. Patent applications represent inventions that are sought to be protected and do not necessarily reflect commercially available products.