The return rail has electrical resistance

A DC train completes its traction circuit through a return path that commonly includes the running rails. Those rails have finite resistance, and their insulation from surrounding ground is imperfect. Some current can therefore take parallel routes through concrete, soil or neighbouring metal. The 1995 Northwestern University report on electrified-rail corrosion describes how a buried gas main can collect current and carry it toward a region where it returns to the railway. A structure need not be deliberately connected to the railway to become part of that circuit.

Sources: sc-northwestern

Metal is lost where current leaves for the electrolyte

The crucial boundary is where conventional current passes out of a metal structure into an electrolyte such as moist soil or concrete. That region acts as an anode: metal can dissolve through an electrochemical reaction. Elsewhere, current may enter the same structure without producing the same local metal-loss mechanism. The Northwestern report consequently separates the current’s pickup point from its discharge point. This explains why finding a nearby railway is not enough to locate damage: the current route and the places where it crosses material boundaries matter.

Sources: sc-northwestern

Two resistances need opposite design treatment

Limiting leakage involves two complementary electrical aims. The intended longitudinal return route should have low resistance, while the route from rail to earth should have high resistance. A 2026 University of Zagreb study explains how individual fastening points contribute to rail-to-earth resistance. There are many of these paths along a route, so apparently small local leakage paths form part of a distributed circuit. This is why the electrical properties of the complete rail support matter alongside continuity of the steel rail itself.

Sources: sc-zagreb

A resilient component is not automatically an insulator

The Zagreb researchers tested two fastening arrangements under a controlled 26 V DC laboratory supply. Both allowed unwanted leakage; insulating anchor bolts in one arrangement did not solve the entire problem because its elastomeric elements had not been specified for high electrical resistance. The study then examined proposed changes with numerical models. Its result concerns those specimens and models, but exposes an important distinction: a component chosen to cushion mechanical load may still provide an electrical path. Mechanical resilience and electrical isolation require separate evidence.

Sources: sc-zagreb

Collection systems give leakage another destination

Andrea Mariscotti’s 2020 technical study describes an additional protection layer: dedicated rebars or grids beneath the rails intercept leakage and connect to a collector returning current toward the traction power station. These collecting elements are arranged separately from the structural reinforcement they help protect. Their effectiveness depends on the relative resistance of competing paths, and some current can escape the collection system too. The engineering objective is to capture leakage before it spreads through vulnerable structures, while preserving the primary insulation that limits leakage at its source.

Sources: sc-mariscotti

A reading changes with both traffic and the surroundings

Stray current is dynamic because trains accelerate, brake and occupy different locations. The 2022 paper “Dynamic Stray Current Measuring Methods in Urban Areas” also identifies soil and track moisture as influential. A single rail-to-ground voltage reading therefore cannot describe every operating and weather condition. The authors compare rail potential, rail current, collection-system measurements and other sensing approaches. These observe different aspects of the circuit: voltage describes an electrical driving condition, while current measurements describe flow. Their combination helps distinguish a changing traffic load from deteriorating insulation.

Sources: sc-measurement