Measurements and calculations of rail potential in D. C. traction power supply system

Abstract

In D.C. traction power supply system, high resistance grounding fault is one of the difficult problems for reliable power supply. Because the load current of trains is much higher than the grounding fault current, it is impossible to detect the fault from the over-current relay. The detection of three phase unbalance which is conventionally applied in A.C. power supply system is not applicable in D.C. traction power supply system, neither. Considering such a peculiarity of D.C. traction power supply, grounding fault detection based on the potential difference between rails and grounding mesh at traction substation is practically used for grounding fault detection at D.C. substation for railway in Japan. The rails of D.C. railway are electrically floating, not grounded intentionally, to avoid stray current caused by D.C. traction loads. Also, the rails are part of a D.C. traction power supply system, and can be an approach path for lightning to the substation. Studying the characteristics of the rail potential and the leakage current to the ground is also important for knowing the tendency of electricity when lightning current flows. In this study, the details of the behaviors of rail potential in urban D.C. railway system are investigated not only by the measurements in practical fields, but also by the calculations of D.C. traction power supply circuit. According to the previous experiences during a few decades, the rail potential near traction substation tends to be negative while it is positive at the location of train because of the voltage drop caused by the current in rails. Our measurement of rail potentials in Tokyo urban area, however, indicated that the rail potentials became positive in many cases. This result contradicts to the common sense of railway power supply engineers. We carried out circuit simulations for many cases and it was found that the rail potential tends to be positive in the traction power supply circuit with many trains. As conclusions, the rail potential becomes positive in the area where train operation density is high while it becomes negative where that is low. In addition, it was confirmed that two factors influence the fluctuation of the rail potential by continuous measurement of the rail potential. First, as the output current of the substation increases, the current leaking from the feeder circuit to the ground increases and the rail potential becomes negative. Second, the rail leakage resistance decreases due to rain, the leakage current increases and the rail potential becomes negative. In this paper, the details of our measurements and calculations will be shown.