Abridgment of effect of transient voltage on power transformer design: The behavior of transformers with neutral isolated or grounded through an impedance

The results of theoretical and oscillographic study of transient voltage phenomena in shell type and core type transformers with neutral isolated from ground, and also with neutral grounded through impedances of different characteristics, are briefly described. In the case of isolated neutral, the majority of lightning traveling waves cause the potential of the entire winding, including neutral, to rise above ground to a value at least equal to the applied voltage. In the case of switching surges, the voltage of the neutral as well as of the rest of the winding is apt to rise to a value considerably in excess of the applied voltage. In the case of transformers excited with damped oscillations similar to switching surges, oscillographic records of both shell and core type transformers indicate internal voltages to ground approximately four times the applied. Grounding of a transformer through resistance, inductance, or capacitance does not reduce the above internal transient voltages, unless a certain relation is established between these constants, the constants of the transformer circuit, and the wave shape of the line surge. If the proper relation of these constants is established, then internal transient voltages are reduced practically to those of a solidly grounded transformer. A grounding device called an “impedor,” can be designed to have impedance at operating frequency equal to any desired value and to act at transient voltage frequencies as if its impedance were practically zero. This device can be designed for each transformer bank or for a number of banks in parallel. If grounded through a suitable impedor a non-resonating transformer retains its uniform transient voltage distribution. During the last year and a half, nearly 800,000 kv-a. of transformers of non-resonating type, have been built so that they can be operated with impedor in the neutral.