Neutral grounding of industrial power systems has always been a controversial topic. Historically, systems with ungrounded neutral were dominant because of the service continuity with a ground fault on the system. This resulted in high system availability because there was no need to trip after the first ground-fault inception. However, as industrial power systems became more complex, transient overvoltage during a ground fault became more severe, making ungrounded-neutral systems less attractive. On the contrary, the ability of grounded-neutral systems to limit overvoltages made them more popular. Over time, ungrounded systems in North America started to disappear, except legacy systems, and almost all new industrial systems are designed with grounded neutral. With myriad grounding methods, the question is which is the most appropriate method to use? Each method has its pros and cons, making the choice of the appropriate one dependent on the application. For each grounding method, this article presents a brief description of selection criteria used for evaluation and their pros and cons.
Challenges with Ungrounded Systems
In the early days of industrial power systems, the convention was to use ungrounded systems. Back then, systems were small enough that ground faults, which constituted the majority of faults, were self-extinguishing . Hence, service continuity was guaranteed, and, therefore, ungrounded systems were more favorable. With the continuous growth in the size and complexity of electrical systems, substantial steady-state and/or transient overvoltages started developing on unfaulted phases, causing neutral-voltage instability.
During a line-to-ground fault, the voltage stress imposed on the insulation of the healthy phases can lead to an insulation failure, escalating the fault to a double line-to-ground fault. The fault escalation is attributed to the neutral-voltage instability because of the significant increase in the neutral voltage to ground (during normal operation, neutral voltage to ground is zero). Fault escalation was constantly observed on ungrounded industrial power systems . Therefore, the majority of new industrial power systems are grounded systems, where an intentional connection of the system’s neutral to ground is made.
Two methods can be employed to connect the system’s neutral to ground: a direct connection, known as solidly grounding, or a connection via impedance, known as impedance grounding. Impedance grounding is further divided into three subcategories based on the nature of the device used to connect the system’s neutral to ground resistance grounding, reactance grounding, and ground-fault neutralizer (also known as tuned-reactance grounding or Peterson coil grounding). Resistance and reactance grounding can either be low or high according to the permitted magnitude of the ground-fault current.
As neutral-grounded systems became the standard practice in North America, the challenge for system designers was to choose the most appropriate grounding method for the application. In other words, the question became should the system’s neutral be connected directly to ground with no deliberate impedance (i.e., solidly grounded), or should a grounding device (i.e., an impedance) be used to connect the system’s neutral to ground? If impedance grounding is selected, then what type (i.e., low- or high-resistance grounding, low- or high-reactance grounding, or ground-fault neutralizer) and what value should be used?
This article tries to answer those questions to help engineers understand the performance differences between grounding methods and then select the most appropriate method for the application at hand.