Welcome to our website.
To view product availability in a specific country, select from the country list below. For Investor Relations, please visit our global site.
Grounding and shielding electrical systems are of key importance to electrical engineers. Understanding the basic operations between grounded and ungrounded electrical systems is necessary for matching the appropriate grounding topology to the desired electrical system performance.
Selecting the proper grounding topology for an electrical distribution system is important to ensure facility occupant safety and health as well as reliable and safe electrical equipment operation. According to NFPA 70: National Electrical Code (NEC), Article 250.4(A)(1), the purpose of electrical system grounding is, “To limit the voltage imposed by lightning, line surges, or unintentional contact with higher-voltage lines that will stabilize the voltage to earth during normal operation.” The focus of Article 250 is to describe the grounding topologies available among grounded and ungrounded systems and how they operate.
The purpose of grounding the electrical system as stated in NFPA 70: National Electrical Code (NEC) is, “To limit the voltage imposed by lightning, line surges, or unintentional contact with higher-voltage lines that will stabilize the voltage to earth during normal operation.” To achieve these goals, the NEC provides the framework for the selection of grounding methodologies in Article 250. The focus of this article is to describe the grounding topologies available among grounded and ungrounded systems and how they operate.
The importance of providing a solidly grounded circuit for safety was recognized in the early editions of the NEC. According to “IAEA Soares Book on Grounding,” 100 years ago, the 1913 NEC committee required that "transformer secondaries of distributing systems must be grounded, provided the maximum difference of potential between the grounded point and any other point in the circuit does not exceed 150 V and may be grounded when the maximum difference of potential between the grounded point and any other point in the circuit exceeds 150 V." The code committee recognized that when a fault occurs on a grounded circuit, the grounded conductor maintains the system voltage at a stable source voltage rather than floating up to a higher potential. This protects individuals from being exposed to a potentially lethal shock were they to touch a faulted line, equipment, or chassis.
Solidly grounded systems
Today, because grounded systems offer greater voltage stability, most of the systems described in Article 250.20 of the NEC require a grounded system, whether it is a solidly grounded system or an impedance grounded system. Historically, the most commonly used system is the solidly grounded system (see Figure 1).
The NEC allows up to 25 ohms of ground resistance, recognizing different soil resistivities found across the U.S. However, the lower the ground resistance (or higher the ground conductivity), the better the ground fault detection system will operate. Typically, 5 ohms is a good design basis for commercial buildings. Lower ground impedance may be required for some medical imaging equipment. In a solidly grounded system, the ground fault system performs better with smaller ground electrode resistance. Article 250.2 of the NEC states that an effective ground fault current path consists of “an intentionally constructed, low impedance, electrically conductive path designed and intended to carry current under ground fault conditions.” Therefore in a solidly grounded system, it is the design intent to provide an earth reference to open a circuit as quickly as possible to isolate the fault based on high current flow. This prevents the fault from escalating and also protects connected motors and equipment from damage (see Figure 2).
There are several types of faults that an electrical system must be designed to withstand. The worst case but less common fault is a 3-phase bolted fault with little or no circuit impedance in the fault path. Equipment is typically sized and noted with a fault current rating based on fault calculations for these situations. With little impedance in a grounded circuit, high fault current levels are possible and arc flash hazards may be present in a solidly grounded system. The high fault current levels are considered one of the main downsides of a solidly grounded system. For example, in a 3-phase line-to-ground fault, voltage remains constant and because the impedance of the system is intentionally minimized, a direct result from the application of Ohm’s Law predicts high fault current flow. A benefit is high fault current will cause the upstream overcurrent protective devices to sense and operate quickly to isolate the faults as they return to the source within pathways designed to have the least resistance. It is up to the designer to provide an adequate pathway to guide the fault properly back to the source with strategies such as compression couplings on raceways, bonding to steel and periodic testing of the ground electrode system.
Because of the importance of this current flow being high enough to trip overcurrent devices, the NEC requires that the neutral-to-ground bond be made within the service entrance equipment. This is essential for the ground fault detection scheme to operate correctly. If the ground is made outside the equipment, the reactance of the circuit will increase. The total impedance of the circuit is expressed as (R+Xj), where Xj is the system reactance. When the total impedance of the system is too high, the overcurrent protective device may not operate as desired. Grounding at a single location at the source also provides benefits for the overall electrical system by preventing circulating currents.
Although a designer must account for the worst-case scenario, the 3-phase fault is quite rare. In fact, line-to-ground faults account for 90% to 95% of all recorded fault events in industrial settings. These faults can manifest themselves as arcing faults, which can cause current flow at a lower level than the overcurrent device rating. This is considered a serious drawback of the solidly grounded system because these faults may go undetected until equipment damage is done. The design remedy is to introduce ground fault detection into the circuit. During the 1970s, the NEC recognized this issue and added language to require that feeders rated 1,000 A or more on solidly grounded 480 Y/277 V wye-connected systems be equipped with ground fault detection. Ground fault detection can get complicated, especially if multiple levels are used within a system. Similar to circuit breaker coordination, it is necessary to coordinate the time-current curves for ground fault overcurrent protection to prevent upstream breakers from tripping prior to the GFI breaker closest to the fault. Otherwise, more systems than desired will be brought offline.
Modern low-voltage transformers are primarily designed and constructed with delta primaries and wye secondaries. In most commercial and industrial applications, the standardized voltage is 480 Y/277 V on the secondary side. Early versions of the NEC
didn't require systems to be grounded on the secondary side for voltages higher than 150 V. Grounding the secondaries of these service transformers for safety and to minimize equipment risk didn't gain momentum until the mid-1930s. A cost-effective solution was to ground a corner of the delta secondary. Therefore, many historic structures still have operating delta-delta service transformers where one corner of the transformer has been grounded to provide 120 V/240 V power within the facility.
The primary goal for a solidly grounded system is to open the circuit as quickly as possible to limit damage and risk to life. For large process and industrial plants, stopping the process can be equally hazardous. Prior to the mid-1930s, the concept of an ungrounded system was still in favor because of the service continuity benefits that the ungrounded system provided. A fault on an ungrounded system doesn't cause the source circuit breaker to trip. In fact, the system will keep operating until the operator tracks down the fault or until a second fault causes a major component in the electrical system to fault to ground, during which large magnitudes of current flow (see Figure 3). While theoretically this system is ungrounded, in reality the three phases are capacitively coupled to ground (see Figure 4).
Rather than a true ground, it is the system capacitance that helps to stabilize the voltage during normal operating conditions. However, during a fault—typically from line to ground (via the system capacitance)—there is no direct ground connection, and there is no high current flow that would otherwise trip the circuit breaker to isolate the fault.
Instead, it causes the phase voltage to rise 1.73 times the voltage on the other phases without tripping the breaker (from “Ground Fault Protection on Ungrounded and High Resistance Grounded Systems,” Post Glover). If cable systems and motor systems were not specified to withstand these higher voltage levels, the electrical systems would be subjected to undesirable stresses that would take their toll over time. Moreover, if an intermittent fault occurs, such as an arc fault, which can strike and restrike, overvoltage of up to 6 times greater than typical line voltage can occur, which can severely damage cable insulation and sensitive equipment. As equipment ages, it becomes more vulnerable to these strikes until, ultimately, it fails and faults to ground through equipment cases—or worse—through a person. Because circuit breakers don't trip, faults in an ungrounded system are difficult to trace and often go undetected until major equipment damage occurs during a second fault. Because of these issues, some industrial plants in the 1930s began converting their electrical infrastructures to grounded systems.
Ungrounded, resistance grounded systems
Although the NEC requires the majority of electrical systems to be grounded, some are actually required to be ungrounded. There are only five different electrical power systems/subsystems noted in NEC Article 250.22 where the code committee has determined the hazards of grounding to outweigh safety benefits of grounding. One of these system types is an isolated power system, which is a distribution power system of limited size, typically for use in hospital operating rooms. These areas are required to have an ungrounded system because it would be considered unacceptable to have a power outage during a surgical procedure. A typical isolated power system consists of a single-phase 10 kVA isolation transformer in which the secondary side remains ungrounded. The transformer’s electrostatic shield is connected to ground and effectively shunts high-frequency noise to ground. The 120 V equipment connected to these systems will continue to operate after the first fault, just as in an ungrounded system. These power systems are particularly suitable for use in operating rooms where there may be water or fluids present and where a GFCI receptacle (required by the NEC in wet areas) would ordinarily be required to be installed. The installation of the isolated power panel is alarmed locally, so if there is a ground fault, the team will be notified, but any ongoing procedures needn't be interrupted.
During the 1970s, language was added to the NEC to require ground fault trip sensors to feeders 1,000 A and above on 480 V grounded electrical systems. The need for electrical service continuity for the industrial process sector drove the need for a hybrid system to combine the stability and safety benefits of the grounded system with the continuous service benefits of the ungrounded system. During this time, resistance grounded systems began gaining traction. Service continuity makes this type of grounding system very attractive today for the traditional pulp and paper industry as well as for high-tech data centers. An impedance grounded system incorporates the benefits of both the grounded and the ungrounded system. The IEEE Green Book identifies the following benefits:
Impedance grounded systems include high resistance ground (HRG) and low resistance ground (LRG) configurations. For a wye-connected transformer, Figure 5 demonstrates how a known resistance is matched to the facility load profile and inserted directly between the secondary of the service transformer and ground. To accomplish this with a delta secondary transformer, an artificial neutral must be created using a zigzag transformer.
In a wye connected HRG system, intermittent faults that cause so much trouble in ungrounded systems will be eliminated by the neutral system ground resistor because its insertion limits the total current flow to ground.
System continuity is maintained because, although ground fault alarms occur, the overcurrent devices do not operate. This current flow in a low-voltage system (480 V to 600V) will be limited typically to 10 A so that the fault can be located and then repaired at a scheduled time without exposing staff to hazardous fault levels (see Figure 6). While HRG systems are a good fit for large data centers, there are pitfalls, such as misapplication of surge protective devices (they must be rated for ungrounded-neutral circuits), and the UPS must be grounded in a compatible method to its input and output wiring. Tracing faults is somewhat difficult and must be accomplished on live circuits using circuit pulsers.
LRG-grounded systems are typically used for 15 kV medium-voltage applications where the charging current may be too high to match an HRG. LRG systems tend to operate more similarly to the solidly grounded system than the ungrounded
system. In this case, the added resistor limits the fault currents between 200 A and 400 A, which is too high to allow continuous operation during a fault. Therefore, ground fault detection equipment must be set to trip as quickly as possible on detection. The advantage of controlling the current is that improved selectivity between overcurrent protective devices in the system may be achieved. It is interesting to note that through the 1999 code cycle, impedance/resistance grounded systems were in the same article as the ungrounded systems because of their operating similarities.
The NEC provides the framework for applying grounded and ungrounded systems. Table 1 summarizes the benefits and drawbacks of these different grounding systems as organized by the NEC. In a facility with a predominant need for line to ground loads, the NEC clearly requires a solidly grounded system. The solidly grounded system is the simplest and the cheapest to implement in the field. It is typically found in commercial buildings of today. In contrast, if a facility only has 3-phase loads and terminating its internal processes is deemed to be too heavy a risk, then an ungrounded system has definite merits. There is a middle ground, however, where service continuity is required, and the benefits of isolating and locating a fault for added safety are required. In these situations, one might consider an HRG system that has a proven track record for use in industrial process plants as well as large data center designs. The HRG system provides a single-point ground system for the facility. However, if and when there is a ground fault, the fault won't cause downtime.
NEC Article 250 has remained largely unchanged over the years, with a few punctuated changes in the 1940s and 1970s. Much credit must be given to the original code committee members for understanding the fundamentals and safety benefits of system grounding. Although grounding is often viewed as being mysterious, adhering to the code will safeguard occupants and facility equipment.
Elizabeth Sharpe, PE, Affiliated Engineers Inc., Seattle
Sharpe is a senior electrical engineer at Affiliated Engineers Inc. She has more than 20 years of design experience in higher education, research facilities, and mission critical projects. Her most recent projects have been for the Fred Hutchinson Cancer Research Center and the University of Washington, School of Medicine Research Facility, both in the South Lake Union neighborhood of Seattle.
Powered by ContentStream®