Achieving effective selective coordination design
The concept of full selective coordination has changed the way engineers must think when designing electrical distribution systems.
Dan Caron and Ron Schroeder
Selective coordination requires integrating different components, technologies, manufacturers, and standards. There is no standard, cookie-cutter approach that can be applied effectively across system designs. Although selective coordination is about to enter its third National Electrical Code (NEC) cycle as a mandated requirement—not left to engineering judgment since the 2005 cycle—issues continue to circulate about the necessity of mandating it, what constitutes compliance, and how other aspects of power system design might be compromised (see “Selective coordination issues”).
The concept of full selective coordination has changed the way engineers must think when designing electrical distribution systems (see Figure 1). For example, when selectively coordinating emergency and legally required standby power systems, overcurrent protective device specifications must accommodate a range of demands. All overcurrent protective devices must be fully selective with all upstream devices for all levels of overcurrent from all sources. Each overcurrent protective device must remain closed long enough for every device below it to clear for all levels of overcurrent, which include:
- Soft, low-current sources
- Stiff, high-current sources
- Low-impedance (bolted) faults
- High-impedance (arcing) faults
Source: Consulting-Specifying Engineer webcast, Feb. 2008
Defining selective coordination
The goal of selective coordination is to isolate a faulted circuit while maintaining power to the rest of the electrical distribution system. Although selective coordination will not prevent problems from occurring, it will help retain system reliability by decreasing the potential for a smaller scale problem to become a larger scale problem. Depending on the location, a fault could still cause a large-scale outage.
According to NEC Article 100, selective coordination is the localization of an overcurrent condition to restrict outages to the affected circuit or equipment. It is accomplished by the choice of overcurrent protective devices and their ratings or settings.
The overcurrent condition may be due to an overload, short circuit, or ground fault. In a selectively coordinated system, only the overcurrent protective device protecting that circuit in which a fault occurs opens. Upstream overcurrent protective devices will remain closed. In other words, they do not open, which averts cutting power to the complete panel.
Low-voltage circuit breakers: When selecting circuit breakers as overcurrent protective devices, tables can help determine proper upstream and downstream circuit breakers. Each manufacturer provides tables only for the overcurrent protective devices it produces (see the online version of this article for an example of a manufacturer’s coordination table). Tables and time-current curves should be used in tandem to meet selective-coordination requirements.
Emergency, legally required, and critical operations power: Emergency, legally required, and critical operations power systems require selective coordination, except when selectively coordinating a system could create safety hazards such as disconnecting fire pumps.
Selective coordination involves trade-offs between personnel safety due to the threat of arc flash, and maintaining power to critical systems while preventing damage to electrical wiring and equipment.
NEC requirements: Selective coordination is mandatory for emergency electrical systems for healthcare facilities, emergency systems, legally required standby systems and critical operations power systems. NEC requirements help ensure electrical circuit and system designs that provide reliable power for life safety and critical loads to help protect life, public safety, national security, and business continuity.
Fault types: Types of faults include bolted, arcing, and ground. Bolted faults are rare. A bolted fault occurs when energized conductors are rigidly connected. The maximum available fault current flows until the overcurrent protective device clears the fault, which protects the circuit.
Arcing faults occur when energized conductors come into proximity. While bolted faults and arcing faults are both short circuits, an arcing fault has significantly higher impedance than a bolted fault, resulting in lower current flow. Because the current flowing through an arcing fault is lower than current flowing through a bolted fault, the overcurrent protective device takes a longer amount of time to clear the fault condition. This is why arcing faults can present significant challenges to selective coordination.
According to IEEE, the most common type of fault is a ground fault. A ground fault occurs when one or more electrical phase conductors come in contact with a grounded conductor, as opposed to a phase-to-phase fault. However, the same principles apply: the lower the impedance, the quicker the overcurrent protective device clears the fault. Bolted faults are comparatively simple to selectively coordinate; arcing faults, not so much.
Protecting both personnel and equipment is vital. Every facility needs protection. Often, selective coordination is only one element in the overall protection scheme.
Effective selective coordination can be determined by two types of studies: a short-circuit current study and an overcurrent coordination study. A short-circuit current study identifies the maximum available short-circuit currents throughout the distribution system at the line-side terminals of each overcurrent protective device. This type of study is typically considered to be part of a facility’s required electrical documentation.
An overcurrent coordination study compares the timing characteristics of various protective devices under consideration in relation to each other. These studies determine the degree of coordination, but only guarantee that selective coordination is achieved if all levels of overcurrent are considered—including bolted line-to-line faults.
Mandated selective coordination demands selective coordination for the full range of overcurrents. It is the responsibility of the design engineer to provide substantiated documentation showing that the design achieves this goal.
The two methods of achieving selective coordination among overcurrent protective devices are the graphical and the table or chart methods. Time-current curves indicate the response of overcurrent protective devices to a range of fault current magnitudes. Typically, time-current curves can be divided into overload and instantaneous or short-circuit regions (see Figure 2). The graphical method examines curves for fuses and circuit breakers. The horizontal axis represents current, while the vertical axis shows the time it takes for the device to interrupt the circuit.
Using the graph method, two circuit breakers crossing at any point in their respective instantaneous trip regions indicates that those two circuit breakers do not coordinate for fault currents above the crossover point.
For current levels in the overload region, time-current curves for overcurrent protective devices can be overlaid for a visual indication of whether selective coordination is achievable. In the overload region, fault currents are relatively low, and device response time is usually not much faster than 1 sec. In this region, selective coordination can be relatively easy to accomplish and the time-current curve is typically an adequate tool for determining selective coordination between devices. The curve must include the level of available short-circuit current. However, the fact of no overlap on the graph does not definitively prove selective coordination.
At higher short-circuit current levels, the time-current curves alone do not show the total picture. The results do not include the effect of the added impedance of the downstream circuit breaker if it begins to open faster than the upstream circuit breaker, as well as the resulting higher coordination levels.
If curves overlap, the consulting engineer should reference the manufacturers’ circuit breaker tables to determine if selective coordination is achieved. The tables show results of tests of overcurrent protective devices connected together.
Overlapping curves can indicate a potential lack of selectivity. Conversely, a lack of overlap indicates selectivity. However, time-current curve analysis alone ignores how current limitation affects the load-side overcurrent protective device. The load-side circuit breaker will react to the peak let-through current allowed to flow by the smaller—or faster—overcurrent protective device for a given prospective fault current.
The true time-current curve for overcurrent protective devices, such as circuit breakers and fuses, is really a band or region extending to either side of a single line. This variation from the ideal is due to the time difference between minimum response time and total clearing time as well as manufacturing and temperature variations. Consideration of all the time-current curve variations is required to eliminate possible errors when examining selective coordination.
A table/chart-based method can also be used to determine coordination. It uses a matrix that shows response time in sec versus current in Amps. This method shows the level of short-circuit current to which the two breakers (upstream and downstream) coordinate.
Both time-current curves and tables are necessary to achieve proper selective coordination.
Imaginary systems, worst-case scenarios
The problem is that design engineers need to conduct a preliminary study based on an imaginary system including worst-case scenarios to ensure the design will be acceptable—before manufacturers and their products are chosen. Engineers typically base the design on standard, generic equivalents. After the contractor chooses the material, the engineer requires manufacturers to conduct their own studies with the selected breakers to ensure they still coordinate and that the correct breaker is provided.
In many cases, achieving coordination with breakers requires specifying breakers with electronic trip, which are more expensive than standard molded case breakers (see Figure 3). Selective coordination can also be achieved with zone-selective-interlocking (ZSI) protection. This method allows two or more ground fault breakers to communicate over a network so a short circuit or fault clears by the breaker closest to the fault in the shortest time possible, regardless of the location of the fault.
If there is a fault on the main bus, there will be more current coming in on the main and less current going out on the feeders. ZSI protection would open the main, which would protect against a bus fault in the substation.
While the design engineer may select an overcurrent protective device that may seem well suited for satisfying the short-circuit study requirements, it may not be the best choice for selective coordination. If the system has been expanded or upgraded over time, it may include both circuit breakers and fuses—possibly from different manufacturers. In these cases, ensuring selective coordination becomes more problematic because a manufacturer’s tables provide data only for its products. Effective selective coordination during a system’s lifecycle could require using the same type of overcurrent protective devices from the same manufacturer over time.
Optimizing selective coordination is an iterative process. Depending on the system’s complexity, the analysis may suggest that device selection indicates an imbalance or tilted trade-offs among selective coordination, equipment protection, and personnel safety (see “General selective coordination guidelines”).
Selectively coordinating power transfer switches
Power transfer switches are essential in emergency, legally required standby and optional standby power systems. Consequently, they are essential to selective coordination systems as well. Optimized selective coordination systems that incorporate power transfer switches achieve fast fault-clearing times and coordination of overcurrent protection at reasonable cost. Transfer switch design affects cost, reliability, maintenance, and personnel safety throughout the system’s lifecycle.
Transfer switches must withstand and close on fault currents until the downstream protective device clears the fault. Its ability to accomplish this is typically rated in terms of the device’s withstand and close-on rating (WCR). The WCRis the highest level of current that can be carried by a given transfer switch for a specific amount of time. This time must be long enough for the upstream overcurrent protective device to clear the fault. Transfer switches with integrated overcurrent protection typically selectively coordinate with other devices.
Whether selected to satisfy requirements of NEC Articles 517, 700, 701, or 708, transfer switch ratings for a specific fault current must be greater than, or equal to, the available fault current and system voltage as determined at the power source terminals of the switch. The WCR can be based on either a specific device rating, or an optional “any-circuit-breaker” rating. A switch may have both types of ratings, which provides greater flexibility in the selection of the overcurrent protective devices.
A transfer switch also may have an optional short-time rating. This rating is intended for use with an upstream circuit breaker having a short time rating of “X” cycles (sec). The overcurrent device’s clearing time is typically provided by trip curves indicated in sec, but is frequently translated to ac cycles in a 60-Hz system. In other words, a transfer switch should be able to withstand a short-circuit current of 65 kA for 0.5 sec or 30 cycles, for example.
Typically, power transfer switches facilitate the selection of fuse-clearing times or breaker settings using increments between 0.5 to 30 cycles, with multiples of two or three cycles being popular. There are no ideal time-delay settings for selective coordination in design schemes. Also, UL doesn’t require a specific time or specific number of cycles to qualify for short time ratings, although it does provide standard recommended values.
The transfer switch location is significant in terms of effective selective coordination. Transfer switches located closer to their loads translate into:
- Higher reliability
- Smaller circuit breaker or fuse sizes for feeding the transfer switch
- Fewer levels of distribution
- Lower fault currents at the transfer switch terminals
- Faster fault-clearing times
- Improved load protection.
Locating a switch closer to the source (as opposed to the load) can lower system or facility reliability, may cause downstream breakers to trip frequently, and may not isolate—or start—an alternate power source. Larger circuit breaker or fuse sizes may be needed. Higher fault currents may be experienced and short time protection may be required. Bottom line: locating a switch closer to the power source typically means relatively poor load protection.
Specifying cycle times
Consulting engineers should recognize that specifying transfer-switch cycle times is necessarily project-specific. Most transfer switches specified by the co-author’s company are 3-cycle switches. The firm has specified 30-cycle switches where at least one of the following conditions exists:
- Larger projects with high emergency-system fault current
- Where the instantaneous trip setting is defeated to achieve selective coordination (note that this may change because of the acceptance of the latest NFPA 99)
- Where transfer switches are served by ANSI switchgear, which also has 30-cycle withstand ratings.
Every project should be considered to be custom. For example, even if two chain restaurants are built in two different locations using the same drawings, the coordination would be different. Effective coordination depends on the electrical utility and the available fault current. Inevitably, electrical characteristics and situations vary from place to place, building to building, design to design, and utility company to utility company.
It might seem easy to specify 30-cycle transfer switches as a cookie-cutter approach for both the ceiling and floor of selective coordination timing. However, that decision introduces safety, cost, and reliability issues. In some installations, personnel safety and equipment integrity may be compromised by letting energy levels flow for 30 cycles within those electrical systems. Using short-time-rated trip units in low-voltage circuit breaker settings may allow fault currents to flow for 30 cycles, perhaps negating equipment protection and increasing arc-flash hazards.
Specifying 30-cycle-rated transfer switches for every application can increase equipment and spare parts costs—sometimes as much as 15% to 30%. These switches may also require rear entry, a larger footprint, and more expensive maintenance.
Specifying 30-cycle-rated transfer switches may also necessitate installing larger feeder cables than normally required in order to safely carry the available short-circuit current for 30 cycles. This can cause a significant increase in the cost of the feeder cabling.
Of the engineers who participated in a survey following a webcast on selective coordination, 88% agreed selective coordination may not be optimal if 30-cycle transfer switches are used for an entire facility.
Considering the custom nature of selective coordination, the better decision is specifying equipment, components, and overcurrent protective devices that precisely satisfy the requirements of that unique design.
Achieving selective coordination can often be accomplished with transfer switches between 3 and 18 cycles. Matching time-based ratings to timing requirements of a given selective coordination design often better serves the design and the facility owner. If the design settings are at 3, 6, 9, 12, or 18 cycles, there is no reason to specify 30-cycle rated switches universally.
Developing a transfer switch schedule that includes fault-current levels and time requirements helps optimize equipment performance and cost.
Mandated selective coordination has changed the way design engineers think about designing electrical distribution systems. Engineers responsible for developing and vetting selective coordination systems that meet NEC requirements often face difficult challenges. Balancing the need for business continuity, equipment protection, personnel safety, and managing costs can be fraught with pitfalls. Knowing NEC requirements, observing design optimization methods, coordinating transfer switch ratings with circuit breakers and fuses, and satisfying special application needs are essential for effective selective coordination.
Caron is principal and the head of the electrical department at Bard, Rao + Athanas Consulting Engineers in Boston. He is also a principal member of NEC Code Making Panel 13.
Schroeder is director of applications engineering and product management for ASCO Power Technologies in Florham Park, N.J.
Case Study Database
Get more exposure for your case study by uploading it to the Plant Engineering case study database, where end-users can identify relevant solutions and explore what the experts are doing to effectively implement a variety of technology and productivity related projects.
These case studies provide examples of how knowledgeable solution providers have used technology, processes and people to create effective and successful implementations in real-world situations. Case studies can be completed by filling out a simple online form where you can outline the project title, abstract, and full story in 1500 words or less; upload photos, videos and a logo.
Click here to visit the Case Study Database and upload your case study.
2012 Salary Survey
In a year when manufacturing continued to lead the economic rebound, it makes sense that plant manager bonuses rebounded. Plant Engineering’s annual Salary Survey shows both wages and bonuses rose in 2012 after a retreat the year before.
Average salary across all job titles for plant floor management rose 3.5% to $95,446, and bonus compensation jumped to $15,162, a 4.2% increase from the 2010 level and double the 2011 total, which showed a sharp drop in bonus.