The Art of Protecting Electrical Systems, Part 7: Equipment Short Circuit Ratings
Editor’s Note: From 1965 through 1970, Consulting-Specifying Engineer ’s predecessor, Actual Specifying Engineer, ran a series of articles on overcurrent protection. Due to the immense popularity of the 31 installments in the series, the authors, George Farrell and Frank Valvoda, P.E., reprised the series in an updated version beginning in the Feb. 1989 issue of CSE. Over the years since the last installment ran in the late ’90s, we have received many requests to re- run this series. Mr. Valvoda passed away in Dec. 2001, but his long-time friend and editorial partner, George Farrell, is reviewing and updating this valuable series of articles, with the assistance of a new collaborator, Barry Feinberg, formerly EE faculty at Purdue University and now an independent consultant. Addenda describing new information that has been added will appear at the end of each article. It seems fitting that this year, 40 years after the series first appeared, we are re-launching it in an electronic format. “The Art of Protecting Electrical Systems” has made the transition from 20thto 21stcentury—and from print to the web.
Part Seven of this ongoing series discusses the importance of overcurrent protection theorems used in short-circuit calculations and component ratings.
Preceding articles explored a number of fundamentals involved in ensuring protection of electrical systems, emphasizing the need for short-circuit calculations and ratings.
This article presents general definitions and comments on electrical-component short-circuit ratings. It points out the need to determine the maximum short-circuit currents as accurately and completely as possible and introduces ways to calculate short-circuit currents.
Fused switch short-circuit (withstand) ratings indicate the rms symmetrical current they can withstand or close into without being damaged. If the rating is not indicated on the switch, it is 10,000 amps. Any rating greater than 10,000 amps must be shown on the label.
The most common ratings are 100,000 and 200,000 amps, although 25,000 and 50,000 are sometimes encountered. These high ratings are dependent on the use of the correct current-limiting fuses, and the fuse class or classes required for these ratings must be stated on the switch label or on the inside cover. Thus, a given switch may have a 100,000-amp short-circuit rating when Class RK5 fuses are installed and a 200,000-amp short-circuit rating when provided with Class RK1, Class J or Class T fuses. This is true even though all of these fuses have a 200,000-amp interrupting rating (A.I.R.).
Motor controllers have little short-circuit interrupting or withstand capabilities by themselves. Because of this, National Electrical Code (NEC) Article 430 details specific motor feeder and branch-circuit protection. While the short-circuit protection may be separated from the motor controller, a common arrangement is the combination starter. It contains a motor branch circuit disconnect device and short-circuit protection, i.e., a fused switch, circuit breaker or motor-circuit protector, a motor controller equipped with motor-running overload protection, and various control components such as control power transformers and push buttons.
Underwriters Laboratories, Inc. (UL) tests individual motor controllers rated 300 volts and less and one horsepower and less at 1,000 amps available short-circuit current. Those controllers that do not fall into the first category but are rated 50 horsepower and less are tested at 5,000 amps. Those rated 50 to 200 horsepower are tested at 10,000 amps. Larger controllers are tested at increasing values up to 85,000 amps for ratings above 900 horsepower. Complete ratings and test procedures may be found in UL Standard for Industrial Control Equipment-UL 508.
The short-circuit protection devices used for tests at 10,000 amps or less (up to 200 horsepower) are non-time delay, UL Class H fuses rated 10,000 A.I.R. These fuses are also known as "one-time," "NEC" or "Code" fuses. Fuse ampere ratings used for the tests are the maximum permitted by the NEC for the largest horsepower motor that may be used with the controller being tested. Since UL has no control over what short-circuit protection will be provided in the field, the 1,000, 5,000 and 10,000 A.I.R. (up to 200 horsepower) are the highest available short-circuit ratings for individual controllers.
Only combination motor controllers may be UL listed at higher short-circuit ratings, unless the controllers are part of UL-listed equipment, which carries an overall short-circuit rating (motor control centers, machine control panels, air conditioning equipment, etc.).
In general, the highest short-circuit rating for combination controllers containing non-current-limiting protection, that is, thermal-magnetic circuit breakers or circuit protectors without limiters, is 5,000 amps in Sizes 0 through 3, 10,000 amps in Sizes 4 and 5, and 18,000 amps in Size 6. With proper current-limiting protection, short-circuit ratings up to 200,000 amps are available.
Transformers may be damaged by both the heating and magnetic effects of a short circuit. When provided with overcurrent protection as required by NEC Article 450, most standard-type liquid-filled and air-cooled transformers will not be seriously affected by the heating effects of a fault. However, these transformers are subject to mechanical damage from magnetic effects. Several faults may cause coils to deform, break solid insulation and break or bend structural components.
Transformer windings cast in epoxy are potted or have other solid insulation, similar to that found in strip-wound coils. They have great mechanical strength and may withstand the magnetic effects of a fault. However, conductors (often bare bus bar) used to interconnect windings of three-phase transformers and/or to connect the windings to transformer terminals are subject to the same mechanical stresses during faults and must be adequately braced to prevent severe damage.
Such transformers, which are part of a unit substation having overall short-circuit ratings, will have been tested and should be able to withstand the stresses of a fault. Many transformers, however, especially air-cooled types, are assembled from purchased coils, and the assembler may not understand the need for adequate bracing of internal buses unless such bracing and testing is specified carefully. Before specifying or installing transformers, short-circuit withstand ratings of the transformers should be obtained, and, if they are not adequate for the system, current-limiting protection must be provided.
One point of caution: Transformer short-circuit protection required by NEC Article 450 does not address the question of protection by circuit breakers without instantaneous trip elements. If short-delay trip elements are used to obtain selectivity, transformer manufacturers should be questioned regarding effects of sustained faults with this type of protection. If any doubts exist about the ability of the transformer to withstand fault currents, other means of obtaining selectivity should be considered.
Short-circuit current effects
Busway (busduct) and the buses in switchgear and similar equipment have withstand rating ranging from 10,000 amps for lighting and trolley busway to 200,000 amps for some switchgear. UL short-circuit ratings for all busway and equipment must withstand the thermal and magnetic effects of the fault.
The short-circuit rating of traditional busway, consisting of bus bars separated by an air gap and supported on insulated bracing, depends on the spacing between the bus braces. This is also true of the bus structures in most switchgear and motor control centers.
A trend in busway construction is toward the use of continuous, solid insulation between the bus bars-often termed "sandwich" construction. This has the advantage of high mechanical strength, resulting in higher short-circuit ratings.
If faults will be sustained for more than three cycles, traditional busways may be damaged by the mechanical stress. The sandwich-type construction may experience thermal damage, since it is less capable of dissipating heat. Maximum short-circuit currents must be known as well as the speed of the protective device. Only then can it be determined if equipment under consideration meets NEC requirements.
Wires and cables
Wires and cables have no specific National Electrical Manufacturers Assn. (NEMA), UL or NEC short-circuit ratings, but may be subject to thermal damage from a severe fault. As stated in NEC paragraph 110-3, equipment and materials must be evaluated for heating effects under normal conditions likely to arise in service. Under some conditions it may be necessary to greatly increase the conductor size to prevent severe damage under short-circuit conditions.
Many other types of equipment are affected by short-circuit currents. Some withstand ratings that are part of their UL listing; others do not. Since fault currents continue to increase and new types of equipment are constantly introduced, engineers, installers and users need to consider short-circuit withstand capability of all equipment.
The intelligent analysis and interpretation of system short-circuit studies will often disclose conditions that require more than the minimums established by the NEC. For example, the code sets these parameters:
NEC Paragraph 90-1 purpose
"(b) Adequacy. This Code contains provisions considered necessary for safety. Compliance therewith and proper maintenance will result in an installation essentially free from hazard, but not necessarily efficient, convenient, or adequate for good service or future expansion of electrical use.
"(c) Intention. This Code is not intended as a design specification or as an instruction manual for untrained persons." (Emphasis added by the authors.)
There is no accurate way to determine what short-circuit current will flow under field conditions. What can be calculated, however, is the current that would flow under certain standardized sets of conditions. These conditions have been adopted by the electrical industry over an extended period of time and are based on fundamental electrical theory as well as empirical data gleaned from observation, myriad testing and manufacturers' data.
In order to meet NEC Section 110-9, the maximum available short-circuit current is required. This usually is determined by calculations for a three-phase bolted fault study using positive-sequence impedances. To evaluate ground faults or other types of unbalanced faults for application of relays, etc., as well as to meet NEC requirements for ground-fault protection, negative-sequence and zero-sequence studies may be necessary.
For a simple commercial building with a radial system, it may only be necessary to determine that the available fault current at the service entrance is sufficiently less than the interrupting rating of the protective devices. This could be determined by a phone call to the utility, or a quick estimate could be made based on transformer kva rating.
On the other hand, large and complex systems can include 1,000 or more nodes (points where available short-circuit current and X/R ratio are determined.) When making a complete system analysis, a separate impedance diagram with consequent calculations must be set up for each point (or node or bus) in the system being studied.
When electrical systems serve large industrial and commercial installations, short-circuit calculations done by hand are susceptible to errors because of the sheer volume of calculations involved. Because of this, short-circuit calculation computer programs have been developed that enable the engineer to accurately study 100-, 300-, 500- and 1,000-bus systems. Before such programs were available, only utility companies could maintain a reasonably accurate model of such systems.
Industrial systems change rapidly and it is impossible to calculate them. The more users of these computer programs know about short-circuit studies, the more accurate will be their interpretation of the computer studies.
Study steps spelled out
The first step in any study is to determine the available fault current at the service entrance (or at on-site generator terminals if there are no utility connections.) When the information is not available from the utility, a quick estimate of the maximum symmetrical short-circuit available from the utility (disregarding contributions from the distribution system) may be made by using the following formula:
This is based on the way transformer impedance is determined. It provides the maximum current that will flow through a transformer with a given impedance when there is an unlimited source of energy. Transformer impedance may be obtained from the transformer nameplate or manufacturer's literature.
In almost all cases, this equation will overstate the available current. However, this method is useful to determine if available fault current is less than the minimum interrupting ratings of the protective devices being considered for use in the system. For example, if it is planned to use circuit breakers with 14,000-amp interrupting rating, and fault current at the service entrance determined by the above formula is 10,500 amps or less, no further calculations would be required. (Conservative practice recommendations are that no protective devices should be applied at more than 75% of their ratings unless detailed system studies indicate that it is safe to do so. Here, 10,500/14,000 = 75%. Among other factors, such detailed system studies necessitate an accurate determination of X/R ratios to determine whether asymmetrical currents exceed device capabilities.)
Connection study required
If the indicated fault current is 16,000 amps, a more detailed study of connections between transformer and circuit breakers will be necessary. If it is 40,000 amps, not only will a more detailed study be required, but also the 14,000-amp rated circuit breaker will certainly not be applicable, at least at the service entrance.
Short-circuit studies consist of creating the Thevenin Equivalent (see Part 2 of this series) of a circuit at the time of the short circuit and solving for the resultant current. In other words, the system is reduced to one impedance in series with one voltage source.
Part 3 of this series presented the concepts of short-circuit power factor and X/R ratio. A short-circuit study enables the calculation of the required X/R ratio at each faulted bus so that asymmetric multiplying factors may be applied to check for proper application of the equipment at each bus.
Mathematically, the Thevenin circuit is solved by the basic equation: I = E / Z, where: I = fault current, E = voltage impressed across the fault, Z = equivalent impedance.
The impedance is a combination of resistance, R, and reactance, X, where:
with the reactance being the arithmetical difference of inductive and capacitive reactances, XL %%MDASSML%% XC. The capacitive reactance in most 60-cycle systems may be neglected except when deliberately introduced, as in power factor correction.
In the expression I = E/Z, it is apparent that if Z equals zero, the current will be infinite. This is an impossibility, since every circuit contains some impedance. However, there are systems where the Thevenin impedance may be as low as 0.001 ohms, resulting in a fault current in excess of 200,000 amps on a 480-volt system. There are many low-voltage networks where the impedance is not over 0.002 ohms, equating a fault current in excess of 100,000 amps.
Even with close estimates for conductor and device impedances, it is not possible to arrive at an exact figure for impedance, because too many variables are present. If impedances of, for example, circuit breakers and fuses are ignored, the calculated values of fault current will be in excess of what will actually occur in a system. For this reason, it must be understood that a short-circuit current study is only an approximation.
The study of short-circuit calculations will begin in Part 8 next month. The first method will be the determination of maximum three-phase bolted fault current through use of the per-unit method. This will be followed by shortcut methods suitable for many simple radial systems. After these fundamentals have been completed, an in-depth study will be made of available computer programs for system studies. Finally, solutions for unbalanced faults will be discussed.