Short-Circuit Study

NOTE This standard governs the short-circuit study as an engineering deliverable: the engineer who performs it and when it is delivered, the calculation method, the system model and the data it is built from, the cases studied, the evaluation of every rating the calculated duty is compared against, the available fault current data for the service equipment, the report, and the model that the coordination and arc-flash studies take over from it. (1.1)

NOTE NFPA 70 110.9 requires every device that interrupts fault current to have an interrupting rating not less than the current available at its line terminals, and 110.10 requires the conductors, the overcurrent devices, and the equipment of every circuit to withstand the fault current the overcurrent device lets through without extensive damage. OSHA 29 CFR 1910.303(b)(4) and (b)(5) state the same two requirements as a condition of the workplace. Neither can be shown from a nameplate; both rest on a calculation of the current available at every point in the system, and this study is that calculation. (1.2)

NOTE The short-circuit study is the first of three studies performed on the same system model. It establishes the magnitude of the fault current at every bus and the X/R ratio behind it; the coordination study under Protective Coordination StudyProtective Device Coordination StudyResolves to the current adopted revision.sync/protective-coordination-study uses those currents to set the protective devices so that the nearest one clears; and the arc-flash study under Arc Flash StudyArc-Flash StudyResolves to the current adopted revision.sync/arc-flash-study turns the bolted fault current and the clearing time at each bus into incident energy. A change to this study's model moves both of the others, which is why the model and the results are delivered in a form the other two can take over without rebuilding. (1.3)

NOTE The following are governed elsewhere and are outside this standard: (1.4)

NOTE The study plan settles, before any data is entered, the questions that otherwise surface when the report is rejected: which buses are in the model and which are not, which cases are run, whose data the model is built from, and whether the study will be back before the switchgear order has to go. A plan returned with the boundary drawn and the schedule tied to the equipment release dates is what makes the study a gate the equipment order waits for rather than a document that arrives after it. (3.1.3)

NOTE A short-circuit model is rebuilt from scratch whenever the native file is lost, and rebuilt wrong whenever the one-line and the data tables in the PDF disagree with what was actually modeled. The native file is the study; the report is a printout of it. (3.2.3)

NOTE The Engineer of Record holds the design model and the utility correspondence and has no stake in the equipment, and performs the study where the fault currents are needed in design to size the equipment, with the result that the study is built on design data and is reconciled to the equipment furnished later; an independent engineer engaged by the Contractor is the arrangement where the study is a construction deliverable priced in the bid, built from the submittals of the equipment actually furnished, and reviewed by the Engineer of Record as a submittal; the equipment manufacturer's engineering services know their own equipment's ratings and series listings best and deliver the study with the switchgear order at the lowest added cost, and their model of the other makers' equipment, the conductors, and any existing system is only as good as the data handed to them; the testing organization's engineering group puts the study, the coordination settings, and the acceptance tests in one firm's hands, so that the party who sets the trip units built the model they were set from. No one arrangement is the norm across projects, and the choice turns on when the ratings are needed and who holds the data. (4.1.2)

NOTE A study delivered before equipment release confirms every interrupting, withstand, and short-circuit current rating while the rating can still be changed on paper, which is why the equipment standards hold the release of switchgear, switchboards, panelboards, transfer switches, and drives for it, and it is built on submittal data and design conductor lengths that the installation departs from; a study delivered after installation is built on the nameplates and the lengths that exist, so its results are the ones the arc-flash labels and the record carry, and a rating it finds short is corrected by replacing equipment that is already installed; a study delivered before release and updated after installation gives both, at the cost of a second pass through the model and a second review, and is the arrangement a project chooses where the system is large, where the conductor routing is uncertain at the time of the submittals, or where the coordination and arc-flash studies will be run on the as-built model. (4.2.2)

NOTE The equipment standards that name this study hold the release of equipment for it, and a Contractor who releases equipment ahead of the study carries the cost of any rating the study then finds short. (4.2.6)

NOTE Commercial analysis software holds the model as a file that the coordination and arc-flash studies open next and that a re-study opens years later, carries the device and conductor libraries the model draws on, and applies the duty multipliers of the ANSI/IEEE method without transcription, which is why it is the tool on nearly every study; a hand calculation by the per-unit or ohmic method of IEEE 551 is reproducible by any engineer with the report in hand and is the tool a project accepts where the system is a single service, one or two transformers, and a handful of buses, with the result that there is no model file to hand to the next study. (4.3.2)

NOTE The ANSI/IEEE method calculates the duty on each device in the terms its rating standard tests it in, with the first-cycle, interrupting, and time-delayed networks and the X/R multipliers of IEEE 551, IEEE C37.010, and IEEE C37.13, and it is the method for every system whose equipment is rated under IEEE and UL standards, which is every system this standard is written for in its home market; the IEC 60909 method calculates the initial symmetrical current, the peak current, the breaking current, and the steady-state current with the voltage factor c and the impedance correction factors of IEC 60909-0, and is the method for a system whose equipment is rated under IEC standards or whose owner's engineering practice is written to IEC. The two methods give different numbers for the same system and a rating tested under one is not evaluated against a current calculated under the other. (5.1.2)

NOTE The three networks are one network with three sets of rotating-machine impedances. The first-cycle network, at one half cycle, carries every motor and generator at subtransient reactance with the IEEE 551 multipliers for motor size, and gives the current that a fuse, a molded-case breaker, a low-voltage power breaker, a bus brace, and the close-and-latch capability of a medium-voltage breaker are evaluated against; the interrupting network, at the contact-parting time of the breaker, lets the induction motor contribution decay by the IEEE 551 multipliers and gives the current a medium-voltage breaker interrupts; the time-delayed network, at about thirty cycles, drops the induction motors and places the generators at transient reactance, and gives the current a time-delayed relay sees and the current the minimum case is built from. On a system with no equipment above 1000 V the interrupting network adds nothing the first-cycle network does not already decide, and a project that leaves it out is leaving out a calculation the software performs anyway. (5.2.2)

NOTE The three-phase bolted fault is the highest current at nearly every point of a system and is the case every interrupting and withstand rating is evaluated against; the single line-to-ground fault on a solidly grounded system can exceed the three-phase fault close to a wye-connected transformer or generator and is the current the ground-fault protection and the equipment grounding conductors are sized against; the line-to-line fault is about 87 percent of the three-phase value and is the lowest phase fault the coordination study uses as its minimum at the end of a feeder; the double line-to-ground fault is calculated where the ground-fault current through a grounded neutral is needed on a system whose grounding method makes it the larger case. On a resistance-grounded or ungrounded system the ground-fault current is set by the grounding resistor or the system capacitance rather than by the fault calculation, and the study reports it from the grounding design. (5.3.2)

NOTE The maximum case decides every rating; the minimum case decides whether the protective devices see enough current to operate at the far end of a feeder, which is the coordination study's question, and sets the lower bound the arc-flash study applies its arcing-current reduction to. A study that reports the maximum only serves the equipment order and leaves the two studies that follow it to build their own minimum case from the same model, and a project accepts that where no coordination or arc-flash study will be run on the model. (5.4.2)

NOTE A pre-fault voltage of 1.0 per unit at the bus nominal voltage is the value equipment is tested at and the value nearly every study uses; 1.05 per unit raises every result by 5 percent and is used where the system is operated above nominal voltage on a regulating transformer or a utility whose delivered voltage sits at the top of its range; 1.1 per unit matches the maximum voltage factor of IEC 60909-0 for a low-voltage system and is used where a study is to be compared with one calculated under IEC. (5.5.2)

NOTE Every point at which a rating is confirmed is a calculation point, and the list is long because the ratings are many: the service equipment and every distribution assembly for their interrupting ratings and bus bracing, every transfer switch for its withstand and closing rating from each source, every busway run for its bracing at the source end and every plug-in unit for its rating at the opening, every controller, drive, control panel, and surge protective device for its marked short-circuit current rating, and every source terminal for the breaker that protects the source. A point of common coupling is added where a drive or a harmonic filter is evaluated under IEEE 519, because the ratio of the short-circuit current to the demand current at that point sets the harmonic limit. The points a project leaves out are the ones whose equipment has no rating to confirm. (6.1.2)

NOTE A written statement from the utility gives the three-phase and line-to-ground contribution and the X/R at the point of delivery for the system as it exists or as the utility plans it, is the basis on nearly every study, and takes a written request and a lead time the study plan has to allow for; a published standard value is what some utilities state for every service of a class in place of a site value, is already a planning maximum, and gives no X/R and no minimum, which the study then assumes; an infinite bus at the primary terminals of the service transformer takes the transformer impedance as the only limit on the current, gives the highest value that transformer can pass, is the basis where the utility provides nothing in time or where the service transformer is the project's and the utility's contribution behind it is large relative to it, and cannot be used where the utility delivers at the equipment voltage with no transformer between. (6.2.2)

NOTE Utility fault current rises over the life of a building as the utility reinforces its system, and equipment ordered today is rated for the contribution stated today. A margin of zero takes the utility's value as given, which is the practice where the utility states a planning or ultimate value rather than a present one; a margin of 10 to 25 percent buys headroom against a reinforcement the utility has not announced, at the cost of the next higher interrupting rating at the service and sometimes one level down; a margin of 50 percent approaches the infinite-bus result on a stiff system and is the figure a project reaches for where the utility has told it a substation upgrade is planned and has not quantified it. (6.2.6)

NOTE A transformer's impedance is the single value that most decides the fault current on its secondary, and a tolerance of 7.5 percent on the impedance is about 8 percent on the secondary fault current. A transformer whose test report shows a lower impedance than its design value is the common way a secondary switchboard ordered to a submittal-phase study ends up with no margin, which is why the tolerance is taken in the maximum case until the tested value is in hand. (6.3.5)

NOTE Published typical values, the tables of IEEE 141 and IEEE 241 or the conductor library of the analysis software built from them, are available at the time the study is run and differ from the furnished conductor by less than the uncertainty in its length, which is why they are the source on nearly every study; the manufacturer's published values for the conductor furnished are the source where the conductor is a cable type with an impedance the tables do not carry, such as a shielded medium-voltage cable, a large interlocked-armor cable, or a busway, or where the study is an as-built update and the submittals are in hand. Busway impedance is the manufacturer's value in every case, because no table generalizes across busway designs. (6.4.2)

NOTE Conductor impedance is what separates the fault current at a panelboard from the fault current at the switchboard that feeds it, and the length of the run matters more than the source of the impedance values. A short run or a run of parallel conductors in steel conduit modeled as if in aluminum conduit carries the fault current of the switchboard down to equipment rated for less, which is why the raceway material and the length are inputs the report has to show. (6.4.6)

NOTE Fifty horsepower is the line IEEE 551 draws between motors modeled one by one and motors grouped at their bus, and the line most studies follow without discussion; a threshold of zero models every motor individually, which a small system or a project with a complete motor list supports and which changes no result at the buses that decide a rating; a threshold above 50 horsepower is the arrangement of a large plant with hundreds of motors, where the individual data below the threshold is effort without effect at the medium-voltage buses and the motors are grouped at each motor control center instead. (6.5.2)

NOTE Lumping the small motors at each bus from their connected horsepower gives the contribution the real load produces and needs a motor list by bus; lumping them as a share of the transformer or bus rating, commonly half of the transformer kVA at a 480 V bus with mixed load, needs no list and is the practice where the load is not yet known or on an existing system with no load inventory; leaving them out of the model understates the first-cycle duty at the buses the small motors sit on and is accepted where the motor load is a negligible part of the bus load, as on a lighting and receptacle distribution. (6.5.4)

NOTE A motor stays a source for the first few cycles after a fault because its rotor keeps turning and its field does not collapse at once. Its contribution lands on the bus it is connected to and on every bus between it and the fault, and at a motor control center the motors can add a third again to the current the transformer supplies, which is why the motors are modeled at the bus they are connected to and not as a single lump at the service. (6.5.9)

NOTE An inverter is a current-limited source and adds a bounded amount to the fault current at the bus it feeds, usually between one and two times its rated current for a few cycles. On a bus fed by a large utility transformer the addition is small; on the load side of a UPS or on a microgrid bus islanded from the utility it is the whole fault current, and the equipment there is rated against a value that the inverter's current limit, not the system impedance, decides. (6.6.4)

NOTE Equipment on the load side of a transfer switch sees a different fault current from each source, and the rating has to cover the higher one; a generator plant alone usually gives less than the utility at the service and can give more at a bus near the paralleling switchgear; the utility and the generation paralleled, during a closed transition or in extended parallel operation, gives the sum of both at every bus between them and is the case that decides the paralleling bus rating; a double-ended substation or a pair of services with the tie closed doubles the source behind every bus on the common side and is a case whether or not the operating procedure forbids it, because the tie can be closed; a UPS in bypass passes the upstream source through to loads that in inverter operation see only the inverter's current limit. A case left out is a configuration the system can take up in which no rating has been checked. (6.7.2)

NOTE A survey of every existing item records what is installed rather than what was drawn, finds the breaker replaced with a lower-rated frame and the transformer retapped since the last study, and costs a survey visit for every item, under the facility's safety program; a survey limited to the items whose nameplate data the Owner's records lack keeps the visit short and trusts the records for the rest, which is the arrangement on a facility with a maintained equipment inventory or a recent prior study; the Owner's records and prior studies alone cost nothing on site and carry forward every change made without a drawing revision, which on an older facility is most of them. On new construction with no existing equipment the question does not arise. (6.8.2)

NOTE An existing system is routinely modified, re-fed, and re-tapped over its life without a revision to the drawings, and the fault current at an existing panel is decided by what feeds it today. The record drawings are the starting point of the survey, not a substitute for it. (6.8.6)

NOTE A short-circuit result is only as good as its weakest input, and the weakest inputs are the ones that were assumed. The register is what lets the Engineer of Record see which results rest on a typical value and which on a nameplate, and it is the list the as-built pass works down. (6.9.4)

NOTE Every rating in the list is one that NFPA 70 110.9 or 110.10 requires to be at least the available fault current, and each is tested on a different basis: an interrupting rating is a symmetrical current at a tested X/R, a close-and-latch rating is a peak or an asymmetrical current in the first cycle, a short-time withstand rating is a symmetrical current held for a stated time, a bus bracing rating is the first-cycle current the structure holds, a marked short-circuit current rating on a transfer switch, a controller, a drive, a control panel, or a surge protective device is a current the assembly was tested at with a specific class or size of overcurrent device ahead of it, and a conductor withstand is an energy that depends on the clearing time of the device ahead of the conductor. A project leaves an evaluation out where the equipment that carries that rating is not within the study boundary. (7.1.2)

NOTE A series rating is a rating of a pair, not of a device. It holds only while the upstream device named in the listing is in place and while the motor contribution between the two devices stays below the limit, which is why the study records the pair and the listing rather than the downstream device's own rating, and why a replacement of either device is a re-study trigger. (7.2.6)

NOTE A device applied above its interrupting rating does not fail to open; it fails to stop opening, with the arc sustained across contacts that cannot clear it. That is the failure NFPA 70 110.9 exists to prevent, and it is the reason a deficiency is a hold on energization rather than a finding for the record. (7.3.7)

NOTE The marking at the service is the one place the available fault current is readable by the inspector, the installer of the next piece of equipment, and the electrician replacing a breaker, and a value marked from a utility letter rather than from the study omits the growth margin and the date the study was built on. The study is the source of the number; the identification standard is the source of the label. (8.3.6)

NOTE The sealed PDF is the record, the native model file is what every later study opens, the editable one-line is what the Owner's drafting staff fold into the record drawings, and the printed copy is what sits in the electrical room for the inspector; a project adds the last two where its record-drawing and closeout practice call for them. (8.4.2)

NOTE Three studies on three models are three answers to one question, and the one that reaches the label on the door is whichever ran last. One model, handed forward with its assumptions, is how the fault current the arc-flash label is calculated from stays the fault current the breaker was ordered to. (8.5.5)

NOTE The fault current at a bus is a property of everything upstream of it, and most of the events that change it are ordinary maintenance: a transformer swapped for the one in stock, a breaker replaced with another make, a feeder re-pulled on a different route. None of them look like a reason to commission a study, which is why the list is written for the operations staff in the words of the work they do rather than in the terms of the calculation. (9.2.5)