IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 48, NO.

2, MARCH/APRIL 2012 575

Applying IEC 60909, Fault Current Calculations

David Sweeting, Senior Member, IEEE

AbstractRather than the short-circuit current that would

occur in a specific instance, International Electrotechnical Com-
mission 60909 derives the maximum and minimum prospective
short-circuit currents in a system for each specific location and
time. This is reported using a series of parameters which relate to
the rated short-circuit current of equipment and the tests required
on equipment to prove that rating. The influence of arc voltage on
short-circuit currents is then discussed.
Index TermsArc voltage, balanced, calculations, dc com-
ponent, impedance, Joule integral, maximum, minimum, peak,
rating, short-circuit current, symmetrical component, system volt-
age, testing, unbalanced, X/R.

I. I NTRODUCTION

A SHORT-CIRCUIT current is the result of an unwanted

event on a power system that needs to be managed without
causing extensive damage. The protection system must clear the
short-circuit current, and all the equipment subject to the short-
circuit current must not be damaged by it.
In order to achieve this outcome, equipment specifications
require testing with short-circuit currents defined by specific
parameters.
The equipment, however, is unlikely to ever see a short circuit
Fig. 1. Typical waveforms: (Solid) Total short-circuit current, (dashed) ac
with the particular parameters specified in its equipment speci- component (decaying if near to generator), and (dotted) transient dc component
fication. Even a test station is unlikely to match the parameters with (dotted) top and bottom envelopes. IEC 60909 parameters:ip (peak),
exactly. Idc (dc component), Ik  (initial
symmetrical component) times 2 2, and Ib
(typical breaking current) times 2 2.
The applications engineer needs to establish whether the
guaranteed short-circuit current capability of the equipment is
likely to be exceeded by any of the events that could occur in
the system where the equipment is or will be deployed. II. S HORT-C IRCUIT C URRENT C OMPONENTS
This requires calculation of the short-circuit currents that In a simple R + jX inductive circuit, a short-circuit current
may occur under all the normal operating conditions of the consists of a decaying ac component and a decaying dc com-
system. This is different to calculating the short-circuit current ponent. These components add together to provide the total
that has or will occur in a very specific case where every piece current.
of data is accurately known. Close to a generator, the generators change in reactance with
Each short-circuit current has a time-varying waveform which time causes the ac current to fall with time. Far from a generator
needs to be reduced to a set of parameters so that it can be spec- in the supply grid, the ac component is constant.
ified for equipment testing and reported in system calculations. The dc component of the short-circuit current is due to
The short-circuit current parameters that are used relate to the fact that current will not change instantaneously in an
how the short-circuit current affects the equipment that must inductance. At the instant that the short circuit occurs, the ac
carry or interrupt it. component will have the same amplitude and phase angle that
it would have had if it had been there for some time (excluding
Manuscript received June 27, 2011; accepted October 17, 2011. Date of ac decay).
publication December 15, 2011; date of current version March 21, 2012. Paper
2011-PCIC-367, presented at the 2011 IEEE Petroleum and Chemical Industry Since the instantaneous ac value is rarely zero, a decaying
Technical Conference, Toronto, ON, Canada, September 1921 and approved dc transient is generated with the opposite amplitude to the ac
for publication in the IEEE T RANSACTIONS ON I NDUSTRY A PPLICATIONS value at the start of the short-circuit period. This allows the
by the Petroleum and Chemicals Industry Committee of the IEEE Industry
Applications Society. short-circuit current to start from the instantaneous value of
The author is with Sweeting Consulting, St. Ives, N.S.W. 2075, Australia current prior to the short circuit.
(e-mail: david@sweeting.com.au). This produces the current waveforms shown in Fig. 1 for the
Color versions of one or more of the figures in this paper are available online
at http://ieeexplore.ieee.org. case of maximum dc component and asymmetry in the total
Digital Object Identifier 10.1109/TIA.2011.2180011 current waveform. On a highly inductive power system, the
0093-9994/$26.00 2011 IEEE
576 IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 48, NO. 2, MARCH/APRIL 2012

maximum asymmetry is produced by a short circuit that begins Note: This is not the peak value for a particular event
just before a voltage zero. but the highest value depending on X/R and the phase
In order to produce this waveform from equations, it is angle when the short circuit starts.
necessary to specify the voltage amplitude, frequency, X/R 5) The symmetrical short-circuit breaking current Ib is
ratio, and impedance of the system plus the decay constant for the rms value of an integral cycle of the symmetrical ac
the ac waveform. The decay of any dc component is set by the component of the prospective short-circuit current at the
X/R ratio. instant of contact separation of the first pole to open of a
These resultant ac and dc components then must be added switching device.
together to produce the total waveform and the peak. Note: This defines how to measure the rms value of
Unfortunately, in real power systems, the reactance near the ac component at the point in time that is relevant
to a generator is a function of time. Also, in power system to switching devices. This only differs from Ik  near to
components, the resistance is a function of frequency due to the generators where the ac component falls with time.
skin and proximity effects interacting with the large physical 6) The steady-state short-circuit current Ik is the rms
dimensions. value of the short-circuit current after the decay of tran-
While the equation parameters can plot simple system re- sient phenomena.
sponses, large complex systems cannot always be easily re- Within the International Electrotechnical Commission
duced to simple equations with only a few parameters. The (IEC) system, these five parameters are used to define
results of complex calculations and tests need to be easily mea- all short-circuit tests and the calculation of the potential
sured and reported in a manner that allows system calculations short-circuit currents in power systems.
to be compared with test data in a manner that correlates with
how the short-circuit currents react with equipment. IV. OTHER S HORT-C IRCUIT C URRENT PARAMETERS
Outside of the IEC sphere of influence, other parameters will
III. IEC 60909 S HORT-C IRCUIT C URRENT PARAMETERS be found that describe short-circuit currents.
IEC 60909 describes how to calculate and measure the 1) The asymmetrical rms short-circuit current is often
resultant short-circuit current waveform at a specified location defined as the rms value of the first half cycle of an
in a network and at a specific time, either the point of initiation asymmetric current.
of the short circuit or contact opening of switchgear. Note: This is calculated from initiation until the current
1) A prospective (available) short-circuit current is the changes sign and can be 65% larger than the symmetrical rms
current that would flow if the short circuit were replaced current (ac component) of the same waveform in highly reactive
by an ideal connection of negligible impedance without circuits where it lasts a similar percentage longer.
any change of the supply.
Note: This is not the current that would flow during V. S HORT-C IRCUIT C ARRYING C APACITY
a short-circuit test on equipment that has impedance or
produces a voltage but the current that would flow in a The short-circuit strength of power system components can
bolted fault at its incoming terminals. This allows for be specified in a number of ways:
the equipment to change the current to help it survive. 1) rated short-time withstand current Irms together with rate
(For example, fuses and current-limiting circuit breakers peak ip withstand current;
are specified using a prospective current that never flows 2) rated prospective short-circuit withstand current Irms ;
during a test due to the arc voltage limiting the current.) 3) rated conditional short-circuit current Irms ;
2) The initial symmetrical short-circuit current Ik  is 4) rated fused short-circuit current Irms .
the rms value of the ac symmetrical component of a The specified Irms current must occur for at least 0.1 s and the
prospective (available) short-circuit current applicable at specified ip achieved during through-fault tests for withstand
the instant of the short circuit. current ratings in many equipment standards.
Note: This is an rms value of only the ac component not For prospective current ratings, the specified Irms occurs at
the total waveform and is measured between the top and the input terminals of the device, and the actual test current can
bottom envelopes of the current waveform. It is defined at be smaller due to the natural impedance of the device (busbar
the instant of short circuit because the ac component can or switchboard).
decay for near to generator faults and a defining value is For the conditional and fused ratings, the protection specified
required. is allowed to limit the actual current during testing.
3) The decaying (aperiodic) component idc of a short- The value of ip determines the peak mechanical forces
circuit current is the mean value between the top and in equipment and is therefore a major parameter in failure
bottom envelopes at the time the short circuit starts. mechanisms.
Note: This describes how to measure the value from a It is common, however, that the test station cannot achieve
measured or calculated waveform. both the specified Irms and the corresponding value of ip
4) The peak short-circuit current ip is the maximum pos- required by the equipment standard at the same time. This is
sible instantaneous value of the prospective (available) because the X/R ratio assumed in the equipment standard and
short-circuit current. the X/R ratio of the test station circuit rarely match.
SWEETING: APPLYING IEC 60909 FAULT CURRENT CALCULATIONS 577

Test stations therefore commonly perform a peak test of a ac and decaying dc components into one combined value,
few cycles followed by a thermal test. which can be 65% larger than the value of the ac component.
Irms needs to be achieved at the start of the short circuit
for a prospective rating and for up to 0.1 s for a withstand VII. S WITCHING E QUIPMENT
rating. Remember that the ac component decays in generator
test stations. While switching equipment must be capable of carrying a
The Joule integral or I 2 t of the test also needs to exceed the through-fault current when it is not called upon to close or trip,
2 there are many more issues to consider than the fault current
rated Irms times the rated time, which may be specified or come
from the equipment standard. The Joule integral of the current withstand requirements for cables, overhead lines, transform-
 2
i dt is a measure of the heat dissipated in the resistances ers, busbars, and the busbar systems of switchboards.
of the system and therefore determines the temperature reached Except for the asymmetrical breaking current test (test duty
by the components. 5), all the breaking current tests are carried out with symmetri-
The rated short-circuit current Irms therefore defines three cal currents, most of which are directly related to Irms , the rated
parameters of a piece of equipment, which need to be satisfied symmetrical current of the device.
by test. This defines one of the most significant parameters of current
extinction di/dti0 or the rate of change of current at current
1) Irms (required) > Irms (specified). zero. Not only does this defines the rate of contraction of the
2) ip (required) > ip (specified). arcing column and the rate of rise of voltage withstand of the
2 2
3) Irms T (required) > Irms T (specified). extinction mechanism after current zero, it also sets the rate of
rise of the voltage applied by the circuit.
It is therefore important, at the instant of contact separa-
VI. C OMPARING S PECIFIED C APABILITY W ITH
tion, to have the rated symmetrical breaking current of the
S YSTEM R EQUIREMENTS switchgear greater than Ib , the symmetrical ac component of
While the equipment capability is proven by test, the system the prospective short-circuit current.
requirements need to be calculated in accordance with a stan- The peak current ip affects the switchgear in a number
dard that relates to the tested equipment capability. of ways. During a making test, after the contacts prestrike,
On a new power system, the rated Irms of the equipment ip defines the maximum pressure that the mechanism must
is normally selected from the R10 range of numbers with overcome to force the contacts to metallic closure (i.e., mech-
a reasonable margin above the calculated Ik  or prospective anism strength). During the asymmetrical opening tests, it sets
short-circuit current of the system. the maximum pressure in extinction chambers. The following
On an existing power system where the calculated short- three tests are therefore equally important in establishing the
circuit current is approaching the ratings of some of the equip- suitability of switchgear.
ment, much more care is required. All three tests need to be 1) Irms (rated) > Irms (calculated) = Ik  .
applied to prevent equipment failure. 2) ip (rated) > ip (calculated) = ip .
2
3) Irms 2
T (rated) > Irms T (calculated) = Ik 2 Tk .
1) Irms (rated) > Irms (calculated) = Ik  .
They are not the only criteria however. Some switchgear
2) ip (rated) > ip (calculated) = ip .
suffers from minimum and/or critical breaking current issues.
2
3) Irms 2
T (rated) > Irms T (calculated) = Ik 2 Tk .
High-voltage back up fuses will explode if subjected to
The peaking factor in equipment standards (such as IEC currents below their minimum breaking current. (When not
61439 [2] and IEC 62271 [3] for switchgear and controlgear) all of the notches in the parallel elements initially clear, this
may be lower than that in the system due to the assumed X/R leads to restrikes inside the fuse cartridge, excessive energy
ratio, so a simple Irms test is not sufficient. dissipation, and eventual rupture.)
If reclosers are used, the heat generated cannot dissipate in Oil circuit breakers have critical breaking currents that re-
the time between recloses so the sum of all the reclose durations quire special tests that need to be monitored.
is required.
To compare with the equipment rated values, IEC 60909 cal-
VIII. U NBALANCED S HORT C IRCUITS
culates Ik  , the initial rms value of the symmetrical component,
2 While three-phase equipment normally has only one rated
ip , the maximum value of the peak current, and Irms Tk where
Tk is the sum of the durations of each short-circuit current. short-circuit current, it relates not only to three-phase faults but
If any of the rated values are conditional, the applications two-phase and single-phase faults as well, and the equipment
engineer must ensure that the conditions (protection required) standards often set out extra tests to cover the different types of
comply with what was used during the certification testing. fault current.
If American equipment is being used, the applications engi- As well as the initial symmetrical (three phase) short-circuit
neer must ensure that he does not compare the asymmetrical current, IEC 60909 sets out how to calculate the following:
fault-current rating of a piece of equipment with a calculated 1) line-to-earth short-circuit currents Ik1  ;
symmetrical short-circuit current because these are two differ- 2) line-to-line short-circuit currents Ik2  ;
ent things with significantly different values. The asymmetrical 3) line-to-line-to-earth short-circuit currents:
value includes the two IEC 60909 components of symmetrical a) Ik2EL2  ;
578 IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 48, NO. 2, MARCH/APRIL 2012

tion even though tap ratios, loads, and power factor correction
are continually changing.
This involves procedures for deriving the short-circuit im-
pedance of the various system components including im-
pedance correction factors, which ensure that results both near
equipment and out in the network represent the most probable
outcomes.
This allows transformer impedance to be calculated in the
main tap-changer position and shunt capacitance and nonrotat-
ing loads to be neglected.
This reduces a million different calculations to one maximum
and one minimum.
It should be noted that most of the data required are often not
Fig. 2. Transformation of unbalanced fault currents across a delta star
transformer.
available and default values are required. These are provided in
the different parts of IEC 60909.
b) Ik2EL3  ;
c) IkE2E  . X. M AXIMUM AND M INIMUM S HORT-C IRCUIT C URRENTS
Note: In line-to-line-to-earth short circuits, the currents
Maximum short-circuit currents need to be calculated be-
in each phase and earth all have different amplitudes.
cause they determine the rating required for the equipment on
As well as the different currents in the fault, each of these the system. They should allow for foreseeable system upgrades
unbalanced faults has different current transformation ratios that could occur independently of repeating the calculations and
across transformers depending on the vector group of the confirming ratings.
transformer (Fig. 2). While IEC 60909 does not call for it, it is useful to also
While with the common delta/star vector group a three- calculate a present maximum, which is needed to check present
phase current transforms with the ratio of the high-voltage protection grading and investigate incidents.
to low-voltage phase-to-phase voltages, a single-phase current Calculation of minimum short-circuit currents is required to
transforms with the ratio of the high-voltage phase-to-phase ensure that protection will pick up and trip with the setting ap-
voltage to the low-voltage phase-to-earth voltage in two of the plied. The protection also must not explode because the current
three high-voltage lines. is below the minimum breaking current of some equipment.
With unganged high-voltage fuses, after the first fuse clears IEC 60909 introduces the concept of a voltage factor c to
in a three-phase low-voltage fault, this leads to a reduction of calculate the maximum and minimum short-circuit currents.
current in the remaining high-voltage fuses. There has been at For the maximum short-circuit current, cmax Un is equal
least one case where the second and third fuses never cleared, to the highest system voltage Um from IEC 60038 (standard
and manual tripping was required hours later. voltages) or the highest voltage for equipment and represents
With a delta/star transformer, a low-voltage phasephase the highest voltage that should appear under normal conditions
current transformed to high-voltage line currents becomes, in at any location in the network at that voltage level.
one phase, two and, in two phases, one times the high-voltage For the minimum short-circuit current, cmin Un is the lowest
phase-to-phase voltage divided by the low-voltage phase-to- supply or utilization voltage from IEC 60038 at that location in
earth voltage. the network.
The different transformation ratios for different faults can The two voltage factors should be used with the most likely
lead to grading issues. impedance of each component. If accurate data are not avail-
able, the most likely value rather than an extreme value should
be used. In other words be conservative once only and not with
IX. IEC 60909 C ALCULATIONS the choice of every parameter.
The IEC 60909 calculation method uses an equivalent source The impedance correction factors are derived for the maxi-
voltage at the short-circuit location driving into the short-circuit mum case and need to be applied with thought for the minimum
impedance of the network with all other voltage sources set cases.
to zero. The supply network is not the same for the maximum and
Symmetrical components are used to define the positive minimum cases.
sequence, negative sequence, and zero sequence impedances For the maximum case, the following data should be
of the system and its components in order to calculate the included.
unbalanced short-circuit currents. 1) cmax Un .
Any power system with multiple voltage levels and voltage 2) The impedance corresponding to the maximum short-
control using transformer tap changers and power factor control circuit current that the supply authority reserves the
has many different configurations per day let alone per year. right to provide to the premises. (New transformer just
The standard sets out how to derive what is the most likely outside.)
prospective maximum and minimum fault currents at that loca- 3) All parallel transformers in service.
SWEETING: APPLYING IEC 60909 FAULT CURRENT CALCULATIONS 579

4) If there is a ring main, it should be closed.

5) Calculation performed at input terminals of a low-voltage
switchboard.
For the minimum case, the following data should be
included.
1) cmin Un .
2) The impedance corresponding to the minimum short-
circuit current that the supply authority guarantees to
provide that connection. (Together with voltage control
this can be derived from minimum capacity.)
3) Minimum parallel transformers in service (at least one out
of service if it can run that way).
4) If there is a ring main, it should be open.
5) Transformer to switchboard impedance included.
6) Calculation performed at the end of a low-voltage switch-
board, where an arc will run to and burn.
7) A calculation including the influence of the arc voltage
for low voltage.

XI. I NFLUENCE OF A RC VOLTAGE ON

S HORT-C IRCUIT C URRENT
Fig. 3. Vector diagrams of a nominal 460-V 20-kA bolted fault and the same
The impedance of an electrical supply network is mainly system with a 350-V equivalent arc voltage. Scales are voltage (1000 V) and
inductive with a small resistive component, and the impedance current (20 000 A).
of loads is mainly resistive with an inductive component (0.8
power factor). This keeps the system losses within reasonable
levels.
The voltage across an arc is however in phase with the current
and out of phase with the reactive-impedance voltage drop.
While arcs on parallel electrodes tend to produce a sine-
wave-like voltage waveform (with plenty of distortion), they are
not resistive because of the negative v/i characteristic. (From
zero to around 1000 A (instantaneous), the arc voltage/cm falls
to around 10 V/cm and thereafter rises slowly with current. This
dominates behavior which is different to a resistor where the
voltage increases with instantaneous current.)
The arc voltage is also only a weak function of electrode
separation and therefore system voltage. The sine-wave-like
waveform is due to the arc lengthening and shortening during
each cycle, and the same behavior dominates over the influence
of electrode separation.
The following vector diagram shows the influence of a
350-Vrms equivalent arc voltage (fundamental frequency com-
ponent phase to phase) on a 20 000-A bolted short circuit in
a nominal 460-V system with an X/R of ten (Fig. 3). It was
calculated at the utilization voltage minimum of 424 V.
The 20-kA prospective fault current at the maximum 508 V
Fig. 4. Vector diagrams of a nominal 600-V 20-kA bolted fault and the same
drops to 17.3 kA at the minimum 424 V. With 350-V arcing system with a 350-V equivalent arc voltage. Scales are voltage (1000 V) and
volts, the current falls to 8.2 kA or 41% of the maximum current (20 000 A).
prospective current and 47% of the minimum nonarcing short-
circuit current. It is a completely different story for a 20-kA bolted short
Even on a nominal 600-V system, the arc voltage has signif- circuit at the maximum 4.44 kV in a nominal 4.16-kV system
icant influence (Fig. 4). The 20-kA prospective fault current at with the same X/R of ten (Fig. 5). The minimum voltage of
the maximum 635 V drops to 16.4 kA at the minimum 520 V. 3.6 kV drops the short-circuit current to 16.2 kA.
With 350-V equivalent arcing volt phasephase, the minimum The 350-V equivalent phasephase arcing voltage only re-
current falls to 11.2 kA or 56% of the maximum prospective duces the 16.2 kA to 16 kA or 99% of that when arcing voltage
current and 68% of the minimum nonarcing current. is not considered. This is because not only the arc voltage is ten
580 IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 48, NO. 2, MARCH/APRIL 2012

system independent of the continually varying load currents and

transformer tap positions but taking into account variations in
system configurations.
The results of IEC 60909 calculations are reported as
follows:
1) the rms value of the symmetrical component of the short-
circuit current:
a) Ik  at the start of the short circuit;
b) Ib at contact opening of protection;
2) the maximum peak current ip ;
3) the Joule integral of the current Ik 2 Tk .
It has been shown how these parameters relate to the rated
short-circuit current Irms of equipment and how short-circuit
tests on equipment, in turn, relate to its rated short-circuit
current.
As well as balanced three-phase short-circuit currents, IEC
60909 provides procedures to calculate unbalanced currents
(Ik1  , Ik2  , Ik2EL2  , Ik2EL3  , and IkE2E  ), which are partic-
ularly important in setting protection.
While IEC 60909 does not take into account arcing voltage,
it has been shown that, while arcing voltage is a minor issue at
high voltage, it has a considerable impact at low voltage.
Fig. 5. Vector diagrams of a nominal 4.16-kV 20-kA bolted fault and the same
system with a 350-V equivalent arc voltage. Scales are voltage (5000 V) and R EFERENCES
current (20 000 A). [1] The Calculation of Short-Circuit Currents in Three-Phase A.C. Systems,
IEC 60909.
[2] Low-Voltage Switchgear and Controlgear Assemblies, IEC 61439.
times smaller in relation to the system voltage but also it is at [3] High-Voltage Switchgear and Controlgear, IEC 62271.
79 to it, and therefore, it has an even smaller influence. [4] D. K. Sweeting, The electric arc, Ph.D. dissertation, School Elect. Eng.,
Univ. Sydney, Sydney, Australia, 1969.

XII. A RCING H AZARDS

With the normal time-graded protection on distribution sys-
tems, the largest arc energy is normally produced by the
minimum arcing fault current because the influence of the
increased time for the protection to trip exceeds the influence David Sweeting (M09SM11) received the B.S.,
B.Eng., and D.Phil. degrees in electrical engineering
of the smaller arcing current. This is the reason that minimum from The University of Sydney, Sydney, Australia.
short-circuit currents including the arc voltage influence are an He spent four years in Switzerland with Brown
essential calculation. Boveri where he worked on circuit breaker develop-
ment and testing before returning to Australia to join
In some systems, the minimum short-circuit current during the Sydney County Council, an electrical distribu-
arcing may not trip the protection at all. tion authority where he held positions in substations
This is not only a case of insufficient current to pick up and overhead transmission and ran the high-power
testing station. In 1980, he joined Bassett Consulting
the protection. The waveform of low-voltage fault currents Engineers where he became an Associate Director. In
can include periods where all three currents remain near zero 1989, he formed Sweeting Consulting, St. Ives, Australia, which is a specialist
for a number of cycles before the insulation fails again. The high-voltage electrical engineering firm. He is a Visiting Professor at the
University of Wollongong, Wollongong, Australia. He has published 70 papers
protection must be able to detect such discontinuous waveforms on circuit breakers, electric arcs, testing, switching over voltages, cable fault
without resetting. location, motor failures, computer-aided design, standards, safety, standard
voltages, power quality, and electric arc burn hazards.
Dr. Sweeting is an Honorary Fellow of The Institution of Engineers Aus-
XIII. C ONCLUSION tralia, a Senior Member of the IEEE Industry Applications Society, and a
member of Consult Australia and the International Council on Large Electric
IEC 60909, short-circuit currents in three-phase ac systems, Systems (CIGRE). He has been a member of a number of International CIGRE
committees, International Electrotechnical Commission working groups, and
calculates the prospective maximum and minimum values of Standards Australias committees covering short-circuit currents and power
the short-circuit current at a particular location in a power quality.