Document No.

GP 12-01
Applicability Group
Date 04 November, 2003

Guidance on Practice for

Power System Design

GP 12-01

BP GROUP
ENGINEERING TECHNICAL PRACTICES

DO NOT COPY
04 November, 2003 GP 12-01
Guidance on Practice for Power System Design

Foreword

This is the first issue of Engineering Technical Practice (ETP) GP 12-01. This Guidance on Practice
(GP) is based on the following heritage documents from merged BP companies:

Amoco (ACES)
EL-SYS-00-G Electrical—Systems—Guide
EL-SYS-00-E Electrical—Systems Design—Specification

BP RPSE
RP 12-3 Electrical Systems and Installations Power System Design

DO NOT COPY
Copyright © 2003, BP Group. All rights reserved. The information contained in this
document is subject to the terms and conditions of the agreement or contract under which
the document was supplied to the recipient’s organization. None of the information
contained in this document shall be disclosed outside the recipient’s own organization
without the prior written permission of Manager, Standards, BP Group, unless the terms of
such agreement or contract expressly allow.

Page 2 of 41
04 November, 2003 GP 12-01
Guidance on Practice for Power System Design

Table of Contents
Page
Foreword............................................................................................................................................2
1. Scope........................................................................................................................................5
2. Normative references ...............................................................................................................5
3. System planning .......................................................................................................................5
3.1. Basic design considerations ..........................................................................................6
3.2. Site considerations ........................................................................................................6
3.3. Energy conservation ......................................................................................................7
4. Power system parameters ........................................................................................................7
4.1. Voltage ..........................................................................................................................7
4.2. Frequency......................................................................................................................7
4.3. Short-circuit rating .........................................................................................................8
4.4. Methods of limiting short-circuit currents .......................................................................8
4.5. Frequency and voltage regulation .................................................................................9
4.6. Reliability .....................................................................................................................11
5. Power system loading.............................................................................................................11
5.1. Equipment rating and diversity factors ........................................................................12
5.2. Types of load ...............................................................................................................12
6. Power utility sources...............................................................................................................13
6.1. General........................................................................................................................13
6.2. Electrical import from a public utility ............................................................................14
6.3. On-site generation with no public utility connection.....................................................14
6.4. On-site generation run in parallel with a public utility ..................................................15
6.5. Emergency power supply equipment ..........................................................................15
7. Power distribution system.......................................................................................................17
7.1. Primary substation .......................................................................................................17
7.2. Site power distribution .................................................................................................18
7.3. Substation location ......................................................................................................21

DO NOT COPY
7.4. Substation design ........................................................................................................21
8. Power system earthing/grounding ..........................................................................................22
8.1. General........................................................................................................................22
8.2. Unearthed/ungrounded (isolated) neutral....................................................................23
8.3. Solidly earthed/grounded neutrals – low voltage power systems ................................24
8.4. Impedance earthed/grounded neutrals – medium voltage power systems .................24
8.5. Power generator earthing/grounding ...........................................................................25
8.6. Earthing/grounding resistors........................................................................................26
9. Power system operation and control ......................................................................................27
9.1. General........................................................................................................................27

Page 3 of 41
04 November, 2003 GP 12-01
Guidance on Practice for Power System Design

9.2. Power generator operation and control .......................................................................27

9.3. Synchronising ..............................................................................................................28
9.4. Load transfer schemes ................................................................................................28
10. Power system performance ....................................................................................................30
10.1. Motor starting...............................................................................................................30
10.2. Over-voltages ..............................................................................................................31
10.3. Harmonics ...................................................................................................................32
10.4. Power factor ................................................................................................................34
11. Power system studies.............................................................................................................35
11.1. General........................................................................................................................35
11.2. Normal power system studies .....................................................................................36
11.3. Short circuit studies .....................................................................................................37
11.4. Stability studies............................................................................................................39
11.5. Harmonic analysis studies...........................................................................................40
12. Power system protection and control......................................................................................40
Bibliography .....................................................................................................................................41

DO NOT COPY

Page 4 of 41
04 November, 2003 GP 12-01
Guidance on Practice for Power System Design

1. Scope

This GP provides guidance for electrical power distribution system design and should be utilised in
conjunction with GP 12-02 and GP 12-03.

2. Normative references

The following normative documents contain requirements that, through reference in this text,
constitute requirements of this technical practice. For dated references, subsequent amendments to, or
revisions of, any of these publications do not apply. However, parties to agreements based on this
technical practice are encouraged to investigate the possibility of applying the most recent editions of
the normative documents indicated below. For undated references, the latest edition of the normative
document referred to applies.

BP Engineering Technical Practices (ETP)

GP 12-02 Guidance on Practice for Power System Installation & Construction
GP 12-03 Guidance on Practice for Power System Protection and Control
GP 12-05 Guidance on Practice for Transformers and Reactors
GP 12-10 Guidance on Practice for Switchgear & Controlgear
GP 12-20 Guidance on Practice for Motors
GP 12-25 Guidance on Practice for Earthing/Grounding
GP 12-35 Guidance on Practice for UPS Systems
GP 12-60 Guidance on Practice for Hazardous Area Electrical Installations
GP 12-80 Guidance on Practice for Power Generators
GP 12-85 Guidance on Practice for Emergency Generators

American National Standards Institute

ANSI C84.1 American National Standard for Electric Power Systems and Equipment
Voltage Ratings (60 Hertz)

Institute of Electrical and Electronic Engineers (IEEE)

IEEE Std C2 National Electrical Safety Code
IEEE Std 399 Recommended Practice for Industrial and Commercial Power Systems
Analysis (IEEE Brown Book)
IEEE Std 519 Recommended Practice and Requirements for Harmonic Control in
Electrical Power Systems

DO NOT COPY
International Electrotechnical Commission (IEC)
IEC 60038
IEC 60909-0
IEC Standard Voltages
Short-circuit currents in three-phase a.c. systems: Part 0 – Calculation of
currents

3. System planning

IEEE Std 141, chapter 2 forms the overall basis of this section.

Page 5 of 41
04 November, 2003 GP 12-01
Guidance on Practice for Power System Design

3.1. Basic design considerations

a. The following points should be considered when establishing a basis for the planning of an
electrical power distribution system:
1. Safety – of personnel and equipment
2. Reliability of primary electrical power source
3. Reliability of power distribution system
4. Simplicity of operation
5. Voltage regulation throughout the power distribution system
6. Maintainability
7. Flexibility – ability to expand and adapt
8. First cost
b. The following points are a guide to what should be included in any power distribution
system planning process:
1. Electrical load definition and load forecasting
2. Selection of an appropriate power distribution system philosophy
3. Locations of major power distribution equipment – in relation to consumer locations
4. Utilisation voltages
5. Primary power utility source
6. On-site generation – if any
7. Single line diagram
8. Power system studies including load flow, short circuit and stability analysis
9. Power system protection and protective device coordination
10. Communications – throughout the power distribution system
11. Maintenance – programming and scheduling
c. If the primary power utility source is from a public utility company, the following
information should be obtained from the utility company:
All negotiations with public utilities should be the sole responsibility of the site
operator.
1. System data – number of phases, nominal voltage, frequency, etc
2. Voltage regulation.

DO NOT COPY 3.
4.
5.
6.
Short circuit ratings
Reliability of utility company supply
Capacity and ratings of incoming feeders
Number of incoming feeders and whether in parallel operation
7. If feeders have automatic transfer, or automatic reclose
8. Protection relay schemes and requirements for coordination with utility company
protection system

3.2. Site considerations

a. Site practices of existing facility should be reviewed and incorporated into design.

Page 6 of 41
04 November, 2003 GP 12-01
Guidance on Practice for Power System Design

b. Electrical equipment numbering, associated data books, and other references should be
coordinated with existing plant numbering system.

3.3. Energy conservation

During the design process associated with power systems attention should be given to the
conservation of energy, if there is an economic advantage when lifetime costs are taken into
account.
The economic evaluation of alternative arrangements should take into account a
lifetime NPV for the project and should include costs for fuel and for any CO2 levy.

4. Power system parameters

4.1. Voltage
IEEE Std 141, chapter 3 forms the overall basis of this sub-section. The clauses in
this whole document refer to the nominal system voltage ranges given in chapter 3
and ANSI C84.1, defined as follows:
• Low voltage 1000V and below
• Medium voltage from 1000V to 100kV
• High voltage from 100kV to 230kV
a. The nominal system (phase-to-phase) voltage should be selected from those detailed in
IEC 60038 or ANSI C84.1 as appropriate.
b. The most economic nominal system voltages should be selected for a power system
bearing in mind the possible need to distribute electrical power over considerable distances
within the site.
The requirement for having the most economic nominal system voltage is intended to
ensure that an economic appraisal of alternatives is undertaken or has been
considered. Such appraisals should include consideration of operational costs
(including cost of losses) over the expected life of the installation.
c. In general the nominal system voltages selected should agree with the voltage levels for
electrical machines (see GP 12-20).
d. For 'green-field' sites or offshore locations with a low possibility of interconnection to an
outside public utility , the use of non-standard nominal system voltages for low voltage
systems should be based on economic considerations.
Selection of a non-standard nominal system voltage (such as 690V) should be based
upon specific economic studies, which take full account of the need to hold spares,
which may be beyond normal stock levels.

DO NOT COPY
4.2. Frequency
a.

b.
The frequency selected for power systems should be either 50 or 60 Hz depending on the
national standards of the site location.
If a site is to be remotely located and powered only from on-site generation, the frequency
selected should be compatible with the frequency of supply of the nearest local electrical
supply source or public utility, particularly if it is envisaged that future inter-connection
may be desirable.
c. If future inter-connection with other sources of electricity supply is not envisaged as being
possible or if economic advantages outweigh the benefits of future possible
interconnections, the frequency of supply should be selected on economic considerations
and on ease of obtaining suitable equipment.

Page 7 of 41
04 November, 2003 GP 12-01
Guidance on Practice for Power System Design

Offshore experience indicates that selecting 60 Hz frequency tends to offer possible

weight, and therefore economic, advantages and since such equipment is widely
available this selection becomes more likely.

4.3. Short-circuit rating

The definition of short-circuit terms used in this sub-section can be found in
chapter 7 of IEEE Std 399
a. All switchgear and distribution equipment on a power distribution system should be
capable of carrying the prospective rms symmetrical breaking short-circuit current for a
specified short-time duration of 1 or 3 seconds without deleterious effect.
The choice between 1 and 3 second durations would be dictated by equipment
availability, economics and fault current protection clearing times.
The backup fault current protection clearing times should always be less than the
equipment short-time current rating.
Generally 3-second short-time current ratings for switchgear would be preferred to
avoid the necessity for rapid protection.
b. The closure of switchgear on to a short-circuit type fault should not result in mechanical
shock load damage to healthy parts of the power system as a result of peak asymmetrical
make currents flowing.
c. Medium voltage power systems should be designed such that the rms, symmetrical
breaking short-circuit current of circuit breakers should be selected from rating values
readily available from equipment manufacturers.
There are situations (e.g. when on-site generation is provided) when switchgear of
higher short circuit capacity (such as 50 kA rms symmetrical) may be necessary,
which may only be available from a small number of manufacturers.
d. The selection of 25 kA rms symmetrical breaking short-circuit current capacity for medium
voltage power system, general-purpose applications, may prove to be an economic ceiling
for the available fault levels of equipment.
If there is an economic advantage in providing switchgear of higher ratings, such as
40 or 50 kA rms symmetrical, then this should be considered.
e. In a green field situation, the short circuit capability of the installation should be based on
economic grounds.
f. The short-circuit capability of switchgear should always be proven by type testing before
being accepted into full duty service.
Type testing would be particularly important if 'enhanced make', peak short-circuit
current or enhanced d.c. content in breaking short-circuit current would be

DO NOT COPY
g.
required. See GP 12-10 for testing requirements.
For low voltage power systems the rms symmetrical breaking, short-circuit current of
general-purpose circuit breakers should not normally exceed 50 kA rms symmetrical.
If economically viable the selection of 80 kA rms symmetrical or above rated
switchgear may be considered for low voltage power systems.

4.4. Methods of limiting short-circuit currents

a. If the power system design indicates prospective short-circuit levels would exceed the
proposed circuit breaker ratings, the following measures should be considered:
1. Increase the power system reactance, provided this causes no technical problems such
as voltage regulation or a commercial problem.

Page 8 of 41
04 November, 2003 GP 12-01
Guidance on Practice for Power System Design

2. Change the power system operating mode by operating with certain circuit breakers
open and provide auto-transfer facilities to reinstate the power supply security and
quality levels.
3. Purchase switchgear and equipment designed and certified for the higher short-circuit
levels if these are available.
4. Provide fault current limiting devices other than fuses.
5. Carry out any combination of the alternatives listed above.
The selection of the most appropriate of the alternatives should be based on least
cost.
Saturable reactors (used as resonant links) could offer a good solution for resolving
short-circuit limitations for medium voltage systems above 20kV, and high voltage
systems and they do not suffer from regulation problems. Their expense generally
precludes their use.
b. The use of Is limiters should be avoided for new installations but could be useful if existing
systems are to be expanded or interconnected with new systems.
Care should be taken in designing systems for Is limiters to ensure supply security
would not be jeopardised and that cost would be justified.
c. Fault-limiting (series) reactors used to increase the reactance between the source and
potential fault location, should comply the guidance given in GP 12-05.
d. If fault-limiting reactors are considered to be necessary, they should preferably be installed
between busbar sections, to limit potential power system voltage profile problems under
normal operating conditions.
e. The impedance of series reactors should be chosen to limit the fault current passed through
the reactor to a level which ensures that total calculated short-circuit levels on either side
of the reactor would be no greater than 90% of the selected equipment short-circuit ratings.
f. Series reactors may be single or three phase, dependent on space and cost considerations.
They should have a thermal ability to carry the rated short circuit symmetrical current for
at least 2 seconds.
The time rating of the reactor would be dependent upon the protection
arrangements.
g. The voltage drop across series reactors under the motor start, restart or re acceleration
conditions should be analysed to ensure that no unacceptable transient under-voltage
would occur.
h. If series reactors are installed between two power systems, the transient stability of the
power generator sets should be analysed.

DO NOT COPY
4.5. Frequency and voltage regulation
a. Power systems should be designed to ensure that the voltage variation experienced at the
terminals of any load equipment should not exceed ± 5% of the declared system nominal
voltage for power system loadings between full load and lightly loaded conditions.
If the system nominal voltage deviates beyond the 5% limit, apart from equipment
life expectation being eroded, there is a danger of exceeding the certification limits
for equipment installed in a hazardous area. This may not actually cause failures or
temperatures likely to cause ignition of a vapour, which may be present, but will
increase the likelihood of such occurrences and will therefore infringe the basic
safety intention.

Page 9 of 41
04 November, 2003 GP 12-01
Guidance on Practice for Power System Design

b. The power system should be arranged to be able to withstand without undue stress a
transient voltage reduction of up to 20% of nominal for typically a 10 second period or
longer.
Motor starting requirements could give rise to such a voltage depression at motor
terminals of 20%, and the power system should be able to successfully accept this
condition for the starting period. Typical motor starting periods can range from 1 to
10 seconds depending on machine and system ratings and machine inertia.
Starting times beyond 10 seconds are possible for particularly high inertia loads
such as fans. These would constitute a special application and should be treated by
exception.
c. Special consideration should be given to power frequency over-voltage effects and the
need for power system equipment to withstand higher over-voltages under some
circumstances.
This requirement ensures that standard equipment parameters would not be
compromised.
Power systems supplied by power generators that have automatic voltage control
could experience short-term over-voltages caused, typically by generator over-
excitation following a direct-on-line start of a large electrical machine. Significant
over-voltage levels may be experienced (120% to 130% of nominal system voltage
are not untypical).
This would require special specification of transformers and other equipment to
avoid problems caused by over-fluxing or increased current due to magnetic
excitation.
d. Special attention should be paid to voltage and frequency tolerance specifications for
generator transformers, which may need to match the run up characteristic of the power
generator.
This could require the transformer to be capable of withstanding full rated voltage
at 80% of rated frequency.
e. Site incoming transformers should be equipped with on-load, automatic tap change
equipment for voltage regulation purposes. (See GP 12-05)
f. Automatic voltage regulation equipment of power systems which have both on-site power
generation and public utility intakes via transformers, should be designed so that there
would be no detrimental inter-action.
Consideration should be given to the deliberate introduction of appropriate
operational time constants to prevent unnecessary hunting between transformer tap
changers and power generator AVR (automatic voltage regulators).

DO NOT COPY
g. The excitation systems of synchronous motors should be arranged to ensure no detrimental
interactive effects with transformer automatic tap changer equipment.
Synchronous motor excitation could be arranged to maintain a constant power
factor to the machine and the transformer tap changer control could be arranged to
maintain system voltage within pre set levels.
h. If a site has no public utility connection, the controls associated with the governors of on-
site power generation should be designed so that generator loading should be automatically
shared between the operating sets and that the nominal site supply frequency should be
maintained within a specified tolerance.
If governor droop control is provided, there should be an operator in attendance
who could take action from time to time.

Page 10 of 41
04 November, 2003 GP 12-01
Guidance on Practice for Power System Design

An automatic power system management system could be provided to ensure that the
system frequency remains within reasonable limits. Such limits should be ± 1% and
the output system frequency time should be within 30 seconds of ‘standard’ time.
A ‘standard’ clock arrangement should be provided for power plants supplying
townships or camp areas. The limit for frequency time may be relaxed for small
power plants, and those where frequency time is considered unimportant.
For unmanned power plants or those where constant operator intervention would be
undesirable, an isochronous controller designed to provide overall frequency and
power generator loadings within prescribed limits should control the system
frequency.
i. The controls associated with the governors of on-site power generation at a site with a
public utility connection should be designed for parallel operation with the public utility
and should incorporate protection devices which should ensure that under public utility
supply disconnection (or unacceptable major disturbance on the local grid network), the
site would retain a defined loading condition. The controls should also incorporate means
for independent operation, with adequate control of frequency and voltage.
If a facility is interconnected with a public utility, the control of the power system
frequency must be under the control of the public utility. Only defined load governor
settings or droop load sharing control systems would be possible. The
interconnection would be always subject to disconnection under automatic
protection system operation, which could leave the on-site power generators
supplying the site load. Under these conditions, whatever the original power
generator control arrangements, the power generators would have to revert to
independent load sharing control.

4.6. Reliability
a. For each power supply arrangement (‘normal’, ‘maintenance’ and ‘unplanned outage’), a
reliability assessment should be carried out to determine the probability of failure of
supply. The results of such an assessment should provide input data for economic analyses.
b. In the early stages of design, a reliability study should be undertaken as part of the
comparison of alternative methods of providing power supplies.
Such studies would also be useful in determining the cost of failures.
c. The terms of reference of these studies relating to objectives, data gathering, and
evaluation should be clearly specified for those undertaking these studies.
These studies may form part of a contractor’s work scope or the work scope of a
project team.
d. Assessment of power distribution system reliability should be carried out as part of an

DO NOT COPY
overall assessment of plant failure and also as part of a comparison of alternative supplies
for new plant. The mean time between failure (MTBF) and the mean time to repair
(MTTR) for a power system failure at any particular switchboard should be determined. In
addition, the quantitative statement should indicate the probability of failure occurring over
a particular timeframe.
For example a value and definition of the probability of zero failures over a 12-
month period should be determined.

5. Power system loading

IEEE Std 141, chapter 3 forms part of the basis of this section.

Page 11 of 41
04 November, 2003 GP 12-01
Guidance on Practice for Power System Design

5.1. Equipment rating and diversity factors

a. A load list should identify all process loads and should further identify the power to be
absorbed by the mechanical drive. This should be adjusted for efficiency of the electrical
drive to arrive at a value of power to be drawn from the electrical system.
b. Power system loading estimates should be based on absorbed demands for machine loads
with appropriate allowances for efficiency and power factor. In addition the effect of
diversity should be assessed using a concept of intermittence or the application of well-
proven diversity factors.
In the absence of either of these methods, an average maximum loading expectation
may be assessed by adding 50% of the intermittent load (that controlled by level
switches or by thermocouples or other control devices) to the continuously running
loads.
c. Assessment of maximum load requirements for an installation should allow for diversity
between various loads, drives or plants. Appropriate factors should consider the likelihood
of loads, drives, production or process trains coincidentally requiring peak demands and
should be based on similar installations.
d. Aggregated plant loading should be carried out taking account of diversity and any other
factors. This ‘after diversity’ load estimate should only include one allowance for
contingency.
e. In addition to this aggregated load, an assessment should be made of operating standby
load.
An estimate of maximum of standby load operation may be obtained by considering
such scenarios as boiler outage (insufficient steam for steam turbine drivers) or a
maximum product throughput condition.
f. Power system equipment should be rated to continuously carry the maximum ‘after
diversity’ load associated with peak design production with an additional 10%
contingency.
g. If ambient conditions are likely to exceed 40°C maximum air temperature and an altitude
of 1000 m, this should be defined in equipment specifications.
The specification of maximum temperature and elevation should ensure that
equipment has sufficient capacity for the most arduous likely conditions. If there is
no possibility of reaching the 'standard' 40°C level at a particular location and there
is economic advantage in relaxing the temperature requirement, the actual
maximum and minimum temperatures should be specified.
h. If a total system load is to be estimated and includes a number of individual switchgear
loads, care should be taken to ensure there is only one overall estimate for contingency.

DO NOT COPY
5.2.

5.2.1.
Types of load
All loads should be presented with clear definition of their types in accordance with the
following:

Basic loading information

a. For each identified electrical load, a description of the service should be given together
with an indication of whether the device is a machine (a dynamic load) or a static load.
1. Dynamic electric motors driving rotating equipment
This type of load is characterised by power being proportional to frequency rather
than voltage. Dynamic loads tend to be constant power regardless of voltage
deviation.

Page 12 of 41
04 November, 2003 GP 12-01
Guidance on Practice for Power System Design

2. Static non-moving types of electrical equipment such as lighting, heating, capacitor

banks, supplies to rectifiers, etc.
Variable speed drives exhibit properties similar to those of dynamic loads viz.
constant power demand despite voltage variations. However, power is also
independent of supply frequency deviations and their contribution to system fault
level is limited by the power electronic system, in some cases to zero contribution.
b. The bulk of the loads on the majority of petrochemical installations comprise dynamic
loads and the proportions of dynamic to static loads are generally in the ratio of 10:1.

5.2.2. Essential loads

a. These are loads which are important to the safety of the installation or the operational staff,
and which require power to permit a controlled shutdown of process plant and facilities in
an emergency. They should have a second independent power source and be generally
associated with no break supplies.
In certain cases, a short supply break of say 15 minutes may be acceptable if this
does not represent a hazard to safety.
Included in the concept of essential loads would be such devices as emergency
shutdown systems, lube oil pumps for gas turbines and large compressors,
ventilation systems, helipad lighting, emergency lighting, some floodlighting
systems, navigational aids, radio systems etc.
b. Some loads, which are essential by this definition, may require the provision of emergency
power supply facilities in order to meet legislative requirements such as NFPA 70
(guidance is available in IEEE Std 446).

5.2.3. Process sensitive loads

a. These are loads whose loss would affect continuity of plant operation, resulting in loss of
revenue but would not result in an unsafe situation arising.
Included in this category would be general lighting systems and process demands.
It would sometimes be possible to include sufficient capacity in an emergency
generator to allow some processes to continue on reduced capacity when main
power generators or main utility supply source would not be available. The
application of such loads should not compromise any emergency services. Each case
would be project specific.
b. Any decision to provide an alternative source of supply for these types of load should be
based on economic assessments.

5.2.4. Non-sensitive loads

DO NOT COPY
a. Non-sensitive loads are those which do not form an important component of a production
or process plant and their disconnection could be borne for relatively long periods with
minimal or nuisance value. They usually form a small proportion of the total connected
load and may have a single power source.
Typical examples of these loads would be the domestic part of accommodation,
offices, workshops, some water injection, some potable water systems, etc.

6. Power utility sources

6.1. General
a. Power supply arrangements are fundamental to any operating site or facility and should
feature in any overall project concept. A power supply philosophy document which details

Page 13 of 41
04 November, 2003 GP 12-01
Guidance on Practice for Power System Design

the arrangements for power supply and the reasoning associated with any selection should
be included in any project concept documentation.
A vendor may have part of their work scope associated with an evaluation and
selection of power supply arrangement. The selection of the power supply
arrangements should form part of a feasibility (or statement of requirements) study.
b. If the power system will be significant there should be an economic comparison of
alternative design arrangements.
c. Large projects should initiate a formal safety review and this process should ensure that the
basic safety needs of the project would be satisfied.
d. Smaller projects or plant extensions where such a formal review may not be included
should consider safety aspects relating to the electrical system and produce an outline
document which presents these issues for later auditing.

6.2. Electrical import from a public utility

a. Agreements with public utility companies will form part of the long-term operating
parameters for a site or facility and it would be necessary for the site operator to be fully
aware of these agreements and have sole responsibility for long-term compliance with the
requirements arising from any negotiations.
Such responsibility should not be delegated to a third party unless the third party
has an ongoing operational responsibility.
Others may request information from public utility companies on behalf of the site
owners but in doing so it should be made clear that any formal agreement which
arises from the information or investigation will be with the site owner.
b. If the principal source of electrical power is selected to be from a public utility, the supply
should be via fully rated duplicate feeders.
An exception to this may be permitted for economic reasons if non-sensitive loads
are to be supplied or the power from the public utility is standby to on-site power
generation.
There should be a cost/benefit statement to justify the use of single feeds.
c. Single supplies with on-site standby generation that would operate to supply full load in
the event of a main power supply failure may be an economic alternative to a dual supply
arrangement.
d. Essential loads should always be provided for by on-site power supply equipment.
e. If economically viable, on-site power generation may be arranged to peak-lop or otherwise
supplement the external power supply, especially if there are tariff advantages.

DO NOT COPY
6.3. On-site generation with no public utility connection
a. If a site or facility is offshore, or remote from a public utility network, or has a surplus of
fuel or process energy, on-site power generation would normally be the principal source of
power. On-site power generation arrangements would be dependent on the following
factors:
1. The fuel source needs to be reliable
2. The nature of the process energy could facilitate dual applications such as co-gen
3. The process steam or other heat requirements, if any could determine co-gen
possibilities
4. The relationship between electric power requirements and the site energy sources

Page 14 of 41
04 November, 2003 GP 12-01
Guidance on Practice for Power System Design

b. The number and arrangement of power generating sets should reflect the particular needs
of the installation in terms of availability and reliability.
c. If an availability of above 80-85% is required, at least 2 power generator sets would
normally be required on sites if there would be no alternative electricity supply. Under
these circumstances the following criteria should be satisfied:
1. There should be sufficient power generation to meet the maximum demand when the
largest single source of electrical supply is out of service at peak demand times due to
maintenance or any other reason.
2. If more than 2 power generators are installed, those loads considered as process
sensitive should be supplied when the largest capacity power generator is out of
service and the second largest generator is coincidentally shut down due to unforeseen
circumstances.
d. Installation of at least 3 power generators would be required for systems demanding higher
reliability such as an availability greater than 96% in a 12-month period to supply 100% of
the load.
Under some circumstances a (higher) reliability requirement could be relaxed if the
consequences of power supply failure would not be severe. In such circumstances
only two (or even one) main power generators could be considered.
e. If multiple power generators are to be provided a philosophy of supply related to ‘normal
operation’, ‘maintenance shutdown’ and to ‘unplanned outage’ should be developed.
If 3 power generators are to be installed it would be expected that substantial
production could be achieved by a single set remaining available with an unplanned
outage of the second set and the third unavailable ( under maintenance).
f. All power generation facilities, which cannot derive alternative power from another source,
should be arranged with black start facilities. The black start arrangements may involve use
of emergency or on-site diesel generators.

6.4. On-site generation run in parallel with a public utility

a. If on-site generation is selected to be the principal source of power and a connection to a
public utility would be available, consideration should be given to the following options
related to connection to the public utility:
1. As a standby source of electric power
2. A means of export of surplus electrical power
3. A combination of both the above
4. As a black start facility

DO NOT COPY
b. The power generation philosophy should consider these options, which should be
contained in a formal design record.
See also clause 6.2.e if there may be economic benefit in peak lopping of the public
utility supply taking advantage of tariff opportunities.

6.5. Emergency power supply equipment

IEEE Std 446 provides further guidance information for this sub-section.
a. One or more of the following should provide emergency power supplies to essential loads:
1. An alternative source of energy such as batteries (see GP 12-35)

Page 15 of 41
04 November, 2003 GP 12-01
Guidance on Practice for Power System Design

2. Normal on-site power generation equipment additionally rated for the essential loads
with an arrangement for duplicate fuel supply which avoids a single contingency
power outage
The power supply to essential loads should be designed to remain available in the
event of at least two of the installed main power generators being unavailable.
The use of on-site power generation for emergency supply duty would not be possible
without at least 3 power generation sets being available.
3. Alternative supply feeds that provide duplication to avoid a single contingency power
loss
4. On-site diesel generation
Essential loads should receive power from battery supplies or should be associated
with a dedicated emergency generator. If this is impractical, leads to low reliability,
or is prohibitively expensive, the least cost of the remaining alternatives should be
adopted.
b. The probability of zero interruptions to the essential load busbar over a 12-month period
should be 99% or better.
c. If main on-site power generating plant or on-site standby plant is to be selected to provide
power to essential loads, it should be either diesel engine or gas turbine driven.
d. On-site power generation additionally rated for essential loads should have spare capacity
of at least 10% over and above that required to meet the maximum ‘after diversity’ load.
e. If necessary a load shedding scheme should be installed to secure the supply to essential
loads.
Essential loads should not form part of the load shedding arrangements.
f. Local standby or emergency generator prime movers should have their own dedicated fuel
supply.
g. Emergency generator sets should be fitted with a 'black start' capability so they would be
capable of starting and running when no alternative source of electrical power is available
(See GP 12-85).
This may be achieved by compressed air starting with air receivers being capable of
six engine starts from one air charge, or by battery starting with a similar
capability, or by both methods. In general two starting methods would be required.
h. Emergency generators should normally be located in a non-hazardous area (see GP 12-60)
with supply air derived from a clean source. For offshore installations the location of the
emergency generator should be on an outside wall, which should have louvers open to
supply combustion air.

DO NOT COPYi.
Care should be taken to ensure that the exhaust and air intakes are located to avoid
problems associated with other adjacent plant or systems.
Emergency generators should be provided with automatic starting and loading facilities. A
simple and reliable method of ensuring that emergency generators could be immediately
switched to their pre-defined load should be arranged.
Manual facilities should also be provided for regular testing purposes. Testing
facilities should permit the loading of standby generator sets.
j. Emergency generators should be capable of running in parallel with the main supply for
testing purposes and for the purposes of transferring load without the need for supply
break.

Page 16 of 41
04 November, 2003 GP 12-01
Guidance on Practice for Power System Design

7. Power distribution system

The switchgear provided for site or facility power distribution systems should comply with
GP 12-10.

7.1. Primary substation

a. The switchgear for primary substations would normally be medium voltage or high voltage
type.
b. On-site power generator circuits (other than local on-site emergency generators) and public
utility power intakes, should be connected together at a common primary substation, the
busbars of which would be used as the main load distribution centre for a site or facility. If
on-site power generators and public utility power intakes are located at different points
within a site, these should be interconnected.
There should be at least one primary substation for a site, and for larger sites
possibly more than one.
c. Busbar arrangements should be selected to be cost effective, operationally flexible and
safe. The following points should be considered:
1. Operational flexibility to permit loads and power supplies to be effectively connected
under scheduled and unscheduled outages of circuits and busbar sections.
2. Minimal switchgear per circuit and simple control and protection schemes
3. Unscheduled loss of busbar sections should not shut down the system beyond the
level designed and provided for.
4. Scheduled maintenance of busbars should be possible without system shutdowns
beyond those designed and provided for.
It may be necessary to locate the incoming circuit breakers at positions along the
busbar section that ensures maximum busbar loading is kept to a minimum.
The possibility should be considered of busbar and bus-section faults in metal
enclosed switchgear, which may cause a full switchboard shutdown for remedial
repairs. Normal switchboard designs do not cater for this eventuality.
d. Single busbar arrangements for primary supply substations should be provided subject to
the following criteria:
1. The switchgear should be indoor metal clad type.
2. The single busbar should be split into a number of sections by using suitable
switchgear. The number of sections of busbars should be consistent with the
acceptable loss of incoming or outgoing circuits under both scheduled and
unscheduled loss conditions.

DO NOT COPY 3. Routine busbar maintenance should not be considered to be necessary or should be
considered to be an infrequent occurrence.
As an example, for 3 incoming transformer feeders, the busbar should normally be
in 3 parts with feeder circuit breakers distributed between each of the sections of
busbar.
e. Double busbar arrangements should be provided in installations if outdoor air insulated
busbars are installed or regular busbar maintenance is considered necessary for
environmental or pollution reasons. They may also be supplied, if the operational
flexibility available from being able to connect incoming power circuits and outgoing load
circuits in a variety of ways is considered to be particularly advantageous.

Page 17 of 41
04 November, 2003 GP 12-01
Guidance on Practice for Power System Design

Earlier refinery primary substation switchgear designs were based on double

busbar arrangements for indoor metal clad medium voltage switchgear. This design
is very rarely justified for modern refinery and other industrial installations.
The use of double busbar arrangements needs to be carefully considered and
advantages balanced against the additional costs of building structural, switchgear
and protective arrangements.
A generally clear case for a double busbar arrangement would be outdoor
installations where maintenance work on busbars would be greater.
f. Ring-bars or mesh busbar arrangements should be used only under circumstances where
future modification is extremely unlikely.
Ring or mesh arrangements are generally difficult to extend and therefore are
considered to have limited application. This type of arrangement has switchgear
saving advantages if the system is unlikely to require modification at a later date.
g. If the power supply is to be obtained from a public utility via equipment such as
transformers or feeders, the busbars should be sectionalised to prevent total power system
shutdown in the event of unscheduled outage of any busbar section.
Coincidental maintenance of a public utility intake transformer should not be
considered.
h. The maximum number of busbar sections should not normally exceed the number of
individual power sources.
This would avoid over-flexibility in a power system, which, may seem useful but is of
little economic benefit in practice. In some installations if emergency power could
be obtained via the primary substation it may be justified to include a separate bus-
section for essential loads.
i. Outgoing load circuits should be connected to busbar sections in such a way as to minimise
power flow across busbar section switches and should also permit unscheduled and
scheduled busbar section outages with the minimum disturbance to the loads being
supplied.
j. On-site power generators may be connected either directly to the primary power supply
busbar or via generator transformers. The type of connection should be selected depending
upon the economics associated with both the generation voltage and the primary supply
busbar voltage.

7.2. Site power distribution

7.2.1. General
a. The site power distribution network should be designed to carry continuously at least

DO NOT COPY
110% of the ‘after-diversity’ maximum demand associated with peak design production at
the maximum ambient conditions (see clause5.1.f).
The requirement for 110% capacity should be applicable to all components within
the power distribution system including transformer capacity. It would not be
applicable to the capacity of load equipment such as induction motors, which should
be rated as close as possible to actual process demands.
Switchboards should be supplied with a degree of spares and room for expansion.
b. The design of the power system distribution should be based on qualitative assessments of
meeting power supply continuity requirements in the event of maintenance and in the event
of failure (see clause 4.6.d).

Page 18 of 41
04 November, 2003 GP 12-01
Guidance on Practice for Power System Design

Since electrical distribution systems are inherently reliable, the design requirements
are usually based on the need to disconnect distribution electrical equipment to
carry out routine maintenance and at the same time keep power supplies available
for operating plant.
c. The power distribution system design applicable to any project should be described in a
design philosophy document. The physical arrangement of the distribution system should
be designed to suit the supply security and other technical requirements of the project.
The design philosophy document should include the reasoning leading to selection
of the preferred power distribution system arrangement. This reasoning should
contain a cost/benefit analysis.
d. The design features associated with a conventional radial or ring power distribution
arrangement should be suitable for most petrochemical site power distribution systems.
Should an alternative arrangement be necessary, it should be compared with a
‘standard’ radial system in terms of operation, reliability, maintainability, and cost.
The results of such a comparison should be included in the design philosophy
document.
e. Site power distribution system designs should be based on a ‘normally closed’ bus-section
and incomer arrangement unless other factors such as public utility requirements, available
switchgear fault ratings, or existing site practice determine an alternative philosophy.
The advantage of continuity of supply in the event of relatively light fault conditions
which would be provided by closed bus-section systems often outweighs the
disadvantages of increased fault disturbance for heavy fault conditions. This
arrangement has inherent safety features since the switchgear fault ratings would be
based on the worst-case supply configuration.
If new transformers are to be provided, increasing the impedance of incoming
transformers could reduce power system fault levels. Increasing the impedance of
transformers increases the voltage regulation between light and full load leading to
a trade-off between fault duty and normal voltage regulation.
A secondary selective, ‘normally open’ bus-section and ‘normally closed’ incomer
arrangement would be the first choice alternative philosophy.
f. If switchgear fault levels are found to be in excess of readily available fault rated
equipment, consideration should be given to operating with bus-section breakers open as
opposed the use of higher fault rated switchgear.
If an open bus-section breaker philosophy is being considered, the need to rapidly
restore supplies to drives should determine whether an automatic secondary
selective scheme should be employed.

DO NOT COPY
7.2.2. Radial power distribution systems
a. Each component of the radial circuit should be capable of supplying 110% of the required
electrical load. Transformers or other plant, which include forced cooling equipment,
should not rely entirely on the forced cooling arrangements to obtain the necessary rating.
b. If the loading of any equipment is of a cyclical nature and forced cooling would not be
expected to operate for more than 2 hours in any 12 hour period, then capacities based
upon forced cooling could be considered. A cooling system failure alarm should be
provided.
c. If triple fed radial systems are to be used, each feeder should be capable of providing at
least 55% of the maximum demand and all busbars should be split into at least three
sections with two bus-section switches.

Page 19 of 41
04 November, 2003 GP 12-01
Guidance on Practice for Power System Design

This arrangement allows for the loss of any one of the three circuits, leaving the two
healthy circuits still capable of providing 110% of the Maximum Demand.
For transformer fed switchgear, fault ratings should be based on at least 2 of the 3
transformer incomers connected. If the design is to be based on 2 out of 3 in service,
an electrical interlock scheme should be arranged to ensure that fault ratings would
not be exceeded.
d. Essential and process sensitive loads should be supplied by two or more identically rated
radial systems including normally closed bus-section switches, or secondary selective
systems with normally open bus-section switches. Each circuit should be capable of
carrying a 110% of the “after diversity” maximum demand. They should be arranged to
ensure that the unscheduled outage of any component of the circuit would not result in loss
of power supply after the faulty equipment has been disconnected from the system.
e. Exceptions to the requirement for all supplies to be maintained could be made in the
following cases:
1. Bus-section circuit breaker faults These may sometimes cause damage, which may
require complete switchboard shutdown.
The occurrence of such faults is sufficiently rare as to be negligible.
2. If non-essential loads are to be supplied from a double radial system. These may
receive a single radial supply from a double radial system and would not receive a
supply in the event of some forms of double radial system outage.
f. If a process plant could be (or would regularly be) switched off, then single (or simple)
radial systems should satisfy process demands.
Maintenance of electrical systems should be scheduled for normal process plant
shutdown times and the reliability of the power distribution system would generally
be much better than required by the process system.

7.2.3. Ring fed power distribution systems

a. A ring fed power distribution system should include isolating means at every load
substation in order that a faulty section of the primary ring may be disconnected and
repaired without power loss during the whole of the repair period. These ring isolators may
be of the off-circuit or on-load type dependent on availability, cost, and the need for rapid
reconnection of load.
b. The need for appropriate power system protection devices to identify fault location should
be considered in order to assist rapid re-connection of healthy components and avoid
possible re-application of the fault.
c. The decision process to select an open ring fed power distribution system should first
consider the comparative reliability and cost of single radial fed systems with non-

DO NOT COPY
automatic standby power supply backup.

7.2.4. Interconnected or mesh systems

The distribution of electrical power where substations are solidly interconnected with each other
should be undertaken only upon specific design justification, which confirms reliability,
operation and safety, in addition to cost advantage.
Mesh connected schemes for medium voltage power distribution systems would be
permissible providing power system protection arrangements would be suitable for
the process plant needs and providing that the fault ratings of switchgear would not
be compromised.
Solidly interconnected mesh systems for low voltage power distribution systems are
rarely used in industrial applications and their response to a fault condition would

Page 20 of 41
04 November, 2003 GP 12-01
Guidance on Practice for Power System Design

be difficult to predict in terms of power system protection operation and level of

disturbance. For this reason and because they would present a point of non-
standardisation if used in hazardous areas they should be avoided.

7.3. Substation location

a. Site substations should be located in areas, which are not classified as hazardous. The
selected location within the site should, whenever possible:
1. Take account of the ability to distribute power to the on-site loads without
unnecessarily high distribution equipment costs.
2. Be suitable for future expansion.
b. Primary substations should additionally be adjacent to on-site power generation or the
public utility intake.
c. If it is impossible to locate a substation outside of a hazardous area classified as Zone 1 or
Zone 2 (see GP 12-60), provision should be made for forced ventilation of the substation to
permit the use of standard industrial equipment. Air for the ventilation system should be
taken from a remote and safe area.
The substation should be equipped with air lock doors, gas detection and
automatically operated measures for isolating the substation ventilation system from
the outside ambient conditions.
Loss of ventilation and pressurisation should initiate an alarm condition, which
although not causing a trip of the switchgear, should be used as an operator based
time dependent system for the effect of repair.
Complete power shutdown of any equipment not suitable for a Zone 1 hazardous
area (see GP 12-60) should be initiated should repairs not be possible within an
operator defined time scale.
d. In the event of gas detection at the air intake to a substation located in a hazardous area, the
philosophy for automatic action should be determined for the site-specific conditions.
It is usually sufficient to trip any air intake and shut down louvers to isolate the
substation or switch room from the outside and thus maintain power to areas not
affected by the gas release in order to achieve a greater overall safety.
e. For switch rooms located in areas classified as Zone 2, standard industrial equipment may
be used, providing there would be no opening from the building into the classified area or
if there would be an opening, the opening should be assessed to determine that a
flammable atmosphere would not enter the room.
f. Smaller non-essential process switchgear could be located outdoors or equipped with a
shelter and if to be located in an area classified as hazardous, the switchgear equipment

DO NOT COPY
should be selected in accordance with the guidance given in GP 12-60.

7.4. Substation design

a. Substation building designs for onshore application should be of the following:
1. Field fabricated building
The substation roof support should be independent of the walls such that a
switchgear explosion that could damage the walls would not cause the roof supports
to be weakened.
2. Prefabricated power center
b. Emergency lighting should be provided which would allow operation of substation
equipment if a power outage should occur.

Page 21 of 41
04 November, 2003 GP 12-01
Guidance on Practice for Power System Design

c. Personnel exit gates from substation switchyards should be equipped with panic bar
hardware.
d. Personnel exit doors from substation buildings should have panic-bar design in accordance
with IEEE C2.
e. Substation designs for offshore installations should additionally consider project specific
parameters, which could include module construction philosophy, economic layout factors
and safety considerations.
f. All substations should normally be connected to their incoming supplies by cables.
g. Transformers should be located as close as possible to the secondary switchgear.
This design requirement would be to keep the more expensive and high capacity
electrical connections as short as possible.
h. Substation building design should ensure that no possibility exists for ingress of surface
water or for any hydrocarbon, which may migrate along the sheath of cables, which enter
the substation.
There could always be a possibility for hydrocarbon contamination to enter
buildings via rainwater, a rising water table or by finding a way through the sheath
of buried cables and travelling beneath the cable sheath and into the switch room.
i. Raising the elevation of the lowest point of a substation building well above the exterior
grade level, would eliminate surface and sub-surface water ingress.
The provision of a cable basement would be a more effective remedy because there
would be no possibility for hydrocarbon vapour (which may be given off from
contaminated water present in an entry duct) from entering the substation.

8. Power system earthing/grounding

8.1. General
a. Earthing/grounding of equipment and systems shall be in accordance with GP 12-25.
The following clauses relate only to neutral earthing/grounding of power systems.
Further guidance is also available from IEEE Std 142, chapter 1.
b. The earthing/grounding of each part of a power system operating at a specific voltage
should be considered individually. If such systems are required to be earthed/grounded
solidly, or via an impedance, the neutrals should be connected to a common plant
earthing/grounding system. This system should have a resistance to earth/ground that
ensures operation of protective devices in the various circuits, and should not be in excess
of 4 ohms.

DO NOT COPY
The selection of 4 ohms as the power system resistance to earth/ground could vary
depending on the basic power system design and safety requirements. The possibility
of earthing/grounding system resistance variation due to varying soil conditions
through time should be considered.
c. If the site power system earth/ground is to be connected to a public utility supply medium
or high voltage system earth/ground, the combined earth/ground resistance of the public
utility and the site power system earthing/grounding system should be less than 1 Ohm.
The requirement for 1Ohm resistance for a combined public utility system
earth/ground and the interconnected earth/ground for solidly earthed neutrals arises
from the need to minimise neutral potentials under public utility earth/ground fault
conditions.

Page 22 of 41
04 November, 2003 GP 12-01
Guidance on Practice for Power System Design

d. The neutral connections for earthing/grounding equipment should be provided at power

generators, transformers, or both, which have their neutrals brought out, and which are the
power source to the distribution power system. If such power sources are delta connected,
and do not have neutrals brought out, neutral earthing/grounding may be carried out at
other star(wye) connected power transformers on the system or by the provision of
earthing/grounding transformers.
e. Power systems operating at a common voltage, which may be normally or abnormally run
unparalleled, should have a neutral earthing/grounding connection facility provided for
each system. The neutral earthing/grounding system should be designed to ensure that
there would be no possibility of inadvertently operating a system with an isolated neutral.
The condition of a power system in operation without a neutral earth/ground could
be just alarmed under some circumstances. Typical of such circumstances would be
if a single earthed/grounded neutral is required (such as to limit circulating currents
between power generators) and an automatic system for earth/ground switching in
the event of power system disconnection would be complex and difficult.
f. The potential between immediate metalwork and a 'true earth/ground' (which may be
transferred to the immediate locale, typically by telephone services) should be calculated
for all representative single phase to earth/ground conditions.
Touch potentials should be less than 430 V under any single phase-earth/ground
fault if there is overcurrent protection arranged to de-energise the live conductor.
The 430V level relates to telephone system limits of exposure. Fundamental safety to
personnel should be considered separate to this requirement. (See IEEE Std 80 and
IEC 60479-1)
If communications and/or protection cables run outside the earthing/grounding
system to another remote location, the need for voltage isolation measures at all
termination locations should be considered.
Control systems usually demand 'clean' or 'reference' earth/ground points which are
associated with earthing/grounding electrodes not connected to the general site
power system earth/ground system.
This practice could give rise to dangerous situations under medium and high voltage
system faults and consideration should be given to interconnecting the
earthing/grounding points of the control system and power system via a high
impedance choke that includes a lightning surge suppressor device.

8.2. Unearthed/ungrounded (isolated) neutral

a. An unearthed/ungrounded or isolated neutral system should be used only subject to
specific design approval.
Such systems could be used if the highest integrity against earth faults would be

DO NOT COPY required, such as oil well drilling power supplies and d.c. shutdown system power
supplies.
If unearthed/ungrounded power systems are installed, the power system conductors
(phase conductors) could constitute a danger of electrical shock or fire as a result of
contact between them and earth/ground.
The danger would be a result of capacitance coupling of the power system
conductors and earth/ground.
b. If an unearthed/ungrounded power system is permitted, an earth/ground fault detection
system should be installed and there should be operator action to clear any earth/ground
faults that occur as a matter of priority.

Page 23 of 41
04 November, 2003 GP 12-01
Guidance on Practice for Power System Design

8.3. Solidly earthed/grounded neutrals – low voltage power systems

a. All low voltage power systems should have their neutrals solidly connected to the plant
earthing/grounding system.
b. For 3-phase, 4-wire power supply systems, solid earthing/grounding should be provided in
order to keep neutral voltages as close to earth/ground as possible. In this case the neutral
earthing/grounding arrangement should be simplified by arranging for the neutral to be
earthed at the incoming low voltage switchboard.
c. Impedance earthing/grounding of low voltage power systems should only be permitted
subject to specific design approval.
Impedance earthed/grounded low voltage power systems may be considered if the
system would be 3 wire without neutral connections and there would be a benefit in
limiting damage during earth/ground fault conditions.
d. If single-phase supplies are taken from low voltage power supply systems, the neutral
should be solidly earthed/grounded (as per NFPA 70 requirements) at the supply point
only.
This arrangement would correspond to a TN-S system according to IEC 60364-1.
If the low voltage power supply is to be taken from a public utility, the neutral would
be solidly earthed/grounded and the supply system would be defined as TN-C-S
according to IEC 60364-1. In this latter case the installation within the premises
should conform to a TN-S arrangement but with no further neutral earths being
required within the installation.
If PME (Protective Multiple Earth) power supplies are taken from a public utility
there should be no need for isolation transformers to achieve the TN-S arrangement
within the site premises, however, the design should conform with national
legislation and this may require that under certain circumstances, isolation
transformers should be provided.
(Refer to regulations governing fuel retail sites e.g. In the UK advice is offered by
the DTI concerning the design of fuel retail outlets. Historically this advice had been
offered by HS(G)4 H).

8.4. Impedance earthed/grounded neutrals – medium voltage power systems

a. Neutral earthing/grounding equipment to limit earth/ground fault currents should be
provided in the neutral connection to the plant earth/ground system on all medium voltage
power systems.
Such provision should limit the overall transient system disturbance caused by
earth/ground faults and should also limit the amount of damage caused by this most
common type of fault.

DO NOT COPY
b. If the power system at the specific voltage contains no direct connected generators under
any mode of operation, the provision of the earthing/grounding resistor and its
earth/ground connection should be as follows:
1. At the source star (wye) connected transformers
2. At other star (wye) connected power transformers
3. At earthing/grounding transformers
c. Neutral earthing/grounding equipment should comprise a resistor with a minimum 10-
second fault rating, selected to reduce the fault current to the full load rating of the power
source transformer.

Page 24 of 41
04 November, 2003 GP 12-01
Guidance on Practice for Power System Design

d. The use of reactors to earth/ground neutral systems should only be used subject to specific
design approval
Systems employing reactor earths are uncommon but have been used in some
countries or under special circumstances. One special circumstance is where a
'Petersen Coil' is used to avoid a circuit trip due to a transient earth/ground fault
(e.g. lightning induced flashover in overhead lines). Such special circumstances
would be unlikely in petro-chemical installations.
If reactance earthing/grounding is considered, its use should be tested against the
cost/benefit compared with resistance earthing/grounding. Special attention should
be taken of the possible tuning effects between the earthing/grounding reactor and
the power system.
e. If direct connected generators could be operated in parallel with source transformers, the
neutral earthing/grounding arrangements should provide for either system operating
independently.
The neutral earthing/grounding equipment should be identically rated for all power
sources. If resistors are provided, these should comprise 10 second fault rated, solid
resistors.
The resistors should reduce the fault current to one half to full load rating of the
lowest rated power source, provided this fault current is sufficient to operate the
distribution system earthing/grounding protection and provide suitable
discrimination.
If the normal ratings of the source transformer and parallel running power
generators would be significantly different, the resistor rating selection should be
dictated by the requirement to ensure that the most insensitive earth/ground fault
protection on any incoming or outgoing circuit operates positively with the smallest
possible source of earth/ground fault current connected to the system.

8.5. Power generator earthing/grounding

a. The neutrals of power generators directly connected to distribution switchgear in 3-phase,
3-wire medium voltage power systems may be solidly or resistance earthed/grounded.
Solid earthing/grounding should be limited to generators if the earth/ground fault capacity
of the generator is approximately equal to the current level required to operate the
distribution power system protection.
If solid earthing/grounding of the neutral is applied to a system with a (relatively)
low capacity generator in parallel with a larger power source, which has resistance
earthing, the generator should have a neutral circuit breaker, which should be
automatically controlled.
If the influence of an earth/ground fault on other equipment is not unduly

DO NOT COPY
b.
compromised by too high an earth/ground fault current, the generator neutral could
be allowed to remain solidly connected.
Resistance earthing/grounding of generators directly connected to the distribution
switchgear should be used. The resistor should be of the 10 second (minimum) fault rated,
solid type, and its resistance should be selected to reduce the fault current to a level
between the half and full load current rating of the machine, provided this is sufficient to
operate the distribution power system protection system selectively.
c. If power generators directly connected to the distribution switchgear would be operated in
parallel and earthed/grounded either solidly or through resistors, the provision of neutral
switchgear for each power generator should be considered to avoid circulating third
harmonic current flows and their associated heating effects.

Page 25 of 41
04 November, 2003 GP 12-01
Guidance on Practice for Power System Design

The neutral of each power generator should be permanently earthed/grounded via a

resistor to limit circulating currents to acceptable levels.
This type of earthing/grounding would mean that earth/ground fault current levels
on the power system would vary depending upon how many power generators were
operating.
d. Neutral switchgear should be connected and operated in a manner to ensure that only one
power generator neutral switch is closed at a time.
Neutral switchgear could comprise contactors on resistance earthed/grounded
schemes instead of circuit breakers.
The provision of neutral switchgear schemes could be avoided if identical pitch
power generators were paralleled, or if the provision of neutral earthing/grounding
resistors sufficiently reduced the prospective third harmonic currents to acceptable
levels. A 1% or less derating as a result of third harmonic currents would be
considered acceptable.
e. If power generators are to be connected to the distribution power switchgear system via
unit transformers, the power generator neutral should be connected to earth/ground via the
primary winding of a single-phase distribution transformer.
A resistor should shunt the secondary winding of this transformer with a resistance
value calculated to be approximately the same as the zero sequence capacitance for
the generator winding system.
Manufacturers’ standard arrangements for high impedance earthing/grounding
systems for these types of generators should be acceptable. The design should allow
5 to 10 A to flow under earth/ground fault conditions.
Selection of a resistance equal to the system-to-earth/ground capacitance under
earth/ground fault conditions would ensure that the system capacitance would
discharge effectively and over-voltages would be eliminated even for arcing
earth/ground faults.
f. The rated primary voltage of power generator earthing/grounding transformers should be
taken as the generator phase voltage. Transformer ratings should be the product of the
primary current and rated phase-to-phase primary voltage using a 30 second (6 times
overload factor) duty cycle.
Voltages of at least 1.5 times generator line to neutral voltages would be acceptable
to obtain standard transformers.
Conservative sizing of the transformer would be preferred, as its reactance would
have little effect on the earthing/grounding system. The secondary resistor could
have a continuous duty as its size and cost would not normally be significant.

DO NOT COPY
8.6. Earthing/grounding resistors
a. Earthing/grounding resistors should be provided with insulation suitable for the phase-to-
phase voltage of the power systems to which they are connected. They should be designed
to carry their rated fault current for the times specified without any destructive effect to
their component parts.
Resistors could be located in an outdoor environment for transformer
earthing/grounding, but could be indoors for power generator earthing/grounding
b. Grid type resistors offer a virtually maintenance free installation and should be employed
on all medium voltage power systems 20kV and below, where resistance grounding would
be required.

Page 26 of 41
04 November, 2003 GP 12-01
Guidance on Practice for Power System Design

This type of earthing/grounding resistor could prove uneconomic for medium

voltage power systems of 20 kV and above liquid neutral earthing/grounding
resistors could offer an optimum choice.

9. Power system operation and control

9.1. General
a. This section should be read in conjunction with GP 12-03.
b. The operational control of a power system should be described in a control philosophy
document. The control philosophy document should clearly define which functions are to
be controlled by automatic means and/or by manual means. The information to be provided
for each of the control functions should also be defined.
The control functions and information dependencies for power systems, which
include on-site power generation operating in parallel with a public supply network
should be carefully defined.
c. The power system should utilise microprocessor-based, multi-function protection relays,
which incorporate means for indication and control, and which could be integrated into an
overall control system. The required form and performance of the multi-function protection
relay system should include for the following:
1. Speed of response of the control functions, including any required time delays to
avoid “hunting”
2. Degree of programmable automation of power system operation, including required
control algorithms
3. Range of protection functions available
4. System architecture
5. Central utilities control room operator information (control screens)
6. Alarm and information conditions
7. Interface arrangements with process control systems

9.2. Power generator operation and control

a. This sub-section should be read additionally in conjunction with GP 12-80.
b. For power systems having multiple on-site power generators, the manual or automatic
control system should control the following functions via a central utilities control room:
1. Stop/Start
2. Synchronising

DO NOT COPY 3.
4.
5.
Generator active power loading (speed control)
Generator excitation (AVR setting) and reactive power sharing
AVR and governor droop settings.
6. The taps on transformer connections with public utility network
If of the on-load tap changer type
c. The central utilities control room operator should also have control of:
1. Primary substation circuit breakers
2. Emergency generator stop/start/synchronisation (if significant)

Page 27 of 41
04 November, 2003 GP 12-01
Guidance on Practice for Power System Design

3. Some downstream distribution system circuit breakers

The central control room operator should have information and means of control of
the entire site power system. This ability to control may be limited in respect of
process loads.

9.3. Synchronising
a. Synchronising and/or check synchronising equipment should be provided if more than one
source of power may be operated in parallel with another. Such synchronising/check
synchronising provisions should be provided for every switch, which has the possibility of
connecting unsynchronised supplies.
A synchronising/check synchronising philosophy (or design document) should be
provided for every installation, which would define the particular arrangements.
b. Manual synchronising arrangements should always form part of the automatic
synchronising provisions for incoming on-site power generator circuit breakers.
This should comprise voltmeters and a synchroscope to show the voltage and
frequency differences between the two systems that need to be paralleled, together
with means of adjusting voltage and frequency of one or other of the sources.
c. A check-synchronising relay should be provided to prevent operator mal-operation. In
order to allow closing a power source onto a dead system under black start conditions, the
check synchronising relay should have a means of manual or automatic override.
d. Check-synchronising facilities should be fitted to bus-section circuit breakers so that it will
be possible to run two systems feeding either section of a busbar completely segregated
from the other. The number of circuit breakers provided with check-synchronising
facilities should be kept to a minimum. A similar logic should be applied to public utility
intake circuits.
As an alternative, circuit breaker interlocking schemes should be installed to
preclude the possibility of paralleling two sources of power if check-synchronising
facilities are excluded.
There should be no need to include check-synchronising facilities on system voltage
levels more than two levels removed from the power generation busbar if the power
system is operated with bus-sections normally closed.
If the system would be operated with open bus-sections for significant periods and
there would be limited central system control or information dissemination, the
check-synchronising arrangement should be reviewed.
e. Synchronising facilities should be provided at the primary power supply voltage levels and
should be avoided at other voltages by use of appropriate circuit breaker interlocking.

DO NOT COPY
The form of interlocking should attempt to avoid the need for break-before-make
action unless absolutely necessary.

9.4. Load transfer schemes

a. Within switchboards employing an open bus-section or incomer arrangement, automatic
beak-before-make (secondary selective) transfer schemes should be provided.
b. No automatic transfer schemes should ever parallel normal and emergency sources of
power.
c. Load transfer schemes may use circuit breakers, or on-load transfer switches/contactors.
d. If automatic transfer schemes are provided, the power system design should be arranged so
that successful transfers could be achieved without causing a general system failure.

Page 28 of 41
04 November, 2003 GP 12-01
Guidance on Practice for Power System Design

The speed of transfer required will determine the cost of the automatic transfer
scheme, and very high-speed schemes could reduce supply interruption times to less
than 100 milliseconds. This could be sufficient to prevent contactor dropout, thereby
enabling induction motors to remain connected.
Economic justification for very high-speed transfer schemes would generally be on
the basis of the cost of lost production and power supply restoration following
longer-term supply interruptions.
e. Other methods of ensuring duplication of supply could include:
1. Feeding alternative loads from different switchboards
2. Providing a parallel operating arrangement of the incomers
f. A make-before-break transfer operation should be considered if the power system is
normally run in synchronism.
This operation should automatically disconnect the transferring (from) circuit on
successful energisation of the transferring (to) circuit, to minimise the duration of
parallel connection.
g. Slow-speed load transfer schemes could be applied to either static loads or induction motor
loads or a combination of the two. They should not be used if synchronous motor loads are
supplied.
The load transfer should be arranged so that the residual voltage of induction
motors would have decayed to less than 25% of the rated source voltage before the
transfer was initiated. The rate of residual voltage decay should be calculated and
the motor's ability to withstand out-of-phase reclosure should be checked.
In general 1 second should be adequate for all low voltage machines and
1.8 seconds should be adequate for medium voltage machines up to 1000 kW. (See
GP 12-20)
h. Transfer schemes associated with induction motors, which would be controlled by circuit
breakers, or contactors of the mechanically latched type, should be designed either to be
capable of re-accelerating the motors if the transfer takes place within the motor under-
voltage protection relay tripping time, or time delaying the transfer to be in excess of the
motor under-voltage protection relay tripping time.
Induction motors, which are controlled by circuit breakers, or contactors of the
mechanically latched type, should include a time delay under-voltage protection
relay. This protection relay should be set to trip the motor controller for power
system voltage dip conditions if recovery would not be possible.
The selection of under voltage trip time should be confirmed by power system
studies, which should be aimed at ensuring reasonable power system recovery

DO NOT COPY
following a fault. In general a setting of, say 2 seconds if the voltage has fallen
below 75% may be appropriate.
i. Induction motors, which are controlled by unlatched a.c. contactors, will inherently
disconnect from the supply on loss of voltage. If it is required to restore power to these
types of motor drives the auto-transfer schemes should be supplemented by contactor
control schemes which restart motors individually or in groups after requisite time delays
determined by process priorities.
j. Load transfer schemes for the start up, run up and loading of standby generators on to a
busbar normally fed from a preferred a.c. source should be initiated by time delayed under
voltage-relaying which should trip the normal a.c. source and auto-start the standby
generator.

Page 29 of 41
04 November, 2003 GP 12-01
Guidance on Practice for Power System Design

The auto start of the emergency (or standby) generator could be initiated before
disconnection of the normal supply in order to provide the main power system
greater chance of recovery (this would take advantage of the finite time it would
take for a standby generator to start).
Emergency generation start initiation would normally be based on main power
system voltage collapse below 75% of nominal for, say 2 seconds and that trip of the
main supply and energisation of the emergency supply would then be initiated.
Energisation would need to be time delayed from the trip signal to allow flux decay
of induction motors, which may have been in operation.
k. Power system re-acceleration and restart studies to determine the most technically
acceptable and cost effective solution should be carried out for each load transfer scheme
considered.
The studies should provide confirmation of under-voltage relay settings.
Consultation with process engineers would be necessary to ensure optimum motor
re-acceleration settings

10. Power system performance

10.1. Motor starting

a. Direct-on-line (DOL) starting of motors should be the norm.
b. The maximum voltage drop at the motor terminals during starting should be limited to 20%
calculated for the minimum generation and system configuration, representing the
minimum duty condition.
The power system should be capable of withstanding this 20% voltage depression.
The motor starting voltage dip acceptability would be contingent on the motor
satisfactorily accelerating, and the voltage dip not causing unacceptable
disturbances to the power system.
c. If connection is made to a public utility, nominal system voltage dips at the public utility
point of common coupling should be kept to a value, which has been agreed between the
site and the public utility.
Public utilities generally require that voltage depressions at the point of common
coupling should be no more than 3% for infrequently started machines (for example
no more frequent than once in 2 to 6 hours) and no more than 1% for frequently
started machines.
d. The DOL starting of the largest and electrically most remote motors (or groups of motors
on an automatic re-acceleration scheme) at all voltages should be investigated to ensure
that no problem exists. If unacceptable motor starting voltage dips could be anticipated, the

DO NOT COPY following solutions should be considered:

1.

2.
Time vary the starting time to be assured of less co-incident starting between
machines
Obtain motor(s) with a lower starting current
3. Increase the fault level within the switchgear rating limits by decreasing system
source impedance to the motor busbars
4. Consider assisted start methods to reduce voltage dips if this solution is limited to a
few specific motor drives
5. Any combination of the items above

Page 30 of 41
04 November, 2003 GP 12-01
Guidance on Practice for Power System Design

6. Increase the system fault level to higher level than normal (see clause 4.3) if higher
fault rated switchgear is available.
The selection of the most appropriate method would normally be made on the basis
of lowest cost and under some circumstances may require provision of a mechanical
driver other than electricity for the largest machine envisaged.
Reduced voltage starting of a few specific motors should be considered only if it is
found to be economical and the additional complication associated with assisted
start equipment is operationally acceptable.
e. Motor load shedding and restart schemes should be developed on the basis of cost
effectiveness for the project under consideration.
If the load is predominantly induction motors, voltage reduction would not
necessarily cause load shedding whereas reduced system frequency could be an
effective load shedding method.
System load shedding would best be initiated either by a frequency-based
arrangement or by an arrangement based on a pre-determined logic scheme.
Power generation deficiency could be detected if total generation capacity was
known together with total system load. Under these conditions immediate load
shedding could be initiated.
Such immediate load/generation matching schemes vary from a simple generator
auxiliary contact used to initiate fixed load shedding to a complex microprocessor-
based system monitoring multiple generators and multiple loads with constant
updates for the amount of load shedding for any specific incident.
The effect that electrical load shedding will have on the process system would need
to be discussed with process engineers.

10.2. Over-voltages

10.2.1. General
IEEE Std 141, chapter 6 gives additional guidance on the subject of voltage surges
from lightning and switching sources.
a. Over-voltages due to static charging should be avoided by effectively earthing/grounding
the power system and all metallic structures that may, or may not, contain electrical
conductors. (See GP 12-25)
b. Using metal enclosures and barriers between the two systems should guard against
physical contact between higher and lower voltage systems.
c. Over-voltages between transformer windings should be limited by effectively
earthing/grounding the neutrals of the secondary voltage power system. Autotransformers

DO NOT COPY
d.
if used, should have the neutral solidly earthed/grounded.
An autotransformer would effectively connect the neutral earthing/grounding
systems of both the primary and secondary power systems.
Over-voltages caused by resonant effects should be investigated for all power systems
operated with an unearthed/ungrounded neutral, including those that may be
earthed/grounded by reactors, or systems with high impedance earthing/grounding.
These effects should also be considered when power system equipment is or could be
operated without a neutral earth/grounded connection for short periods of time.
Resonant inductive-capacitive over-voltages occur when normally
unearthed/ungrounded power systems experience earth/ground faults. Resonant or
ferro-resonance effects could be avoided by solid earthing/grounding the neutrals of

Page 31 of 41
04 November, 2003 GP 12-01
Guidance on Practice for Power System Design

low voltage power systems, and resistance earthing/grounding the neutrals of

medium voltage power systems.
Intermittent earth/ground faults on unearthed systems may cause over-voltages of
the order of five or six times system voltages. Neutral earthing/grounding to allow
for earth/ground fault currents greater than line to earth/ground charging currents
could be used to eliminate these prospective over-voltages.

10.2.2. Switching surges

a. Switchgear and power system equipment should be selected to ensure that transient
recovery voltages produced by switchgear arc extinctions do not exceed the insulation
capability of the system.
Current zero arc extinctions commonly result in transient over-voltages when
switchgear is opened under fault conditions and the healthy side of the system
endeavours to return to the normal system voltage but overshoots. The introduction
of resistance during fault current flow either by special means or by using
switchgear with naturally high arc resistances should be considered to reduce the
value of transient recovery voltage that is generated.
b. Vacuum switches should be designed to limit over-voltages to acceptable withstand levels
for a particular voltage rating.
The production of high over-voltages due to the forcing of a current zero can occur
with fuses and vacuum switches. Current limiting fuses should only be used within
their voltage class.
The provision of surge limiting equipment on vacuum break devices may be required
to limit the over-voltage.

10.2.3. Lightning surges

a. The highest over-voltage, which power systems could be subjected would be those caused
by lightning.
b. Problems caused by lightning strikes on outdoor electrical equipment could be limited by
ensuring all such equipment is earthed/grounded.
c. Lightning over-voltages could reach site equipment by travelling down overhead lines.
Sites, which are fed by a public utility via overhead lines, should be protected against
travelling surges and the methods for protection would be subject to agreement between
the operating site and the public utility.
The public utility should normally provide surge diversion and/or arcing horns on
the equipment connection at the overhead line, such as on a cable connection to the
overhead line or on the primary terminals of a transformer.

DO NOT COPY
d. Overhead lines within a site should have suitable surge diversion equipment provided for
all of the equipment interfaces.
e. Over voltage surge protection should be provided for power generators if these are
interconnected with an overhead line system by short lengths of cable.
f. The switchgear connecting power generators to the power system should be assessed for
switching over-voltage generation and, if necessary, surge diverters should be provided for
the generator stator winding.

10.3. Harmonics
a. IEEE Std 141, chapter 9 gives additional guidance on power system harmonics.

Page 32 of 41
04 November, 2003 GP 12-01
Guidance on Practice for Power System Design

b. For public utility interconnections, the public utility regulations for harmonic loading
should be adhered to.
Harmonics result in power losses and overheating of rotating electrical machinery,
interference on communication and control circuits, overloading of capacitor banks,
and mal-operation of electronic equipment.
c. The total harmonic distortion (THD) of the power system voltage waveform for a medium
voltage (and below) power system should be within the tolerance ± 5% THD as defined in
IEEE Std 519 or alternatively, IEC 60034-1 which specifies telephone harmonic factor
(THF) for synchronous generators.
This level of distortion is compatible with hazardous area certification requirements
and should be adhered to when supplying electrical equipment in hazardous areas.
IEC 60034-1 details irregularities of waveform and although primarily concerned
with synchronous generators, the distortion would be imposed on any equipment
supplied from the generator's output.
Hazardous area equipment employing magnetic effects would be the most sensitive
to THD. These would include electric motors, relays and any equipment with
transformers.
d. If more than 5% THD is anticipated the power system equipment receiving supply should
be rated for the specific harmonic content of the voltage waveform.
A THD voltage waveform distortion greater than 5% may be permitted for supplying
hazardous area equipment provided that the equipment has been suitably certified
or, if waveform irregularities are not significantly different from this tolerance,
assessed and confirmed by the manufacturer as satisfactory for the certification for
the duty.
An example would be a variable speed drive used for a motor with a hazardous area
duty. In this case the motor certification should include for the specific harmonics of
the VSD.
e. If more than 5% THD is anticipated and the manufacturer of the electrical equipment
cannot confirm its suitability for hazardous area duty, the following should be considered:
1. Re-specify the equipment for Ex(p) or pressurised.
2. Revise the electrical system to eliminate the harmonic problem by:
a) Power system supply configuration modifications to the circuits providing
supply to the non-linear equipment causing the problem.
b) Provision of filter equipment at a convenient point on the power system, which
would protect the supply to the hazardous area equipment.

DO NOT COPY
In some cases the voltage waveform interference could be a transient effect, (such as
in the case of drilling activities offshore). If the transient nature of the problem
could be confidently defined, this could ease the equipment manufacturers concerns
over possible adverse harmonic heating effects.
f. Normal operation of power system protection, control, or other equipment should be
caused by harmonic distortion of the power system voltage waveform.
Multiple zero crossings of the voltage waveform could lead to mal-operation of
equipment receiving the power system supply, such as equipment which uses the
mains supply for information transfer or for timing operations.
g. If significant amounts of rectification or inversion equipment are to be used, consideration
should be given to the use of phase-shifted transformers or harmonic filters.

Page 33 of 41
04 November, 2003 GP 12-01
Guidance on Practice for Power System Design

h. The presence of harmonics in common earthing/grounding systems should be assessed and

limited if they are considered to cause prospective hazards.
i. Transformer inrush harmonic current caused by energisation should be considered in the
design of the primary electrical protection system.

10.4. Power factor

IEEE Std 141, chapter 8 gives additional guidance on power factor related issues.
a. Power factor improving equipment should be provided if energy is imported from a public
utility, which applies a tariff associated with low power factor energy provision.
b. Power factor improving equipment may be in the form of capacitors or synchronous
motors, depending on economics and suitability over the range of known operating
condition.
An economic assessment should be made to determine the merit of installing power
factor improvement equipment. The economic case for installing such equipment
would normally be considered as proved if the capital expenditure for the equipment
could be recovered from reduced energy payments within 2 years of commencement
of operation.
c. If the public utility system would normally be operated in parallel with on-site power
generation, the generating equipment should be designed and operated to supply as much
of the reactive load of the site as would be feasible.
The economic case for power factor correction equipment should then be made
based on the assessed availability of the on-site power generation.
If the on-site power generation could supply the site reactive demand, there would
be no need for power factor improving equipment to be installed for the normal
parallel operating mode and its use should only be considered for standby
(unparalleled) operation.
d. The location of power factor correction capacitor banks should be carefully considered if
voltage support and reduction of losses would be the motivators for their use.
A comparison of alternative methods of achieving the same objectives would be
expected as part of the justification for the installation.
e. Plant extension projects should compare the cost of providing power factor improvement
equipment against that of increasing the distribution power system capacity if existing
plant power factors are known to be below 0.8 lagging.
f. For new installations, there should be an economic investigation into the provision of
power factor improving equipment, as a means of reducing distribution power system
capacity.

DO NOT COPYg.

h.
Synchronous motors or capacitors should be located as near as possible to the loads, if the
case for installing power factor improving is proven. If synchronous motors are supplied
for power factor improvement, they should include constant power factor control
equipment.
Synchronous motors should include provision for change from operating in a reactive
power control mode to operating in a constant power factor mode.
This would be necessary should the motor be a small part of the overall installation
demand, or if operation in reactive control mode could make the motor operate for
long periods near an excitation condition which may give rise to stability problems.
i. Induction motors should not be switched as a unit with any power factor improving
capacitors to avoid risks of over-voltages or high transient torque, unless the capacitive

Page 34 of 41
04 November, 2003 GP 12-01
Guidance on Practice for Power System Design

current at full voltage is less than the no-load magnetising current of the associated
induction motor.
Capacitors installed on the motor side of motor control switchgear could act as an
excitation source should the motor be tripped and coasting. Two problems may
result:
• The motor terminal voltage may exceed insulation capability
• The terminal voltage could remain high for a long period and thus compromise
reclosure
If capacitance on the motor side of the switchgear is excessive, either the motor
insulation should be able to withstand high over-voltages (not above 150% as the
induction motor iron circuit could be expected to show some signs of saturation), or
the capacitor needs to be separately switched.
j. Over-voltage protection should be considered for induction motor control circuits if
significant power factor correction capacitors are to be installed on a power system.
If capacitors for power factor improvement are applied to power systems with
nonlinear loads that inject harmonic currents, the capacitors may lower the
resonant frequency of the power system enough to create a resonant condition with
the harmonic currents. Theses currents may be high enough to blow any capacitor
protection fuses fitted, or to damage the capacitors themselves. Potential solutions
to these resonance effects could be:
• Detune the power system by changing the point where the capacitors are
connected
• Change the amount of applied capacitance
• Install specially designed filter reactors

11. Power system studies

11.1. General
a. Power system designs of new installations and extensions to existing plants or facilities
should be analysed, preferably by computer analysis using proprietary software packages
(EDSA, IPSA, SKM Power Tools, etc).
b. A document should be prepared which details the basic design performance requirements
of the power system.
The document could also include or reference the required power system protection
relay systems and settings.
Switchgear load and fault ratings and voltage limits would generally be readily

DO NOT COPYc.
assessed for acceptability.
Power system dynamic performance could be project specific and acceptance would
depend on the economic assessment of 'would it be worth the cost of any
improvement?’
Power systems analysis should be used:
1. To define equipment parameters before purchase
2. To select power system control arrangements and protective relay settings
3. To determine the power system response to normal and abnormal operating
conditions

Page 35 of 41
04 November, 2003 GP 12-01
Guidance on Practice for Power System Design

d. If power system instabilities are predicted by these studies, the following corrective
measures should be considered:
1. Reduce the severity of the disturbance, possibly by decreasing the fault severity
2. Load shedding (with or without load re-acceleration)
3. Detection of the condition and sectionalising the power system into 'islanded ' units
e. Power system behaviour and performance should be studied for steady state, transient
stability following fault disturbance, and voltage recovery following fault disturbance.
Induction motor stability should be included in these studies and the starting
performance of the most critical drives (and groups of drives) should be examined to
ensure that the system should perform satisfactorily under the defined conditions.
f. Power system studies should be carried out within clearly defined terms of reference as
early as possible within a project. The power system study software packages to be used
should be defined at this stage.
The software models provided for power generators, automatic voltage regulators,
governors, induction and synchronous motors, transformers, cables and loads
should be sufficiently detailed and proven to give confidence in the results of the
studies.
Preliminary study work should be carried out using proprietary software packages
to set up a permanent power system model data file, which should be suitable for
detailed design, and could be modified as necessary to include the detailed design
equipment parameters. This data file would then be available throughout the life of
the installation for operational use and could be modified as changes occur.
This process would be considerably eased if the power system analysis software
were the same for preliminary studies through to final studies.
g. The steady state conditions prior to a particular fault condition should be clearly defined
and so should the fault (or onerous) condition, which the system would be required to
experience and recover from. These conditions could include:
1. Minimum power generation, one line (or transformer) out of service for maintenance
and starting of the largest induction motor (presumed direct-on-line)
2. Maximum load, minimum power generation spinning reserve, and a 3-phase
symmetrical short-circuit condition, which removes the largest capacity power
generator from the power system
3. Simple loss of the largest capacity generator from the power system at times of
minimum spinning spare
This may be a less severe condition than if the generator were faulted but could

DO NOT COPY
illustrate the level of load shedding which may occur.
4. Maximum loading and power transfer between two power systems followed by a
3-phase short-circuit which removes one of a number of links between power
generation sources (this may cause transient instability)

11.2. Normal power system studies

a. Normal power system studies include:
1. Load flow analysis to check voltage profiles and circuit loading conditions under
steady state conditions
2. Short-circuit studies to analyse fault currents that might flow under a variety of
symmetrical, asymmetrical and unbalanced short-circuit conditions

Page 36 of 41
04 November, 2003 GP 12-01
Guidance on Practice for Power System Design

These would be used for switchgear specification and power system control and
protective relay application and setting purposes.
3. Stability studies to analyse the transient and dynamic performance of power systems
after large load changes and fault disturbances
These would be used to check:
• The ability of the system to stay in synchronism
• Induction motor stability after start
• Re-acceleration and restart schemes
• The need and effectiveness of under frequency load shedding schemes
They would also be used to consider the merits of:
• Auto-changeover (load transfer) schemes
• Parallel or open operation radial feeders
• Operation of fault limiting devices
• Insertion of switched reactors or capacitors, etc
Motor starting, power system dynamic performance (including load shedding
provision and protection relay arrangements) and short-circuit requirements would
be subject to the specific design requirements of the project.

11.3. Short circuit studies

a. The fault currents that would flow as a result of short-circuits should be calculated at each
power system voltage level for both three-phase and phase to earth/ground fault conditions.
The a.c. symmetrical and d.c. asymmetrical components of short-circuit currents
should be calculated.
b. These calculated currents should be used to select suitably rated switchgear and to allow
the selection and setting of protection devices to ensure that successful discriminatory fault
clearance could be achieved.
c. Short-circuit studies should normally be based on proprietary software packages. The
contribution of induction motors should be included in the study, preferably by direct
dynamic modeling, and the studies should include short-circuit break and make duties.
d. Appropriate manufacturing tolerances for equipment should be considered for short-circuit
studies carried out at the design stage plus allowance for operations at above nominal
system voltage. A design margin should be allowed for later additions.
The allowance should be arranged by undertaking the studies showing later
additional loads on stream and representation of any possible future induction

DO NOT COPY
motors.
e. Short-circuit calculations should be made to determine the peak make short-circuit current,
which occurs 10 ms after fault initiation for 50 Hz power system (8.3 ms after fault
initiation for 60 Hz power systems), as well as the rms breaking short-circuit current
occurring at the instant of circuit breaker interruption (typically five cycles after fault
initiation).
f. Short-circuit calculations should be based on IEEE Std 399, IEC 60909-0, and/or
equivalent national standards, guidelines or recommendations.
The calculation of short-circuit currents should include the fault current
contributions from power generators, synchronous, and induction motors.
g. Public utility short-circuit in-feed information should be obtained if a site or facility would
have a public utility connection.

Page 37 of 41
04 November, 2003 GP 12-01
Guidance on Practice for Power System Design

This information should include any planned future capacity increases and present
maximum and minimum short-circuit values consistent with annual load cycles and
normal public utility switching conditions.
The short-circuit levels of public utility networks could be subject to variation due to
the amount of power generation plant which they may have connected and also
could be subject to variation due to the manner in which the public utility operates
the system (line outages would affect short-circuit levels as would open busbar
systems).
h. Three-phase balanced short-circuit current calculations should be carried out to determine
prospective circuit breaker duties and should include:
1. Peak make short-circuit current capacity
Both a.c. and d.c. current decrements should be included for the half cycle.
2. Rms asymmetric breaking short-circuit current capability
Calculated at a time at which the breaker contacts would be expected to part with an
allowance of 10 ms for instantaneous type protection operation. Both a.c. and d.c.
current decrements should be included for the selected time.
3. Rms symmetrical breaking short-circuit capability
There should be nil d.c. current component but should include a.c. current decrement
for the selected time.
i. If earth/ground fault currents would be limited by neutral earthing/grounding equipment,
the earth/ground fault currents should be assumed to include no decrement and should be
considered constant whatever the level of bonding between the conductor and the faulted
phase.
j. Both the a.c. and d.c. components of motor fault current contributions should be calculated
and included in calculation of prospective short-circuit currents.
At the instant of short-circuit inception the a.c. peak symmetrical component and the
d.c. component should be taken to be identical. Both values should be taken as the
peak direct-on-line starting current, this being dictated by the motor locked rotor
reactance.
Both currents should be assumed to decay exponentially with time using a.c. and d.c.
short-circuit time constants respectively.
The a.c. time constant should be calculated by using the ratio of the locked rotor
reactance and the standstill rotor resistance. The d.c. time constant should be
calculated by using the locked rotor reactance and the stator resistance.
If short-circuits are not directly on motor terminals, these time constants should be

DO NOT COPY
modified (preferably by the power system analysis software package) to take
account of external impedances to the point of fault.
k. The calculation of individual fault current contributions should be carried out for
individual motors of significant ratings on the power system. All other motors on the
system may be treated as a number of typical equivalent motors of total rating equal to the
connected rotating loads, at different locations. The ratings of these equivalent motors
should be selected to be consistent with the actual drives at a given location.
Generally motors with ratings 1000 kW or greater should be represented as
individual machines. However, if there are multiples of these on a single busbar,
they too may be represented by lumped parameters.

Page 38 of 41
04 November, 2003 GP 12-01
Guidance on Practice for Power System Design

11.4. Stability studies

a. Stability studies should be used as support for a particular power system design and also to
ensure that the protection relay arrangements would not compromise the expected power
system performance.
b. The primary objective of transient stability studies is to identify maximum acceptable fault
clearance times.
Secondary objectives would include identification of the best location of power
system open bus-section points, and the relationship between impedance
earthing/grounding to stability.
c. The transient voltage disturbances sustained during short-circuit faults and after fault
clearance should also be studied to ensure that transient disturbances would not result in
loss of supplies due to low voltages or over stressing of equipment insulation due to high
voltages.
d. Transient stability studies should be carried out on power systems to determine whether
synchronous machines would be likely to lose synchronism after the most severe single
disturbance. These studies should be carried out for power systems which include:
1. Dissimilar on-site power generators
2. On-site power generators operating in parallel with a public utility
3. Synchronous motors
4. Power generation busbars interconnected by appreciable impedance
The most severe fault condition would generally be a three-phase short-circuit
applied at the generator busbars with a fault duration determined by the protecting
switchgear, and when cleared results in the disconnection of the largest single fault
contributor from the system. A number of fault locations followed by plant
disconnections should be studied.
These studies should be undertaken assuming the steady state operating condition
before the fault is applied would be one in which the power generation spinning
reserve is at a minimum due to assumed maintenance of the largest on-site power
generator.
e. Dynamic and induction motor stability studies should be carried out to investigate the
voltage and frequency performance of a power system after a major disturbance for the
period from fault inception to the time when steady state equilibrium is reached.
These studies would require detailed automatic voltage regulator (AVR) and
governor modelling for power generators since these items would assist the return
to steady state and would react positively in the time scales likely to be considered.

DO NOT COPY
f. Stability studies should be carried out to consider the effect of the loss of the largest power
supply component under a fault condition, which causes no other electrical disturbance.
Earth/ground faults or mechanical system trip conditions would not normally result
in motor loads being tripped by a.c. contactors dropping off under low-voltage, and
could result in the greatest post-fault power generation deficiency.
If the transient frequency excursion is predicted to exceed 6%, under-frequency load
shedding schemes should be considered.
These studies should also be used to define the minimum number and magnitude of
the various stages of load shedding that would be necessary to keep the frequency
loss within acceptable limits.
g. Induction motor stability studies should be carried out to demonstrate ability to start, re-
accelerate or restart motor loads without stalling or tripping under overload.

Page 39 of 41
04 November, 2003 GP 12-01
Guidance on Practice for Power System Design

h. Induction motor stability studies for motor restart schemes should be used to define the
maximum number and magnitude of the various stages of restart that would be possible
after clearance of faults.
Re-acceleration studies should determine whether motors re-accelerate after
disturbances (fault conditions or under-voltage conditions) have cleared.

11.5. Harmonic analysis studies

a. Harmonic analyses should be based on IEEE Std 519 and/or equivalent national standards,
guidelines or recommendations.
b. Harmonic analyses should be performed on power systems if the connected loads at any
particular power system voltage level would include significant (>20%) non-linear loads,
or for sites of facilities with concern about harmonic levels being excessive.
Non-linear loads would include conversion equipment such as static power
converters for adjustable speed drives, or power supplies for electronic equipment
such as UPS.
These studies should be repeated whenever a significant upgrade to a power system
could include non-linear loads.
c. Harmonic analysis studies could be based on frequency domain methods, which would
involve non-linear load manufacturers providing a Fourier series for the harmonics in the
load current assuming a sine wave voltage input.
The amount of voltage waveform disturbance at any point in the power system
network could be estimated using this manufacturers’ data
If the voltage waveform contains harmonics, then the non-linear load equipment
would actually yield a differing set of load current harmonics.
The frequency domain analysis is at best an approximation and needs to be
compared with actual plant performance to be assured that the system predictions
would be accurate.
d. An alternative harmonic analysis technique would be to use a time domain approach where
system parameters are input into the analysis.
This technique could still have inaccuracies caused by the approximations used for
power system component impedances to each harmonic frequency and a check with
actual system performance should still be carried out.

12. Power system protection and control

Power system protection and control design and installation should be in accordance with GP 12-03.

DO NOT COPY

Page 40 of 41
04 November, 2003 GP 12-01
Guidance on Practice for Power System Design

Bibliography

[1] IEC 60034-1, Rotating electrical machines – Part 1: Rating and performance

[2] IEC 60364-1, Electrical installations of buildings – Part 1: Fundamental principles, assessments of
general characteristics, definitions

[3] IEC 60479-1, The effects of current on human beings and livestock – Part 1: General aspects

[4] IEEE Std 80, Guide for Safety in AC Substation Grounding

[5] IEEE Std 141, Recommended Practice for Electrical Power Distribution for Industrial Plants (IEEE
Red Book)

[6] IEEE Std 142, Recommended Practice for Grounding of Industrial and Commercial Power Systems
(IEEE Green Book)

[7] IEEE Std 446, Recommended Practice for Emergency and Standby Power Systems for Industrial
and Commercial Applications (IEEE Orange Book)

[8] NFPA 70, National Electrical Code (NEC)

[9] NFPA 70E, Standard for Electrical Safety Requirements for Employee Workplaces

DO NOT COPY

Page 41 of 41