Practical Solution Guide To Arc Flash Hazards-XS
Practical Solution Guide To Arc Flash Hazards-XS
Practical Solution Guide To Arc Flash Hazards-XS
to
Arc Flash Hazards
Disclaimer
Warning - Disclaimer: The calculation methods listed in the book are based
on theoretical equations derived from measured test results. The test
results are a function of specific humidity, barometric pressure,
temperature, arc distance, and many other variables. These parameters
may not be the same in your facility or application. The results calculated
from these equations may not produce conservative results when applied to
your facility. PPE recommended by any calculation method will NOT
provide complete protection for all arc hazards. Injury can be expected
when wearing recommended PPE. The results should be applied only by
engineers experienced in the application of arc flash hazards. The authors
make no warranty concerning the accuracy of these results as applied to
real world scenarios.
Using the methods in NFPA 70E or IEEE Std-1584 does not insure that a
worker will not be injured by burns from an arc-flash. Following the NFPA
70E and IEEE 1584 procedures and wearing the proper protective
equipment will greatly reduce the possibility of burns. Using the incident
energy equations developed from the arc flash tests, it is expected that the
personal protective equipment (PPE) classification per the tables in NFPA
70E will be adequate for 95% of the classifications based on test results.
Forward
ESA is pleased to bring you the Practical Solution Guide to Arc Flash Hazards
version 1.0. We believe this will be a valuable tool for electrical engineers, safety
managers, or anyone responsible for implementing and maintaining an arc flash hazard
safety program.
The guide was designed to walk you through the necessary steps of implementing an arc
flash assessment as part of your overall safety program requirements. It will help you
and your team make important decisions concerning the safety of your employees and
how to manage the complex tasks of OSHA and NFPA-70E compliance for arc flash
hazards.
Arc flash hazard analysis and safety program development to protect against arc flash
hazards is in its infancy. Research into the arcing phenomena is ongoing as industry tries
to better understand and model arcing faults. Standards and recommended practices are
changing constantly in order to reflect the added understanding we are gaining and to
better protect workers. Personal protective equipment (PPE) is also changing at a rapid
pace as new and better technology is developed. ESA has created an Arc Flash Resource
Center at the website www.easypower.com to keep you up to date as new information
becomes available and industry advancements are made. Look for new versions of this
guide as we continue to enhance and add new technology to the arc flash assessment
process.
ESA is committed to providing industry with the most advanced state of the art
technology in our EasyPower software product line. We believe EasyPower provides the
self-documenting solution capabilities to keep your safety program current and in
compliance with OSHA and NFPA-70E regulations. ESA can also provide detailed
engineering studies and arc flash assessment programs to help your company get started.
We hope that the Practical Solution Guide to Arc Flash Hazards becomes a valued
resource to your library.
Sincerely,
Chet E. Davis, PE
President, ESA
Table of Contents
1
1.1
1.2
1.3
1.3.1
1.4
1.5
1.6
1.7
1.7.1
1.7.2
1.7.3
1.7.4
1.7.5
1.7.6
1.8
1.9
1.10
Introduction...................................................................................................... 1
Causes of Electric Arcs.......................................................................................... 1
The Nature of Electrical Arcs ................................................................................. 2
Hazards of Arcing Faults........................................................................................ 3
Probability of Survival ........................................................................................ 4
Impacts of Arc Flash .............................................................................................. 4
Potential Exposure to Arc Flash............................................................................. 4
Recent Developments in Addressing Arc Flash Hazard........................................ 5
NFPA 70E and Arc Flash Hazard .......................................................................... 7
Protection Boundaries ....................................................................................... 7
Flash Protection Boundary ................................................................................ 7
Personal Protective Equipment ......................................................................... 8
Classification of Hazard/Risk Category ............................................................. 8
Determining Flash Protection Boundary and Hazard Category ........................ 9
Difference between NFPA 70E and IEEE 1584 Calculations............................ 9
Hazard Assessment Methods .............................................................................. 10
Reducing Exposure to Arc Flash Hazard............................................................. 11
Arc Flash Hazard Program .................................................................................. 12
2.1
2.2
2.2.1
2.2.2
2.2.3
2.2.4
2.2.5
2.2.6
2.3
2.3.1
2.3.2
2.3.3
2.4
2.5
2.5.1
2.5.2
2.5.3
2.5.4
2.5.5
2.5.6
2.6
2.7
7
7.1
7.2
7.2.1
7.2.2
7.2.3
7.2.4
7.2.5
7.2.6
7.2.7
7.2.8
7.3
7.3.1
7.3.2
7.3.3
7.4
7.5
7.6
Appendices
A.
B.
Chapter 1. Introduction
1 Introduction
This chapter provides an overview of arc flash hazards and briefly describes the various
causes, nature, results, standards and procedures associated with arc flash hazards. In
order to deal with the hazard, it is first necessary to develop an understanding of the
phenomena. Details are provided in the following chapters.
An electric arc or an arcing fault is a flashover of electric current through air in electrical
equipment from one exposed live conductor to another or to ground. Arc flash hazard is
the danger of excessive heat exposure and serious burn injury due to arcing faults in
electrical power systems. Electric arcs produce intense heat, sound blast and pressure
waves. They have extremely high temperatures, radiate intense heat, can ignite clothes
and cause severe burns that can be fatal.
The demand for continuous supply of power has brought about the need for electrical
workers to perform maintenance work on exposed live parts of electrical equipment.
Besides the existence of electrical shock hazard that results from direct contact of live
conductors with body parts, there also exists a possibility of electric arcs striking across
live conductors. Although electrical safety programs have existed since the beginning of
electricity, arc flash hazard has not been prominently addressed until recently.
Condensation of vapor and water dripping can cause tracking on the surface of
insulating materials. This can create a flashover to ground and potential escalation
to phase to phase arcing1.
Spark discharge:
Over-voltages across narrow gaps: When air gap between conductors of different
phases is very narrow (due to poor workmanship or damage of insulating
materials), arcs may strike across during over-voltages.
Figure 1.1: (a) Arc blast in box2 ; (b) Arcing fault in electrical panel board
Electric arcs produce some of the highest temperatures known to occur on earth
up to 35,000 degrees Fahrenheit3. This is four times the surface temperature of the
sun.
The intense heat from arc causes the sudden expansion of air. This results in a
blast with very strong air pressure (Lightning is a natural arc).
Chapter 1. Introduction
All known materials are vaporized at this temperature. When materials vaporize
they expand in volume (Copper 67,000 times, Water1670 times4). The air blast
can spread molten metal to great distances with force.
For a low voltage system (480/277 V), a 3 to 4-inch arc can become stabilized
and persist for an extended period of time.
Energy released is a function of system voltage, fault current magnitude and fault
duration.
Figure 1.2: (a) Hand burned by arc flash5; (b) Clothed areas can be burned more
severely than exposed skin
Some of the hazards of arcing faults are:
Heat: Fatal burns can occur when the victim is several feet from the arc. Serious
burns are common at a distance of 10 feet6. Staged tests have shown temperatures
greater than 437F on the neck area and hands for a person standing close to an arc
blast7.
Pressure: Blast pressure waves have thrown workers across rooms and knocked
them off ladders8. Pressure on the chest can be higher than 2000 lbs/ sq. ft.
Blast
Clothing can be ignited several feet away. Clothed areas can be burned more
severely than exposed skin.
Hearing loss from sound blast. The sound can have a magnitude as high as 140 dB
at a distance of 2 feet from the arc9.
% Survival
80
60
40
20
0
20 - 29.9
30 - 39.9
40 - 49.9
50 - 59.9
Figure 1.3: Burn Injury Statistics Probability of Survival (Source: American Burn
Association, 1991-1993 Study; Revised March 2002)
Litigation fees.
Production loss.
Chapter 1. Introduction
cases of reported days away from work due to electrical burns, electrocution/electrical
shock injuries, fires and explosions.
The Census of Fatal Injuries noted 548 employees died from the causes of electrical
current exposure, fires and explosions of 6,588 work related fatalities nationwide.
In the US Chemical Industry, 56% of the fatalities that occurred over a 5-year period
were attributed to burns, fires and explosions, with many of the ignition sources being
related to electrical activity.
Capelli-Schellpfeffer, Inc. of Chicago reported that there are 5 to 10 arc flash injuries per
day resulting in hospitalization. Many arc flash accidents/injuries occur that do not
require a stay or are not properly documented for national tracking purposes. The
number of arc flash accidents is greater than many engineers realize since most arc flash
accidents do not make the daily news.
IEEE Standard 1584, IEEE Guide for Performing Arc Flash Hazard Calculations,
provides 49 arc flash injury case histories in Annex C. A brief description is provided for
each case on incident setting, electric system, equipment, activity of worker, event,
apparel worn by the worker and the outcome of the incident. Readers are encouraged to
read these case histories to gain insights on various conditions leading to such incidents.
The exposure to arc flash depends on the following:
Complexity of the task performed, need to use force, available space and safety
margins, reach, etc.
Tools used.
Condition of equipment.
developed. NFPA 70E is intended for use by employers, employees, and the
Occupational Safety and Health Administration (OSHA). The publication NFPA 70E
(2000) and its proposed revision (May 2003 ROP11) include arc flash hazard as a
potential danger to workers near and around live exposed electrical parts. NFPA 70E and
IEEE Std 1584-2002TM provide guidance on implementing appropriate safety
procedures and arc flash calculations. For the actual wording, see section 6.1.2.
NEC Article 110.16 requires "field marking" of potential arc flash hazards for panels
likely to be serviced or examined in an energized condition. This article also contains a
fine print note (FPN) regarding proper signage and an FPN referencing NFPA 70E.
These FPNs are not technically part of the NEC, but are recommended practices.
OSHA has not specifically addressed arc flash hazards, however, there exists adequate
safety requirements for employers to follow to ensure the safety of the worker in the
workplace (General Duty clause). Some of these are outlined in Table 6.1 in Chapter 6.
The Code of Federal Regulations (Standards 29 CFR) Part 1910 deals with
occupational safety and health standards. Standards on personal protective equipment
(PPE) are outlined in subpart 132. In response to an inquiry on OSHA's stand on arc
flash hazard, Richard S. Terrili, the Regional Administrator for Occupational Safety and
Health, US Department of Labor for the Northwest Region at Seattle, concluded as
follows:
"Though OSHA does not, per se, enforce the NFPA standard, 2000 Edition,
OSHA considers NFPA standard a recognized industry practice. The employer
is required to conduct assessment in accordance with CFR 1910.132(d)(1). If
an arc flash hazard is present, or likely to be present, then the employer must
select and require employees to use the protective apparel. Employers who
conduct the hazard/risk assessment, and select and require their employees to
use protective clothing and other PPE appropriate for the task, as stated in the
NFPA 70E standard, 2000 Edition, are deemed in compliance with the Hazard
Assessment and Equipment Selection OSHA standard."
12
Chapter 1. Introduction
Limited
Approach
Boundary
Flash
Protection
Boundary
Prohibited
Approach
Boundary
Restricted
Approach
Boundary
Energized
Equipment
is a function of the available fault current of the system at that point, the voltage and the
tripping characteristics of the upstream protective device as well as some other
parameters. See Chapter 4 for details.
1.7.3 Personal Protective Equipment
NFPA specifies the requirement of personal protective equipment (PPE) for workers
within the flash protection boundary. All parts of the body which may be exposed to the
arc flash, need to be covered by the appropriate type and quality of PPE. The entire PPE
set may be comprised of FR clothing, helmet or headgear, face shield, safety glasses,
gloves, shoes, etc. depending upon the magnitude of the arc energy. The amount of PPE
required and its quality needs to be determined on the basis of the calculated incident
energy on the worker's body. The calculations need to be performed by a qualified
person such as an engineer. The protective clothing should limit the incident energy
reaching the chest/face of the worker to less than 1.2 cal/cm2. FR clothing provides
thermal insulation and is also self-extinguishing. Protective clothing is rated in cal/cm2.
For details on PPE, see Chapter 6.
1.7.4 Classification of Hazard/Risk Category
NFPA 70E defines 5 levels of risk category for arc flash hazard based upon the calculated
incident energy at the working distance, as shown in Table 1.1. Examples of typical
protective clothing that cover the torso are also provided in this table. Other PPE are also
required to protect various parts of the body.
Table 1.2: Hazard/risk classification as per NFPA 70E-2000
Category
Energy Level
N/A
5 cal/cm2
8 cal/cm2
25 cal/cm2
40 cal/cm2
Chapter 1. Introduction
10
Chapter 1. Introduction
coordination, and arc flash calculations and display results graphically with simple
mouse clicks. EasyPower provides an active self-documenting arc flash assessment
program to meet the needs of todays changing electrical systems.
The results of the assessment can show up on one-line drawings, detailed arc flash
reports and warning labels that can be placed on the hazardous location or equipment.
An additional advantage of EasyPower software is the ability to simulate and modify
the protective device settings in order to reduce exposure to arc flash hazard. The
software can automatically obtain the accurate arcing time from the trip
characteristics of the protective devices. All other methods lack this ability and
therefore need to rely on some approximate value for arcing time.
TX-2
BUS-4
BL-2
67.1" AFB
10.4 cal / cm @ 18"
#3 @ 18"
BL-1
99.6" AFB
18.7 cal / cm @ 18"
#3 @ 18"
BL-3
31.5" AFB
2.7 cal / cm @ 18"
#1 @ 18"
BL-4
67.1" AFB
10.4 cal / cm @ 18
#3 @ 18"
M-1
Figure 1.3: Example of arc flash hazard calculation results on one-line diagram in
the integrated software EasyPower.
Figure 1.4: Example of detailed arc flash hazard report for equipment shown in
Figure 1.3 produced by the integrated software EasyPower.
3. Develop an arc flash hazard program and integrate it into the safety program.
12
Chapter 1. Introduction
5. Training for workers: Workers who are exposed to arc flash hazard should be well
trained to understand what the hazard is, how it is initiated, how to read the
documents and warning labels, how to properly wear PPE, and how the hazard can be
reduced with safer working procedures. Different tasks will require different work
practices.
6. Continual improvement: It is expected with more research and development in arc
flash hazard, that there will be further additions to what we already know. The arc
flash hazard program can be continually improved by including new developments in
standards, industry practices and PPE. Since the power system within a company can
keep changing with time, it is necessary to update arc flash assessment information on
a regular basis. Also, experience can bring in new ideas from workers that can be
included in the program. For this reason, it is necessary to keep the program ongoing
rather than implement it as a one-time project.
7. Safety audit: Safety audits should be performed regularly to evaluate various aspects
of a safety program. The safety audit should include arc flash hazard. If the arc flash
hazard program is in its initial stages, then a closer examination is required.
8. Corporate-wide plan: Corporate-wide plan should be implemented to ensure
consistency in safety practices. It is not advisable to have each plant or division
implement the safety program differently. Communication channels should be
established and responsibility should be distributed between various plants or
divisions, taking a unified approach.
Ralph Lee, "Pressures Developed by Arcs", IEEE Transactions on Industry Applications, Vol.
IA-23, No. 4. July/August 1987, page 760-764.
Source: Thomas E. Neal, Presentation "Insight Into The Arc Hazard", IEEE-PCIC Electrical
Safety Workshop, February, 2003; DuPont Company.
Ralph Lee, "The Other Electrical Hazard: Electrical Arc Blast Burns", IEEE Transactions on
Industry Applications, Vol. IA-18, No. 3 May/June 1987, page 246-251.
See endnote 1.
Source: Danny P. Ligget, Presentation "Electrical Hazards Taking Basics to the Future", IEEEPCIC Electrical Safety Workshop, February, 2003.
6
See endnote 3.
Ray A. Jones, et al, "Staged Tests Increase Awareness of Arc-Flash Hazards in Electrical
Equipment", IEEE Transactions on Industry Applications, Vol. 36, No. 2, March/April 2000, page
659-667.
8
See endnote 1.
See endnote 7.
10
NFPA 70E, Standard for Electrical Safety Requirements for Employee Workplaces10, 2000
Edition, National Fire Protection Association.
13
11
NFPA 70E May 2003 ROP is only a proposed version. The 2003 ROP Revision of the
standard is scheduled to be published as the 2004 version after multiple changes in January of
2004. Readers should note the differences and follow the published standard after it is released.
12
NFPA Online, "Landmark agreement to use NFPA 70E protects electricians in Columbus OSHA, IBEW and NECA contractors forge pact that could lead the nation", September 27, 2002;
( http://www.nfpa.org/PressRoom/NewsReleases/ Landmark/ Landmark.asp).
14
15
3. Number of times workers are exposed to live equipment: When the frequency of
exposure to live equipment is small, an elaborate program may not be needed a few
simple procedures may suffice.
4. Voltage level: For low voltage equipment (240V or less) being fed by small
transformers (125kVA or less), the potential hazard is small, and therefore does not
need to be included in arc flash hazard assessment. The higher the voltage or the size
of transformers, the greater the risk.
5. Continuous processes: Continuously operated facilities may require work on
energized equipment like MCC's and panels. The exposure to risk is higher for such
plants. When possible, schedule work during plant shutdowns.
6. System size: Large systems are likely to have greater arc flash hazard due to the
higher fault currents.
7. System condition: Systems that do not receive periodic planned maintenance are
likely to have a higher risk of arc flash incidents.
8. Changes in electrical system: Since the level of risk depends on the possible
magnitude of arc current, which in turn depends on the interconnections within the
power system, a system that changes with time due to the requirements of the
company will need review of arc flash hazard when the changes are implemented.
Additional effort will need to be incorporated into the safety program to address the
changes. A static electrical power system will require the assessment only once and
the safety procedures will remain the same unless the fault level of the utility changes
or OSHA and NFPA regulations change.
9. Environmental conditions: Are the exposed live parts of electrical equipment subject
to corrosive vapors (such as in chemical plants, sea-side, etc.), oxidation, bees, dust,
rodents or birds causing electrical disturbances resulting in spark and eventually arc
flash? The chance of arc flash exposure is higher in such cases.
2.2.3 Assessing Existing Safety Program
When implementing a new arc flash hazard mitigation program, the additional efforts,
manpower, budget and time that is required will depend largely upon what is already in
place and what resources are available to the company. As mentioned in the previous
sections, the arc flash hazard program is an integral extension of the existing safety
program and is not about just sticking labels on equipment and wearing flash suits. The
following points should be considered in the preliminary planning stage.
17
How often?
Is the training
Has assessment of electrical hazards been carried out? How often? Are
warning labels posted in these areas?
Has safety audit been performed? How often?
Is PPE provided to workers? Is the PPE adequate? Is PPE properly used
and maintained?
Does each facility have up-to-date electrical drawings, short circuit, and
protective device coordination studies?
Is each facility modeled with state of the art graphical power system
software to self-document the system and safety assessment in compliance
with NFPA-70E?
Has the company developed any procedures for safety? Do the workers
follow them?
Are contractors required to follow the same or similar safety procedures?
Is safety training provided to outsourced professionals or contractors?
Willingness of workers to comply with changes in safety program:
Accepting the arc flash program and wearing arc flash PPE is a significant
change to operating habits. Experience shows that workers do not like to
comply with additional clothing.
18
Existence of arc flash hazard program: If some kind of arc flash hazard program is
already in place, then improving the program will not be difficult since the basic
concepts will already have been implemented. Any improvement will come in the
form of better accuracy in hazard evaluation, better documentation, training and
selection of PPE. Since the basic data for calculations will be readily available, the
assessment can be conducted relatively quickly.
Which of the following methods does the company use to assess arc flash hazards:
1.
2.
3.
4.
Integrated software?
Does the company employ electrical engineers? How much time can they devote
to arc flash hazard program? Can they manage this project? Are they trained or
experienced in short circuit studies, protective device coordination and arc flash
hazard assessment?
Does the company employ electrical technicians or use outside technicians for
routine maintenance?
Does the company have safety coordinators for different locations? How much
time can they devote to an arc flash hazard program, for both learning and
implementing?
Does the company have personnel for managing the entire arc flash safety
program?
How much time can workers devote to learning about arc flash hazard and its
prevention as well as implementation of procedures at work?
The source of power supply, multiple grid connection, co-generation and multiple
generators affect the available fault level, the complexity of the arc flash hazard
assessment and the number of scenarios that will be required for analysis. Data is
required from the serving utilities.
Radial distribution versus loops: Radial distribution systems are easier to deal with
and hand calculations can be performed for small systems. Looped distribution
systems require more rigorous calculations.
19
Number of voltage levels: The nature of arcs and therefore the calculation method
varies with the voltage level, and so does the risk.
Number of connection points (buses): Each bus needs to be assessed for arc flash
hazard, and therefore contributes to the total project size.
Number and types of equipment/load: Different data sets are required for different
equipment. The calculation details may also change.
Does the company have up-to-date drawings and equipment data readily available?
Has a short circuit and protective device coordination study been performed
recently for the existing system? This determines how much extra work may be
needed.
20
Protect the workers from potential harm and prevent loss of life.
Comply with Occupational Safety and Health Administration (OSHA) codes and
with National Fire Protection Association (NFPA) standards on employee safety,
NFPA-70E.
2.3.3 Objectives
Objectives are basically an extension of the goals, but are more specific. Typically
companies associate measurable statistics with objectives. Some objectives may be
associated with the end result, (for instance, zero accidents), whereas some may be
objectives of the process involving "what to do" or "how to do".
For example, the process related objectives could be:
Train 50% of workers with a basic level course and 50% of workers with an
advanced level course within 6 months.
Accomplish arc flash hazard assessment in 25% of the plant locations within 6
months.
Select and purchase 50% of the required PPE in three months and complete
distribution in six months.
Reduce the lost work day case incident rate (LWDCIR) by 50% within 1 year.
21
These are general examples for safety related programs, but specific goals may be set for
arc flash hazards.
Some objectives may be set to tie in with the regular safety audits. This is useful for
ongoing programs. For example,
The role of defining objectives is critical because program design and resource allocation
are based on the objectives.
22
NFPA
Tables
Hand
SpreadCalculations sheet
Integrated
software
Number of Buses
< 25
< 25
< 50
50+
1-2
1-2
2-3
3+
Radial/Loop Distribution
Radial
Radial
Radial
Either
Power Sources
Multiple
Frequent Changes in
System
No
No
No
Yes
Diversity in Protective
Devices
Small
Small
Medium
Large
Low
Medium
Medium
High
Yes
Yes
Yes
No
Separate Coordination
Studies
Yes
Yes
Yes
No
Table 2.1 provides a guideline for the selection of assessment method based upon various
system attributes. This guide has been prepared taking the following considerations
Accuracy required.
23
Table 220.6(B)(9)(A) in the proposed NFPA 70E May 2003 ROP provides the
hazard/risk category based on available fault current, voltage, fault clearing time, type of
work to be performed and type of equipment. The available fault current is typically not
known until someone calculates it from the system data. If the available fault current is
not known, then calculations will need to be performed anyway. Once the fault currents
are known, then it is fairly easy to look up the tables to find the hazard category. The
fault clearing time may not be the same as it is assumed in this method. This could lead
to erroneous results and therefore could be risky. The summary of this table for working
on live equipment and for testing voltages is provided in Chapter 3.
2.5.2 Hand Calculations
Hand calculations can be performed either using NFPA 70E Annex B or Annex C
methods, or using IEEE Standard 1584 equations. Please read the individual standards
for details. The summary of the calculation methods for both are provided in Chapter 3.
2.5.3 Spreadsheet Calculator
The IEEE 1584 spreadsheet calculator provides a quick way to obtain arc flash hazard
results. However, like both of the previous methods, this requires the input of available
fault current and the fault clearing time. For some typical protective devices, the total let
through energy can be computed without the need for entering the clearing time. Also if
the protective device is a current-limiting device for which characteristics has been
determined, the reduced arc current and the associated reduced arc flash energy is
calculated. The calculator requires users to enter what percent of arc current is flowing
through the protective device. This can only be determined from short circuit studies.
Often, loads such as motors, contribute to the arcing fault current. The calculator does
not take this into account.
2.5.4 Commercial Integrated Software - EasyPower
EasyPower, power system software provides an extensive array of capabilities to
minimize the effort level required to obtain accurate arc flash analysis in compliance with
NFPA-70E and IEEE-1584. A brief summary of its integrated features are provided
below.
Integrated software may contain several capabilities that reduce the engineering effort to
a single arc flash assessment task. Commercial integrated software can have the
following features that are necessary for arc flash hazard assessment.
24
Short circuit analysis: A single mouse click can calculate the exact short circuit
current at every point (bus/equipment), along with the contributing currents from
every branch including the branch that contains the upstream protective device.
Calculations are provided in accordance with ANSI and IEEE Standards ensuring
proper arc current for both total energy and tripping times.
Arc flash analysis: EasyPower provides total arc flash integration in both short
circuit and protective device coordination modes. Arc flash boundaries, incident
energies, PPE requirements and more, are calculated as you change system
parameters so you instantly know the effect/danger of any system or parameter
change. Assessment reports can be instantly generated and results can also
viewed on the one-line diagrams.
Just like any other method, the basic data is required to operate the software. Once the
necessary data has been entered into the software, results can be obtained with a few
clicks of the mouse button. EasyPowers fast algorithms, accuracy, ability to perform
complex calculations and consider multiple scenarios for various possible connections
make it ideal for both large and small systems and those that have frequent additions or
changes.
EasyPowers complete integration for all critical aspects of arc flash analysis from data
repository, one-line documentation and arc flash results make it the clear choice for a self
documenting safety program. A single source program to maintain compliance with the
many aspects of NFPA-70E arc flash requirements can greatly simplify a safety program.
2.5.5 Accuracy and Conservatism
Before performing an arc flash hazard assessment, it is necessary to determine how
accurate or conservative the assessment needs to be. Arc flash assessment methods rely
on theoretical and empirical equations. It has been observed that there is some random
behavior of arcing faults that may result in actual occurrences that differ from predicted
values. Although the theoretically maximum arcing power has been used by NFPA
methods (which is believed to be safer) arcing currents can vary randomly. This affects
the fault clearing time, and hence the incident energy to which workers may be exposed.
25
Therefore, when carrying out the assessment, it is necessary to cover all aspects of
variability in order to be truly conservative. This requires us to consider a range of
possible values rather than just a single value obtained from short circuit studies. A quick
way to consider a whole range is described in Chapter 4. Appendix A provides more
details on how to deal with the random behavior of arcing faults.
Additional analysis time is required to consider a range of values instead of a single
value. However, this eliminates chances of error and provides greater accuracy.
Recently, it was proposed that over-protection of workers could cause greater chances of
accidents as the workers' movement could be limited due to excess PPE13. The IEEE
Standard 1584 was developed using test results to avoid over-protection from theoretical
equations. Although tests performed may have shown that the theoretical maximum arc
power was not achieved during the tests, the possibility of its occurrence cannot be ruled
out. Therefore, taking theoretical equations to be conservative is valid.
Although not always true, the chart below provides general observations on various
calculations methods.
Description
Conservative
IEEE 1584
Statistically Probable
Scenario Analysis
Improved Accuracy
Scenario analysis is a standard feature in EasyPower and provides easy analysis of all
system configurations and operating conditions and for bracketing arcing current ranges.
2.5.6 Overprotection
When arc flash hazard assessment is too conservative, the assessed hazard/risk category
or the incident energy may require the workers to wear more protective gear than is
practically necessary. Extra layers of thick fire resistant clothing, face shields, and thick
gloves may render the work rather difficult. This situation has the following
disadvantages:
1. The difficulty provided by the PPE may lead to accidents that can be avoided by
slightly less but adequate PPE.
26
2. Longer time is taken by a worker to execute a task when wearing heavier PPE,
therefore reducing overall productivity. Safety should not be compromised to increase
productivity, however, over-protection cannot achieve greater safety.
3. Workers may be discouraged with the chore of having to wear extra PPE.
Task
Category
Data Collection
Hours
Per
Equipment
Substation
0.1
Load
Short Circuit
0.15 0.25
Bus
0.1 0.25
Device
Short Circuit
Study
Analysis
0.1 0.25
Bus
Report
0.1 0.25
Bus
Protective Device
Coordination
Analysis
0.4
Device
Report
0.15 0.4
Device
3-Scenario Analysis
0.25
Bus
1-Scenario Analysis
0.1 0.25
Bus
Report
0.1 0.25
Bus
0.05
Equipment
27
Similarly, estimates can be made for other activities such as worker training, safety audit,
documentation (much of the documentation will already have been done during the
assessment), procurement and distribution of PPE and development of safety procedures
for arc flash hazard.
Table 2.4: Estimate of hours for arc flash hazard assessment for a plant with 350
buses at 4 different voltage levels and 56 substations, using commercial integrated
software for computation.
Task
Hours
% of Total
Data Collection
136
18%
64
9%
80
11%
Protective Device
Coordination
336
46%
120
16%
Total
736
13
28
Applicable Range
0.208 to 15 kV
Frequencies (Hz)
50 or 60 Hz
0.7 to 106 kA
13 to 152 mm
Phases
3 Phase faults
(3.1)
where
log is the log10
Ia = arcing current (kA)
K
=
0.153;
open configuration
=
0.097;
box configuration
Ibf = bolted fault current for three-phase faults (symmetrical RMS) (kA)
V = system voltage (kV)
29
(3.2)
(3.3)
where
En = incident energy normalized for time and distance (J /cm2)
K1 = -0.792;
= -0.555;
open configuration
box configuration
t 610
0.2 D
(3.4)
where
E = incident energy (J / cm2)
Cf = Calculation factor
30
Table 3.2: Distance Factor (x) for various voltages and enclosure types
Enclosure Type
0.208 to 1 kV
>1 to 15 kV
Open air
Switchgear
1.473
0.973
1.641
Cable
t 1
D B = 610 * 4.184C f E n
0.2
E B
(3.5)
where
DB = distance of the boundary from the arcing point (mm)
Cf
= calculation factor = 1.0; voltage > 1 kV
= 1.5; voltage < 1 kV
En = incident energy normalized
EB = incident energy at the boundary distance (J/cm2); EB can be set at 5.0 J/cm2 (1.2
Cal/cm2) for bare skin.
t = arcing time (seconds)
x = the distance exponent from Table 3.2.
Ibf = bolted fault current (kA).
Serious injury due to arc flash burns can occur within this area unless appropriate
PPE is used.
Anyone within this area must wear appropriate PPE regardless of what they are
doing.
The distance from the arc source at which the on-set of a second degree burn
occurs.
31
1.2 Cal/cm > 0.1 sec. is considered a second degree burn threshold.
Medical treatment may still be required if bare skin is exposed to this level of
flash. Full recovery expected.
Defines a boundary around exposed live parts that may not be crossed by
unqualified persons unless accompanied by qualified persons.
Boundary near exposed live parts that may be crossed only by qualified persons
using appropriate shock prevention techniques and equipment.
The theoretical maximum arc power in MW is half the bolted 3-phase fault MVA14,15.
This occurs when the arc current is 70.7% of the bolted fault current. Based on this, the
flash protection boundary is calculated as:
32
where
DB =
V =
Ibf =
t =
3.3.1.2
(3.6)
Incident Energy
1.9593
(3.7)
1.4738
t [ 0.0093 * Ibf
(3.8)
(3.9)
V Ibf t
where
E
Ibf
t
D
=
=
=
=
Equations (3.7) and (3.8) are part of the 2000 edition, and equation (3.9) was proposed in the
2003 draft.
(416 Ia t / 1.2)0.625
1 kV < V < 5 kV
V > 5 kV
0.928 Ibf
Ibf
21.8 Ia t D-0.77
16.5 Ia t D-0.77
(21.8 Ia t / 1.2)1.3
(16.5 Ia t / 1.2)1.3
33
The equations in Table 3.3 apply only to arc in box for short circuit currents between 0.6
kA and 106 kA.
where
E =
Ibf =
Ia =
t =
D =
DB =
System Voltage
Flash Protection
Boundary (feet)
Arc in Air
Arc in Enclosure
10
Arc in Enclosure
20
34
removing and installing bolted covers, applying safety grounds, working on control
circuits, etc.
An example of what 70E (2004) Table 220.6(B)(9)(A) may look like is summarized for
two items in Table 3.4: working on live parts and voltage testing. This table is
preliminary and is for reference purposes only. Refer to NFPA-70E (2004) for final
application guidelines.
The exact short circuit currents for three phase bolted faults can be calculated using
commercial software. A simple approximation described in Annex B.2 of proposed
NFPA 70E 2003 ROP draft is using the upstream transformer data in the following
equation. The actual short circuit current will be less than this calculated value due to the
impedance of the system upstream to transformer.
MVA Base 100
1.732 V %Z
ISC =
(3.10)
where
ISC = 3-phase bolted fault current
MVA Base = rated MVA of the upstream transformer
V = line-to-line voltage at the secondary side of the transformer
%Z = percentage impedance of the transformer.
35
Table 3.4 (a): Hazard / Risk Category Classification for Working on Live Parts as
per Table 220.6(B)(9)(A) of proposed NFPA 70E 2003 ROP (Note: This table is
only a proposed draft and had not been approved or published at the time of writing this
book. Please refer to the upcoming edition of NFPA 70E (2004).)
Equipment Type
Short Circuit
Current (kA)
Fault
Clearing
Time (s)
0.24 kV
42
0.03
< 10
0.03
< 36
0.1
65
0.03
< 10
0.03
Bus
42
0.33
Bus
52
0.2
Bus
65
0.1
Bus
62
0.33
Bus
76
0.2
Bus
102
0.1
Bus
< 10
0.1
Bus
< 10
0.33
35
0.5
42
0.33
52
0.2
65
0.1
< 25
0.33
62
0.33
76
0.2
102
0.1
35
0.5
42
0.33
52
0.2
65
0.1
55
0.35
Equipment Side
Panel Board
Load Side of
Breaker / fuse
0.277 to 2.3 to 1 to 38
0.6 kV 7.2 kV
kV
Other Equipment
36
Table 3.4 (b) Hazard / Risk Category Classification for Voltage Testing as per Table
220.6(B)(9)(A) of proposed NFPA 70E 2003 ROP.
Equipment Type
Short Circuit
Current (kA)
Fault
Clearing
Time (s)
0.24 kV
42
0.03
< 10
0.03
< 36
0.1
65
0.03
< 10
0.03
Bus
42
0.33
Bus
52
0.2
Bus
65
0.1
Bus
62
0.33
Bus
76
0.2
Bus
102
0.1
Bus
< 10
0.1
Bus
< 10
0.33
35
0.5
42
0.33
52
0.2
65
0.1
< 25
0.33
62
0.33
76
0.2
102
0.1
35
0.5
42
0.33
52
0.2
65
0.1
55
0.35
Equipment Side
Panel Board
Load Side of
Breaker / fuse
0.277 to 2.3 to 1 to 38
0.6 kV 7.2 kV
kV
Other Equipment
37
Arcs in 1987, he cites several case histories. In one case, with approximately 100-kA
fault level and arc current of 42 kA, on a 480-V system, an electrician was thrown 25
feet away from the arc. Being forced away from the arc reduces the electricians exposure
to the heat radiation and molten copper, but can subject the worker to falls or impact
injuries. The approximate initial impulse force at 24 inches was calculated to be
approximately 260 lb/ft2 as determined from the equation below.
Pressure =
11.58 * Iarc
D
0.9
(3.11)
where,
Pressure is in pounds per square foot.
D = Distance from arc in feet.
Iarc = Arc current in kA.
14
Ralph Lee's, "The Other Electrical Hazard: Electrical Arc Blast Burns," IEEE Transactions on
Industry Applications, Vol. 1A-18, No. 3, Page 246, May/June 1982.
15
This conclusion is supported by the Maximum Power Transfer Theorem: The power
transferred to a load (an arc in our case) is maximum when the impedance of the load is equal to
the Thevenin impedance of the source. The theorem was first developed by Moritz Hermann
Jacobi.
16
Proposed NFPA 70E, Standard for Electrical Safety Requirements for Employee Workplaces,
2003 ROP Edition, National Fire Protection Association.
17
Ralph Lee, "Pressures Developed by Arcs", IEEE Transactions on Industry Applications, Vol.
IA-23, No. 4. July/August 1987, page 760-764.
38
39
b. Calculate branch currents contributing to the arc current from every branch.
6. Estimate arcing time from the protective device characteristics and the contributing
arc current passing through this device for every branch that significantly contributes
to the arc fault.
7. Estimate the incident energy for the equipment at the given working distances.
8. Determine the hazard/risk category (HRC) for the estimated incident energy level.
9. Estimate the flash protection boundary for the equipment.
10. Document the assessment in reports, one-line diagrams and with appropriate labels on
the equipment.
40
Table 4.1: Typical data needed for equipment for short circuit analysis
Description
Data
Equipment Type
Voltage
MVA/KVA
Impedance
X/R Ratio
Phases/connection
4.2.2 Equipment Data For Protective Device Characteristics
Obtain data on the various protective devices that will determine the arcing time. Table
4.2 shows what kind of information is required. This data may be obtained from existing
drawings, relay calibration data, coordination studies and from field inspection. Obtain
from the manufacturers the time-current characteristics (TCC) for these devices.
Determine whether the protective device is reliable enough. This can be done by
asking the operators, or by testing if necessary. Some companies have periodic relay
testing programs. If the protective device is deteriorating, the data provided by the
manufacturer may not be applicable. If the fault interruption does not occur as expected
then the arc flash assessment cannot be accurate. It will be necessary to repair or replace
such equipment.
Table 4.2: Protective device data to gather
Protective
Device
Data to Gather
Relay
Fuse
Breaker
41
Data
Normal
Operation
Co-generation
Maintenance
Schedule A
Maintenance
Schedule B
Utility
ON
ON
OFF
ON
Generator
OFF
ON
ON
OFF
Motor
ON
ON
ON
OFF
M-2
ON
ON
ON
ON
42
UTILITY
UTILITY
OPEN
0.48KV BUS
0.142
0.346
0.142
Faulted
Bus
TRANSFORMER
MOTOR
OPEN
4.16KV BUS
1.
52
9
1.388
2.
93
4
1.388
4.16KV BUS
Faulted
Bus
GENERATOR
1.067
GENERATOR
TRANSFORMER
MOTOR
0.48KV BUS
M-2
M-2
Figure 4.1: EasyPower example of connected equipment for two possible cases
calculations be done for an equivalent 3-phase system and states that this will yield
conservative results. Based on the data collected for various system operating modes,
arc flash calculations should be performed for each possible case. Traditionally, when
performing short circuit calculations to determine maximum short circuit current,
extremely conservative estimates and assumptions are used. This makes sense if the goal
is to determine maximum breaker or equipment duties. However, for AFH, using overly
conservative short circuit data can yield non-conservative results since a very high fault
current may produce a very short arc duration due to the operation of instantaneous trip
elements. The highest fault current does not necessarily imply the highest possible arc
flash hazard because the incident energy is a function of arcing time, which may be an
inversely proportional function of the arcing current. For AFH determinations, short
circuit calculations should be conservative, but not overly conservative.
4.4.1 Calculate Bolted Fault Current
Calculate the 3-phase bolted fault current in symmetrical rms amperes for all buses
or equipment, and for each possible operating mode. Check for the following while
considering various interconnections at the concerned bus or equipment:
Multiple local generator sources that are operated in parallel or isolated depending
on the system configuration.
Emergency operating conditions. This may be with only small backup generators.
Maintenance conditions where short circuit currents are low but arc duration may
be long.
A short circuit/arc flash case should be developed for each operating mode. This can be a
daunting task for most software or spreadsheet calculators. EasyPower Scenario Manager
provides a simple and easy method to document and analyze each operating mode for
quick repeatable analysis.
Hand calculations and spreadsheet calculators may typically neglect the transient short
circuit values that last for a few cycles. These are higher than the sustained short circuit
values. The generator and motor transients during the fault contribute to the arc fault. To
account for these:
44
Use sub-transient and transient impedances of the generators to find the bolted
fault current if the arcing time is small. For long arcing times, use the sustained
short circuit values. This suggests an iterative process, since the arcing time
depends upon the fault current passing through the upstream protective device.
To cover the variance that can occur in arcs, IEEE procedure suggests the following.
1. Calculate the maximum expected bolted fault condition.
2. Calculate the minimum expected bolted fault condition. The minimum bolted
fault current could be a light load condition with many motor loads or generators
not running.
45
3. Calculate the arcing current at 100% of IEEE 1584 estimate for the above two
conditions.
4. Calculate the arcing current at 85% of IEEE 1584 estimate for the two above
conditions.
5. At these four arcing currents calculate the arc flash incident energy and use the
highest of the incident energies to select PPE. The minimum fault current could
take longer to clear and could result in a higher arc flash incident energy level
than the maximum-fault current condition. The fault current in the main fault
current source should be determined since the current in this device may
determine the fault clearing time for the major portion of the arc flash incident
energy.
Note: Although IEEE recommends considering a range of 85% to 100% of the estimated
arc current, IEEE test data shows that the measured values of arc current vary from 67%
of the estimate to 157% of the estimate. Further analysis of the IEEE test data was
performed by the authors and the results are discussed in detail in Appendix A. Careful
application of tolerances is required for the following reasons:
i. The tripping time of inverse-time protective devices is influenced by the arc
current.
ii. The incident thermal energy is more sensitive to arcing time than it is to arc
current.
iii. A more realistic and reasonably conservative estimate of arcing time can be
obtained by proper selection of tolerances of arc current.
The following section provides guidelines based on statistical analysis of test data for
various voltage levels and enclosure types.
4.5.2.2
Because of the random nature of arc currents, the actual arc current may take any value
within a range of possible values. The calculated arc current is only a single estimate
within the range possible values. The calculated arc current or the highest possible arc
current may not necessarily produce the highest incident energy to which workers may be
exposed. The arcing time may depend upon the arc current due to the tripping
characteristics of the protective device. Therefore, the incident energy may be greater for
smaller arc currents if the contributing branch current of the arc current lies in the
inverse-time section of trip characteristics.
46
Table 4.5: Minimum and maximum tolerances for arc currents obtained from IEEE
1584 test data for confidence level of 95%.
Voltage
Enclosure Type
Minimum Arc
Current
LV
Open
-26.5%
26.0%
LV
Box
-26.9%
33.0%
MV
Open
-6.7%
10.2%
MV
Box
-20.8%
12.3%
When considering the range of the calculated arc current, the simplest way is to take a
tolerance. This tolerance is a percentage of the calculated arc current. The tolerance is
obtained from statistical analysis of the test data. The tolerances differ for IEEE 1584
method and NFPA 70E methods. For further description see Appendix A. Table 4.5
provides tolerances for IEEE 1584 arc current estimate for a confidence level of 95%. A
confidence level of 95% means that there is a probability of 95% that the arc current will
be in the tolerance range. To be more conservative, you could also take a confidence
level of 99%. This would widen the tolerance range.
Example
For a 0.48 kV equipment in open air, the calculated arc current (Iarc) using IEEE 1584
estimate was 40 kA. What is the possible range of the arc current?
From Table 4.5 the minimum and maximum tolerances of arc current are 27.5% and
+31.9% respectively.
Minimum arc current = Iarc * (100 + Min. Tolerance %)/100
= 40 * (100 26.5) / 100
= 29.4 kA.
= 50.4 kA.
If the exact arc gap (or gap between conductors) was used in obtaining the arc current
estimates, then further adjustments need not be applied. However, if the gap is an
average value or an assumed value, then the possible range of arc current may need to be
adjusted. Appendix A describes in detail the effect of gap on the arc current. The gap is
used in calculations in the IEEE 1584 equations. NFPA 70E does not take the gap into
account. Table 4.6 provides the sensitivity of arc current to gap. For every mm of
difference in gap the arc current value is modified by the sensitivity amount. The voltage
across an arc gap is roughly proportional to the length of the gap. Higher voltage means
higher arc power for the same arc current. Since the arc resistance is non-linear, the
47
resistance is not directly proportional to the gap length. Therefore statistical approach is
preferred in the evaluation of the effect of variation of arc gap length on arc current. The
procedure given below should be used only for small differences in arc gap, as with most
other sensitivity analyses.
Table 4.6 Sensitivity of arc current to gap for IEEE 1584 method
Voltage
Enclosure
Sensitivity (% / mm)
LV
Box
-1.0%
LV
Open
-0.7%
MV
Not Required
Example
The exact gap between conductors for various devices are not known. For low voltage
box, it was generally observed that the gap ranged from 25mm to 40mm with an average
of 32mm. There are two ways to deal with this. The first method is to obtain two
estimates for arc current, one for the least gap and the other for the highest gap. The
second method is to adjust the IEEE estimate for the average gap using the sensitivity
shown in Table 4.6. Let us say that the arc current for 32mm gap was found to be 40 kA.
Arc current for min. gap = Iarc * (1+sensitivity*(Min. gap Average gap)/100)
= 40 * (1-1.0*(25-32)/100)
= 42.8 kA.
Arc current for max. gap = Iarc * (1+sensitivity*(Max. gap Average gap)/100)
= 40 * (1-1.0*(40-32)/100)
= 36.8 kA.
If the variation in gap is small, then extensive analysis need not be carried out.
4.5.2.3
After calculating the range of possible arc current, it is necessary to check whether the
calculated values are within the practical range. Check for the following:
48
Upper limit: It is not possible for the arc current to be greater than the bolted fault
current because of the additional impedance of the arc. Therefore, after applying
adjustments for random variations and for gap variations, if the upper limit of the
range of arc current is greater than the bolted fault current, then discard that value
and take the bolted fault current as the upper limit.
Example
For a bolted fault current of 50 kA at medium voltage equipment in box, the arc
current was calculated to be 49 kA using IEEE 1584 equations. To account for
random variations tolerance data from Table 4.5 was applied. This takes the upper
limit of arc current to 12.3% greater than the estimated value. Therefore, the upper
limit of the arc current was calculated to be 49*(100+12.3)/100 = 55 kA. This is
higher than the bolted fault current (50 kA), and therefore, is not possible. Take the
upper limit of arc current as 50 kA.
Lower Limit: Arcs do not sustain when the current is very low. For 480-volt
systems, the industry accepted minimum level for a sustaining arcing fault current
is 38% of the available three-phase fault current21. Test data accompanying IEEE
Standard 1584 shows arc sustaining for 0.2 seconds at 0.208 kV at a current of
21% of bolted fault current. Table 4.7 shows the minimum arc current as a
percentage of bolted fault current obtained during tests. The lower limit of arc
current is not yet clear. Until further information is obtained, it may be reasonable
to use Table 4.7 as the cut-off minimum arc current as a percentage of the bolted
fault current.
Table 4.7: Adjusted minimum arc current as a percentage of bolted fault currents*.
Voltage (kV)
0.2/0.25
21%
0.4/0.48
21%
0.6
28%
2.3
51%
4.16
64%
13.8
84%
*The adjustment is based on maximum measured values taken normalized to the bolted
fault current.
4.5.3 Calculate Branch Currents Contributing to the Arc Current
This is done using the branch current contributions to the bolted fault current obtained
from section 4.4.2. To calculate the contributing currents to the arc fault, use equation
(4.1).
Ix,arc = Ix,BF * Iarc / IBF
(4.1)
49
Where,
Ix,arc = Current through branch x for arc fault
Ix,BF = Current through branch x for bolted fault
IBF = Bolted fault current.
Arc currents have been observed to be non-sinusoidal due to the non-linear nature of the
arc resistance. The harmonic contribution of different branches may vary, however, the
fundamental component can be approximated using the method describe above. It has
been observed that although the voltage waveform is highly distorted, the arc current
however has low harmonic content. Therefore the linear relation (4.1) is a reasonable
approximation.
Section 4.5.2 describes taking upper and lower bounds of the range of arc current. The
branch contribution must be calculated for each case. These are later used to determine
the trip time of protective devices.
50
involved. This must be done before obtaining the trip times of the protective devices
across the transformer.
EasyPowers integrated protective device coordination program automatically determines
the arcing time for each protective device, operating condition, and arcing current level.
Total integration saves you time and resources, and ensures the most accurate solution.
Typically, for any given current, protective devices have a tolerance about the specified
trip time. Many low voltage breakers and fuses specify the upper and lower limits of the
trip time for different current values. For such cases, the time-current curve looks like a
thick band instead of a single line. Relays typically show a single line for the TCC curve,
and specify the tolerance as +/-x% (usually 10% to 15%) somewhere in their product
literature. Some fuse curves provide only the average melting time or the minimum
melting time. Follow the guidelines provided below for determining the trip time.
TCC with tolerance band: Take the total clearing time (upper bound of the band)
corresponding to the branch current seen by the device.
Relays with single line curve: Find within the TCC data or the product literature, the
tolerance for trip time. Add the tolerance to the trip time obtained for the TCC.
Breaker opening time must be added to this value.
Fuse TCC with total clearing time: No adjustment is required since total clearing time
is what we need.
Fuse TCC with average melting time: Obtain the tolerance from the product
literature, TCC data or the manufacturer. Add the tolerance to the average melting
time obtained for the TCC. If tolerance data is not available, make an assumption
using data with similar devices. For most purposes, a tolerance of +/-15% should
suffice. IEEE 1584 suggests taking a tolerance of 15% when average trip time is
below 0.03 seconds and 10% otherwise. Some commonly used fuse curves have been
found to have a tolerance as high as 40%. If the tolerance is known to be small, then
additional computation can be ignored.
Fuse TCC with minimum melting time: Obtain the tolerance from the product
literature, TCC data or the manufacturer. Add the tolerance to the minimum melting
time obtained for the TCC. If tolerance data is not available, make an assumption
using data with similar devices. The tolerance may vary with the slope of the curve.
For smaller melting times the total clearing time may be 30% to 100% higher than the
minimum melting time.
Circuit breaker clearing time: The TCC of relay or trip unit accompanying the breaker
may or may not include the breaker clearing time. If the breaker clearing time is not
included in the TCC data, find the breaker clearing time and add it to the delay of the
trip unit. Breakers typically have a maximum clearing time of 3 to 5 cycles after the
trip coil is energized.
51
9 10
3 4 5 6 7 8 9 100
CURRENT IN AMPERES
2
3 4 5 6 7 8 9 1000 2
700
3 4 5 6 7 8 9 100000
700
500
400
500
400
300
3 4 5 6 7 8 9 10000 2
300
Fuse
200
200
100
100
70
70
50
40
Pickup
50
40
Electronic
Trip LV
Breaker
30
20
30
20
10
5
4
5
4
Total
Clearing
Time
Curve
.7
.5
.4
.3
TIME IN SECONDS
TIME IN SECONDS
10
.7
.5
.4
Instantaneous
Pickup
.3
.2
.2
.1
.1
Minimum
Time
Curve
.07
.05
.04
.03
.02
.07
.05
.04
Thermal
Magnetic LV
Breaker
.03
.02
.01
9 10
3 4 5 6 7 8 9 100
3 4 5 6 7 8 9 1000 2
3 4 5 6 7 8 9 10000 2
.01
3 4 5 6 7 8 9 100000
CURRENT IN AMPERES
The arcing fault currents close to pickup current of instantaneous trip function should be
examined closely. If the calculated fault current value is within the tolerance band of the
pickup, then there is a likelihood of the device not tripping at the expected instantaneous
value. In such cases, if the time-overcurrent function exists in the device then the trip
time for the time-overcurrent function should be taken for arc flash calculations.
It is also important to realize that any changes to the protective device settings can have a
major influence on the arc energy. If device or setting changes are made, the arc flash
calculations must be re-checked and appropriate changes made if necessary.
4.6.2 Trip time for multiple feeds
When a bus is fed from multiple sources, as shown in Figure 4.3, a fault at the bus may
cause a series of breaker operations. The actual fault current will change as the breakers
open, since the sources of power will be sequentially removed from the faulted bus.
Since the current seen by the relays will change over time, further calculations are
required to determine the actual trip time for each breaker. We cannot simply obtain the
trip time corresponding to a single branch current by looking at the TCC data. Protective
devices with time-overcurrent functions typically operate like an integrating device. That
means, the overcurrent or its function is integrated or "added" over time until the sum
reaches a predetermined trip value. This is when the relay trips. For details on how a
relay or fuse integrates the function of current, refer to literature on operation of
protective devices.
53
UTIL-1
GEN-1 BUS
4.
05
8
1.353
0.
01
6
7.
20
8
GEN-1 BUSOPEN
GEN2-BUSOPEN
1.608
GEN2-BUS
1.566
1.353
2.706
2.389
0.
02
7
2.389
1.353
UTIL-BUS
3.212
0.824
UTIL-BUS
GEN-1
GEN-1
GEN-2
(a)
GEN-2
(b)
UTIL-1
Connection Status
GEN-1 BUSOPEN
OPEN
3.
99
8
0.
04
6
3.998
0.000
UTIL-BUS
GEN2-BUSOPEN
Bolted Fault
Current (kA) at 5
cycles
7.208
(b) Generators
disconnected
4.058
3.998
GEN-1
(c)
GEN-2
Figure 4.3: Example showing multiple source fed bus fault and series of operations
with fault current changing with breaker operation.
54
3. Upper bound of arc current due to random variations and its associated trip time.
4. Multiple feed scenarios:
a. Evaluate incident energy for each type of possible connections as noted in Step 4.
b. Evaluate incident energy for arc current changing through series of breaker
operations, as described in step 6. An example is presented below.
Example:
In the example shown in Figure 4.3(a), a 3-phase fault at GEN2-BUS results in a 7.2 kA
bolted fault current. NFPA 70E-2003 ROP Annex C method is used for this medium
voltage arc in box. Working distance of 18 inches is assumed. The contributing branch
currents are shown in Table 4.8. The branch currents I1, I2 and I3 respectively are GEN2
current, current from GEN-1 BUS to GEN-2 BUS and current from UTIL-BUS. The
generators are disconnected in 0.1 second. This reduces the bolted fault current to 4.05
kA. Next, the GEN-1 BUS is disconnected from GEN-2 BUS at 0.2 seconds. The bolted
fault current drops further to 3.99 kA. The last breaker trips at 0.3 seconds. In this
example definite time delay functions have been used to obtain the trip time, only for the
sake of simplicity. In many cases the delays are of inverse-time function, and the arcing
time may be longer as the fault current reduces with sources being removed from the
faulted bus.
Table 4.8: Branch currents and trip times used for example shown in Figure 4.3.
Connection
Status
Bolted
Fault
Current
(kA) at 5
cycles
Arc
Current
(kA)
Branch Currents
(kA)
All
Connected
7.208
7.20
Generators
disconnected
4.058
4.05
Only one
feed
3.998
3.99
I1
I2
I3
T2
T3
0.4
1.35 2.70
0
3.99
0.5
The total incident energy needs to be calculated by adding the incident energies for each
sequence of operation as the sources are removed from the fault.
55
Table 4.9: Total incident energy for example shown in Figure 4.3.
Connection Status
Arc
Current
(kA)
Duration
(s)
Incident
Energy
(cal/cm2)
All Connected
7.20
0.2
2.6
Generators disconnected
4.05
0.2
1.4
3.99
0.1
0.7
Total
4.7
Alternatively, we could take a simpler and more conservative approach of using the
highest arc current and the total arcing time. If we take the highest arc current, 7.20 kA,
and the total fault duration of 0.5 seconds, the calculated incident energy will be 6.4
cal/cm2. For practical purpose, this small difference will not matter. However, the
difference may be significant when the protective devices have inverse-time
characteristics.
4.7.1 Tolerance of Calculated Incident Energy
The selected method may have tolerance about the calculated value of incident energy
because of the random nature of arcs. Different calculation methods will generally yield
different results. For IEEE 1584 equations use Table 4.10 to find a more conservative
estimate that accounts for the randomness of arcs. This table is based on further analysis
of test data accompanying IEEE Standard 1584, and is described in greater detail in
Appendix A. The maximum tolerance should be added to the calculated incident energy.
This procedure is suggested to minimize risk to workers since test data has been found to
have greater incident energy than that yielded by IEEE 1584 formula.
Table 4.10: Tolerances for IEEE 1584 incident energy estimates.
Voltage/ Type of Enclosure
66%
85%
63%
64%
93%
54%
50%
52%
a. This is using the arc current after adjusting for random variations (upper and lower bounds); b. This is
using the arc current from IEEE 1584 formula.
56
Table 4.10 shows incident energy tolerances for two different kinds of arc currents. It
can be seen that there is not much difference in incident energy for arc in box whether the
exact IEEE 1584 arc current or the arc currents adjusted for random variations is used.
The choice of arcing current significantly affects only the arcing time, which in turn
affects the incident energy. From the table it can be observed that the calculated incident
energy is higher than maximum measured values for low voltage open air. However, for
other cases, the estimate may be much lower than the maximum measured values.
Example
The incident energy for low voltage in box was calculated to be 10 cal/cm2 using the
IEEE 1584 equations. The adjusted arc current was used in this estimate. What is the
maximum possible incident energy assuming random behavior of arcs?
From Table 4.10, the maximum tolerance is 63% of calculated value.
Maximum possible value of incident energy with +63% tolerance:
=10 * (100 + 63)/100 = 16.3 cal/cm2.
Energy Level
< 2 cal/cm2
5 cal/cm2
8 cal/cm2
25 cal/cm2
40 cal/cm2
Workers should prepare according to the risk category before commencing work or
inspection near exposed, live conductors. Documentation and warning labels are also
required. Although the incident energy itself may provide a more accurate picture of the
risk, the scale of 0 to 4 for the risk category may convey more meaningful information to
most workers. Employers are required to perform a complete hazard assessment prior to
commencing any work near exposed conductors.
57
where,
DB = distance of the boundary from the arcing point (see note)
D = working distance (see note)
E = maximum incident energy at working distance in cal/cm2
EB = incident energy at boundary, usually 1.2 cal/cm2 for arcing time > 0.1s.
x = distance exponent factor (see Table 4.12)
Note: Distances DB and D must both be in the same units. They can be expressed in
inches or mm.
Table 4.12: Distance exponent "x"
Enclosure Type
IEEE 1584
NFPA 70E
2000
1.9593
Switchgear
1.473
1.641
Cable
Box (0 0.6 kV)
58
2
1.4738
1.6
0.77
Example
The incident energy for a low voltage switchgear at a working distance of 18 inches was
found to be 25 cal/cm2 using the proposed NFPA 70E (2004) method. What is the flash
protection boundary for arcing time greater than 0.1s?
Flash protection boundary is:
DB = 18* (25 / 1.2)1/1.6 = 120 inches.
59
6. If software was used, the name of the software and the version.
7. The results incident energy, hazard/risk category and flash protection boundary
for every equipment.
8. If various modes of operation are possible, document assessment for each mode.
The assessment report should be available to all concerned persons. Some of these may
be:
1. Safety coordinator.
2. Safety division/department.
3. Foremen and electricians.
4. Electrical engineer.
5. Affiliated contractors.
4.10.2 Documentation in One-Line Diagrams
Figure 4.4 shows an example EasyPower one-line diagram. This is the LV part of a
substation showing the results of arc flash hazard assessment. The computation and
drawing was performed by the commercial integrated software EasyPower. Four
circuit breakers are located in the same switchgear lineup. A person working on the
switchgear should inspect the drawing for the arc flash incident energy levels and the
hazard category on all the exposed conductor parts. In this example, the line side of the
main breaker of the panel would produce the highest incident energy if arc flash were to
occur. The arc energy released would depend on the upstream protective devices, relay
R-TX-2 and breaker main breaker. Arrows should be placed to indicate the side of
breaker (line side versus load side) for which arc flash values are noted in the diagram.
This would provide the workers the knowledge of risk at each part of the panel.
The following steps are recommended for a practical documentation of arc flash data on
one-line diagrams:
1. Place the arc flash hazard assessment results on every equipment that poses a risk.
2. Specify the flash protection boundary, also known as arc flash boundary (AFB).
3. Specify the incident energy at the estimated working distance in the standard unit, for
example in calories per cm2. Specify the estimated working distance as well.
Workers should check whether the working distance will be maintained while
working on live equipment. If closer working distances are required, then it may be
necessary to revise the assessment to reflect true working condition. The closer the
distance the more the incident energy, and higher the risk.
60
TX-2
R-TX-2
BUS-4
M-1
MCC-1
78" AFB
10.5 cal/cm2 @ 18"
#3@18"
27" AFB
2.8 cal/cm2 @ 18"
#1@18"
M-1
MCC-1
78" AFB
10.5 cal/cm2 @ 18"
#3@18"
MCC-2
78" AFB
10.5 cal/cm2 @ 18"
#3@18"
MCC-2
78" AFB
10.5 cal/cm2 @ 18"
#3@18"
61
9. If there are various possible sources or interconnections, clearly mention in the oneline, which source is connected and/or which breaker is open or closed. Workers
should first determine if the assumed condition in the diagram reflects the condition
of the power system at the time of work. If the system conditions are different from
those for which the assessment was performed, it is necessary to revise the
evaluation.
4.10.3 Documentation on Equipment
Three types of documentation are recommended for arc flash hazard assessment results
placed on the equipment:
1. Warning labels with arc flash values: Permanent stickers with a warning sign of
adequate size. The label should be located in a place that is easily visible and
readable from some distance. The flash protection boundary and its units, the
incident energy at the estimated working distance and its corresponding risk category
number must be clearly printed on the label. Additional information that is useful for
future revisions are the date of assessment, the method of calculation, and the
software name and version. Warning labels should also be placed just outside the
flash protection boundary, so that workers may see it and prepare accordingly before
they enter the hazardous area.
Figure 4.5: Example of arc flash warning label printed from EasyPower
2. Arc flash assessment results, in the form of a table and a small one-line diagram
as described in the previous section, should be placed on the equipment at a spot
which workers can easily access.
3. Large multi-section equipment may be labeled at various sections of the
equipment. This facilitates hazard communications. Different sections may have
62
different potential arc flash energies. If the same label is to be used on all
sections, the highest possible incident energy must be specified. For example, a
large transformer can have higher incident energy on the low voltage terminals
than on the high voltage terminals. If the incident energy differences are high,
different labels can be placed. This avoids workers having to wear extra PPE
while working on terminals with less incident energy.
NEC, NFPA and IEEE do not specify the details of the information to be placed on arc
flash warning labels. The labels may be more or less detailed than the one shown in Fig.
4.5. The label can be as simple as Warning Arc Flash Hazard. It is up to the
facilities management to decide on detail desired. Arc flash calculations provide the
maximum expected incident energies. There are a number of tasks that can be done
around electrical equipment that does not require the maximum PPE to be worn.
Referring to NFPA 70E-2000 Table 3.3.9-1 the task of reading meters with the door
closed is classified as risk category 0 (zero). While NFPA lists racking in a breaker as
risk category 3, if calculations give the maximum PPE as category 4, then risk category 4
should override NFPA risk class 3. Table 4.13 below is a summary of the NFPA task
table.
63
Energized Equipment
1-15 kV switchgear,
motor starters
Task
Risk Category
Breaker or switch
operating with
door/covers closed
Breaker or switch
operating with
door/covers open
Max Calculated
Max Calculated
Max Calculated
Breaker or switch
operating with
doors/covers closed
Breaker or switch
operating with
door/covers open
Max Calculated
Max Calculated
Max Calculated
Max Calculated
18
"IEEE Red Book" - IEEE Recommended Practice for Electric Power Distribution for Industrial
Plants, ANSI/IEEE Std 141-1986, IEEE, 1986.
20
21
NFPA 70E May 2003 ROP, "Standard for Electrical Safety Requirements for Employee
Workplaces", 2003 Edition, page 57.
64
65
66
67
System Configuration
0.
48
N.C. TIE
MAIN-2
kV
MAIN-1
0.
48
kV
Reducing the fault level depends on the existing system configuration. Double-ended
load centers with a normally closed tie (Figure 5.1) is a prime example where the fault
level can be reduced by either opening the tie or one incoming breaker. The fault current
will be reduced by approximately 50% and the incident fault energy will also be reduced,
although not necessarily in the same proportion. If the bus has two sources or a source
and a normally closed tie as shown in Figure 5.2, opening one of the sources (or tie) will
reduce the fault level while maintenance is done on the equipment. For both situations,
the loading and relay setting should be checked to make sure that the opening of a
breaker does not overload the other source.
Source 1
Source 2
N.C.
Feeder 1
Feeder 2
Feeder 3
Tie
Current limiting fuses/breakers introduce additional resistance within the fuse element
while the fuse is melting. This limits the fault current. Fault currents within the currentlimiting region of the fuse are cleared quickly, usually within half a cycle. Since the
68
incident arc energy is proportional to arcing time, current limiting fuses/breakers limit the
arc energy.
5.2.1.3
Current limiting reactors introduce additional impedance in the system and are used to
limit the fault current. This not only reduces damages caused by faults but also allows the
use of circuit breakers with smaller duty. Limiting the fault current can also increase the
fault clearing time if the fault current happens to lie in the inverse time delay
characteristics of the protective relays. Therefore, protective device coordination analysis
is also required when selecting current limiting reactors.
5.2.2 Reducing Arcing Time
Arcing time can be reduced in several ways. Some changes in the system of settings may
be required for this purpose. Some strategies outlined in this section are as follows.
1. Reducing safety margin for relay and breaker operation with improved solid state
trip devices.
2. Bus differential protection to combine selectivity with instantaneous operation.
3. Temporary instantaneous trip setting during work.
4. Retrofit time-overcurrent relays with delayed instantaneous trip device if needed.
5. Optical sensor to trip breaker in the event of arc flash.
6. Use smaller fuse size if possible; smaller current limiting fuses may clear faster.
Fuses will generally be much faster than breakers at high fault currents even
ignoring current-limiting effect this can greatly reduce arc energy.
7. Protective device coordination study to balance improving reliability with
reducing arc flash hazard.
5.2.2.1
Incident energy increases with time and fault current. Reducing either or both lowers the
incident energy due to an arcing fault. Faster acting relays and trip devices can reduce
the arcing time to some degree. In this regard, a protective relaying review may be
performed in order to see if they can be lowered in time and pick-up. If a protective
device study was done a number of years ago when electro-mechanical relays were the
norm, 0.4 second margin between relay was common. This allowed for breaker operating
time, over-travel, and a time safety margin. Breaker times are now commonly 5 cycles
rather than the 8 cycles of older breakers. Microprocessor relays are now being used, for
which the over travel has essentially been eliminated. The repeatability of the
microprocessor relay is better than that of the electro-mechanical relay. Therefore, the
safety margin can be reduced. The end result is that the relay coordination margins can
69
be 0.2 to 0.25 seconds instead of 0.4 seconds. This is a 25%-35% reduction in arc energy
exposure.
5.2.2.2
Replacement low-voltage trip devices from Satin, Joslyn, Carriere and possibly others,
have an instantaneous unit that that can be turned on or off. This has a high advantage on
the incoming main breakers. In many cases, for coordination purposes, the instantaneous
is not set and fault clearing times are delayed for selectivity. A main breaker clearing
time with a load center tie and feeder breakers could easily have a short time setting of
0.4 seconds. If the instantaneous trip could easily be enabled while work is being
preformed lower fault currents could be tripped and cleared in less than 0.04 seconds.
The incident energy exposure is reduced to 10% of its previous value. During
maintenance, full selectivity of devices may be lost, but the reduction in arc flash
exposure makes it worthwhile. The temporary instantaneous setting should be disabled
and the original protective setting should be restored for normal operations after the work
is completed. Separate instantaneous trip devices with increased protection can also be
added to shunt trip or transfer trip for added protection during work procedures.
5.2.2.4
If bus-differential relaying is not possible then the main relay can be retrofitted with an
instantaneous protective device and a safety control switch. As shown in Figure 5.3, a
selector switch can be used to place the instantaneous in service when maintenance is
being done. Normally the instantaneous protection would not be functional due to the
open contact of the selector switch. However, when work is being done on the energized
equipment, the safety switch would be turned ON and thereby limiting the arc exposure
time to the worker should an arcing fault accident occur. The delayed fault clearing time
could be in the range of 0.4 to 2.0 seconds on the main breaker instead of 0.1 second.
The delayed trip time greatly increases the arc exposure time and amount of radiation a
worker would receive if the arc blast pressure were not enough to propel the worker away
from the fault. The time-selective protection system would be eliminated for duration of
the work in the interest of safety. The selector switch should be lockable in the
maintenance position. Ideally, positive feedback from the trip unit would be used for an
indicating light associated with the switch to confirm the setting change was in effect.
70
Many medium voltage multifunction relays have provisions for different protective
settings for various operating modes. For example, one group of setting is used for
normal operation; a second group of settings is used for emergency mode. Another group
setting could be for maintenance where the tripping and current pick-up settings are
reduced and set as instantaneous. Again, these temporary settings could result in the loss
of selectivity with a gain in human protection.
51
1200/5
Relay without
Instantaneous
Setting
MAIN BREAKER
SWITCHGEAR BUS
FEEDER 1 BREAKER
1200/5
FEEDER 2 BREAKER
50
51
1200/5
Relay with
Instantaneous
Setting
50
51
Relay with
Instantaneous
Setting
Lt
Inst. Overcurrent
Trip Contact
To DC Bus
To Main Breaker
Trip Circuit
Figure 5.4 shows the possible time current curves of a load center. The relay operating
times for both the 100% and 85% fault currents are shown in Table 5.1. If an
instantaneous trip were set on the main breaker at 3 times the long time pick-up, a fault
on the bus would be cleared in approximately 0.05 seconds with the incident energy
being approximately 1.0 cal/cm2 instead of the 4.4 and 5.3 cal/cm2 as shown in Table 1.
71
3 4 5 6 7 8 9 10000
1000
700
700
500
400
300
500
400
300
4160V
250E
200
200
100
1000KVA
70
50
40
30
20
TIME IN SECONDS
10
7
5
4
3
480V
20
600A
.1
.07
Transf Amps
600A
GE RMS-9
Sensor = 800
Plug = 600
Cur Set = 1 (600A)
LT Band = 1
STPU = 4 (2400A)
ST Delay = Min
Inst = 10 (6000A)
.7
250E
S&C
SMU
SMU-40
250E
.5
.4
.3
.2
.1
.07
.05
.04
.03
.05
.04
.03
.02
.02
.01
.5 .6 .8 1
3 4 5 6 7 8 9 10
3 4 5 6 7 8 9 100
3 4 5 6 7 8 9 1000 2
72
10
5
4
3
.7
.2
.5
.4
.3
50
40
30
1600A
1600A
GE RMS-9
Sensor = 1600
Plug = 1600
Cur Set = 1
(1600A)
LT Band = 1
STPU = 2 (3200A)
ST Delay = Min
100
70
TIME IN SECONDS
.5 .6 .8 1
1000
.01
3 4 5 6 7 8 9 10000
Table 5.1: Clearing times and Incident Energy for a 1000-kVA load Center
Main
Feeder
Fuse
Clearing Device
Main
Feeder
Fuse
5.2.2.5
kA at 480V
Operating Time
17.3
0.21
23.2
0.05
17.3
0.42
85% Calculated Fault Current
kA at 480V
Operating Time
14.3
0.30
19.7
0.05
14.3
1.0
Cal/cm2
4.4
1.4
8.8
Cal/cm2
5.3
1.2
17.6
ABB has an Arc Guard System TVOC which has a light sensor to detect an electrical
arc flash. It can be activated by light only or light input supervised with an overcurrent
detector. Its output is used to trip a breaker and has an operating time of 10 milliseconds.
If auxiliary tripping relays are needed to trip several breakers at once, then the auxiliary
relay time needs to be factored in to the total clearing time. Placement of the detector and
its control wiring could be critical. These should be placed close enough to detect an arc
but not be damaged by the initial arc rendering the protection useless.
5.2.2.6
Fuse sizes could be reviewed to determine if smaller fuses can be used. Smaller fuses
reduce the exposure time. This can be significant when the arcing current or 85% of
arcing current is not in the current limiting range of the fuse. Referring to Fig. 4, while
the 250E fuse satisfies the NEC, a smaller 175E fuse would also satisfy the NEC. The
smaller fuse would operate quicker and reduce the arc energy exposure, should the main
breaker fail or should a fault occur between the transformer and main breaker. Speed of
fuses are selected to coordinate with other protective devices and the over-current
capacity of equipment being protected. A disadvantage of lowering the fuse size is the
possibility of fuses not being able to discern a temporary fault from a persistent fault. A
temporary fault, such as those found in overhead distribution lines, exist for a few cycles.
Some fuses are selected such that they allow temporary faults but interrupt persistent
faults. If the fuse size is lowered with the intent of reducing arc flash hazard, then the
fuse may melt upon temporary faults, thus reducing the reliability of supply.
Operating fuses can create sparks and may lead to arc flash accidents. Fuses should not
be temporarily lowered just for the purpose of working on live line.
73
It has to be recognized that the act of changing protective settings on electrical equipment
could place the workman in jeopardy. While the protective devices are at low voltage a
spontaneous fault could occur in the switchgear at this time. Most relay resetting are done
with a keypad and not with screwdrivers, the chance of a fault at this time is extremely
low.
Review protective devices to see if they can be lowered in time and pick-up. Due to
reliability reasons using temporary settings is usually not a preferred practice. Tampering
with settings of protective devices is prohibited. However, if a qualified person, for
instance the engineer, can temporarily provide the alternate settings during the work
period, then the incident energy can be reduced by lowering the trip time.
5.2.2.8
A protective device coordination study is carried out to improve system reliability. This
study can be done on a regular basis, perhaps every few years or whenever there are
changes in the system. Such studies could also include as one of its goals, the reduction
of incident energy from arc flash. The engineer performing the study should
simultaneously evaluate the arc flash hazard, and seek to minimize the hazard by keeping
the arcing time as low as possible.
5.2.3 Remote Operation and Racking
Placing distance between electrical conductors and the worker greatly reduces the arc
incident energy and the arc blast force. The reduction is not linear. For example, a
worker twice as far as another worker from the arc will receive 25 to 50% less energy
than the closer worker. New high voltage equipment can be ordered with the breaker
Open and Close switches remote from the breaker unit. These could be placed on a
non-breaker unit, in a separate control panel, or in a remote room. Older switchgears can
be retrofitted with remote control switches.
New microprocessor-relays can be programmed to supervise manually the closing of a
breaker using a punch and run time, that allows the operator 3 to 10 seconds after
initiating a close to evacuate the vicinity before the breaker is actually closed.
While fully electrically operated low voltage breakers are available they are not the norm.
Low voltage breakers that are fully electrically operated would be useful for remotely
located control switches. As the insurance companies and OSHA begin to demand better
arc flash safety measures, fully operated electrically low voltage breakers may become
more common.
Placing a breaker in or out of a switchgear cubicle exposes the worker to a possible arc
flash hazard. While the breakers mechanical indicator may note that the breaker is fully
open, there have been cases where it was not open due to contact or indicator failure.
Placing a breaker in a cubicle when it is not in the fully open condition can result in an
74
arc. While the distance from live conductors to the worker can be over an arm's length
away, the arc gases can flow around the breaker and result in burns. For breakers that are
being withdrawn from a cubicle, check the following three items before withdrawal the
mechanical indicator shows the breaker open, the breaker indicating lights show the
breaker open, and the ammeter shows all three phases with zero current.
Using a longer operating arm to rack in the breaker can provide the needed distance.
Remotely controlled breaker racking mechanisms are available for some breakers as part
of the new equipment or as retrofits.
Placing a barrier such as a closed door or a portable shield as shown in Figure 5.7 would
limit the arc flash exposure. While the shield as shown would help remove direct arc
burns, radiant energy burns are still possible and PPE is still needed. With a shield the
surface area is increased, therefore making the force exerted by the arc blast more of a
concern.
Switchgear
Cubicle
High Impact
Plastic Shield
with Arm
Breaker
Although not a way to reduce arc incident energy, it is good practice to use the buddy
system. In the event an incident should happen, help can be summoned quickly if a
second person is around.
22
Some of the authors inspected a substation of an irrigation pump in Eastern Oregon in January,
2003. The switchgear had been damaged by arc-flash. It was observed that both rodents and
birds had inhabited the MCC/panel. The leads to the primary side of the potential transformer had
snapped and touched the metal enclosure creating sparks. The arc traveled from the mains side
of the 4.16 kV copper bus bar towards the remote end and melted the bus bar butts and the steel
sheet cover. The recloser at the utility substation tripped several times.
75
77
6.1.1 OSHA
Table 6.1: Various OSHA standards on personal protective equipment
1910.132(a),
1926.95(a)
Application
What?: Protection shall be provided for eyes, face, head and extremities.
When?: Whenever it is necessary by reason of hazards of processes or
environment, chemical hazards, radiological hazards, or mechanical irritants in
a manner capable of causing injury or impairment in the function of any part of
the body through absorption, inhalation or physical contact.
How?: Protective clothing, respiratory devices, protective shields and barriers.
1910.132(b),
1926.95(b)
1910.132(c),
1926.95(c)
1910.132(d)
1910.132(e)
1910.132(f)
Training: When PPE is necessary, what PPE is necessary, how to properly use
the PPE, and how to care, maintain and dispose the PPE. Each affected
employee shall demonstrate an understanding of the training. Retraining may
be required depending upon changes in workplace or PPE. The required
training shall be certified and documented.
1926.100(a)
1926.100(b)
1926.100(c)
78
All employees within the flash protection boundary are required to wear PPE.
FR clothing should be worn when the estimated incident energy at the body may
cause a second degree (curable) burn (1.2 cal/cm2 for arc time greater than 0.1
second or 1.5 cal/cm2 for arc time 0.1 seconds or less).
If incident energy exceeds 4 cal/cm2, heavy duty boots are required to protect the
feet.
6.1.3 ASTM
American Society for Testing and Materials (ASTM) develops standards that specify the
quality of various materials including safety materials such as PPE. The following
standards are applicable to arc flash hazard protection equipment.
ASTM F1506: Standard Performance Specification for Textile Materials for Wearing
Apparel for Use by Electrical Workers Exposed to Momentary Arc and Related Thermal
Hazards, 2002. This standard specifies the requirements for flame resistant clothing.
There are three basic requirements in this standard:
79
a) The fabric under test must self-extinguish in less than 2 seconds after the ignition
source has been removed.
b) Char length for ASTM Test Method D6413 must be less than 6 inches. A fabric
specimen of 12 inch length is hung vertically in an enclosed space and the bottom
is exposed to a methane flame for 12 seconds. The length of the fabric destroyed
by flame is the char length. This test is also known as the standard vertical flame
test.
c) Apart from meeting these pass/fail tests, the fabric is also tested for the Arc
Thermal Performance Value (ATPV) as per ASTMF1959. Manufacturers are
required to report the test results to the end users of the material as an Arc rating
on a garment label.
Any fabric that meets the ASTM F1506 complies with OSHA 1910.269. This
performance specification does not cover coated fabrics commonly used in rainwear.
ASTM F1959: Standard Test Method for Determining Arc Thermal Performance (Value)
of Textile Materials for Clothing by Electric Arc and Related Thermal Hazards. This test
determines how much incident energy is blocked by the fabric before the wearer of the
protective clothing may get a second degree burn. The amount of energy blocked is
reported as Arc Thermal Heat Performance Value (ATPV). If the fabric breaks open the
value is also called the Breakopen Energy Threshold.
ASTM F1891: Standard Specification for Arc and Flame Resistant Rainwear. See
ASTM F1506 for the three basic requirements.
Any fabric that meets the ASTM F1506 complies with OSHA 1910.269.
ASTM F1449: Standard Guide for Care and Maintenance of Flame, Thermally and Arc
Resistant Clothing. This guide provides recommendations for the care and maintenance
of clothing that is flame, thermal and arc resistant. The standard focuses on the industrial
laundering process and also identifies inspection criteria that are significant to the
performance of clothing.
80
Weight: The weight of FR fabric is specified in weight per unit area (ounces/square yard
or g/m2). Higher weights provide more thermal insulation.
Layers: Multiple layers of clothing retain air space between the layers, thus providing
greater thermal insulation than a single layer. Single, thick clothing provide less physical
comfort, whereas multiple layers allow flexibility. Comfort and flexibility are important
in avoiding accidents while working on live equipment.
6.2.2 Care of FR Clothing
Laundering: Obtain complete instructions on care of FR clothing from the manufacturer.
Some cleaning chemicals such as chlorine bleach may affect the finish, reduce the fabric
strength and remove the color of the cloth. Some manufacturers claim that the flame
resistance property is not affected by the bleach23. Follow laundering instructions
provided by the manufacturer.
Contamination: Grease, oil, or other flammable materials catch fire easily and will
continue to burn even after the arc ceases. Therefore FR clothing contaminated with these
substances should not be used. Care should be taken at work to avoid contaminating FR
clothing from such materials.
Storage: The clothing should be stored in a safe condition so that it is reliable.
6.2.3 Useful Life of PPE
The useful life of a PPE may depend on various factors such as the material with which it
is made, the severity of work activity and the abrasion resistance characteristics of the
PPE. Obtain information from the manufacturer to determine the useful life.
The useful life of a PPE is normally stated following some assumptions. It must be
remembered that if the actual conditions are different from these assumptions, then the
stated expected life may not be applicable. It is best if the PPE user obtains from the
manufacturer, the useful life of the PPE for the intended use.
Fabric
Industrial
Launderings
36-50*
18-24#
60-80*
28-38#
81
PPE should be selected according to the needs of the worker and the nature of work
performed. Some of the factors are discussed below.
Comfort: It is vital that the worker is not uncomfortable. Otherwise there could be a risk
of accidents occurring. Comfort is important both physically and mentally. PPE for high
incident energy (hazard/risk category #4 or greater) may have thick and heavy clothing,
headgear and gloves. The comfort level may differ from one individual to another. It is
necessary to ensure that each worker feels as comfortable as possible, wearing the PPE.
Different workers may find different materials more comfortable than others. It may be
beneficial to let the workers try out the PPE to make sure that it is satisfactory in terms of
comfort. It may take some time before a worker adjusts to new PPE. Therefore, it is
recommended that the worker practice wearing the PPE before working on live exposed
equipment. This also ensures that the PPE does not interfere with the task.
Fit: A loose fitting PPE provides more thermal insulation through the air trapped inside.
However, it should not be too tight or too loose so as to interfere with the task.
Layers: As mentioned in the previous section, multiple layers provide additional air
insulation and greater degree of protection. Multi-layer FR clothing is also more
comfortable than a single layer of thick and heavy clothing.
Materials: Choice of fabric material can affect both comfort and weight. There are
different types of treated cotton and synthetic fabric available from various
manufacturers. For multi-layer clothing, the workers may choose to have untreated
flammable fabric such as cotton or wool for inner garments at lower incident energies.
Abrasion Resistance: Some FR clothing is available with high abrasion resistance
quality. Employees who do heavy duty work should use this kind of PPE. Clothing
without such quality can be easily damaged, and may fail to adequately protect the
worker from an arc flash.
82
(a)
(b)
(c)
Figure 6.1: FR clothing (a) jacket; (b) bib overall; (c) complete flash suit (courtesy
W.H. Salisbury & Co.)
Vest/Undergarment: These can be worn underneath shirts, jackets or pants. They
provide an extra layer of protection. Multi-layered clothing is more flexible, easy to
work with and has trapped air to provide additional thermal insulation. Combination of
vest/undergarment with a shirt increases the total arc rating.
Shirt/Pant: FR shirts and pants can be used for incident energy of 4.0 cal/cm2 or below.
These can be multi-layered for higher arc rating.
Bib Overall: Bib overalls worn with a shirt provides higher protection to the chest area
than a shirt worn with a pant. See Figure 6.1 (b).
Coverall: Coveralls are equivalent to shirt and pant.
Jacket: These are usually multi-layered and are like multi-layered shirts. See Figure 6.1
(a).
Hood: The hood is part of the headgear, has face protection and has FR fabric covering
the head, ears, neck and shoulders.
83
84
Proof Test kV
Class 00
500
2.5
Class 0
1,000
5.0
Class 1
7,500
10
Class 2
17,000
20
Class 3
26,500
30
Class 4
36,000
40
Figure 6.3: Gloves (rubber and leather) and boots (courtesy W.H. Salisbury & Co.)
Boots: Heavy duty shoes25 or boots should be worn where incident energies are higher
than 4 cal/cm2.
Hot Stick: Hot sticks are use to operate fuses and switches. These provide insulation
from the high voltage parts. They also allow the worker to maintain increased working
distance, so that the incident energy is less.
Arc Suppression Blanket: This provides a barrier from arc flash.
Ear Muffs: Arc blast can cause severe ear injuries. Ear muffs should be worn to provide
sound insulation and reduce the impact.
Mechanical Barriers: As mentioned in the previous chapter, mechanical barriers can
provide protection from thermal radiation as well as from blast pressure. They can be
used for racking breakers, but are not suitable for most other work.
23
Indura, Nomex.
24
Westex Inc., Product Literature: INDURA Ultra-Soft Flame Resistance Fabrics, September
2002.
25
85
7.2 Training
Training must provide people the knowledge and understanding of the existence, nature,
causes and methods to prevent electrical hazards. The training should also include the
selection and use of appropriate PPE.
As part of regular electrical safety training it would be beneficial to include the following
arc flash related topics. Special arc flash hazard sessions are recommended for
introductory training exercises.
7.2.1 Awareness
b. Existence of the arc flash hazard: Arc flash accidents are not as common as electrical
shock. Therefore, many are not aware of the hazard. Trainers and managers need to
place adequate effort in trying to convince the workers that arc flash hazard is indeed
something to take seriously.
c. Causes: Knowing the causes helps immensely in avoiding the hazard.
d. Nature of arcs: This can relate to the degree of potential damage and possible ways to
reduce hazard.
e. Possible injuries/damage: Findings and statistics from various studies and reported
incidents reveal the gravity of the hazard.
f. Historical cases: Literature on arc flash incidents can be found in many documents.
Review of these cases is illuminating and convincing and is likely to influence
workers to consider taking measures to avoid arc flash injuries.
Chapter 1 provides a brief introduction on the first four topics. IEEE Standard 1584 and
numerous papers on electrical safety provide examples of historical cases. The
awareness training should not be limited to electrical worker only. Since it is the
responsibility of the employer to provide training, PPE and other means of minimizing
87
the hazard for the electrical worker, managers or the responsible persons should be
included in the training.
7.2.2 Standards and Codes
Standards and codes not only provide information on what employers and workers are
required to do, but also suggest solutions/methodology. In following the standards and
codes, compliance should not be the only motive. The following provide the standards
and codes. Some of the related topics are summarized in Chapters 3 and 6.
a. NFPA 70E
b. OSHA
7.2.3 Understanding of Arc Flash Quantities
Workers are expected to read signs/labels, drawings and tables to understand the degree
of hazard a worker may be exposed to. Some of these pertain to the rating of the required
PPE or protection boundaries. It is important that workers understand the following
quantities and their units. An understanding of the physical significance of these
quantities is helpful.
a. Flash protection boundary
b. Working distance
c. Incident energy
d. Hazard/risk category
e. ATPV of PPE.
7.2.4 PPE
Chapter 6 provides some information on PPE. More detailed information can be obtained
from PPE vendors. Use of PPE can restrict visibility and movement, cause discomfort,
and slow down the work. Practice is recommended with PPE before working on
energized equipment. The following topics should be included in training on PPE.
a. Selection of PPE
b. Information/Labels on PPE
c. Training with new PPE
d. Inspection, care & maintenance
e. Useful life & disposal
f. Documentation of use and maintenance of PPE
g. Limitations and potential risks using PPE
h. Limitation of PPE & degree of protection provided
88
Warning signs and labels are part of any safety program. When new labels or signs are
placed, workers need to be able to understand the meaning, and follow instructions
precisely.
7.2.6 Methods to Reduce Risk While Working on Live Exposed Parts
Chapter 5 provides some methods of reducing risk. Many procedures can be developed
in-house to suit the system and type of work the workers may need to perform. Practice
on de-energized systems will greatly augment the knowledge, and make the work safer.
New procedures are introduced during the initial stages of the arc flash program with the
help of experienced workers and safety professionals. Formulation of new procedures as
an ongoing process is expected. Clear communication between different workers who
are part of procedure formulation expedites resolving safety issues.
7.2.7 Arc Flash Hazard Assessment
Arc flash hazard assessment should be carried out by skilled and experienced
professionals. In-house engineers can be provided with the necessary training. Chapter 4
provides practical steps to performing an assessment. It is necessary for the engineer to
apply short circuit analysis and protective device coordination, along with arc flash
energy calculations. Engineers should be aware of the limitations of the standards or
methods they employ and also have a good understanding of how these limitations can be
overcome. It should be noted that only some of the practical issues are discussed in this
book because of time limitations. AFH programs, calculations, and procedures are new
to the industry and changing at a rapid pace. Participation in seminars and forums,
reading of publications, and peer discussions are recommended.
7.2.8 Documentation
Documentation is described in greater detail later in this chapter. Workers should be able
to document any changes performed in the power system, update the single-line
diagrams, and also make note of any discrepancy between the actual equipment/system
interconnection and the single-line diagrams. Workers should also understand the
consequences of the drawings not truly reflecting the actual system condition.
Illustrations can be provided as to how such circumstances may lead to selection of
wrong PPE or wrong procedures.
89
As part of the regular safety audit, the following arc flash hazard related points should be
examined.
1. System changes: Since the last audit, were there any changes in the power system? If
"yes":
a. Was the change properly documented? Were drawings updated?
b. Were any new arc flash calculations performed?
c. Were new warning labels installed at the affected locations?
2. PPE: Do the employees have the PPE needed for the highest level of arc flash
exposure in the facility? What is the condition of PPE?
3. Do workers follow arc flash hazard warning labels, signs and instructions regarding
flash protection boundary and appropriate PPE?
4. Are arc flash hazard procedures followed correctly as they are documented?
5. Do the existing procedures provide protection adequately?
6. Do workers have adequate knowledge and training in arc flash hazard?
7. Have protective device settings been tampered with or modified from the intended
settings? Are the fuse sizes the same as those specified by the engineer?
8. If any equipment requires special operation/maintenance procedures for safety
reasons, is the information readily available to workers? What is their understanding
of the special procedures?
9. Do workers have the tools and equipment for working safely? (Insulated tools,
shields, hot sticks, etc.)
10. Review of accidents and near misses.
90
Work to be performed.
Potential hazards.
Individual responsibilities.
Regular safety meetings are also conducted in companies in which workers are
continually exposed to hazards. Monday morning meetings are held briefly to talk about
safety. Accidents and near misses are discussed. Safety meetings are instrumental in
disseminating information, bringing new issues to attention, and discussing possible
solutions. The Monday morning meetings also provide the workers an opportunity to
focus on safety matters after returning to work from their weekend. Workers should be
encouraged to discuss openly and share their ideas. For example, if a worker drops a tool
on exposed live equipment, it should be discussed, since this event could have led to arc
flash. This discussion could lead to reasoning why the tool was dropped, and whether a
better method or tool would have avoided the incident. From these insights, the
procedures for arc flash hazards can be enhanced.
7.5 Documentation
1. Document all data that was used for the arc flash hazard assessment. This is useful
for implementing changes and future assessment.
2. Prepare a report of the assessment identifying the type, name/ID, incident energy at
working distances, flash protection boundary, hazard/risk category, and other
91
92
26
Ray Jones & Jane Jones, "Electrical Safety in the Workplace", page 147, National Fire
Protection Association, 2000.
27
93
Appendices
95
Table A.1 shows the variation of measured arc current from the IEEE 1584 estimate of
arc current for low voltages. For the test data conditions, the arc current was calculated
using the proposed formula. These values were compared against the actual measured arc
current. The measured values of arc current were found to range between -37.5% and
57.2% of the estimated values. Bare in mind that the equations presented in IEEE 1584
for estimating arcing current are empirical equations based on regression analysis. This
equation is a best-fit curve with an R-square of 98.3%. R-square is a measure of the
equation fit to the data; 100% is perfect28. Therefore, after calculating the arc current
from the equation it is necessary to consider the upper and lower limits of the arc current
due to the random nature of arcs. The upper and lower limits would render the arc
current as a possible range rather than a definite calculated value. Applying the variation
limits presented in Table A.1 may not be the practical approach. It is customary in
engineering practices to define probable ranges based on statistical analysis.
97
Table A.1: Variation of measured data about arc current estimated with IEEE 1584
equation for low voltage (includes both open and box)
Minimum
Maximum
0.2 / 0.24
-10 %
+57.2 %
0.4 / 0.48
-37.5 %
+23.4 %
0.6
-27.6 %
+27.4%
70
Frequency (Count)
60
50
40
30
20
10
0
-40% -30% -20% -10% 0% 10% 20% 30% 40% 50% 60%
Percent Variation
Figure A.1: Distribution of measured arc currents about IEEE 1584 equations for
low voltage arc currents from bolted fault currents.
Figure A.1 shows the distribution of measured arc currents as deviation from the IEEE
1584 estimate for low voltage, in percentages of estimated arc current. This histogram is
the result of further analysis of the data published and used by IEEE for the equations in
1584. For 166 arc tests, the mean deviation from estimated arc current was 2% and the
standard deviation of the variation from estimated arc current was found to be 15.2% of
estimated arc current. The distribution has a skewness of 0.236 and a kurtosis of 1.553.
Although the skewness and kurtosis suggest that the data is slightly off normal
distribution, it may be assumed that with more data samples, the random nature of arc
will follow normal distribution. Therefore, further statistical analysis will follow the
assumption of normal distribution.
Table A.2 presents limits for low-voltage arcing current for various confidence levels.
For a confidence level of 95%, the arc current can have any value between -23.0% and
+27.1% of calculated arc current. The confidence level is a measure of probability or
likelihood. A 95% confidence level implies that out of 100 random samples, 95 samples
98
of observations will be in the specified range, and 5 samples will be above or below the
specified range. Therefore, there is 95% probability that a measured arc current at low
voltage may have a value between 23.0% to +27.1% of the estimated arc current. The
higher the confidence level, the greater the range. It is up to the engineer to select the
confidence level based on how conservative the arc flash evaluation needs to be. A
confidence level of 99% may appear to be too conservative for practical purpose. It is
suggested that 95% be used. As a rule of thumb, the arc current can be considered to be
within the range of +/-25% of the calculated value from IEEE 1584 equations for systems
with voltages less than a 1000V.
Table A.2: Minimum and maximum likely deviations in low-voltage arc currents
from the calculated arc current for various confidence levels.
Minimum Maximum
Mean Standard Confidence
Arc
Arc
Variation Deviation Level
Current Current
2.0%
15.2%
95%
-23.0%
27.1%
99%
-33.4%
37.4%
90%
-17.5%
21.5%
68%
-5.1%
9.1%
Table A.3: Variation of arc currents from estimate for various confidence levels for
2.4 to 15 kV
Arc in
Percent
Variation from
Estimate (%)
Min
Box
Open
-38.9%
-4.2%
Max
Mean
Variatio
n
21.1%
-2.2%
7.6%
0.1%
Standard
Deviatio
n
8.9%
2.5%
Confidenc
e Level
Probable Arc
Current Range
Max
Max
95.0%
-16.8%
12.4%
99%
-22.8%
18.4%
90%
-13.6%
9.1%
95%
-4.0%
4.2%
99%
-5.7%
5.9%
90%
-3.1%
3.3%
For medium voltages, the variation of arc current from the calculated value may be less
than for low voltages, as shown in Table A.3. Arc in open air appears to be more
predictable than arc in box. For a 95% confidence level take a tolerance of approximately
+/-14% from the calculated arc current in box. For arc in open air on medium voltages,
the variation of +/-4% is small enough to be ignored.
There are two reasons for considering the deviation in arc current from the estimate:
a) The incident energy is calculated from the arc current.
99
b) The arc current affects the clearing time of a protective device. As described in
the previous chapter, the incident energy is proportional to the arc time.
Gap Between Electrodes
For low voltages, the gap between electrodes affects magnitude of the arc current. The
gap (G) is part of the IEEE 1584 equation (1, 30, 31) for estimating the arc current. Using
this equation, the graph shown in Figure A.2 was plotted for arcing current as a function
of electrode gaps for various voltage levels, bolted fault current and enclosure type. For
all conditions, the arcing current reduces with increased gap. For medium voltage
systems the gap distance is not an issue for typical switchgear and cable spacing.
The arc voltage is roughly proportional to the arc gap length29. Although the arc
resistance is expected to increase with gap length, it is not a linear relationship. The arc
current decreases with arc resistance. However, since resistance is highly non-linear, it is
preferable to use the empirically derived method of evaluating the effect of gap on arc
current.
0.2kV, 87kA, Box
50.0
45.0
40.0
35.0
30.0
25.0
20.0
15.0
10.0
5.0
0.0
0
10
20
30
40
Gap Betw een Electrodes (m m )
50
60
Figure A.2: Effect of gap between electrodes on calculated arc current for low
voltages (<1000V) using IEEE 1584 equation.
The phase conductor spacing required by NEMA for various voltages is presented in
Table A.4. Similarly, IEEE 1584 guide mentions typical gaps. When we assume a certain
arc gap distance for low voltages, either based on NEMA or some typical value, we must
be aware of the fact that any variation in the actual electrode gap could lead to variation
in actual arc current. Therefore, the incident energy to which a worker may be exposed
may vary. The gap between exposed conductors may vary from equipment to equipment.
Also, at the terminal of a given equipment, the gap between conductors may vary
depending upon the shape and layout. For conductors that are not exactly parallel
(spatially), the minimum and maximum gap may be of interest. Arcs travel away from
100
the power source30. Therefore, it may be sufficient to record the gap between the tip of
the exposed conductor away from the source. For these reasons, it may be very difficult
for the engineer to determine the exact electrode gap for every equipment, especially if
the equipment is live. Although it can be safely assumed that the NEMA gap will be
maintained for all manufactured equipment, the same cannot be guaranteed about the
field installations of conductors, tap-offs and terminations, or repair jobs. If the exact
gaps are known, it is always better to use the data. In case the gaps tend to vary with
equipment, and the engineer may have an idea of the approximate variation from typical
values, then analysis can be carried out for various scenarios such as typical, minimum,
and maximum gaps.
Table A.4: NEMA gap between conductors for various voltage levels
Equipment Type
NEMA Gap
between
conductors
0.1 - 1.0
Switchgear
MCC/Panel
Open Air
Conductor
Switchgear
MCC/Panel
Open Air
Conductor
Switchgear
MCC/Panel
Open Air
Conductor
32
25
32
13
102
102
102
13
153
153
153
13
1.0 - 5.0
5.0 - 15
Gap Unit
mm
mm
mm
mm
mm
mm
mm
mm
mm
mm
mm
mm
Considering multiple scenarios for electrode gaps may be somewhat time consuming,
although more reliable. A quick method of examining the effect of variation in electrode
gap is the sensitivity analysis. Figure A.2 shows the relation between electrode gap and
arc current to be almost linear. Table A.5 shows the sensitivity of arc current to electrode
gap for various voltages, bolted fault currents and enclosure type. The sensitivity is the
percent variation in arc current for 1 mm variation in electrode gap.
The sensitivity varies with the system voltage and the available fault current. However,
for practical purpose the average sensitivity of 1% can be used. The following example
illustrates how sensitivity analysis can be used to quickly determine the possible range of
arc current due to variation in arc gap.
101
Table A.5: Sensitivity of arc current to electrode gap for various conditions
Enclosure
Box
Box
Open
Box
0.208
0.400
0.610
0.485
36
53
88
103
Average
-1.0%
-1.3%
-0.7%
-1.0%
-1.0%
Example:
For a system with line-line voltage of 0.6 kV, assumed electrode gap of 32 mm, and
bolted fault current of 22.6 kA, the arc current in box was estimated to be 15.7 kA using
IEEE 1584 equation. It was also observed that the electrode gaps varied approximately
from 25mm to 40mm at different terminals and connections. What is the likely arcing
current?
Case 1: Gap, G = 25mm.
Variation in gap, g = 25 32 = - 7mm.
Assume sensitivity of -1%/mm.
Variation in arc current = - 7 * (-1%) = 7%
Maximum arc current = 15.7 * (1 + 0.07) = 16.8 kA.
Case 2: Gap, G = 40mm.
Variation in gap, g = 40 32 = 8mm.
Assume sensitivity of -1%/mm.
Variation in arc current = 8 * (-1%) = -8%
Minimum arc current = 15.7 * (1 - 0.08) = 14.5 kA.
The calculated arc current can have values anywhere between 16.8 kA and 14.5 kA.
Further variations can be considered due to randomness of arcs. For a confidence level of
95% the arc current can be within the limit of +/- 25% of the estimated value.
Lower limit of arc current = 14.5 * (1 0.25) = 10.8 kA.
Upper limit of arc current = 16.8 * (1 + 0.25) = 21.0 kA.
102
NOTE: It is not possible for the arc current to be higher than the bolted fault current.
This condition is not violated for an upper limit in arc current of 21 kA. In cases where
the estimated arc current exceeds the bolted fault current, take the arc current as equal to
the bolted fault current. For very low arcing currents, it is beneficial to check whether the
arc current is likely to be self-sustaining under the given system parameters. For 480 volt
systems, the industry accepted minimum level for a sustaining arcing fault current is 38%
of the available three phase fault current31.
The measured incident energy may deviate more widely than the arc current from their
respective calculated values. Figure A.3 shows the deviation of measured incident energy
about the calculated incident energy using the IEEE 1584 equations for low voltages.
First, the incident energy was calculated for various test conditions. Then it was
compared with the measured values. The deviation here is the difference in percent of
calculated values. For low voltage arcs in open air, the measured incident energies during
various tests were found to be lower than the calculated incident energy. The smallest
negative deviation is -25%. It can be concluded that highest possible incident energy is
about 75% of the calculated incident energy when using the IEEE 1584 equations for low
voltage arcs in open air. For LV arc in box, it can be seen from Figure A.3 that the
maximum deviation is very high. The maximum deviation was found to be about 62%.
Therefore, if the calculated incident energy for a case is 10 cal/cm2, there is a possibility
that the actual incident energy may be as high as 10*1.62 or 16.2 cal/cm2.
Although the nature of arcs is highly unpredictable, it is observed that arcs in open air, for
both low voltage and medium voltage, are more predictable than those in box. This can
be concluded from the shape of the histogram (or frequency distribution). The plots for
arcs in air have higher peaks and narrower bases (bottoms) than do the curves for arcs in
box.
The deviation of measured incident energy for medium voltages is shown in Figure A.4.
The maximum deviations are 46% for open air and 49% for box. Therefore if the
incident energy for a case of medium voltage in open air is calculated to be 10 cal/cm2
the actual energy may be as high as 10*1.46 or 14.6 cal/cm2.
103
60%
Low Voltage
LV Open Air
50%
Frequency (%)
LV Box
40%
30%
20%
10%
0%
-100%
-50%
0%
50%
Deviation from Estimated Incident Energy (%)
100%
Medium Voltage
MV Open Air
50%
Frequency (%)
MV Box
40%
30%
20%
10%
0%
-100%
-50%
0%
50%
100%
For the sake of simplicity, the discussions in this section have been based upon univariate analysis. In reality, the variations can be dependent upon multiple factors. To see
the effect of bolted fault current on the deviation let us look at Figure A.5. It can be
observed from this plot that the deviations tend to decrease with increasing bolted fault
current. In Figure A.5, trend lines are drawn for the maximum incident energies. For
higher bolted fault currents, the number of observations are small, and therefore it is not
possible to guarantee that the actual incident energy will not be higher than the predicted
value after accounting for deviations. It is learned at the time of writing of this article
that further tests on arc flash are being planned. It is hoped that more conclusive
104
information will be available in the future. In the mean time, with the available data, the
following procedures can be adopted as a rule of thumb.
The medium voltage trend line in Figure A.5 shows almost constant deviation with
respect to bolted fault current. Therefore, we can assume that the maximum possible
incident energy is 49% higher than the estimate provided by the IEEE 1584 equations.
For low voltage, the maximum possible deviations of incident energy from the calculated
values are about 60% at 10 kA IBF, 40% at 40 kA, and 25% at 60 kA, 10% at 80 kA and
0% at 100 kA. Further breakdown is possible with various enclosure types in air or in
box, but is not discussed in this text. Readers are encouraged to obtain the test data from
IEEE and compare the results. This chapter deals with applying the statistical adjustments
on calculated results without changing the calculation methods proposed in the standards.
Another approach with similar results is to vary the calculation factors (Cf) in the IEEE
1584 equations to obtain more reasonable results directly.
Deviation of Measured Incident Energy about Estimated
100%
LV Measured
80%
HV Measured
60%
Linear (Maximum
MV Mesaured)
Linear (Maximum
LV Measured)
Deviation IE (%)
40%
20%
0%
-20%
-40%
-60%
-80%
-100%
0
20
40
60
80
100
120
Figure A.5: Deviation of measured incident energy from calculated value for various
bolted fault currents using IEEE 1584 equations.
Arc Gap
The arc voltage varies with the arc gap length. A higher arc voltage means a higher arc
power for the same arc current. Therefore the incident energy can be expected to
increase with arc gap length, as long as the arc current is not sharply reduced.
105
In the IEEE 1584 equations, the incident energy is also a function of the gap between
electrodes. As mentioned in the previous section on arc current, the actual gap may vary
from the NEMA gaps or assumed gaps due to various reasons. Further adjustments can
be made to account for the variation in incident energy as a result of variation in gap
between electrodes. Figure A.6 shows the effect of arc gap on the IEEE estimate for
incident energy.
For low voltage, the sensitivity of calculated incident energy is about 0.3% per mm
deviation in arc gap. For a 10mm deviation in gap, the incident energy can be expected to
be 3% higher, which is a rather small increment and can be neglected. In most cases the
variation in gap is not likely to be substantial, and therefore the effect of variation in gap
on the incident energy can be ignored.
2.5
2
1.5
1
0.5
0
0
50
100
150
200
250
300
350
Gap (m m )
Figure A.6: Effect arc gap on incident energy using IEEE 1584 equations.
In obtaining the incident energy based on IEEE equations for the test data, the measured
values of the arc current were used instead of the calculated arc currents. This was done
because we will be using the various adjusted values of arc current rather than just the
calculated arc current. This will cover the entire possible range of arc currents and also
provide more conservative and reliable results in the evaluation of incident energy.
106
worst possible condition. Persons carrying out the arc flash hazard assessment should
read the protective device data documents carefully to ensure that the maximum trip time
has been taken for the calculations and not the average, median or minimum trip times.
Some manufacturers publish fuse TCC for just the minimum melting time of the fuse. In
such cases, it is necessary to contact the manufacturer and obtain the total clearing time
curve or apply a tolerance to estimate the total clearing time. The tolerance may be
provided in the fuse data document or it can be assumed to be similar to the tolerance of
other similar fuses.
.2 .3 .4.5.6 .8 1
700
500
400
300
200
100
70
50
40
30
BL-1
Cutler Hammer Series C
MDLB
Frame = 800(700-800AT)
Trip = 800
Inst = HIGH (6400A)
700
500
400
300
200
Estimated Arc
Current, 7kA
100
70
50
40
30
10
7
5
4
3
10
7
5
4
3
1
.7
.5
.4
.3
TIME IN SECONDS
20
TIME IN SECONDS
20
1
.7
.5
.4
.3
.2
.2
.1
.07
.05
.04
.03
.1
.07
.05
.04
.03
.02
.02
.2 .3 .4.5.6 .8 1
Figure A.7: Trip time at instantaneous pickup current for thermo magnetic
breaker.
Figure A.7 shows the TCC of a thermal magnetic molded case circuit breaker (MCCB).
The estimated arc current of 7 kA shown here as an example, is equal to the pickup of the
magnetic trip. It is likely that the magnetic trip will activate and stop the arcing current at
107
or below 0.04 seconds. However, there is also a possibility that the MCCB may not trip at
that value pickup value. In such a case the thermal unit will trip the breaker at a much
higher time, at 6 seconds. This is a great difference in trip time and hence will result in
great difference in estimated incident energy. Close attention is required in obtaining the
trip times at arc current values close to pickup values.
Figure A.8 shows the incident energies for various values of arc current for a typical solid
state trip device using the IEEE equations. At the IEEE estimate of arc current, the
calculated incident energy is 2.5 cal/cm2. It was mentioned in the previous section that
the actual arc current could be as low as 75% of the IEEE estimate. At 75% of the IEEE
estimate, the trip is much higher and therefore the high incident energy of 9 cal/cm2. The
sharp change in the incident energy at about 82% of the IEEE estimate of arc current is
due to the pickup of short time delay unit of the device, similar to the observation made
for the MCCB in Figure A.7.
12
10
9 cal/cm2
8
6
4
2.5 cal/cm2
75% of
IEEE
Estimate
IEEE
Estimate Iarc
0
50
60
70
80
90
100
Figure A.8: Incident energies for various arc current values for typical solid state
trip device.
108
Theoretical Formulas
Ralph Lee32 proposed the arc flash boundary in terms of the bolted fault MVA. This
assumes the maximum possible arc power, which is half the total available fault MVA or
bolted fault MVA. Calculation of the incident energy is based upon this formula.
DB =
where
DB =
V =
Ibf =
t =
(A.1)
This formula is applicable when definite time trip function is used to interrupt the fault.
A definite time trip function is a fixed time delay, and is independent of the fault current
passing through the protective device. Instantaneous trips are also approximately fixed
time in most devices. If the trip time is independent of the fault current, then making the
assumption that the arc current may have a value that will yield the maximum arc power
is justified. However, this formula needs to be modified if the trip time is a function of
the fault current. Inverse type relays, fuses, thermal trip units and solid state trip units
with I2T time delays have current dependent trip time. Assessment for inverse time
functions can be approached using the same circuit assumptions with which the above
equation (A.1) was derived.
Equivalent Circuit Model
Iarc
Xs
Rs
Vs
Ra
Figure A.9: Equivalent circuit diagram with Thevenin source and impedance, and
arc resistance.
The Thevenin equivalent circuit for the arc fault is shown in Figure A.9. Here, V is the
system voltage at the point of fault, Ra is the equivalent arc resistance, Xs and Rs are
components of the Thevenin impedance Zs, and Iarc is the arc current. When the arc
resistance is zero (a hypothetical case), the arc current is equal to the bolted fault current.
No power is dissipated through the arc. As the arc resistance increases, the arc current
decreases. The arc power reaches a maximum when the arc current is approximately 0.7
109
per unit of the bolted fault current. This holds true only if the X/R ratio of the system is
very high (Rs is negligible). If the X/R ratio is low, then the maximum power transfer
occurs when arc current ratio (Iarc/IBF) is less than 0.7, and the maximum arc power is less
than 0.5 times the bolted fault MVA. A plot of arc power as a function of arc current is
shown in Figure A.10. The arc power and arc current have been normalized in this plot.
1
Parc / Parc[max]
0.8
0.6
0.4
0.2
0
0.30
0.40
0.50
0.60
0.70
0.80
0.90
1.00
Ia / IBF (pu)
Figure A.10: Plot of arc power as function of arc current. (Arc current is expressed
in per unit of the bolted fault current and the arc power is expressed in per unit of
the maximum arc power.)
We can substitute the bolted fault MVA term (1.732 * V * IBF) in equation (A.1) by
2*Parc, since the maximum arc power is released when Parc = 0.5 * MVABF. The
equation now becomes:
DB =
(A.2)
2.65 * 2 * Parc * t
For inverse type trip functions, both the arc power and the arcing time are dependent on
the arc current. Since the total energy released by the arc, Earc, is equal to Parc* t, this too
is a function of the arc current. We are now interested in finding the maximum arc
energy released as allowed by the protective device.
For an inverse square time-current function, the trip curve may be expressed as:
t= K /I
(A.3)
110
Xs2
Rs
(A.4)
Assuming the current seen by the trip device is equal to the arc current, we can express
the total arc energy as allowed to be released by the inverse-square time-current trip
device:
E arc = Parc
* t = K*
I
arc
Xs2
Rs
(A.5)
0.8
0.6
0.4
0.2
0
0.30
0.50
0.60
0.70
0.80
0.90
1.00
Figure A.11: Plot of arc power, trip time and arc energy released as function of arc
current. All values in the plot are normalized: Parc is in per unit of maximum arc energy
(0.5*MVABF); trip time for inverse square curve (I^2T) is 0.1 seconds for Iarc = IBF; Earc
is normalized Parc times arcing time.
The plot of trip time and arc energy released as a function of arc current is shown in
Figure A.11. This shows that the arc current at which the arc power is maximum (70% of
bolted fault current in this case) does not yield the maximum arc energy. Because of the
inverse-square time-current trip device, the highest energy released is for the lowest arc
current. For the example shown in Figure A.11, the arc energy released when the arc
current is 30% of bolted fault current is about five times greater than that when the arc
current is 70% of bolted fault current.
From (A.2) and (A.5), we get the arc flash boundary as a function of the arc current when
inverse-square trip function is used.
DB =
2.65 * 2 * K *
I
arc
Xs2
Rs
(A.6)
111
Summary
Engineers carrying out arc flash hazard assessment need to be aware of the various
uncertainties in the nature of arcs and other factors affecting the evaluation process.
The test result upon which IEEE equations are based is available through IEEE.
Statistical analysis based on this data can provide insight into the deviation of
possible outcomes from the estimated values.
Probability based deviations can provide likely ranges of arc current or incident
energy for a given confidence level. Using these deviations, adjustment can be made
for the IEEE estimates to obtain more reasonable estimates.
Uncertainties in arc gaps and trip time can cause the actual outcome to differ from the
estimate. The highest calculated cal/cm should be considered to provide the worker
maximum safety.
28
IEEE Standards 1584-2002, IEEE Guide for Performing Arc-Flash Hazard Calculations, IEEE
Industry Applications Society, September 23, 2002. (SH95023)
29
A. P. Strom, Long 60-Cycle Arcs in Air, AIEE Transactions, Vol. 65, pages 113-117, March
1946.
30
NFPA 70E May 2003 ROP, "Standard for Electrical Safety Requirements for Employee
Workplaces", 2003 Edition, page 57.
32
Ralph Lee, The other Electrical Hazard: Electric Arc Blast Burns, IEEE Transactions on
Industry Applications, Vol. IA-18, No. 3, May/June 1982.
112
113
equipment types in detail, so there are no reasons for minimized models or built in safety
factors. Secondly, conductor impedances and X/R ratios should be modeled for all
equipment in order to obtain realistic short circuit values. When assessing arc flash
hazards, higher short circuit currents may actually be non-conservative as far as PPE
level is concerned due to fast clearing times. Higher PPE can result at lower fault levels
because of the inverse characteristics of many protective devices. Depending on the
maximum fault level to provide the maximum PPE may result in decreased worker
safety.
It should be noted that the study results will only be as good as the system model. Every
effort should be made to model the actual equipment as found in the field.
Figure B.1: Example showing bus excluded from arc flash assessment
114
Emergency operating conditions. This may be with only small backup generators.
Maintenance conditions where short circuit currents are low and trip time high.
UTIL-2
TX-1
30 / 44.8 MVA
115 - 13.8 kV
9%
11
5
11
5
SWITCHGEAR-A
13
.8
kV
kV
TX-2
30 / 44.8 MVA
115 - 13.8 kV
9%
SWITCHGEAR-B
13
.8
UTIL-1
kV
kV
What is important is that each one of these conditions may change the level of short
circuit current, which in turn changes the clearing time of the protective devices. These
changes can have a significant impact on the arc flash hazard and the PPE requirements
for each piece of equipment.
LOAD-B
12 MW
7 MVAR
Figure B.2: Closed tie breaker increases available short circuit current.
In Figure B.2, an example system with maximum available short circuit current is shown.
Both utility sources are online and the switchgear tiebreaker is closed.
In Figure B.3, an example system with minimum short circuit current is shown. Both
utility sources are online and the switchgear tiebreaker is open, reducing the available
short circuit current on each bus. The examples above consider a double-ended utility tie
system, but the application applies to either low voltage systems with tie breakers or to
emergency generation providing stand alone power or working in parallel with the
normal system.
115
11
UTIL-2
11
UTIL-1
kV
kV
TX-1
30 / 44.8 MVA
115 - 13.8 kV
9%
kV
.8
13
SWITCHGEAR-B
OPEN
13
SWITCHGEAR-A
.8
kV
TX-2
30 / 44.8 MVA
115 - 13.8 kV
9%
LOAD-A
LOAD-B
10 MW
8 MVAR
12 MW
7 MVAR
5 6 7 8 9 10
5 6 7 8 9 100
5 6 7 8 9 1000
40
40
30
30
20
20
10
9
8
7
6
5
10
9
8
7
6
5
1
.9
.8
.7
.6
.5
1
.9
.8
.7
.6
.5
4000A
.4
TIME IN SECONDS
100
90
80
70
60
50
TIME IN SECONDS
100
90
80
70
60
50
.4
.3
.3
.2
.2
14000A
.1
.09
.1
.09
3
5 6 7 8 9 10
5 6 7 8 9 100
5 6 7 8 9 1000
116
The time current curve (TCC) of Figure B.4 shows an extremely inverse relay
characteristic with the trip time increasing as the current decreases. Decreased short
circuit current (opening a tie breaker, removing generation, etc.) can cause longer trip
times and may increase incident energies and the resulting arc flash hazard.
In summary, arc flash assessment should include each operating mode for the power
system to insure correct incident energies are calculated for all system conditions.
The arc flash incident energy and associated protection requirements are based on
potential burns to the persons chest or face, not the hands or arms. The degree of injury
depends on the percentage of the persons skin that is burned and the critical nature of the
burn. Obviously, the head and chest areas are more critical to survival than fingers or
arms.
Appropriate working distances for most operations can be estimated by placing your
elbow at your side and extending your hand to the equipment. A typical average for this
distance is 18 inches. By extending the arm to the full out position, this can be increased
to 24-28 inches for most people, but out-stretched arms are not a typical working
distance. See Figure B.5.
EasyPower provides up to five (5) commonly used working distances for each voltage
level. This allows the user to develop a safety program where distances can be modified
for a specific operation or maintenance function, allowing easy standardization of
117
clothing levels and safety benefits. For certain types of work practices like hot stick
operation, or when the energized bus is set back from the worker, greater distances may
be used to correctly model the reduction in incident energy potential.
Unit of Measure
Working distances and arc flash boundaries are calculated and displayed in various units
of measure including; inches, feet, mm, or meters. Select the appropriate unit that will be
easily recognized and adhered to by workers. Critical safety programs such as arc flash
hazards should not confuse workers with units of measure. Example: For US markets
most workers are more familiar with inches and feet than mm and meters. The opposite
would be true for facilities in Europe.
The arc flash boundary is defined as the distance from the arc source where the onset of
second degree burns can occur. This is typically defined by medical researchers as 1.2
cal/cm2 or 5.0 Joules/cm2. Some research indicates that up to 1.5 cal/cm2 can be used for
exposure less than 6 cycles (0.1 seconds).
EasyPower provides the user with options based on clearing times less than 0.1 seconds
and clearing times greater than 0.1 seconds. EasyPower automatically determines the
operating time from the system protection characteristics, or from user defined times.
118
The arc-flash boundary incident energy must be set at the minimum energy
level in which a second-degree burn could occur. Do not increase the level from
those shown in the dialog box. Reduced values may be used based on your
safety or insurance requirements.
Calculation Standard
119
TX-2
BUS-4
BL-2
67.1" AFB
10.4 cal / cm @ 18"
#3 @ 18"
BL-1
99.6" AFB
18.7 cal / cm @ 18"
#3 @ 18"
BL-3
31.5" AFB
2.7 cal / cm @ 18"
#1 @ 18"
BL-4
67.1" AFB
10.4 cal / cm @ 18
#3 @ 18"
M-1
Figure B.7: Red breakers indicating incident energy exceeds user specified
threshold value.
Table B.1: Proposed NFPA-70E 2004 PPE Requirements
Risk
Category
PPE Requirements
Class #0
0-2 cal/cm2
Untreated cotton
Class #1
2-4 cal/cm2
Class #2
4-8 cal/cm2
Class #3
8-25 cal/cm2
Class #4
25-40 cal/cm2
c) Apply arcing currents and breaker/relay trip times to each device to determine arc
hazard incident energies, arc flash boundaries, working distances, and PPE
requirements.
The steps shown are required for performing the calculations with power analysis
software as well as by hand. Depending on the system size (number of buses) performing
this procedure can be extremely time consuming or nearly impossible without software
tools. Only software based tools that provide true, seamless integration of short circuit,
protective device coordination and arc flash hazard analysis can provide accurate
information for better worker protection and reduced productivity losses due to over
specification of gear. EasyPowers inherent one-line/analysis integration eliminates the
separate steps required by other programs and integrates the short circuit, protective
device, and arc hazard functions, thereby greatly reducing the time and effort to perform
the analysis.
Protective Device Coordination Using EasyPower
Using EasyPower, the process will be broken down into two steps for clarification
purposes.
a) System wide protective device coordination.
b) Arc flash calculations.
While this guide does not provide the details for performing a protective device
coordination study, it should be stressed that this study is the cornerstone to providing
accurate arc flash calculations. Accurate protective device clearing times are essential for
providing correct incident energy calculations and the resulting AF boundaries.
Accurate protective device clearing times are essential for providing correct incident
energy calculations and the resulting AF boundaries.
In Figure B.8, the substation secondary main breaker provides selective coordination
using either setting. However, the arc flash incident energy is increased from 11 cal/cm to
29 cal/cm for the higher short time delay setting. This increases the PPE requirement
from 3 to 4 significantly increasing costs and the probability workers may try to bypass
the higher PPE clothing requirements. This scenario is common to plants where an
accurate protective device coordination study has never been performed, or when workers
121
5 6 7 8 9 10
900
800
700
600
500
5 6 7 8 9 100
5 6 7 8 9 1000
5 6 7 89
900
800
700
600
500
0.161
400
300
400
300
TX-2
1000 / 1288 kVA
6%
200
20.009
100
90
80
70
60
50
TX-2
1 / 1.288 MVA
13.8 - 0.48 kV
6%
40
BL-1
30
BUS-4
20
100
90
80
70
60
50
40
26
.0
24
M-1
250HP
IND
Full Voltage
200
TIME IN SECONDS
TX-2
FLA
30
20
1.614
BL-3
10
9
8
7
6
5
10
9
8
7
6
5
M-1
4
3
2
1
.9
.8
.7
.6
.5
.3
.2
1
.9
.8
.7
.6
.5
Improper Setting
.4
~BL-5
20000A
.3
Proper Setting
BL-3
GE MVT-9
Sensor = 800
Plug = 800
Cur Set = 0.5 (400A)
LT Band = 1
Inst = 4 (3200A)
.1
.09
.08
.07
.06
.05
3
2
BL-1
GE MVT-Plus
Sensor = 1600
Plug = 1600
Cur Set = 0.8 (1280A)
LT Band = 1
STPU = 4 (5120A)
ST Delay = Min
.4
250 HP
IND
~BL-5
GE MVT-9
Sensor = 1600
Plug = 1600
Cur Set = 0.8 (1280A)
LT Band = 1
STPU = 4 (5120A)
ST Delay = Max
TIME IN SECONDS
.6
TX-2
1000 / 1288 kVA
INRUSH
.2
.1
.09
.08
.07
.06
.05
BL-1
20009A
.04
.04
.03
.03
.02
.02
BL-3
31391A
.6
.8
5 6 7 8 9 10
5 6 7 8 9 100
5 6 7 8 9 1000
5 6 7 89
In this next example, Figure B.9, the secondary main breaker is properly set except the I2t
function is left in. This raises the arc flash incident energy from 11.0 cal/cm2 to 16
cal/cm2. If increased arcing impedance is modeled, reducing the arcing current to 80%,
the incident energy is raised to over 20 cal/cm2. This increase in energy can result in an
122
increased cost of personal protective equipment and ongoing worker productivity losses
associated with the increased PPE requirements.
CURRENT IN AMPERES X 100 AT 480 VOLTS
.8
5 6 7 8 9 10
5 6 7 8 9 100
5 6 7 8 9 1000
TX-2
FLA
0.136
1000
900
800
700
600
500
.5 .6
400
5 6 7 8 9 10000
1000
900
800
700
600
500
400
300
300
200
200
TX-2
750 / 966 kVA
8%
11.264
M-1
250HP
IND
Full Voltage
40
100
90
80
70
60
50
40
BL-1
30
BUS-4
17
.2
79
100
90
80
70
60
50
TX-2
0.75 / 0.966 MVA
13.8 - 0.48 kV
8%
30
20
20
10
9
8
7
6
5
1.614
10
9
8
7
6
5
M-1
250 HP
IND
BL-1
GE MVT-Plus
Sensor = 1600
Plug = 1600
Cur Set = 0.8 (1280A)
LT Band = 1
STPU = 4 (5120A)
ST Delay = Min
1
.9
.8
.7
.6
.5
.4
.3
1
.9
.8
.7
.6
.5
.4
.3
BL-3
GE MVT-9
Sensor = 800
Plug = 800
Cur Set = 0.5 (400A)
LT Band = 1
Inst = 4 (3200A)
.2
.1
.09
.08
.07
.06
.05
TIME IN SECONDS
TIME IN SECONDS
BL-3
.2
TX-2
750 / 966 kVA
INRUSH
BL-1
11264A
.1
.09
.08
.07
.06
.05
.04
.04
.03
.03
BL-3
20026A
.02
.02
.01
.5 .6
.8
5 6 7 8 9 10
5 6 7 8 9 100
5 6 7 8 9 1000
.01
5 6 7 8 9 10000
Figure B.9: Typical example of short time I2T trip curve used in place of definite
time delay (flat)
In medium and high voltage systems, it is quite common to find relay settings that are set
far above proper protective boundaries. This is especially true when new systems have
been added to older systems, or when system studies have not been updated on a regular
basis.
123
5 6 7 8 9 100
TX-3
FLA
5 6 7 8 9 1000
5 6 7 8 9 10000
OPTION
0.600
900
800
700
600
500
400
5 6 7 8 9 100000
900
800
700
600
500
400
300
300
TX-3
5 MVA
13.8 - 2.4 kV
6%
M-3
1000HP
IND
Full Voltage
10
9
8
7
6
5
4
3
40
REFINER
30
20
400/5
R-8
GE IAC 53
51/50
Very Inverse
CT Ratio = 1200/5
Tap = 6 (1440A)
Time Dial = 3.1
MF
10
9
8
7
6
5
M-3
1000 HP
IND
TX-3
5000 kVA
6%
4
3
1
.9
.8
.7
.6
.5
R-13
Multilin 469
51/50 SC
Standard Overload
CT Ratio = 400/5
Overload = 1.01 (253A)
Time Dial = 2
SC Pickup = 7.4 (2960A)
TIME IN SECONDS
TIME IN SECONDS
20
51
1200/5
OPTION
Multilin SR745
51/50
Extremely Inverse
CT Ratio = 400/5
Tap = 6 (2400A)
Time Dial = 0.7
Instantaneous = 20 (8000A)
Delay = 0.113
75
3
30
18
.
40
200
100
90
80
70
60
50
14.590
100
90
80
70
60
50
1.302
200
1
.9
.8
.7
.6
.5
.4
.4
.3
.2
.3
.2
TX-3
5000 kVA
INRUSH
.1
.09
.08
.07
.06
.05
.1
.09
.08
.07
.06
.05
R-8
14669A
.04
.04
.03
.03
R-13
26417A
.02
6 7 8 9 10
5 6 7 8 9 100
5 6 7 8 9 1000
5 6 7 8 9 10000
.02
5 6 7 8 9 100000
Figure B.10: Relay trip time compared with solid state device.
Figure B.10 shows an older style induction disk relay providing protection to a 2400 volt
MCC line-up. This unit must be set above the motor protective relays for selective
coordination but low enough to provide proper protection. A standard instantaneous unit
cannot be used without tripping the entire lineup for a motor fault. The tap and time dial
setting shown is a good compromise and typical of many systems. The unit will clear a
bus fault in approximately 0.5 seconds (30 cycles). The arc flash incident energy is over
30 cal/cm2 and requires a PPE of 4. Using a new solid-state relay with delayed
instantaneous setting for selective coordination, the incident energy is lowered to 10
cal/cm2, greatly enhancing worker safety.
124
As can be seen, proper protective device settings can greatly enhance worker safety and
system reliability. Performing an arc flash assessment without first providing proper
protection settings can significantly impact the assessment.
Arc Flash Calculations Using EasyPower
In the previous sections, we have provided the basis for setting up the system model for
proper arc flash calculations. In this section we will provide the details for actually
performing the arc flash study and understanding the results, as well as, some tricks of
the trade.
Arc flash calculations are performed in EasyPowers ShortCircuit focus.
EasyPowers SmartClick interface allows the users to simply double click any bus for
instantaneous results, to fault selective buses, or to Fault All buses.
on the EasyPower
In the example below (Figure B.11) select the ArcFlash button
toolbar. Double click on Bus-4. The results appear on the one-line.
TX-2
BUS-4
BL-2
67.1" AFB
10.4 cal / cm @ 18"
#3 @ 18"
BL-1
99.6" AFB
18.7 cal / cm @ 18"
#3 @ 18"
BL-3
31.5" AFB
2.7 cal / cm @ 18"
#1 @ 18"
BL-4
67.1" AFB
10.4 cal / cm @ 18
#3 @ 18"
M-1
Each protective device displays the required arc flash boundary, let through energy in
cal/cm2, and PPE requirement at a user specified working distance.
The values displayed on the one-line are based on the let through energy of the
protective device, i.e. the energy on the load side of the device, not the line side.
Note: The values displayed are based on the let through energy of the protective device,
i.e. the energy on the load side of the device, not the line side. This important safety
aspect must be understood when applying arc flash results. When working on the line
125
side of a protective device, i.e. the incoming terminals, breaker stabs, or incoming bus
work the incident energy on the line side must be found from the let through energy of
the upstream device, not the device you are working on. For example, when working on
the primary stabs of breaker BL-2, the incident energy available to the worker is found
from the first upstream device protecting BL-2. This is the let through energy of the
secondary main device BL-1, which is 18.7 cal/cm2. If the worker is working on the
load side stabs of BL-2, the let through energy is controlled by BL-2. In EasyPower
these results are associated with that breaker, in this case, 10.4 cal/cm2.
Figure B.12 below shows the same system, but with a primary fuse protecting the
buswork from the TX-2 secondary terminal through the primary or line side bus stabs of
breaker BL-1. Work in this area will require a PPE level 4 requirement and be subject to
a let through energy of 30.8 cal/cm2.
139.9" AFB
30.8 cal / cm @ 18"
#4 @ 18"
TX-2
BUS-4
BL-2
67.1" AFB
10.4 cal / cm @ 18"
#3 @ 18"
BL-1
99.6" AFB
18.7 cal / cm @ 18"
#3 @ 18"
BL-3
31.5" AFB
2.7 cal / cm @ 18"
#1 @ 18"
BL-4
67.1" AFB
10.4 cal / cm @ 18"
#3 @ 18"
M-1
Figure B.12: Primary side fuse protects line side of breaker BL-1 on BUS-4 but has
higher incident energy.
When laying out your safety plan, keep in mind that you will always be working on
either the line side (upstream) or load side (downstream) of a protective device.
When displaying the results graphically, EasyPower provides the user with a clear
picture of line side and load side let through energies, as well as, a visual indication of
problem areas and correct clothing compliance. This information can be posted in the
electrical room providing workers with a clear picture of the system and the hazards that
may not be as easily apparent with just stick on labels. With the click of a mouse you can
change system parameters and compare different operating scenarios. This provides
valuable training information that helps engineers and electricians understand how
system changes impact arc flash hazard ratings.
126
For most large studies, however, it is typically more efficient to display results in
spreadsheet form and print the Arc flash hazard warning labels for each device. To
perform this operation, simply go to Tools ShortCircuit Options ArcFlash Tab,
and check 9 ArcFlash Spreadsheet in the Create Report section of the tab in the
EasyPower menu. See Figure B.6.
Now instead of double clicking on the bus to initiate the fault, select Fault All from the
toolbar, and then from the menu choose Window ArcFlash Hazard Report. A
spreadsheet similar to Figure B.13 below will tile in the foreground of the window.
The EasyPower ArcFlash spreadsheet provides all the data used in the calculations to
determine AF Boundary, Incident energy, and PPE requirements for each protective
device in the system. This data can be applied directly to comply with NEC 2002 and
NFPA-70E by simply clicking on File Print Labels in the menu.
Before you print labels it is recommended that you refer to STEP-2, and review your
modes of operation. It is highly recommended that you save your different operating
modes in EasyPowers Scenario Manager. This will allow you to refer to each case
without affecting the base case system as you make changes and fine-tune your arc flash
assessment.
Summary
1) Run base case arc flash calculations.
2) Switch to different operating modes as defined in Scenario Manager.
3) Run arc flash calculations for each operating mode to determine highest arc
hazard.
4) Compare the highest incident energies from the base case and scenarios. Take the
case with the highest values (there may be multiple cases for different parts of the
system) and modify the arcing current to reflect a high impedance arcing current.
This will lower the arcing current, which may cause longer trip times and result in
higher incident energies. See STEP-2, and Figure B.6. Note: A good starting
place is 80% of the calculated arcing current. Going much lower than this may
result in current values that cannot be realistically maintained.
5) Compare the incident energies of the case selected in task 4 above with the high
impedance values of the same case. Print labels.
127
The following labels can be printed on plastic stock through most laser printers, or via
commercially available label printers. EasyPower provides direct output to selected
label printers so you will avoid hours of data conversion routines.
Figure B.14: Warning label that can be printed directly from EasyPower ArcFlash
program.
Economic Benefits
The economic benefits of performing arc flash assessments using dedicated power system
analysis software becomes readily apparent when the alternative is to use a spreadsheet
calculator like those provided in IEEE-1584. Arc hazard assessments using a spreadsheet
calculator requires the following tasks:
1) Transfer data from the short circuit program to the spreadsheet calculator. This
includes short circuit calculations, bus names, and bus voltages.
2) Determine the arc gap for each calculation or equipment in the spreadsheet.
3) Determine the trip time for each device or bus in the spreadsheet. There are
usually multiple trip times required for each bus.
4) Run the calculation.
5) Apply NFPA-70 PPE requirements to each calculation.
6) Spreadsheet calculations DO NOT provide for a device-by-device analysis, unless
the users accounts for each device in the system.
128
7) Perform the calculation for a change in tie breaker status or generation (mode of
operation).
8) Take the highest results (case) and re-run using a higher impedance arcing fault to
insure accurate results.
As can be seen, the man hours required to perform an arc flash assessment can be cost
prohibitive using a spreadsheet calculator. When applied to large systems, such as those
in the petrochemical or the pulp and paper industries, it becomes almost impossible.
Another consideration is the potential for errors when applying the hand calculations, trip
time look-ups, and spreadsheet work.
EasyPowers complete integration of short circuit, protective device coordination, and
arc flash can be exponential as compared to the use of an IEEE-1584 spreadsheet
calculator. EasyPower simplifies the process, reduces human error and provides a
basis from which system changes and modifications can be modeled and the study results
updated immediately without the extensive work and risk of error associated with a
spreadsheet. EasyPower also helps with safety program requirements for accurate
documentation as it provides reports that become a key part of a corporate arc flash
hazard safety program. The EasyPower ArcFlash program will also be kept up-to-date
with the latest industry standards, helping to ensure the most accurate results.
Conclusion
This guide presents the basic steps for performing an arc flash hazard assessment using
power analysis software. Users performing arc flash assessments should be aware that
reduced short circuit currents could increase arc incident energies for some cases. They
should also fully understand the arc let through energies as applied to protective devices,
before assigning arc flash boundaries and incident energy ratings to equipment.
Power analysis software that provides complete one-line/analysis integration eliminates
the separate steps required by other programs and integrates the short circuit, protective
device, and arc hazard functions. This greatly reduces the time and effort to perform the
analysis.
129
Index
Index
qualified persons, 32
3-phase faults, 43
all possible connections, 42
American Society for Testing and
Materials (ASTM), 79
arc blast pressure, 37
arc current, 29, 45, 97
arc flash analysis, 25
arc flash boundary, 118
Arc flash hazard, 1
arc flash hazard assessment, 39
arc flash hazard program, 15
arc flash hazard report, 127
arc gap, 128
arc resistance, 50
Arc Thermal Performance Value (APTV),
80
arc-flash hazard program, 12
arcing fault incident energies, 120
arcing faults
hazards of, 3
arcing time, 30, 50, 106
reducing, 69
assessing existing safety program, 17
available fault current, 8
available short circuit current, 115
average melting time, 51
awareness, 87
base case, 127
blast shrapnel, 3
bolted fault current, 29, 44, 120
boundaries, 7
box configuration, 29
branch current contribution, 49
breaker response time, 69
burn injury costs, 4
bus differential protection, 70
cable, 31
calculated arc current, 47
calculation factor, 30
calculation standard, 117, 119
care of FR clothing, 81
circuit breaker clearing time, 51
clearing time, 115, 118
clothing compliance, 126
comfort, 82
, 50
131
132
Index
tie breakers, 44, 115
time current curve (TCC), 117
time current curves (TCCs), 120
time-current characteristics, 50
tolerance, 51
total arc rating, 83
total clearing time, 51
training, 13, 87
transformers, 45, 113
type of enclosure, 39
Typical PPE, 8
133