PROTECTIVE CLOTHING GUIDELINES FOR ELECTRIC ARC EXPOSURE

Copyright Material IEEE

Paper No. PCIC-96-34

Dr. Thomas E. Neal Men H. Bhgham Richard L. Doughty

Non-Member Non-Member IEEE Fellow
E. I. duPont deNemours & Co. Bhgham Consultants E. I. duPont deNemours & Co.
Wilmington, DE 19880 Stockbridge, GA 30281 Wilmington, DE 19880

-
Abstract Since 1982 when Ralph Lee wrote about “The Regulations in OSHA 1910 Subpart R require that employees
Other Electrical Hazard: Electric Arc Blast Burns” [l], wear appropriate clothing when they may be exposed to an
industrial and utility engineers have been attempting to electric arc hazard. To meet these regulations, employees
better quantify the degree of risk to personnel associated must not wear clothing that could increase the extent of injury
with electric arc exposure and the appropriate protactive that would be sustained by the employee in the event of
clothing to wear to mitigate serious injury. Recent activity electric arc exposure. Clothing made from acetate, nylon,
by OSHA, NFPA 70E and ASTM F-18 has further polyester or rayon, either alone or in blends, is prohibited by
emphasized the need to protect workers from arc flash burn OSHA 1910.269 unless the employer can demonstrate that
injuries. This paper discusses the results of recent multi- the fabric has been treated to withstand the conditions that
phase arc testing which has enhanced knowledge about: 1) may be encountered, or that the clothing is worn in a manner
potential arc energy as a function of prospective fault that eliminates the hazard. OSHA has issued supplementary
current, and 2) arc protective clothing designs that are guidance for untreated cotton and wool, c l m n g the need
suitable for different levels of incident arc energy. Incident for employers to determine that these materials would not
energy levels are correlated with second degree burn ignite in potential arc exposure situations.
criteria for unprotected human skin. Protective clothing
systems consisting of flame resistant outer layers in single- The most recent edition of NFPA 70E, “Standard for
layer and multiple-layer construction, as well as Electrical Safety Requirements for Employee Workplaces,”
combination systems of jlame resistant outer layers with [5] was completed at the request of OSHA and was issued by
natural and man-made fiber inner layers are discussed NFPA on February 7, 1995. Part 11, Chapter 3 covers
Protection recommendations for the face, head, hands and personal protective equipment and requires for body
feet are also included

INTRODUCTION

In recent years awareness has been increasing that a

significant percentage of electrical injuries, as much as 80%
documented in some cases [2], are burns that result froth
exposure to the radiant and convective energy of electric arcs.
This awareness is reflected in the increasing number of
emerging regulations and standards dealing with prevention
of injuries from electric arc exposure. The medical
community, in addition, is focusing research on
understanding the complex mechanism of injury due to both
electric arc exposure and contact with energized conductors
[3]. The seriousness of the burn hazard is reflected in burn
survival data from a 1991-1993 study of the American Burn
Association [4] shown in Figure 1. Figure 1 indicates that for
a person in the age category 30-39 years, there is only a 50
percent chance of survival if 75 % of the body surface area is
burned. This is a distinct possibility if there is ignition of
untreated clothing as a result of arc exposure. Electric arc
exposure is typically more severe than flame exposure as
shown in Table 1. Figure 1 - Chance of Survival from Burn Injury

ISBN: 0-7803-3587-2 96-CH35988-6/96/0000-0281 $05.00 @ 1996 IEEE

- 281 -
TABLE 1 - COMPARISON OF TYPICAL ELECTRIC ARC AND FLAME EXPOSURES FOR PERSONNEL

protection that “employees shall wear clothing resistant to explosive expansion of surrounding air due to rapid heating
flash flame wherever there is possible exposure to an electric by the high temperature arc, and meltinglvaporization of arc
arc flash.” NFPA 70E is helpful in that it contains flash electrodes and metal components in the vicinity of the arc.
protection boundaries in Part 11, Appendix B based upon Radianffconvective energy from electric arcs can ignite and/or
Ralph Lee’s work [l] and includes sample calculations. Part melt everyday clothing, causing severe injury. Injuries
11, Appendix D, also contains a sample of a typical typically are second and third-degree burns. There is potential
hazardrisk evaluation procedure flow chart. A more detailed for fatal burns because ignited fabrics can cause burn injuries
discussion of OSHA and NFPA 70E requirements, and tables over a high percentage of the body area. Energy from electric
showing a tabulation of curable burn distances for electrical arcs generally produces significant electrical equipment
equipment at commercial and industrial installations, are
included in Jamil’s, Jone’s, and McClung’s work [6].
ASTM F1506-94 Flame Requirement
The American Society for Testing of Materials (ASTM) F- 18 Char Length Limit - 6 Inches
Committee has issued a Standard Protective Clothing Initial Specimen Length - 12 Inches
Specification, ASTM F 1506-94, “Standard Performance
Specification for Textile Materials for Wearing Apparel for
14
Use by Electrical Workers Exposed to Momentary Electric
Arc and Related Thermal Hazards” [7]. The specification
12
uses a vertical flame test for flammability evaluation of
v)
fabrics and requires that a fabric does not continue to burn
after exposure to an ignition source. Only fabrics with a char
$ 10
length less than or equal to 6 inches will pass the test. -
C

Typically, flame-resistant fabrics will meet the requirements

5” 8
ZI)
E
of the specification. However, all weights of everyday cotton
and polyestedcotton fabrics commonly used for work clothing 36
(includmg weights heavier than 11 oz/yd) will not meet this ii
specification. Figure 2 shows char length results from one 6 4
series of vertical flame tests per ASTM F1506 for several
2
fabric types which were laundered once prior to the test.
n
ENERGY PRODUCED BY ELECTlUC ARCS

An electric arc is the passage of electric current between two

electrodes through ionized gases and vapors. Electric arcs in
industrial power systems are normally short in duration,
typically 1 second or less, but produce very high energy
levels. The temperature of the arc plazma is reported as
being as high as 13,000 degrees K, 2.5 times hotter than the Fabric Types
surface temperature of the sun [l]. The electrical energy
supplied to the arc is converted into other forms of energy,
resulting in intense thermal radiation, damaging noise levels, Figure 2 - Fabric Vertical Flame Test Results

- 282 -
damage. and then the energy component of the fundamental and each
harmonic are added to obtain the total energy of the arc.
Theoretically, the energy produced by an electric arc, (Earc),
is equal to the integral of the product of the instantaneous arc If Equation (3) is applied directly using the arc rms voltage
voltage, (Varc), and the instantaneous arc current, (Iarc), and current, the calculated arc energy is higher than the
with respect to time over a gven time interval: actual arc energy since the assumed sinusoidal waveform is
not present. The energy difference increases as the arc
voltage as a percentage of the available system voltage
Earc = Varc Iarc dt increases. For example, in one series of single phase 6 cycle
0 arc tests with a system open circuit voltage of 630 volts rms,
Table 2 shows how the integrated arc energy per Equation (1)
If we assume that the arc voltage and current are both compares to energy calculated using Equation (3).
sinusoidal, the integral simplifies to the commonly known
expression: To insure accuracy in the calculation of arc energy, Equation
(1) was used to calculate arc energy for arc tests described
Earc = (Vrms) (Irms) (PF) (t) *I (2) later in the paper. Instantaneous voltage and current were
multipled by the sampling time interval and the products for
where arc Arc energy in watt-seconds
= each sample summed to obtain the arc energy for the total
Vrms Voltage across arc in rms volts
= duration of the arc.
Irms Arc current in rms amperes
=
PF = Cosine of phase angle between arc current and In a three-phase power system, it is possible to have multiple
voltage arcs occurring simultaneously, phase-to-phase and phase-to-
t = Time period in seconds ground, adding to the complexity of predicting arc energy in
actual installations. Predicting 3- phase arc energy is also
Since the impedance of an electric arc is essentially resistive made more complex by the fact that in electrical equipment,
i.e. the arc current is observed to be in phase with the arc conductor spacings are not uniform between all phases.
voltage, the cosine of the power factor angle becomes 1.0, and Typically phase a-c spacing is twice that of phase a-b and b-c
the expression simplifies further to: spacing.

Earc = (Vrms) (Irms) (t) (3) Maximum Arc Power:

A study of arc current and voltage waveforms, however, Ralph Lee calculated in [l]that during an electrical fault the
indicates that they are not pure sinusoids, but contain maximum available arc power, (PAR-) in kW, occurs when
significant harmonics. The arc voltage especially, tends to be the arc voltage drop equals the electrical supply system
more flat-topped as the arc voltage approaches the available voltage drop, and is equal to:
system voltage. Since any periodic waveform can be broken
down by Fourier analysis into a series of sinusoids with PARCM(kW) = 0.5 x Bolted Fault kVA (4)
increasing harmonic order, Equation (3) can be correctly
applied as long as the arc voltage and current waveforms are Lee stated that the maximum arc power occurs when the arc
first broken down into their respective harmonic components, voltage is 70.7% of the supply voltage and the arc current is

TABLE 2 - CALCULATED ENERGY USING EQUATION (3) VS. EQUATION (1)

395.1 62.7% 6 1080 1200 90.0%

- 283 -
70.7% of the bolted fault level. determine incident energy from electric arcs is direct
measurement.
The maximum arc power calculated per Equation (4) n a y be
different than is experienced in actual fault arcs for some of ARC TESTING PROGRAM
the following reasons:
An extensive testing program was undertaken at an
Equation (4) was derived based upon sinusoidal independent high current laboratory in Canada. The purpose
waveforms. As discussed earlier, the actual waveforms of the testing was to confhm the radiantkonvective energy
are not pure sinusoids, but contain significant harmonics, produced by electric arcs and to determine the effectiveness of
predominantly 3rd order. protective clothing in reducing and minimizing burn injuries.
Both flame resistant and standard work clothing were tested.
Equation (4) assumes that the supply system impedance An effort was made to correlate incident energy levels at set
is purely inductive. Actual power systems have a distances from the arc electrodes with the available
resistive component in their impedance as quantified by prospective short circuit current available at the arc terminals.
the system X/R ratio. Open circuit test voltages were selected at or above the
nominal system voltages of 600 and 2400 volts to allow
The arc voltage drop may not be equal to the supply simulation of phase-to-phase and 3-phase faults on power
system voltage drop during the fault. This assumption is systems. Single phase arcs and phase-to-phase arcs on a
less likely to be satisfied for medium voltage ( > 1000 v) three-phase power system are electrically equivalent as long
and high voltage ( > 46 kV) systems than low voltage as their source Thevenin equivalent circuits are the same.
systems. For medium and high voltage systems, the
normally used bus spacings in electrical equipment The test facility is capable of producing 60 Hertz fault
generally produce arc voltage drops during a fault that currents up to 100 kA rms at voltage levels up to 4 kV. The
are much less than the system voltage drop, and electrical supply feeds through a copper pipe coaxial bus to
consequently the arc power is less than the maximum the arc electrodes to reduce force on the test setup due to
possible. An example of this is mentioned in [8] where magnetic fields. The test facility is able to control the voltage
the authors report that the maximum arc power measured wave switching point to within 0.01 cycle. The closing
during fault testing of arc resistant switchgear at 10 kV switch was controlled to close the circuit at the correct time to
was less than 20% of the maximum 3-phase arc power minimize assymmetry in the current waveform.
calculated per [l]. Test results discussed later in the
paper will confrm this finding. Arc column stability was a paramount concern to insure
repeatability in the testing program. The test apparatus is
If Equation (4) is being used to predict the potential for injury enclosed in a separate test cell structure to eliminate the
to persons from incident arc energy, there are several other effects of rain, wind and non-uniform ambient temperature.
factors to consider. A significant percentage of the arc power In addition, all unnecessary metal plates or structures were
may be consumed in melting and vaporizing electrodes and removed from the test area. Since the test facility power is
other materials surrounding the arc. This portion of the arc supplied directly from a 13.8 kV tertiary winding of the utility
power would not be available as radiant energy. Data from 115/230 kV autotransformer, there was some minor variation
arc tests where electrode consumption was measured indicates in the open circuit voltage measured during a sequence of
that the energy consumed melting and vaporizing electrodes tests. Repeatability of arc current for identical test setups was
is typically 3-20% of the measured arc energy (based upon the generally within 2 percent.
assumption that 75% of the lost electrode was melted and the
remainder vaporized). The highest electrode loss occurs at In order to simulate electrical equipment, hard drawn copper
small electrode spacings, and decreases as electrode spacing electrodes, 314 inch in diameter, were used for the arc testing.
increases. The arc gap could be adjusted up to 16 inches. Arcs were
initiated by a light gauge fuse wire connected between the
In addtion, radiant energy from the arc is sigruficantly ends of the opposing electrodes.
affected by the environment surrounding the arc. Many arcs
occur in electrical equipment enclosures. In those cases, the Copper calorimeters have been used by the protective clothing
enclosure may tend to focus the radiant energy from the arc in industry since at least 1983 for performance testing (ASTM
one direction, increasing the potential hazard to personnel. Standards D4108, F955, and F1060). The center of the arc
The color and cleanliness of the enclosure walls also have an gap was aligned with the calorimeter on each test panel or
impact on the intensity of reflected radiant energy. For all of mannequin. The calorimeter temperature rise data in degrees
the above reasons, it is felt that the most accurate method to C. can be converted into incident energy in cal/cm2 and
be correlated with second-degree burn criteria developed by

-
T
Stoll and Chianta [9]. To calculate incident energy in
calories/cm2, multiply the copper calorimeter temperature r
rise, degrees C, by 0.135 calories/cm*-degrees C. Since
absorbed energy is generally equal to or greater than 90% of
incident energy for copper calorimeters, the two will be
considered as equivalent, and henceforth the term incident
energy will be used.

A data acquisition system recorded voltage, current, and

change in temperature for up to 12 copper calorimeters. Fiber
optic isolating devices eliminated electrical noise pickup by
the data acquisition system. Arc current, voltage, energy and
time duration as well as temperature rise for each individual
calorimeter were recorded for each test. Arc current and
voltage were &gtally sampled ‘every 0.1. milliseconds.
Calorimeter temperature rise was digitally sampled every 20
milliseconds.

ASTM F18.10 Subcommittee has recently approved two test

methods that use single-phase arcs to determine the m= U
ignitability and the thermal performance of fabrics used in
clothing for workers exposed to the electric arc hazard. The
first test method determines the ignitability of a textile
material in single or multiple layers. Fifty specimens of each Figure 3 - Test Setup #1 With A Mannequin In A Partial
textile material are exposed to an electric arc to determine Faraday Cage
their ignitability. This method can also identify textile
materials, such as rainwear, that although they pass the flame surrounding the arc. In practice, the arc is not a straight
exposure requirements of ASTM Specification F 1506, may vertical column. The arc moves horizontally away from the
dmer in ignitability when exposed to an electric arc. center line between the electrodes in a random manner,
leading to a significant variation in calorimeter temperature
The second method quantifies the thermal performance of a rise around the arc in a single test. Therefore, it was
flame resistant textile material. This method exposes a necessary to conduct multiple tests with the same
material to heat energy from an electric arc and measures the experimental parameters to achieve statistically significant
arc thermal performance value of the material. Material results.
performance is determined from the amount of heat energy
transferred by the material, and the observed effect of the Test setup #2 shown in Figure 4 was used per the ASTM
electric arc exposure on the material. procedure to determine the arc thermal performance value of
flame resistant fabrics. The test apparatus again allowed an
For the results presented in this paper, the test procedures open path for radiatiodconvection in all directions from the
used for fabrics and clothing were based on the two arc test arc and utilized a partial Faraday cage. Instead of
methods recently approved by the ASTM F18.10 mannequins, test panels with copper calorimeters were placed
Subcommittee and utilized panels and mannequins around and at fixed distances away from the arc. Both setups
instrumented with calorimeters. At present no other national #1 and #2 include side-mounted sensors for each panel or
or international standard test methods exist for measuring the mannequin to allow simultaneous measurement of incident
radiant energy produced by an electric arc. energy and energy transmitted through the fabric test
specimen.
Three different test setups were utilized. Test setup #1 shown
in Figure 3 is for the ASTM arc mannequin test for untreated Test setup #3 shown in Figure 5 was utilized to determine the
fabrics and was used to determine the ignitability of textile effect of boxlike enclosures on incident arc energy. The test
materials. This test setup provided an open path for apparatus used for the ASTM methods was modfied by
radiatiodconvection in all directions from the arc center line. removing the partial Faraday Cage and installing the test
Mannequins with copper calorimeters were placed around electrodes in a box-like enclosure (22” high x 14” wide x 13”
and at fixed distances away from the arc. The electrodes were deep) similar to electrical equipment enclosures, with a
encloied by a partial Faraday cage (part of the return removable top and bottom to facilitate test setup. The center
conductor) to provide balanced magnetic and electric fields line of the arc electrodes was 3.75 inches from the back of the

- 285 -
Figure 4 - Test Setup #2 With Panels In Partial Faraday Cage

30 Cycle Arc With 4 Inch Electrode

Spacing
2480 V Open Circuit -
Figure 5 Test Setup #3 With Arc Electrodes In Box

500
Radial Variation of Temperature Rise
400 Around Arc

300 I m

200 -Vottage (V rms)

100

0 Arc Arc Arc Arc

0
Figure 6 - Arc Variability Example Test Test Test Test
#I #2 #3 #4

box. Calorimeters on test stands were placed in front of the

box at different angles and distances from the center line of
the electrodes. Other aspects of the test setup were Figure 7 - Variability of Measured Arc Incident Energy
unchanged.
arc energy, current and voltage for each cycle of a 30-cycle
ARC VARIABILITY arc. Figure 7 shows the variability of incident energy when
measured by copper calorimeters at a uniform distance of 1
Arc phenomena have a high degree of variability. Multiple foot from the arc, with nine sensors uniformly spaced around
runs of the same test must be completed to obtain statistically a circle for each test. Each bar on the chart represents the
significant results. Even within a given test there is recorded temperature rise of one calorimeter. Results are
significant variability as evidenced by the plot in Figure 6 of from four tests with an identical initial setup of Arc gap - 4
- 286 -
inches, open circuit voltage - 2440 V, arc duration - 6 cycles,
available prospective fault current - 45.05 kA. The average Radial Variation of Energy Around Arc
of the measured arc power for the four tests is 17,970 kW.

Figure 8 shows a statistical breakdown of calculated incident

energy from readings of twelve calorimeters, located
circumferentially around and at a uniform distance from the
arc, for another series of four tests of 30 cycle duration with a
fault current of 4200 amperes. The setup for each test was
identical. The mean value of the 48 sensor readings was 6.39
cal/cm2,with a sigma of 0.8 1 cal/cm2.

OBSERVED ARC CHARACTERISTICS

An extensive series of single-phase arc tests was conducted

using test setup #2 to determine the characteristics of electric Incident Energy Ranges, cal/cm2
arcs for a specific 600 and 2400 volt system. Figure 9 shows
a trace of arc voltage and current for a typical 6 cycle (99.2
millisec) arc test with the following parameters:
Figure 8 - Radial Energy Distribution Around Arc
As Measured By Copper Calorimeters
Open circuit supply voltage - 630 Volts
Copper electrode spacing - 6 inches
Prospective available (bolted) fault current at the
electrodes - 45.2 kA
0 Arc voltage - 394.8 V rms
-
Arc current 30.6 kA rms

The voltage drop across the arc is 394.8 V or 63% of the open
circuit supply voltage. Note that the arc voltage is flat-topped
versus the arc current which is nearly sinusoidal. The arc
impedance reduced the fault current from the prospective
(bolted) value of 45.2 kA.down to 30.6 kA. The total arc
energy during the 6 cycle period was 1100 kW-sec, yielding
an average arc power of 11,090 kW. The length of copper Time, millisec 1 -Current I
electrode consumed in the test was 518 inch. The maximum I
I
-
Voltage '
possible arc power for this single phase setup per [l] is 0.5 x
630 V. x 45.2 kA or 14,238 kW. The ratio of actual to max.
preQcted arc power was 78%.
-
Fig. 9 Arc Voltage and Current Waveforms

Figure 10 shows the temperature rise of one copper Calorimeter Temperature Rise
calorimeter for the same test shown in Figure 9. The incident
energy is determined from the maximum temperature rise
experienced during the test.

A series of 6 cycle arc tests was conducted to determine the

effects of increasing arc electrode spacing for one 600 volt
system. Figure 11 shows rms arc voltage and current as a
function of electrode spacing.

Arc impedance as a function of electrode spacing for the same 0 2 4 6 8 10

600 volt system is shown in Figure 12. Arc impedance as a
Time, seconds
percentage of the electrical system impedance is shown on the
right hand scale. Based upon Ralph Lee's prediction in [I],
the maximum arc power should occur when the arc
impedance is equal to the system impedance. Figure 12 Figure 10 - Copper Calorimeter Temperature Rise

- 207 -
 600 Volt System - Arc Voltage & 2400 Volt System -Arc Voltage &
Current Current
Prospective Fault Current 45.2 kA Prospective Fault Current 45.1 kA
500 40 600 445
d 400 30 E- -
a, 5 0 0 44.0 *’
ED 2 300 al 0) 400 C
43.5 2
a= 59 5 3 ~a
’
2
> 200
20
10
U
5
5 5 300
: ’ 200 43.0 5
Q 100 G loo 42.5 $
0 0 0 42.0
1 2 3 4 5 6 7

Arc Electrode
Spacing, Inches
L-C)-Arc C u r r q

Figure 11 - Arc Voltage & Current As Function of Electrode Figure 14 - Arc Voltage & Current As Function of Electrode
Spacing For 600 Volt system Spacing For 2400 Volt System

600 Volt System - A r c Impedance 2400 Volt System -Arc Impedance

15.00
6
2 10.00

!ig
0
E 5.00

4
0.00 h l b ( D c 0 ~ ~
- ( u c ) P L o ( D l -

Arc Electrode Spacing, Inches Arc Bectrode Spacing, Inches

Figure 12 - Arc Impedance vs. Electrode Spacing For 600 Figure 15 - Arc Impedance vs. Electrode Spacing For 2400
Volt System Volt System

600 Volt System -Arc Power & Incident 2400 Volt System -Arc Power & Incident
Energy Energy
@ 2 F t Distance From Arc @ 2 Ft Distance From Arc
11500 5.00
l l m 4.00
3.00 5 @ “E
2.00
1 .00
-g2
;
C W 0

0.00
1 2 3 4 5 6 7
2 4 6 8 1 0 1 2
Arc Electrode
Spacing, Inches
-o- Incident kergy Spacing, Inches

Figure 13 - Arc Power & Incident Energy As A Function of Figure 16 - Arc Power & Incident Energy As A Function of
Electrode Spacing For 600 Volt System Electrode Spacing For 2400 Volt System

- 280 -
predcts, on that basis, that the max. arc power for this 600
volt system should occur at an electrode spacing of 6 inches. Incident hergy As A Function Of Distance
-
From The Arc 600 Volt System (!j”Arc Gap)
Figure 13 depicts average arc power and incident energy two 45.2 kA Prospective Fault Current
feet from the arc for the same 600 volt system. The 120 OccGo/o
maximum arc power appears to occur between an electrode
spacing of 3 and 4 inches, a somewhat smaller spacing than
s 100 Om4YO

predicted by Ralph Lee’s theory. The maximum measured

F
O)N
80
OWh
? €
Q ”
incident energy occurred at an electrode spacing of 6 inches. 155 60 m R
It is conjectured that the reason maximum arc power and
- 4
r m
o 4.0
O.m20/0 pg
P,
0a31Yo
$15
incident energy occur at different electrode spacings, may be
due in part, to an improved arc radiation transfer function as
-2 20
00 000CO/o
arc length increases. 1 1.5 2 3 4
Ustance From Arc, Feet
Considering results from all of the 600 volt arc tests (Nonlinear Scale)
simulating a phase-to-phase fault, the maximum arc power
measured was 11,770 kw, equal to 79% of the theoretical
maximum single phase arc power calculated per [l]. -
Figure 17 Incident Energy as a Function of Distance from
the Arc For 600 Volt System
Figure 14, 15 and 16 show the same arc data for a 2400 volt
system. Arc impedance at a given electrode spacing for these Einc = K Earc (5)
600 and 2400 volt systems is nearly the same. Note that in D2.2
Figure 15, the arc impedance as a percent of system
impedance is much lower than for the 600 volt system, less where Einc = Incident Energy in cal/cm2
than 25% for the largest tested electrode spacing of 12 inches. K = 0.0000432 feet/cm2
Based upon Ralph Lee’s theory, this would lead one to believe D = Distance From Arc in Ft.
that the maximum arc power would occur at electrode Earc = Total Arc Energy in calories
spacings greater than 12 inches. The data shown in Figure
16 is inconclusive as to what electrode spacing will produce Ralph Lee predicted that incident energy varied as the inverse
maximum arc power and incident energy for this 2400 volt of the distance squared from the arc in [1], very close to the
system.. experimental exponent value of -2.2 in Equation 10.

Considering results from all of the 2400 volt arc tests EFFECT OF “ARC IN BOX” ON INCIDENT ENERGY
simulating a phase-to-phase fault for electrode spacings of 12
inches and less, the maximum arc power measwed was A series of tests was conducted to determine the effect on
23,460 kw, equal to 41.8% of the theoretical maximum single incident energy of enclosing the arc electrode, in an
phase arc power calculated per [ 11. ungrounded metal box. The tests used to produce the results
in Figure 13 were repeated using test setup #3. Figure 18
VARIATION OF INCIDENT ENERGY WITH DISTANCE shows a comparison of the measured incident energy with the
FROM THE ARC open arc (test setup#2) versus with the arc in a box. Two
curves are plotted for the box tests, one indicating the average
It is important to understand incident energy as a function of incident energy measured using calorimeters positioned at the
distance from the arc. Using test setup #2, heat input to center line of the electrodes, the other indicating the average
copper calorimeters was measured at various distances from incident energy of sensors located both at the center line and
the arc for a 600 volt system. A total of 6 calorimeters were at 7 inches above the center line. Note that at electrode
uniformly spaced in a circle around the arc for each test. The spacings of 4 inches and less, the box has the effect of
results of four 6-cycle arc tests (identical initial test increasing the incident energy 118-142%. These test results
conditions) were averaged (24 calorimeter readings) to are good only for the specific box design and size, and for
determine the incident energy for each distance shown in electrode positions used in test setup #3. Results may vary
Figure 17. Total arc energy for the series of tests was very with different box designs.
consistent and averaged 1084 kW-sec (259082 cal). Incident
energy as a percent of total arc energy is shown on right hand Finite element modelling was used to independently calculate
scale of Figure 17. the impact on incident energy of enclosing the arc in a box.
The arc electrodes, box, calorimeters and mounting panels
The incident energy curve in Figure 17 can be closely were modelled in detail to match test setup #3. An arc
represented by Equation (5):
- 289 -
 fault current (33.5 kA). For this 600 volt system, phase-to-
Incident Energy - Open Arc vs. phase prospective fault current was measured to be 81.3% of
Arc In Box the 3-phase prospective fault current. This measured ratio is
@ 2 Foot Distance From Arc close to the theoretical value of 86.6% predicted by
3 symmetrical component analysis of bolted faults on 3-phase
N power systems.
5 2.5
1
2 2 Results of four tests with the same setup were averaged to
s reduce the effect of arc variability. Incident energy was first
F 1.5 measured for 3-phase arcs in the box, and then the test was
al
c
w 1 repeated for phase-to-phase arcs. Both the 3-phase and
U
C phase-to-phase tests were repeated again with the box
z0 0.5
grounded to B-phase to determine what effect box grounding
QJ

-= o has on incident energy. Figure 19 shows the results of the arc

test series for electrode spacings of 2 and 3.75 inches. The
incident energes shown in Figure 19 are the average of
- Hectrode Spacing, +Open Arc readings from 7 calorimeters in front of the box at a distance
inches - Box(C/L)
-1) of 2 feet from the center electrode.

Several important conclusions can be drawn from Figure 19.

Incident energy from the 3-phase arcs was from 3.2 to 4.3
Figure 18 - Effect of Box On Incident Energy times higher than from the phase-to-phase arcs for the
ungrounded box. This ratio reduced to 2.5 times higher for
temperature of 15,000 O C. was used and the arc was assumed the grounded box. Incident energies were slightly higher for
to be an ideal blackbody. The computer runs simulated 0.1 the 2 inch vs. the 3.75 inch electrode spacing. This may be
seconds of real time. A case was run with and without the due to the effect of magnetic forces lengthening the arc, such
presence of the box around the arc electrodes to allow a that actual arc length for the 2 inch spacing was close to the
comparison of incident energy. Results of the simulation
indicated that when the box surrounded the arc, the incident
energy measured by the calorimeters increased to 129% of the Incident Energy Comparison @ 2 Feet
incident energy measured with an open arc. 3 Phase vs Phase-Phase Arc In A Box
4.0
THREE PHASE “ARC IN BOX” TESTS
3.5
Test setup # 3 was further modified by adding one additional
electrode in the box to allow testing with three-phase faults. 3.0
All three copper electrodes (3/4 inch diameter) were
positioned to enter the top of the box (parallel to the back of
the box) and were terminated at the mid-point of the box.
35 2.5
s
Tests were run with two different electrode spacings, 2 and, ;2.0
3.75 inches between adjacent electrode center lines. This test C
w
setup was different from the other test setups in that the 1.5
electrodes were in parallel and were not aligned end-to-end.
f
‘p
With electrodes in parallel, the magnetic forces produced by e
- 1.0
the adjacent conductors tended to push the arc down to and
away from the end of the electrodes, increasing the total arc 0.5
length and the arc voltage.
0.0
A series of 6-cycle arc tests (610 volts open circuit) was
conducted for one 600 volt system to determine the difference
between incident energy (at a distance of 2 feet) from 3-phase
and phase-to-phase arcs in the box. First, all three phases
were shorted together and the prospective (bolted) fault
current was measured to be 41.2 kA. Then one phase
electrode was removed and the two remaining phases were
shorted together to measure the phase-to-phase prospective Figure 19 - 3 Phase “Arc In Box” Test Results
- 290 -
 “white burns” on rats (similar to second degree burn for
length which produces maximum arc power. human tissue) at exposure times from 0.5 to 10 seconds. In
this work, approximately one cal/cm2 absorbed energy for a
Grounding the box decreased the incident energy from the 3- one second exposure causes a “white burn” on a depilated rat
phase arcs. For the phase-to-phase arcs, grounding the box (with hair removed).
caused a decrease in incident energy with 2 inch electrode
spacing, but an increase with 3.75 inch spacing . For the Stoll and Chianta [9] also report measured and calculated
3.75 inch spacing, the actual arc length in the ungrounded radiant energy exposures for human skin which indcate that
box was approaching the maximum sustainable arc length, slightly more absorbed energy is required to cause a 2nd
and grounding the box may have shortened the arc length and degree or “blister” burn injury for human tissue than is
increased the arc energy since the distance from the outer required for a white burn on a rat. They indicate that human
electrodes to the grounded wall of the bdx was probably less tissue will sustain second degree burn damage at an absorbed
than the phase-to-phase arc length. heat energy of 1.2 cal/cm2for a 1 second exposure time. This
exposure creates approximately a 30°C temperature rise at the
CORRELATION OF INCIDENT ENERGY WITH 2ND base of the epidermis in human s h n which is cited at a skin
DEGREE BURNS OF BARE HUMAN SKIN depth of 80 microns (3.15 mils). The human tissue response
is extrapolated to one second based on exposures of 5.6
Prediction of Burn Iniurv In Human Tissue: seconds and longer. It is reported in very early work [ 161 at a
0.2 second exposure time using a carbon arc search light
It is well known that burn injury in human tissue is a function radiant heat source that 3.2 cal/cm2 heat energy caused a
of elevated tissue temperature and the period of time for second degree burn in human skin.
which the elevated tissue temperature is maintained. Burn
injury is reported to begin at a very slow rate (hours) at a Privette [12][13] empirically extended the work of Stoll to
tissue temperature of 44”C, and the rate of tissue damage exposure times down to 0.01 second and found for very short
increases rapidly (seconds) as the tissue temperature exceeds exposures of from 0.1 to 0.01 seconds that 1.5 to 2.0 cal/cm2
60°C [lo]. is required to cause a white burn on a depilated rat. For short
exposure times of one second or less, the total absorbed
Stoll and Chianta projected, based on an extrapolation of energy required to cause a second degree burn in human
their data, that at an epidermal base temperature of 72”C, tissue, i.e., destruction of the epidermis, ranges between 1.2
“epidermis destruction is virtually instantaneous.” Tissue and 3.2 cal/cm2. The more recent data in [12] and [13]
temperature is reported as the normal skin temperature of converges on the range of 1.2 to 1.6 cal/cm2. Consequently,
33°C plus the temperature rise created by radiant and the focus of this paper will be on this energy range, erring on
convective heat absorption. Lee [l] cites the work of Stoll the conservative side. Since absorbed radiant energy is
and Chianta [9], but utilizes an extrapolation of the work of generally over 90% of the incident energy, we will treat the
Artz [111 to develop a skin temperature of 80°C as the point two as equivalent, again erring on the conservative side.
at which a “curable” burn occurs for an exposure time of 0.1 Approximate values for skm temperature rise AT‘S. skin
seconds. This “curable burn” is comparable to the blister or temperature (ATs + 33 deg. C), copper calorimeter
second degree burn cited by Stoll and Chianta. There is good temperature rise ATc, and total exposure energy for a range of
agreement between Lee’s [l] extrapolation of the calculated heat flux levels and exposure times are provided in Table 3.
skin model temperatures of Artz [ 111, which were used to
determine a “curable burn,” and the measured values of Stoll These values for skin temperature in Table 3 are estimates
and Chianta [9] extended to very short exposure times by since there are many factors which can influence the onset of
Privette [12][13], which were used to determine a second burn injury in human tissue, including variable thickness of
degree burn. Since the work in this paper is based on the epidermis, water content in the skin and the amount of
measured incident energies due to electric arc exposures, the water on the skin surface due to perspiration or other sources
results of this paper are based on the more empirical work of [17]. In addition, burn injury may occur from molten and
Stoll and Chianta [9][14][15] and Privette [12][13] which can vaporized metal released during an electric arc incident or
be used to relate measured heat energy absorbed by human from secondary sources of heat energy caused by flammable
tissue to second degree burn injury. material ignition during the electric arc incident.

Predicting Second Deaee Burn Iniuw Based On Measuring SHORT RADIANT HEAT EXPOSURES &
Heat Enerpy Absorbed By The Skin: TEMPERATURE PROFILE IN HUMAN SKIN

There are a number of reported approaches for predicting Stoll and Chianta [15] relate exposure time to tissue damage
second degree burn injury based on heat energy absorbed by time by considering tissue cool down time as well as exposure
human or animal skin. Stoll and Chianta [ 141 report data for time to obtain a total tissue temperature profile. As exposure
- 291 -
TABLE 3 - SKIN AND COPPER CALORIMETER TEMPERATURE RISE FOR A RANGE O F EXPOSURE
CONDITIONS WHICH RESULT IN A SECOND DEGREE BURN

Incident Exposure Total Max. Est. Max. Est. Copper

Heat Flux Time Exposure Exposed Skin** Calorimeter
cal/cmz- Energy* Skin** Temp. ATc
seconds seconds cal/cm2 ATs, O C OC O C

200.0+ 0.01 2.0 57 90 14.8

30.0+ 0.05 1.5 43 76 11.1
15.0+ 0.1 1.5 43 76 11.1
0.5 1.35 38 71 10.0
1.2- 1.0 1.2 30 63 8.9
0.55-H 3.0 1.7 25 58 12
0.23-H- 10.0 2.3 22 55 17
* Total exposure energy to get a second degree burn.
** Estimated temperature rise and temperature for 80 micron skin depth at the base
of the epidermis.
+ Reference [12]
U Reference [9] - “Stoll Curve”

times decrease, the cool down time becomes an increasing

component of the total tissue damage time. Since human skin Human Tissue - Tolerance to 2nd Degree
is not a good conductor of heat, time is required for heat Burn
incident on the skin surface to penetrate 80 microns to the
base of the epidermis. M e r very short exposures, the skin
surface is rapidly cooling while the heat penetrates to the
epidermis. Consequently, heat transfer rate to the base of the
epidermis is reduced. As a result of this post-exposure cool-
down effect, the heat energy required to cause a second
degree burn increases for very short exposure times as shown
in Table 3 and Figure 20. In addition, for the very high heat
fluxes in Table 3 (times of < 0.1 second), the skin surface
temperature can exceed several hundred degrees C., and
additional heat loss mechanisms may come into play. At
longer exposures of 3 to 10 seconds in which the lower heat
flux results in lower skin temperature, the total exposure
energy required for a second degree burn increases due to the k p o s u r e Time, Seconds
impact of slower kinetics for protein degeneration at the
lower skin temperature.

In order to convert energy values into safe approach Figure 20 - Stoll Curve Extended To Very Short Exposure
distances, the arc hazard must be quantified based on Times
incident arc energy calculated by computer programs or
measured empirically using heat sensors, e.g. copper with the distances reported in NFPA 70E [ 5 ] . The NFPA 70E
calorimeter sensors. Assuming phase-to-phase arcs, Tables 4 data relates to a 3-phase arc exposure and the comparative
and 5 provide some typical arc parameters and the distance data provided in Table 5 relates to a phase-to-phase arc
from the arc for onset of a second degree burn for 0.8, 1.2, exposure. For this comparison, the same arc current was used
and 1.6 cal/cm2incident energy levels. The value 1.6 cal/cm2 for the phase-to-phase arcs as is specified for the bolted fault
is the level at which a second degree burn would occur for a current in NFPA 70E Table B-1. The actual arc current
short time (<0.5 second) exposure of bare skin. The values would be less than 86% of the 3-phase bolted-fault values. A
of 0.8 and 1.2 cal/cm2provide an additional safety factor. comparison of NFPA 70E “safe” distances” with distances
determined by computer software, indicates an inconsistency
Table 5 also compares &stances calculated by this approach as system voltage increases. For example, for 600 volt

- 292 -
TABLE 4 - DISTANCES FROM PHASE-TO-PHASE ARCS AT WHICH THE ONSET OF A SECOND DEGREE BURN
IS PREDICTED ON EXPOSED SKIN

System Arc Arc Arc Distance In Inches From Arc Center Line
Voltage Current Duration Electrode Phase-To-Phase Arc
Spacing Incic nt Energy Lt
kV kA Cycles (sec.) Inches 0.8 caYcmZ 1.2 caYcm2 1.6 cal/cm2
S 60 10 (0.167) 12 142 116 100
5 30 10 (0.167) 12 90 74 64
5 15 10 (0.167) 12 60 49 43
5 8 10 (0.167) 12 42 34 29
0.6 40 6 (0.10) 4 49 40 34
0.6 20 6 (0.10) 4 32 26 23
0.6 10 6 (0.10) 4 21 17 1s
0.6 40 30 ( O S ) 4 109 89 77
0.6 40 l(0.017) 4 20 16 14

TABLE 5 - DISTANCES FROM PHASE-TO-PHASE ARCS AT WHICH THE ONSET OF A SECOND DEGREE BURN
IS PREDICTED ON EXPOSED SKIN COMPARED TO DISTANCES CITED IN TABLE B-1 OF
NFPA 703-1995

System Arc Arc Electrode Distance From Arc Center NFPA 70E
Voltage Current Duration Spacing Line for Various Table B-1
Incident Energies Distance
Phase-To-Phase Arcs Inches”
For
Cycles 3 Phase
kV kA (sec.) Inches Arcs
230 23 6 (0.10) 20 552
13.8 31.3 6 (0.10) 12 169.2
13.8 31.3 l(0.017) 12 69.6
4.16 25 6 (0.10) 6 87.6
4.16 12.6 6 (0.10) 6 66.0
0.6 44 6 (0.10) 4 44.4
0.6 44 0.25(0.004) 4 8.9
0.6 26 6 (0.10) 4 33.6
0.6 17 6 (0.10) 4 27.6
* Source - NFPA-70E, Part 11, Appendix B, Table B-1. Arc length is unspecified for the NFPA 70E data.

systems, the NFPA 70E and computer generated “safe” PERFORMANCE OF NON-FLAME RESISTANT
distances are almost equal, whereas for high voltage systems, CLOTHING
the NFPA 70E distances are an order of magnitude different.
Furthermore, for 600 Volt systems, if one considers the The most serious arc burn injuries result when untreated
potential increases in incident energy due to “arc in the box” clothing is ignited during an arc incident. This is due to the
effects, the NFPA 70E distances may be less than adequate. increased body area which can sustain a second or third
degree burn injury as a result af the extended exposure time
In applying the information in Table 5 , appropriate safety caused by ignited clothing. Table 6 shows an ignition
factors should be built into work rules. An important analysis for a 100% untreated 5.4 ozlyd2 dark blue cotton
consideration for these safety factors is variability due to arc shirt tested using setup #l under controlled laboratory
movement as discussed earlier. conditions.

- 293 -
TABLE 6 - IGNITION ANALYSIS OF 100% PERFORMANCE OF FLAME RESISTANT CLOTHING
UNTREATED 5.4 OZ/YD2 DARK BLUE SYSTEMS
COTTON SHIRT
The use of Flame Resistant (FR) clothing whtch does not
Incident Energy Ignition Analysis ignite can greatly improve the chance of survival by reducing
caI/cm2 the body area which sustains a burn injury. The vital first
6.6 Shirt Ignites 90% of the Time step in the selection of arc protective clothing is to use FR
6.0 Shirt Ignites 50% of the Time clothing as the outer layer of clothing whenever the arc
5.4 I Shirt Ignites 10% of the Time hazard assessment indicates sufficient incident energy for
4.5 I
Shirt Ignites 1% of the Time ignition of untreated clothing. In Table 7, an incident energy
range where untreated clothing is unlikely to ignite, is set at
I 3s I
Shirt Ignites 1% of the Time
(95% Confidence Level) I 0-2 cal/cm2 based on test results in Table 6 (Test setup #1)
and ignition probability considerations discussed above. For
arc hazards with incident energies in excess of 2 cal/cm2, an
Table 6 is based on measured incident energy and does not FR clothing system is indicated. In Table 7 a classification
consider the effect of arc movement (variability). Under the for clothing systems is suggested which relates to the number
controlled laboratory conditions required to meet the new of layers for FR clothing systems and arbitrarily designates
ASTM test methods, the affect of arc movement reduces the untreated cotton as “Class 0,” i.e. zero FR layers.
incident energy required for 1% ignition to approximately 2.5
cal/cm2with a 95% confidence level. The higher variability The use of an FR clothing system does not preclude the
due to arc movement in real arc accidents would be expected possibility of sustaining a burn injury. As shown in Table 7,
to further reduce the incident energy required for 1% ignition the number of layers of FR clothing must be increased as the
probability. Heavier weight cotton fabrics, fabrics with high arc hazard incident energy increases. For instance, if the arc
moisture content (perspiration), and lighter colored cotton hazard assessment determines an incident energy of from 5-
fabrics will require greater incident energy for ignition. 10 cal/cm2, a Class 2 FR clothing system is indicated. The
Conversely, lighter weight cotton fabric and darker colored Class 2 system consists of two layers, e.g. an FR shirt and FR
cotton fabric will require less incident energy for ignition. pants as the outer layer, and untreated or FR underwear as the

TABLE 7 - PROTECTIVE CLOTHING GUIDELINES FOR THE ELECTRICAL ARC HAZARD

Exposure Value (ATPV)”

++ Proposed Range of Incident Energy To Avoid A Second Degree Burn

* ATPV was measured using test setup #2 and is defined in the ASTM F18.10 standard test method for FR
fabrics as the incident energy that would just cause the onset of a second degree burn.
** ATPV was measured using test setup #1.

- 294 -
second, inner layer. Based on test results, the onset of a hazards. For instance, an electrician may be wearing a Class
second degree burn would be expected to occur in the range 2 system (Table 7) for a specific maintenance task, but may
of 10-17 cal/cm' for Class 2 FR clothing systems. For more need to increase to a Class 3 system by adding an FR coverall
severe arc exposures, additional layers of FR clothing are for a brief time period while performing a specific switching
indicated as shown in Table 7. Also it is important to note task.
that areas of the body which are not covered by the clothing
system may sustain burn injury. Protective items for the face, Because of differences in electrical system impedances and
head, hands and feet are discussed in a later section. conductor geometries, and limited knowledge concerning 3-
phase arc incident energies and "arc-in-a-box" effects, arc
The protective clothing classes in Table 7 are provided as hazard analysis is necessarily limited to an estimating process
guidelines based on extensive laboratory testing, and may not at this time. For each specific task involving a potential arc
correlate directly to real arc incidents. Real arc exposures exposure, a hazard analysis may be conducted by the
may be more or less severe than laboratory arc exposures, due following estimating process:
to greater arc movement toward or away from the exposed
worker, unknown arc length, system reclosure, secondary A) Identify the arc parameters listed above, considering the
explosions or fires, weather conditions, and a host of other probable worst case exposure scenario.
factors. The intent of providing the information iii Table 7 is
not to suggest that all burn injuries can be eliminated in arc B) Enter tlie values for the first five arc parameters into
exposures, but rather to indicate clothing systems that will currently available software programs to calculate the
reduce or minimize burn injuries during arc exposures. incident energy for a phase-to-phase (single phase) arc in an
open configuration. If the arc hazard involves a three-phase
ARC PROTECTIVE CLOTHING SELECTION BASED ON system, then additional steps are necessary to estimate the arc
HAZARD ANLYSIS current. First, identify the prospective or bolted fault current
for tlie three-phase system. Then, convert tlie three-phase
The first step in selecting the appropriate arc protective bolted fault current to a calculated phase-to-phase bolted fault
clothing system is to identify and quantify the arc hazards to current by multiplying by the factor 0.866. It is important to
which workers can be exposed. All tasks which involve note that the actual value for the phase-to-phase arc fault
potential exposure to the arc hazard must be evaluated. current can be significantly lower than the bolted fault
current. particularly for low voltage systems as shown in
The following information is required to properly evaluate Figures 11 and 14. For example, for the 600 volt system
each potential arc hazard: tested, the phase-to-phase arc current for the niaxiinuni
incident energy level (see Figures 11 and 13) is 32 kA vs. a
Electrical system voltage phase-to-phase bolted fault current of 45.2 kA. For the 2400
Fault current available at the arc location volt system tested at a maximum 12 inch spacing, the ghase-
Arc time duration i.e. the number of cycles (including to-phase arc current for the maximum incident energy level
possible reclosing). (see Figures 14 and 16) is 44.1 kA vs. a ghase-to-phase bolted
Estimated arc gap based on equipment geometry. fault current of 45.1 kA. The arc fault current value which is
Minimum distance from the center line of the arc to tlie to be entered into the software program is, in the worst case.
worker. the phase-to-phase bolted fault current. As indicated above,
Number of phases, single phase or 3-phase. the actual expected phase-to-phase arc current can be
Configuration of the arc, e.g. open arc or "arc in a box." significantly less than the bolted fault current for low voltage
In an open arc configuration, the arc energy emanates in systems.
all directions (360 degrees around the arc). In the "arc in
a box" configuration, the arc energy emanates out the C) Identify appropriate correction factors If the arc hazard
front of the box (180 degrees or less). The box also tends involves a three-phase electrical system, the value determined
to focus hot gasses and molten metal splatter in one using the software program for a single-phase arc must be
direction. multiplied by a correction factor Based on the limited testing
performed, this factor may be as high as 4 3 for an
The intent of hazard analysis is to determine the probable ungrounded box and 2 5 for a grounded box
worst case exposure which a worker could receive for each
specific job or task. It is likely a worker will be involved in If the arc hazard involves an "arc in a box" configuration
tasks with several different exposure levels during the course instead of an open arc, there is a possible additional
of a work day, and the appropriate arc protective clothing correction factor of 1.5. This "arc in a box" factor is an
may be different for each of the potential arc exposure estimate based on limited data for a single box configuration

- 295 -
(setup #3). The actual factor may be higher or lower When the outer flame resistant layer is unable to resist
depending on the specific geometry of the electrical breakopen, the second layer underneath must be flame
equipment cabinet or box and the arc electrode spacing resistant. Breakopen is defined as the creation of holes,
tears or cracks in the exposed fabric such that incident
Once the incident energy for a specific arc hazard is energy is no longer effectively blocked by the fabric and
determined, Table 7 provides information to assist in the passes to surfaces below the fabric.
selection of the appropriate arc protective clothing system.
For instance, if the total corrected incident energy equals 15 Meltable synthetic fibers must be avoided in fabric
cal/cm*, based on Table 7 a Class 3 protective clothing system underlayers next to the skin. These fibers, when they
would be expected to reduce or minimize burn injury. melt, can adhere to the skin, transfer heat rapidly, and
exascerbate burn treatment.
Bare Skin Exmsure:
For a specific fabric type, protective performance generally
If the arc hazard analysis indicates an incident energy of less increases with fabric weight and thickness when energy
than 0.8 cal/cm2, exposed skm would not be expected to exposure levels are less than required to cause ignition or
receive a second degree burn injury. If the analysis indicates decomposition of the fabric. Multiple layers of clothing are
an incident energy equal to or greater than 1.6 cal/cm2, generally more effective than a single layer of the same
exposed skm would be expected to receive a second degree weight. An effective protective clothing system consists of
burn, and thus, protective clothing and protective equipment one or more flame resistant outer layers without openings,
must be used to minimize burn injury. For incident energy coupled with non-meltable fiber inner layers of sufficient
values between 0.8 cal/cm2 and 1.6 cal/cm2, it is advisable to thickness and weight to minimize burn injuries to the skin.
use arc protective clothing and protective equipment to
minimize burn injury since the actual parameters for an arc PROTECTIVE GUIDELINES FOR THE FACE,HEAD,
fault incident are difficult to predict accurately. HANDS AND FEET

In some cases, an arc hazard analysis may stimulate an Chapter 3 of NFPA 70E-1995 [5] effectively summarizes
assessment aimed at reducing the arc exposure level for a current standards for personal protective equipment and
particular task. Equipment or work procedures may need to commonly used types of protective equipment for the face,
be modified to reduce the exposure levels. head, hands and feet.

GENERAL GUIDELINES FOR ARC PROTECTIVE Typical head protection for arc blast exposure is a
CLOTHING switchman’s hood or a faceshield, both with the viewing
window constructed of 0.080 inch thick polycarbonate. The
The following general guidelines have been developed by the face shield is more appropriate for use in situations with
authors to assist personnel in reducingminimizing burn relatively low radiation exposure. When wearing a face
injury from arc exposure. These guidelines are based on the shield, a person may be subject to burns from hot gasses
knowledge gained from conducting thousands of arc tests and which envelop the head and contact the side of the face and
analyzing the exposed clothing and the associated burn back of the head.
injuries from actual arc and flame exposures:
The switchman’s hood should be used for high exposure
Protective clothing selection must be based on probable situations and when additional protection is desired. Limited
worst case exposure for a task. arc testing indicates that polycarbonate provides an effective
protective barrier for the face for low and moderate levels of
Since severe burn injuries from arc exposure incidents radiant energy exposure. At very high exposure levels and
most often involve the ignition of flammable clothing, times, the polycarbonate can melt, ignite and/or disintegrate.
flammable fabrics on the outer layer of protective For these cases, methods for reducing exposure energy should
clothing must be avoided when the hazard analysis be explored.
indicates that clothing ignition is possible.
Minimum protection for hands is a pair of heavy duty leather
Outer flame resistant layers must not have openings that work gloves. There have been a number of recorded arc
expose flammable underlayers. incidents in which the injured person was wearing leather
gloves and received no injury to his hands, whereas other
Tight-fitting clothing must be avoided. Loose fitting parts of the body were seriously injured. If exposure to
clothing provides additional thermal insulation due to air electric shock is also a hazard, then appropriate rubber
spaces. insulation equipment must also be worn. The authors in [6]

- 296 -
report that normal leather protectors required by Class 0 or 2) Conduct an arc hazard analysis for each task for which
higher rubber gloves provide adequate protection from flash. there is a potential arc exposure. Define the probable
Experience from actual arc incidents shows that the cotton maximum incident energy exposure levels based upon
stitching in leather gloves frequently disappears during the electrical system parameters and the proximity of the worker
arc exposure, but remains in place long enough to provide to the energized components.
adequate protection [6].
3 ) Select protective clothing that is suitable for the probable
The highest level of hand protection is provided by gloves maximum incident energy exposure.
made using layers of flame resistant materials. Since the
hands may be closer to the arc than the body, this additional The employer with the employee has the responsibility to
protection may be important to prevent injury to the hands, assess the arc hazards involved and to select appropriate
wrists and forearms in some arc exposure’situations. protective clothing and equipment. Use of the above
procedure will assist in this selection process.
There have been few recorded incidents where feet have
received burns from arc exposure since the feet are normally ACKNOWLEDGEMENT
below the work area, and are seldom in the line of sight from
the location of an arcing fault. Heavy duty leather work shoes The authors wish to acknowledge the invaluable assistance
normally provide a significant degree of protection to the feet provided by the staff of the Ontario Hydro Technology’s High
The ankles, however, may be subjected to injury if the person Current Laboratory and to thank J. W. Hunt of DuPont
is not wearing adequate leg protection which extends fully Engmeering for preparing the finite element analysis of the
down past the top of the shoes. “arc in the box” setup.

CONCLUSIONS REFERENCES

A comprehensive test program has been carried out to [ 11 Ralph Lee, “ The Other Electrical Hazard: Electrical Arc
quantify the characteristics of electric arcs, the incident Blast Burns,” IEEE Trans. Industrial Applications, Vol.
energy produced at specified distances from electric arcs, and 1A-18, No. 3 , P. 246, MaylJune 1982.
the capability of untreated and flame resistant fabrics to
protect human s h n from injury due to arc exposure, The [21 C. M. Kent and H. L. Floyd, “Managmg the Other
authors have considered the impact of arc variability and have Electrical Hazard: Electric Arcs,” American Society of
discussed the need for safety factors when using the Safety Engneers Forum titled “Safety Technology 2000,”
guidelines. Orlando, Florida, June 19, 1995.

Results of arc testing indicate that the curable burn distances [3] M. Capelli-Schellpfeffer, M. Toner. R. C . Lee, R. D.
predicted by NFPA 70B-1995 [5], based upon Ralph Lee’s Astumian, “Advances in the Evaluation and Treatment of
theoretical calculation of maximum arc power, may need to Electrical and Thermal Injury Emergencies.‘ IEEE
be re-evaluated. “Safe” approach distances in NFPA 70E for Petroleum and Chemical Industry Conference Record,
high voltage systems may be too conservative, whereas for September 1994, pp. 301-307
low voltage systems, the distances may not be adequate. Test
results indicate that for a gven system voltage, maximum arc [4] J. R. Saffle, B. Davis, P. Williams, & The American
power and maximum incident energy from arc exposure may Burn Assoc. Registry Participant Group. “Recent
occur at different electrode spacings. The authors plan to Outcomes in the Treatment of Burn Injury in the
submit results of this testing to the NFPA 70E Technical United States: A Report From the American Burn
Committee for their review. Association Patient Regstry,” Journal of Burn Care &
Rehabilitation, Vol. 14, No. 3 , Part 1, May/June 1995.
The recommended procedure to minimize worker burn injury
and to select adequate arc protective equipment can be [5] NFPA 70E-1995, “Standard for Electrical Safety
summarized as follows: Requirements for Employee Workplaces.” 1995 Edition.

1) If at all possible, plan electrical maintenance work and [6] S . Janiil, R. A. Jones, and L.B. McClung, “Arc Flash
operations to avoid possible exposure to electric arcs. An Burn Hazards at Various Levels of an Electrical System,”
example of this strategy is to purchase arc-resistant instead of IEEE Petroleum and Chemical Industry Conference
conventional medium voltage switchgear. If such alternative Record, September 1995, pp. 3 17-325.
strategies are not possible, then proceed to Step 2.

- 297 -
[7] ASTM F 1506-94, “Standard Performance Specification
for Textile Materials for Wearing Apparel for Use
By Electrical Workers Exposed to Momentary Electric
Arc and Related Thermal Hazards.”

[8] E. W. Kalkstein, R. L. Doughty, A. E. Paullin, J. Jackson

& J. Ryner, ‘The Safety Benefits of Arc Metalclad
Medium Voltage Switchgear,” IEEE Petroleum &
Chemical Industry Conference Record, September
1994, pp. 309-317.

[9] A. M. Stoll & M. A. Chianta, “Method and Rating System

for Evaluation of Thermal Protection,” Aerospace
Medcine., Vol. 40, No. 11, pp.1232-1238, Nov. 1969.

[lo] F. 0. Henriques and A.R. Moritz, “Studes of Thermal

Injury: I. The Conduction of Heat to and Through Skin
and the Temperatures Attained Therein. A Theoretical
and an Experimental Investigation,” Amer. Journal
Pathology No. 23, pp. 531-549, 1947

[ I l l Artz, Moncrief, and Pruitt, “Burns,” W. B. Saunders

Co.; Philadelphia, London and Toronto, 1979.

[12] Privette, Alan, Report Prepared for the ASTM F-18

Committee Task Force F18.10.07 on Electric Arc Test
Method Development, June 10, 1992.

[13] F’rivette, Alan, The Symposium on “Electrical Injury: A

Multidisciplinary Approach to Therapy, Prevention and
Preparation,” University of Chicago, June 12, 1993.

[14] Stoll, A. M. & M. A. Chianta, “Burn Production and

Prevention in Convective and Radiant Heat Transfer,”
Aerospace Med. 39, No. 10, pp. 1097-1100, 1968.

[15] Stoll, A. M. & M. A. Chianta, “Heat Transfer Through

Fabrics as Related to Thermal Injury,” Transactions New
York Academy of Sciences, No. 33, pp. 649-670, 1971.

[16] E. I. Evans, J. W. Brooks, F. H. Schmidt, R. 0.

Williams and W. T. Ham,” Flash Burn Studies on
Human Volunteers,” Surgery, No. 37, pp. 280-297,
1955.

[17] G. R. Ripple, K. G. Torrington and Y. Y. Phillips,”

Predictive Criteria For Burns From Brief Thermal
Exposures,” Journal of Occupational Medicine, Vol. 32,
No. 3, March 1990.

- 298 -