See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/281639860

Protection of fuel storage tanks against lightning - Experimental

developments and risk analysis

Conference Paper · June 2015

CITATION READS

1 562

4 authors, including:

Helio E. Sueta Luis E Caires

University of São Paulo University of São Paulo
25 PUBLICATIONS   16 CITATIONS    12 PUBLICATIONS   41 CITATIONS   

SEE PROFILE SEE PROFILE

Miltom Shigihara
University of São Paulo
16 PUBLICATIONS   22 CITATIONS   

SEE PROFILE

All content following this page was uploaded by Helio E. Sueta on 10 September 2015.

The user has requested enhancement of the downloaded file.

 2015 Asia-Pacific International Conference on Lightning (APL), Nagoya, Japan
PS9-1

Protection of fuel storage tanks against

lightning - Experimental developments and risk
analysis
Hélio E. Sueta, Luís E. Caires, Vinícius Teixeira, Miltom Shigihara and Geraldo F. Burani

lightning: São Luiz plant in Ourinhos, in the state of São

Abstract- This paper presents a study regarding the Paulo on Jan/6th and Rio Claro plant, in the state of Goiás on
protection of fuel tanks against direct lightning discharges. This Jan/16th.
study includes: laboratory experiments in order to verify the hot
spots inside the plates used in tanks; an analysis of the
temperature values obtained was compared with the auto This work presents a specific risk analysis for these types
ignition temperature of some types of fuel; development of a of tanks, considering the results of tests with direct-current
spreadsheet for risk analysis specific to this type of installation pulses (simulating continuing currents) performed directly on
and some recommendations for the protection of fuel storage metal plates and a system to measure the local temperature
tanks against lightning. inside hot face for analysis.

Keywords—Fuel tanks, hot spot, lightning protection, risk II. BASIC CONSIDERATIONS
management.
A. Types of protection for fuel tanks
I. INTRODUCTION Regarding the fuel tank protection against lightning we
must study the effects of lightning in three different

T HIS paper presents a summary of a study to fuel tanks to

protect against the effects of lightning damage including
development of a risk analysis worksheet and the results of
situations: a direct lightning strike to the tank roof or shell; a
strike near the tank area; and a lightning discharge on sensors
or any electrical wire connected to accessory equipment.
tests on steel sheet used in fuel storage tanks.
Lightning discharges to earth may be hazardous to When a lightning strike hits directly a storage tank, the
structures, individuals and also to services such as electricity following effects may occur:
supply, wired telephone, wired computer networks, gas and
water metallic pipes, industrial activities, processes involving • Lightning current will be conducted by metal parts
oil and other types of fuel, etc. of the tank (including parts of the protection system)
The dangers to a structure can result in damage to it and its without causing any damage;
contents, electronics systems failure and damage to life in or • Melting points may cause punctures in the metal
near the structure. sheet and result in fire or explosion;
The effects from damage and failure may extend to the area • A hot spot without perforation of the metal sheet can
near of the structure or affect the environment. ignite the fuel inside the tank;
Dangers to services can result in damage to itself and • Flashover in critical points of the tank, due to the
failures of electronic equipment’s associated. potential difference arising from the current flow or
If a lightning strikes fuel tanks, explosions may occur as induced voltage, initiate fire and/or explosion;
well as onset of fire. Then, the harm can be irreparable, • The plasma temperature of the lightning channel
beyond environmental problems such as the risk of life reaches very high temperatures and therefore turn it
nearby. into a major risk of ignition for explosive
As examples in Brazil on January 2013, two ethanol environments;
storage fuel tanks had exploded after being struck by
In cases when lightning strikes areas near the tank or hitting
sensors or its wires, it may not cause any flaws on equipment,
H.E. Sueta is with the Institute of Energy and Environment of University but it can start a fire or explosions mainly because of
of São Paulo, Brazil (sueta@iee.usp.br). flashovers at critical points of the tank. If the lightning current
L.E. Caires is with the Institute of Energy and Environment of University (in air) flows through inflammable areas, it may be an ignition
of São Paulo, Brazil (luis@iee.usp.br). source, causing fire or explosions that will inevitably reach
V. Teixeira is with the Institute of Energy and Environment of University
of São Paulo, Brazil (vinicius.mpt@gmail.com). nearby tanks.
M. Shigihara is with Institute of Energy and Environment of University of
São Paulo, Brazil (mshigi@iee.usp.br) For protection against indirect effects of lightning,
G.F. Burani is with the Institute of Energy and Environment of University
of São Paulo, Brazil (burani@iee.usp.br). especially against flashover at critical points, the surge

990
 2015 Asia-Pacific International Conference on Lightning (APL), Nagoya, Japan

protection measures (SPM) must be taken, mainly on A.3 External Non-Isolated Protection System
equipotential bonding, grounding system and surge protection
devices (SPD) properly and coordinately installed. In the Non-Isolated system (Fig. 3), the protection is
In order to avoid problems related to direct-strike lightning, provided by lightning rods electrically bonded to the tank
the main types of protection are described below. roof, in order to avoid the strike to hit directly to the tank, but
A.1 Self-protected and self-grounded tanks the current will still flow through the metallic body of the tank
Tanks having a metallic body structure (roof and shell), and to the ground. Existing and permanent metal parts of the tank
plates with minimum thickness to support lightning currents can eventually be used as part of the protection system, such
safely (without causing any punctures or hot spots) are as metal guardrails properly welded to the tank and with
considered self-protected and grounded. Also, the diameter of appropriate dimensions. The air-termination rod should be set
the tank bottom in contact with the soil must have sufficient on the tank roof in a way that it may have a large surface
area to ensure a good grounding, varying according to the soil contact with the plate, ensuring that the lightning current will
type (Fig. 1). This concept of self-grounded is allowed in not cause any damage, corrosion or hot spots in that interface,
some technical standards, including the latest version of when the current flows through it. The design of the rod
ABNT NBR 5419/2005 [1], but this concept will cease in a structure taking into account its size and position should be
future revision of this standard (ABNT NBR 5419-3) that will made using the Rolling Sphere Method.
based on international standard IEC 62305-3/2010 [2]. Then,
the earth-termination system will be required in this case.

Fig. 1. Self-protected and self-grounded tank

A.2 Self-protected tanks with earth-termination system

Tanks considered self-protected with earth-termination
system (Fig. 2) are those which have metal plate’s thickness Fig. 3. External Protection Non-isolated
equal or greater than the minimum safety values, electrically
bonded at an earth-termination system. A suitable
earth-termination system should be designed and installed A.4 External Isolated Protection System
according to standards, which describe techniques using, for
example, grounding rods, grounding ring, grid or mesh The external isolated system (Fig. 4), provides a most
protection, for a single tank or a storage park. likely point to the lightning attachment, but also directs the
total current to the ground through a new path, removing the
tanks parts from it. This type of protection comprises towers
sized according to the Rolling Sphere Method spaced at least
2 meters from the tank shell. These towers have to be
connected on the top by properly sized metallic wires (as
standardized) and, must be connected to the earth-termination
system at the bottom. This protection should be adopted when
the thicknesses of the tank does not prevent either perforation
or the formation of hazardous hot spots when a direct
lightning hit the tank. Furthermore, the lightning current
through the tanks parts may cause dangerous increase of
gradient temperature and/or flashover at points that can
initiate fire or explosion.

Fig. 2. Self-protected with earth-termination system tank

991
 2015 Asia-Pacific International Conference on Lightning (APL), Nagoya, Japan

Current pulses were applied to upper side of new metal

sheets with different thicknesses (Fig. 6).

Electrode

Fig. 4. External protection isolated

GAP
These types of tanks protection were considered in the
development of risk analysis worksheet.

II. EXPERIMENTAL DEVELOPMENTS

This work focused its experimental part on one of the
possible types of damage which occur when the lightning
directly strikes a storage tank of fuel and generates hot spots
in the stroke point of impact.
The main goal of these tests was to determine the critical
temperature points, such as the holes on metal plates were
subject to previous investigations, and tanks used for storage
of fuels have plate’s thickness enough to avoiding such
perforations. These temperature values may be compared
with those reference auto ignition values of several fuel types
Fig. 6. Device under test (Metallic plate)
typically stored in these tanks. In order to verify these thermal
effects it is important to evaluate the continuing current
On underside of the metal plate (Fig. 7), the hot spot
component. This type of current occurs between successive
temperatures were measured through two different methods:
strokes in a flash.
thermocouples and thermal imaging.
The continuing current was simulated from pulses of direct
For this work, the plates used were often used in the
current obtained from a rectifier bridge adjusted to obtain a
construction of tanks, but without any covering or painting. In
pulse of 600 A and variable duration up to 500 ms, thereby
future work, tests using painted and small air termination rods
obtaining pulses of 100, 200 and 300 C. Figure 5 shows an
will be performed.
example of oscillogram obtained in the tests.

Plate under test

Fig. 7. Bottom view of the plate (thermocouple attached with adhesive).

Fig.5. Pulse DC Oscillogram.

The test results are described below.

992
 2015 Asia-Pacific International Conference on Lightning (APL), Nagoya, Japan

A.1 Simulations from the model. Some preliminary applications were made to adjust the
Due to the impossibility of obtaining the real phenomenon measurement system. Regarding temperature, two different
(lightning) data in a controlled laboratory, it is necessary to systems were used: the first one with thermocouples and the
perform simulations in order to obtain results to compare with second with thermal imaging.
observations from the field. In this process it was found that the time for the occurrence
In this case it is imperative to taking into account the value of this event is too short, remaining below the time response
of the temperature rise in an internal point of the plate and of the set type K thermocouple / multimeter YEW TY530,
based on the assumptions previously deducted, it is possible whose time constant is about 300 ms. The stability condition
to establish the initial conditions. according to the response time of thermocouple would be
Therefore, we could draw up a spreadsheet to assist on the about 900 ms.
results evaluation. Thermal images (Fig. 9) showed higher values compared to
The test should be done with direct-current pulses to the system with thermocouples. Table I shows the results of
simulate the component of the lightning known as continuing tests on the plate of 2.75 mm thick with the maximum
current. temperature values obtained by thermography.
Due to the relative low voltage applied in the laboratory TABLE I
between the electrode tip (central rod) and the plate (negative TESTS ON THE PLATE WITH THICKNESS: 2,75 mm
electrode), a copper wire was installed in order to start the U [V] I [A] Q [C] P [W] t [ms] Tmáx [ºC]
electric arc (Fig. 8). Arc
voltage
98 716 111 69839 155 254
94 731 210 68767 288 465
92 742 335 68416 452 >670(*)
(*) Note: It exceeded the scale of the instrument.

igniter-wire The following tables II and III show the results for plates of
4.8 mm and 6.35 mm.
TABLE II
TESTS ON THE PLATE WITH THICKNESS: 4,8 mm
U [V] I [A] Q [C] P [W] t [ms] Tmáx [ºC]
Arc voltage
81 731 114 59254 155 156
85 764 218 65248 286 251
Fig. 8. Copper wire used to start the electric arc.
85 773 305 65903 394 424
In the first series of tests, the positive electrode was used as TABLE III
the central rod and the negative on the plate. TESTS ON THE PLATE WITH THICKNESS: 6,35 mm
Fixing the current value and varying the time of exposure U [V] I [A] Q [C] P [W] t [ms] Tmáx [ºC]
to arc, it is possible to estimate the influence of time on the Arc voltage
phenomenon. We define the current I as 600 A (300 C in 85 729 100 61719 137 111
0.5s). 86 759 235 65040 310 201
The charges Q are defined as 100, 200 and 300 C (values 77 778 430 59779 553 316
tabulated in IEC). In these applications the value of maximum temperature
The device used for thermal imaging was set to shoot the was consistent with the expected which is proportional to
applications and thus to obtain the maximum instantaneous applied charge.
observed. The value of the considered arc voltage was particularly
important and varied depending on thickness of the plate. The
importance of this parameter is meaningful, since it
determines the energy that was transferred or dissipated in the
process. Therefore it is deducted that not all thermal energy of
the arc was used in the concerned phenomenon to determine
the risk of excessive temperature rise. From the empirical data
it was possible to obtain an approximate expression for arc
voltage which was used as input data in a spreadsheet
calculation.
With these values it was possible to improve the model in
order to create a spreadsheet that enables more data to review.
In this case, extrapolation was critical because it is necessary
to determine a critical charge profile according to the auto
ignition temperature of the materials involved. A summary of
Fig.9. Thermal image of application by the backside of the plate

993
 2015 Asia-Pacific International Conference on Lightning (APL), Nagoya, Japan

measured values compared to calculated values is showed in and detailed application of electrical and thermodynamics
Table IV below. theory.
TABLE IV In order to apply the data analysis to practical experience
COMPARISON BETWEEN MEASURED AND CALCULATED VALUES showed that the best results could be performed with different
Plate Current Conduction Charge Maximum Calculated Difference materials and methods. Sheet material samples with greater
thickness [A] Time [ms] [C] Temp.[°C] Maximum [°C] variation in thickness could be tested, otherwise only a few
[mm] Temp. defined thicknesses were tested according the ones used in the
[°C] tanks. For example, in order to obtain a temperature gradient
2,75 716,19 155 111 254 257 -3 it would be ideally to test plates increasing their thickness of 1
2,75 730,91 287,5 210 465 493 -28 in 1 mm. Regarding the methods it should be made a larger
2,75 741,57 452,1 335 >670 782 ???? number of samples per plate to obtain a statistical evaluation
4,8 731,53 155,2 114 156 155 1 of dispersion measures.
4,8 763,64 285,8 218 251 276 -25
It became clear that it is required to obtain functional
relationships between plate thickness and critical
4,8 773,43 393,8 305 424 373 51
temperature, which was in the last instance performed
6,35 728,81 136,6 100 111 114 -3
through a simple mathematical model that was applied in a
6,35 758,83 309,8 235 201 243 -42
spreadsheet.
6,35 777,8 553,2 430 316 424 -108
For the purpose to make it practical for the application, the
In the Table IV, concerning the column “difference”, in results of this experiment need additional statistical
particular the negative values indicate that the spreadsheet has evaluation that correlates the critical charge value to the
calculated higher values than the observed ones in
probability of it be exceeded. This limit can be found
experiments, which is positive to safety. In most evaluations
researching on atmospheric discharges parameters [3].
this case prevails.
The final goal was to obtain the value of maximum charge Based on the results obtained in the tests, the developed
that a plate with certain thickness can absorb so that its spreadsheets and simplified simulations, it was possible to
temperature reaches the critical value depending on stored customize the risk analysis worksheet for the specific case of
fuel. fuel storage tanks.
Table V below shows these values for plate thickness from
1 mm to 10 mm for three types of fuels: gasoline, diesel and III. DEVELOPMENT OF A RISK ANALYSIS PROGRAM APPLIED
ethanol. TO A FUEL TANKS
TABLE V A specific risk analysis was carried out for fuel tanks and a
MAXIMUM CHARGE DEPENDING ON THE THICKNESS OF PLATE spreadsheet based on IEC 62305-2 / 2010 [4]
AND FUEL recommendations was developed to adapt the analysis to the
PRODUCT GASOLINE DIESEL ETHANOL fuel tanks with several modifications.
Auto ignition
The spreadsheet developed was so-called "Tupã (version
temperature [°C]
246 210 363 for tanks)". To facilitate the use of the spreadsheet, many
specific parameters for tanks were fixed, however they are not
THICKNESS [mm] Charge [C] Charge [C] Charge [C]
shown in the spreadsheet, i.e., their values are used, but only
1 6 5 9 the parameters that can vary from the tank to the tank are in it.
2 50 40 80 With this spreadsheet, we can estimate how each tank is in
3 120 100 180 a particular plant to the risks and which are the main
4 170 140 250 protective measures to be taken to ensure that these risks are
within tolerable standard values as described. In addition, we
5 200 170 300
can compare the risks of different tanks and sort them in order
6 230 200 350 of priority for maintenance and refurbishment.
7 260 220 380 In this study we verified the relationship of the parameters
8 280 240 430 indicated in the risk management standard (IEC 62305-2 /
9 310 260 470 2010) [4] and we detail some of the changes made in
10 330 280 510 customization. We also present a customized spreadsheet that
Due to the work of evaluating the results based on their got quite simplified since only specific to parameters for
application, it was observed that the final goal for laboratory tanks with the possibility of variations were presented to
facilitate understanding and handling of the spreadsheet.
tests is to provide a consistent model to describe a critical
The standard IEC62305-2 [4] provides a method for risk
charge profile of lightning discharge, depending on the plate
assessment that includes areas with risk of explosion. Being a
thickness used in fuel storage tank. Such a model need to be generic and comprehensive method and considering that in
better draw up, however to immediate purpose we have to use these areas the risk of economic loss, personal and
a lot of simplification. environmental value is higher, it would be wise to establish
Moreover, even this simplified model pointed variables some method to guide the risk analysis.
and their relations underlining the importance of further Such method should consider the particularities of these
development aiming sedimentary concepts through stricter risk areas as well as the calculation premises recommended

994
 2015 Asia-Pacific International Conference on Lightning (APL), Nagoya, Japan

by the standard, aligning these two points of view. For

example, in the standard the reduction risk factor rf of an
explosion hazardous area (zone 0, 20) is a thousand times
larger than an area with less risk of explosion (Zone 2, 22).
Also the usually area of fuel storage tanks is very large, which
makes the collection area have a very high value. These
associated factors can lead to a risk value which no protective
measure would be enough to ensure the safety of the site.
Because the standard is a generic method for risk assessment,
the variable analysis may lead to outrageous values for not
adequately address the installation details. For example, the
thickness of the side parts and tank roof plating sets whether it Fig. 10b. Risk analysis for tanks - spreadsheet screen – inferior part.
can be regarded as self-protected, but it is not considered in
the standard evaluation. This input data is specific to this .
worksheet and the calculation was based on studies and
results of laboratory tests and analysis of the fuels used. IV. KEY RECOMMENDATIONS FOR THE TANKS PROTECTION
The risk component considering the presence of an
explosive atmosphere shall not increase the risk since the In this section we will present the main recommendations
project installation contains the appropriate protection for the protection of fuel storage tanks as follows: first we will
measures. The risk assessment is related to this premise, that present general recommendations valid for both new and
is, if there is a failure in the protection method it will increase existing tanks; then we will present specific recommendations
the risk. for tanks in operation, and finally the specific
Based on this, the study of the standard can be used by recommendations for new tanks yet to be applied in the
adding the effect of an explosive gas, or rather the probability design phase and deployment.
that this occurs according to the exposure time.
The other variables determine whether the protection These recommendations were based on studies of various
systems operate properly and therefore can be used in the standards, technical papers and results of tests performed.
evaluation. Thus, the aim of this work is to adapt the standard
method for applying in fuel storage tanks. It is very important to mention that the phenomenon of
Among these adaptations we have include the automatic lightning is quite complex and there is no 100% safe
determination of a collection area for tanks by introducing the protection. In this way, even taking 100% of the
height and diameter of the tanks; or by including external recommendations described below it cannot be assured with
towers, in case of isolated external protective; and also absolute certainty that a discharge when reaching a tank or its
including the type of stored fuel and the plate thickness of the surroundings cannot bring any kind of damage.
tank roof.
These new data influencing the risks were taking into A.1 General Recommendations
account the generation of hazardous hot spots, according to
the characteristics of the tank and stored fuel. • The standard for lightning protection recommended for
The main input data are: tank height, body guard height tanks installed in Brazil is, at present, the IEC 62305/2010 [2]
installed in the tank cover, tank diameter, type of LPS used, and, from 2015 (probable year of entry into force of the new
towers characteristics to be used in the LPS, fire or explosion Brazilian standard), the new version of ABNT NBR 5419,
risks, attributes of connected lines, SPD used, the tank both published in 4 parts. This way it is recommended that all
location factor, density of lightning to earth, soil or floor type, fuel storage tanks meet, at least these standards and may also
fuel type and thickness of the roof. meet some additional recommendations outlined in
Below is an example of the spreadsheet screen into two international standards such as the specific rules for these
parts (Fig. 10a and b). structures as American ones.

• We recommend that all sensors and / or other devices in the

tank (or yet to be installed in the same), that have electrical
power cords and/or signal cables should be protected against
surges through Surge Protection Measures (SPM) mainly
through the SPD (Surge Protection Devices) proper and
coordinated, and, when necessary, specific for risk of
explosion areas.

• We recommend that all tanks should be checked in order to

verify the electrical continuity between the roofs (whether
fixed or mobile), the plating and the grounding system,
Fig. 10a. Risk analysis for tanks - spreadsheet screen – superior part avoiding insulating parts susceptible to flashover that can

995
 2015 Asia-Pacific International Conference on Lightning (APL), Nagoya, Japan

cause sparkover. For new tanks, set this continuity in design by the stored product in the tank and/or environmental
and realize the implementation. requirements. The alternatives to provide an adequate
connection between floating roof and tank wall regarding the
• Earth-termination system consists of a grounding grid and conduction of impulse currents associated with lightning will
the number of interconnections between the tanks and the only be permitted if this is successfully demonstrated in tests
mesh according to the standards is defined as the dimensions and the procedures are used to ensure the reliability of the
of it. connection.

• Section D.5.5.2 of ABNT NBR 5419-3 of future Brazilian • It is necessary to make a risk analysis for lightning
standard describes important points about the protection of protection in design phase, after tank installation, during the
fuel storage tanks. In this section, the future standard most critical maintenance and whenever there is any change
indicates that for some structures used to storage liquids in the original tank design project.
which may produce flammable vapor or used to storage gas
are considered self-protected and do not require additional A.2 Specific recommendations for existing and running
protection as long as they are metal contained completely tanks
within, continuous container with a wall thickness greater
than 5 mm steel or 7 mm aluminum, without the spaces that • Never carry out supply operations or emptying of fuel
permit arcing. The standard also indicates that tanks in direct during periods of storms.
contact with soil and pipe lines do not require air-termination
systems. Electrical components and instrumentation used in • Never perform maintenance and / or measuring in the tank
such equipment must be certified for this type of application. or close operations during storm periods.
Measures for lightning protection should be taken as the type
of construction. • Check the thicknesses of tanks and compare the values
found with the design values and normalized ones.
Individual metallic tanks or containers must be bonded to the
grounding electrode according to their horizontal dimensions • Check the status of interconnections between tanks and
(diameter or length) as described in Section 5 of the standard: earth-termination system, compare the amount of
interconnections with standard values and verify electrical
a) up to 20 m: at least two interconnections, arranged continuity between all conductors that make up this system.
equidistantly on perimeter;
• For tank protection purpose against lightning, it is
b) over 20 m: two interconnections plus an additional considered that all possible outputs for pressure of explosive
interconnection at every 10 m perimeter, arranged gases, are equipped with special valves so that any external
equidistantly. spark may never penetrate to interior of the tank.

For tanks grouped in yards, as refineries and storage, • It is necessary to include in the tanks Maintenance Plan,
grounding one point in each tank is sufficient, regardless of specific findings based on the LPS including checking the
greater horizontal dimension. When arranged in courtyards, continuity of the earth-termination grid, coverage of
the tanks must be interconnected. As shown in Tables 7 and 8 air-termination system, thickness of the plates (especially
of future NBR 5419-3, in addition the connection, the pipes roof tanks), electrical continuity between air-termination
that are electrically connected may also be considered system and earth-termination system, functionality of Surge
interconnection according to section 5.3.5 of the standard. In Protection Devices installed, functionality of pressure relief
the case of floating roof tanks, roof should be connected to the systems, among others.
main housing of the tank effectively. The design of junctions
and seals, and their relative locations need to be carefully A.3 Specific recommendations for new tanks still to be
considered so that any possible risk of ignition of explosive applied in the design phase and deployment
mixture by a spark is reduced to the lowest possible level.
When a mobile ladder is installed, equipotential bonding • Checking if the plate’s thicknesses of the design are
cables, flexible (#35 mm²), must be connected at the hinges of specified according to the standard thicknesses reported
the stairs, between the ladder and the top of the tank and (greater than 5 mm for steel).
between the ladder and the floating roof. When a mobile
ladder is not installed in the floating roof tank, one or more • For the design of earth-termination system, before
equipotential bonding flexible conductors (#35 mm²) must be performing the soil stratification onsite of tanks deployment
connected to the main structure of the tank and the floating to check the resistivity and calculate required mesh for
roof considering the dimensions of the tank. Equipotential earth-termination system. All interconnections to the tank and
bonding conductors must follow the ceiling or be installed in details must be indicated in the design.
such a way they do not form loops resulting from its
movement. In floating roof tanks, there must be multiple • Provide all sensors and measuring systems with the
connections at intervals of 1.5 m between the periphery of the appropriate Surge Protection Measures (SPM) suitable for
floating roof and the tank wall. The material selection is given explosive areas.

996
 2015 Asia-Pacific International Conference on Lightning (APL), Nagoya, Japan

V. CONCLUSIONS estimation does not exclude the need for careful analysis of
This study has addressed various aspects related to the design and execution based on standards. The spreadsheet
protection of flammable liquid storage tanks against the is not intended to define the protection system to be used, but
damaging effects of lightning. Because of this, the findings guides you through the data reasonably approximate for a risk
presented here refer to the fuel tanks of protection in general, assessment showing systems with better or worse protection.
based not only on the data presented here as well as in studies
of standards and technical work on the subject. Finally, although the phenomenon of lightning is quite
complex and comprehensive, and a 100% reliable protection
From analysis and studies carried out, we found that the is not possible, this study addresses the various types of
main focus of the study should be the direct strokes to the tank injuries and accidents related to lightning in the fuel tanks,
and the effect of hot spot on the inside surface of the plate. In indicating some key measures of protection. It is always
the study it was also observed that the thicknesses normally important to emphasize that the recommendations of
used in tanks exceed the thicknesses that could be perforated standards and specifications should always be followed to
by lightning. Other possible types of damage such as, for protect people and facilities in Power Plants and Terminals,
example, flashover insulating parts or sensors in tank and the and also to protect of the environment.
damage caused from the passages of lightning by explosive
vapors contaminated areas were also addressed and ACKNOWLEDGMENT
considered in risk analysis and protection measures listed are
traditional and standardized, such as surge protection The authors would like to thank the High Current
measures and recommendation use of special valves that Laboratory of IEE-USP staff, especially Ivan Bueno Raposo
prevent the penetration of any flames into the tank. for the development of the testing device and carrying out the
As for the case of the impact of hot spots inside the plates, tests on plates. Also thanks to Raizen Combustíveis S.A. to
having regard to the little literature available on the subject fund part of this study and provide important technical data,
and the diversity of fuel types, it is necessary further study. especially the engineers Paulo Sérgio Lopes Soares de
Therefore, tests performed in laboratory where been directed Almeida and José Giliard Ansiero.
to this subject and, from these results, studies and specific
worksheets were developed in order to verify the impact of
REFERENCES
critical temperatures in relation to total transferred charges
from long stroke arcs. Assays were performed for various [1] ABNT- ASSOCIAÇÃO BRASILEIRA DE NORMAS TÉCNICAS –
ABNT NBR 5419, Proteção de estruturas contra descargas
sheet metal thicknesses used in tanks and calculations of atmosféricas, 2005, in portuguese.
critical charges for each fuel type studied. [2] IEC – INTERNATIONAL ELECTROTECHNICAL COMMISSION –
IEC 62305 -3, Second edition, Protection against lightning – Part 3:
This particular study allowed us to create the spreadsheet Physical damage to structures and life hazard, 2010.
[3] LIGHTNING AND INSULATOR SUBCOMMITEE OF THE T&D
"Risk Analysis" to fuel tanks, one of this study product that COMMITTEE - “Parameters of Lightning Strokes”. IEEE Transactions
has become more prominent throughout the project. This on Power Delivery, vol. 20, pp 346-358, Jan. 2005.
worksheet assists in organizing priorities for protection of [4] IEC – INTERNATIONAL ELECTROTECHNICAL COMMISSION –
existing tanks and design considering the damaging effects of IEC 62305 -2, Second edition, Protection against lightning – Part 2:
Risk Management, 2010.
lightning. On this point it should be noted that the risk

997
View publication stats