Te 1347 Web PDF

IAEA-TECDOC-1347
Consideration of external
events in the design of
nuclear facilities other than
nuclear power plants, with
emphasis on earthquakes
March 2003
The originating Section of this publication in the IAEA was:
Engineering Safety Section
International Atomic Energy Agency
Wagramer Strasse 5
P.O. Box 100
A-1400 Vienna, Austria
CONSIDERATION OF EXTERNAL EVENTS IN THE DESIGN OF NUCLEAR FACILITIES

OTHER THAN NUCLEAR POWER PLANTS, WITH EMPHASIS ON EARTHQUAKES
IAEA, VIENNA, 2003
IAEA-TECDOC-1347
ISBN 9201028032
ISSN 10114289
IAEA, 2003
Printed by the IAEA in Austria
March 2003
FOREWORD
The design of nuclear facilities other than nuclear power plants in relation to external
events is not a well harmonized practice around the world. Traditionally, the design of these
facilities has either been left to the provisions collected in national building codes and other
industrial codes not specifically intended for nuclear facilities, or it has been the subject of
complex analyses of the type usually performed for nuclear power plants.
The IAEA has recently started a programme of development of safety standards for
such facilities. The need to define the appropriate safety requirements for nuclear installations
prompted a generic review of siting and design approaches for these facilities in relation to
external events. Therefore the assessment methods for siting and design were reviewed by the
engineering community to provide the overall design of such facilities with the necessary
reliability level.
This report aims to provide guidelines for the assessment of the safety of nuclear
facilities other than nuclear power plants in relation to external events through the application
of simplified methods and procedures for their siting and design. The approach adopted is
both simplified and conservative compared with that used for power reactors. It seeks to
provide a rational balance for a suitable combination of sustainable effort in site investigations
and refinement in design procedures, compatible with the assigned safety objectives.
This publication is related to IAEA-TECDOC-348 Earthquake Resistant Design of

Nuclear Facilities with Limited Radioactive Inventory (1985) which focused on the seismic
design of nuclear facilities with limited radioactive inventory. After some 17 years, parts of
IAEA-TECDOC-348 needed modification, as new operational data have become available
from many facilities. In addition, sophisticated design methodologies are now more easily
obtainable, and experts felt that the trade-off between sustainable investment in the facilities
and design conservatism had to be redefined.
A large number of consultants from various States were involved in updating IAEA-
TECDOC-348 and in reviewing draft sections of the present publication their efforts are
greatly appreciated.
The contributions of M. Lebelle (France) and T. Fukuda (Japan) are acknowledged.

The IAEA officers responsible for this publication were P. Contri and H. Tomura of the
Division of Nuclear Installation Safety.
EDITORIAL NOTE
The use of particular designations of countries or territories does not imply any judgement by the
publisher, the IAEA, as to the legal status of such countries or territories, of their authorities and
institutions or of the delimitation of their boundaries.
The mention of names of specific companies or products (whether or not indicated as registered) does
not imply any intention to infringe proprietary rights, nor should it be construed as an endorsement
or recommendation on the part of the IAEA.
CONTENTS
1. INTRODUCTION.................................................................................................................. 1
1.1. Background ................................................................................................................ 1
1.2. Objective.................................................................................................................... 4
1.3. Scope.......................................................................................................................... 4
1.4. Structure..................................................................................................................... 7
2. SAFETY OBJECTIVES AND CLASSIFICATION.............................................................. 8
2.1. Introduction and examples of NFOPs........................................................................ 8
2.2. Safety objectives ...................................................................................................... 10
2.3. Defence in depth ...................................................................................................... 11
2.4. Safety classification of the facilities ........................................................................ 12
2.5. Classification of items in relation to external events ............................................... 16
3. GENERAL APPROACH TO THE DESIGN OF NFOP ..................................................... 19
3.1. Generalities .............................................................................................................. 19
3.2. Safety margin control in the design process ............................................................ 19
3.3. Design procedures and safety assessment................................................................ 21
3.4. Ageing effects on design and re-evaluation ............................................................. 23
3.5. The special case of research reactors ....................................................................... 23
4. DESIGN BASIS FOR EXTERNAL EVENTS .................................................................... 27
4.1. Introduction.............................................................................................................. 27
4.2. Design basis earthquake........................................................................................... 29
4.2.1. Site specific design response spectra based on instrumental data ............... 29
4.2.2. Design response spectra based on historical seismicity .............................. 29
4.2.3. Design response spectral shapes.................................................................. 30
4.3. Design basis for aircraft crash.................................................................................. 32
4.3.1. General ........................................................................................................ 32
4.3.2. Basic data for screening............................................................................... 33
4.4. Design basis for wind............................................................................................... 34
4.4.1. General ........................................................................................................ 34
4.4.2. Basic data .................................................................................................... 34
4.5. Design basis for snow .............................................................................................. 35
4.5.1. General ........................................................................................................ 35
4.5.2. Basic data .................................................................................................... 35
4.6. Design basis for external temperature...................................................................... 35
4.6.1. General ........................................................................................................ 35
4.6.2. Basic data .................................................................................................... 36
4.7. Design basis for flood .............................................................................................. 37
4.7.1. General ........................................................................................................ 37
4.7.2. Basic data .................................................................................................... 37
4.8. Design basis for external explosion ......................................................................... 37
4.8.1. General ........................................................................................................ 37
4.8.2. Basic data .................................................................................................... 38
5. GEOLOGICAL AND GEOTECHNICAL INVESTIGATIONS ......................................... 38
5.1. Generalities .............................................................................................................. 38
5.2. Liquefaction potential .............................................................................................. 40
6. BUILDING DESIGN ........................................................................................................... 40
6.1. Seismic design ......................................................................................................... 40
6.1.1. Layout considerations.................................................................................. 40
6.1.2. Design classes for structures ....................................................................... 41
6.1.3. Geometric modelling................................................................................... 41
6.1.4. Soil structure interaction ............................................................................. 43
6.1.5. Non-structural loads .................................................................................... 44
6.1.6. Spectral analysis .......................................................................................... 45
6.1.7. Simplified dynamic approach...................................................................... 47
6.1.8. Earthquake induced element forces............................................................. 49
6.1.9. Load combination and acceptance criteria .................................................. 50
6.2. Aircraft crash ........................................................................................................... 51
6.2.1. Protection strategies .................................................................................... 51
6.2.2. Local design................................................................................................. 51
6.2.3. Global design............................................................................................... 52
6.3. Wind......................................................................................................................... 53
6.3.1. Load combination and acceptance criteria for extreme wind...................... 53
6.3.2. Load combination and acceptance criteria for frequent wind ..................... 53
6.4. Snow ........................................................................................................................ 54
6.5. External temperature................................................................................................ 55
6.5.1. Layout consideration ................................................................................... 55
6.6. External flood .......................................................................................................... 56
6.7. External explosion ................................................................................................... 56
7. EQUIPMENT AND PIPING DESIGN ................................................................................ 57
7.1. Introduction.............................................................................................................. 57
7.2. Seismic design ......................................................................................................... 58
7.2.1. General methodology for seismic force evaluation ..................................... 58
7.2.2. Simplified analytical evaluation of floor response spectra.......................... 59
7.2.3. Seismic design of containers ....................................................................... 60
7.2.4. Seismic design of pumps and valves ........................................................... 60
7.2.5. Seismic design of equipment and piping..................................................... 61
7.2.6. Other considerations.................................................................................... 61
7.3. Equipment design for aircraft crash loads................................................................ 62
7.4. Anchorage design..................................................................................................... 63
8. SLOSHING EFFECTS......................................................................................................... 64
9. OTHER CONSIDERATIONS ............................................................................................. 65

9.1. Seismic scram system .............................................................................................. 65
9.2. Monitoring systems at the site ................................................................................. 66
9.3. Quality assurance ..................................................................................................... 66
9.4. Evaluation of radioactive exposure to the public during and
after external events ................................................................................................. 66
REFERENCES......................................................................................................................... 67
APPENDIX I: SIMPLIFIED PROCEDURES FOR ASSESSMENT OF

LIQUEFACTION POTENTIAL ...................................................................... 71
APPENDIX II: SOIL STRUCTURE INTERACTIONS.......................................................... 76
APPENDIX III: EARTHQUAKE RESISTANT DESIGN PROVISIONS.............................. 82
APPENDIX IV: COMPUTATION OF PULLING FORCE ON ANCHOR DEVICE ............ 84
APPENDIX V: SIMPLE SEISMIC DESIGN PRACTICE FOR PIPING

SYSTEMS THE CONSTANT PITCH DESIGN METHODOLOGY...... 92
APPENDIX VI: DETAILS OF CONNECTING CABLE AND

PIPING BETWEEN COMPONENTS ........................................................ 102
APPENDIX VII: SEISMIC SCRAM SYSTEMS .................................................................. 105
CONTRIBUTORS TO DRAFTING AND REVIEW............................................................ 107

1. INTRODUCTION
1.1. BACKGROUND
The safety of a nuclear installation requires in general that it be sited, designed,
constructed and operated to protect individuals, society and the environment against an
uncontrolled release of radioactive material. External events (i.e. relevant to natural and
human induced hazard) play a major role in challenging the plant defence and therefore
appropriate design provisions need to be taken to guarantee an adequate safety margin in case
of such events.
In recent years the development of design criteria, design methodologies and

assessment approaches for the implementation of the general safety principles received major
emphasis for nuclear power plants, while other nuclear installations received less attention
even if their radioactive inventory was comparable with NPPs, as is the case for some research
reactors. For fuel reprocessing plants, the associated risk for radiological and chemical
contamination may be rather high, and for many research reactors associated with universities
and research centres, their location is often very close to very densely populated areas.
This report aims at collecting the most relevant experience in Member States for the
design of nuclear installations other than NPPs as set forth in Section 1.3, providing a
coherent framework where safety aspects, site investigations, design procedures and
assessment methodologies are presented in an engineering, ready to use context.
Many IAEA publications address siting and design of nuclear installations other than
NPPs in relation to external events, namely:
Design of spent fuel repositories are discussed in IAEA Safety Series G-3.1 and G-4.1
[1, 2] and in technical publications such as Refs [3, 4].
Design, Safety Assessment and Operation for Research Reactors are discussed in
Safety Series No. 35 [58] and in a number of TECDOCs [912].
Safety requirements and design procedures for fuel cycle facilities are discussed in
[4, 1315].
However, these publications do not specify the methodology to be used in the design
process and therefore, for the objectives of this report, reference is made to the more rigorous
approach applied for the power plants, particularly to the following requirements and Safety
Guides:
Requirements for NPP site evaluation [16].

Evaluation of seismic hazard is presented in Safety Guides [17].
Evaluation of human induced hazard is presented in Safety Guide [18].
Evaluation of flood induced hazard is presented in Safety Guide [19].
Evaluation of hazard induced by extreme events is presented in Safety Guide [20].
Requirements for NPP design [21].
Seismic design is presented in [22].
Design provisions in relation to external events are presented in [23].
Design provisions for foundations are presented in [24].
Safety assessment is presented in [25].
1
The specific design of the nuclear facilities other than NPPs was first addressed by
the IAEA in 1985 with the publication of IAEA-TECDOC-348, Earthquake Resistant Design
of Nuclear Facilities with Limited Radioactive Inventory [12], mainly oriented to seismic
siting and design. The report was widely applied in many States and often it was also
considered as contractual document for some research reactor design.
That publication presented simplified siting and design methods aiming at

minimizing sophisticated calculations and emphasized the importance of construction and
detailing principles. It proposed a simplified design approach alternative to the complicated
and sophisticated methodologies which are associated with and borrowed from nuclear power
plant analysis and design.
Such broad use of [12] in many applications in the world recently suggested a review
of the original report, mainly due to the following, additional, technical reasons:
Evidence from available operating experience: through databases like INES [26], an
analysis was carried out at IAEA on the 43 events recorded since 1991 in research
reactors, reprocessing plants, fuel fabrication plants, laboratories and accelerators.
Among them 10% showed high off-site radioactive contamination, with a sharing
among the different causes that is shown in Figure 1.
External
events
Unknown (3%)
(2%) 1% Bad
16% operation
(8%)
44%
Mech
failure
(54%)
40%
FIG. 1. Causes of INES events in NFOPs. (number in brackets are referred to NPPs, for
comparison).
The analysis of available data shows the low incidence (in terms of number of
events) of external events (3%), in the causes of nuclear events, compared with
mechanical failures (40%) and bad operation (44%). However, the analysis of the
consequences of such events clearly shows how external events (earthquakes, flood,
extreme temperatures, lightning, etc.) are often associated to the most challenging
degradation scenarios of the defence in depth barriers.
It needs to be recalled that the sample is very small and covered a short interval.
Furthermore, the nuclear facilities other than power plants (NFOPs) are often in the
2
most secure areas of the States for strategic reasons and some events probably have
not been reported because of security.
The emphasis on mechanical failures highlighted in the previous statistic shows also
the importance for stringent criteria in component qualification.Feedback from
applications in Member States: in most cases a combination of national codes and
standards for conventional or industrial facilities is applied as a design basis, often
very different for civil structures, mechanical components and equipment design,
with the consequent impossibility to evaluate the safety margin of the facility.
Moreover, little emphasis is usually given to the safety classification and accident
analysis (safety analysis reports are in fact available for very few facilities) with the
consequence of a generic difficulty in the safety assessment of the facilities.
The availability of new and cheap design methodologies and computer codes.
Modern design methodologies allow cheap and reliable design processes with limited
investment of time and money and therefore require new definition of the optimal
balance between design conservatism and capital investment mentioned above.
The growing emphasis on requalification of existing facilities. In the case of research
reactors, among 265 units in operation (2000), 50% are older than 40 years, with
obvious implications in terms of ageing evaluation and requalification.
A first review of the publication [12] was then launched (1998) with the following
targets:
To update the structural typologies to be addressed, their particular design issues and
relevant design limits.
To update the proposed design methodologies in order to be consistent with current
costbenefit ratios and required accuracy level.
To better specify the field of application of each procedure.
To add a graduation in siting and design procedures, trying to provide an average
engineering measurement of the introduced conservatism.
To provide a minimum validation of the proposed assumptions.
Moreover, recently (1999 and 2000) an increased concern for such facilities was
expressed in two IAEA Consultant meetings: the previous targets were confirmed and it was
suggested to extend the scope of the publication to all external events (i.e. flood, aircraft
crash, explosions, etc.) and to all facilities with similar structural design principles.
The following key ideas have driven the preparation of this report:
(1) The safety objectives defined for nuclear power plants need to also be valid for other
nuclear installations and therefore reference is made to the basic IAEA publications
for their presentation.
(2) A so-called graded approach in the design of nuclear installations other than NPPs
(NFOPs) is normal practice in Member States and therefore needs to be considered.
Different values for the probability of exceedance of design basis events might be
used in nuclear facilities according to their radioactive inventory (in general different
than NPPs), to the potential for dispersion in the environment and to other
characteristics of the radiological risk associated to external events.
The investment required for the construction and operation of a NPP is usually some
order of magnitude higher than for other nuclear installations. Therefore in these latter
3
facilities the extension of the site investigation campaigns, the use of expensive design
procedures and the implementation of expensive QA systems in design, construction and
maintenance needs to be reconsidered to provide a sustainable approach able to guarantee a
safe operation.
A common approach for NFOP design is the acceptance of a higher conservatism in

all the mentioned phases, able for example to exclude some sites or to compensate, in the
design phase, for the discrepancies between simplified numerical models and the real
behaviour of the structures in case of external events.
Such conservatism has to be carefully evaluated in order to guarantee that the final
design incorporates the required safety margin.
This report is the final result of such efforts and it confirms the validity of the
structure of the old publication [12], trying to update and extend its technical content, coming
to a completely revised publication on the subject.
1.2. OBJECTIVE
The main objective of this report is to provide information on a rational basis for a
safe design of nuclear facilities other than NPPs in relation to external events. Simplified
methods and procedures are presented here to provide enough conservatism in siting and
design processes in order to avoid the complicated and expensive procedures typical of the
NPPs. Such an approach is in line with the more limited resources typically associated with
NFOPs construction and operation.
The report aims also at providing a consistent basis for the assessment of the safety
objectives assigned to the NFOPs, for the selection of appropriate design standards and codes
consistent with the required safety margin and for the preparation of sound purchasing
specifications for equipment and components.
However, the report is not intended to be an alternative to design codes: its emphasis
is on safety issues related to the design and therefore it merges information for a safe design
with suggestions for quick and easy analytical methodologies.
The overall safety evaluation of the NFOPs needs to refer to the IAEA safety
standards.
Although the publication was not originally developed with reference to the re-
evaluation of existing facilities, most of its content can now be usefully applied to such
assessment.
1.3. SCOPE
This publication is intended for use in siting and design of facilities other than NPPs
in relation to external events. Qualification of components and equipment is not explicitly
addressed.
In particular, the following nuclear facilities have been identified, as potential targets
for this report:
4
(1) Surface facilities of uranium mines, mill factories and uranium tail repositories
(2) Fuel conversion plants
(3) Enrichment plants
(4) Fuel fabrication plants (including MOX fuel facilities)
(5) Research reactors
(6) Radioisotope laboratories
(7) Independent spent fuel storage facilities
(8) Fuel reprocessing plants
(9) Near surface repositories for low and intermediate level waste
(10) Deep waste disposal for permanent storage
(11) Heavy water distillation plants
(12) Accelerators.
However some of them have been excluded from the present report, according to the
following set of criteria:
Peculiar structural typologies or layouts, not comparable with the rest of the facilities:
earth structures, disposal mounds (cask, barrels and earth structures, covered with soil
materials, but without a bearing structure, like No. 9), underground structures and
mines (like No. 1), etc.
Structures with atypical layout and configuration, impossible to be categorized and
rather unique in the world, like Nos. 11 and 12
Facilities with peculiar safety objectives (very long term disposal, strong geological
implications, decommissioning framework, etc.), like No. 10
[Excluded by the Statute]
Therefore, only facilities of the type 2, 3, 4, 5, 6, 7, 8, meeting the above criteria are
considered in the scope of the present report.
In any of the selected facilities, some special equipment is featured which cannot be
designed according to the general principles stated in this report because of its peculiar
functions or atypical configuration. The reactor core in research reactors, the centrifuges in the
fuel enrichment plants are two typical examples.
This report defines the principles for classification and qualification of these critical
components, but does not provide details for their design, installation and operation (e.g.
qualification of welding, internal tolerances, etc.). Appropriate, specific design codes need to
be used in these cases.
Moreover, due to the general objectives of this report, the present publication is
focused on the structural design, which is usually the most expensive segment of the whole
design process.
Other critical tasks of the design, like equipment capacity evaluation (with reference
to all the acceptance criteria defined in Chapter 2, from stability to operability and
interaction), geotechnical assessment, QA, monitoring, etc. are not completely addressed in
this report. Either such tasks are not usually associated with high efforts or they cannot be
simplified, even with conservative approaches, and therefore are not relevant for the
objectives of this report. For guidelines and recommendations concerning the tasks not
5
explicitly mentioned here, general conventional standards or nuclear safety standards are
applicable, according to the risk classification of the facility of interest.
Concerning the applicability of this report to facilities with high radioactive

inventory, it may also be applied to the design of critical and subcritical assemblies
(hereinafter included under the term research reactor) to the extent that is appropriate for these
facilities. Therefore, research reactors with power above several tens of megawatts, fast
neutron spectrum research reactors, and small prototype power reactors may require additional
measures and the use of techniques for power reactors may be more appropriate.
Further guidance on the appropriate application of the simplified approaches

considered in the present report is provided in Section 3.
The resulting set of facilities addressed in this report is identified in the following as
NFOP (nuclear facilities other than NPPs).
External events affecting a nuclear installation could be listed as in the following,

according to the relevant safety standards for NPPs [23]:
(a) Human induced

Aircraft crashes.
Explosions (deflagrations and detonations) with or without fire, originated from
off-site sources and on-site (but external to safety related buildings), like storage
of hazardous materials, transformers, high energy rotating equipment.
Release of hazardous gas (asphyxiant, toxic) from off-site and on-site storage.
Release of corrosive gas and liquids from off-site and on-site storage.
Fire generated from off-site sources (mainly for its potential for smoke and toxic
gas production).
Collision of ships and floating debris (ice, logs, etc.) with the water intakes.
Electromagnetic interference from off-site (e.g. from communication centres,
portable phone antennas) and on-site (e.g. from the activation of high voltage
electric switch gears).
Any combination of the above as a result of a common initiating event (e.g.
explosion with release of hazardous gases, smoke and fire).
(b) Natural
Earthquakes.
Extreme meteorological conditions (temperature, snow, hail, frost, subsurface
freezing, drought).
Floods (from tides, tsunamis, seiches, storm surges, precipitation, waterspouts,
dam forming and dam failures, snow melt, landslides into water bodies, channel
changes, work in the channel).
Landslides and avalanches.
Cyclones (hurricanes, tornadoes and tropical typhoons).
Abrasive dust and sand storms.
Lightning.
Volcanism.
6
This list is not exhaustive and other external events, not included in the list, may be
identified and selected as design basis external events at the site.
However, particularly in case of natural events, some scenarios need to be treated as

exclusion criteria for the site itself (e.g. local volcanism, local capable fault, aircraft crash,
etc.) and therefore they are not discussed here. Such exclusion criteria are usually more
restrictive than those for the NPPs, due to the different balance point between investment and
risk.
However, in case when NFOPs are located at the same site of a NPP, they share the
same hazard and therefore in these cases reference need to be made to the respective Safety
Guide for NPPs.
Other scenarios are addressed preferably through site protection features (e.g. site
drainage, protecting dams and levees, etc.) rather than with plant design measures and
therefore they are mentioned here.
Furthermore, considered external human induced events are of accidental origin.

Considerations related to protection of the plant from malevolent action by third parties are
outside the scope of this report, but they could easily be accommodated into the approach for
human induced events (such as explosions, aircraft crash, etc.).
In general, this report does not provide comprehensive guidance on design for any of
the mentioned external events. In fact, in many cases the same provisions applied for NPPs
need to be implemented for NFOPs (like in the case of extreme meteorological events,
extreme winds, etc.) without particular effort and therefore they are not mentioned here. In
some other cases, like for earthquakes or for aircraft crash, the design procedures usually
require rather high investments and therefore they are in the main scope of this TECDOC,
which proposes simplified approaches for their analysis.
In conclusion, in the following, main emphasis is given to design against earthquake,

flood and aircraft crash which are usually associated with the most expensive protecting
features in siting and design, while other events are discussed only for their specific effects on
NFOPs safety.
1.4. STRUCTURE
This TECDOC consists of nine sections and seven appendices.
Section 2 presents the safety objectives to be met during the design of the NFOPs
considered under the scope of this report and the acceptance criteria usually considered for
design. In addition, this Section shows the safety requirements developed to meet the
objectives and to demonstrate the acceptability of the design. These are used to classify the
facilities and to categorize their structures, systems and components to facilitate the
straightforward application of the different methods included in the present report. Section 3
presents the basis for the design approach and component qualification, in line with the safety
concepts defined at Section 2. Section 4 deals with the design basis definition for a plant:
from the ground motion (earthquake) to other hazards. Based on availability of instrumental
and/or historical data, procedures are suggested for the definition of a reliable design input.
Section 5 contains guidance for site investigations, particularly of geological and geotechnical,
including simplified methods for soil liquefaction analysis. Section 6 presents some simplified
7
methodologies for the seismic design of building structures. Sections 78 deal with design of
equipment and piping with special emphasis on structural detailing and anchoring. Section 9
presents some safety principles for the installation and operation of monitoring systems at the
site, their quality assurance and data processing procedures.
Appendices present experience of some States in design and qualification of

structures and components that can be used to solve specific design problems of NFOPs.
2. SAFETY OBJECTIVES AND CLASSIFICATION
2.1. INTRODUCTION AND EXAMPLES OF NFOPs

In Table I, the results of a very short analysis of the NFOPs in the scope of this
publication is presented, as background information for the identification of the most common
safety requirements for such facilities. The analysis is mainly based on INES data [26] and
bibliographic results [27].
TABLE I. EXAMPLES OF NFOPS WITH ASSOCIATED MAIN SAFETY ISSUES (source:

analysis of SARs and bibliographic references)
Facility Main structural External Safety issues Notes
characteristics Event
design
Fuel Many process buildings Usually 100 Radiation Design for
conversion + support structures year return Chemical hazard decontamination and
plants (pipe bridges, cooling period Fire, explosion decommissioning
towers, chemical tank Physical interaction
farm, boiler, diesel fuel Large quantities of
storage, electrical radioactive waste
substation, settling
basin)
Fuel Process bldg Usually 100 Criticality

enrichment (centrifuges) + feed and year return Radiation
plants withdrawal bldg, plant period Chemical hazard
support facilities and Fire, explosion
utilities Physical interaction
Fuel Process bldg + plant Usually 100 Criticality Confinement in case

fabrication support facilities and year return Radiation of accident
plant utilities period Chemical hazard Full functionality
Fire, explosion during and after DBE
Physical interaction
In case of MOX, highest
concern for the Pu[?]
toxicity
Fuel Process bldg + waste Same return As above. Large quantities of

reprocessing treatment + plant period for hazardous waste
plants support facilities and external
utilities events as in
the NPPs
Independent Storage pool + Sometimes Radiation Leaktightness is
fuel storage transportation systems + same hazard Fire, explosion required to pools (wet
facilities (for plant support facilities of NPPs, solution)
NPPs and and utilities sometimes Storage of casks (dry
8
reprocessing 100 years solution)
plants) return period
Research Reactor bldg + plant Criticality
reactors support facilities and Radiation
utilities Fire, explosion
Containment in some
cases
In addition, recent surveys [28] have been carried out on accidents at NFOPs, also
following Tokai Mura accident, where the following points were highlighted.
Since 1945, nearly 60 criticality accidents of varying degrees of severity have occurred
worldwide, mostly at military related sites in the USA and the former Soviet Union.
All but two of them prior to Tokai Mura took place before the early 1980s. The
two most recent both took place at Russian military related facilities in 1997.
Of those before the early 1980s, 36 (two thirds) occurred either in research reactors or
in laboratories working on critical assemblies. None of them resulted in any
significant release of radioactive material to the environment, but there was a total of
nine fatalities. One further person died in an incident in 1997. The causes and
consequences of each accident varied according to specific operational circumstances
at the various facilities.
The others, one third (21), occurred in nuclear fuel cycle facilities. Seven people died,
including one in 1997, and 40 more received significant exposure to radiation. Again,
there was never any significant release of radioactive material to the environment.
Twenty out of 21 occurred in US or Soviet facilities the exception was a 1970
incident in the United Kingdom. Twenty out of 21 also involved liquid solutions of
fissile materials, and none of them involved either failure of equipment and/or material
or faulty calculations.
The main cause appears in most cases to have been a failure to appreciate the number
of possible accident scenarios, particularly bearing in mind the potential for human
error.
Other statistics, focused on criticality, are available in Ref. [29] with many details
oriented to prevent similar accidents in the future.
Concerning research reactors, data are available in Refs [30, 31].
As a result of these surveys on the main facilities in the scope of this TECDOC,
aimed at the identification of the most common safety concerns for such facilities, the most
recurrent accident scenarios in NFOPs were identified as:
Criticality
Fire and explosions
Chemical hazard: release of radioactive effluents or prevention of safety related
operator actions
Physical interactions among components: drop loads, impairment of operator actions,
particularly as a consequence of external events
Internal misoperation with radioactive material spreading on-site and off-site.
9
These scenarios might be strongly affected by plant conditions that in the case of
NFOPs could be very critical, particularly in relation to:
(1) ageing (some facilities are very old and designed for different purposes, including
support to military oriented tasks),
(2) maintenance (standard of maintenance have proven to be poor in many cases for
NFOPs)
(3) decommissioning procedures (often, extensive facility contamination and presence of
many sources of chemical hazard can complicate the process)
According to this basic survey carried out on the most significant facilities, it comes
out that the most common requirements for structural design of buildings, components,
equipment and distribution systems can be summarized as follows:
Confinement or isolation (ventilation systems): leaktightness
Leaktightness of pools, through liners or prestressing
Avoidance of drop loads
Functionality of safety related items (e.g. heat removal, ventilation, fire control, etc.)
Appropriate anchoring of items
Avoidance of interaction problems: mechanical, chemical or through impairment of
operator action
Radiological shielding.
2.2. SAFETY OBJECTIVES

NFOPs need to meet the general safety objectives as stated for NPPs [21].
The general safety objective stated for all nuclear installations is to protect
individuals, society and the environment from harm by establishing and maintaining in
nuclear installation effective defences against radiological hazards. This general nuclear safety
objective is supported by two complementary safety objectives dealing with radiation
protection and technical aspects.
The radiation protection objective is to ensure that all operational states radiation
exposure within the installation or due to any planned release of radioactive material from the
installation is to kept below prescribed limits and as low as reasonably achievable, an to
ensure mitigation of the radiological consequences of any accident.
The technical safety objective is to take all reasonably practicable measure to

prevent accidents in nuclear installations and to mitigate their consequences if they occur: to
ensure with a high level of confidence that for all possible accidents taken into account in the
design of the installation, including those of very low probability, any radiological
consequences would be minor and below prescribed limits; and to ensure that the likelihood
of accidents with serious radiological consequences is extremely low.
To achieve the above mentioned safety objectives in case of an external event (EE),
the design of the facility needs to prevent any damage to structures and equipment that could
lead to significant exposures to facility personnel or members of the public because of
uncontrolled release of radioactive material.
The consequences to workers and public are discussed in Ref. [12].
10
In the case of the public, the estimated average doses to the relevant critical groups of
members of the public shall not exceed the following limits [32]:
(1) An effective dose of 1 mSv in a year;
(2) In special circumstances, an effective dose of up to 5 mSv in a single year provided
that the average dose after five consecutive years does not exceed 1 mSv per year;
(3) An equivalent dose to the lens of the eye of 15 mSv in a year; and
(4) An equivalent dose to the skin of 50 mSv in a year.
In the case of workers, the consequences need to be expressed in terms of

occupational exposure. The limits are the following:
(1) An effective dose of 20 mSv per year averaged over five consecutive years;
(2) An effective dose of 50 mSv in a single year;
(3) An equivalent dose to the lens of the eye of 150 mSv in a year; and
(4) An equivalent dose to the extremities (hands and feet) or the skin of 500 mSv.
2.3. DEFENCE IN DEPTH

The application of the defence in depth approach to nuclear power plants is well
defined in engineering tradition (see for example [12, 33]), with a specified number of
barriers, levels and features.
For NFOPs, the number of levels could be maintained, but the number and role of the
barriers could be different from that of a power plant, depending very much on the accident
scenarios and the reliability that may be associated with any defence level. There is no unified
approach in Member States and therefore only a case by case approach supported by some
probabilistic evaluation seems to be the only strategy to be followed.
However, some guidelines can be provided for a reasonable implementation of the

approach to NFOPs:
One of the basic ideas of the defence in depth approach is a proper balance between
prevention of an accident and mitigation of its consequences. This balance needs to
be applied also to NFOPs to avoid design strategies in which low radioactive
inventory facilities are designed only with reference to mitigation measures for
potential accident scenarios.
In other words, the low potential for radioactive contamination (i.e. a small source
term) is not to be used to justify a low number of provisions on the prevention
side. The ALARA principle in fact aims at protecting also the workers at the site who
deserve a prevention policy against radioactive contamination.
The reliability of the levels of defence in depth dealing with design measures is
implicitly guaranteed by the application of the single failure criterion. In NFOPs
single failure may be applied to systems (as it is done for NPPs) and also to items in
some cases, but its consequences on the design need to be considered together with
the selected QA approach for design, construction and operation, with maintenance
procedures and with the probability associated to the initiating events.
Particularly for active systems, redundancy, independence, fail-safe design and
11
appropriate auxiliary services can achieve the required reliability levels with
reference to random failures in conjunction with appropriate maintenance procedures.
The balance has to be defined for the specific facility. In any case, the operator
cannot be considered as a redundancy due to the consideration of human errors.
Passive systems (typically structures, vessels, support devices, pipelines, etc.), are
usually designed to perform their safety functions within the prescribed operating life
only on the basis of appropriate QA, design standards, maintenance and surveillance
procedures. Such assumption needs to be discussed in the SAR and reflected into
appropriate procedures covering the plant lifetime.
With special reference to some external events (typically earthquake, flood, aircraft
impact, etc.), also common cause failures may have a high influence on the reliability
of the defence in depth levels. Therefore, a special assessment needs to be carried
out, independently from the application of single failure criterion, aiming at a proper
balance of quality, segregation (physical separation) and diversity requirements to
safety related items.
A typical example concerns seismically classified items where a high level of quality
(i.e. of qualification requirements) can guarantee an appropriate safety level where
redundancy or diversity may lead to less effective solutions.
In case of very rare events (e.g. aircraft crash) or combination of events (also internal
and external scenarios), particularly when enough warning time is guaranteed (e.g.:
heavy storms), or in a re-evaluation of an existing plant, many Member States accept
a reduced level of defence which is limited to guarantee the shutdown of the nuclear
reaction potentially leading to radiological accidents. Such limitation requires a
critical interaction with operating procedures (warning, monitoring, operator actions,
inspections) that has to be defined on a case by case basis.
Guidance for a proper application of these concepts can be found in Ref. [34].
2.4. SAFETY CLASSIFICATION OF THE FACILITIES

As a preliminary screening criterion for the NFOPs to be considered as facilities with
associated radiological risk (in case of external events), it can be assumed that a
conservatively estimated amount of the radioactive inventory is released following an accident
(induced by an external event).
In case the screening shows that the radioactive inventory of the facility is not
significant (i.e. the criterion for the limit to the public is met), conventional standard for the
design of the facility can be applied, according to the national codes for industrial facilities (of
course a parallel evaluation needs to be carried out for other risks, like chemical, fire, etc.).
In case the limits for workers are exceeded, local engineering provisions need to be
put in place for their prevention and mitigation. Conversely, if the release is significant, an
evaluation of the risk associated to the facility needs to be carried out to define an appropriate
safety margin to be associated with the design for EEs. Risk is defined as the product of the
probability of the initiating event (external) with a measure of the consequence generated by a
radioactive release.
The evaluation of the risk posed by a facility needs to analyse continuous emissions,
accidental releases from operation, transportation or human induced events and from natural
hazards. It can be health related or environmental related.
12
The targets of the risk are firstly the workers at the site, the people living near the
site, the people affected by the transportation of contaminants through air, waterways or by
contamination of agricultural products. Secondly, also the ecological systems need to be
considered for their influence on the food chain.
Many procedures have been developed for risk analysis, with emphasis to risk
comparison among different sources in the society and between voluntary and involuntary
causes. Comprehensive information on the state of the art is collected in Ref. [35]. For the
objectives of the present publication, only the radiological risk is addressed and only the risk
induced by external events on the production facilities, without analysis of the effects
associated to the waste which is supposed to be treated and transported according to specific
national regulations.
Following the definition of risk given above, the first term, i.e. the probability of any
EE at the site, needs to be evaluated according to its site specific hazard: detailed information
is provided in this publication for the most relevant EEs in a graded form, with reference to
[16].
The second term of the risk equation, i.e. the probability that EE generate a
radiological consequence depends on characteristics both of the source (facility) and of the
events, such as:
Amount, type and status of radioactive inventory at the site (e.g.: solid, liquid,
processed or just stored, etc.)
Intrinsic reliability and hazard associated to the chemical and physical processes
which take place in the facility (e.g. processes which implies transportation of
hazardous substances might show a higher hazard than cases when fuel is not moved,
like in operating NPPs)
Installed thermal power of the facility
Configuration of the facility for different kinds of productions
Concentration of radioactive sources in the plant (e.g.: in the power reactor, most of
the inventory is in the reactor and in the fuel pool, while in processing plants it is
distributed in the layout)
Facilities designed for experiments and research (such activities have an associated
intrinsic unpredictability) or in any case subjected to frequent configuration and
layout changes (such as activities associated to new product developments)
Need for active safety systems to cope with mitigation of postulated accidents;
amount of engineering features implemented for preventing and mitigating serious
consequences from accidents
Possibility of installation of warning systems able to detect in time the potential
unfavourable development of an event (e.g. meteorological events vs. aircraft
crashes)
Characteristics of the process or of the engineering features which might show a cliff
edge effect (potential small deviations in plant parameters giving rise to severely
abnormal plant behaviour) in case of an accident, without possibility to prevent the
degeneration into radiological consequences
Characteristics of the event (e.g. wind and explosion have a high potential for
dispersion, while earthquake and aircraft crash have a minor contribution on the
dispersion)
13
Environmental characteristics of the site relevant to dispersion (e.g. windy area,
coastal site, etc.)
Easy implementation of emergency planning in relation to the event: access to the
site, evacuation routes availability, time delay between accident and releases, etc.
Potential for long term effect in case of contamination (long lived radionuclides,
persistent effect in the environment)
Number of people potentially affected by an accident at the facility
Potential for off-site versus on-site radiological contamination.
A general evaluation of the risk associated with the NFOPs is difficult due to the high
number of variables depending on the specific layout. In general a reasonable and reliable risk
classification can be made only on a case by case basis, possibly after a detailed PSA analysis
(usually not available at the design stage). Also generic probabilisticdeterministic procedures
have been developed, mainly experience based, for hazard identification, accident modelling,
effect propagation and response evaluation. Some indications on the main tasks are contained
in [35, 36].
In the framework of this report, mainly oriented to an identified group of facilities

and risks, the issues listed above could be interpreted as criteria for such risk classification,
driving the final evaluation of the risk associated to the facility, ranging from a minimum risk
(conventional building) to the highest values (nuclear power plants).
However, a full probabilistic approach in the classification of NFOPs in relation to

the risk they pose to the environment is in general very difficult to be applied because of the
large variety of designs, the wide range of power levels, the different modes of operation and
purposes of utilization, the particularities of the site, and differences among operating
organizations at the facility. Because of this, there is no consensus among the Member States
both on the facility classification and on its effects in the design.
A reasonable and very simplified approach could account only for a reduced number
of the criteria described above, such as:
Class 1 (high hazard): potential for significant off-site radiological contamination
Class 2 (moderate hazard): potential for significant on-site radiological
contamination, with high criticality hazard
Class 3 (low hazard): potential for significant on-site radiological contamination
Class 4 (conventional hazard): industrial risk, conventional industrial buildings.
Following this approach, for some facilities the application of the mentioned criteria
for facility classification shows a strong correlation between risk associated with the facility
and its installed power or radioactive inventory only, instead of all the criteria described
above. This correlation might simplify the classification process at the beginning of the
design. Of course such assumptions need to then be assessed in the safety assessment phase
and justified in the SAR.
According to this criterion, for example, research reactors with powers of several tens
of megawatts, fast neutron spectrum reactors or small prototype power reactors can be
classified as nuclear power plants and they are beyond the scope of this TECDOC in the sense
that their design needs to follow the prescriptions for NPPs (i.e. the relevant safety standards).
14
Fuel reprocessing plants are usually classified in the highest class of risk, but the
engineering features are very different from NPPs and therefore they are included in current
discussion.
Most of the remaining facilities can be classified in Class 2 or below following the
criteria described above and therefore their design could follow the guidance in this report.
The classification of the facility has influence on the following tasks through a
decisional process which is not unified among the Member States:
Selection of the annual exceedance probability for any EE at the site: higher values
can be used for lower class facilities (see Section 4).
Need for a site specific hazard evaluation: the higher the class of the facility, the
more specific it has to be (see Section 4).
The extension of site investigation campaigns: the higher the class of the facility, the
more accurate they have to be (see Section 5).
Definition of the safety margins in the design or, in general, of the reliability levels
that safety functions need to comply with (see below and Sections 6, 7).
Number of safety features: number and quality of defence in depth barriers and
levels, emergency systems, emergency procedures, etc. (in this case, also engineering
tradition, construction practice, economic consideration have strong influence). In
particular the safety classification of systems and items (including their acceptance
criteria in case of an external event) is a very important task affected by the facility
classification and assessed by the Safety Analysis Report (SAR).
Adoption of different QA level in siting, design, construction and operation.
With the support of these concepts, NFOPs need to be sited, designed, constructed
and operated so as not to exceed the recommended limits in all operational states and accident
conditions.
In many cases, it is convenient to develop some conservative assumptions in order to

avoid, even for the highest classes of facilities, expensive investigation campaigns, hazard
studies and design processes. However such conservatism is not to be confused with the
requirements in terms of safety discussed above and associated to the facility classes:
conservatism implies reduced efforts in the design stage, while a different classification in
terms of safety is related to a different amount of safety margin. The analysis of the interaction
between the two aspects, namely the safety classification of a facility and the conservatism
required by a simplified design and siting procedure, is the main target of this report and it is
discussed in the following chapters, for every task.
For all nuclear installations, including NFOPs, a safety analysis report is prepared
which includes safety analysis of the radiological consequences initiated from external events.
These analyses need to include assumptions concerning the safety functions associated with
items important to safety which have to remain operable to keep the consequences within the
above mentioned radiological limits.
In particular, for research reactors, annex to Ref. [5] presents selected safety
functions associated with items important to safety (e.g., buildings and structures, reactor core
and internals, fuel matrix and cladding, reactivity control system including the reactor
15
shutdown system, reactor coolant primary circuit, emergency core cooling system, ventilation
systems) which may be challenged by the occurrence of an earthquake.
2.5. CLASSIFICATION OF ITEMS IN RELATION TO EXTERNAL EVENTS

Following the general criteria defined above for the classification of the facilities
according to the risk they pose to environment and population, the safety analysis of the plant
needs to assess number and characteristics of the safety systems (including the barriers to
prevent radiological accidents) through analysis of the consequences of any postulated
initiating event (including external events) with application of the single failure criterion to
any safety group (see [21] for definitions). Special limitations to redundancy and diversity
may be considered for passive systems, as specified in [21].
The safety analysis needs to be mainly based on the application of the defence in
depth, aimed at maintaining the effectiveness of the physical barriers placed between
radioactive materials and workers, the public and the environment.
Special care needs to be used in the application of the common cause failure
concept in the safety classification as external events may affect many items at the same time
(the case of the earthquake is typical).
The design of systems and components in relation to external events (EE) requires a
classification with the following aims:
(1) the identification of items and systems important to safety. Items needed to perform all
actions required for the postulated initiating events have to be identified to ensure that
the radiological limits are not exceeded, even in case of degeneration into a plant
accident. Items which would lead to radiation exposure to workers or public in case of
failure need to be added to such category. These items have to be protected in case of an
EE.
(2) the identification of the safety functions assigned to them, and therefore their acceptance
criteria (e.g. integrity, stability, operability, etc.). The design has to meet such
acceptance criteria.
For a rational basis in the design, such information might be organized through an EE
classification process of all plant items in order to identify the items who need to be
considered in case of any EE and the relevant requirements.
To this aim, three classes of structures, systems and components (SSCs) can be
identified:
External Event Class 1 (EEC1): safety systems pertaining to EE safety groups [21]
or safety systems which, during and after an EE, interact with items in the safety
group of the EEs
External Event Class 2 (EEC2): safety systems which are not in the EE safety
groups and which, during and after an EE, do not interact with EEC1 items.
External Event Class 3 (EEC3): items not important to safety which could impair
proper functions of EEC1 and EEC2 items or the operator action.
Acceptance criteria need to then be stated for both the operational state of the
facilities and the accident conditions considered in the design of the facility.
16
Such criteria vary among Member States: they may include considerations such as
those listed below.
(a) Radiological criteria, such as:

ALARA levels
Dose limits (or targets) for facility staff and workers at the facility site and the
general public;
Release limits to the environment; and
Risk criteria (where applicable).
(b) Performance criteria, including

Limits to damage of the physical barriers
Limits to damage of safety significant structures, systems and components
Frequency limits for certain anticipated operational occurrences and for
particular accident conditions, including frequency limits for significant
damage of physical barrier (where applicable).
These criteria need to define relevant behaviour limits for any component and
structure in any EE class above according to the safety function which is associated with them
in case of a sequence where an external event is the PIE.
It turns out that the EE classification is not related to the kind of behaviour limit
(acceptance criteria) associated with the component/structure, but it is related only to the
amount of safety margin required (Fig. 2). The more relevant a component is for safety; the
highest safety margin has to be considered in its design. Therefore, the classification aims at
the identification of the safety margin to be applied to any safety function (horizontal arrow in
Figure 4), with grading from the highest associated radiological risk to the lowest. In this
framework, the classification does not mix the behaviour limit (elasticity, integrity, etc.) with
the required amount of safety margin (vertical arrow in Figure 2).
In probabilistic terminology, the deterministic concept of safety margin needs to be

related with the probability of exceedance of the design limit. However, in this report only
reference to the safety margin needs to be made, due to the main objectives stated in Section
1.
This safety margin needs to be intended in a general sense as it might include design
safety margins, but also qualification requirements, requirements for redundancy and
diversity, reliability evaluations, number of safety features, QA prescriptions etc., as described
above.
Moreover, the EE classification should not have any influence in the application of a
specific methodology for the numerical simulation, as it was in the past [12]: numerical
simulation techniques need to be validated in terms of their simplified assumptions and
therefore in terms of their conservatism.
17
Design Safety Safety Interacti Non
limit/Items related systems ng items classified
items (not
( Barriers needed
and for EE)
active)
Functionality
Operability
Integrity
Elasticity
Ext Ev. cat 1 EE-C2 EE-C3
Leaktightness
Stability
Safety Class 1 SC 2
Classification is related to hazard and not to the required
FIG. 2. Interaction between item classification and acceptability limits.
This approach is more in line with the availability of sophisticated calculation

techniques, which can be applied to the design of NFOPs without high additional costs to the
process.
In relation to the component design, an interaction could be stated between the

facility classification and the item classification, as presented above. This interaction is
discussed for example in Ref. [37] for the probabilistic methodology and it is practice in many
Member States.
A possible model for such interaction in the design tasks relying on linear
approaches, may follow the idea that the control of the safety margin is managed through
reduction of the internal element forces only, in a sort of single parameter control. In this
case a design class can be associated to a factor applied to the internal forces. Within each
design class there is the potential for one or more acceptance criteria to be used. Table II
shows a proposal for such interaction.
TABLE II. RELATIONSHIP BETWEEN FACILITY CLASSIFICATION AND ITEMS

CLASSIFICATION IN CASE OF SINGLE PARAMETER CONTROL
DESIGN CLASS SELECTION
Hazard High Moderate Low Conventional

Class 1 2 3 4
EE Class
EE Class 1 Design Class 1 Design Class 2 Design Class 3 Design Class 4
18
3. GENERAL APPROACH TO THE DESIGN OF NFOP
3.1. GENERALITIES
The selection of the most suitable design methodology for the facilities selected in
previous Chapters need to start from the identification of the common structural aspects. A
selection of them is as follows:
Most of NFOPs show a very distributed and irregular layout, mainly composed of flat
buildings. Elevation can be high (about 2030 m), but the lateral extension is almost
always the dominant dimension.
The most common construction technologies rely upon concrete frames, masonry,
shear wall structures combined with steel frames for equipment
Embedment is usually limited to 35 m, but in some cases the whole building is
deeply embedded for more than 50% of their height
Buildings are connected at many points by piping, conveyors, transfer tunnels, etc.
Many equipment are connected in long chains: glove boxes, piping racks and bridges,
etc.
Many pools have liners
Many cranes move heavy and hazardous loads.
These common aspects define also a sort of limit of applicability of the

methodologies suggested in the following or, in many cases, they represent the critical areas
where checks have to be carried out.
3.2. SAFETY MARGIN CONTROL IN THE DESIGN PROCESS

For comparison, the general approach to a safe design is shown in Figure 3 in the case
of NPPs: the possible contributions to the overall safety margin in all the major phases of site
evaluation, design, construction and operation are highlighted.
In the case of NFOPs, in order to meet the safety requirements discussed in Section 2
and to allow simplified and well consolidated design procedures, the requirements for the
design procedures can be modified as in the following:
Hazard Design Load Design Material Constr. Release

eval. Basis Comb meth capac. QA dispers.
Availab. Prob. Coeff. Accuracy Material Deman- Limits

of data target properties ding QA for
Margin Method proce- doses
Accuracy over det. (elastic, Safety dures
(mean, evaluation ductility, margin
median, etc.
best estim,
84%, etc.)
FIG. 3. Steps in the design process where safety margin might be located.
19
(1) A grading has to be defined between nuclear power plants safety approach and
conventional buildings, according to their potential for radiological release and to the
safety functions defined by the safety analysis. Site specific hazard evaluation
procedures and site investigation campaigns, return period for design basis events,
design class concept, as defined in Sections 2, 4 and 5, are affected by grading.
(2) The safety margin in the design needs to be easily proven, even in cases when
different standards are applied. The single parameter concept, as defined in Section
2, can be applied to this concern
(3) An adequate level of conservatism needs to be guaranteed to compensate for
reduced site hazard analysis, site investigation campaigns and for simplified analysis
methods, following the main objectives of this report. Larger areas for investigation
campaigns, simplified, conservative, design basis and penalty coefficients
according to the applied analysis method can be applied to this concern
In this framework, conventional standards may be applied for structural design and
qualification of components in NFOPs, with a big advantage for the management of design
and qualification phases. The role of carrying the safety margin correspondent to the risk
classification of facility and items is left either to the design basis loads, or to a reduction
factor applied on the internal forces, both external to the black box of the standard design.
In either case, the load reduction factor has to be selected according to the
combination of facility and item classification, in a graded way, for example as suggested in
Table I from no reduction up to the building code and standards for conventional buildings.
It is noted that in general internal force reduction has the form of a factor, but in
some cases it may also have a physical meaning, such as ductility in the seismic design of
structures. Ductility usually has a more general meaning, as its main part is related to the
energy absorption in the specific structural configuration. However in this report structural
detailing is not explicitly addressed, for simplicity, and therefore ductility is limited to that
minimum amount that any structure can provide without specific provisions.
With this approach relying on conventional standards, other conservatism sources are
implicitly included, such as: structural damping, soil structure interaction, structural detailing,
etc.
It is common practice to apply different standards to the design of different items,
mainly depending on the supplier. For example, building structures, mechanical components
and electrical equipment, often follow national standards of different States.
In this case it is suggested that a detailed analysis is carried out on the compatibility
of such standards to guarantee the proper management of the overall design safety margin,
according to the general concepts described above.
Another conservative margin is usually applied to the hazard evaluation to
compensate for simplified investigation campaigns or for simplified calculation
methodologies. Such conservatism needs to be demonstrated by the designer and it never
replaces the safety margin required by the classification discussed above, but is to be added on
top of it.
The final combination of various effects is described in Fig. 4, where the mechanisms
suggested to control safety margin and conservatism are listed for the most important siting
and design tasks. It is noted that while facility and item classification are used to calibrate the
20
safety margin, simplified design basis, reduced site investigation campaigns and simplified
calculation methodologies may provide conservatism to the overall design.
In conclusion, in this design approach load combinations and material properties
could be taken by standard design procedures and the whole siting/design process has only
two major modifications compared to conventional building design:
the hazard and the site properties are required to be site specific, even if evaluated
with simplified and conservative procedures. The internal forces need to be corrected
by a pseudo-ductility coefficient (reduction factor to be applied to the internal
forces) according to the interaction of facility and item classification, as shown in
Table I
the assessment of the acceptance criteria (and therefore the design limits associated to
the material capacities) needs to be carried out explicitly in compliance with the
safety objectives and documented in the SAR.
As a consequence, the site exclusion criteria may be revised (i.e.: in case of EE which
require specific, complicated and expensive protection structures and systems, it might be
convenient to exclude the site, when possible) to allow for simplified investigation campaigns
and simplified design rules. Moreover, a requirement on the minimum level of site
investigation needs to be defined and rules have to be defined to compensate for the limited
investigation campaigns and to guarantee a certain conservatism. Section 4 deals with such
aspects.
3.3. DESIGN PROCEDURES AND SAFETY ASSESSMENT

In case the above mentioned approach is followed for structural design, the safety
assessment of the classified items is straightforward and implicit in the design.
However, there are cases in which the safety margin is not easy to be assessed and
other approaches have to be put in place:
in case of equipment and components, where the acceptability criterion is either
functionality or operability, the EE classification has to be reflected into
requirements for direct qualification of the item by testing.
in case of items in EEC3, it is difficult to assess equipment performance during an
EE only by analysis. Therefore, a post-construction walkdown verification is
suggested mainly focused on anchoring (mainly seismic), EE induced fire, flood,
chemical release, drop loads etc., spatial interaction between items and building and
items, impairing of operator access and action.
However, neither acceptance test nor walkdown can provide a measure of the safety
margin (only fragility testing can do that) and therefore, in case such an evaluation is needed,
special procedures can be followed, borrowed (and maybe simplified) from practice in NPPs,
for example in Refs [22, 25].
Moreover, the application of single failure and common cause failure are intended to
provide the design with a high level of reliability. A good balance among redundancy,
diversity and quality requirements of items and systems needs to be guaranteed. A good
reference for such analysis is provided in [34].
21
22
Siting/ Design Facility EE classification of Design basis Site investigation Selection of

Task classification items according to evaluation campaign calculation
according to its their safety methodology
potential for function
radiological release
(1) (2) (3) (4) (5)
Influence on Safety margin Safety margin Conservatism Conservatism Conservatism

the design
Mechanism Modification to the Simplified design Enlargement of Penalty

used for the return period for basis investigation area coefficients on
application EEs (see Section 3) to compensate for element forces,
either of the lack of data and according to the
safety factor Site specific hazard simplified selected method
or of the evaluation and site investigations and conservatism
conservatism investigation (see Section 4, 5) in the design
parameters
(damping, etc.)
Reduction factor on element forces (see Section 6)
according to Design Class
(assuming linear analysis)
FIG. 4. Scheme of different sources of conservatism and safety factors.

3.4. AGEING EFFECTS ON DESIGN AND RE-EVALUATION
Traditionally, most of the NFOPs are not designed with reference to a design
lifetime like in the case of NPPs and a number of layout and plant modifications are normally
expected during their operation. Therefore the ageing aspects need to receive special care,
mainly through the following tasks:
appropriate provisions during design: this needs to focus mainly on appropriate
material selection and development of technical specifications for periodical
inspections
surveillance and testing to assess degradation of components and systems
development of a preventive maintenance programme
optimization of operating conditions
management of repairing, replacing or refurbishing of components.
More details are available in [11], where strong emphasis is given also to the re-
evaluation scenario.
These tasks need to be co-ordinated through an ageing management programme to be

implemented during the plant lifetime.
In view of this, at the design stage appropriate material selection needs to be carried
out and a detailed specification of the safety requirements to structures, systems and
components need to be compiled as input for inspection, maintenance and ageing
considerations.
In the re-evaluation of existing facilities the proposed approach is still valid in terms
of facility and item classification. In case not enough safety margin is shown by the design
assessment, the following two actions are envisaged, in a priority sequence:
(1) more refined procedures need to be applied eliminating excessive conservatism in the
various areas defined above
(2) an easy fixes programme needs to be implemented with the goal of a minimization
of the investment with a maximization of the safety margin increase. Procedures for
such assessment can be taken from the experience in the NPPs, such as [38].
According to general experience in Member States, these two actions provide a

successful re-evaluation of most of the existing facilities.
3.5. THE SPECIAL CASE OF RESEARCH REACTORS

Research reactors show a wide variety of technological solutions, radioactive
inventory, power level and siting environment. A limited selection of them was analysed at
IAEA to support their EE classification and therefore to derive suitable design procedures for
such type of structures.
Table III collects some data from worldwide installations, taken directly from the
safety analysis reports.
Few comments to the Table are summarized in the following:
23
(1) There is a very broad range in nominal power, which is expected to be associated to a
broad range of radiological inventory
(2) There are very big differences in layout, building design and safety system design
even with the same reactor type. This fact shows the difference policy in the safety
approach and design.
(3) There is a big difference in the design basis and in the standards applied for design:
in some cases, even if sited in similar environment, the design basis does not include
minimum requirements for external events and the design is carried out according to
national standards for conventional industrial installations.
Therefore, according to the general principles set up in the previous section, and to
the general background publications, in the case of research reactors the EEC1 and EEC2
systems may be associated to three different groups of items (in increasing order of
importance) according to the installed power in the reactor, radioactive inventory and core
design features, as described in the following:
(1) Only the reactor protection and reactor shutdown systems are required to be always
operable.
Based on safety analysis, it needs to be shown that for all credible accidents, the
radioactive inventory cannot be released from the principal barrier. This is the case of
a research reactor of low power, with power level less than one megawatt, where
therefore only the reactor shutdown system needs to be designed to withstand the EEs
(they are often identified as Class C reactors: low power, low inventory and
extremely low probability of breaking the fuel clad because of the reactor operation
and utilization).
In order to ensure coolability of the core it may be required to ensure no disruption of
the core, i.e., that reactor internals and other equipment or structures such as the
crane bridge or the building roof need not to fail in a way that would obstruct the
cooling of the core.
(2) The reactor protection and shutdown systems plus a system to remove decay heat
from the core (shutdown core cooling system, whose function is executed in most of
the research reactors by the emergency core cooling system) are required to be always
operable.
This is the case of a research reactor of medium power, with power level less than
two megawatts, where in addition to the reactor shutdown system, cooling of the
reactor needs to be incorporated into the design and the capability of the cooling
method to withstand the EEs needs to be demonstrated (they are often identified as
Class B reactors: medium power, intermediate inventory and low probability of
breaking the fuel clad during operation by loss of cooling or coolant).
(3) The reactor protection and shutdown system, a system to remove decay heat from the
core plus a confinement system (building plus normal or emergency ventilation) to
control the release of radioactive material to the environment are required to be
always operable.
This is the case of a research reactor of high or medium power, where a confinement
system (e.g.: containment building plus emergency ventilation) needs to be utilized to
meet the requirements of previous Section (they are often identified as Class A
reactors high power, high inventory and low probability of breaking the fuel clad
during operation because of engineered safety features incorporated during the
reactor design).
24
TABLE III. SITING AND DESIGN DATA FOR SOME RESEARCH REACTORS IN THE WORLD (SOURCE: SARS)
Location type power (MW) typology moderator DBE (g) Design conf. Confinement siting features
Ghana, GHARR1 0.03 pool 0.2 Confinement

Pakistan, China
(4 reactors)
USA UCLA 0.1 pool 0.5

Canada SLOWPOKE 0.5 pool + tank 0
Romania TRIGA 0.5 pool 0.4 (VIII+0.5 MSK)
Mexico TRIGA II 1 pool 0.3 confinement
Slovenia TRIGA II 1 pool 0.3 (IX MCS)
Indonesia TRIGA 1 pool 0.6 Reactor containment
building
Malaysia TRIGA II 1 pool 0.6 confinement ACC, fire, explosion,
earthquake, flood,
storm
Norway JEEP II 2 tank HW 0 diameter 22 m, H=22 Containment (P test 0.15 kg/cm2)
m
Morocco TRIGA III 3 pool 0.35 Confinement
Greece GRR-1 5 pool 0.18 Prestressed pool Confinement
Pakistan PINSTECH 5 pool 0.2 0.1 g for aux bldg. containment fire, lightning, storms,
earthquake
Chile CNCR 5 pool 0.3 Liner 6 mm Confinement (SMA 0.6)
Australia, HIFAR 10 tank HW 0.2 Steel structure on a Confinement (P test 10 kPa)
Denmark, concrete foundation,
Germany (6 diameter 21 m, H=21
reactors) m
Hungary VVR-SZM 20 pool 0.1 Reactor containment Confinement
building
Netherlands HFR 45 vessel + pool 0.1 Containment
(JRC)
25
(4) In addition to those requirements of the so-called Class B, it is necessary to
incorporate an emergency ventilation system to mitigate the consequences if the
radioactive fission products escape the principal barrier in spite of the requirements
imposed by Class B. Based on safety analysis it needs to be shown that a confinement
system is necessary to mitigate the consequences of some design basis accident. The
operability of this confinement system therefore needs to be guaranteed.
For this class, it is much more difficult to assign a range of power levels. Clearly,
however those reactors that are near the limit of applicability of this publication are
in this class, usually above some megawattage (25 MW).
Table IV shows an example of the relationship between classification of research

reactors and their structures, systems and components with regard to their design to EEs.
TABLE IV. CLASSIFICATION OF RESEARCH REACTOR FACILITIES, OF

STRUCTURES, SYSTEMS AND COMPONENTS
CLASSIFICATION OF RESEARCH REACTORS FACILITIES
FACILITY CLASS A B C
RADIOACTIVE INVENTORY HIGH
INTER- LOW
MEDIATE
POWER LEVEL1 HIGH MEDIUM LOW
CATEGORIZATION OF STRUCTURES, SYSTEMS AND COMPONENTS2
SHUTDOWN AND PROTECTION SYSTEMS EEC1 EEC1 EEC1
SHUTDOWN CORE COOLING SYSTEM EEC1 EEC1 EEC3
CONFINEMENT SYSTEM EEC1 EEC2EEC3 EEC3

(BUILDING AND EMERGENCY VENTILATION)
EMERGENCY POWER SUPPLY EEC1 EEC2EEC3 EEC3
WATER PURIFICATION SYSTEM EEC3 EEC3 EEC3
1
For the purpose of the present report, it is assumed that research reactors are grouped into three groups
attending their power level: low, medium and high power research reactors. The first group (low power research
reactors) includes subcritical and critical assemblies and research reactor up to 50 kW of nominal power and
represents about 40 % of the currently operating research reactors; the second group (medium power research
reactors) may include research reactors with powers higher than 50 kW and lower than 1.5 MW and represents
about 20 % of the currently operating reactors; finally, research reactors with power higher than 1.5 MW may be
included in the third group (high power research reactors).
2
the assigned categories are related to the design and construction requirements, particularly referred to civil and
mechanical structures, in accordance with the design level (or free-field acceleration) assigned to the DBE (see
Table VI).
26
More in detail, in the case of the reactor building, according to [5], the external event
scenarios require in general the following functions:
Automatic shutdown signal in case a certain threshold is reached
Shutdown functions and reactor protection systems
Residual heat removal system
Operator access and control (for independent manual scram and reactor monitoring)
Radiation monitoring during and after the design basis event
Long term monitoring of basic reactor parameters
Functionality of all other systems required for defence in depth according to the
specific acceptance criteria assigned to them (e.g. functionality of the emergency
ventilation system, ECCS, etc.).
However, according to engineering practice, the functionality of the items not needed
to bring the system into a safe shutdown might be excluded in many cases from the safety
related list as a consequence of the safety analysis of the reactor itself; this approach is
accepted in case it could be demonstrated that the radiological consequences of the postulated
accident which they are designed for are below the regulatory limits.
Therefore, according to such approach, the emergency ventilation system, the ECCS
and other safety systems might not be required in case of an accident for a reactor of limited
power (usually 23 MW) and therefore, in case they are installed for additional protection,
they might not be EE classified. These conclusions however have to be explicitly
demonstrated in the SAR with an appropriate calculation of the radiation dose which would
affect the site in case of a postulated accident.
As a consequence, in case of a design basis external event (DBEE), only the first part
of the functions listed above could be required for small research reactors, asking for a
classification of the relevant buildings according to the following design requirements:
full service ability of control room to enable operator actions;
full and safe access to the reactor building during and after a DBEE (including
lighting, availability of stairs and necessary housekeeping);
functionality of the safety systems mentioned above, with the redundancy requested
by the design for reliability principle [21];
availability of power supply to the safety systems during and after a DBEE; and
prevention of any interaction accident during and after a DBEE from loss of parts
over safety related components, internal flood, fire and/or hazardous substances
which could impair the required safety functions by components and operator.
These rules provide the basis for classification, and therefore for design, of RR
structures and components.
4. DESIGN BASIS FOR EXTERNAL EVENTS
4.1. INTRODUCTION
One of the main targets of this publication is the evaluation by simplified
methodologies of the design basis for the external events to be considered in the design of the
facilities, trying to limit the high costs associated to the investigation campaigns preliminary
to the traditional NPP design.
27
The evaluation of the site hazard for external events can in general follow the IAEA
recommendations for siting and design of NPPs and research reactors [5, 16]. However, the
site selection process may consider more restrictive exclusion criteria than for NPPs and
research reactors, as a compromise with the investment requested for the facility design,
construction and operation. In this sense, some events which are difficult or expensive to
protect the facility against could be screened out, like aircraft crash (thick shielding and
special equipment qualification would be required for facilities often without a containment),
explosions (blast resisting structures are required), flooding (site protection engineering
structures need to be built and maintained), etc. In fact in NFOPs usually the internal accident
scenarios do not imply high demand to the structures, like in the case of NPPs where, for
example, a containment is normally part of the design. Therefore a protection to external
events would add heavy requirements to the design which might be incompatible with a
rational approach.
In other cases an exclusion of some events from the design basis can be done
according to specific engineering provisions implemented in NFOPs. An example is the
consideration in the safety analysis of the typical spread building layout which might reduce
the consequences of an aircraft crash through physical separation of safety related items.
When NFOPs are at the same site of a NPP, they share the same hazard, but the
design basis might be different due to the considerations above.
Once a screening process on the events to be considered in the design is carried out, a
detailed site hazard has to be evaluated. Facilities in Class 1, 2 or 3 always require site
specific hazard evaluation, which could be simplified for the lowest classes, starting from the
requirements in the IAEA S series for Class 1.
Once the hazard has been evaluated, a grading on the return period is suggested for
facilities in Classes 1, 2, 3. There is no agreement in Member States on this subject, but a
reasonable proposal is shown in Table V, taken from [39], for the major hazard sources. Other
useful references could be found in Refs [35, 36] where a broader review of probabilistic risk
evaluation is described, with reference to the comparison with other sources of hazard for
public and environment.
TABLE V. GRADING OF MEAN ANNUAL PROBABILITY OF EXCEEDANCE FOR

DIFFERENT CLASSES OF FACILITIES (ACCORDING TO [39])
Event Class 1(1)(2) Class 2 Class 3 Class 4
Earthquake 1E-4 5E-4 1E-3 2E-3
Wind 1E-4 1E-3 2E-2 2E-2
Tornado 2E-6 2E-5
Flood 1E-5 1E-4 5E-4 2E-3
Aircraft crash 1E-5 1E-4
Note: (1) definition of classes is slightly different in [39] than in Chapter 2
(2) NPPs usually have an hazard level with an associated exceedance probability value smaller than
Class 1.
The difference in the hazard sources is consistent with the criteria provided in
Chapter 2 based on the evaluation of the impact of the different scenarios on the plant safety.
Site investigation needs to cover in general geology, seismology, geotechnical engineering,
28
hydrology, meteorology, marine environment, human development plans, industrial
installations, communications, naval, train, road and air traffic, etc. However, when data are
not available for the specific site (evaluations procedures on data reliability and requirements
for minimum monitoring are available in the respective safety standards), the extension of the
investigation campaigns might again be compensated by some conservatism in the definition
of the design basis for facilities in Classes 2 and 3, as suggested in the following for different
hazard types.
In any case it is suggested to carry out an estimation of the return period associated
with the selected design basis, at least to allow a comparison with national standards for
design of industrial facilities, and therefore to assess the compliance with the requirements of
the facility classification.
4.2. DESIGN BASIS EARTHQUAKE

In this section some simplified, conservative rules for design basis evaluation are
described for facilities in Class 2 or lower, in case few data are available at the site. Class 1
requires the application of [17] with reference to the exceedance probability defined in
Table V.
Based on availability of instrumental or historical data, one of the following methods

may be used to define design basis ground motions. In any case, the minimum design level
acceleration needs to be 0.1 g.
4.2.1. Site specific design response spectra based on instrumental data
In case some instrumental data, such that the location of seismogenic sources and
zones, attenuation relationships and possibly hazard maps within the region, are available for
the region, then a site specific design response spectra (including site effects) can be generated
by either using the envelope of response spectra (for 5 % damping) calculated from recorded
data or using hazards maps that are constructed using such data.
The Safety Guide for NPP siting [17] needs to be used in this case, particularly in the
evaluation of sufficiency and reliability of available data, with reference to the exceedance
probability in Table V. Appropriate simplifications approved by the Regulatory Bodies might
lead to a reduction of the embedded safety margin according to the facility classification.
4.2.2. Design response spectra based on historical seismicity

In the event that instrumental data are not available, the design basis ground motion
can be evaluated on the basis of the maximum historical intensity in the region. For this
evaluation, the following procedure might be applied for facilities in Class 2 or lower,
provided the site region shows a reasonable uniformity from the seismotectonic point of view:
A zone having a radius of minimum 100 kilometres from the site needs to be
considered. Larger radius of up to 200 km needs to be considered when data is
lacking and there is low seismicity.
Using available publications and catalogues the maximum observed intensity in this
area needs to be established and assumed at the site. The information needs to cover
as much historical data as possible. In any case it needs to be extended to at least 100
years [17].
29
To each seismic design level index, a minimum value of design free-field
acceleration for firm bearing strata (Ab) needs to be assigned as shown in Table VI. These
assigned values need to be compatible with the seismic provisions of several national building
design codes currently adopted by several States. However, intermediate design accelerations
can be assigned based on detailed analyses of the data and other considerations.
The conservatism of such evaluation is embedded in the assumption of a large area of

investigation, much larger than in the case of NPP siting [17]: the reduced investigation
campaign is then compensated by an enlargement of the area of interest and therefore in the
use of the highest seismicity values of the whole area for the site.
TABLE VI. ASSIGNED MINIMUM FREE-FIELD DESIGN ACCELERATIONS

Assigned Design Acceleration for
Range of Maximum Historical Seismic Design Intensity Level Firm Bearing Strata
Intensity, I max*1 Index Ab (g)
I max < VIII 1 0.1
VIII I max < IX 2 0.2
IX I max 3 0.4*2
*1
Modified Mercalli Intensity Scale
*2
For Seismic Design Level Index 3, intermediate design accelerations can be assigned between 0.2~0.4 based
on detailed analyses of the data and other considerations
4.2.3. Design response spectral shapes
In the absence of site specific design response spectra constructed from instrumental
or historical data, Figures 5 and 6 can be used as the normalized design response spectra
representing different design intensity levels and soil types (discussed in the following
chapter), to be used in conjunction with the pga values of Table VI. The numeric values are
given in Table VII, from [54].
TABLE VII. SEISMIC DESIGN INTENSITY LEVELS

Design intensity level A B C D
1&2 0.05/1 0.1/3 0.2/3 2/0.3
(Soil 1,2,3) 0.05/1 0.2/2.5 0.6/2.5 3/0.55
0.05/1 0.5/2.3 1.1/2.3 4/0.81
3 0.05/1 0.1/3 0.4/3 2.1/0.4
(Soil 1,2,3) 0.05/1 0.24/2.5 0.9/2.5 3.5/0.7
0.05/1 0.5/2.3 1.6/2.3 4/0.8
These spectral shapes are given for a 5% damping ratio () and are valid as horizontal
components. In order to get the spectral shape for other damping ratio zeta, the value of
acceleration at points B and C have to be multiplied by the Dd coefficient defined as:
Dd=1.5/(1.0+10*) for 2%< < 20%. In this formula, needs to be set in unit, not in per cent.
For other period T, the acceleration values have to be linearly interpolated within the
bilogarithmic domain.
30
B C
A
FIG. 5. Normalized design response spectra for seismic design intensity level. Indexes 1
and 2.
B C
FIG. 6. Normalized design response spectra for design intensity level 3.
31
a(T)=a(Ti)*(a(Ti+1)/a(Ti)^(Log(T/Ti)/Log(Ti+1/Ti))
Where Ti and Ti+1 are the periods points A to D closest to T. Ti=< T=<Ti+1
For periods less than TA, a(T)=1 for any period and damping ratio.
For periods more than TD, a(T)=a(TD)*(TD/T)^2 (constant spectral displacement in

this period range).
Both horizontal components of earthquake need to be taken equal to the above

defined spectrum.
The vertical component of accelerations needs to be at least 50 % of the horizontal

component.
The earthquake is supposed to have an isotropic probability of occurrence with

respect to the 3 space directions. This means that any combination of the 3 maximum
components of the form alphaX*SX+ alphaY*SY+ alphaZ*SZ with
alphaX^2+alphaY^2+alphaZ^2=1 is equally probable.
In light of recent observed data that exhibits larger than expected peak accelerations
and long duration pulses, if and when a particular site, as identified by appropriate
seismological and geological investigations, is (a) within 10 km of a fault and (b) capable of
generating magnitude 77.5 earthquakes, then steps need to be taken to appropriately increase
the design acceleration level for firm bearing strata, Ab. [4042].
Except for Type 1 soil sites, a two step process needs to be carried out on the seismic
input for the evaluation of the input at the foundation level: a convolution from the bedrock to
the free field and a deconvolution from the free field to the foundation level. Simple models
may be used based on mono-dimensional wave propagation and equivalent linear methods.
However, large modifications of the ground motion at the bedrock needs to be carefully
justified by parametric studies. The use of the free field motion at the foundation level is a
conservative practice and is usually acceptable.
4.3. DESIGN BASIS FOR AIRCRAFT CRASH
4.3.1. General
The aircraft crash scenario (ACC) at a site of a NFOP needs to be generally excluded
through probabilistic considerations and suitable siting procedures. However there are cases
when such a scenario is included in the design, namely:
when a NFOP is located at a site where the ACC cannot be ruled out, either for
strategic reasons or because it has to share the site with other facilities;
when some generic degree of protection is requested to the facility against external
missiles associated to strong wind or even to sabotage.
The effect of an aircraft crash on a NFOP needs to be analysed in order to identify the
areas to be protected.
32
Evaluation of the effects of an aircraft crash needs in general to consider:
global bending and shear effects on the affected structures (overall missile effects);
induced vibrations on structural members and safety related equipment (global
effects), particularly when safety related items are located close to the external
perimeter of the structures;
localized effects including penetration, perforation, scabbing and spalling, by primary
and secondary missiles (local effects);
the effects of fuel fires and possibly explosion on structural members as well as exposed
safety related equipment (ventilation system, containment openings, air baffles).
In general, NFOPs do not show a distributed resistance to a crash, being built with
steel and concrete frame structures; in fact only continuous concrete walls at the external
boundary of the building can provide some degree of protection.
Therefore the analysis needs to consider that the location of the impact is potentially
anywhere on the building (peripheral walls and roof) with any direction inwards into the
building. In principle all exposed structural elements need to be checked against all
mechanisms discussed above. Moreover, the definition of the impacting object is usually very
difficult and needs to consider a wide variety of aeroplanes, helicopters, missiles, etc.
However, such an analysis of the failure mechanisms is very complex and expensive
and it requires very high validation efforts. Therefore a simplified approach is described in the
following for a quick assessment of the consequences of ACC hazard on NFOPs.
4.3.2. Basic data for screening
Basic data for aircraft crash protection implies an extensive knowledge of air traffic
in the vicinity of the facility. The air traffic is conventionally grouped into three categories:
commercial aircrafts more than 5.7 t weight
general aircrafts less than 5.7 t weight (including helicopters)
military aircrafts
As effects from impacts of military aircraft are very severe, it is suggested to avoid
installing an NFOP near a military airfield. Basic studies need to be made in order to be able
to collect the following parameters:
presence of an airfield in the vicinity of the facility site.
probability of crashes per flight from statistical data in the whole region surrounding
the site
number of flight per year
mass and impact characteristics for the different possible aircraft. Impact
characteristics is typically of rigid body or hard shock typically for military jets or
soft shock for civil aviation aircraft.
speed of the aircraft when crashing.
From all these parameters, for each category, a probability of aircraft impact per unit
surface and per year can be derived. From its geometry, a virtual area of the facility can be
defined as the mean normal section of cylindrical projection of the facility under the different
crash angles. Finally, the probability of an aircraft crash on the facility can be evaluated as the
33
product of the probability of impact per unit surface and per year by the virtual surface of the
facility. The need for an aircraft crash protection depends on the probability of crash for each
category. If this probability is higher than 105 per year, the facility design needs to include the
corresponding aircraft crash assessment.
4.4. DESIGN BASIS FOR WIND
4.4.1. General
Effect of wind on constructions are generally grouped into two different types:
Effect of extreme values of wind pressure that can induce high forces in structures
with reference to their ultimate capacity.
Effect of moderate but quasi constant values that can induce resonance effects
through the mechanism of the Karman turbulence and then lower values for capacity
of the structure due to fatigue. This kind of effect depends on the shape of the
structure and generally appears only on tall and slender structures such as chimneys,
perpendicular to the wind direction.
It is important that site statistics of wind speed and direction are well known even in
the low speed range.
The effect of wind on structures mainly depends on the general layout of the
building. It can include over pressures on one side, and under pressure on the lateral, opposite
side and on the roof side. Such a distribution depends also on wind direction.
The structural analysis in relation to extreme wind needs to be carried out for all
structural elements with respect to all directions of wind and combination of possible over and
under pressure compatible with the building geometry.
The environment of a structure has also deep influence on wind effects. Mask effect
may result in reduced loads, group effects in increased and eventually resonant loads. Group
effects need to be accounted for while mask effect is usually neglected.
In general, only frame or slender structures are really sensitive to wind effects. Tall
chimneys either on buildings or on ground are sensitive to wind effects, even more than to
earthquake effects. In order to appreciate the possibility of resonance effects, the dynamic
behaviour of the chimney as well as the behaviour of aerodynamic flow of wind around it has
to be assessed in terms of Karman turbulence frequency.
4.4.2. Basic data
Typically, extreme value, normal (rather frequent) value and frequent value of wind
speed have to be known from a site region monitoring (at least 50 year data). If such a
database is available, the extreme value of wind speed for structure evaluation needs to be the
mean value plus 2.0 standard deviation. The normal (rather frequent) value needs to be the
mean value plus 1.0 standard deviation. For the study of frequent wind, the whole distribution
of wind speed vs. time needs to be accounted for, but mainly the speed(s) which induces
resonance effects. The measure of wind speed needs to be carried out at 10 m height above
ground.
34
In any case, the extreme value cannot be lower than the value provided by the
national building code for the same region.
Any national code giving over and under pressure distribution including variation
with the height of the considered point above ground and relative values with respect to the
building geometry, the wind speed and wind direction can be applied for that purpose,
provided site effects are evaluated.
In case site effects are expected to be significant, a monitoring system has to be

installed and operated for a significant time period for comparison with regional data.
Concerning rotational winds, statistics in the region need to be recovered. In lack of

that, an evaluation of global meteorological data has to be carried out to check if rotational
winds are likely. In case they are, standard procedures as for nuclear power plants needs to be
followed for an identification of a possible scenario.
4.5. DESIGN BASIS FOR SNOW
4.5.1. General
The effect of snow on construction is generally reduced to the weight of the

accumulation of snow on upper horizontal or sub-horizontal surfaces of structures. Generally
few structures are significantly sensitive to such loads. This is however the case for light
frame construction which are rather common among NFOPs.
Evaluation of this load is similar to that of extreme wind loads, but it is less complex
as far as no dynamic effect is expected.
4.5.2. Basic data
Typically extreme and normal (rather frequent) values of snow weight per unit
surface have to be known either from a building code and/or from an investigation in the site
region. If such a database is available (at least 100 years data), extreme value needs to be
taken equal to the mean value of non-zero values of the database plus 2.0 standard deviation,
and the normal value needs to be taken equal to the mean value of non-zero values of the
database plus 1.0 standard deviation. If a building code is used for design purpose, a site
monitoring system operating during the lifetime of facility needs to be installed at the site,
data periodically compared with the data used for design and emergency procedures set up for
the prompt snow removal in case of extreme precipitation.
Building codes may be useful to specify the non-uniform distribution of loads on the
different surfaces according to their geometrical shape, including immediate neighbouring
constructions if any.
4.6. DESIGN BASIS FOR EXTERNAL TEMPERATURE
4.6.1. General
In general, any temperature variation of a structure (steel or concrete) from its
construction temperature induces forces and displacements, as for construction materials
35
thermal dilatation coefficient is not zero. A mean temperature of construction of the structure
has to be identified, as real thermal effects have to be evaluated by difference to the
construction temperature state.
The temperature conditions to which a facility is subjected are due to:

temperature of the air inside the facility
temperature of the air outside the facility
temperature of the structure materials when directly exposed to sunlight, for parts of
the structure that are outside of the building.
For obvious operational reasons, generally the temperature of the air inside the
facility is maintained by the ventilation/isolation system within such limits allowing the
process and the workers to operate in correct conditions. These limits and the external
temperature statistics provide the basis for sizing of the ventilation/isolation system and for
the evaluation of the design basis temperature for all plant parts that are inside the isolation
system.
Other parts of the structure are subjected to the influence of the air temperature
outside the facility and to sunlight.
A different temperature may occur between the two faces of the same structure
element, when this element is in fact a part of isolation system or partitioning of the facility
into different temperature zones.
The structure temperature cannot be taken equal to the air temperature at the contact
because of the heat flow between the outside and the inside the facility, the presence of the
ventilation system and the thermal characteristics of the isolation system, structure, etc.
In the case of non prestressed concrete structures, cracking deeply affects the
structure thermal response. Only approximate approaches are in use by the engineering
community for the evaluation of such effects.
4.6.2. Basic data
Basic data for external temperature need to consist in the measure of two parameters:
air temperature outside under shelter
incident heat flow due to sunlight
This second parameter may have significant influence only if the structural elements
are directly exposed to it because in this case, although maximum heat flow cannot last more
than half a day, the corresponding temperature occurs with a very short time constant and
therefore it can reach the maximum effects as if maximum heat flow is applied constantly.
Conversely, where all the structure members are located inside an isolated and/or
ventilated zone, maximum effects need to be derived from the one day period of the heat flow.
These basic data need to be made over a duration of 1 year at least, and statistically
processed in order to derive design data. Extreme values need to be taken as the maximum
and minimum mean value over half a day in the whole year. Because the real structure
36
temperature is lower and delayed with respect to instant external temperature data, no
conservatism is to be accounted for via standard deviation. Conservatism can be put in the
evaluation of the structure temperature itself by assuming extreme values as permanent values
or at least over a one day period. The external temperature data need to be monitored after
construction of the facility in order to check periodically the validity of design data.
4.7. DESIGN BASIS FOR FLOOD
4.7.1. General
According to the site location, external flood can be caused by different phenomena:
Exceptional sea levels due to waves, tsunamis, storm surge, eventually in conjunction
with tide.
Exceptional river levels due to rain, snow melt and dam failure.
Because of its special consequences on nuclear criticality, presence of water in
nuclear facility has to be controlled with a high reliability level.
4.7.2. Basic data
Basic data for water level need to be derived from a study of site hydrogeological
environment, including all above mentioned phenomena, using either a probabilistic or a
deterministic approach. Typically, highest levels of water due to waves, wind, rain and snow
need to be analysed through a probabilistic approach while tsunamis, tide and dam failure
through a deterministic approach from the knowledge of their cause.
For frequent events, a simplified approach used in some Member States assumes the
historical maximum level (>50 years of measurements) multiplied by a factor of 1.5.
Changes foreseen in the site environment have to be considered if they increase the
mean and maximum water level (new dams, land use, etc.).
4.8. DESIGN BASIS FOR EXTERNAL EXPLOSION
4.8.1. General
External explosion hazard may be related to the presence of other plants in the
vicinity of the facility using or storing explosive materials such as gas, oil, chemical products
or transport of such materials. There are three main types of explosion sources:
clouds of explosive gas
tanks full of gas
solid explosive materials.
A complete study of surrounding industrial activity and transportation by road, river,

sea, train or pipe lines has to be made in order to identify chemical nature and quantity of the
transported substances, geographical location, frequency of occurrence of relevant accidents,
storage or transport conditions and eventually protection against explosions.
As far as the following approach presents significant safety margins, focalization

effects may be ignored.
37
4.8.2. Basic data
According to the above mentioned identification of potential explosion sources, the
design can follow the approach of an equivalent explosion of TNT, particularly if the source is
relatively far from the facility. For this purpose, two coefficients can be applied to the
identified mass of explosive material:
An equivalent TNT mass ratio is applied to the mass of explosive product and gives
the equivalent TNT mass as for its explosive effects. For hydrocarbons, this
coefficient is taken equal to 5. For other special products, this coefficient can reach
higher values up to around 160. Specialized literature needs to be investigated in that
matter, like [43].
A coefficient for gaseous conditions defining the ratio of the total mass present in the
storage or transport that participate in the explosion, depending on storage or
transport conditions. If no detailed estimation is made, for hydrocarbon this ratio can
be taken equal to 20%.
Then, with respect to this estimated TNT equivalent mass and distance from the
facility, and according to specialized literature, an overpressure triangular wave can
be postulated including instant overpressure value and duration. When applying the
pressure wave to the building, it is important to take into account reflection effects on
walls depending on the relative direction of walls and pressure wave propagation
(this coefficient can reach a value of 2) and dynamic effects due to the time
variability of the pressure (this coefficient can also reach a value of 2).
5. GEOLOGICAL AND GEOTECHNICAL INVESTIGATIONS
5.1. GENERALITIES
As in the previous chapters, this section provides suggestions on limited site
investigation campaigns, according to the general strategy defined in Sections 2 and 3.
Geological and geotechnical investigations at the site need to be performed with the
following objectives:
(1) the assessment of possible geological and/or geotechnical problems involving surface
rupture due to faulting, liquefaction, collapse and slope instability as a consequence
of an earthquake.
(2) the evaluation of the soil characteristics so that reasonable soil characterization can
be achieved in the modelling of soil structure interaction, mainly in seismic analysis.
For this, it is necessary to determine depth to bedrock, shear wave velocity variation
and average shear wave velocity for a depth of 25 m, at least.
(3) the evaluation of geotechnical parameters to be used in the design of the foundation
for static and dynamic loads and for radiological dispersion studies through
groundwater.
For facilities in Class 1, the basic reference publication [24] needs to be applied.
For facilities in Class 2 or lower, the simplified following approach might be applied.
The amount of geotechnical investigations to be performed needs to be based on the

extent of potential problems, the available data, the physical size of the facility and its
38
classification. It is suggested that the soil profile is physically identified (i.e. through drilling)
to a depth equal to at least one half of the maximum foundation dimension but not less than
25 m. The depth to firm bearing strata also needs to be determined using geological inference,
boreholes or geophysical methods.
The geotechnical investigations need to include determination of the bearing

capacity, shear wave velocity representative of the site and other soil parameters for
foundation and building design. These need to primarily involve borehole drillings in
sufficient number and to sufficient depth, depending on the soil conditions, but at least one
below every safety related building.
For competent rock sites where the rock formation continues to a sufficient depth,
drilling may not be necessary.
Uncertainties in the mechanical properties of the site materials need to be taken into
account through parametric studies, at least on the shear modulus value. One method is to
vary the shear modulus between the best estimate value times 1.5 and the best estimate value
divided by 1.5.
Table VIII provides a simplified characterization of soil types based on shear velocity
(average value of the upper layers), for use in the selection of the seismic input and in soil
structure interaction.
TABLE VIII. SOIL CHARACTERIZATION

Characterization SOIL TYPE
1 2 3
DESCRIPTION Firm Bearing Strata Other soil than those Alluvium ground which is
(thickness > 25 m) defined as 1 & 3, or with thicker than 25m
lower thickness
Shear Wave Velocity Vs
(m/sec) >1100 m/s 3001100 m/s 150300 m/s
For a more comprehensive treatment of the subject, the reader is referred to

[17, 22, 24].
In parallel with the foundation investigations and the use of available geological and
geotechnical data, studies need to be performed at the site to assess possible hazards resulting
in permanent soil deformations (including surface rupture, liquefaction, collapse, slope
instability). In the event that there are special features of the site such as unusual surface and
subsurface topography, then investigations need to be carried out to determine if there is
focusing of incoming seismic waves.
The walls of high risk facilities need to not be in direct contact with the soil mass
because the evaluation of dynamic soil pressures to embedded structures can be very complex
and may require a substantial effort. Unless it can be shown that safety is not compromised by
having the EEC1 structures in contact with soil, their walls need to be separated from the soil
by EEC2 structures. If this is not practical, then dynamic soil data, which are outside the scope
of this report, are required, in particular for EEC1 buried piping. For applicable methods the
reader is referred to [24].
39
5.2. LIQUEFACTION POTENTIAL
Potential consequences of any liquefaction and soil strength loss need to be assessed,
including estimation of differential settlement, lateral movement or reduction in foundation
soil bearing capacity. Mitigation measures need to be considered in the design of the structure
and can include, but are not limited to, ground stabilization, selection of appropriate
foundation type and depths, selection of appropriate structural systems to accommodate
anticipated displacement or any combination of these measures.
The potential for liquefaction and soil capacity loss needs to be evaluated for site
peak ground accelerations, magnitudes, and source characteristics consistent with the design
earthquake ground motions. The value of peak ground acceleration to be used for liquefaction
studies needs to be evaluated on the base of a site specific study. In the absence of such a
study, peak ground acceleration may be assumed equal to SDS/2.5 where SDS is the spectral
peak acceleration estimated at the site.
Two procedures for evaluation of liquefaction hazard are included in Appendix I. In

addition to these, the procedures recommended in Ref. [41] can also be used. The assessment
of liquefaction potential by any one of the methods needs to yield similar conclusions.
6. BUILDING DESIGN
6.1. SEISMIC DESIGN
6.1.1. Layout considerations
In order to resist earthquakes, the general shape of buildings needs to be as simple

and regular as possible, as well in the horizontal plane as in elevation. This simplicity involves
the volumes, the regularity and symmetry of mass and stiffness distribution, and the
distribution and continuity of any elements called to provide seismic capacity to the structure.
Centroid of masses of floors and centroid of stiffness at each level need to be as

close as possible.
With respect to the earthquake intensity, the overall height-to-width ratio of the
building needs to be limited in order to avoid potential problems related to foundations
behaviour and soil capacity due to uplift.
A foundation raft is suggested. It is suggestedthat the foundation elements for

structures important to safety be located at the same elevation and on the same foundation
type of soil.
Inverted pendulum structures are not allowed for EEC1 structures.
Structural detailing needs to avoid brittle parts, even for those elements that are not
assumed to take significant part of overall forces.
Doors and other openings in floors and walls need to be sized and located in such
way they do not introduce significant discontinuity in the resistant scheme of the structure.
40
Precast panels (or other prefabricated elements) need to be connected in such a way
that they behave as an integral unit during an earthquake.
Some non-structural elements may affect the dynamic behaviour of the structure and
its capacity. It is typically the case of masonry filling in framed reinforced concrete structures
which can lead to shear damage (and rupture) of the so called short column configuration.
In the case of design of new buildings, this solution needs to be avoided. In the case
of re-evaluation of existing buildings, their capacity needs to be assessed together with their
potential effects on the dynamic behaviour of the structure and on the forces in the structural
elements.
6.1.2. Design classes for structures
With reference to the safety classification described above and to the single
parameter control approach, the minimum requirements imposed to structures, during and
after an earthquake, can be summarized as in the following:
For Design Class 1 (DC1) structures, full functionality, is required. No cumulative
damage is allowed in order to withstand next earthquake exactly as the first one.
This implies a quasi-elastic behaviour of the whole structure in an event of the
design basis earthquake (DBE). This behaviour is referred to the global structural
response, ignoring the unavoidable cracking of concrete before and during the
earthquake. This condition is considered sufficient but not necessary to comply
with the safety requirements.
For Design Class 2 (DC2) structures, capability of supporting safety related
components, equipment and systems needs to be maintained. This requires limited
inelastic deformation and is guaranteed by lower ductility coefficients than for
DC3 structures in the event of the DBE. The limited values of the coefficients are
defined in Table IX.
For Design Class 3 (DC3) structures, non-collapse allowing for inelastic behaviour
is required. This is guaranteed by ductility coefficients given in Table IX.
For Design Class 4 (DC4) structures, conventional design standards for industrial
buildings can be applied.
In general, the allowable stresses need to be the same for the different categories of
design class. The difference between design classes needs to be only in terms of different
ductility values associated to the internal forces.
The use of ductility coefficients is associated to a careful application of special
detailing provisions defined in national seismic codes. For ductility values higher than 2, such
provisions need to be put in place and the ductility values justified.
6.1.3. Geometric modelling

Analysis methods for building design have to be adapted to the general layout of the
building. In general, the more complicated is the layout, the higher conservatism needs to be
inserted to compensate for calculation uncertainties.
The main issues which affect the modelling strategy are:
41
the possibility to decompose the building into regions that have consistent
displacements (masses) for each direction of earthquake;
the possibility to easily evaluate the stiffness of the connections between these
regions and to take account of their connectivity order and topology. Inside a single
region, there needs to be a strong connectivity in order to keep displacements
compatible.
TABLE IX. DUCTILITY COEFFICIENT .
DC2 DC3
Type of Structural System
Coefficient Coefficient

SHEAR WALL SYSTEM:
A structural system bearing walls providing support
for all, or major portions of the vertical loads.
Seismic force resistance is provided by
Reinforced concrete walls 2.5 4
Reinforced masonry wall 1.5 2.5
MOMENT RESISTING FRAME

Seismic resistance is provided by Moment Frames
capable of resisting the total prescribed forces. Code
detailing rules are very carefully applied.
Steel frame 3 4.5
Reinforced concrete 2.5 4
DUAL SYSTEM
Mixing of the above systems. max.2.5 max.4.0
INVERTED PENDULUM STRUCTURES:
Structures where the framing resisting the total
prescribed seismic forces acts essentially as isolated 1.0 max.1.2
cantilevers and provides support for vertical load
Between two connected regions, the scheme of force transfer and associated
deformations need to be split into subsets that act in parallel according to the following
categories:
bending beam behaviour;
shear beam behaviour;
compression/tension beam behaviour;
torsion beam behaviour;
composite beam behaviour,
frame behaviour.
In some cases two or more model typologies can be connected to a single mass
representative of the foundation.
42
Such hypothesis of decomposition has to be validated with appropriate sensitivity
analysis.
Concerning the modelling of any region, a stick model can be used provided the
following conditions are met:
The structure shows a global beam type behaviour.
The structure has floors which can be assumed to be rigid in their plane,
compatible with a 6 degree of freedom mass located at its centroid.
Local deformation of walls between floors is assumed to be negligible.
It is assumed that between two consecutive masses, the overall stiffness can be
modelled by equivalent material properties assigned to the equivalent sections.
In case a stick model is not a suitable approach to the modelling of the region, 2D or
3D models need to be developed.
6.1.4. Soil structure interaction
The dynamic response of the coupled model of soil and structure needs to be
evaluated taking into account the behaviour of the soil region around the foundation and of
the seismic waves propagating into it. If SSI effects are deemed to cause beneficial effects,
then the analyses may be carried out without incorporating SSI.
When SSI is considered, the relative behaviour of unbounded soil and foundation
submitted to seismic waves needs to be analysed first and modelled with a set of springs and
dashpots. In a second step, the structure supported on this spring dashpot system needs to be
computed. Appendix II provides the formulas for calculating the stiffness of the equivalent
spring and the damping values to be assigned to a soil spring dashpot system applied at the
base of the structure model for rectangular/circular foundation shape.
Soil modelling might strongly affect the response of structures. Therefore a

variation in soil properties needs to be considered for sensitivity purposes.
It is suggested to take into account the uncertainty on soil modulus, as suggested in

Chapter 5, by considering 3 different values for the soil shear modulus (G). That range is
expected to envelope also the effect of the presence of other buildings of comparable
dynamic behaviour in the neighbourhood.
The design of structural elements needs to be carried out with reference to the
envelope of all the calculated results.
In case of important stratification with a soft upper layer, also the radiation
damping needs to be subjected to sensitivity analysis.
Special attention needs to be paid to the case of the design of a small building
beside a large one: in such a case, the soilstructure interaction of the large building can
determine the motion imposed at the base of the small one.
43
6.1.5. Non-structural loads
Non-structural loads need to be accounted for in the calculation of the seismic

response. These are:
Itemized large equipment and their content (i.e. known individually).
Non-itemized numerous small equipment generally represented as a uniformly
distributed permanent load for each floor area. Note that some are suspended under
floors or on walls (cable trays, ventilation ducts, etc.).
Live operating loads.
Other simultaneous loads as external water or earth static and dynamic pressures.
Static and dynamic thrust of stored material, depending on its storage conditions,
solid, liquid, powdery, etc.
All itemized and non-itemized equipment loads and their content need to be
considered.
Live operating loads have to be considered according to their probability of

occurrence. For example water in a storage pool and the content of the storage itself need to
be accounted for without reduction. For live loads, coefficients taking account for the
probable non simultaneousness of maximum values of the live load in each area of the
building may be considered. Table X suggests some coefficients for live load inclusion into
floor equivalent non structural masses.
The most unfavourable mass configuration needs to be considered in the analysis. In

some cases, large masses (cranes) may have different location in the building, and therefore
the building has to be designed for any possible position of displaceable items. For heavy
storage the designer has to account for several mass configurations of the building.
TABLE X. LIVE LOAD COEFFICIENT
Use of Floor Live Load

Coefficient n
Heavy weight components such as 0.50

large baggage in the wide space
Medium weight components in medium size space 0.30
Light weight components such as desks, books and 0.25

lockers in small size space.
Note:
The weight of the water in the reactor pool needs to be considered as dead load. The weight of permanently
installed equipment and distribution systems at maximum normal operating weight needs to also be considered as
dead load.
44
6.1.6. Spectral analysis
In this section, a general description of spectral analysis is provided. The spectral

analysis can be used to evaluate the dynamic response of a structure which is represented by
a stick model.
In a spectral analysis the calculation steps are as follow:

(1) computation of the eigenmodes (normalized displacements, forces, reactions, strains,
stresses etc. at the different locations in the model) of the structure and associated
frequencies and damping ratios;
(2) computation of modal participation factors for each earthquake direction;
(3) computation of the spectral accelerations for each mode and earthquake direction,
depending on seismic spectra in each direction and modal frequencies and damping
ratios;
(4) computation of the spectral response (displacements, forces, reactions, strains,
stresses etc. at the different locations in the model) of each mode for each earthquake
direction;
(5) combination of spectral response of modes for an earthquake direction, deriving the
overall response of the structure to this earthquake direction;
(6) combination of the three responses to each earthquake direction, deriving the overall
response.
It is suggested that an additional horizontal load eccentricity equal to 5% of the

building horizontal dimension in any direction is added on to account for torsional moments
coming from unmodelled stiffness or mass eccentricity.
In Step 1, the modal damping ratio is calculated by averaging the damping values
of each part of the whole system, using the following procedure:
The structure is divided into p parts, each one having a unique damping ratio
equals to j. K j is the stiffness matrix reduced to the j-th part. i is the i-th modal vector,
reduced to the j-th part.
Averaged modal damping ratio is then given by:

p
E j
i
j
j =1
i = p
E i
j
j =1
where,
E ij = i T K j i
The value of j should not exceed 20 %.
In Step 5, for each earthquake direction, a choice of the number of modes retained
for response analysis has to be made according to the following criteria: the sum of effective
masses of all modes retained in each direction need to be greater than 90% of the total mass
45
of the building. If this criterion is not met, some supplementary modes have to be accounted
for if their frequency is less than the cut-off frequency (fc) of the spectra, or a static
correction (missing mass procedure) has to be used.
The modes need to then be combined according to the Complete Quadratic

Combination (CQC) [44], for example or similar procedures.
Alternatively, if it is shown that the eigenfrequencies of the model are separated

enough, typically more than 10% difference between two consecutive frequencies of
significant modes in the considered direction, the Square Root of Sum of Squares (SRSS)
rule can be applied, as explained in the following.
The maximum response value, Umax, can be calculated according to the following
equation (given for the acceleration):
2
Umax =
j
j j S Dj
where, j j = j-th participation function obtained by eigenvalue analysis
j
: the summation for j which takes the value from 1 to the maximum number
of the superposed mode including possibly the static correction (or missing
mass mode).
SDj = Spectral acceleration of the mode in the considered direction
If a missing mass mode is used, it needs to be considered as having a frequency

equal to fc and a damping ratio equal to 5% for the calculation of the cross coefficient with
other modes in the use of CQC procedure.
Step 5 needs to be applied for each direction. Maximum scalar values needed for
the structural design are evaluated: displacements, accelerations, forces, reactions, strains
and stresses.
In Step 6, the overall response to the three earthquake directions is obtained by

simple quadratic combination of each earthquake direction response. Alternatively, it is
possible to use the Newmark rule:
( 1.0*Gmaxx+0.4*Gmaxy+0.4*Gmaxz)
Gmaxxyz=max (0.4*Gmaxx+1.0*Gmaxy+0.4*Gmaxz)
(0.4*Gmaxx+0.4*Gmaxy+1.0*Gmaxz)
Any variable of interest needs to be first calculated in each mode and direction, then
combined for each direction through Step 5 and at last combined for the three directions
through Step 6.
When checking the structure, maximum scalar values need to be considered with the
most unfavourable sign.
46
The spectral analysis is also the basis for floor response spectra calculation.
However, in case the foundation uplift due to earthquake is more than 30% of the foundation
area, its influence on the floor response spectra due to the non linear behaviour of the
foundation cannot be ignored and specific computation methods have to be applied for their
generation.
The horizontal design response spectra defined in Sections 4 and 5 may be

modified by ductility and damping values to derive the response acceleration for each mode,
SDj in the following way:
SDj = ( ag . Sj . Dj) / j
For Tj 0.1s : j = 1
0.1s < Tj 0.5s : j = (2 - 1)1/2
0.5s < Tj : j =
where, is defined in Table IX, and

ag = design acceleration applicable at the foundation level
Sj = Standard Response Factor for the j-th mode period, Tj. The factor is
determined from 4.2.3
Dj = coefficient related to the damping ratio j for the j-th mode (modal damping)
to be used as indicated at 4.2.3.
The structural damping value is given in Table XI, and the composite modal
damping ratio for the soilstructure system needs to be calculated using the formulas given
in the Appendix II.3 Modal damping needs to be limited to 20%.
TABLE XI. STRUCTURAL DAMPING VALUE
Type of Structure Percentage of Critical Damping (%)
Bolted or riveted steel structure 3
Welded steel structure 2
Reinforced concrete 5
or Steel concrete structure
Prestressed concrete 4
Note: These values are rather conservative because of the variety of structural details. To select more realistic
values to be used with more refined analysis methodologies, appropriate references need to be used.
6.1.7. Simplified dynamic approach

This simplified dynamic analysis method is applicable to all categories of structures
with the condition that they are relatively regular structures, with two perpendicular vertical
symmetry plans and essentially uniform mass and stiffness distribution.
47
The main limitations of the method are listed here below, with reference to the stick
model structures. Such list needs to be used also to verify the applicability of the method to
the structure of interest.
The 6 degrees of freedom (DOF) mass representation is limited to 3 translational
degrees of freedom in any floor and decomposed in two horizontal and one vertical
degree of freedom problems.
Dynamic load distribution is defined according to a single mode shape with a
frequency given by a formula depending on the main parameters of the building
layout only.
The vertical motion is neglected.
Soil structure interaction effects are neglected.
The procedure for the calculation of horizontal loads is as follows:

Determination of floor masses, with the contribution of equivalent non structural
masses.
Determination of the location of the centre of gravity for the whole building.
Determination of the building frequency in the considered direction through the
following formula:
Tb = HB DB
where,
H B = height of the building with respect to the foundation level (metres)
D B = width of the building in the direction parallel to the horizontal excitation
(metres)
= 0.10 for rigid reinforced concrete structures (walls)
0.12 for ordinary reinforced concrete structures (frames).
Determination of the acceleration distribution in the height of the building (each floor
level), as follows:
Fi = ag . CDi(hx) . Wi
where,
ag = design acceleration applicable at the ground surface level as defined in
previous Sections
Wi = the total weight of the building at the i-th floor augmented with the
contribution of non structural masses
CDi(hx) = function related to the building characteristics and defined as :
CDi = ( D1 . DD . D3(hx)) / b
where, b is defined as in 6.1.6
D1 = Standard Response Factor for the fundamental period, Tb. The factor is
determined from Figures 5 and 6 in 4.2.3
DD = coefficient as defined in 4.2.3 related to the damping ratio, , to be
determined from Table XI
48
D3(hx) is the load distribution function throughout the height of the building and is
defined as:
D3(hx) = 1 + 0.5 (hx/Hb) (not smaller than 0.5 when hx<0)
where, hx = height of the point or storey under consideration with respect to the centre
of masses of the whole building (+ sign for above, sign for down)
Hb = height of the centre of masses of the whole building with respect to the
foundation level
If the fundamental period of the structure, Tb is shorter than 0.3 s, the product D1 DD
should not be less than 1.5.
In the vertical direction, a constant acceleration equal to the half of ag can be applied
statically.
The simplified dynamic analysis may be carried out separately for each input
direction. Responses resulting from the two horizontal directions need to be combined using
one of the following methods.
(1) Combination of torsion and orthogonal stresses

(i) Distribution of the torsional moments is carried out according to the
torsional rigidities.
(ii) At the intersection points of X,Y (column, etc.) Especially at the corner
of the building, simultaneous stress in x and y directions is estimated
by combining stresses in each direction using SRSS or CQC method.
(2) Considering the orthogonal effect of resultant internal forces due to defined seismic
input motions, results can be combined as follows:
In the x direction:
gravity loads +/ 100 % of forces due to x direction motions +/ 40 % of
forces due to y direction motions, and
In the y direction:
gravity loads +/ 40 % of forces due to x direction motions +/ 100 % of
motions due to y direction motions.
The +/ sign indicates that the design is performed for maximum member strength.
Responses resulting from horizontal and vertical direction need to be vectorially

combined.
6.1.8. Earthquake induced element forces

The main goal of this section is to define an additional parameter to be applied to
element forces to account for the uncertainties included in the proposed methodologies, in
addition to what has been already made with respect to the category of the structure through
the use of ductility.
This parameter is related to:

the complexity of the building;
the uncertainty due to calculation methodologies;
the objective (design or re-evaluation) of the calculations.
49
Only the earthquake induced element forces, including sloshing effects, underground
water and soil pressure dynamic increments, are affected by this parameter.
The following Table XII provides the coefficient that needs to be used, according to
the selected scenario, as multipliers of the earthquake induced element forces before
combination with other loads and comparison with the element capacities.
TABLE XII. CORRECTION FACTOR FOR SEISMIC RESULTS
Value for
design
Very regular layout Medium regular layout Irregular layout
Value for
re-evaluation
1.0 1.1 1.2

3D analysis method
1.0 1.0 1.0
1.05 1.2 1.3

Stick model
1.05 1.1 1.3
Simplified dynamic 1.1

Not used Not used
analysis method
1.1
6.1.9. Load combination and acceptance criteria
The earthquake loads need to be combined with normal operating loads. In general,
the definition of combinations needs to comply with [22]. Attention needs to be paid to the
combination of static normal operation loads and seismic loads resulting from spectral
analysis, where the signs of different components are lost. If the response in each direction
is governed by the first mode, it is acceptable to consider the relative signs of the first mode.
Thermal loads can generally be ignored in the load combination with seismic
effects for reinforced concrete structures.
The general loading combination for earthquake can be:
1.0*normal loads + 1.0*seismic loads
The normal loads need to take the buoyancy (and lateral pressure for external
walls) due to underground water if any.
The coefficient 1.0 applied to seismic loads does not depend on the design class of
the building because the design class of the building is already taken into account when
applying the ductility coefficients, as stated in para. 6.1.2.
50
In general, the acceptance criteria need to be the same for the different design
classes. They can be taken from any national building code for accidental conditions. The
difference between design classes needs to affect only different ductility values associated
to the internal forces according to para.6.1.2.
National building codes or standards for materials such as concrete and steel
generally specify an increase in allowable stresses for extreme loads such as earthquakes.
These increased values may be used as the elastic limits for the earthquake resistant
design class 1 to 3 also. If such an increase is not specified in the national codes or
standards, then an increase of 33% may be taken in the allowable stresses for concrete
(compressive and shear), structural steel (tensile, compressive and shear) as well as
foundation soil (bearing capacity). However, applied buckling loads need to not exceed 0.75
the critical buckling load.
6.2. AIRCRAFT CRASH
6.2.1. Protection strategies

The protection of NFOPs against ACC needs to take advantage of the generic
characteristics of the NFOP layout (usually very dispersed) and of the low inventory of
radioactive material they store at the same time.
In this framework, the broadest application of the segregation (physical separation)
principle is suggested together with a large use of isolation devices (valves, barriers, etc.) for
the confinement of the damage to the most limited area.
However in many NFOPs there are unique systems/structures, safety related, which
collects most of the radioactive inventory of the facility (centrifuges, reactors, reaction
vessels, etc.). In these cases a protection of this specific system/structure is due as part of the
prevention approach.
Particular emphasis needs to be given to the protection against the effects of fire,
ACC induced, which could spread to other areas of the facilities, impairing personnel
evacuation, operator actions and the confinement of the damage to the hit area.
6.2.2. Local design
Typically, the local design for an aircraft crash relies on the strength of the external
panels of the reinforced concrete structure that is subjected directly to the impact.
Depending on which is the safety function of the panels, the acceptance criteria can
be different, from the tightness that requires an elastic reversible behaviour and possibly leads
to huge protections, to a simple no-penetration criteria which allows significant plastic
excursions and possibly later repairing.
There are two sets of data that could be used in order to make an evaluation of the
behaviour of these local elements:
estimation of mass, stiffness, impact area, fuel amount and speed of the impacting
aircraft.
simplified equivalent estimation of loadtime function and impact area.
51
According to the available data at the site, simplified design formulas for the
protecting structures can be applied from the bibliography (see for example Ref. [45]) for an
estimation of the required stiffness. In particular most of the analytical formulas provided in
the bibliography have been validated in test programs supported by the nuclear industry (e.g.
EDF, Bechtel, Sandia, etc.) and therefore they deem very appropriate for the range of
application requested to NFOPs. The assumption of a rigid missile is considered too
conservative and the relevant formulas need to be disregarded for the present application.
Without an explicit analysis, very expensive for the reasons described above, the
following generic criteria may be applied for the design of a concrete barrier: they represent
the synthesis of many simulations, both numerical and by test, and allow a reasonable sizing
of the structures with a limited amount of conservatism.
The impacting object may be assumed with characteristics similar to one of the
heaviest engine of an aircraft (proved to be the worst scenario for the evaluation of
local effects): around 2000 kg travelling at 215 m/s, deformable, impacting on an
area of 1.5 square metres in perpendicular direction to the external surface
The concrete thickness may be chosen in the following range: 90 cm to stop
perforation of the missile, allowing significant scabbing on the internal surface,
160 cm to avoid most of the scabbing, allowing only a limited amount of penetration.
Reinforcement quantities. Longitudinal reinforcement proved to be not essential for
prevention of local effects: however, it has to be around 0.4 % of the concrete
volume to provide a generic capacity. Shear reinforcement (stirrups) proved to play
an important role in the prevention of shear punching failures: an amount of 1.5% in
volume can limit the plant size of the punching cone to 2 square metres.
6.2.3. Global design

The evaluation of the global dynamic behaviour of the whole building would require
a dynamic analysis of a numeric model subjected to impact forces on any potential impact
location. A simplified approach may again rely on conservative assumptions, limiting the
analysis to the structures supporting the protecting shields. The force transfer needs to
consider the deformation of the impacted area through traditional analytical approaches (see
for example [45]).

For local analysis, the load combination for local stress/strain analysis may be:
1.0 Normal loads (dead + live) + 1.0 aircraft crash loads
The acceptance criteria depend on the function required to each structural element.
For local design, if the only function of the element is to stop the aircraft and maintain the
global stability of the building, it may be designed with plastic excursions of reinforced bars
reaching = 2% deformation.
If the structural element supports equipment that needs to guarantee a safety function,
the plastic excursions need to be limited to = 1% deformation. In both previous cases, the
acceptance criteria for concrete in compression need to be =0.35%.
52
If the element has a tightness function, no plastic excursion can be allowed, and
elastic behaviour has to be guaranteed. In this case, however, it is more convenient to design a
shielding structure able to protect the safety related buildings.
For the global analyses, the load combination for global stress/strain analysis may
be taken as:
1.0 Normal loads + aircraft crash loads
Where depends on the design class of the building according to the following Table.
DC 1 2 3 4
1.0 1.0 1.0 1.0
The coefficient is taken at 1.0 for all the design classes as the screening value for
the design basis set up at chapter 2 is assumed at 105.
The acceptance criteria may be those of any national building code for accidental
conditions.
Concerning the fuel effects, a dedicated analysis needs to be carried out to prevent
fuel access into the facility. Then, the same criteria described above for explosion and fire
may be applied.
6.3. WIND
6.3.1. Load combination and acceptance criteria for extreme wind

Extreme wind loads need to be accounted for simultaneously to the normal loads in a
linear analysis according to the following load combination:
1.0 normal loads + wind loads
in which depends on the design class of the structure according to the following Table
DC 1 2 3 4
1.5 1.3 1.2 1.0
Note: the a coefficient is applied to wind loads, not to wind speed.
Acceptance criteria can be taken from any building code (reinforced concrete, steel
construction, etc.) where extreme wind or other accidental loads are accounted for.
6.3.2. Load combination and acceptance criteria for frequent wind
Frequent wind effects depend on the dynamic behaviour of the structure, frequency,
damping, geometry, the surface state of its external surface, the use of antivortex device and
the wind speed spectra, the life time of the structure, the wind direction if geometry is not
horizontally isotropic (like a chimney for instance). Their evaluation allows the calculation of
cumulative fatigue: stress amplitude over the number of cycles as a function of wind speed
53
and summation of the fatigue for the number of cycles for each wind speed at its time of
occurrence, with summation for all wind speeds. The acceptance criteria need to be selected
according to standard fatigue curves of the structure material.
In general, reinforced concrete chimneys show a rather low sensitivity to fatigue

because of the compression due to dead load, which prevent the reinforcing bars from
tensioning under moderate wind speed. Steel chimneys are much more sensitive to these
effects. Special antivortex devices can be put in place in order to break the Karman
turbulence: the efficiency of such devices needs to be validated by aerodynamic studies before
they are accounted for.
In the calculation of stress range, depending on the design class of the structure, the
frequent wind effects need to be multiplied by an coefficient according to the following
Table:
Design class: 1 2 3 4
1.5 1.3 1.2 1.0
Load combination for stress analysis needs to be:
1.0 Normal loads + wind loads
These wind loads need to take into account resonance effects in the direction
perpendicular to wind direction due to Karman turbulence. Detailed scientific literature has to
be followed in order to calculate this effect. When damping devices are used to reduce these
effects, an in situ validation of their efficiency needs to be carried out before operation.
For the calculation of these effects, if no special damping device is used, damping
ratio needs to be taken equal to 2% for steel construction and 3% for concrete construction.
6.4. SNOW
Extreme snow effects need to be taken into account together with the normal loads
according to the following combination:
1.0*normal loads + *snow loads
with a given by the following Table:
Design class: 1 2 3 4
1.5 1.3 1.2 1.0
The acceptance criteria may be taken from any classical building code for accidental
situations.
Concerning normal snow load (rather frequent), it needs to be considered

simultaneously with normal loads and normal (rather frequent) wind.
54
The load combination may be:
1.0*normal loads + 1.5*(normal wind + normal snow loads)
not depending on the design class of the structure. Capacity needs to be checked
according to acceptance criteria of building codes for normal situations.
6.5. EXTERNAL TEMPERATURE
6.5.1. Layout consideration
It is suggested generally as a good design practice that the whole structure is located
inside an isolation system, in order to reduce the induced thermal stress field.

The thermal loads can be analysed with the hypothesis of a linear behaviour of the
structure for all construction materials.
However, the reinforced concrete case is a special one because of cracking. In this
case, a 0.5 coefficient may be applied on the results of the linear analysis and no further
analysis is necessary if results are correct with respect to acceptance criteria. If not, more
detailed approaches based on theoretical crack width, pattern, and stress redistribution can be
used, with more realistic acceptance criteria.
Since thermal loads is a cyclic one, with one year period, provisions need to be taken
in order to avoid cumulative plastic strain in some localized zone of the structure, whether it is
a steel structure or a concrete structure. For concrete structure reinforcement needs to be
designed to avoid concentrated cracking.
Load combination for extreme thermal loads needs to be:
1.0*normal loads + *thermal loads
Thermal loads have to be evaluated by difference between extreme and construction

temperature conditions.
depends on the design class according to the following Table:
DC 1 2 3 4
Linear approach, reinforced concrete 0.7 0.6 0.5 0.5
Linear approach, steel or prestressed concrete 1.4 1.2 1.0 1.0
Non-linear approach, reinforced concrete 1.2 1.0 1.0 1.0
The acceptance criteria can be taken from any construction code for accidental
conditions in case of linear approach and according to more realistic criteria for the non-linear
approach.
55
6.6. EXTERNAL FLOOD
All items that are important to safety need to be placed at an elevation higher than the
design water level at the site, after consideration of all the protecting measures. All openings
need to be checked: sufficient margin needs to avoid any water inlet into the building.
All external walls need to be designed in order to withstand the external water
pressure on top of the earth pressure. As buoyancy pressure on the foundation may affect the
stability of the building in case of an earthquake, it may be necessary to install a permanent
drainage system.

External flood loads can be used in a static linear analysis of the building that can
consider the local bending of walls due to water pressure and its dynamic effects. If there is a
possibility that flood carries sand or other solid materials, a suitable value for density needs to
be considered.
If the flood is expected to occur with a strong flow, dynamic effects need to be
considered.
Load combination for external flood needs to be
1.0 Normal loads + water pressure loads
with according to the following Table depending on the design class of the building:
DC 1 2 3 4
1.5 1.3 1.2 1.0
Acceptance criteria need to be:

(1) the static stability of the whole building against horizontal sliding displacement and
overturning, taking account of buoyancy.
(2) capacity criteria for accidental loads, taken from any conventional reinforced
concrete building code.
(3) Low cracking criteria for some water tightness if there is no tightness membrane or
similar device. Such criteria may be taken from standard building codes for
reinforced concrete.
(4) In case a water tightness membrane is used, its efficiency has to be proven by test,
with special care to joints between buildings, underground junctions etc.
6.7. EXTERNAL EXPLOSION
Generally, masonry on steel or concrete frame are not efficient against explosions.
Continuous concrete walls and floors are suggested for that purpose.
56
It is possible to put an independent protection between potential locations of
explosion and the facility to be protected. In such a case, the protection effect has to be
validated either by a specific study or by reference to previous experience.
It needs to be noticed that the pressure wave can penetrate inside the facility through
doors or air ducts for instance. This can reach directly safety functions located inside the
building. Such cases have to be investigated and safety functions assessed with respect to that
conditions.
If the analysis is linear, the load combination for an accidental explosion needs to be:
1.0 normal loads + explosion loads
with alpha depending on the design class of the building according to the following Table:
DC 1 2 3 4
1.0 0.9 0.9 0.8
In this case the acceptance criteria may be those of a conventional building code
under accidental loads.
If the analysis is non-linear, = 1.0 for all design classes, but the acceptance criteria
for concrete needs to be selected according to the following Table (max strain in the
reinforcement):
DC 1 2 3 4
(%) 1.0 1.5 1.5 2.0
7. EQUIPMENT AND PIPING DESIGN
7.1. INTRODUCTION
All equipment, components and distribution systems need to be designed to
guarantee their safety function during and after design basis external events. Their capacity
may be demonstrated by one of the following methods:
(1) Analytical methods, simplified, according to this report.
(2) Other analytical methods, as specified in the references.
(3) Numerical analysis using adequate software
(4) Dynamic test.
(5) Pseudo-dynamic test.
(6) Static test.
In this section, only analytical simplified methods (No. 2 above) for the assessment
of structural capacity of equipment, components and piping (with their anchoring) are
described for the case of earthquake and aircraft crash. For other external events, reference is
made to the methodologies set up for the nuclear power plants, as no significant simplification
is foreseen.
57
7.2. SEISMIC DESIGN
7.2.1. General methodology for seismic force evaluation
The seismic coefficient CDi from the simplified dynamic approach (see previous
chapters) may be used to evaluate the static equivalent horizontal and vertical seismic loads on
equipment or piping systems, hereafter labelled as FEh and FEv respectively, at the height of
the equipment, hx [55].
FEh = ag . CDi . D4 . We
FEv = 0.5 FEh
where
We = weight of equipment or equivalent weight of piping system
D4 = Amplification factor due to floor response spectrum as show in Figure 7 and
D4 1.0.
with Af = [(b + e)2 + mr]1/2
b = b b and e = e e for Tb 0.5 s
b = b (2 b 1)1/2 and e = e (2e 1)1/2 for 0.5 > Tb > 0.1 s
b = b and e = e for Tb 0.1 s
b = damping for building as defined in Table XI
e = damping for equipment
examples: light weight welded assembly 1%

welded assembly 2%
bolted assembly 5%
heavily insulated piping 35%
lightly insulated piping 3%
small bore piping 2%
raceways 510%
mr = mass ratio me/mb
me = mass of equipment
mb = effective mass of building interacting with the equipment (in most cases mb may
be reasonably assumed to be the mass of the building at the equipment level).
b = ductility coefficient defined for the building as defined in Table IX
e = ductility coefficient defined for the equipment, with e = 3.0 for distribution
systems and = 2.0 for equipment. Larger values may be used based on
experimental verification.
58
A
f
D4
1.0
0.5 0.67 1.0 1.5 2.0

T /T
e b
FIG. 7. Amplification factor for piping design.
Te: Fundamental period of equipment

Tb: Fundamental period of building
In cases where Te is not evaluated, Te needs to be assumed to be equal to Tb.
Each horizontal component FEh needs to be taken separately along the principal
horizontal axes of the equipment and combined with the vertical component FEv taken in the
most conservative of the two possible directions.
7.2.2. Simplified analytical evaluation of floor response spectra
In case modal analysis is used to evaluate the dynamic response of the structure, the
floor response spectra can be used to evaluate the horizontal and vertical seismic loads on the
equipment or piping systems.
According to [46], the floor response spectrum may be obtained directly applying the
following equations from the two standard design ground spectra which correspond to the
damping value hB of the structure and hA of the equipment or piping systems, using modal
characteristics of the structure obtained as the result of an eigenvalue analysis (TBi is the
natural period of the mode which provides significant excitation at the elevation where
equipment or piping are installed).
(U .S )
2
SE = i Ei
S Ei =
1
{(
TBi / TA ) . S (T
2 2
}
2
. hBi ) + S (TA . h A )
{ }
Bi

(T )
2
1 (TBi / TA ) + 4 (h A + hBi )
2 2 2
Bi / TA

where
59
S E: Floor response spectrum taking into account every evaluated mode of the
structure.
Ui : The i-th mode excitation function value of the floor on which the equipment is
installed. (It is suggested that the eigenvalue of the structure is evaluated from
the first to the i-th mode up to 30 Hz)
SEi: The maximum value of absolute acceleration response of the equipment and
piping system under the i-th mode acceleration of structure.
hA: Damping factor of equipment and piping system.
TA: Natural period of equipment and piping system.
hBi: Damping factor of structure.
TBi: Natural period of structure.
S(TBi hBi): The standard design ground spectrum corresponding to TBi hBi of the structure.
S(TA hA): The standard design ground spectrum corresponding to TA hA of the equipment.
Notes
(1) The mass mA of the equipment and piping system needs to be sufficiently smaller than the mass
mBi of the structure, and the response of the structure needs to not be affected by the response
of the equipment and piping system.
(2) At least 10% broadening of floor response spectrum needs to be taken into account to cope with
the uncertainty of eigenvalue analysis of the structure and the equipment and piping systems.
(3) If the structure and/or the equipment and piping system are not liner system the eigenvalue and
damping value can be evaluated applying the equivalent linear analysis.
As a next step, an eigenfrequency analysis and response analysis for the equipment
can be performed by using the floor response defined above and assuming an appropriate
damping for the piping system. Much software has been developed for this purpose. The
adequacy of software needs to be demonstrated by a comprehensive test programme.
7.2.3. Seismic design of containers
According to the results of the calculation of the natural period obtained using a
model of a single mass system, the design seismic load is calculated by the equations of
7.2.1. The obtained design seismic load is used to perform stress evaluation for the barrel
body, support legs, anchor bolts and other parts.
Buckling analysis, with analytical or numerical methods is usually requested.
Examples of the calculation methods of standard natural period and stress are
presented in Refs [47, 48].
7.2.4. Seismic design of pumps and valves
In principle, calculation of the natural period of each piece of equipment may be

performed using a model of a single mass system under appropriate support conditions
compatible with the shape of the equipment. The motor portion of the equipment mounted on
the shaft can be modelled as a single rigid body, and the natural period does not have to be
calculated. After the determination of the design seismic load by the equations in 7.2.1,
stress evaluation of anchor bolts needs to be performed, according to next sections.
60
Examples of calculation of standard natural period and stress for components are
presented in [39].
7.2.5. Seismic design of equipment and piping
Piping system design, including other distribution systems (raceways and ducts) can
be carried out by one of the following methods.
7.2.5.1. Constant pitch design
This method has been developed for conventional industrial plants and small bore
piping in hot cells of fuel reprocessing plants. The main goal is the adjustment of the pitch of
supports and making the lowest eigenfrequency higher than the dominant response frequency
of floor response. The span of systems evaluated with such method should not be longer than
the length shown in Table V.1. in Appendix V. More details are given in Refs [49, 50].
Constant pitch design is a simple and conservative design method, but care needs to
be used in case of dominant thermal expansion effects. In this case, a dynamic analysis needs
to be applied. Full dynamic analysis needs to also be used when a minimization of the number
of supports for large diameter piping is requested.
7.2.5.2. Simplified response analysis
Piping systems can be modelled as single degree of freedom systems. By using

Rayleigh Method, the fundamental eigenfrequencies can be calculated as well as their
equivalent mass value. Then internal forces can be obtained by a static elastic analysis. For
fundamental modes (one or three), the simplified dynamic approach described in previous
sections can be applied.
7.2.5.3. Allowable stresses
The maximum primary stress in equipment and piping due to earthquake and other
applicable loads should not exceed fy (specified minimum yield stress).
The maximum stress ranges in piping considering combined axial (membrane) and
bending (primary plus secondary) due to earthquake and other applicable loads should not
exceed 2.0 fy (specified minimum yield stress) of the piping material at the design basis
temperature. Higher stresses are allowed if they are substantiated by a fatigue and ratcheting
analysis [39].
7.2.6. Other considerations

(a) Piping running between different buildings/structures
For the portion of piping running between different buildings or structures, the
relative displacement of the two buildings/structures needs to be taken into
consideration.
(b) Pipe connection portion to equipment
In principle, support needs to be made as near to the equipment as possible. Also,
when the operating temperature of the equipment is high, the thermal expansion of
the equipment needs to be taken into consideration. In addition, the nozzle reaction
force acting on the equipment needs to be within the allowable range.
61
Branch connections, connections between components as well as that of component
to wall, ceiling and/or floor need to be flexible. If a component is supported by shock
isolation mounts, the vibratory motion (e.g. rocking and translation) of the
component needs to be taken into account in the design of the shock mounts (e.g. to
avoid components overturning or jumping off its mounts). In addition, all
connections made to such a component need to be flexible enough to accommodate
relative movement without excessive force or stress to such connections.
Some good practices on detailing of connecting cable and piping between
components are shown in Appendix VI.
(c) Buried piping
In this case, the behaviour of the soil during earthquake, the relative displacement
between building/structure and ground, and the thermal expansion of the piping
need to be estimated.
(d) Adjacent piping
Arrangements need to be made to ensure that there is no mutual interference
between piping caused by earthquake induced displacement.
(e) Overhead cranes, bridges and platforms and other overhead items
Overhead cranes, bridges, platforms, etc., located above critical components need to
be designed to resist the seismic force induced by the acceleration specified for the
earthquake resistant structures, even if their operation is not essential.
Provisions need to be taken to preclude damage due to breaking of plaster, window
glass, lighting fixtures, and other brittle components above the critical components.
(see Table III.1. design provision in Appendix III).
(f) Leak tightness requirement
For equipment required to be leaktight during and after an earthquake, such as glove
boxes, leaktightness needs to be assessed by dedicated analysis or demonstration test.
7.3. EQUIPMENT DESIGN FOR AIRCRAFT CRASH LOADS

In case some safety related equipment inside the building have to meet functionality
criteria during and after an aircraft crash, floor response spectra need to be computed from the
building model dynamic response. Local transfer functions need to then be evaluated
according to the equipment location and supporting scheme.
A conservative approach may rely on the following assumptions:

In case a military aircraft is considered as a reference impacting missile, the mass is
around 14 t and the speed is around 200 m/s. An equivalent load function on a rigid
target has peak at 100 Mn for 20 ms. Most of the frequency content is in the range
2040 Hz where most of the structural floors and panels shows their first bending
mode.
A hard impact on a rigid structure induces a high energy transfer to the structure: an
acceleration of 1 g in the range 2040 Hz at 10 m from the impact point represents a
reference value for design of the protection of safety related equipment
The safety related equipment may be protected either by moving it to a more
protected area or by base isolation
Protection needs to be provided against internal debris, dust and fire
62
7.4. ANCHORAGE DESIGN
Most failures related to equipment and distribution systems, including piping
systems, occur at their connections to the supporting structure. These failures are at least as
dependent on the displacement demand imposed by differential support movements, (seismic
anchor motions) as they are on inertia forces in the equipment.
Some simplified design techniques for anchoring and supports are discussed in the
following, but it needs to be reminded that they need to be followed by effective quality
control measures to guarantee their performance under seismic conditions.
In most cases, the capacity of the anchoring device and bolts are based on the
strength of concrete and therefore the design is based on concrete evaluation. However, this
practice cannot be applied for some types of anchor bolts which are installed after concrete
maturing which are sized according to their capacity.
A method for the calculation of pulling force on anchoring device for concrete
sized anchoring is shown in Appendix IV.
Design methods for anchor bolts can follow the procedure outlined here below.
(i) Cast-in-place anchors, with at least 6 diameters embedded length, need to be used
wherever possible, to obtain maximum pull-out strength. Where it is absolutely
necessary to use drilled-in (expansion type) anchors, they need to be of a proven type,
easy to install and resistant to slippage or loosening under severe vibration or impact
loading.
(ii) Where high strength studs or anchor bolts are applied, they need to be preloaded to
90% of their specified minimum yield strength or reduce the risk of prying or uplift
during an earthquake, minimizing fatigue effects and as a convenient means of
pretesting the anchor connection. Where low carbon, structural steel anchors are used
and/or sustained preloading is not dependable, the effects of prying and fatigue need
to be taken into account. In addition, consideration needs to be given to the adverse
effects of flexible mounting plates, long reach bolts or flexible concrete slabs in
which anchor bolts are installed.
(iii) Transverse shear forces need to be assumed to be applied directly to the bolts, unless
shear keys are provided. Without such keys, shearing, as well as tension stresses,
need to be taken into account, using suitable interaction formulae (square law
relationship).
(iv) Consideration needs to be given to the use of redundant anchors.
(v) Care needs to be taken in spacing of anchors and in the distance between anchors and
any free concrete edge, wall or corner, to ensure maintenance of adequate pull-out
strength. Otherwise, a suitable strength reduction factor needs to be applied to such
anchors. Consideration needs to also be given to bolts for anchoring equipment to
floors and especially walls and ceilings, where adequate spreader plates are used on
the opposite side of the floor, wall or ceiling.
(vi) For cast-in-place anchors, the minimum factor of safety against failure in any mode,
including pull out, needs to be 2.5.
(vii) For drilled-in, expansion type anchors, the minimum factor of safety, against failure
in any mode, including pull-out, needs to be 4.
63
It needs to be understood that positive anchorage in the form of anchor bolts needs to
be used to carry uplift due to overturning effects for all equipment. Anchor bolts in the
absence of engineered shear keys need to be designed to carry applicable shear, except where
the shear friction capacity between the component and its foundation can be shown to carry
applicable lateral loads with a safety factor of at least 2.0.
Capacities of anchor bolts of various type and size and under different loading and
geometric conditions are typically given in national codes and manufacturers installation
specification.
8. SLOSHING EFFECTS
Sloshing may be produced in a pool or tanks by strong earthquakes. This

phenomenon may generate waves in the pool which may interact very strongly with the
bridge, from which the reactor and its control system are suspended and submerged structures
near to the water surface. Sloshing can begin several tens of seconds after arrival of the first
higher frequency seismic waves at the site as its typical natural frequencies are usually much
lower than the structural ones.
The evaluation of the sloshing phenomenon can be of interest for two main reasons:
the estimation of the dynamic interaction between the structure and the pool or tank
the evaluation of the wave height (for water runoff) and therefore of the
hydrodynamic pressure on the container wall
For both goals, an equivalent system with two masses and stiffnesses may be used.
The liquid may be replaced with a mass M0 rigidly fixed to the tank at an elevation H0 above
the bottom, plus a mass M1 attached through springs of total stiffness K at elevation H1. For a
cylindrical tank with flat bottom these parameters are given by [51]:
M0 = [(tanh 1.7R/H)/(1.7 R/H)]*M

M1 = [(0.71 tanh 1.8H/R)/(1.8H/R)]*M
H0 = 0.38 H [1.0+(M/M0-1)]
H1 = H[1.0 0.21 M/M1(R/H)^2+0.55R/H [0.15(RM/Hm1)^21.0]^1/2
K = 4.75 g M1^2 * H/(MR^2)
where g is the acceleration of gravity, H the height of the tank and R its radius.
An alternative approach is provided in [52] and [44].

The corresponding solution for a rectangular tank that measures 2L in the direction of
motion is
M0 = [(tanh 1.7 L/H)/(1.7 L/H)]*M

M1 = [(0.83 tanh 1.6H/L)/(1.6H/L)]*M
H0 = 0.38H[1+(M/M0 1.0)]
H1 = H [1.0 - 0.33M/M1(L/H)^2+0.55L/H [0.28 (LM/HM1)^2 1.0]^1/2
K = 3.0 g M1^2 * H/(ML^2)
64
For both shapes of container, = 1.33 and = 2.0, if the hydrodynamic moment on
the tank bottom is to be included in the computation, while = 0.0 and = 1.0, if only the
effects of hydrodynamic pressures on the container walls are of interest.
The solution for a cylindrical tank with a hemispherical bottom may be taken to be
equal to the one for a tank with flat bottom of the same radius and same volume as the tank in
question.
The amplitude of the height of waves set up by the vibration may be taken equal to
the horizontal displacement amplitude x of mass M1 times the factor
= (0.69 K R/M1g)/[1.0 0.92 (x/R)(KR/M1g)^2] in cylindrical tanks,
and
= (0.84 K L/M1 g)/[1.0 (x/L)(KL/M1g)^2] in rectangular tanks.
These expressions are satisfactory provided x does not exceed about 0.2 R, o.2 L, or
0.02 H. Beyond these limits non-linear phenomena become important.
Energy dissipation due to viscosity of the liquid can be expressed as an equivalent

percentage of critical damping. This quantity decreases rapidly with increasing linear
dimensions of the container and is only a small fraction of 1% for tanks of practical interest.
For small values of H/R or H/L, the approximation T1 1.07 R/H^1/2 and T1 1.25
L/H^1/2 are useful for estimating the fundamental period of liquids in cylindrical and
rectangular tanks respectively (T1 is in seconds and H, R, L in metres). The error introduced
by these expressions does not exceed 2% when H/R is smaller than 0.25.
Provision needs to be made for water overflowing pools (reactor or fuel storage) and
potential loss of essential coolant or shielding. In addition, the risk of radioactive material
escaping from the facility needs to be taken into account in the event of earthquake induced
sloshing. Consideration needs to be given to containing such overflow and returning it to the
pool.
Sloshing in closed tanks can produce a strong vacuum behind the surface wave.
Vacuum breakers need to be considered in the tops of tanks to avoid collapse in the event of a
large earthquake.
9. OTHER CONSIDERATIONS
9.1. SEISMIC SCRAM SYSTEM

The design bases for reactor protection and shutdown function associated with
seismic event need to be addressed in the safety analysis report for facilities built in seismic
areas.
As an example, for Class A research reactors (see Chapter 2) a seismic scram system
(automatic seismic trip system (ASTS)) needs to be provided.
65
A monitoring system needs to be provided even in case an ASTS is not installed, to
drive the operator action and the post event inspections. In this framework, the system needs
to be safety related.
For NFOPs, consideration needs to be given to automatic actions to attain a safe state
in case of an earthquake. The facility needs to have protection capabilities in all operating
modes and conditions. Operational limits and conditions of seismic scram system including
surveillance tests and intervals need to be based on safety analysis regarding seismic events.
Appendix VII gives additional information on automatic seismic trip systems for
nuclear power plants, which may drive the decision for post event operator actions also for
other installations.
9.2. MONITORING SYSTEMS AT THE SITE

It is suggested that each facility have a minimum environmental monitoring at the
site for the external events that proved to be sizing for the design of the facility. In case of
seismic areas, at least one strong motion monitoring system at the site is suggested.
Data can be used both to confirm the design basis and as a background information in
case the facility needs periodic safety reviews or life extension.
9.3. QUALITY ASSURANCE

The seismic design process requires the use of sound engineering/scientific
principles and appropriate design standards. Design requirements, input, process, outputs
change, records and organizational interface are controlled.
IAEA safety standards [53] provides detailed recommendations on the QA

management for NFOPs.
9.4. EVALUATION OF RADIATION EXPOSURE OF THE PUBLIC DURING AND

AFTER EXTERNAL EVENTS
Safety features to withstand external events and potential associated accidents need
to be taken into account in the design process.
Therefore it is required that the nuclear facility is designed to accommodate external

events by shutting down the reactor, removing the residual heat and controlling basic
radiological parameters.
A radiological dispersion analysis through air, water and groundwater is required for
any nuclear installation. Results need to comply with basic requirements of Section 2.
The emergency planning needs to be based on such a study and needs to rely on the
availability of escape routes from the site and communication lines during and after any
design basis external event.
The need of an emergency evacuation plan is discussed in Ref. [1], with reference to
the radiological hazard posed by the research reactors. Similar criteria could be used for other
NFOPs.
66
REFERENCES
[1] INTERNATIONAL ATOMIC ENERGY AGENCY, Siting of Near Surface Disposal

Facilities, Safety Series No. 111-G-3.1, IAEA, Vienna (1994).
[2] INTERNATIONAL ATOMIC ENERGY AGENCY, Siting of Geological Disposal
Facilities, Safety Series No. 111-G-4.1, IAEA, Vienna (1994).
[3] INTERNATIONAL ATOMIC ENERGY AGENCY, Report on Radioactive Waste
Disposal, Technical Reports Series No. 349, IAEA, Vienna (1993).
[4] INTERNATIONAL ATOMIC ENERGY AGENCY, Use of Probabilistic Safety
Assessment for Nuclear Installations with Large Inventory of Radioactive Material,
IAEA-TECDOC-711, Vienna (1993).
[5] INTERNATIONAL ATOMIC ENERGY AGENCY, Code on the Safety of Nuclear
Research Reactors: Design, Safety Series No. 35-S1, IAEA, Vienna (1992).
[6] INTERNATIONAL ATOMIC ENERGY AGENCY, Code on the Safety of Nuclear
Research Reactors: Operation, Safety Series No. 35-S2, IAEA, Vienna (1992).
[7] INTERNATIONAL ATOMIC ENERGY AGENCY, Safety Assessment of Research
Reactors and Preparation of the Safety Analysis Report, Safety Series No. 35-G1,
IAEA, Vienna (1994).
[8] INTERNATIONAL ATOMIC ENERGY AGENCY, Safety in the Utilization and
Modification of Research Reactors, Safety Series No. 35-G2, IAEA, Vienna (1994).
[9] INTERNATIONAL ATOMIC ENERGY AGENCY, Guidelines for the Review of
Research Reactor Safety, Services Series No. 1, IAEA, Vienna (1998).
[10] INTERNATIONAL ATOMIC ENERGY AGENCY, Siting of Research Reactors,
IAEA-TECDOC-403, Vienna (1987).
[11] INTERNATIONAL ATOMIC ENERGY AGENCY, Management of Research Reactor
Ageing, IAEA-TECDOC-792, Vienna (1995).
[12] INTERNATIONAL ATOMIC ENERGY AGENCY, Earthquake Resistant Design of
Nuclear Facilities with limited Radioactive Inventory, IAEA-TECDOC-348, Vienna
(1985).
[13] INTERNATIONAL ATOMIC ENERGY AGENCY, Seismic Design Considerations of
Nuclear Fuel Cycle Facilities, IAEA-TECDOC-1250, IAEA, Vienna (2001).
[14] INTERNATIONAL ATOMIC ENERGY AGENCY, Safety of and Regulations for
Nuclear Fuel Cycle Facilities, IAEA-TECDOC-1221, Vienna (2001).
[15] INTERNATIONAL ATOMIC ENERGY AGENCY, Procedures for Conducting
Probabilistic Safety Assessment (PSA) for Non-Reactor Nuclear Facilities, IAEA-
TECDOC 1267, IAEA, Vienna (2002).
[16] INTERNATIONAL ATOMIC ENERGY AGENCY, Site Evaluation for Nuclear
Facilities, Safety Standards No. DS 305, IAEA, Vienna (in preparation).
[17] INTERNATIONAL ATOMIC ENERGY AGENCY, Earthquakes and Associated
Topics in Relation to Nuclear Power Plants Siting, Safety Series No. 50-SG-S1 (Rev.
1), IAEA, Vienna (1991).
[18] INTERNATIONAL ATOMIC ENERGY AGENCY, External Human Induced Events
in Site Evaluation for Nuclear Power Plants, Safety Standards Series No. NS-G-3.1,
IAEA, Vienna (2002).
[19] INTERNATIONAL ATOMIC ENERGY AGENCY, Food Hazard for Nuclear Power
Plants on Coastal and River Sites, Safety Standards No. 50-SG-S10A, IAEA, Vienna
(in preparation) [DS280].
[20] INTERNATIONAL ATOMIC ENERGY AGENCY, Meteorological Events in Site
Evaluation for Nuclear Power Plants, Safety Standards Series No. NS-G-3.4, IAEA,
Vienna (in preparation) [DS184].
67
[21] INTERNATIONAL ATOMIC ENERGY AGENCY, Safety of Nuclear Power Plants:
Design, Safety Standards Series No. NS-R-1, IAEA, Vienna (2000).
[22] INTERNATIONAL ATOMIC ENERGY AGENCY, Seismic Design and Qualification
for Nuclear Power Plants, Safety Standards Series, IAEA, Vienna (in preparation)
[DS304].
[23] INTERNATIONAL ATOMIC ENERGY AGENCY, External Events Excluding
Earthquakes in the Design of Nuclear Power Plants, Safety Standards Series No. NS-G-
1.5, IAEA, Vienna (2003) [DS301].
[24] INTERNATIONAL ATOMIC ENERGY AGENCY, Geotechnical Aspects of Nuclear
Power Plant Site Evaluation and Foundations, Safety Standards Series, IAEA, Vienna
(in preparation) [DS300].
[25] INTERNATIONAL ATOMIC ENERGY AGENCY, Safety Assessment and
Verification for Nuclear Power Plants, Safety Standards Series No. NS-G-1.2, IAEA,
Vienna (2001).
[26] INTERNATIONAL ATOMIC ENERGY AGENCY, INES Users manual, IAEA,
Vienna (2001).
[27] AMERICAN ASSOCIATION FOR CIVIL ENGINEERING (ASCE), A summary
description of design criteria, codes, standards and regulatory provisions typically used
for the civil and structural design of nuclear fuel cycle facilities, ASCE, New York
(1988).
[28] INSTITUTE POUR LA PROTECTION ET LA SRET NUCLAIRE (IPSN), The
Risks of Criticality in the Nuclear Industry, International Conference, Versailles, 15
October 1999.
[29] LOS ALAMOS NATIONAL LABORATORY, A review of criticality accidents, 2000
Revision, LA-13638, May 2000.
[30] AECL INTERNATIONAL SYMPOSIUM, Research reactor safety, operations and
modifications, AECL, Chalk River, Canada, 2327 October 1989.
[31] INTERNATIONAL ATOMIC ENERGY AGENCY, Ageing, decommissioning and/or
major refurbishment of research reactors, proceedings of a seminar held in Bangkok,
Thailand, 1822 May 1992, IAEA, Vienna (1992).
[32] FOOD AND AGRICULTURE ORGANIZATION OF THE UNITED NATIONS,
INTERNATIONAL ATOMIC ENERGY AGENCY, INTERNATIONAL LABOUR
ORGANIZATION, OECD NUCLEAR ENERGY AGENCY, PAN AMERICAN
HEALTH ORGANIZATION, WORLD HEALTH ORGANIZATION, International
Basic Safety Standards for Protection against Ionizing Radiation and for the Safety of
Radiation Sources, Safety series No. 115, IAEA, Vienna (1996).
[33] INTERNATIONAL NUCLEAR SAFETY ADVISORY GROUP, Basic Safety Principles
for Nuclear Power Plants, INSAG Series No. 12 (75-INSAG-3 Rev. 1), IAEA, Vienna
(1999).
[34] INTERNATIONAL ATOMIC ENERGY AGENCY, Application of the Single Failure
Criterion, Safety Series No. 50-P-1, IAEA, Vienna (1990).
[35] INTERNATIONAL ATOMIC ENERGY AGENCY, Guidelines for Integrated Risk
Assessment and Management in Large Industrial Areas, IAEA-TECDOC-994, Vienna
(1998).
[36] INTERNATIONAL ATOMIC ENERGY AGENCY, Manual for the Classification and
Prioritization of Risks due to Major Accidents in Process and Related Industries, IAEA-
TECDOC-727, Vienna (1996).
[37] US/DEPARTMENT OF ENERGY (DOE) standard, Natural phenomena hazards
design and evaluation criteria for department of energy facilities, DOE-STD-1020-94.
68
[38] UNITED STATES ELECTRIC POWER RESEARCH INSTITUTE, A Methodology for
Assessment of Nuclear Power Plant Seismic Margin, USA-EPRI Report No. NP-6041,
October 1988.
[39] PARK, Y.J., HOFMAYER, C.H., Technical guidelines for aseismic design of nuclear
power plants, NUREG/CR6241, BNL-NUREG-52422, NRC, Washington D.C., June
1994.
[40] US/NATIONAL EARTHQUAKE HAZARD REDUCTION PROGRAM (NEHRP),
Recommended Provisions for Seismic Regulations for New Buildings, 1994 editions,
FEMA 223A, May 1995.
[41] US/NATIONAL EARTHQUAKE HAZARD REDUCTION PROGRAM (NEHRP),
Recommended Provisions for Seismic Regulations for New Buildings and Other
Structures, Part 1: Provisions, FEMA 302, Feb. 1998.
[42] INTERNATIONAL CONFERENCE OF BUILDING OFFICIALS (ICBO), Uniform
Building Code, International Conference of Building Officials, Whittier, Ca., 1997.
[43] MINISTRE DE LINDUSTRIE, DIRECTION DE LA QUALIT ET DE LA
SECURIT INDUSTRIELLES, Rgles fondamentales de sret, RFS I.2.d.
[44] AMERICAN SOCIETY OF CIVIL ENGINEERS (ASCE) 4-98, Seismic Analysis of
Safety Related Nuclear Structures and commentary, ASCE, New York, (1998).
[45] BANGASH, M.Y.H., Impact and Explosion: Analysis and Design, Blackwell Scientific
Publications, Oxford (1993).
[46] YASUI, Y., et al., Direct generation method for floor response spectra, K-13/4,
SMiRT-12 (1993).
[47] MINISTER OF TRADE AND INDUSTRY (MITI), Practice #515, Earthquake
Resistant Design Guideline for High Pressure Gas Facilities, Tokyo (1981 rev. 1997).
[48] KANAGAWA, Earthquake Resistant Design Standard for High Pressure Gas Facilities,
Japan (1993).
[49] MINISTRY OF TRADE AND INDUSTRY (MITI), Guides for Petrochemical Industry,
Tokyo (1997).
[50] MINISTRY OF TRADE AND INDUSTRY (MITI), Guide for Reprocessing Plants,
Tokyo (1998).
[51] NEWMARK, N.M., ROSENBLUETH, E., Fundamentals of Earthquake Engineering,
Prentice Hall, Englewood Cliffs, New York (1982).
[52] UNITED STATES ATOMIC ENERGY COMMISSION, Nuclear reactors and
earthquakes, August 1963, TID-4500, USAEC, Washington DC.
[53] INTERNATIONAL ATOMIC ENERGY AGENCY, Quality Assurance for Safety in
Nuclear Power Plants and Other Nuclear Installations, Code and Safety Guides Q1
Q14, Safety Series No. 50-C/SG-Q, IAEA, Vienna (1996).
[54] UNIVERSITY OF TOKYO, EARTHQUAKE RESEARCH CENTER, Anti-earthquake
design code for high pressure gas manufacturing facilities, July 1981, ERS Report
No. III-5.
[55] STEVENSON, J.D., A new procedure for the safe, economical seismic design of
components of nuclear plant facilities at low and moderate seismicity sites, Nuclear
Safety, Vol. 28, No. 2, AprilJune 1987.
69
Appendix I
SIMPLIFIED PROCEDURES FOR ASSESSMENT OF

LIQUEFACTION POTENTIAL
Saturated alluvial sandy layers, which have the water table within 10 m from the
ground surface, and have D50 values on the grain size accumulation curve between 0.02 and
2.0 mm, are vulnerable to liquefaction for the depth between 0 and 20 m. The liquefaction
potential of these layers can be estimated according to any of the two methods outlined in this
Appendix [I.1].
I.1. ESTIMATIONS OF LIQUEFACTION (METHOD 1)

For soil layers which are judged to be vulnerable, liquefaction potential need to be
checked based on liquefaction resistance factor FL defined by the following equations:
FL = R/L (I.1)
where
FL : liquefaction resistance factor
R : resistance of soil elements to dynamic loads, and
R = R 1 + R2 (I.2)
R1 and R2 should be determined in accordance with Figs A-1 and A-2 respectively
L : dynamic loads to soil elements induced by earthquake motion evaluated as:
L=rd*Ks*(v/v) (I.3)
rd=1.0 0.015*z (I.4)
z depth from the actual ground surface (m)
Ks seismic coefficient for evaluation of liquefaction, taken as:

Design Intensity Level 1, Ks = 0.13
v: total overburden pressure (daN/cm2)
v: effective overburden pressure at the static condition (daN/cm2)
Soil layers having liquefaction resistance factor FL smaller than 1.0 need to be judged to
liquefy during earthquakes. Figs. I.1 and I.2 are graphic illustrations of the first term R1 and
the second term R2 represented in the following equations which were proposed based upon
the results of laboratory dynamic triaxial tests on soil specimens taken from several sites in
Japan.
71
N-Value by the Standard Penetration Test
0 10 20 30
Effective Overburden Pressure 0v2 (kg/cm2)

F ir
st T
0.5 er m
R
1=
0.3 0.0%
4
1.0 0 .3
2
0 .3
0
0.
28
1.5
0.
26
0 .2
4
2.0
0.2
2
0.2
0
0.18
0 .1 6
0 .12
0.06
0.10
0.14
0.08
2.5
3.0
FIG. I.1.
0.3
0.2 0.19
Second Term R2
0.1
0.0
-0.05
-0.1
0.6
-0.2
0.01 0.02 0.03 0.1 0.2 0.5 1.0 2.0 3.0
Mean Particle Diameter, D50 - Value on the Grain Size

Accumulation Curve
FIG. I.2.
1
N
2
R=0.0882 +0.19 (0.02mmD50 0.05mm)

' +0.7
1
N 0.35
2
R=0.0882 +0.225log (0.05mm<D50 0.6mm )

' +0.7 D50
1
N
2
R=0.0882 0.05 (0.6mm<D50 2.0mm )

' +0.7
N is the value obtained from the standard penetration test
72
I.1.1. Treatment of soil layers which were judged to liquefy
For those soil layers which were judged to liquefy according to the estimation of
subsection I.1 and are within 20 m of the actual ground surface, bearing capacities and other
soil constants need to be either neglected or reduced in the seismic design, by multiplying the
original bearing capacities by reduction factors DE which are determined in accordance with
FL values in Table I.1.
TABLE I.1. FL VS. DE RELATION

FL Depth, Z (m) Reduction Factor, DE
FL 0.6 Z 10 0
10 < Z 20 1/3
0.6 < FL 0.8 Z 10 1/3

10 < Z 20 2/3
0.8 < FL 1.0 Z 10 2/3

10 < Z 20 1
1.0 < FL 1
I.2. ESTIMATIONS OF LIQUEFACTION (METHOD 2)
An alternative simplified procedure to evaluate liquefaction potential of sandy soils is

as follows:
The cyclic shear stress ratio developed in the field due to earthquake excitations can
be computed from:
(d / o) = n (ag / g) (o / o) d (I.5)
in which
ag = design horizontal acceleration defined in Chapter 6, equation (6.1)

g = gravity acceleration
o = total overburden pressure (daN/cm2)
o = effective overburden pressure (daN/cm2)
d = stress reduction factor defined by d = 1 0.015z
z= depth from the ground surface n meter
n = 0.1 (M 1)
M= magnitude of the biggest earthquake which can conceivably cause
liquefaction at the site
73
The difference in number of cycles of stress due to different magnitude earthquakes
is taken into account in Eq. (I.5) with the reduction factor d.
The soil resistance to liquefaction, 1 / o, can be correlated with some form of

modified penetration resistance according to the following procedure:
(a) Compute SPT Na value normalized to the effective overburden pressure and fines content
by:
Na = 1.7 N/ (o + 0.7) + N f (I.6)
in which N = measured SPT N-value
N f = modification factor in terms of fines content (percentage of fines smaller than

0.074 mm) as shown in Fig. I.3.
10
Nf
0
0 20 40 60 80 100
Fines Content (%)
FIG. I.3. Relationship between Nf and fines content.

(b) Determine liquefaction resistance with limiting strain potential of per cent by Fig. I.4.
The limiting strain potential is the maximum cyclic shear strain likely to be developed by the
earthquake excitations.
FIG. I.4. Relationship between cyclic stress ratio, Na value and limiting strain potential of
sandy soil deposits
74
The factor of safety against liquefaction with limiting strain potential of per cent, F1
can be determined by:
'1 / ' 0
F1 =
d / '0
( A 7(I.7)
)
Soil layers with a safety factor less than one can be considered to liquefy during the
earthquake. Since the damage of liquefaction takes very different forms depending on the
shear strain developed, the following guideline may be tentatively given to specify the degree
of liquefaction.
TABLE I.2. DEGREE OF LIQUEFACTION

Factor of Safety Shear Strain Degree of Liquefaction
F1 <1 20% Extensive

F1 = 1 510% Intermediate
F1 = 1 25% Slight
F1 = >1 2% No significant
Treatment of soil layer which is judged to liquefy. The soil layer which is judged to
liquefy needs to be densified or stabilized so that it can withstand earthquake shaking.
REFERENCE
[I.1] THE JAPAN SOCIETY OF CIVIL ENGINEERS, Earthquake engineering
committee, Earthquake Resistant Design for Civil Engineering Structures in Japan,
Tokyo, 1992.
75
Appendix II
SOIL STRUCTURE INTERACTIONS
II.1. GENERAL
This appendix proposes a method for considering the soilstructure interaction to be

used with the dynamic stick model defined in Section 6.2. It consists in adding a soil spring-
dashpot system at the base of the model. The numerical values of the spring constants
(stiffness) together with the damping coefficients are given in Section II.2 of this Appendix.
Section II.3 gives information to determine the equivalent modal damping factors to be used
in the structural analysis.
In a soilstructure interaction analysis, the dynamic stiffness coefficients of the soil

are originally frequency dependent. Here, however, it is suggested to use the static stiffness
coefficients as spring constants which are frequency independent. The proposed method is
based on classical formulae which may be applied in rather regular subsoil conditions. In case
of very irregular subsoil conditions, special methodology needs to be used in accordance with
specialists.
Formulae are given for a superficial foundation on a homogeneous half-space and on

a soft layer, and for an embedded foundation in a case of homogeneous half-space according
to [II.1-II.6]. The designer needs to compare the actual situation with these cases. Soil
properties results from geotechnical investigations defined in Section 5.
For the application of the formula, the shear modulus of the soil is needed. It can be
evaluated from the shearwave velocity, Vs, by;
G = Vs2 (II.1)
where, is the mass per unit volume of soil, which in lack of data can be taken from
Table II.1 which provides also the Poissons ratio, .
II.2. EQUIVALENT SPRING AND DAMPING
II.2.1. Surface foundation on a homogeneous half-space
The equivalent spring constants and the damping coefficients are given in Table II.2.
TABLE II.1. SOIL PROPERTIES.

Soil type
according to 1 2 3
Table 6 at
5.1.1
(daN/m3) 2200 2000 1800
0.3 0.4 0.45
76
It needs to be noted that the formulae for the spring constants depend on the assumed
conditions such as the soilfoundation interface.
Those in Table II.2 are results based on the assumption that the soil is subjected to
the stress distribution induced by a rigid foundation.
For the damping coefficients, the following applies:
C1 = 0.5 (II.2)
(
C2 = 0.30 1 + B ) (II.3)
where,
3 (1 ) I 0
B = (II.4)
8 R5
For a rectangular foundation, the radius R, to be used in Eq. (II.4), is an equivalent

radius equal to:
R= BL / for translation (II.5)
R = 4 BL3 3 for rocking (II.6)
C3 = 0.8 (II.7)
TABLE II.2. SPRING CONSTANTS AND DAMPING COEFFICIENTS FOR

FOUNDATIONS ON HOMOGENEOUS HALF-SPACE
Direction of Equivalent Spring Constant Equivalent Spring Constant Equivalent Damping

Motion for Rectangular Foundation for Circular Foundation Coefficient
K H = 2 (1 + ) G x BL 32(1 ) G R CH = C1 K H R G
Horizontal KH =
7 8
G 8G R3 C R = C2 K R R G
Rocking KR = B L2 KR =
1 3 (1 )
G 4G R CV = C3 KV R G
Vertical KV = v BL KV =
1 1
____ 16 G R 3 KT I T
Torsion KT = CT =
3 1 + 2 I T R 5
77
= Poissons ratio of soil medium
G = shear modulus of soil medium
= density of soil medium
R = radius of the circulate foundation
B = width of the foundation perpendicular to the direction of horizontal excitation
L = length of the foundation in the direction of horizontal excitation
I 0 = total mass moment of inertia of structure and foundation about the rocking
axis at the base
I T = polar mass moment of inertia of structure and foundation
Constants x , , and v for a rectangular foundation should be evaluated in the

following Fig. II.1.
FIG. II.1. Constants for the formulas in Table II.2.
where:
c = length of the foundation in the direction of horizontal excitation
b = width of the foundation perpendicular to the direction of horizontal excitation
II.2.2. Rigid circular foundation on a stratum over a rigid bedrock
The spring constants are given in Table II.3. The damping coefficients should still be
evaluated from Table II.2.
78
TABLE II.3. SPRING CONSTANTS FOR RIGID CIRCULAR FOUNDATION ON A
STRATUM OVER A RIGID BEDROCK
Direction of Motion Equivalent Spring Constant

(Static Stiffness) Range of Validity
8G R 1 R
Horizontal KH = 1 + H R >1
2 2 H
8G R3 1 R
Rocking KR = 1 + 4 H R >1
3(1 ) 6 H
4G R R
Vertical KV = 1 + 128
. H R>2
1 H
16 G R 3
Torsion KT = H R 125
.
3
H = thickness of stratum
II.2.3. Rigid circular foundation embedded into a stratum over a rigid bedrock
The spring constants are given in Table II.4. The damping coefficients are
calculated from Table II.2.
TABLE II.4. SPRING CONSTANTS FOR EMBEDDED CYLINDRICAL FOUNDATION

ON A STRATUM
Direction Equivalent Spring Constant Range of Validity

of Motion (Static Stiffness)
8G R 1 R 2 D 5 D
Horizontal KH = 1 + 1 + 1 +
2 2 H 3 R 4 H
8G R3 1 R D D
Rocking KR = 1 + 1 + 2 1 + 0.7
3(1 ) 6 H R H
4G R R 1 D D D H
Vertical KV = 1 + 1.28 1 + 1 + 0.85 0.28
1 H 2 R R 1 D H
Coupled
horizontalr K HR = 0.40 K H D
ocking
16 G R 3 D D R<2
Torsion KT = 1 + 2.67 D H 0.5
3 R
D = depth of embedded foundation
79
II.3. MODAL DAMPING
For each foundation motion, the reduced radiation damping factor of a soil is
calculated from the damping coefficient determined in Section II.2:
C
SSI = 0.5 (II.8)
2K
where, C and K are the damping and the stiffness coefficients. The factor 0.5 is intended to
take into account the fact that the actual radiation damping is less than that for a regular
half-space, due to waves reflection in horizontal soil layers.
It should be noted that for low frequencies, the damping factor may decrease to
approximately zero in case of a foundation on an elasticstratum which may trap the reflected
waves. The overall soil damping factor is shown as:
s = g + SSI (II.9)
This value should be limited to 30%
where, g is the hysteretic soil damping defined in Table II.5.
TABLE II.5. HYSTERETIC SOIL DAMPING.

Soil type according to 1 2 3
Table 6 5.1.1
g 3% 5% 10 %
The modal damping factor can be calculated by averaging the damping values of
each element or subsystem, such as the soil, using the following formula:
E
j =1
j
i
j
i = p
(II.10)
E
j =1
i
j
where,
E ij = i T K j i (II.11)
The structure is divided into p parts, each has a damping factor equals to j. K j is
the stiffness matrix reduced to the j-th part. i is the i-th modal vector, reduced to the j-th part.
This value should not exceed 20 %.
80
REFERENCES
[II.1] KAUSEL, E., et al.., Dynamic analysis of embedded structures: Transactions, 4th
SMiRT, SAN FRANCISCO, Paper K2/6, Vol. 1977.
[II.2] KAUSEL, E., USHIJIMA, R., Vertical and torsional stiffness of cylindrical footings,
MIT Research Report R79-6, Dept. of Civil Engineering, Massachusetts Institute of
Technology, February 1979.
[II.3] LUCO, J.E., Impedance functions for a rigid foundation on a layered medium,
Nuclear Engineering Design, Vol.31, n. 57, 1974.
[II.4] GAZETAS, G., Analysis of machine foundation vibrations: state of the art, Soil
Dynamics and Earthquake Engineering, Vol. 2, N 1., 1983.
[II.5] NEWMARK, M., ROSENBLUETH, E., Fundamentals of earthquake engineering,
Prentice-Hall, Inc., Englewood Cliffs, N.J., P. 98, 1971.
[II.6] GAZETAS, G., Foundation vibrations, Foundation engineering handbook, 2nd
Edition, H.Y. Fang, ed., Van Rostrand Reinhold, 1991.
81
Appendix III
EARTHQUAKE RESISTANT DESIGN PROVISIONS
TABLE III.1. EARTHQUAKE RESISTANT DESIGN PROVISIONS
Item Design Provision
Concrete block partition walls Properly reinforce walls, dowel to floor or tie into
steel work, to avoid collapse in an EQ.
Instrument stands and equipment platforms. Add cross bracing; brace back to wall if tall. Anchor
well to resist EQ forces and overturning moments.
Cranes, hoists, jibs, moving bridges Make design provisions for tethering or clamping
working platforms hoists/cranes in a safe position when out of service.
Lower loads onto safe areas when hoisting/handling
operations are over (administrative).
Building to building clearance Provide enough rattle space or use soft caulking.
Safetyrelated equipment close to NSQ Increase normal separation. Cage or barricade NSQ
equipment or structures. equipment. Protect safetyrelated equipment. Secure
NSQ structures or equipment to prevent collapse. Add
redundant or diverse safetyrelated equipment (well
separated). Use fail-safe equipment.
Secure and lock handrails, ladders, etc. Mount

Ladders, handrail, guard rails, stairways, equipment on separate SQ supports
etc.
Secure ceilings and furniture close to sensitive
False or suspended ceilings. Loose equipment. Add curbs and railings around critical
furniture. control consoles to prevent impact from furniture
moving in an EQ.
Qualify by testing. Cast-in anchors suggested.

Drilled-in expansion anchors in lieu of cast- Highstrength anchor bolts preferred (preloaded).
in (where essential) Redundant anchors desirable. Avoid grouted-in
anchors. Through-wall anchors are best.
Long, vertical pipes supported at top and Use lateral restraints at suitable intervals, to provide
bottom only. for horizontal EQ effects.
Small branch pipe or tubing connections Motion limits. Good anchorage. Proper flexibility to
allow for differential movement in an EQ.
82
Item Design Provision
Field-run tubing, small piping and electrical Route carefully or protect well to avoid impact
conduits, small valves and fittings interaction with larger pipes, ducts, etc. during an EQ.
Use adequate clamps and supports.
Conventional pipe hangers
Add lateral restraints at suitable intervals. Replace
rigid rods by swivel type. Avoid use of threading in
the plane of maximum stress.
Tall, overhung valves and valve operators
Add lateral restraints or motionlimiting stops, as
necessary, to limit EQinduced stresses.
Overhead ductwork Strengthen (lock) duct joints. Add end restraints. Use
adequate supports. Use backup supports, where
consequences of falling in an EQ are serious.
Building wall penetrations Use adequate clearance around penetrations, sealed

with flexible, fireproof boots on the inside; or weld
penetrations to embedments on the inside and use soft
bedding, on the outside, with flexible terminations or
bellows.
Underground building wall penetration If underground water level is permanently higher than
the penetration level, qualify tightness device joints
for relative displacement and water pressure
Water, fuel or lubricant lines and storage Adequate support bracing. use protective curbs and
tanks proper drainage, sprinklers, halon or other
fireprotection feature to mitigate effect an EQ.
Tank and equipment supports Add bracing. Double up anchors, with suitable
spacing. Tie back to wall where bracing is insufficient
or tank is too tall.
Highpressure gas storage bottles. Secure bottles to storage racks at top and bottom.
Instrument air reservoirs. Properly support supplyside check valves. Improve

anchorage.
Storage batteries Strengthen battery racks and anchors. Tie batteries to

racks (top and bottom). Place batteries closer to floor
level.
Radioactive fuel/waste storage, hot cells, Secure storage areas. Confine and SQ cooling and
ventilated glove boxes, etc. shielding water systems. SQ ventilation systems,
where necessary for safety
Key: EQ = earthquake; SQ = seismically qualified; NSQ = not seismically qualified
83
Appendix IV
COMPUTATION OF PULLING FORCE ON ANCHOR DEVICE
A general scheme for the evaluation of the actions at the anchoring of equipment is
shown in Fig. IV.1.
FIG. IV.1. Schematic explanation on calculation of pulling force.
The equilibrium of tensile and compressive forces developed in the anchorage system
as well as equilibrating the applied moment to the couple formed by the section tensile force
Tbn and compressive force need to be considered. The distribution of section tensile force Tbn
to individual bolts should be a function of tensile strain in the bolt which in turn is a function
of an individual bolts distance from the neutral axis (N.A.).
The following procedures are considered acceptable methods of determining tensile

and compressive forces on anchoring devices.
84
IV.1. SIMPLIFIED CONSERVATIVE METHOD
M
Tb =
dn
where:
M = applied overturning moment

d = distance between bolts
b = width of base or foundation
n = number of bolts
Tb = tensile force per bolt
IV.2. ELASTIC METHOD - SQUARE FOUNDATIONS
c = ( MC / I ) + ( P / b 2 )
t = ( MC / I ) ( P / b 2 )
t t
x=b =b
c +t 2 MC / I
FIG. IV.2. Bearing force of foundation.
85
then:
b t
Tb n = 1 ( ) b
2 2MC / I t
where:
C = the distance from neutral axis to the
critical side edge
I = moment of inertia of the section.
Allowable
f c 0.35 f c
FIG. IV.3. Reaction force calculation

for anchor bolts.
IV.3. CONCRETE COLUMN ANALOGY
FIG. IV.4. Elastic design and plastic design
Determination of Tbn is based on methods of designing reinforced concrete

columns in accordance with national building codes.
86
IV.4. ISSUES AFFECTING THE PULLING FORCE ON ANCHOR
The geometry of the pits of the bolts can strongly affect the allowed pulling force
beard by an anchor bolt, In the following, some simple rules are collected for an easy and
effective design of anchoring devices [IV.1].
IV.4.1. Pit placement standard
The pit placement standard of anchor bolts is shown in Table IV.1 and Fig. IV.5
which depends upon the type of anchor bolt used.
IV.4.2. Reduced pit placement
When distances between the pits are shorter than the placement distance in the pit
placement standard, the allowable pulling out load for a single anchor bolt should be reduced.
The allowable load for pulling out in this case is to be the value determined
multiplying the value obtained from the type of used bolts by the reduction ratio shown in
Table IV.2. (i) to (iii) (refer to Figs IV.5IV.6).
TABLE IV.1. PIT PLACEMENT STANDARD

Anchor bolt type Pit placement standard
Embedded L and LAtype anchor bolts, At or greater than 10d
chemical expansion anchor bolts d: Anchor bolts nominal diameter
Embedded J and JAtype bolts, At or greater than 2L
headattached bolts, and external thread L: Anchor bolt embedded length
type mechanical expansion anchor bolts
Box-out anchor bolts Placement distance between boxes (A) that is at or
greater than 10cm
87
(a)
(b) (c)
FIG. IV.5. Pit placement standard
TABLE IV.2. (i) Reduction ratio of the TABLE IV.2. (ii)Reduction ratio of the
allowable load for pulling out allowable load for pulling out
according to the placement according to the placement
distance of embedded L and distance of embedded J and
LAtype anchor bolts and JAtype anchor bolts,
chemical expansion anchor headattached bolts and
bolts. externalthreadtype mechanical
expansion anchor bolts.
Number of anchor Reduction ratio () Number of anchor Reduction ratio ()
bolts bolts
2 bolts 1 P 2 bolts 1 P
2 + 80 2.5 + 5
100 d 10 L
3 or 4 bolts 1 P 3 or 4 bolts 1 P
6 + 40 5
100 d 10 L
(Note) 1. P: Placement distance of the (Note) 1. P: Placement distance of the
anchor bolt anchor bolt
d: Nominal diameter of the L: Embedded length of the anchor
anchor bolt bolt
2. It is to be 10d>P>5d. 2. It is to be 2L>P>L.
88
TABLE IV.2. (iii) Reduction ratio of the
allowable load for pulling out
according to the distance
between boxes of box-out
anchors.
Number of anchor Reduction ratio ()
bolts
2 bolts A
10
4 bolts
(Note) 1. A: Distance between boxes of
box-out anchors (cm)
2. It is to be 10cm>A>5cm.
FIG. IV.6. Reduced pit placement.
The allowable shear force for anchor bolts that have been placed at the corner or at
the side of the foundation is described in Fig. IV.7.
Placing location: A firm foundation (corner, side)
FIG. IV.7. Shear force for anchor bolts that have been placed in the corner or side of
foundations.
89
Determine the bolts allowable shear force Qa with the equation that appears below,
and take the smaller value.

Qa= d2fs
4
Qa=3C (C+d)p
Where d: The anchor bolts nominal diameter (cm)

fs: The allowable shear stress of anchor bolts in pure shear mode
C: The distance from the centre of the anchor bolt to the foundations side (cm)
d
(C >5cm)
2
p: The correction factor based on the specified design strength of concrete

1 Fc Fc
p= Min , 5+
6 30 100
Fc: The specified design strength of concrete (daN/cm2)
Note 1. It is suggested to be L>6d. (d: anchor bolt nominal diameter)

2. It is to be h>C. Calculations are to be made with the (3.20) equation if h>C.
IV.4.3. Examples
Fig. IV.8 shows some examples of anchor bolt types. The allowable load for pulling
out of these anchor bolts needs to be higher than that of the allowable load for pulling out
embedded headattached bolts.
FIG. IV.8. Example of effective anchor bolts.
90
IV.4.4. Capacities
a) Ltype bolts
The strength of Ltype bolts is determined on the basis of its bond strength. Due to the fact
that a weakening of bond strength can be expected over many years of use in locations that
are exposed to vibrations, it is suggested to be careful to use safer values.
b) When hanging heavy structure on ceiling slabs or walls, it is necessary to examine not only
the strength of anchor bolts, but it is also suggested to check the slab bearing capacity
REFERENCE
[IV.1] THE JAPANESE ENGINE GENERATOR ASSOCIATION, Guideline Of Seismic

Design, Tokyo, 1999.
91
Appendix V
SIMPLE SEISMIC DESIGN PRACTICE FOR PIPING SYSTEMS THE

CONSTANT PITCH DESIGN METHODOLOGY
V.1. OUTLINE OF SIMPLIFIED SEISMIC EVALUATION PRACTICE OF PIPING
Usually, seismic evaluation of piping is carried out with sophisticated methodologies.

However, when a simplified seismic evaluation is needed, it can be easily carried out through
approximate methods based upon the control of the span length between piping supports and
the capacity to absorb relative displacements [V.1-V.3]. This method is called the constant
pitch design method.
V.2. GENERALITIES ON THE SIMPLIFIED SEISMIC EVALUATION FOR PIPING
A procedure for a simplified seismic evaluation for piping is as follows:
(1) Seismic capacity of piping is evaluated by the span length between piping
supports.
(2) Evaluation of allowable span is carried out:
for the maximum span length of piping
for the span length of piping with concentrated mass.
(3) Evaluation of the capacity of accommodating relative displacements is
performed based on the span length between both ends supported by different
structures, or on the span length to the first support of tee with a branch, whose
diameter is smaller than the half of diameter of the main piping.
(4) Steps (2) and (3) are performed for each direction, two horizontal and vertical.
(5) Evaluation on capacity of absorbing displacements associated with weight and
thermal effects is out of the scope of this evaluation. In the case of piping
containing bellows, capacity of absorbing displacement on bellows should be
evaluated.
The flow chart of simplified seismic evaluation for piping is shown in Fig. V.1.
92
FIG. V.1. Flow of simplified piping design [V.4].
93
V.3. CALCULATION OF ALLOWABLE SPAN LENGTH
V.3.1. Criteria for constant pitch design method
L La
Where, L: Span Length, La: Allowable Span Length.
However, this span length needs to be checked against relative displacements.
V.3.2. Allowable span length
Allowable span length is calculated according to the following rules. The allowable
span length, without considering an additional mass and a concentrated mass, is called basic
allowable span length.
(1) Basic allowable span length values are shown in Table V.1 and V.2. Numerical
diameters in these tables are based on some typical standards. For other sizes,
basic allowable span length can be corrected linearly according to the real size.
However, basic allowable span length of diameter under 48.6 mm is the value
for diameter: 48.6 mm, and basic allowable span length of diameter over 609.6
mm is the value for diameter: 609.6 mm. The applicable range of these tables is
for piping less than 1000 mm diameter.
(2) In the case of an additional mass or a concentrated mass, the basic allowable
span length should be multiplied by Fnd and Fnc to the basic value according to
eq. (V.1) and eq. (V.2) respectively.
(3) In the case of an additional distributed mass, compensation coefficient Fnd is
calculated by the following equation.
0.25

Fnd = 1 + (V.1)

p
or the following value may be taken
Fnd =1.0 at /p10.5,
where, Fnd : compensation coefficient for additional weight

p : total weight of both weight of piping and weight of fluid in piping per unit
length.(N/m)
: additional weight per unit length.(N/m)
(4) In the case of a concentrated weight such a valve, the compensation coefficient
Fnc is obtained by Fig. V.2 or Table V.3. Fig. V.2 shows compensation
94
coefficient for an concentrated weight Fc, and rate of additional weight: rw.
Rate of additional weight: rw is calculated by the following equation. However
Fc=1.0 in the case rw 10.25.
0.75
w
rw = 1 + (V.2)
Wa p
where, rw : rate of additional weight

w : concentrated weight on the span of piping
Wa : basic concentrated weight (referred to Table V.1 or Table V.2)
TABLE V.1. BASIC ALLOWABLE SPAN LENGTH (=0.5)

Nominal Diameter Diameter Basic Allowable Span Length: Basic Concentrated
(in) (mm) (mm) La (m) Weight: Wa(N)
1 40 48.6 6.6 407
2 50 60.5 7.1 605
2 65 76.3 7.9 1,116
3 80 89.1 8.6 1,545
3 90 101.6 9.0 1,986
4 100 114.3 9.5 2,532
5 125 139.8 10.2 3,802
6 150 165.2 10.8 5,357
8 200 216.3 12.2 9,629
10 250 267.4 13.2 15,208
12 300 318.5 14.2 22,361
14 350 355.6 15.0 28,851
16 400 406.4 16.0 40,325
18 450 457.2 16.8 53,612
20 500 508.0 17.6 67,633
22 550 558.8 18.4 83,563
24 600 609.6 19.1 103,946
: Density of fluid: = 1.0 for water
TABLE V.2. BASIC ALLOWABLE SPAN LENGTH (GAS)

Nominal Diameter Diameter Basic Allowable Span Length: Basic Concentrated
(in) (mm) (mm) La (m) Weight: Wa(N)
1 40 48.6 7.0 304
2 50 60.5 7.8 445
2 65 76.3 8.7 839
3 80 89.1 9.5 1,126
3 90 101.6 10.1 1,414
4 100 114.3 10.7 1,775
5 125 139.8 11.7 2,616
6 150 165.2 12.7 3,616
8 200 216.3 14.8 6,349
95
10 250 267.4 16.4 9,863
12 300 318.5 18.0 14,281
14 350 355.6 19.0 18,110
16 400 406.4 20.3 25,339
18 450 457.2 21.5 33,995
20 500 508.0 22.7 41,112
22 550 558.8 23.8 51,143
24 600 609.6 24.9 64,243
1.0
0.5
0
0 1 2 3 4 5
rw
FIG. V.2. Compensation coefficient for concentrated weight.
TABLE V.3. COMPENSATION COEFFICIENT FOR CONCENTRATED MASS

Range of rw Compensational coefficient for concentrated weight:
Fnc
rw 0.25 c = 1.0
0.25 < rw .1 c = 1.130.53 rw
1 < rw c = 0.6360.036 rw
V.4. EQUIVALENT SPAN LENGTH
Equivalent span length of othertypes of piping than the straight one is calculated
according to the following rules:
(1) Equivalent span length of 3-dimensional one (two horizontal directions and one
vertical direction) should be evaluated for 3 directions of the seismic motions.
(2) Equivalent span length is calculated along an axis of piping between the nearest
piping supports.
(3) When an axis of piping is parallel to a direction of seismic motion, equivalent
span length can be calculated without the span of the first leg of the system,
that is, L1 in Fig. V.3.
96
L1
S1
L2
Direction of seismic
S2
motion
FIG. V.3. Equivalent span length for Lshape.
In Fig. V.3, when piping supports S1 and S2 support for direction of seismic motion,
equivalent span length is calculated without considering span length L1, where L is along the
axis of the first leg parallel to the direction of seismic motion, then L = L2.
(4) When there are different diameters of piping in evaluated piping systems, the
equivalent span length is calculated by the following equation.
d
L = l + l1 (V.3)
d1
where,
L: equivalent span length (m)

d: the maximum diameter of evaluated pipings
d1: diameter of specified piping
l: allowable span length for diameter d and
ll: allowable span length for diameter d1.
(5) In case of piping with tee as shown in Fig. V.4, the following is applied:
(a)When a diameter of a branched leg is larger than one half diameter of the
main piping, (L1 + L2), (L1 + Lb) and (L2 + Lb) should be shorter than the
allowable span length. In Figure V.4, piping supports, S1, S2 and Sb provide
support to directions perpendicular to axes of each piping. Moreover, L1, L2 and
Lb are calculated according to items (1) to (4).
a) (b)When a diameter of a branched leg is smaller than one half diameter of
the main piping, both (L1 + L2) and Lb should be shorter than the allowable
span length.
97
S1
ll
Sb
l2
lb
S2
FIG. V.4. Equivalent span length for teeshape piping.
V.5. CALCULATION OF ALLOWABLE RELATIVE DISPLACEMENT OF SUPPORTS
V.5.1. Criteria for allowable displacement
In case both ends are supported by different supporting structures including piping
with tee, evaluation of allowable displacement needs to be performed. Also, in the case of
piping with tee, evaluation needs to take into account the effect from a branch point to the first
support, where its diameter is smaller than one half of diameter of the main piping.
Evaluation of allowable displacement is performed by the following equation.
Ds Da (V.4)
where,
Ds: Relative displacement between supports or between a branch and the first
support in the direction of seismic motion
Da: Allowable displacement in the direction of seismic motion, which is
calculated by Eq. (E-5)
V.5.2. Allowable displacement of piping
Allowable displacement of piping is calculated by the following equation.
Da = Lj. f (V.5)
where
Da: Allowable displacement in the earthquake direction
98
Lj: Projective span length (in mm) in a plane perpendicular to the seismic
motion and that can be calculated according to the method described in
chapter 4.4
f: Allowable displacement per unit length (mm), calculated by the following
equation
C y Lj
f = (V.6)
D
where,
C: Coefficient is defined by allowable strain of piping material, and the value can
be used up to yield strain; C=0.67,
: Yield strain of piping material is calculated by the following equation.
Sy
= (V.7)
E
where,
Sy: Yield stress or 0.2% strength of material at the design temperature

E: Youngs modulus of material at the design temperature
D: Diameter of piping.
V.6. PROJECTIVE SPAN LENGTH
Projective span length is calculated according to the following fundamental rules:

(1) Projective span length of two horizontal directions is evaluated to two seismic
motions.
(2) Projective span length: Lj is span length along an axis of piping between the
adjacent supports in plane which is perpendicular to seismic direction.
(3) When they have parts of different diameters in evaluated piping, projective
span length is calculated by the following equation.
d
Lj = l + l1 (V.8)
d1
where,
Lj: projective span length(m)
d: the maximum diameter of evaluated piping
d1: diameter of specified piping
l: projective span length for diameter : d (m)
99
ll : projective span length for diameter : d1(m)
(4) In case of piping with tee as shown in Fig. V.4,

(a) When the diameter of a branched leg is larger than one half of diameter
of the main piping, evaluation of allowable displacement needs to meet
the following conditions:
Ds (12) Da(12), Ds (1b) Da(1b), and Ds (2b) Da(2b)
where,
Lj(12): projective span length of (L1+L2)
Lj(1b): projective span length of (L1+Lb)
Lj(2b): projective span length of (L2+Lb)
Da(12): allowable displacement of (L1+L2)
Da(1b): allowable displacement of (L1+Lb)
D a(2b): allowable displacement of (L2+Lb)
Ds(12): relative displacement between S1 and S2
Ds(1b): relative displacement between S1 and Sb
Ds(2b): relative displacement between S2 and Sb
In Fig. V.4, piping supports, S1, S2 and Sb support for perpendicular directions to the
axes of piping. Moreover, L1, L2 and Lb are calculated according to items (1) to (4).
(b) When the diameter of a branched leg is smaller than one half of diameter
of the main piping, evaluation of allowable displacement needs to meet
the following conditions,
Ds(12) Da(12)
Ds(1b) + Ds(2b) 4 (V.9)
and ( ) Da(Tb)
L12
2 + 20( )
L
where,
Da(Tb): allowable displacement of the third leg of tee

L: allowable span length of the main piping
L12: span length of the main piping.
V.7. CALCULATION OF RELATIVE DISPLACEMENT
The displacement of support to be used for an evaluation against seismic motion is

the relative displacement between two supporting structures of piping.
Relative displacement of evaluated span length: Ds is calculated by the following

equation.
100
Ds = D1+D2 (V.10)
where,
Ds: relative displacement of evaluated span length

D1: displacement of support 1 under seismic event
D2: displacement of support 2 under seismic event
REFERENCES
[V.1] SHIBATA, H., A proposal for an aseismic design method of equipment and piping
for NPPs in a low seismicity area. Nuc. Eng. Des. 77, 2 (1984) 169180.
[V.2] SHIBATA, H., On anti-earthquake design procedure of equipment and pipings in
near future, in: Proc. Sixth Int. Conf. Structural Mechanics in Reactor Technology,
Paris, (1981), paper K 13/1.
[V.3] SHIBATA, H., On a method of evaluation of anti-earthquake design code of
industrial facilities, Proc. Seventh World Conf. on Earthquake Engineering, Istanbul,
Turkey (1980), paper 4/613.
[V.4] THE JAPAN CHEMICAL ENGINEERING ASSOCIATION, Manual for Seismic
Design, Tokyo, March 1999
101
Appendix VI
DETAILS OF CONNECTING CABLE AND PIPING BETWEEN COMPONENTS
Some suggested good practices detailing connecting cable and piping between
components are given in Figures VI.1.VI.3.
FIG. VI.1. Connection between buildings.
102
FIG. VI.2. Connections for buried cables and pipes.
103
FIG. VI.3. Connections to equipment.
104
Appendix VII
SEISMIC SCRAM SYSTEMS
VII.1. ISSUES AFFECTING THE DECISION TO INSTALL AUTOMATIC SEISMIC TRIP

SYSTEM (ASTS) IN A POWER PLANT
The following issues play a significant role in the decision for an installation of an
automatic seismic trip system (ASTS) in an NPP:
level, frequency and duration of earthquake activity at the facility site;

seismic capacity of facilities structures, equipment and distribution systems;
safety considerations related to spurious scrams;
expected time of seismic scram and comparison of this time to the expected time
for reaching the strong motion part of the earthquake time history;
economic impact of shutting down the facility to the country; or to the research.
issues related to public acceptance of the seismic safety of the facility;
level of operator confidence and reliability.
It is suggested to take these issues into consideration prior to deciding the installation
of an ASTS.
VII.2. ELEMENTS FOR THE SETTING OF THE TRIGGER LEVEL FOR ASTS
Two trigger levels are generally used for ASTS. However for research facilities lower
levels may be selected.
The first level is chosen to be close to the SL-1 level, as defined in [VII.1].
Significant structures, equipment and distribution systems of the NPP are generally expected
not to fail at levels lower than the SL-1 earthquake.
The second level may be close to the SL-2 level [VII.1]. At this level, it may be
possible to have the failure of some parts of the core internals and control rod insertion
mechanism within a few seconds after the seismic scram signal. In cases where such problems
are encountered, alternate ways of scramming the reactor may be considered.
It needs to be noted that earthquake causing the triggering of the second level settings
would have the potential for large scale destruction in the site vicinity, including loss of offsite
power and disruption in the normal cooling water supply.
Settings lower than SL-1 may be selected for interim periods in cases where seismic
capacity assessment and upgrading works for the NPP are ongoing.
Considering the inherent seismic resistance of facility structures, equipment and

distribution systems, lower bound setting level of seismic scram would be suggested.
105
VII.3. ELEMENTS TO BE CONSIDERED IN THE DESIGN OF ASTS
ASTS sensors and associated circuitry need to be designed to withstand the seismic
excitation and to prevent any malfunctioning induced by seismic loads.
Typically free field and foundation level are chosen for locating the sensors.
The ASTS needs to be considered as an element in the safety analysis of the facility.
Accordingly, the system needs to be treated as part of the safety protection system of the plant
and needs to conform to all relevant requirements.
VII.4. GUIDELINES FOR NUCLEAR PLANT RESPONSE TO AN EARTHQUAKE
Plant response to an earthquake needs to be managed regardless of the decision to

install an ASTS. The control room operator needs to record the occurrence of an earthquake
by any information source and needs to follow the plant response monitoring the seismic
instrumentation.
Later, an evaluation of recorded earthquake motion will need to be compared with

the design basis of structures, systems and components, a walk-down evaluation of the
damage experienced at the facility will need to be carried out and an evaluation of the
readiness of the plant to resume operation following the earthquake occurrence will need to be
completed.
In relation to this concern, specific post-event procedures need to be developed,

particularly for the identification of the items to be inspected, for the evaluation of the damage
related quantities and for the involvement of the regulatory body in the decision for resuming
operation [VII.2VII.5].
REFERENCES
[VII.1] INTERNATIONAL ATOMIC ENERGY AGENCY, Earthquakes and Associated

Topics in Relation to Nuclear Power Plants Siting, Safety Series No. 50-SG-S1
(Rev. 1), IAEA, Vienna (1991).
[VII.2] EPRI Report NP-6695, Guidelines for Nuclear Power Plant Response to an
Earthquake, December 1989.
[VII.3] EPRI Report NP-5930, A Criterion for Determining Exceedance of the Operating
Basis Earthquake, July 1988.
[VII.4] EPRI Report TR-100082, Standardization of the Cumulative Absolute Velocity,
December 1991.
[VII.5] EPRI Report TR-104239, Seismic Instrumentation in NPPs for Response to OBE
Exceedance: Guidelines for Implementation, June 1994.
106
CONTRIBUTORS TO DRAFTING AND REVIEW
Alcala-Ruiz, F. International Atomic Energy Agency

Celebi, M. US Geological Survey, United States of America
Chander, H. US Department of Energy, United States of America
Contri, P. International Atomic Energy Agency
Di Meglio, A.F. United States of America
Fukuda, T. TEPCO, Japan
Kojima, N. Mitsubishi Heavy Industry, Japan
Koshiro, En. Obayashi Corporation, Japan
Lebelle, M. Commissariat lEnergie Atomique, France
Shibata, H. Nihon University, Japan
Sollogoub, P. Centre dEtudes de Saclay, France
Tomura, H. International Atomic Energy Agency
Consultants Meetings
Vienna, Austria: 913 November 1998

Vienna, Austria: 15 March 1999
Vienna, Austria: 1014 July 2000
107

Te 1347 Web PDF

Uploaded by

Copyright:

Available Formats

Te 1347 Web PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Te 1347 Web PDF

Uploaded by

Copyright:

Available Formats

IAEA-TECDOC-1347

CONSIDERATION OF EXTERNAL EVENTS IN THE DESIGN OF NUCLEAR FACILITIES

This publication is related to IAEA-TECDOC-348 Earthquake Resistant Design of

The contributions of M. Lebelle (France) and T. Fukuda (Japan) are acknowledged.

9. OTHER CONSIDERATIONS ............................................................................................. 65

APPENDIX I: SIMPLIFIED PROCEDURES FOR ASSESSMENT OF

APPENDIX II: SOIL STRUCTURE INTERACTIONS.......................................................... 76

APPENDIX III: EARTHQUAKE RESISTANT DESIGN PROVISIONS.............................. 82

APPENDIX IV: COMPUTATION OF PULLING FORCE ON ANCHOR DEVICE ............ 84

APPENDIX V: SIMPLE SEISMIC DESIGN PRACTICE FOR PIPING

APPENDIX VI: DETAILS OF CONNECTING CABLE AND

APPENDIX VII: SEISMIC SCRAM SYSTEMS .................................................................. 105

CONTRIBUTORS TO DRAFTING AND REVIEW............................................................ 107

In recent years the development of design criteria, design methodologies and

Requirements for NPP site evaluation [16].

That publication presented simplified siting and design methods aiming at

A common approach for NFOP design is the acceptance of a higher conservatism in

Concerning the applicability of this report to facilities with high radioactive

Further guidance on the appropriate application of the simplified approaches

External events affecting a nuclear installation could be listed as in the following,

(a) Human induced

However, particularly in case of natural events, some scenarios need to be treated as

Furthermore, considered external human induced events are of accidental origin.

In conclusion, in the following, main emphasis is given to design against earthquake,

Appendices present experience of some States in design and qualification of

2. SAFETY OBJECTIVES AND CLASSIFICATION

2.1. INTRODUCTION AND EXAMPLES OF NFOPs

TABLE I. EXAMPLES OF NFOPS WITH ASSOCIATED MAIN SAFETY ISSUES (source:

Fuel Process bldg Usually 100 Criticality

Fuel Process bldg + plant Usually 100 Criticality Confinement in case

Fuel Process bldg + waste Same return As above. Large quantities of

Concerning research reactors, data are available in Refs [30, 31].

2.2. SAFETY OBJECTIVES

The technical safety objective is to take all reasonably practicable measure to

The consequences to workers and public are discussed in Ref. [12].

In the case of workers, the consequences need to be expressed in terms of

2.3. DEFENCE IN DEPTH

However, some guidelines can be provided for a reasonable implementation of the

2.4. SAFETY CLASSIFICATION OF THE FACILITIES

In the framework of this report, mainly oriented to an identified group of facilities

However, a full probabilistic approach in the classification of NFOPs in relation to

In many cases, it is convenient to develop some conservative assumptions in order to

2.5. CLASSIFICATION OF ITEMS IN RELATION TO EXTERNAL EVENTS

(a) Radiological criteria, such as:

(b) Performance criteria, including

In probabilistic terminology, the deterministic concept of safety margin needs to be

Classification is related to hazard and not to the required

FIG. 2. Interaction between item classification and acceptability limits.

This approach is more in line with the availability of sophisticated calculation

In relation to the component design, an interaction could be stated between the

TABLE II. RELATIONSHIP BETWEEN FACILITY CLASSIFICATION AND ITEMS

DESIGN CLASS SELECTION

Hazard High Moderate Low Conventional

EE Class 1 Design Class 1 Design Class 2 Design Class 3 Design Class 4

EE Class 2 Design Class 2 Design Class 3 Design Class 3 Design Class 4

EE Class 3 Design Class 4 Design Class 4 Design Class 4 Design Class 4

These common aspects define also a sort of limit of applicability of the

3.2. SAFETY MARGIN CONTROL IN THE DESIGN PROCESS

Hazard Design Load Design Material Constr. Release