ML20212N054

Regulatory Analysis for Resolution of Unresolved Safety Issue A-46,SEISMIC Qualification of Equipment in Operating Plants

ML20212N054
Person / Time
Issue date:	02/28/1987
From:	Anderson N, Chang T Office of Nuclear Reactor Regulation
To:
References
REF-GTECI-A-46, REF-GTECI-SC, TASK-A-46, TASK-OR NUREG-1211, NUDOCS 8703120330
Download: ML20212N054 (73)

Text

.

'~

~

NUREG-1211

~

Regulatory Analysis for Resolution of Unresolved Safety Issue A-46, Seismic Qualification of Equipment in Operating Plants U.S. Nuclear Regulatory Commission Offica of Nuclear Reactor Regulation T. Y. Chang, N. R. Anderson fa arooy 1

i R M 2E8u 8 *7 22" 1211 R PDR

g..

NOTICE Availability of Reference Materials Cited in NRC Publications Most documents cited in NRC publications will be available from one of the following sources:

1. _ The NRC Public Document Room,1717 H Street, N.W.

Washington, DC 20555

2. The Superintendent of Documents, U.S. Government Printing Office, Post Office Box 37082, Washington, DC 20013 7082

3. The National Technical Information Service, Springfield, VA 22161 Although the listing that follows represents the majority of documents cited in NRC publications, it is not intended to be exhaustive.

Referenced documents available for inspection and copying for a fee from the NRC Public Docu-ment Room include NRC correspoodence and internal NRC memoranda: NRC Office of Inspection and Enforcement bulletins, circulars, information notices, inspection and investigation notices; Licensee Event Reports; vendor reports and correspondence; Commission papers; and applicant and licensee documents and correspondence.

' The following documents in the NUREG series are available for purchase from the GPO Sales Program: formal NRC staff and contractor reports, NRC-sponsored conference proceedings, and NRC booklets and brochures. Also available are Regulatory Guides, NRC regulations in the Code of Federal Regulations, and Nuclear Regulatory Commission issuances.

Documents available from the National Technical Information Service include NUREG series reports and technical reports prepared by other federal agencies and reports prepared by the Atomic Energy Commission, forerunner agency to the Nuclear Regulatory Commission.

Documents available from public and special technical libraries include all open literature items, such as books, joumal and periodical articles, and transactions. Federal Register notices, federal and state legislation, and congressional reports can usua!ly be obtained from these libraries.

Documents such as theses, dissertations, foreign reports and translations, and non-NRC conference proceedings are available for purchase from the organization sponsoring the publication cited.

Single copies of NRC draft reports are available free,to the extent of supply, upon written request to the Division of. Technical Information and Document Control, U.S. Nuclear Regulatory Com-mission, Washington, DC 20555.

Copies of industry codes and standards used in a substantive manner in the NRC regulatory process are maintained at the NRC Library, 7920 Norfolk Avenue, Bethesda, Maryland, and are available there for reference use by the public. Codes and standards are usually copyrighted and may be purchased from the originating organization or, if they are American National Standards, from the American National Standards Institute,1430 Broadway, New York, NY 10018.

NUREG-1211 Regulatory Analysis for Resolution of Unreso ved Safety Issue A-46, Seismic Qualification of Equipment in Operating Plants n"t z r e i;?ctr "

T. Y Chang, N. R. Anderson Division of Safety Review and Oversight Office of Nuclear Reactor Regulation U.S. Nuclear Regulatory Commission W:shington, DC 20666 g

ABSTRACT The margin of safety provided in existing nuclear power plant equipment to-resist seismically induced loads and perform required safety functions may vary considerably, because of significant changes in design criteria and methods for the seismic qualification of equipment over the years. Therefore, the seismic.

qualification of equipment in operating plants must be reassessed to determine whether-requalification is necessary.

The objective of technical studies performed under Task Action Plan A-46 was to establish an explicit set of guidelines and acceptance criteria to judge the adequacy of equipment under seismic loading at all operating plants, in lieu of requiring these plants to meet the criteria that are applied to new plants.

This report presents the regulatory analysis for Unresolved Safety Issue (USI)

A-46.

It includes (1) Statement of the Problem, (2) the Objective of USI A-46, (3) a Summary of A-46 Tasks, (4) a Proposed Implementation Procedure, (5) a Value-Impact Analysis, (6) Application of the Backfit Rule, 10 CFR 50.109

-(7) Implementation and (8) Operating Plants To Be Reviewed to USI A-46 Requirements.

NUREG-1211 iii i-

TABLE OF CONTENTS P, age ABSTRACT..............................................................

iii EXECUTIVE

SUMMARY

ix

- I Statement of the Problem.........................................

1 II Objective of USI A-46............................................

1 III Summary of A-46 Tasks............................................

~1 IV Proposed Implementation Procedure................................

2-

1. Plants Affected...............................................

3

2. Scope of Seismic Adequacy Review..............................

3

3. Requirements for Plant Shutdown...............................

5

4. General Verification Procedure for Plant-Specific Review.........................................

6 A.-Development of' Equipment List..............................

6 B. Comparison of Site Spectra With Appropriate Bounding Spectra...........................................

6 C. Wal k-Through In specti o n....................................

9 D. Review of Equipment Functional Capability..................

11 E. Review of Equipment Unique to Nuclear Plants...............

12 F. Replacement Parts..........................................

12 G. Verification of Anchorage..................................

12 5.

Generic Resolution...........................................

13 6.

Provisions for Resolution for Individual Utilitics...........

15 V

Value-Impact Analysis............................................

16

1. Qualitative Assessment of Safety Benefit...................

16

2. Quantitative Examples of Potential Value Impact...............

18 A. Seismic Failure of an Electrical Cabinet Anchorage.........

19 B. Examples from USI A-45 Analysis............................

20

3. Consideration of Alternatives.................................

24

4. Costs of Alternatives.........................................

25

5. Estimated Costs to Licensees..................................

28 l

6. Costs to the NRC..............................................

29

7. Safety Benefits Compared to Costs.............................

31

8. Impacts on Other Requirements.................................

31

9. Constraints...................................................

31 NUREG-1211 v

TABLE OFLCONTENTS (Continued)

Page s

VI Application of the Backfit Rule, 10 CFR 50.109....................

31 VII Implementation....................................................

34

1. Proposed Method of Implementation..............................

. 34

2. Schedule for Implementation of the Proposed Requirement......................................................

34' 3.-Relationship to Other Existing or Proposed Requirements........

35 VIII References........................................................

35-

~

Enclosure I: Operating Plants To Be Reviewed to USI A-46

. Requirements Appendix A: Generic Letter i

l L

NUREG-1211 vi I

4 LIST OF FIGURES

-Figure Page 1-USI A-46 implementation..........................................

7 LIST OF TABLES Table Page 1

Typical equipment list for USI A-46..............................

8 2

Cost estimates of seismic verification alternatives..............

26 3

Estimated costs to a licensee....................................

30 4-Estimated time for implementation;of the proposed requirement....

34-s.

v 2

NUREG-1211 vii

EXECUTIVE SUMARY cThefssueofSeismiE: Qualification of Mechanical an'd Electrical Equipment.in i

Operating. Nuclear Power Plants was designated as an Unresolved Safety Issue (USI).in. December 1980. The safety concern was that equipment in nuclear plants with construction permit (CP) applications _ docketed before about 1972 has not

'been reviewed according to the current licensing criteria for. seismic qualifi-

- cation of. equipment (Regulatory Guide 1.'100; Institute of Electrical and Elec-tronics Engineers (IEEE) Standard.344-1975 and Standard Review Plan Section 3.10

.(NUREG-0800)).

herefore, this equipment may not have been adequately qualified T

to ensure its survival and functionability in the event of a_ safe-shutdown earthquake (SSE).

Plants with a CP application docketed after about 1972 have been qualified according to the-current licensing criteria and their compliance has been audited by the NRC staff..All plants whose compliance with the current-

. licensing criteria could not be verified are subject to the implementation provisions outlined in.this report.

The NRC staff has determined that it is not feasible to require ' older operating

-plants to meet current licensing requirements.

Therefore, a number of alterna-tive procedures were investigated.

The alternative selected, which forms the basis for the implementation provisions outlined in this report, was the use of earthquake experience data, supplemented by test experience data.

This compila-

. tion of earthquake and test experience data was used to develop a data base,-

With appropriate rules and restrictions, that can be used to verify the seismic capability of most required equipment below certain specified bounds of earth-quake motion. This alternative, in the staff's judgement, provides the most reasonable and cost effective means of ensuring that the purpose of GDC 2 of d

10 CFR Part 50, Appendix A is met.

Development of a seismic experience data base to address USI A-46 was suggested by the Seismic Qualification Utility Group (SQUG) in 1982. -SQUG and its con-tractors performed a pilot study to determine the feasibility of using actual t

earthquake experience data to evaluate-the susceptibility of nuclear power plant equipment to seismic loads. The SQUG concluded, and the NRC agreed, that the use of experience data was feasible.

In 1983, the SQUG proposed the formation of a panel of consultants, the Senior Seismic Review and Advisory Panel (SSRAP),

to independently assess the feasibility of using experience data and to provide expert advice and consultation.

In 1984, the Electric Power Research Institute (EPRI) began an effort to collect and evaluate existing seismic test data on nuclear plant equipment for USI A-46.

The SQUG, EPRI and SSRAP investigations were closely monitored by the NRC staff.

1, On the basis of information developed by the SQUG, with assistance from the Electric Power Research Institute (EPRI), the safety concern was limited to (1) equipment anchorages, (2) functionality of equipment (principally electrical relays) during the period of strong shaking, and (3) outliers (i.e., equipment configurations or locations that preclude their applicability to the experience data base).

I NUREG-1211 ix i

I

1 The NRC has developed implementation requirements to be issued in a generic letter.

In the staff's judgement, the seismic adequacy review and the correc-tion of all deficiencies as required by the generic letter are backfits as defined in 10 CFR 50.109.

The proposed resolution allows for a generic approach to implementation.

That is, the SQUG, with help from EPRI and its consultants, is developing generic guidelines for participating utilities.

Each utility will be required to submit a plant-specific report and will be subject to review and/or audit by the SSRAP 1

and the NRC.

The NRC has formed electrical relay and equipment anchorage review groups that have participated in the development of the requirements and that will perform an audit function during implementation.

The section on implementation procedures in this report is intended to serve as guidance to the SQUG for the development of detailed implementation procedures for the plant-specific reviews.

If a utility that is subject to the implementation requirement elects to not participate in SQUG, that utility must develop its own procedure using the guidance provided herein.

NUREG-1211 x

(

f I

~

REGULATORY ANALYSIS FOR RESOLUTION OF UNRESOLVED SAFETY: ISSUE A-46, SEISMIC QUALIFICATION OF EQUIPMENT IN OPERATING PLANTS I. ~ STATEMENT OF THE PROBLEM The design criteria-and methods for the seismic' qualification of mechanical and electrical equipment in nuclear. power plants have undergone significant change during the history of the. commercial: nuclear power program.

Consequently,'the margins of safety provided in existing equipment to allow the equipment to resist seismically induced loads.and perform its intended safety functions'may vary considerably.< The seismic capability of equipment inioperating plants therefore

~

must be reassessed to ensure _its capability to bring the plant,to a safe shutdown condition when subjected to a seismic event.

-The need for.such alreassessment was identified as a result of: experience with

.the Systematic Evaluation Program (SEP) for 11 older operating plants and the'

-staff's Seismic Qualification Review Team (SQRT) reviews of operating license

. applications.

During-the course of the SEP and SQRT reviews,-the staff identi-fied a concern with_the anchorage and. supports for electrical equipment in the

. SEP plants.

An information notice concerning this issue was sent to all operat-ing plants (NRC May 1980).

The Unresolved Safety Issue (USI) A-46 program investigated the adequacy of seismic qualification methods used for electrical and mechanical equipment installed in older nuclear plants. This investigation determined that it is necessary to develop proposed requirements which could be implemented in a

' practical, cost-beneficial way to ensure that equipment in older plants can adequately withstand a seismic event and to ensure the capability to safely

~

shut down the plant.

II. OBJECTIVE OF USI A-46 The proposed regulatory requirement is needed to verify the seismic adequacy of mechanical and electrical equipment which is required to safely bring the reactor

- and plant to a safe shutdown condition and to maintain it in a safe condition.

The specific objective of the A-46 task was to develop viable, cost-effective alternatives to _ current seismic qualification licensing requirements to be applied to operating nuclear power plants.

III.

SUPNARY OF A-46 TASKS

~A-46 tasks -included investigation of several alternative procedures for ensuring the seismic adequacy of equipment needed to cope with a seismic event.

Some of the alternatives studied did not contribute significantly to the proposed resol-

-ution..

Each of the tasks is described in the A-46 technical findings report, NUREG-1030, and in the references cited in that report.

Tasks included in the A-46 program were.as follows:

LNUREG-1211 1

s s++3 l

(1) JIdentification of Seismic Sensitive Systems and Equipment The objective of this-task was to develop possible methods of generating a

~

. generic' minimum equipment list. :If a methodology could be developed:to

- i evaluatezthe _ risk.importance.of; safety systems and equipmentJthen equip-ment could be ordered by contribution to risk.

~

l

-(2)' Assessment of Adequacy of Existing Seismic Qualification Methods This task involved a study to evaluate past and present methods.to qualify-mechanical and electrical equipment.. The intent was to determine if older

~

qualification procedures could be shown to provide adequate assurance of-seismic adequacy.

~(3) Development and Assessment of In Situ' Test Procedures To Assist in

-Qualification of Equipment This task was intended to develop guidelines for obtaining and using dynamic characteristics-of equipment to assist in verifying seismic adequacy.

~(4) : Seismic Qualification of Equipment Using Seismic Experience Data This task was based on the experience data collected by the Seismic Qualifi-cation Utilities Group.(SQUG) and recommendations made by the Senior Seismic Review and Advisory Panel (SSRAP).

(5) Development 1of Methods To Generate Generic Floor Response Spectra This task led to the development of, and guidelines for using, generic floor response spectra. These generic spectra'can be used in lieu of calculating response spectra for use in determining seismic adequacy.

Task (4), the use'of seismic experience data supplemented by seismic test data, proved to be the most reasonable anf cost effective alternative for verifying seismic adequacy.

The other four tasks either play supporting roles or would l

be used to a limited extent if the seismic experience data base does not per-L tain to a particular item.

The'A-46 implementation plan presented in the fol-

[

lowing paragraphs therefore is based primarily on work completed in Task (4).

IV.

PROPOSED IMPLEMENTATION PROCEDURE l

Alternatives were developed and analyzed for the resolution of the issue. Only l

'three alternatives were considered:

(1) not impose a requirement (i.e., do nothing), (2)? impose current licensing requirements, and (3) require verification t

of seismic adequacy by comparison with the experience data base.

l The alternatives are evaluated in Section V of this report.

On the basis of

~the value-impact analysis in Section V, a decision was made to impose Alterna-tive 3, the use of seismic experience data.

A decision was made to impose the requirement by generic letter.

The staff also concluded that the seismic ade-i quacy review and the correction of all deficiencies as required by the generic letter are backfits as defined in 10 CFR 50.109.

NUREG-1211 2

' U-

~,,. _ _,... m o

<.L.

1..

On the basis of the results of the A-46 tasks summarized above and selection of a proposed resolution position, an implementation plan was developed.

Each licensee of an operating plant that has not been previously reviewed to current licensing criteria would be required to perform a seismic verification review and report the results. The verification review procedure is outlined below.

1.

Plants Affected The current requirements for qualification of equipment used in licensing plants are defined in Regulatory Guide 1.100, Institute of Electrical and Electronics Engineers (IEEE) Standard 344-1975, and Standard Review Plan (SRP) Section 3.10 (NUREG-0800). The importance of equipment support to the qualification of equipment is recognized in current requirements, as evidenced by the following statement:

"The equipment to be tested shall be mounted on the vibration gener-ator in a manner that simulates the intended service mounting. The mounting method shall be the same as that recommended for actual service." The staff believes that plants reviewed to current requirements, with the implementation audited by the Seismic Qualification Review Team (SQRT) as is currently done, have been confirmed to have an adequate level of protection for safe shutdown earthquake (SSE)-level seismic events.

All plants not reviewed to these current equipment qualification requirements, as documented by plant Safety Evaluation Reports (SERs) are included in the A-46 review.

For SEP plants, the structural integrity of equipment were generally covered under the seismic review; however, there are some SEP plants for which the equipment seismic adequacy was left to the resolution of USI A-46.

Most of the SEP plants need only evaluate equipment anchorage and functional capability.

The scope of review should be established in accordance with each plant's Inte-grated Safety Assessment Report and related Safety Evaluation Reports.

A list of plants affected is included as Enclosure I.

2.

Scope of Seismic Adequacy Review Each licensee will be required to determine the systems, subsystems, components, and instrumentation and controls needed during and following an SSE event using the following assumptions:

(1) The seismic event does not cause a loss-of-coolant accident (LOCA), a steam-line break accident (SLBA), or a high-energy line break (HELB), and a LOCA, SLBA, or HELB does not occur simultaneously with or during a seis-mic event.

(2) Offsite power may be lost dtring or following a seismic event.

(3) Plant must be capable of being brought to a safe shutdown condition following a design-basis seismic event.

The equipment to be included in this implementation plan is generally limited to active mechanical and electrical components.

Cable trays are included in the scope of A-46 review because the staff agrees with the SQUG that the long-standing issue of the seismic adequacy of cable trays can be addressed by using the seismic experience data, and it will be an expedient and efficient way to address this concern.

Piping, tanks, and heat exchangers are not included, except that those tanks and heat exchangers that are required to achieve and maintain safe shutdown must be reviewed for adequate anchorage.

Seismic system NUREG-1211 3

i interaction is-included in the scope of review to the extent that' equipment within the scope must be protected from seismically induced physical interaction with all. structures, piping, or equipment. located-nearby.

Lessons learned from studies of nuclear and non-nuclear facilities under earthquake loading indicate that.the effect of fai. lure of certain items such as suspended ceilings'and light fixtures could influence the operability of equipment within the scope of review.

'In addition,-there is some concern that the non-safety-related piping and struc--

tures in close proximity to, or located above, equipment needed for safe shut-down could fail inLsuch a way that it could physically-interact with the required equipment.

Instrument air lines and electrical and instrumentation cabling must be verified to have sufficient-flexibility from the connection to; equipment so that. relative movement of anchor points will not ' jeopardize their integrity.

Air lines and electrical and instrument cabling are not included in the scope of review except for that portion from the equipment item to the first anchor point. 'The failure of masonry walls that could affect the operability.of nearby safety-related equip 1;ent is also of concern.

However, this concern'has been-addressed by IE.Bulletin 80-11 dated May 1980, which requires that all such

. masonry walls be identified and re evaluated to confirm their design adequacy under postulated loads and. load combinations.

This concern is therefore not considered as part of A-46' implementation.

For some pressurized water reactor. plants, the seismic adequacy of auxiliary feed water (AFW) systems has been verified by licensee actions taken in response to Generic Letter 81-14 dated February 10, 1981.

Review of the AFW systems may be deleted from consideration under A-46 if staff acceptance has been documented in an=SER, or if the licensee has committed to meet the requirements of the generic' letter.

For-the purpose of this implementation plan, safe shutdown means bringing the plant to a' hot shutdown condition and maintaining it there for a minimum of 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />.' The 72-hour period is sufficient for inspection of equipment and minor repairs if necessary following an SSE or to provide additional source (s) of water for decay heat removal if needed to extend the time at hot shutdown.

Equipment required includes that necessary to maintain required supporting functions for safe shutdown.

For all equipment within.the defined scope, the verification should closely follow the procedure outlined in Paragraph 4 (General Verifica-tion _ Procedure for Plant-Specific Review) below.

4 i.

Studies are currently being done-as part of USI A-45, " Decay Heat Removal Re-quirements," to review the risk associated with shutdown and decay heat removal systems.

Part of the A-45 effort involves a study to determine the risk asso-I ciated with cold shutdown, including seismic risk.

This is a probabilistic risk assessment study and, as such, includes consideration of seismic ~ hazard well j.

above the SSE level (up to five times the SSE).

Seven~ plant-specific PRA studies will be conducted under A-45.

For each of these studies, plant-specific equip-i ment fragilities are being generated from plant inspections of the equipment.

These plant reviews are specifically looking for anchorage deficiencies a'nd offnormal equipment configurations.' Concerns regardino seismic qualification of cold shutdown equipment are best addressed under USi A-45.

If further A-45 studies show that there is an important reduction in core melt probability if '

L equipment required to reach cold shutdown is seismically qualified to the SSE level, the implementation of these results will be made separately under l

USI A-45.

Accident-mitigating systems were not included within the scope for two reasons:

NUREG-1211 4

. - - --. ~,- -..m y,,,,v.,m.,-m_m-~--e.

_.-,~--.-m.e

(1) Experience data collected by SQUG and others and high-level seismic tests on piping conducted in foreign countries and in the U.S. show that piping is not susceptible to failure resulting from seismic inertia loads.

The only. observed instances of piping failure during the SQUG program to col-1ect seismic experience data were due to relative movement of anchor points and inadequate or nonexistent anchorage of tanks or equipment for sites with zero period acceleration between 0.25g and 0.6g.

s

[

In general, piping is found to have a high margin of safety for almost all F

the piping-if.only seismically induced inertia loads are considered.

High

(

stresses arise where piping runs through walls or is attached to a large f

vessel resulting in relative displacements.

In piping design, seismic stresses are usually held to a small percentage (say 15%) of the overall i

allowable stress.

In addition, seismic risk studies completed to date show

(

that piping is not predicted to fail even at levels two to five times the a

SSE level.

i Furthermore, IE Bulletin 79-02 (NRC, March 1979) requires review of as-built pipe support base plate designs using concrete expansion anchor bolts.

IE F

Bulletin 79-07 (NRC, April 1979) requires review of the proper combination i

of the intramodal responses resulting from the spatial components of a L

multidimensional earthquake,'and the verification of piping system computer y

codes.

IE Bulletin 79-14 (NRC, July 1979) requires the confirmation of 7

"as-built" configuration of safety-related piping systems to their design /

i analysis configuration.

The piping systems, including their restraints, 7M*g, g

were reviewed to the requirements of these IE bulletins, and all operating lJ.J. H '

p plants either met these requirements or were modified to meet them.

g.I 9l.0h

?

V 1

L (2) Seismic experience data collected by SQUG and reviewed by SSRAP, supple-Q g

mented by reviews and literature surveys of strong motion earthquakes,

$p y indicate that mechanical and electrical equipment of the types commonly E

used in nuclear power plants are unlikely to fail at earthquake levels Oc typical of SSEs at U.S. plants east of. California. There is strong evidence M'

E that accident-mitigating systems would function as designed in the unlikely k}P.'

I event they are required following an SSE.

In almost all cases where equip-

%P h

ment damage has occurred, it resulted from failure of the anchorage or

$fG 7

from displacement of unanchored equipment.

It was also observed that some N;T y

equipment with minimal anchorage did not move even though it was subjected Kyi E

to accelerations as high as 0.5g.

jp

@d.Y 3.

Requirements for Plant Shutdown g;

The time a plant can remain at hot shutdown after a seismic event without re-hI.t h

storing offsite power is plant specific.

Each licensee must show a practical

4. ?

means of staying at hot shutdown for a minimum of 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />. If maintaining safe E-shutdown is dependent on a single (not redundant) component whose failure, either J.;n.

b due to seismic loads or random failure, would preclude decay heat removal by the m.u identified means, the licensee must show that at least one practical alterna-W e/

g tive for achieving and maintaining safe shutdown exists that is not dependent

${

g on that component.

y' E

.?

L The equipment to be considered depends on the functions required to be performed.

?~c

%w%

g Typical plant functions would include w

(1) bring the plant to hot shutdown and establish heat removal f{y ee 5

NUREG-1211 5

E-gg4 W -

c

'(2) maintain support systems necessary to establish and maintain hot shutdown j

(3)_Lmaintain control room fun:tions and instrumentation and controls necessary.

to monitor hot shutdown

' (4). provide alternating current-and/or direct current emergency power as nceded on a plant-specific basis to meet the above three functions -

4. LGeneral. Verification Procedure for Plant-Specific. Review

• The: general verification procedure.for plant-specific-review is described below.

Figure 1 outlines this procedure.

It should.be noted that this figure depicts the implementation steps for generic resolution (see paragraph 5 below); however, the logic represented by Figure 1. generally applies to utilities that are not participating members of. the generic group.

The implementation must include:

development of an equipment list; comparison of site spectra with appropriate

-bounding spectra; a walk-through inspection including review of anchorages, re-view of equipment functional capability, review of equipment unique to nuclear

. plants,.and a seismic adequacy review of replacement parts.

A.

Development of Equipment List I

Each licensee will be required to develop an equipment list that includes all equipment iters identified as necessary to perform functions related to plant hot shutdown (see Paragraph.2, Scope of Seismic Adequacy Review,.and Paragraph 3,-

Requirements for Plant Shutdown, above).

l A list of typical equipment required for plant hot shutdown is shown in Table 1.

)

(.

The list is based on polls of SQUG member utilities and is believed to include

(

all of the types of safe shutdown equipment in nuclear power plants; plant-J l

specific lists to be generated as part of the implementation of A-46 are expected

{

j to be shorter.

B.

Comparison of Site Spectra With Appropriate Bounding Spectra The licensee will be required to verify that the appropriate data base _ bounding spectra envelope the site free field spectra at the ground surface defined for the plant. There are a number of nuclear plants whose free field SSE spectra are defined at the foundation level.

For these plants, an estimate of the free field spectra at the ground surface must be made for comparison with the data base bounding spectra.

The licensee must identify all equipment on the equip-ment list which is located at an elevation higher than 40 feet above grade level.** For equipment above about 40 feet, one and one-half times the appro-3 priate data base bounding spectra must envelope the floor response spectra for the equipment. For those cases where floor response spectra are needed,

• Refer to the SSRAP report, "Use of Past Earthquake Experience Data To Show Seismic Ruggedness of Certain Classes of Equipment in Nuclear Power Plants,"

-January 1985.

The SQUG is in the process of expanding the data base to in-clude more recent earthquake experience and 20 classes of equipment which cover all the equipment needed for plant hot shutdown.

The SSRAP report also l

is being revised accordingly.

The final guidance in the SSRAP report may l

differ from that mentioned here. The revised SSRAP report should be followed for implementation guidance.

• • " Grade level" is the top of the ground surrounding the building.

NUREG-1211 6

m

2C

U FT1 C) e Hb)

HH e

Evaluate

• G.G.

G.G.

trial conduct develop Trial

,,ik.thru workshop detailed 4

walk-thru +

h

+

for

+

pr cedure walk-thru inspection fine-tune participating procedure procedure utilities Guidelines Experience for outlier Guidelines Test data Anchorage data base review for for electrical (GERS) review exclusions &

equip. not in relay developed guidelines caveats 8 types but in review by EPRl/SQUG guir elines data base p! ants f

I

[

Pre-Implementation am aus an m-implementation I

N Walk thru 1

F inspection:

(1) review of all anchorages (2) Seismic syst, ems 4

IN interaction review Develop equip.

(3) Identification &

list review of deficiencie' and outliers Modifications Audit by Utilities Staff sg and replacement consultants issue issue Y

7 of equip. and

+

to G.G. and

+

inspection 4

plant-specific anchorages if by staff reports SERs Compare site Review of needed spectra with appropriate electrical relays 4

A m.

h bounding capability g

for functional spectra g

l Leger.d:

y,,g seismic SSRAP G rg,. Generic Group adsquacy limited of equip.

audit equip. - equipment Review of to be modified

+

equip. unique 4

or replaced

• - May be completed after issuance of l

to nuclear formal requirement olants Figure 1 USI A-46 Implementation

Table l' Typical equipment list for USI A-46 1.

Mechanical Equipment 1.

Vertical pumps and motors *

.2.

Horizontal pumps and motors

• 3.

Motor-operated valves

• 4.

Air-operated valves * (including solenoid valves) 5.

Heating,-ventilation and air conditioning items such as fans, blowers, chillers,'and filters 6.

Pumps-(turbine driven, diesel driven, and reciprocating positive displacement types) 7.

Main steam isolation valves 8.

Pilot-operated safety / relief valves.

9.

Spring-operated safety / relief valves 1i 10.

Nuclear steam supply system mechanical equipment (control' rod drive mechanisms) 11.

Power-operated relief valves

12. Air compressors and air accumulators'.

13. Heat exchangers, tanks (anchorage-review only)

14. Atmospheric steam dump valves 2.

Electrical Equipment 1.

Low voltage switchgear*

2.

Metal clad switchgear*

3.

Motor' control centers

• 4.

Transformers * (unit substation type) 5.

Motor generator sets 6.

Distribution panels - ac and dc 7.

Batteries and battery racks 8.

Battery chargers 9.

Inverters 10.

Diesel generators and associated equipment

11. Transformers (other than unit substations) 12.

Automatic transfer switches 13.

Remoteishutdown panels 14.

Cable trays and conduits 3.

Instrumentation 1.

Transmitters ~(pressure, temperature, level, flow) 2.

Switches (pressure, temperature, level, flow) 3.

Resistance temperature detectors and thermcouples 4.

Relays 5.

Control panels and associated components 6.

-Instrument racks and associated components 7.

Instrument readouts (displays, indicators such as meters and recorders)

• The eight. equipment types already included in the SQUG pilot program, (Seismic Experience Data Base).

NUREG-1211 8

--n n

NUREG/CR-3266 entitled " Seismic and Dynamic Qualification of Safety-Related Equipment in Operating Nuclear Power Plants - Development of a Method To Generate Generic Floor Response Spectra" may be used as one alternative to develop the necessary floor response spectra on a case-specific basis. The appropriate bounding spectra for equipment belonging to the original eight types in the data base are defined in the SSRAP report (SSRAP, January 1985), and are also shown as Figure A.1 in Appendix A of the enclosure to the proposed generic letter (see Appendix A).

Fer equipment not included in the eight types in the data base but which exists in the data base plants, and for equipment unique to nuclear plants, appropriate generic bounding spectra are under development.

These generic bounding spectra will not exceed the type A bounding spectrum defined for the data base plants unless other seismic experience or test data collected by EPRI and the NRC Office Of Nuclear Regulatory Research (RES) verify significantly higher seismic capacity.

For equipment outside the data base for which the type A bounding spectrum is used, any caveats or exclusions developed before the implementation of USI A-46 by the SQUG/SSRAP and the NRC must be taken~into account.

The bounding spectra must be developed by SQUG, endorsed by SSRAP, and reviewed and approved by the staff before implementation.

C.

Walk-Through Inspection Each licensee will be required to conduct a plant walk-through and visual in-spection of equipment items in the equipment scope.

The inspection team must consist, as a minimum, of (1) an experienced structural engineer familiar with seismic anchorage requirements (2) an experienced mechanical engineer familiar with plant mechanical equipment (3) an experienced electrical engineer familk sith plant electrical equipment Furthermore, an operations supervisor or a licensed Senior Reactor Operator must be available for consultation before and during the walk-through process.

Not all' members of the inspection team are required to participate in all parts of the walk-through; however, appropriate technical expertise must l'e included for each review area, and a person with the proper structural background must always be present to inspect the anchorage for all equipment.

As an alternative, licensees may use consultants instead of their staff for team members identified in items (1), (2), and (3) above.

The walk-through inspection must cover anchorage review, seismic systems interaction review, and identification of potential deficiencies and outliers, as follows:

(1) Anchorage Review For equipment within scope, verify equipment anchorage (including cable trays and conduits and required tanks and heat exchangers) using guidance provided in Paragraph 5 (Generic Resolution) below and the anchorage guidelines developed by SQUG/EPRI and approved by SSRAP and NRC.

NUREG-1211 9

o 4

[(2)4 iSeismicISystems Interaction Review

~

lDUring the'wal b through inspection, potential seismic systems interactions must beLreviewed to ensurezthat all. equipment within the scope of review-1 will'not be affected by the failure or displacement of adjacent structures,;

l piping..or' equipment due:to physical contact. This review must include-

~

iprotecting cable. trays. and conduitsand ' essential. tanks and heat exchangers,.

_as well-as mechanical-and electrical equ.ipment items.

Particular attention ishould'be given;tofnon-safety-related structures _and equipment.

However,-

the~ potential for physical -impact of ' safety-relatedl items with other: safety-Lrelated items because of.their_close proximity also should be' considered.

~

The review does~notfinclude systemsLinteractions other than direct physical

. impact.

Flooding due to_ failure of' tanks'or_ piping is'not included.

(3); Identification-and Review of " Deficiencies" and Ou'tliers"

" Deficiency" in this context means equipment, components, and their-anchorages / supports,' identified to be obviously' inadequate.by,the A-46;

~

-criteria during plant-specific walk-through revie'ws, and are confirmed to be-inadequate by further engineering studies.

" Outlier" in this' context means equipment items that are subject to the caveats and~ exclusions defined in' the proposed generic letter, or are otherwise not covered by the experi-

~

ence. data.. :Any potential deficiencies must be identified-for all equip-ment'within-the scope.- However, it is'not' expected that many deficiencies-will be found.in any typical nuclear plant.

The treatment of deficiencies is further described in Paragraph 6 (Provisions for. Resolution for Individual' Utilities) below.

For equipment belonging-to the original eight types in the seismic exper-ience data base, identify data base exclusions and caveats from the guidance provided in Paragraph 6 (Guidance on.Use of Seismic Experience Data) of the enclosure to the proposed generic letter.

For equipment which exists in data. base plants but does not belong to the original eight types, collection of additional seismic experience data is not required;'however, the basis for seismic adequacy must be documented for each equipment type.

Guidelines provided in Paragraph 7 (Guidance'on Review of Equipment) of the enclosure to the proposed generic letter and detailed walk-through review guidelines being developed by SQUG and to be-approved by SSRAP and NRC prior to implementation of A-46 should be used for identification and review of " outliers" and " caveats" during the walk-through inspection of this equipment.

As a general guideline, all identified outliers can be deferred for suple-mentati'on for a time period to be negotiated with the NRC Staff.

The proper integration of the proposed work scope into each plant's living schedule for plant modifications will'be taken into consideration.

In 1984, EPRI, with a contractor, began an effort to collect and evaluate existing seismic test data on nuclear plant equipment for A-46. The main objectives of the EPRI program are:

(1) to establish generic equipment ruggedness spectra (GERS) which can be used to demonstrate seismic adequacy of equipment, and (2) to demonstrate functional capability of equipment or components (e.g., relays) which are required to function during the strong NUREG-1211 10

u,

+

a y L

~

notion'part of?an SSE. :All GERS from-the;EPRI program.are sched'uled to be

~

q available for review by'SQUG/SSRAP and the NRC staff by December.1986~.

A=separateLeffort was? initiated by RESiin'1985'to collect and compile,

existing' seismic < fragility;dataLon' nuclear' plant' equipment. 5These-test-data:will/be:used to: improve the' currently available'information on~ equip--

l ment! fragility and seismic margins'.

~'

Even though.the objectives of:the:RES-and EPRI! programs 'areidifferent,. the :

data collection' process and the source; organization e r?thc;dataiare mostly

'the same. cThe cooperative -effort calls' for exchanging collected data andL coordinating collection activ.ities;by.both organizations in ordereto mini-Jaizeicost, preventEduplication,- and maximize the:use~of;available' data

sources. :It has been. agreed that EPRI will.primarily collect dat'a from.

. utilities-and West-Coast testing laboratories,,and'RES will.primarily collect data'.from vendors ~and East-Coast testingLlaboratories.-

"D.

Review of Equipment Functional Capability

-There=may!be' equipment or. components'which are required to function'during th'e strong ground motion.part of'an SSE. 'In these cases,. functional capability of the equipment or components _must be established.

E0n the basis of the seismic experience data gathered to date', the only concern that remains on' equipment' functional-capability is chatter of electrical relays.

Contact' ors and switches are considered as relays in this context.

In' addition,

-mercury switches are.known to malfunction during testing and-should be replaced-

.by other types'of' qualified switches.

Unless the test data being collected by

.EPRI/RES reveal 'otherwise, certain types of relays are the only equipment'whose functional capability must be verified.

It is required.that the review of electrical relays follow these guidelines:

(1) Essential' plant functions that are needed to achieve and maintain hot shut-down conditions during and after an SSE must be identified.

(2) IAssociated systems and electrical circuits needed to provide these functions must be identified.

(3) The circuits that are essential must be evaluated, the relays that are essential must be identified.

Essential relays are relays that must remain functional without chatter during an SSE.

(4) The essential relays must be qualified by test in a manner consistent with current' licensing requirements (SRP Section 3.10, and Regulatory Guide 1.100, IEEE Standard 344-1975), verified by comparison with the test

~

data base being developed by EPRI/RES, or replaced by relays that are qual.ified to current licensing requirements or that can be verified by the

-4:

test data base. As an alternative, the redesign of circuitry, strengthening of relay supports / cabinets to reduce seismic demand, or relocation of relays to reduce demand can be used.

(5) All relays that could potentially change state during an SSE due to contact chatter and preclude the use of equirment needed for shutdown after the SSE NUREG-1211 11

='

W y

y -

m g

~ ~

g y,

7 s-

~ r " '.j 3 s l -Y .p e i (, s ,.i M :' ' f mu'st-be11dentified. : Consideration mu'st be given'to all potential opera ' ' itionalistates_of the relay (i.e.,Jenergized or.deenergized,' tripped orj -.non-tripped);with respect.to the. availability,of.the' equipment (theyicontrol u - 'following1an~ earthquake. LThesei. relays:must also"be; qualified by test or by-icomparisonlwithtestdata:(comparison ~ofLspectralatmountinglocation),or L

replaced by relays: qualified.toLcurrent licensing _ requirements. The 4

q 3 redesign oficircuitry kstrengtheninglof relay! supports / cabinets to reduce fr ^ cdemand, orirelocation_of relays-toLreduce demand can also be.used. : As'any ' alternative',tthef1icenseeimay show'that.chatteringfor_ change of_statetof 1 "the:r_elays;does)not" preclude subsequent l equipment or.-system functions. In. ~ -addition,'creditLcan be_taken for..timelyLoperator action.toires'etsthe relays' ?

in case change ofistatefoccursEduringfan'SSE;'prov_ided theffollowing

-i frequirements :are met: y .L(apiDetailedl relay.resettingproceduresare' developed. ~ ^ - _ b)- There. i~s sufficient time for resetting -the relays. ( L(6);.As;a. general guideline, seismic verification of relays outside 'of the test data. base may be deferred ~until additional! test data is developed. cActual-

schedule dates;will be' based on negotiations with each utility'(whetherla..

4 member ofithe1 generic group or not). ' The proper integration:of. the proposed work scope intoteach plant's'living. schedule for plant modificationsLwill be i -.takeniinto' consideration. 'E; Review of Equipment-Unique to Nuclear Plants For equipment, unique to nuclear plants--such as control rod drive mechanisms, 1 -some power-operated relief-valves, and main steam isolation valves--the test ~ experience data base being developed by EPRI/RES or qualification records for ~

similar; items may be used to verify seismic adequacy.

F. Replacement Parts " Component" in this context means equipment and assemblies-(including anchorages. and supports)'(such as pumps and motor control centers) and subassemblies and. devices (such as motors and' relays) that are part of assemblies. -If components ~ are modified or replaced by the utility as a result of the A-46 review or for .any other reasons,.the' licensee must' verify the seismic adequacy for each modi-fication'or replacement (of an assembly, subassembly, or device) either by using A-46 criteria and methods or as an option, by qualification by current licensing criteria. This provision also applies to future modification or replacements. G.' Verification of Anchorage To verify acceptable seismic performance, the licensee must provide adequate engineered anchorage. There are numerous examples of equipment sliding or over-turning under seismic loading as a result of a lack of anchorage or inadequate -anchorage. Inadequate anchorage can include short, loose, weak, or poorly installed bolts or expansion anchors; inadequate torque on bolts; and improper welding or bending of sheet metal frames at anchors. Torque on bolts can normally be ensured by a' preventive maintenance and inspection program. ~ NUREG-1211 N -5

-In general,.to check equipment anchorages one must estimate the equipment weight .and its approximate center of gravity. Also, one must either estimate.the -equipment's fundamental frequency to obtain the spectral acceleration at this frequency or else use the highest spectral acceleration for all frequencies. 1 When horizontal floor spectra exist, they may be used to obtain the equipment spectral acceleration. 1 Alternatively, for equipment mounted less than about 40 feet above grade,1.5 times the. free-field horizontal design ground spectrum may be used to conserva-tively estimate the equipment spectral acceleration. For equipment mounted more p than about 40 feet above grade, floor spectra must be.used. Equipment anchorage must be not only strong enough to resist the anticipated forces but also be stiff enough to prevent excessive movement of the equipment and the potential resonant response with the supporting structure. The review of anchorages should include consideration of both the strength and stiffness of the anchorage and its component parts. Additional discussion's on seismic motion bounds and equipment' supports and the anchorage for each of the original eight classes of equipment in the experience data base is included in paragraph-6 of the enclosure to the proposed. generic-letter. During the walk-through inspection, anchors and supports of equipment within the scope of review will be carefully inspected using the detailed guidance provided.* If the adequacy of supports and anchors cannot be determined by inspection, an engineering review of the anchorage or support will be required. This engineering review will include a review of design calculations or the performance of new calculations and/or verification of fundamental frequency of equipment to ensure adequate restraint and stiffness. Physical modifications ~ may be necessary if engineering review determines the anchorage or support to be inadequate. 5. Generic Resolution The NRC will endorse and encourage a generic implementation of USI A-46 if the following guidelines are followed: (1) All member utilities of the SQUG would be eligible to participate. (2) The generic group must be responsible for (a) developing procedures to identify relays to be evaluated, (b) defining the functionality require-ments, and (c) develop evaluation procedures for relays. These procedures must be reviewed and endorsed by SSRAP and the NRC staff. (3) The generic group must develop an implementation procedure for systematic and consistent plant-specific reviews. Discussions between the staff, the SQUG, and SSRAP have resulted in a tentative procedure for generic implementation. The procedure is summarized as follows: j

• The detuiled guidance has been developed jointly by SQUG and EPRI.

It was approved by SSRAP and is being reviewed by the NRC staff. It will be approved by the NRC staff before implementation. NUREG-1211 13

.c ..g.; 1 T s q <A s 'r-

(a)-!The generic group must. submit to th'e NRC a.genericischedule for the

~ Ldevelopment of an implementation' procedure and for workshops / training : . g

seminars for participating, utilities.

~ n (b). Each individual. utility must! submit'an implementation ~ schedule to: ~

the-NRC within'60 days of receipt'of1the A-46 generic letterJ 1

(c) L The generic group'aust develop a detailed; walk-through procedure ! based lonthe-implementationrequirementdefinedinthe. generic: letter.

(d)EA trial walk-through'insp'ection must be performed by generic group

. consultants with NRC; participation. ~~(e)f An evaluation'of the' trial walk-through mustLbe made by the staff, 1the SQUG, and SSRAPLand the procedure-fine' tuned. .(f) The generic group must.then conduct workshops / training seminars.for participating utilities. -(g) ; Individual? utilities must perform the plant-specific implementation ~ ' review. This. review must generally follow the guidance given in _ . paragraph 4, General Verification Procedure for Plant-Specific Review,- 'above, u ~ (h),.Each. utility'must-submit to the NRC an inspection report that must include: certification of completion of review, identification.of . deficiencies' and outliers, justification for. continued operation (JCO). for identified ~ deficiencies if these deficiencies are not corrected within 30 days, modifications and replacements of equipment / anchorages

(and supports) made.as a result of the reviews, and the proposed.

schedule for required modifications and replacements not completed at the time of the report submittal. ~ The objective of the requirement to submit a JC0 is to provide-assur-ance that the plant can continue to De operated without endangering the health and safety of the public during the period required to correct the identified deficiency. The JC0 may consider arguments such as imposition of administrative controls or limiting conditions for operation (LCO) or consideration of the importance of the safety function involved and/or identification of alternate means to perform that function. (i) Consultants to the generic group must perform audits of plant-specific reviews. All plants must be audited. The NRC staff will participate in plant audits on a selective basis. The generic group must submit a report of audits performed and results of these audits to the NRC. This report must cover all participating utilities, and must also include results of any reviews and/or audits performed by the SSRAP. (j) The SSRAP and the NRC staff must perform a review of the generic group audit process to evaluate effectiveness. NUREG-1211 14

(k) Final approval of the implementation will be made by the NRC in the form of plant-specific SERs following receipt of a final report from individual utilities certifying completion of implementation reviews and equipment / anchorage modifications and replacements. (4) The generic group must provide for the continuation of the SSRAP as an independent review body. The SSRAP must be consulted during development of the generic program and walk-through procedure, and during implementation audit. (5) NRC staff members must be invited to participate in all meetings'between the generic group, its consultants, and the SSRAP. 6. Provisions for Resolution for Individual Utilities The generic resolution described in Paragraph 5, Generic Resolution, above is the method preferred by the NRC for the implementation of A-46. This paragraph offers provisions for resolution of A-46 for individual utilities not partici-pating in the generic group. Each utility is required to develop a detailed review procedure that must be submitted to the NRC staff for review. This procedure must reflect the guid-ance given in Paragraph 4, General Verification Procedure for Plant-Specific Review, above. Data and procedures developed by the SQUG will not, in general, be available to nonparticipating utilities. All information that has been made publicly available by SQUG or the staff can be used. Each utility is required to perform plant-specific verification reviews according to the guidance in Paragraph 4. It is also required to maintain an auditable record of implementation of USI A-46. Within 60 days of receipt of the A-46 generic letter, the utility must submit to the NRC a schedule for implementation of the A-46 requirements. Utilities who may not have access to SQUG implementation procedures or data base may have difficulty in establishing implementation schedules within 60 days. For these utilities the NRC vi'.1 negotiate time extensions on a case by case basis. An inspection report must be submitted by the utility to the NRC following the plant-specific walk-through inspection. It sheuld consist of the following: (1) Certification of completion of the walk-through inspection and a description of the procedures used. (2) List of equipment included in the review scope. Equipment required to function during the strong shaking period should be identified. (3) Identified deficiencies. (4) Identified outliers. (5) Modifications and replacements of equipment / anchorages (and supports) made as a result of the inspection. (6) Proposed schedule for future modifications and replacements. NUREG-1211 15 i

(7) A justification for cc,atinued operation (JCO) for identified deficiencies if these deficiencies are not corrected within 30 days. ~ When all necessary modifications and replacements of equipment / anchorages are complete, the utility must submit a final report to the NRC. A description of i. the procedures used for the implementation reviews, as well as of modifications and replacements, should be fscluded. The NRC will review the inspection procedure, inspection report, and the final -report, and will audit all plant-specific reviews before final NRC approval. [ Final NRC approsal will be made in the form of plant-specific SERs. 1 V. VALUE-IMPACT ANALYSIS The staff has determined that the requirement to review existing plants for the seismic adequacy of mechanical and electrical equipment is a backfit as defined by 10 CFR 50.-109. The review of existing plants as well as any modifications resulting from the review is subject to the provisions of 10 CFR 50.109. The key factor in deciding if the implementation of A-46 is a backfit (as definec: in i 10 CFR 50.109) is the consideration that the review "model" has changed. That is, all operating plaats are being required to review equipment seismic capabil-ity against a "new model" (i.e., using comparison to seismic experiences or test data as compared to the basis used for their original licensing review). i The backfit. analysis which is presented in the following paragraphs presents justification for performing the seismic adequacy review. Backfit analysis for correction of any deficiency will be performed on a case by case basis if re-quired following completion of the review. Although it is impossible to consi-der the cost benefit of all conceivable deficiencies, several potential defi-ciencies are discussed to establish the possible value impact resulting from i seismic deficiencies. In addition, qualitative safety benefit arguments are i presented to support the proposed requirement. 1 Value-impact analyses involve determination of the net safety bensfit achieved I from implementing a proposeo.'esolution, which is usually a physical change to the plant or a procedural change. The cost of implementing the proposed resolu-tion is then estimated a'J the recommended implementation plan is based on the cost effectiveness, considering how much risk reduction is achieved for the money spent. 1. Qualitative Assessment of Safety Benefit Three factors influenced the staff judgment on safety benefit. l First, subject to certain exceptions and caveats, the staff has concluded that equipment installed in nuclear power plants is inherently rugged and not suscep-tible to seismic damage. In the SQUG pilot program, the eight types of equipment (for which seismic ex-perience data were collected to form the experience data base) are representa-tive of mechanical and electrical equipment in both nuclear and non-nuclear plants. These eight types of equipment generally constitute, in a numerical sense, a large percentage of all safety-related equipment in a nuclear power plant. While conducting the pilot program study, the SQUG looked for damage to all types of equipment resulting from seismic loading. This search was not NUREG-1211 16 3

restricted to the eight types considered for the data base. Of the approxi-mately 3000 pieces of mechanical and electrical equipment surveyed in the data base plants, only one equipment item (an air-operated valve) was damaged because of impact against a nearby structural girder during the earthquake. Although instances of overturned cabinets (such as switchgears and motor control centers) were found, they could all be attributed to inadequate anchorage and restraint. In most instances equipment functioned after the cabinets were made upright. SQUG therefore concluded that, if anchorage and support are adequate, equipment in the data base plants is inherently rugged and not susceptible to damage at the seismic levels experienced. Because of the similarity to equipment in-stalled in nuclear plants, this conclusion was extended to nuclear power plants. The review of seismic experience data (of eight classes of equipment) by a panel of seismic experts (SSRAP) also resulted in similar conclusions. These are as follows: (1) Equipment installed in nuclear power plants is generally similar to, and at least as rugged as, that installed in conventional power plants. (2) With some reservations, when this equipment is properly anchored, it has an inherent seismic ruggedness and a demonstrated capability to withstand significant seismic motion without structural damage. (3) For this equipment, functionality after the strong shaking has ended has also been demonstrated, but the absence of relay chatter during strong shaking has not been demonstrated. The NRC staff has closely followed the SSRAP work and is in broad agreement with its conclusions. The staff has concluded that if the SSRAP spectral conditions are met, it is generally unnecessary to perform explicit seisniic qualification on the eight classes of equipment studied. On the basis of the equipment damage survey conducted in the data base plants and a broad damage survey of strong motion earthquakes around the world, the staff has further concluded that there is no need to collect additional seismic experience data on the remaining type of equipment, provided: (1) anchorage and support adequacy of equipment is ensured, (2) certain caveats or exclusions for this equipment (derived from licensee, SEP, and SQRT review experience) as outlined in paragraph 6 of the enclosure to the proposed generic letter are addressed, and (3) the SQUG docu-ments the basis for seismic capability of each equipment type not included in the original eight types for which detailed data were collected. The second factor influencing the staff judgment on safety benefit is that although equipment is inherently rugged and not susceptible to seismic damage, failures resulting from seismic loads are likely to occur if equipment is not adequately supported or anchored. The need to review anchorage and supports was identified during the Systematic Evaluation Program (SEP) review. Structural adequacy of equipment, including supports and anchorage, was reviewed at each of the SEP plants. These reviews included a plant walk-through by a team of seismic experts. The results of these walk-through inspections included: identification of potential anchorage and support deficiencies such as lack of longitudinal restraints on battery racks at Millstone 1; anchor bolts over-stressed on the containment spray heat exchanger and isolation condenser, a need for positive anchors on switchgear panels, and a need for evaluation of diesel generator anchorage at Oyster Creek; and strengthening of anchors on NUREG-1211 17

battery racks and a need for a general engineering review of anchors at Dresden 2. As a result of the SEP experience, IE Information Notice 80-21 was issued to inform all licensees of the potential problem with anchorages. However, a recent survey of seven operating plants for the purpose of developing plant-specific equipment fragilities indicated that' anchorage deficiencies still exist in operating reactors. The proposed requirement is based on the need to ensure that equipment is ade-quately anchored and supported, that certain equipment (primarily electrical relays) functions as required during the shaking motion, and that other identi-fied exceptions and caveats detailed in the SSRAP report (SSRAP, 1984) and developed by the staff and SQUG are addressed. The safety benefit of verifying the seismic adequacy of equipment by performing the proposed anchorage inspection procedure is principally to ensure that equip-ment needed to safely shut down the plant does not fail because of failure of the anchorage or support or because of mounting configurations or geometry which makes it susceptible to seismic damage. Unanchored equipment or improperly anchored equipment may overturn or move during seismic shaking. Numerous in-stances of overturned and displaced equipment and tanks because of improper or nonexistent anchorage were found during the review of the experience data base plants. This was particularly evident in the review of the Coalinga earthquake data (EQE, 1984). Although equipment anchorages have previously been identified as a problem area, there is evidence that anchorage deficiencies still exist in operating reactors. An inspection program to verify anchorage and supports of safety equipment would ensure that equipment failures due to seismic motion would be highly unlikely. Third, although functional capability after the strong shaking has ended has been demonstrated by the seismic experience data, functional capability (such as absence of relay chatter) during the strong shaking motion (first 30 seconds of an earthquake) cannot be demonstrated by seismic experience data. There is also some equipment that is unique to nuclear plants, for which the seismic experience data base does not apply. Therefore, functional capability of all required equipment and the seismic adequacy of equipment which is unique to nuclear plants can be verified by test experience data. EPRI and RES are currently conducting a program for the collection of test experience data. This program is designed to specifically support A-46 implementation. 2. Quantitative Examples of Potential Value Impact It is very difficult to develop a generic estimate of the safety benefit of performing plant specific seismic verification reviews. However, to demonstrate the potential safety benefit and value impact, several examples are presented. The first example considers the seismic failure of an electrical cabinet anchorage. This is followed by a summary of the risk analyses performed by Sandia Corporation as part of the assessment of decay heat removal systems under USI A-45. NUREG-1211 18

A. Seismic Failure of an Electrical Cabinet Anchorage (1) Frequency Estimates The 'nitiating event ( P(i) ) is assumed to be a SSE with a return frequency of 2.5E-4/RY. Failure of at least one equipment anchorage is assigned a probability of 1.0 ( P(AF) = 1.0). The probability that the equipment with inadequate anchorage is an electrical cabinet ( P(C/AF) ) is assumed to be 0.5. Since only equipment affecting hot shut-down are included in the A-46 program, it is as likely as not, that the cabinet that fails ( P(HF/C) = 0.5) is critical to hot shutdown. Given the above, it is estimated that there is a 20% chance that above events are likely to lead to a core melt (P(CM/HF) = 0.2). Because the equipment is likely to include containment heat removal systems, or affect contain-ment isolation capabilities, the probability of containment failure ( P(CR/CM) ) due to overpressurization, or bypass, is estimated at 0.5. Structural failure of the containment due to the seismic event is not considered. (2) Consequences (Risks) The public risk (R) from the initiating SSE and subsequent events is calcu-lated by the following equation: R = P(i)*P(AF)*P(C/AF)*P(HF/C)*P(CM/HF)*P(CR/CM)*R(D/CR)*L The term "R(D/CR)" = (SE+6) is the conditional public dose, in man-rem, given a containment release. This value is based on a WASH-1400, Cate-gory 2 type release, the fission product inventory of a 1120 MWe PWR, meteorology typical of the Byron (mid-west) site, a surrounding uniform population density of 340 persons per square mile over a 50 mile radius from the plant site, with an exclusion radius of cne-half-mile from the plant. The "L" term is the estimated remaining reactor life (30 reactor-years) of the typical plant. Substituting the estimated values given into the above equation yields: R = 940 man-rem / Reactor as the estimated seismic risk potential per reactor (plant). The safety benefit from the A-46 verification program is assumed to reduce the seismic risk potential by more than an order-of-magnitude. (3) Value/ Impact Assessment The risk reduction potential of 940 man rem / reactor is predicated on assur-ing adequate seismic anchorage and supports for equipments. Examination of the cost estimates shown in Table 3, show that approximately one-half of the A-46 costs are associated with repairs to the anchorages and supports. Because the A-46 verification program includes only equipment that could affect hot shutdown capabilities, the precision of the risk estimates cannot differentiate between the risk reductions that may result from repairs to 5, or 10, anchorages and supports. However, the cost estimates that are l NUREG-1211 19 \\ L__

_,m ~ 4 "[ dominated by anchorage and support repair cost, are approximately propor- -tional to the Aa46. costs.- 1 Based'on.the=above' assumptions, the estimated.value/ impact ratio (V/I)'is- ~the ratio of the estimated potential risi reduction:(940 man-rem / reactor) and the Lestimated average costs shown-laur in Table 3 ($600,000). This ' yields an estimated V/I ratio of

V/I = 1.5

-man-res/ reactor $1000. costs Based'on' the stated assumptions and a rule-of-thumb of One man-rem /$1000 as

an acceptable measure of cost effectiveness, the A-46 program would be cost'

. effective. Itis.unlikelythat~failureestimateswouldbeashighasassumedidthis analysis,.but-the estimated V/I does demonstrate that there.is a potential for significant safety benefit. B. , Examples from USI A-45 Analyses As part of USI A-45,: seismic PRA analyses were conducted to determine the risk . significance of decay heat removal systems. Probabilistic risk assessments were performed for'a number of example plants. A summary of the results of two of these studies is presented here to illustrate the' potential for safety benefit due to conducting the A-46 seismic verification reviews and in correcting' potential deficiencies. The' reports from which the following summaries are taken are in draft form and are still.under review by the NRC. The risk contributions cited, therefore, are. subject to change but should serve to illustrate the potential for safety improvement. The seismic fragilities used were generally taken from a generic data base of fragility functions developed in the seismic safety margins research prog' ram (SSMRP). Since this fragility data base was assembled in the 1980-1982 time frame, a review of plant specific fragilities was performed. This involves plant visits to adjust the generic fragility functions for plant specific application. l (1) Example 1, Combustion Engineerina PWR With SSE of 0.10 A plant visit and walkdown was made in early 1986. It was determined that l the generic fragility functions were adequate for all components. There were three components that could potentially be significant risk contribu-tors, the service water pumps, the refueling water storage tank (RWST) and the condensate storage tank (CST). The service water pumps, however, did not play an important role in the analysis results. The RWST and CST median fragilities were as follows: NUREG-1211 20 L ..I

N r Median' Base-p: Acceleration' Random Component Failure-Mode at Failure Uncertainty RWST Buckling with 0.24g 0.34 Anchor Bolt-Yielding CST Buck 1)ingwith 0.26g 0.34 Anchor Bolt Yielding It should be pointed out that RWST and'the CST have median failure acceler-ation levels above the' SSE. However, they did have less margin of safety above the SSE_than the other components examined during the plant visit, and hence were anticipated to be significant contributors to the accident sequence probabilities. The base case core melt probability due to seismic events was calculated to be 1.3E-05. The calculated base-case seismic risk was: Total man-rem /yr = 1.83 If the two tanks were seismically hardened by improving the anchorage and ~ providing stiffeners at the' tank base, the core melt probability due to seismic failure reduces to 9.6E-07 and the calculated risk is then: Total man-rem /yr = 0.13 If the assumption is made that the remaining plant life is 30 years, then the potential safety benefit achieved by repairing the tank anchorages is 30 (1.83-0.13) or 51 man-rem. With a guideline of $1,000 per man-rem, approximately $50,000 could be justified to make improvements. (2) Example 2, Westinghouse 2 Loop PWR With SSE of 0.12g During a plant visit, the following components were identified as requiring plant-specific fragility derivations: 1) RWST and CST Tanks 2) 4160V Safeguard Busses 3) SIS Pump Busses 4) 480V Safeguard Busses 5) DC Busses 6) DC Battery Chargers 7) Battery Racks The RWST was identified because of its unusually large height to diameter ratio, and the CSTs because of the minimal number of hold-down bolts about their perimeters. The safeguard busses and SIS pump buses were identified because their anchorages consisted of minimal plug welds holding the cabinets to I-beams embedded in the floor. The DC battery NUREG-1211 21 o

chargers'are anchored by 5/16" bolts which appeared minimal.

Finally, the battery racks had minimal wall anchorage, and also utilized wooden battens to restrain the batteries.

Based on these observations, site-specific fragilities were developed for all.of these components. All.the components above were found to have median failure acceleration levels above the SSE. However, they did have-less margin of safety above a the SSE than the other components examined during the plant visit, and l hence were anticipated to be significant contributors to the accident sequence probabilities. Site-specific fragilities were estimated for'the seven items above cs follows: Median Base Random Component Failure Mode ~ Acceleration at Failure Uncertainty RWST Buckling w/ Anchor 0.20g 0.33' Bolt Yielding CST Buckling w/ Anchor 0.96g 0.33 Bolt Yielding 4160V Switchgear Sliding Weld Failure 3.0g 0.25 Overturning Weld Failure 1.5g 0.25 SIS Pump Busses Sliding Weld Failure 3.0g 0.25 Overturning Weld Failure 1.5g 0.25 480V Switchgear Sliding Weld Failure 6.6g 0.25 Overturning Weld Failure 2.0g 0.25 DC Battery Chargers Overturning Bolt Failure 1.65g 0.17 DC Busses Overturning Weld Failure 2.0g 0.25 Battery Racks Anchor Bolt Shear 0.86g 0.20 Based on the plant walk-through and the base case evaluation of the risk of seismically-induced core melt, a number of potential modifications were identified, and the potential reduction in risk resulting from these modifications was evaluated. The refueling water storage tank is located in the containment facade. It is not anchored adequately to survive an earthquake and is estimated to fail at the level of the safe shutdown earthquake. Since refueling water is needed for both high and low pressure injection systems, the result is a significant contribution to core melt frequency. The proposed modification provides a Seismic Category I backup water source that would survive an earthquake 4 times the SSE and allow the operator to rapidly valve water to the HPI or LPI systems. The new source would be the Spent Fuel Pool which, when providing the necessary 140,000 gallons, will have its level lowered approximately 11 feet. Use of the Spent Fuel NVREG-1211 22

y Vb- ' Pool'wculd_ require' pumps;and piping tbEdeliver'the borated water to the ~ [;' suction:of-the HPI and LPI. systems. The total seismically.. induced core melt _ probability with this. modification would be approximately-i '3.1E-5. per year. p 'A'second mod'ification would provide ~ additional anchorage for the 4160V-Lsafeguards buses, 480V transformers and transformer _ buses, safety injection system pump. buses, instrumentation power supply inverters, and battery chargers. Currently,~an earthquake might_ fail the anchor plug . elds 'and then either tip the buses over and shear the wiring, or walk w the cabinets.off'of the pedestals.- The 480V transformers and the instrument inverters could similarly' fail and seriously jeopardize the-AC-power supply. The battery chargers are anchored.by 5/16 inch bolts which can break and allow these chargers to tip over. Loss of battery chargers; will eventually (in 4 hours). lead to a-loss of DC power and core melt. In addition' the battery racks utilize wooden battens and have inadequate wall anchors which should be replaced. An earthquake (causing LOSP) could' knock these batteries over and cause loss of DC power to all safety systems, ~Therefore, this second modification would include improving both the battery racks and wall anchors. The combination of all of these upgrades-would provide sufficient support and anchorage to the electrical components so that their. support systems will withstand loads greater .than.four. times SSE seismic loads. The final-proposed modification to reduce the seismic contribution to core-melt probability is to provide a dedicated safety grade instrument air system to operate the pressurizer power operated relief valves (PORVs). Comparison of the base case with the case in which all modifications are made shows that the annual probability of core damage drops from 6.0E-5 to 5.0E-6, and that the estimated populations dose drops from 34.5 man-REM / year to 2.4 man-REM / year. Thus, implementing the modifica-tions would result in an order of magnitude decrease in risk of core damage and population dose. From a risk reduction standpoint, two items stand out, namely, the RWST and the station battery racks. Upgrading these two items alone would achieve almost all of the reduction in core damage frequency from the base case (6.0E-5 per year) to the lower value (5.0E-6 per year), assuming that implementation of feed and bleed is possible and procedures e for feed and bleed are in place. Any further reduction in risk would result (most effectively) from efforts to decrease the possibility of seismically-induced loss of offsite power. From this example, it is likely that a safety benefit of 34.5-2.4 = 32.1 man-rem /yr could be achieved by providing an alternative to using the RWST and by strengthening the station battery racks. For an assumed re-maining plant life of 30 years, the potential safety benefit could be ~ about 950 man-rem. If a general guideline of $1,000 per man-rem is used, i an expenditure of near one million dollars would be justified which is in ~ the same cost range as performing the A-46 seismic review and making the necessary modifications. NUREG-1211 23

e The'two examples from the USI~A-45 PRA studies represent possibly two ~ ' extremes, a plant where' few if any modifications would be required and a plant with several safety significant deficiencies. 3. Consideration of Alternatives l Several of the proposed procedures investigated as part of A-46 were determined. to be not feasible, or useful only to support other methods. They are discussed l in detail in NUREG-1030. As stated above, the use of seismic experience data was determined to be the most practical way.to demonstrate seismic adequacy. Only three alternative courses of action were' considered, as discussed below. A.- No Actions Required by Licensees This alternative was seriously considered because of the conclusion _that equip- ) ment in nuclear' plants is inherently rugged. The survey of seismic experience in non nuclear facilities that had undergone significant earthquakes indicated that if equipment were properly supported and anchored, it would be expected to survive without damage. Much of this equipment is identical or similar to nuclear plant equipment. This alternative was rejected (1) because the SEP experience during seismic reviews showed that there were some equipment seismic deficiencies particularly with respect to anchorages; (2) because there were several incidences of unanchored or improperly anchored equipment overturning or moving during a seismic event in the data base plants; and (3) because of staff consideration of the recommendations of SSRAP. In addition, the recent experience in nuclear plants discussed previously indicates that anchorage deficiencies still exist. B. Operating Plants Required To Comply with Current Licensing Criteria j l It was recognized from the start that it may not be cost effective or practical l to qualify operating plant equipment using current seismic qualification criteria l and methods because of (1) excessive plant down time, (2) difficulties in ship-ping irradiated equipment to a test laboratory, and (3) difficulties in acquiring identical old-vintage equipment for laboratory testing. In addition, the cost of meeting current criteria would be much greater for an old plant than for a l new one. Meeting current criteria, however, would meet the safety objective; j therefore, cost estimates are provided below for this alternative. I C. Verification of Seismic Adequacy Required by an Onsite Inspection of Anchorage and Supports and by Verifying Equipment Functional Capability During the Strong Shaking Motion Utilizing Seismic Experience Data and/or Test Experience Data This alternative takes advantage of experience gained from the review of facili-ties that have experienced strong motion earthquakes and also provides for ensuring that supports and anchorages are adequate. For equipment not in the seismic experience data plants, for equipment unique to nuclear plants, or for equipment needed to function during the first 30 seconds of an SSE, the test experience data base being developed by EPRI/RES can be used to assess equipment i seismic adequacy and/or functional capability during an SSE. l NUREG-1211 24

4. Costs of Alternatives / A. No Actions Required by Licensees There is no utility cost associated with this option. B. Operating plants Required To Comply With Current Criteria Experience gained from the application of current criteria in new license situa-tions is extrapolated to estimate cost for operating plants. The use of current requirements presents several complicating factors as follows: (1) Equipment would have to be removed'from the plant and sent to a test laboratory for testing. (2) Qualification procedures could result in costly plant shutdowns. (3) Some of the equipment would be irradiated, which would require special procedures for removal, testing, shipping, and reinstallation. It is estimated that the qualification procedure would involve about 40 pieces ef electrical equipment and 70 pieces of mechanical equipment. This is based on the assumption that only equipment required to bring the reactor to a safe shutdown condition is included. A rough estimate of the projected cost to upgrade an operating reactor to meet IEEE Standard 344-1975 is approximately $10 million. This is based on the following assumptions and estimates: (1) About 75% of equipment would require tests and analysis (i.e., structural integrity will not ensure functionality). (2) The average cost of test and analysis per piece of equipment (from Table 2) is $17,300. This amount assumes equipment can be tested in place. If a component is removed, shipped to a test laboratory, and tested, the cost would be much higher. In situ testing would be practical only in a limited number of cases. Recent experience of one utility is that for an ac dis-tribution panel or an instrument rack, the testing cost is $30,000. This does not include removal or shipping which would at least double the cost. For purposes of this estimate, $50,000 per component is assumed for a total of ($50,000)(75% x 110 components) or approximately $4 million. (3) About 10% of the equipment would need to be replaced. The average cost of replacement, based on Table 2 and correctin<j for a more realistic escalation due to inflation, is ($500,000 per piece of equipment) (10% x 110 components) or approximately $5.5 million. (4) 'ihe average cost of analyses for the 25% of the equipment for which analy-sis alone would be acceptable is approximately $20,000 per item or ($20,000) (0.25x 110) or $550,000. This estimate seems reasonable in light of industry experience solicited by the staff on approximate costs to comply with IEEE Standard 323-1974 and IEEE NUREG-1211 25

TABLE 2 a g COST ESTIMATES OF SEISMIC VERIFICATION ALTERNATIVES m m C) s Hfu HH Analysis Teet and Analysis Replacement Camperison Sgpert seedtffcatten Equipment Type High Lew Average Migh Low Average Mfgh Lew Average Mfgh Low Average Mfgh Low Average Air Circ Fan / Motor 10,000 6,000 8,000 44,500 9,900 15,300 75,000 3,500 13.500 600 100 200 7,000 1,300 2,600 Air Cond Unit 200.000 75,000 100,000 118,000 26,200 40,600 260,000 28.000 115.000 1.600 400 800 15,000 2,400 7.000 Cabinet 13,000 7,000 9,000 44,500 9.900 15.300 4,500 1,000 2.500 600 100 200 850 350 '500 600 90 400 600 100 200 350 230 275 Ctreutt Soard 44.500 9.900 15.300 32.580E 2,450E 27,000E 600 100 200 33.700 5.800 13,800 CROE Olesel Generator 200,000 75,000 100,000 118,000 26,200 40,600 750.000 250,000 500,000 2,000 400 1,200 88,600 24.800 49.400 1,300 200 900 600 100 200 370' 240 300 Inverter MSIV 18.000 12,000 15,000 53,600 11,900 18.400 350.000 140.000 200,000 600 100 200 37.400 13,100 21,600 Paness 13,000 7,000 9,000 44.500 9,900 15,300 30.000 1.000 7,000 600 100 200 1,870 -360 710 Small Horiz Pump / 23,000 14,000 17,000 44,500 9.900 15,300 95.000 6.000 54,000 1,200 200 400 8.100 1,460 4.4es Motor stadium Hortz Pump / 23,000 14.000 17,000 44.500 9,900 15,300 160.000 17.000 78,000 1,200 200 400 16,800 3.400 8.400 stator Large Hortz Pup / 23.000 14,000 17,000 44.500 9.900 15,300 245.000 31,000 125,000 1,200 200 400 25.200 5,200 12,800 ,o m Motor Sme11 Vert Pump / 26,000 17,500 22,000 44,500 9.900 15,300 42,000 7,000 24.000 900 100 300 12.100 3.040 6,350 pteter pendium Vert Pump / 26,000 17,500 22,000 44,500 9.900 15,300 87,000 30,000 59.000 900 100 300 18.900 5,200 10,200 Motor Large vert Pump / 26,000 17,500 22,000 44,500 9,900 15,300 160.000 50,000 100.000 900 100 300 31.600 8.500 16,ess Motor Rocks (Instr.) 13,000 7.000 9,000 44,500 9,900 15,300 3,300 750 1,900 600 100 200 800 350 SIS Racks (8at.) 13.000 7,000 9,000 44,500 9.900 15,300 5.000 1.100 2.800 600 100 200 870 360 See 7,500 800 3.400 600 100 200 970 350 579 ~ Strip Chart Rec. 800 130 560 600 100 200 350 230 200 Seleys 53,600 11.900 18.400 73.000 12.000 42,500 600 100 200 9,000 2,140 4,000 IIntal-Clad Sultchgear d 7,100 300' 3.200 600 100 200 600 230 438 ~ Yeltage Swttchgear 10,700 350 3,650 600 100 200 1.270 270 418 stater Centrol Center 1.300 500 1.000 600 _100 200 370 250-300 Transducer 27.400 6,100 9,400 8,500 1,500 5,500 600 100 200 1.530 500 920 Transformer Check Valve 6,000 2.000 4,000 27.400 6.100 9.400 9,000 150 4.800 600 100 200 1,150 350 763 Small Instr. Valve 6.400 3.200 4,800 26.800 6,000 9.200 300 90 125 600 100 200 330 230 260 Small Relier Valve 13.000 8.500 11,000 44.500 9.900 15,300 15,000 1.300 8,000 600 100 200 1,150 340 700 Large Reifei Valve 13,000 8.500 11,000 53,600 11,900 18.40n 45.000 5.200 25,500 600 100 200 3.400 760 1.920 Small Safety valve 11,000 6,500 9.000 44,500 9,900 15.300 6,000 2.800 4,500 600 100 200 1,030 460 670 Large Safety valve 11,000 6,500 9.000 53.600 11,900 18,403 35,000 6,000 14.000 600 100 200 2,500 660 1,200 Equipment with no estimate for a particular method is not suttable for que11fication by that method, a. b. Cabinet only. Contents of cabinet not included. c. E = a 1.000 d. 15 amp-240 V ac 3-pole circuit breaker. e. 600 V 3-phase ac 9 2 hp motor starter.

Standard 344-1975, for both environmental and seismic qualification. This y experience includes: f (1) The upgrade of reactor building pressure. transmitters at a multiunit oper-ating pressurized-water reactor (PWR) will cost about $200,000 for 9 trans-to mitters. The upgrade includes both environmental and seismic qualification and replacement of units. (2) The upgrade of power systems equipment to IEEE Standard 323-1974 (mainly L documentation) at a new PWR will cost the utility about $3.5 million. ) (3) One utility estimated that its share of the cost of a nuclear steam supply program (NSSS) program to upgrade IEEE equipment to IEEE Standard 323-1974 will cost $4 million. (4) At a Combustion Engineering System 80 plant, the estimate to update NSSS ) IEEE equipment to IEEE Standard 323-1974 will cost $15.0 million. (5) At a multiunit operating PWR, the utility estimated that it would cost $20 to $30 million to upgrade equipment to IEEE Standard 344-1975 (seismic qualification only). This estimate did not include documentation or plant down time. Table 2 presents representative costs to verify seismic adequacy. Initial comments on the cost table by an industry group indicate that equipment replace-ment costs are low by a factor of 3 to 5 and in some cases as high as 9. An explanation of the categories in Table 2 is as follows: ~, (1) Analysis The " Analysis" cost estimates were based on experience in estimating analy-sis jobs and on reviews of such analysis performed during Seismic Qualification Review Team (SQRT) audits of qualification reviews performed for operating license applications. Equipment for which no estimate for analysis is given is not suitable for qualification by analysis. (2) Test and Analysis The figures under " Test and Analysis" include the cost estimated by an NRC contractor to determine equipment / support dynamic characteristics via in situ testing. The analysis effort is gr~eatly reduced by using dynamic parameters determined by test. This estimate was compared to actual cost data from the private sector and shown to be high. This was attributed to two factors. First, the estimate was based on a single test per trip, while the actual data involved multiple tests per trip. Second, the esti-mate was based on a full reduction of data, which yields full mass and stiffness matrices in addition to the natural frequency, mode shape, and damping data actually obtained. The numbers in the estimate were reduced by a constant multiplier to account for these factors. Numbers in the " Low" column were obtained by a multiplier that yielded an estimate within 5% of the actual cost for a test contract involving 17 tests in a single trip. Numbers in the "High" column were obtained with a multiplier to account for the more complete data reduction included in the estimate. The numbers in the " Average" column were obtained with a multiplier to NUREG-1211 27

~' = y 3 9n-5 t' 1 . account-for the more complete' data reduction and to adjusttthe estimate to a five-test per-trip basis. 1(3) Replacement ~" Replacement" is the_ cost incurred to replace equipment with qualified equipment. ~This includes purchase of the equipment with qualification Ldocumentation and. installation. : It does not inclMe freight charge's.' Estimates:are primarily. based on," Process Plant-Construction Estimating Standards," by. Richardson Engineering Services, Inc. Two editions of the standard were used,lone: dated.1975 and the other 1981. Estimates taken from the-1975 edition were increased by 30%* to account for inflation. Two components on the list (main steam isolation! valves and control rod u

drive mechanisms) were not covered by the standard.

Estimates for these 'l two were obtained by-contact with vendors. Qualification ^ documentation was assumed to cost 150% of the cost of the unqualified components _for all but three of the components--small instrument valves, transducers,.and relays. These components are produced in large quantities-and required in large quantities in typical plants. Their qualification documentation is assumed to be less costly--50% of the cost of the unqualified component. 'C. Verification of Seismic Adequacy and Equipment Functional Capability Required Through an Onsite Inspection of Anchors and Supports and by Comparing Plant Equipment With Seismic Experience Data and/or Test Experience Data-Two alternatives are considered. If a utility participates in a generic program, the cost will-be substantially less than the cost for a utility that elects to not participate in a generic program. The " Comparison" estimate in Table 2 is the cost of comparing dynamic and func-tional characteristics between equipment in plant and that in the data base. The estimate is based on the assumption that necessary data are readily available. Therefore, no costs resulting from analysis or in situ testing have been included. The_ estimated costs to licensees by using this alternative are discussed below. 5. Estimated Costs to Licensees The least expensive procedure for verifying the seismic adequacy of components is comparison with the experience data. base. This procedure will work for many components; however, additional steps may be required for some components. The estimates presented in Table 2 assume a comparison of the required response spectra and dynamic characteristics of each component with the experience data base. A direct comparison on a component-by-component basis will probably be required for 10% or less of the components.

• Industry comments indicate that actual escalation rates between 1975 and 1984 may be as high as 90%.

For nuclear estimates, the 90% rate is usually multi-plied by 3 to 5, because it does not cover health physics, decontamination, respirator work, etc. NUREG-1211 28 t

If the utilities choose to adopt the generic implementation, costs to individual utilities would be much lower than the cost for each utility to provide a plant-specific verification of seismic adequacy. Shared costs of a generic resolution would depend on the number of utilities participating. The cost to the utility for a plant-specific verification will vary from plant to plant depending on the seismic design basis, the location of equipment, and the type of plant. The following estimate therefore presents a range of costs for each item. Most equipment is located in plant areas where radiation does not present serious problem. for inspection or modification. The cost estimates therefore do not include special considerations for radiation protection. A labor cost of $100,000 per person ymr is assumed. The estimates given in Table 3 do not include plant down time and are for a single power plant unit participating in a generic effort. On the basis of the estimates in Table 3, the total industry cost would be $28 million to $59 million for the approximately 70 plants (units) involved. In addition, the SQUG utilities have spent approximately $200,000 each developing the experience data base, and they anticipate spending an additional $35,000 each before plant-specific implementation. The additional costs are for develop-ment of detailed walk-through procedures, pilot walk-throughs, holding implemen-tation workshops, and documenting the basis for seismic capacity of equipment classes that are not treated in detail in the experience data base. Total SQUG expenditures would be about $2.5 million. If a utility decides to not participate in the proposed generic resolution, additional costs of preparing and submitting a review procedure and for preparing a plant-specific report and the review and audit by the NRC staff would be incurred. This would add an estimated $150,000 to $200,000 to each utility's cost and $10,000 to $30,000 per utility to NRC staff costs. A utility not participating in a generic resolution would be required to develop its own detailed walkdown procedure and would have to spend considerable re-sources in planning and executing the implementation. This plant-specific implementation procedure would have to be reviewed in detail and probably would require several iterations. The data bases and SSRAP and EQE data reports that have been made public would be available to all utilities. 6. Costs to the NRC The principal cost to NRC (for utilities not participating in a generic implemen-tation) would be the cost to review the reports submitted by individual licensees and participation in the plant audits. About 70 plants would be required to submit reports. It would require 0.6 staff month to review each report and 0.5 staff month to prepare an SER, for a total expenditure of 77 staff months. At an estimate rate of $100,000 per staff year, the total cost would be $640,000. If a generic program is implemented by SQUG or a similar utility group, the cost to the NRC would be substantially reduced. NUREG-1211 29 l t

I Table 3' Estimated costs-to a licensee Item Est.imated Cost (Dollars)/ plant . Define. systems, subsystems, and components $17,000 to $35,000 1 required and develop equipment list (2 people, 1 to 2 months) Compare data base spectra with site spectr~a .$ 4,000 to $10,000 (1 person, 1/2 to 1 month) Conduct plant walk-through (4 people,1 to $32,000 to $80,000 y 2 month) Repairs to' anchorage and supports (average $200,000 to $400,000 of $40,000 for support of a electrical or mechanical equipment, for 5 to 10 pieces) Identify relays needed to function during the $27,000 to $55,000 30 seconds of strong motion earthquake, and relays which could potentially chatter or change state during earthquake (2 people, 1 to 2 months, and $1000/ relay (assume 10 to 30 relays)) Miscellaneous modifications to components $10,000 to $20,000 to fit experience data ($2,000 for 5 to 10 items) Collection of test experience data (assumes $50,000 to $100,000 cost to a single utility participating in a generic effort) Generation of floor response spectra (assumes $50,000 to $100,000 a simplified analysis is used in lieu of full-fledged soil-structure interaction and finite element analysis) Auditing performed by an independent con- $ 8,000 to $20,000 tractor (2 people, 1/2 month to 1 month, which includes preparation before audit and documentation for audit) Preparation &nd submission of report to NRC $10,000 to $20,000 TOTAL $401,000 to $840,000 NUREG-1211 30 s . _ _ _. _ _.. _.,. - _ _ -. -..,.,. _. - _. ~. ~ _, _ _.., _.... _. _.. _.

7. Safety Benefit.s Compared to Costs The safety benefit of the proposed seismic verification program is the reduced likelihood of core nelt and radiation release as a result of the seismic failure of equipment required to safely shut down the plant follcwing a seismic event. The principal concern is equipment failure or loss of equipment function as a result of the failure of anchorage or supports or loss of shutdown system functions as a result of relay chatter. The experience data base plus the sur-vey of strong motion earthquakes conducted by SQUG and SSRAP indicate that anchorage failures are possible. Experience gained from SEP reviews and recent staff surveys also indicate that some anchorages in nuclear plants may be sus-ceptible to seismic failure. Although the incremental risk has not been specifically quantified in this study, there is evidence that potential for safety improvement exists. The examples cited in Section V, Paragraph 2 above indicate that substantial safety benefit can be achieved. The staff therefore concludes that substantial increase in the overall protection of the public health and safety can be achieved from the backfit and that the direct or indirect costs of implementation are justified in view of this increased protection. The approximate costs to achieve the safety benefit are (1) to impose current licensing requirements: $10,000,000/per plant (2) to implement a generic program using experience data: $401,000 to $840,000/per plant Because of the lower cost and the more effective treatment of anchorages offered by the proposed seismic verification program, the staff recommends that this program be implemented. 8. Impacts on Other Requirements The proposed requirement would have no impact on current licensing requirements because it would not change the implementation of current requirements on exist-ing license applicants or new license applicants. 9. Constraints Implementation of the proposed requirement could be affected by the limited amount and range of experience data currently included in the data base.

Also, applicable test data have not been collected and organized.

The implementation plan and schedule have, therefore, been developed with the assumption that additional test data will be collected and used to verify seismic adequacy. VI. APPLICATION OF THE BACKFIT RULE, 10 CFR 50.109 The NRC believes that the supporting analyses documented in this regulatory analysis complies with the provisions of 10 CFR 50.109. The following information is provided in answer to the specific information requirements in paragraph (c) of 10 CFR 50.109. 1. Statement of the specific objectives that the proposed action is designed to achieve: NUREG-1211 31 ) I

The specific objective of the proposed A-46 action is to ensure the safety of operating nuclear power plants by requiring the review of seismic adequacy of mechanical and electrical equipment which is required to safely bring the reactor and plant to a safe shutdown condition and to maintain it.in a safe condition. 2. General description cf the acthity that would be required by the licensee or applicant in order to complete the backfit: The resolution of USI A-46 is based mainly on the use of seismic experience and test experience data. The general verification procedure for plant-specific review is described 4 in Section IV.4 of this Regulatory Analysis. Briefly, it includes the following steps: 1 development of an equipment list comparison of site spectra with appropriate bounding spectra walk through inspection, which includes anchorage review, seismic system interaction review, identification and review of deficiencies and outliers. review of equipment functional capability review of equipment unique to nuclear plants + replacement (and/or modification) of equipment and/or equipment + supports. 3. Potential change in the risk to the public from the accidental off-site release of radioactive material: The safety benefit of verifying the seismic adequacy of equipment in operating plants was not specifically quantified in terms of risk reduction. Quantifying the net safety benefit in terms of risk resulting from " qualifying" or verifying the seismic adequacy of equipment proved to be impractical. However, to demonstrate the potential value/ impact, the staff has estimated the potential safety benefit in terms of the risk and the cost benefit potentially achievable (see Section V, Paragraph 2 above). These examples indicate that performing the A-46 review and cor-recting deficiencies can potentially result in reducing the radiation's release over the life of a typical plant by as much as 1,000 man-rem. 4. Potential impact on radiological exoosure of facility employees: A qualitative evaluation of the potential radiological exposure to facil!ty employees was made. The staff concluded that radiological exposure as a result of the review would be minimal for the following reasons: (1) Experience with walkdown type inspections similar to the required review has been that little or no radiological exposure has resulted. Plant personnel who will be involved in the review are routinely in the inspection areas and have free access to these areas. (2) The equipment to be reviewed is generally not located in contaminated areas and is not inside the contaiment building. NUREG-1211 32 I

I i-1- -(3) -A preliminary review of drawings and design information will be con-ducted prior to.the-in plant review which will minimize time spent in the plant. 5. Installation and continuing costs associated with the-backfit, including the cost of facility downtime or the cost of construction delay: ~ 'The' estimated costs-tol licensees for complying with the requirement ~of -A-46 are-presented-in Section V, Paragraph 5 of this Regulatory Analysis. The estimated cost per plant ranges from $401,000 to $840,000. The cost of-facility downtime is not included in this estimate. The implementation schedule will be negotiated with the-licensees in accordance with.the NRC L policy on integrated schedules for plant modifications stated in Generic Letter 83-20 dated May 9, 1983. The proper. integration of.the proposed work scope into each plant's living schedule may allow for the reviews to f .be conducted during' plant outages. 6. The potential safety impact'of changes in plant or operational complexity, including.the relationship to proposed and existing regulatory requirements: None.

7.

The estimated resource burden on the NRC associated with the proposed backfit and the availability of such resources: The cost _ to the NRC for the implementation of A-46 requirement is estimated in Section V, Paragraph 6 of this-Regulatory Analysis. The principal cost to NRC (for utilities not participating in a' generic implementation) would be the cost to review the reports submitted by individual licensee's and for participation in the plant audits. About ~ 70 plants would be required to submit' reports. It would require 0.6 staff months to review each report and 0.5 staff months to prepare an SER, for a total expenditure of 77 staff months. 'At an estimated rate of $100,000 per staff year, the total cost would be $640,000. If a' generic program is implemented by SQUG or a similar utility group, 'the cost'to the NRC would be substantially reduced. 8. The potential impact of differences in facility type, design or age on the relevancy and practicality of the proposed backfit: In general older plants would be expected to have more deficiencies in seismic design, however, most older plants have undergone specific upgrading such as resulted from the SEP program. 9. Whether the proposed backfit is interim or final, and if interim, the justification for impnsing the proposed backfit on an interim basis: The proposed backfit represents the final staff position on USI A-46. I NUREG-1211 33 (

VII. IMPLEMENTATION 1. Proposed Method of Implementation The proposed method of implementation is issuance of a generic letter. The staff is recommending implementation through issuance of a generic letter rather than through a Standard Review Plan. revision or issuance of a Regulatory Guide because the proposed requirements apply only to operating plants that have not been reviewed to current licensing criteria. They call for a one-time review of operating facilities and would not be a continuing requirement. 2. Schedule for Implementation of the Proposed Requirement i The proposed resolution presents the procedure for verifying the seismic adequacy of equipment using (1) seismic experience from non-nuclear facilities that have experienced strong motion earthquakes and (2) equipment test data. The SQUG program to date has been limited to eight classes of equipment. Implementation on the initial eight classes of equipment and a review of all anchorages can proceed as soon as the requirement is established. Additional work is required to collect test data on a number of equipment types. The SQUG has initiated a program to develop the additional information and to continue the SSRAP as an independent review group. Follow-on work needed to complete implementation includes (1) Developing the basis for the seismic adequacy of equipmer:t that is not included in the eight types in the seismic experience data base but that exist in the seismic experience data plants, as well as equipment unique to nuclear plants. (2) Adding qualification test data to the data base. The NRC staff will negotiate the implementation schedule with the generic group taking into consideration the NRC policy on integrated schedules for plant modifications as stated in Generic Letter 83-20 dated May 9, 1983. This policy was reiterated in NRC 1985 Policy and Planning Guidance dated February 26, 1985. Utilities electing to not participate in a generic resolution may negotiate implementation schedules with the staff individually. The actual schedule dates will be based on negotiations with each utility. The proper integration of the proposed work scope into each plant's living schedule for plant modifications will be considered. Table 4 shows the estimated times for completion of each program element. I NUREG-1211 34

b. 8 I.- 1 Table 4-Estimated. time for implementation of proposed requirement' V Elapsed time from date Item. of requirement (months) .g i

Generic group completes final walkdown procedure.

12 and conducts workshops. Start of implementation.. Identify systems, 17 subsystems, and components; conduct walk-through inspection of all anchorages and equipment other ^ than those required to collect additional test data. Complete necessary modifications to all anchorages 23'

and equipment.

Assess seismic adequacy and/or functional capability 36 of equipment and component (including relays) for which collection of.-additional test data is required. Provide report to NRC. 40 3. Relationship to Other Existing or Proposed Requirements The proposed requirement would be imposed on existing plants that were not reviewed to current requirements as an alternative to requiring those plants to meet current requirements. VIII. REFERENCES Brookhaven National Laboratory, " Guidelines for Identification of Seismically Risk-Sensitive Systems 'and Components in a Nuclear Power Plant," draft report, September 1983. EQE, "The Performance of Industrial Facilities and Their Equipment in the Coalinga, California Earthquake of May 2, 1983," August 1984. SSRAP, "Use of Past Earthquake Experience Data to Show Seismic Ruggedness of Certain Classes of Equipment in Nuclear Power Plants," January 1985. U.S. Nuclear Regulatory Commission, Board Notification 83-01A, " Seismic Risk to' Boiling Water Reactor Plants," October, 1983. --, Generic Letter 81-14, February 10, 1981. --, Generic Letter 83-20, May 9, 1983. --, IE Bulletin No. 79-02, " Pipe Support Base Plate Designs Using Concrete Expansion Anchor Bolts," March 1079. NUREG-1211 35

( --, IE Bulletin No. 79-07, " Seismic Stress Analysis of Safety-Related Piping," April 1979. --, IE Information Notice 79-14, " Seismic Analyses for As-Built Safety-Related Piping Systems," July 1979. --, IE Bulletin No. 80-11, " Masonry Wall Design," May 1980. --, IE Information Notice 80-21, " Anchorage and Support of Safety-Related Electrical Equipment," May 1980. --, NUREG/CR-2405, " Subsystem Fragility,"' Lawrence Livermore National Labora-tory, February 1982. --, NUREG/CR-3266, " Seismic and Dynamic Qualification of Safety-Related Elec-trical and Mechanical Equipment in Operating Nuclear Power Plants--Development of a Method to Generate Generic Floor Response Spectra," Brookhaven National Laboratory, September 1983. --, NUREG/CR-3357, " Identification of Seismically Risk Sensitive Systems and Components in Nuclear Power Plants," Brookhaven National Laboratory, June 1983. NUREG-1211 36

L f ~l '1' ENCLOSURE I Operating Plants To Be Reviewed To USI A-46 Requirement This plant list was developed by determining from plant Safety Evaluation Reports whether or not'a seismic qualification review has been performed.using IEEE Standard 344-1975. Plants not documented as meeting the provisions of- .IEEE Standard 344-1975 are included on the list. Alabama

• 1.'

Browns Ferry, Unit 1

• 2.

Browns Ferry, Unit 2

• 3, Browns Ferry, Unit 3 4.

Joseph M. Farley, Unit 1 Arkansas y

• 5.

Arkansas Nuclear One, Unit 1

• 6.

Arkansas Nuclear One, Unit 2 California

• 7.

San Onofre, Unit 1

• 8.

Rancho Seco, Unit 2 Colorado 9. Fort St. Vrain Connecticut

• 10. Haddam Neck
• 11. Millstone, Unit 1
• 12. Millstone, Unit 2 Florida

13. Turkey Point, Unit 3

14. Turkey Point, Unit 4

• 15. Crystal River, Unit 3

16. St. Lucie, Unit 1 Georgia

• 17. Edwin I. Hatch, Unit 1
• 18. Edwin I. Hatch, Unit 2 l
• Plant of utility which is a member of SQUG.

I l NUREG-1211 1 Enclosure I l

r.. Illinois-- -*19. Dresden,. Unit 2

• 20.-Dresden, Unit 3 j
• 21. Zion; Unit 1.-
• 22. Zion, Unit 2'

~*23.' Quad-City,_ Unit 1-

• 24.. Quad-City, Unit 2 Iowa
• 25. Duane Arnold, Unit 1-Maine
• 26. Maine Yankee Maryland
• 272: Calvert Cliffs, Unit 1-
• 28. Calvert Cliffs, Unit 2 Massachusetts

-*29. Yankee Rowe

• 30. Pilgrim, Unit 1 Michigan
• 31. Big Rock Point
• 32.' Palisades
• 33. Donald C. Cook, Unit 1
• 34. Donald C. Cook, Unit 2 Minnesota
• 35. Monticello
• 36. Prairie Island, Unit 1
• 37. Prairie Island, Unit 2 Nebraska

$38. Fort Calhoun, Unit-1

• 39. Cooper New Jersey
• 40. Oyster Creek, Unit 1
• 41. Salem, Unit 1
• 42. Salem, Unit 2 l

NUREG-1211 2 Enclosure I

-.7. New-York

• 43.-Indian Point, Unit 2
• 44. Indian Point, Unit 3
• 45. Nine Mile Point, Unit 1
• 46. R. E. Ginna, Unit 1
• 47.: James'A. Fitzpatrick North Carolina
• 48. Brunswick, Unit 'l-
• 49.-Brunswick, Unit 2 Ohio
• 50. Davis-Besse, Unit 1 Oregon

51. Trojan, Unit 1 Pennsylvania

• 52. Peach Bottom, Unit 2
• 53. Peach Bottom, Unit 3

54. Beaver Valley, Unit 1

• 55. Three Mile Island, Unit 1 South Carolina
• 56. H. B. Robinson, Unit 2
• S7. Oconee, Unit 1
• 58. Oconee, Unit 2
• 59. Oconee, Unit 3 Tennessee
• 60. Sequoyah, Unit 1
• S1. Sequoyah, Unit 2 Vermont'
• 62. Vermont Yankee Virginia
• 63. Surry, Unit 1
• 64. Surry, Unit 2 865. North Anna, Unit 1
• 66. North Anna, Unit 2 NUREG-1211 3

Enclosure I I I

' Wisconsin

67. Lacrosse L*68. Point Beach, Unit 1

• 69. Point Beach, Unit 2 e
• 70. Kewanee NUREG-1211 4

Enclosure I i y

i APPENDIX A GENERIC LETTER (Reference USI A-46) T0: - All Holders of Operating Licenses Not Reviewed to Current Licensing Criteria on Seismic Qualification' of Equipment GENTLEMEN:

SUBJECT:

VERIFICATION OF SEISMIC ADEQUACY OF MECHANICAL AND ELECTRI-CAL EQUIPMENT IN OPERATING REACTORS, UNRESOLVED SAFETY ISSUE (USI) A-46 As a result of the technical resolution of USI A-46, " Seismic Qualification of Equipment in Operating Plants," the NRC has concluded that the seismic ade-quacy of certain equipment in operating nuclear power plants must be reviewed against seismic criteria not in use when these plants were licensed. The tech-nical basis for this conclusion is set forth in References 1 and 2. Direct application of current seismic criteria to older plants could require extensive, and probably impracticable, modification of these facilities. An alternative resolution of this problem is set out in the enclosure to this letter. This approach makes use of earthquake experience data supplemented by test results to verify the seismic capability of equipment below specified earthquake motion bounds. In the staff's judgment, this approach is the most reasonable and cost-effective means of ensuring that the purpose of General Design Criterion 2 (10 CFR Part 50 Appendix A) is met for these plants. Because affected plants are being asked to carry out this evaluation against criteria not used to establish the design basis of the facility, this resolution is a backfit under 10 CFR 50.109. The backfit analysis and findings may be found in the Regulatory Analysis (Reference 2) at pp. 31. Seismic verification may be accomplished generically, as described in the enclosure. Utilities participating in a generic program should so state in their reply to this letter, isientifying the utility group and the schedule established for completion of the effort. The implementation schedule will be negotiated with utility groups or individual utilities in accordance with the NRC policy on irtegrated schedules for plant modifications. See Generic Letter 83-20, May 9, 1983. Utilities not participating in a generic review may be allowed some additional time to complete the review. 1 I i

We therefore request that you provide within 60 days of receipt of this letter a schedule for implementation of the seismic verification program at your facility. Sincerely, Harold R. Denton, Director Office of Nuclear Reactor Regulation

Enclosure:

Seismic Adequacy Verification Procedure

References:

(1) NUREG-1030, " Seismic Qualification of Equipment in Operating Nuclear Power Plants (USI A-46)," August 1986 (2) NUREG-1211, " Regulatory Analysis for Proposed Resolution of Unresolved Safety Issue A-46, Seismic Qualification of Equipment in Operating Plants," August 1986 4 4 4 2 l o o r

ENCLOSURE SEISMIC ADEQUACY VERIFICATION PROCEDURE The proposed procedure for verifying seismic adequacy of equipment is addressed in the following paragraphs. Each licensee will be required to perform the v'erification steps and submit a report to the NRC including an affidavit that the verification has been completed and all equipment within the scope defined below has been found to be acceptable. A generic resolution will be accepted in lieu of a plant-specific verification review subject to the guidance presented herein. 1. Scope of Seismic Adequacy Review Each licensee will determine the systems, subsystems, components, instrumenta-tion, and controls required during and following a design-basis seismic event using the following assumptions: (1) The seismic event does not cause a loss-of-coolant accident (LOCA), a steam-line-break accident (SLBA), or a high-energy-line-break (HELB), and a LOCA, SLBA, or HELB does not occur simultaneously with or during a seis-mic event. However, the effects of transients that may result from ground shaking should be considered. (2) Offsite power may be lost during or following a seismic event. (3) The plant must be capable of being brought to a safe shutdown condition following a design-basis seismic event. The equipment to be included is generally limited to active mechanical and elec-trical components and cable trays. Piping, tanks, and heat exchangers are not included except that those tanks and heat exchangers that are required to achieve and maintain safe shutdown must be reviewed for adequate anchorage. Seismic system interaction is included in the scope of review to the extent that equipment within the scope must be protected from seismically induced physical interaction with all structures, piping, or equipment located nearby. Lessons learned from studies of nuclear and nonnuclear facilities under earth-quake loading indicate that the effect of failure of certain items--such as suspended ceilings and lighting fixtures--could influence the operability of equipment within the scope of reviews. Instrument air lines and electrical and instrumentation cabling must be verified to have sufficient flexibility from the connection to equipment so that relative movement of anchor points will not jeopardize their integrity. Air lines and electrical and instrument cabl-ing are not included in the sco)e of review except for that portion from the equipment item to the first anclor point. The failure of masonry walls that could affect the operability of nearby safety-related equipment is of concern. However, this concern has been addressed by IE Bulletin 80-11, which requires that all such masonry walls be identified and re-evaluated to confirm their design adequacy under postulated loads and load combinations. This concern is, therefore, not considered as part of A-46 implementation. The required seismic interaction reviews will be based on, and consistent with, observations made in the seismic experience data base augmented by expert judgement of NUREG-1211 1 Appendix A Enclosure t~ s

SQUG/SSRAP. The review procedures will be reviewed by the NRC staff and SSRAP prior to plant specific implementation. For some pressurized water reactor plants, the seismic adequacy of auxiliary feedwater (AFW) systems has been verified by licensee actions taken in response to Generic Letter 81-14, dated February 10, 1981. Review of the AFW systems may be deleted from consideration under USI A-46 if staff acceptance has been documented in a Safety Evaluation Report or if the licensee has committed to meet the requirements of the generic letter. For the purpose of seismic adequacy verification, the following guidance is given. Each licensee must identify equipment necessary to bring the plant to a hot shutdown condition and maintain it there for a minimum of 72 hours. The 72-hour period is sufficient for inspection of equipment and minor repairs, if necessary, following a safe-shutdown earthquake (SSE) or to provide additional sour:e(s) of water for decay heat removal, if needed, to extend the time at hot shutdown. Equipment required includes that necessary to maintain the supporting functions required for safe shutdown. For all equipment within the defined scope, the verification must closely follow the procedure outlined in paragraph 2 below. Each licensee must show practical means of staying at hot shutdown for a minimum of 72 hours. If maintaining safe shutdown is dependent on a single (not redun-dant) component whose failure, either due to seismic loads or random failure, would preclude decay heat removal by the identified means, the licensee must show that at least one practical alternative for achieving and maintaining safe shutdown exists that is not dependent on that component. Each licensee must develop an equipment list. This list will include all equip-ment within the required scope. The equipment to be considered depends on the functions required to be performed. Typical plant functions would include: (1) Bring the plant to a hot shutdown condition and establish heat removal. (2) Maintain support systems necessary to establish and maintain hot shutdown. (3) Maintain control room functions and instrumentation and controls necessary to monitor hot shutdown. (4) Provide alternating current (ac) and/or direct current (dc) emergency power as needed on a plant-specific basis to meet the above three functions. 2. General Verification Procedure for Plant-Specific Review The licensee will be required to conduct a plant walk-through and visual inspec-tion of all identified equipment items necessary to perform the functions related to plant shutdown. The inspection team must consist of as a minimum, (1) an experienced structural engineer familiar with seismic anchorage requirements NUREG-1211 2 Appendix A Enclosure I

(2) an experienced mechanical engineer familiar with plant mechanical equipment (3) an experienced electrical engineer familiar with plant electrical equipment Furthermore, an operations supervisor or a licensed Senior Reactor Operator must be available for consultation before and during the walk-through process. Not all members of the inspection team are required to participate in all parts of the walk-through; however, appropriate technical expertise must be included for each review area, and a person with proper structural background must always be present to inspect the anchorage for all equipment. As an alternative, licensees may use consultants instead of their staff for (1), (2), and (3) above. Before the walk-through inspection, the licensee will be required to verify that the appropriate data base spectra envelope the site free-field spectra at the ground surface defined for the plant. There are a number of nuclear plants whose free field SSE spectra are defined at the foundation level, for these plants, an estimate of the free field spectra at the ground surface must be made for comparison with the data base bounding spectra. The ifcensee must identify all equipment on the plant's equipment list that is located at an elevation higher than 40 feet above grade level.* For equipment above 40 feet, one-and-one-half times the appropriate data base bounding spectrum (defined in paragraph 6 below) must envelope the floor response spectra for the equipment. For those cases where floor response spectra are needed, NUREG/CR-3266, " Seismic and Dynamic Qualification of Safety-Related Equipment in Operating Nuclear Power Plants: The Development of a Method to Generat'e Generic Floor Response Spectra," may be used as one alternative to develop the necessary floor response spectra on a case-specific basis. The appropriate bounding spectra for equipment belonging to the original eight types in the data base are defined in paragraph 6 below. For equipment types that are not included in the eight types in the data base but that exist in the data base plants, and for equipment unique to nuclear plants, the appropriate bounding spectra are defined in paragraph 7 below. The walk-through inspection must cover anchorage review and identification of potential " deficiencies" and " outliers." " Deficiency" in this context means equipment, components, and their anchorages / supports that are identified as obviously inadequate by the A-46 criteria during plant-specific walk-through reviews and confirmed as inadequate by further engineering studies. " Outlier" in this context means equipment items that are subject to the caveats and ex-clusions defined in this generic letter, or that are otherwise not covered by the experience data. The treatment of deficiencies is further described in paragraphs 4 and 5 below. The walk-through inspection must cover the following: (1) For equipment within scope, verify equipment anchorage (including required cable trays, tanks, and heat exchangers) using the guidance provided in paragraph 3 below and identify potential deficiencies. Utilities partici-patinginagenerIcimplementationmayusethewalk-throughprocedures being developed by SQUG/EPRI when these are approved by SSRAP and NRC.

• " Grade level" is the top of the ground surrounding the building.

NUREG-1211 3 Appendix A Enclosure . fC_ _ _ _ _ _ _ _ __ _ _ _ __ _J

(2) For equipment belonging to the initial eight types in the c.ata base, ident-ify data base exclusions and caveats (outliers) from the guidance provided in paragraph 6 below. (3) For equipment types that exist in the data base plants but that are not included in the eight types in the data base, the guidelines provided in paragraph 7 below and the guidelines being developed by SQUG (to be approved by SSRAP and NRC prior to implementation) must be used for identification and review of " outliers" and " caveats" during the walk-through inspection for this equipment. The licensee must specify all equipment items that are required to function during the period of strong shaking. The licensee must demonstrate the oper-ability of these items by means other than comparison with the experience data base; otherwise, the licensee must determine that any change of state will not compromise plant safety. The period of strong shaking is defined to be the first 30 seconds of the seismic event and should be considered in conjunction with the loss of offsite power. On the basis of the seismic experience data gathered to date, the only concern remaining on equipment functional capability is the concern regarding relays. Contactors and switches are considered as relays in this context. In addition, mercury switches are known to malfunction during testing and should be replaced by other types of qualified switches. Unless the test data being collected by the Electric Power Research Institute (EPRI) and the NRC Office of Research (RES) reveal otherwise, certain types of relays are the only equipment whose functional capability will need to be verified. The essential plant functions that are required to achieve and maintain hot shutdown during and after an SSE must be identified. The associated systems and electrical circuits required to provide these functions must then be identified. Next, these functions must be evaluated and the essential relays must be identified. Essential relays are relays that must remain functional without chatter during an SSE. These essential relays must be qualified by test, in a manner consistent with current licensing requirements (Section 3.10 of the Standard Review Plan (NUREG-0800), NRC Regulatory Guide 1.100/IEEE Standard 344-1975), verified by comparison with the test data base being developed by EPRI/RES, or replaced by relays qualified to current licensing requirements. As an alternative, the redesign of circuitry, strengthening of relay supports / cabinets to reduce seis-mic demand, or relocation of relays to reduce demand can be used. l The lice'see must identify all relays that could potentially change state during n an SSE as a result of contact chatter and preclude use of equipment needed l after the SSE to place the plant in safe shutdown. The redesign of circuitry, l strengthening of relay supports / cabinets to reduce demandAsanalternatIve,relocat or l relays to reduce demand can also be used. the licensee may i show that chattering or change of state of the relays does not affect system l performance or preclude subsequent equipment or system functions. In addition, j credit can be taken for timely operator action to reset the relays in case change of state occurs during an SSE, provided detailed relay resetting procedures are developed and there is sufficient time for resetting the relays. t HUREG-1211 4 Appendix A Enclosure 1 f 4

For components included in the data base by type but outside the limits of ex-perience data or test data, or of a type not included in the data base, as a general guideline the seismic verification can be deferred until additional test data is developed, endorsed by SSRAP, and approved by the NRC staff, provided that the seismic verification is completed no later than about 36 months from the date of issuance of the USI A-46 final resolution. Actual schedule dates will be based on negotiations with the generic group or with individual utili-ties. The proper integration of the proposed work scope into each plant's schedule for plant modifications will be considered. If a utility replaces components for any reason, each replacement (assembly, subassembly, device) must be verified for seismic adequacy either by using A-46 criteria and methods or, as an option, qualifying by current licensing criteria. This provision also applies to future modification or replacements. " Component" in this context means equipment and assemblies (including anchorages and sup-ports)--such as pumps and motor control centers--and subassemblies and devices-- such as motors and relays that are part of assemblies. 3. Verification of Anchorage To verify acceptable seismic performance, adequate engineered anchorage must be provided. There are numerous examples of equipment sliding or overturning under seismic loading because anchorage was absent or inadequate. Inadequate anchorage can include short, loose, weak, or poorly installed bolts or expansion anchors; inadequate torque on bolts; and improper welding or bending of sheet metal frames at anchors. Torque on bolts can normally be ensured by a preventive maintenance and inspection program. In general, checking of equipment anchorages requires the estimation of equipment weight and its approximate center of gravity. Also, one must either estimate the fundamental frequency of the equipment to obtain the spectral acceleration at this frequency or else use the highest spectral acceleration for all fre-i quencies. When horizontal floor spectra exist, these spectra may be used to obtain the equipment spectral acceleration. Alternatively, for equipment mounted less than about 40 feet above grade, one-and-a-half times the free-field hori-zontal design ground spectrum may be used to conservatively estimate the equip-ment spectral acceleration. For equipment mounted more than about 40 feet above grade, floor spectra must be used. This restriction may be modified if addi-tional data become available to justify raising the 40-foot-limit. Equipment anchorage must not only be strong enough to resist the anticipated forces but must also be stiff enough to prevent excessive movement of the equip-ment and potential resonant response with the supporting structure. The review of anchorages should include consideration of both strength and stiffness of the anchorage and its component parts. Additional discussions on seismic motion bounds and equipment supports and anchorage for each of the original eight classes of equipment in the experience data base are included in paragraph 6 below. This guidance supplements the general guidance above. During the walk-through inspection, anchors and supports of equipment within the scope of review will be carefully inspected. The detailed guidance devel-oped is the preferred method for review of anchorages. The detailed guidance has been developed jointly by SQUG and EPRI. It was approved by SSRAP and is NUREG-1211 5 Appendix A Enclosure

being reviewed by the NRC staff. It will be approved by the NRC staff before implementation. If the adequacy of supports and anchors cannot be determined by inspection, an engineering review of the anchorage or support will be con-ducted. This engineering review will include a review of design calculations or the performance of new calculations and/or verification of fundamental frequency of equipment to ensure adequate restraint and stiffness. Physical modifications may be necessary if engineering review determined the anchorage or support to be inadequate. 4. Generic Resolution The NRC will endorse and encourage a generic resolution of USI A-46 provided the guidelines presented below are followed: (1) All member utilities of the SQUG would be eligible to participate. (2) The generic group must be responsible (a) for developing procedures to identify relays to be evaluated, (b) for defining functionality require-ments, and (c) for developing evaluation procedures for relays. This pro-cedure must be reviewed and endorsed by SSRAP and the NRC staff. (3) The generic group must submit to the NRC a generic schedule for the de-velopment of implementation procedures and for workshops / training seminars for participating utilities. A pilot walk-through must be conducted on a few selected plants to test the procedure. Afterwards, the generic group must hold workshops / training seminars for participating utilities to ensure uniformity in approach. Each individual utility must submit an implementa-tion schedule to the NRC within 60 days of receipt of the A-46 generic letter. Individual utilities must then perform the plant-specific imple-mentation reviews. (4) Each utility must submit to the NRC an inspection report that must include: certification of completion of the review, identification of deficiencies and outliers, justification for continued operation (JCO) for identified deficiencies if these deficiencies are not corrected within 30 days, mod-ifications and replacements of equipment / anchorages (and supports) made as a result of the reviews, and proposed schedule for future modifications and replacements. The objective of the requirement to submit a JC0 is to provide assurance that the plant can continue to be operated without endangering the health and safety of the public during the time required to correct the identified deficiency. The JC0 may consider arguments such as imposition of administrative controls or limiting conditions for operation (LCOs) or consideration of the impor-tance of the safety function involved and/or identification of alternate means to perform that function. (5) Consultants to the generic group must perform audits of plant-specific reviews. All plants must be audited. The NRC staff will participate in plan't audits on a selective basis. The generic group must submit a report of audits performed and results of these audits to the NRC. This report covers all participating utilities, and must also include the results of any reviews and/or audits performed by the SSRAP. NUREG-1211 6 Appendix A Enclosure l l

(6) The SSRAP and the NRC staff must perform a limited review of the generic group audit process to evaluate effectiveness. (7) Final approval of the implementation will be made by the NRC in the form of a plant-specific Safety Evaluation Report for each affected plant after NRC receives a final report from the utility involved certifying completion of implementation reviews and equipment / anchorage modifications and replacements. (8) The generic group must provide for the continuation of the SSRAP as an independent review body. The SSRAP must be consulted during the develop-ment of the generic program and walk-through procedure, and must audit the implementation. (9) NRC staff members must be invited to participate in all meetings between the generic group and the SSRAP. 5. Provisions for Resolution for Individual Utilities The generic resolution described in paragraph 4 above, Generic Resolution, is the method preferred by the NRC for the resolution of A-46. This paragraph offers provisions for resolution of A-46 for individual utilities not partici-pating in the generic group. Each utility must develop a detailed review procedure that must be submitted to the NRC staff for review. This procedure must reflect the guidance given in paragraph 2 above. The data and procedures developed by the SQUG will not, in general, be available to non participating utilities. All information that has been made publicly available by SQUG or the staff can be used. Each utility must perform plant-specific verification reviews according to guid-ance in paragraphs 2 and must also maintain an auditable record of implementation of USI A-46. Within 60 days of receipt of the A-46 generic letter, each utility must submit to the NRC a schedule for implementation of the A-46 requirements. Utilities who may not have access to SQUG implementation procedures or data base may have difficulty in establishing implementation schedules within 60 days. For these utilities the NRC will negotiate time extensions on a case by case basis. The utility must submit an inspection report to the NRC after the plant-specific walk-through inspection. It should consist of the following: (1) Certification of completion of the walk-through inspection and a description of the procedures used. (2) A list of the equipment included in the review scope. Equipment required to function during the strong shaking period should be identified. (3) Identified deficiencies. (4) Identified outliers. (5) Modifications and replacements of equipment / anchorages (and supports) made as a result of the inspection. NUREG-1211 7 Appendix A Enclosure

(6) Thefproposed schedule for future modifications'and replacements. l(7) A JC0 for identified deficiencies if these deficiencies are not corrected' within 30 days. Following the completion of implementation reviews and all necessary modifica-tions and replacements of equipment / anchorages, the utility must submit a1 final . report'to the NRC. A description of the procedures used for the implementation reviews and-the modifications and replacements must be included. The NRC will review the inspection procedure, inspection report, and the final report and will audit all plant-specific reviews before granting final NRC approval. The final NRFapproval will be in the~ form of plant-specific SERS. ' 6. - Guidance on Use of Seismic Experience Data for the Eight-Equipment Types in the Experience Data Base" (1) Seismic Motion Bounds To compare the potential performance of equipment at a given nuclear power plant with the actual performance of similar equipment in'the data base plants in recorded earthquakes, SSRAP has developed seismic motion bounding spectra to facilitate comparison. The purpose of these bounding spectra is to compare the potential seismic exposure of equipment in a nuclear power plant with the esti-mated ground motion that similar equipment actually resisted in earthquakes described in the data base. For convenience, the bounding spectra are expressed in terms of ground response at the nuclear site rather than floor response or equipment response. These bounding spectra represent approximately two-thirds of the free-field ground motion to which the data base equipment was actually exposed. Three different seismic motion bounds (types A, B, and C) are used. Different bounding spectra were developed, not to infer different ruggedness of equipment, but to represent the actual exposure of significant numbers of each class of equipment within the data base to ground motion. These bounds are defined in terms of the 5% damped horizontal ground response spectra shown in Figure A-1. The seismic motion bounds may be used for the equipment class as defined below. Equipment Class Bound Motor control centers Low-voltage (480-V) switchgear Type B Metal-clad (2.4 to 4-kV) switchgear Unit substation transformers

• Guidance in this paragraph is based on the SSRAP report dated January 1985.

l The SQUG is in the process of expanding the data base to include more recent i earthquake experience and 20 classes of equipment which cover all the equip-1 ment needed for plant hot shutdown. The SSRAP report also is being revised I accordingly. The final guidance in the SSRAP report may differ from that l mentioned here. The revised SSRAP report should be followed for implementa-tion guidance. I i l NUREG-1211 8 Appendix A Enclosure

4 g s TEquipment Class Bound Motor-operated valves with large eccentric-operator-Type _C lengths-to pipe-diameter: ratios Motor-operated valves (exclusive of those with large eccentric-operator- - lengths-to pipe-diameter ratios) Air-operated valves Type A Horizontal pumps and their motors Vertical pumps and their motors These spectrum bounds are intended for comparison with the 5% damped design . horizontal around response spectrum at a given nuclear power plant. In other words, if tie horizontal ground response spectrum for the nuclear plant site is less than a bounding spectrum at the approximate frequency of vibration of the equipment and at all greater frequencies (also referred to as the frequency range of interest), then the equipment class associated with that spectrum is considered to be included within the scope of this method. Alternately, one may_ compare 1.5 times these spectra with a given 5% damped horizontal floor spectrum in the nuclear plant. The comparison of these seismic bounds with the design horizontal ground response spectrum is judged to be acceptable for equipment mounted less than about 40 feet

• above grade (the top of the ground surrounding the building) and for moderately stiff structures.

For equipment mounted more than about 40 feet above grade, comparisons of 1.5 times these spectra with the horizontal floor spectrum is necessary. In all cases such a comparison with floor spectra is also acceptable. The vertical component will not be any more significant relative to the horizon-tal components for nuclear plants than it was for the data base plants. There-fore, it was decided that seismic bounds could be defined purely in terms of horizontal motion levels. The criteria are met so long as the 5% damped horizontal design spectrum lies below the appropriate bounding spectrum at frequencies greater than or equal to the fundamental frequency range of the equipment. This estimate can be made judgmentally by experienced engineers without the need for analysis or testing. The recommendation that the seismic bounding spectrum can be compared with the horizontal design ground response spectrum for equipment mounted less than about 40 feet above grade is based upon various judgments concerning how structures respond in earthquakes. However, this 40-foot above grade criterion must be applied with some judgment because some structures may respond in a different manner. (2) Motor Control Centers Motor control centers contain motor starters (contactors) and disconnect switches. They also provide over-current relays to protect the system from l.

• In most cases where numerical values are given in this section they should be considered as either " approximate" or "about," and a tolerance about the stated value is implied, i

NUREG-1211 9 Appendix A Enclosure I f L

overheating. In addition, some units will contain small transformers and dis-tribution panels for lighting and 120 V utility service. Motor control centers of the 600-V class (actual voltage is 480-V) are con-sidered. The general configuration of the cabinets must be similar to those specified in the Standards of the National Electrical Manufacturers Association (NEMA). This requirement is imposed to preclude unusual designs not covered in the data base. Cabinets that are configured similarly to NEMA standards will perform well if they are properly anchored. Cabinet dimensions and material gauges need not exactly match NEMA standards. On the basis of a review of the data base and anticipated variations in condi-tions, it appears that the motor control centers are sufficiently rugged to survive a seismic event and remain operational thereafter provided the following conditions exist in the nuclear facility: (a) The spectrum for the nuclear facility is less than the type B bounding spectrum described in Figure A.1 for frequencies above the estimated funda-mental frequency of the cabinet, and the motor control center is located less than 40 feet above exterior grade and has stiff anchorage, as discussed below. If the motor control center is located higher than 40 feet above exterior grade or does not have stiff anchorage, the floor spectrum shall be compared to 1.5 times the type B bounding spectrum. In all cases a comparison with floor spectra is also acceptable. (b) The cabinets have stiff engineered anchorage. Both the strength and stiff-ness of the anchorage and its component parts must be considered. Stiffness can be evaluated by engineering judgment based on the cabinet construction and the location and type of anchorage, giving special attention to the potential flexibility between the tiedown anchorage and the walls of the cabinet. One concern is with the potential flexibility associated with bending of a sheet metal flange between the anchor and the cabinet wall. Stiffly anchored cabinets will have a fundamental frequency greater than aoout 8 Hz under significant shaking. The intent of this recommendation is to prevent excessive movement of the cabinet and to ensure that under earthquake excitations the natural fre-quency of the installed cabinet will not be in resonance with both the frequency content of the earthquake and the fundamental frecuency of the structure, thereby allowing comparison of the ground response spectra with the type B bounding spectrum. (c) Cabinets with suffi'ciently strong anchorage that do not have the stiff anchorage as recommended above are still considered in the data base; however, the floor response spectrum must be compared to 1.5 times the type B bounding spectrum. (d) Cutouts in the cabinet sheathing are less than about 6 inches wide and 12 inches high including side sheathing between multi-bay cabinets. (e) All internal subassemblies are securely attached to the motor control cabinets that contain them. NUREG-1211 10 Appendix A Enclosure

1.2 nS 1.0 5% DAMPING B h 0.8 h TYPE A TYPE B w U 0.6 U T j YPE A & B N o,4 TYPE C 0.33 W $ 0.2 0.20 i I I I I I 0.0 O 4 8 12 16 20 24 28 FREQUENCY (Hz) i i 4 l Figure A.1 Seismic motion bounding spectra, horizontal ground motion i i NUREG-1211 11 Appendix A Enclosure

R - (f). Adjacent sections of multi-bay cabinet assemblies are bolted together. (g) Equipment and their enclosures mounted externally to motor control center cabinets and supported by them have a total weight of less than 100 pounds. Functional capability (that is, inadvertent change of state or failure to change state on command of relays during an earthquake)-is not considered here. Func-tional capability must be established by other means. The structural integrity of relays contained in the motor control centers and their ability to function properly after earthquakes, as defined in Figure A.1, has been demonstrated. (3) Low-Voltage Switchgear Low-voltage switchgear consists of low voltage (600 V or less) distribution busses, circuit breakers, fuses, and disconnect switches. Low-voltage switchgear of the 600-V class (actual voltage is 480-V) is con-sidered. The general configuration of cabinets must be similar to those speci-fled in Standard C37.20 of the American National Standard Institute (ANSI). This requirement is imposed to preclude unusual designs not covered in the data base. Cabinets that are configured similarly to those defined in the ANSI standards will perform well if they are properly anchored. Cabinet dimensions and material gauge need not exactly match the ANSI standard. All the conclusions, limitations, and bounding spectra for motor control centers are applicable to low-voltage switchgear. (4) Metal-Clad Switchgear Metal-clad switchgear consists primarily of circuit breakers and associated relays (such as over-current relays or ground fault protection relays), inter-locks, and other devices to protect the equipment that it services. Metal-clad switchgear of 2.4 kV and 4.16 kV is considered. The general config-uration of cabinets must be similar to those specified in ANSI C37.20. This requirement is imposed to preclude unusual designs not covered in the data base. Cabinets that are configured similarly to those specified in the ANSI standards will perform well if they are properly anchored. Cabinct dimensions and material gauges need not exactly match ANSI standards. All the conclusions, limitations, and bounding spectra for motor control centers are applicable to metal-clad switchgear, except that the cutouts in the cabinet sheathing shall be less than about 12 inches by 12 inches. (5) Motor-Operated Valves Motor-operated valves consist of an electric motor and gear box cantilevered from the valve body by a yoke and interconnected by a drive shaft. The motor and gear box serve as an actuator to operate the valve. On the basis of a review of the data base and anticipated variations in condi-tions, it appears that motor-operated valves are sufficiently rugged to survive i l NUREG-1211 12 Appendix A Enclosure i .r_._..,-_,, ~~

a seismic event and remain operational thereafter provided the following condi-tions exist in the nuclear facility: (a) The spectra for the nuclear facility are less t.han the appropriate bounding spectrum described in Figure A.1 for frequencies above the estimated funda-mental frequency of the piping-valve system. (b) The valve is located less than 40 feet above exterior grade. If the valve is located higher than 40 feet above exterior grade, the floor spectra shall be compared with 1.5 times the appropriate bounding spectrem. (c) The valve body and yoke construction is not of cast iron. (d) The valve is mounted on a pipe at least 2 inches in diameter. (e) The actuator is supported by the pipe and not independently braced to or supported by the structure unless the pipe is also braced immediately adjacent to the valve to a common structure. The following limitations on operator weight and eccentric length relative to pipe diameter are derived from the data base for motor-operated valves that was provided by SQUG.* (a) A type A bounding spectrum shall be used for the following cases: (see Figure A.2): Valves mounted on 12-inch diameter or larger pipes with a 60-inch or smaller distance from the pipe centerline to the top of the motor actuator, and the approximate actuator weight is less than 400 pounds. Valves mounted on 24-inch diameter or larger pipes with a 100-inch or smaller distance from the pipe centerline to the top of the motor actuator, and the approximate actuator weight is less than 300 pounds. (b) A type C bounding spectrum shall be used for the following cases: (see Figure A.3): Valves mounted on a pipe diameter of at least 2 inches but less than 6 inches, with a 30-inch or smaller distance from the pipe centerline to the top of the motor actuator, and the approximate actuator weight is less than 100 pounds. Valves mounted on a pipe diameter of at least 6 inches but less than 8 inches, with a 40-inch or smaller distance from the pipe centerline to the top of the motor actuator, and the approximate actuator weight is less than 300 pounds.

• The data base contains relatively few heavy operators and small pipe diameters subjected to severe ground shaking.

These limitations could be less restrictive if more motor-operated valves had been located and documented in the areas of i higher shaking. Additional data, either from other earthquake experience or seismic qualification tests, could expand the scope of these recommendations. NUREG-1211 13 Appendix A Enclosure 1 l

i 1 I O OUTSIDE EXPERIENCE DATA ,.B C WITHIN EXPERIENCE DATA iE Ew " 100 $kh6hh$ s p ^ ~ ' 3:g(f,s s_ in: s s ,t: 4)!:l: s gyn,, litt ~ ' :#:Y J. ' - 'g< - s,j:av s < s

• ;!.'

  ;;:5 go

a

ua

!'4 0'b', ( 0 1 ~;' !!' #*i 'h 0 y

• ii 0**p

'-'9

'

. bit?

s ^ ,,7.. 'y' ', !!!.l;.;.s :it:i;if:')g:::;5 ' ,.::: p'.

n.ji.,

f

.f ' ' ^ -':.:!.ig' y ;.; '...,, q.is:!;:;.~ '^' .:. :..,I ' [hi',[ s v.

4., I.31.: ?,'*$:j :. ym ;ig gg, 7

gg,, z?]!e j., e ,s 5J .:;l e. [.... s

.;: :::. v s '

s ' v b: pr:: ?li

u...v.

3

,I;, ii?ii.

x..,'.. .s.. s ?!% ' c,:? ', .t -

!!l i,

::58 :i '<, Y:Y.:i i?<sy Y1:i>

Spic'v'" 3

1:

s ..l ~ 3i:#

Q

?:l lOE @ f

a s*

D 12 24 PIPE DIAMETER (inches) i

• APPRoxNATE MAxNuu OPERATOR wtl0HT f

Figure A.2 Motor-operated valves for which type A spectrum is to be used I I NUREG-1211 14 Appendix A Enclosure ( . =,

I 3 4 I I 4 amn OUTSIDE EXPERIENCE DATA C WITHIN EXPERIENCE DATA Z> W me 100 s s 's

ifiis s ii!%g < n;.y:<: ' j.7(.?

;:ic

.s .s..~...s. .. g 4

s@:;,,..y5;:'s d 300**p..;

,s

3.,

s "ir gs:" .-s -s c "- !ssi:) ~ s ,Ikk :;i;;l;;kq$i J.![hi' hkhs;i.s.!r):S ii "....t is:iss sidi!!!

r....'..'.....,...........<..$s#it.. s.. iolie..J4 :i'. < '

4 O 70 ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ - ~ ~ ~ ~ ' ' ' . i.?;?$ s :t;@ b: f D' Ijijfi$ jy!!ji.ii '.g'..:..,. D * [ f s

.r e.-

i:Y:

s fe w ^ b %;i;;.5g j8?if:

i!EI;;F,

,..p:;I5:!...,.

.:3;

':"v s- ' ""f::8!8i:' Q p 400**

d..:.:.

c.:;::

s < 4: '0W ~, tie-50 tim!. -

i.:$i$;i!.hif.

'"C

( s f:

g 300** 3!$ia:$, YS#'""Nis , pl. '^ sf:.850f.,.iiig s 40 ~~ k J:x4 'iji:ijj~.;: ' <4 e s f.:!!isRs' ?i!!iT%i)!!!!!! ?.,. :::p:in ' /i:- iisis ' esit:", w 100#* ' s,!!. ' s' iilisi)! - " ::i.;;.9 :: : g:: 30 iii !

o.

1. Uc5-(

4$

@i s s s ,i...:::.i+g; ,s s s. s g s '.s ss s

m s

s es; is!.i J.%gi;. a s

.s...

e s

m::

s t 's. '..:

v+ :,.;

f .l,".. d s s s s ~ 'Ay y..iira.:: cz 4 2 4 8 4 10 12 PIPE DIAMETER (inches) 'APPROXWATE MAXIMUM OPERATOR WE10HT Figure A.3 Motor-operated valves for which type C spectrum is to be used NUREG-1211 15 Appendix A Enclosure I I

Valves mounted on a pipe diameter of at least 8 inches but less than 10 inches, with a 50-inch or smaller distance from the pipe centerline to the top of the motor actuator, and the approximate actuator weight is less than 400 pounds. Valves mounted on a pipe diameter of at least 10 inches with a 70-inch or smaller distance from the centerline of the pipe-to the top of the motor actuator, and the approximate actuator weight is less than 640 pounds; or the weight is more than 300 pounds for cases where the distance from the centerline of the pipe to the top of the motor actuator is not greater than 100 inches. For motor-operated valves not complying with the above limitations, the seismic ruggedness for ground motion not exceeding the type A bounding spectrum may be demonstrated by static tests. In these tests, a static force equal to three times the approximate operator weight shall be applied non-concurrently in each of the three orthogonal principal axes of the yoke. Such tests should include a demonstration of operability following the application of the static load. The limitations other than those related to the operator weight and distance from the top of the operator to the centerline of the pipe, given above shall remain in effect. (6) Unit Substation Transformers Unit substation transformers convert the distribution voltage to low voltage. In this discussion, unit substation transformers that convert 2.4-kV or 4.16-kV distribution voltages to 480 V are considered. On the basis of a review of the data base and anticipated variations, it appears that unit substation transformers are sufficiently rugged to survive a seismic event and remain operational thereafter provided the following conditions exist in the nuclear facility: (a) The spectrum for the nuclear facility is less than the type B bounding spectrum described in Figure A.1 for frequencies above the estimated funda-mental frequency of this equipment, and the unit substation transformer is located less than 40 feet above exterior grade. If the unit substation transformer is located higher than 40 feet above exterior grade, the floor spectrum shall be compared with 1.5 times the bounding spectrum. In all cases a comparison with floor spectra is also acceptable. (b) Both unit substation transformer enclosures and the transformer itself must have engineered anchorage. The functiohal capability of properly anchored unit substation transformers during and after earthquakes, as defined above, has been demonstrated. (7) Air-Operated Valves Air-operated valves consist of a valve (controlled by a solenoid valve) operated by a rod actuated by air pressure against a diaphragm attached to the rod. The actuator is supported by the valve body through a cantilevered yoke. NUREG-1211 16 Appendix A Enclosure ) k

On the basis of a review of the data base and anticipated variations in condi-tions, it appears that air-operated valves are sufficiently rugged to survive a seismic. event and remain operational thereafter provided the following conditions exist in the nuclear facility: (a) The ground motion spectra for the nuclear facility are less than the type A bounding spectrum for frequencies above the estimated fundamental frequency of the piping-valve system. (b) The valve body is not of cast iron. (c) The valve is mounted on a pipe of 1-inch diameter or greater. (d) -If the valve is mounted on a pipe less than 4 inches in diameter, the dis-e tance from the centerline of the pipe to the top of the operator shall not exceed 45 inches. If the valve is mounted on a pipe 4 inches in diameter o.m larger, the distance from the centerline of the pipe to the top of the operator shall not exceed 60 inches (see Figure A.4). (e) The actuator and yoke are supported by the pipe, and neither is indepen-dently braced to the structure or supported by the structure unless the pipe is also braced immediately adjacent to the valve to a common structure. The air supply line is not included in this assessment. For air-operated valves not complying with the above limitations, the seismic ruggedness for ground motion not exceeding the type A bounding spectrum may be demonstrated by static tests. In these tests, a static force equal to three times the approximate operator weight shall be applied non-concurrently in each of the three orthogonal principal axes of the yoke. Such tests should include demonstration of operability following the application of the static load. The limitations other than those related to the distance of the top of the operator to the centerline of the pipe given above shall remain in effect. (8) Horizontal and Vertical Pumps Horizontal pumps in their entirety and vertical pumps above their flange are relatively stiff and very rugged devices as a result of their inherent design and operating requirements. Motors for these pumps are also included. Subject to the Ifmitations set forth below, all pumps meet the criteria for the type A bounding spectrum. For horizontal pumps, the driver (electric motor, turbine, etc.) and pump must be rigidly connected through their bases to prevent damaging relative motion. Of concern are intermediate flexible bases, which must be evaluated separately. Thrust restraint of the shaft must also be ensured in both axial directions. The data base covers pumps up to 2500 hp; however, the conclusions appear to be equally valid for horizontal pumps of greater horsepower. For vertical pumps, the data base has many entries up to 700 hp and several up to 6000 hp. However, vertical pumps, above the flange, of any size at nuclear plants appear to be sufficiently rugged to meet the type A bounding spectrum. NUREG-1211 17 Appendix A Enclosure l y

T s 'h + 1 l l 8 i O outsiDe exPaRisNes oATA C C WITHIN EXPERIENCE DATA 5 .J B> 4 60 s s Nd-A$il,'.f!M: *!!if. :*AD D:Y 1:4;;;: c;.s s+ y : :.> , - N. :.,, -*...U. i.;

r

.

v.

.r.,.

2.y'.>,..... 4 y@d gi:;:["I 0J;^. :::?s f . s. 'i.i@S 'f :i.b.g-r.J.:2 %...; ' < i '; ::'ii$ ::! 'YS s. s " " sa : .!;,,;;.:. s

.s.:.:. - <

s.y :c,, - :.. iff0.:i:= i '~ ' I & 4& h {.y:ilQ' $j:!!:.; '#3' I:iA*i,' ', j .i: Q s s a... :np..ii.i:?.. l

in

..1 .S.i:v=.. ,.,3 e

,g

j..;,

• (.,

'2:z'Q;'

st:?.: :.::.

4

. /

s v. .9y,..;. : :y;. ss 'Ds,<i'i 5!:fsy......MM;Ihiii'.:,::ii. ',': l 'y,..

Is

...[.'f ;/ s {:yidri 4':jiy 5 e .p::i:i',. ;.,,, e Vv ' :;7;

i. ' 5.3 'j;;: e f:;ii-'

+

s ':::0!. ' */.5 ;. s s-s ~,'...... . de ^ ^?C : /. Y.;n.v.:: 4.4 .JN:. < e :::*:g: is :.. s

.

$. ii,:

.m ,.s e <.4 pv s.,, : i.

e...p.5:

s-s 2 v .s ).3M. fi-9 9 v.x. : ..,.s,..,v .: s.v<: ' < s

4.-x

g

.y:'.:. -

v. g;+x ;;fgj y.:v;:. e', s

6pss,.s < :::.g::s r.v v.v -

.- p s...

fj

.,7;, 7v '..: ; - + .:s.:y:.y: f2 e .i4:..: - - o 2.J...i' " ';n:y + s.,,., Ag? f 5 ',. :.f.,;.y.,. ;;$.:6 .. :b ki o.:.g g,,.v_' N's

• i offM..:::. -:.:jf.:f

,.:..,,, _,,.4 5 ^. :.: s

j...cif.Y p:<. e.

.!.;;. q:, N:;;, s ss ., : sp :.- v s< ,.v... X ;;.;,;, g ' ?'p; s. -;s' O 1 4 PIPE DIAMETER (inches) I Figure A.4 Air-operated valves for which type A spectrum is to be used j NUREG-1211 18 Appendix A Enclosure 1 h I \\

The variety of vertical pump configurations and shaft lengths, below the flange, and the relatively small number of data base points in several categories pre-einde the use of the data base to screen all vertical pumps. Vertical turbine pumps (deep well submerged pumps with cantilevered casings up to 20 feet in length and with bottom bearing support of the shaft to the casing) are well enough represented to meet the bounding criteria below the flange as well. Either individual analysis or use of another method should be considered as a means of evaluating other vertical pumps below the flange. The chief concerns would be damage to bearings as a result of excessive loads, damage to the in-peller as a result of excessive displacement, and damage as a result of inter-floor displacement on multi-floor supported pumps. 7. Guidance on Review of Equipment that Exists in the Experience Data Base Plants but that Is Not Included in the Eight Types in the Data Base On the basis of the above experience, reviews conducted by the staff in the SEP Program and licensing activities (SQRT. audits), and the observation of the behavior of equipment beyond the original eight classes found in the data base plants, the staff concludes that the seismic adequacy of equipment other than the eight types can be achieved by (1) anchorage verification; (2) a careful I review of caveats, outliers, and exclusions observed; and (3) documentation by SQUG of the basis for seismic adequacy of each equipment type. The SQUG is in the process of broadening the data base to include more recent earthquake experience (notably the 1985 earthquakes of Chile and Mexico). The equipment covered by the experience data base will be expanded from the original eight to twenty which will encompass all equipment needed for plant hot shutdown. The SSRAP report is also being revised accordingly. The guidance in the final revised SSRAP report may differ from that mentioned in the January 1985 SSRAP report. The revis~ed SSRAP report should be followed for implementation guidance. For individual utilities not participating in the generic group, the detailed procedures used to review the seismic adequacy of all equipment should be sub-mitted to the NRC for review. Items such as equipment caveats and exclusions, bounding spectra to be used, and the like should be included in tl.a submittal. I NUREG-1211 19 Appendix A Enclosure L ___.,.h__.___-____

geoa. n. us.Nucuan uvuvoa,co iuioN i e + 1Novee -,~, =c. - ~ a. - %"3# BCuo:RAPHIC CATA SHEET NUREG-1211 Sit IN$taVCTsONS O TMt atytest 2 TsTLt.ND 5USfli 3 LE.vE O L.N. Regulatory nalysis for Resolution of Unresolved Safety Iss A-46, Seismic Qualification of Equipment in Operatin lants I'""""'""... j 1. . Lur oais> Febr ry 1987 T. Y. Chang I

• ^" " """',,,,

l 1987 N. R. Anderson F ruary r e...oa.,~o oas.N,2.i.oN N ..No..,L,No.wa.s.u...,<,c ,aoacra.s.,.oa.uNin uv ea Division of Safet Review and Oversight 2 ,,,,o,,,,,,,,,,,,,, Office of Nuclear actor Regulation U. S. Nuclear Regul ory Commission Washington, D.C. 2 55 10 SPoNSoa:NG OaQ.N12.TsoN N.WE.ND SLING ADDaES$ trac 4 sele coses 11a T

• PE oF aEPossi Technical Report Same as 7

. aea'a cov eaeD u-~~ ~~ 12 SUPPLEwtNT.RY NOTES 13 LESTR.CT (200 worse er 'oul The margin of safety provided in existi nuclear power plant equipment to resist seismically induced loads and erform quired safety functions may vary considerably, because of significant changes 'n des' n criteria and methods for the seismic qualification of equipment over he ars. Therefore, the seismic qualification of equipment in operating plants mu t reassessed to determine whether requalification is necessary. The objective of USI A-46 is to s blish an explicit set of guidelines and acceptance criteria to judge the seismic equ y of equipment at all operating plants, in lieu of requiring these plants to mee the c.iteria that are applied to new plants. This report presents the r gulatory a alysis for Unresolved Safety Issue (USI) A-46. It includes (1) Statement f the Probl , (2) the Objective of USI A-46, (3) a Sum ary of A-46 Tasks, (4) a Pro sed Implement ion Procedure, (5) a Value-Impact Analysis, (6) Implementation,(7) Summary of A-4 Risk Analyses and (8) Operating Plants To Be Reviewed to USI A-46 R uirements. g

i...,. = T.

1.. oocu. Nr *~.Lvi,. -.. v. aso sca..Toas Unresolved Safe Issue A-46, Regulatory Anat sis Unlimited 16 SECua:TY CL.55. F sc.T ON I The pepel . ocNr,,,easesN. .o na s Unci m ifind a r. .m Unclassified 17 NvM8L. 06 P A GE 5 18 Pa>(E

ZC UNITE] STATES , m 7, costanoirees raio 3 NUCLEAR RE ULATORY COMMISSION WASHINGT,0N, D.C. 206Ei6 c, b w a N OFFICIAL BUSINESS ~ aa PENALTY FOR PRIVATE USE. 6300 120555078877 1 1AN US NRC ADM-DIV 0F PUB SVCS POLICY & PUB MGT BR-POR NUREG W-501 g WASHINGTON DC 20555 m 4 OC r.> en -4 \\ mO 5lD E< 5 >Z Oy C>r< C 18.! m en b1 >0 -4 :D O3 Zm

• Ob nc mC O -!

5O mZ E O Z' -1 C _Z Z3 0* 'b m r 3< $mU Zc Q >o o 'm r d >4 2 _4 - M$ C i m>b? t L m t CD l

=

l C l a ~ _ _.... _ _}}