ML20155J948

From kanterella
Jump to navigation Jump to search
TMI Unit 1 Pra,Data Analysis Rept, Vol 5
ML20155J948
Person / Time
Site: Three Mile Island Constellation icon.png
Issue date: 11/30/1987
From: Garrick B, Hubbard F, Iden D
PLG, INC. (FORMERLY PICKARD, LOWE & GARRICK, INC.)
To:
Shared Package
ML20155J777 List:
References
PLG-0525, PLG-0525-V05, PLG-525, PLG-525-V5, NUDOCS 8806210076
Download: ML20155J948 (348)


Text

_ _ _ _ - _ _

Copy Copyright C WW' *

.; PLG-0525 GPU Nuclearr. , in'.

Volume 5 Three Mile Island Unit 1 Probabilistic Risk Assessment DATA ANALYSIS REPORT Project Director B. John Garrick Project Manager Douglas C. Iden Principal Investigator Frank R. Hubbard Task Leaders Mardyros Kazarians Ali Mosleh Harold F. Perla Martin B. Sattison Donald J. Wakefield Prepared for GPU NUCLEAR CORPORATION Parsippany, New Jersey November 1987 ggg62;gggggggg;ge9 P

DCD O Pickarc ,Lowe and Garrick,Inc.

Engineers e Applied Scientists e Management Consultants Newport Beach, CA Washington, DC

O NOTICE This is a report of work conducted by individual (s) and contractors for use by GPU Nuclear Corporation. Neither GPU Nuclear Corporation nor the authors of the report warrant that the report is complete or accurate. Nothing contained in the report establishes company policy or constitutes a commitment by GPU Nuclear Corporation.

O

SUMMARY

OF CONTENTS EXECUTIVE

SUMMARY

REPORT Volume 1 l Acknowledgment j Foreword ,

TFCHNICAL

SUMMARY

REPORT Volume 2 PLANT MODEL REPORT Volume 3 SYSTEMS AMLYSIS REPORT Volume 4 DATA ANALYSIS REPORT Volume 5 HUMAN ACTIONS ANALYSIS REPORT Volume 6 i

I ENVIRONMENTAL AND EXTERNAL HAZARDS REPORT Volume 7 O

l i

.I i

o 0540G123186

1 i

DATA ANALYSIS REPORT CONTENTS Section- Page LIST OF TABLES AND FIGURES y LIST OF ACRONYMS vii 1 INTRODUCTION 1-1 2 DATA ANALYSIS APPROACH 2-1 2.1 Component Failure Rates 2-2 2.1.1 Generic Failure Rate Distributions 2-2 C

2.1.1.1 Generic Distributions Based on Actual Performance Records (Type 1) 2-5 2.1.1.2 Generic Distributions Using Estimates of Available Sources of Generic Data (Type 2) 2-9 2.1.1.3 Generic Distributions Based On a Mixture of Type 1 and Type 2 Data 2-19 2.1.1.4 Development of Generic Failure Rate Distributions 2-19 2.1.2 Data Specialization 2-22 2.2 Common Cause Failure Parameters 2-23 0 2.2.1 Overview of the MGL Method 2.2.2 Estimators for the Parameters of the MGL Model 2-23 2-25 2.2.3 Assessment of Uncertainty 2-27 2.2.3.1 Assessment of Uncertainty Due to Data Sample Size 2-27 2.2.3.2 Assessment of Uncertainty Due to Data Classification 2-30 2.2.3.3 Plant-to-Plant Variability of the MGL Parameters 2-33 2.2.4 Generic Common Cause Parameter Data Base 2-34 2.3 Component Maintenance Data 2-34 2.3.1 Frequency of Maintenance 2-35 2.3.2 Duration of Maintenance 2-35 2.4 Initiating Events Frequencies 2-37 2.5 References 2-37 3 TMI PLANT-SPECIFIC DATA BASE 3-1 3.1 Introduction 3-1 3.2 Component Failure Rates 3-1 3.2,1 Component Failure Data 3-1 3.2.1.1 Definition of Failure 3-1 3.2.1.2 Failure Data Sources 3-3 3.2.2 Component Demands and Operating Hours 3-4 3.2.2.1 Demand Data Sources 3-4 3.2.2.2 Operating Hours Data Sources 3-6 O 3.2.3 Updated Component Failure Rate Distribution 3.3 Common Cause Failure Parameters 3,7 3-7 iii 0421G12318fDAR

_s ______

CONTENTS (continued) g Section Page 3.4 Component Maintenance Datt 3-7 3.4.1 Definition of Maincenance < 3-7 3.4.2 Maintenance Data 'Jources 3-8 3.4.3 Updated Componer.c Maintenance Distribution 3-9 3.4.3.1 Updatad Maintenance Frequency Distributions 3-9 3.4.3.2 Updated Maintenance Mean Duration Distributions 3-9 3.5 Initiating Event Group Frequencies 3-9 3.5.1 Events Quantified, Based on Generic and Plant-S,*aci fic Data 3-10 3.5.2 Initiating Events Whose Frequencies Were Quantified by Plant-Specific Analysis 3-12 3.5.2.1 Loss of River Wcter 3-12 3.5.2.2 Loss of Nuclear Services Closed Cooling Water 3-13 3.5.2.3 Loss of Control Building Ventilation 3-13 3.5.3.4 Inadvertent Opening of DHR Valves (V-Sequence) 3-13 3.6 References 3-18 APPENDIX A: THI-1 COMPONENT FAILURE RATE DATA A-1 A.1 Component Failure Data Summary Sheets A.1-1 A.2 Component Success Data Summary Sheets A.2-1 APPENDIX B: TMI-1 COMP 0NENT COMMON CAUSE FAILURE DATA

SUMMARY

SHEETS 8-1 APPENDIX C: TMI-1 COMP 0NENT MAINTENANCE DATA

SUMMARY

SHEETS C-1 APPENDIX D: TMI-1 INITIATING EVENTS DATA

SUMMARY

SHEETS D-1 O

iv 0421G123186DAR

l l

l LIST OF TABLES AND FIGURES i

)

Tabl e Page 2-1 Example of th'e Calculation of the Effective ' Number of Failures for Various System Impact Categories 2-40 3-1 THI-1 Records Considered in Data Search 3-19 3-2 THI-1 Cold Shutdown Outages - 3-20 3-3 Test Procedures Considered for Success Data Development 3 21 3-4 TMI-1 Component Failure Rate Data 3-28 3-5 TMI-1 Component Common Cause Parameter Data Base 3-31 3-6 TMI-1 Component Maintenance Frequency 3-34 3 TMl-1 Mean Maintenance Duraticn Data Base 3-36 3-8 THI-1 Initiating Event Frequency Data Base 3-38 3-9 Grouping of EPRI Event Categories into TMI-1 Initiating Event Groups 3-39 Figure

'2-1 Populati or. Variability of the Failure Rate .

2-41 2-2 State-of-Knowledge Dist. .bution Over the Set of Frequency Distributions 2-41 2-3 Posterior Distribution for the Parameters-of the Distribution of Pumps' Failure to Start on Demand Rates. 2-42 q 2-4 The Relation Between the Population Variability Curve dnd Uncertainty About Individual Estimates 2-42 U 3-1 Example System for Flow Test Data 3-42 3-2 Cold Leg Injection Path and Hot Leg Suction Path Simplified Diagram 3-43 D

i l

{

g  !

l l

v l

-0421G123186DAR

~.- . - _ - . - - - . . , - _ - . . .

- - - _ - - . . _ m __ - - _ _ ~ _ -

____.4 - - _ -- - --_----- -----w 9

0 O

vi 0421G123186DAR

b LIST OF ACRONYMS pbreviation Definition ACR air-cooled reactor ADV atmospheric dump valve.

A0V air-operated valvo-AT0G abnormal transient operational guidelines ATWS . anticipated transient without scram B0P balance of plant Btu British thermal unit BWR boiling water reactor BWST borated water storage tank ~ <

CARS condenser air removal system {

CAS chemical addition system  ;

CBVS control building ventilation system -

CCF common cause failure  !

CFT core flooding tank CIV containment isolation valve CSF conditional split fraction CST condensate storage tank CR0 control room operator CWS circulating water system

(] -

DHCCW DHR decay heat closed cooling water decay heat removal  :

DHRS decay heat removal system DHRW decay heat river water EFW emergency feedwater E0F emergency operations facility EPRI Electric Power Research Institute ESD event sequence diagram ESAS engineered safeguards actuation system ETC event tree code FSAR Final Safety Analysis Report FTAP Fault Tree Analysis Program GCR gas-cooled reactor GPUN GPU Nuclear Corporation '

HCR human cognitive reliability HPI high pressure injection HPIS high pressure injection system HVAC heating, ventilating, and air conditioning ICCS intermediate closed cooling system '

ICCW intermediate closed cooling water q ICS integrated control system O

n 0517G123186

~ - _

LIST OF ACRONYMS (continued)

Abbreviation Definition O

LBIS line break isolation system LCO limiting condition for operation LER Licensee Event Report LOCA loss of coolant accident LOFW loss of main feedwater LONS loss of nuclear services LORI loss of reactor coolant system inventory LORW loss of river water LOSP loss of offsite power LPI low pressure injection LPIS low pressure injection system LSS low speed stop MCC motor control center MFPT main feedwater pump trip MFW main feedwater MGL multiple Greek letter MOV motor-operated valve MSIV main steam isolation valve MSLB main steam line break MSS main steam system MSSV main steam safety valve '

MSV main steam valve MVP makeup and purification NPE Nuclear Power Experience NRC U.S. Nuclear Regulatory Commission NSCCS nuclear services closed cooling system NSCCW nuclear services closed cooling water NSRW nuclear services river water NSSS nuclear steam supply system OPM Operatinns Plant Manual OTSG once-through steam generator PCL panel center left PCR panel center right PDS plant damage state '

PLF panel left front PLG Pickard, lowe and Garrick, Inc. '

PalD piping and instrumentation drawing PORV power-operated relief valve PRA probabilistic risk assessment PRF panel right front primary to secondary heat transfer i

PSHX '

PSV pressurizer safety valve -

l PWR pressurized water reactor I el viii 0517G123186

_ . _ _ . . _ _ _ - . . . - _ . , . _ . _ . _ _ . . _ _ - . _ _ _ _ _ _ . . ~

t LIST OF ACRONYMS (continued) i f,I Abbreviation Definition t RBCU reactor building cooler unit  ;

RBEC reactor building emergency cooling- l

-RBD reliability block diagram  :

RBS reactor building spray l RBSS reactor building spray system i RCDT reactor coolant drain tank  :

RCP reactor coolant pump -

RCS reactor coolant system _[

RPS- reactor protection -system j t

SCCW secondary closed cooling water I

'SCM subcooled margin SGTR steam generator tube rupture j

SLB steam line break -

SLRDS steam line rupture detection system SR0 senior reactor. operator SRW secondary river water . ,

SSCCS secondary services. closed cooling system SSE safe shutdown earthquake SSS support state system 'i STA shift technical advisor  !

TBV turbine bypass valve  !

TMI-1 Three Mile Island Nuclear Generating Station, Unit 1  !

i ULD unit load demand  ;

i

? [

i l

l i

J O  !

0517G123186 ,

I

1. INTRODUCTION ym This report presents the data developed in support of the TMI-1 PRA systems and-plant analyses and provides a discussion of the techniques used and steps taken in developing the data base.

The following four general areas define the scope of the data analysis as presented in this report:

1. Component Failure Rates
2. Common Cause Failure Parameters
3. Component Maintenance Frequency and Duration
4. Initiating Event Frequencies Several other types of data such as component fragility curves used in the seismic analysis, fire frequencies used in the fire analysis, and human actions are developed and presented elsewhere in the THI-1 PRA l report.

The TMI-1 data are developed by combining in a Bayesian update the cumulative experience from a large population of nuclear power plants documented in the PLG proprietary data base with the comprehensive plant-specific data base developed from a detailed review of the TMI l' Unit 1 records of several years of operation.

, The proprietary PLG generic data base has evolved from all of the PRAs that PLG has performed to date. It is based on data collected from U.S.

reliability data sources and from operating data of U.S. light water reactors evaluated in past PLG PRAs.

The following sections describe in detail the methodology used for data analysis followed by a detailed discussion of the collected

)

plant-specific data. The resulting distributions are presented in I tabular form for each of the five categories of data supported by a  !

series of appendices at the end of the report that provide the detailed '

plant-specific data. All plant-specific distributions are stored in.a computer data base that includes a brief summary of the collected data, the generic distributions used, and several important characteristics of the distributions.

O 1-1 -

0153GU61886DAR

2. DATA ANALYSIS APPROACH D

This section provides a discussion of the techniques used in developing the TMI-1 data base. As mentioned earlier, the data was developed by updating generic information with TMI-specific information, using Bayesian techniques.

Familiarity with certain basic concepts of Bayesian analysis is essential in underst:inding the content of this section. These concepts are briefly reviewed in the following.

The methodoiogy used_ to develop the data for this study is based on the Bayesian in:erpretation of probability and the concept of "probability of frequency" (Reference 2-1). In this context, for example, component failure rates are treated as measurable quantities whose uncertainty is dependent on the state of knowledge of the investigation. The "state of knowledge" is presented in the form of a probability distribution over the range of possible values of that quantity. The probability associated with a particular numerical value of an uncertain but measurable quantity indicates the likelihood that the numerical value is the correct one. ~

l A key issue in developing state-of-knowledge distributions for the parameters of the PRA models is to assure that the information regarding each parameter, its relevance, and its value as viewed by the analyst are

~

presented correctly and that various pieces of information are integrated coherently. "Coherence" is preserved if the final outcome of the process is consistent with every piece of information used and with all assumptions made. This is done by utilizing the fundamental tocl of probabilistic inference; i.e., Bayes' theorem (Reference 2-2).

Mathematically, Bayes' theorem is written as P(xlE,EO ) = k -I L(Elx,E )P(xlE ) (2.1) 0 O where P(xlE,Eo) E probability of x being the true value of an unknown quantity in light of new evidence E and prior body of knowledge Eo.

L(Elx,E0) E likelihood of the new evidence E assuming that the true value is x.

P(xlEo) E probability of x being the true value of the unknown quantity based on the state of knowledge Eo prior to receiving E.

Finally, k is a normalizing factor defined as "3 kE L(Elx,E)P(xlE)dx O 0 (2.2)

(Q all x 2-1 Olb4G0619860AR

In the context of a plant-specific PRA, there are three types of information available for the frequency of elemental events.

Eu = general engineering knowledge such as that of the design and manufacture of equipment.

E l = the historical information from other plants similar to the one in question.

E2 = the past experience in the specific plant being studied.

The information of types E0 and El together constitute the "generic" information, and E2 is the "plant-specific" or "item-specific" information.

Since the THI-1 plant has several years of operating experience, the data developed for the TMI-1 PRA are based on gereric as well as plant-specific information. Any additional plant-specific information collected in the course of operating TMI units in the future can be incorporated into the existing data by applying Bayes' theorem.

It is very important to note that the information Eo brings an element

of plant specificity in the generic data developed for a plant-specific PRA. In general, decisions regarding the relevance and applicability of different pieces of information in develnping each generic distribution are made based on type Eo information. Therefore, a piece of information may be judged as being relevant in developing the generic data in one PHA and not in another. As a result, generic distributions for different plant-specific studies could be significantly different.

2.1 COMPONENT FAILURE RATES 2.1.1 GENERIC FAILURE RATE DISTRIBUTIONS To discuss the way the failure rate distributions were developed based on different types of information, we consider the following cases.

e Type 1. Failure data from operating experience at various nuclear power plants.

e Type 2. Failure rate estimates or distributions contained in various industry compendia, such as WASH-1400 (Reference 2-3) and IEEE-500 (Reference 2-4).

By type 1 information, we mean a set of failure and success data  !

collected from the performance of similar equipment in various power plants. Reference 2-5, for example, provides a detailed list of reported valve failures at various U.S. commercial nuclear power plants for a 2-year period. Also given in this reference are the number of demands and total operating time for the valves at each power plant.

Type 2 information, which could be called processed data, are estimates  !

ranging from the opinion of experts with engineering knowledge about the 2-2 0154G0618860AR '

design and manufacturing of the equipment to estimates based on observed Q performance of the same class of equipment in various applications. For V instance, Reference 2-4 provides failure estimates based on the opinion of several experts. Estimates of Reference 2-5, on the other hand, are based on recorded failures of equipment at various nuclear power plants.

Normally, type 2 data are either a point estimate usually referred to as l the "best estimate," or a range of values centered about a "best '

estimate." In some cases, a distribution is provided covering a range of values for the failure rate with tne mean or median representing the "best estimate" of the source. For instance, IEEE-500 provides a "low," -'

"high," and "recommended" for the failure rates under normal conditions and a "maximum" value under extreme environments. WASH-1400, on the other hand, assesses a probability distribution for each failure rate to '

represent the variability of the available data from source to source.

Such distributions are normally centered around a median value_ judged to  ;

be most representative of the equipment in question for nucle?.-

applications.

The methodology used to develop the TMI-1 failure rate data uses both types of information to generate generic probability distribution for the failure rates. Such distributions represent variability of the failure rates, from source to source (for type 2 information) and/or from plant to plant (for type 1 information). Obviously, as aopiied to TMI-1, these ,

distributions are in fact, prior state-of-knowledge curves for the failure rate of components. The following discussion helps to understand the distinction and serves as a prelude to the discussion of the methodology.

Suppose that we have 100 plants and that for each plant the exact value of the failure rate of a particular type of pump is known. Let Aj be ,

the failure rate of the pump at the ith plant. Suppose further tha the i Ai's can be grouped into a limited number of discj ete values, sag A(1, through AS, w*ith 20 of the Aj*'s being equal to A1 , 35 equal to A2 26 equal tc A3 ,15 equal to A;, and finally, 5 equal to A*5 The frequency distribution of the Aj's is then given by the histogram ,

shown in Figure 2-1.

This histogram represents the "population variability" of the Aj 's because it shows how the failure rate of the particular type of pumps under consideration varies from plant to plant. It is an exact and true representation of the variability of the failure rate at the 100 plants in the population without any uncertainty or ambiguity because the distribution is based on presumed perfectly known failure rates at each and every one of those plants. j Consider now, the case where only estimates and not the exact values of the failure rates are available for some but not all of the 100 plants in the population. With this state of knowledge, obviously we are not able to know the exact population variability distribution (Figure 2-1). The question is how one can use this more limited information to estimate the population variability curve and how close the estimate will be to the true distribution as given in Figure 2-1.

2-3 0154G051886DAR

To answer the question, first note that the desired distribution is a member of the set of all histograms. Because of our limited information, we are uncertain as to which member of that set is in fact the true distribution. This situation can be represented by a probability distribution over the set of all possible histograms expressing our state of knowledge about the nature of the true histogram.

For instance, if the entire space, H, of all possible histograms is composed of only n histograms; i.e., if HE {h y,h2 ,...,h l n

where hj represents the ith histogram, the evidence regarding the pump failure rates at different power plants can be used to assess a prot 5ility distribution over H as follows n

P(H) = {pl,P 2 * * 'PN} with E Pj=1 (2.3) i=1 where pj is the chance that hj is the true histogram.

Figure 2-2 depicts the situation where the variable A is considered to be continuous and the desired distribution is a density function.

For a perfect state of knowledge, we would be able to say which hj is the true distribution; consequently, the corresponding pi would be equal to 1 and all others equal to 0. However, based on the state of knowledge expressed by Equation (2.3), our estimate of the true histogram is n

Ii = E pg h j (2.4) i=1 which is called the "expected distribution." Another histogram of interest is one which is assigned the highest chance of being the true histogram. We call that the "most likely distribution," hm, and we have pm " *

  • IPi i=1,...,n} (2.b)

The problem of obtaining P, defined by Equation (2.3), is formulated in the Bayesian context as follows [see Equation (2.1)]

-I P(hlE)=k 4 L(Elhg )P0 (h4 ) (2.6) where P 0 (h) is the prior state of knowledge regarding the set H defined by Equation (2.3) and P(hj lE) is the posterior state of knowledge in light of the evidence E. The evidence is incorporated via the likelihood term L(Elhj) which is the probacility of observing the evidence given O

2-4 Ulb4GU61886DAR

that the true histogram is h. Finally, k is a normalizing factor defined >

as [see Equation (2.2)]

n k = E L(Elh4) P0(hj) '(2.7) i =1 The expected distribution, Equation (2.4), is our estimate of the true population variability of the failure rate. It shows how the failure .

rates of similar pumps are distributed among plants in the population.  :

Now if all we know about a specific pump before we have any experience '

with it is that it is one member of the population, the population variability curve also becomes our state-of-knowledge distribution for '

the failure rate of that specific pump. In other words, generic distributions representing the population variability can also be used to predict the expected behavior of any member of the population if no other information is available.

For,this reason, the generic frequency distributions developed based on type 1 and type 2 information are used as the state-of-knowledge '

distributions for the components at the IM1-1 plant prior to incorporating the site-specific information.

The following sections describe how types 1 and 2 information can be used to develop generic distribution. i 2.1.1.1 Generic Distributions Based on Actual Performance Reccrds (Type 1)

The following discussion is based on the method presented in Reference 2-6. Consider the case where the following set of information is available about the performance of a generic component in N plants 1 1={<k,T>;i=1,...,N}

g j .

(2.8) where kj is the number of failures of the component in the ith plant in a specific period of time, Tj.

The desired information is, 4(A), the distribution of the failure rate .

of the component, A, in light of evidence 1. 1 This distribution represents the variation of A from one plant to another, and is analogous to Figure 2-1.

Following our discussion in Section 2.1.1, we would like to express a posterior state of knowledge about the true nature of the function $(A).

To make matters practical, it is assumed that 4(A) belongs to a particular parametric family of distributions. Let e be the set of m parameters of Q(A) ,

e = {e 1, m . ,e,} l l

O 1 2-5 0154G062186DAR I

For each value of 0, there exists a distribution 4(Al0) and vice versa. Therefore, the state-of-knowledge distribution over the space of all possible 4(Al0)s is the state of knowledge over all possible values of 6 and vice versa.

Bayes' theorem in this case is written as (see Equation (2.6)]

P(0lIl)=k gy -1L (11 l0,IO) P0 (0lIO) (2.9) where P(0l10II ) " Posterior state of knowledge about 0 in light of evidence I and prior information i g.

t L(11 l0,I O) = the likelihood of evidence 1 1 given that the actual set of parameters of 4(A) is 0.

>P(0lI) 0 O

= prior state of knowledge about 0 based on general engineering knowledge 10 '

and k is a normalizing factor

-1 k =

[L(1l0,I)P(0lI)d0 1 O g O 0

The likelihood term is the (conditional) probability of observing the evidence, I I, given that the data are based on an underlying population variability curve 4(Al0) with 0 as the value of its parameters L = P(<k) , T) > ; i = 1, . . . , N l 0, OI ) (2.10)

Note that L is also conditional on the prior state of knowledge 1. 0 If we assume that the length of operating hours, Tj's, at different plants are independent of one another and that the observed failures, kj 's, also have no dependence (according to our model, each kj is based on a dif ferent underlying failure rate) the joint probability distribution given by Equation (2.10) can be reduced to the product of the marginal distributions as follows N

L(l yl0,I O) = H Pj (k ,Tj le,IO) g (2.11) i=1 where Pj(kj,Tj l0,lo) E probability of observing kj failures of the equipment in question during the period Tj in the ith plant assuming that the set of parameters of the underlying population variability curve is 0.

c>

2-6 0154G0621860AR

If the f ailure rate, Aj, at the ith plant is known exactly, using a.

Poisson model, the likelihood of observing kj in Tj can be calculated j Q(~N from ,

P(k j ,T j % ) = ( I I}k g

    • P I-Aij T )- (2.12) i l

However, Aj is not known. All we know is that Aj is one of possibly many values of variable A which represents the variation of the failure rate from plant to plant. In addition, according to our model, A is distributed according to 4(Al0) with 0 being unknown. For this .

reason, we calculate the probability of observing the evidence,-

<kj, Tj>, by allowing the failure rate to assume all possible values.

This is achieved through averaging Equation (2.12) over the distribution of A i

=

i Pj (kj , Tj je,I )O = Pj (kj ,T jl A) 4(Aje)dA 0

m k ,

=

(ATj ) g i4(Ale)dA (2.13)  !

e U k!j t O '

V Depending on the parametric family chosen to represent 4(Al0), the ,

integration in Equation (2.13) can be carried out analytically or by numerical techniques. For example, if 4(Al0) is assumed to be a garmia i distribution which has the following form

,g) ,r(a)- a-1 e-SA B" (2.14) with a and S, both nonnegative, as its parameters, the integral can be done analytically resulting in (Reference 2-5) kg

. T 4

a f(a+kj )

Pj (kj ,T jla,B ) = (2.15) k!j r a) (B+T )a + kj In developing failure rate distributions, 4(Al0) is assumed to be ,

lognormal with p and o as its parameters 4 (x p ,g ) , 1 exp -

1 En A-En u f (2.16) 2 \ l

,F5 oA (

0

/)

IP '4, Equation (2.13) is calculated numerically.

l h

2-7 0164GU618860AR  ;

. . . . , , . . . , - , N

The total likelihood for all N plants can now be found by using Equation (2.13) in Equation (2.11)

N i, =

(AT3 )ki 'I L(l y l0,1 ) = []

0 dA$(Al0) ki exp(-AT)h 4

(2.17) 1=1 g i J The posterior distribution resulting from using the likelihood of Equation (2.17) in Bayes' theorem, Equation (2.9), is a probability distribution over the m-dimensional space of 0. Any point, O, in this spacehasaone-to-onecorrespondencewithadistribution,4(Al0),

in the space of $(Al0). Figure 2-3 is an example of P(0 lo,I I) constructed for 0 = {a,S), the two parameters of gamma d'stribution based on the pump data from all U.S. nuclear power plants (Reference 2-7).

The "expected distribution" is obtained from [see Equation (2.4)]

$(A) = $(Al0)P(0lIO 'll)de (2.18)

The quantity $(A) "summarizes" the information about A and is used in this study as the model for generic failure distributions.

Sometimes it is also useful to obtain the "most likely distribution" [see Equation (2.4)]. According to the definition., the most probable distribution of A is the one whose paramcters maxf.nize P(0lIgII ). These parameters are, therefore, the solution of the following system of m equations aP(0lI01 I) 3g e i, max = 0; i = 1, . . . , m (2.19)

The methodology discussed above also applies to failure on demand type of data where the evidence is of the form I t = { < k 4,0 >$ , i =1, . . . , N) (2.20) where k; ani Dj are the number of failures and demands in the ith plant, respectively. This can be done if the Poisson distribution used in Equation (2.13) is replaced by the binominal distribution 01 4

k D P(Kj ,04 lA ) = kj !(D$ -kg )!

A ( )$-k p.21) 9 2-8 0154G062386DAR

Example '

d For motor-operated valve failure to start on demand, the following data from six plants were available.

Number of Number of Plant Failures (k) Demands (D) 1 10 1.65 x 10+4 '

2 14 1.13 x 10+4 3 7 1,73 x 10+3 4 42 6.72 x 10+3 5 3 1.26 x 10+3 6 31 9.72 x 10+3

  • I These data, which form a set of type 1 information,1,1 were used in mode 1 of the computer code BEST4 (Reference 2-8), which calculates Equations (2.13) and (2.17) and generates 0(A) based on ,;

Equation (2.18). The result was a 20-bin discrete probability distribution with the following characteristics: ,

i a

5th Percentile: 6.10 x 10-4 50th Percentile: 1.05 x 10-3 95th Percentile: 3.19 x 10-3 t 0 Mean: 2.26 x 10-3 2.1.1.2 Generic Distributions Using Estimates of Available Sources of Generic Data (Type 2)

As mentioned earlier, generic data frequently are not in the fundamental '

form given by Equations (2.7) and (2.20). Rather, most sources report point or interval estimates or even distributions for failure rates l (type 2 information). These estimates are either judgmental (expert  ;

opinion), or based on standard estimation techniques used by the analysts to translate raw data into point or interval estimates, and sometimes into a full distribution. '

An example of such estimation techniques is the well known maximum likelihood estimator given by r

A g=1 (2.22)

T where k is the total number of failures in T units of operating time. ,

Most data sources report AM and not k and T.

To develop a model for constructing generic distributions using this type of data, the following cases are considered. l O

2-9 0154G121786DAR 1

2.1.1.2.1 Estimating an Unknown Quantity Having a Single True Value The following method is adopted from Reference 2-9. Suppose there are M sources, each providing its own estimate of A, which has a single true, but unknown valuet A . An example is the failure rate of a particular component at a given plant. The true value of that failure rate, A o t will be known at the end of the life of the component. Before then, however, the failure rate may be estimated by one or more experts familiar with the performance of the component. Let I j = {A y ;i =1, . . . ,M } (2.23) be the set of such estimates where Ai is the estimate of the ith expert for A t=

The objective is to use information I$ and obtain a state-of-knowledge distribution for A . tObviously, when everything is known about A t; such a state-of-knowledge distribution is a delta function centered at At P(Al Perfect Knowledge) = 6(X-A )

t

( .24 _

Note that in Equation (2.24), A is used as a variable representing the

~

unknown failure rate.

Assuming a prior state of knowledge, Po(A), about the quantity A, Bayes' theorem can be utilized to incorporate information 12 into the prior and obtain an "updated" state of knowledge about A

-I P(Al Aj,...,Ag) = k L(Ay,...,Agl A)Pg D) (2.25)

For N indegendent sources of information the likelihood term, L(Ai, ...,A N l A ) can be written as N

L(Ay,....AglA) = [] P (AylA) j (2.26) i=1 -

where Pj (AjlA) = the probability that the estimate of the ith source is Ay, when the true value of the unknown quantity is A.

The case of dependent sources of information is discussed in Reference 2-9. Obviously, if the ith source is a perfect one, Pj(AilA) = 6(Ai-A) (2.27) which means the estimate, Ai, is the true value. The posterior,  ;

P(AlAi,...,A$), in this case will be entirely determined by the estimate of this source l

P(AlAi,...A$)=6(A-Ai) (2.28) l 2-10 0154GU618860AR '

In another extreme, when it is believed that the source is' totally

/7 unreliable, V Pj(A{lA) = C l (2.29) where C is a constant. This means that if the true value is A, the estimate of the ith source can be anything. Using a likelihood of this '

form in Equation (2.25), will show that the estimate of this source, as expected, has no effect on shaping the posterior state of knowledge.

The likelihood term in this approach is the most crucial element. It reflects the analysts' degree of confidence in the sources of information, their accuracy, and the degree of applicability of their estimates to the particular case of interest. '

As can be seen, the subjective nature of-evaluating and "weighting" of the evidence from different sources fits very well in the above formulation. This becomes clearer in discussing the following models for the likelihood functions in Equation (2.26).

Suppose in estimating the true value of A t ths ith source makes error of magnitude E. Two simple models relating At , E, and A t, an are I

A3A t+E (2.30)-

O c, . x , . E (2.3n In the model of Equation (2.30), if a normal distribution is assumed for '

the error term of the estimate of each source, the likelihood function l will be a normal distribution with mean equal to At + bj, where bj is the expected error or, in other words, a "bias" term about which  !

the error of the ith source is propagated. i Formally, we have '

i g

/ A *-(A 4 t +4 b ))If P(X9 *lAt ) = exp --i (2.32)

(G o; E

\ 01 Iy The variance of the likelihood, of, is the variance of the error distribution. Values of bj and oj are assessed by the data analyst subjectively and reflect the credibility and cccuracy of the source as viewed by the data analyst. Sometimes, certain information provided by the source such as the uncertainty bound for the estimate can be uS d to assess oj.

If, in addition to a normal likelihood function, a normal prior distribution representing the state of knowledge of the data analyst is 2-11 -

Olb4G0618860AR

assumed for Ag with mean A0 and variance o8, the posterior distribution in Equation (2.25) will also be hormal with mean, A ,p given by N

A p = ,E w j(AJ-bj ) (2.33) 1=0 and variance N

I o

p

=

,E 2 (2.34) 1=0 o g where w j , defined as a 2 w - P-i= oj (2.3S) is the weight given to the ith source.

Note that w

9

=1 (2.36) i=0 The mean, therefore, is a weighted average of the individual estimates af ter correcting for their expected biases. Also, as can be seen from Equation (2.35), smaller values of oj result in higher weights, implying that the source which is believed to make errors of smaller magnitudes (oj is the variance of E) is assigned a higher weight; something which is intuitively expected. Extreme cases are when oj = 0 (highest degree of confidence in the ith estimate), for which wj = 1, cod when oj = = (no confidence at all) for which wj = 0.

If, instead of the model of Equation (2.30), the model of Equation (2.31) is applied and the logarithm of the error is assumed to be normally distributed, the likelihood function for the ith source becomes a lognormal distribution i

finAj - (AnAt+ nb ))21 P(AylA)"

j t gg 1

exP 1

7 j[ (2.37) where Anbj is the logarithmic mean error about the logarithm of the true value, AnAt , and oj is the multiplicative standard deviation.

Again, Pj(Aj*lAt ) is the probability that the estimate of the ith source is Ai when the true value of the failure rate is At. Some evidence in 2-12 OlS4G0618860AR

support of the lognormality of Pj(Afl At ) are provided in References 2-9 and 2-10.

O -

V By using the model of Equation (2.37) for individual likelihoods in Bayes' theorem, Equation (2.25), and assuming a lognormal prior -

distribution for A t the posterior state of knowledge will also be a i lognormal with the following median value N ( A*) wg A

50.p

  • N (2.38)  !

i =0 (T)Ii where wi is defined as in Equation (2.35).

The median, then, is a weighted geometric average of the individual '

estimates 6fter correcting for the multiplicative biases. Note that the usual arithmetic and geometric average methods frequently used in the literature are special cases of these Bayesian normal and lognormal models. For instance, Reference 2-4 uses the following geometric average of the estimates provided by several experts N \ 1/N T = (H A (2.39) L (i=1 4)I I

no bias (b = 1), no prior which assumes information, and does equal weights not show g an (W = f)u,ncertainty $ bout the resulting valu

)

Example Reference 2-5 provides a point estimate of 5.60 x 10-3 for the demand '

f ailure rate of motor-operated valves. We would like to use this estimate and obtain a state-of-knowledge distribution for the MOV failure rates. We use the lognormal model of Equation (2.37) to express our confidence in the estimated value 1

exp I

1 finAj-(inA+t nb ))y 21 P(AylA ) = 7 (2.40) where A$ is the estimate (5.60 x 10-3) and A t is the assumed true value of the failure rate which remain ; an unknown variable at this point. Our  ;

subjective judgment about the magnitude of error of the data source is expressed by assigning numerical values to the "bias" term b l and the logarithmic standard deviation ot. ~

We assume that there is no systematic bias (bl=1). We estimate 01 with the aid of range factor (RF) which is a more understandable quantity. Unless otherwise indicated, the range factor here is defined as the ratio of the 95th to the 50th percentiles of the lognormal O

2-13 i 0154G0621860AR -

distribution. Therefore, given the range factor, the value of at is obtained from the following equation 1" NI o 1= 1. 64 b (2.41)

For our example, we assume a range factor of 3. Normally, such a range factor represents a relatively high degree of confidence and means that the source's estimate could be a factor of 3 higher or smaller than the true failure rate and such a statement is made with 90% confidence.

Using this range factor in Equation (2.41) results in a value of 0.67 for ol.

If we now use the likelihood of Equation (2.40) in Bayes' theorem, Equation (2.25), and assume a flat prior distribution, Po(At), the posterior distribution will be

" A ~ E P(AlAj = b.6 x 10-3) = 106.66 exp - f U6 (* )

which has the following characteristics:

bth Percentile: 1.87 x 10-3 both Percentile: 6.6 x 10-3 9bth Percentile: 1.68 x 10-2 Meant 7.01 x 10-3 2.1.1.2.2 Estimating Distributions Using Point Estimates of Various Sources We now go back to our original problem which was estimating the genericfailureratedistribution4(Al0). This time, however, we assume that instead of having the set of <k t,T i> defined from various plants, we are given one estimate, A,in Equation j, for (2.8) each plant.

That is, the evidence is of the form 1 2 = ( A i i =1, . . . , N) (2.43)

The model to be used is a combination of the methods presented in Sections 2.1.3.1 and 2.1.3.2.1 and is fully discussed in References 2-7 and 2-11. A particular family of parametric distributions,

$(Al0), is assumed for X and the information 12 is used in Bayes' theorem to obtain a posterior distribution over the entire set of possible values of 0 and consequently over all possible distributions 4(Xl0).

Formally P(0l12 , 10) = k-1 L(1l0,1)Po(0l1) 2 0 0 (2.44)

See the set of definitions immediately following Equation (2.9) for interpretation of the terms in Equation (2.44).

2-14 -

0164G0618860AR

The total likelihood function in the present case when Ai's'are

/q independently estimated can be written as [see Equation (2.11)]

O N L(1l0,1)*

2 0 O i(Ayl0,I)O (2.45) i=1 where Pj(Ail0,lo) E probability that the estimate provided for (2.46) if the parameter of the ith plant varia the population is A[ bility distribution of '

the failure rates is 0.

t To make matters clearer, note that we are assuming that the ith source of  ;

data is providing an estimate for the failure rate at a particular plant and all we know is that failure rates vary from plant to plant according to,thevariabilitycurve4(Al0). Each Aj, therefore, is an estimate of one point in that distribution. As a result, there are two sources of variability.in the estimates. First, estimates of individual sources are not necessari1 perfect; i.e., they could involve errors and biases as discussed in Section 2.1.1.2.1. Second, even if all the sources were -

perfect, the estimates would still be different due to the actual variation of the failure rate from plant to plant.

Based on our discussion in the previous section, the confidence that we  ;

have in the accuracy of the estimate Ai for the failure rate at  ;

the ith plant can be modeled by a lognormal distribution [see Equation s2.37)]. Assuming no bias, we have  :

P (Ag p g) =

1 ( 1 ftnAi_AnAj32 '

/ j) exp 4 1 (2.47) 4 7 \ i

,/2E o jA y (  !

where Aj is the true value of the failure rate at the ith plant.  :

Again, we really do not know Aj but we assume that it belongs to 4(Al0), the distribution plant to plant. The relationrepresenting between Pj(AIl0,lo) the variability and 4(Al0 of Ai)s from is shown in Figure 2-4.

Therefore, as we did in the case of Equation (2.13) we write Pg (A{l0,10 ) = Pg (AflX) 4 (Al0)dA (2.48) {

U l

As it was mentioned earlier, in developing the failure rate distributions  !

$(Aj0) is sisamed tu be lognormal defined by Equation (2.16). With j i

l l

l l

i 2-15 Olb4GU61886DAR

this assumption, the integration in Equation (2.48) can be done analytically and the result is 1 3 ( AnA*-Any )2 P(A{l0,1)"

4 0 **P 4 ~ ~l ( (2.49) 2/o#+c2 j y A 2 c j,c 2

Equation (2.44), Bayes' theorem, is now written as:

N P(0l Ay, . . . Ag) = k'I H P (Ayl0, 01) P0 (0lIO) 9 (2.50) i=1 The most probable and expected distributions of A can be found in the same way as discussed in Section 2.1.1.1. The expected distribution is calculated by using the result of Equation (2.47) in Equation (2.18). The parameters of the most likely distribution are shown to be solutions of the following system of equations (Reference 2-12) 2 N (ej + a )~I -

En = ,E g AnAy (2.b1) 1=0 2 (cj 2 , o )- 1 i=0 N [(AnA* - Any))2

~ = 0 11 oj 2,c2 aj 2,oz (2.52)

For perfect sources of information (i.e., oj = 0), the above equations simplify and result in the following solution

/N )1N m =1 H

(i=1 Xi / (2.53)  ;

1 N

o = f ,E (En Ay - Rn p) (2.54) 1=0 Note that Equations (2.53) and (2.54) are similar to the conventional results for fitting a lognormal distribution to a set of estimates. It should also be mentioned that the results of this section apply to any set of failure rate estimates from various sources where a true variability is suspected to exist among the actual values being estimated by each source. For instance, if several generic sources of data provide estimates for a particular type of equipment and it is known or suspected 2-16 Olb4G0618860AR

that each source's estimate is based on a different subset of the population, the methods of this section can be applied to obtain a hq generic distribution representing the "source to source" variability of the failure rate. '

Example i

The following set of estimates are available for the demand failure rate of MOVs.

Source Estimate WASH-1400 (Reference 2-3) 1.00 x 10-3 N-1363 (Reference 2-5) 5.60 x 10-3 GCR (Reference 2-12) 1.00 x 10-3 t To'use the model of this section, we need to assign range factors to each source as a measure of our confidence in the estimate provided by that source. In this way, we will be able to determine Pj(A*jl Aj),

Equation (2.47), for each source. -

Following our discussion in the example of Section 2.1.1.2.1, we assign a range factor of 3 to the estimate of N-1363. For the estimate of '

WASH-1400, we as, sign a range factor of 5 which results in a broader

.N likelihood, Pj(AjlAj), for that source and represents a less (d degree of confidence as compared to N-1363. This is due to the fact that the estimate of N-1363 appears to be based on a larger sample of MOV f ailures in nuclear applications than the estimate of WASH-1400. The latter provides a ran median (1.00 x 10-3)we ge have factortaken of 3 for as thethe lognormal distribution estimate. Assigning a whose larger i range factor of 5 also means that we believe WASH-1400 has overstated its confidence in the estimated median value.

The idea of broadening some of WASH-1400 distributions when used as >

generic curves was introduced in an early site-specific PRA study (References 2-13 and 2-14) where the WASH-1400 curves (as given) were used as generic prior distributions. It was then found that several posterior distributions, reflecting the evidence of the specific plant, lay in the tail region of the prior distributions on the high side.

These results led'us to the conclusion that the generic curves had to be broadened to reflect greater uncertainty.

References 2-15 and 2-16 provide further support to our decision. In Reference 2-15, the authors review experimental results that test the adequacy of probability assessments, and they conclude that "the overwhelming evidence from research on uncertain quantities is that people's probability distributions tend to be too tight. The assessment of extreme fractiles is particularly prone to bias." Referring to the Reactor Safety Study, they state "The research reviewed here suggests that distributions built from assessments of the 0.05 and 0.95 fractiles p may be grossly biased."

b 2-17 Olb4G0518860AR

Commenting on judgmental biases in risk perception, Reference 2-16 states:

A typical task in estimating uncertain quantities like failure rates is to set upper and lower bounds such that there is a 98%

chance that the true value lies between them. Experiments with diverse groups of people making many differe'it kinds of judgments have shown that, rather than 2% of true values f alling outside the 98% confidence blunds, 20 to 50% do so (Reference 2-15). Thus, people think that they can estimate such values with much greater precision than is actually the case.

The numerical effect of using a larger range factor is illustrated in the following table Distribution Mem an Mean Perc tile Per e ile ac or

3. 3 x 10'

~

-3 WASH-1400 x 1f 1.2 x 10 3.0 x 10 3 Broadened -

Distribution 2.0 x 10~ 4 ' L ') x ~1 0-3 1.6 x 10 5.0 x 10' 5 l _

We see here that the medians are the same and the mean value increases slightly reflecting the extension of the high side tail of the curve.

For the cases where WASH-1400 we.s the only source used for a failure rate, the above methodology was used to generate a broader generic curve from the distribution of WASH-1400. The applied range factor, however, was not necessarily the same for each case. Several examples of this situation can be found in the detailed failure rate description of Reference 2-17.

Similarly, we assign a range factor of 10 for the GCR estimate. This reflects a lower degree of confidence in the estimate of Reference 2-12.

These range factors can be used to obtain the corresponding oj values by using Equation (2.41). The results are c1=0.67, c1=0.98, and o3=1.40, for WASH-1400, N-1363, and GCR, respectively. These values as well as the estimate from the three sources were used as the main input to the mode 2 of the computer code BEST that calculates Equations (2.47) through (2.b0) and obtains an expected curve based on an integration similar to Equation (2.18).

The resulting histogram has the following characteristics:

5th Percentile: 8.4 x 10-4 50th Percentile: 1.6 x 10-3 95th Percentile: 7.4 x 10-3 Mean: 2.0 x 10-3 2-18 0154G061886DAR

2.1.1.3- Generic Distributions Based On a Mixture of i*ype 1 and Type 2 Data An obvious extension of the situations discussed in Sections 2.1.3.1 and 2.1.3.2 is the case where a mixture of 1 2 and 1 1 information is available. In this case, the equivalent of Equations-(d.9) and (2.44) is 2 1 IO) = k-1 L(Ig,11 l0,Ig)P0 (0lIO)

P(0l1.1 (2.bb)

If 1 1 and 12 are independent pieces of information L(12*I1l0,lo) = L(12l0,lo)L(11 l0,lo) (2.56) where the terms in the right-hand side of the eg" ' inn are defined by Equations (2.11) and (2.45).

The expected distribution of A can now be found from 5(A) = 4(Al0) P (0l1 2 'I1'I O)de (2.57)

O Example As an example, we use the combination of the data given in the examples in Sections 2.1.1.1 and 2.1.1.2.2. This information was used as the main

, 3 input to mode 3 of the computer code BEST3 which calculates Equations (2.bb) through (2.57). The resulting discretized dis 1ribution has the following characteristics:

5th Percentile: 7.49 x 10-4 l

50th Percentile: 2.84 x 10-3 95th Percentile: 1.05 x 10-2 l Mean: 4.30 x 10-3 2.1.1.4 Development of Generic Failure Rate Distributions I Developing a generic data base requires a thorough review, analysis, and tabulation of the available generic data for each of the identified component failure moda. The PLG generic data base is proprietary. It i was updated to its current form during the Seabrook PRA (Reference 2-18),

l and it is documented in Reference 2-17, a PLG proprietary report. This  !

PLG generic data base was used as the generic data basis for the TMI-1 PRA. In addition to generic data sources, several well documented site-specific failure rate data from power plants examined in previous or ongoing risk studies were used in the development of the generic data i base. This assures that the final failure rate distributions accurately l reflect all information currently available. j A practical difficulty in using the available generic estimates in the process of developing generic distributions was the lack of j 2-19 <

0154G061886DAR

standardization in the generic literature. This dictates that utilizing generic sources involves much more than a simple catalog of published failure rate estimates. Each source presents its own unique set of advantages and drawbacks, and these factors must be carefully evaluated before a meaningful comparative analysis may be performed. Typical problems encountered include incomr tibility between failure and test data, inclusion of failures due to ocher than hardware related causes, exclusion of failures due to licensing based reporting criteria, and a general lack of specific documentation of assumptions made, boundary conditions, and methodologies applied. Often it is simply not possible to discern the reasons far significant differences among several sources publishing data for the same component failure mode.

3ecause of the inherent difficulty in ascertaining the direct comparability among these various estimates, the only practical approach to the problem is the assignment of subjective "weighting factors" to each piece of data, based upon the perceived compatibility of the source with the desired failure rate information. These weights are assigned by assessing either a range factor or o parameter for the likelihood functions for each source according to the models discussed in Section 2.1.2. This process is computerized via the camputer code BEST3, which takes as input various point estimates and corresponding subjective range factors as well as plant-specific experience of the component in question at various plants. The code then performs Bayesian calculations based on the models and generates an average distribution for the failure rate representing source to source and/or plant to plant variability of the data. This process involved several iterations in running the code and reviewing the results to ensure that the range of discrete probability distribution was a reasonable representation of the input information and that the binning of the distribution (20 bins or less) was done properly.

In other cases, where only one source of data was available for the component, failure rate distributions were represented as 1. anormal. In general, these failure rate distributions were derived by defining the median value and range factor as the two most physically meaningful parameters of the lognormal distribution (the range factor is defined here as the ratio of the 95th percentile to the median, or the square root of the ratio of the 95th and 5th percentiles). In order to provide traceable documentation of the data sources used in this analysis, the median value of such distributions was based on published data. The range factor was subjectively assigned such that the resulting 5th and 96th percentiles of the distribution represent realistic bounds for expected or observed component failure rates.

The relative magnitudes of the range factors developed for the various distributions were influenced by a set of consistent evaluation criteria. In general, range f actors significantly greater than 10 (i.e.,

a span of more than 100 in failure frequency between the 5th and 95th percentiles) were considered to produce distributions so broad as to convey a nearly uninformed state of knowledge and, therefore, would be of marginal utility in any quantification process. The mean value of such a broad distribution, while defined mathematically, is virtually 2-20 Ulb4G061886UAR

meaningless as a representation of expected component performance

(- because, in truth, very little is known about how the entire population

( behaves. Some distributions were assigned range factors on the order of 10. Typically, these distributions were characterized by sparse generic data not closely correlated to the desired component failure mode and a relatively low degree of confidence in the available source. It is felt that a distribution this broad conveys only marginal knowledge as to the behavior of a populatien and is generally indicative of the application of good engineering judgment to minimal prior information_

Some distributions were assigned range factors on the order of 3 to b (i.e., spans of approximately 10 to 25 between the 5th and 95th probability percentiles). While these distributior,s are still relatively broad, they represent a higher degree of confidence in the failure rate estimate used as the median value.

Treatment of the generic distributions from IEEE STD 500 (Reference 2-4) is discussed in the following. This reference contains data for electronic, electrical, and sensing components. The reported values were mainly synthesized from the opinions of some 200 experts (a form of the Delphi procedure was used). Each expert reported a "low," "recommended,"

and "high" value of the failure rate under normal conditions and a "maximum" value which would be applicable Under all conditions (including ~

abnormal ones). The pooling of the estimates was done using geometric averaging technique, e.g.,

/N ) 1/N A

max N A max,i)I (2.58)

(i =1 >

This method of averaging was considered a better representation of the expert estimates, which were often given in terms of negative powers o f 10. In effect, the usual arithmetic averages of the exponents were used, which, as discussed in Section 2.1.1.2.1, is a special case of the Bayesian model presented in this report.

Reference 2-4 does not recommend a distribution. The method of averaging, however, suggests that the authors have in mind a lognormal distribution.

given inforcation.

Our task now is to determine this distribution from the The recommended value is suggested to be used as a "best" estimate. The word "best" is, of course, subject to different interpretations. We have decided to use it as the median value mainly for two reasons. First, for skewed, lognormal type distributions, the median is a inore representative measure of central tendency than the mean, which is very sensitive to the  :

tails of the distribution. Thus, we suspect that the experts who i submitted their "recommended" estimates actually had in mind median '

values. Experimental evidence (Reference 2-19) also indicates that l assessors tend to bias their estimates of mean values toward the l medians. The second reason is that this choice is conservative, since '

i the mean value of our resulting distribution is then larger than'the "reco, amended" value. The "maximum" value is taken to be the 95th percentile of the lognormal distribution.

]

2-21 1 0154G0618860AR l

For the majority of the components for TMI-1 PRA, generic component failure rates were taken from PLG Generic Data Base (Reference 2-17). In a few cases additional generic distributions had to be developed for some specific types of equipment. Reference 2-17 provides a detailed documentation of the generic distributions used in this study. The mean value of the generic distributions are listed in Section 3 in conjunction with the TMI-specific failure rate distributions.

2.1.2 DATA SPECIALIZATION Data specialization or the development of plant specific failure rate distribution is achieved by applying Bayes' theorem as follows P(AlE2 ) = k-1 L(ElA)P0 2 (A) (2.59) where P(AlE 2 ) is the plant-specific failure rate distribution reflecting the plant-specific experience E 2, and the generic distribution P 0(A) as prior state of knowledge about the failure rate of the component in question. The likelihood term, L(E 2l A), takes the form of a Poisson distribution when A is the rate of failure per unit time and the evidence E2 is k failures in T time units

-U P(kTlA)=I 1

e (2.60)

If A is a demand f ailure frequency and E2 is k failures in D demands, then L(E2 lA) is a binomial distribution yk (2.61)

O P(kDlA)=(U- )! k! (1-A)D-k 1

The magnitude of the effect of adding plant-spe';fic data depends on the relative strength of the data compared with the prior level of confidence expressed in the form of the spread of the prior distribution. Typically both the location and the spread of the posterior or updated distribution is affected by the plant-specific evidence. The mean value of the updated distribution could be higher or lower than the mean of the generic prior but adding the plant-specific data normally reduces the spread of the distribution, as shown in the following example. The generic distribution for the MOV demand failure frequency presented in the example of Section 2.1.1.3 whs updated with 15 failures in 5,315 dem2nds.

Calculations were performed using mode 4 of the computer code BEST3. The following table compares some basic characteristics for the generic prior and updated distributions.

sWudon (per de at:d) Median Perc ntile Percen ile Generic 4. 30 x 10-3 7,49 x 10-4 2.84 x 5. 0- 3 1.0b x IV2 Updated 2.88 x 10-3 1.83 x 10-3 2.82 x 10-3 1. 71 x 10-3 2-22 0154G062386DAR

2.2 COMMON CAUSE FAILURE PARAMETERS In the TMI-1 PRA, dependent failures such as common cause failures at the systems level are treated either explic.itly by.means of identifyir.g causes of dependent failure and incorporating them in the systems or event sequence models, or implicitly by using certain parameters to account for their contribution to the systems' unavailability. Examples of the first category are sharing of common components, fires, floods, and certain types of human error during test and maintenance. This section deals with the econd category, cddressing common cause failures that are nt' covered in the first category, such as design errors, construction errors,

. procedural deficiencies, and unforeseen environmental variations.

The parametric model used in this study to quantify the effect of the second category of dependent failures is known as the multiple Greek letter method (Reference 2-20) which is an extension of the beta factor method (Reference 2-21). The following is an overview of the method and the Bayesian technique used in developing state-of-knowledge distributions reflecti g various sources of uncertainty in estimating the parameters of the method.

2.2.1 OVERVIEW OF THE MGL METHOD in the NGL method, the total failure probability of each component is determined from all irdependent and common cause contributions for that component- For instance, for a component in a system of three. redundant and idelitical components we have Oc ' Q1 + 2Q2 + Q3 (2.62) where Qj is the frequency of simultaneous failure of 1 compc ents.

The common cause paramne.r S is then defined forteach component as the conditional probability of a common cause event involving a second or third unit, given that a specified component failure occurs.

242+03 03 yy + 2y2+N3 '

A second common cause f ailure parameter, y, is defined for each '

component as the conditional probability that a common cause failure involving that component involves all three components in the system N3 YE (2.64) 2y2+N3  :

1 An important observation about the NGL model that is useful in collecting data and estimating parameters is that, for systems having ideni.1 cal components and identical conditions and env ronments acting on the components but with different numbers of com(ponents, the only parameters

p that are "conserved" (i.e., invariant among systems with different O

2-23 0154G061886DAR

component populations) are Q and Q1 All the remaining parameters (Q 2 Q3, S, y) are a function of the number of identical components in the system. For example, consider two systems, one with two components and one with three components. In the two-component system, all common cause events are modeled by Qg inasmuch as no more than two components can fail. However, some of tne common cause events in the two-component system might cause all three components to fail in a three-component system. Inerefore, despite the fact that each component experiences the same causes of independent and common cause events, we have Q2(first system) / Q2(second system)

Q3(first system) = 0 / Q3(second system)

Q2(first system) = 202(second system) + Q3(second system)

To avoid problems, it is recommended that when parameters a e estimated, all data are interpreted to assess the impact of each event in the particular system under investigation. This technique will be illustrated later.

After rearranging Equations (2.62), (2.63), and (2.64), the following identities are obtained Q3 = YSVc Q2= (1 - Y)SQc Q1 = (1 - S)Qc (2.65)

Tnese parameters can now be used to calculate system unavailability due to both iridependent and dependent failures. For example, the following MGL models are obtained for two system configurations. For the one-out-of-three system Q(1/3) = YSQ c +fl-y)(1-S)SQ + h1 - y)2 84 2c (2.66)*

+(1-S)Qf that, to the third order in S and Q, can be further simplified to Q(1/3) = ySQ +f(1-y)SQ2,g3 (2.67)*

The first term on the right side of Equation (2.67) accounts for a triple-component common cause failure, the second accounts for a

  • Approximation comes from the "rare event" approximation.

G 2-24 0154G0618860AR

double-component common cause event in combination with a single q independent failure, and the third represents the case of triple - '

Q independent failures. .j For a two-out-of-three system .

Q(2/3)=h1-y)SQc + YB4c + 3(1 - S)292e (2.68)*

which to the third order in S and Qc, is simplified to 2

Q(2/3) = f(3 - y)SQc + 3(1 - 28)Q (2.69)*

2.2.2 ESTIMATORS FOR THE PARAMETERS UF THE MGL MODEL j To develop estimators for the parameters of the MGL method, we start with the following general formula for the failure frequency Qc of a component in a system of m (identical and redundant) units s

N*c N

( -1)! J (2.70) -

1 (m-l Where Qj d Failure frequency of simultaneous failure of j components in '

the system.

For instance, for a component in a three-unit system (m = 3), we have Nc" 1 (3-j ) (j-1)! NJ 21 21 21

" 21 of 41 + 11 11 N2 + O! 21 4 3  ;

"41 + 2Q2 + N3 (2.71)

An estimator for Qj is j

N* " ml (2.72)

J (m-j )! J ! , D  !

l.

  • Approximation comes from the "rare event" approximation.

O  !

i 2-25 I

Ulb4G061886DAR

where nj E number of events involving j components in failed state.

fl0 5 number of demands on the entire system of m components.

Replacing Qj in Equation (2.70) with the corresponding estimator yields m

l Q**mK c E d"j (2.73)

J J=1 In the following, we develop estimators for the first three parameters of the MGL model for a system of m components. Estimators for the higher order parameters can be developed in a similar fashion. Based on the definition of S as the conditional probability that given a specified component f ailure, at least one other component also failed due to the same cause, we have 0*Q (m- ( -1) Oj (2*74) .

The parameter y is defined as the conditional probability that given a common cause failure of two components, at least a third unit also failed due to the same cause. Therefore, 1 (m-1)!

Y = ggc j 3 (*~d )I (d-1) d Similarly 6 = BY c j 4 (m_ ( _1) ) (2.76)

Therefore, using Equations (2.72) and (2.73), we obtain fm ) {m S=1 E jn .)' (2.77)

(j E

= 2 jn)/

i i

\j=1 U) y =i

{m E jn)i i {mE j n .) (2.78)

(J=3 j) (j=2 J)

{m m 6 =i E jn)i (j=4 j/

l

{(j=3 E jn)))1 (2.79) 2-26 Olb4G0618860AR

For instance, for a three-unit system (m = 3), we have O

V 2n2+3"3

(*

~

83 " n +2n +3n 3 )

1 2 -

3n Y3 " 2n2+ n 3 (* )

2.2.3 ASSESSMENT OF UNCERTAINTY Point estimators developed in the previous section only provide single values for the parameters of the MGL model. However, since the estimates are typically based on limited information, the true value of a parameter may actually differ from the point estimate. The objective of uncertainty analysis is to assess the range of values of each parameter based on the' available information and various sources of uncertainty. .

Variation of the value of a parameter could be due to one or a combination of the following reasons:

1. Size of the data sample.
2. Uncertainty in data classification.

3.

O Variation among the plants in equipment systems and operational philosophy.

The following sections describe how each of the above sources of .

uncertainty was treated in this study.

2.2.3.1 Assessment of Uncertainty Duq to Data Sample Size The first of these sources of uncertainty is a well-known subject in statistics. Larger sets of failure and success data would result in ';

estimates with higher degrees of confidence simply because they are more represenative of the general population. For instance, in  !

Equation (2.77), the larger the total number of failures given by the term i m

N t" j=1 d"j i the more accurate the estimated value of S. The mathematical models i presented in the following Bayesian method provides a mechanism for handling this source of uncertainty. We will limit the discussion to a two-parameter MGL model which applies to a system of three components.

Extensicn of the results to higher order parameters will be a simple task.

Based.on the definition of 8 and y, [ Equations (2.63) and (2.64)], we define b

d Z1E l-6 conditional probability of component failure being a single failure.

2-27 -

Ol64GU61886DAR

Z2 E B ( 1-Y ) conditional probability of a component being involved in a double failure.

Z3 E BY conditional probability of a component being involved in a triple failure.

Note that Zy + Z.g+Z3=1 p.82) as expected.

The likelihood of observing ni single f ailures, 2ng component f ailures due to common cause, and 3n3 component failures due to common cause, can be modeled by a multinomial distribution for Zj's n 2n. 3n P(ni ,2n2 ,3n3 ,lZl ,Z 2,Z3) = C Z 1 1 Z 2

2 3 (2.83) where C is the binomial coefficient. Rewriting Equation (2.83) in terms of S and y gives 2n.+3n g n 3n 2n 2

P(nl ,2n2 ,3n3!0'Y) = C S 3 (1-S ) 1 y 3 (1 y) (2.84)

We now write Bayes' theorem as follows n(B Ylny ,2n2 ,3n3 ) = k P(ny ,2n2 ,3n3 !0'Y)"0(S,y) (2.85) where no and n are the prior and posterior distribution of S and y and k is a normalizing factor defined as k= P (ng,2n 2'3"3l8,Y)"O(6,Y)dSdy (2.86)

As the prior one can use a multinomial distribution Ag-1 B -1 Cg-1 D n0I6'Y) = k S (1-8) g y (1 y) 0-1 (2.87) where k is a normalizing factor.

A noninformative flat prior distribution is obtained by setting A0=B0 = Co = DO"l Using Equation (2.87) in Equation (2.85) results in a posterior distribution for S and y which is also multinomial with parameters A=A0 + 2n2 + 3n3 B = Bo + n1 2-28 OlS4G061886DAR

C = Cu + 3n3 (tj U = Do + 2n2 (2*88)

The mode of the posterior distribution occurs at l A-1 O

mode

  • A+d-2 (2.89)

C-1 Ymode " C+0-2 (2.9D)

The mean values are calculated from A

<0>

  • A+B (2.91)

C

<Y>

  • c+o (2.92) _

Note that for a noninformative prior, the mode of the posterior distribution is 2n2+3n 3 V,q Omode " n +2n,+3n y g (2.93) 3 3n 3

Y

  • mode " 2n 2 +3"3 which correspond to the point estimates developed in Section 2.2.2 for m = 3.

The variance of the posterior distribution for S and y are

^

V(S) = (2.95)

( A+B )l( A+B +1 )

CD V (y ) = -- *

(C+D) (C+D+1)

For instance, for a noninformative prior (2n2+3n3+1)("1+1) v(B) =

(~))

( (n 1+2n2+3n3 +2) (n 1+2n23n3+3)

(2.97) 2-29 0154G0618860AR

As we can see, the variance decreases as the total number of failures (n = n1+2ng+3n3) increases. Since smaller variance means smaller range of; uncertainty, the larger the number of failures in the data sample, the higher the confidence in the estimated value.

2.2.3.2 Assessment of Uncertainty Due to Data Classification An important source of uncertainty is the judgments that are made in the process of classification of data for use in quantifying common cause parameters. Treatment of this type of uncertainty and several other aspects of data classification that have direct impact on the assessment of common cause parameters are discussed below.

Of fundamental importance in a meaningful assessment of the contribution of common cause events to the system unavailability is a detailed review and systematic classification of failure events experienced in the nuclear industry. The data used for this study are based on review and cla.ssification of several thousand failure events reported by U.S.

nuclear power plants, as well as TMI-specific component failure events.

The data classification approach was that of Heference 2-22. In short, events were classified into one of two categories of dependent and independent events. Dependent events are those that involve several ~

component abnormalities that are casually related. All other events are classified as independent. Abnormal states of components a e classified as either failed or functionally unavailable where, in both cases, the component is not capable of performing its function according to a given success criterion. The f ailed state applies to cases where, in order to restore the component to operability, some kind of repair or replacement action on the component is necessary. A functionally unavailable component, however, is capable of operating but the function normally provided by the component is unavailable due to loss of input such as motive power, command signal, cooling water, air, etc.

Sometimes, even though a given success criterion has been met and the component has performed its function according to the success criterion, some abnormalities are observed that indicate that the component is not in its perfect or nominal condition. Although a component in such a state may not be regarded as unavailable, there may exist the potential of the component becoming unavailable with time due to changing conditions, or due to more demanding operational modes. Events involving these potentially unavailable states provide valuable information about causes and mechanisms of propagation of failures and thus should not be ignored. The concept of potentially unavailable states also serves a practical need to enable the consistent classification of "grey area" cases and difficult-to-classify situations. The "potentially unavailable" component state category is defined for this situation. It refers to the cases where the component is capable of performing its function according to a success criterion but an incipient or degraded condition, as defined below, exists.

s Degraded. The component is in such a state that it exhibits reduced performece but insufficient degradation to declare the component unavailable according to the specified success criterion. Examples g 2-30 Olb4GU618860AR

of degraded states are relief valves opening prematurely outside the p technical specification limits but within a safety margin and pumps U producing less than 100% flow but within a stated performance margin.

e Incipient. The component is in a condition that, if lef t unremedied, could ultimately lead to a degraded or unavailable state. An example is the case of an operating charging pump which is observed to have excessive lube oil leakage. If left uncorrected, the lube oil would reach a critical level and result in severe damage to the pump.

A key to distinguishing between degraded and incipient conditions is the knowledge that an incipient condition has not progressed to the point of a noticeable reduction in actual performance, as is the case with a degraded condition.

It is important to recognize that potentially unavailable is not synonymous with hypothetical. Both incipient and degraded conditions are indicative of observed, real component states that, without corrective action, would likely lead to unavailable component states.

Dependent events were further grouped into events in which the cause of l failure of the component (s) of interest is the failure of another '

component (component-caused events) and those where the cause(s) of failure (s) is something other than the state of another component .

(root-caused events). Finally, events in the dependent category were l Screened based on a set of criteria for applicability to PRA type systems  ;

g analysis in general and the TMI-1 PRA in particular. The events that are i not screened out in this process are named "common cause events" and are used to estimate the common cause model parameters.

In estimating the parameters of the MGL model, a particular system size must be considered. The next step is to calculate the number of  ;

component failures for each of the various "system impact" categories.  ;

System impact category refers to the number of components being affected in an event. For instance, if in an event two components are failed, the ,

system impact category for that event is 2. We explain this step with the aid of a hypothetical example.

Suppose that we want to estimate the common cause contribution to the unavailability of a system of three identical redundant components.

Therefore, we need to estimate 6 and y in addition to the failure rate of the component. Suppose, further, that the data after screening -

indicate that there have been 88 independent events involving 70 actual '

and 18 potential failures. In addition, assume that there have been three common cause events:

e Event 1. Common cause failure of two components in a system of two components. However, the cause of failure is such that if a similar event occurred in our example system, it would most likely affect all three components.

i e Event 2. Two components failed within a short period of time but it

( l i

cannot be determined, based on the event description, whether the two  ;

failures shared the same failure cause.

2-31 0154G061886DAR j

e Event 3. One component f ailed and another in degraded condition (potential failure) due to the same cause.

Event 1 involves a situation where the data from a two-component system should be "extrapolated" by postulating the impact of the cause of the event on a three-component system. Therefore, with regard to the "system impact" of this event, there are two hypotheses: (1) the cause only affects two of the three components, and (2) it affects all three.

Weights can be assigned to each of the two hypotheses that reflect the analyst's judgment regarding the two hypotheses. In Table 2-1, this situation is represented by weights of 0.0b and 0.95 assigned to the first and second hypothesis, respectively.

Event 2 also involves two hypotheses: (1) two components were affected independently, and (2) the event is a common cause failure of two components. In Table 2-1 a weight of 0.9 is assigned to the first hypothesis, while the second one is given a weight of 0.1.

In event 3, we are dealing with a common cause situation. However, only one component actually failed, while tne state of the other one was "potentially failed." If we assign a weight of 0310 to the potential f ailure, the effective number of failures in the event is 1 + (0.1)(1) = 1.1, as can be seen in Table 2-1. Note also that there is only one hypothesis regarding the system impact of the cause.

Table 2-1 summarizes the information obtained for the common cause events in the form of ef fective number of component failures in each event for each hypothesis. This effective number can be calculated for the jth event from i

b jj " k=1b W jik (2.98) where i E "system impact" index, which is defined as the number of components assumed to be affected by the cause, wjik E the weight assigned to the state of the kth component in the event j for system impact index 1.

In our example, for the independent events the effective number of failures is 70 + (0.1)(18) = 71.8, where 0.1 is the weight given to potential failures.

In addition to the effective number of components per hypothesis, Table 2-1 lists the weight given to each hypothesis. Finally, the last column of Table 2-1 provides the effective number of failures for each system impact category. For category 1, this number is calculated from J I 5

4= [

j=1 p 5 jj 34 5 effective number of component failures for system impact category i (2.99) l 2-32 l 0154GU61886DAR

where pjj is the weight given to the ith hypotnesis regarding eveat j.

b  !

lie are now ready to calculate point estimates for 8 and y using nj's .

n2 + "3 '

S =_ _ _ (2.100) ny + n2 + "3

  • 3 y =_ , (2.101)

I n

2 + "3 For the present example, based on the values provided in Table 2-1, we ha v,e g

0.41 + 2.85

= 0.04 i 74.59 + 0.41 + 2.85 -

y 2.85

= = 0.87 0.41 + 2.85 t 3 The above estimators reflect the uncertainty due to data classification. l The value of nj's could also be used in the likelihood of Bayes' -

theorem discussed in Section 2.2.3.1 to obtain the combined effect of  :

uncertainties due to data classification as well as data sample size.

2.2.3.3 Plant-to-Plant Variability of the MGL Parameters The third source of uncertainty is the variation of the value of the parameters from plant to plant. This type of variability stems from the f act that similar equipment and systems in various plants may show inherently different failure rates due to a variety of reasons, such as  ;

minor design differences within the same category of equipment and variation in system designs and operating philosophies leading to different coupling mechanisms.

There are two approaches for dealing with this issue. One approach is to assess the variability of the parameters based on statistical evidence from all plants without screening events based on their applicability to the situation under consideration. This results in a wider range of possible values for the parameters. In the second approach, failure events from various plants are reclassified and events not considered to be applicable to the plant or system of interest are excluded from the data base. The result is the formation of a data sample much larger than one based only on the records of the specific plant under consideration.

The resulting uncertainty range for the estimated parameters will obviously be smaller in this case as compared with a distribution 2-33 0154Gu619860AR

representing differences in plants. This reduction in uncertainty is the result of applying the additional information about the specific characteristics of the system being analyzed. This was the approach taken in this study to quantify the common cause parameters.

2.2.4 GENERIC COMMUN CAUSE PARAMETER DATA BASE Based on the approach described in the previous section, the generic data is normally screened for applicability to the particular systems analyses being considered. In that sense, the industry-wide data is specialized to the TMI-1 plant even at the "generic" level. The generic data used for this study and the result of event screening are documented in Reference 2-17. The data base included common cause events for several key components such as reactor trip breakers, diesel generators, pumps, and valves. Mean values of the generic distributions are provided in Section 3, in conjunction with the updated distributions.

2.3 COMPONEh! MAINTENANCE DATA Maintenance activities which remove components from service and alter the normal configurations of mechanical systems can provide a significant contribution to the overall unavailability of those systems. This -

section describes how generic and plant-specific maintenance data are used to develop the distribution of component maintenance unavailability.

These distributions apply to maintenance performed during unit noncold shutdown operating periods (i.e., at power operation or in some cases, at hot shutdown). These include the regularly scheduled preventive maintenance. The specific causes leading to these maintenance activities are not delineated; they include repairs of component failures experienced during operation, repairs of failures during periodic testing, removal from service for special testing or inspection, minor adjustments, hardware modifications, etc.

To quantify maintenance unavailabilities, both the frequency and duration of maintenance are necessary; the frequency of maintenance defines the rate at which components are removed from service wnile the duration and frequency combined determine the component unavailability to be applied in the quantification of system unavailability.

The unavailability due to maintenance is calculated from O

n

=

if7( (2.luz) where f is the maintenanc? frequency and T is the mean duration or, as it is f requently called, mean time to repair.

When f*T " 1, then Qg =f*T (2.103)

Therefore, in order to obtain a state-of-knowledge distribution for the unavailability, QM, one needs to have state-of-knowledge distributions 2-34 Olb4GU61886DAR

for both f and T. Such distributions are developed as described in the p follo: ling.

2.3.1 FREQUENCY OF MAINTENANCE The component maintenance frequency distributions for this study wer^

developed by updating generic maintenance frequency distributions using TMI-specific maintenance frequency data. The method of updating was the same used in updating failure rates described in Section 2.1.2. Five generic maintenance frequency distributions were developed for five general component categories based on the component type, its normal service duty, and the applied technical specifications inoperability limitations. The basis for these distributions is described in Reference 2-16 and their mean values are presented in Section 3 in conjunction with the TMI-specific distributions.

2.3.2 DURATION OF MAINTENANCE As applied in this data base, the duration of a maintenance event includes the entire time period during which the affected component is unavailable for operation. This period is defined from the time when the component is originally isolated or otherwise removed from service to the time when the component is returned to service in an operable state and, in many cases, it may be only weakly dependent on the actual time required for maintenance personnel to effect the repairs.

Five generic distributions for the maintenance duration were used from Ok' the PLG proprietary data base documented in Reference 2-17. The distributions for the TMI-specific mean maintenance duration were developed based on the five generic maintenance duration distributions, updated with TMI-specific component repair times. The following explains the Bayesian technique that was used to develop these distributions.

The following analytical model is used to model the variability of the repair times from plant to plant or from occasion to occasion.

Let t denote the actual repair time in any instance, and imagine that the value of t has been recorded for many, many occasions where this repair operation was performed. From these records, one would be able to plot a curve $(t), showing the frequency distribution of t. The desired mean duration, T, could be immediately computed from this distribution T = t $(t)dt (2.104) 0 To know T, therefore, we need to know 4(t). The problem, of course, is that in real life, one does not usually have the curve 4(t).

Usually, all one has is a small set, E, of values E = { t j; i = 1, . . . . N)

(2.105) t')

() where the tj's represent the observed repair times.

2-35 Olb4G061886DAR

Within the Bayesian framework, the solution to this problem is straightforward. One imagines the true distribution, 4(t), as embedded within a parametric distribution space, 4(tl0), and the probability g distribution is erected on this space using Bayes' theorem and evidence E P(0lE,E 0

) = k-IL(El9,E0 ) P0(0lE0) (2.106) where E is given by Equation (2.105) and P(0lE,Eo) is the poster or probability distribution over 0, the set of parameters of 4(t).

It is assumed that there is a minimum repair time, to, and that the actual repair times are mostly distributed about an average value, with a few much longer than the average. In other words, it is assumed that x=t-t0 (2.107) is_approximately lognormally distributed 4(xlp.o)= I - E" V exp G ox J fE"X\ / l (2.108) -

If x is distributed according to Equation (2.108), the likelihood of observing a particular value, x1, where E t -t x$ 4 0 [2.109) is (g (xg lp.o ) =

1

- exp I1 IE"*1 - E"P\ I q 7 '\ (2.110) i S o xj ( / J(

consequently, L(El0,E 0 ), the total likelihood in Equation (2.106) for 0 = {p.o), becomes N

1 2 *Nlp, ,EO ) " O Di (*ilp, )

L(x ,x (2.111) i=1 The posterior, P(p,olE,E0), which can now be calculated from Equation (2.106), is the probability distribution for different pairs of u and o an.1 consequently for different d(x_lp.o) given by Equation (2.108).

Each such distribution has a mean value, x, which is given by II 2 y exp o (2.112) e 2-36 0154G0623860AR

Therefore, the posterior distribution on p and o is also a probability O distribution about T which is related to the mean repair time by V

T =7+t 0 (2.113)

We now have a probability distribution for T (t0 is a constant) which  ;

represents our state of knowledge about the mean repair time in light of r the observed repair times as given in Equation (2.105).

t The generic information enters the picture through the prior distribution Po(olEO ) in Equation (2.106) that for a lognormal maintenance distribution takes the form Po(plolEO ). Therefore, for each category of component the prior state of knowledge needs to be expressed in terms of a probability distribution for y and o. This is done by transforming the probability distribution over the generic distribution l of the actual repair times to a distribution over the parameters of .

lognormal distribution (p, o). This is done with the aid of computer '

code RTIME2 (Reference 2-23), which transforms state-of-knowledge distributions over the 5th and 95th percentiles of the generic lognormal distribution into a discretized grid for p and o. The details of the -

development of the five generic maintenance frequency distributions are provided in Reference 2-17. The mean values of the generic distributions  ;

are tabulated in Section 3 together with the updated distributions. I 2.4 INITIATING EVENTS FREQUENCIES The initiating events are divided into two groups according to the method using for quantifying their frequencies. The first set is composed of {

those events for which the available data from other nuclear power plants i are judged to be relevant. This includes essentially all initiating  :

events except those involving failure of systems that have configurations l unique to the TMl-1 plant, requiring a plant-specific analysis of those systems. l The methodology used to develop the generic and plant-specific  ;

distribution of the frequencies of the initiating events in the first '

group is similar to one used for component failure rates, as described in Section 2.1. The details of the development of the generic frequencies 4

and the compiled raw data are described in Reference 2-17.

l The details of the development of the frequency of the initiating events in the second group (i.e., those requiring plant-specific analysis of the ,

systems involved) are presented in Section 3.5 of this report.

2.5 REFEREhCES  !

2-1. Pickard Lowe and Garrick, Inc., "Methodology for Probabilistic Risk Assessment of Nuclear Power Plants," PLG-0209, June 1981.  !

2-2. Lindley, D. V., Introduction to Probability and Statistics. '

l Part 1: Probability, Part 2: Inference, Cambridge University g Press, 19/0.  !

2-37 0154G062386DAR i

2-3. U.S. Nuclear Regulatory Commission, "Reactor Safety Study: An Assessment of Accident Risks in U.S. Commercial Nuclear Power Plants," Appendix III, "Failure Data," WASH-1400 (NUREG/7b-014),

October 197b.

2-4. Nuclear Power Engineering Committee of the IEEE Power Engineering Society, "IEEE Guide to the Collection and Presentation of Electrical, Electronic and Sensing Component Reliability Data for Nuclear Power Generation Stations," IEEE STD-500, June 1977.

2-b. Hubble, W. H., and C. F. Miller, "Data Summaries of Licensee Event Reports of Valves at U.S. Commercial Nuclear Power Plants,"

NUREG/CR-1363, EGG-EA-5125 June 1980.

2-6. Kaplan, S., "On a 'Two Stage' Bayesian Procedure for Determining Failure Ratts from Experiential Data," IEEE Transactions on Power Apparatus and Systems, Vol . PAS-102, No.1 January 1983.

2-7. Mosleh, A., "On the Use of Quantitative Judgment in Risk Assessment: A Bayesian Approach," Ph.D Dissertation, University of California, Los Angeles,1981. Available from University Microfilms International, Ann Arbor, Michigan. -

2-8. Pickard, Lowe and Garrick, Inc., "Bayesian Estimation Computer Code 4 (BEST4) Users Manual," PLG-0460, December 1985.

2-9. Mosleh, A., and G. Apostolakis, "Models for the Use of Expert Opinions," Proceedings, Workshop on Low-Probability /High-Consequence Risk Analysis, Arlington, Virginia, June 15-17, 1982, Plenum Press, New York, 1983.

2-10. Dalkey, N. C., An Experimental Study of Group Opinion, The RAND Corporation, Santa Monica, California, (RM-6880-PR), 1969.

2-11. Mosleh, A., and G. Apostolakis, "Combining various Types of Data in Estimating Failure Rate Distributions," Transactions of the 1983 Winter Meeting of the American Nuclear Society, San Francisco, California, 1983.

2-12. Hannaman, G. W., "GCR Reliability Data Bank Status Report "

General Atomic Company, GA-A14839 UC-77, July 1978.

2-13. Pickard, Lowe and Garrick, Inc., Westinghouse Electric Corporation, and Fauske & Associates, Inc., "Zion Probabilistic Safety Study," prepared for Commonwealth Edison Company, September 1981.

2-14. Apostolakis, G., "Data Analysis in Risk Assessments," "Nuclear Engineering and Design, 71, pp. 375-381, 1982.

2-16. Lichtenstein, S., B. Fischhoff, and L. D. Phillips, "Calibration of Probabilities: The State of the Art," in: J. Jungermann and G. de Zeeuw, Editors, Decision Making and Change in Human Affairs, D. Reidel Publishing Co. , Dordrecht, Holland, 1977.

2-38 Olb4G061886DAR

2-16. Slovic, P., B. Fischhof f. and S. Lichtenstein, "Facts Versus Fears: Understanding Perceived Risk," in Societal Risk O Assessment, R. C. Schwing and W. A. Albers, Jr., Editors, Plenum Press, 1960.

2-17. Pickard, Lowe and Garrick, Inc., "PRA Propriety Data."

2-18. Pickard, Lowe and Garrick, Inc., "Seabrook Station Probabilistic Safety Assessment," prepared for Public Service Company of '

New Hampshire and Yankee Atomic Electric Company, PLG-0300, December 1983. '

2-19. Peterson, C., and A. Miller, "Mode, Median, and Mean as Optimal Strategies," J. Exp. Psych., 68:363-367, 1964. ,

2-20. Fleming, K. N., and A. M. Kalinowski, "An Extension of the Beta Factor Method to Systems with High Levels of Redundancy," Pickard,

  • Lowe and Garrick, Inc., PLG-0289, June 1983.

2-21 Fleming, K. N., "A Reliability Model for Common Mode Failures in Redundant Safety Systems," Proceedings of the Sixth Annual i Pittsburgh Conference on Modeling and Simulations, April 1975.

t 2-22. Pickard, Lowe and Garrick, Inc., "Classification and Analysis of i Reactor Operating Experience Involving Dependent Events," prepared for Electric Power Research Institute, Interim Report, PLG-0400, NP-3967, June 1985.  !

O. 2-23. Pickard, Lowe and Garrick, Inc., "RTIME2 Computer Code Users I

Manual," PLG-0439, September 1985.

t l

t i

L 1  :

I 1

I I

t 1

~

0154G061886DAR

O TABLE 2-1. EXAMPLE OF THE CALCULATION OF THE EFFECTIVE NUl!BER OF FAILURES FOR VARIOUS SYSTEM IMPACT CATEGORIES Dependent Events System Independent Effective Impact Events Event 1 Event 2 Event 3 Number of C ategory* ,, _, , , ,

Failurestt p n' p n p n p n 1 1.0 71.8 0.0 0 0.9 2 0.9 1.1 By = 74.59 2 0.0 0 0.05 2 0.1 2 0.1 1.1 62 = 0.41 3 0.0 0 0.95 3 0.0 0 0.0 0 5= 2.85 3

  • Refers to the number of components affected in the event.

9

    • Weight assigned to various hypotheses.

t Effective number of component failures for each event for each hypothesis (Equation 2.98).

17T otal effective number of component failures for each system impact category (Equation 2.99).

O' 0155G060385 2-40

l I

i l

0.35  !

0.25  !

0.20 0.15 i'

0.05 l

A A A A A 1 2 3 4 5 i FIGURE 2-1. POPULATION VARIABILITY l OF THE FAILURE RATE  !

l

. \

-I h

O 1

{

l 1

D pg= PROBABILITY (h;)

P1 P3... Pn FIGURE 2-2. STATE-0F-KNOWLEDGE DISTRIBUTION OVER THE SET OF FREQUENCY DISTRIBUTIONS O

2-41

0 003 - 9 27

[ Ow2 -

i 0 001 -

( .

5 e -

FIGURE 2-3. POSTERIOR DISTRIBUTION FOR THE PARAMETERS OF THE DISTRIBUTION OF PUMPS' FAILURE TO START ON DEMAND RATES O

  1. I 'I8I P;(X$1X;)

i A

A l

FIGURE 2-4. THE RELATION BETWEEN THE POPULATION VARIABILITY CURVE AND UNCERTAINTY ABOUT INDIVIDUAL ESTIMATES 9

2 42

3.- TMI PLANT-SPECIFIC DATA BASE O

3.1 INTRODUCTION

l A comprehensive and well-documented summary of the THI-1 operating experience provides the cornerstone for the Bayesian analysis of the data j developed for this study. This section explains the process of i plant-specific data collection in the areas of (1) component failures, (2) common cause events, (3) component maintenance, and (4) initiating

, events. It also provides the sumary of collected data as well as the '

resulting updated distributions for each of the above four categories.  ;

Detailed listings of the plant-specific data are provided in a series of ,

four appendices at the end of this report. 7 The TMI-l plant started up on September 2, 1974. The data collection effort covered the period from the date of the beginning of commercial operation through June 30, 1984, even though the plant has been shut down  ;

since February 17, 1979. Table 3-1 lists the primary documents and operating records used during the plant data collection task. Each of ,

the documents and sources of information are described in more detail -

under the corresponding topic in the following sections. For component  !

f ailure data, the entire period from September 2,1974, through June 30, 1984, was covered. For reasons explained later, the maintenance i data had to be limited to the noncold shutdown periods. Table 3-2  ;

provides the history of TMI-1 cold shutdown outages including reasons.for j such shutdowns. The initiating events data were collected for the period September 2, 1974, through February 17, 1979.

3.2 COMPONENT FAILURE RATES i

1 Quite simply, the collection of plant-specific failure rate data requires 4

the analyst to count and record, for each component and failure mode  :

being modeled, the number of failures and the corresponding number of  !

demands or operating hours. Unfortunately, these data are virtually  !

never found together in the same plant records. The solution to this

}

problem demands a judicious accounting for each type of data to ensure  ;

accurate and consistent failure rates. i 3.2.1 COMPONENT FAILURE DATA 3.2.1.1 Definition of Failure ~

{

The data presented in this section are used in the system analyses to quantify the frequency of hardware failures that prevent a system from  ;

meeting success criteria defined by the event tree molels. Several failure causes are evaluated in each analysis, and all causes are combined to determine overall system unavailability for each set of success criteria. The failure rate data must be comparable with these ,

applications. The failures must include all events that functionally disable a component for the failure mode being evaluated. This ensures completeness of the failure rate data base. However, failures due to  ;

causes other than internal equipment malfunctions must be closely 3-1 -

i Olb8G061886DAR

examined to determine whether they are evaluated separately in the system model. This avoids potential double-accounting for failures in the g system analysis results. W For example, if a normally closed motor-operated valve must open and remain open for a system to operate properly, equipment failures that prevent the valve from closing after it opens are not relevant for the valve data base. However, these failures are relevant data for a valve that must reclose. If one of the failures occurred because test personel lef t a circuit breaker open, and if this failure cause is evaluated separately in the valve model, the event applies to the evaluation of test personnel error rates. It is not included in the valve hardware failure rate data because doing so would double-account for the test personnel errors. However, if testing errors are not evaluated separately in the valve model, the event is included as a valve hardware failure to ensure complete accounting for all the plant-specific evidence.

A component failure is thus an event in which a piece of equipment fails to perform a function required by the system model. The event is included in the hardware failure rate data base if its cause is not evaluated explicitly in a separate part of the model. If the cause is quantified separately, the event is used as evidence for the appropriate failure cause.

The equipment operating records document a large number of component malfunctions. Many are clearly component failures that should be included in the data base. For example, a motor-operated valve may fail to close because of loose limit switch contacts, or a pump may fail to start because its circuit breaker closing coil has burned out. However, a large number of events documented as malfunctions require additional investigation and subjective evaluation to determine whether they should be included as functional failures. For example, pump shaf t vibration is indicative of possible damage to the pump bearings. Severe vibration is normally included in the data base as functional failure of the pump during operation because shaft seizure or other failures will occur within a few hours if the pump remains running. Observation of minor vibration or bearing noise may be the reason for pump inspection, additional lubrication, or corrective maintenance. These minor problems are sometimes considered as failure "precursors" because they will eventually progress to pump damage if left unattended. However, they are not normally included in the failure data base because the pump will continue to run for several hours or days before experiencing severe damage. Additional information about the types of repairs made, the parts replaced, and the urgency of the repairs often provides importam insight about the severity of these malfunctions. Preventive and corrective maintenance is performed to stop the progression of minor problems, and only the events that actually cause equipment damage are correctly included as failures. The effects on component availability from inspections, preventive maintenance, and minor repairs are included in the maintenance data base described in Sectian 3.4.

The first step in collecting failure data for a component is to determine the failure modes and failure causes to be included in the data base.

3-2 Olb8G061886DAR

t These are defined by the success criteria and types of analyses performed '

for the system models. . The second step is to include only those O malfunctions that cause functional component failure for each failure mode being evaluated. This requires close examination of the plant I

records, discussions with cognizant operating and maintenance personnel, and experienced interpretation of the functional severity of "borderline"

. events. '

i 3.2.1.2 Failure Data Sources Several sources of information were consulted for collecting equipment failure data. All major equipment malfunctions are documented on either a "Work Request" or "Job Ticket" form. Combined together, there are i about 40,000 work requests and job tickets covering the period  !

September 2, 1974, tnrough June 30, 1984. Approximately 21,000 such  !

records are on computer for the period 1977 through June 1984. About ,

19.000 work requests dated prior to 1977 are recorded in a logbook by  !

date of issue and work request number. The majority of logbook entries t include a few words about the nature of the problem and the component '

designation for the components involved. To obtain more detailed ,

information, the original work requests recorded on microfilms had to be i reviewed. ~

A work request or job ticket is written whenever significant maintenance .

is required on any piece of mechanical or electrical equipment. Minor ,

adjustments may be made without work requests, but all work requiring i equipment disassembly, repair, or replacement is documented on a work '

O request or job ticket form. The work requests thus provide a complete history of all significant adjustments, repairs, and replacements of THI-l mechanical and electrical components. They are less complete as a source for documenting electronic and control equipment malfunctions. ,

Each work request identifies the specific component affected, the ,

observed problem, and the desired maintenance activities. Work requests and job tickets are written for both "safety-related" and "nonsafety-related" equipment, and they are written during all plant operating modes. They are, in a sense, the most "pure" and complete j

documentation of component malfunctions available. Since each work request or job ticket is assigned a unique identification number, they  :

also afford easy traceability for all failures recorded in the data base. l l

Use of the work requests and job tickets to collect equipment failures 1 greatly simplifies the collection of compatible component operating time and demand data. Work requests are written for all component malfunctions, regardless of the plant or equipment operating conditions when the problem is observed. The data analyst can, therefore, include all the equipment operating experience as relevant "success" data for the failure rate calculation. Use of more restrictive failure records, such as failures reported only during periodic testing, or failures while the reactor is at power would have required a corresponding reduction of the experience base to provide consistent failure rate data. By restricting the data base, the analyst is also forced to subjectively assess each l O piece of demand and operating time data to determine its applicability to the limited failure experience. j 1

3-3 Ulb8GU618860AR

Another important advantage afforded by the work requests and job tickets is that they provide a record of the component malfunction and the &

corresponding repairs on a single form. In many cases, both types of W information are required for the analyst and plant personnel to determine the relative severity and functional effects of a "borderline" malfunction.

Two drawbacks of the work requests are the large volume of records that must be reviewed and the lack of detail in some of the malfunction and repair descriptions. Each work request form had to be examined first to determine if it applied to a component being modeled in the study. If it did, a more thorough review was performed to determine the exact failure mode and any available information about the cause of failure. Several hundred work requests were actually found to be relevant for the failure data base. Because descriptions of the malfunctions and repairs are often quite brief and abbreviated, it was occasionally difficult for even experienced plant operations and maintenance personnel to reconstruct a specific malfunction. Unless the event could be conclusively discounted as not degrading equipment performance, it was retained in the data base as a functional failure or the component.

All component failures collected for the Tal-1 plant-specific data base -

are documented on the data sheets in Appendix A. Each data sheet includes the specific component affected, the observed failure mode, a brief description of the failure cause, the date of the failure, and the corresponding work request or job ticket identification number.

3.2.2 COMPUNENT DEMAiiDS AND OPERATING HOURS Une of the most difficult tasks in the development of a comprehensive data base is to ensure that the failure events and the successes have been derived from compatible data sources. The documents reviewed for component failure events do not contain any information about the corresponding component success data. Therefore, other documents were used for information about component demands and operating hours.

3.2.2.1 Demand Data Sources The two most important sources for TMl-1 component demand data are the periodic test procedures and the plant operating procedures. Since the work requests and job tickets provide information about component failures during all modes of plant operation, it was not necessary to restrict the demand data to tests or operations performed only during certain plant conditions. Therefore, the TMI-l success data include information obtained from all modes of operation between September 2, 1974, and June 30, 1984.

Table 3-2 summarizes the cold shutdown outages for the TMI-l plant. This information is important for the development of component demand data from the periodic test reports, because many of the testing schedules change when the reactor is placed in cold shutdown.

Many of the periodic test procedures are also used to verify redundant '

equipment operability during maintenance and to verify repaired equipment '

3-4 0158G0618860AR

operability after maintenance. The component maintenance r'ecords i (N described in Section 3.4 and the operability testing requirements were V

used to estimate the number of additional performances of each test for maintenance outages.

Of the 134 test procedures reviewed in detail for the study, 64 provided '

information about relevant mechanical and electrical equipment operations for the component failure rate data base. Table 3-3 lists these tests by number and summarizes the nunber of performances for each test during the- >

data base period. '

For many of the failure moder, in the data base (e.g., failure of motor-driven pump to start on demand), specific operations performed during a test provide direct evidence of component response; e.g., start '

pump X. For a large number of failure modes, however, no analagous specific operations are included to directly verify successful component i performance. In many cases, observed flow rates, pressures, temperatures, or levels can be used as evidence that components have operated suv.essfully or are aligned in their normal positions. As an ,

example of th!s method of test data synthesis, consider the component  ;

failure mode "nenual valve transfers closed". Referring to the example ,

in Figure 3-1, a system flow test is performed to verify the operability ,

of pump X by closing motor-operated valve A and running the pump on recirculation flow. Successful performance of the test requires that the pump start and that adequate flow is observed at flow gauge F. This test, in addition to verifying pump X operability to start and run, '

provides the following test data:

e Motor-operated valve A closes and reopens on demand.

e Tne piping from the suction source through the recirculation line is not plugged. ,

e Check valve C opens successfully.

e Manual valves M1, M2, and M3 are open.

The test does not provide any information about the status of the flow  !

path through motor-operated valve A and manual valve M4. The test would l detect failures to close valve A if a flow path were available downstream' l from valve A, if valve M4 had not failed in the closed position, and if the leakage through valve A were sufficient to degrade the measured flow at gauge F. However, the internal status of valve M4 cannot be determined from the test. If the valve disc had separated from the valve stem, the flow path would be blocked, but since motor-operated valve A is closed throughout the test, this failure mode cannot be detected during normal test conditions. Therefore, the test is not included in the success data for valve M4. This general process was applied to all systems tested during the periodic tests to develop additional information about the demands on, and the operating status of, components not specifically addressed in the periodic test performance steps.

Reactor plant and turbine unit startups and shutdowns also provide an O important source of component demand data. The TMI-l operating 3-b 0158G0618860AR

procedures were reviewed to identify equipment routinely cycled during these evolutions. Operating records from the monthly reports document all reactor plant and turbine unit power changes. The demands from routine plant operations were added to the test nerformances so complete the demand data base fur each component.

3.2.2.2 Uperating Hours Data Sources The operating hours for several large motor-driven components at TMI-1 are provided by Operations Surveillance UPS-594 which has monthly readings of run-time meters for loads on buses D and E. These run time meters provided a source of information on operating hours for most of the large motor-driven pumps and ventilation units in the plant-specific data base.

To calculate plant-specific failure rates for component failure modes like "valve transfers closed" or "heat exchanger plugs during operation,"

the data analyst needs the corresponding number of component operating hours in the unfailed state. For example, if one spurious closure of a motor-operated valve had been experienced in the plant valve population, the data analyst would need to know how many valves were included in the population and, for each valve, how many hours it had remained open.

Detailed success data for these generally passive component failure modes are not directly available from any plant records and are extremely difficult d estimate. The TMI-l plant-specific data were obtained from a detailed analysis of the normal plant operating procedures and paractices and from a review of the periodic test records.

A normal flow path alignment was identified for each system in the plant model. Operation of the system in this alignment verfies that all the associated valves, pfuing, and heat exchangers are open and functioning properly. The plant power operating records and the equipment run time meter logs were used to determine the total number of successful operating hours for each component in the flow path. For example, normal operation of a system might provide continuous flow through a series of three motor-operated valves and a heat exchanger. If the run time meter records indicated that the system had been operated for 1,000 hours0 days <br />0 hours <br />0 weeks <br />0 months <br />, the evidence for the plant data base was 3,000 open valve hours and 1,000 open heat exchanger hours. This accounting process was used to estimate all the passive component operating success hours for normally running systems and for systems operated in a specific alignment during certain plant modes; e.g., the decay heat removal system.

Some systems, such as the reactor building spray system, are operated only during periodic tests but remain unchanged between tests. Each of these systems was also analyzed for component success data. Although these systems are normally in standby, the periodic tests verify the status of several of their components. This testing provides sufficient information to include these components in the success data base if their status remains the same between tests. For example, if a normally open valve was verified to be open because of the flow path established during a test, and if the valve remained open between tests, the total number of hours between the tests was included in the success data for the valve.

3-6 Olb8G0618860AR

r if the valve failed in the closed position during or between the tests,

, its failure would have been discovered during the periodic test and would  ;

-\ have been documented in a work request. Therefore, these standby system component hours were also added to the success data for normally operating systems. l 3.2.3. UPDATEU COMPUNENT FAILURE RATE DISTRIBUi10N  !

As described earlier, the TMl-1 component failure rate distributions were developed by combining two pieces of information; namely, the generic e distributions described in Section 2.1 and the plant-specific failure [

data presented in the previous section as well as Appendix A. The

  • updating process was based on the methods discussed in Section 2 and, in particular, Equation (2.59). The computer code BEST4 (Reference 3-1) was ,

used to perform the calculations. Basic characteristics of the resulting i distributions are listed in Table 3-4. Also presented in the table ara  !

the plant-specific failure and success data for each component, as well [

as,the mean values of the generic distributions. In cases where no  ;

plant-specific data were collected, no updating was performed and the i listed distributions are generic.

3.3 COMMON CAUSE FAILURE PARAKTERS *!

Several common cause events were identified in the course of the c.omponent failure data collection task, as described in Section 3.2. t These events are summarized in Appendix B. Appendix B also provides detailed tables of the screened generic common caused events used to  !

quantify the common cause parameters for this study (see discussion in j l

Section 2.2; especially, Table 2.1). The coamon cause parameter j distributions listed in Table 3-5 were developed by combining the genaric '

and plant-specific data using the approach discussed in Section 2.2. The computations were performed with the aid of the computer code BETA l (Reference 3-2). '

3.4 CUMPONENT MAINTENANCE DATA l i i 3.4.1 DEFINITION OF MAINTENANCE l

. In this study, component maintenance includes much more than unscheduled repairs of equipment failures. A component is considered to be out of '

service for maintenance whenever it is disabled for special inspections, routine preventive maintenance, scheduled overhaul, modifications, replacement, or repairs. The f.equency of these maintenance events is substantially higher than most component failure rates, because the failure rates include only those events that cause severe functional damage to the equipment. The maintenance event durations also depend on the type of activity. The effect.ive "mean time to repair" a component may be only weakly correlated to actual maintenance personnel hours spent repairing failures.

The only maintenance events included in the data base are those that

remove a component from service in a manner that prevents it from performing the function analyzed in its system model. For example, a 3-7 Olb8dO61886DAR

normally closed motor-operated valve may be required to open automatically and remain open for successful system response to an initiating event. Any activity that removes the valve from service in the closed position and prevents it from opening is counted as a g

maintenance event affecting the availability of the valve. If, however, the valve was deenergized in the open position while the mGtor was replaced, the event would not be included in the maintenance data base because it did not prevent the valve from satisfying its required function.

The duration of a maintenance event lasts from the time a component is tagged out of service to the time it is returned to service in an operable conaition. In many cases, this period is only weakly correlated to the actuci number of maintenance personnel hours spent working on the equipment. Factors which make the observed event duration longer than the actual repair time include time for the plant operotors to align the system for maintenance and to realign it after the work is completed, time to perform required operability testing, delays for spare parts, time for coffee breaks and meals, and overnight or weckend periods when maintenance personnel may not be scheduled to work, and where no conflict with technical specifications develops.

3.4.2 MAlfiTEtiANCE DATA SOURCES The most accurate information about component unavailabilities due to maintenance at TMI-l is found in the "Switching and Tagging Orders." The plant-specific maintenance data are derived from a review of all switching and tagging orders written between September 1974 and February 1979.

When a component is realigned in an abnormal configuration, tags are posted locally and in the control room to inform personnel of the equipment status. These tags lenote the repositioned valves, disconnected circuit breakers, special control switch positions, and other actinns required to isolate a component from its system so that personnel may work on it safely. The equipment is considered to be functionally disabled whenever the tags are in place, and it is available for normal service when .'.he tags are removed.

The saitching and tagging orders contain all the information necessary to determine the effect of each maintenance event on functional component availability. A brief description of the reason for maintenance and the specified position for each component realigned for the job are included on each form. The orders also contain signature blanks for work authorization and job clearance. The dates and times of these signatures denote the full period that the component was unavailable for service.

All switching and tagging orders were screened according to the TMI-1 cold shutdown outage periods listed in Table 3-2. Equipment removed from service during cold shutdown was not included in the data base because the nature and duration of these maintenance events do not represent maintenance performed during the operating periods of interest in this study. Some maintenance events were initiated before the reactor was g 3-8 Ulb8G061886DAR

4 placed in cold shutdown or extended beyond the rsactor startup time.

p These were included in the data base for the duration of time above cold Q shutdown, because they affected component availability during the ,

pertinent reactor operating modes. The TMI-1 plant-specific maintenance t data are summarized in Appendix C which provide information about the date of each maintenance event, the corresponding tag order, componert affected, and work performed and its duration.

3.4.3 UPDATED COMPONENT MAINTENANCE DISTRIBUTION As discussed in Section 2, component maintenance unavailability distribution for each component is developed as the distribution of the product of maintenance frequency and maintenance duration for that component. Therefore, separate updated distributions were develuped for component maintenance frequencies and durations based on the generic aistributions discussed in Section 2.3 and plant-specific data presented in Appendix C.

3.4.3.1 Updated Maintenance Frequency Distribution _s, Table 3-6 provides basic characteristics of the updated maintenance frequency distributions. These distributions were developed with the aid of mode 4 of the computer code BEST4 (Reference 3-1), which was used to perform the updating calculations based on Equation (2.59). Table 3-6 also lists the plant-specific data in the form of tu number of maintenance events and the total number of component hours for each '

component.

3.4.3.2 Updated Maintenance Mean Duration DistriMtions Maintenance duration distributions were developed based on the procedure described in Section 2.3.2. Each distribution was developed by establishing a prior distribution for the two parameters of the assumed lognormally distributed maintenance durations based on generic information. This prior parameter grid was then updated using the i plant-soecific data with the aid of RTIM[2 computer mde (Referenc.e 3-3).

The discributions listed in Table 3-7 are the resulting uncertaisi:y distributions of the mean maintenance durations. A total of 48 n,ean

  • maintenance durations was developed for 47 components. The table shows '

that, for NSRW pumps, two distributions were developed: one for short-duration maintenance activities and another for long-duration '

maintenance. This was done to acknowledge that for that component, a single unimodal distribution, such as lognormal, could not represent the data.

3.S INITIATING EVENT GROUP FREQUENCIES Table 3-8 lists the internal ini ,ating event groups chosen for the TMI-1 plant analysis. Each group contains one or more specific initiating events believed to result in the same general plant response as modeled in the plant event trees.

n Details of the selection and grouping of initiators are provided in '

V Section 2 of the Plant Model Report. As was discussed in Section 2.5, 3-9 i Olb8G061886DAR

the internal initiating event groups are divided into two general sets.

The first set is composed of those events for which the available data trom other nucledr power piants are judged to be relevant. This group includes all initiating event groups except groups b, 13,14,18, and

19. The available generic and plant-specific data were input to the Bayesian data aaalysis process to generate a frequency distribution for each of the initiating event groups.

For the second set the data from other power plants could not be used.

This is due to the fact that the systems involved have designs and speci-fications unique to the TMI-1 plant, which require a plant specific analysis. For this reason, the frequency of groups 5,13,14,18, and 19 were quantified by analyzing the corresponding systems and scenarios.

The following sections describe how the frequency of each category was developed.

3.b.1 EVENTS QUANTIFIED, BASED ON GENERIC AND PLANT-SPECIFIC DATA The generic frequency distributions were developed, based on the industry experience with PWRs in general and B&W plants in particular. The loss of off site power data base covers the PWR as well as BWR experience.

Details of how these generic frequencies' are developed are provided in '

Heference 3-4. The mean values of the generic distributions are given in Table 3-8.

Among the sources for the generic plant population event data for initiating events was the Electric Power Research Institute study of pressurized water reactor transients (Ref erence 3-6). The study summarizes the events initiating forced shutdowns at 36 PWR units from their initial year of operation until January 1981.

The 41 initiating event data categories listed in the EPRI study for PWRs were reviewed and 27 of those categories were summed for 4 of the event groups chosen for this analysis as shown in Table 3-9. To the extent possible, the data from the EPRI study was carefully examined. In some cases, incidents not included in the EPRI study were added to the data base, while in other cases, further consultation with other sources resulted in removal of incidents from the data base. The final result of this process is given in Reference 3-4, where tha plant population data for these initiating event categories are listea for each of the 36 PWR units included in the data base.

The EPRI study, because it was done for ATWS analysis, does not provide any data for causes of losses of RCS inventory (groups 1, 2, 3, and 4) or for steam line breaks (groups 6 and 7). Tnese initiators include events which are either the direct result of pipe failures or which have the same effect on plant response as would a pipe failure. Catastrophic pipe failures have occurred in a variety of industrial and nonnuclear power generation facilities. However, the types of piping involved and their operating temperature, pressure, and flow conditions are generally quite different from those of interest in these initiating events. The industry experience adds to the general understanding of pipe failure phenomena, but evaluation of its direct applicability to this data base 3-10 015BG0518860AR

t requires a much more detailed comparative engineering and design analysis than was possible within the scope of this study.

u Several sources of nuclear industry data including Nuclear Power Experience (Reference 3-6) were consulted to obtain plant population data for these initiators. No events applicable to categories 1 and 2 were reported for PWRs. However, the investigation provided information about several events which were judged to be applicable to the data base for groups 3, 4, and 7. These events and the resulting plant population data' are described in Reference 3-4.

In the case of large and medium LOCAs (groups 1 and 2) it was judged that there is little, if any, plant-to-plant variation in the frequency -since the primary piping systems are essentially designed according to the same codes and manufactured based on similar standards. Moreover, these piping systems are not affected as much by the variation of operating  ;

rractices among plants as are other components and systems. Therefore, '

it was decided to use the cumulative experience at U.S. PWRs (zero events in 428 reactor years) as evidence in a one-stage Bayesian updating process. '

The loss of offsite power initiating event (group 7) data was based on extensive review of the history of losses of offsite power at all nuclear >

power plaats in the United States, the details of which are reported in t Reference 3-4.

For the loss of air systcms initiating event (grcup 13), the plant population data were obtained from review of N ar Power  :

% ~

Experience (HPE) (Reference 3-6) for the perio. a70 through 1985. No  !

evidence of total loss of air system was frM ir this review.

Furthermore, it was observed that air contaunation has always only resulted in isolated component failures without any significant impact on the operation of the air system, as a whole, or on the plant operation.

For steam generator tube rupture (group 8), loss of ATA power (group 15),

and loss of one DC power train (group 16), NPE was reviewed and several events were judged to be applicable to the data base.

In the case of steam generator tube rupture initiating event, the review of the industry experience did not reveal any major tube rupture events <

that did not result in an automatic plant trip. i The plant population data for the excessive feedwater flow (group 9) is -

based on the review of the operating history of B&W plants (Reference 3-7).

For TMI-1 plant-specific initiating ever.ts, data was collected by  !

reviewing plant records, such as monthly operating reports and weekly ,

reports for the period of September 2, 1974, through February 19, 1979. .

The HRC Gray Books were also reviewed for the same period.

Appendix 0 summarized the collected data for each of the quantified '

initiating eve.7t categories, based on generic and plant-specific data.

3-11 -

Olb860618860AR

The statistical information is summarized in Table 3-8. Bayesian calculations were done using computer code BEST4 (Reference 3-1). The table also presents the initiating event i frequency distributions resulting from updating the generic distributions with plant-specific data.

3. b. 2 INIIIATING EVENTS WHOSE FREQUENCIES WERE QUANTIFIED BY PLANT-SPECIFIC ANALYSIS 3.5.2.1 Loss of River Water The frequency of loss of river water system is believed to be dominated by external causes that prevent flow of river water to pump intake. The blockage of river water flow has in fact happened in the past at the TMI site. The most severe occurrence happened in February 1979 when, following a heavy rainfall, debris was washed down the river in large quantities and resulted in plugging of the intake screens for S hours.

Unit 2 was operating at the time, and Unit I was shut down for refueling. The Unic 2 intake screens were the only ones plugged during this event. The ficw of river water to the intake stucture was virtually reduced to zero. The source of water for the 6-hour period before the screens were finally cleaned was the water already in the pump house when the blockage occurred.

Events of this type form the basis for the calculatian of the frequency of the loss of river water initiating event (4RW). This frequency is calculated based on the site-specific data for the frequency (fsp) of severe plugging of the intake screens (one event in 12 site-years) and the chance of recovery, given plugging of the screens.

The time available tu unplug the screens depends on the volume of water available in the intake structure (assuming no water flowing from the river) and the number of pumps operating.

The volume of available water ranges between 1.4 x 105 to 4.1 x 105 cubic feet, depending on the river water level, ranging from a normal level at 278 feet to a high level of 303.5 feet. Assuming two river water pumps operating at full capacity (7,250 gpm), the time available for recovery action ranges from 1.3 to 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br />. For one pump operating, this time is doubled.

The following probability distribution is assessed for the frequency of failure of the recovery action HRE4.

Mean: 1.78 x 10-1 Sth Percentile: 3.02 x 10-2 50th Percentile: 3.4 5 x 10-2 95th Percentile: 9.86 x 10-1 For a discussion of the derivation of the duration of HRE4, see the TMI-1 Human Actions Analysis Report.

3-12 01586, '

J860AR

A distribution was developed for fs >

evidence of one event in 12 years. pThe based basic on acharacteristics uniform prior and of this ,

distribution are: ,

Mean: 8.43 x 10-2 5th Percentile: 9.06 x 10-3  ;

both Percentile: 6.12 x 10-2 -

95th Percentile: 1,74 x 10-1 The frequency of the loss of river water is then calculated from '

I

&RW

  • fSP . HRE4 i The main characteristics of the distribution of QRW are listed in -

Table 3-8.

3.5.2.2 Loss of Nuclear Services Closed Cooling Water For detailed discussion of the calculation of the frequency of this .!

initiator, see Section 4 of the Systems Analysis Report. Table 3-8 lists i the basic characteristics of the distribution.

3.5.2.3 Loss of Control Building Ventilation l For detailed discussion of the calculation of the frequency of loss of C.]

) control building ventilation, see Section 6 of the Systems Analysis f Report. Basic characteristics of the distribution are provided in l Table 3-8. i 3.b.2.4 Inadvertent Opening of DHR Valves (V-Sequence)

The following 5.vo scenarios are consi'ared to be the most likely ways of an outside containment LOCA.

e Scenario 1 Failure of two series check valves in the cold leg injection lines of LPI/DHR, followed by failure of lower pressure downstream piping. ,

e Scenario 2. Failure of three normally closed series motor-operated valves in the hot leg suction path of DHR, followed by failure of lower pressure downstream piping.  !

3.6.2.4.1 Scenario 1 (LPI Cold Leg Injection) l Figure 3-2 shows the simplified LPI cold leg injection path arrangement l of the LPI/DHR system at TMI-1.

In general, the frequency of failure for two valves, Vi and V2 s in series (V1 is assumed to be nearest to the RCS) can be expressed as l As " A(V1 ) . P(V2lV1 ) + A(V2 )

  • P(V1lV2) (3.1) 3-13 0158G062086DAR

- _ . . - - ._.. - ... , . . - , . ~ , . .

where As = the frequency of failure of both series valves.

A(VI ) = the frequency of random, independent failure of valve V1 .

P(VlV)=theconditionallikelihoodthatV2 2 1 is failed, given that Vi fails.

A(V2 ) = the frequency of random, independent failure of V2 (events per hour).

P(V1lV2) = the conditional probability that V1 is failed, given that V2 fails.

P(V2lV1) and P(V ilVg) are composed of both random, independent, and demand type failures of the second valve.

In some cases, the random, independent failure frequencies and conditional probabilities for the two valves will be approximately equal, but in other cases, they will not. For example, if V1 leaks slightly but V2 does not, V2 would be exposed to the differential pressure loading to which V1 is normally exposed. In this situation, V1 would have RCS pressure on both sides of the disc and would be expected to have a lower failure rate than V 2, which is exposed to a greater differential pressure. Thus, Equation (3.1) could be written as As

  • A(V1) P(V2 !v1 ) * (1-Pg ) + A'(Vy ) . P'(V2lV1 ) . P g

+ A(V2 ) . P(Vy l 2V ) . (1-Pg ) + A'(V2 )

  • P'(V i l V2 ) . P g ( 3. 2 )

where P[ = the probability that the space between valves is pressurized to RCS pressure.

A'(VI ) = the frequency of a random, independent failure of V1, given that the space between valves is pressurized (events per hour).

P'(V2 lV1 ) = the conditional probability that V2 fails, given that Vi has failed and the space between valves is pressurized.

A'(V2) = the frequency of a random, independent failure of V2, given that the space between valves is pressurized.

P'(Vi lV2 ) = the conditional probability that Vi f ails, given that V2 has failed and the space between valves is pressurized.

O 3-14 0158G0620860AR

-- . _ _ _ . - - . . - - - . ~. .- . - -

On the basis of the loadings across the valve discs, the following assumptions appear to be reasonable for the lines that contain the check valves.

{J} ,

l. A'(V2 ) " A(V1 )-

1

2. A'(V1) is small compared to A(V1 ).
3. A(V2 ) is small compared to A'(V2 )- '
4. P'(VllV2 )
  • P(V2lV1 )- ,

Substituting for A'(V2) and P'(VilV2 )

A3 = A(Vt ) . P(V2 Iv 1 ) . (1-Pg ) + X'(V1 ) . P'(V2 !v1 ) . P g (3.3)

+ A(V2 )

  • P(V y lV 2 ) * (1-Pg ) + A(V 1 ) . P(V 2 Iv l ) . Pg or '

A3 = A(V1 ) . P(V2 !v l )

  • A'(vi ) . P'(V2Iv1 ) P g (3.4) ,

+ A(V2 ) . P(V1 lV 2 ) . (1-P g) ,

Tne third term in Equation (3.4) is small compared to the first; therefore As " X(vl ) . P(V 2 Iv l )

  • A'(vl ) . P'(V2Ivl ) . P g (3.5)

As a conservative upper bound, it can be argued that As " A(V1 ) . P(V2lV1 ) * (1+PI ) (3.6) l Because only a minute amount of leakage is required to pressurize the space between valves, it is assumed that PI approaches 1.0. Therefore As = 2 . A(V1 ) . P(V2lV1 ) ( 3. 7 )  !

Given that Vi has failed independently, V2 could fail upon demand (due to the sudden pressure challenge), or it may fail randomly in time, i sometime after failure of V1 . The latter failure. mode is represented by the standby redundant system model. Equation (3.7) conservatively reflects the potential for discovery of the outboard valve rupture before the next testing opportunity of the inboard valve because of the ability to alarm and indicate this condition to the operator via the accumulator l pressure sensors.

The term P(V2 lV1 ) in Equation (3.7) contains two components: one l representing random failures of the second valve, given that the first  !

valve has failed, and the second representing a demand failure at the time the first valve failed.

O G  :

3-15 0158G062386DAR i

The determination of the frequency of occurrence of random failures is facilitated by assuming that the two series check valves in each path represent a standby redundant system, and failure of the downstream check valve cannot occur until failure of the check valve nearest to the reactor coolant system loop has occurred. The probability of random failure (unreliability) for a single injection path is given by Qpath a 1. e-A t (1 + At ) (3.8) where A is the appropriate failure rate of a single check valve. This expression was then used to derive a failure (or hazard) rate for the path. That is,

-1 A

path (t) = (1 E1 - O ath 3 ( '9) or A

path (t) = L (3.10)

At The plant is expected to go to cold shutdown at least once every 1.5 years at which time these valves will be inspected. If it is determined that the system is not functioning, it is repaired at that time. Therefore, the time-dependent failure rate is bounded at 1.5 years. The average f ailure rate over a time period, T, is given by T

  1. A 1 Adt path per reactor year # *T o

1, L)

At (3.11)

=h[AT-f.n(1+AT))

When AT << 1, this result can be expanded to obtain

<A

  • path (' }

The demand component of the path failure frequency is merely the product of A and the demand failure rate, Ad. Thus,

<A path >

  • A E +A]d (3.13; Finally, the above expression for '-^ sth> is multiplied by a factor of 2 to account for the logic used is developing Equation (3.7). This logic is that the two valves can fail in either sequence because of an assumed high likelihood of inboard valve leakage and pressurization of O

3-16 Olb8G0618860AR

~ -- __ - _- - -. - ._ . _ _ _ _ _ - _ _ _ _ _

the space between valves. Thus, the final expression for the series valves in the injection lines is

~h (O [T+A] t path > " d (3.14)

As an upoer bound, the check valve fail to operate on demand, V7F, will be used for Ad . For A, it was assumed that even though the disc  ;

rupture mode of failure is extremely unlikely, it is implicitly included- ~

in the check valve leak data. The following distribution was developed ,

for the disc rupture mode of failure (V7R) based on extensive review of  ;

PWR, ECCS, and RPS check valve leakage data.

i k

95th Percentile 3.2 x 10-8 '

Mean 8.3 x 10-9 i, Median 2.3 x 10-9  !

Sth Parcentile 1.6 x 10-10 ,f Therefore

<gpn h> = 2 . V7R [Y7N + V7F]

7 2 since there are two injection paths, then the annual frequency of i scenario 1 is j

$ 1 = 2 <A pa th>

A point estimate for $1, using mean values for V7R and V7F and T = 1.5 years is 7.8 x 10-8/ year t

3.5.2.4.2 Scenario 2 (DHR Hot Leg Suction Line)  !

This scenario involves failure of three series MOVs, DH-V1, DH-V2, and l UH-V3. Given that such failures occur, the low pressure-piping and the RHR system components downstream of these MOVs would be exposed to RCS pressure. l The sequential failure of all three values due to random causes is judged i to be very unlikely. The frequency of this scenario, &2, is then  :

calculated based on assuming (1) sequential failure of VI and V2 followed )

by failure of V3 due to failure of VI and V2, and (2) rupture of VI disc, followed by failure of V2 due to V1 failure and finally failure of V3 due to V2 failure. Therefore

$2

  • 2A ( )Ad*AAd V

3-17 0158G062036DAR l

. . __ . . _ . .._ _ ____.,__.__..J

In this case, we use V1F and V7R for Ad and A, respectively.

Therefore 42 = 2 . V7R V1F(V7R T) + V7R . (V1F)2 A

i spoint 42 = 9. estimate 3 x 10- g/ year.ing mean values for V7R and V1F and with T = 1.5 years Finally, the annual frequency of the V-sequence is given by

& = 41 + $ 2 The distribution of 4 is provided in Table 3-8.

3.5.2.5 Loss of Air System For detailed discussion of the calculation of the frequency of this initiator, see Section 18 of the Systems Analysis Report. Table 3-8 lists the basic characteristics of the distribution.

~

3.6 REFERENCES

3-1. Pickard, Lowe and Garrick, Inc., "Bayesian Estimation Computer Code 4 (BEST4) Users Manual," PLG-0460, December 1985.

3-2. Pickard, Lowe and Garrick, Inc., "BETA Computer Code Users Manual,"

PLG-0389, January 1985.

3-3. Pickard, Lowe and Garrick, Inc., "RTIME2 (Repair Time) Computer

' ode Users Manual," PLG-0439, December 1985.

3-4 Pickard, Lowe and Garrick, Inc., "PRA Proprietary Data."

3-5. Electric Power Research Institute, "ATWS: A Reappraisal, Part III, Frequency of Anticipated Transients," EPRI NP-2230, January 1982.

3-6 Nuclear Power Experience, published by S. M. Stoller Corporation, updated monthly.

3-7. B&W Owners Group Probabilistic Evaluation of Pressurized Thermal Shock - Phase 1 Report - BAW-1791, Babcock & Wilcox, June 1983.

9 3-18 0158G121786DAR

O O O TABLE 3-1. TMI-1 RECORDS CONSIDERED IN DATA SEARCH Source Data Monthly Operating Report Plant Power History, Forced and Scheduled Shutdowns, Initiating Events, Vital Equipment Failures Weekly Reports Initiating Events Maintenance Request Logbooks Component Failures Os Work Requests (microfilm) Component Failures Job Tickets (microfilm) Component Failures Computerized Work' Requests and Job Tickets Component Failures Switching and Tagging Orders Component Maintenance Events Component Run-Time Record Component Operating Hours 0161G060485 ,

TABLE 3-2. TMI-1 COLD SHUTDOWN OUTAGES From To Reason D{at October 20, 1974 October 30, 1974 Repair Pressurizer Valve Leaks i 16 November 17, 1974 November 21, 1974 Replace Faulty Control Rod Drive Motor 96 April 5, 1975 April 14, 1975 Repair Control Rod Drive Cable Connector 205 May 25, 1975 June 11, 1975 Repair DR Pump 1B and Control Rod Drive 389 September 27, 1975 October 8, 1975 Repair RCP 1A 285 October 16, 1975 October 19, 1975 Repair Control Rod Drive Stator 84 Ncvember 12, 1975 November 24, 1975 Repair Control Rod Drive Stator and Turbine 295 Valve December 17, 1975 December 21, 1975 Repair Makeup Valve 81 January-16, 1976 January 17, 1976 Replace Control Rod Stator 35 Februa ry 21, 1976 May 24, 1976 Refueling Outage 2,215 November 6,1976 December 2, 1976 Miscellaneous Repair 638 w March 19, 1977 May 15, 1977 Refueling Outage 1,296 4 September 16, 1977 September 25, 1977 Miscellaneous Repairs 224 March 18, 1978 April 30, 1978 Refueling Outage 1,053 June 21, 1978 June 29, 1978 RCP Seal Failure 173 February 17, 1979 March 27, 1979 Refueling 912 March 28, 1979 October 2, 1985 TMI-2 Accident 46,116 NOTES:

1. Unit I started commercial operation on September 2, 1974.
2. Data based on TMI monthly operating reports (January 1,1975 to August 31,1977) and NRC "Gray Book" Reports, 1974 through 1978.

01 42186

O O O TABLE 3-3. TEST PROCEDURES CONSIDERED FOR SUCCESS DATA DEVELOPMENT Sheet 1 of 7 Procedure Humber Title Number of-Tests Performed

  • 1300-3A A/B Reactor Building Spray Pump Functional Test Recirculation 18/25 Mode and Reactor Building Spray System 1300-38 A/B Decay Heat Removal Pump Functional Test and Decay Heat 32/37 Hemoval System Valve Operability Test 1

1300-3C Decay Heat Closed Cooling Water Pumps Functional Test 41 Surveillance Frequency - 92 Days 1300-30 Decay Heat River Water Pump Functional Test and 48 Decay Heat River Valve Operability Test ,

i* 1300-3E Spent Fuel Cooling Pump Functional Test **

D! furveillance Frequency - 92 Days 1300-3F A/B Notor-Driven Emergency Feedwater Functional 20/17 Verification and Valve Operability 1300-3G A/B Turbine-Driven Emergency Feedwater Pump 26/35 Functional Test and Valve Test / Valve Lineup Operability Test 1300-3H A/B Makeup Pump and Valve Functional Tests 13/17 1300-31 A/B NSRW Pump Functional Test and Valve Operability 31/40 Test 1 1300-3J Nuclear Service Closed Cooling Water Pump and Valga 40 Functional Test l

  • Numbers separated by "/" (e.g.,18/26) correspond to the procedures separated by "/" in the procedure number i column; e.g., 1300-3A A/B.

l ** Test procedure reviewed but did not involve components of interest in data analysis.

0161G062386DAR

TABLE 3-3 (continued)

Sheet 2 of 7 Procedure Number Tit 1e "*

  • Performed
  • 1300-3X A/B Reactor Building Emergency Cooling Pump Functional Test 30/41 and Reactor Building Emergency Cooling System 1300-3N Chilled Water Pump (AH-P-3A/B) Functional Test and Valve 44 Operability Test 1300-3P IST of Check Valves During Shutdown 10 1300-3Q Quarterly Inservice Testing of Valves During Normal Plant **

Operations 1300-3R IST of Valves Shutdown and Remote Indication Check **

U 1300-3T Pressure Isolation Test of CF-V-4A/B, CF-V-5A/B, and 3 OH-V-22A/B 1300-4A Hydrostatic Test for ISI **

1300-4C A/B NSRW Pump Functional Test and Valve Operability Test 4/1 1300-4E Nuclear Services Closed Cooling Water Pump and Valve S Functional Test During Refuelings 1301-1 Shift and Gaily Checks **

1301-4.1 Weekly Surveillance Checks **

1301-4.4 Borated Water Storage Tank **

1301-4.6 Station Storage Batteries Required Interval - Weekly **

  • Numbers separated by "/" (e.g.,18/25) correspond to the procedures separated by "/" in the procedure number col umn ; e.g. ,1300-3A A/8.
    • Test procedure reviewed but did not involve components of interest in data analysis.

e 0161Guo2386DAR G S

O O O l

TABLE 3-3 (continued)

Sheet 3 of 7 Procedure Number Title Performed

  • 1301-5.8 Station Batteries Required Interval - Monthly **

1301-6.7 Monitoring of Silt Buildup in River Water Screen Hout **

1301-8.2 Diesel Generator Annual Inspection **

1301-9.7 Intake Pump House Floor, Silt Accumulation **

1301-10.1 Internal Vent Valve Inspection and Exercise **

1301-13.1 Emergency Equipment Readiness **

Y 1302-5.1 and Reactor Coolant Temperature Channels and Pressure / Temperature **

$ 1302-S.$ Comparator 1302-5.2 and RPS High and Low RC Pressure Channels Required Interval - **

1302-S.3 Refueling Interval 1302.5-4 Reactor' Coolant Flux Flow Comparator Required Interval -

Refueling Interval 1302-5.6 RPS Pump / Flux Comparator and RCP Power Monitor Surveillete **

Calibration 1302-S.7 High Reactor Building Pressure Channel **

1302-5.8 High and Low Pressure Injection Analog Channels **

1302-5.10 Reactor Building 4 psig Channels Required Interval - **

Refueling Interval

  • Numbers separated by "/" (e.g.,18/25) correspond to the procedures separated by "/" in the procedure number-column ; e.g.,1300-3A A/B. -
    • Test procedure reviewed but did not involve components of interest in data analysis.

0161G062386DAR

. ._ .- . . _ -. .- - = - - . . . . - - . .

TABLE 3-3 (continued)

Sheet 4 of 7 Procedure Number Numbe $

Title pe for ed*

1302-5.11 Reactor Building 30 psig Pressure Channels Required Interval - **

Refueling Interval 1302-S.18 High and Low Pressure Injection Flow Channel **

1302-5.19 Borated Water Storage Tank Level Indicator **

1302-S.25 Reactor Building Sump Level Required Interval - **

Each Refueling Period 1302-S.26 UTSG Level Channel Calibration **

co 1302-b.30 Diesel Generator Protective Relaying Required Interval - 7 h Refueling Interval 1302-5.31A 4,160V D and E Bus Degraded Grid Undervoltage Relay System **

Calibration 1302-S.318 4,160V D and E Bus Loss of Voltage Relay System Calibration 1302-5.31C 4,160V ID Bus Loss of Voltage / Degraded Grid Auxiliary Timer 3 Calibration 1302-6.31D 4,160V IE Bus Loss of Voltage / Degraded Grid Auxiliary Timer 3 Calibration 1302-5.34 Reactor Trip on Loss of Feedwater/ Main Turbine Trip Requireo **

Interval - Refueling Interval 1302-6.3 EFW Flow Instrumentation Calibration Required Interval - **

Refueling

  • humbers separated by "/" (e.g.,18/25) correspond to the procedures separated by "/" in the procedure number column ; e.g. ,1300-3A A/B.
    • Test procedure reviewed but did not involve components of interest in data analysis.

016196 02386DAR 9 ,

9

O O O TABLE 3-3 (continued)

Sheet-b of 7 Procedure Number Title Perform d*

1302-6.16 PORV Setpoint and Remote Position Check Required Interval - **

Refueling Interval 1303-1.1 Reactor Coolant System Leak Rate **

1303-4.1 Reactor Protection System 140/ Channel 1303-4.11 HPI and LPI Logic and Analog Channels **

1303-4.13 Reactor Building Emergency Cooling and Isolation System Analog **

Channels 1304-4.14 Reactor Building 30 psig Analog Channels 1303-4.16 Emergency Power System 200 DGA, 177 DGB 1303-4.17 Main Steam Isolation Valves Required Interval - Monthly **

1303-4.18 4 kV ES Bus Undervoltage Relay Test Required Interval - Monthly **

. 1303-4.19 HPI and LPI Analog Channels **

1303-5.1 Reactor Building Cooling and Isolation System Logic Channel and **

Component Test 1303-S.2 Loading Sequence and Component Test and High Pressure Injection 23 Logic Channel Test 1303-5.S Control Room Emergency Filtering System Operation Test 7

  • Numbers separated by "/" (e.g.,18/26) correspond to the procedures separated by "/" in the procedure number column ; e.g. ,1300-3A A/B. ~
    • Test procedure reviewed but did not involve components of interest in data analysis.

0161G062386DAR

TABLE 3-3 (continued)

Sheet 6 of 7 Procedure Number Title Pe form d*

1303-5.12 Control Building Emergency Ventilation System Air Distribution ,

Test 1303-6.1 Reactor Building Integrated Leak Rate Test 3 1303-8.4 Reactor Building Spray System Compressed Air Test 2 1303-10.1 Reactor Building Purge System Required Interval - No More than **

Once a Week Prior to Refueling Operation 1303-11.2 Pressurizer Code Safety Valves Setpoint Verification 8 y 1303-11.3 Main Steam Safety Valves 7 1303-11.8 High Pressure Injection 7 1303-11.9 Reactor Building Emergency Cooling System 7 1303-11.10 Engineered Safeguards System Emergency Sequence and Power 8 Transfer Test 1303-11.11 Station Batteries Load Test **

1303-11.13 Control Room Filtering System Test **

1303-11.14 Reactor Building Purge Exhaust **

1303-11.16 Decay Heat Removal System Leakage **

(A,D/8,C/E) 1303-11.19 Turbine Overspeed Testing 211/6/6

  • Numbers separated by "/" (e.g.,18/26) correspond to the procedures separated by "/" in the procedure nun.ber column ; e.g. ,1300-3A A/8.
    • Test procedure reviewed but did not involve components of interest in data analysis.

0161 2386DAR

01 kJ  %) J TABLE 3-3 (continued)

Shee+ 7 of 7 Procedure Number Title Perfo d*

1303-11.21 Core Flooding System Valve Ooerability Test 8 1303-11.22 Main Steam Isolation Valves **

1303-11.26 Reactor Building Isolation Valve Cycle Test Required Interval - **

Cold Shutdown 1303-11.27 Makeup and Purification System Leakage Check 2 l 1303-11.39 Emergency Feedwater Pump Automatic Start 10 1303-11.42 Emergency Feedwater Flow Test From Condensate Storage Tank 3 y Surveillance Frequency m

1303-11.45 PORV Setpoint Check. **

1303-11.S0 Reactor Building Spray System Leakage Check 5 1303-11.63 Emergency Feedwater Flow **

1 1303-11.54 Low Pressure Injection 2 1303-12.3 Venting of MU Pumps and HPI Lines **

1303-12.4 Venting of DH Pumps and LPI Lines **

l r

  • Numbers separated by "/" (e.g.,18/25) correspond to the procedures separated Dy "/" in the procedure number i column ; e.g. ,1300-3A A/8.
    • Test procedure reviewed but did not involve comoonents of interest in data analysis.

4 i l 0161GU62386DAR ,

TABLE 3-4. TMI-1 COMP 0NENT FAILilRE RATE DATA BASE sheet 1 of 3 IMI-I Spectitc Distribution TMI-1 Espertence g

Designator Component Description Failure Mode Mean 5th 95th Falle.re s (hours / demands) Mean p Median g ACR Air Compressor Fatture during Operatton 7 8.61*4* Hours 9.81-5 8.10-5 2.82-5 7.49-5 1.24-4 AC5 Air Compressor Failure To Start on Oceand 3.29-3 3.29-3 2.22-4 1.64-3 1.01-2 AD1 Air Dryer - Compressed Af r $ystee Failure during Operation 0 8.61*4 Hours 1.69-7 1.66-7 1.85-8 9.19-8 4.24-7 AF I Air Filter (ventilation) Failure during operation

  • 5.83-6 5.83-6 1.90-7 1.87 6 1.78-5 AF2 Air Filter (011 removall Failure durfag Operation
  • 1.76-5 1.76-5 5.73-7 5.62-6 5.37-5 AF) Air Felter (Compressed af f system) Failure during Operation
  • 3.54-5 3.54-5 1.15-6 1.13-5 1.08-4 8C Battery Charger Failure during Operation 9 5.17*5 Hours 1.86-5 1.63-5 8.49-6 1.35-5 2.83-5 815F 81 stable Failure To operate on Demand
  • 4.40-5 4.40-5 2.89-6 1.95-5 1.27-4 8T0 Battery (125V DC) Fatlure of Output durf ag Operation 5 1.72*5 Hours 7.53-7 1.29-5 6.02-6 1.18-5 2.38-5 3T0 125V DC Battery Failure of Output on Demand
  • 4.84-4 4.84-4 7.51-5 3.26-4 1.15-3 8US Electrical Bus Failure during Operation
  • 4.98-7 4.98-7 7.73-8 3.36-7 1.17-6 C81FC Circutt Breaker (AC 480V and above) Fallure To Close on Demand
  • 1.61-3 1.61-3 2.80-4 1.22-3 3.23-3 C81F0 Cf rcutt 3reaker ( AC 480V and above Failure To Open on Demand
  • 6.49-4 6.49-4 5.95-5 3.67-4 1.41-3 CSITO Circuit Breaker ( AC 480V and above fransfers Open during Operation
  • 8.28-7 8.23-7 5.08-8 3.99-7 2.36-6 C82FC Circuit Breaker ( AC 480V and above Failure To Close on Demand
  • 2.27-4 2.27-4 6.48-6 8.89-5 6.52-4 C82TO Circuit Breaker ( AC or DC LT. 480V) Transfers Open during Operation
  • 2.68-7 2.68-7 2.50-8 1.41-7 9.11-7 C84F0 Circut t Breaker (reactor trip) Failure To open on cemand 1 876 Demands 4.66-3 2.50-3 1.04-3 2.32-3 4.72-3 CFR 5tn9 1e Control Rod Assembly Failure on Demand 0 610 Demands 3.20-5 3.11-5 1.80-6 1.43-5 1.07-4 CbF Cavitating Ventort Faflure during Operation
  • 2.66-6 7.66-6 8.68-8 8.52-7 8.14-6 g DG5 01esel Generator Fallure To Start on Demand 14 869 Demands 2.14-2 1.58-2 1.04-2 1.51-2 2.38-2 DCRI Olesel Gererator Failure durlag First Hour of Operation 6 983 Hours 1.70-2 6.58-3 3.48-3 6.84 3 1.13-2 ru DGR2 Olesel Generator Failure af ter First Hour of Operation = 2.50-3 2.50-3 2.43-4 1.60-3 5.80-3 (n CIF Pneumatic Camper Failure To Operate on Cesand
  • 1.52-3 1.52-3 2.83-4 1.14-3 3.16-3 Olf Pneumatic Damper Transfers Open/ Closed during Operation
  • 2.67-7 2.67-7 1.78-8 1.20-7 6.71- 7 D2T Fire Camper Inadvertent Actuation
  • 4.20-8 4.20-8 1.69-9 1.41-8 1.31-7 D3F Gravity Damper Failure To operate on Demand
  • 1.52-3 1.52-3 2.83-4 1.14-3 3.16-3 EFVCC EFW Valve Control C' cut. Failure on Demand
  • 2.41-4 2.41-4 1.39-5 1.10-4 7.67-4 (FENA8 [FW [nable Failure during Operetten
  • 4.54-5 4.54-5 5.35-6 2.75-5 1.37-4 (FACT EFW Actuation Cie att Fatture on Demand
  • 2.41-4 2.41-4 1.39-5 1.10-4 7.67-4 EF WL 5W [FW tevel SmitC* Failure during Operation
  • 5.61-6 5.69-6 2.70-6 5.22-6 9.94-6 (FuS! EFW 5fonal Isr stor Failure during Operation
  • 8.75-6 8.75-6 2.25-6 6.90-6 2.06-5 (FSIG IFW Actuattr< iontrol Signal Failure durfng Operatton
  • 2.07-5 2.07-5 6.89-6 1.57-5 4.60-5 fJF [apansion Ant Failure during Operation
  • 1.64-6 1.64-6 1.00-8 2.24-7 6.96-6 FC0 Feed =ater 9and/ Auto Station Failure To 5= Itch to Manual Control on
  • 8.07-4 8.07-4 8.94-5 4.46-4 2.15-3 Cesand FC5 Feedwater Hand / Auto Station Failure during Operation
  • 1.30-5 1.30-5 1.44-6 7.20-6 3.47-5 F5P River Water screen Plugs during Operation 0 12 Hours 1.14-1 4.51-2 2.01-3 2.35-2 9.47-2 FT Flow Transmitter Failure during Operation
  • 6.25-6 6.25-6 6.03-7 4.18-6 1.41-5 Fwf ICS Feedwater Module Faflure during Operation
  • 1.30-4 1.30-4 1.44-5 7.20-5 3.47-4 FZ Fuse Fallure during Operatton
  • 9.20-7 9.20-7 2.83 8 3.16-7 2.83-6 F14 Ventilation Fan Failure durfag Operetton 13 2.58*5 Hours 7.89-6 3.63-5 2.37-5 3.39-5 5.27-5 FIS Ventilation Fan Failure To Start on Demand 5 932 Cesands 4.84-4 2.94-3 8.58-4 2.28-3 5.87-3 HXP Heat Excnanger Plugs durf ag Operation 0 1.59*6 Hours 1.95-6 7.49-7 1.41-7 5.82-7 1.53-6 HER Heat Enchanger teoks/ Ruptures during Operation 0 1.59 6 Hours 1.95-6 7.49-7 1.41-7 5.82-7 1.53-6 IMM ICS Integrated Mast *r Module Failure during Operattoa
  • 5.21-5 5.21 5 5.78-6 2.88-5 1.39-4 INV Inverter Failure during Operette
  • 1.83-5 1.83-5 1.73-6 1.14-5 4.16-5 LC 5tese Generator Water tevel Controller f ailure during Operation
  • 2.66-5 2.66-5 8.68-7 8.52-6 8.14-5
  • No plaat-spectftC data collecte$.

NOTE: Esponential notation is indicated in abbreviated foru; 1.e. 8.61*4 = 8.6? a 104 ; 9.81-5

  • 9.81 x 10-5, e G S

~ m

% b

%J h

%/ }

G}

TABLE 3-4 (continued)

- Sheet 2 of 3 TMI-1 Spectfic Ofstribution TMI-l Empertence Designator Component Description Failure Mode Mean 5th Median 95th Failures (hours / demands) Mean p p L5 ESAS Load Sequencer Failure To Operate on Demand

  • 2.40-6 2.40-6 7.84-8 7.69-7 7.34-6 L5W Limit 5=ttch Failure To Operate on Demand
  • 4.28-4 4.28-4 6.83-5 2.S3-4 1.10-3 LT Level Transmitter Failure during Operation
  • 1.57-5 1.57-5 3.51-6 1.12-5 3.34-5 ML T Manual Leader Failure during operation
  • 2.66-5 2.66-5 8.68-7 8.52-6 8.14-5 kP Reactor Betiding Spray Nozzles Plug during Operation
  • 7.06-8 7.06-8 2.70-9 3.02-8 2.00-7 OGF Cffstte Crto Faflure on Demand. Given Plant Trip
  • 2.66-4 2.66-4 8.68-6 8.52-5 8.14-4 PS Pushbutton Swf tch Fall To Operate on Demand
  • 2.40-5 2.40-5 8.29-7 7.90-6 7.32-5 PPI Piping. Cf. 3-f ach Diameter Failure per Section
  • 8.60-10 8.60-10 1.98-12 1.80-10 2.02 9 t PP2 Pfplag. < 3-Inch Diameter Failure per SectiwA per Hour
  • 8.60-9 8.60-9 1.98-11 1.80-9 2.02-8 PS Power Supply Failure during Operation
  • 1.71-5 1.71-5 1.18-6 7.25-6 4.39-5 P5W Pressure Switch Failure To Operate on Demand
  • 2.69-4 2.69-4 1.41-5 1.25-4 7.69-4 PT Pressure Transaftter Failure during Operation
  • 1.57-5 1.57-5 3.51-6 1.12-5 3.34-5 Pls Normally Operating Motor-Def ven Pump Failure To Start on Demand 2 393 Demands 2.35-3 3.49-3 8.61-4 2.67-3 8.16-3 PIR Normally Operating Motor-Driven Pump Failure during Operation 0 1.56*5 Hours 3.36-5 6.69-6 1.27-6 4.76-6 1.45-5 P25 5tandby Motor-Driven Pump Failure To Start on Demand 3 1731 Demands 3.29 3 1.83-3 5.29-4 1.69-3 2.96-3 P2R standby Motor-Drtwen Pump Fatture during Operatton 6 1.07*5 Hours 3.42-5 4.48-5 2.06-5 3.84-5 6.93-5 P3ts Turbine-Driven toergency Feed Pump Fatture To Start on Demand 0 119 Demands 3.31-2 3.31-2 5.75-3 2.50-2 7.10-2 P3t R Turbine. Driven Emergency Feed Pump Failure To Run 0 33 Demands 1.03-3 9.30-4 5.98-5 4.46-4 2.52-3 P35 Turbine-Driven Main Feed Pump Fativre To start 2 120 Demands 3-31 2 2.23 2 7.49-3 2.03-2 3.74-2 P3R Turbine-Drtwen Main Feed Pump Failure during Operation 3 6.36*4 Hours 1.03-3 6.90-5 2.70-5 6.04-5 1.25-4 w P45 hormally Operating River Water Pump Faftsre To Start 2 430 Demands 2.35 3 3.05-3 7.58-4 2.80-3 5.99-3 s P44 Normally operating River Water Pump Failure during Operation 4 1.18*5 Hours 3.36-5 3.02-5 1.22-5 2.45-5 5.30-5 N P55 Standby River uster Pump Failure To Start 4 814 Demands 3.29-3 4.11-3 1.46-3 3.58-3 7.20 3 C P54 Standby River Water Pump Failure during Operation  ! 5.45*4 Hours 3.42-5 4.41-5 1.27-5 3.99-5 8.56-5 P65 Vacuum Pump Failure To Start
  • 2.35-3 2.35-3 2.51-4 1.44-3 6.42-3 P64 Vacuum Pump Failure To Run
  • 3.36-5 3.36-5 2.75-6 1.64-5 9.00-5 AD Relay Fatture To Operate on Demand .* 2.41-4 2.41-4 1.41-5 1.35-4 6.40-4 R0 Relav Fatture during operation
  • 4.20-7 4.20-7 2.83-o 1.90-7 1.41-6 RSC Reactor Sump Clogs / Falls during operation
  • 1.00-5 1.00-5 3.46-7 3.29-6 3.05-5 SF service Water Strainer Failure during Operation 2 6.02+5 Hours 6.21-6 3.23-6 7.44-7 2.50-6 6.06-6 SIF 5eal Injtction Line Filter Plugging during Operation
  • 3.23-6 3.23-6 3.58 7 1.79-6 8.59-6 5W signal Modiffer Failure during Operetton
  • 2.94-6 2.94-6 4.66-7 2.04-6 6.42-6 SIC Shunt Trip Coll Failure To Operate on Demand
  • 1.40-4 1.40-4 3.27 5 1.05-4 2.94-4 TCF Timing Circuit Failure To operate on Demand
  • 2.40-6 2.40-6 7.84-8 7.69-7 7.34-6 TD0 Time Delay Relay Fa11ere To operate on Demand
  • 2.41-4 2.41-4 1.41-5 1.35 4 6.40-4 ff Temperature Element Fa11ere dortag Operation
  • 7.50-7 7.50-7 1.67-8 1.99-7 2.31-6 TES Turbine Enhaust 8oot Failure during Operation
  • 2.66 6 2.66-6 8.68-8 8.52-7 8.14-6 IM Temperature Monitor Loop ho Output .
  • 3.41-6 3.41-6 3.14-8 5.67-7 1.02-5 TR Tank Rupture during Operation 0 6.89 5 Hours 2.66-8 2.45-8 7.59-10 1.04-8 6.86-8 ULD ICS Unit Lead Demand Module Failure during Operation
  • 1.43-4 1.43-4 1.58-5 7.91-5 3.81-4 VCS Venttistion Cht 1er Failure To Start on Demand 5 375 Demanos 8.07-3 1.11-2 4.72-3 1.02-2 2.00-2 VCR Ventilation Chiller Failure during Operation 3 8.61*4 Hours 9.44-5 4.86-5 2.27-5 4.33-5 7.80-5 VIF Motor-operated Valve Fallure To Operate on Demand 50 1.42*4 Demands 4.30-3 3.51-3 2.78-3 3.40-3 4.25-3 VIT Motor-Operated Valva Transfers Open/ Closed during Operatton
  • 9.27-8 9.27-8 1.03-8 5.02-0 2.37 7 v2F 5elenefd valve Faflure To uperate on Demand
  • 2.43-3 2.43-3 7.64-5 9.79-4 6.94-3 V2T solenold Valve Transfers open/ Closed eersag operation 0 8.12*5 Mours 1.27-6 4.94 7 3.09-8 2.83-7 1.41-6
  • he plant-spectfic data collected.

MTE: Esponentf at notation is indicated in abbreviated Fore; f.e.. 2.40-6

  • 2.40 a 10-0; 1 56*5
  • 1.56 a 105.

0161C122986DAR ,

TABLE 3-4 (continued)

Sheet 3 of 3 TMI-1 Spectfic Otstributton TMI-I f apertence g Designator Component Description Fallure Mode Mean 5th 95th Fattures (hours / demands) Mean Median p ,

V3F Air-Operated Valve Fatlure To Operate on Demand 7 2.72+3 Oceands 1.52-3 2.16-3 1.40-3 2.13-3 3.43-3 V3M Air-Operated Valve Failure To Modulate to Control Pressure

  • 1.62-2 1.62-2 3.99-3 1.20-2 3.50-7 v 3T Atr-Operated Valve Transfers Open/ Closed during Operation 7 1.56*6 Hours 2.67-7 3.24-6 1.36-6 2.87-6 4.78-6 V3TC Af r-Operated valve Transfers Open/ Closed during Operation 0 1.72*4 Hours 2.67-7 2.62-7 1.50-8 1.10-7 7.85-7 V 3FP Atr-Operated Valve Fatture To Transfer to Failed Posttlon
  • 2.66-4 2.66-4 7.57-6 1.04-4 7.62-4 V4F Electrohydraulic Valve Failure To Operate on Demand
  • 1.52-3 1.52-3 2.83 4 1.14-3 3.16-3 V4T Electrohydraulic Valve Transfers Open/ Closed during Operation
  • 2.67-7 2.67-7 1.78-8 1.20-7 6.71-7 V6F Stop Check Valve Failure To Operate on Oceand -

9.13-4 9.13-4 7.01-5 4.21-4 2.35-3 Transfers Open/ Closed during Operation

  • 1.04-8 V67 Stop Check Valve 1.04-8 2.43-9 7.80-9 2.19-8 V7F Check Valve (other than stop) Failure To Operate on Oceand 1 4.96*3 Oceands 2.69-4 2.11-4 7.14 5 1.41-4 3.72-4 V7FI Check valve (Intermediate Cooling) Fatture To Operate on Oceand 1 276 Demands 2.69-4 5.09-4 1.37-4 2.45-4 1.41-3 V7FR Check Valve (river water) Failure To Operate on Cesand 10 3.48*3 Demands 2.96-4 2.08-3 1.16-3 1.79-3 3.22-3 V 7L Check Valve (other than stop) Gross Reverse Leakage during Operattori 1 2.04*5 Mours 5.36-7 9.78-7 1.41-7 5.45-7 2.56-6 V 7t! Check Valve (latermediate cooling) Gross Reverse testage during Operation 7 3.18*4 Hours 5.36-7 1.91-4 8.53-5 1.62-4 3.05-4 V 7t R Check Valve (river water) Gross Reverse Leakage during Operation 1 1.40*5 Hours 5.36-7 1.06-6 1.43-7 5.66 7 2.86-6 V7A Check Valve Cross Reverse Leakage during Operation
  • 7.24-5 7.24-5 1.20-6 1.65-5 2.23-4 V77 Check Valve (other than stop) Transfers Closed; Plugs during Operetton 0 1.23*6 Hours 1.04-8 1.03-8 2.43-9 7.80-9 2.15-8 V771 Check valve (latermediate cooling) Transfers Closed: Plugs during Operetton 0 1.91*5 Hours 1.04-8 1.04-8 2.43-9 7.80-9 2.18-8 V77R Check Valve (rtver water) Transfers Closed: Plugs during Operatton 0 1.74*5 Hours 1.04-8 1.04-8 2.43-9 7.60-9 2.18-8 w V8F Manual Valve Failure To Open on Oceand
  • 7.40-4 7.40-4 1.80-4 4.86-4 1.89-3 a V8T Manual valve Transfers Open/ Closed during Operatton 0 9.47*6 Hours 4.20-8 2.14-8 1.41-9 1.41-8 5.66-8 w V9F Reitef valve (other than PORV or Failure To Open on Demand
  • 2.42-5 2.42-5 7.55-7 9.72-6 6.92-5 O safety)

V 90 Reitef valve (other than PORV or Premature open

  • 6.06 4 6.04-6 1.08-4 3.94-6 1.73-5 i

sa fety) v10F5 Press. rtzer Safety Valve Fatture To Open on Demand (passing 0 131 Demands 3.28-4 2.92-4 1.34-5 1.41-4 8.98-4 stese)

Y10FW Pressurtier Safety Valve Failure To Open on Demand (passing 0 131 Demands 3.28-4 2.92-4 1 34-5 1.41-4 8.98-4 water)

V10R5 Pressurfrer Safety Valve Failure To Rescat on Demand (passing 0 131 Demands 2.87-3 1.53-3 8.84-5 1.07-3 4.30-3 steam)

VICRW Pressurtzer Safety Valve Fallure To Reseat on Demand (passing

  • 1.01-1 1.01-1 2.88-3 1.20-1 2.50-1 water)

V10T Pressurtzer Safety valve Transfers Open/ Closed

  • 3.03-6 3.03-6 5.38-7 1.97-6 8.65-6 V11F5 PCRW Fatture To Open on Demand (passing 0 18 Demands 4.27-3 4.10-3 9.95-4 3.20-3 8.28-3 steam)

V11FW PORY Fatture To Open on Demand (passing 0 18 Demands 4.27-3 4.10-3 9.95-4 3.20-3 8.28-3 water)

Y11R$ PORY Fat ture To Open/ Reseat on Demand 0 18 Demands 2.50-2 2.C5-2 5.85-3 1.77-2 3.85-2 hassing stese)

V11RW PORV Fallure To Restat on Demand 0 18 Demands 1.01-1 1.01-1 2.88-3 1.20-1 2.50-1 (passing water) .

VIIT PORY Transfer Closed during Operatton

  • 3.03-6 3.03-6 5.3S-7 1.97 6 8.65 6 V12F Turtstne Stop/ Control Valve Failure To Operate on Demand
  • 1.25-4 1.25-4 2.92-5 9.37-5 2.63-4 V13T Pressure Controlled Regulating Valve Transfer Closed during Operation
  • 1.69-5 1.69-5 1.88-6 9.37-6 4.51-5 V14F Air Compressor Transfer Valve Fallure to Operate on Cesand
  • 1.52-3 1.52-3 2.85-4 1.03-3 3.57-3 VS Y-Type Stralner Fallure during Operation
  • 2.66-6 2.66-6 8.68-8 8.52-7 8.14-6 X1F Transformer (GST/UAT/ RAT) Fatture during Operation 0 1.72*5 Hours 1.56-6 1.26-6 2.83-7 9.89-7 2.45-6 22F Transformer (Station Service /480V Failure daring Operetton 0 1.38*6 Hours 6.87-7 4.28-7 7.06-8 3.a6-7 8.37-7 to 4.160V)

X 3F Transformer (instrument /120V to Failure during Operetton

  • 1.55-6 1.55-6 7.44-8 6.57-7 4.18-6 480 0

%e plant-specific data collected.

h0TE: (sponenttal notation is indicated in abbreviated fore; f.e.. 2.72+3 = 2.72 x 103 ; 1.52-3

  • 1.52 x 10-3 O

9 .

9

l 7 - /S /~N U U TABLE 3-5. TMI-1 COMPONENT COMMON CAUSE PARAMETER DATA BASE Sheet 1 of 3 Distribution Designator Parameter Component Failure Mode 5th 95th

    • " an Percentile Percentile BACR Beta Factor Air Compressor Fails during Operation 5.00-2 5.11-3 2.98-2 1.28-1 l BACS Beta Factor Air Compressor Fails To Start on Demand 1.00-1 1.02-2 6.98-2 2.49-1 I BBISF Beta Factor Bistable Fails To Operate on Demand 5.00-2 5.11-3 2.98-2 1.28-1 l BCB4F0 Beta Factor Circuit Breaker (R.T.) Fails To Open on Demand 1.85-1 9.74-2 1.76-1 2.58-1 BDGS Beta Factor Diesel Generator Fails To Start on Demand 4.93-2 2.51-2 4.65-2 7.01-2 BDGR1 Beta Factor Diesel Generator Fails during First Hour of 4.09-2 1.59-2 3.75-2 6.39-2 Operation BDGR2 Beta Factor Diesel Generator Fails after First Hour of 4.09-2 1.59-2 3.75-2 6.39-2 Operation BD1F Beta Factor Pneumatic Damper Fails To Operate on Demand 1.00-1 1.02-2 6.98-2 2.49-1 BFI R Beta Factor Ventilation Fan Fafis during Operation 5.00-2 5.11-3 2.98-2 1.28-1 to BFIS Beta Factor Ventilation Fan Fails To Start on Demand 5.00-2 5.11-3 2.98-2 1.28-1 O

BHXP Beta Factor Heat Exchanger Plugs during Operation 5.00-2 5.11-3 2.98-2 1.28-1 BPIS Beta Factor Nonna11y Operating Motor-Driven Fails To Start on Demand 5.63-2 9.06-3 4.72-2 1.07-1 Pump BPIR Beta Factor Normally Operating Motor-Driven Fails during Operation 1.39-2 8.20-4 9.47-3 3.38-2 Pump BP2S Beta Factor Standby Motor-Driven Pump Fails To Start on Demand 1.62-1 8.44-2 1.54-1 2.28-1 BP2R Beta Factor Standby Motor-Driven Pump Fails during Operation 3.35-2 5.25-3 2.80-2 6.41-2 BP3S Beta Factor Turbine-Driven Pump Fails To Start on Demand 2.43-2 6.05-4 1.48-2 6.39-2 BP3R Beta Factor Turbine-Driven Pump Fafis during Operation 3.17-2 1.82-3 2.16-2 7.67-2 "

~

BP4S Beta Factor Normally Operating River Water Fails To Start on Demand 5.63-2 9.06-3 4.72-2 1.07-1 Pump BP4R Beta Factor Normally Operating River Water Fails during Operation 1.39-2 8.20-4 9.47-3 3.38-2 Pump l BPSS Beta Factor Standby River Water Pump Fails To Start on Demand 5.63-2 9.06-3 4.72-2 1.07-1 l BbSR Beta Factor Standby River Water Pump Fails during Operation 1.39-2 8.20-4 9.47-3 3.38-2 NOTE: Exponential notation is indicated in abbreviated form; i.e. 5.00-2 = 5.00 x 10-2, l

I I

_ . _ _ . _ _ . _ _ _- ._ _ . - - _ _ . ~ - . , . . . _ , ~ - _ - - - . . . - - - - - - - ._ -- - - - - - --

TABLE 3-5 (continued)

Sheet 2 of 3 Distribution Designator Parameter Component Failure Mode 5th 95th Mean Median Percentile Percentile BS Beta Cactor EFW Pump (pump portion) Fails To Start on Demand 2.56-2 6.37-4 1.56-2 6.71-2 BR Beta Factor EFW Pamp (pump portion) Fails during Operation 3.42-2 1.97-3 2.34-2 8.26-2 BRD Beta Factor Relay Fails To Operate on Demand 1.00-1 1.02-2 6.98-2 2.49-1 BSF Beta Factor Service Water Strainer Fails during Operation 1.00-1 1.02-2 6.98-2 2.49-1 BTDD Beta Factor Time Delay Relay Fails To Operate on Demand 5.00-2 5.11-3 2.98-2 1.28-1 BVCS Beta Factor Ventilation Chiller Fails To Start on Demand 5.00-2 5.11-3 2.98-2 1.28-1 BVCR Beta f actor Ventilation Chiller Fails during Operation 1.00-1 1.02-2 6.98-2 2.49-1 BVlF Beta Factor Motor-Operated Valve Fails To Operate on Demand 8.07-2 6.29-2 7.99-2 9.41-2 BY6F Beta Factor Stop Check Valve Fails To Operate on Demand 1.00-1 1.02-2 6.98-2 2.49-1 BV9F Beta Factor Relief Valve (not PORV or Fails To Open on Demand 1.00-1 1.02-2 6.98-2 2.49-1 safety) w BV10FS Beta Factor Pressurizer Safety Valve Fails To Open on Demand (steam) 5.00-2 5.11-3 2.98-2 1.28-1 0

N BV10FW Beta Factor Pressurizer Safety Valve Fails To Open on Demand (water) 5.00-2 5.11-3 2.98-2 1.28-1 BV10RS Beta Factor Pressurizer Safety Valve Fails To Reseat on Demand (steam) 5.00-2 5.11-3 2.98-2 1.28-1 BV10RW Beta Factor Pressurizer Safety Valve Fails To Reseat on Demand (water) 5.00-2 5.11-3 2.98-2 1.28-1 GACR Gamma Factor Air Compressor Fails during Operation 5.00-1 2.11-1 5.20-1 7.60-1 GACS Gama Factor Air Compressor Fails To Start on Demand 5.00-1 2.11-1 5.20-1 7.60-1 GBISF Gama Factor Bistable Fails To Operate on Demand 5.00-1 2.11-1 5.20-1 7.60-1 GCB4F0 Gamma Factor Circuit Breaker (R.T.) Fatis To Open on Demand 4.29-1 1.77-1 4.08-1 6.27-1 GFlR Gama Factor Ventilation Fan Fails during Operation 5.00-1 2.11-1 5.20-1 7.60-1 GFl S Garma Factor Ventilation Fan Fails To Start on Demand 5.00-1 2.11-1 5.20-1 7.60-1 GHXP Gama Factor Heat Exchan9er Plugs during Operation 5.00-1 2.11-1 5.20-1 7.60-1 GP15 Gama Factor Normally Operating Motor- Fails To Start on Demand 2.50-1 7.18-5 1.81-1 5.84-1 Driven Pump GPIR Garca Factor Normally Operating Motor- Fails during Operation 5.26-1 4.23-2 4.88-1 9.21-1 Driven Pump GP2S Garma factor Standby Motor-Driven Pump Fails To Start on Demand 3.66-1 1.37-1 3.44-1 5.56-1 GP2R Gama Factor Standby Motor-Driven Pump Fails during Operation 2.49-1 8.12-3 1.80-1 5.81-1 NOTE: Exponential notation is indicated in abbreviated form; i.e. 2.56-2 = 2.56 x 10-2, C161G04168

O O O b TABLE 3-5 (continued)

Sheet 3 of 3

. Distribution i

Designator Parameter Component Failure Mode '

Percentile Percen ile I GP45 Gansna Factor Normally Operating River Fails To Start on Demand 2.50-1 7.18-5 1.81-1 5.84-1

Water Purnp GP4R Gama Factor Normally Operating River Fails during Ope,ation 5.26-1 4.23-2 4.88-1 9.21-1 Water Pump GRD Gama Factor Relay Fails To Operate on Demand 5.00-1 2.11-1 5.20-1 7.60-1 GTDD Gama Factor Time Delay Relay Fails To Operate on Demand 5.00-1 2.11-1 5.20-1 7.60-1 GYlF Gama Factor Motor-Operated Valve Fails To Operate on Demand 2 .01 -1 1.23-1 1.94-1 2.64-1 GV9F Gama Factor Relief Valve (not PORY or Fails To Open on Demand 5.00-1 2.11-1 5.20-1 7.60-1 or safety)

GV10RS Garuna Factor Pressurizer Safety Valve Fafis To Reseat (steam) 5.00-1 2.11-1 5.20-1 7.60-1 Y NOTE: Exponential notation is indicated in abbreviated form; 1.e. 2.50-1 = 2.50 x 10-1 U

l 0161G041686

- _-_ --...- _ _ _ ,.../._ _ _ . - _ . - - - _ , . - _ _ - . _ _ _ - . - _ - _ - . . _ _ . . - . . _ . . _.__

1 1

TABLE 3-6. TMI-1 COMPONENT MAINTENANCE FREQUENCY Sheet 1 of 2 THI-1 Experience

  • Generic Designator Component Mean Events Hours Mean 5th 95th (events / hour) Median p p MFCD1 Main Steam ADV 6 6.36+4 2.75-5 3.25-5 2.23-5 3.10-5 MFCD2 Pressurizer Spray Valve 4.08-5 2 9.54+4 2.75-5 2.68-5 1.81-5 2.55-5 3.45-5 MFCF1 Reactor Building Fan 41 9.54+4 2.19-4 4.02-4 2.94-4 3.85-4 4.89-4 MFCF2 Motor-Operated Cooler Inlet Valve **

2.75-5 2.75-5 1.85-5 2.62-5 3.67-5 MFCF3 Motor-Operated Cooler Outlet Valve ** 2.75-5 2.75-5 1.85-5 2.62-5 3.67-5 MFCF4 Reactor Building Cooling Unit ** 2.75-5 2.75-5 1.85-5 2.62-5 3.67-5 MFCFS Reactor River Water Pump 73 6.36+4 8.42-5 1.06-3 8.44-4 1.03-3 1.24-3 MFCF6 Motor-Operated River Water Discharge Valve **

2.75-5 2.75-5 1.85-5 2.62-5 3.67-5 MFCF7 Reactor River Pump Minimum Flow Valve ** 2.75-5 2.75-5 1.85-5 2.62-5 MFCF8 Reactor River Strainer 3.67-5 2 6.36+4 2.19-4 7.38-5 4.06-5 6.53-5 1.21-4 MFCF9 RR-V5 **

2.75-5 1.85-5 MFCF10 Fan Motor Cooler ** 2.62-5 3.67-5 2.75-5 1.85-5 2.62-5 3.67 5 MFCII Letdown Isolation Valve 2 9.54+4 2.75-5 2.68-5 1.81-5 2.55-5 3.45-5 MFCSI Reactor Building Spray Pump t 11 6.36+4 2.19-4 1.78-4 1.15-4 1.75-4 MFCV1 Control Tower Instrument Air Compressor 2.47-4

" 13 1.27+5 1.08-4 6.80-5 1.16-4 1.46-4 MFCV2 Control Building Fan 32 1.27+5 2.19-4 2.33-4 1.70-4 2.30-4 w MFCV3 Chilled Water Train (pump and chiller) 2.75-4

  • 17 6.36+4 2.19-4 2.53-4 1.60-4 2.43-4 3.39-4 MFC31 MU-V26 **

2.75-5 1.85-5 2.62-5 3.67-5 MFC32 MU-V25 **

2.75-5 1.85-5 2.62-5 3.67-5 MFDA1 Station Battery 0 6.36+4 2.75-5 2.55-5 1.77-5 2.44-5 MFDA2 Battery Charger 3.35-5 2 1. 91 + 5 2.75-5 2.44-5 1.77-5 2.33-5 3.15-5 MFDH1 DHR Pump 30 6.36+4 8.42-5 3.41-4 2.43-4 3.27-4 4.41-4 MFDH2 DHR Cooler **

2.75-5 2.75-5 1.85-5 MFDH3 DHR MOV 2.62-5 3.67-5 10 1.26+5 2.75-5 3.49-5 2.46-5 3.35-5 4.60-5 MFDH4 LPI/HPI Cross-Connect Strainer 5 6.36+4 2.19-4 1.10-4 5.74-5 1.06-4 1.60-4 MFEF) EFW Pump 24 9.54+4 2.19-4 2.43-4 1.79-4 2.42-4 3.14-4 MFEF2 EFW Pump Steam Supply Valve 10 1.59+5 2.75-5 3.35-5 2.35-5 3.26-5 MFEF3 EFW Logic Channel ** 4.16-5 2.75-5 2.75-5 1.85-5 2.62-5 3.67-5 MFFW1 Condensate Pump 19 9.54+4 1.26-4 1.76-4 1.14-4 MFFW2 Condensate Booster Pump 1.68-4 2.30-4 36 9.54+4 1.26-4 3.19-4 2.37-4 3.09-4 4.00-4 MFGA1 Diesel Generator 102 6.36+4 2.75-5 1.57-3 1.29-3 1.53-3 1.82-3 NFGA2 Fuel Oil Transfer Pump 2 6.36+4 2.75-5 2.77-5 1.94-5 2.65-5 3.66-5 MFHA) DHRW Pump 78 6.36+4 8.42-5 1.13-3 9.28-4 1.12-3 1.32-3 MFHA2 DHRW Discharge Valve ** 2.75-5 2.75-5 1.85-5 MFHA3 River Water Strainer 2.52-5 3.67-5 5 6.36+4 2.19-4 1.10-4 5.74-5 1.00-4 1.61-4 MFHA4 Decay Heat Cooler 5 1.27+5 2.75-5 2.90-5 2.00-5 MFHAS Decay Heat CCW Pump 2.'/8-5 3.84-5 14 6.36+4 8.42-5 1.58-4 9.93-5 1.58-4 2.21-4 MFHA6 Decay Heat Service Cooler (scheduled) **

2.75-5 1.85-5 2.62-5 MFHL1 DH-V3 3.67-5 2 3.18+4 2.75-5 2.88-5 1.97-5 2.75-5 3.87-5 MFNP1 Normally Running Makeup Pump 14 3.18+4 1.26-4 2.84-4 1.87-4 2.79-4 3.96-4 cTMI-1 experience is based on the maintenance events documented on maintenance suranary sheets (Appendix C).

C*No plant-specific data collected.

NOTE: Expo al notation is indicated in abbreviated form; f.e 6.36+4 = x 104; 2.75-5 = 2.75 x 10-5, 0161G121786D W t

-s s l

/

TABLE 3-6 (continued) l Sheet 2 of 2 P*# ' " "

TMI-l Experience

  • Generic Designator Component Mean Mean 5 th 95th Events Hours (events / hour) p Median p j ) 9) 9j MFHPA Maintenance Frequency - Two MU Pumps ** 2.75-5 1.85-5 2.62-5 3.67-5 under Maintenance MFHP2 Standby Makeup Pump 27 6.36+4 8.42-5 3.07-4 2.19-4 2.92-4 3.98-4 MFNS1L NSRW Pump - Long Duration 8 9.54+4 2.75-5 3.37-5 2.33-5 3.28-5 4.28-5 MFNSIS NSRW Pump - Short Duration 86 9.54+4 1.26-4 8.71-4 7.20-4 8.61-4 9.74-4
    • 2.75-5 2.75-5 MFNS2 MOV 1.85-5 2.62-5 3.67-5 MFNS3 NSCCW Pump 23 9.54+4 1.26-4 2.10-4 1.49-4 2.03-4 2.70-4 MFNS4 Sin 91e Auxiliary Building 12 1.27+5 2.19-4 1.10-4 6.78-5 1.05-4 1.48-4 MFNS5 Nuclear Services Cooler 0 3.it:4 2.75-5 2.65-5 1.77-5 2.52-5 3.45-5 MFDP1 Auxiliary Station Service Trar.sformer ** 1.26-4 1.26-4 5.46-5 1.10-4 2.15-4 MFP01 Frequency that PORY Is Declared Inoperable ** 2.75-5 2.75-5 1.85-5 2.02-5 3.67-5 with Reactor at Power MFP02 Frequency that PORY Block Valve Is Declared ** 2.75-5 2.75-5 1.85-5 2.62-5 3.67-5 Inoperable with Reactor at Power

'O **

MFRT1 RPS Channel 8.42-5 8.42-5 3.65-5 7.34-; 1.44-4 ca MFSE1 ICCW Pump 5 6.36+4 1.26-4 9.74-5 5.36-5 9.30-5 1.44-4

'" Sump Isolation Valve MFSR1 0 0 2.75-5 2.75-5 1.85-5 2.62-5 3.67-5 DH-v5 **

MFSR2 2.75-5 1.85-5 2.62-5 3.67-5

  • TMI-l experience is based on the maintenance events documented on maintenance sunmary sheets (Appendix C).
    • No plant-specific data collected.

NOTE: Exponential notation is indicated in abbreviated form; i.e. 6.36+4 - 6.36 x 10 4 ; 2.75 2.75 x 10-5,

, . . e- , , , . ---.,-.e.. . . . - . . . - , . - . . - - - . , n .-.,...-,,m . ..-- . .--% , .e-. . - . - - . - 2- ,. ~+

TABLE 3-7. TMI-1 MEAN MAINTENANCE DURATION DATA BASE Sheet 1 of 2 TMI-1 Specific Distribution

  • Designator Component Mean Mean 5th 95th (hours) Percentile Percentile HDCD1 Main Steam ADV 5.6 1.3/+1 6.70+0 1. 59 +1 1.7021 MDCD2 Pressurizer Spray Valve 5.6 6.48+0 3.57+0 5.58+0 1.10 +1 MDCFI Reactor Building Fan 20.9 1.74+1 1.57+1 1.60+1 2.14+1 MDCF2 Motor-Operated Cooler Inlet Yalve 5.6 5.56+0 3.20+0 5.48+0 7.50+0 MDCF3 Motor-Operated Cooler Outlet Yalve 5.6 5.56+0 3.20+0 5.48+0 7.53+0 MDCF4 Reactor Building Cooling Unit 20.9 2.09+1 1.23+1 1.90+1 2.79+1 MDCFS Reactor River Water Pump 20.9 1.46+1 1.15+1 1.42+1 1.82+1 MDCF6 Motor-0perated River Water Discharge 5.6 5.56+0 3.20+0 5.48+0 7.50+0 Valve MDCF7 Reactor River Pump Minimum Flow Valve 5.6 5.56+0 3.20+0 5.48+0 7.50+0 MDCF8 Reactor River Strainer 20.9 2.05+1 1.07+1 1.80+1 3. 26 +1 MDCF9 RR-V5 5.6 5.56+0 3.20+0 5.48+0 7.50+0 MDCF10 Fan Motor Cooler 5.6 5.56+0 3.20+0 5.48+0 7.50+0 HDCII Letdown Isolation Valves 5.6 8.85+0 4.39+0 7.60+0 1.30+1 MDCSI Reactor Building Spray Pump 20.9 1.70+1 1.04+1 1.61+1 2.36+1 Y

w MDCV1 Instrument Air Compressor 20.9 2.04+1 1.91 +1 1.92+1 2.48+1 MDCV2 Control Building Fan 40.4 2.49+1 1.82+1 2.23+1 3.51+1 MDCV3 Chilled Water Train (pump and chiller) 40.4 3.29+1 1.91+1 2.76+1 5.85+1 MDC31 MU-Y26 5.6 5.56+0 3.20+0 5.48+0 7.50+0 MDC12 MU-Y25 5.6 5.56+0 3.20+0 5.48+0 7.50+0 MDDA1 Station Battery 5.6 5.56+0 3.20+0 5.48+0 7.50+0 MDDA2 Battery Charger 5.6 4.74+0 3.03+0 4.35+0 7.02+0 MDDH1 DHR Pump 20.9 1.29+1 1.04+1 1.24+1 1. 56 +1 MDDH2 DHR Cooler 20.9 2.09+1 1. 23 +1 1.90+1 2.79^1 MDDH3 DHR MOV 5, 6 1.18+1 6.32+0 1.17+1 1.70+1 MDDH4 LPI/HPI Cross-Conne t Strainer 20.9 2.14+1 1.22+1 1.87+1 3.28+1 MDEF1 EFW Pump 20.9 1.43+1 1.12+1 1.25+1 1.92+1 MDEF2 EFW Pump Steam Supply Valve 5.6 1.62+1 8.95+0 1.70+1 1.70+1 MDEF3 EFW Logic Channel 10.8 1.08+1 6.91+0 9.54+0 1.58+1 MDFW1 Condensate Pump l16.4 9.96+1 2. 78 +1 9.91+1 1.35+2 MDFW2 Condensate Booster Pump 116.4 4.62+1 2.56+1 3.55+1 9.91 +1 MDFW3 MFW Pump 4G.4 3.C9+1 1.67+1 2.77+1 6.21+1 MDFW4 Feedwater Pump 40.4 1.2b+2 2.30+1 6.26+1 3.29+2 MDFW5 Feedwater Isolation Yalves 40.4 7.76+0 4.26+0 6.97+0 1.20+1 MDGA1 Diesel Generator 40.4 2.17+1 1.81+1 2.16 +1 2.44+1 MDGA2 Fuel Oil Transfer Pump 40.4 3.98+1 1.74+1 3.08+1 7.02+1 MDHAl DHR'd Pump 20.9 1.60+1 1.22+1 1.55+1 1.88+1 MDHA2 DHRW Discharge Valve 5.6 5.56+0 3.20+0 5.48+0 7.50+0 MDHA3 Maintenance Duration - River Water 20.9 2.f4+1 1.22+1 1.87+1 3.28+1 Strainer

  • TMI-1 experience consists of individual outage durations documented on component v"intenance summary sheets (Appendix C).

NOTE: Exponential notation is indicated in abbreviated f.e., 1.37+1 = 1.37 x 10 1 .

O 0161G121786DAR

__w -

f, s Ag mi TABLE 3-7 (t:ontinued)

Sheet 2 cf 2 TMI-1 Specific Distribution

  • l Generic -

Designator Component Mean Mean 5tn 95th (hours) Percentile Percentile MDHA4 Decay Heat Service Cooler 20.9 5.70+1 2.78+# 5.4 5 +1 7.25+1 MDHA5 Decay Heat CCW Pump 20.9 1.98+1 1.90+1 1.92+1 2.21 +1 MDHA6 Decay Heat Service Cooler (scheduled) 20.9 5. 70+1 2.78+1 5.4 5+1 7.2 5+1 MDHL1 DH-Y3 5.3 5.14+0 3.13+0 4.73+0 7.22+0 MDHP1 One MU Pump 20.9 1.30+1 1.11+1 1.24+1 1.55+1 i MDHP2 Two MU Pump 20.9 2.09 +1 1.23+1 1.90+1 2.79+1 MDNS1L NSRW Pump - Long Duration 116.4 3.9 3 +2 3.00+2 3.57+2 4.76+2 MDNSIS NSRW Pump - Short Duration 5.6 1.23 +1 1.15 +1 1.20+1 1.34 +1 MDNS2 MOV 5.6 5.56+0 3.20+0 5.48+0 7.50+0 MDNS3 NSCf,W P mp 116.4 9.84 +1 3.49+1 9.91 +1 1.15+2 MDN54 Stragle h2xiliary Building Ventilation 20.9 2.4 3 +1 1.2 5+1 2.3 4 +1 3.31+1 Train MDNSS Nuclear services Cooler 20.9 5.70+1 2.78+1 5.4 5 +1 7.25+1 MDOP1 Auxiliary Station Service Transformer 116.4 1.17+2 1.60+1 ' 87+1

. 2.77+2 w MDP01 Duration between fold Shutdowns 2.50+3 1.35+3 2.0 7+ 3 3.9 9+ 3 d, MDRTI RPS C;eannel 5.6 5.56+0 3.20 +0 5.48+0 7.50+0 y MDSE1 ICCW Pump 116.4 5.88+1 1.58+1 2.84+1 1.68+2 MDSR1 Sumn Isolation Yalve 5.6 5." +0 3.20+0 5.48+0 7.50+0 MDSR2 BWST Isolation Yalve 5.6 5. 6+0 3. 2 ")+ 0 5.48+0 7.50+0 1

  • TMI-1 experience consists of individual outage durations documented on component me.intenance summary sheets (R7endix C).
    • Based on TMI-1 cold shutdown outage history (Table 3-3). No generic distsibution was used.

NOTE: Exponential notation is indicated in abbreviated form; f.e., 5.70+1 = 5.70 x 10 . 1 4

Y 0161G062386DAR

TABLE 3-8. TMI-1 INITIATING' EVENT FREQUENCY DATA BASE Plant-Specific Distribution Designator Initiating Event Category ,

Events Years Mean*

, Percentile Percentile Percentile LL 1. Large LOCA 2.66-4 0 4.5 1.91-4 7.30-6 7.36-5 5.21-4 ML 2. Medium LOCA 8.00-4 0 4.5 4.20-4 1.91-5 1.86-4 1.32-3

.SB 3. Small LOCA 3.56-3 0 4.5 3.25-3 2.66-5 9.43-4 1.06-2 VSB 4. Very Small LOCA 5.19-3 0 4.5 5.05-3 -2.19-4 2.55-3 1.37-2 VS 5. Inadvertent Opening of DHR Valves ** O 4.5 1.00-7 4.58-10 6.38-9 1.66-7 SLI 6. Steam Line Break in Intermedir.ie Building 8.00-4 0 4.5 4.20-4 1.91-5 1.86-4 1.32-3 SLT 7. Steam Line Break in Turbine Building 6.86-3 0 4.5 6.34-3 1.79-4 2.84 ' 1.58-2

$ TR 8. Steam Generator Tube Rupture 1.39-2 0 4.5 1.13-2 3.95-4 6.43-3 2.82-2 EXC 9. E .cessive Feedwater Flow 2.32-1 0 4.5 1.18-1 2.09-2 7.87-2 2.78-1 FW 10. Total Loss of Main Feedwater 5.48-1 0 4.5 2.33-1 5.11-2 1.83-1 4.81-1 RT 11. Reactor Trip 6.64+0 3 4.5 1.38+0 6.66-1 1.39+0 2.24+0 TT 12. Turbine Trip 1.89+0 7 4.5 1.64+0 7.75-1 1.53+0 2.32+0 LA 13. Loss of Air System ** O 4.5 6.00-3 2.00-4 1.87-3 1.89-2 LC 14. Loss of Control Buf1 ding Ventilation ** O 4.5 1.95-4 5.37-5 1.35-4 4.17-4 ATA 15. Loss of ATA Power 7.16-2 3 4.5 5.42-2 5.18-3 3.61-2 1.73-1 LD 16. Loss of DC Power Train. A 3.33-2 0 4.5 2.77-2 3.73-3 1.87-2 5.99-2 AC 17. Loss of Offsite Power 1.28-it 0 4.5 7.10-2t 1.91-2 5.30-2 1.54-1 LNS 18. Loss of Nuclear Ser '.es ClosN Cooling Water ** O 4.5 1.43-2 4.59-3 1.10-2 2.74-2 LR 19. Loss of River Water ** 0 12.0 7.41-3 3.51 -4 1.26-3 2.25-2

  • Ev:nts pe.- calendar year.
  • fEvent frequency quantified based on analysis of the system (s) involved, t Frcquency per site-year.

NOTE: Expo f al notation is indicated in abbreviated form; i.e., 2.66-4 66 x 10-4 0161Gl??oA6 nap

TABLE 3-1 GROUPING 0F EPRI EVENT CATEGORIES INTO TMI-1 INITIATING EVENT GROUPS Sheet 1 of 3 Plant Specific NP-2230 Initiating Event Category Used Initiating Event Group

1. Large LOCA None.
2. Medium LOCA None.
3. Small LOCA None.

4 Very Small LOCA None.

5. Inadvertent Opening of None.

DHR Valves

6. Steam Line Break in None.

Intermediate Butiding

7. Steam Line Break in None.

Turbine Building

8. Steam Generator Tube None.

Rupture

9. Excessive Feedwater Flow None. .
10. Total loss of Main Feedwater 16. Total loss of feedwater flow (all loops). This trenstent occurs when a simultaneous loss of all main feedwater occurs, excluding that due to loss of station power (see NP-2230 category 35).

24 Loss of condensate pumps (all loops). This transient occurs when all condensate pumps fail, causing a loss of feedwater flow.

(. 25. Loss of condenser vacuum. This transient occurs when either a complete loss or decrease in condenser vacuum results from a hardware or human error.

27. Condenser leakage. This transient occurs when excessive secondary system leakage occurs in the condenser.
30. Loss of circulating water. This transient occurs when circulating water is not available to the plant.
11. Reactor Trip 1. Loss of RCS flow (one loop). This transient occurs when an inadvertent hardware or human error interrupts the flow in one loop' of the reactor coolant system.
2. Uncontrolled rod withdrawal. This transient occurs when one or more control rods are withdrawn inadvertently.
3. CRDM problems and/or rod drop. This transient occurs when failures in the control rod drive mechanism (CRDM) occur which lead to out o, tolerance conditions in the primary system. The transient may include dropping of one or more control rods into the core as part of the CRDM failure.
8. High pressurizer pressure.
11. CVCS malfunction - boron dilution. This transient occurs when bardware or operator error results in a CVCS malfunctior such that reactor power is affected.
12. Pressure, temperature, power imS ? lance. This transient o<,:urs when various primary systems signals indicate pressure, temper.ture, or power imbalances.

14 Total loss of RCS flow. This transient occur., when a hardware or operator error causes a loss of reactor coolant system flow.

0161G060485 3-39

TABLE 3-9 (continued)

Init t ng Ev t roup NP-2230 Initiating Event Category Used

15. Loss or reduction in feedwater flow (one loop). This transient occurs when one feeowater pump trips or when another cccurrence results in an overall decrease in feedwater flow.
17. Full or partial closure of MSIV (one loop). This transient occurs when ore MSIV closes, the rest remain open, or the partial closure of one or more MSlYs occurs.

21 . Feedwater flow instability - operator error. This transient occurs when feedwater is being controlled manually. usually during startup or shutdown, and excessive or insufficient feedwater flow occurs.

22. Feedwater flow instability - miscellaneous mechanical causes. Thi s transient occurs when excessive or insufficient feedwater flow results from hardware failures in the feedwater system.
23. Loss of condensate pumps (one loop). This transient occurs when one condensate pump fails, reducing feedwater flow.
28. Miscellaneous leakage in secondary system. This transicat occurs when excessive leakage occurs in the secondary system, other than ,

the condenser (see NP-2230 category 27).

36. Pressurizer spray failure.
37. Spurious auto trip - no transient condition. This transient occurs when an auto scram is initiated by a hardware failure in instrumentation or logic circuits and no out of tolerance condition exists.
38. Auto / manual trip due to operator error. This transient occurs when an auto scram or manual scram is initiated by human error and no out af tolerance condition exists.
39. Manual trip due to false sinnals. This transient occurs when an operator initiat3s a scram based on infonnation from erroneous instrumentation.
40. Spurious trips - cause unknun. This transient occurs when a scram occur? and no out of tolerance condition can be detected, nor cause of scram determined.
6. High or low pressurizer pressure. This transient occurs when the pressurizer pressure is outside of the required operating limits.
12. Turbine Trip 33. Turbine trip, throttle valve closure EHC problems. This transient occurs when a turbine trip occurs or if turbine problems occur which, in effect, decrease steam flow to the turbine, causing a rapid change in the amount of energy removed from the primary system.

34 Generator trip or generator caused faults. This transient occurs when the generator is tripped due to electrical grid disturbances or generator faults.

18. Closure of all MSIVs. This transient occurs when any one of various steam line or nuclear system malfunctions requires termination of steam flow from the vessel, or by operator action.
13. Loss of Air System None.

O ,

! 0161G060485 3-40 1

t

, ' TABLE.3-9(continued)

( Shee't 3 of 3- r Plant Specific  :[

Initiating Event Group NP-2230 Initiating Event Category Used, j

. 14. ' Loss of Control 9uilding None. ,

Ventilation  !

i i

15. Loss of ATA Power None.

l

-16. . Loss of DC Power Train A None. l

.i

17. Total loss of Offsite Power None. 'i

?

18. Loss of Nuclear Services None. i Closed Cooling Water j
19. Loss of River Water None.

g i

i i

i e

i 1

. 1 l

a

' 3-41 .

0161G042186

9 To recifCulation Flow F --

gauge y Manual Motor-operated il valve M3 valve A To Manual N N  :/ 3 Manual Manual Check ) valve M1 From valve M4 valve M2 valve C suction Pump X i

FIGURE 3-1. EXAMPLE SYSTEM FOR FLOW TEST DATA

\

l l

9 3-42

i l

i FROM CORE  !

FLOODING TANK "B" l i

TO REACTOR VESSEL M M X' DH/LP TRAIN B DH V228 DH V4B i

,TO REACTOR VESSEL M 3 /' DH/LP TRAIN A DH V22A DH V4A '

l FROM CORE -

FLOODING TANK "A"  !

l O

FROM HOT t EG T

>4 P 74 T 7, O DH PUI 4 SUCTION DH V1 DH V2 DH V3 f

i i

FIGURE 3-2. C01.D LEG INJECTION PATH AND l HOT LEG SUCTION PATH St9PLIFIED DIAGRAM I I

I 3-43

_ - . - . - - - . - . - - - . . . . . _ - _ . - . . _ _ - - - - . - . . - . . . - - _ . . - - . - . . _ . ~ - - . -

4 i

4 APPENDIX A J

TMI-1 COMPONENT FAILURE RATE DATA 1

a l

i l

l l

l i

J 1

)

1 i

i

! 0210G0619860AR en,,- y m,, ._ --w,,--,p-wm,a,,y w,-,,,- -.w,w, , , ,

l APPENDIX A

/"' ,

ks T/ -- TMI-1 COMPONENT FAILURE RATE DATA l The data sheets in this appendix summarize the failure events and the corresponding number of demands or operating hours used for the .THI21

-plant-specific component failure rate estimation. The failure events were obtained from a review of Work Requests (WR) and Job Tickets (JT).

The reference WR or JT number is listed by each event. Component demands and operating hours were obtained from plant operating records, periodic tests, and run-time meter log. This appendix is divided into two subsections:

e Appendix A.1: Component Failure Data Summary-Sheets e Appendix A.2: Component Success Data Summary Sheets The data in each section are organized by component type and failure mode as listed in Table 3-4 of this report.

O i i u

I i

O ,

2

) A-1 0210GG619860AR 1

,--.,,.r.,,- - - - . ._

O s

APPENDIX A.1

, COMPONENT FAILURE DATA

SUMMARY

SHEETS

'O 4

J

\i I

h 1

J l

l 4

l I

iO l

l 0210G0619860AR 4

_wm

~ . _ . . __ .. - __ _ . . . .-

THI-1 COMPONENT FAILl'RE DATA

SUMMARY

SHEET DESIGNATOR: ACR SYSTEM: Instrument Air-COMPONENT TYPE: Air Compressor FAILURE MODE: Failure During Operation Site-Specific Data ,

e Failure Data for Given Failure Mode: Seven Failures Date Reported Failure Cause a_ 4-21-83 (JT-CA618) IA-P1A Leaking discharge valve--system pressure could not be~ maintained.

S/5/83 (JT-C2284)IA-PIB "Knocking" piston touching the bottom of cylinder.

9/10/74 (WR-4220)IA-P1A Tripped--would not start.

12/9/74 (WR-6061)IA-P1A Tripped on overload.  !

12/11/74 (WR-6089) IA-P18 Leads leaving breaker badly  !

) burned.

1 5/10/76 (WR-15356) SA-P-1A Bad bearing--replaced.

a 10/18/77 (WR-21679) SA-P-1A Tripped on thermal- l overload--replaced motor. i 1

i 2

e Failure Data Source: Maintenance Request Logbook l Work Requests (microfilm and computer file)  ;

Job Tickets (microfilm and computer file) j

I O  !

A.1-1 l 0210G0619860AR

TMI-1 COMPONENT FAILURE DATA

SUMMARY

SHEET DESIGNATOR: ACS SYSTEM: Instrument Air, Service Air COMPONENT TYPE: Compressor FAILURE MODE: Fail To Start Site-Specific Data e Failure Data for Given Failure Mode: Four Failures Date Reported Failure Cause 6/24/76 (WR-15'33a ) I A-PIB Failed to load--cleaned in next attempt.

6/24/82 (JT-C8702) IA-P1B Would not load at desired setpoint--control switch problem.

10/8/76 (WR-17189)SA-P1A Would not start.

(JT-C6041) IA-P1B Fails to unload and load properly.

e Failure Data Source: Maintenance Request Logbook i Work Requests (microfilm and computer file) l Job Tickets (microfilm and computer file) i 9!l 1

A.1-2 l 0210G061986DAR  !

i

.I E.m a THI-1 j COMPONENT FAILURE DATA

SUMMARY

SHEET  !

OESIGNATOR: AD1 i SYSTEM: Compressed Air ~ System  !

4

COMPONENT TYPE
Air Dryer 4

FAILURE MODE: Fail During Operation Site-Specific Data e Failure Data for Given Failure Mode: Zero Failures  !

i Date Reported Failure Cause-None i

!Oi l

l l

1 4

4 2

I

.i,

)

. l l  !

1 i

e Failure Data Source: Maintenance Request Logbook 2

Work Requests (microfilm and computer file)

Job Tickets (microfilm and computer file)

i I

f A.1-3

0210G061986DAR i

- . . -. _ . .-, , , ,- .- - . . - ,_, . - .,.._ ... _ .-..-.. _ -. - - - , _ _ _ - _ . i

TMI-l COMPONENT FAILURE DATA

SUMMARY

SHEET DESIGNATOR: BC SYSTEM: Electric Power COMPONENT TYPE: Battery Charger FAILURE MODE: Fail During Operation Site-Specific Data e Failure Data for Given Failure Mode: Nine Failures Date Reported Failure Cause 1/5/82 (JT-C7570) B Breaker tripped, replaced two control boards, adjusted current limit.

8/2b/81 (JT-C6631) 1A Would not pick up load.

1/21/80 (JT-C2324) IB Bad AC input breaker.

10/31/79 (JT-C1984)1B AC breaker tripped.

8/3/81 (JT-C6456) 10 Would not carry load.

3/22/77 (WR-19436) IC, 1E 1C tripped off, 1E could not be controlled.

8/17/77 (WR-21089) 1A Tripped on high voltage.

4/9/78 (WR-23422) 1A Tripped on high voltage--float and equalizer out of adjustment.

e Failure Data Source: Maintenance Request Logbook Work Requests (microfilm and computer file)

Job Tickets (microfilm and computer file)

O A.1-4 0210G061986DAR

l t

f THI-l l COMPONENT ' FAILURE DATA

SUMMARY

SHEET j DESIGNATOR: BTO  !

SYSTEM: Electric Power - l

' COMPONENT TYPE: Station Battery  !

FAILURE MODE: Fail During Operation l

Site-Specific Data j e Failure Data for-Given Failure Mode: Five Failures }

Date Reported Failure Cause i i

1/28/81 (JT-C5092)B Zero ground.  ;

i 2/33/81 (JT-C5320) A Grounds < 50 ka--relay in  ;

IP bus grounded. i i

) 12/28/7b (WR-13119)A Grounded. -

l i

12/26/76 (WR-1826b)A A - 70 kn grounds. i (WR-18265) B B - 50 ka grounds.

i i

l i

I l

e Failure Data Source: Maintenance Request Logbook

Work Requests (microfilm and computer file) l Job Tickets (microfilm and computer file)  !

i O  !

i l

l A.1-5 0210G061986DAR

TMI-l COMPGNENT FAILURE DATA

SUMMARY

SHEET DESIGNATOR: CB4F0 SYSTEM: RPS COMPONENT TYPE: Trip Breaker FAILURE MODE: Fail To Open on Demand Site-Specific Data _

e Failure Data for Given Failure Mode: One Failure Date Reported Failure Cause 9/1/76 (WR-16622) CB2 Failed to trip during test.

t G

e Failure Data Source: Maintenance Request Logbook Work Requests (microfilm and computer file)

Job Tickets (microfilm and computer file)

G A.1-6 0210G061986DAR

7- THI-1 (v) COMPONENT FAILURE DATA

SUMMARY

SHEET DESIGNATOR: CRF SYSTEM: Reactor Protection System COMPONENT TYPE: Single Control Rod Assembly FAILURE MODE: Fail To Insert on Demand Site-Specific Data e Failure Data for Given Failure Mode: Zero Failures Date Reported Failure Cause None 7s L,1 e Failure Data Source: Maintenance Request Logbook Work Requests (microfilm and computer file)

Job Tickets (microfilm and computer file)

[D x_ '

l i

1 l

A.1-7 .

0210G0619860AR  !

TMI-1 COMPONENT FAILURE DATA

SUMMARY

SHEET DESIGNATOR: OGR1 SYSTEM: Electric Power COMPONENT TYPE: Diesel Generator FAILURE MODE: Failure To Run during First Hour of Operation Site-Specific Data e Failure Data for Given Failure Mode: Six Failures Date Reported Failure Cause 5/9/77 (WR-19890) DG-1A Low oil pressure.

9/13/78 (WR-25197 ) DG-1B Governor failure.

9/14/78 (WR-25214) DG-1B Governor failure.

(WR-25197) 2/17/79 (JT-C0512) DG-1B Tripped on overspeed.

10,8/79 (JT-C1863) DG-1A Governor linkage failure.

2/11/82

  • DG-18 Breaker tripped open.

e Failure Data Source: Maintenance Request Logbook Work Requests (microfilm and computer file)

Job Tickets (microfilm and computer file)

  • Event listed in GPU memorandum on "Audit of Generator Start Failure Data," December 4, 1984.

A.1-8 02100061986DAR

i THI-1

-( COMPONENT FAILUdE DATA

SUMMARY

SHEET DESIGNATOR: DGS (Sheet 1 of 2)  !

SYSTEM: Electric Power COMFONEllT TYPE: Diesel Generator i

FAILURE MODE: Fail To Start on Demand

  • Site-Spectfic Data

(

e Failure Data for Given Failure Mode: 14 Failures Date Reported Failure -Cause- l t

2/22/79 (JT-C0630)DG-1A Starting air compressor would not load. i 4/11/79 (JT-C0870)DG-1B Failed to start.

i 2/17/76 (WK-14048)DG-1B Starting compressor would not ,

load.  ;

3 2/21/76 (WR-14132)UG-1B Breaker would not close--adjusted j Q governor. l 3/16/76 (WR-14404)DG-1A Would not start from control.

L 3/24/76 (WR-14b09) DG-1A Open air start compressor breaker. !

(

I 6/28/76 (WR-lb955) DG-1B Start compressor failed to start.

8/17/76 (WR-16494)OG-1A Start compressor failed to start.

[

8/23/76 (WR-15124)DG-1A Mould not start.

10/20/?6 (WR-17407)DG-1B Start compressor would not start.

i i

e Failure Data Source: %1ntenance Request Logbook 1 Work Requests (microfilm and computer file) ,

Job Tickets (microfilm and computer file)  ;

t O l A.1-9  !

0210G061986DAR ,

l

TMI-1 COMPONENT FAILURE DATA

SUMMARY

SHEET DESIGNATOR: DGS (Sheet 2 of 2)

SYSTEM: Electric Power COMPONENT TYPE: Diesel Generator FAILURE MODE: Fail To Start on Demand Site-Specific Data e Failure Data for Given Failure Mode: 16 Failures Date Reported Failure Cause 11/8/77 (WR-21871) DG-1A Low lube oil pressure.

4/27/78 (WR-23610) DG-1B Governor failure.

6/12/80

  • DG-1A No voltage control.

8/20/80

  • DG-18 No governor control.

O e Failure Data Source: Maintenance Request Logbook Work Requests (microfilm and computer file)

Job Tickets (microfilm and computer file)

  • Event listed in GPU memorandum on "Audit of Generator Start Failure Data," December 4, 1984.

A.1-10 0210G0519860AR

. -. - . = - - . ._

i TMI-1  !

COMPONENT FAILURE DATA

SUMMARY

SHEET .

, DESIGNATOR: F1R SYSTEM: Air Handling COMPONENT TYPE: Fan Unit ,

FAILURE MODE: Fail During Operation Site-Specific Data ,

e Failure Data for Given Failure Mode: 13 Failures I Date Reported Failure Cause

' 6/24/75 (WR-9777)AH-E-19A Short circuit.

?

8/17/78 (WR-24944) AH-E-1C Fan motor failure.  !

9/29/78 (WR-21489) AH-E-1C Excessive vibration--tripped.

l 2/3/83 (JT-CA110)AH-E-1C Blown fuse. l 3/2b/83 (JT-CA466) AH-E 1A Fan overloaded.

i O 4/27/82 (JT-C8266)AH-E-1A Motor seal failed--bad design.

1/19/79 (JT-C0302)AH-E-1B Motor failed. I 3/22/82 (JT-C8069) AH-E-1B Water leakage into motor.

b/16/79 (JT-C1016)AH-E-1C Water leakage into motor.

i b/28/80 (JT-C3270)AH-E-17A "Outer fan belt is off." l 12/29/82 (UT-C9906)AH-E-17A Fan tripped on overload--wiring  !

error. i b/12/81 (JT-Cb968) AH-E-95B Breaker tripped on thermal

overload. .

3 1

4 1/19/77 'WR-18569)Ah-E-178

, Fan belts flew off. '

1  !

e Failure Data Source: Maintenance Request Logbook

Work Requests (microfilm and computer file) l Job Tickets (microfilm and computer file) i

)

O  !

i i  !

A.1-11 j 0210G0619860AR i 4

,,-.__-.e . . - . . , - , - - - - _ , - , . . y. . ,,, , ,-_,_._-.._,,,rn.

77, _-- --,- . ,.-, .,.em.--rm-- - - , - . - - - _ ,.--.yy ,,c... . , - - , .c-

TMI-l COMPONENT FAILURE DATA

SUMMARY

SHEET DESIGNATOR: FIS SYSTEM: Air Handling COMPONENT TYPE: Fan FAILURE MODE: Fail To Start on Demand Site-Specific Oata e Failure Data for Given Failure Mode: Five Failures Date Reported Failure Cause 3/19/75 (WR-7962) AH-E-32 Breaker tripped--replaced f aulty overload block.

1/7/81 (JT-C4883) AH-E-17A Fan tripped after starting.

b/21/77 (WR-20096) AH-E-19B Would not start.

7/22/77 (WR-20789) AH-E-17B Would not start--damper interlock; damaged damper.

6/16/76 (WR-lS841) AH-E-17A Fan belts off.

e Failure Data Source: Maintenance Request Logbook Work Requests (microfilm and computer file)

Job Tickets (microfilm and computer file)

O A.1-12 0210GU61986DAR

,. TMI-l

(< j) COMPONENT FAILURE DATA

SUMMARY

SHEET DESIGNATOR: HXP, HXR SYSTEM: All COMPONENT TYPE: Heat Exchangers FAILURE MODE: Excessive Leakage, Rupture Site-Specific Data e Failure Data for Given Failure Mode: Zero Failures  ;

Date Reported Failure Cause None f

O\ i

\.)

I i

e Failure Data Source: Maintenance Request Logbook Work Requests (microfilm and cc: puter file) 1 dob Tickets (microfilm and computer file)  !

lT i s_-

V210G0619860AR l

l

TMI-l COMPONENT FAILURE DATA

SUMMARY

SHEET DESIGNATOR: P1R SYSTEM: Nuclear Services Secondary Water COMPONEf;! TYPE: Pump FAILURE MODE: Fail During Operation Site-Specific Data e Failure Data for Given Failure Mode: Zero Failures Date Reported Failure Cause None O

e Failure Data Source: Maintenance Request Logbook Work P,equests (microfilm and computer file)

Job fickets (microfilm a computer file)

O l

A.1-14 l 0210G061986DAR

,-~ TMI-1 (v) COMPONENT FAILURE DATA

SUMMARY

SHEET UESIGNATOR: P1R SYSTEM: Makeup COMPONENT TYPE: Pump 1B (normally operating)

FAILURE MODE: Fail During Operation Site-Specific Data e Failure Data for Given Failure Mode: Zero Failures Date Reported Failure Cause None i

't./

e Failure Data Source: Maintenance Request Logbook Work Hequests (microfilm and computer file)

Job Tickets (microfilm and computer file) 10

'wY A.1-lb 0210G0bl986DAR

TMI-1 COMPONENT FAILURE DATA

SUMMARY

SHEET DESIGNATUR: PlS SYSTEM: Nuclear Services Secondary Water COMP'JNENT TYPE: Pump FAILURE liODE: Fail To Start on Demand Site-Specific Data e Failure Data for Given Failure Mode: Two Failures Date Reported Failure Cause 4/1/78 (WR-23330) NS-P-1A Breaker failure.

7/1/83 (JT-CB174) NS-P-1B Relay failed to pick up after breaker closure.

O e Failure Data Source: Maintenance Request Logbook Work Requests (microfilm and computer file)

Job Tickets (microfilm and computer file)

G A.1-16 0210G0619860AR

, TMI-1

(  ;

COMPONENT FAILURE DATA

SUMMARY

SHEET s ./

UESIGRATOR: Pls SYSTEM: Makeup COMPONENT TYPE: Pump 1B (normally operating)

FAILURE MODE: Fail To Start on Demand Site-Specific Data e Failure Data for Given Failure Mode: Zero Failures Date Reported Failure Cause None i

n V

l i

e Failure Data Source: Maintenance Request Logbook Work Requests (microfilm and computer file)

Job Tickets (microfilm and computer file)

F)

V l

l A.1-17 02)UGU61986DAR J

TMI-1 COMPONENT FAILURE DATA

SUMMARY

SHEET DESIGNATOR: P2R SYSTEM: Makeup COMPONENT TYPE: Pumps lA and 1C FAILURE MODE: Failure During Operation Site-Specific Data e Failure Data for Given Failura Mode: Two Failures Date Reported Failure Cause 2/17/79 (JT-COS13) MO-P-1C Tripped-repaired stator winding.

10/10/74 (WR-4905) MU-P-1A Burned of f motor lead.

O e Failure Data Source: Maintenance Request Logbook Work Requests (microfilm and computer file) deb Tickets (microfilm and computer file)

O A.1-18 0210G061986DAR

. "\ .;

TMI-1 COMPONENT FAILURE DATA

SUMMARY

SHEET DESIGNATOR: P2R SYSTEM: Decay Heat j COMPONENT TYPE: Pump FAILURE MODE: Failure During Operation- .

Site-Specir'ic Data '

e Failure Data for Given Failure Mode: 'Two Failures Date, , Reported Failure Cause 9/19/79 (JT-C1875) DH-P-1A "Broken."

l 7/3/79 IdT-Cl329) DH P-b Pump tripped .-motor grounded.

l l

0 i

l l

e Failure Data Source: Maintenance Request Logbook Work-Requests (microfilm and computer file) l Job Tickets (microfi'-and computer file) l O

A.1-19 0210G061986DAR

TMI-1 COMPONENT FA' LURE DATA

SUMMARY

SHEET DESIGNATOR: P2R SYS TEM: Decay Heat Closed Cooling COMPGNENT TYPE: Pump FAILURE MODE: Fail During Operation Si te-Speci fi c. Data e Failure Data fc.' Given Failure Mode: Two Failures Date Reported Failure Cause 10/21/81 (JT-C7083) DC-P-1A Shaft sleeve replaced.

2/19/82 (JT-C7880) DC-P-1A Packing smoked--replaced.

O e Failure Data Source: Maintenance Request Logbook Work Requests (microfilm and computer file)

Job Tickets (microfilm and computer file) e A.1-20 0210G061986DAR

- ... . . _ . . - . - . - . - - . . - . = . - . . - - - . . . . . . . - . . . - - . . . - . . . - - _ _ - _ .

TMI-l i O COMPONENT FAILURE DATA

SUMMARY

-SHEET i r

DESIGNATOR: _

P2R f

SYSTEM: Building Spray l COMPONENT TYPE: Pump i

' FAILURE MODE: Fail Ouring Operation t

Site-Specific Data o Failure Data for Given Failure Mode: Zero Failures ,

t Date, Reported Failure Caus2 ]I None f r

f

'l l

f i

O :l h

i i

6 l

i I

f l

i I

e Failure Data Source: Maintenance Request Logbook  :

Work Hequests (microfilm and computer file)

Job Tickets (microfilm and computer file)

O l

. l A.1-21 .

0210GU619860AR ,

l

TMI-l COMPONENT FAILURE DATA

SUMMARY

SHEET DESIGNATOR: P2R SYSTEM: Emergency Feedwater COMPONENT TYPE: Motor-Driven EFW Pump FAILURE MODE: Fail During Operation Site-Specific Data e Failure Data for Given Failure Mode: Zero Failures Date Reported Failure -Cause None O

O e Failure Data Source: Maintenance P~uest Logbook Work Requests (microfilm and computer file)

Job Tickets (microfilm and computer file)

O A.1-22 0210G0619860AR

_._ TMl-1

  • I)

N,/

COMPONENT FAILURE DATA

SUMMARY

SHEET DESIGNATUR: P25 SYSTEM: Makeup COMPONENT TYPE: Pumps lA and 1C FAILURE MODE: Fail To Start on Demand Site-Specific Data e Failure Data for Giv.n Failure Mode: Two Failures Date Reported Failure Cause 10/17/74 (WR-bu69) MU-P-1C Tripping latch spring out of position.

1/3/76 (WR-6471) MU-P-1C Did not start on engineered safeguards signal--breaker failure.

7r (f'

)

i j e Failure Data Source: Maintenance Request Logbook l Work Requests (microfilm and computer file) l Job Tickets (microfilm and computer file) 1

.~.

l

( )

v )

I l

A.1-23 l U210G0619860Ag

TMI-1 COMPONENT FAILURE DATA

SUMMARY

SHEET DESIGNATOR: P2S SYSTEM: Decay Heat COMPONENT TYpr: Pump FAILURE MODE: ttil To Start on Demand Site-Specific Data e Failure Data for Given Failure Mode: Zero Failures Date Reported Failure Cause None 9

e Failure Data Source: Maintenance Request Logbook Work Requests (microfilm and computer file)

Job Tickets (microfilm and computer file)

G A.1-24 U210G0619860AR

a  ;

'l I

TMI-1

() COMPONENT FAILURE ~ DATA

SUMMARY

SHEET

-I DESIGNATUR: P2S l

SYSTEM: Decay Heat Closed Cooling l

I COMPONENT TYPE: Pump {

i FAILURE MODE: Fail To Start on Demand i Site-Specific Data I

[

Failure Data for Given Failure Mode:

e Zero Failures-Date Reported Failure 'Cause l l

None i

i 6

l i

( -

i i

'l i

I i

'l i

I e Failure Data Source: Maintenance Request Logbook l Work Requests (microfilm and computer file) i Job Tickets (microfilm and computer file) )

1

()

~ U210G061986DAR.

,- _ .__,_..-.-- _ ..__., . - - -_. , ...._ __- .._-. _ _ _...._......-,,_.,_ .... -..-,~..-_ .._ _ _, . . _ _ , . -

N.

TMI-1 COMPONENT FAILbnE DATA

SUMMARY

SHEET lh DESIGNATOR: P2S SYSTEM: Building Spray COMPONENT TYPE: Pump FAILORE MODE: Fail To Start on Demand Site-Specific Data e Failure Data for Given Failure Mode: Zero Failures Date Reported Failure Cause None O.

i l

J l

l i

e Failure Data Source: Maintenance Request Logbook Work Requests (microfilm and computer file)

Job Tickets (microfilm and computer file) 9 I

A.1-26 0210G061986DAR I

t I

TMI-1 l COMPONENT FAILURE DATA

SUMMARY

SHEET .

DESIGNATOR: P2S t

. SYSTEM: EFW ,

COMPONENT TYPE: Motor-Driven Pump FAILURE MODE: Fail To Start j Site-Specific Data f i

e Failure Data for Given Failure Mode: One Failure Date Reported Failure Cause

{

7/8/82 (JT-C8778) Pump 2A Failed to start during l SP 1303-11.39--fuse failure. l t

5 O .

l 7

1 i,

1 l

e Failure Data Source: Maintenance Request Logbook l Work Requests (microfilm and computer file) l Job Tickets (microfilm and computer file) l t

O  !

i A.1-27  !'

0210G061986DAR

l l

TMI-1  !

COMPONENT FAILURE DATA

SUMMARY

SHEET DESIGNATOR _: P3R SYSTEM: Main Feedwater COMPONENT TYPE: Pump FAILURE MODE: Fail During Operation Site-Specific Data e Failure Data for Given Failure Mode: Three Failures Date Reported Failure Cause 10/7/74 (WR-4870) Pump B Worn lube oil drive gear.

5/22/77 (WR-20102) Pump A Auxiliary oil pump failure.

7/29/83 (JT-CB423) Pump A Bearing failure.

G e Failure Data Source: Maintenance Request Logbook Work Requests (microfilm and computer file)

Job Tickets (microfilm and computer file)

O A.1-28 0210G061986DAR

- THI-1

{j COMPONENT FAILURE DATA

SUMMARY

SHEET DESIGNATOR: P3ER SYSTEM: Emergency Feedwater COMPONENT TYPE: Turbine-Driven Pump FAILURE MODE: Fail During Operation Site-Specific Data e Failure ';ata for Given Failure Mode: Zero Failures Date Reported Failure Cause None

,cx

I

%J l

e Failure D1ta Source: Maintenance Request Logbook ,

Work Requests (microfilm and computer file) l Job Tickets (microfilm and computer file)

'v' A.1-29 i 0210G121786DAR l 1

)

TMI-1 COMPONENT FAILURE DATA

SUMMARY

SHEET l

DESIGNATOR: P3S l

1 i SYSTEM: Main Feedwater i

COMP 0NENT TYPE: Pump FAILURE MODE: Fail To Start on Demand Site-Specific Data e Failure Data for Given Failure Mode: Two Failures Date Reported Failure Cause 4/18/78 (WR-23527) P-B Dirty contact on pressure switch.

10/3/79 (JT-C1832) P-B Turbine trip.

O e Failure Data Source: Maintenance Request Logbook Work Requests (microfilm and computer file)

Job Tickets (microfilm and computer file) e A.1-30 0210G061986DAi!

TMI-1

()

,__x COMP 0NENT FAILURE DATA

SUMMARY

SHEET DESIGNATOR: P3ES SYSTEM: Emergency Feedwater COMPONENT TYPE: Turbine-Driven Pump FAILURE MODE: Fail To Start on Demand Site-Specific Data e Failure Data for Given Failure Mode: Zero Failures Date Reported Failure Cause None (m./

s l

I l

1 i

e Failure Data Source: Maintenance Request Logbook Work Requests (microfilm and computer file)

Job Tickets (microfilm and computer file) l l

A.1-31 Od10G121786DAR

Tfil-1 COMPONENT FAILURE DATA

SUMMARY

SHEET DESIGNATUR: P4R SYSTEM: Nuclear Services River Water COMPONENT TYPE: Pump FAILURE MODE: Fail During Operation Site-Specific Data e Failure Data for Given Failure Mode: Four Failures Date Reported Failure Cause 1/7/76 (WR-13311) NR-P-1A Broken end bell on motor.

6/29/77 (WR-20501) NR-P-1C Broken pump coupling.

8/29/79 (JT-C1646) NR-P-1A Noisy shaft--pump replaced.

2/23/79 (JT-C0b49) NR-P-lC Motor f ailed.

O o Failure Data Source: Maintenance Request Logbook Work Requests (microfilm and computer file)

Job Tickets (microfilm and computer file)

O A.1-32 0210G061986DAR 1

. . . _.. - - - .. . _ . . . . . - - ~ . - . . . - - . .- .- . . - -. . . .

TMI-1 l COMPONENT FAILURE DATA

SUMMARY

SHEET

.i DESIGNATOR: P4S {

SYSTEM: Nuclear Services River Water COMPONENT TYPE: Pump j i

FAILURE MODE: Fail To Start on Demand lt Site-Specific Data-  !

e Failure Data for Given Failure Mode: Two Failures f

Date Reported Failure -Cause  !

I S/17/79 (JT-C1034) NR-P-1A Breaker failed.  !

9/22/83 (JT-CC012) NR-P-1A Tripped upon start--contacts  ;

adjusted. i O ,

t i

l I

i e Failure Data Source: Maintenance Request Logbook

  • Work Requests (microfilm and computer if'le) I Job Tickets (microfilm and computer file) l I

I O  !

l l

A.1-33 1

-0210G061986DAR I

,__._ __ ,.,._. _ _,_- _ _ _ _ _ ..____._.__.___.._._._._..__J

TMI-l COMPONENT FAILURE DATA

SUMMARY

SHEET DESIGNATOR: PbR SYSTEM: Decay Heat River Water COMPONENT TYPE: Pump FAILURE MODE: Failure During Operation Site-Specific Data e Failure Data for Given Failure Mode: Three Failures Date Reported Failure Cause 4/30/79 (JT-C0931) DR-P-1A Excessive vibration (overhauled).

5/14/80 (JT-C3172) DR-P-1A Excessive vibration--new bearings installed.

12/b/79 (JT-C2113) DR-P-1B High v' oration--rebuil t pump.

O e Failure Data Source: Maintenance Request Logbook Work Requests (microfilm and computer file)

Job Tickets (microfilm and computer file) e A.1-34 0210G061986DAR

i i

i THI-l a COMPONENT FAILURE DATA

SUMMARY

SfiEET f

i DESIGNATOR: PSR '

l SYSTEM: Reactor Building Emergency Cooling l CUMPUNENT TYPE: Pump l

FAILURE MOUE: Fail During Operation j i

Site-Specific Data  ;

e Failure Data for Given Failure Mode: Zero Failures Date Raported Failure Cause l, None

.l, I

1 l

i O -

1 l

I i

t 4

i

}

i J

l

)

i e Failure Data Source: Maintenance Request Logbook Work Requests (microfilm and computer file)

Job Tickets (microfilm and computer file)

O A.1-35 U210G061986DAR

. . _ . _ . ._ __ , . _ _ _ _ _ _ , . . _ . _ _ . _ . . - - _ _ _ _ . _ . _ _ _ _ _ _ _ - . . _ . - - . _ . _ _ . ~ . . - _ __

TMI-1 COMPONENT FAILURE DATA

SUMMARY

SHEET DESIGNATUR: P5S SYSTEM: Decay Heat River Water COMPONENT TYPE: Pump FAILURE MODE: Fail To Start on Demand Site-Specific Data e Failure Data for Given Failure Mode: One Failure Date Reported Failure Cause 9/22/83 (JT-C8991) DR-P-1B Pur.p will not rotate because div tr's desilting hose stuck in pump.

O l

l l

e Failure Data Source: Mainte,ance Request Logbook Work Requests (microfilm and computer file)

Job Tickets (microfilm and computer file)

G A.1-36 0210G0619860AR

i TMI-l  !

COMPONENT FAILURE DATA

SUMMARY

SHEET l

DESIGNATOR: PSS  !

SYSTEM: Reactor Cooling Emergency Cooling ,

COMPONENT TYPE: Pump FAILURE MODE: Fail To Start on Demand I t

Site-Specific Data f) e Failure Data for Given Failure Mode: Three Failures l i

Date Reported Failure Cause 1/3/75 (WR-6477)RR-P-1B Did not start on engineered  ;

safeguards signal.

l 12/15/77 (WR-22179) RR-P-1A Cause unspecified. }

J/15/77 (WR-18bb8) RR-P-1B Control switch failure.

O  !

l l

t i

l I

)

l e Failure Data Source: Maintenance Request Logbook j Work Requests (microfilm and computer file)  ;

Job Tickets (microfilm and computer file) l l

O  !

i l

1 l

A.1-37 U210G0619860AR j i

TMI-l COMPONENT FAILURE DATA

SUMMARY

SHEET DESIGNATOR: SF SYSTEM: Decay Heat River Water COMPONENT TYPE: Strainer FAILURE MODE: Fail During Operation Site-Specific Data e Failure Data for Given Failure Mode: One Failure Date Reported Failure Cause 12/9/83 (JT-CC648) 1A Clogged--would not clear by being backwashed.

O e Failure Data Source: Maintenance Request Logbook Work Requests (microfilm and computer file)

Job Tickets (microfilm and computer file)

O A.1-38 0210G061986DAR

4 TM1-1 COMPONENT FAILURE DATA

SUMMARY

SHEET DESIGNATOR: SF SYSTEM: Nuclear Services River Water COMPONENT TYPE: Strainer FAILURE MODE: Fail During Operation Site-Specific Data e Failure Data for Given Failure Mode: One Failure Date Reported Failure Cause 7/25/81 (JT-C6380) SIC Clogged.

O .

e Failure Data Source: Maintenan:e Request Logbook Work Requests (microfilm and computer file)

Job Tickets (microfilm and computer file)

O A.1-39 U210GUbl986DAR

TMI-1 COMPONENT FAILURE DATA

SUMMARY

SHEET DESIGNATUR: TR

, SYSTEM: All COMPONENT TYPE: Surge Tanks FAILURE MODE: Rupture During Operation Site-Specific Data e Failure Data for Given Failure Mode: Zero Failures Date Reported Failure Cause None r

O e Failure Dar.a Source: Maintenance Request Logbook Work Requests (microfilm and computer file)

Job Tickets (microfilm and computer file)

G A.1-40 U210G0619860AR

i I

i i

TMl-1  !

COMPONENT FAILURE DATA

SUMMARY

SHEET DESIGNATOR: VCR l I

SYSTEM: Control Building Chiller Unit l COMPONENT TYPE: Chiller i FAILURE' MODE: Fail During Operation f i

Site-Specific Data  !

Failure Data for Given Failure Mode: Three Failures e  !

l I

Date Reported failure Cause .

.2/3/79 (JT-C0436) AH-C-48 Kept tripping. l 4/12/79 (JT-C0858) AH-C-4B . Tripped on low oil pressure. j 4/2/7S (WR-8197) AH-C-4B Would not run.

i i

l i

O '

I i

l i

l i

l l

l e Failure Data Source: Maintenance Request Logbook 1 Work Requests (micr) film and computer file) {

Job Tickets (microfilm and computer file)

O ,

A.1-41 0210GU61986DAR I

TMI-l COMPONENT FAILURE DATA

SUMMARY

SHEET DESIGNATOR: VCS SYSTEM: Chilled Water COMPONENT TYPE: Chiller Unit FAILURE MODE: Fail To Start Site-Specific Data e Failure Data for Given Failure Mode: Five Failures Oate Reported Failure Cause 11/29/78 (WR-26068) 4A Low oil pressure--no electric power to oil heater.

3/4/79 (JT-CU646) 4A Antirecycle interlock would not clear.

8/29/79 (JT-C1644) 4A Purge unit failed to start.

10/24/79 (dT-C1958) 4A Vane control motor failure.

6/b/80 (JT-C33b9) 4A Low oil pressure, e Failure Data Source: Maintenance Request Logbook Work Requests (microfilm and computer file)

Job Ticke ts (microfilm and computer file)

O A.1-42 0210GU619860AR

i a

TMI-1 COMPONENT FAILURE DATA

SUMMARY

SHEET _

(

DESIGNATUR: V1f i t

t SYSTEM: Reactor Building Emergency Cooler  ;

COMPONENT TYPE: Motor-Operated Outlet Isolation Valve i

FAILURE MODE: Fail ic Operate on Demand l i

Site-Specific Data ,

t e Failure Data for Given Failure Mode: Zero Failures- f Oate Reported Failure Cause 12/13/74 (WR-23964)RR-V-4A Would~not close.  ;

l b/23/78 (WR-6140) RR-V-4B Faulty torque switch.  ;

5 i

O -

l i

i t

i i

-t i

I i

)

1 l

i e Failure Data Source: Maintenance Request Logbook '

Work Requests (microfilm and computer file)

Job Tickets (microfilm and computer file)  ;

e -

O l l

A.1-43 -i 0210G061986DAR

, . . _ , -~ ,. . - , _ . _ _ . _ . _ . ._ _ _ ._.- , .. ~ .. _ .-._ ,. _ .. _ _ . ..-_ _ ..,- ,-..._. ... - -. ...- - _ .... _ _ .-.-

TMI-1 COMPONENT FAILURE DATA

SUMMARY

SHEET DESIGNATOR: VlF SYSTEM: Intermediate Cooling COMPONENT T.PE: MOV FAILURE MODE: Fail To Operatc on Demand Site-Specific Data '

e Failure Dato for Given Failure Mode: Zero Failures Date Reported Failure Cause None O

e Failure Data Source: Maintenance Request Logbook Work Requests (microfilm and computer file) '

Job Tickets (microfilm and computer file)

O A.1-44 0210G0619860AR

t

- t TMI-1  !

COMPONENT FAILURE DATA

SUMMARY

SHEET  ;

DESIGNATOR: V1F ,

i SYSTEM: Main Steam  :

COMPONENT TYPE: MOV ~ -i 1

FAILURE MODE: Fail To Open or Close on Demand I

-i Site-Specific Data e Failure Data for Given Failure Mode: Six Failures  !

Date Reported Failure Cause l 7/1/78 (WR-24C6) MS-V-SA* Both valves failed to operate with operator--both tripped on f (WR-24456) MS-V-SB* overload .

l 5/16/75 (WR-9037)MS-V-108 Burned contacts in opening and'  !

closing circuits.  ;

l 2/12/80 (JT-C2523) MS-V-8A Motor tripped after only 10% open.  ;

11/12/83 (JT-CC484) MS-V-10A Stuck on seat--would not open electrically or manually.  !

3/11/79 (JT-C0688) MS-V-108 Replaced contacts in close i starter.

l

-i

. t i

i e failure Data Source: Maintenance Request Logbook  ;

Work Requests (microfilm and computer file) l Job Tickets (microfilm and computer file)

  • Possible common cause failure.

A.1-45 0210GU61986DAR

TMI-1 COMPONENT FAILURE DATA

SUMMARY

SHEET DESIGNATOR: VlF SYSTEM: Nuclear Services River Water COMPONENT TYPE: MOV FAILURE MODE: Fail To Open or Close on Demand Site-Specific Data e Failure Data for Given Failure Mode: 14 Failures Date Reported Failure Cause 1/6/76 (WR-21535) NR-V-1A Operator failed.

10/4/77 (WR-21535)liR-V-1A Velve disc broken.

$/16/75 (WR-90SS) NR-V-18 Would not operate.

9/17/76 (WR-11383) NR-V-4A Dirty contact on electrical interlock.

8/8/79 (JT-C1bl6) NR-V-lC Dirty contact and relay.

10/17/83 (JT-CC261) NR-V-2 Valve operator.

10/2/82 (JT-C9'35) NH-V-8A Stuck-open auxiliary contact.

4/9/83 (JT-CAS47) NR-V-8C Limtorque out of adjustment.

9/b/79 (UT-C1680) NR-V-108 Motor. ,

1 b/18/83 (JT-CA871) NR-V-lbB Tripped on thermal overload.

7/3/81 (JT-C6286) NR-V-16C Would not open.

2/22/78 (WR-22933) NR-V-1A Would not close or open during test.

(WR-21536)

I 7/16/75 (WK-10151) HR-V-1B Would not close or open.

8/8/7b (WR-10665)NR-V-1B Broken auxiliary block contractor.

e failure Data Sour- 'intenance Request Logbook k Requests (microfilm ano computer file)

'o Tickets (microfilm and conguter file)

O A.1-46 0210G061986DAR

i l  !

l TMI-l COMPONENT FAILURE DATA

SUMMARY

SHEET j i

LESIGNATOR: VlF l' i SYSTEM,: Decay Heat .

COMPONENT TYPE: MOV l

.7 l FAILURE MOL2. Fail To Open or Close on Demand i .i Site-Specific Data e Failure Data for Given Failure Mode: Eight Failures Date Reported Failure Cause l

i 3/6/80 'JT-C2729) 14 V2 Improperly set torque switch. I i.

j . 5/31/83 (JT-CA940)C;-V4A Torque switch opened up when  ;

l valve traveled partly open.  ;

r 1/12/79 (JT-C0204jDH-VSB Breaker tripped.  !

\  ;

11/7/80 lJT-C4463)DH-V56 Breaker tripped.  !

12/b/83 (JT-CC613) DH-V-7 Improperly set torque switch.

3/Z4/76 (WR-14493) UH471 Valve failed to open fully--  ;

replaced stem.

l 4/20/76 (WR-15048)UH-V2 Replaced operator, straightened  ;

valve stem. <

\

l l'/3/74 J (WR-6403)DH-V5A Breaker tripped. i l

i

, .i i  !

I e Failure Data Soarr. ,

i .ntenar 6:ast Lo9 book  !

l - 4 ' m- ,

rofilm and computer file)_ {

(rofilm and' computer file) l i

j. i

@ i i

A.1-47 0210GU619860AR

-+y - + - - - - g or ,--y. ---:- , - , , - = wy # my v-, e v .-v,- es g eo v,

TMI-l C_0MPONENT FAILURY DATA

SUMMARY

SHEET DESIGNATUR: V1F SYSTEM: Fen 1 water CUMPONENT TYPE: MOV FAILURE MODE: Fail To Open or Close on Demand Site-Specific Data e railure Data for Given Failurs Mode: Seven Failures ht_e, Reported Failure Cause 11/19/74 (WR-5676A) FW-V-5A Motor would not run--adjusted limit switch.

11/17/7C (WR-17769) FW-V-5A Would not close from control room.

7/17/80 (JT-C3652) FW-V-bB Would not stay open.

9/8/82 (JT-CB802) FW-V-1B Replaced motor.

11/19/78 (WR-25988) FW-V-1B Would not operate electrically.

(WR-2b966) .

9/6/83 (JT-CB945) FW-V-928 Hard packing--burned or overheated; repacked 10 rings.

11/6/78 (WR-25968) FW-V-5A Would not open electrically.

e Failure Data Scurce: Maintenance Request Logbook Work Requests (microfil:n and computer file)

Job Tickets (microfilm and computer file)

O A.1-48 02VJb0619860AR

i I

TMl-1 i COMPONENT FAILURE DATA

SUMMARY

SHEET DESIGNATOR: V1F  !

SYSTEM: EFW-4 l

i

~ COMPONENT TYPE: MOV

{

FAILURE MODE: Fail To Open or Close on' Demand j Site-Speci fic ' Data f e Failure Data for Given Failure Mode: One Failure i

Date Reported Failure Cause j 11/3/76 (WR-17565)EF-V-2A Would not open or close '

electr.ically.  !

l O

t

l.

(

l e Failure Data Source: Maintenance Request Logbook l Work Requests (microfilm and computer file) t Job Tickets (microfilm and computer file) j O  !

i i

1 A.1-49 l 0210G06A986DAR l

THI-l COMPONEfG FAILURE DATA

SUMMARY

SHEET DESIGNATUR: V1F SYSTEM: Makeup COMPONENT TYPE: MOV FAILURE MODE: Fail To Open or Close on Det..and Site-Specific Data e Failure Data for Given Failure Mode: Eight Failures Date Reported Failure Cause 4/17/78 (WR-23bl0) MU-V-160 Breaker trippec upon opening or closing.

7/2b/79 (JT-C14bL) MU-V-25 Blew control fuses--loose wires.

5/6/77 (WR-19852) MU-V-2A Motor tripped--leads pinched.

10/31/74 (WR-6336) MU-V-16 Improper torque setting--would not open from control.

11/14/76 (WR-12440) MU-V-12 Operator could not engage valve.

b/21/76 (WR-lbb25) MU-V-28 Would not close--stem.

b/21/76 (WR-lbbOl) MU-V-1A Operator could not engage valve--

corrected itself.

3/29/77 (WR-19b01) MU-V-1A Would not operate electrically.

e Failere Data Source: Maintenance Request Logbook Work Requests (microfilm and ccmputer file)  :

Job Tickets (microfilm and computer file) i Gi l

  • 1 A.1-60 l 021060619860AR _

, . I i

I

>THI-l  !

COMPONENT FAILURE DATA

SUMMARY

SHEET UESIGNATOR: V1T I SYSTEM: All l COMPONENT TYPE Motor-0perated Valve FAILUdE MODE: Transfer Open or Closed i S_i' . Speci fic Data e Failure Data for Given Failure Mode: Zero Failures '

Date Reported Failure Cause None d

'I O .

3 i

I 2

1 i

e Failure Data Source: Maintenance .*equest Logbook ,

Work Requests (mic;ofilm and computer file) i

? Job Tickets (microfilm and computer file) l O l i

i A.1-51 i 0210GU61986DAR  !

i i

'.._,. ___.-.,.-.-__-__._-.,__....-,._,m._--,---,.....,_,__.-.---~~.--._-~-..----.

TMI-1 COMPONENT FAILURE DATA

SUMMARY

SHEET DESIGNATOR: V3F SYSTEM: All COMP 0NENT TYPE: Air-0perated Valve FAILURE MODE: Fail To Operate on Demand Site-Specific Data e Failure Data for Given Failure Mode: Ten Failures Date Reported Failure Cause 1/12/81 (.1T-C4897) MU-V-3 Valve binding--Iack of lubrication.

2/17/83 (JT-CA226) MU-V-3 Improper alignment of solenoid valves.

1/16/81 (JT-C4949) MU-V-20 Air not bleeding off.

8/21/83 (JT-C8595) MU-V-26 Failed solenoid.

9/6/?7 (WR 21198) MU-V-17 Wou]d not respond on manual.

9/27/77 (WR-21424) MU-V-17 Stuck 70% open.

9/27/77 (WR-21425) FW-V-178 Valve did not stroke.

6/20/83 (JT-CB130) IC-V-3 Stem to packing tight--would not open fully.

6/18/79 (JT-C0526) IC-V-3 Would not open fully.

5/30/79 (JT-C1147) IC-V-3 Opened half-way.

8/20/79 (WR-10775) MS-V-13B Would not open or stay open--

temperature and pressure switches replaced, e Failure Data Source: Maintenance Request Logbook Work lequests (microfilm and computer file) h5 Tickets (microfilm and computer file)

A.1-52 0210G062086DAR

-~ aa. a .+ - .2 =. - a .,aa. -. - -pa . -. s + .

1 THI-1  !

( ~ COMPONENT FAILURE DATA

SUMMARY

SHEEY l

1 UESIGNATOR: V3T -j 3

SYSTEM: Al l

  • l  ?

COMPONENT TYPE: Air-Operated Valves  !

I 4 FAILURE MODE: Transfer Closed or Open During Operation j

'i Site-Specific Data  !

e Failure D4ta for Given Failure Mode: Seven' Failures i Date Reported Failure. - Cause, S/4/82 (JT.-C8476) DC-V-668 Broken air line.  ;

I b/2/79 (JT-C2715)FW-V-17B Would not stay shu*.--air '

controller leaked. l 4/4/83 (JT-CA606)FW-V-17B .

Air leak in air regulator.  !

i 3/3/83 (JT-CA307) FW-V-17A Air leak.  !

b/2/79 (JT-C0945) FW-V-11B Packing leak.

11/b/74 (WR-5408) MS-V-3C Body to bonnet leak caused vacuum  !

leak. i 1/17/76 (WR-13517) MS-V-3C Air leakage. I r

Q e Failure Dati Source: Maintenance Request Logbook Work Requests (microfilm and computer file)

! Jc5 Tickets (microfilm and computer file) l

i i

4 1

, A.1-53  !

~

0210GU61986DAR

, ~. - , -

TM1-1 COMPONENT FAILURE DATA

SUMMARY

SHEET DESIGNATOR: V7F SYSTEM: All (except intermediate closed cooling and river water systems)

COMPONENT TYPE: Check Valve FAILURE MODE: Fail To Operate on Demand Site-Specific Data e Failure Data fo: Given Failure Mode: One Failure Date Reported Failure Cause

?/2/83 (JT-Ct3203) MU-V-95 Did not seat.

O e Failure Data Source: Maintenance Request Logbook Work Requests (microfilm and computer file)

Job Tickets (microfilm and computer file)

A.1-b4 U210G0619860AR

, ~.\ TMI .l.

( j COMPONENT FAILURE DATA

SUMMARY

SHEET n._s DESIGNATOR _: V7FI SYSTEM: Intermediate Closed Cooling COMPONENT TYPE: Check Valve FAILURE MODE: Fai' To Operate on Demand Site-Specific Data e Failure Data for Given Failure Mode: One Failure Date Reported Failure Cause 3/21/77 (WR-19431) IC-V-138 Stuck open.

hG I

l 1

I e Failure Data Source: Maintenance Request Logbook j Work Regrasts (microfilm and computer file) '

f') '

Job Tickets (microfilm and computer file) )

'x./ l l

A.1-55 0210G061986DAR -

l

TMI-1 COMPONENT FAILURE DATA

SUMMARY

SHEET DESIGNATOR: V7FR SYSTEM: River Water Systems COMPONENT TYPE: Check Valve FAILURE MODE: Fail to Operate on Demand Site-Specific Datl e Failure Data for Given Failure Mode: 10 Failures Date Reported Failure Cause 7/29/81 (JT-C6401) DR-V-21B Stuck closed--cleaned.

3/19/79 (JT-C07b9) DR-V-22A Did not close.

4/21/80 (JT-C3004) DR-V-22A Stuck in closed position.

3/13/79 (JT-C0693) DR-V-6B f%de loud noise, replaced disk.

b/4/79 (JT-C1120) DR-V-6A Stuck open--replaced clapper arm.

5/6/81 (JT-Cb939) DR-V-6A Hur.y open--worn internals.

12/12/79 (JT-C21bO) DR-V-6B Valve did not ', eat--internals replaced.

7/16/78 (WR-24559) NR-V-20B Would not close.

7/ 6/7b (WR-lClb4) NR-V-208 Would not close.

9/15/d2 (JT-C9234) NR-V-20A Would not close.

e Failure Data Source: Maintenance Request Lo9 book Work ReqJests (microfilm arid computer file)

Job Tickets (microfilm and computer file)

A.1-56 0210G061986DAR

,. TMI-1 i  ! COMPONENT FAILURE DATA

SUMMARY

SHEET U

DESIGNATUR: V7L SYSTEM: All (except intermediate closed cooling and river water systems)

COMPONENT TYPE: Check Valve FAILURE MODE: Gross Reverse Leakage, Transfer Open Site-Specific Data e Failure Data 'or Given Failure Mode: One Failure Date Reported Failure Cause 12/7/79 (JT-C2127) DH-V-14A Loose disc.

l

% ,)

i e Failure Data Source: Maintenance Request Logbook Work Requests (microfilm and computer file)

Job Tickets (microfilm and computer file)

/~s N]

A.1-57 0210G0bl9860AR

TMI-l COMPONENT FAILURE DATA

SUMMARY

SHEET DESIGNATUR: V7LI SYSTEM: Intermediate Closed Cooling COMPONENT TYPE: Check Valve FAILURE MODE: Gross Reverse Leakage, Transfer Open Site-Specific Data e Failure Data for Given Failure Mode: Six Failures Date Reported Failure Caus1 8/11/83 (JT-CB534) IC-V-13A Leaked allowing reverse rotation of pump.

9/16/83 (JT-CB905) IC-V-13A Worn disc armhole.

12/21/79 (JT-CO336) IC-Y-13A Leaked back--replaced valve.

1/30/80 (JT-C2241) IC-V-13A Leaked at 200 GPM--repaired.

6/12/81 (JT-C6146) IC-V-13A Leaked--replaced valve inner parts.

9/8/81 (JT-C6744) IC-V-13A Leaked through--repaired.

12/11/80 (JT-C4688) IC-V-13A Leaked through.

e Failure Data Source: Maintenance Request Logbook Work Requests (microfilm and computer file)

Job Tickets (microfilm and computer file)

A.1-58 0210GU619860AR

i e

THI-l f

[l V

COMPONENT FAILURE DATA

SUMMARY

SHEET -

i t

DESIGNATOR: .V7LR l SYSTEM: River Water Systems i

COMPONENT TYPE: Check Valve

  • FAILURE MODE: Gross Reverse Leakage Transfer Open  ;

i Site-Specific Data i e Failure Data for Given Failure Mode: One Failure Date Reported Failure Cause 8/4/78 (JT-C0694) NR-V-208 Leaked back through.

i O ,

i  !

4 f

l i

i i i 1

e Failure Data Source: Maintenance Request Logbook l Work Requests (microfilm a'nd computer file) l Job Tickets (microfilm and computer file) .;

5 j

1 l

l A.1-59  !

0210GU619860AR. l l

TMI-l COMPCNENT FAILURE DATA

SUMMARY

SHEET DESIGNATUR: V7T SYSTEM: All (except intermediate closed cooling and river water systems)

COMPONENT TYPE: Check Valve FAILURE MJOE: Plug / Transfer Closed during Operation Site-Specific Data e Failure Data for Given Failure Mode: Zero Failures Date Reported Failure Cause None O

1 l

l l

l l

l e Failure Data Source: Maintenance Request Logbook l Work Requests (microfilm and computer file) l Job Tickets (microfilm and computer file) l l

ei l

i i

A.1-60 l 0210GU61986UAR

i TMI-1 '

COMPONENT FAILURE DATA

SUMMARY

SHEET i

, DESIGNATOR: V7TI f t

4 SYSTEM: Intermediate Closed Cooling COMPONENT TYPF.: Check Valve

{

FAILURE MODE: Plug / Transfer Closed during Operation Site-Specific Data e Failure Data for Given Failure tiode: Zero Failures- i

)

i Date Reported Failure Cause l None .

4 I

i l

i i

i l

O '

l i

! i a l i

I i

3 j

1 4

\

i 4

I e Failure Data Source: Maintenance Request Logbook 1 Work Requests (microfilm and computer filu) l Job Tickets (microfilm and computer file) i 1

'l l

!O 1 1

1 i

A.1-61

  1. U210G0619860AR l l  !

TMI-1 COMPONENT FAILURE DATA

SUMMARY

SHEET h

DESIGNATOR: V7TR SYSTEM: River Water Systems COMPONEt' TYPE: Checx Valve FAILURE MODE: Plug / Transfer Closed during Operation Site-Specific Data e Failure Data for Given Failure Mode: Zero Failures Date Reported Failure Cause None O

e Failure Data Source: Maintenance Request Logbook Work Requests (microfilm and computer file)

Job Tickets (microfilm and computer file) 9 f

A.1-62 0210G061986DAR 1

l l THI-1 COMPONENT FAILURE DATA

SUMMARY

SHEET  !

f DESIGNATOR: V8T- i SYSTEM:- Al1

['

COMPONENT TYPE: Manual Valve l

FAILURE MODE: Transfer Closed Site-Specific Data-e Failure Data for Given Failure Mode: Zero Failures j Date Reported Failure Cause I

1 None I

)

l 4

J O 1 I

r i

i i

~

1 l

I l

l I j

i l l

1

} 1 j e Failure Data Source: Maintenance Request Logbook i j Work Hequests (microfilm and computer file)

Job Tickets (microfilm and computer file)

I l A.1-6 3

0210GU61986DAR
i

TMI-l COMPONENT FAILURE DATA

SUMMARY

SHEET DESIGNATOR: V10FS, V10FW SYSTEM: RCS COMPUNENT TYPE: Code Safety Valve FAILORE MOUE: Fail To Open on Demand Site-Specific Data e Failure Data for Given Failure Mode: Zero Failures Vate Reported Failure Cause None 9

i e Failure Data Source: Maintenance Request Logbook Work Requests (microfilm and computer file)  !

Job Tickets (microfilm and computer file) '

O A.1-64 i 0210G061986DA3

_' TMI-l t

v) COMPONENT FAILURE DATA

SUMMARY

SHEET DESIGNATOR: V10FS, V10FW SYSTEM: Main Steam COMPONENT TYPE: Code Safety Valve FAILURE MODE: Fail To Open on Demand Site-Specific Data e Failure Oa:a for Given Failure Mode: Zero Failures Date Reported Failure Cause None o

k.

{

e Failure Data Source: Maintenance Request Logbook Work Requests (microfilm and computer file)

Job Tickets (microfilm and computer file) t'h,

/

A.1-65 021000619860AR

TMI-1 COMPONENT FAILURE DATA

SUMMARY

SHEET DESIGNATOR: V10RS SYSTEM: RCS COMPONENT TYPE: Code Safety Valve FAILURE MODE: Fail To Reseat After Opening (passing steam)

Site-Specific Data e Failure Data for Given Failure Mode: Zero Failures Date Reported Failure Cause None O

e Failure Data Source: Maintenance Request Logbook Work Requests (microfilm and computer file)

Job Tickets (microfilm and computer file) 9 A.1-66 U210G061986DAR

~ _ _ _ . . . - - . - . . . - . - . . . - _. -- - . ._- -. - - . . . - ~ - . - . .

l TM1-1 l COMPONENT FAILURE DATA

SUMMARY

SHEET

{

DESIGNATUR: V10RS .;

-i

, SYSTEM: Main Steam j i

l COMPONENT TYPE: Code Safety Valve- l l

FAILURE MODE: Fail To Reseat After Opening (passing steam) ,

Site-Specific Data  !

l >

e Failure Data for Given Failure Mode:' Zero Failures l Date Reported Failure Cause 4

None i f

f i

l O '

\

i 4

f i

4 a

i l

L 4

e Failure Data Source: Maintenance Request Logbook Work Requests (microfilm and computer file)

Job Tickets (microfilm and computer file) l l

2

A.1-67 0210G0bl9860AR

TMI-l COMPONENT FAILURE DATA

SUMMARY

SHEET DESIGNATOR: V11FS, V11FW SYSTEM: Pressurizer, Main Stea'n COMPONENT TYPE: PORV FAILURE MODE: Fail To Open Site-Specific Data e Failure Data for Given Failure Mode: Zero Failures Date Reported Failure Cause None O

e Failure Data Source: Maintenance Request Logbook Work Requests (microfilm and coaputer file)

Job Tickets (microfilm ano computer file) 9 A.1-6B 0210GU61986DAR

,_ THI-1 COMPONENT FAILURE DATA

SUMMARY

SHEET DESIGNATOR: V11RS, V11RW SYSTEM: Pressurizer, Main Steam <

COMPONENT TYPE: PORV FAILURE MODE: Fail To Reseat on Demand Site-Specific Data e Failure Data for Given Failure Mode: Zero Failures Date Reported Failure Cause None

/%

b ,

e Failure Data Source: Maintenance Request Logbook Work Requests (microfilm and computer file)

Job Tickets (microfilm and computer file) r~'s

'N-A.1-b9 0210606198 BOAR

THI-1 COMPONENT FAILURE DATA

SUMMARY

SHEET ll DESIGNATUR: X1F SYSTEM: Electric Power COMPUNENT TYPE: Main Transformer FAILURE MODE: Fail During Operation Site-Specific Data e Failure Data for Given Failure Mode: Zero Failures Date Reported Failure Cause None G

i i

e Failure Data Source: Maintenance Request Logbook Work Requests (microfilm and computer file) '

Job Tickets (microfilm and computer file) 9 A.1-70 0210G0619860AR

l THI-1 l,, ) COMPONENT FAILURE DATA

SUMMARY

SHEET  :

\_/ )

l l'

UESIGNATOR: X2F i

SYSTEM: Electric Po .er l COMPONENT TYPE: Station Trar 'ormer i

FAILURE MD_DE: Fail During Opera 'n l

Site-Specific Data e Failure Data for Given Failure Mode: Zero Failures I Date Heported Failure Cause None i

7s

\

(G I

i l

I l

1 e Failure Data Source: Maintenance Request Logbook ,

{

Work Requests (microfilm and computer file) l Job Tickets (microfilm and computer file)

A.1-71 i U'd10G061986DAR

a 9

)

O

\

APPENDIX A.2 COMPONENT SUCCESS DATA

SUMMARY

SHEETS 1

O I 0211GU619860AR .

i TMl-1 COMPONENT SUCCESS DATA

SUMMARY

SHEET f l

UESIGNATOR: ACR SYSTEM: Instrument Air COMPONENT TYPE: Compressor FAILURE MODE: Fail During Operation One of two compressors is operating at all times. This leads to a total of 8.61 x 104 hours0.0012 days <br />0.0289 hours <br />1.719577e-4 weeks <br />3.9572e-5 months <br /> of operation for the two compressors.

O .

l O

A.2-1 0211G0619860AR

THI-l l COMPONENT SUCCESS DATA

SUMMARY

SHEET I l

DESIGNATOR: AD1 l

l SYSTEM: Compressed Air System COMPONENT TYPE: Air Dryer FAILURE MODE: Fail During Operation One of two air dryers operate at all times. This results in 8.61 x 104 hours0.0012 days <br />0.0289 hours <br />1.719577e-4 weeks <br />3.9572e-5 months <br /> of operation for the two air dryers.

O l

9 A.2-2 0211G061986DAR

i TMI-l i COMPONENT SUCCESS DATA

SUMMARY

SHEET,  ;

DESIGNATOR: BC SYSTEM: Electric Power COMPONENT TYPE: Battery Charger FAILURE MODE: Failure During Operation i

Number of Chargers: 6 '

Number of Chargers on Line: 4 l Number of Hours per Charger: 8.61 x 104 i Total Hours: 4 x 8.61 x 104 = 3.44 x 106 l

O l l

O i A.2-3 0211G061986DAR

I TMI-l

~

COMPONENT SUCCESS DATA SUltiARY SHEET DESIGNATOR: BTO SYSTEM: Electric Power COMPONENT TYPE: Battery FAILURE MODE: Fail During Operation Number of Batteries: 2 Number of Hours per Battery: 8.61 x 104 Total Hours: 2(8.61 x 104 ) = 1.72 x 105 l

r 9

O A.2-4 02116061986DAR

y -. _

THI-l  :

COMPONENT $UCCESS DATA

SUMMARY

SHEET 5

DESIGNATOR: CB4F0 SYSTEM: RPS >

COMPONENT TYPE: Heactor Trip Breaker l FAILURE MODE: Fails To Open on Demand Number of Demands Number of Total Test per Test Tests Demands s 1303-4.1 6 140 840  ;

There were 6 reactor trips during commercial operation of the unit, ,

resulting in 36 breaker demands. Total number of breaker demands is then 876.

I l

i i

r A.2-5 0211G0619860AR i

TMI-l COMPONENT SUCCESS DATA

SUMMARY

SHEET j DESIGNATOR: CRF S YSTEM: Reactor Protection System COMPONENT TYPE: Single Control Rod Assembly FAILURE MODE: Fail To Insert on Demand There have been six trips and four tests. The number of assemblies challenged in each demand is 61. Therefore, total number of demands is 10 x 61 = 610.

9 A.2-6 0211G061986DAR

TMI-l COMPONENT SUCCESS DATA

SUMMARY

SHEET DESIGNATOR,: DGS SYSTEM: Emergency Diesel Generator Services COMPONENT TYPE: Diesel Generator t FAILURE MODE: Fail To Start on Demand Number of Demands Number of Total i Test per Test Tests Demands 1303-11.10 8 8 64 1302-5.30 2 7 14 1303-4.16 1 377 377 1303-5.2 18 23 414 .

869 ,

O e

I i

I O

A.2-7 0211G061986DAR

TMI-l COMPONENT SUCCESS DATA

SUMMARY

SHEET DESIGNATOR: DGR1 SYSTEM: Emergency Diesel Generator Services COMPONENT TYPE: Diesel Generator FAILURE MODE: Fail During Operation Number of Hours Number of Total Run Test Run Time per Test Tests Time-Hours 1303-11.10 0.67 8 11 1302-S.30 0.60 7 4 1303-4.16 0.25 377 . 94 1303-d.2 4.$U 23 104 213 0

l l

l l

9 A.2-8 0211G061986DAR

,,_ THI-l (v) COMPONENT SUCCESS DATA

SUMMARY

SHEET DESIGNATOR: FlR SYSTEM: All Air Handling Systems COMPONENT TYPE: HVAC Fan FAILURE MODE: Fail During Operation One train of control room air handling with three fans is continuously running during unit operation and shutdown. This run time is much greater than thein time calculated successful run this way is times 2.58 duging scheduled tests. Total run x 10 O

L) .

)

i O

LJ A.2-9 0211G0619860AR

TMI-1 COMPONENT SUCCESS DATA

SUMMARY

SHEET DESIGNATOR: FIS SYSTEM: All Air Handling Systems COMPONENT TYPE: HVAC Fan FAILURE MODE: Fails To Start on Demand Number of Demands Number of Total Test per Test Tests Demands 1300-3N* 1 44 44 1303-6.6 12 71 852 1303-11.13 6 6 36 932 O

  • lt is assumed for this test tnat the standby train is started, and the operating train is stopped and lef t in this configuration until the next test.

A.2-10 0211G061'J86DAR

TM1-1 COMPONENT SUCCESS DATA

SUMMARY

SHEET

( )

DESIGNATOR: HXP, HXR SYSTEM: All COMPONENT TYPE: Heat Exchanger FAILURE MODE: Leak Rupture, Plugs Number of Number of Heat Successful System Exchangers Hours Time Span Total Hours Decay Heat 2 8.61 x 104 One train in use 1.72 x 106 Closed during cold shut-Cooling down.

System Decay Heat 2 8.61 x 104 One train in use 1,72 x 105 Removal during cold shut-System down.

Nuclear 4 8.61 x 104 Two trains in use 2.58 x 105 (A

Services River Water during cold shut-down.

System Nuclear 6 8.61 x 104 Six room coolers 4.90 x 105 Services in continuous Closed operation.

Cooling System Reactor 3 8.61 x 104 Last test June 2.58 x 106 l Building 1984.

Emergency l Cooling l System Intermediate 2 3.18 x 104 One heat exchanger 6.36 x 104 Cooling during power  ;

System operation.

Chilled 2 8.61 x 104 One in use 1.72 x 106 Water continuously.

System n

v 1.59 x 106 I

A.2-11 0211G061986DAR

TMI-1 COMPONENT SUCCESS DATA

SUMMARY

SHEET h

DESIGNATOR: P1R SYSTEN: Nuclear Services Closed Cooling COMPONENT TYPE: Nuclear Services Closed Cooling Pump FAILURE MODE: Fail During Operation Two pumps are operating continuously during normal unit operation, and one pump is operating during cold shutdown. Total pump run time is '

1.18 x 103 hours0.00119 days <br />0.0286 hours <br />1.703042e-4 weeks <br />3.91915e-5 months <br />.

/

O l

9 A.2-12 0211G0619860AR

TMI-1 COMPONENT SUCCESS DATA

SUMMARY

SHEET

(

DESIGNATOR: PlR SYSTEM: Makeup l

COMPONENT TYPE: Normally Operating Makeup Pump FAILURE MUDE: Fail During Operation One pump is operating continuously during unit power operation. It is assumed that this is the B makeup pump for most of the time. There is not sufficient information on the operation of makeup pumps during cold ,

shutdown. When on total of 3.18 x 10{y the runistime hours during However, obtained. power operation is has run meter considered recordeda 2.21 x 104 hours0.0012 days <br />0.0289 hours <br />1.719577e-4 weeks <br />3.9572e-5 months <br /> of run time for the B pump between November 1,1976 and July 2, 1984. ExtrapolationtotheperiodSeptegber2,1974,througn November 1,1976, yields an additional 1.59 x 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> of operation for pump B. The total hours is then 2.21 x 104 + 1.59 x 104 = 3.8 x 104 hours0.0012 days <br />0.0289 hours <br />1.719577e-4 weeks <br />3.9572e-5 months <br />.

1 O

l f

A.2-13 U211GU619860AR I i

TMI-1 COMPONENT SUCCESS DATA

SUMMARY

SHEET DESIGNATOR: PlS SYSTEM: Nuclear Services Closed Cooling COMPONENT TYPE: huclear Services Closed Cooling Pump FAILURE MODE: Fails To Start on Demand Number of Demands Number of i3tal Test per Test Tests Demands 1303-11.10 6 8 48 1300-4E 3 5 lb 1300-3J 3 40 120 1303-6.2 6 23 138 321 One pump is operating continuously during cold shutdown, and two pumps are operating during power operation of the unit. This requires one pump to start every time the unit is started from cold shutdown. Total number of demands is then 337.

O 1

0' A.2-14 0211606198bDAR

TMI-1 g COMPONENT SUCCESS DATA

SUMMARY

SHEET DESIGNATOR: P1S

, SYSTEM: Makeup COMPONENT TYPE: Normally Operating Makeup Pump FAILURE MODE: Fails To Start on Demand Number of Demands Number of Total Test per Test Tests Demands 1303-11.10 1 8 8 1300-3H A/B 1 30 30 1303-11.27 1 2 2 40 The 8 makeup pump is started every time the unit is started up from cold shutdown. There have been 16 startups from cold shutdown. Total number of demands on the B makeup pump is then b6.

O f

O A.2-15 0211G0619860AR

TMI-l COMPONENT SUCCESS DATA

SUMMARY

SHEET DESIGNATOR: P2R SYSTEM: Decay Heat Removal COMPONENT TYPE: Decay Heat Removal Pump FAILURE MUDE: Fail During Operation One train of decay heat removal system is in use continuously during cold shutdown. This time far exceeds the run time of the pumps during the scheduled testing. Total run time is 6.43 x 104 hours0.0012 days <br />0.0289 hours <br />1.719577e-4 weeks <br />3.9572e-5 months <br />.

O O

A.2-16 0211G0619860AR

TMI-1

( ) COMPortENT SUCCESS DATA

SUMMARY

SHEET U

UESIGNATOR: P2R I

SYSTEM: Makeup COMPONENT TYPE: Standby Makeup Pumps (A and C)

FAILURE MODE: Fail During Operation Run Hours Number Total Run Test per Test of Tests Hours 1303-11.10 2.0 8 16 1300-3HA 0.50 13 7 l 1300-3HB 1.17 17 20 1303-11.8 1.0 7 7 1303-5.2 2.0 23 46 1303-11.27 2.0 2 4 100 Run meters have recorded 1.13 x 103 hours0.00119 days <br />0.0286 hours <br />1.703042e-4 weeks <br />3.91915e-5 months <br /> of run time on the A and C makeup pumps between November 1, 1976, and July 2, 1984. Extrapolation p of the run time to the period Sept, ember 21, 1974, through November 1, t

) 1976, yields a total of 3.79 x 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> for these pumps. This value was used in the quant'.fication. .

L l

A.2-17 0211G0bl9860AR

TMI-1 COMPONENT SUCCESS DATA

SUMMARY

SHEET DESIGNATOR: P2R

.S_YSTEM: Decay Heat Closed Cooling COMPONENT TYPE: Decay Heat Closed Cooling Pump FAILURE MODE: Fail During Operation One train of decay heat closed cooling system is in use continuously during cold shutdown. This time exceeds the run time of the pumps during the scheduled testing. Total run time is b.43 x 104hours.

O O

A.2-18 0211GU619860AR

TMl-1 COMPONENT SUCCESS DATA

SUMMARY

SHEET OESIGNATOR,: P2R SYSTEM: Reactor Building Spray COMPONENT TYPE: Reactor Building Spray Pump FAILURE MODE: Fail During Operation Run Hours Number Total Number Test per Test of Tests of Hours 1300-3A A/B 0.50 32 16 (pump test) 1300-3A A/B 1.17 11 13 (pump and bearing cest) 1303-S.2 1.50 23 36 1303-11.10 1.00 8 8 1303-11.60 2.0 5 10 82*.0 g Run meters have recorded 9.79 x 101 hours0.00117 days <br />0.0281 hours <br />1.669974e-4 weeks <br />3.84305e-5 months <br /> on both pumps between November 1, 1976, and July 2, 1984. Extrapolation of the run time to the y

period September 2, 1974, through November 1, 1976, yields a total of 121.3 operating hours. This value was used in the analysis.

1 1

l l

()

o i

A.2-19 0211GU619860AR ,

j

TMI-1 COMPONENT SUCCESS DATA

SUMMARY

SHEET OESIGNATOR: P2R SYSTEM: Emergency Feedwater COMPONENT TYPE: Motor-Driven Emergency Feedwater Pump FAILURE MODE: Fail Ouring Operation Run Hours Number Total Number Test per Test of Tests of Hours 1303-11.10 1.00 5* b 1300-3FA/B 0.67 32 21 1300-3FB 1.33 20 27 1300-11.42 1.00 3 3 1300-11.39 U.67 10 l 63 Run meters recorded 200.00 hours0 days <br />0 hours <br />0 weeks <br />0 months <br /> run time between November 1,1976 and July 2, 1984 Extrapolation of run time to the period September 2, 1974, through November 1,1976, yields a total of 248 hours0.00287 days <br />0.0689 hours <br />4.100529e-4 weeks <br />9.4364e-5 months <br /> of operation. This value was used in the analysis.

  • Eight tests have been conducted, but only five include starting motor-driven emergency feedwater pumps.

A.2-20 0211G061986DAR

TMI-l I%.)\ COMPONENT SUCCESS DATA

SUMMARY

SHEET UESIGNATOR: P2S f

SYSTEM: Makeup  ;

COMPONENT TYPE: Standby Makeup Pumps (A and C)  :

FAILURE MODE: Fails To Start on Demand  !

Number of Demands Number of Total Number i Test per Test Tests of Demands j 1303-11.10 8 8 64  !

1300-3H A/B 2 30 60 1303-11.8 4 7 28  :

1303-5.2 8 23 184 1303-11.27 2 2 4 i 340  !

i

( -

. i l

1 L

l 1

A.2-21 l

0211G06198bDAR

s i

THI-1 COMPONENT SUCCESS DATA

SUMMARY

SHEET DESIGNATOR: P2S SYSTEM: Decay Heat Removal COMPONENT TYPE: Decay Heat Removal Pump FAILURE MODE: Fails To Start on Demand Number of Demands Number of Total Test per Test Tests Demands 1303-11.10 6 8 48 1303-11.b4 2 2 4 1303-11.27 2 2 4 1300-38 A/B 2 69 138 1300-5.2 8 23 184 338 The decay heat removal system is put into operation at each cold shutdown. There nave been 16 cold shutdowns at this plant. Alternating the operation of the two trains is assumed to be carried out at th3 time the tests are performed. Total number of demands is then 354 9

O A.2-22 0211GU619860AR

TMI-1

()

COMPONENT SUCCESS DATA

SUMMARY

SHEET DESIGNATOR: P2S SYSTEM: Decay Heat Closed Cooling COMPONENT TYPE: Decay Heat Closed Cooling Pump FAILURE MOVE: Fails To Start on Demand Number of Demands Number of Total Number Test per Test Tests of Demands i 1303-11.10 6 8 48 1303-11.50 2 5 10 1300-3A A/B 2 43 86 1303-11.54 2 2 4 1300-38 A/B 2 69 138 1300-3C 2 41 82 1300-3H A/B 2 30 60 1303-11.8 4 7 28 1303-5.2 6 23 138 694

()

One train of decay heat closed cooling is started for operation during cold shutdowns. There have been 16 cold shutdowns of the unit. Total -

number of demands is then 604.

i I

t LJ

^* ~23 0211GU61986DAR

. TMI-l 20MPONENTSUCCE'[SDATA

SUMMARY

SHEET DESIGNATUR: P2S SYSTEM: Reactor Building Spray COMPONENT TYPE: Reactor Building Spray Pump FAILURE MODE: Fail To Start on Demand Number of Demands Number of Total Number Test per Test Tests of Oemands 1300-3A A/B 2 43 86 1303-5.2 6 23 138 1303-11.10 6 8 48 1303-11.50 2 S J 282 9

9 A.2-24 0211G0619860AR

TMl-1

.O COMPONENT SUCCESS DATA

SUMMARY

StiEET

\

UESIGNATOR: P2S SYSTEM: Emergency Feedwater COMPONENT TYPE: Motor-Driven Emergency Feedwater Pump FA! LURE MODE: Fails To Start on Demand Number of Demands Number of Total Number Test per Test Tests of Demands 1303-11.10 6 5* 30 1300,3F A/B 2 37 74 1300-11.42 3 3 9 1300-11.39 4 10 J 163 l

  • Eight tests have been conducted, but only five include starting motor-driven emergency f eedwater pumps.

A.2-2b 0211G061986DAR

THI-l

_ COMPONENT SUCCESS DATA

SUMMARY

SHEET DESIGNATOR: P3R SYSTEM: Main Feedwater COMPONENT TYPE: Main Feedwater Pump FAILURE MODE: Fail During Operation The feedwater system is not in operation during cold shutdown. It is operating at every other time. The number of hours of operating time for each feedwater component is 3.18 x 104 hours0.0012 days <br />0.0289 hours <br />1.719577e-4 weeks <br />3.9572e-5 months <br />. Two main feedwater pumps are operated for a major portion of tne time. Total "' time for main feedwater pump is then 6.36 x 104hours.

9 G

A.2-26 0211G061986DAR

TMI-1 COMPONENT SUCCESS DATA

SUMMARY

'SHEEl DESIGNATOR: P3ER SYSTEM: Emergency Feedwater COMPONENT TYPE: ' Turbine-Oriven Emergency-Feedwater System

. FAILURE MODE: Fail During Operation Run Time Nuinber of Total Run Test per Test Tests Time-Hours 1300-11.42 0.25 3 1 1303-11.39 0.5 10 5 1303-3G A 0.25 26' 7 3300-3G B 0.58 35 20, 33

=

(::)

I O

A.2-27 0211G061986DAR I

TMI-1 COMPONENT SUCCESS DATA

SUMMARY

SHEET DESIGNATUR: P3S ESTEf1: Main Feedwater COMPONENT TYPE: Main Feedwater Pump FAILURE MODE: Fails To Start on Demand The number of demands on the feedwater system are based on an estimated system startup rate of 10 per calendar year in addition to 16 startups of the unit from cold shutdown. Total number of demands = 60. Two main feedwater pumps are operated for a major portion of the time the unit is operating. Total number of demands on the main feedwater pump is then 120.

O i

l l

9l I

A.2-28  ;

0211G061986DAR 1

l I

TMI-l -

COMPONENT SUCCESS DATA

SUMMARY

SHEET DESIGNATOR: P3ES

'I SYSTEM: Emergency Feedwater  !

)

COMPONENT TYPE: Turbine-Oriven Emergency Feedwater System j i

FAILURE MODE: Fails To Start on Demand  !

Number of Demands Number of Total Number j Test per Test Tests- of Demands .

i 1300-11.42 1 3 3 l 1303-11.39 2 'O 29 1 1300-3G A 1 46 26 I 1300-3G B 2 35 70 l 119 i

o  ;

i i

i i

I i

i l

l l

t O  :

A.2-29 '

0211GU61986DAR

. r. ee w-er t +eew m.rr w ewne- e e worwew-e+=ws -e, wr-t o w +-=ew e w c r.= w-w+ ,ne- w e rwe+-*

l TMI-1 COMPONENT SUCCESS DATA

SUMMARY

SHEET DESIGNATOR: P4R SYSTEM: Nuclear Services River Water COMPONENT TYPE: Nuclear Services River Water Pump FAILURE MODE: Fail During Operation Two pumps are operating continuously during normal unit operation, and one pump {s operating during cold snutdown. Total pump run time is then 1.18 x 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br />.

O 1

l l

l i

9 l

0211G0619860AR

i

~

TMI-1 i COMPONENT Sl!CCESS DATA

SUMMARY

SHEET i

DESIGNATOR: P4S t i

SYSTEM: Nuclear Services River Water-COMPONENT TYPE: Nuclear. Services River Water-Pump' l r

FAILURE MODE: Fails To Start on Demand

[

Number of Demands Number of Total Number  !

, Test per Test Tests of-Demands  ;

1303-11.10 6 8 48 t

1300-4C A/B 3 5 15 1300-31 A/ 6 3 71 213 -

1303-5.2 6 23 138 k 414 I One pump .is operating continuously during cold shutdown, and two pumps  ;

are operating during power. operation of the unit. This requires one pump  ;

to start every time the unit is started from cold shutdown. Total number i

of demands is then 430. f i

I O  !

I e

t i

1 l

i l

O A.2-31 0211G061986DAR

TMI-l COMPONENT SUCCESS DATA

SUMMARY

SHEET UESIGNATOR: PSR SYSTEM: Decay Heat River Water COMPONEfiT TYPE: Decay Heat River War., Pump FAILURE MOUE: Fail During Operation One train of the decay heat river water system is continuously in use during cold shutdown. This run time exceeds the run time fgr pumps during scheduled testing. Total run time is then S.43 x 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br />.

O l

l l

l I

l l

l l

l A.2-32 0211G0619860AR

I 4

i I

TMI-1  !

( COMPONENT SUCCESS DATA

SUMMARY

SHEET DESIGNATOR: P5R l

SYSTEM: Reactor Building River Water '

i COMPONENT TYPE: Reactor Building River Water Pump.  ;

FAILURE MODE: Fail During Operation Number of Hours Number of Total Number-Test per Test Tests of Run Hours {

1300-3k A 0.50- 30 lb 1300-3k B 1.17 41. 48 i 1300-11.9 0.67 7 5  !

1303-5.2 1.50 23 35  !

' 1303-11.10 1.00 8 8 f 111 Run meters have recorded 210 hours0.00243 days <br />0.0583 hours <br />3.472222e-4 weeks <br />7.9905e-5 months <br /> of pump run time between -

l November 1, 1976, and July 2, 1984. Extrapolation of the run time to the t period September 2, 1974, through November 1, 1976, yields a total of i 249 hours0.00288 days <br />0.0692 hours <br />4.117063e-4 weeks <br />9.47445e-5 months <br /> of operation. This value was used in the analysis. i O  !

I i

l 1

i i

i l

I I

l A.2-33 1 0211G0619860AR ,

...,e

TMI-1 COMPONENT SUCCESS DATA

SUMMARY

SHEET DESIGNATOR: PSS SYSTEM: Decay Heat River Water COMPUNENT TYPE: Decay Heat River Water Pump FAILURE MODE: Fails To Start on Demand Number of Demands Number of Total Number Test per Test Tests of Demands 1303-11.10 6 8 48 1303-11.60 2 b 10 1300-3A A/B 2 43 86 1303-11.54 2 2 4 1300-3D 2 48 96 1300-3HA/B 2 30 60 1303-11.8 2 7 14 1303-b.? 6 23 138 456 The decay heat river water system is started up at cold shutdown together with decay heat removal system. There have been 16 cold shutdowns at the unit. Total number of demands is then 472.

O A.2-34 0211G0619860AR

l s

I i

TMI-1  !

.( ) COMPONENT SUCCESS DATA

SUMMARY

SHEET I

UESIGNATUR: PbS li SYSTEM: Reactor Building River. Water .

COMPONENT TYPE: Reactor Building River Water Pump 'j FAILURE MODE: Fails To Start on Demand Number-of Demands Number of Total' Number Test per Test ' Tests of Demands ,

1300-3k A/B 2 71 142 1303-11.9 2 7 14 1303-b.2 6 23 138 1303-11.10 6 8 48 342 l

O .

' i s

I f

t k

i A.2-35 0211G0619860AR

- _ . . . - _ _ . . _ . . _ . - . _ _ - _ _ _ _ . . _ . . _ _ _ _ _ _ . . . _ . _ . _ _ _ _ . . . _ ~ _ _ _

TMI-1 COMPONENT SUCCESS DATA

SUMMARY

SHEET DESIGNATOR: SF SYSTEM: All COMPONENT TYPE: Screen, Filter FAILURE MODE: Plug During Operation Number of Number of Operating Time Span Total System Screens Hours Description Hours Decay Heat 2 8.61 x 104 One train in 1.72 x 105 River Water use during System cold shutdown.

Nuclear 3 8.61 x 104 One train in 2.58 x 105 Services use during River Water cold System shutdown.

Reactor 2 8.61 x 104 Last test 1.72 x 105 Building verified River screens Water on June Systems 1984 6.02 x 105 O

A.2-36 0211G062086DAR

l TMI-1 l h)

%.J COMPONENT SUCCESS DATA

SUMMARY

SHEET l l

l UESIGNATOR: TR SYSTEM: All COMPONENT TYPE: Surge Tank FAILURE MODE: Ruptures Number Number of Successful of Hours (power operation System Tanks and cold shutdown) Total Hours Decay Heat 2 8.61 x 104 1.72 x 105 l Closed Cycle '

System Reactor Coolant 2 8.61 x 104 1.72 x 105 System l

Intermediate Closed 1 8.61 x 104 8.61 x 104 Loop Cooling System rw Nuclear Services 1 8.61 x 104 8.61 x 104 t] Closed Cooling System i Emergency Feedwater 2 8.61 x 104 1.72 x 106 System 6.89 x 106 A.2-37 0211GU61986DAR

TMI-1 CCMPONENT SUCCESS DATA

SUMMARY

SHEET DESIGNATOR: VCR SYSTEM: Chilled Water COMPONENT TYPE: Chiller FAILURE MODE: Fails During Operation One train of the chilled water system is continuously in operation during power operation and cold shutdown. Total operating hours is then 8.61 x 10 4.

9.

O A.2-38 021160bl986DAR

. _ _ _ _ - _ - - . - . . . . - - - . - - . . - . . - - . - . ~ . . - . - - - . . - . - . _ . . . _ _ . . .

.i

. ~ TMI-1

' (, '. COMPONENT SUCCESSI UATA

SUMMARY

SHEET '!

DESIGNATOR: VCS ,

SYSTEM: Chilled Water -

COMPONENT TYPE: Chiller  !

I FAILURE MODE: Fails To Start on Demand  !

Number of Demands Number of Total i Test , per Test Tests Demands l

'1300-3N* 1 44 44  !

1300-5.b* 'l 71, 71 115' l In addition,' the chillers are alternated every other week._ This adds {

another 26. starts for every year.. For the 10-year period covered by the i data collection, the total number of demand is 260 + 11b = 375. ,

+

I l

I i

i I

1 1

1 i

6 i

l i

l l

l

  • For the tests, the standby chiller is started, and the operating pump is  !

stopped. l t

A.2-39 0211G061986DAR l

TMI-1 COMPONENT SUCCESS DATA

SUMMARY

SHEET g UESIGNATOR: V1F SYSTEM: All Systems COMPONENT TYPE: Motor-Operated Valve FAILURE MODE: Fail To Open on Demand N . :3er Number Total of Demands of Number of Test / System Per Test Tests Oemands Other Demands 1300-3P 2 10 20 1303-11.50 2 5 10 1300-3A A/8 2 43 86 1300-11.54 2 2 4 1302-6.16 1 3 3 1300-30 2 48 96 Plus 16 from Unit Shutdowns 1303-11.42 4 3 12 1303-11.39 4 10 40 1300-3 G A 3 26 1300-3 G B 4 36 218 1300-3H A/B 7 17 179 1303-1127 10 2 20 1303-11.8 12 7 84 1303-11.21 2 8 16 1300-3R 15 20 300 1300-4C A/B 3 5 15 1300-3C A/B 3 71 213 Plus 32 from Unit Startups 1300-3k A/B 7 71 287 1303-11.9 8 7 56 1 1300-3y 4 40 160 1303-11.10 20 8 160 1303-11.26 5 24 120 1305-5.2 84 23 1,932 Decay Heat -- -- -- 48 from Unit Removal System Shutdowns Decay Heat River -- -- -- 16 from Unit Water System Shutdowns Total Number of Demands = 4.43 x 10 3 l

l 9l A.2-40 0211GU61986DAR

TMI-1 O COMPONENT SUCCESS DATA

SUMMARY

SHEET U

DESIGNATOR: V1F

{

SYSTEM: All Systems  ;

COMPONENT TYPE: Motor-Operated Valve l l

FAILORE MODE: Fail To Close on Demand ,

Number Number Total *

.( Demands of Number of i Test / System Per Test Tests Demands Other Demands

.1300-31 2 10 20 1300-11.60 l 2 5 10 1300-3A A/B 2 43 86 1300-11.54 2 2 4 1302-6.16 1 3 3

.1300-3D 2 48 96 Plus 16 from Unit Startups 1303-11.42 4 3 12 1303-11.39 4 10 40 l 1300-3G A 3 26  ;

1300 3G B 4 35 218 )

O' 1300 3H A/B 1303-11.27 17 10 17 2

179 20 .

1303-11.8 12 7 84  !

1303-11-21 2 8 16

.1300-3R lb 20 300 I 1300-4C A/B 3 5 15 -

1300-3L A/B 3 71 213 Plus 32 from Unit

~'

Shutdowns

  • 1300-3k A/B 7 71 287 -

1300-11.9 18 7 56 i

1300-3Q 4 40 160 1303-11.10 20 8 160 i 1303-11.26 5 24 120 1306-5.2 84 23 1,932 Decay Heat Removal 48 from Unit System Startups Decay Heat River 16 from Unit  !

Water System Startups Total Number of Demands = 4.43 x 103 O  !

1 A.2-41 0211G0619860AR

_ _.._ _, _ . . - _ . . . . _ . . _ . , . _ _ _ . _ _ _ . __._.___-._.-.__._._J

TMI-1 COMPONENT SUCCESS DATA

SUMMARY

SHEET DESIGNATUR: V1T SYSTEM: All Systems COMPONENT TYPE: Motor-Uperated Valve FAILURE MODE: Transfers Closed Number Number of of System Last Verified Valve Total Valve System Valves Tested / Operated Hours Hours Main Steam 4 2/79 3.18 x 104 1.27 x 105 Makeup 14 2/79 3.18 x 10 4 4.45 x 10 6 Reactor 2 S/84 8.61 x 104 1.72 x 106 Building Spray Decay Heat 1 Open During 5.43 x 104 5.43 x 104 River Water Shutdown Decay Heat 4 Open During 5.43 x 104 2.17 x 105 Removal Shutdown Emergency b 6/84 8.61 x 104 4.31 x 105 Feedwater Nuclear 13 Power Operation 3.18 x 10 4 7.94 x 106 Services River Water 7 Cold Shutdown 5.43 x 10 4 Nuclear 3 2/79 3.18 x 106 9.56 x 106 Services Closed Cooling Reactor 1 2/79 3/18 x 10 4 3.18 x 104 Coolant Inter- S 2/79 3.18 x 10 4 1.39 x 105 mediate Closed Cooling Feedwater 18 2/79 3.18 x 10 4 5.72 x 105 Reactor 3 5/84 8.16 x 104 2.58 x 106 Building River Water Total Valve Hours = 3.36 x 106 A.2-42 0211GV619860AR

i TMI-1 COMPONENT SUCCESS DATA

SUMMARY

S'HEET ,

DESIGilATOR: V1T SYSTEM: All Systems j l

COMPONENT TYPE: Motor-Operated Valve FAILURE MODE: Transfers Open "

Number Number of l

'of System Last. Verified Valve Total Valve  !

Test / System Valves , Tested / Operated , Hours Hours

{

1303-11.27 8 8/83 3.93 x 104 3.14 x 105 i Makeup System  :

and Decay  !

Heat Removal System i

'i 1300-3A A/B 2 5/84 8.61 x 104 1.72 x 105  !

Reactor '

Building  !

Spray i Nuclear 4 Power Operation 3.18 x 104 5.62 x 105 _

Services t River Water 8 Shutdown Operation b.43 x 104-  !

Intermediate 1 2/79 3.18 x 104 3.18 x 104 f Cooling '

System  !

Main Steam 2 2/79 3.18 x.104 6.36 x 105 j System j Total Valve Hours = 1.lb x 106 Note: Although there are many normally closed MOVs at THI-1, only a few -

are verified as staying closed.

i O .

A.2-43  ;

U211G0619860AR ~i j

THI-1 COMPONENT SUCCESS DATA

SUMMARY

SHEET ll l UESIGNATUR: V3F SYSTEM: All Systems COMPONENT TYPE: Air-0perated Valve FAILURE MODE: Fail To Open on Demand Number of Demands Number of Total Number Test per Test Tests of Demands 1300-3Q 5 40 200 1303-S.2 6 23 138 1303-11.42 3 3 9 1300-3G A/B 2 61 122 1303-11.9 2 7 14 1303-11.10 6 8 48 Feedwater System 9 60 590 Startups*

1303-1126 12 24 288 1,359 0

  • There have been 60 feedwater startups, based on an estimate of 10 per calendar year plus 16 cold shutdowns.

ll A.2-44 0211G061986DAR

. THI-1  !

COMPONENT SUCCESS DATA

SUMMARY

SHEET DESIGNATOR: V3F i

SYSTEM: Al1 Systems -j i

COMPONENT TYPE: Air-Uperated Valve l t

FAILURE MODE: Fails To Close on Demand l

Number of Demands Number of Total Numb'er  !

Test per Test Tests of Demands  !

l

-1300-3Q b 40 200 l 1303-5.2 6 23 138 l 1303-11.42 3 3 9  ;

1300-3G A/B 2 61 122 1303-11.9 2 7 14 1303-11.10 6 8 48  :

Feedwater System 9 60 590 l Startups* j 1303-11.26 12 24 288  ;

1,359 l i

O l l

l l

l l

A l

d l

l l

l

  • There have been 60 feedwater startups, based on an estimate of 10 per calendar year plus 16 cold shutdowns.  !

A.2-45 0211G0619860AR l

TMI-1 COMPONENT SUCCE.O DATA

SUMMARY

SHEET g DESIGNATOR: V3T SYSTEM: All Systems COMPONENT TYPE: Air-Operated Valve FAILURE MODE: Transfers Closed System Last Number of Total Number Number of Operated / Verified Valve of Valve System Test Valves Tested Hours Hours Makeup System 5 2/79 3.18 x 104 1.59 x 105 Nuclear 6 2/79 3.18 x 104 1.91 x 105 Services 1 6/84 8.61 x 10 4 8.61 x 104 Closed 6 (90% time) Cold Shut- S.43 x 10 4 3.04 x 10 5 Cooling 2 (107 time) down System Feedwater 9 2/79 3.18 x 104 2.86 x 105 System Decay Heat 2 6/84 8.61 x 104 1,72 x 104 Closed Cycle Emergency 3 6/84 8.61 x 104 5.17 x 105 Feedwater System Total Valve Hours = 1.56 x 106 9

A.2-46 0211G0619860AR

TMI-1 f) v COMPONENT SUCCESS DATA

SUMMARY

SHEET OESIGNATOR: V7F (Sheet 1 of 2)

SYSTEM: All Systems (except intermediate closed cooling and river water systems)

COMPONENT TYPE: Check Valve FAILURE MODE: Fails To Operate on Demand Number of Demands Number of Total Number of Test per Test ,

Tests Demands 1300-3H A/B 12 30 392 (includes 16 unit startups) 1303-11.27 40 2 80 i

1303-11.8 22 7 154 1303-11.54 8 2 16 1300-3P 8 10 80 1300-3A A/B 8 43 344 1303-11.50 4 5 20 1300-3B A/B 12 69 860 (includes 16 unit shutdowns) 1300-3G A 8 26 208  ;

1300-3G B 20 36 700

.1300-3F 10 37 370 1300-11.42 24 3 72 1300-3N 2 44 88 V

A.2-47 0211G061986DAR

TMI-1 COMPONENT SUCCESS DATA

SUMMARY

SHEET DESIGNATOR: V7F (Sheet 2 of 2)

SYSTEM: All Systems (except intermediate closed cooling and river water systems)

COMPONENT TYPE: Check Valve FAILURE MODE: Fails To Operate on Demand Number of Demands Number of Total Number nf Test per Test Tests Demands 1300-3J 6 40 272 (includes 16 startups) 1300-11.21 4 8 32 Feedwater 16 60 Start- 960 System ups 1303-11.39 28 10 280 Total Number of Uemands = 4,928 1

l 91 A.2-48 0211G0619860AR

r l

l 6

TMI-1  !

'( ) COMPONENT SUCCESS DATA

SUMMARY

SHEET l

DESIGNATOR: V7FR  !

SYSTEM: River Water System COMPONENT TYPE: Check Valve j

FAILURE MODE: Fails To Operate on Demand l Number of Demands Number of Total Number of  ;

Test _

per Test Tests Demands -i i

i 1300-30 4 48 224(includes  ;

16 unit I shutdowns )  !

I' 1300-4C A/B 6 5 30 1300-31 A/B 6 71 426 j l

1300-4E 6 S 32  :

1300-3K A/B 4 71 284 }

( 1303-11.9 4 7 28 OPS-S227 2 -- 1,212  ;

i UPS-SS2 2 (pawer -- 1,246 l operation) -

i 4 (cold  !

shutdown)  !

Total Number of Oemands = 3,480  !

i l

l

()

l A.2-49 I 0211G0619860AR l

,-..._w...._- - _ _ , . . - . . _ _ _ _ . , _ . _ , , . , . . . _ . . . . , . , . . _ _ _ , , , - . . _ - _ - , , - - _ . , , , , , , . - . . , _ . - . . . , _ . - , - , - , . , . .

TMI-1 COMPONENT SUCCESS DATA

SUMMARY

SHEET DESIGNATUR: V7FI SYSTEM: Intermediate Closed Cooling COMPONENT TYPE: Check Valve FAILURE MODE: Fails To Operate on Demand Number of Demands Number of Total Number of Test per Test Tests Demands Operation 12 16 startups 276 of System and 7 failures of operating pump train O

l l

l 9\

l A.2-50 '

0211G061986DAR

. .-_ , - -. . .. - _- . . - . . - . . ~ - _. - .

I

~!

TMI-1  ;

() COMPONENT SUCCESS DATA

SUMMARY

SHEET DESIGNATOR: V7L SYSTEM: All Systems (except intermediate closed cooling and river water systems) l COMPONENT TYPE: Check Valve FAILURE MODE: Transfers Open, Gross Reverse Leakage.

Number Number of Total Number

{

[

of System Last Verified of Valve System Valves Tested / Veri fied Hours _

Hours Makeup 2 Power Operation '3.18 x 104 6.36 x 104 l System j i

Nuclear 1 Power Operation 3.18 x 104 1.40 x 106 Services 2 Shutdown S.43 x 104 Closed  !

Cooling l Total Valve Hours = 2.04 x 105 (2) r 4

i E

t 4

0 A.2-61 0211GU619860AR

TMI-1 COMPONENT SUCCESS DATA

SUMMARY

SHEET DESIGtMTOR : V7LI S YSTEM: Intermediate Closed Cooling COMPONENT TYPE: Check Valve FAILURE MODE: Transfers Open, Gross Reverse Leakage Number Number of Total Number of System Last Verified of Valve System Valves Tested / Veri fied Hours Hours Interme- 1 6/84 4.23 x 104 4.23 x 104 diate Closed Cooling 9

l 9 A.2-b2 0211G061986DAR l

t

]

THI-1  !

, COMPONENT SUCCESS DATA

SUMMARY

SHEET l DESIGNATOR: V7LR l SYSTEM: River Water Systems

{

l' COMPONENT TYPE: Check Valve FAILORE MODE: Transfers Open, Gross Reverse Leakage Number Number of Total Number-of System.Last . Verified of Valve ,

. System _ Valves , Tested / Veri fied ,,

Hou.s Hours l Nuclear 1 Power Operation- 3.18.x 104' 1.40'x 105 Services 2 Shutdown S.43 x 104 i River l Water  !

Total Valve Hours = 1.40 x 106 O

1 O

A.2-b3 0211G0619860AR

TMl-1 COMPONENT SUCCESS DATA

SUMMARY

SHEET DESIGNATOR: V7T SYSTEM: All Systems (except intermediate closed cooling and river water systems)

COMPuriENT TYPE: Check Valve FAILURE MODE: Transfers Closed Number Number of Total Number of System Last Verified _

of Valve System Valves Tested / Verified Hours Hours Makeup 12 2/79 3.18 x 104 3.8% x 105 System Decay Heat 3 Cold Shutdown 5.43 x 104 1.63 x 105 Hemoval Chilled 1 6/84 8.61 x 104 8.61 x 104 Water System Nuclear 3 Power Operation 3.18 x 104 1.49 x 105 Services 1 Shutdown b.43 x 104 Closed Cooling System Feedwater 8 2/79 3.18 x 104 2.54 x 105 System Total Valve Hours = 1.03 x 10' 1

i 1

O 1

A.2-54 0211G0619860AR ,

l

i TMI-1

[LJ i COMPONE,NT SUCCESS DATA

SUMMARY

SHEET UESIGNATOR: V7TI SYSTEM: Interrcediate Closed Cooling i

COMPUNENT TYPE: Check Valve FAILURE MODE: Transfers "losed

. Number Number of Total Number of System Last Verified of Valve System. . Valves Tested / Verified Hours Hours Intermediate 6 6/84 4.23 x 104 2.54 x 106 Closed .

Cociing O

V O

A.2-bb 0211G0bl986DAR

TMI-l COMPONENT SUCCESS DATA

SUMMARY

SHEET g DESIGNATUR: V7TR SYSTEM: River Water Systems COMPONENT TYPE: Check Valve FAILURE MODE: Transfers Closed Number Number of Total Number of System Last Verified of Valve System Valves Tested / Verified Hours Hours Decay Heat 1 Cold Shutdown 5.43 x 104 5.43 x 104 River Water Nuclear 6 Power Operation 3.18 x 104 1.18 x 105 Services 1 Shutdown b.43 x 104 River Water Decay Heat 1 OPS-S227 702 702 River Water Nuclear 1 OPS-S$2 623 623 Services River Water Total Valve Hours = 1.74 x 105 9

A.2-56 l 0211G061986DAR l

TMI-1

[)

U COMPONENT SUCCESS DATA

SUMMARY

SHEET DESIGNATOR: V8T (Sheet 1 of 2)

SYSTEM: All Systems COMPONENT TYPE: Manual Valve FAILURE MODE: Transfers Closed Number Number of Total Number of System Last Verified of Valve System Valves Tested / Verified Hours Hours Makeup 28 2/79 3.18 x 104 8.90 x 105 System Decay Heat 12 6/84 8.61 x 104 .1.03 x 106 Removal System Auxiliary 2 2/79 3.18 x 104 7.36 x 104 Spray Line

~ Decay Heat 10 6/84 8/61 x 104 8.61 x 105 -

( Closed Cooling System Decay Heat 2 6/84 8.61 x 104 1.72 x 106 River Water System Reactor 10 b/84 8.61 x 104 8.61 x 105 Building Spray System Chi 11ed 34 6/84 8.61 x 104 1.08 x 106 Water '

System Inter- 8 2/79 3.18 x 104 2.64 x 105 mediate i Cooling System

/~~~s 0 l A.2-57  ;

0211GU61986DAR l J

TMI-1 COMPONENT SUCCESS DATA

SUMMARY

SHEET DESIGt4ATOR : V8T (Sheet 2 of 2)

SYSTEM: All Systenis COMPONENT TYPE: M;nual Valve FAILURE MODE: Transfers Closed Number Number of Total Number of System Last Verified of Valve System Valves Tested / Verified Hours Hours Nuclear 32 During 3.18 x 104 1.02 x 106 Services Plant Closed Operation Cooling System 5 During Shutdown b.43 x 104 1.60 x 105 20 Continuous 8.61 x 104 1.63 x 106 Operation Emergency 7 6/84 8.61 x 104 6.03 x 105 Feedwater System Feedwater 13 2/79 3.18 x 104 4.13 x 106 System Total Number of Valve Hours = 9.05 x 106 S

A.2-58 0211G061986DAR

.. ..- . . . . .-. . . - . - _ - = . - - . ._-. -.. - - - - .-. . . . - . . .--

i  !

.j THI-1 COMPONENT SUCCESS DATA

SUMMARY

SHEET I

DESIGNATUR: V8T  !

r SYSTEM: All Systems l COMPONENT TYPE: Manual Valve i

FAILURE MODE: Transfers Open Number Number of- Total Number of System Last Verified of Valve j System Valves- Tested / Verified , Hours _

Hours i

. Nuclear 2 6/84 8.61 x 104 1.72 x 105 i Services  !

Closed l Cooling i 2 Power Operation 3.18 x 104 6.36 x 104 i

,1 Cold Shutdown 5.43 x 104 5.43 x 104  !

Inter- 1 2/79 3.18 x 104 3,18 x~104  !

mediate  ;

Cooling l System j I

O Makeup 3 2/79 3.18 x 10 4- 9.54 x 104 i

l System j l

Total. Number cf Valve Hours = 4.17 x 10b l

'l 4

l i

2 a

4 J

} l

't i

J I

i i l t

! A.2-59 t 0211GU619860AR  !

TMI-1 COMPONENT SUCCESS DATA

SUMMARY

SHEET DESIGNATUR: V10FS, V10FW SYSTEM: RCS COMPONENT TYPE: Code Safety Valve FAILURE MODE: Fails To Open on Demand Number of Demands Number of Total Test per Test Tests Demands 1303-11.2 6 16 96 O

O A.2-60 0211GU619860AR

_ . . - -.. . . - ~ - . - - - - - . . - - . - .- - . .- .=

i t

i TMI-l i

'COMPUNENT SUCCESS DATA

SUMMARY

SHEET 3

DESIGNATOR: V10FS, V10FW l l

1 SYSTEM: Main Steam I 1

COMPUNENT TY3: Code Safety Valve '

1 FAILURE MODE: Fails To Open on Demand

, t Number of Demands Number of . Total  !

, Test per Test ., Tests Demands l

t

?.

1303-11.3 5 7 3b  !

The relief valves may open on large steam demand reductions and after l

. reactor trips. No record is kept of how many opened. These demands have  :

not been included. Total number of main steam relief valve demands is i then 35. l 4

!O 1

1  !

I l

i I i

4 i i

l r

i I

i,  !

i i

l I

!O I

i A.2-61 4 0211G061986DAR 1

, _ . . - - .- ]

TMI-1 COMP 0tiENT SUCCESS DATA

SUMMARY

SHEET DESIGNATOR: V10RS SYSTEM: RCS COMPutiENT TYPE: Code Safety Valve FAILURE MODE: Fails To Reseat after Opening (passing steam)

Number of Demands Number of Total Test _

per Test Tests Demands 1303-11.2 6 16 96 0

9 A.2-b2 021160bl9860AR

TMI-1

( COMPONENT SUCCESS DATA

SUMMARY

SHEET

% j)

DESIGNATOR: V10RS SYSTEM: Main Steam COMPUNENT TYPE: Code Safety Valve FAILURE HOUE: Fails To Reseat after Opening (passing steam)

Number of Demands Number of Total Test per Test Tests Demands  ;

1303-11.3 5 7 35 l The relief valves may open on large steam demand reductions and after reactor trips. No record is kept of how many opened. These demands have not been included.

i i

I 1

l l

1 O

O l

l A.2-63 I

0211G0619860AR

TMI-1 COMPONENT SUCCESS DATA

SUMMARY

SHEET DESIGNATOR: V11FS, V11FW SYSTEM: Pressurizer, Main Steam COMP 0NENT TYPE: PORV/ ADS FAILURE MODE: Fail To Open on Demand There are three power-operated relief valves: one on pressurizer and two on main steam. They are demanded to open on reactor trips. There have been a total of six trips, leading to a total of 18 demands on PORV/ ADS.

O j

)

i O

A.2-64 0211G062086DAR

l TMI-1 [

COMP 0NENT SUCCESS DATA

SUMMARY

SHEET f]

v DESIGNATOR: V11RS, V11RW ,

i SYSTEM: Pressurizer, Main Steam  :

COMPONENT TYPE: PORV/ ADS FAILURE MODE: Fail To Reseat after Opening l A total of 18 power-operated relief valve demands to reseat after opening is estimated, based on 18 demands to open. (See notes on fail to open on demand success data summary sheet.)

i I

i f

t l

F l

l r

i v  ;

P I

b v

A.2-65 0211G062086DAR

THI-l COMPONENT SUCCESS DATA

SUMMARY

SHEET DESIGNATUR: X1F SYSTEM: Electric Power COMPONENT TYPE: Main Transformers lA and 1B FAILURE MODE: Fail During Operation Both transformers (lA and 18) operate at all times. Total hours per transformer = 8.6 x 104 . Total transformer hours of operation is then 1.72 x 10 5, e

I el<

A.2-66 0211GU619860AR

I

.l l

l TMI-1  !

COMPONENT SUCCESS DATA

SUMMARY

SHEET -

1 DESIGNATOR: X2F SYSTEM: Electric Power j COMPONENT TYPE: Station Service Transformer  !

FAILURE MODE: Fail During Operation All 16 transformers opgrate at all times. Total operating hours per _  !

transformgr = 8.6 x 10 . Total transformer hours of operation is then t 1.38 x 10'J.

l l

1 1

l l

O  !

(

l 1

i i

l l

O l l

A.2-67 l 0211GU619860AR 1 l

4 e

A J

i I

1 i

4 1

1 i

APPENDIX B TMI-1 COMPONENT COMMON CAUSE FAILURE DATA

SUMMARY

SHEETS 1

i f

l 4

'l i

i i

i i

l 1

Y I i

l 1 4 i

l 4

5 0428G1217860AR m -- rm w . ._ _ _ - w a ,-rym m .

i i

1 TMI-l rN SITE-SPECIFIC COMP 0hENT COMMON CAUSE FAILURE DATA

SUMMARY

SHEET N]

SYSTEM: All l COMPONENT TYPE: Motor-Operated Valve  !

FAILURE MODE: Fails To Operate on Demand Site-Specific Data Date Reported Failure Cause l l

7/1/78 (WR-24456)MS-V-5A, Both valves failed to operate-- '

MS-V-bB both tripped on overload.

l l

l Q I V '

l l

l Failure Data Source: Maintenance Request Logbook Work Requests (microfilm and computer file) i dob Tickets (microfilm and computer file) l l

O B-1 0428G061986DAR l

TMI-I COMMON CAUSE PARAMETER S'JMMARY SHEET C0f1PONENT: Standby Motor-Driven Pump (MU, EBS)

FAILURE MODE: Falls To Start on Demand Sheet 1 of 2 Dependent Events

  • Independent Event 1 Evett 2 Event 3 Event 4 Event 5 Event 6 Event 7 Event 8 Event 9 Cctegory - - - - -

Failures p n p p p n p n p n p n p n p n p n 1 26.5 7Fj= 26.5 2 1.0 2 ng = 2.0 3 1 1.2 1.0 3 n3 - 4.2 4 0.0 n4 =

r'v

  • For a complete listing of all dependent events considered in screening for applicability to THI-1, see EPRI-NP-3697 report, Reference 2.22. The following provides cross-references between events listed in this table and those reported in NP-3967 by plant name and date of occurrence:

Etent 1: Salem 1 (November 1979), NP-3967, p. 3-96.

Ee:nt 2: Robinson 2 (July 1973) NP-3967, p. 3-98.

Event 3: Robinson 2 (Novernber 1977) NP-3967, p. 3-98.

O 0428G062186DAR O O

l I

i

[v (J (v TMI-1 COMMON CAUSE PARAMETER

SUMMARY

SilEET COMPONENT: Standby Motor-Driven Pump (DHR, DHCC)

FAILURE MODE: Fails To Start on Demand Sheet 2 of 2 Dependent Events

  • Ind ndent Event 1 Event 2 Event 3 Event 4 Event 5 Event 6 Event 7 Event 8 Event 9 Category - - - - - - - - -

Failures p n p n p n p n p n p n p n p n p n 1 35.45 nj = 35.45 2 1 2 1 2 1 2 n2 = 6.0 3 n3 = 0.0 4 n4 = 0.0 O

  • For a complete listing of all dependent events considered in screening for applicability to TMI-1, see EPRI-NP-3697 report, Reference 2-22. The f ollowing provides cross-references between events listed in this table and those reported in NP-3967 by plant name and date of occurrence:

Event 1: Peach Bottom 2 (April 1978), NP-3967, p. 3-109.

Event 2: Brunswick 1 (April 1979), NP-3967, p. 3-109.

Eeent 3: Browns Ferry 1 (September 1974), NP-3967, p. 3-110.

04?8G062186DAR ._. , _ ,. ,

i TMI-1 COMMON CAUSE PARAMETER

SUMMARY

SHEET COMPCNENT: River Water Pump FAILURE MODE: Fails To Start on Demand Dependent Events

  • Independent Event 1 ec Event 2 Event 3 Event 4 Event 5 Event 6 Event 7 Event 8 Event 9 Category - - - -

ln p n p n Failures p n p p n p n p n p n p n 1 49.3 F=j 49.3 2 1.0 2 2.0 n2 =

3 0.0 n3 =

4 li4 = 0.0 2

CFor a complete listing of all dependent events considered in screening for applicability to TMI-1, see EPRI-NP-3697 report, Reference 2-22. The f ollowing provides cross-references between events listed in this table and those reported in NP-3967 by plant name and date of occurrence:

Ev:nt 1: Cyster Creek (November 1978), LP-3967. p. 3-141.

O 0428G062186DAR G G

O O TMI-1 COMMON CAUSE PARAMETER

SUMMARY

SHEET COMPONENT: Standby Motor-Driven Pump (DHR, DHCC)

FAILURE MODE: Falls during Operation Dependent Events

  • Independent Event 1 Event 2 Event 3 Event 4 Event 5 Event 6 Event 7 Event 8 Event 9 f

Category - - - -

4 l -n - - -

Failures p n p n p n p n p n p p n p n p n 1 33.75 T=j 33.75 2 1 .02 n2 = .02 3

n3 = 0.0 4

n4 = 0.0 E

  • Fer a complete listing of all dependent events considered in screening for applicability to TMI-1, see EPRI-NP-3697 report, Reference 2-22. The following provides cross-references between events listed in this table and those reported in NP-3967 by plant name and date of occurrence:

Event 1: Beaver Valley 1 (January 1978), NP-3%7, p. 3-107.

0428G062186DAR

TMI-I C0t@ ION CAUSE PARAMETER SUMIMRY SHEET COMPONENT: River Water Pump FAILURE MODE: Fails during Operation Dependent Events

  • Independent Event 1 Event 2 Event 3 Event 4 Event 5 Event 6 Event 7 Event 8 Event 9 ,

Category - - - - - - - - -

Failures p n p n p n p n p n p n p n p n p rt 1 87.9 .9 5 ' h'y = 92.40 2 .05 2 .10 n2 =

3 .03 3 .09 n3 =

cn >4 .02 4 1.0 .05 n4 = .13 m

  • F:r a complete listing of all dependent events considered in screening for applicability to THI-1, see EPRI-NP-3697 report, Reference 2-22 The fcllowing provides cross-references between events listed in this table and those reported in NP-3967 by plant name and date of occurrence:

Event 1: Pilgrim (May 1974), NP-3967, p. 3-140.

Event 2: Hatc't 1 ( August 1979), NP-3967, p. 3-141.

O 0428G062386DAR O O

i O ". O N] N(b Y

_TMI-I l COMMON CAUSE PARAMETER

SUMMARY

SHEET l

l COMP 0hENT: EFW Pump (pump portion) l FAILURE MODE: Falls during Operation Dependent Events

  • I Independent Event 1 Event 2 Event 3 Event 4 Event 5 Event 6 Event 7 Event 8 Event 9 i

Impact Number of

S l Category p

p n p n p n p n p p

n p Failures l n n n p n 1 35.7 ny - 35.7 2 E=2 0 3 1.0 .3 n3 = .3 4 "4 = 0 m g 0

  • Ftr a complete listing of all dependent events considered in screening for applicability to TMI-1, see EPRI-NP-3697 report. Reference 2-22. The fellowing provides cross-references between events If sted in this table and those reported in NP-3967 by plant name and date of occurrence:

Event 1: Kewaunee (November 1975), NP-3967, p. 3-127.

i l

l l

i l 0428G062186DAR

JMH COMMON CAUSE PARAMETER

SUMMARY

SHEET COMPONENT: EFW Pump (pump portion)

FAILURE MODE: Falls To Start Dependent Events *

  1. 'C Independent Event 1 Event 2 Event 3 Event 4 Event 5 Event 6 Event 7 Event 8 Event 9 Cctegory -

Failures p n p n p n p n p n p n p n p n p n 1 37.1 nj = 37.1 2 ng = 0 3 n3 = 0 4 n4 = 0 En

  • Fer a cociplete listing of all dependent events considered in screening for applicability to TMI-1, see EPk.'-NP-3967 report, Reference 2-22 None of the events listed in Table 3-28 of NP-3967 involving IFW pumps was considered applicable to TMI-l for this failure mode of EFW pumps.

O 0428G062186DAR O O

1 O O TMI-1 COMMON CAUSE PARAMETER

SUMMARY

SHEET COMPONENT: Diesel Generator FAILURE MODE: Fails To Start on Demand Dependent Events *

,'" In dent Event 1 Event 2 Event 3 Event 4 Event 5 Event 6 Event 7 Event 8 Event 9 Category - - - - - - - - -

Failures p n p n p n p n p n p n p n p n p n l

l 1 245.8 F=

j 245.8 l

l 2 1.0 2 1.0 .2 1.0 .2 1.0 2 1.0 2 1 2 F=

2 8.4 3 1.0 3 n3 = 3.0 y 4 1.0 4 n4 = 0.4 E

  • Fer a complete listing of all dependent events considered in screening for applicability to TMI-1, see EPRI-NP-3697 report, Reference 2-22 The following provides cross-references between events listed in this table and those reported in NP-3%7 by plant name and date of occurrence:

Event 1: Fort Calhoun (July 1973), NP-3%7, p. 3-24 Ercot 2: Zion (July 1974), NP-3%7, p. 3-26.

Event 3: Peach Botton 2 and 3 (June 1977), NP-3%7, p. 3-27.

Event 4: Farley 1 (September 1977), NP-3967, p. 3-28.

Event 5: Cook 1 (December 1977), NP-3%7, p. 3-29.

Event 6: North Anna (February 1981), NP-3%7 p. 3-30.

Event 7: Three Mile Island 1 (April 1974), NP-3%7 p. 3-30.

Event 8: Browns Ferry 1 and 3 (May, June 1981), NP-3967, p. 3-31.

I l

l l

l l 0428G062186DAR

_ ._ . _ _ _ _ . .. _ _ . . _ _ . . . _ _ _ _ _ . .~ _ _.. _ __. .. _ _ . _ _ _ _ _ _ . _ . . _ _ . _.

TMI-1 C0K40ft CAUSE PARAMETER

SUMMARY

SHEET COMPONENT: Diesel Generator FAILURE MCCE: Fails during Operation Dependent Events *

  1. * ' I'C Independent Event 1 Event 2 Event 3 Event 4 Event 5 Event 6 Event 7 Event 8 Event 9 Impact Number of Category -

Failures p n p n 9 n p n p n p n p n p n p n 1 173 .9 2 nj = 174.8 2 1.0 2 1.0 2 1.0 .2 .1 2 n2

  • 44 3 1.0 2.1 n3 = 2.1 c2 >4 ng = 0.0
  • Fcr a complete listing of all dependent events considered in screening for appifcability to TMI-1, see EPRI-NP-3697 report Reference 2-22. The fcllowing provides cross-references between events listed in this table and those reported in NP-3967 by plant name and date of occurrence:

Event 1: Salem (July 1977) NP-3967, p. 3-25.

Eccnt 2: Yankee Rowe ( August 1977). NP-3967, p. 3-25.

Event 3: Millstone 2 (May 1977). NP-3967, p. 3-27.

Event 4: Peach Bottom 2 (February 1978). NP-3967, p. 3-28.

Event 5: Arkansas One 1 (August 1979), hP-3967, p. 3-29.

o 0428G062186DAR O O O

r) v fd' (aD TMI-1 COMMON CAUSE PARAMETER

SUMMARY

SHEET l COMPONENT: Standby Motor-Driven Pump (MU, CBC) l FAILURE MODE: Fails during Operation Dependent Events

  • Independent Event 1 Event 2 Event 3 Event 4 Event 5 l Event 6 Event 7 Event 8 Event 9 Category - - - - - - - - -

Failures p n p n p n p n p n p n p n p n p n 1 52.3 nj = 52.3 2 1 2 2.0 n2 =

3 0.0 n3 =

y 4 n4 = 0.0 C

  • Fer a complete listing of all dependent events considered in screening for applicability to THI-1, see EPRI-NP-3697 report, Reference 2-22 The following provides cross-references between events listed in this table and those reported in NP-3967 by plant name and date of occurrence:

Etent 1: North Anna 1 (July 1978), NP-3967, p. 3-97.

=

3 0428G062186DAR

_TMI-I COMMON CAUSE PARAMETER

SUMMARY

SHEET C04PONENT: Reactor Trip Breaker FAILURE MODE: Fails To Open on Demand Dependent Events

  • Independent Event 2 c( Event 1 Event 3 Event 4 Event 5 Event 6 Event 7 Event 8 Event 9 Cctegory -

Failumes p n p n p n p n p n p n p n p n p n 1 53.1 nj = 53.1 2 1 2 1 2 1 2 1 1 .2 1 .2 1 .2 F=

2 6.6 3 1 .3 T=

3 .3 4 4 .4 4.4 7 1 1 n4 =

5

  • Fer a complete listing of all dependent events considered in screening for applicability to THI-1 see EPRI-NP-3697 report, Reference 2-22. The fcllowing provides cross-references between events listed in this table and those reported in NP-3h67 by plant o.ame and date of occurrence:

Evpnt 1: Connecticut Yankee (December 1981), NP-3967, p. 3-12.

E ent 2: Oconee 1 (February 1979), NP-3967, p. 3-12.

Event 3: St. Lucie (November 1980), NP-3967, p. 3-12.

Etent 4: Salem 1 (February 1933), NP-3967, p. 3-13.

Event 5: Calvert Cliffs 1 (March 1983), XP-3367, p. 3-14.

Erent 6* Calvert Cliffs 2 (March 1983), NP-3967, p. 3-I4.

Event h McGuire 1 (March 1983), NP-3967, p. 3-14.

Ev:nt 8: Maine Yankee (March 1983), NP-3967, p. 3-14.

O 0428G062186DAR O O

i TMI-1 COPMON CAUSE PARAMETER SUPNARY SHEET l

l CGPONENT: Motor-Operated Yalves FAILURE MODE: Fall To Operate on Demand Sheet 1 of 5 Dependent Events

  • In ndent Event 1 Event 2 Event 3 Event 4 Event 5 Event 6 Event 7 Event 8 Event 9 p l Category - - - - -- - - - -

Failures **

l p n p n p n p p p n p n p n p n p n 1 847.7 t

1.0 1.0 2 2 2 1.0 2 1.0 .2 1.0 3 1.0 1.2 1.0 1.2 y >4 1.0 1.3 1.0 1.3 C

  • For a complete listing of all dependent events considered in screening for applicability to TMI-1, see EPRI-NP-3697 report, Reference 2-22. The following provides cross-referer.ces between events Ifsted in this table and those reported in NP-3%7 by plant name and date of occurrence:

Event 1: Cook 2 (January 1979), NP-3967, p. 3-38.

Evint 2: Turkey Point (April 1979), NP-3907, p. 3-38.

Event 3: Arkansas One (April 1980), NP-3%7 p. 3-38.

Event 4: p. 3-39.

Event 5: Palisades Oconee 2 (October (June 1971)5)NP-3967, 197 NP-3987, p. 3-39.

Event 6: NP-3967 p. 3-4 0.

Event 7: Trojan North Anna (October (August 1976)}8),

19 NP-3667, p. 3-40.

Event 8: Kewaunee (September 1975), NP-3%7, p. 3-41.

Event 9: Zion 2 (October 1975), NP-3%7, p. 3-41.

    • See sheet 5 for total failures.

0428G062386DAR *

._ _ _ . _ _ . . _ _ _ . . _ - - _ . _ _ . _ , . _ _ _ - _ _ _ . - _ . _ _ - - _ . . . - - _ . _ _ - . , _ _ _ .. _.. ... _ . - - . _ _ _ .- . . _ _ , . . . _ - - - - ~ ~ _ -. . . . ~ _ - - -

TMI--I COMM0fi CAUSE PARAMETER

SUMMARY

SHEET COMPONENT: Motor-Operated Yalves FAILURE MODE: Fail To Operate on Demand Sheet 2 of 5 Dependent Events

  • Independent Event 10 Event 11 Event 12 Event 13 Event 14 Event 15 Event 16 Event 17 Event 18 Impact Number of Cate9ory - - - - - - - - -

Failures **

p n p n p n p n p n p n p n p n p n 1

2 1.0 2 1.0 2 1.0 2 1.0 2 1.0 2 1.0 2 1.0 2 3

y >4 1.0 5 1.0 2.2 5

'For a complete listing of all dependent events consicered in screening for applicability to THI-1, see EPRI-NP-3697 report, Reference 2-22. The fallowing provides cross-references between events If sted in this table and those reported in NP-3967 by plant name and date of occurrence:

Event 10: Arkansas One 1 (August 1981), NP-3967, p. 3-42.

Event 11: Oconee 1 (November 1975), HP-3967, p. 3-42.

Etent 12: Oconee 2 (December 1975), NP-3967, p. 3-42.

Event 13: Rancho Seco (November 1976), NP-3967, p. 3-42.

E;ent 14: Cook 1 (November 1977), NP-3967, p. 3-44.

Event 05: Oconee 2 (June 1979), NP-3967, p. 3-46.

Etent 16: Oconee 2 (December 1979), NP-3967, p. 3-47.

Er;nt 17: Cook 1 (March 1981), NP-3967, p. 3-47.

Evnt 18: Monticello (July 1972), NP-3967, p. 3-48.

  • cGee sheet 5 for total f ailures.

0 0428G062186DAR 0 0

w G A 3 V U ./

TMI-1 COMMON CAUF' PARAMETER

SUMMARY

SHEET COMPONENT: Notor-Operated Valves FAILURE MODE: Fall To Operate on Demand Sheet 3 of 5 Dependent Events

  • Independent Event 19 Event 20 Event 21 Event 22 Event 23 Event 24 Event 25 Event 26 Event 27 Category - - - - -

Failures **

p n p n p n p n p n p n p n p n p n 1

2 2 1.1 1 1.0 2 1.0 2 1.0 2 1.0 .2 1.0 2 1.0 1.1 3

4 co ->4 1.0 1.3 1.0 1.3 i .L m

  • For a complete listing of all dependent events considered in screening for applicability to TMI-1, see EPRI-NP-3697 report, Reference 2-22. The following provides cross-references between events listed in this table and those reported in NP-3967 by plant name and date of occurrence:

Event *.): Browns Ferry 3 (May 1975), NP-3967, p. 3-48. #

) Evrnt 20: Robinson 2 (Janury 1981), NP-3967, p. 3-49.

Evsnt 21: Surry 2 (July 1981), NP-3967, p. 3-49.

Evtnt 22: Dresden 2 (August 1973), NP-3967, p. 3-50, s Ev:nt 23: Browns Ferry 2 (December 1974), NP-3967, p. 3-51.

Event 24: Pilgrim (September 1974), NP-3967, p. 3-51.

4

~

Evtnt 25: Vermont Yankee (May 1976), NP-3967, p. 3-52.

Event 26: Dresden 3 (September 1975), NP-3967, p. 3-52.

Evsnt 27: Browns Ferry 1 (September 1974), NP-3967, p. 3-53.

    • See sheet 5 for total failures.

4 4

4 l 0428G062186DAR

TMI-1 C0P. MON CAUSE PARAMETER

SUMMARY

SHEET CCMPCNENT: Motor-Operated Yalves FAILURE MODE: Fail To Operate on Demand Sheet 4 of 5 Dependent Events *

  1. 5 ** '#

Independent Event 28 Event 29 Event 30 Event 31 Event 32 Event 33 Event 34 Event 35 Event 36 Impact Number of Category - -

Failures **

p n p n p n p n p n p n p n p n p n 1

2 1.0 2 1.0 2 1.0 2 1.0 2 1.0 2 1.0 .2 1.0 2 1.0 1.1 1.0 2 1

co 0.

cn L4

  • For a complete listing of all dependent events considered in screening for applicability to TMI-1, see EPRI-NP-3697 report, Reference 2-22. The following provides cross-references between events listed in this table and those reported in NP-3967 by plant name and date of occurrence:

E' nt 28: Hatch 2 (September 1978), NP-3967, p. 3-53.

Event 29: Pilgrim (July 1979), NP-3967, p. 3-54.

Event 30: Hatch 2 (May 1980), NP-3967, p. 3-54.

Ev nt 31: Hatch 2 (May 1982), NP-3967, p. 3-54.

Event 32: Dresden 2 (Cctober 1973), NP-3967, p. 3-9.

Event 33: Arnold (March 1976), NP-3967 p. 3-55.

Event 34: Dresden 2 (May 1975), NP-3967, p. 3-56.

Event 35: Cooper (October 1980), NP-3967, p. 3-56.

Evtnt 36: Vermont Yankee (Septerrher 1976), NP-3967, p. 3-56.

    • See sheet 5 for total failures.

T

-w S .

G G

L v) d TMI-I COMMON CAUSE PARAMETER

SUMMARY

SHEET COMPONENT: Motor-Operated Valves FAILURE MODE: Fall To Operate on Demand 1

Sheet 5 of 5 Dependent Events

  • Indep ndent Event 37 Event 38 Event 39 Event 40 Event 41 Event 42 Event 43 Event 44 E ctive

, Event 45 Category P n p n p n p n p n p n p n p n p n

Failures **

1 lij = 799.7 2 1.0 2 1.0 2 1.0 .2 1.0 2 55.7

  • n2 =

3 1.0 3 n3 = 5.4

>4_

n4 = 5.2 w

  • For a complete listing of all dependent events considered in screening for applicability to TMI-1, see EPRI-NP-3697 report, Reference 2-22. The following provides cross-references between events listed in this table and those reported in NP-3967 by plant name and date of occurrence:

Event 37: Dresden 2 (August 1973), NP-3967, p. 3-56.

Event 38: Dresden 1 (October 1978), NP-3967, p. 3-57.

Event 39: Pilgrim (October 1981), NP-3967, p. 3-58. '

. Event 40: Millstone 1 (May 1971), NP-3967, p. 3-59.

Event 41: Pilgrim (April 1973), NP-3967, p. 3-59.

    • See sheet 5 for total failures.

0428G062186DAR

O 1 l

l l

l i

l l

l l

l l

i l

APPENDIX C i i

TMI-1 COMP 0NENT MAINTENANCE DATA

SUMMARY

SHEETS i O

l l

1 l

l l

l l

l O l 0212G0618860AR 4

_ _ _ _ _ . _ _ _ _ _ . _ _ _ _ ____4

e APPENDIX C l TMI-1 COMPONENT MAINTENANCE DATA

SUMMARY

SHEETS The data sheets in this appendix summarize the component maintenance i events.used for the TMI-1 plant-specific maintenance frequency and duration updates. 'The maintenance events were obtained from a review of  ;

the plant tagout orders. The data in this appendix are organized by component type as listed in Tables 3-6 and 3-7.

l I

i l

t l

l 1

O )

O

- C-1 0212G061886DAR

TMI-1 .

COMPONENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: MDCD1 SYSTEM: Main Steam COMPONENT TYPE: Main Steam Atmosphere Dump Valve Site-Specific Data Data include only maintenance performed during periods when the reactor was not in cold shutdown.

e Total noncold shutdown hours for reporting period: 6.63 x 104 Duration Tag Order Component Date (hours) Work Done 5420 MS-V-4A 11/16/74 1.75 Calibrate positioner.

6048, 6050, MS-V-4A 2/20/75 294.75 Repair seat.

6059, 6125, 6141 8167 MS-V-4A 10/9/75 3.5 Check electric /

pneumatic converter and strike valve.

8502 MS-V-3E 11/20/75 125.0 Calibrate.

8721 MS-V-3E 12/30/75 9.0 Repair actuator.

e Maintenance Data Source: Switching and Tagging Orders O

C-2 0212G061886DAR

TMI-1  !

COMP 0NENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: MDC2 ,

i

-SYSTEM: Reactor Coolant i i

COMP 0NENT TYPE: Pressurizer Spray Valve -l I

Site-Specific Data i Data include only maintenance performed during periods when the reactor  !

was not in cold shutdown.  !

t e Total noncold shutdown hours for reporting period: 9.54 x 104 i

Duration  !

Tag Order Component Date (hours) Work Done l

~!

5454 RC-V-1 11/21/74 2.0 Troubleshoot valve.

5871, 5875 RC-V-1 1/23/75 19.5

  • l O l i

l e Maintenance Data Saurce: Switching and Tagging Orders

  • No information provided in tag order forms.

C-3 0212G061886DAR

TMI-1 COMP 0NENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: MDCF1 (Sheet 1 of 3)

SYSTEM: Reactor Building Emergency Cooling COMPONENT TYPE: Containment Fan Cooler Site-Specific Data Data include only maintenance performed during periods when the reactor was not in cold shutdown.

e Total noncold shutdown hours for reporting period: 9.54 x 104 Duration Tag Order Component Date (hours) Work Done 5007 AH-E-1A 9/17/74 5.5 Change overload block.

5137 AH-E-1C 10/15/74 1.3 Drain water out of motor.

5471 AH-E-1C 11/23/74 2.5 Check moisture detector on fan motor.

5475, 5476 AH-E-1C 11/25/74 4.67 Correct problem with water detector alarm.

5491 AH-E-1B 11/26/74 0.5 5498 AH-E-1C 11/27/74 6.75 Clear alarm.

5517 AH-E-1B 12/1/74 1.3 Clear alarm on moisture detector.

5589 AH-E-1B 12/13/74 6.75 Valve RR-V-4B deenergized closed.

6218 AH-E-1B 3/15/75 0.3 Valve RR-V-4B deenergized closed.

6306 AH-E-1C 3/21/75 0.6 Inspect breaker wiring.

6319 AH-E-1C 3/22/75 2.16 Inspect breaker wiring.

6322 AH-E-1B 3/22/75 2.16 8912 AH-E-1C 2/10/76 0.8 Clear vibration alarm.

i

  • No information provided in tz , order forms.

C-4 0212G0618860AR

THI-1 COMP 0NENT MAINTENANCE DATA

SUMMARY

-SHEET DESIGNATOR: MDCF1 (Sheet 2 of 3)

SYSTEM: Reactor Building Emergency Cooling COMPONENT TYPE: Containment Fan Cooler Site-SpecWic Data Duration Tag Order Component Date (hours) Work Done 432 AH-E-1B 8/2/76 3.5 Repair.

434 AH-E-1B 8/4/76 25.2 Repair.

1624 RR-V-4B 1/31/77 48.3 Repair per WR-18072.

649 AH-E-1A 5/14/77 72.0 Remove contactor.

1115 AH-E -1C 9/30/77 15.40 Vibration reading.

1196 AH-E-1C 10/25/77 4.45 Replace vibration switch.

1214 AH-E-1B 10/31/77 4.0 Install new leak detector relay.

1216 AH-E-1B 11/1/77 14.1 Install new leak detector relay.

1224 AH-E-1A 12/2/77 6.0 Install new leak detector relay.

1230 AH-E-1C 11/3/77 7.0 Install new leak detector relay.

7 AH-E-1C 12/13/77 5.45 Adjust vibration switch.

11 AH-E-1C 12/14/77 7.0 Install new vibration switch.

830 AH-E-1A 6/23/78 59.0 Grease motor.

AH-E-1B Grease motor.

AH-E-1C Grease motor.

O C-5 0212G0618860AR

TMI-1 COMPONENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: MDCF1 (Sheet 3 of 3)

SYSTEM: Reactor Building Emergency Cooling COMPONENT TYPE: Containment Fan Cooler Site-Specific Data Duration Tag Order Component Date (hours) Work Done 1030, 1033 AH-E-1C 8/18/78 82.0 Test breaker, check rotation.

~

1034, 1035 AH-E-1C 8/21/78 5.3 Repair motor, amp readings.

1151 AH-E-1A 9/23/78 111.3 Megger motor.

1174 AH-E-1A 9/25/78 14.35 Remove motor.

1287 AH-E-1B 10/18/78 4.1 Grease motor.

1314 AH-E-1B 10/30/78 4.45 Dry water detector.

14C' AH-E-1A 11/27/78 5.50 1417 AH-E-1C 12/1,78 11.0 (estimated) 1419 AH-E-1C 12/2/78 443.0 1485 AH-E-1C 12/22/78 133.0 AH-E-1C 12/28/78 435.0 1557 AH-E-1C 1/15/79 0.45 1569 AH-E-1C 1/18/79 5,50 1576, 1577 AH-E-1B 1/19/79 2.20 1587 AH-E-1B 1/22/79 24.0 (estimated) e Maintenance Data Source: Switching and Tagging Orders

  • No information provided in tag order forms.

C-6 0212G061886DAR

TMI-1

/^)

k/

COMP 0NENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: MDCFS (Sheet 1 of 6)

SYSTEM: Reactor Building Emergency Cooling River Water COMPONENT TYPE: Pump Site-Specific Data Data include only maintenance performed during periods when the reactor was not in cold shutdown.

e Total noncold shutdown hours for reporting period: 6.36 x 104 ,

l Duration ,

Tag Order Component Date (hours) Work Done l

5808 RR-P-18 1/15/75 5.0 I 5860 RR-P-1A 1/22/75 2.5

~

5945 RR-P-1A 2/3/75 1.0 Inspect breaker.

5947 RR-P-1B 2/3/75 2.0 Inspect breaker.

(v 6231 RR-P-1B 2/16/75 0.25 Discharge valve V-1B l

deenergized closed.

6308 RR-S-1A 3/21/75 3.5 Inspect wiring 6350 RR-P-1A 3/24/75 2.0 Megger.

6431 RR-P-1A 3/31/75 7.5 Repair and clean j rotometer; clean i lines. '

6455 RR-P-1B 4/2/75 4.5 Repair and clean.

6655 RR-P-1B 4/21/75 5.2 Repair ESAS.

6664 RR-P-1B 4/22/75 3.0 Check control switch.

6724 RR-P-1A 4/29/75 5.0 Inspect breakers.

  • No information provided in tag order forms.

O C-7 0212G061886DAR

TMI-1 COMPONENT MAINTENANCE DATA

SUMMARY

SHr.ET DESIGNATOR: MDCF5 (Sheet 2 of 6)

SYSTEM: Reactor Building Emergency Cooling tiiver Water COMPONENT TYPE: Pump Site-Specific Data Duration Tag Order Component Date (hours) Work Done 7281 RR-P-1A 6/14/75 0.7 Check keyway on motor shaft.

7524 RR-P-1B 7/27/75 1.2 Megger windings.

7619 RR-P-1B 8/14/75 2.5 Change oil.

7654 RR-P-1A 8/18/75 8.0 Change oil in motor bearings.

8579 RR-P-18 12/3/75 0.5 Megger.

8618 RR-P-1A 12/11/75 4.3 Lube motor. g 8623 RR-P-1B 12/11/75 4.25 Lube motor.

8890 RR-P-1B 2/5/75 3.0 RR-P-2B deenergized.

267 RR-P-1A 6/21/75 10.0 Replace gasket.

375 RR-P-1A 7/20/76 7.0 Repack.

1472 RR-P-1A 12/20/76 220.0 Remove silt from around pump.

1546A RR-P-1B 1/12/77 39.17 Tagged "as is" for silt removal.

O C-8 0212G051886DAR

l TMI-1 Q COMP 0NENT MAINTENANCE DATA

SUMMARY

SHEET  !

%)  :

DESIGNATOR: MDCFS (Sheet 3 of 6)

SYSTEM: Reactor Building Emergency Cooling River Water  ;

COMPONENT TYPE: Pump Site-Specific Data i Duration  !

Tag Order Component Date (hours) Work Done l 1733 RR-P-1B 2/23/77 3.58 Clean strainer. l l

720 RR-P-1A 6/10/77 5.30 Megger motors. .

RR-P-1B l

756 RR-P-1A 6/21/77 3.0 Megger motors, i 1106 RR-P-1B 9/28/77 6.0 Renew expansion  !

joints. t i

1112 RR-P-1B 9/29/77 11.35 Renew expansion joints.

l 1220 RR-P-1B 11/2/77 8.20 Oil change.

74 RR-P-1A 1/5/78 3.15 011 change.

109 RR-P-1A 1/13/78 24.0 Replace strainer.

121 RR-P.'1A 1/17/78 9.0 Replace strainer. I i

133 RR-P-1B 1/19/78 6.15 Replace strainer. I 911 RR-P-1A 7/14/78 6.40 General maintenance, i 1001 PR-P-1B 8/9/78 30.0 General maintenance.  !

1056 RR-P-1A 8/27/78 60,0 Divers removing silt.  !

1058 RR-P-1A 8/30/78 12.30 Divers removing silt. l l

1 l

l O  :

C-9 I

0212G061886DAR

TMI-1 l COMPONENT MAINTENANCE DATA

SUMMARY

SHEET l

DESIGNATOR: MDCFS (Sheet 4 of 6)

SYSTEM: Reactor Building Emergency Cooling River Water COMPONENT TYPE: Pump Site-Specific Data Duration Tag Order Component Date (hours) Work Done 1061 RR-P-1A 8/31/78 15.0 Divers removing silt.

1063 RR-P-1A 9/1/78 7.3 Divers removing silt.

1068 RR-P-1A 9/5/78 13.0 Divers removing silt.

1075 RR-P-1A 9/6/78 16.0 Divers removing silt.

1079 RR-P-1A 9/7/73 17.20 Divers removing silt.

1082 RR-P-1A 9/8/78 15.40 Divers removing silt.

1100 RR-P-1A 9/12/78 14.0 Divers removing silt.

1104 RR-P-1A 9/13/78 5.3 Divers removing silt.

1071 RR-P-1A 9/5/78 9.15 Repai r RR-V-12A, 1145 RR-P-1A 9/20/78 6.0 Installation of lube water modification.

1152 RR-P-1B 9/21/78 10.0 Installation of lube water modfication.

1171 RR-P-1B 9/25/78 11.0 Divers removing silt.

1178 RR-P-1B 9/26/78 14.15 Divers removing silt.

1197 RR-P-1B 9/28/78 13.0 Divers removing silt.

1214 RR-P-1B 10/2/78 15.3 Divers removing silt.

O C-10 0212G0518860AR

TMI-1 COMPONENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: MDCFS (Sheet 5 of 6)

SYSTEM: Reactor Building Emergency Cooling River Water COMPONENT TYPE: Pump Site-Specific Data Duration Tag Order Component Date (hours) Work Done 1223 RR-P-1B 10/4/78 15.4 Divers removing silt.

1231 RR-P-1B 10/5/78 13.0 Divers removing siit.

1241 RR-P-1B 10/9/78 17.45 Divers removing silt.

1248 RR-P-1B 10/10/78 16.45 Divers removing silt.

1251 RR-P-1B 10/11/78 15.20 Divers removing silt.

1256 RR-P-13 10/12/78 11.30 Divers removing silt.

1268 RR-P-18 10/16/78 20.15 Divers removing silt.

1275 RR-P-18 10/17/78 16.15 Divers ra-oving silt.

1282 RR-P-1B 10/18/78 20.25 Divers removing silt.

1293 RR-P-1A 10/23/78 15.3 Divers removing silt.

1296 RR-P-1A 10/24/78 15.15 Divers removing silt.

1297 RR-P-1A 10/25/78 14.0 Divers removing silt.

1305 RR-P-1A 10/26/78 16.0 Divers removing silt.

1313 RR-P-1A 10/30/78 14.3 Divers removing silt.

1316 RR-P-1A 10/31/78 14.3 Divers removing silt.

1320 RR-P-1A 11/1/78 14.3 Divers removing silt.

O C-11 0212G061886DAR

TMI-1 COMPONENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: MDCF5 (Sheet 6 of 6)

SYSTEM: Reactor Building Emergency Cooling River Water COMPONENT TYPE: Pump Site-Specific Data Duration Tag Order Component Date (hours) Work Done 1324 RR-P-1A 11/2/78 14.5 Divers removing silt.

1327 RR-P-1A 11/6/78 11.45 Divers removing silt.

1333 RR-P-1A 11/7/78 15.0 Divers removing silt.

1339 RR-P-1A 11/9/78 8.15 Divers removing silt.

O e Maintenance Data Source: Switching and Tagging Orders O

C-12 0212G0618860AR

TMI-1

("'; COMPONENT MAINTENANCE DATA

SUMMARY

SHEET a

DESIGNATOR: MDCF8 SYSTEM: Reactor Building Emergency Cooling River Water COMPONENT TYPE: Reactor Building River Water Strainer Site-Specific Data Data include only maintenance performed during periods when the reactor was not in cold shutdown, e Total noncold shutdown hours for reporting period: 6.36 y 104 Duration Tag Order Component Date (hours) Work Done 468 RR-S-1A 8/14/76 1.26 Change oil.

470 RR-S-1B 8/15/76 2.1 Change oil.

,e-

\~ /

s e Maintenance Data Source: Switching and Tagging Orders n

N-C-13 0212G061886DAR

TMI-1 COMPONENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: t!DCIl SYSTEM: Makeup COMP 0NENT YPE: Letdown Isolation Valves Site-Specific Data Data include only maintenance performed during periods when the reactor was not in cold shutdown.

e Total noncold shutdown hours for reporting period: 9.54 x 104 Duration Tag Order Component Date (hours) Work Done 6342 MU-V-28 3/24/75 8.5 Inspect wiring in MCC.

6341 MU-V-2A 3/24/75 30 Inspect wiring in MCC.

O e Mainteni:nce Data Source: Switching and Tagging Orders O

C-14 0212G061886DAR

+

l TMI-1 l COMPONENT MAINTENANCE DATA

SUMMARY

SHEET.

.l DESIGNATOR: MDCS1 i SYSTEM: Reactor Building Spray COMP 0NENT TYPE: Pump Site-Specific Data 5

. Data include only maintenance performed during periods when the reactor  ?

was not'in cold shutdown, o Total noncold shutdown hours for reporting period: 6.36 x 104 l t

Duration .

Tag Order Component Date (hours) Work Done .

5811 BS-P-1B 1/16/75 29.5 I 5971 BS-P-1A 2/6/75 1.7 Check and~ inspect breaker.

6343 BS-P-18 3/24/75 8.5 Inspect wiring for -

BS-V-38 for MCC. I s  :

6345 BS-P-1B 3/24/75 6.2 Inspect wiring. l; 6433 BS-P-1A 3/31/75 6.C Change oil.  ;

6436 BS-P-1B 3/31/75 4.0 Change oil.

6457 BS-F-1A 4/2/75 10.0 Troubleshoot i indicating light.

4 6729A BS-P-1A 4/29/75 5.5 Inspect breaker, i

301 BS-P-1A 6/24/76 3.0 Megger. l 33 BS-P-1B 3/9/77 58.0 Check breaker.  !

778 BS-P-1B 6/27/77 6.5 Maintenance, f e Maintenance Data Source: Switching and Tagging Orders  !

  • No information provided in tag order forms.  !

O i i

C-15  !

0212G061886DAR _

TMI-1 COMPONENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: MDCV1 SYSTEM: Control Building Instrument Air System COMPONENT TYPE: Compressor Site-Specific Data Data include only maintenance performed during periods when the reactor was not in cold shutdown.

e Total noncold shutdown hours for reporting period: 1.27 x 105 Duration Tag Order Component Date (hours) Work Done 5858 AH-P-9A 1/22/75 24.0 6031 AH-P-9A 2/18/75 19.5 Repair relief valve AH-P-98 19.5 on air receiver.

6074 AH-P-8A 2/24/75 0.65 Test safety valve.

AH-P-88 0.65 6175 AH-P-9A 3/10/75 0.4 Change control AH-P-9B 0.4 switch.

6418 AH-P-8A 3/27/75 2.12 Inspect and repair.

Ah-P-8B 2.12 7563 AH-P-9A 8/6/75 0.5 Change oil.

AH-P-98 0.5 872 AH-P-9A 9/14/76 0.5 Preventive AH-P-9B 0.5 maintenance.

e Maintenance Data Source: Switching and Tagging Orders

  • No information provided in tag order forms.

C-16 0212G0618860AR

i f

l THI-1 l COMP 0NENT MAINTENANCE DATA

SUMMARY

SHEET i i

DESIGNATOR: MDCV2 (Sheet 1 of 3)

SYSTEM: Control Building Ventilation f COMPONENT TYPE: Control Building Ventilation Fans 17A/B,18A/B >

Site-Specific Data

}

Data include only maintenance performed during periods when the reactor was not in cold shutdown. {

e Total noncold shutdown hours for reporting period: 1.27 x 105 Duration Tag Order Component Date (hours) Work Done .

5093 AH-E-188 10/7/74 36.0 Repair pressure [

indicator.

6407 AH-E-188 3/27/74 5.2 Inspect. breaker. l 680 AH-E-178 5/8/75 4.0 Filter change. i l

6420 AH-E-17A 23.0 Inspect, tighten, and C. 3/27/75 repair.

t 7336 AH-E-17A 6/24/75 4.5 Filter change.

{

7833 AH-E-17A 9/8/75 2.5 Weld supports onto' -l motor base frame.  !

8308 AH-E -188 10/28/75 427.0 Test AH-F-38, 8538 AH-E-17A 11/26/75 3.2 Investigate and [

repair cause for j tripping..

237 AH-E-17A 6/16/76 3.75 Install new belts. ,

1581 AH-E-178 1/21/77 4,3 General maintenance.

1724 AH-E-18A 2/19/77 31.3 Instali flow  ;

totalizer. f 1736 AH-E-18A 2/20/77 l i

1745 AH-E-18A 2/24/77 5.35 Install flow j totalizer. j C-17 0212G0618860AR

, - . - _ _ .--- ~.- _. _ .

TMI-1 COMPONENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: MDCV2 (Sheet 2 of 3)

SYSTEM: Control Building Ventilation COMPONENT TYPE: Control Building Ventilation Fans 17A/B,18A/B Site-Specific Data Duration Tag Order Coinponent Date (hours) Work Done 6 AH-E-18A 3/2/77 4.2 Inspect; calibrate.

8 AH-E -18A 3/2/77 1.1 Inspect; calibrate.

13 AH-E-18A 3/4/77 7.0 Adjust damper.

39 AH-E-18A 3/10/77 5.15 Megger.

40 AH-E-18B 3/10/77 5.15 Megger..

56 AH-E-18A 3/14/77 4.25 Test.

AH-E-188 821 AH-E-178 7/10/77 3.4 Modi fy drain plate.

AH-E-18B E33 AH-E-178 7/16/77 3.2 Repair leaks.

AH-E-188 8 34 AH-E-17A 7/16/77 6.3 Modify drain plate.

AH-E-18A 842 AH-E-178 7/19/77 83.0 Provide access for AH-E-188 condensate removal.

854 AH-E-17B 7/22/77 7.0 Repair damper.

865 AH-E-178 7/25/77 14.0 Repai r damper.

867 AH-E-17A 7/26/77 50.3 Provide better water l AH-E-18A drainage.

l l

l 9

C-18 l 0212G0618860AR 1

TMI-1 (3 COMP 0NENT MAINTENANCE DATA

SUMMARY

SHEET U

DESIGNATOR: MDCV2 (Sheet 3 of 3)

SYSTEM: Control Building Ventilation COMP 0NENT TYPE: Control Building Ventilation Fans 17A/B, 18A/B Site-Specific Data-Duration Tag Order Component Date (hours) Work Done-881 AH-E-178 7/29/77 7.2 Repair float.

AH-E-188 1034 AH-E-17A 9/14/77 4.25 Check belts.

1280 AH-E-178 11/23/77 4.1 Check motor trips.

1289 AH-E-17A 11/30/77 4.0 Install new breaker, ,

1290 AH-E-17A 11/30/77 4.2 Test.  !

1110 AH-E-18A 9/15/78 3.0 Vibration reading. ,

1502 AH-E-17A 1/1/79 19.3 Vibration reading. j i

l I

e Maintenance Data Source: Switching and Tagging Orders )

i O

C-19 0212G051886DAR

TMI-1 COMPONENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: MDCV3 (Sheet 1 of 2)

SYSTEM: Chilled Water System COMPONENT TYPE: Chilled Water System Train (pump and chiller)

Site-Specific Data Data include only maintenance performed during periods when the reactor was not in cold shutdown.

e Total noncold shutdown hours for reporting period: 6.36 x 104 Duration Tag Order Component Date (hours) Work Done 6614 AH-C-4B 4/15/75 5.5 Megger motor.

6618 AH-C-4B 4/16/75 7.15 Investigate loss of oil.

6685 AH-C-4A 4/25/75 15.75 Preventive maintenance.

6733 AH-C-4A 4/30/75 9.75 Preventive maintenance.

6904 AH-C-4B 5/23/75 48.0 Preventive maintenance.

7268 AH-C -4B 6/12/75 153.6 Replace motor.

7299 AH-C-4A 6/18/75 20.0 011 change.

256, 258, AH-C-4A 6/19/76 116.0 Turbopack teardowr.

259, 268, 289 362 AH-C-4A 7/16/76 3.3 Repair purge unit.

496 AH-C-4A 8/25/76 2.5 Repaie control circuit.

820 AH-C-4A 9/1/76 3.0 Replace purge unit.

1105 AH-C-4B 9/13/78 7.2 Calibrate condenser purge gauge.

1656 AH-C-4B 2/12/79 5.2

  • Information provided in tag order forms.

1 C-20 0212G0618860AR

l l Tol-1 COMPONENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: MDCV3 (Sheet 2 of 2)-

SYSTEM: Chilled _ Water System COMP 0NENT TYPE: Chilled Water System Train (pump and chiller)

Site-Specific Data Duration Tag Order Component Date (hours) Work Done 6191 AH-P-3A 3/12/75- 2.3 Lubricate motor.

7307 AH-P-3A '6/19/75 2.0 Inspect breaker.

7680 AH-P-3A 8/22/75 4.7 Lubricate motor.

824 AH-P-3A 9/14/76 1.25 Inspect motor.

O e Maintenance Data Source: Switching and Tagging Orders O

C-21 0212G061886DAR

TMI-1 COMP 0NENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: MDDA2 SYSTEM: Electric Power COMP 0NENT TYPE: Battery Charger Site-Specific Data Data include only maintenance performed during periods when the reactor was not in cold shutdown, e Total noncold shutdown hours for reporting period: 1.91 x 105 Duration Tag Order Component Dat e (hours) Work Done 957 1A 8/18/77 1.0 Replace trip switch. ,

736 1E 5/31/78 1.3 Trouble shoot.

O e Maintenance Data Source: Switching and Tagging Orders O

C-22 0212G0618860AR

TMI-1 -;

COMPONENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: MDDH1 (Sheet 1 of 2) .I SYSTEM: Decay Heat j COMP 0NENT TYPE: Pump Site-Specific Data l Data include c 11y maintenance performed during periods when the reactor  !

was not in cold shutdown. i I

e Total noilcold shutdown hours for reporting period: 6.36 x 104 f I

Duration i Tag Order Compor9nt Date (hours) Work Done 5368 DH-P-1A 10/14/74 2.0 Change oil. .!

5371 DH-P-1B 11/7/74 4.5 Change oil.  !

5578 DH-P-1A 12/12/74 2.0 t

3.25 5635 DH-P-1B 12/19/74 (

5812 DH-P-1B 1/15/75 2.5  !

5849 DH-P-18 1/21/75 12.0 5919 DH-P-1A 1/29/75 3.2 f 6318 DH-P-1A 3/22/75 2.0 Inspect breaker f wiring for V-5B. l DH-P-1B 3/22/75 4.16 ' Inspect breaker wiring for V-48.

6321 DH-P-1B 3/22/75 1.5 Replace breaker wiring for V-6B.

6351 DH-P-1B 3/24/75 7.0 Replace breaker wiring for V-48.

6727 DH-P-1A 4/29/75 5.0 Inspect breaker.

7460 DH-P-1B 7/16/75 3.0 Lubricate.

7477 DH-P '., 7/17/75 2.0 Change oil.

8546 DH-P-h 12/1/75 0.3 Megger.

O

  • No information provided in tag order forms.

> C-23 1 0212G0618860AR

TMI-1 COMPONENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: MDDH1 (Sheet 2 of 2)

SYSTEM: Decay Heat COMP 0NENT TYPE: Pump Site-Specific Data Duration Tag Order Component Date (hours) Work Done 151 DH-P-1B 5/27/76 7.0 Chalige oil and clean sight glass.

157 DH-P-18 5/28/76 6.2 Change oil and clean sight glass.

1003 DH-P-1B 9/3/77 19.15 Inspection.

1004 DH-P-1A 9/4/77 6.45 In spect.i on .

1103 DH-P-18 9/28/77 43.50 Align coupling.

1120 DH-P-1A 10/3/77 9.45 Align coupling. g 1221 DH-P-1A 11/1/77 19.45 Align coupling.

1226 DH-P-1B 12/2/77 8.40 Align coupling.

1228 DH-P-1A 11/3/77 10.50 Check coupling.

1286 DH P-1A 11/29/77 7.30 Oil change.

1287 DH-P-1B 11/30/77 12.15 011 change.

1301 DH-P-1A 12/5/77 9.20 Monthly inspection.

1304 DH-P-18 12/6/78 12.50 Monthly inspection.

104 DH-P-1A 1/12/78 15.0 Check coupling.

199 DH-P-1B 2/13/78 5.3 Check coupling. ]

e Maintenance Data Source: Switching and Tagging Orders O'

C-24 0212G061886DAR

TMI-1 ,

COMPONENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: MDDH3 SYSTEM: Low Pressure Injection COMPONENT TYPE: Low Pressure Injection Line Valves Site-Specific Data Data include only maintene.nce performed during periods when the reactor was not in cold shutdown.

e Total noncold shutdown hours for reporting period: 6.36 x 104 Duration Tag Order Component Date (hours) Work Done ,

5596 DH-V-4A 12/16/74 30-  !

5617 DH-V-4B 12/17/74 2 5707 DH-V-4B 12/31/74 49  !

7337A DH-V-4B 6/24/75 4 Breaker wires  ;

replaced.  ;

6226A Dh-V-7B 3/16/75 2.25 . inspect wiring size !

on breaker. l r

6604 DH-V-78 4/14/75 4.5 Reweld fitting. I 6609 DH-V-7A 4/15/75 17.5 Reweld fitting.

7411 DH-V-7A 7/11/75 18.5 Repair two welds on l vent line.

7677 DH-V-7A 8/21/75 13.75 Repair vent line ,

connection. '

8285 DH-V-7A 10/21/75 11.5 Repair weld leak l upstream of  ;

NU-V-156A. i i

e Maintenance Data Source: Switching and Tagging Orders i I

I 1

  • No information provided i; tag order forms, i C-25 0212G0618860AR

TMI-1 COMP 0NENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: MDDH4 SYSTEM: Decay Heat River Water COMPONENT TYPE: Strainer Site-Specific Data Data include only maintenance performed during periods when the reactor was not in cold shutdown, e Total noncold shutdown hours for reporting period: 6.36 x 104 Duration Tag Order Component Date (hours) Work Done 5753 DR-S-1A 1/7/75 28,25 5948 DR-S-1B 2/3/75 4.0 Check control circuit and test motor.

6001 DR-S-1A 2/12/75 11.0 Open breaker and rotate strainer by hand.

6005 DR-S-1A 2/12/75 7.3 Repair.

7221 DR-S-1A 6/6/75 48 Clean and repair, e Maintenance Data Source: Switching and Tagging Orders

  • No information provided in tag order forms.

O C-26 0212G061336DAR

l TMI-1  :

COMPONENT MAINTENANCE DATA

SUMMARY

SHEET j DESIGNATOR: MDEF1 (Sheet 1 of 2)

SYSTEM: Emergency Feedwater COMPONENT TYPE: Pump (motor-driven EF-P-2A and 28, turbine-driven EF-P-1) . .

Site-Specific Data - j Data include only maintenance performed during periods when the reactor  !

was not in cold shutdown.  !

e Total noncold shutdown hours for reporting period: 9.54 x 104 Duration Tag Order Component. Date (hours) Work Done [

5056 EF-P-2A 9/26/74 1.5 Change oil in motor.

5058 EF-P-2B 9/26/74 2.5 Change oil in motor, f

5056 EF-P-2A 9/30/74 2.75 Change'011.

1 EF-P-2B j i  !

5820 EF-P-2A 1/16/75 0.67 j 5821 EF-P-2B 1/16/75 1.25  !

! 5917 EF-P-2A 1/29/75 3.3  !

5920 EF-P-2B 1/29/75 3.5 '

f 5958 EF-P-2A 2/5/75 2.2 Change oil. j EF-P-28 ,

5995 EF-P-2A 2/11/75 2.5 Quarterly preventive  !

maintenance. I i

6160 EF-P-2A 3/6/75 4.0 Change oil in j EF-P-2B motors.  :

6402 EF-P-1 3/27/75 10.5 Inspect breaker V-1B.

6403 EF-P-1 3/27/75 10.5 Inspect breaker V-28 l

6735 EF-P-1 4/30/75 2.25 011 change. i 4

7757A EF-P-28 5/29/75 4.25 Megger motor for  ;

, preventive  ;

maintenance.

l l

  • No information provided in tag order forms. l

/

C-27 l 0212G061886DAR I

i

.. ,n,.. - ~ _ , . . , - - . . - .,- - ,., ,- . ..- ,- - ... - , , ~ . _ - , , . . . . - . . , - , .

TMI-1 COMPONENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: MDEF1 (Sheet 2 of 2)

SYSTEM: Emergency Feedwater COMPONENT TYPE: Pump (motor-driven EF-P-2A and 2B, turbine-driven EF-P-1)

Site-Specific Data Duration Tag Order Component Date (hours) Work Done 7899 EF-P-1 9/16/75 4.75 Valyes MS-10A, 108, 13A, and 13B deenergized closed.

8724 EF-P-2A 1/2/76 1.0 Change oil.

8725 EF-P-2B 1/2/76 1.5 Change oil, 1079 EF-P-2A 10/29/76 3.5 Change. oil, 1080A EF-P-23 10/29/76 1.5 Change oil.

1081 EF-P-1 10/29/76 3.0 Change oil, g 966 EF-P-1 8/22/77 4.30 Change oil.

1229 EF-P-1 10/4/77 5.4 Change oil.

930 EF-P-2A 8/10/77 1.3 Change oil.

937 EF-P-28 8/12/77 2.3 Change oi1.

e Maintenance Data Source: Switching and Tagging Orders O

C-28 0212G0618860AR

TMI-1 f COMPONENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: MDEF2 ,

3YSTEM: EFW '

i COMP 0NENT TYPE: Pump Steam Supply Valve  !

Site-Specific Data j Data include only maintenance performed during periods when the reactor was not fn cold shutdown..  !

e Toi.31 noncold shutdown hours for reporting period: 1,59 x 10 4 Duration Tag Order Component Date (hours) Work Done T

9309 MS-V-108 3/6/75 1,319.3 Disconnect wiring [

to replace gasket.  !

6849 M3-V-10B 5/16/76 863.5 Correct problem.  ;

7662, 7663, MS-V-13B 8/20/75 29.00 Repair, trouble--

7679, 7668, shoot control  :

7674 circuit. .

462 MS-V-10A 8/13/76 1.0 Troubleshoot l control circuit and i motor. l i

822, 517 MS-V-10B 8/31/76 8.0 Megger. ,

t 873 MS-V-10A 9/14/76 4.5 Preventive  !

maintenance.  !

1704 MS-V-10A 2/17/77 34.0 Electrical ,

maintenance. l Electrical 981 MS-V-10B 8/24/77 5.35 maintenance. i 984 MS-V-10B 8/24/77 0.3 Test.

i 985 MS-V-10A 8/25/77 19.0 , Electrical l maintenance.

l e Maintenance Data Source: Switching and Tagging Orders

{

}

O  !

C-29 0212G061886DAR  !

l

TMI-1 COMPONENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: MDFW1 (Sheet 1 of 2)

SYSTEM: Main Feedwater COMP 0NENT TYPE: Condensate Pump Site-Specific Data Data include only maintenance performed during periods when the reactor was not in cold shutdown.

e Total noncold shutdown hours for reporting perioc': 9.54 x 104 Du ration Tag Order Component Date (hours) Work Done 4946 C0-P-1A 9/5/74 0.67 Clean suction strainer.

5069 C0-r-1C 9/30/74 1.75 Change oil.

5075 CO-P-1B 10/1/74 1.50 Change oil.

5033 C0-P-1A 10/2/74 1.50 Change oil.

6336 C0-P-1C 3/22/75 1.50 Lube flow indication.

7613 C0-P-1A 8/14/75 7.25 Change oil.

7742 C0-P-1A 8/28/75 1.30 Megger motor for preventive maintenance.

7745 CO-P-1B 8/28/75 2.0 Megger motor for preventive maintenance.

7746 C0-P-1C 8/28/75 1.50 Megger motor for preventive maintenance.

8544 C0-P-1C 12/1/75 1.60 Megger motor for preventive maintenance.

515 C0-P-1A 8/31/76 1.50 Change oil.  !

888 CO-P-1A 9/16/76 0.50 Megger motor.  ;

1401, 1404, C0-P-1A 12/3/76 618.2 Remove and install 1405, 1411, motor.  ;

1412, 1495,  !

1497, 1502  !

l C-30 i l 0212G0618860AR  !

TMI-1 COMPONENT MAINTENANCE DATA

SUMMARY

SHEET

(~'))

L DESIGNATOR: MDFW1 (Sheet 2 of 2)

SYSTEM: Main Feedwater COMP 0NENT TYPE: Condensate Pump Site-Specific Data Duration Tag Order Component Date (hours) Work Done 1507 CO-P-1A 12/30/76 5.25 Adjust packing.

1517 C0-P-1A 1/3/77 4.40 Change oil.

1521 C0-P-1B 1/3/77 3.50 Change oil.

1525 C0-P-1C 1/4/77 3.15 Change oil.

1528 C0-P-1C 1/4/77 4.00 Check a,lignment.

1036 C0-P-1A 9/14/77 4.15 Repair suction.

/~N L)i e Maintenance Data Source: S vitching and Tagging Orders v

C-31 0212G0618860AR

TMI-1 COMPONENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: MDFW2 (Sheet 1 of 3)

SYSTEM: Main Feedwater COMP 0NENT TYPE: Condensate Booster Pump Site-Specific Data Data include only maintenance performed during periods when the reactor was not in cold shutdown.

e Total noncold shutdown hours for reporting period: 9.54 x 10 4 Duration Tag Order Component Date (hours) Work Done 4968, 4969 C0-P-2C 9/10/74 20.5 Repair leak.

5448 C0-P-2C 11/20/74 34.0 Maintenance.

5459 CO-P-2C 11/22/74 2.2 Change oil.

5470 C0-P-2A 11/33/74 1.5 Change oil.

5672 C0-P-2A 12/26/74 2.15 5684 C0-P-2B 12/27/74 78.0 5702 C0-P-2A 12/31/74 5.0 6173 C0-P-28 3/10/75 4.0 Change oil.

6177 C0-P-2A 3/11/75 3.75 Change oil.

6186 C0-P-2C 3/11/75 1.0 Change oil.

6487 C0-P-2A 4/3/75 3.1 Charge oil.

6494 C0-P-2A 4/4/75 1.5 Change oil.

6496A C0-P-2A 4/5/75 2.0 Change oil.

6840 C0-P-2A 5/15/75 3.6 Investigate low oil pressure.

7570 C0-P-2C 8/7/75 1.6 Reset relief valve.

  • No information provided in tag order forms.

C-32 0212G0618860AR

TMI-1 COMP 0NENT MAINTENANCE DATA

SUMMARY

SHEET

(~')

\s DESIGNATOR: MDFW2 (Sheet 2 of 3)

SYSTEM: Main Feedwater COMPONENT TYPE: Condensate Booster Pump Site-Specific Data Duration Tag Order Component Date (hours) Work Done 7605 C0-P-2A 8/13/75 2.6 Work on pump motor outboard bearings thermocouple .

7743 C3-P-2A 8/; J/75 0.6 Megger motor for preventive maintenance.

7791 C0-P-2C 9/2/75 10.0 Repai r C0-P-9C.

7880 C0-P-2A 9/12/75 1.6 011 change. 'l 7886 C0-P-28 9/12/75 4.6 011 change.

sl 8224, 8229, C0-P-2A 10/16/75 55 Repair pump.

8274 8290, 8291 C0-P-2A 10/23/75 13.5 Repair seal.

8525 C0-P-2A 11/25/75 2.0 Repair seal.

8535 C0-P-2A 11/26/75 8.5 Repair seal.

876 C0-P-28 9/14/76 2.25 Change oil in motor.

894 CO-P-2A 9/16/76 1.30 Preventive maintenance.

897 C0-P-2C 9/17/76 1.0 Preventive maintenance.

1731 C0-P-2B 2/23/77 0.4 Change oil.

1732 C0-P-2C 2/23/77 1.50 Change oil.

666 CO-P-2B 5/18/77 3.45 Change oil.

f~h)

L l

C-33  :

0212G061886DAR  !

i

TMI-1 COMPONENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: MDFW2 (Sheet 3 of 3)

SYSTEM: Main Feedwater COMPONENT TYPE: Condensate Booster Pump Site-Specific Data Duration Tag Order Component Da'a (hours) Work Done 987 C0-P-2B 8/26/77 3.20 Change oil.

688 C0-P-2A 5/15/78 184.0 Repair seal .

1478 C0-P-28 12/20/78 4.10 1483 C0-P-23 12/21/78 6.30 1511 C0-P-23 1/3/79 26.0 1521 C0-P-28 1/4/79 94.0 O

e Maintenance Data Source: Switching and Tagging Orders

  • No informati o7provided in tag order forms.

C-34 0212G0618860AR

L TMI-1 _

COMP 0NENT MAINTENANCE DATA

SUMMARY

SHEET ~

DESIGNATOR: MDGA1 (Sheet 1 of 8)

SYSTEM: Electric Power COMPONENT TYPE: Diesc1 Generator  ?

Site-Specific Data l

Data include only maintenance performed during periods when the reactor-was not in cold shutdown. 3 i

4 e Total noncold~ shutdown hours for reporting period: 6.36 x 10 ' {

i Duration {

Tag Order Component Date (hours) Work Done l 5320, 5325, DG-1B 10/28/74 56.0 Inspection. l 5335, 5337,  ;

5338, 5343 l 5359 DG-1A 11/4/74 4.5 Pull piston.

5757, 5758, DG-1A 1/8/75 53 Start air valve  :

5771 EG-V-15A closed.

5785, 5787 DG-1A 1/13/75 13.5 t 5800, 5801, 5802 DG-1B 1/14/75 80 Start air valve EG-V-158 closed, j

j l 5830 DG-1B 1/17/75 5.5 Start air valve i EG-V-15B closed.

i 5831 DG-1A 1/18/75 2.67 Start air valve  ;

EG-V-15A closed.

3 5842 DG-1B 1/20/75 9.5 Start air valve -

EG-V-15B closed. l 5848 DG-1B 1/21/75 10.5 Start air valve EG-V-15B closed.

5854 DG-1B 1/22/75 10.0 Start air valve I EG-V-15B closed.

5863 DG-1A 1/27/75 107.25 Start air valve l EG-V-15A closed.  :

  • No information provided in tag order forms.

4 l C-35  !

0212G061886DAR i i

TMI-1 COMP 0NENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: MDGA1 (Sheet 2 of 8)

SYSTEM: Electric Power COMPONENT TYPE: Diesel Generator Site-Specific Data Duration Tag Order Component Gate (hours) Work Done 5952 DG-1B 2/4/75 8.0 Start air valve EG-V-15B closed.

5979 DG-1A 2/7/75 5.5 Start air valve EG-V-15A closed.

6098 DG-1B 2/27/75 10.0 Adjust injection pumps; repair leaks.

6138, 6143 DG-1B 3/4/75 c3.0 Repair governor.

6660 DG-18 4/22/75 4.0 Connect governor motor.

6745 DG-1A 5/1/75 2.0 Inspect breaker.

7287 DG-1B 6/16/75 5.6 Change governor.

7297 DG-1B 6/17/75 1.0 Change megawatt meter.

7399 DG-1A 7/8/75 51.3 Inspect blower drive gear.

7418 DG-1B 7/14/75 53.0 Inspect blower drive gear.

7453 DG-1A 7/16/75 4.95 Inspect intake ai r piping.

7466 DG-1B 7/17/75 2.25 7604 DG-1B 8/13/75 8.0 Clean and check relay cab. l i

i

  • No information provided in tag order forms.

C-36 0212G061886DAR 4

w n

TMI-1 COMP 0NENT MAINTENANCE DATA

SUMMARY

SHEET O'y/

DESIGNATOR: M0GA1 (Sheet 3 of 8)

SYSTEM: Electric Power COMP 0NENT TYPE: Diesel Generator Site-Specific Data Duration Tag Order C_omponent Date (hours) Work Done 7612 DG-1A 8/14/75 3.5 Clean and check relay Cab.

8202 DG-1A 10/14/75 9.0 Starting air ,

compressor deenergized.

8777, 8776, DG-1B 1/9/76 100.67 Preventive 8765 maintenance, calibrate instrument.

8815 DG-1B 1/19/76 17.0 Calibrate C'o instrumentation.

8846 DG-1A 1/24/76 4.5 --

8847 OG-1B 1/24/76 3.75 Preventive maintenance.

8823, 8842 DG-1B 1/23/76 5.0 Surveillance.

8854, 8853 DG-1A 1/26/76 70.0 Repai r.

8869 DG-1A 1/29/76 2.0 Inspect CAM.

8871, 8874 OG-1A 1/30/76 7.0 Calibrate instrumentation.

8880 OG-1B 2/2/76 36.75 Repair leaks in exhaust system.

8892 DG-1B 2/6/76 2.75 Change gasket on exhaust.

  • No information provided in tag order forms.

C-37 0212G061886DAR

TMI-1 COMPONENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: MDGA1 (Sheet 4 of 8)

SYSTEM: Electric Power COMP 0NENT TYPE: Diesel Generator Site-Specific Data Duration Tag Order Component Date (hours) Work Done 8928 DG-1A 2/13/76 5.0 Clean oil on engine.

8843, 8941 DG-1B 2/17/76 7.5 Repair.

242 DG-1A 6/16/76 4.5 Starting air receivers vented.

419 DG-1A 7/30/76 13.0 Replace coolant.

455 DG-1A 8/8/76 4.5 Set governor, change oil.

458 DG-1B 8/11/76 4.0 S governor, change 1534 DG-1A 1/7/77 9.50 Yearly inspection.

1546 DG-1A 1/11/77 17.45 WR-18139.

1547 DG-1A 1/11/77 26.30 Yearly inspection.

1575 DG-1A 1/20/77 20.25 Install servo motor.

1577 DG-1A 1/20/77 20.25 Install resistors.

1585 DG-1A 1/22/77 6.0 Test relays.

1593 DG-1A 1/24/77 1.2 Generator maintenance.

1594 DG-1A 1/24/77 8.15 Generator maintenance.

1596 DG-1A 1/25/77 3.2 Generator maintenance.

O C-38 0212G061886DAR

TMI-1

, COMPONENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: MDGA1 (Sheet 5 of 8)

SYSTEM: Electric Po ar COMPONENT TYPE: Diesel Generator Site-Specific Data Duration Tag Order Component Date ( hou;'s ) Work Done 1603 DG-1B 1/26/77 Preventive maintenance.

1604 DG-1B 1/26/77 27.0 Preventive maintenance.

1607 DG-18 1/27/77 Relay timing.

1627 DG-18 1/31/77 16.50 Annual preventive maintenance.

1642 DG-1B 2/3/77 14.35 Annual preventive maintenance.

t 1665 DG-18 2/8/77 10.50 Annual preventive maintenance.

1671 DG-1B 2/9/77 9.55 Annual preventive maintenance.

37 DG-1B 3/10/77 6.0 Fuel leak.

1030 DG-1B 9/13/77 72.0 Install automatic voltage control .

1121 DG-1A 10/3/77 7.45 Repair air box.

1137 DG-1A 10/5/77 4.25 Install rheostat.

1160 DG-1A 10/12/77 5.0 Repair starting circuit.

1202 DG-1A 10/27/77 11.30 Install new resistors in excitor.

1205 DG-18 10/28/77 7.20 Install new resistors in excitor.

O C-39 0212G051886DAR

TMI-1 COMPONENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: MDGA1 (Sheet 6 of 8)

SYSTEM: Electric Power COMPONENT _ TYPE :Diesel Generator Site-Specific Data Duration Tag Order Component Date (hours) ,

Work Done 1236 DG-1A 11/4/77 47.5 Annual calibration.

1240 DG-1B 11/6/77 38.4 Annual calibration.

1307 DG-1B 12/6/77 53.0 Preventive maintenance.

5 DG-1B 12/12/77 26.15 Preventive maintenance.

10 DG-1B 12/12/77 2.4 Annual inspection.

13 DG-1B 12/12/77 4.45 T mers test.

2.8 DG-1A 12/18/77 29.0 Pre ventive maintenance.

g 33 DG-1A 12/20/77 66 DG-1B 1/3/78 10.15 Clean filters.

70 DG-18 1/4/78 3.2 Inspect leak.

72 DG-1A 1/4/78 17.4 Preventive maintenance.

84 DG-1A 1/6/78 4.45 Replace heat shields.

113 DG-1B 1/15/78 5.10 Check timing.

184 DS-1A 2/8/78 27.55 Set injectors.

251 DG-1B 3/11/78 13.3 Surveillance.

253 DG-1B 3/12/78 3.3 Replace switch.

O C-40 0212G061886DAR

TMI-1

,Q COMPONENT MAINTENANCE DATA

SUMMARY

SHEET U

DESIGNATOR: MDGA1 (Sheet 7 of 8)

SYSTEM: Electric Power COMPONENT TYPE: Diesel Generator Ete-SpecificData Duration Tag Order Component Date (hours) Work Done 916 OG-1A 7/17/78 26.45 Wiring cnanges.

947 DG-1B 7/25/84 11.3 Wiring changes.

1107 DG-1B 9/13/78 12.0 Repair governor.

1131 DG-18 9/18/78 6.0 Install new governor.

1177 DG-1B 9/25/78 6.0 Replace, governor.

1325 DG-1A 11/3/78 5.35 Calibration.

. 1560, 1562 DG-1B 1/16/79 119.0 1586 DG-1B 1/22/79 8.15 1596, 1597 DG-1A 1/24/79 83.0 1613 OG-1A 1/28/79 7.3 1618 DG-1A 1/29/79 14.0 1647 DG-1B 2/7/79 5.0 1649 DG-1A 2/7/79 2.3 1653 DG-1A 2/9/79 3.15 1659 DG-18 2/12/79  ;?6.0 1666 DG-1B 2/13/79 24.3

  • No information provided in tag order forms.

O V

C-41 0212G061886D/A

TMI-1 COMP 0NENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: MDGA1 (Sheet 8 of 8)

SYSTEM: Electric Power COMPONENT TYPC: Diesel Generator Site-Specific Data Duration Tag Order Component Date (hours) Work Done 1582 DG-1A 1/21/77 2.0 Calibrate instrumentation.

1719 DG-1A 2/18/77 1.5 Change oil.

DG-1B 34 DG-1A 3/9/77 0.3 Replace belts.

1314 DG-1B 12/8/77 2,15 Calibra. tion.

21 DG-1B 12/16/77 6.3 893 DG-1B 7/6/78 1.2 Replace belts.

1140 DG-1A 9/19/78 3.0 Reset relief valve.

1233 DG-1A 10/5/78 3.15 Repair leak, o Maintenance Data Source: Switching and Tagging Orders

  • No information provided in tag order forms.

O C-42 0212G061886DAR

TMI-1

(~')

i /

COMPONENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: MDGA2 SYSTEM: Electric Power COMP 0NENT TYPE: Diesel Generator Fuel Oil Transfer Pump Site-Specific Data Data include only maintenance performed during periods when the reactor was not in cold shutdown.

e Total noncold shutdown hours for reporting period: 6.36 x 104 Duration Tag Order Component Date (hours) Work Done 5141 1C 10/16/74 1.25 Test on/off switch.

5559 1B 12/9/74 3,25 Check contacts on auxiliary relay.

(G e Maintenance Data Source: Switching and Tagging Orders v

C-43 0212G061886DAR

TMI-1 COMPONENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: MDHA1 (Sheet 1 of 6)

SYSTEM: Decay Heat River Water COMPONENT TYPE: Pump Site-Specific Data Data include only maintenance performed during periods when the reactor was not in cold shutdown.

e Total noncold shutdown hours for reporting period: 6.36 x 104 Duration Tag Order Component Date (hours) Work Done 5008 DR-P-1A 9/16/74 5.0 Lubricate motor.

5042 DR-P-1B 9/23/74 1.0 Change oil and lubricate motor.

5768 DR-P-1A 1/10/75 27.75 5905 DR-P-16 1/28/75 25.50 g

6239 DR-P-1A 3/17/76 2.3 Lubricate motor.

6359 DR-P-1B 3/19/75 7.6 Lubricate motor.

6365 DR-P-1B 3/25/75 8.0 Inspect breaker wiM ng (DR-S-1B).

6366 DR-P-1B 3/25/75 8.0 Ir.spect breaker wiring (DR-P-28).

6367 DR-P-18 3/25/75 8.0 Inspect breaker wiring (DR-V-1B).

6465 DR-P-1B 4/3/75 3.5 Repair and clean lines.

6488 DR-P-1A 4/3/75 1.2 Repair and clean lines.

6902 DR-P-1B 5/21/75 7.45 Deenergized. 4 6902 DR-P-1B 5/23/75 48 Repair shaft.

7275 DR-P-1B 6/13/15 4.75

  • No information provided in tag order forms.

l C-44 0212G0618860AR 1

TMI-1 G COMPONENT MAINTENANCE DATA

SUMMARY

SHEET

[V DESIGNATOR: MDHA1 (Sheet 2 of 6)

SYSTEM: Decay Heat River Water COMP 0NENT TYPE: Pump Site-Specific Data Duration Tag Order Component Date (hours) Work Done 7279 DR-P-1A 6/14/75 0.5 7302, 7309 DR-P-1B 6/19/75 35.5 Remove motor, repair.

8884 DR-P-1B 2/4/76 3.25 Clean and repair check valve (DR-P-28).

8891 DR-P-1A 2/5/75 2.0 Unclog drain line.

7279 DR-P-1A 6/14/76 24.0 (estimated)

O V

467 DR-P-18 8/13/76 3.0 Clean strainer

( DR-S-28 ) .

871 DR-P-1A 9/14/76 4.75 Semiannual preventive maintenance.

1059 DR-P-1A 10/28/76 6.0 Repack str31ner.

1472 DR-P-1A 10/20/76 220.0 i 1546A DR-P-1A 1/12/77 39.20 Tagged "as is" for silt removal.

58 DR-P-1A 3/14/77 2.5 Repack pump.

  • No information provided in tag order forms.

O O

C-45 0212G061886DAR ,

TMI-1 COMPONENT MAINTENANCE DATA

SUMMARY

SHEET DESIGMTOR: MDHA1 (Sheet 3 of 6)

SYSTEM: Decay Heat River Water COMPONENT TYPE: Pump Site-Specific Data Duration Tag Order Component Date (hours) Work Done 873 DR-S-1A 7/27/77 10.3 Change oil.

876 DR-S-1B 7/28/77 3.3 Change oil.

905 DR-P-1A 8/4/77 4.0 Change oil.

907 DR-P-1B 8/4/77 3.0 Change oil.

1213 DR-P-1A 10/31/77 3.2 Repair strainer.

1285 DR-P-1A 11/29/77 10.3 Replace end bell on motor.

1317 DR-P-1B 12/9/77 17,4 Repack pump.

963 DR-P-1A 7/31/78 5.45 Clean tubes.

986 DR-P-1A 8/4/78 1.25 Megger.

944 DR-P-1B 8/7/78 55.0 Clean tubes.

1010 DR-P-1A 8/13/78 2.45 Unplug drain valve.

1110 DR-P-18 9/13/78 7.20 Divers removing silt.

1113 DR-P-1B 9/14/78 14.0 Divers removing silt.

1100 DR-P-1A 9/12/78 14.0 Divers removing silt.

1100 DR-P-1B 9/12/78 14.0 Divers removing silt.

1104 DR-P-1A 9/13/78 5.3 Divers removing silt.

O C-46 0212GD61886DAR

3 i

r TMI-1 COMP 0NENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: MDHA1 (Sheet 4 of 6)  !*

SYSTEM: Decay Heat River Water COMPONENT TYPE: Pump l

Site-Specific Data i I

)

  • Duration 2

Tag Order Component- Date (hours) Work Done f t

1104 DR-P-1B 9/13/78 5.3 Divers removing silt.  ;

1 1130 DR-P-1B 9/18/78 17.2 _ Divers removing silt. l t

1137 DR-P-13 9/19/78 14.35 Divers removing silt.  !

i 1144 DR-P-1B 9/20/78 17.3 Divers removing silt.

4 1150 DR-P-1B 9/21/78 6.0 Divers , removing silt.

1134 DR-P-1B 9/18/78 12.0 Lube water modi fi cation.*

l t .

1171 DR-P-1A 9/25/78 11.0 Divers removing silt. {

i

1178 DR-P-1A 9/26/78 14.15 Divers removing silt.  !

1193 DR-P-1A 9/27/78 7.3 Lube water

, modi fication.  ;

i 1197 DR-P-1A 9/28/78 13.0 Divers removing silt. l i

! 1214 DR-P-1A 10/2/78 15.3 Divers removing silt.  ;

1223 DR-P-1A 10/4/78 15.4 Divers removing silt.

j 1231 DR-P-1A 10/5/78 13.0 Divers removing silt, f i

1241 DR-P-1A 10/9/78 17.45 Divers removing silt.  !

i l

  • 0verlap with tag order 1130. l t

t i

O  !

3

! C-47 I 0212G061886DAR l

TMI-1 COMPONENT MAINTENANCE DATA

SUMMARY

SHEET g

DESIGNATOR: MDHA1 (Sheet 5 of 6)

SYSTEM: Decay Heat River Water COMP 0NENT TYPE: Pump Site-Specific Data Duration Tag Order Component Date (hours) Work Done 1248 DR-P-1A 10/10/78 16.45 Divers removing silt.

1251 DR-P-1A 10/11/78 15.2 Divers removing silt.

1256 DR-P-1A 10/12/78 11.30 Divers removing silt.

1268 DR-P-1A 10/16/78 20.15 Divers removing silt.

1275 DR-P-1A 10/17/78 16.15 Divers. removing silt.

1282 DR-P-1A 10/18/78 20.25 Divers removing silt.

1285 DR-P-1B 10/19/78 15.30 Divers removing silt. g 1293 DR-P-1B 10/23/78 15.30 Divers removing silt.

1296 DR-P-1B 10/24/78 15.15 Divers removing silt.

1297 DR-P-1B 10/25/78 14.0 Divers removing silt.

1305 DR-P-1B 10/26/78 16.0 Divers removing silt 1313 DR-P-1B 10/30/78 14.3 Divers removing silt.

1316 DR-P-1B 10/31/78 14.3 Divers removing silt.

1320 DR-P-1B 11/1/78 14.3 Divers removing silt.

1324 DR-P-1B 11/2/78 14.5 Divers removing silt.

1327 DR-P-1B 11/6/78 11.45 Divers removing silt.

O l C-48 0212G0618860AR

TMI-1

(~'T. COMP 0NENT MAINTENANCE DATA

SUMMARY

SHEET

<J DESIGNATOR: MDHA1 (Sheet 6 of 6)

SYSTEM: Decay Heat River Water COMPONENT TYPE: Pump Site-Specific Data Duration Tag Order Component Date (hours) Work Done 1333 DR-P-1B 11/7/78 15.0 Divers removing silt.

1339 DR-P-1B 11/8/78 8.15 Divers removing silt.

1357 DR-P-1A 11/30/78 4.0 Test.

1391 DR-P-1A 11/21/78 5.0 Inspection.

1409 DR-P-1A 11/28/78 7.3 1

1521 DR-P-1A 1/8/79 3.5 fi 1530 DR-P-1A 1/8/79 10.0 i G

l e Maintenance Data Source: Switching and Tagging Orders l

I

  • No information provided in tag order forms. l 1

,i")

l 1

C-49 2 i

0212G0618860AR I

TMI-l COMPONENT MAINTENANCE DATA

SUMMARY

SHEET g

DESIGNATOR: MDHA3 SYSTEM: Decay Heat River Water COMPONENT TYPE: Strainer Site-Specific Data Data include only maintenance performed during periods when the reactor was not in cold shutdown, e Total noncold shutdown hours for reporting period: 6.36 x 104 Duration Tag Order Component Date (hours) Work Done 5753 OR-S-1A 1/7/75 28.25 5948 DR-S-1B 2/3/75 4.0 Check control circuit and test motor.

6001 DR-S-1A 2/12/75 11.0 Open breaker and rotate strainer by hand.

6005 DR-S-1A 2/12/75 7.3 Repair.

7221 DR-S-1A 6/6/75 48 Clean and repair.

e Maintenance Data Source: Switching and Tagging Orders

  • No information provided in tag order forms.

O C-50 0212G0618860AR

TMI-1

("N COMPONENT MAINTENANCE DATA

SUMMARY

SHEET C

DESIGNATOR: MDHA4 SYSTEM: Nuclear Services Closed Cooling Systen COMPONENT TYPE: Nuclear Services Coolers Site-Specific Data Data include only maintenance performed during periods when the reactor was not in cold shutdown.

e Total noncold shutdown hours for reporting period: 1.27 x 105 Duration Tag Order Component Date (hours) Work Done 639 NS-C-1A 5/1/78 84.0 Clean tubes.

690 NS-C-1B 5/15/78 183.0 Clean tubes.

709 NS-C-1C 5/22/78 213.0 Clean tubes.

756 NS-C-10 6/5/78 1,274.0 Clean tubes.

(

'v' 948 NS-C-10 7/26/76 2.0 Clean tubes.

e Maintenance Data Source: Switching and Tagging Orders t'3 (v]

C-51 0212G061886DAR

TMI-1 COMPONENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: itDHA5 SYSTEM: Decay Heat Close Cycle COMPONENT TYPE: Pump Site-Specific Data Data include only maintenance performed during periods when the reactor was not in cold shutdown, e Total noncold shutdown hours for reporting period: 6.36 x 10 4 Duration Tag Order Component, Date (hours) Work Done 5031 DC-P-1A 9/20/74 3.8 Repair ground in circuit.

6835 DC-P-1A 5/14/75 0.33 Change oil.

7927 DC-P-1A 9/19/75 3.5 Change oil.

7929 DC-P-1B 9/19/75 2.0 Change oil.

6206 DC-P-1A 3/14/75 1.0 Lubricate motor.

6209 DC-P-1B 3/14/75 1.0 Lubricate motor.

6830 DC-P-1B 5/13/75 0.8 Change oil.

8360 DC-P-1A 11/10/75 0.25 Lubricate motor.

8361 DC-P-1A 11/10/75 0.25 Lubricate motor.

854E DC-P-1B 12/1/75 3.0 Megger.

8567 DC-P-1A 12/3/75 1.5 Megger.

8584 DC-P-1A 12/4/75 1.3 Install new secondary contact block on breaker, 104S DC.P-1B 10/23/76 1.7 Change oil.

1067 DC-P-1A 10/30/76 1.0 Change oil.

e Maintenance Data Source: Switching and Tagging Orders h

C-52 0212G061886DAR

THI-1 COMP 0NENT MAINTENANCE DATA

SUMMARY

SHEET

[}

DESIGNATOR: MDHL1 SYSTEM: DHR C OMPONENT_T leg Suction Line Valves VI and V2 Site-Speci, Data include .ntenance performed during periods when the reactor' was not in col.

- .down.

. e Total noncold shutdown hours for reporting period: 6.36 x 104 Duration Tag Order Component Date (hours) Work Done 5130 DH-V1 10/14/74 4.75 Remove VI automatic interlock bistable, repair.

5378 DH-V1, DH-V2 10/14/74 2.50 Replace triac in ESASS system.

O

, 4 i

J

, o Maintenance Data Source: Switching and Tagging Orders O

C-53 0212G061886DAR 1 i

. _ . . _ _ _ _ _ _ - . _ _ . . . - , . _ . . . . _ ~ . . . _ - . . , . . _

TMI-1 COMP 0NENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: MDHP1, MDHP2 (Shuet 1 of 4)

SYSTEM: Makeup -

COMP 0NENT TYPE: Pumps A, B, and C*

Site-Specific Data .

Data include only maintenance performed during periods when the reactor was not in cold shutdown.

e Total noncold shutdown hours for reporting period: Pump B = 3.18 x 104 Pumps A and C = 6.36 x 104 ,

Duration Tag Order Component Date (hours) Work Done 5112, 5113, MU-P-1A 10/10/74 34.75 Repair pump motor.

5120 5150 MU-P-1C 10/17/74 1.67 Test breaker.

5590, 5591 MU-P-1C 12/13/74 1.25 5592 MU-P-1C 12/13/74 4.0 5726 MU-P-1C 1/3/75 3.2 5762 MU-P-1C 1/8/75 3.0 5813 MU-P-1C 1/15/75 5.0 5923 MU-P-1B 1/30/75 2.0 6003 MU-P-1C 2/12/75 1.0 Inspect BDD for loose connections, nuts, etc.

6006 MU-P-1A 2/12/75 0.5 Inspect BDD.

6045 MU-F-1B 2/11/75 1.0 Inspect 800.

6157 MU-P-1C 3/6/75 5.5 Repair bearing oil leak.

  • Used pump B data for MDHP1 and pumps A and C data for MDHP2.
    • No information provided in tag order forms.

C-54 0212G061886DAR

T TMI-1 COMPONENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: MDHP1, MDHP2 (Sheet 2 of 4)

F SYSTEM: Makeup COMPONENT TYPE: Pumps A, B, and C* f Site-Specific Data Duration  !

Tag Order Component Date (hours) Work Done 6266, 6279 MU-P-1A 3/19/75 7.25 Replace lube oil i cooler temperature  ;

indicator.  !

6310 MU-P-1B 3/21/75 1.0 Inspect breaker f wiring. j 6604 MU-P-1C 4/14/75 4.5 Reweld fitting.  !

. t 6609 MU-P-1A 4/15/75 17.5 Reweld fitting. r 6728 MU-P-1B 4/29/75 5.0 Inspect breaker. l O 6730 MU-P-1A 4/29/75 8.0 Inspect breaker, j

6738 MU-P-1C 4/30/75 8.67 Change oil and check i leaks.  !

6748 MU-P-1C 5/2/75 1.0 Change oil.

6759 MU-P-1A 5/5/75 6.2 Change oil and repair  !

leak.

6787 MU-P-1B 5/7/75 3.15 Change oil, i

6802, 6803 MU-P-1C 5/11/75 5.3 Repair suction valve.

5 6869 MU-P-1B 5/18/75 3.2 Repair auxiliary oil pump pressure switch.  !

l

  • Used pump B data for MDHP1 and pumps A and C data for MDHP2.

O C-55 0212G061886DAR i

I

. --- , - - 1r - ~ - . , - - - . - -

, . - - - , ,-4 , ,,,,,r., w, .____m,-

TMI-1 COMPONENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: MDHP1, MDHP2 (Sheet 3 of 4)

SYSTEM: Makeup COMPONENT TYPE: Pumps A, B, and C* ,

Site-Specific Data Duration Tag Order Component Date (hours) Work Done 7411 MU-P-1A 7/11/75 18.5 Repair two welds on vent line.

7614 MU-P-1B 6/14/75 7.5 Repair oil leaks.

7677 MU-P-1A 8/21/75 13.75 Repair weld leak.

8166 MU-P-1B 10/9/75 2.0 Repair oil leak.

8285 MU-P-1A 10/?1/75 11.5 Repair weld leak upstream of MU-V-156A.

8311 MU-P-1B 10/29/75 2.6 Repair oil .

8901 MU-P-1A 2/9/76 1.75 Replace broken green field connection.

1057 MU-P-1B 10/28/76 4.2 Change oil.

1058 MU-P-1A 10/28/76 3.0 Change oil.

1061 MU-P-1C 10/29/76 3.2 Change oil.

670 MU-P-1A 5/19/77 7.35 Change oil.

1187 MU-V-14B 10/22/77 8.0 Repack valve.

50, 52 MU-P-1B 12/27/77 31.0

  • Used pump B data for MDHP1 and pumps A and C data for MDHP2.
    • No information provided in tag order forms.

O C-56 0212G061886DAR

I i

TMI-1 COMP 0NENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: MDHP1, MDHP2 (Sheet 4 of 4) l i

SYSTEM: Makeup l

COMP 0NENT TYPE: Pumps A, B, and C*

Site-Specific Data  ;

Duration .

Tag Order Component Date (hours). Work Done. ,  ;

l 124 MU-P-1A 1/18/78 14.3 Repair, replace i nipple; repack valve  !

MU-V-74A. j 962 MU-P-1A 7/31/78 12.3 Repair vent line -

nipple. I I

1013 MU-V-148 8/14/78 4.0 Set iimits.  !

1684 MU-P-1C 2/17/78 33.0 '

l O)

% [

I i

l l

l e Maintenance Data Source: Switching and Tagging Orders

  • Used pump B data for MDHP1 and pumps A and C data for MDHP2.
    • No information provided in tag order forms.

O.

C-57 0212G0618860AR

TMI-1 COMP 0NENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: MP;1SIL, MDNSIS* (Sheet 1 of 8)

SYSTEM: Nuclear Services River Water COMPONENT TYPE: Pump Site-Specific Data Data include only maintenance performed during periods when the reactor was not in cold shutdown.

e Total noncold shutdown hours for reporting period: 9.54 x 104 Duration Tag Order Component Date (hours) Work Done 5009 NR-P-1C 9/17/74 4.0 Preventive maintenance M-1.

5657 NR-P-1C 12/23/74 7.0 5692 NR-P-1A 12/28/74 6.0 Screen wash pump 2A deenergized.

6316 NR-P-1A 3/21/75 0.5 Repair automatic vent.

6373 NR-P-1C 3/25/75 22.5 Inspect wiring on breaker NR-S-1C.

6374 NR-P-lC 3/25/75 22.5 Inspect wiring on breakers.

6430 NR-P-1A 3/31/75 7.5 f.epair and clean lines.

6438 NR-P-1C 4/1/75 4.0 Repair and clean lines.

6448 NR-P-1B 4/1/75 5.0 Repair and clean lines.

6814 NR-P-1A 5/12/75 9.0 Repack.

6839 NR-P-1A 5/15/75 8.0 Adjust and repack.

  • Maintenance duration less than 100 hours0.00116 days <br />0.0278 hours <br />1.653439e-4 weeks <br />3.805e-5 months <br />.
    • No information provided in tag order forms.

O C-58 0212G061886DAR

N i

TMI-1  !

COMPONENT MAINTENANCE DATA

SUMMARY

SHEET ,

DESIGNATOR: MONS 1L, MDNSIS (Sheet 2 of 8) ,+-

SYSTEM: Nuclear Services River Water

  • COMPONENT TYPE: Pump Site-Specific Data

(

Duration .

Tag Order Component -Date (hours) Work-Done 7214 NR-P-1A 6/6/75 392.5 Pull motor.-  ;

7363 NR-P-1A 6/27/75 462.75 Replace motor. f f

7442 NR-P-1C 7/15/75 2.5 Change oil and  :

lubricate motor. i 7452 NR-P-1B 7/15/75 4.0 Change oil. '

7574 NR-P-1B 8/11/75 8.7 Repair check valve. l 7580 NR-P-1B 8/12/75 1.5 Repair relay.

O >

V 7588 NR-P-1B 8/12/75 3.4 Inspect check valve I NR-V-208, l

8566 NR-P-1C 12/2/75 1.5 Megger.  ;

8751, 8744, NR-P-1A 1/8/76 312 Repair valve. i 8759, 8808, 8820 146 NR-P-1A 5/26/76 389 Lubricate motor bearings.

202 NR-P-1A 6/10/76 3.5 Check clearance on check valve hinge  ;

pins (NR-V-20A).

341 NR-P-1B 7/8/76 2.0 i I

493 NR-P-1C 8/24/76 8.50 Investigate control l power fuse failure.  ;

  • No information provided in tag order forms.

O  !

C-59 l 0212G061886DAR {

TMI-1 COMPONENT MAINTENANCE DATA

SUMMARY

SHEET ,

DESIGNATOR: MDNS1L, MDNSIS (Sheet 3 of 8) -

SYSTEM: Nuclear Services River Water COMPONENT TYPE: Pump Site-Specific Data Duration Tag Order Component Date (hours) Work Done 1062 NR-P-1A 10/29/76 3.5 Repack pump NR-P-2A.

1472 NR-P-1A 10/29/76 220 1548 NR-P-1C 1/14/77 27.45 Pump tagged "as i s."

2 NR-P-18 3/2/77 1.30 Repack pump.

19 NR-P-1A 3/5/77 5.0 Preventive maintenance, lubricate motors.

NR-P-1B Preventive maintenance, lubricate motors.

NR-P-1C Preventive maintenance, lubricate motors.

61A NR-P-1B 3/14/77 17.40 Repair strainer.

788, 789 NR-P-1C 6/29/77 64.50 Repair coupling.

799 NR-P-1C 7/5/77 386.3 Check pump shaft.

  • No information provided in tag order forms.

O C-60 0212G0518860AR

TMI-1 COMP 0NENT MAINTENANCE DATA S'JMARY SHEET DESIGNATOR: MDNS1L, MDNSIS (Sheet 4 of '8) ,

SYSTEM: Nuclear Services River Water COMP 0NENT TYPE: Pump Site-Specific Data Duration

-Tag Order Component- Date (hours) Work Done 851, 852 NR-P-1C 7/21/77 127.30 Megger motor, couple motor and pump.

855 NR-P-1C 7/22/77 138.50 Remove motor.

871 NR-P-1C 7/27/77 2.15 Check current.

875 NR-P-1C 7/28/77 3.0 Check current.

1291 NR-P-1B 12/1/77 16.15 Replace pump coupling.

94 NR-P-1C 1/9/78 105.0 Repack pump; replace strainer.

917 NR-P-1A 7/17/78 12.15 Replace joint.

938 NR-P-1A 7/24/78 5.45 General maintenance.. <

980 NR-P-1B 8/3/78 8.0 General maintenance.

982 NR-P-1C 8/4/78 16.0 General maintenance.

l 4

O C-61 )

0212G061886DAR

TMI-1 COMPONENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: MDNS1L, MDNSIS (Sheet 5 of 8)

SYSTEM: Nuclear Services River Water COMPONENT TYPE: Pump Site-Specific Data Duration Tag Order Component Date (hours) Work Done 1028 NR-P-1A 8/29/78 33.2 Repair, calibrate.

1056 NR-P-1A 8/27/78 60.0 Divers removing silt.

1058 NR-P-1A 8/30/78 12.3 Divers removing silt.

1061 NR-P-1A 8/31/78 15.0 Divers removing silt.

1063 NR-P-1A 9/1/78 7.3 Divers , removing silt.

1068 NR-P-1A 9/5/78 13.0 Divers removing silt.

1075 NR-P-1A 9/6/78 16.0 Divers removing silt. g 1079 NR-P-1A 9/7/78 17.2 Divers removing silt.

1082 NR-P-1A 9/8/78 15.4 Divers removing silt.

1100 NR-P-1A 9/12/78 14.0 Divers removing silt.

1104 NR-P-1A 9/13/78 5.3 Divers removing silt.

1110 NR-P-1B 9/13/78 7.2 Divers removing silt.

1113 NR-P-1B 9/14/78 14.0 Divers removing silt.

1130 NR-P-1B 9/18/78 17.2 Divers removing silt.

1137 NR-P-1B 9/19/78 14.35 Divers removing silt.

1144 NR-P-1B 9/20/78 17.3 Divers removing silt.

1150 NR-P-1B 9/21/78 6.0 Divers removing silt.

O C-62 0212G061886DAR

I TMI-1 COMPONENT MAINTENANCE DATA

SUMMARY

SHEET

(~')

Lj DESIGNATOR: MONS 1L, MDNSIS (Sheet 6 of 8)

SYSTEM: Nuclear Services River Water COMPONENT TYPE: Pump Site-Specific Data Duration Tag Order Component Date (hours) Work Done 1114 NR-P-1B 9/14/78 9.2 Install lube water modi fication .

1119 NR-P-1A 9/15/78 7.50 Install lube water modi ficati on.

1133 N.R-P-1; 9/18/78 9.4 Repai r val ve NR-V-208.

1171 NR-P-1C 9/25/78 11.0 Divers removing silt.

1178 NR-P-1C 9/26/78 14.15 Divers removing silt.

1197 NR-P-1C 9/28/78 13.0 Divers removing silt.

J 1214 NR-P-1C 10/2/78 15.3 Divers removing silt.

1223 NR-P-1C 10/4/78 15.4 Divers removing silt.

1231 NR-P-1C 10/5/78 13.0 Divers removing silt.

1241 NR-P-1C 10/9/78 17.45 Divers removing silt.

1248 NR-P-1C 10/10/78 16.45 Divers removing silt.

1172 NR-P-1C 9/25/78 10.0 Lube water modi fication .

1220 NR-P-1B 10/3/78 6.45 Rebuild check valve.

1251 NR-P-1B 10/11/78 15.2 Divers removing silt.

1256 NR-P-1B 10/12/78 11.3 Divers removing silt.

1268 NR-P-1B 10/16/78 20.15 Divers recoving silt.

QJ' C-63 0212G061886DAR

TMI-1 COMPONENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: MDNS1L, MDNSIS (Sheet 7 of 8)

SYSTEM: Nuclear Services River Water COMP 0NENT TYPE: Pump Site-Specific Data Duration Tag Order Component Date (hours) Work Done 1275 NR-P-1B 10/17/78 16.15 Divers removing silt.

1282 NR-P-1B 10/18/78 20.25 Divers removing silt.

1265 NR-P-1A 10/14/78 1.15 Repair valve NR-V-22A.

1285 NR-P-1B 10/19/78 15.30 Divers removing silt.

1293 NR-P-1B 10/23/78 15.3 Divers removing silt.

1296 NR-P-1B 10/24/78 15.15 Divers removing silt.

1297 NR-P-1B 10/25/78 14.0 Divers removing silt.

1305 NR-P-1B 10/26/78 16.0 Divers removing silt.

1313 NR-P-1B 10/30/78 14.3 Divers removing silt.

1316 NR-P-1B 10/31/78 14.3 Divers removing silt.

1320 NR-P-1A 11/1/78 14.3 Divers removing silt.

1324 NR-P-1A 11/2/78 14.5 Divers removing silt.

1327 NR-P-1A 11/6/78 11.45 Divers removing silt.

1333 NR-P-1A 11/7/78 15.0 Divers removing silt.

1339 NR-P-1A 11/9/78 8.15 Divers removing silt.

1347 NR-P-1B 11/9/78 3.4 Divers removing silt.

1422 NR-F-1C 12/5/78 23.0

  • No information provided in tag order forms.

O C-64 0212G061886DAR

l TMI  ;

COMPONENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: MDNS1L, MDNSIS .(Sheet 8 of 8) i SYSTEM: Nuclear Services River Water '

COMP 0NENT TYPE: Pump

.l Sito-Specific Data  ;

.I Duration Tag Order Component Date (hours) Work Done (;

  • l 1429 NR-P-1B 12/7/78 5.1 I
  • j 1674 NR-P-18 2/15/79 24.0 ,

(estimated)

O i

l o Maintenance Data Source: Switching and Tagging Orders j i

l l

l l

l l

i

  • No information provided in tag order forms. I O

1 C-65 0212G061886DAR l

)

TMI-1 COMPONENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: MDNS3 (Sheet 1 of 2)

SYSTEM: Nuclear Services Closed Cycle COMP 0NENT TYPE: Pump Site-Specific Data Data include only maintenance performed during periods when the reactor was not in cold shutdown.

e Total noncold shutdown hours for reporting period: 9.54 x 104 Duration Tag Order Component Date (hours) Work Done 5054 NS-P-1A 9/25/74 0.6 Change oil on bearings.

NS-P-18 0.6 Change oil on bearings.

NS-P-1C 0.6 Change oil on bearings.

577 NS-P-1A 1/10/75 10.0 5782 NS-P-1A 1/10/75 187 Repack pump. .

6842 NS-P-1C 5/15/75 0.67 Change oil.

6843 NS-P-1A 5/15/75 0.4 Change oil.

6844 NS-P-1B 5/15/75 0.5 Change oil.

7304 NS-P-1C 6/19/75 28.8 Test thermal overload relay 49X.

7804 NS-P-1B 9/4/75 6.5 Megger motor for preventive maintenance.

8542 NS-P-1A 12/1/75 3.2 Megger.

8557 NS-P-1C 12/2/75 1.0 Megger.

8578 NS-P-1B 12/3/75 0.5 Megger.

343 NS-P-1B 7/9/76 3.2 Preventive maintenance. l l

  • No information provided in tag order forms. '

I

~

C-66 0212G051886DAR

- .- . . _ _ . . - . - . . . . . ._ - - - - . . . . . . . = _ - - . . . .

TMI-1 COMP 0NENT MAINTENANCE-DATA

SUMMARY

SHEET DESIGNATOR: MDNS3 (Sheet 2 of 2)

SYSTEM: Nuclear Services Closed Cycle COMP 0llENT TYPE: Pump Site-specific Data Duration' Tag Order Component Date (hours) Work Dane 1536 NS-P-1B 1/7/77 5.2 Megger control-circuit.

.l 1726 NS-P-1B 2/20/77 1.0 Oil. change.

1730A NS-P-1B 2/23/77 1.2 Oil change.  !

l 48 NS-P-1B 3/11/77 3.0 Bearing.

49 NS-P-1A 3/12/77 4.3 Bearing. I 142 NS-P-1A 1/23/78 10.0 Repack.

1689 NS-P-1B 2/13/77 1.15 Change oil, 1307 (1S-P-1B 10/26/78 16.10 Repack pump.

1309 NS-P-1A 10/27/78 3.35 Change oil..

e Maintenance Data Source: Switching and Tagging Orders O  !

C-67 0212G061886DAR

. - - . . . _ _ _ - . _ . . - . . _ _ . . - . ._.___.._._._,-.,_.~.._..,,._,__a.,.. . _ _ _

TMI-1 COMPONENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: MONS 4 SYSTEM: Air-Handling COMP 0NENT TYPE: Auxiliary Building Ventilation Faris Site-Specific Data Data include only maintenance performed during periods when the reactor was not in cold shutdown, e Total noncold shutdown hours for reporting period: 1.27 x 105 Duration Tag Order Component Date (hours) Work Done 5659 AH-E-15A 12/23/74 11.3 5690 AH-E-15A 12/28/74 1.5 AH-E-15B 5856 AH-E-158 1/22/75 7.1 6250 AH-E-15A 3/17/75 0.5 Inspect breaker and strainer.

g 6383 AH-E-15B 3/26/75 0.75 Inspect breaker.

6778 AH-E-158 5/7/75 26.5 Check noise.

1568 AH-E-158 1/18/77 2.0 General overhaul.

1637 AH-E-15B 2/2/77 10.45 Check vibration.

1073 AH-E-15A 9/5/78 29,30 Change CS12, 1070 AH-E-15B 9/6/78 9.30 Change CS12.

e Maintenance Data Source: Switching and Tagging Orders

  • No information provided in tag order forms.

O C-68 0212G061886DAR

TMI-1 s COMPONENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: MDNS4 SYSTEM: Emergency Feedwater Ventilation COMPONENT TYPE: Ventilation Fans Site-Specific Data Data include only maintenance performed during periods when the reactor was not in cold shutdown.

e Total noncold shutdown hours for reporting period: 6.36 x 104 Duration Tag Order Component Date (hours) Work Done 6421 AH-E-24A 3/27/75 3 Inspect, tighten, repair.

879 AH-E-24A 9/15/76 6 Lubrica.te motors.

AH-E-248 O

e Maintenance Data Source: Switching and Tagging Orders I l

O  !

l C-69  !

0212G061886DAR  !

h TMI-1 COMPONENT MAINTENANCE DATA

SUMMARY

SHEET DESIGNATOR: MDSE1 SYSTEM: Intermediate Cooling Closed Loop COMPONENT TYPE: Pump Site-Specific Data Data include only maintenance performed during periods when the reactor was not in cold shutdown.

e Total noncold shutdown hours for reporting period: 6.36 x 104 Duration Tag Order Component Date [ hours) Work Done 4917 IC-P-1B 9/4/74 3.5 Test molded case breaker.

5985 Filter FIA 2/7/75 2.0 Valves V-14 and V-15 closed; valves V-53 and V-54 open.

5989 Filter FIA 2/10/75 2.0 Valves V-14, V-15, V-53, and V-54 closed; change gasket.

6389 IC-P-2 3/26/76 22.25 Inspect breakers.

678 RM-L9 5/21/77 6.50 Close IC-V-44, V-45, V-65, and V-75.

e Maintenance Data Source: Switching and Tagging Orders O

C-70 0212G0618860AR

I i

l l

O  !

\

1 i

I i

1 i

l l

i 1

l i

APPENDIX 0 TMI-1 INITIATING EVENTS DATA

SUMMARY

SHEETS l

O 0237G1229gSogg

_ _ . . . . _ _ _ . . _ _ . _ . . . - _ . .. ~ , _ . . _ _ _ . . _ . . - _ _ _ _ . . _ . _ . _ . ,__ ._ _ __ _ _-

APPENDIX D

TMI-1 INITIATING EVENT DATA

SUMMARY

SHEETS The data sheets in_ this appendix sumarize the initiating events that have occurred at TMI-l during the period covered by.the data collection effort.

The data sheets ar-e organized by initiating event categories as ' listed in Table 3-8, O

i i

1 O '

D-1 0237G122986DAR

TMI-1 INITIATING EVENTS DATA

SUMMARY

SHEET DESIGNATOR: LL EVENT CATEGORY: Large LOCA NUMBER OF EVENTS: 0 REACTOR YEARS: 4.5 Event Summag Date Event Description None.

O Data Sources: Monthly Operating Reports Weekly Reports Gray Book 9

D-2 0237G122986DAR

TMI-1 f1

%./

INITIATING EVENTS DATA

SUMMARY

SHEET DESIGNATOR: ML EVENT CATEGORY: Medium LOCA NUMBER OF EVENTS: 0 REACTOR YEARS: 4.5 Event Summary Date Event Description None.

O i

5 i

l Data Sources: Monthly Operating Reports Weekly Reports Gray Book O

D-3 0237G1229860AR l

TMI-1 INITIATING EVENTS DATA

SUMMARY

SHEET DESIGNATOR: SB EVENT CATEGORY: Small LOCA NUMBER OF EVENTS: 0 ,

REACTOR YEARS: 4.5 Event Summary Date Event Description None.

O' l

1 l

1 I

l l

Data Sources: Monthly Operating Reports Weekly Reports Gray Book 1

9i D-4 0237G1229860AR 1

TMI-1 ,

INITIATING EVENTS DATA

SUMMARY

SliEET DESIGNATOR: VSB EVENT CATEGORY: Very Small LOCA NUMBER OF EVENTS: 0 REACTOR YEARS: 4.5 Event Sumary Date Event Description f None.  ;

t t

t Data Sources: Monthly Operating Reports l I

Weekly Reports Gray Book j l

l l

0  ;

l 0237G122986DAR

TMI-1 INITIATING EVENTS DATA

SUMMARY

SHEET DESIGNATOR: VS -

EVENT CATEGORY: Inadvertent Opening of DHR V61ves NUMBER OF EVENTS: 0 REACTOR YEARS: 4.5

' Event Summary Date Event Description None.

O Data Sources: Monthly Operating Reports Weekly Reports '

Gray Book O

D-6 0237G122986DAR

n TMI-1 INITIATING EVENTS DATA

SUMMARY

5HEET

]

DESIGNATOR: SLI EVENT CATEGORY: Steam Line Break in Intermediate Building NUMBER OF EVENTS: 0 REACTOR YEARS: 4.5 Event Summary ,

Date Event Description .

None.

O i

Data Sources: Monthly Operating Reports I Weekly Reports  :

Gray Book l

D-7 0237G122986DAR 1

TMI-1 INITIATING EVENTS DATA

SUMMARY

SHEET DESIGNATOR: SLT EVENT CATEGORY: Steam line Break in Turbine Building NUMBER OF EVENTS: 0 REACTOR YEARS: 4.5 Event Summary Date Event Description None.

O Data Sources: Monthly Operating Reports Weekly Reports Gray Book O

0-8 0237G122986DAR

l TMI-1 INITIATING EVENTS DATA

SUMMARY

SHEET

%.)t DESIGNATOR: TR l

EVENT CATEGORY: Steam Generator Tube Rupture i NUMBER OF EVENTS: 0 l

REACTOR YEARS: 4.5 1 1

Event Summary l

Date Event Description i I

None.

l l

l l

O 1

i l

I 1

Data Sources: Monthly Operating Reports 1 Weekly Reports '

Gray Book 1

O l

1 0237G122986DAR

TMI-1 INITIATING EVENTS DATA

SUMMARY

SHEET DESIGNATOR: EXC EVENT CATEGORY: Excessive Fee nater Flow NUMBER OF EVENTS: 0 REACTOR YEARS: 4,5 Event Summary Date Event Description None.

O Data Sources: Monthly Operating Reports Weekly Reports Gray Book D-10 0237G122986DAR l

TMI-1 INITIATING EVENTS DATA

SUMMARY

SHEET DESIGNATOR: FW EVENT CATEGORY: Total Loss of Main Feedwater NUMBER OF EVENTS: 0 REACTOR YEARS: 4,5 Event Sunnary Date Event Description None. l l

l I

1 Data Sources- Monthly Operating Reports Weekly Reports Gray Book O

D-11 0237G122986DAR

TMI-1 INITIATING EVENTS DATA

SUMMARY

SHEET DESIGNATOR: RT EVENT CATEGORY: Reactor Trip NUMBER OF EVENTS: 3 REACTOR YEARS: 4.5 Event Summary Date Event Description 6/25/75 Reactor tripped because control rod swap group 7 phase bus bar faulted to neutral, causing group 7 to drop into the cone. (This was followed by a turbine trip due to water in high pressure turbine - see turbine trip event surraary sheet).

S/27/76 Reactor trip on high neutron flux due to operator switching error in nuclear instrumentation (trying to shift the neutron flux input jacks to the ICS with a large neutron error signal present).

11/14/77 ICS component malfunctioned, causing reactor trip. h Data Soure,es: Monthly Operating Reports Weekly Reports Gray Book 9

D-12 0237G122986DAR

TMI-1 ,

INITIATING EVENTS DAT.i

SUMMARY

SHEET DESIGNATOR: TT EVENT CATEGORY: Turbine Trip NUMBER OF EVENTS: 7 REACTOR YEARS: 4.5 Event Summary Date Event Description 1/23/75 High moisture separator drain tank level caused turbine trip. Reactor did not trip (reduced to 10-8 amps).

3/30/75 A faulty relay gave an erroneous signal that indicated loss of DC power to the turbine EHC system. The signal tripped the turbine. The reactor tripped from high reactor coolant pressure.

5/9/75 Turbine tripped due to a mechanical failure in the "B" .

moisture separator high-level trip device, which allowed O moisture into the switch, causing the switch to short.

6/18/75 A brush recorder monitoring the turbine EHC system caused erroneous voltage spikes resulting in rapid load reduction and reactor trip on high pressure.

6/25/75 Turbine tripped offline due to water in high pressure turbine.

12/21/75 Turbine trip due to actuation of deluge system. The switch was actuated to test the flood valve for the main turbine generator deluge system.

11/15/78 Momentary loss of DC power to the EHC control system caused a turbine trip. The reactor was kept about 15 to 20% at power.

Data Sources: Monthly Operating Reports Weekly Reports Gray Book

O D-13 0237G122986DAR

i l

TMI-1 INITIATING EVENTS DATA

SUMMARY

SHEET DESIGNATOR: LA EVENT CATEGORY: Loss of Air System NUMBER OF EVENTS:

REACTOR YEARS: 4.5 Event Summary Date Event Description None.

O Data Sources: Monthly Operating Reports Weekly Reports Gray Book 9

D-14 0237G1229860AR

TMI-1 g INITIATING EVENTS DATA

SUMMARY

SHEET DESIGNATOR: LC EVENT CATEGORY: Loss of Control Building Ventilation NUMBER OF EVENTS: 0 REACTOR YEARS: 4.5 Event Summary Date Event Description None.

O Data Sources: Monthly Operating Reports Weekly Reports Gray Book O(.>

D-15 0237G122986DAR

TMI-1 INITIATING EVENTS DATA

SUMMARY

SHEET DESIGNATOR: ATA EVENT CATEGORY: Loss of ATA Power NUMBER OF EVENTS: 0 REACT 0P YEARS: 4.5 Event Sumary Date Event Description None.

O Data Sources: Monthly Operating Reports Weekly Reports Gray Book O

D-16 0237G122986DAR

TMI-1 INITIATING EVENTS DATA

SUMMARY

SHEET DESIGNATOR: LD EVENT CATEGORY: Loss of DC Power Train A NUMBER OF EVENTS: 0 REACTOR YEARS: 4.5 Event Summary to Date Event Description None, i

O Data Sources: Monthly Operating Reports Weekly Reports Gray Book O

D-17 0237G122986DAR

THI-1 INITIATING EVENTS DATA

SUMMARY

SHEET DESIGNATOR: AC EVENT CATEGORY: Total Loss of Offsite Power NUMBER OF EVENTS: 0 REACTOR YEARS: 4.5 Event Summary Date Event Description None.

O.

Data Sources: Monthly Operating Reports Werkly Reports Gray Book O ,

D-18 0237G122986DAR

TMI-1  :

INITIATING EVENTS DATA

SUMMARY

SHEET DESIGNATOR: LNS EVENT CATEG,0RY: 1.oss of Nuclear Services Closed Cooling Water NUMBER OF EVENTS: 0 REACTOR YEARS: 4.5 Event Summary Date Event Description  !

l None.

l 3

I i

I i

l Data Sources: Monthly Operating Reports Weekly Reports Gray Book

)

O D-19  !

0237G122986DAR

TMI-1 INITIATING EVENTS DATA

SUMMARY

SHEET DESIGNATOR: LR EVENT CATEGORY: Loss of River Water NUMBER OF EVENTS: 0 REACTOR YEARS: 12 Event Summary Date Event Description See discussion on plugging of intake screens in Section 3.5.

I i

Data Sources: Monthly Operating Reports Weekly Reports Gray Book 1

D-20 '

l 0237G122986DAR