ML20040A474

Midland Plant Auxiliary Feedwater System Reliability Analysis Evaluation

ML20040A474
Person / Time
Site:	Midland
Issue date:	12/31/1981
From:	Roscoe B SANDIA NATIONAL LABORATORIES
To:	Office of Nuclear Reactor Regulation
References
CON-FIN-A-1303 NUREG-CR-2368, SAND81-2164, NUDOCS 8201210157
Download: ML20040A474 (37)

Text

.-

NUREGER-2368 SAND 81-2164

=.

Midland Plant i Auxiliary Feedwater System Reliability Analysis Evaluation Prepared by B. J. Roscoe Sandia National Laboratories Prepared for

~^

U.S. Nuclear Regulatory O'

4' Commission g

fiE C E4' E D dl\\N2 01982n :

y " w g e P

k /i gg fool S oIf 1

820 1210 15 1 A

l

\\

l i

I l

i NOTICE i

j j

This report was prepared as an account of work sponsored by

]

an agency of the United States Government, Neither the j

United States Government nor any agency thereof, or any of -

l their employees, makes any warranty, expressed or implied, or assumes any legal liability or responsibility for any third party's use, or the results of such use, of any information, apparatus product or process disclosed in this report, or represents that i

its use by such third party would not infringe privately owned j

rights.

1 l

I I

l l

i J

t i

t J

l Available from i

GP0 Sales Program Division of Technical Information and Document Control U. S. Nuclear Regulatory Commission Washington, D. C. 20555 Printed copy price: $2.75 and l

l National Technical Information Service Springfield, Virginia 22161 i

r

)

.-,---x-----,

NUREGER-2368 SAND 81-2164 Midland Plant Auxiliary Feedwater System Reliability Ana ysis Evaluation Manuscript Completed: August 1981 Data Published: December 1981 Prepared by i

B. J. Roscoe Sandia National Laboratories Albuquerque, NM 87185 Prepa.'od for Division of Safoty Technology Offico of Nuclear Reactor Regulation U.S. Nuclear Regulatory Ccrnmission W ::hin0 ton, D.C. 20555 NRC FIN A1303 o

v

\\, 3 s

j

,k ~

.c

~,

' i '~'

~

+

x

/

t t

Availabit;ty of Reference Matenals Cited in NRC Publications I Most documents cited in NRC publications will be availabie from one of the following sources:

1.

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

Washington, DC 20555

2. ' The NRC/GPO Sales Program, U.S. Nuclear Regulatory Commission, Washington, DC 20555 1; he National Technical Information Service Springfield, VA 22161 Aliiough the listing that follows represents the majority of documents cited in NRC publications, it is not intended to b? exhaustive.

Refeconced (ocuments available for inspection and copying for a fee from the NRC Public Document Room incude NRC correspondence and internal NRC memoanaa; NRC Office of Inspection and Enforce-ment bulletins, circuars, information notices, inspection and investigation notices; Licensee Event Reports; vendor reports and correspondence; Commission papers; and applicant and licensee documents and correspondence.

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

Documents available from the National Technical Information Service include NUREG series reports and technical reprxts prepared by other federal agencies and reports prepared by the Atomic Energy Commis-sion, forerunne Aency to the Nuclear Regulatory Commission.

Documents available from public and special technicallibrar es include all open literature items, such as books, journal and periodical articles, transactions, and codes and standacds. Federal Register notices, federal and state legislat:en, and congressional reports can usually be obtained from these libraries.

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

Single copies of NRC draf t reports are available free upon written request to the Division of TechnicalInfor-mation ano Document Contr(.l. tJ.S. Nuclear Regulatory Commission, Washington, DC 20555.

e

)

1

- f

ABSTRACT The purpose of this report is to present the results of the review of the Auxiliary Feedwater System Reliability Analysis for the Midland Plant, Units 1 and 2.

i

i d

Acknowledgement i

The author sincerely appreciates the comments on the drafts provided by Jack W.

Hickmar of Sandia National Laboratories.

l This report has extracted freely from the referenced j

documents.

a i

4 1

l i

l l

ii

- ~.

Table of Contents Page l

Abstract i

i j

Acknowledgement 11 List of Figures v

Summary and Conclusions 1

i 1.

Introduction 2

1.1 Scope and Level of Effort 2

1.2 Specific Review 3

2.

AFWS Configuration 3

2.1 System Description

6 j

4 2.2 AFWS Support 8

2.2.1 Power Sources 8

1 l

2.2.2 Alternate Water Sources 10 i

2.2.3 Steam Availability 10 2.2.4 Instrumentation and Controls 11 2.2.5 Initiatio:s Signals for Automatic 12 l

Operation 2.2.6 Testing 13 i

2.2.7 Technical Specifications 13 3.

Discussion 15 3.1 Mode of AFWS Initiation 15 l

l 3.2 System Control Following Initiation 16 1

3.3 Test and Maintenance Procedures and 17 Unavailability 3.4 Adequacy of Emergency Procedures 19 3.5 Adequacy of Power Sources and Separation 19 of Power Sources i

iii

s..

i a

Page 3.6 Availability of Alternate Water Sources 20 3.7 Potential Common Mode Failure 21 3.8 Application of Data Presented in NUREG-0611 22 3.9 Search for Single Failure Point 22 3.10 Human Factors / Errors 23 4

3.11 NUREG-0611 Recommendations Long and Short-Term 23 l

l 3.11.1 Short-Term Generic Recommendations 3.11.2 Additional Short-Term Recommendations 3.11.3 Long-Term Generic Recommendations i

4.

Major contributors to Unreliability 24 5.

Conclusions 26 6.

Glossary of Terms 27

}

7.

References 28 l

i i

i iv

List of Figures Page 1.

Auxiliary Feedwater System Simplified Piping 5

and Instrumentation Diagrams 2.

Comparison of Reliability of AFWS Designs in 25 Plants Using the Babcock and Wilcox NSSS Figure 2(A):

LMEM Figure 2(B):

LMFW/ LOOP Figure 2(C):

LMFW/LOAC v

l 1

Summary and Conclusions j

i The accident at Three Mile Island, Unit 2 (TMI-2) resulted j

in many studies which outlined the events leading to the accident l

as well as those following.

In response to this event, a task group was formed by the Nuclear Regulatory Commission (NRC) to provide an assessment of the generic aspects of the feedwater transient and the related ensuing events at TMI-2 to determine bases for continued safe operation of other reactor plants similar to TMI-2 that were designed by the Babc9ck & Wilcox Company B&W).

l The results were reported in NUREG-0560ll) and NUREG-0565. t 2 i

One of the important safety systems involved in the mitigation i

of such accidents was determined to be the Auxiliary Feedwater System ( AFWS The licensee of each nonoperating plant was instructed (4))

to perform a reliability analysis of his AFWS for three transient conditions involvir-g loss of ma4n feedwater in a manner similar to the study made by NUREG-0611(3) prior to 1

their obtaining an operating license.

Consumers Power Company, I

the licensee fgr Midland Plant Units 1 and 2, submitted a reli-i ability report (5) prepared by Pickard, Lowe, and Garrick, i

Inc., to the NRC in July 1980.

This report was reviewed by i

Sandia National Laboratories (SNL).

The following conclusions resulted from the review:

i

)

1.

Consumers Power Company has satisfactorily complied with I

the requirement to make a reliability study of their AFWS.

)

l 2.

The AFWS of the Midland Plant has medium reliability for j

the first case event, Loss of Main Feedwater, relative to the reliability of operating plants.

Quantitatively, the unavail~

j ability of the system is 1.2 x 10-4 per demand.

Qualitatively,

{

the system is automatically initiated, moderately redundant, has no observed single point vulnerabilities, and is tested periodically to demonstrate flow rate at rated pressure.

Failure on demand is dominated by maintenance and failure of the turbine and pump motor.

The unavailabilitv for the Loss of Main j

Feedwate r and Loss of'Offsite Power,second case event,4 is 5.5 x 10-per demand, which places the reliability in the medium range.

Failure upon i

demand is dominated by maintenance of the turbine pump, failure 1

of the diesel generators to start, and failure of the turbine to I

start.

1 The unavailability for the third case event Loss of Main Feedwater and Loss of All AC Power, is 1.3 x 10-b per demand, I

which places this system in the medium reliability group.

The l

TDP train has no identifiable ac power dependencies and is auto-matica11y actuated.

The dominant failures are maintenance of j

TDP train, failure of the turbine to start and failure of the f

turbine pump.

(

1.

Introduction The results of many studies pertaining to the Three Mile Island Nuclear Power Plant accident conclude that a proper func-tioning Auxiliary Feedwater System is of prime importance in the mitigatign of such accidents.

Therefore, a letter dated April 24, 198054) stating NRC's requirements regarding the AFWS was sent to all operating license applicants with a Nuclear Steam Supply System (NSSS) designed by Babcock and Wilcox (B&W).

The Consumers Power Company (CPCo) of Jackson, Michigan, the applicant for an operating license for the Midland Plant

'Jnits 1 and 2 which has a B&W designed ySSS, provided a response in the form of a reliability analysis (Si which was prepared for them by Pickard, Lowe, and Garrick, Inc. (PL&G).

The analysis was received by SNL for review on April 22, 1981.

The analysis makes a study of the failure of the AFWS to supply sufficient flow to both steam generators (SG).

The method of analysis consists of the construction and i

evaluation of fault trees.

It takes into account actual compo-nent failures, single failure passive components, outage due to test and maintenance, human errors, and common cause failures.

1.1 Scope and Level of Effort This project undertakes a review of those portions of the reliability analysis which (1) satisfy requirement (b) of the letter which states, " perform a reliability evaluation similar in method to that described in Enclosure 1 (NUREG-0611) that was performed for operating plants and submit it for staff review,"

and (2) provide answers to the short and long-term recommendations of NUREG-0611 in response to requirement (c) in hg letter.

The review was conducted according to a Schedule 189 6; which was submitted by SNL to NRC.

Sandia National Laboratories' review addressed the following 2

issues:

i (1)

Mode of AFWS initiation.

(2)

System control following initiation.

(3)

Test and maintenance procedure and unavailability of AFWS.

(4)

Potential common mode factors in the AFWS.

I (5)

Adequacy of emergency procedures for the operation and initiation. -. --

(6)

Adequacy of power sources and separation of power sourcec.

(7)

Availability of alternate water sources.

(8)

Adherence to methodology and data presented in NUREG-0611.

(9)

Search for single failure points.

l.2 Specific Review SNL reviewed the reliability analysis (5) submitted by CPCo.

Particular attention was directed toward determining that the analysis addressed in depth the reliability of the AFWS when subjected to three transient cases (1) Loss of Main Feedwater, LMFW (2) Loss of Main Feedwater/ Loss of Offsite Power, LMFW/ LOOP, and (3) Loss of Main Feedwater/ Loss of All AC Power, LMFW/LOAC.

Also the methods used in NUREG-0611 were compared to those used in the analysis.

This review applies to the AFWSs of Units 1 and of 2 of the Midland Plant.

The applicant has stated that Unit 2 is a replica of Unit 1 and there are no differences in the two AFWSs.

The review began with a meeting among representatives of CPCo, PL&G, B&W, Bechtel, NRC, and SNL on April 30, 1981, in Ann Arbor, Michigan.

The meeting consisted of presentations of the Midland plant AFWS design review by CPCo and Bechtel and of the Midland plant AFWS reliability analysis by PL&G.

Later questions were posed and discussed during telephone conference calls.

2.

AFWS Configuration The AFWS supplies feedwater o the steam generators during normal plant startup, shutdown, and hot standby operations when the main feedwater system is unavailable for service.

The AFWS is also designed to respond automatically to emergency conditions, to supply feedwater to the steam generators in order to remove reactor decay heat, assint in establishing natural circulation, and to cool down the reactor coolant system to the point at which the plant decay heat removal system may be placed into operation.

There are two safety-grade AFWS pumps, one motor-driven pump (MDP) and one turbine-driven pump (TDP), for each of the two units.

Each pump is a horizontal centrifugal unit rated at 885 gpm and 2,700 feet total developed head.

The discharge head is sufficient to establish the necessary flowrate against a steam generator pressure corresponding to the. lowest pressure satpoint of the main stream i

4. _..

safety valves.

The flowrate of each AFWS pump is equal to, or greater than, the flowrate required to remove the decay heat generated at 40 seconds into the transient.

The 40-second time was chosen to allow the AFWS to inject feedwater and begin increasing SG level to the 50% operating range level, required for natural circulation, prior to completing reactor coolant pump coastdown.

Figure 1 shows the simplified piping and flow diagram of the system.

The normal water source of the AFWS is the non-Seismic Category I, 300,000-gallon condensate storage tank (CST).

The CST is sized to accommodate the plant at hot shutdown for approxi-mately 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> followed by a 6-hour cooldown to 280F.

Alternate water sources for the AFWS are the deaerator storage tanks and the condenser hotwell.

Water from the deaerator storage tanks is normally used during hot standby or normal plant cooldown to minimize thermal shock to the steam generators.

Water from the condenser hotwell is considered to be a backup source to be used if water from the deaerators and the CST is unavailable.

A Seismic Category I supply to the AFWS pump suction is provided by the service water system (SWS) to supply feedwater in the event that the CST or other sources of water are not available.

The motor-driven AFWS pump associated with each unit is supplied with power from the Class lE ac power system.

Following an initiation of an auxiliary feedwater actuation signal (AFWAS),

the motor-driven AFW pump is capable of supplying feedwater to the steam generators within 40 seconds, including an allowance of 10 seconds for starting the emergency diesel generators.

The turbine-driven AFWS pump associated with each unit provides system redundancy of supply and diversity of motive pumping power.

Steam supply piping to the turbine driver is taken from each of the main steam lines inside the containment.

The AFW pumps are located in separate flood-protected rooms at elevation 584'-0" of the auxiliary building.

Each AFWS pump room is provided with an engineered safety feature (ESP) unit cooler to control room temperature at a level consistent with-environmental requirements for proper operation of the AFWS system components.

The ESF coolers begin operation in conjunc-tion with the pump they cool, and stop when the corresponding pump stops.

Room temperature is reduced below the room thermostat control setpoint.

The fan of each unit cooler is powered from k

the same train as the pump with which it is associated.

When the pump served by the unit cooler is off, the unit cooler fan is controlled by the pump room thermostat.

Structures, systems, and components required for APWS performance are designed to meet Seismic Category I requirements and to withstand the effects of other credible natural phenomena such as tornados and floods.

The natural phenomena and their _

I "T

(

)

(

d{

6 9

A y

j!

]

I h

i!

h

!I' U

^

Bl,i "ii i

c 1

10 I;

,.b!

.T jr

,r m

i o

18 g

8 l

Om

= w i

m i

=

8

@? T i

+,

,j; o

ig ?

,r 2:

3

@4I.

'O

@y 4

3:.

I*

I-UAI-i i

O U

h

/ !

a i

W i

. i myc 3

1 r m - c m lI,:s iInj^nl t!

I twa cuc s:=

,a w

i il Q

d, o

x x

D.

i.

ib.

i a

rir

!a ll TI a

."a I

i d

X i

i Od i

f l

--1 c

y e

1

!s is u

se gw

.g y

! %. s

'l

1. M! J%,.

d

~

l I,

+'! % %

,i I,

s o

e

o is_

2

+

i il i

i i

,t t

e..

s.,

i 1

1 l

[

f 1

i i

i i

i i

l

-s-

magnitudes are selected in accordance with their probability of occurrence at the Midland site.

2.1 System Description

During startup, the motor-driven AFW pump is used to supply feedwater from the deacrating storage tank to the steam generators.

The AFWS is not activated during normal power genera-tion.

The pumps are placed in the standby mode and are lined up to take suction from the CST if this becomes necessary.

During hot standby, the AFWS is used to provide water to each steam generator to maintain the water level.

Auxiliary J

feedwater pump suction may be taken from the deaerator storage tanks, which maintain the temperature at approximately 229;Fi Feedwa te r flow is pumped into the main feedwater nozzles of the steam generator via the auxiliary-to-main feedwater system.

i Crosstle.

During cooldown, the motor-driven AFWS pump is used. to supply water to the steam generators from the deaerator storage tanks, CST, or the condenser hotwell.

The deaerator stol rage tanks would be the primary source of this water to minimize thermal shock to the steam generators.

Steam generated during normal cooldown is bypassed to the main condenser.

The AFWS pump,.

may be used until the reactor coolant tempetature drops to approxi.

mately 280F, at which point the decay heat rer. oval (DHR) system is activated.

After the DHR system is placed in operation, the SGs are placed in a wet layup condition by using the AFWS.

During wet layup, all required AFW components will be manually controlled to accomplish SG filling.

The events following a postulated break in AFWS piping depend upon the plant conditions at the time of break.

The technical specifications will not permit using the turbine-driven AFMS pump during hot standby except in emergencies.

In the event of a postulated failure in the piping associated with the electric-driven pump, the break is isolated and the turbine-driven pump is started.

Because the turbine-generator is not paralleled to the offsite grid during hot standby, availability of offsite power is assumed.

This permits use of the main feedwater pumps in the event that the turbine-driven pump fails to start.

The l

nafety-grade Auxiliary Feedwater Actuation Signal (AFWAS) auto-l matically starts both the turbine-driven and the motor-driven AFWS pumps.

APWAS also automatically positions the AFWS valves both to mitigate the consequences of a loss of main feedwater or l

loss of offsite power incident and to provide feedwater to allow primary heat removal through the steam generators.

e 2.

i.

4 Under emergency conditions, heat is removed from the reactor i

cooling system- (RCS) by boiling the feedwater in the steam gener-j ators.and venting the steam to the atmosphere through the power-

'h operated atmospheric vent valves and/or the main steam safety valves.

If the main steam isolation valves are open, steam may be' relieved via the turbine bypass system if a condenser is j

available, or through the modulating atmospheric dump valves, if the' condenser is unavailable.

Either method is capable of lowering j

.the RCS temperature to a point where the Decay Heat Removal (DHRS)

{

.can be placed in operation.

l

'The.AFWS pump discharge headers are provided with a double l

crossover piping arrangement for system redundancy.

Each discharge j

header splits into two lines:

one line for the lead-level control j

,valvs of the associated steam generator and another line for l'

the crossover redundant-level control valve of the other steam

!8 genera)or.

The level control valve in the crossover piping j.

normally remains closed, as long as the lead valve is functioning jf properly.

If either the AFWS pump or the lead-level control U,

va,1ve of one train fails to supply the necessary feedwater to

)

its associated steam generator, the pump of the other train L

would then supply feedwater via the crossover piping.

I j

Parallel containment isolation valves are provided on the j

discharge, piping to each steam generator.

One of the parallel valves is ac powered and the other is de powered.

l-j The AFWSpump discharge headers are also provided with minimum i

recirculation and test lines.

The discharge flowpath is to the condensate storage tank or the cooling pond, depending on the suction source.

When AFWS suction is taken from the deaerators, 1

minimum pump recirculation flow is satisfied by recirculation to j

the deaerator storage tanks through the auxiliary-to-main feedwater i

system crosstie.

The AFWS suction piping is arranged to enable the motor-driven pump to operate independently of the turbine-j driven pump.

Normal alignment of the suction is from the non-j Seismic Category I CST when the AFWS is in standby.

Suction can be aligned either to the deaerators or the condenser hotwell by J

opening or closing remote manual valves operated from the main control room (MCR).

Each pump train connects to the SWS through two motor-operated, automatically actuated butterfly valves in series.

l Each train of the AFWAS can be manually initiated from the j

control room and results in the same system response as automatic 1.

initiation.

The number of AFWS components common to both manual and i

automatic initiation has been minimized to the extent practicable.

}

No single failure in either the manual or automatic controls j

will preclude operation of the system.

The AFWS system is equipped i

with a feed-only-the-good-generator (FOGG) control system which operates to terminate AFWS flow to a faulted steam generator.

I i

l !

_. ~ _.

Possible failure of the FOGG system is modeled in the fault tree, but failure of FOGG does not appear in any set of dominant fail-ures.

The FOGG system is powered by the 120 V ac system which is fed from battery driven inverters, i

Cooling for the turbine-driven AFW pump bearings and the turbine lubricating oil is provided by internal recirculation of the pumped fluid through the pump seal coolers and the turbine primary lube oil cooler.

This system is designed to provide sufficient cooling with pumpage temperatures at or below 130F, and satisfies cooling requirements when suction is taken from either the CST or SWS.

Though not intended for normal use, but provided to allow further operating flexibility, a secondary cooler using service water is used when suction is desired from the deaerators.

Valves and controls necessary for the function of the turbine-driven pump and its associated equipment are energized by class lE dc power.

l The AFWS system incorporates the following design features to minimize the effects of hydraulic instability (water hammer) :

a.

AR4S piping rises vertically to the SG AR4S nozzle to prevent drainage of the lines into the SGs.

b.

A FWS lines have check valves to prevent back drainage of the lines.

i

~

c.

Low-temperature AFWS water is fed directly at the upper section of the SGs into the tube bundle, independent of the main feedwater nozzles, so that the injected water is heated to within a few degrees of saturation prior to pooling above the lower tubesheet.

l

2. 2 AFWS Support 2.2.1 Power Sources The AFWS power supplies are derived from Class lE sources.

Each AM4S train ( A and B) is fed from entirely independent Class lE sources.

These sources include:

a.

AC components are fed from trains A and B Class 1E ac I

buses.

I b.

DC components are fed from trains A and B Class lE de

buses, c.

DC buses are normally fed through rectifiers from their respective ac buses, d.

Station batteries feed the de buses whenever ac power is unavailable.

_a_

The train A AFWS consists of the motor-driven pump and its related components.

Major components of the system receive Class lE power supplies as follows:

a.

Motor-driven AFWS pump - ac power b.

Room cooler fans - ac power c.

Level control valves - ac power through inverters from the de bus d.

Parallel containment isolation valves - ac power to one valve, de power to one valve e

Other valves - ac power The train B AFWS consists of the turbine-driven pump and its related components.

Major components of the system receive Class lE power supplies as follows:

a.

Turbine-driven AFWS pump controls - de power b.

Room cooler fans - ac power c.

Turbine steam supply isolation and control valves -

de power / hydraulic d.

Level control valves - ac power through inverters from the de bus Parallel containment isolation valves - ac power e.

to one valve, de power to one valve f.

Other valves - ac power Upon loss of offsite power, all components in trains A and B receive power from the trains A and B emergency diesel genera-tors, respectively.

To provide further AFWS flexibility, the motor-driven pump and associated components (train A) are capabic of being fed of f of the train B diesel generator by manually switching the power supply breakers via mechanical interlocks.

Upon loss of all ac power (station blackout), the train B AFWS will operate using 125V dc Class lE battery-backed sources.

In such an event, the batteries will supply de power to the components listed above and will provide ac power, through inverters, to the ac-powered AFWS level control valves.

System alignment is such that other ac powered valves do not need to operate following the blackout.

The de system has sufficient eapability to supply the required power for AFWS operation during station blackout for at least 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> (battery limitation).

_9_

l l

i 2.2.2 Alternate Water Sources l

Normal alignment of the AFWS suction is from the non-Seismic Category I CST when the AFWS system is in standby.

All suction valves required for system initiation and control are power operated.

Suction can be aligned either to the deaerators or the condenser hotwell by opening or closing remote manual valves i

operated from the main control room.

i Each AFWS pump train connects to the SWS through two motor-

)

operated, automatically actuated butterfly valves in series.

Switchover of the AFWS pump suction to the SWS is accomplished automatically using a two-out-of-four low pump suction pressure logic concurrent with the presence of an AFNAS.

Upon actuation of this switchover, the nonsafety suction sources are isolated and the two butterfly valves to each service water train are opened.

To prevent spurious opening of the service water valves due to normal transients, the low suction pressure must persist for 4 seconds before the transfer is initiated.

The valves

(

admitting service water can siso be opened from the control room i

or auxiliary shutdown panel in response to an alarm of low pump suction pressure.

I 2.2.3 Steam Availability Steam supply piping to the turbine driver is taken from each of the main steam lines inside the containment.

A line from each steam generator, equipped with a normally closed de motor-operated isolation valve, supplies steam to a common header.

This header leads to the turbine through the containment iso-lation valve and throttle trip valve.

The steam lines are designed to prevent the accumulation of condensate in the lines.

The turbine driver can operate with steam inlet pressures ranging from 45 to 1,160 psig.

Exhaust steam from the turbine driver is vented to the atmosphere above the auxiliary building roof.

Following initiation of an APWAS, steam is admitted to the turbine-driven AFWS pump.

The feed-only-good-generator (FOGG) signals are provided to the steam supply isolation valves of both steam generators, ensuring that only the good steam generator pro-vides motive steam to the turbine driver by closure of the steam isolation valves from the faulted steam generator.

This ensures a steam supply to the pump turbine driver.

The time required to open the steam supply isolation valve and bring the turbine-driven pump to speed is less than 40 seconds. -

)

1 1

2.2.4 Instrumentation and Controls i

Instrumentation for the control and monitoring of the AFWS is located in the MCR.

Instrumentation for system operation needed to achieve plant safe shutdown is also contained on the auxiliary shutdown panel (ASP) and may be used in the event the control room is evacuated.

Manual control of any equipment at the ASP overrides the automatic and manual control capabilities j

of that equipment in the MCR.

This allows full control from the ASP regardless of the mode selected in the MCR.

The manual status of the controls at the ASP is indicated by lights on the l

MCR panel.

The following controls are provided both in the MCR and on the ASP:

Motor-driven AFWS pump (start /stop) a.

b.

Turbine-driven AFWS pump (start /stop) c.

AFWS level control valve position i

1 d.

Service water supply isolation valve position

]

(open/close) l Essential power-operated valves in system (open/close) e.

t 1

l f.

AFWS pump turbine speed control valve position j

Alarms are provided in the MCR for the following:

I a.

Condensate storage tank minimum level b.

AFWS pumps low suction pressure c.

Remote control being overridden by local control d.

Service water supply isolation valves and CST recircu-lation block valves open simultaneously.

e.

AFWS low flow i

I j

The following parameters are indicated both in the MCR and on the ASP:

a.

SG water level b.

SG pressure 1

l c.

AFWS pump suction pressure 4

d.

Motor-driven A pump (running / stopped)

Turbine-driven A pump (running / stopped)

e. i l

f.

AFWS pump discharge pressure g.

AFWS flowrate to each SG h.

Turbine driver steam inlet pressure i

1.

Condensate storage tank level j.

Position indicators for:

1.

All AFWS power-operated isolation and control valves (open/ closed) 2.

Service water supply and condensate storage supply isolation valves (open/ closed) 3.

Turbine driver steam inlet isolation valves (open/ closed) 4.

Essential manually operated valves in the recircula-tion line (open/ closed) 2.2.5 Initiation Signals for Automatic Operation ihe safety-grade AFWAS automatically starts both the turbine-driven and motor-driven AFWS pumps.

AFWAS also auto-matically positions the AFWS valves both to mitigate the consequences of a loss of main feedwater or loss of of fsite power incident and to provide feedwater to allow primary heat removal through the steam generators.

The AFWAS will automati-cally start the pumps under any of the following conditions:

a.

Low pressure in either SG b.

Low level in either SG i

c.

Class lE bus undervoltage d.

Loss of reactor coolant flow indicated by loss of power to three out of four reactor coolant pumps c.

Loss of both main feedwater pumps f.

Emergency core cooling actuation signal In addition to automatic initiation, APWS equipment may be manually actuated from the control room or from the auxiliary shutdown panel.

A bypass is provided to avoid actuation of both the APWAS and the main steam line isolation signal systems by a low steam generator pressure during normal startup and shutdown conditions.

nypasses are also provided to avoid actuation of AFWAS either by I

t I

. =.

loss of the main feed pump trip signal or by loss of three out of four reactor coolant pumps during normal startup and shut-down.

The FOGG system continuously monitors the differential pressure between the steam generators.

When a preselected differential pressure is sensed, FOGG automatically closes the l

following:

a.

The AFWS isolation and control valves supplying j

the lower pressure SG b.

The steam valve supplying the turbine-driven AFWS pump from the lower pressure SG The continuous interrogation feature of this system permits isolation any time during a secondary pressure transient and allows the lower pressure SG to be returned to service, should the pressure differential be reduced by corrective action, such as main steam and feedwater line isolation.

The SGs are protected from overfilling by automatic closure of both the AFWS level control and isolation valves feeding the affected SG on high-high level.

2.2.6 Testing The Auxiliary Peedwater System and its supporting systems are tested periodically to satisfy plant technical specifica-tion requirements.

This testing ensures that these systems will be operable when required by various plant conditions.

The plant technical specifications also limit the time that systems, or portions of systems, may be out of service and identify special testing requirements necessary to ensure plant safety while these out-of-service systems or components are being repaired.

Plant procedures concerning this techical specification testing were not yet available for this analysis.

However, test methods and procedures were assumed and included in the analysis.

These are described in Paragraph 3.3.

2.2.7 Technical Specifications The limiting conditions for operation are that two inde-pendent steam generator auxiliary feedwater pumps and associated flowpaths shall be operable with:

j J

as One auxiliary feedwater pump capable of being powered from an operable emergency bus.

b.

One auxiliary feedwater pump capable of being powered from an operable steam supply system..-..

c.

Operation of the steam-driven auxiliary feedwater pump, except for surveillance and testing require-ments und when actuated by station emergency con-ditions, is prohibited unless the electric-driven feedwater pump is inoperable.

Required action is, that with one auxiliary feedwater system inoperable, restore the inoperable system to operable status within 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> or be in hot shutdown within the next 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br />.

The surveillance requirements are:

Each auxiliary feedwater system shall be demonstrated operable:

a.

At least once per 31 days on a staggered test basis by:

1.

Verifying that the steam turbine driven pump develops a discharge pressure of 1,160 psig above suction pressure at a flow of > 850 gpm when the secondary steam supply pressure is greater than 885 psig when tested as required by Specifications.

2.

Verifying that the motor-driven pump develops a discharge pressure of <

(LATER)* psig at a flow of > (LATER) gpm when tested.

3.

Verifying that each valve (manual, power operated, or automatic) in the flowpath that is not locked, sealed or otherwice secured in position, is in its correct position.

b.

At least once per 18 months, during shutdown, by:

1.

Verifying that each automatic valve in the flowpath actuates to its correct position on an auxiliary feedwater actuation test signal.

I

• (LATER) signifies that the numerical values have not yet been established. --.

l 1

2.

Verifying that each pump starts automa-tically upon receipt of an auxiliary feedwater actuation test signal.

3.

Verifying that the auxiliary feedwater steam generator level control valves maintain a steam generator level of (LATER).

4.

Verifying that the auxiliary feed-water pump stops and the auxiliary feedwater crosstie valve clones automatically upon a high level in i

the associated loop steam generator j

of (LATER) feet concurrent with an auxiliary feedwater actuation test signal.

F

)

5.

Verifying that the auxiliary feed-j water pump restarts when the associa-ted steam generator level falls b:. low j

(LATER) feet from the high level in Item 4 above, concurrent with an auxiliary feedwater actuation test j

signal.

l 3.

Discussion The AFWAS consists of two trains.

Redundant auxiliary feed-1 water pumps are provided which utilize diverse and independent i

motive power.

One pump is tubine-driven with steam generated i

in either or both steam generators and the other pump is driven from vital 4160 VAC electric power.

The condensate storage tank j

provides the primary source of water during emergency conditions.

3 The pump output flow paths are cross connected to allow controlled 1

flow to either steam generator from either pump.

i

}

Sandia National Laboratories addressed the following items as part of the review of the reliability analysis.

1 3.1 Mode of AFWS Initiation i

j The AFWAS automatically starts both the turbine-driven and

)

motor-driven AFWS pumps.

The AFWAS will automatically start the AFWS pumps under any of the following conditions:

Low pressure in either SG.

Low level in either SG.

)

Electrical bus undervoltage.

i

l Loss of reactor coolant flow indicated by loss of power to three out of four reactor coolant pumps.

Loss of both main feedwater pumps.

Emergency core cooling actuation signal.

In addition to automatic initiation, AFWS equipment may be manually actuated from the control room or from the auxiliary shutdown panel.

3.2 System Control Following Initiation Both APWS pumps start in less than 40 seconds.

These pumps continuously supply the required water to the steam generators until the flow is terminated by operator administrative control.

Under emergency conditions, heat is removed from the reactor coolant system by boiling the feedwater in the steam generators and venting the steam to the atmosphere through the power-operated atmospheric vent valves and/or the main steam safety valves.

If main steam isolation valves are open, steam may be relieved to the turbine bypass system if a condenser is available, or through the modulating atmospheric dump valves, if the condenser is not available.

Either method is capable of lowering the RCS temperature to a point where the decay heat removal system can be placed in operation.

The FOGG system continuously monitors the differential pressure between the steam generators.

When a preselected differential pressure is sensed, FOGG automatically closes the following:

The AFWS isolation and control valves supplying the lower pressure SG; The steam valve supplying the turbine-driven AFWS pump from the lower pressure SG.

The SGs are protected from overfilling by automatic closure of both the AFWS level control and isolation valves feeding the affected SG on high-high level.

2 4.

i 4

3.3 Test and Maintenance Procedure and Unavailability d

The Auxiliary Feedwater System and its supporting systems will be tested periodically to satisfy plant technical specifi-cation requirements.

The plant technical specifications also limit the time that systems, or portions of systems, may be out

{

of service and identify special testing requirements necessary.

i f

Plant procedures concerning this technical specification l

testing were not yet available for this review.

However, the following were assumed to be applicable, i

i AFWS Pumps.

The auxiliary feedwater pumps are tested

{

monthly on a staggered basis.

This test requires that the AFWS pump successfully pass 100% of the required flow through the 4

l pump test bypass line at the required pump discharge head.

To develop the required pressure, the pumps were assumed to be isolated from the AFWS at the level control valves during this full flow testing.

During the test, if the AFWS is required to operate, the operator at the test bypass valve must close a

this valve to allow AFWS flow to feed the SGs.

a Every 18 months, the auxiliary feedwater pumps are checked to ensure that they start upon receipt of an AFWAS and that the 4

auxiliary feedwater pumps restart after tripping on high level in the steam generators when the steam generator water level is returned to the normal control band.

AFWS Valves.

All manual, power-operated, or automatic valves that are not locked, sealed, or otherwise secured in position are verified in the correct position monthly.

This j

test is assumed to be a visual check rather than a valve cycling check.

Every 18 months each automatically operated valve is checked to ensure the valve cycles to the correct position upon receipt of an APWAS.

The auxiliary feedwater steam generator level control valves are checked to ensure that they maintain l

steam generator water level and that the feedwater stop valves i

are checked to ensure they cycle shut upon receipt of a high level in the associated steam generator.

}

Auxiliary Feedwater Actuation Signal.

The AFWAS is func-l tionally checked monthly.

Channel checks are performed at least l

every 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br />, and the instrumentation channels are calibrated l

at least every 18 months.

l Condensate Storage Tank.

Level in the Condensate Storage Tank is verified at least every 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br />.

With one of the two Condensate Storage Tanks

• inoperable, an auxiliary feedwater pump supply flowpath is demonstrated to be operable at least daily.

l

• (One CST for each power generation unit.)

• Service Water System.

Service water valves (manual, automatic, or power-operated) which service safety-related equipment are verified to be in the correct position monthly if the valves are not locked, sealed, or otherwise secured in position.

Every 18 months each automatic valve is verified to actuate to its correct position upon receipt of an Essential Safeguards Features Actuation Signal (ESFAS) and each service water pump is verified to start on the ESPAS test signal.

Maintenance Ilardware Failures (Mechanical Components).

Packing replacement and adjustment is the dominant cause of maintenance on valves.

In most cases, this maintenance can be performed with the valve in the correct position for system operation (fully open or fully closed).

Valve repairs requiring disas-sembly of the valve, although not frequently occurring, may have a major impact on system availability due to system isolation requirements necessary to safely perform this maintenance.

Thaue valves which require full AFWS shutdown in order for repair also require a plant shutdown (per technical specifications) and, therefore do not contribute to the maintenance unavailability of the AFWS.

Those valves requiring maintenance which only need a single AFWS pump train to be shut down do contribute to maintenance unavailability of the AFWS.

Valves which are periodi-cally cycled, which have a throttling action, or which are in a high energy system are the dominant contributors to this unavail-ability.

These valves are included in the pump train maintenance unavailability.

Pump maintenance consists of a range of actions from major disassembly to packing adjustment.

Most of the maintenance performed requires isolation of the pump from the system and, therefore, contributes to the maintenance unavailability of the pump train.

The maintenance on large motors ranges from inspection and cleaning to major disassembly.

She prevalent failure mode is bearing failure which requires partial disassembly of the motor.

All maintenance of the pump motor contributes to maintenance unavailability and is included in the pump train maintenance unavailability, Turbine maintenance can range from simple adjustments i

to major disassembly.

Turbine failure is included in the maintenance contribution to unavailability of the turbine driven pump train.

From WASH-1400, the expected frequency of pump maintenance is one act every 4.5 months.

This maintenance is assumed to include the pump, the driver (turbine or motor), and associated l

l I

control circuits.

The maintenance duration ranged from a few minutes to several days.

The plant technic 61 specifications limit this maintenance duration to 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />.

The lognormal mean maintenance act duration is 19 hours2.199074e-4 days <br />0.00528 hours <br />3.141534e-5 weeks <br />7.2295e-6 months <br />.

3.4 Adequacy of Emergency Procedures Emergency procedures have not yet been written.

The AFWS is automatically initiated and emergency procedures for manual verification of AFWS flow to the steam generator and of automatic valve switching to the alternate source upon low suction pressure are planned.

Also, operator actions in the event of station blase kaut need to be specified.

3.5 Adequacy of Power Sources and Separation of Power Sources The AFWS system power supplies are derived from Class lE sources.

Each train is fed from entirely independent Class lE sources.

These sources include:

a.

AC components are fed from trains A and B Class 1E ac buses, b.

DC components are fed from trains A and B Class IE de buses.

c.

DC buses are normally fed through rectifiers from their respective ac buses.

d.

Station batteries feed the de buses whenever ac power is unavailable.

The train A system consists of the motor-driven pump and its related components.

Major components of the system receive Class lE power supplies as follows:

Motor-driven pump - ac power Room cooler fans - ac power Level control valves - ac power through inverters from the de bus Parallel containment isolation valves -

ac power to one valve, de power to one valve The train B oystem consists of the turbine-driven pump and its related components.

Major components of the system receive Class lE power supplies as follows:

i 4

Turbine-driven pump controls - de power 1

Room cooler fans - ac power Turbine steam supply isolation and control l

valves - de power / hydraulic Level control valves - ac power through inverters from the dc bus Parallel containment isolation valves - ac power to one valve, de power to one valve Upon loss of offsite power, all components in trains A and B receive power from the trains A and B emergency diesel genera-tors.

To provide further system flexibility, the motor-driven pump and associated components (train A) are capable of being fed off the train B diesel generator by manually switching the power supply breaker via mechanical interlocks.

)

Upon loss of all ac power (station blackout), the train B AFW system will operate using 125 V dc Class lE battery-backed sources.

In such an event, the batteries will supply de power to the components listed above and will provide ac power, through inverters, to the ac-powered level control valves.

System align-ment is such that other ac powered valves do not need to operate following the blackout.

The de system has sufficient capability to supply the required power for AFWS operation during station

]

blackout for at least two hours (battery limitation).

3.6 Availability of Alternate Water Sources Three sources of water are supplied for the auxiliary feed-water system:

The CST serves as the primary source of water during plant emergency conditions.

The condensate system is used for plant startup, hot standby, or cooldown operations.

The service water system serves as a safety-grade backup system to the CST during plant emergency conditions.

l The CST is always aligned to supply water to the AFWS system during plant operation through a normally open motor-operated i

valve (MOV).

This MOV receives an open signal from the AFWAS during plant emergency conditions..

.m

~

The service water system provides a safety-grade backup.to the CST.

Two MOVs in series supply each pump.

The "A" service j

water train supplies the "A" pump and the "B" service water train supplies the "B" pump.

These motor-operated valves open automatically upon receipt of an AFWAS signal in conjunction with a two-out-of-four low suction pressure condition at the

_ _Y associated AFWS pump.

The low suction pressure trip also closes s.>

the associated normal suction valve from the CST.

o

.s 3.7 Potential Common Mode Factors in AFWS

~

s The qualitative failure characteristic of common mode factors N

is a common link when physical barriers cannot be erected to s

prevent the propagation of failures, and procedural barriero cs must then be erected. Typical common links used in a common ~

cause analysis are:

)

Manufacturer Test / Maintenance Operator Motive Power

~

s 4

Instrument Power Installation N

Calibration z

Similar Parts The method used to perform the common cause failure analysis is based on the fault tree.

Qualitative failure characteristics u

are identified for each basic event.

A search is then performed to identify those combinations of basic events that result in system failure and share qualitative failure characteristics.

Barriers between components, both physical and administrative, are considered in the analysis.

A common cause due to human error may occur.

The error can occur after the pump monthly flow testing.

Essentially, after each pump test, the auxiliary plant operator must close the full flow test valve.

If the pumps are tested sequentially (i.e.,

one pump is tested and at the completion of this test the other pump is tested) common human error or combinations of errors are q

possible.

These errors consist of:

the auxiliary plant operator failing to close the full flow test valve for the first pump and failing to close the second pump's full flow test valve (close coupling is assumed); and the main control board operator failing to notice the valve position indication for the full flow test v

s;

> e.

pc -

s--

s

?

r~

c =% m s.

s y -%

Y

?x

.e y

.g.' ^,

- (

vElves on tNe' main control board'(also close coupled if the n^(

first valve. position indication is missed).

y-s

. 3.,8 Applidation oflData Precented in NUREG-0611 ns cm The approach,.taken in the study is to separate the relia-

., ' ^ _ -

Lility analysds Jnto-two steps.

The first step is determination of minimal cutsets of equiprnent failures, and the second step

~

in determination of Sauses that can bring about those equipment fsilures.

-In the-first step, a detailed fault tree is developed

' ~ '

down to the level of, basic' component failure modes such as " valve MOV-3870A fails'to cpen.'"

^-

L'

~

s

- The fault' tree models allures that must occur to

~ '

prevent successful system

_acion.

The Top Event is defined as "No Or Jnoufficient.Floy to Both Steam Generators."

Success is defined as flow from at'least one pump train delivered to at

'least one steam generator.

The complete tault tree includes the pumps,' valves,felectrical supply, motor operators, and turbine and fontrol.mechabism's.

Not modeled are drain lines, drain valveu, piping, and connected lines which are small in aizea 1.e.,' system components whose failure rates are very low j

' compared tos the ones' included in the model.

The AFWS flowpath is mod +1ed'from the water sources to the steam generators.

Electrically, the syst.em it modele'd from the bus to the system.

' ^

n x

~

.The caus'es of equipment failures are random independent

, f ailures, or independent h'umari errors, or test and maintenance, or chamon cause events,' or ^ o.therb.

x

~

sA*

i, T

Computer programs'are'used to', process information in system reliablity' analyses.

The codes are versions of computer packages that have been in use for many years.

Most of the computer progr.ama were used in support of the Reactor Safety Study, WASH-1400,' and hcVe been modified'as developments are made to reduce m

computer c<jst or improve output pre 6entations.

The computer l

program 9 used on this project'are:

Reliability Analysis System, a progre that does qualitative and quantitative fault tree

~analycis'and COMCAN, a routine that performs common cause failure analysis on fault trees.]^ J

~

stTheidata used for the quantification as requested by the NRC,"is taken from Appendix IM of"NURPG-0611.

l

.x>w s.--

~

'ThN jrsliability anal slo pubmitted by the applicant is

~ ' ' " ~ ~ w

^

- compreh'ensive, quantitatiyer und Tt utilizes the techniques and

- ~

data'i'n NUREG-0611.

s Se'arcY YorNSingle Failure -Points

!N 3.9' 1

....e N

s s

There are no single a'ctive component failure points (SFP) associated dith Case 1,,LM)W, or. Case'- 2, LMFW/ LOOP.

For Case 3, LMFW/LAAC, there are uany SFPc, since; Case 3 describes a single

.+

'4 s

L

)

's s

g c -,

,, laA

• h g

T.

]m

\\.

2,,

- -. ~

d I

)

I J

channel system.

In any case, the CST and the piping connected q;

to this tank have the potential to be passive single component i

failure points if any of these components were to have a severe j

leak or rupture.

The failure probability of such an event is j

orders of magnitude smaller than the unavailability of the AFWS and, therefore, is not important to this analysis.

Any single v

l failura. point has a major effect on the reliability of a redun-1 dant system and if any are found, they should be evaluated for j

their likelihood of occurrence and compensated if they are not j

sufficiently rare.

i!

j 3.10 Human Factors / Errors Due to the short period of time between failure of the AFWS to start and loss of the SGs due to dryout, no operator action I

to recover the AFWS war considered.

However, there are some system failures'from which the operator may recover.

The most significant of these is a turbine-driven auxiliary feedwater j

pump trip.

The dominant contributor to turbine-driven auxiliary j

feedwoter pumps failure to start on demand is a failure of the turbine controls; primarily due to turbine trip on overspeed l

during startup.

The operator may manually reset the overspeed trip, or take control of the turbine-driven AFW pump if, during a demand, this pump did not operate.

The probability of failure j

for the operator failing to take action within 30 minutes is Pg

(

~

which has 0.044 mean with 0.005 variance.

Using this value, a j

point value estimate of the system unavailability (failure to start and no recovery) for the double crossover system design is 1

2.5 x 10-5, During the monthly full flow testing of the AFWS pumps, an operator is stationed at the full flow test bypass valve.

After j

the pamp is started, this operator throttles open the full flow test valve to achieve rated pump flow and discharge head.

Should

[L the AFWS be actuated by a plant transient, this operator must s

l close the full flow test valve to allow the AFWS pump to feed the S3s.

The full flow test is assumed to last 15 minutes per month.

4

['

Pump unavailability due to this test is equal to 3.5 x 10-4 The operator error, failing to act correctly during the first j

five minutes after the onset of an extremely high stress situation is'O.9.

The unavailability of a pump train on demand due to

+

this tailure is 3.1 x 10-4 I

Based upon discussions with plant operators and elementary j

analysis, the probability for failure on demand for the common cause human error is estimated to be 8.4 x 10-6, 3.11 NUREG-0611 Recommendations, Long-and Short-Term l

Reference 2 of this report contained Enclosure 1 which j

stated'a number of short-term generic, additional short-term, and long-term generic recommendations.

The response of CPCo to i

these recommendations are contained in Section 10.4.9 of the l

i I

SER for the Midland Plant.

The issues raised by these recommen-dations have been satisfactorily resolved.

l I

t 4.

Major Contributions to Unreliabilty I

The results of the quantitative analysis for the three l

events are given as follows:

l 1.

Loss of Main Feedwater with Offsite Power l

Available -- No single failures that would l

result in insufficient auxiliary feedwater flow were identified.

The dominant failure I

modes are " maintenance on an AFWS pump and

~

system failure on demand without that pump" and " turbine fails to start and pump motor fails to start."

The unavailability of the AFWS for Case 1 is 1.2 x 10-4 per demand, i

which places this system in the medium reliability group relative to operating PWRs.

2.

Loss of Main Feedwater and Loss of Offsite i

Power

-No single failures that would result in insufficient auxiliary feedwater flow were i

identified.

The dominant failure modes are

• turbine fails to start and diesel generator fails," and " maintenance of turbine pump and cyctse failure on demand without this pump."

The unavailability of the AFWS for Case 2 is f

5.5 x 10-4 per demand, which places this system in the medium reliability group relative to operating PWRs.

If 3.

Loss of Main Feedwater and Loss of All AC --

all AC pc:rer is lost, there is only the TDP train J

available.

In this case, the dominant failures t

i are " maintenance of the TDP," " turbine fails to start," and " turbine pump fails."

The unavailability in approximately 1.3 x 10-2 per demand, which places this system in the medium reliability group relative to operating PWRs.

SNL is in agreement with the above results.

These conclusions j

are plotted in Figure 2 along with the operating plant ratings l

which were derived in Reference 7.

l l

l I l l

TRANSIENT LMFW/ LOSS EVENTS LMFW LMFW/ LOOP OF ALL AC*

PLANTS LOW MED HIGH LOW MED HIGH LOW MED HIGH RANCHO SECC o

o O

l OCONEE 1,2,&3 o

3 O

CRYSTAL 0

3 o

RIVER-3 j

b DAVIS-U BESSE-1 O

O O

~

ARKANSAS NUCLEAR o

O o

O NE-1 THREE MILE O

ISLAND-1 O

O MIDLAND 1&2 e

e o

I i

• ON DEMAND O AFTER THIRTY MINUTES

• SCALE FOR THIS EVENT IS DIFFERENT FROM THE OTHER TWO SCALES Figure 2. Rollability Characterizations For AFWS Designs in Plants Using The Babcock &

Wilcox NSSS And Midland 1 & 2 i

j 5.

Conclusions i

It is concluded on the basis of this review that the applicant has completed requirement (b) of the April 24, 1980 letter.

1 i

The AFWS of the Midland Plant has medium reliability relative to the reliability of operating plants for the first case event.

Quantitatively, the unavailability of the system is 1.2 x 10-4 l

per demand.

Qualitatively, the system is automatically initiated, moderately redundant, has no observed single point vulnerabilities, i

and is tested periodically to demonstrate flow rate at rated 1

pressure.

Failure on demand is dominated by maintenance and i

failure of the turbine and pump motor.

The unavailability for the second case event is 5.5 x 10-4 per demand, which places the reliability in the medium range.

Failure upon demand is dominated by maintenance of the turbine pump, failure of the diesel generators to start, and failure of l

the turbine to start.

i The unavailability for the third case event is 1.3 x 10-2 per demand, which places this system in the medium reliability group.

The TDP train has no identifiable ac power dependencies and is automatically actuated.

The dominant failures are maintenance of 2

TDP train, failure of the turbine to start, and failure of the i

turbine pump.

I Improved reliability may be achieved by eliminating any nonessential maintenance, consolidating maintenance, additional i

preplanning of maintenance, and by additional training of main-j tenance personnel.

I I

i I

o j

1 l

~26-m

.. a r.

-.._,.__m

_.m..., _ __,

G. Glossary of Terma AC Alternating Current

}

ac alternating current AFWAS Auxiliary Feedwater Actuation Signal AFWS Auxiliary Feedwater System ASME American Society of Mechanical Engineers ASP Auxiliary Shutdown Panel B&W Babcock and Wilcox CST Condensate Storage Tank DC Direct Current de direct current DilRS Decay IIeat Removal System ESP Engineered Safety Features ESFAS Essential Safeguards Features Actuation Signal FOGG Feed-Only-the-Good-Generator FSAR Final Safety Analysis Report gpm gallons per minute IEEE Institute of Electrical and Electronic Engineers LAAC Loss of all AC power LMFW Loss of Main Feedwater LOOP Loss of Offsite Power MCR Main Control room MDP Motor-Driven Pump MSIV Main Steam Isolation Valve NRC Nuclear Regulatory Commission NSSS Nuclear Steam Supply System psig pounds per square inch gage RilRS Residual IIeat Removal System SER Safety Evaluation Report SFP Single Failure Point SG Steam Generators SNL Sandia National Laboratories SWS Service Water System TDP Turbine Driven Pump TMI-2 Three Mile Island Unit 2 V

Volt l

l l

7.

References 1.

NUREG-0560, " Staff Report on the Generic Assessments of Feodwater Transients in Pressurized Water Reactors Designed by the Babcock & Wilcox Company," dated May 1979.

2.

NUREG-0565, " Generic Evaluation of Small Break Loss-of-l Coolant Accident in Babcock & Wilcox Designed 177-FA l

Operating Plants," dated January 1980.

3.

NUREG-0611, " Generic Evaluation of Feedwater Transients j

and Small Break Loss-of-Coolant Accidents in Westinghouse-Designed Operating Plants," dated January 1980.

4.

Letter to S.

H.

Howell, Consumers Power Company, from D.

F.

Ross, Jr., Acting Director Division of Project Management Office of Nuclear Reactor Regulation, dated 24 April, 1980.

5.

" Midland Plant Auxiliary Feedwater System Reliability Analysis," PLG-0147, Pickard, Lowe and Garrick, Inc.,

October 1980.

l 6.

Schedule 189 No. 1303-1, Title, " Review of Auxiliary Feedwater System Reliability Evaluation Studies for Comanche Peak, 1 and 2, Waterford 3, Watts Bar 1 and 2, and Midland 1 and 2," dated May 7, 1981.

7.

Wenver, W. W.,

R.

W.

Dorman, R.

S.

Enzinna, " Auxiliary Feedwater Systems Reliability Analyses and A Generic Report for Babcock and Wilcox-Designed Plants," BAW-1584, December 1979.

l t - -.

U.S. NUCLE AR REGULATORY COMMISSION NUREG/CR-2368 BIBLIOGRAPHIC DATA SHEET SAND 81-2164

4. TITLE AND SU8 TITLE (Add Volume No, sf eprensari

2. (Leave btwk) l Midland Plant Auxiliary Feedwater System Reliability Analysis Evaluation 3 RECIPIENT'S ACCESSION NO.

7 AUTHORISI

5. DATE REPORT COMPLE TED B.J. Roscoe McNT" lYEAR i

Auaust 10A1 9 PE RFORMING ORGANIZATION N AME AND MAILING ADDRESS (include 2,p Code /

DATE REPORT ISSUED MONTH l YEAR Sandia National Laboratories Decomber 14A1 Albuquerque, NM 87185 6 IL ** **" * '

8 (Leave Nanki

12. SPONSOHING ORGANIZATION N AME AND MAILING ADDRESS (include 20 Codet

10. PROJECT / TASK / WORK UNIT NO.

Division of Safety Technology Office of Nuclear Reactor Regulation M. CONTR ACT NO U.S. Nuclear Regulatory Commission Washington, DC 20555 NRC FIN A1303

13. TYPE OF REPORT PE RIOD COVE RE D (Inctusere darirs) 15 SUPPLEMENTAHY NOTES

14. (Leave o/ek) 1

16. ABSTR ACT CC4 words or less/

i The purpose of this report is to present the results of the review of the Auxiliary Feedwater System Reliability Analysis for the Midland Plant, Units 1 and 2.

l 17 KEY WOHDS AND DOCUMENT AN ALYSIS 17a DESCRIPTORS i

l 17ti IDENTIFIE RS OPEN ENDE D TEHMS l

18 AVAILABILITY ST ATEMENT 10 SECURITY CLASS (This reporff 21 NO. OF PAGES Unclassified Unlimited 20 SE CuRiTY C,L ASS (Thes papf 22 PRICE Unclassified s

N RC F ORY 335 l? ?71

_, ~.

UNITED STOVES F

]

NUCLEAR R E GU LOTO R Y COMMISSIOM W ASHING TON. D. C. 20555 posy AGE AND P E ES P A*D u s. NUCLE AR PEGuL Af om y OF F ICI AL SUSINESS cumu,sseoM PEN ALTY FOR PRIV ATE USE, $300 U. Mall L

_J I

I I

I 1

- -