ML20052F247

Draft Revision 1 to Rept & Safety Evaluation of Browns Ferry Unit 3 Partial Scram Failure of 800628.

ML20052F247
Person / Time
Site:	Browns Ferry, 05000000
Issue date:	07/06/1980
From:	TENNESSEE VALLEY AUTHORITY
To:
Shared Package
ML19219B002	List: ML20052F247
References
FOIA-81-417 NUDOCS 8205120276
Download: ML20052F247 (3)

Text

- - _ ._ _

l

}

l0 5

/

I i

h E ,aia ,.t m L C M T A t l e

7.,II .- * .; - .,.t s. s QJ j s .

.~~~

• * . p } ,

. ~

m._, g j

• tytJ' ,'

'f

' . '$.'f e- .

t:  %, .,.

!? ^t r

Q.

REPORT AND RATETY

'q EVAttIArggg;

• h !*l.48 .

OF THF, *

'y'.

f' LRCMHS FERRY iTNIT 3 PARTIAL SCRAM FAIL 4FRE,

+

OF JONE 28, 1980 DIv1S10N Or NUC1.I'Ak hwzR Jtal y 6, 1900 l

./

Prepared tsy TuMEF.SEE V A1.1.fN AUT1? ORT 1T lle ne t or Engineerista Rranch  ;,

i .. . l

.. I l (* N.

~, 'b^ __

1 i l i

8205120276 811103 l PDR FOIA I, DIGNAN81-417 PDR

'7,

. .: mi&yKn%5%ge?M9%M. _ ,_ L

__ i

,3 v

1. - r m ,.m . ..-

On June 20 1980 power was reduced to 390 96de on Brownn Ferry tinit 3 by derrenning the recirculation flew and inserting 10 power rode in prepaamtsun for a erheduled s hu t. dowv. for feedwater systeur ammintenance.

At 0130 hourn. the operator initiated a umanual er r man to roumplete the as hu t dr.wn operatSon.

All control rud d r i ve* = (CRD's) r e c ei t ved a sir.r am misnel == ohnerved by the ope t e t ieren yarnannal and ( lee whitt technical adysmor. All the went hank rnde (97) s c r auwmed A ss11-in except rod 30-23. which mottled na poestson "07". liowever. un the cast bank. 75 rode out of the Brt withdrawn f ast led en fully insert. and cause to rest at various notch pontrinne f t omi puseltinna 46 en 02 (average insertion - 10 notches.

Tlee t u r s- power wee substantln11y reduced ( t o _8: 2M by the partim1 in-mertfun and level contru2 wan =mintained norummily by the fandwater control as yis t em .

Unllowinn a drain of the neram disclearse volume for periods o f 91 .

maid 53 == rands, re s.ne c t i ve l y . the opermenr emanually s c r a===e d the reactar two add i t t rena l t i sme n witte t o si l aisse r t iers occurring botts t ienes s .

TI,-

m e rmum discharr.e volume van then milowed to drain a third t iene for 1(,0 weronde, t'gisin r es==*vm 1 ut the n eraum discharge Ievel bypann, the reanc t o r nutunrra d am hAgh drain vn l uume . The desmaininn wi t hdrawn control rnde were then oiss o rvest to insert at normal speed. A sequence of er ven t e in o h nswn below sn Table l'. Tise total time elopeed between the initial acrom and finni insertion of the control rode was 14 minutes.

TABLE 1 Sequence of Rvente j Time Event _

"} O men #1 ocram (manual) 272 ace #1 scram reset M4 sam e #2 acram (unanual) 4"3 oee #2 meram remet 4 7(. mee #3 acrase (smanual)

(.07 eec #3 meram res en t 842 . .r e

1. 4 eeramm (auto) - a11' rode et "OO" At thia point. the assermente cuntinued the normel shutdowra opearatioet

- ...- -. i One rod nottb 1 ss equivalent tu t was eneume ric a l positione 4

j

.. . ~ , . . . ., .. . . . .

- I I

9 I

I i l

t

H[dM _ M 14 -

[

r y;~.~ m4 w e f

II. ,C O,k t A N A t.Y 51 !. ,

Tlie firnt seein u n ! peram was initiated froen 3SE power and 40E rectreulation rinw. he elesie ecchnice) edvfnor initiated a procena computer mean ri f the local power vanne emonitore ( LrstM 's ) a r te r comp 3e t t ers of the s e r oes .

The core waa erftical in the vicinity of the pa r tia lly eersummed rode.

se t a power 3 erv e 3 of about 2T or leem. m en estimated by the IEERM readings.

The heet flun at t h t oi pnwer level de very 3aw, and there is no poemi-bility of fisc 1 d e sma n e un vJulmason af fuel werety Itmite. Of r-Smen data end reactor ruteinnt e nmaril e e have been anmEymed, and are within their exp cred vennes.

III. PO't *f_IRC1 DEtt7 I MS P rCT 10_N Fo12 awi nn the eveset uri Juric 7 tt . 1980 TVA conducted various inspections and envieve se e aussunr s me d

1. Hydrau21r caneral Unite - le=amed i a t e ly after the event. operations and a sma l vi t en a n c a crog t ne e r independently verified the valve 3Speup to be norwal on the eenet bank hydrau11e control unite *tt CU 's) .

Ar r uau l a t o r recharge was versfied in the control room. General F3ectric entineerre subner quent l y mieo verified the 3E ct3 ' s 5tneup.

2 U s e, n e r n wi instruveent di ecisa rge header vente and drain valves were oPer-ated an expected. A brief dreiertption of the CRD meram d i sche r ste system i s: inr2nded a ss At t ache.en t 1. A3en. cin e of tire vent ve3ves wee pulled and a 1. % CrH varuuna pump was connected on the drainage side.

A vecuum of B" of ==.rrury was pulled for a enhnet period which dropped nharply to 2". The r.e a m un Enr thte temporary vacuum won not deterusned.

3. 'J h e s c r ari, finrharge 2nutrue=nne voluene level switchen were enlibrated. Tho Inw level (3 gn21on) and rod block (25 gallon) switches did not activate detrinn t he first calibration (113. S oune reesdue was flushed from the i n n t r swwn t tapn and th-se two switches then operated seriefectorily.

Macorded daen and operment observation did ind 1Lat er that theme two swi t c he ss worked at lemut dia r i n.t part of the inc2 dent. The four high icvel s c r a va (5n r.e)1on) ewitchen were enseted with no problems.

4. tic r ata 111 stus y Review - Ther m ereen litetortem for growne Fearry smette

1. ^ 7 and 3 were reviewed for rod insertibn or any other ir re gularitie s .

Out of 320 m e ramaan . marve r al instonews of single rode latching at v "o2" were noted, but cannat be related t re the subject erywn t .

r escen t acrterrence on tu rnwn e Ferry unit 3 wee eseesu s n ad in which the Oria

• Eroup 2 and 2 m e rsion ein i r n ut d e were de-energized and reset before coeptetson or the aeram atroke. Mais action rarmulted in Isalting a nu==w e of rode before full in=ert poestion wee reached. A true screu afenal was not. Isoweve r . present. An this emne, and t he evemt in un re la t e d .

5. Oi.ntrol sud pr2ve vtsstory - A hi ss t nry of recent ved drive perrnrmanne .

vee em inined for possible relationship to the unit 3 occurrence. Deu prohtee= were noted that could contribute to the event.

6. Mas int en a nce Review - A revient wan wia d e of all rece nt unintenance send emod i f i c ss e t on activittee on strovns Terry unit 3 C Rt) esy n t en . Tlee r e to seu avistence thnt t h e ri a. motivatten enuld have a: ore t r ibu ted to the problem of June 2n. 1980.

pg y_ __

G 9

I

lp ..f

. ?,. <.. . sm REPORT ON

. THE BROWNS FERRY 3 _,

~

PARTIAL FAILURE TO SCRAM EVENT ON JUNE 28, 1980 by the

.A~

OFFICE FOR ANALYSIS AND EVALUATION ,

w 0F OPERATIONAL DATA July 30, 1980 Prepared by: Stuart Rubin, Lead George Lanik NOTE: This report documents results of studies completed to date by the Office for Analysis and Evaluation of Opera-tional Data with regard to a particular operating event.

The findings and recommendations contained in this report are provided in support of other engoing NRC activities concerning this event. Since the studies are ongoing, the report is not necessarily final, and the findings and recofmiendations do not represent the position or requirements of the responsible program office of the Nuclear Regulatory Comissicn.

S g' c;#'t 9 f -

^

4 4 g g

, )

e. .

PREFACE The findings, recommendations, and conclusions contained in this report are, for reasons of timeliness, based on information gathered through informal channels between the Tennessee Valley Authority, the General Electric Company, and the US NRC Headquarters and Regional offices. To . _ _

the extent possible, '.he information used in the report has been verified by cross checking wifa other sources. The findings containec in this

• report, including the underlying causes of the partial scram failure which ,

occurred at Browns Ferry Unit No. 3 (BF-3) on June 28, 1980, relate most -

directly to the Browns Ferry reactor. However, similarities among boiling water reactor facilities leads us to believe that the findings and rec-ommendations may be broadly and generically applied to most if not all -

operating BWRs. To this end, we recommend that a plant-by-plant review, ,

not possible in this investigation, be undertaken by others, to assess the

~

applicability of these findings and recommendations to other SWRs and to

• orovide analysis and evaluation of plant-unique design problems not un-covered in this investigation. Additionally, the scope of our investi-gation and recommendations was intentionally limited so as to address only the specific, direct and underlying causes of the partial scram failure at 8F-3. We have not, therefore, taken the broader view, as could be taken by those most directly involved in the ATWS issue. We do believe, however, that some of the infonnation presented in the report can be useful to those involved in this important generic concern. Finally, this in-vestigation was not able to pinpoint a single precise root cause(s) which led to the BF-3 partial scram failure event, beyond to say it was caused by water in the scram discharge volume. However, we believe that, in totality, the various possible cause mechanisms discussed in this report include the actual, albeit, indeterminable root cause(s) of the event.

As a footnote, the writers wish to acknowledge the invaluable and timely information provided by the BF-3 resident inspectors, James Chase and Robert Sullivan, without whose cooperation, timely issuance of this report would not have been possible.

i

, ) i

)

1 i

TABLE OF CONTENTS Page PREFACE ................................................................ i 1 INTRODUCTION ........................................................ I 2 EVENT SEQUENCE ...................................................... 3 3 DESIGN AND OPERATION OF THE BROWNS FERRY UNIT 3 SCRAM SYSTEM ........ 5 ~l"~^

4 CAUSES INVESTIGATED ................................................. 11 -

5 EVENT SEQUENCE ANALYSIS ............................................. 14

}

6 SCRAM OISCHARGE VOLUME / SCRAM INSTRUMENT VOLUME INSPECTIONS AND TESTS ......................................................... 19 7 PREVIOUS BWR EXPERIENCE OF FAILURE TO FULLY INSERT .................. 23 8 FINDINGS ............................................................ 24 9 RECOMMENDATIONS...................................................... 35 10 CONCLUSIONS ....................................~..................... 40 LIST OF FIGURES Figure 2-1 Control Rod Positions Before First Manual Scram ................... 43 2-2 Control Rod Positions After First Scram ........................... 44 2-3 Control Rod Positions After Second Scram .......................... 45 2-4 Control Rod Position After Third Scram ............................ 46 3-1 Control Rod Drive ................................................. 47 3-2 Scram Electrical Diagram ......................... ................ 48 3-3 Control Rod Scram Group Assignment ................................ 49 3-4 S cr am V a l ve Arr an gemen t . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . - 50 3-5 Scram Volume Orain Arrangement .................................... 51 LIST OF TABLES Table 2-1 Event Sequence Recorde'r Printout .................................. 41 5-1 Scram Discharge Volume Drain Time and Total Positions Inserted .... 52 ii

o i

4. o l

I l

1 INTRODUCTION On June 28, 1980, the Browns Ferry 3 reactor experienced a partial failure of the scram system, while shutting down for a scheduled maintenance of the feedwater system. The reactor had been brought down to approximately 35f.

power by reducing recirculation flow and by manual insertion of control __

rods. The subject event occurred when the control room operator initiated .

a manual scram to make the reactor subcritical which was the next step in [-

the normal shutdown evolution. After manual scram actuation, the control ':

rods on the West side of the core were observed to be fully inserted. How- 2 ever, the control rods on the East side of the core did not fully insert.

Most of the East side rods came to rest in notch positions ranging between 00 and 46 after all East side rod motion had ended. Three additional -- -

scrans and about 14 minutes were required to achieve full insertion of the cartially withdrawn East side control rods. .After all rods were com-pletely inserted, the operators resumed normal shutdown operations. #'

On July 2, 1980, a team of NRC Headquarters representatives from IE, NRR, and AE00 went to the Browns Ferry site to gather detailed information on the event, the scram system design and operation, and the results of scram system tests which already had been performed by TVA personnel. With this  ;

initial direct contact at the plant, an independent investigation of the event cause and the recommended corrective actions was begun by the Office for Analysis and Evaluation of Operational Data (AE00). Over the next several days, additional equipment testing was performed on the BF-3 scram system. Testing and analysis was also being conducted during this time by General Electric in San Jose, California to support TVA activities at the plant. During this period, AE00 continued to obtain, analyze, and evaluate  ;

information as it evolved from these and other sources to continue its investigation.

The purpose of this report is to provide the analysis, evaluation, findings,

o i and recommendations which flowed from the investigation of the BF-3 event by the AE00, US NRC. Section 2 of the report contains an event sequence.

Section 3 provides a description of the design and operation of the BF-3 scram system. Section 4 discusses the possible causes of the event which were investicated and the conclusions in each case. Section 5 provides an event sequence analysis. Section 6 provides a summary of the tests and -.

inspections performed at BF-3 which support the event sequence analysis  ;

and some of the findings. Previous operating experience and investigation - :

findinas are contained in Sections 7 and 8, respectively. Specific ' ' ;

recommendations to correct the deficiencies discussed in the findings are provided in Section 9. The conclusions of this investigation are given in Section 10. __ _

~

w _

e

e 1 a ,

2 EVENT SE00ENCE On June 28, 1980, power was being reduced by the control rocm operator at the Browns Ferry Unit 3 nuclear reactor in preparation for a scheduled shut-down for feedwater system maintenance. By 0131 hours0.00152 days <br />0.0364 hours <br />2.166005e-4 weeks <br />4.98455e-5 months <br />, the reactor power had been brought to 390 MWe via decreased recirculation flow and manual con-trol rod insertion. The operating personnel then initiated a manual reactor --

scram to complete insertion of the remaining control rods (which were at ]

the positionsstate.

to a subcritical shown in Figure 2-1 at the time) and thereby bring th Irredictcly after depressing the manual scram buttons, the cporators placed the reactor mcde switch in the SHUTDOWN mode. Control rocm personnel ob- _. _

served that the blue scram lights for all control rod drive scram inlet and

. . r--

outlet valves were illuminated, indicating that all scram valves were open.

~

Control rod position indication also showed that all of the rods on the West 7 side of the core were fully inserted (except for one which had stopped at position "02"). However, position indication showed that 75 rods on the East side of the core were not inserted fully. The East side control rods came to rest at positions ranging from 46 to 00 withdrawn with an average of about 23 positions withdrawn (position 48 corresponds to fully withdrawn). Rod position ,

indications following the first manual scram are shown in Figure 2-2. At this time 18 rods on the East side were fully inserted. As estimated by the LPRM readings, power level on the East side of the core following the first scram appeared to be less than two percent.

Folicwing scram, the Scram Instrument Volume began to fill and the Scram In-strument Volume Hi Level Scram (level switches) actuated. This occurred some-what socner than expected at about 19 seconds. The Hi Level scram condition was subsequently bypassed by the operator (as allowed in SHUTDOWN mode), to permit reactor protection system reset which occurred 4 minutes and 31 seconds folicwing the first scram.

One minute and 33 seconds later a second manual scram was initiated by the O E o

o .

operator. The time following reset allowed partial drainage of the East and West Scram Discharge Volumes. Rod positions following the second scram are shown in Figure 2-3. After this scram, 33 rods on the East side fully l

inserted. The second manual scram was reset after 59 seconds and the scram discharge volume was allowed to drain for 53 seconds at which time a third manual scram was actuated. Upon completion of this scram, 47 rods on the -

East side were fully inserted. Rod positions following the third manual ]

scram are shown in Figure 2-4. The third scram was reset after 3 minutes and ,

26 seconds. The scram discharge level bypass switch was returned to normal 'Q 2 minutes and 40 seconds later. This action initiated a fourth, automatic scram due to a Scram Instrument Volume Hi Level scram condition which had -

not cleared. At this time all rods on the East side were fully inserted. __

A detailed sequence of events as provided by the event sequerce recorder .

is shown in Table 2-1. The total elapsed time between the initial scram and '

~

final insertion of all rods was 14 minutes 2 seconds. 'At this time the s operators continued normal shutdown operations.

~

l-L _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _

e >

3 DESIGN __AND OPERATION OF THE BROWNS FERRY UNIT 3 SCRAM SYSTEM Mechanical and Hydrauli_c Desicn _of_ the Scram System On a GE BWR, such as Browns Ferry Unit No. 3, the Control Rod Drive (CRD) and its associated Hydraulic Control Unit (HCU) provide the means by which each individual control rod can be rapidly inserted upward into the core -

during a reactor scram. A simplified drawing of the CRD mechanism is shown ]

in Figure 3-1. ,

w During periods of no rod motion, the collet fingers are spring loaded into a

~-

groove en the index tube to hold the drive stationary against the force of gravity. High pressure cooling water is applied below the drive piston and _[

equalized without CR0 motion via controlled inle,akage past the CR0 seals and i into the reactor. A CR0 temperature probe is provided internally to monitor -

each CRD to detect CR0 heat-up should cooling water fic~w be interrupted or ~c should excess leakage of high temperature RCS water flow out through the drive, drive insert line and scram outlet valve. Scram outlet valve leakage into the scram discharge volume on the order of 0.1 cpm would raise the probe temper-ature to the alarm setpoint of about 350 F.

At BF-3, water exhausted from the CR0s is routed to either an East or West -

header scram discharge volume. The scram discharge volume (SDV) is sized to provide a volume of approximately 3.3 gallons per CRD (approximately 600 gallens total). The SOV volume is sized to limit the total amount of hot reactor water leakage past the seals during a reactor scram (maximum volume requirement) while providing enough free space at atmospheric pressure so that back pressure on the CR0s does not increase so rapidly as to impede rod insertion speed (minimum volume requirement). In particular, the system design results in a pressure ,

in the SDV imediately following full insertion rod motions of less than 65 psig.

Low pressure in the SOV is necessary to assure that technical specification scram speeds and full-in rod motion are achieved. The volume of water exhausted through the scram outlet valve of a single normal drive for a full stroke is about 0.75 gallons, not including seal leakage and bypass flow. The leakage

, s and bypass flow for a single drive can be in excess of 5 gallons per minute. ,

Nomal scram time; from full out to 90 percent insertion is less than 3 seconds.

Although the SDV is sized for a volume of approximately 3.3 gallons per drive and .

the drive stroke (without bypass) is only approximately 0.75 gallons, only a -~=

~'

single reactor scram is normally possible with respect to the scram discharge, volume. Leakage of reactor water past the seals fills the SDV rapidly as icng ')

as the scram outlet valves are open which would be the case without an RPS .,

reset. This leakage occurs eve.n on rods that are fully inserted. The leakage isl an average of 2 gpm to 3 gpm per CRD. Thus', frcm this source alone, the 3.3 '

gallons per drive of free volume available in the SDV is filled and pressurized within two minutes. Thus, more than one scram would be possible only if the -:

operator were able to reset the scram (closing the scram outlet valves) well j .

within this time period. Without an early reset, the S,DV would be filled and _.

the SDV would have to be drained to attempt a rescram if rod motion is to be

• produced.

The East and West SDV headers are each provided with a vent line and vent valve.

Each header drains via a separate drain line into a scram instrument volume (SIV) where level monitoring instruments are located. The SIV, in turn, has .

~

drain piping and a drain valve.

During normal operation, the vent valves of the East and West SDV headers and the drain valve of the SIV are open. These valves are kept open to allow the leakage past the scram outlet valves entering the SDV to drain continuously into the SIV so that no build-up of water in the SDV occurs which could prevent a reactor scram. These valves close during control rod scram insertion to '

contain and limit the reactor water released through the scram outlet valves. -

During a scram, inflow of water to the SDV normally continues after control rod insertion is completed due to leakage past the CRD seals. Leakage continues until the scram is reset or until the SDV pressure ecuilibrates with reactor pressure.

.t. i A pressure difference of at least 550 psi must be applied between the CR0 -

insert and withdraw lines to drive the rods in during a scram. The pressure difference applied at the beginning of a scram is provided by the 1500 psia '

scram accumulator and atmospheric pressure in the empty SDV. As CR0 scram insertion progresses, pressure losses in the driving fluid due to line losses -

reduce the insert line pressure to below reactor coolant system pressure.

At that time, the ball check valve, integral to the CRD, allows reactor coolant '

system water to come in under the piston to complete the scram, before any " ";

significant build-up in scram discharge volume pressure due to filling from leakage and bypass flow.

RPS El,ectrical Design A simplified sch a atic of the electrical compone'nts of the Reactor Protection " -

System (RPS) is shown in Figure 3-2. It is divided into two independent trip q channels A and B. Each of the channels can be tripped (de-energized) by either the manual scram relays or the two subchannel relays. The subchannel relays are de-energized and opened whenever any one of a variety of trip conditions exist in the reactor or associated equipment. The automatic logic can be described as "one-out-of-two taken twice." For purposes of analysis of the Browns Ferry event, the automatic trip logic will not be discussed because .I this event occurred first with a manual scram.

With reference to Figure 3-2, both scram solenoid valves A and B must change position to provide a scram. Electrically, this requires a trip of both channel A and channel B. De-energizing the two scram solenoids changes the air flow from the control air supply to the vent path. For manual scrams, a separate scram button is provided on the control panel for each channel. A manual scram

~

is initiated by depressing both the channel A scram button and the channel 3 scram button. Because of the power requirements of 185 separate scram solenoid valves on each channel, each channel is divided electrically into 4 separate scram groups. Control rods associated with the HCUs from the four groups are i distributed randomly throughout the core as shown in Figure 3-3. l

Scram Operation The Reactor Protection System performs its design function by de-energizing the 370 scram solenoid air supply valves (2 for each centrol rod drive HCU), ,

de-energizing the two stram discharge volume (SDV) air. supply solenoid valves, ___ c and energizing the four backup scram solenoid valves in the air supply lines as shown in Figure 3-4.

' ~:

Scram insertion is achieved for each individual centrol rod by opening the 7 scram inlet and scram outlet valves. This appliel 1500 psi accumulator pressure to the " insert" side of the control rod drive piston and vents the " withdraw" side of the piston to the SDV which is at atmospheric pressure. -I

~

.A ^

For normal unscrammed conditions, the scram inlet and outlet valves are held shut by control air pressure applied through the energized scram air supply valves (S39A and S398 in Figure 3-4). Tne 50V vent and SIV drain valves are held open by air pressure applied through the energized discharge volume air supply valves (S37A and S37B). The air header which suppiles control air to all of the 372 air supply valves (370 scram, 2 vent / drain) is pressurized through de-energized backup scram valves (S35A and S35B, S70A and S708). The SDV vent ,

and SIV drain valves can be operated manually from the control room.

A scram signal de-energizes both air supply valves for each rod, de-energizes the scram discharge volume air supply valves, and energizes the back-up scram valves, thus venting air pressure from the scram inlet and outlet valves and the SDV and SIV valves. This causes the scram valves to open and the SDV vent and SIV drain valves to close. In the event the individual control rod air supply valves should fail to change position (i.e., mechanical bind-up, etc.), -

the back-up scram valves which were energized and vented air to depressurize the air supply header assure opening of the scram valves. Thus, even if an air supply valve failed to shift, that rod would still scram. A check valve is provided around the downstream back-up scram valve in the air supply line so the upstream valve can assist in the air header venting or assume venting in case the downstream valve fails.

. s

.4 Physical Layout of the Scram System Hydraulic Comoonents at Browns Ferry Unit 3 At Browns Ferry Unit No. 3, the HCUs for all of the CRDs are physically arranged in rows on the " East" and " West" sides of the reactor vessel, outside the drywell and inside the reactor building. The CRDs on the West side of the core are controlled by the West side HCUs and the CRDs on the East side of the core are -

controlled by the East side HCUs. Drives along the interface centerline, T]

between the East and West sides of the care, are alternately routed to the East ':

and West headers. A simplified diagram of the physical arrangement of the "Q HCUs, scram discharge volume, and vent and drain system is shown in Figure 3-5.

TFe HCUs on each side of the reactor are arranged in 4 rocis. Immediately above . . .

tM 4 rows of HCUs are two cross connected " race track" shaped headers fabri-cated with 6" piping into which the discharge from each scram outlet valve is piped. The two connected 6" headers on the East side corsprise the East bank 7

~

scram discharge volume (SDV) and the two connected 6" headers on the West side comprise the West bank scram discharge volume. Each HCU insert and withdraw line is connected to the'CRDs below the reactor vessel with 3/4" Schedule 80 piping through which the scram inlet and scram outlet water flows (and water for normal rod drive motion). These lines average over 50 feet in length.

The lines from the HCU scram outlet valve to the 50V are fabricated with 3/4" - '

Schedule 80 piping and are approximately 10' in length.

The Scram Instrument Volume (SIV) is located on the West side of the reactor at one end of the West side HCus (and SDV). It is configured as a 12" diameter 10' high vertical cylinder. Single float-type level switches are installed to monitor the 3 gallon and 25 gallon levels. Four float-type level switches are provided at the 50 gallon level for the purpose of initiating a reactor scram (SIV Hi Level Scram) before the SDV begins to fill beyond tne point where complete control rod insertion would be prevented.

At Browns Ferry Unit 3, the East bank and West bank SDV each drain via 2" schedule 160 pipe to a single SIV located on the West side. The drain line for the West bank is approximately 15' long while that from the East bank is approximately 150' long. In each run, the total elevation fall in the line

-g.

, 4 , __

I 4

r is approximately l' 7". On the East bank run this is an averige 0.13" fall per foot of horizontal run.

The drain lire piping at the bottom of the scram instrument volume and the vent _,

line piping at the high points of the slightly inclined East bank and West bank -

SOV headers are routed down to the Clean Radwaste Drain (CRW) piping physically 3; located in the reactor building floor. The CRW system is a closed drain system ~:

which discharges underwater in the Reactor Building Equipment Drain Sump at ' ";

4

- the lowest elevation in the reactor building. Many other equipments are drained and vented by this system.

]

k

_4 e

4 CAUSES INVESTIGATED Immediately following the event at Browns Ferry 3, all aspects of the scram system were investigated in an effort to find the cause. The Reactor Pro-tection System (RPS), the air system, mechanical aspects of the CR0 and various valves, the CR0 and HCU hydraulics, and the possibility of air in the hydraulic system were considered. Finally, attention was focused on _

the East bank Scram Discharge Volume.

Electrical Investications -

Following the first manual scram, the operators verified that the blue scram lights were illuminated for all control rods. Both the scram inlet and the outlet valve stem position switches must show an open valve position  ;

to illuminate these lights. Illumination of these lights for all CRDs would indicate that the electrical portion of the RPS'had successfully generated W-a scram signal to open all scram solenoid varves and that all scram valves q had actually opened for all control rods.

The Reacto- Manual Control System (RMCS) which is designed to control only one control rod at a time was reviewed to determine if there could have been possible interference with the scram function. It. was determined that postulated gross failure of the RMCS and initiation of multiple control rod ,"*-

drive withdrawal signals would not prevent insertion during scram since upward scram forces are more than three times the magnitude of the with-

]

drawal forces under these conditions.

By use of reference drawings, hydraulic control units from each of the four rod scram groups were verified to be randomly positioned on both sides of the core as shown in Figure 3-3. Control rod electrical signals to a group

~

1 red on the East side of the core and a group 1 rod on the West side of the core would be identical. The rod insertion pattern during the event shows that on the East side a number of rods from each electrical group did not completely insert while on the West side, rods from all electrical groups did compig:ely insert.

e

Based on this analysis it was concluded that the failure of rods to fully insert only on the East side was not caused by any electrical mal-  !

function in the RPS trip logic.

TVA test (entitled SMI 150) was performed to verify that the response times for the scram actuating relays to fully de-energize were within technical -

specifications. Verification that they were, eliminated the concern that an electrical problem delayed opening of the East side scram valves which w1 in turn resulted in partial insertion. 'Q A test of the voltage on all channel A and channel B scram groups shewed that all went to zerS following a ma1ual scram and all returned to 125 VAC .. .

when reset. This test was run to verify the requirements of US NRC IE .

Bulletin 80-17. A visual and electrical search of the scram circuitry -

cabinetsforspuriousvoltagesourcesandloosewireslthatmighthave .s provided a path for electrical power) to prevent a dropout of the scram relays was perfomed by TVA. None was found.

CR0 Tests Various tests were run on the CRDs on the East side to verify that CRD ,

seal integrity, friction and scram times were within allowable limits. - -

CRD seal integrity was measured via a stall test. Results of these tests did not indicate any unusual amount of flow during stall conditions and, consequently, the CRD seals were judged to be intact. Stall tests could also have provided a means of detecting scram outlet valve leakage. How-ever, this test was not done. Friction tests and single rod scram tests also showed no anomalies.

Non-CondensibleGasiQ'aHydraulicSystem The effects of air or riitrogen in the CRD Hydraulic system were considered.

Upon questioning, GE CRD experts stated that non-condensible gas in the hydraulic system would only cause problems with normal insert and withdraw motions but would not cause problems with scra" insertion. This is because stepping the rods requires intricate timing of rod motion and latching whereas scramming is a single motion. During a stepping function any non- l condensible gas would undergo compressions and expansions much different

- 1 from the behavior of the non-compressible liquid.

The presence of nitrogen gas in the 50V prior to scram would be no different

~ ~ '

than the presence of air which is there routinely. Following initiation of the scram, the vent valve closes slowly enough to allow a good portion -"4J of the non-condensible gas in the 50V to be vented. -h N

we eh W.

l it 4

1 Eg -\

• }

es.

c 6

m

• y-. , .-

5 EVENT SEQUENCE ANALYSIS As discussed previously, the control rod drive HCU exhausts are partitioned into East and West bank scram discharge headers. Control Rod Drives which discharge into the East header are located on the East half of the core while CRDs which exhaust into the West headers are positioned on the West side of _,

the core. 24 The most notable observation of the control rod positions after the first '

manual scram was that all of the control rod drives exhausting into the West header inserted full-in (except for the CRD at position 30-23 which inserted _

to within one notch of full-in) while the control rods exhausting into the - . .

East bank header inserted an average of only 20 positions. This CRD insertion pattern provides strong evidence that the fundam' ental cause* of the extensive 't; f ailure-to-fully-insert of the CRDs on the East side of-the core was hydraulic -- z7 in nature. More specifically, the rod pattern resulted from an inability of the East header CRDs to exhaust properly for some reason.

With respect to possible multiple scram outlet valve failures, all of the East header scram discharge valves were observed by the control room operators to have opene'iupon manual scram actuation. Additionally, all of the manual ,7 isolation valves on the scram discharge lines of the East header HCUs were inspected by the licensee innediately upon shutdown, with each found to be fully open. Accordingly, the remaining possible hydraulic causes could have been blockages in most of the CRD scram exhaust discharge lines or inadequate i free volume (or high back pressure) in the East header SDV. Subsequent scram testing of numerous East header CRDs which failed to fully insert demonstrated, however, that no blockages existed in the CRD exhaust lines. Excessively

~

rapid buildup of back pressure in the East bank SDV, due to multiple CR0 seal failures, could also be postulated as a mechanism which could inhibit full-in control rod motion. However, stall tests performed on the East bank CRDs,

• See Section 4 for a discussion of other possible causes investigated.

l j

together with individual rod and full core scram tests performed prior to restart, demonstrated that an excessively rapid increase in SDV back pressure I resulting from multiple CRD seal failures wat not the cause of the partial scram failure. Accordingly, it was concluded that, for some reason, the East bank SDV had inadequate free volume available to accept the full scram discnarge from all East bank CRDs exhausting into the East header. Thus, the observed East bank control rod insertien behavior can best be explained on tha basis that the East header SDV was at least partially filled with water when the operator manually scrammed the. reactor.

As discussed in Section 3, adequate free volume must be available in both the East and West headers to accomodate water exhausted during control rod scram insertion. Furthermore, water must be exhausted into t,he SDV with low back pressure on the drive piston to assure that technical specification scram speeds are met. A reduction in the free volume in the SDV could tend to in-crease back pressure on the drive pistons too fast which could then increase

the time required to complete scram insertion. Complete rod insertion would still be possible, however, even for significant reductions in the available free volume in the SDV as demonstrated in recent single CRD scram test simula-tions performed by GE. The GE tests showed that for a 40% decrease in the available 50V, a control rod can still fully insert over a broad range of seal 1 leakage values. For a 70% reduction (i.e.,1.0 gal / drive remaining) in available scram discharge header free volume, the rods could still fully insert if seal leakage rates were small enough. For a reduction in SDV of this alagnitude, however, increasing seal leakage rates can cause the CR0 travel (number of positions inserted) to decrease. The tests show that drive travel decreases to only 36 positions (out of 48) when a 70% reduction is coupled with a seal leakage rate of 8.9 gpm. The GE test cases run for 85% reduction in free volume (.5 gal / drive remaining) showed that even with no seal leakage, the drive would insert only 28 positions and decreased to 22 positions for 5.2 gpm and 18 positions for 8.9 gpm leakage. Finally, as expected, the tests showed that the CRDs would not insert at all if there were no free i volume in which to exhaust (0.0 gal / drive) regardless of seal leakage. Thus,

these tests clearly demonstrate that CRD travel during scram insertions can be sharply reduced if the amount of available exhaust volume is reduced suffi-ciently.

Since the manual rescrams (scrams #2 and 3) occurred with East bank scram dis-charge volume almost full of water on each occasion, these later scrams can be used as models for back-checking the cause of the observed East bank control rod insertion behavior during the first scram. That is, the fullness of the East bank SDV during the first scram can be qualitatively and somewhat quan-titatively inferred by comparing it with rod motions during the later scrams.

The available free volumes in the East bank SDV for each of the later scrams can be calculated by multiplying the drain times discussed in Section 2 by the East bank scram discharge volume drain rates discussed,in Section 6. The amount of free volume which would have had to have been available during these later scrams can also be calculated from the observed rod motions during these scrams together with the GE test results. Comparing the volumes calculated both ways can then be used to show whether or not the observed rod motions during each scram were consistent with the amount of discharge volume made available by the drains between scrams. Once these are shown to be consistent, one can infer the limited amcunt of free volume which must have been present in the East bank SDV during the first scram. The East bank drain times, total number of positions inserted, and average number of positions inserted per rod used in this analysis are shown in Table 5-1.

The drain times between the first and second manual scram was 93 seconds and between the second and third manual scrams the drain time was 53 seconds.

Tests at Browns Ferry show that the normal drain rate for the East SDV is  ;

about 11.6 gpm when East ano West scram discharge volumes are draining  ;

simultaneously. Thus, by multiplying this normal drain rate times the drain i time between scrams, we can calculate approximately how much water could have drained out (free volume made available) of the East bank header during

the periods between scrams. Multiplying, one finds that about 18 gallons would have been made available during the first drain (between scrams 1 and

2) for the second scram while about 10.2 gallons would have been made avail-able during the second drain (between scrams 2 and 3) for the third scram.

On the other hand, from the GE tests and the average rod motion given in Table 5-1 to a first approximation and assuming no CR0 seal leakage, an average of .18 gallons per drive was available for the second scram while about .07 gallons per drive was available on average for the third scram.

Thus, for 93 drives, to a first approximation and given no seal leakage, a total of about 17 gallens of free volume was available in the East SOV for the second scram while about 7 gallons was available for the third scram.

However, if every East bank CR0 were assumed to.have a seal leakage of 5 gpm,*

from the GE test results the required volume per drive would have had to have

~

been no more than about 20 percent more than the above values. That is, about 20.5 gallons of free volume would have had to have been available during the second scram and about 9 gallons for the third scram. ,

Comparing the results of the above calculation , it could be concluded that i

the East SOV was draining normally between scrams one and two, and two and three, and that the average rod insertion during the second and third scrams was the amount which one would expect for the amount of volume made available by the

! drain. Thus, the insertion behavior of the East bank control rods logically could be explained on the basis of a virtually filled 50V during the second and third scrams.

This same approach can now be used to infer the cause of limited control rod j motion during the first manual scram. From Figures 2-2 and 2-3, the average j control rod insertion during the first scram was 20 positions. From this value we would infer (using the GE test results) that there was an average  ;

of only .35 gallons per drive available (or about 33 gallons total) in the [

i

• Censervative based on CRD maintenance recommendations.  ;

e

zi East scram volume during the first manual scram. This assumes no seal leakage. With a 5 gpm seal leakage, we would infer that only about .45 gallons per drive (or about 42 gallons total) had been available.

The above calculational results show that the partial scram failure during the first scram can most easily be explained by having an initially partly filled East bank SDV. Similarly, the subsequent CR0 failures-to-fully-insert  !

are explainable based on a partly filled scram discharge volume.

It should be pointed out, however, that there was considerable spread among the control rods in the number of notches inserted after the first scram.

The variation from rod-to-rod could be explained by CRD-to-CRD differences in such paramet'trn as seal leakage (which significantiv effects number of notches inserted), control rod drive friction, nitrogen accumulator pressures, etc.

Finally, evidence that the East bank scram discharge volume was initially partly filled with water can be found in the elapsed time to activate the SIV Hi Level scram switches following the first manual scram. Reactor scrams at BF-3 prior to the June 28, 1980 event resulted in time delays from reactor scram actuation to SIV Hi Level scram actuaticn ranging # rom 42 to 54 seconds.

The first manual scram from the June 28 event had a delay of only 19 seconds.

For a normally empty SDV and SIV, the time delay would represent the time it takes for water exhausted from the CR0s to enter and begin to fill the SOV, travel down the SDV-to-SIV drain lines, and fill the SIV to the 50 gallon level.

If water were already in the East SDV, water exhausted from the CRDs would almost immediately start to push water out of the East SOV and into the crain line. This would cause the SIV to fill more rapidly. Thus, an elapsed time of only 19 seconds to actuate the SIV Hi Level scram switches provides important evidence that the East 50V was already almost completely filled with water at the time of the first manual scram.

^

1 l

1

. 6 SCRAM OISCHARGE VOLUME / SCRAM INSTRUMENT VOLUME INSPECTIONS AND TE5is Following the partial scram failure event at BF-3, TVA, with the assistance of GE, embarked on an extensive inspection and test program. These inspections  ;

and tests were performed to try to pinpoint what caused substantial water to be present in the East bank scram discharge volume on June 28, 1980, ,

while the scram instrument volume level switches were indicating both headers were empty. The inspection program included physical examinations of the drain and vent piping, the scram discharge and instrument volumes, as well .

as the drain and vent valves. These inspections were perfomed in an ettempt j to determine if a vent or drain line blockage had caused the East bank scram discharge volume to not drain properly. Additionally, drain tests were per-formed on both the East and West n0aders to establish the drain characteristics !

of these components. The folicwing paragraphs summarize the results of these' t inspections and tests. ,

Inspections The 2" drain line between the East bank scram discharge volume and the scram instrument volume were checked for blockages. The drain piping was cut at several locations. Metal tape was then inserted through the drain piping seg-ments. These inspections uncovered no obstructions in the piping between the I

SOV and SIV which could have impeded normal draining of the SOV. A fiber optics inspection of the inside of the SOV at the low point of the 6" diameter SOV (where the 2" drain line connects to the 6" SOV) revealed no foreign objects which could have blocked water from draining out of the 6" SOV into the 2" drain line. The vent piping which cross-connects the high points of the East bank scram discharge header was also cut, flushed, and inspected.

No obstruction was found in these vent lines which could have impeded or r prevented nornal draining of the East 50V. l l

Following the event, the vent valve on the East header was removed and a j vacuum pump connected to the Clean Radioactive Waste (CRW) side of the vent i line. Eight (8) inches of mercury was indicated by the 1.35 CFM vacuum pump I

gauge after a few minutes of pumping but fell off sharply to 2 inches shortly thereafter. Neither the validity of this vacuum reading nor the reason for the apparent and brief vacuum pull could be determined by TVA.

Several days later when the 1.35 CFM vacuum pump was reconnected to the same East header common vent pipe, no vacuum could be drawn after the vent line was flushed. A test of the East header vent valve itself showed that it was operable.

l The scram instrument volume was also visually examined with a boroscope by inserting it through the vent and drain line penetrations. No obstructions were found which could have prevented draining into or out of the instrument volume. .

An inspection of the 6" East bank SDV and drain line siiowed that they sloped continuously downward toward the instrument volume, with the exception of a localized 3/4" rise in drain line at the expansion loop in the steam vault.

This might have been a loop seal of greater depth when the steam vault was hot during normal reactor power operation. The overall drop in the drain line between the East SDV and instrument volume was determined to be l' 7" over its 1

150 ft. length. From the inspections discussed above, TVA was not able to  !

locate a blockage, loop seal, valve maloperation, or other impediment to i draining which could be described as the root cause for holding water in East SDV.

Scram Discharge Volume Vent and Drain Tests TVA performed a series of drain tests on both East and West SDV headers over a period of several days immediately following tile partial scram failure event.

The purpose of these tests was to determine the effects of a restricted vent path on East and West bank SDV drain capabilities and to quantify the nonnal drain characteristics of the SDV. Special test procedures were written for these tests.

Typically, these tests involved initially filling the East and West SDV discharge headers and scram instrument volume tank with room temperature demineralized water. During filling operations, the East and West header vent valves were kept open and the scram instrument volume drain valve was kept closed. Nonnal drain times and drain rates for the East and West SDV headers and scram instrument volume were then detennined by recording the elapsed time necessary to empty these volumes with the vent and drain valves open. Vacuum hold tests (simulating vent line blockages) were performed to determine the drain capabilities of the headers with the vent valves closed.

Water level in the SIV and SDV was monitored by ultrasonic equipment and verified by a clear tygon (mancmeter) tube attached to the scram discharge volume headers. Clearing times of the 50, 25, and 3 gallon level switches attached to the SIV tank were also recorded dur.ing the tests.

~

Summary of Tast Results Scram Discharge Volume Vacuum Hold Tests East Head g With the West header drained to empty, the East header was allowed to drain into the SIV with the East header vent valve and SIV drain valve closed and the West header vent valve open. For this condition (which simulated a blocked East header vent), water draineo from the East SOV into the SIV tank at a rate of only 0.6 gpm.

West Header For this test, the East header was first drained to empty by opening its associated vent valve together with the SIV drain valve. The West header was then allowed to drain into the SIV with the West header vent valve and SIV drain valve closed. For this condition (which simulated a blocked West header vent), water drained fror: the West SDV into the SIV tank at a rate of about 3.2 gpm.

East and West Headers For this test, both the East and West headers were allowed to drain simultan-eously into the SIV tank with their respective vent valves closed and the SIV tank drain valve closed. After an initial water surge, the combined drain rates of the two headers into the SIV tank was 0.6 gpm.

Scram Discharge Volume Drain Tests These tests were performed to detennine the drain times and drain rates of the 50V and SIV during nomal draining (open vent and drain) conditions.

Drain tests were performed for both East and West headers draining at the same time. The system was first filled with the 50V vent valves open ano the SIV drain valve closed. At time zero the drain valve was opened.

Ultrasonics indicated that the West header empt'ied after about 9% minutes while the East header emptied censiderably later at about 25 minutes.

Additionally, the 50 and 25 gallon switches in the scram instrument volume cleared at about 94 minutes and 101/4 minutes, respectively. The SIV 3 gallon switch cleared after 11 minutes and 20 seconds had elapsed. Based on the volumes associated with the 50V headers, these tests showed the average drain rate (with both SOV headers draining together) of the East SOV header to be 11.6 gpm while the average drain rate of the West SOV header was shown to be about 35 gpm. The average drain rate for the SIV based on clearing of the SIV level switches was 24.5 gpm. However, this drain rate was with the East 50V header still draining into the SIV at an average rate of 11.6 gpm. That is, the SIV drained 24.5 gpm faster than the East SOV drained.

7 PREVIOUS BWR EXPERIENCE OF FAILURE TO FULLY INSERT  !

A review of previous BWR experience was performed with respect to failure to fully insert control rods and problems with the SDV. The sources of in-formation used were NUREG-0640 and computer searches of LERs. Ccmputer searches via the NRC LER system and the Oak Ridge Nuclear Safety Information Center data base revealed no later events more significant than those re-ported in NUREG-0640.

Most instances of failure of rods to fully insert resulted in a number of rods latching in position 02 (position 00 is fully inserted). Up to the time of publication of NUREG-0640 in April of 1978, 12 scram events where some rods failed to fully insert were tabulated. These events in general involved a relatively small number of CRDs, between 2 and 15. However, one event at Oresden 2 in November of 1974 involved 96 rods. Ninety-three stopped at position 02, one at position 04, and two at position 06. The only cause reported for the failure of rods to fully insert was damaged stop piston seals. Stop piston seal damage can cause excessive leakage past these seals during a scram which could be large enough to fill (and pressurize)the discharge volume in advance of the control rods reaching their full-in position.

F 8 FINDINGS

1. The cartial failure to scram at 8F-3 on June 28, 1980, was aoparently due to the presence of water in the East scram discharce volume header.

As supported by the tests, inspections and analyses discussed in Sections 4 and 5 of this report, the apparent cause of the extensive failure of con-trol rods to fully insert on the East side of the core was the presence of water in the East scram discharge volume. header.

G

2. The BF-3 scram instrument volume Hi level scram function did not and does not orovide protection against the accumulation of water in the East scram discharge volun.e header (with attendant loss of East bank scram function) even for normal venting and draining conditions.

Drain rate tests performed at BF-3 show that water drains out of the scram instrumen't volume tank censiderably faster than water drains into it from the East bank scram discharge volume header even for nomal, free, unobstructed venting and draining. Based on the tests, the average drain rate of the SIV is approximately 35 gpm while the average drain rate of the East bank scram discharge volume header is approximately 11.6 gpm. For these drain characteristics, water will drain out of the SIV leaving it virtually empty while water may still be present in the East bank SDV. This actually occurred in the East header drain tests. During the test, the SIV emptied about 20 minutes before the East header fully drained. -

With these relative draining characteristics, if water were to leak into the 50V faster than 11.6 gpm, water would accumulate in and fill the East header (since water is being added faster than it can drain out). At the same time, the water draining out of the East header (i.e., at 11.6 gpm) will not accumulate in the SIV since the SIV drains at a faster rate (i.e.,

35gpm). This process would result in water filling the East header with-out an automatic SIV Hi Level scram ev'r e occurring. We have also found that water drains out of the SIV so rapidly that the SIV Not Drained alam would not alam in the control room. Thus, there would be neither control room indication that water is filling an East 50V nor automatic reactor scram actuation to provide protection against partial loss of scram capability. j In view of the above, with regard to the SIV Hi Level automatic scram function, we have found that continuous automatic protection against filling the East bank SOV (with subsequent partial loss of scram function) never did and still does not appear to exist at SF-3. Furthermore, any SWR with a SIV normal drain rate significantly faster than its SDV normal drain rate also would be without automatic protection against filling of the SDV. Although i

t

not verified by test, it is likely that the BF-3 West SDV header also would be in this category.

The loss of automatic scraa functicn can be explained in hydrau'ic head terms. The SIV is a high cylindrical tank and the 50 ga.llon SIV Hi Level scram is located over 8' above the bottom of the tank. Thus, it is necessary to build up a head of 8' in the SIV before the Hi Level Trip switches can actuate. If the drain line from the SIV to CRW is a relatively short line (as is the case at SF-3) an 8' driving head, would result in a fairly rapid drain rate. On the other hand, the SDV header is a horizontal pipe with a small slope. Even when filled, the maximum head of water that can be developed above the SDV drain (at 8F-3) is approximately 2h'. Thus, even with a relatively short drain line between the SDV and the SIV, the ,

flow rate in this line would normally be low b'ecause of the low head.

Actually the SDV header drain and the SIV drain are the same size for BF-3, but the SDV drain is considerably longer. As a result,the lower available hydrostatic head in combination with the higher fluid flow resistance re-sults in a much slower drain rate for the East SDV header than for the SIV.

Such an arrangement can never detect accumulation of water in the SDV.

l l

,A

3. A single blockage in the West header vent or drain line could comoletely disable the automatic reactor protection function installed to protect against a loss of scram cacability for all control rods.

For plants like BF-3 which have one SDV which nonnally drains significantly slower than the SIV, it is possible to comoletely disable the protection provided by the SIV Hi Level scram for both the East and West SOV by ,

postulating a blockage on the faster draining SDV. Reduced flow frcm a blockage on this faster draining header SDV, when combined with the normally slower draining header flow, may total less than the scram instrument volume drain rate which wculd then resuit in the SIV emotying with both SDVs still ccntaining water. This would be a serious and undetectable condition if water inleakage were to subsequently develop into both SOV headers such as to keep the headers full at all times. For such a situation, there would

~

be no automatic scram to protect against a total loss of scram function due to CR0 water inleakage since the SIV water level wo~uld never rise to actuate the SIV Hi level scram switches.

l

4. With the current scram discharge volume / scram instrument volume design, a blockage in the SDV vent or drain oath can cause a cartial loss of scram capability and disable the protection function installed to prevent

.I.t .

As discussed in the previous sections, a blockage in the SDV header vent or drain path will drastically reduce the drain rate of the scram discharge volume. Water leaking past the scram outlet valves (or fran other sources) I could then cause the scram discharge volume to fill. Since the CR0 temperature probes would allow about .1 gpm of undetected leakage, as much as 9 gpm could leak into the SDV header undetected from all CRDs. Thus, given a partially blocked West header drain, for example, the West header could easily start to fill with water, leaking in undetected through the West side CR0 scram outlet valves. At the same time, since the drain rate of the West header witP. ? Pain line blocked could now be substantially less than the SIV drain rate, water would not accumulate in the SIV. Therefore, the SIV Hi Level scram switches would not actuate to prevent filling of the header.

Thus, with the present SDV/SIV and Hi level scram arrangement, a single failure such as a blockage of a SDV drain or vent can help initiate a partial loss of scram capability and disable the protective function designed to ,

prevent the loss.

l t

_ 28

5. There are numerous actual and cotential mechanisms for introducino and retaining water in the SOV with no accumulation in the SIV.

Review of the vent and drain paths for the scram discharge volume and the scram instrument volume has shown that there are numerous actual and potential mechanisms which could slow or even stop SOV drainage into the SIV. Since the SIV would still maintain a high drain rate, it would be possible for the SOV to retain water while SIV instrumentation indicates empty.

Possible sources of water are: water from the previous scram; multiple scram cutlet valve leakage; or injection from SOV flush lines.

Mechanisms which retard free draining of water out of the SOV include:

a blockage in the vent piping; a plugged SOV-t6-SIV drain line; a closed SOV vent valve; a vacuum held in the 50V by a loop seal somewhere in the vent line; vent line siphon effects from water in the 50V vent line; venting to the closed CRW system in the Reactor Building Orain Sump below water wi" - vacuum breakers; vacuum effects from fluid flows through the CRW piping system; Vacuum effects from condensing hot water in 50V from the previous scram.

Venting of the 50V to atmospheric pressure while the SIV drains into the closed CRW drain system (which could be pressurized above atmospheric pressure) could also inhibit draining of the 50V headers if there is insufficient downward slope in the 50V drain line. Since the CRW exhausts under water in the Reactor Building Orain Sump and non-condensible gases are present in the fluids draining through the CRW drain system, there is a possibility for pressure to build up in the CRW drain system. This pressure, in conjunction with a small loop seal in the drain line from the SDV to the SIV, could hold up water in the SOV even if the SOV were vented directly to atmosphere. ,

l

, i i

!~

6. The current scram discharge volume / scram instrument volume design results in the automatic Hi Level scram (safety) function being directly deoendent on the nonsafety-related reactor building Clean Radioactive Waste drain system.

For the scram instrument volume Hi Level scram switches to activate, water must accumulate in the scram instrument volume. For water to be able to accumulate in the SIV, it must be able to drain at an adequate rate from the SDV into the SIV. However, from the drain rate tests performed at BF-3, improper venting of the SDV can sharply or totally prevent water from draining out of the 50V. Proper draining of the SOV is directly dependent on the venting function provided by the reactor building Clean Radioactive Waste drain system (a required systems interaction). Accordingly, we would conclude that operability of the SIV Hi Level scram function is dependent on the venting provided by the nonsaf,ety-related reactor building CRW system. Unanticipated adverse venting behavior of the CRW system, which results in reduced ve' ting of air back into the'SOVs, can result in the holdup of water in the S)V with little or no accumulation of water in the SIV. This dependancy appears to be particularly inappropriate if not unacceptable for a reactor protection function which is intended to prevent the loss of reactor scram capability.

l 1

7. The BF-3 partial scram failure event, together with recent events at other BWRs, have shcwn that float-tyoe water level monitoring instruments have a significant degree of unreliability.

The BF-3 partial scram failure event demonstrated on several occasions a significant unreliability of float-type level switches. As shown on the event sequence recorder printcut (Table 2-1), several of the 50 gallon level instruments failed to activate on different occasions. Furthermore,

' during calibration testing of the SIV level switches following plant shut-down, both the 3 gallon and 25 gallon switches were found to be inoperable.

After the inst.*ument taps were flushed of residue, the switches operated satisfactorily. During drain rate testing of the BF-3 SDV, two of the four 50 gallon switches failed to activate twice in.two drain tests. Additicnally, inspections at Brunswick Unit No.1, following a reactor scram on November 14, 1979, revealed inoperable alarm and rod block level switches due to bent float rods. Other surveillances and inspections at Hatch Unit 1 on June 13, 1979, fourid two SIV Hi Level switches inoperable due to bent floats binding against the inside of the float chamber. These recent experiences indicate a significant degree of unreliability of float-type level switches resulting from various causes.

l l l l

8. With the current SWR Reactor Protection System loaic, the cresence of certain automatic scram conditions oreclude SDV drainina (scram reset) to cermit a rescran.

In order to drain the 50V for rescram following a scram actuation, it is necessary to reopen the SDV vent and drain valves and to reclose the scram inlet and outlet valves (RPS reset). This requires the following steps:

1) place the reactor mode switch in SHUTDOWN or REFUEL; 2) actuate the DISCHARGE VOLUME HI WATER LEVEL BYPASS switch; 3) the reactor trip signal must clear or be bypassed in SHUTDOWN or REFUEL modes; and finally 4) reset the RPS. However, the following reactor trip functions cannot be bypassed by the operator in the SHUTDOWN or REFUEL mode:

Drywell High Pressure ,

Reactor Vessel Low Level Main Steam Line Hi Radiation Neutron Monitor System Trip Reactor Vessel High Pressure Condensor Low Vacuum

• Main Steam Line Isolation Valve Closure

• Thus,if any of the above trip conditions are present, resetting the RPS would not be possible.

For example, if a starious MSIV closure event should occur at power with the 50V initially full of water, a reactor scram would occur with the control rods failing to fully insert. If the MSIV closure trip (or Reactor Vessel High Pressure) condition persisted, then a rescram attempt would not be possible since it cannot be bypassed in SHUTDOWN or REFUEL modes. Thus, the trip condition itself would prevent the possibility of rescram. However, we do not consider that any modification is required in the RPS trip / reset circuitry to enable the operator to reset the RPS in the presence of any automatic scram condition, since the capability to reset and rescram has

' not been defined as a required protective action.

• Depends on Reactor System Pressure Interlock setpoint.

32

9. If a scram condition exists which cannot be byoassed in SHUTOOWN or REFUEL mode, then f ailure (to close) of a SOV vent or SIV drain valve can result in an unisolatable blowdown of reactor coolant outside crimary containment.

With the reactor in an unscrammed state, the scram outlet valves provide both a reactor coolant pressure boundary function and a primary containment isolation function. During a reactor scram, the scram outlet valves open (one per control rod drive) and the SDV vent and SIV drain valves close.

Reactor coolant pressure boundary integrity and primary containment iso-lation functions are then transferred to the scram discharge volume vent and SIV drain valves which seal she SDV. We' have found that there are no redundant isolation valves in the vent or drain lines to provide these isolation functions for a scram condition. The failure of any one of these valves in the open position, therefore, could result in an uncontrolled blowdown of reactor water outside primary containment and into the CRW drain lines if the operator could not reset the scram. The blowdown would ultimately discharge to a drain sump which is not des _igned to handle the heat load or pressure buildup. With the present BWR RPS design, the operator would be able to reestablish primary containment isolation (with scram out-let valve closure) only if the RPS could be reset. However, if a reactor scram condition persists and it cannot be bypassed in SHUTDC*WN or REFUEL mode (i.e., any of those listed in Finding #8) it would be impossible to reset the RPS to terminate the blowdown.

Thus, for example, a scram caused by spurious closure of the MSIVs with a failed open scram instrument volume drain valve would result in an uncon-trolled blowdown of reactor coolant outside primary containment and into the drain sump room which contains the engineered safeguard pumps which are required for mitigation. Blowdown wculd continue as long as the MSIV closure scram condition existed (MSIVs not reopened) since this trip cannot be bypassed in SHUTOOWN or REFUEL mode. That is, the scram outlet valves could not be reclosed to isolate the blowdown until the MSIVs could be reopened. For events which result in scrams caused by conditions which cannot readily be cleared, uncontrolled blowdown into the reactor building i (secondary containment) could be sustained for an indefinite period of time with possible envircnmental impact on the required mitigating features.

e 0

10. The emeroency coerating instructions at BF-3 did not include a crocedure or guidance for the coerator to follow in the event of a cartial or comolete scram failure.

The Browns Ferry plants, as perhaps do most (if not all) other SWRs (and probably all other LWRs), do not have emergency procedures for the operator to follow in the event of a partial or complete scram failure. We have found that, although control room operators are trained to verify that the rods have fully inserted upon a scram actuation, procedures do not exist for the operator's inmediate or subsequent actions if full control rod insertion does not occur. Moreover, although operators are fully knowledge-able of the function and operation of the standby liquid control (poison) system, the plant does not have specific procedures which state when the alternate shutdown system must be actuated.

G b

t 9 RECOMMEN0ATIONS

1. The coerability of the Scram Instrument Volume Hi level Scram function should be indeoendent of the Scram Discharge Volume venting and draining reouirements.

The current BWR scram discharge volume / scram instrument volume design configuration requires proper venting of the SDV and proper SDV-to-SIV draining to assure operability of the scram instrument voluae Mi Level scram function. We recommend that the operability of the Hi Level scram be made independent of SDV venting or draining requirew:nts. We make this recommendation because of Finding Nos. I through 6 discussed in the previous section. That is, the hydraulic factors which control '

water level in the SDV and SIV should not be able to negate the response of the Hi Level protection function. We be'11 eve the acceptable configura-tion would be to place the SIV tank directly under- the low end of the 6" SDV header and to connect the top of the SIV tank to t' . bottom of the icw end of the SDV header-by-afshort vertical 6" diamecer pipe (rather than the current 2" diameter horizontal pipe). This arrangement should assure water spillage from the SDV directly down to the tank containing the level monitoring instruments. Furthermore, it would not depend on venting or draining phenomena which are sensitive to blockages. We also recommend two separate scram instrument volume tanks, one on each SDV header bank.

Separate instrument volumes, in immediate proximity to their respective headers, should assure proper water spillage into the SIVs and provide adequate redundancy for protection against a total loss of scram capability.

It is our firm belief that modifications which simply improve the venting of the SDV/SIV volume arrangement to assure operability of the SIV Hi Level scram function are not adequate. We recocriend that this uniquely '

important safety function be made completely indeoendent of any vent or drain arrangements, thereby separating the water accumulation control and protection functions. We further recommend that in situ fill tests be performed to demonstrate that the operability of the protective Hi i Level scram function is insensitive to the vent or drain arrangement for the design configuration finally installed.

2. Scram instrument volume water level monitoring instruments for the SIV Hi Level scram function should be both redundant and diverse.

It is recommended that diversity be added to the redundancy of SIV level monitoring instruments for the SIV Hi Level scram function. Currently, i there are redundant float-type level switches for each RPS channel for the Hi Level scram function. On several occasions recently, as discussed in Finding No. 7, more than one float-type level switch was observed-to be inoperable at once. During and immediately following the BF-3 partial scram failure event, several float-type level switches in the instrument volume f ailed to actuate. In view of these experiences, we recommend that diversity be included in the level monitoring function )

for the SIV Hi Level scram function. The important and unique protection provided by this trip function requires that the presence of water in the SIV be monitored continuously with extremely high reliability. We are recommending that diversity be added in order to assure this reliability.

Monitoring techniques, such as differential pressure cells, ultrasonic detection or conductivity probes, may be considered along with others for this purpose.

i i

i . .

3. All vent and drain paths from the scram discharge volume and scram instrument volume should have redundant automatic isolation valves.

As discussed in Finding No. 9, scrams which occur as a result of autanatic reactor trip conditions which cannot be cleared or bypassed in REFUEL or SHUTDOWN modes can result in unisolatable reactor system blowdowns out-side of primary contain'ent if the SDV vent or SIV drain valve fails to close. To protect against such occurrences, we reccmmend that redundant valves be placed on all vent and drain lines connected to these volumes.

Redundant valves would also protact against equipment damage which might otherwise occur as a result of excessively slow closure or delayed closure of one of~the isolation valves. These valves must be qualified and capable of closing against full reactor pressure, flow, and tenperature conditions in case the lines are not isolated within normally specified time limits. The vent and drain lines and drain supports must also be designed for the hydraulic loads and instabilitics associated with the blowdown of the high pressure / temperature reactor coolant to the drain system. Prolonged blowdown may be ruled out as a design basis with appropriate diverse isolaticn or other acceptable provisions. Blowdown instability due to isolaticn valve time delay is believed to be the cause of failure of the float-type level switches at Brunswick Unit No. 1.

E

4. Emergency operating procedures and operator training should be provided for complete and partial scram failure conditicns.

In view of Finding No.10, we recommend that snergency operating proceoures and training be provided to centrol room operators to respond to partial or complete scram failbre conditions. These procedures should include explicit statements regarding the conditions for which the standby liquid control system must be used. The procedures should incluoe cautions regarding operator actions which should not be taken which could result in a severe transient 'conditica (e.g.,

main turbine trip) being created. The procedures should provide guidance to the operator for starting up safety systems for standby readiness (e.g., HPCI on minimum flow) or for tripping other systems (e.g., re-circulation pumps). The order of operator actions (i.e., immediate, subsequent) should be considered, as well as when the operator should begin attempting to insert rods manually.

We believe that such operations (human factors) aspects can and should be implemented in the near term. Such procedures and training would assure, in the near term, the most appropriate control room operator action during a scram failure event and well in advance of any ATWS modifications which may be required in the long term.

_m.

v

j. - -

i I

5. Consider modifying the 50V vent and SIV drain arrangement to imorove scram discharge volume drain reliability.*

As discussed previously, scram discharge volume draining currently depends on uncertain ver,t and drain functions provided by the reactor building Clean Radioactive Waste drain piping, along with relatively small diameter, ncnredundant vent and drain piping, which are susceptable to blockage. This current, relatively unreliable SDV venting arrangement could be improved by increasing the vent line size and by adding an alternate, reliable, and isolatable vent' path. The alternate vent path could be installed with a check valve and air operated isolation valve to provide an alternate and isolatable path for air inleakage into the scram discharge volume. The alternate' vent path could be vented either directly to the Reactor Building atmosphere or to a gas treatment system with a vacuum breaker. The check valve would provide automatic isolat' ion of this redundant line upon pressurization of the scram dis-charge volume during a reactor scram. The drain function could also

,- be improved by providing a second drain line from the SIV to the CRW floor drain.

We believe that modificationi Gu:5 as those described above, would help improve SIV drain reliabC ity. Ircrovements such as these would thus help l to further reduce the a w;47 challenges to the SIV Hi level protective scrma function.

• Although this recommendation is only for consideration, we do believe that it would further reduce the risks associated with loss of scram capability arising from water accumulaticn in the SDV.

l

10 CONCLUSIONS ,

The Browns Ferry Unit 3 partial scram failure event which occurred on June 28, 1980, demonstrated that the present BWR scram system can be vul-nerable to loss of scram capability while operating at power. Furthermore, the event showed that the loss of scram capability can occur in a way which goes undetected by the operator and unprotected by the reactor protection system.

The information and analysis of the BF-3 partial scram failure, which is provided in this report, concludes that the cause of the loss of scram capability was the presence of water in the East scram discharga header.

Furthermore, our analysis of the scram discharge volume / scram instrument volume design configuration, together with its vent and drain characteristics, leads us to conclude that numerous actual and postulated mechanisms exist which can cause the scram discharge volume to fill undetected and without protection against such filling. Our analyses also show that certain scram events can result in an unisolated reactor coolant blowdown outside of primary containment following a single isolation valve failure.

In view of these design deficiencies, we believe it necessary that modifica-tions be made to the scram discharge volume / scram instrument volume arrange-ment and isolation features. Our specific recommendations for change in the SDV/SIV design which flow from our findings have been provided in this report.

We believe that these recorsnendations should be considered along with those of others who are also reviewing the BF-3 event. We do believe, however, that the design changes described in the recorrcendations are necessary to adequately reduce the risks associated with the present unreliability of the BWR scram system arising from undetected accumulation of water in the scram discharge volume.

Table 2-1 Event Sequence Record:r Printout 01 31 16 A034 _

Reactor Scram Manual B CY 5 34 A034 CY 6 44 A033 Reactor Scram Manual A 01 31 24 A035 Reactor Trio Actuator Al or A2 CY 3 38 A035 CY 3 39 A021 Reactor Low Water Level A CY 3 42 A023 Reactor Low Water Level C CY 3 47 A021 Reactor Low Water Level O CY 3 47 A036 Reactor Trip Actuator 31 or 32 CY 3 56 A022 Reactor Low Water Level 3 CY 5 11 A076 REPT C Tripped 01 31 34 A003 Discharge Volume High Water Level C CY 3 42 A003 CY 5 01 A004 - Discharge Volume High Water Level 0 01 31 37 A002 Discharge Volume High Water Level B CY 6 58 A002 01 31 40 A001 Discharge Volume High Water Level A CY 0 03 A001 .

CY 0 18 A106 Malfunction Bus Energized-CY 0 33 A038 Turb. Stop Valve Closure Scram Trip A CY 0 33 A040 Turb. Stop Valve Closure Scram Trio C CY 0 33 A041 Turb. Stop Valve Closure Scram Trip D CY 0 34 A039 Turb. Stop Valve Closure Scram Trip B CY 0 47 A043 Turb. Gen. Load Rejection Scram Trip B CY 0 47 A045 Turb. Gen. Load Rejection Scram Trip D CY 0 48 A042 Turb. Gen. Load Rejection Scram Trip A CY 0 48 A044 Turb. Gen. Load Rejection Scram Trip C A084 Turb. Tripped - Loss of Hydr. Trip Pressure 01 32 01 N021 Reactor Low Water Level A CY l 12 N021 -

CY l 57 N023 Reactor Low Water Level C CY 2 04 N024' Reactor Low Water Level O CY 3 35 N022 Reactor Low Water Level B 01 34 45 A058 IRM Upscale Trip on Level F CY 4 30 A058 01 34 48 A057 IRM Upscale Trip on Level O CY 7 36 A057

{Y 0 Q7 'l057 CY 8 13 A057 l

o .

= 0 Table 2-1 Event Sequence Recorder Printout CY 0 26 N058 IRM Upscale Trip on Level F CY 0 49 A056 IRM Upscale Trip on Level B CY 0 55 N056 CY l 01 A056 CY 1 16 N056 CY l 41 A056 CY l 48 N056 CY l 56 A056 CY 2 21 N057 IRM Uoscale Trip on Level O CY 4 14 N056 IRM Upscale Trip on Level B 01 42 00 A035 Reactor Trip Actuator Al or A2 CY 9 23 A035 01 2 37 !035 CY 6 35 NO35 CY 6 36 NO36 Reactor. Trip Actuator Bl'or B2 CY 8 05 NO34 Reactor Scram Manual B- .

CY 8 06 NO33 Reactor Scram Manual A 01 c5 17 A035 Reactor Trip Actustor Al or A2 CY 6 47 A035 CY 47 A036 Reactor Trip Actuator B1 or B2 01 45 36 N002 Discharge Volume High Water Level B CY 6 09 N002 CY 6 16 A002 0146 30 A031 Reactor Scram Manual B CY 9 48 A034 CY 9 48 A033 Reactor Scram Manual A 01 47 43 NO35 Reactor Trip Actuator Al or A2 CY 2 37 NO35 CY 2 38 NO36 Reactor Trip Actuator B1 or B2 CY 3 05 NO33 Reactor Scram Manual A CY 3 05 NO34 Reactor Scram Manual B 01 57 04 N002 Discharge Volume High Water Level B CY 3 22 N002 CY 4 08 N001 Discharge Volume High Water Level A 01 57 34 NC04 Discharge Volume High Water Level O CY 4 07 N004 CY 4 19 N003 Discharge Volume High Water Level C 02 29 06 0 E9 TVA BFUP 00 .i t EAST WEST 59 48 48 48 48 48 48 48 55 48 48 48 42 48 42 48 48 48 ,

51 48 48 48 48 48 48 48 48 48 48 48

 ?

48 12 48 0 48 8 48 8 48 0 48 12 48 43 48 48 48 48 48 48 48 48 48 48 48 48 48 48 48 39 38 48 48 48 0 48 48 48 48 48 12 48 48 48 33 35 48 48 48 48 48 48 48 48 48 48 48 48 48 48 48 31 48 48 0 48 48 48 8 48 8 48 48 48 0 48 48 27 48 48 48 48 48 48 48 48 48 48 48 48 48 48 48 23 38 48 48 48 0 48 48 48 48 48 0 48 48 48 38 19 48 48 48 48 48 48 48 48 48 48 48 48 48 48 48 15 48 12 48 0 48 8 48 8 48 0 48 12 48 11 48 48 48' 48 48' 48 48 48 48 48 48 07 48 48 48 42 48 42 48 48 48 03 48 48 48 48 48 48 48 02 06 10 14 18 22 26 30 34 33 42 46 50 54 58 1800 Figure 2-1 Control Rod Positions Before First Manual Scram

Q0 EAST WEST

"" 59 40 4 36 1 42 0 0 0 t

55 36 46 38 2 ' 0, 0 0 0 0

• 51 36 34 0 36 0 0 0 0 0 42 ;10 47 42 0 20 0 24 0 0, 0 0 0 0 0 0 40!44 43 12 0 30 28 40 33 0 0 0 0 0 0 0 39 24 0 40 0 0 34 32l0, 0 0 0 0 0 0 0 35 30 4 38 0 18 8 34 !30 0 0 0 0 0 0 0 31 22 10 0 34 28 2 0 l 0, 0 0 0 0 0 0 0 27 42 36 30 14 26 28 36 ! 24 0 0 0 0 0 0 0 i

23 8 0 38 32 0 12' 36 0 0' 2 0 0 0 0 0 9

19 34 42 34 20 10 28 22 26 0 0 0 0 0 0 0 15 40 0 36 0 0, 0 0 0 0 0 0 38 l 0 11 26 6 14 40 40 24 0 0 0 0 0 07 0 40 2 0 0,, 0 0 0 0 03 '

, 28 42 42 30 0 0 0 02 06 10 14 18 22 26 30 34 38 42 46 50 54 58 1800 Figure 2-2 Control Rod Positions After First Scram

on l

EAST WEST 59 30 0 14 ~18 0 0 0 55 26 42 28 0 0 0 0 0 0 ,

51 22 0 28 34 0 0 0 0 0 0 12 47 30 0 0 0 12 0 0 0 0 0 0 0 0 43 0 30 38 0 14 4 30 26 0 0 0 0 0 0 0 39 8 0 30 0 0 26 14 0 0 0 0 0 0 0 0 35 14 0 30 0 4 0 22 20 0 0 0 0 0 0 0 31 8 0 0 14 12 0 0 0 0 0 0 0 0 0 0 27 26 22 20 2 12 12 16 2 0 0 0 0 0 0 0 23 0 0 30 20 0 0 16 0 0 0 ,0 0 0 0 0 19 10 26 24 0 0 8 ~6 4 0 0 0 0 0 0 0 15 18 0 18 0 28 0 0 0 0 0 0 0 0 11 12 0 0 14 0 0 0 0 0 32l30 07 0 30 0 0 0 0 0 0 0 03 2 34 22 4 0 0 0 i

02 06 10 14 18 22 26 30 34 38 42 46 50 54 58 1800

Figure 2-3 Control Rod Positions After Second Scram l

4 . o Table 5-1 Scram Discharge Volume Drain Time and Total Positions Inserted Orain Time from Total Positions Posi', ions Scram No. Previous Scram. Sec. Inserted Per CR0

~

2 93 956 10.2

3 53 402 4.3 4 160 800 8.6 il 3

l i .

4 a

i l

i j

1 i

4 i

.e-,....-

~

co EAST WEST 59 24 0 'O O 0 0 0 55 20 40 22 o o o o o o ,

51 0 16 0 24 30 0 0 0 0 0 0 47 26 0 0 0 6 0 0 0 0 0 0 0 0

1. 3 0 24 36 0 2 0 24 18 0 0 0 0 0 0 0 39 2 0 24 0 0 20 0 0 0 0 0 0 0 0 0 35 6 0 26 0 0 0- 14 16 0 0 0 0 0 0 0 31 0 0 0 2 4 0 0 0 0 0 0 0 0 l0 0 27 18 16 14 4 2 6 2 0 0 0 0 0 0 0 0 23 0 0 26 12 0. 0 2 0 0 0 0 0 0 0 Q

19 0 16 18 0 0 0 2 2 0 0 0 0 0 0 0 15 4 0 10 0 0 0 0 0 0 0 0. 0 ,0 11 6 0 0 28 26 8 0 0 0 0 0 07 8 0 24 0 0 0, 0 0 0 0 03 0 30 10 0 0 0 0 02 06 10 14 18 22 26 30 34 38 42 46 50 54 58 1800 Figure' 2-4 Control Rod Positions After Third Scram

. , e se

"' ~

^

COUPLIN G

$ PUD BOTTOM OF b [ '

REA".Tei VES$EL L ) Q < _a i- A

/

y r

/ 2 HOU$1NG FlXEO Pl$ TON 5-

] ($ TOP PISTON)

RETURN $PRING .  :

COLLET FINGERS C 4li ;

i _.:t ;i 2!

COLLET PISTON j p ON W8E g

i'

? ,, ,/ ~

7 INCEX TUBE

/

{ ) $' ,

CYLINDER --- ) '

d

l p

)

CRIVE Pl5 TON ORIVE IN$ERT LINE k ) e L LIN l .

, J ,

g - -

VALVE ,

PR = REACTCR REs3URE f Figure 3-1 Centrol Rod Drive

O 480 VAC REACTOR 480 VAC RE ACTOR ,

t.10V BO ARO 1 A MOVBOAnD1B HPs A tJOnMAL SUPPLY llPS 8 NORMAL SUPPLY f.t f.9 I Note: All relays and solenoids G

G shown energized position (non-scram per.it.lon)

. RPS Bus *A* 12e VAC 120 V AC flPS Bus *B*

thnual Scram Relays G . . . .. . -. .

D Subchannel One Relays e C ?!

3[II [ )j Channel B Subchannel Two Relays Channel A G[g 31 e ]g ,_G4 p%

e -

.ll

,?;

46 * +- e d-

.e g -- - - 1 CR0s s c ~ e ~

46 (d

j -*

• CRDs r- - - . . - - - - - -

to l l L .: L ^.. 4" l- 4- 4- -- - + lAAa,r e W. I 2 3 4 4 3 2 1 vgu,

m.  : -, $CR AM 50LENOID VALVES l.

Q Scram l .

Scram '

3 Groups . n ,; 4 l Groups "2 "

g Scram R

{

Discharge 4 l>'~ -

- - - - - 1 In;ide dastied line V0l"*"

O  ! 6.'"O's in-6 -

l'c"c^{',ut Aion typical of all 185 itCus c

- .oustcr r scnau l 1 I -

WATEn -

5' .(. .- e. iNtcr

/. I I

E., u- ---- - - >.

Ei .

RCS Water

e l' q

00 East West  ;

i 59 2 1 3 4 2 1 3 r

35 3 4 2 1 3 4 2 1 3 *.  ;

51 1 3 4 2 1 3 4 2 1 3 4 47 1 3 4 2 1 3 4 2 1 3 4 2 1 43 2 1 3 4 2 1 3 4 2 1 3 4 2 1 3

?

39 4 2 1 3 4 2 1 3 4 2 1 3 4 2 1  !

35 4 2 1 3 4I 2 1 3 4 2 1 3 4 2 1 31 2 1 3 4 2 1 3 4 2 3 4 2 1 1 3 27 2 1 3 4 2 1 3 4 2' 3 4 2 1 1 3 23 4 2 1 3 4 2 3 4 2 4 1 1 3 2 1  ;

19 4 2 1 3 4 2 1 3 4 2 1 3 4 2 1  !

15 1 3 4 2 1 3 4 2 1 3 4 2 1 11 3 4 2 1 3 4 2 1 3 4 2  ;

07 3 4 2 1 3 4 2 1 3 03 4 2 1 3 4 2 1 l

ft 02 06 10 14 18 22 26 30 34 38 42 46 50 54 58 i l

1800 Figure 3-3 Control Rod Scram Group Assignment e --

535 A I 5 35m

" c

@ n O .,, **~'

= = --- ,Cava*

c o. . .. 53.

A

, g.g a*-

I -

$ nn OC f570-FS P51G DC '

5 338 e g.. y g

=

ci1~ '*'"'

5r#- 37A

]h -

.nside dashed line -

typical of all 185 itcus

,,,,,,,,,i 3

,a, , 5 3,,

= = > , ,, I

~, NOIE ' SCa Aed INL & I AND OUT L E I l

.506 AleOh vatvt547 Ita ANO F 1989 AC RPSS AC f Ang Aan TO Cto$t WHett SCRau MPSA Ot1CHA AGE VOL JME btN1 ANO i SCHAM . CH A6N W ALVI s F 3) A,8, Cl n

50tirdOtDl""

vatve5

] - -

ARE Aan IO OPEN

- C A.G.No JSDV soy W *^'t a I _ _ _ _viat

• 1>c vaat

' a=

. . .'.l.' T ',T.**,' ;7,/~,. . ,

2 l I

.a , , , , ,

i w,

_ _ o .. -

Scram Os5 CHARGE V(st,.JtAf

, , rotJ'l= < - .o-

' 5 JSA o .=.acess o . =. ncsa n o y _

r T' "-

1 35 o. E .eCat.o .=. mEst o

,O SCRAM sc n ane

~~

' . '[ f 3, A --

OUIL , , qg $ 70A ( o. .md acatio . %. .ca2 0

. vi O ,2 HO2 Ca0 F ROM OTHE A v0tuus as 5 ,o. I o ....cino . .. .c.u . i 1 SCnAM SC A A4 OUIL E T mos 5 36 c. . =. a 2.a. o 0. .me acers o y sj .Nati g vaLVil 8'aC" 5 3sa e =a nces.o c. .=e ncea.o

< l 7 t* 5 m, A l . .. .s.u o o. . .. .s.u o y l

~

j [ s * "a

__ 5 3, A 1 . . .c.n o .. . .. ...n o

$ Q Ql:-h e l t v Sv. . -. 6.u s o. ... .c.n o

~

g "4 ~ ' c a a" = a' * -

2;-

, l C An'c.No , h, . ,,, , , . . i mo,.. ..

s j l_______ _ _s. SIV Drain ' ' ' ' I " "

• m ,

f 3FA _ o,e = caoss o O f 13C OP. N CLOSED )

Q O

h Water .

we.a = soa . =

.c io. ie n s. o.o es ",*

Pon t aho goe .4oo 1. Oo 6.ris Poa e mesa s. esc.e

' fobe t so=3 0 I

i I I i

- LS -

quawa6uvaay uteac awntoA mcJ35 9-t aan5y

- m

. A

\

n =

ra w

HCU -

V ~

HCU - - - -

xm

.e c.

lF n

v ,

e-m -

HCU V -

5 HCU

\ m -

4 x F'

& \

l I_

o. (c en

- D z

\ wm nn na :r o 3 & QQ l

, Q- o 2an,8 .'

4 M Level Switches -

= <u, on

.o i

C. Wy2

,=.

= _ = ..

. e w .A L

EE ~9 M cu w.

.n c- = . m, e o w

,c m " .

4C 'N' C

v.

==

4* 9

=

!=

i i g- H < -- Other 1

Drains .

n x

=

HCU Discharge ,

HCU .

v Lines /

/ n HCU u a e, L

< w C, e--

=== q

, E HOU

[ -

h g-

'.[.

c

.n- -

n o- HCU m w a es -

wC 7

oo I

=% " b C. 1

[r"1 -- . .. .

j

'ii UNICIAL I!SE ONLY a

PdEi11111i4MY DRAF AUG 8 $30 M

1

1. p#b sg

&f h RE'/IEW 0F RECENT MALFUdCTIONS OF SWR SCRA'i SYSTEMS g,4 '(6 y'6 6 Ao(f 5

g, Db K0\

w Ac" AN INTERIA REPORT TO THE ADVISORY C0dMITTEE ON REACTOR SAFEGUARDS by A TASK FORCE OF ACRS STAFF AND FELLOWS n

ps " -

0FF CIAL USE OLY g

s f oesa

1. w#'o. UNITED STATES

'k

s. ,

NUCLEAR REGULATORY COMMISSION 3 [' 5 t ADVISORY COMMITTEE ON REACTOR SAFEGUARDS

[ WASHINGTON, D. C. 20555

%,.h*'...*%[/

v A'J6 S 1930 i

MEMORA1DUM FOR:

Dr. Miltin S. Plesset, Chairman Subcommittee on Fluid Dynamics FROM: Members of ACRS Task Force A REVIEW OF RECEi4T MALFU;1CTIO 45 0F BWR SCRAM

SUBJECT:

SYSTEM The Task Force has prepared the attached draft report. We look forward to discussing the report with the Subcommittee during the August 19-20 meeting.

dk p' John Stampelos , Okap.1, and 6 -

\J Sec. 3.4 ONW .N M Garry Youhg, Chap. 2 thd 4 Sec. 3.2, 3.5 k A ttA <L % lc M J Dorothy Zugor Apap. 5 Tec. 3.3 00 J P0

a 1 TACLE OF C0iiTENTS Pace Chapter 1-1

1. IN TRODUCT I ON AND

SUMMARY

2-1

2. GE NERIC DESCR IPT ION OF BWR SCR AM SYS TEM . . . . . . . . . . . . . . . . . .

2-1 2.1 Introduction .......................................

Scram Mechanical and Hydraulic Subsystems .. . . . . .. . . 2-1

2. 2 Scram Electrical and Pneumatic Subsystems .......... 2-a

2. 3 3-1 J. DESCR IPT ION OF RECENT CWR SCR AM MALFUNCTIONS . . . . . . . . . . . . .

3-1 [

3.1 Introduction .......................................

B rcwn s Fe r ry -3 E ve nt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-1

3. 2 3-l' 3.2.1 Sequence of Events ..........................

3.2.2 Browns Ferry Equi pment Layout . . . . . . . . . . . .. . . 3-3

3. 3 B ru n s d i c k - 1 E ve nt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3-5 3.3.1 Descriptian of Event ........................ 3-5 3.3.2 Brunswick Equipment Layout .................. 3-6 3.3.3 Discussion.................................. 3-7

3. 4 Ha t c h - 1 E ve n t . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3-8

3. 5 Ot her Recent E vents at BWRs . . . . . . . . . . . . . . . . . . . . . . . . 3-11 3.5.1 Dresden~3 and Browns Ferry-1 Events . . . . .. . .. 3-11 3.5.2 Mi l l s to ne-1 E ve nt . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-12 j

3.5.3 Duane Arnold Event .......................... 3-12 3.6.4 Peach Bottom-2 and -3 Event ................. 3-13 l

4-1

4. NRC REPORTS UN SWR SCR AM SYSTEM MALFUNCTIONS . . . . .. . .. .. . .

5-1

5. COMP ARISON TO W ASH-1400 PREDICTIONS . . . . . . . . . . . . . . . . . . . . . .

6-1

b. ACRS QUESTIONS ...........................................

7-1

7. B l d L I OG R AP H Y . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

t-

e i e

TABLE OF C0i;TE:;TS (con't)

APPE:CIX A f.emoranden to d.F. Ross from P.S. Check , re DSI B'!R SCRA4 DISCHARGE VOLU:C EFFCP.T APPE 0!X B Michelson's Report P

i l

1-1

1. 1:4TR0JUCT10:4 & SU:CWY This report presents the preliminary findings, recommendations, and conclusions of a review of the recent malfunctions of tha scram systems at Browns Ferry 3, drunswick l', Hatch 1, and other SWRs. This report is necessarily preliminary since specific plant information is forthcoming as are the results of tests and analyses of the BWR scram system by the NRC and other organizations.

chapter 2 of this report is a generic descrircion of the BWR scram system.

Chapter 3 describes the incidents at the reactors. The presumed cause of the failure of approximately half of the Browns Ferry control rods to scram was due te insufficient available . volume in the scram discharge header due to poor design -

of the scra.n discharge header vent, drain, and instrument system. Testing has shown that the scram discharge header can drain significantly slower than the scram discharge instrument volume. The presence of water in the scram dis-charge header due to normal leakage into the header from valves or water from previous scrams may not be sensed in the scram discharge instrument volume.

The presumed cause of the failure of the scram discharge instrument volume scram float switches (crushed floats) and damaged pipe supports at Brunswick 1 is water hammer (both flashing of steam and momentum energy transfer of water to structures) in the scram discharge instrument volume. The instrument volume and its ball float switches were not designed for the turbulent flow of water through the system since the instrument volume drain valve was designed to close during scram conditions. A faulty solenoid prevented the instrument volume drain line valve from closing as designed. The valve closed during the incident while water was flowing through the scram discharge instrument volume and drain line.

e i

• l 4

g,3 l

Ine incident 3t Hatch 1, inoperative scram discharge instrument hign-level

• The LER descriuing the inciden* I-suitches, was caused by bent ' float rods.

! states that the rods were probably bent prior to installation or during in-g stallation. The incident at Hatch I will be evaluated on receipt of further The LER is the sole piece of infor. nation presently available. ,

information.

Chapter 4 presents a critique of the recommendations and findings of the AE00's We (Carlyle Michelson's group) analysis of the malfunction at Browns Ferry. .

agree ia principle witn the four recosaendations of the report and have included the report as an attachment. t Unapter 6 presents a suramary of the analysis presented in WASH-1400 of the SWR The scram system. The system nodeled is siinilar to that at Browns Ferry. .

failure to drain the screra discharge header was recognized as an important 1

The WASH-1400. report does not cover the contributor for failure to scram.

possiblity that a blocked vent or that damaged level switches could permit Considering the undetected water hold-up in the scrara discharge volume.

i recent failures, it appears that the probability that these events will occur is considerably higher.

j

! Chapter b lists questions that the ACRS asked at the last full committee Answers are provided where possible.

meeting on the incident at Browns Ferry.

' This Task Force has requested as-built drawings of the scram system, float switch technical manuals, and a list of the required specifications for the scram system as suggested by General Electric.

Some of this information has been received. The portion of the scram system downstream of the scram

. discharge header is supplied by the architect engineer. t i

e m . - ~

w .- -

s 1 88 e'

1-3 l

Our review has shocen tnat there are significant differences in scram discharge

systems. The scram discharge instrument volume has been located contiguous to

, the scram discharge header for more direct _ indication of available volume in the scram discharge header in newer SWRs. The evolution of this system, the utility-architect engineer-vendor interface, General Electric's requirements, and the safety review process for this system nest important to safety will be reviewed by this Task Force in subsequent reports.

• Paul S. Check, Assistant Director for Systems of the NRC Staff, has estab-lished a .nulti-discipline team to review proposed modifications to.the design (especially cn-line monitoring provisions) of the scram discharge' system. A description of the scope of his review is Attachment A.

4 i

l l

l n

I i

i i

I 1

l - --- -- - - - . .--. -. .

f l 1

l 1

.  ?,

2-1

2. GENERIC DESCRIPTION OF BWR SCR/fl SYSTEM 2.1 Introduction The purpose of the BWR scram systen is to quickly insert the control rods from the fully withdrawn position to the fully inserted position in approxi-mately 5 seconds or less. The design to accomplish this task includes mechanical, hydraulic, ' electrical,- and pneumatic subsystems. Each of these To i systems will be described in a general, generic way in this chapter. ,

provide better unaarstanding of the syste.a operation, actual plant parameters I and designs will be used as examples.

2.2 Scram Mechanical and Hydraulic Subsyste:..s I Tne control rod drive nydraulic system consists of three basic subsystems' as follows:

- Su pply Pressure Control Scram The supply and pressure conrol subsystems will not be described in detail since they are' primarily involved with normal operation and cooling of the control rod drive system. However, the interaction of these subsystems with

~the scram subsystea,will be addressed.-

The scram hydraulic subsystem is shown scheuetically in Figure 2.1. The subsystem includes the folicwing major components:

I Charging 1.'ater Header (Including Scram Accumulator)

Scram Valves (Inlet and Outlet)

- Scraa Discharge Volume, Level Switches, and Valves

s .

2-2

^

The scram subsystem utilizes these components to drive the control rods i

}

To produce the rapid insertion, the subsystem rapidly up into the core. I t

pr ovides a hign differential pressure across the control rod drive piston

'(See Figare 2.2). Tne hign differential pressure is obtained by opening a patn for high pressure charging water to the underside of the control rod drive piston while venting the top of the piston to approximately atmospheric pressure. -

Tne design goals for the scram mechanical and hydraulic subsystems are:

- To provide a source of high pressure water to the contrcl rod drive piston. The normal source is the scram accumulator and The accumulatur the backup source is the reactor coolant system.

water pressure is 1400 to 1500 psig.

.-3-5

- To provide a rapid insertion of the control rods (i.e. ,

seconds) by quickly opening the scram inlet and outlet valves.

The scram valves are spring loaded to open quickly (i.e.,

0.1 seconds) after the pneumatic pressure is vented fran d

the valve operators.

- To provide a scram discharge volume (SDV) which is large enough to allow a full scram and yet small enough to prevent exces-sive uncontrolled leakage from the control rod drive system following a scram. The design volume for the SDV is 3.3 This is large enough to gallons per control rod- drive.

accept -the displaced water from a scrammed control rod (i.e. ,

1 i

. i 2-3

~.75 gallons per rod) and yet liinit the subsequent control rod seal and bypass leakage (i.e. ,~ 3 to 6 gpm per rod).

-The purpose of limiting the seal and bypass leakage is to limit and contain the reactor vessel water loss during a scram. R2 ad on the design size of the SDV, it would fill ,

and pressurize to stop reactor water loss in less than -one mi nute. .

- To provice alarms and auto.natic scrams prior to losing the ,

capability for a full scram. Since the SDV must be able to  :

i accept the discharged wacer from a scrammed control rod drive for a successful scraa, the SDV must be instrumented and alarmed to detect water accumulation. This is done by means of a scram discharge instrument volume (SDIV) and water level instruments. ,

The operation of the scram ;nechanical and hydraulic subsystems during a .

scram involves many steps which all occur within approximately 5 seconds.

Following receipt of the scram electrical signal which actuates the pneumatic subsystem (See Section 2.3), the scram inlet and outlet valves open while the B bOV vent and drain valves close as shown in Figure 2.3.

The large differential pressure created by opening the scram valves (initially about 1400 psig) produces a high initial acceleration of the control rod and As provides a large margin of force to overcoine any anticipated friction.

the control rod insertion progresses, the charging water header and ;craa

1 4

. i e

2-4 accumulator pressure falls below the reactor coolant system pressura of 1000 psig. At this point, the ball check valve in each control rod drive housing (See Figure 2.2) allows the reactor coolant system pressure to complete the scram.

Th2 initial insertion of the control rods causes approximately 0.75 gallons of water to be discharged from each rod to the SDV. The subsequent seal and bypass leakage then causes the SDV to fill and pressurize to reactor coolant .

syster, pressure in less than one minute unless the plant operatcr quickly resets the scram signal. Once the SDV has been pressurized or tne scra.:.

valves have been closed by resetting the scram signal, nonnal cooling water flow to the control rod drives is established. .

The SDV/SOIV nest be drained and the water level instrument signals (i.e.,

scram and rod block) cleared before the control rods can be withdrawn from the core. This is accomolished by opening the SDV vent and drain valves after closing the scram outlet valves. During normal plant operation the 1 50V vent and drain valves are open to prevent buildup of water in the SDV from leaking scram outlet valves. The SDIV water level instruments will initiate an automatic scram if water level reaches a predetennined point.

- Additionally, a rod withdrawal inhibit (rod block) and level alarm are provided by the SDIV instruments as shown in Figure 2.4.

The SDV/SDIV layout varies from plant-to-plant. Figure 2.4 and 3.4 show the general layout for Browns Ferry which includes two SDV's which separately 4

t 9

I.

2-5 drain to a single SDIV. Other BWP,'s have two SDV's and two SDIV's located near the SD/'s. Newer designs have the SDIV as an integral part of the SDV's. This is the design for Brunswick and the layout is shown in Figure I

3.10.

2.3 Scra Electrical and Pneumatic Subsystems The control rod scram electrical system is part of the Reactor Protection System (RPS), and it controls the scram pneumatic system which in turn ,

controls the scra.a hydraulic system. A typical RPS scra.n (or trip) logic is shown in Figure 2.5 and a typical electrical and pneumatic arrangement is shown in Fi gere 2.6.

From Figure 2.6, it can be seen that when the scram inlet ad outlet valves

~

open, a driving force to scram the control rods exists. ' These valves are spring-to-open and air-to-close operated. Therefore, the air pressure must bc removed to cause a scra.n.

Af ter initiation of a scram signal the SDV isolates to prevent excessive loss of reactor coolant. This is also accomplished by removing air pressure from the SDV vent and drain valves and allowing spring pressure to close the yalves.

The relays, solenoids, and valves in Figure 2.6 are shown in their normal power operation condition. If any one RPS "A" subchannel trips (and no "B"),

the "A" solenoid valves for every control rod and the "A" solenoid valve for the SDV valves will de-energize. However, the "B" solenoid valves will

a f 2-6 maintain air pressure on the scram valves to hold them shut and or, tne SDV vent and drain valves holding them open. Similarly, if an RPS "B" subchannel trips (and no "A"), only the "B" solenoid valves will de-energize. These two conditions are referred to as half scrams.

If, however, both RPS channels have a subchannel trip present, both "A" and "6" scram scienoids on each control rod will de-energize as well as the SDV

~

solenoid valves, resulting in loss of air pressure to the scram valves and tue SDV vent and drain valves. This allows spring force to open the former and close tne latter valves, scramming the control rods. Tne RPS logic is thus one out of two taken twice.

In the event that any individual solenoid valve fails to vent, the entire scra.: air supply system is independently. vented by the backup scram solenoid valves (See Figure 2.6). The backup scram solenoids only operate'when both 1

RPS " A" and "B" subchannels have trip si gnals.

An RPS mode switch is located in the control room for selecting the necessary scram functions for various plant conditions. The mode switch positions are.

SHUTDOWN, REFUEL, STARTUP, and RUN. For normal power operation the mode switch is in RUN. When the mode switch is placed in SHUTDUW", the reactor is automatically scrammed with a 10 second time reset. The reset causes the scram SHUT 00WN initiating signal to clear. All scram signals (i.e., manual or automatic) must be manually reset by the operator. They can only be reset after the initiating signal has cleared and a 1C second *.i e delay relay has cleared.

W w v i

B coeJ rrat f A'# D -

f Arg RC4c70R Soffe/

hhMEOLS

• V f. 55 t'L -

T P % P IIII' AIR LINE g g ~~l E AA *' '

-a l

I~ d3 myos.

g-  !. , T/ R A#4 pgg gggg I ij

! DRIVE M OPEtJ VALV6_ Frsrod gp g '

l 0 H Ct.osED VALVE l l

,-ctosrD DRWt. . e s;' Hos).s srt G i l SDV -)

r tT I vtyr

-1 cnECK VALVil o F~

SCR AM Of 5CH4RGt-

^Q rQ^I

= l voi.umc (sov) l C.W7Eot kop  !

v>,vc t _

l ~ ~ ~ " ~ ~ ~ '

L O' ' l ,

T' l valve V A LVC i

i (

I

' I s l 4 T

A p V s

1 1

g 40 y'

y l[unt(A l

cnARGING gcgnu lIrnsno.

y

} DRIVE oanoca g I

RccomtJLATOR l

W' sov FROM WArer< U DRAld conocusArE - __-~

sron M E rose 1 L _ _. _.-_ __

TYPEC A L FOR tan R IB6 CONTROL son DReves flWR SCRAM fiYDRAULIC SYSTEM (fl0RMAL VALVE Lit 1EUP, PRE-SCRAM)

FIGURE 2.1 3

4

CONTROL ROD

+

07 BOTTOY OF REACTOR VESSEL

\ (" # "

r' a

. \ ) Q

.l >-

I s

- m

-L.I m t

m -

E W r Mf'[1 l

HOUSING -  : U' - FIXED PISTON dJ ~ .

5 (STOP PISTON; Qu t

7  :

',7; E

• I  ?, u ;

hy y PISTON TUBE y  :

13 7

't. /

a ] , try). lNDEX TUBE f

l g[! .. /

s m

/.

r'  ;

t Q ,1  !

c -

! e j ,ry ,

(T  !

(

ORIVE PISTON

- i 7

~~

h I- 73 WITHDRAW LINE INSERT LINE O + N3 j i f.-

A [' Low HIGH PRESS.

') J g jy i..

L PRESS.

$:, - , e I

L- ^!h  ; 4 SCRAM wgrm IMt.tT 4M....h

'ri nouj w+d i! t BALL CHECK Lt l l

'g  % StRAK OUTLET T)]

S' #! -

VALVE RE ACTOR PRESSURE % a M: WbM f(AQ)

..g

• y (Pal -

1 FIGURE 2.2 BWR CONTROL ROD DRIVE MECHANISM (SIMPLIFIED SKETCH)

8v M Lu) L cou rk ot

• w m r, '**

RCALToll AnR Ana L NE < ygsseL orftY Q rs

<] oecd VALvc _ _ _ _ _ Z f h 3mos.

A 6TM05. l N 0'#5G VA'V#

ectostD I pm Pla TOA

%Q b

'W l We g ~ (stor Aig ~ j.

l g f ENLL Y y cpnosns > l w p l NERDER 6 corte T1 vent yt l

+

l -

es

,,ggsc, N c a r. vatve a I 1

a l fft X+-

0 F---

(ogypot Roo l l SCRAM DISCHARGE pein. t g votomt (sov) cootevG W O ICA l

L r VALVE

$IN VALVE l

- - - -h-y 1

i -

% I I

ns I o I l, Y l

g go #2 i/

i LivEL V

j 3.

CnARGING HERDER g } lIH3W e g g.4p l FROM DRtVE RCComvLATOR CoMDENSATE sroA6GE WA TER PUMP l

l l

NW spy l

rank t-------- - - - - -

p paarta TYF1C 4 L FOR 185 CONTROL ROD DRIVC,$

FIGURE 2.3 BWR SCRAM HYDRAULIC SYSTEM (SCRAMMED VALVE LINEUP)

I 34." L me M

S C

To 4. ,2" Lsac (' R ~ 2' 10 "

SDv A M

D

" AUTO. SCRAM LEVEL s

(.

A ~ 50 GALLONS 12" A

R G

E I

N ~ 4' *7 "

S T

R U

M E

N LEVEL T

V INST.g[) " rod BLOCK LEVEL 0 Nj a ~ 25 GALLONS L

U M

E

~ 3t5 "

S D

I Y [, "

^

ALARM LENE.L

~ 3 G ALLONS A/

~ L' O ODY 2 uut DRAIN ,

r FIGURE 2.4 BRO'lNS FERRY SCRAM OISCHARGE INSTRUMENT VOLUME i

N .

TESISWitCH

@7 t' NtUIRON 148Hueht 5 W ,,

us3mggonggG M80 JEP5flR$I MAI E# ge sNst esDME NI g Hips s on & e nsttsasG (deres !watiW. t tran wErn9w t!)

• • " 's TOMBLht C V {5tAGE" *""" ~~%

e A$i CtosuREs t~- T "sv'>'A55 ~~'

Q CLO51D #OR RE V1OCK WOT . MODE SCR AM DeMH 8vP A55 ANO MODE

" tmWyN $WalCH VOt &4 wet HR <- Swit CH IN E.D OR MGOE SOPt N IN RtiUEL

.AHIO Shui-

-~

St i DOWNI CONOf NSER

&(wwvAC T . CLO5E D FOR MODE Mit s$ot V O .

me O $** $ I SWIICH NOT : NRUN

T T AND < 10% F56G

1. 1 ppg ggggcua,3gg g g D WPHtSS 2 P5sG g gyp.

tsgu," A" y ,,g g, g j*

,g A550GNMt NIS A7 SS 57 a Al

& R, Wu A I E H APHM *A" M5tv s Aae Cao AaC eao

1. LEVEL 60 gy pT[ )T oNSCL.

V _, STOP 3b4 b3 2a4

• - 1 ' HI HI VALVES ib2

1. R. HI PHl.% gyp O M ANU AL

• ~ TSNOP P. g SCHAM CONIHOL , 3 APMM

• A" vAtVES MSL He pt AO APft5fs A.E C.E S.F B. D

-ir---- BR M's A.E C.G sF DH

~ ~

l

-* 1 Hi HB A M $ pups HM

, E- I . ..~- L APftM *E" I 8#

T .Y DNSCL.

SVP l ~ '_ H4 H8 INOP

(,,,, APRM "E" KA9 EA11 K p.3 M43

~

"A3 "A9~ 5 " AI8 KAl RESET r RESET r TT T I

o l RE ACTOR Prof E CitON SYSTEM RE ACIOR PROF ECTION SYSTEM TRIP LOGIC A3 THer EOGIC Al ITVPICAL Of A21 FIGURE 2.5 RPS CHANNEL **A** TRIP LOGIC (TYPICAL OF CHANNEL **B")

m 4

e RPS S NORMAL SUPPLY

• RPS A NORM AL SUPPLY 440 V AC ~R~ a ut nGE NCY HUS

• A~ f MF Rt t PeC* Bus M M UNtf PREF ERRED Mg IR ANM ORME R M

M I I F L YVtHE f L FLYWHfftI 3 G G

INif RUME N T RPS*S~

INTERLOCK

' AIR RPS ~~~~~

BNitpLOCK

_"A-*lh H % _ VOC e r f _ 7 _

SM CLOSf D VWH(N "$

SCR AM BUS " A" 170 V AC iSCR AM l RPSCHANNil I

' VALVf1 T RIPS M i r ^ ', ^ '

$ .S ,

~~ #I OPf M WHE N N O ,O,

~~ ~~

T OPE N WHEN 81 82 83 RP5 " A' SuasCHANNELS ag y 33 + ay ,$UBCHANNfL T ,,,,,

-" SUBCHANNEL gg,pg 82((

R3 * ~~ --

SunCHANNits

~~ --

Q A11 - --,,- ,,- ,-, -

SCR AM OISCH.

,$OL ENOID V ALVES l l

r

( , :'. : '," ~3 g (- VfNY l I g

p$ _l .1_ ^^ ,=:=,r,r, c [ .p 9% . _ L _ _ _ _ _ -l g

i - - .

L___ A SCRAM SOtf MOIO VALVE S VfNT .11VPICAL ONI PAIR Pt R

, f ', f ', f ' , '

VEN1 ' CONTROL ROOD FROM OTHE R To nRg ot NOTE : ALL REL AYS AND torf M A 3 lNG CONogy goN)

H UST 5 I T YP. -

29 C , ,

', f

- v,

$ CHAM - -

M 3CR AM OISCHARGE WAffR ~_ $ CRAM v VOLUME ACCUMUL A10R OUTLEI Q SCHAM WAf tR

, INSTRUME NT INtET VOLUME DR AIN g4 10 RRE OT FIGURE 2.G RPS SIMPLIFIED ARRANGEMENT 9

.~ _ . , _ _ .-_ _ _ _ _ .__ _ ___ _ __ . _ . _ _ _

a . o I 3-1 1

3. LESCRIPTION OF RECENT ddR SCRA4 MALFJHCTIONS i 3.1 Introduction Recent..nalfuncticns in the scram systems at Browns Ferry, Brunswick, Hatch and other BUR's have occurred. These events and the equipment layout will be discussed in this chaptar.

l 3.2 3rowns Ferry-3 Event . ,

aost of the infor,:ation concerning this event is taken from a report by the MRC Office for Analysis and Evaluation of Operational Data (AEOD). Additional l

information, figures, and sketches are from General Electric, !!E, and NRR sources. Browns Ferry-3 is a BWR-4 owned by TVA. It began commercial opera-tion in March, 1977. The Architect Engineer was also TVA.

l . ,

3.2.1 Sequence of Events On June 23, 1960, power was being reduced by the control room operator at the Browns Ferry-3 nuclear reactor in preparation for a scheduled shutdown for feedwater maintenance. At 1:30 a.m. , the operator initiated a manual reactor i scram from about 30" pouar (i.e., 390 M4e). The control room instrumentation 1

verified that all-scrau valves were open. However, the control rod position indicating instruments showed that 76 rods were not fully inserted. Figure 3.1 shows tne rod positions after the manual scrau. All but one of the west side rods (out of 92 total) were fully inserted and the one rod was essential-ly full in at position "U2". Un the east side, 76 rods (out of 9J total)

I were not insertec fully ranging from "46" to "U2" withdrawn-("48" corresponds to fully withdrawn). Later evaluation of the neutron instrumentation indi-cates that the east side of the core was at less than 2% power. Since the i

. - - , , r .-m, ,,--., .- -w, m , - - , --e, , yr, - - - - - - - - - - - - - . 3-~.r- - - --

?

3-2 j i

4 source range and intermediate range neutron instrumentation is not inserted into the core until af ter a scram, no precise indication of criticality at less that 2% power was recorded. Tne RPS mode switch was placed in SHOT 004N l after the scram.  !

Following the scram, the SDI'/ began to fill and initiated a high water level s' scra:n signal somIIhat sooner than expected at about 19 seconds. The Solv level scram was subsequently bypassed by the operator (as allowed in SHOT 00WN  ;

saade), to panait resetting tne RPS logic. }

l A second manual scram was initiated by the operator approximately six minutes af ter the first nunual scram. Rod positions after the second scram are shown in Figure 3.2. All west side rods were fully inserted and 33 of the 93 east i

side rods were fully inserted. This lef t' 60. east side rods not inserted fully ranging from "42" to "02" withdrawn.

The second manual scram was reset and a third manual scram initiated approxi-j mately 2 minutes after the second scrau. Rod positions after the third scra:a are snown in Figure 3.3. Of the 93 east side rods, 47 were fully inserted and 46 were inserted from "40" to "02".

I The third scram was reset and the SDIV level scram bypass switch was changed i

f rom bypass to normal approxiinately 6 minutes after the third scram. This action initiated a fourth, automatic scram due to the SDIV level trip condi-tion. At this time all rods on tha east side were fully inserted. The total elapsed time between the initial scram and the final insertien of all rods

! - was approximately 14 minutes. P 8

--,,c ,, - w , . - - - n-w,

/

3-3 3.2.2 Browns Ferry Eauipment Layout i

The apparent reason for the failure to achieve a full scram on the east side i i

The design of the reactor core was water accumulation in the east side SDV.

of the SDV was intended to pre"ent inadvertent accumulation of water and to Howaver, the layout of the SDV alarm such a condition in the control room.

and piping will not necessarily assure that the design objective is met.

Figure 3.4 snuws an isometric view and Figure 3.5 shows an elevation view of the Browns Ferry equipa.ent laycut. Notice in Figure 3.4 that the west side SUV drains tnrough approximately 20' of 2" piping to tne SDIV while the east side S0/ drains througn approximately 150' of 2" piping. Tnis creates consideraoly nigher flow resistance for gravity draining of the east side SD/. Additionally, notice that in Figure 3.5 the SDV vent lines are not open to atmospnere but are piped into the Reactor Building Equipment Drain Sump.

Inis creates the potential for inadequate venting if the equipment drain header is filled with water.

Drain tests for the Browns Ferry SDV were conducted following the June 23 evant. The results of these tests are as follows:

East Header. With the west SDV drained, the east SDV was allowed to drain This into the SDIV sith the east SDV vent valve and SDIV drain valve closed.

The measured drain rate condition si.nalates a blocked east SUV vent line.

for the east 50/ into the SDIV was 0.o gpm (See Figure 3.0).

West Header. With the east SDV drained, the west 50V was allowed to drain This into the 50lV with the west SDV vent valve and SDIV drain valve closed.

i . .

3-4 condition simulates a blocked west SDV vent line. Tne measured drain rate 4

for the west SDV into the SDIV was 3.2 gpm (See Figure 3.7).

j East and West Header.. With both SJV vent valves closed and the SDIV drain f valvo closed, the 50V's were allowed to drain into the SDIV. This condition After an initial water surge, the simulates blocked vents on both SDV's.

!- coa.bined drain rates of the two SDV's into the LSDIV was 0.6 gpm (See Figure 3.8).

flor.nal Drain Test. Tne SDV's ana SDIV were filled with water and then the  !

i

! <ent and drain valves ware opened. The west SDV drained in 91/2 minutes '

unile the east SUV drained in 25 minutes. The 50 gallen, 25 gallon and 3

gallon level switches in the SDIV cleared at about 91/2,101/4, and 11 1/2 minutes respectively. Based on the volu::es associated with the SDV's, the calculated average drain rate for the east 50V was 11.6 gpm and for the west SDV was 34 gpm. The calculated drain rate for the SDIV was 36 gpm (See t

j Figure 3.9).

2 Based on the above tests it was concluded that a blocked vent line on either or both SDV's could cause water to accumulate in the SDV's while not accumu-I lating in the S0IV because of its hign drain rate. Tne east side SDV is mucn [

slower draining that the west side SDV due to the long pipe connecting to the

~SDIV. Therefore, the east side SDV would be n. ore likely to accumulate water i '

i during normal operation (fro.n such sources as scram discharge valve seat leakage) than the west side SUV. ,

1 i

P

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

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

O 4

.. . r 3-5 t

The utility and HRC approved interin: solution to the Brawns Ferry layout pro-blem for the 50V is to:

l

- Provida a direct vent path to containment atmospnere when the SDV vent valves are open.

- Provide direct measurement of water accumulation in the SUV's  !

by use of ultrasonic level instruments on the SUV's.

' A fin 31 solution to the layout proble:a is being discussed by the NRC Staff and ,

i the Licensecs.

! 3.3 Brunswick-1 Event drunswick is a 621 W.e BWR-4 owned by Carolina Power & Light. It began com-1 mercial operation in March,1977. The Architect Engineer was United Engineers l

s and Constructors Inc. (UESC).

1 i

3.3.1 Description of Event Following a scram fron full power on October 19, 1979, damaged pipe supports were found along the drain line from the Scram Discharge Volume (See Figure 3.10) in Unit #1. Altnough the damage was greater on the south side, both-

.the north and south pipe supports were damaged. Supports were damaged along .

the drain line from the botto.a of the scram discharge volume to the drain valve.  ;

l The damage is believed to be the result of a water hamner. The cause of the event has been traced back to a faulty solenoid controlling the air supply to 4

~both the vent-and the drain valves. This solenoid was causing the air to be ,

bled off slowly when the scram signal was received. As a result of this, l

l l these valves were closing in 4-5 minutes rather than their normal 30 seconds.

I i ThisLis believed to have caused tne water hammer.

i 9

. o e .

J-6

~

Af ter the 15E Regional Of fice was notified of the event, they suggested The licensee reportedly that the SDV switches be inspected for damage.

Since a new solenoid was on order, the plant resumed checked the switches. E operation with the vent and drain valves closed. Periodic draining was to take place initially every hour. Later, it was decided that 501 would be drained when the 50V hign level alarm was received.

On 14cvember 14, 1979, a reactor scram occurred due to high water level in ,

the S3V (See Figure 3.11). Investigations showed that both the SDv high-level alarca and tne rod-block-withdrawal level alarms were inoperable be-The crushed floats ware located cause of bent stems and crushed float-balls.

on the nortn side, the side with the least damage to the piping supports.

j It was concluded that the water hammer event of October 19th damaged these switches. There is a possibility, however, that they may have been damaged j

I i from some other event prior to October 19th.

3.3.2 Brunswick Equipment Layout Figure 3.10 shows the normal valve configuration when the plant is operating.

1 Note that the rod block withdrawal switch and the high level alar.n switch are i connected to the drain line. Figure 3.12 shows the valve lineup wnen the i

reactor has scraamed. The arrows indicate wnere the pipe supports were damaged. There are valves on the system (shown by cross-hatched valves) to isolate the instrument volume from the SDV to test the float switches.

Figure 3.13a shcws a portion of the SDV and the SDIV. Figure 3.13b shows a damaged float.

4 I

~

a i .

3-7 3.3.3 Discussion i

There are a number of explanations for the manner in wnich a slow closing

valve can give rise to a water hammer. First, the delay could have allawed flashing in the drain line and high flow rates throughout the system rather than confining flashing to the 50V and the SDIV. Second, the system could have become filled with water and when the valve finally did close a pressure A

pulse traveling backward could have damaged the piping and the floats.

clear explan3 tion concerning what happened and why it happened is not yet .

available.

Tne float switches at Crunswick were modified in Dece..cer,1974. The floats originally installed in the plant had a self-equalizing hole in the float shaft. This hole would fill with water and tne float would sink. The float switches currently installed in the plant have n'o hole.. The switches are-nydro-testeo to 1625 psi before lea'.ing the factory. However, they are not tested for the dynamic conditions possibly encountered during the water hammer

.I

- event of October 19, 1979. The damage to the floats indicates that they were slammed upward into their housing. It does not appear that tne old-style vented floats would have survived with any less damage. There is no clear explanation as to why the scram switches were not damaged. A recent test

[

done at' Hatch indicates that they are not immune. Their susceptibility may, however, depend upon the system piping. At present, the Brunswick personnel do not plan to replace the float-type level switches witn pressure diaphragm switches.

. t i

1, .

I 3-8 l h fil-

.The switches are tested during plant operation by valving them cut t en The position of the float.

ling tneir volume with a known amount of water.

, t; stems is then co.npared with a calibration curve.

i The water hammer damaged pipe supports for botn the north and south When the specifications for There was no damage to tne vent lines.

lines. i line had not been tne' drain line were checked, it was found that tne dra n ,

The lines were not, therefore, designed to analyzed for dyna:aic support.

The vent line had been analyzed for l

tolerate shock or thermal stresses.

Reactor Controls Inc., who did the first analysis, re-these conditions. d that the analyzed the drain line for seismic and thermal stress and foun i

Brunswick had their Architect-Engineer supports needed to be strengthened. d larger.

' (United Engineers) verify this and the base plates are'l=ft. being ma e United Engineers also did a water haminer study,They andstill they found that a slug of water could hit the 2" piping with a force of 1633 lb.

which hold that the larger pipe supports can take the maximum thermal stress h line. It is not clear would occur for a continuous flow of hot water in t ehammer t case water that the strengthened pipe supports could survive the wors event.

3.4 Hatch I Event _ It started Haten 1 is a 7d6 MWe BWR-4 owned by the Georgia Power Company.

The architect engineers were Bechtel and i

commercial operation in 1975.

The incident under consideration occurred on June 13, Southern Services.

1979 while the reactor was shutdcwn for refueling.

l

. - ~ . . . - .

3-9 Two ' out of four scra:a discharge instru..ent T volume high-level switches were j found inoperable during surveillance testing. The inoperable level switches t were opened for inspection. The float rod, bent on both switches caused the float to oind against the side of the float chamber.

These level switches parform an important function in ensuring scram relia-

.bility. Even though the scrain discharge header, vent, and drain lines are 4

designed to be coupletely empty during nonnal operation, each pneumatic ,. ,

scrar. discnarge valve can leak about 0.1 gpm into tne scram discharge header without causing a high temperature alarm indicative of insufficient control rod seal cooling. (The total leak rate for all scra.a discharge valves at

Browns Ferry was 0.03 gpm according to tests run by General Electric.) This I small leakage is not usually a problem with properly designed and installed vent and crain systens. The scra

n instrument volume high-level switches ,are intended to initiate a scram to shutdown the reactor while sufficient free volume is available to receive the scram discharge.

The LER reported that these switches had been modified prior to the surveil-j lance testing. The modification added a new vent port to the float chamber.

its purpose was to pennit total isolation of the level switch fran the scram 4

discharge header during calibration. After the modification it was found that the float was stuck and would not move. The top of tne float chamber was removed and the float stem was found to be bent. Wear marks on the inside of the float cnamber indicate that the float had been rubbing on the tank for some time.. The switch, however, had never failed to pass surveillance testing.

l

i t

., + , . . - - . . . . - - . - -. .. - . . -

. .. 3-10 .

It is believed that the float rod was originally bent prior to installation or during installation. The bent float rod caused light contact between the float and tank. It is believed that metal particles from the modification performed on the chamber caused the binding whicn stopped the float from moving. Tne rods were straightened, the switches reassembled and tested.

-The vendcr's review did not determine the exact cause. The LER reporting the incident stated that the utility did not plan further action.

~

Tne NRC does not currently have available a written evaluation of this in-cident beyond the LER issued by the utility. Inis Task Force has requested pictures of the float rods (in the correct alignment and bent configurations),

! as-built drawings of the scram system, float switch tecnnical manuals, and a list of the required specifications for the scram system as suggested by ueneral Electric. The portion of the scram system downstream of the scram e

discharge header is supplied by the architect engineer. An assessment of the j

4 proble:n will be attempted af ter the information has been received.

The NRC project manager for Hatch I and 11 has informed us that a similar The preliminary notification incident occurred at Hatch 11 on July 26,193). ,

stated that two out of four scram discharge volume float switches failed to actuate curing testing mandated by NRC as a result of other BUR scram system

, malfunctions. Initial inspection revealed that the scram level floats were l cracked and partially collapsed. This degraded condition allowed them to fill with water 'thus preventing switch activation. Hatch II will remain shutdown until investigations and corrective actions are completed. The Task Force will attempt to assess the problem following the receipt of the infor-mation described above.

a .

3-11

. 5 Other Recent Events at BW45 Tests and inspections of scram systems at SWR's because of the Browns Ferry-3 event have identified some additional problems in the scram system. These events are described in this Section.

3.5.1 Cresden-3 and Brosns Ferry-1 Events As a result of scram tests run to meet the requirements of I&E Bulletin 80-17, Dresden-3 found that the east side SDV was not draining properly following a scri.. The layout for Dresden-3 is similar to that snown in Figure 3.4 for Broans Fe rry. Tha single difference is that the vent lines from the SDV's to the floor drain header have vacuum breakers installed to nelp assure adequate venting.

Investigation of the east side SDV draining problem at Dresden revealed that the east side vent line vacuum breaker was stuck shut. When an operator manually actuated the vacuum breaker, he could hear air rush in. The east SDV then drained properly.

The drain line f ailed to perform its function due to a stuck vacuum breaker and a clogged or obstructed vent line. Subsequent to the Dresden-3 problem, a similar situation was found at Browns Ferry-1. Altnougn Browns Ferry-1 does not have a vacuum breaker, it was found that the SOV was not draining properly due to inadequate venting.

Shortly af ter the Dresden-3 and Browns Ferry-1 proolems were identified, the NRC issued an order to all operating BWRs to provide a direct 50V vent path

3-12 to containment atmosphere independent of vacuum breakers or floor drain headers. This order was given a 48 hour5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br /> implementation schedule. l

3. 5. 2 Millstone-1 Event' i During scram system checkout at Millstone-1,. in response to I&E Bulletin 80-17, it was found that the 10-seconc time delay relay was not properly -installed in tne scram reset circuit. The result was that tne 10-second delay was in-

~

operative and a scram signal could be manually reset in less than 10-seconds. ,

The 10-second relay is described in Section 2.3 of this report. The purpose of tne 10-second relay is to prevent manuai resetting of the scram signal until tne control rods are fully inserted. If an operator '. cts very quickly to reset the scram signal, he could stop the control rods from complet ng i l

thei r insertion. The 10-second time delay would not allow the operator to ll 4

make this mistahe.

I J

! 3.5.3 Duane Arnold Event e As a fesult of walking down the as-built piping for the SDV at Duane Arnold, it was discovered that the SDV drain valve was installed backwards. This did not hinder S0V draining but indicates a lack of proper QA for the scram system installation.

Tne Duane Arnold SDV drain valve is a two-inch globe valve designed to seat when the SJV is pressurized following a scram. The backward installation of l

the valve resulted in an unseating force on the valve plug when the SOV was

! pressuri:ed. This could have resulted 'in excessive leakage from the SDV fol-lowing a scram. Ho.4ever, there is no indication of excessive leakage.

)

w - - - - - , , -, , - - , , - ,n -- , , , - . - - - , y v

l 3 - 1.-

i The drain valve is being reinstallec correctly before Duane Arnold returns to power operation.

J

'3.5.4 Peach Bottom-2 and 3 Event

~

4 t

The backup scram solenoid valves at Peach Bottom-2 and 3 were found to be t

' inoperable.. These valves are discussed in Section 2.3 and are shown in

~

Fi gure 2.6. The valves were inoperable due to the fact that they had 250 VDC l

}

coils installed and a 125 VDC power supply connected to operate _the coils.

o

• Testing verified that 126 VOL would not move the 250 VDC solenoid valves.

The Peach bottom backup scram valves were not listed as safety-related and i 'therefore were not part of the safety-related testing program. These valves are not given credit for operability in safety analysis for the scram system.

The _ installation error, however, identifies another QA problem with the scram i 'systein at a BkR.

1 The backup scram solenoid valves are being replaced at Peach Bottom and an 1 ' operator has been stationed to operate the backup scram valves manually until l

4 the solenoid valves are replaced.

1 1

i 1

f-f

-4, - - ,,-- , ,g---, . --y-- , ,- ,. v - m ->g-

EA5T UEST 40 4 36 42 0 0 0 .

46 35 2 0.,. 0 0 0 0 -

36 42 10 0 0 0 0 0 36 34 0 36 0 0 0 0 0 0 ,

42 0 20 0 24 0 0,.

0 0 0 0 0 0 0 12 40i 44 0 30 28 40 35 0 0 0 0 0 0 0 24 0 40 0 0 34 32 0, 34 30 0 0 0 0 0 0' O

' 30 , 4 3B 0 18 8 0 0 0 0 0 0 22 10 0 34 2E 2 0 0,,. l0 -

24 0 0 0 0 0 0 0 42 36 30 14 26 2S 36 I 8 0 3S 32 0 12' 36 2-W 0 0' O O O O O 0 0 0 0 O O 0 3 34 42 34 20 (JO 28 122 (26

' 40 0 36 0 33 0 0,. 0 0 0, 0 0 5

{0 14 40 40 24 0 0 0 0 0 '

1 l 26 6 40 2 0 0 0 0 7 O 0,,,. l0 28 42 42 30 0 0 0 j i3 02 05 10 14 18 22 26 30 34 '38 42 46 50 54 58 1800

' Control Rod Scram Group Assignment Figure 3.1 Control Rod Positions After First Scre.

gy.

EAST WEST 30 0 14 18 0 0 0

' 26 42 25 0 (! 0 0 0 0 ,

0 28 34 0 0 0 0 0 0 12l22 30 0 0 0 12 0 0, 0 0 0 0 0 0 0 30 35 0 14 4 30 26 0 0 0 0 0 0 0 8 8 30 0 0 26 14 0, 0 0 0 0 0 0 0 14 0 30 0 22 20 0 0 0 0 0 0 0 4l 0 8 0 0 14 12 0 0 Og 0 0 0 0 0 0 0 26 22 20 2 12 12 16 2 0 0 0 0 0 0 0

' 0 0 0 0 0 0 0 0 3: 20 0 0 16 Os 0 O O O

) 10 26 ' 24 0 0 8 l '6 l 4 ' O O O O 5

18 0 la 0 28 0 OJ 0 0 0 0 0 0 1

I 12 0 0 32 30 14 0 0 0 0 0 7 O 30 0 0 OJ0 0 0 0

' 2 34 22 4 0 0 0 3

1 02 OG 10 14 18 22 26 30 34 38 42 4G 50 54 58 1800 Control Rod S: ram Group Assignment Figure 3.2 Control Rod Positions After Second Scram

. e l

t

• • WEST EAS.

i 24 0 0 0 3 0 0 I _

20 40 22 O I 0l s- 0 0 0 0 0 16 0 24 30 0 0 0 0 0 0 26 0 0 0 6 0 0 0 0 0 0 0 0 t

w, 0 24 36 0 2 0 24 15 0 0 0 0 0 0 0 2 0 24 0 0 20 0 0 0 0 0 0 0 0 0 0 0 0 0

. 6 0 26 0 0 0 14 16 0 0l0 4' 0 0 0 0 0 0 -

0 0 0 2 O O OJ 0 15 16 14 2 4 6 2 0 0 0 0 0 0 0 0 1 0 0 26 12 0 0 2 0, 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 3

16 l 15 2l 2 5 4 0 10 0 0 0 0, 0 0 0 0. 0 ,0 1 6 0 0 28 26 8 0 0 0 0 0 7

O 24 0 0 0, 0 0 0 0 3 0 30 10 0 0 0 0 02 06 10 14 18 22 26 30 34 38 42 46 50 54 53 1800 Control Rod Scram Group Assignment figure 3.3 Control Rod Positions After Third Scram 4 -

0 $

e e

C U

b H L z M

]7

/ a:

L' <

3 c3 E

E 5

=

!c ,"

E d

8 e a

> b W $

c a m u

\

1. d $

.\ 3 g

/ / \ c-g N \ a g g , / // :2 [

5 m< /

e Ymb N

p.

f N,'

// fa 55 W f  :~ f y a: # f5 5

a Di N [

E .

'a .5 E

E / *

• j

/ z "

O ,

2 5 ~ 8 e W

E a $r $

w is a

l

a -

n '

T V2: 0 Dt Cv N i u

m 0 -

E .

D I

S 0 m o

s T

s

%, 0 N 3

'o r

t t

u a

E u v o v

s -

z 0E 4

e? $" h7!

f

)

N .

O I

T A

P V E

L GMO t

E

(

uS t T iJ

>zi n U wAR O Y

5 stD A L

n1 o tR tm aa cn Au T

N

,,) mp gQ RE E

M P

e I ru ve is U

Q

• E s 'O u u -

g V 3d) pt l D

w o s H n l

=~ S Y

vh oR sD ~ -

=

R R

E F

~ y, S N

o v i W O

R B

v s

e F O

H E C D O T E

S I

/ AI C

H K

S T

S .

O .

a$

ct 5

3

. c. 8t 4

E ut E W O , R U

/

s G m 'U I T ) C F V4e - H D r.

Sy -

C O'4 ni

- g

8 d 3

,I k'

c; 3,1 r

i,l x/

I y

r h

s =

/.

1  ;

4 0

E n.

C")

g 7NI

%3 V);

M10WC,q,, , .

O aa

- g2 -

m

.T <

N

$ d ki c

$5 1s O

s

.- "" 8 4

$s a < ;a E W

O3h 1 J i [

u  ;

$$$ 3"

• g 4 3 ll "t 5 oc-> ,kgaa - }'

C -.

sm

> . c o a

/ $

) -O- k ..

m i  ;

n m n

sh,> "

w i v r s

,u2 -

\ .

v .-

e c

8? 4 V

% L ..

1

' 4

. e 4 o ~

n / s Y

v

> g o n / s s*

T u

> h b a

'T 7NEW 37H10 4C_,

c.

as

,.s- .,

sa g ut f, % w 12 g

- .- ts sa

=w g 1 a 7 1 m

. 5

$ w a ,

i

{

0> gas :t ,

?Y $

Y" E "

(

I @

@ S

'yftc I je g m

al* *e as/,vi a w

.o a a

\ S m hj03

=

n

.0 ._

g sa5h m 3 r ic

!! a s\ "

1 0-"au=3 r 6

L l

l

0 9 W

,e1 Si s/ s 6

t  ;

/ ._

?

i i i '

3w ia

'4 9NfW M FAQ

• C ,

et W is e 2 a

e d o as d hg 6% w 3 _. - sa gg a g W uES 3s 4

J tf7 S

(J 8ij g

iu m S6 gig -  :.x w 1-@ @ S i8 i e g 3 *

(

eo-> 3++C I m 5

2 8 "/ e 5

1.

3 x ( -_ .

g

- g.

ws i?)h a 3 f g

[. 8 F \ 1

-6 c u

w> 9 e u 0-Xe c c

3 1 l l

l

F . .

1 1

6m

,_.-. .. ., 3. .

-'C

, c1p 3 4 e'

• v/

.s 1

1

/ s s ,

> g Q

Q '

st ,

Eir L a

ss,m ano #_

r$

W C

0 I

c 0

l1a ei n d

dD

~ b- C i E =1 <5 i, 9 g .~'. Go gg e 9 = -

\

~ Z ',

1 1- >

- 2_. g t a g g s I

e c 4 2 g u

d d d >E 3 <[

Y .i b o a 288 a } t ti  ! @ O C I_k C -

( il ' ~ 3o -W ;

g-j m.J,C $

[ -

> O k Q* g 8 'a /Ab 2 o a 2 a

~

\ j I .

e

- g#

sae d

> T f e 9 \ P --

s as m>

O e -

L 1

I l

/ . .

l A

t e

t!

t O 4

~%

W oo Q tLI O

dys V) '

sr I

L spy

+-

F +

D a

._ ( \'S

@ '~

p E

w  %;L u y.

D , a-s

,i p, e s

F gN -

-R 0 Lu O,.

$ N S 5:

> vs 4 m ==

3 vs M, 4

g _g gD "I %9N 2m a .-.

(

%8 g3 ds a g,p R OT > b 9 s, 4

=g S Qwa + -

g uds

,t 0, -

a @o sav o 2" *

,g 3 a xa . + n

~

@ l

_, E 1

I N 7--. lr + $

.J  : "

,,gsp. I u

t l -

l 2 'a" W 1

t- Ed > o 5 > 8 b

0 i O J < k Z - .

z 4

> m 4-- CD -

se Q- g b 2 w 4 k $

i h o $.

i V)

Pi xxD

1 l

4 l

A

\ p.I $ ' ,

I h Au.To SCRAM LEVE L

^ ^8 109 GAlloHS A

LEVEL INST RLA h EfC ,J g ' .

yr Y _ RCD %L OCK LE.V El.-

^ N 69 G AL!.O N S

,/

i I

Nf i

b I ALARM LEVEL.

u "39 G ALLO NS w ~L'

,5[ -+ T O 3 RAIN I

S c. R AM TIS C H AR fi E BRRNS W IC.K RNIT 4l INST RR M E NT VOLKME <

LEVEL SWITCHES, BRUNSWICK UNIT #1 FIGURE 3.11 '

l

To heulee r SOllT H NORTil hv VEtdT VALyg -

p' +, go gg '

c' (< JH //c 1

//

U sW kjh" ,

'- u a

4 0

~

' ='

-3 SDY y ~I scnAM _

) 1 + 5WITutEs 4 1 I

I g.

f ,

8" _

l

[ "

T<i Rob BLotg Q switcu b 1 2."

at Y FN 1

HtGl4 53V LEVEL SWITCH L-. i O+ - sf.,-

L ,_J (AL A R H) mi

,gr 9t Snhals 4 9 9 M f+ 4 + t +

- PlPlNG 8*

VRVE [ ' y 4)

~

+pipsy%T u H clos o VAi.ve a.lo % lame, i

M vnwes re 16cLATf. SmTcHf5 RBE3rT-

.b Flo AT 6 WITCH

'ISOLATIf4G VALVES MlD D4' AGED PIPitlG SUPPORTS, BRlRISWICK ltilT #1 k FIGURE 3.12 4 7 4 M 4 6 E'D H AN 6ERS

e.,

. , ~ .. .

-r

. . ,: - .;  : . ..; 4

' . ._ - a

y. .-

s

,a

.r'.

- .. ,4 . -

,t .y , ,. s. es-v.-

? .. .

i .' , : +; . : ,, ;t A 4.0

,. " " , f. e- ,

4 s ..

.(

=

.,.,.s... m.- A.. .

. g -

4 i

-- ,t ). .' ' i;% i

',[#g/

: )<.t, }E3; # ,1

/. =

h.- 4 ,' , p 3 t s

...- j i}. .y* ,. - (.]'y ;e- -Q

,,e

'.t.,... . .. g

..r *~ e2.. ,g.- 1,7,./ %a

- >- u s. g 4 s .

s 1

-Wr .g :.

• p 4,..<

. s 1, :. e, g;- -\..-

p,. ,

-* g 4 '# * ,

.... . }, . l, .. ..,f. . .s, y..,. . f?a. ; ' T.,

..-..- .x.

m\..\~

- - - _. m _ _2c u . t %  !

f r

FIGURE 3.13a SCRAM DISCHARGE INSTRU."ENT VOLUME l I

g -,wy - .._

.1 +

'. l 9. e

+ .s .

I *,J- [

-% t - l

~7 - '-

. Yes =T.

F e ... g# ;

s . ./ .

p

.3 . : .

,, y' *;O }

.- .- - , , ,-e .

h y,  %

- L,.

, . f 5-- t,. , 4, . '

r., . ,'.,. . $. ' .

j.. .- #. .' s

.y

.e'

• . r. J' *}

.,,t  ?

. .. % + yy:. -

f's ,, ,

T . f';' ' ' .'T e .

l l

. ; r,! i.'5: J;b .S. ../ C. .

.n ' ~

T / ,4, .1 1

'.m. ,

. , Yi ). - $p'?' ' .YJ ,[ * . i .j< 4 . ,a  ; *. ..:

[ *:+ s ;..

,' ' 1, -

..:. ,cn .~,t' M. - ..A.. .

,- m .- .:.y.- 2 :::ui g . w. h _

FIGURE 3.13b DAMAGED FLOAT e

= . _

1 i

4. NRC REPORTS Da B4R SCR A;4 SYSTEM M4;ie..CT 0aS The NRC Staff nas studied the recent SWR scra:n system malfunctions and has issued or will issue reports on those events. The reports provide evaluations and recommendations concerning the event. To date the only report received froc. the NRC Staff concerns the Browns Ferry event. Reports on Brunswick and daten ara. baing prepared by the NRC Staff.

i

) .

Tne NRC Of fice for Analysis and Evaluation of Operational Data issued a report .

on tne oroans Ferry-3 event. Tne report contained the following findings:

1. Tne cause of the partial scram failure was water accumulation in the east SD,.

2. Tne SDIV "Hign Water Level Trip" did not and does not provide protection against filling the east 50V even for normal venting and draining conditions.

3. A single failure (e.g., west side SDV vent or drain line block-age) can completely disable the SDIV "High Water Level Trip" installed to protect against loss of scram capability for the control rods.

4 With the present SDV/SDIV layout, a single failure (blockage) of an SDV vent or drain path can cause a partial loss of scram capability.

5. There are numerous actual and potential mechanisms for introducing and accumulating water in the SDV's with nc! accumulation in the 4,

SDIV.

i l

--n -- -n e, m e - ,, e w ,

42

c. -Tne current SD','/SDIV layout results in tne automatic "Hi Water Level Trip" safety function being directly dependent on the non-safety related reactor building waste drain system.

7 The float-type water level monitoring instruments on tne SDIV have a significant degree of unreliability.

6. The current B'aH RPS logic does not allow scra reset to attempt a re-scram if certain automatic scram signals are present. .

9. Failure to close of a single SDV vent or drain valve during a reacter scram can result in a unisolatable release of reactor coolant outside the primary contain.nent into the secondary containment.

10. The emergency operating instructions at Browns Ferry did not cover a partial or total scram f ailure event...

The report contained the following recommendations:

A. The operability of the SDIV "Hi Water Level Trip" should be in-dependent of the venting and draining requirements.

l d. Modify the 50V vent system to improve drain reliability.

1 C. SDIV instruments should be both redundant and diverse.

O. All vent and drain paths from the SDV should have recundant auto-matic isolation valves.

The AE03 report is attached as Appendix B. It contains the details behind the findings and recommendations listed above. The findings address all the known problems with the Browns Ferry scram system and they apply to many other operating Bh'R's.

J

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

t .

5-1

5. C0'iPARISJ.i TO WASH-1400 PREDICTIO.iS Tne Browns Ferry event, wnere control rods failed to insert due to undetected t

.a*,er present in the Scran Discharge Volume was considered in the WASH-1400 rep 3rt. The layout of the discharge heaoers and the hydraulic control modules is shown in Figure 5.2. A detailed f ault tree was developed for this event 4

and a scae. Mat redaced f ault tree is shown in Figure 5.2. The probability of leakag: into the 50V is given as .12 per denand. This probability is under- .

estima*,ed, because experience indicates that there is always some leakige througn the seals daring normal operation. The probability for a blocked drain is given as 1.3x10~ per de.aand. This is the drain line from the SUV If header into the 531/. For a passive run of pipe, this seems reasonable.

one follows this event on the fault tree (Figure 5.2), one can conclude that the system is not truly single failure proof, because it requires only a block-age in the drain line to cause an unsatisfactory trip since the probability of some undetected leakage into the SDV, particularly at Browns Ferry, is large enough to be very close to 1.

The WASH-1400 report does not cover the possibility that a blocked vent or i damagec level switches could also pennit undetected water hold-up in the scram discharge volume. Considering these failures to date, it appears that the probability that these events will occur is considerably higher than

1. 3 x10-7 per demand. The task force is developing a modified event tree and hopes to provide a new estimate of the probability of this event in a later report.

I

5-2 19 view ot i.ne WA51-14JJ analysis at least two actions can be taken to red;ce the probability of an unsatisfactory trip. The first is to assure tnat a vent path is always clear. This has already been done in most plants.

The second is to regularly (and frequently) test the level switches both during nor,r.a1 operation and following scrams. This procedure should be fol-losed until it is deter,nined when and why such damage occurs. If rugged and reliaole swittnes are available which could withstand such events, then they .

sr oald be ins.alle;. It coe; not seem reasonable, at this stage, to add more instru entation which coula also be damaged and rendered inoperable.

. .a ,

e

.I

>-r-

'" ~

', :z. J

~&

/ /G, f l *

. .1

/ , zw n. m nT e. .

1. /

N +.

O Cb li / 'N 'l ",?.

e D.

a F- .

3 L_._

w,, ,

/ '\ y

. . i

.N  !

[ ,A s \\ I !Ij s .

n..

i

/ / s i I!l . -

., yL ,

i \

\

) ps:M. .

lI 1

...+,0 l ,

g L

k, \ h s. .

.m e -

\ , .

& , nn s4 r5 '

/? -

Jm.

I'*

' %-l- Ngjun

- , , ~ . . . '

r y ..,,

s, >.w s.. s .

FIGURE S.: Reactor Protection System Trip Discharge Headers l

(

t

'Ma ,a- . en w g a

e a i

r <

. . . 4

,...,.-- . = o.

c-

. j r t

-- ~ * -

. ~ .~ w, l .:..

L1 . , . , j i .'. O.T..

~ . .e .,

f,gia

/__ I T t-x = ,,

3 i

i

~

7T i i i e

N

...x.g s

. .:.. . . .. . A.s

. ,. g , i.

... m.s

.a ,

o s..,

e

~;. ,~;. n. ,.. u ,..

(3 ( 3 pi,7

===.o m.wo m.m.: m. ammo TT I W

l

====s==. M .e

• eump M.

then s a_-

"""" g gew ., g tam.as h emunes _

/

W mF W m..

.Mb

• asas j "'" " ! __f ees Tus a

I88 -

go

• n euuur W

jammmmmmum um I

m..,,.,. o...,.,,

(O...

M M

Lan> W.

N4H j d'M y z Q r-v =VIIIOb

,4W ' $ 7 4, P

FIGURE S.1 Reactor Protection System Reduced Tault Tree 4

e r _ ._

'~

e ,

6-1

o. ACRS qJESTI't.S This section lists questions asked at tne ALRS full Committee Meeting held on July 10, 1901 during the NRC Staff oresentation on the malfunctions of the BWM scram system. Short answers are crovided where possible.

1. QUESTION: How and wnen are the scra:a level switciles (in the 3-scra.a discharge instrument volume) tested? How are

- l they tested when the plant is operating? -

A"b E A : Ine Standard Tecnnical Specifications for BWRs state

' that the scra.n discharge volume water level-high i switches are checked monthly.. The float chambers can be valved out of the system and filled with water via instrument taps. (Significant differences exist- -

between plants and the Standard Tech. Specs. do not apply to all BWRs). Brunswick, for example, is

typical as it does not have a guage glass on _the scram discharge instrument volume. Fitzpatrick has a guage glass on its scram discharge instrument volume.

2. QUESTIO.;: What system interactions are involved in tne various positions of the inode (shutdown, refuel, startup, run) j selector switch?

ANSWEPs:

This feature _is described in Section 2.3 (Fage 6).

Additional information is available in the Browns Ferry FSAR. The administratively controlled mode

6-2 switen is provided to select the r,ecessary' scram functions for various plant conditions, In addition to selecting scram functions from the proper sensors, the mcde switch also interlocks such functions as control rod blocks and refueling equip.nent restric-tions, which are not considered here as part of the Reacter Protection System. The switch itself is ,

designed to provide separation between the two trip systems. The node switch positions and tneir related scram functions are as follows:

a. SHJTDOWN - Initiates a reactor scram; bypass main steam line isolation scram and main condenser low vacuum scram if-nuclear system pressure is below 600 psig.

b. REFUEL - Selects Neutron Monitoring System scram for low neutron flux level operation;

)

bypass main steam line isolation scra,a 4

and main condenser low vacuum scram if nuclear system pressure is below 600 psig.

i

c. STARTUP - Selects Neutron Monitoring System scram for low neutron flux level operation; bypass inain steam line isolation scram

{ and main condenser low vacuum scrab if nuclear system pressure is below 600 psig.

i l

l l

l

. o e

1 5-3

d. RJ:4 - Selects Neutron Monitoring Sys cn scram for power range operation.

3. QJESTIO:1: Where did the water ha: amer occur? Where did the piping bend?

Wnat are the generic implications of water ham.ner in this sys-tem?

r A,iS WER : The water ham,ner occurred in the scram discharge instrumeat volwae and its associated drain pipe (Figure 3.10) at S uns- -

wick 1. Pipe supports were torn off tne wall from the bottom of the scram discharge instruraant volume and al? along the drain line up to the drain valve. It is believed tnat the water hammer occurred on October 19, 1979 and that the supports were damaged in a single event. This fact requires further substantiation. The 'aneric implications of tne pos-sicle loss of function in a safety system require further

~

evaluation. The potential for the non-existence of suf-ficient exit volume for the working hydraulic fluid to push ,

the control rods into the reactor core should be examined.  ;

4. QJESTIOT: Is it part of the design to close the vent and drain ,

rapidly in order to provide a pneumatic cushion to pi event water ha.mner? Is the valve closing deliberately delayed?

ANSWER : The design rational of this aspect of the system is not ,

available at this time.

~

6-4

5. QUEST!0'.: Why did it take from October 1979 to June 1960 to issue a ,

corrective bulletin on the scram discharge instrument volute float malfunctions?

ANSWER: Inis question requires evaluation by tne NRC Staff.

i

6. QUEST 10:4: Did the float rods bend during regular operational use?

Was it due to a maintenance or design error? Wnat cleartn=es ,

are provided in tne float track? Does the float have guides or is it completely free to move?

A:is.liR : The answers are based cn preliminary information. The bent float rods at Brunswick occurred when the ficat ball and its attached rods were slam,ed into the top of the float chamber by reverse flow from the drain line. Tne bent rods at Hatch I occurred prior to or during installation of the float balls.

7. QUESTION: Can water get into the scram discharge header without a scram? Is the vent and drain system design adequate to ensure that water will not be held up in the scram dis-charge volume?

l ANSWER: Water can get into the scram discharge header and be con-tained there yet be undetected by the scram discharge in-strument volume instrumentation. This is discussed at length in Section 3.2 and the AE00's report.

I i

L

5 y 6-5

8. QUESTION: What prevented the rods from going in tne first time at Browns Ferry 3?

AN5'WER : The presumed reason was water in the scram discharge volume did not permit the control rod scram outlet water to enter i the scraca discharge system. This is discussed in detail in Section 3.2.

9. QJE57:US: dnat are the suggested design changes to reduce the com-monality of the system venc and drains?

ANS'.?E R : Newer BWR.s have eliminated the drain line between the scram discharge instrument volume. The scram discharge instrument volurae is contiguous to the scram discharge header. The water level in the scrain disnarge header can d be measured directly. Previous difficulties associated with the vent and drain system are reduced with the elimina-tion of the 150 ft. drain line to the scram discharge header instrument volume which exists at Browns Ferry 3 today.

10. QUEST!0": Why is it necessary for the rod alann and rod block switches to be operable?

ANSWER: It is not required for scram but is nrovided as a warning a

to the operator so that sufficient volume is available in the scram diecharge header to receive the control rod work-ing fluid. This is described in detail in Section 3.3.

7-1

7. BIBLIOGRAPHY Final Safety Analysis Report, Browns Ferry Nuclear Plant.

Volume I, Chapter 3 and Volume II, Chapter 7.

~

Final Safety Analysis Report, Brunswick Steam Electric Plant Units 1 and 2. Volume I, Chapter 3 and Volume II, Chapter 7.

Final Safety Analysis Report, Edwin I. Hatch Nuclear Plant Unit 1.

Voluma I, Chapter 3 and Volume II, Chapter 7.

t Inspection and Enforcement Bulletin 80-17.

Inspection and Enforcement Fundamentals Course Manual Boiling Water Reactors.

Report on the Browns Ferry 3 Partial Failure To Scram Event on June 28, 1980 by the Office For Analysis and Evaluation of Operational Data, July 30,1980. Prepared by Stuart Rubin and George Lanik.

4 i

U.S. Nuclear Regulatory Commission, " Reactor Safety Study - An Assessment of Accident Risks in U.S. Commercial Nuclear Power Plants ," WASH-1400 (NUREG-75/014), October,1975.

h a

e,,s

.. p..-f. k$s, Jr. .

deformations) and of the potential for " water hammer" being the cause of the ob:erved deformation at Hatch and. Brunswick.

(A joint IE-NRF, visit to Brunswick is scheduled for July 22, 1950.)

6. We are explorins ways to deal economically with the issue, e.g., thru the B'a'E Owners Group.

r, Jf e 0 3 i Dh-Paul'S. Che:k, Assistant Director for Plant Systems Division of Systems Integration cc: H. Der. ten V. S*.ello N. '.ascly E. Jorden T. Spets L. P. uter. stein W. Kroger V. Panciera M. Hendonca C. Graves D. Thatcher H. Richings W. Pasedag W. Mills D. Eisenhut T. Novak .

T. Ippolito T

,