ML20032C083
| ML20032C083 | |
| Person / Time | |
|---|---|
| Site: | Millstone, Davis Besse, Oconee, Palisades, Saint Lucie, Point Beach, Sequoyah, Pilgrim, Brunswick, Vermont Yankee, Crystal River, FitzPatrick, Trojan  |
| Issue date: | 10/29/1981 |
| From: | Rosen S INSTITUTE OF NUCLEAR POWER OPERATIONS |
| To: | Michelson C NRC OFFICE FOR ANALYSIS & EVALUATION OF OPERATIONAL DATA (AEOD) |
| References | |
| B-81-0275, B-81-275, NUDOCS 8111060681 | |
| Download: ML20032C083 (23) | |
Text
B-81-0275 l
Institut3 of l r-q i
d Nuclear Power gg5 W
Operations 1820 Water Place j
Atlanta, Georgia 30339 4
Telephone 404 953-3600 October 29, 1991 Mr. Carlyle Michelson, Director i
Office of Analysis and Evaluation of Operational Data USNRC Washington, DC 20555
Dear Mr. Michelson:
Per our Memorandum of Agreement, dated June 1, 1981, the following material is forwarded:
SOER 81-15, Partial Loss of DC Power SER 81-81, Velan Swing Check Valves SER 82-81, Brazed Swing Guides-Dresser Electromatic PORVs SER 83-81, Boron Dilution Events SER 84-81, Control and Protection Losses Due to RPV Level Instrumentation Header Isolation SER 85-81, Failure Of Core Spray Valve to Open SER 86-81, Unmonitored Radioactive Liquid Release SER 87-81, Inadequate Reactor Coolant System (RCS) Water Level Indication SER 89-81, Level Instrument Oscillations Due to Reference Leg Flashing SER 90-81, High Occurrence of Degraded Hydraulic Snubbers SER 91-81, Steam Voiding in the Reactor Coolant System During Decay Heat Removal Cooldown Please contact me if you have any questions regarding these items.
Sincerely, f>
S?
R sen, Director Analysis Department RSL/ bas cc:
E.L.
Zebroski
/\\
D.L.
Gillispie 6, \\
Q PDR
LIST OF SIGNIFICANT OPERATING EXPERIENCE REPORTS ISSUED IN 1981 AS OF OCTOBER 8,1981 SUGGESTED SOER NUMBER DATE ISSUED TITLE PRIO RITY 81-1 March 23,1981 1.1 proper Steam Red Generator Level Control & Loss of Ileat Sink Due To A Loss of Instrumentation 31-2 f.; arch 23,1981 System Response And Red 2
Operator Uncertainty Due To Failed Instrumentation 81-3 March 23,1981 Loss of 24VDC Non-Yellow Nuclear Instrumentation Power Supply 81-4 April 9,1981 Pressurizer Level Red Anomalies During Natural Circulation Cooldown 81-5 April 9,1931 Instrumentation To Con-Yellow duct Natural Circulation Cooldown 81-6 April 9,1981 Uiianalyzed Conditions Yellow Encountered During a N atural Circulation 81-7 April 9,1981 Loss of Forced Circu-Yellow lation in the Reactor Coolant System 81-8 April 9,1981 Spurious Actuation Yellow of Safety / Relief Valve
LIST OF SIGNIFICANT OPERATING EXPERIENCE REPORTS ISSUED IN 1981 AS OF OCTOBER 8,1981 Page 2 81-9 April 9,1981 Desiceant Carry-Over Green
~
to the Instrument Air System 81-10 April 9,1981 Event Sequences Yellow Not Considered In Design of Emergency Bus Control 81-11 April 30,1981 Partial Loss of Green Emergency Feedwater Pump Suction 81-12 June 24,1981 Reactor Coolant Red Pump Closure Stud Corrosion 81-13 July 31,1981 Cencurrent Loss Red / Yellow / Green of liigh Pressure Core Cooling System
'81-14 August 5,1981 Cracks in PWR Green Charging Pump Lines 81-15 October 23,1981 Partial Loss
'lellow of DC Power
.7....-.
~
....x s
INPO Y S ^ CSIGNIFICANT OPERATING EXPERIENC N\\
81-15 N ~
October 23, 1981 PARTIAL LOSS OF DC POWER
REFERENCE:
UNIT:
MILLSTONE 2 DCC NO/LER NO:
50-336/81-005 DATE:
1/2/81 NSSS/AE:
COMBUSTION ENGINEERING /BECHTEL RELATED I&E INFORMATION NOTICE: 81-05 NUCLEAR NOTEPAD SIGl!IFICANT EVENT REPORT (SER): 81-5 DESCRIPTION:
The plant equipment operator, intending to take ground readings on System A DC bus, inadvertently rotated the control switch for the System A battery bus main breaker.
This de-energized the System A 125. volt DC bus with the following consequences (refer to attached diagram of AC bus configuration):
o Four of eight reactor trip switchgear breaker undervoltage coils were de-energized, resulting in a reactor scram.
Aturbinetrip,whichnormallyhollowsareactorscram, o
did not occur since the associated logic requires power from DC System A.
o Power was' lost to all control room annunciators.
At 29 seconds'into the event, the control room operator observed that the turbine had not tripped.
He tripped the turbine manu-ally using the master trip on the EHC control panel.
o The turbine trip stopped the reactor cooldown.
Cool-ant average temperature T had dropped f rom 574*F to ay, 525'E, pressurizer pressure from 2250 psia to 1760 psia, and pressurizer level from 56% to off-scale (low).
o The fastatransfer of the in-house loads from the normal station service transformer (NSST) to the reserve station transformer (RSST), which normally follows a turbine trip, did not occur because a portion of the transfer logic is powered from DC System A.
The 4160 and 6900 volt buses in System B were de-energized.
o The generator output breakers and the System A NSST breakers did not trip because DC control power was not availabic.
The System A in-house loads remained con-nected to the main generator and generator step-up t
l transformer through the NSST.
RED IMMEDIATE ATTENTION l
- YELLOW.
PROMPT ATTENTION A
GREEN NORMAL ATTENTION Wog
m
~
81-15 PARTIAL LOSS OF DC POWER Page two o
The loss of power to the 4160-volt emergency bus in System B caused the B diesel generator to start and supply that bus.
o The' generator reverse power relays initiated a 30-second timer.
That time delay would have to be satisfied before a reverse power trip of the generator output breakers would occur.
The System A DC bus was re-energized 51 seconds after the loss of DC.
This initiated a trip of the generator output breakers and a fast transfer of the AC buses in System A from NSST to RSST as well as automatic closure of the Main Steam Isolation valves (MSIVs).
It also allowed the System B 6900-volt bus (25B), which had been de-energized 22 seconds earlier by the turbine trip, to be energized,from the RSST.
Power was applied simultaneously to two Reactor Coolant Pump (RCP) motors and a condensate pump motor.
The starting current from these loads caused the RSST supply breaker for bus 25B to trip on overcurrent.
The B facil-ity emergency diesel generator continued to cupply the System B emergoney safeguard bus.
Approximately 10 minutes into the event,'the "B" emergency diesel generator tripped when a leak at a service water pipe flange sprayed salt water on the machine.
The spray caused a fault in an electrical connector associated with the electronic gover-nor.
This fault is believed to have caused the governor to decrease the. engine speed to the point of tripping the machine on low lubricating oil pressure.
The operator took action to re-energize the System B safeguards bus (24D) from the RSST.
When power was restored to bus 24D, 96 reactor plant and balance-of-plant instruments were discovered to have failed.
Evidently, when the diesel slowed down, increased currents to inductive loads caused fuses to blow in the GE/MAC power supplies for the affected instrumentaion.
It was subsequently determined that the fuse ratings were different from those specified in the equipment documentation.
The plant was maintained in hot standby while completing the recovery from the loss of DC power.
The existing combination of two reactor coolant pumps provided no significant pressurizer spray flow.
The operator did not associate the ineffectiveness of spray with greatly reduced spray control but rather concluded that a "hard bubble" had resulted from a collection of noncon-densable gases in the pressurizer.
Approximately 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> and 15 minutes into the event, pressurizer pressure increased to 2380 psia, causing both power operated relief valves to open for a short duration.
The auxiliary spray valve was subsequently used to control pressure.
81-15 PARTIAL LOSS OF DC POWER Page three SIGNIFICANCE:
Loss of power on one vital DC bus negated the following:
auto-matic opening of the main generator breaker, automatic turbine trip after a reactor trip, and automatic fast transfer of AC buses to the reserve station service transformer providing AC power from off site.
Had power not been restored to the "A" DC bus promptly, only "B" train essential AC power would have been immediately available, and only from the "B" diesel generator, which failed 10 minutes into the incident.
Loss of capability to transfer automatically any of the AC power buses to an off-site power source after a plant trip caused by loss of poweg on a vital DC bus may reduce the immediately avail-able AC power source to only one diesel generator.
The diesel generator corresponding to the de-energized DC bus generally would not be automatically tied to its essential safeguards bus and loaded as required, since DC power from the affected bus would be required to perform the operation.
Consequently, a significant degradation of available AC power could result from loss of a single DC bus.
Failure of the turbine to trip following a reactor trip causes a rapid plant cooling transient.
If unchecked by prompt operator action, a severe overcooling incident could occur.
The operator thought he had spray to the pressurizer, when in fcct, he had little, if any, spray flow.
Without spray, the pressurizer behaves much like it would if it was filled with noncondensable gases. The operator did not realize that the pressurizer behavior was indicative of a lack of pressurizer spray.
Power operated relief valves (PORVs) were subsequently used to control plant pressure increases when earlier use of a different RCP bombination or auxiliary spray flow could have eliminated challenging this equipment.
Restoration of power to bus 25B was hampered because large loads remained on the b*s while the attempt was being made to re-energize the bus.
The inadvertent action of a plant equipment operator in perform-ing his normal routines resulted in a plant trip and temporary loss of normal and off-site AC power sources.
Regarding the loss of control room annunciators, subsequent analysis of the " time to blow" for the fuses versus those which were actually installed, indicates that more time would have been afforded if the proper fuses had been used.
With the proper fuses installed, the diesel may have tripped prior to the fuses blowing, thereby avoiding the loss of instrumentation.
=
- :. -.:.~.=. -. :=
..~-..a.-:.
j*
u L
1 81-15 PARTIAL LOSS OF DC POWER Page four.
i RECOMMENDATIONS:
l i
1.
Plants should review trips and bus transfers on vital power l
logic systems to ensure that when a vital DC bus loses power, the following will be operable:
a.
transfer capability to an off-site. power source of at j
least minimum essential AC loads (one safety-related i
equipment train) 1 b.
starting and loading capability of on-site AC power sources (diese] generators) sufficient to power at least minimum essential AC loads automatic turbine trip and generator trip following c.
reactor trip (Recommendations 1.a and 1.b may be implemented through i
hardware and/or procedural changes.
Where procedural changes are elected, appropriate traini..g should be i
provided to operating personnela) t 2.
Procedures and operator training should address the i
.following:
i a.
the importance of quickly verifying that the turbine has tripped after a reactor trip to preclude a severe over-cooling incident b.
partial pump operation specifically addressing the effect of off-normal RCP combinations on plant response and pressurizer spray flow - (Procedures should include j
specific recommendations on the pump combinations to be used'for maximum spray effectiveness and the desirabil-ity o'f using this RCP lineup whenever possible to mini-1 mize the need for auxiliary spray.)
recovery from loss of'DC power events including guidance c.
on loading and unloading of power supply buses li 3.
The importance of ensuring that specified fuses are used whenever replacement is necessary should be emphasized.
l INFORMATION CONTACT:
W. B. Reuland, NSAC, 415/855-2977 J
A. G. Charbonneau, INPO, 404/953-5425 Please return the enclosed postcard to ensure receipt.
e e
-y-w.m-,r.e-,*,.me-3%.,-
.=-....m
,,,,.,em,,,,en..,,-,,.,,,mm,
-,,-.r.,,_,mor,ww--.-_,
,-e.
y.
Partial Line Diagram of Feeds to AC Busses Generator Step-Up Offsite Power Transformer Main Generator 345KV Reserve Station
/
ServiceTransformer 345KV V
m
__evv,_
B akes
\\
24KV Normal Station Service Transformer g
ww g
_tvem
=
r' I
l i
I I
6900V (A]
6900V [B]
25A 23 D'
l l
4160V (A]
4160V [B]
24A 24B f
l 4160V [k]
4160V [B]
l 24C 240 l
[B] Diesel
[A] Diesel n, Generator q, Generator V
v, l
Breaker Notes:
l G Trips on Turbine Trip; Powered by A DC l
@ Trips on Turbine Trip; Powered by B DC
@ Closes on Turbine Trip; A DC Logic Required
@ Closes on Loss of Normal Power; A DC Powered
@ Trips on Turbine Trip: Both A and B Required
@ Closes on Loss of Normal Power; 8 DC Powered
@ Opens on Loss of Normal Power B DC Powered
@ Opens on Loss of Normal Power; A DC Powered
[ ] = Train Identification
= Bus identification
C121]
Gillispie (INPO) 9-Oct-81 11:48 AM i
NSAC/INPO SIGNIFICANT EVENT REPORT (SER) 81-81
SUBJECT:
VELAN SWING CHECK VALVES UNIT:
POINT BEACH 1 SEQUOYA,H 1 DOC N0/LER NO:
266/81-010 327/80-150 EVENT DATE:
,7/16/81 9/27/80 NSSS/AE; WESTINGHOUSE /BECHTEL WESTINGHOUSE /TVA
REFERENCES:
COMBUSTION ENGINEERING INFO BULLETIN 8107 IE INFORMATION NOTICE 81-30 EVENT DESCRIPTION:
DURING THE PERFORMANCE OF AN INTER-SYSTEM LOCA LEAN TEST AT POINT BEACH 2, IT WAS DISCOVERED THAT CHECK VALVES NEAREST 1
THE REACTOR COOLANT SYSTEM (RCS) IN THE LPSI HEADERS WERE
[
FOUND TO LEAK IN EXCESS OF THE ACCEPTANCE CRITERIA.
THE VALVES WERE DISASSEMBLED AND FOUND TO BE STUCK OPEN.
THESE CHECK-VALVES (VELAN SWING CHECK VALVE ASA 1500 LB-6' DWG 78704 REV B) HAVE A VERY SHORT DISK STAND-OFF POST AND UTILIZE'A SHARP EDGED FACE ON THE PORTION OF THE VALVE BODY THAT SERVES AS THE DISK BACKSTOP.
THE DISK NUT BARELY CLEARS THE SHARP EDGE OF THE DISK BACKSTOP WHEN THE VALVE IS FULLY OPEN.
THE PROBLEM OCCURRED WHEN A LOCK-WIRE ON THE SEGMENT l
ARM DISK NUT JAMMED BETWEEN THE DISK NUT AND THE BACKSTOP OF THE VALVE WHEN THE VALVE WAS OPENED.
THE BACK-UP CHECK VALVE
.*N EACH SAFETY INJECTION LINE WAS ALSO INSPECTED AND CYCLED BY HAND.
ALTHOUGH THESE VALVES WERE PROPERLY SEATED, IT WAS FOUND THEY COULD BE INDUCED TO HANG UP BY THE SAME MECHANISM.
A SECOND SIMILAR EVENT INVOLVING STUCK OPEN FAILURE OF THE SAME TYPE OF VALVE OCCURRED AT SEQUOYAH.
POINT BEACH MODIFIED THEIR VELAN CHECK VALVES BY REPLACING THE LOCK-WIRE WITH A STAINLESS STEEL COTTER PIN WITH THE APPROVAL OF THE VENDOR.
SEGUOYAH MODIFIED THE LOCK-WIRE ARRANGEMENT IN ORDER TO ELIMINATE THE INTERFERENCE PROBLEM.
1 LATER VERSIONS OF DELAN VALVES USE A LONGER STAND-OFF POST AND A BEVELLED FACE ON THE PORTION OF THE VALVE BODY THAT SERVES AS THE DISK BACKSTOP.
OTHER VELAN VALVES HAVING THE SAME DESIGN ARE AS FOLLOWS:
RATING DWG NO.
REV.
1 ASA 600LB-6' 78705 B
.ASA 1500LB-8' 78800 B
COMMENTS:
THIS GENERIC PROBLEM COULD CREATE A FAILURE OF THESE CHECK VALVES WHICH MAY LEAD TO RAPID OVERPRESSURIZATION OF THE LOW PRESSURE SAFETY INJECTION AND SHUTDOWN COOLING SYSTEMS.
INPO IS CONTINUING TO EVALUATE THIS EVENT.
INFORMATION CONTACT
1 entru was found.
(1223 GilliSPie (INPO) 9-Oct-81 12:45 PM
~
NSAC/INPO SIGNIFICANT EVENT REPORT (SER) 82-81
SUBJECT:
BRAZED SPRING GUIDES-DRESSER ELECTROMATIC PORVs dNIT:
PALISADES DOC NO._.ER NO:
255/80-035
.1 EVENT DATE:
8/30/80 NSSS/AE.
COMBUSTION ENGINEERING /BECHTEL
REFERENCE:
COMBUSTION ENGINEERING INFO BULLETIN 81-03 i
EVENT DESCRIP1 ION:
i DURINO PLANT STARTUP, IT WAS DISCOVERED THAT ONE OF THE POWER
^
OPERATED RELIEF VALUES WAS LEAKING.
SUBSEQUENT INSPECTION REVEALED THAT ONE OF TWO BRAZED SPRING GUIDES ON A DRESSER L
ELECTROMATIC POWER OPERATED RELIEF VALVE (MODEL-NO. 31533VX) l THEREBY, CAUSING THE SOLENOID RETURN SPRING TO JAM AND THE
'i HAD SEPARATED FROM ITS SUPPORT PLATE AT THE BRAZED JOINT, VALVE TO STAY OPEN.
DRESSER EVALUATED THIS EVENT AND HAS 1
CONCLUDED THA1 THE GUIDE WORKED ITS WAY OUT OF THE HOLE IN THE SUPPORT PLATE AT THE FAILED BRAZED JOINT AND WHEN THE SOLENOID WAS ENERGIZED TO OPEN THE VALVE, THE GUIDE AND SPRING
' JAMMED TOGETHER WHICH PREVENTED THE SOLENOID FROM RETURNING TO ITS OFF POSITION WHEN IT WAS DE-ENERGIZED.
1 THE PRAZED JOINT SPRING GUIDE DESIGN WAS DISCONTINUED BY f
'~
DRESSER IN 1974 AND HAS BEEN REPLACED BY A DESIGN IN WHACH l
THE SPRING GUIDE IS WELDED TO ITS SUPPORT PLATE.
THIS WELDED d
^
SPRING GUIDE DESIGN HAS BEEN USED IN ALL NEW VALVES AND FOR SPARE PARTS SINCE 1974.
COMMENTS:
I
~
THERE IS CONCERN THAT THIS MODEL DRESSER VALVE HAS A GENERIC DESIGN FLAW THAT COULD CAUSE THE PORV TO FAIL OPEN.
A MORE RELIABLE 'WdLDED SPRING GUIDE' DESIGN IS AVAILABLE' AND CAN BE SUBSTITUTED FOR THE BRAZED JOINT SPRING GUIDE DESIGN.
INPO PLANS NO FURTHER $ VALUATION OF THIS EVENT.
l INFORMATION CONTACT:
ART CHARBONNEAU, INPO, 404/953-5425.-
l l
1 erit r y was found.
j V
l ACTION:
l 9
r' c,
,v i,
C123]
Gillispie (INPO) 9-Oct-81 1:01 PM t
,.'YSAC/INPO SIGNIF.ICANT EVENT' PEFWRT (SER) 83-G1
-l b
SUBJECT:
BORON DILUTION EVENTS i
~
bHIT:
ST. LUCIE 1 DAVIS-BESSE 1 DOC N0/LER N3 50-335/81-002 50-346/1977 EVENT DATE:
12/1/80 5///77 NSSE/AE:
/"-
COMBUSTION BABCOCK & WILCOX ENGINEERING /BECHTEL BECHTEL ll l'
EVENT DISCRIPTION*
>x I
WHILE AT 100% POWER,'St. LUCIE OPERATORS ATTEMPTED TO BLEND BORATED WATER FOR MAKEUP TO THE' VOLUME CONTROL TANK.
A RELIEF VALVE ON THE DISCHARGE OF THE EARIC ACID MAKEUP PUMPS FAILED "i
OPEN AND THUS RETURNED THE CONCENTRATED BORIC ACID TO THE BORIC ACID MAKEUP TANK.
THE MAKEUP WATER CONTINUED TO DE ADDED WHICH DILUTED THE BORON IN THE REACTOR COOLANT AND f
RESULTED IN A GRADUAL POWER INCF: EASE TO 102%' POWER IN FIFTEEN MINUTES.
POWERWASTHENREDUCEDBY{MANUALEMER0ENCYBORATION.
BECAUSE THE RELIEF VALVE WAS DOWNSTREAM OF THE BORIC ACID FLOW ELEMENT, THE OPERATURS!HAD INDICATIONS OF NORMAL BORIC
~
ACID FLOW AND VOLUME CONTROL TANK' LEVEL.
THE ONLY DIRECT INDICATION OF A PROBLEM WAS THE-GPADUAL INCREASE IN REACTOR 3-POWER.
l JWHILE IN COLD SHUTDOWN BEFORE INITIAL CRITICALITY, DAVIE l 0:'
'liESSE.0PCRATORS ATTEMPTED TO RAISE THE uSTER LEVEL IN THE REACTOR"VE'SSEL,IN ORDER TO INSTALL IN-CORE DETECTORS.
AFTER RAJS1HG THE.LEMO., THE VESSEL BORON CONCENTRATION WAS DETERMINED BY4 MEASUREMENT TO HAVE DECF; EASED FROM 1850lTO 1763 PPM.
DUE TO A LEAKING DEMINERALIZED WATER VALVE, THE OPERATORS HAlfifN3ECTED DILUTED BORIC ALID SOLUTION RATHER THAN THE INTEdDED CONCENTRATE? BORIC ACID.
DORIC ACID INJECTION WAS
- THEN;/NITI'ATED AND; CONTINUED UNTIL THE VESSEL BORON CONCENTRATION g
1:ADCtETURNED TO 18G0 PPM.
l i
COMMENTSI l
WHILE NOST BORON DILUTIONS ARE ASSOCIATED WITH LEAKING VALVES, i
SPECIAL CONCERN EXISTS FOR THOSE CASES WHERE THE INDICATIONS l
OF INJECTION FLOWS CANNOT DETECT IMPROPER BLENDING OF BCRIC l
ACID SOLUTION.s IN BOTH CF (HESE CASES, THE LIMITED INSTRUMENTATION FAILED TO GIVE THE OPERAiOP A COMPLETE DESCRIPTION OF THE SITUATICH,,THUS, PREVENTING OUICK DIAGNOSIS OF THE BORON l
D I LU T I 0th, EVE N T.'
^
INP0 PLANS NO FURTHER ANALYSIS OF THESE EVENTS, BUT WILL CONTINUE TO TREND, BORON DILUTION EVENTS.
tINFORnATION CONTACT:
WARREN LIPPITT, INPO. 404/953-7615.
j 4
s 1'entrw was found.
l l
ACTION '
+
+
I E125]
Lavallee (NSAC) 12-Oct-81 1:37 PH NSAC/INPO SIGNIFICANT EVENT REPORT' 84-81
SUBJECT:
CONTROL AND PROTECTION LOSSES DUE TO RPV LEVEL INSTRUMENTATION HEAD,ER ISOLATION
REFERENCE:
UNIT:
BRUNSWICK 1 DOC NO./LER NO:
50-324/81-016 REV 1 DATE:
1/20/81 NSSS/A.E.:
G.E./U.E.1C EVENT DESCRIPTION:
The Brunwsick units have automatic isolation valves on 1
instrument lines that Penetrate the druwell or torus, rather than the excess flow check valves that are genera 11w installed on operating BWR Plants.
These reactor instrumentation Penetration (RIF) valves are air-to-open, spring-to-close valves that fail closed on loss of air, but thew have a slow bleed feature that is designed to hold them open for fifteen minutes.
Their normal air suPPlw swstem has backup conPressors that automoticallw start on loss of normal header Pressure.
Immediat7 repressuri:ation of the RIP valves' control air solenoid would ereclude their ultimate closure on initial loss of air.
The narrow-rande reactor Pressure vessel (RPV) water level instrumentation at Unit i has two sets of hwdraulic headers.
Each header erovides a common connection Point for a set of i
level instruments which Provide control and Protection functions.
An unexplained closure o# one of the RIP valves occurred in Januarv 1981 on the variable les eenetration to one of the narrow-range level instrument headers.
This caused a false level signal to the feedwater control sustem and to the control room recorder.
It also disabled the automatic high level trie of the ' main tu rbine, the feedPumFs, and the HPCI and RCIC turbines.
The false level signal first drifted low and then high over a 45 minute eeriod.
This caused actual RPV water level to first rise above the high level t**rie Point without tries occurring and then to fall and cause a low water level scram from the unaffected i
level instrumentation on the other header.
The RIP valve was reopened immediate1w after the scram, thus restoring feedwater control and the high water level trips.
Actual water level increased rapidlw and the feedeumes tripped.
Actual level then fell to the low-low level initiation of HPCI and RCIC.
RCIC failed to inject and HPCI was used to control level (See Brunswick 1 LER's81-010 and 81-014).
CPtL's response to this and other events involving RIP valvas has included RIP valv'e replacement on the narrow-range Penetrations with excess flow check valves.
H further improvement has.been the installation of two additional hudraulic headers to double the narrow-range headers to eermit seearation of control and -rotection instruments.
e NCOMMENT:
The risnificance' uf this event lies in the ability of a sinsle failure to introduce a control action wherebv a protective system will b r-needed end at the same time remove that prctective feature.
Altinough the Brunswick failure involved air operated RIP valves, other BUR's without inderendent header arransements could e:trerience similar situations caused bv maloeeration of i n s t r un.e n t s, miseositioning of instrument rack valves, or failure of PiFins.
A comparable event involving a RPV water level instrument perturbation has been evaluated by NSAC (See SER 6C-91i SOER 81 13; and NSAC/INPO Analvsis and Evaluation Report 'Hish Pressure Core Cooling System Malfunctions at Hatch l').
Althoush a similar RIP valve failure to the wide-ranse level instrumentation would not involve control and protection function interaction, it could potentially create an upset cotidition and a false oemand for safety system response --
HPCI/RCIC initiation, MSLIV closure and subseauent scram.
The loss of control air to all RIP valves due to failure of the air system could potentially result in the loss of all RPV water level instrumentation and a much more complicated transient.
This subject is being further reviewed by NSAC.
INFORMATION CONTACT:
John William Power, NSAC, (415) 855-2394.
1 entry was found.
ACTION:
F
w 4
2126]
Lavallee (NSAC) 12-Oct-81 1:43 PM.
NSAC/INPO SIGNIFICANT EVENT REPORT 85-81
SUBJECT:
FAILURE OF CORE SPRAY VALVE TO OPEN I
UNIT:
VIRMONT YI.NKEE DOC N0/LER NO:
171/91-018 EVENT DATE:
07/13/01 NS35/A.E.*
GE/EBASCO EVENT' DESCRIPTION:
Durins monthlw surveillance testing, while the Plant was at 100%
eower, it was discovered.that the motor operated sate valve in one of the core spraw lines could not be reopened from the control room becausc the motor Power sueelv breaker in the MCC was tripped.
The condition was corrected bw manual 19 reclosing the breaker.
This event is reported to have occurred because, when the valve was last closed, the control room operator continued holding the manual sprins-loaded control switch in the 'CLOSE' eosition durins and bewond the valve closing stroke action, instead of allowins its Parallel seal-in circuit and the seating toraue limit switch to control the action.
This resulted in toraue limit switch oscillation and repeated surges of motor start
' current at the end of the valve closins stroke which trieeed the motor Power sueelv breaker.
The plant operators at Vermont 4
Yankee have been re-instructed on erecautions to be observed in operation of fast actin,s MOV's.
COMMENT:
This event involved a fast acting Limitoraue operator and sate valve.
The followins features and operating characteristics of this MOV and its control circuit in this Plant are relevant to this event:
Opening of the, seating toraue limit switch contacts in the control circuit tries the motor when the valve has seated.
The sear train and worm of a fast-acting Limitoroue operator are not self-locking against reversal.
Therefore, when the motor has been trieeed bw seatins of the valve, the seating toroue limit switch contacts will auick19 re-close as the toroue sensins belleville sprins pack device in the operator unloads bw forcing reverse rotation of the worm and sear train.
The seating toraue limit switch contact Pair in the valve closing control circuit is the only device in this circuit which, in remote manual operating mode, will open the circuit if the manual switch is held in the 'CLOSE" position when the valve is seatins.
If the manual control switch continues to be held in the 'CLOSE' position, the toroue limit switch contacts will then oscillate and the stalled motor will receive repeated surses of starting current.
-A fast actinc MOV cvstem having the above ehdracteristics aaw be susceptible to rgotor burnout or actor surelv breaker trip if the oeertLor holds the manual control switch in theCLOSE' position after the valve has closed.
In an emergenew situation, such valve orerator failures could potentialls render safets swstems inoperable and delas enecution of emercencu procedures.
This rotential Problem misht be avoided bw operator awareness of the MOV control circuit design or bw chansind the circuit design.
This twee of circuit desido maw be generic to other MOV control circuits.
NSAC will continue to evaluate generic aspects of MOV n lfunctions.
INFORMATION CONTACT:
Jim Duffw, NSAC, (415) 855-2766.
1 entrw was found.
ACTION
- 8 2
P o
r i
[129]
Gillispie (INPO) l'4-Oct-81 6:1? AM NSAC/INPO SIGNIFICANT EVENT REPORT (SER) 86-81
SUBJECT:
UNMONITORED RADI0 ACTIVE LIGUID RELEASE UNIT:
JAMES FITZPATRICK DOC N0/LER N0:
333/NA EVENT DATE:
9/22/81 & 9/25/81 NSSS/AE:
GENERAL ELECTRIC / STONE & WEBSTER EVENT DESCRIPTION:
ON 9/22/81 THE WASTE CONCENTRATJR WAS INADVERTENTLY PLACED IN OPERATION WITH THE CONDENSATE RECEIVER TANK (RECEIVES CONDENSATE FROM WASTE CONCENTRATOR) ISOLATED FROM THE WASTE COLLECTOR TANN.
THIS ARRANGEMENT RESULTED IN THE OVERFILL OF THE CONDENSATE RECEIVER TANK WITH 4500 GALLONS OF LOW LEVEL RADI0 ACTIVE LIGUID BEING RELEASED THROUGH THE TANK VENT, WHICH IS CONNECTED TO AN AUXILIARY STEAM RELIEF VALVE DISCHARGE VENT GUTSIDE THE RADWASTE BUILDING.
(THE CONDENSATE RECEIVER TANK VENT IS CONNECTED DOWNSTREAM OF THE AUXILIARY STEAM RELIEF VALVE DISCHARGE LINE.)
ON 9/25/81, TORUS WATER WAS BEING PUMPED TO THE WASTE COLLECTOR TANK TO MAINTAIN THE WATER LEVEL INSIDE THE TORUS.
THE CONDENSATE RECEIVER. TANK PUMP'S DISCHARGE CHECK VALVES WERE STUCK IN THE OPEN POSITION RESULTING IN OVERFILLING OF THE CONDENSA'.E RECEIVER TANK AND CAUSING ABOUT 200 GALLONS OF LOW LEVEL RADI0 ACTIVE LIGUID TO BE RELEASED THROUGH THE TANK UENT GUT THE AUXILIARY STEAM RELIEF VALVE DISCHARGE VENT OUTSIDE THE RADWASTE BUILDING.'
XN BOTH INSTANCES, THE LOU LEVEL RADI0 ACTIVE LIQUID WAS RELEASED THROUGH AN UNMONITORED AND UNDESIGNATED DISCHARGE PATH.
HOWEVER, IN BOTH CASES, THE AMOUNT OF RADI0 ACTIVE LIQUID DISCHAPGE LIMITS WERE BELOW TECHNICAL SPECIFICATION LIMITS.
FITZPATRICK REROUT$D THE CONDENSATE RECEIVER TANK VENT TO THE WASTE COLLECTOR TANK AND REPLACED THE CONDENSATE RECEIVER TANK PUMP'S DISCHARGE CHECK VALVES.
L1MMENTS:
OTHER PLANTS WITH SIMILAR DESIGNS, SPECIFICALLY BY THE SAME AE, MAY BE VULNERABLE TO THIS UNMONITORED AND UNDESIGNATED DISCHARGE PATH FOR RADI0 ACTIVE LIQUID / GAS EFFLUENTS, INPO PLANS NO FURTHER INVESTIGATION.
INFORMATION CONTACT:
D.
D.
REDDY, INPO, 404/953-5318, 1 entry was found.
Now Joining:
Harper (FPC)
ACTION:
l.
M
[132]
Gilliseie (INPO) 19-Oct-81 8:19 AM
?
NSAC/INPO SIGNIFICANT EVENT REPORT (SER) 87-81
SUBJECT:
INADEQUATE REACTOR COOLANT SYSTEM (RCS)
WATER LEVEL INDICATION UNIT:
TROJAN DOC N0/LER NO:
344/81-12 EVENT DATE:
6/26/81 NSSS/AE:
WESTINGHOUSE /BECHTEL RELATED
REFERENCES:
ISE BULLETIN 80-12 AND IEE INFORMATION NOTICES 80-20 AND 81-09, NSAC/INPO SIGNIFICANT EVENT REPORT (SER) 78-81 EVENT DESCRIPTION:
WITH THE-PLANT IN HODE 5 CONDITION (140 DEGREES F) AFTER COMPLETING REFUELING AND REFILLING OF THE RCS, WATER LEVEL WAS DEING REDUCED AGAIN TO FACILITATE WORK ON THE REACTOR COOLANT PUMP (RCP) SEAL.
WHILE REDUCING RCS LEVEL, WITH THE RCS LEVEL STANDPIPE (TYGON TUBIN0) INDICATING APPROXIMATELY 66' (6' ABOVE CENTERLINE OF THE HOT LEG PIPING) THE RESIDUAL HEAT REMOVAL (RHR) PUMP APPEARED TO BE CAVITATING AS INDICATED BY FLUCTUATING MOTOR CURRENT.
DRAINING OF THE SYSTEM TO THE HOLDUP TANK WAS TERMINATED, THE RHR PUMP WAS STOPPED AND RCS CMARGING ESTABLISHED.
THE INDICATED WATER LEVEL DECREASED SLIGHTLY WHILE CHARGING THE SYSTEM.
IT UAS GUICKLY DETERMINED THAT THE PRESSURIZER VENT VALVE HAD NOT BEEN OPENED EARLIER A9 REQUIRED BY THE DRAINDOWN PROCEDURE.
WHEN THE PRESSURIZER VENT VALVE WAS OPENED, INDICATED WATER LEVEL DEGAN TO DROP RAPIDLY.
THE REACTOR VESSEL HEAD VENT WAS OPENED AFTER WHICH, INDICATED WATER LEVEL STABILIZED AT APPROXIMATE!Y 61'.
THE RHR SUCTION ON THE HOT LEG PIPE IS LOCATED AT THE 59' LEVEL.
PREVIOUS PLANT EXPERIENCE HAD SHOWN THAT CAVITATION UOULD OCCUR WHEN THE RCS LEVEL REACHED APPROXIMATELY 60'.
AN ATTEMPT TO RESTART THE RHR PUMPS FAILED DUE TO AIR ENTRAINMENT IN THE RHR PUMPS.
THE REFUELING WATER STORAGE TANN (RWST)
WAS LINED UP TO THE SUCTION OF THE PUMPS TC PROVIDE A POSITIVE S.UCTION HEAD TO THE PUMPS.
AN RHR PUMP WAS SUCCESSFULLY RESTARTED ONE HOUR AND FIFTEEN MINUTES AFTER INITIALLY BEING SECURED.
AVERAGE COOLANT TEMPERATURE INCREASED LESS THAN 10 DEGREES F DURING THIS EVENT.
COMMENTS:
INPO CONCERNS WITH THIS EVENT ARE AS FOLLOWS:
1)
SUB-ATHOEPHERIC (VACUUM) CONDITIONS WERE PROBABLY DEVELOPED IN THE RCS.
THIS APPEARS TO HAVE RESULTED IN THE ESTABLISHMENT CF A STEAM VOID IN THE REACTOR VESSEL HEAD DURING THE EVENT WHICH WAS NOT APPARENT TO THE OPERATORS, 2)
THE REACTOR UESSEL LEVEL INDICATION (STANDPIPE) WAS INCORRECT AND MISLEADING DUE TO THE SUB-ATMOSPHERIC CONDITIONS
.WHICH EXISTED IN THE PLANT AT THE TIME.
3)
THE SUB-COOLED 1ARGIN MONITOR WHICH UTILIZES THE IN.-CORE THERMOCOUPLES OR WIDE RANGE HOT LEG TEMPERATURE (WHICHEVER IS GREATER) AND RCS WIDE RANGE PRESSURE DID NOT INDICATE THAT A SATURATION CONDITION EXISTED IN THE FLANT.
THE SUB-COOLED MARGIN MONITOR DID NOT FUNCTION CORRECTLY SINCE IT WAS NOT DESIGNED TO OPERATE UNDER SUD-ATHOSPHERIC CONDITIONS.
4)
THE SEVERITY OF THIS EVENT COULD HAVE BEEN GREATER HAD A LARGER DECAY HEAT SOURCE BEEN PRESENT AND APPROPRIATE OPERATOR ACTION NOT BEEH TAKEN BY 1ROJAN PLANT PERSONNEL.
INPO IS CONTINUING TO EVALUATE THIS EVENT.
INFORMATION CONTACT:
ART CHARBONNEAU, INPO, 404/953-5425.
THIS SER SUPERSEDES SER 87-81 ISSUED 14-0CT-81.
1 entrs was found.
ACTION:
d
~
e,
[1343 GillisPic (INPO) 23-Oct-81 12:39 PM NSAC/INPO SIGNIFICANT EtENT REPORT (SER) 89-81
/
SUBJECT:
LEVEL INSTRUMENT OSCILLATIONS DUE TO REFERENCE LEG FLASHING UNIT:-
PILGRIM NUCLEAR POWER STATION DOC N0/LER NO:
'293/81055 EVENT DATE:
9/26/81 i
NSSS/AE:
GENERAL ELECTRIC /BECHTEL RELATED
REFERENCES:
HIGH DRYWEL'L i
TEMPERATURE EFFECT ON REACTOR VESSEL WATER LEVEL INSTRUMENTATION EVENT DESCRIPTION:
ON SEPTEMBER 26 1981, PILGRIM NUCLEAR POWER STATION WAS PERFORMING A NORMAL SHUTDOWN AND C00LDOWN.
APPROXIMATELY FIVE AND ONE HALF HOURS AFTER ENTERING THE SHUTDOWN COOLING MODE, THE CONTROL ROOM YARWAYS COMMENCED OSCILLATING (CONTROL ROOM GEMAC LEVEL INDICATORS WERE ALSO OSCILLATING BUT TO A LESSER DEGREE) RESULTING IN SIMULTANEOUS HIGH.AND LOW WATER LEVEL TRIPS AND ISOLATIONS.
THIS SCENARIO WAS REPEATED THREE ADDITIONAL TIMES AT 20 MINUTE INTERVALS.
THE CAUSE OF THE OSCILLATIONS HAS BEEN ATTRIBUTED TO FLASHING OF THE REFERENCE-LEGS DUE TO THE TEMPERATURE DIFFERENTIAL BETWEEN THE REACTOR COOLANT AND THE DRYWELL AMBIENT TEMPERATURE IN THE VICINITY OF THE REFERENCE LEGS (REACTOR COOLANT 220 DEGREES F, DRYWELL
{-
240 DEGREES F).
THE HIGH DRYWELL EMPERATURE WAS THE RESULT OF DEGRADED DRYWCLL COOLING DUE TO FOULED COOLER COILS AND REDUCED EFFICIENCY OF THE REACTOR BUILDING CLOSED COOLING WATER (RBCCW) SYSTEM DUE TO BYPASS FLOW (LER 294/81049 AND IE BULLETIN 81-03, FLOW BLOCKAGE OF COOLING WATER TO SAFETY SYSTEM COMPONENTS BY ASIATIC CLAMS AND MUSSELS).
COMMENTS:
A RECENT LICENSING CONSTRAINT WHICH PROHIBITS PURGING THE DRYWELL UNTIL THE REACTOR IS OUT OF THE RUN MODE MAY HAVE CONTRIBUTED TO THIS EVENT SINCEr IN THE PAST, THE PURGE PROCESS ASSISTED IN THE REDUCTION OF THE DRYWELL TEMPERATURE.
AT PILGRIM, THE DRYWELL COOLERS ARE PRESENTLY BEING REFURBISHED TO IMPROVE DRYWELL COOLING SYSTEM PERFORMANCE.
ALSO, IN AN ATTEMPT TO RESOLVE T.iE MUSSEL BUILDUP PROBLEM IN THE RBCCW SYSTEM, A PROGRAM OF CHLORINE ADDITION COUPLED WITH INSPECTIONS, IS NOW IN PL ACE.
ADDITIONAL THERMOCOUPLES ARE BEING-INSTALLED IN THE DRYWELL TO MONITOR ACTUAL ZONE TEMPERATURES.
INPO IS CONTINUING TO ANALYZE AND EVALUATE THIS EVENT.
INFORMAT:0N CONTACT:
DICK BAKER, INPO, 404/953-7616.
1 entry was found.
ACTION:
J
4 L135:
L_v:1I9e (NSAC) 23-Oct-81 4:46 PM
- iCNPO ? I SilIF I C AN T EVENT REPORT 90-81
?j5 JECT.
HIGH OCCURRENCE OF DESRADED HYDRAULIC SNUBBERS PEFERENCE:
Ut,! T :
URUNSWICK 2
- JC tJO/LEP NO:
324/81-041 REV. 1 OATE:
3/4/31 HSSE/A.E:
GE/UE & C EVENT DESCRIPTION:
BRUNSWIch UNIfS 1 & 2 HAVE EXPERIENCED A NUMBER OF PIPING SUPPORT ANCMALILES IN RECENT YEARS.
IN MARCH 1981, UtlIT 2 BEGAN A PROGRAM TO TEST SAFETY RELATED HYDRAULIC SNUBBERS IN ACCORDAriCE WITH TECHNICAL SPECIFICATONS.
A TOTAL OF 640 SNUBBERS WERE REMOVED FROM SERVICE AND FUNCTIONALLY TESTED.
130 FAILED THE TESTS AND WERE REBUILT.
AN ADDITIONAL 80 SHUBBERS WERE REBUILT FOR PREVENTATIVE MAINTENANCE REASONS DUE TO LOW NARGIr1 FAISING THE TOTAL NUMBER OF REBUILT SNUBBERS TO 210.
THE REASONS FOR REBUILDING THE 210 SNUBBERS INCLUDED:
45% LOW BLEED; 17% HIGH LOCKUPi 13% NO LOCKUPi 13% HIGH BLEEDi AND 12%
OTHEff.
THE DEGRADED CONDITIONS OF THE SNUBBERS' COMPONENTS WERE nLSO DETERhINED FROM THE TESTS AND INSPECTIONS AND ARE LISTED IN THE FOLLOW [NG TABLE.
THE SNUBBERS WERE FREQUENTLY FOUND DEGRA3ED IN MORE THAN ONE WAY.
% OF REBUILT SNUBEERS WHICH HAD DEGRADED CONDITION WORN PGPPETS 62 SPRING CAPTURE 36 PISTON / CYLINDER WEAR 31 DETERIORATED SEALS 28 GREASE IN FLUlD 7
SIDE LOADIN3 7
NONE FOUND 8
AN ANAL MIS WAS ALSO PgRFORMED TO DETERMINE IF SYSTEM OPERATION HAD A DEGRADING EFFECT ON THE SNUBBERS, I.E.,
DETERMINE IF THE SNUMBERS WERE ' SERVICE SENSITIVE.'
OF THE SNUBBERS REBUILT, 37%
UERE DETERMINED TO BE SERVICE SENSITIVE.
THE MAJORITY OF CAUSES FOR SERVICE SENSIlIVE SNUBBERS WAS ATTRIBUTED TO VIBRATION WITH n NOMINAL NUMBER DUE TO WATER HAMMER.
WORN POPPETS WERE IDCNTIFIED IN THE ABOVE TABLE AS THE PRIMARY DE3RADATICN MODE OF THE SNUBBERS.
THIS UAS ATTRIBUTED TO THE PDPPE"3' LUW RESISTANCE TO PIPE VIBRATION.
REPLACING THE VALVE EODIEC u!TH A NEW, MORE VIBRATION RESISTANT ONE IS EXPECTED TO LESSEN THIS ANOMALY.
THE SECONDARY MODE OF SPRING CAPTURE IS ALSO L :PECTED 10 BE LEGSENED BY REPLACING THE VALVE BODIES.
A 200!. RD'L " C EME t! ! 0F VALVE BODIES WAS SCHEDULED FOR UNIT l'S JUNE S1 OUTALE.
REPLACEMENT WAS SCHEDULED For: A FUTURE CUTAGE AT UDIT 2.
DERGIN-PATTERS 0ll SUPPLIED THE ORIGIrJ AL AND REPLACEMENT SNUebERS AND ACTIVELY PARTICIPATED IN THE INSPECTION FROGRAM.
l
,4, u r1.1 E n T.
r::IS SER Is SEING PROVIDE 0 DEC A'JSE OF THE NUMEROUS SNUBBER
^-
"JEnR;DATIOMC ENCOUNTERED AT DRUNSulCK AND TO NOTE THE EXTENSIVE REFURI:ISHMENT FROGRAM UNDERfALEN AS A RESULT.
OTHER BWR FACILITIE3 ARE EXPERIE'NCING S I M I'_ A R PROBLEMS.
H'/DF AULIC SNUDBERS ARE A t1 IMPORTANT PART OF THE STRUCTURAL fiE S I t:N AND PERFURtiANCE OF SAFETY SYSTEMS' PIPING.
ALTHOUGH A P R t ri A R / F U ti C i I O N OF 5.NUBBERS IS SEISMIC RELATED, DEGRADED 3tiU E HE R PERF0EhANCE tiAY ALSO POSE A THREAT TO A PIPING SYTEM'S INTI3NITY UNDER NORMAL LOADING AND OPERATING CONDITION 5.
DPFRIENCE AT BRUNSWICK APPEARS TO INDICATE THAT EXTERNAL VISUAL ItiE.PECTION OF THE IN-PLACE SNUBBERS tiAY NOT REVEAL OPERATIONAL OR PHY3ICAL DEGRADATIONS.
THIS S'JBJECT IS BEING FURTHER REVIEWED BY NSAC.
INFORMATION CCNTACT:
JOHN WILLIAM POWER, NSAC, 415/855-2394.
1 entrw was found.
Now Joinins:
Seers (AEP)
ACTION:
k I
I I
I I
I I
I I
l I
f I
i s.<-
A
[1383 Lavallee (NSAC) 26-Oct-81 S:35 AM NSAC/INPO SIGNIFICANT EVENT REPORT 91-81 SU3 JECT:
STEAM VOIDING IN THE REACTOR Cn0LANT SYSTEM DURING DECAY HEAT REMOVAL COOLDOWN
REFERENCE:
UNIT:
CRYSTAL RIVER-3 DOC /LER N0*
302/NA (IE CIRCULAR 81-10 DATED JULY 2, 1981)
DATE:
APRIL 21, 1981 NSSE/A.E.:
B&W/ GILBERT ASSOC.
OCONEE-1 269/NA (DUKE LETTER TO NRC REGION II DATED JULY 31, 1981)
JUNE'29, 1981 B&W/ DUKE EVENT DESCRIPTION:
On the above dates, while operating in mode 5, Crwstal River-3 and Oconee-1 developed steam voids in their respective reactor coolant sustems (RCS).
Both Plants were in Mode 5 in the final stages of Plant Cooldown, With reactor Coolant Pumps (RCP's) secured and decaw heat being removed by the residual heat renoval (RHR) sustem.
In this mode of heat removal, continuous flow is maintained through the core between the GHR connections with little or no circulation of coolant in the loops.
Reactor
(,
coolznt was then removed from the RCS via the letdown Path of the volume centrol sustem.
This draining, without RCP flow, caused the misration of hot pressuriner water to the
'A' loor hotles via the surde line.
System pressure reduction a,llowed this essentialls stagnant volume of hot water.to flash to steam.
In both events, a steam bubble of aperoximatels 300 cu.
ft. was formed in the top of the
'A' looe
'J'-les.
This void extended down both sides of the
'J' les about ten feet and filled about half of the uPeer hemispherical head of the
'A' once thronsh steam senerator (OTSG).
At Crustal River-3, RHR was initiated and RCP's secured at an RCS temperature of 270 degrees F, and a Pressure of about 250 Psis.
The Plant was furt 5er cooled and depressurized over the next 15 hours1.736111e-4 days <br />0.00417 hours <br />2.480159e-5 weeks <br />5.7075e-6 months <br /> to 106 degrees F, and 50 Psis.
During this deeressurization, the injection of auxiliarv serav and comperisating c o o l e r.t letdown created a slow Pressuriner outsurse of greater that. 360 destees F water into the
'A' hotles.
Grerators then turned off Precsuriner heaters and continued Pres:uriner auxiliaru sPraw, cooling the Pressuriner and dePressurizind the Plant below saturation Pressure for the temperature of the stannant water volume at the top of the
'A' loup.
The expanding void caused a rapid pressuriner insurse of about 100', indicative of a 300 cu. ft. void formation.
Pressuriner heeters were restored, spraw was terminated and emersenew feedweter flow was initiated to the
'A' OTSG, in an attempt to cool the
'A' hotles.
A similar event occurred again at Crvstal River on 1 Oct 1981.
o M
e At Oconee-1, RHR was initiated and the last RCP secured at en R;S tempersture and.:ressure of 225 decrees F and 310 Psis recsectively.
Pressurizer tempersture was 423 degrees F, and Pressurizer level was 250 inches.
A Procedural stcP had been ouit8ed which tvouired reduction of Pressuriner level to 100" Prior to turning off the last RCP.
Subseauentlvr Pressuri:er level uas 1cwered to 100" to fulfill this reouirement.
This RCS Ictdown moved a larde volume (4000-5000 sal.) of stasnant 423 degrees F water from the pressuri:er into the
'A' hotled.
Operators then turnad off Pressurizer heaters to cool the P r e s ', u r i c e r and depressurire the plant.
After about 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> of FieSSuri:Pr CGQldoWM, swstem Pressure Was lowered below saturation Pressure for the hottest water in the
'A' loop (120 vsis, 350 desrees F).
A rapid Pressuriner insurse of about 100" occurced as the hotled void formed.
Pressurizer hesters were reenerci ed to repressurice the RCS and compress the void.
Addition of feedwater via the emergenes feedrind to hish OTSG levels (scester than 85%) combined with operation of the turbine bypass system also induced coolins of the void area.
Durins both events, RHR coolins of the core was continuouslw m a i n t a i r.e d.
Also, operators at both Plants immediatelv recognized the void formation sumptoms and initiated corrective action.
COMMENTS:
IE Circular 91-10 discusses the Crustal River event, and a similar event at McGuire Unit 1 which ocurred r rior to achievind initial criticalitv.
The dircular discusses the potential for fluid stagnation and stratification, and for retarded coolins of Irrte masses of hot metal (e.s. vessel head) durins RHR ecoldown.
It recommends that utilities conduct an cPeratinc Procedure review to assure sufficient information exists on void fermetion svmetoms and recovers actions.
Void formations in RCS loops are Primarily of concern becsuse under different elant conditions such as a station blackout scenario, they can'erevent natural circulation cooling and thus remove the affected steam generator as a viable heat sink.
These events are also of concern because thew crn become e threat to reliable decas heat removal.
Had oeerstors not recodni<ed sod Proper 19 diadnosed the s v a.p t o m s, void formation uvuld have become nore entensive.
Void formation under the vesrel heed is a possible consequence of this scenario, esPecially if RHR flow is lost.
Significant a::ial temperature iifferences can exist between the recirculating fluid in the essel core region and the relatively stagnant, usualls hotter v
'lu i.i in the RV hezd.
D e '.e r m i ni n d the enset location and entent of void formation can be difficult because of Placement, tweer and anount of existing instrumentation.
Most Plants do not have temperature deteClors in areas must euseertible to void formation (i.e. RV head arer, Loc u#
- J' led, tcP of
'O' tubes).
In one event, oFeTators celied un LSe hutled R "' reedirHs to verify the loor subcoolind mardin.
W :. t h o u t flou, the temr e retu r e of the water at the tue of the
'J' led uas about 45 degrees hotter than the cooler water eda went to the RTDr less the n, ten feet avau.
.,All PUP'; cePeer to be susceptible to this earticular cethod of void formation, although void location will varv.
Hot water r i o s., e Pressurizer tut-surse on a B&W el.snt' will migrate to the tce of the
- A*
luce
'J' l' e s.
On a Westinghouse or CE elant, hot cater from a Pressurizer outsurge could migrate to the RV head area oc in some cases to steam generator
'U' tubes.
Some stees which can reduce the erobabilits of void formation fcon eressuriner outsurge include:
A.
Reducing pressuriner level and loop temperature to low levels (e.g.,
less thar. 100' and less than 200 degrees F) erior t, o securins the last RCP (RCP min. NPSH and temperature limits eermitting).
Normal RHR operation, thoroush coolant mixing, and approximate metal / water thermal eoutlibrium should be achieved Prior to turning off the last RCP.
B.
After RCP's are secured, eressurizer hot water outsurge can be minimized 69 maintaining a constant or slowlw increasing pressuriner level as the pressurl:er is cooled.
If this Pressuri=er cooldown is conducted bw turning heaters off and minimizing auxiliars seras, Pressuriner recirculation and hot water migration can be reduced.
C.
Shifting from saturated steam eressure control to nitrogen control in the eressurizer Prior to securi.ng RCP's is cn option which might be considered in some cases.
Maintaining subcooling margin with N2 eressure while cooling down the pressurizer with RCP's still running is a reliable method of avoiding void formation from this Particular scenario.
Additional insight into void formation and elimination can be gelned from review'of NSAC-16/INPO-2 Report of December 1980:
'Analssis and Evaluation of St. Lucie Unit 1 Natural Circulation Cooldown'.
System repressurization and/or coolant addition, if necessare, are emehasized as the Priorite recovers actions after void formation.
MSAC will continue,to analene these events.
The advantages and disadvantages of restarting an RCP are being analwzed.
INFORMATION CONTACT:
Gars Vint NSAC, 415-955-8903.
This EER supercedes the previous SER on this subject (23-Oct-81) which had an incorrect report number.
1 >> n t r w was found.
ACTION: