ML20084B243
| ML20084B243 | |
| Person / Time | |
|---|---|
| Site: | Quad Cities  |
| Issue date: | 07/03/1972 |
| From: | Riesland J US ATOMIC ENERGY COMMISSION (AEC) |
| To: | Skovholt D US ATOMIC ENERGY COMMISSION (AEC) |
| Shared Package | |
| ML20084B161 | List: |
| References | |
| NUDOCS 8304060202 | |
| Download: ML20084B243 (6) | |
Text
_
. [%h f N)N,'
UNITED STATES ATOMIC ENERGY COMMiddlON
. (Ve&}(
1
- /.L.
WASHING TON. D.C.
20545
\\LQg/
mi9 JUL 3 1972 Donald J. Skovholt, AD/OR, DL TIIRU:
U. A. y:mann, Chief, ORB #2, DL
/pgL Zi ws EVALUATION OF C0( DENSATE PU:!P ROO>f FLOODING AT QUAD-CITIES UNIT 1 DOCKET No. 50--2$~4 On June 9,1972, Quad-Cities Unit 1 condenser circulating water out1ct pipe rubber expanulon joint broke with three circulating water pumps in operation resulting in flooding of the condensate pump room to a level of 15'-6" above the floor and the conderser room to a level of 4' above the floor.
The flooding occurred over a period of about 5 ninutes and involved about 600,000 gallons of water.
The rubber expansion joint rupture resulted fron the discharge valve slamming shut during maintenance on the hydraulic system. The reactor had been shut down for 8 days and required no shutdown cooling. Unit 2 in the startup node at 22 MWt holding 22 psig and was scraimacd was when the control room operator was notified of the probica.
Major equipment in the condensate pump room that became inoperabic were:
. 4 RHR service water pumps 2 dicac1 generator cooling water pumps (#1 and #1/2) 4 condensate-condensate booster pumps 3 condensate transfer pumps scismic detector 4 equipment and floor drain sump pumps hyperchlorite sample analyzer and sample pucp circulating water valve hydraulic power unit The flood water was sanpled'and analyzed prior to pumping to the discharge canal and the radioactivity was no greater than background and well within 30 CFR 20 limits.
1 Temporary corrective measures to prevent reoccurrence were implemented i
after being evaluated and accepted by the Station Review Board and the Nucicar Review Board.
Both review boards concurred in the temporary correction and authorized both units to resume operation.
The corrective action included scaling of the condenser room from the condensate pump
(
room with barriers to prevent, water from entering the condensate pump i
i 1
8304060202 720901 POR ADOCK 05000 S
I 9
0
.s_
Ibnald J. Skovholt JUL 3 1972 room from the condcaser room. A Icvel alarm was to be installed under the condenser that will annunciate in the control room.
The work has been completed and both units are now in operation.
Evaluation The RilR service water pumps are required for two modes of cooling:
(1) normal plant shutdown, and (2) emergency cooling of the suppression pool in the event of a LOCA.
There are four (4) RIIR service water pumps for each unit. One pump and one heat exchanger can accommodate DBA suppression pool coolinC to maintain the temperature of the pool below 181*F based on an initial pool ternperature of 95*F.
At a suppression
- ) dol temperature of 125*F, which will occur in a period of less than
' hours after the start of decay heat rejection to the pool, a temperature t
alarm signals the operator to start a RilR service water pump.
The FSAR shows, in FI ure 6.1-1 of Amendment t;o. 17, that under these C
conditions the pool terperature will remain below 180*F and adequate tipSl! is availabic for the core spray and 1.PCI pumps with containment pressure.up to 1.5 psig.
The accident raode resulting in loss of all four RilR service water puraps has not been analyzed and is not addressed in the FSAR and, therefore, la considered an unreviewed saiety question.
Initial analysin indicates that the June 9 abnormal occurrence coincident with the Joss of the main condenser for the unit would likely have resulted in the follwing conditions had the reactor been operating at maximum licensed power level:
s 1.
Loss of condenser vacuum would have tripped the turbine and scrammed the reactor.
2.
The high pressure transient would have opened the relief valves to discharge, steam to the suppression pool.
3.
'1he lipCI or RCIC system would have been placed in operation, discharging steam through the turbine to the suppression pool and supplying condensate from the condensate storage tanh
,s while maintaining reactor pressure.
4.
About 5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> af ter start of the llPCI or RCIC, approximately 70,000 gallons of condensate would have been discharged through thu reactor to the suppression pool at an average rate of
(
about 200 gpm (makeup rate is 90 gpm) td accommodate decay heat. Without RilR service wa ter, the pool temperature would rise f rom 95'F to 165*F during this period.
./
{
O O
Donald J. Skovholt JUL 3 1972 5.
When the pool temperature reaches 165'F, the LpCI pumps and core spray punps would lose Upsil and become inoperable unicas the containment were pressurined.
If reactor pressure is blown down to less than 100 psi, the lipCI and RCIC would become inoperabic.
At this time all design modes of core cooling for accident conditions would have been lost.
The above conditions did not take place during the occurrence because the unit had been shutdown for 8 days prior to the occurrence and did not require shutdown cooling.
loss of the cooling water pumps and motors for the Unit 1 diesel cencrator and the sparc Units 1 and 2 dicsc1 generators would have prevented operation of the ECCS equipment upon loss of offsite power.
Since flooding and subsequent inoperability of the engineered safety features was due to rupture of a system not designed to perform safety functions, a number of deficiencies uith respect to the General Design Cri teria for nuclear pouer plants were discovered.
These are:
_ Single failure criterion:
A single failure of a passive component results in a loss of capability of essential systems (ECCS and onuite power) to perform their cafety function.
(Multiple failurea resulting from a single occurrence are considered to be a single failure.)
s Criterion 16:
Reactor co6tainment design conditions could have been exceeded and uncontrolled radioactive relcace to the environs would have been expected to occur.
Criterion 17: Onsite.clectric power sources would not have been available for systems and components important to safety because the dicsc1 generatorn 1 and 1/2 cooling watcr pumps werc
~
flooded.
Criterion 33: Reactor coolant makeup would have been lost approximately 5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> after the occurrence.
Criterion 34:
Residual heat removal capability was lost uhen the RilR service water pumps were flooded..
Criterion 35:
Emergency core cooling capability would not have been availabic about 5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> after the occurrence.
4 A
o o
Donald J. Skovholt JUL 3 EU2 Criterion 38: Containment heat rem' oval capability would not have been availabic after the occurrence, even with offsite power available.
Criterion 44:
Cooling water to transfer heat from structures, systems and components inportant to safety was not available from the RiiR service water and the diesel generator cooling water systems.
Criterica 50:
Containment design basis could have been excceded eventually if the reactor had been in operation at the time of the occurrence. Additional studies are required to determine what methods (venting, contain-ment ventilation, etc.) could have been used to maintain conditions within the containment design margins.
In addition to the above, the occurrence demonstrated the need for physical separation of redundant systems conponents important to safety.
In the Guad-Citics Units design, the diesel generator cooling water pumps for Unit 1 and the shared Units 1 and 2 are located side-by-side in the condensate pump room as are the RllR service water pumps for Unit 1 and Unit 2.
Also, these components are located in the same room with the condensate system pumps and piping which are not of Class I design.
Rupture of the 20" condensate line during a seismic disturbance could flood the roca and di' sable equipment essential to safety.
The licensee has submitted a 10-day report of this abnormal occurrence, s
as required by the Technical Specifications, which is considered inadequate regarding corrective action.
The interim corrective action was reviewed and approved liy the Station Revicu Board and the Nuclear Review Board; ooth permitted resumption of operation.
The corrective action partially isolated the condensate pump. room from the condenser room, but the occurrence and corrective measures have not been evaluated adequately by the licensee to provide assurance that the event will not reoccur.
The licensee agreed to discuss the final corrective ncasures at a meeting
.s in Bethesda prior to issuing a final report on this matter by July 17, 1972.
During a telephone discussion with Mr. D. J. Skovholt on June 17, the licensec agreed to discuss the planned corrective measures, interim and final, at a meeting in Bethesda as soon as we desired.
Subsequently, we requested this meeting take place on June 23, but we were advised by the licen'see that he would not be prepared for, such a meeting for at
(.
least 2 additional weeks during which time he will continue station operation.
In a telecon on June 23, A. Giambusso stressed the urgency of an early resolution of the issue and obtained a commitment from the
-3 n
J
O L
O 1
Donald J. Skovholt JUl.
3i9f$
licensee to the effect that a corrective action proposal and schedule would be provided to the Commission by June 26.
On June 26, again by telecon, the licensec advised J. Riesland generally what corrective measures would be taken and that a construction period of about 4-1/2 months would be required to completc them. At the request of J. Riesland, during the telecon, the licensec agreed to presenting their proposal, schedule and interim actions in a meeting on June 30.
This nceting will be in room p-422 starting at 9:00 AM with participation by Commonwealth Edison Company, General E1cetric Company, Sargent and Lundy, Licensing and Regulatory Operations.
Since our evaluation indicates that a severe scismic event could Icad to an unsafe condition of the station, after the condenser rooms and condensate pump rooms are scaled to prevent water flow into the condensate pump room from the condenser room, reasonabic assurance must be provided at the earliest possibic time that a scismic cvent will not damage components or systems important to safety.
The Quad-Citics Station is located in Zonc 1 of the Coast and Geodetic Survey's "Scismic probability Map of the United States" which catagorizes i
this as a zone of minor damage from a scismic disturbance.
In Richter's Scismic Regionalization Map, Quad-Cities is located between zones VII and VIII. The design ground acceleration factor recommended by the licensec's consultants (John A. Blume Associates Engineers) and used in the design for Class I systems is 0.12 g.
They recommended Class II equipment be designed for a ground acceleration of 0.10 g, but the licensec's Class 11 equipment was not constructed to sny specific seismic criteria.
Conclusion l
The proposed interim correct 1ve measure of nealing against yater flow
~
into the condensate pump room has not been shown to be of Class I design.
The adequacy of the interim corrective actions is of concern r,
in consideration of continued operating during the time that final corrective measures are being impicnchted, even allowing for the low probability of serious consequences from a scismic event.
Based on i
the very low probability of a major scismic disturbance, consideration should be given to permitting plant operation for a minimum period necessary to correct deficiencies in the plant design and construction
(
discussed above.
The time required to accomplish this work and the design corrective measures, based on probabic scismic event and damage consequences, must be justified by the licensee to permit continued station operation until the final corrective measures are completed.
t W
(,
/
Donald J. Skovholt JUL 3 1972 Because the 10-day report in inadequate to properly evaluate the event, additional information is required to complete our evaluation.
This information will be provided orally at a meeting with the licensee on June 30,1972, in a written report to be transmitted by July 9,1972, and in a detailed design and construction schedule for the corrective measuren.
Based on the evaluation of this information, regulatory should make a decision whether to permit continued plant operation or shut the facilitics down pending completion of satisfactory correctivp action.
(J.J 's e John I Riesland Operating Reactors Branch #2 Directorate of Licensing 0
4 1
9 s
e
/
I i
r.
e I
L e
e i
\\
y
~