ML19329A134
| ML19329A134 | |
| Person / Time | |
|---|---|
| Site: | Oconee  |
| Issue date: | 12/06/1976 |
| From: | Stello V Office of Nuclear Reactor Regulation |
| To: | Heineman R Office of Nuclear Reactor Regulation |
| Shared Package | |
| ML19329A126 | List: |
| References | |
| NUDOCS 7912300154 | |
| Download: ML19329A134 (8) | |
Text
~
ENCLOSURE l'*
)[]
uNivassTATas p.,;.
f#~ *%,
NUCLEAR REGULATORY COMMISSION y
WASHINGTON, D. C. 20555
($ x)3
?
s,**vfe DEQ 0 61976 i/
R. E. Heineman, Director, Division of Systems Safety, NRR MEMORANDUM FOR:
V. Stello, Jr., Director Division of Operating Reactors, NRR FROM:
OPERATING EXPERIENCE MEMORANDUM NO.
SUBJECT:
FLOODING OF BASEMENT FLOOR IN TURBINE BUILDINGS PROBLEM Partial flooding of the Oconee Nuclear Station's turbine building basement resulted in a water level of two feet in the center of the basement floor and approximately 16 inches at the basement walls.
PRESUMED CAUSE Oconee, Unit 3 was shutdown for re' fueling a'nd"maintena'nce. The Unit 3 turbine condenser was isolated, for inspection, by closure of the six 78-inch manually operated inlet valves and the six 78-inch air Three-inch fackscrews (strong-piston operated butterfly outlet valves.
back) were installed on each of the six butterfly valves to prevent inadvertent opening or drift. Three manways had been removed from the condenser so that an inspection of the condenser tubes could be A loss of AC power to the air controllers for the six performed.
butterfly valves resulted in the introduction of service air to the One of.the six actuator cylinders tending to open these valves.
strongbacks failed to prevent the opening of one of the six condenser outlet valves, resulting in the backflow of water from Lake Keodee to G
4 7912300 6 p
u 8
R. E. Heineman -
DEC06IW5 the condenser through the failed 78-inch outlet butterfly valve.
Flooding of the turbine building, through the open condenser manways, continued for a period of 32 minutes until the plant personnel were able to restore AC power and close the failed valve.
Approximately 200,000 cubic feet of water had entered the turbine building during the 32 minute interval.
This corresponds to a depth c* 16 to 21 inches of water in the turbine
~
building, depending on the location. An electrically driven oil pump for one of the three emergency feedwater pumps and a hotwell pump became inoperable as a result of the flooding.
It is important to note that the turbine building basement floor is at elevation 775 feet (MSL) and t'.e full pond water level,in Lake
,1 Keoweeis800. feet (MSL). This configuration is considered unique to the Oconee Station since in most plant layouts the turbine building basement would be at an elevation higher than the source of circulat-l ing water thus precluding this flooding mode.
This occurrence was caused by a number of errors.
First, a procedural inadequacy was responsible in that the procedures did not require the closing of the air supply valve to the butterfly valves air actuators which would have prevented valve motion.
Secondly, the design of the butterfly valves is such that they will fail open upon loss of power to the four-way solenoid con-trol valve. Thirdly, the jackscrews were not designed to with-stand the forces developed by the valve air actuators.
R. E. Heineman.
DEQ 0 61976 SAFETY SIGNIFICANCE The turbine building, which is 790 _ feet long, is comon to all three units of the Oconee Station. The auxiliary building, which is also common to all three units, is connected to the turbine building via six doors at the lowest elevation. A 21-inch high curb around these doors was provided The to afford some protection against flooding of the auxiliary building.
LPI, HPCI, and auxiliary service water pumps are located in the auxiliary building. As mention previously, loss of one electrically driven oil pump for one of the three emergency feedwater pumps and one hotwell pump had occurred as a r'.v it of the flooding.
If failure of two or more hot-well pumps, on any unit were to occur, it would result in the tripping of he booster and main feedwater p.as i.e., loss of all feedwater to the affected unit, with the exception of ti e emergency feedwater pumps which k-are located at a low elevation on the turbine building basement floor I
and one auxiliary service water pumq also located in the auxiliary building.
Since the emergency feedwater pumps are steam driven, it
.' A < u is difficult to determine at what flooding level would have become inoperative.
Since all the hotwell, booster, and main feedwater pumps are
~
located on the turbine building basement floor, it appears that the simultaneout loss 'of feedwater to all three units could have occurred if the flooding of the turbine building had not been terminated.
If the water level reached 21 inches a,t the walls, water then would have I
I s
,v
O,9 f.h illi.'_2 E DEC 0 619/6 R. E. Heineman.
flowed into the auxiliary building which would have eventually rendered the high pressure coolant injection and the auxiliary service water pumps inoperative. These are the only sources of water to the reactor vessel and the steam generator, respectively.
Loss of the above cited pumps would have removed the only provisions for heat removal from the reactor coolant system.
Portions of the condenser circulating water system piping are not seismically qualified.
Consequently, even if all condenser isolation valves were to function as required during a seismic event, a single passive failure of any one of the six 78-inch pipes immediately downstream of the condenser outlet valves could cause unacceptable flooding for all three Oconee units.
While this issue was reviewed in the past, it appears that the review was limited primarily to evaluating the consequences of failure of the expansion joints in the condenser cooling water system near the condenser waterbox inlet and outlet nozzles without the consideration of a coin-cident failure of one of the non-seismic category I condenser circulat-ing water isolation valves.
REPAIR The pumps which were rendered inoperative were disassembled, cleaned i
and tested satisfactorily.
No other equipment or component damage was experienced.
In order to preclude recurrence of this incident Duke Power states that the-following actions will be taken:
I u
t ar.
j l
R. E. Heineman a DEQ06 W8 1.
The present pilot solenoids on the condenser discharge valves will be replaced with dual-coil, mechanically latched types. This will increase reliability because:
a) Latched solenoids do not change state on loss of control power.
They require electrical powe only when actually changing states, b) Condenser discharge valves controlled by latching solenoids will therefore fail "as-is".
Power failure will not initiate spurious operation of the discharg2 valves.
'c) Latched solenoids avoid co.stiauous coil energization which should increase operating life.
d) Administrative " blocks" of condenser discharge valve aperation can be implemented by the tagging out of control power.
(
2.
The power source for condenser circulating water (CCW) controls will be changed to an ICS power panelboard. This wi11 increase reliability because:
a), Normal power is still derived from a battery backed static inverter.
b) Automatic transfer to a backup power source (regulated power) is made on failure of the normal source. The transfer is made with-out interr.uption of power to the load and without operator action.
3.
Position indicating lights will be added in the control room for the condenser discharge valves. This will help the operator monitor system operating status because:
- % w.s 1
s
(,
i, t
4 Ml R. E. Heineman -
DEC 0 61976 a), The lights will provide additional system status information to
' the existing board displays.
b) Position status of all discharge valves can be determined at once without referring to a printout.
c) The lights will serve as a backup to the computer documentation
/
of valve status.
The local control stations for the condenser discharge valves will be 4.
relocated further from the condenser and on the protected side of a
'colum.
The physical layout of electrical cabling and pneumatic tubing in the 5.
l vicinit) of the condensers will be reviewed to ensure adequate protec-tion from damage by water force.
k 6.
A review will be conducted to determine the feasibility of raising the instrumentation lube oil pump and cooling water pump for the emergency feedwater pump.
7.
The procedures for opening the CCW system inside the Turbine Building are being reviewed and revised as necessary to require that:
a) Inlet and outlet CCW pipes are vented if all CCW pumps are shut-down.
b) Manually operated valves at condenser CCW inlet are closed, tagged d
and mechanically locked.
c) Air to pneumatic piston valves be manually blocked and the piston b'e vented.
l l
l l.
I l
DEC 0 61976 il 1 }ru{I Ej R. E. Heineman i
Screw jacks be in place at the CCW discharge valves to prevent d) valves from drifting open.
The emergency condenser discharge valve be locked closed to the j
e) i gravity drain system.
A design review of the station for susceptability to similar type 8.
flooding incidents and possible means of providing additional assurance of the operability of equipment important to safety is in progress.
- DDR finds these precautionary measures acceptable on an interim basis contingent on prompt implementation of the above cited actions.
DOR ACTION DDR has had a meeting with the Duke power Company to discuss this The licensee has formed occurrence and the proposed correction action.
~
k' a task force which is actively reviewing this occurrence and other potential flooding modes to determine what modifications are necessary to preclude unacceptable flooding levels, regardless of the source of The task force is expected to complete its review and to flooding.
submit a design proposal, within the next four months, for DDR review.
D0R will be reviewing all operating facilities to determine if similar
[
occurrences could arise.
RECOMMENDATIONS 90R recommends that all applications for construction permits and operating licensees be reviewed to preclude this problem.
)
- H j. '-
s Jr Ar.I Pr.
12
- w n,,..,
is
- " v' '. s r.
f*
a,,,I !.'....<.~
n i
r < : -
- t. e,.
acc9
.i
.- c.,.
c,
....e....
u o t s /# / w: J : ~<
S Ar + /c l' e...
i
,e
.1
DEC 0 61976
(
n hj,
R. E. Heineman.
' Principal DOR Personnel Reviewer:
B. C. Buckley, F. Clemenson Project Manager:
D. Elliott Victor Stello, Jr., Director Division of Operating Reactors Office of Nuclear Reactor Reg 01ation
Contact:
B. Buckley, DDR 492.8077 s
.\\
s
, ome
.