IR 05000440/1993006
| ML20035G037 | |
| Person / Time | |
|---|---|
| Site: | Perry  |
| Issue date: | 04/14/1993 |
| From: | Lanksbury R, Martin T NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III) |
| To: | |
| Shared Package | |
| ML20035G033 | List: |
| References | |
| 50-440-93-06, 50-440-93-6, NUDOCS 9304260055 | |
| Download: ML20035G037 (30) | |
Text
7
.
.-
U. S. NUCLEAR REGULATORY COMMISSION
REGION III
l Report No. 50-440/93006(DRS)
Docket No. 50-440 License No. NPF-58
Licensee:
Cleveland Electric Illuminating Company Post Office Box 5000 Cleveland, OH 44101 facility Name:
Perry Nuclear Power Plant Inspection At:
Perry Site, Perry OH Inspection Conducted:
March 27, 1993 - April 2, 1993 Inspectors:
R. B. Landsman, DRP J. F. Schapker, DRS J. G. Guzman, DRS
,
'
G. P. Hornseth, NRR A. Vegel, Resident Inspector - Perry C
Approved By:
( bm 4/Nh3 l
Roger D. L'h bC D'ats-Team Leader TOded '
9 /mlu Approved By:
T. O. Martin, Acting Director Date Division of Reactor Safety Insoection Summary inspection on March 27. 1993. - Aoril 2. 1993 (Report No. 50-440/93006(DRS)
Areas Inspected:
Special Augmented Inspection Team-(AIT) inspection conducted in response to the service water pipe break event at Perry Nuclear Power Plant on March 26, 1993. The review included validation of the sequence of events, evaluation of the root cause for the pipe break, review of the service water system's performance and maintenance history, evaluation of operator response to the event, evaluation of the effects of flooding, evaluation of the
!
licensee's event classification and reporting, and evaluation of the licensee's corrective actions to the December 1991 circulating water pipe break.
Result s: No violations or deviations were identified in any of the areas inspected. No significant operational safety parameters were approached or i
exceeded.
Four radial breaks in the pipe were identified.
Two were about
'
5-inches apart and appeared to be the location where the pipe rupture occurred. Two secondary pipe breaks,' one on either side of the rupture 9304260055 930415 PDR ADOCK 05000440 G
L_
l
-
!
e
.
l location, were also identified. The team concluded that the root cause of the
!
-
failure was an initial perforation that slowly grew and ultimately resulted in
'
the pipe rupture. Several potential root causes for development of the
)
initial perforation were identified. However, the failure analysis effort was
hampered by the fact that substantial portions of the failed pipe were lost at l
the time of pipe rupture. There was evidence that the pipe leaked for a
'
I significant time period prior to the rupture. No pipe joint was involved in the failure. The team did believe that the cause of the pipe rupture was localized and was not the result of general system wide degradation.
The local failure mechanism concept appeared to be supported by the absence of widespread cracking which would result from excessive bending or other gross loads during service.
Visual examination by the team of exposed portions of the pipe, and the portions of the pipe removed during excavation, indicated that no other significant degradation appeared to have occurred. The secondary pipe breaks appeared to be the result of unsupported soil and
,
'
asphalt caving in after the water flow was stopped.
The team concluded that the operators responded in an excellent manner and
that their actions were indicative of a strong knowledge of plant systems and procedures.
!
h
,
!
!
l l
l
i
!
l
'
.
!
'
.
l.0 Introduction 1.1 Event Sumary On March 26, 1993, at about 3:22 p.m. (EST) a nonsafety-related 30-inch
>
fiberglass pipe carrying service water (SW) from the SW pump house to the Unit I turbine building catastrophically failed approximately 13-feet
underground causing the asphalt covered ground to heave up.
The break was located in the west plant yard approximately 50-feet south of the water treatment builaing. Water from the pipe break flooded the western portion of the site and entered various plant buildings causing minor flooding (up to 6 l
to 8-inches) of areas inside the plant.
The entry points were primarily through electrical conduits connected to flooded electrical manholes in the vicinity of the pipe break. Secondary entry points were from flow under various doors on the exterior of the buildings. Peactor operators commenced a fast reactor shut down and manually scrammed the reactor from about 66 percent
,
power at approximately 3:26 p.m.
Under the licensee's emergency plan an Alert i
was declared at 3:35 p.m. due to flooding. The SW leak was stopped approximately 16-minutes after the pipe rupture when the operators stopped the SW pumps. The plant was placed in cold shutdown at 10:10 p.m. on March 26, and the Alert was terminated on March 27 at 1:05 a.m.
'
The resident inspectors responded to the event. A subsequent review by the
!
residents and licensee personnel indicated that no safety-related equipment was affected by the flooding.
Some water entered the high pressure core spray (llPCS) pump room, apparently by dripping down through the pump maintenance
>
access hatches above the pump, and a small amount of water conveyed through electrical conduits entered the emergency service water pump house. No water
!
was observed in any of the remaining rooms containing emergency core cooling
!
system (ECCS) equipment.
Some nonsafety-related equipment, such as the offgas
,
system glycol control panel, was affected.
Radioactive contamination of basement floor areas in several of the site buildings resulted from flood I
waters flowing through existing contaminated areas and from floor drains backing-up.
1.2 Augmented Inspection Team (AIT) Formation Region 111 staffed the Incident Response Center (IRC) and headquarters personnel monitored the event.
Senior NRC managers determined that an AIT was l
warranted to gather information on the SW pipe break.
On Saturday, j
March 27, 1993, an AIT was formed consisting of the following personnel:
Team Leader:
R. D. Lanksbury, Chief, Reactor Projects Section 38,
"
Division of Reactor Projects (DRP)
l
.
.
Team Members:
R. B. Landsman, Project Inspector, Section IA, DRP J. F. Schapker, Reactor Inspector, Materials and Processes Section, Division of Reactor Safety (DRS)
A. Vegel, NRC Resident Inspector, Perry Nuclear Power Plant, DRP J. G. Guzman, Reactor Inspector, Materials and Processes-Section, DRS G. P. Hornseth, Materials Engineer, Materials and Chemical Engineering Branch, Office of Nuclear Reactor Regulation (NRR)
The team leader and two of the team members were on site by the evening of March 27. 'The full team was on site the morning of March 29.
In parallel with formation of the AIT, Region III issued a Confirmatory Action Letter (CAL) (Attachment 1) on March 30, which confirmed certain actions in support of the team.
1.3 AIT Charter A charter was formulated for the AIT and transmitted from Edward G. Greenman to Roger D. Lanksbury on March 27, 1993, (Attachment 2) with copies to appropriate EDO, NRR, AE0D, and Region III personnel.
The AIT was terminated on Friday, April 2,1993.
2.0 Description of the Event 2.1 Service Water System Description
)
The purpose of the nonsafety-related service water (SW) system was to provide cooling water to various auxiliary mechanical equipment throughout the turbine, auxiliary, and radwaste buildings, and the control complex.
The system was capable of removing heat given up by various nonsafety-related heat exchangers including the turbine building closed cooling (TBCC) heat exchangers, the turbine lube oil coolers, and the nuclear closed cooling (NCC)
heat exchangers. The system design included four one-third capacity vertical wet pit pumps, each with a total discharge head (TDH) of 140-feet (nominally 60 psig) at a design capacity of 23,500 gpm. SW flow was through a once-through open loop piping network where lake water was pumped through the tube side of the heat exchangers being cooled and was then returned to the lake by way of discharge tunnel return lines.
The system also supplied makeup water to the cooling tower basins and to the screen wash pumps.
Piping throughout the system was either carbon steel or fiberglass reinforced plastic (FRP) and is non-seismic category I.
FRP piping was used exclusively in the portions of the system which were installed underground or outdoors and was constructed to meet the requirements of the American Society of Chemical Engineers (ASCE) Manuals and Reports on Engineering, Practices #37, for plastic pipe and the American Society of Mechanical Engineers (ASME) Boiler and Pressure Vessel Code,Section X.
SW flow was routed from the SW pump house to the plant through an underground 54-inch diameter FRP pipe (see Figure 1 for a general SW piping arrangement).
The fiberglass supply lines
L
"
!
!
-
were designed to Line Specification P16-7 which had a design pressure of 100 psig.
Normal operating pressures were in the 40 psig range.
As the 54-inch FRP pipe approached the plant, a 30-inch FRP pipe branched off at 90 degrees to direct flow to the Unit 1 turbine building.
It was this line i
that experienced the break. The 54-inch pipe then transitioned to a 48-inch FRP pipe and then to a 42-inch FRP pipe that supplied flow to the control complex. A branch line from the 48-inch pipe that went to Unit 2 had been sealed off.
The FRP piping transitioned to carbon steel piping at the building penetrations.
Piping for the SW system inside the plant was carbon steel. As the SW piping exited the plant, it converted back to FRP piping.
Underground FRP piping was used to return the SW to the discharge tunnel. The lower pressure FRP
'
discharge lines were designed to Line Specification R16-7 which had a design L
pressure of 50 psig. The operating discharge pressure was normally about 9 psig.
2.2 Sequence of Events At 1:14 p.m. (EST) on March 26, 1993, with the plant at 100 percent reactor power, plant personnel reported water coming up from the perimeter of a concrete slab on which a trash compactor and dumpster were sitting and a concrete slab for one of the underdrain system manways. The surrounding areas were covered with asphalt. The concrete slabs were located in the area south of the water treatment building. Witnesses initially reported water bubbling up approximately 4-inches. The water was coming from the asphalt-to-concrete joint around three sides of the concrete slabs. After discovery, the licensee attempted to determine the source of the leak by stopping flow through the underground piping systems known to be in the area (emergency service water (ESW) and condensate transfer) with the exception of service water (SW) which could not be isolated during plant operation. Stopping flow through ESW at 1:28 p.m. showed no affects to the water leaking from the pavement.
Subsequent plant walkdowns showed no-intrusion of water in the plant. At 2:47 p.m., isolation of condensate transfer system underground piping showed no effects on the leakage flowrate.
Plant walkdowns were still reporting no water intrusion. Water continued bubbling up from the pavement and the licensee concluded that the water was the result of a SW leak and that a plant shut down would be required. At 3:22 p.m., low SW discharge pressure (24 psig) was noted and the fourth (of four) SW pumps was started in accordance with plant procedures. At about this same time, people in the vicinity of the leak observed the asphalt heave up with a substantial increase in the amount of water coming from the area.
At 3:25 p.m., the Shift Supervisor ordered a fast reactor shut down in accordance with off normal instruction (ONI)-P41 (Loss of Service Water) due to the rupture of a 30-inch SW line. The fast reactor shut down reduced reactor power to 66 percent and a manual scram was initiated at 3:26 p.m.
Plant emergency instruction (PEI)-B13 (Reactor Pressure Vessel Control) was entered when reactor vessel water level dropped to level 3 (178-inches above the top of active fuel). At approximately 3:27 p.m. the main turbine was manually tripped.
At about this same time plant personnel provided the first
.
.
reports of water intrusion into the plant.
By 3:30 p.m., water had entered the service, intermediate, diesel, radwaste, turbine, offgas, and auxiliary buildings as well as the control complex.
An Alert was declared at 3:35 p.m. based on flooding and at 3:38 p.m. the SW I
system was shut down. The motor driven feedwater pump was also started at l
3:38 p.m. in accordance with the reactor scram procedure (ONI-C71-1).
Reactor level control was shifted from the feedwater system to the reactor core isolation cooling (RCIC) system at 3:41 p.m. due to the impending loss of the condensate system. By 3:52 p.m., both residual heat removal (RHR) pumps "A" and "B" were operating in the suppression pool cooling mode to remove the heat loads added by the RCIC system. At 3:55 p.m. reactor recirculation (RR) pump
"B" was shut down.
At 4:12 p.m. water levels in the plant were reported to be decreasing and were down to 2-inches in the control complex from a high of 6 to 8-inches.
PEI-B13 was exited at 4:20 p.m. and integrated operating instruction (101)-7 (Cooldown l
Following Reactor Scram, Main Condenser Available) was entered. At 4:45 p.m.,
reactor pressure control was shifted from the bypass valves to the safety l
relief valves (SRVs) and the operators began closing the main steam isolation
'
valves (MSIVs) and steam line drain valves. At 4:47 p.m. the
"C" and "D" outboard MSIVs were closed. At 4:48 the inboard
"C" and "D" MSIVs and the inboard and outboard "B" MSIVs were closed. At 4:53 p.m. the inboard and l
outboard "A" MSIVs and the inboard steam line drain valves were closed. The
!
initial SRV opening for pressure control occurred at 4:51 p.m. and main condenser vacuum was broken at 4:55 p.m.
A total of 10 SRV openings occurred during the event. The low condenser vacuum resulted in an MSIV isolation signal at 4:58 p.m.; however, all the MSIVs and the inboard drain valves had already been manually closed.
l r
i A high suppression pool level (18-feet 6-inches) resulted in entering PEI-T23
'
i (Containment Control) at 5:10 p.m..
At 5:29 p.m.,101-6 (Cooldown following i
'
Reactor Scram, Main Condenser Not Available) was entered. The SW system shutdown resulted in high turbine building closed cooling (TBCC) system heat l
exchanger outlet temperatures at 6:19 p.m. and high nuclear closed cooling
'
(NCC) system heat exchanger outlet temperatures at 7:39 p.m.
Equipment that was normally cooled by SW through these heat exchangers was either secured by
[
the operators, or cooling was transferred to the emergency service water (ESW)
system. At 7:08 p.m. indications of high hydrogen concentrations in the l.
drywell and containment were detected.
PEI-M51/56 (Drywell and Containment
!
Hydrogen Control) was entered and at 7:12 p.m. the hydrogen igniters were t
i energized. At 8:15 p.m. shutdown cooling using RHR "A" was established.
l l
'
RCIC tripped at 8:18 p.m. on a reactor water level 8 trip and level control was transferred to the control rod drive system. At 8:45 p.m.,
RR pump "A"
!
was secured.
,
i The plant was placed in cold shutdown at 10:10 p.m. on March 26 and the Alert
was terminated on March 27 at 1:05 a.m.
Af ter analyzing samples of the
'
dr.ywell and containment atmospheres and determining that a hydrogen problem
'
did not exist, the hydrogen igniters were de-energized at 1:25 a.m.
l t
l i
l
'
i
.-
2.3 Precursors to the Event
!
!
Prior to and at the time of this event, the reactor was at steady state i
100 percent power with no plant evolutions in progress that could impact on
the SW system. The SW system had been operating steady state since
March 25, 1993, when-SW pumps were shifted.
SW pumps were routinely shifted
every 2-weeks. Based on the team's review of plant logs and interviews with i
operators, no ongoing activities that could have been precursors to the pipe i
rupture were identified.
j
!
2.4 Operator Response
!
To determine what actions the operators took in response to the event and the
!
suitability of these actions, the team reviewed plant logs, the Post Scram f
Restart Report (1-93-01), appropriate plant emergency and off-normal
'
procedures, and interviewed the operators involved in the event.
!
!
Prior to the event on March 26, 1993, the reactor was at 100 percent reactor
!
power with no major evolutions or plant transients in progress. The operators j
in the control room were cognizant of the water leak adjacent to the water
!
treatment building and were involved in determining the source of the leak.
,
At 3:22 p.m. (EST) a service water (SW) pump discharge header pressure low
!
alarm was received and a fourth SW pump was started in accorknce with off
!
normal instruction (ONI)-P41 (Loss of Service Water). At 3:25 p.m., the shift l
supervisor at the scene reported that the SW leak had increased dramatically, l
and ordered the Unit Supervisor to shut down the plant. A fast reactor l
shutdown was conducted. Core flow was lowered to 52 million pounds mass per
>
hour and the reactor was manually scrammed from approximately 66 percent-
<
reactor power. As a result of the scram, an expected reactor water level 3 i
j actuation was received which resulted in the operators entering plant
!
emergency instruction (PEI)-B13 (Reactor Pressure Vessel Control).
By l
3:30 p.m.,. reactor water level was stabilized and a cooldown was commenced i
utilizing the reactor core isolation cooling (RCIC) system for level control
[
and steam bypass valves for pressure control.
!
i Concurrent with plant cooldown and emergency plant activities, actions were in
'
progress to minimize the impact of the SW pipe rupture.
Plant. equipment i
normally cooled by SW through the nuclear closed cooling or turbine building f
closed cooling systems were secured or realigned.
The condensate and
feedwater systems, and reactor water cleanup system, and one of the
!
recirculation pumps were shut down. Control complex chillers were transferred t
to the emergency-closed. cooling system, and spent fuel pool heat exchangers
!
were aligned to be cooled by the emergency service water system.
In addition,
-
both residual heat removal (RHR) systems were lined up for suppression pool
!
cooling in anticipation of temperature increases due to RCIC operation.
l
!
Approximately 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> after the reactor scram, the steam supply to the main i
turbine gland seals was lost due to the steam seal evaporator having no i
condensate makeup supply. As a result, condenser vacuum was eventually lost.
Prior to breaking condenser vacuum, reactor pressure control was transferred to the safety relief valves (SRVs). With the loss of condenser vacuum, a main
'
steam line isolation signal was received, closing the outboard main steam line
,
l
!
,
'
}
.
drain valves.
The main steam isolation valves and the inboard drain valves were already closed.
At 5:10 p.m.,
suppression pool level reached 18-feet 6-inches, requiring entry into PEI-T23, (Containment Control). The rise in suppression pool level was i
expected due to RCIC and SRV operation. At 7:08 p.m., indications of high
,
hydrogen concentrations in the drywell and containment were received.
Hydrogen analyzer "A" was reading 2.5 percent hydrogen concentration in the drywell head and analyzer "B" was reading 1.5 percent concentration in the containment dome. As a result, PEI-M51/56, (Drywell and Containment Hydrogen Control), was entered and the hydrogen igniters started.
Subsequent samples
,
measured actual hydrogen concentrations of 0.03 percent in the drywell and
'
O.05 percent in containment. The apparent cause for the initial indications of high hydrogen were hydrogen analyzer equipment problems and did not appear to be related to the flooding. At 8:15 p.m., shutdown cooling was established and the plant was subsequently placed in cold shutdown.
Though safety systems were not affected, important nonsafety-related systems, l
including the instrument air compressors and the recirculation pumps, were impacted. The operators took actions to secure or provide alternate cooling methods to equipment to prevent them from overheating and getting damaged.
Due to prompt and effective action, no significant equipment problems were caused by the loss of SW. The operators referenced all of the appropriate ONIs and integrated operating instructions (101s) in responding to the event.
The overall adequacy of procedures and effectiveness of operator actions to mitigate the consequences of the event were excellent. One problem did occur due to operator misinterpretation of a step in procedure ONI-P41. This resulted in the premature securing of both condenser hotwell pumps. The resultant loss of condensate flow resulted in steam jet air ejector exhaust steam being discharged into the offgas system without being condensed. As a result, a potential for damage to the charcoal beds existed.
In addition to the impact on the offgas system, operators in the field noted minor water hammer transients due to the loss of condensate flow. Based on post-event walkdowns, no significant damage to plant equipment was noted.
As discussed above, three PEls were entered during the event.
Based on discussions with plant operators and review of the PEls, no deficiencies were identified in the effectiveness of the PEls to guide the operators in keeping the reactor vessel and containment in a safe condition during the event.
Based on review of this event, the team determined that operator response to this event was prompt, effective, and in accordance with plant procedures.
The operators quickly evaluated the indications and took prompt action to place the plant in a safe condition. Specific strengths included the following:
Good communications and teamwork between operators in the field and in
the control room enabled the shift supervisor to properly evaluate and deal with the event.
i
'
,
-
-
-
-.
-
-
- +
-
,
/
,
t
.
The operators were proactive in responding to changing plant conditions,
!
as reflected in early use of the RCIC system in anticipation of the loss
.
of feed and condensate, and the placement of both trains of RHR in l
suppression pool cooling mode in anticipation of temperature increases l
due to RCIC and SRV operation.
e Good use of procedures assisted in the prioritization and combating of
'
individual equipment problems.
The team concluded that the operators safely responded in an excellent manner to the event and that their actions were indicative of a strong knowledge of
plant systems and plant operating procedures.
l
!
3.0 Event Classification and Reportina i
The licensee classified the event as an Alert in accordance with emergency action level initiating condition K.II.1, accident or hazard which indicates an actual or potential substantial degradation in the level of plant safety.
,
The emergency plan specified that if an event causes visible damage to, or in
,
the judgment of the emergency coordinator (in this case, the shift supervisor)
i threatens safe shutdown equipment, an Alert be declared.
Based on the shift supervisors observation of the rupture initiation, and numerous reports of water intrusion into various plant areas, the Alert was declared at 3:35 p.m.
,
Upon Alert declaration, the Technical Support Center (TSC) and the Operations
Support Center (OSC) were activated. Upon being notified of substantial water ingress into the control complex 599-foot elevation and reports of water in the OSC and water restricting access to the TSC, the emergency coordinator
.
selected alternate facilities. The OSC was relocated to the rear access
!
facility and the TSC was transferred to the Emergency Operation Facility in the Training and Education Center (TEC) within the owner controlled area,
approximately 1.5-miles from the plant.
Subsequently, the OSC and TSC were declared operational at 4:00 p.m. and 4:30 p.m. respectively. The Alert was terminated at 1:05 a.m. on March 27, 1993, upon stabilization of plant i
conditions and establishment of a recovery plan.
The team reviewed the specific circumstances surrounding the event and
!
discussed the event with plant operators and other licensee personnel, i
including the licensee emergency planning staff. Based on this review, the i
team concluded that declaration of an Alert was appropriate, the decision to stay in the Alert for 9.5-hours was conservative, and that required
!
notifications were made in a timely manner.
The licensee held a critique of
Emergency Preparedness during the event and the results will be reviewed in a j
later inspection.
-i
<
,
}
!
!
!
.
__
,
.
l
-
l 4.0 Inspection Results i
4.1 Service Water Piping l
4.1.1 Performance History and Maintenance Since 1985, there have been five unrelated failures in the SW system. An
,
inspection of the SW piping had never been performed because the SW system was
constantly in use, including during outages. The five failures consisted of-I
!
a.
Three failures occurred on the 6-inch SW strainer blowdown line.
i These were:
a leaking joint, a cracked section due to the January 31, 1986, earthquake, and a cracked section apparently
caused by careless work during a previous repair effort.
!
'
,
b.
Three cracks on the 54-inch main header leading out of the SW pump l
house. These were the result of the pipe being struck by a i
backhoe during previous careless repair work on the 6-inch line.
!
c.
A leaking joint on the 42-inch line south of the service building i
due to inadequate preparation of the joint during original installation.
I Preliminary inspection of the SW pipe adjacent to the failed location and the portions of pipe removed during excavation indicated that the system had not
suffered general degradation.
4.1.2 Material Condition of Affected Piping
'
The failure analysis effort was hampered by the fact that substantial portions of the failed area were lost at the time of the rupture.
Pieces from the failed area were broken off and washed away during the flooding that followed the rupture. Although the licensee recovered numerous pieces from the ground j
,
surrounding the eruption site, it appeared that many had been lost in the numerous storm drains in the area. However, the team ascertained that the
cause was local to the failure location and was not the result of general I
system wide degradation.
i Smoothly worn surfaces caused by erosion of the fiberglass, at what appeared
{
to be the initial failure location, indicated that the pipe leaked for a i
significant time period prior to the catastrophic failure. The failure
'
location was through a patch on the inside of the pipe. The patch was a i
single layer of fiberglass approximately 5-inches wide and of some
'
undeterminate length (possibly around the full inside circumference of the
!
pipe). The edge of the patch was not smoothly tapered into the base material.
!
This would tend to create a stress riser on the inside of the pipe. Several l
other smaller patches were seen on the inside and outside of the accessible i
portions of the pipe.
Based on the materials, workmanship, and lack of any
damage under the patches, it appeared that these patches were performed by the l
pipe manufacturer to repair areas containing voids in the resin. No pipe j
joint was involved in the failure.
i i
!
.
.
.
!
!
.
<
The team's investigation considered the following as a possible contributing cause to the pipe failure. The affected pipe crossed the location of the edge of the original foundation excavation. This was a transition from undisturbed
'
natural soil into deep, compacted, clay backfill. This difference in soil compaction may have resulted in differential settlement which could have imposed a bending stress at the break location. This stress may have imposed an increased load on an already weakened location, which by itself may have been within allowable stress levels for the material. After removal of the
'
broken sections of pipe, the licensee compared the elevations for the remaining pipe ends and determined that the section of pipe remaining in the fill material was about 5-inches lower than the pipe remaining in the natural material. The difference in elevations supports the differential settlement theory; however, this may also represent original construction elevations.
The team postulated the following sequence of events leading to the rupture.
.
Assuming a leak existed for months or longer, the water from the small leak would probably have been carried away by the free draining sand backfill placed around the pipe. Over time the leak increased, as evidenced by a smoothly eroded surface around the hole at the failure location. As the size of the eroding hole increased, stresses on the pipe increased.
Eventually,
local tearing rapidly increased the size of the leak.
This caused water to
,
come to the surface because it could no longer be carried away by the sand.
As clay overburden and sand bedding material were being carried away, the flow increased causing more soil erosion. When the licensee noticed. a drop in SW
!
pressure, a fourth SW pump was put in service to increase press'ure. This increase in pressure may have caused the pipe break size to increase or l
possibly caused the complete separation. Recognizing the failure of the SW pipe, the licensee shut down all the pumps in the SW system. The pressure of the escaping water probably initially supported the overburden. When the pumps were shut off, the support was lost for the undermined soil and asphalt paving above the pipe. This resulted in it caving-in.
The cave-in caused two
secondary pipe breaks to occur; one approximately 13-feet and one approximately 16-feet on either side of the original failure.
,
The local failure mechanism concept appears to be supported by the absence of widespread cracking which would result from system wide distress during service. Visual examination of exposed portions of the pipe revealed no other
significant degradation or damage. There was some shallow cracking at the top j
inside of the pipe. This is a common condition in buried fiberglass pipes and l
is not a serious condition.
Fiberglass pipe is a brittle material with roughly the same ductility as cast
iron. As a result, it is susceptible to cracking, instead of bending, in a -
,
limited area as a result of impact or a local overstress condition.
Local
!
damage can exist which would result in a failure, such as the one experienced,
,
even though there is no overall system distress.
Because this material is
'
susceptible to isolated cracking at random locations, it is not possible to
-
judge the overall system condition based upon a limited inspection.
The local damage causing the initial leak may have been the result of one of I
several causes, including:
'
!
T
l
!
{
!
.
.
.
a.
Impact damage during construction b.
Contaminated backfill containing sharp debris c.
Service induced damage resulting from debris in the pipe d.
Failure of the patch 4.2 Flooding 4.2.1 Amount of Water and Flood Path I
Water flooded the ground surface surrounding the SW pipe break and entered the plant primarily through electrical manhole (EM) number 1 at the northwest corner of the radwaste building.
EM 1 was covered with concrete plugs that had gaps of 2 to 3-inches. These gaps readily allowed water to flow into the manhole.
i Discharge from the break ceased when the SW pumps were shutoff approximately 16-minutes after the start of the fourth SW pump.
(The fourth SW pump was started at about the same time of the pipe break.) The licensee estimated
'
that the outflow rate during the 16-minutes approached 89,900 gpm for a total leakage volume of approximately 1.71 million gallons.
Approximately five percent of the total leakage (85,000 gallons) entered the plant and reached the following plant locations by way of EM 1 and by flowing under various roll-up and access doors on the west side of the plant (Note:
all elevations are with respect to sea level, with elevation 620-feet being
ground level):
(1)
Water and silt on floors of the:
Control complex at elevations 599-feet, 574-feet, and 564-feet
Auxiliary building at elevations 620-feet, 599-feet, 574-feet, and 568-feet e
Intermediate building at elevations 620-feet, 599-feet and 574-feet (2)
Water and sand in the turbine power complex at elevation 620-feet, 593-feet and 548-feet.
(3)
Water and silt on the floors of the radwaste truckbay at elevation 620-feet.
(4)
Water in the offgas building at elevation 620-feet and in it's lower levels.
(5)'
Water and silt on the floor of the turbine building laydown area.
(6)
Some leakage into the water treatn.ent building at ground level.
(7)
Water on the floor of the main entrance lobby of the service building.
(8)
The emergency service water pump house (ESWPH) floor was wet and covered with silt.
(9)
The SW pump house lower level had water and silt resulting from a failure of a SW screen wash pump casing. Also, water and silt were noted in the electrical mezzanine floor of the SW pump house.
_
_
_
.
,
(10) Water in the Unit 2 auxiliary building at elevation 568-feet.
f The external surface water paths included (see Figure 2 for a map of the flooded areas):
t South:
The water ran south past the diesel generator fuel storage
'
tanks to all low points in the yard and into storm drains.
The water ran past the service building annex approaching i
the Unit 2 turbine building laydown area.
'
North:
The water ran north up the road area near the maintenance
building. The water also came in contact with the water treatment building and turbine building to the north.
West:
The water ran west past the security fence, northwest to the lake and southwest into and past the primary access control point (PACP).
I East:
The water ran east and came into contact with the offgas, turbine power complex (TPC), auxiliary, radwaste, diesel generator, and service buildings.
4.2.2 Design of the Underdrain System The underdrain system beneath the plant was constructed to prevent excessive uplift pressures from developing beneath the buildings as a result of an extremely high natural ground water table. The system consisted of a porous
,
concrete blanket with a circuit of 12-inch diameter porous concrete pipes beneath the buildings to collect and convey the ground water to manholes, and two systems to remove the water from the manholes and discharge the water to the lake.
One discharge _ system used pumps located in the manholes, and the other used
'
higher elevation pipes to convey the water by gravity to the lake. The pumped
!
discharge system was designed to convey water through the porous concrete
!
blanket and pipes to collection manholes. The water was then discharged from
.
the manholes by submersible pumps, maintaining the water level between
!
elevation 566 and 568-feet.
j The gravity discharge system, which consisted of 36-inch to 48-inch pipes connecting the manholes, was some 20 to 25-feet above the underdrain blanket.
It provided an alternate flow path for drainage in the event of a complete
failure of all the pumps. This system ensured that the water level never
-
exceeded elevation 590-feet.
It incorporated a gravity outfall and was
"
designed to handle a total flow of 60,000 gpm for two units, 30,000 gpm for each.
4.2.3 Water Level flood Design i
The site grading and storm drainage system were designed to preclude subjecting seismic category I structures to water levels greater than 6-inches
<
F above plant grade of elevation 620-feet. Assuming the worst case (i.e.,
&
_
.
~.
.
.
complete blockage of the site storm drainage system and using peak discharge from the most intense hour of the probable maximum precipitation (PMP) the plant site was graded so that overland drainage would occur away from the site buildings and the resulting ponding elevation of 620-feet 5-inches would have no adverse affect upon safety class equipment because the floors at plant grade are set at elevation 620-feet 6-inches.
Portions of safety class structures located below finished grade were protected on their outside surfaces by a continuous waterproofing membrane.
Also, waterstops were provided at construction joints. Additionally, flood
protection for safety class components, equipment and systems located below grade were provided with a floor drain system that would handle leaks of lesser relative magnitude.
.
The design criteria for ensuring the prevention of damage to safety-related equipment by internal flooding caused by a pipe rupture in a moderate energy system, such as the SW system, were:
a)
Plant layout uses separation of seismic category I and non-seismic category components by locating them, where possible, in separate l
buildings.
- b)
Emergency core cooling system (ECCS) equipment was located in separate, water tight compartments.
'
c)
Small leaks were handled by the floor drain system.
Prevention of internal flooding damage had been analyzed in the Updated Safety l
Analysis Report (USAR) and conformance to the analysis is discussed later.
4.2.4 Conformance with the Updated Safety Analysis Report (USAR) Assumed Magnitude and Path
The USAR design basis accidents (DBAs) assumed for the underdrain system were:
(1) a yard break in the circulating water pipe outside the plant near the steam tunnel and auxiliary building, or (2) failed expansion joints occurring
-
inside the turbine building through flow from a fracture in the building base mat.
!
The underdrain system (including the pumped discharge subsystem and the gravity discharge subsystem) was designed to handle the total volume of water released during the DBAs and maintain the underground water level below l
elevation 590-feet.
)
The paths that the water took during the SW pipe break were not specifically i
considered in the USAR underdrain system analysis; however, it was bounded by
both of the underdrain system design basis floods. The internal and external flow resulting from the 30-inch SW line break did not challenge the capacity I
of the underdrain system.
14
)
\\
-
l i
'
(1)
External l
The outflow of the SW break ran off following the slope of the adjacent ground j
surface and either slowly seeped into the ground, was collected by the catch
!
basins located throughout the site for the purpose of collecting storm runoff,
!
or ran into Lake Erie at the northwest corner of the site. The ground seepage j
rate was extremely low due to the 3-feet of impervious fill placed over the
.;
site to reduce infiltration at the ground surface. The resulting inflow from the seepage into the underdrain system would be much lower than for_the design i
basis in-ground circulating water (CW) system pipe break.
Flow into the
.
underdrain system from the underdrain system manholes was not a concern since l
all the manholes were closed as required. Flow resulting from seepage, and any flow through the gasketed manholes, was disposed of by the underdrain
'
system backup pumps which were in operation.
i (2)
Internal
A sizable portion of the water that flowed into the plant entered through EM
1.
Four sets of electrical penetrations were routed through EM 1 at the I
northwest corner of the radwaste building and terminated at the ceiling of
,
control complex elevation 599-feet. Water also entered the plant by flowing i
under various roll-up and access doors on the west side of the plant.
l The vast majority of the water that entered the plant was collected in the
!
radwaste system collection tanks through the plant drain system.
'
Corrective action taken after the December 1991 CW system pipe break precluded flow into or out of the plant by the underdrain system piezometer tubes.
Plant procedures were changed to require that the piezometer tube covers be i
replaced after use of the tubes.
4.2.5 Effects of Flooding
!
The flooding caused by failure of yard piping was found to result in
!
conditions that do not jeopardize safe plant shutdown or adversely affect l
operation of safe shutdown systems.
l Auxiliary Buildina i'
A maximum of 5-inches of standing water was reported'on elevation 568-feet.
Water depths of less than 20-inches on the lower elevations of the auxiliary building would not compromise the operability of safety-related components.
.
An indeterminate level of flooding on elevation 599-feet resulted in leakage
[
into the high pressure core spray (HPCS) pump room through the ceiling concrete hatch plugs. This ingress.resulted in water dripping on the. HPCS pump motor. The motor was subsequently inspected and meggered and no abnormal
. conditions were found.
'j Intermediate Buildina flooding of less than 6-inches in the lowest level of the intermediate
building would not threaten the operation of any safety-related equipment.
In-
!
the case of the SW pipe break, water levels of up to 5-inches were reported on
!
r
$
-.
-
-
)
.
.
elevation 574-feet. Due to the heavy silt content of the flocJ water, the drains appeared to have backed up.
Control Complex (CC_1 l
Equipment required for safe shutdown or for maintaining control room
habitability was located 6-inches above elevation 574-feet.
In the case of the SW pipe break, water level of up to 5-inches were reported on this elevation. The source for leakage into the CC was electrical manhole (EM) 1.
Four sets of electrical penetrations were routed through EM 1 at the northwest corner of the radwaste building and terminated at the CC 599-foot elevation ceiling. Junction boxes (JB) 12503 (Unit 1, Division 1 and 3 cabling) and JB 2473 (Unit 2, Division 3 Cabling) as well as open condui. penetrations for
'
Unit 1, Division 2 cabling and Unit 2, Division 2 cabling were the major sources of water intrusion at the CC 599-foot elevation. Temporary plugs
existed on open conduits but they were forced out by the head of water from
the flooding. The water on elevation 599-feet leaked through floor plugs at that elevation and flowed through the southwest stairway onto elevation 574-feet.
!
Emeraency Service Water Pump House (ESWPH)
-
Water entered the ESWPH as a result of flooding in the yard area through
conduits at the southwest corner of the pump house.
The water sourse for this path was through EM 1 at the northwest corner of the radwaste building. The ESWPH floor was wet or covered with silt over most of the area.
Additionally, the motor fire pump controller was wet on the surface but was not damaged.
Water was also found in an inoperable Unit 2, Division 2 motor control center l
(MCC). No safety-related components were affected.
Turbine Power Comolex (TPC)
)
The turbine power complex elevation 593-feet had water entry from three empty 6-inch penetrations originally slated for fire protection piping. During the
,
flood, water ran on grade level between the interbus transformers and the TPC
!
south wall and was able to flow into the penetrations. Along with water, Grade A fill (sand) also came out of the three penetrations.
- Water also entered the TPC at elevation 593-feet from a penetration that was
,
connected to EM 8.
Water ran down the wall and fell on to fire damper
IM36F571 and subsequently fell on condensate filter control panel IH51P014.
Floodino in Other Plant Buildinas
The offgas and radwaste buildings, the SW pump house, and other detached l
structures contained no components essential to safe shutdown.
No flood protection was required for any of these areas.
\\
4.3 Equipment Problems As stated earlier, no safety-related equipment failures resulted from the SW I
pipe break or from the subsequent flooding. However, various nonsafety-related equipment problems were experienced during the event:
,
.
I f
.
.
.
The glycol skid in the offgas building had water flowing on it that caused erratic operation and instrument indication problems. A work order was issued to investigate for possible damage.
,
The hydrogen analyzers generated what appear to be false elevated readings.
As stated elsewhere, these high readings do not appear to be related to the flooding. The hydrogen analyzers were scheduled to be recalibrated.
As discussed elsewhere, the offgas system charcoal beds may have been damaged due to the discharging of steam into the offgas system.
The south SW screen wash pump upper casing split during the event. The root cause of the casing failure was still under investigation.
The "A" control rod drive (CRD) drive pump experienced minor cavitation /waterhammer due to loss of suction during the event.
<
4.4 Radiological Releases No radiological releases occurred as a direct result of the SW pipe break and subsequent flooding. The 85,000 gallons of water estimated to have leaked into various site buildings was contained within those buildings. Any potentially contaminated water was sent to the radwaste system collection tanks through the floor drains. The licensee took environmental samples at a number of locations, including the major stream, northwest discharge drainage, minor stream skimmer, and at the sanitary sewer lift station, all with no detectable activity.
Sediment samples from the northwest impoundment and the Unit 2 intermediate building sump also resulted in no measurable levels of activity being detected.
One unmonitored gaseous radiological release did occur during the event. The release occurred on March 27, 1993, and was the result of the "B" auxiliary boiler becoming contaminated. When the plant shut down, it was necessary to provide an alternate source of heating steam.
This was accomplished by use of
,
one of the two auxiliary boilers. When the licensee started the auxiliary boiler, the boiler was initially vented to atmosphere. This provided a vent
'
path for the contamination to atmosphere, along with steam. The licensee
indicated that this was the third such occurrence since October 1992. As of
'
this inspection, the licensee had not been able to determine the cause for the auxiliary boiler becoming contaminated and was continuing their investigation.
The initial estimate by the licensee on the amount of the release was a total of 22.25 micro-Curies. This would have resulted in an insignificant exposure at the site boundry.
The team concluded that the gaseous radiological release was minimal and well l
below regulatory limits.
.
5.0 Safety Sianificance
The consequences of the event posed no threat to public health and safety.
A
minor gaseous radiological release occurred but was not directly attributable
!
- to the event.
The release was minimal and well below regulatary limits.
,
e i
,
'
-
Internal plant flooding was limited. The amount of water that entered the plant and the corresponding amount of flooding on various elevations did not
.
!
reach a level that could affect safety-related equipment and was bounded by the design basis flooding analysis. tio significant operational safety parameters were approached or exceeded.
Based on the above, the event was not safety significant; however, response to the event required a rapid reactor
shutdown, including a manual reactor scram, and the consequent actuation of safety equipment.
6.0 Corrective Actions For the December 1991 Circulatina Water
Line Break i
6.1 Background
+
!
On December 22, 1991, the licensee experienced a 36-inch circulating water (CW) system pipe rupture which caused considerable flooding to various buildings. The fiRC dispatched an AIT to evaluate that event (reference liRC Inspection Report 50-440/91026(DRS)).
In response to that event the licensee took or was evaluating corrective actions to minimize internal flooding.
i 6.2.
Inspection l
The team reviewed the corrective action taken by the licensee for the above event. Repair and replacement of the CW pipe, pipe supports, and monitoring activities appeared adequate. Corrective action by the licensee to mitigate i
the flooding of safety-related equipment and buildings had not been completed.
During the previous event, water from the underdrain system piezometer tubes appeared to have leaked into the basement floors of buildings because covers i
for the tubes were either loose or not installed. During this event, the tubes were all closed and no water entered through this path.
Previously some of the water that leaked into the intermediate building 574-foot elevation passed underneath a security door and into a Unit 2 auxiliary building sump.
That water was slightly contaminated,and was pumped from the sump and eventually off site. The licensee had taken no action to preclude water from
passing under the security door into Unit 2; however, they had de-energized
'
and disconnected the sump pump so that the water collected was not pumped out.
During this event, the water that flowed under the security door was retained
'
in the building.
Electrical manholes (ems) were not sealed on the surface and, under flood
[
conditions, water could fill up the EM and flow through electrical conduits.
.
!
EM 1 was. observed to have gaps of 2 to 3-inches by 15-feet. This EM was adjacent to the SW line break and was reported to have approximately 12-inches
,
of flooding at the surface in the area of the manhole.
This flooded the EM.
The water that leaked into the EM flowed through conduits to various areas of-the plant. fio degraded equipment was experienced from the flooding, but the intrusion of water was evident by the amount of silt and moisture still i
present during the inspection. The licensee was evaluating methods of i
preventing flood waters from entering the manholes.
'
The emergency service water pump house (ESWPH) conduit banks on the southeast corner were sealed to minimize water intrusion to the ESWPH in the event of
1 r
.
.
__
.
.
flooding of EM 3.
During this event EM 1 flooded and directed water through conduits to the ESWPH. These conduits, located on the southwest corner of the ESPWH, were not sealed and allowed water to enter. However, the conduits the licensee had sealed prevented water from penetrating into electrical equipment. The water that was directed into the ESWPH ended up on the floor.
The control complex also experienced water intrusion through conduits from EM 1.
Expansion plugs installed at these locations were expelled by the flow of water through the conduits. This caused flooding of approximately 6 to 8-inches at the 599-foot elevation. This location flooded during the previous event because of water entering through these conduits.
The licensee had initiated corrective action documents to seal these conduits, however the time schedule for completion was to be during the fifth operating cycle (1994-1995). At the conclusion of this inspection the licensee was evaluating expediting this schedule.
7.0 Conclusions After completing the AI's Charter, the team was able to make the following conclusions:
(1)
The consequences of the event posed no threat to public health and safety. No significant operational or safety parameters were approached or exceeded.
(2)
The pipe rupture is believed to have resulted from a small perforation and leak in the pipe which slowly grew in size due to erosion. Several potential root causes for development of the
initial perforation were identified. However, a single root cause could not be determined.
The team does believe that the conditions which led to the pipe rupture were localized.
The local failure mechanism concept appears to be supported by the absence of widespread cracking which would result from excessive bending or other gross loads during service.
Inspection of the limited sections of service water system piping that could be observed by the team indicated that no system wide degradation appeared to have occurred.
(3)
The flooding did not effect the function of any safety-related equipment.
(4)
The operators safely responded in an excellent manner to the event and their actions were indicative of a strong knowledge of plant systems and procedures.
(5)
A minor gaseous radiological release occurred but was not directly attributable to the event.
The release was minimal and well below regulatory limits.
(6)
Some additional contamination of the plant occurred as a result of the internal flooding. Water passing through existing
.
!
!
.
!
contaminated areas and floor drains backing up in lower elevations
'
spread contamination to previously clean areas. Overall, the amount and the extent of contamination were not radiologically significant.
.
!
(7)
Corrective actions for the December 1991 circulating water pipe i
break and subsequent flooding had only partiality been implemented l
and, as a result, areas that were flooded before were flooded-l again.
'
r 8.0 Charter Completion
,
The team completed the Charter on April 2,1993, and the AIT was disbanded j
after discussion with Region III management.
i 9.0 Exit Interview
!
!
The team met with licensee representatives (denoted in Attachment 3) on
!
April 2,1993, and summarized the purpose, AIT charter items, and findings of
!
the inspection. The team discussed the results of the inspection, The
licensee did not identify any proprietary documents or processes reviewed by l
the team.
t
!
!
!
'
l
!
i
!
!
$
l
!
l l
.
'
1
.)
.
_.
.
.
ATTACHMENT 3
,
Personnel Contacted Centerior Service Company M. O'Reilly, Senior Attorney f
Cleveland Electric Illuminatino Company D. Igyarto, General Manager, Perry Nuclear Power Plant (PNPP)
B. Beyer, Director, Perry Administrative Support Department (PASD)
S. Kensicki, Director, Perry Nuclear Engineering Department (PNED)
V. Concel, Manager, Systems Engineering, PNED
,
J. Eppich, Manager, Mechanical Design, PNED
,
D. Graneto, Manager, Maintenance Section, PNED W. Kanda, Manager, Electrical Design, PNED F. Vanann, MEU Supervisor, PNED
K. Donovan, Manager, Licensing and Compliance, Perry Nuclear Support
'
Department (PNSD)
H. Hegrat, Supervisor, Compliance, PNSD M. Gymrek, Manager, Operations, PNPP J. Emley, Licensing Engineering, PNSD R. Gaston, Compliance Engineer, PNSD M. Hayner, Licensing Supervising Engineer, PNSD K. Pech, Manager, Integrated Scheduling and Controls, PNSD Nuclear Reaulatory Commission
,
T. Martin, Acting Director, Division of Reactor Safety (DRS)
i R. Lanksbury, Section Chief, Team Leader D. Kosloff, Senior Resident Inspector
!
A. Vegel, Resident Inspector J. Guzman, Reactor Inspector J. Strasma, Public Affairs Officer Other H. Ray Caldwell, Director, Nuclear Activities, Ohio Edison Company T. Reeves, Radiation Analyst, Ohio Emergency Management Agency _
!
J. Vitellas, Energy Specialist, Ohio Public Utilities Commission S. Bair, Reporter, Star-Beacon S. Johnson, Reporter, News Herald
,
J. Kuehner, Reporter, Plain Dealer
!
-
21 t
.
.
L AKE ERIE a
-
'
,
-
r awir s
' M O l? /
F M "!
{
,
~
-
,
!='.'='A
,
.
O t
y
/
"E L"/37 O L.
)* w
,,-
re. c a22.s ion
/
-
15
"
?
Am x
m LO ?E?.'.E
.
-
T
'
s F5 UNIT 1 TURBINE BLDG
'
@
E
~y a'r
,
v og 3ece not DIITsAS BLE RNM M CIwtII
-
,
w a
w APT.A D' Ef'[ AK
'
7 0. - - - -
'y I
I_ E c br--
UNIT 1 N
[
E
_
,
$7mcm Aux]LARY BLDG
.
-
.m m
k
.
kJtL31ser.
11) eN3 FW:Pe UNIT 1
" E*
-==
T
m ac>
x s.,,
u,c - a,.
h: "d, BIL:0L GDERATDR s ta.a s.
cr
,
u
,,
a
_
}!
Lis '7J canet ri UEL gg HANDLING g,wm ca+ux Ks BLDG
^
am w a.
A
'j
.<ust
<
=m Uta nas n
savec l
=>ver uggeg, DbDG
UNIT 2 e
87*
AUXILARY BLDG
WD1E 5r ROLL tr DOORS (L*,RCE IJDDRS)
DFFGAS BLDG TLEB M W X 36 INCH VIIC FTRSDrrJC L D3DPs O CLCC1FICAL CHA5f1 Q
'
tu
"
UNIT 2 TURBitE BLDG SERVICE HATER SYSTD4 PIPr0 IliYOUI'
10 SCTsLE FIGURE 1
- _. --
L A K E-
.n
~ _,
N EN y
y e
~
- n
-
_h EQ E_hl D t., f,'
'~
J
,cro'
.-
.
f
.-
1 * I _D oo_% _
Q j
A.^ = Z - 3 G Oovett Doocs i
}
Y = Pt PC PA L W n.
O t n =p
.uc o-
.
\\
L rect
,
w i o s.
' v(
N
"I'ooDruAREA I
swes s
-
I
'\\\\%Rf@( k,\\ \\
c
= s,c a a,,a
-
\\g
\\
t e 8/Su/
i
^
'
~
Et ]
<
,
k f Guer 1 'rureme %,
1a
!
t
'
'Nie e',' k he " * %-
'
\\
,
[
RAo.WASTT
[g y
-
g r-
.
- '\\
'
y gs
,
m
%12.m gg Gc'
.
,e
'
g '\\ 3 no c connu aI u==9-n x x \\ si
~!,
i
,6
"s t
' sutrier_ hq.
I
.
q __
S smm 1,f
-
"I Aoice'
@
.
l
/
I'A z l l Tr.c?z l
'
hmr *2. Toce.iot B609
,
(
' (
!
,
'
SERVICE KWi131 PIPE BPETsK FLCOD pay NO EC7dZ PIGUrl' 2
i i
- -
.
.-
- -
.
t s ** ar cw
p UNITED STATES g
,g NUCLEAR REGULATORY COMMISSIOto ATTACHMENT 1 o
$c{]hv.
3 REGION Ill
W
, ~,C 799 ROOSEVcLT ROAD l
c,
,
j GLEN El1YN. ILUNOIS Co137-5927 g ~.
s
.....
-
,
MAR 3 01993 i
i CONFIRMATORY ACTION LETTER
!
,
Docket No. 50-440 License No. HPF-58 CAL No. RIII-93-004 Centerior Service Company ATTN:
Mr. R. Vice President Nuclear - Perry c/o The Cleveland Electric j
Illuminating Company
!
_
10 Center Road-i Perry, OH 44081 i
!
.
Dear Mr. Stratman:
SUBJECT: CONFIRMATORY ACTION LETTER
'
Pursuant to a telephone conversation between Mr. T. O. Martin of my staff and
'
'
you on March 30, 1993, related to the service water line rupture which
.!
ncturred on March 26, 1993, it is our understanding that you will take the
.'
following actions:
' ' -
i 1.
Conduct an investigation to determine the cause of the service water
,
,
'i line failure and to evaluate the' decision making and communications associated with the event.
,
2.
Maintain documentary evidence of your investigation effort and make this available to the Augmented Inspection Team.
,
j t;
3.
Evaluate the service water line rupture in light of past service water and circulating water line (fiber glass lines) failures to determine if
additional actions are necessary.
{
,
'
Provide within 30 days to NRC Region III a documented evaluation of the
!
4.
above issues including corrective actions you have taken or plan to j
!
take.
't None of the actions specified herein should be construed to take precedence
over actions which you feel necessary to ensure plant and personnel safety.
_*
.
<
CONFIRMATORY ACTION LETTER
'
I-i:
i
- i
.
-
- -. -
-
-
l.
,
-
-
.
MAR 3 01993 i
.
CONFIRMATORY ACTION LETTER Centerior Sersice Company
'
!
Pursuant to Se: tion 182 of the Atomic Energy Act, 42 U.S.C. 2232, and 10 CTR l
2.204, you are required to:
.
.
1.
Notify me immediately if your understanding differs from that set forth
!
above,
'
2.
Notify me if for any reason you cannot complete the actions within the
!
specified schedule and advise me in writing of your modified schedule in advance of the change, and i
3.
Notify me in writing when you have completed the actions addressed in this Confirmatory Action Letter.
Issuance of this Confirmatory Action Letter does not preclude issuance of an t
i order formalizing the above commitments or requiring other actions on the part
'
of the licensee.
Nor does it preclude the NRC from taking enforcement action for violations of NRC requirements that may have prompted the issuance of this letter.
In addition, failure to take the actions addressed in this Confirmatory Action Letter may result in enforcement action.
The responses directed by this letter are not subject to the clearance procedures of the Office of Management and Budget as required by the Paperwork
!
Reduction Act of 1980, Pub. L. No.96-511.
In accordance with 10 CFR 2.790 of the NRC's " Rules of Practice," a copy i
of this letter and its enclosures will be placed in the NRC Public Document Room.
!
Sincerely,
-
.
.
i
$
L"
,
"
t A. Bert Davis
,
Regional Administrator
See Attachec Distribution A
y
i
!
!
!
!
'
.
.
_ _
,
- - -.
-
'
,
,
l
..
l MAR 3 01993
CONFIRMATORY ACTION LETTER i
Centerior Service Company
Distribution
'
CC:
,
F. R. Stead, Director, Nuclear Support Department D. P. Igyarto, General Manager,
,
Perry Nuclear Power Plant
'
K. P. Donovan, Manager, licensing and Compliance Section S. F. Kensicki, Director, Perry l
Nuclear Engineering Dept.
i H. Ray Caldwell, General i
Superintendent Nuclear Operations OC/LFDCB
'
Resident Inspector, Rill Terry J. Lodge, Esq.
James R. Williams, State of Ohio
Robert E. Owen, Ohio Department of Health
A. Grandjean, Ohio Public Utilities Commission
,
State liaison Officer J. M. Taylor, EDO J. H. Sniezek, DEDR
-
H. L. Thompson, DEDS T. E. Murley, NRR J. G. Partlow, NRR
!
J. W. Roe, NRR
!
,
F. J. Miraglia, NRR I
J. A. Zwolinski, NRR E. L. Jordan, AE00 J. Lieberman, OE J. R. Goldberg, OGC i
J. N. Hannon, NRR i
G. Grant, EDO J. E. Richardson, NRR H. J. Miller, Rlll R. J. Strasma, Rill L. R. Greger, Rlli J
R. D. Lanksbury, Rlll
,
-.
_
,
.
,
e p*"4g UNITED STATES ATTACHMENT 2 i
,
yi fg NUCLEAR REGULATORY COMMISSION
!
,
y
I
' -ef ( i p,
REGION Ill g
'9-
,
- . * "
p 75G ROOSEVELT ROAD J.
- o, g
GLEN ELLYN fWNOIS(0137-5927
o...../
S
,
MEMORANDUM FOR:
Roger D.
Lanksbury, Team Leader, Perry Augmented Inspection Team (AIT)
FROM:
Edward G.
Greenman, Director, Division of Reactor Projects
.
SUBJECT:
DRAFT AIT CHARTER - PERRY SERVICE WATER FIBERGLASS PIPE BREAK Enclosed for your implementation is the Charter developed for the inspection of the events associated with the Perry service water
'
line break which occurred on March 26, 1993.
This Charter was prepared in accordance with the NRC Incident Investigation Manual and the Manual Chapter 0325 AIT implementing procedure, and is based on the discussions you had with Region III personnel on March 27, 1993.
As stated, the objectives of the AIT are to
>
communicate the facts surrounding this event to regional and headquarters management, to identify and communicate any generic safety concerns related to this event to regional and
!
headquarters management, and to document the findings and conclusions of the onsite inspection.
If you have any questions regarding these objectives or the
.
enclosed Charter, please do not hesitate to contact either Tom l
Martin or myself.
!
.
Cd4%
-
I Edward G.
Greenman, Director Division of Reactor Projects
,
t Enclosure:
Draft AIT Charter t
P cc w/ enclosure:
A.
B.
Davis, RIII i
H.
J.
Miller, RIII
'
F.
J.
Miraglia, NRR j
J.
C.
Partlow, NRR
C.
E. Rossi, NRR
J W.
Roe, NRR
B.
K.
Grimes, NRR l
J.
N.
Hannon, NRR l
E.
L.
Jordan, AEOD
J.
E.
Richardson, NRR j
J.
Zwolinski, NRR i
G.
Grant, EDO D.
C Kosloff, SRI
)
,
P
-.
...
.
.
-
.
..
. _
..
-
- -...
.
..
!
!
^
!
.
l PERRY SERVICE WATER LINE BREAK
!
!
DRAFT AUGMENTED INSPECTION TEAM (AIT) CHARTER l
t
INVESTIGATE:
.
1.
The break of the 30" fiberglass service water line and subsequent flooding.
i
2.
Probable root cause(s).
',
3.
Performance history and maintenance on service water piping.
(
4.
Operator response to the event, including use of the Plant
!
Emergency Instructions (PEIs).
.;
5.
Effects of flooding.
!
6.
Event classification and reporting.
7.
Corrective actions.
l 8.
Conclusions.
'!
!
't OUESTIONS FOR PERRY AIT:
'
1.
The break of the 30" fiberglass service water line and i
l subsequent flooding (3/26/93).
!
L 1.2 What was the sequence of events?
t i
1.2 How much water was pumped from the break and where did
it go?
'
f 1.3 What was the flood path (internal and external to the j
plant)?
f
!
1.4 Did the flood path conform to the USAR assumed flood
!
paths and magnitude?
?
1.5 What was the safety significance of : the event? (Include any PRA insights, to.the extent practicable.
Analyze loads on service water system to determine the affect
i of its loss on the safe shutdown of the plant and-maintenance in a safe shutdown condition, including spent fuel pool cooling.)
1.6 Were the licensee's corrective actions from the 12/22/91 circulating water event effective in minimizing the consequences of-this event?
'
i
.. - - -
.
. - -
.
..
.-.
..
-
...
._
. _ _ _ _ _. - ~
. -_
..
__
_
_
. _ _.
-.
_ -. _. _... -... _ _ _
,
'
'
,
i e
2.
Probable root cause(s).
I t
2.1 What was the root cause of the event?
,
.
3.
Performance history and maintenance on service water piping.
I i
3.1 What is the material condition of the affected piping?
I
!
.
3.2 Has there been any history of leakage in the affected line or any related maintenance activities?
i
,
!
3.3 Were there any on-going activities that could have been
!
j precursors to the event?
I
3.4 Has there been any reported damage to the piping during either construction or operation?
i 4.
Operator response to the event, including use of the Plant
[
<
Emergency Instructions (PEIs).
,
,
4.1 What operator actions were taken during the event?
Were they appropriate?
-
!
5.
Affects of flooding.
5.1 Identify all affected equipment.
[
.
!
-
i 5.1.1 Electrical:
cables, switch gear, MCC, etc l
.
5.1.2 Mechanical:
pumps, valves, etc.
!
i 5.1.3 Specifically review water ingress to HPCS,
!
t
,
RHR, RCIC, LPCS rooms and potential for
!
j affecting the operation of these pumps.
l 5.2 Identify the extent of water damage.
i
5.3 Radiological consequences.
j
'
i
.
!
5.3.1 Extent of contamination-from floor drain j
)
backup.
4
.
5.3.2 Offsite releases, if any.
,
6.
Event classification and reporting.
!
-
6.1 Was the event properly classified and were required i
.
-
notifications made in a timely manner *
!
!
7.
Corrective actions and evaluations.
'I
'
l
.
7.1 What are the licensee's short term and long term
'
corrective actions and evaluations?
7.1.1 Service water piping
,
T
.
t J
. -...
.
., _. _...... - -
.,.
--. - -
-
'i
- -
- -. - -.. -.
-.. -----.
- -.
.. -
...
._-.
.. _ _ = _ - ~ _. - - - _.
-
!
. -
,
,
t
. i
!
- o i
i k
1-7.1.2
Electrical components such as cable trays,
{
switchgear, McC's
'
,
J
!
7.1.3 Effected mechanical components i
7.1.4 Radiological consequences, if any
!
a i
d 8.
Conclusions.
t
'
i
1 i
'
s I
.
.
,
- -
I
i
,
'
!,
I s
i il i
I
d
d l
!i i
s
M il n
i
,
l
?
!
!
l I
e i
i
l
!
i l'
l l
i
!
!'
i
!
.
J
.<
$
!
l
]
s
1 h.
4 e---
w
--.,4+.
,,
,
--. -
. --
- -,.
n--<w-
..,,,
-,
m --- - -
--g.,
,,-
,e-.
,.-,.
1,
-
.y
,,