Protecting People and the EnvironmentUNITED STATES NUCLEAR REGULATORY COMMISSION
UNITED STATES
NUCLEAR REGULATORY COMMISSION
OFFICE OF NUCLEAR REACTOR REGULATION
WASHINGTON, D.C. 20555
January 29, 1990
Information Notice No. 90-05: INTER-SYSTEM DISCHARGE OF REACTOR COOLANT
Addressees:
All holders of operating licenses or construction permits for nuclear power
reactors.
Purpose:
This information notice is intended to alert addressees to a potentially
significant problem in identifying and terminating reactor coolant system
leakage in operating modes 4 and 5. It is expected that licensees will
review the information for applicability to their facilities and consider
actions, as appropriate, to avoid similar problems. However, suggestions
contained in this information notice do not constitute NRC requirements;
therefore, no specific action or written response is required.
Description of Circumstances:
On December 1, 1989, Braidwood Unit 1 experienced the unplanned inter-system
discharge of approximately 68,000 gallons of water. The discharge was
caused by the inadvertent opening of a residual heat removal (RHR) system
suction relief valve. The valve failed to reclose, allowing an open flow
path from the reactor vessel, through the RHR system, into the unit's two
recycle hold-up tanks (HUTs).
The unit, which had been in a refueling outage since September 2, 1989, was
heating up in operational mode 5, preparing to enter operational mode 4.
The plant was solid and in the process of drawing a bubble in the
pressurizer. The RHR train "A" pump was in operation and, although the "B"
pump was not running, the "B" train was unisolated and available. The
reactor coolant system (RCS) was at a pressure of 350 psig and a temperature
of 175 F. Charging flow to the vessel was being provided by the "A"
charging pump. Pressurizer heaters were on. The "B" charging pump was
isolated and tagged out of service. (Technical Specifications governing
cold overpressure protection require that only one charging pump be
available. The other charging pump and the safety injection pumps are
required to be tagged out of service, with power supplies removed). To
protect against a pressure switch failure and the subsequent automatic
isolation of the RHR system, the train "A" RHR suction isolation valve was
open and tagged out of service.
9001230126
.
IN 90-05
January 29, 1990
Page 2 of 4
At 1:42 a.m., operators throttled the charging flow and maximized the
letdown flow in preparation for drawing a bubble in the pressurizer. The
RCS pressure was 404 psig and the pressurizer level was off scale, high. At
1:44 a.m., a rapid reduction in the pressurizer level occurred, with the
pressurizer level off scale, low, at 1:52 a.m. Approximately 14,000 gallons
of water drained from the pressurizer and the pressurizer surge line;
however, the reactor vessel level instrumentation system indicated that the
vessel level remained at 100 percent. At 1:49 a.m., the charging flow was
increased and the charging pump suction was switched from the volume control
tank to the refueling water storage tank (RWST).
About 30 to 50 gallons of water were observed on the floor of the auxiliary
building in proximity to the RHR train "A" suction relief valve, leading
plant personnel to believe that this valve had lifted. At 1:53 a.m., the
letdown flow was reduced to minimum and charging was maximized. The RHR
trains were switched from "A" to "B", the "A" pump was stopped, and the
isolation of the "A" train was initiated. At 1:59 a.m., one of the two
running reactor coolant pumps (RCPs) was stopped because of low RCS
pressure.
A second charging pump, "B", was started following completion of the formal
procedure for tagout removal. At 2:35 a.m., the "A" RHR suction isolation
valve was returned to service and closed, completing the isolation of the
"A" train of the RHR system. The pressurizer level began to recover and the
RCS pressure increased slightly, giving operators the impression that the
discharge had been isolated. The "B" charging pump was therefore secured at
2:45 a.m. The pressurizer level, however, did not recover. At 2:54 a.m.,
the "B" charging pump was restarted. At 3:49 a.m., the inter-system
discharge was terminated when the RHR train "A" pump was started, the "B"
pump shut down, and the "B" train was isolated. The level indication for
the HUTs stabilized and the pressurizer level began to recover at 3:52 a.m.
By 5:06 a.m., the pressurizer level had fully recovered and the unit was
stabilized at 360 psi and 175 F. Approximately 68,000 gallons of water had
been discharged from the reactor vessel to the HUTs. (The total amount of
water was composed of 14,000 gallons of initial pressurizer inventory and
54,000 gallons of makeup water).
Following the event, it was determined that the RHR "B" train suction relief
valve had lifted at 411 psi. The lift setpoint for the valve should have
been 450 psi. The valve should have reclosed on reducing pressure but
failed to do so. The premature opening of the valve was attributed to the
presence of foreign material lodged between the valve spindle and the
spindle guide. This foreign material either prohibited the correct
adjustment of the valve or affected the valve's lift setpoint. The valve's
failure to reclose was attributed to improper nozzle ring adjustment. The
reset pressure is strongly influenced by the dynamic forces created by the
nozzle ring. If the ring is located too high on the nozzle, it may result
in an inadequate ventilation area just above the nozzle. Undesirable forces
will develop which may cause a much lower reseat pressure.
The water found near the RHR train "A" suction relief valve had leaked from
a weep hole on a relief valve in a radwaste evaporator line connected to the
.
IN 90-05
January 29, 1990
Page 3 of 4
common discharge header of the train "A" and "B" suction relief valves.
Contrary to original assumptions, there was no evidence that the "A" train
suction relief valve had lifted. The root cause of the problem with the
relief valve on the evaporation line is under investigation but is thought
to be unrelated to the failure of the "B" suction relief valve.
Hampering operators' efforts throughout this event was the lack of an
appropriate emergency operating procedure (EOP) to detect coolant leaks
while in operating modes 4 and 5. However, the operators were able to
combine two related abnormal operating procedures for guidance during this
event. One of the procedures is designed to locate system leaks while in
modes 3 and 4. The other provides guidance for the restoration of the RHR
system following its loss during conditions in which the reactor vessel
inventory is at a reduced level.
Discussion:
The event at Braidwood 1 is significant because it underscores the need to
have EOPs available for use in other than "at power" operating modes. The
fact that over 2 hours were required to locate the stuck-open valve, to
terminate the discharge, and to begin refilling the pressurizer highlights
the need to provide personnel with adequate tools to perform their tasks.
Relying on ad hoc procedures during significant events places an unnecessary
burden on operating personnel. The lack of adequate EOPs could handicap the
most competent operators in their efforts to address significant operational
problems.
Also illustrated by this event is the need for procedures to assure that
adequate RCS makeup capability and cooling options are available in a timely
fashion during shutdown. The discharge through the stuck-open relief valve
exceeded the capability of a single charging pump. Starting a second
charging pump required that formal procedures for tag removal be conducted.
This effort necessitated a considerable amount of time, which may not be
available should a similar event occur while the RCS is at a higher
temperature.
The severity of this event could have been increased if greater decay heat
were present in the reactor vessel or if a gross failure of the relief valve
discharge header had occurred. Greater decay heat would have increased the
potential for voiding in the core. Also, because the header discharges to
the HUTs which are located outside containment, a piping failure could have
resulted in all or a portion of the RCS water being discharged to the
building floor. This event would have necessitated a major cleanup effort
and increased the potential for personnel contamination.
If this event had occurred at one of the nuclear plants that has a single
suction line from the RCS to the RHR system, all shutdown cooling would have
been lost as a result of isolating the failed suction relief valve. An
alternate heat sink would likely have been required; however, in mode 5, an
alternate heat sink may not be readily available.
.
IN 90-05
January 29, 1990
Page 4 of 4
This information notice requires no specific action or written response. If
you have any questions about the information in this notice, please contact
one of the technical contacts listed below or the appropriate NRR project
manager.
Charles E. Rossi, Director
Division of Operational Events Assessment
Office of Nuclear Reactor Regulation
Technical Contacts: Nick Fields, NRR
(301) 492-1173
Julian Hinds, RIII
(315) 388-5575
Attachment: List of Recently Issued NRC Information Notices
.ENDEND