Information Notice No. 91-50: Supplement 1: Water Hammer Events Since 1991
UNITED STATES
NUCLEAR REGULATORY COMMISSION
OFFICE OF NUCLEAR REACTOR REGULATION
WASHINGTON, D.C 20555-0001
July 17, 1997
Information Notice No. 91-50 SUPPLEMENT 1: WATER HAMMER EVENTS
SINCE 1991
Addressees
All holders of operating licenses or construction permits for nuclear power
reactors.
Purpose
The U.S. Nuclear Regulatory Commission (NRC) is issuing this supplement to
Information Notice (IN) 91-50 to alert addressees to continuing occurrences of
water hammer events at nuclear facilities. It is expected that recipients
will review the information for applicability to their facilities and consider
actions, as appropriate, to avoid similar occurrences. However, suggestions
contained in this information notice supplement do not constitute NRC require-
ments: therefore, no specific action or written response is required.
Background
The NRC previously addressed water hammer in Unresolved Safety Issue A-1, as
documented in "Evaluation of Water Hammer Occurrence in Nuclear Power Plants,"
NUREG-0927, Revision 1, dated March 1984. Subsequently, on August 20, 1991,
the NRC issued IN 91-50, "A Review of Water Hammer Events After 1985" to
update the occurrence of water hammer. This supplement is issued to update
various occurrences of water hammer since the issuance of IN 91-50. The
specific events discussed below are recent examples of significant water
hammer occurrences.
Description of Circumstances
Oconee
On September 24, 1996, during the restart of Oconee Unit 2, facility personnel
were manually realigning the moisture separator reheater (MSR) drains at
approximately 50-percent power when an 18-inch second-stage MSR drain line
ruptured. The water/steam was at a temperature of approximately 400 �F.,
250 psig, and burned seven plant workers who were manually realigning valves
to feed-forward to the feedwater heaters. An 18-inch pipe between the second-
stage reheaters and the feedwater heaters ruptured where a 45-degree stub was
attached. It was determined that the procedure for the manual valve alignment
had not been revised to include the guidance about system pressures and valve
opening time,
9707140291. IN 91-50, Supp. 1
July 17, 1997
Page 2 of 5
resulting in the operators opening the valves earlier in the startup than
planned. Also the valves were opened over a period of minutes instead of the
1 1/2-hour period that was used during an earlier startup.
The subsequent reverse flow initiated the destructive condensation-induced
water hammer, which included the combined phenomena of water slug motion and
steam void collapse.
Another water hammer event in the same Unit 2 second-stage reheater drain
system had occurred in May 1996. This water hammer was determined to be
caused by the opening of the high-level divert valve and the resultant reverse
flow. As a result, a support failed on the same pipe that later ruptured.
Review of onsite documentation revealed that water hammers had probably
occurred on the second-stage reheater drain system since initial plant
startup. [See NRC Inspection Report (IR) 50-269, 270, 287/96-15 Accession No:
9612030116.]
Sequoyah
In January 1995, residual heat removal (RHR) pipe supports were found to be
damaged. The cause was determined to be water hammers that occurred during
RHR pump testing. Large gas pockets composed of 99 percent nitrogen were
later found in the RHR lines and probably resulted from leakage of water from
a cold-leg accumulator tank. [See NRC IR 50-327, 328/95-04 Accession No:
9503080217 and License Event Report (LER) 50-327-95-01 Accession No:
9702120251.)
In October 1996, a water hammer in the steam dump piping caused the pipe to
crack where the steam dump piping joins the main steamline. The water hammer
was caused by a failed level switch on the steam dump drain tank. The failed
switch prevented the tank from draining and, because of the common piping
system arrangement, significant portions of the steam dump lines to the
condenser are believed to have been partially full of water before the event.
(See NRC IR 50-327, 328/96-13 Accession No: 9612020167.)
Catawba
In June 1995, pipe caps on the charging system letdown line were loosened
because a water hammer in the letdown line. A portion of the letdown line had
been isolated and then partially depressurized. A small steam void in the
depressurized section is believed to have formed. When the letdown isolation
valve was reopened, the steam void was rapidly collapsed, resulting in a water
hammer. Operators noticed that reactor coolant system leakage rates increased
from 0.2 gpm to 0.7 gpm. Operators conducted a walk down inspection and found
the loosened pipe caps. (See IR 50-413, 414/95-16 Accession No: 9508220388.)
. IN 91-50, Supp. 1
July 17, 1997
Page 3 of 5
Brunswick
A water hammer occurred in July 1996 during Unit 2 startup. Damage included a
drain valve actuator yoke that was severed in two places. The water hammer
occurred in the 4A and 5A feedwater heater (FWH) drain piping during shell
warming of the turbine. The 4A and 5A FWHs heated up faster than normal as a
result of leakage through the extraction steam isolation valves to one or both
heaters. (See IR 50-324, 325/96-13 Accession No: 9610310052.)
Fermi Unit 2
In March 1991, a MSR relief valve discharge manifold drain line ruptured in a
heat-affected zone of a weld because of relief valve discharge loads.
Initially, a U-bolt support failed and allowed excessive cyclic displacement
of the drain line piping, which led to the failure. (See LER 50-341/91-04
Accession No: 9104170157.)
Watts Bar
In November 1994, a severe water hammer occurred in the Train A containment
spray system during Integrated Safeguards Testing. The initiating event was a
momentary interruption of emergency power (intentionally initiated during
testing) after con- tainment spray flow through the full-flow recirculation
line had been initiated by a Phase B containment isolation signal. A blank
flange had been installed in place of a check valve at the top of the riser
adjacent to the ring header to prevent any flow to containment. During the
time power was lost, the vertical containment spray header inside containment
partially drained back through the full-flow recirculation line, voiding the
piping of liquid behind the blank flange. When power was restored the pump
restarted and the voided spray header was rapidly refilled, causing the water
hammer. This event damaged several pipe hangers and was similar to the events
described in IN 87-10, "Potential for Water Hammer During Restart of Residual
Heat Removal Pumps." It was subsequently found that three separate water
hammer events had occurred during the testing. (See NRC IR 50-390/95-60
Accession No: 9510180122.)
In March 1996, a water hammer event occurred on the number 1 feedwater heater
extraction steam system and damaged three steamline motor-operated valves.
The event occurred while draining the steam lines to place the heaters in
service. A pressure differential built up across extraction steam non-
return check valves that was sufficient to overcome the air operator seating
force and let steam into the evacuated lines. This occurrence established the
conditions that caused the condensation-induced water hammer. (See NRC IR 50-
390/96-04 Accession No: 9605140417.)
Vogtle Units 1 and 2
In February 1996, the licensee determined that cracks in the nuclear service
water (NSW) system were the result of multiple water hammer events. Weld
failures had occurred in several locations in the pump discharge lines. The
licensee determined that the weld cracks . IN 91-50, Supp. 1
July 17, 1997
Page 4 of 5
and indications were fatigue related, resulting from flow oscillations or
hydraulic transients when an NSW pump starts while other NSW pumps are already
operating. The transient loading is believed to have been caused by the
impact of the discharge flow from the starting pump against a closed discharge
check valve, which opened it against the pressure from the running pumps.
(See LER 50-424/96-01 Accession No: 9609180013.)
H. B. Robinson Unit 2
In October 1996, a water hammer damaged a seismic restraint on a cold-leg
accumulator injection line. The licensee had completed a check valve leak
test, which partially drained the injection line. The line was refilled by
opening the accumulator isolation valve (10 second stroke time). This step
allowed the 600-psig water in the accumulator to rapidly repressurize the
drain line, thus causing the water hammer. (See NRC IR 50-261/96-12 Accession
No: 9612230323.)
Waterford Unit 3
On three occasions in November and December 1996, a water hammer occurred in
the low pressure safety injection (LPSI) system when the LPSI pumps were
started. The licensee determined that the large pressure transient was caused
by the presence of a nitrogen gas volume in the pump discharge line. The gas
had come out of solution when nitrogen-saturated, 600-psig water leaked from
the safety injection tanks into the lower pressure LPSI piping. When the
pumps started, the compressible gas volume created a transient pressure surge
in the LPSI system that was significantly greater than that expected for a
water-solid system. (See NRC Information Notice 97-40 "Potential Nitrogen
Accumulation Resulting from Backleakage from Safety Injection tanks."
Accession No: 9706230168; and IR 50-382/96-14 Accession No: 9702130178.)
Shearon Harris
In 1991, a high-head safety injection relief valve repeatedly chattered and
caused a water hammer load while performing the mini-flow function to prevent
system pump deadheading. As a result, the relief valves and some system drain
connections were damaged such that a significant portion of the safety
injection flow would have been diverted during a loss of coolant accident.
(See LER 50-440/91-11 Accession No: 9112100271.)
Discussion
A review of the circumstances previously listed indicates that water hammers
are continuing to occur in both high and low temperature systems and/or high
and low pressure systems as a result of a variety of causes and in safety and
nonsafety related systems. In several cases, there were multiple occurrences
of these events. This indicates potentially inadequate root cause evaluations
and corrective actions applied to the design and/or operation of these
systems. The effects of these water hammer events range from piping
deflection and damaged supports to catastrophic system and equipment failures
that sometimes result in personnel injury. Many of the more severe loading
events were caused by column . IN 91-50, Supp. 1
July 17, 1997
Page 5 of 5
separation/void collapse or condensation-induced water hammers. NUREG/CR-5220
"Diagnosis of Condensation-Induced Waterhammer" issued October 1988, provides
reference material and diagnostic procedures concerning condensation-induced
water hammer in nuclear power plants. It should also be noted that some
significant events were apparently caused by the presence of non-condensable
gases (e.g., nitrogen) in the piping systems.
In addition, the potential for water hammer recently was identified in
containment cooling systems. The licensee for Diablo Canyon Units 1 and 2
discovered in February 1996 that cooling water systems serving containment air
coolers might be susceptible to water hammer loadings during a loss-of-coolant
accident (LOCA) with a concurrent loss-of-offsite power (LOOP). Westinghouse
Electric Corporation issued a letter to alert its customers to the potential
safety issue that was identified at Diablo Canyon. Consequently, the licensee
for Haddam Neck plant found its containment air coolers inoperable based on
its analysis of the water hammer effects on the piping and structural limit of
supports. As a result, the NRC staff issued on September 30, 1996, Generic
Letter 96-06. "Assurance of Equipment Operability and Containment Integrity
During Design-Basis Accident Conditions," to notify licensees about the safety
significant issues that could result from the potential water hammer effects
in the containment air coolers.
This information notice requires no specific action or response. If you have
any questions about the information in this notice, please contact one of the
technical contacts listed below or the appropriate Office of Nuclear Reactor
Regulation (NRR) project manager.
signed by S.H. Weiss for
Marylee M. Slosson, Acting Director
Division of Reactor Program Management
Office of Nuclear Reactor Regulation
Technical contacts: Charles Hammer, NRR Thomas Peebles, RII
301-415-2791 (404) 331-5541
E-mail: cgh@nrc.gov E-mail: tap@nrc.gov
Thomas Koshy, NRR
301-415-1176
E-mail: txk@nrc.gov
Page Last Reviewed/Updated Tuesday, March 09, 2021