United States Nuclear Regulatory Commission - Protecting People and the Environment

Information Notice No. 91-85: Revision 1: Potential Failures of ThermostaticControl Valves for Diesel Generator Jacket Cooling Water

                                 UNITED STATES
                         NUCLEAR REGULATORY COMMISSION
                         WASHINGTON, D.C.  20555-0001

                               February 27, 1997

                                           CONTROL VALVES FOR DIESEL GENERATOR  
                                           JACKET COOLING WATER


All holders of operating licenses or construction permits for nuclear power


The U.S. Nuclear Regulatory Commission (NRC) is issuing this revised
information notice to clarify that the failure of the thermostatic control
valve for the diesel generator jacket cooling water at Catawba Nuclear Station
occurred beyond its normal service life and to provide additional information
regarding its shelf life.  It is expected that recipients 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 are not NRC requirements; therefore, no specific
action or written response is required.


On December 26, 1991, the staff issued Information Notice 91-85 to describe a
failure of a thermostatic control valve for the diesel generator jacket
cooling water.  Since the notice was issued, additional details have emerged
on the circumstances related to the valve failure.  This revision reports
these additional details. 

Description of Circumstances

On September 10, 1991, at the Catawba Nuclear Station, Unit 2, the Duke Power
Company (the licensee) removed emergency diesel generator (EDG) 2A from
service for minor corrective and preventive maintenance.  The next day, the
engine was successfully tested for its performance in a no-load condition for
5 minutes and was shut down.  Shortly afterwards, the engine was restarted for
a 1-hour operability test.  After operating the EDG for  15 minutes at full
load, operators observed that the engine cooling water and lubricating oil
temperatures were increasing abnormally.  The engine oil level was checked and
found to be within normal range.  After about 20 minutes, alarms were received
for high lubricating oil inlet and outlet (79.4 �C [175 �F]) and high jacket
water (79.4 �C [175 �F]) temperatures.  Responding to these alarms, an
operator verified that the nuclear service water supply valve was open and
that the flow was within normal range.  After operating for about 28 minutes, 
the engine tripped on a high lubricating oil outlet temperature (93.3 �C
[200 �F]).  After the

9702240073.                                                          IN 91-85, Revision 1
                                                          February 27, 1997
                                                          Page 2 of 3

trip and while the engine was coasting to a stop, an explosion occurred in the
crankcase.  Upon examining the engine, the licensee determined that a
thermostatic valve in the jacket cooling water system had failed, overheating
the engine and causing significant engine damage.

Further details of the event may be found in Licensee Event Report 50-414/91-
10 (Accession No. 9110180010) and in NRC Inspection Report 50-414/91-21
(Accession No. 9111200115). 


At Catawba, the cooling water system for the Transamerica Delaval DSRV-16-4
(16-cylinder) diesel generator engine supplies cooling water to the engine
jacket, the engine lubricating oil cooler, the combustion air aftercoolers,
and the governor lubricating oil cooler.  An engine-driven pump circulates
cooling water through the closed-loop system that includes a three-way
thermostatic control valve (AMOT Model 8D).  When the jacket water temperature
is low, the AMOT valve, which works similarly to the thermostat in an
automobile engine, diverts diesel jacket cooling flow from the heat exchanger,
which is cooled by nuclear service water.  The AMOT valve modulates open to
control the temperature of the diesel jacket water to about 73.9 �C (165 �F)
as the engine reaches operating temperature.  Heat is transferred through the
heat exchanger to the plant's nuclear service water system.

The licensee's investigation revealed that the AMOT thermostat valve had
malfunctioned and caused the engine to overheat.  During the operability test,
the AMOT valve opened only partially from the bypass position, thus allowing
the jacket and lubricating oil temperatures to rise above normal operating
temperatures.  With the engine fully loaded, the temperature of the oil and
water rose to 93.3 �C (200 �F).  At these elevated temperatures, the oil
emitted an increased amount of vapors.  The elevated temperatures also
affected the clearances between the moving parts and reduced the oil's
lubricating qualities.  The heat generated by the friction between the piston
and liner ignited the oil vapors, causing the rapid pressure increase and the
explosion in the crankcase.  

The AMOT valve malfunctioned because two of the four "power elements" (sensing
elements) in the valve had failed.  Licensee staff examined these elements at
the licensee's metallurgical laboratory and attributed the root cause of the
failure to slowly growing intergranular stress corrosion cracking (IGSCC). 
The licensee did not report what caused the IGSCC.  The small cracks allowed
the thermally active medium of the power element to leak out.  This loss of a
thermally active medium directly affects the travel of the actuating rod
(valve stroke).  The licensee also found degraded power elements during an
inspection of the Unit 1 diesel engines.    

The failed elements in Unit 2 had not been inspected or replaced since they
were originally provided with the engine in 1979.  The licensee had vendor's
documentation indicating that the power elements (1) should be inspected at
intervals of 2 to 3 years and (2) have a 15-year shelf life.  The licensee did
not have any documents specifying the service life of the elements and
mistakenly assumed that the 15-year shelf life also applied to the service
life of the elements.  In documents submitted to NRC by AMOT in letter dated
September 19, 1996,   .                                                          IN 91-85, Revision 1
                                                          February 27, 1997
                                                          Page 3 of 3

(Accession No. 9701290173), the elements' service life is stated to be 6�10
years (AMOT Models B, H, and D Steel Bodied Thermostatic Valves, Commercial
Steel Valves, Nuclear "N" Stamp Valves  [Form 757]).  Excessive temperatures,
chemical, electrolytic, or cavitation attack will shorten the service life of
elements and seats.  However, in critical nuclear plant installations, AMOT
recommends the element assemblies be changed every 5�7 years.  Hence, the
power elements at Catawba had exceeded their service life.

AMOT has recently provided NRC a new expectation on the shelf life of the
power element.  In a January 6, 1997, letter to NRC from AMOT (Accession No.
9701160070), the vendor stated that the shelf life of the element is dependent
on the materials used and the storage conditions.  The rubber materials
(general purpose Nitrile) internal to the wax motor are the primary factors
determining the shelf life.  The expected shelf life for the thermostatic
element used in the model H thermostatic valves with the current type of
materials is 5�10 years, provided that the element is stored in the original
box in a cool (15.6 �C�26.7 �C [60 �F�80 �F]) dry area away from sunlight, and
ultraviolet light, and ozone.  Elevated temperatures will reduce the life of
the rubber materials.  Before installing the elements AMOT recommends testing
them in accordance with their procedures.  

The licensee's corrective actions include changing the AMOT power element
replacement schedule and evaluating changing the engine oil to a type with
increased film strength and greater stability at elevated temperatures. 

Related Generic Communications

A similar failure of a thermostatic control valve, but caused by a different
failure mechanism, was discussed in NRC Information Notice 82-56, "Robertshaw
Thermostatic Flow Control Valves," dated December 30, 1982.

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 Office of
Nuclear Reactor Regulation (NRR) project manager.

                                          signed by

                                    Thomas T. Martin, Director
                                    Division of Reactor Program Management     
                                    Office of Nuclear Reactor Regulation

Technical contacts:  Thomas Greene, NRR         John Zeiler, RII
                     (301) 415-1175             (803) 383-4571
                     E-mail:  tag@nrc.gov       E-mail:  jxz@nrc.gov
Page Last Reviewed/Updated Tuesday, November 12, 2013