United States Nuclear Regulatory Commission - Protecting People and the Environment

Information Notice No. 95-20: Failures in Rosemount Pressure Transmitters due to Hydrogen Permeation into the Sensor Cell

                          UNITED STATES
                     WASHINGTON, D.C.  20555

                          March 22, 1995

                                 TRANSMITTERS DUE TO
                                 HYDROGEN PERMEATION INTO THE
                                 SENSOR CELL


All holders of operating licenses or construction permits for
nuclear power reactors.


The U.S. Nuclear Regulatory Commission (NRC) is issuing this
information notice to alert addressees to a potential failure
mode in Rosemount Nuclear Instruments, Incorporated, Model 1152,
1153 and 1154 pressure transmitters due to hydrogen gas
permeation through the isolating diaphragm exposed to process
fluid.  It is expected that recipients will review the
information for applicability to their facilities.  However,
suggestions contained in this information notice are not NRC
requirements; therefore, no specific action or written response
is required.  

Description of Circumstances

On November 22, 1994, St. Lucie Unit 1 was in a cold shutdown
condition and in the process of filling and venting the reactor
coolant system (RCS).  With RCS pressure at 0.45 MPa [50 psig]
and RCS temperature at approximately 38  C [100  F], a safety
injection actuation was initiated when two of the four
pressurizer pressure channels generated high pressure signals. 
When the output from the two transmitters exceeded 11.91 MPa
[1712 psig], the manual safety injection block, which had been
established during cooldown, cleared.  With the safety injection
block cleared, and the two properly functioning pressurizer
pressure transmitters indicating 0.45 MPa [50 psig], the safety
injection actuation logic was satisfied and a safety injection
was initiated.

It was determined that the two pressurizer pressure channels
indicated high pressure because of an erroneous high output from
the pressure transmitters. These transmitters are Rosemount Model
1153 gauge pressure transmitters that had been sent back to
Rosemount for refurbishment because they were susceptible to
sensor cell oil leakage as discussed in NRC Information Notice
89-42, "Failure of Rosemount Models 1153 and 1154 Transmitters,"
NRC Bulletin 90-01 "Loss of Fill Oil in Transmitters Manufactured
by Rosemount," and NRC Bulletin 90-01, Supplement 1.  The failed
transmitters had been in service at St. Lucie since April 1993
(about one cycle) with no apparent symptoms. 

9503200279.                                       IN 95-20      
                                                  March 22, 1995
                                                  Page 2 of 4   

A review of recorded pressurizer pressure channel signals
indicated that the failures involved a gradual increase in
transmitter output over an approximate 5 minute period,
culminating in an output plateau near the upper end of the
transmitter range.  The two failures occurred approximately 10
minutes apart.  Transmitter outputs remained high following the
event.  Loop calibrations were subsequently performed, and both
transmitters showed extremely slow response.  The transmitters
were subsequently replaced and preserved for analysis.  A third
pressure transmitter on the pressurizer that had been refurbished
in the same manner and time frame did not fail.

Discussions between licensee and Rosemount personnel indicated
that the failure mode encountered at St. Lucie was not typical of
oil loss as discussed in the NRC generic communications cited
above.  A preliminary inspection of the transmitter sensing
modules confirmed that no oil loss had occurred.  However, the
high pressure side isolating diaphragm of the sensor cell of each
failed transmitter was bulged.  Rosemount stated that the failure
modes were indicative of gas entrapment in the sensor cell.  A
detailed discussion of the failure is given in a Part 21
notification by Rosemount dated March 21, 1995, (Accession No.


The failed transmitters were sent to the Southwest Research
Institute laboratory for analysis of gases trapped in the sensor
cells.  The laboratory extracted the gas from one transmitter
sensor cell and determined that it was hydrogen.  No corrosion,
galvanic action, water leakage or oil breakdown was observable. 
Gamma back scatter examination was performed to determine the
composition of the isolating diaphragm material.  This
examination indicated that the material of the diaphragms was
Monel metal instead of the Type-316 stainless steel specified for
safety-related Model 1152, 1153 and 1154 transmitters in this
application.  Monel metal is a corrosion-resistant alloy of
primarily nickel and copper which may be used in transmitters of
this type for some plant applications.  Monel is known to be
permeable to monatomic hydrogen.  

Monatomic hydrogen may be generated by a galvanic cell reaction
between Monel and stainless steel, and this may enhance the
permeation of hydrogen from the system through the diaphragm. 
Rosemount has postulated that over a period of months at power,
monatomic hydrogen permeated or diffused through the Monel
isolating diaphragms where it went into solution in the sensor
cell fill oil.  As some of the hydrogen recombined into diatomic
hydrogen (chemical symbol "H2" - the usual form of hydrogen gas),
it became trapped because the isolating diaphragm, being
relatively impermeable to H2, retained it.  Rosemount postulated
that during constant pressure operation, a sensor with H2 under
the isolator diaphragm may not exhibit symptoms or erroneous
output as the H2 may be completely dissolved in the silicone oil. 

.                                                 IN 95-20    
                                                  March 22, 1995
                                                  Page 3 of 4   

This was apparently the case with the two confirmed failures on
November 22, 1994, described above at St. Lucie Unit 1.  The
transmitters reportedly operated normally during the 16 month
period prior to the plant outage.  The monatomic hydrogen
permeating the isolator during this time had no apparent affect
on the transmitter operation.  However, the precipitating
sequence of events leading to the apparent sudden noticeable
failure (as opposed to gradual, but detectable, degradation)
involved (1) plant depressurization which allowed the entrapped
H2 to come out of solution and form a partial pressure within the
sensor cell oil volume which may have caused some deformation of
the relatively flexible isolating diaphragms, followed by (2) a
partial re-pressurization. The repressurization may have caused
the fill oil, which is the capacitor dielectric within the
sensor, to be replaced partially with hydrogen gas.  This would
lead to an increase in the output signal. In addition,
repressurization may have caused a deflection of the center
diaphragm within the sensor, also contributing to the increase in

Based on the St. Lucie Unit 1 experience, conditions most likely
to result in adverse transmitter failure consequences would be
those involving a primary system depressurization followed by a
partial or full repressurization.  Such sequences would include
steam line relief valve openings or breaks, loss-of-coolant
accidents, and steam generator overfeeding events.  In these
cases, the transmitter should function normally during the
initial depressurization.  For pressurized water reactors it is
likely that a safety injection actuation signal would be
generated if primary system pressure went below the low pressure
actuation setpoint.  During any subsequent repressurization,
multiple transmitter failures could lead to erroneously high
pressure signals which could disable interlocks, disable any
automatic reinitiation of safety injection if required, and could
lead to opening of power-operated relief valves.  In addition,
under these conditions the operator could be presented with
conflicting information on the reactor coolant system pressure,
including, for a loss of coolant accident, some information
indicating the primary system was subcooled and other information
indicating a saturated primary system.

For boiling water reactors, failed transmitters could result in
opening of the primary system relief valves and result in system
blowdown.  Failure of the pressure transmitters could also block
automatic injection.  While pump start signals would not be
affected (low level in the reactor pressure vessel or high
pressure in the drywell) low pressure injection could be
precluded by closed injection valves.  This is because low
pressure emergency core cooling system logic typically includes a
permissive which requires indication of low reactor pressure
prior to opening the injection valves.  In such a case, operators
would need to bypass the permissive and open the injection valves
from the control room.  A similar scenario was discussed in IN

Rosemount has made a preliminary determination that about 270
Model 1152, 1153 and 1154 safety-related transmitters constitute
the suspect group.  Rosemount has identified most of these by
serial number and is in the process of informing affected
utilities (see Attachment 1).  The suspect lot is believed at
this time to be limited to those units manufactured (or
.                                                 IN 95-20    
                                                  March 22, 1995
                                                  Page 4 of 4   

by Rosemount after September 1989 and is also limited to the
higher pressure transmitters of pressure range codes 6, 7, 8, 9,
and 0.  The failures at St. Lucie occurred in range code 9
transmitters.  Differential pressure transmitters, as well as
both absolute and gauge-type pressure transmitters could be

Measures such as alerting and briefing operators, conducting
special training sessions and running event scenarios on
simulators may help in responding to a pressure transmitter

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.

                         /s/'d by BKGrimes

                         Brian K. Grimes, Director
                         Division of Project Support
                         Office of Nuclear Reactor Regulation

Technical contacts:  S.V. Athavale, NRR           Mark S. Miller,RII
                     (301)415-2974                (407) 464-7822

                     Stephen Alexander, NRR       Jerry L. Mauck,NRR
                     (301) 415-2995               (301) 415-3248
1.  List of Rosemount Transmitters with Monel 
      instead of Type 316 stainless steel diaphragms.
2.  List of Recently Issued NRC Information Notices

.                                                 Attachment 1
                                                  IN 95-20  
                                                  March 22, 1995
                                                  Page 1 of 1



Arizona Public Service
Baltimore Gas & Electric
Boston Edison
Carolina Power & Light
Commonwealth Edison
Consumers Power
Duke Power
Duquesne Light Company
Ellis & Watts 
Florida Power Corp.
Florida Power & Light
Georgia Power
Gulf States Utilities
Houston Lighting & Power
Illinois Power
Maine Yankee Atomic Power Company
New Hampshire Yankee, Inc.
New York Power Authority
Niagara Mohawk Power Corp.
Northern States Power 
Omaha Public Power District
Pacific Gas & Electric
Pennsylvania Power & Light
Philadelphia Electric Company
Portland GE
Public Service Electric & Gas
South Carolina Electric & Gas
Southern Cal. Edison
Systems Energy
Toledo Edison
TU Electric
Vermont Yankee
Virginia Power
Washington Public Power Supply System
Wolf Creek NOC
Yankee Atomic

Page Last Reviewed/Updated Monday, November 18, 2013