OPERATING EXPERIENCE WEEKLY SUMMARY 93-42
October 15 through 21, 1993
The purpose of the Office of Nuclear Safety (NS) Operating Experience (OE) Weekly
Summary is to enhance safety throughout the Department of Energy (DOE) complex
by promoting feedback of operating experience and encouraging the exchange of
information among DOE nuclear facilities. The OE Weekly Summary is distributed
for information only. No specific actions or responses are required solely as a
result of this document.
Readers are cautioned that review of the OE Weekly Summary should not be relied
upon as a substitute for a thorough review of the interim and final Occurrence
Reports.
The following events were reviewed during the week of October 15 through 21, 1993.
ITEM PAGE
1. LOSS OF AIR HANDLING SYSTEMS 2
2. RADIOACTIVE CONTAMINATION OF WRISTS 2
3. SUSPECT/COUNTERFEIT PARTS 4
4. FIRE ON ROOF OF ERWIN URANIUM FUEL FABRICATION FACILITY 5
5. WATER MAIN RUPTURED BY BACKHOE 6
ADDITIONAL INFORMATION RELATED TO FOLLOWUP ACTIVITIES
1. UPDATE ON RADIOACTIVE LEAD SHIPMENT TO SMELTING COMPANY 7
2. ADDITIONAL INFORMATION FROM UNITED KINGDOM OPERATING
EXPERIENCE 8
FINAL REPORTS - SUMMARIES
1. FINAL REPORT ON EMERGENCY GENERATOR BATTERY CHARGING
SYSTEM EVENT AT ROCKY FLATS BUILDING 559 10
2. FINAL REPORT ON THE DIESEL GENERATOR EVENT AT
ROCKY FLATS BUILDING 779 11
1. LOSS OF AIR HANDLING SYSTEMS
On October 19, 1993, a worker at the Mound Plant inadvertently cut the pneumatic air
control lines for Building 50 ventilation dampers, causing the ventilation system to shut
down. This caused the loss of differential pressure control within the remaining facility
which violated a limiting condition of operation. Facility personnel discontinued all work
when the violation occurred in accordance with the limiting condition-of-operation action
statement. The worker was removing obsolete equipment in accordance with an
approved work package when the lines were cut. At the time of the event, there were
no operations involving plutonium, and no material was released. Continuous air
monitors and the stack monitor remained operational.
Prior to the event, the worker received a work package that included photographs
depicting the equipment to be removed. Facility personnel have not determined whether
the worker performed an activity outside the scope of the work package or if the work
package was deficient. They convened a formal investigation committee to determine
the root, direct, and contributing causes of the occurrence.
There have been numerous cases at DOE facilities where work was performed outside
the scope of maintenance or modification packages or where work packages were not
explicit. The incidents resulted in workers inadvertently cutting energized wires, core
drilling into electrical conduit, and performing unauthorized modifications to systems.
These events continue to emphasize the need for thorough work planning and control.
Unplanned, negative consequences to facility safety and operation can be avoided by
fully understanding the potential effects of planned work on interrelated systems and
components. In addition, these events underscore the importance of staying within the
scope of approved work documents. Workers need to have a thorough understanding
of the bounds of work activity prior to initiating work.
2. RADIOACTIVE CONTAMINATION OF WRISTS
On October 15, 1993, fixed and loose alpha skin contamination measured at up to 10,000
dpm/100cm2 was detected on each wrist of an employee at the Oak Ridge Y-12 Plant.
After repeated washings, one wrist was wrapped because 1,200 dpm/100cm2 fixed
contamination remained on it. The contamination was below the Y-12 Plant action limit
of 1,000 dpm/100cm2 within two days. A whole-body count indicated no internal
contamination, and the results were being confirmed by urinalysis. (ORPS Report ORO--
MMES-Y12DEFPGM-1993-0089)
Facility personnel reconstructed the worker's activities and determined that he probably
became contaminated the previous day while filling bottles with uranium solution.
However, the contamination was not detected during exit frisking with a hand-held survey
instrument nor by a hand monitor, both of which were located at the boundary control
station. After discovering the contamination during a routine survey the next day, health
physics technicians instructed the employee to wash his wrists. After two washings,
4,000 dpm/100cm2 fixed contamination remained on the right wrist. Health services
personnel performed further decontamination efforts on the individual that lowered the
contamination on the wrist to 1,200 dpm/100cm2, and covered the wrist with wrapping.
As a precautionary measure, health physics personnel surveyed the employee's home,
clothing, and automobile and found contamination measuring 400 dpm/100cm2, which is
below the plant limit, on the cuffs of the shirt worn the day before.
Facility personnel investigated the incident and determined that the uranium solution
probably splashed onto the worker's rubber gloves, spread to the surgical gloves
underneath, and was deposited on the wrists while he removed the gloves. Site
personnel indicated the worker was not wearing plastic sleeves on his wrists, which would
have provided additional protection. Facility personnel did not determine why the
contamination was not detected. The worker was surveyed at the boundary control
station by another individual using a hand-held frisker, a technique referred to as the
buddy system. The hand monitor, also at the boundary control station, surveys only what
is placed on top of it, which is normally only the hand. To minimize the potential of a
similar event, facility personnel planned to implement the use of plastic wrist sleeves
when filling bottles with uranium solution and to conduct classes on proper removal of
anti-contamination clothing.
Personnel at other DOE facilities reported many other instances of wrist contaminations.
The wrist area is particularly susceptible to contamination because it is frequently close
to contaminated materials and is at a transition point for anti-contamination clothing, often
becoming exposed. Wrist contaminations have been caused by careless removal of
protecting clothing, insufficient protective clothing, inadequate taping of gloves, and
extreme heat and perspiration that caused tape and gloves to loosen. In some cases,
formal written procedures defining minimum protective clothing for certain operations were
not in place.
These events underscore the need for personnel at DOE facilities to ensure that the wrist
area is adequately protected for the work being performed. Personnel should use extra
protective clothing, such as multiple layers of gloves and plastic sleeves, when the
potential for wrist contamination is high; management should formally define the need for
extra clothing for particular tasks when practical. Workers should also use caution when
removing anti-contamination clothing, particularly gloves, to prevent transferring
contamination from protective clothing to the skin. Appendix 3C of DOE/EH-0265T,
Radiological Control Manual, provides guidelines on selection and removal of protection
clothing, including detailed sequences for removing a full and double set of protective
clothing. Appendix 3D provides guidelines for personnel monitoring with hand-held
survey instruments which, according to the Appendix, should be posted adjacent to
monitoring instruments in accordance with Article 338.8 of the Manual.
While reviewing this event, NS noted that Y-12 management recently changed
radiological conduct practices to improve detection of contamination mechanisms.
Previously, workers washed their hands before surveying for contamination. Current
practice is to survey before washing, which, although resulting in an increased number
of identified and reported contamination-related occurrences, is helping facility personnel
to identify contamination sources and reduce actual contamination events.
3. SUSPECT/COUNTERFEIT PARTS
NS reviewed six incidents reported between October 18 and 20, 1993, involving
identification of suspect/counterfeit equipment at DOE facilities. On October 18, Johnson
Controls, Inc., personnel at Oak Ridge National Laboratory (ORNL) reported discovering
suspect fasteners on a snap-on puller cab, a crane, and a compressor. The suspect
fasteners were grade 5 and 8 bolts with KS, A, hollow triangle, and blank head markings.
Facility personnel replaced the bolts on the puller cab, and removed the other equipment
from service until the suspect bolts were replaced. (ORO--JCI-JCI2005HES-1993-003/4/5)
On October 19, warehouse personnel at the Pantex Plant identified a lot of 300 suspect
fasteners (nuts) received on September 29, 1993, from a distributor in Chicago along with
the manufacturer (San Shing, Taiwan) inspection certificate. Pantex procurement
department personnel contacted the DOE Albuquerque Operations Office in reference to
foreign-made fasteners and were advised that fasteners manufactured by Pacific rim
countries should be considered suspect. The fasteners were segregated and secured
in a locked cabinet until they were either destroyed or proven to meet applicable
standards. (ALO-AO-MHSM-PANTEX-1993-0056)
On October 19, personnel at ORNL reported discovering suspect fasteners in equipment
located in a general stores building. The suspect fasteners were discovered on a sliding
door support, a manlift, and racks supporting spools of electrical cables. Most of the
suspect fasteners discovered at ORNL were grade 5 and 8 fasteners with KS head
markings. Fasteners with hollow triangle, J, and blank head markings have also been
discovered at ORNL. Facility personnel reported that skilled craftsmen at ORNL are
trained to identify suspect fasteners and are involved along with the Quality Department
in the effort to remove suspect parts from service. As an example of their contribution,
craftsmen voluntarily inspected and purged bench stock at several workstations of known
suspect fasteners. (ORO--MMES-X10QUALITY-1993-0001/3/4/5)
On October 18, 1993, Quality Assurance personnel at the Hanford 100 Area K-Basins
discovered that the original factory markings on three piping unions were stamped over
with suspect hand tool markings. Facility personnel issued a non-conformance report to
document the potentially fraudulent material and returned the rejected material to the
vendor.
These events illustrate the increasing level of awareness at DOE facilities of the potential
for counterfeit and suspect parts and the importance of identifying them and removing
them from service. Increased awareness and efforts on this issue is important because
failure of deficient fasteners could result in personnel injury, equipment damage, or failure
of nuclear safety-related equipment designed to prevent or mitigate the consequences
of an accident. As demonstrated at ORNL, training as many people as possible to
identify suspect/counterfeit parts will help raise awareness and expedite the identification
and removal of such parts, resulting in improved safe operation of DOE facilities.
In previous OE Weekly Summaries, NS referenced various DOE and commercial industry
documents related to identifying, assessing, and disposing of suspect and counterfeit
fasteners (OEWSs 92-11,12,14,16,25,26, 93-05,11,20,26,30,40). Environment, Safety, and Health
Bulletin 92-4 (DOE/EH-0266), Department of Energy (DOE) Quality Alert -- Counterfeit
Parts, provides guidance on identification, testing, precautions, and disposition of suspect
counterfeit fasteners and electrical circuit breakers at DOE facilities. NRC Generic Letter
89-02 Actions to Improve the Detection of Counterfeit and Fraudulently Marketed
Products, provides information on programs that the Nuclear Regulatory Commission
considered effective in detecting counterfeit or fraudulently marked products and assuring
the quality of procured products. The document also lists numerous NRC Bulletins and
Information Notices concerning non-conforming materials and equipment and instances
of inadequate dedication of equipment for safety-related applications. Other related
documents include:
General Accounting Office Report RCED-91-6, Nuclear Health and Safety:
Counterfeit and Substandard Products Are a Government Wide Concern;
Industrial Fasteners Institute Research Report, False Grade 8 Engineering
Performance Marks on Bolting and Improper Marking of Grade 8 Nuts; and
Customs Service Report, Fastener Identification Markings.
4. FIRE ON ROOF OF ERWIN URANIUM FUEL FABRICATION FACILITY
On October 13, 1993, personnel at the Uranium Fuel Fabrication facility in Erwin,
Tennessee, declared an alert when a fire occurred on the roof of a building containing
uranium processing equipment. At the time of the event, workers were heat sealing
rubberized material on the roof, and, during this process, the fiberboard or wood ignited.
Facility personnel reported that the smoke was drawn into the building because of the
negative pressure maintained in the building. No radioactive materials were released.
Site personnel quickly extinguished the fire and downgraded the event to an unusual
event in accordance with the emergency plan. (NRC Event 262215)
There have been similar events at DOE facilities involving maintenance activities on roofs
that resulted in fires.
On April 8, 1993, a worker at the Sandia National Laboratory was using a high-
speed metal saw to cut corrugated metal panels on a supply air fan penthouse on
a roof when the fiberglass insulation began to smolder. The worker and his
foreman did not adequately plan and discuss the potential problems and
implications of cutting through fiberglass insulation with a high-speed metal saw
blade. In this instance, the fiberglass insulation backing on the panels should have
been pulled away from the metal prior to the cutting procedure. (ORPS Report ALO-
KO-SNL-NMFAC-1993-0007)
On May 29, 1991, sparks from workers welding structural supports on a roof at the
Princeton Plasma Physics Laboratory caused roof insulation to smolder.
Investigators determined the "Hot Flame Permit" for welding was issued without
considering the effect the welding might have on combustible material that was part
of the roof. The combustible material was on the opposite side of the corrugated
metal roof material, which was itself next to the welding operation. Either heat
conduction or sparks started a small fire. (ORPS Report CH-PA-PPPL-PPPL-1991-1009)
On March 21, 1991, personnel at the Fernald site using an oxygen/acetylene torch
ignited foam insulation during permitted work on a roof penthouse floor of a
laboratory building. Event investigators determined that the fire blanket was
inadequate to prevent total spark penetration and the crawl-space type environment
prevented full view of the cutting area by the fire watcher, contributing to the cause
of the fire. (ORPS Report ORO--WMCO-FEMP-1991-0019)
On September 19, 1990, Idaho Chemical Processing Plant workers were welding
heating and ventilation equipment on the roof of the Process Improvement facility
when roof insulation caught fire. Facility personnel determined that the workers
failed to provide adequate shielding over the opening between the roof flashing and
the structural steel penetrating the roof. As a result, weld spatter entered the
opening and ignited the roof insulation. (ORPS Report ID--WINC-ICPP-1990-0004)
These events underscore the importance of fire hazard permits and proper planning of
work activities when cutting, welding, or grinding on roofs. For these activities, workers
should be trained in the hazards of welding and flame cutting and the duties of a fire
watch.
5. WATER MAIN RUPTURED BY BACKHOE
On October 18, 1993, a worker at the Los Alamos National Laboratory ruptured a ten-
inch water main, releasing approximately 40,000 gallons of potable water. The 100
pounds per square inch of water pressure from the ruptured pipe caused rocks and water
to hit a nearby building, breaking window glass, causing water damage to the interior of
the building, and slightly injuring a worker standing nearby. Water also sprayed onto the
roof and entered the building through air conditioning ducts, soaking ceiling tiles and
accumulating on floors.
Workers were conducting maintenance on an underground gas line when the incident
occurred. They had located the water main below and perpendicular to the gas line, but
the spotter and backhoe operator were unaware of an abandoned valve extending from
the main. The backhoe snagged the valve, causing the rupture. Workers shut off the
water supply which caused a water outage that affected the building fire protection
system. Facility personnel classified the event as an unusual occurrence because of the
performance degradation of the fire protection system. (ORPS Report ALO-LA-LANL-
SERVICESS-1993-0027)
Numerous events were reported by DOE facilities involving damage to underground
cable, piping, conduits, and structures during excavations. NS reported many of these
occurrences in previous OE Weekly Summaries. Several of the events in this category
could have resulted in personnel injury.
On September 16, 1993, personnel at the Idaho Chemical Processing Plant
discovered that a worker had severed the 480-volt buried cable with a fence post.
The cable supplied temporary power to a crane. Workers installing the fence posts
apparently did not properly check the excavation permit prior to driving the posts.
In addition, the cable was buried only two feet deep and was not identified by
marker flags. (ORPS Report ID--MKF-MKNE-1993-0001)
On September 15, 1993, at the Savannah River Effluent Treatment Facility, a
worker damaged an underground pipe while drilling for a soil sample. Facility
personnel conducted a critique and determined that the causes included failure to
prepare an excavation permit, relocation without proper follow-up of the drilling
stakes approximately 20 feet south of the originally surveyed location, and lack of
established work control procedures. (SR--WSRC-ETF-1993-0005)
On October 10, 1992, personnel at the Los Alamos National Laboratory ruptured
an underground natural gas line with a backhoe. Even though workers obtained
an approved excavation permit and all utilities were properly marked for location,
depth of the piping could not be properly measured with available instruments and
facility personnel failed to follow procedures requiring manual exploration prior to
machine excavation. Piping was buried only 10 to 12 inches below the surface.
(ORPS Final Report ALO-LA-LANL-PHYSTECH-1991-1009)
On September 14, 1992, an underground natural gas pipeline ruptured when it was
struck by a backhoe during an electrical utility relocation project at the Lawrence
Livermore Berkeley Laboratory. Facility personnel previously surveyed the area for
underground obstructions in accordance with procedures, but a change in the job
order altered locations. The workers failed to survey the new area and
subsequently punctured an underground gas line with the back hoe. (ORPS Report
SAN--LBL-OPERATIONS-1992-0007)
On September 20, 1991, at the Grand Junction Operations facility, a worker
severed a 13,000-volt underground power cable one foot below the surface while
removing mill tailings at a Uranium Mill Tailings Remedial Action Site. Facility
drawings incorrectly showed the cable buried 4 feet deep and encased in concrete.
(ORPS Report ID--GEO-GJO-1991-1019)
Occurrence reports describing similar events note that facility procedures do not always
require documentation of temporary utilities or excavation permits for all excavation work.
These events demonstrate the necessity to thoroughly investigate the possibility of
underground obstructions prior to commencing excavations. A review of plant facility
drawings may not be adequate for locating underground cables, conduits, tanks, or
piping. In addition, personnel may not be able to obtain information from record drawings
necessary to assure correct marking of buried utilities. Workers should use hand tools
to excavate around utilities. Security and telecommunication personnel should be
consulted prior to commencing digging operations for assistance in locating buried
components. Facility personnel should assure that procedures controlling excavations
provide detailed instructions concerning steps associated with the location of underground
obstructions.
ADDITIONAL INFORMATION RELATED TO FOLLOWUP ACTIVITIES
1. UPDATE ON RADIOACTIVE LEAD SHIPMENT TO SMELTING COMPANY
NS reported an event in OE Weekly Summary 93-41 that occurred on October 4, 1993,
at the Argonne National Laboratory - East facility. Personnel discovered that
approximately 250,000 pounds of scrap lead was shipped to a commercial smelting
company without a Health Physics review. Heath Physics personnel performed a survey
of the remaining material at the lead storage facility and identified lead bricks with
activation levels as high as 1x10-10 curies/gram. NS reported that approximately 40,000
pounds of the scrap shipped off site was not smelted and that it would be surveyed by
Health Physics personnel prior to further processing. On October 15, 1993, NS received
updated information that no contamination/activation was detected on the remaining scrap
material. Facility personnel stated that determination of the root cause, corrective
actions, and lessons learned is still being investigated. (ORPS Report CH-AA-ANLE-ANLESSD-
1993-0001)
2. ADDITIONAL INFORMATION FROM UNITED KINGDOM OPERATING EXPERIENCE
As part of an cooperative information-sharing agreement between the United States
Department of Energy and the United Kingdom, the operating experience information
presented in this section was excerpted from the Statement of Nuclear Incidents at
Nuclear Installations published by the Health and Safety Executive, London, England.
The report describes incidents at nuclear installations during the second quarter of 1993.
A copy of the entire news release may be obtained by contacting: Library, Health and
Safety Executive, Room 021, Baynards House, 1 Chepstow Place, London W2 4TF,
telephone 011-44-071-243 6000, fax 011-44-071-727 6268.
1. INCIDENT 93/2/1
WINDSCALE AND CALDER WORKS (BRITISH NUCLEAR FUELS)
On 18 May 1993 a planned inspection was carried out of the drums stored in
a building known as B300 on the Sellafield site and a pool of syrupy liquid was
observed on the floor. The liquid appeared to originate from a single drum but
two adjacent drums were also affected. The drums concerned were of mild
steel construction and contained plutonium contaminated waste arising from
operations on the site. No release of airborne activity had occurred nor was
there any contamination of personnel.
Following discovery of the event, BNFL made arrangements to sample the
liquor and prevent any further spread beyond the already affected area. The
analysis of the liquor indicated that approximately 0.5 TBq (unquote: or 13.5
curies) of alpha activity had leaked out. This quantity is in excess of column
7 of Schedule 2 of the Ionising Radiations Regulations 1985. The event was
reported to the HSE's Nuclear Installations Inspectorate (NII) on the evening
of 18 May and subsequently to the President of the Board of Trade. BNFL
classified the incident as International Nuclear Event Scale (INES) Level 2.
During the next few days BNFL carried out detailed investigations to establish
the drum history and the material present in each drum. In addition the
sequence of recovery operations and decontamination of the area was
planned. A methodology was established and the safety implications
considered by BNFL in accordance with conditions attached to the nuclear site
license. The operations to recover the drums were successful and the
affected area decontaminated. The leaking drum has been doubly contained
and along with the two other drums placed in an area removed from other
drums within the store and where regular inspection can be undertaken. No
contamination of personnel occurred during these recovery operations which
were completed on 4 June 1993.
BNFL have investigated the cause of the incident and NII are monitoring the
investigation and will follow up on actions arising from the BNFL report and
their own studies.
2. INCIDENT 93/2/4
CHAPELCROSS (BRITISH NUCLEAR FUELS)
During refuelling operations on 26 May 1993 on Reactor 2 an employee was
momentarily exposed to radiation from an irradiated fuel element during an
unusual sequence of operations which followed from a component failure on
the grab in the fuelling machine which is used to handle fuel elements.
An investigation by HSE's Nuclear Installations Inspectorate (NII) has shown
that the unusual sequence of operations, although permitted by written
operating instructions, challenged the installed system of plant interlocks to the
extent that they failed to prevent radiation exposure to the employee.
Modifications to the plant and changes to working procedures are now to be
made to prevent similar occurrences.
It is fortunate that the employee only incurred a radiation dose of 1.1 mSv
during this deviation from normal operation practice. (The annual legal dose
limit to the whole body as set out in schedule 1 part 1 of the Ionising Radiation
Regulations 1985 is 50 mSv with an overall requirement also to keep radiation
exposure as low as is reasonably practicable).
3. INCIDENT 93/2/5
SELLAFIELD (BNFL)
The last quarterly statement of incidents set out the circumstances of an
incident at BNFL Sellafield, in which a Quality Assurance Inspector sustained
a cut to his elbow, whilst inspecting a weld in an enclosed cell on a plutonium
transfer line. At that time, biological monitoring was still in progress and it was
not possible to provide an assessment of the dose he had received. This
work has now been completed and BNFL have reported that the person
concerned has received a committed dose equivalent to the bone surfaces of
628.6 mSv. This is in excess of the annual dose limit for a single organ as set
out in schedule 1 part II of the Ionising Radiations Regulations.
The Nuclear Installations Inspectorate has conducted its own review of the
incident and concurs with BNFL that the most likely cause was the presence
of a sliver of glass, of unknown origin, in the working environment. NII have
concluded that no further regulatory action is warranted in this instance.
Note:
The International Event Scale referenced in the Health and Safety Executive document
was designed by an international group of experts convened jointly by the International
Atomic Energy Agency and the Nuclear Energy Agency of the Organization for Economic
Co-operation and Development. The purpose of the Scale is to promptly and consistently
communicate to the public the safety significance of reported events at nuclear
installations. Events are classified at seven levels, with lower levels 1 through 3 termed
incidents and upper levels 4 through 7 termed accidents. Events that have no safety
significance are classified as level 0/below scale and are termed deviations. Each level
has specific criteria relating to events in terms of off-site impact, on-site impact, and
defense-in-depth degradation. Examples of classified events include the 1986 Chernobyl
accident (level 7), the 1979 Three Mile Island accident (level 5), and the 1989 fire at the
Vandellos nuclear power plant in Spain (level 3).
FINAL REPORTS - SUMMARIES
This section of the OE Weekly Summary discusses events that have been filed as Final
reports on the ORPS. These events contain new or additional lessons learned that may
be of interest to personnel within the DOE complex. The description of each event, its
causes and lessons learned are edited from the original ORPS Final report.
1. FINAL REPORT ON EMERGENCY GENERATOR BATTERY CHARGING SYSTEM
EVENT AT ROCKY FLATS BUILDING 559
On June 22, 1993, two maintenance electricians at Rocky Flats Building 559 discovered
an out-of-tolerance float current supplied to batteries that provide current for starting the
emergency diesel generator. Facility personnel measure the float current monthly as part
of the limiting condition-of-operation surveillance on the emergency diesel generator. The
current flow rate was 953 milliamp (mA), which exceeded the charging current
parameters of 21 and 479 mA. Incident investigators attributed the direct cause to an
external phenomenon when a high ambient temperature caused a high charging current
to the emergency diesel starting batteries. They determined that a contributing cause of
this occurrence was an inadequate procedure that did not clarify the fact that a high
current charge rate alone cannot be indicative of a degradation in long term battery
condition and performance without some indication of low cell voltage. Facility personnel
initiated one of the corrective actions to issue a document modification request requiring
a reading of battery cell voltages when the battery charger current flow rate exceeds the
upper tolerance limit of 479 mA. (ORPS Reports RFO--EGGR-ANALYTOPS-1993-0088)
Another corrective action involved distribution of a lessons learned letter to address the
issues identified as a result of this incident. The following was excerpted from the letter.
The batteries and associated charger were replaced during the Resumption
effort, and utilize a maintenance-free style of battery. Manufacturers
recommendations are used to describe the expected parameters for an
acceptable surveillance. There have been numerous surveillance failures
caused by the lack of definition of appropriate criteria in the surveillance
procedures. This is another example of that same problem. The procedure
acceptance criteria for float current were carefully addressed for the very low
current values experienced at optimum conditions, because of previous
experience with measuring very low currents. The minimum value was
established as just above the minimum capability of the measurement device
to confirm that current flow existed. The upper value was established based
exclusively on the manufacturers calculated value for optimal conditions, and
no allowance was taken for fluctuations in ambient conditions.
This occurrence pointed out that additional data must be gathered if the float
current is above the optimal value, so that the time operability of the starting
batteries can be determined. In this case the ambient temperature of the
room was higher than normal, and the float current was responding to battery
bank current demand. Subsequent battery cell tests indicated that the
batteries were in a fully operable condition. The procedure was revised to
incorporate appropriate tests if float current is found above optimal value.
Surveillance procedures of Preventive Maintenance procedures must
incorporate an adequate set of tests to verify the operability of equipment, and
must not only repeat manufacturers recommendations that are normally based
on a given set of conditions.
2. FINAL REPORT ON THE DIESEL GENERATOR EVENT AT ROCKY FLATS BUILDING
779
On June 15, 1993, operators at Rocky Flats Building 779 observed unacceptable
fluctuations in the speed of the emergency diesel engine. They classified the event as
an unusual occurrence because of degradation of Class A equipment. Incident
investigators attributed the direct cause to deficiencies in the design of the diesel fuel
transfer system that caused the fuel pump to provide inadequate fuel oil flow to the day
tank.
A system engineer identified these deficiencies during the testing of the generator six
months prior to the incident, but facility procedures were not changed to incorporate
corrections. The systems engineer determined there was a large margin of error between
the actual fuel oil day-tank level and the indicated level in the sightglass during operation
of the emergency diesel generator. He attributed the level discrepancy to a combination
of the dynamic properties of the system and installation of the sightglass between the
emergency diesel fuel pump suction piping and the atmospheric vent piping for the day
tank. The engineer observed that when the actual tank level reached approximately one-
third, the indicated level in the sightglass was zero, thus providing a direct flow path from
the day-tank atmospheric vent to the emergency diesel fuel oil pump suction, which
resulted in a sudden and complete loss of fuel oil pressure.
The emergency diesel is fed from the gravity-fed day tank, which holds 11.75 gallons of
fuel, and the emergency diesel burns approximately 30 gallons of fuel per hour operating
under a full load. The systems engineer noted that the pump used for this test was not
capable of maintaining sufficient fuel in the day tank to maintain an uninterrupted supply
to the emergency diesel. Therefore, the tank level dropped to one-third, allowing air into
the fuel supply lines. Facility personnel stated that the engineer's report was not made
formally into a document and transmitted to Operations personnel, thus, the problem was
not corrected. Incident investigators attributed the root cause for the occurrence to facility
management for not taking proper steps to correct the problems after the engineering
evaluation identified the problems and the necessary action to correct the situation.
The systems engineer informed Engineering management of the deficiency in the fuel
pump through his weekly report highlighting the study performed six months earlier. He
distributed the report to Operations personnel to close out a work package and to facility
personnel who requested the original study. On June 24, 1993, facility personnel
completed the systems engineer's recommendations to improve the reliability of the fuel
oil transfer system. (RFO--EGGR-ANALYTOPS-1993-0084)
SAFETY NOTICES UNDER DEVELOPMENT
Note: The Office of Nuclear Safety encourages input related to the development of Safety
Notices. If you have any questions, comments, or information concerning events or
issues similar to the following, please contact Mr. Dick Trevillian, Office of Nuclear Safety
at (301) 903-3074.
1. NS has identified a number of events related to the loss of annunciators and other
safety-related equipment because of problems involving 120-VAC/125-VDC systems
at DOE and commercial facilities. NS is reviewing potential generic problems
associated with the adequacy of 120-VAC/125-VDC systems at DOE facilities.
2. NS is developing a Safety Notice concerning problems with Uninterruptible Power
Supplies (UPS).
3. NS is considering development of a Safety Notice related to control of work at
electrical substations and switchyards.
4. NS is working with Lawrence Livermore National Laboratory and DOE-SF personnel
to develop a Safety Notice on cracking in ventilation ducting.
5. NS is considering developing a Safety Notice related to fuel oil supplies for
Emergency Diesel Generators (EDGs).
6. NS is developing a Safety Notice to address uses of independent verification for
equipment positioning.
7. NS is developing a Safety Notice related to maintaining important alarm and
monitoring systems at facilities undergoing transition and decontamination and
decommissioning.
SAFETY NOTICES PREVIOUSLY ISSUED
Safety Notice No. 91-1, "Criticality Safety Moderator Hazards," September 1991
Safety Notice No. 92-1, "Criticality Safety Hazards Associated With Large Vessels,"
February 1992
Safety Notice No. 92-2, "Radiation Streaming at Hot Cells," August 1992
Safety Notice No. 92-3, "Explosion Hazards of Uranium-Zirconium Alloys," August 1992
Safety Notice No. 92-4, "Facility Logs and Records," September 1992
Safety Notice No. 92-5, "Discharge of Fire Water Into a Critical Mass Lab," October
1992
Safety Notice No. 92-6, "Estimated Critical Positions (ECPs)," November 1992
Safety Notice No. 93-1, "Fire, Explosion, and High-Pressure Hazards Associated with
Drums and Containers," February 1993
Safety Notice No. 93-2, "Control of Temporary Modifications," September 1993
Copies of NS Safety Notices may be requested from: Nuclear Safety Information Center,
Office of Nuclear Safety, U.S. Department of Energy, Room S161, GTN, Washington, DC
20585