ML20128B493
| ML20128B493 | |
| Person / Time | |
|---|---|
| Site: | Fort Saint Vrain  |
| Issue date: | 05/13/1985 |
| From: | Lanning W, Moses D NRC OFFICE FOR ANALYSIS & EVALUATION OF OPERATIONAL DATA (AEOD), OAK RIDGE NATIONAL LABORATORY |
| To: | |
| References | |
| CON-FIN-B-1661 NUDOCS 8507030256 | |
| Download: ML20128B493 (14) | |
Text
-
<>.,...~.y THE ANALYSIS AND EVALUATION OF RECENT OPERATIONAL EXPERIENCE FROM THE FORT ST. VRAIN HTGR
.D. L. Moses Nuclear Operations Analysis Center Oak Ridge National Laboratory Oak Ridge, Tennessee W. D. Lanning Office for Analysis and Evaluation of Operational Data U.S. Nuclear Regulatory Commission Washington, D. C.
T By acceptance of this articse, the publisher or recipient ackr,owledges the U.S. Governinent's right to ret.in. nonexclo.ive. roy.ity.fre.
license in and to any Copyright r,
covering the article.
Paper to be presented at the IAEA Specialists' Meeting on Safety and Accident Analysis for Gas-Cooled Reactors, Oak Ridge, Tennessee, May 13-15,1985
- Work performed for Nuclear Regul atory Commi ssion under Interagency Ag'eement 40-547-75 (NRC Fin No. B1661) at Oak Ridge National Labo-rar.ory, operated by Martin Marietta Energy Systems for the U.S. Depart-me7t of Energy under Contract No. DE-AC05-840R 21400.
8507030256 850513 PDR ADOCK 05000267 p
THE ANALYSIS AND EVALUATION OF RECENT OPERATIONAL EXPERIENCE FROM THE FORT ST. VRAIN HTGR D. L. MOSES Nuclear 0perations-Analysis Center Oak Ridge National Laboratory Oak Ridge, Tennessee United States of America W. D. LANNING Office for Analysis and Evaluation of Operational Data
- United States Nuclear Regulatory Commission Washington, D. C.
United States of America
.lbstract THE ANALYSIS AND EVALUATION OF RECENT OPERATIONAL EXPERIENCE FROM THE FORT ST. VRAIN HTGR The U.S.
Nuclear Regulatory Commission's Office for Analysis and Evaluation of Operational Data has established an extensive program for screening, analyzing, and evaluating the operational experience data from all commercial nuclear power plants in the United States.
This program is designed to provide feedback from field experience with actual operating events to the NRC's continuing efforts to assure the public's health and safety.
Oak Ridge National Laboratory provides technical assistance to AE00 to evaluate the operating experience for Fort St. Vrain.
-In November 1981, Fort St. Vrain operated briefly at 100% power as part of the successful completion of power and flow oscillation detec-tion testing following the installation of region constraint devices.
In February 1982, the redesign of the heliun purification systen into two trains was completed to improve the availability and operability of the Helium Circulator Auxiliary System which had been identified as the major cause of water ingress due to upsets in buffer' helium supply.
Also, in February 1982, a precritical reactor scram resulted in two con-trol rod. pairs failing to insert apparently due to high moisture condi-tions prior to restart; however, this event was believed at the time to be precluded in the future due to the promised lower incidence of mois-ture ingress.
Water ingress events continued to be a frequent problem caused most often by electrical / control systen upsets.
The most recent such event was in June 1984, when the circulator upset led to a moisture ingress large enough to require reduction in temperature to protect graphite.
The precise level of moisture was not determined at that time; however, it was high enough to cause icing of chillers in the helium purification train several hours later.
As a result, the reactor was exposed to m
2 s
several hours of "high" moisture, loss of purified helium flow to control rod mechanisms, and finally a reactor scram in which 6 of 37 control rod pairs failed to insert automatically.
Evidence has also been uncovered that high moisture has caused the transport of volatile chlorides throughout the reactor resulting in corrosion of stainless steel control rod cables and possibly some helium circulator bolts.
Moisture has also caused leaching of B203 contaminant from the reserve shutdown materials, precluding the complete dumping of material during a surveillance test.
1.
INTRODUCTION As described in previous international conferences,1,2 the U.S.
Nuclear Regulatory Commission's (NRC's) Office for Analysis and Evalua-tion of Operational Data ( AE00) has established an extensive program for screening, analyzing, evaluating, and disseminating the safety-related operating expericnce data from all commercial nuclear power plants in the United States.
The NRC also participates in the exchange of opera-tional event information with other countries through the Nuclear Energy Agency and through bilateral agreements.
AE0D evaluates operating ex-perience from foreign reactors through its review of incident reports and provides IAEA member countries with reports of significant U.S.
reactor operating experience including that fran Fort St. Vrain.
The goal of this program is to ensure through feedback that lessons are learned and improvements implemented based upon actual reactor operating experience.
Within this program, particular attention is given to the analysis of those reportable occurrences (R0s) documented in Licensee Event Reports (LERs).3 An LER is required by regulation to be submitted by the licensee for any safety-related or safety-signi fi cant event as defined in the LER Rule (Title 10, Code of Federal Regulations, Part 50.73, or 10 CFR 50.73).
The Nuclear Operations Analysis Center (NOAC) at the Oak Ridge National Laboratory (ORNL) provides technical assist-ance to AEOD in the evaluation and documentation of operating event ex-perience from the Fort St. Vrain High-Temperature Gas-Cooled Reactor (HTGR), which is the only U.S. nuclear plant of its type.
The Fort St. Vrain operating experience
- to be discussed here in-cludes notable safety-related events which have occurred since late 1981 when ORNL was first contracted to provide technical assistance to AE00.
Earlier Fort St. Vrain operating experience through the time of success-ful full-power testing in November 1981 has been summarized by the licensee and the reactor vendor, GA Technologies, Inc. (GA), in papers presented at several different forums during 1982 (Refs. 4-7).
In addi-tion, extensive and very useful detailed evaluations of preoperational
- Considerable information is provided to the NRC by the licensee con-cerning the safety related issues discussed in this paper.
The reader is encouraged to contact the authors for a listing of reports relevant l
to the discussed events.
s 3
and startup testing and of the rise-to-power operating experience through completion of the first refueling outage in August 1979 have been compiled into 'a series of reports under the sponsorship of the Electric Power Research Institute (EPRI).8-10 Finally, the U.S. Depart-ment of Energy's Fort St. Vrain Improvement Planll provides a summary of
- the major operational limits which have affected the plant since start-up.
The' events discussed here are categorized based on the major sys-tems affected, namely, (1) primary systen and reactor vessel, (2) elec-trical systems, and (3) the reactor building.
In all cases to be dis-
. cussed, the lessons to be learned are vigilance and prevention.
These lessons translate into the need for the recognition and control of un-expected situations and of their potential for branching effects.
At Fort St. Vrain, these lessons are found in the effects of moisture ingress, in the challenges experienced to the supply of essential elec-trical power, and in controlling the environment of the reactor build-ing.
2.
DISCUSSION OF REACTOR EVENT EXPERIENCE Since late 1981, Fort St. Vrain has been shut down about 28 of the last 41-months.
The effects of operating events have contributed to much of that shutdown time. Although the following discussion addresses
- safety-related events, the effect on plant availability is also an important consideration.
2.1 Primary System and Reactor Vessel The Fort St. Vrain primary system including the core, control rod mechanisms, coolant ducts, steam generators, and helium circulators is enclosed in a Prestressed Concrete Reactor Vessel (PCRV).
The steam generator and helium circulators sit in PCRV penetrations utilizing double closures.
The interspace between the double closures of these penetrations as well as the control rod penetrations is purged by puri-fied helium at pressures higher than the primary system. The major con-cern raised by recent operational events is that of keeping coolant inside the reactor vessel and contaminants out.
In June 1980 and throughout 1981, problems were experienced with purified purge helium leaking from a steam generator penetration inter-space into the reheat steam of the secondary system.
This leakage led to an excess of noncondensable gases in the condenser.
Since an appar-ent crack in an inaccessible seal weld in the stean generator reheater module was the source of the helium leak, the only practical solution was to reduce pressure in the penetration to just above reheat steam pressure and below primary system coolant pressure.
Radiation monitors on the condenser air ejector are used to detect primary coolant leaks.
A concern was raised by AE00 regarding possible corrosion due to steam leaks into the penetration if purge flow pressure was inadvertently reduced, but the licensee believed that there were adequate moisture
4 monitors upstream in the purge line to detect steam ingress under low
~
heliun flw conditions.
The problen of moisture in the penetration interspaces has recently arisen again.
In September 1984, the licensee confirmed a suspected water leak in the "A" helium circulator penetra-tion interspace.
The moisture detectors in this interspace had alarmed repeatedly for several months but were thought to be malfunctioning by
. the licensee.
The actual presence of substantial moisture was not recognized until the detectors were removed for repair and water ran out.
Subsequently, in November 1984, surveillance testing of a reserve
- shutdown systen (RSS) hopper revealed that half of the boronated graph-
. ite - balls were stuck together with boric acid crystals.
The crystals had resulted from moisture leaching out the B203 contaminant in the B 4,C.
The moisture apparently entered the hopper through the purified heliun inlet purge line which was found to have corrosion upstream, pos-sibly due to moisture entering the purified helium stream after exiting the heliun purification train This failure is similar to an event which occurred several years earlier leading the NRC to require surveillance testing of the kind which detected the present problem. The licensee is replacing the boronated balls with new material with " low" B203 contami-nant in all hoppers.
Moisture ingress continues to be the dominant problem at Fort St.
Vrain. There have only been two minor steam generator leaks, one which occurred in November 1977 with the plant at 50% power and the other in the fall of 1982 following a circulator trip and moisture ingress.
In the latter case, nearly two months were necessary to confirm and locate the leak due to the presence of substantial moisture in the shutdown reactor.
The timing of the water injection of bearing water accumu-lators to ensure lubrication during circulator trips is the primary cause of moisture ingress.
In turn, the loss of bearing water has been initiated frequently by electrical disturbances affecting bearing water supply instru nentation.
Before the separation of the buf fer helium supply into two trains in February 1982, the loss of buffer helium in a single train contributed to multiple circulator trips and frequent moisture ingress events.
However, this modification, which was per-formed as part of the Fort St. Vrain Improvement Plan, 'has failed to alleviate the frequency and severity of moisture ingress.
The most recent and notable moisture ingress event occurred in June 1984.
An. electrical system disturbance caused a circulator trip and bearing water ingress. The operators were apparently unable to diagnose the extent of moisture ingress and continued operating in an attempt to purge the primary coolant of moisture, as allowed by Technical Specifi-cations.
Several hours later, heliun purge flow was lost when the chillers iced up in the only available helium purification train. Ulti-mately, pressure / temperature mismatch was caused during reduction in power by mismatch as "high pressure" was programming downward. Moisture was not the cause of high pressure.
The reactor scrammed on the high pressure / temperature mismatch trip, and six control rod pairs failed to insert automatically.
The trip occurred actually when the pressure was m
5
'l far below normal operating pressure owing to the gas shrinkage during tne cooldown taking place.
The failure of the control rods to insert automatically upon receipt of a valid scram demand is a common mode failure that constitutes a partial ATWS event (i.e., Anticipated Tran-sient Without Scram) - a significant safety concern.
The reactor was well subcritical even with six rods stuck out.
The reactor has since been shut down.
This event is similar to a February 1982 subcritical startup scram in which two control rod pairs failed to insert and in which the manual scram was preceded by high moisture conditions and a itss of control rod
. purge flow. Based on investigations to date, the exact cause of failure has not been determined.
Also discovered was the fact that moist pri-mary systen coolant would penetrate into the affected area of the control rod mechanism even if purified helium purge flow were not lost.
The mechanism housing was not sufficiently well sealed at cable penetrations. The licensee is exploring various design modifications to restrict further moisture ingress from the bearing water system into the primary system and to reduce the pathways for moisture entry into the control rod mechanisms from the primary system.
A related problem associated with the moisture ingress is the leaching of volatile chlorides from various sources within the reactor and their deposition throughout the primary system.
In August 1984, a stainless steel control rod cable broke and was subsequently found to have chloride-induced stress corrosion cracking.
The steel cables are being replaced with corrosion resistant inconel cable.
In January 1985, the disassembly of a Fort St. Vrain circulator under repair at GA found a steel closure bolt apparently affected by chloride-induced stress cor-rosion cracking.
Concern has been raised about the possibility of the pooling of condensed volatile chlorides on the PCRV liner.
The chloride problem is still under invostigation.
The problem represents a lack of making the connection between the level of chloride contaminants in var-ious primary system components, the possibility of moisture-induced leaching, and the potential susceptibility of other components to chloride attack.
In the past, concern about the effects of moisture ingress had been concentrated almost exclusively on graphite corrosion.
In March 1984, the tendon wires of the PCRV were subjected to a five-year surveillance.
A number of corroded and broken wires were found.
Subsequent investigation has determined that a microbiological agent is at work in the presence both of the sulfonate grease used on the tendon wires and of oxygen from air ingress into the tendon enclo-sures.
The steel wires are being attacked by acetic and formic acids formed by the bacteria. The integrity of the PCRV does not appear to be degraded, but the licensee is developing an accelerated surveillance program and is considering the use of a positive pressure nitrogen blanket to prevent oxygen from entering the tendon enclosures. However, recent tests have shown that the tendon enclosures may be too leaky to hold the nitrogen blanket ef fectively.
._____,__m
-- - - - - - - - - - - - - ~ - - ' - " " " - -
6 2.2 Electrical Systems The essential and emergency electrical power sources have been designed for an adequate level of independence, redundancy, capacity, and testability to meet the required level of safety.
The alternate sources of essential and emergency power include:
1.
The main turbine-generator set via the unit auxiliary trans-former (UAT) 2.
Five 230-kv transmission lines via the reserve auxiliary trans-former (RAT) 3.
Two independent standby diesel generator (DG) sets rated at 1210 kw and each stated to be capable of supplying essential loads for safe shutdown and cooling 4.
DC batteries 5.
DC essential instrumentation power from six separate AC busses At Fort St. Vrain, the unlikely long-term loss of AC electrical power can result in a loss of forced cooling (LOFC) due to the loss of elec-trical systems supporting the motive power (steam or water) for the helium circulators.
At least one standby DG set is needed to assure safe shutdown and core cooling without fuel damage.
An extended LOFC could lead to substantial fuel damage due to overheating and the release of fission products into the primary system.
A permanent LOFC consti-tutes a Fort St. Vrain Design Basis Accident (OBA).
Therefore, in re-sponse to NRC concerns on the potential for disruptive faults affecting congested cable areas, the licensee has installed an Alternate Cooling Mode (ACM) electrical systen with independent cabling and a separate 2500 kw DG located away from the main plant structure.
The ACM dupli-cates certain functions of the standby DG sets, but, in the event of the loss of all other power sources, the ACM meets minimum requirements to ensure safe shutdown by manual actuation of the reserve shutdown system, continued cooling of the PCRV liner to contain fission products, depres-surization of the PCRV through the purification systen to limit heat loads on the PCRV upper barrier plates and to filter circulating fission products, and exhausting and monitoring of ef fluents fran the reactor building.
Operation of the ACM alone during a permanent LOFC will not preclude core damage but should ensure that fission products are con-tained within the PCRV.
Because of the importance of the AC power sources, any event af-fecting their availability receives immediate and intensive attention.
Since depressurization must be initiated within two hours of the initia-tion of an extended LOFC fran full-power conditions, troubleshooting and repairing electrical system failures nust be accomplished in less than two hours during emergencies.
Since 1981, the offsite power grid sup-plying Fort St. Vrain has failed once during high winds accompanying a snow storm on May 17, 1983.
At that time, the reactor had been shut i
7 down for about two months so that there was no immediate emergency.
However, during this event, the "A"
standby DG set was unavailable be-cause of repairs on corroded and stuck check valves in a raw water cool-ing supply line.
(The affected valves had apparently not been inspected since initial installation.)
Prior to the event, the "B" standby DG set was running and closed onto essential busses in parallel to offsite power.
The loss of offsite power caused an overload trip of the "B"
standby DG set.
The "B" DG was restarted but could not be closed onto essential loads because time-delay relays failed to reset automatically after load shedding.
The load shedding relays were reset by pulling fuses to deenergize the relays, but this effort took 45 minutes to diag-nose and accomplish.
An AE0D Engineering Evaluation 12 was performed that addressed the potential adverse effects of having offsite and on-site essential power sources closed onto the same loads, especially dur-ing grid disturbances when such alignments may exist. A similar failure of the load shedding relays on the same deenergized busses was experi-enced on December 18, 1984. The reactor was shut down and the plant was deliberately isol ated from the grid in order to perfonn a standby DG load sequencing surveillance.
Both DGs failed to complete raquired load sequencing, and the reconnection to offsite power was delayed because of the failure of the relays to reset.
The licensee has now committed to installing a manual deenergization of the affected relays to allow more rapid operator action.
Other operating experiences with the DG sets show that there have also been occasional problems with engine exhaust temperature sensors.
During the load sequencing surveillance test on December 18, 1984, a failed cell on one bank of new DG batteries resulted in a low exhaust temperature signal, forcing shutdown and declutch of both engines on the "A" DG set. Another independent, random failure of a temperature switch caused the trip of one engine of the "B" DG set.
With only one DG set at SW, capacity, the automatic load sequencing logic was not met, there-by requiring manual action which was to switch to offsite power.
That attempt was delayed by the load shedding relay problen described previ-ously.
However, the situation was compounded but not really affected by a timer motor failure on one of two redundant load sequencing delay timers used to give the DG sets enough time to reach operating status before being closed onto essential busses.
The failure of both timers could have similarly prevented automatic load sequencing. Troubleshoot-ing these concurrent failures took sufficient time that the surveillance could not be completed the same day. Although plant personnel would ex-pedite repairs and exhaust all avenues of recovery during an actual loss of AC power, the potential for common-mode failure of automatic load sequencing and the time delays experienced have raised questions which are still under review by the Of fice of Nuclear Reactor Regulation re-garding the reliability of emergency power systems at Fort St. Vrain.
The ACM DG set has also experienced random failures.
On October 17, 1981, the ACM DG failed to start during surveillance testing because of a faulty starter solenoid.
The Technical Specifications allow the AC'i to be inoperable up to seven consecutive days for maintenance, and two days were required to repair the solenoid since at that time a l
l
3 replacement part was not readily available.
On July 25, 1983, the ACM DG failed to. start because too much water had evaporated from the cells of the starting batteries, which are located'in an adjacent building to which the batteries had been moved to avoid freezing and boiloff prob-lems experienced previously when located in the ACM DG shack.
Ventila-
-tion had been lost in the building, resulting in high ambient tempera-tures.
Recharging the batteries to full charge required about 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br />.
The -incidents experienced with the ACM illustrate the kinds of problems and potential for delay in availability which are part of the evaluation of operating events involving electrical systems.
E During loss of offsite power, the main turbine generator set is ex-pected to remain on line, providing house loads.
The large heat capac-ity of the HTGR core and lack of critical heat flux concerns eliminate 4
the -'need for an anticipatory reactor trip on turbine trip in common use for light water reactor plants with one turbine.
Probabilistic analyses by GA Technologies for similar large HTGRs claim that the main turbine will trip at a rate of only 10-1/ demand during grid failures.
In the past, Fort St. Vrain has experienced difficulty in achieving successful turbine runback from 70% power. The 100% power test performed in Novem-ber 1981 culminated in a successful turbine runback following the suc-cessive trips of two helium circulators; however, this event did not involve loss of grid power. The licensee is still committed to perform-ing a B-series startup test of turbine. generator load shedding from and
. recovery to 100% power.
This test will probably be perf9rmed near the end of the current cycle.
Other electrical system upsets have also occurred.
The mosc notable was caused by a cooling oil pressure sensor on a newly installed 4160/480 volt load center transformer.
The fault occurred twice leading to transients on instrument busses and caused circulator trips which resulted in moisture ingress.
The first occurred on fiay 29, 1984, dur-ing Cycle 4 rise-to-power, and the second occurred on June 22, 1984, initiating the sequence of events leading up to the partial ATWS event on June. 23, 1984.
The second event occurred because plant personnel were unable to diagnose the cause of the first event and apparently had no reason to suspect a faulty sensor.
Since 1981, both the unit auxiliary transformer (UAT) and reserve auxiliary transformer (RAT) have experienced faults.
On March 9, 1983, with reactor at 30% thermal power while moisture was being removed from the primary coolant, a phase-to-ground fault occurred on the VAT due to an arcing short caused by the moisture leakage into a bus duct from the duct cooling - system (not connected with the primary coolant system).
There were no moisture detectors on the ducts.
The damage included burned cables and melted insulators and required ten days for repairs.
Essential loads were being carried by offsite power from the RAT at the time of the incident, so no transient resulted.
On December 8,
- 1983, high winds at the site caused a fire detector to come loose and mal-function, activating the RAT deluge system. Since most of the essential loads were being carried by the UAT at that time, there was only a minor transient involving some building cooling systems, and the RAT was
-w-- -, - - -, - -,, - -
,.v,
,m
,.,,,,-.n,,..,,_m.,,-.,,,.m
l-9 restored within 20 minutes. Both of these incidents illustrate the sus-ceptibility of the plant auxiliary transformers to externally induced events which could have led to more severe transients or loss of essential power if a combination of these or other events had occurred.
In summary, during the past three and a half years, the Fort St.
Vrain electrical systen has been challenged frequently.
Circumstances
.at the time of. each challenge have been such that except for the June 1984 event, the transients were minor and the safety-related implica-tions appear negligible at first glance.
However, a combination or dif-ferent sequence of events has potential safety implications, and efforts
, are under way to improve the reliability of the system.
2.3 Reactor Building (RB)
The reactor building (RB) is a filtered, vented confinement build-ing enclosing the PCRV and essential piping and cabling.
Because the PCRV provides a radiological barrier, personnel access to the RB is available even with the reactor operating at full power.
Also, the layout of essential equipment within the RB but outside the PCRV is made in principle to allow safe shutdown and cooling even if environmental
. conditions in one part of the building cause failure of part of the equipment and limit personnel access.
Some of the notable reactor building events are discussed as follows.
On August 26, 1981, with the reactor shut down, hot slag from weld-ing a pipe hanger fell into oil-absorbent material placed beneath a cable tray to soak up fluid which had leaked fran a hydraulic shock absorber (snubber).
A fire started, damaging 31 of 36 cables in the tray including 16 essential cables.
The fire was extinguished by con-tractor personnel in about five minutes, but was not reported to the licensee for two hours.
On July 26, 1983, a welder was found working inside the RB without a fire-resistant drop cloth to catch hot slag or a fire watch.
In this case, there was no fire. On January 26, 1985, with the reactor shut down, a reactor scram input signal was initiated by high neutron flux rate.
The scram was attributed to the effect of elec-tromagnetic fields generated by a welding machine high frequency starter operating in close proximity to the flux detector cables.
Further investigation" has revealed that the welding machine starter was ap-parently grounded to the cable conduit.
There were apparently no " safe i
grounds" designated for welding machine use.
These events illustrate the importance of controlling the RB environment to hazardous conditions which may be generated by ' workers, tools, or other maintenance equip-ment, and the events provide lessons for' all operating reactors.
During August 6-8, 1983, a reheat steam leak underneath the PCRV caused impedance variations in cables of the helium circulator speed-high trip.
The impedence variations led to loss of one channel of the speed-high trip logic (2/3 channels generate a trip); however, loss of electrical current on a speed-high trip channel did cause the trip of a
+
speed-law channel.
The licensee diverted the steam leak and later repaired the steam leak and replaced the cable with one having higher
- -et 4
+y-s-r--*
-r7i*(--t'e F1r v-*T-
-"='---P'"-"a r-
-W'-
- * * - - - + - - ' * - - * +
9W *r
- '79'w**?ryy++g,g y,up-e" p-=-a--
10
~
quality insulation. This event was reviewedl3 in some detail because of the concern that the unlikely occurrence of a postulated PCRV penetra-tion failure (DBA No. 2) could lead to an extended LOFC (DBA No.1) due to circulator damage because of overspeed.
This situation could only occur if the speed-high trip functions were all lost due to the effects of the hot hellun blowdown and if an independent failure occurred on the speed-low trip logic.
Our review found that adequate cable separation existed to precluda loss of all speed-high cables and that the design pressure differential could not be experienced simultaneously by all four heliu.n circulators for any given PCRV penetration failure.
3.
CONCLUSIONS Fort St. Vrain operating experience since late 1981 has been domi-nated by long periods of shutdown often due to safety implications of operating events which required significant maintenance.
Many events which have obviously not caused shutdown have raised safety-related con-l cerns.
The sources and the effects of moisture ingress are the most significant operational problems.
Although none of the challenges ex-perienced to the essential electrical supplies have proven to be a seri-ous problen with regard to adequate core cooling, there have been a number of potentially serious challenges which were minimized by the operating mode of the reactor.
The evaluation of the operating events at Fort St. Vrain has pro-l vided insight into necessary improvements in HTGR design and opera-tion.
The lessons learned can only be obtained through experience and are particularly valuable for the development of advanced designs such as the HTGR.
The analysis of operating data provides insights into component and systen interactions and reemphasizes the importance of integrating the design, construction, operation, and requirements for emergency response in commercial power reactors.
Such experience illus-trates that diligence, attention to the technical details, an intuitive-ly questioning attitude, and the delegation of responsible authority are necessary to ensure safe operation and availability.
In addition to ORNL's and NRC's independent assessment of operating experience from Fort St. Vrain, the licensee, the U.S. Department of Energy, and indus-try groups maintain vigilance of both domestic and foreign reactor ex-perience applicable to the HTGR.
The continued evaluation and feedback of operating experience is essential to ensuring the viability and success of the HTGR as a safe and economic competitor in the commercial l
nuclear power market.
The licensee for Fort St. Vrain remains optimistic about the operational and safety advantages of Fort St. Vrain and HTGRs.
The licensee has major effort under way to refurbish the control rod drives and to make modifications which would mitigate or eliminate the moisture ingress problem.
The plant is expected to be ready to return to operation this j
summer.
l l
t I
e e
11
~
References 1.
C. Michelson and C. J. Heltemes, "The Use of Operating Experience For Improving Nuclear Power Plant Safety in the United States of America," Paper No. IAEA-CN-42/271, International Conference on Nuclear Power Experience, Vienna, 13-17 September 1982, Interna-tional Atomic Energy Agency, Vienna (1983).
2.
C. Michelson and C. J. Heltemes, "New Trends in the Evaluation and Implementation of Safety-Rel ated Operating Experience," Paper No.
IAEA-CN-39/2.2a, International Conference on Nuclear Power Plant Safety Issues, Stockholm, 20-24 October 1980, International Atomic Energy Agency, Vienna (1981).
3.
Licensee Event Report Sy stem, NUREG-1022, U.S.
Nuclear Regul atory Commission, Washington, D.C. (September 1983).
4.
H. L. Brey and H. G. Olson, " Fort St. Vrain Experience," Conference on Gas-Cooled Reactors Today, Bristol, 20-24 September 1982. British Nuclear Energy Society, London (1982).
5.
R. F. Walker and C. L. Rickard, " Operating Experience with the Fort St. Vrain Nuclear Power Station," Paper No. IAEA-CN-42/279, Inter-national Conference on Nuclear Power Experience, Vienna, 13-17 Sep-tember 1982, International Atomic Energy Agency, Vienna (1983).
l 6.
W. A. Simon and G. C. Bramblett, " Fort St. Vrain Reactor Performance and Operation to Full Power," Third Japan-U.S. Seminar on HTGR l
Safety Technology, June 2-3, 1982, Volume 1,
NUREG/CP-0045 (BNL-NUREG-51674),
Brookhaven National
!.aboratory,
- Upton, New York (1982).
7.
H. L. Brey and W. A. Graul, " Operation of the Fort St. Vrain High-Temperature Gas-Cooled Reactor Plant," American Power Conference, Chicago, April 26-28, 1982, Volume 44, Illinois Institute of Tech-nology, Chicago, Illinois (1982).
8.
K.
R.
Van Howe et al., Operation and Testing Experience During Startup and Initial Operation at the Fort St. Vrain HTGR, EPRI NP-890-SY (Summary Report), prepared for the Electric Power Research Institute by the S. M.
Stoller Corporation,
- Boulder, Colorado
( August 1978).
9.
K. R. Van Howe et al., Fort St. Vrain Experience - Phase 4, Follow-On Studies, EPRI NP-1214, prepared for the Electric Power Research Tiistitute by the S. M.
Stoller Corporation, Boulder, Colorado (November 1979).
- 10. K. R. Van Howe, Fort St. Vrain Experience - First Refueling / Main-tenance Outage, EPRI NP-1292, prepared for the Electric Power Research Institute by the S. H.
Stoller Corporation, Boulder, Colorado (December 1979).
I m.
m
s 12
- 11. Fort St.
Vrain Improvement
- Plan, DOE /SF/10748-T1, Proto-Power Management Corporation, Groton, Connecticut (May 1981).
- 12. M. Chiramal, Engineering Evaluation No. E401, Temporary. Loss of All AC Power Due to Relay Failures in Diesel Generator Load Sheddi ng Ci rcu i try, AE00/E401, U.S. Nuclear Regulatory Commission, Washing-ton, D.C. (January 1984).
- 13. D.
L.
Moses, Technical Review Report No. T423, Inoperability of Helium Circulator Overspeed Trip Channels Due to Impedance Varia-tions in Speed-Sensing Cables Exposed to Steam Leak, AEOD/T423, U.S.
Nuclear Regulatory Commission, Washington, D.C. (October 1984).
\\
\\,
\\
,w
. - ~ -
s
'i /
i pa nsou
{
f g,.
I UNITED STATES e
j,,5
'e NUCLEAR REGULATORY COMMISSION
-f
/
WASHINGTON, D.C. 20555 June 26, 1985 4
i i
i MEM0 FOR: Jim McKnight, RSB i
k j
FROM:
Vicki Yanez, PPMB j
i l
SUBJECT:
TRANSMITTAL 0F SPEECHES i,
t 2
l l
Attached are two copies of a speech to be 1
4 sent to the PDR and DCS. We have filed the i
i NRC Form 426.
.)
1 i
i I
l 1
Speech:
"The Analysis and Evaluation of Recent Operational Experience From the l
l Fort St. Brain HTGR" t
I I
3 b
l I
i t
5 t
i i
I I
i i
W i..
+.
l l
,