ML20210R036

AEOD Annual Rept 1985, Including Review of Operational Data & Summary of AEOD Activities & Accomplishments

ML20210R036
Person / Time
Issue date:	04/30/1986
From:	NRC OFFICE FOR ANALYSIS & EVALUATION OF OPERATIONAL DATA (AEOD)
To:
Shared Package
ML20210R017	List: ML20210R012 ML20210R036
References
TASK-AE, TASK-S601 AEOD-S601, NUDOCS 8605160022
Download: ML20210R036 (168)

Text

{{#Wiki_filter:.=, s A A a Y e AE00/S601 0FFICE FOR ANALYSIS AND EVALUATION g' 0F OPERATIONAL DATA (AE0D) ANNUAL REPORT 1985 1 APRIL 1986 i i g. 8605160022 860500 PDR ORG NEXD PDR g, .j e.

e CONTENTS 1. INTRODUCTION..................................................... 1 2. ORGANIZATION AND STAFFING........................................ 4 3. COMMENTS AND OBSERVATIONS ON 1985 NUCLEAR POWER PLANT OPERATING EXPERIENCE............................................. 5 3.1 Results from Selected AE0D Studies.......................... 5 3.2 Overview of the Three Incident Investigation Team Reports... 8 3.3 Abnormal Occurrences Involving U.S. Nuclear Power Plants.... 13 3.4 Analysis of the 1985 Licensee Event Reports................. 14 3.5 Assessment of Licensee Event Reports Quality................ 18 3.6 Reactor Scram Experience in 1984 and 1985................... 20 3.7 Engineered Safety Features Actuations....................... 30 3.8 Studies of Loss or Unavailability of Safety System Function.................................................... 36 4. COMMENTS AND OBSERVATIONS ON 1985 OPERATING EXPERIENCE AT OTHER LICENSEES........................................................ 46 4.1 No n re a c to r Ev e n t s........................................... 46 4.2 Medical Misadministration Events............................ 54 5.

SUMMARY

OF AE00 ACTIVITIES....................................... 58 5.1 Reactor Operations Analysis Branch (R0AB)................... 58 5.2 Program Technology Branch (PTB)............................. 73 - 5.3 Nonreactor As sessment Branch (NAS).......................... 85 5.4 IncidentInvestigationStaff(1IS).......................... 90 6. STATUS OF AE00 RECOMMENDATIONS................................... 94 7. STUDIES CURRENTLY IN PR0GRESS................................... 151 APPENDIX A -

SUMMARY

OF 1985 ABNORMAL OCCURRENCES APPENDIX B -

SUMMARY

OF 1985 REACTOR SCRAM RATES 9 l l l i

o Office for Analysis and Evaluation of Operational Data Annual Report 1985 1. INTRODUCTION In March 1979, the most serious, and certainly the most costly, accident in the history of the U.S. nuclear industry occurred at TMI-2. As a result of that accident and the associated studies and investigations, it became clear that improvements were required in the way that the NRC and the nuclear community used operating experience to help identify and resolve problems which could jeopardize public health and safety. One of the Commission's early responses to that need was to establish the Office for Analysis and Evaluation of Operational Data (AE0D). This office reports directly to the Executive Director for Operations and is specifically dedicated to the collection, assessment, and feedback of operational data. A specific mission of the office is to analyze and evaluate operational safety data associated with all NRC-licensed activities. This includes commercial power reactors, and radioactive material and fuel cycle licensees. The office also coordinates the overall NRC operational data program and serves as the focal point for interaction with outside and foreign organizations performing similar work. The office's objectives and some of its tasks and activities are highlighted below. 0FFICE FOR ANALYSIS AND EVALUATION OF OPERATIONAL DATA Objectives Collect, screen, analyze, and feed back operating experience to appropriate NRC offices, the nuclear community, and the public for all NRC-licensed activities. Coordinate the overall NRC operational data program. Administer the NRC's Incident Investigation Program. Specific Tasks and Activities Screen U.S. and foreign operational events for significarce. Systematically and independently analyze operational events. Seek trends and patterns and track performance indicators which may indicate potential safety problems.

Develop and support the procedures for and establishment of Incident Investigation Teams. Develop and track recommendations for action by other NRC Offices for resolution of safety issues. Develop and coordinate operational data retrieval systems, including foreign data, j-Prepare and coordinate Abnormal Occurrence reports. Prepare Power Reactor Events reports and other feedback documents. 1 Provide documentation of U.S. events for reporting to the Nuclear Energy Agency's Incident Reporting System. Serve as principal point of contact with ACRS, INP0, and l: NSAC on operational data activities. During 1984 and 1985, the NRC sought to identify potential improvements in the I existing program for the investigation of significant operational events in response to a request by Congress. Brookhaven National Laboratory was 1 contracted by the NRC to parform a study to assess and evaluate improvements-to the program. Subsequent to that effort, the NRC staff identified a number of changes in the approach to investigating significant events. The most note-worthy change is the investig{ation of such incidents by a multi-disciplined Incident Investigation Team IIT), made up of technical experts from the various NRC offices. These teams prepare a single comprehensive report-for each incident describing the event, setting forth the relevant facts, identifying causes, and presenting findings and conclusions. In 1985, AE00 responsibilities were expanded to include the development and support of the Incident Investigation Program. AE0D is part of an overall NRC program to review operating experience in order - to identify specific events and generic situations where the margin of safety established through licensing has been degraded, and to identify and recommend corrective actions that will restore the originally intended margin of safety. AE0D's role in the program is to analyze operating experience. independent of regulatory activities associated with licensing, inspection or enforcement. AE00 maintains an awareness of the studies undertaken by other organizations such as the NRC Offices of Nuclear Reactor Regulation (NRR) and Inspection and - Enforcement (IE), and the Institute of Nuclear Power Operations (INPO). - Normally, AE00 will not overlap or duplicate the study efforts of other organizations unless a particular need or special circumstance exists. Thus, AE00 does not review ~in-depth all' events or operating problems because many problems identified through operating experience are being studied extensively and comprehensively by other organizations. Thus, to be aware of all identified safety problems and lessons of experience, one must integrate the output of many NRC offices and outside programs. 2

4-L i The recomendations contained in AE0D studies are not final NPC positions. They are internal recommendations for action by appropriate NRC program offices (e.g., NRR, IE) or Regional Offices. The program office or Regional Office is responsible for reviewing and, where appropriate, implementing AE0D recommenda-tions. A. written response to each recommendation is required and a formal action-tracking system has been established. 4 t

• 4 3

t 2. ORGANIZATION AND STAFFING During 1985, the Incident Investigation Staff (IIS) was established in AE00 to administer the Incident Investigation Program. The roles and responsibilities of the IIS are described in Section 5.4. Wayne Lanning was selected as Chief of the IIS, and two additional staff members have been selected. In addition, upon the retirement of Thomas A. Ippolito, Frederick J. Hebdon was selected as the Deputy Director of AEOD. Mr. Hebdon had previously been the Chief of the Prngram Technology Branch in AE00. Finally, during FY 85, responsibility for the Accident Sequence Precursor Progran and a number of programs associated with the collection of reliability data'were transferred from RES. As a result, one FTE and $575K in program support funds were transferred from RES to AE00. With the addition of the IIS described above and the transfer of resources from RES to AE00, AE00's staffing level is now 43 FTE. The Commission had requested an increase in AE00 staffing level; however, subsequent budget reductions have required that AE00's staffing level for FY 86 be the same as FY 85. The AEOD organizational structure and current staffing are shown below: Director C. MELTEMES Secretary C. CAtJ ACMER Deputy Director F. NE5 DON Secretary R. McCANN Adelaistrattwe Assistant C. THOMPSON I i i F30 CRAM TECMM01DCT BRANCM REACTOR OPERATIONS HONREACTOR ASSESSMENT INCIDENT INVESTICATION Acting Chief F. stESDON Secretary R. Wit.LIAMg Acting Chief S. RURIN Chief K. 3 TACE Chief W. LANNING Secretary R. RAgat er ry J. ZECh Secretary E. MOFFER e $. PETTIJ0tel RIACTOR SYSTDt3 REACTOR SYSTDt3 E. TRACER Fa0 CRAM DEVE14FMDrf DATA MANACDE3f7 SECTION 1 SECTt001 2 SECT 10It SECT 10er Chlaf S. RUSIN Chief P. IAM Chief B. Dee IC Chief J. CRo0Ks

1. BELL F. SOBE T. CINTURA N. ORNSTEIN t'. MARFER E. 30TLE E. LEEDS D. ZUKOR B. BRAPT

5. MASSARO VACANCY S. SAIAH E. 310 GINS F. CROSS-PRATHER VACANCT R. TR1FATHI T. WOLF F. O'REILLT F. M4fElIIIC DICDIEERING SECT 10er Chief M. CNIRAMAL I'.!!=s" C. 550 Figure 1.

Office for Analysis and Evaluation of Operational Data (4/86) 4 1 1

3. COMMENTS AND OBSERVATIONS ON 1985 NUCLEAR POWER PLANT OPERATING EXPERIENCE AE0D activities include studies to identify and evaluate potentially significant events and safety concerns involving U.S. commercial power reactors, based on events reported to the NRC by nuclear power plant i licensees. During 1985, a number of these events were individually reviewed and studied collectively as part of Trends and Patterns activities in AE00. Studies were conducted that provide a broad overview of the industry and assess the operating experience and characteristics of the industry as well as individual plants. This section provides some observations and perspectives on selected aspects of operating experience in 1985, and provides overviews of Licensee Event Reports as well as the most significant events forwarded to the Consnission for consideration as Abnormal Occurrences. 3.1 Results from Selected AEOD Studies In 1985 a number of AE0D studies were completed or progressed to the point where conclusions could be formed. Several of the significant reports are highlighted in this section. Discussed below are the: (1) AE0D Special Program Study on the effectiveness of licensees' and the NRC's Operational Experience Feedback Programs; (2) AE00 Case Study on the frequency and potential serious-ness of events involving losses of decay heat removal capabilities; (3) AE0D Case Study on the frequency and potential seriousness of BWR system overpressurization events; and (4) AE0D studies on valve operability and reliability. 3.1.1 Operational Experience Feedback Programs Subsequent to the TMI-2 accident, various studies highlighted a need for - increased attention to the feedback of operating experience by the NRC, the industry, and each licensee in order to use the lessons of experience to prevent seriou:: nuclear incidents from occurring. One action taken by the NRC was to require that each nuclear power plant licensee have a formal (i.e., controlled by procedures) program to incorporate the lessons learned from operating experience into plant hardware and procedures, and to assure that these lessons are communicated to plant operators and other personnel. In the fall of 1984, AE00 initiated a survey to assess the usefulness of operational data feedback documents and to determine the characteristics of licensees' cperating feedback programs. AE00 staff made on-site visits to seven licensees, met with INP0 to discuss their programs, and issued a draft report for peer review within the NRC in late December 1985. Copies of the draft report were also provided to INP0 and to the licensees visited, and a copy was also placed in the NRC Public Document Room. The general observations from this study are discussed below. (1) Over the past 5 years, signH icant improvements have occurred in the ' attention paid to operating experience feedback; however, many licensee programs have not yet achieved full performance levels in all areas. Most plants are making moderate, not extensive, use of their in-house operating experience, and in general are making less use of the large body of knowledge associated with events-and concerns that originate elsewhere in the industry. Thus, we still do not have.high. confidence that all 5

+ i-licensee programs are thoroughly exploring and using all of the important lessons learned industry-wide. i (2) The NRC, INPO and others within the industry are expending large amounts of resources on the screening and assessment of operational data and the preparation and dissemination of operating experience feedback documents. However, extensive in-depth assessments of this information by licensees i and expected actions in terms of changes in plant hardware, procedures, and training prograas _ based on this industry-wide feedback were not-readily apparent in some cases. As _a result, similar operational events continue to occur and the yield from these activities in terms of corrective actions and improved operator knowledge and capability seems lower than expected. (3) TMI Action Item I.C.5 intended that the-lessons of experience be routinely provided to all operating personnel in an efficient and effective way. However, plant operators (licensed and unlicensed) and other plant / personnel are not routinely'and effectively provided much operating experience information beyond events that occur in their own plants. (4) The NRC-requirements and guidance in this area are very general'(i.e., few technical program goals and objectives, evaluation criteria, or expected activities are specified). This aspect-has contributed to the l establishment of widely diverse licensee programs with varying effective-i

ness, i

(5) NRC does not routinely verify (i.e, inspect) the effectiveness of current operating experience feedback practices on a plant by plant basis, although the NRC has attached great importance to this activity. At:this. time, however, the general NRC requirements do not provide a sufficiently definitive basis to permit a meaningful evaluation by the NRC. (6) The large volume and the diversity of operational experience feedback may i be degrading the effectiveness of the feedback programs. Duplication and overlap divert resources available to effectively use operating experience feedback. In February 1986, AE00 staff accompanied INPO on a plant evaluation as an observer in the operating experience review area. The results of-the observations and discussions with INPO on their programs,'and AE0D's recommendations for action, will be addressed in the. final AE00 report to be issued in 1986. ~ ' 3.1.2 PWR Losses of Decay Heat Removal Function U.S. PWR operating experience has shown that the loss of decay heat removal (DHR) ' function has been a frequent and potentially ' serious occurrence. One-hundred and thirty losses of operating-DHR systems were reported between 1976 and 1983, involving about 500 reactor-years of operation. The frequency-equates to more than '12 events each calendar year.- ~ The operational data clearly indicate that: (1) human factors are the root cause of most of the reported events; (2) inadequate procedures and personnel-errors during testing, surveillance, maintrnance, and emergency repair operations'are a dominant root cause (almost two-tSirds of the events); (3) the techniques used for planning 6 .. ~.

l h-and coordinating activities during plant shutdown vary widely from plant to plant and are frequently inadequate to prevent the occurrence of the events; (4) the existing procedures and equipment associated with RCS level monitoring during. plant shutdowns are frequently inadequate and failure prone; (5) operator aids are often not readily available to assist in the detection of off-rormal plant conditions while in modes 4, 5 and 6; and (6) alarms and annunciators are not always conveniently located to enable the operators to use normal and emergency procedures during shutdown periods. Furthermore, i operators are not usually provided with, or trained in, the use of emergency procedures associated with equipment failures which may occur during periods of plant shutdown, nor are they provided with time-margin information to assist in recovering from a loss-of-DHR event. l 4 Overall, no significant decrease in the frequency of loss-of-DHR events has been detected on an industry-wide bas's. Although the plants involved in these events have recovered from the loss-of-DHR function before any serious conse-3 quences occurred, the frequency of these events continues to be of concern for 1 PWR safety during periods of plant shutdown. 3.1.3 BWR Overpressurization of Emergency Core Cooling Systems j; In the last decade, at least eight operational events at U.S. BWRs involving an 4 ~ actual or potential overpressurization of an emergency core cooling system have occurred. Each of these operational events involved the failure of a testable isolation check valve on the injection line of an emergency core cooling l-system. Five of the eight events involved an additional failure of the second 1 and final isolation barrier, the motor-operated injection valve. Four of these five events occurred with the plant operating at power, thereby resulting in an actual overpressurization of the unprotected emergency core cooling system. Operating experience indicates that the dominant causes of an inadvertently { open testable isolation check valve were related to problems associated with the nonsafety-related air operator attached to the check valve. Additionally, i. the majority of these problems involved maintenance errors. The causes for the normally closed motor-operated injection valve to inadvertently open during power operation were all related to personnel errors committed prior to or during surveillance testing at power. 4 In all the overpressurization events, the lower pressure emergency core cooling system was pressurized to reactor operating pressure and temperature. Although none of the overpressurization events led to a failure of low-pressure system piping, pumps or valves, such a failure would not be impossible. l Thus, each of the eight operational events can be considered a precursor to a loss-of-coolant accident outside containment due to a potential loss of. integrity of the lower pressure emergency core cooling system. Such an i accident would involve the discharge of high energy reactor coolant outside containment, which could also disable one or more of-the safety systems required to mitigate the accident. Collectively, these operating events ,j' indicate that the likelihood of an interfacing loss-of-coolant accident is higher by at least two orders of magnitude than had been previously assumed. i 7

3.1.4 Failures of Motor-0perated Valves Used in Safety Systems Operating experience from 1981 to 1985 for motor-operated valves shows that failures leading to valve inoperability are continuir g to occur. These failures are similar to the ones observed in earlier years. The overriding conclusions from AE0D studies concerning valve assembly opera-bility and performance / reliability are that the data demonstrate current i methods and procedures at many plants are not adequate to assure that valves will operate when needed, and that this issue of performance and reliability is a very c mplex subject which involves several technical disciplines. It is apparent that valve operation during surveillance testing under conditions which are less arduous than actual operating conditions, can and has provided a false sense of security about valve performance and reliability. Although there have been few examples of valve assembly failure to operate when needed, the June 9,1985 event at Davis-Besse in which auxiliary feedwater isolation valves failed to open illustrates the potential safety concerns. One of these valves also failed to operate under actual demand conditions in 1984. The first failure to operate was investigated, corrective action was imple-mented, and the valve was declared operable after a successful operability test under conditions that were less severe than actual demand conditions. Further, data obtained with signature tracing equipment testing of valves in several nuclear plants indicate a widespread or common mode operability problem. A high percentage of the safety-related motor-operated valves exhibited abnormalities or degradations that could cause failure to operate under some anticipated conditions. Therefore, assurance of operability appears to be strongly dependent upon the diagnostic capability to assess and evaluate failures to operate so that root causes of failures are determined correctly. The data clearly demonstrate there is a need for the capability to determine the actual setpoint of switches, because incorrect settings can render valves inoperable. These steps in turn require a thorough understanding of equipment operation, system interaction loads, and procedures to set switches and protective devices. The entire process requires close cooperation between service technicians, equipment designers, plant systems and component engineers, plant operators, and operations management. 3.2 Overview of the Three Incident Investigation Team Reports The Incident Investigation Program (IIP) was established by the ED0 in June 1985. Between June and December 31, 1985, Incident-Investigation Teams (IITs) were established to investigate three significant nuclear plant operating events: Davis-Besse in June, San Onofre Unit 1 in November, and Rancho Seco in December. These events and the associated plant conditions and practices may not be representative of the industry at large, and three events are a small sample. Nevertheless, the resulting IIT reports note common characteristics in a number of areas and a number of potentially generic issues emerge which 8

4 may warrant increased NRC attention in order to help prevent or minimize such events in the future. These are discussed individually below: (1) The lessons of experience were not being aggressively sought out, assessed and acted upon at these plants. All three IITs commented that for the events studied, the plant personnel did not seem to have aggressive and effective programs to identify and correct the root causes of. equipment failures and operational difficulties, and thus to minimize or eliminate the probability of recurrence. l The.IIT for the Davis-Besse event (NUREG-1154) concluded: "The licensee has a history of performing troubleshooting, maintenance and testing of equipment, and evaluating operating experience related to equipment in a superficial manner, and as a result, the root causes of problems are not always found and corrected." A problem that recurred during the Davis-i Besse event on June 9,1985, was the failure of the AFW isolation valves (the operator could not reopen these valves); a similar problem with one of these valves had been experienced during a previous event in March 1984. t The IIT for San Onofre (NUREG-1190) concluded: "It appears that the 1 Southern California Edison process for evaluating and following_up on 1 events may not be sufficiently thorough and systematic to assure that failed components are detected and adequately explained." For example, during the San Onofre event on November 26, 1985, five safety-related check valves were found to have failed, and as a result there was a potential for a common-mode failure.of the heat sink provided by the three steam generators. Yet, the team did not have high confidence that all five failed check valves would have been found by the licensee without assistance from the team. Further, it was noted that the testing methods and acceptance criteria for these valves were not adequate, and that i records of previous maintenance work on these valves were incomplete, difficult to locate and lacked sufficient detail to determine what was i done. Additionally, in reviewing past experience, the-team learned that j-during the 1970s, a check valve had failed in a similar manner at San Onofre, and that the licensee had the practice of inspecting and refur-b' ', bishing these valves during refueling outages. Thus, the inadequacies of these valves were. known, yet the root cause(s) of the problems had J h evidently not been identified, nor properly corrected, p The IIT for the Rancho Seco event (NUREC-1195) concluded: "It appeared to the team that SMUD personnel found the process of troubleshooting in a highly controlled, systematic and well documented manner, as proposed by the Team, to be quite different from their usual maintenance practices." In this case the IIT noted that the plant weaknesses and vulnerabilities that led to and complicated the event were largely known to the licensee by virtue of a number of previous events and related analyses and studies, but adequate plant modifications were not made so that the event would have been improbable, or that its course or consequences would have been significantly altered. 4 9

1 i-- (2) Operating events can impose c significant burden on the operating personnel on duty. The three events investigated by the IITs indicated that the operating crew can be faced with: common-modd failures of safety systems; multiple independent failures of plant equipment; spurious alarms and incomplete plant instrumentation; the need for prompt actions to be taken outside the control room; fire alarm and fire suppression system actuations; and' interfaces with. security features and systems. Some of the complicating factors.taced and successfully resolved by the operating crew during the three events are discussed individually below. During the Davis-Besse event, the operators were faced with: two commoa-mode failures of the AFW system (closed isolation valves and tripped AFW i. pump turbines); a PORV that remained open; unexpected closure of the MSIVs and loss of the second main feedwater pump; an operator error; inadequate i instrumentation to properly implement critical procedures regarding loss of feedwater; interfaces with security systems and personnel; spurious transfer of an-AFW pump to an alternate water source; an important nonsafety pump physically locked out of service; loss of the source-range nuclear instrumentation; and both of the trains of the safety parameter display system not being available. Altogether, there were i 12 malfunctions within a 30-minute period. The operators had to cope t with these problems in a prompt and effective manner at 1:30 a.m. when f only limited assistance was available. i During-the San Onofre event, the operations personnel were_ faced with: a l complete loss of all inplant ac power; reduced feedwater and loss of stcam generator level indication; a severe water hammer that resulted in a leak i from a feedwater check valve; the. failure of a safety-related check valve which also caused a rupture in a flash evaporator; an actuation of the fire suppression system;. spurious indications that a safety; injection had occurred; spurious actuation of the emergency notification system; j incomplete instrumentation regarding steam generator blowdown. status; two malfunctions of the security access control equipment; and an inability to use normal plant equipment to cool turbine-side loads. All of these difficulties were-overcome in about a 1-hour period between 5 and 6 a.m. when, again, only limited assistance was available. During the Rancho Seco event, the operators had to respond to the loss of the integrated control system (ICS). They found that they had no-control over ainumber of important valves in the plant, and that some of:these valves had moved to new positions (in a number of cases detrimental to plant recovery). Local action was.taken to shut these valves, yet one valve could not be moved, another was not completely shut,.and a _ third was broken and remained halfway open. As a result, a severe overcooling transient occurred; the_ pressurizer emptied; a bubble. formed in the i reactor vessel head; feedwater from!one steam generator overflowed into the steam lines; a makeup penp was severely damaged, and an emergency entry was necessary into a radiation area to manually close a valve; some instrumentation readings were spurious; a smoke detector and a stack j, radiation monitor alarmed; an operator collapsed at the plant control panel; noise made communications in the plant difficult, and two way-r I. X 10 .i .~ n_ _.....,.

i radios did not work; an operator lost his security badge and had to be escorted to the control room; and an equipment operator found it expeditious to climb a 5-foot security fence to get to plant equipment. All of these factors were resolved in an approximately 1.5-hour period starting at 4:13 a.m. in the morning when, again, only limited assistance was available. (3) Maintenance and test practices may not provide high confidence that all plant equipment will perform reliably and predictably in off-normal situations. During the Davis-Besse event, a number of unexpected equipment failures occurred. These included the failure of the two AFW isolation valves to open, the tripping of the two AFW pump turbines due to overspeed, and the unexpected closure of the main steam isolation valves. Each of these equipment failures suggests inadequacies in maintenance and/or test practices. The two AFW isolation valves (as well as a main steam supply valve) had not been teste<1 under actual and credible differential pressure conditions, and the maintenance practices used to set the torque bypass switches resulted in the switches being incorrectly set. The AFW pump turbine tripped on overspeed because of moisture present in the steam. Evidently, the AFW system had never been tested in this configuration and, as a result, the design acceptability of this arrangement had never been verified (i.e, this design inadequacy was not recognized). Finally, the pressure indicators used in the steam feedwater rupture control system (SFRCS) were changed (as part of the equipment qualification program), but the significance and effect on the MSIVs were not detected and corrected through acceptance tests. In addition to these maintenance and test-related failures, two design deficiencies were identified by the. team in their investigation, including a failure to meet the criteria single failure in the AFW system. Previously it was believed that the AFW system was not vulnerable to a loss of function because of a single failure. The San Onofre IIT found that five safety-related check valves degraded to the point of inoperability during a period of less than a year without detection; their failure led to a water hammer and a leak from a feedwater check valve, and a rupture in a flash evaporator /feedwater heater. The Team noted that the potential contributors to this problem were inadequate maintenance, inadequate in-service testing, inadequate component design and inadequate consideration of the effects of reduced power operations. Further, it was noted that: the maintenance records for these valves were either missing or lacked specificity on what was done; the inservice testing records were inconsistent; the testing procedure was not rigorous; the test acceptance criteria were subjective; the testing frequency was open-ended; and the test did not assure detection.of the failures found, in the case of the Rancho Seco event, such factors as the repositioning of I certain valves and the loss of control from the control room were known before the event, and yet there were a number of unexpected factors. The event investigation brought out that no maintenance was being performed on l manual valves such as an AFW manual isolation *;alve (and as a result it gi

.I e was frozen open and could not be closed during the event). Further, the event identified unexpected effects on certain instrumentationfarf plant equipment by the loss of ICS power that were not previously identified through the test programs. The investigation by the Team also identified a number of potential design issues such as: (a) the need for operation of the ICS and the main steamline failure logic (NSFL) for,certain.postu-lated accident sequences;.(b) whether the AFW system will dperate reliably and with full control without ac power; and'(c) cettain design weaknesses in the ICS and how the system responds to restoratiom ot power. (4) Training, plant procedures _ and instrumer.tation may r.ok i e.oroviding all of the knowledge and guidance needed by the operato_r_s.- s The Davis-Besse event was-made mure serious by operator errors in actuating the SFRCS and in resetting the AFW pump turbine overspeed trip and relatching the turbine valves. ThEse operator errors were believed by athe Team to be the result of inadequate training and the poor labeling and placement of the SFRCS-actuation panel. The operator indicated that he had no classroom or simulator traininh.on SFRCS actuation, and had never operated this system or been at a control panel when it was actuated by others. The IIT noted that -the failure of the. equipment operators to - manually reset the overspeed trips and open the trip throttle valves was - due to their lack of" knowledge and experience (although training'had beefr conducted onall three operations), and that the plant instrumentation 'was not adequate to properly implement plant procedures calling for initiation of feed ard bleed cooling. Further, this event emphasized that unless-emergency procedures are very clear.or precise with regard to initiation of " drastic" action, operator delay and reluctance to take action should be expected. -l The San Onofre team observed that the plant procedures did not provide adequate and complete instructions covering the methods and steps to identify and resolve ground faults o'n a transformer. In this situation, the operating staff, with management approval, followed an ad hoc process to deal with the ground, and as a result, the. grounded bus 1C was paralleled with the normally grounded bus IA. ' The Team noted that evidently the training, plant personnel knowledge, or procedures did not alert tra operating staff to the danger of this approach. Further, it was noted that some operators lacked detailed plant knowledge in a number of l areas. ' An" example was the operator errors made in attempting to return L power to the site. Other procedure and training related problems were an incorrect value specified for the RHR valve pressureninterlock and the lack of guidance ~as to when to load the diesel generators or how long they could run without overheating (without ac power). Instrumentation difficulties noted ineduded multiple spurious safety injection (SI) and the lack of indication in the control room of steam generator blowdown (note: there was a delay in the operators rccognizing that steam generator blowdown automatically initiates when containment isolation was reset). Finally, the Team noted that there was no guidance and apparently no training for operators regarding what is an acceptable flow rate into ~a voided feedwater line to prevent' water hammer. 12 ')

The Rancho Seco event was initiated by a single failure in a crucial, but nonsafety-related system, the integrated control system. The Team noted that the emergency operating procedures do not address the loss of ICS, and that neither classroom nor simulator training was provided on the overall plant response to the total loss of ICS de power or the restoration of power. Additionally, several plant operators failed to notice that the two main ICS switches were in the c:(-position. The Team concluded that the operators did not adequately understand the ICS system, and that on-the-job training may not have been adequate. The Team also concluded that operator training and procedures were not adequate to resolve the conflict between avoiding the PTS region and regaining pressurizer level, and to some extent the procedure provided conflicting indications of the appropriate priorities. Procedures had not been modified to direct the operators to alternate controls 'to shut certain valves, and the operators did not recall this option. Further, the operators failed to recall that the suction valve for the makeup pump closes on a safety injection; as a result the pump was severely damaged. During the event, non-licensed operators did not accurately determine the position of the AFW flow control valves and consequently one valve was damaged. These problems were attributed to training weaknesses. Instrumentation did not immediately aid in identifying the cause of the transient. The annunciator covering the total loss of ICS power also serves as a trouble alarm for two other unrelated problems (fan failure and loss of one dc power supply). It was also noted that although the " Annunciator Procedures Manual" was not used, it would have been of no value (L.e., it provided no useful guidance). Additionally, the Team noted that there were spurious instrumentation indications in the control room during the event (e.g., the main feedwater flow recorder) which misled the operating staff. Conclusions IIT reports tend to confirm the observations resulting from other studies. For example, the importance of an effective operational experience assess-ment program at each plant continues to be reaffirmed. Operator training reflecting multiple failures is necessary to assure realism and adequate preparation, and perhaps additional actions by licensees can to help reduce operator stress during significant events. Adequate test and maintenance programs are essential in order to have high confidence that equipment will operate routinely and properly in off-normal, perhaps under accident, j conditions. And proper operator performance depends upon the support and ~ knowledge provided by training, plant procedures and plant instrumentation, yet problems in these areas have been noted by the IITs. 3.3 Abnormal Occurrences Involving U.S. Nuclear Power Plants Each calendar quarter, AE0D prepares and coordinates a Report to Congress on Abnormal Occurrences (A0s). A0s are unscheduled incidents or events which the Commission determines are significant from the standpoint of public health or safety. A0s may be individual incidents, recurring events, generic concerns, or a series of incidents. The criteria for selection of A0s have not changed 13 ~

l I since publication in February 1977. Thus, the number of A0s per year can be viewed as a systematic and reasonably constant index to the perfonnance of the nuclear power industry with regard to potentially serious or significant occurrences. The number of A0s reported for nuclear power plants in the Reports to Congress for each calendar year since 1977 is shown in Figure 2. The history of A0s on ) a per plant basis is shown in Figure 3. The highest number of A0 reports involving U.S. nuclear plants since 1977 were included in -the 1984 reports. An increasing trend in the number of A0s is noted starting in 198?, with a slight drop noted for 1985. One reason for this increase is related-to the increasing number of nuclear power plants. The data in general are sparse and complex, however, making it premature to assess whether the slight drop noted in 1985 is indicative of a trend. i A significant fraction of A0s continues to be associated with plants licensed less than 2 years. For example, of the 38 A0s reported from 1981 through 1985, about 26% (ten events) occurred at plants licensed less than 2 years at the time of the A0. In 1984, 40% of the events (four out of ten A0s) occurred in plants licensed 2 years or less. In 1985, 33% of the events-(three out of nine A0s) occurred in plants licensed 2 years or less. j A summary of 1985 abnormal occurrences, including those still under con-sideration, is provided in Appendix A. This summary includes power reactor, nonreactor, and medical misadministration abnormal occurrences. These latter events are discussed in Sections 4.1 and 4.2. Conclusions In sunnary, despite substantial programs to learn from operating experience and major activities to improve plant operations and per.sonnel proficiency, a meaningful decrease in the number and rate of abnormal occurrences has not occurred. The number of significant items of this magnitude ' remains relatively constant. 3.4 Analysis of the 1985 Licensee Event Reports All Licensee Event Reports (LERs) are coded and stored in a computer readable, j searchable format, and a trend and pattern analysis is subsequently conducted of these reports. The objectives of this analysis are to: provide a quantitative indicator of licensee performance; identify patterns in reported data which may be of interest or concern due to the frequency of occurrence; examine the trend in reportable events over time; and distinguish outliers among the data or other anomalous conditions that would be good candidates for a detailed engineering evaluation. On January 1, 1984, the LER rule (10 CFR 50.73) became effective. For the first time, the NRC and the industry have a description by the licensee, in a reasonably complete and detailed manner, of: all actuations of Engineered Safety Features (ESF), including scrams; all losses of safety function at a system level; all significant systems interactions; all Technical Specification violations; and all significant internal and external threats to plant safety.. I AE00 estimated in 1983 that the number of LERs would probably decrease by 50%, 14

U. S. NUCLEAR POWER PLANTS ABNORMAL OCCURRENCES VS. YEAR 20 0 1G g,. O 12

Y M 10 7

8 I.Il.llll, g 77 78 79 80 81 82 83 84 85 Figure 2. U.S. Nuclear Power Plants Abnormal Occurrences vs. Year U. S. NUCLEAR POWER PLANTS ABNORMAL OCCURRENCES / PLANT VS. YEAR .20 .18 '1G y .14 t h .12 h .10 i .oe l

i.oc

/ / 1 o 77 78 79 SO 81 82 83 84 85 ) CALENDAR YEAR 67 70 70 70 75 81 84 92 97 NO. OF NUCLEAR POWER PLANTS Figure 3. U.S. Nuclear Power Plants Abnormal Occurrences / Plant' vs. Year 15

but that twice as much effort would be required to prepare the new LERs. Thus, no change was estimated in the licensee resources required to implement the LER rule. As indicated in Table 1, this estimate of the number of LERs has proven to be substantially correct. Table 1 LERs Submitted By Year Year LERs Units LERs Per Unit 1981-4016 75 53 1982 4400 81 54 1983 4839 84,,- 57 1984 2435 92 26 1985 2997 97 31

• Dresden 1, Humboldt Bay, and Three Mile Island 2 are not. included in 1985 data.

Palo Verde not included - licensed 12/31/84 The following discussion provides an overview of the LERs submitted in 1985. As noted on a number of occasions, we attach no safety significance to the raw number of LERs per se. The safety significance is assessed through engineering review of each LER, and through trends and patterns analysis of the content of the LERs. Variations in LER counts from plant to plant can result from a host of factors, only one of which is an actual difference in safety performance. Thus, we are examine the reporting pattern with attention to both high and low reporters, in order to gauge the reporting process itself. Analysis of the 2997 LERs submitted in 1985 indicates a broad range in the number of LERs submitted by each licensee (from three LERs to 102 LERs in this 1-year period). The mean number of LERs per unit.is 31, while the median is

25. The distribution of LERs per unit is shown in Figure 4.

In terms of what was reported the largest number. of reports- (45%) was asso-ciated with scrams and-ESF actuations. The number is roughly split between ESF actuations (other than scrams), and scrams. A scram was reported-in about-20% of the 1985 LERs, compared to only 1% of the 1981-1983 LERs (note: scrams were not specifically reportable prior to 1984). The second most frequently reported-type of event (29%) was a condition i prohibited by Technical Specifications (TS) or a shutdown required by the TS. This category covers conditions ranging from a missed surveillance test to the completion of a plant shutdown because of the unavailability of required safety components. e 16

LERs SUBMITTED IN 1985 16 15 - 7 14 - 13 - '/ 7 .,./ g e. l $ 5 35 10 - / y ; g 4 9- / .b B'- 'j 4 E / 7 / / 7_ '/ '/

• / /

j '/ j l/ 'j/ 2 6-w /j (/ 9 / / / 5- ~

r

, j / j j /l / / / p f 7-4~l / / b '/ h, h' 9 / h / 7 h h 3 /: / ) / J / / / / / /

- / / / / / / / / / / / / / i

'/ / / / ~ s l-5' !!215 A ' 311 5 ' 41'45 ' 51'-55'61k5 ' 7t 75 ' 81k5 91'-95 ' 10l'-105 3 d 28 25 - 6-10 16-20 26-30 36-40 46-50 .56-60 65-70 76-80 86-90 96-100 NUMBER OF LERs Figure 4. LERs Submitted in 1985 The third most frequent LER category concerned events that did or could have resulted in a loss of a safety function at the system level. This condition was reported in 8% of the LERs. The percent of LERs by'the individual LER report requirement is shown in Table 2. Table-2 LER Reporting by Report Requirement Refererce Requirement Percent

• 50.73(a)(2)(iv).

RPS/ESF Actuation - 45 50.73(a)(2)(i) TS Shutdown or TS Violation 29 50.73(a)(2)(v) Real or Potential Loss of a Safety System 8 50.73(a)(2)(ii) Unanalyzed Conditions 4 50.73(a)(2)(vii) Failures in Mu?tiple Systems 3 . 50.73(a)(2)(iii) External Threat 1 50.73(a)(2)(viii)(A) Airborne Radioactive Release 1 Other associated reporting requirements 11 '(e g.., Pa rt 21, - 50.36, 73. 71, voluntary)

• Percents sum to greater than 100% because an LER can be reported under more than one reporting requirement.

17

When the LERs are assessed in terms of other classifications, such as NSSS vendor or architect-engineer, typically a wide variation results. For example, the average number of LERs submitted by Babcock & Wilcox plants in 1985 is far below that of the other NSSS vendors. Specific details are provided in Table 3. Table 3 Average Number of LERs by Major NSSS Vendors in 1985 Number of Average Number of LERs NSSS Vendor Units Per Plant B&W 8 15 CE 14 29 W 40 29 GE 33 38 The analysis of 1985 LERs is continuing. A number of major studies are in progress or planned in order to better characterize and understand the nature and significance of these operational events. The preliminary results of some of these studies (e.g., scrams and ESF actuations) are reported in Section 3. 3.5 Assessment of Licensee Event Reports Quality The Licensee Event Report (LER) is one of the most important and widely analyzed documents in the nuclear industry. It is the principal means of identifying and analyzing safety problems and concerns which may not be recognized or properly understood as to potential significance. In order to evaluate the overall quality of the contents of the LERs as part of the its Systematic Assessment of Licensee Performance (SALP) assessments, a representative sample of a licensee's LERs is evaluated against the reporting requirements continued in 10 CFR 50.73. Each sample consists of: LERs with event dates during the SALP analysis period, except that LERs with event dates prior to January 1, 1984 (the effective date of 10 CFR L 50.73) are not included. 1 Of all of the LERs submitted during the period, 50% are selected randomly. 1 However, a minimum of ten LERs and a maximum of 30 LERs are selected. If g the licensee has submitted fewer than ten LERs for the period, all of the LERs for the period are assessed. The evaluation consists of a detailed review of each selected LER to determine how well the content of its text, abstract, and coded fields meet the require-ments of 10 CFR 50.73. The requirements in 10 CFR'50.73 are described in considerable detail in NUREG-1022, and Supplements 1 and 2 to NUREG-1022. The evaluation process for each LER is divided into two parts. The first part of the evaluation consists of documenting comments specific to the content and j presentation of each LER. The second part consists of assessing the text, i abstract, and coded fields of each LER. 18 1

The LER-specific comments serve two purposes: (1) they point out what the analysts considered to be the specific deficiencies or observations concerning the information pertaining to the event, and (2) they provide a basis for a count of general deficiencies for the overall sample of LERs that was reviewed. Likewise, the scores serve two purposes: (1) they serve to illustrate in numerical terms how the analysts perceived the content of the information that was presented, and (2) they provide a basis for the overall score determined for each LER. A separate report is prepared for each licensee which includes a detailed discussion of the strengths and weaknesses of each LER. AE0D asks each Region to forward these reports to the licensee to assist in the preparation of future LERs. In addition, a brief summary statement of the results of the assessment is presented for each licensee. A graphic presentation of the distribution of the overall grades for the licensees assessed to date is provided in Figure 5. AVERAGE SCORE COMPARISON 8 7 1 I l l C I g5 g e o4 5 cab3 5 g z 2 a l l l-ld E O

9. 5 9.0 8.5 8.0 7.5 7.0 6.5 6.0 GRADE Figure 5 19

i 1 Conclusion This data demonstrate that there is still a rather wide divergence in the quality of the LERs submitted by various licensees. Feedback through the SALP process should improve the quality of the LERs submitted by all licensees. 3.6 Reactor Scram Experience In 1984 and 1985 This section analyzes unplanned reactor trips (i.e., scrams) which occurred at U.S. light water power reactors in 1984 and 1985. Data on reactor scrams were extracted from LERs submitted by licensees in conformance with 10 CFR 50.73, which went into effect on JanJary 1,1984. The section includes the results from a completed study of 1984 scrcm data (AE0D/P504) and preliminary results for a similar study for 1985 experience. (See Appendix B for a summary of scram rates by plant.) There are generally three phases to a scenario or sequence of events involving a reactor scram. First, there is the generation of some off-normal plant state which results in operation of the reactor protection system (RPS) or the need for a manual scram. Second, there is the operation of the RPS and control rod drive system. Third, there is the plant and operator response to the scram. Each phase has safety significance. For example, the NRC has concluded that a reduction in the frequency of challenges to plant safety systems should be a prime goal of each licensee. In addition, the NRC believes that large Anticipated Transient Without Scram risk reductions can be achieved by reducing the frequency of transients which call for the RPS to operate. 3.6.1 1985 Scram Experience We define a reactor scram as an actuation of the RPS, whether automatic or manual, which requires control' rod motion. Plants were included in these statistics if they: (1) held a full power operating license, and (2) accum-ulated critical hours for some portion of the calendar year in question. Reactor years were calculated for portions of the calendar year where necessary, based on the date of initial criticality. In 1985 there were e total of 525 unplanned scrams at the 92 U.S. LWRs which were licensed to operate at above 5% power and which had accumulated some critical hours. The corresponding figures for 1984 were 492 scrams at 83 LWRs. Of the 525 scrams in 1985, a total of 61 (12%) were manual. These figures match those for 1984. In both 1984 and 1985 the scram rate for the industry was approximately one scram per thousand hours of critical operation. The 1985 industry average did not show much change from 1984. The comparative statistics for 1983 through 1985 are shown below. Scram Type Average Rate Per Plant Per Year 1983 1984 1985 Manual 0.9 U U Automatic 5.6 5.2 5.0 Total G D D From the NRR study, "1983 Reactor Trip Statistics, " date.1 August 1984. 20 i

3.6.2. Reactor Scram Frequency ~ Reactor scram rates for unplanned reactor scrams occurring in 1984 and 1985 are displayed in Table 4. While there was little change in the PW1 aggregate rate, significant shifts took place at the NSSS-specific level. Table 4 Reactor Scram Frequency 1984 1985 Scrams Per Scrams Per Scrams Per 1000 Critical Scrams Per 1000 Critical Reactor Year Hours Reactor Year Hours PWR W 7.1 1.22 6.5 0.98 CE 5.9 0.96 7.2 -1.19 B&W 3.0 0.44 4.7 0.84 Total 6.3 1.04 6.4 1.01 BWR GE 5.5 1.12 4.7 0.85 AC 7.0 0.94 9.0 1.16 The decreese in the Westinghouse (W) average rate from 1.22 to 0.98 scrams per 1000 critical hours is reflective of a broad-based decrease of the individual plant rates. Of the 35 Westinghouse-designed plar.ts with data for both years, 29 showed a decrease in scram rate, five showed an increase and one remained the same. Also, the four units with initial criticality in 1985 exhibited lower rates in 1985 than the two units initially critical in 1984 did in 1984. The two Westinghouse-designed units that achieved initial criticality in 1984 also showed a large decrease in rates from 1984 to 1985. The increase in the Combustion Engineering (CE) average was'largely due to two units with initial criticality in 1985--Palo Verde 1 and Waterford 3. Elimi-nating these plants from the average results in a decrease for CE plants from 0.96 to 0.76 scrams per 1000 critical hours. The Babcock & Wilcox (B&W) average is based on the smallest number of plants (seven in 1984, eight in 1985), and hence is most sensitive to individual plant behavior. Nonetheless, the increase in average rate from 1984 to 1985 reflects an increase in the rate for six of the seven plants operating in both years. Finally, the General Electric (GE) BWR average shows a sizable decr a from 1.12 to 0.85 scrams per 1000 critical hours. The decrease reflects a drop in rate for 18 of the 28 GE-designed plants which operated in both 1984 and 1985. This outweighed the increases for the other ten plants plus the impact of rela-tively high scram rates from three new plants (two of the plants had initial criticality in 1985; the remaining plant was initially critical on 12/22/84 and was not included with the 1984 data). 21

In both 1984 and 1985 the majority of reactor scrams occurred with the reactor power above 15%: 68% in 1984, and 74% in 1985. We believe this reflects both the minor contribution of startup problems to scram frequency and the very short time spent operating in the low power regime. In fact, 31% and 38% of j total scrams in 1984 and 1985, respectively, occurred while the plant was at 957 power or above. Because of the preponderance of scrams above 15% and the normal greater decay heat removal needs, our analysis is focused on this power regime. Figure 6 is a plot of each plant's reactor scram rate for scrams above 15% power. The axes are scrams per 1000 critical hours and number of critical hours for the plant in 1984 and 1985. One feature noticeable in Figure 6 is the general shift of the reactor population-to higher per plant critical hours in 1985; the average was 5551 critical hours in 1984 and 5878 critical hours in l 1985. For 1984, a scram frequency of 2.0 scrams (above 15% power) per 1000 cHtical hours was selected as a breakpoint for examining relatively poor performance. Ten plants exhibited rates at or above the cutoff, with a maximum rate of 5.7 scrams per 1000 critical hours. Five of these plants had initial criticality in 1984. Proceeding similarly for 1985, a total of eight plants met the criteria, and six of the eight had initial criticality in 1985. Although showing large decreases from 1984 to 1985, Calla'iay 1 (criticality 10/2/84) and Grand Gulf I are the only two plants that were above the cutoff in both 1984 and 1985. Lastly, we note that the naximum scram frequency for 1985 is 4.80 i scrams per 1000 critical hours, less than the maximum for 1984. Conclusions In summary, the scram frequency is slightly improved for 1985 compared to 1984 Plants showing a decreased scram rate outnumbered those showing an increase by approximately two to one. The number of plants exhibiting relatively poor performance (based on their scram rate above 15% power exceeding 2.0 scrams per 1000 critical hours) decreased, although reactor population increased. In addition, the maximum scram rate above 15% power decreased, and new plants showed less extreme behavior in the early months following initial criticality. 3.6.3 Initiating Systems (Above 15% Power) We examined each scram to determine the system containing the root cause of the scram. That system was designated as the " initiating system" if hardware be-longing to that system failed; or if operation, maintenance, or testing of that system led to the reactor trip. The results of this categorization are shown 4 in Table 5. 22

i REACTOR TRIP RATES VS CRITICAL' HOURS BY PLANT Potet eREATGt THAN SEN b 888D5> % g ++ P >++%+++ A + N

>g4 ++ +

g +k + *+ + + McGuire 2 g f + + ++* +>* j g +LaSalle 2 C 40ADg + Hatch 2+4 Salem 2 +wppss 2 A L g + Salem 1 .+ + + Susquehanna 2 , y: u + R 3 + Diablo + Grand +Callaway 1 s gk + Canyon 1 Gulf I M,................................................ -... - e i 2 3 4 s e PLANT TRIP RAW Pet less.S CRmCAL HR5 WAR 1984 i REACTOR TRIP RATES VS CRITICAL HOURS BY PIANT Pouen eneAun THAN 155 3 i,sese.; +t +% i t r Ta11away 1

g *g k+v% + +

A Y i csemel)$+ + ++ 4 4 5 + ++ %+ + Grand Gulf g } j' +$ + Wolf + Byron 1 + c 4ees. Cgeek A i ++ Catawba +Waterford i. ,+++ +Palo Verde 1 + no assei + +Diablo Canyon 2 + u + a 1 e g, M,. e e a s 4 5 e et. Ant vicy MATE pen sees.e catrTzcAL HRe uan sees l Figure 6 23

i Table 5 Initiating Systems Summary Power Greater Than 15% 1984 1985 Number of Percent Number of Percent Systems Scrams of Total Scrams of Total Feedwater 91 27 93 24 Turbine 49 15 45 12 Electrical Distribution 46 14 66 17 , Reactor Protection 32 9 53 14 . Condensate 21 6 15 4 . Main Generator 21 6 28 7 Main Steam 18 5 14 4 Control Rod Drive 10 3 29 8 Other Systems (<10 scrams) 49 15 42 10 Total 337 389 Power Conversion System In 1984, the feedwater system was the single system most responsible for scrams, and this continued to be the case in 1985. Moreover, the major balance-of-plant systems which collectively are sometimes referred to as the power conversion system (i.e., feedwater, turbine, condensate, main generator and main steam) accounted for 59% and 50% of all scrams above 15% power in 1984 and 1985, respectively. The somewhat disproportionate increases for the reactor protection system and control rod drive system (which includes the reactor scram circuit breakers) will be studied in more detail in the ongoing analysis of 1985 data. 3.6.4 Causes of Scrams (Above 15% Power) The LER description of each scram was reviewed to determine the general classi-fication of the root cause or causes as shown in Table 6. Above 15% power, hard-ware failures dominated in both years. Valve malfunctions represented the only large generic component class, accounting for 10% and 18% of hardware failures in 1984 and 1985, respectively. Main feedwater control valves and main steam isolation valves were major contributors. (Note: the boundary definition for a valve includes the valve body, the valve operator, and any attached or dedicated protective devices.) Hardware failures in power conversion system components accounted for 40% of all scrams above 15% power in 1984 and 32% in 1985. 24

Table 6 Cause Summary Power Greater Than 15% 1984 1985 Number of Percent Number of Percent Scrams-of Total Scrams of Total ~ Hardware Failure 204 60 220 57 Human Error 75 22 96 25 Procedure Deficiencies 12 4 20 5 Hardware Failure & Human Error 11 3 4 1 Environmental 10 3 9 1 Unknown 10 3 23 5 Manual Steam Generator Level Control 8 2 5 1 Cause Not Provided 5 1 1 1

System Design

2 1 11 3 Total 337 99 389 99 sDoes not add to 100% due to rounding Personnel related problems (i.e, human error, manual steam generator control problems, procedure deficiencies) accounted for 28% and 31% of reactor scrams above 15% power, making them a substantial but secondary cause. Unlicensed personnel were responsible for close to 10% of all scrams above 15% power, with unlicensed technicians involved in roughly one of every 12 scrams in 1984 and f 1985. The concern is frequently raised that maintenance, testing and calibration 1 at power pose significant risks. In our analysis we found that these three activities contributed to 30% of all scrams above 15% power for both 1984 and 1985. Thus, these activities constitute a substantial contribution to scram frequency. In 1984, the reactor protection system and the turbine system were nearly equal as sources of scrams for these activities. In 1985, the situation appears to have altered slightly; the RPS increased in share while the turbine system's proportion decreased. Electrical distribution system problems increased to become second to the RPS. 3.6.5 Scrams With Associated Failures Unplanned scrams where the recovery was complicated by additional equipment failures or personnel errors can be of concern because of the higher level of stress and demands placed upon the operating personnel and mitigating systems. We define " associated failures" as component failures or personnel errors that did not contribute directly to the cause of the scram, but are associated with 25^

post-scram recovery (e.g., nomally the failure was discovered or occurred when the component was actuated to mitigate the consequences of the scram). In-1984, about 20% of all scrams above 15% power included one or more assoc-iated failures. In total, 77 scrams indicated a total of 123 separate failures. About one-third of the 77 included multiple failures. t In 1985, 18%, or roughly the same percentage as 1984, of scrams above 15% power included associated failures. A total of 85 scrams included a total of 126 1. !~ separate failures. As in 1984, one-third involved multiple associated i failures. Nine of the 21 PWR multiple failure scrams above 15% power involved failure or degraded performance of the emergency (auxiliary) feedwater system.- I 3.6.6 Quantitative Safety Significance Measures The preceding discussion relies on several qualitative assumptions and general-izations concerning the safety significance of a reactor' scram. Specifically, that reactor scram frequency should be minimized; that scram significance is positively correlated with the reactor power level immediately preceding the i i scram; that additional failures and personnel errors following the scram add to the safety significance of the scram. This section describes an approach to obtaining a quantitative measure of the safety significance of an event .{ sequence (transient) involving a reactor scram. l The specialized probabilistic event tree technique called " Accident Sequence i Mcursor (ASF)" analysis was used to review the 1985 reactor scram sequences i as documented in LERs. Based on a reading of the LER, specific sequences were i chosen for more detailed analysis (i.e., supplementing the LER material with i FSAR information, system design drawings, etc.).if the scram sequence included >i; one or more of the following: Any failure to function of a system that should have functioned as a con-sequence of an off-normal event or accident; Any support system failures, including single component failures in cooling water systems, instrument air, and electric power systems; Any instrumentation failures that could impact the performance of core cooling systems or support systems; ~ Any instance where two or more failures occurred; l. Any event or operating condition that was not enveloped by or proceded differently from the plant design basis; or i' Any event that, based on the reviewers experience, could have resulted in or significantly affected a chain of events leading to potential severe i core damage. Twenty-seven scram sequences were selected ~for quantification based on this i' more detailed qualitative review. These sequences were based on LERs received 1 as of January 1986 and, thus, the Rancho Seco event of December 26,~1985 was not included in this analysis. 26

The objective of the quantification is to produce a measure of the conditional risk of core damage or core vulnerability given the observed sequence. To do this we overlay the observed sequence on event trees which provide a reasonable set of alternate paths which the sequence might have taken. An example of such an event tree is shown in Figure 7. The end state of a path through an event tree is labeled " core damage" if the path has been formally analyzed and comprises less than proper operation of a minimum set of components. The end state is labeled " core vulnerability" if core protection is believed to be provided but no analytic basis is generally available to directly support that conclusion. A basic " transient" tree such as Figure 7 was developed for each of a number of sub-classes of light water reactors in order to more accurately portray the general responses (paths) for that class. If the observed scram sequence involved loss of main feedwater, loss of offsite power or loss of coolant (LOCA), more specialized trees were used. Finally, further plant specificity was achieved by manipulating the conditional branch probabilities used.. For example, the likelihood of feedwater system operability following a BWR small LOCA (e.g., a safety relief valve spuriously opening) considered whether the feed pumps at the plant were steam or motor driven. As stated above, a conditional probability of core damage or core vulnerability was calculated for each of the 27 selected scrams. In setting the sequence branch probabilities, this calculation assumed that the failure probabilities for systems that failed during the event were equal to the likelihood of failing to recover from the observed state. Failure probabilities for systems observed degraded during the sequence were set equal to the conditional probability that the system would fail and not be recovered within a 20- to 30-minute period. The failure probability associated with observed successes and with systems unchallenged during an actual sequence were assumed equal to a failure probability based on either available system failure data or fault tree models consisting of typical (i.e., generic) train and common mode failure probabilities. The conditional probabilities of severe core damage for the 27 sequences range from 1.1E-2 [for a loss of feedwater (LOFW) and AFW failure] to 1.7E-6, with many events in the IE-6 to IE-4 range. Because of the uncertainties inherent in the calculations, it is not appropriate to rank events on an event-by-event basis. Instead, events can be grouped into conditional probability ranges to identify the more significant events.. On an order of magnitude basis, the following numbers of events were identified: 27

e e i SWF M Set W W WWr# 8MF# 33 g, a gpg gag game er are 8Hr er WF m Wes SWWE W M emb W M E E I I M Vma 8 M TER R 8 M 9E8 I 4 eiet empust 5 m serum 6 M =a I 7 M tarum I 8 g M tarust I 9 M carum K =E et I 80 M TRA il W= = E N N N 1 I la M TUL5 to e= Tule N to test vus I tg metgaruet to M teruW 9? M TE. I te M carugt 39 M tesem I se est usust M = = i v c .am m a n==== ::: to M anrust t t) I a=====t ti: a = = = = n

==n I N IWitafust n = = =. 3.

==.a t I si== a = nw. 1 l rwe cuss s.c.e.: me a tamaant (t) m for cLASE B Figure 7. Sample Event Tree 28

Conditional Probabilities of Severe Core Damage Events 1E-1 to IE-0 - No events IE-2 to IE-1 - LOFW and AFW Failure at Davis-Besse (1.1E-2)* IE-3 to IE-2 - MSIV Closure and Subsequent SDV Isolation Problems at Oyster Creek - Stuck Open Relief Valve and HPCI/RCIC Unavailability at Hatch 1 1E-4 to IE-3 - Effective LOOP and AFW System Unavailability at San Onofre 1* - AFW Pumps Failure on Demand at Trojan - Reactor Trip, Loss of Feedwater and AFW Train Failure at Davis-Besse ~ - LOFW and RCIC Trip at Hatch 2 - l'ss of Circulating Water and Non-Safety Service Wder due to Expansion Joint Failure at LaSalle 1 1E-5 to IE-4 - 7 events IE-6 to 1E-5 - 12 events The two events above which are identified by an asterisk were investigated by NRC Incident Investigation Teams (IITs). As noted previously, the Rancho Seco event on December 26, 1985 was not included in this analysis, but was investi-gated by an IIT. To provide perspective on the conditional probabilities for these events, and to isolate the safety significance of the scram function per se, the conditional probabilities of a spurious. scram (i.e., the plant. state does not require the scram) with no subsequent complications and a scram due to loss of main feedwater were calculated for each plant class. A " simple" reactor trip with no complications has an associated conditional probability estimated in the IE-7 to low IE-5 range. Losses of feedwater without further complications have associated core damage probabilities in the IE-6 to high IE-5 range. In most cases, the conditional probability associated with loss of i feedwater is not substantially greater than that associated with a spurious trip at the same plant,'primarily because the likelihood of loss of feedwater during a reactor trip is quite high. Based on these estimates, it can be concluded that from a core damage probability standpoint scram sequences with conditional probabilities below IE-4 are not substantially more significant than " simple" trips and losses of feedwater. 29

o Conclusions Our use of these quantitative techniques is still preliminary and exploratory. However, we are encouraged that the approach in some measure is corroborated by independent judgements of safety significance (i.e., the decision to dispatch Incident Investigation Teams for two of the quantitatively most significant events). Further, we feel the analytical approach will enhance our under-standing of the implications of scrams with associated failures. 3.7 Engineered Safety Features (ESF) Actuations AE0D continues to believe that safety systems should work reliably and properly when challenged, but should not be challenged frequently or unnecessarily. In order to gain an enderstanding regarding the need and frequency of the chal-1enges to safety systems, the Commission required, as part of the revised LER system, that actuation of an Engineered Safety Features (ESF) be reported to the NRC as an LER. This reporting requirement became effective on January 1, 1984. Prior to this date, the actuations of these systems were not directly reportable. ESF systems are designed to control and mitigate specific occurrences that might challenge the integrity of the reactor and/or adversely affect plant personnel or the general populace. Generally, these include systems designed to control reactor core reactivity, isolate and cool containment, supply emergency cooling to the reactor fuel, remove residual core heat, assure habitability of the control room under all conditions, control radioactivity releases to the environment, and provide a source of emerger.cy power. As part of the AF0D trends and patterns analysis program, ESF actuations are the subject of periodic detailed review. A study of ESF actuations for the first half of 1984 was published as AE00/P503 in August 1985. The analysis of the second half of 1984 has been completed and will be issued shortly. Data for 1985 are currently being compiled and analysis results will be issued in September 1986. A summary of the findings for the second half of 1984 and a discussion of the trends from the first half of 1984 through the first half of 1985 are presented below. 3.7.1 Frequency of ESF Actuations It was found that 601 actuations occurred during the last 6 months of 1984. Of the units eligible, 72 units reported at least one ESF actuation. On a percentage basis, 21% did not report an ESF actuation during the period, 46% reported one to three actuations, 9% reported four to six actuations and 25% reported more than six actuations. Sixteen units (18%) reported more than ten ESF actuations. The maximum number of actuations at any one unit was 70. Figure 8 shows the unit distribution of total ESF actuations for the second half of 1984. Only about 7% of all reported ESF actuations involved an emergency core cooling system (ECCS), and none of these occurrences were necessary to control an actual loss-of-coolant accident (LOCA). Over 70% of the actuations that occurred in ESF systems were associated with either an isolation function or a ventilation function. 30

U4" DIS"RIBU"0N 0 ENG HEELED i SAFETY EA"URES ACTUATIONS l (July-December ' 984) 1 y gg n y yt, y, << 4 ga g l M g g j ,'(<g j ) NBC L< B j g g i, <m l g 4 >. A Z i i< menn i i e mama e i un== E i E i 0 O !< l i Z l< ""' Z i mua i ,tf i

ed, l

mna L' '. << c p'<mna C l tz m a 3 I 3 i '<i mza a kouns ma i<r is ,<ma i ma Willstone 2 -i 'i< ma Mne Wile Pt. <i k Solem b.<l I rza I ga Q l un bsze i i a l e i 6 e e i e i 5 6 6 6 Number of ESF Actuations Number of ESF Actuotions l Figure 8 l l

^ 3.7.2 Valid ESF Actuations The 601 ESF actuations were found to be the result of several causes. In 259 cases (43%), a measured parameter reached the intended setpoint for ESF actuation. However, only 22 of these occurrences (less than 4% of all ESF 3 actuations) were considered to be valid responses in terms of providing a y l required ESF response to mitigate an actual design basis event. Of these 22 design-basis actuations, nine cases involved loss of offsite power, four involved high radiation events, and five cases involved water level. The remaining 237 cases where a setpoint was reached were considered to be valid, but did not represent a needed response to a design basis event. These actuations resulted from non-design basis conditions such as radioactive trash being moved near a radiation monitor. These valid non-design basis events were primarily the result of water level, loss of power, flow, or toxic gas signals, and resulted primarily in isolation or ventilation actuations such as reactor i water cleanup (RWCU) or containment isolations, or control room ventilation actuations. The distribution of the ESF system functions involved and the measured parameters (signals) for valid ESF actuations are shown in Figure 9. 3.7.3 ECCS Actuations In 45 of the ESF actuations at 33 units, an emergency core cooling system (ECCS) was actuated. Fluid was actually injected into the reactor coolant system in 22 of the 45 events. Twenty units experienced these safety injection events, with no unit experiencing more than two injections. Of the 22 actual injection events, none were needed to control an actual LOCA, and only one was considered a valid design-basis actuation. Twelve of the events resulted in setpoints being reached, about half of them as a result of low water level following reactor scrams. The remaining nine safety injection events were false actuations, primarily the result of personnel errors during maintenance or testing. 3.7.4 False ESF Actuations About 57% (342 out of 601) of the ESF actuations were deemed to be invalid and unnecessary false actuations. These false ESF actuations were caused mostly by spurious signals, equipment failures, or problems related to personnel. Most of the actuations involved isolation or ventilation. 3.7.5 ESF Failures In only three of the 601 ESF actuations did an ESF system fail to actuate properly. In 45 additional cases the ESF actuations performed properly, but one or more failures were associated with the actuation. There was no identifiable trend or pattern for these failures, and redundant systems were available to perform required safety functions. A few of the problems discovered may be significant at other units or under other circumstances. These cases involved: (1) loss of the safety-grade auxiliary feedwater system; (2) trips of electrical heaters in the standby gas treatment system; (3) loss of all offsite and onsite electrical power; and (4) long standing design errors. 32 n- - - + -. y r9 9 v

I Figure 9 Measured Parameters and Associated System Functions for VALID ESF ACTUATIONS General Syst?m Function Involved Measured Heating and Emergency Not Total Parameter Fluid Ventilation Isolation Power Defined Functions (HVAC) Involved Radiation 0 1 18 0 0 19 Loss of Power 2 2 8 39. 0 51 Temperature 0 1 19 0 0 20 Pressure 5 10 5 1 5 26 Level 28 4 22 1 .0 55 Toxic Gas 0 36 0 0 0 36 Flow 0 0 43 0 0 43 Fire Detection 0 5 0 0 0 5 Other 15 2 23 1 3 44 Totals-50 61 138 42 8 299

• Examples: Fluid = High Pressure Coolant Injection (HPCI)

HVAC = Standby Gas Treatment System (SBGT) Isolation = Primary Containment Isolation System (PCIS) Emergency Power = Diesel Generator (DG) 1 d 33

'l 3.7.6 Conclusions Based Upon 1984 Data Analyses of the 1984 data supports the following general conclusions: Ten units were identified as experiencing repeated unresolved ESF actuations which could ultimately challenge continued equipment operability and proper personnel response. These units were: Byron 1, Cook 2, Ft. Calhoun, LaSalle 1 and 2, San Onofre 2 and 3, Sequoyah 1 and 2, and WPPSS 2. The remaining events necessitating ESF actuations, including ECCS actuations, have not been individually significant. This also holds true for failures and problems associated with the ESF actuations. In addition, it was readily apparent that the majority of the ESF actuations were unnecessary and that the rate of these actuations could be greatly decreased by (a). reducing the number of equipment failures during normal operation, (b) reducing the number of personnel errors during maintenance and testing, and (c) revising actuation setpoints to more appropriate protective levels. ESF functions associated with isolation or ventilation shnuld receive first priority in these regards. 3.7.7 Use of ESF Actuations as Performance Indicators It must be recognized that not all plants have the same number and type of ESF systems; thus, there are a number of variables in the reporting of these actuations. Further, as noted above, ESF actuations differ in significance, and the events which are inherently and immediately significant are rare. This point is further demonstrated in Table 7 which indicates that for all of 1984 less than 4% of all actuations involved a valid response to a design basis situation. Also, actual safety injections have been rare. Using a subset of significant ESF actuations as a performance measure is not practical, however, since there are so few that little discrimination in performance would result either among plants or over time for a given plant. Thus, the total number of ESF actuations are being tracked with attention focused on plants with ten or more actuations in a 6-month period. Sustained operation with that rate of i essentially undesired ESF actuations may be indicative of a willingness to accept safety features that are not performing as intended by the designer. 3.7.8 Comparison With 1985 Data Data for 1985 is currently being analyzed in detail, and complete resNlts are not available at this time. However, some preliminary trends can be described using the data for the 'first half of 1985'in Table 7. From early 1984 to early 1965 we can see an increase in'the average number of ESF actuations per reporting plant, and an increase in the number of plants reporting at least one actuation. At the same time we see a decrease in the maximum number of actuations at a single plant, and a general shift in the reactor population toward higher reporting rates. If these raw figures hold up after detailed analysis, we could see generally higher actuation rates for 1985. 34 _i

o Table 7 ESF Actuations for 1984 and 1985 Category 1984 1984 1985 (Jan.-June) (July-Dec.) (Jan.-June) Total ESF Actuations 501 601 790 Units Eligible to Report 87 91 94 Units Reporting 0 ESF Actuations 61 72 81 Units Reporting (%): 0 30% 21% 155 1-3 42% 46% 36% 4-6 9% 9% 15% >6 18% 25% 35% >10 14% 18% 27% Maximum ESFs Actuations at Single Unit 82 70 50 % of Actuations ECCS 10% 7.5% N/A No. of Actual Injections 23 22 N/A No. of Injections for LOCA 0 0 N/A % of Isolation / Ventilation 70% 70% N/A No. Setpoint Reached 143 (28%) 259 43%) N/A Design Basis 12(2%) 22 4%) N/A Non Design Basis 131(26%) 237 39%) N/A Invalid 358 (71%) 342(57%) N/A i Estimated from SCSS. N/A: Not available at this time. 35

O O 3.8 Studies of Loss or Unavailability of Safety System Function During 1985, AEOD conducted two studies that focused on the loss or unavail-ability of systems designed to perform a safety function. The results of the first study, the Loss of Safety System Function (LSSF) Events Study, was published as.AE0D/C504 in December 1985. (The System Unavailability Study will be issued in 1986.) 3.8.1 Conclusions from the 1985 LSSF Study The LSSF Study considered events that resulted in a total loss of safety system function during the time period January 1981 to June 1984. Although the study identified 133 losses of safety system function, the major focus of the analysis was on 87 events (65% of the total) that resulted from human factor i contributions (e.g., personnel errors). The objectives of the LSSF Study were to determine the frequency of this type of event, whether or not these events were occurring more at some plants than at others, and the causes of such events. The study concluded that: (1) Between 1981 and June 1984, no significant trends were observed in the rates of occurrence of LSSF events. However, no large improvement was evident on an industry-wide basis in preventing these events. i (2) Variation in the rates of occurrence of LSSF events was largely the result of variations among a small number of individual plants rather than differences in NSSS type or plant age (except for the first year). (3) With a few exceptions, most events were non-repetitive. (4) About one-third of the PWR events involved either the RHR systems (shutdown cooling mode) (31) or the cnntainment spray system (11). The i BWR events most frequently involved the HPCI (22), RCIC (eight), or RHR (eight) systems. (5) Of the 133 events examined in the LSSF Study, the most frequent events-(87) were the result of human factor contributions; i.e., they were not solely attributable to equipment failure. ~ (6) Only 32 of the 87 events involving human factors were the result of " personnel error"; the remaining 55 were due to factors such as defective management and administrative controls. This indicated that, although training, qualifications, and motivation are important, improvements i were also needed in the areas of management and administrative control, I procedures and planning. 3.8.2 System Unavailability Study The second AE0D study, the System Unavailability Study, is being performed as part of AE0D's Trends and Patterns Program.- This study reviews and evaluates LERs submitted under 10 CFR 50.73 (a)(2)(v), whose intent is to capture those events in which there was, or would have been, a failure of a system to properly complete a safety function, regardless of when the failures were discovered or whether the system was needed at the time. The AE0D System (fnavailability 36

e 1 Study is motivated by the concern that frequent or prolonged unavailability of systems designed to perform safety functions may indicate safety problems which warrant further attention. The purpose of the System Unavailability Study is to: (1) gain an understanding of the frequency and duration of safety system unavailability on both an individual unit basis and on an industry-wide basis, (2) determine the significance and implications of current system unavail-abilities, (3) determine whether specifications on the part of the NRC or by industry appear warranted, and (4) investigate the usefulness of information regarding safety system unavailability as a measure of plant performance. The System Unavailability Study is an extension of the LSSF Study which focused on the 1981-1983 time period. AE00 developed a data base containing pertinent information from LERs on events that occurred during 1984 involving the unavailability of one or more systems that were designed to perform a safety function. AE0D has analyzed the data and formulated a number of conclusions and recommendations for further action. Preparation of the report documenting this study is in progress. Some preliminary analysis of 1985 data has been conducted using LER data on system unavailability for January 1 through j July 31, 1985. These results permit some limited comparisons between the 1984 and 1985 events. l 3.8.3 System Unavailability by Plant l This study covered all plants which had reached criticality by December 31, 1984, except for three plants considered atypical because of size or type (gas-cooled). Only LER's submitted after criticality were assessed. There is a wide range in the number of events reported by individual plants. Figure 10 shows the distribution of the number of system unavailability events submitted by individual plants for 1984 and the first 7 months of 1985, respectively. Of the 82 plants included in the study, 74 submitted a total of 387 LERs which reported one or more occurrences of system unavailability. From these LERs it was determined that there was a total of 346 events in which systems reportable under 10 CFR 50.73 (a)(2)(v) were unavailable. This is equivalent to an average of 0.4 systems unavailability events per month per plant. The number of reported events ranged from 1 (for 5 plants) to 23 (for 1 plant). Of the 90 plants assessed during the first 7 months of 1985, 60 plants submitted a total of 195 LERs which reported one or more occurrences of system unavailability. From these LERs, it was determined that there was a total of 185 events in which systems reportable under 10 CFR 50.73(a)(2)(v) were unavailable. This is an average of 0.3 system unavailability events per month per plant. For this period, the number of reported system unavailability events ranged from 1 (in the case of 12 plants) to 16 (for one plant). Thus, we anticipate that the overall number of system unavailability events for all of 1985 may decrease slightly from the 1984 figure. The study also takes into account the " severity" and the " type" of unavail-ability. The severity is classified as either complete failure ("C"), in which the system is completely unable to perform its function, or degraded ("d"), in which the system operates at less than its specified performance level. The type of unavailability has 11 categories, but these can be assigned to two major categories: unavailable on actual demand (" Actual"), and unavailable on test demand or a period when no demand 37 .o.

r e o i SYSTEM UNAVAILABILITY 1864 IBRs 40 SS - 30 - i g 16 - E 10 6-0 77 777 0 1 3 3 4 5 6'7 8 8 14 15 18 30 33 No. of Srptsza Uneva11stility Ennis s SYSTEM UNAVAILABILITY 1/1-7/J1/96 LERs i' 3e - 30 - l f 36 - an. A f 16 - 10 - ^ ??? O 1 2 3 4 6 6 7 9 to 16 No. of System Unavailability Ennis Figure 10' 38

I e occurred ("Other"). Figure 11 provides the plant by plant statistics for 1984 by showing the type and severity distinctions. For example, Duane Arnold experienced nine events, but only one (a total failure) was an actual demand. In contrast, Dresden Unit 3 had only four events, but.three of these were total failures on actual demand. The type and severity distinction can thus alter the assessment of plant performance in the area of system unavail-ability. The duration of unavailability and the importance of the system are additional considerations. 3.8.4 System Unavailability by NSSS Vendor Table 8 shows the distribution of system unavailability events accordiWg to reactor vendor for 1984 and for the first 7 months of 1985. Two observations from this table are noteworthy: (1) GE BWRs reported about two-thirds of the events in both periods, although they represented only one-third of the plants. This is due primarily to the use of single train safety systems (e.g., HPCI) in BWRs. B&W plants, which repre-sented 9% and 8% of the total number of plants, accounted for only 2% and 3% of the events for the two periods, respectively. (2) There was a significant reduction in the frequency of events reported from CE PWRs during the first 7 months of 1985 compared to 1934 (from an average of 0.3 event per month per plant in 1984 to less than 0.1 per month per plant for the 1985 period). Table 8 DISTRIBUTION OF SYSTEM UN/VAILABILITY EVENTS ACCORDING TO NSSS VENDOR f 1984 LER Data 1/1/-7/1/85 LER Data NSSS No of % of No. of % of No. of % of No. of % of Vendor Plants Total Events Total of Plants Total Events Total B&W 7 9% 8 2% 7 8% 6 3% CE 11 13% 41 12% 13 14% 8 4% GE 29 35% 217 63% 32 36% 124 67% W 35 43% 80 23% 38 42% 47 25% Total 82 100% 346 100% 90 100% 185 100% l 39

0 O f.t O emaeusILift swa 3 -san 43 tot O==amammig any arent

enes, areest seest

.rteel est aus 3 3 13 1. $ 0 $ IO 89 3 3 WN.IE 3 N 18 9. 8 9 9 le 65 3 3 Ett ERIE R 3 el s. $ 0 3 et 33 p n t,,I_8_f_f_f_f_f_1_I_t l_I_f_f_0 l_t_i_t_i_I t,I_t_I_f_I_I_ _1_i_1 8 e I te.ese 3 .I 8 it. ass hut I C4 fragete tales.1 1 6 I I I I 4 .rtssees 3 i El asase I i WlElt freers leise. 3 8 8 8 I.I i .I I t tueser testav 4 0 t Guest 8 i he fttese I i C:Et .CIE 4 I 8 I e i

h. tu..ne.

I (I le.l. 8us. t I tit Inc. Otsee t I .tsE t i I .4 8 W~ Desefee,e e Em:E tm.i. pas. 5 I It tenses tsee i .ta e muuss: I .I i st I-Damefee,8 I time.es 8 18k St. Late i k .Ett et I uss: 4 I i mi Damefeep3 4 M.ns:S Le tulle i O EEtlEEtt! 8t. Late 3 4 8 8 .:a t t D.eusse 3 I ECS

4. hele i 0

.i E.,st CWit.tttitle talse B i i .i ..e.es., i .i .iR E.;-E .w. to w.i .i i.. .-i .i E.m re i 44 Eat. ort Decee t I MK: asse sessee .i 1 I. len heere t tt i i. tesesrt Otose I t IE.E 4 til e l 1 18 Iso beste 3 I (2 I s: O t t .1 tees i E.;E t.a ha.see 3 8 legayah I i 18 i .s.ie .i .i.t (4 i i. i

8. 6, te. t 1

E.s. .I see ele penet I ttt timmer I i .tatt i i L L tese 3 I R$ EIrts.ine 9 8 til twey 3 i i 1 I 8 I t swee e. s e et: are as t E. E tre, 3 i ett: i i i Ostte tenvas i I I tense I i I tungstemme I i C: .ms:.E e a 4 4 1 ..seen i i = ..es i E.4 1 1 1 8 8 .seems 8 I t C!! kenes 8 n i 1rese i 18 t I t i I sees anese keel. 8 CItttt:t trata treet I I Eltttt twee, peset 8 i ttt i i i .s.i i to es.. i

e. ie,.

84 .es e i. 1 1 I I t 8 f ates ! I (4 Peett tutt. I i C:t suresst tesese i E: I l I a I .sse: I fetapeernes l ECM feet tuetse 5 i EE mi .i i i fyt Calhaus 4 ES Pityle t 1 Etttt f.eep time i 15 i = i i i .i

e. -. i i

i i C= i i Et i = si. : i i i I i I I I I t u .e., i i i.. a=a n== = = n 9. t e m e. ee sos.n. e e G

e.., e te aN nm 9 4'

I w as a M LEGEND: Severity C = Complete failure Severity d = Degraded Figure 11 40

o o 3.8.5 System Unavailability by System Figures 12 and 13 show the distribution of system unavailability events according to the system for 1984 and for the first 7 months of 1985, respec-tively. Figures 14 and 15 show the distributions for BWR and PWR systems over the same time periods. The systems most frequently reported to be unavailable were generally BWR systems. Of the ten systems with the highest frequency of occurrence of unavailability for both periods, at least six were BWR systems. PWR Reactor Coolant System (RCS) unavailability events actually are comprised of two basic situations: (1) the RCS leaks, forcing a plant shutdown; and (2) the control feature which trips RCS pumps in certain situations was unavailable. The dominance of BWR systems can probably be attributed to system design. In BWRs, the HPCI and RCIC are single train, turbine-driven systems. Thus, they are vulnerable to a single failure and to the problems associated with turbine-driven equipment. BWR systems were reported unavailable during 1984 at a rate of 0.6 event per month per plant. On the other hand, PWR systems were reported unavailable at a rate of 0.1 event per month per plant. The corresponding system unavailability frequencies for the first seven months of 1985 were 0.6 and 0.1 event per month per plant, respectively. Less than half of the system unavailability events had sufficient information provided in the LER to allow estimates of the duration of unavailability, although this information is required by the LER rule. For these cases, the estimates of the duration ranged from less than I hour to more than 1000 hours. The three systems which had the highest total unavailabilities (summed over all plants) were PWR systems (PWR emergency power, PWR auxiliary / emergency feedwater, and PWR LPSI/RHR/SI accumulator). In general, it was found that the longest total unavailability durations were due to a relatively small number of events. The average duration for BWR system unavailability events was estimated to be 34 hours, approximately with a median of 136 hours. The average PWR system unavailability duration was estimated to be about 184 hours, with a median of 24 hours. In general, the events which had the longest associated duration of system unavailability were of a type in which the system became unavailable under conditions other than an actual challenge or during normal operations (e.g.,testingandmaintenance). Over 90% of the system unavailability events that had an associated duration of unevailability of more than 100 hours, were of the potential type or a type other than a real failure of the system during an actual operational challenge. The dominant root cause category for the 1984 system unavailability data was " unclassifiable." This supports a finding stated in NUREG-1022, Sup-plement 2, that a significant deficiency of current LER texts is the failure of licensees to include failure modes, dates and times, and the mechanism and/or effect(s) of the failed component (s). l 41

e e i I 394 la MIR I I 31551 I le N at te I _t I I i i i I BR ESC n. E I g ag ugtpgg gnuo I SE 15Cl inverrrrrmettenpurrrrn I i IRA UEl/BA/BI hem. 8""'+'"" 8 I BR Essen. Swvies lister 8 ' + + + + + I I sig 33 Inermmer I i R R 1 5 h ateff irrrre m erg i I fim (M5 lerrmm i I IIR Energsm y eer Ir m m eur i p I RA15t3 larmrme i Ipig g'W Cool. Imor. I ' * * * * ' ' ' ' I I ga guyssuv pour I** 8 I fta plast A.C. puser IERIMI I I Stanty Ass Truste nt trarrrrrrr I i flE met. Osrv. theer IBMRII I IRA heet. Elds. Esw.thiellIREDI I I IIA NWWW IIIIE R I I Fire pretsetion ehtsri IIIER - 1 ISA Caskts.Itr!. til.Typel ".." 1 I as rmaw W I.' I I SPf tite ihmer 1'" 8 I AE Ut3 IIIIB 1 1 IIR ISSI IEEE I I BR 3.C. pour IIEE I ISR herter Ascirentallen IRII I asla Am. NdB. Esw. Derl.IIBI I I as asseter pretsetion im i I thdain. Esw. Iludtering IIB I I Essomttal M r IEI I I BE 25 IIE I I SE Bryuel! Esw.thlel. IBI I I flE Roseter protection IBI I I fle IF Artestien Im i I hat diridad itsnitoring IEI I I flR hamle tuntilatten im i SAR Cedain.Aless.Castise im I lisesre d ssure lism. Retrig.I H I I ple Ced alasmut sprey III I ICantaisemat Iashage thtr!.II I I Cudaisant Use. blief II I I Inst isolation Oder. II I I BE Plant II.C. Dietrib. II I I Esq. Sfty. Fest. Wentil. II I I huusciater II I I BR Itset. Itsnitoring II I I hdiationItsnitoring II I I Servies Air II I I fit Centrol And tri w II I I linin Ter6ine II I I Tetim Late Oil II I I lisin Tertine Catrl.Fleid II I I hd.tiste. Bid.Esw.l>tri. II I l ISR Res. Bids. Clod.Cig.utr.it i ISR Enl. Safety Fes.lletes.II I ITu ti m S'.ees Bypass O drl!! I I I I -l i I I e le as .as to aseet F srsal-nsali.TT mens Figure 12. 1984 Data 42

e I i 1/1-7/31/ 5 LER BATA I I MTBI I le 80 30 40 I I I I I I I M ISCI turrrrrrrrrrtugutnunnuun i I M EIC surrumurggggggggung i M L K !/ M lufttrrrrrrtug u g n H I I I M Emergsucy pe er lyrrrrrrrrt I 1 M E3 Imtmr i i Stammy Ess Tresteest IIIRIIH I 1 1 M LfSI/H/S! knes. IMIH I ) 1 M Eergency eer IRInt 8 p 8 M 1981 IIIHH I I FL:a protection mister) tutu I IM Asset. Olds. Ear.Catr!!MH I 8 M IWMFil Im! I I Off46te poner imI I IM Campenent Cas!. liter. Im! l I Instrusut Air topply IIIH I i ! Centain. Env. Itsnitoring I H H I leisst AsJact.Circulat. Iller.imI I I am uts im 1 IM Asseter Aseirculation IIH I I Cable Amesusy IHI I I M Ift3 poner lu I I M l'lantrol And Bri w IH I I M Contalasunt Spray in I IComb.Ess Strl.-escos.TyplII I 169 litr. 4 M Breiss/Uunts! H I IFuel pool Cool./perifica, in i I M milled lister 11 1 I M E.CS II I ! M Essen. Service lister II I I M plant A.C. Distrib. 11 I I M D.C. pour II I I Asseter peer Centrol II I M DCS II I I S.E. Blasism II I ! M Bryuell Ew.Cutr1. II I IFire protection tasmicallit I liertine Osporvisory Cntrl.II I IM Cantain.Lankage Catri.l! I I I I i I I 0 10 20 30 40 IUSER OF SYSTEM t#1MILABILITY EVBffS i i Figure 13. January-July 1985 Data i 43

o BWR SYSTEM UNAVAILABILITY 3904 LBR Deta-lo Eighest { 3-5A EIC m E'" 5 u-i E 7s Y.

• i

=::'9r-

• • • E %t'~t',i.

l m m.ctri.-m i.1,,s g to - i f e-A A 'A A A i i i i i i i i i i s'o s,.t BWR SYSTEM UNAVAILABIllTY 1/2-7/31/96 IERs-lo Eithest inee { f i-as sci ac - h a-Er t a p//, '//>. =ma.t .,i. r/ yi tmi...,, m-f e c atain. s=. amiterig / / t Intraint ai,swely 5 3e _ ,/ s ac - b. b 5k i i i i i i i i i lc arstem i Figure 14 44

e PWR SYSTEM UNAVAILABILITY 19841.ER Date-1C Eichest 3C y M IR y 1-ftA 1751/WA/SI Asem g H. - ,,.//' / HIA 043 >s 16 ,f g g I' -/ /] Him W Cael. Inor y /'/j Hun sesryner puusr 33 - a

y. gig gagg, g,,,, g,ger l

!! -{f,.. g,,j/) j p 11~,/// 0 fin NIVEFW eft, f" 9Onn80g1 LC -// i M,d ;9 d p W P $r / k h h k h h 7 h h [C System a PWR SYSTEM UNAVAILABILITY 1/1-7/31/15 IZRs-10 Hizhest 16 m f 1 f If 1-# Wit Eserisacy oner p c is - He acs 14 Hult LPS!/98/S! kcm i 13 1 HWit 1951 HWR Compoet Cool. IAer. 3

2 d HWil WW/EFW D

1C L 7-Heat Arject.Circulat.lAer. p 9 f/,i H Wit Contairment Seray /, f Mff-Gite peer .i'f 1H:enh.Sas Catr1.-tecemb.Typ f/ $.A,/;1 {,/ F41 p* 7z/ 'h 5l / ) V 3 / C - *l ,' I h l / i 3 {'.'{. Y' f*'f/ / [/,**/: / / / //

-V,/A * * /A,

e r-r s f /) / {// I

• s ',

j s i a 3 (r 6 C ? 6 9 10 Bystem Figure 15 45

4. COMMENTS AND OBSERVATIONS ON 1985 OPERATING EXPERIENCE AT OTHER LICENSEES During 1985, a number of events involving NRC and Agreement State licensed nonreactor facilities and activities were reported to the NRC. This section provides an overview and summary of reported events involving these non-reactor facilities, and medical misadministrations occurring in 1985. ] 4.1 Nonreactor Events The AE0D Nonreactor Event Reports (NRER) data file contains information on licensed nuclear materials and fuel cycle operational events, and on personnel radiation exposure events. The NRER data file management system provides for input, storage, retrieval, and computer-assisted analyses of operational event data and is used in identifying trends in operational safety events which may signal a need for remedial actions by the NRC and/or licensees. 4.1.1 Occurrences in 1985 The NRER data file includes 170 records of events that occurred during 1985. Infomation on these events was contained in reports submitted by nonreactor licensees to the Regional Offices or in other documents, primarily inspection reports. The data file does not include information from certain fuel cycle licensee reports, such as those related to routine effluent releases, nor does it include information from reports of medical misadministrations (see Section 4.2). Table 9 provides information on the types of licensees for which information was entered into the database. ~ An NRER database item may be associated with more than one category of event. For example, a report from a radiography licensee concerning a personnel radia-tion exposure would be counted in the total number of radiation exposure events as well as the total number of events involving radiography. The 170 non-reactor licensee reports were cataloged as 319 events in nine different areas. The details are included in Table 10. Note that, because some reports are associated with more than one event category the total number of events exceeds the total number of reports. The 170 nonreactor licensee reports were most frequently concerned with: radiation exposures (34); lost, abandoned, and stolen material (54 times); leaking sources (29); and medical and tele-therapy (66). Categories in Table 10 labeled " primary" contain events from all types of licensees. These primary categories are: exposure; lost, abandoned, or stolen material; leaking sources; release of material; and actual or potential contamination of consumer products. (With a few exceptions, most events are assigned to only one of the categories.) Categories in Table 10 labeled " secondary" are categories designed to capture events by the type of licensee involved in the event. Many of the events assigned to these categories are also assigned to primary event categories. Secondary categories generally serve as a measure of the frequency with which certain types of licensees make reports to NRC. 46

Table 9 Types of Licensees That Submitted Reports During 1985 Number of License Type Reports Received Academic 1 Medical 50 Comercial/ Industrial Measuring Systems 38 Well Logging (23) Other Measuring Systems (15) Manufacturing and Distribution 17 Industrial Radiography 17 Single Location (In Plant) (2) Multiple Locations (Field) (15) Irradiator~ 3 R&D 9 Source Materials 1 Mills (1) UF6 (0) Other (0) Special Nuclear Material (Including Polonium) 5 Agreement State 16 Other*** 13 Total 170

• Medical misadministration reports are not included.
  • Routine environmental effluent release reports, e.g., reports required by 40.65 and 70.59 were not included in the totals for source and special nuclear materials licensees.
    • Number includes reports received for which no program code was available.

/ 47 g. ,e-

o Table 10 Categorization of Reports Area With Which Report Was Associated Number Of Reports

• Primary Categories:

Personnel Radiation Exposures 34 Lost, Abandoned, and Stolen Material 54 Leaking Sources 29 Release of Material 8 Consumer Products 5 Secondary Categories: Fuel Cycle 6 Radiography 17 '1 Manufacturing and Distribution 36 Gauges / Measuring Systems 15 Other 115 Total 319

• An NRER ' data file item may be associated with more than one type of event.

For example, a report from a radiography licensee concerning a personnel radiation exposure would be counted in the total number of radiation exposure events as well as in the total number of events involving radiography. .g 48

4.1.2 Radiation Exposure Events In 1985, there were 34 reports covering the potential for, or an actual radiation overexposure. Of these 34 events, 21 involved an actual radiation overexposure. The types of licensees associated with actual over-exposures reported during 1985 were as follows: Number of Total Number of Licensee Type Overexposure Events Individuals Exposed Medical / Academic 7 7 Radiography 9 12 Commercial / Industrial 5 14 Total 21 33 Overexposures at Medical or Academic Licensees - The seven reported events at medical and academic licensees involved licensee personnel, and included: three whole body exposure reports, with exposures of 1.75, 2.06, and 3.9 rem / quarter; three skin exposures of 8, 23.73, and 32 rem / quarter; and an extremity exposure of 20.29 rem / quarter. All of these reports came from different licensees. Overexposures at Radiography Licensees - The nine overexposure events at radiography licensees included one extremity overexposure and eight whole body overexposures. Four of the nine radiography events resulted in technical overexposures. In the remaining five, the overexposures ranged from moderate (5-25 rem whole body; 375 rem extremity) to major (25 rem whole body). Although all of the radiography events were reported by different licensees, eight of the nine events, including the events in which there were large overexposures, occurred during radiography at remote sites (i.e., not fixed sites). Overexposures at Commercial / Industrial Licensees - The five overexposures at commercial or industrial licensees consisted of: two extremity exposures (19.1 and 19.41 rem / quarter); an overexposure of 1.76 rem whole body; and an event resulting in a skin exposure (thumb) of 93 rem. The fifth represented a potential or actual exposure at a general licensee. In general, with exception of the overexposures at radiography licensees, most of the overexposures reported during 1985 were minor. Of the nine reported overexposures of radiographers, three resulted in whole body overexposures to individuals-in excess of 20 rem, with one event characterized as resulting in serious injury from radiation. 49

} 4.1.3 Lost, Abandoned, and Stolen Material Licensees are required to report the loss or theft of licensed material that has occurred in such quantities and under such circumstances that the licensee believes a substantial hazard may result to persons in unrestricted areas. Fifty-four events occurred during 1985 that involved lost, abandoned, or stolen licensed material. These events consisted of 32 reports of lost or stolen material, and 22 reports of irretrievable well-logging sources. None of the 54 events resulted in a known rcdiation overexposure.* 4.1.4 Leaking or Contaminated Sources Certain licensees are required to leak test sources and to report leaking sources. Twenty-nine events of leaking or contaminated sources occurred during 1985. None of these were associated with a radiation overexposure. Americium, cesium, iodine, and nickel were the isotopic sources found to be leaking or contaminated. About half of the 29 leaking or contaminated source events were reports of minor, individual sources found to be leaking or contaminated. Two source leakage events were attributed to damage to the source during use. In an additional 13, there was some evidence of a manufacturing or use problem. A preliminary overview of the leaking or contaminated source reports shows that when a generic (or potentially generic) problem arises that results in source leakage, efforts are made by the manufacturer to eliminate or detect the leakage before distribution. 4.1.5 Release of Materials Eight events occurred in 1985 that involved the release of materials: Two Tc-99m generators were accidentally incinerated. Neither event resulted in contamination outside of the licensee facility. Contaminated waste was accidentally incinerated. A fire at a facility resulted in the destruction of two Po-210 static eliminator sources and damage to an Am-241 source. j l A fire at an Army base in Korea resulted in contamination of a building and the area around it by Pm-147 microspheres (3 mci).

• An event, not involving NRC or Agreement State licensed radioactive material, occurred in Morocco. A lost or stolen radiography source is reported to have caused gross overexposure of general members of the public.

50

Three laboratories were contaminated when Os-185 and Os-191 (2 mci) vaporized. i A small amount of UF6 was released when a cylinder valve was opened. Concentrations of uranium were less than 10 CFR Part 20 limits; calculated exposures at the site boundary were less than 0.5 mrem. Cs-137 contamination was found in fly-ash in a steel company bag-house. A gauge containing 200-500 mci Cs-137 may have been melted with scrap steel. The licensee was unaware of the event until a toxic waste shipment from the mill was found to be contaminated. No data are available on the extent of contamination resulting from events involving fires. The event ir which a gauge was melted in a steel mill did result in substantial clean-up costs. The contamination resulting from other events was limited. 4.1.6 Act ual or Potential Contamination of Consumer Products An additional category was defined for the data file in 1985 concerning actual or potential contamination of consumer products. Reports in which radioactive material was found in, or had a reasonable probability for being introduced into, nonlicensed products were categorized as " CON" reports. Five reports of this nature were received in 1985. A gauge originally owned by Del Monte Corporation was discovered at a scrap yard. No contamination resulted from this event. Steel originating in Brazil was found (in Florida) to be contaminated by Co-60. Calculations showed contamination levels of no more than 0.03% of MPC could result if the contaminated steel pipe is used in drinking water systems. Toxic waste from a California steel manufacturer was discovered to be contaminated with Cs-137. Investigation showed that fly-ash was also contaminated. It is probable that a gauge containing 200-500 mci Cs-137 was melted with scrap steel. A gauge licensed by Alabama was found at a scrap yard (three others were still missing). In an overflight made in an attempt to find the above three missing gauges, low level Cs-137 contamination was discovered at U.S. Pipe and Foundry in Bessemer, Alabama. All of the above events concerned the possibility of introducing radioactivity into a steel process. In two cases, gauges were found before they were melted. j In the case of the Co-60 contaminated steel, the contamination level was so low that overexposures would not result from.its use. The State of California is conducting an investigation of the Cs-137 gauge melting event. No information is available on the Alabama events from which to assess the event significance. i 51

4.1.7. Fuel Cycle Facility Event Reports

• The NRER data file contains information from six fuel cycle licensee event reports received during 1985 as follows:

Two events occurred at NFS Erwin. In the first, the top of an 11-liter cylinder containing concentrated uranium solution blew off, permitting 6 to 8 liters to escape. To avoid the problem in the future, NFS is adding instrumentation to the process that generates the solution. In the second event, uranium hexaflouride was released from an " empty" cylinder when a valve was opened. No releases exceeding regulatory limits resulted from this release, and the calculated exposure at the site boundary was 0.5 mrem. A worker at a uranium mill was exposed to 108 MPC-hours in I week. Two waste shipments, each from a different fuel fabricator, contained drums that had holes in them (one each shipment). A waste shipment from a third licensee contained partially solidified waste. These events remain under review, but do not appear to be significant. 4.1.8 Radiography Seventeen licensee event reports concerning 1985 events involved radiography. Two of the events occurred at a fixed radiography site and 15 occurred at remote (field) radiography sites. Thirteen of the reports involved personnel overexposure or potential over-exposure events and have been discussed previously. The number of radiography events reported this period does not differ substantially from the number of events reported during prior 12-month periods. 4.1.9 Manufacturing and Distribution Thirty-six events occurring during 1985 involved the manufacturing and distri-bution of byproduct material (as identified by licensees). One event was significant in view of a possible threat to public safety. The John C. Haynes Company, holding a manufacturing and distribution license, was the focus of extensive NRC activities in March 1985. An arrest was made for illegal possession and use of radioactive material, and for making material false statements to NRC. Ten to 14 curies of americium were seized. In April, the NRC ordered the company to provide sccess to the laboratory for cleanup and removal of contaminated equipment. NRC and EPA initiated a cleanup of areas of significant contamination in the laboratory and surrounding areas.

• The NRER data file does not include information from fuel cycle licensee reports of routine effluent releases.

52

O O 4.1.10 Gauges / Measuring Systems Holders of specific licenses to possess gauges are required to repo'rt failures of, or damage to, shielding, on/off mechanisms, or indicators, or detection of removable contamination on the gauge. In addition, these licensees must make reports required pursuant to 10 CFR Part 20 (lost, or stolen materials, releases of material, etc.). Sixteen events during 1985 were reported by gauge licensees. Most of the 16 events were discussed in other sections of this report: exposures, lost or stolen sources, leaking sources, and release of materials. Only one event has not been previously discussed--an event in which a portable gauge was run over by a paving roller. In this event, although the gauge was damaged, the source was not (i.e., no radioactive material was released). There were no reports of failure of or damage to gauges during 1985. 4.1.11 Abnormal Occurrences In 1985, eight nonreactor events were determined to be abnormal occurrences (three in NRC licensed activities; five in Agreement State licensee activ-ities): Four events involved overexposures of radiographers or radio-grapher's assistants; One event involved overexposcre of an employee at a manufacturer, Gulf Nuclear; One event involved unlawful possession of radioactive material by John C. Haynes Co.; One event involved breakdown of management controls at Pittsburgh Testing Laboratory, a radiography licensee; and One event involved the loss of a large (1.5 Ci Cs-137) well-logging source from a Schulmberger Well Services facility. The source was recovered about 2 months after it was lost. Of the eight events, six were reported to regulatory authorities and two, the abnormal occurrences at Pittsburgh Testing and John C. Haynes, resulted from NRC inspections of the facilities. Of the six reported events, five were over-exposures, with four of the five involving radiographers at field locations. The four radiography A0s involved overexposures of 2000 rem (estimated) to the palm; 1320 rem (estimated) to the hand; 8 and 34 rem (whole body) to two individuals; and 8-31 rad and 15 rad (whole body) to two other individuals. Conclusions A review of events reported in 1985 showed that the number and type of reported events did not differ substantially from those received in other years. In terms of sigr:ificance, it should be noted that the majority of nonreactor abnormal occurrences reported in 1985 that reflected actual or potential health effects resulted from radiography operations. 53 j

4.2 Medical Misadministration Events 7 The NRC regulates certain aspects of the uses of reactor-produced radioisotopes in nuclear medicine and therapeutic radiology. Certain diagnostic and therapy misadministration must be reported to the NRC in.accordance with 10 CFR 35.41. Diagnostic misadministration, as used in NRC regulations, refers to the misadministration of radioisotopes in nuclear medicine studies such as brain scans and bone scans. Therapy misadministration, as used in NRC regulations, refers to the misadministration of radiation from cobalt-60 teletherapy or radioisotopes in radiation therapy. The significance of any event stems from the potential. impact of the event on public health and safety. One dimension of event risk is the frequency of the event; a second is the magnitude of the potential impact of the event. AEOD has used the data collected on misadministrations for 5 years (1981-1985) to estimate error rates for certain of these misadministration events. i l Regarding the frequency of events over the 5-year period, there were 21 therapy misadministration reports that involved teletherapy machines. In these 21 events, a total of 72 patients were overtreated or undertreated. i Using patient statistics from the " Patterns of Care" study of the American j College of Radiology, the error rate per patient is estimated to be about 0.015%. For diagnostic misadministrations, there were about 2000 reported to NRC over this 5-year period. A recent study by-the Technologist Section of the Society of Nuclear Medicine estimated that about 10 million diagnostic procedures are performed annually in the United States. Since NRC regulates only 23 of the 50 states, it is estimated that about 4 million procedures are perfomed annually by NRC licensees. The diagnostic error rate per procedure is estimated to be about 0.01%. Regarding the magnitude of the potential or actual impact of the event, therapy misadministrations are associated with procedures in which large doses of radiation are administered to patients to achieve a therapeutic effect. Diagnostic misadministrations are associated with procedures designed to j permit a diagnosis to be made with little exposure to the patient. Therapy misadministrations have larger potential impacts on the health of the patient than diagnostic misadministrations. Since both teletherapy misadministrations and diagnostic misadministrations have about the same estimated error rate. the therapy misadministrations as a class appear to be individually and col-i lectively more significant than diagnostic misadministrations. For this i reason, AE0D reviews therapy misadministration reports in detail, while. diagnostic misadministration reports are reviewed from a collective or. statistical viewpoint. 4.2.1 Occurrences Reported for 1985 For this period, 297 of the approximately 2700 NRC licensees authorized to 4 perfom nuclear medicine studies or radiation therapy reported one or more misadministrations,.a total of 384 reports involving 414 patients (four were therapy patients). Of the 384 reports -of misadministrations, 380 (99%) reported diagnostic misadministrations, and four-(1%) reported therapy misadministrations. l 54

1 l l 4.2.2 Therapy Misadministrations 1 Four therapy misadministrations were reported in 1985. One of the misadminis-trations involved teletherapy, one involved brachytherapy, one involved a strontium-90 eye applicator, and one involved radiopharmaceutical therapy (iodine-131 for hyperthyroid treatment). The type and probable cause of the misadministrations is shown below: Therapy Misadministrations Reported to NRC for 1985 Type of Procedure /Cause of Error Teletherapy 1 Failure to verify treatment time Brachytherapy 1 Failure to have program to verify accuracy of treatment plans to assure delivery of correct dose Sr-90 Eye Applicator 1 No specific cause known Radiopharmaceutical 1 Dose calibrator range or function switch set to wrong position AE0D issued a case study report (AE00/C505) in December 1985 that analyzed 16 teletherapy misadministrations and two brachytherapy misadministrations that were reported to NRC between January 1981 and July 1984. The general conclusion drawn from the analysis was that the occurrence of therapy misadministrations can be reduced by improvements in licensee quality assurance procedures, especially as they relate to verifying the accuracy of patient dose calculations. Both the teletherapy and brachytherapy misadministrations that occurred in 1985 might have been prevented by quality assurance procedures directed to verifying dose calculatior.s. 4.2.3 Diagnostic Misadministrations This section discusses the 1985 experience in terms of the number of diagnostic misadministrations by the types and causes of the misadministrations. Of the 380 total reports, 294 involved the administration of the wrong radiopharma-ceutical to a patient and 67 involved the administration of a radiopharma-ceutical to the wrong patient (92% of the reported misadministrations were of 55

Q 0 these two types). The remaining diagnostic misadministrations involved four reports of the wrong route of administration, and 15 reports in which the diagnostic dose of a radiopharmaceutical differed from the prescribed dose by greater than 50%. The number, type, and cause of the diagnostic misadminis-trations was about the same as reported in 1984. Although most of the diagnostic misadministrations involved the administration of the wrong technetium-99m compound to a patient or the administration of the technetium-99m compound to the wrong patient, three diagnostic misadministra-tions involved therapy doses of iodine-131 to patients. These misadminis-trations generally resulted from techr. ologist error. In these cases, the technologist was not familiar with the type of study requested or the techno-logist did not follow the physician's prescription. On July 22, 1985, the Office of Inspection and Enforcement (IE) issued an information notice describing previous similar events involving iodine-131 and the precautions that could prevent the occurrence of these types of events. Two iodine-131 events occurred in 1985 after the notice was distributed. (In a third iodine-131 event in 1985, the wrong patient was administered a therapy dose instead of the Tc-99m scan dose for which he was scheduled.) In addition to the diagnostic misadministrations that invcived therapy doses of iodine-131, one 1985 diagnostic administration involved the administration of 200 millicuries of technetium-99m (ten times the prescribed dose) to a patient. This resulted from the technologist making an error in reading the dose calibrator. Effectively, all of the causes for these types of misadministrations involved human error. With regard to administration of wrong radiopharmaceutical, the data show that 16 of the 294 events (5%) resulted from receipt of mislabeled doses from a radiopharmacy. In the remainder of the events, 42 (15%) resulted from misinterpretation of the physician's order, and errors in the preparation or delivery of doses accounted for 140 (47%). Another 85 reports (about 30%), had other causes or contained inadequate information from which to assign a cause. The dominant cause for the wrong patient events was a failure to correlate the patient's identification with the study, 25 event (37%); the patient answering to the wrong name, 15 events (22%); and the wrong patient's name being on the requisition, 15 events (22%). Relatively simple quality assurance procedures (checking the patient identification against the study; asking the patient to state his name) might reduce the frequency of these events. The remaining two types of diagnostic misadministrations, excess dose and wrong l route of administrations, had diverse causes. I 4.2.4 Abnormal Occurrences Included in the A0 Report to Congress for the first three quarters of 1985 were eight misadministration events. Of the eight, four had occurred in 1984. The remaining four 1985 events deemed to be sufficiently serious to be classified l as abnormal occurrences included one brachytheraphy misadministration and three i diagnostic misadministrations; two of the latter involved the administration of near-therapeutic amounts of iodine, and the third diagnostic event involved administering ten times too much technetium. Two diagnostic and one therapy 56 l L

l misadministration have been proposed to the Commission for reporting as A0s.in the fourth quarter 1985 report. (See Appendix A for details.) Conclusions Four therapy misadministrations were reported in 1985. ' Both the teletherapy and the brachytherapy misadministrations that occurred in 1985 might have been prevented by quality assurance procedures directed to verifying dose calculations. Essentially all of the diagnostic misadministrations for 1985-involved either the administration of the wrong radiopharmaceutical or the administration of a radiopharmaceutical to the wrong patient. The number, type, and cause of diagnostic misadministrations are about the same as reported for 1984. The causes reported by licensees are generally the same as have been reported in the past; that is, simple errors associated with (1) preparation of radio-pharmaceuticals, (2) processing nuclear medicine requisitions, and (3) patient identification. 1 i 1 I i 57

~ 5.

SUMMARY

OF AE00 ACTIVITIES During 1985, efforts continued to increase the trend and pattern analysis of operational data and to evaluate the use of lessons of experience within the

rdustry. During the year, AE00 was assigned the responsibility for estab-

~.ishing and managing an Incident Investigation Program (IIP) intended to provide rapid and comprehensive investigations to determine probable causes and consequences of the most significant operating events occurring at nuclear power plants. An overall summary of AE0D reports issued in 1985 and in prior years is shown i below: Summary of AEOD Re) orts Issued January 1,1985 through Jecember 31, 1985 Case Studies 5 issued Engineering Evaluations 15 issued Technical Reviews 15 issued Special Studies 3 issued Trend and Pattern Studies 4 issued IIP Reports: - IIP Program Reports 1 issued - IIT Investigation Report 1 issued Summary by Year 1980 1981 1982 1983 1984 1985 Case Studies 5 5 6 2 5 5 Engineering Evaluations and Technical Reviews 20 41 66 82 55 30 Special Study and Trends and Patterns Reports 0 0 0 1 11 7 i IIP Reports 1 5.1 Reactor Operations Analysis Branch (ROAB) The Reactor Operations Analysis Branch (R0AB) performs the major technical AE0D studies on individual events and potential generic concerns. ROAB provides s.trong engineering and systems capabilities for the review of operational l events involving U.S. and foreign commercial LWR plants. The operating experi-ence of Fort St. Vrain, the only U.S. commerical gas-cooled reactor, is 58

I reviewed for ROAB under contract by ORNL. ROAB is responsible for screening LERs and other pertinent operating information; identifying events involving (, particular safety significance; conducting in-depth engineering evaluations and case studies as warranted; and formulating appropriate recommendations for action by other NRC offices. The full scope of ROAB responsibilities is discussed below. REACTOR OPERATIONS ANALYSIS BRANCH KARL V. SEYFRIT, CHIEF i Responsibilities and Work Products 5.1.1 Systematically screen LERs and other pertinent domestic and foreign event reports and determine their significance 5.1.2 Analyze and evaluate selected individual events and potentially generic safety concerns associated with operational experience 5.1.3 Request follow-up NRC actions including: Recommendations to address potentially significant safety concerns Suggestions for feedback of important lessons learned 5.1.4 Document independent technical assessments including: Case study (including a peer review process) reports. Special study reports Engineering evaluation reports Technical review reports Memoranda 5.1.5 Implement the Memorandum of Agreement with INP0 5.1.6 Provide operational experience perspective and input to related agency activities 5.1.7 Provide input to the regional SALP program (until 9/1/85) j During this reporting period, a number of significant accomplishments and studies were completed. These are summarized below for each of the individual activities identified above. l 5.1.1 Systematically Screen LERs and Other Pertinent Event Reports and Determine Their Significance ROAB screens each LER and other pertinent event reports to identify those events or situations that warrant additional analysis and evaluation. The ROAB screening process is control!sd by means of a written office procedure. From this process ?.0?3 detennines whether: (a) further engineering review of the event by AE0D or others is warranted; (b) the event meets the criteria established for Abnorr31 Occurrence reporting; (c) the event meets the 59

to the European Nuclear Energy Agency (NEA) criteria established for reporting (d) the event should be described in the Incident Reporting System; and/or bimonthly AE0D publication, Power Reactor Events. The fundamental objectivet, of the R0AB screening process are to identify and isolate precursor events and other situations where the margin of safety has been significantly degraded, and to identify from the operational experience generic situations or concerns which may have potential safety significance. A. precursor is considered an event that could have been potentially serious if plant conditions, personnel actions, or equipment failure had been slightly different. Screening accomplishments during this reporting period include: A total of about 3000 LERs were screened by R0AB. Of these, approximately 130 LERs were judged to be significant, warranting additional NRC attention. Of the 130 LERs which were considered significant, about 75% were related to generic issues or to matters being actively pursued by NRR, IE, or RES. The remainder (about 30 events) were designated by ROAB for further analysis and evaluation by AE00. In addition to U.S. experience, R0AB routinely reviews foreign operating experience reports. During this reporting period, 84 NEA/ IRS reports and 2 IAEA reports were screened. Additionally, approximately 100 foreign reports obtained through bilateral agreements were reviewed. Other U.S. operational information reviewed by ROAB during this reporting period included approximately: 3500 regional inspection reports; 160 Part 21 reports; 4000 regional daily report summaries; 475 preliminary notification reports, and 110 INP0 documents. 5.1.2 Analyze and Evaluate Selected Individual Events and Potentially Generic Safety Concerns During this reporting period, ROAB completed three case studies, two special reports, 15 engineering evaluations, and 15 technical review reports. Representative studies are discussed below; a complete listing of completed studies is provided in Section 5.1.4. (Section 5.1 3 provides examples of follow-up NRC actions recommended by ROAB in the case studies discussed below and in selected engineering evaluations.) The AE0D case study on "0verpressurization of Emergency Core Cooling Systems in Boiling Water Reactors" was issued in September 1985 (AE0D/C502). Eight operating events, each involving the failure of a testable isolation check valve on the injection line of an emergency core i cooling system, were identified and evaluated. Five of the eight events involved an additional failure of the second and final isolation barrier, the motor-operated injection valve. Each of the events studied is considered a precursor to a loss-of-coolant accident outside containment, due to a potential loss of integrity of the lower pressure emergency core cooling system. Collectively, these operating events indicate the likelihood of an interfacing loss-of-coolant accident is higher by. a -factor of two to several orders of magnitude than had been previously assessed. AE0D's recommendations to prevent a recurrence of these events were forwarded to NRR for appropriate action. Based on the results of the 60 ~

l .I 4 AE0D case study, NRR established Generic Issue 105 with a "high" priority ranking to address this concern. The AE00 case study on " Safety Implications Associated with In-Plant Pressurized Gas Storage and Distribution Systems in Nuclear Power Plants" was issued in June 1985 (AE00/C501, NUREG/CR-3551). The case study report was prepared by the Oak Ridge National Laboratory. The report presented-many observations that were made during plant site visits. The report highlighted areas where handling, storage, distribution and use of pressurized gases at nuclear power plants were not necessarily carried out in the manner that meets commonly accepted industrial practices. The report also presented instances in which NRC inspectors found the use of pressurized gases different from that which was stated in plants' FSARs or' staff safety evaluation reports. The report presented recommendations for improving the identification, handling and use of pressurized gases in nuclear power plants. AE0D's. recommendations to address the concerns involved in the study were forwarded to IE and NRR for appropriate action. i Following release of the final report an.IE Information Notice 85-87 was issued informing licensees of the availability.of the case study. As noted in the information notice, the case study "provides a detailed, thorough, technical review and offers a broad perspective for many aspects of using compressed gases." The AE0D case study on " Decay Heat Removal Problems at U.S. Pressurized Water Reactors" was issued in December 1985 (AE0D/C503). The study analyzed PWR experiences involving loss of decay heat removal (DHR) systems. Data analysis indicated that the root causes of most. loss-of-DHR events were human factor deficiencies involving procedural inadequacies and personnel errors. Most of the errors were committed during either maintenance, testing, or equipment repair operations. The case study report presented several recommer.dations to improve DHR system avail-ability including: upgrading coordination, planning and administrative control of surveillance, maintenance, and testing activities performed during shutdown; providing operator aids to assist in recovering from loss-of-DHR events; upgrading training and qualification requirements for operations and maintenance staffs; and implementing several hardware modifications. The report was. forwarded to NRR for review and. appropriate action. A preliminary AE0D case study on " Effects.of Ambient Temperature on Electronic Equipment in Safety-Related Instrumentation and Control Systems" was issued for peer _ review in November 1985. The report i documents AE0D's review and evaluation of an event involving a loss of control room cooling and subsequent failure of electronic components at the McGuire Station,.together with other events involving similar failures of heat' sensitive electronic components at other operating plants. Following peer review, a final-report will be issued in 1986. In addition to the above case study reports, AE00 issued a number of other studies which identified potentially significant safety concerns. Some representative examples are as follows: 61

An engineering evaluation (AE0D/E511) was issued in September 1985 on " Closure of Emergency Core Cooling System Minimum Flow Valves." The study found that the minimum flow bypass lines provide an essential pump protec-tion feature for the low pressure ECCS pumps and that the affected safety system trains at both Brunswick and Peach Bottom units should have been l declared inoperable when the minimum flow valves for these systems were closed and deactivated. An evaluation of a data search for similar events at other plants indicated that not all licensees may recognize the importance of minimum flow valves for ECCS pump operability. It was also found that the design of the control logic for the Brunswick core spray system minimum flow valves was inadequate to assure that the valves could perform their containment isolation function in all required situations. The results of this evaluation were discussed in IE information Notice-85-94, " Potential for Loss of Minimum Flow Paths Leading to ECCS Pump Damage During a LOCA." An engineering evaluation (AE0D/E514) on " Core Damage Precursor Event at Trojan" was issued in October 1985. The report evaluated five events that occurred at the Trojan nuclear power plant which could have had serious consequences for equipment or personnel had they occurred under different circumstances. The potentially most serious event occurred on September 20, 1984. The event involved operator errors which initiated a transient and failures of an emergency diesel generator, a diesel-driven auxiliary feedwater pump, and a turbine-driven auxiliary feedwater pump. The multiple, independent, and undetected failures of these safety-related components resulted in a partial loss of emergency onsite ac power and the loss of the safety-grade auxiliary feedwater system. The report also evaluated other events at Trojan including: a May 4, 1984 primary system water level indication failure which caused a temporary total loss of the residual heat removal (RHR) system; a September 26, 1984 reactor trip in which a main steam safety valve stuck open after the control room operators failed to notice that the rods were in manual during a turbine runback; a September 13, 1984 compression fitting failure which occurred during seal table maintenance because the fitting was used incorrectly; and a second compression fitting failure on September 17, 1984, involving a pressurizer instrument line. An engineering evaluation (AE0D/E512) on " Failure of Safety-Related Pumps due to Debris" was issued in September 1985. The evaluation addressed an event at Salem Unit I that occurred on July 13, 1984, during a refueling outage. A centrifugal charging pump (CCP) seized after running for approximately 30 seconds during emergency core cooling system surveillance testing. Seizure of the CCP was attributed to metal filings-which lodged between the impeller and wearing rings. The metal filings resulted from work previously performed on the common header of the CCP vent lines. The loss of the CCP occurred at a time when ECCS operability was not required. However, some safety concerns were identified by AE0D from the review of this event and related operating experiences involving the failure or degradation of ECCS components due to debris ingestion. A technical review (AE0D/T502) on a " Comparative Analysis of Recent Feed-line Water Hammer Events at Maine Yankee, Calvert Cliffs, Salem, McGuire and Palisades" was issued in March 1985. The concerns raised by the study were forwarded to IE and were included in IE Information 62

O s Notice 85-76, "Recent Water Hammer Events," which was issued on September 19, 1985 to all power reactor licensees. An engineering evaluation (AE00/E505) on " Service Water System Air Release Valve Failures" was issued in March 1985. The evaluation addressed a number of air release valve failures at Browns Ferry, Hatch and Duane Arnold. Three potential safety problems were identified as a result of the failures. These were: (1) reduced service water flow to the RHR heat ' exchangers, (2) flooding of the service water pump room, and (3) increased system unavailability due to the need for maintenance. As suggested by the study, Region II followed up on the Hatch Nuclear Plant residual heat removal service water air release valve failures and corrective action program. An engineering evaluation (AE0D/E513) on "High Pressure Core Spray System Relief Valve Failures" was issued in March 1985. The evaluation concluded that excessive hydraulic back pressure from the discharge side of the HPCS discharge relief valve caused the relief valve's internal bellows seal assembly to fail at LaSalle 1 and 2. As a result of this study, it was concluded that River Bend Unit I was the only other BWR potentially susceptible to the kind of relief valve failures which occurred at LaSalle. Following issuance of this engineering evaluation, NRR performed a detailed review of the design and the potential for HPCS relief valve failures at River Bend. An engineering evaluation (AE0D/E510) on " Disabling of a Shared Diesel Generator Set Due to Electrical Power Supply Arrangement for Support Auxiliaries" was issued in July 1985. The report concluded that this situation is not generic to all U.S. nuclear plants, but may be applicable to a few specific plants. An issue of Power Reactor Events included this topic to in order to feed back the identified potential safety concern to industry. An engineering evaluation (AE0D/E504) on " Loss of Actuation of Various Safety-Related Equipment Due to Removal of Fuses or Opening of Circuit Breakers" was issued in March 1985. In view of the potential frequency and safety consequences for such events, the report suggested the elimination of fuse removal practices where practical, during maintenance and/or plant modification activities. The report's findings were discussed in IE Information Notice 85-51, " Inadvertent Loss or Improper Actuation of Safety-Related Equipment." An engineering evaluation (AE0D/E501) on " Motor-0perated Valve Failures D'le to Hammering" was issued in January 1985. Hammering is the phenomenon experienced by a motor-operated valve when the valve is subject to repeated closing attempts after it has already reached the fully closed position. The evaluation concluded that the hammering problem could be generic to many operating reactor units. As a result of this study, IE i Information Notice 85-20, " Motor-0perated Valve Failures Due to Hammering l Effect" was issued. An engineering evaluation (AE0D/E507) on " Electrical Interaction Betweer) Units During a Loss of Offsite Power on August 21, 1984" was issued in May 1985. The event, which occurred at the McGuire Station, was complicated 63 .- l

.~. - by failures that occurred as a result of voltage surges, random component failures, trips of'both units and problems in the common auxiliary control power system. The. report concluded that at multiple unit sites, events initiating at one unit can propagate and affect more than one unit due to problems in systems that are shared by both units. ~ An engineering evaluation (AE00/E515) on " Inadvertent Actuation of Safety. Systems Due to Cross-Talk" was issued in January 1985. The event occurred at the Kewaunee Nuclear Power Plant due to crosy-ta.1k between safety-related instrument channels during surveillance testing that, led to the inadvertent actuation of the containment spray system. A-review of the data-base of. operational events identified ten additional events at other L operating nuclear plants that could be considered to have been caused by-cross-talk interaction between instrumentation channels during testing or maintenance activities. The study concluded that for plants where cross-talk has been a re.urring problem, the licensees have initiated actions to reduce the faquency of occurrence. An engineering evalu& tion (AEOD/E503) on " Partial failures of Control Rod Systems to Scram" was issued in March 1985. This evaluation considered j partial control rod failures that occurred since the Salem ATWS event. Thirteen events which occurred between February 1983 and December 1984 involving failures of control rods to perform their trip function were identified. Three potential comon-cause failure mechanisms were identified by the review. The report. stated that although the failure 3 mechanisms were (in most cases) identified and promptly corrected, their 4 very existence in such an established and safety significant system (i.e., the control rod system) remains-a serious concern. Technical review report AE00/T507,-" Standby Liquid Control System (SBLCS) Pressure Relief Valves Lift at a Pressure Lower Than Reactor Coolant-- 4 Pressure," was issued in August 1985. ;The review of an event which occurred at the Hatch 2 plant found that the SBLCS would have been unable to inject sodium pentaborate into the reactor coolant system'if. it had been called upon to do'so because the relief valves lifted at l 1 600 and 700 psig, rather than the set pressure of 1300 psig. The available operational data suggested the event was isolated and was a i. plant-unique valve degradation rather-than an indication of a generic problem. An engineering evaluation (AE00/E502) issued in January 1985, "RHR Suppression Pool Cooling Valve to Operate" addressed valve failures due to the failure of an anti-rotation device. :The study found that anti-i rotation device failures in valves manufactured by Anchor Darling were not unique because valves of other manufacturers also have experienced similar failures. The defective valves, other than Anchor Darling, found by this study were used in safety systems such~as RHR, AFW and main steam-isola-tion. The failures of the anti-rotation devices were also attributed to a design deficiency..The problem and the corrective actions for. anti-rotation devices in the Anchor Darling valves were addressed in IE Information Notice 83-70. However, the study found that the problem was 'also applicable to valves other than thc3e manufactured by Anchor Darling 1 and that the potential existed for licensees to restrict their review of t i j 64 i .--.-- 4. ., _ _ _._ _-.. _. - -,- -.. _ _ -._. - -_-,. _._- - _, _ ~,_ - -, _ _ a..

the infonnation notice to only Anchor Darling valves. Subsequently, IE l issued Supplement 1 to Information Notice 83-70 on March 4, 1985. An engineering evaluation (AE0D/E506) on " Valve Stem Susceptibility to IGSCC Due.to Improper Heat Treatment" was issued in May 1985. The evaluation. reviewed an event involving stem fracture of a valve in the RHR syster. N e valve stem was made of type 410 stainless steel. Metallurgical examination showed that the stem had failed from. inter-2 granular stress corrosion cracking (IGSCC). Excessive surface hardness, caused by improper heat treatment during manufacture of the stem material was found to be a contributing factor to the stem failure. This study indicated that the valve stems made of 400 series stainless steel and heat treated to high hardness are highly susceptible to IGSCC. Existing codes do not address the actual testing to verify the hardness for.the stem 1 materials after heat treatment. The report suggested that IE consider i issuing an information notice for feedback. purposes, and that RES consider the ~ adequacy of the applicable code requirements for assurance of proper heat treatment.. Subsequently, IE Information Notice 85-59, " Valve Stem. Corrosion Failure" was issued, and a letter wat issued to the ASME by RES requesting that they review this matter and take appropriate action to 4 limit the selection of valve stems to materials that are not susceptible to IGSCC. Two special study reports were issued by ROAB during this reporting period. J A special study (AE00/S501) on " Review of Operational. Experience From Non-power Reactors" was issued in March 1985. The study evaluated operational experience reports submitted from non-power reactor licensees l to determine if the information was. adequate and if the operating experiences 'of non-power reactors should be included in AEOD's program j for the evaluation and feedback of operational data. The review covered operational data submitted over approximately 3 years. The study found variations in reporting due in part to non-standard technical specifications among non-power reactors with licensed power levels less than or equal to i 2 Mwt. As the licenses come up for. renewal, licensees are being encouraged by NRR to adopt the ANSI /ANS standard for technical-specifications for J non-power reactors. The study noted that as licensees adopt this ) standard, reporting will become more uniform. Since postulated events i for non-power reactors have only limited safety significance, and the-l probabilities for an impact on public health and safety is low, the study. i found that the reviews of operating experience'being performed by the j Regions and NRR are considered. sufficient to assure-proper evaluation and feedback of non-power reactor operatinhol. experience. The study concluded that additional independent review by AE0D was no longer warranted. The second special study (AE0D/S503) on " Evaluation of Recent Valve l Motor-Operator Burnout Events" was issued in September 1985. The study evaluated valve motor-operator burnout events subsequent to the 1980 time frame covered by a previous AE0D case study (C203). The_ report identified more than 200 valve operator motor burnout. events including more than 180 events during the approximate 4-year time frame from 1981 to early 1985. This number of events is significantly greater than the 65 l ~ ,,_ _.~ __ _ __.... _ _ _ _ _ _. _., _ -.. _.

0 19 events reported in AE00/C203 for 1978,1979, and 1980. These events pose potential significant safety issues because: (1) motor-operated valves are used extensively in safety systems where performance of the safety function requires that valves must open or closed; (2) the motor failure mechanism can be common mode based on an overall plant philosophy covering thermal overload devices.and surveillance procedures; (3) the failed motors can remain undetected for long periods of time; and (4) motor burnout has resulted in damage to the valve operator in a-manner that prevented valve operation by either the motor or manual drive mechanisms. The report recommended that NRR expedite implementation of the proposed NRR plan, which had been prepared to respond to AE0D/C203, to address motor burnout. 5.1.3 Request Follow-Up NRC Actions Study reports issued by ROAB during this reporting period included a number of formal recomendations for NRC program office follow-up actions related to potentially significant safety concerns or suggestions for feedback of important lessons learned from plant operating experiences. In addition, during this reporting period, a number of actions were initiated or completed by one or more NRC program offices as a result of recommendations or suggestions previously made by ROAB. Selected examples of recommendations or suggestions made by ROAB during this period or actions taken by others as a result of earlier requests are presented below (see Section 6 for a complete status report on all AEOD recommendations that were outstanding during 1985): The case study report on: "0verpressurization of Emergency Core Cooling Systems in Boiling Water Reactors" contained a number of recommendations directed at reducing and eliminating the root-causes of the reported operational events. The dominant causes of the events involved unantici-pated degradation or failure of the pressure isolation protection provided by the valves installed between the high pressure reactor coolant system and the lower pressure emergency core cooling systems. In order to reduce or eliminate these problems the report recomended: disabling the air operator associated with the isolation check valves; additional leakage testing of isolation check valves; improved maintenance and testing procedures involving pressure isolation components; and reduced surveillance testing frequencies of isolation components. The case study report on " Safety Implications Associated with In-Plant Pressurized Gas Storage Systems" contained several recomended actions l aimed at improving the identification, handling and use of pressurized I gases in nuclear power plants. Specifically, the report recommended: procedures to prevent unacceptable damage to safety-related equipment from gas cylinder missiles; protection against hydrogen explosions or fires in areas which could adversely impact safety-related equipment; and identifi-cation of pressurized tanks and lines in safety-related plant areas. The case study report on " Decay Heat Removal Problems at U.S. Pressurized Water Reactors" contained a number of recommended-actions directed at reducing the frequency of the reported operational occurrences and l improving operator response to-loss of decay heat removal events. In particular, the report recommended: an assessment of the need for regulatory requirements to improve the human performance aspects of DHR l 66

a availability; improved reactor vessel level measuring and monitoring during plant shutdown operation; improved mari-machine interfaces related to DHR operation; consideration of DHR suction bypass lines as an alternative to redundant drop lines; consideration to removal of suction line isolation valve autoclosure interlocks; and examination of the need for DHR system redundancy during Modes 4 and 5. Engineering Evaluation and Technical Review Suggestions Engineering evaluation report AE00/E511 suggested that an information notice be issued on closure of emergency core cooling system minimum flow valves. As a result, IE Information Notice 85-94, " Potential for loss of Minimum Flow Paths Leading to ECCS Pump Damage During a LOCA," was issued on this subject. Technical Review report AEOD/T502 suggested that an information notice be issued on a number of recent PWR feedline water hamer events. As a result, IE Information Notice 85-76, "Recent Water Hammer Events," was issued on these and other related water hammer experiences. Engineering evaluation report AE0D/E513 sugge.. " review of the potential for HPCS pump discharge relief valve failure:, u. River Bend in view of the repeated relief valve failures at LaSalle. In response to the request, NRR performed a special evaluation of the HPCS relief valve design application at kiver Bend. Engineering evaluation report AE0D/E504 suggested that an information notice be issued on the causes, consequences and possible preventive measures associated with events involving adverse system consequences resulting from removal of fuses during equipment maintenance or modifica-tion activities. In response, IE Information Notice 85-51, " Inadvertent loss or Improper Actuation of Safety-Related Equipment," was issued. Engineering evaluation report AE0D/E501 suggested that an information notice be issued for the events, causes and possible corrective action associated with MOV failures due to hammering. In response to the request, IE Information Notice 85-20, " Motor-0perated Valve Failures Due to Hamering Effect," was issued. Engineering evaluation report AE0D/E506 suggested that an information notice be issued and applicable ASME code material requirements be reviewed as a result of a valve stem fracture failure caused by intergranular stress corrosion cracking resulting from improper material heat treatment. In response to the requests, IE Information Notice 85-59, " Valve Stem Corrosion Failure," was issued and the ASME was requested by RES to review the requirements for the selection of valve stem materials. i In addition to the requested actions contained in AEOD studies issued during j this reporting period, a number of actions were initiated as a result of past AE00 study suggestions and recommendations. Selected examples of actions taken l by others during this reporting period as a result of previous AE0D report recommendations, suggestions, or requests is presented below. 67 a

AE0D's special study report AE0D/S501 on operational experience from non-power reactors was reviewed by NRR. As a result, NRR agreed to review all data from non-power reactors. AE0D engineering evaluation AE0D/E426 revie'..ed the technical specification requirements associated with low-temperature overpressure (LTOP) potential. AE0D case study AE0D/C401 discussed low temperature over-pressure events at Turkey Point Unit 4. As a result of these two reports, NRR determined that LTOP events should be identified as a new generic l issue. When this issue (No. 94), was prioritized, it was classified as " Medium" priority. AE00 provided additional information which addressed the elevated significance of this issue. As a result of this correspond-ence, the prioritization was upgraded to "High." As a result of reccmendations in a memorandum from C. Michelson to f R. Burnett, dated January 12, 1982, the ED0 established an interoffice comittee in the spring of 1985 to review the assumptions and methodology used for vital area determinations. In addition, NUREG/CR-4392 was issued as the resolution of GI A-29, " Nuclear Power Plant Design for Reduction of Vulnerability to Sabotage." As a result of a recommendation in AE0D case study AE0D/C402 on experience related to moisture intrusion in electrical equipment at commercial power reactors, the criteria for inspection of electrical equipment are being revised. As a result of a memorandum from C. Michelson to R. Mattson, entitled "NRC Action Plan Developed as a Result of TMI-2 Accident--Draft 3, Task II.E.3 f Decay Heat Removal," dated April 24, 1980, work is in progress on a NUREG l l which discusses the reliability of DHR systems. l l As a result of previously identified UHI concerns discussed in a memoran-f dem from H. R. Denton and C. J. Heltemes to J. G. Keppler, Regional l Administrator, Region III, dated 12/3/85 and entitled " Failures in Upper j Head Injection System," the Sandia National Laboratory has been contracted to perfom LOCA calcul6tions for UHI plants to study the effects of r I accumulator nitrogen irjection. A number of suggestions or recomendations made in earlier or preliminary AE0D studies also served as the principal input and/or basis for a number of IE bulletins and information notices issued during this reporting period. 1 IE-B 85-01 Steam Binding of Auxiliary Feedwater Pumps (AE0D/C401) IE-B 85-03 Motor-0perated Valve Comon Mode Failures During l Plant Transients Due to Improper Switch Settings (AE00/C203) 68

IE-IN 85-23 Inadequate Surveillance and Post-maintenance and Post-modification System Testing (AE00/E404) IE-IN 85-89 Potential loss of Solid-State Instrumentation Following Failure of Control Room Cooling (Preliminary Case Study Report) 5.1.4 Document Independent Technical Assessments During 1985, the following studies were completed by R0AB personnel: Reactor Case Study Reports - 1985 No. Subject Date Issued Author C501 Safety Implications Associated with 6/13/85 ORNL In-Plant Pressurized Gas Storage and Distribution Systems in Nuclear Power Plants (Contract) Study C502 Overpressurization of Emergency Core 9/9/85 P. Lam Cooling Systems in Boiling Water Reactors C503 Decay Heat Removal Problems at U.S. 12/23/85 H. Ornstein Pressurized Water Reactors Special Study Reports - 1985 No. Subject Date Issued Author S501 Review of Operational Experience From 3/15/85 D. Zukor Non-Power Reactors S502 AE0D Se:niannual Report 5/2/85 C. J Heltemes 5503 Evaluation of Recent Valve Motor 9/19/85 E. Brown Operator Burnout Events Reactor Engineering Evaluation Reports - 1985 No. Subject Date Issued Author E501 Motor-0perated Valve Failures Due to 1/18/85 M. Chiramal Hamering Prolem E502 Failure of RHR Suppression Pool Cooling 1/25/85 C. Hsu Valve to Operate 69

0 No. Subject Date Issued Author E503 Partial Failures of Control Rod Systems 3/11/85 M. Chiramal to Scram E504 Loss of Actuation of Various Safety-3/29/85 F. Ashe Related Equipment Due to Removal of Fuses or Opening of Circuit Breakers E505 Service Water System Air Release Valve 3/29/85 S. Salah Failures E506 Valve Stem Susceptibility to IGSCC Due 5/20/85 C. Hsu to Improper Heat Treatment E507 Electrical Interaction Between Units 5/17/85 M. Chiramal During Loss of Offsite Power Event of August 21, 1984 E508 Nuclear Plant Disturbances Caused by 5/24/85 S. Rubin Bumped Electro-Mechanical Components E509 Salem Unit 2 Depressurization Event 7/25/85 R. Freeman E510 Disabling of a Shared Diesel Generator 7/30/85 F. Ashe Set Due to Electrical Power Supply Arrangement for Support Auxiliaries E511 Closure of Emergency Core Cooling System 8/9/85 E. Leeds Minimum Flow Valves ) E512 Failure of Safety-Related Pumps Due to 9/4/85 R. Freeman Debris E513 High Pressure Core Spray System Relief 9/16/85 S. Salah Valve Failures i E514 Core Damage Precursor Event at Trojan 10/8/85 D. Zukor E515 Inadvertent Actuation of Safety Systems 12/11/85 M. Chiramal Due to " Cross-Talk" Reactor Technical Review Reports - 1985 No. Subject Date Issued Author T501 Failure of Automatic Protection For 1/22/85 R. Freeman Boron Dilution Event at Callaway Unit 1 l 70

1 i No. Subject Date Issued Author i T502 Comparative Analysis of Recent Feedline 3/18/85 E. Leeds Water Hammer Events at Maine Yankee, Calvert Cliffs, Salem, McGuire and Palisades T503 Pressurizer Level Instrumentation of 5/2/85 M. Chiramal Combustion Engineering Reactor Units T504 Loss of Instrument Air and Subsequent 5/17/85 R. Freeman Pressure Transient T505 Beaver Valley Component Cooling Water 7/17/85 C. Hsu Pump Damage T506 Primary System Release Due to Pres-7/25/85 T. Cintula surizer Degas Relief Valve Lifting T507 Standby Liquid Control System Pressure 8/13/85 E. Brown Relief Valves Lift at a Pressure Lower than Reactor Coolant Pressure T508 Browns Ferry Nuclear Plant HPCI System 8/15/85 E. Leeds Performance T509 Inadequate Surveillance Testing Pro-8/29/85 F. Ashe i cedures for Degraded Voltage and Undervoltage Relays Associated with 4160 Volt Emergency Buses T510 Xenon Induced Power Oscillations at 9/4/85 R. Freeman i Catawba T512 Incorrect Plugging of Steam Generator 10/24/85 R. Freeman Tubes T513 Flooding of Safety-Related Valves in 11/8/85 D. Zukor Pits T514 Potential Loss of Component Cooling 11/25/85 D. Zukor Water Due to Maladjustment of Relief Valves T515 Residual Heat Removal Service Water 12/5/85 S. Salah Booster Pump Air Binding at Brunswick Unit 1 71

5.1.5 Provide Input to the Regional SALP Program (Until 9/1/86) ROAB continued to assess the quality of LERs submitted by each operating unit with respect to clarity, degree of detail, responsiveness and degree of follow-through. General and specific comments were provided to the Regions regarding the nature and adequacy of the LER information submitted. These as-sessments continued until September 1, 1985, at which time review responsi-bility was transferred to Program Technology Branch which arranged to continue these reviews under contract by EG&G, Idaho. During the period from January 1, 1985 to September 1,1985, ROAB completed LER quality reviews for 27 operating plants and forwarded the results to the Regional Offices. The reviews continued to reflect wide variations in the completeness of reports submitted by licensees. There have been notable improvements by some licen-sees, but overall there was little change from previous years. 5.1.6 Implement the Memorandum of Agreement with INP0 AE00 and INP0 share results of completed studies related to review and assess-ment of operational data. Additionally, listings are periodically exchanged for studies planned or in progress. Periodic informal meetings are held to discuss concerns of mutual interest. During 1985, three such meetings were held (March 7. July 18, and November 26). In addition to these meetings, con-siderable informal discussion occurs between the R0AB staff and their INP0 counterparts. ) 5.1.7 Provide Operational Experience Perspectives and Input to Related Agency Activities ROAB is often requested to provide comments and perspectives on a number of technical issues and concerns and to serve on various task forces. Such as-signments are authorized when it is believed that the request offers a desir-able method of communicating the lessons learned from operating experience and related AE0D studies. Examples of this type of activity during this reporting period, included: Participating in a special inspection of Palo Verde in connection with an operational event at Palo Verde Unit 1. Participating in the Electrical Relay Review Group associated with USI A-46, " Seismic Qualification of Equipment in Operating Plants." Participating in an inspection associated with air system degradation that affected the auxiliary feedwater system at Turkey Point. Providing input to NRR on the degradation of safety systems by nonsafety systens for the resolution of USI A-17, " Systems Interactions." Participating in the IAEA technical committee meeting on the subject of combining risk analysis with operating experience and for the development ) of an IAEA technical document on the subject. i 72

~ Providing comments on the task action plan for Generic Issue 105, " Interfacing LOCAs in BWRs." Providing additional information to a foreign organization on' the cor-relation between the indicated temperature in the shutdown cooling system to the unmonitored bulk coolant temperature rise during a loss of shutdown cooling at a U.S. PWR. Providing additional information on potentially sensitive electro-mechanical components related to the resolution of USI A-46, " Seismic Qualification of Equipment in Operating Plants." -Participating in Incident Investigation Teams that reviewed events at-Davis-Besse in June 1985 and San Onofre in November 1985. 5.2 Program Technolooy Branch (PTB) i PTB is responsible for a broad range of program, technical, and administrative activities within AE00, as shown below. PROGRAM TECHNOLOGY BRANCH FREDERICK J. HEBDON, ACTING CHIEF Responsibilities and Work Products 5.2.1 Formulate guidance for and assess the adequacy _of Licensee Event Reports (LERs) 5.2.2 Conduct a comprehensive and systematic trends and patterns program 5.2.3 Conduct Nuclear Plant Reliability Data System.(NPRDS) evaluation program and. coordinate NRC guidance on the NPRD System 5.2.4 Formulate Abnormal Occurrence (AO) guidance and prepare'A0 reports 5.2.5 Issue operational data feedba.ck repprts, including bimonthly Power Peactor Events reports 'and monthiy LER Compilation reports 5.2.6' Prepare reports on U.S. events to the Nuclear Energy Agency's Incident Reporting System (NEA-IRS) and provide support to_the International Atomic Energy Agency's Incident iteport~ing System -(IAEA-IRS)' i 5.2.7 Develop techniques to apply PRA perspectives to the screening and analysis of operational events 5.2.8 Develop guidance for the-NRC's program for operational data review (Manual Chapter _0515) 73

'h 9 5.2.9 Develop and manage data bases, including: Sequence Coding and Search System (SCSS) Foreign Event File 5.2.10 Access INP0 data files, including the NPRD System 5.2.11 Formulate and operate internal management information systems, including: LER screening results (WAMS) Resource accountability (TACS) Incident Reporting System (IRS) During this reporting period, a number of significant milestones and actions were accomplished. These are summarized below for each of the individual activities identified above. 5.2.1 Formulate Guidance for and Evaluate Licensee Event Reports (LERs) On January 1,1984,10 CFR 50.73, " Licensee Event Report System," became effec-tive. This rule modified and codified the LER system which had previously been defined in technical specification requirements. PTB continued to resolve numerous questions regarding the interpretation of the new rule and to conduct assessments of LER reporting patterns and quality. t In order to better assess the quality of the LERs being prepared by licensees, and to bring about improvements where necessary, AE00 contracted with EG&G to assess the quality of LER reporting. The study consisted of a detailed analy-sis of 415 LERs (five from each operating nuclear power plant) that had been prepared under 10 CFR 50.73. The LERs were assessed based on a wide range of criteria from overall quality to completion of each block on the new LER form. The most substantive and frequent deficiencies identified'in LERs involved a lack of detail and clear descriptive information, including the -failure 'to: Describe the root cause(s) of the event (i.e., the cause or causes that, if corrected, will prevent recurrence of the event). 'In many cases, the root cause may not have been described because it was not identified. Explicitly describe the plant response to and the safety consequences, or lack of consequences, of the event. For example, frequently only a state-ment was made, without any stated basis, that the plant responded normally or that there were no consequences. 1 Adequately describe the corrective action. This is particularly important if the lessons from operational experience are to be fed back to other licensees. Include a clear description of personnel errors and other pieces of re-quired data (e.g., type of personnel and their license status, dates and/or times of occurrences, manufacturer and/or model number of failed components, method of discovery of component failures). 74 ~

1 A description of the observed deficiencies and guidance on how to improve the content of LERs was issued to the industry during September 1985 as Supplement 2 to NUREG-1022. In addition, the LER audit program was revised in 1985 to provide specific assessments of each licensee's LERs as input to the Regional based SALP program (see Section 3.) 5.2.2 Formulate and Conduct a Trends and Patterns Program The AE0D Trends and Patterns Programs Plan for FY 86-88 was developed in 1985 and distributed to NRC offices and Regions for information and comment in Janu-ary 1986. This plan describes the AE0D program for the periodic analysis of sets of operational event data reported by commercial power reactor licensees in the Licensee Event Reports (LER3) and to INP0's Nuclear Plant Reliability Data System (NPRDS). " Trends and Patterns" is used to describe a program for analyzing incidents of low individual significance but which may be significant because of their frequency. Trends and patterns analysis usually assesses op-erational data with limited prior formulation of a concern, whereas an engi-neering review usually begins with the formulation of a specific concern. This plan, which revises and extends the earlier plan (AE0D/P402-Trends and Patterns Program Plan, FY 84-86), differs from the previous plan in two re-spects. First, instead of reviewing LER data in a general way at the event l level, the new plan calls for more focus on analysis at the event level (i.e., reactor trips, ESF actuations), and for coverage of these subjects in-greater detail in yearly reports. Second, the program will include trends and patterns analysis of component level data using NPRDS data. The AE0D Trends and Patterns Program consists of three separate program eleinents-(1) Analysis of Event Data events ccvered by the requirements of 10 CFR 50.73 [ gory of operational (1) react Four indepth reports, each treating a different cate ESF actuations, (3) systems unavailability, and (4) technical specifica-tions violations and shutdowns], and a fifth report reviewing the perfor-mance of recently licensed plants will be issued each year covering the events from the most recent calendar year and comparing the latest results with that from earlier years. Further, each report will contain findings, conclusions, and recommendations for correcting problems found to be most significant. The preliminary results of selected 1985 studies are dis-cussed in Section 4. (2) Analysis of Component Data AE0D had developed and pilot tested a program for the systematic analysis of NPRDS data. Between ten and 20 reports on selected key components will be issued each year. The full-scale implementation of this p'rogram began in January 1986. (3) Analysis of LER Failure Data Twc - ts were issued during 1985 (discussed later in this section) pre-v- .he results of AE00's automated analysis of LER data for 1981 ano 75

for the combined years 1981 through 1983 using data attributes stored in the Sequence Coding and Search System. Because of the change in the LER rule in January 1984, further analysis of LERs similar to the analysis in these two reports is now scheduled for FY 88. when several years of opera-tional data (e.g., CY 1984-1987) under the new rule are available for analysis. AE0D issued report P503, " Trends and Patterns Analysis of Engineered Safety Feature Actuations at Commercial U.S. Nuclear Power Plants," in August 1985. This study covers the January-June 1984 time period, and serves as the pilot study for the periodic trends and patterns analysis of ESF data. The analysis of July-December 1984 data will be issued in spring 1986, and the preliminary results from this study are discussed in Section 4. AE0D issued report P504, " Trends and Patterns Report of U.S. Light Water Reactors in 1984," in At.qust 1985. This report is the first in the annual series of scram analyses. The study of 1985 data was ini-tiated in late 1985 and will be issued in mid 1986. Preliminary results from this study are discussed in Section 4. AE00 initiated trends and patterns analyses of system unavailability and violations or shutdowns required by technical specifications as reported in 1984 and 1985. AE00 initiated an analysis of operational data gathered in the first two years of licensed operation for new plants. This analysis will be published in 1986. o A report prepared by INEL under contract to AEOD entitled " Exploratory Trend and Pattern Analysis of 1981 Licensee Event Report Data" was issued as NUREG/CR-4071 in April 1985. This study was expanded to include all LERs from 1981 through 1983; the draft report from the expanded study was completed in January 1985 Observations from the 1981-1983 study include: -(1) even with three years of data (13,000 LERs), the data are sparse and tables of events cross classified by various characteristics contain many zeros; (2) very few of the reported incidents directly affected unit availability (of all inci-dents reported during 1981-1983, only about 5% led to power reductions or shutdowns); (3) reported releases of radioactive effluents to the environ-ment were rare (about 2% of the reported events); and (4) reported inci-dents of personnel overexposure were also infrequent (40 occurrences in this 3-year period). Specific recommendations were made by EG&G for further investigation of component faults based on outlier analysis. ' AE0D subsequently evaluated these recommendations and, based upon the information available to AE00, determined that no further action was necessary at this time. AE0D asked the other NRC offices and regions to independently review these recom-mendations and the AE0D evaluation. The other NRC offices replied that they' believed AEOD's evaluation to be factually correct; none of the offices believed that any further action was necessary. 76

In 1985, AE09 started a program to systematically analyze NPRDS data. The NPRD System is an industry-wide system for tracking the performance of selected sys-tems and components at nuclear power plants. Since almost all U.S. plants in commercial operation supply detailed design data, operating characteristics and performance data, NPRDS provides an extensive data source for analysis of oper-ating experience. AE0D worked with INP0 in developing a list of critical com-ponents, called " key components," which are considered to have the greatest impact on safety system availability and the occurrence of plant transients. The NPRDS Trends and Patterns Analysis Program will focus on these key compo-4 nents. In addition to developing and testing the methods of analysis and nec-essary software, two preliminary tasks were undertaken in 1985 as a pre-requisite to beginning the full scale program in 1986: (1) A ranking is necessary of the key components using importance data from various PRA activities. First, functions considered most important for system safety were ranked, then the components were ranked according to their importance for the selected functions. From the two lists of ranked components developed for BWRs and PWRs, the key components to be analyzed each year will be selected in decreasing order of importance; (2) An estimation is necessary of the current failure rate of key components and a target or threshold failure rate (a preselected unacceptable failure rate) had to be established. The current failure rates were based on lit-erature search and converted into the proper units of reported failures per calendar year per piece of hardware. Based on the needs for probabilistic risk assessment and the precision of current knowledge of failure rates, threshold or alarm rates were tentatively selected which are a multiplier three times and ten times the current failure rate. The statistical analysis of the key components consists of four different methods using the time to failure, engineering information and failure counts in NPRDS: (1) Trend Analysis - to analyze the trend in the failure rate of key compo-nents over time and to detennine if the rate is increasing or decreasing, and the significance of the trend; j (2) The Page Procedure - to detect a shift in the failure. rate of a particular j key component above a preselected target or threshold failure rate; ) (3) Survival Analysis - to determine the factors (pedigree information) which appear to influence the failure pattern of key components and the failure rate of the component over the life of the component; and 1 (4) Reliability Modeling - to fit well-known reliability models to the failure patterns of key components; that is, to model the failure rate of the com-ponent over time, thus providing insight into failure causes and identify-s ing outliers (components whose failure pattern suggests more detailed engineering analysis). The results of this statistical analysis for each key component will be used by i AE0D engineers as a starting point for an engineering evaluation to develop findings, conclusions and recommendations for the selected key components. 77 .m

During 1985 AE0D, with contract assistance, tested these four statistical meth-ods on selected key components and found them to be both feasible and particu-larly useful for NPRDS data analysis. These preliminary tests, the ranking of the key components, and the estimates of the current and threshold failure rates are documented in two reports by INEL (" Preliminary Failure Time Analysis of Selected Nuclear Plant Reliability Data System Components" and " Preliminary Failure Time Analysis of Selected Nuclear Plant Reliability DC,a System Compo-nents, Phase Two"). 5.2.3 Conduct the NPRDS Evaluation Program and Coordinate NRC Guidance.for NPRDS As directed by the Commission, an NPRDS evaluation program continues to be im-plemented. Semiannual evaluation reports on NPRDS progress were forwarded in August 1985 (SECY-85-56A) and January 1986 (SECY-86-35). During the period covered by the first of these reports, INP0 and the utilities were involved in an extensive effort to upgrade the component population and engineering data-base. As a result of this effort, the first evaluation report found a decline in the number of NPRDS failure reports. With the completion of the database - upgrading, the second report noted a vigorous improvement in both the number of failure reports and the number of plants reporting to NPRDS. However, the timeliness of NPRDS failure reports and the quality declined compared to the same period in 1984. This may be due to the utilities reporting a large volume of failures backlogged during the rescoping effort. In general, AE0D now feels that NPRDS has improved to the point it warrants increased use as a source of reliability data to meet agency needs. In 1984 in a sworn affidavit, INP0 stated its position that NPRDS unit-specific data (plant or utility specific NPRDS data) submitted after January 1,1982 are confidential commercial information and requested that NRC withhold NPRDS unit-specific data from public disclosure. AE0D developed and coordinated with other NRC of fices guidance on the public availability of NPRDS data. The ED0 has reviewed the proposed guidance and in November 1985 determined that NPRDS unit-specific data confidential commercial information and therefore was exempt from public disclosure. Such data should only be released if an NRC Office Director determines that the public right to know the basis for a regulatory i decision warrants the release of the data. Normally, INP0 will be notified 30 days in advance of the data release in a letter signed by the Office Director. 5.2.4. Formulate Abnormal Occurrence (A0) Guidance and Prepare A0 Reports AE0D prepares the quarterly A0 Reports to Congress (as well as the associated Federal Register Notices) and, after coordination, forwards them to Congress via the Office of Congressional Affairs. These reports, issued as the NUREG-0090 series, serve as a feedback of significant event information to i Congress, government agencies, licensees, and the public. The reports are videly distributed (PDR, LPDRs, U.S. nuclear plants, Agreement States, and other government agencies) and are available individually or on a bscription basis through the GP0 sales program, i A Handbook on Application of the Abnormal Occurrence Criteria and Examples was prepared, reviewed by the cognizant offices, and subsequently issued on January 11, 1985 to the Office Directors and staff A0 Coordinators. The Handbook was prepared to help assure that events meeting the A0 threshold are a 78

o properly and consistently identified and recomended to the Comission for approval. The handbook provides clarifying guidance to the revised Manual Chapter 0212, describes the background and intent of the A0 criteria, and iden-tifies examples' of the types of events which have been previously reported un-der each A0 criterion. Four quarterly A0 Reports to Congress were issued during calendar year 1985 (third and fourth quarter CY 84, first and second ouarter CY 85). The four j reports described 23 A0s in NRC-licensed activities and seven A0s from Agree-i ment' States. The reports also described seven additional events as other re-ports of interest. Of the 23 A0s in NRC. licensed activities, ten occurred at nuclear power plants and 13 occurred at other licensees (e.g., fuel cycle fa-l cilities, industrial radiographers, medical institutions). The details of these 23 A0s were also published in the Federal Register. Six of the 23 events reported in 1985 occurred during CY 85 and 17 occurred during CY84. I Seven additional events at NRC-licensed activities were identified as possible. A0s for inclusion in the third quarter CY 85 Report to Congress. Four other i events were also' proposed for inclusion in this report as other reports nf in-terest. The document was forwarded to the Comission by SECY-86-38 on February 3, 1986. It was subsequently approved for publication during late February 1986. For the fourth quarter of CY 85, the staff has recommended five additional events at NRC-licensed activities to the Comission for approval. In addition, three other events were recommended for inclusion as other reports of interest. This report was submitted to the Comission for approval early during the j second quarter of CY 86. A sumary of the 1985 A0s is provided in Appendix A. This includes those ap-f proved by the Comission (as of late February 1986) and those still in the l approval chain. i 5.2.5 Issue Operational Data Feedback Reports, Including Bimonthly Power Reactor Events and Monthly LER Compilation Reports ~ In 1984, AE0D initiated a study to review licensees' programs and related NRC feedback documents for operational data review and assessment. The primary purposes of this study were to: (1) sample licensees' operating experience - feedback programs to determine if major deficiencies or problems were apparent; l (2) assess the utilitys' use of various feedback documents, including AE00 pub-i lications; and (3) detemine if changes were warranted in the NRC's opera i tional experience feedback program. Preliminary results of these assessments are discussed in Section 4. The final report will be issued in 1986. Power Reactor Events is a bimonthly newsletter intended to feed back operating experience information and lessons learned to ' personnel'at comercial nuclear power plants. These personnel include licensing' engineers, plant managers, and' training coordinators, as well as reactor operators and support personnel, Other recipients total approximately 1000 and include NRC employees; employees i of DOE and other Federal agencies; applicable State agencies;. various domestic and foreign-industry groups; and interested individuals. 79 -- u.

Each issue contains the following information: " Summaries of Events" provides detailed write-ups of events that may be significant because of their safety implications and/or because of opera-tor and licensee action taken during and after the events. " Excerpts of Selected Licensee Event Reports," added in June 1984, provides direct excerpts from LERs. In general, the information describes conditions or events that are unusual or complex, or that demonstrate a problem or condition that may not be obvious. " Abstracts of Other NRC Operating Experience Documents" provides abstracts and/or titles from pertinent NRC documents reflecting operational experience such as Abnormal Occurrence reports, IE Bulletins and Information Notices, AE0D case studies and engineering evaluations, NRR generic letters, and NRR operating reactor event memoranda. In CY 85, six issues of Power Reactor Events were published. They included 52 event sammaries and 131 LER excerpts. Consideration is being given to termi-nating this publication, since the study of licensee programs mentioned above indicated the document, while read and used by some licensees as source materi-al for training programs, was not resulting in widespread use by licensees to initiate voluntary actions in response to the lessons communicated. A notice to subscribers was included in the issue published in late December 1985. It addressed possible termination of the publication and requested user coments on the proposal. Twelve issues of monthly LER Compilation were published using data from the SCSS/ RECON databases. This compilation provides an abstract of each event sorted by facility name and chronologically by event date. Each event is also cross-referenced by system, component, component vendor, and general keyword indexes. 5.2.6 Prepare Reports on U.S. Events to the Nuclear Energy Agency's Incident Reporting System (NEA-IRS) and Provide Support to the International Atomic Energy Agency's Incident Reporting System (IAEA-IRS) l The U.S. continued to participate during the reporting period with 12 other i countrie's (Belgium, Canada, Finland, France,. Federal Republic of Germany, Ita- ") ly, Japan, Netherlands, Spain, Sweden, Switzerland, and United Kingdom) in a centralized Incident Reporting System (IRS) for exchanging information on oper-ational experience. This system ls operated by the Nuclear Enegry Agency (NEA) under the direction of the international Committee on the Safety of Nuclear Installations (CSNI). AE0D serves as the principal U.S. technical representative to Principal Working Group No.1, " Operating Experience and Human Factors." During this period, PWG-1 met in September 1985 with representatives from IAEA countries to discuss significa'nt operating events, and also met in September 1985 to evaluate the significance of operating events and data in NEA countries. A total of three technical presentations were made by the AE00 representative in these niectings. In addition, AE00 provided suggestions to both organizations regarding ways to obtain better consistency in the quality of reporting. 80

...-~ o 1 4 AE00' screens U.S. operating experience to select and prepare reports on those i events meeting pre-established reporting criteria. During this report period, AE0D prepared and submitted to NEA a total of 77 IRS reports on operational 1 events. In addition, AE0D prepared and submitted to NEA supplemental informa-tion on several U.S. events that were of particular interest to other IRS participants. 1 In 1985, the U.S. formalized its commitment to support the IAEA-IRS. The U.S. is fulfilling its commitment through_the NEA-IRS with NEA coordinating with t IAEA on the exchange of incident reports. AE00, as mentioned above, also pro-vided support in 1985 on significant operating events and in the program direc-tion area. 5.2.7 Develop Techniques to Apply PRA Perspectives-to the Screening and Analysis of Operational Events AE0D has a goal of using the results and methods of probabilistic risk assess-ment (PRA) in reactor operational event assessment. Because of the large number of LERs received, routine use of PRA for-prioritizing LERs may not be practical. Yet such techniques may offer promise to help assess and " calibrate" the significance of selected operating events and component failures. In 1985, AE0D assumed responsibility for the Accident Sequence Precursor (ASP) Program from RES. The objective of this program is to systematically evaluate operational data from U.S. nuclear power plants. The evaluation is-intended to uncover, assess, and rank potentially serious operational incidents. The in -- formation from this program is intended to serve as one element in NRC's as-- sessment of nuclear plant operational safety. The results and findings of this program for FY 1986 will be published as NUREG/CR reports indicating those events which were significant to plant safety over the time intervals 1984 and 1985. This program's quantitative estimate of the significance of individual events will also be used as an input to AE00's efforts to analyze specific cat-i egories or types of operational events such as reactor scrams. The preliminary analysis of 1985 scram data is included in the discussion of scrams in Section 4. i 5.2.8 Develop Guidance for the NRC's Program for Operational Data Review (Manual Chapter 0515) NRC-Manual Chapter 0515, covering the Safety Review of Operational Data, was i revised to update the Manual Chapter in terms of new organizational responsi-bilities that. occurred as a result of regionalization, and to clarify responsi-1 bilities related to the: (a) role and scope of operational safety data l j activities; (b) handling and resolution of potential generic issues; (c) coor-dination of activities; and (d) overall sequence of review activities. In.this period, staff reviews and concurrence actions were completed and the ~ revised Manual Chapter was issued for implementation in June 1985. i 4 l 81

i 5.2.9 Develop and Manage Databases AE00 utilizes a contract with the Nuclear Operations Analysis Center (NOAC) at Oak Ridge National Laboratory to assist in the data storage, retrieval, and evaluation efforts, as discussed below: (1) Sequence Coding and Search System (SCSS) The SCS5 was developed with two objectives: (1) encode all of the relevant technical information provided by the licensee in the LER, and (2) encode the information with sufficient " tags" so that the individual pieces can be precisely retrieved. The SCSS database currently contains data from 1981 to the present. During this period, approximately 3280 LERs were added to the data-base. This increases the number of LERs on the data base to al-most 17,890 LERs. In February 1985, NOAC issued a revision to the controlled copies of the SCSS User's Guide to reflect changes in codes used following the retrofit of the database in December 1984. An amendment to the con-trolled User's Guide describing new software capabilities was issued in August 1985. Amendments I and 2 to Revision 1 of the controlled copies of the SCSS-Coder's Manual were issued in April and July 1985, respectively. Formal documentation on SCSS was issued in April 1985 with the publication of the four volumes of NUREG/CR-3905, Sequence Coding and Search System for Licensee Event Reports. The four vol-umes include the User's Guide (Revision 1), the Code Listings, and the Coder's Manual (two volumes). The Quality Assurance Program for SCSS was finalized and published in April 1985. Amendment I to the Quality Assurance Manual was issued in September 1985. The SCSS Programmer's Manual was published in September 1985. This manual will be updated yearly to keep the program documentation current. Studies were performed to examine the feasibility and desirability of entering LER abstracts by means of an Optical Character Reader (0CR) rather than key punching. During 1985, the system design for routing the OCR-read abstracts to SCSS was finalized, acceptable type styles identified, an OCR upgrade initiated, and an Information Notice to licensees on requirements involving the type fonts and report quality was drafted. The Information Notice was distributed early in 1986. Software development was initiated to provide the requisite software and procedures to permit direct, interactive entry of SCSS coded LERs by coders. On-line coding is expected to be available by March 1986. 82

.o ~ Work continued during this period to backfit the SCSS data base to include LER data from 1980. This work will be completed in 1986. There were 11 requests for on-line access to the SCSS during 1985 from employees of the NRC and the DOE national laboratories. This made a total of 42 individuals who have direct access to the data base. AE0D and its contractor responded to over 180 requests for data out-put from SCSS from individuals without direct access. (2) Foreign Event File (FEF) In late 1985, NOAC developed the Foreign Event File (FEF) database and placed it (as a protected file) on the DOE computerized information system (RECON) to increase its availability to authorized users. The FEF file is similar to the abstract file of LERs that has also been maintained on RECON by the Nuclear Safety Information Center (NS.TC), which is part of NOAC. The scope of the Foreign Event File is limited to events at foreign light water reactors (LWRs) of U.S. design with an output greater than 200 MWe. During the period, 1575 events were added to the FEF database and ten events were identified as potentially safety significant. The poten-tially significant events were forwarded for further review by AE00, 4 NRR, RES, IE, and INP0. FEF currently contains information on over 6000 foreign events (950 with full abstracts). During 1985, the principal users of the system were NOAC, the NRC and INP0. The file was accessed 141 times. NOAC, in early 1985, issued a document for users that describes the database. It also later developed, documented, and implemented for-mal quality assurance procedures. In addition, NOAC reviewed foreign reporting and suggested methods for obtaining consistency in quality in the NEA-IRS and IAEA-IRS. (3) Part 21 and 50.55(e) File Operation of this database continued. During this period, additional training of NRC personnel was conducted in February 1985. Subsequently, responsibility for routine operation of this database was transferred from AE00 to IE. (4) RECON LER File An LER abstract and search capability is also maintained on the DOE RECON system, a computerized information retrieval system designed to provide users with remote terminal access to bibliographic databases. RECON access is available on-line to DOE offices, to institutions holding 00E contracts, to other Federal agencies with energy-related or energy intensive missions, and to State agencies with State-wide responsibilities 83

o for energy programs or information. Printouts from the NOAC database are also available to other organizations on a cost-recovery basis by request. From 1963 to 1983, the LER data for this data base was manually abstracted, and keywords were manually assigned. Once the SCSS became operational, the NSIC data was generated by the computer from the SCSS l database. Therefore, with minimal resources, the NSIC LER file continued to be available on-line'to the government agencies and contractors. ( Over.3370 LERs were added to the RECON database which now contains approximately 46,300 LERs received since 1963. I Of the more than 50 databases included in the RECON system, the LER database is typically in the top six with respect to the frequency of use. For example, for the last 3 months of 1985, the number of ac-cessions to this data base ranged from 71 to 115 per month. AE0D and its contractor responded to over 75 requests for data output from the RECON LER file. (5) Access INP0 Data Files Included in the Memorandum of Agreement with INP0 is a provision to share databases and results of analyses. As a result, AE0D maintains the capa-bility to conduct searches of selected INP0 data files such as NPRDS. During this period, AE0D worked with INP0 to resolve problems concerning the need for NPRDS output. These problems resulted from the reluctance of INP0 to provide access for additional NRC users, to train users on. the on-line NPRDS data base management system SEEK, and to perform NPRDS searches for the NRC. Thus the number of individuals outside of AE0D who were authorized to have access remained at seven. AE0D is continuing to work on the resolution of these problems and to reach agreement with INP0 on reasonable charges for continued NRC access to this INP0 supported data file. AE0D responded to about 55 requests for data output from NPRDS. (6) Formulate and Operate Interna ~, Management Information Systems AE00 maintains several internal databases for such purposes as documenta-l tion of screening results and resource accountability. An additional small database is maintained on reports of U.S. events to NEA/ IRS. Routine reports of LER screening results (WAMS) and manpower utilization (TACS) are prepared periodically and used by AE0D managers. (7). Other Ac,tivities PTB prepared or coordinated office procedures and prepared the office's s input to the NRC Annual Report to Congress. In addition, in 1985. PTB provided significant support to the agency in its defense of the lawsuit', Critical Mass Energy Project vs. NRC (INP0). Program support reports com-pleted by PTB are listed in Table 11. ^ 0 84 L.

Table 11 Program Support Reports Date Subject No. 7/85 Feedwater Transient Incidents in Westinghouse PWRs P501 6/85 Trends and Patterns Analysis of 1981 Through 1983 P502 LER Data. 8/85 Engineered Safety Features Actuations at Commercial P503 U.S. Nuclear Power Reactors January 1 through June 30, 1984 8/85 Trends and Patterns Report of Unplanned Reactor Trips P504 at, U.S. Light Water Reactors in 1984 12/85 Overview of Nuclear Power Plant Operations Experience Feedback Programs (Issued for Peer Review) 5.3 Nonreactor Assessment Staff (NAS) NAS screens and analyzes the operating experience associated with the activities and facilities licensed by the Office of Nuclear Material Safety and Safeguards (NMSS) and by Agreement States (i.e., nonreactor operating experience). In addition, NAS conducts studies from a human factors perspective on both reactor and nonreactor operating events. The major activity of NAS consists of: review of events reported to the five Regional Offices by licensees; a review of inspection reports; entry of coded information on events identified from the review into computer data files; evaluation of the events as a whole; and detailed reviews of specific events and concerns. The full scope of NAS responsibilities is shown below: P NONREACTOR ASSESSMENT STAFF KATHLEEN BLACK, CHIEF Responsibilities and Work Products 5.3.1 Screen individual events associated with NMSS and Agreement State. licensed activities and facilities and determine significance. 5.3.2 Analyze and evaluate individual nonreactor and medical misadministration events and related potentially generic safety concerns 5.3.3 Analyze reactor and nonreactor events from a human factors perspective 85

~ ~ ~ i j 5.3.4 Document independent technical assessments in: Case studies Engineering, Evaluations Technical Reviews Memoranda I 5.3.5 Develop, maintain, and provide updating data to these 4 computerized ~ data files: Nonreactor data file j Medical misadministration data file The milestones and progress on:these NAS responsibilities are summarized individually below. 5.3.1 Screen Individual Events Associated with NMSS and Agreement State -Licensed Activities and Facilities and Determine Significance About 6000 reports per year are received from the approximately 8300 NRC licensees. Of these reports, approximately 400 involved an operational event, i such as an overexposure, spill, or medical misadministration. In' addition, the 8000 Agreement State licensees are required to submit reports to the cognizant ] Agreement State which, in turn,-provides reports on certain of the reported i events to the NRC. These Agreement State reports are also reviewed principally 2 from the standpoint of generic problem identification and from their contribu- ) tion to the review of events reported by NRC licensees. As a result of these NAS screening activities, ~three'1evels of significance are ascribed to the individual events associated with NMSS and Agreement State li-l censed activities and facilities: (a) events that are determined to be A0s; (b) events that:are themselves not of high individual significance but are L events with potential generic significance; and (c) events that appear to have an immediate significance, either specifically or generically. l Of the many hundreds of event reports from NRC and Agre~ement State 4 licensees, 16 were determined by the Commission to meet the A0 reporting 4 criteria. Another.three events occurring in 1985 were recommended by AE00-as meeting the criteria but were not yet acted on by the Commission. Of 4 the 19, 12 of the events occurred in 1985 and seven occurred in 1984. Of these, seven were diagnostic misadministrations, four were therapeutic i misadministrations, and five were occupational. overexposures (four of' { j which were overexposures of radiographers 'or assistant'radiographers). There were three additional A0s: a lost nell logging source; deficiencies in management controls; and an unlawful possession of licensed material. 1 Of the seven diagnostic misadministrations, six involved the misadminis-tration of' iodine-131. The Office of Inspection and Enforcement (IE) is-sued an Information Notice on July 22, 1985 apprising licensees of four diagnostic misadministrations of iodine-131. Two diagnostic iodine events 4 occurred 4fter the issuance of the notice. 4 The four therapy misadministrations all resulted from human factors. ' AE00 published a case study on therapy misadministrations that' had occurred 86 4

s 4 4 i prior to 1985. A major conclusion of this case study--that independent j ' verification of dose calculations and patient chart reviews could reduce the number of therapy misadministrations--is further substantiated by U j _these A0s. p 5.3.2: Analyze and Evaluate Individual Nonreactor and Medical Misadministration Events and Related Potentially Generic Safety Concerns During the reporting period, one case study was completed and issued;.a second case study was issued for peer review; and work continued on a third.- A final case study report, AE0D/C505, ". Therapy Misadministrations Reported j to the NRC Pursuant to 10 CFR 35.42," was issued in December 1985. The j -study documents the analyses of 16 teletherapy and_ two brachytherapy j misadministrations reported to NRC during the period November 1980 through July 1984. The study also contains descriptive information on three addi-i tional. brachytherapy misadministrations and six -radiopharmaceutical thera-py misadministrations reported to NRC during the same period. l The primary conclusions of the study were that: (1) of the 16 teletherapy - j misadministratior.s reviewed in the ~ study,12 could have been prevented by - improved patient' chart reviews or in most cases by independent verifica-tion of patient dose calculations; (2) in addition, one misadministration i that involv'ed 53 patients over three years could have been' prevented by independent verification of the measurements made in calculating wedge filter correction factors; and (3) the brachytherapy misadministrations a 1 analyzed in the study could have been prevented by licensee personnel in-i dependently verifying that correct brachytherapy sources were loaded into the source applicators prior to, or' concurrent with, implanting the brachytherapy sources. Additionally, the report concluded that although professional. medical j groups involved with radiotherapy and related government agencies ' encourage quality assurance programs in radiotherapy facilities,'no government agency, or nongovernmental. accrediting body requires that radiotherapy facilities have quality ' assurance programs that are l consistent with recommendations of medical professional groups involved with radiation' therapy. The salient parts of the primary recommendations contained in the report are: That the Office of Nuclear Material. Safety and Safeguards (NMSS) j communicate the information contained in the report to affected l licensees. That NMSS contact appropriate professional organizations -to encourage 4 and support the initiation of a voluntary, industry-directed physical quality assurance program for radiotherapy facil_ities that includes-2 such things as: independent verification of patient dose calcula-l tions and independent verification of. the activity of brachytherapy sources. l 87

That if substantial progress has.not been made toward completion of the voluntary program by the end of two years, studies should begin as to whether a rulemaking is justified to require radiotherapy facilities to have quality assurance programs. That Part 10 CFR 35.21 be amended to include the calibration of beam modifiers such as wedge filters, shaping filters, trays, etc. 4 These recommendations are under review by NMSS. A preliminary case study report on the " Rupture of an Iodine-125 Brachy-therapy Source at the University of Cincinnati Medical Center" was published in December 1985. The study documents our analysis of an event at the University of Cincinnati that involved the rupture of a high activity iodine-125 seed used in a brachytherapy treatment protocol of brain tumors. The event involved the reuse of the seeds in treating several patients. The primary conclusions of the study are (1) that the i risk of an iodine-125 seed rupture is relatively high when.the sources are reused for several patients because the reuse require the seeds to be removed from old catheters and loaded into new catheters by cutting the catheters open with tools (razor blades, scissors, etc.) that could damage the seeds; and (2) that the consequence of the seed rupture at the University of Cincinnati involving patient and other personnel uptakes and facility contamination could have been mitigated by radiation surveys of the work area and the tools used to remove the seeds from the catheter, and/or by performing a leak test of the seeds. Comments are being assessed and a final report will be issued in 1986. Work continued on a study of medical licensees that had multiple misad-ministrations from November 1980 through December 1984. The data show that licensees had reported anywhere from one to 15 misadministrations 1 over the period. When data for a sample of licensees were transformed i into misadministration rates, it was found that all rates for the sample i set were low, less than or equal to.001. It is expected that the report will be issued for peer review in 1986. NAS (K. Black) served.as the Chair, Data Subcommittee, of the NRC Radiography Task Force. An event reporting form and a regulatory guide explaining its use are being developed by AE0D for publication in 1986. 5.3.3 Analyze Reactor and Nonreactor Events from a Human Factors Perspective During the time period, one case study was issued, and work continued in support of a case study issued in 1984. In December 1985, AE00 issued Case Study Report, AE00/C504, " Loss of i Safety System Function Events." The study identified 133 events involving a loss of safety system function (LSSF).in the 1981 to June 1984 time period, 87 of which (65%)'were the result of human factors contributions. Although the study found no evidence of any previously unrecognized safety problems, it did find that over the time period considered in the study there.is no clear evidence of improvement in preventing loss of safety i system events. The report recommends that IE consider prepar?ng an 88

Information Notice to industry regarding the characteristics of recent LSSF events resulting from human factors contributions. In addition, because it was found that equal numbers of events resulted from errors by licensed operators, non-licensed operators, and other personnel (maintenance personnel and I&C technicians), the report recommended that NRR consider this aspect in the various programs concerning NPP personnel qualification and training. An AE0D study, " Human Error in Events Involving Wrong Unit or Wrong Train," AE0D/S401, was published in January 1984. Because of the implications of wrong unit / wrong train events, NAS continued to track and analyze subsequent events. ~As a result,13 additional events were identified. AE0D initiated followup action by memorandum issued in August 1984 to NRR to again emphasize this subject. NRR connitted to assessing the corrective actions recommended by AE0D to reduce wrong unit / wrong train events in their Maintenance and Surveillance Program Plan (MSPP). AE0D has continued to monitor human errors in events involving the wrong unit, train, or system. AE00 identified 49 additional wrong unit / wrong train events that occurred from mid-1984 through 1985. An additional 41 events involving wrong component were identified. AE0D provided an updat-ing memorandum on these events to NRR early in 1986. Because wrong unit / wrong train events continued to occur, NRR and AE0D undertook a cooperative field survey of the problem during 1985. A series of site visits was undertaken to obtain first-hand knowledge of plant layout, equipment identification, and personnel practices. By the end of 1985, NRR and AE0D staff had completed site visits and reviews at ten multi-unit sites to obtain information on the factors that appeared to have contributed to these types of events. NRR is scheduled to issue ~ a report describing the results of this field survey early in 1986 for review by AE00 and other organizations. 5.3.4 Document Independent Technical Assessments Case Studies: AEOD/C504, " Loss of Safety System Function Events," dated December 30, 1985. AE00/C505, " Therapy Misadministrations Reported to the NRC Pursuant to 10 CFR 35.42," dated December 26, 1985. Technical Reviews: AE00/T511, " Technicians Perform Work on Wrong Control Pod Drive Mechanism" s (Susquehanna 1 Event), dated September 16, 1985. AE0D/T516, "HPCI Overspeed Trip Loss Events and Subsequent Damage Due to Water Hammer" (Pilgrim Event), dated December 12, 1985. 5.3.5 Develop, Maintain, and Provide Updating Data to the.Nonrea'ctor and Medical Misadministration Data Files 89

_.. ~. i [ Computerized data files continued to be updated by NAS on: (1)nonreactor l events; and (2) medical misadministrations. i From the events screened by NAS, approximately 200 nonreactor events and 400 misadministrations were coded by NAS and entered into the nonreactor-i data file during 1985. 4 j 5.4 Incident Investigation Staff l The Jncident Investigation Program.(IIP) was established by the'EDO, and ap-proved by the Connission to upgrade the NRC investigation of significant events j and assure that the. investigation is timely, thorough, coordinated and formally administered. The scope of the IIP includes the investigation of significant i operational events involving reactor and nonreactor activities licensed by the 4 NRC. The primary objective of the IIP.is to ensure that operational events are in-j vestigated in a systematic and technically sound manner to gather information pertaining to the causes of the events. including any NRC contributions or ] lapses, and to provide appropriate feedback regarding the lessons of experience { to the NRC, industry, and the public. i By focusing on causes of operating events and identification of' associated cor-rective actions, the results of the IIP process should improve nuclear safety by ensuring a complete technical and regulatory understanding of significant i events. The IIP has two investigatory responses based on the safety signifi-l l cance of the operational events. Both involve responses by an NRC team to i determine the circumstances and causes of an operational event. For a very j significant event, an Incident Investigation Team (IIT) is established by the i EDO to investigate the event in a manner similar to_ the _NRC response to the-Davis-Besse event. The responses to less significant operational events are i designated Augmented Investigation Teams (AITs) which involve a Regional-directed team complemented by Headquarters personnel, j A three member Incident Investigation Staff (IIS) was established as an i organizational element within AEOD following Commission approval of the Incident Investigation Program. The IIS is responsible for the development 1 and implementation of the IIP, as shown below. ( l l INCIDENT INVESTIGATION STAFF I j WAYNE D. LANNING, CHIEF i Responsibilities and Work Products - 5.4.1 Develop a program plan covering the development of the !l - Incident Investigation Program j 5.4.2 Develop formal guidance for the NRC Incident Investigation Program (NanualChapter0513) 5.4.3-Prepare and maintain personnel rosters'of candidate IIT leaders and members i 90 1' ?'er'--w e9 --fr-gg - pair -ge gd --pg-==a-+Mecuppo ur-t=-t='P+' pewgt*Y D- &wPN-TTi='PT$*P-Ps**y*'?-etc'~"%*-My Ty*Td-'T-(W-'i*'4-9*'W PW -*r*MD" e 8'4 P Mw t D *wt1+ p*,'1T**+1

• TP ft $ -P 9* P--

M#*'eie4 v y t r-gerrumptren Tth ga a t w t

• 5.4.4 Develop and provide training for IIT candidates 5.4.5 Develop procedures that govern and guide IITs 5.4.6 Provide and maintain administrative support for the IITs During the last 3 months of 1985, a number of significant milestones and actions were accomplished. These are summarized for each of the individual activities identified above.

5.4.1 Develop a Program Plan Covering the Development of the Incident Inves.tigation Program A program plan was developed to define and schedule the activities and planned accomplishments that &re required to implement the Incident Investigation Program. The program plan was circulated for comment among NRC offices and published as a final' report in December 1985. The goal called for in the program plan is to have all of the procedures, training and equipment, in place by July 31, 1986 to fully support an IIT for a reactor incident. Similar guidance will be developed for response to a nonreactor incident by March 31, 1987. The program plan defines other long-term activities that will be completed after FY 1986. 5.4.2 Develop Formal Guidance for the NRC Incident Investigation Program (Manual Chapter 0513) A draft of a new NRC Manual Chapter 0513, "NRC Incident Investigation Program," was prepared by the IIS. The draft Manual Chapter defines the objectives, authorities and responsibilities of the NRC offices, and establishes the basic requirements for the investigation of significant operational events involving reactor and nonreactor facilities licensed by the NRC. Because the IIP is an agency-wide program, the Manual Chapter defines the scope and functions of individual offices for investigatory responses by either an IIT or an AIT. Guidance is provided concerning the characteristics and types of operational events warranting an IIT or an AIT response. For each investi-gatory initiative, the Manual Chapter addressses the objective of the team; the scope of the investigation; the schedule; team composition; qualifications; procedural guidance; and follow-up actions. The draft Manual Chapter was circulated for review and comment in February 1986. It is expected that this review will be completed and the new Manual Chapter issued for implementation in 1986. 5.4.3 Prepare and Maintain Personnel Rosters af Candidate IIT Leaders and Members The IIS develops and maintains personnel rosters of candidate IIT leaders and members so that an IIT can be promptly established. The IIT leader roster includes only SES level personnel; the member roster includes senior NRC personnel available to serve as expert members on the IIT. The rosters may also be used to select team members for AITs. 91

A preliminary roster was developed of representative staff members who possessed the leadership and technical abilities to be IIT members. This roster was reviewed and revised, as appropriate, by Office Directors and Regional Administrators, and the roster of approved candidates will be issued in early 1986. Each candidate will be contacted by the IIT and a background statement will be prepared highlighting their areas of expertise and prior regulatory activities. From this roster, the first class of IIT candidates will be selected to receive about 2 staff weeks of specialized training in investigative techniques and methodology. These candidates will then be available to serve on an IIT, and if selected they will be relieved from existing duties for the duration of the investigation (about 6 weeks). 5.4.4 Develop and Provide Training for IIT Candidates The IIS, assisted by the Management Training and Development Staff (MTDS), is developing a training course for IIT leaders and members that will provide state-of-the-art training in incident investigation at reactor facilities. It is expected that the initial two week course will be given in July 1986. The 2-week course will be given by NRC. assisted by a contractor. The course includes: instructions pertaining to NRC investigations and the IIP; accident investigation techniques; lectures by accident investigators; and a simulated investigation of an incident at a reactor site. As part of developing the training plan, the IIS has audited and/or incorporated information from National Transportation Safety Board (NTSB) aviation investigation courses, Ontario Hydro Incident Investigation courses, and courses given by EG&G Services on incident investigations. 5.4.5 Develop and Compile Procedures that Govern and Guide IITs An IIT manual will be developed containing procedures that govern and guide the IIT in its investigation of a significant operational event. In general, these procedures are based on the knowledge and experience gained from the Davis-Besse, San Onofre and Rancho Seco IITs, and related experience from other organizations. For example, techniques employed by the NTSB staff in ineir investigation of transportation accidents are included as appropriate. During this period, a draft procedure for personnel interviews was completed. The following procedures are being developed: (1) Activating an IIT Response, (2) Conduct of Investigation, (3) Guidance for IIT Leader, (4) Records and Doccmentation, (5) Treatment of Quarantined Equipment, (6) Report Preparation,- and (7) Administrative Support. 5.4.6 Provide and Maintain Administrative Support for the IITs The IIS was established in the fall of 1985, and thus did not provide direct support to the IIT that investigated the June 1985 event at Davis-Besse. However, the IIS did support the other two events involving an IIT in 1985. 0n November 21, 1985, the San Onofre Nuclear Generating Station Unit 1 experi-enfed a partial loss of inplant ac electrical power while the plant was operating at 60% power. Following a manual reactor trip, the plant lost all 92

ac power for 9 minutes and experienced a severe incidence of water hammer in the feedwater system. This caused a leak, damaged plant equipment, and challenged the integrity of the plant's heat sink. Because of the potential safety complications of this event, the NRC Executive Director for Operations established an IIT. The Chief of the IIS was assigned as an IIT member to provide both technical and administrative support to the IIT. The IIT issued the results of its investigations in NUREG-1190 in January 1986. The most significant aspect of the event involved the failure of five safety-related check valves in the feedwater system whose failure occurred in less than a year, without detection, and jeopardized the integrity of safety systems. The underlying causes for the event identified issues that were plant specific and generic. The resolution of these issues are under study by the NRC. On December 26, 1985, Rancho Seco Nuclear Generating Station experienced a loss of de power within the integrated control system (ICS) while the plant was operating at 76% power. Following the loss of ICS de power, the reactor tripped automatically on high reactor coolant system (RCS) pressure followed by a rapid overcooling transient and automatic initiation of the safety features actuation system on low RCS pressure. The overcooling transient continued until ICS dc power was restored 26 minutes after its loss. The event involved a number of equipment failures, personnel errors, and a radiological release. An AIT was sent to the site on December 27 and, based on an initial investigatory effort, determined that the event was extremely complex and had potentially significant generic implications. Consequently, on December 30, the ED0 directed the investigation to be upgraded to an IIT and selected the Deputy Director of AE0D to be the IIT Leader. The results of the IIT activity were issued as NUREG 1195. 93

6. STATUS OF AE0D RECOMMENDATIONS This section summarizes the year-end status of all AE0D recommendations which are either new or still outstanding in the last report. The status of a total of 71 recommendations are provided in this section. These have been categorized (as of December 31,1985) as follows: Status of AE0D Recommendations Added Since Last Report 21 Resolved Since Last Report 7 Total Not Resolved 64 Currently Under Review 13 - Proceeding Satisfactorily 49 - Not Proceeding Satisfactorily 2 For tracking purposes, " Resolved Since Last Report" means that formal NRC action has been taken requesting licensee action. " Currently Under Review" means that the recommendation is being assessed by another NRC office and no position has yet been taken by the responsible program office. " Proceeding Satisfactorily" means that the responsible program office has agreed with the recommendation and has initiated appropriate action. "Not Proceeding 1 Satisfactorily" means that the recommendation was not accepted by the program office and additional discussion between the program office and AE0D is being i or will be held. At this time, there are no issues involving AE0D recom-mendations which would warrant ED0 attention. AE00's recommendation tracking i system ensures that all formal AE0D recommendations are tracked until resolution is achieved. In addition to formal recommendations which are tracked and included in this section, additional actions are routinely implemented by NRC program offices on AE0D suggestions contained in engineering evaluations and special reports. These AE0D suggestions are not formally tracked or closed out, and no formal response is required. Section 5.1 highlights a number of examples of where action was initiated based upon AE0D suggestions. 94

AE00 RECOMMENDATION TRACKING SYSTEM i i t RECOMMENDATION SOURCE: Case Study AEOD/C101 R2sponsible AEOD Engineer: T. Cintula i TITLE OR

SUBJECT:

"St. Lucie Natural Circulatf or Cooldown" i RECOMMENDATION 1 7 Provide a supply of cooling water to reactor coolant pump seals that will not be disabled by a single failure. i (Recommendation 4e) RESPONSIBLE CRGR , ER OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS NRR/DSR0/EIB J. Jackson High Not Scheduled Proceeding satisfactorily. ] Will be addressed as part of Generic Issue 65, " Component Cooling Water System Failures." Generic Issue 65 comprises task 2 j of Generic Issue 23, " Reactor 4 Coolant Pump Seal Failures" and will address the reliability i of RCP seal cooling systems. The program will determine what steps, i if any, NRC should take to increase the reliability of the component cooling water supply to the RCP i seals. We will continue to mnnitor this issue to assure that the recommendation is addressed. i j C-101-1 y i l

7 AE00 REC 0094ENDATION TRACKING SYSTEM REC 0f94ENDATION s SOURCE: Case Study AE00/C105 R:sponsible AE00 Engineer: T. Cintula TITLE OR

SUBJECT:

"Calvert Cliffs Unit 1 Loss of Service Water on May 20, 1980" REC 0094ENDATION 1 Installation of dual atmospheric dump valve capability for each steam generator on two-loop PWRs. (Study recomunendation 8(b)3) RESPONSIBLE CRGR OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS NRR/DSR0/RSIB A. Marchese High Not scheduled Proceeding satisfactorily. This . originally was to have been addressed by revision to SRP 15.6.3 E as part of Generic Issue 67.5.1, " Reassessment of Radiological Con-sequences Following a Postulated Steam Generator Tube Rupture." It has now been incorporated into USI A-45, " Shutdown Decay Heat Removal Requirements." RECOP98ENDATION 2 Review of steam generator tube rupture (SGTR) analyses for plants ifcensed prior to the SRP. IStudy recommendation 8(b)2) j RESPONSIBLE CRGR OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS NRR/DSR0/EIB A. Szukiewicz Medium Not scheduled Proceeding satisfactorily. Will be addressed as part of Generic Issue 67.5.2, "Re-evaluation of SGTR Design Basis Event." Tube ruptures are actively being investigated on a generic basis within USI A-47, " Safety Implications of Control Systems" and NTOL licensing issues. C-105-1 t

AEOD RECOMENDATON TRACKING SYSTEM RECOMENDATION ~ SOURCE: Case Study AE00/C105 (continued) R:sponsible AEOD Engineer: T. Cintula I TITLE OR

SUBJECT:

"Calvert Cliffs Unit 1 Loss of Service Water on May 20, 1980" i RECOM ENDATION 3 Revise SRP 9.2.2 to clarify isolation of nonsafety-related portions of service water system. 3 (Study recomendation 8(a)6) RESPONSIBLE CRGR OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS NRR/DSR0/SPEB L. Riani Nearly resolved. Not necessary Proceeding satisfactorily. Will be No value impact addressed in resolution of Generic e needed. Isst!e 36. Completion scheduled for June 1986. The SRP has been revised to address this recommendation and the reconmendation will be processed j as a "no new requirements" issue. l ) 4 1 I e C-105-2

AEOD RECOMMENDATION TRACKING SYSTEM RECOPMENDATION SOURCE: Case Study AEOD/CIO5 (continued) R;sponsible AEOD Engineer: T. Cintula TITLE OR

SUBJECT:

"Calvert Cliffs Unit 1 Loss of Service Water on May 20, 1980" RECOMMENDATION 5 IST of check valves in the instrument air system used to isolate safety-related portions of the system.

(Study recommendation 8(a)2) RESPONSIBLE CRGR l gg 0FFICE/DIV/BR CONTACT PRIORITY REVIEW STATUS NRR/DSR0/SPEB W. Milstead Low / drop None planned Not proceeding satisfactorily. This issue is to be addressed as part of a Generic Issue 43, " Contamination of Instrument Air Lines." AE00 does not agree with the prioritization given. NRR has agreed to review i the prioritization when the AE00 case study on Air System is issued. The AEOD case study is expected to be completed during 1986. i e C-105-3 S

s AEOD RECOMMENDATION TRACKING SYSTEM 1 RECOMMENDATION SOURCE: Memorandum to Harold Denton from C. J. Heltemes, dated May 2, 1983 l R2sponsible AE00 Engineer: T. Cintula i l TITLE OR

SUBJECT:

Response to NRR Comments on AE00 Report, "Calvert Cliffs Unit 1 Loss of Service l Water on May 20, 1980" l RECOMMENDATION 1 Accessibility of ADVs for local manual operations for RCS cooldown following a steam generator tube rupture. RESPONSIBLE CRGR 4 0FFICE/DIV/BR CONTACT PRIORITY REVIEW STATUS

g i

NRR/DSR0/RSIB A. Marchese High Not scheduled Proceeding satisfactorily. I Currently included in USI A-45. A site walkdown survey of nine plants substantiated the AE00 J concern of ADV accessibility. At some plants, the ADVs were readily accessible, while ADVs at other plants were difficult to open manually and may cause personnel j radiation exposure. i k l C-105-4

1 AE00 REC 00MENDATION TRACKING SYSTEM REC 009tENDATION SOURCE: Memorandum dated January 20, 1982 from C. Michelson to Harold R. Denton Responsible AE00 Engineer: M. Chiramal/F. Ashe 4 TITLE OR

SUBJECT:

" Safety Concerns Associated with Reactor Vessel Level Instrumentation in BWRs" j RECOMMENDATION 1 i Safety-related low-low reactor vessel level start of HPCI and RCIC systems should not be prevented or delayed by nonsafety-related high level trip. RESPONSIBLE CRGR OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS I8 l NRR/DSR0/RSIB .l. Joyce High Not scheduled Proceeding satisfactorily. Assigned j as Generic Issue 101. REC 009tENDATION 2 I Protective functions of narrow range level instrumentation must be assured in spite of adverse control system protection system interaction. RESPONSIBLE y CRGR OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS NRR/DSR0/EIB A. Szukiewicz High Not scheduled Proceeding satisfactorily. On-going USI-A47 and Generic Issue 101.

REFERENCE:

Memo dated March 19, 1982 from H. R. Denton to C. Michelson i j C-201-1 l i

AEOD RECOMMENDATION TRACKING SYSTEM RECOMMENDATION i SOURCE: Case Study AE0D/C202 j R2sponsible AEOD Engineer: T. Cintula TITLE OR

SUBJECT:

" Flow Blockage by Bivalve Mollusks at Arkansas Nuclear One and Brunswick" RECOMMENDATION I Capability to measure cooling water flow should be provided for all safety-related equipment. i RESPONSIBLE CRGR I 0FFICE/DIV/BR CONTACT PRIORITY REVIEW STATUS El l NRF/DSR0/EIB C. Hickey Medium Not scheduled

• Proceeding satisfactorily.

Pending generic action on biofouling. Generic Issue 51. RECOMMENDATION 2 Develop snd implement biofouling control strategies. RESP 0hSIBLE CRGR OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS 1 NRR/DSRO/EIB C. Hickey Medium Not scheduled

• Proceeding satisfactorily.

Pending generic action on biofouling. Generic Issue 51.

• May not be required.

If necessary, will be scheduled following completion of Generic Issue 51, " Proposed Requirements for Improving the Reliability of Open Cycle Service Water Systems." Generic Issue 51 is incorporating Task V of Project FIN B-2977 by RES. C-202-1

AE00 RECOMMENDATION TRACKING SYSTEM l i RECOMMENDATION l SOURCE: Case Study AE00/C202 (continued) l R;sponsible AE00 Engineer: T. Cintula l 1 TITLE OR

SUBJECT:

" Flow Blockage by Bivalve Mollusks at Arkansas Nuclear One and Brunswick" RECOMMENDATION 3 Periodic inspection of service water system piping.

RESPONSIBLE CRGR OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS l E; NRR/DSR0/EIB C. Hickey Medium Not scheduled

• Proceeding satisfactorily.

^2 Pending generic action on biofouling. Generic Issue 51. RECOMMENDATION 4 Periodic verification of overall heat transfer coefficient on multiple pass heat exchangers. t i RESPONSIBLE CPGR I 0FFICE/DIV/BR CONTACT PRIORITY REVIEW STATUS NRR/DSR0/EIB C. Hickey Medium Not scheduled

• Proceeding satisfactorily, j

Pending generic action on biofouling. Generic Issue 51. j

• May not be required.

If necessary, wi.ll be scheduled following completion of Generic Issue 51, " Proposed Requirements f r Improving the Reliability of Open Cycle Service Water Systems." Generic Issue 51 is incorporating Task V of Project FIN B-2977 by RES. i C-202-2 3 i 1

AEOD RECOMMENDATION TRACKING SYSTEM RECOMMENDATION SOURCE: Case Study AEOD/C202 (continued) Responsible AEDO Engineer: T. Cintula TITLE OR

SUBJECT:

" Flow Blockage by Bivalve Mollusks at Arkansas Nuclear One and Brunswick" RECOMMENDATION 5 Periodic verification of cooling water flow to all safety-related equipment should be specified in technical specifications. RESPONSIBLE CRGR {} OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS NRR/DSR0/EIB C. Hickey Medium Not scheduled

• Proceeding satisfactorily.

Pending generic action on f biofouling. Generic Issue 51. l I i

• May not be required.

If necessary, will be scheduled following completion of Generic Issue 51, " Proposed Requirements for Improving the Reliability of Open Cycle Service Water Systems." Generic Issue 51 is incorporating Task V of Project FIN B-2977 by RES. l C-202-3 1 i i

l AE0D RECOMMENDATION TRACKING SYSTEM RECOMMENDATION SOURCE: Case Study AE0D/C203 and Memorandum dated May 28, 1982 l R2sponsible AE00 Engineer: E. J. Brown l TITLE OR

SUBJECT:

" Survey of Valve Operator - Related Events Occurring During 1978, 1979, and 1980" (See also Recommendation 1 on page S-503-1)

RECOMMENDATION 1 Existing guidance to bypass thermal overload protective devices associated with safety-related valve motor operators should be reassessed. RESPONSIBLE CRGR l OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS l NRR/DSR0/EIB

0. Rothberg Medium Not scheduled Proceeding satisfactorily. This issue will be addressed in Generic SE l

Issue II.E.6, "In-Situ Testing of Valves," which is being evaluated under contract with BNL. As of mid-April 1986, the contract with BNL had been funded, approved and initiated. This issue will also be further evaluated by NRR during the review of the final AE00 Case Study on MOV performance, to be issued in early 1986. C-203-1 I

AEOD RECOMMENDATION TRACKING SYSTEM l RECOMMENDATION 2 Improved methods and procedures for the setting of torque switches should be developed and evaluated relative l to valve operability and functional qualification. RESPONSIBLE CRGR OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS NRR/DSR0/EIB

0. Rothberg Medium Not scheduled Proceeding satisfactorily. Based on a letter from the Director, NRR dated 2/23/83, an existing RES program was to be expanded to address this subject.

This has been included in RES contract i B3050, " Valve Performance Testing." RECOMMENDATION SOURCE: Case Study AE0D/C203 and Memorandum dated May 28, 1982 (continued) ffR:sponsibleAEODEngineer: E. J.- Brown TITLE OR

SUBJECT:

" Survey of Valve Operator - Related Events Occurring During 1978, 1979, and 1980" RECOMMENDATION 3 Signature tracing techniques should be developed and tried on selected motor-operated valves as part of the @fodic inservice testing program. RESPONSIBLE CRGR OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS NRR/DSR0/EIB

0. Rothberg Medium Not scheduled Proceeding satisfactorily. This will IE/DEPER/EGCB R. Kiessel High Not needed be included as part of a proposed draft plan (not yet approved) for l

Generic Issue II.E.6, "In-Situ Testing of Valves." It will be covered by research programs. A user request memorandum, dated 5/14/84, from the Director, NRR to the Director, RES addressed this item. Valve testing i has been completed and a final report was issued in January 1986, as NUREG/CR-4380. C-203-2

i l AE00 RECOMMENDATION TRACKING SYSTEM i RECOMMENDATION SOURCE: Case Study AEOD/C204 " San Onofre Unit 1 Loss of Salt Water Cooling Event on March 10, 1980," dated July 1982 Responsible AE00 Engineer: H. Ornstein i TITLE OR

SUBJECT:

" Single Failure Vulnerability of San Onofre l's Salt Water Cooling System" l

l RECOMMENDATION 1 Ongoing efforts of the SE) focus on single failure vulnerability and consequences for the salt water cooling i system and other equivalent service and cooling water systems. i RESPONSIBLE CRGR 0FFICE/DIV/BR CONTACT PRIORITY REVIEW STATUS c' NRR/DPLB/SEPB E. McKenna High Proceeding satisfactorily. NRR has reviewed SEP plants for such vulnera-bilities. Modifications have been made at SEP plants as appropriate. San Onofre has made several modifi-l cations and is performing a relia-bility analysis of the modified salt ) water cooling system. The analysis is scheduled for completion in May, 1986. I T C-204-1 j

r-AEOD RECOMMENDATION TRACKING SYSTEM RECOMMENDATION SOURCE: Memorandum, C. Michelson to Chairman Ahearne "New Unresolved Safety Issues" dated August 4, 1980, Memo, C. Michelson to H. Denton, " Resolution of Issue Concerning Steamline Break with Small LOCA," dated June 23, 1982, Case Study AEOD/C205, "ATOG as Applied to the April 1981 Overfill Event at ANO-1" Responsible AE00 Engineer: H. Ornstein TITLE OR

SUBJECT:

" Safety Implications of Steam Generator Transients and Accidents" 'i RECOMMENDATION 1 Combined primary / secondary side blowdown should be a USI for B&W plants. RESPONSIBLE CRGR OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS Es NRR/DPLB/RSB R. Jones High Proceeding satisfactarily. B&W RES B. Beckner licensees /EPRI/NRC are jointly funding a test facility to obtain integral systems test data to resolve the uncertainties associated with B&W plant response to SBLOCA and other transients and accidents. RECOMMENDATION 2 TAP-A47 should focus on equipment modifications or additions to preclude SG overfill as a credible event. RESPONSIBLE CRGR OFFICE /DIV/BR CONTACT PRIOPITY REVIEW STATUS NRR/DSR0/EIB A. Szukiewicz High Proceeding satisfactorily. On-going Generic Issue A-47. NUREG/CR-4385 entitled "Effect of Control System Failures on Transients. Accidents, i and Core Melt Frequencies at a Westinghouse PWR" has been published in support of A-47 resolution. C-205-1

AE00 RECOPMENDATION TRACKING SYSTEM RECOPMENDATION SOURCE: Case Study AEOD/C301 Responsible AE00 Engineer: M. Chiramal TITLE OR

SUBJECT:

" Failure of Class 1 E Safety-Pelated Switchgear Circuit Breakers to Close to Demand" RECC9FENDATION 1 Provide for monitoring the status of the closing circuit of Class I E Circuit Breakers and for appropriately selected breakers such as diesel generator output breakers, make the status indication available to the control room operator. RESPONSIBLE CRGR E OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS NRP/DSR0/SPEB K. Kniel Proceeding satisfactorily. NRR is re-prioritizino Generic Issue 55 based on Reference 3. RECOPMENDATON 2 In the short-term, ifcensees of operating reactors should establish regular local surveillance of Class 1 E switchgear circuit breakers to monitor the readiness status of the spring-charging motor of each unit. ( RESPONSIBLE CRGR OFFICE /DIV/8R CONTACT PRIORITY REVIEW STATUS MRR/DSR0/SFEB K. Kniel Proceeding satisfactorily. (See status of Recommendation 1.) C-301-1

AE00 REC 0009ENDATION TRACKING SYSTEM 4 REC 0fe9ENDATION SOURCE: Case Study AE00/C301 (continued) Responsible AE00 Engineer: M. Chiramal 4 TITLE OR

SUBJECT:

" Failure of Class 1 E Safety-Pelated Switchgear Circuit Breakers to Close on Demand" i

REC 0f0fENDATION 3 In addition to the above, measures that tend to preclude dirty or corroded contacts, poor electrical connections, blown control circuit fuses, and improper return of breakers to operable status should be incorporated into the maintenance procedures and used in actual maintenance practice. f RESPONSIBLE CRGR j 0FFICE/DIV/BR CONTACT PRIORITY REVIEW STATUS 1 NRR/DSR0/SPEB K. Kniel Proceeding satisfactorily. 4 j E (See status of Recommendation 1.) e REC 0mENDATION 4 Shift operating personnel should receive periodic training in the logic and operation of circuit breakers equipped with anti-pumping controls, i RESPONSIBLE CRGR j OFFICE /DIV/BR CONTACT PPIORITY REVIEW STATUS ^ j NPR/DSR0/SPEB K. Kniel y Proceeding satisfactorily. IE V. Thomas NA (See status of Recommendation 1.) i IE issued Inforination Notice 83-50 advising licensees of the problems. l This concern is being reconsidered in NRR re-prioritization of the issue. C-301-2 J 1 1

AEOD RECOMMENDATION TRACKING SYSTEM RECOMMENDATION SOURCE: Case Study AE00/C301 (continued) Responsible AE00 Engineer: M. Chiramal TITLE OR

SUBJECT:

" Failure of Class 1 E Safety-Related Switchgear Circuit Breakers to Close on Demand" i

REFERENCES:

1) Memo to C. H. Heltemes, Jr. from H. R. Denton, June 17, 1983 "AE00 April 1983 Report on Failure of Class 1 E Safety-Related Switchgear Circuit Breakers to Close on Demand

2) Memo to D. G. Eisenhut from R. L. Spessard, June 1, 1984, "Unmonitored Failures of Class 1 E Safety-Related Switchgear Circuit Breakers to Close on Demand" C

3) Memo to R. M. Bernero from H. R. Denton, March 27, 1985 " Scheduled for Resolving and Completing Generic Issue No. 55 - Failure of Class 1 E Safety-Related Switchgear Circuit Breakers to Close on Demand"

4) Memo to H. R. Denten from C. J. Heltemes, Jr., April 12, 1985 " Generic Issue No. 55 - Failure of Class 1 E Safety-Related Switchgear Circuit Breakers to Close on Demand"

5) Memo to C. J. Heltemes, Jr. from H. R. Denton, May 9,1985 "AE00 Concerns Regarding Generic Issue No. 55"

) e C-301-3 ~ l l

AE00 REC 0lWENDATION TRACKING SYSTEM s REC 0pWENDATION SOURCE: Case Study AE00/C401 and Pemorandum from C. J. Heltemes, Jr. to H. Denton, dated March 16, 1984 Responsible AE00 Engineer: D. Zukor TITLE 09

SUBJECT:

" Low Temperature Overpressure Events at Turkey Point Unit 4" l

REC 00MENDATION 1 Correct the LTOP technical specifications for the 5 areas identified in the report. RESPONSIBLE CRGR g 0FFICE/DIV/BR CONTACT PRIORITY REVIEW STATUS NRR/DSR0/RSIB Ed Throm High Proceeding satisfactorily. NRR has ~ identified this activity as new Generic Issue 94. NRR/DSR0 has revised the prioritization to "High" based on our memorandum dated 6/3/85. We agree with the current priority classification. J C-401-1

AEOD RECOMMENDATION TRACKING SYSTEM RECOMMENDATION SOURCE: Case Study AEOD/C402 Responsible AE00 Engineer: H. Ornstein TITLE OR

SUBJECT:

" Operating Experience Related to Moisture Intrusion in Electrical Equipment at Commercial Power Reactors" RECOMMENDATION 1 IE should revise the inspection program to ensure licensee adherence to NRC requirements.

RESPONSIBLE CRGR OFFICE /DIV/BR CONTACT PRIORI (Y REVIEW STATUS IE/DQASIP/ORPB S. McNeill None Proceeding satisfactorily. Criteria for inspection of electrical equip-ment in IE inspection modules 71707 and 71710 are presently undergoing revision in response to this recom-mendation. ) C-402-1

b i AE00 RECOMENDATION TRACKING SYSTEM I RECOMENDATION 50ilRCE: Case Study AE00/C403 Responsible AE00 Engineer: S. Rubin TITLE OR

SUBJECT:

"Edwin I. Hatch Unit 2 Plant Systems Interaction Event on August 25, 1982" RECOMENDATION 1 Evaluate the common mode failure potential of safety systems due to the harsh environment of breaks outside j

containment being back channelled through floor drain systems. RESPONSIBLE CRGR OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS NRR/DSR0/EIB D. Thatcher High Not scheduled Proceeding satisfactorily. This i recommendation is being evaluated 1 as part of Generic Issue No. 77, l C "Back Flow Protection in Common Equipment and Floor Drain Systems." Implementation of Generic Issue No. 77 is currently in progress with an i expected completion date of 8/86. RECOMMENDATION 2 i i Supplemental arrangements should be provided to assure timely isolation of the affected floor drain system j if the results of the above evaluation result in unacceptable common-mode safety system failures. RESPONSIBLE CRGR j OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS i NRR/DSR0/EIB D. Thatcher High Not scheduled Proceeding satisfactorily. This i recommendation is being evaluated i as part of Generic Issue No. 77, "Back Flow Protection in Common i Equipment and Floor Drain Systems." l Implementation of Generic Issue No. 77 is currently in progress with an expected completion date of 8/86. j S C-403-1 l

I AEOD RECOMMENDATION TRACKING SYSTEM REC 0099ENDATION SOURCE: Memorandum from C. J. Heltemes, Jr. to H. Denton, dated July 23, 1984 i Responsible AEOD Engineer: D. Zukor TITLE OR

SUBJECT:

" Steam Binding of Auxiliary Feedwater Pumps" 4

REC 0099ENDATION 1 PWR licensees should establish a method to regularly monitor the AFW system to minimize the potential for steam binding. RESPONSIBLE CRGR 0FFICE/DIV/BR CONTACT PRIORITY REVIEW STATUS NRR/DSR0/RSIB A. Spano High TBD IE/DEPER/EGCB M. Wagner Proceeding satisfactorily. Identified as new Generic Issue 93. Task Action Plan issued 1/22/85. IE Bulletin 85-01 has been issued. 4 k h i s C-404-1 i

AEOD RECOMMENDATION TRACKING SYSTEM l RECOMMENDATION SOURCE: Case Study AE00/C405 ] Responsible AEOD Engineer: S. Pettijohn TITLE OR

SUBJECT:

" Breaching of the Encapsulation of Sealed Well Logging Sources" 4

RECON 9ENDATION 1 Part 39 should require that well logging licensees have specific emergency procedures for handling specific source rupture incidents. RESPONSIBLE CRGR OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS RES/RAMBR M. L. Ernst Proceeding satisfactorily (with one possible exception). The proposed Part 39 has been revised based on i [j comments received during the "public conuent" period. A revised version of the regulation is now undergoing interoffice review. Most of this reconnendation has been incorporated in the revised version. The portion of the recommendation that specifies ...that licensees also have avail-able survey instrumentation capable ) of measuring radiation dose rate i levels of at least 100 mR/hr" has been changed in the revised version j to a requirement for survey instru-mentation capable of measuring at least 50 mR/hr. 4 C-405-1

4 AE00 RECOMMENDATION TRACKING SYSTEM i ! RECOMMENDATION SOURCE: Case Study AEOD/C405 Responsible AEOD Engineer: S. Pettijohn TITLE OR

SUBJECT:

" Breaching of the Encapsulation of Sealed Well Logging Sources" RECOMENDATION 1 (Continued) 1 We belive that licensees should have available on-site survey instrumentation capable of measuring radiation dose rate levels of at least 100 mr/hr. We believe that such instrumentation is required to enable licensees to meet the requirements of Part 20 relative to identifying for the purpose of establishing controls for high radiation areas: 10 CFR 20.203. m It should be noted that both our recommiendation and the proposed Part 39 contain two requirements for survey instruments--one a requirement for an instrument at temporary job sites (this i one is the one specified as capable of ~ measuring 50 mR/hr); the other require- ) ment is a generic one for an instrument (at the home office) capable of detecting radiation and contamination levels that t could be encountered during well logging l operations or in the event of an accident. This part of the proposed regulation I covers an instrument for a wide range of uses and is consistent with our recom-mendation. Our comments on the survey meter capability are under review by RES. j i C-405-2 I i

l AEOD REC 00MENDATION TRACKING SYSTEM RECOMMENDATION SOURCE: Case Study AEOD/C405 { Responsible AE00 Engineer: S. Pettijohn TITLE OR

SUBJECT:

" Breaching of the Encapsulation of Sealed Well Logging Sources" RECOMMENDATION 2 ) l Part 39 should require special physical characteristics for well logging sources, j i RESPONSIBLE CRGR OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS RES/RAMBR M. L. Ernst Proceeding satisfactorily. The 3 proposed Part 39 has been revised based on comments received during ] the "public comment" period. A revised version of the regulation j is now undergoing interoffice review. This recommendation has been incorporated in this revised version indicating that the i recommendation will likely be included in the final version of i the regulation. 1 l C-405-3 1

AEOD RECOMMENDATION TRACKING SYSTEM REC 0PetENDATION s. SOURCE: Case Study AEOD/C405 R2sponsible AE00 Engineer: S. Pettijohn TITLE OR

SUBJECT:

" Breaching of the Encapsulation of Scaled Well Logging Sources" RECOMMENDATION 3 Part 39 should preclude licensees from removing or attempting to remove sources from source holders without specific authorization in the license. RESPONSIBLE CRGR i 0FFICE/DIV/BR CONTACT PRIORITY REVIEW STATUS RES/RAMBR M. L. Ernst Proceeding satisfactorily. The l proposed Part 39 has been revised i based on comments received during Es the "public comment" period. A J revised version of the regulation is now undergoing interoffice review. 1 This recommendation has been i incorporated in this revised version indicating that the recommendation will likely be included in final version of the regulation. ~ 1 C-405-4 W

n i AE00 RECOMMENDATION TRACKING SYSTEM ~ RECOMMENDATION SOURCE: Case Study AE0D/C405 9 1 Responsible AE00 Engineer: S. Pettijohn TITLE OR

SUBJECT:

" Breaching of the Encapsulation of Sealed '4all Logging Sources" 4 RECOMMENDATION 4 A determination should be made whether or not extremity (finger) badges are required for personnel engaged in well logging activities. l RESPONSIBLE CRGR OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS 1 i NMSS/FCMC R. Cunningham Resolved. NMSS responded that while they would "not plan normally to require extremity monitoring for routine sealed source operations, because doses should be very low if u) 4 the licensee follows proper procedures l using remote handling tools". NMSS 3 did state that they intend "to evaluate source removal and maintenance on a case-by-case basis under the provisions of Section 39.43(d) of the proposed rule," and that "under that case-j by-case review, extremity monitoring would be required for any source removal or maintenance action where there is a likelihood of exceeding 25% of the dose values specified in Section 20.101(a), 10 CFR 20". 4 Since extremity exposures are most likely to occur during source removal and/or maintenance operations, we i believe this action adequately addresses the health and safety t issue of concern. We therefore consider this item closed. C-405-5

AEOD RECOMMENDATION TRACKING SYSTEM RECOMMENDATION SOURCE: Case Study AEOD/C405 Responsible AE00 Engineer: S. Pettijohn TITLE OR

SUBJECT:

" Breaching of the Encapsulation of Sealed Well Logging Sources" RECOMMENDATION 5 The interface between the well owner, well logging company, and companies specializing in recovery operations should be defined in order to establish the regulatory responsibility and authority over recovery and well logging operations. J RESPONSIBLE CRGR 0FFICE/DIV/BR

CONTACT, PRIORITY; REVIEW STATUS RES/RAMBR M. D. Ernsts Proceeding satisfactorily. The

,,4 tz, N proposed Part 39 has been revised 's based on comments received during ~ 3 the "public comment" period. A ~ i revised version of the regulation is now undergoing interoffice review. This recommendation has ll's. been incorporated in this revised 3' version indicatino that the recom- ,s mendation will likely be included in + i the final version of the regulation. s o ) ( C-405-6 4

1 AEOD RECOMMEND 4 TION TRACKING SYSTEM RECOMMENDATION SOURCE: Case Study AEOD/C501 (NUREG/CR-3551) R2sponsible AE0D Engineer: H. Ornstein TITLE OR

SUBJECT:

" Safety Implications Associated with In-Plant Pressurized Gas Storage and Distribution Systems in Nuclear Power Plants" RECOMMENDATION 1 Require procedures to provide protection to prevent unacceptable damage to safety-related equipment from portable gas cylinder missiles.

4 RESPONSIBLE CRGR Of 0FFICE/DIV/BR CONTACT PRIORITY REVIEW STATUS 4 IE/EPER D. Kirkpatrick Resolved. Reviewed by IE and NRR. NRR/DSR0/SPEB K. Kniel No action at this time. IE may consider an information notice if "pressarired gas related events" do occur in the future. NRR believes that SRI 3.5.1.1 and 3.5.1.2 are adequate, and that no NRR action is necessary. 1 l C-501-1 4 s

= AEOD RECOMENDATION TRACKING SYSTEM RECOMENDATION SOURCE: Case Study AEOD/C501 (NUREG/CR-3551) (continued) Responsible AE00 Engineer: H. Ornstein TITLE OR

SUBJECT:

" Safety Implications Associated with In-Plant Pressurized Gas Storage and Distribution Systems in Nuclear Power Plants" RECOMMENDATION 2 Require protection to prevent hydrogen explosions or fires in areas containing or impacting operation of safety related equipment.

lh RESPONSIBLE CRGR 0FFICE/DIV/BR CONTACT PRIORITY REVIEW STATUS l-NRR/DSR0/SPEB K. Kniel Proceedino satisfactorily. SPEB is presently prioritizing the matter of backfitting requiring the installation of excess-flow valves on hydrogen lines in operating plants. (Safety issue No. 106 " Piping and the Use of Highly s Combustible Gases in Vital Areas.") I t C-501-2 i i

1 AE00 RECOMMENDATION TRACKING SYSTEM RECOMMENDATION SOURCE: Case Study AE00/C501 (NUREG/CR-3551) (continued) R:sponsible AEOD Engineer: H. Ornstein ' TITLE OR

SUBJECT:

" Safety Implications Associated with In-Plant Pressurized Gas Storage and Distribution Systems in l Nuclear Power Plants" RECOMMENDATION 3 Identification of lines and tanks. i } OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS NRR/DSR0/SPEB K. Kniel Drop Resolved. No NRR action planned. NRR believes that without a risk i assessment demonstrating how the identification of lines and tanks will significantly affect core acit frequencies and/or radiological releases, the NRC would not have the authority to require such identifi-cation. 1 1 1 O e k C-501-3

AEOD RECOMENDATION TRACKING SYSTEM RECOMENDATION SOURCE: Case Study AE0D/C50? Rssponsible AEOD Engineer: Peter Lam TITLE OR

SUBJECT:

"Overpressurization of Emergency Core Cooling Systems in Boiling Water Reactors" RECOMMENDATION 1 Disable the nonsafety-related air operator associated with the testable isolation check valve on the injection line in the emergency core cooling systems. RESPONSIBLE CRGR y 0FFICE/DIV/BR CONTACT PRIORITY REVIEW STATUS NRR/DSR0/RSIB H. Woods High 1986 Proceeding satisfactorily. Assigned as new Generic Issue 105 with an expedited schedule of resolution. RECOMENDATION 2 Perform leakage testing of the testable isolation check valve prior to plant startup after each refueling outage or following maintenance, repair or replacement of the valve. RESPONSIBLE CRGR OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS i NRR/DRSO/RSIB H. Woods High 1986 Proceeding satisfactorily. Assigned as new Generic Issue 105 with an expedited schedule of resolution. C-502-1 O

l ~ j AEOD RECOMMENDATION TRACKING SYSTEM RECOMMENDATION SOURCE: Case Study AE00/C502 R2sponsible AE00 Engineer: Peter Lam TITLE OR

SUBJECT:

'0verpressurization of Emergency Core Cooling Systems in Boiling Water Reactors" RECOMMENDATION 3 Reduce human errors in maintenance and surveillance testing activities.

RESPONSIBLE CRGR OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS bh NRR/DSR0/RSIB H. Woods High 1986 Proceeding satisfactorily. Assigned as new Generic Issue 105 with an expedited schedule of resolution. RECOMMENDATION 4 -Study reducing the frequency of surveillance testing of the isolation barriers of the emergency core cooling systems during power operation. 4 RESPONSIBLE i CRGR OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS 1 NRR/DSR0/RSIB H. Woods High 1986 Proceeding satisfactorily. Assigned 4 as new Generic Issue 105 with an expedited schedule of resolution. C-502-2 2 I a

.,,,,,,,,,,,,,,,,o>< >****'E**"'~~' AEOD RECOP99ENDATION TRACKING SYSTEM RECOMMENDATION SOURCE: Case Study AE0D/C503 R;sponsible AE00 Engineer: H. Ornstein TITLE OR

SUBJECT:

" Decay Heat Removal Problems at U.S. PWRs" i

RECOMMENDATION 1 NRR assess the need for NRC requirements to improve planning, coordination, procedures, and personnel training during shutdown to ensure the availability of the DHR system. ! R' RESPONSIBLE CRGR f"

• y'FICE/DIV/BR CONTACT PRIORITY REVIEW STATUS f

NRR/DSR0/RSIB A. Spano High Currently under review. NRR has stated that this recommendation will i be considered in the resolution of Generic Issue 99. RECOMMENDATION 2 1 NRR require PWR licensees to have a reliable method of measuring and monitoring reactor vessel level during j shutdown modes of operation and corresponding; technical specification requirements for operability. RESPONSIBLE CRGR OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS NRR/DS'O/RSIB A. Spano High Currently under review. NRR has R stated that this reconmendation will be considered in the resolution of Generic Issue 99. i C-503-1 4

... _.. _ _. _ _ _.. _. _ - _ - _.. ~ _ - - 1 AEOD REC 0PMENDATION TRACKING SYSTEM RECOMMENDATION l SOURCE: Case Study AE00/C503 (continued) i Rasponsible AE00 Engineer: H. Ornstein TITLE OR

SUBJECT:

" Decay Heat Removal Problems at U.S. PWRs" RECOMMENDATION 3 NRR require licensees to improve the man-machine interfaces related to DHR operation. RESPONSIBLE CRGR OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS 'b NRR/DSR0/RSIB A. Spano High Currently under review. NRR has stated that this recomendation will be considered in the resolution of Generic Issue 99. RECOMMENDATION 4 NRR should consider DHR suction bypass lines as alternatives to redundant drop lines (if A-45 concludes that single drop line configurations are unacceptable). RESPONSIBLE CRGR OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS NRR/DSR0/RSIB A. Spano High Currently under review. NRR has stated that this recommendation will be considered in the resolution of Generic Issue 99. C-503-2 i

AE00 RECOMMENDATION TRACKING SYSTEM RECOMMENDATION i SOURCE: Case Study AE0D/C503 (continued) Responsible AE00 Engineer: H. Ornstein TITLE OR

SUBJECT:

" Decay Heat Removal Problems at U.S. PWRs" RECOMMENDATION 5 NRR consider removal of autoclosure interlocks to minimize loss-of-DHR events. + RESPONSIBLE CRGR 0FFICE/DIV/BR CONTACT PRIORITY REVIEW STATUS NRR/DSR0/RSIB A. Spano High Currently under review. NRR has stated that this recommendation will be considered in the resolution of Generic Issue 99. RECOMMENDATION 6 NRR should address the issue of DHR system redundancy to ensure that the DHR system is available during Mode 4, and the early stages of Mode 5. RESPONSIBLE CRGR OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS NRR/,DSR0/RSIB A. Spano High Currently under review. NRR has stated that this recommendation will be considered in the resolution of Generic Issue 99. C-503-3 I e i

---

_-x-

AE00 RECOMMENDATION TRACKING SYSTEM RECOMMENDATION SOURCE: Case, Study AE00/C504 R5sponsible AE00 Engineer: E. Trager TITLE OR

SUBJECT:

" Loss of Safety System Function Events" RECOMMENDATION 1 IE issue an Information Notice to feed back the results of the case study to industry. RESPONSIBLE CRGR OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS IE Currently under review. The report was issued in December 1985. M e C-504-1 i

AE00 RECOMMENDATION TRACKING SYSTEM RECOMMENDATION SOURCE: Case Study AECD/C504 Responsible AE00 Engineer: E. Trager TITLE OR

SUBJECT:

" Loss of Safety System Function Events" RECOMMENDATION 2 NRR review the Maintenance and Surveillance Program Plan, the Human Factors Program Plan, and the INP0 training accreditation program to ensure the adequacy of training programs for all types of NPP personnel. RESPONSIBLE CRGR OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS NRR Currently under review. The report was issued in December 1985. U 1 i C-504-2

~ AE00 RECOMMENDATION TRACKING SYSTEM . RECOMMENDATION SOURCE: Case Study AE00/C504 R;sponsible AEOD Engineer: E. Trager TITLE OR

SUBJECT:

" Loss of Safety System Function Events" RECOMMENDATION 3 AEOD determine whether or not to perform further evaluations of losses of ECCS injection systems events, containment spray isolation events, and the Salem loss of component cooling water event. RESPONSIBLE CRGR OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS AE00 Currently under review. The report was issued in December 1985. C. d J C-504-3 4

AEOD RECOMMENDATION TRACKING SYSTEM RECOMMENDATION SOURCE: Case Study AE00/C505 R;sponsible AE00 Engineer: S. Pettijohn TITLE OR

SUBJECT:

" Therapy Misadministrations Reported to the NRC Pursuant to 10 CFR 35.42" RECOMMENDATION 1 The Office of Nuclear Material Safety and Safeguards should communicate the information contained in this report to the affected licensees. RESPONSIBLE CRGR OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS NMSS/FCPC R. Cunningham Currently under review. The case study report was trawsmitted to MMSS for action on 12/26/85. M i e 9 C-505-1 i

AE00 RECOMENDATION TPACKING SYSTEM RECOMENDAT10N SOURCE: Case' Study AEOD/C505 R:sponsible AEOD Engineer: S. Pettijohn TITLE OR

SUBJECT:

" Therapy Misadministrations Reported to the NRC Pursuant to 10 CFR 35.42" ] i RECOMENDATION 2 The Office of Nuclear Material Safety and Safeguards should consider the following j actions in regard to establishing quality assurance requirements for radiotherapy facilities licensed by NRC: Contact appropriate professional organizations to encourage and support the initiation of a voluntary, industry-directed physical 4 lity assurance program for radiotherapy facilities. We believe that the et mitment of the profession organizations in this regard should be assessed by the NRC and a conclusion reached as to the effectiveness of the voluntary program within C two years. w If substantial progress toward completion of the voluntary program, including i a final completion date, has not been demonstrated at the end of two years, we recommend that NMSS initiate the necessary studies to determine whether a j rulemaking is justified to require that radiotherapy facilities licensed by NRC have quality assurance programs to insure the accuracy of patient doses. The program should include such things as: independent verification of patient dose calculations and independent verification of the activity of brachytherapy sources before the sources are implanted. i The voluntary quality assurance program should contain at least the elements outlined above. j RESP 6MSIBLE CRGR 0FFICE/DIV/BR CONTACT PRIORITY REVIEW STATUS 1 NMSS/FCMC R. Cunningham Currently under review. The case study report was transmitted to MMSS for action on 12/26/85. i C-505-2 i

AE00 RECOMENDATION TRACKING SYSTEM RECOMMENDATION SOURCE: Case Study AEOD/C505 R2sponsible AEOD Engineer: S. Pettijohn TITLE OR

SUBJECT:

" Therapy Misadministrations Reported to the NRC Pursuant to 10 CFR 35.42" RECOMMENDATION 3 10 CFR Part 35.21 should be amended to include the calibration of beam modifiers such as wedge filters, shaping filters, trays, etc. RESPONSIBLE CRGR 0FFICE/DIV/BR CONTACT PRIORITY REVIEW STATUS NMSS/FCMC R. Cunningham Currently under review. The case study report was transmitted to MMSS y; for action on 12/26/85. RECOMMENDATION 4 In addition, to the extent that the NRC implements recommendation 3, the action should be made an item of compatibility for Agreement States. RESPONSIBLE CRGR OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS SP W. Kerr y Currently under review. The case study report was transmitted to SP for action on 12/26/85. C-505-3 i l

AEOD REC 0tmENDATION TRACKING SYSTEM RECOPMENDATION SOURCE: Memo: C. Michelson to R. Mattson, "NRC Action Plan Developed as a Result of TMI-2 Accident - Draft 3, Task II.E.3 Decay Heat Removal," April 24, 1980 R:sponsible AE00 Engineer: H. Ornstein TITLE OR

SUBJECT:

" Reliability of DHR Systems" RECOMMENDATION 1 Reliability of DHR systems should be reviewed and where necessary upgraded on an expedited basis.

RESPONSIBLE CRGR OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS NRR/DSR0/RSIB A. Marchese High Proceeding satisfactorily. TAP A-45 Contractor perfoming work which will address this issue - NUREG report expected July 1986. E-001A-1

AEOD RECOPMENDATION TRACKING SYSTEM REC 0pWENDATION SOURCE: Memorandum dated July 15, 1980 from C. Michelson to H. R. Denton R;sponsible AE0D Engineer: M. Chiramal TITLE OR

SUBJECT:

" Operational Restrictions for Class 1 E 120 VAC Vital Instrument Buses" REC 0pWENDATION 1 Impose Technical Specification requirements concerning operational restrictions for Class 1 E 120 Vac Vital Buses on operating plants. RESPONSIBLE CRGR W OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS NRR/DSR0/EIB D. Thatcher High Not scheduled Proceeding satisfactorily. Currently being pursued under Generic Issue 48. i i i i l 4

REFERENCES:

1. Memo dated September 29, 1980 from H. Denton to C. Michelson 2. Memo dated October 25, 1983 from H. Denton to D. Eisenhut t E-005-1 k

AEOD RECOMMENDATION TRACKING SYSTEM RECOMMENDATION SOURCE: Memo dated August 27, 1980 from C. Michelson to H. R. Denton R;sponsible AEOD Engineer: M. Chiramal TITLE OR

SUBJECT:

" Tie Breaker Between Redundant Class 1 E Buses - Point Beach Nuclear Plant Units 1 & ?" RECOMMENDATION 1 Interconnection between redundant safety-related electrical load groups should comply with requirements of Regulatory Guide 1.6. RESPONSIBLE CRGR OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS NRR/DSR0/EIB D. Thatcher Medium Not scheduled Proceeding satisfactorily. Generic Issue 49 has been established for this issue. l \\

REFERENCE:

Memo dated October 16, 1980 from H. R. Denton to C. Michelson E-010-1

l, AE00 RECOMENDATION TRACKING SYSTEM REC 0mENDATION SOURCE: Memorandum to H. R. Denton from C. Michelson, dated January 19, 1981 R;sponsible AEOD Engineer: E. J. Brown TITLE OR

SUBJECT:

" Degradation of Internal Appurtenances in LWPs" RECOMENDATION 1 i j RESPONSIBLE CRGR i 0FFICE/DIV/BR CONTACT PRIORITY REVIEW STATUS i NRR/DSR0/SPEB H. Vander Molen Draft priority Not scheduled Proceeding satisfactorily. Agree-g was detemined ment between NRR and AE00 to include to be low. evaluation of loose parts as possible (Generic Issue 35) missiles during SG blowdowns has been i reached. Additionally, it was agreed that AE00 would conduct a data search to identify loose parts affecting ESF system reliability to assi:;t NRR in evaluating ESF system concerns (Reference). Results of the data ) search will be compiled and forwarded to NRR.

REFERENCE:

Memo, K. V. Seyfrit to W. Minners, February 15, 1985, " Evaluation of of Generic Issue No. 35, Degradation of Internal Appurtenance in LWRs" E-101-1 4

AEOD RECOM4ENDATION TRACKING SYSTEM i REC 0pe9ENDATION SOURCE: Memorandum from C. Michelson to V. Stello and H. Denton, "Immediate Action Memo: Connon Cause Failure Potential at Rancho Seco - Dessicant Contamination of Air Lines," September 15, 1981. ) R::sponsible AE00 Engineer: H. Ornstein 1 TITLE OR

SUBJECT:

" Plant Air Systems" ) 1 RECOMMENDATION 1 Obtain licensees' experience and assessment of this problem and determine course of corrective action if required. RESPONSIBLE CRGR i 0FFICE/DIV/BR CONTACT PRIORITY REVIEW STATUS -M j NRR/DSR0/SPEB W. Milstead Low / Drop Not proceeding satisfactorily. NRR prioritized this issue (Generic Issue 43) and reconmended that it be dropped. AE00 does not agree 4 with the prioritization, and sent NRR a memo (Reference) requesting NRR I hold this issue in abeyance until an AE00 report is written on this subject j including additional operational i experience. The AE00 case study is l expected to be completed during 1986. j NRR has agreed to this approach. A recent ACRS memo related to NRR's j prioritization of generic issues elso i indicated agreement with this approach.

REFERENCE:

Memo from C. J. Heltemes to H. R. Denton, " Contamination of Instrument Air Lines," December 14, 1983 E-123-1 i l

AE00 REC 00mENDATION TRACKING SYSTEM i REC 0pmENDATION SOURCE: Memorandum from C. Michelson to R. Burnett, dated January 12, 1982 R3sponsible AE00 Engineer: W. Lanning t i TITLE OR

SUBJECT:

" Methodology for Vital Area Determination" RECOMMENDATION 1 I

Improve vital area determination methodology by ensuring the completeness and validity of the generic fault trees and guidelines used to help define " key vital areas." i RESPONSIBLE CRGR j 0FFICE/DIV/BR CONTACT PRIORITY REVIEW STATUS l 1 l NMSS/SG/SGPR R. Dube High None Resolved. In the spring of 1985, the ED0 established an interoffice couaittee to review the vital area j assumptions. This effort should be completed by early 1986. 1 i i l i } I i I E-201-1 h

O AE00 RECOMENDATION TRACKING SYSTEM RECOMMENDATION SOURCE: Memorandum from C. Michelson to R. Burnett, dated January 12, 1982 (continued) Responsible AE0D Engineer: W. Lanning TITLE OR

SUBJECT:

" Methodology for Vital Area Determination" RECOMMENDATION 2 Additional resources should be allocated for developing and evaluating practical methods, other than access controls, to minimize insider threats and this activity should receive budgetary priority.

RESPONSIBLE CRGR }{ OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS .) NMSS/SG/SGPR J. Yardumfan High None Resolved. A-29, " Nuclear Power Plant Design for Reduction of Vulnerability to Sabotage" has been resolved and NUREG/CR-4392 has been issued as its resolution. CRGR has approved the " Insider Rule" and it has been transmitted to the Commission by the EDO. E-201-2 1

AEOD RECOMMENDATION TRACKING SYSTEM RECOMMENDATION SOURCE: Memorandum from C. Michelson to R. Vollmer and R. Mattson dated February 24, 1982 Responsible AEOD Engineer: M. Chiramal TITLE OR

SUBJECT:

" Spurious Trip of the Generator Lockout Relay Associated with a Diesel Generator Unit" RECOMMENDATION 1 Should explicitly verify that seismic qualification of all protective devices used in the control and protection circuitry of DG units has been performed with these devices in their energized, de-energized, tripped and non-tripped states. >d r $5 RESPONSIBLE CRGR OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS NRR/DSRO/EIB T. Chang High Not scheduled Proceeding satisfactorily. Issue will be incorporated into 3 USI A-46, 4 i i

REFERENCE:

Memorandum from H. R. Denton to C. Michelson dated May 11, 1982 i i E-212-1 i

i AEOD RECOMMENDATION TRACKING SYSTEM RECOPMENDATION SOURCE: Engineering Evaluation AE00/E215 R;sponsible AE00 Engineer: T. Cintula l TITLE OR

SUBJECT:

" Salt Water System Flow Blockag,e at Pilgrim NPS by Blue Mussels" i RECOMMENDATION 1 Internal inspection of RBCCW HX supply headers RESPONSIBLE CRGR 4' 0FFICE/DIV/BR CONTACT PRIORITY REVIEW STATUS NRR/DSR0/EIB C. Hickey Medium Not scheduled

• Proceeding satisfactorily. Pending generic action on biofouling.

Generic issue 51. RECOPMENDATION 2 Periodic measurement of overall heat transfer coefficient on RBCCW HXs at Pilgrim RESPONSIBLE CRGR OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS NRR/DSR0/EIB C. Hickey Medium Not scheduled

• Proceeding satisfactorily. Pending generic action on biofouling.

Generic Issue 51. REC 0pmENDATION 3 Periodic measurement of Salt Water System flow to RBCCW HXs RESPONSIBLE CRGR OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS NRR/DSR0/EIB C. Hickey Medium Not scheduled

• Proceeding satisfactorily. Pending generic action on biofouling.

Generic Issue 51.

• May not be required.

If necessary, will be scheduled following completion of Generic Issue 51, " Proposed Requirement fer Improving the Reliability of Open Cycle Service Water Systems." The generic issue is incorporating Task V of Project FIN B-2977 by RES. 2 E-215-1 2 i 1

i AE00 RECOMMENDATION TRACKING SYSTEM i RECOMENDATION SOURCE: Memo dated June I, 1982 from C. Michelson to H. R. Denton and R. C. DeYoung i R:sponsible AE00 Engineer: M. Chiramal TITLE OR

SUBJECT:

" Degradation of BWR Scram Pilot Solenoid Valves due to Abnonnal Power Supply Voltages" RECOMMENDATION 1 Implementation of RPS power monitoring system in BWRs should consider ac line losses and voltage drops between RPS power supplies and loads connected to them.

4 RESPONSIBLE CRGR l '0FFICE/DIV/BR CONTACT PRIORITY REVIEW STATUS NRR/DBWRL/EICSB M. Srinivasan High Not required Resolved. Being addressed under MPA C-11 "RPS Power Supplies (BWRs)." Review has been completed for all operating BWR units.

REFERENCE:

Memo dated August 13, 1982 from H. Denton to C. Michelson E-225-1 \\

AEOD RECOMMENDATION TRACKING SYSTEM ~ RECOMENDATION SOURCE: Memo dated June 24, 1982 from C. J. Heltemes, Jr. to E. L. Jordan RIsponsible AE00 Engineer: F. Ashe TITLE OR

SUBJECT:

" Failure of ESF Manual Initiation Pushbutton Switches" RECOMMENDATION 1 Licensees should be made aware of potential problems associated with Cutler Hammer switch assemblies used in low i voltage - low current applications. RESPONSIBLE CRGR l { OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS IE/DEPER/EGCB V. Thomas Low Not needed. Resolved. IE has concluded that no further action is warranted. (Reference) i j j

REFERENCE:

Memo from R. L. Baer to K. V. Seyfrit, March 21, 1985, " Failure of Manual of Initiation Pushbutton Switches in ESF Systems at McGuire Station Units 1 and 2" Y E-227-1 1

AE0D RECOMMENDATION TRACKING SYSTEM RECOMMENDATION SOURCE: Engineering Evaluation AE0D/E304 Responsible AEOD Engineer: T. Cintula TITLE OR

SUBJECT:

"B ck low Protection in Common Equipment and Floor Drain Systems" f RECOMMENDATION'1 Provide backflow protection for drain systems in older operating plants. RESPONSIBLE CRGR J OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS j NRR/DSR0/EIB D. Thatcher High 8/86 (If backfit Proceeding satisfactorily. This is required) issue is being addressed in Generic Issue 77. The task action plan for ] Generic Issue 77 has been prepared and approved. Implementation of the plan is currently in progress with an expected completion date of 8/86. ) 9 E-304-1

AE0D RECOPMENDATION TRACKING SYSTEM RECOMMENDATION SOURCE: Special Study Report - C. J. Heltemes, Jr. to H. R. Denton dated January 13, 1984 and follow-up memorandum dated August 8, 1984 R;sponsible AEOD Engineer: E. Trager TITLE OR

SUBJECT:

" Human Error in Events Involving Wrong Unit or Wrong Train" RECOMMENDATION 1 Consider the need for further clarification or guidance on what constitutes an acceptable independent verification program. RESPONSIBLE CRGR 0FFICE/DIV/BR CONTACT PRIORITY REVIEW STATUS NRR/DHFT/HFIB F. Rowsome High N/A Proceeding satisfactorily. NRR memorandum dated February 13, 1984: "The safety concerns identified in Generic Issue 102 -a Human Error in Events Involving Wrong Unit or Wrong Train, are to be 1 addressed in Generic Issue HF-102, Maintenance and Surveillance Program Plan....If DHFS decides not to include GI 102 in the MSP, then the issue should be submitted for y prioritizing as a new generic issue in accordance with the procedure i outlined NRR Office Letter No. 40." DHFT efforts are continuing on this item. (See status of Recommenda-tion 2.) 4 S-401-1

AEOD RECOPMENDATION TRACKING SYSTEM RECOMMENDATION SOURCE: Special Study Report - C. J. Heltemes, Jr. to H. R. Deriton dated January 13, 1984 and follow-up memorandum dated August 8, 1984 R:sponsible AE00 Engineer: E. Trager ~ _ TITLE OR

SUBJECT:

Human Error in Events' Involving. Wrong Unit or Wrong Train" 47 l RECOMMENDATION 2 NRR review wrong unit / wrong train events and develop appropriate gridance to minimize such events. RESPONSIBLE CRGR OFFICE /DIV/BR CONTACT PRIORITY REVIEW STA1US NRR/DHFT/liFIB F. Rowsome High N/A Proceeding satisfactorily.. F NRR and AE00 staff have completed t ~~ - site visits and reviews at ten 30 multi-unit sites to obtain ~ information on the factors that s appeared to have? contributed to these types of events. The results of this study.will be used by DHFT to help determin? whether WU/WT x-l ^ events should be taken up as an issue separate fr'om Maintenance and L, s Surveillance Program Plen. This ) recommendation is being addressed in Gsneric Issue HF-102. h 1 S-401-2 t

i AE0D RECOMMENDATION TRACKillG SYSTEM RECOP91ENDATION SOURCE: Memorandum - C. J. Heltemes, Jr. to R. DeYoung, May 'll,1984 R2sponsible AE00 Engineer: S. Rubin TITLE OR

SUBJECT:

" Pressure Locking of Flexible Disk Wedge Type Gate Valves" REC 0f#9ENDATION 1 Give imediate consideration to issuing a bulletin on the subject of pressure locking of flexible disk wedge-type gate valves. RESPONSIBLE CRGR OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS IE/DEPER R. Singh High N/A Proceeding satisfactorily. IE prepared a draft bulletin as 1 a part of a.CRGR review package. ] Shortly thereafter, it was learned j that INPO had begun work on an SOER on this subject. AE00 and IE agreed to allow INPO to address the concerns, and activity by IE on the bulletin was suspended with AE00 agreement. On 12/14/84, INPO issued SOER 84-7 " Pressure Locking and Themal Binding of Gate Valves." This document was found by AE00 and IE to be an acceptable substitute for the bulletin originally proposed. IE and AE00 will monitor the LERs, and 50.72 Reports for adequacy of corrective actions taken in response to the INPO SOER. IE and AE00 will review the 1986 LER and NPRDS data related to this issue to monitor the effectiveness of the i industry program. ] S-402-1

AEOD RECOMMENDATION TRACKING SYSTEM A RECOMMENDATION SOURCE: Special Study Report AE0D/S503 and Memorandum, Heltemes to Denton dated September 19, 1985. (See also recommendations related to AE0D/C203 on pages C203-1 and C203-2.) R:sponsible AE0D Engineer: E. J. Brown TITLE OR

SUBJECT:

" Evaluation of Recent Valve Operator Motor Burnout Events" RECOMMENDATION 1 In view of the more than 200 motor burnout events, the NRR plan to address motor burnout should be expedited. G RESPONSIBLE CRGR OFFICE /DIV/BR CONTACT PRIORITY REVIEW STATUS NRR/DSR0/EIB

0. Rothberg Medium Proceeding satisfactorily. This issue will be addressed in Generic Issue II.E.6, "In-Situ Testing of Valves," which is being evaluated under contract with BNL. As of mid-April 1986, the contract with BNL

~~ had been funded, approved and initiated. This issue will also be further evaluated by NRR during the review of the final AE0D Case Study on MOV performance, to be issued in early 1986. S-503-1

1 7. STUDIES CURRENTLY IN PROGRESS Every two months' AE0D issues to cther NRC Headquarter and Regional Offices, INP0, and NSAC a listing of ongoing AE0D studies of operational events involving reactor and nonreactor activities. When completed, each study will normally be documented as either a technical review report, an engineering evaluation report, a special study report, a trends and patterns report, or a case study report. The report type depends on the nature of the study and any associated AE0D recommendations. The status of each study, including the close-out report number of each completed study, is also identified in each status report. If any office or organization wishes to obtain additional information or discussion on any specific study, they are requested to contact either the responsible engineer or supervisor (Stuart Rubin for reactor studies, Robert Dennig for trends and patterns studies, and Kathleen Black for nonreactor studies). Additionally, the other organizations are requested to notify AE0D regarding related work or information that should be considered in either scoping cr completing these studies. A listing of AE0D studies which are in progress and those completed as of - April 1986 is identified in this section. R0AB Study Description Status Operational Experience Involving Failures of Safety-Related Electrical Inverters In Review Interlock Arrangements for Charging Pump Breakers Drafting Check Valve Failure Study in Progress A Review of Motor Operated Valve Performance Issued for Peer Review Effects of Ambient Temperature on Electronic Equipment in Safety Related I&C Systems Study in Progress H Fire and Failure of Detection System Study in Progress 2 Lightning Events at Nuclear Power Plants In Review Corrosion of Safety Related Relays Study in Progress Potential ECCS Pump Damage Due to Postulated RAS Actuation During a LOCA In Review i Inaccessibility of Vital Areas During Operation Issued (E603) i 151

R0AB (cont.) Study Description Status Pressure Sensitive Temperature Switch Results in Unwanted Actions Issued (T601) Inadequate Blocking of Isolation Valves During Equipment Maintenance Activities Study in Progress Delayed Access to Safety-Related Plant Areas Due to Loss of Ventilation Supply Fans Just Assigned Unexpected Criticality Due to Incorrect Calculation and Failure to Follow Procedures Issued E602 Failure of Safety Valves to Reseat In Review Design / Construction Problems at Operating Nuclear Plants Study in Progress Component Failures Due to Vibration Study in Progress MSIV Failures Study in Progress Overspeed Trip of Turbine Driven Pumps Issued for Peer Review Reactor Draindown Caused by Incorrect Valving In Review Foreign Material in Safety Related Systems Drafting Mispositioning of Dual Function Valves Issued (E601) RCIC Spurious Steam Line Isolation Issued (E604) Failure of Emergency Diesel Cooling Water Due to Loss of Air In Review Interactions Between ADC and RCIC and HPCI Study in Progress Reexamination of Water Hammer Occurrences Just Assigned Reversed Sensing Lines or Electrical Leads Drafting Arrangement of Support System Piping and Cooling for ECCS Equipment Just Assigned Problems with Reactivity Control Systems (Mechanical) Drafting PORV/ Block Valve Operating Problems in 4 ' Westinghouse Plants, 1981-1985 Just Assigned 152

ROAB (cont.) Study Description Status High Pressure-Safety Injection System Operational Problems in Westinghouse Plants, 1981-1985 Just Assigned Data Comparison Study for 1984 Events Drafting Loss of SI Capability at Indian Point In Review AFW Pump Unavailability at Turkey Point In Review Degradation of Safety Systems Due to Component Misalignment and/or Mispositioned Control / Selector Switches In Review Compression Fitting Study Study in Progress LOCA Outside Containment Just Assigned Tot:1 System Failure of Service Water Systems at Westinghouse Plants from 1981 to 1985 Just Assigned Plant Air Systems Drafting PTB Study Description Status Trends and Patterns Analysis of Reactor Operational Experience (1981-1983) Drafting Trends and Patterns Analysis of Reactor Scrams (January-December 1985) Drafting An Overview of Nuclear Power Plant Operating Experience Feedback Programs Issued for Peer Review Trends and Patterns Analysis of Engineered Safety Features Actuations (July-December 1984) Drafting Trends and Patterns Analysis of Engineered Safety Features Actuations (1985) Study in Progress Performance of Plants in Operation Less Than Two Years Drafting Trends and Patterns Analysis of Technical Specification Violations for 1985 Study in Progress 153

PTB (cont.) Study Description Status Trends and P&tterns Analysis of System Unavailability for 1984 In Review Trends and Patterns Analysis of System Unavailability for 1985 Study in Progress NAS Study Description Status Overexposure at New England Nuclear Study in Progress Misadministrations--Review of Licensees Having Multiple Misadministrations Drafting Rupture of an Iodine-125 Brachytherapy Being Revised Source at the University of Cincinnati Following Peer Medical Center Review Loss of Ir-192 Seed Study in, Progress Radiography Source Disconnects Study in Progress Five-Year Report on Medical Misadministrations Reported to NRC In Review Follow-up Study of Human Error in Study in Progress Events Involving the Wrong Unit or Wrong Train l 154

\\ Appendix A Sumary of 1985 Abnormal Occurrences (Including Those Approved by the Commission and Those Still in the Staff Office /Comission Approval Chain) A-1 1

Abnormal Occurrences CY 1985 Report No. A0 Crite"lon A0f Title of A0 NUREG-0090 or Example Comments 85-1 Premature Criticality Vol. 8, No. 1 A-9 During a plant startup, Summer During Startup Unit I experienced an unantici-pated transient which resulted in a high flux positive rete trip (automaticshutdown). Cause was due to a trainee (and the super-visor responsible for his actions) to be fully aware of plant status while withdrawing control rods. 2 L, 85-2 Diagnostic Medical Vol. 8, No. 1 G A patient received a radioactive Misadministration material other than that prescribed for a scheduled diagnostic test, resulting in a therapeutic dose to the thyroid in the range of 6,500 to 9,000 rads. As a result, the patient has perhaps a 50% chance of developing hypothyroidism. i 85-3 Diagnostic Medical Vol. 8, No. 1 G A patient received a diagnostic Misadministration radiation exposure that was approxi-mately 10 times the intended exposure (i.e.,1,000 microcuries of iodine-131 rather than 100 microcuries). l I l i f

+ i Report No. A0 Criterion A0f Title of A0 NUREG-0090 or Example Comments I 85-4 Unlawful Possession of Vol. 8, No. 1 G A person was charged with illegal Radioactive Material possession and use of radioactive i material (americium-241) and for making false statements to the NRC. ) AS85-1 Overexposure of an Vol. 8, No. 1 A-1 An employee of a Texas licensee i Employee received a 29.2 rem whoe body exposure while disassembling a radioactive exposure device. The employee did not know the device. contained a 24 curie iridium-192 3, source. s. AS85-2 Radiation Hand ' Burn to Vol. 8, No. 1 A-1 An employee of a Texas licensee i an Assistant Radiographer received an estimated 2000 rems to his right palm due to an unretracted source of 100 curies of iridium-192. i AS85-3 Overexposure of an Vol. 8, No. 1 A-1 An employee of a Texas licensee Assistant Radiographer J received about 1320 rems (beta and { gamma),to his right hand from an j unshielded cobalt-60 source. 1 AS85-4 ' Lost Well Logging Source Vol. 8, No. 1 A-5 A Texas licensee discovered a 1.5 and curie cesium-137 source was missing A-6 from its shield. The source was later found in a cow pasture. The cause is presumed to be due to theft of the source from the shield. f

} I l } l l Report No. A0 Criterion A0# Title of A0 NUREG-0090 or Example Comments 85-5 Inoperable Safety Vol. 8, No. 2 G-2 All three safety injection pumps 4 Injection Pumps were declared inoperable at Indian Point Unit 2, due to boric acid precipitation and possible gas ] (nitrogen) binding. k 85-6 Significant Deficiencies Vol. 8, No. 2 A-11 Applications for reactor operator in Reactor Operator and licenses containing apparently { Training and Material G-3 false information were submitted False Statements by Grand Gulf to the NRC in i September 1981, March 1982, and ? May 1982. = 85-7 Loss of Main and Auxil-Vol. 8, No. 2 G-2 Davis-Besse experienced a complete iary Feedwater Systems and loss of w.ain and auxiliary feedwater A-11 for about 12 minutes during an event involving an automatic shutdown 1 j from operation at 901 power. The 3 event involved several equipment mal-i functions and extensive operator actions, y including operator actions outside the control room. The event was investi- ] gated by an NRC Incident Investigation j Team and results published in NUREG-1154 85-8 ' Diagnostic Medical ~ Vol. 8, No. 2 G A patient received a 10 millicurie dose of Misadministration iodine-131 instead of the intended 400 microcurie dose of iodine-123. i

i l i f Report No. A0 Criterion A0f Title of A0 NUREG-0090 or Example Comments i i l 85-9 Diagnostic Medical Vol. 8, No. 2 G A patient received five milli-l i Misadministration curies of iodine-131 rather than i 10 mil 11 curies of technetium-99m for a routine thyroid scan. This-l resulted in the thyroid receiving r -+ l about 1000 rads rather than the. j expected 30 to 40 rads. 85-10 Breakdown in Management Vol. 8, No. 2 A-11 The NRC issued an Order to remove Controls a licensee's District Manager and i Radiation Safety Officer (same person), and to suspend operations. The action was taken as a result Es nf assigning uncertified people to perform radiography, providing false information to the NRC, and i i falsifying training records. i 85-11 Therapuetic Medical Vol. 8, No. 2 G A patient received a 14,000 rad Misadministration dose to the left lung instead of the l prescribed 5,000 rad dose. 1 ) AS85-5 Overexposure of a Vol. 8, No. 2 A-1 The two people, working for a Texas l Radiographer and an licensee, received 8.3 rems and 34.3 l Assistant Radiographer rems whole body exposures, respec-4 tively, from a disconnected 100 ~ curie iridium-192 source. I i i k

i i i 3 l Report No. A0 Criterion l A0f Title of A0 NUREG-0090 or Example Comments (NUREG-009, Vol. 8, No. 3 was approved for publication by the Commision during late February 1986. The report contains with the following seven events to be i 85-12 Management Control Vol. 8, No. 3 A-11 On November 22, 1985, a 10 CFR Deficiencies Part 50.54(f) letter was issued for LaSalle Nuclear Power Station requesting information on the i licensee's plans to improve its j performance in managing its main-l tenance, operation, and modifica-i tion activities. Numerous problems have occurred between 1982 and 1985. 20 i L 85-13 Inoperable Steam Vol. 8, No. 3 G-2 For about 15 months at Mcine Yankee, Generator Low Pressure 9 of the 12 pressure transmitters l Trip that monitor pressure of the three . steam generators were inoperable due to closed or partially closed root valves. In the event of a l steam line rupture, the subsequent j J reactor trip, main steam isolation, and main feedwater isolation would not i l have initiated automatically on. low steam pressure signals. l ~ ~ ~ 85-14 Management Deficiencies Vol. 8, No. 3 A-11 Because of sericus NRC concern at Tennessee Valley regarding'significant programmatic Authority and management deficiencies at TVA, on September 17, 1985, the NRC issued a 10 CFR Part 50.54(f) letter j to enable the NRC to determine i i l

l 1 Report No. A0 Criterion A0f Title of A0 NUREG-0090 or Example Conments i whether or not the licensee for the Browns Ferry and Sequoyah facilities should be modified or suspended, or the application for the Watts Bar facility should be denied.- Signi-a ficant problems have also occurred a i in the licensee's construction of the Bellefonte facility. 85-15 Therapeutic Medical Vol. 8, No. 3 G A patient treated on a cobalt-60 j Misadninistration teletherapy unit received 3584 rads to a portion of the body rather than the prescribed 2000 rads. l 85-16 Therapeutic Medical Vol. 8 No. 3 G A patient received 15 mil 11 curies of Misadministration iodine-131 rather'than the prescribed i dose of 10 mil 11 curies. 85-17 Exposure of Radiographic Vol. 8, No. 3 A-11 Due to a serious breakdown in management controls and oversight of the licensed Personnel Due to Manage-ment and Procedural Control program, a radiographer and his helper Deficiencies received excessive whole body exposures (about 27.1 rems and 7.4 rees, respec-tively). 85-18 Diagnostic ~ Medical Vol.:8, No. 3 G A patient received 200 mil 11 curies Misadministration of sodium pertechnetate-99m rather than the intended 20 mil 11 curies of the radioactive material. j t i

1 i \\ Report No. A0 Criterion A0# Title of A0 NUREG-0090 or Example Coseents i (NUREG-0090, Vol. 8. No. 4 is in the initial stages of preparation with the following five events under j consideration by the staff for submittal to the Cosutision for approval.) i 4 85-19 Inoperable Main Steam Vol. 8, No. 4 G-2 With Brunswick Unit 2 in cold Isolation Valves shutdown, during surveillance ~ testing three main steam isola-tion valves (two of which were in i the same line) would not fast close. Problem due to contami-nation of the elastomer material (ethylene propylene) in the J associated solenoid valves. l 31 85-20 Nanagement Deficiencies Vol. 8, No. 4 A-11 On. December 24, 1985, the NRC l 5 at Fermi Nuclear Power issued a 10 CFR Part 50.54 (f) i Station letter to the licensee of Femi Unit 2 requesting ~information on i . plans to improve regulatory 4 ' performance. The letter identified l a series of operational and equip-i ment problems and attributed them 1 to ineffective management systems. } G The attending physician for a patient ',85-21. Diagnostic Medical Vol. 8, No. 4- ~ i ~ mistakenly prescribed a dose of 150 Misadministratior. microcuries of I-131 instead of I-123. q The radiopharmacy misinterpreted the prescribed dose as 5 millicuries of I-131. I

l Report No. A0 Criterion A0f Title of A0 NUREG-0090 or Example Coments I 85-22 Therapeutic Medical Vol. 8, No. 4 G A patient was scheduled to receive .i Misadministration 1000 rads to the lateral limbal area of the right eye using a Sr-90 applicator; however, the c medial limbal area of the right eye was treated. 85-23 Diagnostic Medical Vol. 8, No. 4 G A patient received 4.98 millicuries Misadministration of I-131 instead of-a 10 to 15 microcurie dose usually given for a 24-hour thyroid uptake test. ! ? E ) I i J i I i

n Appendix B Sumary of 1985 Scram Data l i i l B -1

APPEull! B 1905 REACTOR TRIP MTES MME RMPJAL AUTO LESS THAN IREATER ~ CRI'TICAL TRIP RATE PER MEM TIME Milt ItEsuAL TMN NOURS 1000 N0dRS BETWEEN TRIPS 151 PONER 151 P0utR POWER ST 15 P0utR ST 151 DIAPLC CANY 3 2 2 11 4 9 1874.2 4.00 209.2 M TERFDRD 3 0 24 15 3343.0 4.49 222.9 PALD VEUE 1 0 12 3 9 2008.0 3.12 320.9 ' BYRON 1 1 22 9 14 4656.4 3.01 332.6 BRAC BULF i 1 12 1 12 5092.1 2:36 424.3 CATANSA 3 9 4 B 3612.4 2.21 451.6 CtLLAWAY l 2 18 3 17 1161.0 2.08 480.1 UDLF CREEK 2 13 6 9 4471.7 2.01' 496.9 5ALEM 2 1 9 1 9 5231.2 1.72 581.2 CRYSTAL RI'.'ER 3 2 6 1 7 4385.3 1.'60 626.5 LASALLE 1 2 7 0 9 5757.5 l' 56 639.7 SAN DNOFRE 2 0 10 2 5 5235.8 1.53 654.5 91ASLO CANYDN 1 1 9 2 I 5295.6 1.51 662.0 RCSUIRE2 3 7 2 8 5490.5 1.46 686.3 PEACH B0iTC5 2 1 5 2 4 2910.6 1.37' 727.7 ARKANSAS 2 0 10 2 8 6377.4 1.25 797.2 SFONG FF m i 1 1 0 2 1647.7 1.21' 823.9 FITZ8ATK;;t 0 7 0 7 5799.6 1.21' 828.5 ~ INDIAN POINT 3 2 8 3 7 5901.1 1.19 843.0 WFP!S 2 1 10 3 8 6999.7 1.16 862.5 ' SECUOYAH2 0 6 0 6 5289.4 1.13 881.6 StPMER 4 8 5 7 6439.9 1.09' 920.0 DAVIS-BESSE I 0 4 1 3 2846.6 1.05 948.9 SAN CNDFRE 3 0 5 0 5 4789.9 1.04 958.0 H.8.RCS!N3M 2 0 3 0 7859.8 1.02 982.5 DRESDEN 2 1 7 3 5 4961.6 1.01 912.3 TU; REY POINT 4 0 9 1 8 7916.8 1.01 189.6 A R AN3s! ! 2 5 0 7 7005.4 400 1000.8 MAIEE Yt& EE I I 2 7 7037.1 0.99 1005.3 TURKEY POINT 3 0 5 0 5 5405.0 0.93 1091.0 CA'VERTCLIFFS1 0 5 0 5 5317.6 0.92 1073.5 LA ;Fi!3E 1 9 3 7 7757.2 0.90 !!08.2 BEGEK ELEY 0 6 1 7 B'45.3 0.55 !!77.9 IN:!AN FCIh! 2 1 8 2 7 8504.1 0.E2 1214.9 OfSTER CFEtt 0 7 2 5 6818.5 0.73' 136!.7 FA' LEY 2 0 5 0 5 6953.1 0.73 1377.6 : j MCEU!;E1 0 5 0 5 6842.4 0.73 1:18.5 $UEQUE m a 1 0 4 0 4 55it.5 0.71-1399.6 RANX 5! 01 0 4 2 2 2674.6 0.70 14:7.3 SL520EU W : 0 5 0 5 7121.2 0.70 1424.2 IRCh15 FE W 3 2 C 1 1 1517.5 0.66 1517.5 8UG CIT!IS 2 0 4 0 4 .4361.8 0.63 1590.4 DRECEN3 0 5 1 4 6715.0 0.60 !!79.7 N!htP!LEFD!!.*1 1 6 2 5 8524.0 0.59 1704.E TR;.4 0 4 0 4 68(4.7 f.59 1701.2 HA;'A"hE*E 2 3 0 5 86E*.4 0.58 173o.5 NAfDI I 2 3 1 4 6907.5 0.58 1726.9 Llr. eel:K 1 0 3 1 2 3420.1 C.59 1710.1 St.LtXIE 2 2 8 2 4 7442.7 0.54 1f60.7 FARuff1 0 5 1 4 7504.1 0.53 1876.0 CCCFI4 1 0 0 1 2057.5 0.49 2057.5 APP (NCl!8 1985 REACTOR TRIP RATES 8-3

o I l ) RAM " EEight AUTS LESS TMN WEATER CRITICAL TRIP RATE PER MANTIME M?lt IR EINE ineN IIMt3 1000 NOUR$ NTEEN ft!PS ISEPOWER 151 POWER PONEA ST 15 POWER ET 151 i Ts!IMILE'll.! 1 1 1 2004.8 0.48 2084.C OCOEE 1 9 4 0 4 H53.3 0.47 2113.3 GCONEE 2 0 4 1 3 6740.3 0.45 2246.B CEVERTCLIFFS2 l 2 6 3 6084.2 0.44 2294.7 BRUNSNICK 2 1 2 6 3 7134.8 0.42 2378.3 NENAUNEE I 5 3 3 7266.5 0.41 2422.2 S.C.CDOK1 0 1 9 1 2595.6 0.39 2595.6 R.E.EINNA I 7 5 3 7838.4 0.38 2612.8 SURRY! 9 5 2 3 7935.4 0.38 2645.1 RONTICELLO 9 3 0 3 5163.0 0.37 2721.0 FILERIM 9 4 1 3 8159.0 0.37 2719.7 9.C.C00K 2 9 2 6 2 5948.I 0.34 2974.4 SAN ONOFRE I I I 6 2 6783.8 0.29 3391.9 DEuhSWICK1 0 1 0 1 3409.6 0.29 3409.6 , NORTH m 4 1 2 4 6 2 6930.I 0.29 3469.4 RlLLSTONE I 6 3 1 2 7324.4 0.27 3662.2 PRISAES 6 2 0 2 7490.2 0.27 3745.1 PRA!RIE15L4ADI e 3 1 2 7363.2 0.27 3661.6 NATCH 2 6 4 2 2 7373.1 0.27 3686.5 SEGUOTAH I 6 3 0 1 3797.2 0.26 3797.2 LASE LE 2 4 2 1 1 3777.6 0.26 3777.6 PEACH BOTitPt 3 6 3 0 1 40:5.7 0.25 4055.7 BUA5 CIT!ElI 2 1 2 B339.0 0.24 4169.5 NORTH AEh4 2 1 1 0 2 8534.4 0.23 4267.2 RILLSTONE 2 0 1 6 1 4460.7 0.22 4460.7 !!ON 1 1 3 3 1 5321.2 0.19 5321.2 SUKAf 2 4 1 9 1 59!6.5 0.17 5936.5 !!ON 2 e 1 0 1 59C9.2 0,17 5909.2 SCONEE 3 0 2 1 1 6140.9 0.16 6140.9 PDINTBEACHI 4 1 0 1 6174.4 0.14 6974.4 St.LUCIEI 1 0 0 1 7134.7 0.14 7134.7 $4' EM 1 0 1 0 1 B361.9 0.12 2361.9 YAHEERCe!L 4 2 2 0 7595.3 0.00 816RDCIPC!NT 2 2 4 .0 6539.5 0.00 FT. CEH11I 4 0 0 0 6466.1 0.00 FEMI 2 1 1 8 0 2400.7 0.00 AIVER I!G 1 1 ( 5 0 589.4 0.03 VEM3TTWii 6 4 6 0 6297.2 0.00 DUa! A50.! 0 1 1 0 4733.2 0.00 FDIN'SEO 2 0 0 0 0 7576.2 0.00 l PRA!RIE156&I2 4 0 0 0 7408.6 0.00 61 464 136 389 9 B-4 -o_ - _.. -.. ~ .}}