ML20247B753
| ML20247B753 | |
| Person / Time | |
|---|---|
| Issue date: | 05/06/1998 |
| From: | NRC (Affiliation Not Assigned) |
| To: | |
| Shared Package | |
| ML20247B617 | List: |
| References | |
| FOIA-98-173 NUDOCS 9805080262 | |
| Download: ML20247B753 (28) | |
Text
{{#Wiki_filter:l I p nag g U.JITED STATES 7 M E NUCLEAR REGULATORY COMMISSION 'f WASHINGTON, o.C. 2C555-0001 %, w 8 ATTACHMENT 2 ,o
- ...+
EVALUATION OF BOILING WATER REACTOR OWNERS GROUP TOPICAL REPORT NED0-32291, l " SYSTEM ANALYSES FOR ELIMINATION OF SELECTED RESPONSE TIME TESTING REQUIREMENTS"
1.0 INTRODUCTION AND BACKGROUND
l The requirement for periodic testing of reactor trip systems is established in Section 50.55a, " Codes and Standards," of 10 CFR, " Domestic Licensing of Production and Utilization Facilities." This states that " protective systems must meet the requirements set forth in editions or revisions of the Institute of Electrical and Electronics Engineering Standard: ' Criteria for Protective Systems for Nuclear Power Generating Stations,' (IEEE-279)." In addition, 10 CFR 50.36 paragraph (c)(1)(ii)(A) requires inclusion in the Technical Specifications the limiting safety systems settings for nuclear reactors, those setting "so chosen that automatic protective action will correct the abnormal situation before a safety limit is exceeded." 10 CFR 550.36 (c)(3) l " Surveillance requirements" states " Surveillance requirements are requirements related to test, calib ation, or inspection to assure that the necessary quality of systems and components is maintained, that facility operation will be within the safety li.mits, and that the limiting conditions of operation l will be met." The Nuclear Regulatory Commission (NRC) in 1975 implemented a program making response time testing (RTT) a requirement of the technical specifications (TSs). IEEE Std. 338-1975, " Criteria for the Periodic Testing of Class lE Power and Protection Systems," and its later version IEEE Std. 338-1977, " Criteria for the Periodic Testing of Nuclear Power Generating I Station Safety Systems," (Ref. 1) provided generic guidance on the conduct of 9005080262 980506 PDR FOIA CHAPMAN 98-173 PDR L
_. _ _ _ _ = _ _ _ _ - - _ _ _ _ - _ _ _ _ _ _ - - - - response time verification tests. The NRC staff endorsed IEEE Std. 338-1977 in Regulatory Guide (RG) 1.118, " Periodic Testing of Electric Power and Protection Systems," Revision 2, June 1978 (Ref. 2). Additional guidance on RTT is provided. in the Instrument Society of America (ISA) Standard ISA- { S67.06-1984, " Response Time Testing of Nuclear Safety-Related Instrument Channels in Nuclear Power Plants," dated August 29, 1986 (Ref 3). ISA-S67.06-1984 has not been endorsed by the NRC staff. In May'1991, the Electric Power.Research Institute (EPRI) issued Report I NP-7243, " Investigation of Response Time Testing Requirements" (Ref. 4). The stated purpose for this report was the nuclear industry's desire to increase plant availability and minimize personnel radiation exposure by justifying the elimination of RTT requirements for the pressure sensors used in l' instrumentation loops. The Boiling Water Reactor Owners Group (BWROG) under the auspices of the General Electric Company, issued Licensing Topical Report NED0-32291, " System Analyses for Elimination of Selected Response Time Testing Requirements," (Ref. 5) in January 1994. In NED0-32291, the BWROG utilized the information in EPRI Report NP-7243, and proposed the elimination of the requirements for . performance of RTT of selected instrumentation in the Reactor Protection System (RPS), Emergency Core Cooling System (ECCS), and Isolation Actuation System (IAS). Table 1 identifies the instrumentation that is addressed by the topical report. 4 i
The BWROG concluded that the RTT requirements for the devices identified in Table 1 can be removed from TSs when the devices are used in systems listed in Table 2. The BWROG provided the basis for the proposed elimination of RTT for the instruments / components identified in Table 1 in NED0-32291. The BWROG noted that within a trip function, redundancy exists in individual instrument channels (e.g.,1-out-of-2 twice) and diversity exists in most safety trip functions (e.g., neutron flux, water level, drywell pressure). The BWR0G stated that the failures which affect response times can be detected during other surveillance tests required by current TSs. In addition, to provide a safety perspective on elimination of RTT, the BWROG evaluated the safety significance of a delayed response using the results'of a BWROG survey which concluded that instrument response time delays of five seconds or greater can be reasonably detected by instrument technicians. The BWROG concluded that a five second increase in the response time of individual specific trip functions has a very low safcty significance based on realistic accident evaluations. This provides further confidence that elimination of RTT for the selected trip functions has no significant safety impact. NED0-32291 concludes that response times are adequately maintained with the l-current practices and that RTT is unnecessary based on plant operating history and experience.- The other existing TS required surveillance tests (i.e., I calibration-tests, functional tests, and logic system functional tests) provide sufficient indication of instrumentation health. The BWROG evaluation
. Table 1 frwtruments/ Components for Which Response Time Testing Can Be Eliminated Trip Units Pressure Sensors (continued) Rosemount 510DU and 710DU Barton 764 Differential Pressure Electronic GE Trip Unit 184C5988 Transmitter Rosemount Differential Pressure Transmitter Relays Models 1151,1152,1153 and 1154 GE liFA Rosemount Pressure Transmitter Models 1151. GE IIM A 11521153, and 1154 GE liga SOR Differential Pressure Switch Bourns SOR Pressure Switch Agastat GP/EGP Family ASEA RXMil2 Sensors GE SAT 6004 Barton 760 GE Type CR105, CR120A, CR205, and CR305 Barksdale TC9622-3 Potter Brumfield MOR and MDR Barksdale BIT-M12SS-GE { Potter Brumlield KII4690 External Devices / Flow Devices (Bailey / GEM AC Modules) 560 Alarm Unit 752 Summer 745 Alarm Unit 750 Square Root Extractor Rosemount!GE Trip Unit Noise Suppression ] Radiation Devices Filter Capacitors l Trip Auxiliary Unit (238X697 Series) - Cornell Dublier WBR 2000-50 I Indicator Trip Unit (12982802 Series) - Sprague 500D-35 NUMAC Log Rad Monitor (304A3700 Series) Miscellaneous Devices Pressure Sensorn Hi Current isolator 133D9947 Barton 288 and 289 Differential Pressure Optical isolator 204B6186AA Indicating Switches Optical Isolator 204B6188AA Barton 763 Gauge Electronic Pressure Transmitter Table 2 Systems for Which Response Time Testing Can Be Eliminated (1) ECCS actuation instrument loops for the (3) Sensors only, for the following RPS following systend; actuations: (a) Low Pressure Core Spray (a) Reactor Vessel Steam Dome Pressure l (b) Low Pressure Coolant injection - liigh (c) High Pressure Core Injection / Spray (b) Reactor Vessel Water level - Low. Level 3 (2) Isolation System actuation instrument loops (c) Reactor Vessel Water Level - High. [ for the following systems: Level 8 l (a) RCIC System (b) IIPCI System (4) Sensors only, for the following MSIV (c) IIPCS System closure actuations: (d) RWCU System (a) Reactor Water Level 1 or 2 l (e) Primary Containment (b) MSL Radiation High j (O Secondary Containment (c) MSL Low Pressure l (g) RilR Shutdown Cooling /Ilead Spray i 1 )
. states that RTT is of no safety significance, causes unnecessary personnel exposure, reduces availability of safety systems during shutdown, and is a significant burden on utility resources. 2.0 DISCUSSION Current Standard Technical Specifications (STS) require nuclear power plants to periodically perform RTT for instrument channels in the RPS, ECCS and IAS. The intent of these tests is to ensure that changes in response time of instrumentation beyond the limits assumed in safety analyses are detected, and combined with instrument calibration, to ensure that the instrument is operating correctly. The response time tests do not demonstrate that the instrument response time design value is met, but rather that the specified i performance requirements of the TSs are satisfied. i l The BWROG stated in NED0-32291 that operational history has shown that significant degradation of instrumentation response times is being detected during the performance of calibrations and other surveillance tests. The BWROG further stated that the performance of conventional response time tests has proven to be of little value in assuring that instrumentation will perform as required or for determining the health of the instrument because the majority of allowable instrumentation response times are system response times rather than instrument response times.
. In NE00-32291, the BWR0G proposed. elimination of RTT for the entire instrumentation loops for the IAS (except main steam isolation valve actuation) and ECCS. For the RPS and main steam isolation valve (MSIV) actuation instrumentation, the BWROG proposed elimination of just the sensor RTT. The response times'for the IAS and ECCS instrumentation are a small fraction of the total system response time requirements (i.e.,10 to 13 seconds for IAS, and 27 to 64 seconds for ECCS). Instrumentation components that may experience response time degradation will continue to respond in the microsecond-to-millisecond range prior to complete failure. Therefore, the BWROG stated that such response time degradation would have no significant adverse effect on system actuation and the instrumentation would continue to meet overall system requirements. For the RPS and MSIV actuation instrumentation loops, the overall TS response time requirements are much shorter (0.33 to 2.0 seconds) than for the IAS and ECCS instrumentation. For the same component, the response time degradation before failure can be significant in relation to the overall system response time requirement. As a result, significant changes in instrumentation response time may not be readily detected for all components during other surveillance tests. The BWROG, therefore, requested the elimination of selected sensor RTTs only, but not the. remaining instrumentation loop components (e.g., trip units and relays) for RPS and MSIV actuation loops. For the RPS, the three selected trip functions considered for elimination of . sensor RTT are:
I Reactor Vessel Steam Dome Pressure - High Reactor Vessel Water level - Low, Level 3 Reactor Vessel Water Level - High, Level 8 l The typical instrumentation response time requirements for these RPS trip functions range from 0.33 to 1.05 seconds. As stated above, because of the more restrictive response time requirements for these circuits than the IAS and ECCS circuits, only the sensors analyzed by EPRI were considered for RTT elimination. For the IAS instrumentation, the following TS response time requirements were considered for elimination: MSIV Closure Reactor Water Level 1 or 2 Main Steamline (MSL) Radiation High MSL Low Pressure Remaining Isolation Actuation Reactor Core Isolation Cooling (RCIC) System High Pressure Coolant Injection (HPCI) System i High Pressure Coolant Spray (HPCS) System Reactor Water Cleanup (RWCU) System Primary Containment Secondary Containment l l 2
. Residual Heat Removal (RHR) Shutdown Cooling / Head Spray The BWROG stated that since the IAS instrumentation logic functions are in parallel with the startup of the diesel generators, the response time requirements (< 10 seconds to < 13 seconds) are not critical for the majority of the isolation circuits. The BWROG stated that this is consistent with the recommendations of Generic Letter 93-05-(Ref. 6) and NUREG-1366-(Ref. 7) for the deletion of RTT where the required' time corresponds to the_ diesel start time. The main steam isolation valve response time requirements (0.5 seconds ~to 2.0 seconds) are an exception. Therefore, the BWROG request for elimination of RTT for the MSIV actuation circuits covers only the sensors. For the ECCS, the following RTT requirements were considered for elimination: Low Pressure Core Spray (27 to 43 sec.) Low Pressure Coolant Injecti'n (37 to 64 sec.) High Pressure C..e Injection / Spray (27 to 35 sec.) ECCS instrumentation response time requirements range from 27 to 64 seconds 1 and, therefore, actual instrument response which is of much shorter duration is not important to meeting these times. Unlike the IAS, the diesel generator,. emergency cooling pumps, and the injection valves response times 1 are in series with the instrumentation response time. For these loops, only the instrumentation will be eliminated from RTT. The overall ECCS system response time requirement, which includes diesel generator, injection valves,
l _g_ l . pumps, and other components, still applies. The diesel generator and the 1 injection valve TS response time requirements are not eliminated. 1 The BWROG indicated that the fundamental approach developed in their study for RTT elimination is consistent with the guidelines of RG 1.118, Revision 2, which endorses IEEE 338-1977, and is stated as follows: " Response time testing of all safety-related equipment, per se, is not required if, in lieu of response time testing, the response time of the safety equipment is verified by functional testing, calibration checks or other tests, or both. This is acceptable if it can be demonstrated that changes in response time beyond acceptable limits are accompanied by changes in performance characteristics which are detectable during routine periodic tests." The BWROG stated that changes in instrument response are detectable prior to significant degradation without the performance of conventional RTT, and will therefore fulfill the criteria of IEEE 338-1977. The BWROG also indicated 'that-the testing which is currently required by TSs, that is, calibrations, functional testing, logic system functional testing and channel checks will identify any significant degradation of instrument response time. A description of each test, including STS definitions, and the relative advantages of each method for detection of degradation was included in Appendix B to the BWROG topical report. Although this description did not
. delineate which test detects which response time failures, or the manner of detection, it did show that response time testing of individual instrument channels had the' longest time interval between performance of surveillance activ.ities, 36 months, when compared to calibrations, functional tests, or channel checks, which are performed at the same or more frequent surveillance ' intervals of 18 months, quarterly or monthly, and once/ shift respectively. The BWROG selected Fermi-2 and River Bend as lead plants to represent typical BWR/4 and BWR/6 plants respectively. TS RTT requirements ("RTT Trip t Functions") were identified for the lead plants, and a detailed listing of specific loop components was compiled for each loop. All components in the instrumentation loops which could potentially affect the loop response time were identified, and failure mode evaluations were conducted in order to determine whether failures could affect response time. In those cases where the BWROG determined that a failure mode could affect the instrument loop response time, the consequences of such a failure were evaluated. Analyses were then performed to determine whether other surveillance testing would identify these potential response time degradations. The BWROG determined l that other surveillance tests could identify these degradations. y l. The BWROG stated that response time degradation is readily detectable by i instrumentation and control (l&C) technicians when performing calibrations and functional tests based on the following: I j
(1) By inputting a standard value over the operating range, the technician can observe any abnormal changes in calibration / functional span and operational range of the instrumentation, i (2) By observing the expected output when an input signal is injected, the technician can determine that the instrumentation is responding properly. (3) By checking calibration / functional points between zero and 100% of
- span, (a) the technician ensures that the instrumentation responds instantaneously with the injection of a test signal and, l
(b) the technician ensures the instrumentation performs in accordance,dith the instruments' design characteristics (i.e., linear or logarithmic). I (4) By observing the response from a trip signal initiated during a l functional test, the technician is able to detect a response time delay in the actuation of an alarm, light, or relay contact status. ) i i
. l l The BWR0G stated that an evaluation of the affected. loop components that exhibit degraded response time during calibrations and functional tests would be performed before the loop is returned to service. In general, the BWROG did not state in which manner (which surveillance type) the determination of response time failure would be made. In most cases, a statement that time response delay could be " detected by other surveillance tests" was used and considered to be sufficient. 1 The BWROG stated that as part of the failure mode evaluations, component experts and vendors were contacted to assist with and verify the analysis. No documentation or letters from these experts and vendors were presented in direct support of the elimination of RTT. The BWROG stated that due to variations in applications and licensee surveillance practices, instrumentation manufacturers were unwilling to make any recommendations concerning acceptability of calibration or other surveillance as a basis for the elimination of RTT. The staff agrees that plant-specific variations in I calibration techniques would preclude generic agreement from instrument manufactures on the appropriateness of elimination of RTT. This is discussed further in Section 3.0 of this report. 1 i J The BWROG performed a review of component failure experience by conducting BWR-specific surveys, and by researching the Nuclear Plant Reliability Data System, NRC. Bulletins, NRC Information Notices, and GE Service Information Letters. This review was used to determine if actual RTT failures had been
_ _ _ = _ - _ _ _ _ - _ _ _ _ _ _ _ _ _ _ . datocted by surveillance testing other than RTT as indicated in this information. Appendix 0 of the topical report summarizes the results of this review, and.shows that in most cases, failures which have an impact upon response time had been or could have been detectable by some other type of surveillance test. The BWROG also referred to-the work documented in EPRI Report NP-7243, " Investigation of Response Time Testing Requirements," dated May 1991 to support RTT elimination. This report evaluated failure modes and effects analyses for many of the sensors employed in the applicable BWR instrumentation loops. However, because not all components and sensors that the BWROG wanted to exempt from response time testing were covered in the EPRI report, the BWROG conducted supplemental evaluations as necessary. Staff review of the EPP.1 report determined that the data contained in the database were inconclusive with respect to the merits or usefulness of RTT, but were generally useful in showing that the majority of component failures which can effect ' response times will also impact the calibration or functional test results of the component. Although the EPRI and BWROG supplemental analyses of component failures, and their relationships to the capability to detect response time failures during other tests were not conclusive, establishing these relationships was not intended to be the sole justification for the elimination of RTT. l .To provide a. safety perspective on the elimination of RTT, the BWROG evaluated the safety significance of delayed response times. The BWROG surveyed l
l I ~ BWROG SURVEY OF INSTRUMENT TECHNICIANS l 40 2 l 5 30 9zIO N 20 u.O E m2 10 z E, m, I 0 I l 1 2 3 4 5 6 7 8 9 10 >10 l TIME TO DETECT SLUGGISH RESPONSE (SECONDS) Figure 1 instrumentation departments at participating BWR plants and at selected pressurized-water reactor plants. The BWROG determined that the technicians believed they can typically qualitatively detect an instrument with a sluggish response as short as five seconds. The technicians were asked: "When performing calibration and/or loop functional tests, what is your estimate of the maximum time it would take you to identify sluggish component performance? l ...Your estimate represents the time elapsed to detect a degradation in 1 instrument response time no matter what individual component is sticking, i sluggish, or failed." A summary of the results of this survey is provided in 1 Figure 1. Figure 1 shows the number of technicians responding to the survey l
as a tunction of the time to detect sluggish component performance. Fifty-percent (median value) of the technicians estimated the time to detect sluggish response as three seconds and eighty-five percent of the technicians' estimates were five seconds. l The BWROG conducted interviews with I&C personnel whose response was that greater th us ten seconds was necessary to determine what problems might exist in the instrument response. The BWROG determined that each one nf these individuals believed that a level transmitter with a capillary seal sensor takes from two to three times as long to determine if a component is sticking, sluggish, or failed depending on the length of the capillary tubes when compared to most other sensors. These individuals used capillary seal sensors as the basis for their answer although this type of sensor is not used in a time response application in nuclear power plants. Based on the survey results, the BWROG realistically evaluated the effect of a response time degradation to a measure of safety significance. The deta.. of these evaluations were included in Section 5.0 and Appendix J of the topical report. 3.0 EVALUATION The argument provided in the BWROG topical report in support for the elimination of RTT is that i.ppropriate alternatives to RTT are in place per the criteria of R.G. 1.118 and IEEE 338-1977, which allow elimination of RTT
I f . i if the response time of the safety equipment is verified by functional l testing, calibration checks or other tests. The report also stated that instrument response degradation of less than five seconds has little safety-significance. It is likely that instrument response of more than 5 seconds will be noted by the instrumentation technicians, it was concluded that elimination of RTT would not result in undue risk. The staff concurs with the BWROG that the majority of instrumentation degradation which will effect response time can be detected by other testing if properly performed. Calibration should be conducted in accordance with the i manufacturers' recommendations to provide a good general indication of the proper functioning of the instrument and channel. If, however, the manufacturer has recommended periodic RTT as well as calibration to ensure correct function of the instrument or device, concurrence must be obtained from the manufacturer that for this application, RTT is either unnecessary or inappropriate. There are known concerns with the elimination of RTT for certain components. The most notable of these is with regard to Rosemount oil-filled pressure transmitters. The BWROG has stated in Chapter 8 of the topical report: " Loss of fill oil in Rosemount transmitters is the subject of NRC Bulletin 90-01 and the associated supplement as well as Rosemount Technical Bulletins 1 through 4. Performance of response time testing is not the preferred method to detect loss of fill oil. l l l 1 l
. 1 l Licensee' actions.to address this failure mode have been taken pursuant to NRC Bulletin 90-01 and the associated supplement." Satisfactory compliance with Supplement I to Bulletin 90-01 would, therefore, be a prerequisite to the elimination of any RTT associated with Rosemount oil-filled pressure transmitters. l A second concern is associated with crimped capillaries in instruments that i use capillaries, which have been shown to affect response time. While not a TS requirement, good engineering practice dictates testing of the response time of new instrumentation used to replace defective equipment or used in . upgrades. This would be true not only for instruments using capillary tubing, but any instrumentation. The staff believes that performing an initial RTT at the time of installation is sufficient to reduce the possibility of crimped capillary tubing or other problems with instrument performance. Degradation in capillaries over time can be detected with the proper calibration methods as discussed below. The BWROG survey, while establishing that the instrumentation technicians believe they can detect an instrument with a sluggish response within five seconds,.did not ask or state how that determination was to be made. It is not unreasonable to expect plant personnel to note a deviation from the standard response of an instrument if the calibration or test is being done in a manner where a single technician can both provide the input to the channel being tested and observe the results of that input. If, however, the
. calibration or test is being conducted in a manner where a single technician cannot provide the input to the channel being tested and observe the results of.that input, a second technician is required. An example of this may be the test of a trip unit, where the stimulus is applied at one end of the equipment, and the trip is observed on the other end of the cabinet, or in a different room. In this case, two technicians in communication with each other would be required. When more than a single iridividual is involved, the individual observing the response may be less sensitive to delays on the order of seconds since that individual must rely on some means other than their own action to sense when the input was applied. For responses of 10 seconds or more, delays can be more accurately perceived since a difference of a second or two in communicating when the input was applied and the output is observed is less critical. Thus, in some cases, the techniques used for calibrations, functional tests, or channel checks may require modification to ensure that the required observation of instrument response is possible. j An additional important issue is the detection method used in the performance of the calibration. If the calibration is done with equipment with an inherent time delay, such as a slow ramp in the process variable, the l additional five second delay in actual instrument response may be difficult to notice. With a slow ramp in the process variable, a sluggish response can be mistaken for a change in setpoint. If, however, the calibration is done with the calibrating equipment using a step function or a fast ramp of I significantly less.than five seconds, the recognition of a five second delay in instrument response is reasonably assumable. l
) l l 1 1 l l l l l l i l A further potential problem inherent in the manner in which calibrations are done is that the sensor is almost always calibrated separately from the trip circuitry. Since both the sensor and trip unit could have a delay in response of up to five seconds, it may be possible for any one function to have a maximum of a ten second delay. However, the staff does not consider this a l significant safety concern, because a response time failure without a corresponding calibration failure is unlikely, and the possibility of two such failures occurring in a single channel is even more unlikely. While the BWROG I has not presented a statistical determination of this possibility, the staff believes the probability is sufficiently small to present no safety significance. Appendix J of the topical report provides a safety assessment of a delayed or sluggish instrumentation response on the order of five seconds for the trip functions selected for RTT elimination. This safety assessment is based on I realistic assumptions and shows that with a five second response time, no safety limit would be exceed d for those transients typically evaluated in the licensing analysis. In addition, within a trip function, redundancy exists in l individual instrument channels (e.g., 1-out-of-2 twice), and diversity exists in most safety trip functions (e.g., neutron flux, water level, drywell pressure). The staff reviewed the effect on plant safety of a delayed instrumentation response on the order of five seconds for the trip functions selected for RTT elimination, and agrees with the BWROG that a five second delay in any of the trip functions discussed in the topical report would not affect the capability of these systems to initiate reactor trip or to provide
- core cooling function. The staff concurs with the BWR0G's conclusion that significant margin exists in the licensing analysis assumptions such that an additional five second instrumentation response delay in the selected trip functions will not significantly impact plant safety. Licensees taking advantage of this elimination of RTT must ensure that operators and technicians continue to be aware of the consequences of instrument response time degradation and are familiar with the alternative means for detecting this degradation. Applicable plant-specific procedures may need to be revised to assure that technicians monitor for response time degradation during the performance of calibrations and functional tests. The staff has determined that 10 CFR 50.36 does not specifically require RTT in those cases where it has been shown that the response time of the instrument or channel will be verified by other means. Requirements related to the operability, applicability, and surveillance requirements, including performance of testing to ensure response times, for RTS and ESFAS systems are retained due to those systems' importance in mitigating the consequences of an accident. However, the elimination of response time testing requirements does not constitute a condition or limitation on operation necessary to obviate the possibility of an abnormal situation or event giving rise to an immediate threat to the public health and safety, in that the ability of the RTS and ESFAS systems to perform their safety functions are not adversely impacted since the response times of the devices are verified by other means. The staff has determined that calibration and other surveillance testing, combined
? l l l with technician awareness of the RTT requirement, will adequately ensure that the response time is verified for the devices identified in Table I when the ( devices are used in systems listed in Table 2. Further, the safety assessment of a delayed response time of approximately 5 seconds provides additional confidence that the elimination of RTT is not safety-significant.
4.0 CONCLUSION
J Most of the instrumentation RTT targeted for elimination in NED0-32291 l involves tests where the instrumentation loop response time is a small fraction of the total allowable system response time requirement. This typically occurs when the total allowable system response time is equal to or greater than ten seconds. In addition, selected pressure and differential pressure sensor RTT can be eliminated on the basis of the failure analysis done by EPRI and supplemented by the BWR Owners Group. On the bases of the above, the following response time tests were evaluated for elimination by the BWROG: (1) All ECCS actuation instrument loops (entire channel). i (2) All isolation system actuation instrument loops (entire channel) -{ except for MSIVs. (3) S.ensors only, for selected RPS actuation instrument loops. (4). Sensors only, for selected MSIV closure actuation. j l i
, Based on its review of the information presented by the BWROG, the staff agrees with the BWROG that, in general, operational history has shown that l degradation of instrumentation response times detected during the performance of calibrations and surveillance tests. Thus the existing surveillance requirements for the instrumentation provide confidence that the safety function of the plant instrumentation will be satisfied. Further, the staff notes that elimination of RTT will reduce the burden on licensees during refueling outages and, therefore, agrees with the BWROG that elimination of RTT will reduce 1) the time that safety systems are out of service or incapable of responding to a degraded plant condition, 2) the potential for inadvertent essential safety function actuations, 3) the complexity of refueling outages, and 4) personnel radiation exposure. The staff has reviewed the BWROG safety assessment of the effects of a five second delay in instrument response and agrees that elimination of response time testing is not safety significant. Based on the above, the staff concludes that elimination of selected RTT as proposed by the BWROG is consistent with 10 CFR 50.36, R.G. 1.118 and IEEE 338-1977 because alternative testing and surveillance will satisfactorily determine instrument response time degradation. Therefore, the staff accepts BWROG Licensing Topical Report NED0-32291, " System Analyses for Elimination of Selected Response Time Testing Requirements," and the letter from T. Green (BWROG) to P. Loeser (NRC), dated April 15, 1994 (Ref. 12), as a basis for
elimination of RTT from TSs for the instruments / components identified.in Table 1 when used in the systems identified in Table 2. When submitting the plant-specific license amendment (TS change) requests, in addition to the request as shown in Appendix I of the topical report, the TS markup tables as shown in Appendix H, and a list of affected instrument loop components as shown in Appendix C.1, licensees should state that they are following the recommendations from EPRI NP-7243 (Ref. 4) and, therefore,.are administratively requiring the following actions: (a) Prior to installation of a new transmitter / switch or following refurbishment of a transmitter / switch (e.g., sensor cell or variable damping components), a hydraulic,RTT shall be performed to determir.e an initial sensor-specific response time value, and (b) For transmitters and switches that use capillary tubes, RTT shall be performed after initial installation and after any maintenance or j modification activity that could damage the capillary tubes. . Licensees must also state the following in their requests: (a) That calibration is being done with equipment designed to provide a step function or fast ramp in the process variable, l l --,--------,--a-a _u__---.au
l (b) That provisions have been made.to ensure that operators and technicians are aware of the consequences of instrument response time degradation, 'and that applicable procedures have been. reviewed and revised as necessary to assure that. technicians monitor for response time degradation during the performance of calibrations and functional tests, (c) That surveillance testing procedures have been reviewed and revised if necessary to ensure calibrations and functional tests are being performed in a manner that allows simultaneous monitoring of both the input and output response of units under test, (d) That for any request involving the elimination of RTT for Rosemount pressure transmitters, the licensee is in full compliance with the guidelines of Supplement I to Bulletin 90-01, " Loss of Fill-011 in Transmitters Manufactured by Ro.tmount," and (e)~ That for those instruments where the manufacturer recommends periodic L RTT as well as calibration to ensure correct function, concurrence is L obtained from the manufacturer that elimination of RTT is acceptable. I i o__________________________________---_--____----______.-----
1 i 1 ) l l
5.0 REFERENCES
i 1. IEEE Std. 338-1977, " Criteria for the Periodic Testing of Nuclear Power i Generating Station Safety Systems." l j 2. Reg Guide 1.118, Rev 2, 1978, " Periodic Testing of Electric Power and Protection Systems," Revision 2, June 1978. 3. ISA-567.06-1986, " Response Time Testing of Nuclear Safety-Related Instrument Channels in Nuclear Power Plants," dated August 29, 1986. 4. Electric Power Research Institute Report NP-7243, " Investigation of l Response Time Testing Requirements," Final Report May 1991. 5. BWR Owners Group Licensing Topical Report NE00-32291, " System Analyses for Elimination of Selected Response Time Testing Requirements," January 1994. 6. NRC Generic letter 93-05, "Line Item Technical Specification Improvement ) to reduce Surveillance Requirements for Testing During Power Operations," September 27, 1993 7. NUREG-1366, " Improvements to Technical Specifications Surveillance Requirements," December 1992 8. Response to the staff questions by letter from T. Green (BWROG) to P. Loeser (NRC), dated March 16, 1994. 9. NED0-31400, " Safety Evaluation of Eliminating the Boiling Water Reactor 'ain Steam Line Isolation Valve Closure Function and Scram Function of the Mair Steam line Radiation Monitor," May 1987, General Electric Company. 10. Memorandum from L. J. Cunningham, Chief, Reactor Protection Branch, " Removal of Reactor Trip Function of BWR Main Steam Line Radiation l Monitor", June 18, 1991. l 11. " Basis for Relaxing ECCS Performance Requirements for BWR4s/6s", EPRI, NSAC-131, September 1988. l 12. Response to the staff questions by letter from T. Green (BWROG) to P. Loeser (NRC),' dated April 15, 1994. ~
L l :. l ATTACHMENT 3 l RESPONSE TO CRGR CHARTER OVESTIONS PROPOSED ACTION: Issue a proposed SER entitled " Evaluation of General Electric Topical Report NED0-32291, ' System Analyses for Elimination of Selected Response Time Testing Requirements'" CATEGORY: 2 RESPONSE TO REQUIREMENTS FOR CONTENT OF PACKAGE SUBMITTED FOR CRGR REVIEW (i) The proposed staff position as it is proposed to be sent out to the BWROG. l Response time testing (RTT) can be eliminated from technical specifications for the selected instrumentation identified in the topical report NED0-32291 can be referenced in license amendment applications with restrictions as discussed in the SER. (ii) Draft staff papers or other underlying staff documents supporting the I requirements or staff positions. 4 None (iii) Each proposed requirement or staff position shall contain the sponsoring office's position as to whether the proposal would increase requirements or staff positions, implement existing requirements or staff positions, or would relax or reduce existing requirements or stsff positions. Use of NED0-3'2291 to modify technical specifications will be a reduction of existing requirements for RTT. (iv) The proposed method of implementation with the concurrence (and any comments) of OGC on the method proposed. The concurrence of affected i program offices or an explanation of any nonconcurrence. ) i 0GC has concurred with the CRGR package and the SER. Its comments i consisted of minor wording changes and are reflected in the attached SER. (v) Regulatory analyses conforming to the directives and guidance of 'NUREG/BR-0058 and NUREG/CR-3568. (This does not apply for backfits that ensure compliance or ensure, define, or redefine adequate protection. l
s 2_ In these cases a documented evaluation is required as discussed in IV.B.(ix).) The SER will allow utilities to reduce their testing requirements by elimination of RTT for some instrumentation and control systems. Implementation of this reductlon is voluntary; therefore, no value/ impact anaysis was performed. (vi) Identification of the category of reactor plants to which the generic requirement or staff position is to apply (that is, whether it is to apply to new plants only, new OLs only, OLs after a certain date, OLs before a certain date, all OLs, all plants under construction, all plants, all water reactors, all PWRs only, some vendor types, some vintage types such as BWR 6 and 4, jet pump and nonjet pump plants, etc.). This SER will apply to BWRs only. (vii) For backfits other than compliance or adequate protection backfits, a backfit analysis as defined in 10 CFR 50.109. The backfit analysis shall include, for each category of reactor plants, an evaluation which demonstrates how the action should be prioritized and scheduled in light of other ongoing regulatory activities. The backfit analysis shall document for consideration information available concerning any of the following factors as may be appropriate and any other information relevant and material to the proposed action: (a) Statement of the specific objectives that the proposed action is designed to achieve; The proposed SER will eliminate some RTT requirements, resulting in 1) reduction of the time that safety systems are out of service or incapable of responding to a degraded plant condition, 2) reduction of the potential for inadvert' nt essential safety function actuatinns, 3) e reduction of the complexity of refueling outages, and 4) reduction of personnel radiation exposure. (b) General description of the activity that would be required by the licensee or applicant in order to complete the action; Licensees that choose to take advantage of the relaxation would review the SER for applicability and submit a technical specification change request, using the SER as justification. (c) Potential change in the risk to the public from the accidental release of radioactive material; j There will be no significant change in the risk to the public because the staff has determined that proper calibration can ensure i
S functionality of the instruments and components listed in the SER. Elimination of RTT will reduce the time that instrument loops are out of service, and will reduce the chance of failure to properly return an instrument to service. l (d) Potential impact on radiological exposure of facility employees and other onsite workers; The radiological exposure to facility employees and other onsite workers will be reduced because elimination of RTT will reduce the frequency that instrument technicians must enter potentially contaminated areas. (e) Installation and continuing costs associated with the action, including the cost of facility downtime or the cost of construction delay; The BWROG estimate is that this change will save at least $50,000 per plant per year in manpower. (f) The potential safety impact of changes in plant or operational complexity, including the relationship of proposed and existing regulatory requirements and staff positions; The elimination of some RTT will reduce maintenance complexity, reduce the time when safety' systems are out of service or incapable of responding to a degraded plant condition, reduce the potential for inadvertent essential safety function actuations, and reduce the complexity of refueling outages. (g) The estimated resource burden on the NRC associated with the proposed action and the availability of resources; The NRC will be required to review the technical specification change request. There were 13 utilities with 17 plants participating in the study; therefore, it is expected there will be 13 requests. It is expected the utilities with similar multiple units will submit multi-plant requests. (h) The potential impact of differences in facility type, design, or age on the relevancy and practicality of the proposed action; The topical and SER consider a variety of specific instruments and system applications, thus each submittal must be tailored to the specific plant. (i) Whether the proposed action is interim or final, and if interim, the justification for imposing the proposed action on an interim basis; This is a final action.
January 10, 1995 MEMORANDUM T0: Jack W. Roe, Director Division of Reactor Projects, III & IV Steven A. Varga, Director I Division of Reactor Projects, I & II Christopher I. Grimes, Chief Technical Specifications Branch Division of Project Support FROM: Jared S. Wermiel, Chief Instrumentation and Control Branch Division of Reactor Controls and Human Factors
SUBJECT:
SAFETY EVALUATION REPORT (SER) APPROVING BWR OWNERS GROUP TOPICAL REPORT NE00-32291, " SYSTEM ANALYSES FOR ELIMINATION OF SELECTED RESPONSE TIME TESTING REQUIREMENTS" Attached is the SER, dated December 28, 1994, approving the BWR Owners Group licensing topical report NED0-32291, " System Analyses for Elimination of Selected Response Time Testing Requirements." It is expected that BWR licensees will be submitting plant specific requests for Technical Specificatation changes based upon this SER. The point of contact in HICB for these plant specific requests will be Sang Rhow. Mr. Rhow can be reached at 504-2826. Please distribute the attached SER to BWR plant project managers for their information and use. l
Attachment:
As stated DISTRIBUTION: l Central File HICB/RF SRhow CSchulten, 0 11-E22 DOCUMENT NAME: A: FWD-SER 95-5 Ta receive a copy of thee docuenent, andcate in the ben:A"jf Copy without ettschment/ enclosure "E" = Copy with attachmentlenclosure
- N' = No copy 0FFICE HICB:DRCH lC SCl:RCB lc BC: $ 8 e L
l l t NAME Ploeser (f JMirV W JWerditT DATE 01/10/95 0J///J /95 01//6 /95 01/ /95 01/ /95 { (/ OFFICIAL RECORD COPY g\\1 --gSo/ 2000 99-J
@ Moo ye fg UNITED STATES 2 E NUCLEAR REGULATORY COMMISSION f WASHINoToN, D.C. 205W0001 %,...../ EVALUATION OF BOILING WATER REACTOR OWNERS GROUP TOPICAL REPORT NED0-32291 " SYSTEM ANALYSES FOR ELIMINATION OF SELECTED RESPONSE TIME TESTING REQUIREMENTS"
1.0 INTRODUCTION AND BACKGROUND
The requirement for periodic testing of reactor trip systems is established in Section 50.55a, " Codes and Standards," of 10 CFR, " Domestic Licensing of Production and Utilization Facilities." This states that " protective systems must meet the requirements set forth in editions or revisions of the Institute of Electrical and Electronics Engineering Standard: ' Criteria for Protective Systems for Nuclear Power Generating Stations,' (IEEE-279)..." In addition, 10 CFR 50.36 paragratih (c)(1)(ii)(A) requires inclusion in the Technical Specifications the limiting safety systems settings for nuclear reactors, those settings "so chosen that automatic protective action will correct the abnormal situation before a safety limit is exceeded." 10 CFR 550.36 (c)(3) " Surveillance requirements" s.ates " Surveillance requirements are requirements related to test, calibration, or inspection to assure that the necessary quality of systems and components is maintained, that facility operation will be within the safety limits, and that the limiting conditions of operation i will be met." The Nuclear Regulatory Commission (NRC),in 1975 implemented a i ENCLOSURE -96b/2000%7-
't . program making response time testing (RTT) a requirement of the technical specifications (TSs). IEEE Std. 338-1975, " Criteria for the Periodic Testing of Class lE Power and Protection Systems," and its later version IEEE Std. 338-1977, " Criteria for the Periodic Testing of Nuclear Power Generating Station Safety Systems," (Ref. 1) provided generic guidance on the conduct of response time verification tests. The NRC staff endorsed IEEE Std. 338-1977 in Regulatory Guide (RG) 1.118, " Periodic Testing of Electric Power and Protection Systems," Revision 2, June 1978 (Ref. 2). Guidance on the l performance of RTT is provided in the Instrument Society of America (ISA) Standard ISA-S67.06-1984, " Response Time Testing of Nuclear Safety-Related Instrument Chanrels in Nuclear Power Plants," dated August 29,1986 (Ref 3). ISA-S67.06-1984 has not been endorsed by the NRC staff, but is widely used by 4 instrumentation technicians in the field. 1 New guidance on the scope of technical specification requirements under 10 CFR 50.36 was promulgated by the NRC with the publication of the Commission's Final Policy S'atement on Technical Specification Improvements in Jiay 1993. RTT is not specifically required by this policy because the essential instrumentation providing indication and actuation functions to mitigate design basis accidents for which RTT elimination has been proposed will l continue to be incorporated in plant technical specifications, and as such, will receive other surveillance to v' ify operability.
{ The Boiling Water Reactor Owners Group (BWR0G) under the auspices of the General Electric Company, issued Licensing Topical Report NE00-32291, " System Analyses for Elimination of Selected Response Time Testing Requirements," (Ref. 5) in January 1994. In NED0-32291, the BWROG utilized the information in EPRI Report NP-7243, " Investigation of Response Time Testing Requirements" (Ref. 4) and proposed the elimination of. the requirements for performance of RTT of selected instrumentation in the Reactor Protection System (RPS), Emergency Core Cooling System (ECCS), and Isolation Actuation System (IAS). Table 1 identifies the instrumentation that is addressed by the topical report. The BWROG concluded that the RTT requirements for the devices identified in Table 1 can be removed from TSs when the devices are used in systems listed in Table 2. The BWROG provided the basis for the proposed elimination of RTT for the instruments / components identified in Table 1 in NED0-32291. The BWROG noted that within a trip function, redundancy exists in individual instrument channels (e.g., 1-out-of-2 twice) and diversity exists in most safety trip functions (e.g., neutron flux, water level, drywell pressure). The BWROG stated that the failures which affect response times tan be detected during other surveillance tests required by current TSs. In addition, to provide a safety perspective on elimination of RTT, the BWR0G evaluated the safety significance of a delayed response using an arbitrarily selected instrument l response time _ delay of five seconds which they believe could be reasonably I u______________________m_
r l l: I Table 1 Instruments / Components for Which Response Time Testing Can Be Eliminated Trip Units Pressure Sensors (continued) Rosemount 510DU and 710DU Barton 764 Differential Pressure Elcrtronic GE Trip Unit 184C5988 Transmitter Rosemount Differential Pressure Transmitter Relays Models 1151,1152,1153 and 1154 3 GE HFA Rosemount Pressure Transmitter Models 1151, j GE HMA 11521153, and 1154 GE HGA SOR Differential Pressure Switch Bourns SOR Pressure Switch Agastat GP/EGP Family ASEA RXMH2 Sensors GE SAT 6004 Barton 760 GE Type CR105, CR120A, CR205, and CR305 Barksdale TC%22-3 3 Potter Brumfield MOR and MDR Barksdale BIT-M12SS-GE Potter Brumfield KH4690 External Devices Flow Devices (Bailey /GEMAC Modules) 560 Alarm Unit 752 Summer 745 Alarm Unit 750 Square Root Extractor Rosemount/GE Trip Unit Noise Suppression Radiation Devices Filter Capacitors Trip Auxiliary Unit (238X697 Series) - Cornell Dublier WBR 2000-50 Indicator Trip Unit (129B2802 Series) - Sprague 500D-35 NUM AC leg Rad Monitor (304A3700 Series) Miscellaneous Devices Pressure Sensors Hi Current isolator 133D9947 Barton 288 and 289 Differential Pressure Optical isolator 204B6186AA Indicating Switches Optical isolator 204B6188AA Barton 763 Gauge Electronic Pressure Transmitter i Table 2 Systerns for Which Respome Time Testing Can Be Eliminated I (1) ECCS actuation instrument loops for the (3) Sensors only, for the following RPS following systems: actuations: (a) low Pressure Core Spray (a) Reactor Vessel Steam Dome Pressure (b) Low Pressure Coolant Injection - High (c) High Pressure Core injection / Spray (b) Reactor Vessel Water level - low, Level 3 (2) Isolation System actuation instrument loops (c) Reactor Vessel Water Level - High, for the following systems: Level 8 (a) RCIC System (b) HPCI System (4) Sensors only, for the following MSIV (c) HPCS System closure actuations: i 1 (d) RWCU System (a) Reacto Water Level 1 or 2 f (e) Primary Containment (b) MSL Radiation High l (f) Secondary Containment (c) MSL Low Presrure (g) RHR Shutdown Cooling / Head Spray
i i . detected by instrument technicians. The BWROG concluded that a 5-second increase in the response time of individual specific trip functions would not have any significant impact on plant safety based on realistic accident evaluations. 2.0 DISCUSSI@ Current Standard Technical Specifications (STS) require nuclear power plants to periodically perform RTT for instrument channels in the RPS, ECCS and IAS. The intent of these tests is to ensure that changes in response time of instrumentation beyond the limits assumed in safety analyses are detected, and. combined with instrument calibration, to ensure that the instrument is operating correctly. The response time tests do not demonstrate,that the instrument response time design value is met, but rather that the specified performance requirements of the TSs are satisfied. The BWROG stated in NED0-32291 that operatior-1 history has show. that significant degradation of instrumentation response times is being detected during the performance of calibrations and other surveillance tests. The BWROG further stated that the performance of conventional response time tests has proven to be of little value in assuring that instrumentation will perform i as required or for determining the health of the instrument because the majority of allowable instrumentation response times are system response times rather than instrument response times. i
. In NED0-32291, the BWROG proposed elimination of RTT for the entire instrumentation loops for the IAS (except main steam isolation valve actuation) and ECCS. The response times for the IAS and ECCS instrumentation are a small fraction of the total system response time requirements (i.e.,10 to 13 seconds for IAS, and 27 to 64 seconds for ECCS). Instrumentation components that may experience response time degradation will continue to respond in the microsecond-to-millisecond range prior to complete failure. Therefore, the BWROG stated that such response time degradation would have no significant adverse effect on system actuation and the instrumentation would continue to meet overall system requirements. For the RPS and main steam isolation valve (MSIV) actuation instrumentation, the BWROG proposed elimination of just the sensor RTT. For the RPS and MSIV actuation instrumentation loops, the overall TS response time requirements are much shorter (0.33 to 2.0 seconds). As a result, changes in instrumentation response time beyond acceptable limits given in TS may not be readily detected during other surveillance tests. To auaress +his, the BWROG referenced EPRI l analyses in NP-7243 as indicating that those failure modes which affect sensor response times beyond acceptable limits can also be detected during calibrations or other tests. In addition, the BWROG i'ndicated that a delay in these trip functions of about 5 seconds, which the BWR0G believes can be detected during other surveillance tests, would not have any significant effect on plant safety and longer delays could be detected during other required surveillance tests, i.
. For the RPS, the three selected trip functions considered for elimination of sensor RTT are: I Reactor Vessel Steam Dome Pressure - High Reactor Vessel Water level - Low, Level 3 Reactor Vessel Water Level - High, Level 8 The typical instrumentation response time requirements for these RPS trip functions range from 0.33 to 1.05 seconds. Only the sensors analyzed by EPRI were considered for RTT elimination. for the IAS instrumentation, the following TS response time requirements were considered for elimination: MSIV Closure Reactor Water Level 1 or 2 Main Steamline (MSL) Radiation High MSL Low Pressure Remaining Isolation Actuation Reactor Core Isolation Cooling (RCIC) System High Pressure Coolant Injection (HPCI) System High Pressure Coolant Spray (HPCS) System Reactor Water Cleanup (RWCU) System
I Primary Containment Secondary Containment Residual Heat Removal (RHR) Shutdown Cooling / Head Spray The BWROG stated that since the IAS instrumentation logic functions are in parallel'with the startup of the diesel generators, the response time requirements (< 10 seconds to < 13 seconds) are not critical for the majority of the isolation circuits. The BWROG stated that this is consistent with the recommendations of Generic Letter 93-05 (Ref. 6) and NUREG-1366 (Ref. 7) for the deletion of RTT where the required time corresponds to the diesel start time. The main steam isolation valve response time requirements (0.5 seconds to 2.0 seconds) are an exception. Therefore, the BWROG request for elimination of RTT for the MSIV actuation circuits covers only the sensors. For the ECCS, the following RTT requirements were considered for elimination: Low Pressure Core Spray (27 to 43 sec.) Low Pressure Coolant Injection (37 to 64 sec.) High Pressure Core Injection / Spray (27 to 35 sec.) The BWROG stated that ECCS instrumentation response time requirements range from 27 to 64 seconds and, therefore, actual instrument response which is of much shorter duration is not important to meeting these times. Unlike the IAS, the diesel generator, emergency cooling pumps, and the injection valves i N
. response times are in series with the instrumentation response time. For these loops, only the instrumentation will be eliminated from RTT. The overall ECCS system response time requirement, which includes diesel generator, injection valves, pumps, and other components, still applies. The diesel generator and the injection valve TS response time requirements are not eliminated. l The BWROG indicated that the fundamental approach developed in their study for RTT elimination is consistent with the guidelines of RG 1.118, Revision 2, which endorses IEEE 338-1977, and is stated as follows: " Response time testing of all safety-related equipment, per se, is r.ot required if, in lieu of response time testing, the response time of the safety equipment is verified by fune.t'ional testing, calibration checks or other tests, or both. Tnis is acceptable if it can be demonstrated that changes in response time beyond acceptable limits are accompanied by changer, in performance characteristics which are detectable during routine periodic tests." The BWROG stated that changes in instrument response are detectable prior to significant degradation without the performance of conventional RTT, and will therefore fulfill the criteria of IEEE 338-1977. The BWR0G also indicated that the testing which is currently required by TSs, that is, calibrations, functional testing, logic system functional testing and channel checks will
' identify any significant degradation of instrument response time. A description of each test, including STS definitions, and the relative advantages of each method for detection of degradation was included in Appendix B to the BWROG topical report. Although this description did not delineate which test detects which response time failures, or the manner of detection, it did show that response time testing of individual instrument channels had the longest time interval between performance of surveillance activities, 36 months, when compared to calibrations, functional tests, or channel checks, which are performed at the same or more frequent surveillance intervals of 18 months, quarterly or monthly, and once/ shift respectively. The BWROG selected Fermi-2 and River Bend as lead plants to represent typical BWR/4 and BWR/6 plants respectively. TS RTT requirements ("RTT Trip Functions") were identified for the lead plants, and a detailed listing of specific loop components was compiled for each loop. All componente in the r instrumentation loops which could potentially affect the loop response time were identified, and failure mode evaluations were conducted in order to determine whether failures could affect response time. In those cases where the BWROG determined that a failure mode could affect the instrument loop response time, the consequences of such a failure were evaluated. Analyses were then performed to determine whether other surveillance testing would
I : identify these potential response time degradations. The BWROG determined l that other surveillance tests (i.e., calibration) could identify these I degradations. In order to ensure that the calibration effectively establishes acceptable instrumentation performance in lieu of a specific RTT, it should include the following features: (1) By inputting a standard value over the operating range, the I i technician can observe any abnormal char.ges in calibration / functional span and operational range of the instrumentation. (2) By observing the expected output when an input signal is injected, l l the technician can determine that +he instrumentation is i responding properly. (3) By checking calibration / functional points between zero and 100% of
- span, l
(a) the technician ensures that the instrumentation responds instantaneously with the injection of a test signal and,
f . (b) the technician ensures the instrumentation performs in accordance with the instruments' design characteristics l (i.e., linear or logarithmic). (4) By observing the response from a trip signal initiated during a functional test, the technician is able to detect a response time delay in the actuation of an alarm, light, or relay contact status. The BWROG stated that an evaluation of the affected loop components that exhibit degraded response time during calibrations and functional tests would be performed before the loop is returned to service. The BWROG stated that as part of the failure mode evaluations, component experts and vendors were contacted to assist with and verify the analysis. No documentation or letters from these experts and vendors were presented in direct sunnort of the elimination of RTT-The BWROG stated that due to variations in applications and licensee surveillance practices, instrumentation manufacturers were unwilling to make any recommendations concerning acceptability of calibration or other surveillance as a basis for the elimination of RTT. The BWROG performed a review of component failure experience by conducting BWR-specific surveys, and by researching the Nuclear Plant Reliability Data System, NRC Bulletins, NRC Information Notices, and GE Service Information
, l Letters. This review was used to determine if actual RTT failures had been detected by surveillance testing other than RTT as indicated in this l l l information. Appendix 0 of the topical report summarizes the results of this 1 review, and shows that in most cases, failures which have an impact upon response time had been or could have been detectable by some other type of surveillance test. The BWR0G also referred to the work documented in EPRI Report NP-7243, " Investigation of Response Time Testing Requirements," dated May 1991 to support RTT elimination. This report evaluated failure modes and effects analyses for many of the sensors employed in the applicable BWR instrumentation loops. However, because not all components and sensors that the BWROG wanted to exempt from response time testing were covered in the EPRI report,'the BWROG conducted supplemental evaluations as they considered necessary. To provide a safety perspective on the e'imination of RTT, the BWROG evaluated the safety significance of delayed response times. The BWROG surveyed instrumentation departments at participating BWR plants and at selected pressurized-water reactor plants. The survey results indicated that most technicians believed they could detect delays of 5 seconds or greater. Five seconds was generally well within the required system response time allowable limit specified in plant technical specifications. Based on the survey results, the BWR0G realistically evaluated the effect of response time
. degradation in order to provide a measure of safety significance. The details of these evaluations were included in Section 5.0 and Appendix J of the topical report. 3.0 flALUATION The primary argument provided in the BWROG topical report in support for the y elimination of RTT is that appropriate alternatives to RTT are in place per the criteria of R.G. 1.118 and IEEE 338-1977, Section 6.3.4, paragraph 3 (page 11), which states: " Response time testing of all safety-related equipment, per se, is not required if, in lieu of response time testing, the response time of the safety equipment is verified by functional testing, calibration checks or other tests, or both. This is acceptable if it can be demonstrated that changes in response time beyond acceptable limits are accompanie.i by changes in performance characteristics which are detectable during routine periodic tests." The BWROG referenced EPRI analyses in NP-7243 as indicating that failure modes which affect sensor response times beyond acceptable limits can be detected during calibrations or other tests. Staff review of EPRI Report NP-7243 and the additior.al failure analysis contained in the WROG report determined that the data contained in the databases were generally useful in showing that the i l l
1 - I5 - great majority of compenent failures which can effect response times will also ' impact the calibration or functional test results of the component (however, the databases were inconclusive with respect to the merits or usefulness of RTT). The BWROG report indicated that instrument response degradation of less than 5 seconds has little safety-significance. The BWROG stated that it is likely that instrument response delays of more than 5 seconds will be noted by the instrumentation technicians, and therefore, it was concluded by the BWROG that elimination of RTT would have no significant impact on plant safety. The staff concurs with the BWROG that delays of about 5 seconds for the selected RPS, ECCS, IAS and MSIV functions would have no significant effect on plant safety and longer delays can be detected by other testing, primarily calibration, if properly performed. Thus, the guidance in IEEE 338 is satisfied by performance of calibration as an alternative to specific RTT for delays greater than five seconds. The staff also believes that calibrations should be conducted in accordance with the manufacturers' recommendations to provide a good general indication of the proper functioning of the instrument and channel. If the manufacturer recommends periodic RTT as well as calibration tt, ensure correct function of the instrument or device, the licensee must ensure that elimination of RTT is i nevertheless acceptable for the particular application: involved.
.- There are known concerns with the elimination of RTT for certain components. The most notable of these is with regard to Rosemount oil-filled pressure transmitters. The BWROG has stated in Chapter 8 of the topical report: " Loss of fill oil in Rosemount transmitters is the subject of NRC Bulletin 90-01 and the associated supplement as well as Rosemount Technical Bulletins 1 through 4. Performance of response time tasting is not the preferred method to detect loss of fill oil. 1 Licensee actions to address this failure mode have been taken i pursuant to NRC Bulletin 90-01 and the associated supplement." i The staff has concluded that elimination of RTT based upon this SER does not relieve licensees of the obligation for satisfactory compliance with the recommendations of Supplement 1 to Bulletin 90-01 for enhanced surveillance monitoring or any other commitments made with respect to Rosemount oil-filled' pressure transmitters. A second concern is associated with crimped capillaries in instruments that use capillaries, which have been shown to affect response time. The staff believes that performing an initial RTT at the time of installation is sufficient to detect the occurrence of crimped capillary tubing or other problems with instrument performance. Degradation in capillaries over time generally can be detected with the proper calibration methods as discussed below. However, in some cases, capillary tubing is not included in L _ _ _ _ __ ___________ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
~
- calibration testing.
For those cases, the staff has concluded that it will be necessary to perform capillary tube testing after any maintenance that might damage the capillary tube. The BWROG stated that the instrumentation technicians believe they can detect an instrument with a sluggish response within five seconds. However, this was not substantiated by testing actual performance. The staff agrees that it is not unreasonable to expect plant personnel to note a deviation from the standard response of an instrument if the calibration or test is being done in a manner where a single technician can both provide the input to the channel being tested and observe the results of that input. If, however, the calibration or test is being conducted in a manner where a single technician cannot provide the input to the channel being tested and observe the results of that input, a second technician is required. An example of this may be the i test of a trip unit, where the stimulus is applied at one end of the equipment, and the trip is observed on the other end of the cabinet, or in a different room. In this case, two technicians in communication with each other would be requircJ. When more than a single individual is involved, the individual observing the response may be less sensitive to delays on the order of seconds since there may be communications delays. Thus, in some cases, the l techniques used for calibrations, functional tests, or channel checks may require modification to ensure that the required observation of instrument response is possible. i l L
. The staff notes that an additional important issue is the detection method used in the performance of the calibration, If the calibration is done with equipment with an inherent time delay, such as a slow ramp in the process variable, an additional delay of about 5 seconds in actual instrument response may be difficult to notice. With a slow ramp in the process variable, a sluggish response can be mistaken for a change in setpoint. If, however, the calibration is done with the calibrating equipment using a step function or a fast ramp of significantly less than five seconds, the recognition of a delay of about 5 seconds in instrument responst is reasonably assumible. A further poter.tial problem inherent in the manner in which cailbrations are done is that the sensor is almost always calibrated separately from the trip circuitry. Since both the sensor and trip unit could have undetected delays on the order of five seconds, it may be possible for any one function to have double the delay. However, the staff does not consider this a significant safety concern, because a response time failure without a corresponding calibration frilure is unlikely, and the possibility of two such fa' lures occurring in a single channel is even more unlikely. While the BWROG has not presented a statistical determination of this possibility, the staff believes the probability is very small. l In Appendix J of the topical report the BWROG provided a safety assessment of a delayed or sluggish instrumentation response on the order of five seconds for the trip functions selected for RTT elimination. The staff reviewed the effect on plant safety of a delayed instrumentation response on the order of
.- five seconds for the trip functions selected for RTT elimination, and agrees with the BWROG that a five second delay would not affect the capability of these systems to initiate reactor trip or to provide the required core cooling function. This includes the RPS and MSIV actuation instrumentation functions with overall response time requirements of 0.33 to 2.0 seconds. The staff concurs with the BWROG's conclusion that significant margin exists in the analysis assumptions such that an additional response delay in the selected trip functions on the order of 5 seconds has minimal safety significance. In order to guard against longer delays, licensees taking advantage of this-elimination of RTT must ensure that operators and technicians continue to be aware of the consequences of instrument response time degradation and are familiar with the alternative means for detecting this degradation. Applicable plant-specific procedures may need to be revised to assure that technicians monitor for response time degradation during the performance of calibrations and functional tests. The staff notes that requirements related to the operability, applicability, and surveillance requirements, including performance of testing to ensure response times, for RTS and ESFAS systems are retained due to those systems i importance in mitigating the consequences of an accident. However, the elimination of response time testing requirements does not constitute a l condition or limitation on operation necessary to obviate the possibility of I an abnormal situation or event giving rise to an immediate threat to the l ' - ' - ' - ' ~
public health and safety, in that the ability of the RTS and ESFAS systems to l perform their safety functions are not adversely impacted since the response times of the devices are verified by other means. Specifically, to meet the guidance of R.G. l.118, RTT is needed unless it has been shown that the response time of the. instrument or channel will be verified by other means. The staff has determined that calibration and other t surveillance testing, combined with technician awareness of the RTT requirement, will adequately ensure that the response time is verified for the devices identified in Table I when the devices are used in systems listed in Table 2.
4.0 CONCLUSION
As noted above, IEEE Std. 338, as endorsed by RG 1.118, states that RTT is not required if in lieu of response time testing, the response time of the safety . equipment is verified by functional testing, calibration checks or other tests, and if it can be demonstrated that changes in response time beyond acceptable limits are accompanied by changes in performance characteristics which are detectable during routine periodic tests. The following response time tests were evaluated for elimination by the BWROG:
- (1)
All ECCS actuation instrument loops (entire channel). (2) All isolation system actuaticn instrument loops (entire channel) except for MSIVs. (3) Sensors only, for selected RPS actuation instrument loops. -(4) Sensors only, for selected MSIV closure actuation. Based on its review of the information presented by the BWROG, the staff agrees with the BWROG that, significant degradation of instrumentation i response times (i.e., delays greater than about 5 seconds) can be detected during the performance of calibrations and other currently required surveillance tests; furthermore, delays on the order of 5 seconds do not have any significant impact on plant safety. This includes the selected RPS and MSIV actuation features which have overall TS response time requirements of 0.33 to 2 seconds. Thus, the staff concludes that the other existing surveillance requirements for the instrumentation provide confidence that the safety function of the plant instrumentation will be satisfied without the need for a epecific RTT. The staff, therefore, accepts BWROG Licensing Topical Report NED0-32291, " System Analyses for Elimination of Selected Response Time Testing Requirements," and the letter from T. Green (BWROG) to P. Loeser (NRC), dated April 15, 1994 (Ref. 12), as a basis for elimination of RTT from ISs for the l instruments / components identified in Table 1 when used in the systems identified in Table 2.
1 i , When submitting the plant-specific license amendment (TS change) requests, licensees must confirm the applicability of the generic analysis of NED0-32291 ) to their plant, and in addition to the request as shown in Appendix I of the i topical report, the TS markup tables as shown in Appendix H, and a list of i affected instrument loop components as shown in Appendix C.1, licensees should state that they are following the recommendations from EPRI NP-7243 (Ref. 4) and, therefore, are requiring the following actions: (a) Prior to installation of a new transmitter / switch or following l refurbishment of a transmitter / switch (e.g., sensor cell or variable damping components), a hydraulic RTT shall be performed to determine an initial sensor-specific response time value, and (b) For transmitters and switches that use capillary tubes, capillary tube testing shall be performed after initial installation and after any maintenance or modification activity that could damage the capillary tube:.. Licensees must also state the following in their requests: (a) That calibration is being done with equipment designed to provide a step function or fast ramp in the process variable, i a
1 1 L. l l I l l (b)- That provisions have been made to ensure that operators and technicians, through an appropriate traini..g program, are aware of the consequences l ~ of instrument response time degradation, and that applicable procedures have been reviewed and revised as necessary to assure that technicians monitor for response time degradation during the performance of calibrations and functional tests, (c) That surveillance testing procedures have been reviewed and revised if l necessary to ensure calibrations and functional tests are being performed in a manner that allows simultaneous monitoring of both the input and output response of units under test, (d). That for_ any request involving the elimination of RTT for Rosemount pressure transmitters, the licensee is in compliance with the guidelines of Supplement I to Bulletin 90-01, " Loss of Fill-Oil in Transmitters Manufactured by Rosemount," and (e) That for those instruments where the manufacturer recommends periodic RTT as well as calibration to ensure correct functioning, the licensee has ensured that elimination of RTT is nevertheless acceptable for the particular application involved.
,? n
5.0 REFERENCES
1. IEEE Std. 338-1977, " Criteria for the Periodic Testing of Nuclear Power Generating Station Safety Systems." 2. Reg Guide 1.118, Rev 2, 1978, " Periodic Testing of Electric Power and Protection Systems," Revision 2, June 1978. 3. ISA-S67.06-1986, " Response Time Testing of Nuclear Safety-Related Instrument Channels in Nuclear Power Plants," dated August 29, 1986. 4. Electric Power Research Institute Report NP-7243, " Investigation of Response Time Testing Requirements," Final Report May 1991. 5. BWR Owners Group Licensing Topical Report NED0-32291, " System Analyses for Elimination of Selected Response Time Testing Requirements," January 1994. 6. NRC Generic Letter 93-05, "Line Item Technical Specification Improvement to reduce Surveillance Requirements for Testing During Power Operations," September 27, 1993 7. NUREG-1366, " Improvements to Technical Specifications Surveillance Requirements," December 1992 8. Response to the staff questions by letter from T. Green (BWROG) to P. Loeser (NRC), dated March 16, 1994. 9. NEDO-31400, " Safety Evaluation of Eliminating the Boiling Water Reactor Main Steam Line Isolation Valve Closure Function and Scram Function of the Main Steam Line Radiation Monitor," May 1987, General Electric Company. 10. Memorandum from L. J. Cunningham, Chief, Reactor Protection Branch, " Removal of Reactor Trip Function of BWR Main Steam Line Radiation Monitor", June 18, 1991. 11. " Basis for Relaxing ECCS Performance Requirements for BWR4s/6s", EPRI, NSAC-131, September '988. 12. Response to the staff questions by letter from T. Green (BWROG) to P. Loeser (NRC), dated April 15, 1994, 1 i l !}}