ML20155G303
| ML20155G303 | |
| Person / Time | |
|---|---|
| Issue date: | 11/05/1998 |
| From: | NRC (Affiliation Not Assigned) |
| To: | |
| Shared Package | |
| ML20155G292 | List: |
| References | |
| GL-83-28, NUDOCS 9811090013 | |
| Download: ML20155G303 (16) | |
Text
, _ _ _ _ - . _ . _ . . . _ _ _ _ . _ - . _ . . _ . , _ _ . . . . _ - . . - _ _ . _ _ _ _ . . _ . . . _ . _ _ . _ . . _ _
b
+#pm ** % o
'. UNITED STATES l ['
5 h
8 NUCLE AR REGULATORY COMMISSION WASHINGTON, D. C. 20555
%...../ 1 l ENCLOSURE 1 SAFET) EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION REVIEW OF BWR OWNERS GROUP REPORTS NEOC-30844 AND 30851P ON JUSTIFICATION FOR AND EXTENTION OF ON-LINE TEST INTERVALS AND ALLOWABLE OUT-OF-SERVICE TIMES FOR BWR REACTOR PROTECTION SYSTEMS '
l l 1.0
SUMMARY
l The staff has reviewed the General Electric Company (GE) Topical Reports l NE00-30844, "BWR Owners Group Response to NRC Generic Letter 83-28 " and NEDC-30851P, " Technical Specifications Improvement Analysis for BWR RPS." '
These reports were issued by the BWR Owners Group to respond to Generic Letter
'83-28, Item 4.5.3 and to support the proposed extension of reactor protection system (RPS) on-line test intervals and allowable out-of-service times (A0Ts)
.for RPS test and repair. The staff has concluded that the analyses presented
.in the Owners Group reports are acceptable for resolving these issues, subject to the limitations and conditiens presented herein.
2.0 BACKGROUND
Or, February 25, 1983, both scram circuit breakers at Unit 1 of the Salem Nuclear Power Plant failed to open upon receipt of an automatic RPS signal. The l operstor terminated the incident about 30 seconds after the initiation of the l
automatic trip signal. The failure of the circuit breakers in this incident was related to sticking of the under-voltage trip attachment. Earlier on February 22, 1983, at the same unit, a steam generator low-low level during
- plant startup resulted in an automatic trip signal. In this case, the reactor
.was trippeo manually by the operator almost coincidentally with the automatic
- trip.
9811090013 981105
l t
l 2 i
l Un February 28. 1983, the NRC Executive Director for Operations directed the i
staf f to investigate and report on the generic implications of these uccurrences. The results of the staff's assessment are in NUREG-1000. " Generic implications of the ATWS Events at the Salem Nuclear Power Plant" (Ref. 1).
Subsequently, the staff issued Generic Letter 83-28 (Ref. 2) requesting that all licensees of operating reactors, applicants for an eperating license, and
~
! holoers of construction permits respond to gener1c issues raised by the analyses of these two anticipated transient without scram (ATWS) events.
Iten. 4.5.3 of this generic letter requested that licensees ar.d applicants review the existir,9 RPS on-line functional test intervals required by Technical Specifications (TS). They are to ensure that current and proposed intervals fur such testing are consistent with a goal of achieving high RPS availability l censidering uncertainties in component failure rates, uncertainties in comon ncde feilure rates, reduced redundancy during testing, operator errors during j testing, and corponent wear caused by the testing.
3.0 APPROACH The BWP Owners Group decided to attempt to resolve these issues generically.
l It comissiorco GE to perform generic analyses and apply the generic results to the individual boiling water reactor (BWR) plants. (The generic analyses are arplicable tu a vast majority of plants that have a relay RPS as well as to the
! rest of the plants that have solid-state RPS.)
Two GE tcpical reports were issued: (1)NEDC-30844(Ref.3),whichanalyzeda representative EvR plant and provided a technical basis for ensurir.g that the
- l. current RPS on-line test intervals meet the recomendations of Generic Letter 83-28. Item 4.5.3, and (2) NEDC-30851P (Ref. 4), which used the base case results from NEDC-30844 to establish a basis for extending the current RPS on-line test intervals and A0Ts. These reports used reliability analyses with fault tree modelling to estimate RPS failure frequency. Sensitivity analyses
, were used to vary the factors that represented the five areas of cuncern delineoteo in item 4.5.? so that their impact was considered appropriately.
The acceptance guideline used by GE for the TS changes is baseo on net increase
. l I
1.
l 3
l l l l
In risk, which'is the difference between the increase in risk that would result f rom the TS changes uno the decrease in risk that would result fror the reduced l 1
likelihood of inadvertent scrams ur exceeding thF U miting. Conditions of l l
l_ Operations. If the net change in risk is determined to be insignificant, the i TS changes would be acceptable. To apply generic plant analyses to specific I
plants, GE collected necessary infortation on the RPS fur each BWR, determined the differences for each plant, and analyzed the effect of each identified l- difference en the RPS foilure frequency (this precedure is cescribed in Appenoix K to NEbC-308 SIP).
1 4.0 HP.C ACTION '
- l. The staff engaged the services of Idaho Nuclear Engineering Lateratory (II:ll) to review the data and methodology used in the two GE reports. This review was l to determine the validity of using the RPS failure frequency as o risk measure; to assess the adequacy of the fault tree analyses and the supporting obtat and te determine reliability calculations adequacy using the W CUT (Ref. 5) and-j FRANilt 111 (Ref. 6) ccmputer codes. INEL notes that, based on conservative assumptions, the estimLted increase in RPS unavailabilit) cue to the prcposed TS changes would contribute a very small increase to estiranted core-melt frequency. Huwever, if the benefits due to the TS changes are taken into account such asithe reduction in the number of inadvertent test-induced scrans.
the net change in risk resulting from the TS changes would be considered insignificant. Using this reasoning, the staff agrees with INEL'on the basis j for acceptance. INELissuedaTechnicalEvaluationReport(TER)(Ref.7) i presenting the o* tails'and results of its review.
l-
- On the basis of its review of the TER, the staff endorses the conclusion that
. . the methodology used ano results obtained in the two GE reports were verified l
for the relay RPS, but were not verified for the solid-state RPS. The staff. -
( therefore ' finds the analyses presented for the relay RPS in NEDC-30844 and hEDC-30851P acceptoble to support a determination that the current or.-line RPS test intervals are consistent with the high RFS availability required by p Generic Letter 83-28, Item 4.5.3. The staff also finds the use of the analyses I
w - -
4 h
bcceptable for supporting the proposed extensions to TS test intervals ana
-increases in A01s.
However. these fincings are limited to plants using GE BWR relay RP5s. The staff review of' additional information and analysis raust be conipleted befcre a position on the solid-state RPS can be provided. The staf f's findings are discussed in the following sections.
5.0 SPECIFIC COMMENTS ON THE TOPICAL REPORTS AND THEIR RESULTS 5.1 Methodology and Data
-The GE analysis evaluates the RPS TS using the RPS failure frequency as the risk measure. In essence, the GE analysis evaluates the changer, in the RP5 failure frequency as a result of the changes in the RPS TS (for example, the changes in the RPS functional testing intervals and in the A0Ts for repairing or testing RPS channels). A brief discussion of the methodology used to calculate the RPS failure frequency follows.
l The calculations of RPS failure frequency depend on two sets ~of parameters.
The-first set consists of initiating events that eventually ledd to actuation of the RPS. The second set consists of "RPS unavailabilitits." which are the prcbabilities that the PPS is unavailable given the demands for RPS actuation.
l Depending on each initiating event, the number of sensors that could actuate the RPS.would vary. Therefore, the RPS unavailability for one ir.itiating event may differ from that for another.
For each initiating event. GE developed a fault tree to quantify RPS unavailability per demand. The f ault tree models the logical relationship of the faults that may contribute to RPS unavailability. The logical representation of the fault tree was used with the computer code WAMCUT to l obtain the dominant cutsets which are the combinations cf faults that cause the RF5 to be unavailable. The dominant cutsets, together with information on l'
y --4 -W P
l I
5 1
l testing and repairis.g the RPS, are then used with the computer code ThAhTIC 111, which calculates ths unavailability of RPS per demand. i To determine the adequacy of f ault tree analyses for estimating RPS l unavailabilities, INEL used the computer code COMCAN III (Ref. 8) to generate cutsets. INEL then compared these cutsets with the cutsets generated by l
WAMCUT, which GE used. In general, INEL detennined that the cutsets obtainec by GE are adequate for estimating the relay RPS unavailabilities. In addition, INEL determined the overall date used in the GE relay RPS analyses arc valid. !
The ir itiating event frequencies used in the GE analyses are mostly in agreement with thuse from other reputable data sources (Ref. 9.10,11). IriEL found that a higher estimate fur pressure regulator failure frequency should be used. However, the use of this higher value would result in only a very small
' increase in overall PPS failure frequency and is not considered significant. '
With respect to component iailure rates used in the GE relay RPS analysis, ll4EL determined that the tailure rates are acceptable for estimating EFS unavailatilities. However, INEL could not verify the random failure retes for the solid-state RPS the basis of the information and reference in the GE reports. In acidition, INEL found that GE used a simplifica fault tree model ;
'for the sclid-state RPS to do WAMCU1 computer runs instead of the fault tree proviced in Appencix 1 of the NEDC-30BblP report. GE did not provide any docun,entation for simplifying the fault tree. The staff agrees with the INEL finoings. On this basis, the staff asked and obtaineo additional information regarding the sulid-state RPS reliability analyses from GE to complete our review.
5.2 Uncertainties in Component Failure Rates l GE has perfonned sensitivity analyses to determine the sensitivity of RPS unavailability to uncertiin ies in component failure rates. To do this, GE I nultiplied the componer.t failure rates with an error factor, which is the ratio cf the upper uncertainty bound and the median value. The re:,ults indicate that uncertainties in the component failure rates have a negligible impact or PM I-t
l 6 I l
l i
unavailability. Therefore. GE concluded that the RPS unavailability is not
. sensitive to.the uncertainties in component. failure rates.
Tc verify GE's conclusion, INEL performed a sensitivity run on the main steam isolation valve (MSIVL closure event. INEL determined that RPS unavailability was ebout the sanic as the RPS unav0ilability for the base case value.
Therefore, the staff concludes that uncertainties in component feilure rates do not significantly affect RPS unavailability and, hence, the failure' frequency.
5.3 Comon Cause Failure Rates GE deterinined that the common cause failure rates of the scram contactors do contribute significantly tc the RPS unavailability for each of the initiating events. GE contends that even when these ccmmon cause failure rates are considered, the results are still lower than other published results.
The staff notes that the proposed changes to the technical specifications would increase to weekly the frequency of testing of the scram contactors by means of the manual scram / test switches in the control room. The staff believes that this is an efficient way of detecting the common cause failures
'of scram contactors.
5.4 Component Wear Out Caused by Testing l
The GE analysis indicates-that among the components in the RPS, the scram !
contactors are_de-energized whenever an individual sensor and its associated relay are tested. Because 11 different types of sensors are tested while the reactor is at full power, the scram contactors would be challenged more ofter than other components in the RPS. For this reason, GE examined the effect of scram contactor wear out caused by testing. The GE analysis indicates that scram contactor wear created by the number of tests required by current TS does not cause any significant increase in RPS failure frequency.
I
4 1
1 7
l Te verify the GE analysis, INFL performed a sensitivity run using the computer code FRANTIC !!!. Thc data it it to the code included a factor that increased the failure rate after each test. The INEL results correspond.to the GE l results and indicate that the RPS unavailability is not significantly impacted by scram contactor wear created by testing.
According to GE, no scram contactors have failed to date and thf.re is ne indication that wear out is becoming a potential problem. With RPS sensor channel testing reduced from monthly to quarterly, the frequency of RPS corpenent actuation would be reduced and RPS components would be less likely to fail due to wear.
Therefore, the staff concludes that RPS unavailability is not sigrificantly impacted by wear of componert due to the proposed test interval.
! 5.5 Sensitivity to Reduced System Redundancy During Testing l
GE examined the irnact of reduced system redundancy during testing en the RPS unaveilabilities by comparing two cases. In the first case, a sensor channel L is "jurpered" during a test and is unable to provide an RPS signal upon actual demand. In the second case, a sensor channel is placed in trip during a test and thus does provide an RPS signal.
l l
The PPS unavailability for the first case is higher than that for the second case. Hon ver, the difference between these two RPS unavailabilities was found i
. to be small and indicates that reduced redundancy during testing has no
_ significant irrpact on RPS unavailability. As an audit of the GE analyses, INEL l performed a sensitivity analysis that showed that a sensor can be unavailable for 3 months without having a significant impact en RPS unavailability.
Therefore, the staff concludes that reduced system redundancy as a result of testing does not have a significant ef fect on RPS unavailabilities.
l l
l
1 l
8 l l
I 5.6 Sensitivity to Human Error Rates During Testinu l
The GE analysis considered two types of human errors during a test: an operator disabling components randomly during a test and an operator causing a common- cause f ailure of all similar components during a test. The GE analysis determined that the first type of operator error does not have significant impact on RPS failure frecuency, but the second type of operator error does.
The GE analysis determined that operator error disabling all scram contactors was the biggest contributor to the RPS failure frequency. Under the proposed TS requirements, the operators will perform channel functional testing of the I manual scram on a weekly basis by actuating the channel manual scram / test switches in the control room. Therefore the staff believes that the weekly testing of manual scram is an efficient way of detecting the common cause failures of scram contacters due to operator errors.
5.7 Changes in RPS Surveillance Testing intervals GE cciculated RPS failure frequency by varying the surveillarce testing intervals of the average power range monitor (APRM) and other sensors from monthly te quarterly. The results showed a small change in RPS failure frequency.
To verify the GE analysas. INEL used the MSIV closure event and varied the testing intervals as above; the !NEL results were the same as the GE results, in view of the small change in RPS failure frequency as a result of the change in test intervals, the staff concludes that the proposed changes in testing intervals are acceptable.
5.8 Changes in Alloweble Outage Times for Test and Repair l The current TS for the relay RPS allow A0Ts of 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> for repairing and ? hours for testing a single sensor channel without placing the channel in a tripped
l
- l l
state. GE believes that the short times allowed by the current TS may cause operator error as a result of stress during repair and testing. GE alsc contends that placirg an individual channel in a tripped condition when repairs and tests cannot be completed within the alloweble outage time may increase the likelihood of an inadvertent scram. Therefore, GE proposes to extend the A0Ts f or repair and for testing. The proposed A0Ts, which are based on the average tir*s needed tc complete trsts and repairs, include sufficient time marqins so that the operators wculd not be placed under undue stress. To support these above changes, GE performed sensitivity analyses and concluded that changing of the A07s had a negligible impact on RPS failure frequency.
INEL audited these GE calculations, using FR/.NTIC III, to analyze the MSIV closure event te account for the extended test and repair times. The INEL results verify GE's fir. dings. On this basis, the staff concludes that the proposed A0Ts for repair and test tires of I hour and 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> fer individuci relay RFS sensor channels can be extended to 12 huurs and 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> respectively.
L.9 Plant-Spccific Application of Generic Results The GE analyses considered the impact en RPS unsvailability resulting from the
. differences in systems / components among the various BWR plants from BWR/3 to BWR/6. GE determined that if the proposed RPS TS changes are implemented, there woulc be no significant increase of RPS failure frenuency for the reviewed FWR plants that use the relay RPS. This determination is based on use of the GE procedure given in Appendix K cf NEDC-30851P for evaluating specific plants against the generic RPS design and analyses.
In general, the staff finds the GE procedure valid for making such plant-specific comparisons. However, the staff cannot verify that the generic results apply to a specific plant without performing (1) detailed comparisons of plant-specific designs with the generic design and (2) sensitivity analyses.
The staff also notes that there is diversity among the analog trip units used in BWRs. The GE reports do not confirm that calibration of the analog trir
l 10 units can be extended from monthly to quarterly without creating excessive drift. In addition, the reports do not supply drift information 'er othar
{
instrumentation used in the PPS. The staff has given e list of plant-specific l 1
conditions licensees must meet to close out item 4.5.3 of Generic Letter 83-28 end to make the plant-specific RPS TS changes. These conditions, listed in 1
Table 1 include a demonstration of adequate drift characteristict..
J As for applying the generic results of the solid-state RPS analyses, the staff asked for and obtained additional information so that it could complete the assessment of the solid-state RPS reliability analyses.
6.0 CONCLUSION
Based on the INEL findings, the staff concludes that the GE analyses demonstrate general compliance with Item 4.5.3 of Generic letter 63-28 for the fdtilities using the relay RPS. In addition, the staff finds that the proposed changes in the. TS for the relay RPS are generally acceptable. With respect to plent-specific implementation of changes in the RPS TS for a plant with a relay PPS. Table I lists piant-specific conditions that each licensee or applicant must meet to complete the resolutier, of Item 4.5.3 and make any proposed TS changes fully acceptable. Further, the results of solid-state RPS review will be provided ir,a future staff Safety Evaluation Report.
7.0 REFERENCES
- 1. US Nuclear Regulatory Comission, " Generic Implications of the ATWS Events at the Salem Nuclear Power Plant, Vols. I and E. NUREG-1000, April 1983.
- 2. Eisenhut, D. G., NRC letter to All Licensees of Opereting Reactors.
Applicants for Operating License, and Holders of Construction Pertnits,
" Required Actions Based on Generic Implications of Salem ATWS Events" (Generic Letter 83-26), July 8, 1983.
me - p
I 11 1
- 3. S. Visweswaren et al., "8WR Owners' Group Response to NRC Generic Letter 83-28, item 4.5.3," General Liectric Company, NEDC-30844 January 1985.
l 4
W. P. Su111 var. ct al., " Technical Specification leprovement Anelyses for l BWP Peactor Protection System," Gener61 Electric Company, NEDC-308 SIP, May 1985.
- b. R. C. Erdr: ann, F. L. Leverenz, and H. Kirch, "WAMCUT; A Cor.ruter Cods
, it,r Fault free Evaluatict.s " EPRI NP-803, Science Applications, Inc.,
June 1970.
(. T. Ginzburg et al., "FPANTIC 111; A Computer Ccde for Tirnt-Depender.t ke11 ability Analysis." Bruukhas n National Laboratory and Science Applications, Inc., April 1984
- 7. B. Colltr.s et al., "A Review of the BWR'0wners Group Technical Specification Improvement Analyses for the BKP Peactor Prutection Syster, " EGG-EA-71CE, corporate pubitsher January 1986.
l T. C. M. Rase.uson et al., "CCPCAN !!!; Use of COMCAN 111 in Syster. Design acd Reliability Analysis, " EGG-2187 corporate publisher October 19BP.
- 9. A. S. McC1ymont et al. , ATWS: A Reappraisal Fart 3: Fr(quency of Anticipated. Transients," EPRI hP-2230, Science Applications, l..c... ,
Jar.uary 1902.
- 10. M. McCann et al., "Probabilistit. Safety Ar.alysis Procedures Guide ."
NukEG/CR 2815. Vol. 1. Rev. 1. August 1985.
- 11. D. P. Mackuwiak et 41.. "Developu nt of Transient initiating Event Frequencies For Use in probabilistic Risk Assetsment " NUREG/CR-38FP IDraf t), Jur e 1984.
CONDITIONS TO CLOSE OUT RELAY PLANTS TABLE 1 For plant-specific application of the TS changes proposed, and for plant-specific closecut of Item 4.5.3 of Generic Letter 83-28, an individual licensee for a plant usirg a relay RPS must:
(1) Confirm the applicability of the generic analyses to its plant.
(2) Demonstrate, by use of current drift information provided by the equipment vendor or plant-specific data, that the drift characteristics for instrumentation used in RPS channels in the plant are bounded by the assumption used in NEDC-30851P when the functional test interval is extended from monthly to quarterly.
(3) Confirm that the differences between the parts of the RPS that perform the trip functions in the plant and those of the Le:a case plant werc included in the analysis for its plant done using the procedures of
', Appendix K of NEDC-30851P (and the results presented in Enclosure 1 te i
letter OG5-491-12 from L. Rash (GE) to T. Collins (NRC) dated hovember 25,1985), or present plant-specific analyses to demonstrate no appreciable change in RPS availability or public risk.
l i
CHANGES TO R.E1M RPS TECHNICAL SPECIFICATION l 3/4.3 INSTRUMD;TATION i
3/4.3.1 REACTOR PROTECTION SYSTEM INSTRUMENTATION LInITINC CONDITION FOR OPEAATION 1
j 3.3.1 As a miniaua, the reactor protection sytes instrumentation channels 1 shown in Table 3.3.1-1 shall be OPERABLE with the REACTOR PROTECTION SYSTEM i
KESPONSE TIME as shown in Table 3.3.1-2.
j APPLICA.BILITY: As shown in Table 3.3.1-1.
l 4
ACTION:
i a
- a. With the number of OPERABLE channels less than required by the Miniana OPEAABLE Channels per Trip Systes requirement for one trip syates, place '
the inoperable channel (s) and/or that trip system in the tripped '
condition
- vithin twelva hours. The provisions of Specification 3.0.4
} are not spp11 cable. i i
l b. With the number of OPDABLE channals less than ' required by the Minimum i
OPRABLE Channels per Trip System requirement for both trip systems, i
place at least one trip systemes in the tripped condition within one hour j and take the ACTION required by Table 3.3.1-1.
i SURVEILLANCE RIQUIRLw.ENTS 4.3.1.1.Each reatter protection system instrumentation channel shall be demonstrated OPERABLE by the perforance of the CHANNEL CHECK, CHANNEL TUNCTIONAL TEST and CHANNEL CALIBRATION operations for the OPERATIONAL
] CONDITIONS and at the frequencies shown in Table 4.3.1.1-3.
5 4.3.1.2 LOGIC SYSTEM FUNCTIONAL TESTS and siriulated automatic operation of all channels shall be performed at least once per 18 months.
4 e
i san a noper abl e channel need not be placed in the tripped conditt when this would cause the Trip Function to occur.
In these cases, the inoperable channel shall be restored to OPERABLE status wathin 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> after the channel was first determined to be inoperable or the ACTION required by Table 3.3.1-1 for that Trip Function shall be taken.
self more chennels are inoperable in one trip system than in t
the other, place the trip system with more inoperable channel s in the tr a pped condt ti on, except when thi s woul d cause the T-1p Function to occur.
I DIANCES TO RELAY RPS TECHNICAL SPECIFICATION Channel Operptional Channel Functional Channel CoM itions in which Functional Unit Check Test Calibration
- Sune111ance Required
- a. Neutron Flux - High S M ,S(b) p g(c) W R 2 S V R 3, 4, 5
- b. Inoperative NA W NA 2,3,4,5
- 2. Average Power Range Monitor:III
- b. Flow Binned Steulated Thermal Power - High S.D(h) gg(c) , q g(d)(e), SA, Ril) 1 l
- c. Neutron Flux - High S Sg(c) , g g(d), SA 1 f
- 4. Inoperative NA Q NA 1, 2, 3, 5 l
- 3. Reactor Yessel Stese Dome Pressure - High S 0 RIS) 1, 2 4 Reactor Yessel Water Level -
Low, Level 3 S O RI RI 1, 2
- 5. Reactor Vesnel Water k vel -
ylgh, Level 8 S O R(R) 1 l
- 6. Main Steam Line InoIntion valve - Clonore HA 0 R 1 t
- 7. Main Steam Line Radletion -
High S Q R 1, 2(.1) f r
~
CHANCES TO RELAY RPS TEGINICAL SPECIFICATION
%ennel Operational Channel Funictional channel Conditione in Which Check' Test Calibrationg ,) Surveillence Required Functional Unit
- 8. Drywell Pressure - High ,
(S) Q (R)(8) 1, 2 l
- 9. Scree Discharge Volume Water level - High 1, 2, Sik)
Type 1 5 Q RI S)
Type 2 S Q R 1, 2, 5(k)
R 1 l
- 10. Turbine Stop Valve - Closure NA Q
- 11. Turblee Control Valve Fast ,
I Closure Valve Trip System 011 Fressure - Low MA Q R 1 l l
- 12. Reactor Node Switch N!a 1,2,3,4,5 Shutdown Position NA R i
NA W NA 1,2,3,4,5l
- 13. Manual Scree l
<.-_______..___=___=.__ _ . _ _ _ _ . _ _ _ . _
i CHANGES 70 gEIAY RPS TECHNICAL SPECIPICATION NOTES 70 TABLE (a) Neutron detectors snay be escluded from CHANNEL CALIBPATION.
(b) Ibe IM and SM channels shall be determined to overlap for at least (1/2) decades during each startup af ter entering OPDAT10NAL C0h71 TION 2 and the 1M and APM channels shall be determined to overlap for at least (1/2) decades during each controlled shutdown, if not performed within the previous 7 days.
(c) Within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> prior to startup, if not performed within the previouc 7 days.
(d) This calibration sha).1 consist of the adjustment of the APM channel to conform to the power values calculated by a beat balance during OPEAATIONAL C0hTITION 1 when THEMAL POWm > 25% of RATED THEAMAL POWIR. Adjust the APM chtsnel if the absoTute difference is greater j than 21 of RATED THD.HAL POWER. Any APM channel gain adjustment made in compliance with Specification 3.2.2 shall not be included in determining the absolute difference.
(e) This calibration aball consist of the adjustment of the APM flow biAsee channel to conform to a calibrated flow signal.
(f) The LPPy.s shall be calibrated at Itast once per 1000 effective full power hours (EFFH) using the TIP system.
(g) Calibrate trip unit at least once per 92 days. l (b) Verify sensured core flow to be greater than or equal to established core flow at the existing flow control valve position.
(1) This calibration shall consist of (verifying) (adjustaent, as required of) the 6+1 second simulated thermal power time constant.
(j) Tnis function is not required to be OPDABLE vben the reactor pressure vessel bend is removed per Specification 3.10.1.
(k) With any control rod withdrawn. Not applicable to control tods removed per Specification 3.9.10.1 or 3.9.10.2.
e
- 3/4.3 1K5TeLHth7 AT10N 855!$
3 / t .3.1 tra:TOs PROTECT 10W SYsitw IW5TRtMth7AT10N ,
The reacter protection systee avtcutically initiates a reactor scrar to:
- a. Preserve the integrity of the fuel cladding.
- 6. Preserve the integrity of the reactor coolatt systet.
- t. Ministre the energy which suit be absorbed following a less of. coolant accident, and i
- d. Prevent inadvertent criticeitty. ;
This specification provides the liriting tenditions for operation l mecessary to preserve the ability of the systre to perfort its intended I fcettien even during periods when instrveent thannels my be out of service j becewse of raintenance. When Neessary, eme channel my be ude inoperable ,
for brief intervals to e.onduct required survel11sace. ,
The reactor protection system is nde up of two independent trip syste.s. There are usually four channels to renitor each parameter with two channels in each trip system. The outputs of the channels in a trip system are cer-bined in a Icgic so that either thannel will trip that trip system.
The tripping of both trip systees will produce a reactor scram. The syster eseets the intent of IEEE-779 for nuclear power plant protection systers.
Specified survet11ance intervals and surveillance and r,tintenance cutage tires have been deterr.ined in accordance with NEDC.30251 P. " Technical Specification Irprovement Analyses for Bh'R Reacter Protection Syster.' as approved by the NRC and doccented in the SER Oetter to T. A. Pickens fro-A. Thadani dated July 15, 1987 . The bases for the trip setting of the RFS are discussed in the bases for Specification 2.2. .
The sensurennt of response time at the specified frequencies provides assurance that the protective functions associated with each thanhel are ccq1sted within the tin 1(Fit esssmed in the stfety Sn:lysts. No tredit was taken for these thannels with response tiets indicated as met applicable.
pesponse tiet t.sy be deecnstrated by any series af sequential overlapping er tetal channel test wasurement, provided such tests denenstrate the total thannel response tin as defined. Senser response tir.e vertittation ray be de.cnstrated by either O) inplace, entite er offsite test sensursrents, er
(?) uttitring replacement sensors with certified response times. ,
r d
y e- -, -, , , , - - , - _ , . .