ML20196D208
ML20196D208 | |
Person / Time | |
---|---|
Site: | Nine Mile Point ![]() |
Issue date: | 11/24/1998 |
From: | NIAGARA MOHAWK POWER CORP. |
To: | |
Shared Package | |
ML20196D198 | List: |
References | |
NUDOCS 9812020177 | |
Download: ML20196D208 (13) | |
Text
~ - ~.. - . - . . , - - -- . - - - - - - - - - - - . . . . . - - - . - - - - - - . - . ~.
t BASES FOR 2.1.1 FUEL CLADDING - SAFETY LIMIT l a
During periods when the reactor is shut down, consideration must also be given to water level requirements, due to the effect of decay
-hent, if reactor water level should drop below the top of the active fuel during this time, the ability to cool the core is reduced. This reduction in core cooling capability could lead to elevated cladding temperatures and clad perforation. The core will be cooled sufficiently to
. privent clad melting should the water level be reduced to two-thirds of the core height. t The Fuel Zone Water Level Monitoring System (FZWLMS) instrumentation has an indicated range which allows continuous indication of ;
r:: ctor water level from below the_ bottom of the active fuel to above the maximum normal water level. The reactor vessel tap for the low-t low-isw water level instrumentation is located approximately 7 feet 11 inches below the minimum normal water level or approximately 4 !'
feet 6 inches above the top of the active fuel. The low-low-low water level trip point is 6 feet 3 inches (-10 inches indicator scale) below the minimum normal water level (Elevation 302'-9"). The 20 inch difference between the reactor vessel tap and the trip point resulted from !
cn svaluation of the recommendations contained in Geners! Electric Service information Letter 299 "High Drywell Temperature Effect on ')
Rsrctor Vessel Water Level Instrumentation." The low-low-low water level trip point was raised 20 inches to conservatively account for
- possible differences in actual to indicated water level due to potentially high drywell temperatures. The safety limit has been established j hers to provide a point which can be monitored and also can provide adequate margin. However, for performing major maintenance as l specified in Specification 2.1.1.e, redundant instrumentation will be provided for monitoring reactor water level below the low-low-low !
wcter level set point. (For example, by installing temporary instrument lines and reference points to redundant level transmitters so that the rocctor water lavel may be monitored over the required range). In addition, written procedures, which identify all the valves which have the !
pot:ntial of lowering the water level inadvertently, are established to prevent their operation during the major maintenance which requires j the water level to be below the low-low level set point.
The thermal power transient resulting when a scram is accomplished other than by the expected scram signal (e.g., scram from neutron flux following closure of the main turbine stop valves) does not necessarily cause fuel damage. However, for this specification a safety limit ;
vio!stion will be assumed when a scram is only accomplished by means of a backup feature of the plant design. The concept of not ;
epproaching a safety limit provided scram signals are operable is supported by the extensive plant safety analysis. I r
6 I
L
! 9812O20177 981124 PDR ADOCK 05000220 '
7 '
P PDR ,_
AMENDMENT NO.142 16 I
( :-
BASES FOR 3.2.5 AND 4.2.5 REACTOR COOLANT SYSTEM LEAKAGE RATE Allowable leakage rates of coolant from the reactor coolant system have been based on the predicted and expenmentally observ~ed behavior of cracks in pipes and on the ability to makeup coolant system leakage in the event of loss of offsite a-c power. The normally expected background leakage due to equipment design and the detection capability for determining coolant system leakage were also considered in establishing the limits The behavior of cracks in piping systems has been experimentally and analytically investigated as part of the USAEC sponsored Reactor Primary Coolant System Rupture Study (the Pipe Rupture Study). Work utilizing the data obtained in this study indicates that leakage from a crack can be detected before the crack grows to a dangerous or critical size by mechanically or thermally induced cyclic loading, or stress corrosion cracking or some other mechanism characterized by gradual crack growth. This evidence suggests that for ler.kage somewhat greater than the limit specified for urudentified leakage, the probability is small that imperfections or cracks associated with such leakage would grow rapidly. However, the establishment of allowable unidentified leakage greater than that given in 3.2.5 on the basis of the data presently available would be premature because of uncertainties associated with the data. For leakage of the order of 5 gpm as specified in 3.2.5, the experimental and analytical data suggest a reasonable margin of safety that such leakage magnitude would not result from a crack approaching the critical size for rapid propagation. Leakage of the magnitude specified can be detected reasonably in a matter of a few hours utilizing the available leakage detection schemes, and if the origin cannot be determined in a reasonably short time, the plant should be shut down to allow further investigation and corrective action.
Inspection and corrective action is initiated when unidentified leakage increases at a rate in excess of 2 gpm, within a 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> period or less. This minimizes the possibility of excessive propagation of intergranular stress corrosion cracking.
A total leakage of 25 gpm is well within the capacity of the control rod drive system makeup capability (page ill-7 of the First Supplement)*.
As discussed in 3.1.6 above, for leakages within this makeup capability, the core will remain covered and automatic pressure blowdown will not be actuated.
Leakage is detected by having all unidentified leakage routed to the drywell floor drain tank and identified leakage routed directly to the drywell equipment drain tanks. Identified leakage includes such items as recirculation pump sealleakage.
The primary means of determining the reactor coolant leakage rate is by monitoring the rate of rise in the levels of the drywell floor and equipment drain tanks. Checks will be made every four hours to verify that no alarms have been actuated due to high leakage. For sump l inflows of one gpm, changes on the order of 0.2 gpm can be detected within 40 minutes. At inflows between one and five gpm, changes on the order of 0.5 gpm can be detected in eight minutas. .
'FSAR t
c AMENDMENT NO.142 103
BASES FOR 3.2.5 AND 4.2.5 REACTOR COOLANT SYSTEM LEAKAGE RATE Another method of determining reactor coolant leakage rate is by monitoring for excess leakage in the drywell floor and equipment drain tanka. This system monitors the change in tank volume over accurate time periods for the full range of tank instrumentation. If the leakage is high enough, an alarm is actuated indicating a leak rate above the predetermined limit (Section V.B)*. .
Additional information is available to the operator which can be used for the shift leakage check if the drywell sumps level alarms are out of service. The integrated flow pumped from the sumps to the waste disposal system can be checked.
Qualitative information is also available to the operator in the form of indication of drywell atmospheric conditions. Continuous leakage from the primary coolant system would cause an increase in drywell temperature. Any leakage in excess of 15 gpm of steam would cause a continuing increase in drywell pressure with resulting scram (First Supplement)*.
Either the rate of rise leak detection system, the excess leakage detection system or the integrated flow can be utilized to satisfy Specification 3.2.5.b.
i
- FSAR AMENDMENT NO.142 104
_______._._______.____m. _ _ _ . - - - -u - - - .-m__ _ - - - - - - - --- ___---t ,-i_ ----
BASES I"JR 3.3.1 AND 4.3.1 OXiGEN CONCENTRATION The four percent by volume oxygen concentration eliminates the possibility of hydrogen combustion following a loss-of-coolant accident 1Section Vil-G.2.0 and Appendix E-II.5.2)*. The only way that significant quantities of hydrogen could be generated by metal-water r action would be if the core spray system failed to sufficiently cool the core. As discussed in Section Vll-A.2.O', each core spray system will deliver, as a minimum, core spray spargst flow as shown on Figure Vll-2*. In addition to hydrogen generated by metal-water reaction.
_ significant quantities can be generated by radie!y;~;. - (Technical Supplement to Petition for Conversion from Provisional Operating License to Full Term Operating License).
At r actor pressures of 110 psig or less, the reactor will have been shutdown for more than an hour and tts decay heat will be at sufficiently low values so that fuel rods will be completely wetted by core spray. - The fuel clad temperatures would not exceed the core sprcy water saturation temperature of about.344*F.
The occurrence of primary system leakage followmg a major refueling outage or other scheduled shutdown is much more probable than the ,
occurrence of the loss-of-coolant accident upon which the specified oxygen concentration limit is based. Permitting access to the drywell i fct bak inspections during a startup is judged prudent in terrns of the added plant safety offered without significantly reducing the margin of safsty. Thus to preclude the possibility of starting the reactor and operating for extended periods of time with significant leaks in tint prim ry system, leak inspections are scheduled during startup periods when the primary system is at or near rated operating temperature -
cnd pressure. ' The 24-hour period to provide inerting is judged to be reasonable to perform the leak inspection and attabhsh the rsquired oxyg:n concentration.
The primary containment is normally'slightly pressurized during periods of reactor operation.' Nitrogen used for inerting could les out cf the containment but air could not leak in to increase the oxygen concentration. Once the containment is filled with nitrogen to the required concentration, no monitoring of oxygen concentration is necessary. However, at least once a week, the oxygen concentration will be det:rmined as added assurance that Specification 3.3.1 is being met.
k
'FSAri AMENDMENT NO.142 126 x e _
ATTACHMENT B NIAGARA MOHAWK POWER CORPORATION LICENSE NO. DPR-63 DOCKET NO. 50-220 Suocortina Information for the Technical Sagp?ication Bases Chanaes Suocortino information for the Chanaes to the " Bases for 2.1.1 Fuel Claddino - Safetv Limit" The change to the Nine Mile Point Unit 1 (NMP1) Technical Specification (TS) " Bases for 2.1.1 Fuel Cladding - Safety Limit" updates the Bases to reflect the installation of the Fuel 1 Zone Water Lavel Monitoring System (FZWLMS) and the change to reactor core fuel bundle types with an increased active fuel length of 145.24 inches. The Modification N1-80-38 installed the FZWLMS ia December 1981 and the initial core fuelload using the longer active fuel bundles occurred in 1979. FZWLMS instrumentstion upgrades were implemented in 1984 and 1995.
The FZWLMS was installed to provide the operator with instrumentation necessary to readily recognize and implement actions to correct or avoid conditions of inadequate core cooling. The system is divided into two redundant channels that provide digital readouts of reactor pressure vessel (RPV) water level. There is also a single channel chart recorder readout of RPV level. The system is designed to continue operation in the event of a loss of offsite power. The instrumentation has an indicated range of -240 inches to + 110 inches which allows continuous indication of RPV water level from below the bottom of the active fuel to above the maximum normal water level. This modification significantly increased the range of RPV levelindication and provided full core monitoring capability.
The FZWLMS is the primary means of raonitoring in-shroud collapsed water level following a Loss of Coolant Accident (LOCA). The system corrects for RPV water density changes resulting from vessel pressure changes and also corrects for the effects of containment drywell and reactor building temperature changes on the reference legs. The system does not provide accurate water levelindication when the reactor recirculation pumps are running because recirculation flow affects in-shroud differential pressure. Thus, the system is in standby during normal operation and initiates when all five recirculation pumps are either manually or automatically tripped.
On April 2,1979, the NRC Otaff issued NMP1 TS Amendment No. 31 incorporating changes 9 fated to the use of General Electric (GE) retrofit 8DNB277 fuel bundles for Cycle
- 6. Ts wre reload was the first use of the GE retrofit "8x8R" fuel bundles by NMP1 and also the first use of fuel bund ltu: with an active fuel length of 145.24 inches. The previous fuel bundle types had an active fuellength of 144 inches. Although the fuel bundle types have changed ouring subsequent fuelloads (the current core contains all GE11 fuel), the maximum !ength of the active fuel continues to be 145.24 inches.
On page 16. the second paragraph of the current TS " Bases for 2.1.1 Fuel Clauaing -
Safety Limit" reads, in part, as follows:
Page 1 of 9
The lowest' point at which the reactor water level can normally be monitored is approximately 7 feet 11 inches below minimum normal waterlevel or 4 feet 8 inches above the top of the active fuel. This is the location of the reactor vessel tap for the low-low-low waterlevelinstrumentation. The actuallow-low-low waterlevel trip point is 6 feet 3 inches (-10 inches indicator scale) below minimum normal water level (Elevation 302 '-9 "). The 20 inch difference resulted from an evaluation of the recommendations contained in General Electric Service Information Letter 299 "High Drywell Temperature Effect on Reactor Vessel Water LevelInstrumentation. "
Niagara Mohawk Power Corporation (NMPC) is revising the paragraph to read, in part, as follows:
The Fuel Zone Water Level Monitoring System (FZWLMS) instrumentation has an indicated range which allows continuous indication of reactor waterlevelfrom below the bottom of the active fuel to above the maximum normal waterlevel. The reactor vessel tap for the low-low-low waterlevelinstrumentation is located approximately 7 feet 11 inches below the minimum normal waterlevel or approximately 4 feet 6 inches above the top of the active fuel. The low-low-low water level trip point is 6 feet 3 inches (-10 inches indicator scale) below the minimum normal waterlevel(Elevation 302'-9"). The 20 inch difference between the reactor vessel tap and the trip point resulted from an evaluation of the recommendations contained in General Electric Service Information Letter 299 "High Drywell Temperature Effect on Reactor Vessel Water LevelInstrumentation. "
The Bases paragraph was revised to replace the statement describing the lowest RPV water level that could be monitored with a statement that describes the current capability of the FZWLMS to monitor water level throughout the entire active fuel region. The revised paragraph also corrects the height between the low-low-low level instrumentation tap and the top of the active fuel to account for the increased active fuel length. The word
" actual was deleted from the third sentence of the current Bases since the actuallow-low-low water level trip setpoint may be above the steted safety limit.
The FZWLMS modification provided NMP1 with unambiguous, easy-to-interpret RPV water level indication. This modification was implemented in response to NUREG-0578, "TMI 2 Lessons Learned Task Force, Status Report and Short-Term Recommendations,"
Recommendation 2.1.3.b, and NUREG-0737, " Clarification of TMl Action Plan Requirements," Item II.F.2, " Instrumentation for Detection of inadequate Core Cooling."
The FZWLMS does not perform any active safety function. Safety Evaluations were performed in accordance with 10 CFR 50.59 and concluded that the original FZWLMS modification and the 1984 and 1995 instrumentation upgrades did not involve a change to the TSs or an unreviewed safety question.
The current NMP1 (Cycle 13) reactor core contains all GE11 type fuel bundles, with active fuel lengths of 145.24 inches. The GE11 fuel design was licensed under NEDE-24011-P-A, " Generic Reload Fuel Application." A Safety Evaluation was performed in accordance with 10 CFR 50.59 and concluded that NMP1 can be operated safely with the Cycle 13 core and that operation with the Cycle 13 core did not involve a change to the TSs or an unreviewed safety question. The operating limits for the Cycle 13 core have been established and the Core Operating Limits Report (COLR) has been submitted to the NRC pursuant to TS 6.9.1.f.
Page 2 of 9
Sunoortina information for the Chanaes to the " Bases for 3.2.5 and 4.2.5 Reactor Coolant System Leakaae Rate" The changes to the NMP1 TS " Bases for 3.2.5 and 4.2.5 Reactor Coolant System Leakage Rate" update the Bases to incorporate NMP1 Configuration Change 1F00193 and Modifications SC1-0064-94 and N1-79-003, which modified the drywell leak detection system. The first two design changes replaced obsolete and maintenance intensive electromechanical equipment with more reliable and accurate solid state digital technology components. The third design change rerouted reactor recirculation pump suction and
]
discharge valve leakage to the drywell floor drain tank. This leakage had previously been ,
routed to the drywell equipment drain tanks. Although the system functions were not j altered, a Bases update is necessary to reflect the design changes. Editorial changes are also included in this Bases update.
Allidentified Reactor Coolant System (RCS) leakage is collected in the two drywell equipment drain tanks and all unidentified leakage is collected in the drywell floor drain tank. The drywell leak detection system provides the following functions:
- Drywell equipment and floor drain tank water level and " rate of rise" monitoring on control room chart recorders.
- Drywell equipment and floor drain tank sump pump control.
- Drywell equipment and floor drain tank " excess leakage" annunciator actuation based on the number of level sv!i;ch actuations for a fixed time duration. l
- Drywell floor drain tank " rate of rise" annunciator actuation based on the measured l time duration between level switch actuations.
These alarm and monitoring functions are the primary means utilized to meet TS 3.2.5.
The previous drywell leak detection system consisted of displacer tubes connected to Linear Variable Differential Transformers (LVDTs) which provided drywell equipment and floor drain tank level signals to the main control room chart recorders and Rochester Instruments " rate of rise" units. Conductivity level probes were used as inputs to General Electric (GE) FANUC Programmable Logic Controllers (PLCs) to control sump pump starting and stopping, and to provide " excess leakage" monitoring and control room annunciation.
The PLCs replaced an obsolete system of electromechanical relays and timers. This design change was implemented per NMP1 Modification SC1-0064-94 during refueling outage (RFO)13, which commenced on February 8,1995.
NMP1 Configuration Change 1F00193 was implemented during RFO14, which commenced on March 3,1997. The modification consisted of the retirement of the existing drywell equipment and floor drain tank level probes, the removal of the LVDTs and associated electronics and control relays, and the removal of the Rochester Instruments " rate of rise" units. The existing drywell equipment and floor drain tank level chart recorders were retained and the control room alarms remained intact. The existing PLCs were also retained, and additionci PLCs were installed. The tank level sensors were replaced with !
new sensors utilizing magnetic reed switch type level probes and signal conditioners to provide the required output level signals for the PLCs. The PLCs, in conjunction with the new level sensors, now provide all of the system monitoring, alarm, and control functions, including providing outputs to drive the control room chart recorders.
Page 3 of 9
1 f>1odification N1-79-003 removed the reactor recirculation pump suction and discharge valve packing gland leakoff isolation valves and capped the associated leakoff lines to the drywell equipment drain tanks. This resulted in the recirculation suction and discharge valve leakage being indirectly routed to the drywell floor drain tank (unidentified leakage) instead of the being directly routed to the drywell equipment drain tanks (identified leakage).
On page 103, the first sentence in the fourth paragraph of the current " Bases for 3.2.5 and 4.2.5 Reactor Coolant System Leakage Rate" reads es follows:
The primary means of determining the reactor coolant leakage rate is by monitoring the rate of rise in the levels of the drywell floor and equipment drain lines.
NMPC is revising the sentence to read as follows:
The primary means of determining the reactor coolant leakage rate is by monitoring the rate of rise in the levels of the dryweV floor and equipment drain tanks.
This revision replaces the word " lines" with the word " tanks" to correct an apparent editorial error. It is clear from context and the discussion on page 104 that the sentence is describing the means by which the " rate of rise" in the levels of the drywell floor and equipment drain tanks are monitored. The level probes which provide the " rate of rise" levelinformation are installed in the drain tanks, not the drain lines.
On page 104, the first paragraph of the current " Bases for 3.2.5 and 4.2,5 Reactor Coolant System Leakage Rate" reads as follows:
Leakage is detected by having all unidentified leakage routed to the drywell floor drain tank and identified leakage routed directly to the drywell equipment drain tanks. Identified leakage includes such items as recirculation pump sealleakage and recirulation pump suction and discharge valve packing leakoff.
NMPC is moving the paragraph to page 103 between the current third and fourth paragraphs and revising it to read as follows:
Leakage is detected by having all unidentified leakage routed to the drywell floor drain tank and identified leakage routed directly to the drywell equipment drain tanks. Identified leakage includes such items as recirculation pump sealleakage.
This paragraph was revised to eliminate the reference to " recirculation pump suction and discharge valve packing leakoff." In addition, the paragraph is being moved to page 103 to enhance the presentation of the Bases information. As previously described, Modification N1-79-003 capped the recirculation pump suction and discharge valve packing gland leakoff lines such that the leakage is now indirectly routed (via the drywell floor) to the drywell floor drain tank instead of being directly routed (via piping) to the drywell equipment drain tanks. As a result, recirculation pump suction and discharge valve leakage is now considered unidentified leakage insteart of identified leakage. Accordingly, the change eliminates the reference to the " recirculation pump suction and discharge valve packing leakoff" from the Bases discussion of identified leakage sources. The design change is acceptable since unidentified leakage to the drywell floor drain tank is monitored by the drywell leak detection system in a manner that is equivalent to that for the drywell '
equipment drain tanks. Moreover, the TS 3.2.5.a leakage limit specified for unidentified Page 4 of 9
- p. _ _ _ _ _ .
f
. , ~.
1 i leakage is more restrictive than the leakage limit for identified leakage (i.e., 5 gpm for i unidentified leakage vs. 20 gpm for identified leakage). A Safety Evaluation was j- perfctmed in accordance with 10 CFR 50.59, which determined that the modification was acceptable for implementation. The modification does not involve a change to the TSs or i an unreviewed safety question. Moving the Bases paragraph from page 104 to page 103 j is for presentational preference reasons only, and is considered editorial.
t l On page 104, the second paragraph of the current TS " Bases for 3.2.5 and 4.2.5 Reactor ,
i Coolant System Leakage Rate" reads as follows:
[
Another method wiH monitor the time required to fiH the tanks between two accurately l determined levels. When the levelin the tank reaches the low-level switch setting, a timer i wiH start and operate for a preset time interval. If the timer resets before the high-level i switch setting is reached indicating a leakage rate within aHowable limits, no action wiH \
result, and the system resets for the next fiHing and timing cycle. If the leakage is high i enough to cause the level to reach the high levelswitch setting before the timer resets i automaticaHy, an alarm is actuated indicating leak rate above the predetermined limit (First [
and Fifth Supplements)*. >
NMPC is revising the paragraph to read as follows:
[
Another method of determining reactor coolant leakage rate is by monitoring for excess l leakage in the dryweH floor and equipment drain tanks. This system monitors the change ;
in tank volume over accurate time periods for the fuHrange of tank instrumentation. If the :
leakage is high enough, an alarm is actuated indicating a leak rate above the predetermined i limit (Section V.8)*. ;
This paragraph was revised to eliminate the description of the timer function as the timers i have been replaced with PLCs. The PLCs now perform the " excess leakage" alarm !
function for the drywell equipment and floor drain tanks by measuring the volume change ;
in gallons (derived from the level osors) that occurs over'a short time period and ;
calculating the rate of change. Volume change is used to determine rate of change !
because of the irregular shapes of the tanks and provides " excess leakage" alarm capability across the entire range of tank instrumentation. Also, the previous reference to the "First l and Fifth Supplements" of the Final Safety Analysis Report (FSAR) has been replaced with ;
the current section of the NMP1 Updated Final Safety Analysis Report (UFSAR), which was
{
revised to describe the upgraded drywell leak detection system. Since the referenced i UFSAR section adequately describes the leak detection methodology for the drywell floor i and equipment drain tanks, a detailed description in the Bases is not necessary. i
\
l On page 104, the fifth paragraph of the current " Bases for 3.2.5 and 4.2.5 Reactor l Coolant System Leakage Rate" reads as follows:
Either the rate of rise leak detection system, the timer leak detection system or the integrated flow can be utilized to satisfy Specification 3.2.5.b.
l l
I Page 5 of 9
~
l l
1 NMbC is revising the paragraph to read as follows:
)
Either the rate of rise leak detection system, the excess leakage detection system or the integrated flow can be utilized to satisfy Specification 3.2.5.b.
This revision replaces the words " timer leak" with the words " excess leakage." A now obsolete system of electromechanical relays and timers had been previously used to provide drywell equipment and floor drain tank sump pump control and " excess leakage" monitoring and annunciation. PLCs replaced this system and now provide the sump pump control and " excess leakage" monitoring and alarm functions. This revision updates the paragraph to reflect this design change and is consistent with the revision to the second paragraph.
Configuration Change 1F00193 and Modification SC1-0064-94 retain the existing two-channel configuration of the drywell equipment and floor drain tank " excess leakage" j alarm, floor drain tank " rate of rise" alarm, and sump pump control functions. The number of level sensors installed provide for full level signal redundancy within each instrument channel. Due to solid state digital technology, the PLCs off.er much greater reliability than l the electromechanical equipment that was removed. In addition, the PLCs are equipped I with self-diagnostics which determine if any problems exist in the software or hardware 1 during all periods of operation. If the self-diagnostics indicate that the drywell leak detection system is experiencing a failure or is in a degraded condition, an alarm will be '
annunciated in the control room. The PLC software program is designed to monitor the deviation between the redundant level sensors and to actuate a control room annunciator if the programmed tolerance is exceeded. The use of redundant sensors within the two instrument channels and automatic fault reporting by the PLCs provides an increased level of reliability and fault tolerance. The use of moisture proof wire connection techniques also adds to the reliability by removing a known failure mechanism.
If a complete failure of the drywell leak detection system occurs (both channels inoperable), the condition remains bounded by TS 3.2.5 which requires that Unit 1 be placed in cold shutdown within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />.
l The drywell leak detection system instrumentation is classified as Category 3 (Regulatory Guide 1.97) and nonsafety-related. The purpose of the drywell leak detection system is to alert the control room of an incipient failure of the RCS pressure boundary in order to allow time for cause determination and plant shutdown (if necessary) before large scale leakage i occurs. Other indications are available (e.g., increasing drywell temperature and pressure, l' decreasing RPV level) to alert the control room of larger scale failures of the RCS pressure boundary. A Safety Evaluation was performed in accordance with 10 CFR 50.59 and j concluded that the modifications to the drywell leak detection system did not involve a change to the TSs or an unreviewed safety question.
Sucoortino information for the Chances to the " Bases for 3.3.1 and 4.3.1 Oxvoen Concentration" The changes to the NMP1 TS " Bases for 3.3.1 and 4.3.1 Oxygen Concentration" update the core spray system flow parameters included in the first paragraph, third sentence, on Page 126 to reflect the results of NMPC Calculation No. S14-40-F005. These results are presented in Figure Vll-2 of the NMP1 UFSAR. The changes to these Bases also include Page 6 of 9
0 minor editorial changes to the first paragraph and the correction of a typographical error in the parenthetical reference at the end of the first paragraph.
TS 3.3.1 limits the oxygen concentration in the primary centainment atmosphere to less than 4% by volume whenever the reactor coolant pressure is greater than 110 psig and the reactor is in the power operating condition. Core spray system flow precludes any significant metal-water reaction for the production of hydrogen within the primary containment following a LOCA. Thus, by limiting the oxygen concentration and assuring core spray flow, the possibility of hydrogen ccmbustion within the primary containment following a LOCA is eliminated. The " Bases for 3.3.1 and 4.3.1 Oxygen Concentration" include the core spray system flow parameters assumed in the LOCA analyses. However, the current values are not consistent with the results of the most recent LOCA analyses, which are presented in Figure Vil-2 of the NMP1 UFSAR.
There are two redundant core spray systems at NMP1. Each system is a separate and independent core spray loop consisting of two independent suction headers, two core spray pumps, two core spray topping pumps, two relief actuated recirculation lines, a test line, a spray sparger, and associated piping, instrumentation, and control valves. The redundant active components in each system are designed and configured to operate as two mdundant subsystems. Thus, each of the two redundant core spray systems contains two redundant subsystems. The core spray system is described in Section Vll-A of the NMP1 UFSAR.
When low-low RPV water level or high drywell pressure is sensed following initiation of a design basis LOCA, all four core spray pumps and all four core spray topping pumps start sequentially. The inside (inside the drywell) isolation valves open (the outside valves are normally open) to allow core spray flow into the RPV when reactor pressure falls below 365 psig. Once core spray has been initiated, it will remain in operation until the core spray pumps and topping pumps are manually shutdown. In the event of a small break LOCA which may not depressurize the RPV, the Automatic Depressurization System (ADS) is available to reduce reactor pressure to below 365 psig and allow the core spray system to function in the desired manner.
Following blowdown from a design basis LOCA. the core spray system removes decay heat and any chemical energy from the reactor core. The core spray system draws water from the suppression chamber and sprays water over the top of the core. With the core .
spray system in operation, the metal-water reaction is negligible and a peak suppression l chamber pressure of 22 psig occurs immediately after blowdown and before the containment spray system (described in UFSAR, Section Vll-B) is initiated. The only way
- that significant quantities of hydrogen could be generated by metal-water reaction would be if the core spray system failed to sufficiently cool the core. This is highly unlikely considering the system and subsystem redundancy. However, in the event that the core spray system did not operate, a peak suppression chamber pressure of 25 psig would occur following a substantial metal-water reaction. The design pressure for the suppression chamber is 35 psig which results in a rnarnin of 10 psig. The occurrence of a i metal-water reaction does not affect the peak drywell pressure, and based on analysis and testing, the maximum drywell pressure would be 50 psig which results in a margin of 12 ,
psig to the 62 psig design pressure for the drywell. '
Page 7 of 9
I Thus, substantial safety margins exist for both the drywell and suppression chamber even in the unlikely event that the core spray system failed to operate. The primary containment design bases are described in Section VI-B of the NMP1 UFSAR.
On Page 126, the first paragraph of the current TS " Bases for 3.3.1 and 4.3.1 Oxygen ;
Concentration" reads as follows:
The four percent by volume oxygen concentration eliminates the possibility of hydrogen combustion foHowing a loss-of-coolant accident (Section Vll-G.2.0 and Appendix E-1 1. 5. 2) *. The only way that significant quantities of hydrogen could be generated would be if all core spray systems failed to sufficiently cool the core. As discussed in Section Vll-A.2.0 and iHustrated in Figure Vll-2 *, the core spray system is capable of design flow of 3400 gpm at a reactor pressure of 113 psig. In addition to hydrogen generated by metal-water reaction, significant quantities can be generated by radiolysis. (Technical Specification to Petition for Conversion from Provisional Operating License to FuH Term l Operating License).
NMPC is revising the paragraph to read as follows: I The four percent by volume oxygen concentration eliminates the possibility of hydrogen combustion following a loss-of-coolant accident (Soction Vll-G.2.0 and Appendix E-l1. 5. 2) *. The only way that significant quantities of hydrogen could be generatsd by metal-water reaction would be if the core spray system failed to sufficiently cool the core. As i discussed in Section Vll-A.2.O*, each core spray system wiH deliver, as a minimum, core '
spray sparger flow as shown on Figure Vll-2 *. In addition to hydrogen generated by metal-water reaction, significant quantities can be generated by radiolysis. (Technical Supplement to Petition for Conversion from Provisional Operating License to FuH Term .
Operating License). !
I The third sentence in the paragraph is revised to reference the current licensing basis flow )
parameters as presented in Figure Vil-2 of the NMP1 UFSAR. The word "the" in the sentence is replaced with "each" to be consistent with the flow data presented in the Figure. The core spray system flow performance pressure versus flow data assumed in the LOCA analyses are presented in Table XV-9A of the UFSAR and the values are reflected in Figure Vll-2.
A typographical error at the end of the paragraph is corrected in the revised paragraph.
The current Bases incorrectly use the word " Specification" in lieu of " Supplement" in the parenthetical reference to the " Technical Supplement to Petition for Conversion from Provisional Operating License to Full Term Operating License." This error was inadvertently introduced when NMPC staff retyped the page and included it in the Application for Amendment to the NMP1 Operating License, dated December 8,1989.
The application was subsequently approved and issued on August 3,1990 as TS Amendment No.115. The typographical error apparently went unnoticed during processing of the Amendment because the error remained in the retyped Bases page issued with the Amendment.
In addition to the changes described above, two minor editorial changes have been incorporated into the first paragraph. The first editorial change corrects the reference
" Appendix E-11.5.2." The correct reference is " Appendix E-II.5.2" (i.e., Section E-Roman Page 8 of 9
i l 1
Numeral l1.5.2) of the original (June 1967) FSAR. This minor error has existed since l
Decbmber 26,1974 when the Bases were originally included as part of the Full-Term i Operating License Technical Specifications. The second editorial change clarifies the l second sentence of the paragraph. The second sentence now clearly states that a failure of the core spray system could significantly increase the quantity of hydrogen generated by I metal-water reaction. This clarification was necessary because a failure of the core spray system will have no impact on hydrogen generated by radiolysis. The control of hydrogen generated by radiolysis is a function of the Containment Atmospheric Dilution (CAD) system which is described in Section Vil-G.2.0 of the NMP1 UFSAR.
A Safety Evaluation was completed in 1995 to address the installation of individual core spray system minimum flow recirculation lines and remote throttling capability. The Safety Evaluation also addressed the incorporation of the results of Calculation S14-40-F005 into Figure Vil 2 of the UFSAR. The Safety Evaluation was performed in accordance with ,
10 CFR 50.59 and concluded that the changes did not involve a change to the TSs or an 1 unreviewed safety question. The core spray system flow parameters and descriptions which currently appear in the UFSAR are bounded by the Safety Evaluation. The changes being incorporated into the " Bases for 3.3.1 and 4.3.1 Oxygen Concentration" are consistent with the UFSAR. Thus, it follows that the changes to the Bases are also bounded by the Safety Evaluation.
The editorial changes and the typographical corrections do not alter the technical content or intent of the Bases and are, therefore, considered administrative in nature. Accordingly, these changes have no safety significance.
i l
l 1
l 1
Page 9 of 9