B14330, Forwards Revised Response to Ieb 80-04 Re Main Steam Line Break W/Continued Feedwater Addition,Covering Containment Pressure Response,Core Reactivity Response,Including Corrective Actions & MSLB Containment Analysis
| ML20127H206 | |
| Person / Time | |
|---|---|
| Site: | Millstone  |
| Issue date: | 01/13/1993 |
| From: | Opeka J NORTHEAST NUCLEAR ENERGY CO., NORTHEAST UTILITIES |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| B14330, IEB-80-04, IEB-80-4, NUDOCS 9301220253 | |
| Download: ML20127H206 (28) | |
Text
r- ,
4 1 y4 NORTHEAST UTILITIES o n.,.i on c.. . s.een su..i. B.,nn. Conn.ct cut 1 .n w un.n.
-, e .c. . m.
"ES'I P.O DOx 270 HARTFORD. CONNECTICUT 06141-0270
_k ' J ; ::C"," *%' "O~. I2m> ses-s000 January 13, 1993 Qgike.t No. 50-336 l
B14330 l Re: NRC IEB 80 04 U.S. Nuclear Regulatory Comission l Attention: Document Control Desk Washington, DC 20555 l
l Gentlemen:
l l
I Hillstone Nuclear Power Station, Unit No. 2 Inspection and Enforcement Bulletin 80-04 Revised Response l
l l In letters dated January 25, 1980"' and July 6, 1982,a2:- Northeast Nuclear l Energy Company (NNECO) responded to NRC Inspection and Enforcement Bulletin (IEB) 80 04. Subsequeht to these letters, analysis associated with the M111 stone Unit No. 2 steam generator replacement project ($GRP) determined that portions of these previous IEB 80-04 responses were in error. The purpose of this letter is to submit the revised response to IEB 80 04, as originally comitted in Licensee Event Report .(LER) ~ 91-10-00, dated November- 15, 1991, and subsequently revised based on telephone- conversation with NRC Region I Staff.
(1) W. G. Counsil letter _ to U.S. Nuclear Regulatory Comission, " Millstone Nuclear Power Station, Unit No. 2 Automatic -Initiation of Auxiliary l- Feedwater," dated January 25, 1980.
(2) W. G. Counsil letter _ to U.S. Nuclear -Regulatory Comission, _" Millstone Nuclear Power Station, Unit No. 2 I&E Bulletin 80-04 on Main Steam-Line Break With Continued Feedwater Addition-Response to Request. for Addi-
- tional Information," dated July 6,1982.
(3) S. E. Scace letter to U.S. Nuclear Regulatory Comission, " Facility Operating Licensing No. DPR-65, Docket No. 50-336, Licensee Event Report 91 010-00," dated November 15, 1991.
L L
2 7 01 1.~ ,
=
9301220253 930113 '
osuu arv.448 PDR ADOCK 05000336 l l_
G- PDR -
e 1
i .
U.S. Nuclear Regulatory Commission B14330/Page 2 January 13, 1993 Backaround in letters dated September 13, 1979,* October 22, 1979,* and October 30, 1979,* the NRC Staf f informed NNECO of the original Staff requirements for automatic initiation of auxiliary feedwater. NNECO's then ongoing evt.luations of the requirement were discussed in detail in a letter dated November 30, 1979.'" In a letter dated December 21, 1979,* the NRC Staff acknowledged correspondence transmitted by NNEC0 regarding automatic initiation of auxiliary feedwater systems upon the loss of main feedwater flow. The Staff-acknowledged that NNECO had submitted the correspondence in response to Short-Term Recommendation 2.1.7.a. " Auto Initiation of Li, " Auxiliary Feedwater System." During preparation of that submittal, NNECO raised the issue of the applicability of the then current main steam line break (HSLB) or main feedwater line break analysis, assuming ea-ly initiation of auxiliary feedwater flow with a failure to limit flow to the affected steam generator.
The basic question concerned whether the changes in assumptions would increase the calculated containment pressure or the likelihood of return to power. In the December 21, 1979,* letter, the Staff requested, among other things, that NNECO resolve the concern by submitting analysis for Staff review.
In response to the NRC Staff request for information on automatic initiation of the auxiliary feedwater system, the design basis steam line break analysis was reevaluated. In the analysis, the additional mass releases to the containment due to auxiliary feedwater addition were added to the Final Safety Analysis Report (FSAR) case and shown to have no impact on the peak (4) D. G. Eisenhut letter to All Operating Nuclear Power Plants, " Followup Actions Resulting from the NRC Staff Revisions Regarding the Three Mile Island Unit 2 Accident," dated September 13, 1979.
(S) D. G. Eisenhut letter to W. G. Council, "NRC Requirements for Auxiliary Feedwater Systems at Millstone Nuclear Power Station Unit 2 " dated October 22, 1979.
(6) H. R. Denton letter to All Operating Nuclear Power Plants, " Discussion of Lessons Learned Short-Term Requirements," dated October 30, 1979.
(7) W. G. Counsil letter U.S. Nuclear Regulatory Commission (J. H. Hendrie),
"Haddam Neck Plant, hillstone Nuclear Power Station, Unit No. 2," dated November 30, 1979.
(8) R. W. Reid letter to W. G. Counsil, " Automatic Initiation of Auxiliary feedwater Systems at Haddam Neck and Hillstone Unit No. 2," dated December 21, 1979.
(9) Ibid.
k U.S. Nuclear Regulatory Comission B14330/Page 3 January 13, 1993 containment pressure and temperature. Since this study was aimed at only assessing the impact of the new automatic feedwater initiation system, the ,
original FSAR assumptions we t not reevaluated. This was supported by evaluations performed by the aclear steam supply system vendor, Combustion Engineering. This comprehensive analysis was submitted to the NRC Staff in a letter dated January 25, 1980."" Since the information requested in IEB 80 04 issued in February 1980 was very similar to the request made in December 1979, NNEC0 presumed that this analysis was also sufficient to respond to the Bulletin. Therefore, no new analysis was performed for the Bulletin._ A Safetf" Evaluation Report -was received from the NRC Staff on October 7, 1982.
On October 18,1991, at 1305 hours0.0151 days <br />0.363 hours <br />0.00216 weeks <br />4.965525e-4 months <br />, with the plant in Mode 1 at 100 percent power, a reportability determination was made concerning a reanalysis of the MSLB event inside the containment. The reanalysis indicated that the assump-tions made for the existing (1979) MSLB analysis were nonconservative with respect to power level, break size, and single active failure. Using more restrictive assumptions, design limits for containment pressure and tempera-ture could have been exceeded. NNEC0 determined that this condition was reportable as a condition outside the design basis of the plant. An Immediate Report was made to the NRC, and the unit inrediately commenced an orderly downpower to apiroximately 3 percent power (Mode 2). The existing MSLB analysis remained valid for Mode 2 operation.
A Justification for Continued Operation (JCO) was developed to allow the unit to return to power operation by stationing a dedicated reactor operator to close the main feedwater block valves following any reactor trip. That JC0-documented the basis for reasonable assurance that, with .the actions of a dedicated operator, containment pressure would remain below the design basis value for all postulated MSLB events. The unit was returned to power opera-tion on October 22, 1991. The details LER 9101040, dated November 15,1991.""of this discovery were discussed in (10) W. G. Counsil letter to U.S. Nuclear Regulatory- Commission (R. Reid),
" Millstone Nuclear Power Station, Unit No. 2 Automatic Initiation of Auxiliary feedwater," dated January 25, 1980.
l (11) R. A. Clark letter to W. G. Counsil, " Resolution of Main Steam Line l Break With Continued Feedwater Addition Even for Millstone Nuclear Power Station, Unit No. 2," dated October 7, 1982.
(12) S. E.. Gcace letter to U.S. Nuclear Regulatory Commission, " Facility l Operating License No. DPR 65 Docket No. 50-336, Licensee Event Report 91-010-00," dated November 15, 1991.
l i-
-r - ~~- ,, , - - - , , , , . , , , - . m . . . - , . - . . - - . - - . , . - - - , _ - , , ,,-1. -, =.--.m--
U.S. Nuclear Regulatory Comission B14330/Page 4 January 13, 1993 Later, in supplemental- LER 91-010-01, dated January 17, 1992,"8' NNECO informed the NRC Staff that the short-term corrective actions, to close the main feedwater block valves given a containment isolation actuation signal (CIAS), had been installed and tested in December 1991. These changes eliminated the need for the dedicated operator. That LER also informed the Staff that changes for which hardware was available would be installed during the 1992 refueling outage. The remaining long-term hardware changes (i.e.,
qualified replacement components) would be completed during the 1994 refueling outage.
During the refueling outage, on August 4,1992, at 1600 hours0.0185 days <br />0.444 hours <br />0.00265 weeks <br />6.088e-4 months <br />, with the plant in Mode 6 at 0 percent power and all fuel stored in the spent fuel pool, a new reportability determination was made which identified two new postulated single failures for the MSLB event inside the containment which resulted in i the calculated containment pressure exceeding the design pressure limit.
The first was a failure of the feedwater regulating bypass valve to terminate flow to the affected steam generator. As ? art of the October 1991 MSLB evaluations, a failure of feedwater regulat'ng bypass valve to close was considered, but was assumed to only provide 10 percent of full power feedwater flow to the affected steam generator. This assumption was consistent with thr.
original analysis and resulted in acceptable containment pressures. There-fore, no provision was made in the JCO, or plant modifications proposed, to isolate the by feedwater flowpath. In response to the October 1991 determination, passa reanalysis of containment pressure response to a MSLB had been initiated, in June 1992, the reanalysis of the failure of the bypass valve to close was performed using actual condensate pump curves and feedwater regulating bypass valve flow characteristics. This reanalysis resulted in the calculated containment pressure exceeding the 54 psig design limit.
The second was a failure of the vital buses to fast transfer to the reserve.
station services transformer (RSST). In that case, power to the condensate pumps would remain available, while power to close the feedwater regulating valves and start the containment pressure control systems would be delayed, due - to diesel start and . sequencing times. These delay times were not previously considered in the MSLB analysis. NNECO committed, in LER 91 010-02, dated September 3, 1992,"* to perform plant modifications, during the outage then in progress, to ensure an acceptable containment (13) S. E. Scace letter to U.S. Nuclear Regulatory Commission, " Facility Operating License No. DPR-65, Docket No. 50-336, Licensee Event Report 91-010-01," dated January 17, 1992.
(14) S. E. Scace letter to U.S. Nuclear Regulatory Commission, " Facility -4 Operating License No. UPR-65, Docket No. 50 336, Licensee Event Report 91-010 02 " dated September 3, 1992. :
l
U.S. Nuclear Regulatory Commission B14330/Page 5 January 13, 1993 pressure response for a MSLB inside containment given these newly identified failures.
The October 1991 MSLB evaluations also considered the aossibility of various loss of power cases. Proposed plant modifications would have eliminated all of the cases identified at that time. However, the newly identified single failure of the vital buses to fast transfer to the RSST introduced delay times that were not previously considered in the HSLB analyses.
In response to the August 4, 1992, reportability determination, a multi-disciplinary task force was established to investigate the issue, ensure that '
all required single failures were considered, and to propose _ modifications which would ensure that the containment response is acceptable. Various design modifications were proposed and evaluated. Based u)on the evaluations concerning all identified single failures, plant modif' cations have been completed which included adding redundant main steam isolation (MSI) signals ,
to MSI actuated components; adding MSI signals to components which did not receive an MSI signal; modifying the HSI logic to actuate on high containment pressure, as well as low steam generator pressure; upgrading power supplies to vital power for selected valves; lowering the containment spray actuation setpoint; and reinstalling the emergency diesel generator start on a safety injection actuation signal. NNECO believes that implementation of these modifications eliminated the need to upgrade 2-FW-42A and 2 FW-42B (feedwater block valves) to full safety grade status as stated in LER 92 010-01.
Following the modifications made during the 1992 refueling outage, the l predicted MSL1 peak containment pressure and temperature will be equal to or less than 54 psig and 426'F. In a letter dated December 4,1992,"" NNEC0 provided additional information, as requested by NRC Staff L. ring their review of the licen:.e amendment associated with the MSLB modifications. Attached to that letter was NNECO's evaluation of equipment qualification (EQ) for postulated MSLB, Based on this evaluation, the required safe shutdown electrical equipment will remain qualified.
Attachnnt 1 provides the revised response to IEB 80-04, based on a comprehen-sive containment reanalysis which was preliminarily reviewed by the NRC Staff on December 17, 1992. Attachment 2 provides an augmented legend to the EQ figure, at requested by NRC Staff, for containment response which was previously submitted in our December 4,1992, letter."*
(15) J. F. Opeka letter to U.S. Nuclear Regulatory Commission, " Millstone Nuclear Power Station, Unit No. 2 Proposed Revision to Technical Specifications Main Steam Line Break Design Limits Response to Request for Additional Information," dated December 4,1992.
(16) Ibid.
e U.S. Nuclear Regulatory Commission I B14330/Page 6
- January 13, 1993
[qnclusion The results in Attachment 1, Table 1, show that for all cases, the peak ,
containment pressure is less than the containment design pressure of 54 psig. -
i figure 6 gives the bounding EQ containment temperature profile. The peak temperature for this profile is 426'F. NNECO notes that Technical Specification 5.2.2 specifies the containment building temperature limit as l 289'F. Because the containment atmosphere exceeds 289'F for only a short ,
period of time, the containment building remains well below 289'F. Further, l NNECO has identified the need to revise the bases of various technical specifications to reflect the results of recent analyses and to improve consistency among the bases. The previously docketed EQ evaluation indicates that the safety-related equipment is qualified for this temperature peak due to the very brief duration during which the temperature spike exists, i If you have any questions on this issue, please contact my staff.
Very truly yours, NORTHEAST NUCLEAR ENERGY COMPANY FOR: J. F. Opeka - .
Executive Vice President BY:
pp ,1, W. D. Romberg?
Vice President cc: T. T. Martin, Region I Administrator G. S. Vissing, NRC Project Manager, Hillstone Unit No. 2 P. D. Swetland, Senior Resident Inspector, Millstone Unit.Nos.1, 2, and 3 Subscribed and sworn to before me this /A'^ day of ./ ,
- v. /, 1993
~ &af hii.o4k Notary Public
(
(
i Date Commission Expires: 3 /3 ! / 0 c; a , ,, -. -..._ _ - - - , , - .
Docket No J 0-336 B14330 Attachment 1 Revised Response to IEB 80 04 January 1993
U.S. Nuclear Regulatory Commission B14330/ Attachment 1/Page 1 January 13, 1993 Revised _ Response to Bulletin 80 04 1.0 Gl{IAINMENT PRESSURE RESPONS1 item #1 of IE Bulletin No. 80 04 pertains to containment response for a main steam line break (HSLB) event. Specifically, Item #1 states:
Review the containment pressure response analysis to deter-mine if the potential for containment overpressure for a main steam line break inside containment included the impact-of runout flow from the auxiliary feedwater system and the impact of other energy sources, such as continuation of feedwater or condensate flow. In your review, consider yo &
ability to detect and isolate the damaged steam generator from these sources and the ability of the pumps to remain operable after extended operation at runout flow.
The analysis of record was reviewed during tt.e reanalysis performed in support of the planned steam generator replacement. It was determined that nonconservative assumptions had been made with respect to power level, break size, and single active failure. Interim analyses, using-more a)propriate assumptions with respect to these parameters, predicted that cesign limits for containment pressure and temperature would be exceeded before the damaged steam generator could be isolated and containment spray would be effective.
A review of_our previous response to IE Bulletin 80 04 determined that, due to these nonconservative assumptions, continuation of feedwater could occur, which had not been analyzed. As described in detail in Section 3.0, this_ finding required corrective actions to ensure isola-tion of feed sources under any single failure scenario. The corrective actions were put in place in three phases. The first phase involved stationing a dedicated operator to close the main feedwater block valves on any reactor trip. The second phase involved replacing the dedicated operator by wiring Containment Isolation Actuation Signals to the main feedwater block va1ves. The third phase involved completing an in-depth review to determine if other single failures existed, . analyses to evaluate the consequences of those failures, and implementing hardware-modifications as necessary.
2.0 CORE REACTIVITY RESPONSE Item #2 of IE Bulletin No. 80-04 pertains to the core reactivity analysis for a MSLB event. Specifically, Item #2 states:
Review your analysis of the reactivity increase which results from a main steam line break inside or outside _ containment. _This
U.S. Nuclear Regulatory Commission B14330/ Attachment 1/Page 2 January 13, 1993 review should consider the reactor cooldown rate and the potential for the reactor to return to power with the most reactive control rod in a fully withdrawn position. If your previous analysis did-not consider all potential water sources (i.e.r runout from the auxiliary feedwater system, continuation of feedwater or conden-sate flow) and if the reactivity increase is greater than previous analysis indicated, the report of this review should include:
- a. The Boundary conditions for the analysis, e.g., the end of life shutdown margin, the moderator temperature coefficient, power level and the net effect of the associated generator water inventory on the reactor system cooling, etc.
- b. The most restrictive single active failure in the safety injection system and the effect of that failure on delaying the delivery of high concentration boric acid solution to the reactor coolant system,
- c. The effect of extended water supply to the affected steam generator on the core criticality and return to power,
- d. The hot channel factors corresponding to the most reactive rod in the fully withdrawn position at the end of life, and Minimum Departure from Nucleate Boiling Ratio (MDNBR) values for the analyzed events.
The analysis of record for the core was reviewed relative to the above items and the following responses are provided:
- a. The end-of-cycle shutdown margin supported in the analysis or record is the Technical Specification _value of 3.6 percentap. The end-of-cycle moderator temperature coefficient is conservatively modeled as -28 pcm/*F. The analysis of record considers four cases, i.e., HFP and HZP both with and without offsite power.
Analysis of this event for these four cases is judged to bound the consequences at intermediate initial power levels. The net effect of the steam generator water inventory determines the potential overcooling that the primary coolant system will sustain which adversely impacts the reactivity transient.
- b. The single failure assumed in the analysis o_f record is the loss of one of two HPSI pumps. The time delay assumed in the analysis of record is 45 seconds for the HPSI pumps, for loss of offsite power cases. The time delay assumed for cases with offsite power is 30 seconds for the HPSI pumps. The analysis accounts for the delay that occurs ' from flushing the nonborated water from the safety injection lines prior to boron injection. These
- - __ . . - --- - _ - . - - . _ - - - - - _ _ ~ -. _ -
I U.S. Nuclear Regulatory Commission B14330/ Attachment 1/Page3 January 13, 1993 assumptions delay the time that highly concentrated boron is injected into the core and, thus, exacerbate the return-to-power. ,
An additional single failure that could impact the core response to a MSLB was identified. Following a reactor trip, plant elec-trical loads will fast transfer from the normal station services-transformer to the reserve station services transformer.- failure of bus 24G would result in power remaining to buses 25A and B which power the reactor coolant pumps. However, power would be lost to the vital buses 24C and D until the diesels are started. -
The HPSI pumps would be sequenced on after the diesels have energized the buses.
This single failure affects cases when offsite power is available by adding a diesel start time delay to the actuation of the safety injection system (SIS). This single failure assumption could be morc limiting than the limiting analysis of record (i.e., loss of one HPSI). For cases with offsite power available, the analysis of record supports a HPSI delay time of 30 seconds. With the plant changes implemented in the 1992 refueling outage, the Technical Specification time delay that includes the start of a diesel generator is 25 seconds for the HPSI pumps. This revised time delay is clearly bounded by the analysis of record.
Additionally, the reactivity transient will be less limiting relative to the analysis of record since two HPSI pumps would be available to deliver borated water to the core sooner and with more capacity. Thus, failure of nonvital bus 24G is considered less limiting than the loss of a HPSI pump in the analysis of record,
- c. The primary effect of an extended water supply to the affected steam generator is an increase in the duration of-the event with potentially a higher return-to-power. Extended water supply can also slightly reduce the primary system pressure which can lead to an earlier SIAS. The analysis of record specifically includes the effect of runout flow from the auxiliary feedwater -system after 180 seconds.
A failure of the feedline isolation system can occur _ only' at-initial power conditions above HZP when the main feedwater system is operating (e.g., HFP). A single failure in the feedwater isolation system that results in extended feedwater supply will-contribute to the cooldown of the primary system by prolonging the event which can lead to a -high return-to-power._ The changes to the feedline isolation system in the 1992 refueling outage ensure a redundant valve alignment and a change to the power supply to ,
the feedwater regulating valves (vital instrument AC powered from vital DC buses). Also, the feedwater regulating . valve bypass w m -,, -w
._ ey+y. 4 i% m-y--gyy -~,u. ,-vw- -y9 -p.p 4 --m-g y,-my,,,e
U.S. Nuclear Regulatory Commission B14330/ Attachment 1/Page 4 January 13, 1993 valves fall in a closed position if power is lost. Thus, a single failure that leads to a continuous supply of feedwater has a low probability of occurrence. However, if a feedline failure that results in an extended water supply were to occur, the additional overcooling would be counteracted by the slower energy _' release through the integral flow restriction in the replacement steam generators relative to the analysis of record. Also, diesel generato: start on SIAS, faster diesel generator start time, and the availability of the HPSI pumps would lead to a less severe reactivity transient compared to-the bounding case for _the analy-sis of record,
- d. The hot channel factors corresponding to the most reactive rod in the fully withdrawn position and Minimum Departure from Nucleate Boiling Ratio (MDNBR) values for the analyzed events are given below. MDNBRs were calculated with the modified Barnett correla-tion with a 95/95 limit of 1.135. The limiting MDNBR case for the analysis of record was initiated from HZP with a coincident loss of offsite J,n er. The limiting peak linear heat rate (LHR) was calculated for the HZP case with offsite power and is less than the conservative centerline melt limit of 21 kW/ft. j Offsite Hot Channel Max. LHR Initial Power Power Factor kW/ft MDNBR HZP Y 10.7 20.9 2.40
}
HZP N ,
8.7 16.5 1.18 HFP Y 12.8 17.1 3.00 HFP N 13.5 5.7 4.60 _
In . summary, the existing MSLB analysis for core response predicts a_
slight return to power, but with no failure of fuel. Assumptions-used _
in the analysis were _ sufficiently bounding that return to power is already maximized.
Those plant design changes which were determined necessary for contain-ment analysis have not been incorporated into the core response MSLB, although doing so would produce less impact to the core.
v.,-- - - - --
- - , . -----,y,- y,,-,,,-m- , ,, y - ---- -
i U.S. Nuclear Regulatory Commission B14330/ Attachment 1/Page 5 January 13, 1993 3.0 CORRECTIVE ACTIONS As discussed in Section 1.0, corrective actions were necessary to prevent exceeding the containment design pressure and temperature.
Since this condition was discovered while the plant was in Mode 1 at 100 percent pownr, interim action was necessary until permanent modifications could be implemented. As described in LER 91-010 00, dated November 15, 1991,"' the interim action included testing the main feedwater block valves to demonstrate adequate closure time and stationing a dedicated operator to close the main feedwater block valves following any reactor trip. The need for the dedicated operator was subsequently eliminated by the implementation of a short-term modification which caused automatic closure of the main feedwater block valves given a containment isolation actuation signal (CIAS). This action was described in LER 91-010 01, dated January 17, 1991. 28 That LER also informed the Staff that changes, for which hardware was available, would be installed during the 1992 refueling outage.
A description of the reanalysis to determine the effect-of MSLB on the containment peak pressure, which incorporates the modifications made during the 1992 refueling outage, is presented in Section 3.1. As noted in Section 2.0, the existing MSLB analysis for core response does not require reanalysis and the plant modifications, which were necessary to obtain an acceptable containment response, would have a beneficial effect on the core response.
3.1 MSLB CONTAINMENT ANALYSIS The analysis to determine the effect of a MSLB on the containment peak pressure and temperature was completed by ABB-CE in October 1992. This analysis reflects the design changes that were imple-mented in the 1992 refueling outage. Of these changes, the following were a direct result of the MSLB analysis:
- 1. Actuation of the main steam isolation signal (MSIS) upon a containment high pressure signal-(CHPS) of 4.75 psig;
- 2. Automatic closure of the feed regulating valve (FRV) block ,
valves (FW-42A and 42E) hnd the feed pump discharge valves (1) S. E. Scace letter to U.S. Nuclear Regulatory Commission, " Facility Operating License No. - DPR 65, Docket No. 50-336, Licensee Event Report 91-010-00,"- dated November 15, 1991.
(2) 5. E. Scace letter to U.S. Nuclear Regulatory Commission, " Facility-Operating License No. DPR-65, Docket No._ 50-336, Licensee Event .. Report 91-010 01," dated January 17, 1992.
E - N v 1- e w - q
1
. . i U.S. Nuclear Regulatory Commission B14330/ Attachment 1/Page 6 January 13, 1993 l
(FW 38A and 388) to include automatic actuation (shutting) l on an MSIS;
- 3. Automatic actuation of MSIS components from either MSI-A or MSI-B;
- 5. Actuation of emergency diesel generator (EDG)-start on SIAS.
In addition, the following Technical Specification changes have been made to ensure that the MSLB accident analysis is bounding:
- 1. Reduction of the minimum containment air recirculation (CAR) 1 Fans starting delay time with normal AC power available to 15 seconds;
- 2. Reduction of the minimum containment spray starting delay time with normal AC power available to 16 seconds;
- 3. Reduction of the minimum EDG starting time to 15 seconds;
- 4. Reduction of the feedwater block valves, FRVs, feed pump discharge valves and FRV bypass valves ~ closure ' times, including stroke time and signal delay time, to 14 seconds, and;
- 5. Reduction of the High-High containment pressure signals to 9.48 psig.
Method of Analysis A complete MSLB spectrum study has been performed to deteraine the limiting cases for peak containment-pressure and for environmental-profiles for Electrical Equipment Qualification (EEQ). The NRC-approved methodology associated with the ABB-CE SGN-III computer program was used. This methodology includes consideration for the following: (a) inclusion of the steam line and feed line volumes into the overall determination of blowdown volume-available; (b) determination of- temperature / pressure - expansion factors and manufacturing tolerances for the steam generators (SGs) and reactor coolant system -(RCS); (c) feed spiking due to the in-creasing pressure imbalance between the ruptured _ and intact SG; (d) inclusion of current core physics and thermal-hydra'ulic data; (e) inclusion of SG shell metal heat transfer as part of the energy release; and lastly, (f) a complete determination of the -
effects of different component single failures. _ during the
_ _ _ __ = _
"- P e- e* n e7+-
- I . .
U.S. Nuclear Regulatory Commission B14330/ Attachment 1/Page 7 January 13, 1993 accident. This methodology is consistent with the Standard Review Plan (SRP) guidance.
Maior Assumotions The major assumptions are as follows:
- 1. Offsite power is conservatively assumed to be available for most of the cases. This increases the primary to secondary heat transfer since _ the reactor coolant pumps- (RCPs)_ are operating. To verify this assum) tion, loss of offsite power cases were included as part of tie single failure analysis.
- 2. For determination of peak containment pressure, the initial containment pressure / temperature is conservatively - assumed to be at the Technical Specification maximum of 16.8 psia
. and 120'F. For determination of peak containment tempera--
ture, the initial containment pressure was conservatively assumed to be 14.7 psia. With a lower initial pressure, containment spray will be actuated later, resulting in a high containment temperature. Relative humidity is assumed at 30 percent except for the EEQ cases where--it is conservatively set at 100 percent.
- 3. Consistent with the NRC a> proved methodology, moisture carryover was determined by tie SGN-III computer code.
- 4. Feedwater spiking is accounted for by conservatively doubling the initial feedwater flow rate for each case.
- 5. Credit is taken for the main steam - nonreturn valves to.
prevent blowdown of the unaffected SG into the containment.
- 6. The maximum RCS flow rate was conservatively assumed to maximize the heat transfer from the primary to secondary side.
- 7. Auxiliary feedwater was conservatively assumed to ;initiatei at 180 seconds based upon the time delay of 3 minutes given in Technical Specification Table 3.3-5.
- 8. RCP heat was-included.
- 9. All actuation , signals are redundant and safety grade. :In some cases credit is taken for actuation of nonsafety grade components initiated by the safety grade- signals. ThisLis fully consistent with original license design bases where l
.. _o
U.S. Nuclear Regulatory Comission B14330/ Attachment 1/Page 8 January 13, 1993 nonsafety components have always been credited for feed isolation.
Results in order to determine the limiting conditions, four different spectrum studies were performed. These are as follows:
- 1. Power level and break size
- 2. Feed system single failures
- 3. Containment heat removal systems single failures
- 4. EEQ spectrum study A summary of the results of these studies is given in Table 1 and Figures 1 through 6. The results of the spectrum studies are sumarized below:
Power level and Break Size A comprehensive sensitivity- study was aerformed to determine the limiting break size for each power leve1. A sensitivity study was g needed because of the interaction of power level with SG inventory and moisture carryover. The limiting break size at a given power level is the largest break size that would result in a pure steam blowdown, since a pure steam blowdown results in the greatest amount of energy being transferred to the containment atmosphere in a short The limiting break sizes were shown to be 3.51 ft' period of time.at 50 percent power and above, 3.35 ft' at 25 percent, and 1.75 ft' at 0 percent. Results are shown as cases Al through A5 in Table 1. No single failures are assumed -in this set of Cases.
Feed System Sinale Failures A comprehensive single failure study was performed for feedwater system isolation. For each single failure, a range of power icvels was analyzed, using the insights from the Power Level / Break Size sensitivity study. The results of this sensitivity study are shown as Cases B1 through B4 in Table 1 and are briefly described below.
Bl. Feed Pump Failura to Trip - The failure of a feed pump to trip on MSIS re.alts in additional feedwater being pumped preferentially into the affected SG until the FRV or the
U.S. Nuclear Regulatory Commission B14330/ Attachment 1/Page 9 January 13, 1993 isolation valves shut. For power levels 50 percent and below, only one feedwater pump was assumed to be running _
when the accident commences.
B2.- Auxiliary Feedwater Regulating Valve Fails Open - A failed open auxiliary feedwater (AFW) regulating valve was assumed to result in the addition of the maximum flow from the two electrically driven AFW pumps to the affected SG. AFW was assumed to commence at 180 seconds. The AFW flow addition results in a slow but steady increase in containment pres-sure and temperature. Credit is taken for operator action at ten minutes to isolate AFW to the affected steam-generator. The results show that the operator has at least 15 minutes to isolate the affected generator.
B3. Feedwater Bypass Valve Fails Open -
The failed open feedwater bypass valve results in additional feedwater bettig pumped preferentially into the affected SG untti the feedwater pump discharge valves shut. In addition, even with the feed pump discharge valves shut, flashing in the feedwater lines continues. to add energy into the affected steam generator. This effect has been taken into account.
This case is the limiting MSLB for peak containment pres-sure. The plant response for this case is shown in Figures 1 through 5 and the sequence of events is given in Table 2.
- 84. Failure of Vital Instrument AC Bus VA-10 or VA This failure could prevent closure of the FRVs and loss of one train of the containment- heat removal systems. Feedwater addition to the affected SG will continue until closure of the feedwater. isolation valves.
Containment Heat Removal Systems Sinale Failures A comprehensive containment heat removal systems single failure study was perforned. For each single failure, a range of power levels was analynd, using the insights from the Power Level Break Size sensitivity study. The results of this sensitivity study are shown as Cases C1 through C5 in Table 1 and are briefly described below.
C1. Failure of Two CAR Fans to Start - This failure is bounded by Case B4 described above.
C2. Failure of One Spray Train to Start -
This failure is bounded by Case B4 described above.
l U.S. Nuclear Regulatory Commission I B14330/ Attachment 1/Page10 January 13, 1993 l l
i i
C3. Failure of the Vital Bus Transfer Mechanism - This failure !
results in a loss of the normal power supply for the vital l buses. Thus, initiation of the containment sprays and CAR :
fans is delayed until the EDGs are powering the vital buses I and auto sequencing has occurred. Since the FRV's- power supplies are powered from a vital DC power source, they are unaffected by this failure and will isolate'the affected SG.
The RCPs and certain other nonvital loads are also unaffected by this failure, which contributes to the severity of this accident by providing more rapid heat transfer from the primary to the affer.ted SG.
C4. Loss of Offsite Power with a Loss of One EDG - A loss of offsite power will result in loss of power to the RCPs, the ,
condensate pumps, and feedwater heater drains pumps. While 1 only one train of containment heat removal systems is available, the -loss of power to these pumps results in a greatly degraded heat transfer in the affected SG and less limiting results. Feedwater isolation will be unaffected since the FRY's power supplies are powered by vital DC supplies.
C5. Loss of Offsite Power with a loss of VA-10/20 - This case is similar to C4, with the exception of the effect on feedwater isolation to the affected SG. With this failure, there is ,
failure of the FRV and the other isolation valves to close.
However, with the loss of the condensate and feedwater heater drains pumps, feedwater addition to the affected SG is terminated. The effect of continued energy addition to-the affected SG from flashing in the feedwater lines has i been taken into account.
EE0 Spectrum Study In order to develop a bounding profile for EEQ, the sensitivity studies (with the exception of Case B2) were repeated with initial conditions selected to maximize peak containmect temperature. The major changes in assumptions for determining the bounding profile are as follows:
- 1. The initial containment pressure was reduced to 14.7 psia.
This results in the maximum delay in containment spray actuation.
L
- 2. The relative humidity was increased to 100 percent.
- 3. The guidelines of NUREG-0588' and IE-IN #84 90 were used to set ' the containment wall reevaporization at 8 percent and p
L
= - . . -
. .- , , , , . - , _ y.
3 l
I
~U.S. Nuclear _ Regulatory Commission i B14330/ Attachment 1/Page 11 January 13, 1993 the modeling of SG superheating as it passes the uncovered 1 portion of the SG tubes before exiting the break. l l
These assumptions result ir: some changes in the timing of j the sequence of events. 1he results of. these casos were used to generate the EEQ containment temperature profile.
Case D1 shows the limiting peak containment temperature.
l 4.0 Conclusion i i
The results in Table I show that for all-cases, the peak containment pressure i s less than the containment design pressure of 54 psig.
Figure 6 gives the bounding EEQ containment temaerature profile. - The peak temperature for this arofile is 426'F. It is noted that Technical Specification 5.2.2 specif< es the containment building temperature limit as 289'F. Because the containment atmosphere exceeds 289'F for only a short period of time, the containment building remains well below 289'F.
Further, NNECO has identified the need to revise the bases of various Technical Specifications to reflect the results of recent analyses and <
to improve consistency among the bases. ;
.iyy.. 7 .,m
TABLE 1 ,
! Peak Power Pressure Peak Temp Case Description Level (psig) ('F)
Al Base Case 102% 48.6 394 75% 49.0 3S9 A2 Base Case A3 Basa Case 50% 49.8 386 A4 Base Case 25% 47.2 379 0% 50.0 345 A5 Base Case Feed Pump Fails to Trip 50% 52.5 385 B1 0% 50.9 345 B2 AFW Regulating Valve Fails Open at 600 sec ,
B3 Feed Regulating Bypass Valve 50% 53.7 3SS Fails Open 102% 53.3 '
395 B4 Vital Bus Cabinet (VA-10/20) a Fails - Preventing FRV Closure I and Eliminating 1/2 of the sprays and Fans 0% 51.8 346 C1 Two Car Fans Fail to Start 51.8 394 C2 One Spray Train Fails to Start 102%
52.2 416 C3 Vital Bus Fast Transfer Fails 102%
25% 48.0 382 C4 Loss of Offsite Power and Loss of one Diesel 50% 51.20 386 C5 Loss of Offsite Power and Loss of VA-10/20 102% NA 426 DI EEQ Calculation - Vital Bus Fast Transfer Fails
U.S. Nucle'ar Regulatory Commission 1 B14330/ Attachment 1/Page 13 January 13, 1993 ;
TABLE 2 SEQUENCE O F EVENTS MP2-MSLB: FEED BYPASS FAILURE CASE 9 50% POWER EVENT !
TIME (seconds) SETPOINT/VALUE 0.00 MSLB occurs from 50% power, break size is 3.51 ft. .
0.01 Fesd spiking occurs which causes feed to ruptured SG to double from its initial flow. ,
1.90 Containment High Pressure Signal (CHPS) is 5.83 psig generated. This will cause a Reactor Trip with uncert, j
and MSIS after a 1.15 second signal delay.
3.05 Reactor trip and turbine trip occur.
An MSIS signal causes the feed pumps to trip off and FRV's, feed isolation valves begin closing. Feed flow begins remping down as the feed pumps coast down.
feed bypass valve fail open.
5.28 Containment High-High pressure signal 11.08 psig :
l (CHHPS) occurs, with uncert.-
l.
l 8.05 feed pump coastdown to " low flow" condition ,
complete,
, 15.90 feed isolation valves shut. Feed to the ruptured SG ceases.
16.90 Containment cooling fans energize.
Time based on CHPS + 15 second delay.
35.58 Containment spray flow commences.- 2600 gpm t Time based on-CHHPS + 30.3 seconds.for-pump- start, valve stroke- time,--and header fill time.
35.58 Peak containment temperature reached. 385 'F 133.98 - Peak containment pressure reached. 53.7 psig 180.00 Auxiliary feedwater flow commences 600 gpm, 120 'F-to the ruptured SG.
600.00 Problem time ends. -
._.,-~._-.a_,_,+.,_ _c ~ .._ _ .;.- - ._. u...-: -.____,;...
~.
U.S. Nuclear Regulatory Comission .
814330/ Attachment 1/Page 14 January 13, 1993 FIGURE 1 .
MSLB CONTAINMENT ANALYSIS 50% Power, Feed Bypass Failure Containment Pressure vs Time 60 -
i -
8 -
50 -
i 9 40 h m .
E _
E2 30 -
m m
m w
~
20 ._
i 10 -
t i
0 '' ' * ' ' ' * ' n s t t l
0 50 100 150 200 250 300 350 400 450 500 550 Time, sec
~.
U.S. Nuclear Regulatory Comissica .
B14330/ Attachment 1/Page 15 >
January 73, 1993 FIGURE 2 .
MSLB CONTAINMENT ANALYSIS 50% Power, Feed Bypass Failure Containment Temperature vs Time 500 _
400 -
t.
g300 _
~
3 .
E O
Q.200-E 0 -
100 --
0 ''''''''''''''''''''''''''''''''''''''''''''''''''
O 50 100 150 200 250 300 350 400 450 500 550 Time, sec
^
i U.S. Nuclear Regulatory Comission .
B14330/ Attachment 1/Page 16 _
I January .13, 1993 ,
- FIGURE 3 .
MSLB CONTAINMENT ANALYSIS 50% Power, Feed Bypass Failure SG #1 Pressure vs Time 1,200 j' -
{
1,000 h-e 800 - -
m -
C. -
~
a 6
- .600 -
t 3,
u -
09 -
O u . ,
0--
400 -
200 l 0 ' ' -
0 50 100 150 200 '250 ~300 350 400 450 500 550 600 Time, sec
. ._ _ __-_________z_____--__
U.S. Nuclear Regulatory Comnission .!
- , - B14330/ Attachment 1/Page 17 ~. ,
- 2 RJanuary 13, 1993 '
4 FIGURE 4
- i. .
t i MSLB CONTAINMENT ANALYSIS '
l ,
50% Power, Feed Bypass Failure i
SG '#2 Pressure vs Time i
. 1,000'
_. I t
)-
800
~
i 4
s-I i
i i soO -
b -
, "3 t m -
i N
e-
- 400 -
4 Q. -
i r
i i 1
i 200 -
4
- t'
- t. .
W 1' -
- '. . . .f....f....'I....l..'..'l..1....t....'f...'.f...1....f....l 0 50 100 150 200 .250 300 350 400 450 500 550 600 Time, sec 6 .
.~,,_
, , . - - , . .-,-s -
. . . , . , , . . , , ,.--,-.m.- - . . ..s.. _ .x . - . .. - . - - , ,- - .. ._sn. ....--__J
i U.S.. Nuclear Regulatory Comission ,
B14330/ Attachment'1/Page 18- *
, January 13, 1993 I '.
l FIGURE 5 . .
c !
. MSLB CONTAINMENT ANALYSIS L 50% Power, Feed Bypass Failure
- . Two Phase Ruptured SG Water Volume vs Time ,
7 -
i .
p _
m6 - -
i n
-C i m -
as -
35 o _- s
.c -
. 6 : j'
- c4 -
e -
E :
.2 O 3 -
m
! 34 2 -
g -
0 -
co 1 -
i t
~
-0.""['
- 0. -50 100 150 200 250- 300 350 400 450 500 550 3 Time, sec 2 :l
- . = -,9 w
- y - -- s +e, , w .: ,-.-.m ,. . . U.. -w -w-t
U.S. Nuclear Regulatory Comission B14330/ Attachment 1/Page 19 January 13, 1993 -
FIGURE 6 MP2 MSLB COMPOSITE EEQ CURVE 500 Maximum Composite Temperature 400 -
u_ _
6300 -
5 -
E
- o. -
E 200 -
0 _
100 -
0 O.1 1 10 100 1,000 Time, sec
Docket No. 50-336 <
B14330 b
i Attachment 2 Augmented Legend for EQ Profile Figure January 1993
-e t y q. sg -.wew ,a - -- g--w. g gy - .g-.- ,-p ,,
U.S. Nucleaf Regulatory Commission <
Bl4330/ Attachment 2/Page 1 l January 13, 1993 Auomented Leaend of E0 Profile Previously Submitted l
103 General Atomic Rad Monitors 121 Rockbestos Coaxlat 107FR Kertte FR cable 116 ideal Setscrews 1
107HTK Korite H IX Cable 111 ASCO Soloniods 1
125 Namco Umit Switch 122 Lirnitorque Motor Operators 115 Westinghouse Motors 102 Conax I enetrations 4
-135 Anamnda Cable <
134 Rosemount Transmitters 128 Litton Connectors 101 Gems-Delaval Transmitters 120 Utton Connectors 'l 4
.119 Weidmuller Terminal Blocks i
- - -