View source for B12632, Forwards Responses to Questions Raised During 870630 Meeting Re Nusco 151 Concerning Reanalysis of non-LOCA Transients & Accidents in Chapter 10 of Facility Description & Safety Analysis

{{Adams
| number = ML20236N159
| issue date = 08/07/1987
| title = Forwards Responses to Questions Raised During 870630 Meeting Re Nusco 151 Concerning Reanalysis of non-LOCA Transients & Accidents in Chapter 10 of Facility Description & Safety Analysis
| author name = Mroczka E
| author affiliation = CONNECTICUT YANKEE ATOMIC POWER CO.
| addressee name = 
| addressee affiliation = NRC OFFICE OF ADMINISTRATION &amp; RESOURCES MANAGEMENT (ARM)
| docket = 05000213
| license number = 
| contact person = 
| document report number = B12632, NUDOCS 8708110392
| document type = CORRESPONDENCE-LETTERS, INCOMING CORRESPONDENCE, UTILITY TO NRC
| page count = 19
}}

=Text=
{{#Wiki_filter:- - _ - _ _ . - _ _ _
e CONNECTICUT YANKEE ATO MIC POWER COMPANY B E R L I N, CONNECTICUT P O. Box 270 e HARTFORD. CONNECTICUT 06141-0270
[".$                                                              August 7,1987 Docket No. 50-213 B12632 Re: 19CFR50.34(b)
U.S. Nuclear Regulatory Commission Attn:    Document Control Desk Washington, D.C. 20555

==Reference:==
(1)  3. F. Opeka letter to C.1. Grimes, "Haddam Neck Plant Reanalysis   of Non-LOCA Design                  Basis     Accidents, NUSCO 151," dated June 30,1986.
Gentlemen:
Haddam Neck Plant Additional Information -
Reanalysis of Non-LOCA Design Basis Accidents On June 30, 1986, Northeast Utilities Service Company (NUSCO), on behalf of the Connecticut Yankee Atomic Power Company, submitted Topical Report NUSCO 151 which described the reanalysis of the non-LOCA transients and accidents contained in Chapter 10 of the Haddam Neck Facility Description and Safety Analysis. The purpose of that submittal was described in Reference (1).
On June 30,1987, a meeting was held between the NRC Staff, its consultants for the review of NUSCO 151 and NUSCO representatives. At the conclusion of the meeting, it was agreed that NUSCO would provide a written response to a number of questions that were raised during the meeting, in accordance with that commitment, these responses are provided in Attachment 1. We trust the Staff will find these responses adequate to resolve their concerns.
Very truly yours, CONNECTICUT YANKEE ATOMIC POWER COMPANY M<, lv E. 3. Mrgcika '
Senior Vice Preside   nt          /
cc:   W. T. Russell, Region 1 Administrator                                                      gp F. M. Akstulewicz, NRC Project Manager, Haddam Neck Plant
: 3. T. Shediosky, Resident inspector, Haddam Neck Plant                                    (8 8708110392 870007 PDR                     ADOCK 05000213 P                                        PDR       ,

      *).
   /-
Docket No. 50-213 4,
         .                                        B12632 Attachment 1 AdditionalInformation Regarding the Reanalysis of the Non-LOCA Design Basis Accidents and its Revisions August 1987
__-____-____-_____-__________A

Introd uction On Juna 30, 1986, Rev. O of NUSCO-151 was submitted to the NRC, describing the results of non-LOCA design basis accidents. These analyses were submitted in order to resolve open isstes related to existing design basis analyses identified during the Systematic Evaluation Program review and to upgrade existing design basis analyses with current nuclear code me thodology.
Subseqtent to the docketing of NUSCO-151, Rev. O, the loss of normal
    ,   feedwater flow analysis had to be revised to account for minimun auxiliary feedwater flow rates that were less conservative than what was assuned in the J une 30, 1986 s ubmittal . The mvised loss of normal feedwater flow analysis was submitted to the NRC on March 10, 1987 as Rev. I to NUSCO-151.
The Haddam Neck Plant is currently shutdown for the C/cle 15 reload.                                                                             The Cycle 15 reload will add 56 fresh fiel assemblies to the core instead of the usual 52 assemblies. As a result, several physics paraneters are more limiting than Cycle 14 and the assumptions used in Rev. O and Rev.1 of                                                                                  1 NUSCO-151. This required that the five affected accident analyses be re vised. These analyses were the uncontrolled rod withdrawal from subcritical, steam line break, boron dilution, rod cl uster control assembly ejection and dropped rod. The revised analyses were submitted to the NRC on May 8,1987 as Rev. 2 to NUSCO-151.
Of the revised analyses, significant changes were made only for the rod withdrawal from subcritical, horon dil ution and steamline break.                                        No significant changes were made for the rod ejection or the rod drop. For the control rod ejection, the anthodology for eval uation was unchanged and the results calculated for Cycle 15 were less limiting than those presented in Revision 0. Because so many of the physics parameters changed, a new analysis was submitted. While the prediction of the nunber of pins in DNB decreased, the assumptions for the radiological evaluation were unchanged and thus are overly conservative. This represents additional margin in the analysis. For the rod drop, the discussion was changod just to reflect a change in te rminology. Previously, the discussion had discussed a penalty factor placed upon the normal peaking factor to take into account the dropped rod. This has been deleted since the dropped rod peaking factors we re used directly. The details of the changes made to the rod withdrawal from subcritical, boron dilution and steam line break are described in Revision 2 and are sunmarized in the discussion provided below.
The accident analyses requiring approval for the startup of Cycle 15 are related to requimd Technical Specification changes. The se accidents were i
identified in the cover letter transmitting Revision 2 to NUSCO-151, and are stmnarized as follows:
1 Uncontrolled Rod Withdrawal from Subcritical The differential rod worth has increased substantially for Cycle 15 heyond the current design basis assunption. The current analysis cannot ba extrapolated to justify the higher diffemntial rod worth. As a msult, consistent with the new analysis, Technical Specifications are proposed to require that the startup rate trip

                                                                                 ~ be - operable' (T.S. 2.4 and Table 3.9-1) whenever the reactor trip                                     ,
breakers am closed and the control rod drive 11ft coils energized, up to the P-7 interlock (10% power). The startup rate trip is now credited for this accidert. Additionally, the Technical Specifications which govern the required nunber of
                      . operating.RCS loops during subcritical conditions (T.S. 3.3) were revised to bel consistent with the analysis.
O
                      ~ Loss of Flow it is- possible.that' the current Technical Specification limit for RCS flow rate (T.S. 3.20)'for four loop operation may not he met considering that there is little available margin and some amount of steam generator tube pl ugging is anticipated. The current x
design basis' accident analysis cannot be extrapolated for lower flow.. -Thus, a new analysis is needed to j ustify the lower proposed RCS flow rate Technical Specification. Since the loss of flow transient'is the limiting moderate fmqmncy DNBR transient, approval of. the new analysis (which assuned a lower flow rate) will allow f ull powe r ope ration.. In addition, a change is being made to the three loop low flow trip setpoint (T.S. 2.4) to eliminate the need for calibration of the setpoint when going from four loop operation to three loop operation. This is consistent with NUSCO 151'but' is not consistent with the current design basis.
O B'oron Dil'ution The shutdown margin (T.S. 3.10.1) is being reduced de to reload physics parameters. Since this will impact the boron dilution accident,' approval of the new analysis is necessary.
O Steam Line Bmak The steam line break analysis is also affected by the change in shutdown margin (T.S. 3.10.1). Since the limiting condition for a steam line break is end of cycle, approval of the steam line break.
analysis is not requimd for the startup of Cycle 15. Howe ve r,
                      . approval of the steam line break analysis is necessary for cperation beyond the middle of the cycle.
                      . Dropped Rod Although it does not directly affect any of the Technical Specifications, approval of the dropped rod analysis is necessary to support disabling of the turbine runback feature. The current analysis takes credit for the turbine runback, and it cannot be extrapolated for Cycle 15. Based on past experience , elimination of this feature should reduce the nunber of unnecessary plant transients caused by spurious dropped rod signals in the future.
Approval of the analysis is necessary for this.

c L
The following paragraphs provide additional information related to other specific qu3stions on NUSCO-151.
: 1. Nodalization used in the RETRAN Analysis In general, the base nodalization used for each transient analysis in the Non-LOCA Transient Analysis Report NUSCO 151 is identical to the RETRAN model shown on Figure II. A.1 in the NUSCO Thermal Hydraulic Model Qualification Vol une I (RETRAN) Topical Report (NUSCO 140-1).
The same nodalization diagram is shown on Figure 1. This model contains 46 vol umes, 75 j tnctions and 25 heat slabs. For 3 loop analysis, loop 2 was deleted from the 4 loop model. The 3 loop nodalization is shown' on Figure 2.
The base model has been developed based on the nesults of the modeling sensitivity studies performed in the RETRAN Topical Report, NUSCO
      .140-1, which represent most of the Chapter 10 transients. The transients' conside red incl uded:
: 1. Steam Line Break
: 2. Loss of Feedwater
: 3. Loss of Load
: 4. Loss of Flow, and
: 5. Dropped Rod In NUSCO 140-1, noding sensitivity studies were provided only for the primary and secondary side of the steam generators, since these are the only regions where nodalization can have a significant impact for the analyzed transients. As discussed below, increasing the nunter of nodes used to model the RCS has very little impact on the predicted re sponse .
In the base model, the core is modeled using point kinetics with one conductor and one vol un2     For the type of transients analyzed in NUSCO 151, the single core model coupled with the point kinetics model is conservative and is consistent with the reactor physics input.
Increasing the nLmber of core nodes with a point kinetics model, will not improve accuracy and will make it more difficult to maintain consistency with the physics input.
The nunber of nodes used for the reactor vessel and loops only affects the calculation of loop transit tine.       In gene ral, loop transit time -is not important for the transients analyzed. In addition, the nodalization used in the analysis is sufficient for modeling loop transit ef fects.

l
                'This has been confirmed by the benchmark analyses for the operational events at CY.- For the 30%' load rejection and the partial loss of feedwater events, the RETRAN analyses using the same base model adequately pmdicted the actual primary pressure and temperature transient response (Section V.B and V.C. in NUSCO 140-1).
               . Based on this ' discussion and-sche sensitivity studies performed (the detailed discussion is. provided in' NUSCO 140-1), we believe that the base model adequately repmsents the CY plant. Howeve r, some' modeling changes were made for the. startup of an idled loop, steam generator                                                                                                           i tube rupture and the S!.B transient, to provide more conservative RETRAN calc ulations.
These- modeling changes made. in the nodalization and assunption are discussed below:
For startup of an idled loop analysis, a split core model'was employed to account for the effeet of imperfeet thermal mixing during the transient. The vessel nodalization is shown on Figure 3.
To make the split core model, the downcomer, plena and com vol unes were divided. into 3/4 and 1/4 parts and scaled proportionally. The moderator feedback was taken from the cold side and applied to both sections of the core. The analysis results showed that incorporating the split com model is mom limiting with respect to MDNBR.
For SLB' analysis, we have supplied information to j ustify the use of a bubble rise- velocity of 1 to 3 ft./sec with a realistic U-tube                                                                                                           '
heat s1ab height in NUSCO 140-1. However, using an infinite velocity with a slab height that ensures that no U-tube uncovery occurs for the time period of interest, is mom . conservative. We decided to take the more conservative approach. In the NUSCO 151 report, we used. an infinite bubble rise velocity. For the limiting case s , the SG t ube s we re al so . ke pt cove m d , d tn to d i ve rsi on of feedwater to the affected SG, which incmased its inventory. Th us ,                                                                                                         ,
both the cooldown rate and- the total cooldown wem conservatively                                                                                                           j maximize d.
O In the steam generator tube ruptum analyses, vol unes and j unctions wem added to model the steam generator tube and primary-to-secondary flow paths.
Besides the modeling changes discussed above, no significant modifications wem made to the base model for the remainder of the                                                                                                               !
Chapter 10 transients.

h -i
: 2.                       Differences in Analysis Assumptions Between Revision 0 and Revision 2 of Non LOCA Transient Analysis Report, NUSCO 151 As~ explained 'in the submittal. ofg revisions _to the Reanalysis of the -
Non-LOCA design basis accidents                                                                                               , four analyses were revised because the assumptions made. in. Revision 0 did note provide enough~ margin for a r
few of the physics parameters for the Cycle 15 reload design. The most significant parameters for which additional margin is required are shutdown margin and rod worth. For most physics parameters, the Cycle
                                                                         -15 design .is bomded by the Revision 0 assumptions. For the bounded parameters, such as moderator coefficient, Doppler coefficient and power distribution, no change' was made to the analysis assunption, even in those analyses resubmitted._ This is reflected in the revised Table 4.0-1 in that there are' very few changes in the Table,, In fact, only the change ~ in minimun scram reactivity insertion worth reflects a
                                                                                                                                                                                                              ~
change in the' analysis assumptions. The two other changes, minimun RCS flow rate and radial peaking factors, am being made to conform to the format of the _new Technical Specifications.
The vessel flowrate measured following the 1986 ref mling outage was 259,000 gpm, incl uding all measurement uncertainties. The margin to the current Technical Specification limit of 257,000 gpm is very small
_ (<1%) and leaves no margin for future steam generator tube pl ugging.
The proposed . vessel flowrate limit of 246,000 gpm provides sufficient margin- to accommodate expected tube pl ugging during the 1987 outage.
For minimun RCS flow rate, in Revision 0, Revision 1 and Revision 2, the following RCS flow rate was used:
4 loop                                                               3 loop core flow rate , gpm                                                                              = 233,870                                                                          184,730 bypass . flow rate , gpm = -23,130                                                                                                                                                           18,270 257,000                                                              203,000 The bypass flow rates noted above are 9% of the total flow rate. A reevaluation of the bypass flow rate has determined that the bypass flow rate is actually 4.5% of the total flow rate.                                                                                                                                                                            The revised bypass flow allowed                                         echnical Specification change that was approved by the NRC in 1986 (1) Letter from E. J. Mroczka to the Nuclear Regulatory Commission "Haddam Neck Plant Revisions, to Reanalysis of Non-LOCA Design Basis. Accidents", May 8,1987.                                                                                                                                                                                                                 ,
(2) Letter from F. M. Akstulewicz to J. F. Opeka, " Cycle 15 Relaod Technical Specifications", April 14, 1986 u.__.      _ . _ _ _ _ _ _        ___ . . . . _ _ _ . _ _ . . _ _ _ _ _ _ _                              . _ _ _ _ _ _ _ _ . _ _                           _ _ _ _ _ . _ . _ . _ _ _ _ _ _ _ _ _ _              _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

Thus, the ' bypass flow rate was over estimated in the. reanalysis and should have been 11,070 gpm for four loop and 8,730 gpm for 3 loop, giving a total flow of '244,940 gpm for four loop and 193,460 gpm for three loop. ..The use of a slightly. higher total flow has a negligible impact on the system response.- The core flow rate is the parameter used in the DNBR . analysis and the core flow rates.given above 'were used in VIPRE.: Thus, the Technical- Specification limits of 246,000 gpm for-four loop and 194,000 gpm for three loop are consistent with the analyses in Revision 0, Revision 1 and Revision 2. Table 4.0-1 has been revised to be consistent with the proposed Technical Specification limits.
The radial peaking factors have been changed to be consistent with the new standard Technical Specification definition of the' parameter. The radial peaking factors mported in Revision 0 were for an all rods out condition. This was. adjusted in the analysis to take into account rod insertion ~ at power. . The radial peaking factors reported in Revision 2 take into account rod insertioa, consistent with the Technical Spe ci fication ' de finition. Thus, the change in the table does not represent a new assunption but repmsents a different way to present the peaking factors.
The physics parameters have been changed for only four accidents. Tte specific changes are shown in Tables 1 through 4. For all other
                                       ' accidents, the physics parameters were unchanged.
: 3. De1ay Times The reactor protection system (RPS) performance was modeled using two times, channel response time and delay time. For the RPS trip functions, the delay times were obtained, where available, from Section 10.1 of the Haddam Neck Facility Description and Safety Analysis (FDSA). These delay times are the same as were used in earlier, approved analyses. For additional conservatism, channel response times were also modeled. Response time is used to model the instrumentation sensor response to changes in the monitored parameter.. This was modeled in RETRAN using a " LAG" control block. This lags the output of the control' block behind the actual RETRAN calculated val m. This control block output was then used in the RETRAN trip cards to model the RPS trips. The RPS trip cards used the delay time to f urther delay the trip. The ' delay time is the time between the sensor generating a trip signal and the control rods beginning to fall.                                                                 i i
Table 5 shows for each RPS function, the delay times assumed in the Chapter 10 reanalysis and in the FDSA analyses, and the channel response time used in the reanalysis.
It is noted that the Reactor Protection System is planned to be L                                        replaced during the 1987 and 1989 ref eling outages. An additional L                                        advantage of the conservative modeling described above, in which both channel response time and delay time are modeled, is that is provides some added margin for the design specifications of the new RPS.

1 The channel response and delay times provided on the attached table and Emodeled. in RETRAN as described above, are judged to be reasonable and acceptable, are . consistent 'with thosef used in previous analyses and in the FDSA, and are typical of.the times. used for other plants.
: 14. - Operator Actions Operator actions were modeled in'some of the transients discussed in
               -NUSCO-151. While most of.the actions help to mitigate the transient, some of. them'make the transient more severe. The. following paragraphs
                ' discuss the10perator action and times used for each transient in which operator actions were modeled.
Uncontrolled Rod Withdrawal from Subcritical The analyses. asssningLthe 2 decades per minute (DPM) rod stop functions
_ credit operatorf action to manually. scram the reactor. .The f unctioning of t he rod stop . na s ul ts in a . 2 DPM powe r ramp. Based on the ' alarms available to the operator, high cotnt' rate and rod stop alarm,. the operator:is. assumed to scram the reactor 5 minties' after the rod stop alarm. This action is accomplished from the control. room.. For reactivity-insertion rates that wouldn't nesult in a. rod stop, a significantly longer ~ tine. is available for the operator to scram the re actor. Even if no credit is taken for operator action, it has been demonstrated that no pins in DNB will occur since backup protection is provided by the overpower trip.
Boron ' Dil ution The analyses asstne that the operator terminates the dil ution prior to
               ' loss of shutdown margin.         This action .is accomplished from the control room. The analyses calculate the times to loss of shutdown margin for the different modes of operation. The time between the first safety
              -grade alarm and loss of shutdown margin are then compared to the acceptance criteria in SRP 15.4.6. The time available is greater than 30 mintdes for nefueling and greater than 15 minties for the other mode s.
Excess Feedwater -
This' analysis credits the operator actions of tripping the reactor and terminating the excess feedwater flow. These actions are accomplished in.the control room. ' The minimtm operator action tine credited in the analysis:is.one mintre after initiation of the event.                                This time is                  ,
consistent- with the FDSA analysis of this event,                                                                   i l
Steam Line Break Event
.               These analyses credit the operator action of terminating au<iliary feedwater' flow to the faulted steam generator ten mintdes after initiation of the event. This action is accomplished from the control room. The tine of peak core power is-less than ten mintres from
               ' initiation of the event for all cases analyzed.

n      n
,              Steam Generator Tube Rupture-Several operator actions are modeled in this event. All of the actions are accomplished from the control room. 'The operator is conservatively
:assuned to trip the reactor ten seconds aiter initiation of the tube r upt ure . This~ results in mom atmospheric releases from the steam generator and-less through the condenser. This increases the iodine released to the environment and, thereby, the. radiological conse g m nce s. The operator is assuned .to start both charging punps one minute after reactor' trip. The charging punps are not automatically started following loss of offsite power. Starting the charging punps wili keep reactor coolant system pressum higher resulting in increased primary-to-secondary leakage. The operator is assuned to isolate 1 au<iliary feedwater to the. affected. steam generator ten minutes after reactor trip. The iodine concentration in- the affected steam generator
 ,            would be: lower if auxiliary feedwater were not stopped. The operator is assuned to terminate primary-to-secondary ' leakage thirty minutes         I after initiation of the event.      This is accomplished by closing the loop isolation valves 'in the loop with the ruptumd steam generator.
Between thirty minutes and'two hours the reactor coolant system is cooled down .to 50 F below the no-load average temperature. Between two hours and.eight. hours the reactor coolant system is cooled down to Residual Heat Removal (RHR) system entry conditions. After this time ,
f urther system cooldown is accomplished by the RHR system.
Loss of Normal Feedwater Flow The analyses assune that auxiliary feedwater is dither automatically initiated or manually initiated, at ten minutes, depending on which occurs first. Manual initiation can be accomplished from the control room. For three loop operation manual auxiliary feedwater initiation was, cre dited.
: 5. - Shutdown Margin Determination The shutdown margin for each operational mode is initially selected based on the experience of past cycle requirements and the known changes inL specific physics parameters for the fwl cycle being eval uated. The steamline break analysis is then performed using the shutdown margin assunption. Based on past experience , the steamline break analysis results have determined the shutdown margin mquirement
             ' for the operational modes (modes 1-3).
             -The boron dil ution event is then analyzed for modes 1-6. For the shutdown modes (modes 4-6), this event has typically governed the shutdown margin requimment. The- acceptability of an assuned shutdown margin da ing the boron dilution event is based on minimun times for operator action (i.e. tensinate the dilution) of 15 minutes for modes 1 through 5 and 30 minutes for mode 6.

r-                                                                           -        -------
                                             -g-For any mode, if the initially selected shutdown margin results in unacceptable analysis results, the required shutdown margin would be adj usted and the transients reanalyzed. For the shutdown modes, the Cycle 15 shutdown margin requirements were adjusted once in order to obtain acceptable results for the boron dil ution events. The initially selected shutdown margins for the operational modes did not require adj ustment based on the results of the steamline break event.
: 6. Boron Dil ution Analysis A significant change has been made between Revision 0 and Revision 2 of Table 4.3-1, Time From First Alarm To Criticality, because credit is no longer being taken for the non safety grade alarms. In Revision 0, credit was taken for the high flow alarm and the VCT high level alarm..
Flow rates greater than 20 gpm would generate an immediate high flow alarm. For flow rates less than 20 gpm, the VCT high level alarm would occur after 37. minutes of dil ution. Since the time to alarm was significantly different, results were presented for both 180 gpm dil ution and 20 gpm dil ution rates.
In Revision 2, credit is taken only for the doubling alarm from the source range monitors. In this instance, the highest dilution flow rate is bounding since a lower dilution flow rate will increase the time between alarm and loss of criticality. Thus, results only for a 180 gpm dil ution rate are pmsented. In Table 4.3-1, different cases of isolated loops and RHR loops available are pmsented to take into account the variations in vol tme available for dil ution. The time to criticality is the time mquired for the dilution to reduce the minimun shutdown margin to 0. The col unn in Table 4.3-1 labeled Time for Operator Action is the time available for the operator action; that is, the time between the first alarm and the loss of shutdown margin.      A time for operator action greater than 15 minutes for modes 1-5 and 30 minutes for mode 6 was used as the acceptance criteria. The se time s are the acceptance criteria given in SRP 15.4.6.
l i
1

   ., . u
: 7. Steam Line Break Methodology
                             - The steam line bmak methodology applied in the revised analysis                                                                     )
was further explained in the response to the qmstions on RETRAN I)                                                                     .
The.'statepoint eval uation that confirms the power and reactivity predicted by RETRAN is consistent with the standard Westinghouse methodology. This' methodology .has been described in tgNUSCO Topical Report 152 Physics Methodology for PWR Reload Design                                             that has been submitted for NRC review. This methodology is identical to thaf4) (5) described in the Westinghouse topica                for physics methodology for which the NRC has provided SERs In this methodology, the cold leg temperatures for the affected and unaffected steam generators along with pressure and reactivity from RETRAN are - used in 'a 3-D physics code to predict the power level. The.
mactivity feedback modeling in RETRAN is then adjusted until agreement is reached between the 3-D physics code and RETRAN. The 3-D physics code is a closed channel model; therefore, no mixing (i.e., cross flow) in the com 'is modeled. The inlet temperatum distribution is based upon conservation of mass and energy and is calculated given the cold leg temperature, inlet pressum and flow rate provided by RETRAN. The cold leg temperatures used as the basis for the inlet temperture distribution used in the 3-D physics code are selected to be the unmixed cold leg temperatures before entering the inlet plenun. We recognize that (1). Letter from E. J. Mroczka to the Nuclear Regulatory Commission                                                                         l "Haddam Neck Plant Revisions to Reanalysis of Non-LOCA Design                                                                        i Basis Accidents", May 8,1987 (2) Letter from E. J. Mroczka to the Nuclear Regulatory Commission                                                                          l "Haddam Neck Plant Response to Reqmsts for Additional Information                                                                     '
l-                                  on RETRAN", May 7,1987 (3) Letter from J. F. Opeka to C. I. Grimes, "Haddam Neck Plant Millstone Nuclear Power Station Unit Nos. 2 and 3 Physics                                                                             j Methodology for PWR Reload Design", September.12,1986                                                                                 4 (4) Bordelon ,    F. M. , et. al, " Westinghouse Reload Safety Eval uation                                                                  i Methodology", WCAP-9272 (WCAP-9272 P-A, ), July 1985                                                                                 j 1
(5) Camden, T. M. et. al, "PALADON - Westinghouse Nodal Computer Code"                                                                       ,
WCAP-9485 (Proprietary), December 1978 and Supplement 1                                                                               1 WCAP-9485-A (Proprietary) September 1987                                                                                              2 (6) Letter from C. O. Thomas to E. P. Rahe " Acceptance for Referencing of Licensing Topical Report WCAP-9272(P)/9272(NP). We stinghouse Reload Safety Eval uation Methodology", May 28, 1985.
L_---       - _ - - -

l cmdit can be taken for some mixing in the inlet plenun, and that using the unmixed cold leg temperatures is overly conservative. Howeve r, since acceptable results were obtained with this assunption, the over conservatism repmsents additional margin available.
Finally, because of the low flow and low pressure conditions, the McBeth correlation was used instead of the W3-L correlation for prediction of DNBR.
l l
1
_ . _ _ _ . _.__U

      .                                                                                                              )
1 Table 1: Change in Assumptions for the Uncontrolled Rod Withdrawal from Subcritical Paramete r        Revision 0      Revision 2             Di sc tssion
                                                                                                                     ]
Di f fe rential        90            135            Increase is needed for the Rod worth,                                         Cycle 15 design                                        ,
pcm/ inch                                                                                                '
        ' Trip f unction ove rpowe r         start up        Because of the increase in trip             rate trip       the differential rod worth, (safety         credit for an earlier trip grade )         is necessary to show acceptable nesults. Add SUR trip to T.S. 2.4 and revise T.S. Table 3.9-1.
Nunber of RC            3               2          An added benefit of taking ptmps running                                      credit for the startup rate trip is that only two operating punps are necessary to meet the DNBR limit. Requires change to T.S. 3.3 Rod block on       not taken         modeled with The rod block must be and without   taken into account because ,
the rod block if operable, it will prevent exceeding the startup rate setpoint and reactor will trip on high powe r. However, rate of power increase is slow, allowing time for operator action.
t                    --

1 TABLE 2:   Change in Asstmption for the Boron Dil ution Event Paramete r  Revision 0     Revision 2          Disc ussion Al arm      Credit          Credit is taken    The accident criterion was taken       only for the       could be met without for non-        safety grade       credit for the non safety grade   doubling alarm      safety grade alarms alarms          for the source     and the analysis is range monitors     also more conservative Sh utdown   Mode 1-2 1700 Mode 1-3             Changes necessary for margin, pcm Mode 3    2000 four loop 1800      the Cycle 15 design Mode 4-5 2500 three loop 2600 Mode 6    5500 Mode 4-5 2600 Mode 6     6000 i
l

1 l
Table 3: Changes to Assumptions for the Rod Ejection Event Paramete r          Re v. O       Rev. 2             Disc ussion Ejected Rod Worth, pcm 4 loop HFP     150            125               As can be seen by this HZP    400           500                comparison, for the 3 loop HFP     300           200                Cycle 15 design some HZP     -
450                parameters are botoded by Revision 0 and Trip Reactivity,                                     some are not. In pcm                                                  performing the 4 loop HFP     3500          3350               analysis we dete rmined HZP    1700           1375              that in fact the Cycle 3 loop HFP     2800          3075               15 ne s ults a re bo toded HZP     -
1775               by Revision 0.
However, we have Post Ejection Fq                                    s ubmitted the revised 4 loop HFP     3.63          3.3                analysis to provide HZP   14.2          10.0                consistent 3 loop HFP     5.28          5.0                documentation.      The HZP     -           10.0                radiological eval ua-tion in Rev. 0 was un-changed, since it is conse rvative .
Preejection Fq 4 loop         2.9           2.67               The preejection Fq 3 loop         2.78          3.10               has been changed becatse of a change in the physics methodology. Ir, Revision 0, the pre-ejection Fq was dete rmined from the assembly with the highest radial peak, The most conservative preejection Fq that should be used is one that is equal to the Technical Specification limit.
In Revision 2, this was     i done for 4 loop.
However, for 3 loop a val e greater than the Technical Specifi-I                                                                            cation was used. While ove rly conse rvative ,
j                                                                            this ass unption was not corrected since acceptable nesults we re achieved.
I

4
[-
Table 3:    Changes. to Assumptions for the Rod Ejection Event                                                                          ,
(Contin ted) i Paramete r                   Rev. O            Re v. 2                                                   Disc ussion Three loop                   HZP not           HZP                                                       In Revision. 0, res ults analysis                     provided          provided                                                  for a three loop HZP case were not provided since it was clearly bounded by the four loop HZP casa.
However, in Revision 2, a three loop HZP case has been provided for completeness, confirming that the four loop case is limiting.
                                                                                                                                           --   -- a

c,
:w .-
R; Table 4: Changes to Assumptions for the Steam Line Break-Paramete r         Re v. 0 -        Re v. 2 -                Discussion Trip Reactivity pcm 4 loop HFP;  4000             4000                     The HZP trip HZP 2200.            1800                      reactivity has been 3_ loop HFP. 3300'            3300                      red Lced d te to the n           >
HZP 3000-            2600-                     revised Cycle 15-
$,                                                                             shutdown margin Dopple r ' feedback Based. Upon;     Based upon          - The Revision 0
                'for nat urn ' to    all rods ott     all rods in       ,
assumption is powe r .                             with .the highest       overly conservative.
worth rod stuck -       Use of the' proper; o ut                    Doppler feedback has nesulted in a significant ned uction in consequences, more than offsetting the change in trip re activity.
State point -       No iteration     Ite ration              Using the new physics Eval uation                          pe rformed             methodology, an
                                                                          . iteration process was pe rformed . to ~ ass ure that the physics and system codes are consistent. This iteration process takes into account incomplete mixing in the inlet plen tm.
c Feedwater isola-     Valve closed     Valve ramped           The description in tion- val ve        70 seconds       closed from            Revision 0 was in-re sponse           after SI AS      65 seconds            correct and has been to 70 seconds          corrected in after SIAS              revision 2.
d
                                                                          ----.------------_----._1.-------x-----------------,---- - -

   ,e
\*
l l                                     Table 5:    Delay and Response Times RPS Fmetion-               Delay Time (sec)                       Response Tine (sec)
Re analysis             FDSA Ass unption             Ass unption Core Power          High        .35                   .35                     .20
(% f ull power) Medi un        .35                                           .20 Low         .35                                           .20 pre ss urize r . High      1.00                   1.0                     .65 Pressure (psia) Low           1.00                                           .65 Pre ss urize r      High      1.00                   1.0                     .65 Level (% span)
Low Reactor - 4 loop           .80                    .80                    .65 Coolant Flow (% of nominal 4 loop flow per loop)
                            - 3 loop        .80                                           .65                                                                              j
(% of nominal 3 loop flow per loop)
Steam Flow High               1.50                Not specified             .65
(% full flow)
Steam / Feed Flow Mismatch 1.50                   Not specified             .65 Coincident with
* 5 Level ( Flow 10 lbm/hr/% NR level )
Startup Rate                   .50                    .50                   .20                                                                              i (DPM)
                                                                                    . _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _}}