ML20236R279
| ML20236R279 | |
| Person / Time | |
|---|---|
| Site: | Byron, Braidwood  |
| Issue date: | 07/15/1998 |
| From: | Krich R COMMONWEALTH EDISON CO. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| NUDOCS 9807210336 | |
| Download: ML20236R279 (12) | |
Text
m
,Commonwecith 12dison Company 14G3 Opus Place a
1] owners Drove. 11,(0515 5701 H
July 15,1998 United States Nuclear Regulatory Commission ATTN: Document Control Desk Washington D. C. 20555 - 0001
Subject:
Byron Nuclear Power Station, Units 1 and 2 Facility Operating Licenses NPF-37 and NPF-66 NRC Docket Numbers 50-454 and 50-455 Braidwood Nuclear Power Station, Units 1 and 2 Facility Operating Licenses NPF-72 and NPF-77 i
NRC Docket Numbers: 50-456 and 50-457 Response to Request for AdditionalInformation Regarding Determination of Dynamic Component Setpoint Uncertainties i
References:
- 1. R. A. Assa (USNRC) to D. L. Farrar (Comed) letter dated May 22, i
1996," Transmitting Request for Additional Information Regardmg Dynnmic Testing ofInstrument Channels."
l
- 2. J. B. Hosmer (Comed) to US Nuclear Regulatory Commission letter dated November 12,1996," Transmitting Response to Request for Additional Information Regarding Dynamic Testing ofInstrument Channels."
- 3. G. F. Dick (USNRC) to I. Johnson (Comed) letter dated May 5,1997,
" Transmitting Request for Additional Infonnation Regarding Dynamic I
Testing ofInstrument Channels."
- 4. J. B. Hosmer to US Nuclear Regulatory Commission letter dated July 1, U
.,,vC 3
1997," Transmitting Response to Request for AdditionalInformation go[
Regarding Dynamic Testing ofInstrument Channels."
1 y
9807210336 990715 PDR ADOCK 05000454 I
P PDR G Dave \\LIC-98-062MEsAC doc
U.S. NRC 2
July 15,1998 The purpose of this letter is to provide the results of a Dynamic Compensation Sensitivity study, initiated by Commonwealth Edison (Comed), to resolve issues raised by the NRC.
The study evaluates the impact of uncertainties in protection system dynamic compensation terms on the safety analyses for Byron and Braidwood stations. A
)
discussion of the specific dynamic compensation terms investigated and the specific transients examined are contained in Attachment 1. The results of the study are contained in Attachment 2. Based on the results of this study, Comed has concluded that setting the
{
dynamic compensation terms at their nominal values in safety analyses, and with the other i.
applied uncertainties and conservatism in the methods and models, provides for a l-sufficiently conservative analysis.
The issues in the initial request for information (i.e., Reference 1) involved the treatment of protection system setpoint uncertainties by the Micro Electronic Surveillance and Calibration (MESAC) system at Braidwood Station. As stated in Reference 1, this request was made on behalf of a NRC staff member even though the NRC had concluded: "...
the health and safety of the public was not affected by the use of the MESAC system and that Comed is in compliance with its licensing bases and technical specifications (TS)."
The Comed response to that request is contained in Reference 2.
In response.to additional issues raised by the same member of the NRC staff regarding the treatment of setpoint uncertainties in the Byron /Braidwood Updated Final Safety Analysis Report (UFSAR) Chapter 15 Safety Analyses, the NRC issued the Reference 3 request for information. Reference 4 provided Comed's response to those questions. In that response Comed concluded that the Westinghouse setpoint methodology determines a "sufficiently L
conservative" protection setpoint to ensure that Safety Analysis Limits are not exceeded.
To further support this conclusion, Comed initiated the attached sensitivity analysis to illustrate the concept of"sufficiently conservative."
Please address any comments or questions regarding this matter to Mr. David J.
Chrzanowski at (630) 663-7205.
1 Sincerely, l
i
- R. M. Krich :
Vice President - Regulatory Services i
' Attachments: ' 1 - Discussion of Events for the Dynamic Compensation Sensitivity Study Results of the Dynamic Compensation Sensitivity Study G Dave \\lJC 98-062.MCSAC. doc
l l
U.S. NRC 3
July 15,1998 g
l cc:
' Regional Administrator-Rill Senior Resident Inspector-Braidwood Senior Resident Inspector-Byron I
l
]
t i
E l
l l
Discussion of Events for the Dynamic Compensation Sensitivity Study The selection of events included in the sensitivity study is a two step process. The first step involves elimination of events which do not credit dynamically compensated
~
protective functions in the safety analyses. The second step then selects events which are l
limiting and represent the different dynamically compensated protective functions l
assumed in safety analyses.
Dynamically Compensated Protective Functions I
As discussed in the J. B. Hosmer to US Nuclear Regulatory Commission letter dated July 1,1997, the Byron and Braidwood safety related protection channels that use dynamic components are:
-i Overtemperature Delta-T l
Overpower Delta-T e
l e Pressurizer Pressure Low Steam Line Pressure Low Low-Low Tavg e
Events that do not model Dynamically Compensated Protective Functions The following events do not credit dynamically compensated protective functions in the i
safety analyses.
UFSAR Eection Title l
15.1.2-Feedwater System Malfunctions Causing FW Flow Increase 15.1.3 Excessive Increase in Secondary Steam Flow 15.1.4 Inadvertent Opening of a Steam Generator Relief or Safety Valve 15.2.6 Loss of Non-Emergency AC Power to the Plant Auxiliaries l
15.2.7 Loss of Normal Feedwater Flow 15.2.8 Feedwater System Pipe Break (with and without power) 15.3.1 Partial Loss of Forced Reactor Coolant Flow I
15.3.2 Complete Loss of Forced Reactor Coolant Flow
(
l5.3.3 Reactor Coolant Pump Shaft Seizure (Locked Rotor) l 15.3.4 Reactor Coolant Pump Shaft Break 15.4.1 Uncontrolled RCCA Bank Withdrawal (Low Power) 15.4.3 Dropped RCCA or RCCA Bank 15.4.7 Fuel Assembly in Improper Location 15.4.8 Spectrum of RCCA Ejection Accidents l
G:\\ Dave \\LIC-98-063.MESAC. doc
__-_______--_--_-__________-_---_-_---_-__a
Events that model Dynamically Compensated Protective Functions but Excluded frbm Study A review of events which model dynamically compensated protective function was performed to select events for the sensitivity study. The reasons for excluding certain events from the study are provided below.
In the analysis for the Westinghouse Original Steam Generators (OSG), no credit is taken for dynamically compensated protective functions. In the analysis for the Backcock &
i Wilcox, International (BWI) Replacement Steam Generators (RSG), the Overpower Delta-T trip is the primary trip. The following potential trip, that does not contain dynamic components and result from this event, is available: Power Range Ifigh Neutron Flux Trip.
The RSG analysis for this event would be sensitive to changes in the dynamic compensation terms. However, a feedwater temperature reduction transient is very similar to a hot full power steamline break transient with respect to the impact on the j
primary and subsequently on the Overtemperature Delta-T protection. Since steamline break is the more limiting event, it is not necessary to include the feed water temperature reduction event in the sensitivity study.
- 2. Steamline Break - Hot Zero Power (UFSAR Section 15.1.5)
Protection for the llot Zero Power Steamline Break event could come from a variety of protective functions. However, protection il only in the form of safety injection, steamline isolation and feedline isolation. The analysis is initiated at hot zero power and j
it is assumed that the rods are in the core. As such, no credit is taken for reactor trip because no reactor trip is needed. Therefore, variation in dynamic response of the reactor trip function has no impact on the hot zero power steamline break event and this event does not need to be included in a sensitivity study.
- 3. Turbine Trin(UFSAR Section 15.2.3)
The Turbine Trip analysis documented in the UFSAR in Section 15.2.3 bounds the Loss of External Load (Section 15.2.2), the Inadvertent Main Steam Isolation Valve (MSIV)
Closure (Section 15.. ') and the Loss of Condenser Vacuum (Section 15.2.5) events.
The analysis for the Turbine Trip event takes credit for the Overtemperature Delta-T protective function which is dynamically compensated. This event is analyzed to demonstrate that the primary and secondary system pressure limits and the DNB design bases are not violated. For the OSG, cases are analyzed both with and without pressure i
GADave\\LIC-98-063.MEsAC. doc Y_____________-___-_____
l i
l l
control systems operable and with minimum and maximum (Beginning of Life (BOL) and End of Life (EOL)) reactivity feedback assumptions. In the cases analyzed for the OSG, the limiting case assumes minimum reactivity feedback without pressure control.
Since this case trips on Iligh Pressurizer Pressure, which is not dynamically compensated, variation in dynamic response of reactor trip function has no impact on the turbine trip event for the OSG and this event does not need to be included in a sensitivity study.
- 4. fhemical and Volume Control System (CVCS) Malftmetion - Decreases in Boron Concentration (UFSAR Section 15.4.6)
The CVCS Malfunction analysis does not explicitly model the setpoint of any dynamic compensation. The analysis determines the time from the initiation of the event until a loss of shctdown margin. The mode I full power operation analysis then uses this time along with the time to reach an Overtemperature Delta-T trip from the Rod Withdrawal at Power analysis to determine the time from a signal until a loss of shutdown margin. The analyses for the other modes do not rely on dynamically compensated protective functions. The mode 1 analysis is not the limiting case for this event and therefore does not need to be included in a sensitivity study.
The Inadvertent ECCS Operation at power event is analyzed for two different purposes.
The first purpose is to demonstrate that the Departure from Nucleate Boiling (DNB) design basis is met. This case is not limiting because the Departure from Nucleate j
Boiling Ratio (DNBR) increases throughout the transient with the minimum DNBR
)
(MDNBR) occurring at the initiation of the event. The other purpose is to demonstrate j
adequate operator action time prior to pressurizer filling. This case models reactor trip at the initiation of the event, i.e. assumed to occur on the " spurious" signal that actuates safety injection. An early reactor trip is the conservative assumption for the pressurizer filling spurious ECCS operation case. As such, the Inadvertent ECCS event does not need to be included in a sensitivity study.
The large break LOCA event models the Pressurizer Low Pressure trip as the primary trip. Ilowever, because of requirements in Appendix K to 10 CFR 50, control rod motion j
is not assumed for this event. Therefore, variation in dynamic response of reactor trip l-protective function has no impact on this event and this event does not need to be included in a sensitivity study.
The protective function assumed ' or the small break LOCA event is from Pressurizer f
l Low Pressure. This analysis applies conservative assumptions to provide a sufficiently conservative analysis. One of the conservative assumptions applied is the reactor trip delay time. The delay from the time the reactor reaches the low pressurizer pressure trip signal setpoint to the time when all the control rods are fully inserted assumed for this event is 6 seconds. This 6.0 seconds delay time includes allowances for signal processing l.
~ delay (2 seconds) and control rod drop time (2.7 seconds). The remaining 1.3 seconds is L.
for added conservatism. The delay in reactor trip due to variation in dynamic response is l4 estimated to be less than 1.2 seconds, which is smaller than the added conservatism in reactor trip delay. Therefore, this event does not need to be included in a sensitivity study.
l Events Included in Sensitivity Study i
- From the previous discussions, Comed has determined that four events should be evaluated for sensitivity to deviations in dynamic response.
- 1. Steamline Break - Hot Full Power (UFSAR Section 15.1.5) l For the hot full power steamline break event, protection comes from the Low Steam Line l
Pressure and the Overpower Delta-T trips. The assumed break size is varied and the first L
protection signal changes from the Overpower Delta-T trip to the Low Steam Line Pressure trip. The case where the protection switches from one function to the next is l
iypically the limiting case (i.e., the case giving the lowest DNBR and highest power).
Since both protective functions are dynamically compensated, variations in dynamic response can impact the results of this event. Therefore, this event is included in the
. sensitivity study. The results are presented in Attachment 2.
' 2. Turbine Trin (UFSAR Section 15.2.3)
In the cases analyzed for the RSG for the turbine trip event, only the minimum reactivity feedback with pressure control case trips on Overtemperature Delta-T. Since the margin to secondary pressure limit is small, this event is included in the sensitivity study. The results are presented in Attachment 2.
- 3. Rod Cluster Contro: Assembly (RCCA) Withdrawal at Power (UFSAR Section 15.4.2)
- For the RCCA withdrawal at power event, protection comes from the Power Range liigh Neutron Flux trip nnd the Overtemperature Delta-T trip. The assumed reactivity insertion rate is varied and the first protection signal changes from the high flux trip to the Ovenemperature Delta-T trip. The case where the protection switches from one function to the next is typically the limiting case (i.e., the ense giving the lowest DNBR). Even if I
OADave\\LIC-98 063.MEsAC40c
i l
l i
'the absolute limiting case trips on high flux, delaying Overtemperature Delta-T will shift l
the case with the minimum DNBR to a difTerent case with a slightly lower minimum DNBR. Therefore, this event is included in the sensitivity study. The results are presented in Attachment 2.
- 4. Steam Generator Tube Ruoture (UFSAR Section 15.6.3)
The Steam Generator Tube Rupture (SGTR) event is analyzed for two purposes. The first case demonstrates Margin to Overfill (MTO) while the second case demonstrates meeting radiological consequence requirements in 10 CFR 100. An early trip is conservative for the MTO case.
In the most recent Comed SGTR methodology approved by the NRC, two cases are analyzed for the MTO case. The first case models protective function from Ovedemperature Delta-T and the second case models protective function from
.1 Pressurizer Low Pressure. The limiting case is the case where protection is from Overtemperature Delta-T. The limiting steam generator type is RSG. Results from the
- limiting case are presented in Attachment 2.
For the offsite dose case, the protective function is Pressurizer Low Pressure. The limiting steam generator type is OSG. Results from the limiting case are presented in.
b 0:\\ Dave'LIC 98 063.MEsAC. doc
s l
l' Results of the Dynamic Compensation Sensitivity Study As discussed in Attachment 1, the four events to be included in the study are hot full power steamline break, turbine trip, Rod Control Cluster Assembly (RCCA) withdrawal at power, and _ Steam Generator Tube Rupture (SGTR). For the first three events and the l
. SGTR offsite dose case, the leads are decreased while the lags are increr. sed to delay l
reactor trip. The first cases modeled a 5% tolerance on both the leads and the lags, reducing the lead / lag ratio by 9.5%. The second cases modeled a 10% tolerance on both the leads and the lags, reducing the lead / lag ratio by 18.2%. For the SGTR margin to overfill (MTO) case, an early reactor trip is conservative; therefore, the leads are j
it creased while the lags are decreased As the results show, the SGTR event is not sensitive to variations in dynamic responses of the protective functions; therefore, only the 10% tolerance case is presented.
In the UFS AR analyses for these events, calculations or evaluations were performed to determine the limiting steam generator type. For the RCCA withdrawal at power and turbine trip events, additional cases were analyzed to determine the limiting reactivity
' feedback case. In this study, additional cases or evaluation were included to confirm that similar sensitivities would result with the different steam generators or with the different
- feedback assumption. The results of the limiting cases are presented in this attachment.
Resuhs
- 1. Steamline Break - Hot Full Power (UFS AR Section 15.1.5)
The primary protective functions for the hot full power steamline break event are the Overpower Delta-T and Low Steam Pressure functions, which are both dynamically compensated. The sensitivity for this event was done in three parts: variation in Overpower Delta-T function only, variation in Low Steam Pressure function only, and variation in both Overpower Delta-T (OPAT) and Low Steam Pressure functions. Table I shows results for the RSG.
Table 1 - Hot Full Power Steamline Break Results l
Case MDNBR Limit Peak Heat Flux Change (%)
- 1. UFSARanalysis 1.724 1.4 1.2278 2.- OPAT,5%
1.713 1.4 1.2329 0.5 l
- 3. OPAT,10%
1.695 1.4' l.2408 1.3 l
- 4. Iow Steam Pressure,5%
1.697 1.4 1.2398 1.2 1
- 5. Low Steam Pressure,10%
1.666 1.4 1.2529 2.5
- 6. OPAT and low Steam Pressure,5%
1.684 1.4 1.2457 1.8
)
- 7. OPAT and Low Steam Pressure,10%
1.640 1.4 1.2650 3.7 The impact from variations in dynamic response of the Overpower Delta-T protective -
function (cases 2 and 3) came from delay in reactor trip only. The remaining cases (4 - 7) resulted in increased impact because the variations resulted in both a delayed l
l OnDave\\UC-98 064.MESAC. doc L_-_-_________________-__-
1
l reactor trip (via Overpower Delta-T) and a different limiting break size (due to Low l
Steam Pressure).
The MDNBR results show that the impact from variations in dynamic response of
_ protective functions is small and that sufficient margin to limit exists to offset the impact.
The peak heat flux results are used to calculate the maximum heat rate, which is used for l
fuel centerline melt evaluation. This type of calculation / evaluation is performed for each reload cycle. A review of recent reload calculations showed that the margin to the maximum heat rate and the conservative method used to calculate the maximum heat rate can more than offset the impact from variations in dynamic responses of protective functions.
- 2. Turbine Trio (UFSAR Section 15.2.3)
The primary protective function for the turbine trip event is the Overtemperature Delta-T function, which is dynamically compensated. Table 2 shows results for the RSG.
. Table 2 -Turbine Trip Results Case MDNBR Limit Peak Secondary Limit (psia)
Pressure (psia)
- 1. UFSARanalysis 2.095 1.4 1314.8 1318.5
- 2. OTAT,5%
2.092 1.4 1314.65 1318.5
- 3. OTAT,10%
2.085 1.4 1315.41 1318.5 The peak primary pressure for this event came from the without pressure control case j
which did not rely on the Ove temperature Delta-T fimetion. Both the MDNBR results and peak secondary pressure resu'ts show that the impact from variations in dynamic response of protective function is insignificant for this event.
It should be noted that the margin to limit for the peak secondary pressure is small.
However, the analysis applies conservative assumptions to provide a sufficiently conservative analysis. One of the conservative assumptions applied is the tolerance for the main steam safety valve. The analysis assumes 4% while the Tech Spec requirements is 3%. If 3% is assumed in the analysis, the peak secondary pressure is 1303 psia. This result confirms that the conservative assumptions used in the analysis provides for a sufficiently conservative analysis.
- 3. RCCA Withdrawal at Power (UFSAR Section 15.4.2)
The primary protective function for the turbine trip event is the Overtemperature Delta-T L
function, which is dynamically compensated. Table 3 shows results for the OSG, l
i L
o:Omve\\LJc-98464.MESAc. doc
' Table 3 RCCA Withdrawal at Power Results Case MDNBR Limit
- 1. UFSAR analysis 1.754 1.4
- 2. OTAT,5%
1.740 1.4
- 3. OTAT,10%
1.737-1.4 The MDNBR results show that the impact from variations in dynamic response of
. protective functions is small and that sufficient margin to limit exists to offset the impact.
- 4. Steam Generator Tube Ruoture (UFSAR Section 15.6.3)
As discussed in Attachment 1, two cases are analyzed for the SGTR MTO case. One case models the protective function from Overtemperature Delta-T and a second case models the protective function from Pressurizer Low Pressure. Both protective functions are dynamically compensated and both cases are included in the study. Table 4 shows MTO results for the RSG.
Table 4 - SGTR MTO Results Case MTO (ft')
- 1. UFSARanalysis(OTAT) 62
- 2. OTAT,10%
60
- 3. UFSAR analysis (Pressurizer Low Pressure) 226
- 4. Pressurizer Low Pressure,10%
226 For the offsite dose case, the primary protective function is Pressurizer Low Pressure.
Table 5 shows results for the limiting OSG case.
Table 5 -SGTR Offsite Dose Results Case Braidwood, EAB Thyroid Limit Dose (rem)
(rem)
- 1. UFSARanalysis 24.93 30
- 2. Pressurizer Low Pr.:ssure,10%
24.%
30 Results from both the MTO and offsite dose cases show that the impact from variations in dynamic response of protective function is insignificant for this event.
Conclusion L
lt is Westinghouse's policy to set the values for most parameters, including the dynamic compensation terms, to their nominal values when performing non-LOCA safety analyses. Selected key parameters which are determined to be important to the analysis results are identified and the values used in the analyses for these parameters are set in a onDave\\1JC-98464.MEsAC. doc
i conservative fashion to demonstrate that the applicable safety criteria are met. This method yields a conservative licensing basis.
Obviously, applying tolerances on the dynamic compensation terms in the direction that will delay reactor trip will result in more severe analysis results. The purpose of this sensitivity study was to quantify the magnitude of the changes on the analysis results due to tolerances on dynamic compensation terms. - Several protective functions are dynamically compensated and several non-LOCA events credit these protective functions. The results of this study show the impact to be insignificant for the turbine trip and SGTR events. For the hot full power steamline break and the RCCA Withdrawal at Power events, the impact is small and sufficient margin to limit exists to offset the impact.
Based on the results of this study, Comed has concluded that setting the dynamic compensation terms at their nominal values in safety analyses, and with the other applied uncertainties and conservatism in the methods and models, provides a sufficiently conservative analysis.
l t-0:\\Deve\\1JC-98-064.MESAC. doc c -
_ - - - - - - - - - - - _ - - - - - - - - - _ - - - - - - -