ML20081J507
| ML20081J507 | |
| Person / Time | |
|---|---|
| Site: | Byron, Braidwood  |
| Issue date: | 03/20/1995 |
| From: | Saccomando D COMMONWEALTH EDISON CO. |
| To: | Office of Nuclear Reactor Regulation |
| References | |
| NUDOCS 9503280002 | |
| Download: ML20081J507 (7) | |
Text
.
Commonw:cith Edison Brcidwood Nucitar Powsr Station
.- ~
- - #1 1 1-Rout 3 #1, Box 84 kV Braceville, Illinois 60407
\\
Telephone 815/458-2801 March' 20, 1995 Office of Nuclear Reactor Regulation U.S. Nuclehr Regulatory Commission Washington, D.C. 20555 l.
Attn:
Document Control Desk
Subject:
Additional Information Regarding Commonwealth Edison Company's Response to Questions Pertaining to the
' Increase in the Interim Plugging Criteria for Byron Unit 1 and Braidwood Unit 1 NRC Docket Numbers: 50-454 and 50-456 l<
References:
1.
D. Saccomando letter to Nuclear Regulatory Commission dated March 15, 1995, transmitting Commonwealth Edison Company Response to i
Questions Regarding the Increase in the l
Interim Plugging Criteria I
2.
D. Lynch letter to Commonwealth Edison Company dated March 14, 1995, transmitting Summary from NRC/ Comed Meeting dated February 23, 1995 3.
D. Saccomando letter to Nuclear Regulatory Commission dated February 7,
- 1995, transmitting Probability of Detection Report and WCAP-14273 In Reference 2, the Nuclear Regulatory Commission transmitted to Commonwealth Edison Company (Comed) four concerns regarding the technical bases supporting the pending license amendments, which I
involve an increase in the interim plugging criteria for steam generator tubes at Byron Unit 1 and Braidwood Unit 1. Reference 1 l
transmitted Comed's response to those concerns.
In that
)
response, Comed stated that information pertaining to concerns 2 and 3 would be provided to the Staff on March 20, 1995.
Attached are:
Attachment A Background information including a discussion of free span leak rate.
Information relating to Comed's proposed analytical model for representing steam generator tube leakage j
(WCAP-14273) (Concern 2 of Reference 2).
This shall be i
I referred to as the " Indication Restricted from Burst (IRB) Method." Results of this analysis are provided in Attachment B.
Kinla\\bybwd\\stgen\\3vitra11\\1 h
9503280002 950320 PDR ADOCK 05000454 P
PDR ig j
i
'r T
l e.
'NRC Document-Controli Desk March.20,'1995" Information relating to an alternate.model'for-representing steam generator tube leakage (Concern.3 of Reference 2). This shall be referred to as the "Mcdified IRB Leak Rate Method."
Results of this analysis are provided in Attachment B.
r Information relating to the Main Steam Line Break Analysis Results.
Attachment B-Summary'of Steam Line Break Leak Rates l Including Indications Restricted from Burst-At the NRC's request, Comed pursued to quantify the' benefit of the use of a leak rate methodology other'than the free span leak rate model. Those'results are included in Attachment B. Review of
~
this data indicates that the results from both the Indication Restricted from Burst' Method (WCAP-14273), and the Modified IRB Leak Rate Method are both within the allowable Main Steam Line Break primary to secondary leak rate values of 9.4 gpm for Braidwood 1 and 12.8 gpm for Byron 1.
Although validation of the results is not complete, we believe the final results will not significantly change and the conclusions will remain uneffected.
Futhermore, we believe that after review of this information, the Staff will agree that the results clearly demonstrates the value of a 3 volt interim plugging criteria for 3/4" steam generator tubes.
In an effort to refine.the bounding value associated with an IRB,.
EPRI is sponsoring'a laboratory test program for indications restricted from burst, Reference 1. Comed will evaluate the test results along with the SGDSM Ad Hoc-committee and revise the modified IRB leak rate methodology, if appropriate.
If you have any questions concerning this correspondence please contact this office.
S
- cerely, f
e-A enise M. Saccbmando Nuclear Licensing Administrator
)
Attachment cc:
D. Lynch, Senior Project Manager-NRR R. Assa, Braidwood Project Manager-NRR G. Dick, Byron Project Manager-NRR S. Dupont, Senior Resident Inspector-Braidwood H. Peterson, Senior Resident Inspector-Byron Office of Nuclear Safety-IDNS-Kinla\\bybwd\\stgen\\3vltre11\\2
-.e 4
er,
yeo r
w n-t
r e
9 Attachment A
Background
The proposed 3.0 volt interim plugging criteria (IPC) is based on limiting the relative tube to tube support plate (TSP) displacements such that the constraining effects of the TSP will preclude tube burst. Therefore, the probability of burst for an ODSCC indication confined within the TSP is zero. However, a finite probability exists for indications that may potentially
" leak" within the confines of the TSP at pressures equal to main steam line break (MSLB) pressure. In such a case, the crack may open up to the limits allowed by the tube to TSP gap and result in leakage greater than that assumed for the indication in the existing freespan voltage correlation. The probability of such an event is defined as equivalent to the probability of a freespan burst.
In previous correspondence, Comed has referred to these types of indications as either "overpressurized indications" or
" indications restricted from burst." From this correspondence forward, Comed and Westinghouse will use the terminology
" indications restricted from burst (IRB)," which best describes the indication and configuration.
To account for the additional IRB leakage which could potentially add to the calculated freespan leak rates, Comed has performed an evaluation using data trom the Braidwood cycle 4 inspection outage. The evaluation was performed using the Draft Generic Letter (94-XX) probability of detection (POD) value of 0.6 and the proposed EPRI voltage dependent POD submitted in Reference 3.
Leak rates were performed using the methodologies listed below, and are summarized in this letter:
- 1. Freespan leak rate (as described in the Draft Generic Letter).
- 2. Indications restricted from burst (IRB) leak rate (as described in WCAP-14273)
- 3. Modified IRB leak rate (as described below)
Free Span Leak Rate The freespan leak rate is taken from the Braidwood IPC report and is described in methods report WCAP-14277, "SLB Leak Rate and Tube Burst Probability Analysis Methods for ODSCC At TSP Intersections."
In essence, the distribution of indications in a SG is simulated many times, e.g.,
100,000, to account for eddy Ki nla\\bybwd\\stgen\\ 3vltra t 1\\3
current uncertainty, probe wear, and growth during the' cycle of pperation.
The total leak rate from the SG indications is retained from each simulation of the SG. These values are placed in ascending order after all of the simulations have been performed.
A specific value from the ordered list is statistically chosen to represent a 95% confidence bound on a 95%
portion of the population of total leak rates.
During the simulation, the leak rate from each indication is based on the simulated voltage of that indication, using the correlation of the leak rate to the bobbin amplitude documented in WCAP-14277.
Hence, the total leak rate predicted is calculated without regard for the potential for any of the tubes with indications bursting during a postulated MSLB.
Since the IPC voltage limit has been selected to result in a very low probability of burst (PoB) for any of the tubes in the SG, this has been the accepted method for estimating a conservative value of the total leak rate that could occur.
Indication Restricted from Burst (IRB) Leak Rate (WCAP-14273)
(Concern 2 of Reference 2)
Indications restricted from burst is the terminology used to refer to the condition of a crack that would have burst had the crack been assumed to be at a free span location. For ODSCC at TSPs then, the value of the probability of an IRB is the same as the calculated value of the probability of a free span burst using the burst pressure to bobbin voltage amplitude correlation.
With an increase in the IPC voltage repair limit, there is a slight increase in the probability of an IRB of one or more tubes in the SG during a postulated MSLB event. Hence, the need to account for the increase in leak rate that would occur from such tubes.
WCAP-14273 submitted as part of Reference 3, documented a method that was based on estimating a bounding leak rate from an indication as a function of the bobbin voltage amplitude. The correlation was derived using information that related leak rate to crack length, and crack length to bobbin voltage.
This method has been brought into question by the NRC because of the use of a crack length to voltage relation; however, estimation of the leak rate using this method has been retained for comparison in this report.
Because of the need to determine if IRB has occurred in order to estimate the IRB leak rate, a determination of whether or not the indication would burst if located in the free span was made. This determination was performed using the methods described in WCAP-14277.
It is noted that, while an estimate of the free span leak rate can be made without simulating the burst pressures, the IRB leak rate cannot. The Monte Carlo simulation of the leak rate is dependent on the simulation of a random value of the burst Kinla\\bytud\\stgen\\3vitrall\\4
pressure for each indication.
If the simulated burst pressure is less than or equal to the MSLB pressure of 2560 psi, the leak rate for that indication is taken as the IRB leak rate from the correlation described above.
The total. leak rates (trecspan + IRB as determined by correlation), from each of the SG simulations are accumulated as for the freespan leak rates, and the ordered value corresponding to the 95% confidence bound on a 95% portion of the population is reported.
Modified Indication Restricted from Burst (IRB) Leak Rate (Concern 3 of Reference 2)
The modified IRB leak rate is based on the premise'that the leakage from any crack that experiences IRB, will be a constant value.
For this evaluation, the constant. leak rate value was conservatively selected to be 7.0 gpm. This value is more conservative than the corresponding free span leakage value of a 15 volt indication.
The application of the 7.0 gpm limiting leak rate for the modified IRB analysis was limited to indications with bobbin voltages greater than the structural limit of 4.75 volts.
Therefore, if during the Monte Carlo simulation an indication was predicted to burst, the 7.0 gpm value was applied to indications greater than or equal to 4.75 volts and the free span leak rate applied to indications less than 4.75 volts.
Then, as in the case of the previously described free span leak rate method, the total leak rates (freespan + fixed IRB Value),
from each of the SG simulations are accumulated, and the ordered value corresponding to the 95% confidence bound on a 95% portion of the population is reported.
MSLB Leak Rate Analysis Results A number of analyses were performed to obtain voltage distributions typical of the 1.0 volt and 3.0 volt IPC repair limit as requested by Byron 1 and Braidwood 1. Analyses were performed for distributions based on a POD of 0.6 and the EPRI voltage dependent POD as submitted in Reference 3.
The analyses were performed starting with the Braidwood 1 EOC 4 voltage distribution and projections made under the different repair limit assumptions. SG D, with SG D growth rates were used for the analyses. For the POD of 0.6, SG D is the limiting SG for EOC 4 and EOC 5 leakage. Although, the most limiting SG at EOC 5 shifts to SG C with the EPRI POD. This shift occurs as the larger number of indications below 3.0 volts is the most influential for the EPRI POD while the largest voltage indications are the most influential on the leak rates for the POD of 0.6. The voltage Krnia\\bybwd\\stgen\\3vltrai1\\5
distributions for repair limits of 3.0-volts.or higher were based on the assumption that all indications above the repair limit' were plugged and no indications below the repair limit were plugged. At EOC 4 for Braidwood 1, all bobbin indications over 3.0 volts were RPC confirmed and plugged, so the plugging assumption is realistic above 3.0 volts and conservative below 3.0 volts.
The results of the various analyses are given in tabular format on Attachment B.
This table lists MSLB leak rates for each leak model and the leak rate results as affected by the POD of 0.6 and j
the EPRI POD. The Maximum EOC voltage corresponds to integrating i
the Monte Carlo distribution tail to 0.3 indications, as typically reported by Comed for projected voltage distributions.
l l
4 l
i l
l i
Kinla\\bytud\\stgen\\3vitra11\\C j
Attachment B Summary of SLB Leak Rates Including Indications Restricted from Burst f
SLB Leak Rate -gpm Case POD Free WCAP Mod.
Span Method Method -
BASED ON ACTUAL EOC-4 EOC-4 1.0 3.9 5.0 7.0 ASSUMED 3.0 VOLTIPC AT BOC-5 BOC-5 0.6 2.7 3.3 4.7 EPRI 0.1 0.1 0.1 EOC-5 0.6 6.9 7.4 8.3 EPRI 3.9 4.6 5.8 1.0 VOLT IPC AT BOC-5 1
MOC-5 0.6 3.1 N.C.
N.C.
EPRI 0.4 N.C.
N.C.
EOC-5 EPRI 3.2 41 r,
4.5 i
N.C. - Not Calculated
- -.