ML20078B342
| ML20078B342 | |
| Person / Time | |
|---|---|
| Site: | Byron  |
| Issue date: | 10/17/1994 |
| From: | Saccomando D COMMONWEALTH EDISON CO. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM), Office of Nuclear Reactor Regulation |
| References | |
| NUDOCS 9410250121 | |
| Download: ML20078B342 (12) | |
Text
__
Commonwealth Edison i
Byron Nuclear Station s
\\v ~
4450 North German Church Road
\\
Byron, Illinois 61010 October 17,1994 Office of Nuclear Reactor Rego.ation U.S. Nuclear Regulatory Commission Washington, D.C. 20555 Attention:
Document Control Desk l
t
Subject:
Additional Information Regarding Application for Amendment to Facility Operating License:
- )
Byron Station Units 1 and 2 (NPF-37/66; NRC Docket Nos. 50-454/455)
" Steam Generator Interim Plugging Criteria"
References:
See Attachment A in references 1 and 2, Comed submitted and supplemented a request to amend the Byron Station license to implement a voltage-based Interim Plugging Criteria (IPC) for Unit 1. Reference 3 further limited IPC applicability to Unit 1, Cycle 7 only, in Reference 2, Comed agreed to implement specific requirements from the draft Generic Letter (GL) on IPC. One of these requirements is a report to NRC prior to restart if the calculated probability of burst based on the projected end o' cycle (EOC) voltage distribution exceeds 1x10 2 Based on the Cycle 6 inspection results and subsequent tube burst assessment, the projected EOC-7 tube burst probability exceeds the 1x102 limit. The attached report (Attachment B) is being submitted in accordance with the proposed Technical Specification requirement to provide an assessment of the significance of this occurrence including the safety significance of the calculated conditional burst probability. Attachment B contains a summary of the Byron Unit 1 inspection results, leak and burst calculational methodology, projected EOC-7 leak and burst results, and a safety assessment of these results.
Due to tube burst considerations, full cycle operation is not justified for Byron Unit 1 Cycle 7 using the proposed methods described in the Byron IPC submittal and supplements. Therefore, Comed is planning a mid-September 1995 inspection outage (see Attachment B).
c.
.c-S Y J
%a v
9410250121 941017 DR ADOCK 0500o454 gp PDR t,
l
l Document Control Desk Page 2 October 17,1994 Although not specifically relied upon to justify operation until the next SG inspection, efforts will continue by both Byron and Braidwood Stations to pursue evaluation and approval of variable Probability of Detection and Limited Tube Support Plate Displacement considerations.
Please address any comments or questions regarding this matter to this office.
Sincerely, f
r
)/
ytDenise M. S 'ccomando Nuclear Licensing Administrator Attachments cc:
G. Dick, Byron Project Manager - NRR H. Peterson, Senior Resident inspector - Byron J. Martin, Regional Adminit,trator - Region ill Office of Nuclear Facility Safety - IDNS
ATTACHMENT A REFERENCES 1.
August 1,1994, letter from J. A. Bauer to W. T. Russell transmitting Byron Station's request for a license amendment to implement an Interim Plugging Criteria.
2.
September 7,1994, letter from J. A. Bauer to W. T. Russell transmitting a supplement to Byron Station's request for a license amendment to implement an Interim Plugging Criteria.
3.
September 17,1994, letter from D. M. Saccomando to W. T. Russell transmitting a supplement to Byron Station's request for a license amendment to implement an Interim Plugging Criteria.
4.
September 16,1994, teleconference between NRC Staff and Byron Station Staff regarding questions and issues related 1, Byron Station IPC submittal / supplements.
5.
September 22,1994, letter from D. M. Saccomando to W.T. Russell transmitting additional information regarding Byron Station IPC submittal / supplements.
6.
September 28,1994, teleconference between NRC Staff and Byron Station Staff regarding questions and issues related to Byron Station IPC submittal / supplements.
7.
August 18,1994, letter from R. Assa to D. Farrar transmitting Safety Evaluation for Use of Interim Plugging Criteria for Braidwood Unit 1.
i 8.
September 2,1994, letter from T. Simpkin to W. T. Russell transmitting WCAP-14046, Rev.1, "Braidwood Unit 1 Technical Support for Cycle 5 Steam Generator Interim Plugging Criteria" (Proprietary) and WCAP-14047, Rev.1, "Braidwood Unit 1 Technical Support for Cycle 5 Steam Generator Interim Plugging Criteria" (Non-proprietary).
9.
September 30,1994, letter from D. M. Saccomondo to W. T. Russell transmitting additional information regarding Byron Station IPC s"bmittals and supplements.
1
ATTACHMENT B Summarv of B1R06 SG Inspection Results During the Byron Unit 1 Cycle 6 refueling outage, bobbin coil eddy current inspections were performed on 100% of the inservice tubes in each steam generator. In addition, rotating pancake coil (RPC) inspections were performed on all ODSCC indications greater than 1.0 volt, a sample of indications less than 1.0 volt and at locations where a 1.0 volt flaw signal may have been masked. These inspections were performed in accordance with the Byron /Braidwood Eddy Current inspection Guidelines that were updated specifically for application of IPC. The updated guidelines include provisions to reduce the eddy current variability due to equipment and analysts and are described in References 1, 2, 5, and 9.
A total of 3076 indications were found at the tube support plates by the bobbin coil examination. All of these indications were verified to be at the outer diameter and were contained within the confines of the tube support plate. 968 of the indications were further inspected with RPC and were verified to be consistent with axial oriented ODSCC. Tables B-1, B-2, B-3 and B-4 provide a summary of the inspection results.
Three tubes with a total of 9 tube support plate intersections and 3 flow distribution baffles were removed to gain additional insight into the root cause of the tube degradation and to assess the crack morphology. Leak and burst testing will also be performed to provide additional data points for the industry leak and burst correlations. The three tubes selected contain 5 flaws of various voltage amplitudes as detected by eddy current. Table B-5 provides a summary of the tubes and indications removed for IPC support.
The required 90 day IPC report will contain more detailed discussion and description of the inspection and tube pull results.
Database for Leak and Burst Correlations The database used to perform the leak and burst assessments to support Byron Unit 1 Cycle 7 operation is described in the August 18,1994, Braidwood SER (Reference 7) with the inclusion of the Braidwood Cycle 4 tube pull results. This j
database is consistent with the database described in WCAP-14046, Rev.1, with the following differences:
B-1
Inclusion of the Braidwood Cycle 4 tube pull results, per Table 3-7 of WCAP-14046, Rev.1, into the leak and burst correlations.
inclusion of model boiler specimen 598-1 into the burst correlation database.
Inclusion of model boiler specimens 598-1 and 598-3 into the leak rate correlation database.
Plant S tube R28C41 data point assigned a leakage value of 24961/hr.
Leak and Burst Assessment Methodoloav and Results Main Steam Line Break (MSLB) leak rate and burst probabilities were calculated using methods described in References 1, 5, 7, and 9. The MSLB leak rate and burst analyses were based on full Monte Carlo methodologies that account for parameter uncertainty per the methods described for leak rate analyses in WCAP-14046, Rev.1, and the August 18,1994, Braidwood SER. Monte Carlo methods were used to develop an end of cycle (EOC) voltage distribution and Monte Carlo techniques applied to the EOC voltage distribution to obtain the MSLB leak rates and burst probabilities. A more detailed description of the analysis methodology will be provided in the 90 day report.
The beginning of cycle (BOC) voltage distributions were initially developed using a Probability of Detection (POD) of 0.6 for all voltages. Subsequent analyses were performed using other POD methods to assess the sensitivity of leak and burst results to POD. Other POD methods evaluated include:
POD based on EPRI data analyst qualification results.
POD of 0.6 at 0.0 volts and increasing linearly to 1.0 at 3.0 volts.
(POPCD POD)
POD = 1.0 for all voltages The POD methods used in these analyses are graphically shown in Figure B-1. A more detailed description of the POD methods and the sensitivity of tube burst results to POD will be provided in the 90 day summary report.
For the tube burst probability analyses, the Monte Carlo analyses calculate, separately, the burst probability of a single tube rupture and multiple tube ruptures, i.e., the probability of one tube, two tubes, three or more tubes rupturing. The Monte Carlo process involves sampling of allindications in the Steam Generator (SG) to obtain the number of burst pressure samples with burst pressures less than the MSLB value of 2560 psid. The sampling process is repeated N times. Without B-2
adjustments for confidence level, the single and multiple tube burst probabilities are the number of occurrences of one, two, etc., tubes with burst pressures less than the MSLB value, as summed over all SG samples, divided by N. The burst probabilities reported are 95% confidence values which include adjustments for the number (N) of SG samples.
Voltage growth rate distributions were obtained for each of the four SGs. A deterministic leak and burst probability analysis was performed to determine the most limiting SG. The analysis determined that SG C was the most limiting SG, therefore, leak and burst calculations contained in this report are based on SG C voltage and growth distributions.
Table B-6 provides a summary of the leak and burst analysis results. A more detailed result description will be provided in the 90 day report.
Safety Assessment The calculated EOC-6 and EOC-7 MSLB leak rate values are significantly less than the site allowable leak rate limit of 12.8 gpm for all calculation methods. Therefore from an accident leakage perspective only, full cycle operation (est.1.3 EFPY) is justified and the.dfsite dose is limited to a small fraction of 10CFR100 limits.
The EOC-6 burst probability based on the actual voltage distribution is 1.9x10,
2 Although this value exceeds the proposed Technical Specification (TS) burst limit 2
of 1x10, it is less than the NUREG-0844 burst probability limit of 2.5x10 2, Likewise, the BOC-7 burst probability exceeds the TS burst limit, but is less than the NUREG-0844 limit (see Table B-6).
The calculated conditional probability of burst for EOC-7 ranges from 1.22x10 2 to 3.29x10 2, depending on POD assumptions, where the latter value is derived from a POD =0.6. These values exceed the proposed TS limit for tube burst probability.
Therefore, from a tube burst probability perspective, full cycle operation is not justified using the methods described in the Byron IPC submittal and supplements.
A reduced period between SG inspections is warranted. Based on the risk assessments and other considerations discussed below, Byron is currently planning an inspection outage prior to full cycle in the mid-September 1995 timeframe.
The results of probabilistic and deterministic approaches were used to evaluate the operating period between SG inspections. The deterministic approach assumes a 1.2 volt indication (1.0 volt indication with 20% eddy current uncertainty) grows linearly over the next cycle at the largest growth rate found (9.86 v/1.279 EFPY).
This results in the 4.54 volt structural limit being exceeded within 5.2 effective full power months. Attachment D of Reference 1 describes a risk assessment that B-3
I I
i i
evaluated the impact of IPC for a full fuel cycle against the insights from the Byron
- ndividual Plant Examination (IPE). This assessment is summarized below.
As stated in the current IPE, the total core damage frequency for Byron is 4
estimated to be 3.09x10 per reactor year, with a total contribution from containment bypass sequences of 3.72x10 8 per reactor year. Operation with IPC with a variable POD is expected to increase the MSLB with containment bypass sequence frequency contribution by a factor of only 10% An upper bound increase of a factor of two is derived when the fixed POD of 0.6 is employed in the calculation. Neither is significant from a risk perspective. This assessment indicates that, for a complete eighteen (18) month fuel cycle, the application of the 1.0 vo!! IPC will not significantly intraase the core damage frequency.
An additional risk assessment, consistent with the August 18,1994, Braidwood SER, was also performed to assess the risk during the next Byron Unit 1 fuel cycle.
This approach uses the EOC-7 Monte Carlo total MSLB burst probability for combined single and multiple tube ruptures with a POD of 0.6 (see Table B-6).
MSLB and Feedwater Line Break (FWLB) frequencies of 1.8x10"/ reactor-year are assumed in this assessment. These frequencies are based on the occurrence of 2 FWLB events in 1370 reactor years of Westinghouse pressurized water reactor operation. An estimated probability of 1x10'3 was used for failure to mitigate the combined effects of such events to prevent core damage. This is consistent with NUREG-0844 and draft NUREG-1477. The risk assessment is presented as i
follows:
(1.8x10' + 1.8x10' )/ year x (1x10 ) x (1.51x10 2) = 5.44x10 / year (BOC) 3 8
(1.8x10~3 + 1.8x10-3)/ year x (1x10~3) x (3.29x10 2) = 1.18x10
/ year (EOC) 4 Based on these values, the estimated frequency of core damage due to induced rupture of combined single and multiple tubes at Byron Unit 1, during Cycle 7, 4
would vary from 5.44x10-8 / year at BOC to 1.18x10 / year at EOC Many probabilistic risk assessments for PWRs estimate total frequency of containment 4
bypass releases on the order of 10 / reactor-year. The estimated incremental risk for Byron Unit 1 full cycle operation would not make Byron an outlier in terms of risk from the accident sequences discussed above. In addition, preliminary results from the Braidwood/ Byron Model D-4 limited tube support plate displacement analysis indicate that the current freespan burst probability would be significantly reduced or even eliminated due to application of the constraining effects of the l
support plate.
The two risk assessments discussed above justify full cycle operation for Byron Unit 1 and any increase in inspection frequency would result in more conservative risk values. However, Comed has opted not to rely on risk assessment alone to B-4
I 2
justify full cycle operation and, consequently, is planning an inspection outage prior to full cycle.
The middle of September 1995 would be the optimum time for Byron to consider a shutdown for SG inspection for the following reasons:
Byron Unit 2 will be entering a refueling outage in mid-February of 1995. A dual unit outage is not desired due to limited resources which will cause delays on both outages and the potential for increased shutdown risk.
Braidwood Unit 1 will be entering an SG inspection outage at the end of February of 1995. The Braidwood inspection results and experience will be evaluated for applicability to Byron. Therefore, it is desired to perform the i
Byron inspection following the Braidwood inspection outage.
From a utility system planning perspective, a Byron Unit 1 shutdown during summer peak electrical demand is not desired by Comed.
The estimated burst probability at the mid-September 1995 shutdown is expected to be less than the NUREG-0844 limit, assuming a FOD =0.6.
Analysis is in progress to obtain this value. The result will be provided at a later date.
i Therefore, Byron is currenUy planning a Unit 1 shutdown in mid-September of i
1995 to perform SG eddy current inspections. Parallel efforts are also underway to assess the applicability of a variable POD and limited tube support plate displacement. The results of the Braidwood Unit 1 inspection outage in February f
of 1995 will be evaluated for applicability and impact on the planned Byron outage.
As additional information becomes available and is assessed, the timing of the Byron Unit 1 inspection outage will be re-evaluated.
1 B-5
i Table B-1:
ODSCC Indications Found By Bobbin Coil Voltage SG A SG B SG C SG D TOTAL Range
< = 1.0 v 645 705 672 395 2417 1.0-2.70 v 150 208 224 63 645 2.70-4.54 v 4
1 2
2 9
> 4.54 v 3
1 1
0 5
Total 802 915 899 460 3076 Table B-2:
ODSCC Indications Confirmed By RPC Voltage SG A SG B SG C SG D TOTAL Range
< = 1.0 v 70 of 124 51 of 85 46 of 76 15 of 24 182 of 309 1.0-2.70 v 126 167 185 43 521 2.70-4.54 v 4
1 2
2 9
> 4.54 v 3
1 1
0 5
- Tubes 123 165 187 45 520 Repaired *
- Note: Tubes repaired due to application of IPC only. Some tubes contained multiple indications at various locations.
l B-6
i ei a
Table B-3:
l Summary of Indications > 2.70 Volts l
3.08 3.01 3.64 3.20 i
3.10 4.56 3.80 4.09 3.37 10.95 i
3.98 5.92 7.10 i
7.64 l
l 4
Table B-4:
I Cycle 6 Growth Rate Summary
}
SGs j
Ave. Voltage Growth 0.41 0.30 0.37 0.28 0.35
]
for Cycle (1.279 EFPY)
Ave. Voltage Growth 0.32 0.23 0.29 0.22 0.27 per EFPY Max, Voltage Growth 6.71 3.65 9.86 3.74 9.86 for Cycle (1.270 EFPY)
B-7
Table B-5:
Summary of Tubes Removed For Laboratory Analysis (IPC)
SG Row-Col FDB 3H TSP SH TSP 7H TSP A
3 -107 NDD 5.96 v NDD 0.25 v C
20-7 NDD 10.95 v NDD NDD C
20-102 NDD 2.58 v 1.25 v NDD Note:
FDB - Flow Baffle Plate TSP - Tube Support Plate NDD - No Detectable Degradation l
Table B-6:
Byron Unit 1 SG C MSLB Monte Carlo Analysis Results Analysis MSLB
- of Total Single Two Tube Three Method Leak Monte Burst Tube Burst Tube Rate Carlo Prob.
Burst Prob.
Burst (gpm)
Samples Prob.
Prob.
EOC-6 Measured 2.3 1000K 1.9x10 2 1.9x10 2 4.74x10~8 3.00x10~8 Voltages BOC-7 POD = 0.6 1.8 450K 1.51 x 10-2 1.51 x10 2 9.26x10 5 6.66x 10^6 POD = 1.0 0.03 450K 3.10x 10~4 3.10x 10~4 6.66x 10 8 6.66x 10~6 EOC-7 POD = 0.6 5.1 950K 3.29x 10-2 3.15x10 1.49x 10'3 3.15x10~6 2
EPRIPOD 3.1 900K 2.29x 10 2 2.28x10 2 1.41 x 10-4 3.33x10~6 POPCD 3.5 300K 2.53x 10-2 2.49x10-2 4.92 x 10
9.99x10 6 POD POD = 1.0 1.7 350K 1.22x10 2 1.21 x10 1.13x 10~4 8.56x10~6 2
B-8
e FIGURE B-1:
Comparison of Probability of Detection Models 100.0 % o a
=
=
=
=
=
=
=
=
=
=
=
=
=
E 90.0%
sx y
+ EPRI
-e-POD =1.0 o 80.0 %
-*- POD =0.6 I
-*- POPCD n
W D
{ 70.0%
60.0 %
=
=
=
=
=
=
1 s
50.0 %
+-
t I-
-+
t - --
0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.50 2.75 3.00 3.25 3.50 3.75 4.00 Voltage
- -