3F0904-03, Response to NRC Request for Additional Information Regarding Special Report 03-01
| ML042710359 | |
| Person / Time | |
|---|---|
| Site: | Crystal River  |
| Issue date: | 09/09/2004 |
| From: | Annacone M Progress Energy Florida |
| To: | Document Control Desk, Office of Nuclear Reactor Regulation |
| References | |
| 3F0904-03, TAC MC1853 | |
| Download: ML042710359 (11) | |
Text
aC Progress Energy Crystal River Nuclear Plant Docket No. 50-302 Operating Ucense No. DPR-72 Ref: 10 CFR 50.36 September 9, 2004 3F0904-03 U.S. Nuclear Regulatory Commission Attn: Document Control Desk Washington, DC 20555-0001
Subject:
Crystal River Unit 3 - Response to NRC Request for Additional Information Regarding Special Report 03-01 (TAC No. MC1853)
References:
- 1. PEF to NRC letter, 3F1003-07, dated October 31, 2003, Crystal River Unit 3 -
Special Report 03-01: Once Through Steam Generator (OTSG) Notifications Required Prior to MODE 4
- 2. PEF to NRC letter, 3F0804-04, dated August 10, 2004, Crystal River Unit 3 -
Response to NRC Request for Additional Information Regarding Special Report 03-01 (TAC No. MC1853)
Dear Sir:
Florida Power Corporation, doing business as Progress Energy Florida, Inc. (PEF) submitted Special Report 03-01 containing the Once-Through Steam Generator (OTSG) inspection notifications required prior to MODE 4 in Reference 1. During a telephone conference with PEF on July 8, 2004, the NRC discussed a Request for Additional Information (RAI) regarding the subject special report. RAI questions 1 and 2 and the corresponding PEF responses were provided in Reference 2. The response to RAI questions 3 and 4 are provided in Attachments A and B to this letter.
This letter establishes no new regulatory commitments.
If you have any questions regarding this submittal, please contact Mr. Sid Powell, Supervisor, Licensing and Regulatory Programs at (352) 563-4883.
Sincerely, Michael J. Annacone Manager Engineering MJA/lvc Attachments:
A. Response to RAI Question 3 B. Response to RAI Question 4 xc: NRR Project Manager Regional Administrator, Region II Senior Resident Inspector Progress Energy Florida, Inc.
Crystal River Nuclear Plant 15760 W. Powerline Street Crystal River, FL 34428
FLORIDA POWER CORPORATION CRYSTAL RIVER UNIT 3 DOCKET NUMBER 50-302 / LICENSE NUMBER DPR-72 RESPONSE TO NRC REQUEST FOR ADDITIONAL INFORMATION REGARDING SPECIAL REPORT 03-01 ATTACHMENT A Response to RAI Question 3
U.S. Nuclear Regulatory Commission Attachment A 3F0904-03 Page 1 of 5 RAI Ouestion 3
- The licensee monitors the growth of intergranular attack (IGA) indications in the first span of the B Steam Generator (SG) in accordance with the plant's technical specifications. The licensee has stated that these IGA indications have limited growth rates and in previous Mode 4 Special Reports (1999 & 2001) has not stated that any indications required plugging. In the 2003 Mode 4 Special Report, the licensee identified that 5 tubes were plugged for greater than 10% growth, 2 tubes were plugged for greater than 40% through-wall indications, and 1 tube was plugged which did not meet the regression technique (new indication).
- The graph Crystal River Unit 3 (CR3) SG B Distribution of Freespan IGA Growth from the 2003 Mode 4 Special Report relates number of indications to growth bins.
Provide a correlation of percent through-wall growth for each of the growth bins.
- Provide a discussion of the purpose, assumptions, inputs, methodology and results of the regression technique identified in the Mode 4 Report.
- Provide the number, location, percent through wall penetration and cause of any new indications identified during the 2003 outage. For the 5 tubes plugged as a result of exhibiting growth greater than 10%, provide the locations, magnitude of growth experienced in these indications broken down by operating cycle over the past operating cycles and the apparent cause of growth.
- For the 2 tubes plugged as a result of having penetration greater than 40% through-wall, provide the locations, percent through-wall penetration values, the magnitude of growth experienced in these indications broken down by operating cycle over the past operating cycles and the apparent cause of growth.
Response
- Information from the 1999 and 2001 Once-Through Steam Generator Inspection Special Reports During Refueling Outage 11 (1 JR) in 1999, nine (9) tubes were plugged in Once-Through Steam Generator B (OTSG-B) due to an assigned percent through-wall (TW) of > 40%. Eight (8) tubes which had percent TW penetrations of < 40% were conservatively plugged in OTSG-B due to an estimated increase in growth of greater than 10% since the last inspection. Twenty (20) first span IGA tubes were plugged due to a lack of historical data for the indications and four (4) tubes were conservatively plugged due to the presence of multiple IGA indications.
During IIR, CR3 examined apparent growth of the first span pit-like IGA indications in OTSG-B. The regression technique indicated percent TW for 1999 was compared to the regression technique indicated percent TW from the 1997 inspection. All regression data was obtained using a high frequency bobbin probe. The growth of the first span pit-like IGA in OTSG-B continued to be essentially zero with the majority of indications exhibiting a - 2% to + 6%
change in indicated percent TW.
U.S. Nuclear Regulatory Commission Attachment A 3F0904-03 Page 2 of 5 The slightly positive increase in indicated percent TW is within the accuracy of the measurement technique used to establish the percent TW depth of the indications.
During Refueling Outage 12 (12R) in 2001, no tubes were plugged due to multiple volumetric IGA indications. Additional tubes (29) were located in the OTSG-B containing IGA indications.
Of these 29 tubes, eight tubes were plugged (i.e., seven tubes had previous sizing data and one tube could not be regressed).
- Correlation of percent TW growth for each growth bin The "growth bins" indicated on the graph in the Attachment to PEF letter 3F1003-07, Page 2 of 5, "CR3 SG B Distribution of Freespan IGA Growth," are equivalent to the percent growth rate of indications based on the 1997 baseline measurement.
- Purpose, assumptions, inputs, methodology and results of the regression technique identified in the MODE 4 Report The regression analysis technique described in the MODE 4 Report is a methodology to measure the TW dimension of volumetric indications. The application of this technique was approved by the NRC in License Amendment 172 to the CR3 Improved Technical Specifications (ITS). This sizing technique is limited to tubes with IGA indications in OTSG-B in the first span Lower Tubesheet (LTSF) to 01 Support Plate (OlS) and allows tubes with volumetric IGA indications
< 40% TW to remain in service.
The regression concept is based on the premise that multiple types of measurements taken on an indication will contain more information than a single measurement. A multiple regression technique mathematically calculates the relationship between multiple measurements from eddy current testing and a single measurement of IGA estimated depth. The regression technique for CR3 is based on the data and destructive analysis from tubes pulled from CR3 in 1992 and 1994 using a specialized high frequency bobbin coil probe.
The variables used are the bobbin differential peak-to-peak voltage and phase measurements from three different frequencies for a given signal. The formula for the regression analysis tool was developed by Framatome-ANP for CR3 and verified on a group of indications from a CR3 pulled tube with confirmed destructive examination data. Specific guidelines were developed and are used every inspection to maintain consistency for the analysis of volumetric indications.
Also, a single analyst per inspection evaluates these IGA indications and obtains the measurement in order to reduce the analysis uncertainty in the measurement accuracy from inspection to inspection.
The regression measurements from the current outage were compared to those measurements from the 1997 inspection to determine a growth rate. Indications that exhibit a growth rate in excess of + 10% TW between the 1997 inspection and the current inspection, or any indication with a TW dimension > 40%, are removed from service. The negative shift in growth rate is attributed to inherent technique (personnel and equipment) uncertainties and margin of error. In 1999, CR3 changed to a low loss cable for bobbin eddy current examination (ECT). This shifted the relative responses between the three frequencies. While the change produced a more dramatic effect on individual frequencies, it produced a very systematic reduction in average depth of approximately 5% TW. When the offset from the cable change was corrected, there was no significant change in the average growth.
U.S. Nuclear Regulatory Commission Attachment A 3F0904-03 Page 3 of 5
- Number, location, percent TW penetration and cause of any new indication identified during the 2003 outage Eight (8) additional indications (shown in shaded font in the table below) were identified in tube 78-95. This tube had three previously identified IGA indications that were recorded during the 2001 inspection. The regression technique was applied for sizing those indications. The cause of the new indications is attributed to the probability of detection as well as an increased sensitivity to recording identifiable indications. This tube was plugged during the 2003 Refueling Outage.
2001 Inspection 2003 Inspection Location Location from from LTS Phase % LTS Phase %
Row Tube (Inches) (Degrees) Volts TW (Inches) (Degrees) Volts TW 78 95 N/A N/A N/A N/A 11.26 -147 - -0.15 23 N/A N/A N/A N/A 9.68 99 0.21 38
_ 8.10 145 0.22 31 8.05 145 0.25 33
= N/A N/A N/A N/A .7.60 129 0.14 26
= = 15.92 155 0.14 15 15.90 157 0.11 15
= = N/A N/A N/A N/A 15.41 140 - 0.33 40 N/A N/A N/A N/A 13.90 134 0.13 16
= _ N/A N/A N/A N/A -12.61 139 0.22 36 10.34 144 0.26 28 10.42 135 0.13 32
= =___ N/A N/A N/A N/A 13.63 50 0.12 21
= = N/A N/A N/A N/A 17.42 37 0.30 N/A
U.S. Nuclear Regulatory Commission Attachment A 3F0904-03 Page 4 of 5
- Location and magnitude of growth experienced in these indications broken down by operating cycle over the past operating cycles and apparent cause of growth for the 5 tubes plugged due to > 10% growth and 2 tubes plugged for > 40% TW The following table identifies seven tubes that were removed from service following the 2003 inspection and the reason (shown in shaded font) they were removed (> 10% growth rate or indication > 40% TW). The average growth rate for each indication is less than 10%. The apparent TW change is primarily attributed to measurement errors associated with the detection and analysis techniques.
1997 Inspection 1999 Inspection 2001 Inspect on 2003 InspectIon Locatlo Location Location Locatlo n from from from n from I LTS Phase % LTS Phase % LTS Phase % LTS Phase %
Row Tube (Inches) (Degrees) Volts TWI (Inches) (Degrees) Volts TW (Inches) (Degrees) Volts TW (Inches) (Degrees) Volts TW
>10% Growth Rate 39 44 7.64 144 0.13 9 7.76 163 0.14 17 7.69' 154 0.12 16 7.65 148 0.14 19 8.26 161 0.23 23 8.35 176 0.36 24 8.33 175 0.36 27 8.28 170 0.43 23 6.08 167 0.38 28 6.28 162 0.21 34 6.09 158 0.26 31 6.12 155 0.28 32
. 16.77 138 0.14 22 16.85 153 0.1 31 16.89 166 0.20 31 16.97 158 0.18 34
. 14.53 121 0.07 29 14.66 123 0.13 35 14.65 127 0.15 34 14.65 130 0.18 32 13.05 137 0.21 27 13.26 151 0.12 33 13.14 149 0.16 28 13.16 150 0.19 24
>10% Growth Rate 43 11 1 49 11.82 14.351 8.591 .67 169 155 0.06 0.50 0.10 32 32 15 11.98 14.41 8.68 174 160 124 0.32 0.38 0.06 32 36 23 11.89 14.37 8.64 174 159 124.
0.31 0.31 0.08.
29 33 241 4
11.87 14.46 8.621 168 147 118 0.34 0.28 0.10 33 32 27
> 40% Throu h-Wall 78 95 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A 11.26 147 0.15 23 T_ 7 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A 9.68 99 0.21 38 N/A N/A N/A N/A N/A N/A N/A N/A 8.10 145 0.22 31 8.05 145 0.25 33 l N/A N/A N/A N/A N/A NMA N/A N/A N/A N/A N/A N/A 7.60 129 0.14 26 N/A N/A N/A N/A N/A N/A N/A N/A 15.92 155 0.14 15 15.90 157 0.11 15 T l N/A N/A N/A N/A N// A N/A N/Al N/A NA N/A I 15.41 140 0.33 40
U.S. Nuclear Regulatory Commission Attachment A 3F0904-03 Page 5 of 5 1997 Inspection 1t999 Inspection l2001 Inspection 2003inspection Locatlo Location Location Locatlo n from from from n from LTS Phase % LTS Phase % LTS Phase % LTS Phase %
Row Tube (Inches) (Degrees) Volts TW (Inches) (Degrees) Volts TW (inches) (Degrees) Volts TW (Inches) (Degrees) Volts TW N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A 13.90 134 0.13 16
_ N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A 12.61 139 0.22 36 N/A N/A N/A N/A N/A N/A N/A N/A 10.34 144 0.26 28 10.42 135 0.13 32 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A 13.63 50 0.12 21 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A NIA N/A 17.42 37 0.30 N/A
>10% Growth Rate 84 96 11.63 161 0.24 28 11.69 150 0.17 34 11.74 154 0.14 30 11.83 151 0.21 28 12.1i 162 0.12 13 12.27 79 4 0.05 -15 12.30 137 - 0.07 22 12:374 143 4 0.09 24 8.97 111 0.24 34 9.03 118 0.22 35 9.05 122 0.22 34 9.11 1 114 1 0.23 35
>10% Growth Rate 95 45 12.58 J 142 0.27 31 12.47 142 0.24 35 12.54 149 0.27 32 12.56 142 J 0.27 33 16.60 4 113 0.10 25 16.58 133 4 0.06 24 16.69 97 0.08 24 16.58 4 81 4 0.07 4 22 11.45 65 0.16 30 11.3 88 4 0.12 34 11.46 76 0.11 28 11.45 65 4 0.154 27 9.36 167 0.23 14 9.34 173 1 0.2 24 9.43 162 0.21 *23 9.431 163 10.22 25
> 40% Throu h-Wall 96 42 11.91 I 143 0.42 34 12 151 I 0.29 37 12.13 153 0.35 37 12.11 f 153 0.32 40 13.00 4 151 0.24 26 13.11 159 4 0.28 34 13.17 156 0.25 30 13.13 163 4 0.32 4 34
__ 9.28 1 151 0.32 30 9.43 133 1 0.16 30 9.52 154 . 0.27 30 9.54 1 158 T 0.29 1 30
>10% Growth Rate 97 49 I 11.83 I 138 0.23 31 11.951 132 0.15 1301 12.02 135 0.17 33 12.14 I 140 f 0.21 I 30 1
15.041'159 0.261 22 15.15 153 0.12 30 5.25' 160 0. 31 15.34 158 0.25
FLORIDA POWER CORPORATION CRYSTAL RIVER UNIT 3 DOCKET NUMBER 50-302 / LICENSE NUMBER DPR-72 RESPONSE TO NRC REQUEST FOR ADDITIONAL INFORMATION REGARDING SPECIAL REPORT 03-01 ATTACHMENT B Response to RAI Question 4
U.S. Nuclear Regulatory Commission Attachment B 3F0904-03 Page 1 of 3 RAI Ouestion 4 Considering that historically it has been assumed that the intergranular attack (IGA) indications in the first span do not grow, discuss the implications of the 2003 findings as they relate to maintaining tube integrity for the new indications or the possible growth of existing indications.
Response
The inspection in 2003 was the fourth consecutive inspection of the tubes with IGA indications.
Tubes that were plugged due to > 10% growth rate are plugged conservatively to assure that the 40% through-wall (TW) criteria would not be jeopardized for the next operating cycle. Of those tubes plugged for indications > 40% TW, one had an indication change from 37% TW to 40%
TW over three cycles. The other tube did not have historical data to perform the regression analysis, but did have three indications with little or no growth.
The Operational Assessment (OA) following the 2003 inspection accounts for the tubes in service with IGA. The OA was performed using deterministic structural performance criteria.
Worst case degraded tubes were projected using a bounding approach with all input at worst case 95<< percentile values and a multi cycle Monte Carlo approach using degradation initiation functions and probability of detection curves.
The approach to a bounding deterministic operational assessment was;
- A worst case Beginning of Cycle (BOC) state of degradation is selected
- An upper 95th percentile growth rate is applied
- The projected worst case degraded tube must exhibit a minimum burst pressure of 3AP and the required axial load capability A Non-Destructive Examination (NDE) repair limit of 40% TW is used for a plug on sizing approach, which is converted to a best estimate actual depth. If necessary, maximum depths are converted to structural average depths using a bounding length/depth profile. An upper 95th percentile sizing error is added to develop a worst case BOC state of degradation. For wear and volumetric degradation, the worst case BOC structural average depth was taken as 0.90 times the maximum depth. When required, an upper bound 95th percentile structural length is considered.
For wear, a linearly tapered profile leads to a bounding length of 0.4 inches. This length also bounds the axial extent of volumetric degradation.
All measured growth rates contain two measurement errors, the initial measurement error and the final measurement error. Typically, measurement errors are normally distributed. For a large data set, the average growth rate will be reliable, as measurement errors will sum to zero. For independent initial and final measurements, the standard deviation of the growth rates will be at least 1.414 times the standard deviation of the measurement error. A standard deviation less than this value suggests measurement errors which are not independent, such as in side-by-side comparisons for growth evaluations by the same sizing analyst. In either case, Monte Carlo simulation of the growth measurement process can provide estimates of actual growth rates compared to measured growth rates compounded by measurement error.
Volumetric degradation at CR3 gives no appearance of growth. Global average growth rates are negative or very close to zero. Apparent positive and negative growth rates are essentially due to
U.S. Nuclear Regulatory Commission Attachment B 3F0904-03 Page 2 of 3 measurement errors. Chance occurrences of under-sizing and over-sizing produce the apparent growth rates. Actual physical growth is essentially undetectable. Very small changes are possible. As a conservative measure, a nominal worst case growth rate of 3% TW per cycle is used in the bounding deterministic operational assessments for both wear depth and volumetric degradation depth in the first span of Once-Through Steam Generator B (OTSG B).
The End of Cycle (EOC) projection of worst case degradation is obtained by simply adding the worst case growth projection to the worst case BOC depth. The results were compared to the EOC Allowable Structural Limits. A conservative bounding analysis demonstrates satisfaction of deterministic structural performance criteria for full cycle operation of CR3 for the next cycle of operation for wear and volumetric degradation. The projected cycle length is 2.0 Effective Full Power Year (EFPY). This is the cycle length used for degradation growth projections in the bounding type operational assessments. Increased margins and allowable cycle lengths can be demonstrated using Monte Carlo techniques described in the EPRI Tube Integrity Assessment Guidelines.
A detailed operational assessment was performed for axial Outside Diameter Stress Corrosion Cracking (ODSCC)/IGA. Consideration of all uncertainties is included.
Bounding deterministic analyses with all input selected at 95 th percentile bounding values are highly conservative. More sophisticated projections of worst case degradation are possible. The processes of initiation, growth, inspection, detection and sizing can be simulated to model the development of specific types of degradation. This technique is termed the multi-cycle Monte Carlo approach. Pertinent variables are selected from distributions defined by measured data and the full range of degradation, including extreme values, is considered.
Summary of Bounding Deterministic Operational Assessments Mechanism Worst 95th Percentile 95th Percentile Projected EOC Margin Case Growth Rate Or Bounding Worst Allowable (%TW)
BOC (%TW/EFPY) Length Case Depth Depth (inches) Depth (%TW)
(%TW) (%TW)
Wear and First Span 50.6 3.0 0.40 56.6 69.8 13.2 IGA The eddy current Probability of Detection (POD) curve and the crack growth rate distribution are the main controlling parameters in the Monte Carlo simulation of corrosion degradation of steam generator tubing. Other pertinent input parameters are the distribution of tubing tensile properties, the distribution of EOC crack lengths and NDE sizing parameters and uncertainties.
U.S. Nuclear Regulatory Commission Attachment B 3F0904-03 Page 3 of 3 The output of the Monte Carlo simulation calculations is a complete description of the state of degradation including detected and undetected instances of degradation. Crack lengths, depths and shapes are tracked and input into burst strength and leak rate calculations.
In terms of structural integrity evaluations, several figures of merit are available from multi-cycle Monte Carlo simulation models. With a deterministic performance criterion, the worst case degraded tube should maintain a burst pressure in excess of 3AP, which is 4050 psi. Just as each simulation of an operating cycle produces a new projection for the number of detected indications, each simulation produces a worst case degraded tube having the worst case burst pressure. Hence, a distribution of worst case degraded tube burst pressures is the end result of many simulations of an operating cycle for a steam generator. The probability of zero bursts at 3AP is 96.6% on a per generator basis.
Volumetric degradation does not contribute to projected Steam Line Break (SLB) leakage at the end of the next cycle of operation. This is evident by the fact that structurally integrity, without wall penetration, will be maintained at 4050 psi compared to the SLB pressure of 2575 psi.
In summary, analyses using both state of the art detailed evaluations and bounding type operational assessments of all operative degradation mechanisms at Crystal River Unit 3 show that deterministic structural integrity and leakage integrity requirements will be met for the full cycle of operation for a cycle length of at least 2.0 EFPY.