RC-12-0173, Technical Basis for Westinghouse Embedded Flaw Repair for V. C. Summer Unit 1 Reactor Vessel Head Penetration Nozzles
| ML12324A168 | |
| Person / Time | |
|---|---|
| Site: | Summer  |
| Issue date: | 11/14/2012 |
| From: | Ching Ng Westinghouse |
| To: | Office of Nuclear Reactor Regulation |
| References | |
| CR-12-04775, RC-12-0173 LTR-PAFM-12-137-NP, Rev 2 | |
| Download: ML12324A168 (29) | |
Text
Westinghouse Non-Proprietary Class 3 Document Control Desk CR-1 2-04775 RC-12-0173 LTR-PAFM-12-137-NP Revision 2 Technical Basis for Westinghouse Embedded Flaw Repair for V. C. Summer Unit 1 Reactor Vessel Head Penetration Nozzles November 2012 Author:
C. K. Ng*
Piping Analysis and Fracture Mechanics Verifier:
A. Udyawar*
Piping Analysis and Fracture Mechanics Approved:
S. A. Swamy*
Manager, Piping Analysis and Fracture Mechanics
- Electronically approved records are authenticated in the Electronic Document Management System
© 2012 Westinghouse Electric Company LLC All Rights Reserved OWestinghouse
Westinghouse Non-Proprietary Class 3 Document Control Desk CR-2-04775 RC-1 2-0173 Record of Revisions Revision Date Description of Changes 0
October 2012 Original Issue I November 2012 Incorporate NRC comments and information from the latest NDE data sheet Included Technical Basis on J-Groove Weld Repair in Section 3.0 Note:
Changes made in the latest revision are indicated by a single line in the right hand margin as shown here.
Page 2 of 29
Document Control Desk Westinghouse Non-Proprietary Class 3 CR-12-04775 RC-1 2-0173 1
INTRODUCTION As a part of the inspection and contingence repair efforts associated with the reactor vessel closure head inspection program at V. C. Summer Unit 1, engineering evaluations were performed to support plant specific use of the Westinghouse embedded flaw repair process to repair unacceptable flaws detected in the head penetration nozzles during the Fall 2012 outage.
The embedded flaw repair process involves depositing a weld material, which is Primary Water Stress Corrosion Cracking (PWSCC) resistant, over the detected flaw on the outside surface of the penetration nozzle of interest as well as over the wetted surface of the attachment J-groove weld. As a result, the surface flaw becomes a sub-surface flaw and is no longer exposed to the primary water environment. The methodology used is based on extensive analytical work completed by the Westinghouse Owners Group, currently the Pressurized Water Reactor Owners Group (PWROG), and a large collection of test data obtained under the sponsorship of Westinghouse, Babcock & Wilcox (B&W) and the former Combustion Engineering Owners groups (CEOG), as well as the Electric Power Research Institute (EPRI). The technical basis of the embedded flaw repair process is documented in WCAP-1 5987-P Revision 2-P-A [1] and has been reviewed and accepted by the Nuclear Regulatory Commission (NRC) in the United States. In the NRC Safety Evaluation Report that was incorporated in WCAP-1 5987-P Revision 2-P-A, the NRC staff concluded that, subject to the specified conditions and limitations, the embedded flaw repair process described in WCAP-15987-P provides an acceptable level of quality and safety. The staff also concluded that WCAP-15987-P is acceptable for referencing in licensing applications.
In this report, the technical basis and the flaw evaluation results to support the use of the Westinghouse embedded flaw repair process for head penetration nozzle number 19, 31, 37 and 52 with unacceptable outside surface flaws in the vicinity of the J-groove weld toes are provided. Engineering evaluations were performed to determine the maximum acceptable initial flaw sizes that can be left behind in a repaired penetration nozzle which would satisfy the ASME Section XI requirements [2]. The purpose of this report is to provide plant-specific technical basis for the use of the embedded flaw repair process and to confirm that V. C. Summer Unit 1 meets the criteria for application of the embedded flaw repair process stated in Appendix C of WCAP-1 5987-P [1].
2 TECHNICAL BASIS FOR APPLICATION OF EMBEDDED FLAW REPAIR PROCESS TO HEAD PENETRATION NOZZLES This section provides a discussion on the technical basis for the use of embedded flaw repair process for head penetration nozzle number 19, 31, 37 and 52 with unacceptable outside surface flaws. Such a repair involves depositing several layers of Alloy 52/52M weld material over the flaw on the outside surface of the penetration nozzle of interest below the J-groove weld as well as the wetted surface of the attachment J-groove weld. Since the Alloy 52/52M repair weld material is PWSCC resistant, the detected surface flaw in the head penetration nozzle of interest is then shielded from the primary water environment and is no longer susceptible to primary water stress corrosion cracking.
Page 3 of 29
Westinghouse Non-Proprietary Class 3 Document Control Desk CR-2-04775 RC-1 2-0173 For the repair of the unacceptable outside surface flaws in head penetration nozzle number 19, 31, 37 and 52, at least three layers of Alloy 52/52M material are deposited (360' full circumference) covering the entire wetted surface of the attachment J-groove weld. The repair weld extends at least 0.5 inch past the interface between the J-groove weld buttering and stainless steel cladding as well as covering the entire outside surface of the head penetration nozzle with at least two layers of Alloy 52/52M material. A schematic of the repair configuration for the repaired outside surface flaw is illustrated in Figure 2-1.
Flaw evaluations were performed based on the flaw sizes and shapes remaining in the repaired head penetration nozzles of interest to demonstrate that the left behind flaws are acceptable for continued operation. The as-found flaw parameters for penetration nozzle number 19, 31, 37 and 52 are shown below in Table 2-1. Since all the indications located on the outside surface of the penetration nozzles in the vicinity of the attachment J-groove weld toes are skewed with respect to the axis of the penetration nozzles, both axial and circumferential flaws are assumed.
Table 2-1 C. Summer Unit 1 Head As-Found Flaw Parameters in V.
Penetration Nozzles Flaw Flaw Flaw Flaw Indications Orientation Length (in)
Depth Location (in)
Penetration No. 19 Circumferential 1.36 Outside Surface/Downhill (Indications #1 & #2)
Axial 0.72 Side Penetration No. 31 Circumferential 0.16 Outside Surface/Downhill (Indication #1)
Axial 0.52 0.122 Side Penetration No. 31 Circumferential 0.16 Outside Surface/Downhill 0.177 Side (Indication #2)
Axial 0.36 Penetration No. 31 Circumferential 0.26 Outside Surface/Downhill (Indication #3)
Axial 0.61 0.256 Side Penetration No. 37 Circumferential 0.31 Outside Surface/Downhill 0.249 Side (Indication #1)
Axial 0.76 Penetration No. 37 Circumferential 0.10 Outside Surface/Downhill 0.214 Side (indication #2)
Axial 0.56 Penetration No. 37 Circumferential 0.16 Outside Surface/Downhill (Indication #3)
Axial 0.52 0.294 Side Penetration No. 52 Circumferential 0.47 Outside Surface/Downhill 0.279 Side (Indication #2)
Axial 0.32 Penetration No. 52 Circumferential 0.21 Outside Surface/Downhill (Indication #3)
Axial 0.12 0.132 Side Page 4 of 29
Document Control Desk Westinghouse Non-Proprietary Class 3 CR-12-04775 RC-1 2-0173 Based on the Ultrasonic Testing (UT) and Penetrant Testing (PT) results at the regions of interest for penetration nozzle number 19, 31, 37 and 52, there are no surface connected indications in the J-groove weld and the detected indications are solely in the base metal of the nozzles. Each of the detected UT indications starts in the nozzle below the toe of the weld.
Some of the measured indication lengths extend slightly above the toe of the weld, but the measurement technique overestimates the lengths due to the large beam spread inherent with tip diffraction probes. The thinnest portion of the weld is the ground contour that blends the weld to the nozzle, so any propagation from the nozzle base metal into the weld metal would be expected to start at that point. In order to determine if the indications grew into the weld metal at this contoured section of the weld, a PT was performed on the J-groove weld and adjacent nozzle. Since none of the PT indications continued into the J-groove weld, it was concluded that the indications only involve the base metal of the nozzle. Nevertheless, technical basis for the embedded flaw repair is provided for the penetration nozzles of interest in this section as well as for the J-groove welds in Section 3.0.
2.1 EVALUATION PROCEDURE AND ACCEPTANCE CRITERIA Rapid, non-ductile failure is possible for ferritic materials at low temperatures, but is not applicable to the nickel-base alloy head penetration nozzle material such as Alloy 600. Nickel-base alloy material is a high toughness material and plastic collapse would be the dominant mode of failure. Therefore the evaluation procedures and acceptance criteria for indications in austenitic piping contained in paragraph IWB-3640 of ASME Section Xl Code [2] are applicable for evaluation of flaws in the head penetration nozzles. The evaluation procedure used is consistent with those in Appendix C of WCAP-1 5987-P [1] and summarized below:
2.1.1 Acceptance Criteria for Axial Flaws For axial flaws, the allowable flaw depth for a given flaw length can be determined from the following expression:
Gh= Gm [i t
where 2=[I+* 1.61 )*,2/]1 a2 1 4-*t/
Page 5 of 29
Document Control Desk Westinghouse Non-Proprietary Class 3 CR-1 2-04775 RC-12-0173 and Su +S
=
Flow stress 2 y (Average of Ultimate and Yield Strengths) 2 o
=
PRm/t
=
Total Flaw Length a
=
Flaw Depth Rm
=
Mean Radius of Penetration Nozzle t
=
Wall Thickness of Penetration Nozzle P
=
Internal Pressure SFm =
Safety Factor for membrane stress:
2.7 for Level A Service Loading 2.4 for Level B Service Loading 1.8 for Level C Service Loading 1.3 for Level D Service Loading The limits of applicability of this equation are a/t < 0.75 and f < f, where ta1ow = 1.58(Rmt)°'
5 [(f IOh) 2 _- 1] 0.5 This limit is chosen such that surface flaws would remain below the critical size based on the plastic collapse condition if they should grow through the wall.
2.1.1 Acceptance Criteria for Circumferential Flaws For circumferential flaws, the following relationship between the applied loads and flaw depth at incipient collapse given by equations in ASME Section XI Article C-5000 [2] is used:
b =
1-2sinp - t sinel 1
=(
Trr - a Tr am where:
= Bending stress at incipient plastic collapse 0
= One-half of the final flaw angle 13 = Angle to neutral axis of penetration nozzle a/t = Flaw depth to wall thickness ratio
= Flow stress 2 Sy (Average of Ultimate and Yield Strengths) am = Applied membrane stress Page 6 of 29
Document Control Desk Westinghouse Non-Proprietary Class 3 CR-12-04775 RC-12-0173 The allowable bending stress, Sc, is as follows, which is used to calculate the maximum allowable end-of-evaluation period flaw sizes and the limit of applicability of this equation is a/t <
0.75.
Sc -C.b
[m ]
where S,
=
Allowable bending stress for penetration nozzle Gm
=
Applied membrane stress SIm
=
Safety factor for membrane stress
=
2.7, 2.4, 1.8 and 1.3 for Service Level A, B, C, and D respectively SFb
=
Safety factor for bending stress
=
2.3, 2.0, 1.6, and 1.4 for Service Level A, B3, C, and D respectively 2.2 Methodology The flaw evaluation considered that the embedded flaw repair process is used to seal the unacceptable flaws from further exposure to the primary water environment. The evaluation began with the determination of the maximum allowable end-of-evaluation period flaw sizes based on the acceptance criteria described in Section 2.1 for the repaired penetration nozzles.
With the embedded flaw repair process, the only mechanism for future sub-critical crack growth is fatigue. The maximum initial embedded flaw size that can remain in a repaired penetration nozzle using the embedded flaw repair process can then be determined by subtracting the predicted fatigue crack growth for future plant operation from the maximum allowable end-of-evaluation period flaw size.
This maximum initial allowable embedded flaw size is then compared with the left-behind flaw in the repaired head penetration nozzle of interest to demonstrate acceptability.
The following provides a discussion of the loading conditions, geometry, thermal transient stress and fatigue crack growth analysis used in the development of the plant specific technical basis for the embedded flaw repair process.
2.2.1 Geometry and Source of Data There are many penetration nozzles in the reactor vessel upper head.
The outermost penetration nozzles (46.0' intersection angle) were selected for thermal transient and residual stress analysis because the stresses in the outermost penetration nozzles are more limiting and can be used to conservatively represent those at penetration nozzle number 19, 31 and 37 and 52.
The dimensions of all the V. C. Summer Unit 1 penetration nozzles are identical, with a 4.00 inch nominal outside diameter and a nominal wall thickness of 0.625 inch [3]. The distributions of residual, thermal transient and pressure stresses in the upper head penetration nozzle were obtained from the detailed three-dimensional plant specific elastic-plastic finite element analyses [4]. The through-wall stress distributions from the finite element analyses were used Page 7 of 29
Document Control Desk Westinghouse Non-Proprietary Class :3 CR-12-04775 RC-12-0173 to determine the fatigue crack growth. The resulting crack growth is then used to determine the maximum allowable initial flaw sizes for the left-behind flaws in the repaired penetration nozzles of interest.
2.2.2 Maximum Allowable End-of-Evaluation Period Flaw Size Determination The requirement for evaluating a flaw using the rules of ASME Section XI is that the loading for normal/upset conditions as well as emergency/faulted conditions be considered.
This is necessary because, as discussed in Section 2.1, different safety margins are used for the normal/upset and emergency/faulted conditions. A lower safety factor is used to reflect a lower probability of occurrence for the emergency/faulted conditions.
Plastic collapse is the governing mode of failure for the head penetration nozzles because the high fracture toughness of the nickel base alloy (Alloy 600) material would prevent brittle fracture from occurring.
Therefore, it is not necessary to consider the effects of secondary stresses resulting from thermal transient stresses and residual stresses. The governing loading for determining the maximum allowable end-of-evaluation period flaw sizes is therefore those due to internal pressure and other applicable external mechanical loads for the normal, upset, emergency and faulted conditions.
2.2.3 Thermal Transients Used in Fatigue Crack Growth Analysis For the fatigue crack growth prediction, the effects of secondary stresses resulting from thermal transient and residual stresses must also be considered. The thermal transients that occur in the upper reactor vessel head region are relatively mild.
The normal and upset thermal transients considered in the fatigue crack growth calculation are shown in Table 2-2 [5].
Page 8 of 29
Document Control Desk CR-1 2-04775 RC-12-0173 Westinghouse Non-Proprietary Class 3 Table 2-2 Reactor Coolant System Transients for V. C. Summer Unit 1 Design Transients Design Cycles Normal Conditions Heat Up/Cooldown 200 Plant Loading/Unloading 18300 Step Load Increase/Decrease 2000 Large Step Load Decrease with Steam Dump 200 Turbine Roll Test 80 Feedwater Heaters Out of Service 40 Steady State Fluctuation (Initial) 150000 Steady State Fluctuation (Random) 3000000 Upset Conditions Loss of Load 200 Loss of Flow 80 Loss of Power 40 Reactor Trip From Full Power 400 Inadvertent Auxiliary Spray 10 Excessive Feedwater Flow 30 Operating Basis Earthquake 400 2.2.4 Crack Tip Stress Intensity Factor One of the key elements in a crack growth analysis is the crack driving force or crack tip stress intensity factor, K1. This is based on the equations available in the public literature. It should be noted that the flaws in the repaired penetration nozzles are conservatively assumed to be surface flaws even though the flaws are embedded after the repair.
For a part-through wall surface flaw, the stress profile is approximated by a fourth order polynomial as follows:
0(x) = Ao + Ajx + A2x2 + A3x3 + A4x4 where:
x a
Distance into the wall from the free surface Stress perpendicular to the plane of the crack Coefficients of the 4 th order polynomial fit, i = 0, 1, 2, 3, 4 For a surface flaw in the penetration nozzle, the stress intensity factor expression from API-579
[6] is used. The stress intensity factor K, (c) can be calculated anywhere along the crack front, Page 9 of 29
Document Control Desk Westinghouse Non-Proprietary Class 3 CR-12-04775 RC-12-0173 where 4 is the elliptical angle of a point on the crack front being evaluated.
The following expression is used in calculating K, (4).
KI=Lna ]0.5 4 Q,
- .G,(a/c, alt, tlR,,) Ajaj j=0 The magnification factors Go, G1, G2, G3 and G 4 can be found in [6]. The parameter "a" is the crack depth, "c" is the half crack length, "t" is the wall thickness, "R" is the'mean radius, "4" is the parametric angle of the elliptical crack, and "Q" is the shape factor.
2.2.5 Fatigue Crack Growth Analysis The applied loads used in the fatigue crack growth analysis include pressure, thermal transients and residual stresses. The normal and upset thermal transients considered in the fatigue crack growth analysis are shown in Table 2-2. The transient cycles are distributed evenly over the entire plant design life. The crack tip stress intensity factor range, AK, which controls fatigue crack growth, depends on the geometry of the crack, its surrounding structure and the range of applied stresses in the region of the crack. Once AK is calculated, the fatigue crack growth due to a particular stress cycle can be determined using a crack growth rate reference curve applicable to the head penetration nozzle material.
The fatigue crack growth rate (CGR) reference curve used in the fatigue crack growth analysis for the Alloy 600 material in air environment is based on that in NUREG/CR-6721 [7] and is shown below.
da 4.CSRAK 4 1 dN =SA C = 4.835 x 10-14 + 1.622 x 10 - 16T - 1.490 x 10- 18T2 + 4.355 x 10- 2 1T3 SR = [1 - 0.82R]- 22 where:
T
=
Temperature of the Transient (°C)
AK =
Stress Intensity Factor Range (MPafM_)
R
=
Stress Ratio (Kmin/Kmax) da
=
Fatigue crack growth rate (meters/cycle) dN Page 10 of 29
Document Control Desk Westinghouse Non-Proprietary Class 3 CR-12-04775 RC-12-0173 Once the incremental crack growth corresponding to a specific transient for a given time period is calculated, it is added to the previous crack size, and the analysis continues to the next time period and/or thermal transient assuming the flaw shape remains constant. The procedure is repeated in this manner until all the significant design thermal transients and cycles known to occur in a given period of operation have been analyzed. For conservatism, R=1 is used in the fatigue crack growth analysis.
2.3 Flaw Evaluation Results The maximum allowable end-of-evaluation period axial and circumferential flaw depths for the V.
C. Summer Unit 1 penetration nozzles of interest are provided for various flaw aspect ratios (flaw depth/flaw length) in Table 2-3. The maximum allowable initial axial and circumferential flaw sizes accounting for fatigue crack growth of 40 years after the repair are shown in Figures 2-2 and 2-3 respectively. The maximum allowable initial flaw sizes are obtained by subtracting the fatigue crack growth for 40 years of service life after the repair from the maximum allowable end-of-evaluation period flaw sizes. Figure 2-4 shows the fatigue crack growth for hypothetical axial and circumferential flaws with initial flaw depth and aspect ratios (flaw depth/flaw length) that bound those for the left-behind flaws in repaired penetration nozzle number 19, 31, 37 and
- 52. The fatigue crack growth curves shown in Figure 2-4 would bound the fatigue crack growth curves for each of the indications in the repaired penetration nozzles of interest. The fatigue crack growth results shown in Figure 2-4 shows that it would take more than 40 years to reach the maximum allowable end-of-evaluation period flaws sizes shown in Table 2-3.
This is consistent with the results shown in Figures 2-2 and 2-3 where the left-behind flaw sizes in the repaired penetration nozzles of interest are below the maximum allowable initial flaw size curves.
As shown in Figures 2-2 and 2-3, the respective maximum allowable initial axial and circumferential flaw sizes are larger than the left-behind flaws in the repaired penetration nozzle number 19, 31, 37 and 52. Therefore, all the repaired flaws are acceptable for continued operation for at least 40 years after the repair. It should be noted in Figures 2-2 and 2-3, the aspect ratios (flaw depth/flaw length) for indications in the penetration nozzles are set to a maximum of 0.5 in accordance with the ASME Section Xl Code.
Table 2-3 Maximum Allowable End-of-Evaluation Period Flaw Sizes (Percentage of Nominal Wall Thickness)
Aspect Ratio (DptLenth Ro Circumferential Flaw Axial Flaw (Depth/Length) 0.20 57%
75%
0.33 73%
75%
0.50 75%
75%
Page 11 of 29
Westinghouse Non-Proprietary Class 3 Document Control Desk CR-12-04775 RC-1 2-0173 600 Ii Outside Surface Flaw I
Alloy 52/52M Repair Weld Figure 2-1 A Schematic of the Repair Configuration for the Outside Surface Flaw Note: The outside surface flaw shown in the figure is for repair configuration illustration purposes and does not intend to represent the actual outside surface flaws detected at V. C. Summer Unit 1 Page 12 of 29
Westinghouse Non-Proprietary Class 3 Document Control Desk CR-12-04775 RC-1 2-0173 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0
Maximum Allowable Initial Axial Flaw Size Curve For I-----~
Repaired Penetration Nozzles with 40 Years of Servlce Life -
-- "--ilr-S.
Penetration 19 Ind. #1&2 a= 0.283" = 0.72" Penetration 37 Ind. #3
- a. = 0.294"1=0.52' Penetration 31 Inc#
a 0.256"1 0.61 enexTation52In.#
a=0.279"1=0.32" I
I Penetration 37 d#
/,
a= 0.249"1 076" Penetration37 Ind. #2
-* ---- I a, = 0.214" 1 o.s6-
'p L
Penetration31 Ind. #2 a= 0.177" I = 0.36" 11 - -- 11 T
, - - 11 T
Penetration 31 Ind. #1 a=0.122"1=0.52"
-- enetrationS52 Ind. #3 a= 0.132"1=0.12" 0.2 0.25 0.3 0.35 Flaw Depth to Flaw Length Ratio (a/I) 0.4 0.45 0.5 Figure 2-2 Maximum Allowable Initial Axial Flaw Sizes for Repaired Penetration Nozzles Page 13 of 29
Westinghouse Non-Proprietary Class 3 Document Control Desk CR-12-04775 RC-12-0173 0.8 0.7 0.6 a
0.5 0.4 ow0.3 0.2 0.1 0
MaximumAllowable Initial Circumferential Flaw Size Curve ForRepaired Penetration Nozzles with 40 Years of Service Life t- +---
L
+
+
L
= _ _
T i
I r
r I
r Penetration 37 Ind. #3 a= 0.294" i =0.157" Penetration 52 Ind. #2 a= 0.279" 1 = 0.471" Penetration 31 Ind. #3 a =0.256" i = 0.262" N
Penetration 19 Ind. #1&2 a= 0.283" I = 1.361"
...11.....
1111_+
_+
-'----1 I-
-I I
IT I1 Penetration37Ind.#1 a= 0.249" I = 0.314"
- .. I.. -----
+:+++++
Penetration 37 Ind #2 a= 0.214"1 =0.105" Penetrationi31 Id, #2 a= 0.177" 1 =0157" I
T - - -
I - - - T I
L_
II 11 -
--- -- ---- I I
I enetration 52 Ind. #3 1--
8--
a=0.132" =0.209.
.... * --+... :.
.. ;-- * - -+ - +I... ; - -r + -; - -;
- - ++ - +
+ -
Penetration 31 Ind. #1 a= 0.1 22" =0.157"..
0.2 0.25 0.3 0.35 0.4 0.45 0.5 Flaw Depth to Flaw Length Ratio (a/l)
Figure 2-3 Maximum Allowable Initial Circumferential Flaw Sizes for Repaired Penetration Nozzles Page 14 of 29
Westinghouse Non-Proprietary Class 3 Document Control Desk CR-1 2-04775 RC-1 2-0173 0.80 0.75 o.7o~~~~~~
!___!_L!2.i_.L!
.. L!
_i.
L._L__- ---- !-...
L
)----
0.70f
~~~
-ntaal=4
=0---
= 0.55
_________Axia Flawfrnta Fa Inta f 0457 a
Inti/
047.2302
- "0.60
-;- - -! - -; - - ;* - L-i -; i i...
n t a a / =. 7 a 1 0 2
-i -I - -
C r i -ii i
= 040
________ii i
i i
i 1
1
! i i 0.3 5 7
0.3 0 0
5 10 15 20 25 30 35 40 45 50 Time (years)
Figure 2-4 Fatigue Crack Growth For Hypothetical Bounding Axial and Circumferential Flaws Page 15 of 29
Document Control Desk Westinghouse Non-Proprietary Class 3 CR-12-04775 RC-12-0173 3.0 TECHNICAL BASIS FOR APPLICATION OF EMBEDDED FLAW REPAIR PROCESS TO PENETRATION NOZZLE ATTACHMENT WELDS This section provides a discussion on the technical basis for the 'use of embedded flaw repair process for potential flaws in the attachment J-groove welds of penetration number 19, 31, 37 and 52. Such a repair process involves depositing Alloy 52/52M repair weld material over the wetted surface of the attachment J-groove welds and on the outer diameter of the penetration nozzles of interest below the attachment J-groove welds in order to seal the crack from the primary water environment. At least three weld layers of Alloy 52/52M repair weld material are deposited (3600 full circumference) covering the wetted surface of the penetration nozzle J-groove weld including 0.5 inch past the J-groove weld buttering and stainless steel cladding interface as well as at least two weld layers covering the entire outside surface of the head penetration nozzle. A schematic of the repair configuration for the attachment J-groove weld is illustrated in Figure 2-1. Since the current available technology cannot characterize the depth of a flaw in the attachment J-groove weld, it is therefore conservatively assumed that the flaw extends radially over the entire attachment J-groove weld cross-section. A flaw evaluation was performed by assuming a flaw of that size in the J-groove welds of the penetration nozzles of interest in developing the technical basis for the embedded flaw repair process.
3.1 EVALUATION PROCEDURE AND ACCEPTANCE CRITERIA 3.1.1 ASME Section Xl Appendix K The evaluation procedure and acceptance criteria used to demonstrate structural integrity of the reactor vessel closure head are contained in Appendix K of ASME Section XI Code [2] as well as Regulatory Guide 1.161 [8]. Although the original purpose of Appendix K was to evaluate reactor vessels with low upper shelf fracture toughness, the general approach in paragraph K-4220 is equally applicable to any region of the reactor vessel where the fracture toughness can be described with elastic plastic parameters. The closure head region of the V. C. Summer Unit 1 reactor vessel has an operating temperature of approximately 5570F. This would result in ductile behavior and therefore the use of the elastic-plastic fracture mechanics method is appropriate.
The approach to evaluating the integrity of a nuclear vessel has been developed over a ten-year period, and has been illustrated with a number of example problems [9] to demonstrate its use.
The extension of this methodology to issues other than the low shelf fracture toughness issue is appropriate when service conditions (temperature) promote ductile behavior. The extension of the Elastic Plastic Fracture Mechanics (EPFM) method to the reactor vessel head is appropriate, as discussed above.
The acceptance criteria are to be satisfied for each category of transients, namely, Service Load Level A (normal), Level B (upset), Level C (emergency) and Level D (faulted) conditions and two criteria must be satisfied.
The first criterion is that the crack driving force must be shown to be less than the material toughness as follows:
Page 16 of 29
Document Control Desk Westinghouse Non-Proprietary Class 3 CR-12-04775 RC-12-0173 Japplied < JO.l where Japplied is the J-integral value calculated for the postulated flaw under the applicable Service Level condition and J0.1 is the J-integral characteristic of the material resistance to ductile tearing at a crack extension of 0.1 inch.
The second criterion is that the flaw must also be stable under ductile crack growth as follows:
aJapplied d
material aa da at Japplied = Jmaterial
- where, Jmateral
=
J-integral resistance to ductile tearing for the material ajapplied
= Partial derivative of the applied J-integral with respect to flaw depth, a caa dmaterial -
Slope of the J-R curve da 3.1.2 Primary Stress Limits In addition to satisfying the above acceptance criteria, the primary stress limits of paragraph NB-3000 in Section III of the ASME Code must be satisfied. The effects of a local area reduction of the pressure retaining membrane that is equivalent to the area of the postulated flaw in the reactor vessel head attachment J-groove weld must be considered to reflect the reduced cross section.
3.2 Methodology The embedded flaw repair process is used to seal any potential flaw in the J-groove weld from further exposure to the primary water environment.
The evaluation was performed to demonstrate the stability of the assumed flaw that encompasses the entire attachment J-groove weld region in the reactor vessel head near the penetration nozzle. The flaw is stable under ductile crack growth if the acceptance criteria in Section 3.1 are met. With the embedded flaw repair process, the only mechanism for sub-critical crack growth is fatigue. Therefore, fatigue crack growth evaluations for the postulated flaw in the reactor vessel head were also performed to demonstrate structural integrity.
The requirement for evaluating flaw stability in the reactor vessel upper closure head in accordance with the evaluation procedures and acceptance criteria in ASME Section Xl code is Page 17 of 29
Document Control Desk Westinghouse Non-Proprietary Class 3 CR-12-04775 RC-1 2-0173 that the governing transients resulting from the normal/upset conditions as well as the emergency/faulted conditions be considered.
3.2.1 Geometry and Source of Data There are many penetration nozzles in the reactor vessel upper head.
The outermost penetration nozzles (46.00 intersection angle) were selected for thermal transient and residual stress analysis because the stresses in the outermost penetration nozzles are more limiting and can be used to conservatively represent those at penetration nozzle number 19, 31, 37 and 52.
The dimensions of all the V. C. Summer Unit 1 penetration nozzles are identical, with a 4.00 inch nominal outside diameter and a nominal wall thickness of 0.625 inch [3]. The distributions of residual, thermal transient and pressure stresses in the upper head penetration nozzle were obtained from the detailed three-dimensional plant specific elastic-plastic finite element analyses [4]. The through-wall stress distributions from the finite element analyses were used to determine the fatigue crack growth as well as determining flaw stability under ductile crack growth.
3.2.2 Stress Intensity Factor One of the key elements in the fracture mechanics evaluation is the determination of crack tip stress intensity factor (K,). The stress intensity factor expression for two corner flaws emanating from the edge of a hole in a plate [6] was used in determining the stress intensity factor for the assumed flaw in the attachment J-groove weld as shown in Figure 3-1.
The stress intensity factor can be expressed in terms of the membrane and bending stress components as follows:
K, =(Mmam + Mbob) (Tra/Q) 112
- where, K,
=
Crack Tip Stress Intensity Factor m=
Remote Membrane Stress Component a
=
Remote Bending Stress Component Mm
=
Membrane Boundary Correction Factor Mb
=
Bending Boundary Correction Factor Q
=
Shape Factor a
=
Depth of the Corner Flaw (See Figure 3-1)
Use of this method requires that the stresses remote from the hole be resolved into membrane and bending stress components.
The attachment J-groove weld shapes were based on the J-groove geometry shown in the head penetration nozzle drawing [10] for the head penetration nozzles of interest.
Page 18 of 29
Document Control Desk Westinghouse Non-Proprietary Class 3 CR-1 2-04775 RC-12-0173 3.2.3 Material Properties One of the most important information on the toughness for pressure vessel and piping materials is the J-R curve of the material, where J-R stands for material resistance to crack extension, as represented by the measured J-integral value versus crack extension. Simply put, the J-R curve to cracking resistance is as significant as the stress-strain curve to load-carrying capacity and ductility of a material.
Both the J-R curve and the stress-strain curve are properties of a material.
Directly measured J-R curves are not generally available for a specific material of interest. The J-integral fracture resistance of the material is determined using the Regulatory Guide 1.161 [8]
and NUREG/CR-5729 [11] from available data such as material chemistry, radiation exposure, temperature and Charpy V-notch energy. The method summarizes a large collection of test data, which were fitted into multivariable models based on advanced pattern recognition technology [11]. Separate analysis models and databases were developed for different material groups, including reactor pressure vessel (RPV) welds, RPV base metals, piping welds, piping base metals and a combined materials group.
The material resistance, Jmat, is fitted into the following equation [8, 11]:
Jmat = (MF)C1 (Aa)C2exp[C3(Aa)C4]
where C1, C2, C3, and C4 are fitting constants, MF is the margin factor and Aa is crack extension For the RPV base metal model, the constants C1, C2, C3, and C4 are defined in Table 11 of NUREG/CR-5729 [11].
Neutron irradiation has been shown to produce embrittlement that reduces the toughness properties of the reactor vessel ferritic steel material. The irradiation levels are very low in the reactor vessel closure head region and therefore the fracture toughness will not be measurably affected.
3.2.4 Applied J-Integral For small scale yielding, the applied J-integral, Jappiied, of a crack can be calculated using the Linear Elastic Fracture Mechanics (LEFM) method based on the crack tip stress intensity factor, K1, calculated as discussed in Section 3.2.2.
However, a plastic zone correction must be considered to account for plastic deformation at the crack tip similar to the approach in Regulatory Guide 1.161 [8]. The plastic deformation ahead of the crack front is regarded as a failed zone and the crack size is, in effect, increased. The K1-values based on the plastic zone adjusted crack depth can then be converted to Japplied by the following equation:
K2 Japplied
'ep
- where, b
Page 19 of 29
Document Control Desk Westinghouse Non-Proprietary Class 3 CR-12-04775 RC-12-0173 Kep is the elastically calculated KI-value based on the plastic zone adjusted crack depth E' = E/(1-v 2) for plane strain, E is the Young's Modulus and v, the Poisson's Ratio.
The plastic zone correction, rp, can be calculated as follows:
rp Sy where Sy is the yield strength of the material and K, is the stress intensity factor calculated in accordance with Section 3.2.2.
Assuming that the original crack depth under consideration is a0, Kep can now be calculated based on the plastic zone adjusted crack depth, a0 + rp.
Once the applied J-integral is calculated, flaw stability for the postulated flaw in the attachment J-groove weld can be determined using the J-R curve developed in Section 3.2.3 for the V. C.
Summer Unit 1 reactor vessel closure head material.
3.2.5 Fatigue Crack Growth The fatigue crack growth analysis involves calculating crack growth for the planar flaws that are assumed to extend radially over the entire attachment J-groove weld cross-section in the reactor vessel closure head which are subjected to a series of operating transient loadings. The loadings included pressure, thermal transients, and residual stresses. The design thermal transients as well as the associated design cycles are listed in Table 2-2. The design transient cycles are distributed evenly over the plant design life. The stress intensity factor range, AKI, which controls the fatigue crack growth, depends on the geometry of the crack, its surrounding structure and the range of applied stresses in the region of the postulated crack.
The methodology used in determining the stress intensity factor, K1, is discussed in Section 3.2.2.
Once AKI is calculated, the fatigue crack growth due to a particular stress cycle can be determined using a crack growth rate reference curve applicable to the material where the crack is postulated.
Once the incremental crack growth corresponding to a specific transient for a small time period is calculated, it is added to the previous crack size and the analysis continues to the next time period and/or thermal transient. The procedure is repeated in this manner until all the significant analytical thermal transients and cycles known to occur in a given period of operation have been analyzed.
3.3 FRACTURE MECHANICS ANALYSIS RESULTS 3.3.1 Comparison of Applied J-lntegral and Material J-R Curves The actual geometry or weld shapes for the V. C. Summer head penetration attachment J-groove welds of interest [10] are shown in Table 3-1 and Figure 3-2, which forms the basis for Page 20 of 29
Document Control Desk Westinghouse Non-Proprietary Class 3 CR-2-04775 RC-1 2-0173 the geometry of the analyzed flaws in the attachment J-groove weld region. For simplicity and conservatism, the largest boundary correction factors for all the J-groove weld flaws of interest, Mm and Mb, were conservatively used in determining the stress intensity factor, Ki.
Table 3-1 Geometry of V. C. Summer Unit 1 Head Penetration Nozzle Downhill Side Attachment J-groove Welds (All dimensions in inches) a from [10]
a from UT data c from [10]
Penetration No.
(without weld (included weld (without weld fillet) fillet fillet) 19 1.264 1.72 1.438 31 1.307 2.32 1.542 37 1.382 1.88 1.730 52 1.504 2.20 2.070 The applied J-integral values were then calculated based on the elastically calculated K, adjusted for the plastic zone correction as discussed in Section 3.2.4. The material J-R curve was obtained as discussed in Section 3.2.3 and the applied J-integral values and the material J-R curve were plotted in Figure 3-3. Using the acceptance criteria discussed in Section 3.1, the structural integrity of the reactor vessel closure head with the planar flaws that are assumed to encompass the entire cross-section of the J-groove region can then be determined.
The key aspects of the structural integrity evaluation are the values of the Jappued versus Jmaterial for the reactor vessel closure head material and the slope of the Japplied curve versus the slope of the material J-R curve as discussed in Section 3.1.1.
Figure 3-3 demonstrated the structural stability of the most limiting flaw shape without the weld fillet for the closure head at penetration nozzle number 19, 31, 37 and 52. In Figure 3-3, it can be seen that for a crack extension of 0.1 inch with an initial flaw depth of 1.504 inch, the applied J-integral value is below that of the material J-R curve. In addition, the slope of the material J-R curve exceeds that of the J-applied curve at the equilibrium point where the Japplied curve intersects the Jmaterial curve. Since the acceptance criteria in Section 3.1 are met, it can be concluded that structural stability can be demonstrated for the assumed flaws with the attachment J-groove shapes tabulated in Table 3-1 for penetration nozzle number 19, 31, 37 and 52.
3.3.2 Primary Stress Limits In addition to satisfying the above acceptance criteria, the primary stress limits of paragraph NB-3000 in Section III of the ASME Code must be satisfied. The effects of a local area reduction of the pressure retaining membrane that is equivalent to the area of the assumed planar flaw in the vessel head attachment J-groove welds of interest must be considered to reflect the reduced cross section. The allowable local area reduction was determined by evaluating the primary membrane stress of a spherical head with reduced wall thickness and based on a maximum operating pressure under various service conditions.
The result shows that the calculated Page 21 of 29
Document Control Desk Westinghouse Non-Proprietary Class 3 CR-12-04775 RC-12-0173 allowable flaw depth is 1.718 inches which is deeper than the maximum attachment J-groove depth of 1.504 inches for penetration nozzle number 19, 31, 37 and 52.
3.3.3 Fatigue Crack Growth into the Reactor Vessel Closure Head Fatigue crack growth into the reactor vessel head was determined for the assumed flaw with the J-groove shape without considering the weld fillet on the downhill side of penetration nozzle number 19, 31, 37 and 52. The flaw is conservatively assumed to be a surface flaw even though it is a subsurface flaw after the repair. The crack growth rate curves used in the fatigue crack growth analysis are taken directly from Appendix A in the ASME Section XI code [2] for ferritic steel material. Since any potential flaws in the attachment J-groove weld are sealed from the primary water environment after the repair, the crack growth rate reference curve for the air environment is used. This curve is a function of the applied stress intensity factor range (AK1) and the R ratio, which is the ratio of the minimum to maximum stress intensity factor during a thermal transient.
As shown in Figure 3-4, the predicted crack growth for the bounding J-groove shape is small and is below the allowable primary stress limit flaw depth of 1.718 inches even after 20 years.
Based on the fatigue crack growth results, the assumed planar flaws which encompass the entire cross-section of the J-groove shape for the head penetration nozzles of interest can be shown to be acceptable for at least 20 years of service life based on the design thermal transients and cycles tabulated in Table 2-2.
3.3.4 Fatigue Crack Growth into the Repair Weld Attachment J-groove weld repair was performed by depositing a minimum of three layers (-3/16 inch) of Alloy 52/52M repair weld material onto the wetted surface of the attachment J-groove weld with potential flaws. The flaw is thus sealed from the primary water environment, and the thickness of the reactor vessel head is conservatively assumed to be locally increased by approximately 3/16 inch. In the fatigue crack growth analysis, an embedded flaw is assumed, which starts conservatively from 5/32 inch above the free surface on the inside surface of the reactor vessel head.
The depth of the embedded flaw is conservatively assumed to be 2.32 inches which consists of the entire attachment J-groove as well as the weld fillet as shown in Table 3-1 and Figure 3-2. This embedded flaw depth would envelop those in head penetration nozzle number 19, 31, 37 and 52 on the downhill side in V. C. Summer Unit 1. In other words, the postulated initial embedded flaw has a total crack depth of 2.32 inch with a crack front that is located at 5/32 inch above the free surface. The fatigue crack growth law used for the repair layer (Alloy 52/52M) is based on the crack growth rate for nickel-base alloy material (Alloy 600) in air environment with a factor of 2 for conservatism and assumed to be the same as those for Alloy 182 weld material in air environment [7]. The fatigue crack growth results are shown in Table 3-2 and the resulting fatigue crack growth for 20 years is insignificant. Therefore the structural integrity of the repaired weld layer is expected to be maintained for at least 20 years of service life.
Page 22 of 29
Westinghouse Non-Proprietary Class 3 Document Control Desk CR-12-04775 RC-12-0173 Table 3-2 Predicted Fatigue Crack Growth into the Repaired Weld Layer Distance of the Crack Front from the Nearest Free Surface (inch) 0 0.156 Downhill 20 0.142 3.4 Conclusion The results of the evaluation have demonstrated that the embedded flaw repair process is a viable method for repairing any potential flaws found in the attachment J-groove welds of the head penetration nozzle number 19, 31, 37 and 52 at V. C. Summer Unit 1. The fracture mechanics evaluation demonstrated that the flaws in the penetration nozzle that are assumed to encompass the entire J-groove shape are acceptable even when the fatigue crack growth for at least 20 years was taken into consideration.
Page 23 of 29
Westinghouse Non-Proprietary Class :3 Document Control Desk CR-12-04775 RC-1 2-0173 Figure 3-1 Geometry and Terminology Used in Stress Intensity Factor Calculation Page 24 of 29
Westinghouse Non-Proprietary Class Document Control Desk
I, I.
Figure 3-2 Definition of J-groove Weld Dimensions Page 25 of 29
Westinghouse Non-Proprietary Class 3 Document Control Desk CR-2-04775 RC-1 2-0173 E" 1.00 C
.* 0.80 0.60 0.00 1.60 1.55 1.60 1.65 1.70 1.75 1.80 Crack Depth (inch)
Figure 3-3 Bounding Applied J-integral and Material J-R Curve Page 26 of 29
Westinghouse Non-Proprietary Class 3 Document Control Desk CR-12-04775 RC-1 2-0173 1.80 1.60 1.40 1.20
,. 1.00 30.80 0.60 0.40 0.20 0.00 0
2 4
6 8
10 12 14 16 18 20 Time (Years)
Figure 3-4 Fatigue Crack Growth Prediction into the Reactor Vessel Head with Bounding Flaw Depth in the Attachment J-groove weld Page 27 of 29
Document Control Desk Westinghouse Non-Proprietary Class 3 CR-12-04775 RC-12-0173 4.0 Conclusions The unacceptable outside surface flaws in the penetration nozzles of interest are isolated from the primary water environment using the Westinghouse embedded flaw repair process. Primary water stress corrosion is no longer a credible degradation mechanism and fatigue is the only credible crack growth mechanism. The left behind flaws in the repaired head penetration nozzle number 19, 31, 37 and 52 including any potential flaws in the attachment J-groove weld have been shown to be acceptable for continued operation for at least 20 years after the repair.
These upper head penetration nozzles will be inspected every refueling outage following the repair. It is therefore technically justified to use the embedded flaw repair process as the repair technique for the reactor vessel head penetration nozzles with the unacceptable outside surface flaws since the criteria for application of such a process as stated in Appendix C of WCAP-15987-P [1] are met.
5.0 References
- 1.
Westinghouse WCAP-15987-P, Revision 2-P-A, "Technical Basis for the Embedded Flaw Process for Repair of Reactor Vessel Head Penetrations,"
December 2003. (Westinghouse Proprietary Class 2)
- 2.
ASME Section Xl Code:
- a.
ASME Boiler & Pressure Vessel Code, 1998 Edition through 2000 Addenda, Section Xl, Rules for Inservice Inspection of Nuclear Power Plant Components.
- b.
ASME Boiler & Pressure Vessel Code, 2007 Edition with 2008 Addenda,Section XI, Rules for Inservice Inspection of Nuclear Power Plant Components.
- 3.
Chicago Bridge & Iron Company Drawing No. 40, Contract No. 71-2631, "157" PWR Control Rod Drive Mechanism Housings Details," Revision 6.
- 4.
Dominion Engineering, Inc. Report C-8849-00-01 Rev. 0, "V.C. Summer RPV Head CRDM Nozzle Welding Residual Stress plus Transient Analysis". (Dominion Engineering Inc. Proprietary Document)
- 5.
Design Specification DS-MRCDA-09-10, Revision 0, Equipment: Reactor Vessel -
Virgil C. Summer Nuclear Station Addendum to Equipment Specification 679105 Rev. 2. (Westinghouse Proprietary Class 2)
- 6.
American Petroleum Institute, API 579-1/ASME FFS-1 (API 579 Second Edition),
"Fitness-For-Service," June 2007.
Page 28 of 29
Document Control Desk Westinghouse Non-Proprietary Class 3 CR-12-04775 RC-12-0173
- 7.
NUREG/CR-6721, ANL-01/07, "Effects of Alloy Chemistry, Cold Work, and Water Chemistry on Corrosion Fatigue and Stress Corrosion Cracking of Nickel Alloys and Welds," April 2001.
- 8.
Regulatory Guide 1.161, "Evaluation of Reactor Pressure Vessel with Charpy Upper-Shelf Energy Less Than 50 ft-lb."
- 9.
"Development of Criteria for Assessment of Reactor Vessels with Low Upper Shelf Fracture Toughness," Welding Research Council Bulletin 413, July 1996.
- 10.
Chicago Bridge & Iron Company Drawing No. 42, Contract No. 71-2631, "157" PWR CRDM Housing Installation," Revision 6.
- 11.
E. D. Eason, J. E. Wright, E. E. Nelson, "Multivariable Modeling of Pressure Vessel and Piping J-R Data," NUREG/CR-5729, MCS 910401, RF, R5, May 1991.
Page 29 of 29