ML20237J610
| ML20237J610 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 04/30/1987 |
| From: | Maxham W, Yoon K BABCOCK & WILCOX CO. |
| To: | |
| Shared Package | |
| ML20237J594 | List: |
| References | |
| BAW-1999, NUDOCS 8708260236 | |
| Download: ML20237J610 (42) | |
Text
_ _ _ ___
BAW-1999 April 1987 TMI-1 Nuclear Power Plant Leak-Before-Break Evaluation of Margins Against Full Break for RCS Primary Piping
\\
r s
f
\\
f
\\
/
\\
h
\\
i i
l l
a s
I s
I
/
\\
4 i
p' l
l
'l l
)
%=
T\\[b
)
- b]
i i
\\
\\
/
t?
U f[kD $f>hh{ f3 babcock &WilCOX
!D t, PI}n
- M' P
l BAW-1999 April 1987 4
\\
TMI-1 Nuclear Power Plant Leak-Before-Break Evaluation I
of Margins Against Full Break for RCS Primary Piping by W. D. Maxham K. K. Yoon THE BABC0CK & WILCOX COMPANY Nuclear Power Division P. O. Box 10935 Lynchburg, VA 24506-0935
~
This document is the property of GPU Nuclear Corporation.
Distribution to or reporduction of this document by individuals or organizations other than GPU Nuclear
~
Corporation is prohibited without the written consent of GPU Nuclear Corporation.
Babcock &Wilcox
_ $$g,
a McDermott company
CONTENTS Page 1.
SUMMARY
1-1 2.
INTRODUCTION AND DISCUSSION....................
2-1 2.1.
Effect of GDC-4 Criteria on RCS Designed Loads 2-1 2.2.
Application of Leak-Before-Break Concept to Postulated Breaks in the RCS Primary Piping.........
2-2 2.2.1.
Leak-Before-Break Concept..............
2-2 2.2.2.
Applicability of Leak-Before-Break Concept to RCS Primary Piping of the B&W NSS Design 2-3 3.
LEAK-BEFORE-BREAK EVALUATION OF RCS PRIMARY PIPING 3-1 3.1.
Scope of Investigations for Generic LBB Evaluations.....
3-1 3.1.1.
RCS Piping Structural Loads.............
3-1 3.1.2.
Leakage Flaw Size Determination...........
3-1 3.2.
LBB Evaluation Criteria...................
3-2 3.2.1.
Leak Rate Criteria 3-2 3.2.2.
Postulated Flaw Size 3-2 3.2.3.
Flaw Stability 3-2 3.2.4.
Limit Load Analysi s.................
3-3 3.3.
LBB Investigation Methodology................
3-3 3.3.1.
Determination of Generic Bounding Loads.......
3-3 3.3.2.
Fatigue Flaw Growth Analysis 3-4 3.3.3.
Determination of Leak Rate Flaw Size 3-4 3.3.4.
RCS Primary Piping Materials Data..........
3-4 3.3.5.
Flaw Stability Analysis..
3-4 3.3.6.
Limit Load Analysis.................
3-6 4.
LBB EVALUATIONS RESULTS FOR THE RCS PRIMARY PIPING 4-1 4.1.
RCS Operating Conditions 4-1 4.2.
Leak Rate Versus Flaw Length 4-1 4.3.
Fatigue Flaw Growth Analysis 4-2 4.4.
RCS Material Properties...................
4-2 4.4.1.
Weld Metal Properties................
4-2 4.4.2.
Base Metal Properties................
4-3 4.5.
Fl aw Stability Analysis...................
4-4 4.5.1.
Applied Loads....................
4-5 4.5.2.
Applied Moment Versus J for 28-Inch Pipe 45 4.5.3.
Applied Moment Versus J for 36-Inch Pipe 4-6 4.5.4.
Limit Load Analysis.................
4-7 4.6.
Summary of Safety Margins for TMI-1 RCS Piping 4-8 Babcock & Wilcox
,j, a McDermott company
N CONTENTS (Cont'd)
Page 5.
CONCLUSIONS............................
5-1 6.
REFERENCES 6-1 LIST OF TABLES i
Table 3-1.
TMI-1:
Leakage Flaw Size for Various RCS Pipes.........
3-7 4-1.
TMI-1 Maximum Applied Forces, Moments, and Flaw Sizes
~
~
used in Fracture Mechanics Analysis...............
4-9 4-2.
TMI-1 Leakage Flaw Size for Various RCS Pipes..........
4-10 4-3.
Maximum Applied Loads for Fracture Mechanics Analysis......
4-11 4-4.
Combined Applied Loads for TMI-1 4-11
('
{
4-5.
Limit Load Calculation for TMT-1 28" and 36" Pipes 4-12 LIST OF FIGURES Figure 3-1, 177-FA Plant Lower Loop Hot Leg Weld Locations 3-8 3-2.
177-FA Plant Lower Loop Cold Leg Weld Locations.........
3-9 3-3.
Locations, Weld Type and Materials in a Typical B&W Designed NSS 3-10 3-4.
Illustration of Stability Assessment Program 3-11 3-5.
Illustration of Pipe Geometry..................
3-12 4-1.
Leak Flow Rate Vs Crack Length for 36" ID Straight Pipe Subjected to an Axial Stress of 6.62 ksi and Various Bending Moments.........................
4-13 4-2.
J-R Curve for Weld Metal 4-14 4-3.
True Stress-True Strain Curve for Weld Metal 4-15 4-4.
J-R Curve for Base Metal 4-16 4-5.
True Stress-True Strain Curve for Base Metal 4-17 4-6.
J Vs. Moment Diagram for 28 Inch Pipe with Base Metal Properties 4-18 4-7.
J Vs. Moment Diagram for 36 Inch Pipe with Base Metal Properties 4-19
~
Babcock &Wilcox
_ jj,
a McDermort company g.
r' l
1.
SUMMARY
One of the many design criteria and requirements for the design of Nuclear Power Plants is referred to in 10CFR50 Appendix A as General Design Criteria 4 (GDC-4).
The Nuclear Regulatory Commission's (NRC) interpretation of GDC-4 requires that components such as the reactor vessel (RV) internals, main reactor coolant system (RCS) piping, and piping attached to the RCS piping be able to withstand loads imposed upon them by certain postulated accidents.
These accidents include main steam line break (MSLB), operating basis earthquake (0BE),
safe shutdown earthquake (SSE),
and loss-of-coolant accident (LOCA).
The LOCA definition assumes that the large diameter, thick wall RCS piping ruptures instantly in a guillotine fashion (double-ended guillotine break, DEGB), allowing full flow from both ends of the ruptured pipe..The DEGB requirement historically has been imposed on structures and components even though no known mechanisms could be identified that would cause the piping to fail so catastrophically.
As a result, with such large breaks, extremely high forces and moments were determined to exist in the piping, and thus, massive restraints were required to limit pipe whip and the jet impingement of discharged fluid on surrounding components and equipment.
Experimental and analytical evidence now exist that shows the main RCS piping will not experience such catastrophic failure, as had been previously postulated.
Indeed, since 1985 when the original B&WOG Leak-Before-Break (LBB) was completed and approved by NRC, the NRC has modified the GDC-4 ruling to allow its application to other high energy piping systems than the main RCS piping.
Thi.s report, using the LBB concept approved by NRC in their 1985 review of l
the B&WOG's LBB report (BAW-1847, Rev. 01), describes the successful applica-tion of the LBB concept to ths RCS main piping of GPU Nuclear TMI-l nuclear plant.
s 1-1 Babcock &Wilcox a McDermott company
This report provides the technical basis for evaluating postulated flaw j
growth in the main RCS piping of the.B&W designed TMI-I NSS under normal-plus-faulted loading conditions.
~
This LBB evaluation of the main RCS piping in the TMI-1 plant has shown that a double-ended guillotine break will not occur and that postulated flaws produce detectable leakage and exhibit stable growth thus, allowing a con-
}
trolled plant shutdown before any potential exists for catastrophic piping failure.
Thus, the THI-1 passive restraints, jet shields, and snubbers presently installed for LOCA only purposes are not required and can be removed.
In addition, any questions as to the impact of Asymmetric Cavity Pressure loading resulting from double-ended rupture of the main RCS piping are no longer valid.
~
. l 1-2 Babcock & WHcox a McDermott company e
2.
INTRODUCTION AND DISCUSSION The investigation described in this report evaluates applying the LBB concept to the RCS primary piping of the TMI-1 Nuclear Power Plant.
This report is written as if it were an appendix to the original B&W Owners Group Report; BAM-1847, Rev. 1.1 The organization of the report is the same as the BAW-1847, Rev.1, herein referred to as the " Original" report.
Since this report uses the methodology discussed in the Owners Group LBB report the paragraph numbering is retained for clarity and to facilitate cross referencing.
The original report and tables are n-t being revised by this appendage but are being supplemented.
Thus, the tables and figures have retained the same identification as the original report with the exception that "TMI-1" has been added to the tables, e.g., Table 3-1 TMI-1.
The scope of this evaluation includes performing structural, leakage and fracture mechanics analyses for those load sets that are not bound by the
" Original" report.
The major licensing changes which have occurred since the generic report was released are that the currently approved requirements of Title 10CFR50, Appendix A, General Design Criteria-4 (GDC-4) states that the dynamic effects associated with postulated pipe ruptures of primary coolant loop piping in pressurized water reactors may be excluded from the design basis when analyses demonstrate the probability of rupturing such piping is extremely low under design basis conditions.
The Nuclear Regulatory Commission is presently propo. ting to expand the scope of GDC 4 to cover all high energy piping in the plants.
"This expanded modification is based on recent research and insights from probabilistic risk analysis."2 2.1.
Effect of 6DC-4 Criteria on RCS Desianed Loads Since the release of the Original Report; GDC-4 has been revised to ack-nowledge the leak before break concept.
This original report discusses the loss of Coolant Accident (LOCA) loads prior to leak-before-break concept o
2-1 Babcock &WHcox a McDermott co.npany
being formalized.
The present GDC-4 allows the application of the leak-before-break concept to piping systems to determine the safety margins existing between flaw sizes that cause detectable leaks and unstable flaws that could lead to double ended pipe rupture.
Thus, a qualified piping system does not need to consider LOCA loads in the design of restraints for that system.
2.2.
Application of Leak-Before-Break Concept to Postulated Breaks in the RCS Primary Pinina The original report furnishes similar information under this subsection and more specifically the applicability of the LBB concept to RCS primary piping of the B&W NSS design.
Since the original report was completed, the NRC has modified GDC-4 of 10CFR50, Appendix A to allow the application of the LBB concept to certain additional piping systems. The main RCS piping system is the system analyzed in this report.
The general methodology for the application of LBB to piping systems is now provided in NUREG-1061.
2.2.1.
Leak-Before-Break Concept The LBB concept was defined in the original report and is included here for cl arity.
The LBB concept has been proposed to justify eliminating the DEGB as a postulated mechanistic event.
Relief from this requirement would eliminate the need to design and install costly and massive restraints.
The LBB concept is based on a plant's ability to detect an RCS leak and applies known mechanisms for flaw growth to piping designs with assumed through-wall flaws.
To ensure adequate margins exist for leak detection, the analysis assumes a leak rate larger than the minimum plant capability.
With various system operating parameters and RCS leak detection capabilities providing the status of the RCS to the operator, a leakage is readi'y detected and thus an orderly and controlled plant shutdown can be perforted before any potential exists for catastrophic pipe failures.
Thus, the objective of the application of the LBB concept is to determine tht t a postulate dflaw remains stable under normal operating plus faulted loads or 2-2 Eabcock &WHcox a McDermott company
that significant margins exist against unstable flaw growth if the postulated flaw is predicted to grow with the applied loads.
2.2.2.
Applicability of Leak-Before-Break Concept to RCS Primary Pioina of the B&W NSS Desian The applicability of LBB to the B&W NSS design was successfully shown in 1985.
This report further extends the application to GPU Nuclear's TMI-1 plant.
Since the original report was issued the B&WOG Utilities have continued to operate the B&W-designed NSS. The accumulated operating time is over 62 reactor years spanning some 15 calendar years, without indications of mechanisms that could cause flaw growth in the RCS piping.
1 l
l
\\
2-3 Babcock & Wilcox a McDermott company
T-e 3.
LEAX-BEFORE-BREAK EVALUATION OF RCS PRIMARY PIPING To apply the leak-before-break concept to any piping systems requires various g
engineering disciplines, computer codes, and various material properties.
In addition, an in-depth evaluation of the assumptions, input data, and method-ology used in the LBB analysis must be assessed to maintain a conservative, yet realistic RCS primary piping evaluation.
It is emphasized that strict manufacturing and nondestructive tests were completed on all RCS piping during fabrication and erection by the NSS vendor, thus the piping is not likely to contain any significant size
~
undetected flaws.
3.1.
Scoce of Investigations for Generic LBB Evaluations The investigation into the applicability of the LBB concept to the RCS primary piping is divided into four technical areas:
e RCS piping structural loads e Leakage flaw size determination o RCS piping material properties e RCS piping fracture mechanics and limit load analysis The four major areas of investigations that contribute to the final results are outlined in the original report and briefly discussed below:
3.1.1.
RCS Pioina Structural Loads The piping loads at various locations on the hot and cold legs of the RCS piping were obtained from existing stress reports for the TMI-1 plant.
The loading combinations of deadweight (DW), thermal loading cases (TH), and the applicable SSE loads are discussed in detail in section 3.3.1.
3.1.2.
Leakaae Flaw Size Determination The leakage flaw size was determined using the minimum TMI-1 plant operating loads applicable for the particular pipe size and configuration. A leak rate 3-1 Babcock &Wilcox a McDermott rompany
of 10 gpm was used for determining the leakage flaw size in order to provide a margin to the actual plant leakage detection system capabilities.
3.2.
LBB Evaluation Criteria 3.2.1.
Leak Rate Criteria To establish the postulated through-wall flaw size for evaluation by fracture mechanics analysis, a conservative flaw size was determined for various locations throughout the RCS primary piping using a minimum detectable leak
~
rate of 10 gpm.
This conservatism provides margins to a plants leakage detection system capability, by a least a factor of 10 while providing a conservative input to t'a fracture mechanics analysis.
' ~ ~
3.2.2.
Postulated Flaw Size Initial flaw size is one of the three fundamental data sets needed to perform a fracture mechanics evaluation of a given component.
The other two are the existing stress field / loads and material properties.
The flaw selected to demonstrate piping integrity is a circumferentially oriented through-wall flaw.
In addition, a longitudinal flaw was considered in BAW-1847, Rev. 1 to represent flaws in the longitudinal seam welds on elbows.
Since all elbow sections are bound by the same nominal size straight sections, only circumferential flaws were evaluated.
The through-wall flaw size was selected to ensure detection by normal plant leakage detection systems and provide a conservative model for fracture mechanics analysis.
A detectable leak rate of 10 gpm has been used for pipes with inside diameters 28 inches or larger to define the through-wall flaw length.
This flaw size is extremely conservative since Technical Specification requirements specify detection of I gpm leakage.
The leakage flaw size was determined using loads for internal pressure plus the axial and bending stresses due to the minimum operating loads.
3.2.3.
Flaw Stability Through a tearing instability analysis, the instability point was determined and the applied J integral evaluated for each pipe size with the maximum loads.
The results are compared with the J at instability to establish safety margins.
3-2 Babcock &Wilcox a McDermott company
3.2.4.
Limit load Analysis For each flawed pipe section, a limit moment was calculated and compared with the applied moment.
In addition, a series of limit moment calculations were darried out to establish the maximum flaw size with which the limit moment becomes equal to the applied moment.
This maximum flaw size (or angle) was compared with the 10 gpm flaw size.
3.3.
LBB Investigation Methodoloav 3.3.1.
Determination of Generic Boundina loads The loads used in both the B&WOG generic and TMI-1 LBB analysis are various combinations of deadweight, thermal, and seismic load cases.
The load cases were obtained from the specific, plant stress reports and represent a plant's piping system loads at the weld locations under a variety of conditions, such as power level and seismic conditions.
The RCS piping loads for the TMI-1 plant were obtained using the same methodology as in the original LBB report and are defined in terms of both force and moment vectors, for the pipe sizes, configurations, load cases, and weld locations specified in Tables 3-1 TMI-1 and 4-1 TMI-1 and Figures 3-1 and 3-2, re.spectively.
Both the initial data sort and the second data sort were made for TMI-1 using the methodology outlined in the original report to obtain the loads and crack lengths, and compared to the B&WOG generic load sets to minimize the number of weld locations that must be considered in the analysis.
An additional data sort was made for THI-1 to ensure bounding load sets were obtained. The sorting process contains the following steps:
(a) Determine the location for maximum loads and then the corresponding minimum loads at that location.
(b)
If there are minimum loads at other locations that are lower than the minimum laads obtained in step (a) then identify those load sets with their corresponding maximums.
lhis method ensures a bounding load set is obtained for a given pipe size, pipe configuration and flaw size.
3-3 Babcock & Wilcox a McDermott company
The results of the leakage flaw size evaluation are provided in Table 3-1 TMI-1.
The maximum loads for the same locations of leakage flaw size is found in Table 4-1 THI-1.
R2.
Fatiaue Flaw Growth Analysis The method of the ASME Section XI, Appendix A, as used in BAW-1847, Rev. 1 is applicable.
Since the THI-I maximum loads are bounded by the generic analysis, the results from Rev.1 of BAW-1847 is also valid for TMI-1.
3.3.3.
Determination of leak Rate Flaw Size The relationship between flaw length and flow rate through the flaw was established using the B&W computer code, KRAKFLO3 which was used in the original analysis.
Details on the applicability and use of this code are provided in the original report.
3.3.4.
RCS Primary Pioino Materials Data In BAW-1847, Rev. 1, base materials and welds in primary piping of the B&W Owners Group plants were categorized.
The results of the material testing program performed for the original B&WOG LBB program on representative carbon steels and welds are documented in report BAW-1889P.4 Locations, material and weld types are shown in Figure 3-3 for a typical B&W designed plant.
The piping materials and field weld data were checked and found to be bound by the materials used in the original LBB analysis.
L3. 5.
Flaw Stability Analysis Considering the operating temperature range and the materials used for fabricating the RCS piping of THI-1 and other B&W0G plants, the fracture mechanics analyses of the flawed piping are in the ductile tearing region.
Therefore, an elastic-plastic fracture mechanics analyses (EPFM) is per-formed.
5 which is a measure of flaw The EPFM analysis is based on the J integra1 driving force in both elastic and elastic plastic range.
This applied J is to be evaluated against the J-R curve which is a material property obtained from compact tension specimen tests.
In BAW-1847, Rev.1, it was demonstrated that the RCS piping elbow sections are bounded by the straight sections with the same nominal diameter, there-1 3-4 Babcock &Wilcox a McDermott company
fore, only straight sections of pipe will be analyzed for tearing instabil-ity.
The EPRI J estimation method 6 for pipes with through-wall flaws is selected for this analysis.
The EPRI method uses the Ramberg-Osgood representation of the material stress-strain relationship.
7 where T is The flaw instability point can be determined by the J-T method the tearing modulus defined as,
-h (3-5)
T-where E is Young's modulus and of is the flow stress.
The tearing modulus is a dimensionless parcL ter.
The condition for a stable crack is that the material's tearing modulus is greater than the tearing modulus obtained from the applied loads.
That is, there:
(3-6)
Tapplied < Tmaterial Evaluation of tearing instability can be illustrated by a plot showing J versus T as shown in Figure 3-4.
The intersection point of the material J-T curve and applied J-T curve is the instability point and the corresponding J value is designated as J from which a critical load value can be instability evaluated.
A safety factor on load is thus defined as the ratio between the critical load and the applied load.
The EPRI method for J integral for through-wall flaws under bending moment M is:
I "+I 2
IM M
a J=fl E + " y 'y g3 (3-7)
C h
where c - R (n -e ) See Figure 3-5 for definition of e a-Re b-Rn 2 (Ia Rf II R
F fl.na y b'Tj le) 1
~
Mo. My cos j y)1 7 sin e-a
(
r a
3-5 Babcock & Wilcox a McDermott company
/
- >?
Mo' - 4a R t f
Mo' is the limit moment of an uncracked cylinder and o y is the yield stress.
The parameters f1 and h1 are dependent on the flaw ratio and the Ramberg-Osgood material constants and n.
The f1 and h1 values were obtained by a series of deformation plasticity finite element analyses and tabulated in reference 6.
L1d limit load Analysis The original RCS piping design analysis included a net section plastic collapse analysis to show that a non-flawed pipe section would not fail by an instability due to the limit load.
In the LBB evaluation, pipes with postulated through-wall flaws are subjected to various loedings, therefore it is prudent to demonstrate that the flawed pipe sections will not fail by plastic instability.
For the LBB analysis, the following limit moment, by Paris and Tada for flawed pipe sections (reference.8), was used; 2
Mp-of (4R t) (cos a - 1/2 sin 0) fa+2af 2 t+
(3-8) where a-P = axial force (kips)
R - mean pipe radius (in.)
t = pipe wall thickness (in.)
p - internal pressure (ksi) f = flow stress (ksi) a - depth of part-through circumferential flaw l
t - pipe thickness 3-6 Babcock & WHcox a McDermott company
Table 3-1.
TMI-1:
Leakaae Flaw Size for Various RCS Pioes Flaw size Minimum (pressure &
Pipe section Weld moment minloads)*
information no.
(ft-kios)
(in.)
36" nominal diameter 1
329.3 11.39 straight section 7
310.9 11.43 inside diameter, 36.5" 1.5 229.1 11.64 thickness, 2.9375" 11 186.9 11.74 36" nominal diameter 1
329.3 12.63 elbow section inside diameter, 36.5" thickness, 3.75" 28" nominal diameter 23 560.0 9.39 straight section 15 508.5 9.53 inside diameter, 28.5" 15.5 479.2 9.61 thickness, 2.375" 14 465.7 9.65 23.5 440.9 9.71 14.5 440.9 9.71 13 260.1 10.23 13.5 241.4 10.30 28" nominal diameter 24.5 1284.2 9.41 elbow section 24 1245.9 9.46 inside diameter, 28.5" 26 962.8 9.92 thickness, 3.125" 18 962.2 9.92 25 932.7 9.97 25.5 903.1 10.02 18.5 891.0 10.05 26.5 887.2
'7.05 17 791.8 10.22 17.5 734.2 10.32 16 644.6 10.55 16.5 597.9 10.71 15 508.5 11.03 15.5 479.2 11.13 14 465.7 11.17 14.5 440.9 11.25 13 260.1 11.72 13.5 241.4 11.77
- Internal pressure - 2150 psi 3-7 Babcock & Wilcox a McDermott company
l Figure 3-1.
177-FA Plant Lower Loop Hot Leg Weld Locations s
Y Z
O O
O O,
o D
~W CD S
O n
t a
^
r 14" 0.D. CORE FLOOD LINE (ROTATED 90 FOR CLARITY)
- RV i
V I
l 3-8 Babcock &WHcox a McDermott company
Figure 3-2.
177-FA Plant Lower Loop Cold Leg Weld Locations 26 28 i
\\
I k i
X d
/
-( ) 0150 g
g-e h
N
'N
(
RV I
'V 12 3-9 Babcock &Wilcox a McDermott company
Y R
I D
4.
C.. 0 T
E 4
7 R
T 0
R E
T A
A 7
R G
T L
T
[
O R.
R U C
O 6 G
- 2. E O8 R
.S.
0 l
l L R
5 E h
1 6
Cl oi O
l Mu L
O S
1 D
o A
8 S
80 n o I
F D
0 O M
S 5
0R MM W
L 7 R n
A S
5 in D
E E T M
E S
N A V MM O
E W
C S
S R L
T N
E N L A
E A I d
"II A
W T L T A R
Y 0
I E
I T S
A e
O 8
W WE N
n R
L f 8
M l
N 0
S g
E I E
1 E
S I
D 0 DR i
O i
R 6
E N
L R A 8 A E C
I s
1 O
A A M E ML E
T I
3 e
T P
I N L L
S U
A T
T I
C S R S I C
D E
B WS C
D O O O f T
S 6
E U R L
L N
W 2
/O 7
[
1 AL 3
S T O E
E o G S
0 ME A
I F
W 5 W M I
W B
SW S
E G
8 n
A 0
L
- O O 0 M 8 Mi T
L M
N
- A 0 A M R
8
[
~
E O
S N
l L
t S
L 9 M E S N S
E L
O a
E C
E i
c N
W o..
i I
A n
l 2
3 4
5 6
12 M
p M
y l8.
/
CC A
T M
D 48 S
L 00 D
[
a E
55 L
W AA 6E n
$, S 1W D
3 i
EI L
l AE s
f 75 Ml C
70 MI l
A 1 2 a
M
((
i M
r r
M
(
j e
I F
ta d
na e
py T
A D
L M
E M
~
(
))
W d
A 8 2 l
D e
- 2. E l.
l EL C C l
W ltCn 8 8 f
80 0 0 A
3 0N 5 5 A
s M
5 A A M
Mh A
M S S M
n MS Av fO SR o
AL A A i
ME
'5 f
SW 7
t s
F 0 a
"r I 2 R
/
( 8 c
E o
A T
M I
L MS RU
/D OB A L ME oi SW MN no 3
MM ij in 3
NM l
)
I I
L L
E 6
N e
.E N3 O
r
- GO3 C
u INCA h
l tI NS I
D g
lGI N
i lR 0)G i
00H1 6N F
N5T 1I f
ID3S. A D2.WE f
UR O
DE E S OlDLPDL fLCEE Y NL A)
RWI 6
8E (El I 6 RE0i Ef 58 5l2Mf 2 08 OA U88A 8 2 CSS 1fS8 (
g.
E O
g M
- k?
=8 d
Figure 3-4.
Illustration of Stability Assessment Program i
Instability Point Instabili ty - - - - -
Unstable Region "c
Staple j
Region T
.5
-i Materi al J-T Appliec J-T Tearing Modulus, T l
3-11 Babcock & WHcox a McDermott company
Figure 3-5.
Illustration of Pipe Geometry Through-wall flaw
/./
+d o
P
=
% t a
i i
Netural Axis N
P
-d o
3-12 Babcock &Wilcox a McDermott company I
7 k;
d l
~'
^'
e 4
4 4.
LBB EVALUATIONS RESULTS FOR THE RCS PRIMARY PIPING This section presents the results of the THI-1 LBB evaluation of the RCS primary piping.
The LBB results were obtained using RCS primary piping N#
bounding loads and material properties for both the base and weld metals.
4.1.
RCS ODeratina Conditio31 For the LBB evaluation, the plant is assumed to be operating under normal full power conditions with a postulated flaw equal to the 10 gpm leakage flaw existing at specific locations in the RCS primary piping.
The SSE loads are then added to the normal operating loads to obtain the resultant moment for use in the fracture mechanics analysis.
Normal system operating conditions used for the LBB analysis were 2150 psi, 600F and RCS flows for the two pipe j
V,.(
sizes of the THI-1 plant. The results are summarized in Table 4-1 for TMI-1.
However, during the search for the maximum and minimum loads, other hot power conditions in addition to 100% power were reviewed to ensure obtaining the bounding loads.
4.2.
Leak Rate Versus Flaw Lenath The leakage flaw length was determined at the location of maximum moment in a given primary size pipe, however, using the minimum moment at that point was used to be conservative.
Since smaller applied moments give larger 10 gpm flaw lengths, additional flaw lengths were calculated at other locations where the minimum moment is less than the minimum moment at the maximum load location.
As scen in Table 4-1 and 4-2 and discussed in section 4-3 and 4.5.1, the location with the maximum moment gave the minimum factor of
~
safety..
This postulated through wall flaw length corresponds to a leak rate of 10 gpm in all cases.
Detailed discussion of the flaw size evaluations is given in the original report.
The data showing the moments and the leakage flaw lengths appear in Table 4-2 THI-1 and Figure 4-1.
4 4-1 Babcock &Wilcox a McDermott company
~
n
,i 4.3.
FaticueFlawGrowthAjdivsis The loads applicable to TMI-bare shown in Table 4-3 and are comparable or less than those considered 'in BAW-1847, Rev. 1.
Therefore, fatigue flaw g
grouth analysis for & postulated inside surface flaw would result in the similar conclusions.
For the 28" ID pipe, any initial flaw depth (circum-ferentiar, flaw) less than sixth-tenths of the pipe wall thickness (0.6t) will not ext $nd through-wall during the design life.
Likewise, any longitudinal flaw whose depth 'is less than 0.4t will not become a through-wall flaw during the design life.
4.4.
RCS Material Properties The results from the B&WOG materials test program defined in section 3.3.4 were used in this analysis. This test was conducted under the B&W QA program for material testing.
The materials in B&WOG plant RCS piping were divided into two major categories: 1) a group of weld metals and 2) base metals i.e. SA106C and SA516.
Tensile, Charpy and compact test specimens were fabricated from materials for a NSS using standard shop fabrication proce-dures and tested at the B&W Alliance Research Center.
The test results for the tension, Charpy and compact fracture data were documented in reference 4.
4.4.1.
Weld Metal Properties A total of 18 compact specimens were tested representing three types of weld metals from six different heats.
The lowest J-R curve is shown in Figure 4-2 represented by a five parameter equation in the following form:
J JR
= -6.51393 (0.085 + A a)-2 + 3687.37 A a + 119 '. 44 where JR is material Jmodified in in-lb/in2 and crack extension Aa in inches.
This J-R curve has a J value of 3940 in-lb/in2 at a maximum crack extension of 0.6788 inch.
The J-R curves for all other weld metals lie above this curve.
The J-R data was obtained at a test temperature of 550F.
Tensile test specimens for the weld metals were machined from the same piece of material used for the compact specimen testing with 0.25 inch diameter.
The tensile specimens were also tested at 550F.
Representative results for the tensile properties of interest are:
4-2 Babcock &Wilcox a McDermott company
o
= 76.0 ksi y
o
= 89.5 ksi u
and af 0.5 (oy + o ) = 82.8 ksi.
=
u While there were son:.c weld metals with lower yield strengths, their toughness were higher than those used in the LBB analysis.
The above choice can be shown to be the cost limiting considering these two competing influences on the J-integral evaluation.
To obtain the Ramberg-Osgood parameters, a true-stress-true strain curve is needed.
From the tensile test records, this curve can be constructed as shoen in Figure 4-3 for the representative weld metal.
The following Ramberg-Osgood equation is fitted to the true stress-true strain curve:
y=*y+a y
( y) ehere o,e - true stress, true strain o
- yield stress y
-o /E cy y
E - Young's modulus a,n
- Ramberg-Osgood material parameters.
The resulting Ramberg-Osgood parameters for the entire strain range of 0 < c
<0.15, where true strain of 0.15 coincides with the maximum load point in the load displacement curve, are:
0.867 a =
n=
14.8.
If one tries to obtain a fit closer to the lower true strain range, it will result in a higher strain hardening parameter n, as is the case in the following base metal discussed in Section 4.4.2.
Since the high range of n (greater than 7) is not defined in the EPRI/GE method, no attempt was made to obtain a better fit in the lower strain range.
e 4.4.2.
Base Metal Properties The test results from four compact test specimens indicate that both base metals, SA106C and SA516, exhibit almost identical J resistance behaviors.
Since the straight pipe sections have the highest stress in RCS piping and 1..
4-3 Babcock &Wilcox a McDermott company
.=
4
are fabricated from SA106, the SA106 properties are chosen for the LBB evaluation.
The J-R data can be expressed in the following equation which shows very good correlation as shown in Figure 4-4.
g JR - -3.12872 (0.06 + aa)-2 + 5172.63 aa + 1146.48 ehere JR is material J modified in in-lb/in2 and aa is crack extension in inches.
This J-R data have a J value of 4280 in lb/in2 at a maximum crack extension of 0.6 inch.
c Representative results for the tensile properties of interest for SA106 are:
y - 39 ksi o
o
- 81 ksi u
and o f - 0.5(oy + "u) - 60 ksi.
For this evaluation the yield strength is used for evaluation of J using the EPRI/GE method.
The flow stress of is used for limit load and tearing modulus determination.
The corresponding true stress-true strain curve is shown in Figure 4-5.
With a least square fit to the entire strain range i.e. O < c <0.15 yields the following Ramberg-Osgood parameters were obtained:
a
- 1.48 l
l n - 5.05 4.5.
Flaw Stability Analysis A discussion of the analysis of the RCS piping elbows and the " Candy Cane" secticn of the hot leg is given in the original LBB report BAW-1847, Rev. 01 section 3.3.5.
They were not re-analyzed in this analysis since the straight 1
sections of piping were shown to be controlling.
To perform a tearing instability analysis, a J versus tearing modulus (J-T) diagram must be generated showing both the material J-T and applied J-T curves for determination of the J at the instability point.
Babcock & Wilcox a McDermott company g
w The EPRI method 6 is used to generate the applied J-T diagrams for tearing instability analysis.
In applying the EPRI method, the applied force including the end cap pressure effect and the applied moment are combined into an equivalent moment by the following equation suggested by reference 9:
9 Meq = M + 0.5 R F /Fb (P) t where M - applied moment R - radius of pipe F - geometry factor for KI in bending b
F - geometry factor for KI in tension, t
P - axial force It was shown in reference 1 that the J f r both weld and base instability 2
metals are greater than a J value of approximately 4000 in-lb/in.
An alternate limit on J, called (JUL), was introduced based on a maximum crack extension from an actual compact test specimen data.
JUL of 3940 and 4280 in-lb/in2 were established for weld and base metals at the maximum crack extensions of 0.77 and 0.6 inches, respectively.
Whenever the applied J is limited by this JUL, tearing stability is assured because J is instability greater than JUL-4.5.1.
Acolied Loads l
The loads given in the original report and the TMI specific maximum loads are shown in Table 4-3.
As seen in Table 4-1 more than one flaw size and moment are given for each pipe size.
However, it was found that the location with maximum moment gave the highest applied J.
Thus only this worst case location is given in Table 4-3.
The final total moments were obtained by adding the equivalent moments from the axial and end cap forces to the maximum momnents from Table 4-3 and are presented in Table 4-4.
4.5.2.
Acolied Moment Versus J for 28-Inch Pioe Figure 4-6 shows the relation between the applied moment and J for the base material metal.
Since it was shown in the original report that the SA 106 base metal is controlling, only the base metal case was analyzed.
The recom-mended safety factor on moment from NUREG-1061 is 1.414.
For the SA 106 base 4-5 Babcock & Wilcox a McDermott company
metal, which is bounding for the weld metal, the value of JUPPER LIMIT is 2
4.280 in-kips /in.
With a factor of safety of 1.414, the applied moment is equal to 57,100 in-kips from Figure 4-5.
The corresponding value of Jtotal 2
is 1.137 in-kips /in, which is well below JUL. At J-JUL, the moment is equal to 78300 in-kips. Therefore, the calculated safety factor on moment is:
Safety Factor = 78300/40362 = 1.94 which is greater than the recommended safety factor of 1.414.
Maximum Flaw Size The recommended safety factor on flaw size is 2.0, as described in reference 2
9.
If the flaw size is increased until J-JUL (4.280 in-kips /in ) at the applied moment of 40362 in-kips, the final flaw size is determined to be 14.65 inches. The calculated safety factor on flaw size is then:
Safety Factor - 14.65/4.7 - 3.12 which is greater than the recommended safety factor of 2.0.
4.5.3.
Acolied Moment Versus J for 36-Inch Pice Figure 4-7 indicates at the maximum applied moment of 64,661 in-kips Jtotal 2
is equal to 0.280 in-kips /in.
With the safety factor of 1.414, the maximum 2
applied moment is 91,400 in-kips which yields a J f 0.656 in-kips /in,
total which is well below J At J-JUL, the moment is equal to 151700 in-kips.
UL.
j Therefore, the safety factor on moment is:
Safety Factor = 151700/64661 - 2.35 which is greater than the recommended safety factor of 1.414.
Maximum Flaw Size 2
If the flaw size is increased until J=JUL (4.193 in-kips /in ) at the applied moment of 64,661 in-kips, the final flaw size becomes 21.25 inches The safety factor on flaw size is:
Safety Factor - 21.25/5.7 = 3.73 which is greater than the recommended safety factor of 2.0, a
4-6 Babcock & Wilcox a McDermott company
4.5.4.
Limit load Analysis 4.5.4.1.
28-Inch Pioe The recommended safety factor on limit load is 3.0, as per reference 9.
From Table 4-5, the limit moment is found to be 105,174 in-kips. Also from Table 4-5, the factor of safety on limit load is found to be 4.3.
This is greater than the recommended value of 3.0; therefore, the limit load criterion is satisfied for the 28-inch diameter pipe.
In addition, a factor of safety on flaw size in the limit load analysis is determined in Table 4-5.
The safety factor is a ratio of the critical flaw size to the 10 GPM leakage flaw size.
The critical flaw is determined as the flaw size which causes the applied moment to be equal to the limit moment of that pipe section with a flaw.
The resulting safety factor, on flaw size is 3.5.
4.5.4.2.
36-Inch Pioe For the 36-inch pipe, the applied moment is 35,352 in-kips and the applied axial force is 348.2 kips.
From Table 4-5, the limit moment is found to be 216,781 in-kips.
The corresponding safety factor on limit load is found to be 6.1, which is greater than 3.0 and thus satisfies the limit load cri-terion.
The safety factor on flaw size in the limit load analysis is determined to be 4.4.
~
Babcock &WHcox a McDermott company f
4.6.
Summary of Safety Marains for TMI-1 RCS Pioino The above analysis can be summarized in the fo'ilowing table:
REQUIRED BAW-1847 (REF. 9)
(REF. 1)
TMI-1 SAFETY FACTOR ON M0 MENT 28" STRAIGHT PIPE 1.414 1.414 1.94 36" STRAIGHT PIPE 1.414
>1.414 2.35 SAFETY FACTOR ON FLAW SIZE 28" STRAIGHT PIPE 2.0 2.2 3.12 36" STRAIGHT PIPE 2.0
>2.0 3.73 LIMIT LOAD ANALYSIS 28" STRAIGHT PIPE
-LIMIT MOM / APP MOM 3.0 2.7 4.3
-CRIT FLAW /10 GPM FLAW 3.5 3.5 36" STRAIGHT PIPE
-LIMIT M0M/ APP M0M 3.0 4.8 6.1
-CRIT FLAW /10 GPM FLAW 6.3 4.4 From this analysis, it is concluded that the Leak-Before-Break concept is applicable to the THI-1 RCS primary piping using the methods described in BAW-1847, Rev. 01.
Babcock & Wilcox a McDermott corr,pany
Table 4-1.
THI-I Maximum Applied Forces, Moments, and Flaw Sizes at Selected locations **
Flaw Axial Maximum Pipe section Weld size force
- Moment information no.
in (kios)
(ft-kios) 36" nominal diameter 1
11.39 348.2 2946.0 straight section 7
11.43 350.7 2654.5 inside diameter, 36.5" 1.5 11.64 348.2 2855.2 thickness, 2.9375" 11 11.74 260.8 1588.0 36" nominal diameter 1
12.63 348.2 2946.0 elbow section inside diameter, 36.5" thickness, 3.75" 28" nominal diameter 23 9.39 426.4 2046.2 5
straight section 15 9.53 126.7 1348.4 I
inside diameter, 28.5" 15.5 9.61 170.2 815.8 thickness, 2.375" 14 9.65 181.8 1326.1 23.5 9.71 388.4 1429.6
--~~
14.5 9.71 170.2 792.4 13 10.23 168.4 1118.7 13.5 10.30 158.0 462.5 28" nominal diameter 24.5 9.41 403.6 1820.2 elbow section 24 9.46 441.1 1642.4 inside diameter, 28.5" 26 9.92 441.1 1715.5 thickness, 3.125" 18 9.92 126.7 1522.0 25 9.97 424.3 1677.8 25.5 10.02 389.9 1158.1 18.5 10.05 132.4 1541.1 26.5 10.05 403.6 1385.3 17 10.22 181.8 1386.2 17.5 10.32 170.2 1248.3 16 10.55 181.8 1308.1 16.5 10.71 170.2 998.4 15 11.03 181.8 1348.4 15.5 11.13 170.2 815.8 14 11.17 181.8 1326.1 14.5 11.25 170.2 792.4 13 11.72 168.4 1118.7 13.5 11.77 158.0 462.5 oInternal pressure is included in the analysis in addition to the axial force o*The first row in each category is the max. moment location.
The remainder was selected based on flaw sizes being larger than that of the next location.
4-9 Babcock &Wilcox a McDermott company
\\
s Table 4-2.
TMI-1 Leakaae Flaw Size for Various RCS Pines Flaw length Minimum (pressure &
Pipe section Weld moment min loads)*
information no.
(ft-kios)
(in.)
36" nominal diameter 1
329.3 11.39 straight section 7
310.9 11.43 inside diameter, 36.5" 1.5 229.1 11.64 thickness, 2.9375" 11 186.9 11.74 36" nominal diameter 1
329.3 12.63 elbow section inside diameter, 36.5" thickness, 3.75" 28" nominal diameter 23 560.0 9.39 straight section 15 508.5 9.53 inside diameter, 28.5" 15.5 479.2 9.61 thickness, 2.375" 14 465.7 5.65 23.5 440.9 9.71 14.5 440.9 9.71 13 260,1 10.23 13.5 241.4 10.30 28" nominal diameter 24.5 1284.2 9.41 elbow section 24 1245.9 9.46 inside diameter, 28.5" 26 962.8 9.92 thickness, 3.125" 18 962.2 9.92 25 932.7 9.97 25.5 903.1 10.02 18.5 891.0 10.05 l
26.5 887.2 10.05 l
17 791.8 10.22 17.5 734.2 10.32 l
16 644.6 10.55 16.5 597.9 10.71 15 508.5 11.03 15.5 479.2 11.13 14 465.7 11.17 14.5 440.9 11.25 13 260.1 11.72 13.5 241.4 11.77 1
cInternal pressure - 2150 psi 4-10 Babcock & Wilcox a McDermott company
Table 4-3.
Maximum Aeolied Loads for Fracture Mechanics Analysis TMI-1 olant specific maximum loads Pipe Maximim Axial Flaw Size Moment Force Size (in,)
Tvoe (ft-kios)
(kios) fin.)
28" Straight 2046.2 426.4 9.39 36" Straight 2946.0 348.2 11.39 BAW-1847. Rev. 1 Loads Pipe Maximim Axial Flaw Size Moment Force Size (in.)
Tvoe (ft-kios)
(kios)
(in.)
28" Straight 3098 250.3 9.2 36" Straight 3952 123.3 8.0 Table 4-4.
Combined Aeolied Loads for TMI-1 Pipe Maximim Axial End Cap Total Equiv.
Total Size Moment Force Force Force Moment Moment (in,)
(in-kio)
(kip)
(kio)
(kip)
(in-k4o) (in-kio) 28" 24554 426.4 1435.4 1862 15807 40362 36" 35352 348.2 2354.3 2703 29309 64661 l
l 4-11 Babcock &Wilcox a McDermott company
Table 4-5.
Limit load Calculation for TMI-128" and 36" Pioes SAFETY FACTORS - LIMIT M0 MENT / APPLIED M0 MENT PIPE RADIUS THICK.
FLAW FORCEX APPL. M SIZE (in)
(in)
SIZE (in)
(kio) lin:_kipl 28 15.4375 2.375 4.7 426.4 24554 36 19.7188 2.938 5.7 348.2 35352 PIPE TH ALPHA LIMIT M SAFETY ita. d_).
Idagl ita41 LM_BAB fin-kio)
FACTOR SIZE a
28 0.30 17.5 0.39 0.774 105174 4.3 36 0.29 26.6 0.37 0.791 216781 6.1 SAFETY FACTORS - MAX FLAW SIZE /10GPM FLAW SIZE (LIMIT MOMENT W/ MAX FLAW = APPLIED MOMENT)
PIPE RADIUS THICK.
FLAW FORCEX APPL. M THETA SJ21 (in)
(in)
SIZE (in)
(kio)
(in-kio)
(dea) 28 15.4375 2.375 4.7 426.4 24554 17.5 36 19.7188 2.938 5.7 348.2 35352 16.6 PIPE MAX THET ALPHA LIMIT M SAFETY j
SIZE (deo) irJdl MBAB fin-kio) Mf_AM FACTOR 28 60.4 0.767 0.286 38802 1.580 3.5 36 73.7 0.865 0.168 46180 1.306 4.4 MATERIAL SA106 FLOW STRESS SIGO =
60 (ksi)
INTERNAL PRESSURE-2.25 (ksi) 1
~
Babcock &Wilcox a McDermott company
Figure 4-1.
Leak Flow Rate Vs Crack Length for 36" ID Straight Pipe Subjected to an Axial Stress of 6.62 ksi and Various Bending Moments Bending Stress (KSI) i 14 20 1 1
5 321 0 ksi_
12 10 R,
f8 a
$6 u.
3 4
2 I
I I
I I
I l
o i
i 0
2 4
6 8
10 12 14 16 Crack Length, Inches 4-13
8 0
o D
6 0
)
l n
a i
t
(
e M
a A
d le W
no r
4 s
i o
f 0
n e
e v
t ru x
C E
R k
-J ca r
2 C
2 4
i 0
e F
rug 0
5 i
F 5
l 1 I I
, I l
p I
I I m
l I
I I
[
e o
T I
l 0
' c 4
3 2
1 0
7%0g s'
.b aElr s OM i
El!=
to O
O
%z t
"o) 3 L
O o
O d
5 z
v
.5 2
4 W d^
O D d h 6H E
8 L
&W N
L Od 4
Eo u.
g g
i I
i I
4 I
I O
O O
O O
O O
O O
O O
O m
O b
O O
4 M
N ism 'SS3 MIS 3n81 4-15 Babcock &Wilcox a McDermott company
0 6
0 5
4 i
0
)
l n
a i
t
(
e a
M A
esa B
n 0
o r
3 i
i o
s f
0 n
e e
v t
r x
u C
E R
k J
ca r
5 C
4 1
F i
4 0
e 0
r 5
ug 5
I i
I I I I I F
I p
I l l
I me 7l l
I I
I I
T I)I I
I l
0 k
5 4
3 2
1 o
q% a.5 E'
?*
aEnR= n EioN f
E ?!= !ix
s.
i (D
O 4
O E
uo E
N o
-K O
m
~
L 0
Q 3
t z
a g
=
O n
3 O w m
a
~
h e
~
e e
T o
m O
m E
a M
YO e
E o
t$
NO 4
O e
L
~
3cn E
O I
i i
i i
i i
a a
O O
O O
O O
O O
O O
O O
m o
b c
c
+
po N
e 1sy 'ss3 sis 3ngi 1
4-17 Babcock & WHcox a McDermott company
0 8
/'
e p
0 i
6 P
h
/
cn x
/
I 8
)
N b
2s e l
ri E
T ot n
f r P
MI
\\
o U
i e
I ao A
R 7
mp PL l
0 l
L rr T
I 4
i E
P m
gP AM P
D a
U E
(
I E
il Da DS I
J T
L te B
P D
N t
A E
P E
nM M
e A
I 0
me 8
LP M
os Ma 2
.B sh Ut i
0 Jw 2
6
-4 e
r ug i
F n0 5
4 5
3 5
2 1
5 O
4 3
2 0
cib,
s(;
>g=$M
! j24
06 1
P
/
04
,/
1
/
/
e p
i 7
0 P
2
/
1 h
/
cn
/
I
/
6 0
3s 0
)
e 1
b ri l
ot fr n
e mp E
o i
ao PL l
rr A
0 l
gP I
T T
8 M
I
/
i PE T
1 a
m il M
(
Da E
J i
\\k t
S I
T t e A
nM 6B R
D N
e E
E E
me 3
P I
0 M
h L
6 0
s os U
M Ma P
D J
P E
.B A
I sh L
Vt P
P i
Jw A
0 4
H 7
u 4
e 0
r 2
ug i
F
\\
0
,m 5
4 5
3 5
5 5
0 4
3 2
0
,{ E vG W*?RR p(= 0M
- 5?!
g24
i, 5.
CONCLUSIONS The results of this analysis demonstrate that all applic-able safety margins recommended by NUREG-1061 are met as were the case for the B&W Owners Group generic analysis (BAW-1847, Rev.1).
Therefore, it is concluded that the leak-before-break approach is justified for the TMI-1 plant and the modified GDC-4 rules are applicable to TMI-1.
t k
s )
e "I
Babcock &Wilcox a McDermott company a
t 6.
REFERENCES 1.
BAW-1847, Revision 1, "The B&W Owners Group Leak-Before-Break Evaluation of Margins against Full Break for RCS Primary Piping of B&W-Designed NSS," September 1985.
2.
Modification of General Design Criterion 4 Requirements for Protection Against Dynamic Effects of Fostulated Pipe Ruptures, NRC 10CFR50, Proposed Rule.
3.
KRAKFLO -- FORTRAN Program for Calculating Flow Through Narrow Cracks, UPGD-TM-38, Babcock & Wilcox, Lynchburg, Virginia, May 1984.
4.
A. L. Lowe, K. K. Yoon, and R. H. Emanuelson, " Piping Material Propeties l
for B&WOG Leak-Before-Break Analysis of the RCS Main Piping," BAW-1889P, October, 1985.
5.
J. Rice, "A Path Independent Integral and the Approximate Analysis of Strain Concentration by Notches and Cracks," Journal of Acolied Mech-anics, June 1968, pp. 379-386.
6.
V. Kumar, et al. " Advances in Elastic-Plastic Fracture Analysis," EPRI NP-3607, August 1984.
7.
P. Paris, Appendix B, NUREG-0744, Vol.1, Rev.1, July 1982.
8.
P.
C.
Paris and H.
Tada, "The Application of Fracture Proof Design Methoda Using Tearing Instability Theory to Nuclear Piping Postulating Circumferential Through-Wall Cracks," NUREG/CR-3464, September,1983.
9.
NUREG-1061, Vol. 3,
" Report of the U.S. NRC Piping Review Committee -
Evaluation of Potential for Pipe Breaks,"
U.S.
Nuclear Regulatory Commission, Nov. 1984.
6-1 Babcock & Wilcox a McDermott company