ML20057E338
| ML20057E338 | |
| Person / Time | |
|---|---|
| Issue date: | 09/30/1993 |
| From: | NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES) |
| To: | |
| References | |
| RTR-REGGD-1.161, TASK-DG-1023, TASK-RE REGGD-01.XXX, REGGD-1.XXX, NUDOCS 9310120037 | |
| Download: ML20057E338 (61) | |
Text
-
U.S. NUCLEAR REGULATORY COMMISS10'4 September 1993
/
o
?,
0FFICE OF NUCLEAR REGULATORY RESEARCH Division 1 ie " M ! ;f Task DG-1023 Io,
,8 DRAFT REGULATORY GUIDE s
Contact:
S. Malik (301) 492-3842 s
i DRAFT REGULATORY GUIDE DG-1023 EVALUATION OF REACTOR PRESSURE VESSELS WITH CHARPY UPPER-SHELF ENERGY LESS THAN SO FT-LB
[
(
I I
I l
l This regulatory guide is being issued in draft form to involse the public in the early stages of the develes-ment of a regulatory Dosition in this arta. 11 has not received complete staff revieg and does not represent an of ficial het staf f position.
Public comments are being solicited on the dra f t guice (including any implementation schedule) and its associ.
ated regulatory ar,alysis or value/ impact statement. Coments should be accomoanied by appropriate si.pporting data. *ritten comments may te submitted to the Regulatory Publications Branch, Dr!PS. Of fice of Admiristra+
tion. U.S. % clear Regulatory Commission Washington. OC 20555. Copies of comments received may be examined
=
at the t4RC Public foc ument Room, 2120 L Stree t NW.. Washington OC.
Comments will be most helpful if received
(
by December 17, 1993.
GeQuests f or single ccples of draf t guides (which may be repeoduced) or for placement on an automatic distri-
[)
bution list for single copies of future draf t guides in specific divisions should be made in writing tp the V
u.s. wctear Pe9ulatory Conmission Washington, oc 20555. Attention: office of Administration, Distribution i
end Mail Services Section.
(
j q 9310120037 930930 QQ66
)Qy I
PDR REGGD O!.XXX.R
_PDR.
TABLE OF CONTENTS s
A.
INTRODUCTION............................
1 8.
DISCUSSION.............................
2 NOMENCLATURE............................
4 C.
REGULATORY POSITION 6
1.
ACCEPTANCE CRITERIA......................
6 1.1 Level A and B Conditions.................
7 1.2 Level C Condition 8
1.3 Level D Condition 9
2.
ANALYSIS METHODS 10 2.1 Level A and B Conditions.................
10 2.2 Level C Condition 15 2.3 Level D Condition 18 l
3.
MATERIAL PROPERTIES......................
19 3.1 Welds Made Using Linde 80 Flux..............
19 3.2 Generic Reactor Pressure Vessel Welds 20 3.3 Reactor Pressure Vessel Base (Plate) Materials......
20 4.
TRANSIENT SELECTION..
23 4.1 Plant-Specific Transients 24 4.2 Bounding Transients 24 D.
IMPLEMENTATION...........................
25 REFERENCES 26 APPENDIX A:
Ex ampl e s o f Me t hod s.................... A-1 APPENDIX B:
Information on Computation of Stress Intensity Factors B-1 REGULATORY ANALYSIS........................... R-1 iii e
1 A.
INTRODUCTION
(
2 Appendix G, " Fracture Toughness Requirements," to 10 CFR Part 50, 3
" Domestic Licensing of Production and Utilization Facilities," requires, in 4
part, that the reactor vessel beltline materials "...must have Charpy upper-5 shelf energy of no less than 75 ft-lb (102J) initially and must maintain 6
upper-shelf energy throughout the life of the vessel of no less than 50 ft-lb 7
(68J), unless it is demonstrated in a manner approved by the Director, Office 8
of Nuclear Reactor Regulation, that lower values of upper-shelf energy will 3
9 provide margins of safety against fracture equivalent to those required by 10 Appendix G of the ASME Code." Charpy upper-shelf energy is defined in Ameri-11 can Society for Testing and Materials' ASTM E 185-79 (Ref. 1) and -82 (Ref.
12 2), which are incorporated by reference in Appendix H, " Reactor Vessel Mate-i 13 rial Surveillance Program Requirements," of 10 CFR Part 50. This guide 14 describes general procedures acceptable to the NRC staff for demonstrating 15 equivalence to the margins of safety in the American Society of Mechanical 16 Engineers' Appendix G of the ASME Code (Ref. 3).
Several examples using these 17 procedures are presented in Appendix A to this guide, and in more detail in
]
18 NUREG/CR-6023 (Ref. 4).
i 19 This draft regulatory guide contains voluntary information collections 20 that are subject to the Paperwork Reduction Act of 1980 (44 U.S.C. 3501 et 1
21 seq.).
This regulatory guide has been submitted to the Office of Management 22 and Budget for review and approval of the information collections.
These i
23 information collections and record keeping are needed for demonstrating 24 compliance with Appendix G to 10 CFR Part 50 for the remaining duration of the 25 plant's license if Charpy upper-shelf energy of the materials in the beltline 26 region may drop, or may have dropped, below the 50 ft-lb regulatory limit.
27 The public reporting burden for this collection of information is j
28 estimated to average 960 hours0.0111 days <br />0.267 hours <br />0.00159 weeks <br />3.6528e-4 months <br /> per response, including the time for reviewing 29 instructions, searching existing data sources, gathering and maintaining the 30 data needed, and completing and reviewing the collection of information. Send 31 comments regarding this burden estimate or any other aspect of this collection 32 of information, including suggestions for further reducing the reporting 33 burden, to the Information and Records Management Branch (MNBB-7714), U.S.
l 34 Nuclear Regulatory Commission, Washington DC 20555-0001; and to the Desk i
q Officer, Office of Information and Regulatory Affairs, NE0B-3019 (3150-0011),
35 36 Office of Management and Budget, Washington, DC 20503.
1 i
1 B.
DISCUSSION 2
The problem of evaluating materials that do not satisfy the 50 ft-lb 3
upper-shelf energy requirement was recognized by the NRC staf f several years 4
ago and was designated Unresolved Safety Issue A-ll, " Reactor Vessel Materials 5
Toughness."
In 1982, the staff completed resolution of USI A-11 by issuing 6
NUREG-0744, " Resolution of the Task A-ll Reactor Vessel Materials Toughness 7
Safety Issue" (Ref. 5), which provided methods for evaluating the fracture 8
behavior of these materials.
Further, Generic Letter 82-26 (Ref. 6) was 9
issued to advise licensees of the USI resolution.
No new requirements were 10 implemented as part of the USI resolution.
However, neither NUREG-0744 nor 11 Generic Letter 82-26 contained criteria for demonstrating equivalence of mar-12 gins with Appendix G of the ASME Code.
Rather, the NRC staff asked Section XI 13 of the ASME Boiler Pressure Vessel Code Committee to develop and suggest to 14 the staff sppropriate criteria.
15 In February 1991, the Chairman of the ASME,Section XI, Subgroup on 16 Evaluation and Standards, provided to the NRC staff the criteria that had been 17 developed by members of the Working Group on Flaw Evaluation (WGFE) and the 18 Working Group on Operating Plant Criteria (WGOPC) (Ref. 7). Although these 19 criteria did not represent ASME Code criteria, they did represent the best 20 opinion of knowledgeable persons familiar with the problem and with the ASME 21 Code.
22 Upon review, the NRC staff found these criteria to be an acceptable 23 method for demonstrating margins of safety equivalent to those in Appendix G 24 of the ASME Code (Ref. 3).
However,specificmetypdsforevaluatingthecri-25 teria still were being developed by the cognizant ASME Code committees.
Fur-26 ther, those efforts were not expected to provide specific guidance on deter-27 mining event sequences and transients to be considered, nor were they expected 28 to provide specific guidance on appropriate material properties.
29 This guide is being developed to provide guidance on evaluating reactor 30 pressure vessels when the Charpy upper-shelf energy falls below the 50 ft-lb 31 limit of Appendix G to 10 CFR Part 50. The analysis methods in the regulatory 32 position are based on methods developed for ASME Code Case N-512 (Ref. 8).
33 The staff has reviewed the analysis methods and finds that they are techni-34 cally acceptable, but the ASME code case is not complete.
It does not provide 35 information on the selection of transients and gives very little detail on 36 material properties selection.
In addition, the code case does not provide 37 specific information on mathematical expressions for calculating the crack i
2
1 driving force (J-integral) for service load Levels C (emergency) and 0 2
(faulted) conditions. This regulatory guide provides guidance on calculating 03 crack driving force for all service load levels, selecting transients for 4
consideration, and appropriate material properties to be used in the analyses.
5 Ductile tearing is the dominant fracture process in the upper-shelf 6
region of the Charpy impact energy versus temperature curve for reactor 7
pressure vessel (RPV) materials.
The conditions governing cleavage mode-8 conversion of ductile tearing process in materials with low Charpy upper-shelf 9
energy are still not well understood, and they are not considered in this 10 draft regulatory guide.
11 The material property needed to characterize ductile tearing in the 12 analysis methods in this draft regulatory guide is the material's J-integral 13 fracture resistance, the J-R curve. This curve is a function of the material, 14 the irradiation condition, the loading rate, and the material temperature.
15 The curve is determined by testing the specific material, under the conditions 16 of interest, in accordance with the American Society for Testing and Mate-17 rials' standard test method in ASTM E 1152-87, " Standard Test Method for 18 Determining J-R Curves" (Ref. 9).
19 Unfortunately, the specific material of interest is seldom available for 20 testing.
Thus, testing programs have used generic materials that are expected 21 to represent the range of actual materials used in fabricating reactor pres-22 sure vessels in the United States. Statistical analyses of these generic data 23 have been performed and reported in NUREG/CR-5729, "Multivariable Modeling of 24 Pressure Vessel and Piping J-R Data" (Ref. 10).
These analyses provide a 25 method for determining the material's J-integral fracture resistance that the 26 NRC staff finds acceptable for use in the methods described in this guide.
27 Other methods for determining the material property may be used on an 28 individual case basis if justified.
29 The statistical analyses reported in Reference 10 addressed a broad range 30 of materials and conditions.
For the purposes of this guide, the NRC staff 31 has concluded that only the ASTM E 1152-87 (Ref. 9) definition of the J-32 integral fracture resistance curve should be used.
This determination 33 requires that a test specimen's net thickness, B, be specified.
Smaller n
34 specimens typically produce more conservative (lower) J-R curves than larger 35 specimens. However, larger specimens are needed to pro ide large amounts of 36 crack growth needed in evaluating certain stability criteria described in 37 Regulatory Position 2 of this regulatory guide.
The NRC staff recommends the 38 test specimen's net-thickness, B, to be 1.0 inch (2.54 cm) for determining n
3 e
}
t I
the J-integral resistance curve using the methods specified in Regulatory 2
Position 3.
This is a reasonable compromise and slightly simplifies the equa-3 tions for the material J-R curve. The neutron fluence attenuation at any 4
depth in the vessel wall (such as near the crack tip) should be determined 4
5 using Regulatory Guide 1.99 (Ref.11).
L 6
NOMENCLATURE i
7 The following terms are used in this regulatory guide and its equations.
s J
The flaw depth, which includes ductile flaw growth (in 8
a 9
inches).
The effective flaw depth, which includes ductile flaw growth 10 a,
11 and a plastic-zone correction (in inches).
1 12 a*
The effective stable flaw depth, which includes ductile flaw 13 growth and a plastic-zone correction (in inches).
The postulated initial flaw depth (in inches).
lf 14 a o I
j 15 B
Net-section thickness of the ASTM E 1152-87 (Ref. 9) test n
16 specimen used in determining material tearing resistance, J-R i
17 curve, behavior (in inches).
18 Cl, C2 Coefficients used in the equation for the material tearing l
19 C3, C4 resistance, J-R curve.
i 20 CR The cooldown rate ( F/ hour).
l 21 CVN Charpy v-notch upper-shelf energy (ft-lb.).
1
(
l 22 E
Young's modulus of elasticity (ksi).
i r
2 i
23 E'
E/(1-v ) (ksi).
l Geometry factors used to calculate the stress intensity 24 F,, F, f3 2
25 factors (dimensionless).
4 i
I L
1 J,,g g The J-integral from the applied loads (in.-lb/in.2).
r s
2 J,, g,t The material's J-integral fracture resistance (in.-lb/in.2),
j 3
J-R curve.
4 J,3 The material's J-integral fracture resistance at a ductile o
5 flaw growth of 0.10 inch ( i n. -l b/ i n.2).
6 K
The mode I stress intensity factor caused by the radial m
7 thermal gradient through the cladding applied to the vessel 8
inner surface, calculated with no plastic zone correction 9
(ksi lin.).
10 K
The mode I stress intensity factor caused by the internal ip 11 pressure, calculated with no plastic-zone correction (ksi 12 li n. ) ; K;p**i*l, K ** are the axial and circumferential ip 13 values, respectively, p 14 Kl, K
calculated with a plastic-zone correction (ksi (in.).
ip U
15 K,,
The mode I stress intensity factor caused by the radial 16 thermal gradient through the vessel wall, calculated with no 17 plastic-zone correction (ksi (in.).
18 K[t K
calculated with a plastic-zone correction (ksi lin.).
it 19 p
Internal pressure (ksi).
20 p,
The maximum accumulation pressure as defined in the plant-21 specific Overpressure Protection Report, but not exceeding 1.1 22 times the design pressure (ksi).
23 R,
The inner radius of the vessel (in inches).
24 SF The safety factor (dirnensionlen).
O
(
25 t
The wall thickness of the vessel's base metal (in inches).
5
1 t'
The sum of the vessel wall thickness, t, and the cladding 2
thickness, tu (in inches).
3 t
The thickness of the stainless steel cladding applied to the u
4 vessel inner surface (in inches).
5 T
Metal temperature, at crack-tip, used in the analysis
( F).
6 o
A reference material's flow stress, specified as 85 ksi in the f
7 ASME,Section XI, Code Case N-512 (Ref. 8) on Charpy upper-8 shelf energy.
9 o,
The material's yield stress (ksi).
Poisson's ratio (dimensionless), specified as 0.3.
10 v
11 C.
REGULATORY POSITION 12 1.
ACCEPTANCE CRITERIA j
13 The following criteria are acceptable to the NRC staff for demonstrating 14 that the margins of safety against ductile fracture are equivalent to those in 15 Appendix G,Section III, of the ASME Code.
Licensees have the option of 16 following the methods in the regulatory guide to determine the equivalent 17 safety margins, or they may use any other methods, procedures, materials data, 18 and transients selection to demonstrate compliance with Appendix G to 10 CFR 19 Part 50.
If licensees choose to follow the regulatory guide, they must use 20 the acceptance criteria, analysis methods, material properties, and transients 21 selection as described in the regulatory guide.
The acceptance criteria are 22 to be satisfied for each category of the transients; namely, service load 23 Levels A and B (normal and upset), Level C (emergency), and Level D (faulted) 24 conditions.
These service load levels are described in Standard Review Plan 25 3.9.3 (Ref. 12).
Because of differences in acceptable outcomes during the 26 various service load levels, different criteria have been developed for Levels 27 A and B, C, and D.
O 6
1 1.1 Level A and B Conditions i
5 2
When the Charpy upper-shelf energy of the base metal is less than 50 ft-3 lb, postulate both axial and circumferential interior flaws and use the tough-4 ness properties for the corresponding orientation.
For a weld with Charpy l
5 upper-shelf energy less than 50 ft-lb, postulate an interior surface flaw 6
oriented along the weld of concern and orient the flaw plane in the radial 7
direction.
Postulate a semi-elliptical surface flaw with an a/t - 0.25 and 8
an aspect ratio of 6-to-1 surface length to flaw depth. A smaller flaw size 9
may be used on an individual case basis if justified. Two criteria must be 10 satisfied as described below. The maximum accumulation pressure, discussed 11 below, is the maximum pressure defined in the Over Pressure Protection Report 12 that satisfies the requirement of Section III, NB-7311(b), of the ASME Code 13 (Ref. 13).
14 1.1.1 The crack driving force must be shown to be less than the material 15 toughness as given by Equation 1:
Jamuw < d.1 (Equation 1) o 17 where J is the J-integral value calculated for the postulated flaw under u
am w 18 pressure and thermal loading, where the assumed pressure is 1.15 times the 19 maximum accumulation pressure, with thermal loading using the plant-specific 20 heatup and cooldown conditions. The parameter J,3 is the J-integral charac-o 21 teristic of the material's resistance to ductile tearing (J,,yi,t), as denoted 22 by a J-R curve test, at a crack extension of 0.1 inch.
23 1.1.2 The flaw must be stable under ductile crack growth as given by 24 Equation 2:
- "d
"'" i ' l (Equation 2)
Ba 6a 25 26 (with load held constant) 27 at a mtied
- material i
28 l
7
1 1
where J is calculated for the postulated flaw under pressure and thermal 2
loading for all service level A and B conditions when the assumed pressure is 3
1.25 times the maximum accumulation pressure, with thermal loading, as defined 4
above.
The material's J-integral fracture resistance should represent a con-5 servative estimate of the data for the vessel material under evaluation (i.e.,
6 mean - 2 standard deviations).
Methods for determining the J-integral frac-7 ture resistance, J-R curve, are discussed in Regulatory Position 3 this guide.
8 Methods for determining the appropriate 'rvice level conditions are discussed 9
in Regulatory Position 4 of this guide.
10 1.2 Level C Condition 11 When the Charpy upper-shelf energy of the base metal is less than 50 12 ft-lb, postulate both axial and circumferential interior flaws and use the 13 toughness properties for the corresponding orientation.
When the Charpy 14 upper-shelf energy of any weld material is less than 50 ft-lb, postulate an 15 interior surface flaw with its major axis oriented along the weld of concern 16 and the flaw plane oriented in the radial direction.
Consider postulated 17 surface flaws with depths up to one-tenth the base metal wall thickness, plus 18 the clad thickness, but with the total depth not to exceed 1.0 inch (2.54 cm) 19 and with an aspect ratio of 6-10-1 surf ace length to flaw depth.
A smaller 20 maximum flaw depth may be used on an individual case basis if justified.
For 21 these evaluations, two criteria must be satisfied.
22 1.2.]
The crack driving force must be shown to be less than the materi::1 23 toughness as given by Equation 3:
J
<J.,
(Equation 3) e%
o 24 25 where J,g is the J-integral value calculated for the postulated flaw in 26 the beltline region of the reactor vessel under the governing Service Level C 27 condition, with a saf ety f actor of 1.0 on the applied loading.
J is the J-ag 28 integral characteristic of the material resistance to ductile tearing 29 (J
ga), as denoted bj a J-R curve test, at a crack extension of 0.1 inch.
30 1.2.2 lhe flaw must also be stable under ductile crack growth as given 11 hy ref tinn a.
8
N N
am t ied materiat
(
(Equation 4)
\\
da Ba w J' 1
j 2
(with load held constant) 3 at applied meterial 5
where J,tig is calculated for the postulated flaw under the governing 6
Service Level C condition, with a safety factor of 1.0 on the applied loading.
7 The material's J-integral fracture resistance should represent a conservative 8
estimate of the data for the vessel material under evaluation (i.e., mean - 2 9
standard deviations).
The J-integral resistance versus crack growth, J-R 10 curve, is defined in Regulatory Position 3 of this guide.
Detr.rmination of 11 the appropriate service level conditions are discussed in Regulatory Position 12 4 of this guide.
13 1.3 Level D Condition
,m i
/
's 14 When the Charpy upper-shelf energy of the base metal is less than 50 15 f t-lb, postulate both axial and circumferential interior flaws and use the 16 toughness properties for the corresponding orientation.
When the Charpy 17 upper-shelf energy of any weld material is less than 50 ft-lb, postulate an 18 interior semi-elliptic surface flaw with the major axis oriented along the 19 weld of concern and the flaw plane oriented in the radial direction.
Consider 20 postulated surface flaws with depths up to one-tenth the base metal wall 21 thickness, plus the clad thickness, but with total depth not to exceed 1.0 22 inch (2.54 cm) and with an aspect ratio of 6-to-1 surface length to flaw 23 depth.
A smaller maximum flaw depth may be used on an individual case basis 24 if justified.
25 For these evaluations, the postulated flaw must be stable under ductile 26 crack growth as given by Equation 5:
M N
amtied materiet (Equation 5)
Sa Sa 27 28 (with load held constant) 9
1 at d
- d,at er i e t amt ied 3
where J,,% is calculated for the postulated flaw under the governing 4
Service level D condition, with a safety factor of 1.0 on the applied loading.
5 Additionally, the flaw depth, including stable tearing, should not be greater 6
than 75% of the vessel wall thickness and the remaining ligament should be 7
safe from tensile instability.
8 The material's J-integral fracture resistance should reflect the mean 9
value of the data representative of the vessel material under evaluation. The 10 J-integral resistance versus crack growth, J-R curve, is discussed in Regula-11 tory Position 3 of this guide. Methods for determining the appropriate ser-12 vice level conditions are discussed in Regulatory Position 4 of this guide.
13 2.
ANALYSl$ METHODS 14 The analysis methods described in this guide are acceptable to the NRC 15 staff for evaluating the criteria described above. Other methods may be used 16 on an individual case basis if justified.
17 2.1 Level A and B Conditions 18 The acceptance criteria discussed in Regulatory Position 1.1 for Level A 19 and B conditions involve a comparison of the applied J-integral to the mate-20 rial's J-integral fracture resistance at a ductile flaw extension of 0.1 inch 21 and a determination that this flaw would be stable under the applied loading.
22 Procedures are detailed below for (1) calculating the applied J-integral for 23 the Service Levels A and 8 flaws and loading conditions and (2) determining 24 that the slope of the material's J-integral resistance curve is greater than 25 the slope of the applied J-integral versus crack depth curve at the equili-26 brium point on the J-R curve where the two curves intersect, as illustrated in 27 figure 1.
28 2.1.1 Calculation of the Applied J-Intearal 29 The calculation of the applied J-integral consists of two steps:
Step 1 30 is to calculate the effective flaw depth, which includes a plastic-zone 10
./ -
A Material J-R Curve h
E en B
.5 Applied J
Evaluation Point Crack Extension, Aa FIGURE 1:
Comparison of the Slope of the Applied J-Integral and J-R Curve.
O 11
1 correction; and Step 2 is to calculate the J-integral for smali-scale yielding 2
based on this effective flaw depth.
O 3
Step 1 4
For an axial flaw with depth 'a' equal to (0.25t + 0.1 in.), calculate 5
the stress intensity factor from internal pressure, p,,
with a safety factor, 6
SF, on pressure equal to 1.15, using Equation 6:
)
K[p*i'
= (SF) p, [1 + (R, /t)] (ra)0 5 f, (Equation 6) 7 F
= 0.982 + 1.006 (a/t)2 3
8 4
9 This equation for K ""i is applicable to 0.05 <; a/t s 0.50, and it includes ip 10 the effect of pressure acting on the flaw faces.
11 For a circumferential flaw with depth 'a' equal to (0.25t + 0.1 in.),
f 12 calculate the stress intensity factor from internal pressure, p, with a 13 safety factor, SF, on pressure equal to 1.15, using Equation 7:
K[p
= (ST) p, [1 + (R, /(2t)] (wa) 5 F (Equation 7)
^
2 T
= 0.885- 0.233 (a/t) - 0.345 (a/t)2 2
~
15 16 This equation for K "" is applicable to 0.05 5 a/t s 0.50, and it includes ip 17 the effect of pressure acting on the flaw faces.
18 For an axial or circumferential flaw with depth 'a' equal to (0.25t + 0.1 19 in.), the " steady-state" (time independent) stress intensity factor from 20 radial thermal gradients is obtained by using Equation 8:
K
= ((CR)/1000) t 5 F (Equation 8) 2 n
3 21 f
= 0.584 + 2.647 (a/ t ) - 6.294 (a/ t)2 + 2.990(a/t)5 3
22 23 This equation for K is valid for 0.2 s a/t s 0.50, and 0 s CR s 100 F/ hour.
n 24 This equation does not include the contribution to K,, from the claduing
{
25 thickness, t If the steady-state values of thermally induced K are used, u.
n l
26 the material J-R curve should correspond to the temperature at the beginning 12
I of the transient, when a uniformly high temperature is present across the m
)
2 vessel wall thickness, leading to the lowest J-R curve.
M 3
Calculate the effective flaw depth for small-scale yielding, a,, using 4
Equation 9:
n)J a, = a
( -- ) [ (K
- K 1
iP 2
(Equation 9) 6n o y 5
6 Sten 2 7
For an axial flaw, calculate the stress intensity factor from internal 8
pressureforsmall-scaleyielding,K[p, by substituting a, in place of 'a' in 9
Equation 6, including the equation for F.
For a circumferential flaw, cal-r 3
10 culateK[p by substituting a, in place of 'a' in Equation 7, including the 11 equation for F.
For an axial or circumferential flaw, calculate the stress 2
12 intensity factor from the radial thermal gradients for small-scale yielding 13 K[1, by substituting a, in place of 'a' in Equation 8, including the equation 14 for F.
3 15 The J-integral f rom the applied loads for small-scale yielding is given
/
16 by Equation 10:
V g = 10 0 0 ( K 'p + K ', ) 2 /r/
(Equation 10)
J i
3 18 Alternatively, in place of the steady-state Equation 8, a thermal 19 transient stress analysis may be performed for the limiting cooldown rate, 20 including the contributions of cladding to thermal stress and the thermal 21 stress intensity factor for this alternative analysis inethod (developed in 22 Ref. 4), the main features for computing K and Km, which are applied in n
23 examples in Appendix A, are given in Appendix B.
The equations providcd in 24 Appendix B may be used if the transient temperature history can be approxi-25 mated adequately by either an exponential or a polynomial equation.
If not, a 26 more rigorous approach should be used. The computer code given in Appendix B 27 is for general illustration.
Licensees assume responsibility for the 28 correctness of the computer code they use.
29 The limiting condition should be determined for the transient time at 30 which the material's J-R curve will be greater than or equal to the J g
for evaluating Equations 1 and l The main steps are:
o ) 31
- a 13
1 a.
Determine the temperature gradient across the vessel wall thickness, 2
in 10 to 20 time steps over the full duration of the transient, and 3
compute the corresponding thermal stress history, taking into 4
account the cladding thickness, t u.
5 b.
For each time step, compute K and K values as a function of the it m
6 crack depth in the range 0.05 5 a/t' s 0.5.
7 c.
For Equation 1, calculate the pressure-induced K and the J g
gpu w '
8 using Equations 9 and 10, at u : rack-tip depth of (0.25t' + 0.1 in.)
9 for each time step.
10 d.
Use Step a to find crack-tip temperature history at each time step.
11 See figure A-1, in Appendix A, for an example.
12 e.
For a given material condition, determine the J-R values at the 13 crack extension of 0.1 inch by using the crack-tip temperature 14 history from Step d.
See figure A-2, in Appendix A, for an example.
15 f.
Compare the material's J-R values as a function of time in Step e, 16 with the J,7,,4,3 val m in Step c.
See Figure A-2, in Appendix A, 17 for an example.
The time at which the J-R value is just equal to 18 the J determines the critical condition for evaluating ouw 19 Equation 1.
20 g.
At the time determined in Step f, evaluate Equation 2 to verify the 21 stability of predicted flaw growth.
22 2.1.2 Evaluation of Flaw Stability 23 Flaw stability is evaluated by a direct application of the flaw stability 24 criterion given by Equation 2.
The applied J-integral is calculated for a 25 series of flaw depths corresponding to iricreasing amounts of ductile flaw 26 growth.
The applied pressure, p, is set equal to the maximum accumulated 27 pressure for Service Levels A and B conditions, p,, with a safety factor, SF, 28 equal to 1.25.
The applied J-integral for Service levels A and B conditions 29 may be calculated using Equations 6 through 10.
Each pair of the applied J-30 integral and flaw depth are plotted on a crack driving force diagram to pro-31 duce the applied J-integral curve as illustrated in Figure 1.
The material's 32 J-R curve also is plotted on the crack driving force diagram.
Flaw stability 33 at a given applied load is demonstrated if the slope of the applied J-integral 34 curve is less than the slope of the material's J-R curve at the equilibrium 35 point on the J-R curve where the two curves intersect.
O 14
1 2.2 Level C Condition
/^A 2
The acceptance criteria discussed in Regulatory Position 1 for Service 3
Level C conditions are similar to those for Service Levels A and B, with the 4
exceptions of the crack size to be considered and the scfety factor applied to 5
the pressure loading.
For Service Level C conditions, flaw sizes up to one-6 tenth the base metal wall thickness, plus the clad thickness tu, but with a 7
total depth not to exceed 1.0 inch (2.54 cm), are to be considered. A safety 8
factor of 1.0 is used for both pressure and thermal loading. As with the 9
Service Level A ind B criteria, for Service Level C it must be demonstrated 10 that the applied v is less than the material's fracture resistance at a crack 11 extension of 0.1 inch, and that the flaw must be stable under the applied 12 loading.
13 Procedures ?re described below for (1) determining the applied J-integral 14 for Service Level C flaw and loading conditions and (2) determining that the 15 slope of the material's J-integral fracture resistance, J-R curve, is greater 16 than the slope of the applied J-integral versus crack depth curve.
17 2.2.1 Calculation of the Applied J-Intearal J
\\
18 The calculation of the applied J-integral consists of two steps:
Step 1 19 is to calculate the effective flaw depth, which includes a plastic-zone cor-20 rection, and Step 2 is to calculate the J-integral for small-scale yielding 21 based on this effective flaw depth.
22 St ep__1 23 Postulate a series of flaws with depths ranging up to cladding thickness 24 plus 0.1 times the base metal wall thickness, but not exceeding 1.0 inch (2.54 25 cm).
The number of flaws and specific flaw sizes to be postulated should be 26 sufficient to determine the peak value of the applied J-integral over this 27 size range.
For each of these postulated flaws, the analysis flaw size "a"
28 should be the sum of the postulated flaw size plus 0.1 inch ductile crack 29 extension.
For axial flaws, at each analysis flaw size, calculate the stress 30 intensity factor arising from internal pressure, p, with a safety factor, SF, 31 on internal pressure equal to 1.0, using Equation 11:
(
15
f l
1 I
K,*p'^ * ' = (ST) p, [l +(R,/t ') ] ( n a) 5 f, (Equation 11) 9 l
T
= 0.982 +1.006 ( a / t ')2 ;
with 0.05 s a / t' s 0.5 3
2 4
l 3
For circumferential flaws, at each analysis flaw size calculate the 4
stress intensity f actor arising from internal pressure, p,, with a safety f
5 factor, SF, on pressure equal to 1.0, us:w Equation 12:
6 K 'p" " = (SF) p, [1 -(R,/t ') ] ( n a) 5 f, (Equation 12)
]
3 I
i T
= 0.885 +0.233 ( a / t # ) +0.345 ( a / t ')2 2
7 i.
l 8
These equations for K,p'i""-
are valid for 0.05 s a/t' s 0.5, and include the 9
effect of pressure acting on the flaw faces.
10 if it can be demonstrated that the actual cooldown rate could be bounded 1
11 by a " constant" cooldown rate, for each crack depth the stress intensity fac-l 12 tor arising from a radial thermal gradient under a constant cooldown rate, f
13 including cladding effects, described in Example 4 of Appendix A is given by 14 Equation 13:
i n = [-3.04979 +134.9688 ( # ) +0.20287 (CR) +0.0lll66 (CR) ( # )
K t'
t' (Equation
-442.6721 ( # )2 +388.9575 ( #t ' )3-0. 000146 (CR)2 t'
15 16 This equation is applicable to 0.05 s a/t' s 0.5, and 100 s CR s 600 F/ hour.
i i
17 lhe CR values less than 100 F/ hour are covered under Service Levels A and B 18 (see Equation 8).
The cladding thickness t
= 5/16 in., R, = 86.875 in.,
u j
19 base metal thickness t = 8.625 in., and R,/t' ratio = 9.72.
Details of the 20 analysis results are given in Example 4 in Appendix A.
21 Calculate the ef fective flaw depth for small-scale yielding, a,, using 22 Equation 14:
i a, = a + ( - ) [ ( K n ) ]2 (Equation 14) 1
- K P
6n o
j y
j 23 1
16 l
3
1 Step _2 2
for each flaw size considered, calculate the stress intensity factor 3
arisingfrominternalpressureforsmall-scaleyielding,Klp, by substituting 4
a, in place of "a" in Equation 11 for the axial flaws and in Equation 12 for 5
the circumferential flaws.
Similarly, calculate the stress intensity factor 6
arising from radial thermal gradients for small-scale yielding, K[1, by 7
substituting a, in place of "a" in Equation 13.
The J-integral arising from 8
the applied loads for small-scale yielding is given by Equation 15:
ong = 1000 (K[p-K[)2 fr/
(Equation 15)
J 9
10 in an actual transient the cooldown rate initially may vary significantly 11 with time.
Therefore, transient-specific peak thermal stress-induced K,1 and 12 K,u computations may be necessary. If so, in place of Equation 13, a thermal 13 transient stress analysis may be performed for the specific transient, includ-14 ing the contributions of cladding to thermal stress and the stress intensity 15 factor.
For this alternative analysis method, the main features for computing 16 K
and Km, which are applied on examples in Appendix A, are given in Appen-n 17 dix B.
The method provided in Appendix B may be used if the transient tem-18 perature history can be approximated adequately by either an exponential or a 19 polynomial equation.
If not, a more rigorous approach should be used.
The 20 computer code given in Appendix B is for general illustration.
Licensees 21 assume responsibility for the correctness of the computer code they use.
22 The limiting condition should be determined for the transient time at 23 which the material's resistance (J-R curve) will be greater than or equal to 24 the J for evaluating Equations 1 and 2.
The main steps are:
um 25 a.
Determine the temperature gradient across the vessel wall thickness, 26 in 10 to 20 time steps over the full duration of the transient, and 27 corrpute the corresponding thermal stress history, taking into 28 account the cladding thickness, t u.
29 b.
For each time step, corrpute K and K,u values as a function of the n
30 crack depth in the range 0.05 5 a/t' s 0.5.
31 c.
For Equation 1, calculate the pressure-induced K and the J ip ouw.
32 using Equations 14 and 15, at a crack-tip depth of {(0.1t 4 t 4
u 33 0.1 in.) 5 1 in.) for each time step.
34 d.
Use Step a to find crack-tip temperature history at each time step.
35 See figure A-1, in Appendix A, for an exarcple.
17 l
l
For a given material condition, determine the J-R values at the 1
e.
2 crack extension of 0.1 inch by using the crack-tip temperature 3
history from Step d.
See Figure A-2, in Appendix A, for an example.
4 f.
Compare the material's J-R values as a function of time in Step e, 5
with the J values in Step c.
See Figure A-2, in Appendix A, wuw 6
for an example.
The time at which the J-R value is just equal to 1
7 the J determines the critical condition for evaluating wuw 8
Equation 1.
l 9
g.
At the time determined in Step f, evaluate Equation 2 to verify the 10 stability of predicted flaw growth.
l i
11 2.2.2 Evaluation of Flaw Stability 12 Flaw stability is evaluated by a direct application of the flaw stability J
13 criterion given by Equati a 4 in Regulatory Position 1.2.2.
The applied J-14 integral is calculated for a series of flaw depths corresponding to increasing l
15 amounts of ductile flaw growth. The applied pressure, p, is set equal to the 16 peak pressure for the Service Level C transient under consideration with a 17 safety factor, SF, equal to 1.0.
The applied J-integral for Service Level C 18 conditions may be calculated using Equations 11 through 15.
Each pair of the 19 applied J-integral and flaw depth are plotted on a crack driving force diagram 20 to produce the applied J-integral curve as illustrated in Figure 1.
The mate-21 rial's J-R curve also is plotted on the crack driving force diagram and inter-22 sects the abscissa at the initial flaw depth, a,.
Flaw stability at a given l
23 applied load is demonstrated if the slope of the applied J-integral curve is 24 less than the slope of the material's J-R curve at the equilibrium point on 25 the J-R curve where the two curves intersect.
26 2.3 Level D Condition 27 The acceptance criteria discussed in Regulatory Position 1 for Level D 28 service conditions involve only the stability of the postulated flaws. Addi-29 tionally, the stable flaw depth must not exceed 75% of the vessel wall thick-30 ness, and the remaining ligament must be safe from the tensile instability.
31 Stability of the ductile crack extension is demonstrated for Service 32 Level D in the same manner as for Service Level C.
However, the material 33 properties should represent only the best estimate (i.e., mean value) of the 34 J-R curve for the vessel material under evaluation.
18 1
1
1 Tensile stability of the remaining ligament is conservatively 2
demonstrated if Equation 16 is satisfied.
O o, > p (R, + a ") / ( t - a,")
(Equation 16) 3 4
3.
MATERIAL PROPERTIES 5
This guide provides methods for determin%g the J-integral fracture 6
resistance of three classes of materials: welds manufactured with Linde 80 7
welding flux, generic welds used in fabricating reactor pressure vessels, and 8
plate materials (low and high toughness). The J-R curves for plant-specific l
9 materials may be used on an individual case basis if justified.
Otherwise, I
10 the material's J-integral fracture resistance may be determined from Equation I
11 17, developed in Reference 10:
f 4 = (SF).( C1 (M)cz exp[C3 (M)"])
(Equation 17) l 12 13 The coefficients in Equation 17 for each material type are discussed below.
14 As noted earlier, the net-section thickness, B, of ASTM E 1152-87 (Ref. 9) n 15 compact-tension (CT) specimens to be considared is snacified as 1 inch.
In 16 addition to the Charpy (CVN) models discussed in this guide, Reference 10 con-17 tains two other models, namely the Copper-Fluence (Cu-(t) models and the pre-18 irradiation Charpy (CVN ) models, which may be used to determine the p
19 material's J-R curves.
l 20 3.1 Welds Made Usina Linde 80 Flux 21 For analyses addressing Service Levels A, B, and C, a conservative 22 representation of the J-R curve is obtained by setting SF - 0.648.
For 23 analyses addressing Service Level D, set SF - 1.0.
C1 = exp [-3.67 + 1.45 ln (CVN) - 0.00308T]
(Equation 18)
C2 = 0.077 + 0.116 lnC1 (Equation 19) 25 19
C3 = -0.0812 - 0.0092 ln C1 (Equation 20) 1 C4 = -0. 5 (Equation 21) 2 3
3.2 Generic Reactor Pressure Vessel Welds 4
for analyses addressing Service Levels A, B, and C, a conservative 5
representation of the J-R curve is obtained by setting SF = 0.629.
For 6
analyses addressing Service Level D, set SF = 1.0.
C1 = exp [-4.12 + 1.49 ln (CVN)- 0.00249 7]
(Equation 22)
C2 = 0.077 + 0.116 in C1 (Equation 23) 8 C3 = -0. 0812 - 0.0092 ln C1 (Equation 24) 9 C4 = -0. 5 (Equation 25) 10 11 3.3 Reactor Pressure Vessel Base (Plate) Materials 12 The elastic-plastic fracture toughness of plate materials may be rela-13 tively high or quite low, depending on a variety of chemical, metallurgical, 14 and thermo-mechanical processing varir.bles.
The statistical analyses reported 15 in Reference 10 included only materials that exhibited a J-R curve with a sig-16 nificantly rising slope, i.e., the higher toughness materials.
However, test 17 results reported in NUREG/CR-5265, " Size Effects on J-R Curves for A-302B 18 Plate" (Ref.14), clearly show J-R curves with very little, if any, increase 19 in slope.
References 14, 15, and 16 provide some insight into the nature of 20 the low toughness issue for the plate materials.
While there are several 21 variables that influence the fracture toughness, sulphur content seems to be a 22 reasonable indicator of the plate toughness, with a " higher" sulphur content 23 indicating " lower" fracture toughness (Ref.16).
A sulphur content of 0.018 24 weight-percent is a good demarkation for high-and low-toughness values.
25 Because of the low-toughness plate issue, and because of the relatively 26 sparse data base that could be used to estimate the fracture toughness for 27 these materials, a f racture toughness model is only provided for high-28 toughness plate materials.
If the sulphur content of the plate is less than 20
1 0.018 weight-percent, the plate models described in Reference 10 may be used.
2 However, if the sulphur content is greater than or equal to 0.018 weight-3 percent, justification should be provided for use of the models in Refer-4 ence 10.
Factors that might justify use of these high-toughness models could 5
include information about the year of manufacture of the plate and any special 6
thermo-mechanical processing that would serve to improve the fracture tough-7 ness of the plate.
If adequate justification cannot be provided, a low-8 toughness plate model should be developed and used.
9 The CVN value should be for the proper orientation of the plate material 10 (see Figure 2).
For example, for axial flaws the CVN value for L-T (strong) 11 orientation in the vessel wall should be used.
Similarly, for circumferential 12 flaws the CVN value for T-L (weak) orientation should be used.
In many cases, 13 the CVN values for both orientations may not be known.
If the CVN value for 14 the T-L (weak) orientation is not available, the L-T (strong) orientation CVN 15 value may be multiplied by a factor of 0.65 (Ref.17) to obtain the CVN value 16 for the T-L (weak) orientation.
However, if the CVN value for T-L (weak) 17 orientation is known and the L-T (strong) orientation is to be estimated, the 18 CVN value for the L-T (strcng) orientation is assumed to be same as that of 19 T-L (weak) orientation.
20 3.3.1 High-Toughness Model (S <0.018 Weight-Percent) 21 For plate material with sulphur content greater than 0.018 weight-22 percent, use of this model should be justified, as discussed above.
23 For analyses addressing Service Levels A, B, and C, a conservative repre-24 sentation of the J-R curve is obtained by setting SF = 0.749.
For analyses 25 addressing Service Level D, set SF = 1.0.
C1 = exp [-2.44 - 1.13 in (CVN) - 0.00277 7 )
(Equation 26)
C2 = 0.077 + 0.116 InC1 (Equation 27) 27 C3 = -0.0812 - 0.0092 in C1 (Equation 28) 28 C4 = -0. 4 0; (Equation 29) 29 30 3.3.2 Low-Toughness Plate (S > 0.018 Weight-Percent}
31 W analyses addressing materials with sulphur content greater than 32 0.018 weight-percent, the J-R curve data are scarce.
Very limited J-R data 21
t O
//
N L
O
[
O S
I A
N T
N W
C I
T O
D A
L E
U TL M
T V
I R
- F T
TL S
P A
R I
A D
I L
GMA na N TI T
N G
OS X n
i E
N LA A s
k/4)/l e
n O
E V
o I
R R
M P
i t
O T
S R
a t
S A
n M
ir O
e T
wa S
l F
A M
TS D
O A
N dn A
a E
M E
S A
M e
h S
t A
fo N
E n
F O
S o
0 I
R W
! /
i t
T i
E A
n C
VLL I.
i N
f L
E S T T
F e
O R
D a
,,i U
I I
N M
/
T D
AMC TS R
A 2
RTI I
TSC E
N K
A R
E F
A V
I G
M P
I F
E E
S R
D_
W A
t<e O
!l l
I
\\
I for a 6-inch-thick compact tension specimen (ASTM 6T CT at 180 F temperature) 2 from an A-302B plate in T-L (weak) orientation, available in NUREG/CR-5265 3
(Ref.14), may be used with adjustments for the specimen temperature and CVN 4
value (Ref. 18), or a material-specific justification should be provided to 5
support the use of other data.
For analyses addressing Service Levels A, B, 6
and C, a lower bound representation (mean - 2a) of the J-R curve should be 7
used.
For analyses addressing Service Level D, the mean value of the J-R 8
curve should be used.
9 Additional J-R curve test data for the low-toughness, A3028, plate 10 material are presently being generated.
Regulatory guidance will be updated, I
11 if justified, based on the results obtained from the test data collected for 12 the J-R curve in low-toughness plate material.
13 4.
TRANSIENT SELECTION 14 Selection of the limiting transients for Service Levels C and D is a key 15 aspect of evaluating the integrity of reactor pressure vessels that contain 16 materials with Charpy upper-shelf energy less than 50 ft-lb.
Generally, Ser-17 vice Levels A and B are limiting. However, there may be plant-specific con-18 siderations that make Service Levels C or D controlling for ductile fracture.
19 To provide reasonable assurance that the limiting service loading 20 conditions have been identified, either of two approaches may be used: a 21 plant-specific transient evaluation or a generic bounding analysis.
It should 22 be noted that plants may be grouped and limiting transients for these groups 23 determined.
The plant-specific transie:.t evaluation is the preferred 24 approach.
However, since some licensees may not have the specific transient 25 information needed for this analysis, a conservative " bounding" analysis may 26 be performed for each service level.
Specific guidance for each of these 27 approaches is provided below.
28 As described in the Discussion section of this guide, ductile tearing is 29 the dominant fracture process in the upper-shelf region. and the possibility 30 of mode-conversion to cleavage (brittle) fracture is not considered in this 31 draft regulatory guide.
The analyses using these bounding transients need 32 only address the transient from its beginning to the time at which the metal 33 at the tip of the flaw being analyzed reaches a temperature equivalent to the 34 adjusted RT plus 50'f.
In this regulatory guide, an adjusted RT,7 plus c7 35 50*f (which typically represents the low-temperature overpressure protection 23
I system's enabling temperature) is taken as the lower temperature limit for 2
upper-shelf behavior.
3 Although ATWS (anticipated transient without scram) is not a design basis 4
transient, it must be considered in evaluating low upper-shelf energy mate-5 rials for compliance with Appendix G to 10 CfR Part 50. ATWS should be con-6 sidered as a Service Level C transient, or a plant-specific justification 7
should be provided for its consideration at another service load level.
8 4.1 Plant-Specific Transients 9
To provide reasonable assurance that the limiting service loading condi-10 tions have been identified on a plant-specific basis, the Service Level C and 11 D design transients and events that are necessary to demonstrate compliance 12 with Standard Review Plan 3.9.3 (Ref. 12) should be used.
13 When this transient list is not available or is incomplete, the most 14 complete list of transients for these service levels that is available for 15 similar plant designs should be used.
Typically, the most complete list of 16 transients would be available for the more recently built plants from a par-17 ticular vendor.
This list should be reviewed, and the limiting transients for 18 the reactor vessel being analyzed should be defined. Once the transients are l
19 defined, system-level thermal-hydraulic analyses should be performed to deter--
20 mine the limiting pressure-temperature-time history for each transient being 21 considered.
This history provides the input to the analyses described in this 22 guide.
23 4.2 Boundina Transients 24 When the plant-specific transients are not available or when developing 25 or updating the pressure-temperature-time history would be an undue burden, a 26 conservative " bounding" pressure-temperature-time history may be used.
This 27 history should anticipate a pressure equal to the shut-off head for the high-l 28 pressure injection system and a cooldown rate of 400 F per hour for Service 29 Level C and 600 f per hour for Service Level D.
These values are based on the 30 NRC staff's experience in performing the bounding analyses (for examples, see 31 Appendix A of this draf t regulatory guide and Ref. 4).
Alternatives to these 32 cooldown rates may be used if justified by the plant-specific safety-injection 33 flows and temperatures.
1 24
l 1
D.
IMPLEMENTATION 2
The purpose of this section is to provide information to applicants and 3
licensees regarding the NRC staff's plans for using this regulatory guide.
4 This draft regulatory guide has been released to encourage public 5
participation in its development.
Except in those cases in which an applicant 6
proposes an acceptable alternative method for complying with specified por-7 tions of the Commission's regulations, the methods to be described in the 8
active guide reflecting public comments will be used by the NRC staff in the 9
evaluation of applications for new licenses and for evaluating compliance with 10 Section IV.A.1 of 10 CFR Part 50 to Appendix G.
25
l 1
RFFERENCES 2
1.
American Society for Testing and Materials, " Standard Practice for 3
Conducting Surveillance Tests for Light-Water Cooled Nuclear Power 4
Reactor Vessels," ASTM E 185-79, July 1979.'
5 2.
American Society for Testing and Materials, " Standard Practice for 6
Conducting Surveillance Tests for Light-Water Cooled Nuclear Power 7
Reactors," ASTM E 185-82, July 1982.'
8 3.
American Society of Mechanical Engineers,Section XI, Division 1, " Rules 9
for Inservice Inspection of Nuclear Power Plant Components," of the ASNE 10 Boiler and Pressure Vessel Code, New York, through 1988 Addenda and 1989 11 Edition.2 12 4.
T.L. Dickson, " Generic Analyses for Evaluation of Low Charpy Upper-Shelf 13 Energy Effects on Safety Margins Against Fracture of Reactor Pressure 14 Vessel fiaterials," NUREG/CR-6023, July 1993.3 15 5.
R. Johnson, " Resolution of the Task A-ll Reactor Vessel Materials 16 Toughness Safety Issue," NUREG-0744, Volume 1 (Revision 1) and Volume 2 17 (Revision 1), October 1982.3 18
' Copies may be obtained from the American Society for Testing and Materials, 19 1916 Race Street, Philadelphia, PA 19103.
20 Copies may be obtained from the American Society of Mechanical Engineers, 2
21 345 East 47th Street, New York, NY 10017.
22 Copies are available for inspection or copying for a fee from the NRC Public 3
23 Document Room at 2120 L Street NW., Washington, DC; the PDR's mailing address 24 is Mail Stop LL-6, Washington, DC 20555; telephone (202)634-3273; fax 25 (202)634-3343.
Copies may be purchased at current rates from the U.S.
26 Government Printing Office, Post Office Box 37082, Washington, DC 20013-7082 27 (telephone (202)512-2249 or (202)512-2171); or from the National Technical 28 Information Service by writing NTIS at 5285 Port Royal Road, Springfield, 29 VA 22161.
26
1 1
6.
Generic Letter No. 82-26, "NUREG-0744 Rev.1; Pressure Vessel Material 2
fracture toughness," Issued by Darrel G. Eisenhut, USNRC, November 12, 3
1982.'
4 7.
Letter from Warren H. Bamford, Chairman of the ASME Subgroup on 5
Evaluation Standards for ASME Section XI, to James E. Richardson, USNRC, j
6
Subject:
Response to NRC Request, A-ll Issue, February 20, 1991.'
i 7
8.
American Society of Mechanical Engineers, " Assessment of Reactor Vessels 8
with Low Upper Shelf Charpy Impact Energy Levels,Section XI, Division j
9 1," Code Case N-512, in Supplement No. 4, " Nuclear Components," New York, i
10 1993.2 11 9.
American Society for Testing and Materials, " Standard Test Method for 12 Determining J-R Curves," ASTM E 1152-87, May 1987.'
13 10.
E.D. Eason, J.E. Wright, and E.E. Nelson, "Multivariable Modeling of l
14 Pressure Vessel and Piping J-R Data," NUREG/CR-5729, May 1991.3 I
i 15
- 11. USNRC, " Radiation Embrittlement of Reactor Vessel Materials," Regulatory l
16 Guide 1.99, Revision 2, May 1988.
l
\\
1 17 12.
A.W. Serkiz, "ASME Code Class 1, 2, and 3 Components, Components Sup-18 ports, and Core Support Structures," Revision 1 to Appendix A to Section 19 3.9.3 of NUREG-0800, " Standard Review Plan for the Review of Safety Anal-j 20 ysis Reports for Nuclear Power Plants," pp. 3.9.3-12 to 3.9.3-20, April 21 1984.3 22
- 13. American Society of Mechanical Engineers,Section III, " Nuclear Power l
23 P1 ant Components," of the ASME Boiler and Pressure Vessel Code, New York, 24 through 1988 Addenda and 1989 Edition.2 25 14.
A.L. Hiser and J.B. Terrell, " Size Effects on J-R Curves for A-302B 26 Plate," NUREG/CR-5265, January 1989.3 27
' Copies are available for inspection or copying for a fee from the NRC Public 28 Document Room at 2120 L Street NW., Washington, DC; the PDR's mailing address 29 is Mail Stop LL-6, Washington, DC 20555; telephone (202)634-3273; fax 30 (202)634-3343.
I 27
1 15.
T.J. Griesbach and E. Smith, "A Review of the ASME Low Upper Shelf 2
Toughness Evaluation Procedures for Nuclear Reactor Pressure Vessels,"
3 Nuclear Engineering and Desian, Volume 130, No. 3, pp. 259-266,1991.
4 16.
Y. Mishima et al., " Manufacture and Characteristics of a Heavy Section 5
Steel Test Plate with Changing Mechanical Properties in the Through-6 Thickness Direction," Nuclear Engineerina and Desian, Volume 137, No. 3, 7
pp. 323-334, 1992.
8 17.
C.Z. Serpan, Jr., (USNRC) Memorandum for C.Y. Cheng, Chief of Materials 9
and Chemical Engineering Branch, Division of Engineering Technology, NRR, 10 USNRC, " Ratio of Transverse to longitudinal Orientation Charpy Upper 11 Shel f Energy," June 25, 1990.'
12 18.
A.L. Hiser, (USNRC) Memorandum for C.Y. Cheng, Chief of Materials and 13 Chemical Engineering Branch, Division of Engineering Technology, NRR, 14 USNRC, " Summary of fracture Toughness Estimates for Irradiated Yankee 15 Rowe Vessel Materials," August 30, 1990.'
O O
i
1 APPENDIX A 2
Examoles of Methods D
3 Several cases are provided here as examples of the methods of analysis 4
described in this regulatory guide.
5 Example 1 (Levels A&B Loadina. PWR Vessel) 6 Consider the following geometric and material properties:
7 Vessel Geometry and Loading Conditions:
8 Vessel internal radius, R, = 86.5 in.; A-533B vessel with generic welds 9
Base metal thickness, t = 1, = 8.444 in.;
Cladding thickness, t 5/32 in.
3 ct 10 Total thickness, t' = (t, + tet) = 8.6 inch; Ratio, (R,/t') = 10.06 -
11 System accumulation pressure, p, = 2.75 ksi; Cooldown transient = 100 F/hr.
12 Base Metal Thermo-Elastic Properties:
13 Modulus of elasticity, E - 27E3 ksi; Poisson's ratio, v = 0.3 14 Yield stress, o 80 ksi; Ultimate stress, o = 90 ksi
=
y u
2 15 Flow stress, o, = 85 ksi; Fluid heat transfer coeff. = 1000 BTU /hr-ft - F 2
16 Thermal diffusivity = 0.98 in / minute; (E.a)/(1 - v) = 0.305 ksi/ F 17 Cladding Thermo-Elastic Properties:
18 Thermal expansion coefficient, a = 9.1E-6/ F; Poisson's ratio, y = 0.3 19 Modulus of elasticity, E = 27E3 ksi; Thern.al conductivity = 10 BTU /hr-ft-F 20 Stress-free temperature of Cladding = 550 F; Initial operating temp. - 550 F 21 The VISA-II code,* with modifications for printing K,p, K,, and K for 3
ct 22 6-to-1 aspect ratio flaws, was used to perform analyses for determining tran-23 sient thermo-mechanical stresses and temperature gradients across vessel wall 24 thickness. An axial flaw with an aspect ratio of 6 to I was postulated to j
25 exist in the vessel internal wall.
To account for the effect of crack-face 26 pressure on stress intensity factor solutions in VISA-II, the accumulation
{
27 pressure was adjusted to be equal to [p.t'.{1 4 R /t'}/R,], 3.02 ksi. At a i
+
t 28
- See F.A. Simonen et al., " VISA-II - A Computer Code for Predicting the 29 Probability of Reactor Pressure Vessel Failure," USNRC, NUREG/CR-4486, March 30 1986.
l l
A-1 l
l
1 fixed crack depth of (0.25t'+0.1) inch, the temperature history prediction is 2
shown in Figure A-1 for a transient with a constant cooldown rate of 100 F/hr.
3 With a factor of safety, SF, of 1.15 on accumulation pressure for Equa-4 tion 1 in Regulatory Position 1.1.1, the applied J-integral history at a crack 5
depth of (0.25t'+ 0.1) inch for mechanical and thermal stresses, including the 6
cladding effects, is shown in Figure A-2.
The applied J-integral reaches the 7
peak steady-state value of 486 in.-lb/in.2 in about 150 minutes. Also shown 8
in Figure A-2 are the J-R curves for generic welds (Equations 17, 24-25) at t
9 three CVN values.
These J-R curves were drawn for a crack extension, Aa, of 10 0.1 inch and for the temperature history, in Figure A-1, at a crack-depth of 11 (0.25t'+0.1) inch. A study of Figure A-2 shows an interesting trend that the 12 crack initiation is predicted to take place at about 45 minutes into the j
13 transient (with crack-tip temperature of 500 F) where the applied-J value 2
14
(= 445 in-lb/in ) is less than the peak steady-state value and is just equal 15 to the material's J-R curve at CVN value of 40 ft-lb.
Thus, the more detailed 16 analysis results in a lower CVN value that satisfies the acceptance criteria.
17 In order to satisfy Equation 2, with a safety factor of 1.25 on 18 accumulation pressure, Figure A-3 shows that CVN value should be greater than 19 or equal to 41 ft-lb.
This is significantly lower than the 47 ft-lb value 20 obtained by using the steady-state applied J-integral approach for analyzing 21 transients with constant cooldown rates.
22 Example 2 (Levels C and D Loadina. PWR Vessel) 23 The problem statement was presented in a meeting of the ASME Code, 24 Section XI, working group on flaw evaluation and operating plant criteria 25 (in Louisville, Kentucky, on December 1,1992), where results of the analyses 26 were compared by the participants. The vessel geometry and material 27 properties are:
28 PWR vessel internal radius, R, = 90.0 inch; A-533B plate material i
9.0 inch; Cladding thickness, t
= 0, R,/t = 10 29 Thickness, t = t
=
u g
30 Copper, Cu = 0.35 wt%; Nickel, Ni = 0.3 wt%;
Initial RT, = 0.0 F 31 Pre-irradiated CVN = 108 ft-lb (L-T orientation) p 2
32 Surface fluence, pt = 3.0E19 n/cm 33 Flaw orientation = Axial, in plate material; Flaw aspect ratio = 6 to 1 34 Fluid temp. at vessel surface, T(tm) = [550 - 250(1 - exp(- 0.1 tm)}] F 35 with time, tm, in minutes.
A-2
1 Heat transfer coeff. - 320 BTU /hr-ft - F; Thermal diffusivity - 0.98 in.2/ min 2
2 Elastic modulus, E - 28E3 ksi; Poisson's ratio, v - 0.3; a - 8.lE-6 in./in.- F 3
Yield stress, o 80 ksi; Flow stress, o, - 85 ksi y
4 J-R curve: J - (SF).[C1.(Aa)'2. exp{C3. ( Aa)C'))
in.-kip /in.2 5
- where, 6
In(Cl) = [-2.89 + 1.22 in(CVf() - 0.0027 T + 0.014 (pt)]
7 C2
- [0.077 + 0.116 In(Cl)]
8 C3
= [- 0.0812 - 0.0092 in(Cl)]
9 C4
= - 0.417 10 SF
- 0.741 for Level C events 11 The VISA-ll code was used to determine thermal stress and temperature I
12 history for the Level C transient specified in the problem.
It was found that 13 at time im = 20 minutes, the peak thermal stresses occur. The corresponding 14 peak thermal stress intensity factor as a function of crack depth to vessel 15 thickness ratio, a/t, of semi-elliptical flaws is given as:
16 K,, = [21.026+374.22(a/t)-1593.56(a/t)242912.l(a/t)3-2029.7(a/t)'] ksilin.
17 with, 0.05 s a/t s 0.5 18 Therefore, at a - 1 inch, K,, - 46.6 ksilin. At an internal pressure, p - 1 19 ksi, the pressure induced K9 = 18.9 ksilin. Now, if the pressure, p, is 20 increased, then at a pressure of 6.75 ksi, the J-applied at a - (0.lt + tct +
21 0.1) inch becomes equal to the material's J-R curve as shown in Figure A-4.
22 This will mark an " initiation" of ductile flaw growth. The temperature at the 23 crack-tip (a - 0.lt + t,t) for time tm - 20 minutes is 400 F.
If internal 24 pressure p is further increased, in Figure A-4 it can be seen that at pressure 25 p - 7.56 ksi the crack growth becomes unstable. That is, the slope of the J-26 applied curve becomes greater than the slope of the material's J-R curve.
27 Example 3 (Levels C and D Loadina. BWR Vessel) 28 The problem statement is the same as in Example 2, except for a BWR 29 vessel geometry. The vessel geometric details are:
30 BWR vessel intecnal radius, R, = 120.0 inch; A-533B plate material 31 Thickness, t - t,
6.0 inch; Cladding thickness, t
= 0; R /t - 20 ct i
32 Flaw orientation - Axial, in plate material; Flaw aspect ratio - 6 to 1.
l A-3
1 The VISA-II code was used to determine thermal stress and temperature 2
history for the Level C transient specified in the problem.
It was found that 3
at time tm = 16 minutes, peak thernal stresses occur.
The corresponding peak 4
thermal stress intensity factor as a function of crack depth to vessel 5
thickness ratio, a/t, of semi-elliptical flaws is given as:
6 K
= (12.243+227.94(a/t)-972.71(a/t)2+1785.2(a/t)3-1249.3(a/t)'] ksilin.
it 7
with, 0.05 s a/t s 0.5 8
Therefore, at a = 1 inch, K, = 27.9 ksilin. At an internal pressure, p = 1 3
9 ksi, the pressure-induced K
= 37.0 ksilin.
If the pressure, p, is g
10 increased, at a pressure of 4.55 ksi, the J-applied at a = (0.lt + tu + 0.1) 11 inch becomes equal to the material's J-R curve as shown in Figure A-5, which 12 will mark an " initiation" of ductile flaw growth.
The temperature at the 13 crack-tip (a = 0.lt + tu) for time tm = 16 minutes is 405'F.
If the 14 pressure, p, is further increased (see Figure A-5), it can be seen that at a 15 pressure p = 4.75 ksi the crack growth has become unstable.
The slope of the 16 J-applied curve is now greater than the slope of the material's J-R curve.
17 Example 4 (Thermal K for Prescribed Levels C and D Loading. PWR Vessel) n 18 For a PWR vessel, thermal K values are determined for a few prescribed n
19 cooldown rate, CR, transients.
The geometric and material properties are 20 given as following:
21 Vessel Geometry and Loading Conditions:
22 Vessel internal radius, R, = 86.875 in.;
A-533B plate material with cladding 23 Base metal thickness, t = t, = 8.625 in.;
Cladding thickness, t
= 5/16 in.
u 24 Total thickness, t' = (t
+tu) = 8.9375 in.;
Ratio, (R,/t') = 9.72 g
25 Thermal Cooldown rate, CR = 100 F/hr to 600 F/hr (constant, for each analysis) 26 Inner wall temperature, T w(R = R,) = 550'F 1% (R = R,) = 150'F 4
27 Base Metal Thermo-Elastic Properties:
28 Modulus of elasticity, E = 27E3 ksi; Poisson's ratio, v = 0.3 29 Fluid-film heat transfer coefficient = 1000 BTU /hr-f t - F j
2 2
30 Thermal diffusivity = 0.98 in / minute; (Ea)/(1 - v) = 0.305 31 Cladding Thermo-Elastic Properties:
32 Thermal expansion coef ficient, a = 9.lE-6/ F; Poisson's ratio. v = 0.3 A-4
1 Modulus of elasticity, E - 27E3 ksi; Thermal conductivity - 10 BTU /hr-ft-F 2
Stress-free temperature of Cladding = 550 F; Initial operating Temp. = 550 F 3
The VISA-II code was used to determine temperature and thermal stress history 4
for constant cooldown rate, CR, transients of 100 F/hr,150 F/hr, 200 F/hr, 5
300 f/hr, 400 F/hr, 500 F/hr, and 600 F/hr.
The corresponding peak thermal 6
stress intensity factor, K,,, as a function of crack depth to vessel thickness 7
ratio, a/t', for 6-to-1 aspect ratio semi-elliptical flaws were computed using 8
the VISA-Il code.
These are shown in Figure A-6, and are presented here in 9
polynomial expressions using least-square fits as:
10 For CR = 100 F/hr, with 0.05 s (a/t') s 0.5:
11 K,, = [27.284 - 5.838 (a/t') - 0.3548 (a/t')2 - 8.3858 (a/t')3] ksilin.
12 For CR = 150 F/hr, with 0.05 s (a/t') s 0.5:
13 K, = [32.003 + 40.012 (a/t') - 138.2 (a/t') - 113.98 (a/t')3] ksilin.
3 14 For CR = 200 F/hr, with 0.05 s (a/t') s 0.5:
(
15 K,, = [36.362 + 82.011 (a/t') - 265.01 (a/t')2 + 226.9 (a/t')3] ksilin.
\\
16 For CR = 300 F/hr, with 0.05 s (a/t') s 0.5:
[43.667 + 150.77 (a/t') - 474.9 (a/t')2 + 415.01 (a/t')3] ksilin.
17 K,1
=
18 For CR = 400 F/hr, with 0.05 s (a/t') s 0.5:
19 K, = [49.254 + 201.12 (a/t') - 632.1 (a/t')2 + 557.87 (a/t')3] ksilin.
3 20 For CR = 500 F/hr, with 0.05 s (a/t') s 0.5:
[53.552 + 237.64 (a/t') - 749.6 (a/t')2 + 666.62 (a/t')3] ksilin.
21 K
=
it 22 For CR = 600 F/hr, with 0.05 s (a/t') s 0.5:
23 K, = [56.927 + 264.21 (a/t') - 838.6 (a/t')2 + 750.88 (a/t')3] ksilin.
3 24 These results were also used in developing the unified Equation 13 for K,,
25 where the constant cooldown rate, CR, and the normalized crack depth, a/t',
26 are used as dependent variables. A least-square statistical fit was performed 27 to obtain Equation 13.
The cross-product term, (CR)(a/t'), was also used in 28 developing this fit, in addition to the polynomial terms in a/t' and CR.
A-5
O O
O W _
l o
mm i
um_
v
$8 l m __
\\
i 3
\\
o 8
o o
m _.
n_
o
~
e 5 h-i
\\
o.
h 8_
PWR Vessel, Service Levels A and B s m_
R,=86.5 in, tet=0.156 in, t'=8.6 in, c/f'=0.25 100 F/hr Cooldown for 150 Minutes
~
a Temperature at the Crack-Tip c
o "O
30 60 90 120 150 Transient Time (Minutes)
FIGURE A-1: Transient Temperature History at Crack Tip for Service Levels A and B.
A-6
\\
)
o a
p e
~
/
o tn -
en c
\\
/
$~
/
m co CO
~
.. / d 'f Lp O.
u t
(
cn
~,)
o 0.156 in x
PWR, Level A&B, Generic Welds, tet=/i'=0.25
-+-
C R;=86.5in, t'=8.6 in, Axial Crack, a
@_ /
100 F/hr Cooldown, Critical Time =45 Minutes
__, to o J (Applied Load, Criteria No. 1) c o
Crack Extension = 0.1 in.
J-R Curve ((CVN = 40 fi-lb))
J-R Curve CVN = 41 ff-Ib
+
+
C D
J-R Curve (CVN = 39 ft-Ib) o U1 i i i i i j iiiii;e i i i i i i i i i i 1 i i iii O
30 60 90 120 150 N
1 Transient Time (Minutes) l FIGURE A-2: J-Applied and J-R Curve History at Crack Tip for Service Levels A
['s and B at a Crack Extension of 0.1 inch.
'x A-7
1 o
to e
8-
/
tn
/
co GQ~
/
r o
~
-c
_ 8-PWR, Level A&B, Generic Welds, ta=0.156 in R =86. Sin, t'=8.6 in, Axial Crack, a/i'=0.25
, "7 i
100 F/hr Cooldown, Critical Time =45 Minutes J (Applied Load, Criteria No. 1)
.o a
J-R Curve (CVN=41 fi-lb, T=500 F)
O O
J (Applied Load, Criteria No. 2) o O
e i i i i i ia i i i iiiiiiijiiiig iiii
".0 0.1 0.2 0.3 0.4 0.5 0.6 Crack Extension (in)
FIGURE A-3: Acceptable Upper-Shelf Energy in a PWR Vessel for Service Levels A and B.
A-8
j
-,)
ao a
sn N
/
o o._
o
<-~s N
C N
m e
~
/
c g~
+
C to
}_ l?
O cn
's
?c oo-PWR Vessel, Level C, High Toughness Plate Rr=90 in, f=9 in, Axial Crack, a=0.9 in T=550-250[exp(
.1 im)] F, fm=20 minutes J (Applied Load, pi = 6.75 ksi) c o
J-R Curve (CVNP=108 ft-Ib, T=400 F) 0 0 :
J (Applied Load, p, = 7.56 ksi) oo-O 0.1 0.2 0.3 0.4 0.5 0.6 Crack Extension (in)
FIGURE A_4: Safety Margin Evaluation in a PWR Vessel for Service Level C.
i A-9
4 8
m N
/
/
oo_
o
~
l/
w 2
.5 0_
- *',/-l
}- /l co
-oo_
. BWR Vessel, Level C, High Toughness Plate R =120 in, t=6 in, Axial Crack, a=0.6 in i
3 T=550-250[exp(
.1 tm)] F, im=16 minutes J (Applied Load, pi = 4.55 ksi) c o
J-R Curve (CVNP=108 fi-Ib, T=405 F)
.c O :
J (Applied Load, pi = 4.75 ksi) o j
o
,,,,,,,,,,,,,,i,,,,,,,,,i,,,,
- 0 0.1 0.2 0.3 0.4 0.5 0.6 Crack Extension (in)
FIGURE A-5: Safety Margin Evaluation in a BWR Vessel for Service Level C.
1 A-10
tt-v O
Thermally Induced Ke (ksi fin) 3 20 30 40 50 60 70 80 90 i i,,
,1 g
g
- D m
I;-
c e
G O
\\
-3 7
k
's N 'N 23 z
s a"
o P-t i
3 N
x i
g g
x g-+
hk
\\
\\
s g
q m a o
+
S C
i
{?
r c'o N
i
't r -
o o
7 j
co
)
2 cl.
+
3" P-t Y
t k
i y,
- ? r o
I 1
L
~
7t, N
b I
l f.
o EE E
,f h
fa OP~
L f
f f
~
f n
a v_
L I
1 5*
D I
/
a n 3 7
0
<g
-+-
Su b
5 p
nm f
/
m x f
f i
/
f,f i
3 %
m f<
o P-t
/
O
\\A ~
/
nS$
~t
/
E'
,/
v
!?
e f'
f O-z,,
=
t t
t t
t t
t gr"
- =n
=
a w
a n
n a
93 d;
o o
o o
o o
g1 o
o o
o o
o u
c3 T. 't,
y, _" D, s,
,5
?,
)
?,
?,
,5 C??
?
g g 7,,
n n
u>
u so
1 APPENDIX B:
Computation of Stress Intensity Factors 2
Information about computing transient temperature gradient across the 3
vessel wall thickness, thermal stresses, pressure, and thermal stress inten-4 sity factors (K K,,) are provided in this Appendix as FORTRAN subroutines i
3p, 5
from the VISA-II code. Additional details on the computational method, theory 6
used, limitations, and names of the variables used are available in NUREG/CR-7 4486* and NUREG/CR-3384.* The computer code provided in this Appendix is for 8
general illustration only. Licensees assume responsibility for the 9
correctness of the computer code they use.
10 A description of cladding-induced thermal stress intensity factor is 11 presented in Appendix A to NUREG/CR-4486.
Limitations of the stress intensity 12 factor correction factors for finite length semi-elliptical surface flaws are 13 indicated in the Appendix C to NUREG/CR-4486.
In developing these correction 14 factors, only uniform membrane and linear bending stresses were considered.
15 In addition, the correction factors for circumferential flaws were assumed to 16 be the same as the ones for axial flaws.
Improved solutions may be used on a 17 case-specific basis if justified.
s i
i i
l l
18
- F.A. Simonen et al., " VISA-II - A Computer Code for Predicting the Probabil-19 ity of Reactor Pressure Vessel Failure," NUREG/CR-4486, March 1986.
D.L.
20 Stevens et al., " VISA - A Computer Code for Predicting the Probability of 21 Reactor Pressure Vessel failure," NUREG/CR-3384, September 1983.
Copies are 22 available for inspection or copying for a fee from the NRC Public Document 23 Room at 2120 L Street NW., Washington, DC; the PDR's mailing address is Mail 24 Stop LL-6, Washington, DC 20555; telephone (202) 634-3273; fax (202) 634-25 3343. Copies may be purchased at current rates from the U.S. Government 26 Printing Office, Post Office Box 37082, Washington, DC 20013-7082 (telephone 27 (202) 512-2249 or (202) 512-2171); or from the National Technical Information 28 Service by writing NTIS at 5285 Port Royal Road, Springfield, VA 22161.
i B-1
1 Taken From: VISA-II Code INUREG/CR-4486 (1986). NUREG/CR-3384 (198311 2
C*****************************************************************************
3 SUBROUTINE SPKI 4
C*****************************************************************************
5 C
Calculate Pressure Values, and, Stress Intensity Factor, PKI 6
DIMENSION CONST(5) 7 REAL I(5), IC(5) 8 INTEGER CRACK, TIME 9
C DETERMINE POLYNOMIAL REPRESENTATION OF PRESSURE 10 CONST(1) = PDATA(1) 11 CONST(2) = ((-25)*PDATA(1)+48*PDATA(2)-36*PDATA(3)+
12 1
16*PDATA(4)-3*PDATA(5))/(3*TMAX) 13 CONST(3) = (35*PDATA(1)-104*PDATA(2)+114*PDATA(3)-
j 14 1
56*PDATA(4)411*PDATA(5))*2/(3*TMAX**2) 15 CONST(4) = ((-5)*PDATA(1)+18*PDATA(2)-24*PDATA(3)+
16 1
14*PDATA(4)-3*PDATA(5))*16/(3*TMAX**3) i 17 CONST(5) = (PDATA(1)-4*PDATA(2)+6*PDATA(3)-4*PDATA(4)+
18 1
PDATA(5))*32/(3*TMAX**4) 19 C Calculate PRESSURE Component of Applied K, PKI, for Each Time & Crack Depth 20 OUTRAD = RAD + TH 21 FACTOR = RAD **2.0 / (OUTRAD**2.0 - RAD **2.0) 22 C
23 D0 120 TIME = 1, 10 24 TT = TMAX* TIME /10.0 25 D0 110 CRACK = 1, ICMAX 26 X = Z(CRACK)/TH j
27 C
CALCULATE INFLUENCE COEFFICIENTS 28 D0 100 M = 1, 5 29 I(M) = ZZ(M,1) + X*ZZ(M,2) + (X**2)*ZZ(M,3) + (X**3)*ZZ(M,4) 30 IC(M) = ZZC(M,1) + X*ZZC(M,2) + (X**2)*ZZC(M,3) + (X**3)*ZZC(M,4) 31 100 CONTINUE 32 PRES (TIME) = CONST(1)+CONST(2)*TT+CONST(3)*TT**2+CONST(4)*TT*
33 1
- 3+CONST(5)*TT**4 34 PKl(CRACK, TIME) = PRES (TIME)*((3.1416*Z(CRACK))**.5)*(10.5238*1(1) 35 1
- 1.1524
- 1 ( 2 )
- X+ 0.172 9
- 1 ( 3 ) * ( X*
- 2 )--0. 0230
- 1 ( 4 )
36 2
- (X**3)+0.0029*I(5)*(X**4))
B-2
1 PKIC(CRACK, TIME) = 5* PRES (TIME)*((3.1416*Z(CRACK))**.5)*IC(1) 2 RATIO = RAD / (10.0*TH)
'd 3
PKI(CRACK, TIME) = RATIO
- PKI(CRACK, TIME) 4 PKIC(CRACK, TIME) = RATIO
- PKIC(CRACK, TIME) 5 C
CALCULATE HOOP STRESS 6
SHOOP(CRACK, TIME) = FACTOR
- PRES (TIME)
- 7 1
(1.0 + (00TRAD/(RAD + Z(CRACK)))**2.0 )
8 110 CONTINUE 9
C CALCULATE LONGITUDINAL STRESS 10 SLONG(TIME) = PRES (TIME)
- FACTOR 11 120 CONTINUE 12 RETURN 13 END 14 C*****************************************************************************
15 SUBROUTINE TPOLY 16 C*****************************************************************************
17 C
CALCULATE WATER TEMPERATURES USING A " POLYNOMIAL" MODEL 18 REAL TEMP (5), CONST(5), S(5), AN(4), Y(4,5), KTEST 19 REAL K, K0, CP(4), SUM (4)
U 20 INTEGER TIME, CRACK, CONSTK, CONSTE 21 INTEGER Q 22 C
" POLYNOMIAL" Modeling of The Water Temperature 23 C
Determine Metal Temperature For EACH CRACK DEPTH AND TIME INTERVAL 24 D0 100 N = 1, 5 25 TEMP (N) = TDATA(N) - TINT 26 100 CONTINUE 27 C
FIT A " POLYNOMIAL" TO THE WATER TEMERATURE 28 CONST(1) = TEMP (l) 29 CONST(2) = ((-25)* TEMP (l) + 48* TEMP (2) - 36* TEMP (3) +
30 1
16* TEMP (4) - 3* TEMP (5))/(3*TMAX) 31 CONST(3) = (35* TEMP (l) - 104* TEMP (2) + 114* TEMP (3) -
32 1
56* TEMP (4) + ll* TEMP (5))*2/(3*TMAX**2) 33 CONST(4) = ((-5)* TEMP (1) + 18* TEMP (2) - 24* TEMP (3) +
34 1
14* TEMP (4) - 3* TEMP (5))*16/(3*TMAX**3) 35 CONST(5) = (TEMP (l) - 4* TEMP (2) + 6* TEMP (3) - 4* TEMP (4) +
36 1
TEMP (5))*32/(3*TMAX**4)
,/ 37 DO 150 TIME = 1, 10 B-3
1 TT = TMAX* TIME /10.
2 C
EQUATION FOR THE TEMPERATURE OF THE WATER 3
lWATER(TIME) - TINT +CONST(1)+ CONST(2)*TT + CONST(3)*TT**2 +
4 1
CONST(4)*TT**3 + CONST(5)*TT**4 5
00 150 CRACK = 1, 5 6
K = K0 7
110 X = ZQ(CRACK)/TH 8
TAU = K*TT/TH**2
)
9 D0 120 M = 1, 5 j
10 S(M) = CONST'M) * ((TH**2/K)*'(M-1))
11 120 CONTINUE
'l 12 DO 130 N = 1, 4 13 ALNQ = AL(N,Q) 14 AN(N) = 2
- SIN (ALNQ)/(ALNQ + SIN (ALNQ)* COS(ALNQ))
15 CP(N) = COS(ALNQ * (1-X))
16 Y(N,1) = 1 - EXP(-(ALNQ**2)* TAU) 17 D0 130 M = 2, 5 18 Y(N,M) = TAU **(M-1) - (Y (N,M-1)/ALNQ**2)*(M-1) 19 130 CONTINUE 20 00 140 N = 1, 4 i
21 ALNQ = AL(N,0) j 22 SUM (N) - AN(N)
- CP(N) * (S(l)
- EXP(-(ALNQ**2* TAU)) + S(2) 23 1
- Y(N,1)/ALNQ**2 +2*S(3)* Y(N,21/ALNQ**2 + 3 *S(4)
- Y(N,3) 24 2
/ALNQ**2 +4 *S(5)*Y(N,4)/ALNQ**2) 25 140 CONTINUE 26 C
EQUATION FOR THE QUARTER POINT TEMPERATURES 27 TQ(CRACK, TIME) = TWATER(TIME) - SUM (1) - SUM (2) - SUM (3) - SUM (4) 28 C
CONTROL FOR THE CONSTANT KAPPA OPTION 1
l 29 IF (CONSTK.EQ. 1) GO TO 150 30 C
TEST FOR THE ACCURACY OF KAPPA FOR THE GIVEN METAL TEMPERATURE, 31 C
IF THE DESIRED ACCURACY IS NOT OBTAINED, ITERATE ON KAPPA 32 C
FOR THIS CRACK DEPTH AND TIME.
33 KTEST = 1.030 - (5.97E-7)*((T(CRACK, TIME))**2) 34 IF ((ABS (KTEST-K)).LE. 0.0001) GO 'O 150 35 K = KTEST 36 GO TO 110 37 150 CONTINUE B-4
1 RETURN 2
END 3
C*****************************************************************************
4 SUBROUTINE TEXP 5
C*****************************************************************************
6 C
Calculate WATER TEMPERATURES Using an " Exponential Decay" Model 7
REAL B, KTEST, K, KO, SUM (4) 8 INTEGER CRACK, TIME, CONSTK, CONSTE 3
INTEGER Q 10 C
EXPONENTIAL DECAY MODEL OF THE WATER TEMPERAT:!RE 11 D0 130 TIME = 1, 10 12 TT = TMAX* TIME /10.
13 C
EQUATION FOR THE TEMPERATURE OF WATER 14 TWATER(TIME) - TO + DT * (1-EXP(-BE*TT))
15 D0 130 CRACK = 1, S 16 K = K0 17 100 WSQ = BE*TH*TH/K 18 TAU - K*TT/(TH*TH) 19 D0 120 N = 1, 4 f
20 ALNQ = AL(N,Q) 21 B
- -DT*((2* SIN (ALNQ)/(ALNQ+(SIN (ALNQ))*(C0S(ALNQ))))
22 1
- ( EX P (- ( ALNQ* *2
- TAU) )- EX P (-WSQ*T AU) ) / ( ( ALNQ ** 2 /WSQ)- 1 ) )
23 X
- ZQ(CRACK)/TH 24 SUM (N) = B
- COS(ALNQ*(1-X))
25 120 CONTINUE 26 C
EQUATON FOR THE " QUARTER POINTS" TEMPERATURE VALUES 27 TQ(CRACK, TIME) - TWATER(TIME) - SUM (1) - SUM (2) - SUM (3) - SUM (4) 28 C
CONTROL FOR THE CONSTANT KAPPA OPTION 29 IF (CONSTK.EQ. 1) GO TO 130 30 C
TEST FOR KAPPA ACCURACY AND CONTROL OF KAPPA OPTION 31 KTEST - 1.030 - (5.97E-7)+((T(CRACK, TIME))**2) 32 IF ((ABS (KTEST-k)).LE. 0.0001) GO TO 130 33 K = KTEST 34 GO TO 100 35 130 CONTINUE 36 RETURN 37 END B-5
1 C*****************************************************************************
2 SUBROUTINE SKIT 3
C*****************************************************************************
4 C
Calculate Stress and Temperature at Crack-Tip and Thermal Stress 5
C Intensity Factor, Skit 6
REAL E(5,10), CC(5), 1(5), IC(5) 7 INTEGER CRACK, TIME 8
INTEGER Q, CONSTE, CONSTK 9
C DETERMINE POLYNOMIAL REPRESENTATION OF TEMPERATURE PROFILE 10 C
CONVERT CLAD THERMAL CONDUCTIVITY 10 INCH AND MINUTE UNITS 11 CCOND = CCOND / (12.0*60.0) 12 COND - COND / (12.0*60.0) 13 D0 105 TIME - 1, 10 14 TQ1 - TQ(1, TIME) 15 TQ2 - TQ(2, TIME) 16 TQ3 - TQ(3, TIME) 17 TQ4 - TQ(4, TIME) 18 TQS - TQ(5, TIME) 19 C1 - TQ1 20 C2 - (-25*TQ1+48*TQ2-36*TQ3+16*TQ4-3*TQ5)/(3*TH) 21 C3 - (35*TQl-104*TQ2+114*TQ3-56*TQ4+11*TQ5)*(2.0/3.0*TH**(-2))
22 C4 - (-5*TQ1+18*TQ2-24*TQ3+14*TQ4-3*TQ5)*(16.0/3.0*TH**(-3))
23 C5 - (TQl-4*TQ2+6*TQ3-4*TQ4+TQ5)*(32.0/3.0*TH**(-4))
24 C
CALCUATE TEMPRATURE AT THE CRACK TIPS 25 00 100 CRACK = 1, ICMAX 26 T(CRACK, TIME) - Cl+C2*Z(CRACK)+C3*(Z(CRACK)**2) 27 1
+C4*(Z(CRACK)**3)+C5*(Z(CRACK)**4) 28 100 CONTINUE 29 IF (CTH.LE. 0.0) GO TO 105 30 T(1, TIME) - T(2, TIME) - (COND/CCOND)*(T(2, TIME)-T(1, TIME))
31 105 CONTINUE 32 IF (CONSTE.EQ. 1) GO TO 120 33 D0 110 TIME - 1, 10 34 DO 110 CRACK - 1, 5 35 r(CRACK, TIME) = 0.286+(5.400E-5 * (TQ(CRACK, TIME)))
36 1
-(2.600E-8 * (TQ(CRACK, TIME))**2) 37 110 CONTINUE B-6
1 G0 10 140 2
120 D0 130 TIME = 1, 10 3
DO 130 CRACK = 1, 5 l
4 E(CRACK, TIME) = EDATA
]
5 130 CONTINUE 6
C DETERMINE POLYNOMIAL REPRESENTATION OF STRESS DIST 7
140 D0 170 TIME = 1, 10 8
D0 150 CRACK = 1, 5 9
CC(CRACK) = E(CRACK, TIME)*TQ(CRACK, TIME) 10 150 CONTINUE 11 Al - CC(l) 12 A2 = (-25*CC(l)+48*CC(2)-36*CC(3)+16*CC(4)-3*CC(5))/3.0 13 A3 = (35*CC(l)-104*CC(2)+114*CC(3)-56*CC(4)+11*CC(5))*(2.0/3.0) 14 A4 = (-5*CC(l)+18*CC(2)-24*CC(3)+14*CC(4)-3*CC(5))*(16.0/3.0) 15 A5 = (CC(l)-4*CC(2)+6*CC(3)-4*CC(4)+CC(5))*(32.0/3.0) 16 SIG1 - A2/2.0 + A3/3.0 + A4/4.0 + AS/5.0 17 SIG2 - -A2 18 SIG3 - -A3 m
19 SIG4 - -A4 20 SIG5 - -A5 21 C
CALCULATE STRESS AT CRACK TIPS 22 D0 170 CRACK = 1, ICMAX 23 X = Z(CRACK)/TH 24 STRESS (CRACK, TIME) = SIG1 + SIG2*X + SIG3*(X**2) 25 1
+ SIG4*(X**3) + SIG5*(X**4) i 26 C
CALCULATE INFLUENCE FUNCTIONS 27 D0 160 M = 1, 5 28 1(M)
= ZZ(M,1) + X*ZZ(M,2)+ (X**2)*ZZ(M,3)+ (X**3)*ZZ(M,4) 29 IC(M) = ZZC(M,1)+X*ZZC(M,2)+(X**2)*ZZC(M,3)+(X**3)*ZZC(M,4) 30 160 CONTINUE 31 A = Z(CRACK) 32 C
EQUATION FOR THE THERMAL STRESS INTENSITY 33 TK(CRACK, TIME) = ((3.1416*A)**.5)*(SIGl*I(l) l 34 1
+SIG2*I(2)*X+SIG3*1(3)*X**2 35 2
+SIG4*I(4)*X**3+SIG5*I(5)*X**4) 36 TKC(CRACK, TIME) = ((3.1416*A)**.5)*(SIGl*1C(l)+SIG2*1C(2)
(
)
37 1
- X+SIG3*1C(3)*X**2+SIG4*IC(4)*X**3+SIG5*IC(5)*X**4) i L./
B-7
_____.__._____J
i i
170 CONTINUE 2
RETURN 3
END 4
C*****************************************************************************
5 SUBROUTINE KICLAD 6
C*****************************************************************************
7 C
THIS SUBROUTINE CALCULATES STRESSES AND STRESS INTENSITY FACTORS 8
C DUE TO THE PRESENCE OF " CLADDING" ON THE I.D. SURFACE OF THE VESSEL 9
INTEGER CRACK, TIME 10 INTEGER CONSTE, CONSTK, Q 11 REAL 10, Il 12 DO 170 TIME = 1, 10 13 C
CALCULATE STRESS DISTRIBUTION THROUGH VESSEL WALL 14 C
TEMP AT CLAD / BASE METAL INTERFACE 15 T1 - 0.5*(T(2, TIME) + T(3, TIME))
16 C
TEMPERATURE AT THE VESSEL 1.D.
17 TO - T(1, TIME) 18 C
STRESS-FREE TEMPERATURE 19 TI - SFREET 20 C
CALCULATE STRESS DISTRIBUTION DUE TO CLAD 21 C
SIGCI - STRESS IN CLAD AT VESSEL I.D.
22 C
SIGC2 - STRESS IN CLAD AT CLAD / BASE METAL INTERFACE 23 C
SIGB1 - STRESS IN BASE METAL AT CLAD / BASE METAL INTERFACE 24 C
SIGB2 - STRESS IN BASE METAL AT VESSEL 0.D.
25 DELEA - CLADE*CALPHA*(1-ARATIO)/(1-CLADNU) 26 C
CALCULATE STRESS IN CLAD (KSI) 27 SIGCl - DELEA * (TI - TO) 28 SIGC2 - DELEA * (TI - TI) 29 C
CALCULATE FORCE DEVELOPED IN CLAD 30 FCLAD = CTH*0.5*(SIGCl + SIGC2) 31 C
CALCULATE STRESSES IN BASE METAL (KSI)
M.
R0 - RAD 33 R1 = RAD + CTH 34 R2 - RAD + TH 35 CONST = 1.0/((R2/RI)**2.0-1.0)*(RO-RI)/Rl*DELEA 36 1
- (TI-0.5*(T0+Tl))
37 SIGB1 = CONST * (1 + (R2/Rl)**2.0)
B-8
l 1
SIGB2 - CONST
- 2.0 l
2 C
CALCULATE FORCE DEVELOPED IN BASE METAL 3
FBASE - (CTH-TH)*0.5*(SIGBl+SIGB2) 4 C
ADJUST SIGB1 AND SIGB2 TO BALANCE FORCES FCLAD AND FBASE 5
SIGINC - 0.5*(SIGB1-SIGB2) 6 SIGAVE = 0.5*(SIGBl+SIGB2)*FCLAD/FBASE 7
SIGB1 - SIGAVE + SIGINC 8
SIGB2 - SIGAVE - SIGINC i
9 C
CALCULATE CONSTANTS DESCRIBING STRESS DISTRIBUTION 10 C
QI - SLOPE OF CLAD STRESS DISTR.
11 QI - (SIGCl-SIGC2)/SIGC1/(CTH/TH) 12 C
P - SLOPE OF BASE METAL STRESS DISTR.
13 P - (SIGB2-SIGB1)/SIGCl / ((TH-CTH)/TH) 14 C
-R - INTERCEPT OF BASE METAL STRESS GRAD. AT VESSEL I.D.
15 R - -(SIGBl/SIGCl - P*CTH/TH) 16 C
CALCULATE STRESS AND KI DUE TO CLAD FOR ALL Z(CRACK)'S 17 C
KI AT THE I.D. SURFACE EQUALS ZERO (I.E.,CRACKDEPTH = ZER0) 18 SCLAD(1, TIME) - SIGCI 19 CLADK(1, TIME) - 0.0 20 C
KI IN CLAD NEAR CLAD / BASE METAL INTERFACE 21 SCLAD(2, TIME) - SIGC2 22 ALP - Z(2)/TH 23 10 = 1.12240.9513* ALP-0.624* ALP **2.0+8.3306* ALP **3.0 24 11 = 0.682540.3704* ALP-0.0832* ALP **2 0+2.8251* ALP **3.0 25 CLADK(2, TIME) - SQRT(3.14159*Z(2))*SIGCl*(IO-QI* ALP *II) 26 C
CALCULATE KI IN BASE METAL 27 XI - CTH/TH 28 D0 170 CRACK = 3, 35 29
. ALP = Z(CRACK)/TH 30 SCLAD(CRACK, LIME) - (-R+ ALP *P)*SIGCl 31 10 - 1.122+0.9513* ALP-0.624* ALP **2.0+8.3306* ALP **3.0 32 CLADK(CRACK, TIME) = SQRT(3.14159*Z(CRACK))*SIGCl*l.751938 33 1
- ((IO-0.63662)*((1.0+R)*ASIN(XI/ ALP)+ ALP *((QI4R*P) 34 2
- SQRT(1.-(XI/ ALP)**2.)-QI)-1.570796*R)+(10-1.0)*(((1.0+R)-XI/2.
l 35 3
- (Q I + R* P) ) *SQRT (1. -(X I /AL P) *
-1.0-0.7894*R*P* ALP))
O 37 170 CONTINUE B-9
I RETURN 2
END 3
C*****************************************************************************
4 SUBROVIINE FACMB (AAA, BBB, THH, FMA, FMB, FBA, FBB) 5 C*****************************************************************************
6 C
THIS SUBROUTINE CORRECTS FOR " FINITE LENGTH" SEMI-ELLIPTICAL FLAWS 7
DIMENSION ZM(2,4), ZB(2,4), Z(2) 8 DIMENSION X1(12), YM(12,4), YB(12,4), Y(4) 9 DATA X1/0.,.0125,.025,.0375,.05,.075,
.1,.15,.2,.3,.4,.5/
10 DATA Y/.05,.25,
.5,.8 /
11 DATA YM/ 1.0,.99,.98,.96,.95,.91,.87, 80,.75,.66,.60,.55, 12 1
1.0,.94,.88,.83,.80,.76,.73,.68,.63,.55,.49,.44, 13 2
1.0,.88,.77,.69,.64,.59,.55,.49,.44,.36,.31,.27, 14 3
1.0,-.72,.56,.48,.43,.38,.35,.29,.24,.18,.15,.13 /
15 DATA YB/ 1.0,.98,.97,.95,.9*,.92,.89,.85,.82,.74,.66,.58, 16 2
1.,.93,.88,.84,.80,.75,.72,.67,.63,.57,.50,.43, 17 2
1.,.84,.71,.63,.57,.49,.45,.39,.35,.29,.23,.18, 18 3
1.,
.69,.50,.38,.29,.20,.14,.08,.05,.02,.01,.04/
19 DATA Z/
0.0, 0.5 /
20 DATA ZM/.44,.55,.40,.48,.31,.31,.23,.17 /
21 DATA ZB/.50,.62,.63,.67,.58,.50,.43,.32 /
22 AOL = AAA/(2.0*BBB) i 23 A0T = AAA/THH 24 D0 100 I - 1, 3 25 J-I 26 IF( Y(I+1).GT. A0T ) GO TO 110 17 100 CONTINUE 28 110 N1 - J 29 N2 - J+1 30 DO 120 I = 1, 11 31 J=1 32 IF ( X1(I+1).GT. AOL ) G0 10 130 33 120 CONTINUE 1
34 130 M1 - J 35 M2 - J+1 36 FACl - (AOL-X1(M1))/(XI(M2)-X1(M1))
37 XXI
= YM(M1,N1)+FACl*(YM(M2,N1)-YM(M1,N1))
B-10
1 XX2 = YM(M1,N2) + FACl*(YM(M2,N2) - YM(M1,N2))
2 FAC = (A0T -Y(N1))/(Y(N2)-Y(N1))
3 IF (A0T.LT. 0.05 )
FAC = 0.0 4
IF ( A0T.GT. 0.80 ) FAC = 1.0 5
FMA = XXI + FAC*( XX2 - XXI )
6 XXI - YB(M1,N1) + FACl*(YB(M2,N1)-YB(MI,N1))
7 XX2 = YB(M1,N2) + FACl*(YB(M2,N2)-YB(M1,N2))
8 FBA = XXI + FAC*( XX2 - XXI )
9 FACl - AOL/0.5 10 XXI = ZM(1,N1) + FACl*(ZM(2,N1)-ZM(1,N1))
l 11 XX2 = ZM(1,N2) + FACl*(ZM(2,N2)-ZM(1,N2))
12 FMB = XXI + FAC*( XX2-XX1) 13 XXI = ZB(1,N1) + FACl*(ZB(2,N1)- ZB(1,N1))
14 XX2 = ZB(1,N2) + FACl*(ZB(2,N2)- ZB(1,N2))
15 FBB = XXI + FAC*(XX2 - XX1) 16 RETURN 17 END I
O B-11 i
~ - - -
1 RfCULATORY ANALYSIS 2
1.
STATEMENT OF THE PROBLEM 3
Appendix G, " Fracture Toughness Requirements," to 10 CfR Part 50, 4
" Domestic Licensing of Production and Utilization Facilities," requires, in 5
part, that the reactor vessel beltline materials "
.must have Charpy upper-6 shelf energy of no less than 75 f t-lb (102J) initially and must maintain 7
upper-shelf energy throughout the life of the vessel of no less than 50 ft-lb 8
(68J), unless it is demonstrated in a manner approved by the Director, Office 9
of Nt: lear Reactor Regulation, that lower values of upper-shelf energy will 10 provide margins of safety against fracture equivalent to those required by 11 Appendix G of the ASME Code." Draft Regulatory Guide, DG-1023, " Evaluation of 12 Reactor Pressure Vessels with Charpy Upper-Shelf Energy Less Than 50 ft-lb,"
13 is being developed to provide acceptance criteria and analysis methods 14 acceptable to the NRC staff for demonstrating margins equivalent to those in 15 Appendix G of the ASME Code.
16 Publication of regulatory guidance is necessary because no guidance on 17 Appendix G currently exists.
There are reactors, both pressurized water reac-18 tors and boiling water reactors, with upper-shelf energy that is projected to
~
19 fall below the 50 f t-lb regulatory limit before the end of the current license 20 period. Without regulatory guidance, each affected licensee will have to 21 submit a plant-specific analysis, including acceptance criteria and evaluation 22 methods, and the staff will have to evaluate each submittal without the 23 benefit of stated acceptance criteria and approved evaluation methods.
24 2.
OBJECTIVES 25 The objective of this gside is to provide acceptance criteria and 26 evaluation methods acceptable to the NRC staff for demonstrating margins 27 equivalent to those of Ap podix G of the ASME Code for those beltline 28 materials whose Charpy upper-shelf energy falls below the regulatory limit 29 provided in Appendix G to 10 CFR Part 50.
30 3.
AtTERNATIVES f
31 Two alternatives to issuing the evaluation procedures for pressure 32 vessels with Charpy upper-shelf energy less than 50 ft-lb were considered:
R-1
l 1
1 (1) endorse actions being implemented by the ASME Code,Section XI, and (2) 2 take no action.
3 3.1 Endorse ASME Code.Section XI Code Case 4
The ASME Code, in Code Case N-512' in Section XI, provides acceptance 5
criteria and evaluation procedures for pressure vessels with Charpy upper-6 shelf energy less than 50 ft-lb. However, the code case evaluation procedures 7
currently address only Service Levels A and B, and no guidance on specific 8
materials properties is provided.
It is important that all four service 9
levels be considered in the evaluations, and it is important that specific 10 guidance on estimating material properties be provided.
Given the ASME codi-11 fication process, and the process whereby the NRC endorses ASME Code Cases, 12 the time delay in obtaining suitable guidance would be excessive. At present, 13 ASME Code Case N-512 does not provide complete guidance. As discussed above, 14 the code case does not provide information on the selection of transients, and 15 it gives very little detail on material properties selection.
As such, a 16 request for revision of ASME Code Case N-512 will have to be made.
17 3.2 Take No Action 18 As discussed in SECY-93-048,2 " Status of Reactor Pressure Vessel Issues 19 Including Compliance With 10 CFR Part 50, Appendices G and H," using the NRC 20 staff's generic criteria for estimating Charpy upper-shelf energy, there are 21 15 plants that currently would have calculated upper-shelf energy less than 50 22 ft-lb and 3 others that would have upper-shelf energy below 50 ft-lb before 23 the end of their operating licenses.
Appendix G to 10 CFR Part 50 requires 24 licensees to submit analyses to demonstrate margins equivalent to those in 25 Appendix G of the ASME Code 3 years before the upper-shelf energy of any 26
'American Society of Mechanical Engineers, " Assessment of Reactor Vessels 27 with Low Upper Shelf Charpy Impact Energy Levels,Section XI, Division 1,"
28 Code Case N-512, in Supplement No. 4, " Nuclear Components," New York,1993.
2 29 James M. Taylor, Executive Director for Operations, SECY-93-048, Policy 30 Issue (Information) for The Commissioners, USNRC, February 25, 1993.
Copies 31 are available for inspection or copying for a fee from the NRC Public Document 32 Room at 2120 L Street NW., Washington, DC; the PDR's mailing address is Mail 33 Stop LL-6, Washington, DC 20555; telephone (202)634-3273; fax (202)643-3343.
R-2
1 beltline materials falls below 50 f t-lb.
Therefore, taking no action is not a 2
viable alternative.
3 4.
COSTS AND BENEFITS OF ALTERNATIVES 4
The cost and benefits of the two alternatives discussed above are 5
presented here.
6 4.1 Endorse ASME Code Section XI Code Case 7
The acceptance criteria proposed ir the ASME Section XI code case are 8
identical to those proposed in this draft regulatory guide. The regulatory 9
guide analysis procedures for Service Levels A and B were taken from the code 10 case. However, the guide provides procedures applicable to Service Levels C 11 and D.
The guide proposes specific guidance on appropriate material proper-12 ties and on the selection of transients for consideration, whereas the code 13 case does not provide these procedures and guidance. Thus, without this 14 guide, each affected licensee would develop appropriate procedures for Service 15 Levels C and D, to justify the choice of transients, and to develop plant-16 specific material properties.
O 17 It is estimated that without the guidance proposed in this regulatory 18 guide, developing plant-specific procedures and material properties and apply-19 ing them to check and report the analysis results would require an additional l
20 6 staff-months (1040 hours0.012 days <br />0.289 hours <br />0.00172 weeks <br />3.9572e-4 months <br />) for each affected licensee. Assuming that one-l 21 half of the affected licensees either belong to owners' groups or could make l
22 use of common data, the total additional burden on the licensees that would be l
23 incurred by plant-specific anaTyses is estimated as 9 plants x 6 staff-months 24 per plant, or 54 staff-months (9360 hours0.108 days <br />2.6 hours <br />0.0155 weeks <br />0.00356 months <br />).
25 In addition to the increased burden on the licensees, it is estimated 26 that an additional 1.5 NRC staff-month would be required to review each plant-27 specific submittal. Thus, the total increased burden on the NRC staff, 28 assuming that one-half of the affected plants can be grouped, is estimath to 29 be 9 plants x 1.5 staff-month per plant, or 13.5 staff-months (2340 hours0.0271 days <br />0.65 hours <br />0.00387 weeks <br />8.9037e-4 months <br />).
30 This estimate assumes that there would be only minor discussions with the 31 licensees.
O R-3
1 4.2 Take No Action 2
As discussed in Section 3.2 above, taking no action is not judged to be a 3
viable alternative.
4 5.
DECISION RATIONALE 5
It is recommended that the regulatory guide be issued because it would 6
offer a comprehensive set of acceptance criteria, evaluation procedures, and 7
material properties that can be used to perform the analyses required under 8
Appendix G to 10 CFR Part 50 for those pressure vessels with Charpy upper-9 shelf energy of any beltline material that falls below 50 ft-lb.
Issuing the 10 regulatory guide is recommended over the alternative of endorsing the ASME 11 Section XI code case because the code case does not currently include: (1) 12 analysis procedures for Service Levels C and D, (2) guidance on selecting the 13 transients for evaluation, nor (3) details on temperature-dependent material 14 properties.
Further, it is estimated that preparing plant-specific analyses 15 that include the necessary procedures and data that are not addressed in the 16 code case would require approximately 54 staff-months of effort for the indus-17 try and approximately 9 staff-months for the NRC to review the additional 18 information.
19 The NRC staff considered the possibility of working with the ASME Code 20 Section XI working group to modify the code case to include the missing proce-21 dures and data.
However, given the number of plants that could need the guid-22 ance in the near term, and given the ASME codification process and the NRC's 23 process for endorsing ASME code cases, the time needed to modify and endorse 24 the code case was judged to be excessive.
25 The efficacy of the proposed regulatory guide procedures was demonstrated 3
26 by generic bounding calculations performed by the NRC staff in preparing 27 SECY-93-048.
These calculations demonstrated that the requirement in 3
28 Charles Z. Serpan, Jr., Memorandum to Jack Strosnider, January 15, 1993, 29
" Generic Bounding Analyses for Evaluation of Low Charpy Upper-Shelf Energy 30 Effects on Safety Margins Against Fracture of RPV Beltline Plate and Weld 31 Materials"; and Charles Z. Serpan, Jr., Memorandum to Jack Strosnider, Feb-32 ruary 8, 1993, " Additional Information Regarding Results of Generic Bounding 33 Analyses for Evaluation of Pressure Vessels Fabricated Using Low Charpy Upper-34 Shelf Energy Materials." Copies are available for inspection or copying for a 35 fee from the NRC Public Document Room at 2120 L Street NW., Washington, DC; 36 the PDR's mailing address is Mail Stop LL-6, Washington, DC 20555; phone 37 (202)634-3273; fax (202)634-3343.
R-4
=
l 1
REGULATORY ANALYSIS 2
1.
STATEMENT OF THE PROBLEM 3
Appendix G, " Fracture Toughness Requirements," to 10 CFR Part 50, 4
" Domestic Licensing of Production and Utilization Facilities," requires, in 5
part, that the reactor vessel beltline materials "...must have Charpy upper-6 shelf energy of no less than 75 ft-lb (102J) initially and must maintain 7
upper-shelf energy throughout the life of the vessel of no less than 50 ft-lb 8
(68J), unless it is demonstrated in a manner approved by the Director, Office 9
of Nuclear Reactor Regulation, that lower values of upper-shelf energy will 10 provide margins of safety against fracture equivalent to those required by 11 Appendix G of the ASME Code." Draft Regulatory Guide, DG-1023, " Evaluation of 12 Reactor Pressure Vessels with Charpy Upper-Shelf Energy less Than 50 ft-lb,"
13 is being developed to provide acceptance criteria and analysis methods 14 acceptable to the NRC staff for demonstrating margins equivalent to those in 15 Appendix G of the ASME Code.
16 Publication of regulatory guidance is necessary because no guidance on 17 Appendix G currently exists.
There a e reactors, both pressurized water reac-18 tors and boiling water reactors, with uprer-shelf energy that is projected to 19 fall below the 50 ft-lb regulatory limit before the end of the current license 20 period. Without regulatory guidance, each affected licensee will have to 21 submit a plant-specific analysis, including acceptance criteria and evaluation 22 methods, and the staff will have to evaluate each submittal without the 23 benefit of stated acceptance criteria and approved evaluation methods.
24 2.
OBJECTIVES 25 The objective of this guide is to provide acceptance criteria and 26 evaluation methods acceptable to the NRC staff for demonstrating margins 27 equivalent to those of Appendix G of the ASME Code for those beltline 28 materials whose Charpy upper-shelf energy falls below the regulatory limit 29 provided in Appendix G to 10 CFR Part 50.
30 3.
ALTERNATIVES i
31 Two alternatives to issuing the evaluation procedures for pressure 32 vessels with Charpy upper-shelf energy less than 50 ft-lb were considered:
R-1
i 1
1 (1) endorse actions beirg implemented by the ASME Code,Section XI, and (2) 2 take no action.
3 3.1 Endorse ASME Code.Section XI Code Case 4
The ASME Code, in Code Case N-512' in Section XI, provides acceptance 5
criteria and evaluation procedures for pressure vessels with Charpy upper-6 shelf energy less than 50 f t-lb.
However, the code case evaluation procedures 7
currently address only Service Levels A and B, and no guidance on specific 8
materials properties is provided.
It is important that all four service 9
levels be considered in the evaluations, and it is important that specific 10 guidance on estimating material properties be provided.
Given the ASME codi-11 fication process, and the process whereby the NRC endorses ASME Code Cases, 12 the time delay in obtaining suitable guidance would be excessive. At present, 13 ASME Code Case N-512 does not provide complete guidance. As discussed above, 14 the code case does not provide information on the selection of transients, and 15 it gives very little detail on material properties selection.
As such, a 16 request for revision of ASME Code Case N-512 will have to be made.
17 3.2 Take No Action 18 As discussed in SECY-93-048,2 " Status of Reactor Pressure Vessel Issues
'9 Including Compliance Witt 10 CFR Part 50, Appendices G and H," using the NRC 20 staff's generic criteria for estimating Charpy upper-shelf energy, there are 21 15 plants that currently would have calculated upper-shelf energy less than 50 22 f t-lb and 3 others that would have upper-shelf energy below 50 ft-lb before 23 the end of their operating licenses.
Appendix G to 10 CFR Part 50 requires 24 licensees to submit analyses to demonstrate margins equivalent to those in 25 Appendix G of the ASME Code 3 years before the upper-shelf energy of any 26
'American Society of Mechanical Erigineers, " Assessment of Reactor Vessels 27 with Low Upper Shelf Charpy Impact Energy Levels,Section XI, Division 1,"
28 Code Case N-512, in Supplcment No. 4,
" Nuclear Components," New York, 1993.
2 29 James M. Taylor, Executive Director for Operations, SECY-93-048, Policy 30 Issue (Information) for The Commissioners, USNRC, February 25, 1993.
Copies 31 are available for inspection or copying for a fee from the NRC Public Document 32 Room at 2120 L Street NW., Washington, DC; the PDR's mailing address is Mail 33 Stop LL-6, Washington, DC 20555; telephone (202)634-3273; fax (202)643-3343.
R-2
1 beltline materials falls below 50 ft-lb.
Therefore, taking no action is not a 2
viable alternative.
3 4.
COSTS AND BENEFITS OF AtTERNATIVES 4
The cost and benefits of the two alternatives discussed above are 5
presented here.
6 4.1 Endorse ASME Code Section XI Code Case 7
The acceptance criteria proposed in the ASME Section XI code case are 8
identical to those proposed in this draft regulatory guide. The regulatory 9
guide analysis procedures for Service Levels A and B were taken from the code 10 case.
However, the guide provides procedures applicable to Service Levels C 11 and D.
The guide proposes specific guidance on appropriate material proper-12 ties and on the selection of transients for consideration, whereas the code 13 case does not provide these procedures and guidance.
Thus, without this 14 guide, each affected licensee would develop appropriate procedures for Service 15 Levels C and D, to justify the choice of transients, and to develop plant-Ip 16 specific material properties.
pd 17 It is estimated that without the guidance proposed in this regulatory 18 guide, developing plant-specific procedures and material properties and apply-19 ing them to check and report the analysis results would require an additional 20 6 staff-months (1040 hours0.012 days <br />0.289 hours <br />0.00172 weeks <br />3.9572e-4 months <br />) for each affected licensee. Assuming that or.e-21 half of the affected licensees either belong to owners' groups or could make 22 use of common data, the total additional burden on the licensees that would be 23 incurred by plant-specific analyses is estimated as 9 plants x 6 staff-months 24 per plant, or 54 staff-months (9360 hours0.108 days <br />2.6 hours <br />0.0155 weeks <br />0.00356 months <br />).
25 In addition to the increased burden on the licensees, it is estimated 26 that an additional 1.5 NRC staff-month would be required to review each plant-27 specific submittal.
Thus, the total increased burden on the NRC staff, 28 assuming that one-half of the affected plants can be grouped, is estimated to 29 be 9 plants x 1.5 staff-month per plant, or 13.5 staff-months (2340 hours0.0271 days <br />0.65 hours <br />0.00387 weeks <br />8.9037e-4 months <br />).
30 This estimate assumes that there would be only minor discussions with the i
31 licensees.
g R-3
1 4.2 Take No Action 2
As discussed in Section 3.2 above, taking no action is not judged to be a 3
viable alternative.
4 5.
DECISION RATIONALE 5
It is recommended that the regulatory guide be issued because it would 6
offer a comprehensive set of acceptance criteria, evaluation procedures, and 7
material properties that can be used to perform the analyses required under 8
Appendix G to 10 CFR Part 50 for those pressure vessels with Charpy upper-9 shelf energy of any beltline material that falls below 50 ft-lb.
Issuing the 10 regulatory guide is recommended over the alternative of endorsing the ASME 11 Section XI code case because the code case does not currently include: (1) 12 analysis procedures for Service levels C and D, (2) guidance on selecting the 13 transients for evaluation, nor (3) details on temperature-dependent material 14 properties.
Further, it is estimated that preparing plant-specific analyses 15 that include the necessary procedures and data that are not addressed in the 16 code case would require approximately 54 staff-months of effort for the indus-17 try and approximately 9 staff-months for the NRC to review the additional 18 information.
19 The NRC staff considered the possibility of working with the ASME Code 20 Section XI working group to modify the code case to include the missing proce-21 dures and data.
However, given the number of plants that could need the guid-22 ance in the near term, and given the ASME codification process and the NRC's 23 process for endorsing ASME code cases, the time needed to modify and endorse 24 the code case was judged to be excessive.
25 The efficacy of the proposed regulatory guide procedures was demonstrated 3
26 by generic bounding calculations performed by the NRC staff in preparing 27 SECY-93-048.
These calculations demonstrated that the requirement in 3
28 Charles Z. Serpan, Jr., Memorandum to Jack Strosnider, January 15, 1993, 29
" Generic Bounding Analyses for Evaluation of Low Charpy Upper-Shelf Energy 30 Effects on Safety Margins Against Fracture of RPV Beltline Plate and Weld 31 Materials"; and Charles Z. Serpan, Jr., Memorandum to Jack Strosnider, Feb-32 ruary 8, 1993, " Additional Information Regarding Results of Generic Bounding 33 Analyses for Evaluation of Pressure Vessels Fabricated Using Low Charpy Upper-34 Shelf Energy Materials." Copies are available for inspection or copying for a 35 fee from the NRC Public Document Room at 2120 L Street NW., Washington, DC; 36 the PDR's mailing address is Mail Stop LL-6, Washington, DC 20555; phone 37 (202)634-3273; fax (202)634-3343.
R-4
1 Appendix G to 10 CfR Part 50 to demonstrate margins equivalent to those in 2
Appendix G of the ASME Code could be satisfied for materials with Charpy 3
upper-shelf energy less than 50 ft-lb for all the generic vessel geometries 4
and material combinations considered.
I 5
The acceptance criteria in the regulatory guide were taken directly from 6
the ASME efforts. The criteria were developed by the ASME Code Section XI 7
working group over an ll-year period and represent the collective judgment of 8
a body of experts representing the NRC staff, research contractors, nuclear 9
utilities, nuclear power plant vendors, consultants, and academia.
Similarly, 10 the evaluation procedures for Service Levels A and B were developed by this 11 group.
The procedures in the regulatory guide for Service Levels A and B are l
12 essentially identical to those in the ASME Section XI code case.
Thus, the 13 acceptance criteria and the evaluation procedures for the service levels that 14 generally control the analyses are based on the consensus technical opinion of 15 a large group of technical experts and were developed over an extended period.
16 The evaluation procedures for Service Levels C and D were developed by 17 the NRC staff and build on the procedures for Service Levels A and B.
As part 18 of a continuing effort by the ASME Section XI working group, the NRC staff has O
19 compared the regulatory guide procedures to other procedures that are being d
l 20 developed by various organizations. The comparison was very favorable, with i
I 21 the procedures proposed in the regulatory guide predicting lower acceptable l
l 22 Charpy upper-shelf energy values than would be predicted by the other I
23 procedures, which were less rigorous and, consequently, more conservative.
l 24 The transient selection procedures are based on procedures that have 25 already been endorsed by the staff.
Alternatively, generic bounding 26 transients can be used if justified.
27 The guidance on material properties is based on a state-of-the-art 28 statistical evaluation of all available fracture toughness data.
A broad 29 range of alternatives is offered in the regulatory guide so that methods 30 acceptable to the NRC staff are offered for virtually every situation and 31 combination of circumstances.
32 The proposed regulatory guide provides timely, cost-effective guidance 33 that is based on the consensus of a large group of technical experts repre-34 senting diverse backgrounds and interests.
The guidance is comprehensive and 35 would provide an effective and definitive approach to performing equivalent 36 margin analyses.
R-5
t O
t Printed on recycled paper O
Federal Recycling Program l
l
UNITED STATES FIRST CL ASS Mall NUCLEAR REGULATORY COMMISSION POSTAGE AND ff ES PAID WASHINGTON, D.C. 20555-0001 usNnc i
PERMIT tJo. G 67 l
OfflCIAL BustNESS PENALTY FOR PRIVATE USE, $300 I
US NPC-CTR4 DOCUMENT 1
D"CtJ"E'!T C n N T r< C L f!r e r c s "1-1240 CON T R CL Dreg WaSHINGTCN OC "rE55 l
l l
.......