ML20247G091
| ML20247G091 | |
| Person / Time | |
|---|---|
| Site: | Brunswick  |
| Issue date: | 03/08/1989 |
| From: | Giannuzzi A, Gustin H, Riccardella P STRUCTURAL INTEGRITY ASSOCIATES, INC. |
| To: | |
| Shared Package | |
| ML20247G066 | List: |
| References | |
| EER-89-0055, EER-89-0055-R01, EER-89-55, EER-89-55-R1, NUDOCS 8904040120 | |
| Download: ML20247G091 (54) | |
Text
U Yr E R. N 8@
\\.
- w. \\
? acy b 1
JUSTIFICATION FOR CONTINUED OPERATION OF THE CAROLINA POWER & LIGHT BRUNSWICK STEAM ELECTRIC PLANT UNIT 2 STRU
>w<
ASSOCATES3C l
l l
3150.
ADEN EXPRESSWAY SUITE 226 e SAN JOSE, CAIRDRNIA 95118 e (408) 978-8200
- TELEX 184817 STR 8904040120 890313 PDR ADOCK 05000324 P
PNU
_ _ ~ - - _ - - - ~ - - - - -
EER B") - co S5
'2.e4
\\
7
?W
~
Report No.: SIR-89-008 Revision No. O Project No.: CPL-02Q-2 March 8, 1989 JUSTIFICATION FOR CONTINUED OPERATION j
OF THE CAROLINA POWER & LIGHT BRUNSWICK STEAM ELECTRIC PLANT UNIT 2 Prepared by:
Structural Integrity Associates f
Prepared for:
Carolina Power & Light Co.
3 !I E7 Prepared by:
Date H.
L. Gustin
/
Date Nh Prepared zi Reviewed by:
/1 1
Date 3 % @9 C.
Riccarde'lla
^
3!8!b3 Approved by:
Date D. R.
Pitcairn
Rt+ i P3
?
TABLE OF CONTENTS Section EjlLqg
1.0 INTRODUCTION
1-1 2.0 FLAW EVALUATIONS AND ANALYSES 2-1 2.1 Safe End Materials 2-1 2.2 Stress Components 2-1 2.3 Allowable Flaw Size Determination 2-2 2.4 Flaw Growth Calculations.
2-3 3.0 DISCUSSION OF SAFETY MARGINS IN ANALYSES 3-1 4.0 BASIS OF CRACK GROWTH LAW FOR INCONEL 600 4-1 4.1 Development of Inconel Crack Growth Rate Curve.
4-1 4.2 Correlation of Crack Growth Rate Curve with Field Observations 4-4
5.0 CONCLUSION
S 5-1
6.0 REFERENCES
6-1 SIR-89-008, Rev. O i
gg ASSOCWEINC
$LY Of '
LIST OF TABLES Table Pace 2-1 BSEP Unit 2 - Applied Stress Summe.ry C Riser, Thermal Sleeve Attachment Weld 2-6 2-2 Allowable Flaw Size Calculations Using IWB-3641 Source Equations 2-7 2-3 Flaw Growth Time Predictions 2-8 4-1 Test Data for Crack Growth of Nickel Base Alloys in Pure Water at 288*C.
4-8 SIR-89-008, Rev. O 11 ASSOCIATESINC
F_.ER. B9 - oo S5 Rev.1
(*
Pa}e la J
LIST OF FIGURES Ficture Eggg 1-1 Thermal Sleeve-to-Safe End Geometry.
1-2 2-1 Finite Element Model of Recirculation System For Use in Determination of Weld Overlay Shrinkage-Induced Stresses.
2-10 2-2 Residual Stress Distribution at BSEP and DAEC Thermal Sleeve Attachment Weld Locations [3]
2-11 2-3 Stress Intensity Factors Used in Flaw Evaluation Analyses 2-12 2-4 Flaw Observed in BSEP Unit 1 Riser F Thermal 4
Sleeve to Safe End Weld Location.
2-13 2-5 BSEP Unit 2 Evaluation Cases, 360* Flaw (No MISP) 2-14 2-6 BSEP Unit 2 Evaluation Cases, 360* Flaw (MISP) 2-15 4-1 Crack Growth Rate Data 4-9 4-2 Crack Growth Rate Data for Inconel 600, 82, 182 4-10 4-3 General Electric Company Proprietary Information.
4-11 4-4 Furnace Sensitized Alloy 600 SCC Growth Rates 4-12 1
4-5 Crack Growth Rato Data for Inconel 600, 82, 182 Including GE Proprietary Data.
4-13 4-6 Comparison of Predicted and Observed Crack Growth Rates Vs. Stress Intensity.
4-14 4-7 Comparison of Predicted and Field Growth, BSEP Unit 1 Riser F and DAEC 4-15 SIR-89-008, Rev. O iii 6
ASSOCMTESINC
gggs bus u'w@
1.0 INTRODUCTION
kg U
During the 1988-89 maintenance / refueling outage at Carolina Power
& Light's Brunswick Steam Electric Plant Unit 1,
weld overlay repairs were applied to seven of the ten recirculation inlet nozzle-to-safe end welds.
The details of the design of these repairs are discussed in Reference 1.
The repairs and a portion of the nozzle and safe end material were examined ultrasonically following repair completion, using both shear wave and refracted longitudinal wave techniques.
The refracted L wave examination of the Inconel safe ends upstream of the weld overlay repairs detected cracking originating at the inside surface of the thermal sleeve-to-safe end wcld in all of the inlet safe ends, including those three safe ends to which no repair had been applied.
The component geometry, including identification of the flawed location, is illustrated in Figure 1-1.
Since the BSEP Unit 2 plant is essentially identical to the BSEP Unit 1 configuration, and since prior UT examinations did not utilize the refracted L
wave technique which detected the cracking in Unit 1,
it is reasonable to consider the potential for the existence of similar undetected cracking in Unit 2.
The purpose of this report is to determine the effects of such potential on the safety of continued operation of Unit 2 until the next scheduled refueling
- outage, currently planned for September, 1989, and to demonstrate that continued operation of Unit 2 uncil the September outage is justified.
The approach to justifying continued Op6 cation of Unit 2 is to assume that the most severe flaw found in Unit 1 also exists in the most highly stressed riser in Unit 2.
With the above assumption, a
fracture mechanics crack growth analysis is performed to demonstrate that the postulated defect will not grow to an unacceptable size before the next scheduled outage.
The crack growth law employed for this analysis, based on laboratory data, was benchmarked to the cracking experienced in BSEP-1 and SIR-89-008, Rev. 0 1-1 ASSOCIATESINC L_ _ __ __
EEE B9'o 55 R.e v. I t '2-p,9 DIEC Inconal thermal slesvo-to-safo and
- walda, in ordar to further validate its use.
It is felt that the above justification for continued operation of Unit 2
is quite conservative, since both mechanical stress improvement (MSIP) of these safe ends and hydrogen water chemistry (HWC) controls to mitigate IGSCC have been implemented at Unit 2.
SIR-89-008, Rev. 0 1-2 M
ASSOCWESINC
L 7
Z 8
YZ OO 6
LS V
E E
L S
NI I
N m
IT E
\\m R
E F
5 A
7
~
y S
3 m
r L
t 6
E e
N p
m O
o C
m e
N G
I d
s D
n EL E
/,pRA VE EW e
5 f
ELT' 2
a S
1 SN
~
E o
1 LM t
AI I
MC ev
)
ET e
1 iT l
e TA T
l W]
S AN LU l
F(
a m
DN r
EO e
VI h
~
1 y7 e
R r
u 4
g 8
9 i
2 F
6 5
0 N7:A,'
o y8
EE32. fa - 00 55 ha. \\
'2.O FLAW EVALUATIONS AND ANALYSES The observed flaws on Unit 1 were assumed to be representative of the flaws which could exist on Unit 2.
These flaws appear to originate at the crevice formed by the thermal sleeve to safe end attachment weld, as illustrated in Figure 1-1.
In order to determine the potential effects of similar flaws on the safety of continued operation of Unit 2,
fracture mechanics flaw evaluations were performed, assuming the worst case Unit 1 flaws l
existed on Unit 2.
2.1 Safe End Material The recirculation system inlet safe ends are Inconel 600 material.
The safe ends were welded to the nozzle butter following the vessel post weld heat treatment.
As cracking progresses into the safe end, it grows into mill annealed material.
The flaw evaluations described in this report used material properties for this type of material.
2.2 Stress Components The analyses used stress contributions from applied stresses due to piping loads and operating conditions taken from Reference 2.
The applied stress components assumed in the analyses are summarized in Table 2-1.
These are representative of the C riser, which had the highest combined loads of any of the ten nozzles.
A finite element analysis of the BSEP Unit 2 recirculation system was performed to evaluate the effects of weld overlay induced shrinkage stress on the locations of concern.
The
- model, illustrated in Figure 2-1, included the reported axial shrinkages SIR-89-008, Rev. 0 2-1 ASSOCIATESINC
U2g gs 'tyv m
bvI pq g
t
.from all wold ovorlays in tho cyctam.
Tho 28-inch rceirculation outlot nozzles were assumed to be rigidly fixed in the model.
The 12-inch recirculation inlet nozzles were assumed to be constrained by appropriate stiffness boundary elements in each of l
the six degrees of freedom.
Stresses at the locations of concern were determined from the cutput forces and moments at the model nodes nearest to the locations of thermal sleeve attachment j
The maximum value of the calculated weld overlay shrinkage stress for thermal sleeve attachment weld locations (10 total) on Unit 2 (6.24 ksi) is included in Table 2-1, and was used in the evaluation of hypothetical flaws at the limiting thermal sleeve ; attachment. region on Unit 2.
The residual stresses due to the welding of the thermal sleeve-to-safe end attachment were taken from Reference 3.
These stresses were derived from elastic-plastic thermal stress analyses performed for the Brunswick system and geometry in response to the flaws identified at Duane Arnold Energy Center (DAEC) in 1979.
Figure 2-2 presents the axial residual stress distribution through the safe end wall at the thermal sleeve attachment weld location.
Similarly calculated stresses for the DAEC geometry are also shown for comparison.
2.3 A.L1owable Flaw Size Determination Review of Table 2-1 shows that the primary membrane plus bending stress ratio for the limiting location on Unit 2 is very low (approximately 0.25).
For this case, the allowable flaw depth in the component is 75%
of component wall thickness.
This is determined by reference to ASME Section XI IWB-3640 and Appendix C to Section XI.
In fact, the source eg Lons which underlie the tables in IWB-3641 (contained in Appendix C) of Section XI would allow a much deeper flaw were it not for an artificial 75%
limit which is imposed in the tables.
Calculations have been performed, as presented in Table 2-2, using the source equations, SIR-89-008, Rev. 0 2-2 ASSOCIATESINC
, w_ v
~ -
f& \\
9Cy with a cafoty factor of 2.773 on cpplicd lond, cnd racult.in an allowable crack depth of 86% of the safe end wall thickness.
i As illustrated in Figure 2-4, the deepest of the flaws observed in the vicinity of the thermal sleeve attachment welds in Unit 1 was approximately 61% through-wall, with an average depth around the circumference of 49%.
Thus there would be significant margin between the observed and allowable flaw depths, if the Unit 2 cracking were the same as that observed in Unit 1.
Consequently, the observed flaws would be acceptable without repair today.
In order to determine whether the bounding flaws would grow to an
. unacceptable level prior to the next refueling outage (scheduled j
to begin in approximately 6 months),
fracture mechanics crack growth calculations were performed, as.ciiscussed below.
2.4 Flaw Growth Calculations The observed flaws are assumed to have been initiated by a crevice corrosion mechanism, with -transition to intergranular ctress corrosion (IGSCC) when the flaw size becomes large enough to produce a significant crack tip stress intensity factor.
Conservatively, a bounding IGSCC growth law from Reference 4 for Inconel material was used in the analysis to bound the crevice corrosion rates.
This rate relationship is:
-8 K.26 2
da/dt =1.078 x 10 The origin and justification of this crack growth expression is discussed in Section 4.0 of this report, along with a correlation of this law to observed field crack growth rates at this location at BSEP-1 and DAEC.
The crack growth analyses were performed using the SI-developed computer program pc-CRACK [5).
Calculation of stress intensity factors for each loading condition or stress distribution is SIR-89-008, Rev. 0 2-3 ASSOCIATESINC
g 99 owesD Rcv l Paey ?
psrformad internal to the program.
A p l c *. which precenta the stress intensity factors K as functions of distance through the wall due to each contributing load or stress is included in Figure 2-3.
The hypothetical cracking evaluated on Unit 2 was assumed to be similar to the observed cracking on Unit 1.
That is, the worst case flaws from Unit 1 were used as the bounding flaws for the Unit 2
evaluation.
The Unit 1
Riser F
flaw was mapped ultrasonically, and is illustrated in Figure 2-4.
This flaw had a maximum depth of 61%, and an average depth of 49%.
This worst case flaw vas used for evalut. tion of hypothetical flaws of Unit 2.
The analysis assumes a 360*, circumferentially oriented crack in a pipe with thickness to radius ratio of 0.2 (actual t/r at the observed flaw location in Unit 1 is 0.176).
Applied stresses due to internal pressure and piping loads were taken from Reference 2,
and the worst case stresses in the vicinity of the thermal sleeve attachment weld (Riser C) were used in the analyses (Table 2-1).
The largest value of weld overlay induced shrinkage stress predicted at any of the inlet safe end locations was also used in the analysis.
In addition, the residual stress distribution derived from Reference 3 for the Brunswick system was used.
The combination of worst case stresses and worst case Unit 1 flaws is assumed to conservatively represent the limiting location on Unit 2
for the purposes of evaluating the hypothetical flaws which could exist on Unit 2.
Three cases were analyzed.
These were:
1.
As-welded residual stress, bounding crack growth law 2.
As-welded residual stress, realistic crack growth law based on BSEP-1 and DAEC field observations SIR-89-008, Rev. 0 2-4 6
@JC
RMi g
3.
As-waldad residual stress, factor of 5 improv in crack growth due to HWC In addition, the same three cases were also run for an assumed zero residual stress pattern to take partial credit for the MSIP treatment of these welds performed in 1988.
The analytical crack growth predictions are illustrated in Figures 2-5 and 2-6, with the predicted times to grow the assumed flaw to the allowable value for each case summarized in T' ble a
2-3.
The times in Table 2-3 are based on starting with a 360*
flaw. equal to the worst average depth of flaw observed in Unit 1, which as in Figure 2-4 is 49% through-wall.
Use of the 61% peak through-wall would yield shorter times, but is considered to be overly conservative because of the 360*
model used in the analysis.
Under these assumptions, Case 1
above predicts grevth to
}
allowable depths in 5-7
- months, which indicates marginal acceptability for the remainder of the fuel cycle.
It should be noted, however that the crack growth correlation for Inconel used in the analysis is a very conservative one.
An attempt to i
benchmark the crack growth correlation against the observed flaw growth on Unit 1 and DAEC is discussed in Section 4.2.
The benchmarking leads to a predicted reduction in crack growth rate by approximately a factor of 3,
which leads to an allowable period of continued operation of 17 to 23 months.
Also, BSEP-2 is currently operating on hydrogen water chemistry (HWC), which yields still a further reduction in expected crack growth rates (factor of 5 used in this analysis).
For all but the Case 1 crack growth law, even assuming the maximum local crack depth obser-ed in Unit 1 (61%) with the 360* model would result in an acceptable period of continued operation.
SIR-89-008, Rev. 0 2-5 ASSOCATES INC
gg g,,,gg Rev i n
t9 y
Table 2-1 BSEP UNIT 2 - APPLIED STRESS
SUMMARY
C RISER THERMAL SLEEVE ATTACHMENT WELD PIPE CD (in) 15.00 PIPE THICKNESS (in) 1.125 PIPE ID (in) 12.75 PRESSURE (psi) 1000.00 (operating) 1325.00 (design) 2 X-SECT AREA (in )
49,o4 3
SECT. MOD (in )
158.38 Stress Type Fx My Mz Tot. Mom.
Axial (kip)
(in-kip)
(in-kip)
(in-kip)
Stress (ksi)
PRES (design) 3.45 PRES (operat) 2.60 DW 1.10 10.80 102.26 102.83 0.67 OBEl,x 1.49 39.54 196.2' 200.21 1.29 OBE2,y 0,22 4.38 44.11 44.32 0.28 OBE3,z 1.32 31.19 164.51 167.44 1.08 COMB.OBE 1.71 43.93 240.37 244.35 1.58 Thermal 1.63 256.76 257.21 3G3.43 2.33 Shrinkage 8.86 5.33 959.10 959.11 6.24 Pm 2.60 Pb 2.25 SIR-89-008, Rev. 0 2-6 ASSOCIATESINC
EEP_ se ooSS R e+ l to TABLE 2-2 Pau-0 ALLOWABLE FLAW SIZE CALCULATIONS USING IWB-3641 SOURCE EQUATIONS OUTPUT. TEX Thursday, March 2, 1989 Page 1 ALLOWABLE FLAW SIZE EVALUATION
{
ALLOWABLE FLAW SI2E USING SOURCE EQUATICHS FOR CIRC
(
CPLO2Q UNIT 2 JCO:
ALLOWABLE WITH CPERATING PRESSURE = 1000 WALL THICKNESS:
1.1250 l
MEMBRANE STRESS =
2.6000 SAFETY FACTOR =
2.7730 BENDING STRESS =
2.2500 SAFETY FACTOR =
2.7730 STRESS RATIO:
0.5772 ALLOWABLE STRESS: 23.3000 FLOW STRESS:
69.9000 L/CIRCUH 0.00 0.10 0.20 0.30 0.40 0.50 0.80 0.70 ALLOWABLE A/T 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 0 9541 0 9092 t
0.80 0.90 1.00 ALLOWABLE A/T 0.8898 0.8838 0.8838 b
9
{
ALLOWABLE FLAW SIZE USING SOURCE EQUATIONS FOR CIRCUM 0t CPLO2Q UNIT 2 JC0; ALLOWABLE WITH DESIGN PRESSURE = 1325 WALL THICKNESS:
1.1250 MEMBRANE STRESS:
3.4500 SAFETY FACTOR:
2.7730
()
BENDING STPISS:
2.2500 SAFETY FACTOR:
2.7730 STRESS RATIO:
0.6784 ALLOWABLE STRESS: 23.3000 FLOW STRESS =
69.9000 L/ CIRCUM 0.00 0 10 0.20 0.30 0.40 0,50 0.60 0.70 ALLOWABLE A/T 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 0.9287 0 8857 I
0.80 0.90 1.00 1
ALLOWABLE A/T 0.8662 0.8823 0.8823 SIR-89-003, Rev. 0 2-7 6
fg_g. M -ooSS T2.e.V I
'2.1 TABLE 2-3 p3 FLAW GROWTH TIME PREDICTIONS Predicted Time to Grow From Worst Unit 1 Average LOAD 360* Crack Depth (49%) to Allow. Flaw Size CRACK CASE 360' Crack 360* Crack I
As-Welded Residual Zero Residual GROWTH Stress (No MSIP)
Stress (MSIP)
LAW Allow = 75% Allow = 86% Allow =75%
Allow = 86%
Bounding Law 5 months 7 months 15 months 18 months Fiold Correlated Law 17 months 23 months
> 5 years
> 5 years HWC Law 26 months 35 months
> 5 years
> 5 years 1
a l
$LR-89-008, Rev. 0 2-8 M'M 5Amt I.-
1
EE1L ga -oe55 ken. k p-p9 _
A
/
b i
r
~
[
f if
[
N, h
l
\\
)
l s
/
r
/
\\)
I-Figure 2-1.
Finite Element Model of Recirculation ti System for Use in Determination of Weld overlay Shrinkage Induced Stresses l
l I
f SIR-89-003, Rev. 0 2-9 R,
" L., " ;
o
-_m
_m Rev. \\
p Y ~13 Outside t Surface l
t i
3t/4 i
i ft
=E 4
i i
i 5%
I TH i
=. ?!
u=
i g 'i l
t/:
\\
=h S
8 i
l 2MC E U l
[
i i
E
-?s -
d I
i m
E5 3RUNSWICI I
t/4 I
u
'3 i
I I
l Inside Surface 0 i
-60
-40 00 0
00 40 60 RESIDUAL STRESS (ISI)
Figure 2-2.
Residual Stress Distribution at BSEP and DAEC Thermal Sleeve Attachment Weld Locations (3)
SIR-89-008, Rev. 0 2-10 N
ASSOCWESINC
.- g g u m Fs t-W 5 5 R e+ 1 g
92 7 -
k b
m 3
m U
m
_a 03 o
m m
r%
m o
O 8
B O
m e
a v
m 21 U
w m
o
_=
x Z h 23 o
QC m
3 Ha m
2
+
2 w
OW t
.5 g
n b-c
+m
/
m
=
/
m O
/
3 m
be, d
/
+
m w
a
+
m A
W e
g m
~d bi
~
eu 3
m cd o
Z m
- e z W 9-8 m
o pm
+m I
N ZS -
gm u+
3
~y 02 m
(/)
El 4
n m-d M
U1 h
O m
H m
a G3 t
u E
-m m
D 7
e
~
i i
i i
i i
i i
i i
i i
i i
i i
iO e
4$$$$$N$$$
U3***"O
.?
w S*0 HONI IS>l SIR-8 9-008, Rev. 0 2-11 l
\\
1
i b
t 1
5 0
4 f
7 n-
=
3 w
f a
I l
f l
S e
U l
l baw 1
o g
F t
h i
R T
l l
t l
b E
d 1
l a
T l
o h9 N
P W
E t4 C
e p
v ey 8
I D
Pe dl A
le e
E i
ES et i
D R
l ga 3
t Sl am O
a P
1 Brm r
i 0
T e x 2
t M
i e
v o O
n h
a r q
i R
U T
p F
p P
a T
E S
S I
B D
n i
d t
3L e
g-vr r
t e
e s
b t
O w
q a
lP
, i 0
g 4
0 0
0 0
0 0
0 0
0 2
9 8
7 5
4 3
2 1
er ug i
y o
P 3
E wb
.3
m5 C
W j
H o
/
- 4
- 0 g
v g
+
+
+
+
+
+
0
+
i 3
/
M,
)
s d)TA n SL asR E
.T l i
iI
- 'i'l i
,I
,1 uU R o0 R
'iI I I I
h1 O 1
((C T
ED 0
ML E
a i
I t
2 T1
.r F
h 8
M 0
7 0
1
+
0, >
3 =
i 1
i+:i,g o
i m I'
+
3 r
+
6 9
0mo 1
+
r4
+
r h7 i
+p/
m0 y0>=
02 0
D 0
g L
2 E
~
f l
5 W
4 t
E SA
~
0 O
9 7
6 5
4 3
1 O
I 0
~
0 O
0 O
0 O
nb otO wp EH
~ toS y o n
e
W~--------__--_
~o D
4>
1>
I 4>
<>o
(
i O
M l
o l
o i
D o
ib
- o
- A>
g g
i l
o Z
1>
)
g
~
1
$.1; 7
s$
x 4>
==
-> (>
0
- s. o 89 -
3 o
3D y)'
6 OQ e
1
--4) hC%
g) o o
v E
=<>
WO o
o Q
20 c
N p
9 o
4>
m o
A>
[
M>
3 4>
e4 g>
c
~
o g>
+
D3
(>
t>
N 1) t>
M l>
o I
d
-l>
P C
C a) b b
N 5
M 2
d l
i C
i i
i i
i i
i i
.- o n
$o o.
b.
N D
e o
o o
o o
o o
3 O
w e
a s=0 b
(u) H.Lasc w ti SIR-89-008, Rev. 0 2_14 i
arrw arrv
2cw-l l
pa
. 2 % ---
3.0 DISCUSSION OF SAFETY MARGINS IN ANALYSES The above analyses demonstrate that Unit 2 can continue to be operated safely until the next refueling outage (approximately 6 months), assuming that the thermal sleeve attachment weld region of the safe end contains flaws similar to those detected on Unit 1.
The arguments are based upon crack growth predictions of the hypothetical flaws presented in Section 2.0 of this report.
Other arguments also support this position.
These include:
- 1. The safe end at the location of interest has a
wall thickness which is almost double that of the adjacent pipe wall (1.125-inch vs. 0.628-inch).
In the location of the postulated bounding
- flaw, the remaining ligament is comparable to the nominal pipe wall thickness.
If the flaw were arrested at its current depth, the section modulus and cross sectional area of the remaining ligament are comparable to those of the pipe wall adjacent to the flaw location, which must sustain the same loading.
Since net section collapse is the prevailing failure mode for the ductile wrought Inconel safe end material, this demonstrates that design basis safety margins would not be degraded by the observed Unit 1 cracking.
- 2. The bounding crack growth law used in the Section 2.0 analysis is conservative.
A modification of this law to correlate with actual observed flaw growth at DAEC and BSEp Unit 1 is discussed in Section 4.2.
This correlation would predict growth rates approximately 1/3 as fast as those predicted with the bounding crack growth law in Section 2.0.
The resulting times to reach allowable flaw size are considerably larger with the field-correlated crack growth law.
SIR-89-008, Rev. 0 3-1 DrG,GRITY awy sm w
Er1L B") - 0055' R.ca. I 9aaf-2A _
- 3. Mechanical Stross Improvement (MSIP) was applied to all 10 thermal sleeve attachment weld locations on Unit 2
in Spring, 1988 [6).
MSIP has been shown to be effective in producing compressive residual stress distributions at treated locations.
The analyses summarized in Section 2.0 conservatively ignored the beneficial effects of this process on crack growth at the thermal sleeve attachment weld location.
A second set of analyses presented in Section 2.0 assumed that residual stress was reduced to zero by MSIP.
These results show significant improvement in predicted crack growth.
In
- fact, even more favorable residual stress improvement would be expected.
- 4. A hydrogen water chemistry system was placed in operation at BSEP Unit 2 in late February, 1989, and will be used during the balance of the fuel cycle.
Hydrogen water chemistry has been shown to be effective in retarding the initiation and growth of IGSCC flaws.
It is also expected to be effective in retarding crevice corrosion mechani~sms, since the reduction of oxygen to very low levels due to recombination with hydrogen effectively
" turns off" the differential aeration
- cell, which is the
" battery" which drive.,
the aggressive ionic species into the crack.
Combined with the bulk diffusion out of the crevice, the crevice chemistry is eliminated in the crack.
A reduction in the crack growth rate of a factor of at least 5 is expected to be achieved.
~
Section 4.0 discusses this effect in greater detail.
5.
Finally, even if a flaw continued to grow and eventually propagated 'brough the safe end wall, or were deeper than the obsen ad Unit 1
flaws at this
- time, the predicted failure mode would be one of leak-before-break.
This expectation is supported by the experience with flaws in a similar location at
- DAEC, which did in fact cause a
detectable leak which was detected by plant safety systems, leading to plant shutdown, prior to structural failure of SIR-89-008, Rev. 0 3-2
EEft e-coSS R ev. t hof 3,
the pip 2.
Since the materials and geometries of the BSEP and DAEC components are
- similar, leak-before-break assumptions are justified for this flaw location.
l i
SIR-89-008, Rev. 0 3-3 INTE,GRITY acerv mc ~e
EE.Y-. M -@a@
R.ev - l Pay - ?>L_ --
4.0 BASIS OF CRACK GROWTH LAW FOR INCONEL 600 The Nuclear Regulatory Commission issued NUREG-0313 Revision 2,
and Generic Letter 88-01, in January, 1988, in order to provide materials selection and processing requirements for austenitic stainless steel components in BWRs (7,
8].
These documents provide the requirements for acceptance of austenitic stainless steels containing IGSCC-like non-destructive indications for continued operation in BWRs.
The guidance for acceptance is based in part on the use of an NRC approved crack growth rate for austenitic stainless steel, illustrated in Figure 4-1 curve
[7).
While the NUREG provides detailed technical requirements for continued use of austenitic stainless steels containing; IGSCC-like indications, it states that the use of nickel-base alloys in BWR water will be evaluated on an individual case basis.
No crack growth rate curve similar to Figure 4-1 is recommended in the NUREG for nickel-base alloys.
Consequently, continued operation of nickel-base austenitic materials containing IGSCC-like indications requires the development of such a crack growth rate curve.
The purpose of this section of this report is to illustrate the basis for development of that crack growth rate curve for nickel based components used in the BWR.
4.1 Development Of Inconel Crack Growth Rate Curve The development of a crack growth rate curve for nickel-base alloys required a review of open literature and proprietary data to develop typical and bounding crack growth rate curves in a
'1 manner similar to that presented in Figure 4-1 for austenitic stainless steels.
Figure 4-2 presents the open literature Inconel 600 crack growth rate date as a
function of stress l
intensity.
Included on this curve are data points for Inconel 82 SIR-89-008, Rev. 0 4-1
Rev1,
OY and Inconal 182 weld
- matals, whore appropriate to the understanding of the Inconel 600 data.
Figure 4-2 shows that a rGpresentativo crack growth rate curve can be drawn bounding the available data.
The imREG-0313 Revision 2 curve for austenitic ctainless steel is also presented on Figure 4-2 for comparative purposes.
Table 4-1 presents the actual test conditions for the open literature crack growth data illustrated in Figure 4-2.
Table 4-1 shows that the test data include both bolt-loaded and live loaded fracture mechanics tests, in the laboratory and in roactor.
The material condition is generally a
severely sensitized condition, as contrasted to the Inconel 600 safe ends at Brunswick Unit 2, which are in the mill annealed condition.
Tho water chemistry includes high oxygen tests with a very high impurity level and water conductivity (1
ppm sulfuric acid providing a water conductivity of 8 s/cm, an oxygen level of 7 ppm and a pH of 4.8, (4]), and in reactor tests, as well as tests in high purity water containing 200 ppb oxygen (7, 8].
The curve bounding this data is the bounding crack growth law used in the flaw growth analyses, described in Section 2.0 above.
The top five data points in Table 4-1 represent displacement-controlled tests in which average crack growth data is used to predict crack growth rates.
The remaining five data points represent live-load instantaneous crack growth rate laboratory data.
Following the initial literature review, Carolina Power & Light staff were informed by General Electric Company that additional, GE proprietary crack growth data for Inconel 600 in BWR environments would be made available to CP&L in order to augment the current data base.
That data was made available to structural Integrity for use solely in support of CP&L during the NRC review of BSEP Unit 1 and 2 thermal sleeve / safe end cracking concerns.
The data is presented here as Figures 4-3 and 4-4.
The proprietary data include laboratory and in-reactor data, p;rformed on live-load specimens measuring instantaneous crack growth using the GE crack arrest verification system (CAVS) i approach.
The data in Figure 4-3 are for low temperature SIR-89-008, Rev. C 4-2 6
6ccmSM W
y g.,ggC iW * \\
Pa r e
concitized Inconel 600 tested in 200 ppb oxygen water in the laboratory.
The water conductivity for these tests was 0.3 to 0.5 S/cm.
The data in Figure 4-4 include both laboratory and in reactor test result.s on furnace sensitized Inconel 600 in various environments including the hydrogen water chemistry environment at an operating power plant.
Figure 4-4 shows that at low water l
conductivity (excellent water quality), the crack growth rates are significantly slower than in less pure water.
In the hydrogen environment the crack growth rate is a factor of 5 to 10 less than in the normal water environment at the same water conductivity level.
In less pure water containing eggressive species (believed to be sulfate ion), the crack growth rate is of the order of hundreds of mils par year.
Upon combining the GE proprietary crack growth rate data from i
Figures 4-3 and 4-4 with the open literature curve of Figure 4-2, i
one observes (Figure 4-5) that the crack growth rates indicated by some of the proprietary data fall well above the open literature Inconel curve, particularly at low stress intensities It is believed that the use of live loads in the GE tests, the use of severely sensitized Inconel 600 in the GE tests, and the less pure water used in the GE tests may explain the higher crack growth rates exhibited by these data.
The effect of water purity on the crack growth rates of Inconel 600 is illustrated by Figure 4-6 taken from Reference 14.
Figure 4-6 shows that changes in water purity of from 0.1 to 0.3 or from 0.3 to 0.5 S/cm results in increases of IGSCC crack growth rates in Inconel 600 of one-half an order of magnitude or more.
The crack growth rate law used in this analysis is the curve presented in Figure 4-2.
From Figure 4-5, it is seen that this does not represent a bounding curve for all Inconel 600, when the GE proprietary data are included.
Nonetheless, it is believed to be a realistically conservative crack growth rate curve, given the material condition, type of loading on the thermal sleeve attachment weld at Erunswick and the water chemistry conditions SIR-89-008, Rev. 0 4-3 6
l m_____
w w w-v o w w 2)
RevI Pay-
'N prccont in the th9rmal sleeve attachment welds at the Brunswick units.
Note in Figure 4-5 that the three results representing in-reactor CAVS fall significantly below the bounding SCC growth curve, and are more representative of in-reactor water chemistry conditions.
4.2 Correlation of Crack Growth Rate Curve With Field Observations In previous sections of this report, a crack growth rate law which described IGSCC in inconel 600 (4) has been presented and applied in the evaluation of hypothetical flaws in the thermal alcove attachment welds at BSEP Unit 2,
as a portion of the justification for continued operation (JCO) for that unit.
The purpose of this section of the report is to determine whether that law reasonably predicts the observed
- cracking, and to "bonchmark" it against the observed crack growth rates at BSEP Unit 1 and at Duane Arnold Energy Center (DAEC).
The relation under consideration is
= 1.078 x 10
?g, 6 (in/hr) t for Inconel 600 and weld metals I-182 and I-82.
'or the BSEP-1 worst flaw (F riser) the total stress intensity
! actor is the sum of contributions due to pressure, bending
- daadweight,
- thermal, shrinkage) and residual stresses.
K calculations were run using pc-CRACK and assuming a 360* flaw.
The operating pressure of 1000 psi was assumed, a unit bending casa (scaled up for actual loading) was used, and the residual etross distribution for BSEP thermal sleeve attachment weld locations was taken from [2].
The total K is determined from K (pressure) + 9.24 K (bending)
+K residual "
total SIR-89-008, Rev. 0 4-4
mm wy ~m R.CV' q)07 36 A kcy accumption in correlating crack growth to field exparience in situations such as this is the portion of operating time to ob:erved failure associated with initiation (versus growth), and tho size of the flaw once initiated.
For purposes of this anclysis, it was assumed that crack initiation of crevice Inconel I
I 600 might reasonably be expected in operating times of
(
approximately one year, and that the flaw size, when initiated would be large enough so as to sustain an applied K greater than tho IGSCC crack growth threshold.
Based on the crack growth data precented in Section 4.1 of this report, a reasonable IGSCC crack growth threshold for Inconel in a BWR environment is on the order of 15 ksi M.
}
If bhreshold is assumed to be 15 ksi M, then the initiated flaw aize for IGSCC growth would be approximately 0.018 inch.
Using I
the load combination presented above, a prediction of the time rsquired for an assumed initial flaw of this magnitude to grow to ths worst observed flaw depth in Unit 1 (F riser) was performed I
uning pc-CRACK.
The results are presented in Figure 4-7.
Using tha above crack growth relation, the time required to grow a l
0.018 inch initial flaw to the maximum observed crack depth af O.685 inch is approximately 1.64 years.
since Unit 2 has been operating for 6.8 EFPY and Unit 1 6.4 EFFY, Thsce results suggest that either:
1.
The observed flaws took approximately 4.5 years to initiate, then grew at the bounding rate, or 2.
The flaws actually grew at a slower rate than assumed, with relatively rapid initiation.
For purposes of this correlation evaluation, an initiation time of 1 year will be assumed.
This leads to an actual growth rate i
of (0.685 in.-0.018 in.)/5.4 year, or 0.124 in./ year, versus the SIR-89-008, Rev. 0 4-5 6
am am nr I
Rat - \\
2y prcdicted rato of (0.685 in.-0.018 in.)/1.64
- years, or 0.407 in./ year.
There is, therefore, a potential conservative factor of 3.29 on predicted crack growth in the bounding crack growth law.
A revised crack growth correlation has been prepared, in an attempt to more closely match the crack growth rate associated with observed flaws at BSEP-1.
The revised law was developed by scaling the coefficient in the crack growth correlation as fallows:
" da ~
0.124 in./ year
~8 2
x K.26 x
1.078 x 10
- dt new 0.407 in./ year
~9 2
K.26
= 3.271 x 10 If this new correlation is used in pc-CRACK, with the same stress assumptions as previously used, the predicted growth time is 45260 hours or 5.17
- years, which considering a
one year initiation time, is within the accuracy of the assumptions of this analysis.
This result is also shown in Figure 4-7.
If the proposed crack growth correlation
~9 2K.26 (in./hr)
= 3.271 x 10 is used with the DAEC geometry and residual stresses, but with BSEP loads (adjusted for geometry effects) assumed since DAEC loads / stresses are not well defined, the predicted crack growth time is approximately 37 months to grow through-wall or about 3 years.
This also compares reasonably well with the field observations, since ?.hrough-wall cracking was observed in about j
four years of plant operation.
)
on the basis of this evaluation, a second,
" field-correlated" crack growth law was useo to make remaining life predictions in Section 2.0 of this report.
These predictions are considerably SIR-89-008, Rev. 0 4-6 6
ASSOCIATFS WC
--Ed %9 - 0055 9su \\
sidwar than those gancrated using the bounding crack growth law, Pay.
p and are believed to be more representative of what might actually be expected in the field, in the absence of hydrogen water chemistry.
l SIR-89-008, Rev. 0 4-7
> r> : 4 AEAmTAP
l 11 k O*%V 1
[P e
c
)
)
)
)
)
,* 4 (Il
)
n
)
]
]
0 0
3 a
1
]
3 t
e 4
4 4
1 1
1 l
1 9
1 r
(
[
[
[
[
[
[
[
[
(
e f
m e
c R
/
r r
S 4
4 o
o m
m m
p m
O O
O t
t c
c c
c S
S S
c c
/
/
/
7
/
2 2
2 a
a S
S S
S H
H H
e e
p p
0 p
R R
t m
m m
1 5
3 1
1 n
P P
p n
n e
P P
p i
i 0
0 0
0 0
mn 1
1 1
2 2
2 3
2 2
2 o
0 0
0 0
0 0
0 r
i 2
2 2
v 0
0 0
b b
b b
b b
b n
p p
p p
p p
p E
e m
a p
p p
p p
p p
p p
p p
p p
0 0
0 0
0 0
0 0
0 0
0 0
0 0
7 7
7 2
2 2
2 2
2 2
r s
h y
/
o
~
'0
'0
~
~
~
~
l n
0 0
0 0
0 l
e i
1 1
1 1
~
1
~
1 1
~
A gh x
x x
x 0
x 0
x x
0 at.
5 8
4 5
5 2
4 1
1 1
d rw e x
x x
e eot 4
3 2
2 1
7 3
1 2
9 s
vr a a
AG R B
le k
K cC a6 5
2 9
i*
l N8 9
1 1
8 ni 3
4 4
1 f2 i s o
F k 4
t ha E
t L
wr h
B oe t
A rt w
T Ga o
W r
8 0
5 k
G m cn m
4 4
2 ai k
r c
C ar r
C a
o a
f o
L a
0 t
K c3 a
i D
iM l
l -
4 a
c t
5 5
5 5
0 3
y1 0
0 2
s t
i 4
4 4
2 5
3 C2 3
3 3
e i s T
nk I
eru
. s 9
9 9
0 0
se r 2
2 2
0 0
om u 1
1 1
0 0
pi o 4
4 4
7 7
xT
}
[
E
)
)
?
?
t
(
(
n d
d e
e e
e v
v s
t S
S S
i i
a T
T T
e e
e L
L L
c c
r e
e T
+
+
+
R R
S S
S S
S t
S S
S s
s T
T T
T T
a F
F F
A A
L L
L L
L e
H 0
0 0
0 0
0 0
0 l
0 0
0 0
0 6
0 a
6 6
6 6
6 6
i y
r y
2 y
y y
y 2
o y
e o
2 8
o o
o o
8 l
o t
l 8
1 l
l l
l 1
l l
1 l
l l
l l
A l
1 A
I I
A A
A A
I 1
l vN cO l8 s
'll!l[
EER 89 -ee55 R ev- (
$9 f3 V'
-3 10 l
NRC CURVE 3
linlyr
/_
10-4 A
8 f
3 AD,-
f' /
l-
-5
-c 10
/
O.
V 0.2 one 0 ; 5'"5jttrea at 1150'F/2 h 2
W (EPR a 15 C/cm ) - GE A 0.2 ons 0,; sensitized at 1150*r/2 h 2
C (EPR = 10 C/cm ), gg O 0.2 ppm 0 ; 5'"5 "' "d 't '150/2' "
2 H
. GE 3
O 0.2 pas o ; severely sensttired g
@ l0-6 O 8 em 0 ' 5'"5'""8 a t '150/2' "
2 CD GE HITACHI GENED AML g
GECR0 y
<:C rdr48 ppm 0;sensitizedbywelding; 2
- 0.04 in /yr tr5.t 932 r/24 h (5R n e
U 3 8 ppe 0 ; sensittred at (1292*F/10 min)'
2 2
+ (932*F/24 h) (EPR = 4 C/cm )
l O 8 pas 0 ; sensitized at (1292'F/10 min)
~~
2 2
+ (932*F/24 h) (EPR = 4 C/cm )
10-7 I
D 8 ppm 0 ; sensitized at (1292*F/10 min) l 7
t (842'F/257 h) (EPR
- 15 C/ca )
+
l
@ 8 ppm 0 ; sensitized at (1292'F/10 min) RECENT 2
l
+ (842*F/257 m) (EPR a 15 C/ce!)
ANL f = 0.1 Mr. R
- 0.94 l
O 8 ppa 0 ; sensitized at 1292'F/14 h DATA 2
(t/R = 20 C/ce!) f = 0.008 Mz, l
R = 0.95 l
X 8 p p 0 ; sensittred at 1292*F/14 h 2
I (EPR = 20 C/cs!) f a 0.08 H2 I
R = 0.95 s
i O
10 20 30 40 50 60 70 STRESS INTENSITY, K (ksid5.)
Figure 4-1.
Crack Growth Race Data SIR-89-008, Rev. 0 4-9 il 11:
i-
> c. 4 ASSOCIATES DC
Sev \\
g*f 'ko j
1 0 -'.-
4 Alloy 600 o Corrosion '87, Paper f84
_~
EPRI RP 2006-17 EPRl/GE a Dresden 1 - In Reactor Information Exchange Hi h og 7 ppm) o EPRI RP 2293-1 gg (1 ppm H SO )
2 4
10-':
NUREG-0313 inconel 1.078x10 (K)"
d T
Stainless St
(
3.59x10-*(K)g' f
c
/
e c
/
/
.c3 10 -':
a 2
/
0
/
~
\\
g 4<___.__>i o
o 10 -'::
4 l
I l
10 _,
i.4 4 i. iii
.............i.
.......i.........
10 20 30 40 50 60 Stress Intensity Factor, kaidn i
Figure 4-2.
Crack Growth Rate Data for Inconel 600, 82, 182 l
l SIR-89-008, Rev. 0 4_10 M
ASSOCIATES WC L_____________________
STRE,55 INTOEUT, K (MPc@
Rev. t to 20 30 40
@at 4I
-2 10 t
en Stainless Stati i !C '
l Uc;:tr bound da/dt 3
i, Inconal 600 t
l-t 0 + --
~~ / ~ ~.,,,., Uf':'
.~~1; V
,,, ;7,a,.f.
T
\\_
,/ /!It ' '
- - -4. n - 3 5
2 t
---;.c a
I z
r-4 7
a a
c t
a s
[
T 7
u u
I
-r r
j c
=
-s -
e 2
[g q
t E
5 5
t
=
=
O i
2 0
e
_ <., 4 e
o L
.6 a
i x
x 4
g u
4 C
i
- =
C l
l t
l
-s 1 10 t
e.
s I
L i
-a 12 I
1 4
0.2 bpn 0 3
2
-7 I
I I
~2 t
O 10 20 30 40 50 ST.:ESS INTENSITY, K CKs1 6 3 GENERAL ILECTRic COMPANY PROPR!ETARY INFORMATION N
DITEGRITY ASSOCIKfEEINC
- z_z,
ll!!\\l 1i'l NpjT ' g M N
Ngl Tmf 8
me 1
Y c
E e
R e
W w
T r
o.
a S
oS S.
e f
r S
S H
E E'
S 0 S
nS n'
V V
9 1
1 E
A C A
9 Gl G
C C
GC n'
R r
8
)
A E a"
y r i
eH o
0 m r
P S"
k o
t N e
E S H
s" w
t c W c
/c t
a a
a n
T.
S Ot E "'
r e H e
0 p
O i
t Ml L
o R f
s
(
A l'
nN n
- e. Y u
a t
eT m
L i I
u IV ly W s O$l I
,5 T
n l
E 0 C o
e NI U
t ES 4 D g
G cN V
nn O
3 C e
oe C /[/, 2 c
'6 W
'e l./
P
- P e
'[
0 I
4 CE
[9fo n
4 e
{c-r i: -
u 5
g 4
/
k'8 i
~
n 8
F g
F i
1 g
10 0 4 eqg m
"(g0
'd n
m,.
D o,
E n
C Y.
t i
Z nV Wt l
n l
e A
I m
l T
ni T
I l
i is GNC 0nk UEC IS 0r0 OS S o,
o S
l N6i3 0
E l
v~
I 0
- i l
n T N6 m,
E YEy P OY t
CT O.
I Sori EAU o
e s T
o ell n GN A c
CLae NES c.
Wtn H MT o
I S
A A E E. E i
I c
Cs WlP H m
H O M N
s LI A 8e R
0 r t
o 2
U S
m F
N7
,o@ w%-
r '.
iB
~.
(fltjl!;lIllll
g yy~~o RevI g
- 7 6
1 0 -' __
A!!y! RP 2006-17 o Corrosion '87, Paper 784 600 4
EP EPRig e Dresden 1 - In Reactor information Exchange High Oxygen (7 ppm) e EPRI RP 2293-1 e
Resin intrusion
"~
(1 ppm H SO4)
NUREG-0313 Stainless *(Stg' 3.59x10 K) 10~';
s' g
g Inconel p
g 1.078x10-*(K)*
s
+
c c
/
+
.c l
li 10 :
/
4o a
e
/
e y
g 4
e o
o 10 -*:
GE Proprietary Data In-Reactor CAVS Laboratory CAVS
+
Laboratory /cm 0.3-0.5 S 10-',......................................
10 20 30 40 50 60 Stress Intensity Factor, kaidn Figure 4-5.
Crack Growth Rate Data for Inconel 600, 82, 182 Including GE Proprietary Data permoema SIR-89-008, Rev. 0 4_13
Rev. l 9ay y
MPa [m 10 20 30-40
~
M d/'"
10
- " Arm 'consr. como
- car
-3
- o a' 10
- @ [j s/.m y, ru :
roc o,
- ~
o t.o f
~
,sf,
MEDCTED CURW3 (tt) 1.0
- SAK3 QM 30455 ISC/em2 200ese 0 2
-7
-3smvine m 10 N
0.5
_4 iv c
v b.
w 4 a
\\=
C
.U O3
~
-8 e
x
-s g 'C c
o
-5 6
10 e
0.8 o
c:
[
o o
x u
=
_3 U
10
- /
/
-o
/
- 10 f
-10 10 10 20 30 40 STRESS INTENSITY, ksivTd FIGURE 15. Comoarison of the predicted and observed crack growin rates vs stress intensity for available cata on inconels 600 and 182 tested at or near constant load in 200 ppb 0xygen, 288 C water. It is possicle to infer some trends in the data which are not predicted (e.g., a plateau in stress intensity or conductivity),
although conclusions are not mented from the data (see text).
SIR-89-008, Rev. 0 4-14
- ,s.
e,;
e.
.- r : 4 5 CCm me me
i
)
D I
T L
E W
I F
i
(
O C
E R
A D
G 0
0 D
L 6
E IF C
E
&A D
D E
Dl
)
l
)
s EA d) w F
n 7
T a
S 0asR t
l 4uU)
CfE w
4 oOD S
se hl L I
e I
rN DR T (lE r
a,
(
l e
EF u
E1 IM (P y1 g
T Ri
- 2. E i
l P
T E
F t
S i
PU 5S B
D P
5
(
E FS B
+
O
)
w 0
O L
g a
2 N
s r w i
a e S
eN A
I y
R s
C r
O i
A
.E y
g C
I 3 A
,n P
D I
1 i 6
d
.Pn)
M
(
E uw l
)
O S oa D
BDI l
,2
(
l 0
U C
OD 9,
8 7
5 4
3 2
1 0
(
j o
0 0
0 O
0 0
0 P
7 ES 4
D e
r
^
3 })O 3 u
i g
i F
m ;;
- 0"
- nk-o 7G a= #/
exn ea-ooSS Re.V-l 4 (,
9ay -
5.0 CONCLUSION
S On the basis of this evaluation, it is concluded that BSEP Unit 2 can continue to be operated safely for the remaining six months of the current fuel cycle, even if flaws of the same severity as those recently observed in the Unit 1 recirculation inlet thermal cleeve attachment welds are present in Unit 2,
and have gone undetected in prior inspections.
l This conclusion is supported by the following observations and cvaluation results.
1.
Cracking in the Unit 2 thermal sleeves, if any, should not be significantly worse than that observed in Unit 1.
2.
The worst of the Unit 1 flaws would not currently pose a safety concern if they existed in Unit 2.
Due to low applied stresses at the flaw location, they satisfy ASME Section XI allowables, which are based on no loss a design basis safety margins.
3.
Crack growth predictions, based on laboratory data and correlated against the observed field crack growth rates in Unit 1 and in a similar BWR thermal sleeve configuration (Duane Arnold Energy Center) indicate that the worst of the observed Unit 1 flaws would not be predicted to exceed the Section XI allowables during j
the remainder of the Unit 2 fuel cycle.
4.
The unit is currently on and is planned to be on hydrogen water chemistry (HWC) for the remainder of the i
fuel cycle.
The beneficial effects of HWC are expected to arrest or greatly reduce crack growth rates at the l
thermal sleeve attachment weld location.
l SIR-89-008, Rev. 0 5-1 g
EER SA- 0o55 Revl pd p 41 5.
Field experience at DAEC indicates the hypothetical failure mode at this location to be one of leak-before-break.
,$ s!
9 wsa n
n 4
o
,y s
i m
l 42,'
$$a
.u 56i
.., l SIR-89-008, Rev. 0 5-2
- .e si; e
- c..
EER. 29)- oos5 R e. v-l 4%
Pap
6.0 REFERENCES
1.
Structural Integrity Associates " Weld Overlay Repairs of Recirculation Inlet and Core Spray Nozzle-to-Safe End Weld, Brunswick Steam Electric Plant Unit 1," Revision 0, January, 1989.
66 2.
General Electric Co.,
" Design Report: Recirculation Piping and Equipment Loads," 23AS485, Revision 0, October 2, 1985.
3.
Nutech Engineers, "Ra.sidual Stress Analysis of Thermal Sleeve-to-Safe End Welds:
Brunswick Steam Electric Plant Units 1 and 2," XCP-01-003, April 18, 1979.
4.
Structural Integrity Associates, " Development of Inconel Weld overlay Repair for Lcv Alloy Steel Nozzle-to-Safe End Joint," EPRI Projec#. RPT 303-1, final report, June 1988.
5.
Structural Integrity Associates, "pc-CRACK User's Manual,"
Version 1.2, March 1987.
6.
Telecon from R. Jordan (CP&L) to D. Pitcairn (SI), February 1989.
7.
Technical Report on Material Selection and Processing Guidelines for BWR Coolant Pressure Boundary Piping, NUREG-??l3, Rev.
2.,
January 1988.
8.
NRC Position on IGSCC in BWR Austenitic Stainless Steel Piping (Generic Letter 88-01), January 25, 1988.
p:
9.
EPRI/GE Information Exchange.
10.
Reactor Primary Coolant System Pipe Rupture Study, U.S.
Energy Research & Development Administration Contract AT (04-3)-189, Project Agreement 37.
11.
P.
L. Andressen, Corrosion
'87, Paper #84, "Effect of Dissolved Oxygen, Solution Conductivity and Stress Intensity on the Interdendritic Stress Corrosion Cracking of Inconel 182 Weld Metal."
12.
EPRI Research Project RP 2006-17, General Electric Contractor, In Progress.
Personal Communication with Project Manager R.
Pathania.
13.
EPRI Research Project RP 2293-2, General Electric Contractor, In Progress.
Personal Communication with Project Manager R.
- Pathania, v.
SIR-89-008, Rev. 0 6-1 DgEyB.JIT._
u Rov l Pa jg 14' P. L. AndroacGn, Corrosion, Volum3 44, No. 6, Juno 1988, p
- p. 376, " observation and Prediction of the Effects of Water Chemistry and Mechanics on Environmentally Assisted Cracking of Inconals 182 Wald Metal and 600."
s 19 s
8s a
m
~
-s
'ZC'g m
v
.: s
=r J*! 4
- Ni
- i l
SIR-89-008, Rev. 0 6-2 1
l 3cery,an e w
EER 8") o055 R ev..i_
'P$e-le ATTAC}DfENT 1 NUCI. EAR SAFETY EVALUATION CHECKLIST Identification and Description of Iten Being Evaluated:
EER 8#)-oo5s. Fev.L :
Mfih$ L G( b ri ~^ D&vaNsNn dr[7 2 dNL. % b af Gor Ciud k TLo h'es%l$is Tlx+ AotEe SAw Gds (dzh w
1 Circle one 1.
Does this item represent a change to the facility Yes h as described in the FSAR7 If yes, describe the change and the FSAR section(s) involved.
l 2.
Does this item represent a change to procedures as described g the FSAR?
Yes h
~
If yes, describe the change and the FSAR section(s) involved.
3.
Does this item represent a test or experiment not described in the FSAR7 Yes h If yes, describe the test or experiment.
l l
4.
Does this ites require a revision to the FSAR?
Yes h (If yes, submit appropriate change (s) per RCI-04.1, as applicable) 5.
Does the proposed change, test, or experiment require a change to the technical specifications?
Yes @
(If yes, submit appropriate change (s) per RCI-02.1) 0-RCI-03.1 Rev. 5 Page 11 of 14 l
ggj$, M 0066 Rev. 1 Pap E ATTACHMENT 1 (Cont'd)
Circle Onj 6.
The change, test, or experiment shall be tested against the following criteria:
'6.1 Will the probability of occurrence of any accident previously evaluated in FSAR (Chapter 15) be increased?
Yes Unceetain
' BASIS
- b e_
& b nh l
6.2 Will the consequences of any accident previously evaluated in the FSAR (Chapter 15) be increased? Yes h Uncertain BASIS
- 6M 7 6
6.3 Will the probability of occurrence of malfunction of equipment important to safety previously Yes h Uncertain evaluated in the FSAR be increased?
BASIS
- bee 'T1p 6.4 Will the consequences of malfunction of equipment
]
important to safety previously evaluated in th,e FSAR be increased?
Yes N Uncertain BASIS
- b
'l Ac N=
6.5 Will the probability of an accident or possibility for malfunction of equipment important to safety of a different type than already evaluated in FSAR Yes h Uncertain be created?
bgFJ BASIS
- fte t,
- BASIS REQUIRED - UTILIZE ADDITIONAL SHEETS AS NECESSARY
)
EsER Boi - ooss Ra.q.\\
P'ag &
ATTACHMENT 1 (Cont'd)
'1 circle one 6.6 Will the margin of safety as defined in the basis to any technical specifications be reduced?
Yes @ Uncertain BASIS
- 6 p dP_
8
~2 Fr 7.
If any response to Question 6 is yes, it is to be Yes h assumed that the proposed changs or test or experiment constitutes an unreviewed safety question within the meaning of 10CFR50.59.
Based on this determination, does the subject change or test or experiment constitute an unreviewed safety question?
8.
Will the change, test, or experiment have a significant Yes @
adverse effect on the environment?
l 9.
Will the change, test, or experiment raise a potential safety concern on the unit to which it applies, or to
~~
the other unit?
Yes @
10.
Does the change, test, or experiment affect or bring about a change to other plant procedures?
Yes h NOTE:
A copy of this Safety Evaluation shall accompany the package through review.
11.
Indicate the sections of the FSAR researched to confirm the determinations l
made in Items 1, 2, 3, 4, 6, and 8 above.
I chan6-is. 3 2._, 3.i. s.2.
v.3. s.4.1. s'.4. 3, 6 2-12.
Indicate the sections of the technical specifications, researched to confirm the determinations as applicable, made in Items 5 and 6 above.
7
- 94. 4. I, X 4.4.3 3J4.4 8. 3. 4 3,2 i
Prepared by:
61 4 Date:
3 -/ 0 - N C) 5e. id.16o.4ak l
Title:
PNSC Approval:
Date:
(if required)
- BASIS REQUIRED - UTILIZE ADDITIONAL SHEETS AS NECESSARY r
.Q__ i Y30. _QC1 fl n-D n--_
oo A oA
l EER No. 89-0055 Rev. 1 Page 85%
1 SAFETY ANALYSIS 6.1 There is no reason to believe that cracks in Unit 2 would be significantly worse than those in Unit 1.
The basis for this assumption is that the operational. time for each of the Units is comparable; is, through December, 1988, Unit 1 was at approximately 6.39 EFPY (Effective Full Power Years) and Unit 2 was at approximately 6.64 EFPY.
In addition, the water chemistry in the two units has been similar.
The cracks in Unit 1 are currently not a safety concern in that they are within the code allowables.
Projected crack growth rates without Hydrogen Water Chemistry (HWC) indicate that Unit 2 can be operated for:the remainder of the current cycle (approx. 6 months) without exceeding the code allowables for flaw indications.
The addition of HWC should significantly reduce the crack growth rate.
In addition, the expected failure mode would be leak before break.
The BSEP Technical Specifications limit the unidentified leakage to 5 gpm averaged over any 24-hour period or a 2 gpm increase within any 24-hour period during normal plant operation.
Should cracks in the recirculation piping propagate throughwall, the unidentified leakage rate would increase and the plant could be safely shut down using normal design safety systems required to be operable in accordance with plant technical specifications.
Based on the above, it is concluded that the probability of occurrence of any accident or malfunction of equipment previously evaluated in chapter 15 of the FSAR will not be increased.
6.2 Pipe breaks are analyzed in chapter 15 of the FSAR.
All possibilities for pipe break sizes and locations were investigated.
The most severe nuclear system effects, and the greatest postulated release of fission products to the primary containment, would' result from a complete circumferential break of one of the recirculation loop pipelines (28").
This accident has been established as the design basis LOCA.
A circumferential break of a recirculation discharge riser (12") is bounded by the design basis LOCA.
In addition, it would require five severed 12" discharge risers to equal the leakage area posed by one severed 28" pipe.
Therefore the consequences of any accident or equipment malfunction previously evaluated in chapter 15 of the FSAR will not be increased.
6.3 See 6.1
f l
EER No. 89-0055 Rev. 1 Page
>$'4 6.4 See 6.2 6.5 Pipe breaks are analyzed in chapter 15 of the FEAR.
No new accident scenarios would be created by the presence of cracks in the recirculation discharge risers.
Therefore, the probability of an accident or possibility for malfunction of equipment important to safety of a different type than already evaluated in the FSAR will not be created.
6.6 There is no reason to believe that cracks in Unit 2 would be significantly worse than those in Unit 1.
A crack growth analysis indicates that Unit 2 can operate for the remainder of the current fuel cycle (approx. 6 months) without exceeding the code allowables for flaw indications.
It is therefore concluded that the margin of safety as defined in the basis to any technical specifications will not be reduced.
(
EERJ4 Q O) - 00 66 R0v. No.
?bt
$6 FORM 3 ENGINEERING EVALUATION REPORT ENVIRONMENTAL QUALIFICATION IMPACT FORM (EER-EQIF) l Will the evaluation:
1.
Justify the deletion of equipment / common components from the BSEP EQ program?
Yes No 1
2.
Justify the addition of (already existing) equipment / common components to the BSEP EQ program?
[
Yes No 3.
Authorize the repair of EQ equipment / common components with other than qualified like-in-mind equipment / component parts?
Yes E No 4.
Affect the existing installation or interface (of EQ equipment / common component applications) as designated in EDBS and/or in the qualification data package?
$ No Yes 5.
Justify the (quality class) upgrade of equipment / common components or component parts which could be utilized in EQ applications?
Yes No 6.
(Re) Define qualification parameters (e.g., normal or LOCA/HELB environ-mental conditions, postaccident operating time requirements, essential passive / active postaccident operating requintments, qualified life assumptions /results, etc.) for specific EQ equipment?
O Yes M No 7.
Provide an EQ-related jus'.ification for continued operation (as required per PLP-02, Section 4.4.3.3 g 4.4.4)?
O Yes
$ No 8.
Provide the resolution of a qualification problem (as required per PLP-02, Section 4.4.4)?
Yes No NOTES:
1.
If all no, then no further EQ consideration is required.
Mark I
the EER Traveler accordingly as required by ENP-12 and include this completed EER-EQIF within the EER package.
An EQ Technical Review is not required.
2.
If any yes, an EQ impact assessment (per Section 5.3) must be performed during the evaluation process.
Mark the EER Traveler accordingly and include this completed EER-EQIF within the EER package. An EQ technical review is required.
l BSEP/Vol. XX/ENP-34.1 19 Rev. 3
w_-----,-----------__.___.
(.
ENCLOSURE 2 GENERAL ELECTRIC AFFIDAVIT j
j
d GENJRAL ELECTRIC COMPANY AFFIDAVIT I, Rudolph Villa, being duly sworn, depose and state as follows:
1 1.
I am Manager, Consulting Services, General Electric Company, and have been delegated the function of reviewing the information described in paragraph 2 which is sought to be withheld and have been authorized to apply for its withholding.
2.
The infomation sought to be withheld is contained in the Seneral l
Electric presentation entitled " Thermal Sleeve Attachment Weld Overlay Hi s tory. "
3.
In designating material as proprietary, General Electric utilizes the definition of proprietary information and trade secrets set forth in the American Law Institute's Restatement of Torts, Section 757. This definition provides:
"A trade secret may consist of any formula, pattern, device or compilation of information which is used in one's business and which gives him an opportunity to obtain an advantage over competitors who do not know or use it....
A substantial element of secrecy must exist, so that, except by the use of improper means, there would be difficulty in acquiring information....
Sons. factors to be considered in determining whether given information is one's trade secret are: (1) the extent to which the information is known outside of his business; (2) the extent to which it is known by employee.; and others involved in his business; (3) the extent of measures taken by him to guard the secrecy of the information; (4) the value of the information to him and to his competitors; (5) the amount of effort or money expanded by him in developing 1.he information; (6) the ease or difficulty with the which the information could be properly acquired or duplicated by others."
4.
Some examples of categories of information which fit into the definition of proprietary information are:
a.
Information that discloses a process, method or apparatus where prevention of its use by General Electric's corgtitors without license from General Electric constitutes a cocr.stitive economic advantage over other companies; b.
Information consisting of supporting data and analyses, including test data, relative to a process, method or apparatus, the application of which provide a competitive economic advantage, e.g., by optimization or improved marketability;
u c.
Information which if used by a competitor, would reduce his expenditure of resources or improve his competitive position in the design, manufacture, shipment, installation,~ assurance of quality or licensing of a similar product; d.
.Information which reveals cost or price information, production capacities, budget levels or commercial strategies of General Electric, its customers or suppliers; e.
Information which reveals aspects of past, present or future General Electric customer-funded development plans and programs of potential comercial value to Gene.a1 Electric:
f.
Information which discloses patentable subject matter for which it may be desirable to obtain patent protection; g.
Information which General Electric must treat as proprietary according to agreements with other parties.
5.
Initial approval of proprietary treatment of a document is typically made by. the Subsection manager of the originating component, who < r most likely to be acquainted with the value and sensitivity of th information in relation to industry knowledge. Access to such, documents within the Company is limited on a "need to know" basis and such documents are clearly identified as proprietary.
6.
The procedure for approval of external release of such a document typically requires review by.the Subsection Manager, Project manager, Principal Scientist'or other equivalent authority, by the Subsection Manager of the cognizant Marketing function l(or delegate) and by the Legal Operation for' technical content, competitive effect and-determination of the accuracy of the proprietary designation in accordance with the standards enumerated above. Disclosures outside General Electric are generally limited to regulatory bodies, customers and potential customers and their agents, suppliers and licensees then only with appropriate protection by applicable regulatory provisions or proprietary agreements.
7.
The document mentioned in paragraph 2 above has been evaluated in accordance with the above criteria and procedures and has been found to contain information which is proprietary and which is customarily held iniconfidence by General Electric.
8.
The information to the best of my knowledge and belief has consistently been held in confidence by the General Electric Company, no public disclosure has been made, and it is not available in public L
sources. All disclosures to third parties have been made pursuant to regulatory provisions of proprietary agreements which provide for maintenance of the information in confidence.
E' I
9.
Public disclosure of the information sought to be withheld is likely l
to cause substantial harm to the competitive position of the General Electric Company and deprive or reduce the availability of profit making opportunities because it would provide other parties, including i
i competitors, with valuable information concerning stress corrosion cracking in low alloy steel /Inconel 182 and the impact of plant water chemistry, which were obtained at considerable cost to the General Electric Company.
i STATE OF CALIFORNIA
)
COUNTY OF SANTA CLARA
) 88 Rudolph Villa, being duly sworn, deposes and says:
l That he has read the foregoing affidavit and the matters stated therein are true and correct to the best of his knowledge, information, and belief.
Executed at San Jose, California, this f A day of MI M, 198 I.
Rudolph Villar General Electric Company Subscribedandswornbeforemethis7 day ofkMt3-1981
}]1 m / 9 U b fA.
1 NOTARY PllBLIC, STA"E OF CALIFORNIA OFFICIA!. SEAL
[
l MARY L KENDAlt NoTAIM PUSUC = CAUFORNIA f' SANTA clARA COUNTY s
. _ _ _ -.. _