ML20043C745
| ML20043C745 | |
| Person / Time | |
|---|---|
| Site: | FitzPatrick  |
| Issue date: | 05/31/1990 |
| From: | Mehta H GENERAL ELECTRIC CO. |
| To: | |
| Shared Package | |
| ML20043C742 | List: |
| References | |
| SASR-90-43, NUDOCS 9006060130 | |
| Download: ML20043C745 (19) | |
Text
e Attcchment 2 to JPN-90-040 DRF # B11-0497 SASR # 90 43 DRF # 137 0010 May 1990 i
I STRUCTURAL EVALUATION OF INDICATIONS IN THE REACTOR TOP HEAD AT THE JA FITZPATRICK POWER STATION i
Prepared by:
H S. Mehta, Principal Engineer Materials Monitoring &
'l Structural Analysis Services
(
Verified by:
T A. Caine, Senior Engineer Materials. Monitoring &
Structural Analysis Services
{
i Approved by:
S. Ranganath, Manager Materials Monitoring &
Structural Analysis Services i
9006060130 900525 DR ADOCK 05000333 ff PDC I'
GENuclearEnergy
-* I r
j TABLE OF CONTENTS A
i
SUMMARY
.l
1.0 INTRODUCTION
l 1
i i
2.0 SECTION XI FRACTURE MARCIN ASSESSMENT 1
1 2.1 ' Applied Stresses 2.2 Stress' Intensity Factor Calculations 2.3 Allowable Flaw Depth Calculation
.2.4 Crack Crowth Assessment
3.0 CONCLUSION
S i
4.0 REFERENCES
)
i
l
SUMMARY
t This report evaluates four weld - flaws identified during routine inservice inspections of the James A.
FitzPatrick reactor pressure vessel head.
These inspections were conducted during the 1990
~
Refueling Outage and used ultrasonic equipment to detect flaws.
Two of the four flaws..are acceptable without repair under. the provisions of. ASME Section XI.
No further evaluations were performed l
on these two flaws.
t For the purposes. of this evaluation, the two remaining flaws were. r single. larger,. bounding flaw.
The model also assumes modeled as a that these flaws are surface cracks ' subject - to growth as a result exposure to the environment inside the reactor ve'ssel.
Both of these-assumptions are conservative since they would tend to accelerate crack growth and increase stress levels..
All examination data suggests that these indications are not due to cracks, but are attributable to the original vessel manufacturing process.
The evaluation duonstrates that even if these flaws are assumed to be cracks, ASME Section X1 code allowable stresses will not be exceeded and all code margins will be maintained during the next operating cycle.
l 11
1 r
- 1. Introduction l
ll During the 1990 Refueling _ Outage at the James A.
FitzPatrick Nuclear 1
Power Plant, selected reactor pressure vessel head welds were examined as part of routine ASME Section XI code - inspections. - Four flaws (or
'I indications) were identified as a-result of these ultrasonic inspections.
These indications were. located in the: circumferential weld that joins the circular upper head plate (or - dollar plate) -and the curved i
trapezoidal plates (meridian ~ platas) that together form the-hemispherical head. Figure 1 illustrates the location of this weld.
All four flaws are located in the dollar plate, within the weld and heat-affected zone regions.
t l
Ini'ial examinations used manual ultrasonic techniques from the t
exterior of the head.
Supplemental. manual examinations from the interior and exterior surfaces 'and automated examinations from the exterior surfaces were used to further examine-and define the final indication. sizes. After careful review of all of the inspection data,,
four separate indications were: identified (Ref. 1)..All examination data suggests that these indications are not due to cracks, but are attributable to the vessel manufacturing process. These are described on Table 1 and their location illustrated in Figure 2.
l I
Two of the four flaws (Nos. 1 and'3) are acceptable without repair
)
under the flaw acceptance criteria of ASME Section XI.
No further
-I evaluations were performed on these two flaws.
The other two flaws (Nos. 2 and 4) exceede'd ASME Section XI' allowable.
flaw criteria.
These two welds were evaluated using Section XI, IWB 3600 criteria to determine'if repairs were necessary.
l 1
~
l
-By_ comparing radiographs of the flaw areas to pre service radiographs, it was apparent that flaw Nos. - 1 and 4.were present during head manufacturing.
Both sets of radiographs for flaw No. 4 were computer enhanced to improve the ability to detect - flaws. -This comparison further confirmed that these flaws were not service induced.
i The interior surface in the region'of flaw No. 4 was. lightly surface l
polished and then re inspected using liquid penetrant and magnetic particle techniques.
No seidence of cracking or connection to the_
subsurface indications was observed in either test.
Despite the absence of a
surface
- flaw, the model analysis conservatively assumes that the bounding ' flaw - is a - surface crack j
subject to growth as a result of exposure to. the environment-inside
{
the, reactor vessel.
Both of these assumptions are conservative since they would tend to accelerate crack growth and increase stress levels'.-
i i
To facilitate analysis, individual indications were first translated into two flaw combinations.
Flaw combination A represents flaw 2 and l
flaw combination B-represents flaw 4 The dimensions of flaw combinations A and B are shown in Figure 3.
The flaw analyzed conservatively._ bounded the dimensions of both flaw j
combinations.
This bounding flaw was assumed to be open to the interior of the reactor vessel, 0.55 inches-deep and 5.5 inches long.
3 Even with these conservative assumptions, the evaluation demonstrates that if these flaws are indicative of cracks, ASME Section: XI code allowable stresses will not be exceeded and all code margins will be maintained during the next op3 rating cycle.
l 2
w
.y 2.
SECTION XI FRACTURE MARUIN ASSESSMENT'
.The UT examination has characterized the subject. indications toi be subsurface.
Nevertheless, for the-purpose of fracture margin:
assessment they. were conservatively characterized as surface indications.
Thus, crack-growth due'to stress corrosion cracking was-included in calculating the final flaw depth for comparison with allowable value.
- Thus, the final ' crack. sizes determined' here.
represent bounding values based'on' SCC growth.
A11L evidence supports..
a fabrication flaw in existence since plant startup. 'The details of fracture margin assessment are described next.
2.1 Applied Stresses For the purpose of determining'the allowable: flaw size the-vessel i
hydrotest was the governing transient since. the. pressure Lis highest and the vessel temperature was lower (than-that 'for operating l
condition), thereby providing the -limiting. fracture loading for both
.)
normal and faulted conditions.
l 1
-The stresses in the region of indications are primarily due-to:
j (1) pressure stress and (2) weld residual stress. The pressure stress magnitude corresponds to.PR/2t.,
or about 15.1 ksi based on the j
measured thickness of 4 in, and the 1100 psi - pressure ' for ' the-- vessel l
1 hydrotest.
Residuni stress at the weld between the dollar plate and.
the trapezoidal plates, is reduced significanth m.
result of post weld heat treatment (PmtT).
Ilowever, some weld ru.idual. stress'still remains af ter PWIT, so a bending stress value of 8 ksi was assumed for this analysis.
]
2.2 Stress Intensity Factor (K) Calculations Stress intensity factors were calculated for a semi elliptic.
surface flaw with an aspect ratio a/1 0.1 for the hydro test
_ i condition.
In the absence of specific stress intensity solutions for the spherical vessel head geometry, the evaluations were performed using two different idealizations:
3
-. - J
1.
Section XI method using the flat plate solution The solution for a surface flaw (Ref. 2) is applicable for both membrane and bending stresses. Thus, pressure and residual stresses are included.
2.
Solution for a semi elliptic axial flaw in a cylinder This solution developed by Zahoor (Ref.
3) for j
different radius to thickness values and aspect ratios is only valid for internal pressure.
The bending j
residual stress was evaluated using the Section XI flat j
plate solution.
i Figure 4 shows the comparison of the calculated stress intensity factor as a function of crack depth.
It is seen that the two solutions are close, but for deeper flaws the solution for the axial crack in a cylinder is more realistic, since it represents the stiffness of the vessel head more accurately.
In any case, for the flaw sizes of _ interest, ' the differences between the two solutions are insignificant.
2.3 Allowable Flaw Depth Calculation 1
The allowable flaw depth is determined using the Section XI IWB 3612 acceptance criteria based on the applied stress intensity factor.
According to the FitzPatrick technical specifications (Figure 5) at 12 EFPY, the temperature for the hydro test is 190 F.
The RT the plate is 10 F, but for the NDT dollar weld material, specific values were not available.
However based on the measured Charpy energy values in the original material certs (in excess of 70 ft-lb at 10 F), the RT was conservatively assumed to be 0 F.
The available NDT 4
4
4 toughness is 200 Ksi/in ' at 190 F.
With a safety factor of- /10 on toughness, the allowable K value (according to IWB 3612) is 200//10 -
y 63.25 ksi/in.
The allowable crack depth corresponding to this is 1.667 in. (Fig. 6).
This is the maximum allowable crack depth at the end of the next inspection interval.
I 2.4 Crack Crowth Assessment l
4 As stated earlier, the indicat. ions are subsurface and are proba.
bly due to fabrication defects.
Thus, there is no evidence of any l
environmental effect that would cause crack growth.
Nevertheless, a conservative crack growth analysis was performed
. n.. a. surface i
flaw. The crack growth rate data from Ford (Ref. 4) shown in Figure 4
7, was used for the growth prediction.
Since the applied stress intensity factor is low, the Equation 8 relationship shown in Figure 7 4
was used.
Figure 8 shows the predicted SCC growth for the flaw combination B, which assumed a surface crack of total depth 0.55 in.
The predicted SCC growth was 0.086 in, in the 18 month period, corre-l sponding to the next fuel cycle.
In addition to this,- fatigue crack growth due to 10 cycles of pressurization were also included.
The incremental growth due to the 10 cycles of fatigue crack growth is conservatively estimated to be 0.010 in.
This gives a total crack depth increment (due to SCC and fatigue) of 0.086 + 0.010 - 0.096 in.
The predicted final depth is 0.55 + 0.096- ~ 0.646 in., which.is well within the allowable flaw size of 1.667 in.
i l
5
- l
S
3.0 CONCLUSION
S The result of the conservative crack growth prediction, assuming-a suriace flaw shows a final depth 0.646 in.: at the end of the next fuel cycle.
This is well below the allowable depth of 1.667 in.,
which assures the safety factor of /10. Thus, continued operation for the next fuel cycle can be justified and the' full ASME code _ margins are maintained.
r s
6
4.0 REFERENCES
1.
Ultrasonic Examination Data Package for Veld VC TH 1 2, copy contained in CE DRF B11 00497.
2.
ASME Boiler and Pressure Vessel Code,Section XI, " Rules for In Service Inspection of Nuclear' Power Plant Components,'" 1989 Edition.
3.
A.
- Zahoor,
" Closed Form Expressions for Fracture Mechanics Analysis of Cracked Pipes,"
Journal of, Pressure Vessel-Technology, Vol. 107, May 1985.
4.
F. P. Ford et al, " Stress Corrosion Cracking of 1.ow Alloy Steel /
Water Systems at 288'C, "EPRI Contract RP 2006 7-Interim Technical Technical Progress Report No. 2, August 1988.
I:
l i
7
1 L..
Td)le 1 I
l
ACCEPTABLE'BY RG INDICATION #2 AZIMUTH 209*
PARALLEL TO WELD IN HAZ SHALLOW DEPTH-(.1")
NEAR ID SURFACE INDICATION #3 AZIMUTH 185*
s l
TRANSVERSE TO WELD ON FUSION LINE SHALLOW DEPTH (.3")
VISIBLE SURFACE SCAR ACCEPTABLE BY RG INDICATION #4 AZIMUTH 113' MutTIoLE REFLECTORS IN HAZ PARALLEL TO WELD SUBSURFACE FLAW SURFACE GROOVE 8
DOLLAR PLATE DOLLAR PLATE i
ae 4w A
- o ooaoaaaaoooo*
i
\\
FLANGE Figure 1. RPV TOP HEAD ASSEMBLY 9
r i
l-O OOO O O
O O
O
.O' O
O O
1 O
9
.O t
s O
O l-O
+
O
.o O
o O
i o
O Q
s o
O
}e o
O et 5
?
-o O
"j
.o.
O.
\\O o
o-O O O OO Fimre 2, l.ncatim n' the Indicqtims 10
4 I
0.333"-
N 3:
5" u
1r
/
/
1r s
\\
j i i l
- 0.17 "
.a
=
0.50"-
r (assuming = surfoce flow) i Flow Combination #
A l
-0.35" N
N 2.3."
1r1 r p
/
(
I f
\\
\\
- g J L J L-0.2"-
4 a
=
0.5 5" -
(assuming surfoce flow)
Flow Combination 8
cin 3 FitzPatrick Top Head Flow Combinations 11
4 e
100 P = 1100 PSI 90 -
RNITE FLAW, c/ = 0.1 b
8 KSI RESIDUAL (BENDING) STRESS.
O k
80 -
,s s
5 a-70 -
S
'./. -
o 6
60 -
D 50 -
s K
40 -
tnw h
30 -
./~
lo 20 -
s ZAHOOR MEMBRANE SECTION Xi' BENDING 10 -
SECT 10N 'XI MEMBRANE AND BENDING 0
i i
i 0
' O.2 0.4 0.6 0.8 1
1.2 1.4
'1.6 1.8 -
2 CRACK DEPTH (in):
ri, I Comparison.of Flat. Plate ond-Cylindrical Models for. Membrane l Stress.
s c._,
w
1600 VAUD TO 12 EFPY
^~ Trr"vc N # E tL
~
w s - NON-NUCLEAR WTING k
B C
1400 -
I f
I e - gLtAn (cont CRricAL) f f
1
(
I I
- 1200 -
L N 3 '=o e,
/
/ g
=s,
/
/ /
gc=
jj
(
(
9 E
600 -
f f
t::
/
f w
400
/
/
- Nn "'r J
NA won-sCLTUNt y
/ /
sOLTUP so r i
j 0
1 0
50 100 150 200 250 300 350 MINIMUM REACTOR VESSEL METAL TEMPERATURE ( F)
Pressure-Temperature Limits for Tit: Patrick through 12 ETPY Figure 5 I
13
a
.a-
- 1 F
100 c
P~= 1100 PSI I-90 -
FlNITE Fl_AW, c/l = 0.1 8 KS1 RESIDUAL (BENDING) STRESS -
"O
?
ZAHOOR (MEMBRANE) AND SECT XI (BENDING) APPROACHES l_
l; 80 -
g RTndt = 0 F x
v e.-
70.-
S TEST TEMP = 190 F-(Klo//lT) =63.25)-
m o
.....................=
if 60 -
.j -
h__
- i. '
M-z w
W
- z
- 4 40 -
~ i m
m
- w
' E
- [g 30 --
~..
- 9 = 1.667 IN
~
- 4 o
20 -
4 10 -
i; 0-0 0.2 -
0.4 0.6 0.8 -
1.2 1.4 :
1.6
- 1.8.
2
-u
~
- CRACK DEPTH -(in)-
Fig 6; Kz vs.:a. for Pr. essure: Test, FitzPatrick~ Top 1 Head' Flow : Region N --
n
\\
egqC C
- p
'E%-q,s.g a
q.
.gs 4-y 9 - lag +9o-9,94'p ya r-qcg+,pg,.
i99,p,ag..(
geg #.g
~y-+%"ag
%^.m+gp m.
%g etc s '+sagn,My us.'y9 ry eg-b-
g
-'"M%
,,epe.
sym,9 9-.g p y g p g y+,p g 9,y-g a %+- yq.9 gy g wg -. g g-gg ggg e.q,- g g my q w9ir y, p-y g.,
-gy. p y. -ingis g-vy ysiyywhgpq h my.g.--,w, 9,P y-q
I l
.e l
l l
l WPo /6 j
20 30 40 50 60 70 80 90 100 110 120 l
l l
I l
l l
l l
l l
l 1NEODfhCAL WeD WSERWD Matut ##0PASAn0w RAltaittl$ INTEWSTY 1 Q
)
16 F
flELAnomasePS Ftut A&33B/Atos/Ates as 300soe OrYSENATED WAftyR E m.S n.A acurTT l-1
~
Nats a
'l f KurTT M
,10-6 3
9 W MALL AutTT. ON g
l F e 080ADon o.oies sukPaum l
J
~
f i
r.
%v s,.
.s 10
j g
Low SULPwum I
jg*
LANT WN4 tt0VATION 01 ::
g.
i e
E Of DISPOSITON LWt POn
(
g CRACKWG (EQUATION 91 W
I
~
i 0 I 16
?
9/C 5
i
-+-
vasec enue g
4 Esth MR90 BON 5
~
6 e,wc O
put
+
10
O 9
-u id a
r m
+,
P
+
"I i
e i
I el i
I i
1 I
,20 30' 40 50 IO 70 80 90 100 110 -
gg7 f 9%
A STRESS INTENSITY Kadn 10
Figare i. Theoretical and observed (64-72.79) ersek propagation rate / stress intensity relationships for.01058 steel in oxygenated water for estrosion potentials between 0 and -100 eVebe.
15
9 1.0 0.9 -
0.8 -
- do = 0.086 in.
0.7 -
l I q
o 0.6 -
i 5
0.55 -
l O
0.5 -
j
- 0.35" o
i 2.3" a "I xo a
i 0.4 -
i il i
e r-O i
0 I h j
- 0. 2 "-
0.3 _
a = 0.55". M j
(m uoc 0~2 -
i Flow Combination f.. g dl = 18 Months mi 0.1 -
i 0.0 i
0 2
4 6
8 10 12 14 16 18 20 22 24 26 28 30 OPERATING TIME (Hours)
(Thousands) ng8 Predicted SCC Cro<.k Growth for FitzPatrick Top Head Flow SfMTZ2 DPW
.=
. = -.
...-