ML20100M526
| ML20100M526 | |
| Person / Time | |
|---|---|
| Site: | Hatch  |
| Issue date: | 12/10/1984 |
| From: | Gucwa L GEORGIA POWER CO. |
| To: | Stolz J Office of Nuclear Reactor Regulation |
| References | |
| 841210, GL-84-11, NED-84-605, TAC-55236, NUDOCS 8412120355 | |
| Download: ML20100M526 (90) | |
Text
_ _ _
'Georpia Power Company 333 Piedmont Avenue Attanta, Georgia 30308 Telephcie /04 526-6526 Mail;ng Address.-
Post Office Box 4545 Atlanta, Georg:a 30302 Georgia Power L. T. Gucwa l'e southem electnc system Manager Nuclear Engineering and Ch;ef Nuclear Engineer
-NED-84-605 Deceber 10, 1984 Director of Nuclear Reactor Regulation Attention: Mr. John F. Stolz, Chief Operating Reactors Branch No. 4 Division of Licensing U. S. Nuclear Regulatory Comission Washington, D. C.
20555 NRC DOCITT 50-321 -
OPERATING LICENSE DPR-57 EDWIN I. HA'ICH NUCLEAR FIANT UNIT 1 INSERVICE INSPECTION OF STAINLESS STEEL PIPING -
1984 REEUELING OUTAGE Gentiment Georgia Power Company (GPC) herein submits the results of the inservice inspection of stainless steel piping and the corrective actions taken for those welds reported to have crack-like indications during the Fall 1984 maintenance / refueling outage.
Enclosed is a report which details, but is not limited to, 1) the scope of exminations, 2) procedure and personnel qualification related to IGSCC detection and sizing, 3) inspection results,
- 4) flaw evaluations and repairs, and 5) future plans.
The greater part of the information contained in the enclosed report was discussed previously, although sme was in preliminary form, with NRC on Novmber 9, 1984 and Novmber 15, 1984 during meetings in your offices located in Atlanta and Bethesda, respectively.
Pursuant to the requirments of the NRC's Safety Evaluation Report issued following repairs of stainless steel recirculation piping at Hatch Unit 1 during the Fall 1982 maintenance / refueling outage, GPC sutaitted by letter dated May 31, 1984 for your review and approval our proposed plans for the inservice inspection of the subject piping during the Fall 1984 maintenance / refueling outage.
Based on your reviewer's coments concernjng exmination smple size by weld category as defined in NRC Generic Letter 84-11, the aformentioned sutaittal was supplemented, as appropriate, by letter dated Septaber 26, 1984.
While an additional question concerning weld saple size was later raised by your reviewer, it is academic now since 8412120355 041210 gDRADOCK 05000321 eda c
p i
I (
)
'f, y.
t' 1
.G orgiaPbwerd gy f/,;'
Director of Nuclear Reactor Regulation Attentions Mr. John F. Stolz, Chief
. Operating Reactors Branch No. 4 December 10, 1984-j.
. Page 'IWo E &
1 one. hundred percent' of. the stainless steel circunferential and branch connection welds 'in the Recirculation, Residual Heat Removal -(RHR),.and Reactor Water Cleanup -(BWCU) systens were ultimately exanined as a result of observing reportable, crack-like indications during the current outage.
i i
Included as part of the May 31, 1984 subnittal was a justification for continued service with six weld overlays which were applied during the
' previous: maintenance / refueling outage.. 'Ihe justification provided in the aforementioned subnittal and the favorable inspection results of.the v
- overlaid welds inspected during the current outage should aid in your granting approval for another cycle's service with those overlays.
A sweepol'et weld (lB31-lRC-22AM-1BC-1) in the Recirculation System was
- t found to have seven short, shallow axial ultrasonic indications. when the -
weld was examined during the Fall 1982 maintenance / refueling outage.
Analysis showed that this weld could be left unrepaired.
GPC voluntarily
~
installed an acoustic anissions device to monitor for any leakage from this unrepaired sweepolet weld.
GPC intends to remove the device during the current oute based on the results of the weld's reexamination,-the leak-before-break. concept, and the aucynented. reactor coolant-leakage surveillance requirenents currently in place.
While installation of the device is not a. licensing condition, your concurrence with~ its renoval is s
requested since it was discussed rather extensively in the safety evaluation report issued after the 1982 outage.
Details of the reexanination of the subject weld and our justification for removal of the acoustic anissions
_. leakage detection device are discussed in the enclosed report.
- a.
.GPC intends to return the unit to power operation upon completion-of-the 4'~
necessary repairs, analyses, baseline examinations of the new weld overlays,
,~
hydrostatic testing, your approval for continued service for another ' cycle with the' existing weld overlays,. and your concurrence with the renoval of o
E, 1
the. acoustic anissions device on the unrepaired sweepolet weld from the j;
previous outage.
Return to power operation is'~ currently scheduled for Decenber 28, 1984.
While your review and approval of the scope of our
-inspection ; plan was. a condition specified in the aforenentioned -. safety ievaluation. report, it may be a moot point since all stainless steel circunferential and branch connection piping welds in the Recirculation, RHR, and BWCU systems were ultimately exanined during the outage. Criteria for flaw evaluation, weld overlays, leakage limits, and leakage detection as they pertained to our May 31, 1984 subnittal were consistent with the intent of IRC Generic Letter 84-11 and should therefore be acceptable to you.
We l-l'
{
i
~~700775 4
. -... _ _,. ~. - -. _. _ ~,.. - _,. -,.. -. - ~....- _,..- - -. _.. ~ m.- _... _ -
.,m.
GeorgiaPower d Director of Nuclear Reactor Regulation Attention: Mr. John F. Stolz, Chief Operating Reactors Branch No. 4 Decenber 10, 1984 Page Three believe the results of the inspections and repairs progran provide an adequate basis for the safe operation of the unit.
Seven copies of this letter and enclosed report are provided for your convenience.
Since the final version of the design report will not be available until all repairs are ceplete, the enclosed NUTEEH design report is a preliminary draft version.
The final design ' report will be provided for your review when it becmes available to GPC.
By copy of this letter, NRC Region II is concurrently being provided this report to assist you, as appropriate, in your review.
Should you have any questions in this regard, please contact this office.
Sincerely yours, h,c b &
L. T. Gucwa JAE/mb Attachnents xc:
J. T. Beckham, Jr.
H. C. Nix, Jr.
J. P. O'Reilly (NRC-Region II)
Senior Resident Inspector roorn a
- =, - -
7,a ;
I.5
' 'f.f i
ENCIO6URE 1 Inspartion of Hatch Unit 1 Recirculation, RHR, And RWCU Piping Welds -
1984 Maintenance / Refueling Outage 4
-December 10,'1984
r TABIE OF COMITNTS 1.0 Introduction'
'2.0 Scope._of Exminations 2.1 -Original Scope 2.2 Expanded Scope 1.0 Inspection Procedures and Personnel Qualifications 3
3.1 Procedures 3.1.1~
IGSOC Detection 3.1.2 IGSOC Sizing
'3.2 Personnel Qualification
3.2.1 IGS(X
Detection 3.2.2 IGSOC Sizing 4.0 Inspection Results
-4.1 Welds with Reportable Indications 4.2 Nature of Reportable Indications
% ird-Party Review 4.3 4.4 Exmination of Existing Weld Overlays and Unrepaired Sweepolet Weld 5.0 Repairs and Flaw Evaluations 5.1 Flaw Evaluations and Overlay Design 5.2 h ird-Party Review 5.2.l~
Review Process 5.2.2 Results of Third-Party Review
~
12 Inch Dimeter Riser Welds 5.2.2.1 5.2.2.2 28 Inch Dimeter Suction and Discharge Welds 5.2.2.3 22 Inch Dimeter Sweepolet' to Ring Header Welds 5.2.2.4 24 Inch Dimeter Residual Heat Renoval Weld References i
6.0 Future Plans 6.1 Modifications /Replacment 6.2 Ieakage Limits and Detection 7.0 Stenary and Conclusions
-i-
~
December 10, 1984
i b
1.0 INTRODUCTICN During the inservice inspection conducted during the Fall 1984 maintenance / refueling outage at Hatch Unit 1, several stainless steel piping welds in the Recirculation and Residual Heat Reoval (RHR) systes were found by ultrasonic exmination to have reportable, crack-like indications.
i Georgia Pwer. Company (GPC) hereby sutnits the following information concerning the inservice inspection of stainless steel welds, inspection i
procedures and personnel qualifications, weld repairs.and flaw evaluations, and future plans.
2.0 SCOPE OF EXAMINATIONS 2.1 Original Scope
'Ihe original scope. of exminations at Hatch Unit 1 during the Fall 1984 maintenance / refueling outage included the ultrasonic inspection of approximately seventy five (75) stainless steel welds (included several longitudinal see welds) in the Recirculation, RHR, and Reactor Water Cleanup (RWCU) systas.
Core Spray piping and Control Rod Drive (CRD) hydraulic return line piping were not required to be exmined in the original scope of exmination of stainless steel welds since they were either of a different material type (Core Spray-carbon steel) or had been capped at the reactor vessel nozzle and rerouted to another system outside primary contairnent (CRD capped and rerouted to RWCU return).
'Ihe exminations were conducted for GPC by Southern Ctnpany Services (SCS) and its contractor, Sonics Systes International (SSI).
Procedures and personnel qualifications relative to the detection of intergranular stress corrosion cracking (IGSOC) and depth sizing of any such cracking will be discussui later in this report.
Basically, the welds exmined during the original scope of stainless steel weld examinations were esprised of three groups. The three groups involved were as follows:
Welds normally scheduled to be exmined throughout the ten-year 1) inservice inspection interval to meet ASFE Section XI Code requirments, 2)
Welds (e.g., weld overlays, unrepaired weld) required by the Hatch Unit 1
Safety Evaluation Report (SER) issued following analyses / repairs during the Fall 1982 maintenance / refueling outage and by NRC Generic letter 84-11, and 3)
Exminations connitted to NRC as a result of cracking observed at other boiling water reactors,.
(e.g.,
inconel-buttered Recirculation jet pmp Recirculation safe ends / nozzles,.
instr mentation nozzles).
Ciremferential welds exmined in the original scope typically were chosen based on crack experience, where available. Where such information was not available, high stress rule index nmber and/or high carbon content was used for selecting the welds to be examined. 'Ihe welds in the original scope of exminations included the following sizes of stainless steel piping by syste: Deceber 10, 1984
}
Ultimately,136 ciremferential and branch connection welds (includes the 4 safe end-to-nozzle welds and 2 jet pmp instrumentation nozzle penetration seal-to-safe end welds) in the Recirculation, RHR, and RWCU systes were exmined as a result of observing reportable, crack-like indications.
Of that neber exmined, 21 piping welds in the Recirculation and RER systes were found by ultrasonic inspection to have reportable, crack-like indications. Is noted in the previous section, exmination results of those welds found to have reportable, crack-like indications will be discussed in Section 4.0.
Since reportable, crack-like indications were not observed in the inconel-buttered Recirculation safe end-to-nozzle welds exmined in the original scope of exminations, the scope of exminations was not expanded for that particular ASME category weld (i.e., Category B-F).
Both of the s
Recirculation jet pmp instrmentation nozzle penetration seal-to-safe end welds were exmined with acceptable results; consquently, scope expansion for those welds was not necessary.
%e Recirculation jet pump instrmentation nozzle safe end-to-nozzle welds had been examined during the previous maintenance / refueling outage with acceptable results.
The exmination scope during the Fall 1984 maintenance / refueling outage was expanded, as appropriate, to meet or exceed ASME Section XI Code and NRC requirments.
3.0 INSPECTION PROCEDURES AND PERSONNEL CUALIFICATIONS 3.1 Procedures 3.1.1 ICSOC Detection SCS NDE procedure Ur-H-400 was qualified in October 1982 at Battelle Colmbus Laboratories in Colmbus, Ohio. This procedure was qualified under the guidelines of NRC I&E Bulletin 82-03 (for Hatch Unit 1) and later a@ roved for use under NRC I&E Bulletin 83-02. In addition, IGSOC detection methods and techniques used during the 1982 refueling outage were essentially the sme as those used during the 1984 refueling outage.
Ultrasonic examiners were rquired to record all flaw indications regardless of splitude and all geometry 50% DAC and greater.
SCS Procedure Ur-H-400 is essentially the se e procedure as that used NorE:
for detection qualification at the EPRI NDE Center.
3.1.2 IGSCC Sizing SCS NDE Planar Flaw Sizing procedure Ur-H-470 was developed incorporating techniques and methods deonstrated and approved for use at the EPRI NDE Center in Charlotte, North Carolina.
Such methods include SLIC-40 (Multi-pulse Observation Sizing Technique-MOST),
450 and 520 Shear Wave-Pulse Arrival Travel Time (PATP), Satellite Pulse Observation Technique (Spor), 500 and 700 Refracted Longitudinal Wave, and the "I.D.
Creeping" Iongitudinal Wave.
, Deceber 10, 1984
3.2 Personnel Qualification 3.2.1 IGS(I Detection Level II and III personnel performing ultrasonic exminations and/or evaluations were qualified at the EPRI NDE Center for detection of IGSOC.
In addition, Level I,
II, and III contractor (SSI) personnel were administered SCS.NDE procedure exminations on site.
Also, an EPRI IGSCC Pipe Crack Seple was prc,vided for Level I personnel to ensure an understanding of procedure requirments, e.g.
scanning, detection, and search unit location requirments, and to build confidence.
3.2.2 IGS(I Sizing Personnel performing ultrasonic sizing of indications were qualified at the EPRI.NDE Center in Planar Flaw Sizing and were certified to Level III in ultrasonics. In all, five such individuals were used on Hatch Unit 1; three fra SG and two fra Nuclear Energy Services (NES). Wird-party examiners were used for their opinion as discussed in Section 4.3 of this report.
However, their results were not used as a basis for repair decisions.
Initially, sizing was performed by SG AND SSI.
If GPC Plant Hatch engineering determined that the weld may not require repair, then finite flaw sizing was performed by NES to provide more detailed information for disposition.
Finite flaw sizing was performed at rande locations around the weld.
his approach was taken to minimize personnel radiation exposure on welds that absolutely required repair.
4.0 INSPfrTION RESULTS 4.1 Welds with Reportable Indications Of the 136 ciremferential and branch connection (sweepolets, etc.) welds examined, 21 welds (19 ciremferential and 2 branch connection) were found to have reportable, crack-like indications.
These welds and their results are tabulated in Table 4.1.
4.2 Nature of Reportable Indications of the 21 welds -identified with reportable indications,18 welds contained l
circmferentially oriented indications and 3 contained r..fally oriented indications.
The ciremferential indications were detected essentially 3600 intermittent around the ciremference.
I 4.3 %ird-Party Review Based on the scope of reportable indications detected during the early l
stages of the outage, the SG Inspection, Testing, & Engineering (ITE)
Department decided to obtain third-party review of a specific scope of welds. The third-party vendors were as follows: One cmpany identified as Team #1, Kraftwerk Union (KWU), and NES.
This third-party review was to confirm evaluations of crack-like indications and estimate depth of the
! l I
Deceber 10, 1984
r indications.
However, final evaluation and disposition of sizing results was made by SCS and/or NES. %e most conservative estimate was reported to GPC by SCS and was used in analysis for decisions concerning repairs.
It should be noted that.this was not a research effort but a means to show-that results were conservative. See exmples of the third-party review results are shown in Figures 4.3 and 4.4.
L4 Examination of Existing Weld Overlays and threpaired Sweepolet Weld i
During the Fall 1982 maintenance / refueling outage at Hatch Unit 1, 7 welds (6. ciremferential welds and 1 sweepolet weld) were identified as having
' reportable,. crack-like indications.
Analysis revealed that the 6~
ciremferential welds required repair by weld overlay while the sweepolet weld could be left unrepaired. % e welds are as follows:
SYSITM NEID NO.
Recirculation IB31-lRC-227M-1 1B31-lRC-22AM-4 1B31-lRC-22EE-1 l
1B31-lRC-22IM-4 1B31-lRC-22PM-1BC-1 (sweepolet) lEll-lRHR-20B-D-3 RHR lEll-lRHR-24B-R-13
%e existing weld. overlays were examined during the Fall 1984 maintenance / refueling outage and showed no evidence of cracking in the overlay material. They were ultrasonically exmined to verify the integrity of both the weld metal and its bond to the pipe base material, in a manner l
In addition, a liquid consistent wih ASME Code,Section V, Paragraph 1550.
and 1" of base penetrant examination was conducted on the weld overlay Any new weld overlays to be material on either size of the weld overlay.
applied during the 1984 outage were to be exmined in a similar manner.
As part of our letter dated May 31, 1984, GPC provided NRC justification for an ' additional cycle of operation with the existing six weld overlays in the Recirculation and RHR systes.
The six weld overlays applied previously at '
[
Hatch Unit 1 were full structural overlays. Five of the six overlayed welds contained only axially oriented flaws. Since.the flaws are due to IGSOC and thus depend on the presence of sensitized material for continued growth, their growth in the axial direction is restricted to the heat affected
%is means that axial flaws cannot grow axially and thus will never zone.
present a significant pipe break threat. We overlay welds consist of 308L weld metal with controlled ferrite which has been demonstrated to be highly resistant to IGSCC. With this barrier to continued IGSOC at the outer pipe surface and the resistant 304 stainless steel base metal limiting the axial growth, the axial flaws are effectively contained. With regard to the sixth overlayed weld, it had two-relatively short circumferential flaws on the
. order of 11/2" in legth with the deepest indication having a maximm depth
. of 33% of the unrepaired pipe wall. With the beneficial effect of the weld
-S-1 10, 1984 Deceber
overlay induced residual
- stress, calculations. predict that these cirem ferential cracks will have essentially no growth during their five-year design ' life. - However, even if. these calculations of the 33%
through wall sizing are significantly in error, the overlay for the joint is
_ ould accomodate much
- longer, deeper substantially overdesigned and w
Reference:
Section ciremferential flaws with no loss in safety _ margin.
(
3.1.2 of Attachment 1 to GPC May 31, 1984 letter). %erefore, GPC requests that NRC grant approval for an : additional cycle of operation for the existing six weld overlays in the Recirculation and RHR systens in light of -
their successful examination during the Fall 1984 maintenance / refueling outage. and 'our - having provided in the aforementioned letter justification for their continued service.
With regard to the cweepolet weld, 1B31-lRC-22AM-1BC-1; it was examined during the Fall. 1984 maintenance / refueling outage and found to have reportable,~. crack-like indications other than those observed during the previous outage.
%e indications observed during the previous outage were shallow, axially oriented, and lay outside of the heat affected zone of the weld.
%e indications observed during the Fall 1982 maintenance / refueling outage were not observed during the current outage.
%e area of _the indications was reexamined with advanced techniques (e.g.,
sizing techniques) to try to confirm cracking; however, no cracking was detected and the-previous indications were thought to be I.D. roll (noise) _rather than IGSCC.
We new indications were observed to be two ciremferential indications totaling approximately 9 inches-in length with a maxima through wall depth of 11%.
In addition, several spot indications were observed with-the deepest indiction being 18% through-wall.
Analyses conducted by the primary contracter and a third party indicated that the subject weld could be left unrepaired for a period of at least one fuel cycle based on flaw limits imposed by ASME Section XI Code, Paragraph INB-3640 and NRC Generic Letter 84-11.
Based on these exanination results and analyses, it is the intention of GIC to recove the acoustic anissions leakage device installed on this solution annealed sweepolet weld joint during the current outage.
While the device was voluntarily installed by GPC and was not a licensing condition, NRC's concurrence with its removal 'is requested since the device is discussed rather extensively in the Hatch Unit 1 SER issued - for - the analyses / repairs conducted during the previous outage. It is the opinion of GPC that local leakage detection is no longer reluired for this weld.
%e extranely high inherent toughness and ductility of the stainless steel-piping material; the tendency of cracks in such piping to grow through-wall and leak before affecting' its structural load carrying capacity (i.e.,
" Leak-before-break" concept) ;
and augmented reactor coolant leakage surveillance requirements currently in effect support our desired renoval of the acoustic emissions leakage detection device on the subject weld. December 10, 1984
^ 's
- +
Table 4.1 Edwin I. Hatch Nuclear. Plant Unit 1 Results of Exminations'.
1984 Exmination of Stainless Steel Piping Results-
-Weld No.
Last Exm-Fall 1982 Outage Current Results
-Recircallation Systen J12AR-F-2 Baseline Not exmined Elbow 30%*
12AR-F 11/82 Pipe-3600 I.D. Geo.
Pipe / Elbow 20-30%*
Elbow-3600 O.D. Geo.
12AR-H-2 11/82 Pipe-No Indication Elbow 20-30%*
Elbow-C.D. Geo. due to Counterbore 12AR-H-3
-11/82 Pipe-No Indication Elbow 20-30%*
Elbow-O.D. Geo. due to Counterbore 12AR-J-3 11/82 Pipe - No Indication Elbow 20 - 30%*
Elbow - I.D.'Geo. 3600 12AR-K-2 5/79 Not exmined Elbow - 30%*
12AR-K-3 Baseline Not exmined Elbow 30%*-
12BR-C-2 11/82 Pipe - No Recordable Pipe / Elbow-49%*
Indication Elbow-3600 I.D.
Geo. Counterbore 12BR-C'3
-Baseline Not examined Elbow - 66%*
'12BR-D-3 11/82 Elbow O.D. Geo. due Elbow 20%*
to I.D. Countertere 12BR-E-2 Baseline Not examined Elbow - 25%*
12BR-E-3 11/82 Elbow-3600 I.D. Geo.
Elbow -30%*
Counterbore 22AM-1BC-1 11/82 7 Axial Indications 18% Spot indication, No Overlay max. circ. indica.
tion 11%
"* Maximun detected depth - 3600 intermittent December 10, 198'4
d-Table 4.1-Edwin I. Hatd) Nuclear Plant Unit 1 Results of Exminations 1984 Exmination'of Stainless Steel Piping (Cont'd)
Results-Weld No.
Last Exm Fall 1982 Outage Current Results 22EM-1BC-1 11/82 I.D. Geo.
29% Circ.
28A-6 Baseline Not ex mined Elbow 16% Axial 28A-10 Baseline Not exmined Elbow 50%*
28B 11/82 Pipe I.D. Geo. 3600 Elbow 32%*
Elbow I.D. & O.D. Geo.
3600
-28B-4 3/81 Not exmined Elbow 31%*
~28B-11 Baseline Not exmined Elbow 49%*
28B-16 Baseline Not exmined Pipe 17% Axial RHR System 24B-R-13 11/82 Axial Scan No Indication Pipe 50% Axial Circ. Scan I.D. Geo.
3600
- Maximum detected depth - 3600 intermittent l.
I i December 10, 1984
i rn
- co l
FIGURE 4,1 j(
g 3
"A" Loop - Recirculation & Attached
_,h
<-s 3'
9 r5 l
3
,JK J.,_g.
.7 l
93 2,
- c. J K y
m w m
-*1 r tr.
[
qu y
y y
4" 4
l 12AR i
h (Typ. - 5 places) l j b h
g g
g
[3Bc-1 I
i f l
17 f3 3BC-2 F
/
1
,E. IBC-2 l lBC-1 s
"C C
Y
/
k l
/
t
(
i
)
p 5
2
-w i
- 22m I
o Ell-
{
I is OMA
{
lE11-lRHR-24A-R l
RC i
e
=
PUMP 4l N 1B31-lRC-28A B3 l-ig -
C001A-tsgyyy II 991-
'/
sF031A) j weias aepeirea witn oyeriey N/
e r
m
+
' N;/
~
B31-Welds with crack-like indi-(F02%.Wl}
t I
cations L12 13 s-Unrepaired sweepolet weld from y
lQ g
i=i
-JN-previous outage 7
8 9J Q Existing weid overlays from l
previous outage
m 9
.9 FIGURE 4.2
{
T "B" Loop - Recirculation & Attached Systems B
9 9
A E
g
-sk
-s rs
' gs 3
-s g3 g_
N;f h; E4 l
f "1
7 7
~~
0 l[_[~1
~
4 4' g.l 4g 4 g./
gf 4
4' 3
M i
h O
l 12BR Q
I f l
l h
(
(Typ. - 5 places!
=
7
+
(
-4
?.
-y-h
(- y 2
]
.- 22BM l
1E11-1RHR-l3 20B-D e
Ll l2 i s' s
RC
- 1E11-1RHR-24B-R i
PUMP j l B3 l~
C00lB-IS 9 h IB31-1RC-28B f
fl2 831-
{
./
,F03iB l
LEGEND r
x/
/
93l-t n
j
~
+
7 4
F0238 jkly i
We ds repaired with overiay g
!9 f
j Weids with crack-11ke indica-
)
gl g
- a tions j
81 9>
10 l
Existing weia overiays trom j
previous outage
con
~
6 2
.$ m W e0 is E S,$
.TEB o
E W
o g
m.a o#
5#v 8
o w
Ev*
M E 8 1030 V,a h9$l>
5
.n
.ll e
N g
N a
+
g oN m
o r-U h'
' h.
n m
O t~
N 3 DEC 10984
-~
'A
), c, 6
09 NES AS SHOWN FIGURE 4.4 W KWU* Wo-72'- 7tf 6% ;31% ^
i 1222 TEAM 88 l 15 % - 20'/o 72'-78' l
CJ MINAC POSSIBLE LINEAR 72"-7tf l
i S
l o
i b
+
90' 270
~-
ae e
1 e
c a
/g 9
- g%
IB3HRC-288-4 WELD AREA-END VIEW 18 0o RECORDED % OF THRU-WALL t
+-
5.0 REPAIRS ADO FIAW EVAIDATIONS Prior to the Fall-1984 maintenance / refueling outage, GPC contracted _ with NUIECH to perform design activities and overlay welding repairs of piping at Hatch _ Unit 1 should _ reportable, crack-like indications be detected during the inservice inspection of the stainless steel piping in the Recirculation,
Application of any, required overlays was subcontracted. to Welding Services, Inc. by NUIECH.
Activities under.the N(TIECH scope of work included, but was not limited to, fracture mechanics analysis, overlay sizing calculations, preparation of the ASME Section XI
' repairs progra, and preparation of a final design report and analysis. GPC instructured NUTECH that flaw evaluations and repairs performed would be consistent with the criteria specified in NRC Generic Letter 84-11 and were performed accordingly by the. %e following sections discuss the analyses and repairs performed by NUIET and review thereof by a third-party.
5.1 Flaw Evaluations and Overlay Design As noted in Section 4.0, 21 piping welds in the Recirculation and RHR systes were found to have reportable, crack-like indications.
Analyses performed by NUIECH have deonstrated the need to repair 17 welds at this time in the form of a weld overlay. %e welds requiring repair at this time are:
SYS794 WEID PO.
Recirculation 1B31-lRC-12AR-F-2 1B31-lRC-12AR-F-3 1B31-lRC-12AR-H-2 1B31-lRC-12AR-H-3 l
1B31-lRC-12AR-J-3 1B31-lRC-12AR-K-2 1B31-lRC-12AR-K-3 1B31-lRC-12BR-C-2 IB31-lRC-12BR-C-3 1B31-lRC-12BR-D-3 l
1B31-lRC-12BR-E-2 1B31-1BC-12BR-E-3 1B31-lRC-28A-10 i
[
1B31-lRC-28B-4 1B31-lRC-28B-ll RHR lEll-lRHR-24A-R-13 In addition, the indications in 4 piping welds in the Recirculation Syste
-were' analyzed by NUTECH and were determined not to require repairs at this time.
We welds dispositioned by analysis are as follows:
~ 1B31-lRC-22AM-1BC-1, 1B31-lRC-22EM-1BC-1, 1B31-lRC-28A-6, and 1B31-1RC-28B-16.
Included as an attachment to this report is a copy of the preliminary draft NUIECH design report on the Recirculation and RHR systes weld overlay repairs and flaw evaluations for your review. %e draft report is December 10, 1984
incmplete since weld overlay repairs were still being performed at the time of this writing.
ne final version of the NtTIEC11 design report will be submitted for your review upon its completion.
-5.2 Third-Party Review Structural Integrity Associates, Inc. (SI) was contracted by GPC to provide an independent third-party review of crack-like indications identified during the inservice inspection of the primary pressure boundary austenitic stainless steel piping at Hatch Unit 1 (see Section 4.0).
SI has performed similar third party reviews for weld overlay designs at Peach Botte 2 and 3 and at the Caorso Nuclear Power Stations.
SI has also been the principal weld overlay design and flaw evaluation contractor for several other utilities including New York State Power Authority. (Fitzpatrick), Yankee Atmic Electric Corporation (Vermont Yankee),
Northeast Utilities (Millstone 1) and Tennessee Valley Authority (Browns Ferry 1, 2,
- 3).
In addition, SI has been a consultant to GPC for IGSCC concerns at both Hatch units during the past two_ years.
%e following sections describe the third-party review function provided by SI for GPC in support of the current Hatch Unit 1 outage.
5.2.1 Review Process
%e principal subject area of SI's independent review as the evaluation and disposition of all crack-like indications in the piping resulting fra the -
ultrasonic inservice inspections.
%is included evaluation of the ultrasonic results, fracture mechanics analysis of the flaws to determine potential for crack growth, cm parison of final flaw sizes to allowable flaw size, design of weld overlay repairs where such repairs are demed necessary, and metallurgical / welding consultation with regard to the application of weld overlay repairs. ne criteria for the review consisted of the flaw evaluation procedures for austenitic piping cmponents of A91E Section XI, IWB-3640, suppleented by the reccnnendations of NRC Generic Letter. 84-11.
%e independent review included an independent verification of the design input to the flaw evaluation or overlay design. Stresses were-obtained frm a recently empleted design stress report for the Hatch Unit 1 recirculation syste, prepared by General Electric Company to the NRC I&E Bulletin 79-14 requirements (Refen.ce 1). Material properties used in the evaluation were independently
- verified, and conservative, bounding interpretations of the ultrasonic results were used in sizing of the defects for analysis and repair.
%e flaw evaluation process for both axial and circumferential defects used representative weld residual stresses for both axial and ciremferential stresses for each pipe size. Upper bound crack growth rates for weld December 10, 1984
sensitized material'were utilized in the analysis and all normal operational loads,. including thermal expansion loads, were used for crack growth.
%e L'
allowable flaw size and weld overlay thickness calculations included use of all design basis (pressure, dead weight and seimic) primary stresses.
As r+2 M+3 in NRC Generic Ietter 84-11, no credit was taken for the initial weld overlay layer for structural reinforement, and any l ~
ciretaferential indications with substantial length were overlay repaired j
regardless of their depth. Care was taken in selection of the weld material
~
to obtain Type 308L stainless steel weld wire containing a minista of 10 FN
~
ferrite and in welding by using the high quality machine Gas 'Ibngsten Arc 1
Process (GmW), with controlled heat input.
No credit was. asstmed for part-through defects in the overlay sizing, all defects were assumed to be through-wall for the entire crack length.
hus, all overlays are full structural overlays with the exception of the RHR weld overlay as discussed in Section 5.2.2.4.
As a final step in the evaluation, weld overlay shrinkage stresses will be considered when as-built shrinkage measurments are available.
Details of the flaw evaluations and the weld overlay sizing calculations are stunarized below.
5.2.2 Results of %ird-Party Review 5.2.2.1 12 Inch Diameter Riser Welds Cradc-like indications were identified in a total of twelve welds in the recirculation riser piping at Hatch Unit 1 (see Section 4.0 for inspection details).
Since. the indications were identified as 3600 intermittent indications for all joints, NRC Generic IAtter 84-11 recmeendations result in full structural overlays for all affected 12 inch riser piping weld joints.
%e resulting SI weld overlay designs for the 12 inch welds are presented in Table 5.1.
%ese overlays, as with all the Hatch Unit 1 overlays, have been reconciled with the overlay sizes specified by NUIECH Engineers, the overlay design contractor, and any differences were resolved by taking the more conservative of the two approaches.
%e overlays have been installed using machine GTAM technique and Type 308L SS filler containing 10. EN minimum.- Table 5.1 also presents overlay thickness requirements for the conservative case in which thermal expansion stresses are included as a primary stress component.
Note that in this case, the design overlay - thickness is sufficient if the first layer of overlay is included.
5.2.2.2 28 Inch Dimeter Suction and Discharge Welds
%e ultrasonic examination of the 28 inch diameter recirculation suction and discharge welds at Hatch Unit 1 revealed circumferential crack-like indications in four joints and axial crack-like indications in two joints.
Due to the depth and length of the circumferential indications and the reconnendations of NRC Generic Ietter 84-11, it was decided to overlay these four joints using full structural overlays.
As is the case of the 12 inch diameter riser welds, the welding was performed using the machine GrAW Deceber 10, 1984
- e
.s technique and Type 308L' SS. filler wire containing 10.PN minimun ferrite.
%e SI overlay sizes were again compared to the NUTECH Engineers. design and the 2 designs were in essential agreement.
Table 5.2. presents the overlay sizes for these four welds.. Once again, --if the first layer of weld overlay
.is included, the overlay; designs. are of sufficient thickness to include
~
tnermal. expansion stress as a primary stress component.
The two axial crack-like indications were of a depth so that flaw evaluation was able to deonstrate continued successful operation for at least one additional fuel cycle with these joints unrepaired.
%e crack. growth analysis - utilized representative circumferential residual stress data and analysis for this pipe size obtained fra EPRI reports (Refs. 2, 3, 4). %e analysis technique and results were again compared to the NUTECH Engineers results and agrement was obtained.
Crack growth curves versus allowable flaw size for the two joints are presented in Figures 5.1 and 5.2.
It is seen that'in both cases the indications are acceptable for a period well in excess of one fuel cycle.
5.2.2.3 22 Inch Diameter Sweepolet to Ring Header Welds Crack-like circumferential indications were also observed in two of - the sweepolet to ring heeder welds during the inservice inspection of these -
we16s.
These crack-like indications were evaluated using the flaw evaluation methodology described above.
%e weld residual stress used assumed that the sweepolet to header joints had been solution annealed ~
following welding, (a condition which was verified as a result of a review of shop fabrication records) and therefore produced a zero through thickness residual stress state.
The flaw evaluation results indicated that all of the flaws would reain within the limits imposed by the A9tE Code,Section XI, Paragraph IWB-3640 and NRC Generic letter 84-11 for a period of at least one fuel cycle (see Figures 5.3 and 5.4).
%ese results were compared to those developed by NUHDI Engineers and were essentially in agreement.
5.2.2.4 24 Inch Diameter Residual Heat Reoval Weld one axially oriented indication was discovered in a 24 inch RHR weld.
%e depth of this indication was such that crad growth calculations could not deonstrate acceptable crack growth in one fuel cycle.
nus, even though such an indication does not result in a reduction in piping syste structural safety margins, a weld overlay repair was applied to this joint.
nis overlay repair was not a full structural overlay, but instead was used
- merely to arrest further crack growth and prevent potential leakage fra the joint.
%erefore, detailed structural design calculations were not performed, and an overlay of two weld ~ 1ayers with a width sufficient to cover both weld heat affected zones was specified by NL7FECH Engineers. %is overlay was also installed using machine GTAW technique ' and Type 308L-SS filler containing 10 FN minimum.
SI is in agreement with this overlay concept for short axial flaw indications, and has used an essentially similar procedure in the past at other plants. D W '10, 1984
e REFERDKIS 1.
GE Report "Results of Seimic Evaluation:
'As-Built' Recirculation Piping Including Replacment Actuator for F031 Discharge Valve", Hatch Unit 1 Plant Piping Analysis, Design M ao 170-113, Septaber 26, 1984.
2.
R.
J.
Deuth &
B.
- Doll, "Last Past Heat Sink Welding Process Developnet", EPRI report NP-3414, January 1984.
3.
F. W. Brust & R. B. Stonesifer, "Effect of Weld Parmeters on Residual -
Stresses in BWR Piping Systes", EPRI report NP-1743, March,1981.
4.
E.
F.
Rybidci, et al, "Caputational Residual Stress Analysis for Induction Heating of Welded BWR Pipes", EPRI report NP-2662-ID, Decenber,1982.
i Decenber 10, 1984. _.. _ _ _ _ _
.s TABLE 5.1 Results of Independent Review Calculations for 12 Inch Riser Weld Overlays 1
Recuired Overlay Thickness (inches)
Mfnfmum Overl 1
w/o Thermal Stresses
!w/ Thermal Stresses Length (inches F2 0.19 0.24 2.85 H2 0.19 0.24 2.85
,J3 0.19 0.23 2.85 X2 0.19 0.22 2.85 K3 0.20 0.27 2.85
~
C2 0.20 0.25 2,85 C3 0.20 0.32 2.85 E2 0.20 0.23 2.85 D3 0.20 0.29 2.85 F3 0.20 0.29 2.85 H3 0.20 0.31 2.85 j
E3 0.20 0.30 2.85 1
j l
a e
DEC 10'1984 I
e TABLE 5.2 Results of Independent Review Calculations for 28 Inch Wald Overlays Required Dverlay Minimum Overlay
'~ Weld-Thickness finches)
Length (Inches) w/o Thermal w/Themal 28-A10
.40
.43 5.8
- 28-B11 40
.42 5.8
- 28-B3
.40
.44
-5.8 28-84
.39
.42 5.8 Actual length should be half of this length because cast stainless steel Iump side of weld was not overlay repaired.
> DEC101984
m o
.y D="i" o
g IfRBt CHNSIM CRAG GREN s 8 *- - - - -
1118-3640 ALttalA8tE C
R A7 C
E.6 l
gj 2/3 x IWB-3640 ALLOWA8tE P
f 7
I.4 i
sh
~
l i
N.2
~
d i
l
.1 n
g
^
l 68 g
IIR ( Igg gggy FIGRE 5.1 Projected Crack Gmerth Versus Allowable For Weld 28.A6 l
i I
4.-
a.
_4.
a
.,s.
-4 u..
e
%,_ a m
6 m
9 b
nem 2
i 2
i 3
~
n
\\
l f
l a
i I
8 I
gg I
l I
5 8
j i
i
=
l-a i
8 15
=
E i
! =E olii!
g 1
~g 3
g lE
=
s is I
3 a
,v 8
ja E
i I?
I
~
I R.
I m
I
\\
C I
I q*
~
l
[L I
i M
K M
N M
unamum e.nas.~m
-, DEC 101984
C3 l
g STES CIRMllE CBG CIRBI E
l W
sh C
R
=-asa ALLOWABLE i
A.7 l
0 sh D
E.5 f
2/3 x Ilm-3640 ALLOWABLE f
u l.4 i
I I
I I.3
~
i ll i
.2 f
1 I
2 -
f -
I g
=
^
8 19 29 38 48 fl E (flellA D ll0 HS) l FIGURE 5.3 Pro.jected Crack Growth Versus Allowable For Samep-o-let Weld 22-18Cl-AMI i
i b.. -
o si SfEBS CORB08IM CB G CI0llfI e
S2
}
C
~
l.I
,e w
IWB-3640 At LOMABLE D.6 E
l.5._ g 7
g ALLOWABLE I.3 l
N
.2
,[
t a
i t
8 18 29 38 (9
flNI(flellAB 1101R8) r
~
FIGAtE 5.4 Projected Crack Growth Versus Allowable For Sweep-o-let Weld 22-18ci-BMI W -
- o 3
0-6.0 FITIURE PIANS 6.1 Modifications / Replacement GPC has no. firm plans at-this time with, regard to modification / replacement.
-NRC is advised that GPC has procurred piping for the Recirculation, RHR, and RWCU: systems should it be decided to replace the existing piping-at sme point in the future. Should replacment be undertaken, the new piping would be of a similar configuration to that installed at Hatch Unit 2 during the
.1984 maintenance / refueling outage for that particular unit.
6.2 Leakage Limits and Detection our May 31, 1984 and Septaber 26, 1984 sutaittals. addressed, this point.
Proposed Technical Specification changes to augment then existing-reactor coolant leakage detection requirments were sutaitted to _ NRC by '. letters dated February 10 and 11, 1983. He proposed changes sutaitted as a result of crack-like indications being. observed during the Fall _1982 maintenance / refueling outage were subsequently reviewed and approved as discussed in the NRC Hatch Unit 1 SER dated February 11, 1983. %e proposed changes meet the intent of the leak detection and leakage limits discussed in Attachcent 1 to NRC Generic Letter 84-11.
With the exception of our requesting your concurrence of.our desired reoval of the acoustic mission leakage detection devices installed.
during the previous maintenance / refueling outage as discussed above in Section 4.4, no changes other than those discussed in Section 2.5 of Attactnent 1 to our May 31, 1984 sutaittal are planned.
J i
, t December 10, 1984
_ - _. ~ _. _ - _, - - _.,., _. _.. _..... _. _. _. _ _
- c-g-
e 7.0 SWHARY MD CONCLUSIONS o
Inspections proposed for the Recirculation, RHR, and RWCU systes piping during the Fall 1984 maintenance / refueling outage meets the intent of the Hatc'. Unit 1 SER, NRC I&E.Bulletin 83-02, NRC letter SECY 83-267C, and NRC Generic Letter 84-11.
o As a result of observing reportable, crack-like indications in the original scope of exminations, the exmination scope was expanded to include 100% of the 136 circtnferential and branch connection welds in the Recirculation, RHR, and BWCU syste s.
o
'Ihe detection and sizing of IGSOC was performed by qualified inspection personnel.
o
'1Venty one (21) welds were observed to have reportable crack-like indications in the Recirculation and RHR syste s.
o Analyses revealed that 17 of the 21 welds rs;uired repair at this time in the form of a weld overlay while the reaining four welds were left unrepaired. A stmoary of weld disposition is as follows:
PIPING NLMBER DISPOSITION 12" Recirculation 12-Circs.
Weld Overlay Risers 22" Recirculation 2-Circs baluation,leftun-Sweepolets repaired 28" Recirculation 4-Circs Weld Overlay 2-Axial Evaluation, left un-repaired 24" RHR l-Axial Weld Overlay o
Flaw evaluations and repairs were performed consistent with the criteria specified in NRC Generic Letter 84-11.
o
'Ibird-party reviews were conductai in the areas of nondestructive exmination (detection / sizing) and flaw evaluations / repairs.
o Continued op ration with the existing six weld overlays in the l
Recirculation and RHR systes is requested and justified based on their acceptable examination results during the current outage and GPC's justification provided in our May 31, 1984 sutaittal to NRC.
o While crack-like indications were observed in unrepaired sweepolet weld 22AM-1BC-1 during the current outage, the indications were other than those observed previously. 'Ihe previously observed indications were no
[
. December 10, 1984
o g.
y J.
longer present.
The area of f indications was reexmined with v ianced techniques _ (e.g., _ sizing techniques) to try to confirm the 3
cracking;
- however, no cracking was detected and the previous.
M
-indications were thought to be I.D.
roll (noise) rather than IGSCr..
'l The :new indications were ciremferential totaling approximately 9 inches in length with a through-wall depth of 11%.
Several spot indications were observed the deepest of which uas 18%.
Based on the unrepaired sweepolet exmination results discussed above, o
the leak before break concept, and augmented reactor coolant leakage surveillance requirments currently in place, GPC intends to reove the acoustic emissions device installed on _ the subject weld during the current outage and rquests NRC's concurrence in its reoval.
o Regarding future plans, there are no firm plans at this time for modification / replacement, however, piping has been procurred should it be decided to replace the existing Recirculation, RHR, and RWC11 systes piping at sme point in time.
'With the exception of our requesting NRC concurrence of our desired o
reoval of the acoustic missions device on the previously discussed unrepaired sweepolet weld, no changes to leakage limits and detection other than those discussed in our May 31, 1984 subnittal are planned, jo Upon empletion of the necessary
- repairs, analyses, baseline exmination of the new weld overlays, successful empletion of the hydrostatic
- test, and granting. of requested approvals and/or concurrence by NRC concerning continued se' vice with existing weld overlays and reoval of the acoustic missions leakage. detection device, it is the intention of GPC to return the unit to power operation on or about Dec e ber 28, 1984.
GPC concludes that the inspections and repairs progra conducted during o
the, Hatch Unit 11984 maintenance / refueling outage provides an adequate basis for the safe operation of the unit.
+.3
(,
4.
,iY
( i i
-Deceber 10,-1984-
6
- n...
43 XGP-09-106 Revision A December 1984 XGP009.0106 DESIGN REPORT FOR t?.\\
EVALUATION AND DISPOSITION L'i
,s
~
OF IGSCC FLAWS AT Q.]w;Q as PLANT E.
I.
HATCH UNIT 1
- .h W-A2 Yp3 v
- h % p;'d p.
er:7,.
.cu.
f::. "N; 3x n:,t3
's u v
v.
vs Prepared for:
Q
- $j e,
Georgia Power Cobhany Q yj7
,e, GA S
em e1 y *.,. j Prepared bp:' #
[:/..
NUTECH Engineers, Inc.
S an'fJi>se,S3 California
.y
,a. y v.
- , *.'t 4
ff f,_
W
".?
[
f v;[,_
Issued by:
Prepared by:
'Q ty T'
(([?
1 Acoj[cjb,P.E.
H. L. GustindP.E.Ya B.
D.
Technical.Leadern D'
Project Matrager Applied Mechanics 2'
";F I
Y y #s_
.]
Approvedcby:
w
.:3 Date:
8 i
T. J. Wenner, P.E.
Engineering Manager Structures & Materials Engineering
N 9
REVISION CONTROL SHEET 4
9 s -ts
. ~,,
k.s t
n
.4 9..
k.^y
[
f h *
~
wri h
'e
'?
+
I n
rs 4
i' 2
(
%W" 3
I L
4
+
1
').
r k
i l
w I
i i
t
(
i
(
i XGP-09-106 11 N
Revision A
A
.-s.
.,.,A gb E
CERTIFICATION BY REGISTERED PROFESSIONAL ENGINEER em
~3 a,
P3 E_ff
- %.*, i
%i,, +a. %.
4'..
y@<..,. 4rm.
t$ a a
e, -qgs s
[,.Y'9 Yr,k, c'll wd4
- E,'
y-5
_.s
~
w 7
h.f-!
.?
ff%.v, y::. ea s%., %.
%ch
.'.qr -w j(,$
s%lh Y
-..a p.,
tA va a
tv po
..?-YRt AI..D-5!$
3 f_ f. -:-s4
%.5;%.,;> u ?
p
.w
- s. : c:
C ir r
My
$'i.,'
(.A 6;AlE u.,-
- t. g
- n.,
Y'Q-fyr//
gy
.*A
%w:p
~a
[
.h*
f >f '
,[~
,, > h,%,
v +A vgv
\\g 4
'., fx
=
/ ;p
%g%f
- ,5
. qj' A
Y 't.,1.;..-a,~fd
?
-p ws
)
- h g?'%
5:g Vp-w s.wA,, "q/
'L5 t -> e Q:y',.v{.p+ *%Q?y
.f~$ %
"(L; 2
I
-pf ~
- 4. ~'jA
',. = *l; y-
,_f f
w N #f y '.
n't
%s' *f
%Q^.y, PYg %
vg-.dg g.99 g, s,
- -g
-._., g g,y'
' .' :\\
y')
r-Ns 3:4
%f..
.'? Y
'u.
2, I
y i
l, e
XGP-09-106 iii Revision A
,=
w es-
.e -w o rw + y r
+---ry
---~r--ye e,-mm-rsw,-T-y-ww-y--y-w,s-wewe,--=vn-*y-k-=c--w-,~r--w-w.e-w--e-e,-ent-,a w
- r-ee.-,e,v"~
ew=.,--e- -, + + - w w emev ww-
--v---v
s TABLE OF CONTENTS Page LIST OF TABLES LIST OF FIGURES 3
15
1.0 INTRODUCTION
h
,rlNY j, Q, NK) 2.0 REPAIR DESCRIPTION ga m
3.0 EVALUATION CRITERIA f."
fd NU 3.1 Weld Overlay Repair Criteria
, #{~%' % fh T.X) 4 3.1.1 Strength Evaluation "tl.$
3.1.2 Fatigue Evaluation
. Q gj 3.1.3 Fracture MechanicsiEvaluation/
s f -1
.-Q 3.2 Flawed Pipe Analysis Criteria Q 3
a x
4.0 LOADS
%,[],^ ' '
4.1 Mechanical and Indernal Pressure Loads f
6 i 4.2 Thermal Loads ~
.Ns s
,g
%f 4.3 Weld 05erlay Shrinkage-Induced Loads j
5.0 EVALUATION METHODSfAND RESULTS
~
_fi.
y Descripti;onfof Geometries Analyzed 5.1 pn y'.
5.2 ' Code Stress Analysis
.v
,. =.
5.3N Fracture Mechanics Evaluation fw.s
[P5I4, Treatment of Axial Flaws 0
s qM -5.5f, Effect on Recirculation Systems
.?
525 Evaluation of Flaws in Unrepaired Welds XGP-09-106 iv Revision A f
--,e
,,--.,-..w.
..-v.,
---v
.2
. TABLE OF CONTENTS (Continued)
Page 6.0 LEAK-BEFORE-BREAK ASSESSMENT A(5 fy. N 6.1 Net Section Collapse b y?t 6.2.
Tearing Modulus Analysis "k
g
,y.. ~2 %
'Q 6.3 Leak Versus Break Plaw Configuration:
ef s
e
.-.es
- g..%e ;f 6.4 Axial Cracks
-m>
' i
,:q,s n g x3 c :>.
y 6.5 Multiple Cracks D]
yd W4 %-e 1rl b
6.6 Nondestructive Examination S%
,, y ~ J
,uy g')g i
6.7 Leakage Detection w.
/
6.8 Historical Experience i y'/v w-
,3 7.0
SUMMARY
AND CONCLUSIONS.[
J t 's, A.*y
8.0 REFERENCES
y s
g>
1 w
~ ;4
.h
..14
- i' y
[?,p, c x..
.*; n, 7 "
'::g.
'y en a %
T 3 f.a wr i
.;n y ~,.
',)
g.t.g 9-
^
. [e:-
w, i
se t
..3..'
"7 u.
f (j.
- S t
..y u-
+ [: 1
+'
N.'.. ?
Y -..,
,;~.
sq-
..s
. + i
%. 6,.?
g>
XGP-09-106 v
Revision A
?
-a,.
LIST OF TABLES Number Title Page 1.1 Plant E.
I.
Hatch Unit 1 Fltw Disposition -
4 Fall 1984 Outage f[
1 Flaw Disposition," i \\u
-id*t g 1.2 Plant E.
I.
Hatch Unit i
%ny Fall 1982 Outage
,es
- .,<p
'?[j{s,[7 f
2.1 Weld Overlay As-Built Dimensions R;dh 5.1 12" Pipe-to-Elbow Code Stress ResUlts
\\
w+p g,
x;z;r,-,
ts 5.2 28" Pipe-to-Elbow Code Stress Results?!}
ts z
nn.
v :!m
,2;f 5.3 28" Elbow-to-Pump Code S, tress"Resg{ts' ll r
5.4 24" Tee-to-Valve Code Stress Results t's,6 5.5 22" Pipe-to-End Cap Code Stress Results v
5.6 24" Pipe-to-Pipe; Code Stress Results J
4 '
4-5.7 Thermal Gradient Results
%?
m 5.8 Plant E.II4 Hat'ch Unit 1 Weld overlay Induced Shrinkage' Stresses (Later) fN N4 (
6.1 Effect(offPipe* Size on the Ratio of the Crack Length?for45"GPM Leak Rate and the Critical tcrack' Length (Assumed Stress s = Sm/2 )
j
<e 3
s
~v 1.
xo
..L c > ;,;,
,e w C
4
. - +,
o
- w
'5)x i
~
f,=,f,
.XGP-09-106 vi Revision A
=
3 LIST OF FIGURES i
Number Title Page 2.1 Typical Configuration of 12" Elbow-to-Pipe 4
{$
v, 5.1 Finite Element Model of 12" Pipe-to-Sweep $let'id
~"
Weld e,
,.N 5.2 Weld Overlay Thermal Model 9
kf
,7' 5.3 Thermal Transients
-I '~ s y 3'g %)>
p; 5.4 Typical Axial Weld Residual Stress i
Distribution in 12" Pipe Af_ter Overlay 3
[f 5.5 Plant E.
I.
Hatch Unit l', Recirculation System Piping Model
{1 73 en g,
6.1 Typical Result of, Net Se'ction'.2 Collapse Analysis of Cracked Stainle's's Steel Pipe
~
6.2 Stability Analysiasfor BWR Recirculation System (Stainless' Steal)
. ?'
'j'
^
6.3 Summary of 5 Leak-Before-Break Assessment of BWR Recirculation System fs.
Typi' cal 3 Pipe'X 6.4 Crack Failure Locus for Combined Through-Wall?Plus 360' Part-Through Crack v
t
"2
'd}
v
%;. c 4 =
4 XGP-09-106 vii Revision A nutgich
1.0
-INTRODUCTION This report summarizes the analyses performed by NUTECH to establish the basis for weld overlay repairs and to evaluate unrepaired flaw indications in the (t1 RecirculationSystematGeorgiaPowerCompanyYd[FNantE.
M?5 N74 s
I. Hatch Unit 1.
Ultrasonic (UT) examinatioelof'the'
- QA
/Y weldsinthissystemduringboththeFall$1982andFall
,fQ 'gg 1984 outages identified flaws which'werehjudge' d to be
- in cracking AIGSCC).
intergranular stress corrosion $$ ' Q Y f}
61
(
m Weld overlays have been applied to 17 of these welds
~%?5N
duringtheFall1984f0'utage.
The purpose of each w
ns overlay is to arrest-Many further propagation of the fy Nijh cracking and,;to"restoresthe original design safety margin to thd*% if fweld.
All of the flaws are in Type 304
/2 s.
't@s stainless?sseel material.
Tables 1.1 and 1.2 contain a U />@
description?of each flaw indication as well as its 4 "p;
'ig
/ disposition.
-g
.y
,c % ~; g 6: / r
- q
- -A Y
J r
N;p,,a.
Flaw evaluations have been performed for 4 welds during
,a
~\\lf,ithis outage which were determined not to require weld overlay repair.
The purpose of the evaluations is to assure that the original design safety margins for these welds has not been degraded.
Tables 1.1 and 1.2 contain a description of these flaw indications.
XGP-09-106 1.1 Revision A
Tablo 1.1 PLANT E. I. HArOf UNIT 1 FLAW DISPOSITION FALL 1984 OLTrrE Overlay Design
-a c-3 Weld Number Flaw Description tN
. I/2(2) wa tj s
hf 1-B31-lRC-12AR-F-2 Cire. 20-30% x 360*
(E ; 0.23-2.0 f~7% [s0[23 2.0 1-B31-lRC-12AR-F-3 Cire. 20-30% x 360' 1-B31-lRC-12AR-H-2 Circ. 20-30% x 360' A 7 " c% 0.23 2.0 lQ, M..j0.23 2.0 l
%\\
1-B31-lRC-12AR-H-3 Cire. 20-30% x 360*
1-B31-1RC-12AR-K-2 Cire. 30% x 360'
-1 D Y_
/ / 0.23 2.0
'wr
,r
~
1-B31-lRC-12AR-K-3 Cire. 30% x 360' ;,
0.23 2.0 4.)
r.1 1-B31-lRC-12AR-J-3 Cire.20-30%x360*(
0.23 2.0 1-B31-lRC-12BR-C-2 Circ. 20-30% x 360 N,-
0.23 2.0 1-B31-lRC-12BR-C-3 Circ. 25% I 360' O.23 2.0 1-B31-lRC-12BR-D-3 Circ.(20% x 360' O.23 2.0 1-B31-1RC-12BR-E-2 Cire. 25% x 360*
0.23 2.0 1-B31-lRC-12BR-E-3
' Circ.dk0%x360' O.23 2.0 1-Ell-lRHR-24AR-13; Axiab50%x1.75" two layers (2) 1-B31-1RC-28A-10 k [ Circ.'50% x360*
0.42 4.25(3) 1-B31-lRC-28B-11 5,7 SC'hc. 49% x360' O.42 4.25(3) i 1-B31-lRC-28B-3 I',
.Cire. 32% x 360*
0.44 3.0 1-B31-lRC-28&-4 ~ "@
Cire. 31% x 360' O.44 3.0-1-B31-1RC-288-16 Axial 17% x l' No Overlay 1-B31-lRC-28A-6 Axial 16% x 0.5" No Overlay 1-B31-lRG22AM-1BC-1 Cire. Under 10%
No Overlay 1-B31-lRC-22BM-1BC-1 Intermittent Circ.; Max. 30%
No Overlay Notes:
1.
The effective thickness, t, is exclusive of the thickness of the initial layer.
2.
I/2 = C/2 + 0.75" on valve side; I/2 = C/2 + 2.25" on tee side.
3.
I/2 is entirely on the elbow side of the groove weld centerline.
XGP-09-106 1.2 Revision A
Tablo 1.2
_PLMrr E. I. HAIO{ UNIT 1 FIN DISPOSITION PALL 1982 OUrA3E Overlay Design t$
t 4($^'%p.,N d I/2 Weld Number Flaw Description yp l-B31-lRC-22-AM-1 Axial 63% x 1/2" e J 0.25'
.s.y 1-B31-lRC-22-AM-4 Axial 72% x 1/2"
%0.25
[1 kb.25 l-B31-lRC-22-BM-l Axial 64% x 1/2"
(
I??0.25 l-B31-lRC-22-BM-4 Axial 67% x 1/2"
,, TO
[:~f.,']0.4 l-Ell-lRHR-20-BD-3 Axial 94% x 3/8" 3.5 l
Circ. 33% x l-1/2"li A
0.4 3.5 a
1-
~
d 1-Ell-lRHR-24-BR-13 Axial 47% x 1/2" 'a 0.3 4.0 M y,.
7:
.t,.s y n
- I/2 = 3.0" on pipe sider I/2 = 3l5" on end cap side
,y
.k
+
s 4
j
),
~
"( hs,.
1 w
\\ a-f
'",+ '
m
,s
~/
,jr s
s j
XGP-09-106 1.3 Revision A
2.0 REPAIR DESCRIPTION The UT flaw indications' requiring repair have been remedied by increasing the pipe wall-thickness through hither the deposition of weld metal 360* around and to(N side of the existing weld.
This is shown for%4aatypical
%Q gm geometry in Figure 2.1.
The weld-deposit'ediband pro-
?![ ll$
i vides additional wall thickness to restorefthe original
- p % %
~
design safety margin.
In addition / the$ weld,ing process PT produces a strong compressive redihual dtress pattern on f$ll*
' ldywQ'
'~
the'inside portion of the pipe wall which prevents s
iM 93 further crack growth.
The' deposited weld metal.is. Type
_~m,
+=/r 308L with controlled; delta ferrite content so as to be.
/ ~L resistant to propagation of IGSCC.
Design and as-built lb'
%; ~
information for all overlays.is shown in Tables 1.1 and
( i,.1) 2.1.
"J"
/
1" m
%l%
3/
' { c A,
. a,
%h Tho r nonde'5str'uctive examination of each weld overlay c
6 ~43
/ consisted of:
, ?:;a p..
[
n *'t-g
_ 3.. q.
- 's 1)
Delta ferrite content measurement of the first t
- )
c "y
f overlay layer.
x
< i V
2)
Surface examination of the completed weld overlay by the liquid penetrant examination technique in accordance with ASME Section XI (Reference 1).
XGP-09-106 2.1 Revision A
i h
S 3)
Volumetric. examination of'the completed weld overlay by the ultrasonic examination technique in 4
accordance with Plant E. I. Flatch ' Procedure.
,j b
4
<4;o,.
tQ l
,3
. AqN ;q
%;v e%.A t
$-m M;h) qf 4
[l CX 4
%.v % ay
<y
{l" %Q ~Q().
4
'bg
.r:4 g
i v
V, y3
,c; Yl*.q u a f,f~ r) l f.%%
.g
,e v
< *,y:4 w
+
1 Fi gh i
yn
. o) i, b 7'
t 4 A.
52/
y~
c r S fiOY e
1R i
4
,9 /
l p
a
'W
,.g-
,3 eg;/ n.s.,z-s y
'n h
/g
/T/
%b tv
>-.7 f,
y
~,;ft;a
,3 ga y
4 n.
L :%
Tc. A
.N,A 4,4 I
- 5..
s t.,. N v
n
.. ;x.,/
l 4;h,(jhr q v,7! g y,,' [
3
,/T %,b e
- 5. A l
8.."*
- i stv i;.#
,.1 v; *y 4
y qs 4
' 4,, p4 ;
r xs t e
+
, 2 w--
1
'l-
,.f??h a,.
a.
,1 l4 h,..f q-N 'g V
i
? f s
a p.
- it s;.x
- f
(
\\; i<! f y;
i f
f r
i I
I.
I r
XGP-09-106 2.2 r
Revision A i
i e
L
Table 2.1 WELD OVERLAY AS-BUILT DIMENSIONS A
As-Built As-Built Weld I.D.'
Thickness Le'nath
'Wyf <:~it
~
~Rl \\
4".up N;p 1
5 p
l-B31-lRC-12AR-F-2 Later O[s
[f Later acg 1-B31-lRC-12AR-F-3 iX
%" %;)g %j>
l-B31-lRC-12AR-H-2 ff 9
(x
,f
[f;h' %g <
1-B31-lRC-12AR-H-3
/:,
a 1-B31-lRC-12AR-K-2 34
'J::1 r
w
. c?
""k?
y l-B31-lRC-12AR-K-3 n.+;1
,e l-B31-lRC-12AR-J-3 f(
m 1-B31-lRC-12BR-C-2
,9
't
,y 1-B31-lRC-12BR-C-3
'w r
i.
~
+ ;'
l-B31-lRC-12BR-D--3,
'O,.
t.
p 1-B31-lRC-12BR-E-2i
, Jj'>
v;yl 1 1-B31-lRC-12BR-23
\\
(~
l-Ell-lRHR-24AR-13" l-B35-l'RC28A-10 A
~
. o,.
3
-1 1-B31-lRC-288-ll l-1-B31-lRC-28B-3 1-B31-lRC-28B-4 o
y XGP-09-106 2.3 Revision A
Figure 2.1 TYPICAL CONFIG. RATION OF 12" ELBOW-TO-PIPE WELD OVERLAY 4
c f:
,s M ~~ -
f W TYPICAI. #
% fr TYPE 3mL WELD OVERLAY
S
/>
t[;;
v.
k i
2.0" WN =
,' ?~ '
A 12" 0.508" THICK a
l WALL 304 SS PtPt a tse-rr i,1
?
p s.o~ mN
,o TYPICAL CRACK i >
AsWeLoan suePAes LoNo aamprAats PoR my7g >
RActus staow l
PCouse.o2 t
(
PATENT APPUSD POR l
i l
l l
XGP-09-106 2.4 Revision A l
nutggh
O a
3.0 EVALUATION CRITERIA This section describes the criteria used to establish the acceptability of the weld overlay repairs and flawed pipe analyses.
All evaluations and repairs wereNi performedinaccordancewithNRCGenericLetddrSk-ll, m.
dated March, 1984.
" is
'(I
&?5 %, _ JA
. m,
~,f 3.1 Weld overlay Repair Criteria
/[ " '!h ' ?.)
II iIt
,4 7, $(, /
~>
- v Because of the nature of thes'e repairs 7 the geometric l',
Ej configuration is not specifically{ covered by Section III of the ASME Boiler an'd Pressure'Nessel Code, which is intendedfornew.;bnotruction.
However, the materials,
- n
+
fabrication procedures > Land quality assurance require-1 7
ments used forfweld overlay repairs are in accordance withaphl{ Sable'sectionsofthisCode.
The intent of thedesign$riseriaistoassureequivalentmarginsof
{
- 3
.t
, safety-for strength and fatigue considerations as
.provided'in the ASME Section III Design Rules.
In c>
,~
'addidion, because of the IGSCC conditions that led to
,3
}<(the need for repairs, IGSCC resistant materials have been selected for the weld overlay.
As a further means of assuring structural adequacy, criteria are also provided for fracture mechanics evaluation of the repairs.
4 XGP-09-106 3.1 Revision A
Highly conservative assumptions were used for all evaluations.
Axial and circumferential flaws in pipes with diameters less than or equal to -12" were assumed to be through-wall for the measured length and were evaluated in accordance with the criteria of References
'O.,
I and 2.
Flaws in larger' pipes were evaluatid?:ini
"% ^
g v.
accordance with the requirements of NRC _ Letter 84-11'.
?.
NA N;+,4.g
~
2N 3.1.1 Strength Evaluation
['
s,['3 Np i:
Ei v
- T'p *, yty The structural adequacy of.thh weld ~'ove'rlay repairs from f
0*
(0 a strength viewpoint with respectyto applied mechanical
m
+
loads was demonstrate ld by performing ASME Boiler and il Pressure Vessel CodebrSection III, Class 1 (Reference 3)
.r;f N;s analyses.
These analysi,s bounded each weld overlay
.r.
u n
' '; i repair.
i y"",7
.a
,h s
G) 1, 3.1.2 Fatiaue Evaluation 5
)
4 l
b i
TheJatress va ues o ta ned from the above strength evaluation were combined with thermal and other 4 -
' secondary stress conditions to demonstrate adequate 4
fatigue resistance for the design life of each repair.
The criteria for fatigue evaluation includes XGP-09-106 3.2 Revision A i
P 1.
The maximum range of primary plus secondary stress was compared to the secondary stress limits of Reference 3.
A 2.
The peak alternating stress intensity, including 1a w
all primary and secondary stress terms andfaj,s(
m..
NA fatigue strength reduction factor ofC,,520hto account 6x f?'
for the existing crack, was determined [using
,c w y y conventional fatigue analysisitechniqu'es';
The M
Q total fatigue usage factor,~Ofinedt as the sum of d
. - - 9,
~ 3 ny the ratios of applied.mmmh'er of"cy'cles to allowable IR i3 t
number of cycles at sich,..stres's level, must be less v
than 1.0 for the) design life of each repair.
The allowable number}of cycles was determined from the A'
stainless $s' teel f'a igue curve of Appendix I of "I > / l' Reference 3.'
J' 9:N, p
ri
')
s 3.1.3 Fracture Mechanics' Evaluation v
1 t.. a s+
l T
/
- ,y s'"^A highly conservative nethod was used to demonstrate the f.
'1 g,
(l.
adequicy of the weld overlay repair.
The growth of the j
m L'/hssumed flaw was calculated using a conservative crack 7
growth correlation combined with the predicted residual stress and applied stress distribution.
The weld overlay was designed such that the net section limit XGP-09-106 3.3 Revision A
would be satisfied for the predicted flaw size.
3.2 Flawed Pipe Analysis Criteria The flawed welds which were determined not to require weld overlay repair meet thecriteriagiveninPfhagraph m
IWB-3600 of Reference 1 and Reference 4.
A highlyfs Q,*
,o n, conservative crack growth law has been usedtto sp f:
"?
demonstrate that IGSCC induced flaws will$not grow to a
~~1 criticalflawsizeduringthenext'fue1{cyc1hi
~
il
\\M U%
1)
,,,,,, (
2./
.c F'.
C:1 A
I2 'i Hi t;
v,
'~
- f.,
f f
.y
- s. :5
/
?.
r
_f l g
9
-y' ry
'i s'
- ^$,
i
,e ",.
I l
.s s
}
's' i
? -j v
XGP-09-106 3.4 Revision A
4.0 LOADS The loads considered in the evaluation of the UT flaw indications included mechanical loads, internal e
pressure, differential thermal expansion loads,t,and weld
- )
As overlay shrinkage-induced loads.
The mechanical <and "n\\/
,, s internal pressure loads used in the analyses 're a
4" J
described in Section 4.1.
A discussion of:the thermal m
+
transient conditions which cause differential
- thermal 4
h expansion loads is presented in,5Section3.2.
The loads v ~
induced by weld overlay shrinkage are" discussed in L4 f
Section 4.3.
(' f;)
e,
%;._.u.
i 4.1 Mechanical and Internal Pressure Loads 7;
V
~
The design pressures of 1325 psi (discharge side) and
<'s 1050 psiE(suctionJside) were obtained from Reference 5..tThe deadwe'i'ght and seismic loads were obtained from (Reference 6>.
[
y
\\
w 4.'2 Thermal Loads 2
The thermal loads for each weld were obtained from Reference 7.
4 XGP-09-106 4.1 Revision A
4.3 Weld overlay Shrinkage-Induced Stresses Each weld overlay causes a small amount of axial shrink-age underneath the overlay.
This shrinkage induces
<4 bending stresses in the remainder of the piping " system.
.@w.s.
These shrinkage-induced stresses are calculatedcusing f?
NUTECH computer program PISTAR (Referencef8fijiThe #
%g ll actual as-built shrinkages are used,in thejanalysis.
m The resulting stresses are included in % y.y fjn; the crack growth (L i es
- y. -\\
f.!j analyses.
. q 3 A.jg
,~;;
ua w
'~
[!
j}
'l'N.ll m
i:
..G 9' ' ( q.
4
. b. '
N J, >
e e2 e
.A rf
's
/
a b
4 x
s L.
s
%l
'8.,
a-
}-};g, g.
m s s'
f f.'-
y
( O.,6 - N y,
'+
. "'y 27-( 4 v
<A y :3
- sj
.t !
s-s
<a s
1
'Ig
XGP-09-106 4.2 Revision A
i 5.0 EVALUATION METHODS AND RESULTS l
The flawed welds shown in Table 1.1 were identified by UT inspections during the Fall 1984 outage at Plant o
E.
I. Hatch Unit 1.
The flawed welds shown in Ta'le 1.2 b
wereidentifiedduringtheFall1982outagek$NUNit,1.
^
For each flawed weld, the methods of Sect' ion [3 were c
,~
applied to determine if an overlay was r'equired to meet
,;c.
a therequirementsofReferences1,and4;>,[N iI' A
V
??
5 c'
~v The evaluation of each weld overlay Ye' pair consists of t- \\
'd a Code stress analysis per'Section/III (Reference 3)
+
and a fracture mechani~cs evaluation per Section XI (Reference 1) and; Reference 4.
The flawed pipe analysis
/;
was performed pernReferences 1 and 4.
s
,,, ^ 3 The application,of' weld overlays produces a small amount o,
, s.-
of-axial shrinkage which in turn imposes steady state
' i
(,
" secondary, stresses in the affected systems.
Analyses to fl7' quantify this effect and address its significance are
'k '
'f discussed in Section 5.5.
?
W 5.1 Description of Geometries Analyzed Six distinct flaw geometries required weld overlay at E.
I. Hatch Unit 1.
Those were:
12-inch diameter pipe-to-XGP-09-106 5.1 Revision A
elbow-(14 cases), 28-inch diameter pipe-to-elbow (2 cases), 28-inch diameter elbow-to pump (2 cases), 24-inch diameter tee-to-valve (1 case), 22-inch diameter pipe-to-end cap (4 cases),'and 24-inch diameter pipe-to-4 pipe (1 case).
Code stress and fracture mechanics e
evaluations for these cases (which envelope th'e%/i
- Q geometries listed in Tablesl.1 and 1.2) areisummarized 48s Il in Tables 5.1 to 5.6.
Analysis results'forfthe e3 xj locations with axial flaws only are? discussed >in Section A
~
'f ll 5.4.
- y y
l-n L
5.2 Code Stress Analysis
,N, ]l
- L%
g Finite element modelslof the weld overlaid regions were
?'
.\\
developed using #>he ANSYS" (Reference 9)
/
t computer E
=t program.
The"models (Figure 5.1 is an example) were e
x based on aIcomposEte worst case flaw and on design
\\"i: l, minimum overlay
- thickness.
The as-built thickness is 7
s$
g'reater'thsn or equal the design thickness.
The
, ' ' s ;f metresses in the overlaid weld due to design pressure and appliid moments as described in Sections 4.1 and 4.2
',' were calculated using the models.
s.
The results of Code stress analyses per Reference 3 are given.in Tables 5.1 through 5.6.
The allowable stress XGP-09-106 5.2 Revision A
values from Reference 3 are also given.
The weld overlay repair satisfies the Reference 3 requirements.
The weld overlay thermal model was taken to be axisym-a metric (Figure 5.3).
The exterior boundary was* assumed J4 The temperature distributioks[in(the to be insulated.
r w,
'..t weld overlay, subject to the thermal transients defined c
s in Section 4.2, were calculated using Charte 16 and 23 r
2:%
t ), '
of Reference 10.
E' s
N i
e.,_.,f
,. %j The maximum thermal stress :for useain"the fatigue i
b:
analysis was calculated using s.the(following equation s q,
EmaT Be&T y
2 e erence 3) a = 2(1 vin lv
+
l '.2 V
,r g.
,1 where:
t ;,f g 9/
s
+.r 6
{28.3 x 10 psi (Young's Modulus)
E
- =
g
- e~
2 :.
=.",',,h[Ilx10-6.p-l
- a,
(Coefficient of Thermal Expansion) g v
AT Equivalent Linear Temperature Difference
=
y Peak Temperature Difference
'AT
=
2 0.3 (Poisson's Ratio) v
=
The values of AT1, AT, and o are given in Table 5.7 for 2
the thermal transients.
XGP-09-106 5.3 Revision A
ir A conservative fatigue analysis per Reference 3 was performed.
A fatigue strength reduction factor of 5.0 l
was applied due to the crack.
The fatigue usage factor A
i was then calculated assuming the thermal transients These results are'alson\\
i discussed in Section 4.2.
[\\
+s summarized in Tables 5.1 through 5.6.
'p')
'u t xv g% NiN 5.3 Fracture Mechanics Evaluation f'
^%)
.;g
+
3 IN g ~,,,. 5 q q,e
'q
~
r r
The allowable crack depth was' calculated based on il References 1 and 4.
Crack growth 4due to fatigue was
+ DV determined based on.RSference 11.
Calculation of crack growth due to IGSCC was, based on References 12 and 13.
,,)
Q Y
s,
,x From Referende[11$ the calculated fatigue crack growth
/..
1 -
is less 'than-0.010 inch during the next five years.
The weld overlays applied to welds at Plant E.
I. Hr.tch Unit
'\\
1 produce a highly compressive residual stress pattern
% ^ r; r,y ein' the inside portion of the pipe wall.
For the
.)
,f-circumferential flaws identified, this residual stress ~
Epattern will effectively arrest further IGSCC crack growth.
Thus, no further degradation of the integrity of these welds is anticipated.
Crack growth analyses using the calculated residual stress pattern (Figure l
5.4) and a conservative crack growth law (see Section XGP-09-106 5.4 Revision A
5.6) were performed with NUTECH computer program NUTCRAK (Reference 13).
I Growth of a flaw due to IGSCC requires a susceptible material.
Since the cracks found at Plant E. I.[ Hatch is Unit 1 are assumed to go through the original'.:wallk any
,.m._, 'A additional postulated IGSCC growth wouldshave,sto occur
(,
O into the weld overlay, which is not susceptible due to a ~m, ya the controlled ferrite content offitN" dup}exi(austenite Us ti and delta ferrite) structure.
Measurements of delta n~n n,
.nr y
ferrite for the first overlay layer'show a delta ferrite f) 4 content which is generally ' greater fthan 7.5 FN.
This is x-judged to be sufficiehtly higd tb prevent IGSCC propagation into the> weld overlay.
s y
KR v
5.4 Treatment of Asial Plaws
}
~,.; '
Axial IGSCC > crack length is limited on either end by the q
\\
original weld and the extent of sensitized material in
.'theLweld heat affected zone (HA2).
The tabulated allowable axial crack sizes in Reference 1 are truncated
/,
at a maximum depth of 75% of the pipe thickness and are therefore very conservative for axial IGSCC.
The ASME Code minimum thickness is based on maintaining a factor of safety of three against pipe failure.
Pipe XGP-09-106 5.5 Revision A
thickness in excess of the Code minimum thickness provides a reserve margin which can be used to tolerate short, through-wall (or less than through-wall) axial cracks.
The length of through-wall axial cracks which A
results in a factor of safety of 3.0 during normal Na
-s operation is calculated in Reference 2 as a-function of sw the applied stress.
f/~d c
%y g?
9.,
y.g Whenever the combination of an axiAf~ crack"ahd applied 79 F1 stress results in a factor of safety ofJ3*.0 or more, the
,:Q x yvyy axial crack, even if it is shrough-wall, maintains the M
u originally required Code safety factor.
~
fy
' N-M '
~
l, 5.5 Effect on Recirculation System
- 2
~O,
- ['
)
V' The effects of the radial shrinkage are limited to the region ' adjacent to' and underneath the overlay.
Based on Reference 1,1, hhe stresses due to the radial shrinkage a
3
!are less.than yield stress at distances greater than 4 y
^"' inches from the ends of the overlay.
Weld residual
- l '
?
n c
~,
stresses are steady state secondary stresses and are not 5
611mited by the ASME Code (Reference 3).
a.
The effect of the axial weld shrinkage on the repaired systems were evaluated with the NUTECH computer program PISTAR (Reference 8) using the piping model shown in XGP-09-106 5.6 Revision A
0.
9 Figure 5.5.
The measured shrinkages of all weld overlay repairs were imposed as boundary conditions on this model.
Since the ASME Code does not limit weld residual stress, all stress indices were set equal to 1.0.
1bd rm r s Since weld shrinkage-induced stresses are notelimited by 3_,y
^f the ASME Code, the Code acceptability of.-fthese welds is A/
not in question.
It is judged that stresessTof the
~
fl$, %
magnitude calculated will have negligible effect on the xs
- s integrity or IGSCC susceptibility Sf thes$ welds.
5
% i,l w
R' A
- u. \\
kl 5.6 Evaluation of Flaws in Unrepaired/ Welds
~ ~.,J '
es
,/
The prediction of crack growth for the flaws in
- n v'
unrepaired wel'ds requided the following inputs:
.j 1)
Stdady:!staterapplied stress.
,,.]
2). "WeldNresidual stress.
4r 7,_
- a
'3)$,j; Flaw characterization.
p Mf4),TCrack growth model.
m'
'y 5))
Crack growth law.
q.
q?
v' The approach was to use conservative input for applied stress, residual stress, crack growth model and crack growth law.
Thus, the result of the analysis is a very conservative prediction of crack depth versus time.
XGP-09-106 5.7 Revision A
The steady state moment due to operating pressure, dead weight and thermal expansion was obtained from Refer-ences 5, 6, and 7 In addition, the raoment due to the axial weld shrinkage of the overlays was added to the other steady state moments (Section 5.5).
The sdress in Y3 the unrepaired weld due to weld shrinkage isf[ Negligible.
- b_-
't
- p. m,
g 61 O
A conservative crack gr wth correlation forIweld 4
- 7,?..%
sensitized material was used and,is#diven below:
4 3
u 1
3 t;.
1: ~
9/
_ w Y,x _j/
!L = 1.697 x 10-8 2.q r "
d g
dT
- t,
- < {
t,
'7 [-
4.
a
.,v where:
//
3' t
fy h AV Q.s n ->
~,
Differential crack size (inch) da
=
cjf. p Diffefe,nti~al time (hour) dT
=
s%
,s
'Apdliedht'ressintensityfactor (ksi[)
K
=
y r,
y.' \\ g, ' ' %*
m
- n..e f)
.l
' I -1 (Tite crack growth model is a linear interpolation between m v. ;-
t f:< San dinsidi diameter (I.D. ) cracked cylinder and an edge-s x.
s
..y 3
c>
j/
Q),, e crack'ed plate.
The crack growth model assumes a 360*
"t 7 O.4 ' ' '
\\$3#c$ack.
The magnification factors for both an I.D.
\\y
, 3, cracked cylinder and an edge-cracked plate were obtained
,t from Reference 13.
I l
4 L
i XGP-09-106 5.8 l
Revision A I
L
The predicted crack growth for the unrepaired flaws was calculated with the NUTECH computer program NUTCRAK (Reference 13).
Allowable crack depth was obtained by taking 2/3 of the Reference 1 source equation values, as a
required by Reference 4.
This analysis-indit.ctes that i.
the flaw indications will not grow to their allowable
%" u,
i.h ge;x sizes.
fjh? lh
%l el?
[i,!
. s syy
,c m nu f h"34';. A.3. k ", _ ;4,
'g S
%,y
"\\
65 N3 v
$J}t tJ:?
t-3; ep y.O;-
9,,,,/
,< r
... e
~<-
lfj gs
. ~.
's 1 ft) g' x g y v aar
,7
/, c
,,z hp?
Y.
p M
/
N[:fl3
/n.
3, f w,.
/
9 Y
'(
%,[-
,3 Y,. 4
? ;-
g:fy%&)
y-. ;-...
r *.-
gx
,/ m.a vc r
's
'I
([i
<[;
- s y
%. w ' ?~.. ' ;r g
M 4,
,9.;,
N.$ _'
u) f:)
%Q. yl v
XGP-09-106 5.9 Revision A
e.
Table 5.1
'12" PIPE-TO-ELBOW CODE STRESS RESULTS i
i 4
a..q ghy t 1
wz;%:M
.9 y:.:
f.W
%; ? -..
f.,
- ?&
- .
%;y.
n..
4w'N,./.e. s4 es,
.2.,'k; s
+r z -. f,.
a nv i'
vf,' ja av
%..sy f.pn%,':S%f;+$
- Nh gg gJ 4:/
q yf Ea 9
- j..:J.
k
<y; A n;+g*:,,,iQ:;_, ;o*
g,p, m p -T.
1
.; -- Q.
.sf_
- <w>r l
k,g A
i f-
)
a.
,.o, n
'x t 0 %.ws& 'y LATER N;; y A
i f:-. ;'
f a
e,-
} -
)*
t y' Qm\\
' -)
i
.O.
of a%
c.f L6 Dp
'1 /
9
/Q 44
%,1
.J: s
%.
- Wll
.y+
f
+'q(n jf.o,,_
..=,s.
. n.,,%
g k9.f"'sQ'.:g
- g. g
- L
- h.%
s W
2
,.c t,
- A, xs Ay 4-48 I ' tlp
/ n;? %*
t r[v[.w pl'9
% ~u,s
?
,.t
- y l~ $%.
'V h i.' h.f ~#' < i%
w ::./
%G?
^
%[.,(
g*%
L,,;,~...,
A ;>a,%
.,,.A..4, f.4 p.
s..a (e
y,a
'1-g yy Nh..e
-a-
+,J Y-,x?g
,4 i.
'\\ J;h. fE.9 e
mxy
.h .
.v t.
t l
l l
t I
i-i.
. XGP-09-106 5.10 Revision A g
y y.,
..y,
..,g.,
m,.,.,..%..,r
,,,y#.,,,,.4__,.,,m,.%.,,m._,.,,my.,mp.rm,%
_cym,..,,,,,,.,,,.,,_
m fo s
i Table 5.2 4 -
1 28" PIPE-TO-ELBOW CODE STRESS RESULTS
+
r j
t 4
N IN se j
Ndi. ' ;,, ? \\-}
f
.%:s
%g-Ag,. :as 1
- '"x VA
. ppa,-IA
%P i
Jy
- y.h l
%&y. )
g+x~,,.l
,[,.e ~...
c.
[.
- g., -y' * ( " '.~%' 1--a
., ^ j$g
[
.,y
,, f:
y= ;.
va.
{.(
jj, v q' a
.5t pr,
.pygog@
g
,.e e.
vus y y< ~ c i
hr
+
i SSN 4
p ds; af, k.
ca k$
4 ri 4
- f. ;
i '
A; k
,#.. f g-
?
4LATER %:iy
/ tl-r L
,1
.1.
A-j,s %.~.
. R. 5.. <
A
- v.,
y, ;,.
,e,..,i c*
M r
m ei Kf. l]?
l cy,.
- ,.c. %,.,
f,; \\
mb
'f p. m;x:.
v<
x' t
g Q p;f 7 p
- p. m
,,un
.,s
}.; k
../O>
- - t '
.I
- b. _y.,s
- c-S t 'v.a, V
g~;
W. B's f~%j:p
- f 1.\\
.s; %
(.,
a ;q~,s.3 A..wp,.
[m. y-#
V I[' (%
w, id
%:y.y ?;
m.
af l
l i
i I
t l
XGP-09-106 5.11 l
Revision A
- <- 4-ve -r w,r.
y e+,.,,.
-..e,~,,.,..--.,w,w._,,,qm_._,,,,,_,
~ __ _.. _. _.. _ _.
h 4
t Table 5.3 28" ELBOW-TO-PUMP CODE STRESS RESULTS
- t f.
3
? >
I i.-
6,m p3 a
m 4 khdk e*n s
%u ff%v. z. Th vf 4
y
[-ij v ;w f.y w s sy c
,g?%
p.'j+*%Q %.xv%
j
\\
- p;s y!.A w
e v-
}
Q:4 q
Wk f] -
/i+p"+% % h./
?
Ng;p i
n:
s l
N Yk o
n.%
i-sg"[': 77l%'f:
i JTLATER sw 4':,-v l,
f-p
+ ~.vx..m,.. f_\\
j' ' A-' Y
{.
'W
- J:-,_
j.
j[A
$$$A.
yv g:9 sRf' up f: ; R,,.&.r s:
- t h
s.::Q p. %,... n MU5d
/
+
y : h, a
k 4.[
f V;( (
_A.
4
- q r.%n..2v.
i 4: S.
' V '.~. t y *, $/y *%;f, 2:T
g s
1 < '-
NL
_.4rAc),.
8 -
n +,5 O
_w,,
A.v;s w :\\
C:i,
- J.
f y-g S: -'
1'
}. 4:\\
jY~
f A
-...a
'* j' :.h_ ff
- s' *;[%y 4#
.. l ' lv' 9
%,.h
< c'rs
- ,L
h M~.
${ })*
VM ';'yh
'v (gf
%.f V
' % s.
M.i i --
%: ci f ';uf M
__,. +,
. ys s-y y
,.p-t t
i 6
a
)
4 4
XGP-09-106 5.12 i
Revision A l
nutagh
+,--.O..
~._-,,.,-...
J -
__.-_,.-,,m--.,.,_,,,w,-.~,m,-_,_..
y.
.,,,,...,-,r__,,..,,_ww.,.-
z:
s Table 5.4 1
l' 24" TEE-TO-VALVE CODE STRESS RESULTS 1
l 8
ga CQ-I!4 sw l%gg<:s~;.: g
-: sm g1
.e i
.rT?k Q]k
~
ffql$
Y.h fif V:
a:f i.
4s i %. ix/;-@
-'j% ':
t
\\.
.t &~-6 (n,./w, g
g,Y ':
.r cy s, ij l'.'Ea 'Y f
[i g
.-m}
V ')
?',.en n
eW% Dw..dN
- ,, se
. yL;p
&ly%a@v3 n!
4.a%:.,.;> M:
1
,;:r L
p.
mW s
l
+A.. P' LATER g,c.. w.
- c. A
- 4. s e, a'Yg,A i.
$? <g
w --,%.llp
"\\.-
f sy)i f x v
c-fy
. rer 4.:m,-
i.~
e %~,c,s,.,%, -
s t
n i.
mp tg 4.-
%24
,, q^l 4 \\ps W;" aW.,$f i
n s
J.',...)
- _).,'
- my y
....J-%;
s
+ -
- r.,a n.
t -
1,/
gi: t.,
y::@g c'
e
- h J.D
....m
'd-l* n!
' Mff g,. - m,;.s..
% x,
~
v :
Q :,53 ' s -g.a. ;', -
?
'a..:)l%
F6/
\\9 g(.?.
s-s,T.-: 9w.
4.'- (.
.s A
-c t
L
/
- s#
a t g
L XGP-09-106 5.13 Revision A
11 t
Tab'le S.5 4
22" PIPE-TO-END CAP CODE STRESS RESULTS
+
i e
1 4
4 2
gh
.m
'ij ; 4 4
- 44. %- &,;9 f%
2
%s v 4 m-rPu
%.... $.s r
N h; -
ma 4
/e Q%
gh.,7
+
%s.M..y7 i,
e-y
- g.,%%gsw A:p.
i
.f T s 1
y j-lk
\\c {'s I?{
-Q tv v,.
V'h jz f t?
A WO> Y
~wi w
(g@
A es h *$L h-?
5e., yi.,,)
7:X,' e
.4-4%
y,,9 :,I.
b.
A) LATER'WW a
l
.,di#
+f'j Ni?*b.,
.u v
%., m j~
ff
-w.9 A'
s.y
~
Ih s' v.f
'-; W
\\
fyw
%lhj%
M -
l k# ?[,$[e*N..,-. db g
y 3~
b v 5!
t.
- d. \\.) fry
~ ~.
.((
v',d
., r l, $$ Wh '-?
t'g-) L,' }' %'. . A _d t> %:. 4.,Jy (Sy.y%g j-,,: -l;y ^'g .g /l/.~ '?% y ? h %'E% 4(.4.V.:-? yn. k "%47 9.14 ". <Q, %9.( ? 7 J- 'tiig ,;,'. 3 k e,? w-3,;y'f.. -[ '%'I ' _t "al f.'ru XGP-09-106 5.14 i , Revision A I l __.me,_ _,r,y.,w.,y.,,,_,,.w--y%,.-,-m+.-.p-,,.-m.m.,,,,--c +t-,,. . -.wv.--..---..1-, w-w.--- .-e.--
Table 5.6 24" PIPE-TO-PIPE CODE STRESS RESULTS s. O, \\ r (, '. h. m. V4 4:.ms.' ar Q~.g...#n, e.m,.pg n.... u, 34 %'-g9 py,f*;;% - r.;s} . %<t v e s ' '(g.fhikV
- 1. y f f'QD
% ~ s:% gg# % Qr Nyh v uw u M ga m - -a plh.Y %.-.,nyf 1:9 p,.x p %, g .q % k;SZ Mp _, r /$ .: g r:9 4+.
- q, o-
- IU%w#A[/
t
- 9LATER A.1
.e ,4 j
- .a ef4,',h -qs
-e AV %JA AF %?' % -4 L.? ,g; - 9p J Mh
- .,% f,
- e^.V T -
%;; :u,4 f#.. A g. Y
- ig h
"C 'h' -..f.,,;.. 3 g% v c 3 -+e ( u, wg,C.$$ NI.h + ? en w;.7 t,. p. t -. t. <wp-y8': : - - + s
- 1 e i.>
e ?Q M,9 A=.. s r ~ %,h f 't,;- 4}^% 29 d i f/9e g-y -' %: a.a. b '( e' t '.J
- V
=_/ % :.. a D[' c',
- .".4,
..9} '( .J 9'.,* 'a r, ,' :.j 4 ? if I M 0 Y. .,,M V 6 3 p 4 XGP-09-106 5.15 Revision A
Table 5.7 THERMAL GRADIENT RESULTS i' w. $ e.y ~~ J ~, +-- 7 N-t I a A \\ l
- n e
,' LATER '3: ..s 5 T = -9 .r i a e-XGP-09-106 5.16 Revision A nutg,gh
Table-5.8' PLANT E. I. HATCH UNIT 1 WELD OVERLAY INDUCED SHRINKAGE STRESSES
- a d'i us k&
v4 /p%M m' : 3,, we.,,s , s n., Weld ID Stress ( ps'i ) 4s T9v .a. %23 v :'% L! N-s f-i v
- h
% c.\\, g',7?N, i. [,t ;;.w ' s ' a,;
- 9., 9 c
f J. #" ' <\\ ) '.i ;.
- /
' [< j ^ 5 m i. m* c% C' e n g' %;y 9 l:g.p ,n'. e: .g ~;'f m 1:" q C .4 4 Nk lb, E w a, m 4 ; '*, .. t y;* [ IATER %' *f,ts;w1*;P .a y f.j# ,r,n
- ;~
3 5: b "i. /. x f.'4 'f' y .. I*D. j.r;f ' n'Vy f4/ A y* 1 .1-v.sr f a 6-5,_.- q Q?g c i , 8;;. g s.. ' k ~ > ;, 5. I "5
- s. zf %2
- t, 4
- "3.{s j-gg
<e ^ V'. e
- l-gr
- n. ; %
3 Nn ,
- e y, h.s, s
ur.s 'g 2 .,v' Y. '. s. y, a% ',
- l_ '->:_y
\\ g. ;' j -s- .+ XGP-09-106 5.17 Revision A
ll' ii li li 9 a l e.] --\\ + l '~'" b, i v I ) p h Et ) s3 \\\\\\\\ l l ( )
- ?
r <E umass.am Figure 5.1 Finite Element Mcdel of 12" Pipe-to-Sweepolet Weld XGP-09-106 5.18 Revision A nutggh
INSULATION +A-WE.D JA sem sm
- 5 esilit ses'
$k,,,,,, A\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\M. s w, 5 e7 m,. s ..n s,, s. j' a _.4, ,,-w 3. I ~., , }, ,, +, t (_ 'j t & A - ~, ~ 'r. d :
- r -
4,;.+
- i
- 3...;
gs ~._._c y .0 'k 'N;., s
- s. -
g. >h .t.. h = 738 BTU /hr-ft - F fi./, k = 10 STU/hr-ft. F g7 i b 1 y: v-e a.
- i, g
s s ,,4,- I SICTION A-A n. ) ecrim.as.zz Figure 5.2 Weld Overlay Thermal Model l i i I XGP-09-106 5.19 Revision A nutggh
< :a 44 EMERGENCY -- 3 . / c. t ~~ ~ 5 fif G 7 s ,e , ; r.> y e', SMALL t TEMPERATURE - En' CHANGE m ~ STARTUP SHUTDOWN - 1 NORMAL - 1 ] ] OPERATION I I k j RESIDUAL - I' 38. 25. ~ 5 CYCLES CYCLES YEARS =.. = ENTIRE -9 ~ 'l' SEGUENCE ,N." CYCLE REFEATS l s. ?> TIME -j pcy., na.w I Figure 5.3 Thermal Transients l i XGP-09-106 5.20 Revision A nutaq, h
30 l t 20 - 4 ll \\; t i /?", ' j m.,. s.% p' p, 10 - /
- ?
N' e q. . a 1. +v'A v-E I I I l' '1 x 0 y x /t *, ^.e9 -31.0 0.2 0.4 0.8 QA g N w.f.W g ['] 4, .a i, '. jJ g b- .1 N 0.25" OVERLAY Q = 23.52 KJ /IN - 9., ft = 75% 's t = ORIGINAL PfPE WALL 30 - v. ~,
- s. 3 t
,9 ' r . ;p ,-i. o r y 4p -e 'g3 mass.sta v Figure 5.4 v l Typical Axial Weld Residual Stress Distribution l in 12" Pipe After Overlay l l XGP-09-106 S.21 Revision A nutggb
t n ,c o 193 i i.. s . A..
- v. z.
s 'i <<N,.- c %- v,. i.. .a %i-i.. a% 1" n , e,, s.- %,s a, > %. s i. Ji 1 si. uv (g g ra in %e,. l3 ^ o ss. -m ,a m. . ot ,g,*4 in s /i .. h m';#*E;v%3, i p. pgg s ,n c l y gu i...f, y o. o
- n. s n.
o na ..!' ;,.. j ' k,' v.c d n. in. we b.js;p&.e ni o . m (, r/ i."' { n.g ~ n. ..i ,s% o -v .j. R..g-* " <. c : 1p -li 4 s9
- e c
r xy 8.. s,. n. .v , m P.. q.c -y 5,. M3 y u.- M,5k s - [\\...,. v < 9 'A.vwx, V s... ws u Jh. ,8Aj y.] in v,-,, ~ w;.. t l ,- 7 g ,s. a,, [ S y' l
- . A, /p N 3 9 a
yy / j 3. x 4r t ms,' Vg j, g, M u j. ,s. ,; _ ) s, p - ;,; j. 97 y l Figure 5.5 i l' E. I. Hatch Unit 1 Recirculation System Piping Model i XGP-09-106 5.22 Revision A-l . _ _ _ _,..... - _.. _,.,.. -. -. _ _,.. - ~, --.
l 6.0 LEAK-BEFORE-BREAK ASSESSMENT For welds with undetected flaws, or containing IGSCC indications which are judged to be small enough to not require a repair, thefollowingconsideraNions 1\\ -a form the basis for continued plant operat= ion 9for e, another fuel cycle. ,e / l 'l-l ,3 \\ ;y b,~' ^ [ 'k) 6.1 Net Section Collapse j- .- 4 Jc, 4 l y rn The effect of IGSCC on the structural integrity of s-i (; j piping is evaluated through thefuse of a simple , w+n ) n,.- " strength of materials" approach to assess the load-i carrying capacityfof a piping section after the 'N 's cracked portion has1 Deen removed. Studies have shown 1 l t. (ReferencesJ12 and 14) that this approach gives a J6s ^ ^. conservative,\\1ower-bound estimate of the loads which y, a; W would cause unstable fracture of the cracked sectioni Typical results of such an analysis are l 7 e ("" ' shown in Figure 6.1 (Reference 12). This figure deIinesthelocusoflimitingcrackdepthsand lengths for circumferential cracks which are predicted by the net section collapse method to cause l failure. Curves are presented for both typical L piping system stresses and stress levels equal to ASME Code limits. Note that a very large percentage l l l l XGP-09-106 6.1 i Revision A l nutagh L --,wg v-, w w .p y_.s y ,,,,,4m. g-w.7
~ .of pipe wall can be cracked before reaching these limits (40% to 60% of circumference for through-wall-cracks, and 65% to 85% of wall thickness for 360' part-through cracks). b Also shown'in Figure 6.1 is a sampling ofEcracks 3-which have been detected in service, either through E' fi UT examination or leakage. In each cade:[there has ,r m a been a significant margin between thd?sfadaof crack Fi Vii ~ observed and that predicted toscause. failure under
- m. N v..
,e f 2W # -.- *-'a service loading conditions. Also','~as discussed e
- +
t.2 f below, there is still considerable margin between g_~ey these net sectionfhollapse' limits and the actual ,r, cracks which wouldtcause instability. N Nw / O 11 'i a% -V^y (; 6.2 Tearing Modulus Analysis .c.p. s% w %, $/ _ Elastic-p.~lastic fracture mechanics analyses are a j /(3.phNsentedinReference14whichgiveamoreaccurate N v b. , f[ ' a ' representation of the crack tolerance capacity' of <T l
- ) sti;>inless steel piping than the net section collapse O'
l ,t [f approach described above. Figures 6.2 and 6.3 l graphically depict the results of such an anaAysis 1 [ (Reference 14). Through-wall circumferential defects of arc length equal to 60' through 300* were assumed at various cross sections of a typical BWR 'XGP-09-106 6.2 Revision A
Recirculation: System. Loads were applied to these sections of sufficient magnitude to produce net section limit load, and the resulting values of tearing modulus were compared to that required to 'cause unstable fracture (Figure 6.2). Note that in u
- ~
L :t all cases there is substantial margin, indi,cating '~%i3 ,m., that the net section collapse limits,ofdth'e pre'vi'ous (f M Fi'gure 6.3 ~ section are not really failure limits.Q*$3x e?5}38 summarizes the results of all such analyses performed through-wall cracks in\\{ terms bf margin on b a l for 60* fly Th4p/ tearing modulus for stability. 4The" margin in all in 1 cases is substantial. VMs s1r K-WY nd a l,l rc 1 6.3 Imak Versus Brhak$ Flaw Configuration "Oh W V r 'C, // of perhapsim~re significance to the leak-before-break o A, 'A argument [is,the' flaw configuration depicted in Figure y g @ figuration addresses the concerns This con j6.4. jf%;3 n;'% <" raised by the occurrence of part-through flaws ,( "; N. Q7e 2470>, -growing circumferentially before breaking throagh the
- r y m
g < >. s vA 4 %_j. Nj outside surface to cause leakage. Figure 6.4 y ,) ~N q}' presents typical size limitations on such flaws based wr l on the conservative net section collapse method of Section 6.1. Note that very large crack sizes are predicted. Also shown on this figure are typical detectability limits for short through-wall flaws 'XGP-09-106 6.3 Revision A 4
(which are amenable to leak detection) and long part-through flaws (which are amenable to detection by 1 UT). The margins between the detectability limits, and the conservative, net section collapse failure A limits are substantial. It is noteworthy that the ^d d likelihood of flaws developing which are %, sQ ,se*s gf g characterized by the vertical axis shownX,i'n Figure a. -s %-;: ?, , ~ -' 6.4 (constant depth 360* circumferentiA1:_3 cracks) is CQ% V soremoteastobeconsidered. impossible.[dMaterial !?. II and stress asymmetries always' tend to; propagate one , --~g y s...r.; - u. portion of the crack faster than th,e bulk of the crack front, which will e'ven_tu' ally result in " leak-t xi before-break." /This observation is borne out by extensive fiel'd'exp rience with BWR IGSCC. 3 s '.? e ? 7 6.4 AxialCrack'sk' ~ a v s .d> l_ --Many ofithe IGSCC occurrences at Plant E. I. Hatch ~,I.N v-Unit 1 were short, axial cracks. These can grow (7 ,j' /' I sthrough the we.ll but remain short in the axial .c' s/ <?N . direction. This behavior is consistent with 5 n., 1 ' [ d/~ expectations for axial IGSCC since the presence of a sensitized weld heat-affected zone is necessary, and this heat-affected zone is generally limited to approximately 0.25 inch on either side of the weld. Since the major loadings in the net section collapse XGP-09-106 6.4 Revision A i nutggh ._.2
x analysis are bending moments on the cross section due to seismic loadings, and since these loads do not ex'ist in the circumferential direction, the above leak-before-break arguments are even more' persuasive s for axially oriented cracks. There is no known W mechanism for axial cracks to lengthen befsre2 growing w% ~ through-wall and leaking, and the potential rupture C )9 loading on axial cracks is less than 'thatton <3 z, circumferential cracks. fr' N IA. N3 v n g7;, N.. % 4:l b +. = 6.5 Multiple Cracks / g_ m ,u:. w Analyses performed)for EPRINNeference 15) indicate that the occurdence of multiple cracks in a weld,-or y qt crackinginjmultipleyeelds in a single piping line t.a /o does not inhalidate the leak-before-break arguments ,. - x e s. discu/ssed:,above'. 1 l .. 3 - f,9%# e , _j M l 6.6 Ndddest'rt$ctive Examination l r. ~^ V" y _;, s t',-'1 C i i T primary means of nondestructive examination for s - o i m IGSCC in BWR piping is ultrasonics. This method has M i i l been the subject of considerable research and development in recent years, and significant l improvements in its ability to detect IGSCC have been achieved. Figure 6.4 illustrates a significant 1 XGP-09-106 6.5 Revision A nutagh
aspect of UT detection capability with respect to leak-before-break. The types of cracking most likely to go undetected by UT are relatively short circumferential or axial cracks which are most 4 amenable to detection by leakage monitoring. p(
- \\
Conversely, as part-through cracks lengthen'/ sand thus m . 3. f,.. become more of a concern with respecteto(leak-b'efore-g- 4 ,a break, they become more readily detectable by UT. \\ V ,r f g m c 6.7 Leakage Detection t% t j;< 9,'q[~ g_, ],/ = p, !!s4 Typically, leakage detection for BWR reactor coolant .~., system piping injihrough s'umi' level and drywell activity monit$ ing. These systems have gf sensitiviti/d on tiejorder of 1.0 gallon per minute _%n a (GPM). Plant / technical specification and adminisErat,ive) limits typically require ,t..-N , investigation / corrective action at 5.0 GPM j y, LL .'" unidentified leakage, or when there is a 2.0 GPM G )
- increase in unidentified leakage in a 24 hour period.
, '. / .A c) v ~ ^ Table 6.1 provides a tabulation of typical flaw sizes which cause 5.0 GPM leakage in various size piping assuming a membrane stress of S,/2 (Reference 10). i XGP-09-106 6.6 Revision A
Also shown in this table are the critical crack lengths for through-wall cracks based on the net section collapse method of analysis discussed above. For conservatism, the leakage values are based on pressure stress only, while the criti" cal .+ ( crack lengths are based on the sum of all4 combined l
- '< A e
loads, including seismic. Consideringiother normal
- )
operating loads in the leakage analysiSswould result p 8, ~ in higher rates of leakage for;a give.n cra'ck size. ,3 y4 4 NotethatthereisconsiderableimargiN)betweenthe ( ..gp %;%,,7y j crack length which produces 5.0 GPMaleakage and the V4 critical crack length, andthatjthismarginincreases .~ with increasing pihe size. l e; <Jf t.' ~ /p x H istorical-Expe rie n;ce^> 6.8 l N :s e s - _' e L The $bodesthhoNies regarding crack detectability have , bee sOppo' reed by experience (Reference 15). Indeed,. ,n;'s,f)the "large number of IGSCC incidents to date in BWR o ~ ipipind,nonehavecomeclosetoviolatingthe ~ h to <f J1 structural integrity of the piping. L 9 8 l XGP-09-106 6.7 Revision A i l nutggb I = . - ~,, .,-r- .,,y-, . - ~. v
i l l Table 6.1 EFFECT OF PIPE SIZE ON THE RATIO OF THE CRACK LENGTH FOR 5 GPM LEAK RATE AND THE CRITICAL CRACK LENGTH (ASSUMED STRESS o = S /2) 4 m 7
- ['..,. l w..,
W' %, 5 4 ., 2>. sj i f .g s- ., -i,, :- g t r s s NOMINAL CRACK LENGTH FOR CRITICAL CRACKi gjg l PIPE SIZE 5 GPM LEAK (in.) LM A liS) C c. 4" SCH 80 4.50 4 6f54 0.688 10" SCH 80 4.86 1~
- 15.95 0.305
~. 24" SCH 80
- 4. 9 72:
35.79 0.139 ecrtas.as.co m.- N t.' z l l ,i ' ~ ,,9 t,.
- e l.
XGP-09-106 6.8 Revision A nutggb
I i eM t I i g 4 + 1.0 " N -O ,O \\ c 4 N P = 6 kai, Pb=0 0.8 4 g m 4 s. % ~% 5 es i . n*'*y== ~ P,J=Sb,P Pb = 1.5 S 0.6 m m } 8 @ Reid Data - Part. j .4 0 i Througt: Raws j
- g M'
C Reid Data - Laaks C S, = 16.0 kai of = 48.0 kai e s 0.2 - 0 Values at 550 F ^ 0 '"0 0.2 0.4 0.6 0.8 1.0 Frasnan of Circumference, e/r Pcet.as.co.2s Figure 6.1 Typical Result of Net Section Collapse Analysis of Cracked Stainless Steel Pipe XGP-09-106 6.9 Revision A l nutggh t i
8 55 -- ] 2108 gp e 4 " f" L JVM N 1 g.ag5 B' ~ ..y- 'A 300 . ', J. N .c, A, ^ ggg. 'q g 9 n 200 - ^ (Unsmhief a r x 1 m. n L { 180 - (Sandel S 3 = 12Do 8 'I
- E 20 = 240. -
s 100-- 4 3=1W8 ' q-Jfc g. j ,s 29 = 300s ll 0, -5ILO, 12.5 1 3.0 82J 100.0 137J 175.0 212.5 250.0 T Pert.as.as.to ( Figure 6.2 Stability Analysis for BWR Recirculation System l (Stainless Steel) l XGP-09-106 6.10 Revision A
l-i 20 i 60 TilROUGII CRACK PLASTIC LIGAMENT 4: .c 15 - / RANGE OF MATERI AL- O TEARING MODULUS RANGE OF APPLIED,,',' j rTEARING MODULUS un 10 - 3 2 -[' / MARGIN 0 U 5O 100 150 200 250 '/ TEARING MODULUS 4 Pcetsam ti Figure 6.3 Summary of Leak-Before-Break Assessment ._of 3WR Recirculation System XGP-09-106 6.11 Revision A nutggb
i i i ? 2a t d FIPE CROSS SECTION
- m,.
? * 'k ;,. { O. 7. - :g+ p _,, r 0.6 ~ ~ ~ ~ - ~ ~ ~ -- i:: g lw ', s O l '.' r -d 0.s. l . ?, 'i;I
- 0. 4 -
,g-, 'g u N 0.3 - l i 7 v i 0.2 -- c' iSi I 1 O1 ~ ~~ LEAK MONITOR I 10.0 10.0. 0.1 0.2 0.3 0.4 0.5 0.' 6 0.7 l =/w PCPt.ELos 12 l Figure 6.4 i Typical Pipe Crack Failure Locus for Combined Through-Wall Plus 360* Part-Through Crack i l I XGP-09-106 6.12 Revision A nutggh L
W 7.0
SUMMARY
AND CONCLUSIONS Evaluation of the repairs to the Recirculation System r; ported herein shows that the resulting stress levels are acceptable for all design conditiods. The ,:f stress levels have been assessed from the(standpoint of load capacity of the components, fadigue, add 3the
- 0
$<Mp. /. - . w resistance to crack growth. e %f k ~ Ngg
- S Acceptance criteria for the an$ lyses have been 2,
,,,. ~ 8 ~ established in Section 3?O of this" report which II di demonstrate that: ,Q nm ~- 33 f 1. ThereKi's no loss of design safety margin x;., ' N4 Joderthat(providedbythecurrentCodefor (Cla'ss 1 piping and pressure vessels (ASME js .. -.. s.SecSion III). 9, /;;. sg, (-/ 1 fj.i 7 '2.+. V During the design evaluation period of 1 m. 7 cycle for each repair, the observed cracks ~, / a will not grow to the point where the above
- i safety margins would be reduced.
s ;,. Analyses have been performed and results are presented which demonstrate that the repaired welds satisfy these criteria by a large margin. Analyses have also been performed which demonstrate that the XGP-09-106 7.1 Revision A
e .o unrepaired welds satisfy these-criteria. Furthermore, it is concluded that IGSCC experienced in the Reactor Recirculation System at Plant E. I. Hatch Unit 1 does not increase the probability of a a design basis pipe rupture at the plant. This!', conclusion expressly considers the nature [df e m s cracking which has been identified at(Pl' ant E. I.S (3
- s Hatch Unit 1, and the likelihood thatiothei similar zmg 'y>x v
cracking,may have gone undetected'.;The ' conclusion is
- 9 based primarily on the extremelp@shighYinherent f.y
-T/ toughnessandductilityjfthest'ain'lesssteelpiping ia 'i i material.t Cracks in such jpipir}g grow through-wall (.: gm' n:,. w and leak before affecting it,, structural load s ... n. carrying capacity.7.: + Af %Q'V p; i, +, NQ.y ' ,N "df s,, %,D, g/ t 3 - l %) m y,. e,
- f-
[' l., %;;% /j.m ig ' . '}^ * %., -7 \\._
- ji
,e:
- q..
- y. '
iC XGP-09-106 7.2 Revision A 11
y h. h +
8.0 REFERENCES
1. ASME Boiler and Pressure Vessel Code Section XI, 1983 Edition, Winter 1983 Addendum. 2. NUTECH Report COM-76-001, " Weld Overlay Design Criteria for Axial Cracks,"3 Revision 0, March 1984, NUTECH File. COM076.0208. 6' 4$:M 3. ASME Boiler and Pressure Vessel CodeJA Section III, I, Subsection - NBfandg Appendix I) 1980 Edition with Addenda through Summer 1980.- 'Q f g% 9lh 4. NRC Generic Letter,84all',i,idated April 19, 1984, NUTECH File No'. COM096.0010.0001. '%gma~h) Y.: d? 5. Later dy? 9 v 6. GELetterNok-[G-GPC44-500, " Hatch 1. Seismic Evaluation,for Replacement Recirculation'Valvef0perators", NUTECH File XGP009.0025;"' As 7. NUTECH? Calculation Package No. GPC-04-303, " Weld Ovierlay Thermal Analysis, PISTAR x$Pi i Analysis", NUTECH File XGP009.0025.
- d$ p ng U*
8. '";NUTECH Computer Program PISTAR, Version c3 2.0, Users Manual, Volume 1, TR-76-002, k;.; Q @,. Revision 4, File No. 08.003.0300. % a; s 9.., N?'? ANSYS Computer Program, Swanson Analysis i "i j E 6 Systems, Revision 4, NUTECH File No.- t[e Ajs V 08.061. "- yy y 4%. N 10. Schneider, P.J., " Temperature Response m .g-@ " ( :;s Q Charts," John Wiley and-Sons, 1963. h; _ 'O 11. NUTECH Report NSP-81-105, Revision'2, Mid! / " Design Report for Recirculation Safe End and Elbow Repairs, Monticello Nuclear Generating Plant," December 1982,' File No. 30.1281.0105. 12. EPRI-NP-2472, "The Growth and Stability of Str.acs 'Corcosion Cracks in Large-Diameter BWR Piping," July 1982. XGP-09-106 8.1 Revision ~A
y 13. NUTECH Computer Program NUTCRAK, Version 2.0, Revision 2, December 1983, File No. 08.039.0005. 14. EPRI-NP-2261,'" Application of Tearing Modulus Stability Concepts to Nuclear Piping," February 1982. A 15. Presentation by EPRI and BWR Owne[s\\ Group to U. S. Nuclear Regulatory Connais'sion, " Status of BWR IGSCC DevelopmenteProgram," October.15, 1982. fe '~% f ,y m ;q g. y a r-q,. j:1 %.q:[a f i* f? t- .:U; n :ca mi) aQ.}'., 4/ '6? 'r a gn %l%,.,g,L'd p.\\ gea q %;i? m .g i t; V.' %
- y
-:3 my.<-: y _ ;g R.bd' y: {. A v 'd $\\ p,w .s..J A v %{. '[%, j;, f .JI Jo R-.~ ) g'i. - J L y,;,A # . j.; .s u'; r_ iyQg. y;a + n. w [ f;L af g -.f ,w. .3 y1 ~p 1-.-:e. .e,,;, i 1V W5 ~
- t
'K ' n, p 'y .y v :. y [. ,. x ~ n i e.
- 1
- p. :)-
,V v g_ ) ^*g-l \\ ? 'f..?..e._ ~q ;l 1<E3 -<> t y.; ../ ' # s <<../ 'Q t I r l XGP-09-106 8.2 1 Revision A .., _ _ _ _ _,.. _. - _ _ _ _.}}