ML18026B151
| ML18026B151 | |
| Person / Time | |
|---|---|
| Site: | Browns Ferry  |
| Issue date: | 08/09/1984 |
| From: | Domer J TENNESSEE VALLEY AUTHORITY |
| To: | Harold Denton Office of Nuclear Reactor Regulation |
| References | |
| NUDOCS 8408150213 | |
| Download: ML18026B151 (130) | |
Text
ATTACHMENT 2 Report No:
SIR-84-022 Revision 0 SI Project No:
TVA-04 July 20, 1984 ANALYSIS OF BROWNS FERRY UNIT 3 JET PUMP INSTRUMENTATION NOZZLE SAFE-END REPAIR Prepared by Structural Integrity Associates, Inc.
Prepared for Tennessee Valley Authority
/'
prepared by:
yM ~/7 R.
tones> er Date:
Prepared by:
K nate: 7 Prepared by:
er er I
Date:
1 /M Reviewed by:
V. F.
. F.
op an Date:
7 3~
@+
Aoproved by:
T. L.=Gerber Project Manager Date
+ M cF QH~S ISPB Ib STRUCTURAL INTEGRITYaaeaeaaree
LIST OF FIGURES FIGURE TITLE PAGE 1-'2 1-3 Jet Pump Instrumentation Nozzle Configuration for Browns Ferry Unit 3 (El. View) 1-3 Jet Pump Instrumentation Nozzle Configuration for Browns Ferry Unit 3 (Perspective) 1 4 Cross Section of Safe-End Showing the Worst Observed Crack Indication
~
1-5 1-4 Final Overlay Design..................
1-6 2-1.
Circumferential Flaw Size Limits Versus Axial Stress 2-10 3-1 3-2 3-3 3
4 3-5 3-6 3-7 3-8 3-9 3-10 3-11 3-12 Finite Element Grid for the Short Overlay Analysis 3-5 Residual Hoop Stress Isobars in the Thin Portion of the Safe-End After Completion of the Short Overlay 3-6 Residual Hoop Stress Isobars in the Thick Portion of the Safe-End After Completion of the Short Overlay
. '-7 Residual Axial Stress Isobars in the Thin Portion of the Safe-End After Completion of the Short Overlay 3-8 Residual Axial Stress Isobars in the Thick Portion of the Safe-End After Completion of the Short Overlay 3-9 Finite Element Grid for the Full Length Overlay Analysis...................
3-11 Nozzle to Safe-End Weld Residual Hoop Stress Before Application of the Overlay'........
~
~
~
~
~
3-13 Nozzle to Safe-End Weld Residual Axial Stress Before Application of the Ovelray Nozzle to Safe-End Weld Residual Hoop Stress After Application of the Overlay...............
3-15 Nozzle to Safe-End Weld Residual Axial Stress After Application of the Over lay.......
~
~
~
~
~
~
~
3-16 Residual Hoop Stress Isobars in the Thin Portion of the Safe-End After Completion of the Full Length Overlay (with tube support plate)
~
~
~
~
~
3-18 Residual Axial Stress Isobars in the Thin Portion of the Safe-End after Completion of the Full Length Overlay (with tube support plate)
~
~
~
~
3-19 STRUCTURAL INTEGRITY~res
0
LIST OF TABLES TABLE TITLE PAGE Applied Loading to Jet Pump Instrument Nozzle Safe End NBA........ "..............
2-11 3
Loading to Jet Pump Instrument Nozzle Safe End N4B Calculation of Applied Stresses at the Thinnest Section of Safe-End at Nozzle NBA.
2-11
. Calculation of Applied Stresses at the Thinnest Section of Safe-End at Nozzle NSB......-......
2-13 STRUCTURAL INTEG RITYassocu rss
1.0 INTRODUCTION
Cracking suspected to be IGSCC, has been observed in both of the Browns Ferry Unit 3 jet pump instrumentation nozzle safe-ends.
The nozzle/safe-end geometry is illustrated in Figures 1.1 and 1.2, and consists of an approximately 5 inch long, tapered, stainless steel safe-end attaching the low alloy steel reactor vessel nozzle to a'pair of eccentric'educers.
The qafe-end and nozzle are nominally 4 inch in diameter.
The reducers join the safe-end to a 12 inch nominal diameter blind-flanged pipe through which the instrumentation tubes are routed.
The UT examinations indicate axially oriented cracks located intermittently.
around the circumference of the safe-end.
The worst crack indication's illustrated in Figure 1.3.
The crack is essentially through-wall over the thinnest portion of the safe-end and extends about 3.75 inch from the safe-end/reducer weld toward the nozzle weld.
Based on a field replica-tion/metallography method, the entire safe-end is believed to be sensitized.
- 1. 1 Repair Design Evolution The initial assessment of the extent of cracking in the safe-ends suggested that cracking was limited to the thin portion of the safe-end.
Since applying the overlay to the entire length of the safe-end would presumably make future UT examination more difficult and would require welding near the low alloy nozzle, an overlay design which stopped the overlay at the thick end of the taper was initiallyconsidered.
When later examination of the UT data resulted in revised and larger crack length estimates (with the largest crack extending well into the thick portion of the safe-end as illustrated in Figure 1.3),
the short overlay design was abandoned and a design which extended the overlay over the entire safe-end was adopted.
The th'.ckness of the overlay has been revised two times.
The original design thickness of 0. 16 inch was based on the understanding that the cracks were limited to the thinnest portion of the safe-end.
After the crack lengths were revised
- upward, the design calculations were repeated using an upper bound crack length equal to the length of the safe-end (5.5 inch).
This STRUCTURAL INTEGRITYassocuvcs
I significant increase in the assumed length of the crack resulted in a 0.02 inch increase in the design thickness to 0.18 inch.
This overlay thickness was to be achieved with three automatic GTAW heat sink layers.
e The second revision to the overlay thickness was to increase it to the final design thickness of 0.25 inch.. This increase in the thickness required four weld layers.
This increase in the design thickness was adopted in response to NRC concerns relative to multiple axial flaws and some new (unreviewed) information from BNI concerning the appropriate flow stress to use when estimating the structural margin for pipes with axial flaws.
Increasing the weld overlay thickness was judged to be the most expeditious way to address.
these concerns.
1.2 Final Repair Design The final design configuration is shown in Figure 1.4.
The effective weld overlay thickness is 0.25 inch and it extends from the transition region of the eccentric reducer to the middle of the original nozzle to safe-end weld.
As indicated in the Figure 1.4, the repair includes one layer of SMAW over the cracked region followed by one layer of GTAW over the entire safe-end.
Both of these layers are applied with no water in the safe-end.
Additional overlay was applied to produce a uniform diameter in the region of the reducer/safe-end weld.
Four layers of GTAW were applied to produce the 0.25 inch design thickness over, the entire length of the safe-end and a portion of the eccentric reducer.
These final GTAW layers were applied with water inside the safe-end.
1-2 STRUCTURAL INTEGRITY*ssoce~cs
Bio SNmO dg z V) fT) X 9 C p 0 Q c r
FIGURE 1-1 Jet Pump Instrumentation Nozzle Configuration for Browns ferry Unit 3 (El. View)
0
'L
)'h y/'Reactor Pressure Vessel I
'4..
r
. V,i' FIGURE 1-2 Jet Pump Instrumentation Nozzle Configuration for Browns Ferry Unit 3 (Perspective)
~I
~I
/4'sec FIGURE 1-3 Cross Section of Safe-End Showing the Worst Observed Crack Indication
zS Q C 2 -l.
0 FIGURE 1-4 Final Weld Overlay Design Ql Stick (SWAW),
1 Pass, Dry Pipe Q2 Auto (GTAW, 1 Pass; Dry Pipe Q3 Auto (GTAW), 2 or 3 Passes, Wet Pip O
4 Auto (GTAW), 4 Passes, Wet Pipe 0.25, Inch Min. Thickness
2.0 OVERLAY DESIGN The weld overlay thickness was sized to meet the requirements of ASNE,Section XI, IWB-3640.
Both through-thickness, circumferential cracks and through-thickness; full safe-end length (5.5 inch),
axial cracks are considered.
Using conservative assumptions for safe-end dimensions and
- loads, an effective overlay thickness of 0.18 inch was found to be adequate.
Increasing the overlay thickness from 0.18 to 0.25 inch further increases the structural margin.
The design life of the overlay is one fuel cycle, however, it is likely that further analysis, including consideration of subcritical crack growth due to stress corrosion and fatigue, would justify extending the life.
- 2. 1 Safe-End Loads and Stresses Loads for the analysis were obtained from recent design calculations provided by TVA.
A design pressure of 1250 psi and axial stresses computed from forces and moments as given in Tables 2. 1 and 2.2 (for nozzles NBA and N8B, respectively) were employed to size the repair.
Tables 2.3 and 2.4 sumarize the calculations of the applied axial stresses at the thinnest section of the safe-end.
2.2 Overlay Sizing The overlay thickness was determined by calculating the thickness required for a worst case axial flaw and for a worst case circumferential flaw and then taking the larger thickness of the two.
In the following two subsections, it is shown that the required thickness for the assumed axial flaw is 0. 18 inch and the required thickness for the assumed circumferential flaw is 0. 16 inches.
Therefore, the axial flaw is the one which determines the required overlay thickness of 0.18 inches.
In accordance with NRC SECY-83-267C, the effective overlay thickness is defined as the thickness of overlay deposited after the first weld layer that clears dye penetrant inspection.
- Also, the minimum effective overlay 2-1 STRUCTURAL INTEGRITYsssac tss
thickness permitted was two weld layers after the first layer to clear PT inspection.
2.2. 1 Axial Flaws The methodology of ASME Section XI Table IWB-3641-3 was used to establish the overlay thickness for a 5.5 inch long through-wall axial crack in the safe-end.
This was the approximate length of the safe-end.
The stress ratio for use in table IWB-3641-3 was computed'rom the hoop stress due to pressure:
O'H
= pR/t
= 9,881 psi where:
O'H
= hoop stress p
= design pressure, 1250 psi R
= safe-end
- radius, 5.375/2 inch, to bound largest radius t
= thickness of safe-end, 0.34 inch, to bound thinnest section For the Code allowable stress Sm = 16,950 psi (ASME Section III), the stress "ratio was conservatively computed as:
O'H/Sm
0.583 Using an axial crack length (lf) of 5.5 inch and a minimum value of ~t
0.825
= 6.7=7 lf By employing the above values for the stress ratio and flaw size, Table IWB-3641-3 was used to iterate and reach a solution for the minimum overlay thickness as follows:
( 1)
Assume the overlay thickness
= 0.18 inch 2-2 STRUCTURAL
. INTEGRITV~pcs
(2)
The ratio of crack depth to repaired wall thickness (a/t) was 0.34/(0.34
+ 0.18)
= 0.654.
(3)
The reduced stress ratio due to the overlay thickness was (0.583)x (0.654)
= 0.381.
(4)
For a stress ratio
= 0.381, IWB-3641-3 permits an a/t= as large as 0.66.
Since the repaired crack depth ratio was 0.654, the overlay thickness of 0.18 inch is sufficient for this assumed axial crack.
2.2.2 Circumferential Flaw The methodology of ASME Section XI Table IWB-3641-1 was used to determine the required overlay thickness for a bounding 360 degree through-wall crack in.
the safe-end at the 0.34 inch thickness weld zone.
The axial stress ratio for use in Table IWB-3641-1 were taken from Tables 2.3 and 2.4.
The resulting primary stress ratios are:
NBA:
(Pm + Pb)/Sm
= (4163 + 7998)/16,590
= 0.717 NBB:
(Pm + Pb)/Sm
= (4186
+ 7476)/16,950
= 0.688 where Pm was the primary membrane stress and Pb was the primary bending stress.
(Thermal expansion stresses are not included in the stress ratio based on reasoning given in Section 2.4).
Employing these stress ratios and a 360 degree crack size, Table IWB-3641-1 was used to iterate and reach a solution for the minimum overlay thickness as fo1 lows:
(1)
Assume the overlay thickness to-be 0.16 inch (2)
The ratio of the crack depth to repaired wall thickness (a/t) was 0.34/(0.34
+ 0.16)
= 0.68.
(3)
The reduced stress ratio due to the overlay thickness was (0.717)x (0.68)
= 0.488.
2-3 STRUCTURAL INTEGRITYassocuvcs
(4)
For a stress ratio of 0.488, IWB-3641-1 permits an a/t of up to 0.687, by extrapolation verified with source equations for net section plastic collapse (Figure 2.1).
Since the repaired crack depth ratio was 0.68, the overlay thickness of 0.16 inch was sufficient for the assumed circumferential crack.
This 0.16 inch thickness was smaller than the 0.18 inch thickness required by the assumed axial flaw and thus the required overlay thickness was 0. 18 inch.
2.3 Structural Margin Calculations As described
- above, the weld overlay, thickness was initially sized to meet the requirements of IWB-3640.
Both through-thickness, 360 degree, circum-ferential cracks and through-thickness, full safe-end length (5.5.inches),
axial cracks were considered.
Using conservative assumptions for safe-end dimensions and loads, an effective overlay thickness of 0.18 inch was found to be adequate.
Increasing the overlay thickness from 0.18 to 0.25,in.
increases the structural margin.
The margin above IWB-3640 requirements can be found by calculating the stress ratio allowed by Code (allowable stress divided by Sm) and comparing this with the applied stress ratio (applied stress divided by Sm) for the repaired safe-end.
Three idealized geometries were considered for the structural margin analysis.
In the first, the original safe-end was assumed to be a straight
- pipe, 0.34 inch thick with a 4.50 inch outside diameter.
The second case treats the safe-end as a pipe of 0.92 inch thickness with an outside diameter of.5.38 inches.
For these two cases, calculations are made for a through thickness, 5.5 inch long axial crack and a through thickness, 360 degree, circumferential crack.
The third case neglects the original safe-end entirely (or assumes the orioinal safe-end is completely cracked with multiple cracks and thus carries no stress) and looks at the structural margin in the overlay alone.
Both. hoop and axial stresses were ~considered.
.2-4
')
STRUCTURAL INTEGRITYsssosuvss
It should be noted that in the following calculations, no credit was taken for the SMAW and GTAW weld overlay layers appl ied with the pipe dry (approximately 0.10 inches thickness).
These layers increase the thickness of the original safe-end and provide even greater structural margin.
2.3. 1 Margin for a Through-.Wall Axial Flaw Acceptable stress ratios for piping with axial cracks are provided in ASME Section XI Table IWB-3641-3.
This table is based on an empirical formulation attributed to Eiber et al and a safety factor of 3 on the stress to'rupture for normal operating conditions (I).
This formulation,gives the allowable stress ratio as O'H
$t/d-I Sm Lt/d-l/M where M is a curvature correction factor given by I+ 1.611 I/2 4Rt and t = wall thickness R
= pipe mean radius d
= crack depth 1
= crack.length h
= hoop stress Sm
= Code stress limit If t is the thickness of the original safe-end, let the thickness of the safe-end with the weld overlay repair be t'.
In this case, the depth of a through-thickness crack will equal t.
The allowable stress ratio for the repaired safe-end is 2-5 STRUCTURAL INTEGRITYissocuvas
Treating the safe-end as a-4.50 inch diameter pipe with an original wall thickness of 0.34 inch, a through-wall axial crack 5.5 inch long, and a weld ov'erlay repair of 0.25 inch thickness, allowable o'H/Sm ='.51.
I Assuming a design pressure of 1250 psi, the applied stress ratio for this configuration is applied <H/Sm
= 0.28.
The ratio of the allowable stress to the applied stress is then 1.82.
Treating the safe-end as a 5.38 inch diameter pipe with an original wall thickness of 0.92 inch, a through-wall axial crack 5.5 inches
- long, and a
weld overlay repair of 0.25 inch thickness, the Code allowable stress ratio is allowable <H/Sm
= 0.35 while the applied stress ratio for this configuration is applied O'H/Sm
= 0.17.
'n this case, the ratio of the allowable stress to the applied stress is 2.06.
Review of these factors shows that with either of these idealized safe-end configurations, there is a large margin between the applied stress ratio and the Code allowable stress ratio.
Realizing that the Code itself contains a
safety factor of 3 on rupture, these margins are significant.
2.3.2 Margin for Hoop Stress (neglecting the original safe-end)
- Next, consider the case where the original safe-end is neglected or is assumed to have multiple axial cracks.
In this case no credit for the original safe-end material is taken and one simply looks at the stress in the 0.25 inch thick overlay.
Using the largest overlay radius, allowable
<H/Sm
= 1.00 2-6 STRUCTURAL INTEGRITY~source l
while applied
<H/Sm
= 0.83 and the allowable to applied stress ratio is 1.20.
Consequently, even if one neglects the original safe-end material, the overlay itself satisfies Code design requirements (which includes a safety factor of 3 on stress to rupture).
2.3.3 Margin for a Through-Wall Circumferential Flaw Acceptable stress ratios for piping with circumferential cracks are provided in ASME Section XI Table IWB-3641-1.
Consider first the idealized case in which the safe-end is treated as a 4.5 inch diameter pipe with an original wall thickness of 0.34 inch and a weld repair of 0.25 inch thickness.
The
,original wall is assumed to have a through-wall, 360 degree, circumferential flaw.
The repaired flaw depth to thickness ratio is then 0.58.
The stress
- ratios, computed from the stresses of Tables 2.3 and 2.4 and corrected for the repaired wall thickness, are then NBA:
applied (Pm + Pb)/Sm
= 0.413 NBB:
applied (Pm + Pb)/Sm
= 0.396 while the Code allowable stress ratio is allowable (Pm + Pb)/Sm
= 0.70.
Therefore, the ratio of allowable axial stress to the. applied axial stress is 1.69 and 1.77 for the NBA and NBB nozzles, respectively.
- Next, consider the case in which the safe-end is treated as a 5.38 inch diameter pipe'with an original wall thickness of 0.92 inch.
Again, assuming a through-wall, 360 degree, circumferential flaw and a weld repair of 0.25 inch thickness, the repaired flaw depth to thickness ratio is 0.79.
(The IWB-3640 maximum permitted flaw depth of 0.75 is not-of concern here sin'ce'he actual flaw depth in the thick portion of the safe-end is much less than
'he assumed through-wall flaw in this discussion.)
The applied stress
- ratios, computed from the stresses of Tables 2.3 and 2.4 and,corrected for the repaired wall thickness, are then, NBA:
applied (Pm + Pb)/Sm
= 0.214 N88:
applied (Pm + Pb)/Sm
= 0.206 while the Code allowable ratio is found by an extrapolation of Figure 2-1 2-7
~ STRUCTURAL INTEGRITY*ssncuus
0'
allowable (Pm + Pb)/Sm
= 0.28.
Therefore, the ratio of allowable axial stress to the applied axial stress is 1.31 and 1.36 for the NSA and NSB nozzles, respectively.
2.3.4 Margin for Axial Stress (neglecting the original safe-end) finally, consider the case where the load bearing capacity of the original safe-end is neglected such that the repaired safe-end is essentially 0.25 inches thick.
The axial stress ratios for the 4.5 inch diameter portion of the safe-ends are then NSA:
applied (Pm + Pb)/Sm
= 0.851 NSB:
applied (Pm + Pb)/Sm
= 0.822 while the Code allowable stress ratio is allowable (Pm + Pb)/Sm
= 1.5.
Therefore, the ratio of allowable axial stress to applied axial stress is 1.76 and 1.82 for the NSA and NSB nozzles, respectively.
In fact, even including thermal stresses of 10,556 psi and 9,566 psi for safe end overlays at nozzles NSA and NSB, respectively, the resulting stress ratios are still less than the, Code allowable stress, ratio of 1.5.
The applied stress ratios for the 5.38 inch diameter portion of the, safe-end are smaller than those shown above.
2.4 Weld Metal Toughness Consideration's Experimental and field evidence exists'which suggests that some austenitic stainless steel weld metal may be significantly less tough than 304 stainless
,steel base metal.
This raises questions concerning the applicability of the IWB-3640 tables which assume that failure will be due to plastic collapse rather than unstable fracture.
Weld metal toughness is not an issue, for the current overlay design because no credit is taken for pre-existing weld metal.
That is, the overlay design 2-8 STRUCTURAL IN TEG RITYmsoa*lss
is based on assuming through-wall, 360 degree circumferential flaws (as well as long axial flaws).
In the current overlay design, the loading can be supported entirely by the high toughness Tungsten Inert Gas (TIG) weld
- overlay, and thus the IWB-3640 analysis for limit load failure is considered appropriate.
Another issue which is associated with that of toughness is whether secondary
- stresses, such as thermal
- stresses, should be included in the IWB-3640 evaluation.
If the overlay is sufficiently tough then the failure mechanism will be plastic collapse, the secondary stresses will be, in effect, relieved by the associated large plastic strains.
Since the TIG overlay weld is considered to be tough enough that any failure will be by plastic collapse;
'it'ould be inconsistent (and overly conservative) to include secondary stresses in the IWB-3640 evaluation.
2-9 t
" STRUCTURAL INTEGRITY~arcs
0
~ 0
.8
~
~
ae~(zb
~ ae -N OV&PCAw
%Wc'Ahfi5/
O I
IO CL (n
. QlC I
I
.7
.5 28
= 180 to 360o L..
~ -
e 2& BO r...
~
~
'o QP I
OPlg CL Ol
.4 U
~ 3 TAbM IMh-3( O'I-I ygic rs; socgS cu/V~s cps.cacgreb VALE/$
(rrF. g pq~wonrs):,
0-26
~ P(gO x
a &"= (gd sAF&ry FAc..TOW' 2,.171 J
I
~
~
~
~
~
'i 4
~
~
~
,I
.2
~ t
~
~
~
I
~
4 lq I
~
~
4
.3
.4
.5
.6
.7
.8
.9 1.0 1.1, 1.2
. 1.3 1.4 1.5 1.6
.1.7 Stress
- Ratio,
~
Pm+
Pb m
FIGURE 2-1 Circumferential. Flaw Size Limits Versus Axial Stress 2-10 STRUCTURAL INTEGRITYissoaivcs
TABLE 2-1 Applied Loading to Jet Pump Instrument Nozzle Safe End NBA Load Oeadweight Thermal OBE (+)
(+)
Forces lbs Fx F
Fz 33
-619
-3
-231
-1046 57 87 178 268 154 297 444 Moments ft-lbs Mz Mx My
-18 50
-1682 132
-424
-3666 335 1042 968 559 1716 1595 These loads are applied forces and moments from the piping outboard of the safe-end/reducer interface.
TABLE 2-2 Loading to Jet Pump Instrument Nozzle Safe End NBB Load Forces lbs Fx Fz Moments (ft-lbs) x M,
Mz Oeadweight Thermal OBE (+)
SSE (+)
0
-524 0
52
-1210 0
224 360 228 241 396 320 13 13
-1233 39 39
-3359 55 880 1293 72
~
1182 1391 These loads are applied forces and moments from the piping outboard of the saf e-end/reducer interface.
2-11 S TRUCTURAL INTEGRITY~pcs
TABLE 2-3 Calculation of Applied Stresses at the Thinnest Section of Safe-End at Nozzle NBA DIMENSIONS 00=
ID=
T=
A=
I=
C=
4.50 IN.
3.82 IN.
.34 IN.
4.44 (IN.) ~2 9.68 (IN,) n4 2.25 IN.
LOADS WITHOUT THERMAL STRESS P=
FX=
FY=
FZ=
MX=
MY=
MZ=
1250 PSI 120 LBS.
797 LBS.
271 LBS.
353 FT.LBS.
1092 FT.LBS; 2650 FT.LBS.
AXIAL STRESSES SM=FX/A + P*R/2*T 4163 PSI SB=MR*C/I 7998 PSI 2-12 STRUCTURAL
~
INTEGRITY~jocu lcs
TABLE 2-4 Calculation of Applied Stresses at the Thinnest Section of Safe-End at Nozzle NBB DIMENSIONS OD=
ID=
T=
A=
I=
C=
4.50 IN.
3.82 IN.
.34 IN.
4.44 (IN.) " 2 9.68 (IN.)n4 2.25 IN.
LOADS WITHOUT THERMAL STRESS P=
FX=
FY=
FZ=
MX=
MY=
MZ=
1250 PSI 224 LBS.
884 LBS.
228 LBS.
68 FT.LBS.
893 FT.LBS.
2526 FT.LBS.
AXIAL STRESSES PM=FX/A + P*R/2*T 4186 PSI PB=MR*C/I 7476 PSI 2-13 STRUCTURAL INTEGRITYissooivcs
3.0 RESIOUAL STRESS ANALYSIS This section describes the two welding residual stress analyses which were conducted as part of the overlay design process.
The analyses provide a
means of 'verifying that the favorable residual stress effects of the overlay procedure do indeed occur for the jet pump instrumentation nozzle safe-end.
The unusual aspects of the safe-end which make the residual stress calculations necessary include the proximity of the low alloy steel
. nozzle and the associated residual stresses from the original nozzle to safe-end weld, the nonuniform thickness of the safe-end, and.the presence of a tube sheet which restrains the inward shrinkage'f the eccentric reducer during overlay welding.
Two weld overlay lengths were analyzed.
In the shorter overlay design, the overlay stops at the thick end of the safe-end outer surface taper, while in the full safe-end length overlay design, the overlay stops on the original safe-end to nozzle weld.
Due to the fact that the designs continued to evolve during the time that the overlay residual stress analyses were being
- made, the analyses do not reflect the final weld overlay designs in all details.
However, the results of the calculations
" clearly show that the beneficial residual stress effects of the heat sink weld overlay process do occur for both overlay lengths.
3.1 Background
on the Overlay Welding Analysis Methodology The methodology for numerical modeling of pipe welding techniques was initially developed at Battelle's Columbus Laboratories (2,
3,
- 4) with support from the Electric Power Research -Institute (EPRI) and the U. S.
Nuclear Regulatory Commission (NRC).
As a result of numerous complexities associated with modeling the welding process and the extreme expense that would be incurred if each aspect of the welding process were modeled with s ate of the art analytical
- tools, the methodology uses a
number of simplifyin'g assumptions.
Most of these assumptions are difficult to
'ustify based solely on analytical reasoning and therefore extensive experimental verification of the methodology has resulted (2-7).
3-1 STRUCTURAL INTEGRITYsssocuvcs
While the'weld modeling methodology was developed for pipe girth welds, and the bulk of experimental verification has been for girth welds, the methodology has been applied to other geometries and welding techniques (3, 7).
The methodology has recently been applied,to the overlay weld repair of a sweepolet and has been found to provide residua'I stresses which are in good agreement with surface and through-thickness residual stress measurements obtained from a sweepolet mock-up (8).
The methodology for predicting welding residual stress involves the use of two basic models.
The first model is used to compute the transient 3D temperature history due to welding.
The second model uses this temperature history as input and provides residual stresses,
- strains, and deflections.
3.1.1.
Temperature Model The temperature model is based on the closed form analytica'I solution for a point heat source moving along a straight path in an infinite medium (2).
This solution assumes that thermal properties are independent of temper-ature and does not include the effect of phase transformations.
The principal advantage of using this analytical solution instead of more state of the art numerical solutions is the tremendous cost savings.
Three-dimensional, nonlinear, transient thermal analysis is quite expen-sive, both computationally and in terms of the time to set up the finite element or finite difference model, The analytical solution on the other hand is amenable to small desktop calculators or microcomputers and requires much less time to exercise.
The most important justification for using this simple model, however, is the fact that it provides a realistic 3D transient temperature history which can be made to fit experimental temperature data through the use of a weld heat input efficiency factor and various other modifications which take advantage of the principle of superposition (e.g., the solution for a doubly insulated plate is easily obtained through superposition methods).
3-2 STRUCTURAL INTEGRITY~pcs
i
3;1.2 Residual Stress Model The stress analysis model is more sophisticated than the temperature models The stress model is based on the incremental finite element approach for solving boundary value problems with thermal-elastic-plastic material behavior.
Unlike the temperature
- model, the stress model uses fully temperature dependent mechanical properties.
A primary assumption of the weld modeling methodology is that the welding process can be adequately represented by a two-dimensional stress analysis mode.
This assumption is motivated primarily by cost considerations, but through many comparisons with experimental data has been shown.to result in good residual stress predictions.
The use of axisymmetric models is most common for pipes and similar geometries while plane strain is more appropriate for plates'he residual stress modeling of the safe end is via two-dimensional axisymmetric finite elements..
3.2 Short Overlay Welding Analysis This analysis models the application of three weld overlay layers to the safe-end.
The first layer is a 30 kJ/in stick weld which is made without water inside the safe-end.
The model assumed that this layer was 0. 10 in.
thick.
The second and third weld layers are automatically deposited 37 kJ/in TIG welds which are each 0.08 inches thick.
The safe-end was assumed to be filled with water during these automatic welds.
The welding speed was taken as 3 in/min for all weld layers.
All three layers extend from approximately 1.0 inch on the eccentric reducer side of the safe-end to
, reducer weld centerline to about 3:4 inches on the opposite side.
This end
'point coincides with the end of the outer surface taper of the safe-end.
The welding direction was assumed to be circumferential for all welds with
. the initial weld passes being on the eccentric reducer end of the safe-end and the final weld passes being on the nozzle end.
The total thickness of the overlay is 0.26 inches while the thickness of the automatically deposited portion which uses a water heat sink is 0.16 inches.
3-3 STRUCTURAL tNTEGRITY*ssocurcs
The finite element grid for the short overlay is shown in Figure 3-1.
The two boundries at the upper and left corner of the model are not actual boundaries of the nozzle and are thus provided with "rollers" so as to represent the constraint due the portion of the nozzle which is not modeled'.
The opposite end of the model is allowed to translate axially but is not allowed to rotate.
The safe-end to nozzle weld is represented by a planar transition in material properties from stainless steel to low alloy steel.
. The nozzle to safe-end and safe-end to reducer weld induced residual stresses were not simulated in this analysis.
In the case of the nozzle weld, this was deemed appropri,ate due to the separation distance from the overlay.
The reducer
- weld, on the other
- hand, is directly under the overlay.
Experience has shown that final residual stresses are not significantly influenced by the original residual stresses when the butt welded region is totally covered by overlay weld.
The tube support plate inside the reducer was not included in this model.
However, it was included in the analysis of the full 'length overlay.
The results of this full length overlay analysis indicate that the tube support plate does not inhibit the residual stress improvement effects of t'e overlay application for the safe-end, and affects the reducer residual stresses to only a small extent (as will be illustrated in the section pertaining to the full length overlay analysis).
3.2. 1 Short Overlay Residual Stress Analysis Results Figures 3-2 and 3-3 show the final hoop residual stress distributions for the short overlay analysis.
Figure 3-2 shows the hoop stress isobars for the thin portion of, the safe-end and Figure 3-3 shows the hoop stress isobars for the tapered portion of the safe-end.
Figures 3-4 and 3-5 show the corresponding plots for the axial stress component.
Both stress components are characterized by compressive stress on the inside portion of the section and tensile stress on the outside portion.
This type of distribtuion is the intended result of using a, heat sink weld.
3-4 STRUCTURAL INTEGRITY~res
Weld Nozzle~
{Low Allo Safe-End (Stainless Steel)
Eccentric Reducer (Stainless Ste 1)
Weld Over-la Weld rI, FIGURE 3-1 Finite Element Grid For The Short Overlay Analysis
fr J'P I N1. 580 SIGNA T
2 QJ B 50000. 00 G 40000.00 30000.00
+ 20000. 00 X 10000.00 0.00
-10000.00 R -20000.00 Y -30000.00 X -40000. 00 3K -50000.00 z0 r7l
'2 O
Zl Vl uj Z
z V)
Pl à FIGURE 3-2 Residual Hoop Stress Isobars in the Thin Portion of the Safe-End After Completion of the Short Overlay
J'P.I N 1..580 SIGMA T
Cl 50000.00 6 40000.00 30000.00
+ 20000.00 X 10000.00 0.00
+ -10000. 00-Z -20000.00 Y -30000.00 X -40000.00
-50000.00 e
FIGURE 3-3 Residual Hoop Stress Isobars in the Thick Portion of the Safe-End After Completion of the Short Overlay
3PIN1;580 SIGNA T
6 o
5 Ql
+
X
+
R N
y O
50000.00 40000.00 30000.00 20000.00 10000.00 0.00
-10000.00
-20000.00
-30000. 00
-unnnn.nn
-SUOUO.UU 7
ol r
J
~P f
INZ Z
V FIGURE 3-4 Residual Axial Stress Isobars in the Thin Portion of the Safe-End After Completion of the Short Overlay
3P IN1. 580 0
0 30000.00 20000.00 10000. 00
.00
-10000.00
-20000.00
-30000.00
-40000.00
-50000.00 R
Y X
N SIGMA Z
0 m S0000.0 8 40000.0 FIGURE 3-5 Residual Axial Stress Isobars in the Thick Portion. of
.the Safe-End After Completion of the Short Overlay
Based on previous experience, the overlay induced residual stresses do not change significantly after the first heat sink layer is deposited.
There-
~
fore, depositing one or two additional layers of TIG weld to the overlay would not be expected to significantly alter the residual stresses.
Thus the current results are expected to be representative of overlays with as many as four overlay layers.
3.3 Full Length Overlay Welding Analysis This analysis models an overlay which extends from near the transition in the reducer to the middle of the original nozzle to safe-end weld.
Due to the original nozzle to safe-end weld being at the edge of the region to be
- overlaid, any effects of the butt weld residual stresses and/or bimetal interface were included in the analysis by first simulating the nozzle to safe-end butt weld.
Figure 3-6 shows the finite element grid for this full length overlay analysis.
The effect of the tube support plate was included in this analysis of the full length overlay.
The support plate was modeled as being a rigid support at the inner reducer surface which prevents any inward radial deflection but allows unrestrained outward deflection.
The nozzle to safe-end weld was simulated using ten weld passes applied in six layers.
The welding was assumed to be a 50 kJ/in, stainless steel stick weld, being placed at a rate of 3 in/min.
While the short overlay analysis approximated the low alloy/stainless steel interface as a radial line, this analysis accurately represents the interface including the weld butter.
After the nozzle to safe-end weld simulation was complete, the application of three weld overlay layers were simulated.
The first portion of the first layer.is a 50 kJ/in stick weld which is'0.10 in. thick and is made without water inside the safe-end.
This stick weld starts at the safe-end to reducer weld and stops at the end of the safe-end outer surface taper.
The remainder of the first layer and the second and third layers were modeled as 37 kJ/in GTAW welds with a water heat sink.
The second and third 3-10 STRUCTURAL
- INTEGRITYissomvcs
z Vl Q C
.z g FIGURE 3-6 Finite Element Grid for the Full Length Overlay Analysis
layers are again 0.08 inches thick so that the final overlay thickness, including the non-heat sink layer, is 0.26 inches thick.
The overlay modeled here differs from the final design in two respects'irst, in the final design, a non-heat sink weld was applied to the thick portion of the safe-end adjacent to the nozzle while in the model this was applied with a heat sink.
This difference does not significantly alter the final residual stresses because the effect of the heat sink is very small at this section due to the large wall thickness.
The second difference is the number of overlay layers and the final overlay thickness.
As discussed above for the short overl'ay model, the residual stresses do not change significantly with additional overlay layers and thus the model is believed to provide a
good representation of the final overlay 'design residual stress state.
3.3.1 Full Length Overlay Residual Stress Results Figures 3-7 and 3-8 show the hoop and axial residual stresses in the vicinity of the nozzle to safe-end weld before application of the overlay repair.
Note that both stress components are tensile on the inside surface near the weld.
The hoop component is highly compressive to either side of the weld and may have been a contributing factor in the tendency for less cracking in the thick portion of the safe-end.
Figures 3-9 and 3-10 show the residual stress in the vicinityof the nozzle to safe-end weld after the overlay repair.
In this case, both hoop and axial stresses are compressive over the inner one-half to two-thirds of the original wall thickness.
As expected, tensile residual stress is produced in the overlay itself and in the outer portion of the original wall. It is seen that the presence of the nozzle to safe-end weld residual stresses prior to application of the overlay do not adversely affect the overlay induced residual stresses.
Furthermore, it would seem that including the nozzle to safe-end weld residual stresses had little effect o'n the final residual stress state.
3-12 STRUCTURAL INTEGRITYacsouasal.
t f" I-r-
r r
TVA04.200 SIGMA T
n soooo.oo O 90000.00 3OOOO.OO
+ 20000.00 X 10000.00
- 0. 00
+ -10000.00
-20000.00 Y -30000.00 X -(10000.00 q.
% -50000.00
( t'"
I f
t""
(
f'sJ if III I
V co%'~HE 8 04
/+~5gyg gp~~~~~ qr z N Q C x g Q c
~AC<-CuQ FIGURE 3-7 Nozzle to Safe-End Weld Residual Hoop Stress Before Application of the Overlay
I
(
I TVA04.200 SIGMA Z
r'-
r I
r G. 50000.00
<3 40000.00 30000.00
+ 20000.00 X )0000.00 0.00
. + ->0000.00
-20000.00
-30000. 00 X -90000. 00
-50000.00
.au c 0+Pwcwi~~
r
< a~prPChS.oJ
~4 WSrocJ nrod8 c<
~M Iq Figure 3-8 Nozzle to Safe-End Weld Residual Axial Stress Before Application of the Overlay
. TVA04.980 SIrgA T
7 n
+~
g Y
50000.00 93000.CO 000.00 2'-900.00 1-.:000.00 0 ~ 00'
- 0000.00
-20000-00
-:0000.00
-F0000.00
-50000.00 Q4<C- ~M@
. FIGURE 3-9 Nozzle to Safe-End Weld Residual Hoop Stress After Application of the Overlay
I I '
TYAOQ. 980 SIGMA Z
I Q
%)
soooo.oo 6
u.000-00 30000.00
+ 20000.09 X 10000.00
". 00
+
!0000.00 R -20000.00 Y -30000.00
-40000.00
-50000.00 a
~~<~ sir'0 FIGURE 3-10 Nozzle to Safe-End Weld Residual Axial Stress After Application of the-Overlay
Figures 3-11 and 3-12 show the predicted residual hoop and axial stress-isobars near the safe-end to reducer weld.
These results reflect the presence of the tube support plate.
These figures can be compared to Figures 3-2 and 3-3 which do not include the effect of the tube support plate.
In the reducer, the hoop stresses are less compressive for the case which includes the tube support plate but the hoop stresses in the safe-end are largely unaffected.
The axial stresses are largely unaffected by the presence of the tube support plate.
j 3-17 STRUCTURAL INTEGRITY~res
TVA04.980 SIGNA T
~
SOOOO:Oo 6 40000.00 30000.00
+ 20000.00 X 10000.00 e 0.00
.+ -10000.00
-20000.00 Y -30000.00 X -40000.00 x -50000.00 cl Scr PPpdr PcA TC,
~<5'+cf7RrC Rr Gare 8 FIGURE 3-11 Residual Hoop Stress Isobars in the Thorn Portion of the Safe-End After Completion
-of the Full Length Overlay (with tube support plate)
0
TVA04..980 SIt N~
Z p 2 SOC:":.00
+~ 6 AGE -"'.00 r
lg A 30e.1.00
+ 200 '.00 X iOOC:.00 00
+ -1C=:.0.00 R
-20".~~0. 00
'( -30C50.00
~<
~ -~00=:0.00
-se;ao.oo
~g
+ <Cc".KMD
~tKV%'IC FIGURE 3-12 Residual Axial Stress Isobars in the Thin Portion of the Safe-End after Completion of the Full Length Overlay (with tube support plate)
4.0 DISCUSSION AND CONCLUSIONS Weld overlay repairs have been designed for Browns Ferry Unit 3 jet pump instrumentation nozzle safe-ends in which IGSCC cracks have been iden-tified.
The repair configuration consists of multiple layers of weld overlay covering the entire safe-end and applied from the change-in-section region of the eccentric reducer to the center of the safe-end to nozzle weld.
The weld overlay repair was designed based on ASME, Section Xl, IWB 3640 criteria.
Full safe-end thickness, 360 degree, circumferential cracks and full. safe-end thickness, full safe-end length (5.5 inch), axial cracks were assumed in the analyses.
Safe-end stresses were calculated from TVA supplied loads and conservative combinations of safe-end dimensions.
These analyses showed that an effective weld o'verlay thickness* of 0. 18 inch would be sufficient to maintain Code margins reflected in IWB 3640 criteria.
Following initial weld overlay design, a decision was made to increase the effective overlay thickness from 0.18 inch to 0.25 inch.
Calculations were made to show the additional structural margin introduced by increasing the overlay thickness.
These calculations show that even if one assumes the original safe-end material is removed and only the weld overlay material
- remains, Code margins are maintained.
Residual stress calculations were made for a short overlay design (which was not used) and for the final full length overlay design which was employed.
These calculations show compressive hoop and axial stresses in the inner one-half to two-thirds of the safe-end and reducer wall over the entire length of the overlay.
The presence of the tube support plate in the eccentric reducer near the reducer to safe-end weld does not produce tensile residual stress near the inside surface of the reducer or safe-end.
- Thickness of overlay applied-after a clean dye penetrant examination of initial overlay layers.
4-1 f
STRUCTURAL L
INTEGRITY~ssocuvas
~,
Although the weld overlay repairs are designed for one fuel,cycle, it is anticipated that a longer design life could be demonstrated by showing that crack extension by stress corrosion and/or fatigue is not a concern.
4-2 STRUCTURAL
'a
~NTEGRITY~wc~
5.0 REFERENCES
1.
Ranganath, S.
and Norris, D., "Evaluation Procedure and Acceptance Criteria for Flaws in Austenitic Steel Piping".
2.
- Rybicki, E. F., et al.,
"Residual Stresses at Girth-Butt Welds in Pipes and Pressure Vessels", Final Report to U. S. Nuclear Regulatory Commission, Division of Reactor Safety, Research under Contract No.
AT (49-24)-0293, NUREG-0376, published
- November, 1977.
3.
- Rybicki, E. F.,
et al.,
"Residual Stresses Due to Weld Repairs, Cladding and Electron Beam Welds and Effect of Residual Stresses on Fracture Behavior",
Final Report to U.
S.
Nuclear Regulatory Com-
- mission, Division of Reactor
- Safety, Research under Contract No.
AT(49-24)-0293, NUREG-0559, published
- December, 1978.
4.
- Brust, F.
W.
and Stonesifer, R. B., "Effect of Weld Parameters on Residual Stresses in BWR Piping Systems",
Final Report to Electric Power Research Institute, NP-1743, Research Project 1174-1, March 1981.
5.
Rybicki, E. F., et al.,
"A Finite Element Model for Residual Stresses and Deflections in Girth Butt Welded Pipes",
Journal of Pressure Vessel Technology, Vol. 100, No. 3, August 1978, pp. 256-262.
6.
- Rybicki, E.
F.
and Stonesifer, R.
B.,
"Computation of Residual Stresses Due to Multipass Welds in Piping Systems",
Journal of Pressure Vessel Technology, Vol. 101, No. 2, May 1979, pp.
149-154.
7'.
- Rybicki, E.
F.
and Stonesifer, R.
B.,
"An Analysis Procedure for Predicting Weld Repair Residual Stresses in Thick-Walled Vessels",
Journal of Pressure Vessel Technology, Vol. 102, No. 3, 1980, pp. 323-331.
8.
Report "Design Report for Recirculation Piping Sweep-o-lets Repair and Flaw Evaluation, Browns Ferry Nuclear Plant Unit 1", SIA Report SIR-83-006, Rev. 0., 10/21/83.
5-1 r;
STRUCTURAL INTEGRITYissocurcs
ATTACHMENT 3 In addition to the aforementioned welds inspected pursuant to the unit 3 shutdown order, the following welds were examined ultrasonically with satisfactory results.
A.
N1A and N1B, 28-inch recirculation outlet nozzle-to-safe-end.
B.
N2A, N2B, N2C, N2D, N2E,
- N2F, N2G, N2H, N2J, and N2K.
Twelve inch recirculation inlet nozzle-to-safe-end.
Co N-8A and N-8B get pump instrumentation nozzle-to-safe-end weld D.
N-10 2-inch standby liquid control - core differential pressure nozzle-to-safe-end and safe-end-to-pipe welds E
N-12 A&B 2-inch instrumentation nozzle-to-safe-end and safe-end-to-pipe welds F-N-11 A&B and N-16 A&B 2-inch water level nozzle-to-safe-end and safe-end-to-pipe welds.
ATTACHMENT 4 METALLOGRAPHIC EVALUATION OF UNIT 3 REACTOR VESSEL SAFE ENDS BROWNS FERRY NUCLEAR PLANT The sub.1ect evaluations were performed as a followup after crack indications were found in the.iet pump instrumentation nozzle safe ends (N8A, N8B).
The.]et pump instrumentation nozzle safe ends (N8A, N8B), both unit 3 recirculation outlet nozzle safe ends (N1A, N1B), and
? of the 10 recirculation inlet nozzle safe ends (N2C, N2H) were examined.
Since the 10 recirculation inlet safe ends were fabricated from only two heats of material and the 2 safe ends that were checked were from each of these
- heats, this sample gives adequate confidence of the condition of all recirculation in)et safe ends.
The metallographic determination used is an electrolytic oxalic acid etch to simulate the ASTM A262 practice A
sensitization test.
This teat is a visual determination based on the difference 1n corrosion hehavior in the grain boundary area of the sensitized material vei sus the nonsensittzed material.
The grain boundaries will be preferentially attacked in a sensitized
- material, resulting f n a "ditched" appearance.
The nonsensitized material will be attacked uniformly across %he grain, resulting in a "stepped" appearance.
Figures 1 and 2 give the visual results of the testing on nozzles N8A, N8B, N2C, and N2H taken from field replicas.
The recirculation outlet nozzles were visually inspected in the field with a microscope since the replication failed to give adequate results..
Figure 1 is the get pump instrument safe ends.
Both of these microstructures would be classified as "ditched," although not all the grain boundary area was attacked.
(The Practice "A" guidel,i.ne is that sensitization is indicated when any grain is completely surrounded by ditched grain boundary.)
The inlet nozzles (figure 2) and the outlet nozzles,(from visual microscopic examinations) were found to have a "stepped" structure, indicating no sensitization, The attached table gives the carbon content of all the nozzles exami.ned, The examinations included each nf the heats of mater ial involved.
The N8A and N8B safe ends hnd <'xcessively high carbon contents
(.08 and
,09
- percent, respectively) and arc made. from the same heats of material as the outlet nozzle safe ends.
The NAA and NAB safe ends apparently received some post-solut/on annealed h< sting, possibly during the local stress rel'<ef of the pirth welds by Ishikawa,\\ima-Harima Heavy Industries (IHI).
These safe ends were replaced after the furnace stress relief of the individual shell cour~<'.s hut before the final assembly of the shell
courses.
The recirculation inlet and outlet nozzles'aximum temperatures
>were recorded during the local girth weld stress r'elief.
No information is currently available on the maximum temperature that the.]et pump instrumentation nozzles or the standby.liquid control (SLC) nozzles attained during local stress reliefs.
Summar and Recommendations The.1et pump instrumentation nozzle safe ends appear to have been heated into the sensitization
- range, probably during a stress relief operation of the vessel at IHI.
the e
2.
The metallographic results agree with the ultrasonic testing results su s on ese safe ends.
Both the NSA and NAB safe ends appeared sensitized, and both have crack-like hndications extending beyond the weld heat-
~
affected zones.
The large nozzles did.not appear sensitized and have not shown cracking during the inspections current to this report.
3.
The SLC safe end has been ultrasonically 1nspected and is free of defects.
We therefore do not expect that this safe end's sensitized.
The unit 2 vessel was also assembled at IHI, so the same nozzles may be similar lv affected-.
In addition, the Jet pump instrumentation nozzle safe ends have carbon contents of.08 and
.09 (information supplied by GE).
We therefore recommend that these nozzles be metallographically and ultrasonically examined dur ing the next refueling outage.
The unit 1 safe ends are of >ower carbon content and were replaced by BKW (with the exception of the recirculation outlet safe end whi h ID en s w
c were ID c a
) after total stress re].ief of the reactor pressure vest
.1 W
ther efore do not expect. that the unit, l,ie<
pump instrumentation nozzle safe ends are sensitized.
TABLF.
Heat Number~ and Carbon'ontent.
of Unit 3 Nozzle Safe Ends Nozz1r.
Wn.
Rent Vo.
C.'arbon (5)
Recl rculat ion Out,let Recirculation Inlet Jet.
Pump Tnntrumrnt.at.ion N1A N1R N2A NPB N2C N2D NPi; NPF Nc"0 NPH NPJ N c"K NRA NAi1 FS-781-4 FS-781-5 VL-274-1 VL-274-1 VL-274-1 VL-274-1 YL-274-1 VL-274-:?
VL-274-~
VL-274-2 VL-274-:?
VL-274-2 FS-781-4 F.S-78 1-")
o.o8 0.09 o.n6 n.o6 0.06 0.06 0.06 0.07 0.07 o.07 0.07 0.07 0.08 0.09
g~
~
'I'*
c
~<;,
~ i
"- ~
~
/00rc I:Igure IA:
Result.s of t:h<<oxalic acid etch on norw.le NSA.
All the areas appearc<l t,o b<'nder-<'tchod
<ared with ASTM I'r~ctice A, l>nt.:>re <'tchc>out;h t.o m;tice a reasonable determinat ion.
No(c th<
<<li I'f< rene<
in et< hlnl, behavior th<so no><><I<'s (I'i}',or<s IA and II3) <nld th<'nlet noz/I<-'s and 2>I3).
prepared
- A262, between (I:it,ures 2A t
),
~
~
~
~
'I
~
f
~
g
~
r i
I
~ W
/
/
~ ~
~
\\':iI,'ure III:
NoÃF.Ic NHII
L) 1.
~I
~
~
I'igure 2A:
Noir.le N2(;
IOOx Cg I
D L
~
~
- "dyL 4:.>> i -.
L I'inure 28:
NorzIi N:!II
/00 x
ATTACHMENT 5 Browns Ferry Nuclear Plant Unit 3, Cycle S
Xnduction Heating Stress Improvement (XHSX) of XGSCC Susceptible 304 Stainless Steel (SS) Welds 1.0 Introduction Results of the IEB 83-02 XGSCC Ultrasonic (UT) examinations indicated that the recirculation, residual heat removal (RHR), core spray, and reactor water (RWCU) cleanup piping systems were free of indications of IGSCC.
To pr event IGSCC initiation, induction heating stress improvement (IHSX) was implemented on all susceptible 304 SS Class 1
welds in those systems.
General Electric Company.was contracted to perform IHSI under a two-phase workplan.
Phase I consisted of' site survey to evaluate the implementation of IHSI on candidate welds.
Phase II included coil development, scheduling, equipment setup, and all other work necessary to complete the XHSI treatments on welds identified as treatable in Phase X.
2.0 Phase I - Site Surve The site survey was,conducted from January 22, 1984 through February 3,
1984.
The following work was per formed during the survey:
- evaluation of candidate welds designated by TVA for treatment
- collection of weld contour data
- verification of weld accessibility and identifica tion of obstructions
- measurement of piping systems - study of potential XHSI equipment locations The survey information was then evaluated and a workplan for Phase XI was laid out.
,It was determined that XHSI could be implemented on 147 welds.
The treatable welds are listed in Tables 1 through 5 and their locations are shown in f'gures 1 through 3.
Weld DRWC-3-5B, the first outboard isolation, valve to flued head weld on the reactor water cleanup system, was added to the scope in May 1984 bringing the total number of treatable welds to 148.
Seven recirculation, 20 core spray, and two RHR welds were excluded f'rom the treatment work scope; these are given in table 6.
.The recirculation welds which were excluded are fillet welds which could not be treated by.then available IHSI methods.
XHSI techniques to treat fillet welds are currently under
development, and these welds will be treated at a future outage when such XHSI methods can be used successfully.
The core spray and RHR welds ~hich were excluded are carbon steel or low-carbon SS and are not considered susceptible to IQSCC.
They will requ5re no further disposition.
2.2 Induction Coils The survey results indicated that 57 induction coils would be needed to perform the 147-weld IHSI work scope.
This required 22
~ new coils in addition t'o the 35 coils already available to GE.
2.3 Interferences Forty-two interferences were identified during the site survey.
The list below gives the type and number of each obstruction identified.
~Te No. of Obstructions Hanger Lug Hanger Pad 3
Hanger Bracket 6
Hanger Rod 1
Hanger Clamp 5
Mhip Restraint 4
Electrical Conduit Carbon Steel Pipe Chain Falls and Mire Rope 4
Snubber Lug 1
Temporary Scaffold 2
Clamp 2
The obstructions created by the four whip restraints were avoided by coil redesign.
Mater-cooled shielding was fabricated to protect the whip restraints from overheating during treatment but was found to be unnecessary as the treatments progressed.
All other interferences were removed prior to the treatment of each weld.
Plant equipment, such as hanger components and
- conduit, was restored following treatment of the associated weld.
2.4
~Eau1 ment The equipment locations were also determined during the survey.
Equipment needed for XHSI consisted of a 4160/480V three-phase transformer, a frequency converter (power supply), work stations, a cooling water system, and a data acquisition system.
Each work station consisted of a voltage-reducing transformer, a capacitor
-bank, and a variable transformer that matches the converter output power to the impedance of the induction coil.
The cooling water system was a self-conta'ned closed loop supplying cooling water to the frequency converter, work station, coils and
electrical cables.
The data acquisition system monitored and documented the pipe temperature during each XHSX treatment.
Thermocouples were attached to the pipe's outer surface and connected to the data acquisition system.
Two work stations were located outside of the drywell, one at each equipment hatchway.
The XHSX control room, which housed the data aquisition hardware as well"as the process control panel, was located at the personnel air lock.
The power supply and cooling supply system pump skid were placed on elevation 593.
Xn addition, a direct line communication system was established between the power supply, pump skid, heat station, and IHSI control room.
A communication line between the IHSX control station and the reactor control room was also established.
3.0 Phase XI - IHSX Treatments The IHSI treatments were performed from March 14 through May 25, 1984.
The following table shows the time taken to complete each system.
Svstsm No. Welds Date First Thermocouple Installed Date Last Thermocouple Removed RWCU CS Recirc RHR Weld DRWC-3-5b 13 3/14 9
, 3/16 q6 3/23 29 5/1 1
5/21 3/22 3/22 5/14 5/17 5/25 An overall average of 3 treatments were performed each day.
All treatments were judged to be successful.
In general, the treatment sequence for each weld included thermocouple (TC) installation, coil installation, low-power idle run, coil adjustment, treatment, coil removal, TC removal, and PT of TC tack welds.
Selected welds were also ultrasonically examined following the IHSX treatment.
3
. Eleven TCs were attached to each weld to record temperature data during XHSI treatment.
Five TCs were p'ositioned on one azimuth, parallel to the center axis of the pipe, with one centered on the weld crown and two on either side placed in the heat-affected zone (HAZ) and at the edge of the XHSI heat zone.
Two TCs were also attached on the HAZ on the three remaining azimuths spaced 90o apart.
On some welds, a twelfth TC was used to monitor the temperature of permanent obstructions positioned close to the XHSI heat zone.
The data acquisition system had a 12>>channel
4 input, allowing all data to be recorded on tapes, and provided individual TC temperature printouts every 4 seconds.
A temperature profile plot was also provided during each IHSI treatment.
The TCs were resistance welded to the pipe in accordance with ASME Section IXX, NB4311-3.
Following the IHSX treatment, the TCs were removed and the affected areas were blended smooth and liquid penetrant examined in accordance with ASME Section IXI NB5000.
3-2 Low-Power Idle Run A low-power pre-treatment at 250oC + 50oC (482oF'+ 90oF) was performed on each weld gust prior to the full IHSI treatment to verify that the TCs were operative, the coil was positioned correctly, the water was cooling effectively, and load controls were operative.
On some welds several low-power tests were required to precisely align the 'coil.
3 '
IHSI Treatment To obtain a successful IHSI treatment, the minimum throughwall temperature difference of 275oC (495oF) was effected within the treatment zone for the minimum heating time (see Table 7 for process control parameters).
This was achieved by heating the pipe outer surface within the treatment zone to between 400oC (752oF) and 575oC (1067oF) while simultaneously cooling the inner surface with system water flowing at the specified rates.
Several welds required more than one attempt to obtain a successful treatment.
In the treatment of 31 welds, there were deviations from the process control parameters;,
these were all determined acceptable by GE engineering and documented on NCR and FDDR reports.
Post XHSI Ultrasonic Examination A 25-percent sample of IGSCC susceptible welds were ultrasonically examined following the IHSI treatments.
The welds were selected for examination based on the following factors:
1.
Welds which had recordable indications and/or underwent evaluation and were found to have geometric reflectors during initial examination for ISGCC.
2.
Welds in the same location where defects were found during the unit 1, cycle 5 ISGCC examinations.
The welds in the sample are listed in Table 8.
Weld No.
DSRWC 4 was found to have linear indications.
This weld was dispositioned by the Nuclear Central Office Metallurgy and Codes Section as acceptable for continued service. It will be reexamined during the unit 3, cycle 6 inservice inspection.
4.0 Conclusions The IHSI program undertaken on Br owns Perry unit 9 was successfully completed within schedule and without major problems.
All IGSCC susceptible 304 SS welds on the subject systems inboard of the penetrations received successful IHSI treatments.
The seven recirculation line 4-inch fillet welds, which were not treated, are considered less susceptible 'to IGSCC than butt welds.
As stated previously, these welds are under consideration for future treatment as the technology becomes available.
IHSI has been shown to offer a level of mitigation against IGSCC.
Treatment of these recirculation, RHR, core spray, and RWCU will be cost effective by providing one or more cycles of operation with relative freedom from cracking and associated repair activities.
Current speculation is that IHSI combined with other mitigation measures, e.g., alternate water chemistry is required to provide life-of-plant immunity.
TABLE.1 RECIRCULATION LOOP A&B BFNP-3 SIZE (IN. )
CONFIGURATION STP/SE STP/LREL STP/TEE STP/STP VLV/LREL VLV/STP.
STP/SREL PMP/SREL STP/PMP LOOP A
B B
B B
B B
TVA VELD IDENT.
GR-3-53 GR-3-59 KR-3-45 GR-3-54 KR-3-47 KR-3-2 KR-3-50 GR-3-60 KR-3-51 KR-3-24 GR-3-55 KR-3-46 KR-3-3 KR-3-25 GR-3-61 GR-3-56 GR-3-3 GR-3-62 GR-3-29 GR-3-57 GR-3-2 GR-3-63 GR-3-28 KR-3-48 KR-3-52 GR-3-58 GR-3-64" GR-3-1 GR-3-27 Page 1 of 4 2/27/84
TABLE 1 (Cont.)
PECIRCULATION LOOP A&B (Cont.}.
BFNP-3 SIZE (IN.)
28 22 I~
~ I 12
~ I
- 28 CONFIGURATION CRS/TEE CRS/RED HDR/ECP HDR/CRS HDR/VLV HDR/SOL STP/SOL CRS/TEE LOOP B
B TVA VELD IDENT.
GR-3-34 KR-3-11 KR-3-33 KR-3-15 KR-3-37 KR-3-12 GR-3-18 KR-3>>34 GR-3-44 GR>>3-25 GR-3-26 GR-3-51 GR-3-52 KR-3-13 KR-3-14 KR-3-19 KR-3-20 KR-3-35 KR-3-36 KR-3-41 KR-3-42 GR-3-9 GR-3-12 GR-3-19 GR-3-22 GR-3-35 GR-3-38 GR-3-45 GR-3-48 GR-3-8 Page 2 of 4
- REVISED 3/14/84
)
TABLE 1 (Cont.)
REQIRCULAT10N LOOP AKB (Cont.)
BFNP-3 SIZE (IN.)
12 tc4
~ ~
CONFIGURATION STP/SE STP/RE!l l"T/ECP STP/LRE'
~
~ I LOOP A
B B
A B
B B
TVA MELD IDENT.
GR-3-11 GR-3-14 GR-3-17 GR-3-21 GR-3-24 GR-3-37 GR-3-40 GR-3-43 GR-3-47 GR-3-SO GR-3"15 GR-3-41 GR-3-63A GR-3-4 k
GR-3-7 GR-3-30 GR-3-33 GR-3-10 GR-3-13 GR-3-16 GR-3-20
-GR-3-23 GR-3-49 GR-3-46 GR-3-42 GR-3-39 GR-3-36 KR-3-16 page 3 of 4
- REVISED 3/19/84
TABLE 1 (Cont.)
RECIRCULATION LOOP AKB (Cont.)
BFNP-3 SIZE (IN.)
CONFIGURATION STP/LREL LOOP A
B TVA hIELD IDENT.
KR-3-17 KR-3-18 KR-3-21 KR-3-22 KR-3-44 KR-3-43 KR-3-40 KR-3-39 KR-3-38 Page 4 of 4
2/27/84
BFNP-3 SIZE (IN.)
CONFIGURATION STP/TEE STP/LREL STP/LREL STP/LREL LREL/VLV STP/VLV STP/VLV STP/SOL (6")
DRHR-3-19 DSRHR-3-9 DSRHR-3-10 DSRHR-3-11 DRHR-3-21 DRHR-3-22 DRHR-3-23 DSRHR-3-8 2/27/84
TABLE 3 RHR LOOP B (DIS CHARGE)
BFNP-3 SIZE (IN.)
24 CONFIGURATION TEE/STP STP/VLV VLV/SREI SREL/STP STP/STP STP/VLV SREL/VLV SREL/STP/SREL TVA VELD IDENT.
DRHR-3-18 DRHR-3-17 DRHR-3-16 DSRHR-3-7 DSRHR-3-6 DRHR-3-15 DRHR-3-14 DSRHR-3-5A DSPHR-3-5 STP/SREL DRHR-3-13 RHR LOOP A (DISCHARGE) 2-STP/TEE STP/VLV SREL/VLV-SREL/LREL STP/LREL STP/STP STP/VLV VLV/LREL STP/LREL STP/SREL STP/SREL DRHR-3-9 DRHR-3-8 DRHR-3-7 DSRHR-3-4A DSRHR-3-4 DSRHR-3-3 DRHR-3-6 DRHR-3-5 DSRHR-3-2 DSRHR-3-1 DRHR-3-4 2/27/84
TABLE 4 CORE SPRAY BFNP-3 SIZE (IN. )
12 CONFIGURATION STP/STP STP/LREL LREL/LREL STP/LREL STP/VLV STP/STP STP/LREL STP/LREL STP/VLV TVA WELD IDENT.
DCS-3-13 DSCS-3-7 DSCS-3-8 DSCS-3-9 DCS-3-14 DCS-3-4 DSCS-3-1 DSCS-3-2 DCS-3-5 2/27/84
TABLE 5 REACTOR VATER CLEAN-UP BFNP-3 SIZE (IN. )
CONF IGl.'RATION SOL/VLV VLV/STP STP'/LREL LREL/VLU VLV/STP STP/LREL LREL/STP STP/LREL STP/LREL STP/LREL LREL/STP STP/LREL STP/LREL
/ ~
TVA h'ELD IDENT.
DRh C-3-1A DRh'C-3-1 DSRVC-3-1 DRhC-3-2'Rh'C-3-3 DSRhC-3-IA DSR'nC-3-2 DSRh'C-3-3 DSRv'C-3-4 DSRhC-3-5 DSRh'C-3-6 DSRh'C-3-7 DRhC-3-<
2/27/Sa
TABLE 6 fields Not Treated By IHSI Process Recirculation Loop A KR-3-49 KR-3-4 KR-3-1 Recirculation Loo B
KR-3-26 KR-3-23 KR-3-53 GR-3-63b Core S ra S stem TCS-3-426 TSCS-3-425X TSCS-3-423 TSCS-3-424 TCS-3-422 TCS<<3-421 TSCS-3-420 TCS-3->>9 TSCS-3-418 TCS-3-417
~RRR B stem.
TRHR-3-191 TRHR-3-192 TCS-3-410 TSCS-3-409 TSCS-3-408
>SCS-3-407 TCS-3-406 TCS 3 405 TSCS-3-404 TCS-3-403 TSCS-3-402 TCS-3-401
TABLE 7 IHSI PROCESS COl'tTROL PARAMETERS 1.
Pipe Outer Surface Temperature Within Treatment Zone 2.
Minimum Throughwall Temperature Difference
( AT) 3.
Maximum Temperature of Weld Crown Minimum Width of Zone Heated to AT 500 +
75 C (932 + 135oF) 100oC
- 180oF 275oC (495oF) 600oC
< 1112oF) 1.5 Rt 5.
Minimum Distance from Weld Center to Boundary of T Minimum 6.
Minimum Heating Time 7.
Frequency
- 8. Induction Coil Length 9.
Minimum Water Velocity (in vertical or horizontal flooded pipe) 10.
Minimum Water Velocity (in pipe with air pockets) 15 mm (OR6 in) or t/2 (whichever is larger)
'0.7 t2/
seconds 3 to 4 kHz, 3~Rt Minimum 0.5 m/s (1.64 ft/s) 1.2 m/s (4 ft/s)
R= radius t= wall thickness a =thermal diffusivity
TABLE 8 WELD SAHPl E FOR EXAMINATION AFTER IHSI Recirculation GR-3-1 (28n)
GR-3-2 (28<<)
GR-3-54 (28 r)
GR-3-55 (28>>)
GR-3-57 (28<<)
GR-3-60 (28<<)
GR-3-62 (28n)
KR-3-63 (28<<)
KR-3-3 (28<<)
KR-3-25 (28 )
KR-3-45 (28 r)
KR-3-46 (28<<)
KR-3-47 (28 r)
KR-3-50 (28<<)
Residual Heat Removal GR-3-44 (22<<)
KR-3-15 (22rr)
KR-3-34 (22<<)
KR-3-37 (22>>)
GR-3-13 (12<<)
GR-3-16 (12<<)
GR-3-20 (12>>).
GR-3-36 (12<<)
GR-3-39 (12<<)
GR-3-42 (12n)
KR-3-39 (12<<)
GR-3-14 (12n)
DRHR-3-8 DRHR-3-13 DRHR-3-14 DRHR-3-15 DRHR-3-2'l DRHR>>3-19 (24n)
(24n)
(24<<)
(24
)
(24n)
(ppn)
Core S ra DSRHR-3-2 (24<<)
DSRHR-3-4A (24<<)
DSRHR-3-5 (24<<)
DSRHR-3-5A (24<<)
DSRHR-3>>10 (24<<)
DCS-3-4 (12<<)
DCS-3-14 (12<<)
DCS-3-1
.(12<<)
Reactor Mater Cleanu TCS-3-417
( 10 <<}
TCS-3-422
( 12>>)
DRWC-3-1A (6<<)
PT, best effort UT DSRMC-3-3 (6<<}
DSRWC-3-4 (6<<)
60-3-37 'A'-3-4o 6g-3-~3 6d-3 07 Dk' kW-3-38
Ã-3 +3
<I-3-'o H-h-43 X.A' / ~
k'P-3-Z/
N-3-22 gg-)-20 A'-3-2/
XA'-3;6 dk'-3-5 <
gg-3 $0 S4-3-6o N'-3-4 A'
~1 6'4-3-02
- p-3-46 A'-3-$3
~e-3-5~
64-3-Z3 64-3-$
uV-3-/0 6'g-3-6/
6c'-8-35'p-3-0/
64-3-3S /
gp 3 ~
4'63-33 u'-3-4S
<83-4' g-3-PZ GP-3-c6 i
'm-3-25/
gg-3-22 6'k'- 3-19 gg-S-Zo
<l-3-IP
/fg-3-//
6d-3-Q g-3-I3 4'J-3-/4'63-~7 kE-Z-34 6&8-3/
Q'-3-4+
K8-3-25 5A-s/
&'-3-Q e'-s-46 r
~C-3 ~B..
64-3-8 4p b-r2 K4-3-3
-0 6'4-3-7 j'4-3.s I 60-3-t'2 P-3-Aa
,6Z-~-2Z
&P gf 8-3-sp',
sg gg XA'g-3 52 X gf'-3->0 3
gt-3-4'7 Q' J6 HAEC/k c'5 lATi'04' OOP A 3417 Qg-3-I r
~-3-~
M-3-2
/ /6'o'/'E /
- Revised 3/14/84 BsAP -3
Ok'eg -g -p DSeAg D
'%'-3-Iq ou'ue os'-a-2I Dseh'R -p-g DE/M 3 2-2-~
Zo'seeg-g-rg 88rrR g Sb'd7/0/V
85PA/g-3-7 BSRpfd-3-6 DSfarp-3-5 Dsp/C -3 35R eC - 3.-3 DCS-3-/3 DCS-3-IQ DSCS-3-9 OSCS-3 "8 Oscs-3-7 C~~E S'P<dV - 8 zoos DNA-B-2 DQ'sC-3-I QWC-8-I D$Rfr6'- 3-Z
~DA urd-3-lh 88s'g-g-g OCS-3-S oscS-3-2 DCS"3-f DScs azoic-s-e Cog'py/Y - I Looc
~r~emR e~r~C ~~Em-V~
~
RAP 3 FSuCF 8.
TABLE IA IGSCC SUSCEPTIBILITY RANKING FOR BROWSES FERRY 3
RECIRCULATION SYSTEM PIPE NELDS LOOP A IGSCC Carbon Susceptibility Id.'II Pipe Model GR-3-11 KR-3-16 GR-3-10 GR-3-9 GR-3-1 4 KR-3-17 GR-'-13 Gh-'3 12 GR-3-17 KR-3-1 8 16 8-3-15 KR-3-11 KR-3-12 GR-3-1 8 GR-3-8 GR-3-21 KR-3-21 GR-3-20 GR-3-19 GR-3-24 KR-3-22 GR-3-23 GR-3-22 1.65
- 1. 82 1.75 2.14 1.70
- l. 87
- 1. 80 2.21
- l. 84 2.10
- 1. 91 2.17 1.42 1.63 2.00 1.68 1.62
- l. 83
- l. 81 2.13 1.65
- l. 82 1.67 2.03 0.056 0.063 0.063 0.060 0.056 0.063 0.063 0.060 0.056 0.063 0.063 0.058 0.060 0.060 0.060 0.060 0.056 0.063 0.063 0.060
- 0. 056
- 0. 063
- 0. 063
- 0. 060 Transition To Safe Ena Buttweld Elbow Buttweld Elbow Buttweld Transition To Safe End Buttweld Elbow Buttweld Elbow Buttweld Transition To Saf e End Buttweld Elbow Buttwelo Elbow Buttweld Reducer Buttweld Cross Buttweld Cross Buttweld Cross Buttweld T-Cross Transition To Safe End Buttweld Elbow Buttweld Elbow Buttweld Transition To Safe End Buttweld Elbow Buttweld Elbow Buttweld 10 11 13 14 20 21 23 24 40 41 43 39 31 51 95 60 61 63 64 80 81 83 84
TABLE IA (CONTINUED)
IGSCC SUSCEPTIBILITY RANKING FOP.
BROWNS FERRY 3
RECIRCULATION SYSTEH PIPE WELDS LOOP A IGSCC Carbon Susceptibility hQdM Rl Pipe Yiodel GR-3-25 GR-3-26
'R-3-52 GR-3-51 KR-3-3 KR-3-2 GR-3-3 GR-3-2 GR-3-1 R-3-58 48 57 GR<<3-56 KR-3-47 KR-3-46 GR-3-55
'R-3-54 KR-3-45 GR-3-53 KR-3-14 KR-3-13 KR-3-19 KR-3-20 KR-3-15 KR-3-1 KR-3-4 KR-,3-49 I
1.21
. 1.20 1.17 1,09 1.36 1.24 1.23 1.04 1.03 1.24 1.23 1.00 1.19 1.20 1.34 1.48 1.30 1.60 1.24 1.12 1.60 1.62 1.53 0;048 0.050 0.050 0.048 0.055 0.055 0.053 0.064
- 0. 064
- 0. 063
- 0. 064 0.064
- 0. 062
- 0. 062 0.057 0.064 0.064 0.064 0.064 0.060 0.060 0.060 0.060
- 0. 070
- 0. 060
- 0. 060
- 0. 050 4
(SHT) 4 (SHT) 4 (SHT) 4 (SHT)
Buttweld Valve Buttweld Valve Buttweld Valve Buttweld Valve Buttweld Tee Buttweld Elbow Buttweld Valve Buttweld Valve Buttweld Pump Buttweld Pump-Elbow Buttweld Elbow Buttweld Valve Buttweld Valve-Elbow Buttweld Elbow Buttweld Tee Buttweld Tee Buttweld Elbow Buttweld Elbo~
Transition Buttweld Buttweld Buttweld Buttweld Buttweld Cap Butt Weldolet Butt Weldolet Weld Neck Flange 257 256 253 252 117 121 123 124 158 157 155 154 153 151 149 144 143 141 140 15 19,29 59,69 78, 89'59 15 159 119 154
~SRI Values Not Calculated For Solution Heat Treated Welds 0
Cl
TABLE IB IGSCC SUSCEPTIBILITY RANKING FOR BROWNS FERRY 3 RECIRCULATION SYSTEM PIPE WELDS LOOP B IGSCC Carbon Susceptibility il 'L XII Pipe Model GR-3"37 KR-3-3 8 GR-3-36 GR-3-35 GR-3-40 A-3-3 9
, GR-3-39 GR-3-38 R-3-47 43 R-3-46 GR-3-45 GR-3-'50 KR-3-44 GR-3-49 GR-3-48 GR-3-43 KR-3-40 GR-3"42 GR-3-41 KR-3-33 1.56 1.72 1.72
'.09 1.67
- 1. 88
- 1. 81 2.'22 1.50 1.65
- l. 81 2.15 1.61 1.78 1.70
. 2.06
- 1. 88 2.12 1.92 2.17 1.35 0.056 0.063 0.063 0.060 0.056 0.063 0.063 0.060 0.056 0.063 0.063 0.060
- 0. 056 0.063 0.063 0.060 0.056 0.063 0.063
- 0. 058
- 0. 060 Transition To Safe End Buttweld Elbow Buttweld Elbow Buttwelo Transition to Safe End Buttweld Elbow Buttweld Elbow Buttweld Transition to Safe End Buttweld Elbow Buttweld Elbow Buttweld Transition To Safe End Buttweld Elbow Buttwelo Elbow Buttweld Transition To Safe End Buttweld Elbow Buttweld Elbow Buttweld Reducer Buttweld Cross 170 171 173 174 180 181 183 184 220 221 223 224 240 241 243 244 200 201 203 204 199
TABLE IB (CONTINUED)
IGSCC SUSCEPTIBILITY RANKING FOR BROWNS FERRY 3
RECIRCULATION SYSTEM PIPE WELDS LOOP B Carbon Keld.~
&RE IGSCC SusceptibilitY Pipe Model GR-3-44 KR-3-3 4 GR-3-3 4 KR-3-25 KR-3-24 GR-3-29 GR-3-28 GR-3-27 GR 3-64 KR-3-51 GR-3-61 GR-3-60 KR-3-50 GR-3-59 1.48
- l. 95 1.65 1.31 1.20 1.19 1.02 1.01 1.22 1.25 1.01 1.01 1.20 0.91 1.21 1.32 1.07 0.060'0.
060 0.063 0.064
- 0. 064
- 0. 053
- 0. 055
'0.055 0.063 0.064 0.064
- 0. 062
- 0. 062 0.064
- 0. 064
- 0. 064
- 0. 064 1
1 1
1 1
1 2
2 1
1 2
2 1
3 1
1 2
Buttweld Cross Buttweld Cross Buttweld Tee-Cross Buttweld Tee Buttwelo Elbow Buttweld Elbow-Valve Buttwelo Valve Buttweld Pump Buttweld Pump-Elbow Buttweld Elbow Buttweld Valve Buttweld Elbow Buttweld Elbow Buttwelo Buttweld Elbow Buttweld Elbow Transition 191 211 265 287 291 293 294 326 325 323 322 321 318 314 313 311 310 KR-3-42 KR-3-41 KR-3-35 KR-3-36 KR-3-37 KR-3-26 KR-3-23 KR-3-53 GR-3-63B 0.060 0.060 0.060 0.060 1.12 0.070 1.60 0.060 1.58 0.060 1.54 0.050 1.54 0.050 4
(SHT) 4 (SHT) 4 (SHT) 4 (SHT)
Buttweld Buttweld Buttweld Buttweld Buttweld Cap Butt Weldolet Butt Weldolet Weld Neck Flange Weld Neck Flange 239,249 219,229 179,189 175 175 289 329 322'22
- SRI Values Not Calculated For Solution Heat Treated Welds 10
TABLE IC IGSCC SUSCEPTIBILITY RANKING FOR BRONNS FERRY 3
RECIRCULATION SYSTEH SAFE KK)S Held No.
Carbon IGSCC Susceptibility Inlet Nozzle:*
N2Ag N2B
- N2C, N2D N2E
- N2F, N2G
- N2H, N2J N2K Outlet Nozzle:
i (I) 2 (I) 2 (0)
(I) 2 (I) 2 (0) 1.25 1.42
- 0. 97 1.25 1.42 0.97
- 0. 060, 0.060
- 0. 060
- 0. 070
- 0. 070 0.070 N1B 3 (I) t N1A 3 (I)
- See Figures 10 and 11
- I = Inside 0 = Outside 1.33 1.33 0.09 0.08
Cl
TABLE IIA IGSCC SUSCEPTIBILITY RANKING FOR BROWNS FERRY 3 RESIDUAL HEAT REMOVAL SYSTEM PIPE WELDS
'LOOP A IGSCC Carbon Susceptibility LH Pipe Model DRHR
<<3-1 8 DRHR 17 DRHR 16 DSRHR 7 DSRHR 6 DRHR 15 DSRHR SA DRHR 14 RHR
'-13 SRHR 5 DRHR 19 DSRHR 8 DSRHR 9 DRHR 22 DRHR 21 DSRHR 3:0 DSRHR 11 DRHR 23 TRHR;.191 TRHR
'-3-192 TRHR
" 193
- l. 90 1.30 1.51 1;51 1.04 1.09 1.47
- 1. 47 1.45 1.45 1.74 1.84
- 1. 90
- 0. 060 0.060 0.060 0;060 0.060 0.060 0.060 0.060 0.060 0.060 0.060 0.060 0.060 1.35 1.33 1.33 0.060 0.060 0.060 1.04 0.060 1.79 0.060 1.65 0.060 1.79 0.060 1.79 0.060 Buttweld Tee Buttweld Valve Buttweld Valve Buttweld Buttweld Buttweld Valve Buttweld Elbow Buttweld Elbow Buttweld Elbow Buttweld Buttweld Buttweld Buttweld Valve Buttweld Valve Buttweld Elbow Buttweld Elbow Buttweld Elbow Buttweld Elbow Buttwelo Elbow Buttweld 113 110 109 108 107 3,06 106 105 105 169 168 167 166 165 165 165 164 163 163 160 12
TABLE IIB IGSCC SUSCEPTIBILITY RANKING FOR BROWNS FERRY 3
RESIDUAL HEAT REHOVAL SYSTEH PIPE FIELDS LOOP B Carbon H~J2u IGSCC Susceptibility Pipe lhodel DRHR 9 DRHR 8 DRHR 7 DRHR 4A 5SRHR 4'SRHR 2 DSRHR 1 DRHR,4 DRHR 3B
- 1. 82 0'. 060 1.31 0.060 1.56 0.060 1.56 0.060 1.61 0.060 1.15 0.060 1.27 0.060 1.58 0.060 1.58 0.060 1.48 0.060 1.48 0.060 0.97 0.060 Buttweld Tee Buttweld Valve Buttweld Valve Buttweld Elbow Buttwelo Elbow Buttweld Buttweld Buttweld Valve-Elbow Buttweld Elbow Buttweld Elbow Buttweld Elbow Buttweld 282 280 279 279 278 277 276 275 275 274 274 270
- Assumed Carbon Value.
13
0 TABLE III IGSCC SUSCEPTIBILITY RANKING FOR BROWNS FERRY UNIT 3 REACTOR HATER CLEANUP SYSTEM PIPE WELDS Carbon M~nk'GSCC Susceptibility DRWC-3-1A DRWC-3-1 TRWC-3-1 TRWC-3-2xl
'TRWC-3'-4 SRWC-3-1A SRWC-3-2 DSRWC-3-3 DSRWC-3-4 DSRWC-3-5 DSRWC-3-6 DSRWC-3-7 DRWC-3-4 R
DRWC-1-4A 1.12 1.14
- 1. 23 1.24 1.15 1.22 1.21 1.20 1.1 8 1.30 1.32 1.30 1.24 1.61
- 0. 060
- 0. 060 0.060
- 0. 060 0.060 0.060 0.060 0.060 0.060 0.060 0.060 0.060 0.060 0.060 Buttweld Valve Buttweld Valve Buttweld Elbow Buttweld Valve Buttweld Valve Buttweld Elbow Buttweld Elbow Buttweld Elbow Buttweld Elbow Buttweld Elbow Buttweld Elbow Buttweld Elbow But tweld Buttweld
.*Assumed Carbon Value.
14
<r r
4 P
gl r i+<
/
g 4
~
w
~
.