ML20112J042
| ML20112J042 | |
| Person / Time | |
|---|---|
| Site: | South Texas  |
| Issue date: | 06/30/1996 |
| From: | WESTINGHOUSE ELECTRIC COMPANY, DIV OF CBS CORP. |
| To: | |
| Shared Package | |
| ML19344C988 | List: |
| References | |
| WCAP-14654, NUDOCS 9606190152 | |
| Download: ML20112J042 (100) | |
Text
.
, WESTINGHOUSE NON-PROPRIETARY (CLASS 3)
WCAP-14654 SG-96-05-013 i
SPECIFIC APPLICATION OF LASER WELDED SLEEVES FOR THE SOUTH TEXAS POWER PLANT STEAM GENERATORS i
June 1996 ;
WESTINGHOUSE ELECTRIC CORPORATION NUCLEAR SERVICES DIVISION P.O. BOX 158 ,
MADISON, PENNSYLVANIA 15663 C1996 WESTINGHOUSE ELECTRIC CORPORATION ALL RIGHTS RESERVED WPF2341:49\060496 9606190152 960614 PDR ADOCK 05000498 P PDR
(
i ABSTRACT !
Under Plant Technical Specification requirements, steam generator tubes are periodically l inspected for degradation using non-destructive examination techniques. If established inspection criteria for tube integrity are exceeded, the tube must be removed from service by plugging, or the tube must be brought back into compliance with the Technical Specification l
Criteria. Tube sleeving is one technique used to return the tube to an operable condition, j The purpose of this evaluation is to establish the applicability of a generic laser welding l sleeving analysis for 3/4 inch diameter tube feedring-type and Westinghouse preheater steam ;
generators (WCAP-13698, Rev. 2) to the South Texas steam generators. (Note: The terms !
" South Texas", " South Texas Project", STP, Houston Lighting and Power and HLP will be used interchangeably in this document.) l The sleeve design, mechanical testing, stress corrosion resistance testing and evaluations,
- in'stallation processes and nondestructive examination discussed in the generic report apply. l directly to South. Texas. Information developed subsequent to the publication of WCAP- !
' 13698, Rev. 2 and the few exceptions to the generic WCAP are noted. j t
Based on the combined results of this evaluation and the generic evaluation (WCAP-13698, j Rev. 2), the laser welded sleeves are concluded to meet applicable ASME Boiler and Pressure -
Vessel Code and regulatory requirements for South Texas. The allowable plugging margin (
for sleeve degradation is 42% !
i i
l e
i l
4 5
-WPF2341:49\060496 i
i TABLE OF CONTENTS Page No.
Abstract . i .
i 1.0 - Introduction 1-1
- 2.0 Sleeve Design and Description 2-1 !
' 3.0 ' Analytical Verification 3-1 3.1 Structural Analysis 3-1 3.1.1 Geometry 3 3.1.2 Materials 3-2 2
3.1.3 Applicable Loading Conditions and Structural Assessments 3-2 3.1.4 Minimum Required Sleeve Wall Thickness 3-4 3.1.5 Determination of Plugging Limits 3-6 3.1.6 Sleeve / Tube Contact Pressures 3-7 3.2 Sleeved Tube Rc ative Flow Induced Vibration Assessment 3-24 i 3.3 ' Hydraulic Equivalency 3-27 4.0 Mechanical Tests 4-1 ,
4.1 Establishment of STP ETS Lower Joint Process Based 4-2
- on Previous Qualifications -
4.1.1 ETS Lower Joint for the Model E2 SG Factory Roll 4-2 Expanded Tube Joints of STP1 ,
4.1.2 STP1 ETS Leak Test Equivalence 4-4 4.1.3 Justification for Using the Model El ETS Lower Joint 4-5 l Pullout Resistance for STP1 -
4.1.4' Justification for Using M. Yankee ETS Leakage Resistance 4-6 Results as Bounding Results for STP1 '
4.1.5 Conclusions 4-7 4.2 ETS Lower Joint for the Factory Hydraulic Expanded Tube 4-13 l Joints of STP2 :
4.3 Details of Previous 3/4 Inch Tube Sleeve Lower Joint Process 4-15 Qualifications 4.3.1 Elevated Alloy 690 Sleeve-FSG (Maine Yankee) 4-16 -
- 4.3.2 Elevated Alloy 690 Sleeve-PSGs (Doel 4) 4-18 5.0 Stress Corrosion Testing of Laser Welded Sleeve Joints 5-1 ;
, 5.1 LWS Process and SG Design Variables 5-2
- 5.2 - Residual Stresses vs. Stress Relief Temperature in LWS 5-3 Sleeve Repairs
' 5.3 Corrosion Test Description - .
5-3 5.4 Corrosion Resistance of Free-Span Laser Weld-Repaired Tubes 5-4
- As-Welded Condition i
~ 5.5 . Corrosion Resistance of Free-Span Laser Weld Repaired Tubes 5-5 l
- with Post Weld Stress Relief ' !
- )
WPF2341:49\060696 ii .
i L
TABLE OF CONTENTS (Continued)
Page No. ;
5.6 Corrosion Resistance of Free-Span Laser Weld-Repaired Tubes-with 5-5 4
Post Weld Stress Relief and Conditions of Axial Load During Test 5.7 Estimated Sleeve Performance at South Texas 5-7 5.8 Outer Diameter Surface Condition 5-9 6.0 Installation Process Description 6-1 7.0 NDE Inspectability 7-1 7.1 Inspection Plan Logic 7-1 ,
4 7.2 General Process Overview of Ultrasonic Examination 7-2 7.2.1 Principle of Operation and Data Processing of 7-2 Ultrasonic Examination 7.2.2 Laser Weld Test Sample Results 7-3 7.2.3 Ultrasonic Inspection Equipment and Tooling 7-3 7.3 Eddy Current Inspection 7-4 7.3.1 Cecco-5/ Bobbin Principles of Operation 7-4 7.4 Alternate Post Installation Acceptance Criteria 7-5 4 7.5 Inservice Inspection Plan for Sleeved Tubes 7-5 1
8.0 References 8-1 ,
j b
3 WPF2341:49\060596 iii ,
LIST OF TABLES 3.1-1 Materials Used in Reference 1 for the Structural Evaluation of Laser 3-12 Welded Sleeves 3.1-2 Normal Operation Parameters, Reference 3, Upgraded Fuel,10*F 3-12 T-hot Reduction,10% SGTP 3.1-3 Pressure Loads for Design, Faulted, Test, and Emergency Conditions 3-13 Considered in Reference 1 3.1-4 Pressure Drop Loads for Design, Faulted, Test, and Emergency Conditions 3-14 Considered in Generic Evaluation of Reference 1 versus South Texas Values 3.1-5 Summary of Maximum Primary Stress Intensity Elevated Tubesheet Sleeve 3-15 3.1-6 Summary of Generic Structural Evaluation of Normal, Upset, and Test 3-16 Loads, Reference 1.
3.1-7 Limiting Upset Transients Pressure Loads for South Texas LWS 3-17 Installation Used in Minimum Sleeve Wall Calculations (Data from References 2,3) 3.1-8 Strength Propertica (ksi) - Sleeve Material Thermally Treated Alloy 690, 3-18 Reference 5 3.1-9 Summary of Minimum Wall Thickness Calculations for Laser Welded 3-18 Tubesheet Sleeves For Use in 3/4 inch OD Tubes for South Texas Model E2 S/G 3.1-10 Summary of Recommended Plugging Limit for Laser Welded Tubesheet 3-18 Sleeves For Use in 3/4 inch OD Tubes for South Texas Model E2 S/G 3.1-11 Comparison of South Texas and Doel 4 Maximum Tubesheet Radial 3-19 Displacement (inches) 3.1 12 Minimum Contact Pressures Between Sleeve and Tube 3-19 3.2-1 Relative Flow Induced Vibration Evaluation Results for a 22 Inch 3-26 Long Elevated Tubesheet Sleeve in Model E2 SGs With Various Boundary Conditions (Reference 1) 4.1-1 Applicability of Doel 4 (Model E1) LWS* Lower Joint Process to 4-8 STP 1 & 2 Normal Operation 4.1-2 Applicability of Doel 4 (Model E1) LW Elevated Tubesheet Sleeve
- 4-10 Lower Joint Process to STP 1 & STP 2 Faulted Condition 4.1-3 Maine Yankee Toll-Last ETS Lower Joint as Basis for STP 1 Roll-Last 4-12 ETS Lower Joint 4.2-1 Maine Yankee Toll-Last ETS Lower Joint as Basis for STP 2 Roll-Last 4-14 ETS Lower Joint 4.3 1 Verification Test Results - Mechanical Interference Fit Lower Joint 4-23 Elevated Tubesheet Sleeve Alloy 690 Sleeve for 3/4 Inch Tube (Feedring SG) 4.3-2 Verification Test Results - Mechanical Interference Fit Lower Joint 4-26 Elevated Tubesheet Sleeve Alloy 690 Sleeve for 3/4 Inch Tube (Preheater SG) 5-1 Summary ofImpact of Laser Welded Sleeve Operations on Stresses 5-10 5-2 Far-field Stress as a Function of Stress Relief Temperature 5-11 5-3 Results of 750'F Doped Steam Tests for Nd:YAG Laser 5-12 Weld Repaired Mockups 5-4 Doped Steam Corrosion Test Results for Tube-Sleeve Mockups - Tested 5-13 Without Axial Load 5-5 Summary of Fabrication Parameters, Temperatures, Stresses and 5-14 Corrosion Test Results for 3/4 inch Sleeve Mockups - Tested with Applied Axial Load WPF2341:49\060496 iv
i r
LIST OF FIGURES 3.1-1 Finite Element Model of Model E Channel Head /Tubesheet/Shell 3-20 l 3.1-2 Contact Pressures for Normal Conditions with an Intact Tube 3-21 3.1-3 Contact Pressures for Normal Conditions with a Separated Tube 3-22 !
3.1-4 Contact Pressures for Faulted Conditions with Intact or Separated Tube 3-23 !
5-1 - Accelerated Corrosion Test Specimen for Welded Joint Configuration 5-15
- 5-2 Test Stand for Fabrication of LWS Mockups Under Locked Tube Conditions 5-16
. 5-3 LWS Mockup Corrosion Sample Test Assembly 5-17 5-4 IGSCC in Alloy 600 Tube of YAG Laser Welded Sleeve Joint After 109 5-18
- ' Hours in 750*F Steam Accelerated Corrosion Test ;
7-1 Ultrasonic Inspection of Welded Sleeve Joint 7-6 j 7-2 Typical Digitized UT Wavefona .
7-7
. 7-3 A,B,C, and Combined C-Scan Display for Weld in UT Calibration Standard 7-8 7-4 UT Calibration Standard 7-9 7-5 Cecco-5 Sleeve Calibration Standard 7-10 7-6 Strip Chart Display for Cecco/ Bobbin Data 7-11
. 7-7 Response of Cecco-5 Probe to 60% OD Axial Notch in Parent Tube 7-12 !
Located at Expansion Transition
- 7-8 Response of Cecco-5 Probe to 60% OD Circumferential Notch in Parent 7-13 ,
Tube Located at Expansion Transition ,
i
! i l
i j
1 1
, i 1
I i
WPF2341:49\060596 y i
t h
4
1.0 INTRODUCTION
This report documents the results of an analysis to evaluate the applicability of tlw generic laser welded sleeving analysis for 3/4 inch diameter tube feedring-type and Westinghouse preheater steam generators (WCAP-13608, Rev. 2), Reference 1, to the South Texas Units 1 and 2 steam generators. In performing the generic analysis, transient loads are used that umbrella the steam generators to be sleeved. Included in the generic analysis are calculations to determine minimum wall thickness requirements for the sleeves. These calculations are a function of plant operating parameters, which vary from plant to plant, and which can change with the implementation of operating or system modifications. The purpose of this evaluation then, is to compare the current set of transient and operating parameters for South Texas to those used in the generic analysis, with the intent of confirming that the generic analysis provides a bounding analysis for South Texas, and to also remove any conservatism in the generic analysis for minimum wall thickness,if possible.
The results of this analysis are based on transient data supplied by Reference 2.
In establishing the structural adequacy of the laser welded sleeves in the generic analysis, criteria were evaluated for primary stress limits, maximum range of stress intensity and fatigue, and minimum wall thickness requirements. The load conditions applicable to each ,
of these areas are reviewed in this analysis to establish the applicability of the generic analysis. In general, the discussions to follow provide only a brief overview of each area.
More in-depth discussions are contained in Reference 1 for the generic analysis.
1 T
i J
?
b WPF23411:4&O60496 y
I 2.0 SLEEVE DESIGN AND DESCRIPTION The sleeve which is elevated in the tubesheet (ETS), as shown in Reference 1, including the design documentation, is directly applicable to South Texas. The other sleeve design documentation in Reference 1, i.e., for the full length tubesheet sleeve (FLTS) and the tube support sleeve (TSS) was also evaluated and determined to be applicable to South Texas.
S I
2 4
e 1
4 i
J 4
i l
f L
r 4 WPF23412:49060496 g.1
l 3.0 ANALYTICAL VERIFICATION j 3.1~ Structural Analysis - '
l This section provides the structural analysis basis for installation of Alloy 690 laser welded -
sleeves (LWS) in 3/4 inch nominal outside diameter Alloy 600 tubes with [ l*## !
nominal wall . thickness in the. South Texas Model E2 steam generators. The generic
- structural evaluation oflaser welded sleeves for 3/4 inch tubes with [ ]'## walls is ;
documented in hference 1 and essentially covers the South Texas application, as stated j
. above in Section 2. The major structural topics covered are: the ASME Code evaluation, the .
1 sleeve / tube contact pressure evaluation (for the elevated tubesheet sleeve lower joint), and the minimum sleeve wall thickness requirements which define the associated plugging limits.
.While the bulk of the verification is based on data in hference 1, the sleeve / tube contact ,
- pressure assessment is based on the evaluation performed for equivalentlaser welded sleeves !
' installed in the Doel 4 Model El steam generator, as documented in hference 4.
- 3.1.1 Geometry
F The laser welded sleeve and tube geometries for use in South Texas Model E2 steam l generators are the same as the sleeve and tube geometries considered in the generic LWS l structural evaluation in hference 1, namely, nominal [ j l'## sleeves installed in nominal 0.750 inch OD x 0.043 inch wall thickness tubes.
The nominal width (interfacial axial extent) of the laser weld joining the tube and sleeve for f L alljoints is [ ]'##. However, qualification tests for the weld process show that the -
welds may be as small as [ ] *##. Thus, in performing the analysis in Reference 1, a weld width of[ l*## was conservatively assumed. Therefore, the stress and fatigue results in hference 1 are for the limiting weld width geometry of [ l'##. Thus, with
,. respect to sleeve, tube, and weld geometry, the results and conclusions of Reference 1 apply directly to the South Texas LWS installation. :
- . l l
As stated above, verification of the sleeve / tube contact pressure at the ETS lower joint is !
I based on the evaluation performed for the installation of equivalent laser welded sleeving in ;
the Doel 4 Model El steam generator using a finite element model of the tubesheet, i channelhead, cylinder, and external support ring. The South Texas Model E2 tubesheet, l
- channel head, and cylinder are essentially the same as the Doel 4 counterparts. However, l
. the external support rings, which have the same outer diameters, have different thicknesses ;
and locations. The 9.75 inch thick South Texas Model E2 support ring is attached to the ,
channel head near the tubesheet, and the 14.5 inch thick Doel 4 Model El support ring is !
attached to the tubesheet. These differences in external support ring geometry are ;
considered in Section 3.1.6. I e
i WPP23413:4&O60496 - 3-1
i j i 4 3.1.2 Materials i
Table 3.1-1 lists the materials considered in the generic structural evaluation of the laser welded sleeves in Reference 1. These same materials apply to the South Texas installation except for the channel head,' which is SA-533, Grade B, Class 2 for South Texas compared to l SA-216 Grade WCC for the generic analysis in References 1 and 4. The differences in l channel head materials have virtually no effect on the sleeve / tube interactions. Therefore, !
with respect to sleeve, tube, tubesheet, channel head, and cylinder materials, the results from
$ Reference 1 are directly applicable to South Texas. ;
The material of construction for the 3/4 inch tubes in the South Texas Model E2 steam .
generators is nickel b' ased Alloy 600 in the mill annealed (MA) condition with a 35 ksi minimum yield. All of the sleeves are thermally treated (TT) Alloy 690, which has a 40 ksi minimum yield. Since the laser weld process joins the sleeve and tube by direct fusion, l
'.without using a filler metal, the weld section was conservatively assumed (in Reference 1) j
' to have at least the lower strength properties of the weaker MA Alloy 600 material. The l fatigue curves used in the analysis of the sleeve, tube, and laser weld are the ASME Code ;
- fatigue design curves for nickel-chromium-iron (Alloys 600 and 690) given in Figures I-9.2.1 !
t and I-9.2.2 of Appendix I of Reference 5. j
! 3.1.3 Applicable Loading Conditions and Structural Assessments !
The umbrella loadinr, conditions, used in the generic analysis of Reference 1, are defined in Reference 2 and include transient loads from the applicable design specifications for ABB-CE j feedring steam generators and for Westinghouse Model D3, D4, D5 and E1/E2 steam !
generators. Therefore, the transient loadings in References 1 and 2 cover the installation of LWS in the South Texas Model E2 steam generators. In addition, those transients that initiate at 100% power conditions, assume the normal operating parameters listed in Table 3.1-2 for South Texas as given in Reference 3. l Note that the seismic loads (the OBE and the SSE) result in negligible stresses in the tube and sleeve compared to the pressure and temperature loads. Also, the fatigue usage due to !
J the OBE is insignificant, as shown in Reference 1. Therefore, the specified seismic loads are j
- negligible and are not explicitly considered in the following discussions. 3
.1 3.1.3.1 Pressure Loads Table 3.1-3 lists the pressure loads, specified in Reference 3 for design, faulted, test, and j
- emergency conditions and considered in the generic LWS structural analysis reported in '
. Reference 1. Pressure stresses in the sleeve were evaluated for each classification at the '
limiting AP values shown in Table 3.1-3. The latest parameters for South Texas in Table 3.1-2, give the full power steam pressure (P,) to be 975 psia, compared to a nominal initial design ;
i
?
WPF2341-3:41WD60496 3-2 :
~
i
-- _ . _ . . - . . - . - ~
5
value of 1100 psia. Therefore, those @ values in Table 3.1-3 that are > 0 and that initiate relative to 100% full power conditions, should be conservatively increased by the amount of 1100 - 975 or 125 psi to account for the reduced steam pressure. Table 3.1-4 shows a comparison of the pressure drop loads for the generic case, Reference 1, versus the values for South Texas which have been increased by 125 psig where applicable. Note that only the pressure drop values in the four heavily outlined boxes in Table 3.1-4 have increased by 125 psi for South Texas. There are no changes to the limiting @ for design or test conditions or for those transients which initiate at zero load (hot standby) conditions. As also shown in Table 3.1-4, none of these increased values exceeds the existing limiting & in each classification. That is, the [ ]*'# value continues to limit for faulted conditions, and the[ ]* value continues to limit for emergency conditions. Therefore, the results for the pressure stress evaluations in Reference 1 remain valid and applicable for South Texas, and it may be concluded that the ASME Code pressure stress limits are satisfied.
The umbrella loads for the primary stress intensity evaluation have been identified in Table 3.1-3. From Reference 6, the largest magnitudes of the ratio of Calculated Stress Intensity / Allowable Stress Intensity are [
] * *#. The analysis results show the primary stress intensities for the laser welded ETS sleeved tube assembly to satisfy the allowable ASME Code limits. A summary of the limiting primary stress intensity conditions are provided in Table 3.1-5 with the tube intact.
3.1.3.2 Normal, Upset, and Test Loads )
The maximum range of primary plus secondary stresses and the cumulative fatigue usage are calculated and evaluated in Reference 1 for the normal, upset, and test loads specified in Reference 3. While these same transients apply to South Texas, the lower secondary side steam pressure (Table 3.1-2) increases some of the stress ranges somewhat. Table 3.1-6, taken directly from Reference 1, lists the limiting ratios of calculated to allowable stress '
intensities for the sleeve, tube, and weld and the highest limiting cumulative fatigue usage factor conservatively calculated for the worst case geometry conditions of a totally separated tube fixed at the first tube support plate. In addition, as stated previously, the maximum weld stresses were conservatively calculated assuming the limiting minimum sleeve / tube weld width of [ ]* (compared to the average expected weld width of [
3..c ),
Since the maximum calculated cumulative fatigue usage factor is very small, only [ ]*'"
compared to the limit of one, the small increase in some of the pressure drop loads (due to the lower steam pressure), will not significantly increase the cumulative usage. Therefore, the ASME Code fatigue limit for cyclic stresses is satisfied.
1 l
l l
WPF2341-3:49M60496 3-3 )
1
The same conclusion may be made regarding the range of primary plus secondary stresses.
Much of the stress ranges are due to thermal stresses, which are somewhat smaller since the primary and secondary fluids are at slightly lower initial full power normal operation temperatures, as listed in Table 3.1-2. Therefore, the slight increase in pressure stress range does not limit. The limiting stress intensity ratios in Table 3.1-6 [ ]"'
, indicate that sufficient margins remain to accommodate the small increase in stress range associated with the lower secondary pressures. Therefore, the ASME Code limit for primary plus secondary stresses is satisfied.
3.1.3.3 Upset Loads Used in Minimum Sleeve Wall Calculations Table 3.1-7 gives the primary and secondary pressures for the significant upset conditions specified in Reference 3. The secondary pressures are for the latest parameters in Table 3.1-2.[
L ju 3.1.4 Minimum Required Sleeve Wall Thickness In establishing the safe limiting condition of a sleeve in terms ofits remaining wall thickner. ,
the effects of loadings during both the normal operation and the postulated accident conditions must be evaluated. The applicable stress criteria are in terms of allowables for the primary membrane and membrane plus bending stress intensities. Hence, only the
, primary loads (those necessary for equilibrium) need be considered. Since primary bending stresses are negligible in the sleeve and tube, the pressure stress equation NB-3324.1 of the Code, Reference 5,is used to calculate t . That is, i '
AP, x R,
~
P, - 0.5 (P, + P,) ;
where: Ri = maximum inner radius of sleeve = [ ]"#
Pi = internal pressure = Pp = primary pressure (psig), !
P, = external pressure = Ps = secondary pressure (psig).
APi = Pi - P, = Pp - Ps>
l P, = allowable maximum value of primary membrane stress intensity (psi). '
wpm 413:49060496 3-4 l
3.1.4.1 Normal Operation
- From Table 3.1-2, the load parameters for normal steady state operation resulting in the maximum primary to secondary pressure drop are:
[
l us 3.1.4.2 Maximum Upset Condition From Table 3.1-7, the load parameters for the maximum upset condition occur for the loss ofload event (at time 10 seconds) and are:
'[ ,.
-us 3.1.4.3 Accident Conditions The dominant loading for LOCA and SSE loads occurs [
ju
. WPF2341-3:49M0496 3-5
0
[
-u, ,
A summary of the normal, upset and accident minimum required wall thicknesses is given in Table 3.1-9.
3.1.5 Determination of Plugging Limits The minimum acceptable wall thickness and other recommended practices in Regulatory Guide 1.121, Reference 7, are used to determine a plugging limit for the sleeve. The Regulatory Guide was written to provide guidance for the determination of a plugging limit for steam generator tubes undergoing localized tube wall loss and can be conservatively applied to sleeves. Tubes with sleeves which are determined to have indications of degradation of the sleeve in excess of the plugging limit, would have to be repaired or removed from service.
As recommended in paragraph C.2.b of the Regulatory Guide, an additional thickness degradation allowance must be added to the minimum acceptable tube wall thickness to establish the operational sleeve thickness acceptable for continued service. Paragraph C.3.f of the Regulatory Guide specifies that the basis used in setting the operational degradation allowance include the method and data used in predicting the continuing degradation and consideration of NDE measurement errors and other significant eddy current testing parameters. An NDE measurement uncertainty value of [ 1*, Reference 1, of the sleeve wall thickness is applied for use in the determination of the operational sleeve thickness acceptable for continued service and thus determination of the plugging limit.
Paragraph C.3.f of the Regulatory Guide specifies that the bases used in setting the operational degradation analysis include the method and data used in predicting the WPF2341-3:#060496 - 3-6
. _ - . ._ . - - . - . - . . - - _ . . . _ . .._.- ~ , - - - - - . - - . -- -..-
o continuing degradation. To develop a value for continuing degradation, sleeve experience l must be reviewed. To date, no degradation has been detected on Westinghouse designed mechanical joint sleeves and no sleeved tube has been removed from service due to ,
degradation of any portion of the sleeve. This result can be attributed to the changes in the sleeve material relative to the tube and the lower heat flux due to the double wall in the sleeved region. Sleeves installed with the laser weld jcint are expected to experience the same performance. As a conservative measure, the conventional practice of applying a value of [ ]*^* of the sleeve wall, Reference 1, applied as an allowance for continued degradation,is used in this evaluation. :
From Section 3.4.2 and Table 3.1-9, the structural limiting minimum sleeve wall thickness ,
isL ;
i
) a.c.e A summary of the resulting plugging limit, as determined by Regulatory Guide 1.121 recommendations, is given in Table 3.1-10.
3.1.6 Sleeve / Tube Contact Pressures Inside the tubesheet, it is important to maintain adequate contact pressure at the hard rolled sleeve / tube interfaces to prevent pullout and leakage in the elevated tubesheet sleeve configuration. Loads are imposed on these interfaces as a result of tubesheet rotations under
, ~ pressure and temperature conditions. The effect of tubesheet rotations on the contact )
l pressures at the sleeve / tube interface was determined in Reference 4 for the Doel 4 Model El l
tubesheet. In the limiting load cases considered in Reference 4, (namely, normal operation !
! and the steam line break faulted condition), the net effect of tubesheet rotations, thermal !
expansions, and fluid pressures is to increase the contact pressure between the sleeve and tube which enhances leakage prevention.
The primary end of the Doel 4 Model El steam generator geometry and materials (tubesheet, channelhead, and cylinder) are essentially the same as the South Texas Model E2 steam i generator primary end geometry except for the thickness and location of the external support rings and the channel head materials (see 3.1.2). Since the South Texas normal and SLB loads are the same as the Doel 4 normal and SLB loads, then the effects of thermal :
expansions and fluid pressures should be the same and the results from Reference 4 should also apply to South Texas, if the differences in location and thickness of the external support
?
9 i
WPF23413:49960496 3-7 i E
i e a,,-a- >' - , .,. ~ .,-ce-- , ,-n, . -<m a , . , . - - - - - . , ,n.-- -, , -, - , -
, ,,n-- , ,+- . -
I~
rings do not signi6cantly influence the tubesheet rotational displacements at the sleeve / tube contact locations.
The major factor in determining the tubesheet rotational effect is the dilation of the tube hole diameter in the tubesheet (near the secondary surface) due to tensile hoop and radial strains.
In axisymmetric analyses, such as were performed in Reference 4, the hoop strain is the calculated radial displacement divided by the radius, and the radial strain is the derivative of the radial displacement with respect to the radius. The derivatives are estimated by fitting a fifth-order polynomial to the calculated radial displacements of the b Aesheet versus the radius. The resulting radial and hoop strains are then used to find the hole dilations, which, in turn, are used to estimate the dianges in the sleeve / tube contact pressures. Based on paavious experience, the radial and hoop strains are nearly equal. Thus, to assess the
. applicability of the Reference 4 results to South Texas, it is only necessary to compare the calculated radial displacements (i.e., the hoop strains) of the South Texas Model E2 tubesheet with the calculated radial displacements of the Doel 4 Model El tubesheet. This was done using two axisymmetric finite element shell models of the tubesbeet, channelhead, cylinder, and external support ring. The support ring in the shell model of the South Texas Model E2 SG is 9.75 inch thick and is attached to the top of the channelhead just below the tubesheet. The support ring in the shell model of the Doel 4 Model El SG is 14.5 inch thick and is attached to the tubesheet outer rim. Tubesheet displaceme its were calculated for both normal and limiting SLB faulted conditions.
Table 3.1-11 shows the resulting maximum radial displacements of the tubesheet at the location of the hard rolled sleeve / tube interface for normal operation and the maximum faulted SLB conditions. The combined results indicate that the South Texas Model E2 tubesheet maximum radial displacements are slightly less than the Doel 4 Model El maximum radial displacements. Therefore, the contact pressure at the sleeve / tube interfaces ,
should be slightly higher than those calculated for Doel 4 in Reference 4, thereby increasing the interface contact pressure and enhancing leakage and pullout resistance. Since the Doel 4 contact pressures were acceptable, then the slightly higher South Texas contact pressures ,
are also acceptable.
Loads are imposed on the sleeve as a result of tubesheet rotations under pressure and temperature conditions. A 2-D axisymmetric finite element analysis of a Model El tubesheet, channel head, and lower shell was performed in Reference 4. It has been established sat the results fmm that analysis are conservative when applied to the South Texas Units 1 and 2 steam generators. The model is shown in Figure 3.1-1. This yielded displacements' throughout the tubesheet for two pressure and three thermal unit loads. The three WPM 3413:49960496 3-8 I
1 4
0
)
.L temperature loadings consist of applying a uniform thermal expansion to each of the three component members,~ one at a time, while the other two remain at ambient conditions.
Previous calculations performed with a 3-D finite element model of this region of a Model D-4 steam generator showed that the displacements at the center of the tubesheet when the divider plate is included are [
j a,c.e, The radial deflection at any point within the tubesheet is found by scaling and combining the unit load radial deflections at that location according to:
l ..c l
l l
This expression is used to determine the radial deflections along a line of nodes at a constant axial elevation (e.g. neutral aris) within the perforated area of the tubesheet. A fifth order polynomial is fit to the Ug versus R data so that Un may be expressed as a continuous function of the tubesheet radius.
The expansion of a hole of diameter D in the tubesheet at a radius R is given by:
Radial: AD = D (dug (R)/dR)
Circumferential: AD = D (Ug(R)/R}
WPF2341-3:49460496 3-9
__________--_-_=_-___-_________-_________________-_
The maximum expansion of a hole in the tubesheet is in either the radial or circumferential direction. Typically, these two values are within ( )* of each other. Since the analysis for calculating contact pressures is based on the assumption of axisymmetric deformations with respect to the centerline of the hole, a representative value for the hole expansion must be used that is consistent with the assumption of axisymmetric behavior. Taking the average of the radial and circumferential hole expansions is a reasonable choice. This hole expansion includes the effects of tubesheet rotations and deformations caused by the system pressures and temperatures. It does not include local effects produced by interactions between the sleeve, tube, and tubesheet hole. Thick shell equations from Reference 10 in combination with the hole expansions from the finite element model displacements are used to calculate the contact pressures between the sleeve and tube, and between the tube and tubesheet.
For a given set of primary and secondary side pressures and temperatures, the contact pressure equations are solved for selected elevations in the tubesheet to obtain the contact pressmes as a function of radius between the sleeve and tube and the tube and tubesheet.
The elevations selected were the neutral axis of the tubesheet and five elevations spanning the section from the bottom of the ETS to two inches from the top surface of the tubesheet.
3.1.6.1 Normal Operation i
The temperatures and pressures for normal operating conditions at South Texas Units 1 and 2 are:
a,c,e Primary Pressure =
Secondary Pressure =
Primary Fluid Temperature (T ,1)3
=
l Secondary Fluid Temperature =
l l
For this set of primary and secondary side pressures and temperatures, the contact pressures between the sleeve and tube and the tube and tubesheet are obtained as functions of radius for selected elevations in the tubesheet for both intact tubes and tubes separated above the tubesheet.
WPF23413:49/060496 3-10 I
3.1.6.2 Faulted Condition The temperatures and pressures for the limiting faulted condition are:
a,c,e i
Primary Pressure =
Secondary Pressure =
Primary Fluid Temperature (Tw) =
Secondary Fluid Temperature =
These conditions are for late in the feed line break transient. The AP early in the transient is higher [ l'" but the primary and secondary side temperatures are much closer together. That case was considered, but the contact pressures based on the above parameters are lower than those calculated with the higher AP.
For this set of primary and secondary side pressures and temperatures, the contact pressures between the sleeve and tube and the tube and tubesheet are obtained as functions of radius
%r selected elevations in the tubesheet for both intact tubes and tubes separated above the tubesheet.
l 3.1.6.3 Summary of Results The contact pressures between the sleeve and tube, and between the tube and tubesheet are plotted versus radius in Figures 3.1-2 through 3.1-4. Results from these figures are summarized in Table 3.1-12.
These pressures are for the elevation [ l'" below the top of the tubesheet, which corresponds to the top of the hard roll of the ETS. They are conservative for any lower elevation in the tubesheet.
Note that, in all cases, the net effect of the [
l'". This contact pressure is [
)u WPF2341-3:4WD60496 - 3-11 L_ . _.__..._m.. __ .i.
l Table 3.1 l 1
Materials Used in Reference 1 for the {
Structural Evaluation of Laser Welded Sleeves l
! )
ASME Designation, Component (s) Appendix I of Reference 5.
Tube & Weld SB-163 - Mill Annealed Alloy 600 (35 ksi min yield) ;
Sleeve SB-163 - Thermally Treated Alloy 690 (40 ksi min yield)
Tubesheet &
External Support SA-508 Class 2a Ring Channel Head SA-216 Grade WCC (Generic Analyses)
SA-533 Grade B Class 2 (South Texas)
Cylinder Shell SA-533 Grade A Class 2 Table 3.1-2 Normal Operation Parameters, Reference 3, Upgraded Fuel,10*F T-hot Reduction,10% SGTP Parameter Operating Range l NSSS Power a,c.e RCS Vessel T ,,
Vessel Two#,oia Reactor Coolant Pressure Steam Pressure ,
_ Steam Temperature Zero Load Temperature .
P
?
WPF23414:4&W0496 3-12 i
Table 3.1-3 Pressure Loads for Design, Faulted, Test, and Emergency Conditions Considered in Reference 1 Pressure Load, psig, l Considered in Reference 1 l Classification Condition l' Primary Secondary AP =
P, P, P, P, a,c,e Design Design Primary 2485 235 2250 Design Secondary Faulted Reactor Coolant Pipe Break (LOCA)(50 see)
Feedline Break (FLB)
Steam Line Break (SLB)
RC Pump Locked Rotor (2 see)
Control Rod Ejection (2 see)
Test Primary Side Hydrostatic Pressure TesL Secondary Side Hydrostatic Pressure Test Tube leak Test A Tube Leak Test B Tube Leak Test C Tube Leak Test D Primary Side Leak Test l Secondary Side leak Test Emergency Small LOCA (10000 see)
- l. Small SLB (4000 see)
Complete Loss of Flow (2 sec)
WPF23414:4WD60496 3-13 l
Table 3.1-4 Pressure Drop Loads for Design, Faulted, Test, and Emergency Conditions Considered in Generic Evaluation of Reference 1 versus South Texas Values a,c.e i
I C
h l
i l
l WPF2341-3:4WO60496 3-14 {'
i
Table 3.1-5 Summary of Maximum Primary Stress Intensity Elevated Tubesheet Sleeve a,c,e l
i
. WPF2341-3:49960496 3-15
Table 3.1-6 Summary of Generic Structural Evaluation of ;
Normal, Upset, and Test Loads, Reference 1.
Values Listed are for Maximum Ratios (Minimum Margins) j and Usage which Occur for Tube Separated and Dented and a Minimum Sleeve Tube Weld Width of [0.015 inch)*#"
I
~
a,c,e D
6 l
l P
f 1
t
~ WPF2341-3:4WD60496 3-10
Table 3.1-7 Limiting Upset Transients Pressure Loads f for South Texas LWS Installation ;
Used in Minimum Sleeve Wall Calculations !
1 (Data from References 2,3)
I 8,C,0 i l
l l
l I
l l
I WPF2341-3:49M)60496 3-17 ,
Table 3.1-8 Strength Properties (ksi) - Sleeve Material Thermally Treated Alloy 690, Reference 5 a,c.e Table 3.1-9 Summary of Minimum Wall Thickness Calculations for Laser Welded Tubesheet Sleeves For Use in 3/4 inch OD Tubes for South Texas Model E2 S/G a,c,e Table 3.1 10 Summary of Recommended Plugging Limit for Laser Welded Tubesheet Sleeves For Use in 3/4 inch OD Tubes in South Texas Model E2 S/G a,c,e l
l wm3u.3-ma 3 18
4 Table 3.1-11 Comparison of South Texas and Doel 4 Maximum Tubesheet Radial Displacements (inches) a,c,t.
1 J
Table 3.1-12 Minimum Contact Pressures Between Sleeve and Tube a,c,e I
1 i
l l
WPF2341-3 49460496 3-19 l l
i l
l
a,c,e Y
I d
a Figure 3.1-1.
Finite Element Model of Model E Channel Head /TubesheetShell WPF23413:4WD60496 3-20
k a,c,e .
9 t
t e
i 4
t 4 !
W I
(
r E
i i
l t
4 i
Figure 3.1-2.
Contact Pressures for Normal Conditions with an Intact Tube WPP28413:4WD60496 3-21 l
i
-. . . . ~ .
I t
a,c,e
-?
4 i
s t
i k
i I
I t
1 i
l l
l l
Figure 3.1-3.'
Contact Pressures for Normal Conditions with a Separated Tube WPF2341-3:41W60496 3-22
)
___ _ . _ . _ , _ _ _ I
a a,c,e l 1
l t
,t f
c ,
i i
s d
P I
. 1 s'
f t
I i
i 4
1 L
I 4
i i !
Figure 3.1-4.
Contact Pressures for Faulted Conditions with Intact or Separated Tube !
r s
?
1 MAS 413:4e m o4ss 3-23 :
I
3.2 Sleeved Tube Relative Flow Induced Vibration Assessment This section evaluates the flow induced vibration characteristics of the elevated tubesheet sleeves (ETS) for South Texas Units 1 and 2. In the generic laser welded sleeving evaluation for 3/4 inch tubes (Reference 1), the vibration characteristics of a [
ju, As discussed in Reference 1, two vibration mechanisms were considered: cross flow turbulence and fluidelastic excitations. Other vibration mechanisms such as axial flow turbulence and vortex shedding were not considered viable based on field experiences and were not addressed further. The Reference 1 tube vibration assessment demonstrates that [
jus 4
In performing the Reference 1 evaluations, a relative analysis method was used. The worst case from a vibration perspective is to postulate that the tube becomes severed somewhere in
! the section of tubing between the upper laser weldedjoint (LWJ) and the top of the tubesheet.
When the tube becomes severed, the sleeve (with smaller cross sectional properties and lower
- stiffness than the tube) becomes an active load-bearing member for vibration-induced loadings, and the span length increases as the fixed lower end of the tube span moves to the top of the sleeve / tube hydraulic expansion joint (approximately [ ]"' below the top of the tubesheet). The relative evaluation of Reference 1 compares the fluidelastic stability ratios (FSR) and turbulent amplitudes (TA) associated with the nominal tube and the worst case sleeved and severed tube. The detailed methodology for this relative evaluation is summarized in Reference 1. The results of this relative evaluation are shown in Table 3.2-1.
Ten cases were run with various pinned and unsupported (open) conditions postulated for the bottom four tube support plates, for the nominal case and a sleeved and severed tube. Ratios of the sleeved and severed tube to nominal tube stability ratios and turbulent amplitudes for each case are presented in Table 3.2-1. Of the ten cases considered, only two cases resulted in FSR ratios exceeding [ ]"# and only three cases resulted in turbulent amplitude ratios exceeding [ ]"". These occurred for the lowest TSP open condition (which showed about WPF23414:G060496 3-24
1 a[
] *".
' Based on the information in Table 3.2-1, it is evident that [
j..e In conclusion, [
]*** Model E2 SGs.
I i
1 1
WPF2341-3:49/060496 3-25
Table 3.21 Relative Flow Induced Vibration Evaluation Results for a
- 22 Inch Long Elevated Tubesheet Sleeve in Model E2 SGs With Various Boundary Conditions (Reference 1) a,c,e f
e WPF2341-3:4WO60496 3-26 1
^O I
k 3.8_ Hydraulle Equivalency i
The hydraulic equivalency number (N ,4) 3 for a sleeve is defined as the number of such : [
i sleeves which will be equivalent to one plugged tube in terms of primary flow resistance.
The generic sleeving report for 3/4" tubes (Reference 1) calculated bounding values for
' hydraulic equivalency number. These values could be used for all possible sleeve .
configurations in 3/4" tubes. "The bounding hydraulic equivalency value for a tubesheet i sleeve was [
i
)a,c.e 0
t 6
?
l t
l i
WPF2341-3:4M60496 - 3-27 r
i
4.0 MECHANICAL TESTS .
Section'4.1'of Reference 1 includes the results of two FLTS lower joint development test j programs for 3/4 inch tube, non-Westinghouse SGs. Section 4.2 of Ref.1 addresses one FLTS ;
lowerjoint development test program for 7/8 inch Westinghouse SGs and one program for the
-laser weld upper joint of 7/8 inch Westinghouse SGs. The basis for the ETS lowerjoint in -
STP Unit'l is provided below; the basis for STP Unit 2 ETS lower joint is provided below, I plus a minor program to be performed for confirmatory strength and leakage resistance testing.
l The lower joint of the FLTS was developed for the Model El LWS (Doel 4) program and is - ,
directly applicable to the full-depth roll expanded tube joints of STP Unit 1; no leakage was !
recorded for this joint in the qualification test. The FLTS lower joint for the hydraulic expanded joints of STP Unit 2 will be similar to that of STP Unit 1, however, a minor- t confirmatory test program will be performed to verify this application at an appropriate point l
in a Unit 2 sleeving program. !
Although the FLTS lower joints for 3/4 inch tube sleeves shown in Reference I have been completely satisfactory for the respective applications, the ETS lowerjoints for 3/4 inch tube sleeves have been developed separately. Both types ofjoints must meet the same pullout and ,
t leakage resistance requirements for the respective applications. One of the reasons for 4
separate developments in the past was that [ j 3 i
! i k I i
t t
y ..
The upper joint testing (7/8 inch tube sleeves) in' Section 4.2 of Ref. I applies to the upper .
joints of the STP 3/4 inch tube sleeves. i i
l This report documents the development of two non-welded lowerjoints, based on testing and )
evaluation, to be used for laser welded elevated sleeves in the South Texas Project (STP) l Units 1 and 2 Model E2 steam generators (SGs). (Note: Where Unit 1 is to be differentiated ;
l WPF23414:dO60496 41 i
m d
i from Unit 2, the terms STP1 or STP2, respectively, will be used in this document.) Although ;
there are three types of sleeves which may be considered for installation in STP, only one j type, the sleeve elevated in the'tubesheet (ETS), is evaluated in this document. The longest l ETS which has been generically evaluated in Ref.1 is [ l'". The maximum length i is stated because the maximum length is bounding in terms of stress in the sleeve / tube !
structure; sleeves shorter than these involve lower stresses in the sleeve / tube structure.
The lower joint of the ETS is located within the tubesheet portion of the tube and is a [
mechanical, non-welded joint known as a mechanical interference fit (MIF) joint. Sleeve !
pullout resistance and leakage resistance at this lower joint are a direct result of the interference fit radial CP between the tube inside surface or " diameter" (ID) and sleeve i outside surface or diameter (OD). Changes to the as-installed CP of the structure result from f the four types ofloading conditions, normal operation (N.Op.), faulted, upset and test and'are ,
evaluated to ensure adequate pullout and leakage resistance.
Qualifications !
The STP1 Model E2 ETS lowerjoint consists of a [ i f
I j s.t.e i
The STP2 tubes have the same dimensions as the STP1 tubes and are also fabricated of Alloy i 600. However, the STP2 tubes are hydraulic expanded in the tubesheet holes. The lower j joint of the STP2 ETS will be very similar to an existing ETSjoint made in non-roll expanded ,
(nonrolled) 3/4 inch diameter tubes. This is the Maine Yankee process. This joint will be ;
verified by a minor program of confirmatory testing at the appropriate point in the !
prepr. rations for the field sleeving operation. ,
4.1.1 ETS Lower Joint for the Model E2 SG Factory Roll Expanded Tube Joints of !
STP1 f One of the previous ETS lowerjoints is the Westinghouse Model El SG configuration at Doel l
- 4. It was used in the 100 percent sleeving program at that site. Consisting of the same ;
- sleeve, as well as the factory roll expanded tube as the STP1 joint, it was developed for the i
- 3/4 inch x 0.043 inch (nom.) wall, Alloy 600 tube, in 1994. The tubesheet unit cells, the i
! quantity of tubesheet material which is considered part of the tubejoint surrounding a single :
tube, are the same for the Model El and Model E2 SGs; the pressure boundary materials, ;
WPF2341-4:49060496 42
{
t
.. . = - - - - . .. . .- . . . . ,
l i.
l including the sleeve material, are also the same. The sleeve lowerjoint fabrication, including the roll expansion and hydraulic expansion processes, roll expander type and torque are the l
' same for sleeves for the two SG models. However, the Doel 4 sleeve installation sequence was - !
slightly different from the planned STP1 sequence; [
i l
l j.,, ;
i The [ !
i l
j e, i i
t The MYjoint consists of a [ .
]'## installed in a 3/4 inch (nom.) x 0.048 inch (nom.) wall thickness, Alloy 600 tube, !
explosive expanded in the tubesheet in the factory. The MY ETS results are used herein to l show that the qualification results for the [ ;
]*## Refer to Tables 4.1-1 and !
4.1-2. The MY evaluation considered both [ ]*## sequences and these l results show the effect of installation sequence on sleeve-to-tube contact pressure and '
. leakage. Contact pressure [ l l*## The MY leakage results are shown in Section 4.3. :
The thermalhydraulic conditions, tubesheet dimensions and tubesheet configuration differ j slightly between the STP1 and MY SGs. Therefore, the sleeve lowerjoint undergoes different !
contact pressure changes for the two SGs; the effect of these changes on leakage and pullout !
resistance for STP1 are shown below and in Tables 4.1-1 and 4.1-2. The changes for MY are l
- also shown in these Tables. !
i I
l
.h WPF234144WO60496 '4-3 f i
f 4.1.2 STP1 ETS Leak Test Equivalence Leak testing at room temperature, along with minor adjustments of the results to relate to the plant, is demonstrated to be applicable for operating and accident conditions on the following basis.
Normal Operation The change in CP for STP1 was determined for the most stringent conditions. This is for a tube location in the tubesheet at a radius of approximately [ '
i
).u ;
Applied Ie, d Effects on Sleeve-to-Tube Lower Joint Roll Expansion - N.Op.
a,c,e
)
i P
J r
l l
l 4
i The laboratory room temperature (RT) leak testing results for the Model El SGs of Doel 4 were applied to the Model E2 SGs of STP. In this application, the [
l'" Refer to Table 4.1-1.
WPF2341-4:4WO60496 ' 44 l
l
__._________.___..___J
Faulted -
An evaluation for the faulted test conditions, similar to that made for the N.Op. conditions above, was made and indicates a [ ] *"
Applied Load Effects on Sleeve-to-Tube Lower Joint Roll Expansion - SLB a,c,e The laboratory RT leak test results from the Model El (Doel 4) program were applied to the Model E2 SGs of STP1. In this application, the [
l'" Refer to Table 4.1-2.
4.1.3 Justification for Using the Model El ETS Lower Joint Pullout Resistance for STP1 The Model El ETS lower joint process pullout resistance applies to STP 1 because the as-installed contact pressures will be approximately equal. Further, the N.Op. and faulted cps, for the limiting location in the bundle will be approximately equal. Refer to Table 4.1-3.
l 1
WTT2341449060596 - 4-5 :
l
e 4.1.4 Justification for Using M. Yankee ETS Leakage Resistance Results as Bounding Results for STP1 The change from the [
34 Bounding criteria for permissible primary-to-secondary side leakage during N.Op. were determined by allocating one-half of the 150 gpd limit, i.e., 75 gpd, to the lowerjoints oflaser weld sleeves and by conservatively assuming that all sleeved tubes develop throughwall degradation. This allocation permits the other 75 gpd to be assigned to other potential primary-to-secondary leakage. This flow, averaged over all 4851 hot leg tube joints, for the 100 percent sleeving case, results in a per-sleeve leakage of approximately 0.81 drops per minute (dpm) per sleeved tube. With the per-sleeve leakage projected at [
j.e Similar assumptions, i.e., allocation of one-half of the permissible flow, for 100 percent sleeving, for the faulted condition, for the usual 1 gpm (1440 gpd) primary-to-e;condary leakage for the plant. results in an average permissible flow of approximately 3.9 dpm per sleeve. This is a factor of approximately 22 over the leakage of [ ]*e#
projected above. The allocation factor in this case could be significantly reduced.
It is estimated that, for a STP1 SG with 500 ETSs, [
) ..c In the test, the projected ETS average leakage was [
l'## For the sake of discussion, using the ETS test criterion average leakage of 0.25 dpm per tube at N.Op., the selected portion of the MY Administrative Leak rate of 0.459 dpm per tube, the total primary-to-secondary leakage for [
) .,c,e i
WPF2341-4.4WO60496 4-6
I
[
ja.e 4.1.5 Conclusions .
l The existing ETS joint processes for 3/4 inch x 0.043 inch wall thickness tubes (Model E1)'
- will provide acceptable pullout resistances for N.Op., faulted, test and upset con'ditions in the
-[. l'## sequence for the Model E2 SGs of STP1. ,
The ETS lower joint also provides acceptable leakage resistance; the leakage will be
- negligible assuming that [
l'#d in the tube portion spanned by the sleeve.
P e
b d
1 i
WPF2H144%60496 47 l
1
.1
4 (Page 1 of 2)
Table 4.1-1 Applicability of Doel 4 (Model E1) LWS* Lower Joint Process to STP 1 & 2 Normal Operation a,c,e i
1 L
1 i
i e
a a
1 J
WPF23414:49/060496 ' 4-8 1
. i 1
Table 4.11 Applicability of Doel 4 (Model E1) LWS* Lower Joint Process to STP 1 & 2 Normal Operation :
a,c,e f
t a
4 i
r 1
5 2
a t
b 4
5 WPF23414:4WO60496 ' 4,9 t
t Table 4.12 (Page 1 of 2) t Applicability of Doel 4 (Model EI) LW Elevated Tubesheet Sleeve
(
- Lower Joint Process to STP1 & STP2 Faulted Condition '
a,c.e t I j
l P
I l
l l
i I
l l
I i
wpFM41-4:49M60496 4-10 i
- Table 4.12 (Page 2 of 2)
Applicability of Doel 4 (Model EI) LW Elevated Tubesheet Sleeve
- Lower Joint Process to STP1 & STP2 Faulted Condition a,c.e l
WPF23414:4&W496 4 11
Table 4.1-3 Maine Yankee Roll Last ETS Lower Joint - l l
as !
Basis for STP1 Roll Last ETS Lower Joint i A,C,e d
9 n
, i 6
i
)
i i
i I
WPF23414:4WD60496 4 12
4.2 ETS Lower Joint for the Factory Hydraulic Expanded Tube Joints of STP2 W
The STP2 tubes are also Alloy 600 tubes. However, the tubes were expanded into the !
- tubesheet by the hydraulic expans )n process. This process is similar to the explosive ,
expansion process used in the MY Si s. The [ l t
I E
j.
t i
i i
i I
i WPF2341-4:4WD60496 4 13
.i
I i
. Table 4.2-1 l t Maine Yankee Roll Last ETS Lower Joint I as 1 Basis for STP2 Roll Last ETS Lower Joint ,
e 8,C,e ,
l a
a A
a i
i >
i 4
I h
a i !
i 1
I i
i i
J A
i l
I i
1 e
WPM 3414:4&O60496 4 14 l
. i i
4.3 Details of Previous 3/4 Inch Tube Sleeve Lower Joint Process Qualifications Mechanical tests are used to provide additional information related to sleeve joint i performance. Unit test cells are used for mechanical testing. A unit test cell or specimen is !
one which is a single sleeve joint and sufficient tube and sleeve length to bound transition l effects. For tubesheet specimens, a collar is used to simulate the effect of the tubesheet. The >
wall-thickness of the collar has been selected to simulate the radial stiffness of the steam l generator tubesheet. ;
Mechanical testmg was previously applied to both HEJ (lower joint) and laser welded (free i span and lowerjoint) sleeving to confirm analyses that evaluated the interaction between the -
sleeve _ and tube. ' Mechanical testing is primarily concerned with leak resistance and joint ;
. strength, including fatigue resistance. A consistent characteristic observed in the testing of HEJ, a.k.a. mechanical interference fit (MIF) lowerjoints for sleeves, is that I '
i p.e l
[
Sections 4.3.1 and 4.3.2 summarize previous mechanical tests and results for 3/4 inch tube j sleeves. The 3/4 inch sleeve results show the adequacy of obtaining the required strength of l the roll expanded portion of the MIF lowerjoint, based on optimal roll thinning of the sleeve. )
This same method is used to achieve the required strength of the roll expanded portion of the i
- the STP2 sleeve MIF joints (see Section 4.2).
f t
In previous testing documentation, Reference 1, some of the 3/4 inch tube sleeve lowerjoint ,
i
- - specimens were also subjected to cyclic thermal and mechanical loads, simulating plant transients. The magnitudes of these forces and temperatures were determined from plant ;
normal operating and postulated accident conditions. The force loadings assumed locking of l
- the tube to the support plate structure and accounted for differential thermal expansion '
between the steam generator shell and the tube bundle. Other specimens were subjected to tensile and compressive loads to the point of mechanical failure. These tests demonstrate that the required joint strength exceeded the loading the sleeve joint would receive during normal plant operati ms or accident conditions.
J l
WPF23414:49460496 4 15 '
O Note: In the following test portions of this report, the units of primary-to-secondary side differential pressures are listed simply as " psi," rather than "psid." The secondary side pressures were zero psig.
4.3.1 Elevated Alloy 690 Sleeve-FSG (Maine Yankee)
A sleeve elevated in the tubesheet (ETS)is shown in Figure 2-2 of Ref.1. The lower joint of this sleeve is a mechanical, non-welded joint known as a mechanical interference fit (MIF) joint. Sleeve pullout resistance and leakage resistance at this lowerjoint [
ju The ETS for this FSG consists of a [
}"" installed in a 3/4 inch (nom.) outside diameter x 0.048 inch (nom.) wall thickness, Alloy 600 tube, explosive expanded in the tubesheet in the factory.
4.3.1.1 Leakage and Pullout Resistance Tests It has been determined that for ETS MIF lowerjoints, [
l r
i i
l l
i j a.c.e Similar to the method used in the pullout resistance calculation, the changes from the as- ;
installed CP were also used to project leakage for the RT leak test. The result of the beneficial and detrimental changes to the as-installed CP were used to bound the projected leakage at elevated temperatures.
WPF2341-4:4WO60496 4-16 l
l
il 4.3.1.2 Acceptance Criteria ,
Primary to-Secondary Leakage The acceptance criterion for average leakrate at normal operating conditions was 0.25 drops :
per minute (dpm) per sleeve. This was based on a fraction of the Administrative permissible
- leakrate of 50 gpd per SG. A similar limit for faulted conditions was not specified. (Note:
There are approximately 75,000 drops in one gallon.)
Pullout Resistance - Normal Operation -
A pullout resistance of three times tbs maximum primary-to-secondary pressure differential, times the tube cross sectional area,i.e., the "endcap" load, for normal operation has been used as the requirement for sleeve MIF lower joints and it is consistent with the ASME B&PV Code. Based on this approach, the limiting required resistance to pullout, upward, for the FSG in this case was 1605 lbs. for the most stringent case,1500 psi pressure differential, at the design condition, and the largest tube ID, resulting from installation of a tube in the ;
largest hole in the tubesheet.
Pullout Resistance - Faulted Condition A pullout resistance of 1.43 times the "endcap" load for the corresponding primary-to-secondary pressure differential, for the limiting faulted condition, SLB, was used as the requirement for sleeve MIF lower joints and it is consistent with the ASME Code. For the FSG in this case, the maximum pressure differential for the limiting faulted condition was 2520 psi; the largest tube ID, [
p Pullout Resistance - Upset and Test Conditions These conditions had been bounded by normal operation and SLB conditions in previous evaluations and it was assumed that the same was true for this case.
Pullout Resistance Limiting Condition - Conclusion The limiting axial load for the joint design is the greater of the "3 AP" endcap load for the normal operation condition and the "1.43 times the largest faulted endcap load". In this case, the normal operation condition endcap load,1605 lbs. caused the largest load and was the limiting condition.
WPF2341-4:49/060496 4 17
4.3.1.3 Results and Conclusions of Verification Tests Leakage Resistance - Normal Operation i The leak test results are shown in Table 4.3-1. The average leakage under normal operation .
l conditions, i.e.,1900 psi AP and RT, was [
je Leakage Resistance - SLB 4
The same sleeve / tube /tubesheet simulant samples used for the normal operation testing were
~also used for SLB testing at a AP of 2650 psi. Therefore, SLB leakrate testing was performed for Engineering Information only. The average leakage under SLB conditions,i.e., 2650 psi AP was [
f j .c Pullout Resistance I
j ' As stated above, the objective of the secondary-to-primary side differential pressure testing was to determine the sleeve-to-tube interference fit CP. The pressure at which the leakage became significant was a conservative measurement of the contact pressure; the actual contact pressure was higher than the leakage initiation pressure. The test pressures were 4 [
l 1
i
- ja.c 4 4.3.2 Elevated Alloy 690 Sleeve PSGs (Doel 4) ,
Two previous interference fit, ETS lower joints have been developed for sleeves for Westinghouse 3/4 inch diameter tube steam generators (SGs).
. One of these is the Westinghouse Model El configuration which consists of the same sleeve and tube as the second conSguration, the Model D4, ETS lower joint, i.e. a [
l*#' installed ,
in a 3/4 inch (nom.) outside diameter x 0.043 inch (nom.) wall thickness, Alloy 600 tube, roll I expanded at the factory. The tubesheet unit cells are also the same for the Model El and l
WPF23414:4&O60496 4-18 l
I
r Model D4 SGs. The sleeve lowerjoint fabrication, including the roll expansion and hydraulic expansion processes, roll expander type and torque are the same for sleeves for the two SGs.
However, the Model El sleeve installation sequence was slightly different from the Model D4 ,
sequence; [
j a,c.e 4.3.2.1 Leakage and Pullout Resistance Tests The same tests used to qualify the ETS MIF lower joints for MY were'used to qualify the lower joints for Doel 4. This included the RT leakage resistance test, the CP test and previous testing. Additionally, elevated temperature leak tests and RT pullout resistance
. tests were also performed. ,
The objective of the leak tests was to determine potential primary-to-secondary side leakage for the rare case where the tube became completely degraded within the sleeve length. (The upperjoint, the laser weld, was taken as leaktight in this and in all cases.) Leak tests were performed for this case at RT at [
).
In the pullout resistance related test, the [
j a.c WPF23414:49960496 4 19
1 1
l I
F S
1 jo 4.3.2.2 Acceptance Criteria Primary to-Secondary Leakage i The acceptance criterion for average leakrate per sleeve at normal operation conditions was not listed specifically for this PSG site.- However,in the absence of an Administrative Leak l limit in this non-domestic plant, a typical value would be [
4 j.
Pulleut Resistance - Normal Operation ,
As discussed previously, a pullout resistance of [
ju ,
t a
f L
WPF23414:4&O60496 4 20
Pullout Resistance - Faulted Condition The corresponding pullout resistance at the most stringent faulted condition, FLB, was calculated to be [
j .e..
Pullout Resistance - Upset and Test Conditions These conditions had been bounded by normal operation and the most limiting faulted conditions in previous evaluations and it was assumed that the same was true for this PSG
.- case.
Pullout Resistance Limiting Condition - Conclusion The limiting axial load for the joint design was the greater of the [
l was the limiting condition.
4.3.2.3 Results and Conclusions of Verification Tests Refer to Table 4.3 2.
Leakage Resistance - Normal Operation Reference to Table 4.3-2 shows that, for the 20 samples, no leakage was observed at RT or the normal operation temperature of 626 F. The permissible leakage guideline (criterion) of
[ l'## per sleeve was met.
Leakage Resistance - FLB The same 20 sleeve / tube /tubesheet simulant samples used for the N.Op. testing were also used for FLB testing at AP's of[
, l*## per sleeve was met.
Pullout Resistance Related Test l
As stated above, the objective of the secondary-to-primary side pressure testing was to ;
determine the sleeve-to-tube interference fit CP. The pressure at which the leakage became significant was a conservative measurement of the contact pressure; the actual contact ;
pressure was higher than the leakage initiation pressure. The test pressures were [ l*##
WPF2341-4:49/060496 4-21 l 1
e e
I j
1 i
1 1
J l
l J
l l
I WPF2341-4:M60496 . 4 22 4
Table 4.3-1 (Page 1 of 3)
Verification Test Results - Mechanical Interference Fit Lower Joint Elevated Tubesheet Sleeve Alloy 690 Sleeve for 3/4 Inch Tube (Feedring SG )
Room Temp. Leak Rate (dpm)
Sample Identification Sleeve i D. Pressure, PSI
, Siv. Roll Torque A%T*
Collar Tube Sleeve Collar 1.D. Lgn. in. Len, in. in. # % 1900 2650 3100 a,c.e WPF2341-4:49/060496 4 23
Table 4.31 (Page 2 of 3)
Verification Test Results - Mechanical Interference Fit Lower Joint Elevated Tubesheet Sleeve Alloy 690 Sleeve for 3/4 Inch Tube (Feedring SG )
Room Temp. Leak Rate (dpm)
Sample Identification Sleeve i D. Pressure. PSI Siv. Roll Torque AWT*
Collar Tube Sleeve Collar I.D. Lgn. in. Len. in. in. # % 1900 2650 3100 18 14 202 Maximum 12.0 1.50 150 5.88 0.040 0.176 0.120 a c.e
>. l r
l l
1 i
l WPF23414:4WO60496 4 24 i
I Table 4.31 (Page 3 of 3)
Verification Test Results - Mechanical Interference Fit Lower Joint Elevated Tubesheet Sleeve Alloy 690 Sleeve for 3/4 Inch Tube (Feedring SG )
Room Temp. Leak Rate (dpm)
Sample Identification Sleeve I D. Pressure PSI Siv. Roll Torque AWT*
Collar Tube Sleeve Collar I.D. Lgn. in. un. in. in. # % 1900 2650 3100 a,c.e e
}
WPF2341-4:4920496 4-25 l
Table 4.3-2 (Page 1 of 5) 3 Verification Test Results - Mechanical Interference Fit Lower Joint 5 Elevated Tubesheet Sleeve Alloy 690 Sleeve for 3/4 Inch Tube (Preheater SG) a,c,e i
i, i
l l
. I s
t____ ____ _ _ _ _ _ _ _ _ _ - -______--- __________i. ____i- _ -____ - _ _ _ _ , . _ _ _ _ _ _ -
m____. _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
5 g Table 4.3-2 (Page 2 of 5) 3 Verification Test Results - Mechanical Interference Fit Lower Joint
$ Elevated Tubesheet Sleeve i Alloy 690 Sleeve for 3/4 Inch Tube (Preheater SG) a,c,e i
E i
l l
l l
I t
[ i l
i
5 g Table 4.3-2 (Page 3 of 5) g Verification Test Results - Mechanical Interference Fit Lower Joint c Elevated Tubesheet Sleeve i Alloy 690 Sleeve for 3/4 Inch Tube (Preheater SG) a,c,e k
2 3
M
_ _ . _ . __.__.m_ ___ --____ - m._ __ - _ _ _ _ _ _ - -- _ _ _ _, i___w + _-a w w' + -' Fv' w
Table 4.3-2 (Page 4 of 5) l$ Verification Test Results - Mechanical Interference Fit Lower Joint
_ ;2
$ Elevated Tubesheet Sleeve i Alloy 690 Sleeve for 3/4 Inch Tube (Preheater SG) . a,c,e k
e .
8-n
Table 4.3-2 (Page 5 of 5)
- 5 Verification Test Results - Mechanical Interference Fit Lower Joint
'E Elevated Tubesheet Sleeve i Alloy 690 Sleeve for 3/4 Inch Tube (Preheater SG) a,c,e i
S O
a
_ _ _ _ _ . _ _ . _ _ _.___m __ ___.__m.. _ _ _ __. _ _______ . _ _ _ _ _ _ _ ._____________._.__._ _ _______ m - ___ . . . - -
5.0 STRESS CORROSION TESTING OF LASER WELDED SLEEVE JOINTS
- The Alloy 690 TT (thermally treated) sleeve material exhibits exceptional resistance to stress corrosion cracking in steam generator environments (Reference 11). Based on all available corrosion test results, Alloy 690 TI' appears immune to stress corrosion cracking in primary water (PWSCC), and offers substantial advantage over other candidate SG tube alloys in faulted secondary side environments. For this reason it has been the preferred alloy for heat transfer tubing in new and replacement SGs since approximately 1988;its use for sleeving extends back as far as 1984.
i The resistance, therefore, of the laser weld-repaired sleeve joint is dictated by the resistance of the Alloy 600 tubing at the repair elevation. Hence, the major threat to the operational integrity oflaser welded sleeve repairs is the magnitude of the stresses residual to the sleeve installation process. These stresses are the combined results of: (a) the hydraulic expansion of the sleeve and tube, (b) the stresses associated with the welding process, cnd (c) the far-
_ field stresses that develop during post-weld thermal stress relief.
The purpose of the thermal stress relief operation is to reduce the peak residual stresses in the fusion weld and, for certain installation geometries, the peak stresses in the upper (free-span) hydraulic expansion transitions. However, under conditions where the tubes are {
axially restrained by locking and/or denting at the tube support plates, the thermal stress relief can elevate substantially the far-field stresses that develop in the tubing. These stresses would be additive to any remaining unrelaxed stress at the laser weld / hydraulic expansion locations. As discussed in subsequent paragraphs of this section, the role of these stresses on the corrosion resistance of tube-sleeve assemblies has been recognized and an attempt made to evaluate their effects.
In view of the role played by the stress level in determining the service performance of weld-repaired SG tubing, a discussion is presented in the next subsection of the influence of the ,
LWS process parameters and SG design variables on stress. A subsequent section reviews l briefly the effects of thermal stress relief on stress levels and is followed by a summary of the results of corrosion tests performed to evaluate the resistance to PWSCC oflaser welded sleeve-repaired tube mockups. Included in this summary are the results of tests on as-welded mockups (i.e., without post-weld stress relief), mockups tested under conditions without applied axial loads, and mockups tested under conditions believed to reflect the conditions which might exist under conditions of axial restraint. The laser welding processes used to prepare the test specimens are representative of the field processes using the neodymium-YAG (Nd:YAG) pulsed laser currently in use by Westinghouse for sleeve welding.
1 2341-5:49/060496 5-1
J 4
5.1 LWS Process and SG Design Variables i
The influence of the sieeve process parameters and steam generator design features or tube conditions on stress levels is summarized in Table 5-1.
- As installed, i.e., prior to thermal stress relief or final hard rolling, the far-field stresses are !
- - generally low, on the order of a few ksi. The peak residual stresses at the laser weld, -
! however, are quite high; they have been estimated as approaching 80 - 85% of the tensile t l yield strength of the sleeve-tube assembly. [ j
?
j p.e j l
- The residual stresses at the upper hydraulic expansion are somewhat lower - estimated as j
- - 35 kai (however, this region may also be subject to relatively early corrosion failures if these ;
! stresses are not reduced). :
f' The most practical means to relax these peak residual stresses is by thermal stress relief.
For conditions in which the tube is free to expand axially, i.e., no fixity or restraint at the j support plate locations, stress reliefis an efficient process, and has a negligible impact on far-i field stresses. However, recent experience with operating steam generators suggests this j condition may not always exist, and it is useful to assume for conservatism that the tubes (
- may in fact be locked at the tube support plate (s); most recent corrosion tests have thus been performed under the more conservative assumption, i.e., under conditions of applied axial stress.
1 The consequence of a locked tube condition is that thermal stress relief, while lowering peak l 4
residual stresses at the laser weld and at the hydraulic expansion, may increase the far-field !
3 1 axial stresses in the tube and may lead to bulging distortion of the tube at and above the !
elevation of the weld. Since this response is a consequence of the thermal expansion of the tube, the higher the stress relief temperature or the greater the axial extent of the region '
. being stress relieved, the greater the axial far-field stresses.
l Hence, the thermal stress relief process must be carefully tailored to achieve a trade-off >
' between reduction of the peak stresses at the weld and hydraulic expansion transition while ,
. at the same time minimizing the far-field stresses.
l In view of the influence of the tube-tube support plate span length on tl.e magnitude of the ,
far-field stresses, optimization of the sleeve installation r.nd abiss rdief process must be !
defined on a plant (or SG design)-specific basis. ,
t i
^
t p as41.s:4em6o496 5-2 e
n e- . . . - , , , -<-n a-. -,-~n--
5.2 Residual Stresses vs. Stress Relief Temperature in LWS Sleeve Repairs Table 5-2 summarizes the expected range of far-field stresses that result as a function of the stress relief process. These are conservative stress values from strain gage measurements above and balow the laser weld location and are for temperatures measured at the weld and upper hydraulic expansions of sleeve mockups; the values shown are appropriate to Model E preheat SGs [
y.c These data show the substantial reduction of far-field stress that can be realized in LWS-repaired SG tubing by controlling the stress relief temperature to be in the lower portion of t the allowable range.
5.3 Corrosion Test Description Since approximately 1988, Westinghouse has used the doped steam corrosion test to evaluate the resistance of test mockups or repair assemblies to primary water stress corrosion cracking
, (PWSCC). This test is conducted in dense steam in an autoclave operating at 750*F (400*C).
The steam is doped with 30 ppm each of fluoride, chloride and sulfate ions in addition to 11 psig of dissolved hydrogen. For test mockups of the type considered here, the doped steam, at a pressure of 3000 psig, is in contact with the ID surfaces; the environment on the OD surfaces is pure steam at 1500 psig.
This test provides an extreme acceleration of the corrosion process relative to that which occurs in an operating steam generator. In some respects, the doped steam test can be viewed as a stress-indexing test; failure times in the doped steam test can generally be analyzed in terms of the stresses (residual and pressure) present in the test articles. In view of the dominant role stress plays in PWSCC of Alloy 600, this is a particularly valuable feature of the test.
The acceleration of the corrosion process provides the opportunity to evaluate the corrosion resistance of configurations appropriate to the repair process ofinterest, and avoids the need to rely on such stress-indexing tests as the stainless steel-MgCl2 or Alloy 600-sodium
, tetrathionate tests which require surrogate materials or nonrepresentative microstructures.
As mentioned above, corrosion tests have been performed on tube-sleeve mockups in the as-welded condition, and for conditions representing weld stress relief with and without the addition of axialloading.
Generally, two types of specimen have been tested. The first of these, illustrated in Figure 51, has been used to test laser weld joints in the as-welded condition, or in the condition following thermal stress relief of the joint, but without additional axial load.
l 2341-5:49/060496 5-3 l
l l
l
i
~
i l
l The second configuration is somewhat more complex. In this mockup test, the specimen is ,
fabricated using a test stand as shown in Figure 5-2. The purpose of the test stand is to pennit the sleeve installation, hydraulic expansion, welding, and post weld thermal stress relief under locked tube conditions. The nominal span length between supporting plates is varied to simulate the appropriate values for the SG model/ design ofinterest. The stresses that result from the several stages of fabrication are measured by placing strain gages above and below the weld location. Temperatures are recorded throughout the stress relief process.
Following all specimen fabrication steps, the specimens are unloaded and prepared for corrosion testing. The configuration of the test assembly used for these tests is shown in Figure 5-3. The specimen is loaded axially in a tensile machine to the strain values noted in the fabrication sequence. By means of the threaded end fitting at the top of the assembly and the compression cylinder, the axialload is established and maintained on the sleeve joint throughout the carrosion test.
To facilitate interpretation of the corrosion test results and to provide verification of the aggressiveness of the test environment, roll expansiontransition mockups, prepared of Alloy 600 tubing with known low resistance to cracking, are included in the test autoclaves.
5.4 Corrosion Resistance of Free-Span Laser Weld Repaired Tubes
- As-Welded Condition Corrosion testa have been performed on laser weld-repaired tube assemblies prepared using both the CO, and the Nd:YAG laser processes. The former process is no longer ofinterest and will likely not be used for field operations; hence, data are presented here only for the Nd:YAG process.
The corrosion tests on as-welded mockups have been performed on specimens of the configuration shown in Figure 5-1; i.e., without added axial load. The doped steam test results are summarized in Table 5-3. (Table 5-3 also includes some data for stress-relieved Nd:YAG welds.) [ ]
j.
A limited number of as-welded 3/4 inch tube-sleeve mockups have also been tested (ca.1994) to support a field sleeving campaign. [ j j
) s.c.e l
Figure 5-4 is a micrograph showing the typical failure location in these test specimens. The failures invariably occurred in the Alloy 600 base metal adjacent to the weld. The cracking !
is intergranular, typical of PWSCC, and is circumferential in orientation. This failure mode i
2341-5:49/060496 5-4
, l l
i l
has been observed in essentially all laser weld-repair mockups tested, irrespective of whether or not the specimen was stress relieved, or subjected to additional axial load during the test.
5.5 Corrosion Resistance of Free Span Laser Weld Repaired Tubes
- with Post Weld Stress Relief In addition to the results presented in Table 5-3 referred to in the previous subsection, doped steam corrosion tests were performed on 3/4 inch tube-sleeve mockups to support the 1994 field sleeving campaign. These specimens were tested without the imposition of axial loading. One of the objectives of this test program was to evaluate the effectiveness of the post weld thermal stress relief over the temperature range 1275 - 1675'F (for the relevant ;
sleeving campaign, the process specification was 1400 - 1600*F). The results of these doped steam tests are presented in Table 5-4.
These tests were, for the most part, terminated at 200 - 227 hours0.00263 days <br />0.0631 hours <br />3.753307e-4 weeks <br />8.63735e-5 months <br />, a time period agreed upon
-with the utility as sufficient to demonstrate adequate resistance to in-service degradation through the remaining service performance of the steam generators. All specimens were ;
post-test destructively examined by splitting and flattening. [
p 5.6 Corrosion Resistance of Free-Span Laser Weld-Repaired Tubes
- with Post Weld Stress Relief and Conditions of Axial Load During Test .
Experience related to a field sleeved tube inspection campaign indicated that restraint to I axial expansion due to locking of the tube at the tube support plate (TSP) elevations could
- lead to " bulging" of the tube above the sleeve, and the introduction oflarge axial "far-field" stresses. This provided the incentive to include conditions ofrestraint both during fabrication of mockups for testing and during corrosion testing.
The degree of axial restraint varies (see discussion in Subsections 5.2 and 5.3) with span
, length (e.g., the distance from tubesheet to TSP) and installation / fabrication parameters - in particular, the thermal stress relief. Hence, most recent tests have used conditions which recognize these factors for the specifle plant or sleeve application ofinterest.
Doped steam corrosion tests have recently been performed on 3/4 inch tube LWS-repaired
- mockups prepared to simulate sleeving installations for two different models of operating ,
steam' generators. A summary of the fabrication parameters, pertinent measurements made l during the mockup fabrication, and the results of corrosion tests is provided in Table 5-5. j i
2341449m6o496 5-5 I
For each of these mockups, only the laser weld regions were stress relieved. This minimizes the increase in far-field axial stress while providing a more efficient field installation process by avoiding the need to separately stress relieve the upper hydraulic expansion (UHE) transitions. This is only practical, however, when the distance between the weld and the UHE is sufficiently small that the stress relief of the weld provides a measure of stress relief to the UHE as well. [
ju A consequence of this difference can be seen in the corrosion test results for these specimens, Table 5-5. All failures in the CAE set of specimens occurred at the UHE locations, whereas the failures for the MHE specimens occurred equally at the weld and UHE elevations, at longer times than for the CAE specimens.
The axial stresses imposed on these specimens during testing were determined from strain gage measurements made during specimen fabrication. The variation seen for the four CAE specimens reflects the fact that CAE-001 and -003 experienced a maximum temperature of 1520 - 1530*F, whereas for CAE-002 and -004 the maximum temperature was approximately 1355*F. The lower stresses in the MHE mockups reflects the use of a slightly lower weld stress relief temperature range and the substantially greater span length relative to the CAE mockups (47 inches vis-a-vis 36.7 inches).
The experience accrued in the fabrication and testing of tube-sleeve mockups has been used to optimize the field sleeving process so as to minimize field installation time while at the same time arriving at a configuration in which the local weld stresses and far-field tube stresses are controlled so as to maximize field service performance of the sleeve repairs. This optimization involves modifying the equipment such that the distance between the laser weld and the UHE is kept to a practical minimum, thereby permitting effective stress relief of both regions at the same time. [
ju Note re. Current Field Installed Laser Welded Sleeves l The performance oflaser welded sleeve repairs in operating steam generators has been excellent. Tubesheet and TSP sleeves have been in service in a domestic nuclear power plant for over four years with no indications of degradation. These sleeves are in tubes known to ,
have some degree oflock-up at the TSPs; [ }'^*
l Stress relief was limited to the weld region and was performed at approximately 1400*F.
1 2341-5:49/o6o496 5-6 I
In a non domestic plant, approximately 5 years of operation had been attsined with LWS-repaired tubes at the time the repaired SGs were replaced, again with no incidents of degradation. In another non-domestic plant, over 11,000 elevated tubesheet sleeves have been in service for approximately 21 months. After approximately 10 months of operation, NDE of all sleeved tubes and destructive examination of ten pulled tube-sleeve assemblies revealed no in-service corrosion degradation of the laser welds, the hydraulic expansion regions, or the tube bulges that resulted from stress relief under locked tube conditions.
In the following subsection an estimate is provided of the service performance that might reasonably be expected for sleeve installations in Model E SGs such as those at South Texas.
5.7 Estimated Sleeve Performance at South Texas Two conditions were considered. These were: (a) the tubes are completely free to expand axially upon sleeving and thermal stress relief; and (b) the tubes are rigidly fixed at the first tube support plate (TSP).
In performing the follawing estimates, the operating temperature of South Texas (T -
626*F) is taken into consideration.
All estimates of sleeve performance were based on stresser ineasured in prototypic mockups for which the laser weld stress relief region experienced five minute exposures at 1350*F.
Stress relief of the upper hydraulic expansion transitions is not anticipated to be necessary.
Tubes Free to Expand Axially In this case, following thermal stress relief of the laser weld region, the primary stresses acting on the tube-sleeve assembly are the remaining residual weld stress and the operating pressure stress. [
j a.c .
I y.e 2341 5:49/060496 5-7
i l
Tubes Fixed at the First Tube Support Plate l In the South Texas Model E steam generators, the first TSP is at an elevation approximately :
37 inches above the top of the tubesheet. For fixed conditions at this elevation, the far-field stresses after thermal stress relief of the weld will be in the range of [ }'##
Corrosion testing of mockups under this condition of stress, again from comparison with roll transition mockups exposed at the same time, indicates degradation-free sleeve performance ,
for periods approximately twenty times those required to initiate PWSCC in roll transitions.
Adjusting these results to the case of South Texas provides an estimate of [ l*#" years
. of service for the laser welded sleeves. ,
A summary of the estimates for the service performance oflaser welded sleeve-repaired tubes at South Texas, for the different conditions assumed for tube fixity, is provided below.
l a,c,e I
t i
i i
2341 4:4& O60496 5-8 l
i
5.8 Outer Diameter Surface Condition Because the sleeving involves operations only on the primary side, no aspect of the sleeve !
installation directly involves the tube OD surfaces. In operating SGs, however, the OD surfaces undergo surface corrosion and may collect deposits. These are typically oxides or related minerals in the thermodynamically stable form of the constituent elements;in PWR secondary water, magnetite is the most prominent oxide that forms. At the temperatures experienced during sleeve welding and thermal stress relief, these compounds are stable and do not thermally decompose. All such compounds have crystal structures that are too large j to permit diffusion into the lattice of the Alloy 600. Reactions between these stable oxides 4
and minerals and the alloying elements of Alloy 600 are thermodynamically unfavorable.
j Consequently, their presence during sleeve installation is not expected to produce deleterious j tube-sludge / scale interactions.
J This judgment has been evaluated by installing and laser welding sleeves into tubes removeci from operating plants. Following the sleeving operations, microanalytical examinations were performed to verify the lack ofinteractions. Prior to welding, the tubes had oxide deposits which contained Cu, Ti, Al, Zn, P and Ca as measured by EDAX analyses on an SEM. '
Following welding and stress relief the maximum penetrations of the OD surfaces were on j the order of 7 to 8 pm (less than a grain depth).
Additional evaluations were performed on three areas of an Alloy 600 U-bend section which was coated with sludge and heat treated in air for 10 minutes at 1350*F. The sludge was a simulant of SG secondary side sludge (Fea4O , Cu, CuO, ZnO, CaSO4 and MgCl 2) and was applied to the U-bend using acrylic paint as a binder. Post-thermal exposure evaluations indicated no general or intergranular corrosion had occurred.
r P
2341.s:49/060496 5-9 2
6
.- . . - -_ .. - . . . . - . -. . ~ _ . . . .- - . - . .-. .. . - . . . . .- ...
l I
f a
Table 5-1 .I Summary of Impact of Laser Welded Sleeve Operations on Stresses l a,c.e 1
1 1
l 4
P
, i i
d
\
4 9
1 i
3 4
j i
4 ss414:4emso4es 5-10
l D
- Table 5 i Far-fleid Stress as a Function of Stress Relief Temperature f a,c,e -!
t
' t, i
1 i
i I
4 I
l 8
I i
f 3
5 i
?
t i
t
(
i b
s a
7 T
i ;
i
- l. !
a l !
i J f 4
i A
I 4
).. '
h' s I
i sse14:4woeosos :
5-11 i'
1 I
1 Table 5-3 :
Results of 750 F Doped Steam Tests for !
Nd:YAG Laser Weld Repaired Mockups a,c,e f
f
)
6 t
?
k i-4 1
i 1
d 1
~
i i
i 2s41-5.4seso496 5-12 l l r l
P r ~
Table 5-4 Doped Steam Corrosion Test Results for Tube Sleeve Mockups - Tested Without Axial Load a,c,e 3
4 i
k 1 t 4
4 9
4 o
J
)
k'
, 1 f
- 23414:4Wo60496 - 5-13 [
b
\
. = . _
l g Table 5-5 g Summary of Fabrication Parameters, Temperatures, Stresses and
-' Corrosion Test Results for 3/4 Inch Sleeve Mockups - Tested With Applied Axial Load a,c.e l ,f
?'
4 4
_ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _- _ m__.__ _ _ . _ _ _ _ _ _ . __ _ _ _ _ _ _ _ _ . - - m __ - - - ~ s _m_ _, - -. ---s.-. _ _. __m_. ._-_- nm.___ _ _ _ _. _ _ _ _ _ _ __ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _
.- .=_.. _ .._.- _ . __ _ ._ _
4 i
.4.. >
i, a,c,e J
4 s )
7 i
t i
4 7
d
, 1 j s. ' ,
t J
J t
P e ;
i
]
}
. t 1'
1 1
1 t
i Figure 5-1 Accelerated Corrosion Test Specimen for _ Welded Joint Configuration 2341 4:4awo496 5-15 .
1
. l j
t
- a,c,e I
l I
1 1
1 l
t i
I 1
l i
!' I a
t-i f
- j. i
(
i i
1 l
I Figure 5-2 i
- Test Stand for Fabrication of LWS Mockups Under Locked Tube Conditions 2341 s:4woso4ss 5-16
+ :,
t t
a,c,e
.i.
' ?
1
- o. -
.e
?
4 s
. I.
f 4
4 i 1
. I i
4
' t.
i i
t I
r L
i i
t i
i i
f t
i i
i Figure 5-3 i r
LWS Mockup Corrosion Sample Test Assembly n414:4emso40s - 5-17 i
i S
C a,c,e i 1
I l
t d
1 4
I i
f Figure 5.4 i .
4 IGSCC in Alloy 600 Tube of YAG Laser Welded Sleeve Joint After 109 Hours in 705 'F Steam Accelerated Corrosion Test ,
7 t
23414:4Woso496 5-18.
l 1
h
8.0 INSTALLATION PROCESS DESCRIPTION The outline of the installation processes in Reference 1 applies to South Texas sleeve installation with one exception; the lower sleeve to tubejoint is formed by [
]*** The discussion in Reference 1 provides an overview of the process and the detailed installation process steps are all specified in the individual applicable field service ' procedures that are provided by Westinghouse as part of thejob. The sequence of the installation steps, [
]*^'is optimized to minimize tube far-field stresses.
Heat treatment of the sleeve to tube weld joint is also an important part of the upperjoint formation. The length of tube heated to temperatures above the minimum stress relief temperatureis [
j.
k t
4
- WPF234144WD60496 61
I 7.0 NDE INSPECTABILITY f i
Note: Much of the information of Section 7 of Reference 1 has been updated. Information l in Section 7 of Reference 1 not requiring updating is noted below.
i The welding parameters are computer controlled at the weld operator's station.~ The essential variables, per ASME Code Case N-395, are monitored and documented to produce i trepeatability- of the weld process. In addition, two non-destructive examination (NDE) {
capabilities have been developed to evaluate the success of the sleeving process. One method is used to confirm that the laser welds meet critical process dimensions related to structural requirements. The second method is then applied to provide the necessary baseline data to ,
facilitate subsequent routine in-service inspection capability.
7.1. Inspection Plan Logic The basic tubesheet sleeve inspection plan shall consist of: !
9 A. Ultrasonic Inspection (Section 7.2) [ 1*" or alternate methods (Section 7.4). ,
- 1. . Verify minimum required weld width.
B. Eddy current' examination (Section 7.3) [ ]""
. 1. Demonstrate presence of upper and lower hydraulic expansions. l
- 2. Demonstrate lower roll joint presence.
- 3. Verify weld is located within the hydraulic expansion. :
- 4. Verify Presence of a post weld heat treatment as applicable.
- 5. Verify lack of process anomalies such as blow holes or weld cracking.
- 6. Record baseline volumetric inspection of the sleeve, the sleeve / tube joint, and the parent tube in the vicinity of the welded sleevejoint for future inspections. ;
i C. Weld Process Control [ '}"* !
- 1. Demonstrate weld process parameters comply with qualified weld process l specification.
l I
WPP2341 t4woso49s 7 1-i l
I i
7.2 General Process Overview of Ultrasonic Examination The ultrasonic inspection process is based upon field proven techniques which have been used on laser welded sleeves for 3/4 inch and 7/8 inch OD tubing installed by Westinghouse. l The inspection process developed for application to the laser welds uses the transmission of ultrasound to the interface region (the sleeve OD / tube ID boundary) and analyzing the amount of reflected energy from that region. An acceptable weld joint should present no acoustic reflectors from this interface above a predetermined threshold.
i Appropriate transducer, instrumentation and delivery systems have been designed and techniques established to demonstrate the ability to identify welds with widths below the structural requirements. The entire weld interface (100 per cent of the axial and circumferential extent) will be examined. Acceptance of welds is based upon application of criteria which is qualified by destructive examination oflaboratory produced marginal welds.
7.2.1 Principle of Operation and Data Processing of Ultrasonic Examination The ultrasonic examination of a laser-weld is schematically outlined in Figure 7-1. An ;
ultrasonic wave is launched by application of an electrical pulse to a piezoelectric transducer.
The wave propagates in the couplant medium (water) until it strikes the ID of the sleeve.
Ultrasonic energy is both transmitted and reflected at the boundary. The reflected wave returns to the transducer where it is converted back into an electrical signal which is amplified and displayed on the UT display.
The transmitted wave propagates in the sleeve until it reaches the sleeve OD. If fusion between the sleeve and tube exists, the wave continues to propagate through the weld joint into the tube. This wave then reaches the outer wall (backwall) of the tube and is reflected back to the transducer. The resulting UT display from a sound weld joint is a large signal from the sleeve ID, followed by a tube backwall " echo" spaced by the time of travel in the sleeve-tube-weld assembly (T1 .u). If no fusion between the sleeve and the tube exists, another ,
pattern is observed with a large signal from the sleeve ID followed by a reflection from the sleeve OD. The spacing of these echoes depends on the time of travelin the sleeve alone (T u )-
Additional reflections after the sleeve OD reflections are considered " multiples" of the sleeve OD reflection. These are caused as the sound energy reflected off the sleeve OD bounces back and forth between the sleeve ID and OD, and decays over time.
I yu l WPF2341-7:4WO60496 72
[
ju Criteria. for the acceptance of a laser weld is based upon combination of the observed ultrasonic response at the at the weld surface, the sleeve / tube interface, and the tube OD.
An' automated system is used for digitizing and storing the UT wave forms (A-Scans).
[
]"" The ultrasonic response from the weld is then digitized for each pulse. A typical digitized A-scan is shown in Figure 7-2. Time intervals known as " gates" are set up over the signals ofinterest in the A-Scan so that an output known as a "C-Scan" can be generated. The C-Scan is a developed view of the inspection area which maps the amplitude of the signals ofinterest as a function of position in the tube. A combined C-scan which shows the logical combinations conditions of signals in two gates with respect to predetermined threshold values can also be displayed. Figure 7-3 shows the A, B, C, and combined C-scan display for a weld in a calibration standard.
7.2.2 -Laser Weld Test Sample Results Ultrasonic test process criteria are developed by [
J a.c.e i-Field application requires calibration to establish that the system essential variables are set per the same process which was qualified. Elements of the calibration are to:
. Set system sensitivity (gain).
j -
Provide time of flight reference for sleeve ID, OD and tube OD signals.
Verify proper system function by examination of a workmanship sample. 1 Figure 7-4 depicts a calibration standard for the sleeve weld UT exam.
7.2.3 Ultrasonic Inspection Equipment and Tooling i The probe is. delivered with the Westinghouse ROSA III zero entry system. The various subsystems include the water couplant, UT, motor control, and data display / storage. l 1
, WPF2341-7:49460496 - 7-3
L I
I~
The probe motion is accomplished via rotary and axial drives which allow a range of speeds and axial advances per 360' scan of the transducer head (pitch). The pitch provides a high degree of overlapping coverage without sacri 5cing resolution or sensitivity.
The_ controls and displays are configured for_ remote location in a trailer outside of '
containment. The system also provides for periodic calibration of the UT system on the steam generator platform.
7.3 ' Eddy Current Inspection Upon conclusion of the sleeve installation process, a final eddy current inspection is performed on every installed sleeve to meet the process verification and baseline inspection .
- requirements outlined in Section 7.1 B. The combined Cecco-5/ bobbin probe is utilized towards this end to provide an enhanced baseline inspection without sacrificing data acquisition speed. The bobbin probe provides the inspection to verify the presence and -
location of the expansions, as well as weld location. The Cecco- 5 probe provides baseline examination of the sleeve and tube.
7.3.1 Cecco-5/ Bobbin Principles of Operation The standard bobbin probe configuration consists of two circumferentially wound coils which are displaced axially along the probe body. The coils are connected in the differential mode; that is, the system responds only when there is a difference in the properties of the materials surrounding the two coils.
The Cecco-5 (C5) design operates as a transmit-receive probe. The C5 configuration is designed to provide detection of both circumferential and axial degradation. There are two bracelets of coils, each consisting of an array of transmit-receive sets. Each bracelet is capable of achieving 50 percent coverage of the circumference of the tube. This is due to the fact that there is no coverage directly underneath the coils of a transmit receive probe. For this region, the second bracelet is offset relative to the first to achieve full coverage.
Transmit-receive probes are, by nature of their operational principles, lecs sensitive to lift-off effects than a comparable impedance coil. By virtue of this feature, probes can be designed in such a fashion that the coils do not have to ride the surface of the tube in order to achieve a reasonable level of detectability in a region of geometric change. The coupling of the probes with instrumentation and software designed to take advantage of their specific design features makes transmit-receive probes an attractive technology for the inspection of sleeved tubes.
The ~ calibration standard used for Cecco 5 sleeve inspection includes various axial and circumferential notches as depicted on Figure 7-5. Notches are located in the expansion WPF23417:4946o496 " 7-4
I transitions as well as in the tube and sleeve freespan. Figure 7-6 depicts a 20 channel strip chart plot of the calibration standard. The analysis software allows the data from the two bracelets and the bobbin coil to be displayed in an aligned fashion. The channels may be selected so that data from each sensing point is viewed, enabling viewing of an entire tube circumference on a single screen.
l Figures 7-7 and 7-8 show the response of the Cecco 5 probe to [
]*" notches in the parent tube at the sleeve expansion transition.
Cecco-5 Probes have been qualified to EPRI Appendix H requirements for detection in 3/4 inch and 7/8 inch sleeved tubing.
7.4 Alternate Post Installation Acceptance Criteria No Change From WCAP-13698 Rev. 2 7.5 Inservice Inspection Plan for Sleeved Tubes No Change From WCAP-13698 Rev 2 l
%TF2341-7:49/060496 7-5
l 1
smw ._
%/
mrum
~3
, 7 ~s ~,
N SIA
' /
h/ st.rrvt i
-- JODdT
-TUSC IDCALIZCD vavcang i
, ta , _
NQ JOINT j
T, , , _
coco- 7 l JOINT ,
l l
l Figure 71 l
Ultrasonic Inspection of Welded Sleeve Joint i i
WPF23417:49M60496 76 l
s a,c,e i
?
r I
f I
i i
I E
b i
i 2
Figure 7-2
' Typical Digitized UT Waveform i
.wmut 7:4emio4es . 77 f
4 d, - - , - , , - -- - . -
. l I
a,c,e ;
l 4
l a
')
t h
f i
)
s i
l
?
i
.i i
i 1
P t
I
. I 1
~l Figure 7-3 I
a
. A, B, C, and Combined C-Scan Display for Weld in UT Calibration Standard ]
l WPF2341-7:4WO60496 .7 8-1
8 - A .y- a iJA+3 W 4 AM 4h, =4 4 41 d44 - 4 s de A # k - eA44k. .~ LFs.3-J "A J J4 .4 4 =
t h-f a,c e I
f J
i; J
J i
h I
r P
i
-i b
I
+
~
. :t i
t i
I t
4 5
t t
i
' Figure 7-4 :
t
' UT Calibration Standard.
.l -
i WPF2341-7:49460496 79 t
n -
a o
a,c,e -
l i
i s
Figure 7-5 Cecco-5 Sleeve Calibration Standard l
WPF28417:49 960496 7,yg.
L
L
. 1 i
I l
I a,c,e !
I 1
f
?
1 l
i
.t i
1
~!.
4 i
t I
i i
i i
t t
s s
v f
i t
i i
i i
l t
Figure 7-6 i r
. . i i
Strip Chart Display for Cecca/ Bobbin Data !
WPF28417:49/060496 7 11 i; I
I
4 a,c,e 4
4 4
J i
4 4
1 1 <
1 1
+
J %
r J -
l P
i t
P Figure 7 7 Response of Cecco-5 Probe to 60% OD Axial Notch in Parent Tube Located at Expansion Transition
- WPF2341-7
- 41WD60496 7 12 4
A a,c,e b
\
f I
t i
Figure 7-8 Response of Cecco-5 Probe to 60% OD Circumferential Notch in Parent Tube Located at Expansion Transition i
- WPF23417:49/060496 7-13
- - I l
g- l
8.0 REFERENCES
l 1. WCAP-13698,'Rev. 2, " Laser Welded Sleeves for 3/4 inch Diameter Tube Feedring- l Type and Westinghouse ~ Preheater Steam Generators Generic Sleeving Report", 1 Westinghouse NSD, Pittsburgh, PA, April,1995. ,
- 2. Design Specification 412A24, " Laser Welded Sleeves for 3/4 inch O.D. Tubes of Combustion Engineering Feedring Steam Generators and for Westinghouse Model DS, D4, D5, and E1/2 Steam Generators", Dated 4/30/93.
I j : 3. Letter PCWG-1805, from Westinghouse NATD Fluid Systems Engineering to PCWG l
Members / Alternates, " Category III Approval of South Texas Fuel Upgrade .
Parameters", November 4,1991. l
- 4. WCAP-14107, Rev. O, " Specific Application of Laser Welded Sleeves For Doel Unit 4 ,
Steam Generators", Westinghouse NSD, Pittsburgh, PA, April 1994.
- 5. ASME Boiler and Pressure Vessel Code,Section III, " Rules, For Construction of ,
Nuclear Power Plant Components," The American Society of Mechanical Engineers :
New York, NY,1989.
- 6. Thurman, A. L., " Primary Stress Evaluation of 3/4" Elevated Tubesheet Sleeves",
NSD-JLH-6149, Westinghouse NSD, Madison, PA, May 9,1996.
- 7. USNRC Regulatory Guide 1.121, " Bases for Plugging Degraded PWR Steam Generator Tubes (For Comment)," August 1976.
- 8. WNET-142, Volume 8 "Model D4-2 Steam Generator Stress, Report Divider Plate
! Analysis", Westinghouse Tampa Division, September,1977.
s
- 9. WNET-150, Volume 8, "Model E-2 Steam Generator Stress Report, Divider Plate ;
Analysis", Westinghouse Tampa Division, September,1978.
k i
- 10. Timoshenko, S., Strength ofMaterials, Part II, Third Edition, Van Nostrand Company, Princeton, NJ,1956.
- 11. " Alloy 690 for Steam Generator Tubing Applications", EPRI Report NP-6997-SD, Final Report for Program S408-6, October 1990.
4 WPF23414:4&O60496 8-1 t
4