ML20214J932
| ML20214J932 | |
| Person / Time | |
|---|---|
| Site: | Comanche Peak  |
| Issue date: | 05/31/1986 |
| From: | WESTINGHOUSE ELECTRIC COMPANY, DIV OF CBS CORP. |
| To: | |
| Shared Package | |
| ML19292G307 | List: |
| References | |
| WCAP-11128, NUDOCS 8612020059 | |
| Download: ML20214J932 (43) | |
Text
- -
7 WESTINGHOUSE CLASS 3 WCAP-11128
~
t SHOTPEENING LICENSING REPORT FOR COMANCHE PEAK UNIT 1 MAY, 1996 1 .
1 A
~
t 6 8612020059 861124 5 DR ADOCK 05000 WESTINGHOUSE ELECTRIC CORPORATION
- NUCLEAR ENERGY SYSTEMS l P. O. BOX 355
( PITTSBURGH, PA. 15230-0355 l
SHOTPEENING LICENSING REPORT FOR
- COMANCHE PEAK UNIT 1 TABLE OF CONTENTS 4
1.0 INTRODUCTION
2.0 SHOTPEENING OBJECTIVES AND PROCESS DESCRIPTION 2.1 Objectives 2.2 Process Description (General) 3.0 SHOTPEENING TOOLING SYSTEM DESCRIPTION AND VERIFICATION 3.1 Shotpeening Tooling System Description 3.2 Shotpeening Control System
- 3.3 Hardware and Process Verification Program 4.0 SHOTPEENING PROCESS 4.1 Process Development and Qualification 4.2 Process Verification 4.3 Background of Qualification of Process Parameters 4.4 Magnesium Chloride Testing 4.4.1 Sample Preparation 4.4.2 Test Procedure 4.4.3 Post Test Examination 4.4.4 MgCl2 Test Results and Discussion 4.5 X-Ray Diffraction 4.5.1 Sample Preparaion 4.5.2 Scope of Examination 4.5.3 Results and Discussion w 4.6 Strain Gauge Measurements 4.6.1 Experimental Method 4.6.2 Results and Discussion 4.7 Almen Intensity Studies 4.7.1 Experimental Details and Results 5.0 SHOTPEENING POST PROCESS EDDY CURRENT INSPECTION 5.1 Introduction 5.2 ECT Verification Program Description 5.3 Test Results 5.4 ECT Inspectability Summary 6.0 SAFETY EVALUATION 6.1 Introduction 6.2 Tube Bundle Integrity Evaluation 6.2.1 Shotpeening Program Implementation 6.2.2 Tube Structural and Leak Before Break Considerallons 6.2.3 Post Process Eddy Current Baseline Inspection 6.3 Conclusions l
9486Q:10/051286
1.0 INTRODUCTION
Primary water stress corrosion cracking (PWSCC) of mill annealed Nickel i Chromium Iron Alloy 600 steam generator tubing has been identified as having a potential impact on steam generators. Steam generator tubes have been plugged due to leakage or eddy current indications. As part of an ongoing steam generator PWSCC preventive measure program, Westinghouse combined effor.ts with Framatome to develop and implement a process that provides additional margin against inner diameter (ID) PWSCC occurring in the mechanically expanded portion of the steam generator tube located within the tubesheet and at the transition area between the expanded and non-expanded portions of the tube at the top of the tubesheet by reducing the residual tensile stresses at or near the inner surface of the tube. This reduction is effected by the application of a mechanical stress modification process known as shotpeening. The shotpeening process and equipment have been developed and qualified by Framatome and have been successfully used on both non operating plants and operating plants. Westinghouse has verified the Framatone design through a
! complete verification program for application on the Model D steam generators at Comanche Peak.
The shotpeening process and system design concept is based on observations to date that PWSCC appears to develop in the mechanical roll and roll transition
, zones of the tubesheet region of the steam generator tube. Primary water stress corrosion cracking is intergranular in nature and the cracks are generally short, tight (low leakage rates) and axial in orientation. This report presents a discussion of Framatome's shotpeening process development and equipment design, and Westinghouse's verification and implementation of the preventive measure program. Included is a discussion of the shotpeening process and tooling, the associated parameter acceptance criteria, a summary of the shotpeening tooling and process verification testing, and eddy current inspectability of tubing after implementation of the shotpeening technique.
l, Most importantly, a safety evaluation is provided where it is demonstrated i that the application of the shotpeening PWSCC preventive measure process to a the ID of a steam generator tube does not represent a potentially unreviewed safety question as defined in 10 CFR 50.59 (a) (2).
9486Q:10/051686 1 -1
2.0~ SHOTPEENING OBJECTIVES AND PROCESS DESCRIPTION 2.1 OBJECTIVES
,s ,
The primary objective of the shotpeening program at Comanche Peak Unit 1 is to l i
provide a means of reducing the residual tensile stresses that appear to contribute to PWSCC on the ID of a full depth mechanically rolled steam generator tube on the hot leg end of the tube.
One hundred percent (100%) of the active steam generator tubes will be shotpeened in the hot and cold leg within the tubesheet region. Peening is applied on the tube inside surface of the roll expanded portion the tubesheet area and the expansion transition zone at the secondary side of the tubesheet. The impact of the fluidized shot produces 6 compressive stress state on the ID surface, reducing the potential for PWSCC.
2.2 PROCESS DESCRIPTION The shotpeening tooling is designed and qualified to shotpeen the inside surface of the roll expanded portion of the steam generator tubesheet area and the expansion transition zone of the steam generator tube at the secondary
~
side of the tubesheet. Peening application is to be applied as shown in a
Figure 2-1. Peening will begin within a distance of [ l c.e inches above the tubesheet primary surface. The peening will extend up through the tack-roll to hard roll transition located at the primary face, through the tubesheet region, and through the upper hard roll transition, to an elevation of approximately [ l ac.e inches above the top of the tubesheet.
[
h 9486Q:10/051686 2-1 f
j a.c.g The efficiency of the process is realized with this parameter selection as it ,
allows for maximum rapidity of the treatment.
v The shotpeening process involves the introduction of high compressive stresses on the inside tube surface by peening the surface with small [
35 ,c.e.g.
[
v j a,c.e.g 1
The shotpeening process is explained in greater detail in Section 3.2 of this report. For the Model 04 steam generators at Comanche Peak, the shotpeening process is applied to mill annealed Inconel 600 tubing with nominal dimensions of .75 inch 00 and 43 mil wall.
4 9486Q:10/051986 2-2
2.3 SHOTPEENING PROCESS PARAMETER DETERMINATION The selection of process parameters for field implementation of the 1 shotpeening process has been based on the developmental testing and plant application performed by Framatone and the Westinghouse verification testing program. Process parameter verification was accomplished by applying the shotpeening action to specially prepared tube samples configured for conditions representative of the tubesheet region. Tube specimens were examined for surface conditions in the peened and unpeened states to obtain a visual surface comparison. Several groups of specimens were exposed for controlled time periods to accelerated, environmentally severe stress corrosion tests in boiling magnesium chloride with stainless steel samples for
~
parameter verification. Since field tooling was used, these tests serve also as the basis for verification of the operation of the production tooling.
Similar tests were used as a basis for final confirmation during field tooling verification. Shotpecning process and tooling system verification is discussed in detail in Section 4.0 of this report.
m l
l 1
J i
! 9486Q:10/051686 2-3
e
~
b,c.e g 1
o o
1 l _
l l
Figure 2-1 Shot Peening Application
1 l
3.0 SHOTPEENING TOOLING SYSTEM DESCRIPTION AND VERIFICATION
. )
3.1 SHOTPEENING TOOLING SYSTEM DESCRIPTION s -
\
Equipment has been designed and developed for robotic application and automatic control of the shotpeening nozzle assembly in steam generator tubes. The shotpeening operating equipment and operating principle are shown in Figures 3-1 and 3-2.
The shotpeening tool assembly includes a [
?
j a c.e.g A predetermined weight of microbeads is placed in the shot generator, where the shot is ( ,
o
~
j a c.e.g 9486Q:10/051686 3-1
Following each tube treatment the beads are weighed to verify recovery. The bead shot is drawn back into the bead generator by the vacuum pump while the a nozzle head is positioned under the next tube to be treated.
The vacuum bead recovery system and tooling design minimize loss of the bead shot to the channel head area. [
]I Procedures require appropriate covers over the steam generator nozzles to close off the primary system. A complete inspection and cleaning of the channel head surface will be performed after.shotpeening.
3.2 SHOTPEENING CONTROL SYSTEM DESCRIPTION The shotpeening control system monitors operation of equipment to verify appropriate travel speed, pressurization, and mass flow of shot.
The controller unit is based on a programmable logic control technique that provides commands to relays. The relays, in turn, regulate operation of all equipment. The units can be operated automatically for the complete peening .
cycle or manually stopped and started to complete an interrupted automatic cycle of the shotpeening process.
The DC motor on the [ ]a,g has an encoder and tachometer that provide two different ways to measure speed. The tachometer is employed to give independent speed measurements of the pusher / puller system, which is internally monitored by the control unit microprocessor. The operator at the console can also verify peening time, vacuum pressure, tool contact with the tubesheet, and shot weight.
i .
l 9486Q:lD/051686 3-2
3.3 HARDWARE AND PROCESS VERIFICATION PROGRAM The shotpeening equipment and process has been qualified and successfully
, s implemented by Framatome on both operating and non-operating plants.
Westinghouse has verified the tooling operability and process parameter selection for application at Comanche Peak. Process verification testing was accomplished utilizing actual field tooling systems. All field tooling was tested to meet pre-established functional requirements to prove that the shotpeening process and integrated tooling system provides an appropriate technique for modifying residual stress distribution.
A key element of the qualification program is the acceptance of the hardware for field readiness. Field tool assemblies, completely assembled and integrated with the delivery system, were tested to verify acceptance of the process parameters. Testing was conducted on a steam generator channel head mock-up, representing the channel head and tubesheet geometry.
The tooling functioned to established design criteria. , The design criteria I included application of the shotpeen process to the inside surface of the roll expanded portion of the tubesheet area and to the expansion transition zone at the secondary side of the tubesheet.
Tooling acceptability included verification of proper ID surface coverage and penetration within the identified range. The parameters that required verification are [
j a.c.e A test stand containing process control samples was used to monitor the consistency of the shotpeening applications. The test samples were set up to provide monitoring of the shotpeening [
j a,e,g The test stand contains two Almen strips and two pre-split sample tubes. The Almen Strip is a standardized strip of spring steel, which, when exposed to the shot stream, bows convex to the stream. The height of the resultant arc provides a measurement of peening intensity. Almen intensity has been 9486Q:10/051686 3-3 i
l
demonstrated to correctly characterize the ef fect of shotpeening on the tube.
The split tube sample provides visual verification of the shotpeening elevation and axial coverage. -
Each tooling system to be used at site was tested using the process control samples for verification of operability. The evaluation and acceptance criteria for the shotpeening system and process are listed in Table 31 along with an objective description for each requirement.
9486Q:lD/051686 3-4
TABLE 3-1
SUMMARY
OF EVALUATION / ACCEPTANCE CRITERIA TESTING / EVALUATION
- REQUIREMENTS OBJECTIVES MINIMUM ACCEPTANCE A. Mechanical Performance of the Complete System
~
a,c.e b
94860:10/051686 3-5
l l
i TABLE 3-1 (Continued)
SUMMARY
OF EVALUATION / ACCEPTANCE CRITERIA TESTING / EVALUATION REQUIREMENTS O_BJECTIVES MINIMUM ACCEPTANCE B. Shotoeenina Distance. Axial Coveraae and Effectiveness a,c.e I
l i C.
af l
9486Q:10/051686 3-6
i i
TABLE 3-1 (Continued)
SUMMARY
OF EVALUATION / ACCEPTANCE CRITERIA TESTING / EVALUATION
. REQUIREMENTS OBJECTIVES MINIMUM ACCEPTANCE D. Shotoeenina of Process Verification Samples a,c.e e .
b 9486Q:10/051686 3-7
'a,c.e 9 .
o
.4 1
4 e
i
~
l i
J Figure 3-1 Shot Peening Operating Principle 1
a,c e,g
. - 7 Figure 3-2 Shot Peening Principle Diagram
. . = - - - _ _ ___
a,c.e g l
. =
1 i
l i
1 Figure 3-3 Shot Peening Spray Nozzle
4.0 SHOTPEENING PROCESS 4.1 PROCESS DEVELOPMENT AND QUALIFICATION
,=
The shotpeening process and tooling has been developed and qualified by Framatome. The qualification program supported optimization of the 4
shotpeening parameters and development of the tooling and process. The program included a variety of testing to characterize the effects and efficiency of shotpeening. Parameter optimization was achieved through Almen intensity measurements. Process effectiveness was characterized through accelerated stress corrosion testing, strain gage measurements, and X-ray diffraction measurements of residual stress.
Almen intensity is the standard parameter for measuring shotpeening intensity. Almen intensity measurements were applied to determine and verify the optimal parameters for shotpeening. The shotpeening parameters considered were the bead material, the bead size, and bead velocity and shotpeening time. [ ]a,c.e.g j
diameter have been selected for 3/4 in. tube treatment. Testing showed that this bead type does not become embedded in the tube wall or result in fragmentation. Functional testing utilizing Almen Intensity measurements also determined the injection pressure and vacuum pressure equilibrium, the saturation time, as well as the optimal nczzle and orifice size.
i Based on the selected parameters, testing was done to check process efficiency and characterize the effect of shotpeening as a function of applied Almen intensities. The testing consisted of outer wall stress measurements and inner wall stress profile. Strain gauge measurements showed that changes in the OD tensile stresses due to shotpeening were acceptably low. Magnesium chloride stress corrosion tests on 3/4 in, stainless steel tubing demonstrated the efficiency of the process in eliminating stress corrosion cracking (SCC) of the inner tube wall. Inner tube wall X-ray diffraction measurements of shotpeened tube samples showed that the initial tensile stresses are
<. completely eliminated by shotpeening. Based on these tests, a specified Almen intensity range was determined for optimal compressive stress achievement.
l 9486Q:10/051686 4-1
After the results of the tests demonstrated the ef fectiveness of the process, shotpeening was applied to steam generator tubes at three operational plants. -
The field experience provided progressive implementation of enhancements of the tooling and process.
4.2 PROCESS VERIFICATION Following the Framatome qualification program and field application,
- Westinghouse conducted a test program to verify the efficiency of the selected shotpeening parameters, process, and tooling. This program consisted of 4 types of stress index characterization tests that included
- 1) MgCl 2 immersion tests of peened and unpeened stainless steel tubes expanded into Inconel collars; 2) X-ray diffraction of residual stress distributions; 3) strain gauge measurements of changes in hoop and axial 00 stresses due to the i shotpeening; and 4) Almen intensity measurements. The testing and results are discussed below, preceded by a brief discussion of the logic in selecting surrogate Type 304 tubing for the stress indexing tests.
4.3 BACKGR0llND OF QUALIFICATION OF PROCESS PARAMETERS The objective of shotpeening is to reduce residual tensile stresses in the .
Inconel 600 tubing to stress levels that may not be susceptible to PWSCC in the primary coolant at normal operating temperatures. The stress dependence -
of primary water SCC (PWSCC) is not well quantified because of the long i
" incubation" periods required for PWSCC at operating temperatures and because most experiments have been conducted on very highly strained samples under thermally accelerated (360 degrees Celsius) conditions. Field experience with mill annealed Inconel 600 tubing indicates that PWSCC is a concern only in those regions where fabricational stresses exceed the yield strength, such as
! at mechanically expanded tubesheet transition regions from expanded to unexpanded tubing. The reduction of residual surface tensile stresses by peening of expanded surfaces should therefore reduce the susceptibility of these regions to PWSCC.
i 9486Q:10/051686 4-2 I
~ - - - , - - - - - - - - , - . . - , . _ . _ _ _ _ _ . . . _ _ _ _ _ _
_- ~. - - .- . .-.
Experience with tensile stress reduction has indicated that direct measurements of residual tensile stresses are typically less useful than a direct indexing of the resistance to SCC. The direct indexing of additional resistance to SCC has been particularly successful ir. shotpeening of austenitic stainless steels to mitigate against chloride-ion SCC. Type 304 stainless steel is so susceptible to this attack that the well-known laboratory test of boiling 42 percent MgC1 2can crack even modestly stressed Type 304 samples in a few hours.
Mill-annealed Inconel 600 remains generally resistant to SCC degradation that short-tern tests such as MgC1 for Type 304 (or mercurous nitrate for brass, 2
~
etc.) do not exist. To index either residual tensile stress reduction or increased resistance to SCC in Inconel 600 in short-term tests, recourse is made to either (1) sensitization of the Inconel 600 followed by exposure at room temperature to certain aqueous sulfur oxyanion solutions such as tetrathionate or (2) the use of a surrogate material such as Type 304 stainless steel with the MgC1 test.
2 One limitation of the use of sensitized Inconel is the need to identify a heat
! that will reproducibly sensitize and exhibit high susceptibility to the intergranular SCC process in exposure to the aqueous sulfur species. Such an identification of heat and sensitization heat treatment can be a time-consuming critical factor if large numbers of specimens must be prepared in a short duration for a wide matrix of testing with assurance that there are not inherent materials variables. Sensitized Inconel was therefore not adopted for the present program.
Type 304 stainless steel " surrogate" tubing was used for the shotpeening
- parameter development of the present effort. Westinghouse experience with this material and with MgC12 testing is extensive from rotopeening l
development programs. For the heat of Type 304 stainless steel in those i
programs, a calibration was developed that correlated the time-to-failure in MgC1 2 with applied tensile stress in laboratory annealed tubing. This test material experience and the data base favored the use of the surrogate Type 304 stainless steel tubing for the present shotpeening qualification effort.
i I 9486Q:10/051686 4-3 i
4.4 MAGNESIUM CHLORIDE TESTING 4.4.1 SAMPLE PREPARATION The test samples consisted of reference 3/4 in. 00 x 0.043 in. wall thickness Type 304 tubing, roll expanded into holes in a thick-walled Inconel 600 cylinder, or " collar", simulating the tubesheet. The hole diameter in the collars was [ j a.c.e, which is the specified high tolerance size for
- normal tubesheet holes in Model D steam generators. To double the number of representative transitions per sample, the configuration depicted in Figure 4-1 was adopted for the specimens of the present program. The first roll step was expanded at the bottom to produce a normal transition that began within the collar. The rolling tooling was then reversed, and the second and third roll steps were made, resulting in a second normal transition that began within the collar. The " skip" between roll steps No. 1 and No. 2 was then repaired with roll steps Nos. 4 and 5, as shown in Fig. 4-1. All expansions were performed using shop equipment and procedures in accordance with Westinghouse Steam Generator Materials Procedure 5.1.1, " Procedure for Expanding Tubes by the Mechanical Rolling Process".
The expanded specimens that were to be shotpeened were dimensionally inspected ,
and sent to the Training and Readiness Center where production shotpeening equipment is assembled and qualified. The specimens were shotpeened in a test .
stand that simulated a Comanche Peak steam generator tubesheet, utilizing production equipment and the shotpeening procedures that will be applied at the site. [.
\
94860:10/051686 4-4
P j a.c.e.g All peened specimens and eight unpeened control specimens were then comitted to the NgC1 test as described below.
2 4.4.2 TEST PROCEDURE l In general, the procedures and equipment used are as described in ASTM G 36-73. The pre-heated specimens were immersed in the boiling MgC12 solution when the temperature was 154 to 155 degrees Celsius. Time from start of i'amersion to return of bath to temperature was approximately 15 minutes for the pre-heated samples. Samples were boiled for 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />.
For the heat of 304 stainless steel used in the present program, tests with C-rings indicated that reproducible stress corrosion cracking (SCC) occurred in boiling MgC1 in 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> imersions for all samples stressed to 30 ksi 2
and for all samples of the present program at a stress level of 20 ksi.
j Relatively few C-rings that were stressed to 10 ksi cracked in 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />, however. This behavior is consistent with uniaxially loaded tensile specimens of a different heat of 304 stainless steel in which an effective 24-hour stress threshold of 13 ksi applied for MgC12 tests. A second previous heat l
of 304 stainless steel also exhibited an effective 24-hour stress threshold above 10 and below 20 ksi in C-ring calibration tests. The heat-to-heat
- reproducibility of effective threshold stresses indicates that the effective threshold stress for SCC in 24-hour MgC1 tests of the present samples was 2
also approximately 13 ksi.
4.4.3 POST-TEST EXAMINATION i
Following the 24-hour exposure to MgCl 2
, each sample was first split axially into two approximately equal halves and the tubing removed from the collar.
I, One tube half was then flattened and examined on its 10 and 00 surfaces at 10X 9486Q:10/051986 4-5
with the stereomicroscope; it was then further examined by bending to strain the OD in tension,.followed by re-examination under stereomicroscopy for .
circumferential and axial cracks. Subsequently, the mating half sections of nearly all samples were also flattened and examined to increase the statistics of the results.
4.4.4 MgC1 2
TEST RESULTS AND DISCUSSION Table 4-1 presents the details of the MgC1 2 test results of the eight unpeened and the 24 shotpeened specimens. Of the latter, twelve represented transition region and twelve represented tubesheet region configurations and peening parameters. Also included are the results of C-rings stressed to 10, 20 and 30 KSI, and immersed in a kettle with three test specimens. (The purpose of the C-rings was to confirm the potency of the MgC12 test solution; each of the 30 KSI stressed rings and at least two of the three C-rings stressed to 20 KSI in each kettle must have cracked to validate the test. The C-rings stressed at 10 KSI would not be expected to crack frequently since the stress threshold for cracking of the type 304 heat used was somewhat greater than 10 KSI. Nevertheless, a low frequency of cracking would not be unusual, given the statistical nature of stress corrosion cracking.) ,
The critical region where high residual stresses could be present (and result -
in cracking in the MgC1 2 solution) are the transitions at the top and bottom of each specimen. Both the 10 and the 00 at these locations were typically cracked, after 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> exposure, in the expanded but unpeened condition.
Table 4-1 details the observation by treating separately each axially split
" half" section, denoted (1) and (2), each of which had two transitions; then combining the two halves, (1) and (2), for a total of four transition areas per specimen. The column "By Specimen" simplifies the findings by noting the number of cracked transitions per specimen, regardless of whether a transition -
was cracked in both halves or only one half of a specimen.
I 9486Q:10/051686 4-6
As shown, both transitions of each unpeened specimen were cracked on both the
- , ID and the 00. (One of the transitions of Specimen TBX-014 was uncked on only one of the half sections). This is the condition against which the
, peened specimens are to be compared. As shown, none of the transitions of any of the peened specimens were cracked on the ID, indicating the benefit of the peening process, over the range of parameters applied. Thus 1/2T, or 100%
coverage, was as effective as the reference T, or 200% coverage, or 2T, which is 400% coverage.
On the 00, all of the transitions were cracked, indicating residual stress levels above the threshold, but this situation was present in the unpeened condition also. Peening of the ID inevitably results in some higher OD residual stresses, so it is not reasonable to expect a lower frequency of OD cracking after peening. The MgC1 test, as applied, is not capable of 2
assessing the incremental 00 tensile stress induced by the peening but the level is low, 6 KSI or less, as shown by the strain gauge' readings.
The potency of the test solutions was confirmed by the C-ring performances; each of the 30 KSI and 20 KSI stressed C-rings cracked as did several of the 10 KSI stressed C-rings, indicating an aggressive environment.
i The conclusions of the MgC1 tests are that the reference shotpeening 2
parameters reproducibly reduce the ID residual stresses to below the threshold i level, and that this g rotection is afforded over a range of peening times, without a significant increase in the severity of the OD stresses.
, 4.5 X-RAY DIFFRACTION The purpose of X-Ray diffraction measurements is to obtain a profile of the tube wall stresses resulting from the specified application of the process.
Residual stresses and the degree of cold working on the shotpeened samples are measured from the shif t or broadening of dif fraction peaks in the X-Ray lines from beams directed under controlled conditions on layered sections of the
- , sample surface.
9486Q:10/051686 4-7
This technique is used to evaluate the process effectiveness on the tube internal surface by providing a measurement of the compressive stress layer.
The results are evaluated in conjunction with the data obtained for 00 strain gauge measurements (Section 4.6) and the conclusions on MgCl2 samples as discussed in paragraph 4.4.4.
- 4. 5.1 SAMPLE PREPARATION
(
The Westinghouse X-Ray diffraction samples consisted of mill annealed Inconel 600 tubing of 3/4 in. OD x 0.043 in. wall roll expanded into carbon steel collars, simulating the tubesheet. The collars were geometrically similar to those used for MgCl specimens. Sample preparation was performed in 2
accordance with Westinghouse Steam Generator Materials Procedure 5.1.1 and the m
- rolling was done as described in paragraph 4.4.1 of this report. <
Prepared samples were shotpeened using functionally qualified tooling to be used at Comanche Peak. System operating parameters were the same as for the MgC1 samples and are described in Table 4-2. Two samples were prepared.
2 One sample simulated tubesheet transition conditions and the other represented the process results obtained in the tubesheet.
4.5.2 SCOPE OF EXAMINATION The requirements and objectives for X-Ray diffraction testing have been defined previously in Table 3-1, item D-3. The intent of the testing is to substantiate the Framatome qualification results that show a depth of' penetration in excess of two mils and a significant peak 10 surface compressive stress on samples of 3/4 in Inconel 600 tubing using the specified peening intensity of ( ]' ' C ' D 9 Measurements have been made in both the expanded and unexpanded portions of the Westinghouse prepared samples.
4.5.3 RESULTS AND DISCUSSION Independent measurements made by Westinghouse indicate that for both expanded and unexpanded specimens, the shotpeening process results in a residual stress b
9486Q:10/051686 4-8
profile that exceeds a penetration depth of 2 mils and there is no significant difference on specimens produced in the transition and tubesheet area of treatment. The results also indicate that a significant compressive stress exists on the tube ID surface after the application of the specified i shotpeening intensity.
l
. 4.6 STRAIN GAUGE MEASUREMENTS An important measurement of process effectiveness is the resulting stress I changes on the 00 after the application of the specified shotpeening intensity on the tube 10 surface. Processed parameters were selected to limit the 00 stress changes to below 5.8 ksi tensile. The resulting stress condition on I the tube 00 surface was verified independently by Westinghouse using the specified parameters selected by the Framatome process qualification.
4.6.1 EXPERIMENTAL MCTHOD In order to determine the resulting stress addition to the 00 surface both circumferential and longitudinal stresses were calculated after the l
application of shotpeening using the specified intensity on the ID surface.
Strain gauge rosettes were mounted on the outside surface of mill annealed 3/4
!' in Inconel 600 tubing and the resulting strain was measured for the referenced parameters used to provide 10 surface stress indexing samples. A direct measurement of resulting 00 stress after the application of shotpeening is best obtained on unexpanded tubing that is relatively free of manufacturing stresses. This measurement provides an indication of the stress change without the interaction of residual stresses existing prior to the shotpeening i application.
The system operating conditions for the strain gauge measurements are given in Table 4-3 and are a result of the requirements and objectives as stated in Table 3-1, item D-2.
94860:10/051686 4-9
4.6.2 RESULTS AND DISCUSSION The calculated stress change after the application of specified shotpeening parameters was found to be below the recommended limit of 5.8 ksi on all .
unexpanded tube sections. An example of the results is provided in Figure 4-2. The data also showed similar stress values for both the longitudinal and circumferential direction. These results are consistent with the Framatome qualification of the process.
4.7 ALMEN INTENSITY STUDIES The consistency of shotpeening is measured by Almen intensity. This technique is un'iversally applied for methods of mechanical surface treatment where stress redistribution is a result of cold working by controlled particle interaction. It is also the standard parameter by which shotpeening system performance can be evaluated. Process development and qualification involved a detailed evaluation of the methods used to control intensity for the individual systems applied in the shotpeening process. The end results are to relate all parameter evaluations to measurement of intensity at the work surface.
.\
Almen intensity readings are a function of the momentum of the particles used in the treatment and the integrated exposure time of the application. The ,
control of the resulting' intensity is system specific. Procedures have been developed to calibrate the individual systems to obtain controlled Almen values.
4.7.1 EXPERIMENTAL DETAILS AND RESULTS Each shotpeening system applied at Comanche Peak was functionally qualified for adequate control of shotpeening intensity. Testing was performed to show individual system capabilities to maintain the specified range of intensity .
for shotpeening. This was accomplished by the development of intensity versus system nozzle injection pressure curves. Figure 4-3 shows a representative .
curve for one of the systems used at Comanche Peak. This curve is used to l
9486Q:10/051686 4-10 1
1
determir.e the nominal nozzle injection pressure needed to obtain the average target intensity of [ ]*9 This nominal injection pressure is then
- used to obtain intensity measurements as a function of nozzle application time as shown in Figure 4-4. Similar curves have been developed for all systems to be applied at Comanche Peak. Evaluation of the data for all tested systems shows that the specified intensity of application can be correctly maintained.
I i
1 I
v l
94860:10/051986 4-11
TABLE 4-1 SNOTPEENING PROCESS VERIFICATION TESTS T 24-N00R BOILING NAGNESIUN CHLoeIDE RESULTS.
REFERENCE TWD-SPEEB P90 CESS *
[
E It RESULTS C-alNGS 08 RESULTS D Num6er of SCC Transitions Number SCC Number of SCC Transttless O Number of Tested Transitions Number Tested Number of Tested Iransitions
,0 (la same pot with transitions)
Sample Peening By Nelves By By Helves By a,b,e.g Number Dete* m m (1) +(21 Seecimen 30 ksi 20 ksi 10 ksi m M (11+(21 Seec ture 1
4
)
i l
i em m
- 6 . g
TABLE 4-2 o
SYSTEM OPERATING PARAMETERS FOR STRESS INDEXING SAMPLES a,b,c.e,f i
l 94860:10/051686 4-13
TABLE 4-3 SYSTEM OPERATING PARAMETER FOR 00 STRESS MEASUSEMENTS
' a,b,s,e,f i
4 d
e I
)
i
) .
1 i
,i
- 94860:10/050586 4-14
t
_ b,c.e l
l d
i i
. l 1
1 w -
a Figure 4-1 Rolling Sequence for Fabrication of Stress Gauge Specimens
-~~,,v- ,_,------e,--,-,n--.----.,- , - - , - - - - - , - - - - - - . - - - - - - - , , . - - - -
o.
.o.
e '
\
4 t
i i
l 1
1 e
v i
I i
1 ,
v L
j
?
, 1 I
r Figure 4-2 Changes in OD Stress with Peening Time for unexpanded Tubes . - - - -
1 r
M
)
i 4
1 a,
4 d
i f .
4, ,
N 9
9 1
n a,b,e 4
4 1 -
t -
I_
i i
i 1
4 e i
}
i i
1 d 1
i i
1 1
i i.
I i-i
. t 4
Figure 4-3 Almen Intensity vs Injection Pressure _
i
e.
n.
I i
. \
+
4 i
s .
1 i
- t t
?
f t
e l
r Figure 4-4 Almen Intensity vs Shotpeening Cycles Transition
5.0 SHOTPEENING POST PROCESS EDDY CURRENT INSPECTION
5.1 INTRODUCTION
4 ;
i Eddy Current (EC) testing was conducted to determine the impact of shotpeening on the inspectability of steam generator tubing. The objective of the testing was to determine whether shotpeening introduces additional signals or alters the existing baseline signature of the steam generator tubing.
5.2 EDDY CURRENT VERIFICATION PROGRAM DESCRIPTION j The testing was conducted on mill annealed Alloy 600 tubing with dimensions of 3/4 in. OD by .043 in, wall thickness, which represent the material and size of tubing in the Comanche Peak Model 04 steam generators. Two types of tube configuratons, unexpanded and mechanically roll expanded, were included in the eddy current evaluation. All samples were without cracks or defects. EC
< inspections were performed using a [ ] diameter bobbin probe and a MIZ-18 instrument. Test frequencies of [
]' d were used. Evaluations were based on the [ ]" # data.
. EC testing was done before and af ter shotpeening to ascertain whether shotpeening would distort the baseline EC response of a steam generator
. inspection. In the peening operation, a section of each tube was left ,
unpeened to create a transition of peened-to-unpeened tubing in the inspected region of each tube. This provided an evaluation of the effect of shotpeening l to the Alloy metal.
l l Four sample configurations had been mechanically roll expanded into clamshell I collars and the collars removed. This testing evaluated the effect of shotpeening on the eddy current signature of a tube section with l characteristics representative of the rolled tube geometry in the tubesheet
. . region of the steam generator.
1=
1 1
d 5-1 9486Q:10/052786 I
Four samples were unexpanded straight sections of tubing. This testing evaluated the ability of eddy current to detect any change in the tube due to shotpeening, without the complicating influence of other tube signals f rom roll transitions and roll overlaps that might mask a signal ef f ect due to .
peening.
5.3 TEST RESULTS For all EC examinations that were performed on these Alloy 600 tube specimens, no change in signal could be attributed to the shotpeening process for the baseline signature. In the test of the four expanded tube samples, peening l produced no detectable change in the characteristic signals associated with the rolled tube geometry. In the more limiting test of unexpanded tubes, there was no detectable eddy current signal associated with the peened-to-
- unpeened transition.
l 5.5 EC INSPECTABILITY
SUMMARY
On the basis of this testing, shotpeening has been shown to cause no alteration of baseline EC signatures for the uncracked MA Alloy 600 tube.
This conclusion applies also to the transition of peened-to-unpeened tube area. Therefore, it is concluded that standard EC inspection systems will
- yield baseline results essentially equivalent for steam generator tubing before and after shotpeening.
i I
i l
l 94860:10/042586 5-2
_ _ _ _ _ _ . _ _ _ _ . - _ _ _ - - - _ _ _ _ _ __ .._._ -_ _ _ _ _ - _ _ _ _ _ _ _ _ _ . . _ . _ . _ - _l
COMANCHE PEAK UNIT 1 SHOTPEENING PROGRAM SAFETY EVALUATION
~
6.1 INTRODUCTION
Primary water stress corrosion cracking (PWSCC) of mill annealed Inconel 600 I steam generator tubing has been' identified as having a potential impact on steam generator availability. Steam gerierator tubes have been plugged due to leakage or eddy current indications of PWSCC. PWSCC has been investigated in the laboratory and is attributed to the combination of three inservice factors: high operating temperatures, susceptibility of tube microstructure, ,
and high local strain conditions. The potential censequences of PWSCC that result in primary-to-secondary leakage are secondary side contanination and
- the possibility of exceeding the plant technical specification' allowable leakage limit. TheseeventscouldproduceunneceEsaryplantshutdownsandthe potential of having to derate a plant's electrical output if the number of
~ tubes plugged exceeds a plant's available tube plugging level margin.
i Eddy current-inservice inspection results from some operating Model D steam generators indicate the possibility of tube degradation in the mechanical hardroll region of the tube below the top of,the tubesheet because of residual stresses. Many of the signals observed have been generally classified as
,' undefined because of the distortion observ'ed in th,e lissajous figures. This signal distortion has been presumed to arise from coincidence with the roll I' step signals and possibly magnetic interference. Nevertheless, possibility of SCC within the tubesheet region warrants a proactive approach to maintaining i
margin of the steam generator. Texas Utilities, Generating Company has implemented a PWSCC preventive measure programilor the tubesheet region of the j i Comanche Peak Unit i steam generators. One hundred percent of the tubes in l all steam generators will be shotpeened in the hot and cold leg in the tubesheet region.
For full depth mechanically rolled tubes, an effective means of minim'izing
~
potential PWSCC is to modify the residual tensile stresses in the hardroll region of the steam generator tube. This is accomplished through the 4
reduction of residual tensile stresses at or near the inside surface of the tube resulting from the tube expansion process and is affected by the ,
! application of a mechanical stress modification process known as compressive 1
9486Q:10/050586 6-1 t
d stresses on the inside tube surface by peening the surface with small nickel alloy beads. The shotpeening process is applied on the length of the tube ,
i extending f rom near the tube end at the primary face of the tubesheet. A portion of the unexpanded tube above the top of the tubesheet is also .
! shotpeened. The following safety evaluation is provided to demonstrate that the application of the shotpeening process to the inside diameter of the hot and cold leg of the Comanche Peak Unit 1 steam generators does not adversely impact steam generator tube integrity and therefore does not represent an unreviewed safety question pursuant to 10 CFR 50.59(a) (2) criteria.
6.2 TUBE BUNOLE INTEGRITY EVALUATION 4
6.2.1 SHOTPEENING PROGRAM IMPLEMENTATION The shotpeening program at Comanche Peak Unit 1 provides an effective means of reducing the residual stresses that contribute to PWSCC occurring on the inside diameter in the tubesheet region of a full depth mechanically rolled steam generator tube.
L The shotpeening equipment and process has been qualified and successfully implemented in Europe by Framatome on both operating and non-operating ,
plants. The Framatome qualification program supported optimization of the shotpeening parameters and the development of the tooling and process. ,
I Parameter optimization was achieved through Almen intensity measurements.
i Process effectiveness was characterized through accelerated stress corrosoin cracking testing, strain gauge measurements, and X-ray dif f raction
! measurements of residual stress.
Westinghouse has verified the tooling operability and process parameter selection for application at Comanche Peak Unit 1. Westinghouse process
{
verification testing was accomplished utilizing actual field tooling systems, f
i All field tooling was tested to meet pre-established functional requirements ,
to prove that the shotpeening process and integrated tooling system provide an appropriate technique for modifying residual stress distribution. ,
- Specifically, process parameter verification has been accomplished by l shotpeening specially prepared tube samples that have been configured for i
9486Q:10/042986 6-2 i
i
~-- _ _ . . . - . .. . - . - _ . . . _ . . _ _ _ _ _ _ _ _ _ , _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ , _ _ _ _ , _ _ __
1 conditions representative of the tubesheet region. Tube specimens were examined, for surface conditions in both peened and unpeened states to obtain a visual surface comparison. Moreover, specimens were exposed to MgCl2 stress indexing corrosion testing for controlled time periods to demonstrate the efficacy of the shotpeening process in reducing susceptability stress corrosion cracking (SCC) of the inside diameter tube wall. Process l
effectiveness was also measured through strain gauge measurements and X-ray diff raction measurements of residual stress.
6.2.2 TUBE STRUCTURAL AND LEAK 8EFORE BREAK CONSIDERATIONS Upon application of the shotpeening process, steam generator tube integrity
. has been assessed at several tube locations within and immediately above the tubesheet region of the Comanche Peak Un'it 1 steam generators. Particular attention focused on the effects of shotpeening on: 1) the tube-tol ubesheet t
weld; 2) the hardroll expanded tube throughout tha tubesheet; 3) the expansion transition zone of the full depth hardroll; and 4) tube outside diameter stress generated in the unexpanded region immediately above the top of the tubesheet. These aspects as well as leak before break considerations are addressed in the following paragraphs.
' The effect of peening over the inside surface of the tube-to-tubesheet weld has been evaluated. The weld performed during steam generator fabrication is a structural weld thatris conservatively considered to be acted upon during design basis loading of the tubing. ,In the stress analysis of the Comanche Peak Unit i steam generators, no credit is taken for any added structural integrity to the tube bundle for mechanical hardroll expansion of the tube in the tubesheet. Af ter fabrication, each weld was inspected per ASME Boiler and Pressure Vessel Code requirements. The Code does not prohibit the use of shotpeening to enhance service performance for austenitic or high nickel alloy
' materials. The ASME Code does not require reanalysis of the shotpeened weld.
The tube end on the model 04 steam generator is recessed from the primary tubesheet surface cladding and the tube-to-tubesheet weld is a standard recessed weld on the ID of the tube. The shotpeening process involves light a
contact of shot with the inside diameter surface of the weld. However, the 9486Q:10/050586 6-3
I structural integrity of the weld is not compromised. Peening is a shallow application on the weld surface that does not involve structural deformation, .
removal of material, or change in weld dimensions.
Throughout the mechanical hardroll expanded portion of the tube in the tubesheet, the presence of the tubesheet acts to constrain the tube and complement its integrity in that region by precluding tube deformation beyond its expanded cutside diameter, i.e., neither tube rupture or collapse can occur. Therefore, the compressive stress loading due to the shotpeening process cold working the inside surface of the tube wall will not result in tube deformation beyond the presently expanded diameter. Secondly, based on tubesheet hardroll tests, the hardroll joint is expected to be leak tight,
. i.e., the plant would not expect to experience leak sources emanating f rom the hardroll expanded region of the tubesheet. Thus, leak tightness is not compromised. These factors indicate that application of the shotpeening process within the mechanical expanded portion of the tube in the tubesheet up to the expansion transition zone does not compromise tube integrity.
Moreover, throughout the mechanical roll expanded portion of the tubesheet, the results of MgC1 2 stress indexing corrosion tests have demonstrated a significant benefit obtained through the reduction of inside diameter tube '
wall stresses by shotpeening. As mentioned above, based on selected parameters, testing (MgC1 2 testing on stainless steel tubing) has been done ,
by Westinghouse and Framatome to verify process efficacy and to characterize the effect of shotpeening as a function of applied Almen intensities. Inside tube wall X-ray diffraction measurements of shotpeened tube samples have revealed that the initial tensile stresses are completely eliminated by shotoeening.
Relative to shotpeening of the expansion transition region of the tube, cracking of the roll transition in service is essentially limited to the length of the transition region. For mechanical rolling, this length is on .
the order of 0.25 inch. The high stresses and significant plastic strains needed to crack Inconel 600 in primary water are expected only over this ,
length. While shotpeening has a large effect on surface stresses, this effect 94860:1D/050586 6-4
diminishes quickly with depth from the inside diameter tube will surfaces.
The local effect of shotpeening is not expected to drive the extent of cracking beyond that experienced by unpeened roll transitions.
Experiments conducted in France by EdF and Framatome have examined the effects of shotpeening on the leakage rate and crack propagation characteristics of through-wall cracked Inconel 600 steam generator tube specimens. Sensitized Inconel 600 tube specimens were through-wall cracked in sodium tetrathionate; half of those were shotpeened on the inner diameter surfaces. Both the peened and unpeened specimens were internally pressurized to cause through-wall leakage. Test results showed that the pressure threshold, defined as the internal pressure required to cause leakage, was greater for peened specimens
} ,
relative to unpeened specimens. The value of the pressure threshold, however, j was lower for all tests than the operating differential pressure that is characteristic of Comanche Peak steam generators (1250 psi). The change in threshold pressure is believed to reflect the smearing over action of the shotpeening process on the inner diameter surface cracks. For all tests, the postpeened leakage rates were comparable with those measured for unpeened specimens. Therefore, although the threshold pressure value increased, this value is still lower than the operating pressure differential and leakage characteristics remained stable. Moreover, there was no observed extension, axially or circumferentially, of the cracks on the inner diameter surface.
These test results are judged by Westinghouse to support the contention that l
the introduction of inner diameter surface compressive residual stresses due to the shotpeening application does not have an adverse affect on leak before break considerations. Thus, as cracking of roll transitions inservice is essentially limited to the length of the transition and, for full depth mechanically rolled expansions, this length is on the order of 0.25 inch, it is expected on ther basis of leak rate monitoring during normal operation that unstable crack growth is unlikely to oc::ur in the event of a postulated limiting accident condition.
l t
94860:10/051686 6-5
As the peening process is applied in each tube a distance cf up to approximately two inches above the secondary surface of the tubesheet, the effect of the peening process on the unexpanded portion of the tube above the -
roll transition region was evaluated. For tube locations above the expansion transition zone, while providing the increased margin to PWSCC on the tube inside diameter, strain gauge test results demonstrate that the slight increase in tube wall stress on the outside surface of the tube as a result of shotpeening at optimized intensities does not result in an increased propensity for tube outside diameter stress corrosion cracking.
6.2.3 POST PROCESS EDDY CURRENT BASELINE INSPECTION
~
Following the shotpeening process application, Texas Utilities Generating Company will conduct an eddy current inspection of both legs of all Comanche Peak Unit 1 steam generators within the shotpeened regions. Consistent with Regulatory Guide 1.83 " Inservice Inspection of Pressurized Water Steam Generator Tubes" reconnendations, the shotpeening post process eddy current sampling will serve as a reference for future inspections. A program of periodic inspection of the steam generators is essential in order to monitor tube bundle integrity by characterizing the nature and cause of any tube degradation so that corrective actions may be taken. It has been verified .
through a series of tests on expanded and unexpanded tube samples that the shotpeening operation does not cause any alterat'on of the baseline signatures for uncracked mill annealed Inconel 600 tubing.
6.3 CONCLUSION
S The application of the shotpeening process in the tubesheet region up to and including the roll transition zone has been shown to provide a significant increase in margin to PWSCC. It has been determined based on testing that the shotpeening process does not adversely impact steam generator tube integrity or leak before break considerations, i.e., both the adequacy of peening intensity on tube inside diameter and outside diameter surfaces has been demonstrated. Also, as per Regulatory Guide 1.83 recommendations, the 9486Q:10/051686 6-6
l application of the shotpeening process to the tubesheet region of the steam generator tubes at Comanche Peak Unit i does not interfere with periodic !
inservice inspection and interpretation to assess tube structural and
, leaktight integrity. The shotpeening process field procedures and inherent quality assurance checks further substantiate that the application of the shotpeening process to the Comanche Peak Unit 1 steam generators does not represent a potentially unreviewed safety question pursuant to 10 CFR 50.59 (a) (2) criteria.
s 4
9486Q:lD/051686 6-7
-. -