ML20137N655
| ML20137N655 | |
| Person / Time | |
|---|---|
| Site: | Peach Bottom  |
| Issue date: | 11/30/1985 |
| From: | Chapman T, Clark J, Delwiche D GENERAL ELECTRIC CO. |
| To: | |
| Shared Package | |
| ML20137N652 | List: |
| References | |
| NEDC-31115, NUDOCS 8602040192 | |
| Download: ML20137N655 (68) | |
Text
_ _ _ _ _ _ _ _ _ _ _ - - - -
NOVEMBER 1985 i
EVALUATION OF THE LOW CARBON TYPE 316 STAINLESS STEEL RECIRCULATION OUTLET SAFE ENDS AT PEACH BOTTOM UNIT 3 GENER AL $ ELECTRIC
NEDC-31115 Class II November 1985 EVALUATION OF THE LOW CARBON TYPE 316 STAINLESS STEEL RECIRCULATION OUTLET SAFE ENDS AT PEACH BOTTOM UNIT 3 T. L. Chap =an J. P. Clark D. E. Delviche R. M. Horn Reviewed: AI/ , 'h -
G. M. Gordon, Manager Plant Materials Technology 1
Approved:
E. Kiss, Manager Plant Technology NUCLEAR ENERGY BUSINESS OPERATIONS + GENERAL ELECTRIC COMPANY SAN JOSE. CALIFOANIA 95125 GENER AL $ ELECTRIC
NEDC-31115 IMPORTANT NOTICE REGARDING THE CONTENTS OF THIS REPORT Please Read Carefully The only undertakings of General Electric Company respecting information in this document are contained in the contract between the Philadelphia Electric Company and General Electric Company, as identified in Purchase Order PB394987-N for this report and nothing contained in this document shall be construed as changing the terms and conditions of that contract. The use of this information by anyone other than Philadelphia Electric or for any purpose other than that for which it is intended, is not authorized; and with respect to any unauthorized use, General Electric Company makes no representation or warranty, and assumes no liability as to the completeness, accuracy, or use-fulness of the information contained in this document.
NOTICE The information contained in this document is not to be used for other than the purposes for which this document is furnished by the General Electric Company, nor is this document (in whole or in part) to be reproduced or furnished to third parties (other than to carry out said purposes) or made public without the prior express written permission of the General Electric Company.
Neither the General Electric Company nor any of the contributors to this document makes any warranty or representation (express or implied) with respect to the accuracy, completeness, or usefulness of the information contained in this document. General Electric Company assumes no responsibility for liability or damage of any kind which may result f rom the use of the information contained in this document.
11 ,
NEDC-31115 CONTENTS Pa ge
- 1.
SUMMARY
1-1
- 2. BACKGROUND 2-1
- 3. UT EVALUATIONS 3-1 3.1 Initial UT Evaluation 3-1 3.2 Core Sample Location 3-7 4 CORE REMOVAL PROCEDURES 4-1 4-1 Core Samole Removal 4-1 4-2 Core Sample Hole Examination and Repair 4-2
- 5. CORE SAMPLE EVALUATION 5-1 5.1 Meta 11ographic and Chemical Results of Safe End 0.D.
Boat Sample 5-1 5.2 Plan for Metallurgical Analysis 5-1 5.3 Visual Examination of Core Sample 5-2 5.4 Meta 11ographic Evaluation 5-3 5.5 Cold Work Effects 5-4 5.6 UT Reflectors in Weldment 5-4 5.7 Summary 5-7
- 6. UT REASSESSMENT 6-1 6.1 NDE of Core Sample 6-1 6.2 Other Occurrences 6-2 6.3 Suma ry 6-2 6.4 Recommendations 6-3
- 8. REFERENCES 8-1 APPENDICES A. PLAN FOR METALLURGICAL ANALYSIS OF PEAcil BOTTCM-3 CORE SAMPLE A-1 B. EFFECTS OF FIELD WELD PREPARATION ON JOINT CROSS-SECTION B-1 111/iv
NEDC-31115 ILLUSTRATIONS Figure Page 3-1 Peach Bottom-3 Recirculation Outlet Safe End Configuration 3-2 3-2 Schematic of Ultrasonic and Radiographic Records for Peach Bottom-3 Outlet Safe End to Pipe Weld 2-BS-2 (a) UT crack indications around circunference; (b) construction radiograph interpretation around circumference. 3-3 3-3 Schematic of Weld 2-BS-2 UT Plots Showing Indication Metal Path from Safe End Side at Different Circumferential Locations 3-5 3-4 Schematic of Weld 2-BS-2 UT Plots Showing Indication Metal Path from Pipe Side at Different Circumferential Locations 3-6 3-5 Schematic Showing Core Sample Location Plan and Top and Cross-Section Views of Possible Sample 3-8 3-6 Schematic of Location Technique for Core Location and Accompanying UT Signal for I.D. Crack Indication 3-9 3-7 Layout of Weld 2-BS-2 Cross-Section Displaying UT Plots at Core Sample Location 3-11 4-1 Core Sample Removal Tooling and Cutter Arrangement 4-3 5-1 Photograph of Core Sample Af ter Removal 5-9 5-2 Photograph Showing Grinding on the Pipe Inner Surface 5-10 5-3 Plug Sample Cutting Plan (all samples A, B, C and D were mounted for optical microscopy with no cracking found in any section) 5-11 5-4 Cross Section Photomicrograph of 2-BS-2 Safe End to Pipe Weld Core Sample 5-12 5-5 Composite Photograph Letailing Meta 11ographic Sample Location and Results. No cracking found at any location. 5-13 5-6 Photographs of Pipe and Safe End Microstructures Etched to Detect Sensitization 5-14 5-7 Actual Cross-Section of Weld 2-BS-2 Displaying UT Plots Used in Core Sample Selection (Note: Plots show that indications lie in weld metal / fusion line region.) 5-15 5-8 View of Cold Work on Inner Surface of Safe End. Note Depth of Grain Defor=ation is Less Than 0.004 inch. 5-16 v
NEDC-31115 ILLUSTRATIONS (Continued)
Figure Page 5-9 Photomicrograph and Plot of Hardness Traverse of Safe End Surface to Evaluate Cold Work 5-17 5-10 Comparison of Hardness Found in Peach Bottom-3 Safe End Core Sample with Hardness Data from General Electric Pipe Test Laboratory 4 inch Pipe Specimens that were Ground 5-18 5-11 Views of Slag Inclusion on Planes Separated by 0.050 inch Length of Inclusion Estimated to be 1/8 inch. 5-19 5-12 High Magnification SEM View of the 316L/308 Weld Interface (Note the Abrupt Microstructural Change and Lack of Transition Zone at the Interface) 5-20 5-13 High Magnification View of the Weld Matrix Approximately 100 mils from the 316L Fusion Line 5-21 5-14 Dispersive X-ray Energy Scan Confirming the Mo Enrichment in the Light Second Phase 5-22 6-1 Schematic of Location and Signal f rom UT of the 316L Safe End Meta 11ographic Sample 6-4 6-2 Cross Section of Weld A Associated with UT Indications 6-5 6-3 Cross Section of Weld B Associated with UT Indications 6-5 6-4 Cross Section of Weld C Associated with UT Indications 6-6 vi
- 1.
SUMMARY
During 1985 refueling outage inspections at Peach Bottom Unit 3, significant ultrasonic (UT) indications were reported on both sides of both 28-inch recirculation outlet safe end to pipe welds 2-AS-2 and 2-BS-2. These indications suggested the presence of extensive intergranular stress corrosion cracking (IGSCC). Since the safe ends are low carbon 316 stainless steel and the weld design is uncreviced, based on extensive laboratory testing and con-siderable field experience, IGSCC was not expected.
l In order to confirm the presence and depth of cracks reported by UT, a 1-inch diameter core sample was removed f rom the B-Loop safe end to pipe weld-ment 2-BS-2. The location of the sample was carefully selected to coincide with the region where UT signals indicated the presence of the deepest cracks in the safe end. The resulting hole was replugged and a full structural weld overlay repair of the entire joint was performed.
Following receipt of the core sample at General Electric' Vallecitos Nuclear Center special analysis and metallurgical secticning was performed.
The results are as follows:
- 1. Meta 11ographic examination of the sections of the core sample at high magnification showed no intergranular stress corrosion cracking (IGSCC) in the low carbon 316 material. In two of the three samples, small lack of fusion / slag inclusion type defects were found at a distance of 5/8 inch from the I.D. (mid-wall). This coincides with the crack depth and location indicated by UT in this section.
- 2. Sectioning of the core confirmed the actual weld cross-section and I.D. location of the weld root. Based on this and knowledge of the UT signal paths, it was determined that the UT indications reported in the 316L heat affected zone core were actually at the weld fusion line or in the weld metal itself. This conclusion is consistent with UT data plots at various locations around the circumference of both A and B-Loop recirculation safe end to pipe welds.
1-1
- 3. Visual observation, metallographic evaluations and hardness measure-ments determined the presence of I.D. grinding with a relatively shallow level of surface cold work (0.004 inch maximum depth.). No evidence of abusive grinding was found in the core sample, which is consistent with the absence of cracking.
- 4. It is believed that an unusual straight sided weld root fusion line (vertical cross-section in the weld root pass for a significant distance,1/8 inch, f rom the I.D.) provided a UT reflector on the inner surface that appeared to have characteristics of IGSCC. The vertical root geometry may have resulted f rom weld preparation modifications that could have been performed during field fit-up of these closure weld spools.
In summary, no cracking of the 316L material was found in the core sample. Evidence was found that the UT indications reported on the low carbon 316 side of the weld are actually in the weld fusion line or in the weld metal. The ultrasonic test indications previously reported as IGSCC are believed to be due to the unusual straight sided weld root geometry. In turn ,
the reported " crack depths" are related to small lack of fusion type defects in the 308 weld metal itself. The fact that there was no IGSCC in the low carbon 316 material is consistent with laboratory data and field performance of L grade austenitic stainless steel piping.
1-2
- 2. BACKGROUND Extensive experience has established that high carbon type 304 austenitic stainless steel which is sensitized by welding can be susceptible to inter-granular stress corrosion cracking when exposed to oxygenated high temperature water environment typical of the Boiling Water Reactor coolant.1 However, if the carbon level of the steel is reduced below 0.035%, typical of L grade austenitic stainless steel, the steel will not sensitize during the welding process and is therefore highly resistant to IGSCC. The resistance of these steels in recirculation coolant environments has been documented thoroughly.
Further reduction of carbon level to <0.02% to obtain additional margin against sensitization was the basis for the selection of the type 316 Nuclear Grade (316NG) replacement alloy reported in Reference 2. Although the L grade, nuclear grade and stabilized grade stainless steels are highly resistant to IGSCC in the as-welded condition, laboratory and field results have indicated that cracking can occur even in these materials under severely cold worked and/or creviced conditions even in the absence of sensitization. For example ,
field data did establish that cracking occurred in the crevice region of the Peach Bottom-2 recirculation inlet safe ends where the thermal sleeve was attached to the safe end. This region was significantly removed f rom the butt weld and the heat affected zone where IGSCC typically occurs in sensi-tized high carbon stainless steel. Examinations also established that shal-lower IGSCC was associated with local areas of severe surface cold work near the attachment weld attributed to grinding or slag removal hammer peening during the fitup and welding of the thermal sleeve to the safe end.
During the recent inspections of the recirculation piping system at Peach Bottom-3 (including both piping and safe end components), all of the recirc-ulation inlet and outlet safe end to pipe butt welds were inspected as they had been at Peach Bottom-2. While all of the safe ends were manufactured 2-1
NEDC-3,1115 ,
of low carbon type 316 stainless steel
- considered to be conforming material i by NUREG-0313, Rev.1, the piping that was joined to it was a high carbon type 304 stainless steel that was susceptible to IGSCC. This prompted the required inspection. This inspection established that several of the high carbon type 304 heat affected zones were cracked. The inspection also established that all of the recirculation inlet safe end heat affected zones in the 316L were free of UT indications. ' Similar findings had been made at Peach Bottom-2 dur-ing the pipe replacement activities. These results were consistent with the expected performance of low carbon, non-sensitized austenitic stainless steels. However, the inspections of the two outlet safe end heat affected zones revealed IGSCC type UT indications in both the 304 and the 316L material.
These findings were not expected for the 316L material and contradicted laboratory and field data on the high IGSCC resistance of these steels. In particular, earlier liquid penetrant inspections at the Peach Bottom-2 plant 1
of the same type of outlet safe end to pipe butt velds had established the 3
outlet safe end HAZ's to be free of cracking after prepping for pipe installa-tion. Secondly, inspections of the other outlet safe end welds where the safe end was attached to the low alloy steel nozzle at both plants did not reveal 1 any crack-like indications.
Due to the unexpected reported occurrence of the indications in the 316L material, i.e. , their presence as well as their apparent depth, a core sample was taken from one of the outlet safe end to pipe weld HAZ region to evaluate the nature of the cracking. The objective of this report is to document all the work performed in this evaluation of the apparent IGSCC in the low carbon type 316 outlet safe end material. The report will cover the following topics:
- a. Initial UT Inspection Results
- b. UT Evaluations
- Material certified as Type 316 with low carbon for design purposes but is referred to in this report as 316L.
2-2
, c. Core Sample Removal Procedures
- d. Core Sample Evaluation
.; e. Reassessment of U.T. Indications
- f. Reassessment of Low Carbon Stainless Steel Performance I
It should be noted that following the removal of the core sample, a plug was seal welded into the hole and the outlet safe end to pipe welds were overlayed with a full structural overlay as a repair for the IGSCC type UT indications found in both the pipe and safe end heat affected zones. To complete the required repairs without extension of the outage, the weld overlays had to be applied following the core sample removal and prior to completion of the entire core sample evaluation. ;
i i
J I
2-3/2-4
! NEDC-31115 1
t
- 3. UT EVALUATIONS 3.1 INITIAL UT EVALUATION A cross sectional sketch of the Peach Bottom-3 recirculation outlet l nozzle, safe end, and pipe configuration is displayed in Figure 3-1. The l results of the weld 2-BS-2 inspections are displayed in Figure 3-2. This figure shows the depth of the indications around the circumference for both i the pipe side and the safe end side of the weld.
The indications on the safe end side were originally detected and evaluated as IGSCC by General Electric, and were independently confirmed by Southwest Research Institute personnel. A total of five different examiners made the same evaluations. The indications had all the identifying characteristics for IGSCC which are the following:
Initiation on the I.D.
Depth Sharp characteristics with multiple tips Readily detected at transducer skew angles Detectable with an I.D. creeping wave
" Crack tips" with deeper cusps detected with fracted longitudinal wave The indications were known to be at or near the fusion line but field experience had shown that cracks in large diameter pipes would frequently occur there. This information - combined with the construction radiographic evidence also displayed in Figure 3-2 that showed the inside to be ground smooth eliminating geometric reflectors at those locations - lef t little doubt j that the safe ends were cracked.
3-1
WELD q
WELO
= 8.1&6n. MIN ALLOY 182 NER -
\ /" dh~3 6 Gen +
'(
I N 29 3 75 6n.-demn N _, =
E NOZZLE '
(LOW ALLOY) g 75 SAFE END ( i OM LOW CARBON l
]
316 SS f 1.22 6n.
PIPE NOM I 26.375
' 304 SS in. damn SST CL AD I jf ?
{ I
\ 25 M M.smn 25 93 km. dann I
Figure 3-1. Peach Bottom-3 Recirculation Outlet Safe End Configuration
(el OUTLET 2 85-2 PIPE SIDE 100 76
........:....- - nim g
= a wimmwiinww/wwenwwwwi t i VfffM/M/MA VfM i
a VfMf/MM/MMM/AV//fM/M//f/A VMMMA 7/fME/M/MA Vf/A i
j 0 10 20 30 40 60 60 70 80 90 I WMMlf/A W/4VM////A W//A W///A h?llllllb ff ffkffff/Yf iffff 4 Y 1 2 76 300 I l SAFE END SIDE l BOTTOM M tra l O TOP l CIHCUMF ERENTI AL LOCATlON TOP
" OF COHE SAMPLE w LONGITUDIN A L W (b) SEAM PIPE HT OHIENTATION 90 80 70 60 60 40 30 20 10 0 m SLIGHT WE LD HEPAlH C TER SOLID HOOT , @
. -i
@cD@@@OO
- s# @@
-l -
SH A LLOW HOOT SOtto HOOT 0 10 20 30 40 60 60 70 80 90 UT OHIENTATION SAFE END HOOT GROUND FLUSH Q AREAS OF DEEPER GHINDING -
PROB ABLY o.d.
INTERMITTEN T GRINDING 360 degrees Figure 3-2. Schematic of Ultrasonic and Radiographic Records for Peach Bottom-3 Outlet Safe End to Pipe Weld 2-BS-2; (a) Irr crack indications around circumference; (b) construction radiographic interpretation around circumference
l d
NEDC-31115 The recirculation outlet safe end to pipe weld 2-B.4-2 was selected for obtaining the core sample. This selection was based on both UT and construc-tion Radiographic (RT) data. The construction radiographs showed that there had been considerable grinding on the inside surface as indicated in Fig-ure 3-2. This was important since it was felt that cold work as a result of the grinding was the only potential method of crack initiation in this uncreviced joint. According to the radiographs weld 2-BS-2 had an area where all of the root and counterbore had been removed. This corresponded to an area indicated by UT to have a relatively long and deep crack indication on the safe end side of the weld. The layout of the UT indications for weld 2-BS-2 is also shown in Figure 3-2 and, for reference, the core sample centerline is noted. (The method of selecting the core sample location is described later in this section.)
Figure 3-3 shows cross-sectional plots of representative UT data from the safe end side of weld 2-BS-2. Similar UT examination results were observed for the companion A-Loop weld 2-AS-2. The plotting in Figure 3-3 shows the safe end crack indications to be at or near the fusion line or in the weld metal. Because the actual I.D. weld root centerline and weld cross-section geometry cannot be accurately predicted based on 0.D. weld crown location, these indications were evaluated as IGSCC in the 316L heat-affected zone.
Figure 3-4 shows cross-sectional plots of the pipe side indications. Note I that the indications are typical of piping IGSCC and are located in the pipe heat-affected zone area, except for the indication at the 31 inch location which could be an indication of cracking on the safe end side.
The combination of the pipe side and safe end side UT data layouts gave a clear indication of apparent IGSCC on both sides of the joint. As with any field weld, typical weld cross-sections and assumed I.D. root locations can be applied to the layouts, but the data must be interpreted with the appropriate allowance for skewed weld crowns and various as-built cross-sectional geom-etries. Based on these allowances, and based on prior large pipe experience with IGSCC near the fusion line, the data in Figures 3-3 and 3-4 (and the IGSCC signal characteristics associated with the indications) led to the evaluation of IGSCC.
3-4
NEDC-31115 E SAFE ENO CIRCUMFERENTIAL "
LOCATION 1 1
4
. I
- r 1
l l
'}
2J in.
4 I
i
) I $
i l
l 46 in, l j i l l T
l "t 2 l
1 i
52 in.
I Figure 3-3. Schematic of Weld 2-3S-2 UT Plots Showing Indication Metal Path from Safe End Side at Different Circumferential Locations 3-5
I l
PIPE SAFE END CIRCUMFERENTIAL -
LOCATION g 4
31 in. ,,
I l
i wn ,,
I e
50 in.
l
'i i
1 71 in.
Figure 3-4. Schematic of Weld 2-BS-2 UT Plots Showing Indication Metal Path from Pipe Side at Different Circumferential Locations 3-6
- - , - - - - . __m__ --- -
~ _ - - - .
l !
3 NEDC-31115 3.2 CORE SAMPLE LOCATION As described previously, the crack indication near the bottom of the 2-BS-2 joint was selected for core sample removal. A zone located from the 40 to 54 inch azimuth (clockwise looking toward vessel centerline) was selected for further evaluation. This area was selected to contain the longest, deep-est crack indication on the safe end side and confirmed evidence of I.D.
grinding, which was felt to be the only known mechanism for crack initiation in the low carbon material. The layout of this zone and the proposed sample l location with respect to the weld centerline is shown in Figure 3-5. Note that this is only a reference location, and that final positioning of the sample centerline and the exact azimuthal location was to be determined by r additional UT, which is described next.
4 i To assure that the sample was located to contain a representative portion j-of the entire crack indication, and to finalize the exact location with respect to the weld centerline, additional UT examinations were performed. These j included 45* shear wave and 45' refracted L-wave examinations. Using this 1
data, the precise azimuth location and an 0.D. weld centerline reference point
- were marked by the UT personnel performing the examination. The location I selected was the 47 inch azimuth (looking clockwise toward vessel centerline).
This was about the midpoint of the crack indication on the safe end side at a location showing confirmed IGSCC signals on the I.D. and a measured crack tip at 50% through-wall.
Extreme care was taken in locating the precise position for the core sample that was to be removed. A UT transducer was placed on the safe end and aimed directly at the indication. While this indication was being observed on the instrumentation, the centerline of the weld was marked with a punch directly in front of the transducer. This arrangement is shown in Figure 3-6.
Next, cross-section layouts and UT data plots were prepared to determine the axial position of the core sample cutter. The primary objective was to include both the initiation and the through-wall crack tip points in the 316 side of the joint. A second objective was to include the root pass and some 2
of the pipe side crack indications. It was decided to locate the center of i
3-7 4
. . - -w. - - - - - - - - s,r,,--. ,--,-w --
NEDC-31115 NOZZLE PIPE SAFE END y
N <
v
=
2 N~= 2
\ i L
40 in. (refl ' l
- 54 *. C ~_ {
/ l 1 .
,~
/ Q C
% -1 :
, u -
/*%-
1800 LOOKING TOWARD' VESSEL CL CORE SAMPLE CL WELD CL (LAYOUT USING UT)
+ e t/8 in. (DETERMINED USING UT DATA PLOTS)
NO. 7 ORILL a 3/8 in. DEEP l/
g 1/4 20 TAP
^
r F
f[f h i
PlPE > SAFE END 1.22 4.
/ U m
^
V/ W/ A d e MATERI AL REMOVED l BY CUTTER 1.o in. e l
$u* Pts[ohT I I .
a ~ 1375 'a *
,i . < \
t l
\
Figure 3-5. Schematic Showing Core Sample Location Plan and Top and Cross-Section Views of Possible Sample 3-8
k NEDC-31115 I
908
/
/
/
/ \
/
/ , .= 08 tTOPt s
180TTOMI i 8
,. /. ~s -
\ S'. /
s,
- WELD 80%
AMPUTUD E o I.8 2 METAL PATM ten.)
Figure 3-6. Schematic Showing Technique for Core Sample Location and Accompanying UT Signal for I.D. Crack 3-9
NEDC-31115 the core sample cutter 1/8 inch toward the safe end side of the 0.D. weld centerline reference point as shown in Figure 3-5. The data plots and sample cross-section layout are shown in Figure 3-7. Note that up to 3/8 inch of material would be removed on the pipe side of the weld root centerline, and at '
least 5/8 inch of material would be removed on the safe end side depending on the actual location of the weld root. This location satisfied the objectives and would include more of the safe end material than the pipe. The actual core sample later verified that the location was selected properly.
It should be noted that the 0.D. weld centerline was used as a reference point for UT data plotting and sample cutter location; however, it could not be assumed that this point represented the I.D. root weld centerline. The weld crown had been ground nearly flush and there was considerable 0.D.
mismatch between the pipe and safe end. With these conditions, as with any typical field weld, the I.D. root could actually be located on either side of this apparent 0.D. weld centerline. Similarly, the UT cross-section data plots (Figure 3-7) show the crack tip in the weld metal, but a slight shif t of weld centerline would place this tip right along the fusion line, typical.of large diameter piping IGSCC.
4 1
3-10
CORE SAMPLE CENTERLINE f WELD CENTERLINE
( 460RL 460M L 450 SW 460SW 0.125 in. % e I U U if If i
I PtPE SAFE END l
l l w I 1 an.-dem COME SAMPLE e %
00 47 in.
j VIEW LOOKING TOWAR D VESSEL 2700 900 1800 Figure 3-7. Layout of Weld 2-BS-2 Cross Section Displaying UT Plots at Core Sample Location i
3-11/3-12
- 4. CORE REMOVAL PROCEDURES 4.1 CORE SAMPLE REMOVAL In this section, the location and removal of a 1 inch diameter core sample from 28-inch pipe to safe end weld 2-BS-2 will be discussed.
k The purpose of the core sample was to metallographically confirm the crack indications in the low carbon 316 safe end material. By obtaining a through-wall sample, the exact location, extent and depth of cracking could be confirmed. In addition, the I.D. surface could be examined to determine the method of crack initiation.
Tooling Preparation Once the need for a sample was identified, commercially available cutting tool vendors were reviewed and equipment was selected for mockup testing. The tooling selected was a special milling cutter and drive motor assembly, as shown in Figure 4-1. The tool was modified to shorten its working height to clear the 18" minimum distance between the pipe and the biological shield.
The mill cutter and attachment arbor were also modified to install a capture bolt to the core sample. This assured that the sample would not inadvertently separate from the tool.
(
The tool was set-up and functionally tested in the GE San Jose facilities.
All tests were performed using a spray coolant of de-ionized water. The tool performed very successfully, and high quality cuts could readily be made through a pipe butt weld. On-site, additional mockups were performed under l
full access restrictions to train craf t labor crews and formally qualify the sample removal procedure.
Selection of the sample removal location was conducted by reviewing the l
original construction radiographs and the ultrasonic in-service-inspection results as discussed in Section 3. Following verification of the sample cut-ting location a small pilot hole was drilled to a depth of 3/8 inch and tapped 4-1
e o 28-in. PIPE . ;- s ,
l l
- PILOT DRILL TOOL MOCK UP
,e .
MOUNTING PLATE x CAPTURE BOLT TOOL BASE AND SLtDE
^ h DRIVE MOTOR CUTTER y ,;. . g,s f.iyN7
$a
/
't 4
. E.
a) TOOLING INSTALLED ON MOCK UP b) CORE SAMPLE CUTTER AND AND DRILLING PILOT HOLE CAPTURE BOLT ARR ANGEMENT l
Figure 4-1. Core Sample Removal Tooling and Cutter Arrangement r
i
NEDC-31115 to attach the sample capture device. The cutter was installed in the tool and the sample was removed using spray de-ionized water cooling. Sample removal went very well and required only about 15 minutes cutting time.
The sample was removed from the drywell and shipped to General Electric Vallecitos Nuclear Center for destructive examination. No decentamination was performed to preserve the specimen I.D. and crack surfaces for future residual element contamination studies.
4.2 CORE SAMPLE HOLE EXAMINATION AND REPAIR While the hole was in the pipe, a visual examination of the inside sur-face of the joint was attempted by Philadelphia Electric. This examination was complicated by seepage of water through the hole due to the only practical method of draindown that could be applied for this work. Draindown was accom-plished by lowering the water level in the shroud annulus using the 28-inch ,
outlet piping B-Loop. A water level control and monitoring scheme was applied to the pipe, but remaining water in the vessel could not be drained lower than the 28-inch outlet nozzles, and therefore seepage from higher level nozzles would of ten cause water to run down the pipe.
Viewing was also limited by the complexity of locating and manipulating fiber optic devices in a 28-inch pipe through the 13/8 inch diameter core sample hole. The depth of field and magnification tradeoffs with this device, combined with the access and manipulation problems, made it impossible to per-form a meaningful viserl examination for cracking at other azimuth locations.
The only pertinent examination results were obtained by viewing directly through the hole to the opposite side of the pipe. With this method, a dark line was observed on the pipe side of the root pass that is believed to be the 43 inch long pipe side crack indication reported from the 85 to 38 inch UT azimuth (see Figure 3-2). No such indications were observed on the safe end side of the joint. Although this observation is by no means conclusive, it does support the overlay repair of the joint due'to pipe side cracking, and 4-3
NEDC-31115 showed no anomalies on the safe end side that would contradict the subsequent core sample examination results.
The repair of the core sample hole was performed under the approved ASME Code Section XI repair program in place for the weld overlays. A circular I plug was installed and seal welded in position. The vessel / pipe was then f filled and a full structural overlay was applied over the joint and seal plug area. This repair was evaluated against ASME Code Section III rules for open-ings and. reinforcement (NB-3330), and against Section III allowable stresses by a finite element analysis. Results showed that the design with the weld overlay satisfies the intent of the Section III requirements (without taking credit for the plug seal weld itself), and complies fully with ASME Code Sec-tion XI IWB-3640 requirements for evaluation of piping.
9 4-4
l l
l
- 5. CORE SAMPLE EVALUATION The core sample evaluation followed an evaluation of a boat sample cut from the 0.D. of the safe end to determine the chemical and metallurigical condition of the safe end material. Following a brief discussion of these findings, the core sample evaluation will be presented.
i 5.1 METALL0 GRAPHIC AND CHEMICAL RESULTS OF SAFE END 0.D. BOAT SAMPLE Af ter the initial UT indications were found, a boat sample was removed f rom the 0.D. surface of the 2-BS-2 outlet safe end. Prior to removal, Electro Chemical Potentiokinetic Reactivation (EPR) measurements using a dual i scan unit established that the material was not sensitized. This sample was removed from above the uncracked region of the safe end near the top of the weld. The sample was 1/8" deep, 1-1/4" long, and 1/4" wide. The sample included both the 316L safe end material as well as the weld material. The sample was then transmitted to General Electric's Vallecitos Nuclear Center (VNC). A metallographic evaluation was made of the material for micro-structure, sensitization and hardness. A chemical analysis was also performed on the material. The evaluation established that at the 0.D. surface, the material was annealed, not sensitized, and had a typical grain size of ASTM 3.5. The hardness level was also typical at R 71. The chemistry check B
established the heat to be 316L. Table 5-1 gives a comparison of the heat certification chemistry and the check chemistry from the boat sample.
5.2 PLAN FOR METALLURGICAL ANALYSIS A cylindrical core sample, approximately 1-inch in diameter was removed from the safe end to pipe weld 2-BS-2 of the 28-inch recirculation outle t nozzle N1B to confirm the nature of UT indications. The sample was removed from a location near the center of the largest and deepest UT crack indication.
4 (The depth was estimated to be 50% wall at this location.) The circumferential location of the core sample (Figure 3-2) also coincided with a location of heavy weld root grinding, as determined by the construction radiographs for the veld.
l 5-1
_ _ . - _ __ _ __ _ _ _ _ _ _ _ _ _ . _ . . . _ m _ . .
The core sample located as shown in Figure 3-5 encompassed the 316L safe end material, the weld, and the 304 SS pipe material, with 5/8 inch from the i weld centerline on the 316L side and 3/8 inch fro = the weld centerline on 304 l pipe side. The biasing of the sample location to the safe end side of the weld was for the purpose of focusing on IGSCC indications detected on the safe )
end side.
1 Attachment A is the plan for metallurgical analysis of the core sample.
The plan, which included macroscopic examination, optical and scanning electron microscopy, hardness profiles and sensitization measurements was focused on (a) the determination of the nature of cracking and the identification of propagation mechanism in the 316L safe end material; (b) the verification of the degree of I.D. surface cold work and (c) the metallographic confirmation of the ultrasonic crack signals, including confirmation of crack depth rela-tive to the weld fusion line.
5.3 VISUAL EXAMINATION OF CORE SAMPLE d
Immediately following removal of the core sample from the safe end it was examined visually (Figure 5-1). Cracking or weld root fusion lines were not evident. Evidence of grinding, however, was present. A ferrite meter (for weld metal ferrite determination) was used to confirm that the core sample was taken at the desired location and the sample contained both safe end and pipe material. The sample was packaged and shipped to Vallecitos Nuclear Center for metallurgical examination.
i At Vallecitos, in accordance with the established plan for analysis, the inner surface was visually examined and the macroscopic evidence of weld root grinding was photographed (Figure 5-2). The weld root region had been ground in a direction transverse to the fusion lines, obliterating the weld root.
The ground surface appeared typical, with no evidence of aggressive, abusive grinding.
l.
+
l 5-2 i
e
.= - - . . - _ _ ._
NEDC-31115 5.4 METALL0 GRAPHIC EVALUATION The sample was then decontaminated by ultrasonic cleaning and examined under a stereo microscope. At magnifications up to 33X evidence of cracking was not observed at any locations along the I.D. surface. The core sample was then sectioned as shown in Figure 5-3. Each cutting plane was oriented normal to the weld fusion line such that a polished and etched section would be a transverse cross section of the weld. Figure 5-4 is a polished and etched cross section of the sample. Close examination of this section, and the three other sections prepared from the core sample (Figure 5-5) clearly showed the following:
- a. No IGSCC was found in the entire sample. There is no cracking in the 316L safe end material, or in the pipe material contained within the core sample.*
- b. No weld heat affected zone sensitization was found on the safe end j side of the weld, as expected with a 0.019 carbon content. The pipe side of the weld (0.054% carbon) did exhibit sensitization as expected. Figure 5-6 are views of the pipe side of weld (RA2) and the safe end of the weld RA2, respectively.
I
- c. A replotting of the UT metal path assumptions on the actual cross sectional weld geometry showed the safe end UT I.D. indications were actually located in the weld metal / fusion line region and the crack tip indications were located well into the weld metal. This is based on the confirmed location of the root centerline on the I.D.
relative to the centerline of the weld on the O.D. surface. Fig-ure 5-7 displays a sketch of the UT metal paths sketched relative to the actual core sample cross-section.
- At this azimuth location, the pipe side indications were up to 1/2 inch from the fusion line (outside the edge of the core sample), and were shallower and lower amplitude than at other azimuth positions. Therefore, pipe side findings at this azimuth are not necessarily representative of the remainder of the pipe side indications.
5-3
! NEDC-31115 I
i 5.5 COLD WORK EFFECTS Severe surface cold working, and increased surface hardness levels can be an important causative factor in the initiation and growth of stress corrosion l j cracking in 316L stainless steels. The macroscopic visual examination of the -
i inner surface of the core sample showed that there had been some weld root grinding at the time of fabrication. The degree of surface working was judged !
to be normal. There was no sign of aggressive or abusive surface working, as would be evidenced by surface metal smearing and deep, bhort grind marks or f
l depressions. The surface was found to be smoothly ground with shallow grind marks normal to the fusion line. The weld root was ground flush. t on the through-wall section (marked A in Figure 5-5) polished and etched f to reveal the microstructure, the section view of the safe end inner surface showed a cold worked layer, with grain deformation to a depth of less than j 0.004 inch. This shallow cold work layer is characteristic of light-to-
! moderate surface grinding (Figure 5-8). A microhardness traverse was prepared
! for the safe end at a location approximately 5/16 inch from the fusion line of-j the weld, to be outboard of the annealing effect of the weld. Figure 5-9 dis-plays the location of indentations and gives a sketch of the hardness profile.
! Note that within 0.010 inch into the surface the hardness level drops from R 26 near the surface to less than R, 92 for the bulk interior material.
- l. The hardness profile found in typical grinding in Pipe Test Laboratory studies are shown in Figure 5-10 compared to the cold work in the Peach Bottom-3 outlet safe end. The figure shows the Peach Bottom-3 safe end is l bounded by the Pipe Test Lab data. ,
5.6 UT REFLECTORS IN WELDMENT i
j Metallurgical analysis has shown the safe end weldment to be free of a
4 IGSCC. This finding is consistent with the metallurgically and composi-I tionally correct 316L safe end material and 308 weld deposit. Since the weld- ,
, ment was found to be uncracked, later core sample evaluations focused on I understanding the source of UT signals detected at the 316L/308 weld interface.
i f
5-4 i
NEDC-31115 The ultrasonic (UT) signals detected on the 316/308 weld region apparently resulted from some effect other than IGSCC. The core sample sections prepared for microscopy were studied by optical microscopy, scanning electron microscopy (backscattered electron imagery) and electron microprobe (wave dispersive x-ray system) to identify possible UT reflectors.
While an obvious source of UT reflection was not identified, the follow-ing facts are apparent:
i
- a. On the four transverse sections examined by optical microscopy, no j evidence of IGSCC cracking was found, and the wild / base metal inter-face was metallographically normal, such that a m. crostructural dif-ference would not be expected to cause a UT reflec; ion.
- b. Under optical examination at higher magnification (100 to 500X) some porosity, slag, and minor lack of fusion was found in the weld metal. The largest weld defect was a slag inclusion in the weld metal found approximately 1/8 inch from the safe end weld fusion line. The location coincided with the location where UT crack tip indication. As shown in Figure 5-11 the plane, on which this "large" defect was found, was ground to expose a new plane 0.050 inches below the original surface. The inclusion was still visible j on this new surface, though much smaller in size. A rough estimate l for length is 1/8 inch. In other regions in the weld small " normal" amounts of porosity were also found.
- c. Following optical microscopy, the core sample pieces were radio-graphed (RT) to detect the possible presence of additional weld defects. Further evidence of small porosity and inclusions was found.
- d. By optical microscopy, multiple non-metallic (presumably MnS) inclu-sions were found along the 308/316L fusion line approximately 1/8 inch from the I.D. surface. While not observed in this section, 5-5 l
such inclusions at sufficiently high densities could cause UT reflections similar to IGSCC. Radiography confirmed that the density of these small inclusions was low and therefore an unlikely UT reflector.
- e. The weld geometry was somewhat unique in that the weld root region had vertical fusion lines for a distance of about 1/8 inch from the I.D. Although the UT signal reflection characteristics of this geometry have not been fully evaluated, false UT crack indications were found on the removed core' sample as well as in a laboratory test weldment with vertical fusion lines,
- f. The sectioned core sample was examined on the SEM and microprobe equipment in an effort to identify possible compositional or con-stituent gradients at the weld fusion line that could account for UT reflection. Results were negative.
The sample was examined on the scanning electron microscope using the back scattered electron imagery (an energy dispersive detection device). No abrupt corpositional gradient other than Mo was obaerved at the fusion line, and no unexplainable elements, inclusions or accumulations associated with the fusion line were found.
A linearly decreasing gradient of diffused Mo was observed in the weld metal when moving from the 316L fusion line towards the pipe side fusion line.
In addition, there was an enrichment of Mo in the ferritic phase of the weld.
The enrichment gradient from the 316L fusion line to the 304 fusion line was linear, suggestive of mechanical mixing. Figure 5-12 is a photo of the 316L weld interface boundary and Figure 5-13 is a SEM photo of the Mo enriched fer-ritic phase within the austenitic matrix. Figure 5-14 is an energy scan made on the SEM with the energy dispersive X-ray detection equipment providing confirmation of Mo enrichment in the light second phase regions of Figure 5-13.
! 5-6 i
, Electron microprobe (wave dispersive 'X-ray imagery) techniques have con-firmed no compositional uniqueness associated with the fusion line, or weld deposit. The microprobe also confirmed the Mo is likely present in the weld structure in an elemental form rather than in the form of a complex carbide.
This suggestion is supported by the lack of carbon enrichment in the ferritic l phase, or inclusions.
i i
l The SEM studies have shown the absence of compositional or microstructural
] gradients or enrichment associated with the 316L fusion line that could have contributed significantly to lJT reflection. The observed Mo gradient is gradual and the minor effect is probably unobservable by UT methods.
1 1
5.7
SUMMARY
.i No IGSCC was found in the core sample. The lack of 316L cracking is
- consistent with the absence of excessive cold work, and/or sensitization.
i j The available evidence indicates the UT indications resulted from the j unique weld root geometry and the presence of small weld defects.
I l
t i
i I 1 I I l
)
5-7 i
- -. - _--.-..---_ _.._._ . _ -._ __~_.. --. - ,_--___ . - , , _ . - _ - - , - - - - . , _ - _ - _ - , - - - - . - _ ~ , - - . . -
4 NEDC-31115 Table 5-1 MATERIAL CHEMISTRY VERIFICATION Mill Certification in Laboratory i Material Test Report (w/o) Verification * (w/o)
Carbon 0.019 0.016**
Chromium 17.05 17.20***
Nickel 13.49 13.72***
Molybdenum 2.19 2.19***
Silicon 0.80 0.28 Sulfur 0.020 0.019 Phosphorus 0.031 0.028 Manganese 1.67 1.79 Cobalt 0.11 Columbium 0.02
- Copper -- 0.33 Titanium 0.005 Vanadium -
0.04 i Boron --
0.005 Nitrogen 0.60-0.100 i
i h
- By emission spectroscopic method ,
- 0.024% by combustion method t
- By X-ray fluoresence method i
- 5-8
! e i
..e Jj- %- ,w?
' k';.
p.,
N' Ef{k .f ~ b , [
> , +
, a ;;, \@r '([_*, +" .
[
.;',',i,.
_ y3,.,-.y3 a __
. l; c'y 7, .
-.,Q w*
- t. s3 G, ..
Q?-
. q. 4\ '
k .s , .
w
~ A ,*
j' \
\
1 5 \
i
\ ,
, Ay T w so i
y e
i *
\
\
\
j
...,L;:.
1 .[~
., ~
3 ,.
- 1, ,> - --
. \
.o .+
- -e
\
) , , ;
i
@y i
\
\
(f:A1%.
w _ t i
1 Figure 5 Photograph of Core Sample After Removal i l
, i
1
. ~ ,. .
i
- R i
? ~
i
' + ; 2.. .
,)
.y . -4 ' .!
l , ,
9 h4
,'r# . J< ' [ i', t I .,a
, T)'<#; '
i
' ff
- : 1 , . 2. . + a: ..
t i , ' 'a j, s'gi: w; if 2. .'/f pp' {' I
- w. . . . . - . .. ,
'f, r "g [
.-N" . .l.N_'* . h.y' h
- M Y' -' ' , 8, o -,W+
- , w r .,
f' ,,_ H
.$, *M:2.Jjf s .
~
. ;... vi mfi ;. ~ ,, .: 7 Jc -
e j , : .s.mj f[.g .
0 ;[ X M. p ' a w 4 ;
t Q ;~
(
t 1
l l
l i
i l I~ .Mr4 ..i.
l Figure 5-2. Photograph Showing Grinding on the Pipe Inner Surface i
l - - - - .- . . , _ _ . _ _ _ _ _ _ _ _ _
i l
l NEDC-31115 .
I I
l
.i?
b %< . > ' b *,('
~. .
- . =, ,
SAFE END
(-Y*.'
S .
ge h ..
- g. 5:.
. , ;. s . . . $ . '. .,, + ' .
+. ,
I. .-
- j. r . 1 4
- t t -aq - c.ge s . .
s 1g* ,
g
. . . ,g
- , a,
'rIllg..['f$@ - ,.,,..,7,,-
, 3, .y ,- ,
i i?? k .s'
- .[ [ L;, ,. . . . Q
,g? ;.% J. -M ',tf [,.'.
'- ~
.,, jyn' ' . .
en.
. .' .. ., , !. .~
~
e-
, f .
' +' .
(h ... :
PIPE K4 y . '
N ._
.m> '
s ..
- g. _
+
i 7 35. .- z
,19. ; ;
'ai A. ,
-r f.. ,' ^ ~
, , , , , - w r, e
- ', ; e % ,
,; , ;j .; Q'2K..,3~- Mc. m; .ij; QT-
.' O . ,Y , $. y~f ~ .,,s
.-- ?Y .. ,
i5+&M'*,bb,*. ;- , .?
- e. - n ' ll. Q $$$$5 % b
- i
'~ '* " ' " " ~ '
Figure 5-3. Core Sample Cutting Plan (all samples A, B C and D were mounted for optical microscopy with no cracking found in any section) 5-11 1
l l
l
- - - . _ -o -
- x. ,-
.- 4
. s a u y-
- E * ,
, j', ' # ,,
L,-
c m.
,, T'I.'[ -
a : )
rT[W
~, W. . - ;
W 4;,,m b
s ,- . ; ';c. o e
.,/
n, v.
b .
- ; _. L',a ., f .- ,e
,~'
XL r;
., $rto f,;\.,
4 es, t1 . iy.
^
y di.x. s.
.V
,i ?y%, ,
.,a y ,.
_^ '
- 1 N?*
.~
5, G%.
s . _. . <:-;* ~ /,
. av e :
' -' r A_. _ 7. ;.5%. c.,i ,
s t. .,.
aw.
..,.~g,..,-y .y i' [
.3
~. .,
.,e_J e m, g; x-
.? gy
'~ ~ y ;.p' .--'* w *, % i ,; ~
.;;e , ; ,; .<7 e R.y, . = - V .. - - e '
L 'i~ n '~;A
'tk e+*'".i _ .
3 g , , QM<.j; % .4 f { % s 9[J[p l
- p - i
- r, n ,i
+ sy x J /
y\.- -: ., .. .,y M n f . /
s, n - t wJ
.y ;,
~
.je,jQ y., .j J ,,
Q< sa
- ('
.( + '<.
Nr . s.s g;,7 +; -.x q="7**'
.,.:,a a c n.~, . , . ,,~'., ,
> ., ?
g~ si i,g;,3...
- y. ' *'7 g .~...
'.,.r [;.
, .; w v- .. #
.y.
.; t
.g , . ! '.
" As
.%. %g
- y j.or T[y.. .C'
> ~,,;;y d 4 gtrA v 1
(.[
, - 's W . f. .
t L1E4. .C.sihidh, Figure 5-4. Cross Section Photomicrograph of 2-BS-2 Safe End to Pipe Weld Core Sample 5-12
CORE SAMPLE CUTTING PLAN
} \..
. .ib7k. I:, y'!. - ' ' it '.4, f5h* Q h ,_ %
, p f << * ~ . .
s .1 Of Iq4 ';;g >
.. '. 's -
, , 'N3Ad.,, . O ; y (,' " '
, )
^*
..+.s. '
304 PIPE GRINO MARKS
' Xl, ,
~
N h f
._t 6 .
'. i 4
.- 316L SAFE END WELDROOT ,
,. 3
- -y
/ "
i s .,-
le g g3
, , y- ' -
E
- ' SC 51 bl t.n is_ - ---- W f;;; - ga.
',- i
-)
lr- 6
/
. E'ry P.E 3
f7 " A, .
k ,,d $
~
~
Sc.52 Figure 5-5.
w \_
Composite Photograph Detailing Metallographic Sample Location and Results.
l No cracking found at any location.
l I
NEDC-31115 y-k.
u-N . . .p: . 3
, c^]
.-l q . .:y.
n,
, ' : ; b, .--
. ..3.}4
.~ ; sd g ,17 5.4 . u +.
W
. s, . -
gp
) , ,, . . .
/..,
r t ,.
r e
. j .
- a. . _; ,,s s : .;q ;;[.. , . . .
M, = . - _-u- . :
,7 p} .-
7
-3 . ,
..s
- s. .. ; .- .
p , - * -
fj * [,.; ^:_ >; - ;
, RK , ' j
?^ ;
.2 > - ;;.n '
g . . rV , . . , ':. ..y. e t ; .. : . ,,. u. y .
7_ .. % , y s . . - .
,-. :-, . .c .iy
. 3 . .
3- x
- ,4
- 4. & A, , .* .w.
v.sn.
n .,,q.
.e h'. : : .x -y
-w.
3,. e.t"**..,.
...,- . ,a. s' g.., s
.,.4 q w %,. &. .c; , :.
w 3 3 - ,.. m s K, . v . . . . . ..>: -
b ~ I., .lfN
- f % 4n.. q;b..cY '. r .. .
i f d As .
c7,.
,, ,;7 wyw L
. +
. y. 4fr r:
v .,
M . - 7. .gg.,.g 3
V t, ? ,Tv ... a4 . v
-<,x . .
i-
.4, .,f .ij
.y m; g 4 E t i 4 .' 1 v' A ,n i3 m - .
6 pr.
pa-
- y
- #.ga. ,. . As r i 2:4s. ,
moo.
.c
. tw. s - . mQ y .g + w u9
..y A5 ; 1 - c,-ew .r. .
w.
1%
g MN$m.a ?1!d E M ih b.1 m yb8hedd$bsihh, 250X (a) PIPE SIDE OF WELD (HAZ) 0.054 CARBON 304 S/S m 44 % @
f C o Q::YCI' I
Wv, q7d.yn@~
~
W v4 E U~t y @c.
.- t
%O. e;j
/.. %
.: :pM p [N Q W.;. m
%* vn.
%n Tm 7 s .;;%;;aww%s;@#
- t W w m Qw u p ., MJ4.3 7 m p g; m , ,t,yf _y
\"
..-w .
- .-T s ; y;
..a .. ,'j: wy:c W,i
. "sp
- -x. y n, ;y.%,._.n.. :.. p y?, ~..r ,. ,
r m m, . -
e 3 ;
w u ne.
~
- . n,= 3 .py 7 p.
~ r.;+ Q: p% y h.. , m m.,
, +, .1.,q, ;. -;;,w.p g .M, < > ;y w,.- en.y,.. ,.wycw :
.. ..t,. . . ,\
- m. ,. y. u,,
^
. -p ,, - .
e .~
y . qy /, . .. e .; q/>r 7,. -
y;,a - y : ,v : m.+,a~s 3 4.g p3 t u: n., g. m g u _-9 3 W.
m.w m W..
^
,n.z ,
}r%,-,#. -d ., . . f, l ^ , },:
- _., q;; m ._
' w'u. _ O
- Q :. w w .~ w,.m .. sJ.,
C_ , .m
- .n. :, e , s _w -
j [hh.,. yp ;.; _
g,g 7^Qa :., ;. {.w,y,p. ., j [ ~, ;.. ' , j,,,
.,.m',. e: . w.' n ..n g s g.:
QS " J .7.[; !
h e ,
, se s . .m h ,
(
. f[ 2f [m ^ :1 J m . .W..'
-m 4
y :p' f . 3 [ .
t,4 .r.~..
-f.' l
- I .e' ;.
chN k a'e%, f ,,e%q , 79 i. a,%
~
,j% y. ,
- w x,.-
q.'% '
gz. - . s
, > m, .e- ,, ,'
._ A. b u ,-jp w, m i):..,3 y ; y <: @l . ( g f, . l p.* ' . , -,:8,jE~, qg
[. m,,2 - _
r , :.
s
,3 .- zev s
e-. , m: n . .. . . ,,.
. e <
- ra .
a7..y:g. a,-,. ;,wnmmw- ,. v ,
_ A.2 g,47 .% 7y " s am y p 7 ~ p , ,7 ,
'--5= -w-
"pj.h.q:Q ;lBl*aia Q,,. .4T4% ,yn.
jf Q ;p %dQhfQl9je y p;j^$' [~f[. ?
Q' f 'j&q
, , . T :B:
~s
- ',.. Mgp
.,x;.J ? ; y.I y@; w -
. t ' . a o, N y 3.
.f,i.m. :; ' +. .
1 ..1~ , . % ./, --
'N -, =,. .rfv.-;.;> :
.s x >
+-- g ; ' p 3 ;A.A-- /. - . . .
.--- *.c. f. .f , .
g.- 5s * -- -
- 4 ,,
,rs w . .g.s .v,
~ < ., g _q 5,,-__y,; ,_ _,
i c% '
,yu , z, 4e g _
v s&.wh, n.2m: n] .' j , s .&'wwww g ..c s
. ~ . .
2 sox (b) SAFE END SIDE OF WELD (NOTE NO SENSITIZATION) 0.019 CARBON 316L S/S Figure 5-6. Photographs of Pipe and Safe End Microstructures Etched to Detect Sensitization 5-14
CORE SAMPLE CENTERLINE WE LD CENTER LINE 450 RL 450RL 0.125 in.+ e 450SW 450SW lI I r i f U N dq r ,
SAFE END
~
N' i 1-in.-deem CORE SAMPLE e Figure 5-7. Actual Cross-Section of Weld 2-BS-2 Displaying UT Plots Used in Core Sample Selection. (Note: Plots show that indications lie in weld metal / fusion line region.)
5-15
~
\ . ,fC N s
+x !
e
/ [ /
/
' ,/
, 6 /
, e, ,
= 4:
., i
, x n '
\ * -
. /, L
.' ,. / . .
. 3 . . ,
,,/ _ .I
- ' .. 4. s~ o.004in.
- M jf
[7 3 4 r ,
- 2 sox Figure 5-8. View of Cold Work on Inner Surface of Safe End. Note Depth of Grain Deformation is Less Than 0.004 inch.
5-16
l l
l
' NI' h Or '
.gd .j\ + Z*?;
~; . . . +
~
) -. .
,, , 'v
,. s -m
. 6-t, .
- h N~. g ' 5 ?]
s
\
c T. - c {, .~
.s y] _s :
-.g . ,
-$. _x .,
I
- !l ;&.N. ^,-
k '
s
, ~ _
\ 'n y.,
3 pu:
7'
\ , Y:y' [.'
1, 6
?q
- j. ,-
. 5:'s n - +
f \,
/' / x.
100X I
I 28 -
l 2 l l
22 m
O (100) -
8
C cI I @
94 - 5 @ ,
n @
c m - @
l l i I I O O 01 0.02 0.03 0.o4 0 05 0 06 1.0. SUR F ACE DEPTH INTO SURFACE (m.)
Figure 5-9. Photomicrograph and Plot of Hardness Traverse of Safe End Surface to Evaluate Cold Work 5-17
700 -
O PEACH BOTTOM-3 01, g PIPE SPECIMENS 600 -
AFTER GRINDING V AS-RECEIVED PIPE 500 -
_L 8
M 42 RC400
'e E 2 O
5 9
y 27 RC300 -
7 os _
- O D
3 Q>*x
__ d ;~
90 RB200
._~~~~~-
'O
- M,-g~ C
, ~ -
O 43 Rg 100 -
PE ACH BOTTOM-3 OUTLET SAFE END 0 I I ' ' I I I I I I I I I I I I I I I I I O 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 35 40 45 50 55 60 65 DISTANCE FROM 6 d. SURF ACE mm (mile)
Figure 5-10. Cocparison of Hardness Found in Peach Bottom-3 Safe End Core Sample with Hardness Data from General Electric Pipe Test Laboratory 4 inch Pipe Specimens that were Ground
, ,: .p wmm ..v -w
.n,:,,.. w. .. q,( ~ . .p:, n f
N,..-
s,r < r o.. ,,p.?it.*
- n. 3. t . y. . s.4f .
b ~
',>7 h
' %c , [md,W-,'? ?
w.,f. k#d:
e 4
':1 Qi?fhiMQ@aa:%&m'G'AiQM&Q g4.W8h l
n e %i' Wit &W@%Q&Y1QsMM MMM%2 c h&&M%@ghib$kMh&W bN $
iWM>.,e#!
4
'h. @NNhhkh
% w@?m:sNME k$h@b -
$ i 4fp.tpr $ ~%M ,9 d
n y 7ln . ,.v M, 0r-%,..'p;f. .
n %n. n.$c ;- M J $. u.m.: m y %.. u,.,
me,, s.,* .w. w. ,
s m' .m .
^
.qx. . . .. ,. ;
c i '[. ?, L Q , r*.f C
. e, y ~- z.z.a .C u
$ ~ a
- e
. y.i ,.~)em gg(,p,sy aw,. '. , }%c wswc. 2. c pv,.~ % 94 f . ,. '.4.i. . b.'t m.?*Q' p.
y
,.3 .
w
- .- . w. Wry:og,m; l'zQ:h*fyQ;y : p C. . :.m b ,~
q+%, t y. w g n,,s:, :t ' .x~, ;",~,4','. ,'
.=
. .n. p .. . . 8
- w. . o
.p;,. 4. . . .r , .. . .r. ~ . . . . .
, e. -
w c e.m. v.ee9 a -
.,n,, t, , h +ge.+ . s. .,.e :w > x:.y -c:,~ . .
z
- ~.-.
- w wy:p. c.-.nse,,,., . e
- .- 5 -
- a sn v' p.at,;,. ;' .N. piv: *.,4 h +. ,3. , . . : x, . w
.n ,~ . m-n. .,.a.
.,..y
-. m. . m
< o.
m ., .c,. . . .. . .
r m . .ers . ~.t
. . ,:e m ~ , 7 ., . .\ o g y
- . n .y
- L .
4,,,7.
p .,T,3 %.; .\v. >- ,, :;.
6,, y< p , .. . .,.z.~.-,r-.;.,,, ,o
,., . v. ,. ... ,, , , - 3. , , ' .gc4. g w g . ,
. : a. . .%s , . : .~.Q
.v , .
- . y, ,
m
's.a*
e ,
. . .. ?,9% ., . . n.
, . f.
g ,.", ; f:
w ,
4 (,
.a
-.n ,.p,7 .
,'*' + w. '
.pr*, ._ _ .
,A :.. . :l l *1
. ea, _ . . .
I ;f p g
. 9 d C.
q a w , s v, . , ,
.: . w am . .
~w- g,.. %.c ..%jy.
~,u.. s:w ,
t.n=.u. . . , n., y . ... M; 9
.. m z.3 w Ny') M $ m ,.7"Q"
-4 ;)
(
m t.@M::n..,:.
r f5 5 0 M . Y -
a.
- r. ~'t *.-. .,: ~. :,-
W,.;r. w E. m: tt.m1-)
..' . fy**a~Qi$e:s,ae u..)pK.w . .,m?e g;y;2.* .
a f .s 4r v'.,:
e z w- ,n :* ,< .
!?w f O .
N,..'*b yh.,. 7. U .E -8," b DE .c
- . *'t' . e m/."., a ' ,, e' % ",. 4 ..
t ,e .g n.. w
, O cU
'h_ f N . . y W f fk o oe C.( "fshD b.,2,s.: ,*yf:M. . n'.&, pgu j,, p.J,h p-. e(.
4
'"m*"*
{c. - r 3.; pm.s -
mf.I,,n.~Ye,: a .m;ypvf. ~ypf; me 3x f
e e :,G ww .2u.
e a__ __
u-2oa 4
f @._ o y a
- T >;wus, ., . ,: : - q. . . . . , . .. ..p , ...o.
- mg;pt~ ,{ ., g,
- . m *y ,y ,. -u eo I Jge .,Qn . y- ,K y*V* >; a.<.
w gmg
- /
O .' . t,c:.
.om. .,- r s ' .
O u y'r C
~
,e.. h {,. 44..c-5... .
,. .,.~. .yw f;- .. m
)
'g.e ev.s, .t .
t 0
- ! ;g ;
. .g:.',Y lm~ ;- p
. .. au N')... m ,
J ;. .-
'a (' lo..$ : k h .
{$
.c,7"' N. ,
de hQ[ ' fti g ., '
-@', ;, ?., . - .1. . ,
~
e
,. ~*
J.Q-'fW. ., .
. 'J, . .;q.' '.. , ' . . 'g.i (;, . . .
~ ,,
tir, '.n . , -J i
.v . , LO Jf h *' . ff , ,Yl . . . *. l .'- /
4 .p. <
4.:
s nejk s ,.
. f. ,> e c
- s.ry g.(o[,s'>s f> .. ,2 b
, . g.gY ~ # o .? .
c,fwn.gA~W~Mg'+ Yr$;G(p .e1,Tf,( g St9 W p,4.D . ~ . . .
.n x.:e ., -
3
~
N'k' ,7, ti .b'* s" #/ $. ,
',) 7.t.
5-19 l
1
~
s ., }g ,
.. g 4 m , ,. . , ~
~
f, e
- . ;'+.
M .
- i. .'
s ,
- p. x ..
. S, p, . ..n' .p?. n?,g:~g1 %n N
~y / ' "^
If
.r eg, T .; \
g.
Cl.=. '[.. *j
, .@ #u -
di{ ,d *
/ -
j'4
- 193~ 7.
l I
i l
Figure 5-12. High Magnification SEM View of the 316L/308 Weld Interface (Note the Abrupt Microstructural Change and Lack of Transition Zone at the Interface) 5-20
e
'%M',R '
l t 'V 1"3 i 'b
.';pi,yg w-, '.
4 def.:'
1 Figure 5-13. High Magnification View of the Weld Matrix Approximately 100 mils from the 316L Fusion Line. (The light phase in the dendrite boundaries has approximately a 4 to 1 enrich-ment of Mo over the austenitic phase.)
5-21
l NEDC-31115 i
l l
i l
r l
l i
i 42 Pt. S SISEC 84913 'HI' i u=512 H=44 KEY 1 1H A0s44KdU 1H l
6 f
i i
i I
l
\
l l
l l
l l
l Figure 5-14. Dispersive X-ray Energy Scan Confirming the Mo Enrichment
]
in the Light Eccond Phase l
5-22 l
- 1
.- 6. UT. REASSESSMENT l I
6.1 NDE OF CORE SAMPLE After the core sample was sectioned and examined metallographically further NDE was performed. Radiographs were taken of the slices which were 1/2 inch to 1/4 inch in thickness. The radiographs were taken using a 1975 KVP X-ray machine and fine grain film (Kodak type R) to maximize the sensi-tivity. Small discontinuities such as lack of fusion and minor inclusions were detected, but no cracks.
Additional UT was also performed on the metallographic slices. First a straight beam (0* longitudinal) was directed from the safe end side of the sample, such that the sound beam was perpendicular to the weld root. A direct reflection was obtained from the fusion line. Secondly a 45' shear wave was aimed from the face of the sample towards the weld root such that it inter-cepted the fusion line at the opposite face of the sample thus forming a corner trap. 'Again a direct reflection was obtained from the fusion line area. In both of these experiments calibration of the UT instrument was performed utilizing a 1/16 inch diameter side drilled hole at a depth of 1/2 inch in a stainless steel block. The reflection from this hole was set at 100% of full screen height. The resultant reflections from the fusion line were 50% of full screen height at a metal path of 0.4 inch in the case of the shear wave and 60% of full screen height at a metal path of 0.5 inch in the case of the straight beam. This arrangement is shown in Figure 6-1. These experiments clearly show that the source (s) of the reflections are still contained in the metallographic sample and are somehow related to the fusion line at the weld root. The only aspect noted to be unusual in this case is the long (1/8 inch),
vertical side walls of the root.* The indication that was detected with refracted longitudinal waves in the field and thought to be due to diffraction at a crack tip is now believed to be related to small slag stringers with associated lack of fusion located near mid wall in the weldment. Records
- A possible explanation for this geometric condition is shown in Appendix B.
6-1
~ - . . - . - . . - - , - - - . -
}
- indicate and metallography shows that this is the height in the weld where a l different size electrode was used and as a result slag stringers might be expected.
4 j- 6.2 OTHER OCCURRENCES I~
This is not the only occurrence where indications similar to those expected of IGSCC have been detected from a weld root with vertical fusion i lines.
! 1 1
! Recently at a foreign BWR site UT indications were evaluated as IGSCC in l
- j. 10-inch diameter, 304 SS recirculation riser welds. The evaluation was made by qualified examiners from the ISI contractor and confirmed by qualified i examiners f rom another contractor. Upon removal no cracks were found;
] however, the welds did have a vertical root / fusion line condition similar to 4
j that seen on the Peach Bottom core sample. Cross sections of these welds are I shown in Figures 6-2, 6-3 and 6-4. Another occurrence was observed when UT ,
was performed on a special CE weld which had vertical side walls. Again, indications were obtained which were similar to those detected at Peach Bottom.
l 6.3
SUMMARY
)
j In summary, it has been shown that certain weld root or fusion line j orientations or specific metallurgical conditions can result in UT signals
- similar to those obtained from IGSCC. Additionally, weld defects, such as
! lack of fusion and/or slag stringers can give indications similar to those i
obtained from crack tips when using refracted longitudinal waves. To overcome ;
! this situation precise cross sectional plotting of indications would be i
required to show that they are located in the weld metal. The root configura-tion and orientation would also be needed. As stated, with the weld geometry information obtained from the core sample, UT indications on the safe end side j of the veld are clearly at the fusion line or in the weld itself. However, j indications on the 304SS pipe side that are located away from the fusion line are obviously in the heat affected zone, and as such are indicative of ,
i
! t j 6-2 3
. - - .s-, w. - . .m.-.--me,w. --,.-~r-%, -. - -
-~w.-c. -..---,,4 - - . - - , , , - , - . , .---..-,....----,--m.~ . - . - - -
. - - ..~- _ _.. - ._ - - - - . . .-- .- - - . - - .- - - - _ - _ _ _ - - - . .
NEDC-31115 significant IGSCC, thus necessitating the weld overlay repair. From a generic standpoint it is clear that the cause or source of the indications munt be determined and an effective evaluation technique developed.
6.4 RECOMMENDATIONS I
4 A special UT procedure is needed to discriminate IGSCC from Peach
, Bottom-3 weld type indications. This would require mockups with vertical root i
welds and with IGSCC located near the root. Samples f rom the field might also be helpful when and if they are available. Such a study could be funded by the BWR Owners Group through EPRI. Certainly the objective of such a program j
would have to be to understand the source of the reflections and the qualifi-cation of discriminating techniques or methods.
i i
I
]
1 I
i j
j l
l I
i
! 6-3 1
/
i
- . - - . . _ . , _ _ _ . _ _ . . - . _ . - . _ . _ _ _ - - _ - - _ - - - _ _ _ . . , . . . . . . _ - ~ . . - .
4 NEDC-31115 WELD ROOT (a) SHEAR WAVE EX AMINATION CUTTING BURR CUTTING BURR ,
g 50%
\,'s ['ss , '
\
'\'
- s
\ 1 i \ E i '
?
I \ 3 s $
I <
l \s I l
' 0 .45 1.0 j
g METALPATH I
I
/ TRANSDUCER WELD ROOT (b) LONGITUDINAL WAVE EXAMINATION (b)
% %g g %
,, I
. ,,- a n\
g
\
l 1 so% i l i \g l
i s
=
sl I
/ 'N B
3
- I g I
< l
/
/
/
/
0 .45 1.0 #
p METALPATH TRANSDUCER Figure 6-1. Schematic of Location and Signal from UT of the 316L Safe End Meta 11ographic Sample 6-4
WW 0 <
fwc M %; 34i u d_ .- N a
[
- GM<v;.7 m , h@~e;rdW-g y
p~ i?p 4 us. ?wgph
. .W ?.g. 4c'. r.s. ..:.
a.;. . d. ~ -:c. u. y. n. se_: ,.,., mas,%y. .
p .,
lR !?$q y
,c,:.y%:f.,p; 0 5- @ : ? .i? g x- :2&'th,
/
?$-??
7:?.OlE5-:, %. .f:dM..p$hf$1&{Y %w&jgl j,$3%$.W
.y.
y:.c
.n, i w *
- .> . '. ; :. .. : . s. _ gA ,
r.;
.7..;
q..
WI'bg,..apS WQ.L'
, .(.rsl %
.gdg
.g, y.
. , y_ 11 :g"d..!..-1;l __ . - :
. . 4* g . . * ; ._ q.* y,.3, sq].g.. &y
~- ; ..~ .s .: <:. -
.dy;
- y. .
yy. .?p:u,,3 t! :115 -QQw.. Q.',;},9
, . ep r, - . , ..
- ,i y e -1
.- ay. s .
Q g> -,L-Q Figure 6-2. Cross Section of Weld A Associated with UT Indications
. c ,*~2 2Tf[@,.7,7'.2.
ar ..~gy
.:.2,' ? . ,,s, Y.-kh, ,
.N. .,'[N. r-N
... . .. ,i: ' r.~. m 'm y- ; g+':
. - . J. W ws -
s '):
- f n. 't ., 3, .
- 6,94.f...s.
- r.n - - ;% ..t'M.-:lW u,,:
>fh. % , T. ki? i, p.a,1 ..t.J '4.ln;:MM,fN. 2. .<
~%
.: ,.* #,s%.:. r.:J:.. p'.; . a c.;
'....>-r.,..-.n
- ^.
. .b'^ . n. 'N'? %v-r..';;, Q , .i.i.~m.
. ;; . Q :, ..;: g, q
-v. A k.-:r .. . ;, x:r
.- , .3. p;c:.
- - .c .e-. l; 1. ,s .w.%
m y, .g
- ,%..
- .,'."::':s.;:..y..A,.,.2,Aga.f.g*~
~ . . . . . .
.+..
7%ry. rig],, : ' ~%,;z~. 1v f;j;.: g} '.%'.m-;9.
,w . ,s
, , . ... ...~. .
- .: J .h : % : ;. ; . . . =T:s.;., r.v ;: . .
g.Q; y~. g's-w.s1. %%' a~".;,; "'f.. ,:. '. y) ["e.,.. ,..
.g;..
, X ;"^%o'y.
- c. ' ' ' ' '
Figure 6-3. Cross Section of Weld B Associated with UT Indications 6-5
~
b%f
~
,, "(( ^ ,
}
A ,2 ,
ll~ . ' '
)
' " Q .44 , ..
- g' i @"# a
.kk * .- .
.e
%e, ) gy. . :. wag t b. .4 +^ *~ "
~u efi%
Ob*Lfh ~' p_p. 3483Ee,qf/Erlg$f,.
n D.Mx5 F $ 4- d Y;A, -
wr m V s.%qw:m.vW.h- am%en.as.
12, Y-(: .. .,s..
h$ ,b I$ h [5 b k 1
Figure 6-4 Cross Section of Weld C Associated with UT Indications 6-6
- 7. LOW CARBON STAINLESS STEEL PERFORMANCE Following the confirmation of no cracking in the low carbon type 316 stainless steel outlet safe end material, the overall experience of low carbon steels was reviewed. In summary, there is a large body of field data and laboratory data that shows that nuclear grades of low carbon material are highly resistant to IGSCC in high temperature oxygenated environments. Each
, set of data will be briefly reviewed.
7.1 FIELD PERFORMANCE Presently, there are approximately 1000 L grade stainless steel pipe welds in GE BWR operating plants. One third of these L grade welds are in small diameter piping (diameters less than 6 inches) where IGSCC generally leads to leakage in short times. None of these welds have leaked to date. In contrast, the behavior of the high carbon material in these size lines has been quite different with a significant fraction of the welds exhibiting cracks or leakage. The balance of the successful L-grade weld experience is in the larger pipe (diameter of 10 inches or greater) with over 200 of the larger welds having greater than 8 years of successful operation. The current overall experience at Peach Bottom 2 and 3 where these welds were inspected, verifies the superior cracking resistance.
7.2
SUMMARY
OF LABORATORY QUALIFICATION A laboratory qualification effort to develop alternate recirculation piping alloys was extensive as discussed in NUREG-10611 and Section 2.0.
The majority of testing was carried out by General Electric in a program sponsored by EPRI as part of the BWR Owners Group program. This program focused on low carbon austenitic stainless steels for a variety of reasons including ease of fabrication, ASME Code acceptability, ease of inspection, and familiarity. The program established that all the low carbon stainless steels of the 304 and 316 type were highly resistant to IGSCC in actual full size pipe tests conducted in oxygenated environments. A large factor of improvement over conventional high carbon type 304 stainless steel was i
7-1 i
NEDC-31115 determined for all the alloys (greater than 55 for the 316NG material). In addition, in transient water chemistry the low carbon 316NG steel performed better than type 304. Microstructural evaluations established the difficulty in sensitizing low carbon materials. Studies also established the higher
~
resistance of low carbon 316 materials to cold worked induced cracking.
! However, studies in bolt loaded fracture mechanics tests did establish that all non-sensitized austenitic stainless steels including both low carbon and stabilized grades could crack in highly stressed creviced locations.
1 7.3
SUMMARY
f All of these results are consistent with the findings at the Peach Bottom 2 and 3 plants. While cracking has been found in the creviced region where the thermal sleeve was attached to the safe end, all of the non-creviced HAZs
{
in the low carbon material have been evaluated to be free of any IGSCC.
4 d
4 S
a f
I 7-2
- 8. REFERENCES I
l
- 1. " Investigation and Evaluation of Stress Corrosion Cracking in Piping f of Boiling Water Reactor Plants", U.S. Nuclear Regulatory Commission Report, NUREG-1061, August 1984.
NP-3671-LD, October 1982.
4
- 3. "r. valuation of Crevice Cracking in Peach Bottom Atomic Power Station Uait 3 Recirculation Inlet Safe Ends", General Electric Company 1
- NEDC-31086-P, September 1985.
j
- 4. Private Communication f j S. J. Kuniya et al, Private Communication, 1984.
i . .
i 1
1 1
l 1
1
- 8-1/8-2 1
i
- _ . - , , , - , , , . - , _ _ _ - . - - _ - . _ _ _ _ . - . - . . - _ - ~ . - - - . . - . _ . _ ~ . _ _ _ _ _ - - . - - _ _ . .
NEDC-31115 Appendix A PLAN FOR METALLURGICAL ANALYSIS OF PEACH BOTTOM-3 CORE SAMPLE INTRODUCTION l
A cylindrical core sample, approximately 1 inch in diameter will be l removed from the safe end side of the safe end-to-pipe weld 2-BS-2 of the 28 inch recirculation outlet nozzle N1B. The sample will be removed as shown in Figure 1.
- 1. MACROSCOPIC A. Visual exam of core sample inner surface. Examine, and photograph macroscopic evidence of weld root grinding.
B. Map the fracture plane in the core sample. Identify and docu-ment the crack on both the inner surface and the cylindrical side of the core. (Note: Do not penetrant examine (PT} or use any material other than deionized water on the sample, so as to avoid masking of possible contaminants on the f racture surface.)
C. Prepare the cutting map and prepare sections for the tests identified below. (See Figure 2.)
- 2. OPTICAL MICROSCOPIC A. On a through wall section, polished, and etched with an appropriate reagent, characterize the fracture relative to the following:
microstructure cold work sensitization (if present)
A-1
i NEDC-31115 1-crack morphology, which emphasis on -
mode, and relation to microstructure branching crack tip
! oxide thickness and distribution a
character of MnS inclusions t
f' -
microhardness profile - and changes with depth below inner pipe surface i
' - examine weld fusion line region for evidence of carbon diffusion or dilution of the 316L by the 308 weld deposit f
(See also Test 4-E) 1 I
microhardness variations with distances from the weld 1 fusion line l
B. Prepare a second section, again on a plane normal to the crack i
plane. The purpose of this section is to provide a more
! detailed examination of the crack tip, and to provide a second 3
4 plane of examination to support the results of the through wall i
section.
l
- 3. SCANNING ELECTRON MICROSCOPY A. Prepare the portion of the core sample selected for surface fractography. Cool the sample in liquid nitrogen, and quickly
.{
< before the sample warms to room temperature, forcibly pry open
- the sample to expose the fracture faces. Back cutting may be 4 necessary to accomplish this.
B. Evaluate the macroscopic features of the fracture, with special emphasis in the regions of crack initiation, stable crack i growth, and the crack tip.
1 A-2 I
,e- r-q-r+-w e+.- g -vemw+-. e,,,wg..e--.+,mven-., -, .
,mg. r,-n.,-,m-m-e w,-=-wmy%.e-%
NEDC-31115 C. Evaluate the microscopic features of the fracture. By noting the surface fractography, verify the fracture mode.
D. Perform an analysis of the oxide compositional gradients, from l the crack mouth to the crack tip.
j E. Perform a microprobe analysis of the inclusions (by the boat sample analysis, the inclusions were found to be MnS).
F. Perform a surface fractography of the pipe inner surface. The intent of this test is to identify a possible correlation between crack pattern and the surface cold work.
- 4. OTHER TESTS l
A. Take surface hardness measurements on the pipe inner surface in the vicinity of the crack. This measurement will identify the possible presence of surface working.
- B. Establish the cold work depth by making microhardness measure-ments on the section polished for optical microscopy. Begin-ning approximately 5 mils from the pipe inner surface, take herdness measurements at 10 mil increments, moving in the pipe through wall direction. At a depth of 70 mils, change the measurement spacing to 50 mils, and continue to the pipe 0.D.
surface.
C. Perform a sensitization measurement by the dual scan EPR method near the I.D. surface on the section polished for optical microscopy (and micro hardness measurements).
D. Establish the strain hardening profile by making microhardness traverses on a polished and etched through wall section.
Beginning at the weld fusion line, make measurements at 10 mil increments (to 70 mils then at 50 mil increments out to a A-3
. _ _ . . _ . = .__ _ _ _ _ _ _ _ _ . . . _
1 i NEDC-31115 i distance of 0.500 inches from the weld fusion line). Make these measurements at the 25%, 50%, and 75% wall positions of the safe end.
i i
- E. Test for the possible presence of Martensite, or other magnetic phases (ferrite, or sigma) on the section prepared for optical microscopy. .
I 1
F. With one of the remaining non-cracked portions of the core sample, decontaminate and perform a wet chemical analysis, with particular emphasis on Carbon, Boron, and Nitrogen. ,
r NOTE: WHILE THIS IS THE EXPECTED PLAN FOR THE METALLURGICAL ANALYSIS OF THE PEACH BOTTOM-3 SAFE END SAMPLE, CHANGES TO THE PLAN MAY BE NECESSARY.
AS WITH ANY FAILURE ANALYSIS, RESULTS OBTAINED EARLY IN THE ANALYSIS
! MAY SUGGEST ALTERNATE TESTS TO BE PERFORMED TO CLARIFY SOME UNFORESEEN ,
t
! OR UNEXPECTED FINDING.
1 4
i l
i 4
j t i
! A-4 I i
. . _ _ . . _ _ _ - . , . . _ . _ . _ . . _ _ _ _ , _ _ _ _ _ . _ _ _ . _ _ _ . . . _ . _ . _ ~ . _ . _ . _ _ _ _ _ _ _ . . . . - . _ _ _ _ . _ _ _ . . _ _ _ . _ . _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _
NEDC-31115 NOZZLE 00 FIPE SAFE END
~
f%I l
i l
=~ -
h lE ;
%,+
4
% s ek.y_,
=:- y CORE SAMPLE CL WELDCL (LAYOUT USING UTI
-.es. e 1/8 in. (DETERMINED U$tNG UT C ATA PLOTS)
NO. ORlLL a 3/8 6n. DEEP r ff
/ Il l
PIPE # s SAFE END 1.22en,
/ if y '-: w I
( M ATER AL REMOVED (7 i 7s i ....n..
' V Q$
=LE , , ,5 .n .
lQ
\ -
Figure i A-5
NEDC-31115 e e a h ,
c h l 5 t 7,
) s h J
) l 4
N- -
e OPTICAL MICROSCOPY e OPTICAL MICROSCOPY ((
9 FERRITIC PHASE e HARONESS MEASUREMENTS STUDIES e SENSITIZATION TESTS CRACX .
M
_/ %
I 3 i eSEM S e OXIDE STUDIES 5 e CHEM AN ALYSIS g 3
OTHER TESTS 3 i
INNER SURP ACE F M ACTOG A APHY Figure 2. Suggested Cutting Plan for Core Sample A-6
NEDC-31115 Appendix B EFFECTS OF FIELD WELD PREPARATION ON JOINT CROSS-SECTION The piping at Peach Bottom-3 was installed using the manual Gas-Tungsten-Arc and the Shielded-Metal-Arc welding processes. The root pass was installed using the extended land weld preparation and an "open-butt" welding technique. This technique uses a 1/8 inch minimum root gap as shown in Figure B-1(a). With the extended weld preparation and the 1/8 inch gap, a normal weld root cross-section as shown in Figure B-1(b) would be expected.
Welds 2-AS-2 and 2-BS-2 were closure welds for the 28-inch piping loop between the recirculation pump and vessel safe end. Accordingly, the closure spools (vertical segment, 90* cibow and horizontal segment) were match mach-ined to the required dimensions using a templating technique. The installa-tion procedure required fit-up of the spool and welding of the vertical weld first, allowing the pipe to shrink vertically until the horizontal weld joint (2-AS-2 and 2-BS-2) was axially aligned. Weld sequence on the vertical welds would also have to be controlled to maintain the 1/8 inch root gap at 2-AS-2 and 2-BS-2.
Realizing the difficulty of accurately templating and machining such a large spool piece, the need to modify the weld preparations in the field would not be unexpected. Such a modification would most likely be performed by hand grinding or filing the extended lands as shown in Figure B-2(a). In this case, the cross-section shown in Figure B-2(b) could result, which is similar to that found in the core sample.
Field construction records are insufficient to show whether this actually occurred, but the practice was commonly used to help obtain the 1/8 inch minimum gap specified in the "open-butt" welding procedures. This weld prep modification could have been made anywhere on the 28-inch piping, but the closure weld spools would be the most probable locations requiring such a field modification. Based on this, the observed root cross-section and anom-alous UT crack indication are not considered typical of the other 28-inch pip-ing in the system.
B-1
NEDC-31115 aoo 3/32 in.+
- I 1
- /
v r 1/Sen GAP NORMAL ROCT (b) R ESULTING WELD (a) NORMAL EXTENDED LAND PREPAR ATION CROSS SECTION FOR "CPEN BUTT" WELDING (OPTION AL REMOVAL OF SCUARE EDGES OF LAND TO 08TAIN MORE UNIFORM ROOT FUSION ALSO SHOWN)
Figure B-1. Nor:nal Weld Preparation and Weld Cross-Section EXTENOED LAND REMOVED BY flung OR GalNDING I
e t /S.n. G AP VERTICAL SiOE W ALL ROOT (b) RESULTING WELD (a) MOOlFIED PREPAR ATION TO CROSS SECTION CORRECT JOINT FIT UP Figure B-2. Modified Weld Preparation and Cross-Section B-2
GENER AL h ELECTRIC
_ . _ . . _ _ _ _ _ _ _ _ _ _ _ _ .